Gated ion channels: Held open by glycans?

Glycosylation analysis and homology modeling of a receptor for the inhibitory neurotransmitter GABA reveals regulation by glycosylation.

The main inhibitory neurotransmitter system in the brain, the gamma-aminobutyric acid (GABA) type A receptor system, is a pharmacological target in the treatment of anxiety, epilepsy and other neurological conditions. However, greater understanding of the regulation and function of GABA_A receptors is needed. Investigating the role of GABAA subunit glycosylation, Robert Macdonald and colleagues now report in the Journal of Biological Chemistry that glycans regulate the stability, assembly and function of GABAA channels.

GABA_A receptors, part of the Cys-loop receptor superfamily of ligand-gated ion channels, are heteropentameric assemblies of alpha and beta subunits. Preventing the N-glycosylation of Cys-loop receptors impedes their cell-surface expression, but many details about GABA_A receptor glycans and their roles are lacking. Macdonald and colleagues used multiple sequence alignment, homology modeling and analysis of glycosylation consensus sequons to predict three glycosylation sites in each of the GABA_A receptor β1–3 subunits. Using transfected human embryonic kidney cells, they investigated the glycosylation efficiencies of these sites for the β2 subunit, and the importance of glycosylation for surface targeting and function of α1β2 receptors. The authors mutated each of the predicted glycosylation sites singly or in combination, and analyzed the molecular masses of the recombinant proteins. All three sites were glycosylated, but glycans at the third site, N173, had lower molecular masses than those attached at the other two. These glycans remained sensitive to N-acetylglucosaminidase H (endo H) digestion as they passed through the Golgi apparatus, and glycomic profiling revealed that they are high-mannose structures. Removing the third glycosylation site reduced total β2 levels, consistent with a structural role for the glycan in proper protein folding and stability.

By contrast, the first and second sites carried complex type glycans that were not required for stability of the individual β2 subunit. However, the second site, N104, was important for assembly of the receptor: when β2 subunits were co-expressed with α1 subunits — as binary α1β2 receptors — mutation of N104 decreased their surface expression. This appeared to be caused by a reduction in pentamer stability or assembly, and not by impaired forward trafficking from the ER. The first site, N32, had a lower efficiency of glycosylation, consistent with the presence of Ser, not Thr, in the sequon, but mutation of this site did not significantly affect subunit levels or receptor assembly.

To find out how β2 glycans regulate GABA_A receptor function, the authors measured peak current amplitudes after agonist stimulation. The α1β2(N173Q) receptor was less sensitive to low levels of GABA, and deactivated more rapidly than wildtype or α1β2(N32Q) receptors. Current amplitudes of α1β2(N104Q) were too small for detailed analysis. Cell-attached single channel recording showed that although single channel conductance was not affected by lack of any of the three glycosylation sites, the mean open times of all three mutant channels were reduced. Thus, mutation of the first site, which had no effect on stability, did impair channel function, suggesting that receptor function might be fine-tuned by glycosylation of the weaker sequon. A mutation near to N32 is associated with childhood absence epilepsy and reduction of peak current amplitudes of the receptor channels. Whether this mutation affects glycosylation remains to be determined. Nevertheless, this study provides important insights into the nature and function of GABA_A glycans.

Emma Leah
Functional Glycomics Gateway

Reference:
Lo, W.-Y. et al.
Glycosylation of beta2 subunits regulates GABA_A receptor biogenesis and channel gating.
J. Biol. Chem. July 16 (2010)
Full text

Integrin degradation: An ESCRTed mechanism controlling cell migration

Integrin ubiquitination mediates lysosomal degradation of fibronectin–integrin complexes via the ESCRT machinery, thereby controlling cell migration.

Integrin endocytosis and recycling have important roles during cell migration; however, it is not known whether the degradation of integrin proteins is also involved. In Developmental Cell, Herald Stenmark and colleagues now report that ubiquitination of α5 integrin and binding of fibronectin to α5β1 integrin mediates lysosomal degradation of both proteins via the endosomal sorting complex required for transport (ESCRT) machinery, and that this process is required for cell migration.

In human fibroblast cells, the authors first examined the intracellular localization of fibronectin and α5β1 integrin and demonstrated that a portion of α5 integrin colocalizes with fibronectin in the multivesicular endosomes (MVEs). This suggests that fibronectin and a fraction of α5β1 integrin are trafficked together from endosomes to lysosomes for degradative processing. Furthermore, they found that endogenous α5 is ubiquitinated at the cytoplasmic lysine residues in response to fibronectin binding, and that mutation of the cytoplasmic lysine residues in α5 reduced the rate of its degradation, showing that fibronectin-mediated ubiquitination of α5 integrin promotes its lysosomal degradation.

Next, they used confocal imaging to show that both α5 and β1 integrin are localized in the lumen of MVEs upon fibronectin binding. In contrast, mutant α5 integrin could not be sorted into MVEs even after fibronectin treatment, suggesting that ubiquitination of α5 is crucial for its proper endocytic sorting. Importantly, wild-type α5 integrin colocalized with 'active' β1 integrin inside the lumen of the MVEs, indicating that endocytic sorting of integrin proteins is dependent on ligand binding. Moreover, depletion of α5 integrin using short interfering (si)RNA inhibited sorting of fibronectin to the MVEs, with no effect on its endocytosis, and expression of mutant α5 protein could not restore fibronectin sorting. This confirms that fibronectin and its receptor α5β1 integrin are trafficked together and that this depends on ubiquitination of α5 integrin.

As ubiquitinated proteins directly interact with the components of the ESCRT, the authors next tested whether ESCRT machinery is involved in α5β1 degradation. Inhibition of ESCRT complexes with siRNA resulted in accumulation of ubiquitinated α5β1 integrin on endosomal compartments, and this was enhanced in the presence of fibronectin. Furthermore, ESCRT depletion induced accumulation of fibronectin in early endosomes, which confirms that fibronectin is trafficked together with α5 integrin, and its degradation is also dependent on the ESCRT machinery. Consistent with these results, immuno-electron microscopy showed that fibronectin and α5 integrin colocalize in enlarged early endosomes in ESCRT-depleted cells.

Western blotting demonstrated that protein levels of fibronectin and α5β1 integrin accumulated in the cells upon inhibition of ESCRT machinery, whereas no change was detected in their mRNA expression. This indicates inhibition of protein degradation and is consistent with an ESCRT-dependent mechanism.

Finally, live-cell time-lapse imaging showed reduced cell migration speed in the absence of α5 integrin; transfection with siRNA-resistant wild-type α5 integrin restored the cells' speed, whereas expression of mutant α5 did not have any effect. Taken together, these results demonstrate that ubiquitination of α5 integrin mediates proper migration of fibroblast cells through trafficking and degradation of fibronectin-bound α5β1 integrin via the ESCRT pathway. Further research will help elucidate the role of integrin degradation in cancer cell migration.

Iley Ozerlat

Reference:

Lobert V.H. et al.
Ubiquitination of α5β1 integrin controls fibroblast migration through lysosomal degradation of fibronectin–integrin complexes.
Dev. Cell, 19, 148-159 (2010)
Full text