Discussion
H. polygyrus infection reduces RSV infection peak viral load, secondary immuno-pathology, and associated lung function impairment6. We now describe how this helminth infection induces enhanced monopoiesis in the bone marrow, elevated circulatory monocytes and increased recruitment of monocyte derived mononuclear phagocytes to the lung. Using monocyte ablation and transfer we show the functional importance of circulatory monocyte derived cells in the H. polygyrus -induced anti-viral effect against RSV. Elevated numbers of circulatory leukocytes over the first two weeks ofH. polygyrus infection, including lymphocytes, neutrophils and monocytes have previously been described but only in C57BL/6 mice29,30. The correlative observations of bone marrow myelopoiesis to drive this expansion in circulatory monocytes, and the subsequent elevation of lung mononuclear phagocytes are novel to our study. We expect that the mononuclear phagocyte population in the lung is directly derived from circulatory monocytes. This is supported by the circulatory increase being observed from 7 dpi, whereas lung mononuclear phagocytes remain unchanged until 10 dpi. The absence of the increase in lung mononuclear phagocytes when circulatory monocytes are ablated supports this interpretation. However, due to the potential for persistence of CCR2 on tissue mononuclear phagocytes31we cannot discount the possibility that we directly ablated cells in the lung as well.
A growing body of work has assessed the interplay between parasitic and viral infections32. H. polygyrus infection has been previously described to suppress influenza A (IAV) infections although the mechanism was undetermined33.Trichinella spiralis infection improves disease outcome to IAV in a gut damage dependent mechanism, but this was through suppression of secondary inflammation, rather than reduction in viral titres and no alteration in the mononuclear phagocyte compartment was observed34. Schistosoma mansoni (S. mansoni ) infection is protective against secondary infection with IAV and in pneumonia virus of mice (PVM) infection it gave a small reduction in peak PVM load35. S. mansoni larvae pass through the lung in mice so their effect on respiratory viral infection is likely due to local inflammatory responses rather than to the systemic response observed with the strictly enteric H. polygyrus . A mononuclear phagocyte response was not reported but others have described S. mansoni to drive recruitment of alternatively activated macrophages36. Their local immunosuppressive effect is often detrimental to viral infection control, with bothH. polygyrus and S. mansoni infection reactivating latent murine γ-herpesvirus infections by suppressing the antiviral IFN response37.
The promotion of alternative activated macrophages in helminth infection has also been described in tissues peripheral to the gut. H. polygyrus induces heart macrophages with a strong type 2 polarisation which appear at 28 dpi38. The absence of elevated blood monocytes at this late time point, and the very small increase in the number of monocytes and mononuclear phagocytes expressing the M2 marker CD301 (predominantly regulatory macrophages, data not shown) at 10 dpi in our study, further indicate that different mechanisms are responsible for the early induction of lung mononuclear phagocytes afterH. polygyrus infection and a later type-2 immunity driven expansion of alternatively activated macrophages.
An active factor driving the early H. polygyrus effects observed in our study has not yet been identified but the fact that both the antiviral effect and the expansion of both blood monocytes and lung mononuclear phagocytes depend on IFNAR signalling likely implies a role for IFN-Is either as direct effectors or stimuli for a secondary signal. The monocytic response to IFN-I signalling is key to myeloid responses in many other infections. IFNAR-deficient mice have been used widely in SARS-CoV-2 models where this disruption impairs recruitment of Ly6C+ monocytes to the lung39,40. In other inflammatory conditions this may be due to decreased turnover to Ly6C- monocytes, but the overall increase in total monocyte and mononuclear phagocyte numbers in H. polygyrusinfection suggests other additional effects14. IFNAR signalling has also been linked to mononuclear phagocyte function in IAV infection41,42 and to inflammatory monocyte accumulation in mucosal herpes simplex virus infection43. IFN-I signalling is also key to the accumulation of inflammatory Ly6C+ mononuclear phagocytes in the lung associated with older age, with a phenotype similar to that seen in viral infections such as IAV44. In PVM infection IFNAR signalling has no effect on monocytic cells but is required for the induction of inflammatory conventional dendritic cells, indicating a likely infection specificity of myeloid IFN-I signalling requirements45. However, dendritic cells express CD11c and may contribute to the CD11c+ macrophages we have found to expand afterH. polygyrus infection. Indeed the ongoing difficulty in characterising the mononuclear phagocyte sub-compartments in murine pulmonary inflammation (recently reviewed46,47) due to shared expression of the key markers used for steady state characterisation is also a limitation in this study. It is important to note that similar to the work in our study the use of global IFNAR knockouts in the studies above means that despite the usually high expression of IFNAR on monocytes, their response may be dependent on secondary cells.
Effects of IFN-Is on blood monocytes are less well studied including inIfnar1 deficiency. The historical characterisations of IFNAR-deficient mice showed elevated circulatory monocytes, counter to our model, but the increased risk of infection in IFNAR-deficient mice, at a time before modern infection control measures were available, may have contributed to this effect20. In human peripheral blood mononuclear cells (PBMCs) classical monocytes are more responsive to IFN-Is than non-classical monocytes due to differential abundance of IFNAR48 and culture of PBMCs with IFN-Is will alter their phenotype49-51. An expansion of circulatory monocytes and lung CD64+ cells, similar to our findings, was observed in idiopathic pulmonary fibrosis patients and associated with increased circulatory IFN-Is52. Exogenous treatment with IFNα for two to four weeks in asthma patients also increased the numbers of circulating monocytes and increased their antigen presenting phenotype53.
Clearly, from the array of responses described above the context of IFN-I signalling is crucial. The outcome of IFN-I stimulation is modulated by concomitant signalling/environmental factors that certainly also play a role in H. polygyrus responses54. One such environmental factor is the microbiome and it’s metabolic products, the presence of which is essential for the H. polygyrusantiviral effect6. While we were unable to test the impact of the microbiome in the present study, interactions between IFN-I production and microbiome components have been well described55,56 and the microbiome is known to be modulated in helminth infections including H. polygyrusinfection57,58.
The apparent lack of dependency on IFNAR signalling for bone marrow monopoiesis in this model suggests alternative signalling requirements. Interferon treatments have been described to affect lymphopoiesis59, IFN-I suppresses neutrophil differentiation60,61, and IFNγ promotes myelopoiesis62,63. Parasitic modulation of the bone marrow has also been described, with Trichuris muris andTrichuris gondii infections modulating haematopoiesis in an IFNγ dependent manner64,65. There is also substantial evidence that interferon regulatory factor 5 (IRF5) is a key regulator of hematopoietic development, particularly in myeloid lineages66,67. H. polygyrus has been recently described to induce IFN-γ (IFN-II) early in infection68 which may be responsible for the initial induction of monopoiesis. The deeply interlinked feedback loops between IFN-I and IFN-II signalling may explain why we see a reduced effect on monocytosis in the bone marrow69.
The mechanism of the anti-viral effect of H. polygyrus -induced mononuclear phagocytes is as yet unclear. In RSV infection mononuclear phagocytes are known to produce TNFα which is important for control of viral expansion, providing a possible mechanism13 but their direct effects on viral load were not assessed. Our findings directly demonstrate, for the first time, that elevating mononuclear phagocyte numbers in the lung prior to RSV infection can reduce viral load. Monocytes/macrophages are also known to produce IFNβ, viperin and OAS in response to viral exposure, thus the expansion of mononuclear phagocytes in the lung may explain the previously observed increase in these antiviral factors in whole lung samples during H. polygyrusinfection19,70-73.
In conclusion, we show that intestinal helminth infection can induce transient systemic monocytosis and increased numbers of mononuclear phagocytes within the lung that are essential to reducing the burden of viral respiratory infection. Due to the IFNAR dependency of this expansion in the blood and lung we hypothesise that helminth induced IFN-Is or secondary signalling products play a key role in these effects. Future work will be required to identify the specific factors that act upon monocytes to drives these effects and to further characterise the specific anti-viral mechanisms of these cells in RSV infection.