Targeting host tyrosine kinase receptor EphA2 signaling via small-molecule ALW-II-41-27 inhibits macrophage pro-inflammatory signaling responses to Pneumocystis carinii β-glucans

ABSTRACT

Pneumocystis jirovecii, the fungus that causes Pneumocystis jirovecii pneumonia (PJP), is a leading cause of morbidity and mortality in immunocompromised individuals. We have previously shown that lung epithelial cells can bind Pneumocystis spp. β-glucans via the EphA2 receptor, resulting in activation and release of proinflammatory cytokines. Herein, we show that in vivo Pneumocystis spp. β-glucans activation of the inflammatory signaling cascade in macrophages can be pharmacodynamically inhibited with the EphA2 receptor small-molecule inhibitor ALW-II-41-27. In vitro, when ALW-II-41-27 is administrated via intraperitoneal to mice prior to the administration of highly proinflammatory Saccharomyces cerevisiae β-glucans in the lung, a significant reduction in TNF-alpha release was noted in the ALW-II-41-27 pre-treated group. Taken together, our data suggest that targeting host lung macrophage activation via EphA2 receptor-fungal β-glucans interactions with ALW-II-41-27 or other EphA2 receptor kinase targeting inhibitors might be an attractive and viable strategy to reduce detrimental lung inflammation associated with PJP.

INTRODUCTION

EphA2 receptor is a member of the receptor tyrosine kinase family. It is a transmembrane receptor containing an extracellular region that binds activating ligands (ephrins) and more recently discovered to bind fungal β-glucans (1–3), leading to the activation of the intracellular tyrosine kinase domain (4). Others have shown the specificity of EphA2 receptor for fungal β-glucans and activation (phosphorylation) via binding of this receptor to zymosan-coated beads as well as Candida albicans, Aspergillus fumigatus, and Rhizopus delemar fungal organisms and the absence of binding and phosphorylation of the receptor by Staphylococcus aureus and Escherichia coli bacteria (3). More recently, our lab has shown that purified recombinant EphA2 protein alone can specifically and significantly bind Pneumocystis β-glucans, verifying that the protein is a receptor for fungal β-glucans (1). Traditionally, the EphA2 receptor kinase pathway has important roles in carcinogenesis, pathological angiogenesis, and inflammation in atherosclerosis (5–7). Expression of EphA2 receptor is high in both epithelial and endothelial cells (4). More recently, this receptor pathway has also emerged as an important regulatory pathway for host defense against microbial pathogens, including bacterial, viral, and fungal pathogens (2, 8–11). For example, it has been demonstrated previously that the EphA2 receptor is active in the binding and trapping of the hookworm Nippostrongylus brasiliensis by bone marrow-derived macrophages, suggesting a role for the EphA2 receptor in macrophage/microbial pathogenesis (12).

ALW-II-41-27 is a small-molecule inhibitor that has been demonstrated to selectively bind to the ATP-binding pocket of the EphA2 receptor kinase domain (13). Although the compound has been shown to have EC₅₀ values on a number of kinases in vitro at 10 uM (including CSF1R, DDR1/2, Kit, Lck, and PDGFRα/β) (14), this and other recent studies show EC₅₀ effects of the inhibitor at 200 nM or less on EphA2 kinase activity, allowing more directed and targeted therapeutic dosing (14–16). The inhibitor has been used in the past to inhibit cancer cell growth in vitro and in vivo (17–20) and more recently to inhibit uropathogenic bacteria adherence to bladder epithelial cells and to reduce the oxidative stress and proinflammatory host response in LPS(lipopolysaccharide)-treated colonic cells (16). Therefore, we evaluated this inhibitor to determine whether it might serve as an alternative agent to reduce proinflammatory events resulting from interactions with Pneumocystis β-glucans through the EphA2 receptor kinase signaling pathway in macrophages in vitro (1), as well as the inhibitor’s ability to reduce yeast β-glucan-driven proinflammatory cytokine release in the lung. Measured and timed therapeutic inhibition of EphA2 signaling may aid in mitigating harmful downstream inflammatory events that result during anti-Pneumocystis pneumonia (PCP) treatment. We and others have shown that killing of fungal organisms as a result of this action exposes highly inflammatory β-glucan carbohydrate (21–29).

Cells [2 × 10⁵ RAW macrophages (American Type Culture Collection)] for each experimental condition were plated in five wells of a 96-well microtiter plate and incubated for 4 hours. After 4 hours, ALW-II-41-27 purchased from Sigma-Aldrich was pre-incubated with the RAW cells for 60 minutes. Next, 100 ug/mL of Pneumocystis carinii (Pc) β-glucans (28) was added to the wells, and the plates were centrifuged at 500 × g to synchronize the carbohydrate/macrophage interactions. Plates were then placed at 37°C for 60 minutes. Next, cells were washed with 1× PBS, lysed, and protein quantification determined. Total proteins in equal amounts were loaded and separated by polyacrylamide gel electrophoresis. Finally, proteins were transferred to nylon for Western blotting and incubated with antibodies to phospho-p38 or ERK1/2 as well as total p38 and ERK1/2 (Cell Signaling Technology) to demonstrate equal loading control. Protein phosphorylation kinetics were quantified with Image Studio Lite (LI-COR). All experiments were repeated four to five times. Activation of MAPK (mitogen-activated protein kinase) is well documented in macrophage responses to Pneumocystis infection (29–31). Herein, we demonstrate that the specific EphA2 receptor inhibitor ALW-II-41-27 can indeed significantly inhibit Pc β-glucan-induced phosphorylation of both p38 (Fig. 1A) and ERK1/2 (Fig. 2) in a dose-dependent manner measured by Western blot with similar ALW-II-41-27 concentrations previously published for cervical, endometrial, nasopharyngeal, and colonic cells inhibitor studies (13, 15, 16, 32). Western blots for both phospho-p38 (Fig. 1B) and phospho-ERK1/2 (Fig. 2B) were quantified by densitometry analysis against their respective total proteins. Next, we wanted to determine whether ALW-II-41-27 inhibition would not only result in decreased MAPK phosphorylation but also result in downstream reduced secretion of the proinflammatory cytokine TNF-alpha. To determine this, RAW macrophages were seeded as above. After 4 hours, ALW-II-41-27 was incubated with the RAW cells for 60 minutes. Cell media were removed, and 100 ug/mL of Pc β-glucans plus ALW-II-41-27 compound was applied to the cells, centrifuged as described above, and incubated for 18 hours at 37°C. Supernatants were then collected and assayed for TNF-alpha by ELISA (26). As shown in Fig. 3A, ALW-II-41-27 significantly reduced TNF-alpha release in a dose-dependent fashion. To determine if ALW-II-41-27 could significantly alter the inflammatory potential of macrophages already undergoing activated proinflammatory cytokine release via Pc β-glucan stimulation, we also added ALW-II-41-27 post 60 minutes after Pc β-glucan stimulation. Similar to Fig. 3A, we also noted significant suppression of RAW cell TNF-alpha release in a dose response fashion in these experiments (Fig. 3B). Next, to confirm these findings in primary cells, mouse lung alveolar macrophages were isolated as previously described (31). After allowing the macrophages to bind for 2 hours, ALW-II-41-27 was incubated with the macrophages for 60 minutes. As stated above, the supernatant was removed, and Pc β-glucans (100 ug/mL) plus ALW-II-41-27 compound was added to the cells, centrifuged as described above, and incubated for 18 hours at 37°C . Supernatants were then collected and assayed for TNF-alpha by ELISA as noted above. Similar to what was shown in Fig. 3, ALW-II-41-27 addition to Pc β-glucan-induced primary mouse lung alveolar macrophages significantly reduced TNF-alpha cytokine release from these cells (Fig. 4), confirming the ability of the compound to inhibit proinflammatory response in native lung alveolar macrophages. To determine if ALW-II-41-27 could also affect TNF-alpha release from RAW macrophages infected with live P. murina organisms, 2 × 10⁵ cells were plated in duplicate wells of a 96-well plate for 4 hours as above. Next, ALW-II-41-27 was incubated with the RAW cells for 60 minutes. Finally, P. murina organisms in the presence of ALW-II-41-27 were applied at a multiplicity of infection of 2:1 to the cell supernatant, and the plates centrifuged at 500 × g to synchronize the fungal organism/RAW cell interactions and incubated for 18 hours at 37°C. Supernatants were then collected and tested for TNF-alpha release. Similar to what we reported above for Pc β-glucan alone, TNF-alpha was significantly induced in the presence of live fungal organisms. The addition of 1,000 nM of ALW-II-41-27 to the media prior to the addition of the fungal organisms significantly suppressed TNF-alpha release from the cultured macrophages (Fig. 5). These results suggest the exciting possibility of targeting the EphA2 tyrosine kinase receptor pathway in those individuals with active PCP to reduce exuberant lung inflammation. On this note, as an initial proof-of-concept experiment to determine if ALW-II-41-27 administered to mice could inhibit fungal β-glucan-driven proinflammatory cytokine response in the lung, we pre-treated mice with intraperitoneal (IP) injections of either ALW-II-41-27 or the vehicle control 20 hours prior to the addition of Saccharomyces cerevisiae β-glucans. After 20 hours, the yeast β-glucans were administered via intratracheally (IT). The following day (24 hours), the mice were sacrificed, and total lung protein lysates measured for TNF-alpha by ELISA. Remarkably, we noted that ALW-II-41-27 could indeed significantly reduce TNF-alpha protein levels in the lungs versus the vehicle control group in yeast β-glucan-challenged mouse lungs (Fig. 6).

Fig 1.

(A) ALW-II-41-27 can significantly inhibit Pc β-glucan-induced ERK1/2 phosphorylation in a dose-dependent manner. Representative blot of four separate experiments. (B) The phospho ERK1/2 signals were quantified with Image Studio Lite software and normalized to total ERK1/2 levels. Initial analysis was first performed with ANOVA. If ANOVA indicated overall differences, subsequent group analysis was then performed by a two-sample unpaired Student t test for normally distributed variables. Error bars show SD from the mean. *P < 0.05, ns, non-significant; ANOVA, analysis of variance.

Fig 2.

(A) ALW-II-41-27 can significantly inhibit Pc β-glucan-induced p38 phosphorylation. Representative blot of five separate experiments. (B) The phospho p38 signals were quantified with Image Studio Lite software and normalized to total p38 levels. Similar to phospho ERK1/2 quantitation above, analysis of phospho p38 quantification by ANOVA was first performed. If ANOVA indicated overall differences, subsequent group analysis was then performed by a two-sample unpaired Student t test for normally distributed variables. Error bars show SD from the mean. ****P < 0.0001. ANOVA, analysis of variance.

Fig 3.

ALW-II-41-27 administered 60 minutes prior to (A) or 60 minutes after (B) Pc β-glucans significantly dampens RAW 264.7 production of TNF-alpha in vitro. Data are the ±SEM for at least four separate experiments. The TNF-alpha data analysis was initially first performed with ANOVA. If ANOVA indicated overall differences, subsequent group analysis was then performed by a two-sample unpaired Student t test for normally distributed variables. Error bars show SD from the mean. *P < 0.05, **P < 0.01, and ****P < 0.0001. ANOVA, analysis of variance.

Fig 4.

In primary mouse lung alveolar macrophages, ALW-II-41-27 significantly dampens production of TNF-alpha in vitro in the presence of Pc β-glucans. Data are the ±SEM for at least three separate experiments. Initial analysis was first performed with ANOVA. If ANOVA indicated overall differences, subsequent group analysis was then performed by a two-sample unpaired Student t test for normally distributed variables. Error bars show SD from the mean. *P < 0.05, and ****P < 0.0001. ANOVA, analysis of variance.

Fig 5.

ALW-II-41-27 can significantly inhibit P. murina-induced TNF-alpha secretion in mouse macrophages. Data represent the mean ± SEM for at least three separate experiments. Initial analysis was first performed with ANOVA. If ANOVA indicated overall differences, subsequent group analysis was then performed by a two-sample unpaired Student t-test for normally distributed variables. Error bars depict the SD from the mean. *P < 0.05 and **P < 0.01. ANOVA, analysis of variance.

Fig 6.

The effects of IP injection of ALW-II-41-27 on Saccharomyces cerevisiae β-glucan proinflammatory response in the lung. Twenty hours prior to administering 100 ug/mL of S. cerevisiae β-glucans via IT, mice were administered 0.1 mg/kg ALW-II-41-27 or the vehicle (Methocel) control via IP. After 18 hours post drug or vehicle treatment, mice were administered another 0.1 mg/kg of ALW-II-41-27 via IP or vehicle stated. After 2 hours, mice were administered 100 ug/mL of S. cerevisiae β-glucans via IT administration. The following day, mice were sacrificed, and total lung protein lysates (200 ug total) measured for TNF-alpha by ELISA. Bar graph represents the results from 11 to 12 mice per group. If ANOVA indicated overall differences, subsequent group analysis was then performed by two-sample unpaired Student t test for normally distributed variables. Error bars show SD from the mean. ****P < 0.0001. ANOVA, analysis of variance.

A number of reports have shown the importance of the EphA2 receptor pathway in organism attachment and host immune recognition to microbial pathogens (12, 33–35). Recently, we also have reported that EphA2 can bind Pneumocystis glucans and is involved in lung epithelial cell proinflammatory response to the organism’s cell wall carbohydrate (1).

In vitro and in vivo data presented here suggest promising preliminary evidence that therapeutically targeting the EphA2 receptor in PCP-infected individuals undergoing standard anti-PCP treatment may provide additional anti-inflammatory relief as a result of fungal killing and the release of highly proinflammatory β-glucans during the treatment of Pneumocystis pneumonia.