Hello, bon jour, konichiwa, ni hao. Thank you for joining us today. I’d like to welcome you to the webinar entitled Autophagy from Fundamental Mechanisms to Mechanical Stress in Physiology and Disease. Before we begin, I want to check that the audience can see the title slide. If you can’t see it, please be sure you have Flash enabled.
My name is Darcey Miller, and I will be serving as your moderator for this webinar. I’m a product manager at Novus Biologicals, one of the Bio-Techne brands. And our featured speaker is Dr. Patrice Codogno. Dr. Codogno is a research director of the French Institute for Health and Medical Research and a group leader at the Institut Necker Enfants-Malades. He is also an honorary professor at the University of Buenos Aires in Argentina and associate editor of the Autophagy journal since it was established. His research is dedicated to studying the basic aspects of autophagy, including membrane dynamics, autophagy signaling, and the role of autophagy and the stress response to mechanical stress in cancer treatment. One of his most important contributions to the field of autophagy is from his 2000 Journal of Biological Chemistry paper with Fred Meijer in which hey provided the first evidence for opposing roles for class I PI3K and class III PtdIns3K in mammalian cells. Before handing over control to Dr. Codogno, I would like to make a brief announcement about the sponsor. Bio-Techne brings together the prestigious life science research brand of R&D Systems, Novus Biologicals, Tocris Bioscience, ProteinSimple, and Advanced Cell Diagnostics to provide the scientific research community with a comprehensive and world class product portfolio of reagents, assays, instruments, and custom manufacturing and testing services. I would like to highlight that Novus’ LC3B antibody is the most widely trusted and cited antibody to monitor autophagy induction, and Tocris offers a wide range of small molecules for the study of autophagy.
Also, we’d like to invite you to ask questions during the webinar by using the Ask A Question Box, which is located just below the presentation screen. During the Q&A session directly following the presentation, I will then ask Dr. Codogno your questions. If you are interested in additional educational content on autophagy, please go to the Resources tab located in the upper right-hand corner of your screen for further information.
And with that brief introduction, I’ll hand over control to Dr. Codogno. So thank you, Darcey, for the introduction. Hi, everybody. So, today in this webinar, I would like to discuss two lines of research we have in the lab concerning autophagy and more precisely macroautophagy. First, we are interested in the regulation of the process. And in the second part of the talk, I will discuss more physiological and pathophysiological aspects related to the role of autophagy in stress response and more precisely on mechanical stress.
So probably, as most of you know, macroautophagy runs in eukaryotic cells and starts from a membrane named phagophore, and the membrane elongates to form a sealed vacuole named autophagosome. And this autophagosome is able to trap either in a bulk manner or in a selective manner a fraction of the cytoplasm. And the end of the macroautophagic pathway is a fusion of autophagosome with an acidic compartment endosomal compartment and lysosomal compartment and at the end of the day, cargo are degraded in the lysosomal compartment. So, of course, autophagy run at the basal level in most cell types, and the process can be induced by various stimuli; and in this sense, macroautophagy is a stress response. Macroautophagy can be stimulated, for example, by nutritional stress (meaning by starvation of nutrient), by oxidative stress, by intracellular stress situations such as ER stress (endoplasmic reticulum stress), and also accumulation of aggregate of protein or accumulation of damaged organelles.
So, as a consequence, autophagy is very important to provide energy by degradating proteins and lipids, for example, to produce amino acids and fatty acids to maintain cellular metabolism. But it’s also quality control for the cytoplasm by the degradation of damaged organelle and the removal of protein aggregate. So the process isorganelle and the removal of protein aggregate. So the process is associated with a lot of membrane remodeling and is also able to modulate signaling pathway by degradating some protein involved in signaling.
So, as a consequence, its mechanism is very important in biology and, for example, autophagy is important for cell survival and to keep the balance between survival and cell death. It’s also engaged in differentiation development in the regulation of metabolism, and we know now that autophagy plays a role, also, in innate immune response and acquired immune response. And a phase of autophagy is also engaged in sequestrating microorganisms that invade the cytoplasm – bacteria or virus cells.
So, of course, this regulation of autophagy is associated with many diseases, and some of these diseases are characterized, in fact, by mutation on autophagy protein, and autophagy plays a role from cancer to chronic inflammatory diseases such as Crohn’s disease, neurodegenerative disease, and also metabolic diseases such as obesity and type 2 diabetes.
So, as I told you in the introduction, I will first discuss some aspects of the regulation of autophagy we are interested in the lab. And as Darcey mentioned, our story started with the discovery, in 2000, in collaboration with Fred Meijer, in Amsterdam, that two PI3 kinases are engaged in the regulation of autophagy. So class I phosphatidylinositol 3-kinase downstream of the insulin receptor, in fact, in most cells has an inhibitory effect on autophagy because it activates mTOR, and the kinase mTOR is known to repress autophagy.
On the other side, in fact, autophagy is stimulated by class III phosphatidylinositol 3-kinase. And I would mention, you know, that in the seminal discovery by Paul Gordon nd Pierre Seglen that 3-methyladenine, a well-known inhibitor of autophagy, was, in fact, an inhibitor of PI3 kinase. And following on with this discovery, Fred Meijer and Edward Blemmaart show, in fact, that wortmannin (another PI3 kinase inhibitor) are also able to block autophagy by blocking the activity of class III PI3 kinase. Of course, on this slide, I would like to mention the seminal discovery by Beth Levine of Beclin 1. And Beclin 1 is one of the autophagy proteins that interact with class III PI3 kinase.
So, the product of class III PI3 kinase, , as you know, is phosphatidylinositol-3 phosphate. This lipid is very important in autophagy, but it’s not restricted to the autophagic pathway. In fact, the major reservoir of this lipid is the endocytic pathway early endosome and inner vesicle of late endosome. But PI3P is also present at the cell surface close to the primary cilium and also associated with phagocytosis.
So an important step to the understanding of the function of PI3P in autophagy and to the selectivity for autophagy was brought by the discovery by Ohsumi and colleagues that, in fact, class III PI3 kinase – also known as Vps34 – can be in different complexes. In fact, complex I where Vps34 associates with its partner, Vps15, interact with ATG14 and Beclin 1 or Vps30. And this complex controls the formation of autophagosome. Whereas when Vps34 and Vps15 are associated with the protein UVRAG in complex II, this complex control endocytosis and/or late-stage of the autophagic pathway. And the activity of this complex is regulated by a protein called Rubicon that inhibit the activity of this complex.
Our first understanding on the activity of this complexes was recently brought by structural data showing, in fact, that these two complexes can act on different membranes. For example, complex II is able to produce a PI3P at the surface of large liposomes and small liposomes. Whereas complex I is only able to produce PI3P at the surface of small liposomes. These results show the importance of the geometry of the membrane to determine the activity of these complexes.
And moreover, it has been recently shown that complex II bind more weakly to phosphatidylinositol containing membrane that complex I. So in fact, phosphatidylinositol is a substrate for Vps34. And this resulted, very interestingly, in light of the recent report by Mizushima and colleagues, in Japan, and this group showed, in fact, that the ULK1 complex, which is a first complex engaged in the formation of the autophagosome, initiate autophagosome formation at phosphatidylinositol synthase enriched endoplasmic reticulum subdomain.
So with this in mind, we can now put all the players in the picture. And you can see on these slides that the autophagy machinery with ATG protein are recruited at the endoplasmic reticulum. So, the PI3 kinase complex I, as I told you previously, is controlled by the activity of the ULK1 complex by phosphorylation and ubiquitination. The PI3 kinase complex I, in fact, produces PI3P. And this lipid, in fact, has at least two interactors in the autophagic pathway. First, PI3P interacts with a protein named DFCP1. And DFCP1, in fact, is localized in a part of the ER called the omegasome. And this was discovered by Nicky Ktistakis and colleague, in UK. And, as you can observe from the reconstitution, in fact, the ER (in purple) and the omegasome serve as a sort of cradle to elongate the phagophore (in yellow).
Another important partner for PI3P is a protein named WIPI. And WIPI proteins are, in fact, the yeast ortholog of Atg18. And this protein is important to recruit on the phagophore the first ubiquitin-like conjugation system where Atg5 is covalently linked to Atg12, and this conjugate interacts with the protein Atg16. And in fact, Atg16 interact with WIPI. And, this conjugate is very important to activate the second conjugate in the autophagic pathway; meaning the Atg8 or in mammalian cells, the LC3 conjugate. And, in this conjugate, in fact, the C-terminus of LC3 is conjugate to the polar head of PE or phosphatidylethanolamine. And at the end of the day, when the autophagosome is formed and released, only a fraction of LC3 remains associated with the autophagosome. And only the LC3 associated with the inner membrane will be transported to the lysosome, and the part of LC3 associated with the external membrane is released to the cytosol.
So, another important player is the protein Atg9. And Atg9, in fact, is the only transmembrane Atg protein. And vesicles containing Atg9 contribute to the very early stage of the autophagosome formation. Roughly, we can say that Atg9 controls a number of autophagosome forms, and LC3 controlling factor size of the autophagosome. So of course, the things are more complex. We know that the ER is important to contribute to autophagosome formation. But in the literature, now we know that many membranes can contribute to the autophagosome formation from Golgi apparatus but also plasma membrane, endosomes, and also I would like to focus on the importance of contact site.
And in fact, Yoshimori and colleagues showed that contact site between the endoplasmic reticulum and mitochondria are important to recruit the autophagy machinery to initiate the autophagosome formation. And probably, as you know, the endoplasmic reticulum is able to have contact site with most of the organelles themselves from endosome, Golgi apparatus, lysosome, peroxisome, lipid droplet, and also with plasma membranes.
So, since the plasma membrane has also been shown to to contribute to autophagosome formation, in the lab, we ask the question whether ER plasma membrane contact site can contribute to autophagosome biogenesis. So the project in the lab is under the supervision of a young PI, Etienne Morel, in collaboration with a postdoc, Anna Chiara Nascimbeni, and this is part of a collaboration with Francesca Giordano, in France, who is an expert in ER contact site and also with the lab of Maria Vaccaro in Buenos Aires, Argentina.
So briefly, what are ER plasma membrane contact sites? So the tethering of ER with the plasma membrane is dependent on protein name extended synaptotagmin, and there are three E-Syts, E-Syt 1, 2, and 3. And these contact sites are important for lipid transfer but also for calcium movement. So on the upper slide, we can observe the localization (in green) of E-Syt2 and the protein by confocal microscopy is only observed let’s say close to the substratum, meaning close to the plasma membrane, and we have no localization of E-Syts in the perinuclear area.
So, in our first series of experiments using HeLa cells, we ask the question whether in response to an autophagy stimulus there is a dynamic of this contact site. So for this, we transected HeLa cells with horseradish peroxidase targeted at the C-terminal with KDEL motif, which is an ER retention motif, and thereafter, by using a substrate for horseradish peroxidase we stain the endoplasmic reticulum. And at the EM level, we made quantification of ER plasma membrane contact site. And this was mainly in controlled cells and in cells after starvation.
And as I told you previously, starvation is an inducer of autophagy. And for example, you can observe from the Western blot in C that we have an accumulation of the protein LC3-II during a period of starvation, and this reflects, in fact, the increase in autophagosome formation; and at the same time, we observe the degradation of the protein p62. And this protein is a cargo for the autophagic pathway.
So when we starve cells, we observe an increase in the number of contact sites. And interestingly, we also observe an increase of the level of E-Syt2 and E-Syt3. We failed to observe any significant change in E-Syt1. And more importantly, we failed to observe any change in other protein from the endoplasmic reticulum such as calnexin or syntaxin 17. So it seems that in response to an autophagy stimulus, there is a dynamic of this contact site and increase. I have no time to show this, but we observed this phenomenon with other stimulus and starvation.
So the question was what is going on when we modulate this contact site? Are we able to modulate the autophagic response? So for this, we knocked down E-Syts in HeLa cells, and from Francesca’s work, we know that to block the formation of ER plasma membrane contact site we have to knock down the three E-Syts. So we knock down the three E-Syts. And thereafter we starve cells and we analyze autophagy. And you can observe that when we knock down E-Syts, we have an impairment in the accumulation of LC3-II. And on the contrary, you can observe an accumulation of the protein p62; meaning that there is an impairment in the autophagic pathway, and we have less degradation of an autophagic cargo. So we also observed, according to the decrease in the level of LC3-II, this protein reflect, in fact, the number of autophagosome we observe less LC3 protein when we knock down E-Syts when we transfected GFP-LC3 we have less LC3 puncta than in control cells after starvation. But interestingly, we observed the decrease in the number of puncta is mainly associated at the cell periphery and not too much in the perinuclear region.
So, the question was well, are we able to identify, in fact, protein engaging the autophagic process that contacts ER plasma membrane contact site? And we are able to detect in fact, PI3P, which is an important lipid in the biogenesis of autophagosome, as previously told. So just a technical point, we use, in fact, a GST-FYVE probe. So as you know, the five domains of protein is a domain that recognizes PI3P. And we use a GST-FYVE that we can use as a primary antibody. Another way to detect PI3P is to transfect cells with a GFP-FYVE probe. But, we and other reported that when you transfect cells with a GFP-FYVE you modulate, in fact, the autophagic pathway, you induce an artifact. So we prefer to use this approach to detect PI3P.
So doing this, when we starve cells, we wanted to detect, let’s say, the autophagosome pool of PI3P and for this you can observe in the lower panel that we have a co-distribution of PI3P with LC3 (in blue) and also with E-Syts (in red). Whereas, when we analyze the endosomal pool of PI3P and this PI3P, in fact, colocalize with EEA1 so the black arrow here, you can observe the colocalization of PI3P with EEA1. However, for this guy you failed to observe any colocalization or co-distribution with LC3 and E-Syts.
So, PI3P engaged in autophagy seems to be recruited at ER plasma membrane contact site and co-distributed ith E-Syts. So the question was are the proteins that interact with PI3P – meaning DFCP1 and WIPI2 – are also recruited at this contact site. And the answer is yes, on the upper panel, you can observe that DFCP1 (in green) co-distributed with LC3 (the autophagy marker) but also co-distributed with E-Syts as an ER plasma membrane marker. And in red, of course, also co-distributed with the endoplasmic reticulum stain with the SEC61beta antibody. The same holds true for WIPI2. And WIPI2, in fact, (in the lower panel) co-distributed with the ER and also co-distributed with E-Syts (in green).
So following on these results, we wanted to know whether we were able to detect autophagic structure close to the plasma membrane and also the ER. So, our first technique we used was to use confocal microscopy to detect LC3 and the endoplasmic reticulum close to the plasma membrane, and we observed this by confocal microscopy. You can observe LC3 dots and ER close to the substratum. However, to be sure that these events occur close to the substratum, we use another technique. We use a total internal reflection fluorescence or TIRF microscopy. And this technique allows us to detect events very close to the plasma membrane, in fact, in the range of 100 nanometer. And you can observe (on the left) so the ER is present close to the plasma membrane. And when we starve cells, we start to observe LC3 dots, and we have the quantification. So we are quite sure that autophagic structure are localized close to the plasma membrane.
So we went a step further and analyzed the process at the level of electron microscopy. An example is given on the upper panel (on the left side); we observe the plasma membrane, the cortical ER (in black), and autophagic structure pointed by the blue or red. Similarly, we performed some ImmunoGold with an anti-LC3 and also to detect E-Syts, and both protein protein are detected with the cortical ER endoplasma membrane. We also applied confocal microscopy and reconstitution and super resolution microscopy. And with super resolution microscopy, you can observe that LC3 and E-Syts are very close, in fact, and this protein are associated with the endoplasmic reticulum (in green).
So following on these results, we wanted to know how the system works at the molecular level. So, first, we analyzed two complexes that act very early in the autophagic pathway because we detected PI3P and PI3Ps quite early events during the autophagosome formation. So first we analyzed the ULK1 complex. And the ULK1 complex, as I told you, is the first complex that are during autophagosome formation in response to starvation. And, we observe this. We analyzed this complex in controlled cells both in basal condition under starvation – meaning with a plus – and we performed the same analysis in cells with a triple knockdown of E-Syts where we have an impairment, in fact, in the autophagic pathway in autophagosome formation.
So what we observed is that whatever the condition, there is no change in the stability of the protein of this complex. For example, we have no change in the level of ATG13, which is one component of the complex, and then 3 we have no change in the level of the ULK protein. And moreover, this complex seems to be functional because it is known that when you starve cells, you have a dephosphorylation of ULK1 and other sites sensitive to mTOR because mTOR is inhibited, autophagy is induced, and it is known that this induces dephosphorylation of ULK1 at the mTOR sensitive site. And we observed the same phenomenon in controlled cells or in cells with a triple knockdown of E-Syts.
So we turn our interest to the second complex – meaning the complex with Beclin 1, ATG14, and class III PI3 kinase, so the PI3 kinase complex I. And what we observed under the situation is that when we triple knockdown E-Syts, even in basal condition and also after starvation, we have less Beclin 1 and less ATG14. We have not analyzed yet, in detail, whether this is degradation, but we have some evidence for that. But in contrast, we failed to observe no difference in the level of Vps34.
So with this in mind, we analyzed more precisely, in fact, the level of PI3P engaged in autophagy and engaged in endocytosis in starved cells. So for this, we were able to discriminate between the endosomal pool of PI3P and the autophagosomal pool of PI3P, as previously shown by co-staining with EEA1 and LC3. So when we starve cells, we failed to observe any difference in the level of the endosomal pool of PI3P between controlled cells (the black bar) and triple knockdown E-Syts cells (the hash bar). However, when we focus on the autophagosomal pool of PI3P, we observe, in fact, a decrease in the accumulation of the autophagosomal pool of PI3P when we knock down E-Syts protein. So as I told you previously, we failed to observe any change in the level of Vps34. And according to the results, when we analyze the activity of Vps34 in vitro, we failed to observe any difference between controlled cells and triple knockdown cells. So something is going wrong with the complex I that produces PI3P in the autophagic pathway.
So, to make the conclusion, what we showed is that the autophagy machinery can be recruited at the ER plasma membrane contact site. Here we have another example with ATG16, for example, on the left side of the slide. We have E-Syts and ATG16 that are co-distributed. And in fact, we came with this with complementary experiment I have no time to show is that, in fact, the PI3 kinase complex I is known to interact with a transmembrane protein of the ER named VMP1 discovered by Maria Vaccaro in Buenos Aires. And this slide show, in fact, that VMP1 interact with Beclin 1. And on the other hand, in our work, we have been able to show that VMP1 also interact with E-Syt2. So, in fact, this protein makes a bridge between the the PI3 kinase complex I and ER plasma membrane contact site, and thereafter there is production of PI3P and maybe downstream formation of autophagosome. So here what is interesting is that in this work we show that ER plasma membrane contact site contribute to the formation of autophagosome. And as I previously told you, ER mitochondria contact site contributes to autophagosome formation. So what we made is a quantification of the contribution of this contact site in the formation of autophagosome during starvation. And we came to the conclusion that, in fact, about 30% of autophagosome in starved cells emanate from ER plasma membrane contact site. And also, 30% of autophagosome emanate from ER mitochondria contact site.
So that means that this contact site contributes to 60% of the autophagosome form in starvation, and the remaining 40% are associated with the endoplasmic reticulum, but we don’t know yet whether other contact sites can contribute to the formation of the autophagosome. So, at the end of this part, I hope I convinced you that ER contact site are important in the autophagic pathway to control autophagosome biogenesis, and also probably also to control autophagosome maturation as probably recently shown with ER lysosome contact site depending on the level of cholesterol in cells.
So now I would like to change the topic and go to the second topic of autophagy as a stress response and more precisely to focus on shear stress. Of course, you know – and we know – that autophagy is a stress response conserved to yeast to mammals and including also in plants but also in Dictyostelium, in Drosophila, in C. elegans, also in zebrafish. In most systems, autophagy is a stress response. We know well our organism responds to starvation to hypoxia to growth to induce autophagy and also in pathological situations this is an important topic in the cancer field that autophagy, in fact, is induced by therapy.
However, less is known about the role of mechanical crosstalk between autophagy and mechanical stress. So, we know in physiology, for ages, from the pioneering work in the autophagy field starting in the 50s, that hormones such as glucagon and insulin can regulate autophagy, for example, in the liver. Glucagon stimulates example, in the liver. Glucagon stimulates autophagy and insulin inhibits autophagy. And of course, nutrient – and for example, amino acid in the liver – are inhibitors of the autophagic pathway.
However, there is another player in physiology, which are mechanical stress. And mechanical stress are very important in physiology. For example, compression in bone and muscle and stretching, for example, in muscle during exercise, but also stretching is involved in the heart, in blood vessel, and lung tissue. And in the lab, we focus on shear stress. And shear stress is important in the circulatory system due to blood flow and and also in the kidney due to the urinary flow.
So, in the lab, we investigate irst shear stress in the kidney, and this project is under the supervision of an assistant professor, Nicolas Dupont, and in collaboration with PhD student (who is now a postdoc), Idil Orhon. And the work was a collaboration in our institute with a group of Fabiola Terzi, Thierry Capiod, and also we have on this project a collaboration with a group of Wolfgang Kuehn in Germany.
So briefly, as you know, in the kidney, cells lining the proximal tubule of the kidney are subject to shear stress induced by the urinary flow. And this can be reproduced in vitro by using the device shown on the right part of the slide. So we have a chamber; in this chamber, you can replace kidney epithelial cells repaired from the experiment with various cell line. And these cells are polarized. And of course, after you apply a flow using the device and we control the flow and we apply a laminar shear stress of 1 dyne/cm2, which reflect roughly the flow observed in vivo in the proximal tubule of the kidney.
So what we observe, in fact, is that when you apply shear stress to kidney cells, for example, MDCK cells, what you observe, in fact, is a time-dependent reduction in cell size. So, I should mention that this reduction in cell size is not an increase in cell proliferation, first point. And the second point, as you can observe on the upper panel with staining with E-cadherin, which is a marker for cell polarity, we retain the polarity when we induce shear stress.
Of course, next we wanted to analyze autophagy in the response to shear stress, so we used MDCK cells but also other cell lines, and these cells were stably transfected with GFP-LC3. And what we observed is that when we induce shear stress for several days, we have the black bar, an increase, in fact, in the level of LC3-II. And as a control, we use the same cells in static condition without flow, and we failed to observe significant ncrease in the level of LC3-II. I just mentioned on the upper panel that in red we have stained the primary cilium because kidney epithelial cells are ciliated and I will come back to this point a little bit later.
So of course, as you know, when you observe an accumulation of LC3 puncta, that could reflect easily an increase in the autophagic pathway from the formation of autophagosome to the degradation in the lysosome or a blockade in the autophagic pathway and an accumulation of autophagosome. To answer this question, we repeated this experiment by transfected cell by the so-called tandem probe. So this probe is an RFP-GFP-LC3 probe. So the basis for the use of this probe is that in autophagosome you have yellow dots because you have the green and the red fluorescence; whereas when you move to the autolysosome, you have only the red fluorescence, you have red dots because in the lysosomal compartment due to the acidic environment, there is a quenching of the green fluorescence.
So we used this probe. And as you can observe, the black bar in static and the black bar after one day, for example, of shear stress reflects the total number of dots. So we have an increase in the total number of dots. And, we also observe an increase in the number of autophagosome (the white bar); but more interestingly is that under shear stress, we observe an increase, in fact, (in the gray bar, meaning in the red dots, meaning in autolysosome. So that means that we have more autophagosome formed, and we have more autolysosome formed when we trigger shear stress.
So next, if autophagy is instrumental in regulating, in fact, the cell volume in response to shear stress, to approach this we use a genetic approach by knocking down Atg5 or Atg16 two proteins involving an autophagosome formation by shRNA approach. And we use the stem cells and the expression of the shRNA can be induced when we add IPTG to the cells. And we use also a pharmacological approach, for example, by using 3-methyladenine, an inhibitor of autophagosome formation.
So after four days of shear stress, what we observed when we knock down Atg5, Atg16, or when we add 3-methyladenine is, of course, as expected, a reduction in LC3 dots and a reduction in autophagy. But interestingly, when we analyze he cell size or cell volume, what we observe is, in fact, in response to an impairment of autophagy, cells are no longer to correctly decrease the cell size.
So the question next was to understand how shear stress can be integrated at the cell surface to control the autophagic pathway and downstream to control cell size. In fact, it has been shown some years ago by Kuehn and colleagues that the primary cilium present at the apical side of kidney epithelial cells control cell size. Moreover, in a collaborative effort with the lab of Ana Maria Cuervo, in New York, and with a postdoc Olatz Pampliega, we have been able to show some years that, in fact, there is a cross talk between autophagy and the primary cilium. And in response to cilium, in fact, the primary cilium is able to trigger the autophagic pathway.
So briefly, the primary cilium is a macrotubular based structure present at the surface of many cell types. And in the case of kidney epithelial cells, present at the apical site. And, this structure is very important to coordinate many signaling pathway, for for example, the Hedgehog pathway, the PDGF pathway, the Wnt pathway. But, the primary cilium is not only a chemical sensor is also a mechanical sensor, as was shown by Kuehn and colleague in the kidney that, in fact, the primary cilium is important to integrate shear stress induced by the urinary flow.
So, we ask the question whether the primary cilium can be upstream of the autophagic pathway to control cell size. So for this in vitro, we use different strategy to block ciliogenesis. And one of the strategies was to knock down by shRNA approach he protein Kif3a. And Kif3a in fact, is a protein that is part of kinesin-2 and kinesin-2 is a motor that moves along macrotubule to the apex, and when you knock down you block the activity of kinesin-2. And when you knock down Kif3a, it’s known that you block ciliogenesis.
So, what is going on with autophagy when we knock down ciliogenesis, we observe a reduction in autophagy in response to shear stress. And again, interestingly, when we analyze the cell size and volume we observe, in fact, that cells with a knock down of Kif3a. So the black bar after four days, the cells are less prone and controlled cells; the white bar to controlled cell size and cell volume.
So we move from this to in vivo because Wolfgang Kuehn in his lab the conditional mouse model was in validation of Kif3a in the kidney. So we translate our in vitro results in vivo. And what we observe, in fact, in the proximal tubule in the kidney is, of course, in the mutant animal less primary cilium (shown in red on the upper panel). And we also observe less LC3 dots (shown in green in the upper panel). And here, again, in vivo when we block ciliogenesis and downstream we block autophagy, what we observe is an impairment in cell size regulation. You can observe that cells in mutant animal are bigger than in controlled cells. So autophagy seems to play an important role in controlling cell size in the proximal tubule of the kidney and this event is very important to control the geometry, in fact, of the proximal tubule.
So I will not emphasize, but just mention that we have been able to analyze the signaling pathway coming down from the primary cilium in response to shear stress to trigger autophagy and to trigger cell regulation. We observe, in fact, I would say two waves of autophagy. The first wave is dependent on the protein present in the primary cilium called polycystin-2, and this is a calcium channel. And we observe an early autophagic response before the cell size regulation. But, this polycystin-2 dependent autophagy does not seem in our system to be engaged in cell volume regulation. We don’t know yet what is the function of this wave of autophagy.
But we observe a second wave, I should say, that triggers the activation of a kinase named LKB1. LKB1 activates the kinase AMP kinase. And you know probably, AMP kinase blocks inhibit autophagy. And this is upstream, and the inhibition of autophagy trigger autophagy and induce the cell size and cell volume regulation.
So, to close up this talk, I would like just to show that we extend these results to another model to the endothelium. And the work, in fact, was performed in Chantel Boulanger and Pierre-Emmanuel Rautou, in Paris, at the Cardiovascular Institute. And in the lab, we collaborate with Nicolas Dupont to this project.
So interestingly, in the aorta for example, it is known that endothelial cells are subject to high shear stress in the descending part of aorta and low shear stress at the level of the aortic arch in the bifurcation. So we analyze the level of autophagy in response to high shear stress and in response to low shear stress due to the blood flow.
So first, in vivo using mouse model analyzing the autophagy in zone of high shear stress and in zone of low shear stress, we observe more, in fact, LC3 dots in the zone of high shear stress and in the zone of high shear stress and in the zone of low shear stress. And this was also reproduced in vitro by using an endothelial cell line, HUVECs cells. And in fact, you can observe that under low shear stress – meaning 2 dyn/cm2 – and quite high shear stress – 20 dyn/cm2 – which roughly accommodate to the in vivo parameter, we observe at EM more autophagic structure in high shear stress than in low shear stress. And accordingly, when we analyze the level of LC3-II, we observe more accumulation in a time-dependent manner of LC3-II in the presence of high shear stress than in the presence of low shear stress.
However, as previously told, when we observe an accumulation of LC3-II, we have to answer the question whether or not there is an increase in the is an increase in the autophagic pathway or a blockade in the autophagic pathway. So for this, we use, again, in vitro in HUVEC cells tandem probe. And here again, we observe much more red dots (meaning autolysosome in the zone f high shear stress than in the zone of low shear stress). And moreover, we analyze the event of fusion between autophagosome and autolysosome by performing double staining with an anti-LC3 and an anti-LAMP-2; LAMP-2 is a marker of the lysosomal compartment we have much more event of fusion of autophagosome with lysosome in the zone of high shear stress than in the zone of low shear stress.
And in fact, the same holds true in vivo. And, to analyze the autophagic flux in vivo, we injected mice with chloroquine. And chloroquine is a lysosomal inhibitor, it’s a lysosomotrope, it accumulate in the lysosome and quench the lysosomal pH and block the degradation of LC3-II. And you can observe that in the zone of high shear stress when we injected chloroquine, in fact, we have more accumulation of LC3-II; meaning that we block, in fact, the autophagic flux. Whereas, there is not too much change in the zone of low shear stress in vivo.
So, these results were also translated into pathophysiology because by using, in fact, human carotid atherosclerotic plaque plaque obtained after endarterectomy, what is known, in fact, is that upstream of the plaque there is a high shear stress. And downstream of the plaque, there is a low shear stress. And here again, interestingly, in this setting, we observed the same difference; meaning more autophagy, more accumulation of LC3 dots in the zone of high shear stress, upstream of the plaque; than in the zone of low shear stress downstream of the plaque symbolized on the graph by the black square.
So what we did next, what Chantel and Pierre-Emmanuel did, in fact, was to analyze the process in vivo in a model of atherosclerosis. A model widely used in the atherosclerosis field is apolipoprotein E-/- mice. And it is known that these mice accumulate lipids in the endothelium, the aorta, and mainly at the level of the aortic cross where there is a low shear stress, as you can observe on the left aorta. So, what was done next was to cross back this animal with an autophagy deficient model with a conditional invalidation of Atg5 in endothelial cells. And we can observe on the very right part of the slide is now you can observe fat
deposit in the descending part of the aorta. And on the left, we have the quantification, meaning we have under high shear stress an increase in the fat deposit in animals with invalidation of Atg5 with reduction in the autophagic process. Whereas with not too much change, in fact, in the zone of low shear stress.
So, [cuts out] underchange, in fact, in the zone of low shear stress.
So, [cuts out] under under high shear stress and this autophagy in endothelial cells is that they are protective. And here again, I did not detail the results; we observed that the induction of autophagy is dependent on the activation of AMP kinase and on the inhibition of mTOR. Whereas in the [cuts out] stress. When you block autophagy in the zone of high shear stress, you have a low level of autophagy. And we report in this study, in fact, that this low level [cuts out] apoptosis, senescence, the release of inflammatory cytokine and also impair the endothelial cell alignment, which is a hallmark, in fact, of atherosclerosis.
So this last slide summarizes the results on this part. We analyze the effect of shear stress on autophagy in the kidney. We can conclude that the urinary flow is [cuts out] epithelial cells and the pathway is dependent on activation of AMP kinase, inhibition of mTOR, and the function is to regulate cell size. The shear stress is integrated of the cell surface by the primary cilium to trigger autophagy. In the endothelium, the blood flow is able to trigger autophagy here again, by activating AMP kinase/blocking mTOR, and this autophagy is atheroprotective. However, in this setting, we don’t know yet what is cell surface transducer to trigger autophagy.
So very lastly, I would like to conclude with some open questions in the field of autophagy related to this talk and beyond. So from very basic science, a question still open is what is the origin of autophagosomal membrane; what are the lipid sources? Also, what is the interaction between autophagy and intracellular trafficking (endocytosis, exocytosis)? What so-called non-canonical autophagy can tell us about to understand better the role of ATG proteins? With non-canonical autophagy – I mentioned the role of ATG protein in the secretion of proteins or secretion of small molecules such as ATP or also the role of ATG protein in phagocytosis through LC3-associated phagocytosis – what is the function of this non-canonical pathway in physiology and pathophysiology?
Of course, still big question on should be better defined is a crosstalk between autophagy and cell death but also autophagy and cell proliferation. I would mention we known for ages that autophagy is a cell autonomous function process. But now we know that autophagy also non-cell autonomous function. And this has been clearly shown in cancer. What is the role of non-cell autonomous function in physiology? And lastly, what is the role of autophagy in integrative physiology, for example, the relation between autophagy and circadian clock? And something what is emerging, what is fascinating is the role of autophagy in longevity. Now there is a large panel of evidence to state that autophagy is an anti-aging mechanism.
So very lastly, I would like not to detail this slide, but just to thank all the people of my team – those who I presented the work today but also other people. Thanks to our collaboration and leave you with these two views of phagophore – one from artist autophagy let’s say; and the second one is a natural one, is a very nice beach in the south part of Greece. So thank you very much for your attention. Thank you for that very informative presentation, Dr. Codogno. We’ve already received a great number of questions. But before we begin the panel discussion, I’d like to remind attendees about the resources tab located on the upper right-hand corner of the screen, which contains literature and posters on autophagy from Novus Biologicals and Tocris Bioscience. This includes two pieces that were co-authored by Dr. Codogno. And also, this presentation has been recorded and will be available for viewing following the webinar. So if you still have any questions for Dr. Codogno, you can submit them using the Ask A Question Box, which is located just below the presentation screen.
So with that, the first question I’d like to start with is what is the contribution of mitochondria to ERPM interaction? The contribution of mitochondria to ER plasma membrane contact site, we don’t know yet. Just what I can say is that we observe – by different techniques – that some mitochondria can be closely localized near ER plasma membrane contact site. But at the moment, we don’t know whether this dialog between these three partners – meaning plasma membrane, ER, mitochondria – can can work together at the same place. That’s an interesting question because we observe such structure, but we have no answer to that at the moment. Another question – in the normal heart, [cuts out] what could be the percentage of cargos from ER versus [cuts out] plasma membrane? Please repeat the question. heart, so it’s a mechanically stretching organ, what would be the percentage of cargos from ER versus mitochondria or plasma membrane? You mean the percentage of cargo in…well, the percentage of cargo in autophagosome that’s the question? Yes. I would say [cuts out] autophagy and this can, of course, if you treat your mitophagy you will [cuts out] also organelle. But, [cuts out] come back to a very old study by Per O. Seglen and colleagues. When you starve cells, in fact, it made the quantification of the organelle it can identify inside the autophagosome; and of course, it made the ratio with organelle outside of autophagosome. And he came down to the conclusion that the ratio is the same. So that was one of the evidence for the bulk aspect of autophagy. Did I answer your question? Yes, I think [cuts out] trouble hearing. Is anyone else having trouble hearing? Can you send a message if that’s the case? Okay, well let’s try another question; maybe it’s just a few people that are having difficulty. I know you had mentioned earlier that you were looking at other forms of autophagy induction, and so someone asked have you tried using inhibitors of mTOR, like rapamycin. And if yes, do they mimic the effects of starvation that you observed on the autophagosome formation? In the first part of the talk, you mean if we use other inhibitors or stimulators? Yes, we repeated this experiment by using taurine, for example, which is an inhibitor of mTOR. We also used mechanical stress to trigger autophagy. And in all these settings, what we observe is an increase in ER plasma membrane contact sites. So, it’s not unique to starvation. Now, one of the questions we want to answer is, is there any selectivity from the localization of ER plasma membrane contact site to sequester some cargo, in fact? I think people are still having some difficulty hearing, so we’ll do the best to answer a couple more questions, and then I’ll just let everyone know that they will have access to this recording later on to hear the remaining questions. Can you see some relation between BCL2 levels in autophagosome formation? We have not investigated this directly. But, of course, it is known from the literature from Beth Levine’s lab that BCL2 can interact with Beclin 1 and block autophagy. In the work I presented today, we have not investigated this aspect. We don’t know whether BCL2 can be a regulator of shear stress-dependent autophagy. We don’t know, we have not investigated this point, no. I’ll ask a couple more questions. One is how does autophagy affect crosstalk between cells in terms of adhesion molecules, for example? For example, it has been clearly shown in cancer that, for example, cells from the microenvironment, autophagy can produce via the degradation of protein some amino acids And these amino acids can be used by the tumors to fuel tumor cells. And this is one of the crosstalk. Meaning the product of degradation in one cell type can be used by another cell type. So this is the notion of non-cell autonomous function of autophagy. Okay, I’m going to end it with one more question. And that is, is the role of cell size regulation by autophagy a general phenomenon? Yes, this has been reported by other group, in fact, one group reported that the metabolism is very important to ontrol cell size. And for example, if you go to the literature, when you inhibit autophagy and you look carefully, for example, in Drosophila, work by Eric Baehrecke clearly showed when you block autophagy you have an increase in cell size. So there is a relationship between autophagy and cell size. Yes, this seems to be acquired general phenomenon. And of course, now we wanted to understand how autophagy is able to control cell size. We’re We’re investigating several possibilities on that. Okay, well, I I think we’ll end it with that. Hopefully everyone was able to hear the questions. So that concludes our Q&A session. I’d like, again, to thank the audience for attending and Dr. Codogno for his engaging presentation. So be on the lookout for an email from Novus with a link to the recorded webinar, and this concludes our webinar. Have a great day or evening, everyone. Thank you so much.