Lactic Acid Suppresses IL-33–Mediated Mast Cell Inflammatory Responses via Hypoxia-Inducible Factor-1α–Dependent miR-155 Suppression

Mast cells are sentinels of the innate immune system, guarding the body against selected bacterial and parasitic infections. However, mast cells are best known for the major role they play in allergies and allergic asthma. The interaction of allergen with mast cell–bound IgE and subsequent signaling through the IgE receptor, FcεRI, result in a signaling cascade provoking release of early and late phase mediators (1–5). The early phase mediators, released within minutes of activation, include tryptases, chymases, histamine, PGs, leukotrienes, and platelet-activating factor, whereas the late phase mediators, released hours later, consist of cytokines and chemokines, including IL-1β, IL-4, IL-5, IL-6, IL-10, IL-13, TNF, MIP-1α, and MCP-1. These factors yield the clinical symptoms of immediate hypersensitivity, including the wheal-and-flare response, itching, and vasodilation/edema. Mast cells can also promote chronic diseases such as asthma upon repeated Ag exposure, which results in airway remodeling due to sustained inflammation (1–5).

Although IgE crosslinking is the best studied form of mast cell activation, many stimuli elicit a mast cell response. IL-33 is a recently discovered alarmin in the IL-1 family. Produced by endothelial cells, epithelial cells, fibroblasts, mast cells, and keratinocytes in response to damage or stress, IL-33 promotes a Th2 response (6–9). Binding to the ST2/IL-1RacP receptor on mast cells results in the release of cytokines, chemokines, and lipid mediators (6, 7). IL-33 has also been shown to promote mast cell survival, maturation, and adhesion (8, 10). Although it is a poor inducer of degranulation, IL-33 augments degranulation triggered through the IgE receptor (6, 7). IL-33 has beneficial effects in atherosclerosis, cardiac remodeling, and helminth infection (8). However, it has been linked to asthma, rheumatoid arthritis, multiple sclerosis, type I diabetes, and skin inflammation (7).

Inflammation causes important changes in the cellular microenvironment, which can be beneficial if temporally and spatially controlled, but pathological when chronic. A well-known example of chronic inflammation is the tumor microenvironment. Tumors are known to preferentially undergo anaerobic glycolysis, even in the presence of sufficient oxygen, resulting in a hypoxic microenvironment with high (40 mM) lactic acid (LA) concentrations (11–19). These unique environmental factors can alter cellular responses, allowing tumors to escape immune surveillance (16, 19–21). There is evidence that the tumor microenvironment uses LA to promote tumor-associated macrophages to take on an M2 phenotype, which is anti-inflammatory and has reduced Ag presentation ability (12, 19). Tumor-derived LA has also been shown to inhibit dendritic cell function, resulting in decreased proliferation and reduced Ag presentation (14). Among cytotoxic T cells, LA decreases proliferation and inhibits their cytotoxic function (16). Other examples of altered microenvironments that have increased lactate levels are obesity, hypertension, and type II diabetes, as well as tissues suffering injury, infection, or ischemia (20–23).

Because LA is known to alter cellular responses in inflammatory environments, we tested the effects of physiological LA concentrations on mast cell function. LA exposure for 24 h was sufficient to suppress IL-33–mediated cytokine secretion, effects that were pH- and monocarboxylate transporter (MCT)-1–dependent. Suppression of IL-33–induced inflammatory cytokine and chemokine secretion was accompanied by reduced activation of several signaling intermediates. Expectedly, LA treatment enhanced hypoxia-inducible factor (HIF)-1α expression. This correlated with a decrease in proinflammatory miR-155-5p, which was reversed by HIF-1α blockade. miR-155-5p overexpression abolished LA suppressive effects, demonstrating the critical nature of the HIF-1α–miR-155 cascade. These results provide insight into microenvironmental effects during wound healing, chronic inflammation, and tumorigenesis.

C57BL/6 male and female mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and used at a minimum of 6 wk old, with approval from the Virginia Commonwealth University Institutional Animal Care and Use Committee.

Mouse bone marrow–derived mast cells (BMMC) were derived by harvesting bone marrow from C57BL/6 mouse femurs, followed by culture in complete RPMI 1640 (cRPMI) medium (Invitrogen Life Technologies, Carlsbad, CA) containing 10% FBS, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 1 mM sodium pyruvate, and 1 mM HEPES (all from Corning, Corning, NY), supplemented with IL-3–containing supernatant from WEHI-3B cells and stem cell factor (SCF)–containing supernatant from BHK-MKL cells. The final concentrations of IL-3 and SCF were adjusted to 1.5 and 15 ng/ml, respectively, as measured by ELISA. BMMC were used after 3 wk of culture, at which point these primary populations were >90% mast cells, based on staining for c-Kit and FcεRI expression. Mouse peritoneal mast cells were obtained by collecting peritoneal lavage from C57BL/6 mice, which was cultured in cRPMI supplemented with recombinant mouse IL-3 and SCF at 10 ng/ml each. Peritoneal mast cells were used after 10–12 d, at which point these ex vivo–expanded cells were ∼85% mast cells, based on staining for c-Kit and FcεRI expression.

Recombinant mouse IL-3, SCF, and IL-33, recombinant human IL-33, as well as mouse IL-6, TNF, and MCP-1 (CCL-2) ELISA kits were purchased from BioLegend (San Diego, CA). Mouse MIP-1α (CCL-3) and vascular endothelial growth factor ELISA kits were purchased from PeproTech (Rocky Hill, NJ). Mouse IL-13 ELISA kits were purchased from eBioscience (San Diego, CA). l-(+)-LA and sodium l-lactate were purchased from Sigma-Aldrich (St. Louis, MO). Human IL-6, TNF, and MCP-1 ELISA kits (BD OptEIA) were purchased from BD Biosciences (Franklin Lakes, NJ).

For IL-33 activation, BMMC (2 × 10⁶ cells/ml) were cultured in 20 ng/ml of IL-3 and SCF in cRPMI. An equal volume of 25 mM LA in cRPMI was added to the cell suspension, resulting in a final cell concentration of 1 × 10⁶ cells/ml, 10 ng/ml IL-3 and SCF, and 12.5 mM LA. Control conditions received cRPMI in place of LA. After 24 h of pretreatment in LA media, cells then received 100 ng/ml IL-33 for 16 h, after which supernatants were collected. pH was measured for media alone, LA, and lactate-conditioned media using the Beckman Phi 45 pH meter.

Cells were cultured at 2 × 10⁶/ml and lysed in lysis buffer (Cell Signaling Technology, Danvers, MA) supplemented with 1.5× ProteaseArrest (G-Biosciences, Maryland Heights, MO). Protein concentration was determined using the Pierce bicinchoninic acid assay protein assay kit (Thermo Scientific). Proteins were resolved by SDS-PAGE using 30 μg of total protein per sample on 4–20% Mini-Protean TGX gels (Bio-Rad, Hercules, CA). Transfer was made onto nitrocellulose membranes, which were then blocked for 1 h at room temperature with 2% BSA in PBS. Membranes were rinsed in PBS and then incubated overnight at 4°C in PBS with Tween 20 containing 2% BSA and primary Ab diluted 1:1000. Membranes were probed with Abs purchased from Cell Signaling Technology, including anti–phospho-TGF-β–activated kinase-1 (TAK1; Thr^184/187, catalog no. 4508), total TAK1 (catalog no. 5206), phospho–NF-κB p65 (Ser⁵³⁶, catalog no. 13346), total NF-κB p65 (catalog no. 4764), phospho-JNK (Thr¹⁸³/Tyr¹⁸⁵, catalog no. 9251), total JNK (catalog no. 9258), phospho-p38 (Thr¹⁸⁰/Tyr¹⁸², catalog no. 9216), total p38 (catalog no. 9212), phospho-ERK1/2 (Thr²⁰²/Tyr²⁰⁴, catalog no. 9101), total ERK1/2 (catalog no. 4695), and GAPDH (catalog no. 2118). Membranes were washed the next day with PBS with Tween 20 every 5 min for a total of 30 min, then incubated with a 1:15,000 dilution of either goat anti-rabbit DyLight 800 (catalog no. 5151) or goat anti-mouse DyLight 680 (catalog no. 5470) infrared-labeled secondary Abs (Cell Signaling Technology). Membranes were rinsed a final time before being analyzed with an Odyssey CLx infrared scanner (LI-COR Biosciences, Lincoln, NE). Normalization was done using Image Studio 4.0 software (LI-COR Biosciences).

TAK1 inhibitor (5Z)-7-oxozeaenol (5 μM; Tocris Bioscience, Bristol, U.K.), JNK inhibitor SP600125 (10 μM; EMD Millipore, Billerica, MA), NF-κB inhibitor BAY 11-7085 (2 μM; Tocris Bioscience), ERK inhibitor FR180204 (25 μM; Cayman Chemical, Ann Arbor, MI), and MCT inhibitors α-cyano-4-hydroxycinnamic acid (CHC, 5 mM; Sigma-Aldrich) and AR-C155858 (100 nM; Tocris Bioscience) were solubilized in DMSO. Under the conditions used, none of these inhibitors caused significant cell death. Inhibitors were added to culture 1 h prior to activation with IL-33 (100 ng/ml). Supernatants were collected 16 h later, and ELISAs were used to determine cytokine production.

After BMMC were treated in 12.5 mM LA, total RNA was extracted with TRIzol reagent (Life Technologies, Grand Island, NY) and later measured using the Thermo Scientific NanoDrop 1000 UV-Vis spectrophotometer (Thermo Scientific, Waltham, MA) according to the manufacturer’s recommended protocol. For RNA extract that would be used to measure microRNA expression, polyadenylation was done prior to cDNA synthesis using the qScript microRNA cDNA synthesis kit (Quanta Biosciences, Gaithersburg, MD). For samples used to measure mRNA expression, cDNA was synthesized using the qScript cDNA synthesis kit (Quanta Biosciences) following the manufacturer’s protocol. Quantitative PCR (qPCR) analysis was performed with a CFX96 Touch real-time PCR detection system (Bio-Rad, Hercules, CA) and SYBR Green detection using a relative Livak and Schmittgen (24) method. Each reaction was performed according to the manufacturer’s protocol using 10 ng of sample cDNA, with the following primers: mmu-miR-155-5p, 5′-UUAAUGCUAAUUGUGAUAGGGGU-3′ (catalog no. MMIR-0155; Quanta Biosciences); mmu-miR-155-3p, 5′-CUCCUACCUGUUAGCAUUAAC-3′ (catalog no. MMIR-0155*; Quanta Biosciences); HIF-1α forward, 5′-TGAGGCTCACCATCAGTTAT-3′, reverse, 5′-TAACCCCATGTATTTGTTC-3′; MCT-1 forward, 5′-GCTGGAGGTCCTATCAGCAG-3′, reverse, 5′-CGGACAGCTTTTCTCCTTTG-3′; MCT-2 forward, 5′-TTACCGTATCTGGGCCTTTG-3′, reverse, 5′-CCAAAGCAGTTTCGAAGGAG-3′; GAPDH forward, 5′-GATGACATCAAGAAGGTGGTG-3′, reverse, 5′-GCTGTAGCCAAATTCGTTGTC-3′; SNORD47, 5′-GUGAUGAUUCUGCCAAAUGAUACAAAGUGAUAUCACCUUUAAACCGUUCAUUUUAUUUCUGAGG-3′ (catalog no. MM-SNORD47; Quanta Biosciences); β-Actin forward, 5′-GATGACGATATCGCTGCGC-3′, reverse, 5′-CTCGTCACCCACATAGGAGT-3′. Amplification conditions for microRNA detection were set to heat activation at 95°C for 2 min followed by 40 cycles of denaturation at 95°C for 5 s, annealing at 60°C for 15 s, and extension at 70°C for 15 s. All other reactions consisted of a heat activation step at 95°C for 10 min followed by 40 cycles of 95°C for 15 s, 55°C for 30 s, and 60°C for 1 min. Fluorescence data were collected during the extension step of the reaction.

BMMC were transfected with 100 nM HIF-1α small interfering RNA (siRNA; catalog no. GS15251) and scrambled FlexiTube siRNA (catalog no. 1027280) from Qiagen (Valencia, CA). miR-155-5p (catalog no. 470919), miR-155-3p (catalog no. 471999), and negative control/mock (catalog no. 479903) microRNA mimics were transfected at 50 nM and purchased from Exiqon (Woburn, MA). Transfections were done with Amaxa Nucleofector technology from Lonza (Allendale, NJ) using program T5 in the following transfection media: DMEM, 20% FBS, and 50 mM HEPES buffer. BMMC were incubated in IL-3 and SCF (10 ng/ml each) and used 48 h after transfection.

C57BL/6 mice 12–16 wk old were first injected with 1 mg/kg ketoprofen from Spectrum Chemical (New Brunswick, NJ) and then 30 min later injected with 4 mg/kg in 4% (w/v) LA. Sixteen hours later, mice were then injected with 1 μg of recombinant mouse IL-33 from eBioscience. Four hours later, mice were euthanized and blood was collected via cardiac puncture and plasma was prepared from collected blood. All animal protocols were approved by the Virginia Commonwealth University Institutional Animal Care and Use Committee.

As approved by the Internal Review Board at the University of South Carolina, surgical skin samples were collected from the Cooperative Human Tissue Network of the National Cancer Institute. Skin mast cells were harvested and cultured from five human donors as previously described (25). After 6–10 wk, mast cells were used when purity was nearly 100%, as confirmed with toluidine blue staining.

Data were presented as means ± SE and analyzed using GraphPad Prism 6 software (GraphPad Software, La Jolla, CA). Comparisons between two groups were done using an unpaired Student t test, and comparisons between multiple groups were done using one-way ANOVA with a Tukey post hoc test. A p value <0.05 was deemed significant.

In a normal physiological state, LA is present in peripheral tissue at 1–2 mM (26), whereas in pathological conditions LA concentrations often exceed 10 mM (19, 27). Elevated levels of LA are present in tissues where pathological conditions also promote mast cell infiltration and IL-33 expression (7, 13, 14, 28). Therefore, we decided to first investigate how LA alters IL-33–mediated activation of mast cells by measuring the impact on cytokine secretion. The timing and dose of LA treatment were determined by measuring changes in IL-33–induced cytokine and chemokine production. LA exposure greatly reduced IL-33–mediated IL-6 secretion, with maximal effects at 6–24 h of pre-exposure. Adding LA simultaneously or 48 h prior to IL-33 yielded no inhibition (Fig. 1A). In response to varying doses of LA, an IC₅₀ between 6 and 12.5 mM was observed for different cytokines (Fig. 1B). Whereas 25 mM LA yielded further inhibition of some cytokines, it also resulted in ∼30% cell death (data not shown). Hence, we employed 12.5 mM LA, which elicited no changes in cell viability (data not shown). The inhibitory effect of LA was consistent among proinflammatory cytokines, because BMMC pretreated in 12.5 mM LA for 24 h prior to IL-33 activation showed significantly decreased IL-6, TNF, MCP-1, MIP-1α, and IL-13 production, yet increased vascular endothelial growth factor production (Fig. 1C). Lastly, peritoneal mast cells harvested from C57BL/6 mice were also pretreated with LA and then activated with IL-33. As shown in Fig. 1D, LA significantly suppressed the IL-33–induced production of IL-6, MIP-1α, and MCP-1 by peritoneal mast cells as well. These findings demonstrate that LA can selectively suppress inflammatory cytokines from in vitro–differentiated as well as ex vivo–expanded primary mast cells.

Previous studies have demonstrated that pH plays a critical role in the ability of LA to alter cellular function. For example, LA has been shown to suppress LPS-induced TNF secretion and delay NF-κB activation in monocytes, while promoting an anti-inflammatory M2 phenotype in macrophages (19, 29, 30). However, lactate, the salt form of LA that does not lower pH, enhances LPS-mediated inflammatory responses in macrophages (21). We noted that media pH decreased to 6.5 immediately after adding LA, and returned to 7.2 within 6–8 h (data not shown). To investigate the importance of pH on LA effects in mast cells, BMMC were cultured in 12.5 mM LA or sodium lactate for 24 h prior to IL-33 activation. The results showed a clear difference, as LA suppressed IL-33–mediated IL-6, TNF, IL-13, and MCP-1 production, whereas sodium lactate had no effect (Fig. 2A), indicating the importance of pH on cytokine production.

Other work has demonstrated that cellular entry and egress of LA is controlled by MCT, a group of transporters within the solute carrier family (SLC16). MCT-1 and -2 can transport lactate into the cell (31–33). Whereas MCT-2 expression is largely restricted to the testis, heart, and brain, MCT-1 is widely expressed, including in hematopoietic cells (31, 34, 35). To determine whether MCT-1 is important for LA effects on mast cells, we treated BMMC with two different MCT inhibitors, CHC (a pan-MCT inhibitor) and AR-C155858 (an MCT-1 and -2 inhibitor) and measured IL-33–induced cytokine production. In the presence of these inhibitors, LA did not suppress IL-33–mediated production of IL-6 (Fig. 2B). To determine MCT-1 and -2 expression in mast cells, RNA was harvested from BMMC and analyzed via qPCR. Functionality of MCT-1 and -2 primers was confirmed by using kidney tissue cDNA with these primers (data not shown). Whereas MCT-1 transcripts were easily detected in BMMC within 40 cycles, MCT-2 transcripts were not detectable at all (Fig. 2C). Thus, these data demonstrate that LA effects on IL-33 activation are likely dependent on acid pH and the MCT-1 carrier.

We next sought to elucidate how LA alters IL-33 signaling. Flow cytometry revealed a modest (∼17%) decrease in surface ST2 expression after 24 h of LA treatment, which seemed unlikely to explain the inhibitory effects (data not shown). Therefore, we examined LA-mediated alterations in downstream signaling events. BMMC were pretreated in 12.5 mM LA for 24 h and activated with IL-33 before lysis. These lysates were then used to assess changes in TAK1, JNK, p38, NF-κB p65, and ERK phosphorylation, pathways suggested to be important for IL-33–mediated mast cell function (36). LA suppressed IL-33–induced activation of NF-κB p65, JNK, ERK, and TAK1, without altering p38 phosphorylation (Fig. 3). To determine whether blocking the suppressed pathways alone is sufficient to mimic the effects of LA on cytokine production, we treated BMMC with JNK, TAK1, ERK, or NF-κB chemical inhibitors. Whereas the JNK inhibitor did not reproduce the same level of suppression, TAK1, NF-κB, or ERK inhibition completely abolished IL-33–mediated cytokine production, mimicking LA effects (Fig. 4). Thus, LA inhibits multiple IL-33 signaling cascades, yielding redundant suppression of the proinflammatory stimulus.

The microRNA miR-155 has recently emerged as a powerful proinflammatory regulator by virtue of its ability to target and suppress inhibitory proteins (37). Because LA is known to promote HIF-1α production (19, 38, 39) and HIF-1α can control miR-155 expression (40–42), we hypothesized that LA-induced HIF-1α might suppress miR-155, yielding a net negative effect on IL-33 signaling. After 6 h of LA treatment, HIF-1α mRNA increased nearly 3-fold, whereas miR-155-5p was suppressed >50% in BMMC. Interestingly, the miR-155-3p species was unaffected, indicating selectivity of these effects (Fig. 5A). HIF-1α siRNA transfection prevented LA-mediated miR-155-5p suppression (Fig. 5B), demonstrating that this process is HIF-1α–dependent. To test the functional importance of miR-155-5p suppression, BMMC were transfected with miR-155-5p or miR-155-3p mimics (Fig. 5C). BMMC transfected with an miR-155-3p mimic prior to LA treatment still showed suppression of IL-33–induced cytokine production (Fig. 5D, upper graphs). However, transfecting the miR-155-5p mimic eliminated LA-mediated suppression, suggesting that miR-155-5p is critical for LA effects (Fig. 5D, lower graphs).

Thus far, we have shown that LA suppresses IL-33–mediated mast cell inflammatory responses in vitro and ex vivo. Next, we investigated LA effects on IL-33–mediated inflammatory responses in vivo. Intraperitoneal IL-33 injection has been shown to elicit mast cell–mediated inflammation coupled with cytokine production (43). Mice were first injected i.p. with LA and then 16 h later injected with IL-33. IL-33 greatly elevated plasma MCP-1 and IL-13 levels, an effect that was nearly completely reversed by LA pretreatment (Fig. 6). Thus, LA antagonizes the proinflammatory effects of IL-33 in vivo as well as in vitro.

To determine whether our mouse data are consistent in human mast cells, we assessed the effects of LA on IL-33–induced cytokine production using primary human skin mast cells (SkMC). SkMC from five healthy donors were harvested and cultured in the presence of LA for 24 h. Because SkMC respond poorly to IL-33 alone, we measured IL-33–mediated enhancement of IgE responses (7). LA pretreatment suppressed both IgE-mediated cytokine production and the enhancing effects of IL-33. Furthermore, LA effects were completely or partially reversed by the MCT inhibitor CHC (Fig. 7). These data suggest that LA consistently suppresses IL-33 signaling and mediates these effects via MCT transporters in both murine and human mast cells.

LA is a byproduct of anaerobic glycolysis and is known to be increased in a variety of pathological states, including cancer, obesity, type II diabetes, and wound healing (20–23). LA concentrations, which are 1–2 mM in plasma at rest, can reach 20 mM during acute exercise and 40 mM at tumor sites (19, 26). Although LA has been shown to inhibit cytotoxic T cell–mediated killing, promote M2 macrophage differentiation, and prevent dendritic cell Ag presentation, its effects on mast cell function have not been investigated (14–16, 19). Mast cells are known to participate in allergic disease, parasitic infection, and resistance to bacteria. They have less defined roles in cancer and wound healing but are known to participate in both (2–5, 44–47). Our data indicate that LA alters inflammatory cytokine production in mast cells stimulated with IL-33, a cytokine elevated in many pathological conditions. In the context of cancer, this could promote tumor escape from immune surveillance, as is true for tumor-associated macrophages in response to tumor-derived LA (19). In wound healing, it may prevent a destructive chronic inflammatory response, but might also reduce pathogen clearance.

BMMC exhibited decreased inflammatory cytokine production when exposed to LA prior to IL-33 activation. Similar to reports on other cell types, the inhibitory effects of LA in our assays were tightly linked to acidity (15). Whereas sodium lactate increases NF-κB signaling and transcription in macrophages stimulated with LPS (21, 48), LA decreases LPS-mediated signaling in macrophages (29, 30). Sodium lactate is the salt of LA, with the deprotonated carboxyl group linked to sodium via an ionic bond. Our results show that without the decrease in pH, IL-33–induced cytokine and chemokine production was unaltered. Additionally, we demonstrated that the MCT-1 transporter is critical for LA-mediated suppression. Others have shown that MCT-1 employs proton cotransport with lactate (49), supporting our data showing LA but not its salt is suppressive. Interestingly, MCT-1 was recently shown to require a chaperone protein, CD147, for its function (32). This warrants further investigation of mast cell CD147 expression and how this protein contributes to LA effects.

LA has been shown to suppress TLR4-mediated signaling and cytokine production (29, 30). IL-33 signaling is still being unraveled, but it shares common molecular pathways with TLRs. IL-33 activates the MAP3K TAK1 as an apical kinase, with resulting downstream activation of MAPKs, NF-κB, and AP-1 in mast cells (36). Our data show that although p38 phosphorylation is not affected, TAK1, NF-κB p65, ERK, and JNK phosphorylation was significantly diminished, and this correlated with decreased cytokine production. Chemical inhibitors of TAK1, ERK, or NF-κB each mimicked LA effects on IL-33 signaling, whereas JNK inhibition had no effect. Whether LA effects on multiple signaling proteins are simply due to apical TAK1 blockade yielding multiple downstream consequences and how LA inhibits phosphorylation are issues that require further study.

In response to LA treatment, HIF-1α expression was elevated in mast cells, as predicted by previous studies (19, 38, 39). A known molecule targeted by HIF-1α in other systems is miR-155, which is proinflammatory in many lineages (41). Studies show that miR-155 possesses a hypoxia response element in its promoter region, further connecting HIF-1α and miR-155 (42). We demonstrated that LA specifically suppresses miR-155-5p while leaving miR-155-3p unaltered. This selectivity appears consequential, because miR-155-5p but not miR-155-3p overexpression reversed LA effects. Despite originating from the same transcribed primary microRNA, it is not unusual for 3p and 5p strands from the same primary microRNA to be expressed differently in mature form in response to different stimuli. By nature of being complementary to one another, 3p and 5p strands can have very different targets, eliciting effects with negative or positive feedback on the degradation of these strands (50–52). Additionally, HIF-1α silencing restored miR-155-5p levels, indicating that in mast cells, HIF-1α negatively regulates miR-155-5p expression. This is another example of lineage-restricted effects, because HIF-1α induces miR-155 in populations such as epithelial cells (42). Clearly our understanding of hypoxia and tissue acidity will need to account for variations in lineage, allowing for nuanced effects.

The functional relevance of these findings is supported by consistent effects in vivo and on primary human mast cells. Several features of these experiments warrant further discussion and investigation. First, a recent study showed that reduced pH enhances IgE-mediated mouse mast cell cytokine production (53). We incidentally noted the opposite while stimulating human mast cells with IgE (Fig. 7). The differences between these studies could be due to the acids employed, assay parameters, or species variation. Kamide et al. (53) did not specify the acid they employed. A strong acid might have different effects than LA (pKa = 3.86) and may not be transported by MCT-1. The previous study also employed a 3-h incubation, whereas we cultured for 24 h. However, our IL-33 studies showed no enhancing effects of inflammatory cytokines at time points between 0 and 48 h. More work should be done to examine the effect of LA on IgE signaling. Importantly, note that our in vivo assay did not limit LA effects to mast cells, although mast cells are activated by systemic IL-33 (43). A variety of cytokine-producing cells respond to IL-33. This list continues to grow and currently includes mast cells, basophils, eosinophils, group 2 innate lymphoid cells, some Th2 cells, and macrophages (54–58). Therefore, our results demonstrate that LA can antagonize IL-33–induced systemic cytokine production in vivo, but they do not restrict these effects to a specific lineage. Although it is striking to find nearly complete suppression of cytokines in vivo, further study is needed to reveal how LA acts on various lineages.

In conclusion, LA is able to suppress inflammatory responses among IL-33–activated mast cells in a pH- and MCT-1–dependent manner. This suppression requires an HIF-1α–dependent blockade of miR-155-5p. The ability of LA to suppress IL-33–mediated inflammatory responses was reproduced in vivo and in human mast cells. These data provide fundamental insight into how tissue microenvironments, especially in pathological conditions, can greatly alter mast cell responses. Given our expanding comprehension of myriad activities played by IL-33 and mast cells, understanding and intervening in these signaling cascades is likely to be clinically important.

The authors have no financial conflicts of interest.