Transcription factor EB (TFEB) is a new therapeutic target for Pompe disease

A recently proposed therapeutic approach for lysosomal storage disorders (LSDs) relies upon the ability of transcription factor EB (TFEB) to stimulate autophagy and induce lysosomal exocytosis leading to cellular clearance. This approach is particularly attractive in glycogen storage disease type II [a severe metabolic myopathy, Pompe disease (PD)] as the currently available therapy, replacement of the missing enzyme acid alpha-glucosidase, fails to reverse skeletal muscle pathology. PD, a paradigm for LSDs, is characterized by both lysosomal abnormality and dysfunctional autophagy. Here, we show that TFEB is a viable therapeutic target in PD: overexpression of TFEB in a new muscle cell culture system and in mouse models of the disease reduced glycogen load and lysosomal size, improved autophagosome processing, and alleviated excessive accumulation of autophagic vacuoles. Unexpectedly, the exocytosed vesicles were labelled with lysosomal and autophagosomal membrane markers, suggesting that TFEB induces exocytosis of autophagolysosomes. Furthermore, the effects of TFEB were almost abrogated in the setting of genetically suppressed autophagy, supporting the role of autophagy in TFEB-mediated cellular clearance.

Thank you for the submission of your manuscript to EMBO Molecular Medicine. We have now heard back from the three referees whom we asked to evaluate your manuscript. You will see that they consider your manuscript to be of potential interest. However, they also raise significant concerns about the study, which would have to be addressed in a substantive revision of the manuscript.
Reviewer #1 highlights that additional data on the function of TFEB-treated Pompe diseased muscles is required to strengthen the significance of TFEB as a potential therapeutic target.
Importantly, reviewer #2 requires in-depth characterisation and quantification of exocytosis in myofibers overexpressing TFEB. In addition, this reviewer would like to see more careful characterisation of the observed apoptosis and an investigation into possible necrosis phenotypes. Finally, this reviewer is concerned about the observed mitochondrial changes and requests a more thorough analysis of the effect of TFEB transfection on mitochondria.
Reviewer #3 highlights the potential for toxicity of the proposed treatment, which we would ask you to address experimentally.
In our view the suggested revisions would render the manuscript much more compelling and interesting to a broad readership. We therefore hope that you will be prepared to undertake the recommended experimental revision. TFEB to clearance of glycogen filled lysosomes by using adult myofiber and myoblast obtained from GAA knockout mice. The overexpression of TFEB in myotubes and myofibers ameliorates the glycogen load and decrease the size of lysosomes. By using different tools the authors have also shown that TFEB ameliorates autopahgosome-lysosome fusion and induces lysosome exocytosis. Finally they extend in vivo the results obtained in cell culture by overexpressing FTEB in skeletal muscles of GAA knockout mice. The overall paper is well written but the findings are not novel. Indeed, Ballabio group have already published that TFEB overexpression decrease glycogen content in fibroblast of Pompe patients (Medina DL et al. 2011, Transcriptional Activation of Lysosomal Exocytosis Promotes Cellular Clearance. Dev. Cell. 21, 421-430). The authors should consider the following points. Point 1. Lysosomal exocyotsis is the main issue of this paper. However all the findings that support exocytosis are based on imaging system to detect relocation of Lamp1 positive vesicles from the cytoplasm of myotube to plasma membrane. However, Lamp1-cherry has been never shown on the plasma membrane of GAA-/-myofibers that overexpress TFEB. The authors must better characterize that truly exocytosis happens in their condition. Release of lysosomal enzymes (acid phosphatase, b-galactosidase, b-exosaminidase) have to be shown to occur in TFEB overepressing myofibers. Moreover, the most important piece of data is the clearance of glycogen-filled lysosomes in GAA-/-myofibers. The release of glycogen in the cell culture medium must be studied. Furthermore, exocytosis inhibitors should be used to block the beneficial effect of TFEB expression on glycogen load. Finally the authors should use EM to show the lysosomal/autophagosomal exocytosis event in TFEB overexpressing fibers.
Point 2. Lamp1 is sorted to lysosomes via endosomes pathways. The authors should monitor whether TFEB overexpression affect endosome trafficking that place Lamp1-cherry in the secretory pathways via Tubular Sorting Endosome.
Point 3. Fig. 3B. There re two bands and the lower one is indeed increased in ad-TFEBmut samples. Explain why two bands and whether this increase is significant and what it means. Point 4. Fig 3C. TUNEL technique can not distinguish necrosis from apoptosis unless it is coupled with morphological analyses and caspase 3/6/7 activation. Moreover the blebbing present in TFEBoverexpressing fibers is also an index of necrosis or necroptosis. Authors must characterized better apoptosis by monitoring caspase 3 activation and chromatin fragmentantion and condensation. Moreover also necrosis should be studied by determining LDH release, propidium iodide incorporation or membrane permeabilization (Evans blue uptake), RIP1/3 accumulation.
Point 5. Fig. S5A, B). The figure of autophagy activation in Torin1/2 treated cells are too small to sustain authors claim. Please show higher magnification in order to reveal LC3 positive puncta and quantify the increase of vesicles. Alternatively show LC3 lipidation by western blotting. The authors claim: "A striking increase in the levels of p-ERK1/2 in TFEB-treated PD myotubes points to such a possibility in muscle cells" is incorrect. The data are instead supporting the concept that ERK1/2 are downstream TFEB. To sustain their claim the authors should block ERK by treting cells with inhibitors (or knockdown ERK) . Point 6. The authors showed an increase of lysosome motility and fusion by TFEB overexpression in myofibers of GAA knockout mice. They also found that the number of large lysosome is reduced. However the movies and the pictures also show the appearance of huge lysosomes that came out from the fusion of enlarged lysosomes. Therefore, it looks like that TFEB induces the fusion of lysosome into giant vacuoles. Please quantify and show the distribution of lysosome sizes in control and TFEB overexpressing fibers.
Point 7. Fig 4B and movies. Most of the time TFEB overexpression induces the formation of blebs, The blebs are index of cellular swelling and therefore of pre-necrotic events ( Fig 4B, Fig 6B, Movie 1, Movie 6). Necrosis should be studied (see above point 4). Moreover, many movies of TFEB overexpressing fibers show movement of myofibers that causes changes of focus and area of observation (fiber become larger or smaller) (Movie 1, Movie 6, Movie 7). Therefore, it is very difficult to follow vesicles trafficking and fusion events and to interpret the data. Movie 8 is not working. Point 8. Fig. 8 A and B. Mitochondria morphology and distribution looks different in TFEB overexpressing muscles. Indeed the mitochondria localized nearby the Z-line (the correct localization) are bigger in panel B (TFEB transfected muscle). Conversely the accumulation of big and abnormal mitochondria closed to giant lysosomes that is depicted in panel A is absent in panel B. Is TFEB affecting mitochondria number, morphology and localization? In the case of an increase of mitochondria, are mitochondria uncoupled? Is glucose and lipid homeostasis altered? In fact, decrease of glycogen load in lysosome might be a consequence of a decrease glucose uptake or an increase of glucose utilization from uncoupled mitochondria. All these conditions must be analyzed in order to give a clear picture of TFEB action in muscles.

Minor points:
The authors often use the words Lamp1 and Lamp as synonyms. This is wrong since it exists also Lamp2 proteins that have different functions (CMA) from Lamp1.

Referee #3 (Comments on Novelty/Model System):
This is an outstanding paper. The only reservation is with regard to the potential medical impact, because the treatment might have unsuspected toxicity.

Referee #3 (General Remarks):
This is an outstanding manuscript describing application of a new therapeutic strategy to Pompe disease.
Major concerns: 1) The potential for toxicity from induction of apoptosis in PD muscle is high. It would be reassuring to know that toxicity in TFEB-treated mice was absent (central nuclei, elevated transaminases or creatine kinase) in AAV experiment. 2) Lacking the information in response to my first concern, it would be better to highlight the potential for toxicity in vivo by addressing this subject in the discussion. A long-term experiment that addressed these questions would be the only definitive response, and that might not be available for the current manuscript. Therefore the possibility of unexpected toxicity must be addressed in the discussion.
Minor concerns: 1) Abstract: Why is Glycogen Storage Disease capitalized? 2) Page 5, paragraph 3: "studying the lysosomal defect" would be correct. 3) Page 5, paragraph 4: "with an adenovirus vector expressing" would be correct. The reviewer has only minor points. 1. The author said TFEB is new therapeutic target in title. The author showed that the clearance of autophagic build-up in myofibers well, but they did not supply the property of treated PD muscles. Providing information of the treated mouse muscles, motor performance, contractile force, muscle size and pathology, will help the readers to consider the therapeutic possibility of this treatment.
We have addressed the point raised by the reviewer experimentally, and the data have been included in the revised manuscript. According to the reviewer's suggestions we evaluated fiber pathology of treated muscles. Hematoxylin-eosin staining did not show gross alterations of the muscular architecture in TFEB-treated gastrocnemii, compared to untreated muscles, or signs of toxicity (Supporting information; Figure S9 A). No differences were seen in the number of centralized nuclei in treated and untreated muscles. TUNEL and caspase 3 staining of muscle preparations from TFEB-treated muscles did not show an increase of apoptotic cells (Supporting information, Figure  S9 B and C). Muscle size was also evaluated and no significant differences were observed between untreated and TFEB-treated muscles (Supporting information, Figure S9 D and E).

The authors showed the increase of the motility of lysosome after TFEB treatment. Beside of the importance of movement of lysosomes toward the plasma membrane for reducing their numbers in myofibers, what is the impact in increment of the velocity of lysosomes?
As the reviewer correctly points out, an increase in lysosomal "dynamics" may impact not only on lysosomal exocytosis but may have more general consequences on intracellular trafficking. For example, we have previously demonstrated that TFEB enhances the fusion between lysosomes and autophagosomes (Settembre et al. Science 2011). However, while we recognize the importance of this point, it would be very difficult, and out of the scope of this paper, to determine the overall impact in increment of the velocity of lysosomes.
3. Discussion is too long. The first paragraph in Discussion, which is repetitive to Introduction, can be omitted.
We shortened the discussion in the revised manuscript. We disagree with the reviewer's opinion on the lack of novelty of our paper for the following reasons:

TFEB expression solely could make a clearance of the accumulated glycogen with lysosomes
1. If one wants to test any new therapeutic approach to Pompe disease, the most relevant systems to use are multinucleated muscle cells, muscle fibers, and whole muscle, the tissue that is phenotypically affected and most difficult to treat by the currently available drug. We have used all three systems. The data on TFEB in Pompe disease that we published in a previous study (Medina et al. Dev. Cell 2011) were extremely limited (one figure subpanel). Indeed they consisted of preliminary experiments in fibroblasts from a single patient. No data on muscle, which is the major site of pathology, and no in vivo data from Pompe mouse models were previously published. Therefore, all findings in the current manuscript are novel.
2. We feel that the Referee overlooked the importance of the data on autophagy, in particular the results of TFEB expression in autophagy-deficient mice. The idea that functional autophagy is required for fully efficient TFEB-mediated clearance is an entirely new concept. (acid phosphatase, b-galactosidase, b-exosaminidase) has to be shown to occur in TFEB overexpressing myofibers. Moreover, the most important piece of data is the clearance of glycogen-filled lysosomes in GAA-/-myofibers. The release of glycogen in the cell culture medium must be studied.

Furthermore, exocytosis inhibitors should be used to block the beneficial effect of TFEB expression on glycogen load. Finally the authors should use EM to show the lysosomal/autophagosomal exocytosis event in TFEB overexpressing fibers.
We have added the data on the release of lysosomal acid phosphatase in the medium following TFEB treatment of myotubes (Supporting information, Table S1). These data, combined with the appearance of LAMP1 on the cell surface, provide strong evidence of lysosomal exocytosis (as now emphasized in the revised manuscript). As for the glycogen measurement in culture medium, the biochemical assay for glycogen (which is not terribly sensitive in the first place) is based on the conversion of glycogen to glucose, followed by glucose measurement. Myotubes grow and differentiate in the medium with high concentration of glucose, which precludes accurate measurements due to the high background. The data in myotubes strongly suggest that TFEB induces lysosomal exocytosis in muscle. In addition, we tried to measure lysosomal exocytosis in live TFEB-transfected fibers but soon realized that the separation of transfected from nontransfected fibers (a procedure required for any comparison) is not feasible, particularly given the relatively low transfection efficiency. Picking up only transfected fibers (which are surrounded by huge numbers of non-transfected fibers) and subsequently exposing them to multiple rounds of replating invariably results in fiber contraction.

Point 2. Lamp1 is sorted to lysosomes via endosomes pathways. The authors should monitor whether TFEB overexpression affect endosome trafficking that place Lamp1-cherry in the secretory pathways via Tubular Sorting Endosome.
In Pompe disease glycogen accumulates in hugely enlarged Lamp1-positive structures. It is well known that these structures are lysosomes and not tubular sorting endosomes, which do not accumulate glycogen. In addition, in a previous study we demonstrated that TFEB overexpression has no effect on endocytic secretory pathways (Medina et al. Dev Cell 2011).

Point 3. Fig. 3B. There are two bands and the lower one is indeed increased in ad-TFEBmut samples. Explain why two bands and whether this increase is significant and what it means.
LAMP1 never runs as a single band on western; additional bands most likely represent different glycosylation forms of the protein. The increase in LAMP1 in TFEB-treated cells is not unexpected since TFEB was shown to stimulate lysosomal biogenesis. We have added this information to the revised manuscript.
Point 4. Fig 3C. TUNEL technique cannot distinguish necrosis from apoptosis unless it is coupled with morphological analyses and caspase 3/6/7 activation. Moreover the blebbing present in TFEBoverexpressing fibers is also an index of necrosis or necroptosis. Authors must characterized better apoptosis by monitoring caspase 3 activation and chromatin fragmentation and condensation. Moreover also necrosis should be studied by determining LDH release, propidium iodide incorporation or membrane permeabilization (Evans blue uptake), RIP1/3 accumulation.
We have followed the reviewers' suggestions and measured LDH release and caspase-3 activity. No signs of apoptosis or necrosis were detected according to these methods. This information was added to the paper. As for the blebbing in TFEB-overexpressing fibers, we would like to emphasize that this phenomenon is not seen in freshly isolated fibers or fibers maintained for days in culture. The blebbing is only seen when the fibers are exposed for hours to time-lapse confocal microscopy. We say this very clearly in the revised manuscript.
Point 5. Fig. S5A, B). We provide a new figure S5 with insets for enlarged views and with additional data on LC3. We have also modified the discussion to address the referee's suggestion regarding ERK.
Point 6. The authors showed an increase of lysosome motility and fusion by TFEB overexpression in myofibers of GAA knockout mice. They also found that the number of large lysosome is reduced. However the movies and the pictures also show the appearance of huge lysosomes that came out from the fusion of enlarged lysosomes. Therefore, it looks like that TFEB induces the fusion of lysosome into giant vacuoles. Please quantify and show the distribution of lysosome sizes in control and TFEB overexpressing fibers.
We have added the graph on distribution of lysosomal sizes in control and TFEB-overexpressing fibers. The difference between the two conditions is highly significant.

Point 7. Fig 4B and movies. Most of the time TFEB overexpression induces the formation of blebs,
The blebs are index of cellular swelling and therefore of pre-necrotic events (Fig 4B, Fig 6B,  Regarding "blebbing" see point 4 above. It is no surprise that the fibers may move a bit, and indeed it is tricky to track vesicular movements. Time-lapse microscopy of live fibers is in general not a trivial endeavour. Suffice it to say that during each session, out of ~30 selected fibers, only one or two fibers (and often none) survive. However, when we analyse the fibers for vesicle velocity, we very carefully select the fibers and the time points with minimal twitching. This information was included in the original manuscript (Supporting Information): "To minimize the contribution of fiber movement to lysosomal velocity measurements, manual tracking was performed only between time points when fiber twitching was not ostensible." We fixed Movie 8 -thank you very much for pointing to the problem.
Point 8. Fig. 8 A  We agree with the reviewer on the fact that TFEB may have an effect on glucose/glycogen metabolism. However, the fibers are isolated and analysed 4 days after TFEB transfection, and any "short term" change in glucose/glycogen metabolism would have negligible effects on muscle glycogen accumulation, which takes much longer to build in Pompe disease. The Referee also raises an interesting question about mitochondria in general. We have looked at mitochondria using MitoTracker in live fibers and immunostaining of fixed fibers with cytochrome c. We found no difference in the mitochondrial distribution and abundance in control and TFEBtreated fibers.
As for the EM analysis, the purpose of Figure 8 is to show the striking reduction in size and number of glycogen-filled lysosomes. We provide additional images (see below) to the point-by-point response. These show that big mitochondria (arrows in panel A, left) can also be seen in preparations from untreated muscles.
The other concern raised by the reviewer is the accumulation of giant mitochondria around lysosomes in untreated muscle. We regularly see hugely enlarged lysosomes without adjacent clusters of mitochondria (arrowhead in panel A, right) in untreated samples. Conversely, we could see accumulation of mitochondria (arrows in panel B, left) near giant lysosomes in TFEB-injected mice. This illustrates the heterogeneity of muscle tissue rather than a strict pattern of mitochondrial distribution.
To address the reviewer's concerns about the morphology of mitochondria we performed additional analyses. Mitochondria in treated and untreated muscles look quite similar in terms of size, cristae morphology, and electron density. A quantitative analysis of the number and size of mitochondria has now been added (Figure 8 C).
It is worth noting, though, that such accumulations of mitochondria near lysosomes in TFEBexpressing muscles are less prominent likely due to the increased mitophagy. Indeed, the only major difference in the appearance of mitochondria between control and TFEB-injected mice was the presence of mitochondria inside an autophagosome (see panel C). We think however that increased mitophagy is just a consequence of general increase in autophagy triggered by TFEB overexpression rather than a result of the mitochondria dysfunction.