Endocrinology and Metabolism

Gustducin couples fatty acid receptors to GLP-1 release in colon

Yan Li, Zaza Kokrashvili, Bedrich Mosinger, Robert F. Margolskee


Sweet taste receptor subunits and α-gustducin found in enteroendocrine cells of the small intestine have been implicated in release of the incretin hormones glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) in response to glucose and noncaloric sweeteners. α-Gustducin has also been found in colon, although its function there is unclear. We examined expression of α-gustducin, GLP-1, and GIP throughout the intestine. The number of α-gustducin-expressing cells and those coexpressing α-gustducin together with GLP-1 and/or GIP increased from small intestine to colon. α-Gustducin also was coexpressed with fatty acid G protein-coupled receptor (GPR) 40, GPR41, GPR43, GPR119, GPR120, and bile acid G protein-coupled receptor TGR5 in enteroendocrine cells of the colon. In colon, GPR43 was coexpressed with GPR119 and GPR120, but not with TGR5. Treatment of colonic mucosa isolated from wild-type mice with acetate, butyrate, oleic acid, oleoylethanolamide, or lithocholic acid stimulated GLP-1 secretion. However, GLP-1 release in response to these fatty acids was impaired in colonic tissue from α-gustducin knockout mice.

  • α-gustducin
  • glucose-dependent insulinotropic polypeptide
  • glucagon-like peptide 1
  • fatty acid receptors
  • enteroendocrine cell

the incretin hormones glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are estimated to account for ∼50–70% of total insulin release after oral glucose administration (13, 30, 59). The meal-related GLP-1 response in type 2 diabetes is decreased, which may contribute to the decreased incretin effect in this population (58, 86).

The sweet taste receptor subunits T1r2 and T1r3, and the G protein α-subunit α-gustducin, initially found in taste cells (48, 54, 55, 60, 91), have also been shown to be present in enteroendocrine and brush cells of the small intestine (3, 14, 29, 51). α-Gustducin and T1r receptors are also present in the small intestine's L- and K-type enteroendocrine cells, which release GLP-1 and GIP, respectively, in response to glucose in the gut lumen (35). In response to glucose gavage, α-gustducin null mice do not show the elevation of serum levels of GLP-1 found in gavaged wild-type mice (35). Furthermore, sweetener-stimulated GLP-1 release from L cell lines can be inhibited by small-interfering RNA against α-gustducin or pharmacological blocking of the T1r2+T1r3 sweet receptor (35, 51). In addition, other mechanisms independent of α-gustducin and T1rs, including sodium glucose cotransporter-1 (SGLT1) and ATP-sensitive K+ channels (KATP), are known to regulate GLP-1 release from primary L cells (69).

α-Gustducin is also expressed with peptide YY (PYY) and GLP-1 in L cells of the human colon (74), although its functions in these cells are unknown. The lumen of the colon lacks dietary sugars, but is filled with microbiota, nondigestible matter, and fermentation products, such as short-chain fatty acids (SCFAs), ammonia, phenols, amines, and bile acids (12, 23, 50, 84). Medium- and long-chain fatty acids (LCFAs) are also detected in fecal content from patients with small bowel resection on a high-fat diet (36) and from colectomized patients (1). Interestingly, the SCFA receptor G protein-coupled receptor (GPR) 43 (6, 47) is coexpressed with GLP-1 in both human and rat colon L cells (37). The LCFA receptors GPR40 and GPR120, the G protein-coupled bile acid receptor TGR5, and oleoylethanolamide (OEA) receptor GPR119 have been detected in L cells and/or in L-type enteroendocrine cell lines; agonist stimulation of these receptors promotes GLP-1 secretion (8, 15, 28, 38, 46, 69, 85). However, whether these receptors couple to α-gustducin in native cells is unknown.

Given α-gustducin's involvement with sweetener-stimulated GLP-1 release from L cells of the small intestine and its unexplained presence in L cells of the colon, we examined coexpression of α-gustducin with fatty acid receptors in colon. Here, we show that α-gustducin is coexpressed with SCFA receptors GPR43 and GPR41, LCFA receptors GPR40 and GPR120, OEA receptor GPR119, and bile acid receptor TGR5. In addition, using α-gustducin null mice, we show that α-gustducin is involved in GLP-1 release mediated by these fatty acid receptors. α-Gustducin is likely to contribute to colonic release of incretins to augment insulin release from pancreas.


Tissue preparation.

All experimental protocols and procedures were approved by Monell's Institutional Animal Care and Use Committee in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Adult male C57BL/6 mice (2–10 mo old) were killed by cervical dislocation, and the entire small intestine and proximal colon were removed quickly. The duodenum, jejunum, ileum, and proximal colon were flushed with phosphate-buffered saline (PBS, pH 7.4). For in situ hybridization, tissues were freshly frozen in Tissue-Tek optimum cutting temperature (OCT) mounting media (Sakura) and then sectioned within 1 h. For immunohistochemistry, tissues were fixed for 2 h in 4% (wt/vol) paraformaldehyde/1× PBS and then cryoprotected in 20% (wt/vol) sucrose/1× PBS overnight at 4°C before embedding in OCT. Sections (8–12 μm thick) were prepared using a CM3050S cryostat (Leica Microsystems) and placed on precoated microscope slides (Superfrost plus; Fisher Scientific). Sections were dried at 40°C for 20 min before immunohistochemistry.

In situ hybridization.

In situ hybridization was performed as previously described (96). Briefly, tissues were freshly frozen in Tissue-Tek OCT mounting media (Sakura) and then sectioned within 1 h. Next, the sections were fixed for 10 min in 4% paraformaldehyde and then permeabilized in 10 μg/ml proteinase K (Boehringer Mannheim) solution. After postfixation and acetylation, the slides were treated with DNase and then prehybridized for 1 h at room temperature in prehybridization solution containing 50% (vol/vol) deionized formamide, 5× saline/sodium citrate (SSC), 5× Denhardt's solution, 500 μg/ml salmon sperm DNA, 250 μl/ml of yeast tRNA, and 2.5 M EDTA in diethyl pyrocarbonate-treated water. Digoxigenin (DIG)-labeled RNA probe was incubated at 65°C overnight for hybridization. Next, the slides were incubated with anti-DIG-alkaline phosphatase (1:1,000; Boehringer) after blocking with 10% (vol/vol) heat-inactivated normal goat serum. Alkaline phosphate labeling was detected by a nitroblue tetrazolium plus 5-bromo-4-chloro-3 indolyl-phosphate mixture (Roche) with levamisole (Sigma). Antisense and sense RNA probes were used at equivalent concentrations and run in parallel in the same experiment to ensure equivalent conditions. For each experiment, in situ hybridization with positive controls with T1r3 or α-gustducin antisense probes was done on taste tissue to ensure the hybridization worked properly. The following primers were used to make RNA probes: GPR40 RNA probes, 5′-AGTGTCCCACGCTAAACTGC-3′ (forward) and 5′-TGAGTCCCAACTTCCTCCAG-3′ (reverse); GPR41 RNA probes, 5′-ATGGGGACAAGCTTCTTTCTTGGC-3′ (forward) and 5′-TCTGAGTGACAGCAATGTCTGC-3′ (reverse); GPR43 RNA probes, 5′-ATGACCCCAGACTGGCACAGTTC-3′ (forward) and 5′-TCTAGGTGGCATTTCCAAGC-3′ (reverse). The cloned product was verified by DNA sequencing and then used for making the DIG-labeled probes by DIG-RNA kit (Roche).


Double indirect immunohistochemistry techniques were performed as previously described (96). Briefly, frozen sections were rehydrated with PBS. Nonspecific binding was blocked with a blocking buffer [3% (vol/vol) BSA, 0.3% Triton X-100, 2% (vol/vol) donkey serum in 1× PBS] at room temperature for 1–2 h. Sections were incubated with two of the following primary antibodies from different species: rabbit anti-α-gustducin (1:250; sc-395), goat anti-GLP-1 (1:150; sc-7782), goat anti-GIP (1:300; sc-23554), goat anti-GPR40 (1:200; sc-28416), goat anti-GPR41 (1:200; sc-131166), and goat anti-GPR43 (1:200; sc-28424) from Santa Cruz Biotechnology; rabbit anti-GPR120 (1:150; NLS2004) or rabbit anti-GPR119 (1:200; NB110–92716) from Novus Biologicals; or rabbit anti-TGR5 (1:500; ab72608) from Abcam. Incubations with primary antibodies were carried out overnight at 4°C in a humidified chamber. After three 15-min washes with 1× PBS with 0.1% Tween 20 (PBST), the following secondary antibodies were added to the sections: Alexa488 donkey anti-rabbit and Alexa594 donkey anti-goat or Alexa647 donkey anti-goat (1:800–1:1,000; Invitrogen) for 2 h at room temperature in the dark. Next, the slides were washed three times with PBS and mounted with Vectashield mounting medium (Vector). Negative controls (e.g., minus primary antibody, plus blocking peptide, or comparison with nonimmune serum) were included for antibodies against fatty acid receptors (see Supplemental Figs. 1 and 2).1 Antibodies against α-gustducin (sc-395) (35, 51), GLP-1 (sc-7782) (35), and GIP (sc-23554) (24, 95) have been validated previously in the above-mentioned papers from our laboratory and/or from others.

For double-immunofluorescent labeling using primary antibodies from the same species, after the first primary incubation, the slides were incubated for 1 h at room temperature with secondary Fab donkey anti-rabbit antibody conjugated to Alexa488 (1:250; Jackson ImmunoResearch). Rabbit serum (10%) was applied for 1 h at room temperature to cover the first primary antibody. The slides were washed and incubated with an excess of unconjugated Fab anti-rabbit antibody for 2 h at room temperature. Next, the second primary antibody was applied overnight at 4°C after three washes with PBST. The slides were then incubated with the second secondary Fab donkey anti-goat conjugated to Alexa594 (1:250; Jackson ImmunoResearch) for 1 h at room temperature. Negative controls included the omission of both primary antibodies and the omission of the second primary antibody to confirm complete blocking of the first primary antibody. Only green fluorescence was observed when the second primary antibody was omitted.


Image visualization and capture were performed as previously described (96). Briefly, bright-field images were visualized using a SPOT digital camera (Diagnostic Instruments) attached to a Nikon SA Microphot microscope and processed using Image-Pro Plus image analysis software (Media Cybernetics). Acquisition parameters were held constant for in situ hybridization with both antisense and sense probes. Fluorescence images were captured with Ar, GeNe, and HeNe lasers as well as appropriate excitation spectra. Scanware software (Leica Microsystems) was used to acquire z-series stacks captured at a step size of 0.25–0.35 μm. Digital images were arranged using Photoshop CS (Adobe Systems).

Cell counting.

Quantitative measurements were conducted to determine the percentage of α-gustducin-labeled cells that coexpressed incretins GLP-1, GIP, or fatty acid receptors GPR40, GPR41, GPR43, GPR119, GPR120, and TGR5, and vice versa. The coexpression of GPR43 with GPR119, GPR120, or TGR5 was also quantified. The sections were visualized using a Nikon DXM1200C digital camera attached to a Nikon eclipse 80i fluorescence microscope. Only those cells for which the entire cell bodies could be visualized were counted. The data were expressed as the mean percentage from three sections ± SE.

Secretion of GLP-1 from proximal colon mucosa.

The protocol of Jang et al. (35) was modified to assess GLP-1 release from proximal colon mucosa. In brief, mucosa was obtained from the proximal colon by scraping with mild pressure from the short edge of a glass slide. The isolated tissue was allowed to settle at the bottom of a tube on ice and then washed three times with Dulbecco's PBS. After the final wash, purified mucosal tissue was suspended in serum-free Dulbecco's modified Eagle's medium (DMEM) with antibiotics and 20 μl/ml dipeptidyl peptidase IV inhibitor, separated into aliquots, and incubated with fatty acid agonists diluted in glucose-free Hanks' buffered salt solution with 10 mM HEPES (pH 7.0) in 5% CO2 for 2 h at 37°C. In a control set of experiments, ∼80% of cells were viable based on trypan blue staining after incubation for 2 h with 32 μM bile acids. The media were collected, and the released active GLP-1 levels were measured. At the end of experiment, the tissue was centrifuged and lysed with lysis buffer with dipeptidyl peptidase IV inhibitor. Levels of active GLP-1 in the supernatant were measured as a fraction of the cell GLP-1 content. GLP-1 secretion was calculated as the stimulated GLP-1 content of medium normalized for the total amount of GLP-1 in the medium plus cells. Dilutions were carried out so that active GLP-1 levels could be measured within the standard range of an ELISA (Millipore).

Statistical analysis.

GLP-1 data presented are means ± SEM. Data were analyzed, as appropriate for the data set, by ANOVA with Dunnett's post test, or by unpaired two-tailed Student's t-test (GraphPad Prism). P < 0.05 was considered significant.


Coexpression of α-gustducin with incretins in colon.

We examined coexpression of α-gustducin with GLP-1 and GIP in enteroendocrine cells ranging from duodenum to proximal colon. α-Gustducin immunofluorescence was detected in the mucosal epithelium of both small intestine and colon and some α-gustducin-expressing cells costained with GLP-1 and/or GIP (Fig. 1A). The number of α-gustducin-positive enteroendocrine cells increased markedly from duodenum (5.5 ± 1.5/section) to proximal colon (40.6 ± 4.5/section). More GLP-1-expressing cells were located in duodenum (17.7 ± 8.2/section), jejunum (34 ± 0.6/section), and proximal colon (38 ± 7.2/section); fewer were found in ileum (8.3 ± 3.0/section) (Fig. 1B). Similar results were found with GIP-positive cells: higher in duodenum (36.7 ± 2.9/section), jejunum (28 ± 0.6/section), and proximal colon (37.5 ± 4.5/section) but fewer in ileum (16.7 ± 4.5/section) (Fig. 1B). From duodenum to colon, both GLP-1-positive and GIP-positive cells showed increased coexpression with α-gustducin (α-gustducin+/GLP-1+ cells increased from 20 ± 7 to 70 ± 9%; α-gustducin+/GIP+ cells increased from 2.3 ± 2.3 to 83 ± 12%) (Fig. 1C). In small intestine, almost all α-gustducin-positive cells coexpressed GIP; in colon, ∼71 ± 17% of α-gustducin-positive cells coexpressed GIP; 90 ± 5% of α-gustducin-positive cells coexpressed GLP-1 in jejunum, the highest level of coexpression in gut. The lowest coexpression occurred in duodenum, where 48 ± 10% of α-gustducin-positive cells coexpressed GLP-1 (Fig. 1D). In small intestine, α-gustducin was mainly located in L/K cells and K cells, with relatively few L cells. However, in colon α-gustducin expression was nearly evenly distributed among L, K, and L/K cells (Fig. 1E).

Fig. 1.

Coexpression of α-gustducin with glucagon-like peptide 1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) in small intestine and colon. A: indirect immunofluorescent confocal imaging showing coexpression of α-gustducin (α-gust) with GLP-1 and GIP in jejunum (a–f) and proximal colon (g–l) (scale bar, 16 μm). B: quantitation of cells positive for GLP-1, GIP, and α-gustducin in different segments of gut. C: percentage of GLP-1- or GIP-positive cells in mouse proximal colon coexpressing α-gustducin. Values are means ± SE from three sections. D: percentage of α-gustducin-positive cells in mouse proximal colon coexpressing GLP-1 or GIP. Values are means ± SE from three sections. E: calculation of α-gustducin distribution in enteroendocrine L (GLP-1+), K (GIP+), and L/K (GLP-1+/GIP+) cells. The calculation was based on the same data set used to generate Fig. 1D.

Coexpression of α-gustducin with fatty acid receptors in colon.

Given the chemosensory function of α-gustducin in taste buds and small intestine, we speculated that, in colon, α-gustducin could be involved in luminal sensation through interactions with one or more G protein-coupled receptors. Because of the large quantity of SCFAs in the lumen of the colon (50), the known expression of SCFA receptors in L cell lines (37), and the expression of α-gustducin in L cells, we examined coexpression of SCFA receptors with α-gustducin in colon. Double-immunofluorescence confocal microscopy confirmed the presence of GPR43-positive cells in colon, ∼70% of which expressed α-gustducin (Fig. 2). We extended our studies to the other SCFA receptor, GPR41, and other receptors with agonists found in colon: LCFA receptors GPR40 and GPR120, bile acid receptor TGR5, and OEA receptor GPR119 (Fig. 2). Approximately 45% of α-gustducin-positive colon cells were also positive for GPR43 or TGR5, and <20% of those cells expressed GPR120 or GPR119 (Fig. 2C). Because in colon over 70% of α-gustducin-positive cells expressed GLP-1 and/or GIP (Fig. 1D), then we can infer that, as long as >30% of α-gustducin-positive cells express fatty acid receptors, expression of these fatty acid receptors must also overlap with that of GLP-1 and/or GIP. Thus, cells coexpressing fatty acids and α-gustducin are most likely endocrine cells expressing one or both of these gut hormones. Specificity controls for the antibodies are shown in Supplemental Figs. 1 and 2. Quantitation of the colon endocrine cells coexpressing α-gustducin and these receptors showed that >60% of GPR41-, TGR5-, and GPR119-positive cells coexpressed α-gustducin, 40% of GPR40-positive cells coexpressed α-gustducin, and 17% of GPR120-positive cells coexpressed α-gustducin (Fig. 2B).

Fig. 2.

Coexpression of α-gustducin with fatty acid and bile acid receptors in mouse colon. A: immunofluorescent detection of α-gustducin (green) and short-chain fatty acid (SCFA) G protein-coupled receptor (GPR) 43 and GPR41, long-chain fatty acid (LCFA) receptors GPR40 and GPR120, bile acid receptor TGR5, and oleoylethanolamide (OEA) receptor GPR119 (red) in mouse proximal colon (scale bar, 16 μm). B: percentage of fatty acid or bile acid receptor-positive cells in mouse proximal colon coexpressing α-gustducin. Values are means ± SE from three sections. C: percentage of α-gustducin-positive cells in mouse proximal colon coexpressing fatty acid or bile acid receptors. Values are means ± SE from three sections.

To independently confirm that these fatty acid receptors are expressed in mouse colon, we performed in situ hybridization for SCFA receptors GPR41 and GPR43 and LCFA receptor GPR40. mRNAs for the GPR40, GPR41, and GPR43 receptors were found to be expressed in a subset of epithelial cells in mouse colon (Supplemental Fig. 3). Note that the number of positive cells for GPR41 mRNA was much lower than that of GPR40 and GPR43. Taken together, the frequent coexpression of α-gustducin with fatty acid and bile acid receptors suggests that, in colon, α-gustducin might couple to these receptors to mediate signal transduction.

Coexpression of GPR43 with GPR120 and GPR119 but not with TGR5 in colon.

Next we examined whether these fatty acid and bile acid receptors were coexpressed in colon epithelial cells. Because of the high percentage of cells coexpressing GPR43 and α-gustducin, and limitations from the species of origin of our primary antibodies, double-labeled immunofluorescence was performed only with primary antibodies against GPR43 (raised from goat) and GPR120, GPR119, or TGR5 (all raised from rabbit). As shown in Fig. 3A, a subset of GPR43-positive cells also expressed GPR119 and GPR120. More than 60% of GPR119- and GPR120-positive cells expressed GPR43, and ∼80% of GPR43-staining cells also expressed GPR119 or GPR120 (Fig. 3, B and C). These data suggest that multiple fatty acid-sensing pathways could be present in a single cell. Note, however, that we found GPR43 and TGR5 to not be coexpressed together in mouse colon epithelial cells (Supplemental Fig. 4).

Fig. 3.

GPR43 is coexpressed with GPR119 and GPR120. A: immunofluorescent detection in mouse proximal colon of SCFA receptor GPR43 (red) and OEA receptor GPR119 or LCFA receptor GPR120 (green) (scale bar, 16 μm). B: percentage of GPR119-positive or GPR120-positive cells coexpressing GPR43. Values are means ± SE from three sections. C: percentage of GPR43-positive cells coexpressing GPR119 or GPR120. Values are means ± SE from three sections.

GLP-1 release in response to fatty acids in colonic tissue from α-gustducin knockout mice.

Given the coexpression of α-gustducin and GLP-1 with fatty acid or bile acid receptors, we hypothesized that activation of one or more of these receptors could couple to α-gustducin to induce secretion of GLP-1. To examine this hypothesis, GLP-1 secretion in response to fatty acids and bile acids was measured in colonic tissue isolated from wild-type and α-gustducin knockout mice. Basal GLP-1 release was not significantly different between the two groups (Fig. 4A). Treatment of wild-type tissue with acetate (80 or 160 mM), oleic acid (9.6 or 14.4 μM), α-linolenic acid (α-LA, 96 μM), OEA (4.8 μM), and lithocholic acid (LCA, 9.6 and 32 μM) stimulated GLP-1 secretion significantly compared with the nontreated group (Fig. 4, B–D). In contrast, α-gustducin knockout colon tissue did not display any increase of GLP-1 release in response to the tested SCFAs and LCFA oleic acid (Fig. 4, B and C). α-LA (96 μM), a potent agonist for both LCFA receptors GPR40 and GPR120 (28, 34), only slightly increased release of GLP-1 in α-gustducin knockout colon tissue (Fig. 4C). Surprisingly, with the concentrations of propionate and butyrate tested, we did not observe an increase of GLP-1 secretion from wild-type tissue. This contrasts with receptor activation seen in heterologous cell lines cotransfected with GPR41 or GPR43 and Giα (6, 47). This difference could be because of higher expression in heterologous systems or differences in ligand response due to coupling of the receptors to α-gustducin in primary colon tissue vs. Giα in the cell lines.

Fig. 4.

Impaired GLP-1 secretion from colon tissue of α-gustducin knockout mice in response to fatty acids. A: basal GLP-1 release in glucose-free, BSA-free Hanks' buffered salt solution from colon mucosa of α-gustducin knockout (filled bar) vs. wild-type (open bar) mice. B–D: GLP-1 secretion in response to SCFAs, LCFAs, OEA, and bile acids from colon mucosa of α-gustducin knockout (filled bar) vs. wild-type (open bar) mice. Data were generated in triplicate for each mouse, with n = 3–6 mice for each genotype, and are expressed as means ± SE. Statistical significance was determined by ANOVA and Student's unpaired two-tailed t-test. *P < 0.05, **P < 0.01, and ***P < 0.001, gustducin knockout vs. wild-type mice. #P < 0.05 and ###P < 0.001, fatty acid-treated vs. basal GLP-1 release.

OEA at high concentration (9.6 μM), but not at low concentration (4.8 μM), decreased GLP-1 secretion from colon tissue of wild-type mice (Fig. 4D). A similar trend has been shown in an L-type enteroendocrine cell line (46), increasing GLP-1 release at a low dose (5–10 μM) and decreasing it at a high dose (15 μM), although the primary colonic tissue in our study exhibited higher sensitivity to OEA, with an inhibition of GLP-1 secretion at 9.6 μM. The opposite effect was observed in α-gustducin knockout colon tissue where the high concentration of OEA (9.6 μM) induced GLP-1 secretion (Fig. 4D). This could be because of a right shift of the OEA response curve in the α-gustducin knockout tissue to become less sensitive to the low concentration (4.8 μM) of OEA. The bile acid LCA increased GLP-1 secretion in colon tissue from both wild-type and α-gustducin null mice. However, colon tissue from α-gustducin null mice showed a significantly lower fold increase in the GLP-1 response to LCA compared with that of wild-type tissue (Fig. 4D).


The G protein subunit α-gustducin is expressed in ∼20% of taste receptor cells, where it is a key mediator of bitter, sweet, and umami (glutamate) taste transduction (27, 55, 91). α-Gustducin is also present in enteroendocrine cells and brush cells of the gastrointestinal tract in both humans and rodents (3, 29, 74, 82). In enteroendocrine cells of the small intestine, α-gustducin is frequently coexpressed with the incretin hormones GLP-1 and GIP (35, 51). α-Gustducin is also present in GLP-1-positive enteroendocrine cells in colon (3, 74). Mice lacking α-gustducin display impaired GLP-1 secretion in response to glucose gavage (35). Furthermore, sweetener-stimulated GLP-1 release from isolated small intestine from α-gustducin null mice is deficient. Experiments with enteroendocrine cell lines implicate both α-gustducin and the T1r2+T1r3 sweet receptor in GIP and GLP-1 release (35, 51).

GLP-1 is secreted from intestinal endocrine L cells, which are located mainly in the distal ileum and colon. In contrast, GIP is released from intestinal K cells, which are localized to more proximal regions (duodenum and jejunum) of the small intestine. However, endocrine cells that produce either GLP-1 or GIP, as well as cells that produce both peptides (L/K cells), can be found throughout all regions of the small intestine (see review in Refs. 2 and 30). In the present study, we find that there are many more α-gustducin-expressing cells in colon than in small intestine. In colon, α-gustducin is frequently coexpressed with GLP-1, GIP, or both. Interestingly, in duodenum and jejunum, very few of the α-gustducin-expressing cells were found to be L cells (i.e., expressed only GLP-1), instead these cells expressed either GIP (K cells) or both GIP and GLP-1 (L/K cells). Importantly, the percentage of L-type cells among the α-gustducin-expressing cells increased from near 0% in small intestine to ∼29% in colon, suggesting that colonic L cells could histologically and functionally differ from the L cells in the small intestine.

In human subjects with ileostomies, the circulating GLP-1 levels after a fat load are significantly reduced compared with healthy control subjects, suggesting that loss of colonic endocrine tissue is an important determinant in regulating postprandial GLP-1 levels (72). In vivo studies of rodents and humans have shown that increased SCFA production after dietary fiber ingestion increases GLP-1 levels (19, 70, 97). A recent study has found that SCFAs trigger secretion of GLP-1 by increasing cytosolic Ca2+ in L cells in primary culture. Mice lacking SCFA receptors GPR41 or GPR43 exhibit reduced SCFA-triggered GLP-1 secretion in vivo and ex vivo (88). These studies suggest that direct stimulation of colonic L cells with SCFAs could promote GLP-1 secretion.

GPR41 and GPR43 mRNAs are widely expressed in various tissues, including adipose tissue, intestine, and immune cells (6, 22, 31, 41, 47, 52, 61). In addition, GPR43 is expressed in GLP-1-positive colon cells in both humans and rats (37). In vivo studies have shown that GPR41 is tightly involved in energy homeostasis by stimulating leptin production (93), reducing caloric extraction (76), and promoting activity of the sympathetic nervous system (41). GPR43-deficient mice display exacerbated inflammation in dextran sulfate sodium-induced colitis (52) and are protected from high-fat-diet-induced metabolic syndrome (4). Ex vivo studies show that activation of GPR43 leads to neutrophil chemotaxis (47, 89) and inhibition of lipolysis in adipocytes (22, 31). Human GPR41 and GPR43 receptors can be activated by short carboxylic acids containing one to six carbons (6, 47, 61). In heterogeneously expressed cell lines, activation of GPR41 or GPR43 inhibits cAMP accumulation and induces inositol trisphosphate formation, intracellular Ca2+ release, and phosphorylation of p42/44 mitogen-activated protein (MAP) kinase (6, 47). In addition, pertussis toxin abolishes the Ca2+ response elicited by GPR41 but not that of GPR43. This suggests that GPR43 may activate Gq in addition to Gi family G proteins that include gustducin and the transducins (6, 47).

The LCFA receptor GPR40 is highly expressed in pancreatic β-cells (5, 18, 34, 42) and also found in endocrine cells in gut where it is frequently coexpressed with gastrin, the incretin hormones GLP-1 and GIP, ghrelin, cholecystokinin (CCK), PYY, secretin, or serotonin (15). GPR40 plays a role in acute free fatty acid (FFA)-induced secretion of insulin (15, 34, 43, 45, 57, 77, 78, 81), GIP, GLP-1 (15), and CCK (49). Medium-chain fatty acids and LCFAs activate GPR40, which in cell lines and rat β-cells couples to insulin release via a Gq-phospholipase C (PLC)-Ca influx pathway (5, 21, 34, 42, 77, 78). In addition, activation of GPR40 by FFA also reduces the voltage-gated K+ current in rat β-cells through protein kinase A, leading to an increase in intracellular Ca2+ and insulin secretion (16).

GPR120 is highly expressed in mature adipocytes, macrophages, enteroendocrine GLP-1-positive cells, and K cells (28, 63, 65). GPR120 is involved in adipogenic processes (25), inhibition of cytokine release (63), and secretion of GLP-1 and CCK from enteroendocrine cell lines (28, 32, 83). GPR120-deficient mice on a high-fat diet display more severe insulin resistance (63), obesity, and liver steatosis (33), suggesting a protective role of GPR120 in metabolic syndrome. In addition, GPR120 and GPR40 are also expressed in taste papillae of mice: GPR120 is mainly in type II taste cells (53); GPR40 is in type I and type II cells (7). In vivo studies have shown that GPR40 and GPR120 mediate the taste of fatty acids (7). Medium- and long-chain FFAs activate GPR120 and evoke a rise in intracellular Ca2+ and increase the amount of phosphorylated extracellular signal-regulated p42/44 MAP kinase in heterogeneously expressed cell lines and in the STC-1 enteroendocrine cell line (28). However, FFA-induced GLP-1 secretion in GLUTag cells is involved in PKCζ but not protein kinase B, MAP kinase, or Ca2+ responses (32), suggesting that GPR120 could couple to diverse downstream signaling elements to mediate GLP-1 release.

High levels of TGR5 mRNA are detected in spleen, lung, liver, stomach, intestine, adipose tissue, gallbladder, and bone marrow in human and mouse (39, 66). Activation of TGR5 induces GLP-1 secretion in vivo and ex vivo (20, 38, 85), improves glucose tolerance and enhances insulin secretion (85), and prevents and reverses diet-inducing obesity (85, 90). In addition, TGR5 inhibits cytokine production in macrophages and inhibits macrophage foam cell formation, reduces vascular lesion formation (66), and attenuates colon inflammation in rodent models of colitis (10). Activation of TGR5 in response to bile acids induces cAMP production and increases MAP kinase activity in heterogeneously transfected cell lines (39, 40, 90), suggesting that it couples via a G protein to increase cAMP.

GPR119 is expressed in β-cells of pancreatic islets (9, 44) and GLP-1-positive cells in the gastrointestinal tract (8). In vivo studies have shown that activation of GPR119 stimulates GLP-1 and GIP release, improves glucose tolerance, and increases insulin release (8, 11, 17, 46). In addition, GPR119 also plays a role in decreasing food intake and body weight gain in diet-induced obese rats (64), reducing gastric emptying (17), and inhibiting Cl secretion in colonic mucosa (11). OEA, the most efficacious ligand for both human and mouse GPR119, increases cAMP levels in heterogeneous cells (64). Activation of GPR119 increases insulin secretion in pancreatic β-cell lines via both cAMP accumulation and increased intracellular Ca2+ (62, 80) and stimulates GLP-1 release from intestinal cell lines via cAMP accumulation (8, 46).

In the present study, we have found that intestinal α-gustducin is coexpressed with several fatty acid receptors and the bile acid receptor TGR5. In taste cells, heterotrimeric gustducin decreases cAMP (via α-subunit activation of phosphodiesterase) (56, 75) and increases intracellular Ca2+ levels (via βγ-subunits' actions on PLC) (94). It is likely that gustducin is involved in GLP-1 release from intestinal cells in response to SCFAs and LCFAs. This is supported by our observation of impaired GLP-1 release in mucosal tissue from α-gustducin knockout mice in response to fatty acids. We also found that intestinal tissue from α-gustducin knockout mice displayed impaired GLP-1 release in response to OEA and bile acid, suggesting that α-gustducin also participates in GPR119- and TGR5-activated pathways in enteroendocrine cells. Coexpression of multiple fatty acid receptors together with α-gustducin in intestine suggests that multiple sensing pathways are present in endocrine cells in the gut for efficient stimulation of GLP-1 release. Given the broad range of fatty acids in colon lumen under physiological conditions, a combination of SCFAs and LCFAs could act additively as potent agonists to induce GLP-1 secretion.

After oral nutrient ingestion, plasma GLP-1 levels rise in a biphasic pattern starting with an early phase (peak within 10–15 min) followed by a longer (30–60 min) second phase. Two models to explain GLP-1 release after meal ingestion have been proposed: a proximal-distal loop and a direct stimulation of L cells by luminal sugars (87). The proximal-distal loop proposes that nutrient detectors located in upper regions of the gastrointestinal tract control the release of GLP-1 from distal L cells by neural, most likely vagal (73), or hormonal (for example, GIP) (71) mechanisms. The potential mechanisms underlying direct stimulation of L cells by luminal sugars include KATP (68), SGLT1 (26, 69), and sweet taste receptors (35). Given the abundance of GLP-1-positive cells in the colon, GLP-1 release in response to local fatty acids could contribute to the plasma GLP-1 increase after a meal. Meals can induce colon motor activity within 10 min through the gastrocolic reflex (67, 79, 92). This reflex could change the colonic fatty acid concentrations and stimulate L cells in colon to release GLP-1 during the early phase of a meal.

Our data indicate that α-gustducin in colon is a key signaling molecule coupling fatty acid and bile acid receptors to GLP-1 secretion. Agonists that enhance GLP-1 secretion from the α-gustducin-expressing colon cells that respond to FFAs may serve as a novel target for the treatment of diabetes.


This research was supported by National Institutes of Health (NIH) Grant DK-081421 to R. F. Margolskee. Imaging was performed at the Monell Histology and Cellular Localization Core, which is supported, in part, by funding from the NIH Core Grant 1P30DC-011735–01.


The authors declare that they have no conflicts of interest.


Author contributions YL, ZK, BM and RFM designed experiments; YL collected and analyzed data; all authors contributed to writing the manuscript.


We thank Dr. Karen K. Yee for confocal microscopy training.


  • 1 Readers are herein alerted to the fact that supplementary figures for this paper may be found at the institutional website of one of the authors, which at the time of publication they indicate is: [http://www.monell.org/supplemental/rm0113]. These materials are not a part of this manuscript, and have not undergone peer review by the American Physiological Society (APS). APS and the journal editors take no responsibility for these materials, for the website address, or for any links to or from it.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
  69. 69.
  70. 70.
  71. 71.
  72. 72.
  73. 73.
  74. 74.
  75. 75.
  76. 76.
  77. 77.
  78. 78.
  79. 79.
  80. 80.
  81. 81.
  82. 82.
  83. 83.
  84. 84.
  85. 85.
  86. 86.
  87. 87.
  88. 88.
  89. 89.
  90. 90.
  91. 91.
  92. 92.
  93. 93.
  94. 94.
  95. 95.
  96. 96.
  97. 97.
View Abstract