The study, led by a team investigating the links between SARS-CoV-2 infection and hematologic abnormalities, was built around a central and stubborn question: COVID-19 is frequently accompanied by changes in blood counts, inflammatory disturbances, and signs of altered hematopoiesis, yet it has remained unclear whether the earliest hematopoietic compartments and the stromal cells that support them are directly touched by the virus or merely respond to it from a distance. To address this, the researchers designed in vitro exposure systems that allowed them to compare how adult mesenchymal stromal cells from two anatomical sources, and umbilical cord blood CD34-positive hematopoietic stem and progenitor cells, responded to wild-type SARS-CoV-2, the Delta variant, and the Omicron variant.

Mesenchymal stromal cells are a natural place to look for such interactions. They form part of the microenvironment — the niche — in which hematopoietic stem cells reside, divide, and commit to lineage fates. The team examined both adipose-derived mesenchymal stromal cells, obtained from fat tissue, and bone marrow-derived mesenchymal stromal cells. When these cultures were exposed to the three viral variants, the researchers looked for the viral nucleocapsid protein, a reliable marker of viral presence inside cells. What they found was a striking pattern of source- and variant-dependence: the proportion of cells showing nucleocapsid positivity varied depending both on where the stromal cells came from and on which viral variant was used, and it was highest with Omicron in the adipose-derived cells. Notably, however, infectious virus in the culture supernatants was undetectable, indicating that the stromal cells carried viral components without supporting a productive, replicative infection.

Why would Omicron preferentially leave its signature in adipose-derived stromal cells? A candidate emerged from a comparison of surface receptor expression. The adipose-derived cells expressed substantially higher levels of neuropilin-1, a host factor that has been implicated in SARS-CoV-2 entry, than their bone marrow-derived counterparts. Neuropilin-1 is known to bind the CendR motif, a polybasic sequence at the S1–S2 boundary of the viral spike protein, and earlier work had suggested it could act as an alternative or auxiliary entry route, particularly relevant to variants carrying characteristic polybasic cleavage-site features. To test whether neuropilin-1 mattered functionally, the researchers knocked the gene down in adipose-derived stromal cells. The result was telling: reducing neuropilin-1 levels diminished the Omicron-associated nucleocapsid signal, implicating an NRP1-linked entry process in the variant’s engagement of these cells. This connects the observed pattern of susceptibility to a defined molecular mechanism rather than leaving it as a mere correlation.

Beyond viral entry, the study examined how the stromal cells responded functionally to viral exposure. A particularly interesting finding involved indoleamine-2,3-dioxygenase 1, or IDO1, an immunoregulatory enzyme that catabolizes tryptophan along the kynurenine pathway and is known to modulate immune responses and contribute to immune escape and tolerance. SARS-CoV-2 exposure induced IDO1 in the stromal cells, with the most prominent induction occurring in the adipose-derived population. Because IDO1 activity can suppress T-cell responses and alter the local immune environment, its induction in niche-supporting cells suggests a mechanism by which viral encounter could subtly reconfigure the immunological tone of the stromal compartment, even in the absence of productive infection.

The most provocative part of the study concerns the hematopoietic side of the niche relationship. The researchers exposed umbilical cord blood CD34-positive hematopoietic stem and progenitor cells — the population containing the most primitive blood-forming cells — to the viral variants. With Omicron, several coordinated changes emerged. Nucleocapsid positivity appeared within the CD34-positive population and, critically, was enriched in the long-term hematopoietic stem cell fraction, the most primitive and self-renewing compartment. At the same time, the frequencies of primitive fractions were reduced, suggesting a shift in the composition of the population away from its most immature elements. Heat-inactivated virus did not produce these effects, indicating that active viral material, not merely the physical presence of viral particles, drove the response.

Functionally, the Omicron-exposed progenitors did not simply decline. Instead, they reorganized their output. Colony formation — the standard assay in which progenitors are plated in semisolid medium and give rise to clusters of mature blood cells — increased after Omicron exposure, and the cultures showed an accumulation of early erythroid progenitors. This combination of reduced primitive fractions, heightened colony-forming activity, and erythroid-skewed output is characteristic of stress-responsive hematopoiesis, the emergency mode of blood production in which the stem cell compartment is pushed toward differentiation to meet acute demands. The implication is that Omicron exposure can promote the remodeling of early blood formation toward a stress-response program, at least in this in vitro setting.

To understand the molecular underpinnings of these changes, the team performed transcriptional profiling using the NanoString platform. The analysis revealed activation of host-response programs and, importantly, alterations in cell-cycle and differentiation programs within the exposed cells. Such transcriptional shifts provide a coherent framework for the phenotypic observations: changes in cell-cycle gene expression could explain shifts in the balance between quiescence and proliferation, while altered differentiation programs align with the observed erythroid bias and increased colony output. Together, the profiling data and the functional assays support a model in which viral engagement of primitive hematopoietic cells triggers a coordinated transcriptional and functional remodeling, rather than a nonspecific toxic effect.

The authors are careful about what their data do and do not show. The exposure systems were in vitro, and infectious virus was not recovered from the stromal cultures, so the findings do not establish that mesenchymal stromal cells or hematopoietic stem cells support viral replication in the body. The cord blood experiments likewise demonstrate viral protein presence and transcriptional and functional changes rather than a productive infection. Nevertheless, the combination of results — variant-dependent nucleocapsid detection, a receptor-linked entry mechanism, immunomodulatory enzyme induction, and lineage-remodeling in primitive progenitors — indicates that Omicron can physically engage both stromal and early hematopoietic compartments in a biologically meaningful way. These observations offer a plausible mechanistic bridge to the hematologic abnormalities reported in COVID-19 patients and identify neuropilin-1 and the stress-hematopoiesis axis as candidate targets for future investigation into the long-term consequences of SARS-CoV-2 exposure on the blood system.

Subject of Research: Omicron SARS-CoV-2 interaction with mesenchymal stromal cells and cord blood CD34-positive hematopoietic stem and progenitor cells

Article Title: Omicron SARS-CoV-2 shows NRP1-linked susceptibility in adipose-derived MSCs and alters primitive cord blood CD34⁺ HSPCs

Article References: Bhukhai, K., Tanhuad, N., Srimorkun, P., Manopwisedjaroen, S., Panvongsa, W., Vacharathit, V., Kanjanasirirat, P., Thongsa-ad, U., Kongsomros, S., Sutjarit, N., Srisaowakarn, C., Tawonsawatruk, T., Wongtrakoongate, P., Boonmuen, N., Borwornpinyo, S., Anurathapan, U., Chairoungdua, A., Asavapanumas, N., Hongeng, S., & Thitithanyanont, A. (2026). Omicron SARS-CoV-2 shows NRP1-linked susceptibility in adipose-derived MSCs and alters primitive cord blood CD34⁺ HSPCs. PLOS One, 21(10), e0360439. https://doi.org/10.1371/journal.pone.0360439

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0360439

Keywords: SARS-CoV-2, Omicron, NRP1, mesenchymal stromal cells, CD34-positive HSPCs, hematopoiesis, cord blood, IDO1, nucleocapsid protein, erythroid progenitors, Delta variant, viral entry