The paracellular channel for water secretion in the upper segment of the Malpighian tubule of Rhodnius prolixus.

Lumen to bath J12/C1 and bath to lumen J21/C2 fluxes per unit concentration of 19 probes with diameters (dm) ranging from 3.0-30.0 A (water, urea, erythritol, mannitol, sucrose, raffinose and 13 dextrans with dm 9.1-30.0 A) were measured during volume secretion (Jv) in the upper segment of the Malpighian Tubule of Rhodnius by perfusing lumen and bath with 14C or 3H-labeled probes. Jnet = (J12/C1-J21/C2) was studied as a function of Jv.Jv was varied by using different concentrations of 5-hydroxy tryptamine. Jnet for 3H-water was not different from Jv. We found: (i) A strong correlation between Jnet and Jv for 8 probes dm = 3.0-11.8 A (group a probes), indicating that the convective component of Jnet is more important than its diffusive component and than unstirred layers effects which are negligible. Therefore group a probes are solvent dragged as they cross the epithelium. (ii) There is no correlation between Jnet and Jv for 11 probes with dm = 11.8-30 A (group b). Therefore these probes must cross the epithelium by diffusion and not by solvent drag. (iii) In a plot of Jnet/Jv vs. dm group a probes show a steep linear relation with a slope = -0.111, while for group b probes the slope is -0.002. Thus there is a break between groups a and b in this plot. We tried to fit the data with models for restricted diffusion and convention through cylindrical or parallel slit pathways. We conclude that (i) group a probes are dragged by water through an 11.0 A-wide slit. (ii) Most of Jv must follow an extracellular noncytosolic pathway. (iii) Group b probes must diffuse through a 42 A-wide slit. (iv) A cylindrical pathway does not fit the data.