Author Response
The following is the authors’ response to the original reviews.
eLife assessment
This fundamental study advances our understanding of the cell specific treatment of cone photoreceptor degeneration by Txnip. The evidence supporting the conclusions is convincing with rigorous genetic manipulation of Txnip mutations, however, there are a few areas in which the article may be improved through further analysis and application of the data. The work will be of broad interest to vision researchers, cell biologists and biochemists.
Reviewer #1 (Public Review):
Summary:
This is a follow-up study to the authors' previous eLife report about the roles of an alpha-arrestin called protein thioredoxin interacting protein (Txnip) in cone photoreceptors and in the retinal pigment epithelium. The findings are important because they provide new information about the mechanism of glucose and lactate transport to cone photoreceptors and because they may become the basis for therapies for retinal degenerative diseases.
Strengths:
Overall, the study is carefully done and, although the analysis is fairly comprehensive with many different versions of the protein analyzed, it is clearly enough described to follow. Figure 4 greatly facilitated my ability to follow, understand and interpret the study.
Weaknesses:
I have just one concern that I would like the authors to address. It is about the text that begins at line 133: "We assayed their ability to clear GLUT1 from the RPE surface (Figure 2A)". Please provide more details about this. From the figure it appears that n = 1 for this experiment, but given how careful the authors are with these types of studies that seems unlikely. How did the authors quantify the ability to clear GLUT1 from the surface? Was it cleared from both the apical and basal surface? (It is hard to resolve the apical and basal surfaces in the images provided). The experiments shown in Fig. 1H and Fig. 1I of PMID 31365873 shows how GLUT1 disappears only from the apical surface (under the conditions of that experiment and through the mechanism described in their text). It would be helpful for the authors to discuss their current results in the context of that experiment.
We repeated all eight AAV-Best1-Txnip alleles for RPE GLUT1 staining with more than three eyes of each condition. We also quantified the GLUT1 intensity on the RPE basal surface. A new Figure 2-figure supplement 1 with these data has been added to this submission. The results and conclusions are similar to those in our initial submission.
As mentioned in our provisional responses: GLUT1 on the basal surface of the RPE is more easily scored than that on the apical surface. The photoreceptor inner segments and Müller glia microvilli also have GLUT1, and their processes are juxtaposed and/or intertwined with the apical processes of the RPE, making the apical process GLUT1 staining of the RPE much more difficult to score. In some sections where the RPE and the retina separate, we can score the apical process GLUT1 staining of the RPE, but we do not always have this situation in our sections. The current quantification in the new Figure 2-figure supplement 1 thus concerns only the basal staining.
As a separate issue, Reviewer #1 mentioned the work of another group (Wang et al., 2019, PMID: 31365873), which claimed that, on the apical surface of the RPE, GLUT1 is down-regulated in a RP mouse strain, RhoP23H. We have not consistently observed such a down-regulation of GLUT1 in other RP mouse strains such as rd1, rd10 or Rho-/- (unpublished data; see review Xue and Cepko, 2023, PMID: 37460158). However, as we pointed out above, it is difficult to score GLUT1 staining on the RPE apical surface. It is even more difficult in the degenerating retina where RPE and photoreceptor processes degenerate. For reference, one can see images of degenerating RPE apical processes in Wu et al. 2021 (PMID: 33491671).
Reviewer #2 (Public Review):
The hard work of the authors is much appreciated. With overexpression of a-arrestin Txnip in RPE, cones and the combined respectively, the authors show a potential gene agnostic treatment that can be applied to retinitis pigmentosa. Furthermore, since Txnip is related to multiple intracellular signaling pathway, this study is of value for research in the mechanism of secondary cone dystrophy as well.
There are a few areas in which the article may be improved through further analysis and application of the data, as well as some adjustments that should be made in to clarify specific points in the article.
Reviewer #3 (Public Review):
Summary:
Xue et al. extended their groundbreaking discovery demonstrating the protective effect of Txnip on cone photoreceptor survival. This was achieved by investigating the protection of cone degeneration through the overexpression of five distinct mutated variants of Txnip within the retinal pigment epithelium (RPE). Moreover, the study explored the roles of two proteins, HSP90AB1 and Arrdc4, which share similarities or associations with Txnip. They found the protection of Txnip in RPE cells and its mechanism is different from its protection in cone cells. These discoveries have significant implications for advancing our understanding of the mechanisms underlying Txnip's protection on cone cells.
Strengths:
(1) Identify the roles of different Txnip mutations in RPE and their effects on the expression of glucose transporter
(2) Dissect the mechanism of Txnip in RPE vs Cone photoreceptors in retinal degeneration models.
(3) Explore the functions of ARrdc4, a protein similar to Txnip and HSP90AB1 in cone degeneration.
Weaknesses:
(1) Arrdc4 has deleterious effect on cone survival but no discussion on its mechanism.
(2) Inhibition of HSP90 is known to cause retinal generation. It is unclear why inhibition enhances the protection of Txnip.
As mentioned in our provisional responses, little was known about the function of Arrdc4 or HSP90AB1 in cones. We summarize some of the recent discoveries regarding these two proteins in the new Discussion:
“Arrdc4, the most similar α-arrestin protein to Txnip that also has Arrestin N- and C- domains, accelerated RP cone death when transduced via AAV (Figure 1). This observation suggests that Txnip has unique functions that protect RP cones. Recently, Arrdc4 has been proposed to be critical for liver glucagon signaling, which could be negated by insulin (Dagdeviren et al. 2023). The implication of this potential role in RP cone survival is unclear, but interestingly, the activation of the insulin/mTORC1 pathway is beneficial to RP cone survival (Punzo et al. 2009; Venkatesh et al. 2015).”
“Little is known about the function of HSP90AB1. Knocking down Hsp90ab1 improved mitochondrial metabolism of skeletal muscle in a diabetic mouse model (Jing et al. 2018). Knocking out HSP90AA1, a paralog of HSP90AB1 which has 14% different amino acids, led to rod death and correlated with PDE6 dysregulation (Munezero et al. 2023). Inhibiting HSP90AA1 with small molecules transiently delayed cone death in human retinal organoids under low glucose conditions (Spirig et al. 2023). However, the exact role of HSP90AA1 in photoreceptors needs to be clarified, and the implications for HSP90AB1 in RP cones are still unclear. ”
In addition, we used AlphaFold Multimer, an AI algorithm based on AlphaFold-2, to explore the possible interaction between TXNIP, PARP1 and HSP90AB1 in the revision. One of the predicted models is shown as the new Figure 5-figure supplement 2. The C-terminus of Txnip is predicted to link HSP90AB1 and PARP1 together in this model.
Recommendations for the authors:
Reviewer #1 (Recommendations For The Authors):
I have just one concern that I would like the authors to address. It is about the text that begins at line 133: "We assayed their ability to clear GLUT1 from the RPE surface (Figure 2A)". Please provide more details about this. From the figure it appears that n = 1 for this experiment, but given how careful the authors are with these types of studies that seems unlikely. How did the authors quantify the ability to clear GLUT1 from the surface? Was it cleared from both the apical and basal surface? (It is hard to resolve the apical and basal surfaces in the images provided). The experiments shown in Fig. 1H and Fig. 1I of PMID 31365873 shows how GLUT1 disappears only from the apical surface (under the conditions of that experiment and through the mechanism described in their text). It would be helpful for the authors to discuss their current results in the context of that experiment.
See our responses to Review #1’s public review section above.
Also, is the clearance from the RPE plasma membrane homogenous throughout the RPE monolayer?
In the area of AAV infection, the effects are very homogenous. In the uninfected area, the clearance does not occur, and we consider the uninfected area of the same eye to be an excellent internal control.
A statistical analysis (as was provided for other experiments in the manuscript) would help to make the surprising conclusion about C.Txhniip.C247S more convincing.
In this revision, we used the Mann-Whitney U test with the Bonferroni correction for GLUT1 intensity quantification. For the cone survival statistics, we used the t-test or ANOVA with Dunnett multiple comparison test. The information has been added to each figure legend.
Another improvement I suggest for this figure is to include normal full length Txnip as a positive control to show how completely it removes GLUT1 from the surface.
Added. See the new Figure 2-figure supplement 1.
Another point that should be discussed is - when Txnip prevents GLUT1 from reaching the surface does all the GLUT1 get fully degraded within the cell. A brief description of how Txnip influences GLUT1 stability and localization would be helpful.
We are unable to track the fate of the GLUT1 after it is removed, i.e. we do not see definitive intracellular staining. We do not know if this is due to degradation or a hidden epitope.
Minor point
(1) Confusing citation on lines 99-100: "We previously showed that overexpressing the Txnip wt allele in the RPE using an RPE specific promoter, derived from the Best1 gene (Esumi et al. 2009),.." makes it sound like Esumi et al. is the citation for their previous study, which is not correct.
We have amended this to: "We previously showed (Xue et al. 2021) that overexpressing the Txnip wt allele in the RPE using an RPE-specific promoter, derived from the Best1 gene (Esumi et al., 2009), did not improve RP cone survival."
Reviewer #2 (Recommendations For The Authors):
Regarding the manuscript, here are some suggestions that authors can take into consideration for the completeness of the study:
(1) The text references the relationship between α-arrestin and glucose metabolism in cone cells, but fails to provide an explanation for its specific involvement in glucose metabolism. Consequently, readers may struggle to discern the targeted metabolic pathway.
We understand this point from Reviewer, and would love to know more about its mechanism, which is one reason why we undertook the current study. The mechanism(s) by which Txnip affects metabolism remains to be elucidated. To summarize our findings from our previous study, we showed that LDHB, which converts lactate to pyruvate, was required for Txnip-mediated rescue. Addition of the LDHB gene, however, did not boost rescue. We also showed that mitochondrial size and membrane potential were improved, and the Na/K pump function was improved, in Txnip-treated cones. Improved mitochondria were not sufficient, however, as revealed by a PARP-1 KO mouse with improved mitochondria that did not extend cone survival. In addition, using a Txnip mutant that does not remove the glucose transporter, we still saw cone rescue, so this function cannot be required for Txnip-mediated rescue. How does Txnip lead to improved mitochondria and to a reliance on lactate? We do not know.
(2) Although the author conducted an experiment on arrdc14 due to its similarity to Txnip, the lack of clarification on why arrdc4, with a 60% amino acid similarity, did not yield the same effects as Txnip remains unaddressed. Highlighting structural disparities or differences in intracellular signaling pathways could potentially shed light on this incongruity. Subsequently, an additional experiment may be warranted to test the hypothesis regarding the effective component of α-arrestin for cone rescue.
Additional experiments are needed to learn of the relevant differences between Arrdc4 and Txnip, but are beyond the scope of our work at the present. However, we have added a paragraph on newly published data on the function of Arrdc4 in the new Discussion:
“Arrdc4, the most similar α-arrestin protein to Txnip that also has Arrestin N- and C- domains, accelerated RP cone death when transduced by AAV (Figure 1). This observation suggests that Txnip has unique functions that protect RP cones. Recently, Arrdc4 has been proposed to be critical for liver glucagon signaling, which could be negated by insulin (Dagdeviren et al. 2023). The implication of this potential role regarding RP cone survival is unclear, but interestingly, the activation of the insulin/mTORC1 pathway is beneficial to RP cone survival (Punzo et al. 2009; Venkatesh et al. 2015).”
(3) The utilization of distinct mutant Txnip variants to impact RPE, cones, and their combined influence is noted. A comparative table elucidating the impact of cone rescue on these three targets would greatly enhance clarity.
We presented these data in Figure 4 in a table format.
Additionally, the text does not definitively establish whether Txnip.C247S.LL351 and 352AA, as well as Txnip.C247S, indeed manifest discrepancies when exclusively affecting RPE.
We edited a sentence in Results to: “Similar to Best1-wt Txnip (Xue et al., 2021), Best1-Txnip.C247S did not show significant improvement of cone survival, ruling out the C247S mutation alone as promoting the cone survival by Best1-Txnip.C247S.LL351 and 352AA.”
(4) While the text mentions that Txnip stimulates lactate utilization within cones, it remains unclear whether this effect extends to RPE. If applicable, this trait could potentially contribute to its role in cone rescue.
We agree with the Reviewer, and hope to address this question in our next study.
(5) The discussion introduces the notion that one potential mechanism for cone rescue by Txnip.C247S involves facilitating unhindered movement of Thioredoxin for redox processes. To validate this hypothesis and elucidate the mechanics of Txnip's involvement in cone rescue, it may be prudent to conduct further experiments concentrating on the interaction between Txnip and thioredoxin. Alternatively, an experiment aimed at upregulating Thioredoxin expression would be a valuable addition.
We hope to address this question in the future. However, the effect may be more complicated than our simple hypothesis regarding release of Thioredoxin. More than a dozen proteins were found to differentially interact with Txnip vs. Txnip.C247S (Forred et al. 2016).
Reviewer #3 (Recommendations For The Authors):
(1) Glucose transporter 1 is identified as an important mechanism in the protection of cone degeneration. It is unclear why GLut1 is upregulated in retinal cells although the expression of Txnip mutants are specifically in the RPE in Figure 2.
This retinal GLUT1 upregulation was not consistently observed in the treated eyes, so we did not comment on it in the text.
(2) Mutant N. Txnip was mentioned in the discussion that it causes obvious retinal degeneration. The quantification of retinal thickness from Figure 2 will be more rigorous.
Unlike the robust effects of Best1-N.Txnip on RPE GLUT1 level, this negative effect of Best1-N.Txnip on ONL thickness was not consistent. This result does not undermine the other major conclusions. Therefore, we deleted the related sentence of the original text: “This hypothesis is supported by the observation that N.Txnip led to an obvious thinning of the outer nuclear layer of the wt retina, reflecting a loss of photoreceptors”. We did leave in the related finding as follows:
“The N-terminal half of Txnip (1-228aa) might exert harmful effects in the RPE, that negate the beneficial effects from the C-terminal half, suggested by the observation that its removal, in the C-terminal 149-397 allele, led to better cone survival when expressed in the RPE (Figure 2). In cones, the C-terminal half, including the C-terminal IDR tail, may cooperate with the N-terminal half, or negate its negative effects, to benefit RP cone survival. However, the C-terminal half is not sufficient for cone rescue when expressed in cones, as the 149-397 allele did not rescue.”
