Anaphylatoxin signaling activates macrophages to control intracellular Rickettsia proliferation

ABSTRACT Pathogenic Rickettsia species proliferate within the cytoplasm of permissive host cells in vivo. The cytoplasm of these host cells is adequate to support the complex metabolic and physiological needs for Rickettsia growth. However, a dramatic host/pathogen interplay occurs when Rickettsia encounter innate immune cells, whereby the bacteria can proliferate as normal or the host can restrict bacterial growth. This interplay is most divergent within myeloid host cells, where intra- and extracellular factors can produce either successful Rickettsia parasitism or innate immune control of bacterial proliferation. With the prior knowledge that the mammalian complement system is activated during mammalian infection, we sought to determine if extracellular complement activation and anaphylatoxin signaling can modify the fate of Rickettsia within mononuclear host cells. Results indicate that supplementation of growth media with either C3a or C5a anaphylatoxin peptides is sufficient for many myeloid cells to control the proliferation of multiple different Rickettsia species. Chemical or genetic disruption of anaphylatoxin signaling or anaphylatoxin receptors eliminates complement-induced restriction of bacterial proliferation. Finally, anaphylatoxin signaling modifies macrophage physiology by inducing inflammatory phenotypes that ultimately control the intracellular proliferation of these pathogens. IMPORTANCE Pathogenic Rickettsia species are extremely dangerous bacteria that grow within the cytoplasm of host mammalian cells. In most cases, these bacteria are able to overpower the host cell and grow within the protected environment of the cytoplasm. However, a dramatic conflict occurs when Rickettsia encounter innate immune cells; the bacteria can “win” by taking over the host, or the bacteria can “lose” if the host cell efficiently fights the infection. This manuscript examines how the immune complement system is able to detect the presence of Rickettsia and alert nearby cells. Byproducts of complement activation called anaphylatoxins are signals that “activate” innate immune cells to mount an aggressive defensive strategy. This study enhances our collective understanding of the innate immune reaction to intracellular bacteria and will contribute to future efforts at controlling these dangerous infections.

Rickettsia species are Gram-negative obligate intracellular bacteria that proliferate within the cytosol of eukaryotic host cells (7).These organisms are maintained in nature through an endozoonotic cycle of parasitizing hematophagous arthropods and transmission to mammals during arthropod feeding (8,9).After transmission to mammalian hosts, Rickettsia species primarily invade and proliferate within cells of the endothelial system.While infected endothelial cells detect the intracellular bacteria and respond accordingly, Rickettsia species are able to sufficiently manipulate host cell signaling and innate immune responses to generate a permissive growth environment (10)(11)(12).As the disease progresses, immune recognition of Rickettsia and endothelial cell signaling contribute to the recruitment of innate leukocytes (13).Subsequent encounters between mononuclear cells and Rickettsia produce vastly different outcomes, with either Rickettsia invading into and successfully colonizing the monocyte or the monocyte effectively controlling bacterial proliferation with associated proinflammatory signaling (14)(15)(16)(17)(18)(19).
The complement system has traditionally been considered an arm of the innate immune system, whereby soluble proteins circulate through bodily fluids as inactive precursors waiting to be activated by innate immune sensors of infection (20,21).The ensuing complement activation can directly temper infections by generating lytic pores on the surface of the pathogen, by coating the particle in complement opsonins that enhance uptake by innate immune cells, or by generating massively proinflammatory anaphylatoxin peptides (21).The complement system is activated directly by Rickettsia in vitro and during mammalian infection in vivo (22)(23)(24)(25).However, the bacilli have evolved mechanisms to combat complement-mediated killing (26)(27)(28).While the antibacterial membrane attack complex and opsonization are dispensable for the immune response to Rickettsia infection, complete ablation of complement functionality drastically reduces the veracity of the immune response to infection (24).We therefore initiated this work with the knowledge that the complement system contributes to the effective immune response to infection but lacked knowledge of the molecular or cellular mechanisms of efficacy.
In the present work, we have examined the effects of anaphylatoxins and anaphyla toxin receptors on Rickettsia intracellular growth within myeloid cells.We have devel oped different in vivo and ex vivo approaches to evaluate macrophage responses to anaphylatoxin stimuli by evaluating cellular responses and bacterial survival.Our findings advance our understanding of the immune response to Rickettsia infection by establishing the role of the anaphylatoxin-anaphylatoxin receptor axis in controlling Rickettsia proliferation.

Anaphylatoxin-activated RAW264.7 murine macrophage cells limit intracellu lar growth of pathogenic Rickettsia
Previous studies have demonstrated that genetic ablation of the murine complement system results in increased sensitivity to Rickettsia australis infection, with mortality occurring prior to the development of a fulminant adaptive immune response (24,45).Therefore, Rickettsia-induced complement activation likely contributes to innate immune control of infection in vivo.Related molecular and cellular analyses of the direct interaction between pathogenic Rickettsia and the complement system have also indicated that these bacteria are not susceptible to either the antibacte rial membrane attack complex or complement receptor-mediated phagocytosis, thus indirectly implicating the anaphylatoxin/anaphylatoxin receptor system as the arm of the complement system that contributes to the control of Rickettsia infection (24).We therefore sought to assess the hypothesis that anaphylatoxin signaling is essential for restricting Rickettsia growth within host cells.
The most direct method for eliminating complement activity from murine serum is to genetically ablate the central complement component C3 (46).While complement activation is readily inducible in normal mouse serum (NMS) during Rickettsia infection, C3-deficient mouse serum (C3 −/− MS) is incapable of generating complement effector mechanisms, including forming the membrane attack complex, opsonizing bacteria, or, importantly, producing the proinflammatory anaphylatoxin peptides C3a and C5a (24).As shown in Fig. 1A and B, both C3a and C5a are produced when R. parkeri-infected RAW264.7 cells are cultured for 3 days in media containing NMS.However, neither C3a nor C5a is produced when the same cells are grown in media containing C3 −/− MS, demonstrating that C3 −/− MS does not stimulate C3a nor C5a anaphylatoxin production and therefore cannot induce signaling through the reciprocal anaphylatoxin receptors on mammalian cells.
The murine RAW264.7 cell line has previously been employed to investigate the hostpathogen interplay occurring within Rickettsia-infected macrophages (47,48).RAW264.7 cells express the anaphylatoxin receptors on their surface and are capable of responding to extracellular anaphylatoxins (42,(49)(50)(51)(52).To determine if complement influences Rickettsia proliferation in these macrophages, we infected RAW264.7 cells with the causative agent of Rickettsia parkeri rickettsiosis in the presence (NMS) or absence (C3 −/− MS) of functional complement.DNA was extracted at 2 and 3 days post infection to assess R. parkeri proliferation as determined by calculating the quantitative PCR (qPCR) ratio of Rickettsia DNA (sca1) to mouse DNA (actin).As shown in Fig. 1C, RAW264.7 cells cultured in media containing complement-active serum and complement-deficient C3 −/− serum did not produce differences in R. parkeri growth through 2 days of culture.However, there was significantly more Rickettsia at 3 days post infection when grown in the absence of complement as compared to complement-active NMS.It is also important to note that there is no statistical difference in the survival of RAW264.7 cells infected with R. parkeri and cultured in the presence of NMS or C3 −/− MS throughout the 3 days of culture (Fig. S1), so the changes in Rickettsia load are not associated with the death of the host cells.To validate this finding in other Rickettsia, we infected RAW264.7 cells with the causative agent of Rocky Mountain spotted fever, R. rickettsii, in the presence of comple ment-active NMS or complement-deficient C3 −/− MS.As shown in Fig. 1D, R. rickettsii also proliferated better in the absence of complement activation than in complement-active normal human serum (NHS).These results suggest that complement activation induces murine RAW264.7 macrophages to restrict the growth of Rickettsia.
To examine the potential that anaphylatoxin-independent complement activities were responsible for restricting Rickettsia growth, we performed R. parkeri infection in the absence of complement activation by using C3 −/− MS but supplementing the media with a physiologically relevant concentration of purified C3a and C5a peptides to reconstitute anaphylatoxin signaling (53).C3a and C5a are small polypeptides that act as potent inflammatory mediators targeting a broad spectrum of immune and nonimmune cells, including macrophages (54).R. parkeri and R. rickettsii were again cultured in RAW264.7 cells with C3 −/− MS supplemented with phosphate-buffered saline (PBS), C3a peptide, C5a peptide, or both C3a and C5a at 1 µg/mL.The addition of these peptides does not affect the survival of RAW264.7 cells (Fig. S2).Quantitative PCR of DNA extracted from the cells on day 3 demonstrates that anaphylatoxin supplementation significantly restricted R. parkeri and R. rickettsii growth as compared to permissive growth in C3 −/− MS (Fig. 1E  and F).These data directly implicate C3a and C5a peptides in enhancing the restriction of Rickettsia growth.The anaphylatoxin peptides C3a and C5a exert their inflammatory effects through interaction with specific anaphylatoxin receptors called C3aR, C5aR1, and C5aR2 (55).To determine if anaphylatoxin-induced restriction of Rickettsia growth is mediated by the anaphylatoxin-anaphylatoxin receptor axis, we cultivated RAW264.7 cells in complementactive NMS, complement-deficient C3 −/− MS, or complement-active NMS in the presence of the anaphylatoxin receptor inhibitors PMX53 and SB290170.SB290157 is an antago nist of C3aR (56), whereas PMX53 is a competitive antagonist of the C5aRs (57).After 3 days of growth in RAW264.7 cells, R. parkeri growth was restricted in complement-active media (NMS) as compared to complement-deficient media C3 −/− MS (Fig. 1G).However, in media with reconstituted complement activity (NMS), chemical inhibition of the anaphylatoxin receptors restores R. parkeri proliferation, demonstrating that restriction of Rickettsia growth in macrophages requires the anaphylatoxin receptors.Taken together, the data in Fig. 1 demonstrate that complement activity can restrict Rickettsia growth in murine RAW264.7 macrophages, that C3a or C5a is sufficient to decrease bacterial growth, and that signaling through the anaphylatoxin receptors is required for comple ment-mediated restriction of Rickettsia proliferation.

Anaphylatoxin-mediated reduction of Rickettsia growth in human cells
As multiple Rickettsia species are intracellular pathogens in humans, we sought to determine if the anaphylatoxin-anaphylatoxin receptor axis also supports the constraint of Rickettsia proliferation within human monocytic cells.To this end, we examined the influence of the anaphylatoxins on R. parkeri growth with human THP-1 cells.These macrophages are of particular relevance to Rickettsia as bacterial proliferation in these cells correlates with mammalian virulence (14,58,59), and there appears to be a dynamic interplay between the bacteria and host occurring within these cells (15,60).We employed normal human serum to drive THP-1 differentiation/adherence in the absence of exogenous THP-1 activators like PMA.Also, in the absence of the ability to genetically ablate complement activity as with C3 −/− mouse serum, we instead generated complement-deficient human serum through physical heat inactivation (heat-inactiva ted human sera, hiHS) or chemical inhibition with ethylenediaminetetraacetic acid (NHS + EDTA) (61,62).Notably, neither heat inactivation nor EDTA supplementation is toxic to THP-1 cells (Fig. S3 and S4).R. parkeri-infected THP-1 was cultured for 3 days in either NHS or hiHS.As shown in Fig. 2A, complement activity in the NHS reduces R. parkeri growth as compared to complement-inactivated hiHS.Subsequent chemical inhibition of complement activity (NHS + EDTA) also permitted greater proliferation in THP-1 cells than that of NHS (Fig. 2B).These data demonstrate that complement activity is capable of driving human THP-1 cells to limit the growth of Rickettsia.
To determine if anaphylatoxin stimulation is essential for complement-induced restriction of Rickettsia proliferation in THP-1 cells, we infected R. parkeri in the presence of human anaphylatoxin peptides.The complement system was nonfunctioning in hiHS, but the addition of anaphylatoxins is sufficient to decrease Rickettsia proliferation at 3 days post infection (Fig. 2C).Additionally, chemical inhibition of anaphylatoxin receptors eliminates the decreased Rickettsia proliferation induced by complement activation without affecting the health of the host cells (Fig. 2D and S5).Together, these data demonstrate that anaphylatoxin and anaphylatoxin receptor interactions restrict R. parkeri proliferation in human monocytic cells.
While monocytic cells are of particular interest for their differential interaction with pathogenic Rickettsia, these bacteria have a primary tropism for endothelial cells in vivo (13,63,64).To this end, we investigated the effect of complement activation on the growth of R. parkeri in EA.hy926 human endothelial cells.As shown in Fig. 2E, comple ment-active NHS is not sufficient to restrict R. parkeri growth in EA.hy926 at 3 days post infection as compared to complement-inhibited NHS + EDTA.This finding contrasts with the observed complement-mediated restriction of growth in murine and human monocytes (Fig. 1C and D; Fig. 2B).This endothelial versus monocyte dichotomy highlights the possibility that anaphyla toxin-activated monocytes could secrete factors to induce endothelial cells to restrict Rickettsia proliferation.This concept is conceivable because exogenously activated endothelial cells have been demonstrated to directly reduce Rickettsia growth (13,65,66).To assess the validity of the hypothesis that anaphylatoxin-treated monocytes can indirectly control Rickettsia growth, we set up an EA.hy926 endothelial + THP-1 macro phage co-culture.As shown in Fig. 2F, there is no significant difference in R. parkeri growth at 3 days post infection within EA.hy926 or EA.hy926 + THP-1 co-culture, so long as the infected cells are cultivated in media lacking functional complement.However, R. parkeri proliferation is reduced in EA.hy926 + THP-1 co-culture when propagated in complement-active NHS; this demonstrates that complement activation is not sufficient to restrict bacterial growth in endothelial cells, but the addition of complementactivated macrophages is sufficient to modulate the growth environment to decrease Rickettsia proliferation.

Anaphylatoxin-mediated restriction of R. australis proliferation in primary murine bone marrow macrophages
Having demonstrated a role for anaphylatoxins and anaphylatoxin receptors within immortalized murine and human macrophage cell lines, we sought to further examine these phenotypes within primary cells.To this end, we isolated tibial and humeral bone marrow macrophages (BMMs) from C57BL/6J mice.We deduced that it was important to employ R. australis in this model because this is the sole Rickettsia species capable of causing disease in the murine B6 background (45).Macrophage colony-stimulating factor (M-CSF)-differentiated C57BL/6J BMMs were infected with R. australis in media containing complement-active NMS or complement-deficient C3 −/− MS.While R. australis propagated equally well in media containing NMS or C3 −/− MS through day 2 of propagation, there was a significant restriction of R. australis growth at day 3 of infection within BMMs treated with NMS as compared to C3 −/− MS (Fig. 3A).Microscopic imaging with the nuclear fluorophore 4′,6-diamidino-2-phenylindole (DAPI) to illuminate BMM nuclei and an anti-Rickettsia antibody to visualize the bacteria showed dramatically fewer R. australis within BMMs cultivated in NMS than in C3 −/− MS at 3 days post infection (Fig. 3B).Quantification of fluorescent bacilli within 600 BMMs further demonstrates the reduced R. australis growth resulting from complement activation (Fig. 3C).
To investigate if complement-associated restriction of R. australis growth in BMM was due to anaphylatoxin activities, we propagated R. australis in BMM with complementdeficient media (C3 −/− MS) supplemented with anaphylatoxin peptides.As shown in Fig. 3D, supplementation of the complement-deficient media with C3a, C5a, or C3a + C5a anaphylatoxins resulted in decreased R. australis proliferation at day three post infection as compared to mock-treated BMM.This is further visualized by microscopic analysis of Rickettsia content in BMM (Fig. 3E) and quantified as the ratio of bacilli per host nucleus in 600 individual BMMs (Fig. 3F).Together, the data presented in Fig. 3 demonstrate that complement and, more specifically, anaphylatoxin activation are sufficient to restrict R. australis proliferation in B57B/6J BMMs.
The major advantage of querying the R. australis + C57BL/6J system is the availability of genetically modified derivatives within that murine genetic backbone.This allows investigation of anaphylatoxin efficacy in BMMs lacking the relevant complement receptors C3aR, C5aR1, and C5aR2.To that end, we investigated the role of the different anaphylatoxin receptors in R. australis infected BMMs.BMMs from wild-type (WT), C3aR −/ − , C5aR1 −/− , and C5aR2 −/− mice were infected with R. australis in the presence of NMS or C3 −/− MS.As shown in Fig. 4A, BMMs cultivated in C3 −/− MS were more permissive for the growth of the bacteria than NMS regardless of the absence of any single anaphylatoxin receptor, demonstrating that elimination of any single receptor is not sufficient to eliminate the growth control phenotype.
By pairing individual anaphylatoxin peptides with BMMs lacking the corresponding anaphylatoxin receptor(s), we were able to interrogate each individual anaphylatoxin/ anaphylatoxin receptor pair.To identify which peptide/receptor pairs are responsible for restriction of Rickettsia proliferation, we infected complement receptor-deficient BMMs within complement-deficient media (C3 −/− MS), but supplemented with either C3a or C5a peptides.As shown in Fig. 4B and C, the addition of C5a peptide to C3aR −/− BMM is sufficient to restrict bacterial proliferation, demonstrating that the interaction of C5a pair in mammalian cell biology is unresolved, but existing data often suggest a lack of immune activation through C5aR2-mediated signaling (67)(68)(69).

Anaphylatoxin agonism modulates macrophage immune phenotypes to control Rickettsia proliferation
Having established a role for anaphylatoxin/anaphylatoxin receptors in restricting the proliferation of multiple Rickettsia species within multiple different host cell types, we sought to identify the physiological changes occurring within infected monocytes that ultimately block Rickettsia proliferation.A central cell signaling phenotype associated with anaphylatoxin agonism is phosphorylation of ERK1 and ERK2 protein-serine/threo nine kinases (42,(70)(71)(72)(73).To investigate if exogenous anaphylatoxin stimulation also induced ERK1/2 phosphorylation in R. australis-infected BMMs, the host cells were infected with R. australis and treated with anaphylatoxin peptides for a total of 10 min.As shown in Fig. 5A, R. australis-infected BMMs treated with anaphylatoxins demonstrate greater ERK1/2 phosphorylation after 10 min of infection than mock-treated R. aus tralis-infected BMMs, suggesting that anaphylatoxin agonism of anaphylatoxin receptors induces rapid signal transduction resulting in the phosphorylation of ERK1/2.
With the knowledge that some innate immune cells are capable of preventing the essential rickettsial pathogenic act of phagosomal escape, we subsequently queried for anaphylatoxin-induced changes to Rickettsia localization within host cells (5,66,74).C57BL/6J BMMs cultured in complement-deficient serum were treated with either PBS or C3a + C5a anaphylatoxins.After 10 min of anaphylatoxin stimulation, the BMM were infected with R. australis.After an additional 1 h of incubation, the samples were fixed with paraformaldehyde to halt all progression through the invasion process.Phagosomal escape was assessed by confocal microscopy using lysosomal-associated membrane protein 1 (LAMP-1) and anti-Rickettsia antibodies.As shown in Fig. 5B, more bacilli colocalize with LAMP-1 in anaphylatoxin-treated BMMs than their mock-treated counter parts.Quantifying the frequency of colocalization indicates that anaphylatoxin treatment results in a significant increase in LAMP-1-positive Rickettsia (Fig. 5C), suggesting that anaphylatoxin signaling induces BMMs to block or delay R. australis phagosomal escape.
Finally, anaphylatoxin/anaphylatoxin receptor activation is known to induce global changes to the transcriptional profile of leukocytes (75,76).Previous studies have shown that secretion of cytokines, including IL-6, TNF-α, and IFN-γ, is associated with inhibition of Rickettsia growth (48,(77)(78)(79)(80)(81)(82).To determine if anaphylatoxin treatment changes the transcription of inflammation-associated transcripts within infected cells, we again infected WT BMMs with Rickettsia in the presence of complement-active NMS or complement-deficient C3 −/− MS.Infected BMMs were harvested 24 h post infection to assess the quantity of mammalian transcripts by qPCR.The quantity of each transcript was equilibrated by comparison to actin mRNA.As shown in Fig. 5D, the quantity of transcripts encoding for the cytokine inducible nitric oxide synthase iNOS (NOS2), the M1 macrophage marker CD38 (CD38), and the cytokines IFN-γ (IFNG), TNF-α (TNFa), and IL-6 (IL6) were all increased with complement activation, indicating that anaphylatoxin agonism induces transcriptional reprogramming within the infected BMMs.mRNA encoding for the anaphylatoxin receptors was also increased, illuminating a potential feedback loop.Together, the observed increase in (i) ERK1/2 phosphorylation, (ii) localization of Rickettsia in phagosomes, and (iii) levels of mRNA encoding for proinflammatory markers indicates that anaphylatoxin stimulation activates macrophages to control the proliferation of pathogenic Rickettsia.

DISCUSSION
Herein, we report a role for anaphylatoxins and anaphylatoxin receptors in reducing intracellular growth of pathogenic Rickettsia in monocytic host cells.Stimulation with either C3a or C5a is sufficient to induce murine RAW264.7,human THP-1, and murine bone marrow macrophages to directly control the proliferation of three different rickettsial pathogens, including R. parkeri, R. rickettsii, and R. australis.Additionally, anaphylatoxin stimulation of THP-1 macrophages impacts Rickettsia survival in endothe lial/macrophage co-cultures.The growth restriction phenotype is induced by the presence of both anaphylatoxin and cognate receptors, with C3a/C3aR, C5a/C5aR1, and C5a/C5aR2 pairs each contributing to efficacy.Finally, anaphylatoxin stimulation of Rickettsia-infected bone marrow macrophages produces proinflammatory phenotypes that correlate with a reduction in intracellular growth.Together, these data demonstrate that Rickettsia-induced anaphylatoxin production acts upon leukocytes to control intracellular Rickettsia replication.
When viewing the host-pathogen interaction through the lens of the host cell, we currently understand that macrophages have the capacity to mount an innate immune response when infected by Rickettsia pathogens (83).In this manuscript, we have established that activation of the anaphylatoxin-anaphylatoxin receptor axis modifies monocytic cells to control Rickettsia infection.The individual steps that connect anaphylatoxins to restricted Rickettsia proliferation include the following: (i) the presence of extracellular anaphylatoxins C3a or C5a; (ii) recognition of these peptides by the cognate anaphylatoxin receptors C3aR, C5aR1, or C5aR2; (iii) induction of intracytoplas mic cell signaling involving ERK1/2 phosphorylation; and (iv) changes to the physiology of the mammalian cell that include restriction of rickettsial phagosomal escape and differential transcriptional programming.
The requirement for extracellular anaphylatoxins correlates with other exogenous signals that can activate both infected leukocytes and endothelial cells to control Rickettsia proliferation.The chemokines CCL5, CXCL5, and CXCL10 and the cytokines IFN-γ, TNF-α, IL-1α, IL-1β, IL-6, IL-8, IL-12, and IL-18 have all been associated with activation of Rickettsia-infected endothelial cells or leukocytes (16,17,(84)(85)(86)(87)(88).The primary difference that separates anaphylatoxins from these previously identified classical immunostimulants is that chemokines and cytokines are largely produced de novo by cells after detection of a pathogen, whereas anaphylatoxins are produced by proteolysis of preformed precursors independent of a host cell.Due to the speed of anaphylatoxin production, it is reasonable to postulate that anaphylatoxin stimulation precedes the activation of innate immunity in earnest.
Extracellular C3a and C5a peptides are detected by the cell anaphylatoxin receptors C3aR, C5aR1, and C5aR2 (Fig. 1G; Fig. 2D; Fig. 4; Fig. 5A).Additionally, each individual peptide/receptor pair was sufficient to mediate restriction of Rickettsia growth and ERK1/2 phosphorylation, with no observed additive effect with the application of both anaphylatoxins (Fig. 1E and F; Fig. 2C; Fig. 3D).Previous studies have observed a synergetic effect when endothelial cells are treated simultaneously with C3a and C5a to release IL8 ( 89), but not IL-6 or IFN-γ and IL-2 responses in the whole blood of healthy donors after stimulation by hepatitis B virus antigens (90).Although we never observed a differential or synergistic effect of different anaphylatoxins, it may remain worthwhile to investigate how each anaphylatoxin integrates into our understanding of the immune response to Rickettsia infection.The anaphylatoxin receptors therefore join an ever-increasing repertoire of innate immune receptors that can indirectly or directly sense the presence of Rickettsia bacilli.Previous findings have demonstrated that the innate immune system can directly detect Rickettsia through preexisting IgM molecules that activate the complement system to produce anaphylatoxins (91).Therefore, the pathway from (i) IgM recognition of the bacterial surface, (ii) anaphylatoxin production, and (iii) anaphylatoxin receptor agonism is a strong surveillance pathway for alerting mammalian cells about the presence of Rickettsia.Anaphylatoxin-mediated immunosur veillance joins other previously identified mechanisms that include Toll-like receptor 4 and a cytoplasmic sensor connected to the NLRP3 inflammasome (83,(92)(93)(94)(95).Our current knowledge posits that these three surveillance mechanisms are the initiators of the innate immune response to Rickettsia infection.
In Fig. 5A, we have established that anaphylatoxin receptor activation induces cell signaling changes within the infected cell that include phosphorylation of ERK1 and ERK2.ERK1/2 signaling has previously been connected to other related Rickettsiales.Anaplasma phagocytophilum and Orientia tsutsugamushi both induce ERK1/2 phosphor ylation, but Ehrlichia chaffeensis infection of macrophages decreases ERK1/2 phosphory lation (96)(97)(98)(99)(100)(101).As such, anaphylatoxin-induced ERK1/2 activation is correlative to the "activated" cell signaling profile that supports host control of Rickettsia growth (102).Many different cell signaling proteins, including MyD88, NF-κB, ISG15, SOCS1/UBP43, and p38, have been identified as correlates of productively activated monocytic or endothe lial cells (103)(104)(105)(106)(107). The relative importance of each of these host activation markers has yet to be determined, but together, these proteins serve as the foundation for examining how Rickettsia "deactivates" infected host cells.Rickettsial manipulation of host signaling is currently exemplified at a molecular level by modulation of PI3K and Arf6 activity but is also apparent by noninflammatory changes to transcriptional and proteomic profiles induced by infection (15,60,(108)(109)(110).It is clear that there is an ongoing battle for cell signaling mechanisms occurring within the cytoplasm of Rickettsia-infected leukocytes.
The final outcome of anaphylatoxin stimulation is the induction of changes to the physiology of the host monocyte.At the most basic level, the conflict between Rickettsia and a host leukocyte is summarized as bacterial proliferation within the host cytoplasm versus intracellular killing of the bacteria through (i) inducible nitric oxide synthesis, (ii) hydrogen peroxide and reactive oxygen species production, and (iii) sequestration of tryptophan (48,111,112).The time it takes to fully implement these antibacterial strategies may explain why we only observed a restriction of Rickettsia growth after 3 days of infection.Our observed anaphylatoxin-induced increase in Rickettsia-LAMP1 colocalization (Fig. 5B and C), induction of iNOS transcription (Fig. 5D), and induction of the macrophage activation marker CD38 (Fig. 5D) strongly supports the concept that anaphylatoxin stimulation leads to intracellular killing of Rickettsia (113)(114)(115)(116).This model implies that the cell that detects anaphylatoxins is the same cell that induces bactericidal mechanisms.As demonstrated by the changes to the transcription of multiple proinflammatory genes (Fig. 5D) and activated-monocyte restriction of rickettsial proliferation in endothelial cells (Fig. 2F), anaphylatoxin stimulation also has effects on the larger population.Indeed, increased transcription of genes encoding for the cytokines IFN-γ, IL-6, and TNF-α all support the concept that anaphylatoxin production is a comprehensive activator of innate immunity beyond the specific monocyte that detects the presence of C3a or C5a (16,(117)(118)(119).Our data show that complement activa tion and anaphylatoxins drive Rickettsia-infected macrophages toward an inflammatory phenotype through the modulation of cytokine transcription.
Activation of endothelial cell signaling by Rickettsia infection induces the production of cytokines and the recruitment of many different innate and adaptive immune cell types (13,117).As shown in Fig. 2E and F, EA.hy926 is unable to restrict Rickettsia proliferation after anaphylatoxin stimulation.However, the addition of anaphylatoxinactivated macrophages is sufficient to reduce Rickettsia proliferation in the entire culture.While the mechanism of monocyte/endothelium communication is not elucidated in the present study, previous studies have highlighted that co-culture of macrophages modified the transcriptional profile of endothelial cells (146).Additionally, infection of endothelial cells with Rickettsia attracts leukocytes to adhere to the infected endothelial cell, presumably to help resolve the local infection (147,148).As such, there is a clear two-way communication between endothelial cells and monocytes, whereby infected endothelial cells recruit leukocytes, which in turn activate the bactericidal mechanisms within the infected endothelial cells (149)(150)(151)(152).As the anaphylatoxin/anaphylatoxin receptor axis enhances macrophage-dependent restriction of Rickettsia growth within endothelial cells, anaphylatoxin production is likely to be one of these chemical messengers responsible for endothelium/macrophage communication.
Two reports have linked anaphylatoxin receptor overexpression to disruption of macrophage functionality (153,154).As shown in Fig. 5C, anaphylatoxin stimulation increased transcription of the three anaphylatoxin receptors, C3aR, C5aR1, and C5aR2.Therefore, in the C57BL/6J BMM model system, anaphylatoxin signaling induces a feedback loop that increases expression of the anaphylatoxin receptors.Anaphylatoxinstimulated monocytes may elaborate increased sensitivity to further anaphylatoxin stimulation, raising the specter of immunopathogenesis associated with overabundant anaphylatoxin/anaphylatoxin receptor axes.
The data presented in this manuscript integrates into multiple ongoing investigations into the Rickettsia-monocyte interaction.These studies are largely derived from the concept that endothelial cells are "susceptible" to Rickettsia parasitism, but activated leukocytes have the ability to both directly combat infection and activate nearby cells to generate an overall environment less hospitable to Rickettsia infection.Supported by the observation of pathogenic Rickettsia within monocytic cells in vivo and Rickett sia akari tropism for monocytes (28,155,156), the Rickettsia-monocyte host-pathogen interaction has developed into a significant line of inquiry (5,47,59).From the patho gen perspective, some Rickettsia have evolved mechanisms to survive and replicate in macrophages.Indeed, proliferation within cultured macrophages is a hallmark of pathogenic species that is lacking in the nonpathogenic species (15,59).Rickettsia infection of THP-1 macrophages leads to the activation of the lipid catabolic pathway and induces reprograming of the cells toward an anti-inflammatory profile (15,157).Furthermore, R. australis can establish a cytosolic niche in macrophages by inhibiting the bactericidal effect mediated by IL-1β via the Atg5-dependent autophagy response, leading to systemic infection of R. australis (158,159).Finally, R. parkeri OmpB blocks polyubiquitylation to promote autophagy evasion in macrophages and is shown to be essential for bacterial growth (160).Together, these data demonstrate that Rickettsia species have evolved molecular mechanisms to deactivate monocytic cells.However, when examining the complex interplay ongoing within Rickettsia-infected monocytes, anaphylatoxin signaling lies well within the host repertoire of antibacterial mechanisms.
In summary, we have utilized models of Rickettsia infection in monocytes to elucidate the role of complement activation and the anaphylatoxin/anaphylatoxin receptor axis in modulating the response to infection.Anaphylatoxins were found to activate macrophages to restrict Rickettsia infection by limiting phagosome escape.Activated macrophages accumulate pERK1/2 and increase transcription of monocyte activation markers and pro-inflammatory cytokines.Overall, our work illuminates the role of the anaphylatoxin/anaphylatoxin receptor axis in the innate immune control of Rickettsia parasitism.
R. parkeri Portsmouth, R. australis Cutlack, and R. rickettsii Sheila Smith were cultivated in Vero cells, lysed using blunt cannulas, and sucrose-gradient purified as previously described (165).Rickettsial stocks were aliquoted in sucrose-phosphate-glutamate buffer and stored at −80°C until use.The quantity of bacteria was determined by titration, as has been previously described (166).

Serum and tissues
Pooled NHS was purchased from Innovative Research.NHS was heat-inactivated at 56°C for 30 min or with 20 mM EDTA where noted (62,167).NMS (C57BL/6J) and C3 −/− MS were isolated by cardiac puncture and Z-gel (Sarstedt) centrifugation.All serum samples were aliquoted, snap-frozen, stored at −80°C, and thawed immediately before use.

Rickettsia infections
RAW264.7 cells were cultured in 96-well plates for 24 h in cDMEM.The media were removed from cells and replaced with DMEM media with 10% NMS or C3 −/− MS.After 20 min, cells were infected with Rickettsia spp. at multiplicity of infection (MOI) = 1.THP-1 and EA.hy926 cells were seeded in 96-well plates for 24 h in cRPMI or cDMEM.The media were removed from cells and replaced with media with either 10% normal human serum, heat-inactivated human serum, or EDTA-inactivated human serum.After 20 min, the cells were infected by Rickettsia species at MOI = 1.
For BMM infections, the M-CSF-induced BMMs were seeded in a 96-well plate at 5 × 10 4 cells/well.Prior to infection, the media were removed and replaced with DMEM media containing NMS or C3 −/− mouse serum in the presence or absence of anaphy latoxin.The BMMs were infected with R. australis at MOI = 1 and incubated for an additional 3 days before analysis.

Anaphylatoxin receptor antagonists
Media were supplemented with the C3a receptor antagonist 60 nM SB290170 (Sigma) or C5a receptor antagonist 40 nM PMX53 (Sigma) as described (56,57).Cells were pretreated for 20 min before infection by Rickettsia at MOI = 1 and were incubated for an additional 3 days before DNA extraction.

Assessment of cell viability
At the time points indicated for each cognate qPCR, a CyQUANT MTT Cell Viability Assay (Thermofisher) was performed on the live cells following the manufacturer's recommendation.The absorbance was read at 540 nm using a Spectramax M2 micro plate reader (Molecular Devices), and the data are graphed as the absorption of at least four different samples.

Immune correlates of activation
C3a and C5a concentrations were determined by an enzyme-linked immunosorbent assay as previously described (91).Appropriate media were harvested on day 3 and used for the evaluation of complement activation.

Analysis of bacterial loads and expression of murine genes
Total genomic DNA (mouse and Rickettsia) was isolated using the PureLink Genomic DNA Mini Kit (Thermofisher).DNA was quantified by PCR utilizing GreenLink No-ROX QPCR Mix (BioLink) and a CFX-Connect Real-Time System (BioRad) using mouse actin (24) and human GADPH (171) primers that were previously described.
RNA was isolated using the RNA Mini Kit (Invitrogen), DNA was removed using the TURBO DNA-free Kit (Thermo Scientific), and cDNA was generated using the RevertAid RT Reverse Transcription Kit (Thermo Scientific).Primers used for the quantification of mouse IFNG, IL6, NOS2, TNFa, CD38, C3aR, C5aR1, and C5aR2 are listed in Table S1.All samples were graphed against a standard curve of the specific amplicon cloned into the pCR2.1 plasmid.The data are expressed as the ratio of the targeted gene to Mus musculus actin cDNA.

Confocal microscopy
Cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked in PBS containing 2% bovine serum albumin (BSA).Rickettsia species were detected with rabbit anti-Rickettsia (RcPFA) (172), followed by goat anti-rabbit Alexa Fluor 488 antibody (Invitrogen).Mammalian nuclei were identified with DAPI.LAMP-1 was stained with mouse anti-CD107a clone 1D4B (Bioscience) and followed by goat anti-mouse Alexa Fluor 546 (Invitrogen).To image total Rickettsia, images were captured using a Zeiss LSM 800 Laser Scanning Confocal Microscope system at 40×.Rickettsia/LAMP1 co-localization was assessed by consistently obtaining five slices of optical sectioning of cells using a Z step using two channels simultaneously with 63× oil immersion.Acquired images were processed on ImageJ software.

Statistical analyses
Statistical analysis was performed on Excel and GraphPad PRISM software (version 8.  2F were analyzed with a one-way ANOVA with Tukey's multiple comparison test of all data sets.Data from Fig. 4A were analyzed with a one-way ANOVA with Dunnett's multiple comparison of each NMS + C3 −/− MS pair. an approved Assurance Statement (D16-00172) on file with the Office of Laboratory Animal Welfare.

FIG 1
FIG 1 Murine anaphylatoxins reduce Rickettsia proliferation in RAW264.7 cells.(A and B) Presence of the anaphylatoxins C3a (A) and C5a (B) in the media after 3 days of culture of R. parkeri-infected RAW264.7 cells in the presence of NMS and C3 −/− MS, as determined by an enzyme-linked immunosorbent assay.(C and D) Proliferation of R. parkeri (C) and R. rickettsii (D) in RAW264.7 cells cultivated with complement-active serum (NMS) or complement-deficient serum (C3 −/− MS), as determined by the quantitative PCR ratio of Rickettsia sca1 to murine actin DNA.(E and F) Quantity of R. parkeri (E) and R. rickettsii (F) after 3 days of culture in RAW264.7 cells with complement-deficient serum (C3 −/− MS) supplemented with phosphate-buffered saline, mouse C3a peptide, mouse C5a peptide, or both C3a and C5a peptides.(G) Presence of R. parkeri after 3 days of culture in RAW264.7 cells with complement-active serum (NMS), complement-deficient serum (C3 −/− MS), or complement-active serum with anaphylatoxin receptor antagonists PMX53 and SB290170 (NMS + inhibitors).*P < 0.05 by (A,B) Student's t-test; (C,D) one-way ANOVA with Sidák's multiple comparison of matched days; (E,F,G) one-way ANOVA with Dunnett's multiple comparison test to the NMS/ mock-treated control.All columns represent at least five data points, and experiments were repeated to ensure reproducibility.

FIG 2
FIG 2 Human anaphylatoxins restrict Rickettsia proliferation in THP-1 cells.(A) R. parkeri proliferation in human THP-1 monocytes cultivated with either complement-active NHS or complement-deficient hiHS, as determined by the qPCR ratio of Rickettsia sca1 to human gapdh DNA.(B) Growth of R. parkeri cultivated for 3 days in THP-1 cells with either NHS or complement-deficient EDTA-inactivated human serum.(C) Quantity of R. parkeri in THP-1 cells cultivated in complement hiHS supplemented with PBS, human C3a peptide, human C5a peptide, or both C3a and C5a peptides.(D) Quantity of R. parkeri after 3 days of growth in THP-1 cells cultivated in complement-active serum (NHS) or NMS plus the anaphylatoxin receptor antagonists PMX53 and SB290170 (NHS + inhibitors).(E) R. parkeri proliferation in EA.hy926 human endothelial cells cultivated complement-active NHS or complement-inactivated NHS + EDTA.(F) R. parkeri growth in endothelial (EA.hy926) and macrophage (THP-1) co-culture.R. parkeri proliferates in ea.hy926 or co-culture in complement-inactive serum, but growth is restricted in EA.hy926/THP-1 co-culture containing complement-active serum.*P < 0.05 by (A) one-way ANOVA with Sidák's multiple comparison of matched days; (B,D,E) Student's t-test; (C) one-way ANOVA with Dunnett's multiple comparison test to mock-treated control; (F) one-way ANOVA with Tukey's multiple comparison test of all columns.All columns represent at least five data points, and experiments were repeated to ensure reproducibility.

FIG 4
FIG 4 The anaphylatoxin receptors C3aR1, C5aR1, and C5aR2 are essential to the restriction of Rickettsia australis growth in murine BMM.(A) Quantity R. australis after 3 days of growth in WT, C3aR −/− , C5aR1 −/− , and C5aR2 −/− BMMs cultivated in the presence of complement-active NMS or complement-deficient C3 −/− MS as determined by the qPCR ratio of R. australis sca1 to murine actin DNA.(B, C) R. australis growth in C3aR −/− BMM cultured with complement-deficient C3 −/− mouse serum supplemented with PBS or mouse C5a peptide as determined by qPCR (B) and Rickettsia (green) and DAPI (blue) microscopy (C).(D,E) Quantity of R. australis after 3 days of growth in C5aR1 −/− BMM cultured with complement-deficient C3 −/− mouse serum supplemented with PBS, mouse C3a peptide, or mouse C5a peptide as determined by qPCR (D) and Rickettsia (green) and DAPI (blue) microscopy (E).(F, G) Quantity of R. australis after 3 days of growth in C5aR2 −/− BMM cultured with complement-deficient C3 −/− mouse serum supplemented with PBS, mouse C3a peptide, or mouse C5a peptide as determined by qPCR (F) and Rickettsia (green) and DAPI (blue) microscopy (G).*P < 0.05 by (A) one-way ANOVA with Sidák's multiple comparison of matched receptor mutants; (B,D,F) Student's t-test.Scale bar = 20 µm.All columns represent at least five data points, and experiments were repeated to ensure reproducibility.

FIG 5
FIG 5 Anaphylatoxins modulate the phenotype of murine BMM, leading to decreased Rickettsia proliferation.(A) ERK1/2 phosphorylation as compared to total ERK1/2 and actin from R. australis-infected BMMs treated with PBS or anaphylatoxins C3a and C5a.(B) Representative images of R. australis cultured for 1 h in BMM pretreated with complement-deficient C3 −/− MS or C3 −/− MS supplemented with mouse C3a and C5a peptides as determined by confocal microscopy using the endosomal/lysosomal marker lysosomal-associated membrane protein 1 (LAMP-1), anti-Rickettsia, and DAPI.(C) Subsequent quantification of R. australis-LAMP1 co-localization in over 100 single BMMs.(D) Quantification of cDNA from Rickettsia-infected BMMs cultured in the presence of complement-active NMS or complement-deficient C3 −/− MS.Macrophage activation markers, pro-inflammatory cytokines, and anaphylatoxin receptor mRNA were quantified by qPCR as compared to actin cDNA.*P < 0.05 by Students t-test.Scale bar = 5 µm.All columns represent at least five data points, and experiments were repeated to ensure reproducibility.LAMP-1 colocalization was observed for 100 individual cells in each column and was repeated to ensure reproducibility.
3).Data from Fig. 1A, B, Fig. 2B, D, and E Fig. 3C, F, Fig. 4B, and Fig. 5C and D were analyzed by the Student's t-test.Data from Fig. 1C, D, Fig. 2A, and Fig. 3A were analyzed by a one-way ANOVA with Sidák's multiple comparison of matched days.Data from Fig. 1E, F, and G Fig. 2C, Fig. 3D, and Fig. 4D and F were analyzed by a one-way ANOVA with Dunnett's multiple comparison test to mock-treated control.Data from Fig.