Acanthamoeba proteases contribute to macrophage activation through PAR1, but not PAR2

Acanthamoeba infections are characterized by an intense localized innate immune response associated with an influx of macrophages. Acanthamoeba protease production is known to affect virulence. Herein, the ability of Acanthamoeba trophozoite proteases, of either the laboratory Neff strain or a recently isolated clinical strain, to stimulate IL‐12 and IL‐6 and to activate protease‐activated receptors, PAR1 and PAR2 expressed on murine macrophages, was investigated.

of endogenous and exogenous proteases. 7 PARs are expressed in a ubiquitous manner in various immune, as well as non-immune cells, and have a wide range of physiological functions associated with the maintenance of tissue integrity, as well as with inflammatory and immunological responses. 8 Interestingly, PARs, especially PAR 1 and PAR 2 , are expressed on human corneal epithelial (HCE) cells, predominantly in the outer cell layers of the corneal epithelium, and their activation leads to production of pro-inflammatory cytokines, as well as wound healing processes. 9 Consequently, the role of innate immune receptors in Acanthamoeba recognition and response at the ocular surface has been an area of recent interest and it has been demonstrated that Acanthamoeba plasminogen activator factor stimulates, in HCE cells, the production of IL-8 in a PAR 2dependent, but PAR 1 -independent manner. 10 Once Acanthamoeba trophozoites reach the deepest layers of the tissues, macrophages as well as neutrophils are the predominant cell types present at the site of infection, 11 where they are found alongside trophozoites and cyst forms. Macrophages are therefore considered the major cells involved in Acanthamoeba clearance. 6 Although macrophages are heavily recruited during the development of both GAE and AK, little is known about how Acanthamoeba influences macrophage function.
Our recent studies have demonstrated the role of TLR4, MyD88dependent events in the activation of murine macrophages. 12 The potential role of proteases released by Acanthamoeba, and their interaction with PARs expressed by macrophages have not until now been reported.

| Acanthamoeba strains
In this study, we examined the effects of co-cultivation of trophozoites of either a laboratory strain of Acanthamoeba (Neff), isolated from soil over 60 years ago, and kindly donated by the late Prof.
Keith Vickerman (University of Glasgow, UK), or a clinical isolate (clinical) of Acanthamoeba with murine bone marrow-derived macrophages (MBDM). Both Acanthamoeba strains are of the T4 genotype. 13 Trophozoites of either strain were maintained in culture and prepared for the co-incubation experiments as previously reported. 12. of the neck," was used to sacrifice the animals to obtain tissue. This was followed by a confirmation method (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_ data/file/662364/Guidance_on_the_Operation_of_ASPA.pdf). The procedure was undertaken by a trained person. The name of the staff member performing the performing the Sch1 is held locally on a register of competent people.

| Inhibition of secreted amoebic proteases
In order to investigate the role of Acanthamoeba secreted proteases in influencing trophozoite-induced macrophage cytokine production, MBDM were obtained from the femurs of 7-week-old male BALB/c mice as previously reported, 14 and subsequently treated, prior to infection, with either leupeptin (Sigma Chemical Co, Poole, UK), a serine and cysteine proteases inhibitor, or E64 (Sigma Chemical Co) that selectively inhibits cysteine proteases. The choice of these two protease inhibitors relied on the specific composition of the Acanthamoeba secretome that mainly comprises serine and cysteine proteases. 15 Macrophages in complete RPMI (cRPMI) were considered the negative inhibitor control. Immediately after addition of protease inhibitors, macrophage cultures were incubated with either Neff or the clinical isolate trophozoites (ratio 1:1). IL-12 and IL-6 production was evaluated after 24 hours, by enzyme-linked immunosorbent assay (ELISA).

| Infection of C57BL/6 macrophages with Acanthamoeba castellanii trophozoites
The study was performed using C57BL/6 wild-type mice and C57BL/6 mice deficient for the PAR 2 gene. MBDM were obtained from both mouse strains and co-incubated with either Neff or the clinical isolate trophozoites at a ratio of 1:1. In addition, macrophages were stimulated with LPS as a control for their ability to produce IL-12 and IL-6.  trophozoites. Furthermore, a difference in IL-6 production was found between the protease inhibitors in macrophage cultures incubated with Neff trophozoites; leupeptin (0.303 ng/mL) was more efficient at inhibiting IL-6 than E64 (P < 0.005). This difference in macrophage IL-6 production was not demonstrated following incubation with the clinical isolate trophozoites ( Figure 1B). Neither of the inhibitors significantly influenced LPS induced macrophage IL-12 or IL-6 production indicating that their activities were targeting trophozoite protease activity specifically ( Figure 1A,B).

| Acanthamoeba-induced IL-12 and IL-6 production by murine macrophages is PAR 2independent
After demonstrating that amoebic serine and cysteine proteases could stimulate IL-12 and IL-6 production by murine macrophages, we investigated if PARs were involved in this event. PAR 2 was considered a potential target for our investigation, since it is an extracellular receptor widely expressed on immune cells and involved in inflammatory and immunological processes. 18 Both IL-6 and IL-12 production by PAR 2 −/− macrophages co-incubated with trophozoites of either Neff strain or the clinical were similar to that produced by WT macrophages (Figure 2).

| Acanthamoeba castellanii clinical strain induces IL-12 production by murine macrophages in a PAR 1dependent manner
In the absence of evidence of PAR 2 -dependent activation of macrophages by Acanthamoeba, further investigations were undertaken to determine whether PAR 1 was involved. Similar to PAR 2 , PAR 1 is an extracellular receptor widely expressed on immune cells and involved in inflammatory and immunological processes. 18 As expected, pre-treatment with RWJ 56110 did not induce either IL-12 or IL-6 production by murine macrophages (Figure 3A,B).
Stimulation of macrophages with LPS induced IL-12 and IL-6 production and this was significantly diminished by pre-treatment with RWJ 56110 (P < 0.05; Figure 3A,B). Pre-treatment of macrophages with RWJ 56110 significantly reduced IL-12 production in response to the clinical isolate trophozoites compared with control macrophages (P < 0.05); however, it did not significantly reduce Neff LPS at 200 ng/mL (LPS) was also included in the experimental design as positive control. Uninfected macrophages (Control) were considered the negative control. The experiment was performed twice. Results represent the mean ± standard error of n = 3. Mann-Whitney U test was applied to evaluate differences between C57BL76 WT and C57BL/6 PAR 2 −/− conditions. Macrophages derived from C57BL/6 PAR 2 −/− did not show reduced IL-12 or IL-6 production, relative to macrophages derived from wild-type mice, when co-incubated with trophozoites. Thus, the production of these pro-inflammatory cytokines by macrophages, in response to Acanthamoeba, is not PAR 2 -dependent F I G U R E 3 IL-12 (A), and IL-6 (B) production by murine macrophages, pre-treated with the PAR 1 antagonist RWJ 56110, after 24 h coincubation with Acanthamoeba. 1 × 10 6 murine macrophages, obtained from BALB/c mice, were pre-treated with 20 μM RWJ 56110 solution (+ RWJ56110) or not pre-treated (− RWJ56110). Macrophages were incubated for 10 min at 37° 5% CO 2 . Subsequently, macrophages were stimulated with LPS at a concentration of 200 ng/mL (LPS). Unstimulated macrophages, treated or not treated with RWJ 56110, were included in the experimental design (Control). The experiment was performed twice. Results represent the mean ± standard error of n = 3. Student's t test was applied to evaluate differences between − RWJ56110 and + RWJ56110 conditions. In the graphs, significances between − RWJ56110 and + RWJ56110 conditions are indicated as follow: for values of P < 0.05*. Values below the detectable levels are indicated in the graphs as ND (not detected). Data shown in the graphs indicate that IL-12 production by murine macrophages induced by clinical isolate trophozoites is at least in part PAR 1 -dependent, although it does not appear to be the main receptor involved in this event trophozoite-induced IL-12 production ( Figure 3A). Pre-treatment of macrophages with RWJ 56110 did not modify the Acanthamoebainduced IL-6 production ( Figure 3B).

| D ISCUSS I ON
During parasitic infections, proteases are known to be important virulence factors and to be involved in cell differentiation, tissue penetration, nutrient acquisition and immune evasion mechanisms. 19 The importance of Acanthamoeba proteases in pathogenicity, 4,20,21 immune evasion 22  However, during Neff strain infection of macrophages, leupeptin was more effective than E64 in inhibiting IL-6 production, indicating that serine proteases might be the major protease responsible for inducing this cytokine. By using PAR 2 deficient macrophages, we found that Acanthamoeba-induced IL-12 and IL-6 production was PAR 2 -independent. On the other hand, Acanthamoeba-induced IL-12 production by macrophages was shown to be induced through PAR 1 -dependent mechanisms. Interestingly, a very recent study has reported that Acanthamoeba plasminogen activator factor stimulates IL-8 production by human corneal epithelial cells in a PAR 2dependent manner. 10  have been reported and known to be associated with bacterial, viral and fungal infections, but not as yet during protozoan infections. 26 Consequently, the present study suggests previously unobserved interactions between PARs and TLRs during Acanthamoeba infections that influence cytokine production by macrophages, although more investigations are needed to elucidate these interactions.
In conclusion, the current study demonstrates for the first time a contribution of Acanthamoeba serine and cysteine proteases and host PAR 1 in macrophage activation and inflammatory responses. Such knowledge might ultimately provide insight into the pathogenesis of Acanthamoeba-induced diseases and their treatments.

ACK N OWLED G EM ENTS
We are grateful to Professor Robin Plevin for providing us with PAR 2 deficient mice.