Marked differences in tight junction composition and macromolecular permeability among different intestinal cell types.

BACKGROUND: Mammalian small intestinal tight junctions (TJ) link epithelial cells to one another and function as a permselective barrier, strictly modulating the passage of ions and macromolecules through the pore and leak pathways, respectively, thereby preventing the absorption of harmful compounds and microbes while allowing regulated transport of nutrients and electrolytes. Small intestinal epithelial permeability is ascribed primarily to the properties of TJs between adjoining enterocytes (ENTs), because there is almost no information on TJ composition and the paracellular permeability of nonenterocyte cell types that constitute a small but significant fraction of the intestinal epithelia. RESULTS: Here we directed murine intestinal crypts to form specialized organoids highly enriched in intestinal stem cells (ISCs), absorptive ENTs, secretory goblet cells, or Paneth cells. The morphological and morphometric characteristics of these cells in organoids were similar to those in vivo. The expression of certain TJ proteins varied with cell type: occludin and tricellulin levels were high in both ISCs and Paneth cells, while claudin-1, -2, and -7 expression was greatest in Paneth cells, ISCs, and ENTs, respectively. In contrast, the distribution of claudin-15, zonula occludens 1 (ZO-1), and E-cadherin was relatively homogeneous. E-cadherin and claudin-7 marked mainly the basolateral membrane, while claudin-2, ZO-1, and occludin resided in the apical membrane. Remarkably, organoids enriched in ENTs or goblet cells were over threefold more permeable to 4 and 10 kDa dextran compared to those containing stem and Paneth cells. The TJ-regulator larazotide prevented the approximately tenfold increases in dextran flux induced by the TJ-disrupter AT1002 into organoids of different cell types, indicating that this ZO toxin nonselectively increases permeability. Forced dedifferentiation of mature ENTs results in the reacquisition of ISC-like characteristics in TJ composition and dextran permeability, suggesting that the post-differentiation properties of TJs are not hardwired. CONCLUSIONS: Differentiation of adult intestinal stem cells into mature secretory and absorptive cell types causes marked, but potentially reversible, changes in TJ composition, resulting in enhanced macromolecular permeability of the TJ leak pathway between ENTs and between goblet cells. This work advances our understanding of how cell differentiation affects the paracellular pathway of epithelia.

Proteasome inhibition decreases cardiac remodeling after initiation of pressure overload.

We tested the possibility that proteasome inhibition may reverse preexisting cardiac hypertrophy and improve remodeling upon pressure overload. Mice were submitted to aortic banding and followed up for 3 wk. The proteasome inhibitor epoxomicin (0.5 mg/kg) or the vehicle was injected daily, starting 2 wk after banding. At the end of the third week, vehicle-treated banded animals showed significant (P<0.05) increase in proteasome activity (PA), left ventricle-to-tibial length ratio (LV/TL), myocyte cross-sectional area (MCA), and myocyte apoptosis compared with sham-operated animals and developed signs of heart failure, including increased lung weight-to-TL ratio and decreased ejection fraction. When compared with that group, banded mice treated with epoxomicin showed no increase in PA, a lower LV/TL and MCA, reduced apoptosis, stabilized ejection fraction, and no signs of heart failure. Because overload-mediated cardiac remodeling largely depends on the activation of the proteasome-regulated transcription factor NF-kappaB, we tested whether epoxomicin would prevent this activation. NF-kappaB activity increased significantly upon overload, which was suppressed by epoxomicin. The expression of NF-kappaB-dependent transcripts, encoding collagen types I and III and the matrix metalloprotease-2, increased (P<0.05) after banding, which was abolished by epoxomicin. The accumulation of collagen after overload, as measured by histology, was 75% lower (P<0.05) with epoxomicin compared with vehicle. Myocyte apoptosis increased by fourfold in hearts submitted to aortic banding compared with sham-operated hearts, which was reduced by half upon epoxomicin treatment. Therefore, we propose that proteasome inhibition after the onset of pressure overload rescues ventricular remodeling by stabilizing cardiac function, suppressing further progression of hypertrophy, repressing collagen accumulation, and reducing myocyte apoptosis.