Metabolic Adaptations: The Physiology of Cold Acclimatization and Exercise in Rodent Models

Sarah: Good morning, good afternoon, and good evening, everyone, and welcome to our webinar titled “Metabolic Adaptations, the Physiology of Cold Acclimatization and Exercise in Rodent Models”. This webinar has been sponsored by Sable Systems International, so a big thank you to them for helping to make this event possible. Joining us today, we’re very fortunate to have Dr. Jonas Treebak, an associate professor at the University of Copenhagen, and Dr. Alexander Bartelt, a Professor of Cardiovascular Metabolism at the Ludwig Maximilians University in Munich. Their presentations will discuss the physiological mechanisms underlying metabolic adaptations to exercise training and cold exposure. And so with that, I’m very pleased to welcome Dr. Jonas Treebak. Jonas, thanks so much for joining us today and the floor is yours whenever you’re ready.

Dr. Treebak: Okay, thank you very much. So, I thought I wanted to take this opportunity to talk about some of the recent work that we’ve done, detailing the role of NAD and NAD in skeleton muscle, and especially in response to exercise and exercise training. So, we know, or well, we work from the hypothesis that NAD is important for the ability to adapt to metabolic stress. And we all know NAD as this cofactor responsible for more than 400 different chemical reactions in the cell. It’s involved in redox reactions, being oxidized and reduced continuously, but it’s also involved in regulating circadian metabolism, epigenetic regulation and DNA repair, among other things. So, levels of NAD in the cell is balanced between consuming and producing processes. And there is this thinking in the field that with aging and pathological conditions, this balance can shift. And that with stimulation of NAD biosynthesis, with NAD precursors, for instance, or by inhibiting some of the consuming enzymes, we can regain the balance, maybe even obtain higher levels of NAD in the cell, which we think is beneficial.

So, our work has primarily been trying to figure out what the role of NAMPT is. So, the biosynthetic side of the scale. And NAMPT is this enzyme in the NAD salvage pathway. It’s a rate limiting enzyme in converting glutinamide to NAD. There also the tryptophan pathway, the De Novo pathway, whereby NAD is generated from tryptophan, the Preiss-Handler pathway where nicotinic acid is the precursor for NAD, and common for these two pathways is that they rely on the NAD synthase enzyme that is the last step in converting in AAD to NAD. NAD synthase is a tissue specifically or cell specifically expressed, so some tissues express this and some tissues don’t. Skeletal muscle is one of those cell types that doesn’t express NAD synthase. So that’s why we thought targeting NAMPT would be an interesting approach in order to perturb NAD metabolism in this cell type.

So, we generated muscle-specific NAMPT knockout mice, and these mice, they have very significantly reduced NAD level. They also have reduced NADH levels, and they die prematurely. They have, starting from four weeks of age, at 12 weeks of age, we have only 30 to 35% of these animals left. So, why do they die? We think this has to do with impaired mitochondrial function and also impaired integrity of the plasma membrane. This leads to disturbances in calcium homeostasis that in turn induces the MPTP response which is the mitochondrial permeability transition pore response that basically tries to buffer excess calcium in the cytosol. And if you have very high calcium levels in the cell, then you may have the mitochondria swelling so much that they burst. And if this occurs, the cell becomes necrotic and the cell would die. And this leads to muscle dysfunction and dystrophy. This is why we think that these mice, they die.

We have also a tamoxifen inducible model in the lab, and we have these mice, these mice are viable. They have the same reduction in NAD levels, so 85 to 90% reduced NAD levels, but they live fine, they have no visible phenotypes. We have mice that are more than two years of age. So, one question is why one model is viable and the other is not. We think that this may have to do with the very remarkable remodeling of the tissue that goes on in early life of the mouse. So, muscle fiber area, as you can see, has a fourfold increase in the time span between two weeks of age and six weeks of age in the wild-type mice, if you look at the black bars here. And this may cause stress of the membrane that needs to be repaired. And we can actually see in the constitutive NAMPT knockout mice at six weeks of age, we have staining of IgG inside the cell, indicating that the plasma membrane is actually…the integrity is lost in many of the cells.

So why is integrity lost? So, in order to keep membrane integrity, you need to have a solid base, you need to connect the intercellular cytoskeleton with the external matrix. And this is the primary anchor, if you will, is the laminin receptors, and these are the integrins and the distal glycan illustrated here in this figure. They bind laminin and they are ADP-robosylated, so integrin alpha-7, for instance, is ADP-robosylated in muscle, and we also have that this ADP-robosylation actually increases the binding of the integrin two laminin. And ADP rebosylation is an NAD requiring process. So you can imagine if in a wild type setting, where you have excess of NAD in the cell, if you have a small leak, a small break in the membrane, you may have NAD leaking out of the cell, then ADP ribosyl transferases will then ADP rebosylate the integrin alpha-7, for instance, and then make sure that the membrane stays intact, but in the knockout situation where you have very little NAD, this may not occur and you have calcium influx that may end up killing the cell because of the MPTP response. So, this is something that we are still working on to study in more detail.

But the inducible model for us is, we thought this was the perfect model to study because the mice are viable, they have no visible phenotype, and they have very low levels of NAD. So how does or do muscle from this inducible skeletal muscle in other mice, do those cells adapt to metabolic stress? So, we isolated mitochondria from these cells, and we could see that while whole cell energy levels were reduced by around 90%, mitochondrial levels of energy was only reduced by 50%. When we stimulate with TCA cycle intermediates, substrates, and also with ADP, we can see a large increase in NADH, which is, I guess, something that we would expect to happen. And this also happens in the knockout mice, but to a lower extent. So, this may suggest, maybe because they come from a lower level, or it may suggest that the TCA cycle is less efficient in generating NADH.

However, if we take out muscle and isolate single intact fibers and expose those to the major oxygen consumption in the ouroboros system, we can actually see that there’s absolutely no difference between knockout and wild-type fibers, either at sub-maximal ADP doses or with succinates. So, coupled respiration and uncoupled respiration is similar in these conditions. So, these are intact fibres, what about mitochondria, isolated mitochondria? So, in order to stress further the cell, we wanted to do an experiment where we exercise or acutely exercise these animals. Normally, when we do this type of exercise protocol, we have three days of acclimatization before they rest for two days, and then we do sort of a graded exercise protocol where they end up at a submaximal running intensity. So, we did this with the inducible NAMPT knockout mice and we could see that both, well with submaximal doses of ADP we can actually see a genotype effect, no effects of running, but when we stimulated with higher doses of ADP, sub-maximal dose of ADP, and also with succinate to maximally induce complex II activity, there was no difference between genotypes, no difference with running. So, they may have…the sensitivity to ADP may be different between knockout and wild-type mice, but maximal respiratory capacity is not affected by a lack of length.

So, from the same mitochondria, we also measured ADP synthesis rate. That was a nice induction with exercise, and we saw a corresponding increase in membrane potential, which of course is necessary in order to increase ATP synthesis rate. So, exercise increases ATP synthesis rate and also membrane potential, but this is independent of NAMPT. We could then, from the oxygen consumption rates and the ATP synthesis rate, we calculate the P-O ratio, which tells you about, is a measure of efficiency in a way for the mitochondria to produce ATP and we can see that there’s actually no difference between runners and sedentary mice and no other differences between genotypes. So, mitochondria seems to be doing okay with around 50% less NAD.

So next we asked whether training-induced adaptations would be affected in the knockout mice. So, in order to do this, we had mice running in running wheels for four weeks inside indirect calorimetry cages and we did a maximal exercise capacity test at the end of the four-week period. And we had sedentary and trained animals. So, we could see from just measuring running distance, there were no differences in either dark or light cycle in terms of the distance that the mice covered. And this is of course nice in order to be able to compare the two genotypes. We also measured all the standard measures that you can get from these indirect calorimetry cages. We have oxygen consumption, and you can see that there is kind of an adaptation period where the animals adapt to the wheels and this takes about a week before they reach a sort of a steady state. And what was actually remarkable with these data was that the levels of oxygen consumption that we observed with the running wheels is actually very similar to what we obtain when we do a maximal running test on a treadmill. So, with the running wheels, these mice are actually performing very high intensity exercise. And I think we can safely call this exercise training, at least maybe even HIIT training. There was no difference when we calculated this per day or per phase between sedentary and trained, between knockout and wild-type mice, of course there were differences between sedentary and trained.

RER is the measure of substrate utilization and it’s basically the ratio between CO2 produced and O2 consumed. And again, we can see we have this adaptation period and the steady state period. And also, it’s quite interesting to notice that while the mice are actually running at a very high intensity, RER is actually lower in the exercised animals compared to the sedentary animals. And this, of course, may be due to the differences in how they handle lipids in the substrates, so lipid oxidation may be increased, and lipid storage may be decreased in the exercising animals. So, this may explain this, but something that we could look further into. There was no difference between genotypes either in the sedentary or trained state.

So next we wanted to know whether the efficiency of the animals, the running efficiency was different. In order to do this, we simply plotted the distance run versus the energy expended for that particular bout of exercise. These are actually plotted values for all four weeks, so you can see that this is just a representative plot, but there’s a very nice linear correlation between distance and energy expenditure, so yeah, so the more the animal, the longer the animal run, the more energy it expends, or it uses. So, these are just to show you different examples of this. We took the slopes of these curves we generated for each animal, and if you do that you can actually see how there’s absolutely no difference between genotypes in the efficiency, so how many joules per meter they consume. If you look at this over the four-week period we can see that there’s a little drop, there’s no difference between genotypes, but there is a drop in energy expenditure indicating that these trained mice are actually becoming more efficient when they run in the wheels. So, I guess that is also expected.

So, as I said we also tested the maximal exercise capacity at the end of the training period, and we did this using a graded exercise performance test. We had 10 minutes in the chambers to measure basal oxygen consumption then a graded protocol, where the intensity of the running is increased, and then at the end, after a specific criteria for terminating the experiment, they had 10 minutes of recovery measurements as well. So, these are the different graphs from the baseline, the running and the recovery and you can see that after being inserted into the treadmill you can there’s an increase oxygen consumption, but this stabilizes and then in the…during the running, we can see an increase in oxygen consumption, and then in the recovery this is very rapidly going to baseline again. But one thing that I think is very striking here is that that we actually obtain the maximum or the peak oxygen consumption rate at around four minutes of exercise. And even if we increase exercise intensity, we don’t see a corresponding increase in VO2. So, this is something that I think we need to work more on and see whether this is also reproducible with other models. But in any case, what we found was that with training there was a small increase in VO2 peak, independent of genotype, but the maximal exercise capacity was reduced in the knockout mice. We didn’t have a significant interaction here, but I think we will repeat the experiment and we may be able to see a significant difference here between how they respond to exercise.

All right, so in order to detail even further how these mice responded to trace size training, we took another cohort of trained animals, and we took the quadriceps muscle, trypsinized it and subjected it to LC-MS. And so, we did a proteomic analysis and then and we identified 1,742 unique proteins. Of these, 564 were significantly regulated. If we look at the gene ontology analysis of those significantly regulated proteins in the wild types and the knockouts, there was a striking similarity between the cellular compartments where these significantly-regulated proteins belong. I think actually top seven or top eight of these terms are completely identical between the genotypes and they all relate to the mitochondria.

So, looking a little bit further into this, we could see that there are 205 commonly regulated proteins, 221 proteins are significantly regulated in the knockouts only, and 138 are regulated only in the wild type. But if we, for instance, only look at the upregulated proteins in the knockout mice, we can actually see that, I think you can appreciate here that nothing really goes on. And we have the, in the green, we have the wild-type sedentary, then the wild-type trained, then the knockout sedentary and the knockout trained. So, nothing goes on in the wild-type mice, but we have a clear reduction in the baseline level of many of these proteins. And you can see that we kind of have a normalization of the protein levels if we compare the wild-type trained to the knockout trained. So, what might be occurring here is that the knockout mice are able to normalize proteins within the mitochondria and thereby obtain the ability of the mitochondria to respire normally. But while they are able to exercise to a similar extent in the sort of untrained state is maybe a little bit weird and something that we are looking into more in terms of how mitochondrial super-complexes are regulated. So, this is something that is working in progress.

All right, so this is my final slide except for the acknowledgements, but the take-home messages, NAMPT in muscle is crucially important to maintain NAD levels. Ablation of NAMPT muscle during embryonic development results in premature death, probably due to inability to maintain membrane integrity. Skeletal muscle is resilient to ablation of NAMPT in adult mice, so respiratory capacity is the same, mitochondrial respiratory capacity and bioenergetics acute exercise is similar, running efficiency is the same when they train, and improvements in the VO2-peak is actually also similar with training. Maximal exercise capacity after training appears to be reduced in the knockout mice and what may be occurring actually is that the exercise training in the knockout mice normalizes the abundance of these mitochondrial proteins. But yeah, I guess you are wondering the same as this mouse to the left, do I really need all that energy in my muscle? It may be that there is a lower threshold for how much energy is needed in order for the cell to survive, but I think it’s actually quite low, and adaptation occurs irrespective of these lower levels of energy, which I think is quite remarkable.

So, with that, I just want to thank all the collaborators, I don’t have time to mention everyone. I will say that Sabina Chupanava was instrumental in generating the inducible NAMPT knockout mice and all the studies that are shown was run by her. So, I’ll be able to take any questions and thank you for your attention.

Sarah: Okay, so with that, I’m very pleased to welcome Dr. Alexander Bartelt. Alex, thank you so much for joining us today and the floor is yours whenever you’re ready.

Dr. Bartelt: Yeah. Thank you everyone for joining this webinar today, and thanks so much for the invitation to share some of our research and some of the fundamental concepts we’re interested in understanding metabolism. I’m a professor at the Ludwig Maximilians University in Munich at the Institute for Cardiovascular Prevention and like with every good presentation, I would first like to thank the people that have contributed to the work I will be presenting today, which are the members of our lab and the people that do fund our research.

So, the topic of today’s presentation is cold air acclimatization and as you are all very well aware cold is a very strong environmental stimulus that particularly for homeotherm mammals represents an environmental challenge to which we have to adapt, which means that whenever the temperature outside is lower than our body temperature, we have to actively produce heat in order to maintain our body temperature. And this process can be depicted in this simple diagram here, which is particularly shown for mice now. And you can see here that in the thermo-neutral zone, the basal metabolic rate, the BMR, is enough to produce enough heat to maintain body temperature. Now, when a mouse is exposed to temperatures outside of this thermo-neutral zone, the so-called lower critical temperature, then it needs to recruit extra metabolism in order to produce extra heat to maintain body temperature. And that obviously is also associated with an increase in oxygen consumption. And this is actually also what most people study when they talk about cold adaptation or cold acclimatization. However, it’s very important to point out that cold goes along with a lot of physiological changes, which is predominantly driven by increased sympathetic tone. And as I said, this means there’s increased oxygen consumption, but also, for example, a major hallmark here is increased blood pressure and particularly to serve the energy demands, increased food intake.

So, thermogenesis is a process that has multiple components and one of these is non-shivering thermogenesis. And this is a process that particularly drives increases in metabolic rate and one outside temperature drop corresponds to about 8% to 10% increase in energy expenditure. And in most small rodents, this is actually an effect that is largely mediated by the activity of brown adipose tissue. And this is an electron scanning microscopy picture where you can see the more raspberry-like looking adipocytes here with the multilocular lipid droplet structure. And as we know, the tissue is brown because it has so many mitochondria and so much iron. Now, what are unique features of these cells? These thermogenic adipocytes have a couple of properties that make them distinct from any other cell in our body. And this is particularly mechanisms that lead to thermogenesis, which are UCP1 and other futile cycling mechanisms. And the particular combination of high mitochondrial lipid content makes these cells little heater devices. And the end, it’s not also only about the cells, also the tissue has particular properties. So classical brown adipose tissue is highly innervated, which enables a direct line from the brain that, in the end, is the main controller of the response to cold, and it can immediately, through sympathetic activation, lead to lipolysis and thermogenesis. And it’s also highly vascularized, so the oxygen can be delivered, and also the heat that is then released by this tissue can be transported into the periphery of the body.

So, cold acclimatization has different phases. And this is depicted here in this classical picture by Cannon and Nedergaard, where you can see that true cold adaptation takes weeks. And on the left-hand side, you see a mouse that is probably housed at warm temperatures, something like thermoneutrality. And in the upper panel, you can see here, when you inject these mice with norepinephrine, you will only get a very small increase in metabolism. And that is because these mice are adapted to warm temperature, and they haven’t recruited a lot of thermogenic capacity. So once these mice are put into colder temperatures, then there’s an immediate increase in metabolism, which is initially derived from shivering thermogenesis. And then slowly, this is being replaced by brown fat-derived thermogenesis. And this is in a regular mouse. And this goes hand-in-hand with recruitment of UCP1 expression and remodeling of the tissue. And as you can see here, a significant recruitment of brown fat in mice, you can see after something like a week or so.

I mentioned there is a significant remodeling of the tissue, and this is also a classical picture where you can see that in the beginning, in an unstimulated condition, there are a couple of brown adipocytes, fewer pre-adipocytes and precursor cells, but with enhanced and prolonged adrenergic stimulation, there’s an increased proliferation and increased production of new cells. But also the ones that are already there become more powerful in their thermogenic capacity in a way that they make more mitochondria, more UCP1, and then in the end this manifests and this histological difference that you can see here on the right hand side, where you just have more cells and more power. In addition to that, there’s also a process called adipose tissue browning. So, you see here on the top panel, the white and the classical brown fat, but there’s also the intermediate phenotype where in the white adipose tissue, you find islands of thermogenic fat cells. And this is a classical process that can also be observed when mice are adapted to cold.

How is this working? Well, you have again a white fat that undergoes browning either through cold or selective beta-3 agonism, and then the reversal is also possible. For example, when the temperature rises again when the cold stimulus is taken away or particularly when mice are fed high-fat diets, and then this process is reversed and leads to adipose tissue whitening. And we know both are possible. There are precursors that are being recruited to become inducible beige adipocytes, but also there’s a trans-differentiation of existing, so probably masked inducible fat cells that look like white fat cells more likely. And this is a process that is particularly relevant for humans, as you might all know, humans do not have a lot of brown fat. And this is particularly because of the conditions under which we live and our thermoregulatory processes are very different. And if you really want to study it, mouse brown fat can also be humanized in a way by, for example, housing mice at higher temperatures and feeding them a high-fat diet. And then morphologically and functionally, mouse brown fat becomes more similar to human brown fat. This is a process that cannot only be observed macroscopically, but there’s also very strong remodeling of metabolism. For example, here in a recent publication of ours, we showed that the remodeling of lipid and oxidative metabolism is actually wired. And this is a study here where we’ve shown that CHREBP beta can drive the adaptation to warmer temperatures through de novo lipogenesis, and for example, if the de novo lipogenesis is not working very well, then these tissue retain their thermogenic profile. And on the lower bottom, you can see here that mice that do not have CHREBP beta in their brown fat have higher levels of UCP1 and are resistant to the involution of brown adipose tissue. And in addition to that, glucose and lipid metabolism are also wired to mTORC2, and this is a recent publication from the Girton lab, where they showed that glucose can cycle both into the CCA cycle, generate intermediates like acetyl-CoA, and this can also feedback on the gene regulatory level, showing again that both anabolic and catabolic pathways are linked together.

So, this all emphasizes that there’s very strong remodeling of metabolism and brown fat during cold acclimatization. And our particular interest here is quality control. How can cells ensure that during these complicated and weeks-long processes, the cells do retain their metabolic capacity and with very low error rates? And a process that is very important here is called proteostasis. It describes the healthy life cycle of a protein from translation to folding to degradation. And the three main pillars of proteostasis are the unfolded protein response, the ubiquitin proteasome system, and autophagy. And all these three pillars have been demonstrated to play very important roles, not only in metabolism, but also to a certain degree in adipocytes. And I want to now focus more on the ERA pathway, which is shown here in the middle column, the ER associated protein degradation, where you can see that there’s a transcription factor called Nfe2L1, which is proteolytically activated by protease called DEI2. And this transcription factor then can stimulate proteasomal breakdown of obsolete, misfolded and unwanted proteins. And this is a key process actually before preventing the accumulation of waste. It’s a little bit like you’re throwing a big party and then the trash needs to be picked up afterwards. And this is the responsibility of the ERA pathway during cold acclimatization. So, if this doesn’t work, then the brown fat actually turns white in the cold. So, it’s adipose tissue whitening that is induced by cold somewhat paradoxically, and this is also seen in beige adipocytes. Also, adipose tissue browning does not occur. It’s important to point out that this is an adaptive process. So, in the absence of any cold, these mice do not display phenotype because stem neutrality is actually a state of natural Nfe2L1 deficiency. And if there is no cold, these mice do not really have a phenotype.

What’s always important to see is your adipose tissue actually active, and we’ve recently established a method by which we can measure the metabolic activity of brown adipose tissue by using semiconductor quantum dots. And these hydrophobic particles can be embedded into the core of recombinant lipoproteins. And in particular here, we use short wave infrared sphere quantum dots. And these particles that emit in a range where actually biological tissue becomes translucent. And we use this tool, for example, here in mice that were lacking Nfe2L1 in brown fat specifically. And you can see here on the left-hand side and the intrascapular region, the uptake of the tracer, the lipoproteins. You’re basically looking into the mouse here and you can see here in the knockout mice, there was really not much happening. And also when you look into the brown fat here under visible life, both tissues are there, but only the wild type brown fat is able to accumulate and metabolize large amounts of lipoproteins. So, these mice do not have any thermogenic capacity.

And now the last part of my talk, I want to spend talking about some fundamental misconceptions that we often encounter when we review papers and studies, and you’d expect that in the absence of any brown fat activity, mice should probably become more susceptible to obesity. But just at regular room temperature, somewhat surprisingly, even on a high-fat diet, we did not see any of this. The mice gained similar weights, became as obese, and when you actually measure their energy expenditure by indirect calorimetry, we also found that there were no baseline differences in these animals. And this is really surprising because the general notion should be that, because brown fat is such an important contributor to energy balance, the absence of its activity should have a tremendous impact on energy expenditure. Nevertheless, when you actually test for brown fat activity with the injection of, for example, beta-3 agonist, which you see here, then you see that wild type mice mount a very strong response, which was completely absent in the knockout mice. So, indicating that, yes, there was no brown fat activity, but the mice were somewhat able to compensate for this so that in the end baseline, actually, there was not much difference.

So, what could be one of those mechanisms? And what is very obvious, and I mentioned this in the beginning, cold acclimatization is dependent on many systems, And here, when you measure temperature, for example, in the core, there’s not a difference between wild type and knockout mice, but both the brown fat temperature and the skin temperature are reduced, which might indicate that there is a decreased heat loss. And actually, the most important role here is also of the sympathetic tone to prevent heat loss by limiting the blood flow into the periphery. And I have to point out here, very important for mice is actually the tail because mice can really dictate how much heat they lose via the tail, which makes the tail a major thermoregulator. And this is probably not the only example where you can see a strong disconnect between cold acclimatization and brown fat activity. There are other examples in the literature.

This is a very interesting paper by Kaipat et al., where they looked into the energy expenditure at decreasing temperatures of FGF21 knockout mice and UCP1 knockout mice – two models where you would expect that brown fat activity is lower. And you can nicely see in this diagram that they just adapt fine to the cold because of compensatory mechanisms. And I think this again shows that there are important exceptions to the rule that brown fat is such a strong contributor to cold-induced thermogenesis. So, a couple of take-home messages here. What you should really keep in mind is that cold acclimatization is a complex physiologic and particularly systemic process that pretty much impacts on every tissue in the body and that there are two major phases which you can distinguish in your research. One is the acute exposure to cold which is something like up to 24 hours which impacts existing cells and tissues, and anything beyond that could be considered some sort of cold acclimatization, and I tend to say that it takes at least seven days until you see strong recruitment of brown adipose tissue. And keep in mind that cold acclimatization does not require non-shivering thermogenesis. And because of that, also cold is an inappropriate testing regimen for non-shivering thermogenesis. So, in order to overcome this, I recommend that you use indirect calorimetry to measure no epinephrine or CL-induced respiration. And at the same time, you use tracer studies to measure nutrient disposal directly into your tissues. And this can be done with fatty acid tracers or glucose tracers. And there are a lot more that are being developed over time.

So, with this, I thank you for your attention. Thank you very much.

Sarah: Okay. Thanks so much for that awesome presentation. And before we move on to our Q&A, we are going to run one more quick audience poll. So, this question is, do you currently use metabolic phenotyping in your research – yes or no? So, that’s a quick one to answer, and I will give everyone about 30 seconds. So, it’s great to see those answers coming in. It looks like we have a lot of people doing metabolic phenotyping, so that’s really great. Okay, thanks again so much for answering these polls. And while we wait for those last couple answers to come in, I’m going to welcome back our speakers. Jonas and Alex, are you with us?

Dr. Treebak and Dr. Bartelt: Hi Sarah, yes.

Okay, so we are going to kick off our Q&A. So, the first question here is for you, Jonas. This question is, in your publication, you describe the wild type and skeletal muscle NAMPT knockout mice. Did you also phenotype the heterozygote animals? And do they have an intermediate phenotype or are levels still sufficient in these animals?

Dr. Treebak: That’s a good question. I mean, we, in the initial phase, we did do some studies in the heterozygous mice. But when we found a very profound phenotype in the full knockouts, then we sort of went in that direction. But we didn’t really pursue the potential intermediate phenotype that we would expect from the heterozygous mice. But to be honest, as I recall the data, we didn’t see much of a phenotype in the heterozygous mice. They were mostly like the wild-type mice, which I also would expect from the NAD levels, which were reduced, but not to the same extent as in the knockouts.

Sarah: Okay, great. Thank you. This next question here is for you, Alex. What do you think is the role slash contribution of shivering thermogenesis to cold acclimatization?

Dr. Bartelt: Yeah, I think this is a matter of debate and therefore there’s a great question to kick this off. I think there are multiple papers that now have actually measured shivering, for example, by motion detectors, and see how really muscle shivers. And these have a great resolution so you can also measure small peaks and muscles shivering, and there are the papers that show that definitely cold increases shivering, but it’s very difficult to really take that away and measure its actual contribution. I remember there’s a paper, I think it came out in Nature Medicine some time ago, where they use one of the South American arrow toxins, curare, that would, you know, block shivering and use that to determine the relative contribution. So, if you refer to that, probably it is important during the first couple of hours. How much it is, I’m not quite sure – it’s difficult to measure. Certainly, we have also in the past done models that impact on muscle metabolism in the cold and clearly if muscle shivering does not work properly, cold exposure is severely dangerous for the mouse and it cannot survive.

Sarah: Okay, great. Well, not great for the mouse, but great answer. My next question here is for you, Jonas. This question is, do you have plans to do future treadmill studies? And if so, will you include the knockout mice and the heterozygote mice?

Dr. Treebak: I think, to be honest, I don’t think we will include the heterozygous mice. I mean, all the exercise studies that we are doing and the work that we are doing with the NAMPT in muscle mice will be with the inducible model. And I can say for that model at least, the heterozygous mice don’t show any phenotype whatsoever. So no, we will only do the wild type and the knockout mice in the future. But it’s a good suggestion, but I don’t think we would see anything to be honest.

Sarah: Okay. Great. Alex, this question is for you. What is the reason for why failure to remove misfolded proteins results in the whitening of brown adipose tissue? Is it that these misfolded proteins lead to loss of mitochondria?

Dr. Bartelt: Yeah, so it’s actually a relatively slow process. We’ve done experiments where the mice were born at thermal neutrality in the absence of any cold. And there you can see that the tissue is completely indistinguishable from wild type tissue. And if you then place these mice at colder temperatures, for example, room temperature, the mice adapt fine, but it takes about a week or so. Again, the regular window of brown fat recruitment, where you start to see that the tissue actually, instead of becoming browner, is whiter. And the whiter histologically actually does not reflect so much how white it is actually. And it is absolutely true that a major effect that we see is a loss of mitochondria, a loss of iron that is quickly observed. And there are also hallmarks of mitophagy in this tissue. We’ve done also ubiquitin studies where we have explored what actually are the proteins that are found in a hyper ubiquitinated state and about 40% of the brown fat ubiquitin comprises mitochondrial proteins and not only those are actually 40% of the mitochondrial proteins and many of those that are, yeah, have not been known to be turned over by the proteasome. Is it only mitochondria? I don’t think so. It’s very difficult to really pinpoint one specific mechanism, but I’m certain that the mitochondrial dysfunction that we observe under these conditions is a main driver of the brown fat dysfunction.

Sarah: Okay, great. Great answer. Jonas, we have another question here for you. This question is how long do knockout mice survive?

Dr. Treebak: So, I can take both models. I mean, the constitutive mice die prematurely. We have only had one mouse so far that has lived to actually be able to breathe. But they die, and we only included data for the first 12 weeks in the paper, but they still die eventually. And the inducible model, we do see animals, as I was saying in the presentation, I think we have had mice that have lived for more than two years, so they don’t have a mortality phenotype.

Sarah: Okay. Great. The next question here, first says, Alex, thank you for the talk. And second says, when culturing adipocytes in vitro, how important is the temperature of the incubator? Can a decreased incubator temperature induce browning and is 37 degrees Celsius appropriate with respect to thermoneutrality?

Dr. Bartelt: Yeah, that’s a fascinating question. And I think from a cell biology standpoint, we don’t really have a good answer. There’s a PNAS paper by the Spiegelman Lab where they show that also in-vitro adipocytes can sense temperature. That’s very provocative. I don’t think it has been picked up in the literature ever since. At the end of the day, what do cells need to be happy? And I think also it’s a misconception that in the mouse, the brown fat cells are actually cold, rather the opposite, they’re actually warmer. The brown fat tends to have a warmer temperature than the rest of the mouse because it needs to produce a lot more heat that is being then transported away into the body. So, I don’t think these terms of thermoneutrality would apply to a cultured adipocytes. Nevertheless, yeah, it does probably make sense to study the effects of temperature when culturing cells a little bit in more detail, and it could very well be that cells have a different, let’s say, preference zone at which temperature they are being best cultivated. For sure, you can say that at the regular incubator temperature, cells differentiate fine and work fine. So, it’s definitely a feasible model.

Sarah: Okay. My next question is for you again, Alex. This question is, you mentioned the differences between mice and humans. There have been many attempts at translating these to therapies for humans. What are your thoughts on this?

Dr. Bartelt: Well, I think that using brown fat and thermogenic adipocytes as a therapy for humans is a long shot goal. I mean, we’re all aware of the small amounts of thermogenic adipocytes in humans and a very small contribution to overall energy expenditure. But I study brown fat not because of its translational potential, rather because of its unique biology. And I do think that there are a lot of secrets that we can extract from these cells that maybe teach us something about general metabolic adaptation. And I’ve shown some results on Nfe2L1, for example, which clearly has a very interesting role in brown fat, but we use it more now as a starting point to explore its role in more translational work, which for example, is in white fat cells and obesity or other metabolically relevant cells. So, I think there’s still a lot we can learn from brown fat to understand general concepts of metabolism. But if you think in terms of does brown fat matter in humans, I think the clear answer is yes. And people, even though these are mostly correlations, people that have more brown fat are usually healthier. And a thermogenic lifestyle, to my opinion, does really, if you pursue this on a long-term scale, does really contribute to a healthier metabolism.

Sarah: Okay, great. So, I have one last question here and it’s for both of you. So, I’m going to start with Jonas, but then after Alex, you can weigh in as well. This question is, in your opinion, what is the future in this research field? What direction will your lab go based on your current data?

Dr. Treebak: So, as I also alluded to in my talk, I think for, you know, for how the inducible NAMPT knockout mice adapt in order to maintain respiratory capacity, even with low levels of NAD, I think one thing that we haven’t done yet is to study the mitochondrial super-complexes and how mitochondria may adapt in that aspect, I think is very fascinating. And doing that both in the base state, but also with acute exercise, even exercise training, I think could be something that could potentially explain the lack of phenotype in our mice. So yeah, that’s one future research direction that we will take.

Sarah: Great. And Alex, same question to you. In your opinion, what is the future of this research field and what direction will your lab go based on your current data?

Dr. Bartelt: I think that all mammals with or homeotherm animals and therefore temperature is a very important denominator of metabolism. And also considering climate change, I think temperature has become or has come again into the spotlight of not only public health, but the entire human race. And therefore, I’m very much interested in cold as one overarching principle that directs our metabolism. And we’re now pursuing whether there are unique, overarching concepts that we can translate, for example, the biology of brown fat into other tissues like muscle or the heart and can see whether those really help us to understand the physiology of temperature and the metabolic adaptation to cold, and how we can maybe exploit that also to fight the other big pandemic, the other public health crisis, which is obesity and cardiovascular disease. So, this is the research direction we’re taking at the moment. And it’s very exciting because there’s a large overlap between things that you would have never thought are related to each other.

Sarah: Great. Awesome. Well, I just wanted to thank you both so much again for all of your fantastic insights today. It really was a pleasure to have you both with us.

Dr. Treebak and Dr. Bartelt: Thank you very much, Sarah. And thanks to the audience. Thank you.

Sarah: In closing, thank you again for taking part in this Inside Scientific webinar, sponsored and made possible by Sable Systems. We look forward to having you with us again soon.