Genetic control of human brain transcript expression in Alzheimer disease.

We recently surveyed the relationship between the human brain transcriptome and genome in a series of neuropathologically normal postmortem samples. We have now analyzed additional samples with a confirmed pathologic diagnosis of late-onset Alzheimer disease (LOAD; final n = 188 controls, 176 cases). Nine percent of the cortical transcripts that we analyzed had expression profiles correlated with their genotypes in the combined cohort, and approximately 5% of transcripts had SNP-transcript relationships that could distinguish LOAD samples. Two of these transcripts have been previously implicated in LOAD candidate-gene SNP-expression screens. This study shows how the relationship between common inherited genetic variants and brain transcript expression can be used in the study of human brain disorders. We suggest that studying the transcriptome as a quantitative endo-phenotype has greater power for discovering risk SNPs influencing expression than the use of discrete diagnostic categories such as presence or absence of disease.

CR1 is associated with amyloid plaque burden and age-related cognitive decline.

OBJECTIVE: Recently, genome-wide association studies have identified 3 new susceptibility loci for Alzheimer's disease (AD), CLU, CR1, and PICALM. We leveraged available neuropsychological and autopsy data from 2 cohort studies to investigate whether these loci are associated with cognitive decline and AD neuropathology. METHODS: The Religious Orders Study (ROS) and Rush Memory and Aging Project (MAP) are longitudinal studies that enroll nondemented subjects and include annual clinical evaluations and brain donation at death. We evaluated CR1 (rs6656401), CLU (rs11136000), and PICALM (rs7110631) in 1,666 subjects. We evaluated associations between genotypes and rate of change in cognitive function as well as AD-related pathology. Lastly, we used pathway analysis to determine whether relationships between single nucleotide polymorphisms and cognitive decline are mediated through AD pathology. RESULTS: Among our study cohort, the mean years of follow-up were 7.8 for ROS and 4.3 for MAP. Only the CR1 locus was associated with both global cognitive decline (p = 0.011) and global AD pathology (p = 0.025). More specifically, the locus affects the deposition of neuritic amyloid plaque (p = 0.009). In a mediation analysis, controlling for amyloid pathology strongly attenuated the effect of the CR1 locus on cognitive decline. INTERPRETATION: We found that common variation at the CR1 locus has a broad impact on cognition and that this effect is largely mediated by an individual's amyloid plaque burden. We therefore highlight 1 functional consequence of the CR1 susceptibility allele and generalize the role of this locus to cognitive aging in the general population.

GAB2 alleles modify Alzheimer's risk in APOE epsilon4 carriers.

The apolipoprotein E (APOE) epsilon4 allele is the best established genetic risk factor for late-onset Alzheimer's disease (LOAD). We conducted genome-wide surveys of 502,627 single-nucleotide polymorphisms (SNPs) to characterize and confirm other LOAD susceptibility genes. In epsilon4 carriers from neuropathologically verified discovery, neuropathologically verified replication, and clinically characterized replication cohorts of 1411 cases and controls, LOAD was associated with six SNPs from the GRB-associated binding protein 2 (GAB2) gene and a common haplotype encompassing the entire GAB2 gene. SNP rs2373115 (p = 9 x 10(-11)) was associated with an odds ratio of 4.06 (confidence interval 2.81-14.69), which interacts with APOE epsilon4 to further modify risk. GAB2 was overexpressed in pathologically vulnerable neurons; the Gab2 protein was detected in neurons, tangle-bearing neurons, and dystrophic neuritis; and interference with GAB2 gene expression increased tau phosphorylation. Our findings suggest that GAB2 modifies LOAD risk in APOE epsilon4 carriers and influences Alzheimer's neuropathology.

Tart cherry juice decreases oxidative stress in healthy older men and women.

Compared with young adults, older adults have significantly impaired capacities to resist oxidative damage when faced with acute stress such as ischemia/reperfusion. This impairment likely contributes to increased morbidity and mortality in older adults in response to acute trauma, infections, and the susceptibility to diseases such as atherosclerosis, cancer, diabetes, and Alzheimer's disease. Consumption of foods high in polyphenols, particularly anthocyanins, have been associated with improved health, but the mechanisms contributing to these salutary effects remain to be fully established. This study tested the hypothesis that consumption of tart cherry juice containing high levels of anthocyanins improves the capacity of older adults to resist oxidative damage during acute oxidative stress. In a double-blind, placebo-controlled, crossover design, 12 volunteers [6 men and 6 women; age 69 +/- 4 y (61-75 y)] consumed in random order either tart cherry juice or placebo (240 mL twice daily for 14 d) separated by a 4-wk washout period. The capacity to resist oxidative damage was measured as the changes in plasma F(2)-isoprostane levels in response to forearm ischemia-reperfusion (I/R) before and after each treatment. The tart cherry juice intervention reduced the I/R-induced F(2)-isoprostane response (P < 0.05), whereas placebo had no significant effect. The tart cherry juice intervention also reduced basal urinary excretion of oxidized nucleic acids (8-hydroxy-2'-deoxyguanosine, 8-hydroxyguanosine) (P < 0.05) but not urinary excretion of isoprostanes. These data suggest that consumption of tart cherry juice improves antioxidant defenses in vivo in older adults as shown by an increased capacity to constrain an oxidative challenge and reduced oxidative damage to nucleic acids.

Comparison of biomarkers of oxidative stress and cardiovascular disease in humans and chimpanzees (Pan troglodytes).

In the oxidative stress hypothesis of aging, the aging process is the result of cumulative damage by reactive oxygen species. Humans and chimpanzees are remarkably similar; but humans live twice as long as chimpanzees and therefore are believed to age at a slower rate. The purpose of this study was to compare biomarkers for cardiovascular disease, oxidative stress, and aging between male chimpanzees and humans. Compared with men, male chimpanzees were at increased risk for cardiovascular disease because of their significantly higher levels of fibrinogen, IGF1, insulin, lipoprotein a, and large high-density lipoproteins. Chimpanzees showed increased oxidative stress, measured as significantly higher levels of 5-hydroxymethyl-2-deoxyuridine and 8-iso-prostaglandin F(2alpha), a higher peroxidizability index, and higher levels of the prooxidants ceruloplasmin and copper. In addition, chimpanzees had decreased levels of antioxidants, including alpha- and beta-carotene, beta-cryptoxanthin, lycopene, and tocopherols, as well as decreased levels of the cardiovascular protection factors albumin and bilirubin. As predicted by the oxidative stress hypothesis of aging, male chimpanzees exhibit higher levels of oxidative stress and a much higher risk for cardiovascular disease, particularly cardiomyopathy, compared with men of equivalent age. Given these results, we hypothesize that the longer lifespan of humans is at least in part the result of greater antioxidant capacity and lower risk of cardiovascular disease associated with lower oxidative stress.

Sorl1 as an Alzheimer's disease predisposition gene?

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by progressively disabling impairments in memory, cognition, and non-cognitive behavioural symptoms. Sporadic AD is multifactorial and genetically complex. While several monogenic mutations cause early-onset AD and gene alleles have been suggested as AD susceptibility factors, the only extensively validated susceptibility gene for late-onset AD is the apolipoprotein E (APOE) epsilon4 allele. Alleles of the APOE gene do not account for all of the genetic load calculated to be responsible for AD predisposition. Recently, polymorphisms across the neuronal sortilin-related receptor (SORL1) gene were shown to be significantly associated with AD in several cohorts. Here we present the results of our large case-control whole-genome scan at over 500,000 polymorphisms which presents weak evidence for association and potentially narrows the association interval.

Relationship between sunlight exposure, housing condition, and serum vitamin D and related physiologic biomarker levels in captive chimpanzees (Pan troglodytes).

In primates, the primary source of vitamin D is synthesis in the skin through sun exposure. Decreased sun exposure may lead to vitamin D deficiency and consequently other health issues. In laboratory, sanctuary, and zoo settings, chimpanzees (Pan troglodytes) may be housed indoors for prolonged periods of time without regular exposure to unfiltered sunlight. However, little research has examined the relationship between housing conditions and vitamin D serum levels in captive chimpanzees. In this study, we retrospectively compared serum levels of total vitamin D, calcium, ionic calcium, phosphorous, albumin, and alkaline phosphatase in 18 female and 12 male chimpanzees as they cycled between indoor-only and indoor-outdoor enclosures. Total vitamin D was significantly lower and alkaline phosphatase significantly higher when subjects were in the indoor-only enclosures compared with when they had regular access to outdoor enclosures. A vitamin D effect occurred only in young and prime-adult animals. Changes were significant in female but not in male chimpanzees. Calcium, ionic calcium, phosphorus, and albumin did not differ between indoor-only and indoor-outdoor enclosures. However, female chimpanzees exhibited significantly lower calcium and phosphorous levels while in the indoor-only enclosures. These results suggest that adult captive chimpanzees experience vitamin D deficiency when housed without regular access to unfiltered sunlight and that these effects may be more acute for adult female animals.

Negligible senescence: how will we know it when we see it?

The recent public claim that "SENS is a practical, foreseeable approach to curing aging" has stirred considerable controversy among bio-gerontologists. Testing this hypothesis will not only require precise definitions for the somewhat subjective terms "practical," "foreseeable," and "curing," it will require a precise definition of the term "aging." To facilitate proper experimental design, this definition must focus on the nature of aging itself, not its causes or consequences. Aging in mammals is a process that begins early in adult life and continues steadily thereafter until death. It is manifested by a decline in the functional capacity (or, more precisely, reserve capacity) of a variety of vital physiologic systems leading to increasing risk of morbidity and mortality over time. Aging, however, cannot be measured by simply monitoring morbidity and/or mortality. Aging can only be measured by monitoring the decline of global functional capacity itself. This, in turn, will require an operational definition of aging expressed as a rate function (i.e., it will have units expressing aging as an overall rate of functional change per unit time). Widespread acceptance of such global indexes of aging rate in animal models and humans will greatly facilitate research activity specifically designed to increase the understanding of aging mechanisms and antiaging interventions.

The effects of aging on hormone and reproductive cycles in female chimpanzees (Pan troglodytes).

In contrast to those for human females, observational cycle data available for chimpanzees suggest that menstrual cycling, and thus reproductive potential, continues until near death. This study documents age-related changes in estrous cycling and hormone profiles in 14 female chimpanzees (Pan troglodytes) ranging in age from 31 to 50 y. Estrous data were analyzed from daily cycle charts, averaging 13.3 y of cycle data per subject, after omission of gestational and postpartum amenorrhea. Concentrations of total luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol (E2), and other hormones were assayed in serum samples taken biannually. Sample collection times were chosen to avoid the ovulatory LH and FSH peaks of the female's cycle and yielded a mean of 19.6 serum samples over an average of 14.4 y per subject. Analysis of cycle charts revealed a negative relationship between age and the percentage of cycle days at maximal tumescence. There also were positive relationships between age and the length of the estrous cycle and age and the percentage of cycle days at complete detumescence. Analysis of hormonal data revealed curvilinear relationships between age and both LH and FSH. These cycle and hormonal changes mirror those in perimenopausal and menopausal women. Our data provide evidence of perimenopause (at 30 to 35 y) and menopause (at 35 to 40 y) in the chimpanzee.

Discrimination of breast cancer by anti-malignin antibody serum test in women undergoing biopsy.

PURPOSE: The anti-malignin antibody serum (AMAS) test (Oncolab, Boston, MA) has been reported as 97% sensitive and 95% specific for malignancies. To objectively assess accuracy of this test for discrimination of breast cancer, we studied a series of women undergoing core breast biopsy. SUBJECTS AND METHODS: Seventy-one core-needle breast biopsies were classified as malignant, suspicious, or benign by two independent pathologists blinded to AMAS results. Corresponding sera were read as AMAS positive, negative, or borderline by criteria used by Oncolab and also using criteria derived from receiver-operator curves based on values for slow (S-tag), fast (F-tag), and their difference (Net-tag) antibody reported by Oncolab. We calculated sensitivity and specificity and analyzed distributions by Fisher's exact test. RESULTS: Biopsies were read as 42 (59%) benign, 12 (17%) suspicious, and 17 (24%) malignant. By Oncolab criteria, sensitivity (59%) and specificity (62%) were maximized by pooling suspicious with malignant and AMAS borderline with positive (P = 0.098). Receiver-operator curves showed best sensitivity (62%) and specificity (69%) for the criterion AMAS positive if Net-Tag > 135 microg/mL or S-Tag > 220 microg/mL (P = 0.015). CONCLUSIONS: The AMAS test discriminates suspicious and malignant from benign lesions, but sensitivity is insufficient to identify patients to be spared biopsy and false-positive rates are too high for population screening.

Quantification of 8-iso-prostaglandin-F(2alpha) and 2,3-dinor-8-iso-prostaglandin-F(2alpha) in human urine using liquid chromatography-tandem mass spectrometry.

Quantification of 8-iso-prostaglandin F(2alpha) (8-iso-PGF(2alpha)) has been suggested to be a reliable indicator of lipid peroxidation that may be related to in vivo free radical generation, oxidative damage, and antioxidant deficiency. We have developed a LC-MS/MS method to quantify 8-iso- PGF(2alpha) and its dinor metabolite, 2,3-dinor-8-iso-prostaglandin F(2alpha) (2,3-dinor-8-iso-PGF(2alpha)), in human urine samples. After an initial purification step using an automated C18 solid phase extraction procedure, the urine sample was injected directly into a liquid chromatography (LC) system and detected with tandem mass spectrometry. The detection limit of the assay was 9 pg for 8-iso-PGF(2alpha) and 3 pg for 2,3-dinor-8-iso-PGF(2alpha) with both inter- and intraday variations of less than 12%. The inaccuracies were less than 3% for both analytes at three different levels. The urinary excretion rate of 2,3-dinor-8-iso-PGF(2alpha) was higher than that of 8-iso-PGF(2alpha), and changed in proportion to the parent compound (R = 0.70, n = 60). Values obtained with this method showed good linear correlation to duplicate 8-iso-PGF(2alpha) measurements performed with GCMS (R = 0.97, n = 15). The mean excretion rates of 8-iso-PGF(2alpha) and 2,3-dinor-8-iso-PGF(2alpha) were significantly higher in smokers than in nonsmokers (0.53 +/- 0.37 vs. 0.25 +/- 0.15 microg/g creatinine, p = 0.002 for 8-iso-PGF(2alpha) and 8.9 +/- 3.8 vs. 4.6 +/- 2.6 microg/g creatinine, p = 0.003 for 2,3-dinor-8-iso-PGF(2alpha), respectively). The excellent accuracy, reproducibility, and high throughput of this method should permit it to be used in large clinical studies and standard clinical laboratories.

Urinary excretion of three nucleic acid oxidation adducts and isoprostane F(2)alpha measured by liquid chromatography-mass spectrometry in smokers, ex-smokers, and nonsmokers.

To assess novel liquid chromatography/mass spectrometric methods for measuring oxidative damage to nucleic acids and lipids, we compared urinary excretion of 8-hydroxy-2'-deoxyguanosine (8-OHdG), 5-hydroxymethyl-2'-deoxyuridine (5-OHmU), and 8-hydroxyguanosine (8-OxoG), and an isoprostane, 8-iso-prostaglandin F(2)alpha (IsopF(2)alpha) in 234 healthy men (n = 113) and women (n = 121), 80 current smokers, 96 never-smokers), and 58 ex-smokers (no tobacco use for 3 years). The 8-OHdG and 8-OxoG did not differ significantly by group; 5-OHmU was higher in smokers, compared with ex- (p <.003) and never- (p <.0001) smokers and in ex- vs. never-smokers (p =.014) at, respectively, 13.5 +/- 0.7, 11.3 +/- 1.0, and 8.7 +/- 0.3 microg/g creatinine. IsopF(2)alpha was higher in smokers, compared with ex- (p =.007) and never-smokers (p <.0001) and in ex- vs. never- smokers (p =.002) at, respectively, 1.1 +/- 0.10; 0.74 +/- 0.07, and 0.51 +/- 0.04 microg/g creatinine. There were significant correlations among all three nucleic acid adducts and between IsopF(2)alpha and both 5-OHmU and 8-OHdG. Many smokers and ex-smokers had high levels of either 5-OHmU excretion or IsopF(2)alpha excretion, but not both. We conclude that 5-OHmU and IsopF(2)alpha are more discriminating of oxidative stress from tobacco smoke than the other two compounds measured. Whether characteristic patterns of excretion of these indicators forecast differential disease risk should be explored in future research.

Time to talk SENS: critiquing the immutability of human aging.

Aging is a three-stage process: metabolism, damage, and pathology. The biochemical processes that sustain life generate toxins as an intrinsic side effect. These toxins cause damage, of which a small proportion cannot be removed by any endogenous repair process and thus accumulates. This accumulating damage ultimately drives age-related degeneration. Interventions can be designed at all three stages. However, intervention in metabolism can only modestly postpone pathology, because production of toxins is so intrinsic a property of metabolic processes that greatly reducing that production would entail fundamental redesign of those processes. Similarly, intervention in pathology is a "losing battle" if the damage that drives it is accumulating unabated. By contrast, intervention to remove the accumulating damage would sever the link between metabolism and pathology, and so has the potential to postpone aging indefinitely. We survey the major categories of such damage and the ways in which, with current or foreseeable biotechnology, they could be reversed. Such ways exist in all cases, implying that indefinite postponement of aging--which we term "engineered negligible senescence"--may be within sight. Given the major demographic consequences if it came about, this possibility merits urgent debate.

Is there an antiaging medicine?

In spite of considerable hype to the contrary, there is no convincing evidence that currently existing so-called "antiaging" remedies promoted by a variety of companies and other organizations can slow aging or increase longevity in humans. Nevertheless, a variety of experiments with laboratory animals indicate that aging rates and life expectancy can be altered. Research going back to the 1930s has shown that caloric restriction (also called dietary restriction) extends life expectancy by 30-40% in experimental animals, presumably at least partially by delaying the occurrence of age-dependent diseases. Mutations that decrease production of insulin growth factor I in laboratory mammals, and those that decrease insulin-like signaling in nematodes and fruit flies, have increased life expectancy as well. Other general strategies that appear promising include interventions that reduce oxidative stress and/or increase resistance to stress; hormone and cell replacement therapies may also have value in dealing with specific age-related pathologies. This article reports the findings of a consensus workshop that discussed what is known about existing and future interventions to slow, stop, or reverse aging in animals, and how these might be applied to humans through future research.

The Women's Health Initiative could not have detected cardioprotective effects of starting hormone therapy during the menopausal transition.

The Women's Health Initiative (WHI) randomized controlled trial failed to show cardioprotection by estrogen plus progestin treatment of postmenopausal women. But by design, the WHI population was 10-fold underpowered to show cardioprotection of women starting hormone treatment during the menopausal transition. Thus, observational studies that showed cardioprotection in such women remain the only applicable clinical guide to this issue. Randomized controlled trials are urgently needed to test cardioprotection in women starting treatment during the menopausal transition.

Whole genome association analysis shows that ACE is a risk factor for Alzheimer's disease and fails to replicate most candidates from Meta-analysis.

For late onset Alzheimer's disease (LOAD), the only confirmed, genetic association is with the apolipoprotein E (APOE) locus on chromosome 19. Meta-analysis is often employed to sort the true associations from the false positives. LOAD research has the advantage of a continuously updated meta-analysis of candidate gene association studies in the web-based AlzGene database. The top 30 AlzGene loci on May 1(st), 2007 were investigated in our whole genome association data set consisting of 1411 LOAD cases and neuropathoiogicaiiy verified controls genotyped at 312,316 SNPs using the Affymetrix 500K Mapping Platform. Of the 30 "top AlzGenes", 32 SNPs in 24 genes had odds ratios (OR) whose 95% confidence intervals that did not include 1. Of these 32 SNPs, six were part of the Affymetrix 500K Mapping panel and another ten had proxies on the Affymetrix array that had >80% power to detect an association with α=0.001. Two of these 16 SNPs showed significant association with LOAD in our sample series. One was rs4420638 at the APOE locus (uncorrected p-value=4.58E-37) and the other was rs4293, located in the angiotensin converting enzyme (ACE) locus (uncorrected p-value=0.014). Since this result was nominally significant, but did not survive multiple testing correction for 16 independent tests, this association at rs4293 was verified in a geographically distinct German cohort (p-value=0.03). We present the results of our ACE replication aiongwith a discussion of the statistical limitations of multiple test corrections in whole genome studies.

Reproductive aging in female chimpanzees (Pan troglodytes).

Published age-specific fertility rates document a sharp decline in female chimpanzee fertility after age 35 years. However, in contrast to data on human females, little else is known regarding reproductive aging in chimpanzees. We documented age-related changes in estrous cycling, hormone profiles, and reproductive physiology in 14 female chimpanzees (Pan troglodytes) ranging in age from 32 to 50 years. Estrous data were analyzed from daily cycle charts, averaging 14.1 years of cycle data per subject, after omission of gestational periods and postpartum amenorrhea. Concentrations of total luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, and progesterone were assayed in serum samples taken biannually. Sample collection times were chosen to avoid the ovulatory LH and FSH peaks of the female's cycle and yielded a mean of 16.2 serum samples over an average of 14.0 years per subject. Analysis of cycle charts revealed a negative relationship between age and the length of the follicular phase (percentage of cycle days at maximal tumescence). There also were positive relationships between age and the length of the estrous cycle, the percentage of cycle days at complete detumescence (i.e. amenorrhea), and the frequency of irregular cycles. Analysis of hormonal data revealed curvilinear relationships between age and both LH and FSH, with peaks above the menopausal threshold occurring around ages 35-40 years. Cycle and hormonal changes were similar to those in perimenopausal and menopausal women, providing evidence of perimenopause (at 30-35 years) and menopause (at 40 years) in the chimpanzee.

A high-density whole-genome association study reveals that APOE is the major susceptibility gene for sporadic late-onset Alzheimer's disease.

OBJECTIVE: While the apolipoprotein E (APOE) epsilon allele is a well-established risk factor for late-onset Alzheimer's disease (AD), initial genome scans using microsatellite markers in late-onset AD failed to identify this locus on chromosome 19. Recently developed methods for the simultaneous assessment of hundreds of thousands of single nucleotide polymorphisms (SNPs) promise to help more precisely identify loci that contribute to the risk of AD and other common multigenic conditions. We sought here to demonstrate that more precise identification of loci that are associated with complex, multi-genic genetic disorders can be achieved using ultra-high-density whole-genome associations by demonstrating their ability to identify the APOE locus as a major susceptibility gene for late-onset AD, despite the absence of SNPs within the APOE locus itself, as well as to refine odds ratios (ORs) based on gold-standard phenotyping of the study population. METHOD: An individualized genome-wide association study using 502,627 SNPs was performed in 1086 his-topathologically verified AD cases and controls to determine the OR associated with genes predisposing to Alzheimer's disease. RESULTS: As predicted, ultra-high-density SNP genotyping, in contrast to traditional microsatellite-based genome screening approaches, precisely identified the APOE locus as having a significant association with late-onset AD. SNP rs4420638 on chromosome 19, located 14 kilobase pairs distal to the APOE epsilon variant, significantly distinguished between AD cases and controls (Bonferroni corrected p value = 5.30 x 10(-34), OR = 4.01) and was far more strongly associated with the risk of AD than any other SNP of the 502,627 tested. CONCLUSION: This study provides empirical support for the suggestion that the APOE locus is the major susceptibility gene for late-onset AD in the human genome, with an OR significantly greater than any other locus in the human genome. It also supports the feasibility of the ultra-high-density whole-genome association approach to the study of AD and other heritable phenotypes. These whole-genome association studies show great promise to identify additional genes that contribute to the risk of AD.

A survey of genetic human cortical gene expression.

It is widely assumed that genetic differences in gene expression underpin much of the difference among individuals and many of the quantitative traits of interest to geneticists. Despite this, there has been little work on genetic variability in human gene expression and almost none in the human brain, because tools for assessing this genetic variability have not been available. Now, with whole-genome SNP genotyping arrays and whole-transcriptome expression arrays, such experiments have become feasible. We have carried out whole-genome genotyping and expression analysis on a series of 193 neuropathologically normal human brain samples using the Affymetrix GeneChip Human Mapping 500K Array Set and Illumina HumanRefseq-8 Expression BeadChip platforms. Here we present data showing that 58% of the transcriptome is cortically expressed in at least 5% of our samples and that of these cortically expressed transcripts, 21% have expression profiles that correlate with their genotype. These genetic-expression effects should be useful in determining the underlying biology of associations with common diseases of the human brain and in guiding the analysis of the genomic regions involved in the control of normal gene expression.

Disparate LDL phenotypic classification among 4 different methods assessing LDL particle characteristics.

BACKGROUND: Our study seeks to clarify the extent of differences in analytical results, from a clinical perspective, among 4 leading technologies currently used in clinical reference laboratories for the analysis of LDL subfractions: gradient gel electrophoresis (GGE), ultracentrifugation-vertical auto profile (VAP), nuclear magnetic resonance (NMR), and tube gel electrophoresis (TGE). METHODS: We collected 4 simultaneous blood samples from 40 persons (30 males and 10 females) to determine LDL subclasses in 4 different clinical reference laboratories using different methods for analysis. LDL subfractions were assessed according to LDL particle size and the results categorized according to LDL phenotype. We compared results obtained from the different technologies. RESULTS: We observed substantial heterogeneity of results and interpretations among the 4 methods. Complete agreement among methods with respect to LDL subclass phenotyping occurred in only 8% (n = 3) of the persons studied. NMR and GGE agreed most frequently at 70% (n = 28), whereas VAP matched least often. CONCLUSIONS: As measurement of LDL subclasses becomes increasingly important, standardization of methods is needed. Variation among currently available methods renders them unreliable and limits their clinical usefulness.