Functional Characterization of Human Receptors for Short Chain Fatty Acids and Their Role in Polymorphonuclear Cell Activation*

Short chain fatty acids (SCFAs), including acetate, propionate, and butyrate, are produced at high concentration by bacteria in the gut and subsequently released in the bloodstream. Basal acetate concentrations in the blood (about 100 (cid:1) M ) can further increase to millimolar concentrations following alcohol intake. It was known previously that SCFAs can activate leukocytes, particularly neutrophils. In the present work, we have identified two previously orphan G protein-coupled receptors, GPR41 and GPR43, as receptors for SCFAs. Propionate was the most potent agonist for both GPR41 and GPR43. Acetate was more selective for GPR43, whereas butyrate and isobutyrate were more active on GPR41. The two receptors were coupled to inositol 1,4,5-trisphosphate Tissue Distribution of GPR41 and GPR43— Reverse transcription (RT)-PCR experiments were carried out using a panel of total RNA (peripheral blood mononuclear cells (PBMC), dendritic cells, monocytes, T lymphocytes, small intestine) and poly(A) (cid:2) RNA (thymus, spleen, lymph node, bone marrow, lung, stomach, adipose, breast). The total RNA from neutrophils was prepared from venous blood of healthy donors using TriPure (Roche Diagnostics). Approximately 50 ng of poly(A) (cid:2) RNA or 500 ng of total RNA was reverse-transcribed with Superscript II (Invitrogen) and used for PCR. Human GPR43 and GPR41 receptors transcripts were detected by PCR using the following primers: 5 (cid:1) -TTCTACAGCAGCATCTACTG-3 (cid:1) (GPR43 forward), 5 (cid:1) -GA- AGCACACCAGGAAATTGA-3 (cid:1) (GPR43 reverse), 5 (cid:1) -TACGTCATAGAA-TTCTCAGG-3 (cid:1) (GPR41 forward), and 5 (cid:1) -TGTTCACTGGTCTTTCTTT- C-3 (cid:1) (GPR41 reverse). The expected sizes of the amplified DNA bands were 439 and 508 bp for GPR43 and GPR41, respectively. PCR was performed using the Taq polymerase under the following conditions: 94 °C for 5 min; 30 cycles at 94 °C for 1 min, 53 °C for 1 min 30 s, and 72 °C for 40 s (for GPR43); or 94 °C for 5 min, 30 cycles at 94 °C for 1 min, 52 °C for 1.5 min, and 72 °C for 35 s (for GPR41). A control was performed with glyceraldehyde-3-phosphate dehydrogenase cDNA fragment (509 bp) as described previously (16). Aliquots of the PCR were analyzed by 1% agarose gel electrophoresis. Polymorphonuclear Cell Pharmacology: Intracellular Calcium Re- lease and Chemotaxis Assays— Polymorphonuclear cells Cells and

G protein-coupled receptors (GPCRs) 1 constitute one of the largest gene families yet identified (1). In addition to about 160 characterized receptors, around 125 human genes encode proteins obviously belonging to this family of receptors, but their ligands and functions remain to be determined. These so far uncharacterized receptors are referred to as orphan GPCRs, but they are expected to play, by analogy with characterized members of the family, important roles in the regulation of physiological processes. For some orphan receptors, sequence similarity with well known receptors allows to construct hypotheses regarding the chemical nature of their ligands or their involvement in physiological processes. However, many orphan receptors are clustered in subfamilies with low similarity to characterized receptors.
Orphan receptors led in a number of cases to the discovery of molecules that were not recognized previously as functional extracellular mediators. The chemical diversity among endogenous ligands of GPCRs is unique as it includes ions, bioamines, lipids, peptides, and large proteins, as well as a large number of odorant molecules. The recent identification of prokineticins (2), UDP-glucose (3), lysophosphatidylcholine (4,5), sphingosylphosphorylcholine (5,6), relaxin (7), eicosanoid (8), kisspeptin (9,10), and psychosin (11) as new ligands for GPCRs has uncovered completely new extracellular pathways regulating cellular and tissue functions.
Among orphan receptors, a cluster of receptors poorly related to other subfamilies includes four members, GPR40, GPR41, GPR42, and GPR43 ( Fig. 1) (12). GPR41 and GPR42 have the same length and share 98% amino acid identity. The four genes encoding these receptors are intronless and are clustered onto chromosomal region 19q13.1. Although little information is available concerning these receptors, GPR41 was shown to induce apoptosis via a p53/Bax pathway in an ischemia/reperfusion paradigm (13). In addition, GPR43 and its murine orthologue LSSIG are induced during the differentiation of leukocyte progenitor cells to monocytes or neutrophils and were found mainly in hematopoietic tissues, suggesting that this receptor could have an important function in the differentiation and/or activation of leukocytes (14).
In this report, we showed that propionate, acetate, and other SCFAs act as specific activators of GPR41 and GPR43. An extensive pharmacological study revealed differences in the rank order of potency of SCFA toward each receptor, as well as in the G protein coupling leading to intracellular cascade activation. Pharmacological data obtained on human polymorphonuclear cells involved GPR43 as the main functional SCFA receptor on these cells.
Cloning and Sequencing of Human GPR43 and Human GPR41-Oligonucleotide primers were synthesized on the basis of the sequence of the human GPR41 and GPR43 receptors (GenBank TM accession numbers AF024688 and AF024690, respectively). For GPR43 cloning, sense primer 5Ј-CCGGAATTCACCATGCTGCCGGACTGGAAG-3Ј and antisense primer 5Ј-CTTGTCTAGACTACTCTGTAGTGAAGTC-3Ј were used in a PCR amplification using human genomic DNA as template and Pfu DNA polymerase (Stratagene) under the following conditions: 1 min at 94°C, 30 s at 52°C, 1 min at 72°C, 3 cycles; 1 min at 94°C, 30 s at 63°C, 1 min at 72°C, 30 cycles. A fragment of 1 kb containing the entire coding sequence of human GPR43 gene was amplified, digested by EcoRI and XbaI, and cloned in the bicistronic pEFIN5 expression vector. In the bicistronic vector, designated pEFIN5, both the recombinant receptor and the neomycin phosphotransferase selection marker are transcribed from a single promoter element.
For GPR41 cloning, sense primer 5Ј-CCGGATATCACCATGGATAC-AGGCCCCGAC-3Ј and antisense primer 5Ј-CTTGTCTAGACTAGCTT-TCAGCACAGGC-3Ј were used in a similar strategy, except that the coding sequence was cloned using EcoRV and XbaI. All the inserts of the resulting plasmids were sequenced on both strands using the Big-Dye Terminator cycle sequencing kit (Applied Biosystems). Sequence alignment was performed using the ClustalX software, version 1.8 (15), and dendrograms were constructed using TreeView.  Tissue Distribution of GPR41 and GPR43-Reverse transcription (RT)-PCR experiments were carried out using a panel of total RNA (peripheral blood mononuclear cells (PBMC), dendritic cells, monocytes, T lymphocytes, small intestine) and poly(A) ϩ RNA (thymus, spleen, lymph node, bone marrow, lung, stomach, adipose, breast). The total RNA from neutrophils was prepared from venous blood of healthy donors using TriPure (Roche Diagnostics). Approximately 50 ng of poly(A) ϩ RNA or 500 ng of total RNA was reverse-transcribed with Superscript II (Invitrogen) and used for PCR. Human GPR43 and GPR41 receptors transcripts were detected by PCR using the following primers: 5Ј-TTCTACAGCAGCATCTACTG-3Ј (GPR43 forward), 5Ј-GA-AGCACACCAGGAAATTGA-3Ј (GPR43 reverse), 5Ј-TACGTCATAGAA-TTCTCAGG-3Ј (GPR41 forward), and 5Ј-TGTTCACTGGTCTTTCTTT-C-3Ј (GPR41 reverse). The expected sizes of the amplified DNA bands were 439 and 508 bp for GPR43 and GPR41, respectively. PCR was performed using the Taq polymerase under the following conditions: 94°C for 5 min; 30 cycles at 94°C for 1 min, 53°C for 1 min 30 s, and 72°C for 40 s (for GPR43); or 94°C for 5 min, 30 cycles at 94°C for 1 min, 52°C for 1.5 min, and 72°C for 35 s (for GPR41). A control was performed with glyceraldehyde-3-phosphate dehydrogenase cDNA fragment (509 bp) as described previously (16). Aliquots of the PCR were analyzed by 1% agarose gel electrophoresis.
Aequorin Assay-The functional response to SCFAs was analyzed by measuring the luminescence of aequorin as described previously (17). For all assays, data were analyzed with the PRISM software (Graph-Pad Prism Software, San Diego, CA) using nonlinear regression applied to a sigmoidal dose-response model, as for all assays used in this work.
cAMP Assay-CHO-K1 cells were spread on Petri dishes (250,000 cells per 35-mm dish) and cultured overnight in Nutrient Ham's F-12 medium containing 10% FBS, 100 units/ml penicillin, 100 g/ml streptomycin, and 400 g/ml G418. Cells were preincubated for 30 min in Krebs-Ringer HEPES buffer composed of 25 mM HEPES, pH 7.4, 124 mM NaCl, 5 mM KCl, 1.25 mM MgSO 4 , 1.45 mM CaCl 2 , 1.25 mM KH 2 PO 4 , and 8 mM glucose and then incubated for 20 min in the same medium supplemented with 10 M forskolin and variable concentrations of agonists. The reaction was stopped by removing the supernatant and the addition of 100 l of lysis buffer in each well. The cAMP contents were quantified by enzyme-linked immunosorbent assay (CS200 kit; Applied Biosystems).
[ 35 S]GTP␥S Binding Assay-The measurement of agonist-stimulated [ 35 S]GTP␥S binding to membranes of cells expressing human GPR41 or human GPR43 was performed as described previously (9). Briefly, membranes (10 g) from CHO-GPR41 or CHO-GPR43 cells were incubated for 15 min at room temperature in binding buffer (20 mM HEPES, pH 7.4, 100 mM NaCl, 3 mM MgCl 2 , 3 M GDP, 10 g/ml saponin) containing different concentrations of SCFA in 96-well microplates (Basic FlashPlates; PerkinElmer Life Sciences). After addition of 0.1 nM [ 35 S]GTP␥S, microplates were shaken for 1 min and further incubated at 30°C for 30 min. The incubation was stopped by centrifugation of the microplate for 10 min, at 800 ϫ g and 4°C, and removal of the supernatant. Microplates were counted in a TopCount (PerkinElmer Life Sciences) for 1 min per well.
Fluorescence-based Intracellular Ca 2ϩ Mobilization-CHO/GPR43 and CHO/GPR41 cells were seeded into black 96-well assay plates the day before the experiment. Cells are loaded for 1 h with 4 M Fluo-3/AM (Molecular Probes) in Hank's balanced salt buffer and washed before being transferred to a microplate reader (FDSS; Hamamatsu Photonics KK, Hamamatsu, Japan) for compounds injection and simultaneous fluorescence recording for 3 min (excitation and emission wavelengths, 340 and 510 nm, respectively). Results were expressed as a percentage of fluorescence change as compared with basal level.
Phosphoinositide Accumulation-Cells expressing GPR41 or GPR43 receptors were labeled overnight with 10 Ci/ml [ 3 H]myo-inositol in culture medium. Cells were then washed three times and incubated in Krebs buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 4.2 mM KCl, 1 mM CaCl 2 , 0.5 mM MgCl 2 , 5.6 mM D-glucose) for 15 min in medium containing 1 mM LiCl. SCFA agonists were then added for 30 min. The reaction was stopped by replacing the incubation medium with 1 ml of 5% HClO 4 . The total inositol phosphate pool was then extracted and purified on Dowex columns as described by Joly et al. (18). Results were expressed as the ratio between the radioactivity collected in the inositol phosphate fraction over the radioactivity recovered from the cellular membranes solubilized in 10% Triton and 0.1 N NaOH and used as standard. The use of this ratio allows for greater homogeneity in the data, as it reduces variations resulting from differences in cell numbers from individual wells.
Polymorphonuclear Cell Pharmacology: Intracellular Calcium Release and Chemotaxis Assays-Polymorphonuclear cells were purified from buffy coats of healthy volunteers. PMN chemotaxis was performed in Boyden microchambers (Neuro Probe, Gaithersburg, MD) with polyvinylpyrrolidone-free polycarbonate membranes (5-m pore size; Corning Separations Division, Acton, MA) as described previously (19). For intracellular Ca 2ϩ assays, cells were incubated in Hanks' balanced salt medium containing 0.1% bovine serum albumin and 2.8 g/ml Fura-2/AM (Molecular Probes) at 37°C for 45 min. Cells were washed, resuspended at 5 ϫ 10 6 cells/ml, and allowed to re-equilibrate for 10 min at 37°C. Cells were then transferred to cuvettes, and calcium transients were monitored through fluorescence measurements using an LSB 50B spectrofluorimeter (PerkinElmer Life Sciences).

Identification of GPR43 and GPR41 as Receptors for Short
Chain Fatty Acids-In the frame of a general strategy for characterizing ligands for orphan G protein-coupled receptors, a CHO-K1 cell line coexpressing GPR43, G␣ 16 , and apoaequorin (GPR43-WTA11) was established and screened in an aequorin-based functional assay against a large collection of reference compounds comprising peptides, lipids, carbohydrates, and small chemical compounds. A biological activity specific for GPR43-expressing cells was observed for a number of peptide solutions containing acetate as the counter ion. Control tests revealed that the acetate ion itself, and not the peptides, was responsible for the agonist activity on GPR43. A pH effect could rapidly be excluded, as neutral acetate buffers displayed the same activity as acetic acid. Furthermore, different salts of acetate (Na ϩ , NH 3 ϩ , and K ϩ ) confirmed the activity of the acetate anion. We also excluded that the effect of acetate on intracellular calcium release was subsequent to the acidification of the cell cytosol, as several organic acids with a similar pK a were not active on GPR43-expressing cells. This specificity of the GPR43 response to acetate was further demonstrated by testing in the same assay a set of about 60 orphan and characterized receptors belonging to different GPCR classes and subclasses, in addition to the wild-type cell lines WTA11. None of these responded to acetate. The determination of concentration-action curves allowed us to estimate the EC 50 of GPR43 for sodium acetate at 268 Ϯ 26 M (Fig. 2A). Additional SCFAs were tested at this stage and were found to activate GPR43, as well, in the aequorin-based assay (data not shown).
We also screened the receptors structurally related to GPR43 and found that GPR41, which shares 42% amino acid identity with GPR43, was also activated by acetate, although to a lower extent (Fig. 2B). Indeed, the EC 50 of GPR41 for acetate was estimated at 1390 Ϯ 220 M, and the other SCFA propionate was found to display a higher potency on this receptor, with an EC 50 of 11.6 Ϯ 1.4 M (Fig. 2B).
Intracellular Coupling of GPR41 and GPR43-The natural coupling properties and the intracellular signaling pathways activated by GPR43 and GPR41, upon stimulation by acetate or propionate, were investigated in CHO-K1 cells expressing the human receptors but devoid of aequorin or additional coupling proteins such as G␣ 16 (CHO/GPR43 and CHO/GPR41 cells). We first demonstrated that both receptors coupled negatively to adenylyl cyclase through a Pertussis toxin-sensitive G protein (G i/o class), while being unable to promote accumulation of cAMP in the absence of forskolin (not shown). This cAMP accumulation assay was used to characterize further the detailed pharmacology of the receptors (see Fig. 3, A and B, and see below). A [ 35 S]GTP␥S binding assay was used alternatively as a functional test to monitor G i/o coupling of these receptors in a cell-free assay, to exclude further the possibility that cell activation would result from nonspecific actions on the cell metabolism or cytoplasmic components. The activity of acetate and propionate on GPR43 and GPR41 was confirmed in this assay (Fig. 4, A and B). Propionate was equipotent on GPR43 (EC 50 ϭ 259 Ϯ 67 M) and GPR41 (EC 50 ϭ 274 Ϯ 75 M), whereas acetate was more potent on GPR43 (EC 50 ϭ 537 Ϯ 31 M) than on GPR41 (EC 50 ϭ 1299 Ϯ 65 M). PTX treatment inhibited the response to SCFAs for both receptors. Stimulation of GPR41 and GPR43 also resulted in the release of intracellular calcium, with a similar rank order of potency (see Fig.  4, C and D and Table I). Although the EC 50 values observed in both assays were similar for GPR41, they were higher in the Ca 2ϩ assay as compared with the cAMP assay for GPR43. Furthermore, PTX abolished the response of GPR41 but not of GPR43 (Fig. 4, C and D). This suggested a unique G i/o coupling for GPR41 and a dual coupling through the G i/o and G q families for GPR43.
The activity of SCFAs was confirmed following expression of GPR43 and GPR41 in other cell lines (COS-7 and human embryonic kidney 293), with similar EC 50 values observed for acetate and propionate. Moreover, transient expression of GPR43 in COS-7 cells led to the accumulation of inositol phosphate products in a PTX-independent mechanism with an EC 50 of 325 Ϯ 36 M for propionate and 132 Ϯ 40 M for acetate (Fig.  4E). Stimulation of GPR41 expressed in COS-7 cells resulted in the accumulation of inositol phosphates, only following the cotransfection of the chimeric G protein G qi5 (Fig. 4F). Furthermore, the cotransfection of GPR43 and G qi5 increased significantly the basal level of inositol phosphates, as compared with control conditions, or the expression of one of the plasmids alone, suggesting that GPR43 is endowed with constitutive activity (data not shown).
Stimulation of GPR43 and GPR41 expressed in CHO-K1 cells with 10 mM propionate induced a time-dependent phosphorylation of p42 and p44 mitogen-activated protein kinases  4. Intracellular cascades activated by GPR41 and GPR43. A and B, the binding of [ 35 S]GTP␥S to membranes of CHO-K1 cells expressing GPR43 (A) or GPR41 (B) was measured following stimulation by acetate or propionate. C and D, intracellular calcium mobilization was measured in CHO-K1 cells expressing GPR43 (C) or GPR41 (D), after culturing the cells in the presence or absence of Pertussis toxin. The data represent the mean Ϯ S.E. for triplicate data points, and the displayed curves are representative of at least three independent experiments. E and F, inositol phosphate accumulation was determined in COS-7 cells transiently expressing GPR43 (E) or cotransfected with GPR43 or GPR41 and the chimeric G qi5 protein (F). G and H, phosphorylation of ERK1/2 mitogen-activated protein kinases following stimulation of GPR43 (G) and GPR41 (H) expressed in CHO-K1 cells by acetate or propionate for 5 min. The cellular extract (20-g proteins) was separated by SDSpolyacrylamide gel electrophoresis, transferred to nylon membranes, and labeled with antibodies specific for the phosphorylated forms of ERK1 and ERK2.
(ERK1/2), reaching a maximum by 3 to 5 min and returning to basal levels within 10 min (Fig. 4, G and H). On GPR41, stimulation of p42/p44 phosphorylation by acetate was less efficient than by propionate at the same concentration (10 mM). This activation was abolished after PTX pre-treatment, demonstrating the involvement of G i/o family proteins in this pathway, as well. On GPR43, similar levels of p42/p44 phosphorylation were obtained following acetate or propionate stimulation, and its blockade by PTX pre-treatment was also efficient (Fig. 4H).
In Vitro Pharmacology of Human Recombinant GPR41 and GPR43-The pharmacology of both receptors was extended by assaying a number of short carboxylic acids and related molecules in a cAMP accumulation assay, using CHO-K1 cells, expressing stably GPR41 or GPR43, and stimulated with forskolin. The tested compounds included carboxylic acids containing 1 to 8 carbons, either linear or branched. The full list of active compounds is displayed in Table I, together with the functional parameters.
Among the compounds displaying an agonist activity on GPR43, acetate, propionate, and butyrate were characterized by a similar micromolar potency (Fig. 3A). The rank order of potency was propionate Ͼ acetate ϭ butyrate Ͼ isobutyrate Ͼ caproate Ͼ isovalerate Ͼ valerate Ͼ formate Ͼ pivalate Ͼ L-OH-butyrate, caprylate, and heptanoate, which were active only at concentrations of 10 mM (Table I). For human GPR41, the rank order of potency showed notable differences as compared with GPR43. The order was propionate Ͼ isobutyrate Ͼ butyrate Ͼ valerate Ͼ isovalerate Ͼ caproate Ͼ pivalate Ͼ acetate (see Fig. 3B and Table I). Formate was totally inactive on this receptor. Other molecules including acetaldehyde, acetamide, acetate esters, acetone, ethanol, pyruvate, lactate, and glyoxylate were tested and found to be inactive on both receptors.
Distribution of GPR41 and GPR43 in Tissues and Leukocyte Populations-The tissue distribution of the receptors was determined by RT-PCR in a number of human peripheral tissues, brain areas, lymphoid organs, and leukocyte populations purified from human blood. GPR43 expression was characterized by a high specificity. Transcripts were found at high level in PMN and at lower levels in PBMC, as well as in purified monocytes. GPR43 transcripts were also detected in bone marrow and spleen (Fig. 5). GPR41 expression appeared more widespread, with significant and grossly similar expression in spleen, lymph node, bone marrow, and PBMC. No expression could be detected in PMN, monocytes, dendritic cells, and a number of peripheral organs (Fig. 5).
Functional Effects of SCFAs on Polymorphonuclear Cells-SCFAs have been described as displaying activities on leukocyte populations, particularly PMN cells (20 -22). Given the identification of specific receptors for these molecules and the characterization of their expression profile we investigated the effects of SCFAs on human leukocytes. PMN cells were particularly investigated, provided the strong and specific expression of GPR43 in this population and the previously described activity of SCFAs on these cells. Stimulation of PMN cells by propionate or acetate resulted in a transient increase of intracellular calcium (Fig. 6, A and B), as described before the identification of specific receptors for these agents (23)(24)(25). The Ca 2ϩ response was similar to that obtained with the bacterial peptide formyl-Met-Leu-Phe (fMLP), the prototypical agonist of the formyl peptide receptor. Prior stimulation of the cells with acetate or propionate inhibited subsequent responses to propionate or acetate (Fig. 6B), suggesting a homologous desensitization of a common receptor for the two agonists. Acetate or propionate did not affect, however, the functional responses to fMLP, interleukin-8, or leukotriene B4 (Fig. 6A). Conversely, a first stimulation of PMN with fMLP, interleukin-8, or leukotriene B4 reduced or abolished, in a concentration-dependant manner, the subsequent response to acetate or propionate (Fig. 6C). Concentration-action curves were established for propionate (Fig. 6D) and acetate (Fig. 6E)  The functional parameters of GPR43 and GPR41 activation by SCFAs and related molecules were determined using a cAMP accumulation assay (in the presence of forskolin) and a fluorescent-based calcium mobilization assay. Values are the mean pEC 50 and S.E. for at least three independent determinations. The number in parentheses indicates the potency order for each assay and each receptor. ND, not determined.  PMN cells were also tested for their chemotactic response to sodium acetate and sodium propionate. Both SCFAs resulted in a classical bell-shaped dose-response curve with an optimal concentration of 1 mM (Fig. 6F). The potency of SCFAs in neutrophil chemotaxis was significantly lower than that of fMLP, which was still fully active at concentrations of 1 nM (data not shown). Furthermore, the efficacy of fMLP was also higher than that of SCFAs, as the maximal chemotactic index obtained with fMLP was on average at least 3-fold higher than with SCFAs. Additional experiments indicated that, at optimal agonistic concentrations (1 mM), SCFAs could inhibit neither spontaneous neutrophil migration nor chemotaxis induced by 10 nM fMLP (data not shown). DISCUSSION GPR43 and GPR41 belong to a cluster of four orphan GPCRs, together with GPR40 and GPR42, which share little structural similarities with other subfamilies of G protein-coupled receptors. GPR43 is mainly expressed on leukocyte populations, particularly neutrophils, whereas GPR41 expression is more widely distributed in tissues. The involvement of GPR43 in leukocyte function and host defense is supported by the induction of its mRNA during the differentiation and activation of monocytes and PMN cells (14).
The screening of a library of reference bioactive molecules allowed us to identify acetate and then other short chain fatty FIG. 6. Functional activation and chemotaxis of polymorphonuclear cells, in response to acetate and propionate. Freshly prepared neutrophils were tested for their functional response to acetate, propionate, and fMLP, as well as cross-desensitization, using a fluorescence-based Ca 2ϩ mobilization assay. A, neutrophils were stimulated successively with 30 mM propionate and 100 nM fMLP. B, neutrophils were stimulated twice with 30 mM acetate. C, neutrophils were stimulated first with 100 nM fMLP and then with 30 mM propionate. Each panel is representative of at least three independent experiments. D, concentration-action curve for propionate. E, concentration-action curve for acetate. F, neutrophil chemotaxis assay in response to acetate and propionate. The data represent the index relative to basal migration across the membrane. The chemotaxis data represent the mean Ϯ S.E. for triplicate wells of five independent experiments. acids as specific agonists for GPR43. The SCFAs were then used to screen the structurally related receptors, and we found that GPR41 was also activated by SCFA with a different rank order of potency, although its amino acid identity with GPR43 was somehow limited (39%). A cell line expressing GPR40 did not respond to SCFAs in our hands, whereas GPR42, which shares 98% amino acid identity with GPR41, has not been tested in this study. Acetate was 100-fold less potent on GPR41 as compared with propionate and butyrate, whereas the three ligands activated GPR43 with a similar potency. A number of observations in various cell environments and functional assays suggested that the two receptors are coupled differently to intracellular signaling cascades. GPR41 appears to be coupled exclusively to the to G i/o family of proteins, whereas a dual coupling through G q and G i/o families is observed for GPR43. In both cases, the resulting intracellular pathways activated by the receptors included inositol 1,4,5-trisphosphate generation, intracellular Ca 2ϩ release, ERK1/2 activation, and inhibition of cAMP accumulation.
Among the compounds tested, a set of ten carboxylic acids and related molecules were found to be active on both GPR41 and GPR43 in cAMP accumulation and intracellular Ca 2ϩ release assays. Structure-activity relationships of the active compounds showed that the carboxylic moiety is required for activity. Indeed, aldehydes, ketones, or alcohols with similar carbon chains were totally inactive, as well as ester derivatives of the active compounds. A carboxylic moiety has to be located at the end of aliphatic chain comprising one to six carbon atoms, linear or branched. The optimal length is two to three carbon atoms for activity on GPR43 and three to five carbon atoms for GPR41. The presence of two (or more) carboxyl groups is, however, not tolerated as a number of di-acids were inactive, whatever the length of the carbon chain. These inactive compounds included oxalate (C2), malonate (C3), succinate (C4), aspartate (C4), glutamate (C5), and citrate (C6).
In vivo relevant SCFAs include, but are not limited to, acetate, propionate, and butyrate. These molecules are produced in considerable amounts by microbial fermentation in the hindgut, where they reach local concentrations as high as 70 to 100 mM (26). SCFAs are also produced as metabolic by-products of anaerobic bacteria present in the periodontal pocket. SCFAs are rapidly transferred from these compartments to the bloodstream, and the usual concentration in peripheral blood is around 100 to 150 M for acetate, 4 to 5 M for propionate, and 1 to 3 M for butyrate (27). Moreover, the plasmatic concentration of acetate can increase up to 10-fold as compared with basal values, following ethanol administration (28,29). A few inherited diseases, caused by specific enzyme defects, result in the accumulation of propionate and butyrate (30,31).
The high and specific expression of GPR43 in neutrophils can be related to the well established effects of SCFAs on these cells. Indeed, the observed calcium mobilization and chemotaxis of neutrophils, in response to SCFAs, confirms an abundant literature describing activation of PMN by SCFAs, which includes morphological changes and cell polarization, Ca 2ϩ release, actin cytoskeleton remodeling, and cytoplasmic pH oscillations (20,25). Our pharmacological data suggest that GPR43 is the functional receptor responsible for these previously reported actions of SCFAs on neutrophils.
The identification of functional receptors for SCFAs, expressed on leukocytes, opens new perspectives in the modulation of immune system functions in various pathophysiological situations. The average concentrations of propionate and butyrate in blood are too low to activate GPR41 or GPR43, but the blood concentrations reached by acetate are well within the active range for GPR43. Moreover, the main sources of SCFAs are bacteria in the gut, where their local concentrations are known to be much higher than in the blood. It has been established that products of the gut commensal flora can promote the activation of the mucosal immune system in the presence of an impaired or injured mucosal barrier, contributing to the development of inflammatory bowel diseases (32). Leukocytes also infiltrate diffusely the intestinal wall, in the absence of obvious morphological, clinical, or endoscopic evidence of inflammatory processes. Neutrophils play a key role in the early steps of a number of inflammatory processes of the gastrointestinal tract, including gastritis, enterocolitis, ulcerative colitis, and ischemia reperfusion injury. Following migration from the systemic circulation into the mucosal interstitial space, neutrophils undergo activation to produce reactive oxygen intermediates and chemokines, leading to the perpetuation of the inflammatory response, as well as contributing to the ultimate mucosal injury. The development of therapeutic strategies to block neutrophil recruitment and activation might be highly beneficial in a number of diseases characterized by an inflammatory component. Cyclosporin A, which inhibits neutrophil and T cell recruitment, is an efficient treatment for inflammatory bowel diseases. The identification of GPR43 as a functional receptor for SCFA could potentially lead to new therapeutic strategies.
Acetate concentration can increase up to 10-fold and reach millimolar concentrations after alcohol ingestion, as a consequence of the conversion of 60 to 75% of alcohol to acetate (29). It has been established that acute and chronic alcohol intake increases the susceptibility to infections caused by bacterial and viral pathogens (33,34). The impaired host defense associated with alcohol is the consequence of a defective inflammatory response, with altered cytokine production and decreased neutrophil function (35,36). Chronic ethanol intake was shown to affect fMLP-induced chemotactic activity, superoxide production by neutrophils, and their bactericidal activity (37,38). Some of these effects might involve the desensitization of neutrophil SCFA receptors by high blood concentrations of acetate, resulting in a reduction of their recruitment to sites of bacterial infection. Leukocyte infiltration is also an important component of alcoholic liver disease. In acute alcoholic hepatitis, PMN are selectively recruited to the liver where alcohol is metabolized into acetate (37,39).
As a conclusion, we have identified two specific receptors of SCFAs. These receptors could play a role similar to that of the fMLP receptor (40), by recognizing a bacterial metabolite and promoting leukocyte recruitment to the site of infection, as well as their activation. The pharmacological profile of GPR43 and its specific expression in neutrophils suggest, therefore, an early role in the induction of immune and inflammatory responses, with a possible involvement in inflammatory bowel diseases, as well as in alcoholism-associated disease susceptibility. The precise function of GPR41, given its broader distribution, is less clear. Additional studies, including in vivo pharmacology and the generation of knockout models, will be necessary for identifying further the precise roles of these two receptors and their potential applications as therapeutic targets. During the completion of the present work, two publications describing the independent identification of GPR40 as a receptor for medium fatty acids and of GPR41 and GPR43 as receptors for SCFAs became available on-line (41,42). These authors describe a pharmacological profile of SCFAs for GPR41 and GPR43 that is similar to that reported here.