Identification and characterization of the scl gene encoding a group A Streptococcus extracellular protein virulence factor with similarity to human collagen.

Group A Streptococcus (GAS) expresses cell surface proteins that mediate important biological functions such as resistance to phagocytosis, adherence to plasma and extracellular matrix proteins, and degradation of host proteins. An open reading frame encoding a protein of 348 amino acid residues was identified by analysis of the genome sequence available for a serotype M1 strain. The protein has an LPATGE sequence located near the carboxy terminus that matches the consensus sequence (LPXTGX) present in many gram-positive cell wall-anchored molecules. Importantly, the central region of this protein contains 50 contiguous Gly-X-X triplet amino acid motifs characteristic of the structure of human collagen. The structural gene (designated scl for streptococcal collagen-like) was present in all 50 GAS isolates tested, which together express 21 different M protein types and represent the breadth of genomic diversity in the species. DNA sequence analysis of the gene in these 50 isolates found that the number of contiguous Gly-X-X motifs ranged from 14 in serotype M6 isolates to 62 in a serotype M41 organism. M1 and M18 organisms had the identical allele, which indicates very recent horizontal gene transfer. The gene was transcribed abundantly in the logarithmic but not stationary phase of growth, a result consistent with the occurrence of a DNA sequence with substantial homology with a consensus Mga binding site immediately upstream of the scl open reading frame. Two isogenic mutant M1 strains created by nonpolar mutagenesis of the scl structural gene were not attenuated for mouse virulence as assessed by intraperitoneal inoculation. In contrast, the isogenic mutant derivative made from the M1 strain representative of the subclone most frequently causing human infections was significantly less virulent when inoculated subcutaneously into mice. In addition, both isogenic mutant strains had significantly reduced adherence to human A549 epithelial cells grown in culture. These studies identify a new extracellular GAS virulence factor that is widely distributed in the species and participates in adherence to host cells and soft tissue pathology.

Late relapsing childhood lymphoblastic leukemia.

Childhood lymphoblastic leukemia (ALL) is usually assumed to have been permanently eradicated in patients in long-term remission, but occasionally can recur after many years. To learn more about the problem, we studied a group of children whose leukemia had been in remission for 10 or more years before relapse and tried to determine whether they had true recurrences or second malignancies. We studied children treated on Medical Research Council ALL protocols between 1970 and 1984 and followed up by the Clinical Trial Service Unit in Oxford. Detailed clinical and laboratory data was collected from the centers concerned on all who were reported to have had a recurrence of their leukemia after 10 or more years from the time of achieving first complete remission (CR1). To prove that the relapse was a true recurrence rather than a second or secondary leukemia, DNA extracted from archived marrow smears was subjected to polymerase chain reaction (PCR) analysis for the presence of an identical Ig heavy chain (IgH) or T-cell receptor (TCR) gene rearrangement at initial diagnosis and subsequent relapse. A total of 1,134 of 2,746 children had survived 10 years or more (range, 10 to 24 years) in CR1 and of those, 12 (approximately 1%) had subsequently relapsed. Relapse blast cells were shown to express the common ALL antigen (CD 10) in all cases and an identical clonal IgH or TCR gene rearrangement was found on PCR analysis of DNA from diagnosis and relapse in all eight cases where DNA extraction was successful. A further program of therapy was successful in inducing a second CR in all patients, four of whom have succumbed to a second relapse after 12 to 27 months. The remaining eight are in continuing CR2 at a follow-up of 12 to 108 months (median, 52) from relapse. Although the risk of relapse of childhood ALL after 10 years in remission appears to be small (around 1%), it persists. This raises questions about how blasts can survive quiescent for so long and when we can truly be confident of cure, if ever.

Gonococcal opacity protein promotes bacterial entry-associated rearrangements of the epithelial cell actin cytoskeleton.

Neisseria gonorrhoeae enters cultured human mucosal cells following binding of a distinct gonococcal opacity (Opa) outer membrane protein to cell surface proteoglycan receptors. We examined the route of internalization that is activated by Opa-expressing gonococci (strain VP1). Microscopy of infected Chang epithelial cells showed that gonococcal uptake was insensitive to monodansylcadaverine (150 microM), which interferes with clathrin-mediated endocytosis. Similarly, indirect immunofluorescence staining for clathrin in infected cells showed distribution of cellular clathrin unaltered from the distribution in noninfected cells. The microtubule inhibitors colchicine (50 microM) and nocodazole (20 microM) but not the microtubule-stabilizing agent taxol (10 microM) caused a moderate (30 to 50%) reduction in gonococcal entry without affecting bacterial adherence. The most dramatic effects were obtained with the microfilament-disrupting agent cytochalasin D (3 microM), which totally blocked bacterial entry into the cells. Double immunofluorescence staining of gonococci and actin filaments in infected cells demonstrated bacterium-associated accumulations of F-actin as an early signal of bacterial entry. The recruitment of F-actin was transient and disappeared once the bacteria were inside the cells. Cytochalasin D disrupted the actin cytoskeleton architecture but did not prevent the recruitment of F-actin by the bacteria. Adherent, noninvasive gonococcal Opa variants lacked the ability to mobilize F-actin. Recombinant Escherichia coli expressing the gonococcal invasion-promoting Opa of gonococcal strain MS11 (Opa50) adhered to the epithelial cells in an Opa-dependent fashion but was not internalized and did not recruit detectable amounts of F-actin. Coinfection with the E. coli recombinant strain and gonococci resulted in specific entry of the diplococci, despite the presence of large numbers of adherent E. coli cells. Together, our results indicate that Opa-mediated gonococcal entry into Chang cells resembles phagocytosis rather than macropinocytosis reported for Salmonella spp. and sequentially involves gonococcal adherence to the cell surface, Opa-dependent and cytochalasin-insensitive recruitment of F-actin, and cytochalasin D-sensitive bacterial internalization.

Fetal plasma erythropoietin concentration in severe growth retardation.

OBJECTIVE: The aim of this study was to determine whether hypoxemia induces an increase in plasma erythropoietin concentration in human fetal life and, if so, whether this response stimulates fetal erythropoiesis. STUDY DESIGN: The plasma erythropoietin concentration in blood samples from 33 small-for-gestational-age fetuses at 26 to 38 weeks' gestation was measured. Measurements were compared with the reference range for gestation, and associations with PO2, pH, and erythroblast and erythrocyte counts were examined. RESULTS: The mean plasma erythropoietin concentration in the small-for-gestational-age fetuses was significantly increased, and the degree of increase was significantly associated both with fetal acidemia and, more strongly, with fetal erythroblastosis. CONCLUSION: Erythropoietin production in response to tissue hypoxia occurs from at least 26 weeks' gestation with measurable physiologic effects on erythropoiesis. Furthermore, more accurate assessment of tissue oxygenation may be obtained by measuring the erythroblast count rather than the blood pH.

Fetal plasma erythropoietin in pregnancies complicated by maternal diabetes mellitus.

OBJECTIVE: Our purpose was to investigate the relationship between fetal plasma erythropoietin concentration and measures of short-term and long-term glycemic control and fetal oxygenation in pregnancies complicated by maternal diabetes mellitus. STUDY DESIGN: A cross-sectional study was performed at The Harris Birthright Research Centre for Fetal Medicine, London. Cordocentesis was performed in 31 diabetic pregnancies for the measurement of umbilical venous blood pH, PO2, PCO2, lactate and glucose concentration, erythroblast count, hemoglobin, and plasma erythropoietin concentrations. RESULTS: The mean pH was significantly lower and the PCO2, lactate, erythropoietin, hemoglobin, and erythroblast counts were significantly higher than the appropriate normal mean for gestation. There were significant associations between (1) fetal erythropoietin and erythroblast count, (2) fetal erythroblast count and hemoglobin, (3) fetal hemoglobin and maternal glycosylated hemoglobin, and (4) maternal glucose and fetal glucose, pH, and lactate. CONCLUSIONS: We postulate that maternal hyperglycemia causes fetal hyperglycemia and acidemia. Increased erythropoietin may be caused by tissue hypoxia or hyperinsulinemia. The increase in fetal hemoglobin may be the consequence of increased erythropoiesis, mediated by either erythropoietin or hyperinsulinemia.

Fetal plasma erythropoietin concentration in red blood cell-isoimmunized pregnancies.

OBJECTIVE: The aim of this study was to investigate the relationship between fetal anemia, plasma erythropoietin concentration, and erythroblastosis in red blood cell-isoimmunized pregnancies. STUDY DESIGN: Fetal plasma erythropoietin concentration in umbilical venous blood samples from 68 red blood cell-isoimmunized pregnancies at 18 to 35 weeks' gestation was measured. Measurements were compared with the appropriate reference range with gestation, and associations with blood pH, erythroblast count, and hemoglobin concentration were examined. RESULTS: The mean fetal plasma erythropoietin concentration and erythroblast count in red blood cell-isoimmunized pregnancies were significantly increased only in severe fetal anemia (hemoglobin deficit > 7 gm/dl). Furthermore, some severely anemic fetuses were hydropic and acidemic. The degree of increase in plasma erythropoietin was significantly associated with both fetal acidemia and, more strongly, fetal erythroblastosis. CONCLUSION: These findings suggest that in fetuses from red blood cell-isoimmunized pregnancies the ability to prevent tissue hypoxia is present until anemia becomes severe, presumably by an increase in cardiac output and tissue perfusion. In severe anemia tissue hypoxia occurs, and the data indicate that fetuses respond by increasing erythropoietin production from at least 20 weeks' gestation. Furthermore, more accurate assessment of tissue oxygenation may be obtained by measuring the erythroblast count rather than the blood pH.

Erythropoietin concentrations in obstructive sleep apnoea.

Eight patients with obstructive sleep apnoea and a normal haemoglobin concentration underwent nocturnal studies during which oxyhaemoglobin saturation was recorded continuously with an ear oximeter and serum erythropoietin concentration was measured hourly by means of a radioimmunoassay. Serum erythropoietin concentrations remained within the normal range throughout the study despite falls in oxyhaemaglobin saturation in individuals to 33-78%. There was no relation between the degree of nocturnal hypoxaemia and serum erythropoietin concentrations. The brief cyclical episodes of hypoxaemia typical of obstructive sleep apnoea may not be a sufficient stimulus for erythropoietin secretion.

Serum erythropoietin changes in autologous and allogeneic bone marrow transplant patients.

Sequential changes in serum erythropoietin (sEPO) levels were measured by radioimmunoassay in six patients receiving autologous rescue (AR) and 11 patients receiving an allogeneic bone marrow transplant (BMT) for malignant disease. Longitudinal studies showed an inverse relationship between sEPO and haemoglobin levels in the autologous rescue and allogeneic transplant patients throughout the 130 d post-transplant study period. Early post-conditioning EPO responses were normal for the haemoglobin level in both groups, but after day 14 post-transplant, erythropoietin production in response to anaemia became impaired in one autologous rescue patient and eight of the 11 allogeneic transplant patients. There was no clear association between late impairment of sEPO production and conditioning therapy, infection, graft-versus-host disease, immunosuppressive therapy or serum creatinine. Blood transfusion requirements were similar for both groups in the first month after transplantation, but from days 31 to 90 post-transplant, BMT patients required an average of 5.5 units per patient compared with 1 unit per patient for the autologous group. Marrow transplant procedures do not affect early EPO responses but may diminish late responses. The potential value of exogenous rHuEPO in hastening engraftment and decreasing transfusion requirements, particularly for those patients who appear to have impaired EPO responses, remains to be shown by clinical trials.