WO2008030278A2 - Assays for radiation exposure - Google Patents

Abstract

The present invention provides a method of detecting or determining, and the use of a means of detecting or determining, if a biological sample has been exposed to ionizing radiation. The invention is carried out by detecting a change in expression of at least one preselected gene in a biological sample, the change in expression indicating that the biological sample, or a subject from which the biological sample has been collected, has been exposed to a ionizing radiation.

Description

Assays for Radiation Exposure

Derek L. Stirewalt

This invention was made with Government support under Grant No. U19-AI06777 from the National Institutes of Health. The US Government has certain rights to this invention.

Field of the Invention

The present invention concerns methods for detecting if a biological sample, or a subject from which a biological sample has been collected, has been exposed to a dosage of ionizing radiation.

Background of the Invention

Ionizing radiation has medical, industrial and military uses. Although ionizing radiation can be used in the therapy of diseases such as cancer, exposure to biologically significant levels of such radiation can also cause genotoxic stress. Similarly, many industrial processes (such as the production of nuclear power) and military uses (such as nuclear weapons) can expose individuals to hazardous levels of ionizing radiation. Such radiation can elicit a variety of cellular responses, ranging from cell-cycle arrest to mutation, malignant transformation, or cell death. Many of the responses (such as genotoxicity) are often subtle, and exposed persons may be unaware or unsure if they have been exposed. Moreover, it may require years to evince an untoward effect (such as the development of a malignancy) caused by the exposure. See, A. Fornace et al., Method for Detecting Radiation Exposure, US Patent No. 7,008,768 (US DHHS).

Summary of the Invention

The present invention provides a method of detecting or determining if a biological sample has been exposed to ionizing radiation {e.g., at a dosage or amount of at least 1 Sievert or 100 REM, where "REM" means "Rδntgen equivalent man"). The invention is carried out by detecting a change in expression of at least one preselected gene (e.g., a gene of Table 1-3) in a biological sample, the change in expression indicating that the biological sample has been exposed to ionizing radiation.

In some embodiments the biological sample is collected from a subject suspected of receiving a dosage of ionizing radiation, and the change in expression indicates that the subject has been exposed to a dosage of ionizing radiation.

In some embodiments the biological sample is a blood sample, such as a T-cell containing blood sample. In some embodiments the biological sample is whole blood.

In some embodiments the change in expression is determined by polymerase chain reaction (or "PCR"), particularly by reverse-transcriptase PCR (or "RT-PCR"). In some embodiments RT-PCR is performed directly on the biological sample (e.g., directly on whole blood) without intervening separation steps.

In some embodiments the change in expression is detected quantitatively so that the radiation dosage may be determined quantitatively (e.g., by comparing the change in expression in the biological sample to changes in expression observed in control samples that have received a known dosage of radiation). In some such embodiments the radiation dosage may be determined to an accuracy of about 2, 1, or 0.5 Sieverts (that is, the embodiment of the method can distinguish between dosages that differ by 2, 1 or 0.5 Sieverts).

A further aspect of the present invention is a kit useful for carrying out a method as described herein.

A futher aspect of the present invention is the use of a means of detecting a change in expression of at least one preselected gene (e.g., a gene of Table 1-3) in a biological sample (such as a biological sample from a subject), in determining if said sample (or determining if the subject from which said sample is taken) has been exposed to a dosage of ionizing radiation.

The present invention is explained in greater detail in the drawings and specifications set forth below. The disclosures of all United States patent references cited herein are to be incorporated by reference herein in their entirety.

Brief Description of the Drawings

Figure 1. Differentiating between two Contiguous γ-irradiation Doses and Cell Viability with respect to Radiation Doses. A. Shows the number of probe sets (log scale, y-axis) that display significant expression differences between two doses (x-axis) for the following two group comparisons: 0 vs. all γ-irradiation doses (>0); 0 - 0.15 vs. 2 - 12 (>0.15); 0 -2 vs. 4 -12 (> 2); 0 - 4 vs. 6 -12 (>4); 0 - 6 vs. 9 -12 (>6); and 0 - 9 vs. 12 (>9). B. Shows how the viability of T-cells (y-axis) decrease in dose-dependent manner with time (x-axis) after the following specific doses of γ-irradiation: no γ-irradiation (0, blue), 0.15 Gy (pink), 2 Gy (yellow), 4 Gy (bright blue), 6 Gy (purple), 9 Gy (brown), and 12 Gy (green).

Figure 2. Linear Dose-dependent Radiation Responses in T-cells. A. Shows an example of a gene (PDHB) that displays a decrease in expression (y-axis, log₂ expression) in a dose-dependent manner 24 hours after radiation exposure (x-axis, Gy). Ages of the subjects are provided (right upper comer): 30 y/o (red); 39 y/o (green); 43 y/o (dark blue); 55 y/o (light blue); and 61 y/o (orange). The gender of the subjects is not shown, but there were 2 males (ages: 39, 55) and 3 females (ages: 30, 39, 61). B. Shows an example of a gene (RAP2B) that displays an age- and dose-dependent expression change 24 hours after radiation exposure.

Figure 3. Time Dependent Nature and Overlap of linear Expression Changes in T-cells. A. Shows the number of genes (x-axis) with linear expression changes as a factor of the amount of time that has lapsed after γ-irradiation exposure (y-axis). B. Shows the number of genes with linear expression changes post γ-irradiation that overlap between the three time points (3 hrs, blue; 8 hrs, yellow; and 24 hrs, green). Not surprising, the 8 and 24 hour time points had the most overlapping genes (N = 77), which is probably due to the relative small number of linear expression differences identified at the 3 hour time point.

Figure 4. Probe sets Differentiating between two Contiguous γ-irradiation Doses and Cell Viability with Respect to Radiation Doses. A. Shows the number of probe sets (log scale, y-axis) that display significant expression differences between two doses (x-axis) for the following two group comparisons: 0 vs. all γ-irradiation doses (>0); 0 - 0.15 vs. 2 - 12 (>0.15); 0 -2 vs. 4 -12 (> 2); 0 - 4 vs. 6 -12 (>4); 0 - 6 vs. 9 -12 (>6); and 0 - 9 vs. 12 (>9). B. Shows how the viability of white cells from whole blood (y-axis) decrease in dose- dependent manner with time (x-axis) after the following specific doses of γ-irradiation: no γ- irradiation (0, blue), 0.15 Gy (pink), 2 Gy (yellow), 4 Gy (bright blue), 6 Gy (purple), 9 Gy (brown), and 12 Gy (green).

Figure 5. RASGRP2 Expression at 8 and 24 hours after Radiation Exposure. A. Compares the log₂ expression of RASGRP2 (y-axis) between white blood cells (WBCs) from subjects without in vitro exposure to radiation (N = 5 subjects/5 samples, red) and those WBCs at 8 hours after in vitro exposure to γ-irradiation (N = 5 subjects/83 samples, green). WBCs exposed to irradiation received a wide range of doses (0.15 - 12 Gy). B. Compares the log₂ expression of RASGRP2 (y-axis) between white blood cells (WBC) from the same subjects without exposure to radiation (N = 5 subjects/5 samples, red) and 24 hours after in vitro exposure to γ-irradiation (N = 5 subjects/83 samples, green) at the all doses > 0.15 Gy. For the microarray studies, some baseline samples were collected at time 0 and 24 hours later to control for possible time-dependent effect.

Figure 6. DHRS4L2 Expression at 24 hours after Radiation Exposure. A. Compares the log₂ expression of DHRS4L2 (y-axis) between white blood cells (WBCs) that received 0 - 4 Gy (red) vs. 6 - 12 Gy (green). B. Shows the Iog₂ expression of DHRS4L2 (y- axis) at each dose level of γ-irradiation: no γ-irradiation (0, red), 0.15 Gy (green), 2 Gy (dark blue), 4 Gy (light blue), 6 Gy (orange), 9 Gy (pink), and 12 Gy (brown).

Figure 7. Time Dependent Nature and Overlap of Linear Expression Changes in Whole Blood Cells. Shows the number of genes with linear dose-dependent responses that are significant (NFD <1) at the different time points after radiation exposure. There are 89 genes at 3 hours after exposure, 97 genes at 8 hours after exposure, and 178 genes at 24 hours after exposure. The number of genes common to two or more time points are also shown. 23 genes overlap at both the 3 and 8 hour time points. 27 genes overlap at the 8 and 24 hour time points and 2 gene overlap at 3 and 24 hour time points. 2 genes (CDKNlA and CABIN) overlap at all three time points.

Figure 8. Linear Age- and Dose-dependent Radiation Responses in WBCs from Whole Blood. Shows an example of a gene (ASNS) that displays an age- and dose- dependent expression change 24 hours after radiation exposure. Ages of the subjects are provided (right upper comer): 30 y/o (red); 39 y/o (green); 43 y/o (dark blue); 55 y/o (light blue); and 61 y/o (orange). The expression of ASNS (x-axis, Iog2 scale) decreases with dose (y -axis) but this expression is significantly associated with the age of the subject (upper right corner): 30 y/o (red); 39 y/o (green); 43 y/o (dark blue); 55 y/o (light blue); and 61 y/o (orange).

Figure 9. Quantitative RT/PCR Assay for FDXR. Shows the average expression of FDXR (y-axis) in RNA from whole blood from 5 subjects as the dose of γ-irradiation increases (y-axis). FDXR expression is corrected for RNA integrity using B₂M expression. The ranges of FDXR expression for each radiation group are also shown (arrow bars). As found in the microarray studies, there was no overlap in the expression between no irradiation (0) and any of the other radiation doses (0 vs. any treatment, p < 0.00001). In addition, FDXR expression was able to discriminate between no irradiation and even the lowest dose irradiation (i.e. 0.15 Gy, p =.008).

Figure 10. Standard and Quantitative RT/PCR Assays of CDKNlA for Canine Model. A. Shows the RT/PCR product for CDKNlA in four canine RNA samples (Sl, S2, S3, S4). No template (H₂O) lane is negative for amplification. B. Shows expression of CDKNlA (y-axis) in RNA from whole blood from 2 dogs prior to radiation exposure (x axis, No Tx) and 24 hours after 5 Gy of radiation exposure (x-axis, 5 Gy). The expression value was adjusted for RNA integrity using dog G3PDH standard. As found in the microarray human studies, there was evidence of increased expression of CDKNlA after in vivo radiation exposure.

Figure 11. DDB2 Dose Dependent Expression Response at 8 and 24 hours after Radiation Exposure. Shows the Iog2 expression (y-axis) change for DDB2 for 10 patients after varying radiation doses (x-axis) at 8 hours (panel A) and 24 hours (panel B). As demonstrated by the figures, DDB2 alone has fairly good power to discriminate between 0 radiation exposure, 0.15 Gy and higher levels of exposure.

Figure 12. PCNA and RASGRP2 Dose Dependent Expression Response after Radiation Exposure. A. Shows the Iog2 expression of PCNA at various doses of radiation. PCNA expression increases with increasing doses of radiation exposure. B. Shows the Iog2 expression of RASGRP2 at various doses of radiation. RASGRP2 expression decreases with increasing doses of radiation exposure.

Figure 13. Algorithm to Determine Radiation Exposure Levels Based on Dose Dependent Expression Changes for a subset of 11 Genes. Shows an algorithm using subset of 11 genes that can discriminate between various levels of radiation exposure (no γ- irradiation, 0.15 Gy, 2 Gy - 9 Gy, and 12 Gy) at 24 hours after exposure. The 11 genes used in this algorithm included: CDKNlA, FDXR, DDB2, RPS27L, PCNA, XPC, CRSP7, CD79A, GPR89A, IGL@, and CABINl.

Figure 14. Algorithm to Determine Exposure to Radiation Based on Dose Dependent Expression Changes for a subset of 5 Genes. Shows an algorithm using subset of 5 genes that can discriminate between no radiation exposure and some level of radiation exposure at 8 hours (panel A) and at 12 hours (panel B) after exposure. The genes used for this model included: CDKNlA, FDXR, DDB2, RPS27L, and PCNA. Detailed Description of the Preferred Embodiments

"Subject" as used herein include both human subjects and animal subjects (particularly mammalian subjects such as dogs, cats, cattle, sheep, pigs, horses, monkeys, mice, rats, etc.) for veterinary medical purposes. Also, animals can be utilized as an indicator of human exposure where the animals received a like dosage as humans, and as an indicator of exposure in a region or physical location of interest.

"Ionizing radiation" as used herein may be from any natural or non-natural source (the latter as encountered in, for example, medical and industrial applications). Examples of ionizing radiation include but are not limited to alpha, beta, gamma rays, x-rays, particle bombardment from accelerators, electron beams, nuclear decay, and combinations thereof thereof. Doses of ionizing radiation may be "acute" in that they occur in a specific natural or non-natural environment in which the subject or sample is present (e.g., for from one minute to one hour; from one hour to one day; from one day to one week), and from which the subject or sample may be removed. In some embodiments the dose of ionizing radition is not "background" or natural environmental radiation, although in some embodiments (such as suspected high level of radon gas contamination in a building or mine) the invention may be utilized to detect acute exposure to high levels of "background" radiation.

"Biological sample" as used herein maybe taken or collected from any type of subject as given above. A "biological sample" is generally a cell or tissue sample that contains proteins or nucleic acids from which gene expression can be determined. Biological samples may be of any cell or tissue type, including but not limited to skin, muscle, bone, nerve, milk (e.g., in the case of dairy cattle), and blood. In some embodiments a blood sample is preferred.

"Blood sample" as used herein may be whole blood or any fragment thereof, such as blood platelets, white blood cells or T-cells, red blood cells, blood plasma, and combinations thereof. In some embodiments whole blood is preferred; in some embodiments a blood sample that contains T-cells is preferred (hence encompassing both whole blood and blood fractions that contain T-cells).

Any suitable biological sample can be utilized to carry out the present invention, with a preferred embodiment being blood samples and particularly whole blood. The sample can be collected from a subject (e.g., a human or mammalian subject) by any suitable means, such as swabbing, scraping, lance, needle, syringe, etc., depending upon the particular type of sample being collected.

The present invention can be carried out in accordance with known techniques (including but not limited to those described in US Patent No. 7,008,768), or variations thereof that will be apparent to those skilled in the art given the present disclosure. The disclosures of all United States patent references cited herein are to be incorporated by reference herein in their entirety.

As noted above, the present invention provides a method of detecting if a biological sample has been exposed to a dosage of ionizing radiation. In general, the method involves detecting a change in expression of at least one (e.g., one, two, three, four, five, or more) preselected gene (such as disclosed in Table 1-3 below) in a biological sample. In some embodiments the change in expression is determined by comparison to expression seen for that gene in a like or "control" biological sample that has not been exposed to ionizing radiation). A change in expression indicates that the biological sample (or a subject from which the biological sample has been collected) has been exposed to a dosage of ionizing radiation.

The change in expression may be an increase or decrease in expression, depending upon whether the gene is one which increases or decreases expression in response to ionizing radiation (particularly in the case of linearly responsive genes such as as set forth in Table 1 below) as discussed further below. In some embodiments the change in expression can be simply an activation or inactivation of expression (particularly in the case of "threshhold responsive" genes such as set forth in Table 2 below).

Determining the change in expression may be carried out directly or indirectly out by any suitable means or method. The amount of encoded protein expressed may be determined by any suitable means or method, such as by chromatography, immunoassay, isoelectric focusing, etc., including combinations thereof. The amount of encoded RNA transcribed (as an indicator of gene expression) may be detected by any suitable means or method, such as polymerase chain reaction, and particularly reverse transcriptase-polymerase chain reaction. See, e.g., US Patent Nos. 6,858,395;'6,777,210; 6,706,874. PCR primers or primer sets for the one or more genes selected for detection are generated in accordance with known techniques from the sequence of the genes themeselves.

As noted above, genes that may be used to carry out the present invention may be linearly responsive genes or threshhold dependent genes. Table 1 below describes genes that display or exhibit a radiation dose dependent expression change that is linear (including positively and negatively correlated responses) and which can be used to carry out the present invention. Table 2 below describes genes that display a radiation dose dependent expression change that is threshold dependent and which can be used to carry out the present invention. Note that a quantitative assay can be carried out by (a) detecting the amount of expression of one or more linearly responsive genes, (b) detecting the activation (or inactivation) of expression of a plurality of threshhold dependent genes which are activated (or inactivated) at different threshholds or different levels of radiation, and combinations thereof. Table 3 describes subset of genes of Tables 1-2 that can be used to carry out the present invention.

In some embodiments, the invention is capable of detecting a dose or dosage of ionizing radiation of at least 4 Seiverts (400 REM); in some embodiments the invention is capable of detecting a dose of ionizing radiation of at least 3 Seiverts (300 REM); in some embodiments, the invention is capable of detecting a dose of ionizing radiation of at least 2 Seiverts (200 REM); in some embodiments the invention is capable of detecting a dose of ionizing radiation of at least 1 Seivert (100 REM); in some embodiments, the invention is capable of detecting a dose of ionizing radiation of 0.5 Seiverts (50 REM).

In those cases where the invention is used to quantitatively determine the dose of ionizing radiation, in some embodiments the invention distinguishes between doses that differ by at least 2 Seiverts (200 REM); in some embodiments distinguishes between doses that differ by at least 1 Seivert (100 REM); and in some embodiments distinguishes between doses that differ by at least 0.5 Seiverts (50 REM).

Kits for carrying out methods of the invention may be embodied in a variety of different forms. In some embodiments the kit may simply comprise a blood collection device (such as a lance, syringe, needle, etc.) including a receptacle for storing or retaining the blood sample and instructions for forwarding the sample to a laboratory for performance of the determining methods or assays described above. In some embodiments the kit may comprise a primer or primer set for a particularly predetermined gene or genes as described herein, optionally packaged with any necessary instructions for carrying out the method. In some embodiments the kit may comprise reagents and apparatus for carrying out the methods, such as reagents and apparatus for carrying out PCR as described herein and in accordance with known techniques. Numerous variations of such kits will be readily apparent to those skilled in the art based upon the disclosure herein, and the particular application or need for which the kit is to be adapted.

203725 at GADD45A 6.28849 0.10304

213280_at GARN L4 -5.32144 0.70969

208913 at GGA2 7.63628 0.00863

214430 at GLA 6.94645 0.02977

218147 s at GLT8D1 5.25568 0.81264

202615 at GNAQ -5.39529 0.6099

201921 at GNG10 /// LOC552891 -5.94002 0.20367

202382 s at GNPDA1 5.55513 0.44036

206655 s at GP1BB /// SEPT5 -6.11877 0.14331

204137 at GPR137B 5.30509 0.73396

207651~at GPR171 5.93815 0.20443

221767 x at HDLBP 5.4329 0.56474

204689 at HHEX -7.85061 0.00595

212642~s at HIVEP2 6.53846 0.06385

211732~x at HNMT 6.37531 0.08718

204647 at HOMER3 5.49714 0.49541

201163 s at IGFBP7 -5.4703 0.52324

204627 s at ITGB3 -5.28835 0.75969

220266 s at KLF4 -5.91306 0.21484

200785 s at LRP1 5.27633 0.77876

209348 s at MAF 6.71221 0.04601

200797 s_at MCLt -5.38982 0.61676

202617 s at MECP2 5.16307 0.98429

201126 s at MGAT1 5.59183 0.40881

214770 at MSR1 5.42927 0.56895

213511 s at MTMR1 -5.35997 0.65569

220319 s at MYLIP -6.7005 0.04703

200759 x_at NFE2L1 7.72207 0.00743

204081 at NRGN -6.81892 0.0377

208690_s_at PDLIM1 -6.21016 0.11993

218019 s at PDXK 5.96819 0.19264

218319_at PELH -5.34394 0.67762

209302_at POLR2H 5.62281 0.38399

203680_at PRKAR2B -7.42661 0.01249

204054 at PTEN -6.56248 0.061

210367_s_at PTGES 6.36095 0.08962

222077 s at RACGAP1 5.74186 0.30221

217776 at RDH11 6.47699 0.07177

209568_s_at RGL1 5.31935 - 0.71274

200885_at RHOC -5.78878 0.27512

201785 at RNASE1 6.5334 0.06446

212155 at RNF187 6.74762 0.04306.

213414 s_at RPS19 8.06888 0.0041

202649 x at RPS19 5.62422 0.3829

218307 at RSAD1 -5.39282 0.61299

217728 at S100A6 -5.93273 0.20663

200832 s at SCD 5.6021 0.4004

212154_at SDC2 6.58344 0.05863

212158_at SDC2 6.43691 0.07747

204614 at SERPINB2 7.15065 0.02049

207351 s_at SH2D2A 5.2949 0.74951

217691 x at SLC16A3 5.72157 0.31476 007/01

-23-

217825 s at UBE2J1 -6.28551 0.10364

217750_s_at UBE2Z 5.2318 0.85374

200931 s at VCL -5.58158 0.41737

204787 at VSIG4 6.44944 0.07564

218753 at XKR8 -5.51484 0.47789

219059 s at XLKD1 5.5851 0.41441

37943 at ZFYVE26 5.43627 0.56087

203585 at ZNF185 -7.17817 0.0195

219123 at ZNF232 5.16846 . 0.97334

218401 s at ZNF281 -5 49206 0.50056

39891 at ZNF710 5.81461 0.2613

Z scores provide a measure of statistical significance along with the direction of expression change. A negative Z score indicates a decrease in expression after radiation exposure and positive Z scores indicate an increase in expression. The cut-off for statistically significant Z scores varies with analyses, but as a general rule a Z scores greater than 4.75 or less than -4.75 are considered statistically significant. For some analyses, it is perhaps more appropriate to determine the group of expression changes with significant expression differences based on the number of false discoveries (NFD). The NFD correlates closely with p-values, increasing in number as the chance of the gene having a false positive increases. The NFD indicates the statistical chance with a group of genes for there to be false positive. For example, if you have 20 genes and the NFD within those 20 genes increase to 1, there high statistical chance for 1 of the 20 genes to be a false finding. For these analyses, we have chosen to include all genes with an NFD of less than or equal to 1.

EXPERIMENTAL

Developing quantitative RTVPCR assays for candidate genes to be validated using an in vivo radiation dog model. We have started developing quantitative RT/PCR assays for selected genes to be validated in our in vivo dog model. Details of Results and Studies.

1) Obtain additional subjects for in vitro radiation studies. Over the past 9 months, we have enrolled and obtained blood samples from 16 additional subjects for the proposed experiments. These blood samples were obtained under an IRB-approved protocol. These additional subjects bring the total accrual on protocol 1882 to 141 enrollees. 2) Optimization of whole blood conditions for in vitro radiation experiments.

Because peripheral blood samples are drawn in heparin to prevent coagulation, we examined whether washing out the heparin from the samples and/or adding fetal bovine serum (FBS) with RPMI to cells may improve the viability of white blood cells over 24 hours. If there were significant numbers of dying or apoptotic cells at 24 hours in our no treatment control, it would introduce time-dependent expression changes, which would confound our results. To measure dying and apoptotic cells, viability studies were performed on whole blood using both flow cytometry assays with propidium iodine and manual examination with trypan blue staining. We found that the viability of whole blood samples remained highest if the cells were left in the heparin with the subject's serum rather than trying to wash out the heparin or add other media. The unprocessed WBC had a viability of approximately 97% after 24 hrs, while other combinations of washing out the heparin and adding media +/- FBS reduced the viability to < 90% after 24 hours in most cases.

3) Microarray studies examining dose-dependent expression changes in hematopoietic cells. Microarrays were performed as per Affymetrix's standard protocol using 5 micrograms of total RNA (Affymextrix, 3420 Central Expressway Santa Clara, CA 95051. Customer Service 888-DNA-CHIP (888-362-2447)). Biotin-labeled target was hybridized onto Affymetrix HG-Ul 33 A arrays. Individual arrays were screened for quality as described in our original proposal. CEL files were analyzed using robust multi- array average (gcRMA), which adjusts for background, normalizes, and generates log₂ transformed values based upon the perfect match signal (Izzary, Biostatistics, 2003). These expression values were imported into GenePlus software (Enodar Biologic, Seattle, WA), and analyses were performed separately using T-cells and whole blood data. Analyses examined expression differences between defined radiation cut-offs. For example, a comparison was made between samples exposed < 4 Gy and samples exposed > 4 Gy. These analyses were designed to identify potential candidate genes that may discriminate between specific doses of γ-irradiation. Other analyses examined the expression changes as a continuous variable over all doses. These later analyses sought to identify genes that displayed a linear expression change after radiation exposure that was dose-dependent. Expression changes for probe sets were considered statistically significant if the absolute Z expression difference (Z- difference) was >4.75 or <-4.75, which has been used in other similar microarray studies (Xu, Hum MoI Genet, 2002). Of note, an individual gene may have more than one probe set (sequences from the gene on the array used to detect expression); therefore, the number of significant probe sets is usually slightly more than the number of significant genes. a) Discovery of dose-dependent expression changes in negatively selected human T- cells.

Analyses to discriminate between 2 contiguous doses of radiation. We have examined the expression changes in T-cells from 5 subjects with varying demographics (age range 30 - 61 yrs old; 2 males and 3 females). These preliminary studies identified a large number of probe sets from genes that displayed significant expression differences between contiguous doses of γ-irradiation (e.g. 0 Gy vs. > 0.15 Gy). The number of "discriminating" probe sets was highest when comparing no γ-irradiation to all other doses of γ-irradiation (>0, Figure IA). At higher doses of γ-irradiation, there was a trend for the 8 and 24 hour time points to identify more "discriminating" genes than the 3 hour time point (Figure IA). This later finding may be related to the "time dependent" nature of the molecular pathways activated by higher doses of γ-irradiation exposure. For example, the majority of cells receiving the higher doses of γ-irradiation were still alive at 3 and 8 hours after exposure, and it was not until 24 hours after exposure that a large percentage of these cells were dead (Figure IB), suggesting that transcription of some genes associated with apoptotic and death pathways takes time.

Linear dose-dependent responses. - Linear dose-dependent responses in T-cells are prevalent at 8 and 24 hours after -^-irradiation. Analyzing the T-cell microarray data for linear dose-dependent responses, few genes displayed a significant changes at the 3 hour time point (N = 9). However, at eight hours after in vitro γ-irradiation exposures, 206 genes displayed significant linear dose-dependent responses, and at the 24-hour time point, 575 genes displayed a significant linear dose-dependent response (Figure 3B), with the vast majority of significant genes (412 of 575, 72%) at the 24 hour time point exhibiting a decrease in expression. This finding may be secondary to a decrease transcription in the dying cells or reflect the biology of these dying cells.

Although most genes were not associated with the age of the subject, approximately 15% - 20% of the potential candidate genes with dose dependent responses also displayed significant age-dependent responses (Figure 2). In addition, there were many genes with potential age associated expression changes that did not quite meet statistical significance. Genes with age- and dose-dependent expression changes may not be optimal for the development of radiation dosimetry assays because their age-dependent expression changes may be more related to "functional" age than "chronological" age. If so, it would be very difficult control for functional age in any future dosimetry test. The validation and potential implications of these findings will require further investigations, and some of these findings of age associated expression changes may be due to more to inter-individual variations or even random chance. However, the potential for age-associated expression changes reiterates the importance of examining subjects across different age groups during the process of the development of any dosimetry assay that will be applied to the entire population.

- Overlap between the 8 and 24 time points for dose-dependent responses in T-cells. As previously described, the number of genes with significant linear expression changes increased with time (Figure 3A), suggesting that quantitative RT/PCR assays for dose dependent responses may be most useful if the assays are performed at least 3 hours after exposure. This finding may not be of real practical consequence, given that in the event a large-scale radiation exposure, the chance of obtaining blood on patients within 3 hours after exposure may be small. As one might predict, there were fewer genes that overlapped between the 3 hr time point and the other two time points, with (approximately 10%) genes displaying the same linear dose-dependent expression differences between the 8 and 24 hr time points (Figure 3B). Many of these "overlap" genes are known to be involved in DNA- damage repair pathways and apoptosis (e.g. FAS, GADD45A, TP53I3, and MDM2). b) Discovery of dose-dependent expression changes in whole blood using microarrays. Although T-cells may be an excellent model system for examining dose- dependent responses after in vitro radiation exposures, subproject 3 seeks to develop quantitative RT/PCR-based radiation dosimetry assays that can be utilized within 24-48 hours after in vivo radiation exposure; therefore, RT/PCR assays using whole blood would be more practical, eliminating the requirement of lymphocyte selection. The major limitation to using whole blood is that there will be inter-individual variations in the cell types that could interfere with the reliability of the assay. To begin exploring if whole blood can be used for the development of RT/PCR-based dosimetry assays for radiation exposures, we have examined the global expression changes of whole blood from 5 healthy donors after γ- irradiation. Just as with the T-cells, microarrays were performed after γ-irradiation at specific time points and after previously defined doses. These 5 subjects are the individuals used for the T-cell experiments and were chosen specifically to have varying demographics (age range 30 - 61, gender 2 males and 3 females).

Analyses to Discriminate between 2 Contiguous Doses. As with the T-cell experiments, we examined if there were genes with expression changes that differentiated between two contiguous doses of radiation. At the higher doses of radiation exposure, the overall number of probes demonstrating a statistically significant difference was somewhat lower than with a pure population of T-cells (Figure 4A). Similar to the T-cell model, there was a dose-dependent effect on the cell survival in the WB model (Figure 4B), but the magnitude of this effect was not as dramatic as found in the T-cells, which were under proliferative pressure at the time of γ-irradiation exposure and may be more sensitive to radiation than other cell types in the peripheral blood.

Almost 500 genes displayed significant expression changes between no γ-irradiation exposure and any radiation exposure. Many of these genes displayed significant expression changes at both the 8 and 24 hour time points (Figure 5A and 5B), suggesting that these genes may be excellent biomarkers for radiation dosimetry assays over an 8 — 24 hour time window after exposure.

Even within the middle doses of radiation, some genes displayed significant expression differences between two contiguous doses. For example, DHRS4L2 expression was significant different between 0 - 4 Gy versus 6 - 12 Gy groups at 24 hours after exposure (Figure 6). c) Linear dose-dependent responses are identified in whole blood cells. Similar to the T-cell experiments, linear dose-dependent expression responses after irradiation were identified at all three time points. A total of 89 genes displayed a linear dose-dependent expression change at 3 hours after exposure. At the 8 hour time point, 97 genes had significant dose-dependent expression changes. Similar to the T-cell experiments, there was an increase in the number of genes with dose-dependent expression changes as more time lapsed from the time of γ-irradiation exposure, such that at the 24 hour time point, 178 genes displayed dose-dependent expression changes (Figure 7). There was overlap among some of the genes at the different time points 23 genes overlap at both the 3 and 8 hour time points. 27 genes overlap at the 8 and 24 hour time points and 2 gene overlap at 3 and 24 hour time points. 2 genes (CDKNlA and CABIN) overlap at all three time points (Figure 7). As found in the preliminary T-cell experiments, there were also genes at each time point that displayed significant expression changes associated with age. For example, at the 24 hour time point, 4 of the 52 genes displayed significant expression changes associated with age and radiation doses (Figure 8).

4) Development of quantitative RT/PCR assays for candidate genes to be validated in the human in vitro model. We have started the process of developing human quantitative RT/PCR assays for a selected number of genes with dose-dependent expression changes after γ-irradiation. FDXR displayed excellent discrimination between no γ- irradiation and any γ-irradiation exposure in the whole blood microarray analyses at both the 8 hr (pO.OOOl) and 24 hr (p < 0.0001) time points. Quantitative primers and an ABI probe were developed for the human FDXR gene. Quantitative RT/PCR analyses found no overlap in expression between the no treatment group and the other samples treated with varying doses of γ-irradiation (Figure 9). Based on these findings, we are moving forward with the development of quantitative RT/PCR assays for this gene, creating a plasmid with FDXR that will be used to determine copy numbers in future assays. These quantitative RT/PCR assays will be performed in samples from other subjects that have not been used for the microarray studies.

5) Development of quantitative RT/PCR assays for candidate genes to be validated in the canine in vivo model. CDKNlA displayed a significant linear increase in expression at both 8 hours and 24 hours after γ-irradiation exposure in our human in vitro microarray experiments, suggesting that CDKNlA may be an excellent candidate gene to examine in our in vivo canine model. Therefore, we developed CDKNlA dog specific quantitative PCR primers. These primers showed specific amplification of the dog CDKNlA gene via standard RT/PCR assays (Figure 10A). We then developed a dog specific CDKNlA probe for a quantitative RT/PCR assays. We had access to RNA from 2 dogs at baseline and 24 hours after 5 Gy of total body irradiation. Quantitative RT/PCR analysis from these samples demonstrated an increase in expression of the canine CDKNlA, without any overlap between the two groups (Figure 10B). Despite the small number of subjects, we confirmed that quantitative RT/PCR assays can be developed for the in vivo canine model based on the human in vitro microarray data and that some of the genes identified in the in vitro model may display dose-dependent responses in the in vivo canine model. The foregoing is illustrative of the present invention, and is not to be construed as limiting thereof. The invention is defined by the following claims, with equivalents of the claims to be included therein.

Claims

THAT WHICH IS CLAIMED IS:

1. A method of detecting if a biological sample has been exposed to ionizing radiation, comprising: detecting a change in expression of at least one preselected gene in a biological sample, a change in expression indicating said biological sample has been exposed to ionizing radiation.

2. The method of claim 1, wherein said biological sample is a blood sample.

3. The method of any preceding claim, wherein said biological sample is whole blood.

4. The method of any preceding claim, wherein said biological sample is collected from a human or mammalian subject.

5. The method of any preceding claim, wherein said biological sample is collected from a subject suspected of receiving said dosage of ionizing radiation, and wherein said change in expression of said at least one preselected gene indicates said subject has been exposed to said dosage of ionizing radiation.

6. The method of any preceding claim, wherein said radiation is at a dosage is at least 1 Seivert (100 REM).

7. The method of any preceding claim, wherein said detecting is carried out by PCR, particularly RT-PCR.

8. The method of any preceding claim, wherein said detecting is carried out by chromatography on the protein encoded by said gene.

9. The method of any preceding claim, wherein said at least one preselected gene is selected from the group consisting of: CABIN1 GNLY

CD79A IGHM

CDKN1A KLRF1

CRSP7 MR1

DDB2 PLAGL2

FDXR PLSCR3

GPR89A RASGRP2

IGL@ RSAD1

PCNA SIGIRR

RPS27L TBCB

XPC TNFAIP6

ATF5 TNFSF4

ATP6V0E2 TUBA3

BIRC3 AGPAT7

CCL4 CENTB1

CLEC4E KCNQ1

CTSW S100A4

DRAM SPON2

10. The method of any preceding claim, further comprising quantitatively determining the dosage of said ionizing radiation to an accuracy of at least 1 Sievert.

11. The use of a means of detecting a change in expression of at least one preselected gene in a biological sample in determining if said sample has been exposed to ionizing radiation.

12. The use of claim 11, wherein said biological sample is a blood sample.

13. The use of any preceding claim, wherein said biological sample is whole blood.

14. The use of any preceding claim, wherein said biological sample is collected from a human or mammalian subject.

15. The use of any preceding claim, wherein said biological sample is collected from a subject suspected of receiving said ionizing radiation, and wherein said change in expression of said at least one preselected gene indicates said subject has been exposed to said ionizing radiation.

16. The use of any preceding claim, wherein said ionizing radiation is at a dosage is at least 1 Seivert (100 REM).

17. The use of any preceding claim, wherein said detecting is carried out by PCR, particularly RT-PCR.

18. The use of any preceding claim, wherein said detecting is carried out by chromatography to assess the protein encoded by said gene.

19. The use of any preceding claim, wherein said at least one preselected gene is selected from the group consisting of:

CABIN1 GNLY

CD79A IGHM

CDKN 1 A KLRF1

CRSP7 MR1

DDB2 PLAGL2

FDXR PLSCR3

GPR89A RASGRP2

IGL@ RSAD 1

PCNA SIGIRR

RPS27L TBCB

XPC TNFAI P6

ATF5 TNFSF4

ATP6V0E2 TUBA3

BIRC3 AGPAT7

CCL4 CENTB1

CLEC4E KCNQ1

CTSW S100A4

DRAM SPON2

20. The use of any preceding claim, further comprising quantitatively determining the dosage of said ionizing radiation to an accuracy of at least 1 Sievert.