WO1993025897A1

WO1993025897A1 - Diagnostic method for detection of active atherogenesis

Info

Publication number: WO1993025897A1
Application number: PCT/US1993/005520
Authority: WO
Inventors: Jan J. Opplt
Original assignee: Opplt Jan J
Priority date: 1992-06-12
Filing date: 1993-06-09
Publication date: 1993-12-23
Also published as: EP0645010A4; EP0645010A1; AU4631193A

Abstract

A method of detecting atherogenesis from a serum comprising: electrophoretically separating the lipoproteins of the serum sample on a step gradient gel; quantitating each lipoprotein; and comparing the level of each lipoprotein in the sample with the levels of each lipoprotein of individuals with active atherogenesis.

Description

Title: DIAGNOSTIC METHOD FOR DETECTION OF

ACTIVE ATHEROGENESIS

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates generally to a method of detecting active atherogenesis. More particularly, this invention relates to a diagnostic method for the reliable detection of active atherogenesis involving the separation and quantification of atherogenic and other lipoproteins.

2. Description of Related Art

Atherosclerosis is a degenerative disease of the arteries characterized by thickening and hardening of their walls. Since atherosclerosis occurs in virtually epidemic proportions, the need for a diagnostically accurate screening method cannot be overstated.

It is important to the prevention of atherosclerosis to develop a practical and accurate screening test for significant changes in atherogenic lipoproteins. This goal has not been achieved because of the cumbersome and expensive methodology r^eeded for isolation and quantification of serum lipoproteins, and therefore the i.-edical community's focus on the study of much simpler lipoprotein components, such as: total cholesterol, HDL-cholesterol, triglycerides and apolipoproteins A and B.

These approaches were not surprising in view of the known practical limitation of the two basic methods for lipoprotein separation: lipoprotein agarose gel or cellulose acetate electrophoresis and analytical ultracentrifugation using Philpott-

Svenson's optical system. Both methods are extremely time consuming and, although they can be relatively well standardized, their resolution of the main lipoprotein classes is suboptimal. Thus, it is understandable why quantification of the simple components of lipoproteins insufficiently replaced the measurement of the concentration of actual lipoprotein moieties separated into their classes.

The history of lipoprotein electrophoresis illustrates these problems. Analytical filtration paper was introduced in 1949 as the first oldest supporting medium for the separation of serum lipoproteins. Unfortunately it only resulted in two distinct fractions, α and β. (Table I, pattern 0). The introduction of 1 % agarose gel allowed separation of a pre-/? complex composed of lipoproteins of higher molecular weight of approximately 4,500,000 to 15,000,000 daltons (Table I, pattern 1 ,2). Further refinements of the agarose gel (1.2% gel in 0.035 % EDTA buffer, 1 % sucrose) allowed a partial separation of pre-β subfractions. (Table I, pattern 3,4). According to combined ultracentrifugal and electrophoretic studies, the α₂ -fraction is formed under pathological conditions by chylomicrons of liver origin and their remnants, or NNLDL (very very low density lipoproteins), the pre-/3₂ fraction by VLDL (very low density lipoproteins) and the p-β_ fraction by IDL (intermediate low density lipoproteins, DDL, and Lp(a). (Table I, pattern 4 and Table II, pattern 6). The α, -electrophoretic fraction revealed mostly α_1Y-subfraction composed from VVHDL (very very high density lipoproteins) and an albumin-FFA-complex, as well as of α_lx-subfraction (main mass of HDL,), and αix-_TRAi_L containing HDL, (high density lipoproteins) and associated lipoproteins (Table I, pattern 5).

Serum cholesterol levels are insignificantly different or similar in groups of patients with proven coronary heart disease, when compared to groups of clinically healthy people, or to groups of individuals with proven (angiographically) absence of coronary heart disease. The easiest explanation my be visible in pure analytical mathematics: both classes of serum lipoproteins, LDL and HDL, extremely rich on cholesterol, are behaving in cases with CAD (coronary artery disease) in adverse, opposite directions, so that an analytical gain in LDL-cholesterol concentrations may be almost, or completely scored out by an analytical loss of HDL-cholesterol concentrations. Thus, the analysis of serum cholesterol for screening of active atherogenesis is inferior to the screening method of molecular electrophoresis.

Clinical correlation studies have emphasized the importance of the subtractions of the electrophoretic pre-3-complex as markers of the altered lipoprotein metabolism associated with coronary artery disease. Even with this improved separation of fraction and subfraction, the method could only provide measurements of lipoprotein classes. Polyacrylamide gel electrophoresis also led to a somewhat better separation of lipoproteins, but did not offer substantial advantages. Similarly, gradient polyacrylamide gel electrophoresis provided an incomplete separation of LDL nd IDL classes, and is not easily standardized for reproducible results, nor does it offer correct and convenient separation of HDL class (Table II, pattern 7).

This brief review highlights the need for method that afford excellent separation of lipoprotein classes and subclasses so as to allow their quantification. Ideally, these methods would be suitable for routine clinical application. Separation of natural biomolecules from their associations with other serum proteins, glycoproteins, macroglobulins and/or macromolecular complexes is a major challenge because of the range of macromolecular size and weight of the lipoproteins (10,000 daltons to. approximately 350,000,000 daltons and from about 30 A to perhaps 6,000 A). BRIEF DESCRIPTION OF DRAWINGS

Figure 1, patterns 0-5 illustrate the separation of lipoproteins into fractions and

subfractions;

Figure 2 patterns 6 and 7 illustrate the incomplete separation of lipoprotein

classes;

Figure 3 is a schematic representation of molecular electrophoresis based on use of novel systems of discrete nonsequential multitudinous gelous molecular sieves with specific separation coefficients;

Figure 4 illustrate the ability of molecular electrophoresis to accurately reflect changes in the quantities of all lipoprotein classes and subclasses;

Figure 5 illustrates different pathologic values in different lipoprotein classes;

Figure 6 is a spectrophotometric pattern from a known healthy human being;

Figure 7 is a spectrophotometric pattern from a human being with mildly active atherogenesis; and Figure 8 is a spectrophotometric pattern from a human being with severe active atherogenesis.

SUMMARY OF INVENTION The present invention relates to a method of determining mean values of lipoprotein classes and subclasses comprising the steps of:

(a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro- molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by K_D and K_R for the analyzed group of molecules which direct the concentration of

SUBSTITUTE SHEET the medium and density of netting of each gelatinous layer, and wherein the value of

K_D and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value;

(b) adding a sample of biological particles and macromolecules from a human being to system (a) for molecular electrophoresis;

(c) separating said biological particles and macromolecules into lipoprotein classes and subclasses; and

(d) quantifying said separated lipoprotein classes and subclasses;

(e) repeating steps (b), (c) and (d) with different human beings; (f) averaging the quantities of lipoprotein classes and subclasses.

The preisnt invention further relates to a method of detecting active atherogenesis from a human sample comprising the steps of:

(a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro- molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by Kø and K_R for the analyzed group of molecules which direct the concentration of the medium and density of netting of each gelatinous layer, and wherein the value of K_D and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value;

(c) separating said biological particles and macromolecules into atherogenic lipoprotein classes and subclasses; and (d) quantifying said separated atherogenic lipoprotein classes and

subclasses;

(e) comparing said quantity of atherogenic lipoprotein classes and subclasses with predetermined mean values; (f) recognizing diagnostic conditions of active atherogenesis from specific differences in atherogenic classes and subclasses.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Atherosclerosis is a degenerative disease of the arteries characterized by thickening and hardening of the artery walls. An atherogenic lipoprotein is a lipoprotein which, if not present in a human sample at a mean value, may lead to atherosclerosis. A mean value may be a normal value. Active atherogenesis is a process characterized by abnormal values of atherogenic lipoproteins. Thus, it is of critical importance to identify and quantify atherogenic lipoproteins in order to detect active atherogenesis, thereby enabling one to prevent atherosclerosis. The present invention relates to a method of determining mean values of lipoprotein classes and subclasses comprising the steps of:

(a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro¬ molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by K_D and K_R for the analyzed group of molecules which direct the concentration of the medium and density of netting of each gelatinous layer, and wherein the value of K_D and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value; (b) adding a sample of biological particles and macromolecules from a human being to system (a) for molecular electrophoresis;

(c) separating said biological particles and macromolecules into lipoprotein classes and subclasses; and (d) quantifying said separated lipoprotein classes and subclasses;

(e) repeating steps (b), (c) and (d) an effective number of times with samples from different human beings;

(f) averaging the quantities of lipoprotein classes and subclasses. For the purposes of the present invention the term molecular electrophoresis shall mean a process of separating particles and macromolecules by transporting said particles and macromolecules in a homogeneous electric field through a system of molecular filters of increasing density, wherein the specific K_D and/or K_R directs the particle size of retained particles. The use of an electric field to facilitate the transportation of particles makes the process of the present invention similar but not identical with classic electrophoresis.

Once mean values of lipoprotein classes and subclasses are determined, particularly atherogenic lipoprotein classes and subclasses, the detection of active atherogenesis is possible. Thus, the present invention further relates to methods of detecting active atherogenesis from a human sample comprising the steps of: (a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro¬ molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by K_ø and K_R for the analyzed group of molecules which direct the concentration of the medium and density of netting of each gelatinous layer, and wherein the value of

K_ø and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value;

(c) separating said biological particles and macromolecules into atherogenic lipoprotein classes and subclasses; and

(d) quantifying said separated atherogenic lipoprotein classes and subclasses; (e) comparing said quantity of atherogenic lipoprotein classes and subclasses with mean values according to claim 1;

(f) recognizing diagnostic conditions of active atherogenesis from specific differences in atherogenic classes and subclasses.

The samples of biological particles and macromolecules from human beings to be added to the system for molecular electrophoresis may be collected in any manner. Generally, the samples may be obtained from blood. The blood may be pretreated, using ultracentrification methods such as centrification, and the like.

In determining mean values of lipoprotein classes and subclasses molecular electrophoresis is used because of its superior ability to separate and quantify. Generally, a statistically valid number of samples is used to determine mean values.

Preferably, the statistically valid number of samples is collected from human beings of a given geographical area.

Alternatively, molecular electrophoresis is used an effective number of times with samples from different known healthy human beings. A known healthy human being is a person with a proven absence of artery diseases, as indicated by conventional testing methodology. An effective number of times should be construed to mean at least about 25 times, more preferably at least about 35 times. After molecular electrophoresis has been conducted an effective number of times, the resulting quantities of separated lipoprotein classes and subclasses amy be averaged to give mean values. The mean values may also be referred to as predetermined mean values.

It should be understood that mean values of lipoprotein classes and subclasses may vary, depending upon the given geographical area of the human beings who provide the biological samples. Furthermore, the mean values may vary depending upon the race, culture or ethnicity of the human beings who provide biological samples. For purposes of this invention, the term "given geographical area" is intended to include groups of human beings classified by common culture, race or ethnicity. For instance, the mean values determined from human beings in Japan may differ from those of Scandinavia. Pre- drably, the values obtained from a sample are compared to mean values obtained from samples p: vided by human beings of the same given geographical area.

A description of lipoprotein classes and their subclasses is presented for proper understanding of th.. significance of the present invention. There is a certain, although not yet fully established, difference between a subclass and a remnant. Subclasses are regularly appearing (e.g. under normal almost constant and well reproducible flotation rates). Remnants are physically and/or physico-chemically not well-defined mixtures of metabolites, originated at specific metabolic steps (e.g. chylomicrons, very low density lipoproteins, etc.). Chylomicrons (CHY) have mean concentrations up to 50 mg% in the serum of a fasting, healthy person. Postprandially, CHY may increase up to 500%, from

20 to 250mg%, as solitary metabolites; therefore distinguishable from all other pathologic changes. Significant pathologic hyper-chylomicronemia may be characterized by an increase of 500 to 1000%, or even higher concentrations of these metabolites in the patient's circulation. Isolated hyper-chylomicronemia depicts the postprandial stage.

Until recently, the chylomicrons remnants (CHY-R) were largely unknown and/or ignored. Since there particles lack significant amounts of cholesterol and their arterial permeability is very low, the few individuals with a deficiency of lipoprotein lipase, or apoprotein C-II, did not suffer from premature atherosclerosis, although they have enormous amounts of these particles in the circulation. The above- described large lipoprotein particles are also know as very very low density lipoproteins (VVLDL). In pathologic conditions, however, the CHY, CHY-R, NNLDL and WLDL-R may easily exceed 1000% of their mean concentrations. Inherited enzyme defects, such as chylomicronemia syndrome, or Apo-C-II-Deficiency, are examples of the above-described conditions. These remnants are not atherogenic.

Very low density lipoproteins (VLDL) may exceed their mean concentrations of 120-240 mg% in fasting serum by 200-1000%. The highest levels of VLDL are observed as active atherosclerosis in severe cases of familial hypertriglyceridemias.

The NLDL subclasses, as probably all other subclasses and/or remnants, represent a scale of similar metabolic products of which each subclass is analytically distinguishable and has reproducible characteristics. While the chemico-physical composition of these metabolic by-products may be qualitatively, but not quantitatively similar, their patho-physiological features may be quite different.

The first group of remnants of VLDL is characterized by "larger" molecules, as compared to the second group that are still very heterogeneous, characterized by a density of less than about 1.006 g/ml and particle size equal or greater than about

400 A. This group was not characterized by preparative and/or analytical ultracentrifugation at density 1.063 g/ml, but it was well separated using analytical micro-ultracentrifugation with UV-light at density 1.21 g/ml. These remnants may be composed of about 58% triglycerides and about 14% cholesterol. The hydrated densities of these large VLDL range from about 1.006 to about 1.008 g/ml, as compared to the entire VLDL class of about 1.006 to about 1.010 g/ml.

The second subpopulation of the VLDL remnants with a particle size of approximately 350-320 A, and hydrated densities of about 1.010 g/ml is richer on apolipoprotein B and on cholesterol (approximately 30%), but, in contrast to the first group depleted in triglycerides (approximately 31 %) and the apolipoprotein E and C concentrations when related to that of apo B. The differences in apo E content are known to influence the metabolic fate of these remnants. The hydrated densities of small VLDL are about 1.010 g/ml.

Metabolism of triglyceride-rich lipoproteins largely determines the high density lipoprotein (HDL) cholesterol levels. The delayed clearance of dietary fat, that is, an extended postprandial lipemia, may induce the uptake of triglyceride-rich remnants by arterial cells, resulting in an intracellular accumulation of cholesterol esters and in the subsequent development of atherogenic foam cells. Subjects with low levels of subclass HDL_. are more prone to CAD because of their delayed clearance of triglyceride-rich chylomicrons.

The concentration of intermediate density lipoproteins (IDL) are critical markers of active atherogenesis, especially if their total concentration exceeds 50% to 100% of their mean value, a mean value being defined as 30 to 45 mg%. The subclasses of IDL represent additional markers of diagnostic sensitivity. The particle size of IDL is between about 250 and about 300 A.

The IDL class of serum lipoproteins, is a heterogenous fraction containing VLDL remnants and a S_f of less than about 20 LDL. It has been considered as a metabolic intermediate positioned between VLDL and LDL and as the end product of the action of the plasma-lipolysis. In hypertriglyceridemic subjects, apolipoprotein-

B enters directly into IDL and that about 1/4 to 1/3 of newly synthesized apo-B enters plasma by this route. Separating the IDL by the ultracentrifuge at a density less than about 1.019 g/ml, determined S_f ^* of about 12-20^* . The IDL class is quantitatively characterized by the flotation coefficient (Fj ₂₁ = 70-40) and by an electrophoretic mobility (μ) identical with pre-/3, lipoprotein fraction in 1 % agarose gel (Pfizer Pol-E

Film System) and is titled LDL, or an intermediate between low density and very low density lipoproteins.

Two electrophoretic fractions "large IDL" and "small IDL.," exist, and correspond to the ultracentrifugal subclasses as follows: IDL, and IDL;, (with F_{1 21} = 50 and 45 respectively) comprise the smaller molecules of the IDL class.

Increased levels of IDL mass and IDL cholesterol are related to an increased risk of coronary artery disease in humans. Additionally, in patients with peripheral vascular, disease, the sum of IDL molecules had an average molecular weight of about 4.50 ± 0.22 x 10⁶ daltons, which was not significantly different from about 4.59 ± 0.21 10⁶ daltons found in a control group. An analysis of the isotope kinetics of apo-

B in VLDL, LDL and IDL indicates that an IDL compartment consisted of two closely connected subcompartments, one of which was outside the immediate circulation. Furthermore, part of IDL is accepted directly by peripheral cells. Concentrations higher than about 35 mg% of IDL class of lipoproteins in human serum are pathologic. Analytically determined concentrations of IDL in the human circulation express only part (e.g. the intravascular portion) of the level of this lipoprotein. Nevertheless, it is presently the most valuable metabolic sign of the atherogenic conditions which may be characterized by relatively low serum concentrations of total cholesterol and total triglycerides as well.

The LDL class of plasma lipoproteins flotate in the density range 1.019 g/ml -

1.063 g/ml, whose flotation rate is S_{f 1,053} = 0-12. This physical continuum of LDL forms a spectrum of particles, varying in size, hydrated density and chemical composition. LDL in the 1.019 g/ml to 1.063 g/ml density range is usually referred to as low density classes of lipoproteins, LDL^ whereas in the 1.006 g/ml to 1.019 g/ml density range was in the past called low density lipoprotein, LDL,.

A fractionation and ultracentrifugation of the 1.006 g/ml to 1.09 g/ml density range (e.g. IDL and LDL) yielded two subspecies of lipoprotein are present in the 1.006 to 1.019 g/ml density range and three subspecies in the 1.019 to 1.063 g/ml density range. The concentrations of low density lipoproteins (LDL) are primary and equally critical markers as the IDL of active atherogenesis. The LDL concentrations are dependent upon movements of their LDL₃ subclass. A 50% increase in LDL concentration, or in concentration of its LDL₃ subclass, is pathognomonic.

The LDL subclasses may also possess different physico-chemical properties (e.g. chemical composition, particle size, molecular weight and conformation) under different pathological conditions. Heterogeneity exists in human LDL group of subspecies.

Similarly, inborn and/or acquired disorders in lipoprotein biosynthesis provide quantitatively different metabolic rates for different lipoprotein subclasses. The metabolic state of certain lipoprotein unit (e.g. a certain lipoprotein subclass) depends upon a variety of factors.

For instance, the kinetics of the subclasses depend, very importantly, on the content of apolipoprotein-B which is known to have slower kinetic than those of the other apolipoproteins (e.g. apolipoprotein E and C). Therefore, subclasses with higher level of apo B (e.g. in molecularly smaller LDL₃ subclass) are characterized by slower metabolic kinetic rate than those with lower levels of apo B, mixed even with apo E and apo C (e.g. in molecularly larger LDLj subclass). These kinetic conditions, however, do not fully explain why the LDLj subclass is, under normal conditions, quantitatively the most concentrated of the LDL subclasses, as it would provide some kind of LDL pool.

Human liver cells secrete apo B_JOQ (and not apo B₄₈). Human apo B₁₀₀ is acylated and regulated by corresponding m RNA. Its synthesis is regulated by levels of circulating LDL or its metabolites, if the subclass LDL₃ is taken into consideration. The cellular uptake of subclass LDL₃ regulates the synthesis of m RNA and directly influences the spectrum of LDL,-₃ subclasses.

Particles of LDL₃ subclass are distributed into two compartments : about 40% into the circulation and the rest into the intercellular space of the human body. An obstruction to this natural metabolic trend results in the accumulation of the relatively molecularly small and dense LDL₃ subclass particles in circulation (mean molecular weight about 1,500,000 to about 2,400,000 daltons). On the contrary, the buoyant lipoprotein particles, which belong to the LDL class, are in fact the LDL, subclass of mean molecular weight of about 3,000,000 to about 3,500,000 daltons, having on the surface of its molecules not only apolipoprotein B₁₀₀, but also some apo E and apo

C and being richer on lipids.

The LDL in humans is largely resistant to dietary changes. Even an addition of large quantities of cholesterol to the diet does not change the composition, nor particle size of LDL. Nevertheless, as patients with hyper-apo-B lipoproteinemia undergo weight reduction, the composition of LDL particles commonly changes.

They increase in particle size, because these molecules acquire more cholesterol- esters.

The concentrations of medium density lipoprotein (MDL) are not, to a significant extent, diagnostically useful with regard to atherosclerosis. The concentrations of high density lipoproteins (HDL) are, of course, important in the prediction of active atherogenesis. However, it is the decrease of HDL concentration over 50% of mean values (e.g. < 100 mg% of the moiety) of this lipoprotein which is diagnostic.

The HDL class, having a density of about 1.063 to 1.21 g/ml, may be separated into the HDL, subclass, defined by a density of about 1.09 g/ml, as well as with F_1-2, less than about 9 and further into the HDL₂ subclass, characterized by a density of about 1.15 g/ml and F_{1 2}, of about 9-3.5. In addition to these minor subclasses, the major HDL₃ subclass has a density of about 1.125 to 1.210 g/ml with the mean density of about 1.18 g/ml and Fι.₂. of about 3.15 - 0.0. The lipoprotein class HDL has a simple relationship to atherogenesis. A significant decrease in the concentrations of the above HDL metabolites is directly related to atherogenesis. Thus, the lower the concentration of HDL, or HDL-CH, the higher the predictive value of this phenomenon. The multilayered molecular filtration systems in accordance with the invention comprise a series of at least about 5 discontinuous molecular sieves which are effective to separate particles according to a broad range of molecular weights when arranged in a regular or irregular, i.e., sequential or nonsequential, successions manner such that selected molecules or molecular groups are retained by the specific "binding" or sieving layers of the filtration system. The largest molecules are engaged by the first sieving layer, while all other molecules of less size pass into the next layer. The next sieve selectively keeps groups of the next large molecules, but releases the rest of the molecules of lesser size until separation of the molecular groups is completed. Optionally, the filtration system may contain a number of non- sieving or "nonbinding" layers which function to increase the resolution of the separation in certain cases. Figure 3.

As previously stated, the multilayered molecular filtration systems comprise a series of at least about 5 discontinuous molecular sieves. Preferably, the multilayer molecular filtration systems comprise a series of at least about 8 discontinuous molecular sieves. In another preferred embodiment, the multilayer molecular filtration systems comprise a series of at least about 12 discontinuous molecular sieves. The precise number of molecular sieves, or layers, to be used may depend upon the identity of a specific lipoprotein class or subclass to be separated and quantified.

SUBSTITUTE SHEET The particles or molecules which may be separated are of various origin

(synthetic or occurring in the nature) with unrestricted significance, e.g., in chemistry, pharmacology, biochemistry, pathophysiology, plant or animal physiology, or in medicine. The separating gelatinous layers may be prepared from all known media including, but not limited to, cellulose acetate, starch gels, agar-gels, agarose-gels, gels of agarose derivatives, polyacrylamide gels, gradient polyacrylamide gels of different gradient (e.g., 2% to 27% of PAA and 4% to 30% of PAA), pore gradient polyacrylamide gels, mixtures of, or combinations of polyacrylamide gels and different derivatives of agarose or other polysaccharides, immunofixative or biospecific sorbants, as well as any of the above gels with addition of protein- complexes-forming variants (SDS), urea, or emulators (TRITONS), or glass adhesives, etc. The gels may be solid, semifluid or fluid.

The system is most advantageous in separations and/or determinations of molecular weights of unknown particles, or those mixtures of particles with very broad molecular sizes and weights or mixtures of very small as well as very large particles.

The discrete system of layers filters-up, or otherwise separates macromolecular mixtures according to their molecular sizes in such a manner that each of the layers excluded (or retains, or delays) groups of molecules of higher order and releases all other molecules of lower order, e.g., of smaller molecular sizes, until the groups of the smallest molecular size are retained at the end of the mediae while none of the molecules are lost.

An optimal composition of discrete layers may be selected or developed using adequate molecular standards and/or molecular markers covering the entire molecular

scale of separated mixtures or complexes.

The above specific conditions are best defined by the partition coefficient K_D and by the retardation coefficient K_R "{designed by K.A. Ferguson : Metabolism, 13 : 985, 1964). The retardation coefficient of Ferguson is given by:

where C = (0.434ιr V)^1A where 1' is the length of the fiber chain per unit volume, R the radius of the partitioning molecule, and r the radius of the gel fiber. According to the theories of gel filtration, K_R should depend only on the gel structure and the size of the partitioning protein.

The partition coefficient, K_D, of the protein between the gel and solution, as obtained from gel filtration experiments, can be related to K_R, by the following relationship : log K_D = - K_RT where T is a gel concentration. If the values of κ. of analyzed macromolecules are plotted against the size of radii (R) of these partitioning molecules (in A) at a standard pH, a linear relationship is obtained. Figure 3.

These systems are particularly efficient in the form of discrete layers of identical length or widely varying length depending on the composition, structure, network, shape or gradation of these changes in their quality or composition. The unique discrete and discontinuous multiple arrangement of molecularly specific, e.g., specific macromolecules and/or sizes of macromolecules, retaining filters or sieves constitutes the basic principle of the system.

SUBSTITUTE SHEET The procedure for making a discrete system of specific molecular sieves or filters includes: empiric or experimental selection between sieves and specific binders

(e.g., based on immuno-affinity, complexing, physical or chemical interactions, etc.) in particular relation to the macromolecules to be isolated. Empiric or experimental selection of molecular dimensions of molecular mixtures for which the discrete system should be effective. Empiric or experimental selection of molecular standards and/or markers indicating critical margins of the dissolved molecular spectrum and preparation of discrete multitudinous systems of specific, discontinuous sieves or filters first in a working form of non-polymerized or non-gelatinized status. The discrete gelatinous polymers are organized into discrete multiple layers, each of them designed to have exclusive and/or strictly specific properties, which are selectively effective in the molecular separation and/or in molecular characterization of the chosen mixture of macromolecules.

In one embodiment, the discrete layers of the gelatinous polymers are obtained by polymerization (at elected various degrees) of recrystallized N-methylol-acrylamide with a bifunctional acrylic or allylic compound (e.g., BIS or DATD) forming a three- dimensional, cross-linked sieve of the resulting polymer having different acrylamide monomer concentrations, e.g., up to 36% of the acryl monomer. To activate the above reaction, a free-radical catalyst, e.g., potassium persulfate, riboflavin or TEMED, was added, as well as a sodium chloride and/or TRIS, or barbital, or similar buffer solution to provide the intended concentration of acrylamide monomer

(T).

Usually, each layer is prepared individually and casted (layered) into pre- prepared glass column, capillaries, circular dishes, or pre-assembled macro- or micro- plates in the molds. Where indicated, layers containing gradients are prepared using conventional linear or exponential gradient makers (e.g., LKB-217-MULTIPHOR II, or others). Repel Silane and/or Bind Silane (LKB) may be applied, if the gels are to be bound or repelled off the glass plate. Where needed, the above-described PAA gels may be modified and/or filled with agarose, sepharose, agarose derivatives, etc.

The following abbreviations have the meaning indicated:

ACRYL Acrylamide

BIS N, N ' -methylenebisacrylamide

DATD N,N'-diallyltartardiamide

DMAPN 3-dimethylaminopropionitrile

TEMED N,N,N' ,N'-tetramethylenediamine

TRIS tris (hydroxy-methyl) aminomethane

T range of acrylamide monomer and bimer con¬ centration in % g ACRYL + g BIS

% T x 100

100 ml gel

PAA polyacrylamide

PAAG polyacrylamide gel(s)

PAAGE polyacrylamide gel(s) electrophoresis

GPAAGE gradient polyacrylamide gel(s) electrophoresis

C crosslinking degree g BIS

% C x 100 g ACRYL + g BIS

Where suitable, the layers are filled starting from the bottom of the molds. The "stacking gels" are layered as the last layer.

At the end of this operation, any known techniques can be employed for the construction of the applicator gels with precast sample wells (capable of absorbing from about 5 μl to about 80 μl sample volumes). Construction of the final system of specific layers is prepared in any quantity, quality, physical form, composition and sequence. The borderlines of the layers may be constructed with sharp, visible edges (for example if the previous layer is fully gelatinized, when the following one is being overlaid) or in unbroken confluent fashion. Semiliquid or liquid gelatinous forms which do not change their physical status in time, although fully polymerized, have to be fitted into their containers (of any shape) in order to t vent a mixing of materials.

The separation lines of the above layers (of the same, or different concentra¬ tions and/or compositionsjare rectilinear when electrophoretic plates or columns are elected or strictly circular (when circular microelectrophosis, or circular diffusion techniques are picked-up). All possible mixtures of gelatinizing materials, with or without any additional "filling" gels, can be used without restrictions.

The discrete layers may be particularly efficient as system of almost identical or widely varying layers, because of their optimal composition, structure or length or shape or gradation of changes in their quality or composition.

The final shape of these discrete systems of molecular sieves may be manufactured in any thickness and usable forms in large or small (e.g. microplates) plates with glasseous or plastic supports, columns in glass tubes, or in capillaries, chromatographic columns, etc. In this way, gel-plates of various sizes (from 125 x 250 mm, allowing the simultaneous separation of up to 96 biological samples - down to 50 x 50 mm, allowing the simultaneous separation of 9 samples and only one standard) may be prepared. The preservation of gels in glass plates, containing analytical systems of discrete and/or discontinuous gelatinous layers may be preserved by immersion into a buffer solution as well as by packing them in storage bags. In such manner, it will be possible to handle their shipment and other usual commercial handling. The above-described gel-casting forms (e.g. plates, tubes, capillaries, etc.) should be useful with any known electrophoretic procedure.

The optimal composition of the multitudinous, discrete, nonsequential gelatinous layers may be prepared or developed by determination of partition (K_D) and/or retardation (K_R) coefficients, calculated for the selected molecular groups of the analyzed mixture. All known forms of low or high voltage (vertical, horizontal, radial, capillary, etc.) electrophoresis, electrofocussing, chromatography, etc. may be utilized for separation of macromolecules. The mode of distribution of mixed macromolecules throughout the mediae may also be strongly dependable on the available instrumenta¬ tion. However, once the best fitting instrumentation is established, it may be easily standardized. Specific electrophoretic mA, V, V-hrs. (buffer strength, pH continuity, discontinuity, etc.) have to be established. For the separation of genetic material a Pulse Field Electrophoresis may also be employed.

The unique discrete and discontinuous nonsequential and multitudinous arrangement of molecularly specific (e.g. specific for certain macromolecules and/or sizes of macromolecules) molecular filters or sieves constructed in agreement with separation coefficients, constitutes a basic principle of the system.

The largest particles are engaged by the first sieving layer, while all other molecules of less size pass into the next layer. The next sieve selectively keeps groups of the large molecules, but releases the rest of the molecules of lesser size until separation of the molecular groups is completed. Optionally, the filtration system may contain a number of nonsieving or "nonbinding" layers which function to increase the resolution of the separation in certain cases.

All other smaller lipoprotein particles, not to mention other much smaller protein molecules are passing through the critical variation of the last pore size of the zone (with continuously increasing, for example polyacrylamide-agarose-destran, gradients of this pore gel) and enter into next discrete gelatinous layer, where they are further selected and concentrated - . they move from the rear margin to the front margin of the unique zone. In order to compensate for more balanced distribution of individual lipoproteins, the speed of the electrophoretic movement on some parts of the mediae in some specific migrating distances is decreased (retarding gels, containing saccharose), or increased (accelerating gels of specific composition).

The noncontinuous pore size gradients created by variations in acrylamide concentrations are most critical factors in function of the discontinuous, nonsequential molecular filters, because they represent a sieving barrier for movement of entire classes of lipoproteins, characterized by certain molecular sizes. Therefore, only molecules of certain critical hydrated density and/or molecular size are selectively retained and retarded by a unique layer, thus all others are passed through the next unique, equilibrial layer for further separation.

Very sharp zones of single or individual lipoprotein subunits are first predetermined by discrete gradients, which limit their migration throughout gradients with continuously increasing polyacrylamide concentrations, characterized by fradually reducing pore size and increasing viscosity, caused the front of the molecular group to move more slowly than the rear of the group (trapped in the net of this kind of gel) counteract the diffusion and compresses the molecular sorority into tight zone again.

The typical lipoprotein spectrum in proven coronary artery heart disease in North America is characterized by significant increase (p< 0.001) of HDL. These are the main signs of the corresponding metabolic disorders (which are more than one involved, all covered under this simple metabolic inversion of LDL/HDL concentrations).

Coronary heart or artery disease in North America is further characterized by insignificant differences in total lipoprotein concentrations and, what is even more peculiar, by equally insignificant differences in total cholesterol concentrations.

However, the differences in total triglyceride levels are highly significant.

Screening and diagnosis is enormously important with regard to class IDL, although less significantly (p<0.01) elevated in CAD. Most recently, it has been recognized that the IDL class (S_f"_{1 063}12-6O, or F_{1 210} 70-40) has similar modes of entry and exit into the intima of the human carotid artery, as the atherogenic class

LDL, which has been known for a long time, to transfer from plasma into the arterial wall in vivo.

There is a great diagnostic advantage in analyzing the complete spectra of serum lipoproteins for early detection of atherogenesis. The complete spectra of serum lipoproteins includes four main classes and as many as eight subclasses.

Serum cholesterol concentrations analyze only one case, while serum lipid levels consider three classes. Additionally, calculated values may be avoided, for example of "β- cholesterol", or so called "LDL-cholesterol", handicapped by unacceptable analytical errors (up to ± 30%) and, in addition, to extreme sensitivity (e.g. triglyceride determinations) to strict 12-14 hours fasting specimens.

Most determinatively, even a minor presence of atherogenesis is closely related to analytically recognizable elevations of concentrations of main lipoprotein classes, and what is most elucidating, lipoprotein subclasses.

There are different conditions for pathologic values in different lipoprotein classes, as is demonstrated by Figure 6.

CHY and VVLDL are pathologic, only if their concentration in serum exceeds 500% of their mean concentration, and of course, in correlation with other classes, as well as VLDL. Both classes may, under pathologic conditions, far exceed even 1000% of their mean concentrations, however.

The analysis of IDL is difficult in relation to the behavior of VLDL and LDL, resp. LDL_j subclass. Generally, about a 50% or 75% increase of IDL signals metabolic disorders related to a pathology. Nevertheless, its total concentration seldom exceeds 500% of its mean value.

LDL is a diagnostically "sensitive" class. A 50% increase of LDL concentration is pathognomonic for the active stage of atherosclerosis.

The sensitivity of the new analytical tool, ME, for detecting elevations in LDL

SUBSTITUTE SHEET cholesterol, defined as greater than 150 mg% , is near 100% since it is directly assaying LDL and its subclasses. All other screening methods attempt to calculate

LDL cholesterol on the basis of a number of imprecise assumptions leading to ±

30% CV (CV% is coefficient of variance in %, ideally CV% should be less than 5. Mean values for the lipoprotein classes, as determined by analytical ultracentrifugation have been derived by many investigators over the past two decades. Thus, mean values for all subclasses are now available so that a patient's lipoprotein spectrum, based on molecular electrophoresis, can be appropriately classified. MDL has a predictive value only if in this class has been quantitatively determined the Lp(a) lipoprotein.

HDL class is sensitively regulated genetically. Only in familial hyper-α- lipoproteinemia is this class elevated over 300 mg%, and in familial Tangier disease the class is almost completely absent. Pathognomonic is the case of human active atherogenesis characterized by a significant decrease of the concentration of this class

(which we determined in this clinical study as HDI.^ class). A decrease of the HDL level under 150 mg% in serum has to be taken in serious consideration, although active atherogenesis may be characterized with values as low as 50 mg%.

The new technique, a screening tool, is called "molecular electrophoresis" (ME). ME can provide similarly accurate and reproducible separation and quantitation or semiquantitation of atherogenic and other lipoproteins.

The separation of classes and subclasses of lipoprotein using ME is illustrated, and the ability of ME to accurately reflect changes in the quantities of all lipoprotein classes and subclasses is shown in Figure 4.

SUBSTITUTE SHEET The quantification of the separated lipoprotein classes and subclasses may be accomplished using any acceptable means. Spectrophotometric techniques and the like are often used. Methods of quantification are readily apparent to those skilled in the art. Molecular electrophoresis constitutes not only a new analytical method, but a new analytical aim for practical and inexpensive molecular separation of lipoproteins into classes and diagnostically important subclasses. ME offers the only practical way for detection and definition of small metabolic changes, leading to and characterizing active phases of atherogenesis. The method of detecting active atherogenesis utilizes molecular electrophoresis. Molecular electrophoresis is used to separate and quantify lipoprotein classes and subclasses. Preferably and more importantly, it is used to separate atherogenic lipoprotein classes and subclasses. The quantity of the lipoprotein classes and subclasses, especially the atherogenic ones, are compared to predetermined mean values. A diagnostic condition, which may be recognized, is a condition wherein the predetermined mean values of a lipoprotein class and subclass differ from the values obtained from a single sample.

More specifically, recognizing diagnostic conditions of active atherogenesis may be accomplished by comparing the values of lipoprotein classes and subclasses of a sample with the determined mean values of lipoprotein classes and subclasses.

Generally, diagnostic conditions are characterized by a difference in the quantified values of a sample as compared to the determined mean values. In one embodiment of the invention, diagnostic conditions are characterized by a sample's similarity to pathologic values of lipoprotein classes and subclasses. Pathologic values for lipoprotein classes and subclasses are defined elsewhere in this disclosure. It should be understood that pathologic values, just as mean values, may vary dependent upon the characteristics of a given geographic area. Furthermore, because human beings are individuals, specific pathologic values inherently vary from person to person. Figures 6, 7 and 8 demonstrate the effectiveness of the diagnostic method disclosed herein. The three spectrophotometric patterns indicate the presence of discrete lipoprotein classes and subclasses. Figure 6 is a spectrophotometric pattern of macromolecular serum lipoprotein classes from a known healthy human being. Figure 7 is a spectrophotometric pattern of macromolecular serum lipoprotein classes from a human being with mildly active atherogenesis. Figure 8 is a spectrophotometric pattern of macromolecular serum lipoprotein classes from a human being with severe active atherogenesis.

Example 1 Lipoprotein flotants may be separated on the precasted discontinuous, nonsequential agarose-polyacrylamide gel systems.

The following separation of serum lipoproteins is achieved using an eight layer system:

Blood samples in EDTA are obtained on the morning of the cardiac catheterization after a 14 hour fast. Collected plasma samples are adjusted to 1.21 g/ml density by using NaCl and KBr, and centrifuged using Beckman Model L

Ultracentrifuge, at 100,000 x g for 16 hours in a Beckman type 50 rotor. Lipoprotein flotants are removed in the top 2 ml with Pasteur pipettes (polycarbonate tubes) and dialyzed (using Minicon B-15, sample concentrator, Amicon, or S and S Collection Bags : UH- 100/75, Schleicher-Schuell) against the electrophoretic buffers used in this

SUBSTITUTE SHEET study. The dialyzed products are then quickly brought to the planned analysis.

The gelatinous mixture for each layer of gel is prepared separately. A basic solution is provided which contains 1.6 ml of TRIS-Borate-Na₂EDTA (TEE) buffer

(Ph = 8.3), 1.0 ml of 6.4% 3-dimethyl-amiopropionitrile (DMAPN) later replaced with TEMED (N,N,N',N',-tetramethylenediamine), 3.9 ml of 25 agarose (or in some experiments of Sepharose 2B, or 4B, or 6B), 4.25 ml of an acrylamide, plus bisacrylamide solution, the concentration of which corresponds to those designed for the given gelatinous layer. Distilled and deionized water is added to each gelatinous mixture in order to adjust the total volume to about 15.0 ml. All the above-described gelatinous mixtures are kept hot in a water-bath up to about 90 ^*C, so that the agarose remained soluble. Before the use of the above mixture for a preparation of corresponding gelatinous layer, the mixture is removed from the water bath and cooled to 50^* - 55 ^*C. Next, 0.5 ml of 1.6% ammonium persulfate is added to each gelatinous mixture to quickly promote transfer to the gelatinous mold. The components are mixed and the selected portion (for example

1.0 ml) is carried over into the casting form, but not before the previous layer gelatinized.

The electrophoresis of lipoprotein flotants and molecular weight standards is performed in a 10 cm long, 10 cm large and 2.0 nm thick vertical slab gels (casted in GSC-8 apparatus) at about 14 ^*C for the period of about 16 hours, 1 hour at about

30 V and about 15 hours at constant voltage of about 125 V, in TBE buffer (pH = 8.3), utilizing Pharmacia-Uppsala analytical vertical system (GE-2/4-LS cell).

The lipoproteins separated in gels may be prestained (by mixing of equal volume of flotant with 5 g per L solution of Sudan Black B in ethylene glycol), or stained after a standard fixation (200 g trichloracetic acid in 1000 ml of distilled water) with Coomassie Blue G 250 for apolipoproteins (in a mixture of equal portions of acid copper sulfate solution and solved Coomassie Blue in diluted methanol), or

Sudan Black B for lipidic moiety (using SBB in ethyleneglycol or within 60% ethanol), respectively with Oil Red 0 stain. The gels may be thereafter destained and preserved in adequate preserving solution, or dried.

After the electrophoresis is completed, the separated lipoproteins are fixed by heating at 72 ^*C. The gels are cut into two parts, so that they contain nearly identical samples. One part is stained with a solution of 0.1 % Coomassie Brilliant Blue R-250 in 50% ethanol and 9% acetic acid (v/v); the other part is stained with a solution of

0.2% Amido Black 10B in 5% acetic acid, and the final component with Fat Red 7B in alkaline solution. Four liters of Fat Red 7B stain are prepared by dissolving 0.9% g in absolute methanol containing stabilizer Triton-X-100. Working stain is prepared just before use. Staining is standardized by adjusting stain concentration against permanent colorimetric controls at 600 nm. After 15 minutes of staining with Fat

Red 7B, the plates are cleared in a mixture of MeOH and H₂O (3 : 1) for 20 seconds, rinsed in 2% glycerol, and dried at 72 ^*C.

The above procedure is further standardized by the concurrent use of the direct standard described by applicant. This standard contains known quantities of ultracentrifugally isolated and purified lipoprotein classes (S_f°0-400). This allows verification of the precise mobility of each ultracentrifugal class and confirmation that their mobilities are identical with each analysis.

The molecular weight standards used during molecular electrophoresis are Carboxylated Latex beads (DOW Chemicals), Thyroglobulin Dimer and Tyroglobulin (Pharmacia) with molecular diameters of 380 A, 236 A and 170 A respectively. The molecular weights of the above standards are : 5,388,333 daltons, 1,400,000 daltons and 669,000 daltons respectively. The marker proteins are: 0-lipoprotein (about 2 x 10⁶ daltons), α-macroglobulin (about 820,000 daltons), thyroglobulin (about 669,000 daltons) apoferritin (about 443,000 daltons) and finally catalase (about 240,000 daltons). In most cases, albumin is used (m.w. 68,000 daltons and size 71 A) as an electrophoretic migration marker (the albumin existed in a dimer form).

In the early stage, the films are scanned at 520 nm on a scanning densitometer (microzone Digital Integrator, Model R-lll, Beckman Instruments, Inc. Spinco Division). The raw densitometric values are used to determine the relative proportion belonging to each major fraction, reported as percentage of the total area of the entire scan. The total counts per delimited area are also obtained and no correction factor is applied.

The scanning may be enhanced by using the Shimadzu Dual-Wavelength Scanner Model CS-930 (Shimadzu Corporation, California, USA) for densitometric

(at 280 nm, or 620 nm) or reflectometric (at 600 nm wavelength) scans.

A semiquantitative analysis of fractions and subtractions is also performed. Electroforetograms, stained with Fat Red 7B in alkaline solution under standard conditions, are correlated with lipoprotein fractions that had been ultracentrifugally prepared, at standardized densities, and quantitated and then stained with Fat Red 7B under identical conditions : molecular weight standards of four main lipoprotein classes CHY + VNLDL, VLDL, LDL₂₍₃ are simultaneously analyzed electrophoretically and always stained on precasted nonuniform, discontinuous and nonsequential agarose - polyacrylamide gels. Analytical values of 35 normal subjects with angiographically proven absence or coronary artery disease (CAD) and 106 patients with angiographically proven severe coronary artery disease, e.g. with grade 2 (= 50-74% reduction in vessel diameter), grade 3 (= 75-99% reduction in vessel diameter), and grade 4 (= total occlusion) using Sones technique and the classification of CAD-grading independently by two experts is performed. The age of the group of normals is 46.2 (± 8.0) years, the age of the patient group is 52.0 (± 7.1) years. The difference in age is slightly significant (p< 0.05). The mean severity of the CAD in the patient's group could be expressed by the mean grade 10. The products of molecular electrophoresis from the above group are characterized as follows:

1. The first very thin and porous gelatinous layer (of 1.0 - 1.5% T) separates out only the largest molecules (e.g. 1,000 - 4,000 A) e.g. 100,000,000 daltons) to 350,000,000 daltons m.w.), representing exogenous chylomicrons (CHY). Usually, they constitute fraction "a", or its prefraction, if their concentration is extremely high.

The levels of chylomicrons (0.05 - 0.1 mg%) are in human circulation so low in the fasting stage that they become inseparable from endogenous chylomicrons and/or from their fragments. In this case, only one fraction appears in the T = 2% molecular sieve.

2. The second gelatinous layer (of 2.0% T) is characterized by K_R, generally suitable to retard chylomicrons of hepatic origin of smaller particle size (600 - 900 A). Under physiological conditions, even the next group of lipoproteins (VVLDL) may be separated in this molecular sieve. 3. The third gelatinous layer (of 2.5% T) retains the chylomicron fragments, or NVLDL, characterized by a range of sizes (550 - 450 A) and molecular weights (approximately 80,000,000 - 25,000,000 daltons). Similarly, under physiologic conditions, even this class may contribute to the fraction "a". 4. The fourth layer (of 3.0% T) represents the specific molecular sieve with retardation coefficient for VLDL forming few subclasses in the broad range of 600 to 400 A, e.g. in the range of 18,000,000 - 6,000,000 daltons (mean 12,000,000 daltons). Under physiologic conditions this class has only one, but may reveal up to four subclasses in pathologic stages. Fraction b. 5. The fifth layer (T = 3.2 (± 0.1)%) has to fit for the retardation coefficient for IDL, diagnostically second most important class in the spectrum of serum lipoproteins. It may belong metabolically to VLDL, but separates on the invented systems of gels well, producing up to 2-4 subclasses under pathologic conditions in the range of sizes 350 - 250 A and molecular weights of 5,000,000 - 4,500,000 daltons. Interlayers of standard network of T = 2% enhances the clarity of separation. Fraction c.

6. The sixth cluster of layers (in fact a group of layers ranging in their netting from T = 3.5 - 4.5%, eventually combined with interlayers of standard network of T = 2%) is equipped with retardation coefficient K_R, fitting for retardation of all LDL subclasses. The undivided LDL class includes lipoprotein particles with sizes ranging from 206 - 210 A and molecular weights from 4,000,000 - 1,500,000 daltons.

7. The seventh layer (T - 4.8% ± 0.2%) is adequate for K_R and a proper

density of netting may be calculated for Lp(a) if desired. The MDL render particles of size 200 - 190 A and m.w. 1,000,000 - 900,000 daltons and mean density 1.076 g/ml. Fraction h.

8. The eighth cluster of selective layers, eventually with combination of intermediate, alternating standard layers, offers the separation of the complicated HDL class. However, there is some confusion in the present nomenclature.

Therefore, the minimal physical parameters have to be given in order to correctly describe which subclasses are in fact determined of HDL (daltons = 1.0635 - 1.210 g/ml).

The system of molecular sieves consisting from polyacrylamide gels is calculated from corresponding K_D and K_R for HDL, (T = 5.0%), HDLz (T =

11.6%), HDL₃ (T = 16.3%) and VHDL (T = 18.0%). With the polyacrylamide- agarose gels adequate separation is experienced with regard to the above subclasses in the gelatinous sieves of the range of 5.0-8.0% T, mainly after shorter electrophoretic runs. The HDL, = 2b (d 1.070-1.100 g/ml) represents a less known lipoprotein class. Protein content 25%. Mean banding position 1.090 g/ml. M.W. = 4,200,000 ± 200,000. Peak F^*,._2J rate 5.40 ± 0.30. The HDI^ = 2a (d 1.100-1.125 g/ml) includes lipoprotein in the size range 160-140 A. Cholesterol: protein ratio 0.50 to 0.40. Protein content 52%. Mean banding position 1.110 g/ml. M.W. = 2,630,000 ± 100,000. Peak F\₂, rate 3.15 ± 0.06. The HDL₃ = 3 a-c

(d 1.125-1.200 g/ml) represents a lipoprotein of size range from 113-115 A. Cholesterol: protein ratio 0.37-0.14. Protein content 68%. Mean banding position 1.145 g/ml. M.W. = 1,770,000 ± 90,000. Peak F^* ₂, rate 1.56 ± 0.13.

The NVHDL (mean banding position 1.342 g/ml), protein content 98%. Complex i - 1.

The discussed metabolic study is methodologically more complex, but here are commented only the partial results of basic lipid analyses, performed routinely at that time (that of serum cholesterol and serum triglycerides) and the quantitative results of molecular electrophoresis.

They are graphically and numerically depicted on Figure 4. They are noted significant differences between finding in clinically healthy population and patients with proven CAD in the following manners : *** = on the level of high significance, p< 0.001 ** = on the level of low significance, p<0.01

* = on the level of marginal significance, p<0.05

Two-way analysis of variance (fixed-effects model) and multiple comparisons is used to determine the significance of differences between groups in Figure 4. The two factors for analysis of variance are sex and the severity group (three levels). The chi-square statistic is used to assess differences between group 1 and group 2. The correlation between lipid determinations and severity of disease is examined by using stepwise linear regression analysis. The sensitivity and specificity of serum lipid levels and lipoprotein electrophoretic abnormalities in detecting patients with CAD are defined as follows:

Sensitivity (%) = true positives _{χ 10}Q true positives + false negatives

Specificity (%) = true negatives _χ JQQ true negatives + false positives

Predictive value (%) = true posi^tives _{χ 1(χ)} true positives + false positives

For use of the determinations of total serum cholesterol and total serum SUBSTITUTE SHEET triglycerides for the definition of the risk of coronary heart disease have the following

diagnostic values:

Predictive Sensitivity Specificity Value

Cholesterol > 250 mg% 58% 65% 81 %

Triglycerides > 150 mg% 58% 48% 77%

The use of molecular electrophoresis for the determination of the risk of coronary heart disease has the following diagnostic value:

Predictive Sensitivity Specificity Value

Molecular electrophoresis 95% 85% 95%

None of the presently know screening methods for active atherogenesis ever reached or exceeded the predictive value of 85%, not to mention 90%. The molecular electrophoresis has a minimal predictive value of 95%.

The specific changes in the spectrum of serum lipoproteins are in coronary artery disease are also extremely sensitive (95% = sensitivity) and satisfactorily specific (usually more than 85%, closer to 91 - 93%).

While the invention has been explained in relation to its preferred embodiments, it is to be understood that various modifications thereof will become apparent to those skilled in the art upon reading this specification. Therefore, it is to be understood that the invention disclosed herein is intended to cover such modifications as fall within the scope of the appended claims.

Claims

I claim:

1. A method of determining mean values of lipoprotein classes and subclasses comprising the steps of:

(a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro¬ molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by K_ø and K_R for the analyzed group of molecules, which direct the concentration of the medium and density of netting of each gelatinous layer, and wherein the value of K_ø and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value;

(d) quantifying said separated lipoprotein classes and subclasses;

(e) repeating steps (b), (c) and (d) an effective number of times with different human beings;

(f) averaging the quantities of lipoprotein classes and subclasses.

2. The method of claim 1, wherein one or more separating gelatinous layers of system (a) comprise polyacrylamide or derivatives thereof.

3. The method of claim 2, wherein one or more layers of the polyacrylamide gel contain agarose, dextran, or derivatives thereof in an amount ranging from about 0.5% to about 5.0% by weight.

4. The method of claim 1, wherein at least one gelatinous layer in which the concentration of a copolymer as defined by %T is sequentially increasing linearly or exponentially.

5. The method of claim 1, wherein system (a) comprises gel structures and gel concentrations (%T) which are effective to separate biological particles or macromolecules comprising lipoproteins into their classes and subclasses according to their specific partition (K_D) and retardation (K_R) coefficients.

6. The method of claim 1, wherein system (a) comprises from about 6 to about 18 separating gelatinous layers with variable concentrations of copolymers in

the range of about 2 to about 20% by weight.

7. The method of claim 1 wherein said samples in step (b) are collected from human beings of a given geographical area.

8. The method of claim 1 wherein pathologic values are determined.

9. The method of claim 1 wherein said sample of step (b) is pretreated.

10. The method of claim 1 wherein system (a) comprises at least about 8 discrete separating gelatinous layers.

11. The method of claim 1 wherein system (a) comprises at least about 12 discrete separating gelatinous layers.

12. The method of claim 1 wherein system (a) comprises at least about 24 discrete separating gelatinous layers.

13. A method of detecting active atherogenesis from a human sample comprising the steps of:

(a) providing a system of multilayered, discrete, discontinuous, nonsequential gels for molecular electrophoresis of biological particles and macro- molecules, comprising at least about five discrete separating gelatinous layers, wherein each gelatinous layer is defined by specific separation coefficients represented by K_ø and K_R for the analyzed group of molecules, which direct the concentration of the medium and density of netting of each gelatinous layer, and wherein the value of K_ø and K_R for each discrete separating gelatinous layer differs from an adjacent layer by a significant discontinuous and nonsequential value;

(d) quantifying said separated lipoprotein classes and subclasses;

(e) comparing said quantity of lipoprotein classes an subclasses with predetermined mean values;

(f) recognizing diagnostic conditions of active atherogenesis from specific lipoprotein classes and subclasses.

14. The method of claim 13, wherein one or more separating gelatinous layers of the system comprise polyacrylamide or derivatives thereof.

15. The method of claim 14, wherein one or more layers of the polyacrylamide gel contain agarose, dextran, or derivatives thereof in an amount ranging from about 0.5% to about 5.0% by weight.

16. The method of claim 13, wherein at least one gelatinous layer in which the concentration of a copolymer as defined by %T is sequentially increasing linearly or exponentially.

17. The method of claim 13, wherein system (a) comprises gel structures and gel concentrations (%T) which are effective to separate biological particles or macromolecules comprising lipoproteins into their classes and subclasses according to their specific partition (K_D) and retardation (K_R) coefficients.

18. The method of claim 13, wherein system (a) comprises from about 6 to about 18 separating gelatinous layers with variable concentrations of copolymers in the range of about 2 to about 20% by weight.

19. The method of claim 13 wherein said sample in step (b) is collected from a human being of the same given geographical area from which mean values are predetermined in step (e).

20. The method of claim 13 wherein said sample of step (b) is pretreated.

21. The method of claim 13 wherein system (a) comprises at least about 8 discrete separating gelatinous layers.

22. The method of claim 13 wherein system (a) comprises at least about 12 discrete separating gelatinous layers.

23. The method of claim 13 wherein system (a) comprises at least about

24 discrete separating gelatinous layers.

24. A method of detecting active atherogenesis from a human sample comprising the steps of:

(b) adding a sample of biological particles and macromolecules from a human being to system (a) for molecular electrophoresis; (c) separating said biological particles and macromolecules into atherogenic lipoprotein classes and subclasses; and

(d) quantifying said separated atherogenic lipoprotein classes and subclasses;

(e) comparing said quantity of atherogenic lipoprotein classes and subclasses with predetermined mean values;

25. The method of claim 24, wherein one or more separating gelatinous layers of system (a) comprise polyacrylamide or derivatives thereof.

26. The method of claim 25, wherein one or more layers of the polyacrylamide gel contain agarose, dextran, or derivatives thereof in an amount ranging from about 0.5% to about 5.0% by weight.

27. The method of claim 24, wherein at least one gelatinous layer in which the concentration of a copolymer as defined by %T is sequentially increasing linearly or exponentially.

28. The method of claim 24, wherein system (a) comprises gel structures and gel concentrations (%T) which are effective to separate biological particles or macromolecules comprising lipoproteins into their classes and subclasses according to their specific partition (K_D) and retardation (K_R) coefficients.

29. The method of claim 24, wherein system (a) comprises from about 6 to about 18 separating gelatinous layers with variable concentrations of copolymers in the range of about 2 to about 20% by weight.

30. The method of claim 24 wherein said sample in step (b) is collected from a human being of the same given geographical area from which mean values are predetermined in step (e)

31. The method of claim 24 wherein said sample of step (b) is pretreated.

32. The method of claim 24 wherein system (a) comprises at least about 8 discrete separating gelatinous layers.

33. The method of claim 24 wherein system (a) comprises at least about 12 discrete separating gelatinous layers.

34. The method of claim 24 wherein system (a) comprises at least about 24 discrete separating gelatinous layers.