 OK, I'm going to make the introductions very brief. And our first speaker is Chris Masterjohn, who will describe resolving the vitamin D paradox. And we'll have a punchline at the end of his talk. Thank you very much, everyone. So the title of my talk today is Resolving the Vitamin D Paradox, or Vitamins A and K required to render vitamin D a heart-protective nutrient. And so I'll first have to try to convince you that there is a vitamin D paradox to be resolved. And then I will try to convince you that vitamins A and K are part of resolving that paradox. So before I start getting into what this vitamin D paradox actually is, I'd like to make sure we're all on the same page just by giving a brief overview of what these nutrients we're talking about are and where they're found in foods. And you can see that the main source of the fat-soluble vitamins A, D, and K in general is in liver, sunshine, fatty animal products, fermented foods, and colorful vegetables. Because vitamin A is stored in the liver of animals, that's where we keep vitamin A when we're not using it. That's the resource that we tap into to get vitamin A when we need it. Then if you're looking at terrestrial animals, the vitamin A is primarily found in the liver. It's also found in the livers of fish. And so if you look at fish liver oils, cod liver oil is given as an example here. That's also a great source of vitamin A. Vitamin D is primarily found, well, not found in the sun, but when the sun strikes our skin, converts a precursor in our skin to vitamin D, and then we then absorb it. But fish are unique in that they also store vitamin D in their livers. And so fish liver oils, including cod liver oil, are also very rich in vitamin D as well. And so that makes fish liver oils unique in being a great source of both vitamins A and D. There are lesser amounts of vitamins A and D found in terrestrial animal fats, especially in the fats related to reproduction, like dairy fat or egg yolks. Dairy fat being meant to nourish a growing animal, and egg yolks meant to become a young animal. We can also supplement vitamin A by obtaining carotenoids from colorful vegetables. Beta carotene is a prototypical example of a carotenoid that can be converted into the physiologically essential form of vitamin A. And because it plays a role in photosynthesis, it's associated with the green color of photosynthetic tissue. And because these carotenoids also provide yellow, orange, and red pigments, then whenever you see the colors red, yellow, orange, or green, that's a good sign that vegetables can be a good source of vitamin A. However, the ability to make the conversion of those carotenoids into vitamin A can vary dramatically between different people because of genetics, and it can also vary dramatically between different people because of their life stage or health situation. And so it's really animal products, especially liver and cod liver oil, that are the most reliable sources of vitamin A. But depending on who you are and what your condition is, these colorful vegetables can supplement. Vitamin K is found in two forms, vitamin K1 and vitamin K2. Leafy greens are the best source of vitamin K1 because it also plays a critical role in photosynthesis. But vitamin K2 is primarily found in animal fats or fermented foods. Animals will convert K1 in these leafy greens into K2. And on top of that, bacteria produce their own vitamin K2. And I gave as an example here cheese, which is both an animal fat and a fermented food. So when the cow eats grass, which is a form of leafy greens, it gets plenty of K1, then it converts that K1 to K2. And then when someone takes the dairy product and ferments it into cheese, the bacteria make additional vitamin K2. The K2 found in animal foods and fermented foods and the K1 found in plant foods all vary in structure and in their precise roles in the body. So it's really best to get all of these. So I really think that eating a mix of all these different foods is the best way to get a synergy of these fat-soluble vitamins. But now that we kind of understand what we're talking about when we speak of them, I'd like to get into what is this paradox and why do we want this synergy? Why don't we just care about each of these vitamins individually? So here is a graph showing a meta-analysis of the association between 25 hydroxyvitamin D in the serum or plasma of humans and their risk of cardiovascular disease over time. 25 hydroxyvitamin D is a metabolite of vitamin D that is produced in the liver. But when we measure it in the blood, that's our best look into someone's vitamin D status. So I'll just refer to this as vitamin D status for the sake of simplicity. So when you're looking at this graph, you can see that as we go up the vertical axis, the risk of heart disease is increasing. And if we go across the horizontal axis, the concentrations of 250 HD or vitamin D status is increasing. And if you look at these circles, each individual circle represents an individual risk estimate from a particular study. And because this is a meta-analysis, it means that it's pooling the results of many studies together. Now, if you look at this line here, what it's trying to do is generate an average to see the pooled trend from all these different studies. And when you see the shaded area, this represents the confidence interval. And that means that the narrower the shaded area is around that line, the greater our confidence is that that's where the actual risk is. But the wider that gray area is, the less confident we are about that data. And so you can see a couple of things from this graph. One is that as we get higher and higher in vitamin D status, there are fewer and fewer of these circles and the circles that are there are very small. That means that there's not much data in that range. The data that we have are from some pretty small studies with pretty few people actually having vitamin D status that high. Consequently, this shaded area is very large. When we look at concentrations between eight and 28 nanograms per milliliter, we have a real lot of data. So the shaded area is very tight around the line. And what we can see is that between eight and 24 nanograms per milliliter, we have a clearly decreasing risk of cardiovascular disease with increasing vitamin D status. But as we get higher and higher away from 24 nanograms per milliliter, it kind of looks like that risk stops decreasing. But what we really see is that's where the gray area starts really increasing, which means there's so little data up in that range that we really have no idea what's going on. And the reason this is a little concerning is because many advocates of vitamin D supplementation are saying we should all be at least this high or over here somewhere. And that's the place where we don't really have any data about heart disease, so we don't really know what's going on. After this meta-analysis was published, a new study came out in cardiovascular, cardiac surgical patients. So these are patients with a very high risk of heart disease and consequently, there was a very high proportion of them who had major cardiac events in the subsequent year after they measured vitamin D status. So they were able to get a real lot of data in a very short amount of time. And what they found was that between 20 and 40 nanograms per milliliter, there was a very large drop in the risk of major cardiac events compared to low vitamin D status. However, if you had vitamin D status over 40 nanograms per milliliter, then you were just as bad off as if you had very low vitamin D status under 12 nanograms per milliliter. So this indicates that there's a sweet spot here and either higher or lower vitamin D status is harmful. Again, this is concerning because many people are trying to get their vitamin D status over here. Now I should say that these are observational studies. They don't have the ability to show cause and effect. And so naturally they don't show that having higher or low vitamin D status is actually causing heart disease. However, I'd like to use this data to frame further evidence for a causal hypothesis that I'd be exploring today with you. Now, at first glance, and from what I've presented so far, this isn't necessarily all that paradoxical. Yes, it's a little bit odd that high vitamin D and low vitamin D can be harmful, but really it's not that much different from ideas we have like moderation and everything in moderation, too much or too little of a good thing can be bad, too much water, too little water can be bad. This idea of a U-shaped risk curve isn't that paradoxical. However, it's a little paradoxical that the harmfulness or apparent possible harmfulness of having high and low vitamin D status is the exact same thing. And it gets even more paradoxical when we look at the mechanisms by which we can postulate that vitamin D status is having these effects because it seems that the mechanism whereby high vitamin D status might contribute to heart disease and low vitamin D status might contribute to heart disease may be the same mechanism. And so that's where it really starts to get paradoxical. Now, before we get into that, let's look at some parallel data from kidney stones. We know from animal studies that if you give toxic quantities of vitamin D to an animal, you get widespread pathological calcification of soft tissues. That means calcium's going into the heart and the kidneys where it doesn't belong and it's not go and possibly at the expense of going to the bones and teeth where it does belong. And when it goes to the kidneys, what we see is kidney stones. And this is data from Israeli lifeguards who had twice as high vitamin D status as non-lifeguards in Israel. And they had about 20 fold higher incidents of kidney stones. And you can see that in Southern Israel, the proportional increase in risk was not as high. It was 12 fold compared to 19 fold. But that's because the general population in Southern Israel has twice the incidence of kidney stones as in Northern Israel. And that's consistent with lots of observational studies showing that as you get closer to the equator, you get higher risks of kidney stones. And that could be related to increasing vitamin D status. Now, this 53 nanograms per milliliter isn't that much different from the higher than 40 nanograms per milliliter that we saw potentially contributing to heart disease in cardiac surgery patients. These authors noted that these Israeli lifeguards also had signs of dehydration and of skin damage. And they said that probably all of these contributed to the increased susceptibility to kidney stones. Again, that's probably not that different from the cardiac surgical patients who are not the general population. They are people who have specifically greater risk of heart disease because of other interacting factors. So we might propose from this that perhaps high vitamin D status is contributing to kidney stones and to heart disease when there are other interacting negative factors predisposing to those conditions maybe not harmful in every specific person but we'll get into these contextual factors more soon. First, let's take a look at the mechanism. What we see is that animal experiments show that both low and high vitamin D can contribute to pathological cardiovascular calcification. Most of the calcium in our body belongs in the extracellular matrix of bones. And when it starts accumulating in the blood vessels and the heart valves, this can contribute to heart disease. And what we're seeing from this study is that when animals were fed either 5% of the minimal requirement of vitamin D or 10 times higher than the minimal requirement of vitamin D, the only the animals that were fed the deficient amount of vitamin D at an increased area of calcification found in their cardiovascular tissue and an increased number of calcification spots. By contrast, we see another study in pigs where when they fed various doses of vitamin D, all of the pigs that were fed 25-fold higher than the minimal requirement of vitamin D had atherosclerosis. And when they looked specifically at calcification, they saw a dose-dependent increase in the proportion of arterial segments with atherosclerotic plaque shown on the left and pathological calcification shown on the right. So these animal experiments are suggesting that high and low vitamin D does, in fact, causally contribute to the risk of heart disease. And what's particularly paradoxical is that the mechanism seems to involve pathological calcification, both in high vitamin D and in low vitamin D. Now this gets even more paradoxical when we start to consider some of the evidence about the other contextual factors that modify this process and the actual mechanism going on. So if we look at the conventional understanding of vitamin D toxicity, we have the idea that normal vitamin D leads to normal serum calcium, and that when serum calcium is normal, calcium goes into all the right places. It stays in the soft tissues very little, it goes primarily into the bones and teeth. However, the conventional idea is that when we have too much vitamin D, we get hypercalcemia, which is too much calcium in the blood, also too much phosphate, and this excess of calcium and phosphate circulating in the blood spontaneously precipitates and lands into the soft tissues. In 2007, I suggested that we redefine this mechanism because there are a number of observations that seem to contradict that this is the only explanation for the mechanism of vitamin D toxicity. Shown here is the paper that I published, and I'll just briefly explain the rationale. So I initially got involved in looking into fat-sable vitamin interactions because in the 1990s, we started to see the development of the idea that too much vitamin A would contribute to osteoporosis. And as I looked into the literature, I was seeing that this idea came from observational studies conducted at northern latitudes where access to sunlight for vitamin D and dietary vitamin D was very scarce and people were very deficient in vitamin D. And that if you looked into the animal experiments, you saw that toxic doses of vitamin A would cause bone loss and spontaneous fractures in animals. But if you gave the same animals massive doses of vitamin D alongside the vitamin A, you would not see that effect. On the other hand, if you gave these animals toxic doses of vitamin D, they would get pathological calcification of all the soft tissues, the kidneys, the aorta, the heart valves, the lungs, the bladder, and so on. But if you also gave them massive doses of vitamin A, you would not see that toxicity. This was shown across many species. It was shown in rodents. It was shown in birds. It was shown in cows. And it was shown with less mechanistic detail even in humans in a 1941 study showing that when you gave massive quantities of both vitamins, you could increase immunity to the common cold and get no toxicity. But when you just gave one vitamin or the other, you would not get any benefit for the common cold and you would get toxicity symptoms and people would quickly drop out of the study because they couldn't handle the toxicity symptoms. Now, if you also look at these studies, what you see is that vitamin A, when it protects against soft tissue calcification caused by vitamin D, it does not protect against the hypercalcemia. And so this is showing that the excess calcium persists in the blood and yet for some reason the soft tissues don't calcify. And so that shows that there is something else going on besides just hypercalcemia. On the other hand, there was a study in the 1970s where vitamin D in the diet caused kidney stones in chickens at doses that were not high enough to produce hypercalcemia. So you can get vitamin D causing soft tissue calcification without causing hypercalcemia and you can get vitamin D causing hypercalcemia without causing soft tissue calcification depending on the dietary context in which that vitamin D is given. And so this becomes even more paradoxical if we're to look at it solely from the conventional paradigm that hypercalcemia is what's causing the soft tissue calcification. But at the same time, this data can help us resolve the paradox if we allow it to start exploring some potential alternative ideas to how this mechanism might be playing out. And of course this isn't to say that hypercalcemia doesn't contribute to the soft tissue calcification, it's just to say that there are other things going on. And the mechanism that I proposed concerns the regulation of vitamin K dependent proteins. As it turns out, it was long thought that soft tissue calcification was a relatively passive process where when you have too much calcium and phosphorus in the blood, they spontaneously precipitate and deposit the soft tissues when they shouldn't. But in fact, all extracellular tissues of all vertebrate animals are super saturated with calcium and phosphorus which means that there's always enough to spontaneously precipitate and it's needed to be maintained that high so that bone can mineralize. And the reason that tissues don't calcify normally isn't because calcium and phosphorus isn't high enough, it's because we have numerous proteins and non-protein molecules whose purpose is to actively defend the soft tissues from calcification. And one of those protective components is a vitamin K dependent protein called matrix glop protein. Now my hypothesis was that based on cellular experiments showing that vitamins A and D regulate vitamin K dependent proteins and that vitamin K activates those proteins, I suggested that when these vitamins are obtained in the proper balance, you get properly regulated and activated vitamin K dependent proteins and this leads to strong bones and teeth and protects the soft tissues against calcification. By contrast, when you have a great excess of vitamin D relative to vitamin A, you get a massive over-production of vitamin K dependent proteins. This leads to an overwhelming of the cells' capacity to activate them and then you get a bunch of inactive proteins that can't do their job. The soft tissues will not be protected and calcification ensues. If we were to take an analogy and we would say there's an assembly line and my job is to assemble widgets. Stefan is close to me, he's right next down the line so his job is to paint the widgets and I make the widgets at the same pace that Stefan's painting them, we get a bunch of painted widgets. But if I'm making widgets at 10 times the rate that Stefan's painting them, they start to build up on the table, eventually can't keep them organized, they all start falling on the floor, people walk around, start breaking all the widgets and then we get a bunch of unpainted and broken widgets that are useless. So it's a similar idea here if you view the production and activation of the vitamin K dependent proteins as an assembly line, if you're overwhelming the capacity of the downstream elements of that production to keep up, you're going to wind up with a bunch of ineffective products. Vitamin K's role in this process is to activate these proteins by adding carbon dioxide to them. Carbon dioxide then gets rearranged and adds an extra negative charge to the protein. By increasing the negative charge, it better allows the protein to bind to calcium, which is a positively charged ion. In the case of some proteins, binding to calcium just allows calcium to act like a structural glue that maintains the shape of the protein so it can contribute to its proper function. But in the case of matrix gloprotein, which is a protein that defends soft tissues against calcification, it's actually allowing the protein to directly bind to that calcium and prevent the salts from growing and depositing into the tissues. If you look at the matrix gloprotein or MGP knockout mouse, what's the first thing you notice about the picture? Yeah, the MGP knockout mouse is smaller. Well, what's going on in this mouse is that the growth plates, which are soft tissues made of cartilage are calcifying early and so the animal isn't growing. But also that calcium is not going into the bones and teeth so they're getting osteopenia and spontaneous fractures and it is going into the soft tissues, including the arteries. So they die within two months because they're heavily calcified arteries rupture. If we look at the epidemiological evidence, we see that high intakes of vitamin K2, which is the form found in animal fats and fermented foods that's better able to make it to the vascular tissues where that protein is produced. Higher intakes of vitamin K2 are associated with a lower risk of coronary heart disease incidents, mortality and severe aortic calcification. And you can see in animal experiments if you inhibit the recycling of vitamin K with warfarin, which is a rat poison and also anticoagulant used in human clinical medicine, you can see that the warfarin leads to cardiovascular calcification shown by the black staining and that if you add vitamin K1 to the food, which is found in leafy greens, it has no protective effect, whereas if you add vitamin K2, it completely protects against this effect. In actuality, this is probably species and strain dependent because some animals are better able to convert K1 to K2, but this shows experimental evidence that vitamin K, especially vitamin K2, is activating these matrix gloprotein and protecting the cardiovascular tissues from calcification. So if we go back to my hypothesis and make it a little more detailed, my hypothesis is that if we have vitamins A, D and K in balance, we get properly activated matrix gloprotein that provides strong bones and teeth and protects strong tissues from calcification, including the cartilage growth plates thereby allowing for adequate growth. But if we get a great access of vitamin D relative to vitamin A, we overproduce MGP. It can't be adequately carboxylated using vitamin K and then it can't protect soft tissues from calcification. So in 2007, I published this hypothesis and then I gave it to Zheng Dong Wang at Tufts University to suggest that he investigate the hypothesis in an experimental model that he had used showing that in mice, pharmacological metabolite of vitamin D could protect against lung cancer, but it would cause kidney stones. And if you gave a pharmacological metabolite of vitamin A, you could still get the benefit of protection against lung cancer, but the mice would not get kidney stones. And so they then investigated this hypothesis in the tissues that they had frozen from that experiment and provided strong support from my hypothesis. The graphs from this study are all shown with the control group, the group given retinoic acid, which is a metabolite of vitamin A, the group given calcitrile, the metabolite of vitamin D and the group given both metabolites together. Here we're looking at the MGP mRNA expression, which is the message that the cell is telling to the cellular machinery to produce this protein. And you can see only with D alone, are they getting an increased message to produce more of the protein. Then if you look at the undercarboxylated form of the protein, which is the defective form that can't do its job, you see that this defective form was only increased in the group getting D alone. And when you look at the carboxylated form of the protein, which is the form that is activated and can do its job, you see that it was highest in the group that got both metabolites alone. And this is actually going beyond what my hypothesis predicted, because it suggests that there's something more than the overproduction going on, and perhaps the carboxylation or activation efficiency is actually improved when you're getting both vitamins together. If you look at the overall results as the proportion of MGP that was defective, you can see that it was highest in the group that got the metabolite of D alone and lowest in the group that got the two metabolites together. Again, suggesting that D alone is harmful, but also suggesting that this synergy between A and D is actually improving the situation over just getting a, quote unquote, normal diet. Some questions that were left unresolved from this study. Can these mechanistic findings be replicated using dietary vitamins instead of the pharmacological metabolites? Is vitamin K protective? Maybe it is, but maybe the vitamin K dependent enzyme isn't there in adequate amounts. So maybe adding vitamin K wouldn't be protective. I think it would be, but that's up in the air. How do these treatments affect the enzymes that recycle vitamin K or use it to activate proteins? Perhaps there are other elements of vitamin K metabolism that are affected, and perhaps that could explain better why the result is better when you get lots of both vitamin instead of a normal control diet. Can the effect on vitamin K dependent proteins be definitively separated from the effect on serum calcium? Are these really two independent mechanisms? And then, can these effects also be seen in arterial tissue and in humans? So working as a postdoc at the University of Illinois over the last two years, I started investigating this study by feeding rats a diet for six months that got one, where each group got one of five different doses of vitamin D. The lowest dose was the normal recommended amount for the rats, and the highest dose was 50 times higher than that. You can see that vitamin D status dose-dependently increased, verifying that there actually was more vitamin D in the diets like I thought there was. And when we looked at the tissues, we didn't see increased calcium in the cardiovascular tissues, but we did see it in the kidneys. So I followed that up with a quantitative analysis showing that calcium was deposited more in the kidney tissue of the animals that's shown on the left. On the right, you can see that phosphorus was only slightly increased. That's probably because there's just normally so much phosphorus in the cell that there's a lot of noise that's drowning the signal. But this seems to support that calcium phosphate crystals were being deposited in the kidney. However, I was hoping that this would produce a model independent of hypercalcemia, and it did not. Serum calcium, oh, excuse me. That's the next slide. Showing here, the undercarboxylated osteocalcin, which is a marker in serum of poor vitamin K status was increased, and the ratio of undercarboxylated osteocalcin was also increased, and that confirms that the model does seem to involve some sort of impaired vitamin K status. But on the other hand, serum calcium phosphorus and the calcium phosphorus product were also all increased, and this did not turn out the way I expected it to. So this does show that vitamin K status is impaired, but it doesn't definitively separate it from serum calcium. Now this is sort of surprising, because I only saw these effects in the highest dose group where you're seeing very high vitamin D status, much higher than what's seen in cardiac surgery patients with increased heart disease and in Israeli lifeguards with increased kidney stones. And it's also surprising because the literature has previously shown that rats with even this high vitamin D status shouldn't get hypercalcemia and mine did. And so I went back and looked at earlier blood samples and I don't have the data to show you, but one interesting thing is that even at eight and 16 weeks, these animals did not have hypercalcemia and they only did after 24 weeks. So one of the important implications is that a study that's less than six months long isn't adequate to see the true effect of vitamin D in these animals. In the next week, I should be completing measurements of all of the vitamin K dependent proteins, how they're expressed in the kidney and I'll be looking at their carboxylation status over the next month or two. And also over the summer, I had Grace who's in the audience right there working with me on a research project to see what happened to the bones in these animals. And surprisingly, it seems that these animals had better bones, worse kidneys and no difference in cardiovascular tissues. So far this is consistent with all my other dietary studies that they never turn out quite the way that I expected them to. I will say however that I had very high expectations for Grace based on her previous accomplishments and in talking to her supervisors at the CDC where she had formerly worked as a research chemist. She surpassed all of my expectations. I hope she's able to present what she did this summer next time at AHS. If you see her, you should say hi to her and if you ever have the opportunity to collaborate with her or hire her, you'd either be an ignorant or a fool not to do so. Moving on. So future directions that I wanna look at are will vitamins A and K protect against the harms that vitamin D caused in these animals at this dose? But I'm really starting to wonder if maybe a more productive route is to look at other risk factors to see these animals had such high vitamin D status that it might be irrelevant. And so is it possible that if we can better replicate the human situation where instead of having a purified refined diet that's been continually improved over the course of the 70s through the 90s to make sure all of the rats have perfectly adequate nutritional status, can we instead look at something that actually looks like the situation in humans where we have genetic polymorphisms that interfere with our ability to harness nutrients from foods. We have widely varying nutrient intakes and we have widely varying other risk factors that are contributing to these diseases. So I would like to maybe look at a model where I'm actually removing the other vitamins to see if that causes harm at lower doses of vitamin D and harm in other tissues like the cardiovascular tissues rather than to see, rather than seeing whether the other vitamins are protective in this particular context because I'd like to see if I can make this model more relevant to human disease and eventually actually look at it in humans. I'll be looking at these types of questions in my new position as Assistant Professor of Health and Nutrition Sciences at Brooklyn College where I'll be teaching classes on nutritional chemistry to undergraduates who wanna be registered dietitians accepting master's students who wanna work with me and moving on to conduct this type of research in the new laboratory that they're renovating for me which should be ready in the next couple of months. So I'd just like to briefly acknowledge the Westinay Price Foundation for funding some of my research, Fred Kumaro and the rest of the department of Comparative Biosciences at University of Illinois where I conducted most of this research to date and you for attending my talk, so thank you very much. Do we have any time for questions or no, okay. Okay, so I'll be, I think I'll just make myself available for questions during the break, okay. Thank you.