Hsp47 promotes biogenesis of multi-subunit neuroreceptors in the endoplasmic reticulum

  1. Ya-Juan Wang
  2. Xiao-Jing Di
  3. Pei-Pei Zhang
  4. Xi Chen
  5. Marnie P Williams
  6. Dong-Yun Han
  7. Raad Nashmi
  8. Brandon J Henderson
  9. Fraser J Moss
  10. Ting-Wei Mu

  • Curated by eLife

    eLife logo

    eLife assessment

    This important study defines new functions for the ER-resident protein HSP47 in the quality control of multi-pass membrane receptor proteins. The evidence supporting the conclusions is solid, with rigorous biochemical assays employed in appropriate models. However, additional consideration regarding the mechanism of HSP47-dependent regulation of membrane protein quality control would have strengthened the study. This work will be of broad interest to cell biologists and biochemists interested in the fields of proteostasis membrane protein quality control, and neuroreceptor signaling.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Protein homeostasis (proteostasis) deficiency is an important contributing factor to neurological and metabolic diseases. However, how the proteostasis network orchestrates the folding and assembly of multi-subunit membrane proteins is poorly understood. Previous proteomics studies identified Hsp47 (Gene: SERPINH1 ), a heat shock protein in the endoplasmic reticulum lumen, as the most enriched interacting chaperone for gamma-aminobutyric acid type A (GABA A ) receptors. Here, we show that Hsp47 enhances the functional surface expression of GABA A receptors in rat neurons and human HEK293T cells. Furthermore, molecular mechanism study demonstrates that Hsp47 acts after BiP (Gene: HSPA5 ) and preferentially binds the folded conformation of GABA A receptors without inducing the unfolded protein response in HEK293T cells. Therefore, Hsp47 promotes the subunit-subunit interaction, the receptor assembly process, and the anterograde trafficking of GABA A receptors. Overexpressing Hsp47 is sufficient to correct the surface expression and function of epilepsy-associated GABA A receptor variants in HEK293T cells. Hsp47 also promotes the surface trafficking of other Cys-loop receptors, including nicotinic acetylcholine receptors and serotonin type 3 receptors in HEK293T cells. Therefore, in addition to its known function as a collagen chaperone, this work establishes that Hsp47 plays a critical and general role in the maturation of multi-subunit Cys-loop neuroreceptors.

Article activity feed

  1. Version published to 10.7554/elife.84798 on eLife
  2. eLife

    eLife assessment

    This important study defines new functions for the ER-resident protein HSP47 in the quality control of multi-pass membrane receptor proteins. The evidence supporting the conclusions is solid, with rigorous biochemical assays employed in appropriate models. However, additional consideration regarding the mechanism of HSP47-dependent regulation of membrane protein quality control would have strengthened the study. This work will be of broad interest to cell biologists and biochemists interested in the fields of proteostasis membrane protein quality control, and neuroreceptor signaling.

  3. eLife

    Reviewer #1 (Public Review):

    In this manuscript, Wang et al focused on defining the importance of the ER proteostasis factor HSP47 in regulating the folding, assembly, trafficking, and activity of GABAA receptors. Previous mass spectrometry-based interactomics identified HSP47 as the most enriched GABAA interacting chaperone in HEK293T cells. Here, the authors expand this study, demonstrating that HSP47 interacts with GABAA subunits in mouse brain homogenates and in vitro, demonstrating that HSP47 binds the alpha1 GABAA subunit with high affinity. They went on to show that depletion of HSP47 reduces the surface expression and activity of GABAA receptors in primary hippocampal neurons. Alternatively, overexpression of HSP47 increased the trafficking of GABAA receptors in HEK293T cells. Interestingly, chemical or genetic disruption of critical disulfide bonds within the alpha1 subunit of GABAA decreased interactions with HSP47, while increasing interactions with the ATP-dependent ER chaperone BiP, suggesting that HSP47 binds folded GABAA subunits in the ER lumen and promoting receptor assembly. Consistent with this, the authors employed a FRET-based system and biochemical assays to demonstrate that HSP47 enhances the assembly of GABAA receptors. Further, they demonstrate that overexpression of HSP47 could enhance the trafficking and surface activity of the epilepsy-associated alpha1(A322D) GABAA mutant. Finally, the authors show that HSP47 promotes the assembly and activity of other Cys-loop receptors including nAchR.

    Overall, this work expands our understanding of how membrane receptors including GABAA and nAchRs are folded and assembled. In particular, the demonstration that HSP47 works after canonical ATP-dependent chaperones such as BiP in promoting the assembly of GABAA receptors is intriguing, as it is beginning to demonstrate the sequence of events important for ER quality control of these membrane proteins. The use of multiple biochemical and genetic approaches to manipulate GABAA receptors and following the assembly, trafficking, and activity of these receptors is also a strength. However, one outstanding question is how does manipulation of HSP47 (with either overexpression or depletion) influence overall ER proteostasis and function. Can these effects be specifically attributed to HSP47 or does this reflect more global impairment of ER function induced by altered signaling through pathways such as the UPR? This is important because, while the authors do demonstrate direct interactions with HSP47, the direct importance of these interactions on the assembly and activity of Cys loop receptors remains somewhat unclear. Additional efforts addressing the specific impact of HSP47 manipulation on overall ER proteostasis should address this comment and allow for a more complete understanding of the role of HSP47 in regulating the assembly and trafficking of these proteins. Regardless, this is a solid manuscript that reviews new insights into membrane protein quality control and the importance of HSP47 in regulating ER proteostasis and function.

  4. eLife

    Reviewer #2 (Public Review):

    Wang and colleagues previously characterized the protein interactome for GABA subunits and identified HSP47 chaperone as a top interacting protein. Here, they follow up to assess the function of this HSP47-GABA interaction. Using primarily HEK293 cells, they provide evidence that the ER-resident HSP47 chaperone promotes the folding of GABA receptor subunits and the assembly of GABA subunits into multimeric ion channels. Interestingly, they demonstrate HSP47 can rescue the folding and function of a missense mutant A332D epilepsy-associated GABA subunit. They also demonstrate similar enhanced folding/function for acetylcholine receptor assembly. Overall, the experimental data are well-presented and provide insight into new ion channel clients whose folding and assembly are dependent on the HSP47 chaperone. The study also identifies HSP47 expression as a potential strategy to target and enhance the function of misfolded ion channels, and this may have broader biomedical therapeutic significance beyond GABA channels.

  5. eLife

    Reviewer #3 (Public Review):

    Wang et al. show a new role for the small heat-shock protein Hsp47 in the assembly and plasma membrane trafficking of GABAA receptors and other heptameric neuroreceptors. Hsp47 (SERPINH1) is primarily known as a collagen-specific molecular chaperone, but it has been increasingly recognized as important for other protein clients. In a prior mass spectrometry study from the same group, Hsp47 was identified as the most enriched interaction partner of GABAA neurotransmitter-gated ion channels. In this study, the authors now follow up on the functional role of Hsp47 for the GABAA heteromer assembly and its cell-surface trafficking.

    Strengths:
    The authors show convincingly that Hsp47 plays an important role in promoting the cell surface expression and activity of GABAA receptors. Knockdown of Hsp47 in rat primary neurons decreases endogenous GABAA protein subunits on the cell surface and GABA-induced currents. Overexpression of Hsp47 in HEK293T increases abundance and cell surface trafficking of exogenously expressed GABAA subunits. Importantly, the overexpression of Hsp47 also rescues cell surface expression and channel currents of epilepsy-associated mutant GABAA receptors (alpha1 A332D), which could point to a future avenue to ameliorate pathogenic misfolding. The authors use a variety of experimental approaches to glean the mechanism by which Hsp47 promotes GABAA cell surface expression. In vitro GST pulldown experiments confirm a direct interaction between Hsp47 and the alpha1 and beta2 subunits. Site-directed mutagenesis and DTT addition indicate that the formation of a disulfide bond in the alpha1 subunits is critical for the Hsp47 interactions, leading the authors to conclude that Hsp47 is likely to bind to a more folded state of the subunit. In contrast, the ER Hsp70 chaperone BiP binds more strongly when the disulfide bond is disrupted, which corresponds to a more misfolded state as indicative of more alpha1 in the insoluble fraction. FRET assays and non-reducing gels to monitor GABAA receptor assembly again show that Hsp47 overexpression promotes the formation of the alpha1-beta2 complex. However, while these experiments are generally carried out thoroughly and the data is presented well, the results are interpreted too narrowly to only support their proposed models without considering alternative possibilities (see more below). Lastly, the authors show that Hsp47 overexpression also enhances the cell-surface expression and peak currents of another heteropentameric Cys-loop superfamily neuroreceptor, namely the a4b2 nicotinic acetylcholine receptor.

    Weaknesses:
    The authors propose a compelling model in Figure 7 by which Hsp47 binds to a late-stage, largely folded alpha1 or beta2 subunit essentially acting as a holdase to promote assembly into larger dimers or other folding intermediates. However, the data in the manuscript would also support alternative models that the authors should more carefully consider. For instance, Hsp47 overexpression leads to a buildup of additional alpha1 and beta2 subunits (as described in lines 256-258 and seen in Fig. 4C), suggesting that Hsp47 may instead prevent subunits from getting degraded. Conclusions about Hsp47 binding after BiP to a largely folded state are indirectly based on shifts in the steady population of WT or misfolded GABAA subunits, but Hsp47 overexpression may in turn influence this equilibrium. Without any experiments examining the kinetics of protein interactions, degradation, or cell surface expression conclusions are difficult to interpret. Lastly, most experiments are carried out in HEK293T, which does not endogenously express GABAA or other neuroreceptors. There is a disconnect between the knockdown studies in rat primary hippocampal neurons and the overexpression experiments in HEK293T cells. The loss of GABAA receptor trafficking and function in the neurons could result from the secondary effect of the Hsp47 knockdown.

    Overall, the study provides valuable new insights into the client scope of the ER small heat shock protein Hsp47, advances our understanding of neuroreceptor proteostasis, and provides potential corrective strategies to enhance the expression of epilepsy-associated mutations through targeting Hsp47. Hence, the paper should have broader relevance for a readership interested in proteostasis, membrane protein trafficking, and neuroreceptor signaling. However, I recommend addressing the following comments, mainly because the study in its current form only incompletely corroborates the authors' conclusions about the mechanism by which Hsp47 facilitates the neuroreceptor subunit assembly:

    • For the in vitro experiments in Fig. 1, it would be important to show controls that the recombinantly expressed alpha1(ERD) adopts a well-folded state. Similarly, how did the authors ensure that the alpha1- and beta2-GST proteins adopt a folded (or near-folded) conformation?
    • In several experiments (e.g. Fig. 2A, Fig. 4B-C, Fig. 5B) IF staining or Western blots for the alpha1 and beta2 subunits are taken as a proxy for full GABA receptor assembly. Are the other subunits (e.g. gamma2) present and can they be detected?
    • Does Hsp47 knockdown in the primary hippocampal neurons leads to other changes in proteostasis network composition, e.g. UPR activation? This will be important to quantify to ensure that the reduced GABAA function can be directly attributed to the loss of Hsp47.
    • How are the Hsp47 knockdown and overexpression phenotype in the 2 different cell lines connected? If Hsp47 abundance is a limiting factor for GABAA proteostasis, it would be helpful to show (e.g. by lentivirus transduction) that additional Hsp47 can increase GABAA surface expression in the primary neurons.
    • Increased alpha1 and beta2 monomers in Fig. 4C suggest that the increase in receptor complex formation is likely due to more subunits being present when Hsp47 is overexpressed. Does Hsp47 prevent the degradation of excess or misfolded subunits? This can be easily tested with cycloheximide-chase or pulse-chase assays.
    • Does Hsp47 overexpression lead to more alpha1(A332D) monomer build up in cells (similarly to the WT alpha1)? The total level of alpha1(A332D) should be quantified for Fig. 5B. Similarly in Fig. 6A, does Hsp47 overexpression stabilize the abundance of nAChR subunits? The authors could easily quantify the abundance of individual subunits by Western blot.
    • Did the authors test the effect of Hsp47 overexpression on the trafficking of other misfolding-prone GABAA subunit variants? For therapeutic purposes, it will be important to evaluate a broader set of variants. Even if Hsp47 only restores select variants, these results would be useful for pinpointing a mechanism by which Hsp47 binds to the receptor subunits.

  6. Version published to 10.1101/2022.10.24.513629v2 on bioRxiv
  7. Version published to 10.1101/2022.10.24.513629v1 on bioRxiv