Human Herpesvirus 6A and 6B inhibit in vitro angiogenesis by induction of Human Leukocyte Antigen G

Abstract

We have previously reported that human herpesvirus 6 (HHV-6) infection of endothelial cells (ECs) induces the loss of angiogenic properties, through the expression of HHV-6 U94, possibly associated to the release of a soluble mediator. It is also known that the soluble isoform of HLA-G exhibits an anti-angiogenic function, important in implantation, transplantation and neoplastic development. In this study, we analyzed the expression of HLA-G in HHV-6 infected ECs, showing that both HHV-6A and HHV-6B infection induce a potent up-modulation of HLA-G, including both membrane and soluble isoforms. Interestingly, HHV-6A and HHV-6B induced different isoforms of HLA-G. The virus-induced increase of HLA-G was likely due to the expression of the U94 viral gene, that by itself was able to reproduce the effect of whole virus. The effect of U94 was mediated by human transcription factor ATF3, that induced HLA-G activation by recognizing a consensus sequence on its promoter. Virus-induced inhibition of ECs angiogenic ability directly correlated to HLA-G expression and release, and the addition of anti-HLA-G antibody restored the angiogenic properties of HHV6-infected ECs. The induction of HLA-G expression in ECs might represent an important mediator of HHV-6 induced effects.

HHV-6A Infection and Systemic Sclerosis: Clues of a Possible Association

“…Although a similar prevalence of anti-HHV-6A/B antibodies in the SSc group and healthy controls (76.8 vs. 75%), the prevalence and titer of anti-U94 antibodies were significantly higher in the SSc population (100%, median titer 1:400) compared to controls (46.7%, median titer 1:166) (p < 0.004), confirming previous similar observations in multiple sclerosis patients [45]. Furthermore, consistently with previous observations showing the ability of virus infection to inhibit angiogenesis by a U94-mediated sHLA-G production in endothelial cells [37], we detected increased levels of sHLA-G in HHV-6A/B positive SSc patients compared to negative ones (p < 0.0001). Notably, SSc patients who were both HHV-6A/B and HLA-G positive, also displayed the highest titer of anti-U94 antibodies (p < 0.05), supporting a possible role of HHV-6A/B infection in the disease progression.…”

“…Aiming to clarify this aspect, HUVEC were infected in vitro with HHV-6A or 6B, and analyzed for the expression of a panel of 84 fibrosis-associated factors. The results showed that HHV-6A and 6B infection induced the upmodulation of several factors, suggesting that infection per se may interfere with the normal functions of endothelial cells [11,37]. Although the TGFβ2 factor, recently recognized as a key factor in fibrosis induction, did not result upregulated by virus infection in HUVECs, other factors belonging to TGFβ superfamily were promoted in particular by HHV-6A, including GREM1 and INHBE, IL-4, IL-5, TNF, and MMP9.…”

“…Human umbilical vein endothelial cells (HUVEC; ATCC CRL-1730), obtained as previously described [11,37], were grown in EGM-2 medium (BioWhittaker, Walkersville, MD, USA) in collagen-coated plates (Biocoat Collagen, BD Biosciences, Bedford, MA, USA). All the experiments were performed on third to fifth passage HUVECs.…”

“…Since we have already observed that HHV-6A U94 can induce HLA-G in HUVEC cells, inhibiting angiogenesis [37], we analyzed HLA-G plasma levels in correlation with HHV-6A/B presence in SSc patients. As shown in Figure 5, we observed an increase in plasmatic HLA-G in HHV-6 positive patients, compared to HHV-6 negative ones (p < 0.0001).…”

“…Furthermore, the positivity for HLA-G and HHV-6 also correlated with the anti-U94 antibody titer (p < 0.05), as HLA-G/HHV-6 positive patients exhibited the highest anti-U94 IgG plasma levels (p < 0.05). Since we have already observed that HHV-6A U94 can induce HLA-G in HUVEC cells, inhibiting angiogenesis [37], we analyzed HLA-G plasma levels in correlation with HHV-6A/B presence in SSc patients. As shown in Figure 5, we observed an increase in plasmatic HLA-G in HHV-6 positive patients, compared to HHV-6 negative ones (p < 0.0001).…”