The different effect of alpha and gamma interferons and interleukin 2 on the expression of CD2, CD3, CD4 and CD8 antigens in comparison to histocompatibility antigens of human lymphocytes

The effects of alpha- and gamma-interferons (IFN-alpha, -gamma) and of interleukin 2 (IL-2) on the expression of certain differentiation antigens were compared with those of major histocompatibility antigens on human lymphocytes. IFN-gamma and IFN-alpha in high doses significantly increased the expression of T11 (CD2) differentiation antigen, but did not affect the expression of T4 (CD4), T8 (CD8), T3 (CD3) and Leu-7 antigens (HNK-1). Both natural and recombinant IFN-alpha and -beta apparently increased the expression of HLA-ABC antigens and of beta-2 microglobulin (beta 2m) after 16 h incubation. The amount of HLA-DR antigen, however, doubled in a few hours following IFN-gamma treatment. IL-2 affected the expression of CD2 and CD8 antigens only marginally, but did not affect that of CD3 and Leu-7; however, it strongly enhanced the expression of HLA-ABC, HLA-DR, and beta 2m antigens.     

Identification of two amino acid residues on Ebola virus glycoprotein 1 critical for cell entry

Using site-directed mutagenesis and retroviral vector pseudotyping of the wild type or mutated glycoprotein of Zaire ebolavirus (ZEBOV), we analyzed 15 conserved residues in the N-terminus of the filovirus glycoprotein 1 (GP1) in order to identify residues critical for cell entry. Results from infectivity assays and Western blot analyses identified two phenylalanine residues at positions 88 and 159 that appear to be critical for ZEBOV entry in vitro. We extended this observation by introduction of alanines at either position 88 or 159 of Ivory Coast Ebolavirus (CIEBOV) and observed the same phenotype. Further, we showed that introduction of each of the two mutations in a recombinant full-length clone of ZEBOV (Mayinga strain) that also carried the coding sequence for GFP could not be rescued, suggesting the mutants rendered the virus non-infectious. The two phenylalanines that are critical for both ZEBOV and CIEBOV entry are found in two linear domains of GP1 that are highly conserved among filoviruses, and thus could provide a target for rational development of broadly cross-protective vaccines or antiviral therapies.     

Differentiation antigens on rhesus monkey lymphocytes. I. Identification of T cells bearing CD3 and CD8, and of a subset of CD8-bearing cells

Rhesus monkeys provide an excellent preclinical model to test the effect of monoclonal antibodies (mAb) in vitro and in vivo. So far, mostly mAb have been used which were originally raised against human cell surface antigens but cross-reacted reasonably well with homologous antigens on rhesus monkey cells. However, to optimize the model, it was necessary to produce mAb which react specifically with subsets of rhesus monkey lymphocytes. In this report, three mouse anti-rhesus monkey mAb are described, specific for different subsets of rhesus monkey T lymphocytes. None of the reagents cross-reacts with human lymphocytes. Characterization of these mAb was based upon indirect immunofluorescence, using a simultaneous staining technique, and immunoprecipitation of the specific target antigens. One antibody (GM9) reacts with the same subset as is recognized by mAb specific for human CD8+ cells. The second mAb (GM13) is specific for a subset of CD8+ cells. A third mAb (FN18) was of particular interest: it identifies a cell surface complex, RhT3, expressed on mature T lymphocytes, of which the polypeptide chains have a molecular mass of 22 and 27 kDa. The data strongly suggest that RhT3 is a CD3-like determinant, so far unidentified in the rhesus monkey.     

Identification of amino acid residues critical for endonuclease and integration activities of HIV-1 IN protein in vitro.

HIV-IN protein, tagged with a hexahistidine tail was expressed in Escherichia coli and purified by a one-step nickel chelate affinity chromatography procedure. The purified IN protein was characterized in terms of its endonuclease and integrase properties in vitro. Specific cleavage and integration of HIV U5 LTR ends were observed in the presence of 2-5 mM Mg2+ or Ca2+. In the presence of 2 mM Mn2+, cleavage and integration occurred at additional sites indicating a decreased specificity. The properties of mutant IN proteins were examined in vitro. Deletion of 39 amino acids from the N-terminus and a minimum of 25 residues from the C-terminus impaired IN-mediated cleavage and integration activities. The results of site-directed mutagenesis experiments showed that residues critical for IN function are highly conserved. Mutations of conserved residues Asp64, Pro109, Asp116, and Glu152 adversely affected IN function in vitro. Mutations of nonconserved residues Gly189 and Thr112 had no effect. Mutation of a conserved Thr115 to Ala caused a near complete loss of Mg(2+)-dependent integration activity, but only partially effected endonucleolytic cleavage activity of IN. These results suggest that not all conserved residues are involved in both endonucleolytic cleavage and integration activities of HIV-IN.     

CEACAM1 (CD66a) mediates delay of spontaneous and Fas ligand-induced apoptosis in granulocytes

Granulocytes form the first and fastest line of defense against pathogenic infections. Their survival is limited by apoptosis, a process that is critical for the resolution of inflammation. Pro-apoptotic and pro-inflammatory cytokines, as well as several receptors, can alter the lifespan of granulocytes. Here we report that the carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1, CD66a) is involved in the regulation of granulocyte survival. Until now CEACAM1 is described to control cell proliferation, cell migration, tumor growth, angiogenesis and diverse leukocyte functions. However, very little is known about its role in granulocytes. We found that CEACAM1 expression in resting rat granulocytes is significantly higher than in other leukocyte subtypes. Stimulation led to a strongly increased CEACAM1 cell surface expression and to release of soluble CEACAM1. DNA fragmentation assays and annexin V staining revealed that binding of CEACAM1-specific antibodies, Fab fragments and soluble CEACAM1-Fc constructs to cell surface-expressed CEACAM1 causes a delay of spontaneous and Fas ligand (CD95L)-induced apoptosis. Tyrosine phosphorylation of CEACAM1-L, its association with SHP-1, the activation of Erk1/2 and caspase-3 appeared to be crucial for the CEACAM1-mediated anti-apoptotic effect. These findings provide evidence that CEACAM1 influences the resolution of inflammation by prolonging the survival of rat granulocytes.     

降钙素原、中性粒细胞CD64、IL-6和CRP检测在新生儿感染性疾病中的诊疗价值

为探讨降钙素原(procalcitonin,PCT)、中性粒细胞CD64、IL-6和CRP早期诊断新生儿感染性疾病及判断预后的价值,选择感染患儿,根据临床表现及血常规、细菌学培养等检查结果确定为细菌组60例、病毒组50例,分别在入院治疗前及治疗3 d后采集空腹血进行PCT、中性粒细胞CD64、IL-6及CRP水平测定,对照组(60例,同期住院的非感染患儿)检测其PCT、中性粒细胞CD64、IL-6及CRP水平,进行统计学分析。采用双抗体夹心免疫层析法测定血清PCT和IL-6水平;采用FCM测定中性粒细胞CD64百分比;采用Aristo特定蛋白分析仪测定CRP水平。结果显示,治疗前细菌组PCT、中性粒细胞CD64、IL-6和CRP水平均明显高于病毒组和对照组(P0.05);病毒组与对照组比较,PCT、中性粒细胞CD64、IL-6和CRP水平差异无统计学意义(P 0.05)。PCT、中性粒细胞CD64、IL-6和CRP诊断新生儿细菌感染性疾病的截断值分别为:0.56 ng/mL、7.01%、8.79 pg/mL、14.5 mg/L。治疗后有效组PCT、中性粒细胞CD64、IL-6和CRP水平与治疗前比较均明显下降(P0.05),治疗后无效组PCT、中性粒细胞CD64和IL-6水平与治疗前比较均无明显变化(P 0.05)。提示PCT作为早期检测感染性疾病的标志物,对新生儿细菌感染更为敏感,而PCT、中性粒细胞CD64、IL-6和CRP可作为新生儿感染性疾病早期诊断、疗效判断的重要指标。     

Modulation of CD8 and CD3 by HIV or HIV antigens

To investigate whether human immunodeficiency virus (HIV)-1 and HIV-1 antigens modulate surface and cytoplasmic CD8 or CD3, as well as CD4, we used cell permeabilization reagents, surface/cytoplasmic fluorescent staining, multiparameter flow cytometric techniques and an in vitro culture system in which relatively few lymphocytes are actively infected with HIV. Human peripheral blood lymphocytes were: not stimulated, not stimulated but HIV-inoculated, phytohaemagglutinin (PHA)-stimulated, PHA/HIV-inoculated (PHA/HIV), or placed into media with soluble gp120, Rev or Nef. HIV inoculation and Nef had striking modulatory effects on CD8. The cytoplasmic CD8 median fluorescent intensity (MFI) of positive lymphocytes was lower for cells in unstimulated/HIV-infected cultures than unstimulated cultures (44 versus 62% of ex vivo value, P = 0.032) and lower for cells in PHA/HIV cultures than in PHA cultures (56 versus 100% of ex vivo, P = 0.041). The surface CD8 MFI values for Nef were significantly lower than the ex vivo value (75% of ex vivo, P = 0.006). At days 2-7 of culture, Rev was associated with slight reductions in surface CD4 MFI (58% of ex vivo versus 78% of ex vivo for unstimulated cultures, P = 0.047) and greater effects on cytoplasmic CD3 MFI (131 versus 179% of ex vivo for unstimulated cultures, P = 0.035), and surface CD8 MFI (70% of ex vivo, P = 0.006 versus ex vivo value). The globality of Rev's effects suggests these are related to a shared processing pathway, i.e. not due to direct interaction with CD3, CD4 and CD8; the effects of HIV inoculation and Nef on CD8 expression appear to be more CD8 specific. Because CD8 is essential for cytotoxic T-cell function, its down-modulation could inhibit this activity, including anti-HIV cytotoxicity. Given the critical roles of CD3 and CD8 in T-lymphocyte signal transduction and antigen responsiveness, the effects of HIV, Rev and Nef on these molecules have clinically significant implications concerning the pathogenesis and treatment of HIV.     

Biochemical comparison of the T6 antigen and HLA-A,B antigens

The human thymic differentiation antigen T6, which was found to be associated with beta 2-microglobulin, was compared to the HLA-A,B antigens. Using a heteroantiserum prepared against denatured heavy chains of HLA-A,B antigens, no cross-reactivity with denatured T6 could be detected. The molecular weight of the protein backbone of T6 was found to be 34,000 as compared to 40,000 for the HLA-A,B antigens. Also, not only was the percentage of carbohydrate of T6 (25-35%) different from the HLA-A,B antigens (10%), but lectin binding studies showed that their sugar composition may differ. The two forms of T6, which previously had been found on MOLT-4 cells, appeared to have different levels of glycosylation, but apparently had the same protein backbone. T6, like HLA, has a hydrophobic domain, since it could be labeled with [125I]iodonaphthylazide. We conclude from these studies that T6 may be a class I MHC antigen which is different from the classical HLA-A,B antigens.     

Unusual cytotoxic activities of thymus-independent, self-antigen-specific CD8(+) T cells

We compared the cytotoxic activities of thymus-dependent and thymus-independent CD8(+) T cells. Thymus-dependent CD8(+) T cells, which are foreign antigen specific, acquired cytotoxic activity to tumor cells with a basal dose of the antigen peptides and to hybridoma cells expressing anti-TCR mAb only after differentiation into effector cytotoxic T lymphocytes (CTL). In contrast, thymus-independent CD8(+) T cells, which have been shown to be self-antigen specific, never showed cytotoxic activity to the target cells with a basal dose of the self-antigen peptide, while they could lyse hybridoma cells expressing anti-TCR mAb even without prior antigenic stimulation. Furthermore, the ex vivo cytotoxic activity of thymus-independent CD8(+) T cells was also observed against the target cells with high doses of the antigen peptides, which were not lysed by freshly isolated thymus-dependent CD8(+) T cells. Thus it is revealed that thymus-independent, self-antigen-specific CD8(+) T cells already acquire mature CTL functions in situ but have an increased threshold of TCR-mediated signaling for activation. These differences in cytotoxic activities between thymus-dependent and thymus-independent CD8(+) T cells suggest distinct roles of the two subsets of CD8(+) T cells in vivo.     

Identification of residues critical for enzymatic activity in the domain encoded by exons 8 and 9 of the human inducible nitric oxide synthase

Overproduction of nitric oxide (NO) by inducible NO synthase (iNOS) has been implicated in the pathogenesis of several diseases including airway inflammation of asthma. iNOS is active only as a homodimer. We previously demonstrated that the region encoded by exons 8 and 9 is critical for dimerization. In this study, alanine-scanning mutagenesis was used to identify critical amino acids in that region by expression of mutant proteins in human embryonic kidney 293 cells. All iNOS mutants yielded iNOS protein as detected by Western analysis. Four iNOS mutants with alanine replacing Trp260, Asn261, Tyr267, or Asp280 did not generate NO. Dimer formation was tested by sodium dodecyl sulfate polyacrylamide gel electrophoresis at 4 degrees C, followed by immunoblotting. Wild-type iNOS migrated both as monomers and dimers. iNOS mutants with alanine replacing Trp260, Asn261, or Tyr267, however, migrated only as monomers, suggesting that their inability to produce NO is related to a defect in dimer formation. Interestingly, the Asp280 mutant retained the ability to dimerize, indicating that it represents an inactive form of an iNOS dimer. These data identify four amino acids in exons 8 and 9 critical for iNOS activity, three of which also influence dimerization. These residues are strictly conserved among all NOS isforms and across species. Thus all NOS isoforms share general structural similarities, including specific amino acids critical for dimerization and catalytic activity. These data increase our understanding of the structural elements critical for NO synthesis and lay the groundwork for future studies aimed at downregulation of iNOS activity.     

Characterization of two sheep lymphocyte differentiation antigens, SBU-T1 and SBU-T6

The monoclonal antibodies 25.91 and 20.27 define two lymphocyte cell surface antigens of sheep. 25.91 is reactive with 60-80% of lymphocytes and 98% of thymocytes, and only stains surface immunoglobulin-negative peripheral lymphocytes. 25.91 immuno-precipitates a 67,000 MW protein from lymphocyte lysates under both reducing and non-reducing conditions, whereas immunoprecipitation of thymocyte lysates reveals a 67,000, 62,000 MW complex. The tissue distribution and molecular weight analysis reported here for the antigen recognized by 25.91 indicate that this antigen is the sheep homologue of the human T1 and mouse Ly 1 antigens. The monoclonal antibody 20.27 is reactive with 80% of thymocytes and the majority of cell surface immunoglobulin-positive peripheral blood lymphocytes (B cells), but is unreactive with peripheral blood T cells. 20-27 also stains Langerhans cells in skin tissue sections and large dendritic-like cells in the paracortex sections and large dendritic-like cells in the paracortex of lymph node tissue sections. Immunoperoxidase staining of thymus tissue sections with 20.27 shows intense staining of cortical thymocytes and an absence of staining within the medulla. Molecular weight analysis of the 20.27 antigen reveals two major bands of 46,000 and 12,000 MW under both reducing and non-reducing conditions. The 20.27 antigen has properties resembling MHC class I-like antigens such as T6 in the human and TL in the mouse.     

Presentation of epstein-barr virus latency antigens to CD8(+), interferon-gamma-secreting, T lymphocytes

Epstein-Barr virus (EBV) infects more than 95 % of the human population and causes an asymptomatic life-long infection in the majority of EBV carriers. Cell-mediated immunity provides resistance to EBV, as demonstrated by the occurrence of EBV-induced post-transplant lymphoproliferative disease in immunosuppressed patients. Here we looked for IFN-gamma-producing T lymphocytes in the blood of healthy donors with a rapid enzyme-linked immunospot (ELISPOT) assay, comparing as antigen presenting cells monocytes and dendritic cells (DC) infected with recombinant vaccinia virus (rVV). We found a strong CD8(+) ELISPOT response to one or more of the EBNA 3A, 3B and 3C antigens in the PBMC from 14 / 18 donors. The sensitivity of the overnight ELISPOT assay was increased using DC as antigen-presenting cells, including 3 / 3 individuals who lacked ELISPOT in PBMC. In addition, DC could markedly expand EBV-specific spots after a 7-day culture. In a smaller number of donors, we documented recognition of the subdominant LMP 1, LMP 2 and EBNA 1 antigens that are expressed in a variety of EBV-associated malignancies. Therefore our data provide more evidence for the efficacy of DC in eliciting rapid responses to EBV latency antigens in circulating CD8(+) T cells.     

Structural and mutational analyses of a CD8alphabeta heterodimer and comparison with the CD8alphaalpha homodimer

The crystal structure of a recombinant mouse single chain CD8alphabeta ectodomains at 2.4 A resolution reveals paired immunoglobulin variable region-like domains with a striking resemblance to CD8alphaalpha in size, shape, and surface electrostatic potential of complementarity-determining regions (CDR), despite <20% sequence identity between the CD8alpha and CD8beta subunits. Unlike the CD8alpha subunit(s) in the heterodimer or homodimer, the CDR1 loop of CD8beta tilts away from its corresponding CDR2 and CDR3 loops. Consistent with this observation, independent mutational studies reveal that alanine substitutions of residues in the CDR1 loop of CD8beta have no effect on CD8alphabeta coreceptor function, whereas mutations in CD8beta CDR2 and CDR3 loops abolish CD8alphabeta coreceptor activity. The implications of these findings and additional CD8alpha mutational studies for CD8alphabeta- versus CD8alphaalpha-MHCI binding are discussed.     

Identification of chromosomes 3, 6, and 8 aberrations in uveal melanoma by microsatellite analysis in comparison to comparative genomic hybridization

In uveal melanoma, monosomy 3 is strongly associated with metastic disease and poor prognosis. Cytogenetic analysis and comparative genomic hybridization (CGH) have been used to identify chromosomal aberrations in uveal melanoma. As these methods are costly and time consuming in routine diagnostic settings, we evaluated whether tumors with monosomy 3 can be reliably identified by microsatellite analysis (MSA). In addition, we also tested if aberrations of chromosomes 6 and 8, which have also been associated with the course of the disease, can be detected by MSA. We established a protocol for MSA of 23 markers, 3-4 on each arm of chromosomes 3, 6, and 8. Twenty tumors were analyzed by CGH and MSA, and 10 tumors were analyzed by MSA only. For chromosome 3, the results of CGH and MSA were concordant, thus indicating that the dosage of this chromosome can reliably be determined by MSA. However, MSA failed to detect copy number gains at 6p in some tumors. Moreover, despite quantitative evaluation of allele ratios, it was not possible to discern 8p losses and gains reliably. We thus conclude that while MSA can be used to determine monosomy 3 in uveal melanoma, careful interpretation of results for chromosomes 6 and 8 is recommended.     

Identification of tyrosine residues crucial for CD200R-mediated inhibition of mast cell activation

CD200 and its receptor CD200R are type-1 membrane glycoproteins, which contain two immunoglobulin-like domains. Engagement of CD200R by CD200 inhibits activation of myeloid cells. Unlike the majority of immune inhibitory receptors, CD200R does not contain an immunoreceptor tyrosine-based inhibitory motif but contains three tyrosine residues (Y286, Y289, and Y297) in the cytoplasmic domain. Y297 is located in an NPxY motif. Previously, we have shown that engagement of CD200R in mouse mast cells induces its tyrosine phosphorylation and recruitment of inhibitory adaptor proteins Dok1 and Dok2, leading to the inhibition of Ras/mitogen-activated protein kinase activation. In the present study, we examined the roles of these three tyrosines in CD200R-mediated inhibition by site-directed mutagenesis in mouse mast cells. Our data show that Y286 and Y297 are the major phosphorylation sites and are critical for CD200R-mediated inhibition of mast cell activation, and Y289 is dispensable. Our data also suggest that the Src family kinase may mediate the phosphorylation of CD200R and Dok.     

Identification and characterization of CD8+ suppressor T cells

It has long been appreciated that certain subsets of T cells are capable of suppressing immune reactions. Initially, such T cells were described as CD8⁺ suppressor T cells (Ts) and there is a vast body of research spanning 30 years that describes the immunobiology of CD8⁺ Ts. However, studies on CD8⁺ Ts have suffered from the inability to distinguish CD8⁺ Ts from CD8⁺ T cells of other phenotypes. Here we present a brief history of CD8⁺ Ts, a review of recent progress distinguishing CD8⁺ Ts as a unique subset of CD8⁺ T cells, and an overview of the evolving immunological context in which CD8⁺ Ts function.     

Identification and characterization of CD8+ suppressor T cells

It has long been appreciated that certain subsets of T cells are capable of suppressing immune reactions. Initially, such T cells were described as CD8+ suppressor T cells (Ts) and there is a vast body of research spanning 30 years that describes the immunobiology of CD8+ Ts. However, studies on CD8+ Ts have suffered from the inability to distinguish CD8+ Ts from CD8+ T cells of other phenotypes. Here we present a brief history of CD8+ Ts, a review of recent progress distinguishing CD8+ Ts as a unique subset of CD8+ T cells, and an overview of the evolving immunological context in which CD8+ Ts function.     

Determination of co-expression of activation antigens on proliferating CD4+, CD4+ CD8+ and CD8+ lymphocyte subsets by dual parameter flow cytometry

The incidence of activation markers on proliferating CD4+, CD4+ CD8+ and CD8+ lymphocyte subsets was determined in a single laser Epics-C fluorescence-activated cell sorter system, using a series of double staining combinations. Experiments were performed after 3 days of culture with PHA on cell fractions enriched for CD4+ or CD8+ lymphocytes before initiation of culture. The percentage of CD4+, CD4+ CD8+ and CD8+ lymphocytes in the total population was determined using double staining with Leu3 PE for the detection of CD4+ cells, and Leu2 FITC for the detection of CD8+ cells. Next, double stainings with Leu3 and Leu2 antibodies conjugated with PE and antibodies directed against activation markers (M) IL-2 receptor, transferrin receptor, HLA-DR antigen and CALLA conjugated with FITC were performed, using the following combinations: Leu3 and Leu2/M, Leu3/M and Leu2/M. The expression of activation markers on CD4+ CD8+ lymphocytes was calculated from the results. Our findings indicate that CALLA is expressed on most CD4+ and all CD4+ CD8+ cells, and on a small percentage of CD8+ lymphocytes; the IL-2 receptor was expressed on most CD4+ cells, on approximately three-quarters of CD4+ CD8+ cells and half the CD8+ cells; HLA-DR was expressed on a small percentage of CD4+ cells, all CD4+ CD8+ cells and half of CD8+ cells. The transferrin receptor was almost exclusively expressed on CD4+ CD8+ cells. The standard deviation of the calculated values did not exceed 13% and this analysis can generally be applied to determine the co-expression of a third marker in a mixture of single and double stained cells using conventional methods.     

Subcellular localization of CD66, CD67, and NCA in human neutrophils.

CD66 and CD67 are granulocyte-specific activation antigens; their surface expression is up-regulated when neutrophils are activated. CD66 antibodies recognize an approximately 180-kd neutrophil surface protein that is also recognized by anti-carcinoembryonic antigen (CEA) antibodies and is therefore a nonspecific cross-reacting antigen (NCA). CD67 antibodies recognize an approximately 100-kd neutrophil surface protein that is attached to the membrane via a glycosyl-phosphatidylinositol anchor. To identify an intracellular pool from which CD66 and CD67 could be up-regulated, the subcellular distribution of proteins recognized by CD66 and CD67 monoclonal antibodies and polyclonal anti-CEA was studied. Neutrophil plasma membranes, granules, and cytoplasm were prepared by nitrogen cavitation and differential centrifugation and then analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting. Most of the 180-kd protein recognized by CD66 antibodies and the 100-kd protein recognized by CD67 antibodies were located in the secondary granule fraction, with lesser amounts detectable in the plasma membrane fraction. Several NCA species ranging from approximately 40 to 200 kd were identified, and the distribution of these NCAs was different in the primary granules, secondary granules, and plasma membrane fractions. The major NCAs in the plasma membrane fraction were of approximately 95 to 100 and approximately 180 to 200 kd; the secondary granule fraction contained major NCAs of approximately 42, 85, 95 to 100, and 180 to 200 kd. NCAs were also detected in the primary granule fraction, the most prominent being of approximately 90-100 kd; no NCA of approximately 180 to 200 kd was detected in the primary granules. The presence of CD66, CD67, and NCAs in the secondary granules suggests secondary granules as a likely source from which these antigens could be recruited to the cell surface with activation. The potential role for NCAs in the primary granules is unknown.     

Dichotomy of two CD8+ lymphocyte subsets in HIV infection. Depletion of CD8+ CD3- and expansion of CD8+ CD3+ subsets: consequence on the CD4/CD8 ratio.

In order to highlight the underlying mechanism(s) of the CD8 lymphocyte expansion in the HIV infection, two distinct CD8 subsets were analysed: T CD8bright+ CD3+ with MHC-restricted activity, and non-T CD8dim+ CD3-, which performs natural killer (NK) activity. It consists of a cross-sectional study including 168 HIV-infected patients (74 CDC stage II, 48 CDC stage III and 46 CDC stage IV) compared among them and to 60 healthy individuals. We observed an expansion of CD8+ CD3+ cells which masks a depletion of CD8+ CD3-. The comparative study showed that the expansion of the CD8+ CD3+ is relatively higher than that of total CD8+ lymphocytes and that the depletion of the CD8+ CD3- subset is severe, begins early and remains constant through the HIV progression. The comparison of CD4/CD8 and CD4/CD8+ CD3+ ratios showed that the latter could possibly be a better indicator in the HIV infection. The mechanism of inverted CD4/CD8 ratio in healthy individuals was also clarified. The CD8+ CD3+, CD8+ CD3- and CD4/CD8+ CD3+ parameters would be more specific markers than total CD8 and CD4/CD8 ratio especially in therapy trials.