*4.3. Nested Polymerase Chain Reaction*

A nested polymerase chain reaction (nPCR) was used for the qualitative detection of the viral genomic sequence in DNA isolated from whole blood, synovial membrane tissue samples (a marker for persistent latent infection), and cell-free blood plasma (a marker of active infection). Total DNA was isolated from WPB and synovial membrane tissue samples using the standard phenol-chloroform extraction. The QIA amp DNA Blood Mini Kit (Qiagen, Hilden, Germany) was used to extract DNA from plasma. The concentration of extracted DNA was measured spectrophotometrically (Nanodrop ND-1000 Spectrophotometer, Thermo Fisher Scientific, Waltham, MA, USA). To assure the quality of the whole blood, cell-free blood plasma, and synovial tissue DNA, as well as to exclude the contamination of plasma DNA by cellular DNA debris, a β-globin PCR was carried out using a polymerase chain reaction (PCR) (C1000 Touch Thermal Cycler, BioRad, Hercules, CA, USA). One microgram of whole blood and synovial tissue DNA, as well as 10 μL of plasma DNA, were subjected to nPCR with the HHV-7-specific primer, as described previously [71]. Positive (HHV-7 genomic DNA; ABI, Columbia, MD, USA) and negative controls (DNA obtained from practically healthy HHV-7-negative donors and a reaction without template DNA), as well as water controls, were included in each experiment. In the experiments, the sensitivity of HHV-7-specific primers corresponded to one copy of HHV-7 per reaction [72].

#### *4.4. Light Microscopy and Immunohistochemistry*

Synovial membrane tissue specimens (*n* = 54) were obtained from all OA patients undergoing joint replacement surgery. Two series of histological sections of 4–5 μm were cut from 10% formalin-fixed, paraffin-embedded tissue samples and mounted on SuperFrost Plus slides (Germany Menzel GmbH, Braunschweig, Germany) for histopathological and immunohistochemical evaluation. Before immunostaining, deparaffinization and hydration were conducted in xylene and graded alcohol to distilled water. During hydration, a 5 min blocking process for endogenous peroxidase was conducted with 0.3% (*v*/*v*) H2O2 in 95% methanol. Heat-induced epitope retrieval was accomplished with the sections immersed in 10 mM sodium citrate buffer, pH 6.0, at 96–98 ◦C for 5 min in a vapor lock.

Immunohistochemistry was performed conventionally using a monoclonal anti-HHV-7 antibody (Advanced Biotechnologies, Columbia, MD, USA, 1:500) raised against the tegument protein pp85 of HHV-7 [73,74]; the polyclonal rabbit anti-human TNF antibody (Biorbyt, Cambridge, UK, 1:100), which labels a certain peptide of human TNF [75]; monoclonal mouse anti-human CD68 (DacoCytomation, Glostrup, Denmark, clone PG-M1, 1:50), which labels monocytes/macrophages via the recognition of lysosome proteins,; and monoclonal rabbit anti-human CD4 (Cell Marque, Rocklin, CA, USA, SP35, 1:100), which recognizes a 55 kD glycoprotein expressed on the cell surface of T-helper/regulatory T-cells.

The amplification of the primary antibody and visualization of reaction products were performed by applying the HiDef Detection HRP Polymer system and diaminobenzidine tetrahydrochloride substrate kit (Cell Marque, Rocklin, CA, USA). The sections were counterstained with Mayer's hematoxylin, washed, mounted, and covered with coverslips. Immunohistochemical controls included the omission of the primary antibody. Sections were photographed by a Leitz DMRB bright-field microscope using a DFC 450C digital camera or scanned with a Glissando Slide Scanner (Objective Imaging Ltd., Cambridge, UK) with a 10×, 20×, and 40× objective. Reproducible measurements of tissue markers were obtained using the, Aperio ImageScope program v12.2.2.5015, Leica Biosystems Imaging, Vista, CA, USA and images were processed with the ImageJ program (National Institute of Health, Bethesda, MD, USA). Assessment of the histopathology and immunostaining was performed by two independent observers blinded to clinicopathological data.

Cells that were labeled with the anti-HHV-7, anti-TNF, anti-CD68, and anti-CD4 antibody and displayed brown reaction products were considered as immunopositive. The total number of immunopositive cells appearing within the microscopic field, depicting a certain synovial region, was estimated quantitatively in 10 randomly selected visual fields of each sample (magnification 400×).

Additionally, to better visualize the cellular distribution and localization of the HHV-7 antigen, the synovial tissue specimens were processed for fluorescent immunohistochemical staining and confocal microscopy. The sections that immunoreacted with the primary antibody overnight at 4 ◦C were washed in PBS, followed by incubation in goat anti-mouse IgG-FITC: sc-2010 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA 1:300) as the secondary antibody. Then, sections were counterstained with 4',6-diamidino-2phenylindole (DAPI) (Thermo Fisher Scientific, Invitrogen, Renfrew, UK, 1:3,000) and mounted in Prolong Gold with DAPI (Thermo Fisher Scientific). Imaging was performed using an Eclipse Ti-E confocal microscope (Nikon, Tokyo, Japan).

#### *4.5. Scoring of Synovitis by Krenn*

To define synovitis, involving inflammatory changes of the synovial membrane depicting intra-articular changes of a joint, we graded it using the scoring system introduced by Krenn and Morawietz [76]. Routinely (with hematoxylin and eosin), stained slides were used, and the lesions found in the synovial membrane were assessed. The following histopathological features were evaluated and scored: The cellular hyperplasia of the lining layer; the cellular density of the sublining layer; and the presence of inflammatory infiltration: 0—absent, 1—mild, 2—moderate, and 3—strong. The sum obtained provided the synovitis score, which was interpreted as follows: 0–1, no synovitis; 2–4, low-grade synovitis; and 5–9, high-grade synovitis.

#### *4.6. Statistical Data Analysis*

To better interpret molecular virology, serology, histopathology, and immunohistochemistry data, statistical analyses were performed using The GraphPad Prism 8 demo version (GraphPad Software, La Jolla, CA, USA). The D'Agostino and Pearson, Anderson–Darling, and Shapiro–Wilk tests were used to evaluate whether the collected numerical data were normally distributed. If data were not normally distributed, we used nonparametric one-way ANOVA on ranks or Kruskal–Wallis test followed by the two-stage step-up method of Benjamini, Krieger, and Yekutieli as post hoc tests when comparing medians instead of means. The chi-square test was performed for categorical variables. Categorical parameters were expressed as frequencies and percentages. The results of the histopathological assessment of Krenn scores in the synovial membrane samples of the study groups are expressed as violin plots, the median, and the interquartile range (IQR) as dispersion characteristics. To compare numerical values between two groups, the nonparametric two-tailed Mann–Whitney U test was applied. In the case of paired group comparisons, the Wilcoxon matched-pairs signed rank test was used. Correlations between the numbers of immunopositive cells were determined using either parametric Pearson's or nonparametric Spearman's correlation analyses, depending on the data distribution. The correlations were considered as follows: 0.2 to 0.4—weak; 0.4 to 0.7—moderate; and 0.7 to 0.9—strong. A *p*-value of less than 0.05 (*p* < 0.05) was considered statistically significant.

**Author Contributions:** Conceptualization and design of research: V.G. and M.T.; formal analysis: V.G.; data curation: V.G., M.T., S.S. (Sofija Semenistaja), Z.N.-K.; and S.S. (Simons Svirskis); writing—original draft preparation: V.G.; writing—review and editing: V.G., M.T., S.S. (Sandra Skuja), Z.N.-K., and M.M.; visualization: V.G., S.S. (Sandra Skuja), M.T., and S.S. (Sofija Semenistaja); prepared figures: V.G., M.T., S.S. (Simons Svirski), and S.S. (Sandra Skuja). All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This study was supported by the Latvian Council of Science Grant Nr. Izp-2018/1-0149 and the National Research Programme Biomedicine for the Public Health (BIOMEDICINE), project 7.2. The authors would like to thank Anda Kadisa, Rheumatologist at Riga East University Hospital Clinic "Gailezers", Latvia, for advising on the diagnosis and recruitment of osteoarthritis patients.

**Conflicts of Interest:** The authors declare no conflict of interest.
