Ebola virus defective interfering particles and persistent infection.

Ebola virus (Zaire subtype) is associated with high mortality disease outbreaks that commonly involve human to human transmission. Surviving patients can show evidence of prolonged virus persistence. The potential for Ebola virus to generate defective interfering (DI) particles and establish persistent infections in tissue culture was investigated. It was found that serial undiluted virus passages quickly resulted in production of an evolving population of virus minireplicons possessing both deletion and copyback type DI genome rearrangements. The tenth undiluted virus passage resulted in the establishment of virus persistently infected cell lines. Following one or two crises, these cells were stably maintained for several months with continuous shedding of infectious virus. An analysis of the estimated genome lengths of a selected set of the Ebola virus minireplicons and standard filoviruses revealed no obvious genome length rule, such as "the rule of six" found for the phylogenetically related Paramyxovirinae subfamily viruses. Minimal promoters for Ebola virus replication were found to be contained within 156 and 177 nucleotide regions of the genomic and antigenomic RNA 3' termini, respectively, based on the length of authentic termini retained in the naturally occurring minireplicons analyzed. In addition, using UV-irradiated preparations of virus released from persistently infected cells, it was demonstrated that Ebola virus DI particles could potentially be used as natural minireplicons to assay standard virus support functions.

Genetic diversity and distribution of Peromyscus-borne hantaviruses in North America.

The 1993 outbreak of hantavirus pulmonary syndrome (HPS) in the southwestern United States was associated with Sin Nombre virus, a rodent-borne hantavirus; The virus' primary reservoir is the deer mouse (Peromyscus maniculatus). Hantavirus-infected rodents were identified in various regions of North America. An extensive nucleotide sequence database of an 139 bp fragment amplified from virus M genomic segments was generated. Phylogenetic analysis confirmed that SNV-like hantaviruses are widely distributed in Peromyscus species rodents throughout North America. Classic SNV is the major cause of HPS in North America, but other Peromyscine-borne hantaviruses, e.g., New York and Monongahela viruses, are also associated with HPS cases. Although genetically diverse, SNV-like viruses have slowly coevolved with their rodent hosts. We show that the genetic relationships of hantaviruses in the Americas are complex, most likely as a result of the rapid radiation and speciation of New World sigmodontine rodents and occasional virus-host switching events.

Isolation, genetic diversity, and geographic distribution of Bayou virus (Bunyaviridae: hantavirus).

Bayou hantavirus, previously implicated in human hantavirus pulmonary syndrome in Louisiana, was isolated from a rice rat (Oryzomys palustris) captured in Georgia. The presence of antibody among rice rats captured throughout the southeastern United States and the extent of diversity among the genetic variants of Bayou viruses suggest that the rice rat is the most likely natural reservoir of the virus and that both virus and host have probably co-evolved for some years.

Isolation, characterization and geographic distribution of Caño Delgadito virus, a newly discovered South American hantavirus (family Bunyaviridae).

Rodents collected from the Venezuelan llanos (plains) during field studies of viral hemorrhagic fever were tested for evidence of hantavirus infection. Hantavirus antibody was found in one (7.7%) of 13 Oryzomys bicolor, one (3.4%) of 29 Rattus rattus, 10 (6.0%) of 166 Sigmodon alstoni and one (2.2%) of 45 Zygodontomys brevicauda. Hantavirus-specific RNA was detected in lung tissues from four antibody-positive rodents: two S. alstoni from Portuguesa State and one S. alstoni each from Cojedes and Barinas States. A hantavirus isolate (herein identified as VHV-574) was recovered from lung tissue from a hantavirus RNA-positive S. alstoni collected from Portuguesa State. The results of serological tests and analyses of small and medium RNA segment nucleotide sequence data indicated that VHV-574 represents a novel hantavirus (proposed name 'Caño Delgadito') that is distinct from all previously characterized hantaviruses. The results of analyses of nucleotide sequence data from the four hantavirus RNA-positive S. alstoni suggested that Caño Delgadito virus is widely distributed in the Venezuelan llanos.

Naturally occurring Sin Nombre virus genetic reassortants.

Genetic reassortment has been shown to play an important role in the evolution of several segmented RNA viruses and in the epidemiology of associated diseases. Sin Nombre (SN) virus is the cause of hantavirus pulmonary syndrome throughout the western United States. Like other hantaviruses, it possesses a genome consisting of three negative-sense RNA segments, S, M, and L. Recent analysis has demonstrated the presence of at least three different hantaviruses in Nevada and eastern California, including SN, Prospect Hill-like, and El Moro Canyon-like viruses. In addition, two distinct lineages of SN virus can be found in Peromyscus maniculatus rodents (sometimes in close proximity) trapped at study sites in this region. Data obtained by phylogenetic analysis of sequence differences detected among the S, M, and L genome segments of these SN viruses are consistent with reassortment having taken place between SN virus genetic variants. The results suggest that M (and to a lesser extent S or L) genome segment flow occurs within SN virus populations in P. maniculatus in this region. No reassortment was detected between SN virus and other hantavirus types present in the area. This finding suggests that as genetic distance increases, the frequency of formation of viable reassortants decreases, or that hantaviruses which are primarily maintained in different rodent hosts rarely have the opportunity to genetically interact.

Complete genetic characterization and analysis of isolation of Sin Nombre virus.

This study reports completion of the genetic characterization of the entire genome of Sin Nombre (SN) virus (NMH10) detected in autopsy tissues from a patient who died of hantavirus pulmonary syndrome (HPS). The large (L) genome segment was found to be 6,562 nucleotides in length and encoded a putative L polymerase that was 2,153 amino acids in length. No evidence of segment reassortment with other well-characterized hantaviruses was obtained. The sequence of the entire S, M, and L genome segments of SN virus (strain NMR11) isolated from a mouse (trapped in the residence of the patient infected with SN virus [NMH10]) by passage two times in Peromyscus maniculatus and then by five passages in E6 Vero cells was determined and compared with that of the virus detected in autopsy tissues. Only 16 nucleotide differences were detected between the virus genomes, and none of these resulted in virus protein amino acid substitutions. Determination of the exact 5'- and 3'-terminal sequences of all genome segments of SN virus and representatives of other serologic groups in the Hantavirus genus, family Bunyaviridae, showed the existence of conserved nucleotide domains that may be involved in important regulatory mechanisms, such as RNA encapsidation, polymerase binding, and control of transcription and replication.

Coexistence of several novel hantaviruses in rodents indigenous to North America.

Three genetically distinct members of the Hantavirus genus have been detected in Nevada rodents by RT-PCR and nucleotide sequence analysis. These include Sin Nombre (SN), El Moro Canyon (ELMC), and Prospect Hill (PH)-like viruses which are primarily associated with Peromyscus maniculatus (deer mouse), Reithrodontomys megalotis (western harvest mouse), and Microtus spp. (voles), respectively. Although this region of the United States is ecologically diverse, rodents infected with different hantaviruses appear to coexist in several different geographical and ecological zones. In two widely separated states, Nevada and North Dakota, PH-like viruses are present in three different species of vole. In addition, ELMC-like virus has been detected in both R. megalotis and M. montanus (mountain vole). SN virus is a cause of hantavirus pulmonary syndrome throughout much of the United States. SN virus RNA is found in 12.5% of P. maniculatus in Nevada and eastern California. Two lineages of SN virus coexist in this region and differ from SN viruses originally found in infected rodents in New Mexico, Arizona, and Colorado. These data show the complexity of hantavirus maintenance in rodents. Distinct hantaviruses or virus lineages can coexist either in different or the same rodent species and in either different or the same geographic or ecological zones.

Differential effects of sequential, simultaneous, and single agent interleukin-3 and granulocyte-macrophage colony-stimulating factor on megakaryocyte maturation and platelet response in primates.

Recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) following interleukin-3 (IL-3) priming has been shown to increase thrombopoiesis. To elucidate the comparative abilities of IL-3 and GM-CSF in influencing megakaryocyte development in vivo, serial bone marrow analyses were performed on rhesus monkeys treated with 5 micrograms/kg/d of IL-3 and 5 micrograms/kg/d of GM-CSF sequentially for 4 days each, simultaneously for 8 days, and as single agents for 8 days. Platelet counts maximally increased to a mean of 7.5 x 10(5)/microL (n = 3) on days 11 through 12 in monkeys treated with sequential IL-3/GM-CSF. In contrast, neither IL-3 alone nor simultaneously administered IL-3/GM-CSF elicited increases in thrombopoiesis between days 3 and 15. GM-CSF elicited a variable platelet response. Megakaryocyte ploidy distributions were significantly (P < .001) shifted between days 7 and 10 in monkeys treated sequentially and between days 3 and 15 in monkeys treated with combined IL-3/GM-CSF and with GM-CSF alone but not in monkeys treated with IL-3 alone. The changes in mean DNA content and megakaryocyte size, as determined by digital image analysis, were larger in monkeys treated with sequential IL-3/GM-CSF and with GM-CSF alone than in simultaneously treated monkeys. In addition, sequentially but not simultaneously treated monkeys showed increased numbers of megakaryocytes on bone marrow biopsy. We conclude that administration of IL-3 followed by GM-CSF treatment increases thrombopoiesis by sequentially increasing megakaryocyte numbers and maturation and that these effects are diminished by simultaneous administration of the two cytokines.

Recombinant human granulocyte-macrophage colony-stimulating factor promotes megakaryocyte maturation in nonhuman primates.

Megakaryocytes are responsive to several nonlineage-specific cytokines in vitro. In this study, we examined the in vivo effects of recombinant human granulocyte-macrophage colony-stimulating factor (rh GM-CSF) on late stages of megakaryocytopoiesis in the rhesus monkey. Four rhesus monkeys were given 10 micrograms/kg body weight/day of rh GM-CSF s.c. in two divided doses daily for 8 days. Megakaryocyte maturation was evaluated serially by measuring nuclear ploidy and cytoplasmic size. GM-CSF-treated monkeys developed significant shifts in ploidy distribution from days 3 through 15 (p less than or equal to 0.001), with increased frequencies of 64N and 128N megakaryocytes. Mean megakaryocyte size increased 92.5% on day 9, paralleling the increase in DNA content, although megakaryocyte size within ploidy groups did not increase. Megakaryocyte number remained unchanged following rh GM-CSF treatment. The platelet count responses were variable, and mean platelet volume did not change. The present study demonstrates that therapeutic doses of rh GM-CSF stimulate megakaryocyte endomitosis and increase mean size. The data indicate that GM-CSF is effective in promoting the maturation stage of megakaryocyte development but does not result in a consistent thrombopoietic response.

Maintenance of normal platelet mass in anemic Belgrade rats and their response to iron.

The Belgrade (b/b) rat has severe hypochromic, microcytic anemia accompanied by mild thrombocytopenia and a 49% reduction in megakaryocytes (MKs). The platelet counts are decreased only 34%, but relative platelet size measured by two independent methods averages 50% greater than controls. Thus, the platelet mass of the b/b rat is within the normal range. The marrow MK progenitors (MK colony-forming units, CFU-MK) respond linearly to increased colony-stimulating activity in vitro, but they are reduced 68% and form smaller colonies than normal. Flow cytometric analysis of MK ploidy indicates that significantly more MKs are distributed into the low and high ploidy classes compared with the normal, and the mean ploidy is similar. The b/b rat maintains effective thrombocytopoiesis in spite of a severe reduction in MK progenitors, primarily by an increased rate of maturation of the endomitotic compartment. Iron treatment partially arrests the b/b anemia and is associated with a significant increase in CFU-MK, a normalized MK ploidy distribution, and a significant decrease in platelet size. The favorable response to iron therapy suggests that the megakaryocytopenia is secondary to the severe anemia and results from stem cell commitment to the erythroid lineage.

Postoperative sore throat: effect of oropharyngeal airway in orotracheally intubated patients.

The incidence of postoperative sore throat was evaluated prospectively in 203 orotracheally intubated patients undergoing general anesthesia for surgical procedures. Patients were randomly assigned to have either a plastic oropharyngeal airway or a gauze bite-block in place during the operation and were evaluated for the occurrence of postoperative sore throat by questionnaire the day after surgery. The incidence of postoperative sore throat was 35.2% in the oropharyngeal airway group and 42.5% in the gauze bite-block group, not a statistically significant difference (P greater than 0.05). The incidence of postoperative sore throat was significantly higher when blood was noted on the airway instruments (64.5%) than when it was not (30.9%) (P less than 0.01). There was an association, although not statistically significant, between the incidence of postoperative sore throat and intubation by an anesthesia resident with less than 1 yr experience (P = 0.064). The data from this study indicate that the intraoperative use of hard plastic oropharyngeal airways, compared with the use of soft gauze bite-blocks, does not increase the incidence of postoperative sore throat. These data also suggest that pharyngeal trauma may contribute significantly to the development of postoperative sore throat. We suggest that aggressive oropharyngeal suctioning may contribute to this pharyngeal trauma.

Preliminary trial of minidose heparin prophylaxis for painful sickle cell crises.

Four patients with well-documented severe recurrent painful sickle cell crises received 12-month courses of self-administered minidose heparin (5000-7500 units every 12 hours). Patients were evaluated weekly or bi-weekly for symptoms and signs of sickle crises. The observations were compared with identical observations during 12 months off heparin (control). Prior to starting heparin, baseline laboratory values were established, including hematocrits, platelet counts, and plasma coagulation factors. Bone density was evaluated by radiologic measurement of metacarpal cortical width and by forearm transmission osteodensitometry. Laboratory values were repeated at each visit; bone density was reevaluated every 6 months. At each visit clinical status for the preceding 1-2 weeks was classified as: no pain (I), minor pain controllable by non-prescription medications (II), pain requiring office or emergency room admissions for parenteral narcotics (III), hospitalized for about 3 days for pain crisis (IV). One patient completed a single 2-year cycle; one patient completed 2 cycles (followed 4 years), and 2 patients have undertaken 3 cycles (followed greater than or equal to 6 years). No treatment-related complications occurred requiring discontinuation of heparin. Thrombocytopenia was not observed. There was no evidence of progressive osteopenia. All patients improved while receiving heparin; 1 moderately, 3 markedly. Cumulative data (8.7 patient years on heparin, 12 control years) revealed a 73% reduction in days of hospitalization per year and 74% reduction in hours spent in emergency rooms per year during heparin administration. Pretreatment pain patterns recurred when heparin was discontinued.

Endotoxin-induced production of plasminogen activator inhibitor by human monocytes is autonomous and can be inhibited by lipid X.

Peripheral blood mononuclear cells (PBMs) produce both tissue factor and plasminogen activator inhibitor type 2 (PAI-2) in response to gram-negative bacterial lipopolysaccharide (LPS). The cellular roles in the tissue factor response have been previously elucidated, and we now report those roles in PAI-2 production. Monocytes are the only cells among LPS-stimulated PBMs that produce PAI-2 as assessed by measurement of PAI-2 activity and antigen. Concomitant immunohistochemistry demonstrated that monocytes contain PAI-2, with a greater number staining positively and more intensely after exposure to LPS. LPS-stimulated monocytes produced increased amounts of PAI-2 with or without addition of lymphocytes. Lymphocytes prestimulated with LPS and then washed did not induce PAI-2 production in monocytes to which they were added. Lipid X, a precursor in the biosynthetic pathway of lipid A and LPS, was able to inhibit LPS induction of monocyte PAI-2 in a dose-dependent manner. This inhibition was not due to cellular toxicity, the phospholipidlike nature of lipid X, interference with the PAI-2 assay, or monocyte production of a substance interfering with PAI-2. Lipid X was an effective inhibitor of PAI-2 production even when added up to 30 minutes after LPS.

Increased release of plasminogen activator inhibitor type 2 accompanies the human mononuclear cell tissue factor response to lipopolysaccharide.

Human peripheral blood mononuclear cells (PBM) respond to lipopolysaccharide (LPS) with increased release of a plasminogen activator (PA) inhibitor. This response is dose dependent and parallels the LPS-induced expression of PBM tissue factor activity. The PA inhibitors of control and LPS-stimulated PBMs appear identical as both are identified by antibodies to PA inhibitor type 2 of human placenta, but not by antibodies to type 1 inhibitor of bovine aortic endothelial cells. The PA inhibitor is specific for urokinase type PA as determined by the 125I-fibrin plate assay, and direct cleavage of 125I-plasminogen; it does not effectively inhibit tissue-type PA. The inhibitor forms an active site-dependent complex with 125I-urokinase, which then demonstrates an increase in mol wt from 33 kd to 68 kd on reduced sodium dodecyl sulfate (SDS) polyacrylamide gels. PBMs neither secrete nor express active PA. Hence, the exposure of PBMs to LPS results in conditions highly favorable to fibrin deposition and persistence: increased procoagulant and antifibrinolytic activities, accompanied by no measurable PA. Such modulation of these effectors may be important in the pathogenesis of fibrin characteristically found in tissue lesions of endotoxin-initiated processes.

Human platelet aggregation is initiated by peripheral blood mononuclear cells exposed to bacterial lipopolysaccharide in vitro.

Platelet consumption is a prominent feature of disseminated intravascular coagulation. We investigated whether monocyte procoagulant activity (PCA) might play a role in platelet consumption associated with gram-negative septicemia. Human mononuclear cells exposed in vitro to lipopolysaccharide demonstrated parallel dose-dependent increases in PCA and ability to induce platelet aggregation. Induction of platelet aggregation required the generation of thrombin dependent on coagulation Factors VII, X, and II, and calcium. This is consistent with monocyte tissue factor initiating thrombin generation. A specific monoclonal antimonocyte antibody was used to identify monocytes via indirect immunofluorescence, and demonstrated that all monocytes were included in platelet aggregates. Mononuclear cells that did not express PCA did not induce platelet aggregation and monocytes were not surrounded by platelet clumps. These data suggest that monocytes induced to express tissue factor on their surface may be important mediators of endotoxin-induced platelet, as well as fibrinogen, consumption.

Comparisons of pooled polyclonal rabbit anti-human C3d with four monoclonal mouse anti-human C3ds. I. Preparation, purification and binding properties.

Performance characteristics of pooled rabbit IgG polyclonal anti-C3d are compared with one mouse IgM and three mouse IgG monoclonal anti-C3d antibodies (MAs). IgG MA,s employed singly or in combination, failed to precipitate C3d; by contrast, IgM MA and polyclonal anti-C3d precipitated C3d. Measurements of polyclonal anti-C3d concentration by chemical means and by 125I-C3d radioimmunoassay (RIA) agreed closely. RIA values were 50% of chemical measurement values for three of the four MAs. Use of sucrose density gradient ultracentrifugation to assess MA C3d/anti-C3d molar combining ratios for soluble anti-C3d/C3d was not possible because fast-sedimenting multimeric C3d/anti-C3d complexes did not form. Dissociation and competitive binding studies indicate that (1) two MAs had substantially lower affinities than the other anti-C3d antibodies, and (2) polyclonal anti-C3d recognizes more C3d epitopes than are recognized by individual MAs. The results demonstrate antigenic complexity of C3d fragment and illustrate the difficulties of predicting individual MA performance based on prior experience with polyclonal antibodies.

Comparisons of pooled polyclonal rabbit anti-human C3d with four monoclonal mouse anti-human C3ds. II. Quantitation of RBC-bound C3d, and characterization of antiglobulin agglutination reactions against RBC from 27 patients with autoimmune hemolytic anemia.

Labelled polyclonal IgG anti-C3d and monoclonal IgM and IgG anti-C3d antibodies (MAs) were employed at increasing antibody excess to measure the number of C3d molecules on human red blood cells (RBC) coated by complement in vitro and in vivo. Values for the number of C3d sites per cell determined with polyclonal anti-C3d were at least 4-fold higher than when MAs were used. The results suggest that the molar combining ratio for polyclonal anti-C3d with a single RBC-bound C3d fragment is more likely greater than 4.0 than 1.0 as previously assumed. Antiglobulin agglutination studies compared polyclonal and monoclonal anti-C3d antibodies against C3d-coated RBC from 27 patients with autoimmune hemolytic anemia. All four MAs showed striking prozones, requiring their use over a 25-fold higher range of dilutions than polyclonal anti-C3d. Polyclonal anti-C3d produced stronger agglutination reactions than any of the IgG MAs. Only the IgM MA produced agglutination as strong as, or stronger than, polyclonal anti-C3d. While IgM MA always gave the strongest MA agglutination reactions, no consistent ranking of the three IgG MAs was observed. Agglutination was not enhanced when all IgG MAs were combined; addition of IgG MAs to IGM MA reduced the strength of agglutination seen with IgM alone, suggesting blocking of IgM binding by competing IgG anti-C3d.

In vivo instability of red-blood-cell-bound C3d and C4d.

Until now, there have been no measurements of the in vivo stability of red-blood-cell-bound C3d and C4d subfragments of the third and fourth components of human complement. We have recently described a radiolabeled antiantiglobulin method for measuring RBC-bound C3d and have demonstrated that small amounts of C3d are present on RBC of all normal subjects tested. In the present study, the method was applied to follow the increments above baseline of RBC-bound C3d and C4d produced by autotransfusing 3 normal volunteers with 160-200 ml of RBC strongly coated in vitro by C3d and C4d. Posttransfusion measurements were carried out over 21-34 days. Immediate and long-term in vivo survival of the transfused RBC was unimpaired by C3d and C4d coating. Of the bound C3d antigen, 85%-95% disappeared from circulating RBC in 5-8 days; the remainder disappeared more slowly, with half-times in the range of 8-29 days. C4d antigen disappeared substantially more slowly, describable by a single exponential function in 2 of the 3 subjects, with half-times in the range of 12-31 days. Recognition of the in vivo instability of RBC-bound C3d helps in interpreting steady-state and changing levels of RBC C3d coating in a variety of alloimmune and autoimmune disorders.

Further studies of the C3g component of the alpha 2D fragment of human C3.

Recent studies employing sheep red cells, rabbit anti-sheep cell IgM, guinea-pig R3 reagent and purified human C3, factor I, factor H and human serum have led to the discovery of a new C3b breakdown product antigen, C3g, which is present on the alpha 2D fragment of C3 but is removed by trypsinization. The present studies were undertaken to confirm the above findings in entirely human red cell and serum systems, including studies on cells from nine patients with a variety of complement-mediated autoimmune haemolytic anemias. The results confirm the recently reported findings in every respect. The presence of both C3d and C3g on all patient's red cells so far tested supports the concept that red cells coated by C3d alone are likely to be a laboratory artefact and that the alpha 2D fragment (C3d,g) is the natural end product of cell bound C3b breakdown in vivo.