In vivo exchange of glucose and lactate between photoreceptors and the retinal pigment epithelium
Curation statements for this article:-
Curated by eLife
eLife Assessment
This research is valuable as it investigates metabolic shuttling between photoreceptors and retinal pigment epithelium (RPE) using in vivo infusion techniques and mouse models. The authors find that the retina significantly relies on circulating glucose, with photoreceptors being the primary consumers of glucose, which is convincing. However, the study has incomplete evidence to support the claims that photoreceptors can use lactate as a fuel source, that lactate exported from photoreceptors is utilized by the RPE, and that lactate contributes to the TCA cycle in the RPE. These claims need substantial revision to include potential alternative explanations or perform key experiments.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Photoreceptors in the retina of a vertebrate’s eye are supported by a tissue adjacent to the retina, the retinal pigment epithelium (RPE). The RPE delivers glucose to the outer retina, consumes photoreceptor outer segments discs, and regenerates 11-cis-retinal. Here we address the question of whether photoreceptors also provide metabolic support to the RPE. We use complementary approaches and animal models to show that glucose is the primary fuel for the retina, that photoreceptors are the primary cell type in the retina to consume glucose, and that lactate derived from photoreceptor glucose consumption is transported to and catabolized by the RPE. These data rigorously support and extend the concept of a metabolic ecosystem between photoreceptors and RPE.
Article activity feed
-
eLife Assessment
This research is valuable as it investigates metabolic shuttling between photoreceptors and retinal pigment epithelium (RPE) using in vivo infusion techniques and mouse models. The authors find that the retina significantly relies on circulating glucose, with photoreceptors being the primary consumers of glucose, which is convincing. However, the study has incomplete evidence to support the claims that photoreceptors can use lactate as a fuel source, that lactate exported from photoreceptors is utilized by the RPE, and that lactate contributes to the TCA cycle in the RPE. These claims need substantial revision to include potential alternative explanations or perform key experiments.
-
Reviewer #1 (Public review):
Summary:
In this manuscript, the authors sought to build upon their prior work, which suggested the presence of an outer retinal metabolic microenvironment using ex vivo and in vitro systems, by using in vivo methods and a multitude of genetic models. The authors convincingly demonstrate that the retina prefers circulating glucose to some other circulating fuel sources and that photoreceptors are the main consumers of glucose in the retina. However, the claims regarding the ability of photoreceptors to utilize lactate as a fuel source, that lactate exported specifically from photoreceptors is taken up by RPE and further utilized to support the TCA cycle in the RPE are incomplete or inadequate and would benefit from further experimentation to convince the reader of such biological processes. Considering …
Reviewer #1 (Public review):
Summary:
In this manuscript, the authors sought to build upon their prior work, which suggested the presence of an outer retinal metabolic microenvironment using ex vivo and in vitro systems, by using in vivo methods and a multitude of genetic models. The authors convincingly demonstrate that the retina prefers circulating glucose to some other circulating fuel sources and that photoreceptors are the main consumers of glucose in the retina. However, the claims regarding the ability of photoreceptors to utilize lactate as a fuel source, that lactate exported specifically from photoreceptors is taken up by RPE and further utilized to support the TCA cycle in the RPE are incomplete or inadequate and would benefit from further experimentation to convince the reader of such biological processes. Considering alternative explanations and performing key experiments to confirm or refute these claims would substantially improve the impact of this study.
Strengths:
The major strengths of this study are its in vivo infusion methodologies and utilization of mouse models that are devoid of photoreceptors or are photoreceptor-specific conditional knockouts to provide convincing evidence that the retina utilizes circulating glucose to a significant degree and photoreceptors are the main consumers of glucose in the retina. These in vivo studies are complemented by ex vivo experiments in retinal explants.
Weaknesses:
While the in vivo infusion methodologies are a clear strength, not utilizing these techniques or other in vivo methodologies with the genetic models that lack photoreceptors or photoreceptor-specific proteins and not providing in vivo metabolomics data from these infusions in the RPE is a major weakness. Also, some circulating fuel sources may not get into the retina in appreciable amounts, impacting some of the authors' claims. Another major weakness is that for many of the claims noted by the authors, alternative explanations have not been considered nor have the proper experiments been conducted to fully support or refute these claims. For example, the authors claim it is photoreceptors that utilize lactate upon knockout of Glut1. However, other cells in the retina, such as Muller glia, may be the ones actually catabolizing lactate based on prior studies and enzyme expression patterns and their kinetics to support photoreceptors via the production of other metabolites from lactate. This alternative has not been considered nor have experiments been conducted to refute this possibility. Additionally, the authors claim lactate exported from photoreceptors is being taken up by RPE. The models used to support this claim lack photoreceptors, or their ability to take up glucose. None of the models specifically address lactate export from photoreceptors. Finally, the authors claim lactate exported from photoreceptors can be oxidized to TCA cycle intermediates in the RPE in vivo. No experiments specifically addressed the downstream path of lactate exported by photoreceptors in RPE TCA cycle metabolism in vivo, so this conclusion is also not well supported. Hence, the claims need to be significantly amended with an acknowledgment of potential alternatives or with some key experiments performed.
-
Reviewer #2 (Public review):
Hass et al. use in vivo and ex vivo mouse models to explore and validate the use of glucose and lactate by the outer retina. While the authors' conclusions are not totally novel, their work uses powerful in vivo models to validate, strengthen, and support their conclusions. This data is an important step forward in the field's understanding of retinal metabolism.
They performed in vivo metabolite tracing with 5 different fuel sources and found that glucose was the primary fuel for TCA in the retina. While performing these experiments they measured the circulating levels of the tracer metabolites to ensure steady-state labeling which aids in the interpretation of the results. Showing the levels of the labeled tracer in the retina would be a nice addition to establishing if the tracer is getting into the …
Reviewer #2 (Public review):
Hass et al. use in vivo and ex vivo mouse models to explore and validate the use of glucose and lactate by the outer retina. While the authors' conclusions are not totally novel, their work uses powerful in vivo models to validate, strengthen, and support their conclusions. This data is an important step forward in the field's understanding of retinal metabolism.
They performed in vivo metabolite tracing with 5 different fuel sources and found that glucose was the primary fuel for TCA in the retina. While performing these experiments they measured the circulating levels of the tracer metabolites to ensure steady-state labeling which aids in the interpretation of the results. Showing the levels of the labeled tracer in the retina would be a nice addition to establishing if the tracer is getting into the target tissue.
To support their conclusions that the photoreceptors are the primary consumers of glucose in the retina, the authors used multiple mouse models either with photoreceptor degeneration or a retina lacking the primary glucose transporter. While the photoreceptor degeneration mouse model has some caveats that make interpreting the data challenging, the glucose transporter KO models are a powerful tool to show the changes in metabolite levels between the retina and RPE in a retina. These retinas are not degenerated and have more subtle metabolic rearrangements. Therefore decreases in glucose consumption and lactate export can confidently be attributed to the changes in the photoreceptor metabolism. This model also allowed the authors to show that when glucose uptake is limited the photoreceptors can use lactate.
The authors show in vivo data to support that the RPE uses lactate from the photoreceptors as a fuel source. They do very short-term tracing in vivo to show that the RPE has reduced lactate levels and TCA labeling in a mouse model lacking photoreceptors. There is no deficiency when the RPE is measured ex vivo. These data clearly show that the adjacent photoreceptor activity is impacting RPE metabolism.
The manuscript is well-written, and thorough and does a very good job detailing and explaining methods and concepts that are not straightforward. The authors address (and do not bury) confusing data that does not necessarily support their conclusions (for example glycolytic intermediates in Figure 3C being elevated. The authors even perform additional experiments to clarify artifacts they observed in the tracing of the degeneration model due to short-term ischemia.
-
Reviewer #3 (Public review):
This work addresses the metabolic interplay between photoreceptors and the adjacent supporting layer of the vertebrate retina, the retinal pigment epithelium (RPE). Prior work from the Hurley lab and others provided evidence, mainly in acutely dissected mouse retina and in cell culture, for the idea that although glucose enters the retina via the RPE, the photoreceptors use most of this glucose via glycolysis, producing lactate that is used by other cells such as Müller cells and RPE cells. In the current study, they build on this by showing that these same principles hold true in vivo, using organism-level stable isotope tracing, as well as in intact retina preparations. They also use several mutant mice that lack photoreceptors, or that lack glucose transporters in either rods or the whole retina, to …
Reviewer #3 (Public review):
This work addresses the metabolic interplay between photoreceptors and the adjacent supporting layer of the vertebrate retina, the retinal pigment epithelium (RPE). Prior work from the Hurley lab and others provided evidence, mainly in acutely dissected mouse retina and in cell culture, for the idea that although glucose enters the retina via the RPE, the photoreceptors use most of this glucose via glycolysis, producing lactate that is used by other cells such as Müller cells and RPE cells. In the current study, they build on this by showing that these same principles hold true in vivo, using organism-level stable isotope tracing, as well as in intact retina preparations. They also use several mutant mice that lack photoreceptors, or that lack glucose transporters in either rods or the whole retina, to examine the contribution of photoreceptors to retinal glucose uptake. While many of the concepts were introduced in earlier work, it is an important expansion of this work to show these same mechanisms function in vivo. The authors also use other labeled fuels, lactate, and palmitate, to characterize their use in the presence or absence of glucose transport.
The paper presents a nice combination of in vivo experiments (with a steady infusion of labeled metabolites into the circulation of a living mouse) with ex vivo experiments that allow the monitoring of lactate production and temporal control of labeling.
Overall, the work provides convincing evidence that in the eye of a living mouse, photoreceptors are the main consumers of glucose in the retina, and the main producers of lactate. It seems less clear that the incorporation of labeled glucose into TCA metabolites in the RPE is dependent on the photoreceptor processing of glucose to lactate. Figure 5D is cited as the evidence that "much less m+3 lactate reaches the RPE-choroid in AIPL-/- mice than in controls," and indeed there is much less labeled lactate; but the downstream labeling of citrate is not substantially affected. It is also hard to discern whether these in vivo experiments provide evidence that photoreceptor-derived lactate suppresses glucose oxidation in RPE cells (as shown in vitro in Kanow et al., 2017).
-
Author response:
We thank the reviewers for their thoughtful reading and review of our manuscript. These reviews make clear that, for this work to be complete, we must make progress on the following fronts:
(1) Expand the discussion to better incorporate alternate explanations of our data
(2) Improve data visualization and experimental support or an experimental refutation for the following concepts
a. Photoreceptor-derived lactate exported specifically from photoreceptors is utilized in the RPE TCA cycle
b. Photoreceptors can utilize lactate as a fuel source when starved of glucose
To address these concerns, we will focus our efforts on infusing 13C6-glucose into rodΔglut1 mice. Lactate is not made without glucose, so this experiment should indicate whether glucose utilization in photoreceptors provides lactate to the RPE, and whether …
Author response:
We thank the reviewers for their thoughtful reading and review of our manuscript. These reviews make clear that, for this work to be complete, we must make progress on the following fronts:
(1) Expand the discussion to better incorporate alternate explanations of our data
(2) Improve data visualization and experimental support or an experimental refutation for the following concepts
a. Photoreceptor-derived lactate exported specifically from photoreceptors is utilized in the RPE TCA cycle
b. Photoreceptors can utilize lactate as a fuel source when starved of glucose
To address these concerns, we will focus our efforts on infusing 13C6-glucose into rodΔglut1 mice. Lactate is not made without glucose, so this experiment should indicate whether glucose utilization in photoreceptors provides lactate to the RPE, and whether that lactate is used in the TCA cycle.
The reviewers also noted that changes in 13C labeling of RPE TCA cycle intermediates downstream of lactate is not obvious (between C57BL6J mice and AIPL1-/-). We think that at least in part, this is a consequence of the way we presented the data. We will improve how we display our data so that the differences of incorporation of 13C in TCA cycle intermediates in control and AIPL1-/- RPE is clearer.
-
-
-
-