WO1998024472A1 - Uses of thioredoxin - Google Patents

Uses of thioredoxin Download PDF

Info

Publication number
WO1998024472A1
WO1998024472A1 PCT/US1997/022292 US9722292W WO9824472A1 WO 1998024472 A1 WO1998024472 A1 WO 1998024472A1 US 9722292 W US9722292 W US 9722292W WO 9824472 A1 WO9824472 A1 WO 9824472A1
Authority
WO
WIPO (PCT)
Prior art keywords
thioredoxin
cells
trx
human
cell
Prior art date
Application number
PCT/US1997/022292
Other languages
French (fr)
Other versions
WO1998024472A9 (en
Inventor
Garth Powis
Original Assignee
Garth Powis
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Garth Powis filed Critical Garth Powis
Priority to EP97952292A priority Critical patent/EP0946192A1/en
Priority to CA002274116A priority patent/CA2274116C/en
Publication of WO1998024472A1 publication Critical patent/WO1998024472A1/en
Publication of WO1998024472A9 publication Critical patent/WO1998024472A9/en
Priority to US12/694,577 priority patent/US20100166724A1/en

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5011—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing antineoplastic activity
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
    • A61K48/00—Medicinal preparations containing genetic material which is inserted into cells of the living body to treat genetic diseases; Gene therapy

Definitions

  • the present invention generally relates to the use of thioredoxin as, inter alia,
  • a cell growth stimulator as well as a screen for agents that are useful in reducing or
  • Thioredoxin is a low molecular weight (M r 11,000 - 12,000) redox protein
  • thioredoxin reductase Human thioredoxin, which has 5 cysteine (Cys)
  • thioredoxin contains 3 additional Cys residues not found in bacterial thioredoxin that give it
  • cysteine residues that undergo reversible oxidation to cystine.
  • Cys92, Cys69 and Cys73 are found in mammalian but not in bacterial thioredoxins. Cys73
  • Thioredoxin reduces a variety of intracellular proteins including
  • thioredoxin has remarkable extracellular cell growth stimulating properties. It has been
  • Thioredoxin was first studied for its ability to act as reducing co-factor for
  • Transcription factors regulated by thioredoxin include NF- ⁇ B
  • thioredoxin can regulate AP-1
  • eosinophil cytotoxicity stimulating factor protein called eosinophil cytotoxicity stimulating factor (Silberstein DS, et al., J. Biol Chem
  • ADF adult T-cell leukemia-derived factor
  • ADF has been reported to be secreted by virally transformed
  • Thioredoxin mRNA has been found to be over expressed by some human
  • thioredoxin protein was present in certain human tumor cells, and it has not been known that
  • thioredoxin protein played any role in preventing or enhancing tumor cell growth. While thioredoxin itself is known, its use in identifying agents that inhibit cell
  • Human thioredoxin reductase has been characterized as a protein (Oblong JE,
  • MDS myelodysplastic syndromes
  • leukemia may occur in one-third of the patients.
  • the underlying defect is decreased
  • GM-CSF granulocyte-macrophage colony stimulating factor
  • G-CSF granulocyte colony stimulating factor
  • IL-3 displays multilineage progenitor stimulatory effects in normal marrow clinical
  • the present invention relates to the use of thioredoxin as, mter alia, a cell
  • tumor cell growth in a tumor cell that over-expresses thioredoxin comprising contacting said
  • Such agents can include, ter alia, small molecular compounds that complex
  • Such agents can include, inter alia, antibodies to this redoxin, compounds that inhibit the
  • thioredoxin comprising measuring thioredoxin expression or activity in a first sample of said
  • the present invention is based, at least in part, on the discovery that
  • thioredoxin protein is over-expressed in certain human tumor cells; that thioredoxin stimulates the growth of cancer cells; that thioredoxin inhibits apoptosis; that thioredoxin is
  • the present invention involves the new uses of thioredoxin, thioredoxin
  • reductase and mutant forms of thioredoxin for use in screening for anti-tumor agents. It has
  • anti-thioredoxin and/or anti-thioredoxin reductase agents for use as anti-tumor
  • the present invention further relates to the use of thioredoxin and/or
  • thioredoxin reductase antibodies for use as anti-tumor agents.
  • the present invention further relates to the use of anti-sense thioredoxin or
  • anti-sense thioredoxin reductase compounds for use as anti-tumor agents.
  • the present invention further relates to the use of thioredoxin nucleic acid
  • probes and/or thioredoxin antibodies in a diagnostic assay for certain cancers.
  • the present invention further relates to the use of thioredoxin as a target for
  • agents to be used in combination with existing and new treatment therapies such as drugs
  • tumor cells or to increase the sensitivity of tumor cells to these modalities.
  • mutant forms of thioredoxin provide proteins with additional
  • Fig. 1 shows a chart that illustrates the stimulation of human bone marrow
  • CFU-GEMM multilineage progenitors
  • BFU-E erythroid progenitors
  • V V
  • CFU-GM myeloid progenitors
  • Fig. 2 shows a chart that illustrates potentiation of IL-2 induced MCF-7 breast
  • Fig. 3 shows a chart that illustrates the inhibition of thioredoxin stimulated
  • Figs. 4A-B illustrate comparative charts showing the effects of thioredoxin
  • Nucleotide means a monomeric unit of DNA or RNA consisting of a sugar
  • nucleoside base and sugar is called a "nucleoside".
  • the base characterizes the nucleotide.
  • DNA bases are adenine ("A”), guanine (“G”), cytosine (“C”), and thymine (“T”).
  • A adenine
  • G guanine
  • C cytosine
  • T thymine
  • uracil substitutes for T.
  • an A on one strand pairs with
  • DNA comprises deoxyribose as the
  • RNA comprises ribose
  • amino acids are shown either by a three letter or one letter abbreviation as
  • DNA Sequence means a linear array of nucleotides connected one to the
  • Codon means a DNA sequence of three nucleotides (a triplet) which
  • ATG is a translation start signal.
  • Proteins are composed of a linear array of amino acids
  • Gene means the entire DNA of an organism, cell or a virus. It includes,
  • Gene means a DNA sequence which encodes through its template
  • mRNA messenger RNA
  • cDNA means a complementary or copy DNA prepared by using mRNA as a
  • oligonucleotide primer and a mixture of nucleotides.
  • PCR means a polymerase chain reaction whereby a specific DNA sequence, either genomic or cDNA, can be preferentially amplified by the enzyme Taq polymerase
  • Transcription means the process of producing mRNA from a gene or DNA
  • Translation means the process of producing a polypeptide from mRNA.
  • “Expression” means the process undergone by a gene or DNA sequence to
  • Plasmid or phagemid means a nonchromosomal double-stranded DNA
  • a plasmid may be changed or transformed as a result of the DNA of the plasmid.
  • a plasmid may be changed or transformed as a result of the DNA of the plasmid.
  • plasmid carrying the gene for ampicillin resistance transforms a cell previously
  • a cell transformed by a plasmid is
  • Recombinant DNA Molecule or “Hybrid DNA” means a molecule
  • Apoptosis is programmed cell death activated by a genetic program to
  • Oncogene is a gene that encodes a protein able to transform cells in culture to induce cancer in animals.
  • FBS Fetal bovine serum
  • NIH 3T3 cells In a non-limiting embodiment of the present invention, NIH 3T3 cells
  • a nuclear localization signal causes malignant transformation of the cells.
  • lymphoid cells with human thioredoxin DNA has been shown to inhibit apoptosis induced by
  • glucocorticoid N-acetylsphingosine, staurosporine, thapsigargin
  • the thioredoxin gene acts as an oncogene according to the
  • an oncogene a gene that encodes a protein able to transform cells in
  • the thioredoxin gene offers an increased
  • the cells differentiate and move down the villi to eventually be shed into the gastric lumen
  • thioredoxin levels decrease.
  • human MCF-7 breast cancer cells with cDNA for human wild-type thioredoxin or with
  • thioredoxin increases the density to which the NIH 3T3 cells grow in culture and stimulates anchorage-independent colony formation by MCF-7 breast cancer cells.
  • the redox-inactive thioredoxin increases the density to which the NIH 3T3 cells grow in culture and stimulates anchorage-independent colony formation by MCF-7 breast cancer cells.
  • mutant thioredoxin acted in a dominant-negative manner, so that transfected MCF-7 cells
  • embryonic cells with human thioredoxin cDNA increases their growth rate and cell saturation
  • the cells are inoculated into immunodeficient (scid) mice.
  • Trx was originally studied for its ability to act as a cofactor for ribonucleotide reductase. the first unique step in DNA synthesis (1). Human Trx was subsequently shown to modulate the DNA binding of several transcription factors that regulate cell proliferation, including nuclear factor KB ( 2). the glucocorticoid receptor (3), and. indirectly through the nuclear redox protein ReM . activator protein- 1 (Fos Jun heterodimer. Ref. 4). Cloning and sequencing of human Trx have shown that it has a predicted amino acid sequence (5. 61 identical to that of a growth factor secreted by virus-transformed leukemic cell lines, termed adult T-cell leukemia-derived factor (7. 8). Human Trx. but not bacterial Trx. added to the culture medium stimulates the growth of a variety of normal and cancer cell lines (8-10). The addec Trx is not taken up by cells ( 1 1 ) and appears to stimulate cell growtr
  • Trx is required for growth stimulation, and redox-inactive mutant Trxs do not stimulate cell growth ( 10 ) .
  • Trx mRNA levels are increased compared with corresponding normal tissue in almost half human pnmar ⁇ lung ( 5 ) and colon tumors examined ( 13).
  • Trx protein has been reported to be increased in human cervical neopiastic squamous epithelial cells ( 14) and hepato- cellular carcinoma ( 15 ).
  • Trx is excreted from cells ( 16- 18) using a leaderless secretory pathway ( 17). and we have suggested that Trx might be a growth factor for some human cancers ( 1 1 ). However, it remains to be unequi ocally demonstrated that endogenously produced Trx can affect cell proliferation. The role Trx plays in the transformed phenotype of cancer cells also is not known.
  • NIH 3T3 cells Human MCF-7 breast cancer cells and munne NIH 3T3 cells were obtained from the American Tissue Type Collection iRockville. MD). maintained in DMEM containing I0 ⁇ % FBS under 6 c CO : at 37 * C. and passaged before confluence.
  • NIH 3T3 cells were transfected wuh Trx:pRXneo. Trx. pDC30 neo. C32S/C35S.pDC304neo. or pRXneo alone.
  • MCF-7 cells were transfected with Trx:pDC304neo. C32S/C35S:pDC304neo. or pDC3Wneo alone.
  • Transfection used liposomes of .V-(l ⁇ .3-dioleolyl)propyl
  • Cells were selected by growing for 4 weeks in DMEM with lOt FBS and 400 ⁇ g/ml G4I8 sulfate ( Life Technologies. Guthersburg. MD). Cell colonies were isolated by ir simzati ⁇ n onto small squares of sienle filter paper and expanded by growing in (he same medium. All studies were conducted on clonal cell lines between passages 3 and 10.
  • Fig. 2 Effects of transfection with Trx or C32S C35S cDNA on the growth of NIH 3T3 cells.
  • Cells were plated in platuc dishes at a density of 2 x 10* cells cnv in DMEM with lOf' FBS and cell number measured daily.
  • NeoC vector atone-nnsfectcd cells.
  • the apparent decrease in the number of ceils after day 3 is due to detachment of ceils from the plastic surface: T. ⁇ . and ⁇ i. Thioo. Thr ⁇ AD. and Trao9 ceils transfected with Trx cONA: D and ⁇ .
  • Tumor tormati ⁇ n b ⁇ transiected NIH 3T3 cells as studied bs the s.c injection of 10' transected cells in 0 l ml of ste ⁇ le 09'r N ⁇ CI into the backs of groups ol tour male SCID mice or itx nude mice.
  • Tumor fom tiun bv MCF-7 cells was studied b> injecting : ⁇ 10' cells in 0.1 ml ol stenle Q.9 ⁇ > NaO and 0.1 ml of M tngel tBecton Dickinson. Bedford. MAi s.c. into the backs of groups of four tem ⁇ le SCID mice that had been implanted s.c.
  • Stausucal analysis was by Student s nonpaired r test unless otherwise stated. Tumor growth rates in SCID mice were linearized using the cube root of the tumor volume by day for each mouse, and ANOVA was performed using Dunnett's test to determine significant differences from the vector alone- transfected (control) cell line.
  • transfected Trx mRNAs probably because the transfected Trx mRNAs also contain portions of the vector promoter region, the 5' leader sequence, or the polyadenylate tail.
  • the level of transfected Trx mRNA expression was relatively low. being only 0.2- 1.4-fold the endogenous mouse Trx mRNA.
  • Western blomng showed no significant increase in the level of Trx protein tn the cells compared with wild-type or vector aione-transfected cells (results not shown).
  • Trx-transfecied NIH 3T3 cells grew at the same rate on a plastic surface, but reached saturation densiues up to twice that of the
  • vector aione-transfected NIH 3T3 cells (Fig. 2).
  • the vector aione- transfected cells had the same growth characie ⁇ sucs of * ⁇ ld-rype NIH 3T3 cells.
  • NIH 3T3 cells transfected with the redox-inacuve C32S/ C35S Trx grew more slowly and reached a lower saturation density on a plastic surface than the vector alone-iransfected cells Neither the vector aione-transfected NIH 3T3 cells nor the Trx or C32S/C35S Trx transfected cells formed colonies in soft agarose (results not shown ) .
  • Trx-transfected NIH 3T3 cells The ability of transfected NIH 3T3 cells to form tumors when inoculated into tmmunodeficiem mice is used to identify neopias ⁇ c transforming genes (24).
  • Trx-transfected NIH 3T3 cells Thio ⁇ or ThioAD were injected i.m. into SCID or nude mice, there was no tumor formation over 40 days (results not shov.nl. Thus. Trx expression, at least at the level obtained in this study, was not. by itself, transforming.
  • Trx and C32/C35S Transfection of MCF-7 Breast Cancer Cells Human solid cancer cells generally show a greater proliferation response to added Trx than do mouse fibroblasts ( 10. 1 1). This is shown for MCF-7 human breast cancer ceils compared with NIH 3T3 cells in Fig. 3. Thus, we also studied the effects of Trx transfection using MCF-7 breast cancer cells. Transfection of MCF-7 cells with Trx: pDC304neo yielded 31 clones that stably overexpressed Trx mRNA. and transfection with C32S/C35S.pDC304neo yielded 45 clones stably expressing C32S/C35S mRNA.
  • Fig. 4 Expression of transfected mRNAs by some of the clones is shown in Fig. 4 As seen previously with the mouse cells, the transfected human Trx mRNAs tn MCF-7 cells were larger than endogenous human Trx mRNA. The level of Trx mRNA expression was up to 0 8-fold and C32S/C35S mRN up to 2 1 -told the endogenous Trx mRNA levels. Light microscopy showed no difference m the appearance of vector aione-transfected and Trx- transfected MCF-7 cells growing on glass overslips ( Fig ⁇ . and both were similar to wild-type MCF-7 cells In contrast. C32S/C35S- transfected MCF-7 cells appeared more rounded and had ⁇ reduced cytoplasm-to-nucleus ratio
  • Trx-transfected MCF-7 cells showed linear growth characteristics on plastic surfaces over 7 days.
  • the Trx-transfected cells grew at the same rate as the vector aione-transfected MCF-7 cells (Fig. 6). However, when grown in the absence of. or with 0.5 ⁇ t FBS for 2 days, the Trx-transfected cells grew at twice the rate of the vector aione-transfected cells (results not shown).
  • the C32S/C35S-trans- fected cells grew at a significantly slower rate that was 56-78 ⁇ of the Fit 5 lifht microicops ol Trx wdCj'.S/C.'JS cDN'A-tniuiecied MCF-7 breau cancer cell!.
  • the cells were crown to 731 conlluet e on
  • Tissues from the injection site and other organs were taken for histologies! examination at the end of the study.
  • the animals injected with vector alone or Trx-transfected cells showed large solid tumors.
  • the animals injected with Q-2 C33S-transfeeted cells showed small microscopic rumor cell deposits. There was no evidence of tumor metastasis to other organs in any of the animals.
  • Northern analysis of the tumor taken from animals injected with Trx-transfected cells showed the presence of transfected Tn mRNA as determined by its large size ⁇ results not shown i.
  • Trx regulates the redox state and activity of a rtumoer of intracellular proteins that control cell irowth. including ⁇ oonucleoude reductase ( I) and the DNA binding of several transcnotion factors (2— i ⁇ Recombinant human Trx added to normal and cancer ceils in culture stimulates their proliferation 11 1 ).
  • ⁇ oonucleoude reductase I
  • 2 i ⁇
  • Trx i ⁇ Recombinant human Trx added to normal and cancer ceils in culture stimulates their proliferation 11 1
  • Trx may play tn malignant transformation of cells is not known. The present study was undertaken to address some of these questions.
  • NIH 3T3 ceils transfected with T ⁇ t showed an increased cell saturation density when grown as a monolayer on plastic surfaces. Loss of contact inhibition is a feature of transformed cells (24).
  • Trx-transfected NIH 3T3 cells did not form rumors when inoculated into tmmunod- eticient mice.
  • the Trx-transfected MCF-7 ceils did not show increased growth on plastic surfaces in normal FBS. but exhibited significantly increased anchorage-independent growth measured by colony formation in soft agarose. It is surprising that when the Trx-transfected MCF-7 cells were grown as xeno rafts in SCID mice, they exhibited decreased growth rate compared with vector aione- transfected MCF-7 cells.
  • Trx can stimulate the immune system of mice' so that T ⁇ t secreted by the transfected MCF-7 celts might promote some immune rejection, even by the SCID mice, which, although deficient in mature B and T lymphocytes, have natural killer-, myeioid-. and antigen-presenting ceils ( 26 ⁇ .
  • SCID mice which, although deficient in mature B and T lymphocytes, have natural killer-, myeioid-. and antigen-presenting ceils ( 26 ⁇ .
  • Both NIH 3T3 and MCF-7 breast cancer cells transfected with the C32S C23S Trx showed slowed growth rates on a plastic surface.
  • colony formation by MCF-7 breast cancer cells in soft agarose was considerably decreased.
  • the C32S/C33S- ⁇ ransfected MCF-7 cells formed only microscopic tumors.
  • C32S C33S is a rcdoi-inactive mutant Trx that acts as a competitive inhibitor of T ⁇ reductase ( 10).
  • Trx acts as a competitive inhibitor of T ⁇ reductase ( 10).
  • Our X-rav crystallographic studies have identified a highly conserved 1. mmo acid hydrophobic surface on mammalian, out not bacterial. Tr ⁇ ». whicn t- ⁇
  • C31S/C35S is very similar to that of Trx (27) so that C32S/C3SS is likely to participate in the formauon of a heterodimer with Trx and thus might lower Trx monomer concentrations or affect the biological activity of the dimer.
  • C32S/C35S does not stimulate cell growth when added to the culture medium.
  • C32S/C35S might also act as a competitive inhibitor to the normal redox-acnve substrates of Trx Whatever the mechanism, it appears that C32S/C35S acts in a dominant-negative manner to inhibit the effects of endogenous Tn and. in so doing, inhibits cell growth and reverses the transformed phenotype of MCF-7 breast cancer cells.
  • Trx is known to be secreted from cells by a Ieaderless secretory pathway ( 17). The concentrations of Trx found in the medium, up to 10 nvt after 6 h. are lower than those required to directly stimulate cell proliferation ( I I ).
  • Trx at nanomoiar concentrations will potentiate the growth effects of cytokines such as ⁇ nterieuk ⁇ n-2 and basic fibroblast growth factor 4 . It remains to be established whether the extra Trx is producing its effects on cell proliferation through an intracellular or an extracellular action.
  • Trx binds to the urface of cells ( 1 1. 28) so that secreted Trx could have a local effect at the outer cell surface although concentrations tn the medium are low
  • the levels of transfected Tn mRNA in cells were not high, only up to 1-fold endogenous Trx mRNA levels, and were independent of the mammalian transfection vector used Typically. mRNA levels resulting from transfection using such vectors are 10— 50-fold or higher (29). It may be that high levels of Trx gene expression are toxic to cells. We have found only a low expression of the human Tn gene in ⁇
  • transge ⁇ ic mice 4 In .oniraM, some human tumors show very hi-j levels ⁇ i Tn mRN ⁇ compared with the normal tissue: mure than 1 1 -told in human p ⁇ murv lung tumors i 30) and even higher in human p ⁇ mars ol n tumors ⁇
  • thioredoxin was over-expressed in tumor cells compared to normal mucosa, and in all cases
  • lymphocytes Levels of thioredoxin significantly higher than in normal dividing cells, were
  • nuclear proliferation antigen was detected by Ki67 antibody and
  • TUNEL in situ terminal deoxynucleotidyl transferase
  • thioredoxin protein is directly associated with highly proliferative tumors.
  • MCF-7 human breast cancer cells were transfected with cDNA for thioredoxin
  • thioredoxin protein production measured by quantitative Western blotting, up to 100% that
  • thioredoxin can reverse the transformed phenotype and inhibits tumor growth in vivo
  • thioredoxin is a known protein, it has not been disclosed or
  • thioredoxin causes inhibition of anchorage-independent growth of the cells in soft agarose
  • the redox-inactive mutant is
  • agents may be antibodies, drugs or antisense.
  • imidazolyl disulfides inhibit thioredoxin-dependent cell growth (Oblong JE, et al., Cancer
  • NCI was conducted in order to identify compounds with a similar pattern of growth
  • thioredoxin reductase is the only known way for thioredoxin to be
  • reductase could also be a target for the development of anti-cancer drugs.
  • modified thioredoxin does not undergo spontaneous oxidation and/or
  • dimer formation or protected against breakdown by blood and tissues, may have therapeutic
  • transgenic animals in need of stimulation of body growth; (7) in need of simulation of the
  • the underlying defect in myelodysplastic syndrome is decreased multilineage
  • progenitor cell growth associated with decreased sensitivity to growth factor stimulation.
  • Thioredoxin acts to increase the sensitivity of cells to growth factors and stimulates
  • multilineage progenitor cells which provides a beneficial utility in individuals with MDS.
  • Thioredoxin may be used to protect
  • Wild-type and Cys73 ⁇ Ser mutant thioredoxin also stimulates the growth of
  • fibroblasts which are important components of wound healing process.
  • transgenes with wild-type or mutant forms of thioredoxin, with or without tissue specific
  • inducible promotors could be used to stimulate the development of the animal or the
  • cytokines such as IL-2 and fibroblast growth factor (FGF).
  • IL-2 factors given directly or themselves as gene therapy, for example IL-2.
  • modified thioredoxin also stimulates cell growth.
  • thioredoxin has a highly conserved hydrophobic dimer forming surface and that Cys73
  • Cys73 ⁇ Ser thioredoxin is as effective as wild-type thioredoxin at stimulating cell
  • Cys73 ⁇ Ser mutant thioredoxin can prevent the death of lethally y-irradiated mice.
  • mice There were 6 mice in the control group and 4 mice in the treated group. The
  • Cys73 ⁇ Ser mutant thioredoxin stimulates colony formation by the muitilineage progenitor
  • CFU-GEMM but does not stimulate the lineage specific erythroid progenitor (BFU-E)
  • CFU-GM myeloid progenitor
  • Fig. 1. illustrates the stimulation of human bone marrow colony formation by
  • Cys73 ⁇ Ser mutant thioredoxin in accordance with the present invention.
  • Human bone marrow was obtained as excess material from normal allogeneic bone marrow donors.
  • CFU-GEMM CFU-GEMM
  • BFU-E erythroid progenitors
  • V myeloid progenitors
  • IL-2 interleukin-2
  • FGF fibroblast growth factor
  • Fig. 2 illustrates potentiation of IL-2 induced MCF-7 breast cancer cell growth by
  • Cys73 ⁇ Ser mutant thioredoxin in accordance with the present invention. Cells were growth
  • antibodies to the receptors for the growth factors can block the
  • thioredoxin 1 ⁇ M; monoclonal antibodies to FGF receptor, IL-2-receptor and
  • the EGF and EGFR were added as a negative
  • modified thioredoxin does not undergo spontaneous oxidation and/or dimer formation has a
  • proteolytic cleavage to prevent breakdown by plasma enzymes.
  • Thioredoxin/mutant thioredoxin may have use after bone marrow
  • thioredoxin might be more effective in this regard.
  • the use of mutant thioredoxins may not be
  • Thioredoxin is a wiJelv disrnruted redox protein ch t reyulate* sever.ii intracelluiar redov - "cea.e. and stimulates the proliferation of both normal and tumor cells. Wc have round :hat when ihsence of reducing auents. human recombinant Trx undergoes spontaneous oxidation, iosin it ' uinre ;ell yjrowth. bur is still a substrate tor NADPH-derendenr reduction human thioredoxin :re '..
  • Trx ro a that ⁇ > nor a su strat r ' or • eduction T reductase and thar does not stimulate cell proliferation. Both conversions can be induced .ancs and are reversible bv treatment with the chiol reducins auent dithiothrettol. STS- PAGE Trx undernoes oxidation to monome ⁇ c form,s> precedin ⁇ dimer formation.
  • Trx forms a Jtmer that is stahilt.ed by an tntermolecular Cvs '-Cvs .
  • C73S C73S was prepared to determine the tele of Cys' 1 in oxidative rowth stimulation.
  • C73S was as effective a» Trx in s ⁇ mula ⁇ ns cell growth and was a comparable iioredoxin reductase.
  • C73S did not sh ⁇ w spontaneous or oxidant-induced loss of acnvttv and did ner.
  • the results suggest that Trx can exist in monome ⁇ c forms, some of which are mediated bv not stimulate cell proliferation but can be reduced by thioredoxin reductase.
  • Cys' is also involved of an en:ymattcally inactive homodimer. which occurs on long term storage or by chemical us. although clearly involved m protein inactivation.
  • Cvs 1 ' is not necessarv for the zrowth ttvitv of Trx.
  • Trx is a redox protein found in both eukaryotes and pro- aryotes [1J.
  • the redox activity of Trx arises from a highly conserved T -Cys-Gly-Pro-Cys-Lys active site sequence where the 2 cysteine residues (Cys) undergo reversible oxidation to cystine.
  • Reduction of Trx is cataly_ «d by thiore ⁇ doxin reductase (2J.
  • Trx was originally identified in EscJi- ai ia coU as a hydrogen donor for ribonucleotide reductase [3J.
  • Trx has since been found to act as an intracellular dithiol-disulfide reductase and to modulate the activity of a number of intracellular proteins (4-61 including the DNA binding of transcription factors [7-10J. Trx-like sequences are found in other proteins including protein disulfide is ⁇ m- erase (111. There is evidence that Trx may play a role in the growth and transformed phenotype ot some cancers. Trx is
  • Trx act* .w ⁇ ⁇ - enously as a redo -active growth factor.
  • Trx * » • act by a helper mechan ⁇ >m that - ⁇ n « ⁇ t ⁇ :e » the cell? inrowth factors secreted bv the cell.- c> 11 H- Mutant human Trxs.
  • ana could be .tctinu as an autocnne factor tor tne umwth ut Mime cancer ells 1131-
  • T ⁇ t can also orm a homodimer with a 1100 ⁇ 2 inter ace domain and a disulfide bund between Cys 71 from each monomer (20).
  • Cys' - ⁇ Ser mutant human Trx (C73S) was prepared from single-stranded, sense strand human Trx cDNA ligated by polvethylene glycol precipua- non into the pBluescnpt KS vector (Stiatagene. La Jolla, CA) using R408 helper phage.
  • the single-stranded cDNA was used fur uligonucleoride-dtrected m vtrro mu ⁇ genesis (Version 2.1 Kit. Amersham. Biickinghamshirc.
  • the Tn solution was kept at 4* and used within 2 hr (fresh) or stmed in water or 0.1 M potassium phosphate-buttered 0.9% NaQ at 4* or -20* tor specified times.
  • Oxidised Trx for cell trrowth studies was prepared by addinu a 5-told molar excess ⁇ H ; 0» to a 25 ⁇ M Trx stock solution without DTT and I hr later removtnc unreacted H : 0 ; ustnu a PD-10 column.
  • MCF-7 human hre.tst cancer cells were obtained from the
  • Trx reduced/min/mg protein Human placenta thioredoxin reductase, specific activity 33J ⁇ mol Trx reduced/min/mg protein, was prepared as previously described [23]. Reduction of Trx and C73S by thioredoxin reductase was measured by the oxidation of NADPH at 340 nM with insulin as the final electron acceptor as described by Luthman and Holmgren (21.
  • Trx. mutant C73S or C32S/ C35S Trxs Trxs that had been aged at room temperature for 48 hr, 7 davs, 90 days; or Trxs treated for I hr with I mM diamide, 10 mM DTT, 3 mM 2-merca ⁇ toethanol or 2:1 (v:v) HsO : , was mixed with an equal volume of loading buffer containing 3% SDS. 10% glycerol and 0.1% brom- phenol blue in 0.05 M Tns-HCl, pH 6.8.
  • Cys' ⁇ Ser mutant Trx (C73S) stimulated the proliferation of human MCF-7 breast cancer cells.
  • the EC W for gmwth stimulation bv C73S was 350 nM and the maximum effect was seen at 1 ⁇ M. which is similar to values we have previously- reported tor stimulation t MCF-7 t-ell proliferation bv recombinant numan T ⁇ . [IS].
  • FIG. I Stimulation of MCF-7 breast cancer call growth by fresh and aged Trx and C73S.
  • Also shown for reference is the effect of 10% fetal bovine serum. Each value is the mean of 3 dete ⁇ ninations, and bars are SEM.
  • Trx scored in the presence of bovine catalase at 1 unit/ml did not lose biological activity over a 5-day period (results not shown).
  • C73S was a good substrate for reduction by human ptacen- tal d ⁇ oredoxin reductase with a K of 0.20 ⁇ M and a V m m of 6.3 run ⁇ Vr ⁇ in/ ⁇ g. These values are similar to those we have previously found for fresh Trx, which were a K favor of 0 J3 ⁇ M and a V ⁇ of 5.9 M ⁇ l/mm/ ⁇ g (23).
  • Trx When stored in H : 0 either at -20* or at room temperature Trx showed a loss of activity with a hatf-Ufe of
  • the lower pH of the Hilutum in water could stabilize Trx or the increa s e in ionic strencth of phosphate-buffered 0.9"; NaCl could enhance the formation of the inacnve homodimer oi Trx.
  • the aaed Trx showed a slow, delayed reduction bv thioredoxin reductase that was stimulated by catalytic amounts of fresh Trx (Ftp. 2). It is important to note that the loss at activity of Trx as a substrate for thioredoxin reductase was much slower than the loss of activity as a stimulator of cell growth.
  • C73S did not sh ⁇ w a loss of activity as substrate for thioredoxin reductase upon storaee for up to 30 davs.
  • the ability of Trx to act as a substrate for thioredoxin reductase was completely inhibited by treatment wtrh 5 molar equivalents of H 0 : . whereas C375 remained fullv active after treatment with 100 molar equivalent? oi H : 0 : (Fig. 3).
  • Electrophoretic analysis of freshly prepared human Trx stored in DTT showed a mixture of 5 bands of apparent molecular weights tanging from 6.1 to 11 kDa (Figs. 4, 5, and 6, lane 1 ).
  • Storage of Trx at room temperature without DTT resulted in a change in the banding pattern with disappearance of the 8.1 -kDa band by 48 hr (Fig. 4, lane 2).
  • Storage of Trx without DTT for 7 days resulted in the loss of additional bands and the appearance of a new band at 23 kDa due, apparently, to a Trx dimer (Fig. 4, lane 3).
  • Storage of Trx without DTT for 90 days at 4* resulted un almost complete conversion to the Trx dimer (Fig. 4, lane 4).
  • FIG. 3 The effect of H t O t on the reduction of Trx (filled ban) and C73S (open ban) by thioiedoxin reductase.
  • Trx solutions were treated with varying amounts of H t 0 2 for 18 bra at room temperature.
  • Reductaie activity WH measured by adding treated samples to a solution of 0.1 HEPES buffer, pH 7.6, 5 mM EDTA, 17 ⁇ M insulin, 100 ⁇ M NADPH. 15 ⁇ g/ml human thioredoxin reductase and measuring the race of NADPH oxidation at 340 tun at room temperature.
  • Trx dimer a protein thiol oxidizing agent [251 (Fig. 5, lane 5).
  • the formation of Trx dimer following diamide treatment was also confirmed by gel permeation chromatography (results not shown). H j O; treatment of Trx also caused dimer- izati ⁇ n but produced a different banding pattern to that pnxjuced by diamide (Fig. 5, lane 6).
  • Treatment of Trx with NEM. a thiol alkylating agent [26] gave a single band with a slightiy elevated apparent molecular weight, but no dimer formation (Fit:. 5. lane 4).
  • Trx undergoes at least 2 levels of spontaneous and induced oxidative trans, formation.
  • the first oxidation occurs spontaneously within a few days to a form(s) that can n ⁇ longer stimulate ceil growth but remains a substrate for thioredoxin reductase.
  • the slower oxidation occurs over a period of weeks, or can be induced by the thiol oxidizing agent diamide, and leads to a disulfide bonded homodimer which not only fails to stimulate cell growth but is a poor substrate for thioredoxin reductase.
  • Trx can be induced by chemical oxidation, are protected against by catalase and are reversed by the thiol reducing agent DTT is consistent with the interpretation that the changes in Trx are due to oxidation.
  • Cys 71 appears to play a critical role in both levels of oxidant-induced inactivation since C73S does not lose the biological activity or its ability to act as a substrate for thioredoxin reductase upon aging.
  • Trx fresh human recombinant Trx can exist in at least five different states, which probably reflect the fully reduces state of the protein as well as different intramolecular disulfide bonded states due to the five cysteine residues present in the protein. While the specific nature of these intramolecular disulfide bonds is not known, it is likely that some, at least, are due to non-
  • FIG. 4 Effect of storage on Trx studied by SDS-PAGE. Protein was stained with silver stain. Lane 1, fresh Trx: lane 2, Trx 48 hrs at room temperature without DTT; lane 3. Trx 7 days at room temperature without DTT; and lane 4. Trx stored 90 davs at 4 ⁇ without DTT. Position of molecular on the left. narurai disultide ⁇ on ⁇ -. structure?
  • the number uf free thiuls in fresh Trx was determined to be 4.5 to 4.6/molecule by Ell- man ' » reauent [2S] (data nut shown), indicating that all five cvstemes are in the sulfhydryl form.
  • Treatment of NEM- alkvlated Trx with oxidizing or reducing agents produces no change in the banding pattern (data not shown), which is runner evidence that all 5 sulfhvdrvls have been alkvlated.
  • FIG. 6 Oxidation and reduction of mutant Trxs studied bv SDS-PAGE.
  • Trx may be undergoing "native" mtramolecular disulfide bond formation prior to ele ⁇ rophoresis, which preven ⁇ the formation of random dtsulfide-bond formation seen with denaturation and electrophoresu of fresh Trx.
  • bovine pancreatic trypsin inhibitor [27, 31).
  • FIG. 7 Position of cysteines in human Trx. Ribbons and ball-and-s ⁇ ck representation showing the relative positions of Cys", Cys 11 . Cys", Cys M and Cys", based on the crystal coordinates for the wild type redneed protein (20). None of tne thiols are in a position for rtrnilfirlir bond formation except for the redox active pair Cys 12 and Cys 1 *. The inter- molecular r ⁇ wln ⁇ V bond requiring the least distort-on in the protein would be between Cys 31 and Cy»".
  • Trx homodimer 23-lcDa Trx homodimer. Reducn ⁇ n of the Trx dimer by thioredoxin reductase is slow and delayed, and is stimulated by low concentnin ⁇ ns of fresh Trx, suggesting there may be an autocatalvtic process. A similar conclusion was reached by Ren et al. [34
  • Trx undercoes a ⁇ >ter loss of activity with thioredoxin reductase in PBS versus water indicates that iron-induced oxidation or an increase in ionic strength mav stabilize and enhance dimer formation, which is consistent with the hvdrophohic nature ⁇ i the dimer interface observed in crystals of human Trx
  • Trx is secreted bv cells into the extracellular environment, which is ptedominandv oxidizing, and might be expected to undergo monomenc oxidation. Conside ⁇ ng its ease of formation, it is reasonable to assume that monomenc oxidation will precede oxidative homodimer formation. Whether this might be sufficient to prevent Trx from acting as a growth factor is not known.
  • the formation of the oxidized monomer inside the cell is less likely since it s li can be slowly reduced by thioredoxin reductase and the interior of the ceil is highly reducing.
  • Trx homodimer formation The physiological significance of homodimer formation is also unknown. What might be Trx homodimer has been reported in diamide-treated Jurkat celb [35]. We have observed small amounts of the Trx homodimer bv immuno- bloitmg of untreated MCF-7 breast cancer and other cell lysates (Powis et al., unpublished observations). It is in- tnguing to speculate that formation of an oxidized Trx monomer or homodimer in response to intracellular oxi- dants such as H ; 0 ; might be a wav mammalian cells detect oxidant formation.
  • Trx is believed to exist in normal celb at concentrations from I to 10 ⁇ M [2, 12], though in selected tissues and specific cell compartments this value could be much higher. It is therefore not unreasonable to assume that Trx will undergo homodimer formation in vivo. As we observed with the enhanced inacttvation of Trx tn phosphate buffered saline, we expect dimer formation to precede faster m mvo than we observe m wrro m water. Whether dimer formation m wvo would prevent the taster oxidation to an intramolecular form is unknown. The slow autocatalvtic reduction of the Trx homodimer to the monomer would be a way to restore the cell to normal operating conditions after the induction of oxidative stress.
  • Trx undergoes relatively rapid (over a few days) spontaneous and oxtdant-induced conversion to a form(s) that doe « not stimulate cell proliferation, but is still a substrate for reduction bv thioredoxin reductase. There is much dimer (over a period of weeb) spontaneous oxidation of Trx to a Cvs 7, - rab ⁇ li:ed homodimer form that is not a substrate for thioredoxin reductase and that also does not stimulate cell proliferation. Both conversions can be reversed by treatment with the thiol reducing agent DTT. and K i th appear to involve the Cys 71 residue.
  • the redox protein thioredoxin plays an important role In controlUng cancer cell growth through regulation of DNA synthesis and transcription factor activity. Thioredoxin is overexpressed by a number of human primary cancers and its expression is decreased during dexamethasone- induced apoptosis of mouse WEHT7.2 thymoma ceils. We examined the ability of WEHT7.2 cells stably transfected with human thioredoxin cDNA showing increased levels of cytoplasmic thioredoxin to undergo apoptosis in vitro and in vivo.
  • the cells were protected from apoptosis induced by dexamethasone, staurosporine, etoposide, and tharnigargin, but not by JV-acetyl-sphingosine.
  • the trx-transfected cells formed tumors that showed increased growth compared to wild-type, as well as ⁇ c/-2-transfected, WEH17-2 cells.
  • the trx- and £>c/-2-transfected cell tumors both showed less spontaneous apoptosis than tumors formed by the wild-type ceils.
  • f ⁇ r-translected cell tumors did not show growth inhibition upon treatment with dexamethasone. This study suggests that increased thioredoxin expression in human cancers may result in an increased tumor growth through inhibition of spontaneous apoptosis and a decrease in the sensitivity of the tumor to drug-induced apoptosis.
  • Trx 3 is a low molecular weight redox protein found in both pro- karyotic and eukaryouc cells (1).
  • the cysteine residues at the conserved Cys 32 -Gly-Pro-Cys 35 -Lys active site of Trx undergo reversible oxidauon-reduction catalyzed by the NADPH-dependent selenium- containing flavoprotein Trx reductase (2).
  • Human Trx is a protein of ⁇ f r 1 1.500 with 27% amino acid identity to Eschenchia coli but containing three additional Cys residues not found in bacterial Trx that give the human protein unique biological properties (3).
  • Trx was originally studied for its ability to act as a reducing cofactor for ribonucleotide reductase. the first unique step in DNA synthesis (4). More recently. Trx has been shown to exen redox control over a number of transcription factors, including nuclear factor KB (5), transcription factor I11C (6). BZLF1 (7), and the glucocorticoid receptor (8). and indirectly, through nuclear redox factor Ref-1/HAPE, Trx can regulate AP-1 (Fos/Jun heterodimer; Ref. 9).
  • Trx is also a growth factor with a unique mechanism of action.
  • Human Trx stimulates the proliferation of both normal fibroblasts and a wide va ⁇ ety of human solid and leukemic cancer cell lines (10, 11).
  • Redox activity is essential for growth stimulation by Trx.
  • mutant redox-inactive forms of Trx lacking the active sue Cys 3: and Cys 35 residues are devoid of growth stimulating activity (1 1 ).
  • Studies with ' ⁇ Mabeled Trx have revealed no high-affinity binding sues that
  • Trx thioredoxin scid. severe combined im unodefi- ⁇ rnl CAT chlnramnhenicni acetvltran ferase might suggest receptors for Trx on the surface ctfcaaccr cells (12). Trx appears to -» «"»ni— cell raOliferanon by ittcRating the sensiuviry of the cells to growth factors secreted by the ceUs themselves (12).
  • Trx overexpresaion may be a factor in the growth of some i ⁇ ftff 1 cancers.
  • Trx gene expression is decreased during n>xa ⁇ nethasone-inch ⁇ ced apoptosis of mouse triytnorr-a-derived WEHT7-2 cells (17).
  • Trx cDNA btixnan Trx cDNA
  • fxtminrri the effects on both spontaneous and drug- induced apoptosis in vitro and with the cells growing in scid mice.
  • Trx cDNA Human wild-type Trx cDNA was prepared as described previously, cloned uao the ⁇ forl sne of the pDC-MM ⁇ co mammalian tiw-sfccnon veci ⁇ r ( 16) and transfected by clecsoporsflon uao mouse W ⁇ H ⁇ .2 trjy ⁇ nc*na-derived cells (18).
  • Tunife ⁇ rd cells were maintained at cuhure de-tancs up to 10* ccUs/ml in DMEM romainrnj 10% fetal b ⁇ vme serum supplemented with 800 ⁇ ml C 18 surfate, and clo ⁇ ea were laolased ia soft afarose and then maintained u. 200 uf/ml G418 sulfate.
  • AU studies were nmriurirri on clonal hues between passages 3 and 20.
  • Stably transfected M-2 WEH-7-2 celb (W_Hbl2 cells) were obtained from Dr. Roger Miesfeld (Umversrry of Arizona. Arlington. AZ: Ref. 19).
  • Drup were added at a culture density of I X 10* to 2 X 10 s cxUs ml.
  • Stock solunons (10 mM) of (texam-that ⁇ ne were utepsiui ia ethanci, whereas -taurotporme. etoposide. th- ⁇ st ⁇ rtui. and N-acesyl-sphuigostne were prepared tn DMSO. Further d ⁇ oons were made unng culture medium.
  • Gsacscart-coid BareptTi The level of funcnonal fjucocorucoid receptors was assessed using a transient ccvansfecuoo of cells with a glucocorocoid response elemenf CAT reporter p-tsrmd (p ⁇ unCAT; Ref. 20) and ⁇ -falac ⁇ o- ndase. After a 22-h recovery period, the cells were treated with I ⁇ M dexa- me aaone. and CAT protein was measured after in additional 24 h unng s CAT EL1SA (Boeh ⁇ nger Mannheim. lncJanapolu. IN). An aliquot of the transfecied cells was stained for 0-salsct ⁇ uda-e activity and CAT activity normahzed for transfection efficiency.
  • Apiptnr-i Apoptotu was measured by an E ISA for htsto ⁇ -a-sociated DNA frasmema (21). by morphotofy and by flow cytometry (22).
  • the cntena uaad for the ⁇ >orphc of ical idennficauon of apoptonc ceils included condensation and rr-crfinsoon of the chromc ⁇ n with the formation of crescents, cellitn aie. increased -uinint. nuclear frsfToentanon.
  • the coverslips were exposed lo a I 1 (X) dilution ot goat aniirabbit biotinsiaied IgC for I h. washed with PBS. exposed to a 1.30 dilution ot fluorescein streptavidm fluorochrome. and again washed with PBS Cells were examined using a Leica TCS-ID laser scanning contocai microscope with an excitation wavelength of 488 nm. For subcellular localization studies ut Trx. Cy5 i indodicarbocyanme) streptavidm was used as the fluorochrome.
  • Tumor lormation by wild-type and iranstected WEHI7 2 cells was studied bv injecting 2 x 10' ceils in 0.1 ml ot mat ⁇ gel s.c into the flanks ol groups ot 20 temale scid mice. Tumor volume was measured with calipers, and mice were euthanized when the tumor volume exceeded 2 cm"
  • 10 mice trom each group were injected i p with I mg/kgdav dexamethasone in lO ⁇ r ethanol in 0 T- NaCI Control mice were injected with vehicle alone. On dav 14. three mice tmm each group were euthanized with CO.. and the tumors were excised and immediately fixed in glutaraldehsde
  • Tissue for Bright-Field Examination The glutaralde- hyde-fixed tissue was posifixed in osmium tetroxide. dehydrated in a graded se ⁇ es ot alcohols and embedded in epoxv resin One- ⁇ m-lhick sections were prepared and stained with loluidine blue for b ⁇ ght-field examination
  • VVEHI7 2 cells were stablv transfected with human Trx cDNA in the pDC304neo mammalian transfection vector
  • Tr ⁇ 5 and Trx6 Fig IA
  • Tr ⁇ 5 and Trx6 Fig IB
  • the rrt-transfected WEHI7.2 cells were resistant to apoptosis induced by 1 ⁇ M dexamethasone as measured by histone-associaied DNA fragmentation (Fig. 2A l or by flow cytometry (Fig 23). Histological examination of the W ⁇ H17.2 cells revealed a classic apoptotic morphology in response to dexamethasone However, only a small fraction of the cells undergo apoptosis at any one time, and they rapidly progress to fragmented cells For this reason, results are expressed as relative apoptosis rather than percentage of apoptotic cells.
  • Glucocorticoid receptor activity measured using a glucocorticoid receptor/CAT reporter plasmid was not decreased in the rrt-transfected cells (results of three studies not shown)
  • Trx is prestm in ine isioplasm and ihe nucleus of wild-type WEHI7 _ cells ⁇ 5> and 7> ⁇ o in transiecied cells toi
  • the cells were treated with 0.01* ethan ⁇ l vehicle ⁇ > or I ⁇ M dexamethasone (Q). md spoptotis was cute ⁇ urad 24 later. Coiwiwu. men of four detcrmina- uoni; ban. SE. •. P ⁇ 0.05 compared to Neo controi. B. apoptosis measured by flow cytometry showi ⁇ f typical lesulu. Ref rani ⁇ l. R2. and ⁇ J of the vanerpraros ate live nonapoptotK. early apoo- lotic. and late apoptotic cellt. respectively. /. pOC3 4neo vector-alone- transfected control cells: 2.
  • the trx- transfected WEHI7.2 celK formed tumors that grew more rapidly than tumors formed b> cither wild-type or bcl-2- transfected WEHI7.2 cells rFig. 3 ⁇ ) Upon histo- > o
  • tumors formed by the vwld-type cells showed fields of apopiouc cells adjacent lo fields ot viable cell , as well as apoptotic cells ad ⁇ uxed with viable-appearing cells (Fig 3fl ⁇
  • the cells undergoing apoptosis exhibited the classic appearance of condensed and argmated chromaun. some in the form of crescents, and j dense cvtoplasm accompanied by vacuolization
  • the rn -transfecied EHI7.2 cell tumors showed minimal numbers ot cells undergoing O 98/24472 £>
  • Trx may be able to directly reduce redox-regulated nuclear transcription factors, such as AP-1 (Fos Jun heterodimer. Ref. 9). If Trx can enter the nucleus, it may not need other nuclear redox factors, such as Ref- 1 /HAP 1. as has been suggested (9).
  • the rrr-tra ⁇ sfected cells were resistant to apoptosis induced by dexamethasone. Trx has been reponed to be necessary for assembly of the glucoconicoid receptor (8). However, glucoconicoid receptor activity was not decreased in the transfected cells, suggesting that the effects of Trx on apoptosis appear to lie downstream of the glucocorticoid receptor.
  • the /nr-iransfected cells also showed resistance to apoptosis induced by siaurosporine. etoposide. ⁇ /-acetyi-sphingosine. and thapsigargin.
  • Exogenously added human Trx has been reported to inhibit apoptosis induced by tumor necrosis factor a in U937 human lymphoma cells (26). However, we found that exogenously added human Trx did not protect WEHI7.2 cells against apoptosis induced by dexamethasone (27). Tumor necrosis factor a and dexamethasone are thought to trigger apoptosis by different signaling pathways. It may also be that exogenous Trx is not taken up by WEHI7.2 cells. We have found that other tumor cells take up Trx poorly, if at all (12). Clearly, an increase in intracellular Trx achieved by transfection of trx in the present study is associated with resistance of the WEHI7.2 cells to apoptosis induced by dexamethasone and other agents.
  • Bcl-2 is believed to exe ⁇ its inhibitory effects upstream of the activation of the cysteine aspanate proteases cascade (caspase) responsible for the final stages of apoptosis (28).
  • caspase cysteine aspanate proteases cascade
  • the protective effects of Bcl-2 against apoptosis have been suggested to involve an antioxidant mechanism (29). although this is disputed based on the ability of Bcl-2 to block apoptosis caused by agents that are thought not to act by an oxidant mechanism (30) or caused by hypoxia (31 ).
  • WEHI7.2 cells transfected wiih trx formed tumors in scid mice that grew considerably faster than tumors formed by the wild-type parental celK or b nr/-2-transfected cells. This may be due. in pan. to a decreased rate of spontaneous apoptosis that occurred in ihe trx- transfected cell tumors.
  • High levels of Bcl-2 have been found in a wide variety of human cancers (34). Although transfection with bcl-2 is known to confer resistance to apoptosis induced by anticancer drugs and radiation, the effects of bcl-2 on tumor growth are less clear. Transfection with bcl-2 gives a survival advantage to cells in culture ( 351.
  • trx a gene found to be overexpressed in a number of human cancers, can inhibit apoptosis of cancer cells in culture induced by a variety of agents.
  • the rrx-transfected cancer cells show an increased growth, decreased spontaneous apoptosis, and decreased sensitivity to apoptosis induced by dexamethasone. If similar effects occur in patient tumors, then trx could be a new human proto-oncogene.
  • FIG. 1 Stimulation of human bone marrow colony formation by Cys73 ⁇ Ser mutant thioredoxin. Human bone marrow obtained as excess material from normal allogenetc bone marrow donors. The effects of Cys73 ⁇ Ser thioredoxin on colony formation by: (o) multilineage progenitors (CFU-GEMM); (•) erythroid progenitors (BFU-E); and (V) myeloid progenitors (CFU-GM), was measured over 10 days as described. Values are the mean of 4 dete ⁇ inations and bars are S.D.
  • CFU-GEMM multilineage progenitors
  • BFU-E erythroid progenitors
  • V myeloid progenitors
  • FIG. 1 Potentiation of IL-2 induced MCF-7 breast cancer cell growth by Cys73 ⁇ Ser mutant thioredoxin.
  • Cells were growth arrested for 48 hr in medium with 0.5% serum (10 s cells) then stimulated in the absence of medium with either IL-2 or Cys32 ⁇ Ser mutant thioredoxin at the concentrations shown. Cell number was measured after 48 hr. Each point is the mean of 3 deteiminations and bars are S.E. The dotted line shows stimulation by 10% serum.
  • FIG. 3 Inhibition of thioredoxin stimulated MCF-7 cell growth by receptor antibodies. Cell proliferation was measured as described in Figure 2. The concentrations of agents used were Cys73 ⁇ Ser mutant thioredoxin (Trx) 1 ⁇ M; monoclonal antibodies to FGF receptor, IL-2-receptor and EGF-recep * .or 4 ⁇ g/rnl; and EGF 10 nM. The EGF and EGFR were added as a negative control. Values are the mean of 3 detexi-iinations and bars are S.E. The dotted line shows the effect of 10% serum alone.
  • Trx Cys73 ⁇ Ser mutant thioredoxin
  • FIG. 2 Effects of Trx and C32S/C35S cDNA transfection on proliferation of MCF-7 cells.
  • Left panel 3 x 10* cells were plated in 3.8 cm 2 plastic culture dishes in DMEM with 10% fbs and cell number measured 7 days later.
  • Right panel 10* cells were plated in 2 cm 2 wells containing soft agarose and colonies measuring >60 ⁇ . counted 7 days later.
  • Control the Neol vector-alone tranfected MCF-7 cells; Trx 9, Trx 12, and Trx 20, MCF-7 cells transfected with human Trx cDNA; Serb 4, Serb 15 and Serb 19, MCF-7 cells transfected with C32/C35S cDNA.
  • Values are the mean of 3 determinations and bars are S.E. **indicates p ⁇ 0.01 compared to vector-alone transfected cells, •" ⁇ indicates p ⁇ 0.01 compared to vector-alone transfected cells.
  • Human thioredoxin is a putative oncogene that may confer both a growth and survival advantage to tumor cells. Over-expressed thioredoxin mRNA has been found in both primary human lung and colorectal cancers. To determine the intratumor distribution and amount of thioredoxin protein in human primary tumors and to determine if its overexpression is related to proliferation or apoptosis, we studied primary human gastric carcinoma samples. An immunohistochemical assay for thioredoxin in paraffin embedded blocks was developed. We studied ten patients with primary high risk gastric carcinoma.
  • thioredoxin protein overexpression to apoptosis we utilized a paraffin based in situ assay (Tunel) and to delineate proliferation we utilized the nuclear proliferation antigen detected by Ki67.
  • Thioredoxins are low molecular weight redox proteins found in both prokaryotic and eukaryotic cells (1).
  • the cysteine (Cys) residues at the conserved -Cys-Gly-Pro-Cys-Lys active site of thioredoxin undergo reversible oxidation-reduction catalyzed by the NADPH-dependent selenium containing flavoprotein thioredoxin reductase (2).
  • Human thioredoxin is an 11.5 kDa protein, with 27% amino acid identity to E. coli thioredoxin. It contains 3 additional Cys residues not found in bacterial thioredoxin that give it unique biological properties (3).
  • Thioredoxin was first studied for its ability to act as reducing cofactor for ribonucleotide reductase, the first unique step in DNA synthesis (4). More recently thioredoxin has been shown to exert redox control over a number of transcription factors, including NF-KB (5), FIHC (6), BZLF1 (7), the glucocorticoid receptor (8) and, indirectly through another redox factor Ref-1, AP-1 (Fos/Jun heterodimer) (7). Thioredoxin modulates the binding of the transcription factors to DNA and thus, regulates gene transcription.
  • Thioredoxin is also a growth factor with a unique mechanism of action.
  • the predicted amino acid sequence of thioredoxin is identical to that of a previously identified growth factor secreted by HTLV-1 transformed leukemic cell lines, called adult T-cell leukemia-derived factor (ADF) (3).
  • ADF stimulates growth of lymphoid cells (9,10).
  • human recombinant thioredoxin stimulates the proliferation of normal f ⁇ broblasts and human solid tumor cancer cells even in the absence of serum (1 1, 12). It does this by increasing the sensitivity of the cells to growth factors secreted by the cells themselves (13).
  • thioredoxin at nM concentrations increases the sensitivity of human breast cancer cells to interleukin-2 (IL-2) and basic fibroblast growth factor (bFGF) by 1000 and 100 fold, respectively (unpublished observations).
  • IL-2 interleukin-2
  • bFGF basic fibroblast growth factor
  • the term "voitocrine”, from the Greek “to help”, has been coined to describe this growth stimulating activity of thioredoxin (13).
  • Mutant redox-inactive forms of thioredoxin lacking the active site cysteine residues and E. coli thioredoxin are devoid of growth stimulating activity (12).
  • Human thioredoxin is known to be secreted from cells by a leaderless secretory pathway (15) so that it could be acting extracellularly to stimulate cancer cell growth.
  • thioredoxin is important for the growth, death and transformed phenotype of some human cancers. Stable transfection of normal fibroblasts with human thioredoxin cDNA (trx) increases their growth rate and transfection of human MCF-7 breast cancer cells with trx increases their colony formation in soft agarose (16). Transfection of the MCF-7 cells with a dominant negative redox inactive mutant trx causes inhibition of colony formation and almost complete inhibition of tumor formation when the cells were inoculated into scid mice.
  • Immunohistochemistrv Five micron thick sections were deparaf inized and then subjected to antigen unmasking with one of two methods with heat plus citrate buffer at pH 6.6 or microwave plus EDTA buffer at pH 8.0 as previously described (20,21). The best signal to noise ratio was established by judging reactivity with cell lines known to be a high expressor of thioredoxin (A549 human lung cancer) and a low expressor of thioredoxin (SK BR3 human breast cancer)(19).
  • the degree of thioredoxin expression in tumor cells was judged at 400x magnification as 4+ (very intensely positive), 3+ (moderately intensely positive), 2+ (moderate), 1+ (faint), or 0 (completely negative) throughout the sample.
  • a single investigator (TG) was responsible for scoring all the samples.
  • Ki67 proliferation antigens
  • biotin-avidin labelled method after avidin blocking 212.
  • the degree of Ki67 staining was classified as the percentage of nuclear positive tumor cells listed as: absent (0), >0-5% (+), 6-25% (++), 26-50 (+++), >51% (++++).
  • Apoptosis Assay Apoptotic cells were detected utilizing the TUNEL assay (23, 24) adapted to an automated in situ hybridization instrument (gen II, Ventana Medical Systems, Inc.).
  • the TUNEL assay utilizes recombinant terminal deoxynucleotidyl transferase (Tdt) (GLBCO BRL) for adding homopolymer tails to the 3' ends of DNA which are more abundant in apoptotic cells(23, 24).
  • Biotin-16, 2'-deoxyuridine-5'-triphosphate (Biotin 16-dUTP) was the label used for terminal transferase in this DNA 3' -end labelling reaction.
  • Avidin-Horseradish Peroxidase and 3,3'-diaminobenzidine as chromogen(23, 24).
  • the instrument utilized deparaffinized sections with subsequent digestion with Protease I (Ventana Medical Systems Arlington, Az.) for 8 minutes VMS1). Incubations were performed per Ventana Gen II protocol on the instrument with the final steps being as above using avidin-horse radish peroxidase and DAB detection method to visualize the apoptotic nuclei as an intense brown color (diaminobenzidine). As an enzyme control we utilized two sections from each tissue: one with Tdt
  • the TUNEL assay result was scored by the number of brown - apoptotic tumor nuclei per high power field (400x objective). The values were: 0 (absence of apoptotic cells), + (>0-2/hpf), ++ (2-4/hpf), +++ (>4-8/hpf), ++++ (>8/hpf).
  • Thioredoxin expression was correlated with Ki67 expression and with apoptosis measured by the TUNEL assay using Spearman's nonparametric rank correlation test.
  • the optimum signal to noise ratio was found by using the following antigen retrieval conditions: microwaving at pH 8.0 in EDTA as tested by a high thioredoxin expressor (A 549) and low thioredoxin expressing (Sk BR3) cell line.
  • the lower level cells in the pits showed cytoplasmic and scattered nuclear staining, while the higher mid-level graduation staining was typically lighter and restricted to the cytoplasm. The significance of this differential distributions is not known.
  • Thioredoxin does not have a known nuclear localization sequence (3).
  • thioredoxin is specifically located within neoplastic gastric carcinoma cells and not in stromal cells or admixed B or T lymphocytes or macrophages.
  • the tumor cell thioredoxin density typically exceeded that of the adjacent normal mucosa.
  • the hematoxylin and eosin stains reveal a pleomorphic carcinoma invading the gastric wall.
  • the thioredoxin expression (right upper and lower panels) is present in both the nuclei and cytoplasm of tumor cells in malignant glands and in rare associated leucocytes. Thioredoxin expression is absent in the adjacent stroma (lOOx to 400x).

Abstract

The present invention relates to the use of thioredoxin as, inter alia, a cell growth stimulator, as well as a screen for agents that are useful in reducing or preventing thioredoxin-associated apoptosis inhibition and agents that are useful in inhibiting thioredoxin stimulated cell growth.

Description

USES OF THIOREDOXIN
1. BACKGROUND OF INVENTION
The present invention generally relates to the use of thioredoxin as, inter alia,
a cell growth stimulator, as well as a screen for agents that are useful in reducing or
eliminating thioredoxin-associated apoptosis inhibition and agents that are useful in
inhibiting thioredoxin stimulated cell growth.
Thioredoxin is a low molecular weight (Mr 11,000 - 12,000) redox protein
found in both prokaryotic and eukaryotic cells. (Holmgren A., J Biol Chem
264:13963-13966 (1989)), that undergoes reversible thiol reduction by the NADPH-
dependent enzyme thioredoxin reductase. Human thioredoxin, which has 5 cysteine (Cys)
residues, is a 11.5 kDa protein with 27% amino acid identity to E. coli thioredoxin. Human
thioredoxin contains 3 additional Cys residues not found in bacterial thioredoxin that give it
unique biological properties. (Gasdaska PY, et al, 1218:292-296 (1994)). Cys32 and Cys35
are the conserved catalytic site cysteine residues that undergo reversible oxidation to cystine.
Cys92, Cys69 and Cys73 are found in mammalian but not in bacterial thioredoxins. Cys73
appears to be particularly important for maintaining the biological activity of thioredoxin in
an oxidizing environment. Thioredoxin reduces a variety of intracellular proteins including
enzymes such as ribonucleotide reductase which is important for DNA synthesis, and critical
Cys residues in transcription factors such as NF-κB, AP-1 and the glucocorticoid receptor,
thus, altering their binding to DNA. In addition to its intracellular actions, human
thioredoxin has remarkable extracellular cell growth stimulating properties. It has been
reported (Gasdaska PY, et al., 1218:292-296 (1994)) that thioredoxin is identical to a growth
factor reported to be secreted by human HTLV-1 transformed leukemia cell lines (Fox JA, et
al., Proc. Natl Acad Sci USA 84:2663-2667 (1987)). It has also been found that human recombinant thioredoxin will stimulate the growth of a wide variety of fibroblast and human
solid tumor cell lines in culture (Gasdaska JR, et al. Cell Growth Differ 6: 1643-1650 (1995);
Oblong JE, et al. J Biol Chem 269:11714-11720 (1994)). E. coli thioredoxin does not
stimulate cell proliferation.
Thioredoxin was first studied for its ability to act as reducing co-factor for
ribonucleotide reductase, the first unique step in DNA synthesis. (Laurent TC, et al., J Biol
Chem 239:3436-3444 (1964)). More recently thioredoxin has been shown to exert redox
control over a number of transcription factors modulating their binding to DNA and thus,
regulating gene transcription. Transcription factors regulated by thioredoxin include NF-κB
(Matthews JR, et al., Nucl Acids Res 20:3821-3830 (1992)), TFIIIC (Cromlish JA, et al., J
Biol Chem 264:18100-18109 (1989)), BZLF1 (Bannister AJ, et al. 6:1243-1250 (1991)), the
glucocorticoid receptor (Grippo JF, et al. J Biol Chem 258:13658-13664 (1983)) and,
indirectly through a nuclear redox factor Ref-1/HAPE, thioredoxin can regulate AP-1
(Fos/Jun heterodimer) (Abate C, et al. Science 249:1157-1161 (1990)). Thioredoxin is also a
growth factor with a unique mechanism of action.
Human thioredoxin has been sequenced and cloned. (Gasdaska PY, et al.,
1218:292-296 (1994); Deiss LP, et al., Science 252:117-120 (1991)). It has been shown that
the deduced amino acid sequence of thioredoxin is identical to that of a previously known
protein called eosinophil cytotoxicity stimulating factor (Silberstein DS, et al., J. Biol Chem
268:9138-9142 (1993)) or adult T-cell leukemia-derived factor (ADF) (Gasdaska PY, et al,
1218:292-296 (1994)). ADF has been reported to be secreted by virally transformed
leukemic cell lines and to stimulate their growth (Yodoi J, et al., Adv Cancer Res 57:381-411
(1991)). These observations have been extended to show that human recombinant thioredoxin stimulates the proliferation of both normal fibroblasts and a wide variety of human solid and
leukemic cancer cell lines. (Gasdaska JR, et al. Cell Growth Differ 6:1643-1650 (1995);
Powis G, et al., Oncol Res 6:539-544 (1994); Oblong JE, et al. J Biol Chem
269:11714-11720 (1994)). It has been shown that thioredoxin stimulates cell proliferation by
increasing the sensitivity of the cells to growth factors secreted by the cells themselves.
(Gasdaska JR, et al. Cell Growth Differ 6:1643-1650 (1995)).
Recombinant modified thioredoxins, otherwise called mutant thioredoxins,
have been developed, but no indications of uses were known in the art for any particular
mutant form. In a wild type thioredoxin, the cysteine (Cys) residues at the conserved -Cys32-
Gly-Pro-Cys35-Lys active site of thioredoxin undergo reversible oxidation-reduction
catalyzed by the NADPH-dependent flavoprotein thioredoxin reductase. (Luthman M, et al.,
21 :6628-6633 (1982)). It has been reported that mutation of the active site Cys32 and Cys35
residues to serine (Ser) residues, either singly or together (C32S/C35S thioredoxin), results in
a compound that is redox inactive (i.e., it is not a substrate for reduction by thioredoxin
reductase) and that does not stimulate cell proliferation (Oblong JE, et al. J Biol Chem
269:11714-11720 (1994)).
Thioredoxin mRNA has been found to be over expressed by some human
tumor cells (Gasdaska PY, et al, 1218:292-296 (1994); Grogan T, et al., Cancer Res (1997,
in press)) and since it is secreted from cells by a leaderless secretory pathway (Rubartelli A,
et al., J Biol Chem 267:24161-24164 (1992)) it could be a growth factor for some human
cancers (Gasdaska JR, et al. Cell Growth Differ 6:1643-1650 (1995)). However, the
mechanism for cell growth stimulation by thioredoxin mRNA has been examined and found
not to promote cell growth. Recombinant human thioredoxin is not taken up by cells and does not bind to high affinity cell surface receptors but appears to enhance the sensitivity of
cells to endogenously produced or other growth factors, a mechanism termed voitocrine
(Greek, voithos = helper) (Gasdaska JR, et al. Cell Growth Differ 6:1643-1650 (1995)).
The in vitro cell growth stimulating activity of human thioredoxin has been
previously reported for human lymphoid and solid tumor cancer cells (Gasdaska JR, et al.
Cell Growth Differ 6:1643-1650 (1995); Oblong JE, et al. J Biol Chem 269:11714-11720
(1994)) and for mouse fibroblast cells (Oblong JE, et al. J Biol Chem 269:11714-11720
(1994)). The production of a Cys73→Ser mutant thioredoxin has been previously reported.
In one study it did not act like wild-type thioredoxin as a component of a complex cell
growth stimulating factor called "early pregnancy factor" (Tonissen K, et al., J Biol Chem
268:22485-22489 (1993)). In another study it was reported that Cys73→Ser mutant
thioredoxin did not form a dimer, but cell growth stimulating activity by the mutant
thioredoxin was not investigated in this study (Ren X, et al., Biochem 32:9701-9705 (1993)).
However, the ability of the Cys73→Ser mutant and other mutant thioredoxins to stimulate
cell proliferation has not been reported. There have been no prior reports of administration of
mutant thioredoxins in vivo.
It has been known that certain human tumor cells were found to over-express
thioredoxin mRNA compared to normal lung tissue from the same subject. (Gasdaska PY, et
al., 1218:292-296 (1994)). It has also been known that human primary colorectal tumors have
exhibited elevated levels of thioredoxin mRNA compared to normal colonic mucosa
(Berggren M, et al., Anticancer Res 16:3459-3466 (1996)). It has not been known that
thioredoxin protein was present in certain human tumor cells, and it has not been known that
thioredoxin protein played any role in preventing or enhancing tumor cell growth. While thioredoxin itself is known, its use in identifying agents that inhibit cell
growth stimulated by thioredoxin has not been previously shown.
Human thioredoxin reductase has been characterized as a protein (Oblong JE,
et al., Biochem 32:7271-7277 (1993)). In addition, the general properties and the cDNA base
sequence of human thioredoxin reductase is known in the art. However, it has not been
disclosed or suggested in the art that thioredoxin reductase be used as an anti-tumor drug
target.
The myelodysplastic syndromes (MDS) are a heterogeneous class of life
threatening diseases characterized by ineffective hematopoiesis and progressive, reractory
cytopenia (List AF, et al. J Clin Oncol 8:1424-1441 (1990)). Transformation to acute
leukemia may occur in one-third of the patients. The underlying defect is decreased
multilineage progenitor cell growth associated with decreased sensitivity to growth factor
stimulation (Merchav S, et al. Leukemia 5:340-346 (1991) ). Very high doses of recombinant
granulocyte-macrophage colony stimulating factor (GM-CSF) and recombinant human
granulocyte colony stimulating factor (G-CSF) can ameliorate neutropenia but do not
improve red blood cell or platelet function (List AF, et al. J Clin Oncol 8:1424-1441 (1990)).
Although IL-3 displays multilineage progenitor stimulatory effects in normal marrow clinical
trials have shown limited ability to improve hematopoiesis in MDS (List AF, et al., Blood 82
(Suppl. l):377a (1993)). Thus, current treatment for MD is limited by the ability of cytokines
to stimulate hematopoietic progenitor cells and the decreased sensitivity of these cells to
growth factors.
2. SUMMARY OF THE INVENTION The present invention relates to the use of thioredoxin as, mter alia, a cell
growth stimulator, as well as a screen for agents that are useful in reducing or preventing
thioredoxin-associated apoptosis inhibition in tumor cells and agents that are useful in
inhibiting thioredoxin stimulated growth of tumor cells.
A non-limiting embodiment of the invention involves a method of inhibiting
tumor cell growth in a tumor cell that over-expresses thioredoxin comprising contacting said
tumor cell with a cell growth inhibiting effective amount of an inhibitor of thioredoxin
expression. Such agents can include, ter alia, small molecular compounds that complex
with and interfere with the biological action of thioredoxin, preferably those that complex
with active Cys residues, antisense inhibitors of thioredoxin expression, antibodies, or
inhibitors of nucleic acid expression.
A further non- limiting embodiment of the invention involves a method of
reducing inhibition of apoptosis in tumor cells that over-express thioredoxin comprising
contacting said tumor cells with an effective amount of an agent that inhibits thioredoxin.
Such agents can include, inter alia, antibodies to this redoxin, compounds that inhibit the
activity of this redoxin, preferably those that inhibit the activity of active Cys residues in the
protein, cross-linking agents and the like.
A further non-limiting embodiment of the invention involves a method of
identifying an agent that inhibits tumor cell growth in cells that over-express thioredoxin
comprising measuring thioredoxin expression or activity in a first sample of said cells;
contacting a second sample of said cells with an agent to be tested; measuring expression or
activity of thioredoxin in said second sample; comparing expression or activity of thioredoxin
in said first sample and said second sample; whereby a decrease in expression or activity of thioredoxin in said second sample is indicative of an agent that inhibits tumor cell growth.
A further non-limiting embodiment of the invention involves a method of
identifying an agent that reduces inhibition of apoptosis in a tumor cell that over-expresses
thioredoxin comprising measuring thioredoxin expression or activity in a first sample of said
cells; contacting a second sample of said cells with an agent to be tested; measuring
expression or activity of thioredoxin in said second sample; comparing expression or activity
of thioredoxin in said first sample and said second sample; whereby a decrease in expression
or activity of thioredoxin in said second sample is indicative of an agent that reduces
inhibition of apoptosis.
A further non-limiting embodiment of the invention involves a method of
identifying an agent that reduces thioredoxin induced inhibition of apoptosis in a tumor cell
growth.
A further non-limiting embodiment of the invention involves a method of
stimulating cell growth comprising introducing a nucleic acid encoding a human thioredoxin
having Ser at amino acid reside 73 under conditions whereby said nucleic acid is expressed.
A further non-limiting embodiment of the invention involves a composition
comprising an agent that is useful in reducing or eliminating thioredoxin-associated apoptosis
inhibition and an acceptable carrier.
A further non-limiting embodiment of the invention involves a composition
comprising an agent that is useful in inhibiting thioredoxin stimulated cell growth and an
acceptable carrier.
The present invention is based, at least in part, on the discovery that
thioredoxin protein is over-expressed in certain human tumor cells; that thioredoxin stimulates the growth of cancer cells; that thioredoxin inhibits apoptosis; that thioredoxin is
over-expressed in some human primary tumors and is correlated with increased tumor cell
growth and decreased apoptosis; and that agents that inhibit thioredoxin also have anti-tumor
activity.
The present invention involves the new uses of thioredoxin, thioredoxin
reductase, and mutant forms of thioredoxin for use in screening for anti-tumor agents. It has
not been known in the art to use thioredoxin and/or thioredoxin reductase in a screening assay
for anti-thioredoxin and/or anti-thioredoxin reductase agents for use as anti-tumor
compounds.
The present invention further relates to the use of thioredoxin and/or
thioredoxin reductase antibodies for use as anti-tumor agents.
The present invention further relates to the use of anti-sense thioredoxin or
anti-sense thioredoxin reductase compounds for use as anti-tumor agents.
The present invention further relates to the use of thioredoxin nucleic acid
probes and/or thioredoxin antibodies in a diagnostic assay for certain cancers.
The present invention further relates to the use of thioredoxin as a target for
agents to be used in combination with existing and new treatment therapies, such as drugs
and radiation, that reduce or prevent the thioredoxin-induced inhibition of apoptosis in tumor
cells or to increase the sensitivity of tumor cells to these modalities.
In addition, mutant forms of thioredoxin provide proteins with additional
growth stimulating activity.
These and still further objects as shall hereinafter appear are readily fulfilled
by the present invention in an unexpected manner as will be readily discerned from the following detailed description of the preferred embodiments of the invention, especially when
read in conjunction with the accompanying drawings.
In contrast to the present invention, none of the above cited references teach or
suggest the use of thioredoxin protein according to the claimed invention.
3. BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 shows a chart that illustrates the stimulation of human bone marrow
colony formation by Cys73→Ser mutant thioredoxin in accordance with the present
invention, wherein the effects of Cys73→Ser thioredoxin on colony formation are shown by
(o) multilineage progenitors (CFU-GEMM); (•) erythroid progenitors (BFU-E); and (V)
myeloid progenitors (CFU-GM), measured over 10 days as described. (Values represent the
mean of 4 determinations and bars represent S.D.)
Fig. 2 shows a chart that illustrates potentiation of IL-2 induced MCF-7 breast
cancer cell growth by Cys73→Ser mutant thioredoxin in accordance with the present
invention, wherein cells were growth arrested for 48 hr in medium with 0.5% serum (105
cells) then stimulated in the absence of medium with either IL-2 or Cys32→Ser mutant
thioredoxin at the concentrations shown and cell number was measured after 48 hr. (Each
point represents the mean of 3 determinations and bars represent S.E., and the dotted line
shows stimulation by 10% serum.)
Fig. 3 shows a chart that illustrates the inhibition of thioredoxin stimulated
MCF-7 cell growth by receptor antibodies in accordance with the present invention, wherein
cell proliferation was measured as described in Fig. 2; the concentrations of agents used were
Cys73→Ser mutant thioredoxin (Thioredoxin) 1 μM, monoclonal antibodies to FGF receptor, IL-2-receptor and EGF-receptor 4 μg/ml, and EGF 10 nM; and the EGF and EGFR were
added as a negative control. (Values represent the mean of 3 determinations and bars
represent S.E., and the dotted line shows the effect of 10% serum alone.)
Figs. 4A-B illustrate comparative charts showing the effects of thioredoxin
and C32S/C35S cDNA transfection on proliferation of MCF-7 cells. In Fig 4A, 3 x 104 cells
were plated in 3.8 cm2 plastic culture dishes in DMEM with 10% fbs and cell number
measured 7 days later. In Fig. 4B, 104 cells were plated in 2 cm2 wells containing soft
agarose and colonies measuring >60 microns counted 7 days later. (Control, the Neol vector
alone transfected MCF-7 cells; Thioredoxin 9, Thioredoxin 12, and Thioredoxin 20, MCF-7
cells transfected with human Thioredoxin cDNA; Serb 4, Serb 15, and Serb 19, MCF-7 cells
transfected with C32S/C35S cDNA. Values are the mean of 3 determinations and bars are
S.E. **indicates p <0.01 compared to vector-alone transfected cells.)
Figs. 5 to 27 are briefly described in the text accompanying the respective
figure.
4. DETAILED DESCRIPTION OF THE INVENTION
All of the various publications cited in the present specification, including
those publications that are referenced by a numeral corresponding to the citation listed
thereafter, are incorporated by reference in their entireties.
4.0 DEFINITIONS
In order that the invention herein described may be fully understood, the
following definitions are provided: -^H
"Nucleotide" means a monomeric unit of DNA or RNA consisting of a sugar
moiety (pentose), a phosphate, and a nitrogenous heterocychc base. The base is linked to the
sugar moiety via the glycosidic carbon (1' carbon of the pentose) and that combination of
base and sugar is called a "nucleoside". The base characterizes the nucleotide. The four
DNA bases are adenine ("A"), guanine ("G"), cytosine ("C"), and thymine ("T"). In RNA
uracil ( "U") substitutes for T. In double stranded molecules, an A on one strand pairs with
T(U) on the other, and G with C. As is conventional for convenience in the structural
representation of a DNA nucleotide sequence only one strand is shown in which A on one
strand connotes T on its complement and G connotes C. DNA comprises deoxyribose as the
sugar while RNA comprises ribose.
"Amino acids" are shown either by a three letter or one letter abbreviation as
follows:
Abbreviated Designations Amino Acid
A Ala Alanine
C Cys Cysteine
D Asp Aspartic acid
E Glu Glutamic acid
F Phe Phenylalanine
G Gly Glycine
H His Histidine
I He Isoleucine
K Lys Lysine
L Leu Leucine
M Met Methionine
N Asn Asparagine
P Pro Proline
Q Gin Glutamine
R Arg Arginine
S Ser Serine
T Thr Threonine
V Val Valine w Tip Tryptophan Y Tyr Tyrosine
"DNA Sequence" means a linear array of nucleotides connected one to the
other by phosphodiester bonds between the 3' and 5' carbons of adjacent pentoses.
"Codon" means a DNA sequence of three nucleotides (a triplet) which
encodes through mRNA an amino acid, a translation start signal or a translation termination
signal. The code however, is degenerate, with some amino acids being encoded by more than
one triplet codon. For example, the nucleotide triplets TTA, TTG, CTT, CTC, CTA and
CTG all encode the amino acid leucine ("Leu"), TAG, TAA and TGA are translation stop
signals and ATG is a translation start signal.
"Proteins", "peptides" and "poly peptides" are composed of a linear array of
amino acids connected one to the other by peptide bonds between the alpha-amino and
carboxyl groups of adjacent amino acids.
"Genome" means the entire DNA of an organism, cell or a virus. It includes,
inter alia, the structural genes coding for polypeptides, as well as regulatory regions
including operator, promoter, terminator, enhancer and ribosome binding and interaction
sequences.
"Gene" means a DNA sequence which encodes through its template or
messenger RNA ("mRNA") a sequence of amino acids characteristic of a specific
polypeptide.
"cDNA" means a complementary or copy DNA prepared by using mRNA as a
template for synthesizing the first strand of DNA using reverse transcriptase, an appropriate
oligonucleotide primer and a mixture of nucleotides.
"PCR" means a polymerase chain reaction whereby a specific DNA sequence, either genomic or cDNA, can be preferentially amplified by the enzyme Taq polymerase
using synthetic, oligonucleotide sense and antisense primers, (which specify a target
sequence), a mixture of nucleotides and a temperature thermocycling regime which allows
sequential denaturing, annealing and synthesis of the target DNA between the primers.
"Transcription" means the process of producing mRNA from a gene or DNA
sequence.
"Translation" means the process of producing a polypeptide from mRNA.
"Expression" means the process undergone by a gene or DNA sequence to
produce a polypeptide and comprises a combination of transcription and translation.
"Plasmid" or "phagemid" means a nonchromosomal double-stranded DNA
sequence comprising an intact "replicon" such that the plasmid is replicated in a host cell.
When the plasmid is placed within a unicellular organism, the characteristics of that organism
may be changed or transformed as a result of the DNA of the plasmid. For example, a
plasmid carrying the gene for ampicillin resistance (AMPR) transforms a cell previously
sensitive to ampicillin into one which is resistant to it. A cell transformed by a plasmid is
called a "transformant".
"Recombinant DNA Molecule" or "Hybrid DNA" means a molecule
consisting of segments of DNA from different genomes which have been joined end-to-end
outside of living cells and able to be maintained in living cells.
"Apoptosis" is programmed cell death activated by a genetic program to
implement a series of events that cause the death and disposal of a cell. This is in contrast to
cell death occurring by necrosis, usually as a result of injury to the cell.
"Oncogene" is a gene that encodes a protein able to transform cells in culture to induce cancer in animals.
"Over-expression" in the context of determining over-expression of
thioredoxin or recombinant modified thioredoxin is characterized by a two fold increase or
more of the levels of thioredoxin or recombinant modified thioredoxin in a target sample
compared with a control sample.
The following abbreviations may be used throughout the disclosure:
"C32/C35S" = Cys32 → Ser32 / Cys35 → Ser35
"DTT" = dithiothreitol
"FBS" = Fetal bovine serum
"NEM" = N-ethylmaleimide
"Scid" = Severe combined immunodeficient
"Trx" = thioredoxin
4.1 USE OF THIOREDOXIN AS AN ONCOGENE
In a non-limiting embodiment of the present invention, NIH 3T3 cells
transfected with human thioredoxin DNA that has been directed to the nucleus of the cells by
a nuclear localization signal causes malignant transformation of the cells.
In a further non-limiting embodiment, stable transfection of mouse WEHI7.2
lymphoid cells with human thioredoxin DNA has been shown to inhibit apoptosis induced by
a variety of agents including glucocorticoid, N-acetylsphingosine, staurosporine, thapsigargin
and etoposide, which is similar to the pattern of inhibition of apoptosis caused by the anti-
apoptotic oncogene bcl-2 in these cells. The thioredoxin transfected WEHI7.2 cells form
tumors in scid mice that grow more rapidly and show less spontaneous apoptosis than WO 98/24472 ^ΛC PCT7US97/22292
vector-alone or bcl-2 transfected cells, and are resistant to growth inhibition by
glucocorticoid. Therefore, the thioredoxin gene acts as an oncogene according to the
standard definition of an oncogene: a gene that encodes a protein able to transform cells in
culture or to induce cancer in animals (Lodish H, et al., Cancer. In: Lodish H, Baltimore D,
Berk A, Zipursky SL, Matsudaira P, Darnell J, eds. Molecular Cell Biology. New York:
Scientific American Books, pp.1258 (1995)).
In a further non-limiting embodiment, the thioredoxin gene offers an increased
survival advantage as well as a growth advantage to tumors in vivo, unlike the known anti-
apoptosis oncogene bcl-2 which offers only a survival advantage and requires other genetic
changes for tumor growth (McDonnell TJ, et al., Nature 349:254-256 (1991)).
4.2 THIOREDOXIN IS OVER-EXPRESSED IN CERTAIN HUMAN TUMOR CELLS
It has been discovered that thioredoxin DNA is over-expressed in certain
human tumor cells resulting in the over production of thioredoxin. According to a technique
for reproducibly retrieving antigens for immunohistochemical studies from archived paraffin,
human tissue pathology samples were embedded and used it to study thioredoxin protein
levels with a panel of human primary gastric carcinoma tissue samples, it has been found that
thioredoxin is present in dividing normal basal crypt cells. It has been further learned that, as
the cells differentiate and move down the villi to eventually be shed into the gastric lumen,
thioredoxin levels decrease. By stably transfecting murine NIH 3T3 fibroblast-like cells and
human MCF-7 breast cancer cells with cDNA for human wild-type thioredoxin or with
cDNA for a redox-inactive mutant thioredoxin, it has been found that transfection with
thioredoxin increases the density to which the NIH 3T3 cells grow in culture and stimulates anchorage-independent colony formation by MCF-7 breast cancer cells. The redox-inactive
mutant thioredoxin acted in a dominant-negative manner, so that transfected MCF-7 cells
showed inhibited growth and a reversal of the transformed phenotype, assessed by growth in
vitro and in vivo.
It has been shown that stable transfection of mouse NIH 3T3 normal
embryonic cells with human thioredoxin cDNA increases their growth rate and cell saturation
density in culture (normal NIH 3T3 cells are highly contact inhibited) which is in vitro
evidence of transformation. It has also been shown that transfection of MCF-7 human breast
cancer and HT-29 human colon carcinoma cells with human thioredoxin cDNA increases
their colony formation in soft agarose and tumor growth by HT-29 colon cancer cells when
the cells are inoculated into immunodeficient (scid) mice.
Trx was originally studied for its ability to act as a cofactor for ribonucleotide reductase. the first unique step in DNA synthesis (1). Human Trx was subsequently shown to modulate the DNA binding of several transcription factors that regulate cell proliferation, including nuclear factor KB ( 2). the glucocorticoid receptor (3), and. indirectly through the nuclear redox protein ReM . activator protein- 1 (Fos Jun heterodimer. Ref. 4). Cloning and sequencing of human Trx have shown that it has a predicted amino acid sequence (5. 61 identical to that of a growth factor secreted by virus-transformed leukemic cell lines, termed adult T-cell leukemia-derived factor (7. 8). Human Trx. but not bacterial Trx. added to the culture medium stimulates the growth of a variety of normal and cancer cell lines (8-10). The addec Trx is not taken up by cells ( 1 1 ) and appears to stimulate cell growtr
by enhancing the action of other growth factors ( 1 1. 12). The redox activity of Trx is required for growth stimulation, and redox-inactive mutant Trxs do not stimulate cell growth ( 10).
Trx mRNA levels are increased compared with corresponding normal tissue in almost half human pnmarγ lung (5 ) and colon tumors examined ( 13). Trx protein has been reported to be increased in human cervical neopiastic squamous epithelial cells ( 14) and hepato- cellular carcinoma ( 15 ). Trx is excreted from cells ( 16- 18) using a leaderless secretory pathway ( 17). and we have suggested that Trx might be a growth factor for some human cancers ( 1 1 ). However, it remains to be unequi ocally demonstrated that endogenously produced Trx can affect cell proliferation. The role Trx plays in the transformed phenotype of cancer cells also is not known.
To provide some answers to these questions, we have stably transfected murine NIH 3T3 fibroblast-like cells and human MCF-7 breast cancer cells with cDNA for human wild-type Trx or with cDNA for a redox-inactive mutant Trx. We have found that transfection with Trx increases the density to which the NIH 3T3 cells grow m culture and stimulates anchorage-independent colony formation by MCF-7 breast cancer cells. The redox-inactive mutant Trx acted in a dominant- negative manner, so that transfected MCF-7 cells showed inhibited growth and a reversal of the transformed phenotype. assessed by growth in vitro and in vivo.
MATERIALS AND METHODS
Human wild-type Trx cDNA and cDNA for C32/C35S redox-inactive Trx. in which both active-site cysteine residues are replaced by seπne ( 10). were prepared as described previously (5. 10). The cDNAs were cloned into either the Kpn\ or Sort sites of the pRXneo mammalian transfection vector, under consutuuve control of the Rous sarcoma virus promotor ( 19). or into the Sou site of the pOC304neo mammalian transfecuon vector, in which constitutive expression is driven by the cytomegalotirus and SV40 promoters 120). Correct onentaαon of the cONAs in (he vectors was confirmed by restriction digestion The pRXneo and pDC30 neo vectors were obtained from Dr. Roger Miesfeld (University of Arizona. Tucson. AZ).
Human MCF-7 breast cancer cells and munne NIH 3T3 cells were obtained from the American Tissue Type Collection iRockville. MD). maintained in DMEM containing I0<% FBS under 6 c CO: at 37*C. and passaged before confluence. NIH 3T3 cells were transfected wuh Trx:pRXneo. Trx. pDC30 neo. C32S/C35S.pDC304neo. or pRXneo alone. MCF-7 cells were transfected with Trx:pDC304neo. C32S/C35S:pDC304neo. or pDC3Wneo alone. Transfection used liposomes of .V-(l^.3-dioleolyl)propyl|-ιV V.(V.-tnm- ethyiammoniumrnethylsulfate i Boehnnger Mannheim. Indianapolis. IN) according to the manufacturer' s instructions. Cells were selected by growing for 4 weeks in DMEM with lOt FBS and 400 μg/ml G4I8 sulfate ( Life Technologies. Guthersburg. MD). Cell colonies were isolated by ir simzatiυn onto small squares of sienle filter paper and expanded by growing in (he same medium. All studies were conducted on clonal cell lines between passages 3 and 10.
Northern hybridization analysis of Trx and C32S/C35S mRNA used a full-length [α-,:P]dCTP-labeled human Trx cDNA probe as described prev iously (5). and the blo(s *ere quantified using a Phosphorlmager ( Molecular Dvnamtcs. Sunn vale. C λ i Transfected Trx and C32S/C35S mRNA lev els
»Λ >
Figure imgf000021_0001
Time (days)
Fig. 2. Effects of transfection with Trx or C32S C35S cDNA on the growth of NIH 3T3 cells. Cells were plated in platuc dishes at a density of 2 x 10* cells cnv in DMEM with lOf' FBS and cell number measured daily. •. NeoC vector atone-nnsfectcd cells. The apparent decrease in the number of ceils after day 3 is due to detachment of ceils from the plastic surface: T. ▼. and <i. Thioo. ThrøAD. and Trao9 ceils transfected with Trx cONA: D and ■. NIHBH and NTHBF cells transfected with C32S/C3SS cONA. Values are the mean of three determinations: ban. SE. •. P < 0.03 compared with vector alone- transfected cells, shown for days 3 and only. The study is ryptcal of three repeat experiments. WO 98/24472 < JD PCT/US97/22292 I ΠN tormαtuin bs MCF-" vc'l- was measured in Λit asai *ι v DMEM and 1(11 FBS cr " Jass. XS described prcsinusls ill i Gruwth ol cell, on plastic surtaxes *ιιn DMEM -nil lO FBS was measured Jails lor NIH-3T3 cells user 4 da»s and tor MCF-7 cell
Figure imgf000022_0001
7 dass. as described previously (22). All cell jrow th studies were conducted in the absence i G41X sulfate.
Tumor tormatiυn b\ transiected NIH 3T3 cells as studied bs the s.c injection of 10' transected cells in 0 l ml of steπle 09'r NαCI into the backs of groups ol tour male SCID mice or itx nude mice. Tumor fom tiun bv MCF-7 cells was studied b> injecting : < 10' cells in 0.1 ml ol stenle Q.9<> NaO and 0.1 ml of M tngel tBecton Dickinson. Bedford. MAi s.c. into the backs of groups of four temαle SCID mice that had been implanted s.c. 2 da>s previously with 21-dav release pellets ot 0.22 mg of 17-B-estradιol i Innovative Research. Sarasota. FLi. The 17-0-esιradιol pellet was replaced at 21 days Tumor volume was measured with calipers ι23) twice a week for 40 days, λt the end of the study, the animals were lulled, and tumors and other organs were taken for histological analysis.
Stausucal analysis was by Student s nonpaired r test unless otherwise stated. Tumor growth rates in SCID mice were linearized using the cube root of the tumor volume by day for each mouse, and ANOVA was performed using Dunnett's test to determine significant differences from the vector alone- transfected (control) cell line.
RESULTS
Tπt and C32S/C35S Transfection of NIH 3T3 Cells. Transfection of mouse NIH 3T3 cells with Trx:pRXneo yielded 6 clones stably expressing Trx mRNA. transfection with Trx:pDC30*neo yielded 4 clones, and transfection with C32S/C35S:pDC30 neo yielded 12 clones. The levels of transfected mRNA in some of the ciones is shown in Fig. I. The human Trx and C32S/C35S mRNAs were larger than the endogenous mouse Trx mRNA. probably because the transfected Trx mRNAs also contain portions of the vector promoter region, the 5' leader sequence, or the polyadenylate tail. The level of transfected Trx mRNA expression was relatively low. being only 0.2- 1.4-fold the endogenous mouse Trx mRNA. Western blomng showed no significant increase in the level of Trx protein tn the cells compared with wild-type or vector aione-transfected cells (results not shown).
The Trx-transfecied NIH 3T3 cells grew at the same rate on a plastic surface, but reached saturation densiues up to twice that of the
Figure imgf000022_0002
MCF- NIH 3T3
Fif 3 Sumulauon of the proliferation of MCF-7 human breast cancer tell* and mouse NIH 3T3 cells bv human Tπ Cells were trow in arrested in DMEM containing 0 ft FBS for 48 h w thai there were 0 α » 10' .ells, at which lime the medium »a« replaced with fresh DMEM with ιβι or »ιuιouι < _> I μM human TiΛ Cell numbers » ere measured JS ft later Values are the mean oi three Oeierrnir iions. ban. SE -2^
Fiy α AuuwaUinimm r»ι»ιnιi s,ιnnemαfuι>- ^ 4 JCV α ^jT Cj *5 o1 t,* t,4, Jr
«ι> of SI V-T breast cancer . lls »ι_hlv iransMctetl wuh cDNA M* wiki-iypc r* IΛI mi miMHft-.- lue mutani Tπ l iSV sSi nsbnϋi/nt SSILΓ.. tull'lcncth '-P.faitietcu numan Tn cDNA ι«> B VCFneu are cells iranslecieu »uh pDC MM secut mine. The IXMI.HU u l » cnooeenυui Tn mRNA
Figure imgf000023_0001
and the r«ø 4MJH<4 it live transiected Trx mRNAs V lues heitiw are the ϋt irαnxtecied Tn mRNA lo enuotnous mRNA ueterminc. n> •I pnotoflonnueer analysis
Figure imgf000023_0002
- 0.5 0.2 0.4 0.8 2.1 1.7 0.1 0.2 1.1 -
Figure imgf000023_0003
2f
vector aione-transfected NIH 3T3 cells (Fig. 2). The vector aione- transfected cells had the same growth characieπsucs of *ιld-rype NIH 3T3 cells. NIH 3T3 cells transfected with the redox-inacuve C32S/ C35S Trx grew more slowly and reached a lower saturation density on a plastic surface than the vector alone-iransfected cells Neither the vector aione-transfected NIH 3T3 cells nor the Trx or C32S/C35S Trx transfected cells formed colonies in soft agarose (results not shown). The ability of transfected NIH 3T3 cells to form tumors when inoculated into tmmunodeficiem mice is used to identify neopiasαc transforming genes (24). When the Trx-transfected NIH 3T3 cells Thioό or ThioAD were injected i.m. into SCID or nude mice, there was no tumor formation over 40 days (results not shov.nl. Thus. Trx expression, at least at the level obtained in this study, was not. by itself, transforming.
Trx and C32/C35S Transfection of MCF-7 Breast Cancer Cells. Human solid cancer cells generally show a greater proliferation response to added Trx than do mouse fibroblasts ( 10. 1 1). This is shown for MCF-7 human breast cancer ceils compared with NIH 3T3 cells in Fig. 3. Thus, we also studied the effects of Trx transfection using MCF-7 breast cancer cells. Transfection of MCF-7 cells with Trx: pDC304neo yielded 31 clones that stably overexpressed Trx mRNA. and transfection with C32S/C35S.pDC304neo yielded 45 clones stably expressing C32S/C35S mRNA. Expression of transfected mRNAs by some of the clones is shown in Fig. 4 As seen previously with the mouse cells, the transfected human Trx mRNAs tn MCF-7 cells were larger than endogenous human Trx mRNA. The level of Trx mRNA expression was up to 0 8-fold and C32S/C35S mRN up to 2 1 -told the endogenous Trx mRNA levels. Light microscopy showed no difference m the appearance of vector aione-transfected and Trx- transfected MCF-7 cells growing on glass overslips ( Fig ι. and both were similar to wild-type MCF-7 cells In contrast. C32S/C35S- transfected MCF-7 cells appeared more rounded and had α reduced cytoplasm-to-nucleus ratio
Quantitative Western immunoblotting showed no significant increase in the level of Trx protein in the transfected MCF-7 cells compared with vector alone-iransfected MCF-7 cells, except for one clone (Table 1 ). There was. however, a significant 60^ increase in the secretion of Trx into the medium by three of the clones compared with the vector aione-transfected MCF-7 cells. The three other clones showed a 20-50% increase of Trx in the medium, but this was not statistically significant. Thus, it appears that most of the extra Trx and C32S/C35S produced by the transfected cells is secreted into the medium.
All of the transfected MCF-7 cells showed linear growth characteristics on plastic surfaces over 7 days. The Trx-transfected cells grew at the same rate as the vector aione-transfected MCF-7 cells (Fig. 6). However, when grown in the absence of. or with 0.5<t FBS for 2 days, the Trx-transfected cells grew at twice the rate of the vector aione-transfected cells (results not shown). The C32S/C35S-trans- fected cells grew at a significantly slower rate that was 56-78^ of the Fit 5 lifht microicops ol Trx wdCj'.S/C.'JS cDN'A-tniuiecied MCF-7 breau cancer cell!. The cells were crown to 731 conlluet e on |la cos- eπlrpv f»ed wiw mctnanol. iiaineα wita i o- anowilu-ivpc dve (Diff-Quick. Baxteri. and or served wiin a x 100 oil immersion oto cαse A. pOCJOα vector alone-mωlecicd MCF-7 cells: B Trx X Tπ-mmfecud MCF-7 cells: C. Serb α C32SVC33S-ιransleeted MCF-7 cells: 0. Serb 19 O-S OJS-iπuislecied MCF-7 cells
Figure imgf000025_0001
Figure imgf000025_0002
^
cted
Figure imgf000026_0001
Figure imgf000026_0002
vector alone -transtecteα cells I Fig 'olonv formation was signif?... ; icanUY increased between '3- and' Jld ' for the Trx-transfecte ' " MCF-7 cells compared with the vector aione-transfected cells, and significantly decreased up to 73 - for the C32S C3 JS-eransiected cell- when the cells were grown tn sou agarose. The vector alone-iransfected cells showed growth characteristics identical to those of wild- type MCF-7 cells under all conditions i results not shown).
Tumor Formation by Trx- and C3-cVC35S>trusιeeted MCF-7 CeUs. The vector alone-πansfected MCF-7 cells injected mto SOD mice formed rumors that grew at the same rate as aoαtraαsfected MCF-7 cells we have seen in other studies. Trx-transfected MCF-7 cells formed rumors in SOD mice, although they grew at a significantly slower rate than rumors formed by vector aione-transfected •ells: 37% for Trx 12 cells and 8% for Trx 20 cells (both P < 0.0$ by least squares regression analysis: Fig. 7). Tumor formation by the C32S C3.S-transrecιed MCF-7 ceils was almost completely suppressed. Tissues from the injection site and other organs were taken for histologies! examination at the end of the study. The animals injected with vector alone or Trx-transfected cells showed large solid tumors. The animals injected with Q-2 C33S-transfeeted cells showed small microscopic rumor cell deposits. There was no evidence of tumor metastasis to other organs in any of the animals. Northern analysis of the tumor taken from animals injected with Trx-transfected cells showed the presence of transfected Tn mRNA as determined by its large size < results not shown i.
Table I Tπt irrrtt m mκ≠κn4 MCF-7 trrtm t rtr ctill M -m- Tn or nαo-Hucmt CJ25/C3 JS iSerβMrmlmrd MCF.7 bra caαr Mis 110*1 m nn ιιι--»ιl a i ml of OME.M for β nr ma Tr« m ihe animal ml tron or a iw meiMMi ii-wrirt bv qu-Wiiauve Wes-tra lawMaM-outf. Vab-n av s -woe of dm fcfloan-i cxpm-rt rcUMivc to vcnor aMne^r-mlonoe ceti-.
Claof Ceir . *-.
MCF-7 vector 1.0 - 0.0 1.0 =0.0
Tn * 0» ϊ 0.0 ι.ι :0
Tn 12 i : s o i- 1.2 ϊ 0J
Figure imgf000027_0001
DISCUSSION
Trx regulates the redox state and activity of a rtumoer of intracellular proteins that control cell irowth. including πoonucleoude reductase ( I) and the DNA binding of several transcnotion factors (2— iι Recombinant human Trx added to normal and cancer ceils in culture stimulates their proliferation 11 1 ). However, tt has not been demonstrated that endogenously produced Trx can stimulate cell proliferation. Furthermore, the role Trx may play tn malignant transformation of cells is not known. The present study was undertaken to address some of these questions. NIH 3T3 ceils transfected with Tπt showed an increased cell saturation density when grown as a monolayer on plastic surfaces. Loss of contact inhibition is a feature of transformed cells (24). and the cell saturation density of the Trx-transfected NIH 3T3 cells was similar to that seen with other transformed, weakly tumongenic mouse 3T3 ceil lines (231. However, the Trx-transfected NIH 3T3 cells did not form rumors when inoculated into tmmunod- eticient mice. The Trx-transfected MCF-7 ceils did not show increased growth on plastic surfaces in normal FBS. but exhibited significantly increased anchorage-independent growth measured by colony formation in soft agarose. It is surprising that when the Trx-transfected MCF-7 cells were grown as xeno rafts in SCID mice, they exhibited decreased growth rate compared with vector aione- transfected MCF-7 cells. This may be because human Trx can stimulate the immune system of mice' so that Tπt secreted by the transfected MCF-7 celts might promote some immune rejection, even by the SCID mice, which, although deficient in mature B and T lymphocytes, have natural killer-, myeioid-. and antigen-presenting ceils (26ι. Both NIH 3T3 and MCF-7 breast cancer cells transfected with the C32S C23S Trx showed slowed growth rates on a plastic surface. In addition, colony formation by MCF-7 breast cancer cells in soft agarose was considerably decreased. When injected into SCID mice, the C32S/C33S-ιransfected MCF-7 cells formed only microscopic tumors. C32S C33S is a rcdoi-inactive mutant Trx that acts as a competitive inhibitor of Tπ reductase ( 10). Our X-rav crystallographic studies have identified a highly conserved 1. mmo acid hydrophobic surface on mammalian, out not bacterial. Tr\». whicn t-έ
Figure imgf000028_0001
Time (days)
Fig 7 Grouth of Trx and C32S/C35S-0aιιsfected MCF-7 breast cancer cells in SCID mice Female SCID mice implanted t c. 2 davs prcMousiv with a 21-dav release pellet ol 0 23 mg of 17-S-esu-nlιol were injected s c with . 10* iransfected MCF-7 cells in 0 1 ml of 09"> NaCl and 0 I ml of Vlatngei. MCFneo. pOC30α vector alone-iranstected MCF-7 cells ▼ Trx I I Trx-transteciexl cells at Trx 20. Ihraredoxin-transtected cells. ▼ Serb C32S/05S-ιr_nsfecιed cells. . Serb 13 C32S/C35S-transιected cei There »ere four mice per group Tumor growth was measured twice a week lor 40 davs The 17-0-estradιol pellet was replaced ol 21 davs Values are mean Ban. SE
stabilizes the Cys73 -mediated disulfide-bortded dimer (27V The physiological function of this Cys73-Cys'3 linked Tπt dimer is not known The surface structure of C31S/C35S is very similar to that of Trx (27) so that C32S/C3SS is likely to participate in the formauon of a heterodimer with Trx and thus might lower Trx monomer concentrations or affect the biological activity of the dimer. Unlike wild-type Trx. C32S/C35S does not stimulate cell growth when added to the culture medium. C32S/C35S might also act as a competitive inhibitor to the normal redox-acnve substrates of Trx Whatever the mechanism, it appears that C32S/C35S acts in a dominant-negative manner to inhibit the effects of endogenous Tn and. in so doing, inhibits cell growth and reverses the transformed phenotype of MCF-7 breast cancer cells.
Most of the added Trc or C32S/C35S protein that is produced by the transfected cells appears to be secreted into the medium Whether the transfected Trx is produced in a different compartment to endogenous Trx. allowing it to be secreted, or whether a constant proportion of Trx is secreted is not known. Trx is known to be secreted from cells by a Ieaderless secretory pathway ( 17). The concentrations of Trx found in the medium, up to 10 nvt after 6 h. are lower than those required to directly stimulate cell proliferation ( I I ). However, we have recently found that Trx at nanomoiar concentrations will potentiate the growth effects of cytokines such as ιnterieukιn-2 and basic fibroblast growth factor4. It remains to be established whether the extra Trx is producing its effects on cell proliferation through an intracellular or an extracellular action. t should be noted that Trx binds to the urface of cells ( 1 1. 28) so that secreted Trx could have a local effect at the outer cell surface although concentrations tn the medium are low
The levels of transfected Tn mRNA in cells were not high, only up to 1-fold endogenous Trx mRNA levels, and were independent of the mammalian transfection vector used Typically. mRNA levels resulting from transfection using such vectors are 10— 50-fold or higher (29). It may be that high levels of Trx gene expression are toxic to cells. We have found only a low expression of the human Tn gene in ^
transgeπic mice.4 In .oniraM, some human tumors show very hi-j levels υi Tn mRN λ compared with the normal tissue: mure than 1 1 -told in human pπmurv lung tumors i 30) and even higher in human pπmars ol n tumors ι | .» ι It is not known why higher Trx RN λ levels could not be obtained in transected cells. «.ere unable to obtain transformation oι NIH 3T3 cells with Tn. It remain , to be demonstrated whether ihe much higher levels of Trx expression seen in some human tumors might be transforming.
The observation that α redox-inactive dominant-negative Trx reverses the transformed phenotype of MCF-7 cells suggests that drugs that inhibit the redox activity of Tn might offer a novel approach to treating some forms of human cancer. Alternatively, inhibiting the enzyme responsible for the reduction of Trx. the lavoprotein Trx reductase. might also lead to a selective inhibition of cancer cell growth. We have shown previously that some antitumor quinones. including doxorubicin and diaziquone. are mechanism-based (suicide substrate) inhibitors of Tn reductase both in the purified form and in intact cells (31). However, the antitumor quinones have many other effects that could contribute to their antitumor activity (32). On the basis of our transfection studies, it would be of great interest to see what effect selective inhibitors of Tn or its reductase have on cancer cell growth and transformation.
In summary, our results have shown that stable transfection of nontransformed mouse fibrobtast-like cells and human breast cancer cells with human Tn leads to low levels of overexpression and increased cell saturation densities but no transformation, measured by tumor formation of NIH 3T3 cells in tmmunodeficient mice. Stable transfection with redox-inactive mutant Trx results in a dominant- negative effect with inhibition of mouse cell and human breast cancer cell growth and reversion of the transformed phenotype of human breast cancer cells, measured by their ability to form colonies .n soft agarose and to form tumors in mice. The Trx produced appears to be secreted mostly from cells, and whether the Trx is having an intracellular or extracellular action remains to be determined. REFERENCES
Figure imgf000030_0001
8 wS 'Vag.va. Y . **».... A.. MH**. M . Maeda. M . Yodoi. J and T ruTrTi. T r A ΛdβuUlltt |-ceeHil l lecun-»ecmπιiιa.--d«e.n..v.e..d. fac.or,th,ored —o,,n p ■ ro uce _d_ I *b-v, b.o.,,th. h jurnmaan.
T-lvrnphotroptc virus type I and Epstem-Barr ™^"f^ J^™^ an au.ocπne gπ.»<h '*<<» »""f'ιed ^ ""erieukjn-1 »" ""CTleul"n- *»
leukemia and hiirmiluiiiHis κι ihnired in imnlsemrni
Figure imgf000030_0002
2≤>
1 λllcs M C . and Licrxrr M \l Measurement m human 'umnr .cil r"»ιn n umirnjicr as»ι>ιed iiume analssis Br I Canter •*_■ .'' •
Figure imgf000031_0001
2 Po is. G . Lee See. K. Sanume. K S . Meldcr. D C . and Ht-Jnetl. t M ι uιnnn eimines as -uhstrates lor quinofle retiucta.se I NADI P 'H iquiniine .sweptoπ ι\ι doretlucto.se I and live effect ol Jieumaml on iheir cytototicits Biochem Pharmacol in. 2473-2479. I K7 .3 Geran. R. I.. Grecnberg. N H.. Mjcdoiuld. M. M.. Schumacher. A M . and hιs.«t B J. Protocols lor screening chemical agents and natural pπiduets against jnin l tumors and other biological systems. Cancer C emoiher. Rep.. 3 1 - 103. 1972 24 Pilot. H. C. Fundamentals of θncolιv.v p. 149 New York. Marcel Dekker. Ins
1981 23 Schlager. J. J . Hoerl. B J.. Riebow. J . Scott. D P . Gasdaska. P . Scott. R E.. and Powis. G Increased NADPH.tquinonc-acceptori oxi oreductase i DT-Jtaphurαsc ι activity is associated with density-dependent growth inhibition .it normal hut nm transformed cells. Cancer Res.. 53: 1338-1342. 1993 26. Shulu. L. D Immunυlogieat mutants ot ihe mouse. Am. J λnai.. 191 )<)?-.' I 1
1991 27 Weichsel. A.. Gasdaska. J. R.. Powis. G . and Monitor!. W R. Crystal structures m reduced, oxidized, and mutated human liiiorcdoxins: evidence for a regulators hυ- modimer. Structure, in press. i996. 23. Ifversen. P.. Zhang. X. M.. Ohlin. . Zeuthen. J.. and Borrebaeck. C. A. Effect ol cell-denved growth factors and cytokines on the clonal outgrowth of EBV-miccted B cells and established lymphoblastotd cell lines. Hum. Antib. Hybrid.. 4 1 I S- 1 3 1993.
29 Powis. G.. Gasdaska. P Y . Gαllegos. A.. Shernll. K.. and Goodman. D Os cr expression of DT-diaphorase in transfected NIH 3T3 cells Joes not lead to increased αnticancer quiπoπe drug sensiiisuy a questionable role for ihe cn.s rnc as a target for bioreductisely activated αnncαncer drugs Anticαnccr Res . ' ;
1 141-1 146. 1993.
30 Gasdaska. P.. Oblong. J. E.. and Powis. G. Cloning and expression ot ihioredύxin < T ι cDNA from human colon cancer cells Identity with ADF and lesels of T mRN a. ,n human rumors. Proc. Am. Assoc. Cancer Res.. 34- 62. 1 93
31. Mau. B-L.. and Powis. G. Inhibition of cellular t ioredotin reductase bs duiiquune and doxorubtcin. Biochem. Pharmacol.. 4! 1621 -1626. 1992.
32. Powis. G. Metabolism and reactions ol quinoid amicancer agents Pharmacol Thcr 31: 157-162. 1987.
In the majority of the subjects tested (8/10), human primary gastric carcinomas
thioredoxin was over-expressed in tumor cells compared to normal mucosa, and in all cases
the over-expression was found only in the cancer cells and not in stromal cells or infiltrating
lymphocytes. Levels of thioredoxin significantly higher than in normal dividing cells, were
found in 5/8 of the over-expressing carcinomas. To relate thioredoxin over-expression to cell
proliferation and apoptosis, nuclear proliferation antigen was detected by Ki67 antibody and
apoptosis by the in situ terminal deoxynucleotidyl transferase (TUNEL) assay were measured
in the same tissue samples. (See Table 1). Thioredoxin expression was significantly and
highly positively correlated with nuclear proliferation antigen (p<0.01) a marker of
aggressive tumor growth and highly negatively correlated with apoptosis (p<0.001) a form
programmed cell death that is presumed to limit tumor growth. Thus, thioredoxin is over-
expressed at the mRNA and protein level in a number of human primary tumors. Further, the
expression of thioredoxin protein is directly associated with highly proliferative tumors.
' λ
Table 1. Staining of Thioredoxin in Human Gastric Cancers:
Comparison with Cell Proliferation and Apoptosis
Staining scored as absent (0) or weak (+) to intense (++++)
Figure imgf000033_0001
* = = gastric pits; NE = non evaluable
MCF-7 human breast cancer cells were transfected with cDNA for thioredoxin
or with a catalytic site redox-inactive mutant thioredoxin, C32S/C35S. using two constitutive
eukaryotic expression vectors (pRXneo and pDC304neo) and a number of clones were
selected for each. The level of transfected thioredoxin and C32S/C35S thioredoxin mRNA
was up to 2-fold the endogenous message. Both types of transfected cells showed increased
thioredoxin protein production, measured by quantitative Western blotting, up to 100% that
of mock-transfected cells. There was little difference in the growth of the transfected cells formed up to
4-fold more colonies when grown in soft agarose and the C32S/C35S transfected cells formed
up to 80% fewer colonies as illustrated in Figs. 4A-B.
When these cells were injected into scid mice the thioredoxin transfected cells
formed tumors while the C32S/C35S transfected cells did not form tumors, as illustrated in
Fig. 5. This was confirmed by histology. Thus, a dominant-negative redox inactive
thioredoxin can reverse the transformed phenotype and inhibits tumor growth in vivo
providing molecular biology evidencing that thioredoxin is a novel target for anti-cancer drug
development.
4.3 USE OF THIOREDOXIN AS AN ANTI-TUMOR DRUG TARGET
Although thioredoxin is a known protein, it has not been disclosed or
suggested that thioredoxin be used as a screen for anti-tumor agents. It has now been shown
that stable transfection of the MCF-7 breast cancer cells with a redox-inactive mutant
thioredoxin causes inhibition of anchorage-independent growth of the cells in soft agarose
and causes complete inhibition of tumor formation in vivo. The redox-inactive mutant is
formed from thioredoxin where the catalytic site cysteine residues are replaced with serine.
Further, it was shown that the mutant thioredoxin did not inhibit monolayer growth of the
cells, i.e., does not inhibit normal cell growth, while it causes inhibition of anchorage-
independent growth of the cells in soft agarose, i.e., does inhibit an in vitro characteristic of
tumor cell growth. This is the activity that would be expected from drugs that inhibit
thioredoxin. 4.3.1 EXAMPLES OF AGENTS THAT INHIBIT THIOREDOXIN
Agents that inhibit thioredoxin have been identified in accordance with the
present invention, such agents may be antibodies, drugs or antisense. A series of
unsymmetrical 2-imidazolyl disulfides were investigated as inhibitors of the thioredoxin
system and as potential anti-tumor agents. Although these agents were originally identified
as competitive inhibitors of thioredoxin reductase (Oblong JE, et al., Cancer Chemother
Pharmacol 34:434-438 (1994)) but it has now been shown that they also to bind irreversibly
to Cys73 of thioredoxin and to block its reduction by thioredoxin reductase. A number of
these disulfide compounds have been studied and have demonstrated anti-tumor activity
against human tumor xenografts in scid mice with up to 90% inhibition of MCF-7 breast
cancer and HL-60 promyelocytic leukemia growth. It has now been demonstrated that the
imidazolyl disulfides inhibit thioredoxin-dependent cell growth (Oblong JE, et al., Cancer
Chemother Pharmacol 34:434-438 (1994)) and that their growth inhibitory activity in the
National Cancer Institute 60 human tumor cell line panel correlates with levels of thioredoxin
mRNA in these cell lines (Berggren M, et al, Anticancer Res 16:3459-3466 (1996)). A
COMPARE correlative analysis of the activity of the lead disulfide compounds in the NCI
cell line panel with over 50,000 compounds already tested for cell growth inhibition by the
NCI was conducted in order to identify compounds with a similar pattern of growth
inhibitory activity. Some of the compounds identified in this way were inhibitor of
thioredoxin reductase and some were inhibitors of thioredoxin. 4.4 USE OF THIOREDOXIN REDUCTASE AS A TARGET FOR INDUCING ANTI-
PROLIFERATION
Although the general properties of human thioredoxin reductase as a protein
and the cDNA base sequence of human thioredoxin reductase has been known in the art, it
has now been discovered that thioredoxin reductase is useful as an anti-cancer drag target. It
has now been shown above that redox activity is necessary for the growth stimulating activity
of thioredoxin. Since thioredoxin reductase is the only known way for thioredoxin to be
reduced biologically it is an obvious extension of the above observations that thioredoxin
reductase could also be a target for the development of anti-cancer drugs.
4.5 USE OF RECOMBINANT MODIFIED THIOREDOXIN FOR STIMULATING
CELL GROWTH
It has been discovered that human thioredoxin, and specifically recombinant
modified thioredoxin (mutated thioredoxin), does not undergo spontaneous oxidation and/or
dimer formation, or protected against breakdown by blood and tissues, may have therapeutic
utility in situations where stimulation of cell growth is preferred or required. In a non-
limiting embodiment of the present invention, such new uses include, and are not limited to,
the beneficial use of thioredoxin and/or recombinant modified thioredoxin in stimulating cell
proliferation in individuals (1) with myelodysplastic syndrome; (2) in need of bone marrow
transplantation; (3) with post-chemotherapy to stimulate bone marrow growth; (4) in need of
stimulation of the immune system; (5) in need of stimulation of would healing; (6) such as
transgenic animals in need of stimulation of body growth; (7) in need of simulation of the
responses to sytokines and growth factors for growth stimulation effects; and (8) in gene 33 therapy techniques.
The underlying defect in myelodysplastic syndrome is decreased multilineage
progenitor cell growth associated with decreased sensitivity to growth factor stimulation.
Thioredoxin acts to increase the sensitivity of cells to growth factors and stimulates
multilineage progenitor cells which provides a beneficial utility in individuals with MDS.
In individuals in need of bone marrow transplantation, it would be of great
utility to promote the growth of transplanted cells. Thioredoxin may be used to protect
hematopoietic progenitor cells and to expand cells ex vivo for bone marrow cell growth. This
would rely on a selective effect for bone marrow since tumor cells might also be stimulated
by the thioredoxin.
It would provide a great benefit to individuals subject to chemotherapy
treatment to selectively stimulate bone marrow cell growth post-chemotherapy.
It has been found that Cys73→Ser mutant thioredoxin will stimulate the
proliferation of human immune cells in culture, which can provide a great benefit to
individuals in need of stimulation of immune system cells.
Wild-type and Cys73→Ser mutant thioredoxin also stimulates the growth of
fibroblasts, which are important components of wound healing process. There would be a
great advantage of using thioredoxins to stimulate wound healing, for example after surgery.
It has been found that wild-type thioredoxin expressed as a transgene in mice
may be lethal or is only weakly expressed. Therefore, it is possible that construction of
transgenes with wild-type or mutant forms of thioredoxin, with or without tissue specific
and/or inducible promotors, could be used to stimulate the development of the animal or the
growth of selected organs. It has been found that thioredoxin can potentiate the response of cells in
culture to +growth factors and cytokines such as IL-2 and fibroblast growth factor (FGF).
Combinations of thioredoxin with other growth factors or cytokines therefore increases the
therapeutic usefulness of these growth factors where increased cell proliferation is the desired
therapeutic effect.
Introduction of the thioredoxin or mutant thioredoxin genes into human cells
provides a mechanism of improving the therapeutic usefulness of other cytokines or growth
factors given directly or themselves as gene therapy, for example IL-2.
4.5.1 EXAMPLES OF STIMULATION OF CELL GROWTH USING THIOREDOXIN
PROTEIN
Although thioredoxin mRNA has been found to be over expressed by some
human tumor cells, it has been discovered that thioredoxin, specifically recombinant
modified thioredoxin also stimulates cell growth.
A novel mechanism or over-expression and secretion from the cells by a
leaderless secretory pathway has important consequences for potential therapeutic uses of
thioredoxin as explained below. It has been discovered that human recombinant thioredoxin
undergoes spontaneous oxidation in air to give a form that will not stimulate cell growth.
This spontaneous oxidation appears to involves Cys73 since a mutant thioredoxin where this
residue has been converted to serine (Cys73→Ser thioredoxin) does not undergo this loss of
activity. X-ray crystallography studies of wild-type and C73S thioredoxin show that
thioredoxin has a highly conserved hydrophobic dimer forming surface and that Cys73
stabilizes homodimer formation through a Cys73-Cys73 disulfide bond (Weichsel A, et al. Structure 4:735-751 (1996)). The active site Cys residues become relatively inaccessible in
the thioredoxin homodimer so that it is a very poor substrate for thioredoxin reductase. The
thioredoxin homodimer does not stimulate cell proliferation. The half life of recombinant
human thioredoxin in phosphate buffered 0.9% NaCI at -20°C is 6-8 days. Thus, the wild
type thioredoxin is not a good protein for therapeutic use because of its tendency to oxidize
and lose biological activity.
It has been found that wild-type thioredoxin loses its ability to stimulate cell
proliferation even over a few days even before formation of the Cys73-Cys73 disulfide
stabilized dimer. This appears to be due to modification of the monomeric form of
thioredoxin possibly involving reversible dimerization without covalent linkage, or to other
oxidative events in the protein. In contrast, it has been found that Cys73→Ser thioredoxin is
stable in solution over several weeks, even at room temperature, and does not dimerize.
Cys73→Ser thioredoxin is as effective as wild-type thioredoxin at stimulating cell
proliferation and retains this ability with no loss over many days and, thus, appears to be
more suitable as a potential therapeutic agent.
In order to investigate whether thioredoxin and mutant thioredoxin proteins
have activity in intact animals I studied the ability of the Cys73→Ser mutant thioredoxin to
prolong the life of mice that had been lethally γ-irradiated and which, if untreated, die from
bone marrow suppression as shown in Table 2. Mice that had been injected with the
Cys73→Ser mutant thioredoxin survived 850 Gy γ-radiation whereas non-injected mice died.
Thus, Cys73→Ser mutant thioredoxin can prevent the death of lethally y-irradiated mice.
While not wishing to be bound to any particular theory, it is presumed that this effect is due
to stimulation of bone marrow cell growth. ~*&
Table 2. Protection Against Radiation Induced Death
by Cys73→Ser Mutant Thioredoxin
Mice received 8.5 Gy -irradiation. One group of mice was treated with Cys73→Ser
thioredoxin in 0.9% NaCI 0.85 mg/mouse injected i.v. 30 min before and 24 hr after
radiation. There were 6 mice in the control group and 4 mice in the treated group. The
study was terminated on day 30.
Figure imgf000040_0001
Evidence that Cys73→Ser mutant Thioredoxin stimulates the growth of bone
marrow was obtained directly by adding Cys73→Ser mutant thioredoxin directly to human
bone marrow and studying its effects on colony formation by the cells, as illustrated in Fig. 1.
Cys73→Ser mutant thioredoxin stimulates colony formation by the muitilineage progenitor
cells (CFU-GEMM) but does not stimulate the lineage specific erythroid progenitor (BFU-E)
and myeloid progenitor (CFU-GM) cells.
Fig. 1. illustrates the stimulation of human bone marrow colony formation by
Cys73→Ser mutant thioredoxin, in accordance with the present invention. Human bone marrow was obtained as excess material from normal allogeneic bone marrow donors. The
effects of Cys73→Ser thioredoxin on colony formation are shown by (o) multilineage
progenitors (CFU-GEMM); (•) erythroid progenitors (BFU-E); and (V) myeloid progenitors
(CFU-GM), as measured over 10 days as described. (Values are the mean of 4
determinations and bars are S.D.)
It has further been found that Cys73→Ser thioredoxin can stimulate cell
proliferation by increasing the response of the cells to other cytokines or growth factors such
as interleukin-2 (IL-2) and fibroblast growth factor (FGF) as illustrated in the chart in Fig. 2.
Fig. 2 illustrates potentiation of IL-2 induced MCF-7 breast cancer cell growth by
Cys73→Ser mutant thioredoxin, in accordance with the present invention. Cells were growth
arrested for 48 hr in medium with 0.5% serum (105 cells) then stimulated in the absence of
medium with either IL-2 or Cys32→Ser mutant thioredoxin at the concentrations shown.
Cell number was measured after 48 hr. Each point on the chart represents the mean of 3
determinations and bars represent S.E. The dotted line shows stimulation by 10% serum.
In addition, antibodies to the receptors for the growth factors can block the
response to these agents, in accordance with the present invention as shown in Fig. 3. Fig. 3
illustrates the inhibition of thioredoxin stimulated MCF-7 cell growth by receptor antibodies,
in accordance with the present invention. Cell proliferation was measured as described above
in the context of Fig. 2. The concentrations of agents used were Cys73→Ser mutant
thioredoxin (Thioredoxin) 1 μM; monoclonal antibodies to FGF receptor, IL-2-receptor and
EGF-receptor 4 μg/ml; and EGF 10 nM. The EGF and EGFR were added as a negative
control. Values represent the mean of 3 determinations and bars represent S.E. The dotted
line shows the effect of 10% serum alone. Therefore, the discovery that human thioredoxin, and specifically recombinant
modified thioredoxin, does not undergo spontaneous oxidation and/or dimer formation has a
tremendous potential in vivo utility in situations where stimulation of cell growth is required.
In addition, it may be advantageous to modify the thioredoxin structure to increase the
potency and therapeutic usefulness, such as changing the amino acid sequence at the site of
proteolytic cleavage to prevent breakdown by plasma enzymes.
Thioredoxin/mutant thioredoxin may have use after bone marrow
transplantation of cancer patients or together with chemotherapy to stimulate bone marrow
recovery, or to stimulate the immune system in patients with AlDs. There may be other
potential therapeutic applications for thioredoxin/mutant thioredoxin such as increasing the
rate of wound healing. If a thioredoxin or mutant thioredoxin gene could be introduced into
an animal as a transgene this might result in an increased growth rate of the animal. A
thioredoxin transgenic mouse has been developed, although the levels of gene expression are
very low and the animal does not show an increased growth rate. However, a gene for mutant
thioredoxin might be more effective in this regard. The use of mutant thioredoxins may not
be limited to the Cys73→Ser mutant. Mutation of the other Cys residues can also affect
biological activity (30). There are also other amino acid residues on the hydrophobic domain
of the molecule that X-ray crystallographic studies have shown might also be important for
dimer formation.. Mutation of these and possibly other amino acid residues, might alter the
biological activity of thioredoxin.
The in vitro cell growth stimulating activity of human thioredoxin has been
previously reported for human lymphoid and solid tumor cancer cells (Gasdaska JR, et al.
Cell Growth Differ 6:1643-1650 (1995); Oblong JE, et al. J Biol Chem 269:11714-11720 (1994)) and for mouse fibroblast cells (Oblong JE, et al. J Biol Chem 269:11714-11720
(1994)). The production of a Cys73→-Ser mutant thioredoxin has been previously reported.
In one study it did not act like wild-type thioredoxin as a component of a complex cell
growth stimulating factor called "early pregnancy factor" (Tonissen K, et al., J Biol Chem
268:22485-22489 (1993)). In another study it was reported that Cys73→Ser mutant
thioredoxin did not form a dimer, but cell growth stimulating activity by the mutant
thioredoxin was not investigated in this study (Ren X, et al., Biochem 32:9701-9705 (1993)).
However, the ability of the Cys73→Ser mutant and other mutant thioredoxins to stimulate
cell proliferation has not been reported. There have been no prior reports of administration of
wild-type or mutant thioredoxins in vivo.
4.5.1 ROLE OF OXIDATIVE INACTIVATION OF THIOREDOXIN AS A CELLULAR
GROWTH FACTOR
Thioredoxin (Trx) is a wiJelv disrnruted redox protein ch t reyulate* sever.ii intracelluiar redov - "cea.e. and stimulates the proliferation of both normal and tumor cells. Wc have round :hat when ihsence of reducing auents. human recombinant Trx undergoes spontaneous oxidation, iosin it' uinre ;ell yjrowth. bur is still a substrate tor NADPH-derendenr reduction
Figure imgf000044_0001
human thioredoxin :re '.. α slower spontaneous conversion of Trx ro a
Figure imgf000044_0002
that ι> nor a su strat r'or •eduction T reductase and thar does not stimulate cell proliferation. Both conversions can be induced
Figure imgf000044_0003
.ancs and are reversible bv treatment with the chiol reducins auent dithiothrettol. STS- PAGE Trx undernoes oxidation to monomeπc form,s> precedinϋ dimer formation. We have recentlv av crx-stallosraphv that Trx forms a Jtmer that is stahilt.ed by an tntermolecular Cvs '-Cvs . A Cvs7' — ► Ser mutant Trx (C73S) was prepared to determine the tele of Cys' 1 in oxidative rowth stimulation. C73S was as effective a» Trx in sπmulaαns cell growth and was a comparable iioredoxin reductase. C73S did not shυw spontaneous or oxidant-induced loss of acnvttv and did ner. The results suggest that Trx can exist in monomeπc forms, some of which are mediated bv not stimulate cell proliferation but can be reduced by thioredoxin reductase. Cys' is also involved of an en:ymattcally inactive homodimer. which occurs on long term storage or by chemical us. although clearly involved m protein inactivation. Cvs1 ' is not necessarv for the zrowth ttvitv of Trx.
^
Trx" is a redox protein found in both eukaryotes and pro- aryotes [1J. The redox activity of Trx arises from a highly conserved T -Cys-Gly-Pro-Cys-Lys active site sequence where the 2 cysteine residues (Cys) undergo reversible oxidation to cystine. Reduction of Trx is cataly_«d by thiore¬ doxin reductase (2J. Trx was originally identified in EscJi- ai ia coU as a hydrogen donor for ribonucleotide reductase [3J. Trx has since been found to act as an intracellular dithiol-disulfide reductase and to modulate the activity of a number of intracellular proteins (4-61 including the DNA binding of transcription factors [7-10J. Trx-like sequences are found in other proteins including protein disulfide isυm- erase (111. There is evidence that Trx may play a role in the growth and transformed phenotype ot some cancers. Trx is
over expressed by a number of human canceπ compared with normal tissue (12-141. We have recently shown that transfection of human cancer cells with a dominant- negative mutant human Trx inhibits anchorage- independent growth m vitro and tumor formation m mo (15J.
As well as having intracellular actions, Trx act* .w^β- enously as a redo -active growth factor. Hum.ιn Trx ι« identical ιo the leukemic cell αutocnne crowth fa t r . lult T- ell leukemic factor (131. and stimulate the uf wth »< both normal fibroblasts (16| and human heπ t, > and solid tumor cancer cells in culture (17. M. Trx
Figure imgf000045_0001
*» • act by a helper mechanι>m that -<n«ιtι:e» the cell? inrowth factors secreted bv the cell.-
Figure imgf000045_0002
c> 11 H- Mutant human Trxs. where the
Figure imgf000045_0003
and C >' residue? Λt the catalytic site? are converted v serine* ι «•:*. either ΠCIV . <r together, are redox inactive and do not stimulate cell growth (161- Trx is secreted from cel bv leaderless secre- ^
tun,- r-arnwuy [ 1 | ana could be .tctinu as an autocnne factor tor tne umwth ut Mime cancer ells 1131-
We have t'ound that £. coil Trx. unlike human Trx. does nut stimulate the crnwth -it human .olid cancer cells ( IS) Tne structures of E. cult and human Trx are similar, and both are substrates tor human thioredoxin reductase. How* ever, the surface resumes ot the two t'orms van* considerably (201. Human Trx has 3 additional cysteine residues. CysA:, Cys"- and Cys73, in addition to those in the active site, which do nut normally form intramolecular disulfide bonds (20. 21]. Tπt can also orm a homodimer with a 1100 Λ2 inter ace domain and a disulfide bund between Cys71 from each monomer (20). During our studies of cell growth stimulation by Trx we observed that storage of the Trx without a reducing agent for even a few days resulted in a loss of its growth-stimulating acπvity, although the Trx ' remained a substrate for reduction by thtoredoxm reductase. e have, therefore, examined the role of spontaneous and induced oxidation of Trx and cysteine-deieted mutant Trxs. and their ability to stimulate cell prolireranυn.
MATERIALS AND METHODS Preparation of Thioredoxins
Recombinant human Trx and Cys,: -→ Ser/Cys" -+ Ser mutant Trx (C32S/C355) were prepared and purified as previously described (16). Cys' -→ Ser mutant human Trx (C73S) was prepared from single-stranded, sense strand human Trx cDNA ligated by polvethylene glycol precipua- non into the pBluescnpt KS vector (Stiatagene. La Jolla, CA) using R408 helper phage. The single-stranded cDNA was used fur uligonucleoride-dtrected m vtrro muαgenesis (Version 2.1 Kit. Amersham. Biickinghamshirc. U.K.) using olig nucleotide 5 ' -TGTTOGC ATGG ATTT- GACTTC-3'. Point mutagenesis was confirmed hy Jideuxy sequencing of base-denatured double-stranded DNA using the Sequenase Version 2.0 kit (USB. Cleveland. OH). Novel αel and 3-tmHl restncted sites were introduced at the 5' and 3' ends of the mutant Trx cDNA bv olipv- nucieotide-directed PCR. Ndel/BαmHl restncted fracments were extracted fhim aμnntse icels. ligated into a suitably restncted pET-3-a expression vector (22], rxansrocmed into E. coli BL21 cells and confirmed by didenxv seuuencinc. C73S Trx was expressed and purified as previously de- scnbed (161. AH Trxs were stirred at -20' as a 25 μM stock solution tn 5 mM DTT. Before use. the DTT was removed bv passtni! the Trx solution thnmuh a TD-10 desnlnnii column (Pharmacia. Uppsala. Sweden). The Tn solution was kept at 4* and used within 2 hr (fresh) or stmed in water or 0.1 M potassium phosphate-buttered 0.9% NaQ at 4* or -20* tor specified times. Oxidised Trx for cell trrowth studies was prepared by addinu a 5-told molar excess ^ H;0» to a 25 μM Trx stock solution without DTT and I hr later removtnc unreacted H:0; ustnu a PD-10 column.
Cell rourth ScucLies
MCF-7 human hre.tst cancer cells were obtained from the
American Culture Colle ti n |Rt>cla ille. MD) . maintained in DME containint; 10% tetal bovine serum at 37" and 6 CO.. and passaued at 75 confluence. The effect ot Trx and modified
Figure imgf000047_0001
on MCF-7 cell growth as determined as previousU descπced [IS]. Bπetlv, 10' cells were plated in a 35 -mm culture dish tn DMEM containing 10% tetal bovine serum and. arter attachment for 24 hr, growth arrested using DMEM with 0.5% fetal bovine senim for 48 hr. The medium was then replaced with DMEM containing Trx or other additions for 2 davs and cell number measured with a hemυcvtometer. All incubations were conducted in triplicate.
Thioredoxin Reductase Assay
Human placenta thioredoxin reductase, specific activity 33J μmol Trx reduced/min/mg protein, was prepared as previously described [23]. Reduction of Trx and C73S by thioredoxin reductase was measured by the oxidation of NADPH at 340 nM with insulin as the final electron acceptor as described by Luthman and Holmgren (21.
Electrophoresis
A 25 μM solution of fresh Trx. mutant C73S or C32S/ C35S Trxs; Trxs that had been aged at room temperature for 48 hr, 7 davs, 90 days; or Trxs treated for I hr with I mM diamide, 10 mM DTT, 3 mM 2-mercaρtoethanol or 2:1 (v:v) HsO:, was mixed with an equal volume of loading buffer containing 3% SDS. 10% glycerol and 0.1% brom- phenol blue in 0.05 M Tns-HCl, pH 6.8. Approximately 0.02 μg of the protein was loaded in each lane of a 24 * 45 cm 16.5% polvacrvlamide resolving gel (pH 8.4) containing 0.3% SDS, a 10% spacer gel and a 6% stacking gel and separated by elecrxophoresis using an anode buffer of 0.2 M Tns-HCl. pH 3.9 and cathode buffer of 0.1 M Tns-HCl, 0.1% SDS, pH 8.2. The gel was run for 1 hr at 400 volts before loading the samples and then at 400 volts for 24 hr before ixing in 50% methunol. 7.5% acetic acid for 20 mm, followed bv 5% methunol. 7.5% acetic acid for 20 mm, followed by 10% glutaraldehvde for 20 mm. The gel was soaked in 2 L H 0 ovemieht to remove unbound SDS and then silver stained (ICN Silver Srain Kit, Irvine. CA). Similar observations were made when the uels were stained with Coomassie Blue.
RESULTS Qro itfi Stimulation
Cys' → Ser mutant Trx (C73S) stimulated the proliferation of human MCF-7 breast cancer cells. The ECW for gmwth stimulation bv C73S was 350 nM and the maximum effect was seen at 1 μM. which is similar to values we have previously- reported tor stimulation t MCF-7 t-ell proliferation bv recombinant numan Tπ. [IS]. Stoniue of Trx in the absence ot I rejuetnc acenr -uv.h .is DTT at 4" fur 5 davs resulted in a 7^ \ι l , and for T aw ,ι S".ι loss of' cell urowth sπ ul.itini; ι F._: P In contrast. C37S
Figure imgf000048_0001
Stored at 4*C (days) 0 S 90 0 3 90
FIG. I. Stimulation of MCF-7 breast cancer call growth by fresh and aged Trx and C73S. MCF-7 caflf ware growth arrested and the stimulation of ctfl proi-ieratioα measured over 2 days using I μM Trx or C73S that was trash or had been stored as a 25 μM stock solution without reducing agent tor 3 days or 90 days at 4*. Also shown for reference is the effect of 10% fetal bovine serum. Each value is the mean of 3 deteπninations, and bars are SEM.
showed no loss of activity when stored under these conditions. Trx scored in the presence of bovine catalase at 1 unit/ml did not lose biological activity over a 5-day period (results not shown).
Reduction of T uorcαoxBU bv Tfuorcdoxtn nfrπict-sse
C73S was a good substrate for reduction by human ptacen- tal dύoredoxin reductase with a K of 0.20 μM and a Vm m of 6.3 runυVrαin/μg. These values are similar to those we have previously found for fresh Trx, which were a K„ of 0 J3 μM and a V^ of 5.9 M∞l/mm/μg (23).
The effect of storing Trx without DTT on its ability to act as a substrate for duotedoxin reductase was investigated (Table I). When stored in H:0 either at -20* or at room temperature Trx showed a loss of activity with a hatf-Ufe of
TABLE 1. Effect of storage of
In H,0 In PBS
-20* ♦«• -20* ♦II* t,„ (days) n (days) n (days) n (da )
Trx 30.5 20.1 8.2 U
C73S stable" staWe' stable* s aMe* ZT-ϊ
Figure imgf000049_0001
> Tne los ot Trx ictivir. MJJ more rapid „ nen st.-reJ in phosphate-buttered C. "n N.iCl. with a h.iit ine >r :
Figure imgf000049_0002
Phi-rhjte butter is know n to contain .small a ounts ot iron [Z-|. which could catalv:e an oxidative pru o in- creJMnu* the l ss iii Trx activity. Alternatively, the lower pH of the Hilutum in water could stabilize Trx or the increase in ionic strencth of phosphate-buffered 0.9"; NaCl could enhance the formation of the inacnve homodimer oi Trx. The aaed Trx showed a slow, delayed reduction bv thioredoxin reductase that was stimulated by catalytic amounts of fresh Trx (Ftp. 2). It is important to note that the loss at activity of Trx as a substrate for thioredoxin reductase was much slower than the loss of activity as a stimulator of cell growth. C73S did not shυw a loss of activity as substrate for thioredoxin reductase upon storaee for up to 30 davs. The ability of Trx to act as a substrate for thioredoxin reductase was completely inhibited by treatment wtrh 5 molar equivalents of H 0:. whereas C375 remained fullv active after treatment with 100 molar equivalent? oi H:0: (Fig. 3).
Multiple Forms of Thioredoxin
Electrophoretic analysis of freshly prepared human Trx stored in DTT showed a mixture of 5 bands of apparent molecular weights tanging from 6.1 to 11 kDa (Figs. 4, 5, and 6, lane 1 ). Storage of Trx at room temperature without DTT resulted in a change in the banding pattern with disappearance of the 8.1 -kDa band by 48 hr (Fig. 4, lane 2). Storage of Trx without DTT for 7 days resulted in the loss of additional bands and the appearance of a new band at 23 kDa due, apparently, to a Trx dimer (Fig. 4, lane 3). Storage of Trx without DTT for 90 days at 4* resulted un almost complete conversion to the Trx dimer (Fig. 4, lane 4). Treatment of 7-day aged Trx (Fig. 5, lane 2) with 2-mer- captoethanol resulted in the reappearance of the fresh Trx banding pattern, except for the 8.1 -kDa band, which did not reappear (Fig. 5, lane 3). Loss of the smaller bands and dimer formation was seen when Trx was treated with di-
Figure imgf000049_0003
^
Figure imgf000050_0001
HjO. (mole equivalents)
FIG. 3. The effect of HtOt on the reduction of Trx (filled ban) and C73S (open ban) by thioiedoxin reductase. Trx solutions were treated with varying amounts of Ht02 for 18 bra at room temperature. Reductaie activity WH measured by adding treated samples to a solution of 0.1 HEPES buffer, pH 7.6, 5 mM EDTA, 17 μM insulin, 100 μM NADPH. 15 μg/ml human thioredoxin reductase and measuring the race of NADPH oxidation at 340 tun at room temperature. One hundred percent of dύαreaoxin reductase activity ia defined as 0.1 absorbance iinufmin/rnM Trx or C73S Trx. HjO, had no effect on the oxideuon of NADPH.
amide, a protein thiol oxidizing agent [251 (Fig. 5, lane 5). The formation of Trx dimer following diamide treatment was also confirmed by gel permeation chromatography (results not shown). HjO; treatment of Trx also caused dimer- izatiυn but produced a different banding pattern to that pnxjuced by diamide (Fig. 5, lane 6). Treatment of Trx with NEM. a thiol alkylating agent [26], gave a single band with a slightiy elevated apparent molecular weight, but no dimer formation (Fit:. 5. lane 4). Treatment of 7 day aged Trx with NEM pnxiuced both the higher molecular weight hand us in Fin. 5. Lane 4. and the hands illustrated in Fig. 5. Lane 2 (data not shttwn), suggesting that in the aged Trx not all the ilfhydryls are available for covalent modification. None of the changes caused by NEM were reversed with 2-mercaptoethanol treatment (data not shown). 2-Mercaproethnnol reversed Trx dimer formation caused by both diamide and H;0: treatment (Fie 5. lanes 7 and 8) but was less effective at reversing changes in the monomeπc banding pattern of Trx pnxiuced by H;0: (Fig. 5. lane 8).
Freshly prepared C73S Trx and C32S/C35S Trx showed fewer band* than wild tvpe Trx (Fig. 6. lanes 2 jnd 3, compared with lane 1 ). Treatment of C32S/C355 Trx with diamide resulted in the formation of a 2 >-kD imer (Fιι*. 6. Iune e Treatment ot C73S Trx wirh diamide caused the bunds to oal s into a in le band ot around 10 kDa. but <*?> there was no 23-kDa dimer formed (Fu». 6, lane 4) τ_ effects ot diamide un C37S and C32S/C35S were reverseri bv creatment with DTT > F11 . 6, lanes 5 and 7'ι.
DISCUSSION
The study shows that human recombinant Trx undergoes at least 2 levels of spontaneous and induced oxidative trans, formation. The first oxidation occurs spontaneously within a few days to a form(s) that can nυ longer stimulate ceil growth but remains a substrate for thioredoxin reductase. The slower oxidation occurs over a period of weeks, or can be induced by the thiol oxidizing agent diamide, and leads to a disulfide bonded homodimer which not only fails to stimulate cell growth but is a poor substrate for thioredoxin reductase. The fact that similar changes in Trx can be induced by chemical oxidation, are protected against by catalase and are reversed by the thiol reducing agent DTT is consistent with the interpretation that the changes in Trx are due to oxidation. Cys71 appears to play a critical role in both levels of oxidant-induced inactivation since C73S does not lose the biological activity or its ability to act as a substrate for thioredoxin reductase upon aging.
We have shown by SDS-PAGE that fresh human recombinant Trx can exist in at least five different states, which probably reflect the fully reduces state of the protein as well as different intramolecular disulfide bonded states due to the five cysteine residues present in the protein. While the specific nature of these intramolecular disulfide bonds is not known, it is likely that some, at least, are due to non-
23.3
Figure imgf000051_0001
FIG. 4. Effect of storage on Trx studied by SDS-PAGE. Protein was stained with silver stain. Lane 1, fresh Trx: lane 2, Trx 48 hrs at room temperature without DTT; lane 3. Trx 7 days at room temperature without DTT; and lane 4. Trx stored 90 davs at 4β without DTT. Position of molecular on the left. narurai disultide ^on <-. structure? wπich orm durinu Je- natur tion and the oxιdι:ιnu condition, ot extended elec- trop-v-resu [271- The observation :nat C375 and C32S' C35S exhibit a simpler banding pattern than wild-tvpe Trx upon SDS- PAGE also suggests that the banding pattern is due to disulfide bond formation. X-rav structural analysis indicates that in addition to a disulfide Kind between Cvs'" and Cvs . the onlv other intramolecular disulfide bund that could form in the non-denatured Trx is between Cys' ' and Cvs '. although even this would require a different conformation of the protein [20|. With the exception uf a pι>>5irle sliuht modification in C *, there is no evidence that Cys,;. Cys" or Cys*" are oxidύed in Trx crystals formed tn the presence of 5 mM DTT |20|. The fact that treatment of Trx with NEM pnxluces only one band implies that pnor to denaturatton and electrophoresis fresh Trx exists as a single species. The number uf free thiuls in fresh Trx was determined to be 4.5 to 4.6/molecule by Ell- man '» reauent [2S] (data nut shown), indicating that all five cvstemes are in the sulfhydryl form. Treatment of NEM- alkvlated Trx with oxidizing or reducing agents produces no change in the banding pattern (data not shown), which is runner evidence that all 5 sulfhvdrvls have been alkvlated.
23.3
Figure imgf000052_0001
S
23.3
Figure imgf000053_0001
*"' 1 2 3 4 S 6 7
FIG. 6. Oxidation and reduction of mutant Trxs studied bv SDS-PAGE. Lane 1, fresh Trx; lane 2, fresh C73S Trx; Lane 3. fresh C32S/C35S Trx: lane 4, C73S Trx treated with 1 mM rfwnnVisn mnc 5, C73S Trx as in lane 4 treated with 10 mM DTT; mac 6, C32S/C35S Trx treated with 1 mM di- amide; and lane ?, C32S C35S Trx as in lane 6 rreated with 10 mM DTT. Position of molecular mass markers in kDa are shown on the left and right.
Oxidation of cysteines to sulfenic or sulfinic acids is unlikely ω occur spontaneously [29], It is noteworthy that HsOz treatment of Trx gives rise to a different monomeric banding pattern than that of spontaneously oxidhed Trx. The original monomeric banding pattern is also not regenerated by treatment with DTT. As has been previously suggested for NADH peroxidase [30]. we speculate that HjO; oxidizes the cysteines to sulfenic acids and to the irreversible suihnk or sulfon acid states.
During the same time interval that there was a loss of the growth stimulating activity of Trx. there was a shift of the ehxtrophoreric banding pattern. There was a collapse of the banding pattern with Ion of some of the Trx monomeric bands over 7 days, suggesting that Trx may be undergoing "native" mtramolecular disulfide bond formation prior to eleαrophoresis, which prevenα the formation of random dtsulfide-bond formation seen with denaturation and electrophoresu of fresh Trx. A similar phenomena has been observed with bovine pancreatic trypsin inhibitor [27, 31). Alkytation of aged Trx with NEM gave more than one protein product, indicating that aged Trx exists in multiple forms and not all the sutmydry are available for reaction. Since C73S does not undergo a similar shift in banding pattern and does not undergo loss of growth stimulating activity, it can be assumed that Cys'' is involved in this intramolecular disulfide Kind formation, perhaps with Cys'2 (Fip. 7). Thus, spontaneous aging of Trx over a few davs results in the inability of Trx to stimulate cell growth, nlthinieh Trx is "rill a su rrnte for reduction hv thioredoxin
Figure imgf000054_0001
FIG. 7. Position of cysteines in human Trx. Ribbons and ball-and-sάck representation showing the relative positions of Cys", Cys11. Cys", CysM and Cys", based on the crystal coordinates for the wild type redneed protein (20). None of tne thiols are in a position for rtrnilfirlir bond formation except for the redox active pair Cys12 and Cys1*. The inter- molecular rϋwlnπV bond requiring the least distort-on in the protein would be between Cys31 and Cy»". The sutmydryb for these residues are 9.1 A apart in the model, but could possibly approach each odser through local distortions in nearby residues , Both reiiriuci are in loops, making necessary distortions of energetically lower cost. The region near Cys'1 has already been shown to adopt alternate conformations [20], in support of this μuesihflliy. This figure was made with MOLSCRIPT [35].
reductase. Analysis of the X-ray structure υf Trx shows that Cys7' is by far the most accessible cysteine residue and possibly the most reactive [33].
If a solution uf Trx is left long enough, or upon treatment with a strong oxidising agent such as diamide or H>0:. there is formation of 23-lcDa Trx homodimer. Reducnυn of the Trx dimer by thioredoxin reductase is slow and delayed, and is stimulated by low concentninυns of fresh Trx, suggesting there may be an autocatalvtic process. A similar conclusion was reached by Ren et al. [34|. Formation of the Trx homodimer appears to involve the Cvs' ' residue since C73S, where Cys' 1 is replaced with senne. does not undergo oxidation- induced homodimer formation as dυ Trx and C32S/C35S. Ren a al. [34) have shown C73S does not undergo oxidative homodimer formation induced by se- lenodithioglutathione. We recently reported the X-ray crystal structure of Trx and idenπfied a lareelv hvdmphobic dtmer forminu interface that is stabιlt:ed bv a Cvs '-Cvs' 1 disulfiJe bond [20]. Our observation that Trx undercoes a ι>ter loss of activity with thioredoxin reductase in PBS versus water indicates that iron-induced oxidation or an increase in ionic strength mav stabilize and enhance dimer formation, which is consistent with the hvdrophohic nature ^i the dimer interface observed in crystals of human Trx
The importance of the monomeric oxidative torm(s) ot Trx is unknown. While the structural nature is vet to be identified, it does have different hioloyicjl activity in our m wrro system. Trx is secreted bv cells into the extracellular environment, which is ptedominandv oxidizing, and might be expected to undergo monomenc oxidation. Consideπng its ease of formation, it is reasonable to assume that monomenc oxidation will precede oxidative homodimer formation. Whether this might be sufficient to prevent Trx from acting as a growth factor is not known. The formation of the oxidized monomer inside the cell is less likely since it s li can be slowly reduced by thioredoxin reductase and the interior of the ceil is highly reducing.
The physiological significance of homodimer formation is also unknown. What might be Trx homodimer has been reported in diamide-treated Jurkat celb [35]. We have observed small amounts of the Trx homodimer bv immuno- bloitmg of untreated MCF-7 breast cancer and other cell lysates (Powis et al., unpublished observations). It is in- tnguing to speculate that formation of an oxidized Trx monomer or homodimer in response to intracellular oxi- dants such as H;0; might be a wav mammalian cells detect oxidant formation. Trx is believed to exist in normal celb at concentrations from I to 10 μM [2, 12], though in selected tissues and specific cell compartments this value could be much higher. It is therefore not unreasonable to assume that Trx will undergo homodimer formation in vivo. As we observed with the enhanced inacttvation of Trx tn phosphate buffered saline, we expect dimer formation to precede faster m mvo than we observe m wrro m water. Whether dimer formation m wvo would prevent the taster oxidation to an intramolecular form is unknown. The slow autocatalvtic reduction of the Trx homodimer to the monomer would be a way to restore the cell to normal operating conditions after the induction of oxidative stress.
In summary, we have found that human recombinant Trx undergoes relatively rapid (over a few days) spontaneous and oxtdant-induced conversion to a form(s) that doe« not stimulate cell proliferation, but is still a substrate for reduction bv thioredoxin reductase. There is much dimer (over a period of weeb) spontaneous oxidation of Trx to a Cvs7,- rabιli:ed homodimer form that is not a substrate for thioredoxin reductase and that also does not stimulate cell proliferation. Both conversions can be reversed by treatment with the thiol reducing agent DTT. and Kith appear to involve the Cys71 residue. A Cvs7' → Ser mutant Trx. which stimulates cell pπiliferation and u as etfecπve a substrate for thioredoxin reductase as Trx. did not show aw or oxidation- induced loss i these activities. Thus, with time Trx «raduallv los» its ahtltrv to stimulate cell proliferation and to he ,ι «uhscr.ue for thioredoxin reduer.ι>e. unlike the C\s"' -+ er mutant Trx. which retains ιhe>e activities with no l . Thus. Cvs7' ι> not critical for holocical JL -nrir-tci IΠUMUWIC-.- ' -.rise jitt v. etn«.- <n nuinjn ch ire- Ji.iin rr. jiwi i (f rri ui? inhibitor* ar rwroirrtrutiifedtfxin rcuukMse ru ciimin.itu.n .1 micucnic (smperties of thtore- Juxin ,' 3 » Owm 269: ! 1714-1 1 20. 1994 W.ι .ιsuuι N. Tauava V. W.ikasum A. MICWI M. .Maeda M. Yodoi J jn Tuπz T. Adult T-S. '11 leukemia-denved factor/ Thioreαoxin produced both human T.lvmphocropic virus tvpe I and Epstein-Barr vtπa-πansrormed Ivmphocytes. acts
1. Holπurren A. Thioredoxin. Annu R v Bk-J m 54: 237-171. as an auco nne tmm-rh racror and svneηpzed with interieu- 1 )}5. km- 1 and ιnterieukιn-2. Proc Sad Acad Sa L'SA 87: 8232-
2. Luchrrun M and Holmitren A. Rat liver thioredoxin and 8236. 1990. thioredoxin redu ta-ei punπcaciυn and characterization. Bκ»- It). G-udadu JR. Be nrnm M and Powι» G. Ceil puwth aαmula- than 21: 6623-66? 3. 1902. tiun bv the redox protein thuiredoxin occurs by a novel
3 aurenr TC. Moore EC and Reicrurd P. Enzvαutic svnthesu helper mechanism. Gill Uπm-th Differ 6: 1643-1650. 1995. ot deoxvnbonucleotides VI. Isolation and hafacteruaαon of 19. Ru uteili A. Boierto A. Allavena G. Wullπum E and Site R. thionsdixon. the hydruiten donor mm Exhtndβ coii B. Biol Secretion of thwredtwm by normal and neopiaacic celb Ohrm 239t 3436-3444. 1964. rhπaath a leaderles* secretory parhway. / Biul Chan 267:
4. Fountoulakis . UnhiUing inrermeuβre of the extracellular 24161-24164.1992. domain of rhe inrerfenm uamma receptor, J Bad GWm 267: 20. ctchicl A. Gasdaska JR. Powts G and Mondbrr R. C - 7095-7100. 1992. tal structures of reduced. oxnJized. and mutated human
5. tiumer A. S ndeπ D and Habcrmann E Disulfide fυrmanort rhumrdoxiru: Evidence rivr a teαuiatory homodimer. Srnicntre in reduced tetanus toxin bv muiredαxin: The pharmacuiogi- 4: 735-751. 1996. cai role of interchain covalent and nαncυvalem bonds. Ton- 21 rorman-Kav JD. CUm GM. ingneU PT and Gronenbom urn 31: 1423-14)4. 1993. AM. Htιrh-f«ι>iutιon rhree-dimer-iiυnal structure of reduced ά. Silverman RS and NanJi OL Reduced duureduxin: A posrec mbiiunr human thuweskixui in solution. Biocnem 30: sible phvjioloffcai cotacωr tor vtααun K cpoxalc reductase. 2685-2698. 1991. Further support for an active iite αuulnde. Bmenrm BiuprVu 21 Studier F . Rownbenj AH. Dunn JJ and Dubendorff JW, Res Cimmun 153: 1248-1254. 1988. Use of T7 RNA pυlvmcrase to direct exprcsuon of cloned
7. Havashi T. Ueno Y and Okamoto T, Oxakeeducαve regulagenes. Methods ErtQrmul 18Sι 60-89. 1991. tion of nuclear factor xB: Invoh-emcnt o( a cc-kuar reducing 23. CMx*}Ε. Gι*i*k* r .5^ll K v^ ?ι G. r r&c2∞r, caalvst thioredoxin. / Biol Chen 26βt 11380-11388, 1993. of human riiioiidoxin rauuctase Ptopernes and dhavacter-ca-
8. Gaiter D. Mih S and Oroge W. Dtaanα effect of ijucathi- non bv shsorpoorv and αrαutx dkhrotsm spe iuacupy. B v one disulfide on the nuclear tnucnpπøn factoo xB and the ώm 32. 7271-7277. 1993. aαivaαw protein- 1. E r J Biochem 221: 639-648. 1994. 24. Poyer JL nd McCav PB. Reduced σιriΛxexιopyndine nucleo-
9. Gπrro JF. Tienrungroi W, Dahmcr MK. Housley PR and ode oxslm nrilvtirl aiteranons of meπuirane ihoapholioids. Ptaσ wTB. Evidence that the endogenous heat-stable gjuco- J Bsoi Chem 246: 263-269. 1971. comcoid receptor-acnvacing factor α th-ocβdoxin. } Biol 25. Kosower NS and Kusowcr EM. Diamide: An oxidant probe for Chem 258: 13658-13664. 1983. thsob. Metfnb Eritjmoi 23 li 123-132. 1993.
0. Cromitth JA and Roeder RG, Human αanaeripoon factor 26. Gilbert HF. Thiol/d-iu-firJe exchange eaudibna and disuinae 1IIC (TFIIC). Punfk-ioon. rx>fypepode srnxnsre. and the inbond stability. MeA tt vjmol 251: 8-30. 199S. volvement of duel groups tn spec c DNA binding. J Biol 27. Cr iahton TE. Disulfide bond foαnaαøn m rAorjnns. MeAodi Chrm 264: 18100-18109. 1989. Eτcτ oi 107: 303-329. 1984.
1. Freedman RB. Hint TR and Tuitt MF. Protein diaulphidt 28. EDaan GL Tissue Sul ydryi groups. Arch Biochem βioprra isomerase: budding bridges un proct toidiDf, Trench Biochem 82: 70-77. 1959. Sa 19t 331-336, 1994. 29. Claibomβ A. Miller H. f^rsonate D and Ross RP. Proonn-
2. Berggren M. Gallegos A. Gasdaska JR. Gasdaska PY. su-fenac aαd itabiuanon and function m enzyme catalysis Wa JM j arid PowwG. TTιιors-k>xm and rhsr-iaonx and gene regulation. FASEB J 7ι 1483-1490. 1993. rase gene expression in human tu on and ctfl tines, and the 30. Poole LB and Oaiborne A. The non-Aavin redox center of effects t ( serum gimuraoon and hypo-oa. Aaβxjancer Res (in the srflttisβcoccal NADH peroxidase II. Evxiencc far a sαbf press). lizad criorm-si-iferuc acid. / Biol Chan 264: 12330-12338.
3. Gaadaska PY. Oblong JE. Cagmve IA and Powis G. The 1989. predicted ammo sod sequence of hwaean rreosssloain a akm- 31. Weaaasn JS and Kim PS. Reexanunaαon of the folding of ncal to that of the aumcnnt growth factor huaaan aduk T- BPTk r oαunancc of move mrermediann. Sαencc 253: ceil derived factor ( ADrT: Triβreoβxm saRNA ■ elevated tn 1386-1393. 1995. some human tumors. Sirxfe Bfaαin* Aαa 1218s 292-296. 32. Krautis PJ. MO ESCRIrT: A program to produce both de1994. tailed and Khemanc plots of proonn stnanuπα. ) Appl Cnst
4. Nakamura H. Masutani H. Tag Y. Yasaasjc i A. Luamoto 24t 946-950. 1991. Y. Nanbu Y. Fujii S. Own K and Yodoi J, Expression and 33. Page DL. Prognosis and breast cancer. Recognition of lethal frewm-promoting effect of aduk T-ctfl leukiss— Jtnved <ac- and favorable pwtnosiic rrpe-. Am / Swg Pβthol 15t 334-349, tor. A human rhturedoxin hom iopur tn hefwuceikilar car1991. cinoma. Cancer 69ι 2091-2097. 1991 34. Ren X. Bjornstedt M. Sh n B. Encson ML and Holrngren A.
>. Galleeos A. Gasdaska JR. Guodman 0. Gasdaska PY. Berg- Muαaencsis uf structural half-cvsnnc residues in human men M. Bπehl MM and Powβ G. Tian feiasan with human thioredoxin and erfeea on the reaulatuwi uf acnviry by w thioredoxin increases ceil proliferation and a dueatnant negalenodieiutathtυne. 3 -.-fvm 32: 9701-9705. 1993. tive mutanr thioredoxin reverse the rrarua-rtaed phenocvpe 35. Sato N. Iwara S. Nakamura K. Hoπ T. Mon K and Yixkx J. ot brenst cancer cells. Cancer Rts on press). TriMil-πιedιaκd re ox reuuiaruin ot apoptosis. ) imm r A 154: i Oblonc IE. Betmiren M. Gasdaska PY and Poms G. Site- 5194- »203. 19U5 4.6 REDUCING } TIBITION OF APOPTOSIS IN TUMORCE S TIiAT.OYER- EXPRESS THIOREDOXIN
ABSTRACT
The redox protein thioredoxin plays an important role In controlUng cancer cell growth through regulation of DNA synthesis and transcription factor activity. Thioredoxin is overexpressed by a number of human primary cancers and its expression is decreased during dexamethasone- induced apoptosis of mouse WEHT7.2 thymoma ceils. We examined the ability of WEHT7.2 cells stably transfected with human thioredoxin cDNA showing increased levels of cytoplasmic thioredoxin to undergo apoptosis in vitro and in vivo. The cells were protected from apoptosis induced by dexamethasone, staurosporine, etoposide, and tharnigargin, but not by JV-acetyl-sphingosine. When inoculated into severe combined Immuno- defl ent mice, the trx-transfected cells formed tumors that showed increased growth compared to wild-type, as well as ύc/-2-transfected, WEH17-2 cells. The trx- and £>c/-2-transfected cell tumors both showed less spontaneous apoptosis than tumors formed by the wild-type ceils. Unlike tumors formed by the wild-type and M-2-u-ansfected WEHI7.2 cells, fπr-translected cell tumors did not show growth inhibition upon treatment with dexamethasone. This study suggests that increased thioredoxin expression in human cancers may result in an increased tumor growth through inhibition of spontaneous apoptosis and a decrease in the sensitivity of the tumor to drug-induced apoptosis.
INTRODUCTION
Trx3 is a low molecular weight redox protein found in both pro- karyotic and eukaryouc cells (1). The cysteine residues at the conserved Cys32-Gly-Pro-Cys35-Lys active site of Trx undergo reversible oxidauon-reduction catalyzed by the NADPH-dependent selenium- containing flavoprotein Trx reductase (2). Human Trx is a protein of Λfr 1 1.500 with 27% amino acid identity to Eschenchia coli but containing three additional Cys residues not found in bacterial Trx that give the human protein unique biological properties (3). Trx was originally studied for its ability to act as a reducing cofactor for ribonucleotide reductase. the first unique step in DNA synthesis (4). More recently. Trx has been shown to exen redox control over a number of transcription factors, including nuclear factor KB (5), transcription factor I11C (6). BZLF1 (7), and the glucocorticoid receptor (8). and indirectly, through nuclear redox factor Ref-1/HAPE, Trx can regulate AP-1 (Fos/Jun heterodimer; Ref. 9).
Trx is also a growth factor with a unique mechanism of action. Human Trx stimulates the proliferation of both normal fibroblasts and a wide vaπety of human solid and leukemic cancer cell lines (10, 11). Redox activity is essential for growth stimulation by Trx. and mutant redox-inactive forms of Trx lacking the active sue Cys3: and Cys35 residues are devoid of growth stimulating activity (1 1 ). Studies with '~ Mabeled Trx have revealed no high-affinity binding sues that
Received 2 7Λ7. accepted sVI7 )7
The costs of publication of this article v. ere defrayed in pan by the payment of page charges This article must therefore be herebv marked advemstmt in accordance with I K U.S.C Section 17*4 solels to indicate this fact
' This won, was supported bs NIH Gram CA4S725 do G. P.I A B was panly supported bs grant from the American Foundation for Pharmaceutical Education
' To whom requests for repnnts should be addressed, at Anzona Cancer Center ~ Lmxersm of Anjona I 5 | North Campbell Avenue. Tucson. AZ 85724-5024 Phone (52(>ι 62h-h4(lκ Fat (520) h2t>-4«48. E-mail gpowis@azcc.anzona.edu
' The abbreviations used are Trx thioredoxin. scid. severe combined im unodefi- πrnl CAT chlnramnhenicni acetvltran ferase might suggest receptors for Trx on the surface ctfcaaccr cells (12). Trx appears to -»«"»ni— cell raOliferanon by ittcRating the sensiuviry of the cells to growth factors secreted by the ceUs themselves (12).
We have found thai Trx mRNA it elevated cotnoared to paired normal tissue in almost half of the r-uman rirnary lung and colon tumors we examined (3. 13). Other studies have found increased Trx ia human necφlastic cervical s-nιaBious epitheliurn ceUs and henato- cellular cminσma (14, 15). We have recently shown that human breast cancer cells nnsfected with a *» i«nι negative, redox-inactive mutant Trx show reduced aικhorage-irtrter»no m growth at vitro and an almost complete inhibiόoα of tumor foπnation in vrvo (16). Thus, Trx overexpresaion may be a factor in the growth of some iππftff1 cancers.
We previously reported that Trx gene expression is decreased during n>xaτnethasone-inchιced apoptosis of mouse triytnorr-a-derived WEHT7-2 cells (17). To further study the effects of Trx on apoptosis, in this study we stably transfected WEHT7.2 cells with btixnan Trx cDNA and fxtminrri the effects on both spontaneous and drug- induced apoptosis in vitro and with the cells growing in scid mice.
N TERIALS AND METHODS
Cans. Human wild-type Trx cDNA was prepared as described previously, cloned uao the Λforl sne of the pDC-MMαco mammalian tiw-sfccnon veciαr ( 16) and transfected by clecsoporsflon uao mouse WΕHΠ.2 trjyτnc*na-derived cells (18). Tunifeπrd cells were maintained at cuhure de-tancs up to 10* ccUs/ml in DMEM romainrnj 10% fetal bσvme serum supplemented with 800 μ ml C 18 surfate, and cloαea were laolased ia soft afarose and then maintained u. 200 uf/ml G418 sulfate. AU studies were nmriurirri on clonal hues between passages 3 and 20. Stably transfected M-2 WEH-7-2 celb (W_Hbl2 cells) were obtained from Dr. Roger Miesfeld (Umversrry of Arizona. Tucson. AZ: Ref. 19). Drup were added at a culture density of I X 10* to 2 X 10s cxUs ml. Stock solunons (10 mM) of (texam-thatαne were utepsiui ia ethanci, whereas -taurotporme. etoposide. th-φstμrtui. and N-acesyl-sphuigostne were prepared tn DMSO. Further dϋαoons were made unng culture medium. a-RNA Eipuaatiiii Ncethera blc< hybtvtiτ.-inn analysis was performed as rtramhrd prevtoutly unng a fuU- ngm lo>J,P)<-CTP- be d human Trx cDNA probe (3). Blots were quantified uung a Molecular Dynamics Phos- porrlmaccr.
Gsacscart-coid BareptTi. The level of funcnonal fjucocorucoid receptors was assessed using a transient ccvansfecuoo of cells with a glucocorocoid response elemenf CAT reporter p-tsrmd (pπunCAT; Ref. 20) and β-falacιo- ndase. After a 22-h recovery period, the cells were treated with I μM dexa- me aaone. and CAT protein was measured after in additional 24 h unng s CAT EL1SA (Boehπnger Mannheim. lncJanapolu. IN). An aliquot of the transfecied cells was stained for 0-salsctøuda-e activity and CAT activity normahzed for transfection efficiency.
Apiptnr-i Apoptotu was measured by an E ISA for htsto∞-a-sociated DNA frasmema (21). by morphotofy and by flow cytometry (22). The cntena uaad for the π>orphc of ical idennficauon of apoptonc ceils included condensation and rr-crfinsoon of the chromcαn with the formation of crescents, cell ihnn aie. increased -uinint. nuclear frsfToentanon. cytoplasπuc vacuoliza- ooc and apoptooc body formation. Cells were incubated with 20 μf/ml 7-aramo aconomyc-n D for 30 nun si 4*C before beιn| analyzed by flow cytometry.
Isfπanfliiui israaiu Statn a Cell- were centn uied onto 0.17-mm- tmc quartz covenlrps. air dried for 10 nun. fixed with 4% rncthanoMree fonnak hyde for 20 mm at room terr enuure. washed for IS mm in PBS. pH WO 98/24472 r_L PCT/US97/22292
<≡>3-
7 2. and permeubilized with lOOt meihanol at -20°C for 6 mtn. The cover- slips were then stored at — 20°C until immunostaimng. when thev were allowed to come to room temperature and blocked with 1% BSA in PBS. This was tolloweϋ bs a I IU dilution ot goat serum in PBS. before reacting tor I h with a 1.100 dilution ot tmmunoalfinuy purified rabbit annhuman Trx polv- cloπal anti xx ι ! 3) Λlter being washed with PBS. the coverslips were exposed lo a I 1 (X) dilution ot goat aniirabbit biotinsiaied IgC for I h. washed with PBS. exposed to a 1.30 dilution ot fluorescein streptavidm fluorochrome. and again washed with PBS Cells were examined using a Leica TCS-ID laser scanning contocai microscope with an excitation wavelength of 488 nm. For subcellular localization studies ut Trx. Cy5 i indodicarbocyanme) streptavidm was used as the fluorochrome. tollowed by digestion tor I h at room temperature with l(X) μg/ml RNase A and DNA stained with 25 ΠM YOYO-! iodide tor 10 min. Cells were then examined bv laser scanning contocai micro copy at excitations ot 48H nm i YOYO- I I and 647 nm lCvSi Relative fluorescence intensities ot groups of 20 cells were measured at the same laser power photomultiplier tube voltage, and line averaging setting as grav level intensities using SigmaScan soitware ijandel Scientific. Cone Madera. CA) Because ihe transfected cells exhibited an uneven distribution ot fluorescent staining, a template of a regular array of dots was placed over the image, and multiple I up to 90) nuclear and cytnplasmic measurements were made
In Vim Tumor Growth. Tumor lormation by wild-type and iranstected WEHI7 2 cells was studied bv injecting 2 x 10' ceils in 0.1 ml ot matπgel s.c into the flanks ol groups ot 20 temale scid mice. Tumor volume was measured with calipers, and mice were euthanized when the tumor volume exceeded 2 cm" Nine davs alter tumor cell injection. 10 mice trom each group were injected i p with I mg/kgdav dexamethasone in lO^r ethanol in 0 T- NaCI Control mice were injected with vehicle alone. On dav 14. three mice tmm each group were euthanized with CO.. and the tumors were excised and immediately fixed in glutaraldehsde
Preparation of Tissue for Bright-Field Examination. The glutaralde- hyde-fixed tissue was posifixed in osmium tetroxide. dehydrated in a graded seπes ot alcohols and embedded in epoxv resin One-μm-lhick sections were prepared and stained with loluidine blue for bπght-field examination
RESULTS
VVEHI7 2 cells were stablv transfected with human Trx cDNA in the pDC304neo mammalian transfection vector We examined multiple clones and found the maximal increase in Trx mRNA compared to endogenous levels ot mouse Trx mRNA. was 1 8-fold for clones Trχ5 and Trx6 ( Fig IA ) As determined by immunofluorescent staining and confocal microscopy, the »rt-traπsfecιed cells showed increased levels of Trx (Fig IB) The relative fluorescence intensity of wtld-rype WEHI7.2 cells ( z SE. rt = 20) was I 00 r 0.05. ot Trx5 cells. 2.15 = 0 14 ( P < 0 001 compared to wild rype i: and of Trx6 cells. 1 87 = O i l {P < 0 001 compared to wild type) Trx-hke fluorescent staining was observed in the nucleus as well as the cytoplasm of the cells (Fig l O In the wild-type cells. 60.1 r 5 1 % of the fluorescent staining was in the nucleus, in the TrxS cells it was 59.8 ~ 2.5%. and in the Trx6 cells it was 36.1 = 1.8*.
Compared to both wild-type or vecior-aione-rxanstected cells, the rrt-transfected WEHI7.2 cells were resistant to apoptosis induced by 1 μM dexamethasone as measured by histone-associaied DNA fragmentation (Fig. 2A l or by flow cytometry (Fig 23). Histological examination of the WΕH17.2 cells revealed a classic apoptotic morphology in response to dexamethasone However, only a small fraction of the cells undergo apoptosis at any one time, and they rapidly progress to fragmented cells For this reason, results are expressed as relative apoptosis rather than percentage of apoptotic cells. Glucocorticoid receptor activity measured using a glucocorticoid receptor/CAT reporter plasmid was not decreased in the rrt-transfected cells (results of three studies not shown) We also studied the effect of rrt trans- tection on other agents known to induce apoptosis (Table I ) Compared to vector-alone-rraπsfected cells, rrt-transtected cells were resistant to apoptosis induced by siaurospoπne. a general kinase S8
Figure imgf000060_0001
Figure imgf000060_0002
Figure imgf000060_0004
Figure imgf000060_0003
Fi| I λ Northern blot hvbndijation analvsis of total RNA exiracted from wild-type mow * EHI7 : cells Irom pDOt-tneo vtcισ>-amne iransfecied WEHI7 . cells i/veoi and Irom ihe rri-transfcv-ied V.EHI" : s wes Trx? and Trxb Λ full-length ,:P-labeled rrt DSΛ prow was used lor hshnditaiion r.'i> nun_ transiecttd human Tn mRNΛ. otuiπm haΛU mouse Trx mRSΛ The taiwrs on ihe nt .now ine position of molecular weight markers tkbi 8 fluorescence tmmunoriisiochemic-l staining of Trx tn cells using itnmu- noalfinus.puπfied rabbit nuhun n Trx poivclorul aniibodv bionnvlated goat antirabbit IgG fluorescein ttnrputidin and User wanning honrocal microscopy / wtld-iype V.EHI* : sdl> .' pOC3oaneo sector alone transtected WEHI7 cells. J TnS in- transtecied cell* J Tnft tn tr-nsiected cell* C lluorescence immunohistochemical staining ut Trx using C*3-sireρta*H)ιπ fluorochrome and YOYO- 1 to coumentain nuclear DSΛ showing that Trx is prestm in ine isioplasm and ihe nucleus of wild-type WEHI7 _ cells ι5> and 7>τo in transiecied cells toi
inhibitor ι23 l, bv a cell-permeant sphingσsine analogue. /V-acetvl sphingosine ι 4 ι bv ihapsigjrgin which blocks the uptake of intracellular Ca: * resulting in an increase in intracellular free Ca " concentration ι ! 4ι and hv etoposide a topoisomerase II inhibitor
fig. 2. Effects of iπc and *c7-2 miufc ion n WEH17-2 cells on dcumethasone-induced ftpopto- . A, apoptotii measured by an EUSA for hmone- associated DNA frafmems. expressed u relative πuctco»oeo*l enrichment. Wild, wi Hype WEHI7.2 cells: *V«. pDC304neo ve or-irone-transf-Kfe EHI7.2; W.HbU. b i-2-τmxt tά WEHI7.2: Trxi and Trτό. rrx.transfected WEHI7.2 cells. The cells were treated with 0.01* ethanσl vehicle <■> or I μM dexamethasone (Q). md spoptotis was meu< urad 24 later. Coiwiwu. men of four detcrmina- uoni; ban. SE. •. P < 0.05 compared to Neo controi. B. apoptosis measured by flow cytometry showiπf typical lesulu. Ref rani Λl. R2. and ΛJ of the vanerpraros ate live nonapoptotK. early apoo-
Figure imgf000061_0001
lotic. and late apoptotic cellt. respectively. /. pOC3 4neo vector-alone- transfected control cells: 2. pDC304neo vecior-alonc'innsfected cells treated f r 48 h wtth I μM dexamethasone: J. Trxύ trx-mrnitcua WEHI7.2 celh: J Trxb celK treaied for 48 h with 1 μ dexamethasone.
125 WEH17.2 cells txansfecie with the bcl-2 antiapoptoiie proio- υncogene f .HB12 cells) showed a similar pattern of proteciioπ against apoptosis induced by the various agents as did ihe rn-tninv- fected cells (Table ).
When inoculated into scid mice, the trx- transfected WEHI7.2 celK formed tumors that grew more rapidly than tumors formed b> cither wild-type or bcl-2- transfected WEHI7.2 cells rFig. 3Λ ) Upon histo- >o
Figure imgf000062_0001
Figure imgf000062_0003
Wild Neo W.Hb12 Trxδ Trx6
Figure imgf000062_0002
Forward light scatter
logical examination, tumors formed by the vwld-type cells showed fields of apopiouc cells adjacent lo fields ot viable cell , as well as apoptotic cells adπuxed with viable-appearing cells (Fig 3flι The cells undergoing apoptosis exhibited the classic appearance of condensed and argmated chromaun. some in the form of crescents, and j dense cvtoplasm accompanied by vacuolization The rn -transfecied EHI7.2 cell tumors showed minimal numbers ot cells undergoing O 98/24472 £>
Figure imgf000063_0001
apoptosis scattered throughout the tumor, mass. Tumors formed by tc/-2-ιransfected WEHT7.2 ceils also showed very few cells undergoing apoptosis (not shown). Areas of necrosis were seen in wild- type, (rr-transfected. and έr -2-rransfected cell tumors, usually adjacent to fields of viable-appearing rumor cells or. in the case of the wild-type cells, adjacent to areas that show extensive apoptosis or next to viable-appearing cells. Treatment of t e ice with dexamethasone starting at day 9 had no effect on the growth of the rrx-rxansfected cell tumors but markedly inhibited the growth of the wild-type rumors and the £>c/-2-transfected cell rumors (Fig. 3Λ). Histological examination revealed no evidence of increased apoptosis caused by dexamethasone
Figure imgf000063_0003
Figure imgf000063_0002
treatment of wild-type, trx-transfected. or ->c/-2-transfecιed cell tumors.
DISCUSSION
WEHI7.2 cells stably transfected with human rrx showed a maximal increase of 1.8-fold in Trx mRNA compared to endogenous levels of mouse Trx mRNA. This relatively low level of overexpression is similar to our experience with trx transfection of mouse NIH 3T3 cells and human MCF-7 breast cancer cells (16), suggesting that higher levels of unregulated trx expression may be toxic to cells. As deter-
Figure imgf000064_0001
Time (days)
B
Figure imgf000064_0002
mined by immunofluorescent staining and confocal microscopy, the rrv-transfected cells showed approximately 2-fold increased levels of Trx. The finding that Trx is present in the cytoplasm and the nucleus of cells confirms an earlier immunohistochemical study using conventional light microscopy of cervical tumor cells that reported eyto- plasmic. nuclear, or cytoplasmic and nuclear localization of Trx ( 14). This is an important observation because Trx may be able to directly reduce redox-regulated nuclear transcription factors, such as AP-1 (Fos Jun heterodimer. Ref. 9). If Trx can enter the nucleus, it may not need other nuclear redox factors, such as Ref- 1 /HAP 1. as has been suggested (9).
The rrr-traπsfected cells were resistant to apoptosis induced by dexamethasone. Trx has been reponed to be necessary for assembly of the glucoconicoid receptor (8). However, glucoconicoid receptor activity was not decreased in the transfected cells, suggesting that the effects of Trx on apoptosis appear to lie downstream of the glucocorticoid receptor. The /nr-iransfected cells also showed resistance to apoptosis induced by siaurosporine. etoposide. Λ/-acetyi-sphingosine. and thapsigargin. Exogenously added human Trx has been reported to inhibit apoptosis induced by tumor necrosis factor a in U937 human lymphoma cells (26). However, we found that exogenously added human Trx did not protect WEHI7.2 cells against apoptosis induced by dexamethasone (27). Tumor necrosis factor a and dexamethasone are thought to trigger apoptosis by different signaling pathways. It may also be that exogenous Trx is not taken up by WEHI7.2 cells. We have found that other tumor cells take up Trx poorly, if at all (12). Clearly, an increase in intracellular Trx achieved by transfection of trx in the present study is associated with resistance of the WEHI7.2 cells to apoptosis induced by dexamethasone and other agents.
The pattern of resistance to drug-induced apoptosis caused by trx transfection is similar to that produced by transfection with the human proto-oncogene bcl-2. Bcl-2 is believed to exeπ its inhibitory effects upstream of the activation of the cysteine aspanate proteases cascade (caspase) responsible for the final stages of apoptosis (28). The protective effects of Bcl-2 against apoptosis have been suggested to involve an antioxidant mechanism (29). although this is disputed based on the ability of Bcl-2 to block apoptosis caused by agents that are thought not to act by an oxidant mechanism (30) or caused by hypoxia (31 ). The aπtioxidants Λ'-acetyl-cysteine. pyrrolidine dithio- carbamat.. Trolox (a water-soluble vitamin E analogue), and buty- laied hydroxytoluene protect rat thymocyies against drug-induced apoptosis (32. 33). We have previously reported that Trolox. catalase. and superoxide dismuiase protect murine WEHI7.2 cells against dex- amethasone-induccd apoptosis (27). It is intriguing, therefore, that trx. a gene that codes for a known redox-active protein, also inhibits apoptosis. The mechanism by which Trx inhibits apoptosis remains to be established, but its pattern of amiapopiotic activity similar to Bcl-2 suggests that it also may act upstream of the cysteine proteases.
WEHI7.2 cells transfected wiih trx formed tumors in scid mice that grew considerably faster than tumors formed by the wild-type parental celK or b nr/-2-transfected cells. This may be due. in pan. to a decreased rate of spontaneous apoptosis that occurred in ihe trx- transfected cell tumors. High levels of Bcl-2 have been found in a wide variety of human cancers (34). Although transfection with bcl-2 is known to confer resistance to apoptosis induced by anticancer drugs and radiation, the effects of bcl-2 on tumor growth are less clear. Transfection with bcl-2 gives a survival advantage to cells in culture (351. Transgenic mice overεxpressing Bcl-2 under transcπptional regulation of the immunoglobulin heavy chain enhancer develop benign lymphoma that eventually progresses to high-grade malignant disease (36ι This suggests that bcl-2 also provides a survival advantage in cells in vivo but that an additional change, most frequently rearrangement of mvc (36). is necessary for tumor growth. Our studies using -EH37.2 thymoma cells show that bcl-2-tτanstccιti, cells formed tumors that grew faster than tumors formed by wild-type WEHI7.2 cells. This may be due to a reduction in the rate of spontaneous apoptosis observed in the -> -2-transfected cell tumors compared to the wild-type tumors. It was not possible to distinguish a difference in the rates of spontaneous apoptosis between the trx and £>t7-2-transfected cell tumors. Paradoxically, the ύc/-2-transfected cell tumors still showed growth inhibition by high-dose dexamethasone treatment, as did wild-type cell tumors. There was no evidence for increased apoptosis caused by dexamethasone treatment of wild-type, frr-transfected, or M-2-tτansfected cell tumors, so the possibility remains that in vivo dexamethasone does not inhibit tumor growth in vivo by a mechanism that involves increasing the rate of apoptosis. The results of this study and our previous work ( 16) suggest that the Trx system offers a novel target for agents to promote apoptosis and inhibit tumor growth, as well as to reverse the drug resistance of some cancers. It is interesting, therefore, that some 2-imidazolyl disulfide inhibitors of Trx (37) have been shown to induce apoptosis in cancer cells (38) and, in animal studies, to have antitumor effects (38).
In summary, we have shown that transfection with trx. a gene found to be overexpressed in a number of human cancers, can inhibit apoptosis of cancer cells in culture induced by a variety of agents. In animals, the rrx-transfected cancer cells show an increased growth, decreased spontaneous apoptosis, and decreased sensitivity to apoptosis induced by dexamethasone. If similar effects occur in patient tumors, then trx could be a new human proto-oncogene.
REFERENCES
) Holmgren. A Thioredoxin and gluurcdoxin systems. J. Biol. Chem.. 2M: 13963—
13966. 1989 2 Luthman. M . and Holmgren. A R-l liver thioredoxin and thioredoxin reducuue: purification and characten-ation Biochemistry. 21 6628-6633. 1982.
I Gasda-ska. P Y.. Oblong. J E Cotcreave. I A . and Powis. G. The predicted amino acid sequence of human thioredoxin is identical lo lhat of Ihe autocnne powth factor human adull T-cell densed factor (ADF I Ihioredoain mRNA is elevated ui some human tumors Biochim Biophys Acta. 1218 292-296. 1994
4 Laurent. T C. Moore. E C. and Reichard. P. Enzymatic synthesis of deoxyribc- nuckondes VI Isolation and characterization of thioredoxin. the hydrogen donor Irom iΛ hrnch culi B J Biol Chem.. 23V. 3436-3444. 1964.
5 Matthews. J R . Wak sugi. N . Virelizier. J-L.. Yodoi. J.. and Hay. R- T. Ttυoredoiin regulates the DN binding activity of NF-«cB by reduction of a diiulprudr bond insol ing cyueine 62 Nucleic Acids Res.. 20 3821-3830. 1991
6 Cromlish. J A- and Rocdcr. R C Human tπutscnpuon factor LUC fTFlIIC). PuπfKHion. potypereidr structure, and the involvement of thiol groups tn specific DNA binding J BK- Chem.. 2Λ4 18100-18109. 1989
7 Bannister. A J.. Cool. A., and Kotuandn. T. In mm DNA binding acuvrry of Fos/Jun and BZLFI but not C EBP it affected by redoi changes. Oncogene. 6. I24 230. 1991
8 ϋnppu I F . Tienrunrmi W . Dahmcr. M K . Housley. P R_ and Pran. W B EsH-rnce thai the endogenous heai-s h glucoconicosd rcceptor-eαivatinf factor ts lhH«ron>ιn J Bwl Chem . 2Vt 13658-13664. 1983
9 Abate. C . Paiel. L. Rauschet F J III. and Curran. T Redox regulation of fos and run DNA-rπndinp actisns in mm Science ( W ashingion DO. 2*9 1157-1161. 1990
10 Pimis C. . Oblong. J t . Gasdaslj. P Y.. Berggtm. M.. Hill. S.. and Kirkβ-tnci. D L The ihiorcdυiinrtriinred in reducuse redos system and control of cell gro th. Oncnl Res . 6 5'9- 4 I9SM
I I Oblong. 1 E . Berggren. M . Gasdas . P Y.. and Pr is. C Sne-dirtct-d muugov esι» o( active site cvsteines in human thwredoiin produces competitive wi-bnors of human Ihioredosin reductase and elimination ol mnogenic pro enies of thMreomtn. ) Biol Ovtm . -'ftv 1 1714 - 1 1720 I9SVI i : jasdasia. J R Berggfrn M and Prmis G Cell rrm-ih utmulation bs the redoi nmein ihioredoiin ticcurs h* a isel helper mechanism Crll Grr ih Difler.. 6 InJ i- IΛVI 1995
I ' Berggrm M . Ciallefos A (lasdisl-. J R . (iasdasVa. P >. V. ame-e. J . and Pπβiv G Thioredoxin and thioreυ un reducuse gene expression in human lurnors and cell lines and Ihe eflects ol srnini stimulation and hs-pm" Aimcancπ Res . '6 34.SS)- λ4(* 1096
N hum S Nanbu V Nrsmγ ., II <<πιshl."P mT.'T M.'siitanrΗ"' ιW\ !•<•'•« I Cι«rspnrssκιn ol -dull T-sell lcsi-emia-densed laclor. a human Uuoreooim hnmo- logue. and human papillomavirus DNA in πcoplastic cervical sututmous epithelium Cjncer tPhila.i. 6Λ 1583-1591. 1991
Kawahara. N . Taπaku. T . Yokomizo. A . Nunπ. H.. Ono. M. Waco. M . Kuhno. Takenaka. K.. Sugimachi. K.. and Kuwano. M. Enhanced coexpression υt thioredoxin anil high mobility group protein I acnes in human hepjto ellular carcinoma and the possible association with ecrexsed sensiiivuv to cisplatin Cancer Res.. SΛ 5330- 5333. 19%
Gullegos. A.. Gasdaska. J R.. Taylor. C W.. Paine-Mumetx G D . Goodman. D . G-.sda.ska. P Y . Berggren. M.. Bnehl. M M.. and Powis. G Translectton wuh human thioredoxin increases cell prolileπmon and a dominant negative mutant thioredoxin reverses the transformed phenotvpe nt breast cancer cells Cancer Res . 56 5765-5770. 19%
Bnehl. M M.. Cotgreave. 1. A., and Powis. G. Downrcgulation of the antioxidant etense dunng glucoco coid-mediaied apoptosis Cell Death Differ . 2 41-46. 1995
Hams. A W . Bankhurst. A. D . Mason. S.. and Warner. N L Differentiated functions expressed by cultured mouse lymphoma cells II. β antigen, surface immu- noglobulin and a receptor for antibody on cells of a thymoma cell line J Immunol . 110 431-438. 1973
Lam. M.. Dubyak. G.. Chen. L... Nuiiez. G . Miesfeld. R. L... and Distelhorst. C W Evidence that BCL-2 represses apoptosis by regulating endoptasmic retιculum-a.sso- ciated Ca- * fluxes Proc Natl Acad Sci USA. 9/ 6569-6573. 1994 Rundlett. S E.. Gordon. D A., and Miesfeld. R. L. Characterization of a panel of rat ventral prostate epithelial cell lines immortalized in the presence or absence of androgen- Exp Cell Res.. 203. 214-221. 1992
Leist. M.. Gartner. F . Bohlinger. I.. Tiegs. G.. and Wendel. A Application of the cell death detection ELISA for the detection of tumor necrosis factor-induced DNA fragmentauon in muπnc models of inflammatory organ failure Biochemica. / / 20-22. 1994
Philpott. N H . Turner. A J C. Scopes. J Wesiby. M.. Marsh. J C W . Gordon- Smith. E C . Daigleish. A. G . and Gibson. F. M The use of 7-amιno actinomycin D in identifying apoptosis simplicity of use and broad spectrum of application compared with other techniques Blood. 6 2244-2251. 19%
Kondo Y . Woo. E.. Michalska. A E . Choo. A C. and Lazo. J S Metallothionein null cells have increased sensitivity IO anticancer drugs Cancer Res . 55 2021-2023 1995
Pushkareva. M . Obeid. L M . and Hannun. Y A Ceramide an endogenous regulator of apoptosis and growth suppression Immunol Today 16 294-297. 1995 Onishi. Y . Azuma. Y . Sato Y Mizuno. Y . Tadakuπu T . and Kizaki H Topoi-
somerase inhibitors induce apoptosis in ihvmocyies. Biochim. Biophys. Acta. 1175 147-154. 1993 Maisuda. M.. Masutam. H . Nakamura. H . Mivajiπu. S.. Yamauchi. A.. Yonehara S . Uchida. A.. Inmajin. K.. Honuchi A . and Yodoi. J Protective activity ot aduli T cell leukemia-denvcd lactor I ADFi against tumor necrosis factor-dependent ivm- toxicily on U937 cells J Immunol.. 147 3837-3841. 1991 Baker. A. F . Bnehl. M M.. Dorr. R.. and Powis. G Decreased antioxidant elense and increased oxidani stress duπng dexamethasoπe-induced apoptosis bcl-2 selectively prevents ihe loss ot caialasc activity Cell Death Differ.. 3 207-213. 19% Shimizu. S.. Eguchi. Y . Kamnke. W . Maisuda. H . and Tsujimoto. Y Bcl-2 expression prevents activation ot the ICE protease cascade Oncogene. 12. 2251-2257 19% Hockenberry. D M . Olivai. Z N . Yin. X.. Milliman. C. L.. and Korsmeyer. S J Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell. 75 241-2 1 1993 Jacobsen. M . Burne. J F . and Raff. M C Programmed cell death and rxl-2 protection in the absence ot a nucleus EMBO J . 13 1899-1910. 1994 Jacobson. M D . and Raff. M C Programmed cell death and Bcl-2 protection in verv low oxygen Nature (Lond I. 374 814-816. 1995 Wolfe. J T . Ross. D . and Cohen. G M A role for metals and free radicals in the induction of apoptosis in Ihvmocyies FEBS Lett.. 352 58-62. 1994 Salgo. M G.. and Pryor W A Trolox inhibits peroxyniinte-mcdiated oxidanve stress and apoptosis in rat thymyocyies Arch. Biochem. Biophys.. 333 482-488. 19% Reed. 1 C . Miyashita. T . Takayama. S . Wang. H . Saio. T.. Krajewslu. S . Aime- Sempe. C. Bodrug. S . Kuada. S . and Hanαda. M BCL-2 family proteins regulators of cell death involved in the paihogenesis of cancer and resistance to therapy J Cell Biochem . 60 23-32. 19% Maeyama. Y Enhanced expression of bcl-2 inhibits apoptosis in cultured human keratinocytes Kurume Med. J.. -12 291-297. 1995 McDonnell. T J . and Kor mever. S J Progression from lymphoid hyperplasia to high-grade malignant Ivmphoma in mice transgenic for the M 14.18) Nature (Lond I. 349 254-256. 1991 Kuperus. M . and Kirkpaiπck. D L Interaction of redox active disulfides with Ihe auiocnne growth factor human thioredoxin Proc Am Assoc Cancer Res . 36 426 1995 Powis. G . Kirkpatnck. D . Angulo. M Gasdaska. J R . Powell. M B . Salmon S E. and Montfort W R Thioredoxin redox signalling a novel target for anucancer drug development Anticancer Drugs 7 I Suppl 3) 121-126. 19%
Figure imgf000069_0001
T dorβ oxin (M)
Figure 1. Stimulation of human bone marrow colony formation by Cys73→Ser mutant thioredoxin. Human bone marrow obtained as excess material from normal allogenetc bone marrow donors. The effects of Cys73→Ser thioredoxin on colony formation by: (o) multilineage progenitors (CFU-GEMM); (•) erythroid progenitors (BFU-E); and (V) myeloid progenitors (CFU-GM), was measured over 10 days as described. Values are the mean of 4 deteπ inations and bars are S.D.
Figure imgf000069_0002
£> %
Figure imgf000070_0001
Figure 2. Potentiation of IL-2 induced MCF-7 breast cancer cell growth by Cys73→Ser mutant thioredoxin. Cells were growth arrested for 48 hr in medium with 0.5% serum (10s cells) then stimulated in the absence of medium with either IL-2 or Cys32→Ser mutant thioredoxin at the concentrations shown. Cell number was measured after 48 hr. Each point is the mean of 3 deteiminations and bars are S.E. The dotted line shows stimulation by 10% serum.
Figure imgf000070_0002
έ±s
Figure imgf000071_0001
+ + + ÷ Trx
+ + EGF
Antibodies
FGF?.
IL-2?.
EGF?.
Figure 3. Inhibition of thioredoxin stimulated MCF-7 cell growth by receptor antibodies. Cell proliferation was measured as described in Figure 2. The concentrations of agents used were Cys73→Ser mutant thioredoxin (Trx) 1 μM; monoclonal antibodies to FGF receptor, IL-2-receptor and EGF-recep*.or 4 μg/rnl; and EGF 10 nM. The EGF and EGFR were added as a negative control. Values are the mean of 3 detexi-iinations and bars are S.E. The dotted line shows the effect of 10% serum alone.
Figure imgf000071_0002
3<
Figure imgf000072_0001
Figure 2 Effects of Trx and C32S/C35S cDNA transfection on proliferation of MCF-7 cells. Left panel; 3 x 10* cells were plated in 3.8 cm2 plastic culture dishes in DMEM with 10% fbs and cell number measured 7 days later. Right panel; 10* cells were plated in 2 cm2 wells containing soft agarose and colonies measuring >60 μ. counted 7 days later. Control, the Neol vector-alone tranfected MCF-7 cells; Trx 9, Trx 12, and Trx 20, MCF-7 cells transfected with human Trx cDNA; Serb 4, Serb 15 and Serb 19, MCF-7 cells transfected with C32/C35S cDNA. Values are the mean of 3 determinations and bars are S.E. **indicates p <0.01 compared to vector-alone transfected cells, •"■indicates p <0.01 compared to vector-alone transfected cells.
Figure imgf000072_0002
f
Figure imgf000073_0001
Time (days)
Figure 3 Growth of Trx and C32S/C35S-tτansfected MCF-7 breast cancer cells in scid mice. Female scid mice implanted s.c. 2 days previously with a 21 day release pellet of 0.25 mg 17-β-estradiol were injected subcutaneously with 2 x 10* transfected MCF-7 cells in 0.1 ml 0.9% NaCl and 0.1 ml matrigel. (O) MCFneo, pDC304 vector-alone transfected MCF-7 cells; (▼) Trx 12, thioredoxin transfected cells; (■) Trx 20, thioredexin transfected ceUs; (V) Serb 4, C32S/C35S transfected cells; and (O) Serb 15, C32S/C35S transfected cells. There were 4 mice per group. Tumor growth was measured twice a week for 40 days. The 17-β-estradiol pellet was replaced at 21 days. Values are mean and bars S B.
Figure imgf000073_0002
^
Human thioredoxin is a putative oncogene that may confer both a growth and survival advantage to tumor cells. Over-expressed thioredoxin mRNA has been found in both primary human lung and colorectal cancers. To determine the intratumor distribution and amount of thioredoxin protein in human primary tumors and to determine if its overexpression is related to proliferation or apoptosis, we studied primary human gastric carcinoma samples. An immunohistochemical assay for thioredoxin in paraffin embedded blocks was developed. We studied ten patients with primary high risk gastric carcinoma. To relate thioredoxin protein overexpression to apoptosis we utilized a paraffin based in situ assay (Tunel) and to delineate proliferation we utilized the nuclear proliferation antigen detected by Ki67. In this survey we found thioredoxin was localized to tumor cells and overexpressed compared to normal gastric mucosa in 8 of 10 gastric carcinomas. The thioredoxin was found at high levels in 5 of the 8 overexpressing carcinomas. The overexpression of thioredoxin was typically found in both a nuclear and cytoplasmic location in the neoplastic cells. There was a significant positive correlation (P=0.0061) with cancer cell proliferation measured by Ki67. There was a significant negative correlation (P=0.0001) with apoptosis measured by the Tunel assay. Thus.human primary gastric tumors that are highly expressive of thioredoxin have both a higher proliferative rate and a lower rate of spontaneous apoptosis than tumors that do not express thioredoxin. Whether this thioredoxin-related combined growth and survival advantage translates into poor clinical outcome remains to be determined.
WO 98/24472 ^\ ^ PCT/US97/22292 >
Thioredoxins are low molecular weight redox proteins found in both prokaryotic and eukaryotic cells (1). The cysteine (Cys) residues at the conserved -Cys-Gly-Pro-Cys-Lys active site of thioredoxin undergo reversible oxidation-reduction catalyzed by the NADPH-dependent selenium containing flavoprotein thioredoxin reductase (2). Human thioredoxin is an 11.5 kDa protein, with 27% amino acid identity to E. coli thioredoxin. It contains 3 additional Cys residues not found in bacterial thioredoxin that give it unique biological properties (3).
Thioredoxin was first studied for its ability to act as reducing cofactor for ribonucleotide reductase, the first unique step in DNA synthesis (4). More recently thioredoxin has been shown to exert redox control over a number of transcription factors, including NF-KB (5), FIHC (6), BZLF1 (7), the glucocorticoid receptor (8) and, indirectly through another redox factor Ref-1, AP-1 (Fos/Jun heterodimer) (7). Thioredoxin modulates the binding of the transcription factors to DNA and thus, regulates gene transcription.
Thioredoxin is also a growth factor with a unique mechanism of action. The predicted amino acid sequence of thioredoxin is identical to that of a previously identified growth factor secreted by HTLV-1 transformed leukemic cell lines, called adult T-cell leukemia-derived factor (ADF) (3). ADF stimulates growth of lymphoid cells (9,10). We extended these observations and showed that human recombinant thioredoxin stimulates the proliferation of normal fϊbroblasts and human solid tumor cancer cells even in the absence of serum (1 1, 12). It does this by increasing the sensitivity of the cells to growth factors secreted by the cells themselves (13). For example thioredoxin at nM concentrations, as are found in human serum (14), increases the sensitivity of human breast cancer cells to interleukin-2 (IL-2) and basic fibroblast growth factor (bFGF) by 1000 and 100 fold, respectively (unpublished observations). The term "voitocrine", from the Greek "to help", has been coined to describe this growth stimulating activity of thioredoxin (13). Mutant redox-inactive forms of thioredoxin lacking the active site cysteine residues and E. coli thioredoxin are devoid of growth stimulating activity (12). Human thioredoxin is known to be secreted from cells by a leaderless secretory pathway (15) so that it could be acting extracellularly to stimulate cancer cell growth.
Our work has shown that thioredoxin is important for the growth, death and transformed phenotype of some human cancers. Stable transfection of normal fibroblasts with human thioredoxin cDNA (trx) increases their growth rate and transfection of human MCF-7 breast cancer cells with trx increases their colony formation in soft agarose (16). Transfection of the MCF-7 cells with a dominant negative redox inactive mutant trx causes inhibition of colony formation and almost complete inhibition of tumor formation when the cells were inoculated into scid mice. Our recent studies have shown that stable transfection of mouse thymoma cells with human trx inhibits apoptosis induced by a variety of agents including glucocorticoid, staurosporine, N- acetylsphingosine, thapsigargin and etoposide, which is similar to the pattern of inhibition seen with the antiapoptotic oncogene bcl-2 in these cells (17). The trx transfected cells form tumors that when inoculated in scid mice grow more rapidly and show less spontaneous apoptosis than vector alone or bcl-2 transfected cells, and are resistant to growth inhibition by glucocorticoid(17). These results suggest that trx offers a survival as well as a growth advantage to tumors in vivo, unlike bcl-2 which offers only a survival advantage and requires other genetic changes for tumor "3S growth (18).
We have previously reported that almost half of human primary lung cancers we examined overexpress thioredoxin mRNA compared to normal lung tissue from the same subject (3). Recently we have found that more than half of human primary colorectal tumors have elevated levels of thioredoxin mRNA, up to over 100 fold for one subject, compared to normal colonic mucosa taken from within 5 cm of the tumor from the same subject (19). In these studies, however, thioredoxin mRNA was extracted from pieces of tumor and nothing is known of its mtratumor distribution, or even if the increased thioredoxin mRNA leads to an increase in thioredoxin protein. It remains to be determined if thioredoxin overexpression is related to proliferation or apoptosis in human primary tumors. These are clearly important questions that we now address in the current studies utilizing primary human gastric carcinoma samples.
In the current study we sought to develop an assay for thioredoxin in paraffin embedded blocks allowing survey of human tumors in archival tissue banks. To this end the Southwest Oncology Group (SWOG) Gastrointestinal Biology Laboratory made available a relevant archival paraffin block bank of gastric carcinomas. Furthermore, to relate thioredoxin to apoptosis we also sought to refine a paraffin-based in situ assay of apoptosis (20,21). Finally, to relate thioredoxin to proliferation, we utilized the previously developed assay of the nuclear proliferation antigen detected by Ki67 (22).
Methods Patient Sampies: Paraffin blocks from ten gastric carcinoma resections were studied. These pathology samples derived from ten patients on Southwest Oncology Group (SWOG) protocol 9008 (also known as intergroup study # 116). This is a study of high risk gastric carcinoma comparing gastrectomy only versus gastrectomy plus adjuvant therapy. The patients ranging in age from 42 to 75; all had previously untreated, stage π to HI B gastric carcinoma. They had biopsy proven adenocarcinoma of the stomach which had a high risk for recurrences due to evidence of carcinoma extension beyond the muscularis propria and/or having lymph node involvement. Patients with Stage O, IA or any stage with Ml were not eligible. As of December 1996 this study has accrued 486 patients.
Immunohistochemistrv: Five micron thick sections were deparaf inized and then subjected to antigen unmasking with one of two methods with heat plus citrate buffer at pH 6.6 or microwave plus EDTA buffer at pH 8.0 as previously described (20,21). The best signal to noise ratio was established by judging reactivity with cell lines known to be a high expressor of thioredoxin (A549 human lung cancer) and a low expressor of thioredoxin (SK BR3 human breast cancer)(19).
All tumor samples and control cell lines were stained using a standard immunohistochemical method as previously described (20-21). To obviate biotin receptor reactivity, biotin-avidin blocking was performed first Then the primary antibody (polyclonal rabbit anti-human thioredoxin) (19) was utilized at a titer of 1/200 after titration of control cell lines. The best signal to noise ratio was found following microwaving at pH 8.0 with EDTA buffer. Sections were treated with biotinylated goat-anti-rabbit antibody and then with avidin-peroxidase complex, each for 30 minutes at 42* C in an automated immunostainer (VMS ES, Ventana Medical Systems, Tucson, Arizona) (20,21). Sections were counterstained with methyl green, dehydrated, rinsed in xylene and coverslipped.
The degree of thioredoxin expression in tumor cells was judged at 400x magnification as 4+ (very intensely positive), 3+ (moderately intensely positive), 2+ (moderate), 1+ (faint), or 0 (completely negative) throughout the sample. A single investigator (TG) was responsible for scoring all the samples.
Additional immunohistochemical assays employed antibody to proliferation antigens, Ki67 (Ventana, Tucson, AZ), also using the biotin-avidin labelled method after avidin blocking (22). The degree of Ki67 staining, again judged at 400x magnification, was classified as the percentage of nuclear positive tumor cells listed as: absent (0), >0-5% (+), 6-25% (++), 26-50 (+++), >51% (++++).
Apoptosis Assay: Apoptotic cells were detected utilizing the TUNEL assay (23, 24) adapted to an automated in situ hybridization instrument (gen II, Ventana Medical Systems, Inc.). The TUNEL assay utilizes recombinant terminal deoxynucleotidyl transferase (Tdt) (GLBCO BRL) for adding homopolymer tails to the 3' ends of DNA which are more abundant in apoptotic cells(23, 24). Biotin-16, 2'-deoxyuridine-5'-triphosphate (Biotin 16-dUTP) (Boehringer-Mannheim, Indianapolis, IN) was the label used for terminal transferase in this DNA 3' -end labelling reaction. Avidin-Horseradish Peroxidase and 3,3'-diaminobenzidine as chromogen(23, 24).
The instrument utilized deparaffinized sections with subsequent digestion with Protease I (Ventana Medical Systems Tucson, Az.) for 8 minutes VMS1). Incubations were performed per Ventana Gen II protocol on the instrument with the final steps being as above using avidin-horse radish peroxidase and DAB detection method to visualize the apoptotic nuclei as an intense brown color (diaminobenzidine). As an enzyme control we utilized two sections from each tissue: one with Tdt
enzyme and one without enzyme (negative control).
The TUNEL assay result was scored by the number of brown - apoptotic tumor nuclei per high power field (400x objective). The values were: 0 (absence of apoptotic cells), + (>0-2/hpf), ++ (2-4/hpf), +++ (>4-8/hpf), ++++ (>8/hpf).
Statistical Analysis
Thioredoxin expression was correlated with Ki67 expression and with apoptosis measured by the TUNEL assay using Spearman's nonparametric rank correlation test.
Results
The optimum signal to noise ratio was found by using the following antigen retrieval conditions: microwaving at pH 8.0 in EDTA as tested by a high thioredoxin expressor (A 549) and low thioredoxin expressing (Sk BR3) cell line.
Immunohistochemical localization of thioredoxin (positive staining) was found in the tumor cells of 8 of 10 gastric carcinoma samples (see table 1). Seven of these eight showed both nuclear and cytoplasmic staining (see Figure 1). The two cases with no tumor thioredoxin showed positive staining in the adjacent normal mucosa and are important controls, suggesting these are true, not false-negative, tumors (see Figure 2).
Among the eight thioredoxin positive gastric carcinomas there was a range of posirivity from faint (+) to intense (++++) with five cases having high level thioredoxin (+++ to + ι i i ) and three having low level (+ to ++) (see Table 1).
In all samples there was the adjacent normal mucosa where the strongest staining was found in the gastric mucosal pits (++) while faint staining was found in the superficial mucosa (+). The localization differed based on site with gastric pits showing both nuclear and cytoplasmic staining while the middle mucosa had only cytoplasmic staining (see Figure 3).
Increased levels of thioredoxin levels positively correlated with increased cell proliferation as measured by Ki67 expression (r=0.861, p=0.0061) and negatively correlated with apoptosis as measured by the Tunel assay (r=0.949, p=0.0001) (see table 1, and figures 4 and 5).
Discussion
An important aspect enabling this study is the development of two methodologic refinements: (1) the use of heat-based antigen unmasking methods to allow optimal, reliable measurement of thioredoxin by IHC in archival paraffin embedded tissues; and (2) adaptation of the Tdt based TUNEL assay to an automated procedure on an automated in situ machine. The heat-based antigen optimization of IHC entailed heating the paraffin section in 5mM EDTA in 0.1 M TRIS, pH 8.0. The specificity of the reaction was assured by the finding of high positive signal in A549 human lung carcinoma, a known high level thioredoxin expressor as determined by prior Western blotting (19). SK BR3 breast carcinoma cells likewise served as low level expressor control also established by prior Western blotting. The gastric tumor samples themselves also served as important positive and negative, same-slide controls. In particular, within the entrapped normal gastric mucosa gastric pits and mid-level mucosal cells showed thioredoxin signal while surface mucosal cells were negative. There was a clear difference in the subcellular localization of thioredoxin in normal positive gastric cells, the lower level cells in the pits showed cytoplasmic and scattered nuclear staining, while the higher mid-level graduation staining was typically lighter and restricted to the cytoplasm. The significance of this differential distributions is not known. Thioredoxin does not have a known nuclear localization sequence (3). From our IHC studies it is clear that thioredoxin is specifically located within neoplastic gastric carcinoma cells and not in stromal cells or admixed B or T lymphocytes or macrophages. The tumor cell thioredoxin density typically exceeded that of the adjacent normal mucosa. The minimal background staining and strong signal to noise in all the samples, as illustrated in figures 1-5, demonstrate the refinement of the thioredoxin paraffin assay we have developed.
In this survey of 10 primary human gastric carcinomas we have determined the extent of thioredoxin overexpression and determined its localization and its relationship to proliferation and cell survival (apoptosis) status. We found thioredoxin is overexpressed, compared to normal gastric mucosa, in the malignant cells of 8 of 10 gastric carcinomas. The thioredoxin protein was found at high levels in five of the eight overexpression carcinomas. The expression was typically found in both a nuclear and cytoplasmic location in the neoplastic cells. There was a significant positive correlation (p<0.01) between increased levels of thioredoxin expression and cell proliferation measured by Ki67 expression. There was also a significant negative correlation (p<0.0001) between increased levels of thioredoxin and apoptosis measured by the TUNEL assay. Thus, human primary gastric tumors highly expressive of thioredoxin have both a higher proliferative rate and a lower rate of spontaneous apoptosis than tumors with absent or low thioredoxin (see figure 4). This finding is consistent with our experimental observation that the stable transfection of mouse WEHI7.2 cells with human wild type thioredoxin leads to increased tumor growth rate in vivo associated with a decreased rate of spontaneous apoptosis (17). We have also found that transfection of human MCF-7 breast cancer cells with a dominant-negative redox inactive mutant thioredoxin inhibits tumor growth in vivo (18). Thus, overexpression of thioredoxin in gastric carcinoma is associated with increased cell growth and cell survival giving the cells doubly immortalizing properties. Whether this will translate in patients into more aggressive tumor growth, as seen in animals with thioredoxin transfected tumor cells, and a poor prognosis remains to be determined.
There have been 2 reports of the immunohistochemical distribution of thioredoxin in human primary tumors. Fujii et al (25) reponed that, while the squamous and glandular cells of normal human cervix showed no thioredoxin IHC, the intermediate and superficial layers of cervical squamous neoplastic tissue, as well as invasive squamous cell carcinoma showed cytoplasmic and nuclear staining for thioredoxin. A study by Kawahara et al (26) has reported enhanced expression of thioredoxin in human hepatocellular carcinoma compared to adjacent non-cancerous liver, with both a nuclear and cytoplasmic localization of the staining. Thus, thioredoxin overexpression appears to be a common phenomenon among a diversity of human neoplasms.
Future studies are required to confirm the relationship between thioredoxin overexpression, increased gastric cancer proliferation and increased cell survival. Our newly developed ability to simultaneously perform combined TUNEL and IHC assays on a single tissue section should allow more precise definition of the relationship of thioredoxin to cell proliferation or cell death, since the phenotype of individual apoptotic or proliferative cells may now be discerned by these double labelled assays.
Finally, we anticipate correlative clinical studies to relate thioredoxin expression, Ki67 and apoptosis index to pathogenic grade, response to chemotherapy, disease free survival or overall survival thus defining the impact of thioredoxin on human carcinomas. Our patient data on the SWOG 9008 (Intergroup 0116) study of high risk gastric carcinomas which is now ongoing and has accrued 486 gastric carcinoma patients would seem to be the ideal patient cohort to study. Now that we have developed paraffin-based assays for thioredoxin, Ki67 and Tdt apoptosis by the Tunel assay in a standardized optimized manner, the full clinical study testing the clinical impact of thioredoxin is now feasible.
S
Table 1. Gastrointestinal Carcinoma Staining Results
Case # Thiored o in i-67 Tunel
Tumor Normal* Tumor Tumor
1. ++++ -H- +-M- +
2. +++ ++ +++ +
3. ++++ -H- NE +
4. ++++ ++ ++++ +
5. +++ ++ +++ ++
6. 0 ++ + ++++
7. + ++ + +++
8. 0 ++ + ++++
9. ++ ++ +++ +++
10. ++ ++ NE ++
NE : = not evaluable
* = gastric pits
"S
1. Holmgren, A. 1989. Thioredoxin and glutaredoxin systems. J Biol Chem, 264:13963-13966.
2. Luthman, M., Holmgren, A. 1982. Rat liver thioredoxin and thioredoxin reductase: purification and characterization. Biochem, 21:6628-6633.
3.Gasdaska, P.Y., Oblong, J.E., Cotgreave, I.A., Powis, G. The predicted amino acid sequence of human thioredoxin is identical to that of the autocrine growth factor human adult T-cell derived factor (ADF): Thioredoxin mRNA is elevated in some human tumors. Biochim Biophys Acta, 1218:292-296, 1994.
4.Laurent, T.C., Moore, E.C., Reichard, P. Enzymatic synthesis of deoxyribonucleotides VI. Isolation and characterization of thioredoxin, the hydrogen donor from Escherichia coli B. I Biol Chem, 239:3436-3444 1964.
5. Matthews, J.R., Wakasugi, N., Virelizier, J-L., Yodoi, J., Hay, R.T. Thioredoxin regulates the DNA binding activity of NF-kappaB by reduction of a disulphide bond involving cysteine 62. Nucl Acids Res, 20:3821-3830, 1992. ό.Cromlish, J.A., Roeder, R.G. Human transcription factor I]IC (TFRIC). Purification, polypeptide structure, and the involvement of thiol groups in specific DNA binding. J Biol Chem, 264:18100-18109, 1989.
7.Bannister, A.J., Cook, A., Kouzarides, T. In vitro DNA binding activity of Fos/Jun and BZLF1 but not C/EBP is affected by redox changes. Oncogene, 6:1243-1250, 1991. δ.Grippo, J.F., Tienrungroj, W., Dahmer, M.K., Housley, P.R., Pratt, W.B. Evidence that the endogenous heat-stable glucocorticoid receptor-activating factor is thioredoxin. J Biol Chem, 258:13658-13664, 1983.
9.Wakasugi, N., Tagaya, Y., Wakasugi, A., Nfitsui, M., Maeda, M., Yodoi, J., Tursz, T. Adult T-cell leukemia-derived factor/Thioredoxin produced by both human T-lymphotropic virus type and Epstein-Barr virus- transformed lymphocytes, acts as an autocrine growth factor and synergized with interleukin-1 and interleukin-2. Proc Natl Acad Sci USA, 87:8282-8286, 1990.
10. Yodoi, J., Tursz, T. ADF, a growth-promoting factor derived from adult T cell leukemia and homologous to thioredoxin: Involvement in lymphocyte immo tion by HTLV-1 and
EBV. Adv Cancer Res, 57:381-411, 1991. l Powis, G., Oblong, J.E., Gasdaska, P.Y., Berggren, M., Hill, S., Kirkpatrick, D.L. The thioredoxin thioredoxin reductase redox system and control of cell growth. Oncol Res, 6:539-544, 1994.
12.Oblong, J.E., Berggren, M., Gasdaska, P.Y., Powis, G. Site-directed mutagenesis of active site cysteines in human thioredoxin produces competitive inhibitors of human thioredoxin reductase and elimination of mitogenic properties of thioredoxin. J Biol Chem, 269: 11714-11720, 1994.
13.Gasdaska, J.R., Berggren, M., Powis, G. Cell growth stimulation by the redox protein thioredoxin occurs by a novel helper mechanism. Cell Growth Differ, 6:1643-1650, 1995.
14. Kitaoka, Y., Sorachi, K., Nakamura, H., Masutani, R, Mitsui, A., Kobayashi, F., Mori, T., Yodoi, J. Detection of adult T-cell leukemia-derived factor/human thioredoxin in human serum. Immunol Lett 41:155-161, 1994.
15. Rubartelli, A., Bajetto, A., Allavena, G., Wollman, E., Sitia, R. Secretion of thioredoxin by normal and neoplastic cells through a leaderless secretory pathway. J Biol Chem 267:24161-24164, 1992.
16. Gallegos, A., Gasdaska, J.R., Goodman, D., Gasdaska, P.Y., Berggren, M., Briehl, M.M., Powis, G. Transfection with human thioredoxin increases cell proliferation and a dominant negative mutant thioredoxin reverses the transformed phenotype of breast cancer cells. Cancer Res, 56: 5765-5770 1996.
17. Baker, A., Briehl, M., Payne, C, Powis, G. Thioredoxin, a gene found overexpressed in human cancer, inhibitis apoptosis in vitro and in Vivo. Cancer Research, (in press) 1996.
18. McDonnell, T.J., Korsmeyer, S.J. Progression from lymphoid hyperplasia to high-grade malignant lymphoma in mice transgenic for the t(14; 18). Nature, 349:254-256, 1991.
19. Berggren, M., Gallegos, A., Gasdaska, J.R., Gasdaska, P.Y., Warneke, J., Powis, G. Thioredoxin and thioredoxin reductase gene expression in human tumors and cell lines, and the effects of serum stimulation and hypoxia. Anticancer Res 16: 3459-3466, 1996.
20. Grogan, T.M., Casey, T.T., Miller,T.P., et al. Automation of immunohistochemistry, In Weinstein RS (ed): Advances in Pathology and Laboratory Medicine, vol 6. St. Louis, Mosby, 1993, pp 253-283.
21. Grogan, T.M., Rangel, C, Rimsza, L., Bellamy, W., Martel, R., McDaniel, D., McGraw, B., Richards, W., Richter, L., Rodgers, P., Rybski, J., Showalter, W., Vela, E., Zeheb, R. Kinetic-mode, automated double-labeled immunohistochemistry and in situ hybridization in diagnostic pathology. Advances in Pathology and Laboratory Medicine, 8:79-100, 1995.
22. Miller, T., Grogan, T., Dahlberg, S., et al. Prognostic significance of the Ki67 associated proliferation antigen in aggressive non-Hodgkin's lymphomas: A prospective Southwest Oncology Group Trial. Blood, 83:1460-1466, 1994.
23. Gavrieli, Y., Sherman, Y., Ben-Sasson, S.A. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J of Cell Biology, 119:493-501, 1992.
24. Grasl-Kraupp, B., Ruttkay-Nedecky, B., Koudelka, H., Bukowska, K., Bursch, W., Schulte-Hermann, R. In situ detection of fragmented DNA (TUNEL assay) fails to discriminate among apoptosis, necrosis, and autolytic cell death: A cautionary note. Hepatology, 21:1465- 1468, 1995. <?6
25. Fujii, S., Nanbu, Y., Nonogald, H., Konishi, 1., Mori, T., Masutani, H., Yodoi, J. Coexpression of adult T-cell leukemia-derived factor, a human thioredoxin homologue, and human papillomavirus DNA in neoplastic cervical squamous epithelium. Cancer 68:1583-1591, 1991.
26. Kawahara, N., Tanaka, T., Yokomizo, A., Nanri, H., Ono, M., Wada, M., Kohno, K., Takenaka, K., Sugimachi, K., Kuwano, M. Enhanced coexpression of thioredoxin and high mobility group protein 1 genes in human hepatocellular carcinoma and the possible association with decreased sensitivity to cisplatin. Cancer Res, 56:5330-5333, 1996.
? ^
Figure Legends
1. Thioredoxin positive gastric carcinoma (Case 2)
A. The hematoxylin and eosin stains (left, upper and lower panels) reveal a pleomorphic carcinoma invading the gastric wall.
B. The thioredoxin expression (right upper and lower panels) is present in both the nuclei and cytoplasm of tumor cells in malignant glands and in rare associated leucocytes. Thioredoxin expression is absent in the adjacent stroma (lOOx to 400x).
2. Thioredoxin negative gastric carcinoma (Case 8)
A. Upper left: gastric carcinoma with complex glands in lower field of view with overlying normal gastric mucosa and submucosa. Hematoxylin and eosin (lOOx).
B. Lower left: same section as (A), negative control stained after biotin-avidin block to eliminate biotin receptor affect and stained with substituted irrelevant isotype matched monoclonal antibody (lOOx, Diaminobenzidine).
C. Upper right: same section as (A) and (B) showing faint reactivity in normal upper mucosa (+), moderate reactivity in the submucosa (++) and the underlying gastric carcinoma appears negative for thioredoxin (0) (lOOx).
D. Lower right: higher magnification of (C) showing detail with gastric pit cells having both nuclear and cytoplasmic stain and absent tumor staining (250x).
3. Normal gastric mucosa - thioredoxin and Ki67 (proliferation antigen) expression
A. Upper left: normal gastric mucosa and gastric pits with underlying muscularis propria hematoxylin and eosin (lOOx).
B. Upper right: same section as (A) stained for thioredoxin with faint (+) mucosal staining and moderate (++) gastric pit staining (lOOx).
C. Lower left: same section as (B) at higher magnification. Note the faint mucosal staining is solely cytoplasmic, while lower lying gastric pit cells are both cytoplasmic and nuclear (250x).
D. Lower right: nuclear Ki67 expression notable in lower mucosa and upper gastric pits
(250x).
4. Thioredoxin intense gastric carcinoma related to strong proliferation and weak apoptosis
(Case 4)
A. Upper left: Complex adenocarcinoma cell glands (400x, hematoxylin and eosin)
B. Upper right: Intense thioredoxin expression in gastric carcinoma cells (400x)
C. Lower left: A high percentage of Ki67 positive cells indicating high proliferation
(400x).
D. Lower right: A rare Tdt+ - apoptotic cell indicating weak apoptosis (Tunel assay,
400x, same slide).
5. Thioredoxin negative gastric carcinoma related to weak proliferation and strong apoptosis (Case 6)
A. Upper left: Complex adenocarcinoma glands (400x, hematoxylin and eosin).
B. Upper right: Absent thioredoxin expression (400x).
C. Lower left: A low percentage of Ki67 positive cells indicating a low proliferative rate
(400x). er right: A very high Tdt+ apoptotic cell rate indicating strong apoptosis (TUNEL assay, 400x, same slide).
"«5>
Figure imgf000091_0001
Figure imgf000091_0002
Figure imgf000091_0003
^
Figure imgf000092_0001
^l
Figure imgf000093_0001
Figure imgf000093_0002
Figure imgf000093_0003
Figure imgf000093_0004
All of the various publications cited above are hereby incorporated by
reference in their entireties.

Claims

I CLAIM:
1. A method of inhibiting tumor cell growth in a tumor cell that over-expresses
thioredoxin comprising contacting said tumor cell with a cell growth inhibiting effective
amount of an inhibitor of thioredoxin expression.
2. A method of reducing inhibition of apoptosis in tumor cells that over-express
thioredoxin comprising contacting said tumor cells with an effective amount of an agent that
inhibits thioredoxin.
3. A method of identifying an agent that inhibits tumor cell growth in cells that
over-express thioredoxin comprising
measuring thioredoxin expression in a first sample of said cells;
contacting a second sample of said cells with an agent to be tested;
measuring expression of thioredoxin in said second sample;
comparing expression of thioredoxin in said first sample and said second
sample;
whereby a decrease in expression of thioredoxin in said second sample is
indicative of an agent that inhibits tumor cell growth.
4. A method of identifying an agent that reduces inhibition of apoptosis in a
tumor cell that over-expresses thioredoxin comprising
measuring thioredoxin expression in a first sample of said cells;
contacting a second sample of said cells with an agent to be tested; measuring expression of thioredoxin in said second sample;
comparing expression of thioredoxin in said first sample and said second
sample;
whereby a decrease in expression of thioredoxin in said second sample is
indicative of an agent that reduces inhibition of apoptosis.
5. A method of identifying an agent that reduces inhibition of apoptosis in a
tumor cell growth.
6. A method of stimulating cell growth comprising introducing a nucleic acid
encoding a human thioredoxin having Ser at amino acid reside 73 under conditions whereby
said nucleic acid is expressed.
7. A composition comprising an agent that is useful in reducing or eliminating
thioredoxin-associated apoptosis inhibition and an acceptable carrier.
8. A composition comprising an agent that is useful in inhibiting thioredoxin
stimulated cell growth and an acceptable carrier.
PCT/US1997/022292 1996-12-06 1997-12-05 Uses of thioredoxin WO1998024472A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP97952292A EP0946192A1 (en) 1996-12-06 1997-12-05 Uses of thioredoxin
CA002274116A CA2274116C (en) 1996-12-06 1997-12-05 Use, of an inhibitor of thioredoxin, for inhibiting tumor growth
US12/694,577 US20100166724A1 (en) 1996-12-06 2010-01-27 Uses of thioredoxin

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US3199596P 1996-12-06 1996-12-06
US60/031,995 1996-12-06

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US09319292 A-371-Of-International 1999-06-03
US09/875,578 Continuation US6689775B2 (en) 1996-12-06 2001-06-06 Uses of thioredoxin

Publications (2)

Publication Number Publication Date
WO1998024472A1 true WO1998024472A1 (en) 1998-06-11
WO1998024472A9 WO1998024472A9 (en) 1999-02-04

Family

ID=21862543

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1997/022292 WO1998024472A1 (en) 1996-12-06 1997-12-05 Uses of thioredoxin

Country Status (3)

Country Link
EP (1) EP0946192A1 (en)
CA (1) CA2274116C (en)
WO (1) WO1998024472A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6372772B1 (en) 1997-08-01 2002-04-16 Prolx Pharmaceuticals Corporation Inhibitors of redox signaling and methods of using same
US6552060B1 (en) 1997-08-11 2003-04-22 Prolx Pharmaceuticals, Inc. Asymmetric disulfides and methods of using same
US6566514B1 (en) 1998-01-30 2003-05-20 Genesense Technologies Inc. Oligonucleotide sequences complementary to thioredoxin or thioredoxin reductase genes and methods of using same to modulate cell growth
WO2004013283A2 (en) * 2002-08-02 2004-02-12 University Of Rochester Thioredoxin mutants and uses thereof

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1991004320A1 (en) * 1989-09-12 1991-04-04 Aktiebolaget Astra The use of thioredoxin in the treatment of malignantly transformed cells in animals and man

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1991004320A1 (en) * 1989-09-12 1991-04-04 Aktiebolaget Astra The use of thioredoxin in the treatment of malignantly transformed cells in animals and man

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
BAKER A ET AL: "Thioredoxin, a gene found overexpressed in human cancer, inhibits apoptosis in vitro and in vivo.", CANCER RESEARCH 57 (22). 1997. 5162-5167, XP002062161 *
BERGGREN M ET AL: "Thioredoxin and thioredoxin reductase gene expression in human tumors and cell lines, and the effects of serum stimulation and hypoxia.", ANTICANCER RESEARCH, (1996 NOV-DEC) 16 (6B) 3459-66, XP002062159 *
GALLEGOS A ET AL: "Transfection with human thioredoxin increases cell proliferation and a dominant-negative mutant thioredoxin reverses the transformed phenotype of human breast cancer cells.", CANCER RESEARCH 56 (24). 1996. 5765-5770, XP002062160 *
GASDASKA J R ET AL: "Cell growth stimulation by the redox protein thioredoxin occurs by a novel helper mechanism.", CELL GROWTH AND DIFFERENTIATION, (1995 DEC) 6 (12) 1643-50, XP002062158 *
POWIS G ET AL: "Selenium and the thioredoxin redox system: Effects on cell growth and death.", ONCOLOGY RESEARCH 9 (6-7). 1997. 303-312, XP002062162 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6372772B1 (en) 1997-08-01 2002-04-16 Prolx Pharmaceuticals Corporation Inhibitors of redox signaling and methods of using same
US6552060B1 (en) 1997-08-11 2003-04-22 Prolx Pharmaceuticals, Inc. Asymmetric disulfides and methods of using same
US6566514B1 (en) 1998-01-30 2003-05-20 Genesense Technologies Inc. Oligonucleotide sequences complementary to thioredoxin or thioredoxin reductase genes and methods of using same to modulate cell growth
WO2004013283A2 (en) * 2002-08-02 2004-02-12 University Of Rochester Thioredoxin mutants and uses thereof
WO2004013283A3 (en) * 2002-08-02 2005-08-11 Univ Rochester Thioredoxin mutants and uses thereof

Also Published As

Publication number Publication date
CA2274116A1 (en) 1998-06-11
CA2274116C (en) 2009-05-05
EP0946192A1 (en) 1999-10-06

Similar Documents

Publication Publication Date Title
Gasdaska et al. The predicted amino acid sequence of human thioredoxin is identical to that of the autocrine growth factor human adult T-cell derived factor (ADF): thioredoxin mRNA is elevated in some human tumors
US6737523B1 (en) Nucleic acids comprising regions of the rat PEG-3 promoter that display elevated expression in human cancer cells and uses thereof
US20100166724A1 (en) Uses of thioredoxin
Leff et al. In VivoL-DOPA Production by Genetically Modified Primary Rat Fibroblast or 9L Gliosarcoma Cell Grafts via Coexpression of GTPcyclohydrolase I with Tyrosine Hydroxylase
Tonello et al. Role of sympathetic activity in controlling the expression of vascular endothelial growth factor in brown fat cells of lean and genetically obese rats
Miyashiro et al. Midkine expression in human breast cancers: expression of truncated form
Gasdaska et al. Human thioredoxin reductase gene localization to chromosomal position 12q23–q24. 1 and mRNA distribution in human tissue
Pai et al. Helicobacter pylori culture supernatant interferes with epidermal growth factor-activated signal transduction in human gastric KATO III cells.
US7018631B1 (en) Compositions and methods for sensitizing and inhibiting growth of human tumor cells
WO1998024472A1 (en) Uses of thioredoxin
Guo et al. Effect of regulated expression of the fragile histidine traid gene on cell cycle and proliferation
JP2009544305A (en) Recombinant mycobacterial strain expressing mycobacterial FAP protein under the control of a promoter active under hypoxia and its application for cancer therapy
WO1998024472A9 (en) Uses of thioredoxin
EP1539189A2 (en) Method of using anti-apoptotic factors in gene expression
Watanabe et al. Analysis of protooncogene c-erbB-2 in benign and malignant human prostate
Cook et al. Differential immunohistochemical localization of xanthine oxidase in normal and neoplastic human breast epithelium
Kruse et al. Production of trypsin by cells of the exocrine pancreas is paralleled by the expression of the KH protein vigilin
KR102282107B1 (en) Vector library for yeast two-hybrid screening and method for identifying deubiquitinating enzyme bound to target proteins
EP1309603A1 (en) Nucleic acids comprising regions of the rat peg-3 promoter and uses thereof
KR20180056275A (en) Immunomodulator using CD200 gene and use thereof
KR100489188B1 (en) A Method for Detecting Senescent Cells, a Compostion for Inducing a Senescent State of Cells and a Method for Inducing a Senescent State of Cells Using Caveolin
Lim et al. Increased drug resistance following retroviral gene transfer of a chimeric P-enolipyruvate carboxykinase (GTIP)-bacterial O 6 alkylguanine-DNA alkyltransferase gene into NRK cells
AU677589B2 (en) Methods and compositions for increasing the sensitivity of acell to a DNA damaging agent
KR20050042162A (en) Gene therapy for critical limb ischemia with wild type or mutant enos
US6982148B2 (en) Progressive elevated gene-3 (PEG-3) induces aggressive cancer phenotype and regulates angiogenesis

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AL AT BA BB BG BR CA CH CU CZ EE GE HU ID IL IS JP KP KR LC LK LR LT LV MD MG MK MN MX NO NZ PL RO SG SI SK SL TR TT UA US UZ VN YU AM AZ BY KG KZ MD RU TJ TM

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH KE LS MW SD SZ UG ZW AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN ML MR NE SN TD TG

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
COP Corrected version of pamphlet

Free format text: PAGES 1-92, DESCRIPTION, REPLACED BY CORRECT PAGES 1-87; PAGES 93 AND 94, CLAIMS, REPLACED BY CORRECT PAGES 88 AND 89; PAGES 1/5-5/5, DRAWINGS, ADDED

WWE Wipo information: entry into national phase

Ref document number: 09319292

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2274116

Country of ref document: CA

Ref country code: CA

Ref document number: 2274116

Kind code of ref document: A

Format of ref document f/p: F

WWE Wipo information: entry into national phase

Ref document number: 1997952292

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1997952292

Country of ref document: EP

WWW Wipo information: withdrawn in national office

Ref document number: 1997952292

Country of ref document: EP