Oral alpha-tocopherol supplementation inhibits lipid oxidation in established human atherosclerotic lesions

BACKGROUND: Much experimental evidence suggests that lipid oxidation is important in atherogenesis and in epidemiological studies dietary antioxidants appear protective against cardiovascular events. However, most large clinical trials failed to demonstrate benefit of oral antioxidant vitamin supplementation in high-risk subjects. This paradox questions whether ingestion of antioxidant vitamins significantly affects lipid oxidation within established atherosclerotic lesions. METHODS AND RESULTS: This placebo-controlled, double blind study of 104 carotid endarterectomy patients determined the effects of short-term alpha-tocopherol supplementation (500 IU/day) on lipid oxidation in plasma and advanced atherosclerotic lesions. In the 53 patients who received alpha-tocopherol there was a significant increase in plasma alpha-tocopherol concentrations (from 32.66 +/- 13.11 at baseline to 38.31 +/- 13.87 (mean +/- SD) micromol/l, p < 0.01), a 40% increase (compared with placebo patients) in circulating LDL-associated alpha-tocopherol (p < 0.0001), and their LDL was less susceptible to ex vivo oxidation than that of the placebo group (lag phase 115.3 +/- 28.2 and 104.4 +/- 15.7 min respectively, p < 0.02). Although the mean cholesterol-standardised alpha-tocopherol concentration within lesions did not increase, alpha-tocopherol concentrations in lesions correlated significantly with those in plasma, suggesting that plasma alpha-tocopherol levels can influence lesion levels. There was a significant inverse correlation in lesions between cholesterol-standardised levels of alpha-tocopherol and 7beta-hydroxycholesterol, a free radical oxidation product of cholesterol. CONCLUSIONS: These results suggest that within plasma and lesions alpha-tocopherol can act as an antioxidant. They may also explain why studies using < 500 IU alpha-tocopherol/day failed to demonstrate benefit of antioxidant therapy. Better understanding of the pharmacodynamics of oral antioxidants is required to guide future clinical trials.     

Myeloperoxidase targets apolipoprotein A-I, the major high density lipoprotein protein, for site-specific oxidation in human atherosclerotic lesions

Oxidative damage by myeloperoxidase (MPO) has been proposed to deprive HDL of its cardioprotective effects. In vitro studies reveal that MPO chlorinates and nitrates specific tyrosine residues of apoA-I, the major HDL protein. After Tyr-192 is chlorinated, apoA-I is less able to promote cholesterol efflux by the ABCA1 pathway. To investigate the potential role of this pathway in vivo, we used tandem mass spectrometry with selected reaction monitoring to quantify the regiospecific oxidation of apoA-I. This approach demonstrated that Tyr-192 is the major chlorination site in apoA-I in both plasma and lesion HDL of humans. We also found that Tyr-192 is the major nitration site in apoA-I of circulating HDL but that Tyr-18 is the major site in lesion HDL. Levels of 3-nitrotyrosine strongly correlated with levels of 3-chlorotyrosine in lesion HDL, and Tyr-18 of apoA-I was the major nitration site in HDL exposed to MPO in vitro, suggesting that MPO is the major pathway for chlorination and nitration of HDL in human atherosclerotic tissue. These observations may have implications for treating cardiovascular disease, because recombinant apoA-I is under investigation as a therapeutic agent and mutant forms of apoA-I that resist oxidation might be more cardioprotective than the native form.     

c-kit-immunopositive vascular progenitor cells populate human coronary in-stent restenosis but not primary atherosclerotic lesions

Progress in the treatment of human in-stent restenosis (ISR) is hampered by an imprecise understanding of the nature of the cells that occlude vascular stents. Recent studies suggest that circulating vascular progenitor cells may mediate vascular repair and lesion formation. Moreover, functional endothelial progenitor cells appear to play a protective role in attenuating vascular lesion formation. Hence, we sought to answer two important questions: 1). Are primitive cells found in ISR lesions? 2). Is the abundance of cultured angiogenic cells (CACs) in patients with ISR different from that in patients with non-ISR lesions or normal controls? Human coronary atherectomy tissue from 13 ISR, 6 postangioplasty restenosis (RS), and 14 primary (PR) atherosclerotic lesions, as well as 15 postmortem coronary artery cross sections from young individuals without atherosclerosis, were studied. All 13 ISR and 4 of 6 RS tissue specimens contained cells that immunolabeled for the primitive cell marker c-kit and smooth muscle alpha-actin, whereas the intima and media of PR lesions and normal arteries were devoid of c-kit-immunopositive cells. The abundance of peripheral blood mononuclear cell-derived CACs was assessed in 10 patients with ISR, 6 patients with angiographically verified patent stents, and 6 individuals with no clinical evidence of coronary artery disease. CACs were less abundant in ISR patients than in non-ISR controls (13.9 +/- 3.1 vs. 22.3 +/- 6.7 cells/high-power field, P < 0.05), and both of these groups had fewer CACs than non-coronary artery disease patients (37.6 +/- 3.8, P < 0.05). These findings suggest a unique pathogenesis for ISR and RS lesions that involves c-kit-immunopositive smooth muscle cells. Moreover, the paucity of CACs in patients with ISR may contribute to the pathogenesis of ISR, perhaps because of attenuated reendothelialization.     

Oxysterols in cap and core of human advanced atherosclerotic lesions

Objective: Different parts of the advanced atherosclerotic lesion have characteristic differences in lipid content, but the distribution of lipid oxidation products has not been reported. This study provides novel data on oxysterol and hydroxyoctadecadienoic acids quantification in core versus cap. It compares the lipid composition of core and cap to assess the topographical distribution of evidence of lipid oxidation.

Methods: Lipids and oxidised lipids were analysed by gas chromatography (GC) and GC-mass spectrometry (GC-MS) in samples of human atheromatous lipid core and fibrous cap of individual advanced atherosclerotic plaques (Stary, Type V) in necropsy samples.

Results: The total lipid was of course massively greater in the core than in the cap. The oxidation products, cholest-5-en-3β,26-diol (26-OH-CHOL) and cholest-5-en-3β,7β-diol (7β-OH-CHOL) were detected in all the samples. 26-OH-CHOL was more abundant in the core than in the cap when related both to wet weight and to cholesterol. 7β-OH-CHOL levels were significantly higher in the core than in the cap when related to wet weight but not when related to cholesterol. Because the processing included a sodium borohydride reduction step, the 7β-OH-CHOL detected could partly originate from 7-ketocholesterol or 7-hydroperoxy-cholesterol. Several isomeric hydroxyoctadecadienoic acids were detected in both core and cap, more in the cap when related to cholesterol content. Most of the components of the cap showed a high degree of cross-correlation on linear regression analysis, but cross-correlations were weaker for the core. The core samples contained a larger proportion of linoleate relative to oleate than the fibrous cap.

Conclusion: The findings suggest that the different lipid and oxidised lipid contents of cap and core may be due to variations in oxidative activity in different parts of the lesion.     

Oxysterols in cap and core of human advanced atherosclerotic lesions

Objective: Different parts of the advanced atherosclerotic lesion have characteristic differences in lipid content, but the distribution of lipid oxidation products has not been reported. This study provides novel data on oxysterol and hydroxyoctadecadienoic acids quantification in core versus cap. It compares the lipid composition of core and cap to assess the topographical distribution of evidence of lipid oxidation.

Methods: Lipids and oxidised lipids were analysed by gas chromatography (GC) and GC-mass spectrometry (GC-MS) in samples of human atheromatous lipid core and fibrous cap of individual advanced atherosclerotic plaques (Stary, Type V) in necropsy samples.

Results: The total lipid was of course massively greater in the core than in the cap. The oxidation products, cholest-5-en-3β,26-diol (26-OH-CHOL) and cholest-5-en-3β,7β-diol (7β-OH-CHOL) were detected in all the samples. 26-OH-CHOL was more abundant in the core than in the cap when related both to wet weight and to cholesterol. 7β-OH-CHOL levels were significantly higher in the core than in the cap when related to wet weight but not when related to cholesterol. Because the processing included a sodium borohydride reduction step, the 7β-OH-CHOL detected could partly originate from 7-ketocholesterol or 7-hydroperoxy-cholesterol. Several isomeric hydroxyoctadecadienoic acids were detected in both core and cap, more in the cap when related to cholesterol content. Most of the components of the cap showed a high degree of cross-correlation on linear regression analysis, but cross-correlations were weaker for the core. The core samples contained a larger proportion of linoleate relative to oleate than the fibrous cap.

Conclusion: The findings suggest that the different lipid and oxidised lipid contents of cap and core may be due to variations in oxidative activity in different parts of the lesion.     

Esterified lipid hydroperoxide-derived modification of protein: formation of a carboxyalkylamide-type lysine adduct in human atherosclerotic lesions

We have recently identified Nepsilon-azelayllysine (AZL) as a carboxyalkylamide-type novel lysine adduct in the reaction of linoleic acid hydroperoxides with the lysine derivative. To examine the formation of AZL in vivo, a novel monoclonal antibody (mAb19D5) specific to AZL moiety was prepared. The mAb19D5 scarcely recognized oxidized low-density lipoprotein (oxLDL), whereas the treatment of oxLDL with alkali or phospholipase A2 significantly increased the immunoreactivity. Similarly, the immunopositive materials were detected in alkali- or phospholipase A2-treated sections from human atherosclerotic aorta but not in untreated sections. These results suggest that esterified lipid hydroperoxide-derived modification of protein may serve as one mechanism for the oxidative modification of LDL and subsequent formation of atherosclerotic lesions in vivo.     

Staphylococcus aureus extracellular adherence protein (Eap) reduces immune cell phenotype in developing but not in established atherosclerotic lesions

Atherosclerosis is a chronic, inflammatory disease of the vessel wall where triggered immune cells bind to inflamed endothelium, extravasate and sustain local inflammation. Leukocyte adhesion and extravasation are mediated by adhesion molecules expressed by activated endothelial cells, like intercellular adhesion molecule 1 (ICAM-1). Extracellular adherence protein (Eap) from Staphylococcus aureus binds to a plethora of extracellular matrix proteins, including ICAM-1 and its ligands macrophage-1 antigen (Mac-1, α_Mβ₂) and lymphocyte function-associated antigen 1 (LFA-1, α_Lβ₂), thereby disrupting the interaction between leukocytes and endothelial cells. We aimed to use Eap to inhibit the interaction of leukocytes with activated endothelial cells in settings of developing and established atherosclerosis in apolipoprotein E (ApoE) deficient mice on high-fat diet. In developing atherosclerosis, Eap treatment reduced circulating platelet-neutrophil aggregates as well as infiltration of T cells and neutrophils into the growing plaque, accompanied by reduced formation of neutrophil extracellular traps (NETs). However, plaque size did not change. Intervention treatment with Eap of already established plaques did not result in cellular or morphological plaque changes, whereas T cell infiltration was increased and thereby again modulated by Eap. We conclude that although Eap leads to cellular changes in developing plaques, clinical implications might be limited as patients are usually treated at a more advanced stage of disease progression. Hence, usage of Eap might be an interesting mechanistic tool for cellular infiltration during plaque development in basic research but not a clinical target.     

Similar molecules spatially correlate with lipofuscin and N-retinylidene-N-retinylethanolamine in the mouse but not in the human retinal pigment epithelium

The accumulation of lipofuscin in the retinal pigment epithelium (RPE) has been implicated in the development of age-related macular degeneration (AMD) in humans. The exact composition of lipofuscin is not known but its best characterized component is N-retinylidene-N-retinylethanolamine (A2E), a byproduct of the retinoid visual cycle. Utilizing our recently developed matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI–IMS)-based technique to determine the spatial distribution of A2E, this study compares the relationships of lipofuscin fluorescence and A2E in the murine and human RPE on representative normal tissue. To identify molecules with similar spatial patterns, the images of A2E and lipofuscin were correlated with all the individual images in the MALDI–IMS dataset. In the murine RPE, there was a remarkable correlation between A2E and lipofuscin. In the human RPE, however, minimal correlation was detected. These results were reflected in the marked distinctions between the molecules that spatially correlated with the images of lipofuscin and A2E in the human RPE. While the distribution of murine lipofuscin showed highest similarities with some of the known A2E-adducts, the composition of human lipofuscin was significantly different. These results indicate that A2E metabolism may be altered in the human compared to the murine RPE.     

Effect of sphingomyelin versus dipalmitoylphosphatidylcholine on the extent of lipid oxidation

Phospholipids with sites of unsaturation are targets of peroxidation. We investigated the effect of two types of lipids with identical headgroups, sphingomyelins (SMs) and dipalmitoylphosphatidylcholine (DPPC), on the extent of oxidation of stearoyl-arachidonoyl phosphatidylcholine (SAPC) with four double bonds. Peroxidation was induced with tert-butylhydroperoxide and FeCl(2) at 35 degrees C. The decrease of SAPC versus DPPC, or N-palmitoyl SM, or N-stearoyl SM, was monitored by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) at various reaction times. For corresponding molar ratios of DPPC:SAPC and SM:SAPC, SAPC oxidized faster and to a greater extent when DPPC, rather than N-palmitoyl SM or bovine brain SMs, was present. However, at 35 degrees C the hydrophobic tails in SM mixtures were more disordered than in those of DPPC. The slower oxidation of SAPC in SM-rich vesicles may result from the presence of a tight network of H-bonds that bridge neighboring SM molecules and poses a stronger interfacial barrier to the passage of oxidants.     

Phosphorylated, but not native, tau protein assembles following reaction with the lipid peroxidation product, 4-hydroxy-2-nonenal

A correlation between hyperphosphorylation of tau protein and its aberrant assembly into paired helical filaments has lead to suggestions that phosphorylation controls assembly, but lacked a mechanistic basic. In this work, we have found that phosphorylated, but not native, tau protein is able to form polymers after the reaction with 4-hydroxy-2-nonenal, a highly toxic product of lipid peroxidation. Phosphorylation of tau by both proline or non-proline directed kinases, was able to assemble it into polymers.     

Salivary DNA,lipid, and protein oxidation in nonsmokers with periodontal disease

Reactive oxygen species (ROS) are implicated in the destruction of the periodontium during periodontitis. The imbalance in oxidant activity may be a key factor.The aim of this paper is to determine whether periodontitis is associated with increased oxidative damage to DNA, lipids, and proteins and modification of total antioxidant capacity (TAC) in saliva. Saliva was collected from 58 periodontitis patients and 234 healthy controls, all nonsmokers. Periodontal disease status was characterized using the Community Periodontal Index of Treatment Needs (CPITN). Assays for 8-OHdG (ELISA), 8-epi-PGF2α (ELISA), and total protein carbonyls (ELISA), and oxy-blotting (Western)/mass spectrometry were performed to quantify oxidative damage to nucleic acids, lipids, total and individual proteins, respectively, in whole nonstimulated saliva. Salivary TAC was measured by inhibition of ABTS oxidation by metmyoglobin. We observed (i) significantly higher levels of 8-OHdG, 8-epi-PGF2α, and carbonylated proteins in saliva of periodontal patients as compared with controls (P = 0.0003, < 0.0001 and < 0.0001); (ii) 8-OHdG, 8-epi-PGF2α, and carbonylated proteins were independently negatively associated with CPITN (P = 0.004, 0.02, and < 0.0001); (iii) a positive correlation between salivary TAC and periodontal disease status in the study group (P < 0.0001); and (iv) specific oxidation of transferrin, human IgG1 heavy chain fragment, and salivary amylase in periodontitis. Periodontal disease is associated with increased oxidative modification of salivary DNA, lipids, and proteins. Augmented salivary total antioxidant capacity may represent an adaptive response to oxidative stress. Salivary amylase, transferrin, and human IgG1 heavy chain fragments are particularly prone to enhanced oxidation in periodontitis.     

Histochemical evidence for inducible nitric oxide synthase in advanced but non-ruptured human atherosclerotic carotid arteries

In response to cytokine stimulation, the inducible isoform of the nitric oxide synthase (iNOS) produces large amounts of nitric oxide with potential consequences in the pathophysiology of atherosclerosis. Previous investigations have demonstrated the presence of iNOS in human atherosclerotic lesions. The goal of this study was to evaluate the occurrence of the expression of iNOS in ruptured versus non-ruptured human carotid atherosclerotic plaques. Using plastic-embedded sections, we performed in situ hybridization and immunohistochemistry on very advanced atherosclerotic lesions type V (non-ruptured) and type VI (ruptured) from 12 atheromatous carotid arteries from endarterectomy and six non-atherosclerotic internal mammary arteries from aorto-coronary bypass. Only one internal mammary artery expressed iNOS in the endothelium. In contrast, iNOS mRNA and protein were repeatedly expressed in advanced lesions type V in 5/7 cases, particularly in inflammatory regions. Specific cell markers identified iNOS-positive cells as macrophages and T-lymphocytes but also as smooth muscle cells and endothelial cells adjacent to these inflammatory regions. Nitration of protein tyrosines was not always associated to iNOS expression but more likely to the presence of inflammatory cells. In complicated lesions type VI, the occurrence of iNOS mRNA and protein expression diminished drastically (1/5 cases). Combined expression of iNOS mRNA and protein is frequently found in advanced but non-ruptured human atherosclerotic carotid lesions while it becomes rare after the plaque has ruptured. These findings suggest that iNOS could be an active participant in the plaque rupture event.     

C-reactive protein levels are associated with the progression of atherosclerotic lesions in rabbits

Elevated plasma levels of C-reactive protein (CRP) are associated with increased risk of cardiovascular disease. CRP immunoreactive protein is also detected in the lesions of atherosclerosis. However, it is not known whether the CRP contents of atherosclerotic lesions are associated with the initiation and progression of atherosclerosis. To examine this hypothesis, we investigated different types of atherosclerotic lesions of rabbits fed with a cholesterol-rich diet for 6, 12, 16, and 28 weeks and examined their relationship with CRP. We measured the aortic atherosclerotic area, macrophages, and smooth muscle cells along with CRP contents in the lesions. Atherosclerotic lesions of aortas began to form at 6 weeks and were characterized by accumulation of macrophages in the intima, and lesions became more fibrotic in the advanced stage. Both plasma CRP levels and the lesional CRP contents were associated with the lesion size. Our results suggest that plasma CRP, as well as lesional CRP, associated with the formation and progression of atherosclerotic lesions.     

Serum leptin levels in septic men correlate well with C-reactive protein (CRP) and TNF-alpha but not with BMI

Leptin, an adipocyte-derived signaling factor, is a member of the IL-6 cytokine family. However there is no direct evidence of leptin stimulation of the acute phase protein (APP) synthesis which is typical for all other IL-6-like factors. The purpose of this study was to characterize the dynamics of circulating leptin in relation to ten APPs. We used postoperative septic patients as a model of cytokine network hyperstimulation and intensive APP reaction. The prospective study was performed on 22 patients with proven postoperative intraabdominal sepsis after large abdominal surgery. Plasma levels of leptin, TNF-alpha, IL-1beta, soluble IL-2 receptor (sIL-2R), IL-6 (ELISA analysis) and ten APPs (nephelometric analysis) were estimated. We have demonstrated a statistically significant elevation of plasma leptin concentrations in the septic group compared with healthy subjects (p<0.001). The correlation of plasma leptin and BMI during postoperative sepsis was diminished. The regression coefficient was the highest for leptin and CRP (r=0.48, p<0.05), and for leptin and alpha-1-antitrypsin (r=0.46, p<0.05) in the septic group. There was significant correlation between TNF-alpha and leptin (r=0.47, p<0.05) and between IL-6 and leptin (r=0.45, p<0.05) in septic patients. No significant correlation was found between leptin and "negative" APP and between leptin and IL-1beta. Leptin has thus been shown as an acute phase reactant with a potential hematopoietic, immunomodulatory and hepatocyte stimulating activity during the infectious and non-infectious stress response. The significant correlation between leptin and CRP and leptin and alpha-1-antitrypsin indicates that leptin can participate in APP synthesis regulation during a systemic inflammatory response.     

Correlation of iron and zinc levels with lesion depth in newly formed atherosclerotic lesions

Several studies have indicated a relationship between body iron content and cardiovascular disease, although other studies have not. There are also suggestions that zinc has an antioxidant and antiatherosclerotic effect. We have used Nuclear Microscopy, using the combination of Scanning Transmission Ion Microscopy (STIM), Rutherford Backscattering Spectrometry (RBS), and Proton Induced X-ray Emission (PIXE) to map and quantify iron and zinc levels in newly formed atherosclerotic lesions. Sixteen New Zealand White rabbits fed on a high cholesterol diet were divided into four groups of 4 rabbits each. Six weeks into the high cholesterol diet, two groups were treated with the iron chelating agent desferrioxamine, for 2 weeks and 4 weeks, respectively, by surgically implanting with Alzet osmotic pumps (Alza Corporation, Palo Alto, CA, USA) containing desferal (0.5 g/ml). The other two groups served as controls, and were surgically implanted with osmotic pumps containing saline. Tissue sections were taken from the aortic arch, flash frozen, and air-dried. Analysis of atherosclerotic lesions indicated a trend (p =.07) to a reduction in the progression of the lesion after 4 weeks of desferrioxamine treatment. For each of the control and desferrioxamine-treated animals however, the more extensive lesions contained a higher concentration of iron and a lower concentration of zinc. Our results are consistent with the view that early lesion formation may be accelerated by free radical production caused by increased iron levels, that zinc might antagonize such effects, and that more prolonged desferal treatment might have an antiatherosclerotic effect.     

Modified low density lipoprotein isolated from atherosclerotic lesions does not cause lipid accumulation in aortic smooth muscle cells

Foam cells in atherosclerotic lesions are derived not only from blood monocytes but also from smooth muscle cells (SMC). To better understand the mechanisms by which SMC may become lipid-laden, we have studied the catabolism by cultured rabbit aortic SMC of LDL derived from atherosclerotic lesions (A-LDL) previously shown to be chemically modified. A-LDL was isolated either from homogenates of atherosclerotic plaques in human aortas by affinity chromatography and gel filtration, or from nonhomogenized extracts of plaque minces by ultracentrifugation and gel filtration. Internalization of A-LDL by SMC or fibroblasts appeared to be mediated primarily via the LDL receptor since: 1) either unlabeled LDL or A-LDL could inhibit the degradation of 125I-labeled A-LDL or of 125I-labeled LDL, 2) the uptake of both A-LDL and LDL, as estimated by their abilities to stimulate cholesterol esterification, was reduced in cells in which LDL receptor expression was down-regulated; and 3) the uptake of both [3H]cholesteryl ether-labeled A-LDL and LDL by normal fibroblasts was significant and could be inhibited by excess LDL, but was negligible in receptor-negative fibroblasts. At saturating concentrations of lipoproteins, maximum cholesterol esterification in SMC was greater for LDL than for A-LDL. Over a 48-h incubation, A-LDL, like LDL, was unable to induce cellular cholesteryl ester accumulation. Cross-competition studies suggested that either the affinity of A-LDL for the LDL receptor was less than that of LDL, or that some particles in A-LDL are not internalized by SMC. The latter alternative was supported by the observations that some A-LDL particles had undergone aggregation, especially at high concentrations, and that aggregated forms of A-LDL or plasma LDL failed to be internalized and degraded by SMC. Collectively, these results are consistent with recognition of some of the A-LDL particles by the LDL receptor, but also suggest that, at least under in vitro conditions, A-LDL is unlikely to induce lipid accumulation in SMC resulting in SMC-derived foam cells.     

Expression of caspase-3 and -7 does not correlate with the extent of apoptosis in primary breast carcinomas

Association between the rate of apoptosis and expression of the several relevant molecules (Bcl-2, pro- and active caspase-3, and caspase-7) was studied in 61 primary breast carcinomas. The rate of apoptosis detected both morphologically and by the TUNEL assay appeared to be high in 18 (30%), moderate in 14 (23%), and low in 29 (48%) carcinomas. High apoptotic index was strongly associated with advanced tumor grade and estrogen receptor positive (ER+) status but not with other investigated clinical or morphological parameters. Among the molecules studied, only the Bcl-2 protein expression demonstrated strong (inverse) correlation with the apoptotic index (p = 0.032). The data of this expected correlation was served as internal control in the study. Interestingly, high levels of the anti-apoptotic protein Bcl-2 was frequently co-incident with increased expression of pro-apoptotic molecules, such as active caspase-3 (p = 0.004) and caspase-7 (p = 0.001). However, expression of caspase-3 or caspase-7 did not show correlation with the extent of apoptosis or any clinico-morphological features, except overrepresentation of ER+ status in tumors expressing caspase-3 (p = 0.009). Thus, these findings indicate a general dysregulation of spontaneous apoptosis in primary breast tumors.     

Attenuation of a human rotavirus vaccine candidate did not correlate with mutations in the NSP4 protein gene.

The NSP4 protein of a simian rotavirus was reported to induce diarrhea following inoculation of mice. If NSP4 is responsible for rotavirus diarrhea in humans, attenuation of a human rotavirus may be reflected in concomitant mutations in the NSP4 gene. After 33 passages in cultured monkey kidney cells, a virulent human rotavirus (strain 89-12) was found to be attenuated in adults, children, and infants. Nucleotide sequence analysis of the NSP4 protein gene revealed only one base pair change between the virulent (unpassaged) and attenuated 89-12 viruses, which resulted from a substitution of alanine for threonine at amino acid 45 of the encoded NSP4 protein. Because both threonine and alanine have been found at position 45 of NSP4 in symptomatic and asymptomatic human rotaviruses, neither amino acid in this position could be established as a marker of virulence. Therefore, attenuation of rotavirus strain 89-12 appears to be unrelated to mutations in the NSP4 gene.