Plasma membrane delivery, endocytosis and turnover of transcobalamin receptor in polarized human intestinal epithelial cells

Abstract

Cells that are metabolically active and in a high degree of differentiation and proliferation require cobalamin (Cbl: vitamin B(12)) and they obtain it from the circulation bound to transcobalamin (TC) via the transcobalamin receptor (TC-R). This study has investigated the plasma membrane dynamics of TC-R expression in polarized human intestinal epithelial Caco-2 cells using techniques of pulse-chase labelling, domain-specific biotinylation and cell fractionation. Endogenously synthesized TC-R turned over with a half-life (T(1/2)) of 8 h following its delivery to the basolateral plasma membrane (BLM). The T(1/2) of BLM delivery was 15 min and TC-R delivered to the BLM was endocytosed and subsequently degraded by leupeptin-sensitive proteases. However, about 15% of TC-R endocytosed from the BLM was transcytosed (T(1/2), 45 min) to the apical membranes (BBM) where it underwent endocytosis and was degraded. TC-R delivery to both BLM and BBM was inhibited by Brefeldin A and tunicamycin, but not by wortmannin or leupeptin. Colchicine inhibited TC-R delivery to BBM, but not BLM. At steady state, apical TC-R was associated with megalin and both these proteins were enriched in an intracellular compartment which also contained Rab5 and transferrin receptor. These results indicate that following rapid delivery to both plasma membrane domains of Caco-2 cells, TC-R undergoes constitutive endocytosis and degradation by leupeptin-sensitive proteases. TC-R expressed in apical BBM complexes with megalin during its transcytosis from the BLM.

Figure 1. TC-R and megalin expression in the plasma membrane domains of Caco-2 cells

BBM and BLM proteins (100 μg) were separated on non-reducing SDS-PAGE (5%) and transferred proteins were detected with antiserum to TC-R (A) and megalin (B). The bands were visualized by the ECL technique. Triton X-100 extracts of the BBM were first immunoprecipitated with antiserum to TC-R (C) or megalin (D) and the immunoprecipitate was subjected to SDS-PAGE (5%) and probed with antiserum to megalin (C) or TC-R (D). The data show a typical representative blot obtained from six experiments using BBM and BLM isolated from three separate membrane preparations. Other details are provided in Methods.

Figure 2. Detection by immunoblotting of an endosomal-rich fraction

Fractions 8 and 9 collected from the metrazimide gradients in 4 aeparate experiments were pooled and precipitated protein in these fractions (50 μg protein) was separated on SDS-PAGE (7.5%) and immunoblotted using primary antibody for Tfr (A), TC-R (B), Rab5 (C), or megalin (D). The bands were visualized by the ECL technique. The blot shown is a typical representative from 4 separate blotting experiments.

Figure 3. Kinetics of TC-R delivery to BLM

[³⁵S]-TC-R present in BLM (A) and intracellular pool (B) following a pulse–chase experiment are shown as a percentage of [³⁵S]-TC-R (60 000 c.p.m.) present in the total cellular extracts prepared from cells harvested from one filter. Non-reducing SDS-PAGE of the immunoprecipitated radioactivity in the two fractions are shown. C, biotinylated [³⁵S]-TC-R transcytosed to BBM as a percentage of that present in the BLM at 0 time (60 000 c.p.m. per filter) is shown. Non-reducing SDS-PAGE of immunoprecipitated apical biotinylated [³⁵S]-radioactivity was carried out by pooling cells from 4 filters at each time interval. The bands were visualized by fluorography. The data shown are mean ±s.d. from 4 separate pulse–chase and domain labelling experiments. Other details are provided in the Methods.

Figure 4. Effect of processing and trafficking inhibitors on cell surface expression of TC-R

Caco-2 cells grown on culture inserts were treated with the indicated reagents: (1) none, (2) colchicine, (3) tunicamycin, (4) wortmannin, (5) brefeldin A, and (6) leupeptin. Cells were labelled with [³⁵S]-methionine (25 μCi ml⁻¹) and labelled TC-R was extracted with TC-R antiserum added to either the apical or the basolateral side at 4°C. Top panels, immunoprecipitated [³⁵S]-TC-R. Bottom panels, non-reducing SDS-PAGE of BLM- and BBM-associated [³⁵S]-TC-R. Other details are provided in Methods. Data shown are mean ±s.d. from 4 separate experiments.

Figure 5. Pulse–chase labelling of TC-R in leupeptin-treated and -untreated Caco-2 cells

A, post-confluent Caco-2 cells were treated with (█) or without (•) leupeptin (1 mg ml⁻¹) for 1 h prior to pulse labelling with [³⁵S]-methionine (200 μCi per flask). The cells were then chased for 0–16 h with DMEM with or without leupeptin (1 mg ml⁻¹). B, the immunoprecipitated radioactivity (4 500–88 000 c.p.m. in the absence of leupeptin and 68 000–88 000 c.p.m. in the presence of leupeptin) was liberated and subjected to non-reducing SDS-PAGE and the bands were visualized by fluorography. Other details of labelling, cell extraction and immunoprecipitation are provided in the Methods. The data shown are mean ±s.d. of 4 pulse–chase experiments.

Figure 6. Kinetics of endocytosis of TC-R from the BLM and BBM

Filter grown cells pulsed for 60 min with [³⁵S]-methionine (25 μCi per filter) were subjected to domain-specific biotinylation at 4°C and then chased at 37°C for 0–120 min. At each time interval, biotinylated [³⁵S]-TC-R present in either the BBM or the BLM was determined by immunopreciptation using TC-R antiserum added either to BBM or BLM. The immunoprecipitated radioactivity was released and treated with streptavidin–agarose to separate biotinylated from non-biotinylated [³⁵S]-TC-R. Left panel, biotinylated [³⁵S]-TC-R in the BLM (•) and BBM (▪) are shown and are the mean ±s.d. of 5 separate labelling experiments (P≤ 0.01). Right panel, SDS-PAGE of biotinylated [³⁵S]-TC-R (1 000–60 000 c.p.m.) in the BLM (top) and 200–10 000 c.p.m. in the BBM (bottom) were used during the indicated chase time intervals.