Isolation of plasma lipoproteins by zonal ultracentrifugation in the B14 and B15 titanium rotors

Lipoproteins were isolated from plasma of man, dog, rabbit, rat, and chicken by ultracentrifugation in continuous density gradients using the B14 titanium and B15 titanium zonal rotors. Both the VLDL and the LDL of human plasma were separated easily from the HDL and from the other more plentiful plasma proteins by centrifugation for only 1 or 2 hr in the B14 or B15 rotor, respectively. Satisfactory separation of the HDL from the more dense plasma proteins was not achieved with these rotors. The human LDL achieved isopycnic equilibrium (d 1.04) on prolonged periods (> 24 hr) of centrifugation in a sucrose-KBr density gradient. The pattern of distribution of cholesterol and phospholipid throughout the density gradient coincided with the pattern of distribution of the lipoprotein-protein measured spectrophotometrically or chemically. The concentration of cholesterol and phospholipid in the lipoproteins isolated by zonal ultracentrifugation agreed with analyses reported for lipoproteins isolated by sequential centrifugation in solutions of increasing density. The lipoproteins isolated by zonal ultracentrifugation were characterized further by their electrophoretic behavior. The fractions which were identified as the LDL (d 1.04-1.05) from all species migrated on paper as a beta-globulin; the LDL from plasma of dogs contained an additional component which has been designated as an alpha(2)-globulin. The fractions which were identified as the HDL from all species migrated as an alpha(1)-globulin. Reaction of human LDL with either rabbit antihuman beta-lipoprotein or rabbit antihuman serum resulted in a single immunodiffusion band. The S(f, 1.063) of the human LDL was calculated to be 6.0. When plasma from humans or rabbits was centrifuged in the B15 rotor, the HDL was not visible as a distinct peak and was not separable from the bulk of the more dense plasma proteins; when plasma from dogs or chickens was centrifuged under identical conditions, the HDL was clearly detectable. Even though the mean density of the HDL from dogs or chickens was not different from that of man or rabbits, the visibility of this lipoprotein in dogs and chickens was probably due to its high concentration in the plasma of these species. When plasma from the rat was centrifuged under similar conditions, the HDL was also clearly in evidence. Although rat plasma contained a relatively small concentration of HDL, the lipoprotein had a lower mean density than did the HDL of the other species and was therefore more easily separable from the dense plasma proteins. The procedure of zonal ultracentrifugation for the isolation of lipoproteins by flotation is simultaneously preparative and analytical and should find useful application in the investigation of the soluble lipoproteins from plasma and tissues.     

Rapid purification of photosystem I chlorophyll-binding proteins by differential centrifugation and vertical rotor

Photosystem I (PSI), which consists of a core complex and light-harvesting complex I (LHCI), is an important multisubunit pigment-protein complex located in the photosynthetic membranes of cyanobacteria, algae and plants. In the present study, we described a rapid method for isolation and purification of PSI and its subfractions. For purification of PSI, crude PSI was first prepared by differential centrifugation, which was applicable on a large scale at low cost. Then PSI was purified by sucrose gradient ultracentrifugation in a vertical rotor to reduce the centrifugation time from more than 20 h when using a swinging bucket rotor to only 3 h. Similarly, for subfractionation of PSI into the core complex and light-harvesting complex I, sucrose gradient ultracentrifugation in a vertical rotor was also used and it took only 4 h to obtain the PSI core, LHCI-680, and LHCI-730 at the same time. The resulting preparations were characterized by sodium dodecyl-sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), absorption spectroscopy, and 77 K fluorescence spectroscopy. In addition, their pigment composition was analyzed by high-performance liquid chromatography and the results showed that each Lhca could bind 1.5-1.6 luteins, 1.0 Violaxanthins, and 0.8-1.1 beta-carotenes on average, demonstrating that fewer carotenoids were released than with the slower traditional centrifugation. These results showed that the rapid isolation procedure, based on differential centrifugation and sucrose gradient ultracentrifugation in a vertical rotor, was efficient, and it should significantly facilitate preparation and studies of plant PSI. Moreover, the vertical rotor, rather than the swinging bucket rotor, may be a good choice for isolation of some other proteins.     

Purification of very high density lipoproteins by differential density gradient ultracentrifugation

Differential density gradient ultracentrifugation procedures, utilizing a vertical rotor, were developed for the preparative purification of very high density lipoproteins (VHDL, density greater than 1.21 g/ml). The VHDLs of several insect species were purified as follows. An initial density gradient ultracentrifugation step removed lipoproteins of lower density from the VHDL-fraction, which partially separated from the nonlipoproteins present in the infranatant. A complete separation was achieved by a second centrifugation step employing a modified gradient system. The use of a vertical rotor and specially designed discontinuous gradients allows a relatively fast, efficient, and economical isolation of the class of very high density lipoproteins. Similar gradient systems should be useful for the detection and purification of VHDLs from other sources.     

Plasma lipoprotein separations by zonal ultracentrifugation.

Procedures for the separation of plasma lipoprotein classes and subclasses by zonal ultracentrifugation are described. The main density classes, very low density lipoproteins (VLDL), low density lipoproteins (LDL) and high density lipoproteins (HDL), in plasma can be separated in a single run for 20 hours. For the isolation of VLDL-LDL a centrifugation time of only 90 minutes is needed. Separations can be performed on plasma volumes varying from 10 to 400 ml in the Ti-14 rotor used; VLDL can in this way be isolated from 400 ml plasma in 30 minutes. The advantages and disadvantages of zonal ultracentrifugation in comparison with the commonly employed differential ultracentrifugation for separation of lipoproteins are discussed.     

Specific elimination of albumin in serum lipoprotein preparations.

In the preparation of pure serum lipoproteins by means of ultracentrifugation techniques, it is necessary to perform consecutive centrifugations in order to reduce the content of contaminating serum albumin (1) impairing apolipoprotein determination and purification and turnover studies with ¹²⁵I-labeled lipoproteins. Extensive centrifugation, however, may cause in vitro changes in the composition of the lipoprotein particles (1,2). The present report describes the use of matrix-bound anti-albumin for specific elimination of albumin in very low-density lipoprotein (VLDL), low-density lipoprotein (LDL), and high-density lipoprotein (HDL) preparations from human serum.     

Estimation of lipoprotein and apoprotein distribution in rat plasma by cumulative density ultracentrifugation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Lipoprotein distribution in rat plasma determined after sequential ultracentrifugation (requiring 8 days of centrifugation to separate lipoproteins in five density classes), was compared to estimates based upon cumulative density ultracentrifugation (46 hr of ultracentrifugation). In general comparable values were obtained by the two methods with regard to protein, total cholesterol, cholesteryl ester, free cholesterol, and triacylglycerol distribution. However, the HDL3 protein concentration found by sequential ultracentrifugation was only about 50% of that found after the cumulative procedure. Apolipoproteins in lipoproteins isolated by the two methods were well separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis. Color of the stained bands was extracted and read photometrically. A linear standard curve was obtained with albumin. Absorbance corresponding to 1 microgram/ml was 0.057. Below d = 1.100 g/ml (HDL2b) the two ultracentrifugation methods gave comparable results for all apoproteins. In contrast to this the level of apo A-I, apo E, and apo A-IV in the more dense types of HDL was higher when estimated by cumulative than by sequential ultracentrifugation. In HDL3 isolated by sequential ultracentrifugation the apo A-IV, apo E, and apo A-I concentrations were 51, 31, and 45% respectively, of values found after cumulative ultracentrifugation. The results indicate that cumulative density ultracentrifugation, followed by colorimetric determination of apoproteins separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, is a useful approach when studying lipoprotein distribution in rat plasma.     

Isolation of serum chylomicrons prior to density gradient ultracentrifugation of other serum lipoprotein classes

A method for the removal of serum chylomicrons before density gradient ultracentrifugation of the other serum lipoproteins using an SW 41 swinging bucket rotor is presented. In a preliminary spin, the chylomicrons with an Sf greater than 400 X 10(-13) s float to the top of the gradient, whereas the other lipoproteins are retained in the infranatant fraction. After removal of the chylomicrons, the other serum lipoproteins are subsequently fractionated by isopycnic density gradient ultracentrifugation. Analysis of the separated lipoprotein fractions suggested that this procedure permits isolation of a chylomicron fraction consisting solely of chylomicrons but that the very low density lipoprotein fraction subsequently isolated also contains chylomicrons or chylomicron remnants with an Sf less than 400 X 10(-13) s, and that there is considerable overlap in flotation rate and particle size of very low density lipoproteins and chylomicrons.     

Very fast ultracentrifugation of serum lipoproteins: influence on lipoprotein separation and composition

A very short run time and small sample volumes in the separation of lipoproteins by preparative ultracentrifugation are needed for several investigations. Recently, a very fast sequential separation method was described that needs only 100 min for one run in a centrifugal field of 625 000 × g. We studied the influence of centrifugal fields of this dimension on lipoprotein separation and lipoprotein particle integrity using a Beckman Optima^TM TLX ultracentrifuge with a TLA-120.2 rotor. Rotor speed (120/90/60/30 · 10³ rev./min) and run time (100 min/3 h/6.7 h/27 h) were selected in such a way that the product of centrifugal field and run time remained constant. The first conditions correspond to the very fast ultracentrifugation (VFU) procedure with a centrifugal field of 625 000 × g. Thirty different plasma samples covering a wide range of lipid and protein concentrations were separated in the course of two centrifugal runs at densities of 1.006 and 1.063 kg/l which yielded very-low-density lipoproteins (VLDL), low-density lipoproteins (LDL), and the subnatant of low-density lipoproteins, including high-density lipoproteins (HDL) and concomitant sedimented plasma proteins. The major lipid components of the lipoproteins, triacylglycerols, free and esterified cholesterol, phospholipids and the apolipoproteins B and A-I, were estimated considering the masses of the tube contents after a slicing procedure. Measurements of lipids and proteins showed a very good recovery of better than 94% and 91%, respectively, and precision-within-series (coefficient of variation) of better than 4.2% and 6.5%, respectively. The effects of the rotor speed on the lipoprotein structure appeared to be weak. With increasing rotor speed, VLDL and LDL lipid constituents principally tended to decrease, whereas they increased in the subnatant of the LDL-run. The mean lipoprotein mass composition, considering the mass percentage of each measured particle constituent, did not show significant alterations. Total protein decreased in VLDL and in LDL and increased in the subnatant of the LDL-run. As checked by an enzyme-linked immunosorbent assay (ELISA) and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), the protein effects were due to nearly complete disappearence of contaminating plasma proteins, especially albumin as the major contamination of VLDL and LDL. The apolipoproteins (apo) B-100, A-I, E and C-I to C-III remained nearly unaffected. The main advantages of VFU were the very short run time (cumulative flotation time is 3.4 h) and the elimination of albumin without repeated runs. The procedure was suitable for the assessment of lipid and protein constituents in lipoproteins from very small plasma samples (500 μl).     

Apolipoprotein and lipid distribution between vesicles and HDL-like particles formed during lipolysis of human very low density lipoproteins by perfused rat heart

A study was undertaken to determine the relative association of lipid and apolipoproteins among lipoproteins produced during lipolysis of very low density lipoproteins (VLDL) in perfused rat heart. Human VLDL was perfused through beating rat hearts along with various combinations of albumin (0.5%), HDL2, the infranatant of d greater than 1.08 g/ml of serum, and labeled sucrose. The products were resolved by gel filtration, ultracentrifugation, and hydroxylapatite chromatography. The composition of the lipoprotein products was assessed by analysis of total lipid profiles by gas-liquid chromatography and immunoassay of apolipoproteins. A vesicle particle, which trapped and retained 1-2% of medium sucrose, co-isolated with VLDL and VLDL remnants by gel filtration chromatography but primarily with the low density lipoprotein (LDL) fraction when isolated by ultracentrifugation. The vesicle was resolved from apoB-containing LDL lipolysis products by hydroxylapatite chromatography of the lipoproteins. The vesicle lipoprotein contained unesterified cholesterol (34%), phosphatidylcholine and sphingomyelin (50%), cholesteryl ester (6%), triacylglycerol (5%), and apolipoprotein (5%). The apolipoprotein consisted of apoC-II (7%), apoC-III (93%), and trace amounts of apoE (1%). When viewed by electron microscopy the vesicles appeared as rouleaux structures with a diameter of 453 A, and a periodicity of 51.7 A. The mass represented by the vesicle particle in terms of the initial amount in VLDL was: cholesterol (5%), phosphatidylcholine and sphingomyelin (3%), apoC-II (0.5%), apoC-III (2.2%). The majority of the apoC and E released from apoB-containing lipoproteins was associated with neutral-lipid core lipoproteins proteins which possessed size characteristics of HDL. The vesicles were also formed in the presence of HDL and serum and were not disrupted by serum HDL. It is concluded that lipolysis of VLDL in vitro results in the production of VLDL remnants and LDL apoB-containing lipoproteins, as well as HDL-like lipoproteins. A vesicular lipoprotein which has many characteristics of lipoprotein X found in cholestasis, lecithin: cholesterol acyltransferase deficiency, and during Intralipid infusion is also formed. The majority of apolipoprotein C and E released from apoB-containing lipoproteins is associated with the HDL-like lipoprotein. It is suggested that the formation and stability of the vesicle lipoprotein may be related to the high ratio of cholesterol/phospholipid in this particle.     

Comparison of two ultracentrifugation procedures for separation of nonhuman primate lipoproteins.

A comparison of nonhuman primate plasma lipoproteins isolated by swinging bucket rotor density gradient or fixed angle rotor differential ultracentrifugation is described. Whereas these two methods produced comparable results for the composition of low density (LDL) and high density (HDL) lipoproteins, the very low density lipoprotein (VLDL) fraction isolated with the swinging-bucket rotor contained relatively more cholesterol (free and esterified) and less phospholipid and protein than that fraction obtained with the fixed-angle rotor. Estimations of lipoprotein concentration by agarose gel electrophoresis and particle size by electron microscopy coupled with molar ratios of surface to core constituents indicate that the swinging bucket procedure resulted in a more complete harvest of VLDL particles, especially those in the larger size range.     

Agarose isoelectric focusing of plasma low and very low density lipoproteins using the PhastSystem

Human plasma lipoproteins, fractionated by density gradient ultracentrifugation, and very low density lipoproteins, subfractionated by cumulative rate centrifugation, were subjected to agarose isoelectric focusing in small format thin gels prepared in the laboratory for the commercially available PhastSystem (Pharmacia). From preparation of the gels to their staining, the procedure took less than 3 h. The pH gradient was found reproducible and the apparent average pI of individual low density lipoproteins could be measured with a coefficient of variation of less than 5% between and less than 2% within the same run. The method appears especially suitable for the exploration of charge properties of multiple lipoprotein samples, or other large macromolecules as low density lipoproteins and very low density lipoproteins, with considerable economy of time and reagents.     

Interaction of cremophor EL with human plasma

1. Interaction of cremophor EL (CRM) with human plasma lipoproteins and nonlipoproteins has been investigated by ultracentrifugation. 2. VLDL has only a low or negligible capacity to bind CRM, i.e. there is little or no change in the optical absorption at 280 nm of VLDL when CRM is added. 3. A low density subfraction of low density lipoproteins seems to associate substantially with CRM at relatively low CRM concentrations (1-3 mg/ml), but such association is not evident for CRM concentrations in the region 12-116 mg/ml. 4. Low density lipoproteins (LDL) may act as a carrier for CRM-emulsions, yet there seems to be no concomitant change in the 280 nm optical absorption of the proteins of LDL. 5. The position in the gradient (i.e. in the centrifugation tube after centrifugation) of high density lipoproteins (HDL) is shifted towards lower density in the presence of 1-4 mg CRM/ml. For higher concentrations of CRM, a destruction of HDL can be observed: the HDL distribution is converted into a bimodal distribution of respectively lighter and heavier "HDL"-particles than the normal ones; the densities at the peaks of these distributions are approximately 1.07 g/ml (light), 1.20 g/ml (heavy) and 1.11 g/ml (normal HDL). The optical extinction coefficient is apparently the same for the proteins of normal--and modified HDL. 6. Even high CRM concentrations (less than or equal to 116 mg/ml) have no perceptible effect on the gradient positions and profile of human serum albumin (HSA) and/or other heavy proteins. 7. The possible biological significance of these findings is briefly touched upon.     

人脐血血浆外泌体提取方法的比较北大核心CSCD

为探寻高效且稳定的提取人脐血血浆外泌体的方法,利用超高速离心法、蔗糖垫密度梯度离心法、改良超速离心法和聚乙二醇(polyethylene glycol, PEG)沉淀法提取人脐血血浆外泌体,并比较4种方法的优劣。利用透射电镜、动态光散射技术观察外泌体的形态、结构及大小;聚氰基丙烯酸正丁酯(bicinchoninic acid, BCA)法测定外泌体蛋白总量;Western blotting检测外泌体表面标志蛋白CD63、HSP70以及外泌体阴性蛋白GM130 (高尔基标志蛋白)的表达。结果表明,与提取外泌体的“金标准”,即超高速离心法相比,蔗糖垫密度梯度离心法稳定性好,获取的外泌体粒径较均一,但操作较复杂,耗时长;改良超速离心法操作较简单,纯度较高;PEG沉淀法提取的外泌体蛋白量最高,操作时间最短,但杂质较多。结果表明,4种方法均能从人脐血血浆中获取外泌体,但在操作时间、纯度、提取量等方面存在一定差异。因此,应根据实验目的和具体要求选择合适的提取人脐血血浆外泌体的方法。     

Rapid quantitative apolipoprotein analysis by gradient ultracentrifugation and reversed-phase high performance liquid chromatography

A new methodology for the analysis of lipoprotein composition using a combination of gradient ultracentrifugation and high performance liquid chromatography was used to determine the differences in lipoprotein composition between non-hyperlipidemic men and women. Lipoproteins from each subject were separated into six subfractions: VLDL, IDL, LDL, and three subfractions of HDL by a single gradient ultracentrifugation spin of less than 5 hr. The HDL subfractions were designated HDL-L (the lightest density subfraction, rich in apoCs and poor in apoA-II), HDL-M (the middle subfraction, rich in apoA-II), and HDL-D (the most dense, relatively poor in both the apoCs and apoA-II). The concentrations of the water-soluble apolipoproteins in each subfraction were determined using reversed-phase HPLC. The concentrations of apoB and the lipid components of the lipoproteins were determined by chemical and enzymatic methods. This methodology proved to be highly reproducible when performed on fresh plasma samples and we were able to identify many sex-associated differences in lipoprotein composition. This methodology is the only nonimmunological technique available for analyzing lipoprotein composition that offers such a combination of accuracy, speed, and completeness.     

Ultracentrifugation micromethod for preparation of small experimental animal lipoproteins

Sequential flotation ultracentrifugation is commonly used in the preparation of plasma lipoproteins. However, protocols often require prolonged centrifugation time (48-72 h) and large plasma volumes (2-20 ml), which makes them unsuitable for studies on small laboratory animals. Although analytical techniques such as FPLC have often small sample requirements, further fraction analysis is often limited to the small fraction volume obtained. A sequential ultracentrifugation micromethod is described to obtain rat lipoprotein fractions from 400 microl of plasma in a cumulative centrifugation time of 7.5 h. Fraction volumes were determined and densities were adjusted to those of rat plasma lipoproteins. Polyacrylamide gel electrophoresis and enzymatic measurements of triglycerides, total cholesterol, and phospholipids were used to assess the purity of the lipoprotein fractions. The results were compared with those obtained from a classical sequential ultracentrifugation protocol. The micromethod presented here can be further adapted to other experimental animal species with little modifications.     

An effective method for plasma lipoprotein separation: studies of various animal species

1. Plasma lipoprotein separation by density gradient ultracentrifugation largely depends on visual examination based on the natural yellow pigments of lipoproteins. 2. In non-human species and in humans with dyslipoproteinemia, some lipoproteins are not well visualized due to the lack of pigments. 3. Using a fluorescent probe in minute quantity (1,6 diphenyl-1,3,5-hexatriene) we were able to demonstrate an effective plasma lipoprotein separation using a discontinuous density gradient ultracentrifugation technique. 4. Plasma lipoproteins of human, chicken, rat and carp were compared showing the unique character of carp HDL.     

Separation of the main lipoprotein density classes from human plasma by rate-zonal ultracentrifugation

The major lipoprotein density classes (chylomicrons-VLDL, LDL, HDL(2) and HDL(3)) were isolated from human plasma in a two-step ultracentrifugal procedure using the Ti-14 zonal rotor. The isolation of the two major high density lipoprotein subclasses (HDL(2) and HDL(3)) was achieved in a 24-hr run using a nonlinear NaBr gradient in the density range of 1.00-1.40. The lipoproteins with a density < 1.063 found in the rotor's center were isolated in a second run of 140 min duration using a continuous linear NaBr gradient in the density range of 1.00-1.30. The isolated lipoproteins were analyzed for chemical composition and for electrophoretic mobility; purity of isolated fractions was checked by immunochemistry. The lipoproteins exhibited flotation rates, chemical compositions, and molecular weights similar to those found with the common sequential procedures in angle-head rotors. The amount of lipoprotein lipids in the bottom fraction of the zonal rotor was comparable to that of the angle-head rotor. The described method yields the main lipoprotein density classes free from albumin in a very short running time; compared with the rate-zonal techniques already in use, this method allows the quantitative separation of an additional lipoprotein density class (HDL(2)) without increasing the running time. Furthermore, this procedure proved to be suitable for isolation of plasma lipoproteins from subjects with various types and varying degrees of hyperlipoproteinemia.     

A rapid single-step centrifugation method for determination of HDL, LDL, and VLDL cholesterol, and TG, and identification of predominant LDL subclass

Determination of the circulating levels of plasma lipoproteins HDL, LDL, and VLDL is critical in the assessment of risk of coronary heart disease. More recently it has become apparent that the LDL subclass pattern is a further important diagnostic parameter. The reference method for separation of plasma lipoproteins is ultracentrifugation. However, current methods often involve prolonged centrifugation steps and use high salt concentrations, which can modify the lipoprotein structure and must be removed before further analysis. To overcome these problems we have now investigated the use of rapid self-generating gradients of iodixanol for separation and analysis of plasma lipoproteins. A protocol is presented in which HDL, LDL, and VLDL, characterized by electron microscopy and agarose gel electophoresis, separate in three bands in a 2.5 h centrifugation step. Recoveries of cholesterol and TG from the gradients were close to 100%. The distribution profiles of cholesterol and TG in the gradient were used to calculate the concentrations of individual lipoprotein classes. The values correlated with those obtained using commercial kits for HDL and LDL cholesterol. The position of the LDL peak in the gradient and its shape varied between plasma samples and was indicative of the density of the predominant LDL class. The novel protocol offers a rapid, reproducible and accurate single-step centrifugation method for the determination of HDL, LDL, and VLDL cholesterol, and TG, and identification of LDL subclass pattern.     

Lipolysis products promote the formation of complexes of very-low-density and low-density lipoproteins

In the course of lipolysis, surface lipid products may accumulate on very-low-density lipoproteins (VLDL). To investigate potential lipoprotein interactions mediated by such products, radiolabeled low-density lipoproteins (LDL) were incubated with VLDL and bovine milk lipoprotein lipase in the presence of limited free fatty acid acceptor. With partial VLDL degradation, association of radiolabeled LDL with VLDL remnants or larger aggregates of VLDL density was demonstrated by gradient gel electrophoresis, agarose chromatography, and density gradient ultracentrifugation. VLDL-LDL complex formation was also observed in incubations with lipid extracts from lipolyzed VLDL or with purified palmitic acid in the absence of lipolysis. Complex formation was inhibited by addition of increasing amounts of albumin as free fatty acid acceptor, but could be detected at molar ratios of free fatty acids/albumin that occur in vivo. Composition analysis of LDL reisolated following incubation with VLDL and lipase under conditions favoring partial complex formation revealed enrichment in glycerides and depletion of cholesterol. We conclude that lipolysis products can promote the formation of stable complexes of LDL and VLDL, and that physical interactions of this nature may play a role in the transfer of lipids and apolipoproteins between lipoprotein particles.     

Single spin density gradient ultracentrifugation method for the detection and isolation of light and heavy low density lipoprotein subfractions

A single spin density gradient ultracentrifugation method in a swinging bucket rotor has been applied for the detection and isolation of low density lipoprotein (LDL) subfractions. The visualization of the LDL heterogeneity was facilitated by prestaining the serum with Coomassie Brilliant Blue R prior to density gradient ultracentrifugation for 19.5 hr. A total of 13 human serum pools was analyzed. In each pool, two LDL subfractions, a lighter LDL1 subfraction, occasionally showing a subdivision into two bands, LDL1A and LDL1B, and a heavier LDL2 could be clearly distinguished by the banding pattern in the density gradient. Physicochemical characteristics of the isolated LDL subfractions were determined. The simple method for detection and isolation of these subfractions presented here may facilitate future studies on LDL heterogeneity.