Macromolecule Translocation across the Intestinal Mucosa of HIV-Infected Patients by Transcytosis and through Apoptotic Leaks

Abstract

Based on indirect evidence, increased mucosal translocation of gut-derived microbial macromolecules has been proposed as an important pathomechanism in HIV infection. Here, we quantified macromolecule translocation across intestinal mucosa from treatment-naive HIV-infected patients, HIV-infected patients treated by combination antiretroviral therapy, and HIV-negative controls and analyzed the translocation pathways involved. Macromolecule permeability was quantified by FITC-Dextran 4000 (FD4) and horseradish peroxidase (HRP) flux measurements. Translocation pathways were addressed using cold inhibition experiments. Tight junction proteins were characterized by immunoblotting. Epithelial apoptosis was quantified and translocation pathways were further characterized by flux studies in T84 cell monolayers using inducers and inhibitors of apoptosis and endocytosis. In duodenal mucosa of untreated but not treated HIV-infected patients, FD4 and HRP permeabilities were more than a 4-fold increase compared to the HIV-negative controls. Duodenal macromolecule permeability was partially temperature-dependent and associated with epithelial apoptosis without altered expression of the analyzed tight junction proteins. In T84 monolayers, apoptosis induction increased, and both apoptosis and endocytosis inhibitors reduced macromolecule permeability. Using quantitative analysis, we demonstrate the increased macromolecule permeability of the intestinal mucosa in untreated HIV-infected patients. Combining structural and mechanistic studies, we identified two pathways of increased macromolecule translocation in HIV infection: transcytosis and passage through apoptotic leaks.

Keywords: HIV; apoptosis; intestine; macromolecule passage; tight junction; transcytosis.

Figure 1

Macromolecule permeabilities in the duodenal and colon mucosa of HIV-infected patients and HIV-negative controls. Permeabilities were determined from FD4 and HRP flux measurements performed on duodenal (A,B) and colon (D,E) mucosa obtained from HIV-negative control individuals, from untreated (HIV naive), and from suppressively treated (HIV cART) HIV-infected patients. The transepithelial resistance (TER) of the mucosal samples was monitored in parallel to the flux measurements (C,F). m = 25–28, n = 20 (control), m = 23–32, n = 19 (HIV), m = 14–18, n = 10 (HIV cART), and m = 8–13, n = 10 (control), m = 10–11, n = 5 (HIV), m = 11–12, n = 9 (cART) for duodenal and colon mucosal samples, respectively. * p < 0.05, ** p < 0.01.

Figure 2

Expression of tight junction proteins involved in paracellular macromolecule passage. For comparison, protein expression was expressed in relation to that of controls (100%). (A) Expression of occludin (Occl), tricellulin (Tric), and lipolysis-stimulated lipoprotein receptor (LSR) in duodenal mucosa obtained from HIV-negative control individuals (n = 6), and from untreated (HIV naive; n = 10) and suppressively treated (HIV cART; n = 7) HIV-infected patients. (B) Expression of occludin, tricellulin, and LSR in colon mucosa obtained from HIV-negative control individuals (n = 4), and from untreated (HIV naive; n = 3) and suppressively treated HIV-infected patients (HIV cART; n = 3–4). Representative immunoblots of occludin, tricellulin, and LSR in membrane preparations from duodenal (C,D) colon mucosa. The band covered with background at the very right position of the blot was excluded from the densitometric analysis for tricellulin. * p < 0.05, ** p < 0.01.

Figure 3

Effect of apoptosis induction and cold on macromolecule permeability of T84 monolayers. Induction of apoptosis by camptothecin (1 µg/mL) (A) reduced the TER of T84 cells (n = 13) and (B) increased the activity of caspases (n = 4; for comparison, the caspase activity of camptothecin-treated monolayers was expressed in relation to that of untreated control monolayers). It also affected permeability of T84 monolayers at 37 °C and 14 °C to (C) FD4 and (D) HRP. n = 13 and 10 for FD4 and HRP fluxes at 37 °C, respectively, and n = 10 and 7 for FD4 and HRP fluxes at 14 °C, respectively. *** p < 0.001, camptothecin versus control. ## p < 0.01, 37 °C versus 14 °C.

Figure 4

Effect of inhibitors of apoptosis or endocytosis on camptothecin-induced macromolecule permeability of T84 monolayers. Effect of camptothecin (1 µg/mL) on permeability of T84 monolayers to FD4 and HRP in the absence or presence of the caspase inhibitor QVD-Oph (20 µM) (A,B), or the endocytosis inhibitor MitMAB (20 µM) (C,D). * p < 0.05, ** p < 0.01, *** p < 0.001 versus untreated controls. # p < 0.05, ### p < 0.001 without versus with inhibitor. n = 7–9.

Figure 5

Effect of cold on macromolecule permeability of the duodenal and colon mucosa. Comparison of (A) FD4 and (B) HRP permeabilities of the duodenal and the colon mucosa at 37 °C and 14 °C. Mucosal samples for these experiments were obtained from HIV-negative individuals. * p < 0.05, n = 7–8 for duodenal samples, n = 3 for colon samples.