In vivo role of complement-interacting domains of herpes simplex virus type 1 glycoprotein gC.

Immune evasion is critical for survival of viruses that establish persistent or recurrent infections. However, at the molecular level, little is known about how viruses evade immune attack in vivo. Herpes simplex virus (HSV)-1 glycoprotein gC has two domains that are involved in modulating complement activation; one binds C3, and the other is required for blocking C5 and properdin (P) binding to C3. To evaluate the importance of these regions in vivo, HSV-1 gC mutant viruses were constructed that lacked one or both gC domains and studied in a murine model of infection. Each gC region of complement regulation contributed to virulence; however, the C3 binding domain was far more important, as virus lacking this domain was much less virulent than virus lacking the C5/P inhibitory domain and was as attenuated as virus lacking both domains. Studies in C3 knockout mice and mice reconstituted with C3 confirmed that the gC domains are inhibitors of complement activation, accounting for a 50-fold difference in virulence between mutant and wild-type viruses. We conclude that the C3 binding domain on gC is a major contributor to immune evasion and that this site explains at a molecular level why wild-type virus resists complement attack.

Viral infection of vascular endothelial cells alters production of colony-stimulating activity.

Viral infections in humans are frequently associated with granulocytopenia and/or granulocytosis. Such changes in myelopoiesis could result from infection of the granulocyte-macrophage colony-forming cell (CFC-GM) or changes in the production of colony-stimulating activity (CSA). Endothelial cells are a known source of CSA and may be transiently or persistently infected during a number of viral infections, including infection with herpes simplex virus type I (HSV-I) and measles virus. Therefore, we examined the effect of endothelial cell infection with these two viruses on the production of CSA. Uninfected passaged endothelial cells produce CSA when stimulated by the continual presence of a factor present in medium conditioned by peripheral blood monocytes (MCM). Within 4 h of infection with HSV-I, endothelial cells no longer produced CSA in response to MCM. In contrast, measles virus infection induced CSA production by passaged endothelial cells even in the absence of MCM. Measles virus-induced CSA production was maximal at 24 h and required the presence of live virus within the endothelial cells. The effects of HSV-I and measles virus on CSA production were not dependent on alterations in the production of alpha- or gamma-interferon by the infected endothelial cells. Infection with HSV-I did not stimulate endothelial cells to release any detectable interferon. In contrast, the supernatants of the measles-infected cells contained only beta-interferon, a known inhibitor of CFC-GM development. These studies suggest that CSA production by endothelial cells is directly altered by infection with HSV-I and measles virus. An alteration in CSA production might contribute to changes in myelopoiesis that frequently accompany viral infection in humans.

Virus infection of endothelial cells increases granulocyte adherence.

Adherence of human granulocytes was measured on endothelial monolayers of human and bovine origin, grown in 35-mm Diam petri dishes and in cluster wells. Adherence to human endothelium in petri dishes using 1.0 ml of whole blood averaged 17.9+/-3.7%, and to bovine endothelium was 20.3+/-3.7%. Cluster wells required only 1/5 the endothelial cells needed for petri dishes, and 0.25 ml of whole blood yielded average adherence of 26.2+/-3.4 to human cells and 28.0+/-3.7 to bovine in the wells. The impact of infection of the endothelium by different viruses on subsequent granulocyte adherence was measured. Polio virus produced an acute lytic infection of human endothelial cells, with associated increased adherence to 185.4% of control 24 h after inoculation. Significantly increased adherence was noted at 6 h, before detectable cytopathic effect. Herpes simplex type I caused a similar rapidly lytic infection of bovine endothelium associated with increased adherence to 213.7% of control 6 h after inoculation. This augmented adherence could be demonstrated when granulocytes were suspended in physiologic saline solution, showing that antibody and complement need not be present. Trypsin treatment of infected monolayers did not prevent the augmentation, and supernate from infected monolayers increased the adherence of polymorphonuclear leukocytes to normal, uninfected monolayers. Chronic, slowly lytic infections, lasting 7 d or more, were induced with adenovirus in human endothelium and with measles virus in bovine cells. Adherence increased as virus was noted in the cell cultures on day 4, several days before cytotoxicity was seen. Thus, chronic viral infection of the endothelium appears possible, and results in increased granulocyte adherence. In naturally occurring disease, such an infection may act synergistically with adherent granulocytes to damage the endothelium, and may represent an in vitro model of vasculitis.

Fc and C3 receptors induced by herpes simplex virus on cultured human endothelial cells.

The mechanism by which immune complexes deposit in vascular tissue is uncertain. Several human viruses, including herpes simplex virus, have recently been demonstrated to replicate in human endothelial cells. Such viruses may injure vascular tissue and could play a role in the pathogenesis of immune complex deposition. Therefore, we studied the expression of receptors for immune complexes containing IgG and C3 on endothelial cells after infection with herpes simplex virus type I.Human umbilical vein endothelial cells were incubated with (51)Cr-labeled sheep erythrocytes sensitized with IgG, IgM, or IgM plus complement. Preferential binding of IgG or complement-coated erythrocytes to uninfected endothelial monolayers was not observed. In contrast, significant binding of erythrocytes coated with IgG or IgM plus complement was observed after viral infection. Phase-contrast and scanning electron microscopy demonstrated erythrocyte adherence around the infected endothelial cells in a rosette pattern. Binding of IgG-coated erythrocytes was fully inhibited by Fc (0.31 mg/ml) but not Fab' fragments of nonimmune IgG. Binding of complement-coated cells was unaffected by the presence of IgG (1 mg/ml). With purified individual components, binding of complement-coated erythrocytes depended on the presence and was proportional to the concentration of C3. Binding of IgG-or C3-coated cells was detected beginning 4 h after infection. These studies indicate that herpes simplex virus type I infection can induce IgG and C3 receptors on human endothelial cells. These receptors may promote the deposition of immune complexes in vascular tissue after certain viral infections.

Thrombin enhances herpes simplex virus infection of cells involving protease-activated receptor 1.

BACKGROUND: We have previously shown that the surface of purified herpes family viruses can initiate thrombin production by expressing host-encoded and virus-encoded procoagulant factors. These enable the virus to bypass the normal cell-regulated mechanisms for initiating coagulation, and provide a link between infection and vascular disease. OBJECTIVE: In the current study we investigated why these viruses may have evolved to generate thrombin. METHODS: Using cytolytic viral plaque assays, the current study examines the effect of thrombin on human umbilical vein endothelial cell (HUVEC) or human foreskin fibroblast (HFF) infection by purified herpes simplex virus type 1 (HSV1) and type 2 (HSV2). RESULTS: Demonstrating that the availability of thrombin is an advantage to the virus, purified thrombin added to serum-free inoculation media resulted in up to a 3-fold enhancement of infection depending on the virus strain and cell type. The effect of thrombin on HUVEC infection was generally greater than its effect on HFF. To illustrate the involvement of thrombin produced during inoculation, hirudin was shown to inhibit the infection of each HSV strain, but only when serum containing clotting factors for thrombin production was present in media. The involvement of protease-activated receptor 1 (PAR1) was supported using PAR1-activating peptides in place of thrombin and PAR1-specific antibodies to inhibit the effects of thrombin. CONCLUSION: These data show that HSV1 and HSV2 initiate thrombin production to increase the susceptibility of cells to infection through a mechanism involving PAR1-mediated cell modulation.

Use of a glucocorticoid-inducible promoter for expression of herpes simplex virus type 1 glycoprotein gC1, a cytotoxic protein in mammalian cells.

Abundant expression of herpes simplex virus type 1 glycoprotein gC (gC1) in transfected mammalian cells has not previously been achieved, possibly because gC1 protein is toxic to cells. To approach this problem, the gC1 coding sequence was placed under the control of the weak but inducible glucocorticoid-responsive promoter from the mouse mammary tumor virus (MMTV) long terminal repeat (LTR). As controls to evaluate for gC1 cytotoxicity, the MMTV LTR promoter was used to express glycoprotein gD1, and a strong, constitutive promoter from the Moloney murine sarcoma virus LTR was used to express gC1. L cells were transfected with these constructs, and a clone expressing gC1 from the inducible MMTV LTR promoter was analyzed. In the absence of glucocorticoid (dexamethasone) stimulation, only a low level of gC1 mRNA expression was detected; after overnight stimulation with dexamethasone, transcription increased approximately 200-fold. Abundant gC1 protein that was functionally active in that it bound complement component C3b, was produced. From passages 5 through 26 (70 cell population doublings), the gC1-producing clone became less responsive to overnight dexamethasone stimulation. The block to gC1 expression occurred at the level of transcription and was associated with hypermethylation of the MMTV LTR DNA. Treatment of the clone with 5-aza-2'-deoxycytidine partially reversed the block in gC1 protein production. Late-passage cells assumed a gC1-negative phenotype that appeared to offer a selective growth advantage, which suggested that gC1 was cytotoxic. Several findings support this view: (i) some cells expressing gC1 after overnight stimulation with dexamethasone assumed bizarre, syncytial shapes; (ii) continuous stimulation with dexamethasone for 5 weeks resulted in death of most cells; (iii) cells transfected with gC1 under the control of the strong Moloney murine sarcoma virus promoter assumed bizarre shapes, and stable gC1-expressing clones could not be established; and (iv) cells induced to express gD1 retained a normal appearance after overnight stimulation or 15 weeks of continuous stimulation with dexamethasone. The inducible MMTV LTR promoter is useful for expressing gC1 and may have applications for expressing other cytotoxic proteins.

Assessment of in vitro drug sensitivity of human tumor cells using [3H]thymidine incorporation in a modified human tumor stem cell assay.

We have developed a method for performing in vitro drug testing on primary human tumor explants which is a variant of the human tumor stem cell assay (HTSCA) described previously by Salmon et al. (N. Engl. J. Med. 298: 1321, 1978). The method utilizes a cell-containing liquid top layer and a soft-agar bottom layer. Tumor growth is measured by [3H]thymidine incorporation into material precipitable by 5% trichloroacetate. Results show linear correlations with number of cells plated and with a number of colonies per plate measured using the HTSCA, when cell aliquots from one sample are used. In vitro drug sensitivity, as determined by inhibition of [3H]thymidine incorporation, correlates with HTSCA results in 54 of 61 determinations (89%). Of 22 experiments in which drug sensitivity curves were compared, 21 (95%) were similar in both systems. The [3H]thymidine method yields results more quickly (5 days after samples are plated) and with smaller variances than those measurements obtained using the HTSCA. Normal human skin muscle, lung, and colon tissue and a human fibroblast cell line do not incorporate significant amounts of [3H]thymidine into trichloroacetic acid-precipitable material. Thus, normal cell components plated in tumor samples do not interfere with assay results. Standard scintillation counting is used; optical counting, either visual or automated, is not required. Therefore, the measurement of in vitro drug sensitivity by inhibition of incorporation of radiolabeled precursors deserves further evaluation as a predictor of in vitro response.

Human leukemia-derived cell lines and clones as models for mechanistic analysis of natural killer cell-mediated cytotoxicity.

Tumor target cells (TC) are lysed by natural killer (NK) cells provided that they (1) form conjugates with the effector cells, (2) activate effector cells to release cytotoxic factors, and (3) they are susceptible to the lytic effect of these factors. While this cascade of events that leads to TC killing has been defined, the signal molecules responsible for each of the steps remain largely undetermined. A variety of human leukemia-derived TC lines and clones were analyzed for their sensitivity to NK cell-mediated lysis and for their ability to bind and activate NK cells. These characteristics have been correlated with TC surface expression of differentiation antigens and carbohydrate residues. Of the cell lines and clones tested, K562, SPI-802, MOLT-4, MOLT-4/C8-1, ZS, KG-1/A-3, and HL-60S were sensitive to NK cell-mediated lysis, while KG-1, THP-1-0, HL-60R, and LFM were resistant. KG-1, THP-1-0, HL-60R, and LFM cells were further studied to determine mechanisms responsible for their resistance to NK cells. It was found that HL-60R and LFM cells were unable to bind NK cells. In contrast, KG-1 and THP-1-0 cells were able to bind to and activate NK cells. Therefore, it is likely that the NK-resistance of KG-1 and THP-1-0 cells may be related to their lack of sensitivity to cytotoxic factors released by bound NK cells. All of the TC cell lines and clones capable of binding NK cells expressed the 3-fucosyl-N-acetyl-lactosamine hapten (Lex or SSEA-1 antigen) recognized by the monoclonal antibody Leu M1. These TC consistently lacked surface L-fucose residues, as shown by lack of Ulex europaeus agglutinin binding. In contrast, HL-60R and LFM which did not form conjugates with NK cells, did not express surface Lex determinants and avidly bound the Ulex agglutinin. Distinct subpopulations of NK-resistant KG-1 cells expressed Lex antigens or bound Ulex. We compared KG-1/A-3, a NK-sensitive cell clone, with the parental NK-resistant KG-1 cell line. KG-1/A-3 lost the ability to bind the Ulex lectin displayed by the parental cell line and showed increased expression of Lex determinants. Results from these phenotypic analyses suggest that expression of Lex determinants and Ulex binding sites on the TC membrane are mutually exclusive and their expression or absence may correlate with mechanisms which regulate TC-NK cell interactions.

Entry of human immunodeficiency virus-1 into glial cells proceeds via an alternate, efficient pathway.

Although the CD4 molecule is the cellular receptor for human immunodeficiency virus-1 (HIV-1) in cells of the lymphocyte/monocyte lineage, a number of investigators have also been able to infect cells, including several of central nervous system (CNS) origin, that do not express CD4 protein or mRNA. These infections are generally nonpermissive. To ascertain whether the nonpermissive nature of infection in glial cells is due to an inefficient entry pathway, we prepared a permanently transfected U373-MG cell line expressing the CD4 molecule and demonstrated that HIV-1 still replicates at a low level. Furthermore, a virus uptake assay indicated that HIV-1 enters glial cells effectively, even in the absence of CD4. These results demonstrate that HIV-1 entry is efficient and that the restrictive nature of the infection in glial cells is due to postentry mechanisms. In addition, these findings support the existence of an alternate, efficient, entry pathway in some glial cells.

Combination chemotherapy and radiation therapy. The medical management of epidural spinal cord compression from testicular cancer.

Radiation and surgery are the mainstays of treatment for epidural spinal cord compression. There are reports in the literature, however, in which the use of chemotherapy alone or in combination with radiation and surgery has resulted in improved patient response. We present two patients with spinal cord compression from testicular cancer who developed progressive paraplegias that were successfully reversed with combination chemotherapy, steroids, and radiotherapy. In one case, the patient was unable to receive more than 600 rad (6 Gy) of radiation and his improvement was clearly related to the chemotherapy he received. In treating tumors causing spinal cord compression that are thought to be chemosensitive, one should consider adding chemotherapy to radiation or using chemotherapy alone if the options of surgery or radiation are not available.

Effect of adherence to newly initiated antiretroviral therapy on plasma viral load.

OBJECTIVE: To determine whether differences in adherence to newly initiated antiretroviral therapy exist between subjects who do and do not achieve undetectable plasma viral loads. DESIGN: Observational cohort study monitoring adherence and virological and immunological parameters over the initial 4 months of therapy with nelfinavir. Adherence was measured using the microelectronic monitoring system (MEMS; APREX Corporation, Menlo Park, California, USA). SETTING: General Clinical Research Center at a tertiary care center. PARTICIPANTS: Forty-one protease inhibitor-naive subjects with viral loads > 10 000 copies/ml newly starting a regimen including nelfinavir, referred from HIV clinics in Philadelphia. MAIN OUTCOME MEASURES: The primary outcome was undetectable viral load (< 50 copies/ml) after 4 months. Secondary measures included changes in viral load and CD4 cell counts. We hypothesized that adherence would be greater in subjects who achieved undetectable viral loads. RESULTS: Adherence was greater in undetectable subjects, who took a median of 93% of prescribed doses [interquartile range (IQR) 84-96%], whereas detectable subjects took a median of 70% (IQR 46-93%). Adherence correlated with viral load decrease (Spearman's rho = 0.38, P < 0.01) and CD4 cell count increase (Spearman's rho = 0.25, P = 0.06). Despite differences between the groups over 4 months of therapy, there were no adherence differences over the first month [undetectables, 95% (IQR 88-98%) versus detectables, 94% (IQR 87-98%), P > 0.50]. CONCLUSIONS: Adherence is important in determining whether or not individuals achieve suppression with a newly initiated antiretroviral regimen. Adherence begins to wane after the first month of therapy. Therefore, closer assessment of adherence particularly after this first month is important.

The effect of neonatal thymectomy on Tamiami virus-induced central nervous system disease.

Tamiami virus, a member of the arenavirus group, produces an acute CNS disease in suckling mice manifested primarily by cerebellar ataxia, paralysis, convulsions, and death. Animals that survive are left with an asymptomatic cerebellar heterotopia. Neonatal thymectomy prevents both acute CNS disease and the resultant cerebellar heterotopia despite equivalent titers of virus and concentrations of viral antigen in the brains of both thymectomized and nonthymectomized infected mice. Inflammatory CNS disease and cerebellar germinal cell necrosis do not develop in thymectomized mice examined more than three months after infection. Viremia and complement-fixing antibody occur in both groups of mice with slightly higher antibody titers in nonthymectomized mice. Tamiami virus-induced cerebellar heterotopia appears to be immunologically-mediated, but the immunopathologic cerebellar lesion differs from the frank necrosis of the brain produced by both Tacaribe and LCM virus in newborn mice.

Hydrocephalus produced by the 6/94 virus; A parainfluenza type 1 isolate from multiple sclerosis brain tissue.

The 6/94 virus, parainfluenza type 1 isolate from multiple sclerosis brain tissue, produced hydrocephalus in newborn Syrian hamsters. All animals developed clinical disease and died within a week. Ependymal cells lining the aqueduct of Sylvius became necrotic and fused, resulting in obstructive hydrocephalus. The 6/94 virus antigen was seen in ependyma and meninges. Paramyxovirus nucleocapsids were seen within cytoplasm of ependymal cells. Virus was recovered from hamster brains for only two days. Infectious virus could be recovered from brains grown in vitro as explants for 21 days. No evidence of rising hemagglutination-inhibiting antibody was noted for up to one month after infection. Intraperitoneal or subcutaneous injection of 6/94 virus did not produce hydrocephalus. HA2 virus and the temperature sensitive mutant of HA2 virus failed to produce hydrocephalus, while Sendai virus caused lesions similar to those of 6/94 virus.

Conjugal temporal arteritis.

We report the simultaneous occurrence of biopsy-proven temporal arteritis in husband and wife. Serologic and viral studies were negative, including viral culture of the wife's temporal artery. The concurrent incidence of giant cell arteritis in a married couple would suggest a common exogenous exposure.

Flow cytometry evaluation of cell-mediated cytotoxicity.

A novel flow cytometry method for the evaluation of cell-mediated cytotoxicity is described. This method uses flow cytometry analysis to distinguish target cells from effector cells by differences in volume and light scatter characteristics. Non-viable target cells, following their interaction with effector cells, are determined via propidium iodide (PI) dye exclusion and then expressed as a percentage of the total target cell population. This assay is suitable both for analysis of systems which allow recycling of cytotoxic effector cells (total cell cytotoxicity assays, TCCA), and of systems in which recycling does not occur (single cell cytotoxicity assays, SCCA). Natural killer (NK) cell-mediated cytotoxicity evaluated by flow cytometry is significantly correlated with the standard 51Cr release assay. Flow cytometry can also be used to evaluate the competitive inhibition that certain cell types exert on the cell-mediated killing of NK-sensitive targets. A prerequisite for this assay is that competitor cells and target cells are distinguishable through their volume and light scatter characteristics. Advantages and pitfalls of the flow cytometry method are discussed, in comparison with the 51Cr-release assay.

Natural killing of cytomegalovirus-infected targets in renal transplant recipients.

The ability of peripheral blood mononuclear cells of renal transplant recipients to lyse cytomegalovirus (CMV) infected fibroblasts was determined in an 18-hr 51Cr release assay. Natural killing (NK) against CMV-infected targets was generally depressed for the first two years after transplantation. In three individuals who developed CMV disease after transplantation, NK activity against CMV-infected and uninfected targets rose to high levels following reductions in immunosuppressive therapy and in temporal association with resolution of disease.

The role of pretransplant immunity in protection from cytomegalovirus disease following renal transplantation.

To determine the extent to which pretransplant immunity resulting from natural infection protects against cytomegalovirus (CMV) disease, we analyzed CMV serology on 153 kidney donor and recipient pairs and followed transplant patients to determine incidence and severity of CMV disease. The overall incidence of CMV disease was 22%. Significant differences occurred in CMV disease incidence and severity, depending on the immune status of the kidney donor and recipient. Among recipients of kidneys from seropositive donors, immunity offered significant protection from CMV disease, reducing its incidence from 61% in nonimmune to 24% in immune patients (P less than 0.01). Pretransplant immune patients also had fever CMV-related complications. Among recipients of kidneys from seronegative donors, pretransplant immunity conferred a significant risk of CMV disease; immune patients had a 20% incidence of CMV disease compared with 2% in nonimmune patients (P less than 0.02). Disease was generally mild in all patients receiving kidneys from CMV infection had a 3-fold higher incidence of CMV disease than patients with reactivation infection (P less than 0.01). The incidence of CMV disease was similar in immune patients, whether they received a kidney from a seropositive or a seronegative donor. However, an important observation was that disease was significantly more severe in immune patients receiving a kidney from a seropositive donor (P less than 0.05). This indicates that if kidneys from seropositive donors are selected for use only in seropositive recipients, this places the immune patient at a higher risk for severe CMV disease. We conclude that pretransplant immunity offers a significant advantage to patients receiving kidneys from seropositive donors.

Vulvovaginitis in premenarcheal girls: clinical features and diagnostic evaluation.

Fifty-four premenarcheal patients (median age 5.8 years) with symptoms or signs of vulvovaginitis were studied, and the results of cultures of vaginal secretions were compared with those from an age-matched control group. Vaginal discharge was found on examination in 24 of 42 patients with a complaint of discharge, and in two of 12 patients without a complaint of discharge. Convincing evidence of bacterial or monilial infection was found in 14 of the 26 patients with discharge on examination, but in none of the 28 patients without discharge (P less than .001). In the latter group pinworm infestation was present in one patient. Moniliasis occurred exclusively in girls who were pubertal (P less than .001). Four patients were found to have gonorrhea. No patient appeared to have symptoms or signs caused by Bacteroides sp, Chlamydia trachomatis, viruses, or Trichomonas vaginalis. Noninfectious causes were identified in four patients with and 13 without discharge (P less than .025); the most common cause was poor hygiene, implicated in six patients. Bubble bath use was implicated in only one patient. In 22 patients, no specific cause could be identified. All patients with poor hygiene as the only cause, and most with no demonstrable etiology, recovered after being advised to institute improved perineal hygiene. Patients with vaginal discharge are likely to have specific infections, and therefore cultures should be taken, in particular for Neisseria gonorrhoeae. Genital pruritus in prepubertal girls has little or no etiologic specificity, but in pubertal girls with vaginal discharge it suggests the presence of monilial vaginitis.

Outbreak of cryptosporidiosis in a day-care center.

An outbreak of diarrhea due to infection with Cryptosporidium occurred in a day-care center. During a period of 2 months, 23 of 53 (43%) children attending the day-care center and 15 of 104 (14%) household contacts had diarrhea. Cryptosporidium oocysts were identified in 13 of 20 (65%) symptomatic children tested compared with three of 27 (11%) asymptomatic children (chi 2 = 12.56, P less than .001). Enteropathogenic bacteria, enteroviruses, rotavirus, and other protozoan parasites were ruled out as the cause of the diarrhea. A history of diarrhea in household contacts was associated with excretion of Cryptosporidium oocysts by the children. Human-to-human transmission of the infection was suggested by the epidemiology.

Human submandibular saliva specifically inhibits HIV type 1.

Studies from a number of laboratories have shown the presence of factor(s) in whole, parotid, and submandibular human saliva capable of inhibiting HIV-1 infectivity in vitro. Data from our laboratory suggested that the level of anti-HIV-1 activity is higher in submandibular than parotid or whole saliva. Previous results obtained with pooled submandibular saliva from seronegative individuals included a filtration step following saliva-virus interaction. In this article, we present data on the HIV-1 inhibitory activity of individual submandibular saliva samples collected from 15 donors. We show that although anti-HIV activity is quantitatively similar in most individuals (9 of 15), some (4 of 15) are much less active than others and some (2 of 15) lack inhibitory activity. We also show that for most individuals the level of anti-HIV inhibitor is similar with or without a filtration step. However, 2 of the 15 samples demonstrated activity only after filtration. The quantitative and qualitative anti-HIV activity of individual saliva samples appeared to reflect differences in the individual donors. We further show that the anti-HIV activity of submandibular saliva is demonstrated not only against laboratory strains of HIV-1 but is similarly active against three clinical HIV-1 isolates. In contrast, submandibular saliva had little effect on the infectivity of HIV-2 or SIV.