Lutzomyia longipalpis Saliva Induces Heme Oxygenase-1 Expression at Bite Sites

Sand flies bite mammalian hosts to obtain a blood meal, driving changes in the host inflammatory response that support the establishment of Leishmania infection. This effect is partially attributed to components of sand fly saliva, which are able to recruit and activate leukocytes. Our group has shown that heme oxygenase-1 (HO-1) favors Leishmania survival in infected cells by reducing inflammatory responses. Here, we show that exposure to sand fly bites is associated with induction of HO-1 in vivo. Histopathological analyses of skin specimens from human volunteers experimentally exposed to sand fly bites revealed that HO-1 and Nrf2 are produced at bite sites in the skin. These results were recapitulated in mice ears injected with a salivary gland sonicate (SGS) or exposed to sand fly bites, indicating that vector saliva may be a key factor in triggering HO-1 expression. Resident skin macrophages were the main source HO-1 at 24–48 h after bites. Additionally, assays in vivo after bites and in vitro after stimulation with saliva both demonstrated that HO-1 production by macrophages was Nrf2-dependent. Collectively, our data demonstrates that vector saliva induces early HO-1 production at the bite sites, representing a major event associated with establishment of naturally-transmitted Leishmania infections.


INTRODUCTION
The leishmaniases are a group of diseases caused by protozoan parasites from more than 20 Leishmania species (1). There are three main forms of the disease: visceral leishmaniasis (VL, also known as kala azar), cutaneous leishmaniasis (CL), and mucosal leishmaniasis (MCL). While CL is the most common form of the disease, VL is the most serious and can be fatal if untreated. Most VL cases occur in Brazil, East Africa and in South-East Asia with an estimated 50,000-90,000 new cases occuring worldwide each year (2).
Phlebotomine sand flies (Diptera: Phlebotominae) are blood feeding insects of medical importance that transmit parasites of the genus Leishmania. Natural transmission of Leishmania to humans occurs during blood acquisition by female phlebotomine sand flies. As the sand fly bites, it introduces saliva into the vertebrate host dermis alongside metacyclic infective promastigotes and other vector-derived factors (3)(4)(5)(6). Studies have demonstrated that saliva promotes Leishmania infections. L. major coinjected with Lutzomyia longipalpis or Phlebotomus papatasi saliva resulted in a more severe disease reflected by larger lesions when compared with a group of mice receiving parasites alone (7). This initial observation was supported by additional studies demonstrating the enhanced infectivity of L. major when coinoculated with saliva from the sand fly L. longipalpis (8,9). Apart from antihemostatic properties, sand fly saliva is chemotactic for different immune cells, such as macrophages (10), neutrophils (11), dendritic cells (12), and lymphocytes (13). In addition, many cell types, including monocytes, interact with sand fly saliva, thereby modifying inflammatory processes at the blood feeding site (3). It has been proposed that such effects on the host immune system contribute to increased parasite loads in mice exposed to sand fly bites compared to animals infected through needle injection (3). The specific mechanism underlying the effect of vector saliva on the host immune response is not fully understood.
Heme oxygenase-1 (HO-1) is a key enzyme triggered by cellular stress, exhibiting cytoprotective, antioxidant, and antiinflammatory properties (14). HO-1 is the rate-limiting enzyme in the catabolism of heme. It breaks down the porphyrin ring to yield equimolar amounts of biliverdin, free iron (Fe +2 ), and carbon monoxide (CO) (15). Pharmacological induction of HO-1 or administration of the end products of its activity can exert therapeutic effects in a variety of immune-mediated inflammatory diseases (16). The transcriptional induction of HO-1 occurs in response to multiple forms of cellular stress and tissue damage. Oxidative stress activates the transcription factor nuclear factor erythroid 2-related factor-2 (Nrf2), which in turn binds antioxidant response element (ARE) enhancers and induces expression of protective antioxidant genes including HO-1. Activation of Nrf2 requires its translocation to the nucleus and binding to ARE enhancer motifs (17,18). In addition to Nrf2, IL-10 has been shown to directly mediate HO-1 expression (19). Importantly, IL-10 is highly upregulated in the skin up to 18 h after sand fly bites (6).
Previous studies from our group have shown that HO-1 favors Leishmania infection (20,21). Based on our findings, we have proposed that HO-1 plays two major roles following Leishmania infection: (1) it prevents host cell damage, and (2) it decreases the ability of the host to limit intracellular growth of the parasite. Of note, we have also found that patients with VL presented higher systemic concentrations of HO-1 than healthy individuals, and that the levels of this enzyme substantially decreased after Abbreviations: SGS, salivary gland sonicate. leishmanicidal treatment, suggesting that HO-1 is a potential biomarker of active VL (21,22).
In the present study, we first demonstrated that experimental exposure to sand fly bites in humans triggers robust HO-1 protein expression in situ, in the skin, but not in blood. Further experiments in mice revealed that experimental exposure of mice ears to sand fly bites induced Nrf2-dependent HO-1 production by resident macrophages after 24-48 h. Similarly, injection of saliva induced HO-1 production in the ear dermis. Lastly, in vitro assays using macrophage cell lines showed that saliva drives Nrf2 translocation to the nucleus and ARE activation, leading to augmented HO-1 expression. Together, our findings identified HO-1 as a novel inflammatory mediator triggered by saliva deposited into skin during sand fly bites. This further facilitates our understanding of the early events during Leishmania infection that contribute to the successful establishment of the pathogen and onset of leishmaniasis.

Experimental Exposure to Lutzomyia longipalpis Bites Induces in situ Expression of HO-1 and Nrf2 in Humans
Having previously demonstrated that HO-1 is induced during Leishmania infection and identified its association with active VL (21), we asked whether it plays a part in the early stages of parasite establishment after vector-transmission. Immunohistochemistry of paraffin-embedded skin sections, obtained 48 h after the last of several experimental exposures of human volunteers to sand fly bites, revealed that both HO-1 (Figures 1A,B and Supplemmentary Figure 1A) and Nrf2 (Figures 1C,D, Supplemmentary Figure 1B) were produced in the dermis in situ. Expression of both HO-1 and Nrf2 was not detected in control specimens of skin not bitten by sand flies taken from the same individuals ( Figures 1E,F).
Next, we tested whether circulating concentrations of HO-1 protein levels in plasma of the human volunteers is also altered after exposure to sand fly bites. We observed that levels of HO-1 were undistinguishable between samples collected prior to exposure to bites and after up to nine repeated exposures ( Figure 1G). Thus, HO-1 induction by bites of L. longipalpis in humans is restricted to the skin and is not followed by changes in its systemic levels. The real but focal HO-1 response driven by the sand fly bites is reinforced by the fact that the biopsies came from volunteers whose serum values remained unchanged (shown in blue and purple lines in Figure 1G).

Bites of Lutzomyia longipalpis Induce Nfr2-Dependent HO-1 Production in Resident Skin Macrophagess
To investigate whether sand flies also induce HO-1 during blood feeding on mice, we exposed 20 Lu. longipalpis adult females to mice ears and followed the kinetics of HO-1 production in host skin by western blot. Compared to naïve, unbitten, skin, HO-1 protein levels peaked at 24 h after sand fly bites and remained high up to 48 h before returning to basal levels at 1 week (Figure 2A). Of note, levels of HO-1 were comparable to controls at 6 h after sand fly bites. No differences were observed in the sand fly feeding score among the groups (Supplemmentary Figure 2A). Next, we assessed the involvement of Nrf2 or IL-10 in HO-1 induction at 24 h after sand fly bites using animals genetically deficient in either of these molecules. In contrast to wild-type and IL-10 −/− animals, HO-1 induction was substantially diminished exclusively in Nrf2 −/− animals ( Figure 2B). No differences were observed in the sand fly feeding score among the groups (Supplemmentary Figure 2B).
To identify the source of HO-1, we exposed mice ears to bites of 20 sand flies and recovered the cells 24 h later. Ear cells were sorted into non-myeloid cells, neutrophils, resident macrophages and inflammatory monocytes according to the gating strategy shown in Figure 2C and Supplemmentary Figure 3. Western blots of the total protein extracted from each population demonstrated that HO-1 is mainly produced by resident macrophages with a minor contribution from inflammatory monocytes ( Figure 2D). Taken together, our data shows that sand fly bites trigger Nrf2-dependent expression of HO-1 by resident macrophages in host skin.

SGS Induces HO-1 Expression in Mice and in Human Macrophages
We next tested whether sand fly saliva, deposited into host skin during blood feeding, could be the component contributing to the increased production of HO-1 in the dermis. Mice ears challenged intradermally with SGS exhibited high levels of HO-1 compared to animals challenged only with PBS at 24 h post-injection (Figures 3A,B). To verify if saliva was acting directly on macrophages, we conducted immunocytofluorescence assays using human THP-1 derived macrophages. Macrophages incubated with SGS induced a positive intracellular staining for HO-1 ( Figure 4A). This finding was further confirmed using a WB assay, which also revealed that HO-1 induction in vitro occurred as early as 2 h post incubation of cells with SGS and was sustained up to 8 h later ( Figure 4B). Together, these results demonstrate that HO-1 is robustly induced by SGS in vivo and in vitro.
Expression of HO-1 is described to be dependent on induction of the canonical antioxidant response element (ARE), which is recognized by the transcription factor Nrf2, which translocates from the cytosol to nucleus when activated (23). To confirm that the ARE/Nrf2 pathway is involved in SGS-driven HO-1 induction in our system, we evaluated Nrf2 protein expression in nuclear extracts from THP-1 FIGURE 2 | Bites of Lutzomyia longipalpis sand flies induce Nfr2-dependent HO-1 protein expression by resident macrophages in mice. Ears of C57BL/6 mice were exposed to bites of 20 uninfected L. longipalpis sand flies (UB, uinfected bite) and tissue was processed for total protein extraction before or after sorting by flow cytometry. (A,B) Western blots of total protein extracts from mice ears. Hsp90 was used as a loading control. (A) kinetics of HO-1 expression following sand fly bites. Total protein extract from a mouse spleen was used as a positive control. (B) HO-1 expression in wild-type, Nrf2 knockout (Nrf2 −/− ) and IL-10 knockout (IL-10 −/− ) mice 24 h after exposure to sand fly bites. (C,D) Ear cells recovered 24 h after sand fly bites were stained with antibodies for population sorting by flow cytometry prior to total protein extraction. (C) Gating strategy for sorting populations of non-myeloid cells (I), neutrophils (II), resident macrophages (III) and inflammatory monocytes (IV). (D) Western blot of total protein extract from populations sorted according to (C). Naïve total ear cells were used as a negative control (CT). Histone H3 was used as a loading control. Blots are representative of two to three independent experiments. macrophages incubated with SGS. As expected, Nrf2 expression was substantially increased as early as 2 h post incubation ( Figure 4C). Participation of this pathway was further illustrated by using a luciferase reporter assay. In this experimental system, plasmid constructs containing the canonical ARE promoter response element or the Nrf2 promoter are transfected into RAW264.7 cells. Using this approach, we found that ARE and Nrf2 luciferase activities were significantly increased after incubation with SGS ( Figure 4D). These findings reveal that SGS induces HO-1 in cells through activation of the ARE-Nrf2 pathway.

DISCUSSION
Modification of host inflammatory responses at the site of the sand-fly bite is critical for successful blood feeding (24), but it also plays an important role in the establishment of Leishmania infection (25). It is also well-known that experimental exposure to Leishmania-infected sand fly bites causes more severe disease associated with higher parasite loads in mice, whereas needle injection of Leishmania alone results in milder outcomes (3). Although several mechanisms have been proposed to explain this discrepancy (26), the exact pathway directly associated with the enhancing effect of Leishmania transmission by sand fly bites on infection has not been fully characterized. The results from our study adds to our current knowledge of this topic by demonstrating that sand fly saliva robustly induces HO-1 in the host dermis at the bite site during blood feeding.
In the present study, we observed a significant induction of HO-1 expression in the dermis of human volunteers experimentally exposed to L. longipalpis bites. HO-1 induction was localized to the bite site with no change in HO-1 serum levels after several repeated exposures of individuals to sand fly bites. Thus, the sand fly bite alone does not trigger systemic inflammation, not surprising given the relatively small area of tissue damage at the bite site. Of note, only two volunteers consented to provide skin biopsies for analysis. Nevertheless, both biopsies showed a clear induction of both HO-1 and Nrf2. Further, induction of HO-1 in human macrophages in vitro as well as its induction in mice reinforce the results obtained from human biopsies.
HO-1 is thought to be an intracellular enzyme (27), and its detection in plasma would indicate substantial tissue damage. In this scenario, the release of HO-1 occurs after cellular lysis during robust inflammation associated with infectious diseases such as VL (21,28,29), tuberculosis (30) and malaria (31,32). For sand fly bite-induced HO-1 expression in the skin, our data demonstrates that it is triggered in resident skin macrophages by saliva and as such could play a critical role in the early establishment of Leishmania infection. Production of HO-1 in the dermis microenvironment may counterbalance the host antiparasite effector mechanisms such as free radial production. Additional studies are warranted to directly test this hypothesis.
The direct induction of HO-1 by sand-fly bite is intriguing. During blood feeding the vector probes in search for capillaries to produce the hemorrhagic pool upon which it feeds. Therefore, it is possible that mechanical or inflammation-driven (25) tissue damage could result in release of heme from damaged cells. Heme is an extremely toxic molecule, which can induce significant oxidative stress and cell death (33). Indeed, we have previously published that heme induces inflammatory cell death in neutrophils (34) and in macrophages (22) in the context of Leishmania infection in vitro. Importantly, heme is the canonical substrate of HO-1 (35), and accumulation of heme in tissues is directly associated with induction of this enzyme (36). Thus, it is reasonable to hypothesize that an increase in heme availability in the dermis at the site of a sand-fly bite could be the trigger for the local induction of HO-1 to minimize the deleterious effects of free heme during blood feeding. However, results from our in vitro and ex vivo experiments clearly demonstrate that HO-1 is directly induced by sand-fly saliva. This is an important observation because it indicates that HO-1 production may not be due to non-specific inflammation from tissue injury caused by the bite. Alternately, both tissue damage and saliva could be contributing to the robust induction of HO-1 after bites. Further studies are needed to elucidate this aspect of the innate immune response to sand fly bites. Either way, sand fly bite-mediated induction of HO-1 in the dermis early after sand fly bites is likely to support the establishment of Leishmania in the host. The specific molecule from saliva responsible for HO-1 induction remains to be determined.
One pathway for the induction of HO-1 requires activation and nuclear translocation of Nrf2 (37). This transcription factor is activated during cellular stress and it binds to the ARE motif in the nucleus, which promotes transcription of HO-1 mRNA (38). Other triggers of HO-1 expression have also been reported including IL-10 that induces HO-1 expression through activation of the signal transducer and activator of transcription 3 (STAT-3) (19). IL-10 is induced by saliva (26) and is present at high levels in mice ears after bites of L. longipalpis sand flies (6). Nevertheless, our data confirms that saliva directly induces HO-1 in cells through activation of the ARE-Nrf2 pathway and not indirectly through its induction of IL-10. Previous studies from our group have described that Leishmania infection interferes Frontiers in Immunology | www.frontiersin.org with macrophage signaling pathways triggering Nrf2 activation (23). As such, the induction of Nrf-2 by both sand fly saliva and Leishmania could serve as a mechanism to deal with the altered oxidative state of infected macrophages resulting in increased susceptibility to Leishmania infection.
Altogether, our findings demonstrate that sand fly saliva induces HO-1 production at the bite site. This suggests that in addition to its importance in disease pathogenesis, HO-1 may play a critical role in the early establishment of Leishmania parasites after vector-transmission.

Study Subjects
Healthy human volunteers were experimentally exposed to bites of 10 laboratory reared Lutzomyia longipalpis sand flies as approved by the Institutional Review Boards of the Walter Reed Army Medical Center, the Uniformed Services University of the Health Sciences, and the National Institute of Allergy and Infectious Diseases (NIAID). All clinical investigations were conducted in accordance with the Declaration of Helsinki revision 2013 principles. Written informed consent was obtained from all participants.

Sand Fly Rearing and Salivary Gland Homogenate Preparation
L. longipalpis sand flies were collected in Mali (39,40) and reared at the Laboratory of Malaria and Vector Research (LMVR), NIAID. Salivary glands were dissected from 50, 5-7-day old colonized female L. longipalpis sand flies. The salivary glands were pooled, aliquoted, and used as needed to perform in vitro experiments. Salivary gland sonicate (SGS) was prepared by ultrasonication, followed by centrifugation at 10,000 g for 3 min at 4 • C. Supernatants were collected and dried using a vacuum concentrator (Thermo, Asheville, NC). SGS were shipped to the Gonçalo Moniz Institute (IGM-FIOCRUZ, Salvador, Bahia-Brazil) and rehydrated with ultra-pure water (KD-Medical, Columbia, MD) immediately before use.

Exposure of Human Volunteers to Uninfected Sand Flies
A secure, custom-designed Plexiglas capsule (Precision Plastics, Beltsville, MD) with a meshed surface was used in the exposure experiments. Healthy Lutzomyia saliva antibody negative volunteers, 18-50 years old, were exposed to bites of 10 colony-reared female sand flies on a bi-weekly basis for the first 2 months (± 1 week) and once every 2 months (± 1 week) for the following 18 months. The capsule was strapped to the upper arm of each individual for 20 min and covered with a dark fabric to simulate nighttime conditions. All sand flies were accounted for and killed after each exposure. Two of 15 subjects volunteered 2 mm dermal specimens obtained by biopsy 48 h post-bite after the final exposure to sand flies.

Exposure of Mice to Uninfected Sand Flies
Mice were anesthetized i.p. with a combination of 150 mg/kg ketamine and 20 mg/kg xylazine. Both mouse ears were placed inside vials containing 20 L. longipalpis females. Sand flies were allowed to feed for 1 h in the dark. Feeding counts were obtained by counting fully blood fed flies using a stereoscope.

Immunodetection of HO-1 and Nrf2 in Skin Specimens Obtained by Biopsy and THP-1 Cells
A 2-mm punch biopsy was obtained from the bite site of 2 volunteers 48 h after the 9th exposure to L. longipalpis sand flies. Normal skin samples was obtained from healthy individuals by plastic surgery. Biopsied samples were stored in 10% buffered formalin solution, and paraffinembedded and sectioned (Histoserv, MD) for histological staining.
For immunofluorescence analysis, THP-1 cells (5 × 10 5 ) were added to wells containing phorbol myristate acetate (PMA) for differentiation, plated on coverslips and allowed to adhere for 3 days and left for an additional 3 days. Cells were stimulated with SGS (equivalent to 1 pair/well) or CoPP (30 µM); Cobalt protoporphyrin IX (CoPP) (Frontier Scientific, Logan, UT) is a pharmacologic inducer of HO-1. CoPP was dissolved in 0.1N NaOH and RPMI 1640 medium and adjusted to concentrations of 30 µM for in vitro assays. After 24 h, cells were fixed in 4% paraformaldehyde (Sigma), followed by permeabilization with 0.1% Triton X-10, and blocking with 1% BSA. Primary antibody against HO-1 overnight. Cells were then incubated with the secondary antibody, Texas Red R anti-mouse IgG (H+L) (1:3,000) (Vector Laboratories, Burlingame, CA), followed by nuclear staining with 1 µg/mL of DAPI. A control staining protocol employing only the secondary antibody (i.e., no primary antibody) was implemented following the same procedure described above. Coverslips were analyzed by the fluorescence microscopy Nikon E600, and digital images were obtained using an Olympus Q-Color 1 digital camera with the Image Pro Plus software.

HO-1 Quantification by ELISA
Seven out of 15 volunteers completed nine exposures to sand fly bites. Serum samples were obtained from each individual before (a) and at 1, 4, and 9 weeks after exposure to L. longipalpis sand flies. Serum levels of HO-1 were measured using a commercially available ELISA kit (Enzo Life Sciences, Farmingdale, NY, USA).
Immunoblotting THP-1 cells (1 × 10 6 cells) were washed twice with ice-cold PBS and then lysed in 100 µl of lysis buffer (50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 10 mM EGTA, 50 mM NaF, 20 mM β-glycerophosphate, 250 mM NaCl, 0.1% Triton X-100, 1 µg/ml BSA and a 1:100 dilution of protease inhibitor cocktail, Sigma-Aldrich, St. Louis, MO, USA) for total protein extraction. For nuclear protein extraction, after infection and/or treatment, the cells were washed twice with 1x PBS and then lysed with 100 µL of buffer A (HEPES 10 mM pH 7.9. 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, NP-40 0,25% (v/v); cocktail of protease inhibitors) for 10 min on ice. The lysed cells were centrifuged at 14,000 g for 1 min at 4 • C, and the pellet was resuspended in 60 µL of buffer C (20 mM HEPES pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 20% glycerol, protease inhibitor cocktail) and incubated on ice for 20 min. The lysate was centrifuged at 14,000 g for 5 min, and the supernatant containing nuclear proteins was collected in a new tube. The protein extracts were subjected to electrophoresis on 10% SDS-polyacrylamide gels and transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ, USA). After blocking with 5% nonfat dry milk in TBS with 0.1% Tween-20 (TBS-T), the blots were incubated over-night with antibodies against Nrf2 (12721) and Lamin (2032) from Cell Signaling Technology; and β-actin (47778), followed by anti-rabbit (2004) or anti-mouse (2005) horseradish peroxidase-conjugated IgG (1:4,000) from Santa Cruz Biotechnology. The membranes were then submitted to 3 washes with 0.1% TBS-T after each incubation, and the proteins were detected using the ECL chemiluminescent detection system (Amersham Biosciences).
For in vivo experiments samples, C57BL/6 mice were injected with PBS, 1 pair of SGS or CoPP (1 mg/kg) in 10 uL PBS in both ears. Non-stimulated (naïve)mice were kept in a separate cage. After 24 h, both ears from each mouse were harvested, ground and homogenized in 1 ml PBS supplemented with protease inhibitors. Whole-tissue extracts were prepared in 1% Nonidet P-40 lysis buffer (1% Nonidet P-40, 150 mM NaCl, 20 mM Hepes (pH 7.5) supplemented with protease and phosphatase inhibitor cocktails (Sigma). Proteins were separated by SDS-PAGE, transferred to a nitrocellulose membrane and probed with Abs against HO-1 (ADI-OSA-110-J, Enzo, Ann Arbor, MI) and β-actin (47778) from Santa Cruz Biotechnology. Bands were visualized using an Image Quant 4010 apparatus (GE healthcare).
After exposure to sand fly bites, mice ears were harvested and dermal sheets separated with forceps and placed in RIPA buffer (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with Halt TM Protease Inhibitor Cocktail (Thermo Fisher Scientific, Waltham, MA, USA) and PMSF (Sigma-Aldrich, St. Louis, MO, USA). The tissue was then submitted to mechanical disaggregation in a BD TM Medimachine (BD Biosciences, Franklin Lakes, NJ, USA). Any remaining large particles were removed using a 50 µm Filcon filter (BD Biosciences, Franklin Lakes, NJ, USA). Whole tissue extract was kept on ice to insure complete lysis. After 15 min, the extract was centrifugated at 8,000 g for 10 min and the soluble fraction was collected. Proteins were separated by NuPAGE TM

Transient Transfections and Luciferase Assays
To

Cell Sorting
After exposure to sand fly bites, mice ears were harvested and disinfected with ethanol. The dermal sheets were then separated using forceps and digested in PBS containing Liberase TL (Roche) at 37    Disclaimer: SK and JV are government employees and this is a government work, the work is in the public domain in the United States. Notwithstanding any other agreements, the NIH reserves the right to provide the work to PubMed Central for display and use by the public, and PubMed Central may tag or modify the work consistent with its customary practices. You can establish rights outside of the U.S. subject to a government use license.

Conflict of Interest Statement:
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. The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.