Human Osteoblast-like Cells Are Permissive for Zika Virus Replication

To the Editor:

We read with great interest the letter by Roimicher, et al¹. The authors reported a case of a woman with a 4-year diagnosis of rheumatoid arthritis who achieved therapy-induced clinical remission 4 months before acquiring Zika virus (ZIKV) infection. The woman presented with fever, maculopapular rush, conjunctivitis, and arthralgia, and was tested for Chikungunya virus and ZIKV infection. The ZIKV genome was detected in blood and knee synovial fluid (SF). ZIKV viremia become negative after the end of clinical signs (Day 7); however, the patient experienced articular pain about 1 week later, and tested positive for ZIKV RNA in SF, while ZIKV viremia remained negative. These findings characterized for the first time a ZIKV infection localized within the joint of a patient, and raised a more general question about the ZIKV ability to directly target cells of the osteoarticular tissues, thus contributing to the establishment of joint-related symptoms.

Here we provide experimental observation that may support and strengthen the notion of the ability of ZIKV to productively infect cells belonging to the bone lineage. Specifically, we investigated the ability of ZIKV to infect and replicate in human osteoblast-like SV-40 T antigen immortalized cell line (HOBIT; kindly provided by Professor M.C. Re, University of Bologna). For this study, ethical approval was not required, in accordance with the policy of the Italian National Institute for Infectious Diseases L. Spallanzani (Istituto Di Ricovero e Cura a Carattere Scientifico).

HOBIT cells were exposed to either an Asian (ZIKV strain PRVABC59, Genbank KU501215) or an East African (ZIKV strain MR766, Genbank LC002520.1) ZIKV strain for 1 h at 37°C, at a multiplicity of infection (MOI) of 0.1 and 1 TCID₅₀/cell. A clear cytopathic effect was observed for both strains starting from 48 h postinfection (hpi). The production of infectious viral particles was monitored by back-titrating supernatants collected at 6, 24, 48, 72, and 96 hpi. Table 1 shows a steady increase of ZIKV titer in HOBIT infected with the 2 viral strains at both MOI, thus indicating that these cells are highly permissive to productive infection by ZIKV, releasing up to 10^7.9TCID₅₀/ml. The virus yield titer peaks slightly earlier and reaches higher levels with the African strain, suggesting possible differences in the replication kinetics.

Our data extend current knowledge about the broad ZIKV tissue and cell tropism².

Indeed, studies on both animal models and human cases reported the presence of ZIKV in skin fibroblasts, epidermal keratinocytes, dendritic cells³, human endometrial stromal cells⁴, and cells of the placenta^5,6. ZIKV also targets neuronal cell types^7,8 and cells of the reproductive tract, including spermatogonia, Sertoli and Leydig cells, and vaginal epithelium^2,9. This extensive tropism is paralleled by the detection of ZIKV in multiple body fluids, including tears, saliva, semen, cervical mucus, and urine^2,10.

Overall, the present study, together with the detection of ZIKV in SF¹, supports the hypothesis that joint tissue may be a possible reservoir of ZIKV replication, thus expanding the spectrum of ZIKV pathogenesis.