Regulation of the Axillary Osmidrosis-Associated ABCC11 Protein Stability by N-Linked Glycosylation: Effect of Glucose Condition

Abstract

ATP-binding cassette C11 (ABCC11) is a plasma membrane protein involved in the transport of a variety of lipophilic anions. ABCC11 wild-type is responsible for the high-secretion phenotypes in human apocrine glands, such as that of wet-type ear wax, and the risk of axillary osmidrosis. We have previously reported that mature ABCC11 is a glycoprotein containing two N-linked glycans at Asn838 and Asn844. However, little is known about the role of N-linked glycosylation in the regulation of ABCC11 protein. In the current study, we investigated the effects of N-linked glycosylation on the protein level and localization of ABCC11 using polarized Madin-Darby canine kidney II cells. When the N-linked glycosylation in ABCC11-expressing cells was chemically inhibited by tunicamycin treatment, the maturation of ABCC11 was suppressed and its protein level was significantly decreased. Immunoblotting analyses demonstrated that the protein level of the N-linked glycosylation-deficient mutant (N838Q and N844Q: Q838/844) was about half of the ABCC11 wild-type level. Further biochemical studies with the Q838/844 mutant showed that this glycosylation-deficient ABCC11 was degraded faster than wild-type probably due to the enhancement of the MG132-sensitive protein degradation pathway. Moreover, the incubation of ABCC11 wild-type-expressing cells in a low-glucose condition decreased mature, glycosylated ABCC11, compared with the high-glucose condition. On the other hand, the protein level of the Q838/844 mutant was not affected by glucose condition. These results suggest that N-linked glycosylation is important for the protein stability of ABCC11, and physiological alteration in glucose may affect the ABCC11 protein level and ABCC11-related phenotypes in humans, such as axillary osmidrosis.

Fig 1. Characterization of ABCC11 fused with EGFP, expressed in MDCKII cells.

(A) Schematic illustration of ABCC11 protein. ABCC11 has two N-linked glycosylation sites, N838 and N844, in an extracellular loop between transmembrane helices 7 and 8. (B) Glycosylation status and expression of ABCC11-EGFP and EGFP-ABCC11 in MDCKII cells. Cell lysates were prepared 72 h after the transfection, and subsequently subjected to immunoblotting after treatment with or without PNGase F. The immunoreactive bands, corresponding to the glycosylated forms (Gs) of EGFP-fused ABCC11 protein, were disappeared after PNGase F treatment. The apparent molecular weight value of the non-glycosylated form (Non-G) of EGFP-fused ABCC11 protein was about 180,000. α-Tubulin: a loading control. (C) Glycosylation status and expression of non-tagged ABCC11 and ABCC11-EGFP in HEK293 cells. Cell lysates were prepared 72 h after the transfection and treated with or without PNGase F, and subsequently subjected to immunoblotting using the anti-ABCC11 antibody. α-Tubulin: a loading control. (D) Apical localization of ABCC11-EGFP and EGFP-ABCC11 in MDCKII cells 72 h after the transfection. An endogenous apical membrane marker gp135 was immunostained using anti-gp135 antibody (red). Nuclei were stained with TO-PRO®-3 iodide (gray). All panels show the Z-sectioning images. Mock (EGFP): a control vector; GLUT-9b-EGFP: a positive control for apical localization; GLUT-9a-EGFP: a negative control for apical localization. (E) 5-fluorouracil (5-FU) resistance activity of MDCKII/mock cells and MDCKII/ABCC11 WT-EGFP cells. Data are expressed as mean ± S.E.M. n = 8. Where vertical bars are not shown, the S.E.M. was contained within the limits of the symbol. Statistical analyses for significant differences were performed according to Student’s t test (*, P < 0.01 compared with mock).

Fig 2. The effects of tunicamycin, an N-linked glycosylation inhibitor, on lowering the protein levels of ABCC11.

Forty-eight hours after the transfection, MDCKII cells transiently expressing ABCC11 wild-type (WT) were cultured with tunicamycin (TNM) (0, 0.8, and 4.0 μg/mL) for further 24 h, then cell lysate samples were prepared after the treatment without (A) or with (B) PNGase F, and then subjected to immunoblotting. α-Tubulin: a loading control. P: PNGase F-treated ABCC11 WT as a control for the band position of non-glycosylated ABCC11 (details are described in Materials and Methods). The signal intensity ratio (ABCC11/α-tubulin) of the immunoreactive bands was determined and normalized to that in TNM-untreated cells (0 μg/mL in B). Data are expressed as mean ± S.D. n = 5. Statistical analyses for significant differences were performed according to Student’s t test (**, P < 0.01 vs control).

Fig 3. The effect of the disruption of N-linked glycosylation sites on the protein levels of ABCC11.

(A) Glycosylation status and expression of ABCC11 wild-type (WT) and its mutants expressed in MDCKII cells 72 h after the transfection. The immunoreactive bands, corresponding to the glycosylated forms (G, glycosylation at N838 or N844; GG, at both N838 and N844) of the ABCC11 protein, disappeared with PNGase F treatment. α-Tubulin: a loading control. (B) mRNA levels of ABCC11 WT and its mutants. The mRNA levels were normalized to the WT (control) level. (C and D) Densitometoric analysis of protein levels of ABCC11 WT and glycosylation-deficient mutants. After PNGase F treatment, cell lysate samples were subjected to immunoblotting. The signal intensity ratio (ABCC11/α-tubulin) of the immunoreactive bands (C) was determined and normalized to the WT level. Data are expressed as mean ± S.D. n = 4. Statistical analyses for significant differences were performed according to Dunnett’s test (*, P < 0.05; N.S.: not significantly different as compared with WT).

Fig 4. The effect of the disruption of N-linked glycosylation sites on the localization of ABCC11.

Confocal microscopic images were obtained 72 h after the transfection. All samples show the apical localization of ABCC11 in MDCKII cells. An endogenous apical membrane marker gp135 was immunostained using the anti-gp135 antibody (red). Nuclei were stained with TO-PRO®-3 iodide (gray). All panels show the Z-sectioning images.

Fig 5. The effect of N-linked glycosylation on the protein stability of ABCC11.

(A and B) The effect of the inhibition of protein translation with cycloheximide (CHX) on the protein levels of ABCC11 wild-type (WT) and the Q838/844 mutant. Seventy-two hours after the transfection, MDCKII cells transiently expressing ABCC11 WT or Q838/844 were further incubated with 50 μM CHX for 0, 3, 6, and 12 h. At the indicated time, cells were collected and stored at -80°C until use. Simultaneously, all cell lysates were prepared and treated without (A) or with (B) PNGase F, and then subjected to immunoblotting. α-Tubulin: a loading control. The signal intensity ratio (ABCC11/α-tubulin) of the immunoreactive bands corresponding to the non-glycosylated form (Non-G) of ABCC11 protein was determined and normalized to the control (t = 0 h) level. Gs, glycosylated forms. (C) Densitometoric analysis of protein levels of the ABCC11 WT and the Q838/844 mutant. Values are expressed as % of control (values at 0 h). Data are expressed as mean ± S.D. n = 3 (WT), 4 (Q838/844 mutant). Statistical analyses for significant differences were performed according to Student’s t test (*, P < 0.05 in indicated time points).

Fig 6. The effect of bafilomycin A₁ or MG132 treatment on the protein levels of ABCC11.

(A) Forty-eight hours after the transfection, MDCKII cells transiently expressing ABCC11 wild-type (WT) were cultured with 10 nM bafilomycin A₁ (BMA) or 2 μM MG132 (MG) for a further 24 h. Then, cell lysates were subjected to immunoblotting after treatment with or without PNGase F. Gs, glycosylated forms; Non-G, non-glycosylated form. (B and C) Densitometoric analysis of protein levels of ABCC11 WT (B) and Q838/844 mutant (C) in BMA or MG-treated cells. After PNGase F treatment, cell lysates were subjected to immunoblotting. α-Tubulin: a loading control. The signal intensity ratio (ABCC11/α-tubulin) of the immunoreactive bands was determined and normalized to the non-treated control (Non). Data are expressed as mean ± S.D. n = 4. Statistical analyses for significant differences were performed according to Student’s t test (*, P < 0.05; **, P < 0.01 compared with control). (D) Z-sectioning images of cycloheximide (CHX)-treated MDCKII cells transiently expressing ABCC11 WT or Q838/844 mutant in the presence of MG132. Seventy-two hours after the transfection, MDCKII cells were cultured with 50 μM CHX in the presence or absence of 2 μM MG132 for further 12 h. Nuclei were stained with TO-PRO®-3 iodide (gray).

Fig 7. The effect of extracellular glucose on the protein levels of ABCC11.

(A and B) The decrease of the glycosylation and the protein levels of ABCC11 wild-type (WT) expressed in MDCKII cells cultured with a low-glucose medium. (C) The little effect of low-glucose culture on the protein levels of the Q838/844 mutant expressed in MDCKII cells. Forty-eight hours after the transfection, mostly confluent MDCKII cells transiently expressing ABCC11 were cultured with high (H) or low (L) glucose medium for further 24 h. Then, cell lysates were prepared and subsequently subjected to immunoblotting after treatment without (A) or with (B, C) PNGase F. The immunoreactive bands disappeared with PNGase F treatment, corresponding to the glycosylated forms (Gs) of ABCC11 protein. α-Tubulin: a loading control. In densitometoric analyses of protein levels of ABCC11 WT and the Q838/844 mutant, the signal intensity ratio (ABCC11/α-tubulin) of the immunoreactive bands was determined and normalized to the control (high glucose) level. P: PNGase F-treated ABCC11 WT as a control for the band position of the non-glycosylated form of ABCC11 (details are described in Materials and Methods). Data are expressed as mean ± S.D. n = 4. Statistical analyses for significant differences were performed according to Student’s t test (**, P < 0.01). N.S.: not significantly different among groups.