Role of Gpcpd1 in intestinal alpha-glycerophosphocholine metabolism and trimethylamine N-oxide production

Abstract

Glycerophosphocholine (GPC) is an intracellular metabolite in phosphatidylcholine metabolism and has been studied for endogenous choline supply in cells. GPC, as a water-soluble supplement, has been expected to play a role in preventing brain disorders; however, recent studies have shown that intake of high levels of choline-containing compounds is related to trimethylamine N-oxide (TMAO) production in the liver, which is reportedly associated with the progression of atherosclerosis. In this study, we aimed to explore the mechanisms underlying the intestinal absorption and metabolism of GPC. Caco-2 cell monolayer experiments showed that exogenously added GPC was hydrolyzed to choline in the apical medium, and the resulting choline was transported into the Caco-2 cells and further to the basolateral medium. Subsequently, we focused on glycerophosphodiesterase 1 (Gpcpd1/GDE5), which hydrolyzes GPC to choline in vitro and is widely expressed in the gastrointestinal epithelium. Our results revealed that the Gpcpd1 protein was located not only in cells but also in the medium in which Caco-2 cells were cultured. Gpcpd1 siRNA decreased the GPC-hydrolyzing activity both inside Caco-2 cells and in conditioned medium, suggesting the involvement of Gpcpd1 in luminal GPC metabolism. Finally, we generated intestinal epithelial-specific Gpcpd1-deficient mice and found that Gpcpd1 deletion in intestinal epithelial cells affected GPC metabolism in intestinal tissues and partially abolished the increase in blood TMAO levels induced by GPC administration. These observations demonstrate that Gpcpd1 triggers choline production from GPC in the intestinal lumen and is a key endogenous enzyme that regulates TMAO levels following GPC supplementation.

Keywords: GDE5; Gpcpd1; choline; glycerophosphocholine; trimethylamine N-oxide.

Figure 1

GPC metabolism in mammalian cells. GPC is an intracellular water-soluble metabolite of PC metabolism. In mammalian cells, GPC is hydrolyzed to choline and subsequently oxidized to betaine. In addition, choline can be metabolized to trimethylamine (TMA) by the intestinal microflora and subsequently converted to trimethylamine N-oxide (TMAO) in the liver. GPC, glycerophosphocholine; PC, phosphatidylcholine.

Figure 2

Administration of glycerophosphocholine elevates circulating trimethylamine N-oxide level.A, circulating TMAO, (B) GPC, (C) choline, (D) betaine post oral GPC supplementation gavage (500 mg/kg body weight, n = 6). All values are presented as means ± SD; significance was determined using one-way ANOVA with Tukey’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 3

GPC metabolism within the intestinal tract after intraduodenal GPC injection. Intraduodenal injections of GPC (500 mg/kg body weight) changing choline metabolites in the (A) portal vein and in the (B) small intestine segment (n = 4–5). All values are presented as means ± SD; significance was determined using one-way ANOVA with Tukey’s multiple comparisons test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. GPC, glycerophosphocholine.

Figure 4

Extracellular GPC hydrolysis for choline absorption in a Caco-2 cell monolayer system.A, schematic illustration of transwell system and experiment design. Choline metabolite levels in the (B) transwell system or in the (C) Caco-2 cells after apical addition or basal addition of GPC (100 μM, 300 μM, and 1 mM). D, betaine level in the transwell system after apical addition or basal addition of GPC. E, GPC-hydrolyzing activity in conditioned medium obtained from Caco-2 cells after 20 h of culture (without Caco-2 cells), conditioned medium with Caco-2 cells, and control medium. All values are presented as means ± SD (A) and (E), one-way ANOVA with Tukey’s multiple comparisons test. B–D, two-way ANOVA with Tukey’s multiple comparison test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. GPC, glycerophosphocholine.

Figure 5

Gpcpd1 knockdown in Caco-2 cells affects its GPC-hydrolyzing activity.A-C, Caco-2 cells were transfected with a negative control siRNA (CON) or Gpcpd1 siRNA (siGpcpd1) (n = 4–5). A, Gpcpd1 mRNA level in Caco-2 cells determined using quantitative PCR. Choline metabolite changes in the (B) Caco-2 cells lysate (n = 6) and (C) medium (n = 6) were quantified using LC-MS. D, Western blot analysis of medium and cell lysate from Caco-2 cells either transiently transfected with an empty vector or Gpcpd1 expression vector. E and F, choline metabolite level after GPC addition (200 μM) to the medium from control Caco-2 cells or siGpcpd1-transfected Caco-2 cells quantified using LC-MS. All values are presented as means ± SD, (A), (B), and (C), unpaired two-tailed Student’s t test. E and F, two-way ANOVA with Tukey’s multiple comparison test. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. GPC, glycerophosphocholine; Gpcpd1, glycerophosphodiesterase 1.

Figure 6

Gpcpd1-KO mice show disrupted choline metabolism in the small intestine.A, epithelium from gastrointestinal segments (n = 6) was isolated and subjected to Gpcpd1 mRNA analysis using qPCR. B, schematic illustration of Gpcpd1^flox/flox, villin-cre (Gpcpd1-KO) mice. C, Gpcpd1 mRNA levels of Gpcpd1-KO mice (KO) or its WT littermate (CON). Choline metabolite levels in the (D) intestine segment and (E) liver from Gpcpd1-KO mice (KO, n = 5) or its WT littermate (CON, n = 6) quantified using LC-MS. F, RNA-seq analysis of small intestine tissue from Gpcpd1-KO mice (KO) and its WT littermate (CON). Heat maps of genes associated with glycerophospholipid catabolic process and alkaloid catabolic process are shown. All values are presented as means ± SD, (A), one-way ANOVA with Tukey’s multiple comparisons test. B, two-way ANOVA with Tukey’s multiple comparison test. D and E, unpaired two-tailed Student’s t test ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Gpcpd1, glycerophosphodiesterase 1.

Figure 7

Decreased circulating TMAO levels in Gpcpd1-KO mice.A, circulating TMAO, (B) GPC, (C) choline, and (D) betaine levels post oral GPC supplementation gavage (500 mg/kg body weight) in Gpcpd1-KO mice (KO, n = 6) or its WT littermate (CON, n = 4). E and F, changes in choline metabolite levels in response to intraduodenal injections of GPC (500 mg/kg body weight) with or without antibiotics treatment. G, intestinal Gpcpd1 mRNA expression level with or without antibiotics treatment. All values are means ± SD, (A), two-way ANOVA with Tukey’s multiple comparison test. B–F, mixed-effects analysis with Šídák's multiple comparison test. G, unpaired two-tailed Student’s t test ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. GPC, glycerophosphocholine; Gpcpd1, glycerophosphodiesterase 1; TMAO, trimethylamine N-oxide.