Genotypic, phenotypic and functional analysis of CD4+CD7+ and CD4+CD7- T lymphocyte subsets in Sézary syndrome.

The expansion of CD4+CD7- T cells in the peripheral blood of Sézary syndrome (SS) is well known. It remains unclear whether this population contains the dominant T cell clone. Peripheral blood mononuclear cells (PBMC) of five SS patients were sorted by fluorescence-activated cell sorting into CD4+CD7- and CD4+CD7+ populations. These populations were analysed separately for clonality of the T cell receptor gamma chain (TCR-gamma) by PCR-DGGE. The cytokine profile of both populations was investigated by RT-PCR ELISA for IFN-gamma, IL-2, IL-4, IL-5, IL-10, IL-13 and IL-15. In three other patients with known Vbeta-usage, the dominant T cell clones were phenotypically characterized by double staining. PCR-DGGE of TCR-gamma demonstrated that all patients had a clonal population in their blood and that this population was present in CD4+CD7- and CD4+CD7+ populations. Concerning mRNA cytokine transcription, the two populations did not show any consistent differences. In three patients with identified clones (Vbeta 3.1, 5.3 and 6.7), double staining revealed positivity for CD2, CD3, CD4, CD5, CD45RO and CD7 in a significant proportion (at least 35%). We conclude that the CD4+CD7- population does not represent the dominant T cell clone in patients with SS. An increase in this population of PBMC in SS might account for deviations in the T cell functions of the patients.

The role of in vitro-induced lymphocyte apoptosis in feline immunodeficiency virus infection: correlation with different markers of disease progression.

Human immunodeficiency virus infection is characterized by a progressive decline in the number of peripheral blood CD4(+) T lymphocytes, which finally leads to AIDS. This T-cell decline correlates with the degree of in vitro-induced lymphocyte apoptosis. However, such a correlation has not yet been described in feline AIDS, caused by feline immunodeficiency virus (FIV) infection. We therefore investigated the intensity of in vitro-induced apoptosis in peripheral blood lymphocytes from cats experimentally infected with a Swiss isolate of FIV for 1 year and for 6 years and from a number of long-term FIV-infected cats which were coinfected with feline leukemia virus. Purified peripheral blood lymphocytes were either cultured overnight under nonstimulating conditions or stimulated with phytohemagglutinin and interleukin-2 for 60 h. Under stimulating conditions, the isolates from the infected cats showed significantly higher relative counts of apoptotic cells than did those from noninfected controls (1-year-infected cats, P = 0.01; 6-year-infected cats, P = 0.006). The frequency of in vitro-induced apoptosis was inversely correlated with the CD4(+) cell count (P = 0. 002), bright CD8(+) cell count (P = 0.009), and CD4/CD8 ratio (P = 0. 01) and directly correlated with the percentage of bright major histocompatibility complex class II-positive peripheral blood lymphocytes (P = 0.004). However, we found no correlation between in vitro-induced apoptosis and the viral load in serum samples. Coinfection with feline leukemia virus enhanced the degree of in vitro-induced apoptosis compared with that in FIV monoinfected cats. We concluded that the degree of in vitro-induced apoptosis was closely related to FIV-mediated T-cell depletion and lymphocyte activation and could be used as an additional marker for disease progression in FIV infection.

Can low density lipoprotein influence microvascular caliber?

The intracellular free calcium concentration ([Ca2+]i) plays an important role in the regulation of vascular tone. In vascular smooth muscle cells (VSMCs) LDL causes changes in vascular tone by increasing [Ca2+]i. Pericytes are regarded as the microvascular counterpart of VSMCs and implicated in the regulation of microvascular cell biology under normal and pathological conditions (e.g., diabetes mellitus, arterial hypertension, arteriosclerosis). For this reason pericytes and VSMCs were compared in their ability to increase [Ca2+]i after stimulation with LDL. Single VSMCs and pericytes were loaded with 2 microM of the Ca2+-sensitive dye Indo-1/AM. Fluorescence was recorded at 405 nm (Ca2+-bound) and 485 nm (Ca2+-free). Cells in suspension were loaded with 2 microM of the calcium ionophore FURA-2 AM (excitation wavelengths: 340 and 380 nm, emission 505 nm). Basal [Ca2+]i levels were significantly higher in single pericytes (165 +/- 38 nmol/L, n = 41) than in VSMCs (150 +/- 39 nmol/L, n = 40, P = 0.0038). In cell suspensions the following values were obtained: Pericytes (113 +/- 27 nmol/L, n = 36) and VSMCs (109 +/- 26 nmol/L, n = 28), which are statistically not significant. For all concentrations of LDL used (except at 1 microg/ml n-LDL), the increase above basal values was significant and both cell types showed a clear dose-dependent reaction pattern. This study shows for the first time that pericytes and VSMCs increase their [Ca2+]i in a similar way after LDL stimulation. In analogy to aortic smooth muscle cells, our results indicate that LDL mediated [Ca2+]i changes in pericytes in the microvascular bed may cause vasoconstriction leading to impairment of blood flow in the microvasculature.

P-Glycoprotein (P-gp) mediated efflux in Caco-2 cell monolayers: the influence of culturing conditions and drug exposure on P-gp expression levels.

The influence of cell culture conditions and previous drug exposure on P-glycoprotein (P-gp) expression levels in Caco-2 cells was determined. In this study, the expression of P-gp is demonstrated (i) visually by confocal laser scanning microscopy (CLSM), (ii) functionally by transport studies with substrates of the efflux pump, and (iii) quantitatively by flow cytometry (FCM) analysis using specific monoclonal antibodies (anti P-gp MRK 16 as an external antibody and P-GlycoCheck C219 as an internal antibody). Trypsinization of the cells after reaching confluence led to a decrease of P-gp expression levels, while trypsinization before reaching confluence led to an increase after long-term cultivation. Culturing the cells on polycarbonate filters did not elicit a significant change of P-gp expression over time in culture, whereas in plastic flasks (polystyrene) a decrease was detected. Using CLSM a strong fluorescence on the apical side of Caco-2 cell monolayers was observed, as a result of incubation with MRK 16 as primary and IgG Cy5 as secondary antibody. Previous drug exposure of the cells showed that verapamil, celiprolol, and vinblastine induced the P-gp expression, while metkephamid (MKA) decreased the P-gp expression level as compared to the control. Permeation studies consolidated the theory that P-gp is expressed in the Caco-2 cells examined. For talinolol and MKA, a higher transport from basolateral to apical side than from apical to basolateral could be measured. Incubation of the cell monolayer with MRK 16 reduced the secretion process to the apical side, but did not influence [3H]mannitol flux. Caco-2 cells seem to be a suitable cell line model for P-gp-mediated secretion studies. However, the variability of the P-gp expression requires careful control when this model is to be used in quantitative structure/secretion studies.

Analysis of the phenotype and phagocytic activity of monocytes/macrophages from cattle infected with the bovine leukaemia virus.

The bovine leukaemia virus (BLV) is a retrovirus that infects mainly B lymphocytes of cattle, but proviral DNA can also be isolated from monocytes/macrophages. This study investigated the effect of BLV infection on surface antigens on freshly isolated peripheral blood monocytes and cultured monocyte-derived macrophages, with and without lipopolysaccharide (LPS) stimulation. The effect of BLV infection on phagocytic activity of CD14+ monocytes was also assessed. The percentage of monocytes expressing the surface antigens CD11b, CD32 (FcgammaRII), MHC class II and the surface antigen recognised by mAb DH59B were increased in BLV-positive cattle. In contrast, expression intensity of all markers was low in samples from BLV-positive cattle. CD14+ monocytes from BLV-positive cattle showed less Fcgamma-receptor-mediated phagocytosis compared to monocytes from BLV-negative cattle. After 7 days in culture, there was evidence for shedding/downregulation of surface antigens on monocyte-derived macrophages, in particular on cells from BLV-positive cattle. LPS stimulation decreased the percentage of cells expressing the measured markers in monocyte-derived macrophages taken from BLV-negative cattle, but not in cultures derived from BLV-positive cattle. The results provide further evidence for an altered function of monocytes and macrophages in BLV-infected cattle.

Sézary syndrome, T-helper 2 cytokines and accessory factor-1 (AF-1).

Mycosis fungoides and the Sézary Syndrome are characterized by clonal accumulation of well differentiated T-helper memory cells in the skin and, in the case of the Sézary Syndrome, also in the blood and lymph nodes. Well known immunological abnormalities in patients with MF and SS include reduced delayed type hypersensitivity, decreased proliferation upon stimulation with mitogens of the peripheral blood mononuclear cells, elevated IgE or IgA serum levels and eosinophilia. These abnormalities can be explained by the predominance of T-helper 2 cytokines. Clonal T-lymphocytes purified from the peripheral blood of SS patients transcribe mainly IL-10 and IL-5. They can be CD7 positive or negative. These clonal T cells express the accessory factor-1 (AF-1) or Interferon gamma receptor beta-chain that is essential for Interferon gamma signalling. These results imply perspectives for new therapeutical approaches, such as IL-12 or chimeric fusion proteins.

Single pericytes and pericytes in suspension are stimulated in a similar way by low-density lipoprotein.

BACKGROUND: Pericytes are regarded as the microvascular counterpart of smooth muscle cells and implicated in the regulation of blood pressure at the microvascular level. Ca2+ plays an important role in biochemical processes involved in blood pressure regulation and can be activated by low-density lipoprotein (LDL) cholesterol. OBJECTIVE: To determine whether stimulation either of single cells or of cells in suspension by LDL would produce any difference in the increase in cytosolic free calcium levels ([Ca2+]i). DESIGN AND METHODS: Single pericytes were loaded with 2 micromol/l of the Ca2+-sensitive dye Indo-1/AM. The Indo-1 fluorescence was recorded at 405 nm (Ca2+-bound) and 485 nm (Ca2+-free) after stimulation with LDL. Pericytes in suspension were loaded with 2 micromol/l of the Ca2+-sensitive dye FURA-2/AM. The FURA-2 fluorescence kinetics were recorded at 340-380 nm. Ratios of fluorescence at the two wavelengths were transformed to [Ca2+]i. RESULTS: Basal [Ca2+] levels appeared to be higher in single cells (148+/-13 nmol/l, n = 20) than they were in cells in suspension (128+/-8 nmol/l, n = 25; P= 0.0078). After stimulation with LDL the increase in [Ca2+]i in both systems was about 220% above baseline. A clear dose dependency was seen for both systems. CONCLUSIONS: Single pericytes and pericytes in suspension increase their [Ca2+]i after stimulation with LDL dose-dependently. Even though single-cell measurements revealed some technical limitations, their responses were comparable to those obtained in a cell suspension. In analogy to aortic smooth muscle cells, our results indicate that LDL might also play a blood-pressure-regulatory role in the microvasculature.

Interleukin-10 counterregulates proinflammatory cytokine-induced inhibition of neutrophil apoptosis during severe sepsis.

Neutrophils play a key role in the pathophysiology of septic multiple organ dysfunction syndrome (MODS) through excessive release of toxic granule components and reactive oxygen metabolites with consequent tissue destruction. The increase of senescent neutrophils during sepsis indicates a potential breakdown of autoregulatory mechanisms including apoptotic processes to remove activated neutrophils from inflammatory sites. Therefore, neutrophil apoptosis of patients with severe sepsis and its regulatory mechanisms were investigated. Spontaneous neutrophil apoptosis from patients with severe sepsis was significantly reduced in comparison to healthy individuals. Cytokines detected in the circulation during sepsis (tumor necrosis factor-alpha [TNF-alpha], interferon-gamma [IFN-gamma], granulocyte colony-stimulating factor [G-CSF], granulocyte-macrophage colony-stimulating factor [GM-CSF]) inhibited neutrophil apoptosis in both groups, though the effect was more distinct in neutrophils from healthy humans. Addition of lipopolysaccharide (LPS) to neutrophils from healthy humans markedly (P < .05) reduced apoptosis which was partially restored through addition of anti-TNF-antibody. Interleukin-10 (IL-10) counteracted (P < .05) inhibition of neutrophil apoptosis induced by LPS, recombinant human (rh) TNF-alpha, rhIFN-gamma, rhG-CSF, and rhGM-CSF, whereas rhIL-4 or rhIL-13 were ineffective. Reduced neutrophil apoptosis during sepsis was concomitant with increased tyrosine phosphorylation, while IL-10 markedly inhibited tyrosine phosphorylation in LPS-stimulated neutrophils. These results identify proinflammatory cytokines and IL-10 as strong regulators of spontaneous neutrophil apoptosis during sepsis. Inhibition as well as acceleration of neutrophil apoptosis seems to be associated with alterations of signal transduction pathways.

Flow cytometric detection of the Golgi apparatus using antibodies to glycosyltransferases.

Intracellular glycosyltransferase protein expression can be assessed by flow cytometry. We report the detectability of the Golgi associated beta1,4 galactosyltransferase (GT) and alpha2,6 sialyltransferase (ST) upon permeabilization in Jurkat and EBV-JY cells representing a T- and B-lymphoid cell line, respectively. The method employs fixation with paraformaldehyde and permeabilization with saponin. It ensures reliable internalization of the antibody and little background staining and does not cause leakage of the antigen. We first applied monoclonal antibodies to GT for establishment of this method by flow cytometry. The obtained flow cytometric signal could be localized to the Golgi apparatus by confocal laser scanning microscopy. To exclude interference from possible cell-surface staining, measurements were carried out on non-permeabilized cells. No signal was found in Jurkat cells, while low but measurable ecto-galactosyltransferase was found on EBV-JY cells. F(ab)'2 fragments of polyclonal antisera to GT and ST were generated and shown to be useful for double indirect staining with the monoclonal antibody. The method described here permits relative assessment of Golgi glycosyltransferase expression in lymphoid cells by flow cytometry.

T-cell derived cytokines co-stimulate proliferation of CD40-activated germinal centre as well as follicular lymphoma cells.

Follicular lymphomas, the malignant counterparts of normal germinal centre (GC) B-cells, grow in vivo in close association with polyclonal T-cells, predominantly from the T-helper cell type. T-cell-derived growth factors are involved in the development of GC B-cells. However, their role in the pathogenesis of follicular lymphomas has not been clearly defined. We investigated the co-stimulatory activity of 14 cytokines (interleukin-1 to -8, IL-10, IL-13, INF-alpha, TNF-alpha, GM-CSF and SCF) on the proliferation of CD40-activated follicular lymphoma cells in comparison to tonsillar GC B-cells. Tonsillar GC B-cells (n = 4), malignant cells from diagnostic lymph node biopsies of patients with follicular (n = 4) or transformed (n = 4) lymphomas were grown on irradiated CD40-ligand transfectants, with and without cytokines. [3H]-thymidine uptake was measured at day 7. IL-10 and IL-4 proved to be the most potent co-stimulators of proliferation of tonsillar GC B-cells, whereas proliferation of follicular lymphoma cells was co-stimulated by IL-4. The fact that IL-4 is a T-cell derived cytokine, suggests that lymphoma infiltrating T-cells play a role in the growth of these malignancies. Moreover, proliferation of both non-neoplastic tonsillar GC B-cells and follicular lymphomas is co-stimulated by T-cell derived cytokines, indicating that responsiveness to paracrine factors may not be a characteristic of the malignant phenotype.

The proinflammatory cytokine interleukin-12 occurs as a cell membrane-bound form on macrophages.

Phagocyte-derived interleukin-12 (IL-12) is a proinflammatory cytokine promoting cell-mediated immune responses in inflammatory and infectious disorders. Based on sequence homology of the p40 subunit with the interleukin-6 receptor it has been speculated that IL-12 could also exist as a membrane-bound form, but thus far only soluble (secreted) IL-12 has been identified. We have therefore analyzed human monocytic U937 and mouse P388D1 macrophages for membrane-bound IL-12 by flow cytometry. IL-12 is constitutively expressed on the cell surface of both cell lines. IL-12 cell surface staining is enhanced in response to stimulation with IFN-gamma plus LPS. IL-12 is also present in the supernatant of cultured P388D1 macrophages. Thus, in addition to a soluble form IL-12 occurs as a membrane-bound molecule on monocyte/macrophage cell lines.

Flow cytometric detection of activation-induced cell death (apoptosis) in peripheral blood lymphocyte subpopulations from healthy cats.

Human peripheral blood lymphocytes (PBLs) from healthy individuals are resistant to in vitro-induced apoptosis, but activated human lymphocytes can readily undergo apoptosis. The activation of human lymphocytes is accompanied by the upregulation of a cell surface antigen, the major histocompatibility complex (MHC) class II-antigen. Only a minority of PBLs are usually MHC class II-antigen-positive in healthy humans. In contrast, in healthy cats the majority of feline PBLs are MHC class II-antigen-positive. We have, therefore, investigated the sensitivity of feline peripheral blood lymphocytes obtained from specified pathogen free (SPF) cats to the induction of apoptosis. Feline PBLs from SPF cats (n = 16) and human PBLs from healthy donors (n = 2) were isolated. After short-term culture, cells were examined for the presence of fragmented DNA as a result of apoptosis by a DNA agarose gel electrophoresis method and for the presence of DNA double strand breaks by in situ 3' end labeling. In addition, relative DNA content per cell was flow cytometrically determined using propidium iodide (PI) or 7-actinomycin-D (7-AAD) and apoptotic cells were identified on the basis of a reduced DNA content. Cell surface antigens and cellular DNA were analyzed simultaneously by dual-color flow cytometric analyses in order to study lymphocyte subsets. Single- and dual-color analysis revealed that, in contrast to human lymphocytes, feline lymphocytes rapidly underwent apoptosis when cultured overnight in medium. Furthermore, the majority of apoptotic cells was found within the MHC II-positive cell subject.

A sex-specific number of germ cells in embryonic gonads of Drosophila.

Male first instar larvae possess more germ cells in their gonads than female larvae of the same stage. To determine the earliest time point of sexual dimorphism in germ cell number, we have counted the germ cells of sexed embryos at different developmental stages. We found no difference in germ cell number of male and female embryos at the blastoderm and early gastrulation stage, or when germ cells are about to exit the midgut pocket. We find, however, that males have significantly more germ cells than females as soon as the germ cells are near the places where the gonads are formed and in all later stages. Our results show that germ cells are subject to a sex-specific control mechanism that regulates the number of germ cells already in embryos.

Separation of a cysteine-rich surface antigen-expressing variant from a cloned Giardia isolate by fluorescence-activated cell sorting.

Clonally derived trophozoites (clone O2-4A1, from a sheep) of the morphologically defined group of Giardia duodenalis change their cysteine-rich surface proteins (CRISPs) spontaneously during in vitro culture. This phenomenon constitutes a serious obstacle for studies that rely on a pure population of cells bearing a particular variant CRISP. We describe herein the successful separation and quantitation of a trophozoite subpopulation expressing a 90-kDa major surface antigen (CRISP-90) from a heterologous population using fluorescence-activated cell sorting (FACS) on the basis of fluorescein-tagged antibodies directed specifically against O2-4A1 CRISP-90. During subsequent in vitro culture, the purified cell population exhibited a progressive decline in the proportion of cells labeled by CRISP-90 specific antibodies. After 46 generations, only one-third of the total trophozoite population reacted with the anti-CRISP-90 antibodies.

Immunization-induced decrease of the CD4+:CD8+ ratio in cats experimentally infected with feline immunodeficiency virus.

In a previous experiment a group of 15 specified pathogen free (SPF) cats were experimentally infected with a Swiss isolate of feline immunodeficiency virus (FIV). A group of 15 SPF cats served as FIV negative controls. Nine cats of each group were vaccinated with a recombinant feline leukemia virus (FeLV) vaccine, six cats in each group with a placebo vaccine. All vaccinated cats developed high antibody titers to FeLV and were protected against subsequent FeLV challenge infection. In both control groups five of six cats became persistently infected with FeLV. Unexpectedly, the primary immune response to the vaccine antigen was significantly higher in the FIV positive group than in the FIV negative. The secondary response was stronger in the FIV negative cats. The goal of the present investigation was to further study the immune response in these 30 cats. They were immunized twice with the synthetic peptide L-tyrosine-L-glutamic acid-poly(DL-alanine)-poly(L-lysine) (TGAL) 21 days apart. Blood samples were collected on four occasions during the immunization process. They were tested for antibodies to TGAL, complete blood cell counts and CD4+, CD8+ and pan-T-lymphocyte counts. The following observations were made: (1) in contrast to the FeLV vaccine experiment, the primary immune response to TGAL was not significantly stronger in the FIV positive cats when tested by enzyme-linked immunosorbent assay (2). The absolute size of the CD4+ lymphocyte population was distinctly smaller in the FIV positive than in the FIV negative cats. The lowest CD4+ values were found in the dually FIV/FeLV infected cats. (3) A population of CD8+ lymphocytes was identified that was characterized by a distinctly weaker fluorescence. The size of this population increased in FIV positive and decreased in FIV negative cats during the TGAL immunization experiment. (4) The CD4+:CD8+ ratio increased in FIV negative cats during TGAL immunization from 1.9 to 2.3. In contrast, in FIV positive animals the CD4+:CD8+ ratio decreased significantly from 1.9 to 1.3 during the same period. From these and earlier data it was concluded that in short-term FIV infection the immune response to T-cell dependent antigens may be increased over that of the controls. Immune suppression develops gradually with duration of the infection. The significant drop of the CD4+:CD8+ ratio over a 5 week immunization period suggests that antigenic stimulation may accelerate the development of immune suppression in FIV positive cats. If this is a general feature, FIV infection may provide a particularly interesting model for studying the pathogenesis of AIDS.

The primary structures of the core antenna polypeptides from Rhodopseudomonas marina.

The antenna complex B880 of Rp. marina has been isolated by applying ion-exchange chromatography on Whatman DE-52 resin and sucrose density centrifugation of LDAO-solubilized photosynthetic membranes. The antenna polypeptides B880-alpha and B880-beta were prepared by organic solvent extraction of extensively dialyzed and freeze-dried B880 antenna complex material or photosynthetic membranes. Gel filtration on Sephadex LH-60 and ion-exchange chromatography on Whatman DE-32 resin in the presence of organic solvents and an additional step on a C-8 reversed phase column yielded pure alpha- and beta-apoproteins. Their complete primary structures have been elucidated using automated Edman degradation and carboxypeptidase digestion. According to quantitative Edman degradation the ratio of B880-alpha and B880-beta has been determined as 1:1 in the isolated antenna complex as well as in the photosynthetic membrane. B880-alpha of Rp. marina, presumably N-formylated, consists of 52 amino acid residues and is 75, 56, 52 and 44% homologous to the corresponding core antenna polypeptides of Rs. rubrum, Rp. viridis, Rb. capsulatus and Rb. sphaeroides. In contrast, B880-beta (56 amino acid residues) is less homologous to the corresponding core beta-antenna polypeptides of the same strains (57, 51, 41 and 42%). It shows an extended N-terminal domain as compared to the B880-alpha polypeptide. Apart from the typical structural features of bacterial membrane-bound antenna polypeptides (three domain structure. His-residue in the hydrophobic stretch) the antenna polypeptides of Rp. marina are structurally related to polypeptides of core antenna complexes with strong near infrared circular dichroism signals.

The primary structure of Bacillus cereus neutral proteinase and comparison with thermolysin and Bacillus subtilis neutral proteinase.

The complete amino-acid sequence of a neutral proteinase, produced by Bacillus cereus, was determined by protein sequencing. The neutral proteinase consists of 317 amino-acid residues. The primary structure is 70% homologous to thermolysin, a thermostable neutral proteinase and 45% homologous to Bacillus subtilis neutral proteinase. The zinc-binding site and the hydrophobic pocket of the active site are highly similar in all three proteinases. B. cereus neutral proteinase which is 20 degrees C less thermostable (60 degrees C) than thermolysin (80 degrees C) shows only minor differences in calcium binding sites and salt bridges compared to thermolysin (known from its X-ray diffraction analysis), whereas B. subtilis neutral proteinase (50 degrees C) differs considerably. Therefore it was assumed that the difference in thermostability between B. cereus neutral proteinase and thermolysin is not caused by different metal binding properties, or differences in the active site, but by changes within the rest of the molecule. Calculation of secondary structure potentials according to Chou & Fasman, hydrophobicity and bulkiness of the different structural elements and preferred cold----hot amino-acid residue exchanges indicated, that the thermostability of thermolysin compared to B. cereus neutral proteinase is caused by small effects contributed by numerous amino-acid exchanges distributed over the whole molecule, resulting in increased hydrophobicity of beta-pleated sheet and higher bulkiness of alpha-helical regions.

[Bacterial meningitis in newborn infants. A retrospective study from a pediatric clinic 1967-1978]. / Méningites purulentes du nouveau-né. Etude rétrospective dans un service de pédiatrie, 1967-1978.

A total of 22 newborn infants (14 boys, 8 girls) have been admitted and treated for bacterial meningitis in the University Pediatric Service of Geneva over a period of 11 years (May 1967 to May 1978). The three most common infectious agents were: Group B beta-hemolytic Streptococcus (6/22 cases), Escherichia coli (6/22 cases), and Listeria monocytogenes (4/22 cases). Thirteen of the 22 infants died (a 59% mortality, in keeping with that observed in other centers). Follow-up of the nine survivors showed a relatively favorable course from a developmental and neurological point of view. Only two of the infants have significant sequelae. Factors predisposing toward the occurrence of neonatal meningitis are a small birth weight, premature rupture of the membranes, and perinatal maternal infection. Prevention of neonatal meningitis is therefore very much dependent upon good perinatal care of the mother. Treatment of neonatal meningitis is impaired by the poor diffusion of antibiotics through the blood-brain barrier. Intrathecal antibiotics were used in 4 cases in this series, and three of these 4 patients died: intrathecal antibiotherapy is obviously not a good solution. Molecules with a better diffusion such as chloramphenicol should be considered with renewed interest.