A synthetic peptide corresponding to the extracellular domain of occludin perturbs the tight junction permeability barrier

Abstract

Occludin, the putative tight junction integral membrane protein, is an attractive candidate for a protein that forms the actual sealing element of the tight junction. To study the role of occludin in the formation of the tight junction seal, synthetic peptides (OCC1 and OCC2) corresponding to the two putative extracellular domains of occludin were assayed for their ability to alter tight junctions in Xenopus kidney epithelial cell line A6. Transepithelial electrical resistance and paracellular tracer flux measurements indicated that the second extracellular domain peptide (OCC2) reversibly disrupted the transepithelial permeability barrier at concentrations of < 5 microM. Despite the increased paracellular permeability, there were no changes in gross epithelial cell morphology as determined by scanning EM. The OCC2 peptide decreased the amount of occludin present at the tight junction, as assessed by indirect immunofluorescence, as well as decreased total cellular content of occludin, as assessed by Western blot analysis. Pulse-labeling and metabolic chase analysis suggested that this decrease in occludin level could be attributed to an increase in turnover of cellular occludin rather than a decrease in occludin synthesis. The effect on occludin was specific because other tight junction components, ZO-1, ZO-2, cingulin, and the adherens junction protein E-cadherin, were unaltered by OCC2 treatment. Therefore, the peptide corresponding to the second extracellular domain of occludin perturbs the tight junction permeability barrier in a very specific manner. The correlation between a decrease in occludin levels and the perturbation of the tight junction permeability barrier provides evidence for a role of occludin in the formation of the tight junction seal.

Figure 1

Expression and junctional localization of occludin correlated with the development of tight junctions in Xenopus A6 kidney epithelial cells. A6 cells were allowed to grow to confluency in Transwell filters in normal medium and were subsequently changed to low calcium medium for 18 h. Then the medium was replenished with normal calcium (t = 0), and the formation of tight junctions was monitored by measuring TER at t = 0 h (TER = Ωcm²), t = 15 h (TER ∼0 Ωcm²), t = 3 d (TER ∼100 Ωcm²), and t = 5 d (TER > 1,000 Ωcm²). (a) Indirect immunofluorescence of occludin in A6.2 cells at t = 0, 15 h, 3 d, and 5 d. (b) Western blots of A6 cell lysate for occludin, cingulin, and ZO-1 at t = 0, 15 h, 3 d, and 5 d.

Figure 2

A synthetic peptide (OCC2) corresponding to the entire second extracellular domain of chick occludin decreased TER of A6 cell monolayers. (a) Effect of various synthetic peptides on TER. OCC1 (corresponding to the entire first extracellular domain of chick occludin), OCC2 (corresponding to the entire second extracellular domain of chick occludin), OCC2(S) (corresponding to the scrambled sequence of the entire second extracellular domain of chick occludin), and DMSO solvent control were used. Newly confluent A6 cell monolayers (starting TER ∼1,000 Ωcm²) that were still developing TER were used. Cell monolayers were treated with a final concentration of 5 μM OCC1, 5 μM OCC2, 5 μM OCC2(S), or DMSO (0.05%) for 66 h, and peptides were replenished every 24 h. At the end of the 66-h peptide incubation, TER for control monolayers reached ∼5,000–6,000 Ωcm². n = 6 for each condition. (b) Time course of effect of OCC2 peptide on TER of A6 cells that were still developing TER. Cell monolayers that had attained TER ∼750 Ωcm² were treated with a final concentration of 5 μM OCC1 (n = 4) or 5 μM OCC2 (n = 5) at t = 0. Peptides were replenished at 30 h. (c) Dose dependency of OCC2 peptide on TER in A6 cell monolayers that were still developing TER. A6.2 cells were allowed to grow to confluency in normal medium and were subsequently changed to low calcium medium for 18 h. The low calcium medium was replaced with normal calcium medium containing a final concentration of 0.2, 0.5, 2, and 5 μM OCC2. TER were measured after 4 d when control cell monolayers developed TER of ∼3,000 Ωcm². n = 3 for all concentrates tested. (d) Time course of OCC2 peptide on TER of steady-state A6 cell monolayers that were confluent for ∼2 wk (TER ∼8,000 Ωcm²). Cells were treated with a final concentration of 5 μM OCC2 at t = 0. Untreated monolayers were done in parallel as control. Peptides were replenished at 22 and 76 h. n = 3 for all conditions. (e) Dose dependency of OCC2 peptide on TER in steady-state A6 monolayers (TER ∼6,000 Ωcm²). Cell monolayers were treated with a final concentration of 0.5, 1, 2, 5, and 10 μM OCC2. TER was measured at 40 h after peptide addition. The TER of each individual monolayer is plotted. Each concentration of OCC2 was done on duplicate monolayers. All error bars represent standard error.

Figure 3

OCC2 increased the paracellular flux of membraneimpermeant tracer molecules. (a) Effects of OCC2 on the flux of [³H]mannitol, [¹⁴C]inulin, Texas red–conjugated neutral dextran (mol wt 3,000), and Texas red–conjugated neutral dextran (mol wt 40,000). OCC1 was used as control peptide. A6 cell monolayers were allowed to grow until TER reached ∼1,200 Ωcm². Cell monolayers were then treated with a final concentration of 5 μl OCC1 or OCC2 for 36 h. TER was measured and tracers flux assays were performed as described in Materials and Methods. For all four tracers, n = 8 and error bars represent SEM. (b–e) The relationship between tracer flux and TER changes induced by OCC2 treatment. Absolute flux values for individual A6 cell monolayers were plotted against TER of the same monolayer. (b) [³H]mannitol, (c) [¹⁴C]inulin, (d) neutral dextran (mol wt 3,000) conjugated with Texas red, and (e) neutral dextran (mol wt 40,000) conjugated with Texas red.

Figure 4

OCC2 reduced junctional stainings of occludin but not ZO-1, cingulin, ZO-2, and E-cadherin. A6 cell monolayers from the paracellular tracer flux assays described in Fig. 3 were processed for indirect immunofluorescence microscopy at the end of the flux assays. OCC1-treated monolayers had TER of ∼2,500 Ωcm², and OCC2-treated monolayers had TER of ∼250 Ωcm². OCC1-treated (a, c, e, g, and i) and OCC2-treated (b, d, f, h, and j) monolayers were immunostained in parallel for occludin (a and b), ZO-1 (c and d), cingulin (e and f), ZO-2 (g and h), and E-cadherin (i and j).

Figure 5

OCC2 specifically decreased total cellular occludin levels. (a) Western blots of occludin, cingulin, ZO-1, ZO-2, and E-cadherin of total cell lysates from monolayers that were treated with OCC1, OCC2, or DMSO solvent control. A6 cells were allowed to grow until TER reached ∼1,000 Ωcm², and monolayers were treated with 10 μM of OCC1, 10 μM OCC2, or 0.1% DMSO for 24 h. (b) Only the peptide that decreased TER also caused a decrease in occludin levels. Western blot of occludin in A6 total cell lysates of monolayers that were treated with OCC1, OCC2(U) (unmodified), OCC2, or OCC2(S) (scrambled). A6 cells were allowed to grow to confluency in normal medium and were subsequently changed to low calcium medium for 18 h. A6 cells were then replenished with normal calcium media containing peptides at a final concentration of 5 μM. OCC2(U), unmodified OCC2, and OCC2(S), scrambled sequence of OCC2. Peptides were replenished every 24 h, and cells were extracted for analysis at 4 d after initial peptide treatment. (c) Occludin synthesis was not reduced by OCC2 treatment. A6 cells that were either untreated or treated for either 2 or 22 h with a final concentration of 5 μM OCC2 were subsequently labeled for 2.5 h with [³⁵S]methionine followed by immunoprecipitation (IP) of occludin. (d) Turnover of occludin was enhanced by OCC2 treatment. A6 cells were metabolically labeled 20 h with [³⁵S]methionine. At the end of the labeling period (t = 0), fresh media (without [³⁵S]methionine) containing 10 μM OCC2 was added for 12 h followed by immunoprecipitation (IP) of occludin. Untreated A6 cells were used in parallel as a control.

Figure 6

The effects of OCC2 on TER and occludin accumulation were reversible. (a) Reversibility of TER after OCC2 removal. A6 cell monolayers that had TER of ∼1,700 Ωcm² were treated at t = 0 with a final concentration of 5 μM OCC1 or OCC2. At t = 24 h, peptides were either replenished (OCC1 and OCC2) or removed (OCC2 Recovery) from the cells. OCC1 (n = 6), OCC2 (n = 6), and OCC2 recovery (n = 3). (b) Recovery of junctional stainings of occludin after OCC2 removal. A6 cell monolayers that had TER ∼1,000 Ωcm² were treated at t = 0 with a final concentration of 5 μM OCC1 or OCC2. At t = 24 h, peptides were either replenished (OCC1 and OCC2) or removed (OCC2 recovery) from the cells. At t = 60 h, cells were processed for indirect immunofluorescence microscopy of occludin. OCC1 (TER ∼2,200 Ωcm²), OCC2 (TER ∼250 Ωcm²), and OCC2 recovery (TER ∼2,300 Ωcm²).

Figure 7

OCC2 did not cause morphological changes in A6 cell monolayers as observed by scanning EM. Confluent A6 cells grown on polylysine-coated coverslips were treated 24 h with a final concentration of 10 μM OCC1 (a and d), 10 μM OCC2 (b and e), and 0.1% DMSO (c and f). Cells were then processed for scanning EM. A6 cells were ∼7–10 μM in diam.