Primary deficiency of microsomal triglyceride transfer protein in human abetalipoproteinemia is associated with loss of CD1 function

Abstract

Abetalipoproteinemia (ABL) is a rare Mendelian disorder of lipid metabolism due to genetic deficiency in microsomal triglyceride transfer protein (MTP). It is associated with defects in MTP-mediated lipid transfer onto apolipoprotein B (APOB) and impaired secretion of APOB-containing lipoproteins. Recently, MTP was shown to regulate the CD1 family of lipid antigen-presenting molecules, but little is known about immune function in ABL patients. Here, we have shown that ABL is characterized by immune defects affecting presentation of self and microbial lipid antigens by group 1 (CD1a, CD1b, CD1c) and group 2 (CD1d) CD1 molecules. In dendritic cells isolated from ABL patients, MTP deficiency was associated with increased proteasomal degradation of group 1 CD1 molecules. Although CD1d escaped degradation, it was unable to load antigens and exhibited functional defects similar to those affecting the group 1 CD1 molecules. The reduction in CD1 function resulted in impaired activation of CD1-restricted T and invariant natural killer T (iNKT) cells and reduced numbers and phenotypic alterations of iNKT cells consistent with central and peripheral CD1 defects in vivo. These data highlight MTP as a unique regulator of human metabolic and immune pathways and reveal that ABL is not only a disorder of lipid metabolism but also an immune disease involving CD1.

Figure 1. Impaired presentation of self-derived antigens in ABL but not FHBL.

Monocyte-derived DCs from healthy controls and ABL patients (A and B) or FHBL patients (C) were cultured at the indicated concentrations with 5 × 10⁴ CD1a-, CD1b-, and CD1c-restricted T cells or CD1d-restricted iNKT cells. Cytokine release of T cells was determined after 18–20 hours of coculture by ELISA. Culture of DCs or T cells alone did not lead to detectable cytokine production. CD1 restriction is demonstrated by blocking of the T cell response by anti-CD1 antibodies (B). One representative experiment is shown. Two additional experiments with cells from independent ABL, FHBL and control subjects gave comparable results.

Figure 2. Impaired presentation of exogenous antigens in ABL.

DCs (A and D–H), monocytes (B) or B cells (C) (2 × 10⁴) were cocultured with 1 × 10⁵ CD1d-restricted NKT cells (A–D and H) or CD1b- (E) or CD1c-restricted T cells (F) in the presence of αGalCer (A–C, 100 ng/ml in C), GMM (E), mannosyl-phosphomycoketide (MPM; F), BbGL-IIf (10 μg/ml, D), or GSL-1 (10 μg/ml, D). In G, DCs were cocultured with CD1c-restricted T cells at the indicated days after infection with M. tuberculosis H37Rv. In H, DCs were cocultured with CD1d-restricted iNKT cells in the presence of the indicated heat-killed bacteria (MOI of 20). In some cases, blocking antibodies against CD1d or IL-12 were added. Cytokine release of T cells was determined after 18–20 hours of coculture by ELISA. Culture of DCs or T cells alone did not lead to detectable cytokine production. One representative experiment is shown. Two additional experiments with cells from independent ABL and control subjects showed comparable results. bac, bacteria.

Figure 3. Unaltered MHC class I– and class II–restricted antigen presentation in ABL patients.

(A) PBMCs (2 ×10⁵) were cultured in the presence or absence of 2 μg/ml CEF peptide mixture on IFN-γ ELISPOT plates. Plates were developed after 20 hours of culture, and the stimulation index (spots in the presence of CEF/spots in the absence of CEF) was calculated. (B) DCs (2 × 10⁴) were cocultured with 1 × 10⁵ tetanus toxin–reactive HLA-DR–restricted SPF3 T cells in the presence or absence of 25 μg/ml tetanus toxin. Cytokine release of T cells was determined after 18–20 hours of coculture by ELISA. Culture of DCs or T cells alone did not lead to detectable cytokine production. One representative experiment is shown. Two additional experiments with cells from independent ABL and control subjects showed comparable results.

Figure 4. Proteasomal degradation leads to impaired group 1 CD1 expression in ABL.

(A) Flow cytometric analysis of expression of the indicated molecules on DCs. For analysis of total (surface and intracellular) expression, cells were permeabilized before antibody staining. For surface staining, cells were stained and then fixed. Shown is the mean fluorescence intensity on a log₁₀ scale. (B) Flow cytometric analysis of CD1 and MHC class I expression after culture in the presence or absence of the indicated proteasomal inhibitors (10 μM) for 10 hours. Cells were permeabilized before antibody staining to allow for detection of total protein expression. Shown is the mean fluorescence intensity on a log₁₀ scale. One representative experiment is shown. One (B) and two (A) additional experiments with cells from independent ABL and control subjects showed comparable results.

Figure 5. Restoration of CD1 expression and function in ABL upon MTP reconstitution.

(A) Flow cytometric analysis of CD1 expression on ABL or control (CTR) DCs 3 days after infection with lentiviruses expressing MTP or the empty control virus. Shown is the mean fluorescence intensity on a log₁₀ scale. (B) Analysis of CD1-restricted presentation of self lipids 3 days after infection with lentiviruses expressing MTP or the empty control virus. DCs from healthy controls or ABL patients were cocultured with CD1-restricted T cells or iNKT cells. One representative experiment is shown. One additional experiment with cells from independent ABL and control subjects showed comparable results.

Figure 6. Reduced iNKT levels and an altered iNKT phenotype in ABL.

(A and B) iNKT levels in peripheral blood were determined by staining with αGalCer/CD1d tetramers or empty CD1d tetramers as control and are expressed as percentage of PBMCs. (C) Flow cytometric analysis of expression of cell surface markers on iNKT cells (upper panel) or conventional (tetramer-negative) T cells (lower panel) in peripheral blood. Shown is the mean fluorescence intensity. Dots represent individual subjects. Statistical significance was calculated using the 2-tailed Mann-Whitney U test. Analysis of iNKT surface markers was limited to patients with detectable iNKT cells. Bars indicate the median.