Lymphatic filariasis and onchocerciasis are parasitic helminth Neglected Tropical Diseases (NTDs) evoking severe morbidity. These filarial diseases are being prioritized for elimination. Despite encouraging results accomplished by mass drug administration programs over the last thirty years, the number of people infected with these filarial parasites remains significant in endemic regions (approximately 90 million combined) (1). The recent CoVID-19 pandemic significantly impacted on preventive chemotherapy treatment coverage and highlights the vulnerability of this long-term elimination strategy with preventative chemotherapy (2).

Filarial nematodes induce polarized host type 2 immune responses in experimental models which are associated with killing of developing larvae and induction of immunopathology in the lymphatics or ocular tissues (3, 4). Clinically, type-2 immune responses are also evident and associated with pathology, albeit as part of a more heterogeneous immune response including induction of regulatory immune networks and classical inflammation evoked by Wolbachia endobacteria and secondary microbial skin infections (5–7). Cellular mechanisms of type-2 anti-filarial responses include the proliferation of non-classical ‘alternatively activated’ polarized macrophages at the site of infection (8, 9), driving a profound chemotactic eosinophil recruitment from the blood (10–13) which in turn leads to granulocyte-dependent larvicidal effector response and granuloma formation (14–17). However, little is known on the pathogen sensing and sources of type-2 priming signals involved in the early stages of the infection. We have recently identified a clear IL-4/IL-13/IL-4Rα dependent mechanism for tissue macrophages proliferation and activation (13). However, we have also demonstrated that there are IL-4Rα-independent/IL-5- and CCR3-dependent pathways mechanisms that can also lead to eosinophil-dependent effector responses prior to patency (12), thus indicating that other early sources of Type 2 associated cytokines and ligands might be involved in this mechanism.

Innate lymphoid cells (ILCs) were first described as innate counterparts of T lymphocytes, being able to produce an array of effector cytokines but lacking adaptive antigen receptors (18, 19). Amongst these, type 2 ILCs (ILC2) are known for producing IL-4, -5, -9 and/or -13 in the context of helminth infection (20, 21) which promote CD4⁺ T-helper (Th)2 differentiation (22). In addition, these cells have also been reported to proliferate in sterile tissue sites following filarial nematode infection (20). Recent studies have shown using RAG2^-/-γc^-/- immunodeficient animals (lacking functional lymphocytes and innate lymphoid cells) that chronic susceptibility to certain filarial infections is dependent both on lack of adaptive immune function and ablation of the ILC compartment (23–25). Beyond ILC2s, NK cells, which belong to the ILC1 family, have also been shown to produce IL-4 and IL-5 cytokines against Brugia malayi microfilaria (26) and contribute to the production of IL-13 in the context of a Trichuris muris infection (27). Furthermore, NK cells depletion led to increased Litomosoides sigmodontis filarial worm load (28).

Our present study focused on the exploration of ST2⁺ ILC2, IL-12Rβ2⁺ ILC1 (non NK) and NKp46⁺ NK cell populations dynamics and protective immune functional roles in the context of the human lymphatic filaria, B. malayi, using experimental infections of immunocompetent, RAG2^-/- and RAG2^-/-γc^-/- immunodeficient animals and via antibody mediated cell depletions.

Infection with filarial nematodes remains a major cause of global morbidity via induction of immunopathology. Further, chronic filarial nematode infection modulates immune responses in the context of co-infection, allergy, and autoimmune disease (41), yet our understanding of the fundamental cellular mechanisms of immunity during infection is limited. We have previously demonstrated, through the use of RAG-deficient and SCID mice, a pivotal role for the innate compartment in the immunological control of filarial parasite infection (13, 25). However, the relative contributions made by specific innate cell populations is poorly understood. The data presented here provides the first evidence of a major role for NKp46⁺ NK cells in protection against B. malayi and O. ochengi infection in immunocompromised animals. NKp46⁺ cells expanded rapidly during the early stages of the infection and their depletion, via the use of γc^-/- animals and neutralizing antibodies, rendered mice susceptible to chronic infection. We also demonstrated that these cells further expand and exhibit an activated/memory-like phenotype associated with an augmented innate immune protection against B. malayi challenge infection in the absence of adaptive immunity.

There is a well-established role for NK cells in the defense against infection by viral, bacterial, and protozoan organisms (42). In these contexts, NK cells act as rapid first responders, with key roles in the production of cytokines and the direct lysis of cells. However, their role in defense against macroparasites is poorly defined. In previous studies investigating permissiveness of filarial infection in immune-deficient mouse strains, it was noted that depleting NK cells resulted in reduced B. malayi burden in C57BL/6-scid/scid mice (43). Complementary to this was the observation that enhancing NK cell activation, via injection with the TLR agonist poly(I:C), increased the permissiveness of NOD/LtSz-scid/scid mice (which normally show low NK cell activity and complement deficiency) to B. malayi infection (43). This suggested that the level of NK cell activity positively correlated with the permissiveness of the host to infection. However, in a study using BALB/c mice infected with L. sigmodontis, NK cells (identified as Dx5⁺CD3^-) expanded significantly from 35 days post infection in the pleural cavity and down-regulated the inhibitory receptors Ly6C, Ly6A and Ly6G2, suggesting they are functioning here as effector cells rather than immunoregulators (28). Furthermore, their depletion with anti-asialoGM antibody increased both adult and mf burdens indicating that NK cells have an anti-filarial effect. In addition, ex vivo culture of NK cells (identified as CD3^-CD56⁺) with B. malayi mf and L3 resulted in a contact-dependent increase in expression of the activation markers CD69 and CD71 and the pro-inflammatory cytokines IFNγ and TNFα (26). These data indicate that NK cells respond to filarial infection by homing to the site of infection and increasing their activation state in a manner that is detrimental to the survival of the parasite. This is consistent with our data showing that depletion of NKp46⁺ NK cells is sufficient to increase susceptible to B. malayi infection in the absence of adaptive immunity. We expanded these observations by demonstrating NKp46⁺ NK cells developed an enhanced activation phenotype/memory-like capacity within the peritoneal cavity which was associated with innate immunological resistance against subsequent challenge following anthelmintic drug clearance of the primary infection. Such a phenomenon of ‘innate resistance’ has been prior described for challenge infection with Onchocerca lienalis skin mf infection following microfilaricidal drug clearance in lymphopenic SCID mice (44). This apparent NKp46⁺ cell-intrinsic memory-like phenotype and its contribution to the resistance to filarial infection requires further interrogation, using, for instance, adoptive transfer and recall activation assays.

The mechanism by which NKp46⁺ NK cells mediate parasite reduction remains unclear. In previous publications we have demonstrated that eosinophils are larvicidal effector cells during B. malayi infection, and that their IL-5-/CCR3-dependent recruitment is essential for parasite expulsion in immuno-competent BALB/c mice (12, 13). It was observed that in susceptible immunocompromised RAG2^-/-γc^-/- mice, which lack a functional NK cell compartment as well as being deficient in T and B cells, the granulocyte population in the peritoneal cavity during L. sigmodontis infection skewed toward neutrophils when compared to immunocompetent mice in which eosinophils predominate (23). Since then models of adoptive immune cell transfer into RAG2IL-2Rγ-deficient C57BL/6 mice have been developed, allowing for further pin-pointing of cellular mechanisms of anti-filarial immunity (24). We observed that when NKp46⁺ NK cells were depleted using an anti-asialoGM1 antibody, eosinophil but also neutrophil recruitment to the peritoneal cavity was reduced. Thus, in the absence of an intact type-2 adaptive immune response, NKp46⁺ NK cells may facilitate filarial larval parasite killing by co-ordinating the recruitment and activation of eosinophil and/or neutrophil effector cells alongside with subsequent granuloma formation and extracellular traps release (17, 45, 46). However, with recent reports on the anti-asialoGM1 antibody also having off-target effects on in vivo basophil populations (47) and on the use of anti-NKp46 antibody resulting in blocked activation rather than cellular depletion (48), further investigations are needed to assess ILC activation in the context of these NK depletion studies. Regarding off-target effects of anti-asioloGM1 on basophils, we and others have demonstrated that eosinophils rather than basophils are the major effector cell-type mediating attrition of filarial infections in experimental models and that basophils play a negligible role in filarial immunity (12, 13, 49). Thus, it is doubtful that anti-asioloGM1 treatment is increasing susceptibility via and off-target depletion of this granulocyte population.

Beyond parasitic infection, there is a clear link between NK cell activity and regulation of eosinophilia during airway hypersensitivity reactions. In humans suffering from severe asthma, NK cell activation measured by expression of CD69 and NKG2D correlated strongly with blood eosinophilia (50). Consistent with a role for NK cells as drivers of the eosinophilic response, in mouse models of allergic inflammation sensitization challenge with ovalbumin (OVA) resulted in a concomitant increase in NK cells and eosinophils (51, 52). Further, when NK cells were depleted prior to sensitization, lung eosinophilia was impaired (51, 52). In vitro co-culture of NK cells and eosinophils demonstrated that activated NK cells are capable of triggering eosinophil degranulation (53). Interestingly, NK cells have also been shown to act as negative regulators of eosinophils, capable of triggering their apoptosis under in vitro co-culture conditions (50, 53). Further, depletion of NK cells using anti-asialoGM1 at the peak of OVA-induced airway inflammation delayed the resolution of lung eosinophilia (54). Thus, under type-2 inflammatory conditions, NK cells perform a dual function first recruiting eosinophils and then subsequently inducing their apoptosis to resolve inflammation. Therefore, it will be interesting to define the mode through which NKp46⁺ NK cells recruit eosinophils during filarial infection, whether this is direct, via the secretion of chemoattractant, or if they support the development of M2 macrophages known to be key recruiters of eosinophils during B. malayi infection (13). There is also some evidence that NK cells express IL-5 in response to contact with B. malayi mf (26). However, this data was derived from a single ex vivo co-culture model of NK cell-B. malayi mf interaction. Additionally, B. malayi L3 did not provoke an IL-5 response under the same conditions. Therefore, further work is required to determine whether NK cell-secreted IL-5 is a direct recruiter of eosinophils during in vivo infection.

Our data illustrated a lack of ST2⁺ ILC2 expansion in the peritoneal infection site following B. malayi infection, which was surprising given the predominant protective type-2 polarized immune response triggered in this filarial experimental infection model (12, 13). In murine models of small intestinal-dwelling nematodes, ILC2s are known to play a role in the priming of Th2 responses, functioning as potent producers of Th2 cytokines (55). Depletion of ILC2s results in a failure to expel Nippostrongylus brasiliensis (34, 56), and impairs IL-4-driven differentiation of Th2 CD4⁺ T cells following infection with Heligmosomoides bakeri (22). Further, priming of ILC2s in the gut, via infection with H. bakeri or T. spiralis, can trigger ILC2 migration to distal tissues such as the lung and enhances subsequent larval trapping during the lung migratory stage of N. brasiliensis (57, 58). However, the role of ILC2s during filarial nematode infection remains poorly understood. During Loa infection in humans, the frequencies of cKit⁺ILCs (ILC2s and ILC3s) and cKit⁺IL-13⁺ILCs (ILC2s) in peripheral blood were observed to increase (59). The cKit⁺ILCs expressed an array of TLRs and were capable of producing Th1, Th2, and Th17 cytokines upon stimulation by cytokine cocktail ex vivo. Interestingly, cKit⁺IL-13⁺ILC frequency did not correlate with the frequency of Th2 CD4⁺T cells suggesting that ILC2s might not be required for sustaining peripheral Th2 responses (59). However, the authors derived samples from already chronically infected individuals and therefore this data does not rule out a role for ILC2s in the early induction of Th2 CD4⁺T cell responses. Despite ST2 being expressed by a subset of ILC2s only (18), during chronic infection of L. sigmodontis infection in Balb/c mice, efficient clearance of L. sigmodontis microfilariae is ST2 (IL-33R)-dependent, which might implicate a role for ST2-dependent ILC2 expansion in the control of filarial microfilaremias (60). It has been demonstrated that IL-5 producing ILC2s expanded specifically in the pleural cavity infection site of L. sigmodontis, whereas in secondary lymphoid tissue no change in ILC2 frequency is observed (20). Further, the frequency of ILC2s in the blood reduces during L. sigmodontis infection (20), indicating that an increase in ILC2 frequency in the pleural cavity may be a consequence of recruitment from peripheral tissues. The ILC2 expansion following L. sigmodontis but not B. malayi infection in serous cavities might be explained by the L. sigmodontis larvae migration through the lymphatics and pulmonary blood vessels before the thoracic cavity invasion (61). Conversely, in our studies, B. malayi larvae are introduced directly into the peritoneal cavity. Considerable damages to the endothelium and lung parenchyma are evidenced as part of the L. sigmodontis larval migration route. Damaged tissues caused by nematode invasion are known to trigger a variety of ‘alarmin’ release from stromal cells, such as IL-33, which are key triggers of robust ILC2 expansion (62).

Consistent with our data however, ILC2s are not ubiquitously required for immune control of all nematode parasites. Whilst ILC2s respond strongly to small intestinal nematodes (22, 34, 56) they fail to expand during infection with the cecal dwelling Trichuris muris and their depletion has no effect on parasite expulsion (63). Thus, it seems that the role of ILC2s in nematode infection is highly contextual and further work is required to establish causality between ILC2 activation and immunity to filarial nematodes, especially considering the diversity of clinically and veterinary important filarial nematodes and variability in their migration routes and adult niches of parasitism.

The present study implicates a functional role for NKp46⁺ NK cells in innate anti-filarial immune-protection in the absence of adaptive immunity. This mechanism may be relevant in situations whereby adaptive immunity is impaired, anergized or actively suppressed (for instance genetic variability or chronic co-infection situations in humans). Further, our data provides a mechanistic rationale why additional common gamma-chain deficiency is a requirement for chronic susceptibility in lymphopenic mouse models of human/veterinary filarial pathogens.

Follow-up investigations are needed to demonstrate their functional interplay with effector cells such as eosinophils and the full repertoire of memory-like signals harbored by NKp46⁺ NK cells following a secondary infection which may exert a degree of innate-memory protection from challenge. Research on NK cell biology and plasticity is progressing and reveals a wider scope of NK cell functional fates than initially thought (18, 64). Selecting the relevant phenotypic markers will be therefore crucial to allow a comprehensive scrutinization of the relevant NK subsets and their dynamics following filarial primary or secondary infection.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by the Liverpool School of Tropical Medicine ethical committee, the University of Liverpool ethical committee and the REFOTDE Animal Care Committee.

Conceived and designed the analysis: NP, SW, MT, and JT. Collected the data: NP, JF-S, AM, VC, BN, HS, JA, and AS Contributed data or analysis tools: NP, JF-S, SC, AM, HS, SW, and JT. Performed the analysis: NP and JF-S. Wrote the paper: NP, JF-S, SC, and JT. All authors contributed to the article and approved the submitted version.

This work was supported by a Bill and Melinda Gates Foundations Grand Challenges Explorations grant (OPP1119043) to JT, SW, and MT, and a Medical Research Council New Investigator Research Grant (MR/L018756/1) to JT. The funding bodies had no roles in the design of the study and collection, analysis, and interpretation of data. This work was supported, in whole or in part, by the Bill & Melinda Gates Foundation [OPP1119043]. Under the grant conditions of the Foundation, a Creative Commons Attribution 4.0 Generic License has already been assigned to the Author Accepted Manuscript version that might arise from this submission.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.