CD127 expression and regulation are altered in the memory CD8 T cells of HIV-infected patients--reversal by highly active anti-retroviral therapy (HAART).

HIV infection activates abnormally the immune system and the chronic phase is accompanied by marked alterations in the CD8 compartment. The expression of CD127 (IL-7R alpha chain) by memory CD8 T lymphocytes in HIV-infected patients is analysed and reported. The memory CD8 T cell subset was characterized by expression of CD45RA and CD27 markers, and CD127 cell surface expression was measured ex vivo by four-colour flow cytometry. HIV infection was associated with a fall in the proportion of CD127(+) cells among memory CD8 lymphocytes that resulted in a higher CD127(-) CD45RA(-)CD27(+) CD8 T cell count in HIV-infected patients. Diminished CD127 cell surface expression [mean fluorescence intensity (MFI)] by positive cells was also observed in this subset. The data suggest that these defects were reversed by highly active anti-retroviral therapy (HAART). The regulation of CD127 expression was also studied in vitro. Down-regulation of CD127 by interleukin (IL)-7 was observed in memory CD8 lymphocytes from healthy donors and HAART patients. Expression of CD127 by memory CD8 lymphocytes cultured in the absence of IL-7 confirmed that IL-7R regulation is altered in viraemic patients. Under the same experimental conditions, memory CD8 lymphocytes from HAART patients were shown to express CD127 at levels comparable to cells from healthy individuals. Altered CD127 cell surface expression and defective CD127 regulation in the memory CD8 T lymphocytes of HIV-infected patients are potential mechanisms by which these cells may be impeded in their physiological response to endogenous IL-7 stimulatory signals. Our data suggest that these defects are reversed during the immune reconstitution that follows HAART.

Leishmania major reaches distant cutaneous sites where it persists transiently while persisting durably in the primary dermal site and its draining lymph node: a study with laboratory mice.

So far, studies of Leishmania persistence in mice have used injections of parasites administered either intravenously in the tail vein or subcutaneously in the footpad. These routes poorly reflect the natural conditions when the sandfly delivers metacyclic promastigotes intradermally. In this study B10D2 and BALB/c mice were inoculated within the ear dermis with 10(4) Leishmania major metacyclic promastigotes. The parasite load was monitored by quantitative PCR in different tissues from the dermal inoculation site to distant tissues. The two sites of multiplication and persistence of parasites were the site of L. major inoculation and the draining lymph node (DLN), with a different pattern in the two mouse inbred lines. These two organs were the only sites harboring parasites 12 months postinoculation, with the DLN of BALB/c mice harboring around 10(7) parasites, a stable load from months 3 to 12. In these two sites, 8 and 12 months after inoculation, interleukin 4 (IL-4), gamma interferon, and inducible nitric oxide synthase transcripts parallel the parasite load while IL-10 transcript levels remain high. In addition, at early time points until month 3, parasite DNA was also detected in distant tissues such as the contralateral noninoculated ear or the tail skin, indicating that blood was at least transiently disseminating the parasites. In contrast, L. major DNA in liver, spleen, and femoral bone marrow remained sporadic in mice of both lines. This study is discussed within the framework of Leishmania transmission from the vertebrate host to the sandfly vector, a complex process still poorly understood.

Transient IFN-gamma synthesis in the lymph node draining a dermal site loaded with UV-irradiated herpes simplex virus type 1: an NK- and CD3-dependent process regulated by IL-12 but not by IFN-alpha/beta.

Our previous studies have shown that UV-inactivated, non-replicating herpes simplex virus type 1 (UV-HSV-1) triggers early and transient synthesis of IFN-alpha/beta in the mouse regional lymph node when delivered upstream (i.e. in the ear dermis). In this study, it is demonstrated, by use of a quantitative RT-PCR readout assay, that IFN-gamma mRNA expression was rapidly and transiently upregulated in draining lymph nodes when UV-HSV-1 was delivered in the ear dermis of C57Bl/6 mice. An increased number of IFN-gamma-producing cells was also detected in the lymph node by flow cytometric analysis. Two different subsets of cells, namely DX5(+) NK cells and CD3epsilon(+) T cells, accounted for this early IFN-gamma synthesis. Prompt upregulation of IFN-alpha and IL-12p40 mRNA was also recorded. We took advantage of IFN-alpha/beta-receptor knockout and wild-type 129 mice to study a potential role of IFN-alpha/beta in the signalling pathway leading to IFN-gamma transcription/translation. IFN-gamma mRNA upregulation still occurred in IFN-alpha/beta-receptor(-/-) mice, showing that IFN-alpha/beta was dispensable. The use of IL-12-neutralizing antibodies, prior to UV-HSV-1 delivery, confirmed the major role played by IL-12 in the early/transient IFN-gamma burst.

Listeria monocytogenes as a short-lived delivery system for the induction of type 1 cell-mediated immunity against the p36/LACK antigen of Leishmania major.

Listeria monocytogenes has been used as an experimental live vector for the induction of CD8-mediated immune responses in various viral and tumoral experimental models. Susceptibility of BALB/c mice to Leishmania major infection has been correlated to the preferential development of Th2 CD4 T cells through an early production of interleukin 4 (IL-4) by a restricted population of CD4 T cells which react to a single parasite antigen, LACK (stands for Leishmania homologue of receptors for activated C kinase). Experimental vaccination with LACK can redirect the differentiation of CD4(+) T cells towards the Th1 pathway if LACK is coadministrated with IL-12. As IL-12 is known to be induced by L. monocytogenes, we have tested the ability of a recombinant attenuated actA mutant L. monocytogenes strain expressing LACK to induce the development of LACK-specific Th1 cells in both B10.D2 and BALB/c mice, which are resistant and susceptible to L. major, respectively. After a single injection of LACK-expressing L. monocytogenes, IL-12/p40 transcripts showed a rapid burst, and peaks of gamma interferon (IFN-gamma)-secreting LACK-specific Th1 cells were detected around day 5 in the spleens and livers of mice of both strains. These primed IFN-gamma-secreting LACK-reactive T cells were not detected ex vivo after day 7 of immunization but could be recruited and detected 15 days later in the draining lymph node after an L. major footpad challenge. Although immunization of BALB/c mice with LACK-expressing L. monocytogenes did not change the course of the infection with L. major, immunized B10.D2 mice exhibited significantly smaller lesions than nonimmunized controls. Thus, our results demonstrate that, in addition of its recognized use for the induction of effector CD8 T cells, L. monocytogenes can also be used as a live recombinant vector to favor the development of potentially protective IFN-gamma-secreting Th1 CD4 T lymphocytes.

Cytokine transcripts in pediatric tuberculosis: a study with bronchoalveolar cells.

Pediatric tuberculosis (TB) differs from adult TB in many features. To date, cytokine expression has not been studied in children with TB. The relative amounts of the various cytokines released at the site of infection may be important determinants of TB disease development and pathology. We determined cytokine transcripts in bronchoalveolar cells (BACs) recovered from 9 children presenting with TB and from 9 children with pulmonary diseases other than TB. An RT-PCR-based method was developed to quantify the mRNAs encoding six cytokines (IFN- gamma, IL-12, TNF- alpha, IL-10, IL-4, TGF- beta 1) known to play key roles in mycobacterial infections. Expression of mRNA encoding TGF- beta, TNF- alpha and IFN- gamma was statistically significantly higher in BACs from children with TB than in BACs from children with other pulmonary diseases; whereas the levels of mRNA transcription for TGF- beta is high, the levels of mRNA transcription for IFN- gamma and TNF- alpha remain low. All children had low levels of mRNA for IL-12(p40). IL-4 was barely detectable in all cases. Children with miliary TB had high levels of IL-10 transcripts and low levels of mRNA encoding TGF- beta. The immunosuppressive cytokines TGF- beta and IL-10, are overproduced in children with non-miliary TB and miliary TB respectively and are probably involved in the progression of the disease. These data suggest that Th1 responses are reduced in children with TB.

IL-4 mRNA transcription is induced in mouse bone marrow-derived mast cells through an MHC class II-dependent signaling pathway.

We have previously shown that mouse bone marrow-derived mast cells (BMMC) can process and present immunogenic peptides to CD4 T cells. Here, we report on a T cell-dependent MHC class II-mediated mast cell activation resulting in IL-4 transcription and protein release. Presentation of optimal doses of ovalbumin peptide 323-339 resulted in IL-2 production by a specific T cell hybridoma and increase in IL-4 mRNA transcription in mast cells. IL-4 mRNA transcription increased by 200-fold in mast cells treated in IL-3/IL-4/granulocyte-macrophage colony-stimulating factor (high presenters) whereas only a tenfold increase or no increase were obtained with IL-3/IL-4/IFN-gamma- or IL-3-treated mast cells (low presenters), respectively. Induction of IL-4 mRNA transcription in purified mast cells by direct ligation of MHC class II molecules, using anti-I-A and anti-I-E-coated beads, indicates that MHC class II molecules are critical in this signaling pathway. However, when compared to T cells, anti-MHC class II-coated beads were less efficient, indicating a potential role of accessory molecules in this mast cell activation process. IgE-independent IL-4 production by mast cells as a result of cognate interaction with CD4 T cells could be critical for the development of type 2 responses. This novel mechanism may contribute to the induction and/or amplification of specific IgE-mediated allergic responses.

Quantitation of messenger RNA by competitive RT-PCR: a simplified read out assay.

A competitive RT-PCR method that permits reliable quantification of minute amounts of reverse-transcribed mouse lymph node mRNA is described. Using this technique, an absolute number of cDNA copies ranging from 10(3) to 10(5) can be determined, with a precision superior to 25%. The standard templates described in the present study permit the quantitation of beta-actin, IFN gamma, IL2, IL3, IL4, IL10, IL12 (p40 subunit), TGF beta 1, inducible nitric oxide synthase, ELAM-1, VCAM-1, and ICAM-1 mouse mRNA. The expression of a particular transcript is normalized to an arbitrary number of actin transcripts. The standard templates and wild-type cDNA have nearly identical sequences, but they can be distinguished by unique restriction sites. Known amounts of these standard templates, are co-amplified with serial dilutions of the cDNA derived from the mRNA of interest. Oligonucleotide primer pairs possessing 3' octamers found infrequently in the mouse genome (< or = 0.26 x 10(-6)) are used to amplify sequences, chosen to contain no GC stretches longer than 8 (PCRare software) (Griffais et al., 1991). Samples of each PCR product are digested separately with restriction endonucleases unique either for the wild-type or the standard amplicon. The quantitation of the test product and the standard product is easily carried out following their electrophoresis in an ethidium bromide-stained agarose gel.

The antigen-specific cell-mediated immune response in mice is suppressed by infection with pathogenic lyssaviruses.

Responsiveness of T cells (RTC) was studied in BALB/c mice intramuscularly infected with various lyssaviruses. After infection by this peripheral route, two types of viruses could be classified according to their effects: 1) pathogenic viruses, including fixed rabies Pasteur virus (serogenotype 1) and wild viruses belonging to serogenotype 1 (from a rabid fox in France and from a cow infected by a vampire bat in Brazil) or to serogenotype 5 (European bat lyssavirus 1); and 2) non-pathogenic viruses, including Mokola virus (serogenotype 3). RTC was tested by analysing in vitro the capacity of splenic T cells from infected BALB/c mice to produce cytokines after antigenic (purified lyssavirus antigens) or polyclonal stimulation (concanavalin A). Cytokine production was followed by assaying the biological activity of interleukin-2 and by testing for interleukin-2, interleukin-4 and interferon-gamma (IL2, IL4 and IFN gamma ) messenger RNAs (mRNA) by transcription into complementary DNA and amplification by the polymerase chain reaction. The initial biologically active IL2 and cytokine mRNA production was observed in mice infected with pathogenic or non-pathogenic lyssaviruses. Only mice with symptoms (infected with pathogenic viruses) lost the capacity to produce cytokines in vitro after antigen-specific stimulation. No such loss was observed after polyclonal stimulation. In mice peripherally infected with non-pathogenic viruses, no loss was observed after stimulation with lyssavirus antigens. Thus, infection with pathogenic lyssaviruses by the peripheral route induces in BALB/c mice a loss of T-cell responsiveness after antigen activation, but not after polyclonal activation.

Parasite-host relationships: in-situ study of Leishmania spp. in resistant and susceptible mice.

The host's skin is a critical tissue in the natural life cycle of the Leishmania spp. known to cause an 'asymptomatic' infectious process or cutaneous or visceral leishmaniasis in mammals. The dermis, once disturbed by the inoculation of infective parasites, becomes a site of dynamic events, the progression of which depends upon both host and parasite characteristics. Whatever the final site of the morbidity caused by the parasites, whether it be cutaneous, visceral or muco-cutanous, this site reflects the parasite and host's ability to create a pro- or anti-parasite micro-environment. The characteristics of this environment are now amenable to analysis in situ, as illustrated by the study of the cutaneous processes initiated by inoculation of Leishmania major in laboratory mice.

Human glycoprotein HGP92 induces cytokine synthesis in mouse mononuclear phagocytes.

HGP92 has been shown to enhance in vitro and in vivo the bactericidal and tumoricidal activity of mouse macrophages. In this study we investigated the effect of HGP92 on the accumulation of cytokine mRNA in mouse inflammatory, peritoneal macrophages and the monocytic cell line J774. HGP92 significantly enhanced the level of cytokine mRNA for IL-1 alpha, IL-1 beta, IL-6, IL-10, IL-12, TNF-alpha and GM-CSF during the first 24 h of the incubation. This effect triggered by HGP92 was comparable to that obtained with lipopolysaccharide (LPS), which is a strong cytokine inducer. This accumulation of cytokine mRNA in macrophages was correlated with secretion of IL-6 and TNF-alpha in cell supernatant. The release of IL-6 was HGP92 concentration dependent over a range of 0.3-10 micrograms/ml with a maximum production obtained after a 24 h incubation of inflammatory macrophages with HGP92. This effect was shown not to be due to contamination of HGP92 by LPS since inflammatory macrophages from C57BL/6 mice were responsive to HGP92 pretreated with polymyxin B sulfate and unresponsive to heated HGP92. Stimulating activity of HGP92 was confirmed using macrophages from C3H/HeJ mice. These results suggest that HGP92 might modulate the immune responses by increasing cytokine production by macrophages.

Transient inducible events in different tissues: in situ studies in the context of the development and expression of the immune responses to intracellular pathogens.

Intracellular pathogens whether facultative like Mycobacterium sp., e.g. Bacillus Calmette Guérin, Listeria monocytogenes or strictly intracellular like Leishmania sp. initiate either asymptomatic infectious processes or disease depending both on factors of the host (genetic as well as environmental ones) and the infectious/pathogenic agents. In this contribution, we first summarized informations which justify to develop in situ analysis to decipher the sequential events that result in different modes/classes of immune responses. How the mode of the immune response is determined remains a main question to address. Although it has recently become clear, in vitro, that immunocompetent cells and their cytokines are critical to set on a stable mode of immune response, acting on naive T cells, this area deserves more in vivo studies. Indeed, peripheral T cells, at different stages of differentiation, may exist in vivo (a) naive/virgin, (b) experienced, (c) effector T cells, depending on the level of stimulation of the immune system by either endogenous or exogenous (e.g. gut flora) signals. The three chosen examples illustrate our contributions in this field focusing on three different non-lymphoid tissues which may become infected: bone marrow (Bacille de Calmette Guérin), liver (Listeria monocytogenes), skin (Leishmania major). These three illustrations also allow to attract attention on the interest of using mice of genetically different strains the immune response of which is set up under different modes.

Abnormal selection of antibody repertoires in retrovirus-induced murine AIDS.

The mature B cell repertoire in the course of murine AIDS (MAIDS) was investigated. The polymerase chain reaction (PCR) was used to amplify a large diversity of rearranged Ig H chain genes in normal or infected mice, 2 and 8 wk after virus inoculation. Libraries were constructed from the polymerase chain reaction products. By sequencing V-D-J clones in these libraries and analyzing the respective complementary determining region 3 (CDR3), we have shown at 8 wk the emergence of a population of B cells with significantly less N diversity, some sequences lacking any N addition, a typical feature of fetal repertoires known for degeneracy, and autoreactivities. This decreased N diversity was not present 2 wk after inoculation and could not be related to a defect in terminal deoxytransferase expression because the steady-state levels of terminal deoxytransferase mRNA were found normal in MAIDS bone marrow 8 wk after inoculation. FACS analyses revealed a decreased number of bone marrow B cells (B220+, sIgM+) in MAIDS already present at 2 wk, suggesting an alteration in the pathway of B cell differentiation and resulting in a decrease of peripheral B cells renewal. A relative enrichment of spleen cells in long lived B cells as a consequence of this blockade may participate in the abnormal antibody repertoire selection occurring in MAIDS. These data suggest in the MAIDS pathogeny the relationship between an abnormal repertoire selection and the pathologic process.

High frequency of T lymphocytes committed to interferon-gamma transcription upon polyclonal activation in spleen from lymphocytic choriomeningitis virus-infected mice.

Splenic T lymphocytes from C3H/HeOur mice infected for 7 days with lymphocytic choriomeningitis virus (LCMV) do not proliferate in response to concanavalin A (Con A). Although the IL-2 gene remained silent after polyclonal activation, the gene encoding the p55 chain of the IL-2 receptor was normally transcribed. These data indicated that the co-ordinated expression of the unique wave of cytokine and cytokine receptor expression, associated with T cell triggering, did not occur in T lymphocytes from LCMV-infected mice. In a first attempt to characterize the potential of these cells to initiate the transcription of cytokine genes, we have focused our attention on interferon (IFN)-gamma, a cytokine displaying multifocal activities on the immune response. We found that the IFN-gamma encoding gene, silent before Con A activation, was transcribed after triggering in normal and LCMV-infected cells. Notably, the level of induction was approximately 10-fold higher in LCMV mice than in non-infected control mice. IFN-gamma gene was induced in both CD4 and CD8 subsets. Induction was sensitive to cycloheximide addition and thus required de novo protein synthesis. The high level of IFN-gamma mRNA transcripts was correlated with a high frequency of cells transcribing this gene. By in situ hybridization we showed that the majority (approximately 70%) of the splenic T lymphocyte population were positive for IFN-gamma mRNAs. A matching increase in IFN-gamma protein corresponded to this elevated IFN-gamma mRNA level. This observation revealed the existence in LCMV-infected mice of a preponderant peripheral T lymphocyte population which displayed unusual activation and proliferative characteristics.

Altered cytokine genes expression by conA-activated spleen cells from mice infected by lymphocytic choriomeningitis virus.

The intravenous injection of mice with lymphocytic choriomeningitis virus (LCMV) induces a rapid and long-lasting immunodeficiency. T lymphocytes from 7-day-infected mice do not proliferate in vitro in response to ConA stimulation, do not produce IL-2 but display high affinity IL-2 receptors on their membrane. The non-coordinated regulation of these genes suggested that other cytokine-encoding genes may also be affected in their regulation. We have thus analyzed the expression of the genes encoding different cytokines transcribed during spleen cell activation by ConA. The genes encoding T lymphocyte-derived cytokines can be classified in three groups: the genes expressed similarly by normal and LCMV-cells (the p55 and the p75 chains of the IL-2 receptor [1]), the genes under expressed in LCMV-cells (IL-2, IL-3, IL-4 and IL-5) and the genes over expressed by these cells (GM-CSF and IFN-gamma). These results show that the viral infection has provoked a profound alteration of the overall regulation of the genetic program that follows T lymphocyte activation. Since T cell activation depends strictly on accessory cell-derived cytokines, we measured the level of transcription of IL-1, IL-6 and TNF-alpha; and our data show that the expression of these genes is equivalent in normal cells and in cells from LCMV-infected mice.

Mouse T-lymphocyte activation by Urtica dioica agglutinin. I.--Delineation of two lymphocyte subsets.

Urtica dioica agglutinin (UDA) is a mouse T-lymphocyte-specific mitogen endowed with proliferative characteristics different from ConA, the prototypic T-lymphocyte mitogen. In particular, UDA induces 2-3-fold-reduced thymidine incorporation as compared to ConA. In an attempt to define the basis of this reduced proliferation, we analysed whether UDA binds to a unique subset of T lymphocytes or whether it activates only a T-cell subset. Cytofluorimetric analysis showed that this lectin binds uniformly to all T lymphocytes and does not, on this criterion, distinguish a particular T-cell subset. We next analysed whether UDA provokes the activation of all T lymphocytes. This was carried out by measuring the increase in cell size and the induction of the p55 chain of the IL2 receptor. The analysis showed that, throughout the kinetics of cell activation, only one subset of T lymphocytes increased in size and expressed the p55 chain of the IL2 receptor, suggesting that UDA activates only a subpopulation of T cells. This conclusion was strengthened by the analysis of 5-bromo-2-deoxyuridine (BrdU) incorporation into the DNA of UDA-activated cells. Two populations were easily identifiable: a BrdU-negative subset consisting of all the small p55-negative lymphocytes, and a BrdU-labelled subset including all the large p55-positive cells. BrdU was incorporated in both CD4+ and CD8+ cells, indicating that UDA did not distinguish helper from cytotoxic T lymphocytes. In addition to the p55 chain of the IL2R, all cycling cells expressed the Pgp-1 activation marker. The T lymphocytes, which bound UDA but did not proliferate, remained fully susceptible to subsequent stimulation by ConA. In conclusion, the capacity to proliferate upon UDA binding differentiates a UDA-sensitive from a UDA-refractory subset among splenic mouse T lymphocytes.

Mouse T-lymphocyte activation by Urtica dioica agglutinin. II.--Original pattern of cell activation and cytokine production induced by UDA.

Urtica dioica agglutinin (UDA) is a T-lymphocyte-specific polyclonal activator that differs from ConA, the classical mouse T-cell mitogen, by inducing a late and limited proliferation of a distinct T-cell subset recruited among both CD4+ and CD8+ lymphocytes. We investigated the possibility that the particular kinetics may originate from UDA-specific activation processes in which the known early mandatory signals were completed only after an extended delay. We report that the time of contact required between lectin and the cell membrane to acquire the capacity to proceed into cell cycle was much longer (36-40 h) for UDA than for ConA (8-10 h). Addition of phorbol ester, which artificially induces PKC translocation, or ionomycin, which provokes Ca2+ mobilization, did not accelerate the proliferative kinetics, suggesting that these early mandatory signals are not the limiting factors in the delayed proliferation. The induction of c-myc was retarded in the UDA group, and there was a good correlation between the kinetics of c-myc induction and the kinetics of cell proliferation. The comparison of the level of transcription of the genes encoding different cytokines revealed additional differences between the two mitogens: the whole wave of cytokine gene expression was delayed with UDA. In particular, IL2, IL3 and IFN gamma gene expression was retarded compared to the ConA-induced single wave. An even later transcriptional wave took place at around 72 h for IL4 and IL5. Finally, this particular kinetics corresponded to an unusually high level of IL3 and IFN gamma and a low level of IL4 and IL5 gene transcripts.(ABSTRACT TRUNCATED AT 250 WORDS)

Activated T lymphocytes from mice infected by lymphocytic choriomeningitis virus display high affinity IL-2 receptors but do not proliferate in response to IL-2.

The i.v. injection of mice with lymphocytic choriomeningitis virus (LCMV) initiates a rapid and long lasting immunodepression which can be monitored in vivo or in vitro. Splenic T lymphocytes taken from mice infected for 7 days with LCMV are characterized by a low proliferative capacity in response to Con A stimulation in vitro. In an initial attempt to understand the molecular mechanisms regulating the general anergy induced by the viral infection, we have analyzed the transcription of IL-2 and of p55 IL-2R alpha gene, two genes involved in T cell proliferation. IL-2 gene transcripts were hardly detected after Con A activation of spleen cells from LCMV-infected mice. In contrast, the expression of the gene encoding IL-2R alpha chain was induced as in control noninfected cells. In addition, the expression of the p75 IL-2R beta chain was not modified. The transcripts of the IL-2R alpha and of the IL-2R beta genes were normally translated as high affinity. IL-2R were expressed on the membrane of T lymphocytes from LCMV-infected mice. Despite the finding that these receptors could also internalize IL-2, the exogenous addition of this growth factor did not induce cell proliferation, indicating that the virus-induced blockade is multifocal.

Interleukin 2/interleukin 4-independent T helper cell generation during an in vitro antigenic stimulation of mouse spleen cells in the presence of cyclosporin A.

Cyclosporin A (CsA), an immunosuppressive drug which completely suppresses the humoral response to thymus-dependent antigens, does not affect the generation of T helper (Th) cells during an in vitro stimulation of murine spleen cells with sheep red blood cells. The appearance of Th cells depends on time and their development can be prevented by X-ray irradiation performed at the onset of the culture and up to 3-4 days later; however, beyond this time, irradiation is ineffective, suggesting that cell proliferation is essentially completed by this time. The activity of Th cells generated in the presence of CsA is resistant to irradiation, indicating that the effector cells belong to a memory subset. Limiting dilution analysis has shown that the frequency of the Th cells recovered from 6-day sheep red blood cell-stimulated and CsA-treated cultures is of approximately 1.4 X 10(-3), representing an increment of about 500-fold compared to naive spleen cells. The increased frequency of Th cells and the sensitivity to irradiation of the generation of these cells demonstrate that Th cell precursors proliferate in the presence of CsA. Lymphokine genes transcription analysis confirms that the inhibition of interleukin (IL) 2/IL 4 gene expression is one target of CsA action. That the generation of Th memory cells can nevertheless take place strongly argues in favor of the existence of an IL 2/IL 4-independent pathway for murine T cell proliferation. Our finding that the transcription of the IL 7 gene is not inhibited by CsA raises the possibility for a role of this T cell growth factor in the generation of memory Th cells.