*2.4. Histopathology and Immunohistochemical Detection of Antigens within the Synovial Membrane*

To better explore the extent of synovitis and the contribution of cells to the development of inflammation, we performed a microscopical analysis of the synovial membrane tissue samples. In the first set of histopathological examinations, we assessed the synovial morphology in HHV-7 PCR+ and HHV-7 PCR− groups when inflammation was not confirmed microscopically. Histopathologically, the lining cells formed one layer, the synovial stroma revealed normal cellularity, and no inflammatory infiltrates were present. In contrast, the synovial lesions consistent with low-grade synovitis demonstrated an increase in thickness of the lining layer and stromal cellularity, and the presence of a few, mostly perivascular lymphocytes or/and plasma cells (Figure 3c). Comparatively, high-grade synovitis was distinguished by the presence of a greatly thickened lining; the appearance of ulceration and multinucleated giant cells; greatly increased stromal cellularity; and, finally, the presence of numerous lymphocytes and plasma cells, often forming follicle-like aggregates (Figure 3d).

**Figure 3.** The presence of synovitis assessed by Krenn scores (**a**), statistically (**b**), and morphologically (**c**). (**a**) Violin plots depict the differences in median values demonstrated in the histopathological assessment of Krenn scores from the study groups. Synovitis revealed in the tissue samples of HHV-7 PCR+ OA patients presented with a median value of 3 (IQR 2–4), whereas HHV-7 PCR− OA patients presented with a median value of 2 (IQR 1–4). There was no significant difference found in the synovitis scores estimated for the HHV-7 PCR+ and HHV-7 PCR− groups (*p* = 0.483). Simultaneously, Krenn scores tended to be higher in the HHV-7 PCR+ group when compared to the HHV-7 PCR− group. (**b**) Frequencies of the absence of synovitis and the presence of low- and high-grade synovitis detected during histopathological assessment of the synovial membrane tissue samples from the study groups. The estimation confirms that OA patients commonly present without histopathologically detectable synovitis or demonstrate low-grade synovial inflammation. (**c**) A representative image depicting low-grade synovitis in OA. The synovial lining layer is slightly thickened, and the stromal density is slightly increased; few perivascular lymphocytes are evident (arrow). Hematoxylin and eosin staining. (**d**) A low-power image demonstrating high-grade synovitis in OA. The synovial lining layer is moderately thickened, and some lymphocytes are evident; stroma reveals moderate activation, whereas perivascular inflammatory cells and lymphatic follicle characterize the inflammatory component. Hematoxylin and eosin staining. Scale bars: 50 μm.

To further explore synovitis, we specified the cellular contributors by the use of immunohistochemistry. The small number of synovial CD4-positive lymphocytes found in the samples of both study groups, consisting of HHV-7 PCR+ and HHV-7 PCR−, was in line with low-grade synovitis (Figure 4a). Even in the absence of severe synovial inflammation, a statistically significant difference between the number of CD4-positive cells in HHV-7 PCR+ and HHV-7 PCR- groups was confirmed (Figure 4b).

To better assess the local expression of the pro-inflammatory marker TNF, we compared the numbers of TNF-positive cells in HHV-7 PCR+ and HHV-7 PCR− OA, and found no statistically significant differences between the groups. Similarly, no correlation was established when plasma cytokine levels were compared to the data depicting immunohistochemistry findings. Furthermore, we used the Wilcoxon matched-pairs signed rank test to compute the matched pairs, TNF-positive cell number, and HHV-7-positive cell number, submitting synovial samples of the HHV-7 PCR+ group to the test (Figure 4c). Simultaneously, in the HHV-7 PCR+ group, changes in the TNF-positive and HHV-7-positive cell count had a significant, positive correlation (*r* = 0.593, *p* = 0.0453) when assessed by Spearman's rank correlation (Figure 4d).

**Figure 4.** Assessment of synovial CD4-positive lymphocytes and TNF-positive cells in both study groups—HHV-7 PCR+ and HHV-7 PCR−—and HHV-7-positive cells in the HHV-7 PCR+ group. (**a**) The absence of synovitis graded as 0–1, and the presence of a small number of synovial CD4-positive lymphocytes found in the samples of both study groups—HHV-7 PCR+ and HHV-7 PCR−—consistent with low-grade synovitis (grade "2"). Grade "2" synovitis presented with a higher number of CD4-positive lymphocytes when compared to the lower grades. I.I. Krenn—inflammatory infiltration as a substantial part of the Krenn score (inflammatory infiltration, cellular hyperplasia of the lining layer, and cellular density of the sublining layer summed up to provide the Krenn score). Each dot represents a single data point. (**b**) The median number of CD4-positive cells per visual field in the synovial samples obtained from the HHV-7 PCR+ group is significantly higher than that in the HHV-7 PCR− group. Asterisks represent the significance level (\*\*\*\* *p* < 0.0001). (**c**) Median numbers assessed for immunohistochemically confirmed that positive cells are plotted for TNF and HHV-7 antigens. WMP—Wilcoxon matched-pairs signed rank test. The asterisk represents the significance level (\* *p* < 0.05). Each dot represents a single data point; blue dots represent HHV-7+ cells, and violet squares represent TNF+ cells. (**d**) Correlation of the median number of HHV-7+ and TNF+ cells expressed per visual field and detected in the samples of the HHV-7 PCR+ group; *r* = 0.593, *p* = 0.0453. An increase in the number of HHV-7+ cells reveals the elevation in the number of TNF+ cells. (**e**) CD4 immunohistochemistry. A representative image demonstrating T-lymphocytes decorated by the anti-CD4 antibody and recognized by the presence of brown reaction products in a follicle-like lymphocytic inflammatory infiltrate found in the synovial sample of HHV-7 PCR+ subjects. Scale bar 20 μm. (**f**) Through the use of TNF immunohistochemistry, the synovial lining presents TNF-positive cells interspersed by TNF-negative cells, whereas the sublining demonstrates mostly perivascular positivity observed in a sample of HHV-7 PCR+ patients. Scale bar: 50 μm.

When submitting the synovial samples obtained from both study groups for microscopical analysis, the lymphocytes and plasma cells colonizing the sublining layer demonstrated either diffusely scattered patterns of distribution, or compact and mostly perivascular patterns. Often, the presence of small follicle-like lymphocytic inflammatory infiltrates was confirmed (Figure 4e). Simultaneously, when assessed immunohistochemically, TNF-positive cells were distributed across the synovial lining and sublining and often demonstrated perivascular localization (Figure 4f).

To better recognize and estimate residential cells, synovial macrophages, and their role in the production of pro-inflammatory cytokines, we labeled cells with the anti-CD68 antibody. Furthermore, we compared the presence and number of CD68-positive and CD4-positive cells. The distribution of inflammatory cells bearing CD68 and CD4 labeling varied in both HHV-7 PCR+ and HHV-7 PCR− groups; however, the difference was not statistically significant (Figure 5a). Opposingly, the distribution of synovial CD68-positive cells and CD4-positive cells differed to a greater extent when latent and active HHV-7 infection was referred (Figure 5c). Simultaneously, no correlation was established when CD68 immunohistochemistry data were compared to the results depicting plasma pro-inflammatory cytokine levels (TNF and IL-6). Under the microscope, CD68-positive cells presented in both synovial subdivisions, the lining and sublining layers were diffusely distributed in both HHV-7 PCR− (Figure 5b) and HHV-7 PCR+ (Figure 5d) groups, and more local patterns of distribution were acquired when contributing to follicle-like inflammatory infiltrates.

**Figure 5.** Comparison of the distribution of synovial CD68- and CD4-positive inflammatory cells by the use of statistics (**a**,**c**), and the microscopical assessment of CD68-positive cells (**b**,**d**). (**a**) The distribution of CD68- and CD4-positive cells in both HHV-7 PCR+ and HHV-7 PCR− groups: CD68-positive cells constituted almost two-thirds (70%) of inflammatory cells found in HHV-7 PCR− samples, whereas, for CD68- and CD4-positive cells, 60 and 40%, respectively, were more equally distributed in HHV-7 PCR+. (**c**) CD68-positive cells represented a major part (71%) of inflammatory cells in latent HHV-7 infection, whereas these were opposingly distributed in active HHV-7 infection, with 68% and 32% of CD4- and CD68-positive cells, respectively. (**b**,**d**) CD68 immunohistochemistry; a representative image ((**b**), HHV-7 PCR− sample; (**d**), HHV-7 PCR+ sample) demonstrating CD68-positive cells decorated by the anti-CD68 antibody and developed brown reaction products in the synoviocytes of lining layer and the macrophages of sublining layer. The immunohistochemical decoration reflects the presence of lysosome-specific proteins involved in sorting in the trans-Golgi region, targeting to lysosomes, and fusion with the plasma membrane.

Fourteen of 19 HHV-7 PCR+ OA patients presented with synovium applicable for further immunohistochemical studies and tissue detection of the antigen. Furthermore, HHV-7-positive lymphocytes and macrophages were distinguished by their cytological appearance. To better explore the relationship between synovial cells bearing the HHV-7 antigen and TNF-producing cells, we computed the matched pairs' TNF-positive cell number and HHV-7-positive cell number, similar to Figure 4c,d but stratified into cellular types (Figure 6a,c). We determined a statistically significant, positive correlation (*r* = 0.6629, *p* = 0.0367) between HHV-7-positive lymphocytes and TNF-positive cells (Figure 6b), whereas a negative correlation (*r* = −0.6797, *p* = 0.0351) was found between HHV-7-positive endotheliocytes and TNF-positive cells (Figure 6d).

**Figure 6.** Assessment of synovial HHV-7-positive cells—lymphocytes (panel (**a**) and (**b**)) and vascular endothelial cells (panel (**c**) and (**d**)), and TNF-positive cells in the HHV-7 PCR+ group. (**a**) Median numbers assessed for immunohistochemically confirmed positive cells are plotted for the TNF and HHV-7 antigen. Each dot represents a single data point; blue dots represent TNF+ cells, and violet squares represent HHV-7+ lymphocytes. (**b**) Correlation between the median number of TNF+ cells and HHV-7+ lymphocytes expressed per visual field and detected in the samples of the HHV-7 PCR+ group; *r* = 0.6629, *p* = 0.0367. Each dot represents a single data point; blue dots represent TNF+ cells, and violet squares represent HHV-7+ lymphocytes. Correlations determined by Pearson's rank correlation test. (**c**) Median numbers assessed for immunohistochemically confirmed positive cells are plotted for the TNF and HHV-7 antigen. Each dot represents a single data point; blue dots represent TNF+ cells, and violet squares represent HHV-7+ endotheliocytes. (**d**) Correlation between the median number of TNF+ cells and HHV-7+ endotheliocytes expressed per visual field and detected in the samples of the HHV-7 PCR+ group; *r* = −0.6797, *p* = 0.0351. Each dot represents a single data point; blue dots represent TNF+ cells, and violet squares represent HHV-7+ endotheliocytes. Correlations were determined by the Spearman's rank correlation test.

Finally, HHV-7 immunohistochemistry data were compared for patients presenting with latent and active infection. When assessed quantitatively, HHV-7-positive lymphocytes, endotheliocytes, and macrophages constituted 58, 27, and 15% and 32, 67, and 1% of the cases of latent and active HHV-7 infection, respectively (Figure 7a,b). The immunohistochemical estimation of synovial HHV-7-positive cells applied for the HHV-7 PCR+ group demonstrated that cells labeled with the anti-HHV-7 antibody were localized in the sublining layer, both peri- and intravascularly (Figure 7c,d), and in the lining layer

(Figure 7e,f). Moreover, some endothelial cells constituting the internal lining of blood vessels found in the synovial membrane stroma were positively stained with the anti-HHV-7 antibody. Furthermore, evidence of the presence of HHV-7 expression in synovial tissue correlated with nPCR data.

**Figure 7.** The assessment of the HHV-7 antigen in the synovial membrane. (**a**,**b**) Frequencies of HHV-7-positive lymphocytes, vascular endotheliocytes, and macrophages based on immunohistochemistry data and compared for patients presenting with latent (**a**) and active infection (**b**). When assessed quantitatively, HHV-7-positive lymphocytes constituted a significant cellular fraction affected by the virus in latent HHV-7 infection and reached 58%. In contrast, vascular endotheliocytes forming the innermost layer of vascular beds appeared the most vulnerable in active HHV-7 infection, demonstrating a 67% involvement. There was a significant difference found between the distribution of HHV-7-positive endotheliocytes estimated for latent and active HHV-7 infections (*p*=0.028). (**c**) Immunohistochemical detection of the HHV-7 antigen in the case of latent HHV-7 infection. Intravascular HHV-7-positive cells recognized by brown coloration (blue arrowheads) localized in the lumen of congested blood vessels. Cellular nuclei counterstained with Mayer's hematoxylin (blue). Scale bar: 20 μm. (**d**) Numerous perivascular (blue arrow) and vascular endothelial HHV-7-positive cells localized in the sublining layer. Cellular nuclei counterstained with Mayer's hematoxylin (blue). Scale bar: 20 μm. Detection of the tegument protein pp85 of HHV-7 by immunofluorescence (HHV-7-immunopositive products, green), confocal microscopy; 1% toluidine blue was added to the fluorophore, and it resulted in near infra-red fluorescence in the cellular cytoplasmic compartment. Green arrows indicate the presence of viral protein at the top of synovial macrophages (**e**) and within the cell cytoplasm (**f**). Scale bar: 10 μm.
