Saliva enhances infection of gingival fibroblasts by herpes simplex virus 1

Abstract

Oral herpes is a highly prevalent infection caused by herpes simplex virus 1 (HSV-1). After an initial infection of the oral cavity, HSV-1 remains latent in sensory neurons of the trigeminal ganglia. Episodic reactivation of the virus leads to the formation of mucocutaneous lesions (cold sores), but asymptomatic reactivation accompanied by viral shedding is more frequent and allows virus spread to new hosts. HSV-1 DNA has been detected in many oral tissues. In particular, HSV-1 can be found in periodontal lesions and several studies associated its presence with more severe periodontitis pathologies. Since gingival fibroblasts may become exposed to salivary components in periodontitis lesions, we analyzed the effect of saliva on HSV-1 and -2 infection of these cells. We observed that human gingival fibroblasts can be infected by HSV-1. However, pre-treatment of these cells with saliva extracts from some but not all individuals led to an increased susceptibility to infection. Furthermore, the active saliva could expand HSV-1 tropism to cells that are normally resistant to infection due to the absence of HSV entry receptors. The active factor in saliva was partially purified and comprised high molecular weight complexes of glycoproteins that included secretory Immunoglobulin A. Interestingly, we observed a broad variation in the activity of saliva between donors suggesting that this activity is selectively present in the population. The active saliva factor, has not been isolated, but may lead to the identification of a relevant biomarker for susceptibility to oral herpes. The presence of a salivary factor that enhances HSV-1 infection may influence the risk of oral herpes and/or the severity of associated oral pathologies.

Fig 1. Human salivary proteins enhance HSV-1 infection of human gingival fibroblasts.

(A) Infection assay of HSV-1 KOStk12, measured as activity of virus-encoded β-galactosidase. Saliva samples (30K retentate) from the indicated donors were added into the growth medium of AG09319 gingival fibroblasts for 48 h at the indicated concentrations. Growth medium containing salivary proteins was removed and cells were exposed to the reporter virus HSV-1 KOStk12 (MOI = 5) for 5 h. Cells were lysed, and the activity of β-galactosidase produced by the virus was measured to monitor infection. Black symbols represent native salivary samples. Open symbols represent heat-denatured (HD) saliva samples. The dashed line represents the baseline level of β-galactosidase activity in infected fibroblasts in the absence of saliva stimulation. Although saliva stimulation is consistently observed, the extent of the enhancement varied between experiments. Furthermore, the background levels of infection of AG09319 fibroblasts varied as well. The cause of this variation remains unclear. Thus, normalization of effect over background has proven to be uninformative in comparing experiments. Therefore, we show a representative of 2 or more experiments. (B) Saliva samples (30K retentate) from donor 4 were incubated with Proteinase K agarose for 4 h at 37°C or mock treated, and then serially diluted in culture medium for cell stimulation. The x axis is labeled according to protein concentration prior to PK treatment. AG09319 fibroblasts were treated and infected as in panel A. As a control without saliva stimulation, fibroblasts were mock treated with buffer (triangles). Error bars represent standard deviations between at least 3 independent experiments. (C) AG09319 gingival fibroblasts were pretreated with P. gingivalis LPS at the indicated concentrations prior to infection with HSV-1 KOStk12 (MOI = 2). An average of three experiments is shown. Error bars represent standard deviations. For comparison, infection levels in the absence of any stimulation, which are shown as a dashed line (panel A) or black triangles (panels B-C), correspond to equivalent “mock treatment” conditions.

Fig 2. Human saliva enhances infection of human gingival fibroblasts by HSV-1 but not HSV-2.

Diploid human oral fibroblasts AG09319 were exposed to salivary proteins (donor 1) at the indicated concentrations or mock treated with PBS (buffer), and exposed to HSV-1 KOStk12 or HSV-2(333)gJ- reporter viruses at MOIs of 5 and 1.5 pfu/cell respectively. The activity of virus-encoded β-galactosidase is used to monitor infection. A representative of at least three experiments is shown.

Fig 3. HSV-1 production by gingival fibroblasts stimulated by human saliva.

AG09319 fibroblasts were stimulated with saliva proteins (donor 2) at 4 mg/ml for 48 h prior to infection. Cell culture supernatants were collected at 8, 24 and 48 h post-infection and titered on Vero cells. Each dot represents the ratio of titers of stimulated over unstimulated cells. A ratio of 1 (dashed lines) corresponds to similar titers in stimulated and control cells. Data from four stimulation experiments are shown. The horizontal bars indicate average fold increase. Statistical t-test analysis showed that the average increase in titer in stimulated cells is statistically significant at 24 h post-infection (p = 0.0169), indicated by an asterisk.

Fig 4. Human saliva promoted entry into receptor deficient B78H1 cells.

(A) Entry assay. Salivary proteins (30K retentate) from the indicated donors were added into the growth medium of B78H1 cells for 48 h. Growth medium containing salivary proteins was removed and cells were exposed to HSV-1 KOStk12 for 5 h. Cells were lysed, and β-galactosidase activity was measured to monitor infection. Black symbols represent native saliva proteins. Open symbols represent saliva proteins which were heat-denatured (HD) by boiling for 5 min. The dashed line represents the baseline level of β-galactosidase activity in infected B78H1 cells in the absence of saliva stimulation. (B) Plaque assay. B78H1 cells were pre-treated with saliva proteins (donor 1) for 48 h. Medium was removed and cells were exposed to 10⁴ pfu of HSV-1 KOS. Cells were overlaid with methylcellulose and plaque were allowed to form for 48 hours. Infected cells were immunostained with anti-HSV glycoprotein antibodies. As a control for infection, susceptible C10 cells (i.e. B78H1 cells expressing human nectin-1) were exposed to only 10² pfu of HSV-1 KOS and showed plaques comprising multiple infected cells (arrow, right panel). The left panel shows unstimulated B78H1 cells and no signs of infection. (C) Entry assay in the presence of anti-receptor antibodies that specifically block usage of nectin-1 (monoclonal antibody CK41) or HVEM (polyclonal serum R140). B78H1 cells were stimulated (black) or not (white) with saliva proteins from donor 2 (30K retentate, 1.8 mg/ml). After removal of saliva, cells were preincubated with purified immunoglobulins from R140 serum or CK41 monoclonal antibody at 100 mg/ml for 1 h before the addition of HSV-1 KOStk12. Non-immune immunoglobulins were used as control. The activity of virus-encoded β-galactosidase is used to monitor infection. An average of three experiments is shown with error bars representing standard deviations.

Fig 5. Variation and dose response of salivary activity between donors.

Salivary proteins (30 kDa retentate) was added to B78H1 cell culture medium at the indicated concentrations for 48 h. Stimulating medium was removed and cells were exposed to HSV-1 KOStk12 for an entry assay. Infection was monitored by measuring β-galactosidase activity after a 5 h infection. Results from a representative of at least two independent experiments is shown.

Fig 6. Partial fractionation of salivary components.

(A) Diagram of fractionation steps through cation (Mono S) and anion (mono Q) exchange chromatography. FT: Flow-through. (B) Saliva activity determined by a virus entry assay. B78H1 cells were incubated in the presence of various concentrations of salivary proteins (donor 1) from the indicated fractions during 48 h. Saliva-containing growth medium was removed and cells were exposed to HSV-1 KOStk12 for 5 h. Infection was monitored by β-galactosidase activity by measuring absorption of substrate (CPRG) at 595 nm. A representative of at least two experiments is shown. (C) Silver stained gel showing proteins in the indicated fractions. + and–signs indicate the presence of infection-enhancing activity as determined in C. Molecular weights (MW) of marker proteins are indicated. (D) Size exclusion chromatography. Fraction eluted from the monoQ column with 250mM NaCl was loaded on a Superdex200 column. Proteins from size exclusion fractions were subjected to SDS-PAGE. A composite image of silver stained gels from the same experiment is shown. P₁ to P₄ indicate pooled fractions used in stimulation assay. (E) Elution fractions were pooled (P₁ to P₄), proteins were quantified and used at two concentrations to stimulate B78H1 cells. Infection by HSV-1 KOStk12 is reported as β-galactosidase activity as in B. A representative of at least 2 experiments is shown. (F) Fractions from activity peak 1 (panel D and E) were pooled and loaded on a polyacrylamide gel under native (N) or denaturing conditions (D). (G) Unfractionated saliva sample (30 kDa retentate) was incubated with Jacalin-agarose for 4 h at 4°C. Treated and untreated salivary proteins were tested on AG09319 fibroblasts. After 48 h pre-treatment with salivary proteins (1 mg/ml) cells were infected with HSV-1 KOStk12. Infection is reported as β-galactosidase activity 5 h post-infection. The dotted line indicates background signal in the absence of infection. Error bars represent standard deviations.