The Peter Attia Drive - #147 - Hussein Yassine, M.D.： Deep dive into the ＂Alzheimer's gene＂ (APOE), brain health, and omega-3s

00:00:00.000 Hey, everyone. Welcome to the drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:15.480 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:19.800 into something accessible for everyone. Our goal is to provide the best content in health

00:00:24.600 and wellness, full stop. And we've assembled a great team of analysts to make this happen.

00:00:28.880 If you enjoy this podcast, we've created a membership program that brings you far more

00:00:33.280 in-depth content. If you want to take your knowledge of the space to the next level at

00:00:37.320 the end of this episode, I'll explain what those benefits are. Or if you want to learn more now,

00:00:41.720 head over to peteratiyahmd.com forward slash subscribe. Now, without further delay,

00:00:47.740 here's today's episode. My guest this week is Dr. Hussain Yazin. Hussain is an associate professor

00:00:55.520 in the Department of Medicine at the Keck School of Medicine at USC. His lab has focused for almost

00:01:01.180 a decade now on lipid metabolism and nutrition and the effect these have on cognition and the

00:01:07.220 risk of developing Alzheimer's disease, paying specific attention to the ApoE4 allele. Many

00:01:13.160 of you listening to this podcast will no doubt be familiar with ApoE4 genotypes. We've spoken about

00:01:18.820 this at great length in a number of podcasts. This is really the deep dive into E4, what the

00:01:25.600 implications are. I actually wanted to speak with Hussain after we were part of a journal club

00:01:31.060 where he presented a couple of papers that he authored that were simply beyond fascinating.

00:01:36.240 And it was definitely one of the deepest dives I'd ever seen into the inner workings of EPA and DHA's

00:01:42.580 role in the brain. Now, we do get to that near the end of the podcast, but I realized that it's

00:01:47.540 important for people to really understand a lot about the brain. And so I would consider this kind

00:01:51.780 of a fundamental core coursework on brain biology. We really explain the architecture of the brain,

00:01:59.580 the energy systems of the brain, and what goes wrong. What are the various things that go wrong?

00:02:04.500 Because it's not just one thing as a person develops Alzheimer's disease. And then we get into

00:02:08.680 what ApoE is, and then why we might have these three different variants or isoforms of it,

00:02:15.140 E2, E3, and E4. And what turns out to be different depending on which of those you have.

00:02:20.440 The obvious thing, of course, is your risk for Alzheimer's disease, but why?

00:02:24.080 We have a little side tangent that goes into sort of the role of EPA in cardiovascular disease,

00:02:29.880 with or without DHA, and then kind of bring it back to this discussion around the role of DHA

00:02:35.300 in Alzheimer's disease, and also what some of the specific things that we understand to date

00:02:41.160 with respect to patients that have E4 and what they can do to reduce risk. I'm going to give you

00:02:45.340 a warning up front. Unfortunately, if you're anything like me, you're going to come away

00:02:49.400 from this episode, I think, frustrated that our state of knowledge is still so pedestrian. But

00:02:55.160 nevertheless, this is a very important discussion, and I certainly learned a lot in it, and it will

00:02:59.800 actually have an impact on how I think about taking care of my patients. So without further delay,

00:03:04.500 please enjoy my conversation with Hussein Yazeen. Hussein, thanks so much for making time to sit

00:03:15.380 down today. I'm really looking forward to this. Obviously, I had the pleasure of doing Journal

00:03:19.700 Club with you a few weeks ago, which is what made me realize we needed to do this for a much broader

00:03:24.080 audience. Let's kind of go back to the beginning a little bit, though. I'm kind of curious as to

00:03:28.340 what got you interested in this space. I know you grew up in Lebanon. Did you do medical school

00:03:33.980 there as well as college? Yeah. Thank you, Peter, for the invitation. I'm happy to be part of your

00:03:39.420 show. Yes, I grew up in Beirut, Lebanon. That's my medical school. And then shortly after, moved to

00:03:46.700 the U.S. to do some residency training, followed by a fellowship. And then I got to where I am right now.

00:03:56.860 Tell me a little bit about that path. Your fellowship was in what?

00:03:59.360 I trained in endocrinology. I come from a family with strong heart disease and diabetes. So I decided

00:04:07.600 to subspecialize in lipid disorders. And I studied lipids for a good of two to three years before

00:04:15.920 I had the opportunity to study brain lipids. And I switched from blood lipids to the brain.

00:04:23.180 Well, this is a topic that I think a lot of people are interested in because I remember 10 years ago

00:04:30.900 when you would check an ApoE genotype on somebody, it was a very unusual thing to do.

00:04:39.740 Today, it's not as unusual. It's probably still not incredibly common that people are walking around

00:04:45.000 knowing their ApoE genotype. But it's certainly, I mean, it seems a log order more common than it was

00:04:51.020 a decade ago. And I suspect a decade from now, that might be one of the genes that kind of

00:04:55.740 everybody knows about themselves. So where do you think is a good place to start? I think this topic

00:05:01.820 is complicated enough that it probably warrants at least some discussion of what makes up the brain.

00:05:09.660 Like we talk about neuronal cells, we talk about glial cells. Can you maybe help explain the

00:05:15.260 architecture of the brain so that as we get into the different types of cells, what a blood brain

00:05:20.940 barrier means, where CSF resides? These things will become important in this discussion. And I think

00:05:26.380 we should assume that most people listening don't necessarily understand those things.

00:05:30.060 Sure. So you first asked about the ApoE genotype. And I have seen in my practice calls,

00:05:39.740 I would say even monthly from people who did 23andMe, and found out that they are E4 carriers,

00:05:46.540 and they wanted to know more. So I not uncommonly get calls to figure out what is it that I have ApoE4,

00:05:56.060 what does that mean for my brain? You typically see in the report, some discussion that they are at

00:06:02.060 an increased risk of Alzheimer's disease, and people are interested to figure out if there's anything they

00:06:07.260 can do. So yes, ApoE4 will be something important to deal with. Now regarding the brain itself, the brain

00:06:16.220 is a unique compartment that is mostly a lipid organ. So by weight, the brain contains a large amount of

00:06:24.620 lipids, from cholesterol, which is a sterol, to different forms of fatty acids. And the brain is

00:06:34.700 largely composed of three types of cells, although it's much more than that. But the main cells are

00:06:39.500 the neurons, which are responsible for firing and forming synapses, which regulate how the brain

00:06:48.060 functions. The neurons cannot do their own function, had it not been for helper cells. And these include,

00:06:55.420 in general, glial cells. Glial cells could be astrocytes, and these are cells which are tightly

00:07:03.340 associated with neurons, and they regulate the energy storage production and so forth that the

00:07:11.260 neuron would need to keep firing. Astrocytes have a lot of functions, but they're mostly viewed as

00:07:18.540 taking up glucose or fat and providing substrates that the neuron could use to be able to generate

00:07:27.260 ATP and keep firing. And then you've got the glial cells, the microglial cells, which have gained a lot

00:07:34.460 of traction in the last decade or so because they have been linked to neurodegenerative diseases such

00:07:40.220 as Alzheimer's disease. And glial cells are immune cells responsible for cleaning up. Whether it's going

00:07:48.380 to be a beta or an infection, these immune cells have the responsibility of making sure that the

00:07:56.460 house is in order. There's also the oligodendrocytes and other kinds of cells. But those three cells are

00:08:04.140 the main type of cells that we study in the brain. The brain is surrounded by a blood-brain barrier, and

00:08:12.300 this is an important concept because the brain is protected from the outside of the brain. And for a

00:08:19.100 good reason. The brain needs to have a stable environment to be able to function. The blood-brain

00:08:25.180 barrier regulates what gets in and out of the brain and stands as a barrier to prevent toxic proteins,

00:08:33.020 infectious organisms, and other things from getting free access. The blood-brain barrier separates the

00:08:40.460 blood vessels and the blood inside the blood vessels from the brain. And it's composed of capillary

00:08:46.700 cells. These are endothelial cells which have a lining of tight junctions. These are very tight

00:08:57.660 in the sense that they would act as a barrier, followed by pericytes which surround those

00:09:05.180 endothelial cells. And these are supporting cells for the endothelium to maintain the integrity of the

00:09:11.340 blood-brain barrier. And around those you have the mural cells or wall cells which make up the matrix

00:09:18.860 surrounding those two cell types and maintains the integrity as well of the blood-brain barrier. So

00:09:25.580 you've got this elaborate system that is designed to protect the brain and to maintain a stable environment.

00:09:34.780 Now finally, cerebrospinal fluid is in a simplistic way the sewage system of the brain. When the brain

00:09:42.620 receives its metabolites, its nutrients from the circulation, it crosses the blood-brain barrier through

00:09:50.220 the junction of the endothelial cells into brain cells. There is a fluid surrounding brain cells that we

00:09:57.980 call the interstitial fluid. Brain cells will take up whatever they need, whether it's an astrocyte,

00:10:04.140 a microglia, or a neuron, and produce some byproducts. One of the commonly studied byproducts

00:10:10.540 are A-beta proteins which are accumulating in diseases like Alzheimer's. Those byproducts will then

00:10:17.260 get cleared into the CSF. The CSF, or cerebrospinal fluid, as I mentioned before, is a form of a system that allows drainage

00:10:27.660 from the brain cells from the interstitial fluid. It's run by a pump. That pump

00:10:35.980 is in a system known as the choroid plexus.

00:10:40.140 The choroid plexus will be pumping clear water-like fluid into the CSF, and this motion of

00:10:48.380 movement of the fluid throughout the brain washes the brain off. At some point, the choroid plexus

00:10:55.260 interferes with the blood circulation to clear off these metabolites from the CSF back into the blood

00:11:04.220 for excretion. So if we take A-beta amyloid peptides, for example, they're produced by the neuron

00:11:11.580 as the neuron is firing. These are peptides which come off the membrane of a neuron, and then they

00:11:19.020 leak into the CSF, and then from the CSF they get into the blood, and the blood clears those A-beta

00:11:26.220 peptides through the liver or other organs. And this constant CSF activity is critical for maintaining

00:11:37.100 the sewage system of the brain in order. So I hope I have addressed some of the questions that you

00:11:43.820 were asking. Yeah, absolutely. And we can even sort of talk about some illnesses that occur that get in

00:11:50.060 the way of this. So for example, sometimes children are born with abnormalities that make it very

00:11:55.900 difficult for them to clear CSF, or children can be born with malformations that obstruct the flow of

00:12:04.300 CSF, and they develop conditions where some of these compartments within the brain expand and enlarge.

00:12:10.380 And it's a very intricate system. In situations like that, for example, you have to come up with other

00:12:15.340 ways to drain the system. And some of these children's require shunts where you have to place

00:12:20.140 an exogenous tube into their brain and drain it into their abdomen or something like that.

00:12:25.660 It's all such an amazing organ. I remember I had one rotation in general surgery where we did a

00:12:30.940 month of neurosurgery. And it is quite remarkable when you're operating on the brain and you remove the

00:12:37.580 dura, the brain is a bloodless organ. So it looks unlike any other organ, like when you're operating on the

00:12:44.940 kidney and you get right down to the kidney, you still see the sort of pulp-like redness of the

00:12:50.140 kidney or the lung or the liver or any of these other organs. But the brain is quite distinct in

00:12:54.620 that it doesn't have that. And once that dura is peeled back and you see it bathing in this CSF,

00:13:00.540 it looks unusually sterile. Which is not to say, of course, as you cut into the parenchyma,

00:13:05.420 you wouldn't get blood, but at that level, it's not. You mentioned something earlier, and I want to just

00:13:10.620 make sure we put it in context, which is that by weight, the brain is disproportionately lipid-laden

00:13:17.420 relative to other organs. So let's make sure people understand what that means.

00:13:22.780 Every cell in our body has a bilayer of cholesterol that makes up its cell membrane. That's what allows

00:13:29.420 it to have fluidity. That's what allows it to hold transporters across its cell membrane, etc.

00:13:36.220 How does, say, a cardiac myocyte, the muscle cell of a heart, differ in terms of its lipid content

00:13:45.740 from either the neuron or the astrocyte or the microglia? What is it about these cells in the brain

00:13:53.260 that structurally requires such an abundance of lipid and cholesterol?

00:13:59.180 That's a fundamentally important question. And it actually begs to ask perhaps a slightly different

00:14:05.900 question. How is the brain functionally different than a myocyte? And one way to look at this question

00:14:13.980 is to understand that the brain operates by firing. So you're looking at perhaps an electric system

00:14:22.860 where you have wires connected to each other. And these wires have to fire constantly. And by firing,

00:14:31.100 you need electricity to move from one piece of the wire to the other.

00:14:36.460 This requires specific cables. In our modern-day language, you need a form of fiber optics or high-speed

00:14:45.660 wiring that allows this firing to happen in instantaneous manners. And it's carefully regulated

00:14:53.740 and separated by location, by speed, and by changes in a microenvironment.

00:15:01.100 To be able to achieve this delicate and elaborate map of wires attached to each other, you need lipids.

00:15:12.540 Specifically, for example, if we're talking about a neuron, we're talking about myelin.

00:15:18.860 If we're talking about a synapse, we're talking about a synaptosone. That's a synapse body. The synaptosone

00:15:25.820 by itself to be able to conduct and depolarize, meaning to change the electric potential across the cell

00:15:33.980 membrane. It requires a certain environment of lipids that allows it to do the signaling, 0.75

00:15:41.500 allows the glutamate or glutamine to go inside the cell, and then release certain mediators to open

00:15:49.420 certain channels. For example, calcium channels. So this particular environment necessitates that you

00:15:57.260 have an elaborate system of lipids that facilitate this process. This is a very taxing system, meaning

00:16:06.700 that to be able to constantly talk to me through this podcast, you would need an immense amount of ATP.

00:16:12.940 And the extraction of glucose and the utilization of lipids is adapted to be able to answer this

00:16:20.220 particular environment. In contrast, a cardiac myocyte, outside of its being regulated by

00:16:28.140 the foci that regulates heartbeats in the atria and so forth, the AV node and other areas,

00:16:34.780 the cardiac myocyte is simply a mechanical cell that has to pump all the time to be able to maintain

00:16:42.780 a certain pressure and blood flow. So that environment by itself requires strong cytoskeletal elements,

00:16:50.780 such as actin or myosin, that are efficiently packaged for pumping and may require less lipids than

00:17:01.180 the wiring that we discussed in the brain. Yeah, so I think that's a very elegant way to explain

00:17:07.820 that basically every cell in the body, whether it be, as you said, a myocyte in the heart or a cell

00:17:15.340 in the kidney or a cell in the liver or a cell in the lung or a cell of skeletal muscle, all of those

00:17:21.100 things have a function that is contractile, gas exchange, filtration, something like that.

00:17:29.980 The brain's entire function is electron potentials and electricity. So it would make sense, as you said,

00:17:40.540 that every facet of the brain has to be optimized for signal transduction. And that includes not just

00:17:46.460 the myelination of the sheaths in the neurons, but even all of the other cells that would facilitate for

00:17:54.620 rapid potential and recurrence of normal electron gradients. The brain, I mean, I think there's a

00:18:01.420 statistic that we all learn in medical school, right? It weighs about 2% of our body weight,

00:18:06.620 and yet is probably responsible for something on the order of 20% of our energy consumption.

00:18:12.620 Again, you've provided a very elegant explanation for why that's the case. Why is it that this tiny,

00:18:19.340 tiny, tiny organ is so energy demanding? How does the brain utilize energy? So what,

00:18:26.780 let's talk a little bit about how substrate makes its way to the brain. And let's contrast it just

00:18:32.380 very quickly with, you know, pick a myocyte in your leg, right? It can take glucose out of circulation.

00:18:40.220 It can either with or without insulin, bring that glucose molecule through a GLUT4 transmitter

00:18:46.380 directly into the cytoplasm. It can, depending on the speed at which it requires ATP, it can very

00:18:54.140 rapidly take that to pyruvate and ultimately to lactate, or it can more slowly take the pyruvate

00:18:59.420 into the mitochondria and generate more ATP there. It can do the same thing with fatty acids. Of course,

00:19:05.420 with fatty acids, it's going to go from an acetyl-CoA pathway into the mitochondria and take

00:19:11.500 these two carbon units into ATP. So the muscles have lots of options for fuel. Even in a case of

00:19:18.700 starvation, they can use ketones, but basically glucose, fatty acids, lactate, all of these things

00:19:24.060 are fuels and they go directly to the cell. And then within the cell, you have aerobic and anaerobic

00:19:30.860 pathways. If you're talking about glycolysis, you sort of alluded to it already, but the brain already

00:19:36.460 has one thing that makes it an extra step, which is there's a blood-brain barrier. So it's not even

00:19:42.380 like the glucose goes directly there through the capillary. So how does the brain extract energy

00:19:49.660 from circulation and how does it differ from the rest of the body in its fuel partitioning?

00:19:56.220 I would say that I am not an expert in this field, but I'll give you my two cents. This is a complex

00:20:02.380 question and I would beg to help the listeners understand that this is a complex topic that

00:20:08.460 even as we speak, there's a huge gap in knowledge. But my understanding is that the brain prefers glucose

00:20:17.260 as the predominant source of energy. And that by itself is regulated differently than the rest of the

00:20:25.900 body. The rest of the body have a system by which insulin regulates specific transporters known as

00:20:33.900 GLUT4s, as you mentioned. And GLUT4 allows an insulin-dependent mechanism to enter glucose into pumping

00:20:44.860 muscles or adipocytes. And that responds to the outside environment, meaning you can induce GLUT4 expression

00:20:55.820 by exercising. You can induce GLUT4 expression by changes in body weight. At the blood-brain barrier,

00:21:04.780 GLUT4 plays a minor to no role in glucose uptake. The blood-brain barrier, in contrast, has a

00:21:12.220 predominance of GLUT1s. GLUT1 expression is not controlled by what you eat, it's not controlled by how

00:21:22.220 you exercise to a large extent, and it's not controlled by the same mechanisms that govern insulin-regulated

00:21:28.940 glucose signaling. Partly because your brain cannot be moody in its choices. It has to have a constant supply

00:21:38.780 of glucose. Now GLUT1 expression can be regulated, and it's largely regulated to protect the brain.

00:21:47.980 Meaning, if you are faced with a situation that is systemic hyperglycemia, such as type 2 diabetes,

00:21:56.220 the brain protects itself by reducing GLUT1 expression at the blood-brain barrier.

00:22:01.340 The opposite holds true. If you are, for some reason, going through a prolonged fasting period,

00:22:10.140 where you're dipping into glycogen stores and you're hypoglycemic, the brain upregulates GLUT1

00:22:17.900 at the blood-brain barrier to extract as much as possible all the glucose in your circulation to

00:22:25.420 maintain a relatively constant amount of glucose in the brain. Now, during physiological states,

00:22:34.620 the glucose in the brain is relatively not, does not fluctuate that much, although it does fluctuate,

00:22:40.460 but it doesn't fluctuate as much as that in plasma. During diseased states such as Alzheimer's,

00:22:46.860 for example. Leakage or destruction of the blood-brain barrier associates with destruction

00:22:52.700 in the GLUT1 transporters, and the brain now struggles in capturing glucose from the system,

00:22:59.500 which leads to a whole new cascade that we might argue is a diseased cascade. It's compensatory,

00:23:06.380 but it's responding to a glucose shortage. So in that regards, glucose regulation in the brain

00:23:14.140 differs from the circulation. Now another fundamental aspect to understand is that the brain is not

00:23:21.660 efficient in utilizing fat as a source of ATP. One thing that you notice after you dissect a brain

00:23:28.860 is that you don't find fat depots that you find in your adipose tissue. Yes, it's a fatty organ,

00:23:37.660 but there aren't really storage sites of fat. It does not store fat. To a large extent, there are lipid

00:23:45.980 droplets in the brain, but they are more dynamic than the adipose tissue that we have outside.

00:23:52.780 And it prefers not to use fat as a source of ATP. It prefers to use fat to integrate it into the myelin

00:24:01.340 sheath or the synaptosome to regulate membrane fluidity to help with depolarization. Nevertheless,

00:24:08.940 when the brain goes through crises and they can't use glucose, they have a mechanism to extract ATP from

00:24:15.660 fat. But the result is a price tag. The result is oxidative stress. The brain has a complicated

00:24:23.180 system, as we mentioned before, that consists of astrocytes and neurons and glial cells and all sorts

00:24:28.300 of stuff. And they have partitioned their roles. A neuron is like your athlete who needs to run a

00:24:36.460 marathon at a certain speed. Now, for that athlete to succeed, he needs to have water bottles staged at

00:24:45.740 every juncture or a mile. He used to have helpers with towels, helpers with whatever it is for the

00:24:53.660 athlete to be able to do the whole 23-mile if it's a marathon. Astrocytes, the helper cells of neurons,

00:25:02.300 would take up the glucose and process it to make lactate and then shuttle the lactate into the neuron

00:25:09.100 because the neuron says, I can digest lactate more efficiently than the glucose and I can produce ATP

00:25:15.900 with less damage. So the neuronal favorite food is lactate and not glucose. But glucose gets into

00:25:24.220 the astrocyte where it is shuttled and changed into lactate. When glucose is no longer there, and we're

00:25:33.580 talking about prolonged fasting and glycogen depletion, the brain becomes very efficient in extracting ketone

00:25:41.900 bodies through the blood-brain barrier to maintain its firing and to avoid someone from getting into a

00:25:47.820 coma-like state. So in a nutshell, to summarize what I just said, the brain does its best to regulate its

00:25:57.740 glucose content that is not dependent on insulin signaling. Number two, the brain prefers glucose

00:26:06.620 and not fat as a source of energy. And number three, when glucose is not available, it extracts

00:26:14.540 all the ketone bodies that are produced from fatty acid oxidation outside the brain to maintain itself

00:26:21.260 going. So a couple of other things to follow up on that. So actually, I didn't realize that GLUT1 was

00:26:26.940 basically regulated by peripheral glucose concentration. That's actually super interesting.

00:26:32.380 And I also didn't realize that the neurons were getting their lactate from the astrocytes. So I

00:26:39.980 assume the implication of that is as the astrocytes take glucose to make pyruvate, to make lactate,

00:26:49.100 the astrocytes are keeping the ATP from that process rendered to themselves while transporting or donating

00:26:56.780 the lactate to the neuron. Is that what's happening? Yeah. And then what about the microglial cells?

00:27:02.860 Are they using glucose or are they using the residual lactate? I don't know what the source of

00:27:09.020 energy for microglial cells. I am not sure how they use that energy production. And do you have a sense

00:27:15.580 just broadly what the distribution is of ATP consumption between the glial cells and the neurons?

00:27:23.260 I mean, I know that the glial cells grossly outnumber the neurons in terms of the numbers of cells,

00:27:29.180 but do you have a sense, is it like a 50-50 division of energy consumption?

00:27:33.100 I don't know. I don't want to provide an inaccurate response, so I don't know.

00:27:38.460 Okay. So we have a pretty good handle on what's going on here now. You've done a great job, I think,

00:27:44.940 giving us a primer on both the structure of the brain and now energy. So let's get into this idea of

00:27:52.460 what is ApoE? And let's talk about it not as a gene right now, but let's talk about it as an

00:27:58.540 apolipoprotein, and then we can talk about the genetics of it later. But what is this thing called

00:28:03.020 ApoE? Your interest is also in cardiovascular disease. This is a podcast in which we talk a lot

00:28:08.620 about it, so we can even contrast for the listener what ApoB is. Most people listening to this are very

00:28:14.140 familiar with what ApoB is, this apolipoprotein that wraps lipoproteins and defines a subclass of them,

00:28:20.940 the morphologically distinct lineage of the VLDL to the IDL to the LDL. But what's ApoE? What is that

00:28:29.180 apolipoprotein? That's a fascinating question, and that has been the focus of my own research for the

00:28:35.660 past six or seven years. So ApoE is, I like to give analogies, if you don't mind. Do you like music?

00:28:45.180 Sure. Before COVID, how often do you go listen to an orchestra playing the mat or any of the

00:28:53.020 fancy places? And New York has beautiful music too. Well, I'm not a fancy music guy, but I like to go

00:29:00.060 to concerts prior to COVID, but I don't know if that'll fit in with the analogy. Well, the analogy I

00:29:05.180 was trying to make is that for an orchestra to play efficiently, it needs conductors. My father

00:29:13.420 asked me a few months ago, and he's by no means a musician or any of you have never played a music

00:29:20.780 instrument, but he was telling me, what is that conductor doing? I mean, can't they play without

00:29:26.060 the conductor? So how, if I were to ask you, and if you allow me a minute to sit in your shoes,

00:29:32.000 what is the conductor doing?

00:29:33.900 That's a great question. First of all, I would assume he is vital. And secondly, I would assume

00:29:38.960 that he is somehow managing the timing of all of the different sections within the orchestra.

00:29:46.780 And not just the timing, but if one is crescendoing and the other is decrescendoing.

00:29:52.720 So I suspect there's just many more moving pieces that say a rock band where you have

00:29:57.300 one guitar, one bass, one drum, maybe a keyboard and a singer.

00:30:01.600 Absolutely. And in the lipoprotein orchestra, APO-B would be one of the principal players. Let's

00:30:11.000 assume it's your violinist. APO-A would be your piano player. And you need both, you know,

00:30:17.880 sometimes you need the violinist to run the show, and you need piano for certain pieces.

00:30:23.640 But what puts the orchestra together in the peripheral circulation, largely two conductors in this case,

00:30:30.820 APO-E and APO-C3. What APO-E and APO-C3 do is that they regulate the speed of which things are

00:30:39.240 happening. They're different than APO-B because you can argue they're promiscuous. APO-B is married 1.00

00:30:47.120 to LDL particles, maybe to some extent VLDL, LDL, and this whole pipeline. APO-A1, to a large extent,

00:30:55.860 is married to the HDL family of particles. Now, APO-E and APO-C3, also known as exchangeable

00:31:03.420 lipoproteins, are not. They can jump from different populations of lipoproteins. And by taking different

00:31:11.840 roles, they can decide whether a lipoprotein is going to stay in the circulation, and that would

00:31:17.540 be your APO-C3. Or get cleared, and that would be your APO-E. So this is a simplified version because I think

00:31:25.380 in real life it's way more complex. But the reason why APO-E has been elusive and difficult to study

00:31:31.400 is because of its multiple roles. And it appears on HDL, it can appear on LDL, it can appear on IDL,

00:31:39.320 it can appear on VLDL, it changes whether you're fasting or postprandial, and it determines to some

00:31:47.260 extent the fate of your lipoprotein. So if we take a second to imagine, you know, what happens after we

00:31:53.940 drink a milkshake. So when you drink a milkshake, you're absorbing all the fat from the milk, and you're

00:32:00.400 packaging it into chylomicrons, which are known to carry APO-B48. APO-B48 will then have to circulate

00:32:08.920 and eventually get cleared by forming remnants. As APO-B48, the chylomicron containing particle is

00:32:17.440 circulating, it starts changing its apolipoprotein composition. In the beginning, it would have

00:32:25.980 a larger amount of APO-C3, which would maintain it from getting lipolyzed. And at some point,

00:32:34.940 when it gets to a certain size, HDL will donate the APO-E and allow that remnant to get taken up

00:32:43.320 by an APO-E receptor into the liver. Typical APO-E receptors include LRP1. That also happens with VLDL

00:32:54.000 particles. When they are produced, when they are large, they acquire an APO-C3 that blocks

00:33:00.940 lipoprotein lipase. And as they shrink in size, APO-C3 will fall off, APO-E would get reassembled,

00:33:09.020 and that will assist this VLDL particle of getting cleared. One of the fundamental aspects of metabolic

00:33:18.120 syndrome is an inefficient process where these particles are, the APO-Es and the APO-C3s are not

00:33:25.500 as mobile, and you end up having an APO-C3-enriched, small, dense LDL particle that is not cleared.

00:33:34.060 So we're talking now about APO-E in the blood. So that APO-E in the blood has maybe a different

00:33:41.700 role than the APO-E in the brain.

00:33:45.500 By the way, before we go to the brain, Hussain, I'm very familiar with APO-C3 in the scenarios you've

00:33:52.200 described it. And frankly, I'm quite hopeful that in the near future, we'll have APO-C3 assays in

00:33:59.380 the periphery as an increase in our armamentarium as we try to use biomarkers to better understand

00:34:06.200 cardiovascular disease risk. And can APO-C3 levels be a part of residual risk? Once you've minimized

00:34:12.180 APO-B, will APO-C3 tell us something above and beyond it? And as you said, it tracks so closely

00:34:18.240 with insulin resistance and metabolic syndrome. I'm less familiar with APO-E in the periphery and

00:34:25.040 wasn't actually aware of just how promiscuous it was. It seems to have affinity for a number of

00:34:31.280 receptors at the liver. Doesn't it also have affinity for the LDL receptor itself?

00:34:36.060 It does have an affinity for the LDL receptor, but its affinity is higher for LRP1, which is one of

00:34:41.740 the LDL receptor family members. Yeah, APO-E is a bit more complicated than the rest of the

00:34:48.940 lipoproteins because it can take many shapes and forms and the affinities can change based on lipid

00:34:54.960 binding. It has two arms, an N-terminus, a C-terminus. Typically, the N-terminus binds to the receptor and

00:35:02.660 the C-terminus is buried inside the lipid core of a lipoprotein particle. It can dislodge and become

00:35:09.560 lipid-free or it can lipidate. And these interactions are critical to our understanding

00:35:17.320 of how APO-E functions. And the reason why we've had an investment in science in APO-E largely is due

00:35:25.060 to APO-E4, because that mutation that substitutes the arginine, the cysteine with the arginine,

00:35:32.760 changes how APO-E folds and unfolds and starts to carry lipids.

00:35:36.920 Now, we can measure the concentration of APO-B in the circulation. We can measure the concentration

00:35:45.320 of APO-A in circulation. And in fact, as you said, APO-B is so predictably linked to LP little a,

00:35:55.320 VLDL, IDL, and LDL that we effectively use APO-B concentration as a surrogate for LDL concentration.

00:36:02.460 What does APO-E concentration mean? If, I mean, to my knowledge, there is no commercial assay to

00:36:15.760 measure APO-E concentration in the periphery, but should you measure it in the lab, what would it

00:36:22.440 tell you? It's like asking me, what instrument does the conductor play? Does he?

00:36:31.100 Not at the time he's conducting. He plays them all, I suppose, but indirectly.

00:36:36.240 Right. So the APO-E concentration is elusive because it doesn't matter what the concentration

00:36:43.840 is. That doesn't tell you the same amount of information that APO-B or APO-A1 does.

00:36:51.720 There is a one-to-one correlation ratio between every APO-B particle and every LDL particle. So you

00:36:59.040 you're really closely estimating how much LDL particles there is by measuring APO-B.

00:37:06.180 APO-A is a little bit more complex. It's not one-to-one. It could be one-to-four, but you can estimate

00:37:12.560 with every four APO-A1 APO-Lipoproteins, one HDL particle. Now, when we talk about APO-E,

00:37:22.440 what are we talking about? APO-E HDL, APO-E-free, APO-E VLDL, APO-E IDL, APO-E LDL. So if you have

00:37:31.780 at one given concentration tons of APO-E on HDL, you have a completely different phenotype than if

00:37:37.920 you have it on VLDL. Because as we mentioned, the conductor is trying to now make all the orchestra

00:37:45.360 work. They have to stretch themselves and they have to make the violinist play harder, or they have to

00:37:51.060 make the piano pick up at a certain point. You can't make an assessment of the function of APO-E by just

00:37:58.120 measuring its concentrations. And to some extent, that holds true for APO-C3. Although APO-C3 at extremes

00:38:06.120 of measurements, it's more predictable and could be useful toward understanding residual risk in

00:38:11.780 metabolic diseases. But APO-E is more elusive. So what is APO-E? Again, the APO-Lipoprotein,

00:38:20.680 we'll wait until we understand this a bit better before we get down to the genetic isoforms. What is

00:38:25.840 APO-E doing in the brain? How does the conductor regulate things in the brain?

00:38:30.120 Oh, so that conductor analogy may not hold. Because we don't have our traditional LDL,

00:38:38.100 VLDL in the brain. We only have APO-Lipoprotein particles that you may argue are completely

00:38:47.600 different than what you're finding in plasma. Some group of scientists believe that APO-Lipoproteins in

00:38:55.600 the brain are HDL. But that by itself also is not accurate. They could have the size of HDL,

00:39:02.260 but the composition might be very different. So APO-E in the brain is not facilitating clearance

00:39:08.860 of lipids or the rate by which lipolysis is happening. Its role in the brain is largely

00:39:16.840 supporting the astrocytes. So astrocytes are supporting neurons. But for astrocytes to support

00:39:25.400 neurons, they need a mechanism that can efficiently crosstalk different cell types. And APO-E happens

00:39:34.960 to be able to do many different things. For example, an APO-E in a glial cell can regulate

00:39:43.220 through the exchange of lipids how inflammatory the glial cell is going to be.

00:39:49.880 An APO-E in an astrocyte can take out cholesterol from the astrocyte and give it through an LRP1

00:39:56.840 mediated uptake in the neuron to the neuron itself. APO-Es are notorious for being robust.

00:40:04.760 One of the earliest studies I have done, maybe seven, eight years ago, was cutting the nerve of

00:40:10.680 a rat model. And at that point, we did a mass spec experiment and found out one of the largest

00:40:16.600 changes in hundreds, if not thousands of proteins after you just take a neuron and cut it are drastic

00:40:23.600 changes in APO-E. APO-E exponentially goes up after damage. Whether it's a stroke, whether it's an

00:40:32.360 artificial severing of a nerve, APO-E will be released at very high rates to make sure repair

00:40:39.980 is going to happen. APO-E uses its flexibility, structural flexibility to transport lipids,

00:40:47.480 uses its structural flexibility to regulate what receptors and transporters the cell can actually

00:40:54.460 express. One fundamental difference between APO-B and APO-E is that APO-B, once it's taken up by a liver

00:41:04.980 cell, typically goes and gets a lysosomal degradation pathway. And that biology by Goldstein and others

00:41:12.800 have led to therapies, including statins and later PCSK9 inhibitors, because APO-B can get degraded

00:41:20.440 in the lysosome. And if you can figure out how APO-B is degraded, you can make a medication.

00:41:26.460 Now, one fundamental difference about APO-B from APO-E is that APO-E, to a large extent,

00:41:31.480 escapes degradation. And once it's taken up by a liver cell or an astrocyte or a neuron,

00:41:38.060 it recycles back. So it's a notorious APO-Lipoprotein that can get inside the cell

00:41:43.880 and then outside the cell, recycling most of the time. And by recycling, maybe the conductor analogy

00:41:51.960 or the thermostat analogy works. It controls or fine-tunes how much lipids there is in the cell,

00:41:57.960 and it fine-tunes a lot of different pathways which are tied to lipid metabolism.

00:42:04.400 Most importantly, the inflammatory pathway.

00:42:08.000 Now, I assume the blood-brain barrier prevents peripheral APO-E from entering the brain and vice

00:42:15.340 versa?

00:42:16.420 The dogma says yes, but our research, recent research suggests that APO-E4, but not APO-E3,

00:42:22.280 might be able to slip.

00:42:23.880 Slip through all of those tight junctions?

00:42:25.840 No, not through the tight junctions. The affinity of APO-E4 to the LRP-1 is quite higher than APO-E3

00:42:33.460 and APO-E2. And LRP-1 is expressed at the blood-brain barrier. So it could hack the system

00:42:39.000 and get through endothelial LRP-1 into the brain.

00:42:42.120 And do we have a sense of what the half-life is of APO-E? I mean, APO-B, as you said,

00:42:46.620 is a relatively short half-life. I mean, it's even without medication, its life, its residence time

00:42:52.120 in circulation is probably less than a week before it's going to get recycled vis-a-vis the liver.

00:42:58.160 How long is APO-E hanging around?

00:43:00.260 It's a complicated question. The answer to that is which particle are we talking about?

00:43:04.420 Well, I'm actually saying the APO-E itself in any form, whether it be free or bound to something

00:43:10.500 else, the actual APO-E can, it sounds like what you're saying is it can be around for a very long

00:43:15.660 period of time and take multiple forms. Yeah. So if it's a lipid-free form, APO-E can be rapidly

00:43:21.580 excreted or disappear from the circulation. If it's bound to HDL, it tends to hang longer. And if

00:43:29.620 it's bound to VLDL, it tends to hang out less because VLDL clearance rates are faster than that

00:43:36.340 of HDL. And this is a marriage. So APO-E is conferring those properties to the lipoprotein

00:43:45.320 that's actually going to. All right. Let's now talk a bit about the genetics that regulate

00:43:51.860 these different isoforms because APO-E is a protein that exists in different forms. And that's not true

00:44:01.680 of all of our genes. Now, thousands and thousands of years ago, say hundreds of thousands of years

00:44:08.100 ago, there was really only one isoform to the best of our knowledge, correct? And that's the one that

00:44:12.900 today we give the moniker E4 to. Is that correct? That's right.

00:44:18.840 So just for the sake of argument, a couple hundred thousand years ago, all of our ancestors had one

00:44:25.180 version of this gene. So you had two copies of the E4 because you have two copies of every gene.

00:44:31.320 And what do we believe was the reason for that? I mean, presumably evolution has strong pressure.

00:44:39.320 Do we have a sense of what an APO-E4? Well, actually, before we do that, Hussein, let's ask 0.64

00:44:45.680 maybe the more relevant question, which is when did that change and what was the next isoform to show up?

00:44:52.640 And do we have a sense of why? So I'm not an evolutionary biologist to give you accurate

00:44:58.140 responses, but I can probably ask the first question better than the second question. And I

00:45:03.440 think fundamentally, I was alluding to in a few minutes ago to an important inflammatory function

00:45:10.500 for APO-E. APO-E is capable of turning a cell into an acute inflammatory state, especially glial cells

00:45:21.520 or immune cells, such as the macrophage. When the cell expresses APO-E, it changes quiet substantially,

00:45:30.660 its lipid content. And it might be better equipped to become an inflamed cell. Or the opposite.

00:45:40.700 When the cell exchanges APO-E, it's also better equipped to be a less inflamed cell.

00:45:48.300 I know these are contradicting statements, but you have to realize that APO-E can regulate things.

00:45:56.340 You know, you can turn your thermostat on and off. It's not one thing or the other. It could be in

00:46:02.360 different forms depending on the situation, on the interaction, on the time. It can flip because of

00:46:08.400 its structural flexibility. Now, APO-E4, which is our ancestral variant, has a greater capacity to

00:46:17.700 aggregate than APO-E3 and APO-E2. This aggregation means that they're not suspended in solution,

00:46:26.380 favors a strong inflammatory response. So when APO-E4 starts aggregating inside a microglia,

00:46:35.920 it makes an acute inflammatory response much more efficient, and much more directed

00:46:42.920 toward what the microglia is fighting. So let's step back and imagine that we are 300,000 to

00:46:50.920 100,000 years ago. And then out of every 10,000, 20 births, few are surviving and the others are dying

00:46:58.560 from puerperal sepsis, from other forms of infection. We know now that APO-E4 women had a better chance 0.95

00:47:08.500 of giving childbirth and protection from puerperal sepsis compared to non-E4 women in the slums of 0.99

00:47:18.320 Brazil, or in certain areas in Africa. We know that, based on studies, that E4 confers some form

00:47:28.260 of an advantage. We also know that those who have had E4 may have had better luck surviving not only

00:47:39.060 parasitic infections. So in certain countries in Nigeria, people who have the E4 and ezenophilia

00:47:47.060 tend to be much healthier than E4 carriers without ezenophilia. And more importantly, non-E4 carriers

00:47:57.740 with ezenophilia. So if you are living in a place where parasites are common, and they're constantly

00:48:06.180 testing your immune system, you are much better off carrying the APO-E4 allele than not. Locally,

00:48:14.820 the macrophage is much better equipped to deal with the parasite. And importantly, your brain is

00:48:20.300 much better equipped to deal with whether it's TB, meningitis, and the mother would be better equipped

00:48:27.360 to fight a septic event during childbirth. So from that perspective, APO-E4 is our ancestral gene.

00:48:38.840 Now, what happened over the last few hundred thousand years, perhaps maybe over the last few

00:48:44.960 hundred years, and let's not go far, we turned from a septic environment to gradually an aseptic

00:48:52.380 environment. And the second change that happened is that we've prolonged our lifespan. And I know

00:48:59.200 that you have a dear interest in aging. So we have now, it's surprising to me reading from

00:49:05.640 history books that just a hundred years ago, we're not talking a hundred thousand years ago,

00:49:10.380 a hundred years ago, the mean age was something of the 47. Am I making things up? Does that make sense

00:49:15.900 to you? It depends on where you look, but it was a little more than 47, a hundred years ago in the

00:49:21.160 developed world. But you're right. I mean, I think you could say about a hundred to 150 years ago was

00:49:26.320 really when we saw a big uptick in lifespan. Right. So if that's the case, and let's assume

00:49:33.660 we're not talking about the kings and queens in England, and let's talk about those people living in

00:49:40.320 Nigeria or in Africa, anywhere, or in Brazil. And let's imagine that these people have several

00:49:47.800 milestones that they have to go through to be able to survive to their next decade.

00:49:53.100 And you would notice two things. One, that APOE4 carriers are surviving better. Two, that you're not

00:50:00.500 seeing a disadvantage of APOE4 because these people are not living to 80. Some of them would

00:50:08.500 probably die at 50 or 60 or 70, but certainly they may not have the resources to make it to 96.

00:50:18.700 So what happens to E4 carriers when they're dying at 60 or 65? I mean, what I'm trying to say is,

00:50:25.620 what doesn't happen to APOE4 carriers when they're dying at 65? They don't get Alzheimer's.

00:50:31.820 In that regards, it might provide a plausible explanation of why this ancestral gene worked for us

00:50:40.500 in the past, but may not work for us in this modern aging population.

00:50:47.400 It's always been confusing to me why there was any evolutionary pressure to create variants of E3 and E2,

00:50:56.220 because the best of my reading is that the E3 showed up about 50,000 years ago and E2 showed up

00:51:03.800 about 10,000 years ago. And again, from a functional standpoint, 10,000, 50,000, and 100,000 are basically

00:51:11.840 the same. Nobody was really living long enough to recoup the benefit of Alzheimer's disease risk

00:51:19.220 reduction going from 300,000 years ago to 50,000 to 10,000.

00:51:22.820 No, I would look at that from a different perspective. I'm, again, no anthropologist,

00:51:27.840 and my knowledge is really as mediocre as any of your readers or listeners. I would look at it as

00:51:33.000 mass movement and change in environment. From 200,000 years to 100,000 years ago, there was mass

00:51:39.820 movement from the African continents all the way to Europe, and that was associated with a drastic change

00:51:46.800 diet. And, you know, my understanding is that there was a heavily favored meat consumption,

00:51:54.800 including fish, in the Rift Valleys in Africa and in the savannas. And when they moved to a plant-based

00:52:01.840 diet in Northern Europe and where they started farming, I think that put pressure on APOE4 and

00:52:09.360 allowed APOE3 and APOE2 to be more successful. And this is a slightly different twist because APOE4,

00:52:17.280 as much as it's very strong against inflammation, it thrives on a specific dietary lipid-providing

00:52:25.620 environment. And remember that APOE4, one of the fundamental aspects that we learned from Alzheimer's

00:52:32.860 is that APOE4 makes GLUT1 less successful at the blood-brain barrier. So what APOE4 does,

00:52:41.760 it somehow tells the brain, you know, I don't want you to be eating sugar all the time.

00:52:48.240 You have to be more resilient and rely on fat. And again, this is a hypothesis. This is not based on

00:52:57.880 true hard science. But the idea is that our ancestral APOE4 diets matched APOE4. And then

00:53:06.700 when they moved to a plant-based diet, the effect of APOE4 on GLUT1 was now counterintuitive. Now

00:53:14.480 there's more carbohydrates and you want something that will help GLUT1 and not oppose it. And APOE3 and

00:53:21.200 APOE2 would actually favor a more robust GLUT1 expression of the blood-brain barrier, which means

00:53:27.080 they favor a more glucose utilization of the brain. So when did it first become apparent either to

00:53:35.000 epidemiologists or neurologists or anybody else studying it for that matter, that these three

00:53:42.960 isoforms predicted a very different risk of Alzheimer's disease in modern humans?

00:53:50.380 The story is we're in the late 80s, and this is a time where genetics is gaining traction. You know,

00:53:58.820 I recently watched a very nice series, I think on PBS Masterpieces, discussing how a revolution in

00:54:06.400 genetic fingerprinting changed how we do policing, figure out who is guilty and who's not. And that

00:54:13.240 happened. What's interesting is that this science was happening between 81 and 84. So around

00:54:20.380 88-89, a researcher at Duke by the name of Alan Roses was running an Alzheimer's group and had access to a

00:54:29.300 very robust, at that time, robust enough, obviously now it's not robust at all. But he had the ability

00:54:36.040 to genetically profile a large number of common variants of people dying with and without Alzheimer's.

00:54:44.040 At that point in time, he publishes a case series, a few hundreds, with and without Alzheimer's, and

00:54:52.380 identifies that patients with Alzheimer's had substantially higher ApoE4 compared to non-Alzheimer's. And

00:55:00.700 you know, remember that we have known ApoE4 since the 60s. So this is not a new discovery, but we didn't know

00:55:08.240 anything about ApoE4 and Alzheimer's disease until that time. So he meets a lot of backlash because the prevailing

00:55:16.840 hypothesis at that time is that Alzheimer's is amyloid accumulating and leading to the amyloid cascade

00:55:23.580 hypothesis where you form amyloid plaques and then you get tau tangles. And the field at that point was heavily

00:55:30.380 vested in this amyloid hypothesis. So Alan Roses later on, a few years later, publishes a larger study

00:55:39.680 which contains now longitudinal follow-up people, you know, progressing to disease and confirms the

00:55:47.720 observation that if you are an ApoE4 carrier, chances of getting Alzheimer's are substantially higher

00:55:55.020 than if you are a non-ApoE4 carrier. And we're talking about late-onset Alzheimer's disease, the

00:56:00.600 most common form which happens after the age of 60. So at that time, Alan Roses cemented his name as

00:56:09.040 among those who identified the association of ApoE4 with AD. Do you have a rough sense at the time,

00:56:18.720 and I think it's probably quite different today, but do you have a broad sense at the time of what the

00:56:23.580 hazard ratios were of E44 versus 34 versus 33? I don't think the hazard ratios would change

00:56:34.020 substantially in a matter of two to three decades. So what we know now is that if you have two copies

00:56:41.660 of ApoE4, your chance of getting Alzheimer's increases 12-fold. If you have one copy of ApoE4,

00:56:49.800 your chance of having Alzheimer's increases anywhere from two to four-fold. And what does that matter,

00:56:56.500 by the way, if it's a 4-2 or a 4-3? Yeah, so 4-3 means you have one copy of ApoE4.

00:57:04.660 And the question becomes a little bit difficult when you have 2-4 because 2 is supposed to be

00:57:10.280 protective. Those who have ApoE2 are protected from getting Alzheimer's disease. So the 2-4 carriers

00:57:17.520 sometimes behave S4, sometimes behave S2, and this might have to do with how well these proteins are

00:57:25.060 getting expressed inside the cells. And potentially other genes as well, like TOM40 or FGF and things

00:57:33.740 like that, that seem to also play a role here, or frankly, things that we haven't understood. Like,

00:57:39.420 you see families with a very involved family history, and they might only have one copy of

00:57:46.440 the E4, but it behaves in a very virulent way. I mean, you see the same thing in cardiovascular

00:57:51.680 disease where some families have LP little a, and the level is not particularly high, but it still

00:57:58.300 behaves very aggressively. And you'll see other families where LP little a is, I mean, astronomically 0.96

00:58:05.560 high, and yet they seem largely spared of premature atherosclerosis.

00:58:10.580 Correct. So I think one way to rephrase the question is, how do we explain that not all ApoE4

00:58:19.000 carriers have the same risk of Alzheimer's disease? And that's largely seen by ethnicity. So people who

00:58:26.820 live in Nigeria, for example, the risk of having Alzheimer's disease is substantially less

00:58:31.700 than people who live in the U.S. or in Japan. The Hispanic, Latino, Latinx population carrying an

00:58:40.480 E4 does not produce the same risk of AD compared to white people or Japanese people carrying the ApoE4.

00:58:50.140 So how do we explain that? And that's a fundamentally important question. And as you just alluded to or

00:58:57.240 noted, ApoE4 is part of a gene locus. In other words, it's part of a haplotype. That's a section

00:59:06.860 of a chromosome where a bunch of gene variants are getting co-inherited. They are within the same

00:59:15.180 locus. So ApoE4 is found on chromosome 19. Around ApoE4 is approximately 20 to 30 different gene

00:59:23.980 variants that happen to be co-inherited with ApoE4. And the description between these gene

00:59:30.880 variants are described by what we call linkage disequilibrium. If the linkage disequilibrium is

00:59:38.360 closer to 1, it means every time you have ApoE4, let's assume now we're talking about the long

00:59:45.020 version of Tom 40, you can have the long version of Tom 40 expressed with it. Now that linkage

00:59:52.080 disequilibrium is what differs between ethnicities and can largely also explain why certain ethnicities

01:00:00.060 develop disease and certain don't. So as humans or generations are getting interbred and crossed,

01:00:08.880 the fidelity of this linkage disequilibrium can start getting breached. Meaning now you can start

01:00:15.680 introducing the short version of Tom 40 next to ApoE4 or you can start introducing another version of

01:00:22.320 ApoC1 or ApoC2 or ApoC3. All these are in the same location as the ApoE together with again another 20 to 30

01:00:33.660 other genes. So that doesn't diminish the importance of ApoE4 but tells us that ApoE4 by itself may not be

01:00:41.980 sufficient to provide disease. It has to be in an environment that promotes its pathogenicity.

01:00:49.220 And all things equal, from a clinical perspective, we are more worried when someone of Asian or 1.00

01:00:56.860 Caucasian ethnicity has it versus someone of Latin or African descent. Is that a fair assessment? Again,

01:01:06.060 it's an oversimplification but just based on the epidemiology and also based on frankly potentially

01:01:12.980 what you said earlier which is the origin of the gene, right? This is a gene that takes its roots

01:01:20.680 very early in our ancestry and offered an important protection and potentially by the time we migrated to

01:01:29.360 a more northern climate, some of that protection was less necessary. Again, that's a teleologic sort of

01:01:35.640 hand-waving rationalization but is that a safe assessment that everything you're saying risk is just

01:01:43.400 going to be higher in Caucasians and Asians? I wouldn't go with a blank statement. Maybe in general,

01:01:51.700 yes, but maybe there are what we call either gene-gene interactions or gene-environment interactions

01:01:57.560 that have to take into account. Let me give you a simple example. In a Colombian study, they show

01:02:03.700 that ApoE4 by itself may not substantially increase the risk of Alzheimer's disease but since diabetes

01:02:11.340 is prevalent in a certain city in Colombia, diabetics who were ApoE4 carriers had substantial

01:02:17.840 increase in AD risk compared to diabetics alone or ApoE4 carriers alone. And this is an example of a

01:02:25.880 gene-by-environment interaction. So just to be sure we got that right,

01:02:30.540 E4 with diabetes was higher risk than E4 without diabetes and higher risk than diabetes without E4.

01:02:39.580 Yes.

01:02:40.320 By way of comparison, do you recall what the risk was of E4 without diabetes to diabetes without E4,

01:02:47.600 which was second?

01:02:48.800 I don't recall the numbers, but I can tell you they were not striking at all. So when I actually

01:02:53.160 looked at the data, the E4 risk by itself seemed barely above non-E4 carriers. So if you looked at

01:03:01.500 it and you'd say this is a healthy Colombian who is farming all day long, lean, and they have

01:03:07.940 absolutely no metabolic diseases, the E4 is doing nothing.

01:03:11.820 Yeah. You mentioned something earlier that I thought was super fascinating. I want to come back

01:03:16.660 to it, which is some of the diet interactions. So let's go back to where you left it, which was when

01:03:21.800 talking about it through the lens of the E4 carrier. The first thing you mentioned is, look,

01:03:26.760 E4 makes GLUT1 less successful at the blood-brain barrier. And you've already established for us

01:03:32.440 that GLUT1 is basically regulated by peripheral glucose concentration. Its purpose is to make sure

01:03:39.280 that when glucose is in low abundance, the brain gets first dibs on it. And that when glucose is in

01:03:45.760 overly high abundance, the brain is protected from it. So now you have an E4, you have a person who's

01:03:51.540 an E4 variant, they're less sensitive to that mechanism. Presumably that means in an environment

01:03:58.020 richer in glucose, they're less able to appropriately partition fuel. Is that a safe assessment?

01:04:05.860 Yes. You can argue that in an environment that is rich of glucose, or maybe in a different term,

01:04:12.280 an insulin resistant environment, ApoE4 carriers are less capable of regulating by many mechanisms,

01:04:20.660 including GLUT1, the fluctuations in glucose, and they're more susceptible to disease. But

01:04:25.740 by no means, this is a restricted relationship of ApoE4 to GLUT1. Because we also know from studies

01:04:32.960 that we have done and others, that ApoE4 also affect the transporters that transport omega-3 fatty

01:04:39.320 acids into the brain. Now we're going to come to that in a moment, because that of course is,

01:04:44.200 I mean, in many ways, one of the most interesting things to now talk about. But before we go there,

01:04:50.900 I do want to come back to this idea of, do we have an understanding of type 2 diabetes, which is

01:04:59.320 simply a very extreme version of insulin resistance and metabolic dysregulation?

01:05:03.720 Does it disproportionately then, I hate to use the word, but punish carriers of ApoE4 versus E3 versus

01:05:12.640 E2? Is that clearly established in the epidemiology? I don't think so. I wouldn't look at it as type 2

01:05:23.900 diabetes is a distinct disease that punishes ApoE4, because type 2 diabetes to me is a syndrome,

01:05:30.300 not a disease. So what are we talking about? And it's frankly a pretty significant spectrum

01:05:35.380 for that matter. Correct. So are we talking about an individual who is consuming a large

01:05:40.960 amount of carbohydrates, who gets type 2 diabetes? Are we talking about ApoE4 itself? Because as you

01:05:47.640 probably may know, ApoE4 by itself creates a system with aging that makes an individual insulin

01:05:54.480 resistant. So in one regards, ApoE4 could be a factor for why a certain person may get type 2

01:06:03.100 diabetes. So in that regards, ApoE4 is not conspiring with type 2 diabetes. It might be behind

01:06:10.040 type 2 diabetes. So you have to look at type 2 diabetes from a more individualized causal. What's

01:06:18.500 causing type 2 diabetes here? Is it the diet? Is it genetics? And how is ApoE4 interacting with this

01:06:25.980 phenotype is probably more complicated than an A1c and a genotype in a large population. Because

01:06:33.840 when you look deep enough into these studies, you find quite a bit of discordant results that makes it

01:06:40.380 hard to explain. You sort of alluded to it earlier that the E4 carriers, as they migrated

01:06:49.740 ancestrally, say out of Africa or even out of South America, were primarily consuming a diet lower in

01:06:57.640 carbohydrates, presumably higher in protein and fat. Do we have any reason to believe today

01:07:03.100 that there is a rationale for matching diet to genotype?

01:07:11.580 So right now, in major studies that looked at ApoE4 dietary patterns in the Western world,

01:07:18.460 they don't differ by genotype. So if you're looking at the US or Europe, an ApoE4 carrier

01:07:24.400 may well likely be eating a similar diet to a non-carrier. Now, would that lead to the same detrimental

01:07:30.340 effect? Yeah, yeah. Sorry. My question is less about the existing dietary pattern and more of

01:07:35.560 the prospective ask of, is there a reason to change dietary pattern to produce a better outcome?

01:07:43.000 Yes, of course. So this is part of what we're studying is how does ApoE4 interact with the diet

01:07:50.820 and how that interaction can change disease processes. And what we know is that the first concept that

01:07:59.140 your listeners should be aware of is that ApoE4 is a disease of aging.

01:08:04.360 Do you mean Alzheimer's disease is a disease of aging or?

01:08:07.800 Alzheimer's disease is a disease of aging, but ApoE4's risk of developing Alzheimer's and aging-related

01:08:13.700 diseases is fundamentally based on the aged model. Younger 35-year-olds, ApoE4 carriers, to my knowledge,

01:08:22.940 do not have frank presentations of a disease. A younger ApoE4 carrier, who could be myself or you or

01:08:32.080 anybody else, could be fully functional. It's only when you hit a certain age that ApoE4 carriers

01:08:38.280 start to have problems. And not all of them. So the majority of ApoE4 carriers, you may argue,

01:08:46.160 are surviving, they're thriving, they're doing okay. So there's only a subset of ApoE4 carriers,

01:08:52.940 who are aging and getting disease. So then the question is, who are these people? How do we

01:08:58.120 explain them? Is it only genetics? Or is there a dietary genetic interaction? And I think the answer

01:09:04.900 to this question has to do with what is happening with these people as they age? Are they developing

01:09:12.660 other diseases? Are they, for example, we know that an ApoE4 carriers who happen to be the victim of a

01:09:20.720 traumatic brain injury has much worse outcomes than that without traumatic brain injury? We know

01:09:26.680 that an ApoE4 carrier, possibly with type 2 diabetes, may have worse outcomes than an ApoE4 carrier

01:09:34.340 without. But again, as I mentioned to you, this is a little bit more complicated because not all type

01:09:39.660 2 diabetes is the same. We know that an ApoE4 carrier who may have a second hit, could be a genetic

01:09:47.380 second hit. They're inheriting another protein that increases the risk of Alzheimer's disease.

01:09:53.220 Now their path toward disease is much more accelerated. So what we also know is that as we age,

01:10:01.800 our ability, our cells lose the ability to regenerate and to sustain a certain status.

01:10:10.420 And that is associated with lapses in energy production. That is associated with several

01:10:18.740 complications, which include, at the blood-brain barrier, a reduced ability to express those

01:10:25.960 GLUT1 transporters, omega-3 transporters, and even ketone body transporters. So the aging ApoE4

01:10:35.880 individual who has second and third hits is now straining the brain and leading to an environment

01:10:44.740 where, in this situation, the diet may make a difference. Another angle to look at this question

01:10:52.820 is, there are certain diets, and coming from the cardiovascular background, we both agree that

01:10:58.820 there are certain diets that can accelerate aging or create a form of stress that can

01:11:05.820 be the second hit that ApoE4 carriers are exposed to, to accelerate toward an aging brain that is

01:11:14.180 diseased. I know the answers may not be crystal clear, but the diet interaction with ApoE4 is

01:11:22.740 contingent upon aging and a second or a third hit. But what seems interesting in what you just said to me

01:11:30.740 is a big part of the manner in which ApoE4 transmits its risk to the individual is through an energy

01:11:43.240 crisis. Much of what you said had to do with substrate. It had to do with reducing the utilization

01:11:53.100 of and or access to substrate. Is that a fair assessment? That is one hypothesis. It's a prevailing

01:12:01.280 energy hypothesis that links ApoE4 with disease and aging. There are other hypotheses which do not have

01:12:10.380 to be exclusive of this hypothesis, meaning that you could also have an energy slash inflammation

01:12:19.400 inflammation differences that predispose ApoE4 to neuroinflammation. And that by itself could lead

01:12:27.680 to an accelerated disease pathway. You could also have an ApoE4 backslash. It could be a vascular

01:12:35.980 component here as well. Yeah, vascular or leakage component. And there's a group at USC where I work

01:12:40.840 that has shown that there is leakage in the blood-brain barrier allowing toxins to get into the brain and

01:12:46.720 shrink cells. And that is by itself non-exclusively related to the dietary or the inflammation.

01:12:55.540 We do not want to say this is it and that's only it. It is plausible that there are multiple hits,

01:13:01.380 but you could look at it from different angles. Well, not only that, I mean, those hits are synergistic

01:13:06.620 and in the wrong direction, right? So if you have leaky blood-brain barrier and you're more predisposed

01:13:12.740 to inflammation, which again, as you pointed out, was very helpful 300,000 years ago when we were

01:13:19.040 fraught with parasites, you now have two hits within the same hit, which is you're more likely to get

01:13:24.740 toxins across the blood-brain barrier, which is bad in and of itself. And then secondly, you're more

01:13:29.260 likely to have an overreactive immune response to it. And if you couple that with say a third hit of

01:13:34.920 neuronal energy starvation, these things begin to circle. So let's pivot for a moment now to

01:13:42.280 another enormous area of focus for you, which is the role of omega-3 fatty acids. This is such an

01:13:50.900 interesting topic. There's nobody listening to this who hasn't heard of omega-3 fatty acids. And even if

01:13:56.220 they haven't heard of them in exactly those terms, people have certainly heard the terms EPA and DHA.

01:14:01.680 And there's no shortage of confusion about these things. So I guess for the odd person who maybe

01:14:08.120 just doesn't know exactly what we're talking about, can you give just a brief overview of what EPA and

01:14:12.800 DHA are and why we spend so much time and energy trying to study and understand their role in human

01:14:19.060 health? There isn't really a short answer to a complicated question like this one. But in a nutshell,

01:14:25.840 when you break down the components of the brain, you find that a significant portion, in some studies

01:14:33.860 50 or 40 percent of the brain is composed of polyunsaturated fatty acids such as DHA, EPA,

01:14:42.400 and arachidonic acid, AA. The EPA, DHA, and add to that alpha-linoleic acid, ALA, are known as omega-3s.

01:14:53.340 In contrast, arachidonic acid is in omega-6. So the brain is highly enriched in both omega-3s and omega-6.

01:15:02.580 And you'd contemplate why. Why do we have so much of these fatty acids in the brain?

01:15:08.140 And the fundamental answer has to do with membrane fluidity. Having polyunsaturated fatty acids in

01:15:16.560 neuronal membrane facilitates to a large extent the neuronal firing that we just talked about in the

01:15:23.360 beginning. So this composition of fat allows the brain or the neurons to conduct their work very

01:15:31.880 efficiently. So DHA is the predominant building block, the most commonly found omega-3 in the brain.

01:15:41.580 EPA is not that common. DHA, or docosahexanoic acid, is given that name because of the number of fatty

01:15:49.400 acids and the number of double bonds. It's six. EPA has five of, and we're talking about the location of

01:15:58.860 three double bonds in the structure of DHA or EPA. EPA is not very abundant in the brain, but that

01:16:09.020 doesn't mean it's not very important. EPA is less abundant, but has perhaps stronger anti-inflammatory

01:16:15.780 effects than DHA. There is limited interconversion between DHA and EPA, but DHA can become EPA.

01:16:24.400 And also EPA can become DHA, but that process is also limited. The reason people talk about omega-3s

01:16:33.980 and they make a big deal out of it is because the human body doesn't have an efficient system

01:16:39.200 to make them from scratch. In fact, you can't make them from scratch. If you want to make them,

01:16:45.940 you need a precursor known as alpha-linoleic acids, or ALA. By a group of desaturases,

01:16:52.440 ALA can get transformed into DHA and EPA. However, it is widely believed that only 0.5%

01:17:00.880 of DHA and EPA are made from conversion of ALA. That number, though, you have to take it with a

01:17:08.960 grain of salt because there might be evidence that in diets, vegetarians or vegans who do not eat fish,

01:17:16.000 that conversion rate may actually be much higher. But despite that, that conversion rate has not been

01:17:21.780 documented to exceed 5%. So at the most, 5% of DHA and EPA can come from ALA, although most people

01:17:30.380 argue that it's only 0.5%. So what does that mean for us? If we can't make DHA and EPA,

01:17:39.540 and we rely on diet to get them, what does it mean that we are not getting them?

01:17:44.480 So what I've just alluded to is the U.S. diet. And you probably know as much as I do even more,

01:17:52.420 that the U.S. diet is not enriched in DHA or EPA. The consumption of DHA and EPA in the U.S.,

01:18:01.240 I can give you an example, DHA is averaging 100 milligrams per day, is low. That consumption

01:18:08.180 may not provide enough omega-3s to the brain. And the question is, does that result in disease?

01:18:18.620 And the answer is likely, yes. Now, what we don't know is the supplementing people with omega-3s

01:18:28.380 make a difference. And the answer to that question is, at this point, we don't know that that makes a

01:18:36.080 difference. Simply because research on supplements have not panned out. It's very conflicted. There are

01:18:44.360 positive studies, there are negative studies. And the supplements themselves are quite distinct and

01:18:50.580 different in production and concentrations and quantities and qualities. So I know I have alluded to

01:18:57.260 a lot of different aspects of this field, but the answer to your question is that these omega-3 fatty

01:19:04.260 acids are important for the brain. We can't make them efficiently, and we're not consuming enough of them.

01:19:11.420 And as you also alluded to, when we supplement them, they can come in various forms, right? We can

01:19:19.440 get them from microalgae, which tend to be triglyceride-based. We can refine them from fish

01:19:27.040 oils directly, which are ethyl esters, or they can come from krill oils, which are phospholipid-based.

01:19:34.600 Does that sound about right?

01:19:36.980 Yes, yes. You can get omega-3s from all these sources.

01:19:41.260 It basically has to be marine. I mean, that's sort of the bottom line, is it has to have some sort of

01:19:47.100 tie to algae or microalgae, either being consumed by other things like fish. And it's funny, I sort of

01:19:54.140 remember many years ago reading something very interesting that explained why EPA and DHA had to

01:20:01.340 have some sort of a marine origin predominantly, and it had to do with the formation of the third

01:20:07.300 carbon double bond. And I can't remember any more than that. Does that ring a bell, anything about this?

01:20:13.240 I can tell you that the complexity of the structural distribution of these double bonds

01:20:18.980 requires a certain machinery to make them that is present in algae. Now, why wouldn't other species

01:20:27.180 make it? I don't know. Yeah, it had to do exactly with the very, very complex carbon fixation that only

01:20:35.980 algae could do. Let's start with what we know. You have a pretty good sense that people who have

01:20:43.220 higher levels of EPA and DHA from consumption of fish have better outcomes than people who don't.

01:20:50.960 Do we know that pretty well? I don't know. I honestly don't know if this is very clear.

01:20:59.300 What my knowledge is, is that the opposite might be closer to truth, meaning that people who don't eat 0.78

01:21:06.480 at all fish or seafood or omega-3 rich diets might be at a higher increased risk of disease,

01:21:15.780 as opposed to the other statement, which means people who eat a lot of them have less disease. So

01:21:21.500 what I'm trying to point out is that there could be a deficiency state that predisposes somebody to

01:21:27.620 disease. But once you meet that threshold, may or may not get much more benefit.

01:21:32.600 Yeah. So in other words, you can get scurvy from insufficient vitamin C, but there's probably

01:21:39.420 little evidence to suggest that supplementing vitamin C offers health protection and benefit

01:21:44.380 beyond the RDA. By the way, for what it's worth, I believe that's true with vitamin C. I think there's

01:21:49.420 no evidence to suggest it's helpful in massive doses. Nevertheless, there's been no shortage of

01:21:57.080 attempts to study this question. And there have been some significant studies that have gone to

01:22:03.180 great lengths to do this. Now, although we're talking about the brain, you are no stranger to

01:22:08.780 the heart. So can we pivot for a moment to talk about a very interesting study published probably

01:22:16.000 about 18 months ago that looked at a very high dose of EPA, four grams in individuals with elevated

01:22:23.860 triglycerides and a number of other risks. And I have to be honest with you to my surprise produced

01:22:30.360 kind of an amazing outcome. It was not what I expected. Of course, I'm speaking about the reduce

01:22:35.920 it study. Do you want to give people just a quick overview of that study and what it found and what

01:22:41.540 your thoughts are on that study?

01:22:42.640 So the background is some Japanese studies over a decade ago has suggested that a certain EPA to

01:22:50.820 arachidonic acid ratio was associated with substantially less cardiovascular disease in

01:22:58.880 certain regions in Japan. And there were some trials in Japan at that point that have attempted

01:23:06.540 supplementing EPA. And the results were positive, but they were not conclusive. So they were using

01:23:16.460 something around two grams of EPA. And they found out some trends which were very exciting. But again,

01:23:23.280 they were not strong enough to make an official recommendation. So somehow, this led to a concerted

01:23:31.000 effort in the US to try to answer this question. Can we give high doses of EPA to mimic this ratio that

01:23:41.700 they found in Japan of EPA consumption and translate into less cardiovascular events? So it became an

01:23:51.100 interesting question to pursue. And the reduce it investigators decided to go ahead and run a large

01:23:59.900 multi-center trial where they gave four grams of pure EPA. And they selected people who they thought would

01:24:09.200 benefit the most from this intervention. And those included people with diabetes, high triglycerides, because

01:24:15.580 EPA is known to lower triglycerides. And they monitored them for cardiovascular events over a period of a few

01:24:22.860 years. And they found out that compared to their placebo arm, those who were given

01:24:29.720 4 grams of EPA per day, that's substantially better. The study had one flaw. And I don't know if it's a fatal flaw

01:24:38.640 or not. I think that needs to be figured out. The placebo in that arm was not your typical placebo.

01:24:49.040 So typically, when you do omega-3 or fatty acid intervention trials, the most common placebo is corn oil,

01:24:56.700 soybean oil, soybean oil. And in this case, so they used mineral oil. And they were criticized at the

01:25:04.160 time of publication, whether EPA indeed reduced cardiovascular events or did mineral oil increase

01:25:11.780 cardiovascular events.

01:25:13.420 Before we leave that, let's understand that a bit more. So corn oil, safflower, canola,

01:25:18.260 those are more E6 than E3. They have a little bit of ALA in them, and they still have some monounsaturated

01:25:25.260 fats in them. But is it safe to say that they're probably more than 50% omega-6?

01:25:30.340 Yes, but the concentrations given in them is not high. The concentration of E6 is really a low amount. These people

01:25:38.440 are not consuming grams of E6.

01:25:40.380 And what's mineral oil's breakdown? What's it formed from? I mean, my only interaction with mineral

01:25:46.540 oil is to polish my cutting board, you know, or to make sure my cutting board doesn't get too dry.

01:25:52.760 Yeah. I mean, to be honest with you, I don't know. But all of this came after the fact. We've never

01:25:58.880 considered in our trials mineral oil because it was never an option. And it was very obscure that that

01:26:05.780 mineral oil was chosen in that trial to be the placebo. And I don't think they even thought about

01:26:11.600 this until the study was published. And then people started asking.

01:26:16.360 I see. Yeah. And I mean, I think mineral oil, frankly, is totally different. I mean, I don't,

01:26:21.240 it's probably more resembles like a petroleum-based product or something like that. I don't think

01:26:26.580 it's like a fatty acid in the sense that corn oil is or anything like that. Yeah, that is odd

01:26:31.920 that they would use that. But they were given presumably four grams of mineral oil every day

01:26:36.800 as well in a placebo. Right. And I think the decision to make or to create mineral oil as the

01:26:44.520 placebo at that point in time was not thought of from a health perspective, but mostly from maybe

01:26:51.900 what's available or economic or matching the taste of the intervention itself. So there wasn't deep

01:26:59.760 thinking about what would happen if you added so much mineral oil in a study like this one. And I

01:27:06.920 think the investigators have regretted that decision, but, you know, hindsight is always 20-20.

01:27:13.000 Yeah. And it's certainly a plausible explanation given just how significant the effect was in

01:27:19.480 reduce it. I mean, it was, wasn't the absolute risk reduction something on the order of 4% or

01:27:27.480 something like that. I mean, it was in excess, it was certainly on par with what you see in some of

01:27:31.980 the most potent drug trials. Yes. Yes. And this, this was surprising. And what's important is in

01:27:39.160 science, you always have to be skeptical and you always need to find whether any particular study

01:27:47.420 can be replicated or not before you make strong recommendations and adopt any drastic change in

01:27:55.640 how you are consuming certain things or whether it's a medication or a diet or a supplement.

01:28:00.740 So the company was able to make very strong claims that high doses of EPA are important addition to

01:28:10.420 the standard management of diabetes, cardiovascular disease, atherosclerosis at the time of its publication

01:28:16.760 and even urged the FDA to add that indication, which the FDA appeared to agree to.

01:28:22.900 But at the same time, other studies were going on the pipeline to confirm, although they did not use

01:28:32.940 the same identical ethyl ester EPA, because that is perhaps patented by the company that made

01:28:41.140 that particular drug. And I don't want to mention any particular drug in the show because I'm not

01:28:47.040 trying to promote anything. So long story short, what we have learned from the American Heart

01:28:52.380 Association a few months ago is that high doses of EPA did not translate into cardiovascular benefit

01:28:59.880 when compared to corn oil or the other standard form of placebo.

01:29:06.260 Now you're referring to the STRENGTH trial, correct?

01:29:08.620 Yes.

01:29:09.300 Now, was that a high dose of EPA or was that EPA and DHA combined?

01:29:15.360 It was predominantly a high dose EPA, although I think there was DHA in it. The predominant form

01:29:22.280 that was given was EPA.

01:29:24.200 And what was the dose? That was four grams omega-3. Yeah, I think you're right. I think

01:29:31.380 it was probably about three grams and one gram is my recollection.

01:29:35.840 Yeah, I can look it up. I do not want to give your listeners any inaccurate information. So if you

01:29:40.820 like, we could just look up, but it was a predominantly EPA supplement.

01:29:44.780 Yeah. And as you said, when compared to a corn oil, that study showed essentially zero difference

01:29:50.220 between them. So again, what's different, right? So it begs the question, is the mineral oil the

01:29:57.160 problem in the first study? Was the dose not correct or the ratio of three to one of EPA to DHA?

01:30:06.120 Is it simply irrelevant? To your point earlier, maybe these people didn't have a deficiency

01:30:11.880 and supplementing to excess made no difference. The other thing is these were relatively sick

01:30:18.320 patients, right? These were patients that I can't recall off the top of my head what the

01:30:23.280 strength patients looked like, but if they were anything like the reduce it patients, which I

01:30:27.880 remember more, I mean, almost 80% of these patients had type two diabetes. All of them had

01:30:33.040 dyslipidemia. So the other argument is, is four years of treating patients that are at high, high

01:30:39.980 risk of major adverse cardiac event, enough time to steer the ship? Are you better off treating people

01:30:45.440 in their forties for, you know, two decades to try to mitigate that risk? And I think, you know, frankly,

01:30:51.620 that question probably becomes even more interesting from the standpoint of brain health and heart

01:30:56.000 health, right? Where we have fewer options.

01:30:59.000 Let me tell you that how I think about this, and it may or may not make sense to you or to your

01:31:04.960 readers. When we're talking about reduce it or EPA, are we talking about a supplement or are we talking

01:31:13.080 about a drug?

01:31:14.300 In the case of reduce it, it's a drug. I mean, that's a pharmaceutical prescribed drug.

01:31:18.800 Right. So this distinction is important because if we're talking about a drug,

01:31:23.160 drug, then we are thinking about a cascade. We're thinking of, you give a statin, you inhibit

01:31:29.220 HMG-CoA reductase, you decrease the expression of the, or you increase the expression of the LDL

01:31:36.060 receptor in the liver, and then you suck cholesterol circulating by LDL or VLDL particles, and you excrete

01:31:43.920 it, you lower LDL cholesterol, you decrease small dense or atherogenic LDL cholesterol particles

01:31:51.160 that translate into less atherosclerosis, and then this translates into less cardiovascular

01:31:56.780 events. This is the prevailing hypothesis of how we view statins work. Now, if I were to ask you,

01:32:03.900 how does EPA work? How does it reduce cardiovascular events? And if we can come up with a cascade

01:32:11.180 similar to that of statins, I think we can argue that you could use high-dose EPA as a drug.

01:32:16.380 But if we're stuttering and we can't really map the actual cycle, we might be fishing for results.

01:32:24.740 Yeah. And it's also interesting when, I mean, I don't think this analysis has been done, but

01:32:29.840 it would be a clever study, would be like a pairing study where you take the same patients, of course,

01:32:38.340 and you pair one group to phenifibrate to another with high-dose EPA such that they get equal reduction

01:32:47.600 of triglycerides. And you ask the question, does that make a difference? In other words, how much

01:32:52.700 of this benefit, if any, comes from the triglyceride reduction, which all things equal, is going to

01:32:57.980 lower ApoB concentration as you reduce the lipid carrying capacity lipoprotein? And did that, you

01:33:04.320 know, is that the mechanism? Now, again, that's more of a thought experiment. I can't imagine anybody

01:33:07.900 would go and do the actual study of trilipics versus facepa, but you're right. We don't have

01:33:15.200 a very clear mechanistic pathway for why this would make sense.

01:33:21.720 One possibility, although I'm not trying to be against or for EPA, I'm just trying to think

01:33:28.160 out loud. One possibility is that EPA has potent anti-inflammatory effects. And it is plausible

01:33:34.680 that EPA, not by any triglyceride lowering potential, but by a local anti-inflammatory effect

01:33:41.220 on the plaques, has resulted in less plaque rupture and less events. But again, these are all ideas,

01:33:47.460 because we don't fully understand what happened and why they did get substantial reduction in events.

01:33:53.980 Now, this is irrelevant because, you know, to me, I don't study right now cardiovascular disease, and

01:33:59.780 I'm not trying to understand exactly an application for EPA on patients at risk of heart disease. But my lab

01:34:11.640 is more interested in figuring out a link between omega-3s and the brain. And the reason why I

01:34:18.660 brought to reduce it, or we brought this example of high-dose EPA in cardiovascular disease,

01:34:23.700 is to contrast completely different systems. Unlike cardiovascular disease, where you have a very

01:34:31.820 well-described process, where you have a plaque that ruptures and you get an event, your typical

01:34:37.920 Alzheimer's dementia is often a chronic, low-grade, slow process that happens over years, and that

01:34:47.100 eventually manifests in loss of function of daily activities, cognitive deficits, and so forth.

01:34:55.420 The role of DHA in this disease is more closer to the analogy that you used with scurvy.

01:35:04.660 In landmark studies done decades ago at NIH, they showed when you took a mouse model and you deprived

01:35:12.200 the pregnant mice from DHA completely, and then you looked at their offsprings, they found that

01:35:20.100 these baby litters, the offsprings of the mouse, had a condition that we call microcephaly, where the

01:35:27.980 brain failed to form and develop and was very small. When you looked under the microscope, you found

01:35:34.540 out that these neurons formed from a DHA-deficient diet. We're not making enough synapses.

01:35:42.200 DHA-deficients under the microscope compared to neurons who were exposed to a deficient or a

01:35:48.880 non-deficient or a regular chow diet when the mother was nursing or having the children in utero.

01:35:56.480 So that by itself told us that DHA-deficiency is detrimental to the function of the brain.

01:36:04.560 Now, we also know from many studies that in humans during development, not having enough

01:36:14.740 polyunsaturated fatty acids, whether it's lactating breast milk or in the diet itself, was associated

01:36:22.640 with poorer outcomes in school. And these are large studies published in the 90s, which have led the

01:36:29.040 FDA in 2001 or 2002, I believe, to agree to the recommendation of supplementing infant formula

01:36:36.920 with both DHA and arachidonic acid, AA. So, you know, any listeners who has bought infant formula or

01:36:46.060 have used infant formula, and if they look at the actual formula, they see the tag DHA and AA fortified.

01:36:53.360 So, the concept that there is something called DHA or EPA or omega-3 deficiency exists is probably

01:37:04.340 solid. Now, what's less known or more confusing or more difficult to ascertain is that what age range

01:37:13.760 does this deficiency make the largest impact? And how does aging affect that deficiency question?

01:37:22.260 So, to put this question in a different perspective, the brain develops rapidly from conception to

01:37:29.600 possibly three to five years. And by the age of six, the human brain is almost fully developed. And then

01:37:36.180 you start slowly accreting. Accretion refers to the tension of lipids and material such as protein in

01:37:45.240 the structure of the brain, where it's not exchanging with plasma anymore or blood. So, it's kind of stuck

01:37:51.320 in the brain. So, the accretion of lipids or omega-3s in the brain slows considerably between the ages of

01:37:58.420 six all the way to 12 and 13. And then beyond that, the pools are relatively stable and expansion of

01:38:05.380 brain size. Brain size is very slow. And then after certain age cutoffs like 60s and above, you start

01:38:12.520 seeing the opposite where atrophy starts to happen with aging. So, we can clearly visualize that figure

01:38:19.220 and grasp the concept that during rapid accretion of lipids, it's really important for, you know,

01:38:26.620 babies and kids to get exposed to enough omega-3s to allow the brain to fully develop. And that doesn't

01:38:33.100 have to be supplementation. That could well be a good diet. Now, what happens between the ages of six

01:38:39.480 and 60 is quite elusive, meaning that what happens to a population that is 35 that is completely not

01:38:48.300 consuming any omega-3s? Are they in any diseased states? And I can't point out a single large study

01:38:57.520 that can say yes. All what I can point out to is a myriad of small studies which show discrepant

01:39:05.680 results. Some studies suggest there are subtle cognitive impairments. Other studies suggest anxiety

01:39:13.140 disorders, mood disorders, depression. But other studies don't find these associations. So, that tells us

01:39:21.440 that the role of omega-3 in the diet between the ages of six and 60 is more difficult to understand.

01:39:31.600 Now, that doesn't mean it's not important. Is there any scenario in which having lower consumption

01:39:37.880 of EPA and DHA is beneficial in those small studies? Or is this basically a difference of neutral

01:39:44.980 to negative? I think it's the latter, neutral to negative. But I think the concept that I haven't

01:39:52.960 discussed yet, but might be important to understand, is that the half-life of lipids in the brain is quite

01:39:59.900 different than the half-life of lipids in different compartments in the body. So, for example, if you take

01:40:06.700 DHA or EPA and you study after somebody is injected or consumed a label DHA or EPA, how long does it take

01:40:15.720 for the DHA or EPA to disappear from blood? You understand that it takes a matter of weeks. Within

01:40:22.060 three to six weeks, that dose will disappear. And the opposite is true. It takes three to six weeks

01:40:28.360 for the dose to reach plateauing or saturation kinetics after being fed a maximal dose of DHA or EPA.

01:40:35.340 So, that has to do with the half-life of these lipids on the transporting proteins such as HDL. So, HDL may

01:40:45.360 define the half-life because it's going to be the major carrier of phospholipids. Now, when we talk about

01:40:52.800 other compartments such as the brain, that may not be the case. The amount of time that DHA hangs out in

01:41:00.660 neuronal membranes is substantially longer than the amount of time that DHA hangs out on the surface of

01:41:08.300 a phospholipid on HDL or with an albumin carrier in blood. And, you know, if we think about this more

01:41:15.440 deeply, that makes sense. We do not want brains that are fluctuating with DHA and AA. We want a stable

01:41:21.880 the pool of lipids that does not dramatically fluctuate. So, what does that mean? That means

01:41:29.320 if you took somebody in a trial where they're not consuming DHA or EPA, and then you gave them

01:41:35.800 high doses of DHA and EPA, and, you know, if you look at these trials, they range anywhere from 12

01:41:41.560 weeks to a year, and the majority are three months or six months, it's not surprising that these trials

01:41:47.600 are finding not much. And in fact, when they find something, most of the time, it's either a gut

01:41:53.200 effect or an inflammatory effect. Because some of these omega-3s have anti-inflammatory effects.

01:41:59.660 So, that presents a distinct challenge to the omega-3 field, is that we may not have the resources to do a

01:42:06.480 five or a ten-year study. But we are privy to epidemiology studies, which can give us a longer

01:42:14.280 vision of what happens to those who consume enough omega-3s versus those who don't consume omega-3s at

01:42:22.600 all. And then I'll just layer in the question to bring it back to our discussion of APOE.

01:42:28.720 Do we have a flavor to add to this, which is, how does everything you say change in the context of

01:42:37.140 APOE type, if at all? It does. So, this is what my lab is particularly interested in solving.

01:42:44.280 Which is, what is the best diet that we can give an APOE4 carrier to actually prevent cognitive

01:42:51.020 decline? So, we've been doing studies to examine this from multiple angles. And one of the studies

01:42:58.560 that we found is that we took younger 35-year-old APOE4 carriers and injected them with labeled DHA

01:43:05.040 into their blood. And we estimated how much DHA is getting into the brain using a PET scan.

01:43:11.360 And what we found is that younger APOE4 carriers had greater uptake of DHA in their brain compared

01:43:20.040 to non-carriers. Meaning that a younger cognitively normal APOE4 carrier is sucking all the DHA it can

01:43:28.280 take from blood to maintain a certain form of cognitive profile, to maintain your normal cognitive

01:43:38.020 behavior. And APOE4 carriers love the DHA it's seeing in blood. And the brain of an E4 carrier is on fire.

01:43:48.160 It's firing more and it is using the DHA that is sucked from the blood. And that could probably provide us

01:43:55.140 some deep insights about our ancestors and their diets. I mean, we know that APOE4 carrier rate is highest in

01:44:02.020 Africa. And we know that E4 carriers in Africa were consuming possibly more fish and meat than E4

01:44:08.500 carriers in Asia or E4 carriers in Europe. So in that regard, we looked at epidemiology studies and

01:44:15.840 found, you know, a relatively large epidemiology study from Finland suggesting that E4 carriers who

01:44:22.220 consumed a good amount of fatty fish combined with red leafy vegetables, antioxidants, when you looked

01:44:32.100 at these individuals two decades or more into the future when they were 60, those who were consuming

01:44:38.660 the highest amounts of omega-3s from fish and other sources in the diet had substantially and significantly

01:44:46.480 less diseased than those who were not. So to us, that gave us insights that perhaps the younger APOE4

01:44:56.960 brain loves to consume omega-3s. Now, I hope the story ends here and, you know, that gives us a nice

01:45:05.920 recommendation, but it doesn't. Because we were continuously studying APOE4 carriers across the age

01:45:13.360 spectrum. And what we found is that after a certain age, and I would say this goes anywhere

01:45:19.120 from 55 to 70, the ability of the APOE4 brain to capture APOE4 from blood gets compromised. And this

01:45:28.200 happens around the same time that the GLUT1 receptors are compromised. This happens around the same time

01:45:34.980 that the blood-brain barrier itself starts to get compromised. Because you have to imagine that the

01:45:40.160 blood-brain barrier is tightly coupled. So when it starts disintegrating, your GLUT1s are not working,

01:45:47.420 your omega-3 transporters are not working. So what does that mean? So we were part of a large

01:45:53.180 randomized clinical trial where patients with mild Alzheimer's disease were fed high doses of DHA

01:46:00.760 versus placebo over a period of a year and a half. And what that trial found is that in this trial,

01:46:07.660 two grams of omega-3s had no effect on everyone. Zero effect. The trial was negative, was published

01:46:15.020 in 2010 in JAMA. And the lead author was Joe Quinn. When we looked at that trial, we tried to understand

01:46:23.620 how did APOE4 affect the response. And in that trial, APOE4 carriers had zero response. And non-carriers

01:46:32.340 appeared to improve on the primary outcome, the ADAS-COG, and on the secondary outcome, which is the MLSC.

01:46:38.660 So in that trial, APOE4 carriers were less likely to respond than non-carriers.

01:46:45.580 And again, the age of those participants was what?

01:46:48.120 These people were between 65 and 80.

01:46:50.680 Got it. And was that DHA only or EPA and DHA in that two grams?

01:46:54.760 That was DHA only.

01:46:56.360 And what you said earlier about the young people massively assimilating DHA, was that also shown

01:47:02.360 with EPA or only DHA?

01:47:04.220 We were only studying DHA for the main reason is that the DHA is the main omega-3 that makes the

01:47:10.560 building blocks of synapses in the brain. And EPA more behaves as a signaling molecule that's

01:47:16.160 anti-inflammatory. So an application for EPA in the brain may be quite different than DHA.

01:47:22.060 So if you're more focused on vascular disease and perhaps the small embolic or thrombotic

01:47:27.760 atherosclerotic strokes, then an EPA application may make more sense. But if you're looking at the

01:47:34.040 synapse and what it requires to form neurotransmission, we think from a DHA perspective.

01:47:39.600 So what you just described fits with the previous observation, which is there may be a critical

01:47:45.980 window in which DHA is essential. And once you're outside of that window, you lose the ability to

01:47:55.080 integrate or assimilate DHA at about the same time your blood-brain barrier is failing and your ability

01:48:01.240 to regulate even glucose is beginning to diminish.

01:48:05.840 Absolutely. So you hit it right on. We are actively doing a large trial that we have called

01:48:12.380 PreventE4 to partially address this question. If we took younger ApoE4 carriers before they have

01:48:18.560 dementia or before they have clinical symptoms, can we show any effect on the brain? And we had this

01:48:26.680 immensely challenging question of should we look at 30 to 40 year olds or should we look at 55 to 70

01:48:34.340 year olds? And in 2017, we made the decision to look at the 55 to 70 year olds because those people

01:48:42.800 do cognitively decline with time. So we have the opportunity within two to three years to figure out

01:48:48.700 if they're dropping. If we did this trial in a 35 year old, we may end up with a futile study because

01:48:55.540 neither arm would change. So we are running a big risk though. It is plausible that between the ages

01:49:01.880 of 55 and 70, you're still too late. So where do you hold this? Where do you go is unknown, but

01:49:08.460 PreventE4 is ongoing. It started in 2019. And unfortunately now because of COVID, we had to hold

01:49:14.820 and Los Angeles had a substantial share of COVID cases. So we have, we had to hold recruitment in the

01:49:21.320 last six months. But hopefully, you know, we hope to open recruitment again in the next few months. The

01:49:27.320 study will possibly wrap up in 2023 and get published between 2024-2025. So that's an ongoing effort where

01:49:35.600 we're giving two grams high dose DHA based on a pilot study that we have done before showing that you need

01:49:43.300 higher doses of DHA to get into the brain. In APOE4 carriers who do not have cognitive disease yet,

01:49:50.260 they may have mild disease, but definitely not dementia. And to see whether APOE, you know,

01:49:55.140 high dose DHA supplementation can slow down the progression of disease in this population.

01:50:00.560 So again, of course, the experiment that can't be done is taking the 35 year old E4 carriers,

01:50:06.840 putting them on DHA for the rest of their lives. Because you're basically giving them DHEA at the

01:50:14.520 time when you know they can assimilate it and following them for long enough to see a hard

01:50:18.440 outcome, which is, is there a meaningful reduction in the actual disease we care about studying?

01:50:23.700 Since that experiment can't be done and will never be done, we then have to basically think about this

01:50:30.880 through the lens of this is a question that medicine does not have a tool to solve through the

01:50:35.460 neat package called the clinical trial. Yeah, the way to approach this problem would actually be

01:50:41.580 finding surrogate biomarkers that you believe and trust that will hold long and behold. For example,

01:50:50.300 we now, maybe in the last five years, have finally agreed that LDL cholesterol is a good target to lower

01:50:57.680 for cardiovascular disease prevention. Although every now and then you'd get somebody who shows you

01:51:03.480 lower LDL cholesterol may not translate into cardiovascular events. But for the most of

01:51:08.840 people in the field agree that lowering an LDL cholesterol by a statin or by azetimibe or by any

01:51:15.020 other measurement can associate with less atherosclerosis. Can we find a biomarker in 35-45 year old

01:51:23.720 individuals, a brain biomarker, where if you change it, you would convince most people that this may

01:51:30.360 protect you against dementia or cognitive decline a couple of decades. That's a fundamentally important

01:51:36.140 question that needs to be tackled because without brain biomarkers, we may not be able to answer these

01:51:43.040 complicated questions. I mean, I'm a bit skeptical that we're going to find anything anytime soon

01:51:49.160 that fits exactly that description. And I think it's worth noting that one of the first drugs ever

01:51:54.700 approved to lower LDL did so, but did not reduce events. I'm blanking on the name of the drug, but it

01:52:03.160 actually was a drug that inhibited the enzyme that converted desmosterol to cholesterol. I think it's

01:52:11.220 something called delta-60 saturase or something. And in doing so, it did lower cholesterol and by

01:52:18.020 extension, LDL cholesterol. Although frankly, at the time, they probably only looked at total cholesterol,

01:52:23.060 but events actually went up, presumably because desmosterol went up so much and acted as a sterol

01:52:30.220 that had the same oxidizing properties. So in the end, events matter. But you're right. Look,

01:52:36.220 most drugs today that lower ApoB also reduce events. But in the absence of such a biomarker

01:52:41.920 or functional study, it doesn't have to be a biomarker, I guess. If you had an imaging study or

01:52:47.720 a functional study that could matter, hippocampal volume, if we believe that that was

01:52:52.820 sensitive enough. But absent that, it still comes down to a question of what's the asymmetry of

01:52:58.620 risk on each side, right? So option one is do nothing. Option two is either through diet or

01:53:06.760 supplement, take a given amount of DHA. What is the downside of the latter? What is the downside of

01:53:12.560 the former? I think those are questions that we have to spend more time thinking about because I don't

01:53:18.040 think that clinical trials can give us answers to some of the most important questions facing disease

01:53:23.400 prevention.

01:53:23.840 Yeah. And so this is exactly where diseases like Alzheimer's disease or vascular dementia or any

01:53:30.380 form of dementia represents a deep painstaking challenge compared to atherosclerosis where you

01:53:36.880 can have an event within a couple of years and still see an effect on plaque reduction,

01:53:43.640 lower event rates, whether it's MI or amputations or strokes. Dementia, on the other hand, is much more

01:53:49.940 subtle, much more complicated and requires a completely new set of tools. And as we speak,

01:53:57.280 there's a massive amount of brain imaging studies from functional MRIs to PET scans that are looking

01:54:03.500 at different tracers, including amyloid and tau, which to date have not really panned out in the AD

01:54:09.080 field. But with the premise that we can find a biomarker that if you can modulate, can predict

01:54:15.200 something a decade later. So one more thing that I just want to share with you before we wrap up is

01:54:20.900 that, you know, my lab has also done very exciting studies that will be published in the next two to

01:54:25.840 three months, showing that the older ApoE4 brain, once it gets into the dementia stage, starts

01:54:34.220 upregulating enzymes that facilitates its own autodigestion. Because of the failure in utilizing

01:54:41.840 glucose as a source of energy, we find that enzymes like phospholipase A2 are upregulated and activated in

01:54:51.180 the ApoE4 brain to perhaps extract fatty acids from its own myelin sheath to produce ATP.

01:54:59.540 And that tells us about that the recommendations that we should give the younger ApoE4 carrier may be

01:55:07.740 completely different than a recommendation that we give an older ApoE4 carrier who is in a state of energy

01:55:14.620 deprivation and energy failure.

01:55:17.460 Say a little bit more about that. I mean, you're talking about someone who now actually is in the stages of

01:55:22.400 early cognitive decline, who presumably, I mean, when you say autodigesting, I mean, you're referring to an enzymatic

01:55:30.740 degradation of the cell?

01:55:34.460 Yes. So what we did is in the last three years, we've looked at enzymes that control omega-3s and omega-6s

01:55:42.500 in the brain. And we found out that a specific enzyme known as phospholipase A2, which is the calcium

01:55:49.340 dependent version of phospholipase A2, is strongly upregulated in Alzheimer's disease brains who carry the ApoE4

01:55:58.900 genotype compared to non-Alzheimer disease brains who carry the ApoE4 genotype. So we had access to

01:56:08.240 a large database of brain tissues from Rush. Rush has one of the largest Alzheimer's disease research

01:56:15.000 centers in the country. And we looked at many enzymes and stumbled across CplA2 and showed that

01:56:22.600 CplA2 is hyperactivated or phosphorylated in A4-AD brains. So you've got the ApoE4 carriers who developed AD as opposed to

01:56:32.300 the ApoE4 carrier who died without AD. And we found that the ApoE4 AD brain activates these enzymes which take away

01:56:42.360 fatty acids from the phospholipid membrane and then perhaps degrades the fatty acid to be used for many

01:56:50.940 reasons. To us, the most obvious reason why you are extracting fatty acids from phospholipid is possibly

01:56:58.640 to generate ATP. Now, there is a side effect to doing that, and that would be neuroinflammation and oxidative

01:57:04.320 stress.

01:57:04.700 And I assume that this is something you're not seeing in the non-E4 carrier who is in the comparable

01:57:12.000 stage of dementia?

01:57:13.940 We are not seeing the same pattern of activation in a non-E4 carrier who has a comparable state of

01:57:20.440 dementia. That's correct.

01:57:22.120 Now, there's a lot of people out there who are talking about things like ketone supplementation,

01:57:26.780 so not necessarily using ketosis in its nutritional or starvation form, but even using exogenous ketones

01:57:33.220 as a supplemental fuel, is there any reason to believe that that or a nutritional approach to

01:57:41.040 ketosis would be a benefit to somebody with E4, especially when they're in that second wave?

01:57:48.500 Or is the answer, well, anything that's going to prevent glucose from getting into the brain

01:57:53.220 sufficiently is probably going to prevent ketones from getting there too?

01:57:57.020 You have fantastic insights and questions. So I think people have thought about

01:58:02.720 ketone meals, ketone body meals for treatment of dementia, and the largest studies to date have

01:58:09.660 shown that E4 carriers, once they have the disease, are reluctant or less likely to benefit from this

01:58:16.360 supplementation compared to non-carriers. And exactly to your point, it is possible that the

01:58:21.820 degenerating blood-brain barrier is compromising the transport of ketone bodies the same way it has

01:58:27.760 compromised glucose and the same way it has compromised omega-3 fatty acids. So that makes us think that the

01:58:35.500 aging Alzheimer's disease brain is a late stage that is refractory to the, unless you are trying to fix the

01:58:45.300 fundamental issue that is related to nutrient transport by a medication or an intervention. Dietary

01:58:52.380 interventions at this stage may not be very effective and highlights the importance of getting the right

01:59:00.320 intervention before you get to that stage to prevent it from happening.

01:59:05.200 I hate being asked questions like I'm about to ask you, so forgive me, but you know that everybody

01:59:11.040 listening to this, and by definition about 25% of these people listening to this are going to be a

01:59:15.720 carrier of at least one variant of E4. So this is a very relevant question. What do you recommend to

01:59:22.380 people? Let's just limit this to people who are, you know, have no stage of disease. I mean, you've

01:59:28.380 certainly made the case for DHA. Yeah, so I don't recommend omega-3 supplementation for a simple reason

01:59:34.560 we don't have the evidence to actually support. Once we publish or others publish high-quality

01:59:41.880 evidence supporting going to a pharmacy and picking up a supplement, until that happens,

01:59:49.120 I don't think we have the evidence to support omega-3 supplements. But what I do recommend,

01:59:54.480 though, is for people to take at least one serving of fatty fish per week. We know from epidemiology,

02:00:01.240 APOE4 carriers who consume one serving of fatty fish per week may be providing enough omega-3s to

02:00:09.320 their brains and helping themselves long-term. Is there anything else that you have learned in

02:00:15.580 your travels vis-a-vis exercise, other things within nutrition? I know that's not the focus

02:00:21.460 of your research, but do we know anything about the E4 carriers being more benefited, if such a term

02:00:27.780 exists, through exercise? I mean, we know the importance of exercise in Alzheimer's prevention

02:00:32.000 is enormous, but do we know if E4s bend one way or the other in that wind?

02:00:37.420 We don't have high-quality randomized clinical trials, but we may have some epidemiology studies.

02:00:44.140 Morris and his team at WashU have shown in a PET study, this is an imaging study where you scan

02:00:50.560 amyloid load in the brain, that APOE4 carriers who exercise have less amyloid plaque buildup

02:00:58.500 compared to APOE4 carriers who do not exercise. Now, again, we would need an intervention to

02:01:06.300 demonstrate that, but as we discussed in a minute, it's very difficult to find younger

02:01:12.780 APOE4 carriers, put them in a lifestyle intervention, and then look at the outcome three to four to a

02:01:20.220 decade later. Logistically, it's complicated and expensive. We also know from a Swedish group

02:01:26.840 that looked at multiple factors of what determines pathology in APOE4 carriers, and they concluded that

02:01:35.660 a high level of education, hypertension control, protected APOE4 carriers from dementia compared to

02:01:44.360 APOE4 carriers who had less education and worse blood pressure control. So, to answer your question,

02:01:51.500 I think if you knew at birth that you had APOE4, if your parents knew, and you have the ethnicity,

02:02:00.720 the risk, the family history, the best advice that we have is to maintain an adequate omega-3

02:02:06.240 consumption from fatty fish, not supplements, to make sure that the blood pressure is controlled,

02:02:12.800 and to make sure that there is some at least minimal amount of exercise carried out throughout

02:02:20.460 the lifespan to provide a modifiable element to the APOE4. And when we talk about modifying APOE4 risk,

02:02:29.580 we often are referring to modifying the vascular effects of APOE4. So, not only will you get

02:02:36.220 cardiovascular benefit, but you will also get brain benefit, because the blood vessels in Alzheimer's

02:02:43.420 disease have as much of a role as the actual neurons and astrocytes and so forth.

02:02:50.080 Hussain, this has been very interesting. Again, I think a lot of people are going to find this to be

02:02:54.720 relevant to themselves or to people that they care about. I won't lie, it's a little frustrating that

02:02:59.780 we're still so early in the study of this, something that is so relevant, and I believe

02:03:06.140 probably quite preventable if we had a better sense of what the interventions were. And I think

02:03:10.060 you're absolutely right. Biomarkers and or other functional imaging studies would really speed

02:03:16.100 things up. If we could do clinical trials in two years and get answers, we'd be a lot better off than

02:03:22.140 if we had to study people over the course of their lifetime or rely on epidemiology, which is

02:03:25.980 fraught with so many challenges that make it very difficult to disentangle causal relationships.

02:03:32.220 But nevertheless, this has been illuminating and I thank you greatly for your generosity of time

02:03:37.140 and effort. Absolutely. Thank you for inviting me. It was a pleasure talking to you, Peter.

02:03:41.540 Thank you for listening to this week's episode of The Drive. If you're interested in diving deeper

02:03:45.780 into any topics we discuss, we've created a membership program that allows us to bring you more

02:03:50.520 in-depth exclusive content without relying on paid ads. It's our goal to ensure members get back

02:03:55.960 much more than the price of the subscription. To that end, membership benefits include a bunch

02:04:01.180 of things. One, totally kick-ass comprehensive podcast show notes that detail every topic,

02:04:06.600 paper, person, thing we discuss on each episode. The word on the street is nobody's show notes rival

02:04:11.940 these. Monthly AMA episodes or ask me anything episodes, hearing these episodes completely.

02:04:18.060 Access to our private podcast feed that allows you to hear everything without having to listen to

02:04:23.340 spiels like this. The Qualies, which are a super short podcast that we release every Tuesday

02:04:28.680 through Friday, highlighting the best questions, topics, and tactics discussed on previous episodes

02:04:33.400 of The Drive. This is a great way to catch up on previous episodes without having to go back and

02:04:38.560 necessarily listen to everyone. Steep discounts on products that I believe in, but for which I'm not

02:04:44.220 getting paid to endorse. And a whole bunch of other benefits that we continue to trickle in

02:04:48.700 as time goes on. If you want to learn more and access these member-only benefits, you can head

02:04:53.020 over to peteratiamd.com forward slash subscribe. You can find me on Twitter, Instagram, and Facebook,

02:05:00.360 all with the ID peteratiamd. You can also leave us a review on Apple Podcasts or whatever podcast

02:05:06.880 player you listen on. This podcast is for general informational purposes only and does not constitute the

02:05:12.620 practice of medicine, nursing, or other professional healthcare services, including the giving of

02:05:17.920 medical advice. No doctor-patient relationship is formed. The use of this information and the materials

02:05:24.000 linked to this podcast is at the user's own risk. The content on this podcast is not intended to be a

02:05:30.320 substitute for professional medical advice, diagnosis, or treatment. Users should not disregard or delay in

02:05:37.660 obtaining medical advice from any medical condition they have, and they should seek the assistance of their

02:05:43.180 healthcare professionals for any such conditions. Finally, I take conflicts of interest very seriously. For all of

02:05:50.160 my disclosures and the companies I invest in or advise, please visit peteratiamd.com forward slash about where I

02:05:58.740 keep an up-to-date and active list of such companies.

02:06:13.180 Thank you.

02:06:18.660 Thank you.

02:06:33.640 Thank you.

The Peter Attia Drive - February 01, 2021

#147 - Hussein Yassine, M.D.： Deep dive into the ＂Alzheimer's gene＂ (APOE), brain health, and omega-3s

Episode Stats

Summary

Transcript