*3.6. Changes in Activity of IFN-*α/β *and in Phagocytic Activity of Immune Cells of Immunized Mice*

Next, the biological activity of type I IFNs and the phagocytic activity of leukocytes in groups of the immunized mice were studied. Activity of type I IFNs was evaluated by measuring relative production of IFN-α/β in response to stimulation of leukocytes by the Newcastle disease virus. There was no statistically significant difference in the production of IFN-α/β by peripheral blood cells between the groups. Mouse splenocytes more actively produced type I IFNs; the highest production was found in group 1, whereas the smallest was in group 3 (Figure 6a).

**Figure 6.** The differences in the biological activity of type I IFNs and phagocytic activity of peripheral blood leukocytes and splenocytes from immunized mice. (**a**) The activity of type I IFNs was evaluated by the production of IFN-α/β in response to stimulation of leukocytes by the Newcastle disease virus in vitro and was expressed in IU/ml; (**b**) phagocytic activity of immune cells was determined by luminol-dependent chemiluminescence (CL) and was expressed as activation index—the ratio of the intensity of induced CL to the intensity of spontaneous CL. The values on each diagram are means ± SD of six measurements done in three independent experiments. \* *p* < 0.05 compared to control group 4.

Phagocytic activity between the groups was compared by quantification of the production of reactive oxygen species (ROS), determined by luminol-dependent chemiluminescence (CL). For analysis, we used a suspension of splenocytes and a leukocyte preparation from heparinized whole blood. The activation index (AI) was reduced in the blood and in splenocytes of group 3 mice (Figure 6b) due to an increase in the level of spontaneous CL and a decrease in induced CL. Spontaneous CL shows the level of ROS production by cells, and an excessive increase in CL could be associated with a hazardous effect of ROS towards the cells. Induced CL reflects a potential ability of cells to respond to stimuli. The inability of immune cells to stimulate CL (AI ≤ 1) can point to inhibition of the bactericidal activity of phagocytes upon immunization with a plasmid. In the remaining experimental groups, the AI values were >1 and did not differ from the control group.

#### **4. Discussion**

Advances in cell therapy in recent years are associated with the use of the immunosuppressive properties of MSCs in transplantology, oncology, and some other areas of medicine, although many issues remain unresolved [46]. Depending on the microenvironment, MSCs can exhibit both immunosuppressive and immunostimulating properties [47,48]. However, the mechanisms of stimulation of the immune response by MSCs remain vague and have been studied mostly in vitro in mixed leukocyte reactions (MLR) [47,49].

To date there are only two papers investigating immune response to MSCs that express viral proteins. The first one describes the cells that express human immunodeficiency virus (HIV) Gp120 [11]. The second, recently published by Bolhassani et al. (2019) showed immunization of mice by mMSC that express E7 protein of human papillomavirus (HPV E7 antigen) in a complex with small heat shock proteins leads to a strong T-cell immune response and to a partial protection of animals against HPV-induced tumors [50]. Our data present a first evidence that modified MSCs (mMSCs) expressing HCV proteins affect innate and adaptive immune responses in mice. The experimental data show that the administration of naïve or modified MSCs to mice increases both spontaneous and ConA-induced levels of lymphocyte proliferation. Assessment of the proliferative response to mitogens is one of the most universal tests to assess the lymphocyte function; a weak reaction indicates the failure of cellular immunity. The functional activity of activated lymphocytes (production of IFN-γ) increased. The biological activity of IFN-α/β increased, as established in the test with the inhibition of the cytopathogenic effect of the encephalomyocarditis virus. Type I IFNs are known to play an important immunoregulatory role in relation to both the innate and adaptive immune responses to viral infections. For example, IFN-α/β induces the cytotoxicity of NK cells and increases the expression of MHC class I and co-stimulating molecules on antigen-presenting cells (APCs) [51]. A comparison of the effectiveness of these parameters of the innate immune response to mMSC with the response to immunization with a DNA construct shows that in the latter case the immune cells induce IFN-α/β and IFN-γ at a lower level. When mice were immunized with DNA, the phagocytic activities of monocytes/macrophages, neutrophils and dendritic cells decreased as well. In contrast, the functions of phagocytic cells during immunization with MSCs and mMSCs remained at the level of healthy intact mice.

The adaptive immune response to mMSCs was also significantly higher than that to the plasmid. For instance, mMSC induced a humoral response to all viral proteins expressed in MSCs. Almost all antibodies to non-structural HCV proteins were of IgG2a isotype. Switching to the synthesis of IgG2a antibodies is controlled by the Th1 cellular component of the immune response: a correlation between the level of IgG2a, virus-neutralizing properties, and IFN-γ synthesis has been established [52,53]. The DNA construct also caused the formation of IgG2a antibodies, but their activity was less than that in response to modified MSCs. The most significant difference in antibody titers (40-fold) was found in response to HCV NS3 and NS4 proteins. It should be noted that in combination with the gene adjuvant, pcGM-CSF plasmid, the pcNS3-NS5B construct induced a more active formation of IgG2a antibodies [30]. The immunomodulatory orientation of MSCs with respect to the B-cell response has been suggested to depend on the level of stimulation with viral antigens: the weaker the signal, the greater the stimulating potencies of MSCs [54]. A high level of IL-6 produced by mMSC can stimulate an active humoral response to HCV: this cytokine has been shown to be necessary for differentiation of B cells and secretion of immunoglobulins [54,55].

Production of IgG1 antibodies against HCV was detected in groups immunized with the non-transfected MSCs; however, these antibodies were detected sporadically and with low activity. This may be caused by a cross-reaction between the recombinant proteins obtained in the bacterial system and the antibacterial antibodies in the blood sera of mice.

We characterized the cellular response by the proliferation of lymphocytes and their functional activity—secretion and intracellular content of IFN-γ. These methods are considered as the most informative by a majority of authors [17]. The cellular immune response to the modified MSC significantly exceeded the response to the plasmid pcNS3-NS5B in the proliferative response to the HCV sequence, as well as in the synthesis and secretion of IFN-γ. Differences in the signal intensity between groups 1 and 3 were 2.6–4.2 times in T-cell proliferation and ELISpot and up to 30 times in ELISA when determining the concentrations of produced IFN-γ. All non-structural HCV proteins elicited a cellular response; the maximum response was observed for NS5B. This protein is an RNA-dependent RNA polymerase, the key component of HCV replicase. NS5B is a target for direct-acting antivirals in the treatment of hepatitis C [56]; NS5B has the largest number of conserved T-cell epitopes that are important for vaccine design and induction of an effective immune response [57]. Thus, when immunizing with mMSCs, we achieve a functionally active T-cell response to several HCV proteins simultaneously, including different genotypes. This result is very important as an HCV vaccine should elicit multiantigenic, multigenotypic responses that should protect against challenge with the range of genotypes and subtypes circulating in the community [29]. We also hope that usage of more conserved viral proteins (i.e. nonstructural) will allow induction of a pangenotype response.

One of the mechanisms of the suppressor action of MSCs on the adaptive immune response in inflammation and cancer is believed to be their inhibition of maturation of antigen-presenting dendritic cells under the action of various soluble factors - TGF-β, IL-10, NO, and PD-1 [48]. We showed that when immunizing healthy animals in all experimental groups, the number of dendritic cells did not change compared to the control.

A very interesting fact is the data that immunization with MSCs and mMSC causes a twofold decrease in the number of MDSCs, a heterogeneous population of immature myeloid cells with a powerful suppressor potential. This phenomenon for immunization with MSCs is described for the first time. The role of MDSCs in viral infections has not been adequately studied [58]. In patients infected with HCV, an increase in the MDSC population is observed; these cells inhibit the proliferation of CD4<sup>+</sup> and CD8<sup>+</sup> lymphocytes, NK cells, and IFN-γ production [59,60]. Similar results have been obtained in the study of cells from patients with HIV and hepatitis B infections [61,62]. Most experimental data show that the administration of MSCs in oncological and autoimmune diseases results in MDSC accumulation and immunosuppression mediated by certain chemokines and growth factors [63–65]. On the other hand, when modeling cancer in mice, a dependence of the immunomodulatory "phenotype" of MSCs on the injection site was found: the simultaneous injection of MSCs with tumor cells led to immunosuppression, distal injection led to immunostimulation; the immune response was shown to correlate with a decrease in the proportion of MDSCs and T-regulatory cells (Treg) [66]. Thus, one of the mechanisms of stimulation of the innate and adaptive immune response to naïve and modified MSCs in our experiments may be the suppression of MDSCs.

During inflammation, the immunosuppressive properties of MSCs are manifested: they suppress both innate and adaptive immunity, weakening the maturation and ability to present antigens by dendritic cells, inducing the polarization of macrophages in the direction of the alternative phenotype, inhibiting the activation and proliferation of T and B lymphocytes and reducing the cytotoxicity of NK cells [48]. We administered MSCs to healthy mice. On the one hand, the biological properties of MSCs depend on the microenvironment; on the other hand, the immunomodulating effect of MSCs themselves is mediated by the secretion of various soluble factors. We compared the production of several cytokines by naïve and modified MSCs in vitro. It turned out that the expression of HCV proteins influenced the production of at least three pro-inflammatory cytokines - IFN-γ, IL-2, and IL-6. For example, after two weeks of cultivation of transfected cultures of MSCs in the presence of G-418, the concentrations of secreted cytokines IFN-γ and IL-2 decreased by 2–3.5 times, and that of IL-6 increased by eight times compared with MSCs. These stably transfected cells were injected into mice. Interestingly, the accumulation of IL-6 leads to the activation of the pro-inflammatory phenotype of the MSC population—MSC-1 [47] and promotes the formation of Th17 cells that activate the immune response [67]. Most likely, fluctuations in cytokine production are associated with the action of viral proteins on cell metabolism but not with changes in MSCs epigenetics because mesenchymal cells have been shown to maintain the genetic stability for at least seven to nine passages [68]. The spectrum of

cytokines that are produced by cells transfected with HCV genes has been previously found to change with time, depending on the type of cells and specific viral proteins [69–71].

The data on the effect of exogenous IFN-γ on MSCs are contradictory. Several authors have noted an increase in the antigen-presenting properties of MSCs as a result of IFN-γ pretreatment [72]. Other studies have shown that the "priming" of MSCs in vitro with IFN-γ, TNF-α, or IL-1β leads to the formation of the immunosuppressive phenotype MSC-2 [47,73,74]. Our results showed, for the first time, that after administration in mice, modified MSCs treated with IFN-γ cause a pronounced (tenfold) decrease in the cellular response. This means that an excessive concentration of pro-inflammatory cytokines does not stimulate, but instead inhibits the immune response to antigens presented by MSCs.

The major technique that is recommended for evaluation of T-cell response to novel HCV vaccines is quantification of IFN-γ production by ELISpot assay that shows activity of antiviral response [19,75,76]. Though we have not shown protection against HCV infection using mMSCs, such experiments could be performed in future using the respective models. So, we consider our results as a basis for subsequent preclinical (and clinical) studies of protective effect of mMSCs in future. Human mMSCs that express non-structural HCV proteins could be evaluated as a prophylactic vaccine that triggers a strong T-cell response. A growing trend in human MSC clinical trials is the use of allogenic and culture-expanded cells [1]. Application of mMSC to chronic hepatitis C patients may enhance therapeutic response to direct acting antivirals (DAA) via enhanced T-cell immune response that can clear the infected cells. Studies of a combined usage of various candidate HCV vaccines and antiviral agents including DAA are one of the current trends in the field (as can be exemplified by [77,78]. Despite the fact that the results do not always show enhanced clearance of the infection, this approach is still considered promising [19,79,80].
