A novel mutation, Ile344Asn, in microsomal triglyceride transfer protein abolishes binding to protein disulfide isomerase

Abstract

Microsomal triglyceride transfer protein (MTP) plays crucial roles in the assembly and secretion of apolipoprotein B-containing lipoproteins and loss of function MTP variants are associated with abetalipoproteinemia, a disease characterized by the absence of these lipoproteins. MTP is a heterodimeric protein of two subunits, MTP and protein disulfide isomerase (PDI). In this study, we report a proband with abetalipoproteinemia who was monitored annually for 10 years in her third decade and had very low plasma lipids and undetectable apoB-containing lipoproteins. Genetic testing revealed biallelic variants in the MTTP gene. She has a well-documented nonsense mutation Gly865∗ that does not interact with the PDI subunit. She also has a novel missense MTP mutation, Ile344Asn. We show that this mutation abrogates lipid transfer activity in MTP and does not support apolipoprotein B secretion. This residue is present in the central α-helical domain of MTP and the substitution of Ile with Asn at this position disrupts interactions between MTP and PDI subunits. Ile344 is away from the known MTP:PDI interacting sites identified in the crystal structure of MTP suggesting that MTP:PDI interactions are more dynamic than previously envisioned. Identification of more missense mutations will enhance our understanding of the structure-function of MTP and the role of critical residues in these interactions between the two subunits. This knowledge may guide us in developing novel treatment modalities to reduce plasma lipids and atherosclerosis.

Keywords: abetalipoproteinemia; apoB; hypobetalipoproteinemia; lipoproteins; variants.

Fig. 1

Ile344Asn does not support lipid transfer and apoB100 secretion. A–C: Mko-3 cells were transfected with 3 μg of plasmid expressing either WT or Ile344Asn MTP. After 48 h, media was replaced with fresh DMEM with 10% FBS. After overnight incubation, cell homogenates (25 μg protein) were used to measure MTP and β-actin (control) protein levels (A) by Western blot analysis and to measure triglyceride (TG) transfer activity (B). Overnight-conditioned media was used to measure apoB-100 secreted by the cells by performing ELISA (C). The amounts of apoB were normalized to cell protein levels. D–F: For lipid transfer assay and MTP:PDI interaction studies, 400 μg protein from the cellular homogenate was incubated overnight at 4°C with anti-Flag M2 affinity gel and the purified MTP protein was eluted with 100 μM Flag peptide in buffer K. Purified protein (20 μl) was separated on SDS-PAGE and probed with anti-hMTP antibodies (top), stripped and reprobed with anti-PDI antibody (bottom) (D). Amounts of MTP protein bands were quantified using ImageJ and values were used to normalize lipid transfer activities. A different aliquot was used to measure TG and PL transfer activities (E–F). The bars and error bars represent mean ± SD. To calculate the significance one-way ANOVA nonparametric (multiple comparison) or two-way ANOVA was used, ∗∗∗ and ∗∗∗∗ represent P < 0.001 and P < 0.0001, respectively. The data are representative of three independent experiments performed with biological triplicates.

Fig. 2

Increased expression of Ile344Asn did not improve the lipid transfer activity and apoB secretion in Mko-3 cell. Mko-3 cells were transfected with either 2 μg of plasmid expressing WT MTP or 4 μg, 6 μg and 9 μg of Ile344Asn MTP. After 48 h, the media was replaced with fresh DMEM with 10% FBS and incubated overnight. The cells were collected, homogenized, and centrifuged at 12,000 rpm or 13,500 g for 10 min at 4°C. From the clear supernatant, 25 μg protein was separated on 8% SDS-PAGE and probed with anti-hMTP (A, top), stripped, and probed with anti-β-actin antibodies for control (A, bottom). For the TG transfer assay, 25 μg protein was used from the clear homogenate (B). Overnight-conditioned media was used to measure apoB-100 secretion by ELISA (C). For lipid transfer assays and MTP:PDI interaction studies, purified proteins were separated on SDS-PAGE and probed for anti-hMTP followed by anti-PDI antibody (D). Different aliquots of purified proteins were used to measure TG and PL transfer activity (E–F). The bars and error bars represent mean ± SD. To calculate the significance, one-way ANOVA nonparametric (multiple comparison) or two-way ANOVA was used. ∗∗∗ and ∗∗∗∗ represent P < 0.001 and P < 0.0001, respectively. The data are representative of three independent experiments performed with biological triplicates.

Fig. 3

Substitution of Ile with Asn at 344 may disrupt interactions between helices within the central α helical domain of MTP. A: Ile344 is identified in the MTP:PDI crystal structure. Ile344 is away from the predicted MTP:PDI interacting interface (colored in blue). B: Ile344 is present in the third stand of the central α helical domain. C: Ile344 interacts with Val363 on Helix 4 and Ser371 and Ala374 of helix 5. D: Ile was replaced with Asn using the rotamer command in Chimera X and its interactions with nearby amino acid were studied. In contrast to Ile344, Asn344 appears to interact with Leu304 in helix 1, Leu336 in helix 2, and Ala339 residue present in the connecting coil between helices 2 and 3.

Fig. 4

Possible explanations for the lack of PDI binding in Asp361Tyr and Arg540His missense mutations. A: Asp361 and Arg540 are identified in the MTP:PDI crystal structure. Asp361 is away from the predicted MTP:PDI interacting interface while Arg540 is present in helix 14 close to the MTP:PDI interacting site (colored in blue). B: Asp361 is present in the fourth strand of the central α helical domain and interacts with Arg392 on Helix 6. C: Tyr361 which instead appears to interact with Phe329 in helix 2. D: Arg540 in helix 14 is predicted to interact with the side chains of Gln530 and Leu567. E: When replaced with His, it was predicted to interact with Tyr528 in helix 13.