Identification and characterization of a novel DGAT1 missense mutation associated with congenital diarrhea

Abstract

Acyl-CoA:diacylglycerol acyltransferase (DGAT)1 and DGAT2 catalyze triglyceride (TG) biosynthesis in humans. Biallelic loss-of-function mutations in human DGAT1 result in severe congenital diarrhea and protein-losing enteropathy. Additionally, pharmacologic inhibition of DGAT1 led to dose-related diarrhea in human clinical trials. Here we identify a previously unknown DGAT1 mutation in identical twins of South Asian descent. These male patients developed watery diarrhea shortly after birth, with protein-losing enteropathy and failure to thrive. Exome sequencing revealed a homozygous recessive mutation in DGAT1, c.314T>C, p.L105P. We show here that the p.L105P DGAT1 enzyme produced from the mutant allele is less abundant, resulting in partial loss of TG synthesis activity and decreased formation of lipid droplets in patient-derived primary dermal fibroblasts. Thus, in contrast with complete loss-of-function alleles of DGAT1, the p.L105P missense allele partially reduces TG synthesis activity and causes a less severe clinical phenotype. Our findings add to the growing recognition of DGAT1 deficiency as a cause of congenital diarrhea with protein-losing enteropathy and indicate that DGAT1 mutations result in a spectrum of diseases.

Keywords: acyl CoA:diacylglycerol acyltransferase; diet and dietary lipids; genetics; intestine; lipid droplets; protein-losing enteropathy; triglycerides.

Fig. 1.

The protein-losing enteropathy associated with DGAT1 deficiency correlates with levels of dietary fat. Stool α-1 antitrypsin levels, a possible marker of disease activity, correlates with varying amounts of dietary fat over time for twin B.

Fig. 2.

DGAT1 p.L105P protein is found at reduced levels in cells. A: DGAT1 transcript levels are similar in control and patient fibroblasts. Relative mRNA levels determined by quantitative real-time PCR are shown here. B: Immunoblot shows less endogenous DGAT1 protein in mutant fibroblast lysate than in controls. C: FLAG-tagged wild-type and p.L105P DGAT proteins were expressed in SUM159 cells. The cells were treated with cycloheximide and lysed at different times. Immunoblot shows wild-type p.L105P DGAT1 protein may be degraded more rapidly compared with wild-type. D: Quantification of data shown in (C) and a second independent experiment. Statistical significance was determined by linear regression; *P < 0.05.

Fig. 3.

Cells with DGAT1 p.L105P enzyme have reduced, but not absent, DGAT activity. A: Lysates from DGAT1 p.L105P patient fibroblasts exhibit decreased DGAT in vitro activity. Incorporation of [¹⁴C]oleoyl CoA into TG in control and patient fibroblasts was measured. Lipids were separated by TLC, and the TG band for different reaction times is shown. B: Quantification of data shown in (A). Statistical significance was determined by Wilcoxon rank sum test; **P < 0.005; ***P < 0.001. C: Residual DGAT activity in patient fibroblast lysates can be further decreased by treatment with a DGAT1 inhibitor (D1i). Statistical significance was determined by Wilcoxon rank sum test; **P < 0.005; ***P < 0.001. D: TG synthesis is decreased in intact fibroblast cells of patients. The amount of [¹⁴C]oleate incorporated into TG in counts per minute per milligram protein is shown here. Statistical significance was determined by Wilcoxon rank sum test; ***P < 0.0001; *P < 0.05. D2i, DGAT2 inhibitor.

Fig. 4.

A: Mutant fibroblasts have fewer lipid droplets (LD) under DMSO and DGAT2 inhibitor (D2i) conditions after oleate treatment. Cells were stained with BODIPY 493/503 and imaged for neutral lipids. Representative images are shown. Scale bar: 30 μm. B: Lipid droplet numbers were quantified using CellProfiler. Quantification represents data from six to eight images per cell line (three controls and two patients) per condition. Statistical significance was determined by Wilcoxon rank sum test; ***P < 0.0001. D1i, DGAT1 inhibitor, D1iD2i, DGAT1 and DGAT2 inhibitors.