Postprandial Stimulation of GIP-Secretion 2 2 3 3 4 4 Frank Reimann, Eleftheria Diakogiannaki, Catherine E Moss, Fiona M

Home , G cell

*Manuscript Click here to download Manuscript: Reimann et al GIP secretion 24sep19.docx Click here to view linked References

1 1 Postprandial stimulation of GIP-secretion 2 2 3 3 4 4 Frank Reimann, Eleftheria Diakogiannaki, Catherine E Moss, Fiona M. Gribble 5 6 5 7 6 From the Wellcome Trust/MRC Institute of Metabolic Science (IMS), 8 9 7 University of Cambridge, United Kingdom 10 8 11 9 12 10 13 11 14 15 12 16 13 17 14 Corresponding authors: 18 15 Frank Reimann and Fiona M. Gribble 19 20 16 Wellcome Trust/MRC Institute of Metabolic Science (IMS) 21 17 MRC Metabolic Diseases Unit 22 18 University of Cambridge, 23 19 Addenbrooke's Hospital, Hills Road 24 20 Cambridge CB2 0QQ, 25 26 21 Tel: 0044-1223-746796 or 336746 27 22 email: [email protected] or [email protected] 28 23 29 24 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 1 64 65 25 Abstract: 1 2 3 26 Glucose-dependent insulinotropic polypeptide (GIP) is a gut hormone secreted from the 4 5 27 upper small intestine, which plays an important physiological role in the control of glucose 6 7 8 28 metabolism through its incretin action to enhance glucose-dependent insulin secretion and 9 10 29 has also been implicated in postprandial lipid homeostasis. GIP is secreted from 11 12 13 30 enteroendocrine K-cells residing in the intestinal epithelium, which sense a variety of 14 15 31 components found in the gut lumen following food consumption, resulting in a circulating 16 17 18 32 plasma GIP signal dependent on the nature and quantity of ingested nutrients. In this review, 19 20 33 we explore the mechanisms underlying the control of GIP secretion, which have been 21 22 34 23 identified through combinations of in vivo, in vitro and molecular approaches. 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 2 64 65 35 Introduction: 1 2 36 3 4 5 37 GIP was originally isolated from the proximal small intestine as a peptide inhibiting gastric 6 7 38 acid secretion, and named accordingly gastric inhibitory peptide [1]. However, based on the 8 9 10 39 sequence homology of this peptide to other secretin family members known to stimulate 11 12 40 insulin secretion it was hypothesised that GIP might actually be an incretin, a hormone 13 14 41 15 released from the intestine in response to oral, but not intravenous, glucose, that stimulates 16 17 42 insulin secretion and indeed it was subsequently shown that exogenous GIP co-administration 18 19 43 with glucose had an insulinotropic effect [2]. The effect on gastric acid production, whilst 20 21 22 44 clear in dogs, was only seen at very high (supraphysiological) concentrations in man, which 23 24 45 resulted in its renaming to Glucose-dependent insulinotropic polypeptide, thus keeping the 25 26 27 46 GIP acronym (see [3] for a historical review). 28 29 47 Soon after the amino acid sequence of the GIP peptide was published [1] (a minor correction 30 31 32 48 to the sequence reducing it from 43 to 42 amino acids was published later [4]), antisera were 33 34 49 raised and used to identify the cellular origin of this hormone. Enteroendocrine cells (EECs) 35 36 50 found in the duodenum and slightly less frequently in the jejunum [5] staining for GIP were 37 38 39 51 originally thought to be a population known as D1-cells, but were subsequently identified as 40 41 52 K-cells, both named to distinguish them from other EECs which differ in their vesicular 42 43 44 53 appearance in “ultrastructural” electron-microscopic preparations, with some other EECs 45 46 54 already named after the hormones associated with them, such as the gastrin producing G-cell 47 48 49 55 [6, 7]. In recent years the classification of EECs into many different cell types based on 50 51 56 vesicular appearance after diverse staining techniques and correlation with one predominant 52 53 57 hormone has, however, been challenged, based on at the time unexpected co-localization of 54 55 56 58 hormones, such as GIP and the sister incretin glucagon-like peptide-1 (GLP-1) occasionally 57 58 59 in the same cells [8, 9], and even more promiscuous hormone expression profiles observed in 59 60 61 62 63 3 64 65 60 transgenic animals expressing genetic tags in specific hormone expressing cells [10, 11]. 1 2 61 Indeed, EECs are now thought to change their expression profiles whilst maturing along the 3 4 5 62 crypt villus or crypt-surface epithelial axis [12-14]. Nonetheless, hormones have preferential 6 7 63 expression profiles along the proximal to distal gut axis and GIP is found predominantly in 8 9 10 64 the very proximal small intestine, the duodenum and jejunum [15, 16]. In mice the overlap 11 12 65 with GLP-1 in the same cells is relatively limited, such that most duodenal GLP-1 expressing 13 14 66 15 cells do not co-express GIP and vice versa [17, 18] and we will continue to use the K-cell 16 17 67 terminology for GIP-expressing cells throughout this review, in which we will focus on the 18 19 68 sensing of secretory stimuli by these cells. 20 21 22 69 23 24 70 Sensing of macro-nutrients by K-cells 25 26 27 71 28 29 72 GIP levels in the circulation rise rapidly following food ingestion, or when nutrients are 30 31 32 73 infused into the duodenum. Infusion studies in humans have demonstrated that GIP secretion 33 34 74 is strongly determined by the rate of nutrient infusion (e.g. of glucose), suggesting that 35 36 75 circulating GIP levels predominantly reflect the rate of nutrient delivery from the stomach 37 38 39 76 into the duodenum, and hence the rate of gastric emptying [19] although the glucose 40 41 77 absorption rate in the duodenum/jejunum will also affect the signal [20]. As described below, 42 43 44 78 a range of different nutrient and non-nutrient components have been found to modulate GIP 45 46 79 secretion in different model systems. 47 48 49 80 50 51 81 Carbohydrates: 52 53 82 54 55 56 83 Oral, but not intravenous glucose was shown to be a strong stimulant of endogenous GIP 57 58 84 secretion as early as 1974 [21], which led to its classification as an incretin, as exogenous 59 60 61 62 63 4 64 65 85 GIP had also been shown to stimulate insulin secretion [2]. In the following decade it became 1 2 86 clear that carbohydrate that were substrates for sodium-dependent glucose transporters 3 4 5 87 (SGLT; Slc5a1), such as glucose, galactose and even non-metabolisable glucose analogues 6 7 88 such as -methyl-glucose pyranoside (MDG) were good GIP stimulants and that inhibition 8 9 10 89 of active sodium-dependent glucose uptake with phloridzin prevented glucose stimulated GIP 11 12 90 release [22]. We revisited this once we had established mixed intestinal epithelial cultures 13 14 15 91 and fluorescently labelled K-cells [23]. When cultures were co-treated with forskolin and 16 17 92 IBMX to elevate cytosolic cAMP levels, glucose and MDG were reliable stimulants of GIP 18 19 20 93 secretion (2-3-fold stimulation compared to forskolin/IBMX alone) and these responses were 21 22 94 abolished by phloridizin. SGLT1 was restricted to the apical surface of K-cells [23] and thus 23 24 25 95 ideally placed to respond to luminal changes in glucose concentration – comparable to our 26 27 96 previous observations in GLP-1 secreting L-cells [24, 25] - and we concluded that direct 28 29 97 depolarization through coupled glucose and sodium uptake into K-cells resulted in activation 30 31 2+ 32 98 of voltage gated Ca -channels triggering GIP secretion. This mechanism is supported by the 33 34 99 fact that SGLT1 knock-out animals lacked the rapid rise in plasma GIP levels normally 35 36 37 100 observed after glucose ingestion [26]. Interestingly, even at later time points (up to 2 h after 38 39 101 oral glucose challenge), no GIP response was observable in SGLT-1 knock-out animals, 40 41 42 102 contrasting with GLP-1 responses, which whilst compared to wild-type animals are reduced 43 44 103 at early time points (<15 minutes after glucose ingestion) are exaggerated at later time points 45 46 104 (1 and 2 h), probably reflecting delivery of glucose to the more distal intestine where it could 47 48 49 105 stimulate a greater number of L-cells presumably after being converted to SGLT-1 50 51 106 independent stimuli such as short-chain fatty acids by the microbiota [27]. This different 52 53 54 107 effect of GIP inhibition, but GLP-1 elevation following an oral glucose tolerance test is also 55 56 108 observed in human volunteers pre-treated with a mixed SGLT-1/2 inhibitor [28]. 57 58 59 109 60 61 62 63 5 64 65 110 Despite good evidence for a predominantly SGLT-1 dependent mechanism, other possible 1 2 111 glucose sensing pathways have been suggested to play a role in GIP secretion. The finding 3 4 5 112 that K-cells express all the machinery employed by the pancreatic beta cell, including 6 7 113 glucokinase [29], which channels glucose into glycolysis to eventually result in closure of 8 9 10 114 ATP-sensitive potassium (KATP) channels, that otherwise keep beta-cells hyperpolarized thus 11 12 115 preventing activation of voltage-gated Ca2+-channels under conditions of low ambient 13 14 116 15 glucose, suggested that this pathway is also employed for glucose sensing in K-cells. Indeed 16 17 117 tolbutamide, a KATP-channel inhibitor, stimulated GIP secretion from mixed epithelial 18 19 118 cultures, but this effect was lost in forskolin/IBMX treated cultures [23] and as tolbutamide 20 21 22 119 also affects Epac2 [30], a cAMP sensor expressed in K-cells, this might be an off-target 23 24 120 effect. In wild-type mice glimepiride, a sulfonylurea, failed to elicit a GIP-response [31] and 25 26 27 121 although the authors observed some role for KATP in GIP secretion in streptozotocin treated 28 29 122 diabetic mice [31] no obvious change in the GIP response to an oral glucose tolerance test 30 31 32 123 was observed in patients with type 2 diabetes after initiation of sulfonylurea treatment [32]. 33 34 124 Work with STC-1 derived cell lines, which express high levels of GIP [33], has also disputed 35 36 125 any role of K -channels in GIP secretion, but, although the particular cell lines employed in 37 ATP 38 39 126 these studies lacked any glucose-stimulated GIP secretion, nevertheless a metabolic sensing 40 2+ 41 127 mechanism involving AMPK-dependent kinase and Ca -release from intracellular stores was 42 43 44 128 postulated [34-36]. It seems the KATP-channel whilst expressed in K-cells plays at best a 45 46 129 modulatory role in glucose-stimulated GIP secretion. 47 48 49 130 50 51 131 GIP-secretion was also reported in response to sucralose, an agonist on the sweet taste 52 53 132 receptor (Tas1R2/3), from GLUTag cells, which however do not express high levels of GIP. 54 55 56 133 This effect was blocked by the Tas1R2/3 antagonist gurmarin, and the sweet taste receptor 57 58 134 was thus put forward as a luminal glucose sensor for incretin release [37]. In our laboratory 59 60 61 62 63 6 64 65 135 we do not see any significant GIP response to sucralose from primary intestinal cultures [23] 1 2 136 nor is there any significant contribution of this mechanism to incretin release in the perfused 3 4 5 137 rat intestine [38] and most reports do not support an acute GIP (or GLP-1) response to 6 7 138 artificial sweetener intake in man (for example [39, 40]). A luminally facing sweet taste 8 9 10 139 receptor seems also difficult to reconcile with the lack of a GIP response in SGLT1 knock- 11 12 140 out animals, given that glucose would linger around at the luminal side for longer in these 13 14 141 15 animals, providing greater opportunity to stimulate such a receptor. However, others have 16 17 142 come to different conclusions, as sweet taste receptor activation can upregulate SGLT-1 18 19 143 expression and there might be a complicated interplay between these pathways [41]. 20 21 22 144 Theoretically it remains possible that there is a complex interaction in which sweet taste 23 24 145 receptor activation only stimulates GIP (or GLP-1) secretion, when the EECs are 25 26 27 146 concomitantly depolarized by SGLT1-dependent glucose uptake – however, this is very 28 29 147 different from the idea that Tas1R2/3 encodes the critical glucose sensing moiety in K-cells, 30 31 32 148 and the selective sweet-taste receptor antagonist gurmarin did also not reduce glucose 33 34 149 stimulated GIP secretion in the perfused rat small intestine [38]. 35 36 150 37 38 39 151 There is mixed evidence for fructose, a (sweet) monosaccharide that is not a substrate for 40 41 152 sodium-dependent glucose transport, as a stimulant for GIP secretion. K-cells do express the 42 43 44 153 fructose transporter GLUT5 and fructose did modestly stimulate secretion in primary 45 46 154 intestinal epithelial cultures, possibly reflecting a metabolic component of stimulation 47 48 49 155 observable at basal cAMP levels and comparable to the tolbutamide triggered response under 50 51 156 these conditions [23]. However, no significant fructose stimulated GIP secretion was 52 53 157 observed in perfused intestinal preparations from rats [22] or after oral ingestion in rats, mice 54 55 56 158 and humans [42], although some response was seen when fructose was administered orally to 57 58 159 diabetic mice (after streptozotocin treatment) [43] or very obese ob/ob mice [44]. Whether 59 60 61 62 63 7 64 65 160 this latter observation has anything to do with a reported increase in K-cell density in this 1 2 161 model [45, 46] or the reported over-secretion of GIP in obese humans in response to other 3 4 5 162 stimuli including glucose and fat [47], has to our knowledge not been fully investigated. 6 7 163 However and in contrast, reduced GIP-responses to a mixed meal have been reported in 8 9 10 164 morbidly obese individuals when compared to lean subjects, which could be correlated to 11 12 165 delayed gastric emptying [48]. 13 14 166 15 16 17 167 Lipids: 18 19 168 20 21 22 169 Even though GIP is considered an incretin, boosting insulin secretion in the context of 23 24 170 elevated plasma glucose, when isocaloric intraduodenal lipid and glucose infusions are 25 26 27 171 compared head to head more GIP is secreted in response to the fat stimulus[49]; as blood 28 29 172 glucose is not elevated under these conditions, this does not result in an elevation of plasma 30 31 32 173 insulin [49, 50]. This finding, whilst not reproduced in every study (e.g. [51]) presumably 33 34 174 reflects an important role of GIP in the control of adipocyte metabolism discussed elsewhere 35 36 175 in this issue. Early investigations into the mechanisms underlying lipid stimulated GIP 37 38 39 176 secretion established that only longer chain fatty acids (>C10) were effective stimulants and 40 41 177 that non-saturated fatty acids were more effective compared with their saturated C18- 42 43 44 178 counterpart [52]. Once it became clear that K-cells express G-protein-coupled receptors for 45 46 179 long-chain free fatty acids (GPR40/FFAR1 and GPR120/FFAR4) [23], research focused on 47 48 49 180 these, as has been reviewed previously (e.g.[53]). Some controversy still exists about the 50 51 181 importance of these receptors, especially FFAR4, for GIP secretion; whereas Ekberg et al. 52 53 182 reported, consistent with the literature [54], a GIP secretory defect in response to oral olive 54 55 56 183 oil in FFAR1 single-knock out mice, they failed to observe a similar defect in FFAR4 single 57 58 184 knock-out mice [55]. By contrast, Iwasaki et al. reported a 75% reduction of GIP secretion in 59 60 61 62 63 8 64 65 185 response to lard oil in FFAR4 knock-out mice [56], although, subsequently the group 1 2 186 reported that the FFAR4 effect likely resided in cholecystokinin (CCK)-secreting I-cells, as 3 4 5 187 CCK-replacement was able to recover corn oil induced GIP-secretion in FFAR4, but not 6 7 188 FFAR1 single knock-out mice [57]. The effect on GIP secretion was proposed to be a 8 9 10 189 reduction in lipid digestion, as the reduced CCK secretion would result in reduced bile and 11 12 190 pancreatic lipase delivery to the intestinal lumen. Both groups, however, showed that FFAR1 13 14 191 15 and FFAR4 are highly expressed in K-cells and that double knock-out of the two receptors 16 17 192 had a more profound effect than either single knock-out on lipid stimulated GIP secretion 18 19 193 [55, 58]. In our laboratory we did observe a reduced responsiveness to a mixed corn oil/olive 20 21 22 194 oil gavage in FFAR4 single knock-out mice in vivo (Figure 1a). As free fatty acids proved to 23 24 195 be a fairly poor stimulus of GIP secretion when tested in isolation in mixed epithelial cultures 25 26 27 196 [23], we investigated this further in vitro using more complex mixtures of lipids, aiming to 28 29 197 simulate the composition of post prandial micelles (PPM), which contain bile acids and 30 31 32 198 mono-acyl-glycerides in addition to free fatty acids. This mixture triggered a profound 33 34 199 secretory GIP response in mixed epithelial cultures (Figure 1b), probably reflecting 35 36 200 simultaneous stimulation of free fatty acid receptors and receptors for the other PPM 37 38 39 201 components, like GPR119 (mono-acylglyceride) and GPBAR1 (also known as TGR5; bile 40 41 202 acid). These predominantly Gs-coupled receptors are also expressed in K-cells, as synergy 42 43 44 203 between Gs- and Gq coupled receptors for incretin secretion has been demonstrated [55]. 45 46 204 Note, however, that raising cAMP with forskolin and IBMX was insufficient to enable strong 47 48 49 205 responses to free fatty acids in mixed epithelial intestinal cultures [23], suggesting that 50 51 206 additional signaling pathways other than adenylate cyclase stimulation downstream of Gs 52 53 207 might play a role. Interestingly, in this experimental setup both FFAR1 and FFAR4 single 54 55 56 208 knock-out significantly reduced PPM-stimulated GIP secretion (Figure 1b/c). As, in this in 57 58 209 vitro system, no change in bile or lipase availability would occur, fatty-acid dependent 59 60 61 62 63 9 64 65 210 signaling of both FFAR1 and 4 within K-cells seems likely, although we cannot exclude that 1 2 211 paracrine cross-talk between different cells within the culture contributes to the observed 3 4 5 212 effects. In L-cells, FFAR1 has been linked to activation of TrpC3 [53, 59], whereas FFAR4 6 7 213 has been linked to TrpM5 activation in linoleic acid stimulated CCK-release from the STC1 8 9 10 214 cell line [60], which also produces GIP and GLP-1. Future research should address whether 11 12 215 FFAR1 and FFAR4 can recruit different signaling cascades in K-cells. 13 14 216 15 16 17 217 In GLP-1 secreting L-cells, fatty acids appear to stimulate the cells from the basolateral 18 19 218 direction, rather than the intestinal lumen facing apical cell surface, consistent with a small 20 21 22 219 molecule agonist of FFAR1 only stimulating secretion when applied from the 23 24 220 basolateral/vascular side [61]. We are unaware of similar studies showing a dependence of 25 26 27 221 lipid stimulated GIP secretion on lipid absorption, but interference with chylomicron 28 29 222 formation by co-application of the surfactant pluronic L-81 virtually abolished GIP-secretion 30 31 32 223 in response to intraduodenal lipid application in rats [62]. It has been suggested that free fatty 33 34 224 acids might be locally released from chylomicrons produced and secreted by nearby 35 36 225 enterocytes [63], stimulating both GIP and GLP-1 secretion, although there was clearly also 37 38 39 226 an FFAR1-independent sensing mechanism for chylomicrons observable in mixed epithelial 40 41 227 cultures [63]. A dependence of GIP-secretion on fat absorption is further supported by studies 42 43 44 228 interfering with MGAT2 and DGAT1, two enzymes involved in the step-wise re- 45 46 229 esterification of mono-acylglycerides to tri-acylglycerides inside the intestinal epithelium 47 48 49 230 during lipid absorption. Both MGAT2 and DGAT1 knock-out animals show reduced GIP- 50 51 231 responses to an oil gavage at all time points [64] and similar blunted GIP-responses were 52 53 232 observed after pretreating mice with a DGAT1 inhibitor [65]. Lipid stimulated GLP-1 and 54 55 56 233 PYY excursions in these models were more complex, with elevated plasma levels at later 57 58 234 time points, presumably reflecting increased delivery of un-absorbed lipids to the distal 59 60 61 62 63 10 64 65 235 intestine, where more of these hormones are produced, similar to the observations in SGLT-1 1 2 236 knock-out animals after a glucose gavage. 3 4 5 237 6 7 238 8 9 10 239 Despite the apparent importance of the G-protein coupled receptors for sensing of long-chain 11 12 240 fatty acids (FFAR1/4) and mono-acylglycerides (GPR119), some other lipid sensing moieties 13 14 241 15 have been proposed. K-cells express, compared to the surrounding epithelial cells, higher 16 17 242 levels of fatty-acid binding protein 5 (FABP5), an intracellular chaperone for lipophilic 18 19 243 compounds, and FABP5 has been implicated in responses to oral lard [66]. Although the 20 21 22 244 exact intracellular mechanism, dependent on solubilization of lipids by bile, remains unclear, 23 24 245 FABP5 knock-out animals are partly resistant to HFD induced obesity, if to a lesser extent 25 26 27 246 than GIP knock-out mice [66]. CD36 and FATP4, transporters for fatty acids, have been 28 29 247 implicated in CCK, secretin [67] and GLP-1 (REF) release and are thus likely to play some 30 31 32 248 role in GIP-secretion, possibly indirectly through effects on chylomicron formation. The 33 34 249 related scavenger receptor class-B member-1 (SCARB-1), implicated in intestinal cholesterol 35 36 250 absorption, is also highly expressed in K-cells, but we have not been able to demonstrate a 37 38 39 251 role of this receptor for K-cell lipid sensing (unpublished observations). 40 41 252 42 43 44 253 Protein: 45 46 254 47 48 49 255 Although protein and its digestion products are considered the less important macronutrient 50 51 256 stimulus for GIP secretion, an early study showed that a mix of amino acids stimulated GIP 52 53 257 secretion in man when perfused intraduodenally, but not when given intra-venously [68]. 54 55 56 258 Recently it has been reported that the GIP response to intraduodenally infused hydrolyzed 57 58 259 whey protein, also containing a mix of amino acids and small peptides, is more pronounced 59 60 61 62 63 11 64 65 260 in older compared with younger healthy men [69]. In mice, a range of different amino acids 1 2 261 has been found to stimulate GIP-secretion either in vivo in the leptin deficient ob/ob mouse, 3 4 5 262 which shows exaggerated GIP-responses (alanine, arginine, cysteine, histidine, lysine and 6 7 263 hydroxyproline) [70] or in mixed epithelial cultures (glutamine) [23], and intraduodenal 8 9 10 264 glutamine infusion (without the addition of other amino acids) also stimulates GIP secretion 11 12 265 in man [71]. The underlying cellular sensing mechanisms for amino acids in K-cells have not 13 14 266 15 been widely studied, but it seems likely that mechanisms identified in GLP-1 secreting L- 16 17 267 cells could also be present in K-cells. These include active transport of di/tripeptides by 18 19 268 PEPT1 (Slc15a1), an electrogenic H+-dependent transporter [72], and electrogenic Na+- 20 21 22 269 dependent uptake of amino-acids for example via BOAT1 (Slc6a19) and SNAT2 (Slc38a2) 23 24 270 [73], which could, in analogy to the glucose sensing mechanism via SGLT1, explain the 25 26 27 271 selective stimulation by luminal, but not intra-venously supplied amino-acids. Note, however, 28 29 272 that Slc6a19 knock-out mice have been reported to have exaggerated rather than reduced GIP 30 31 32 273 responses to refeeding [74], which seems unexpected if BOAT1 would act similarly to 33 34 274 SGLT1. Alternatively, amino acids can be sensed by G-protein coupled receptors expressed 35 36 275 on K-cells. These include the calcium-sensing receptor (CASR), shown to be important for 37 38 39 276 amino-acid sensing in L-cells [72, 75], and GPR142, for which a synthetic agonist elicited a 40 41 277 strong GIP response when given orally, which was absent in GPR142 knock-out mice [76]. 42 43 44 278 45 46 279 Non-nutrient regulators: 47 48 49 280 50 51 281 The K-cells also express a number of receptors for luminal constituents that may rise 52 53 282 postprandially, but are not macronutrients. These include the G-protein coupled receptors for 54 55 56 283 bile acids (GPBAR1/TGR5) and bile triggers a strong GIP secretory response in the perfused 57 58 284 intestine [77]. Similarly, short chain fatty acids, products of bacterial fermentation, can be 59 60 61 62 63 12 64 65 285 sensed by the K-cell expressed G-protein coupled receptors FFAR2 and FFAR3, but 1 2 286 interestingly, this has been linked to inhibition of GIP secretion, predominantly through 3 4 5 287 FFAR3 [78], and stimulation of GLP-1 secretion [78, 79], predominantly through FFAR2. 6 7 288 In parallel to the direct postprandial sensing of luminal constituents by K-cells, GIP secretion 8 9 10 289 is also under the control of other gut hormones. When purifying canine K-cells by elutriation, 11 12 290 Kieffer et al. noted that the enteroendocrine enriched cultures had a strong inhibitory 13 14 291 15 somatostatin (SST) tone and SST-immuno-neutralization resulted in elevated GIP-secretion 16 17 292 [80]. We confirmed SST to be a potent inhibitor of GIP secretion in mixed epithelial cultures, 18 19 293 which could be partly blocked by a SSTR5 antagonist, that by itself had, however, no effect 20 21 22 294 in the absence of exogenously added SST in these cultures [81]. In the perfused pig intestine 23 24 295 it was shown that GIP stimulates SST release [82], so it seems likely that there is a 25 26 27 296 complicated cross talk and signal integration at the level of the intact epithelium. Other 28 29 297 inhibitory signals converging on K-cells include endocannabinoids, which inhibit GIP, but 30 31 32 298 not GLP-1 secretion through CB1 receptors [81], and galanin [83], a common 33 34 299 neurotransmitter in the enteric nervous system, which inhibits both GIP and GLP-1 release. 35 36 300 Gastrin-releasing peptide was shown to stimulate elutriated canine K-cells [80], however, 37 38 39 301 mouse K-cells lack the bombesin 2 receptor and in mice secretion of GIP is not stimulated by 40 41 302 bombesin application, which, however, stimulates GLP-1 secretion in the same preparation 42 43 44 303 [17]. Although K-cells express muscarinic receptors such as M4, there is little evidence for 45 46 304 cholinergic modulation of GIP secretion [84]. Whilst the apparent over-secretion of GIP 47 48 49 305 observed in obesity and type 2 diabetes was at one stage thought to reflect a defective feed- 50 51 306 back inhibition by insulin [47], this might be of minor relevance and more related to the 52 53 307 difference in meal size in obese volunteers [85]. 54 55 56 308 57 58 309 Conclusion: 59 60 61 62 63 13 64 65 310 GIP-secreting K-cells, found at highest number in the duodenum, are well placed to respond 1 2 311 to luminal macronutrient rises after a meal. However, they do not just detect the presence of 3 4 5 312 these components in the intestine, but secretion of GIP is closely coupled to nutrient 6 7 313 absorption. This is either achieved by the use of the same molecular mechanism used to 8 9 10 314 extract the nutrients from the lumen for detection, as in the case of SGLT1, or by expressing 11 12 315 nutrient sensitive G-protein coupled receptors on the basolateral side, shielded from the 13 14 316 15 lumen, but presumably exposed to high nutrient concentrations being released from 16 17 317 enterocytes in their vicinity, as exemplified by FFAR1. The predominant location of K-cells 18 19 318 in the proximal intestine in contrast to the more distal location of the sister incretin GLP-1, 20 21 22 319 can result in different postprandial profiles, especially when digestion and/or nutrient 23 24 320 absorption is compromised (as discussed above for SGLT1 inhibition) or delivery to the more 25 26 27 321 distal intestine is accelerated, as after bariatric surgery [86]. These conditions have been 28 29 322 shown to have little impact or even reduce GIP-secretion, whilst they enhance GLP-1 30 31 32 323 secretion. 33 34 324 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 14 64 65 325 Figure legends: 1 2 326 3 4 5 327 Figure 1: Effects of FFAR knock-out on lipid stimulated GIP secretion 6 7 328 a) Plasma hormone changes in response to an oral lipid tolerance test in wild-type (WT) 8 9 10 329 and FFAR4 knock-out (FFAR4-/-) mice were examined as previously described [87]: 11 12 330 In brief, mice were fasted overnight (<16 hr). Intragastric gavage of a 1:1 mix of 13 14 331 15 olive:corn oils or phosphate buffered saline (control) was administered at 10ml/kg 16 17 332 body weight. 25 min later, mice were anaesthetized with isoflurane, and terminal 18 19 333 blood samples taken at 30 min by cardiac puncture. 20 21 22 334 b) Hormone secretion from primary mixed murine duodenal cultures were performed as 23 24 335 previously described [81]: In brief, supernatants were collected after a 2 h incubation 25 26 27 336 in either control buffer (containing in mmol/l: 138 NaCl, 4.5 KCl, 4.2 NaHCO3, 1.2 28 29 337 NaH2PO4, 2.6 CaCl2, 1.2 MgCl2, 10 HEPES, pH 7.4 with NaOH and supplemented 30 31 32 338 with 0.1 % bovine serum albumin) or control buffer with the addition of post-prandial 33 34 339 micelles (PPM; containing: oleic acid (200 μM), 2-monooleoyl-glycerol (70 μmol/l), 35 36 340 L-α-lysophosphatidylcholine (70 μM), cholesterol (17 μM) and taurocholic acid (700 37 38 39 341 μM)), or control buffer with the addition of forskolin and 3-isobutyl-1-methylxanthine 40 41 342 (Fsk/IBMX; 10 and 100 μM, respectively). Remaining cells were also lysed and both 42 43 44 343 supernatant and lysate GIP content was measured, enabling calculation of the % 45 46 344 secretion of the total well content. (Left) Responses in cultures established from 47 48 49 345 FFAR4 knock-out and matched wild-type animals. (Right) Responses in cultures 50 51 346 established from FFAR1 knock-out and matched wild-type animals. 52 53 347 Plasma and culture GIP (supernatant and cell lysate) were assayed by ELISA (GIP Total 54 55 56 348 ELISA Kit; Millipore, USA). FFAR1 [88] and FFAR4 [89] knock-out and matching 57 58 349 wild-type tissues were kind gifts from AstraZeneca. Statistical analysis was performed in 59 60 61 62 63 15 64 65 350 Graphpad Prism version 8 with a one-way ANOVA and Sidak’s multiple comparison 1 2 351 compensation. **** p<0.0001, *** p<0.001 compared to control in the same genotype; 3 4 5 352 #### p<0.0001, ## p<0.01 compared to the same stimulus in the other genotype, as 6 7 353 indicated. 8 9 354 10 11 355 Figure 2: Schematic of K-cell stimulation 12 13 14 356 Macronutrients, broken down by digestion in the intestinal lumen into monomeric 15 16 357 components, either stimulate K-cells directly from the luminal side through electrogenic 17 18 19 358 uptake into K-cells, via sodium-coupled glucose (SGLT1) or amino-acid (BOAT1) transport 20 21 359 or proton coupled di/tri-peptide (PEPT1) transport, triggering changes in the membrane 22 23 2+ 24 360 potential () and voltage gated Ca entry. The same transporters also play key roles in 25 26 361 absorption of these nutrients into the interstitial space, where they can stimulate G-protein 27 28 29 362 coupled receptors, such as the calcium sensing receptor (CASR) or GPR142, both sensitive to 30 31 363 amino acids. Triglycerides are absorbed as long chain free fatty acids (LCFA) and mono- 32 33 34 364 acyl-glycerides (MAG) and released after re-esterification in chylomicrons. Free-fatty acids 35 36 365 released at the basolateral side of enterocytes, possibly in chylomicrons, stimulate their 37 38 366 respective G-protein coupled receptors (FFAR1 and 4) on the basolateral side of K-cells. See 39 40 41 367 text for further explanation. 42 43 368 44 45 369 Acknowledgements: 46 47 48 370 Research in the Reimann/Gribble laboratory is currently funded by the Wellcome Trust 49 50 371 (106262/Z/14/Z, 106263/Z/14/Z) and the Medical Research Council 51 52 372 (MRC_MC_UU_12012/3). CEM was funded by a BBSRC-iCase PhD studentship in 53 54 55 373 collaboration with Astra Zeneca. We would like to thank Carol Lenaghan and David M. 56 57 374 Smith for provision of FFAR1 and FFAR4 knock-out- and corresponding wild-type tissue. 58 59 60 375 61 62 63 16 64 65 376 References 1 377 2 378 [1] J.C. Brown, J.R. Dryburgh, A gastric inhibitory polypeptide. II. The complete amino acid sequence, Can J 3 379 Biochem 49(8) (1971) 867-72. 4 380 [2] J. Dupre, S.A. Ross, D. Watson, J.C. Brown, Stimulation of insulin secretion by gastric inhibitory 5 381 polypeptide in man, J Clin Endocrinol Metab 37(5) (1973) 826-8. 6 382 [3] D.L. Sarson, Gastric inhibitory polypeptide (GIP), J Clin Pathol Suppl (Assoc Clin Pathol) 8 (1978) 31-7. 7 383 [4] H. Jörnvall, M. Carlquist, S. Kwauk, S.C. Otte, C.H. McIntosh, J.C. Brown, V. Mutt, Amino acid sequence 8 384 and heterogeneity of gastric inhibitory polypeptide (GIP), FEBS Lett 123(2) (1981) 205-10. 9 385 [5] J.M. Polak, S.R. Bloom, M. Kuzio, J.C. Brown, A.G. Pearse, Cellular localization of gastric inhibitory 10 386 polypeptide in the duodenum and jejunum., Gut 14(4) (1973) 284-8. 11 387 [6] R. Buffa, J.M. Polak, A.G. Pearse, E. Solcia, L. Grimelius, C. Capella, Identification of the intestinal cell 12 388 storing gastric inhibitory peptide., Histochemistry 43(3) (1975) 249-55. 13 389 [7] A.M. Buchan, J.M. Polak, C. Capella, E. Solcia, A.G. Pearse, Electronimmunocytochemical evidence for the 14 390 K cell localization of gastric inhibitory polypeptide (GIP) in man., Histochemistry 56(1) (1978) 37-44. 15 391 [8] M.J. Theodorakis, O. Carlson, S. Michopoulos, M.E. Doyle, M. Juhaszova, K. Petraki, J.M. Egan, Human 16 392 duodenal enteroendocrine cells: source of both incretin peptides, GLP-1 and GIP., Am J Physiol Endocrinol 17 393 Metab 290(3) (2006) E550-9. 18 394 [9] K. Mortensen, L.L. Christensen, J.J. Holst, C. Orskov, GLP-1 and GIP are colocalized in a subset of 19 395 endocrine cells in the small intestine., Regul Pept 114(2-3) (2003) 189-96. 20 396 [10] A.M. Habib, P. Richards, L.S. Cairns, G.J. Rogers, C.A. Bannon, H.E. Parker, T.C. Morley, G.S. Yeo, F. 21 397 Reimann, F.M. Gribble, Overlap of endocrine hormone expression in the mouse intestine revealed by 22 398 transcriptional profiling and flow cytometry, Endocrinology 153(7) (2012) 3054-65. 23 399 [11] K.L. Egerod, M.S. Engelstoft, K.V. Grunddal, M.K. Nøhr, A. Secher, I. Sakata, J. Pedersen, J.A. Windeløv, 24 400 E.M. Füchtbauer, J. Olsen, F. Sundler, J.P. Christensen, N. Wierup, J.V. Olsen, J.J. Holst, J.M. Zigman, S.S. 25 401 Poulsen, T.W. Schwartz, A major lineage of enteroendocrine cells coexpress CCK, secretin, GIP, GLP-1, PYY, 26 402 and neurotensin but not somatostatin, Endocrinology 153(12) (2012) 5782-95. 27 28 403 [12] J. Beumer, B. Artegiani, Y. Post, F. Reimann, F. Gribble, T.N. Nguyen, H. Zeng, M. Van den Born, J.H. 29 404 Van Es, H. Clevers, Enteroendocrine cells switch hormone expression along the crypt-to-villus BMP signalling 30 405 gradient, Nat Cell Biol 20(8) (2018) 909-916. 31 406 [13] H. Gehart, J.H. van Es, K. Hamer, J. Beumer, K. Kretzschmar, J.F. Dekkers, A. Rios, H. Clevers, 32 407 Identification of Enteroendocrine Regulators by Real-Time Single-Cell Differentiation Mapping, Cell 176(5) 33 408 (2019) 1158-1173.e16. 34 409 [14] L. Billing, P. Larraufie, J. Lewis, A. Leiter, J. Li, B. Lam, G. Yeo, D. Goldspink, R. Kay, F. Gribble, F. 35 410 Reimann, Single cell transcriptomic profiling of large intestinal enteroendocrine cells in mice – identification of 36 411 selective stimuli for Insulin-like peptide-5 and Glucagon-like peptide-1 co-expressing cells, Molecular 37 412 Metabolism (under revision), 2019. 38 413 [15] K. Sjölund, G. Sandén, R. Håkanson, F. Sundler, Endocrine cells in human intestine: an 39 414 immunocytochemical study, Gastroenterology 85(5) (1983) 1120-30. 40 415 [16] G.P. Roberts, P. Larraufie, P. Richards, R.G. Kay, S.G. Galvin, E.L. Miedzybrodzka, A. Leiter, H.J. Li, 41 416 L.L. Glass, M.K.L. Ma, B. Lam, G.S.H. Yeo, R. Scharfmann, D. Chiarugi, R.H. Hardwick, F. Reimann, F.M. 42 417 Gribble, Comparison of Human and Murine Enteroendocrine Cells by Transcriptomic and Peptidomic Profiling, 43 418 Diabetes 68(5) (2019) 1062-1072. 44 419 [17] B. Svendsen, R. Pais, M.S. Engelstoft, N.B. Milev, P. Richards, C.B. Christiansen, K.L. Egerod, S.M. 45 420 Jensen, A.M. Habib, F.M. Gribble, T.W. Schwartz, F. Reimann, J.J. Holst, GLP1- and GIP-producing cells 46 421 rarely overlap and differ by bombesin receptor-2 expression and responsiveness, J Endocrinol 228(1) (2016) 39- 47 422 48. 48 423 [18] L.L. Glass, F.J. Calero-Nieto, W. Jawaid, P. Larraufie, R.G. Kay, B. Göttgens, F. Reimann, F.M. Gribble, 49 424 Single-cell RNA-sequencing reveals a distinct population of proglucagon-expressing cells specific to the mouse 50 425 upper small intestine, Mol Metab 6(10) (2017) 1296-1303. 51 426 [19] A.N. Pilichiewicz, R. Chaikomin, I.M. Brennan, J.M. Wishart, C.K. Rayner, K.L. Jones, A.J. Smout, M. 52 427 Horowitz, C. Feinle-Bisset, Load-dependent effects of duodenal glucose on glycemia, gastrointestinal 53 428 hormones, antropyloroduodenal motility, and energy intake in healthy men, Am J Physiol Endocrinol Metab 54 429 293(3) (2007) E743-53. 55 430 [20] J. Schirra, M. Katschinski, C. Weidmann, T. Schäfer, U. Wank, R. Arnold, B. Göke, Gastric emptying and 56 431 release of incretin hormones after glucose ingestion in humans, J Clin Invest 97(1) (1996) 92-103. 57 432 [21] S. Cataland, S.E. Crockett, J.C. Brown, E.L. Mazzaferri, Gastric inhibitory polypeptide (GIP) stimulation 58 433 by oral glucose in man, J Clin Endocrinol Metab 39(2) (1974) 223-8. 59 60 61 62 63 17 64 65 434 [22] S. Sykes, L.M. Morgan, J. English, V. Marks, Evidence for preferential stimulation of gastric inhibitory 1 435 polypeptide secretion in the rat by actively transported carbohydrates and their analogues, J Endocrinol 85(2) 2 436 (1980) 201-7. 3 437 [23] H.E. Parker, A.M. Habib, G.J. Rogers, F.M. Gribble, F. Reimann, Nutrient-dependent secretion of glucose- 4 438 dependent insulinotropic polypeptide from primary murine K cells, Diabetologia 52(2) (2009) 289-98. 5 439 [24] F.M. Gribble, L. Williams, A.K. Simpson, F. Reimann, A novel glucose-sensing mechanism contributing to 6 440 glucagon-like peptide-1 secretion from the GLUTag cell line, Diabetes 52(5) (2003) 1147-1154. 7 441 [25] F. Reimann, A.M. Habib, G. Tolhurst, H.E. Parker, G.J. Rogers, F.M. Gribble, Glucose sensing in L cells: a 8 442 primary cell study, Cell Metab 8(6) (2008) 532-9. 9 443 [26] V. Gorboulev, A. Schürmann, V. Vallon, H. Kipp, A. Jaschke, D. Klessen, A. Friedrich, S. Scherneck, T. 10 444 Rieg, R. Cunard, M. Veyhl-Wichmann, A. Srinivasan, D. Balen, D. Breljak, R. Rexhepaj, H.E. Parker, F.M. 11 445 Gribble, F. Reimann, F. Lang, S. Wiese, I. Sabolic, M. Sendtner, H. Koepsell, Na+-d-glucose Cotransporter 12 446 SGLT1 is Pivotal for Intestinal Glucose Absorption and Glucose-Dependent Incretin Secretion., Diabetes 13 447 (2011). 14 448 [27] D.R. Powell, M. Smith, J. Greer, A. Harris, S. Zhao, C. DaCosta, F. Mseeh, M.K. Shadoan, A. Sands, B. 15 449 Zambrowicz, Z.M. Ding, LX4211 increases serum glucagon-like peptide 1 and peptide YY levels by reducing 16 450 sodium/glucose cotransporter 1 (SGLT1)-mediated absorption of intestinal glucose, J Pharmacol Exp Ther 17 451 345(2) (2013) 250-9. 18 452 [28] Y.L. He, W. Haynes, C.D. Meyers, A. Amer, Y. Zhang, P. Mahling, A.E. Mendonza, S. Ma, W. Chutkow, 19 453 E. Bachman, The effects of licogliflozin, a dual SGLT1/2 inhibitor, on body weight in obese patients with or 20 454 without diabetes, Diabetes Obes Metab 21(6) (2019) 1311-1321. 21 455 [29] A.T. Cheung, B. Dayanandan, J.T. Lewis, G.S. Korbutt, R.V. Rajotte, M. Bryer-Ash, M.O. Boylan, M.M. 22 456 Wolfe, T.J. Kieffer, Glucose-dependent insulin release from genetically engineered K cells, Science 290(5498) 23 457 (2000) 1959-62. 24 458 [30] C.L. Zhang, M. Katoh, T. Shibasaki, K. Minami, Y. Sunaga, H. Takahashi, N. Yokoi, M. Iwasaki, T. Miki, 25 459 S. Seino, The cAMP sensor Epac2 is a direct target of antidiabetic sulfonylurea drugs, Science 325(5940) (2009) 26 460 607-10. 27 461 [31] H. Ogata, Y. Seino, N. Harada, A. Iida, K. Suzuki, T. Izumoto, K. Ishikawa, E. Uenishi, N. Ozaki, Y. 28 462 Hayashi, T. Miki, N. Inagaki, S. Tsunekawa, Y. Hamada, S. Seino, Y. Oiso, KATP channel as well as SGLT1 29 463 participates in GIP secretion in the diabetic state, J Endocrinol 222(2) (2014) 191-200. 30 464 [32] J.W. Stephens, T.B. Bodvarsdottir, K. Wareham, S.L. Prior, R.M. Bracken, G.D. Lowe, A. Rumley, G. 31 465 Dunseath, S. Luzio, C.F. Deacon, J.J. Holst, S.C. Bain, Effects of short-term therapy with glibenclamide and 32 466 repaglinide on incretin hormones and oxidative damage associated with postprandial hyperglycaemia in people 33 467 with type 2 diabetes mellitus, Diabetes Res Clin Pract 94(2) (2011) 199-206. 34 468 [33] R.E. Kuhre, N.J. Wewer Albrechtsen, C. Deacon, E. Balk-Møller, J.F. Rehfeld, F. Reimann, F. Gribble, J.J. 35 469 Holst, Peptide production and secretion in GLUTag, NCI-H716 and STC-1 cells: a comparison to native L-cells, 36 470 J Mol Endocrinol (2016). 37 471 [34] L. Li, B.M. Wice, Bombesin and nutrients independently and additively regulate hormone release from 38 472 GIP/Ins cells, Am J Physiol Endocrinol Metab 288(1) (2005) E208-15. 39 473 [35] E.B. Ramshur, T.R. Rull, B.M. Wice, Novel insulin/GIP co-producing cell lines provide unexpected 40 474 insights into Gut K-cell function in vivo, J Cell Physiol 192(3) (2002) 339-50. 41 475 [36] S.Y. Wang, M.M. Chi, L. Li, K.H. Moley, B.M. Wice, Studies with GIP/Ins cells indicate secretion by gut 42 476 K cells is KATP channel independent, Am J Physiol Endocrinol Metab 284(5) (2003) E988-1000. 43 477 [37] R.F. Margolskee, J. Dyer, Z. Kokrashvili, K.S. Salmon, E. Ilegems, K. Daly, E.L. Maillet, Y. Ninomiya, B. 44 478 Mosinger, S.P. Shirazi-Beechey, T1R3 and gustducin in gut sense sugars to regulate expression of Na+-glucose 45 479 cotransporter 1, Proc Natl Acad Sci U S A 104(38) (2007) 15075-80. 46 480 [38] M.Y. Saltiel, R.E. Kuhre, C.B. Christiansen, R. Eliasen, K.W. Conde-Frieboes, M.M. Rosenkilde, J.J. 47 481 Holst, Sweet Taste Receptor Activation in the Gut Is of Limited Importance for Glucose-Stimulated GLP-1 and 48 482 GIP Secretion, Nutrients 9(4) (2017). 49 483 [39] Y. Fujita, R.D. Wideman, M. Speck, A. Asadi, D.S. King, T.D. Webber, M. Haneda, T.J. Kieffer, Incretin 50 484 release from gut is acutely enhanced by sugar but not by sweeteners in vivo, Am J Physiol Endocrinol Metab 51 485 296(3) (2009) E473-9. 52 486 [40] T. Wu, B.R. Zhao, M.J. Bound, H.L. Checklin, M. Bellon, T.J. Little, R.L. Young, K.L. Jones, M. 53 487 Horowitz, C.K. Rayner, Effects of different sweet preloads on incretin hormone secretion, gastric emptying, and 54 55 488 postprandial glycemia in healthy humans, Am J Clin Nutr 95(1) (2012) 78-83. 56 489 [41] D. Kreuch, D.J. Keating, T. Wu, M. Horowitz, C.K. Rayner, R.L. Young, Gut Mechanisms Linking 57 490 Intestinal Sweet Sensing to Glycemic Control, Front Endocrinol (Lausanne) 9 (2018) 741. 58 491 [42] R.E. Kuhre, F.M. Gribble, B. Hartmann, F. Reimann, J.A. Windeløv, J.F. Rehfeld, J.J. Holst, Fructose 59 492 stimulates GLP-1 but not GIP secretion in mice, rats, and humans, Am J Physiol Gastrointest Liver Physiol 60 493 306(7) (2014) G622-30. 61 62 63 18 64 65 494 [43] Y. Seino, H. Ogata, R. Maekawa, T. Izumoto, A. Iida, N. Harada, T. Miki, S. Seino, N. Inagaki, S. 1 495 Tsunekawa, Y. Oiso, Y. Hamada, Fructose induces glucose-dependent insulinotropic polypeptide, glucagon-like 2 496 peptide-1 and insulin secretion: Role of adenosine triphosphate-sensitive K(+) channels, J Diabetes Investig 6(5) 3 497 (2015) 522-6. 4 498 [44] P.R. Flatt, P. Kwasowski, C.J. Bailey, Stimulation of gastric inhibitory polypeptide release in ob/ob mice 5 499 by oral administration of sugars and their analogues, J Nutr 119(9) (1989) 1300-3. 6 500 [45] J.M. Polak, A.G. Pearse, L. Grimelius, V. Marks, Gastrointestinal apudosis in obese hyperglycaemic mice, 7 501 Virchows Arch B Cell Pathol 19(2) (1975) 135-50. 8 502 [46] C.J. Bailey, P.R. Flatt, P. Kwasowski, C.J. Powell, V. Marks, Immunoreactive gastric inhibitory 9 503 polypeptide and K cell hyperplasia in obese hyperglycaemic (ob/ob) mice fed high fat and high carbohydrate 10 504 cafeteria diets, Acta Endocrinol (Copenh) 112(2) (1986) 224-9. 11 505 [47] W. Creutzfeldt, R. Ebert, B. Willms, H. Frerichs, J.C. Brown, Gastric inhibitory polypeptide (GIP) and 12 506 insulin in obesity: increased response to stimulation and defective feedback control of serum levels, 13 507 Diabetologia 14(1) (1978) 15-24. 14 508 [48] N.Q. Nguyen, T.L. Debreceni, J.E. Burgess, M. Bellon, J. Wishart, S. Standfield, C.H. Malbert, M. 15 509 Horowitz, Impact of gastric emptying and small intestinal transit on blood glucose, intestinal hormones, glucose 16 510 absorption in the morbidly obese, Int J Obes (Lond) 42(9) (2018) 1556-1564. 17 511 [49] T. Wu, C.K. Rayner, L.E. Watson, K.L. Jones, M. Horowitz, T.J. Little, Comparative effects of 18 512 intraduodenal fat and glucose on the gut-incretin axis in healthy males, Peptides 95 (2017) 124-127. 19 513 [50] J.M. Falko, S.E. Crockett, S. Cataland, E.L. Mazzaferri, Gastric inhibitory polypeptide (GIP) stimulated by 20 514 fat ingestion in man, J Clin Endocrinol Metab 41(2) (1975) 260-5. 21 515 [51] M.R. Lucey, P.D. Fairclough, J.A. Wass, P. Kwasowski, S. Medbak, J. Webb, L.H. Rees, Response of 22 516 circulating somatostatin, insulin, gastrin and GIP, to intraduodenal infusion of nutrients in normal man, Clin 23 517 Endocrinol (Oxf) 21(3) (1984) 209-17. 24 518 [52] P. Kwasowski, P.R. Flatt, C.J. Bailey, V. Marks, Effects of fatty acid chain length and saturation on gastric 25 519 inhibitory polypeptide release in obese hyperglycaemic (ob/ob) mice, Biosci Rep 5(8) (1985) 701-5. 26 520 [53] F.M. Gribble, E. Diakogiannaki, F. Reimann, Gut Hormone Regulation and Secretion via FFA1 and FFA4, 27 521 Handb Exp Pharmacol 236 (2017) 181-203. 28 522 [54] S. Edfalk, P. Steneberg, H. Edlund, Gpr40 is expressed in enteroendocrine cells and mediates free fatty acid 29 523 stimulation of incretin secretion, Diabetes 57(9) (2008) 2280-2287. 30 524 [55] J.H. Ekberg, M. Hauge, L.V. Kristensen, A.N. Madsen, M.S. Engelstoft, A.S. Husted, R. Sichlau, K.L. 31 525 Egerod, P. Timshel, T.J. Kowalski, F.M. Gribble, F. Reiman, H.S. Hansen, A.D. Howard, B. Holst, T.W. 32 526 Schwartz, GPR119, a Major Enteroendocrine Sensor of Dietary Triglyceride Metabolites Coacting in Synergy 33 527 With FFA1 (GPR40), Endocrinology 157(12) (2016) 4561-4569. 34 528 [56] K. Iwasaki, N. Harada, K. Sasaki, S. Yamane, K. Iida, K. Suzuki, A. Hamasaki, D. Nasteska, K. Shibue, E. 35 529 Joo, T. Harada, T. Hashimoto, Y. Asakawa, A. Hirasawa, N. Inagaki, Free fatty acid receptor GPR120 is highly 36 530 expressed in enteroendocrine K cells of the upper small intestine and has a critical role in GIP secretion after fat 37 531 ingestion, Endocrinology 156(3) (2015) 837-46. 38 532 [57] A. Sankoda, N. Harada, K. Iwasaki, S. Yamane, Y. Murata, K. Shibue, Y. Thewjitcharoen, K. Suzuki, T. 39 533 Harada, Y. Kanemaru, S. Shimazu-Kuwahara, A. Hirasawa, N. Inagaki, Long-Chain Free Fatty Acid Receptor 40 534 GPR120 Mediates Oil-Induced GIP Secretion Through CCK in Male Mice, Endocrinology 158(5) (2017) 1172- 41 535 1180. 42 536 [58] A. Sankoda, N. Harada, T. Kato, E. Ikeguchi, K. Iwasaki, S. Yamane, Y. Murata, A. Hirasawa, N. Inagaki, 43 537 Free fatty acid receptors, G protein-coupled receptor 120 and G protein-coupled receptor 40, are essential for 44 538 oil-induced gastric inhibitory polypeptide secretion, J Diabetes Investig (2019). 45 539 [59] D.A. Goldspink, V.B. Lu, L.J. Billing, P. Larraufie, G. Tolhurst, F.M. Gribble, F. Reimann, Mechanistic 46 540 insights into the detection of free fatty and bile acids by ileal glucagon-like peptide-1 secreting cells, Mol Metab 47 541 7 (2018) 90-101. 48 542 [60] B.P. Shah, P. Liu, T. Yu, D.R. Hansen, T.A. Gilbertson, TRPM5 is critical for linoleic acid-induced CCK 49 543 secretion from the enteroendocrine cell line, STC-1, Am J Physiol Cell Physiol 302(1) (2012) C210-9. 50 544 [61] L.W. Christensen, R.E. Kuhre, C. Janus, B. Svendsen, J.J. Holst, Vascular, but not luminal, activation of 51 545 FFAR1 (GPR40) stimulates GLP-1 secretion from isolated perfused rat small intestine, Physiol Rep 3(9) (2015). 52 546 [62] W.J. Lu, Q. Yang, L. Yang, D. Lee, D. D'Alessio, P. Tso, Chylomicron formation and secretion is required 53 547 for lipid-stimulated release of incretins GLP-1 and GIP, Lipids 47(6) (2012) 571-80. 54 55 548 [63] A. Psichas, P.F. Larraufie, D.A. Goldspink, F.M. Gribble, F. Reimann, Chylomicrons stimulate incretin 56 549 secretion in mouse and human cells, Diabetologia 60(12) (2017) 2475-2485. 57 550 [64] M. Okawa, K. Fujii, K. Ohbuchi, M. Okumoto, K. Aragane, H. Sato, Y. Tamai, T. Seo, Y. Itoh, R. 58 551 Yoshimoto, Role of MGAT2 and DGAT1 in the release of gut peptides after triglyceride ingestion, Biochem 59 552 Biophys Res Commun 390(3) (2009) 377-81. 60 61 62 63 19 64 65 553 [65] B.S. Maciejewski, T.B. Manion, C.M. Steppan, Pharmacological inhibition of diacylglycerol 1 554 acyltransferase-1 and insights into postprandial gut peptide secretion, World J Gastrointest Pathophysiol 8(4) 2 555 (2017) 161-175. 3 556 [66] K. Shibue, S. Yamane, N. Harada, A. Hamasaki, K. Suzuki, E. Joo, K. Iwasaki, D. Nasteska, T. Harada, Y. 4 557 Hayashi, Y. Adachi, Y. Owada, R. Takayanagi, N. Inagaki, Fatty acid-binding protein 5 regulates diet-induced 5 558 obesity via GIP secretion from enteroendocrine K cells in response to fat ingestion, Am J Physiol Endocrinol 6 559 Metab 308(7) (2015) E583-91. 7 560 [67] S. Sundaresan, R. Shahid, T.E. Riehl, R. Chandra, F. Nassir, W.F. Stenson, R.A. Liddle, N.A. Abumrad, 8 561 CD36-dependent signaling mediates fatty acid-induced gut release of secretin and cholecystokinin, FASEB J 9 562 27(3) (2013) 1191-202. 10 563 [68] F.B. Thomas, E.L. Mazzaferri, S.E. Crockett, H.S. Mekhjian, H.D. Gruemer, S. Cataland, Stimulation of 11 564 secretion of gastric inhibitory polypeptide and insulin by intraduodenal amino acid perfusion, Gastroenterology 12 565 70(4) (1976) 523-7. 13 566 [69] C. Giezenaar, N.D. Luscombe-Marsh, A.T. Hutchison, S. Standfield, C. Feinle-Bisset, M. Horowitz, I. 14 567 Chapman, S. Soenen, Dose-Dependent Effects of Randomized Intraduodenal Whey-Protein Loads on Glucose, 15 568 Gut Hormone, and Amino Acid Concentrations in Healthy Older and Younger Men, Nutrients 10(1) (2018). 16 569 [70] P.R. Flatt, P. Kwasowski, R.J. Howland, C.J. Bailey, Gastric inhibitory polypeptide and insulin responses 17 570 to orally administered amino acids in genetically obese hyperglycemic (ob/ob) mice, J Nutr 121(7) (1991) 1123- 18 571 8. 19 572 [71] J. Chang, T. Wu, J.R. Greenfield, D. Samocha-Bonet, M. Horowitz, C.K. Rayner, Effects of intraduodenal 20 573 glutamine on incretin hormone and insulin release, the glycemic response to an intraduodenal glucose infusion, 21 574 and antropyloroduodenal motility in health and type 2 diabetes, Diabetes Care 36(8) (2013) 2262-5. 22 575 [72] E. Diakogiannaki, R. Pais, G. Tolhurst, H.E. Parker, J. Horscroft, B. Rauscher, T. Zietek, H. Daniel, F.M. 23 576 Gribble, F. Reimann, Oligopeptides stimulate glucagon-like peptide-1 secretion in mice through proton-coupled 24 577 uptake and the calcium-sensing receptor, Diabetologia 56(12) (2013) 2688-96. 25 578 [73] F. Reimann, L. Williams, G.D. Xavier, G.A. Rutter, F.M. Gribble, Glutamine potently stimulates glucagon- 26 579 like peptide-1 secretion from GLUTag cells, Diabetologia 47(9) (2004) 1592-1601. 27 580 [74] Y. Jiang, A.J. Rose, T.P. Sijmonsma, A. Bröer, A. Pfenninger, S. Herzig, D. Schmoll, S. Bröer, Mice 28 581 lacking neutral amino acid transporter B(0)AT1 (Slc6a19) have elevated levels of FGF21 and GLP-1 and 29 582 improved glycaemic control, Mol Metab 4(5) (2015) 406-17. 30 583 [75] R. Pais, F.M. Gribble, F. Reimann, Signalling pathways involved in the detection of peptones by murine 31 584 small intestinal enteroendocrine L-cells, Peptides 77 (2016) 9-15. 32 585 [76] O. Rudenko, J. Shang, A. Munk, J.P. Ekberg, N. Petersen, M.S. Engelstoft, K.L. Egerod, S.A. Hjorth, M. 33 586 Wu, Y. Feng, Y.P. Zhou, J. Mokrosinski, P. Thams, F. Reimann, F. Gribble, J.F. Rehfeld, J.J. Holst, J.T. 34 587 Treebak, A.D. Howard, T.W. Schwartz, The aromatic amino acid sensor GPR142 controls metabolism through 35 588 balanced regulation of pancreatic and gut hormones, Mol Metab 19 (2019) 49-64. 36 589 [77] R.E. Kuhre, N.J. Wewer Albrechtsen, O. Larsen, S.L. Jepsen, E. Balk-Møller, D.B. Andersen, C.F. Deacon, 37 590 K. Schoonjans, F. Reimann, F.M. Gribble, R. Albrechtsen, B. Hartmann, M.M. Rosenkilde, J.J. Holst, Bile acids 38 591 are important direct and indirect regulators of the secretion of appetite- and metabolism-regulating hormones 39 592 from the gut and pancreas, Mol Metab 11 (2018) 84-95. 40 593 [78] E.Y. Lee, X. Zhang, J. Miyamoto, I. Kimura, T. Taknaka, K. Furusawa, T. Jomori, K. Fujimoto, S. 41 594 Uematsu, T. Miki, Gut carbohydrate inhibits GIP secretion via a microbiota/SCFA/FFAR3 pathway, J 42 595 Endocrinol 239(3) (2018) 267-276. 43 596 [79] G. Tolhurst, H. Heffron, Y.S. Lam, H.E. Parker, A.M. Habib, E. Diakogiannaki, J. Cameron, J. Grosse, F. 44 597 Reimann, F.M. Gribble, Short-Chain Fatty Acids Stimulate Glucagon-Like Peptide-1 Secretion via the G- 45 598 Protein-Coupled Receptor FFAR2., Diabetes (2011). 46 599 [80] T.J. Kieffer, A.M. Buchan, H. Barker, J.C. Brown, R.A. Pederson, Release of gastric inhibitory polypeptide 47 600 from cultured canine endocrine cells, Am J Physiol 267(4 Pt 1) (1994) E489-96. 48 601 [81] C.E. Moss, W.J. Marsh, H.E. Parker, E. Ogunnowo-Bada, C.H. Riches, A.M. Habib, M.L. Evans, F.M. 49 602 Gribble, F. Reimann, Somatostatin receptor 5 and cannabinoid receptor 1 activation inhibit secretion of glucose- 50 603 dependent insulinotropic polypeptide from intestinal K cells in rodents, Diabetologia (2012). 51 604 [82] J.J. Holst, S.L. Jensen, S. Knuhtsen, O.V. Nielsen, J.F. Rehfeld, Effect of vagus, gastric inhibitory 52 605 polypeptide, and HCl on gastrin and somatostatin release from perfused pig antrum, Am J Physiol 244(5) (1983) 53 606 G515-22. 54 55 607 [83] A. Psichas, L.L. Glass, F. Reimann•, F.M. Gribble, Galanin inhibits GLP-1 and GIP secretion via the 56 608 GAL1 receptor in enteroendocrine L and K cells, 2015. 57 609 [84] C.F. Deacon, What do we know about the secretion and degradation of incretin hormones?, Regul Pept 58 610 128(2) (2005) 117-24. 59 611 [85] R. Ebert, W. Creutzfeldt, Gastric inhibitory polypeptide (GIP) hypersecretion in obesity depends on meal 60 612 size and is not related to hyperinsulinemia, Acta Diabetol Lat 26(1) (1989) 1-15. 61 62 63 20 64 65 613 [86] P. Larraufie, G.P. Roberts, A.K. McGavigan, R.G. Kay, J. Li, A. Leiter, A. Melvin, E.K. Biggs, P. Ravn, K. 1 614 Davy, D.C. Hornigold, G.S.H. Yeo, R.H. Hardwick, F. Reimann, F.M. Gribble, Important Role of the GLP-1 2 615 Axis for Glucose Homeostasis after Bariatric Surgery, Cell Rep 26(6) (2019) 1399-1408.e6. 3 616 [87] C.E. Moss, L.L. Glass, E. Diakogiannaki, R. Pais, C. Lenaghan, D.M. Smith, M. Wedin, M. Bohlooly-Y, 4 617 F.M. Gribble, F. Reimann, Lipid derivatives activate GPR119 and trigger GLP-1 secretion in primary murine L- 5 618 cells, Peptides (2015). 6 619 [88] R. Brownlie, R.M. Mayers, J.A. Pierce, A.E. Marley, D.M. Smith, The long-chain fatty acid receptor, 7 620 GPR40, and glucolipotoxicity: investigations using GPR40-knockout mice, Biochem Soc Trans 36(Pt 5) (2008) 8 621 950-4. 9 622 [89] V.M. Stone, S. Dhayal, K.J. Brocklehurst, C. Lenaghan, M. Sörhede Winzell, M. Hammar, X. Xu, D.M. 10 623 Smith, N.G. Morgan, GPR120 (FFAR4) is preferentially expressed in pancreatic delta cells and regulates 11 624 somatostatin secretion from murine islets of Langerhans, Diabetologia 57(6) (2014) 1182-91. 12 625 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 21 64 65 Figure

Figure 1

in vivo b) in vitro a) ## #### WT

WT WT

**** 30 -/- 30 #### FFAR1-/- FFAR4-/- **** FFAR4 1000

****

content)

- /ml) - 20 **** 20 pg *** **** **** **** 500

10 10

Plasma GIPPlasma (

Secreted (% GIP Secreted Secreted (% GIP Secreted

0 0 0 control oil control PPM Fsk/IBMX control PPM Fsk/IBMX Figure 2

FABP5 MGAT2 DGAT1 Chylomicron MAG CD36 + FATP4 LCFA

FFAR1 LCFA FFAR4

SGLT1 DY Glucose L-cell Ca2+ GIP Dipeptides PEPT1 Amino acids BOAT1 CASR GPR142

Amino acids Amino acids BOAT1 Dipeptides Dipeptides PEPT1 Glucose Glucose SGLT1