Activation of the epidermal growth factor receptor initiates innate immune responses

Affiliations 8 1 Division of Infectious Diseases, Harbor-UCLA Medical Center, Torrance, California 90502, USA 9 2 Institute for Infection and Immunity, Los Angeles Biomedical Research Institute at Harbor-UCLA 10 Medical Center, Torrance, California 90502, USA 11 3 David Geffen School of Medicine at UCLA, Los Angeles, California 90024, USA 12 13 # Current address: Center for Infectious Disease Imaging (CIDI), Radiology and Imaging Sciences, 14 NIH Clinical Center, National Institutes of Health (NIH), Bethesda, Maryland 20892, USA 15 16 *Correspondence to: mswidergall@labiomed.org; sfiller@ucla.edu 17

oral epithelial cell fungal burden. EphA2 plays a central role in stimulating the epithelial cell 23 release of proinflammatory mediators that mediate resistance to OPC. Another receptor for C. 24 albicans is the epidermal growth factor receptor (EGFR), which interacts with candidal invasins 25 such as Als3 interact and induces epithelial cells to endocytose the fungus. Here, we investigated 26 the interactions between EGFR and EphA2. We found that EGFR and EphA2 constitutively 27 associated with each other as part of a physical complex. Activation of EGFR by C. albicans Als3 28 was required for sustained EphA2 phosphorylation and for induction of CXCL8/IL-8 and CCL20 29 secretion by epithelial cells. Treatment of uninfected epithelial cells with IL-17A and TNFα also 30 induced EGFR phosphorylation, which was necessary for epithelial cells to respond to these 31 cytokines. In mice with OPC, pharmacological inhibition of EGFR during caused a modest 32 reduction in oral fungal burden, markedly impaired production of proinflammatory cytokines and 33 significantly decreased accumulation of neutrophils and inflammatory monocytes. Thus, while C. 34 albicans activation of EGFR mediates fungal invasion of the epithelium, it also sustains EphA2 35 signaling, inducing the epithelial cell proinflammatory response to the fungus. 36 37 38

Introduction 53
The human oral cavity hosts a multifaceted microbiome comprised of an estimated 600 54 bacterial and 100 fungal species (1). Among these fungal species is Candida albicans, which 55 grows as a harmless commensal in at least 50% of healthy adults. When there is an imbalance 56 of local or systemic immune homeostasis, C. albicans can proliferate, causing oropharyngeal 57 candidiasis (OPC) (2). While this disease is relatively uncommon in healthy adults, it causes 58 significant morbidity in a large, diverse population of patients, including those with HIV/AIDS, 59 xerostomia, corticosteroid use, diabetes mellitus, and cancer of the head and neck. Each year, 60 there are nearly 10 million cases of OPC in patients with HIV/AIDS world-wide, and nearly one 61 fifth of these cases have esophageal involvement (3). 62 At least 80% of cases of OPC are caused by C. albicans. When this organism overgrows 63 in the oropharynx, there is invasion of the superficial epithelium, leading to host cell death (4). 64 One mechanism by which C. albicans invades oral epithelial cells is receptor mediated 65 endocytosis. In this process, invasins, such as Als3 and Ssa1 expressed on the surface of C. 66 albicans hyphae interact with epithelial cell receptors such as E-cadherin and a heterodimer 67 composed of the epidermal growth factor (EGFR) and HER2. This interaction triggers 68 rearrangement of the epithelial cell cytoskeleton, leading to the formation of pseudopods that 69 engulf the organism and pull it into the host cell (4-7). 70 The epithelial cells that line the oropharynx sense the presence of C. albicans and 71 orchestrate the host inflammatory response to fungal overgrowth. In addition to producing host 72 defense peptides that have direct antifungal activity, oral epithelial cells secrete alarmins, 73 proinflammatory cytokines, and chemokines that recruit phagocytes to foci of infection and 74 enhance their candidacidal activity to limit the growth of the invading fungi (8-10). This epithelial 75 cell response to OPC is amplified by interleukin (IL)-17, which is secreted by γδ T cells, innate 76 TCRαβ + cells, and type-3 innate lymphoid cells (11)(12)(13)(14). Recently we determined that the ephrin 77 type-A receptor 2 (EphA2) is expressed on oral epithelial cells and senses exposed β-glucan on 78 the fungal surface. When C. albicans proliferates on the epithelial cell surface, EphA2 is activated 79 and oral epithelial cells secrete host defense peptides and proinflammatory mediators. In mice 80 with OPC, EphA2 also induces the production of IL-17A, and EphA2 -/mice are highly susceptible 81 to OPC (15). 82 Although EphA2 is required for the normal host defense against OPC, exposure of oral 83 epithelial cells to purified β-glucan induces only transient EphA2 activation and is not sufficient to 84 initiate a significant inflammatory response. By contrast, exposure to live C. albicans induces 85 sustained EphA2 activation and a strong inflammatory response (15). In the current study, we 86 sought to elucidate how C. albicans infection prolongs EphA2 activation and induces a 87 proinflammatory response during OPC. We found that activation of host EGFR by C. albicans 88 Als3 maintained EphA2 phosphorylation and was required for the fungus to stimulate oral 89 epithelial cells to produce the chemokines, CXCL8/IL-8 and CCL20. Treatment of uninfected 90 epithelial cells with IL-17A and TNFα induced transient phosphorylation of EGFR, which was 91 necessary for epithelial cells to respond to IL-17A. In mice with OPC, pharmacological inhibition 92 of EGFR caused a modest reduction in oral fungal burden and a markedly impaired inflammatory 93 response. Thus, while C. albicans activation of EGFR mediates fungal invasion of the epithelium, 94 it is also plays a central role in inducing the local inflammatory response to this fungus. 95 96

97
The cellular fate of epithelial EphA2 differs depending on the type of stimulation. 98 When oral epithelial cells are infected with yeast-phase C. albicans, the organisms germinate and 99 begin to form hyphae within 60 min. Previously, we found that EphA2 is autophosphorylated on 100 serine 897 within 15 min of infection, and it remains phosphorylated for at least 90 min. By 101 contrast, when oral epithelial cells are exposed to β-glucan in the form of zymosan or laminarin, 102 EphA2 is phosphorylated for only the first 30 min of contact; at later time points, EphA2 103 phosphorylation returns to basal levels, even though β-glucan is still present (15). To investigate 104 how oral epithelial cells respond to prolonged EphA2 stimulation, we compared the response of 105 these cells to the native EphA2 ligand, ephrin A1 (EFNA1) and to C. albicans. After 15 min of 106 exposure, both stimuli induced phosphorylation of EphA2 ( Fig. 1A; S1). However, after 60 min, 107 the cells exposed to EFNA1 had minimal phosphorylation of EphA2, and total EphA2 levels had 108 declined ( Fig. 1B; S1). By contrast, the cells exposed to C. albicans had sustained EphA2 109 phosphorylation and no decrease in total EphA2 levels. Thus, exposure to C. albicans not only 110 induces EphA2 phosphorylation, but prevents subsequent EphA2 dephosphorylation and 111 degradation that normally occurs with prolonged exposure to its native ligand EFNA1 (16). 112

EGFR sustains C. albicans-induced EphA2 activation and constitutively interacts 113
with EphA2. Previously, we determined that EphA2 and EGFR function in the same pathway to 114 mediate the endocytosis of C. albicans by oral epithelial cells. Also, EphA2 is required for 115 C. albicans to activate EGFR because siRNA knockdown of EphA2 in oral epithelial cells blocks 116 phosphorylation of EGFR induced by the fungus (15). Cross talk between EphA2 and EGFR has 117 also been observed in other cell types (17,18). To investigate the nature of this cross-talk, we 118 tested the effects of inhibiting EGFR on C. albicans-induced EphA2 activation. When EGFR was 119 either knocked down with siRNA or inhibited with the specific EGFR kinase inhibitor, gefitinib (19), 120 EphA2 phosphorylation was transient, occurring within 30 min of infection, but declining to basal 121 levels by 90 min (Fig. 1C, D; S1). Therefore, activation of EGFR by C. albicans is required to 122 sustain EphA2 phosphorylation. 123 One potential mechanism for the cross-talk between EphA2 and EGFR is a physical 124 interaction between these two receptors. We investigated this possibility by immunoprecipitation 125 experiments. When EphA2 was immunoprecipitated from lysates of epithelial cells, EGFR was 126 pulled down; the amount of EGFR that was associated with EphA2 did not increase when the 127 epithelial cells were infected with C. albicans (Fig. 1E, F). In reciprocal experiments, we 128 determined that immunoprecipitation of EGFR from infected and uninfected epithelial cells pulled 129 a constant amount of EphA2 (Fig. 1E, F). These results indicate that EphA2 and EGFR 130 constitutively associate with each other, likely forming part of a complex that enables each 131 receptor to influence the activity of the other. 132

EGFR induces a subset of EphA2-mediated pro-inflammatory responses in oral 133
epithelial cells. The physical association of EphA2 with EGFR suggested the possibility that 134 EGFR may also induce the epithelial cell proinflammatory response to C. albicans infection. To 135 test this hypothesis, we compared the effects of siRNA knockdown of EGFR and EphA2 on 136 cytokine production in oral epithelial cells that were infected with C. albicans. We found that 137 knockdown of EGFR inhibited the secretion of CXCL8/IL-8 and CCL20 similarly to knockdown of 138 EphA2 ( Fig. 2A). Although knockdown of EGFR had no effect on the release of IL-1α, and IL-1β, 139 knockdown of EphA2 blocked both of these cytokines. Treatment of the epithelial cells with 140 gefitinib had a similar effect to siRNA knockdown of EGFR, resulting in inhibition of CXCL8/IL-8 141 and CCL20 secretion but no reduction of IL-1α, and IL-1β release (Fig. 2B). These results suggest 142 that EGFR governs a subset of the epithelial cell inflammatory response to C. albicans infection. with the als3Δ/Δ null mutant also caused significantly less secretion of CXCL8/IL-8, CCL20, IL-1α, 151 and IL-1β (Fig. 3D). Collectively, these results indicate that C. albicans Als3 activates EGFR, 152 which prolongs EphA2 phosphorylation and stimulates oral epithelial cells to secrete chemokines 153 and pro-inflammatory cytokines. 154 Inhibition of EGFR impairs the host inflammatory response during OPC. To further 155 assess the role of EGFR in mediating the host inflammatory response to C. albicans during OPC, 156 we treated immunocompetent mice with gefitinib (20, 21) and then orally inoculated them with 157 C. albicans. Using immunostaining with phosphospecific antibodies, we assessed the effects of 158 gefitinib on EGFR and EphA2 phosphorylation in the oral epithelium. Similar to its effects in vitro, 159 treatment with gefitinib reduced the phosphorylation of both EGFR and EphA2 after 1 d of infection 160 ( Fig. 4A, B). At this time point, the gefitinib-treated mice had a 3-fold reduction in oral fungal 161 burden compared to untreated mice (Fig. 4C). The oral tissues of gefitinib-treated mice also 162 contained significantly less IL-1α, IL-1β, CXCL1/KC, IL-17A, and the p19 subunit of IL-23 than the 163 control mice (Fig. 5A). By contrast, the levels of CCL2, CCL3, CCL4, and the host defense peptide 164 S100A8 in the gefitinib treated mice were not significantly different from the control mice, 165 indicating that gefitinib selectively inhibited a subset of the host inflammatory response. Treatment 166 with gefitinib also caused a dramatic reduction in the number of neutrophils and inflammatory 167 monocytes in the oral tissues relative to the mice ( Fig. 5B, C, S4). Thus, in immunocompetent 168 mice, gefitinib inhibits C. albicans-induced phosphorylation of EGFR and EphA2, thereby reducing 169 both oral fungal burden and the host inflammatory response. 170 Because gefitinib impaired the host inflammatory response to C. albicans, we investigated 171 its effects on the candidacidal activity of neutrophils and macrophages. We found that gefitinib 172 did not decrease the capacity of human neutrophils or mouse bone marrow derived macrophages 173 to kill C. albicans in vitro (Fig. S5). Taken together, these results indicate that while EGFR 174 signaling is required for epithelial cells to mount a pro-inflammatory response to C. albicans, it is 175 dispensable for governing phagocyte killing of this organism. 176 EGFR activity is required for the epithelial cell response to IL-17A. Many of the 177 chemokines that were suppressed by gefitinib treatment are known to be induced by IL-17A, a 178 cytokine that plays a central role in the host defense against OPC (22-24). We therefore 179 investigated whether EGFR is required for oral epithelial cells to respond to IL-17A. When oral 180 epithelial cells were stimulated with IL-17A and TNFα in the absence of C. albicans, EGFR was 181 autophosphorylated for 60 min, after which phosphorylation returned to basal levels ( Fig. 6A, 182 Oral epithelial cells play a central role in orchestrating the host defense against C. albicans 189 during OPC (2, 4, 25). Here, we found that EphA2 and EGFR form a complex in oral epithelial 190 cells. When these cells are infected with C. albicans, EphA2 and EGFR functionally interact to 191 induce a pro-inflammatory response. Previously, we found that EphA2 is required for C. albicans 192 to stimulate EGFR (15). Here, we determined that EGFR is in turn necessary for C. albicans to 193 induce prolonged activation of EphA2. This finding suggests that the initiation of an antifungal 194 proinflammatory response in the oral cavity requires two signals, one induced by EphA2 binding 195 to fungal β-glucans and the other induced by EGFR interacting with fungal invasins. 196 EphA2 and EGFR are known to interact in epithelial cell cancers, especially those that 197 have become resistant to EGFR inhibitors (26,27). In these cells, siRNA knockdown of EphA2 198 restores sensitivity to EGFR inhibition. Interestingly, treatment of these malignant cell lines with 199 soluble EFNA1 has the same effect as EphA2 siRNA, presumably because EFNA1 induces 200 EphA2 endocytosis and subsequent degradation. Our data indicate that C. albicans activates 201 EphA2 differently than EFNA1 because binding of C. albicans stabilizes EphA2 and prevents its 202 degradation. The sustained EphA2 protein levels likely contribute to the prolonged EphA2 203 signaling induced by C. albicans infection. 204 Functioning as a receptor for fungal β glucans, EphA2 initiates the epithelial cell production 205 of chemokines and pro-inflammatory cytokines in response to C. albicans overgrowth (15). The 206 finding that EphA2 and EGFR interact prompted us to investigate whether EGFR also mediates 207 the production of pro-inflammatory mediators. Indeed, we found that knockdown of EGFR with 208 siRNA or inhibition of EGFR with gefitinib significantly reduced C. albicans-induced production of 209 CXCL8/IL-8 and CCL20 in vitro. Infection with the als3Δ/Δ mutant also induced less chemokine 210 secretion by oral epithelial cells, suggesting that activation of EGFR is necessary for secretion of 211 CXCL8/IL-8 and CCL20. However, gefitinib treatment had no effect on epithelial cell release of 212 IL-1α and IL-1β, whereas infection with the als3Δ/Δ mutant decreased the release of these 213 cytokines, an effect that was similar to siRNA knockdown of EphA2 (15). A potential explanation 214 for this result is that treatment with gefitinib caused a 20%±8% increase in epithelial cell damage 215 due to the wild-type strain (n=9, p < 0.0001 by the Student's t-test), whereas infection with the 216 als3Δ/Δ null mutant caused significantly less damage to oral epithelial cells relative to the wild-217 type strain (6). Similarly, siRNA knockdown of EphA2 reduced the extent of C. albicans induced 218 epithelial cell damage(15). Thus, we speculate that while the secretion of CXCL8/IL-8 and CCL20 219 in response to C. albicans is induced by activation of EGFR, the release of IL-1α and IL-1β is 220 stimulated by epithelial cell damage. 221 The central role of EGFR in the host inflammatory response was also demonstrated in the 222 mouse model of OPC, where treatment with gefitinib inhibited phosphorylation of both EphA2 and 223 EGFR, and reduced the tissue levels of CXCL1/KC, IL-1α, IL-1β, IL-17A, and IL-23p19. As a 224 result, the accumulation of neutrophils and inflammatory monocytes in the oral tissues was 225 dramatically decreased. It was notable that treatment with gefitinib only inhibited a subset of the 226 inflammatory mediators induced by C. albicans infection; the tissue levels of CCL2, CCL3, CCL4 227 and S100A8 in the gefitinib treated mice were not significantly different from control mice. This 228 results suggests that EGFR activation is require for the production of a subset of inflammatory 229 mediators in vivo and that other inflammatory mediators are induced by an EGFR-independent 230 pathway. EGFR has previously been found to be important for the production of CXCL8/IL-8 and 231 CXCL10 by pulmonary epithelial cells infected with influenza and rhinovirus. In contrast to 232 C. albicans, which appears to activate EGFR directly (5), these viruses activate EGFR indirectly 233 by stimulating a metalloproteinase that cleaves an EGFR proligand that in turn binds to EGFR 234 (28,29). 235 We observed that stimulation of uninfected epithelial cells with IL-17A and TNFα induced 236 the phosphorylation of EGFR. Moreover, gefitinib blocked the release of CXCL8/IL-8, CCL20, 237 and GM-CSF by uninfected cells that had been stimulated with IL-17A and TNFα. These data 238 indicate that EGFR phosphorylation is required for at least some of the epithelial cell responses 239 to IL-17A stimulation. Lee et al. (30) found similarly that in a colonic epithelial cell line, the 240 combination of IL-17A and TNFα induces EGFR phosphorylation and that treatment with an 241 EGFR kinase inhibitor decrease IL-17A-induced release of CXCL8/IL-8 and CXCL10. These 242 authors suggested that EGFR potentiates and prolongs ERK signaling induced by IL-17A, leading 243 to chemokine secretion. 244 Our current results combined with our previous data (15) suggest a more nuanced model 245 for how oral epithelial cells respond to C. albicans overgrowth during OPC (Fig 7). In this model, 246 C. albicans β-glucans initially activate EphA2. Subsequently, when C. albicans forms hyphae, it 247 expresses invasins such as Als3 that activate EGFR and sustain EphA2 activation, leading to 248 activation of the MEK1/2, c-Fos, and STAT3 signaling pathways, ultimately resulting in the release 249 of proinflammatory mediators by the infected epithelial cells. EGFR is also activated by the IL-17A 250 that is produced by intraepithelial lymphocytes, leading to further amplification of the epithelial cell 251 proinflammatory response. The overall result is the secretion of chemokines, proinflammatory 252 cytokines, and host defense peptides as well as recruitment of phagocytes to the focus of 253 infection, leading to inhibition and eventual killing of the fungus. 254 While receptor-mediated induction of the proinflammatory response is likely beneficial to 255 the host, activation of EphA2 and EGFR may also be beneficial to the fungus. The interaction of 256 EphA2 and EGFR with C .albicans activates the clathrin-dependent endocytosis pathway in the 257 epithelial cells, leading to rearrangement of the actin cytoskeleton and formation of pseudopods 258 that engulf C. albicans and pull it into the epithelial cell (31) . This process contributes to the 259 pathogenicity of the fungus because the internalized organism is hidden from phagocytes and 260 can utilize the epithelial cell as a source of nutrients (32). 261 This model explains the modest effect of gefitinib on oral fungal burden during OPC. 262 Although EGFR inhibition impaired the host inflammatory response to C. albicans, this negative Immunoblotting. OKF6/TERT-2 cells in 24-well tissue culture plates were switched to 295 KSF medium without supplements for 1 h and then infected with 1 × 10 6 C. albicans yeast or 296 incubated with IL-17A and TNFα for various times as described previously (15). Next, the cells 297 were rinsed with cold HBSS containing protease and phosphatase inhibitors and detached from 298 the plate with a cell scraper. After the cells were collected by centrifugation, they were boiled in 299 sample buffer. The lysates were separated by SDS-PAGE, and phosphorylation was detected 300 by immunoblotting with specific antibodies against pEphA2 (#6347, Cell Signaling) and pEGFR 301 (#2234, Cell Signaling). Next, the blot was stripped, and the total amount of each protein was 302 detected by immunoblotting with antibodies against EphA2 (D4A2, Cell Signaling) and EGFR 303 (#4267, Cell Signaling). Each experiment was performed at least 3 times. 304 Co-immunoprecipitation. OKF6/TERT-2 cells were grown in 75 cm 2 flasks to confluency 305 and then switched to KSF medium without supplements for 3 h and then infected with 1x10 8 306 C. albicans yeast. After 30 or 90 min. OKF6/TERT-2 were washed with ice-cold cold PBS (with 307 Mg 2+ , and Ca 2+ ), scraped, and lysed with 100 µl ice-cold 5.8% octyl β-D-glucopyranoside (0479-308 5g; VWR) in the present of protease/phosphatase inhibitors. Whole cells lysates were precleared 309 with 20µl of protein A/G plus (sc-2003; Santa Cruz Biotechnology) at 4°C for 30minutes. Bead-310 protein mix was centrifuged at 3000rpm for 30 sec at 4°C and supernatants were collected. 2 μg 311 of anti-EGFR antibody (sc-101; Santa Cruz Biotechnology), or anti-EphA2 antibody (#6347, Cell 312 Signaling) respectively, was added to 500 µg of proteins, and incubated on a rotator at 4°C for 2 313 hours. 25µl of protein A/G plus was added to each immunoprecipitation sample and incubated for 314 an additional hour at 4°C. Samples were pelleted at 3000 rpm for 30 sec, and washed 3 times in 315 500 µl of ice-cold 1.5% octyl β-D-glucopyranoside. Proteins were eluted with 30 µl of 2X SDS 316 buffer, and heated at 90°C for 5 minutes. Samples were centrifuged at 3000 rpm for 30 sec, and 317 supernatants were collected, and separated by SDS-PAGE, and analyzed as described above. 318 Measurement of epithelial cell endocytosis. The endocytosis of C. albicans by oral 319 epithelial cells was quantified as described previously (36). OKF6/TERT-2 oral epithelial cells 320 were grown to confluency on fibronectin-coated circular glass coverslips in 24-well tissue culture 321 plates and then infected for 120 min with 2 × 10 5 yeast-phase C. albicans cells per well, after 322 which they were fixed, stained, and mounted inverted on microscope slides. The coverslips were 323 viewed with an epifluorescence microscope, and the number of endocytosed organisms per high-324 power field was determined, counting at least 100 organisms per coverslip. Each experiment was 325 performed at least 3 times in triplicate. 326

Cytokine and chemokine measurements in vitro.
Cytokine levels in culture 327 supernatants were determine as previously described (15). Briefly OKF6/TERT-2 cells in a 96-328 well plate infected with C. albicans at a multiplicity of infection of 5. After 8 h of infection, the 329 medium above the cells was collected, clarified by centrifugation and stored in aliquots at -80 °C. 330 The concentration of inflammatory cytokines and chemokines in the medium was determined 331 using the Luminex multipex assay (R&D Systems). weighed and homogenized. The homogenates were cleared by centrifugation and the 355 concentration of inflammatory mediators was measured using a multiplex bead array assay (R&D 356 Systems) as previously described (15,37). 357 Flow cytometry of infiltrating leukocytes. The number of phagocytes in the mouse 358 tongues were characterized as described elsewhere (38). Briefly, mice were orally infected with 359 C. albicans as described above. After 1 d of infection, the animals were administered a sublethal 360 anesthetic mix intraperitoneally. The thorax was opened, and a part of the rib cage removed to 361 gain access to the heart. The vena cava was transected and the blood was flushed from the 362 vasculature by slowly injecting 10 mL PBS into the right ventricle. The tongue was harvested and 363 cut into small pieces in 100 µL of ice-cold PBS. 1 mL digestion mix (4.8 mg/ml Collagenase IV; 364 Worthington Biochem, and 200 μg/ml DNase I; Roche Diagnostics, in 1x PBS) was added after 365 which the tissue was incubated at 37°C for 45 min. The resulting tissue suspension was then 366 passed through a 100 μm cell strainer. The single-cell suspensions were incubated with rat anti-367 mouse CD16/32 (2.4G2; BD Biosciences) for 10 min in FACS buffer at 4°C to block Fc receptors. The stained cells were analyzed on a 2-laser LSRII flow cytometer (BD Biosciences), and the 374 data were analyzed using FACS Diva (BD Biosciences) and FlowJo software (Treestar). Only 375 single cells were analyzed, and cell numbers were quantified using PE-conjugated fluorescent 376 counting beads (Spherotech). 377 Phagocyte killing assay. The effects of gefitinib on neutrophil killing of C. albicans were 378 determined by our previously described method (20). Briefly, neutrophils were isolated from the 379 blood of healthy volunteers and incubated with gefitinib or diluent in RPMI 1640 medium plus 10% 380 fetal bovine serum for 1 h at 37°C. Next, the neutrophils were mixed with an equal number of 381 serum-opsonized C. albicans cells. After a 3 h incubation, the neutrophils were lysed by 382 sonication, and the number of viable C. albicans cells was determined by quantitative culture. 383 To obtain bone marrow-derived macrophages, bone marrow cells from BALB/c mice 384 (Taconics) were flushed from femurs and tibias using sterile RPMI 1640 medium supplemented 385 with 10% fetal bovine serum (FBS) and 2 mM EDTA onto a 50 ml screw top Falcon tube fitted 386 with a 100 μm filter (39). 6x10 6 bone marrow cells per 75 cm 2 were seeded in RPMI 1640 387 supplemented with 20% FBS, 100 μg/ml streptomycin, 100 U/ml penicillin, 2 mM Glutamine, and 388 25 ng/ml rHu M-CSF (PeproTech). After 7 days, the bone marrow derived macrophages 389 (BMDMs) were treated with gefitinib or the diluent and then incubated with serum-opsonized 390 C. albicans cells (multiplicity of infection 1:20) for 3 h. Next, the BMDMs were scraped, lysed by 391 sonication, and the number of viable C. albicans cells was determined by quantitative culture. 392 Indirect Immunofluorescence. To detect phosphorylation of EphA2, and EGFR in the 393 tongue of C. albicans infected mice, 2-μm-thick sections of OCT-embedded organs were cut with 394 a microtome and then fixed with cold acetone. Next, the cryosections were rehydrated in PBS 395 and then blocked. They were stained with primary antibodies against phosphorylated EphA2   C. albicans; ctrl, control; GEF, Gefitinib; UNINF, uninfected). Statistical significance is indicated 557 by *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (Mann-Whitney test with Bonferroni 558 correction for multiple comparisons). 559 infection. Results are median ± interquartile range of at total of 14 mice per group from two 578 independent experiments. The y-axis is set at the limit of detection (100 CFU/g tissue). Statistical 579 significance is indicated by *P < 0.05, **P < 0.01 (Mann-Whitney test). 580 show median and interquartile range of a total of seven mice in each group from two independent 584 experiments. Statistical significance is indicated by *P < 0.05, **P < 0.01 (Mann-Whitney test with 585 Bonferroni correction for multiple comparisons). 586 C. albicans interacts with EphA2 and EGFR on oral epithelial cells, there is prolonged 597 phosphorylation of EphA2, which activates the STAT3 and mitogen-activated protein kinase 598 (MAPK) pathways, leading to the production of chemokines, proinflammatory cytokines, and host 599 defense peptides. The chemokines recruit proinflammatory monocytes and neutrophils to the 600 focus of infection. Activation of EphA2 and EGFR also causes the epithelial cells to endocytose 601 C. albicans, which subsequently causes epithelial cell damage and leads to the release of IL-1α 602 and IL-1β. In response to C. albicans infection, leukocytes such as γδ T cells, innate TCRαβ + 603 cells, and type-3 innate lymphoid cells secrete IL-17A, which requires EGFR activation to amplify 604 the epithelial inflammatory response. TNFα for the indicated times (min). Data are the mean ± SD of 3 independent immunoblots. An 642 images of a representative immunoblot is show in Fig. 6A. Statistical significance is indicated by 643 *P < 0.05, **P < 0.01 (two-tailed Student's t-test assuming unequal variances). 644