Impaired Barrier Function and Autoantibody Generation in Malnutrition Enteropathy in Zambia

Abstract

Intestinal damage in malnutrition constitutes a threat to the survival of many thousands of children globally. We studied children in Lusaka, Zambia, with severe acute malnutrition (SAM) and persistent diarrhea using endoscopy, biopsy and analysis of markers and protective proteins in blood and intestinal secretions. We carried out parallel investigations in apparently healthy adults, and analyzed biomarkers only in apparently healthy children. Villus height and crypt depth did not differ in children with SAM and adult controls, but epithelial surface was reduced in children with SAM (median 445, interquartile range (IQR) 388, 562μm per 100μm muscularis mucosae) compared to adults (578, IQR 465,709; P=0.004). Histological lesions and disruptions of claudin-4 and E-cadherin were most pronounced in children with SAM. Circulating lipopolysaccharide, a marker of bacterial translocation, was higher in malnourished children (251, IQR 110,460EU/ml) than in healthy children (51, IQR 0,111; P=0.0001). Other translocation markers showed similar patterns. Anti-Deamidated Gliadin Peptide IgG concentrations, although within the normal range, were higher in children with SAM (median 2.7U/ml, IQR 1.5-8.6) than in adults (1.6, 1.4-2.1; P=0.005), and were inversely correlated with villus height (ρ=-0.79, n=13, P=0.001). Malnutrition enteropathy is associated with intestinal barrier failure and immune dysregulation.

Keywords: Autoantibodies; Enteropathy; Environmental enteric dysfunction; Glucagon-like peptide 2; HIV; Malnutrition; Microbial translocation; Tissue transglutaminase serology, deamidated gliadin peptide serology.

Fig. 1

Mucosal histology in severe acute malnutrition. A, histological section from a child with severe acute malnutrition (SAM) showing moderate villus blunting (villus height 247 μm). B shows a higher magnification of A at the point of the arrow, an early epithelial discontinuity measuring 62.5 μm. C and D, detachment of pale and irregular epithelial cells in biopsies from two adults with EE; at the site of these more extensive discontinuities or microerosions (black arrows) the basement membrane is exposed to the luminal stream. E, a biopsy from a child with SAM showing subtotal villus atrophy.

Fig. 2

Epithelial histology and disruption of claudin 4 and E-cadherin. Epithelial breaks imaged in pairs in nearby sections from three children with severe acute malnutrition, each in a separate column. Top row: haematoxylin/eosin (H&E) stains with scale bar representing 50 μm. Second row, magnified H&E images of boxes from the first row with scale bars representing 20 μm. Third row, images of same sections shown in first row, with E-cadherin immunostaining in green, claudin-4 immunostaining in red, and nuclei (DAPI) in blue; scale bars at 50 μm. Fourth row, further magnification of boxes from third row, using the same colour code, and scale bars at 20 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Molecules protective against microbial translocation. (A) GLP-2 in adults was negatively associated with circulating lipopolysaccharide (LPS; ρ = − 0.35; P = 0.01 in the whole group, but in HIV seronegatives the correlation was stronger (ρ = − 0.51; P = 0.001). (B) Presence of TFF3 immunoreactivity in duodenal aspirates of children was associated with greatly reduced circulating LPS (P = 0.03).

Fig. 4

Coeliac type antibodies. Anti-deamidated gliadin peptides (DGP) and anti-tissue transglutaminase (TTG) antibodies in children. (A) Correlation between anti-DGP antibody concentration in serum and villus height (ρ = − 0.79; P < 0.0001); two points with anti-DGP concentrations above 40 U/ml omitted for clarity. (B) Correlation between anti-TTG antibody concentration (Inova system) in serum and circulating lipopolysaccharide binding protein, LBP (ρ = 0.83; P < 0.0001).