At his annual checkup, a ninety-two-year-old man said, “Doctor, do you think I’ll make it to a hundred? I don’t drink, smoke, eat rich food, or have a lot of sex.”
His doctor asked, “So why do you want to live to be a hundred?”
Studying the biology of longevity and aging has led to some surprising and promising insights about how and why we age and, more important, how we can age later and slower. With each new discovery geroscientists make, we get closer to creating a future in which the golden years are truly golden. As a physician with a background in molecular genetics and endocrinology, I am well versed in the medical problems that plague the elderly, but until I met my wife’s grandmother Frieda, I had never deeply questioned the biology of longevity. Her vitality as she grew older was in such stark contrast with people twenty years younger, so I couldn’t help wondering: Just what is aging?
We know that our bodies, organs, and even our cells experience wear and tear, consistent with a law of physics—the second law of thermodynamics. Like appliances, all these physical objects break down over time. We also know that environmental factors like what we eat and drink, whether we exercise or smoke, and how well we sleep may slow or accelerate aging. So we have a variety of interesting theories about the root causes. One of the more controversial theories suggests that aging is programmed—that our cells receive biological signals that tell them when to deteriorate and when to die. At an event on aging at Gordon Research Conferences, I led a panel discussion about whether this is true—whether our bodies are programmed to age and, if so, whether they can be unprogrammed. Even colleagues who are typically provocateurs were subdued, with few wanting to defend the notion that the indignities of aging are inevitable. One scientist, Hong Gil Nam from South Korea, presented illuminating data showing how tree leaves are programmed to age, not because of wear and tear or external environmental factors but because the tree itself sends out a signal telling the leaves to change color, die, and drop from the branch. I can still remember the silence that blanketed the room after he delivered his conclusions. What did this mean for humans? Are we programmed to decay, too? The implications were unsettling. But as I gave it more thought, I realized that the leaves are programmed to die, but the tree itself is not, and neither are we. Although our cells can activate in response to stress and aging by way of a program to die, known as apoptosis, or stop dividing, if they are damaged and cannot be repaired, that does not directly program our death.
Jan Vijg at Einstein conducted a study with Vera Gorbunova, University of Rochester; Laura Niedernhofer and Paul Robbins, University of Minnesota; and others that showed that in many species, as we age, more of our cells develop mutations. Genomic instability—a high frequency of mutations—is one of the hallmarks of aging, and this team showed that mutations accumulate in part because repair mechanisms are increasingly impaired. If repair is impaired, cells can die or stop dividing, leading to a decrease in the sizes of organs. And if the repair mechanisms fail, cells can become cancerous.
At least part of the problem lies in the mitochondria, whose number and function decline with age. As the mitochondria suffer free radical damage and accumulate damage in their own DNA—which is commonly believed to be the primary culprit in the decline—their energy production may become impaired and cause apoptosis. Mitochondrial dysfunction is associated with many health risks, including heart, kidney, liver, and gastrointestinal diseases. Decline in mitochondrial numbers and function is also implicated in declines in vision, hearing, and skin condition. And impaired mitochondrial function is associated with major metabolic conditions, such as obesity and diabetes.
In our initial studies with Ashkenazi Jews, we did not find any differences between centenarians’ mitochondria subtypes and control groups’ mitochondria subtypes, but Joseph Attardi, California Institute of Technology, had previously observed a mitochondrial mutation that occurs much more frequently in Italian centenarians than in the rest of the Italian population. This prompted us to analyze our own population more deeply, and while we could not confirm the genetic finding from Italian centenarians, we did find an aging-related increase in the incidence of another mutation that was not inherited and occurred by chance during early development. This finding alerted us to a new biology of aging related to the mitochondria, which we’ll cover in chapter 5.
Evolution has faced a lot of challenges and perfected our biology through trial-and-error experiments over billions of years. So when scientists make a discovery, we first ask, “Why did evolution create it this way?” Evolution prioritizes reproduction, so that may be why the aging process picks up speed after reproduction. While it’s true that men can reproduce well into their seventies and beyond, the majority of men fathering babies are in their twenties. Once we’ve passed our DNA to our children, there’s nothing else we can do to contribute to the evolution of our ability to age no matter how long we live. However, there’s a really interesting hypothesis about that. Based on a combination of biology and sociology, the “grandparent hypothesis” suggests that orphans (whose parents generally weren’t resilient enough to withstand the aging process and died relatively young) don’t do as well as people who grew up with parents and grandparents. The grandparents who had survived were biologically resilient, and because of that, they had time to accumulate wealth and wisdom and have more children and grandchildren. So in evolutionary terms, the DNA that allowed the grandparents to age better had the effects of populating and strengthening the biology for the next generations. That’s one example of the many aging theories out there that are correct, but it’s important to remember that there’s much more to the story.
One of my favorite explanations for aging is the antagonistic pleiotropy theory, which suggests that biological resources we need for reproduction when we’re young may end up hurting us when we’re old. In 1957, George C. Williams offered this hypothesis that rapid aging is the price paid to achieve better reproduction. For example, we need cholesterol to build the membrane of cells, including those associated with reproduction, so it’s possible that people who have high cholesterol levels are going to have more reproductive success. But after their children are born, this cholesterol can eventually turn against them, harm their blood vessels, and increase their risk of heart diseases and stroke. On the other hand, we need cholesterol throughout our lives to build cell membranes and for healthy brain function. But scientists have found that nature provides many examples of exchanges between longevity and reproduction that support this theory. In numerous research studies conducted with a variety of animals, the animals that reproduced did not live as long as those that were prevented from reproducing. And the animals with the most offspring died first. So it looks like there is an exchange between reproduction and longevity, but the evidence is far from conclusive. For instance, there are studies showing that increasing reproduction in fruit flies extends their life span.
GENOTYPE
A genotype is a change in the sequence of a gene (also known as a variant or SNP). It is important if it is linked to a single trait like red hair, or a disease like diabetes.
When we began our Longevity Genes Project, we were still curious about the exchange between reproduction and longevity. I jokingly proposed that maybe for humans, it’s more about raising kids than birthing them. Research conducted by Alan Shuldiner, endocrinologist, geneticist, and vice president of Regeneron Pharmaceuticals, has shown that the more offspring that Amish parents have (up to thirteen), the longer the parents live. A friend and colleague, Alan has conducted genetic studies with the Amish because they have only two hundred to three hundred founders. It’s likely that there’s a social explanation for the results, considering that in such a society the more kids you have, the more kids are available to take care of you as you age.
In our medical history survey, we ask the offspring of the centenarians and the members of the control group how many children their parents had and found that, besides having them later in life than their peers who lived to an average life expectancy, the centenarians had significantly fewer children—both the men and the women in our study had on average from one-half to one fewer child than the average. And since the centenarians—who have longevity genotypes—have fewer offspring, the number of people who possess a longevity genotype may have decreased with every generation. While income, education, and other factors affect birth rate, the centenarians in our study are more alike than different when it comes to education and income levels. They also did not have access to very effective birth control methods, so there seems to be a lot of evolutionary pressure to eliminate longevity genes. With this in mind, it’s possible that the ages of Abraham—175—and Moses—120—recorded in the Old Testament and the Torah were true. Not to mention Methuselah, who’s said to have lived to 964, and Noah, said to have reached 960. Maybe we’re losing longevity genes because of a low rate of reproduction among those with exceptional longevity (Abraham had only two sons, one delivered when he was one hundred years old).
We also wanted to see if there was any correlation between longevity in women and the age of menopause. As I looked through the stack of completed surveys on my desk, I noticed that nearly all the female centenarians had begun menopause exactly at age fifty. Wow, this is impossible, I thought. And there were too many cases to be a coincidence, so I called our nurse, Bill, who had administered the surveys, to ask if he had any insight into why so many of the women started menopause when they were exactly fifty.
“Well, I ask them when they stopped having their menses, and most of them don’t remember,” he said. “So I say, ‘Fifty?’ And they say, ‘Yes, fifty.’”
We had a good laugh about that, but it’s a perfect example of why research survey questions are standardized and need to be asked a certain way, which we eventually did once we homed in on the questions that were most important for us to ask.
When I tell people that I study centenarians, one of the questions I get is: “If I want to live to one hundred, should I move to a blue zone?” Blue zones are the places with the world’s longest-living populations, as identified by National Geographic fellow Dan Buettner. He found these populations in Loma Linda, California; Ikaria, Greece; Okinawa, Japan; Nicoya, Costa Rica; and Sardinia, Italy, and he looked for commonalities among the people there to see what lifestyles and backgrounds contributed to the longest lives. These locations don’t all have more centenarians on average than other places, but the people who live there stay healthy for longer and live longer on average than people in other locations.
Buettner found nine common denominators among the people in the blue zones:
All these common denominators can contribute to a longer health span and life span, but I think the most significant common denominator is genetics. For example, in Sardinia, there are three villages that are very close to each other that have a population of one-hundred-year-old males. But there are many other small villages nearby that have the same environment—the same climate, trees, and pretty much everything else—that don’t have populations that live longer than others on average.
The same situation exists on Ikaria. There are more centenarians on the island per capita than on nearby islands, which have the same climate, plants, and animals. So while blue zones may show an interaction between genes and environment, unless the genetics of these populations are studied, we can’t know whether it’s the environment, the genes, or the interaction of the genes with the environment that’s responsible for the longevity. I think the blue zones are great, and there’s good reason to believe that living in a blue zone may help to increase health span and life span. But no matter where centenarians live, they are usually genetically unique, and we need to keep in mind that not everyone who lives in a blue zone makes it to a hundred years old.
So the jury is still out on the major underlying factors that contribute to blue zones, but they’re not the only hot longevity topic to make the news. Dr. Elizabeth H. Blackburn, who coauthored The Telomere Effect with Dr. Elissa Epel, was awarded a Nobel Prize for discovering the molecular nature of telomeres, which are DNA extensions at the ends of chromosomes that carry no genetic information but keep the chromosomes compact and protected, like the tips of shoelaces, and for codiscovering the enzyme called telomerase. As we age, telomeres become shorter, and Blackburn and others promoted the hypothesis that this shortening drives aging. In our studies at Einstein, we’d found that centenarians had longer telomeres than people who were eighty-five years old, and the centenarians’ offspring also had longer telomeres than their control groups.
Blackburn often mentioned our high-profile study because it showed some alterations in the genes of telomerase, although we did not know their functional importance. That seemed to support the theory that the longer our telomeres are, the longer we live. But we need to do long-term studies to better understand this conclusion, because we don’t know if the people who had longer-than-average telomeres at age eighty-five started out with longer ones or if they became shorter more slowly than other people’s. Maybe our centenarians started out with telomeres of the same length as the average person’s but they didn’t lose as much length over their lifetimes. Or maybe their telomeres were much longer than average when they were born. With more research, I think we’ll find that when people age more slowly, the length of their telomeres is longer because they started with longer telomeres at birth or had less telomere “attrition.” So while having long telomeres may predict good health that doesn’t mean they cause longevity.
In a lab next to mine at Einstein, Dr. Ron DePinho studied mice in which telomeres are made shorter by knocking out the telomerase enzyme—the enzyme that elongates telomeres—and the first generation of these mice did not age more than those that were not manipulated. On the other hand, lengthening telomeres by telomerase overexpression will usually cause cancer. Blackburn carefully measured average telomere length in the blood of thousands of people and showed that those with the shortest telomeres were at risk for heart disease and those with the longest telomeres were at risk for cancer. Interestingly, mice have much longer telomeres than humans, and Ron argues that his mice did not get old because their shortest telomeres were longer than humans’. When telomeres are manipulated to become shorter over generations, the mice do age more quickly. However, the problem with this argument is that while mice age in ways similar to humans, they do it in a much shorter time period and die at around three years of age with longer telomeres, suggesting that telomeres are really not a major factor in aging for animals or humans.
So it seems that even though the health of centenarians and their offspring is characterized by longer telomeres, having average telomere length is optimal for most of us. Also, telomere length can be rapidly regulated by stress, which shortens them, and prayers, which lengthen them, suggesting flexibility in their length even if they become shorter on average. So measuring the length has no predictive value.
One of the latest theories of aging—and my favorite—is presented by my friend and colleague David Sinclair in Lifespan: Why We Age—and Why We Don’t Have To. The information theory of aging proposes that we age and become more susceptible to diseases because our cells lose information. DNA stores information digitally, but the cells have an analog format that can modulate the function of genes in the sequence of the DNA. The analogy David uses to explain this theory is that of a CD player (or, for us older folks, a record player). The digital information is the “song,” and the scratches on the surface of the disc represent the accumulated effects of aging in the DNA. Another way to think of it is that the DNA, including longevity genes, is the hardware and epigenetics is the software. And the question we’re asking is how to remove the scratches. We can find the mechanisms of aging through observing and experimenting with the changes in the digital information. The analog information is changing all the time because of our interaction with the environment. It’s the epigenetic information that tells a cell whether it’s a liver cell or a hair cell. When we grow old, cells become confused because epigenetic mechanisms are scrambling the information. That’s why fat cells start showing up in the liver and hair cells develop instead of skin cells. The last example is why older people often have hair sprouting from their ears and other places where it didn’t used to grow.
Methylation is an epigenetic process—specifically, a chemical reaction between DNA and a unit of organic compounds known as a methyl group—that can either activate or deactivate genes. Each of the body’s cells have exactly the same chromosomes, but we don’t need to use every one of the genes in every cell. Methylation helps to decide what type of cell is formed, such as a fat cell, a liver cell, or a cell for another specific part of the body.
Currently at Einstein, Gil Atzmon, my first postdoc fellow in genetics and now the core director of the longevity projects, is looking at epigenetic methylation in the stem cells of our families of centenarians. The methylation patterns in centenarians’ stem cells are dramatically different from those of the control group, and those of centenarians’ offspring are different from those of the age-matched control group. Since methylation occurs across the genome in millions of locations, this work is expensive, and its analysis is ongoing.
Meanwhile, hundreds of researchers worldwide are measuring proteins and metabolites, including hormone levels, to discover biomarkers that change as we age. We are finding that the changes in some of these measurements don’t do harm—they don’t seem to drive aging—while the changes in others may cause aging. What’s harder to decipher is which changes are responsible for the breakdown that occurs with aging, which ones are protecting us, and which ones do both. And it’s critical that we figure this out—if we mistakenly change something that’s protective, we make more trouble. Some mechanisms such as inflammation, which becomes more common with age, can be protective or destructive under different circumstances. We know that inflammation can contribute to chronic disease, so it might seem that we should find a way to stop it entirely, but we need inflammation to help us fight infection. Without it, we’d die.
Inflammaging, a term coined by immunologist Claudio Franceschi, reflects biological measures of inflammatory factors that are expressed in the elderly mainly by musculoskeletal pain that most of them suffer from. Some of the inflammation is caused by cells that stop dividing but do not die. These senescent cells—a.k.a. “zombie” cells—accumulate with age and can go rogue and secrete inflammatory factors and other proteins known as SASP (senescence associating secretory proteins) that may change their local environment and cause cancer.
We are also learning that some mechanisms, including sex hormones and growth hormones, fade as we age to protect us, and replacing or supplementing them has not proved to increase health span or life span. Estrogen and testosterone are prime examples of hormones that decline with age, and as an endocrinologist, I used to think that a lot of aging had to do with this decline and that hormonal treatment might slow aging. But this theory has proved to be largely erroneous. I believe evolution has a good reason for menopause. Tom Kirkwood, a scientist, colleague, and friend, argued in a provocative paper that if reproduction did not have an age cutoff in humans, many mothers would die relatively young because childbirth becomes riskier with age, and in many cases, mothers and grandmothers would therefore be unavailable to support the younger offspring. Many animals do not go into menopause, so this effect might be explained as a “side effect” of humans’ evolving to stand and walk on two feet instead of hands and feet. This position led to a narrowing of the pelvis, which had been wider when weight was distributed over four limbs rather than two. With more available nutrition, the babies also became bigger, and in combination with the erect position, this made large babies difficult to deliver and could have claimed the lives of many mothers. While a number of factors contribute to the rise in Cesarean sections in the past decades, one of the factors is overnutrition in pregnant women relative to their pelvis size.
Since we always need to consider the reasons that evolution planned our biology as it is, it’s usually best to proceed with caution when we attempt to alter those plans. For example, while hormone replacement therapy was all the rage for decades, a long-term national study conducted by the Women’s Health Initiative (WHI) found that many effects of estrogen were harmful in some postmenopausal women. In the study, which involved women who were an average age of sixty, the women who were given estrogen had more clusters of age-related diseases—such as heart attacks, more breast cancer, and more cognitive decline—than women in the same age range who were given a placebo. So while women who received estrogen did prefer not to have hot flashes, sleep disturbances, and other distressing symptoms of menopause and liked some of the positive effects, such as improvements in their skin and prevention of colon cancer, the estrogen replacement was harmful in terms of protecting against some major age-related diseases. Since the publication of this study, other studies have reported a decrease of up to 25 percent in the incidence of breast cancer in the United States.
Declining estrogen and resultant menopause appear to be protective against aging, and if that’s true, averting these conditions is the wrong thing to do. By the way, the same goes for declining testosterone in men—replacement therapies among men with very low testosterone improve some symptoms but increase the risks for other conditions. In fact, the risk-to-benefit ratio led doctors to stop recommending it even to men who are highly deficient in testosterone.
People who had been staunch believers in hormone replacement therapy were shocked by the results of the WHI study and are looking for windows of opportunity where the treatments will have the desired effects without doing harm. There are studies that suggest there may be a time frame between ages fifty and sixty when hormone replacement therapy could be beneficial, but as I explained earlier, estrogen appears to be a hormone that is beneficial in young women but harmful in older women, so we still have a lot to learn. And since people biologically age at different rates, we would need to pinpoint the right time frame for each person. Beyond that, even if we can provide the therapy during the optimal period, the benefits may end shortly after that time frame ends.
The WHI was a billion-dollar study that was controversial from the start. Many scientists and clinicians asked why it was necessary to spend so much money on a clinical study when there was already strong evidence from association studies. They had concluded from those studies that there was evidence that although estrogen carries a risk of thrombosis and strokes for some women, it does many good things for postmenopausal women, chief among them clearing the symptoms of menopause—hot flashes, sleep disturbances, mood changes, and vaginal dryness. Some animal models—most of them mice or rats that were supplemented with estrogen or a placebo after having had their ovaries removed at a young age—supported the notion that estrogen has positive effects. (Studies involving old mice and rats, whose health usually declined after being treated with estrogen, were largely ignored based on the argument that they were already too old or sick by then.) But the problem with association studies is that they do not account for the varying behaviors and habits of the participants. While many of the women who had been taking estrogen saw improvements, it was because they were taking very good care of themselves in other ways, not because they were receiving estrogen. Many women on estrogen were exercising regularly, did not smoke, took vitamins and supplements, carried less excess weight, and had healthier lifestyles in general than the women who were not taking estrogen. Admittedly, they also took estrogen very early on, rather than starting to take it in their sixties, and that may make a difference.
The WHI, on the other hand, was a clinical study (i.e., some patients received placebos), and it was also double-blind, meaning that neither the doctor nor the patient knew their assignment. And this kind of controlled study is the only kind of study that can specifically determine whether a drug works. In the case of estrogen, there’s no doubt that it has a youthful effect on a young body, but the question was, can it restore youthfulness to an old body? And the answer is no—an answer that had to come from a double-blind clinical study.
When I began studying the science of aging, the theory of cutting calories to increase longevity was being tested in research labs around the globe. Eating less than what we’d normally eat is called caloric restriction, and in animals, it turned out to have one of the most reproducible effects on slowing aging, increasing average and maximal life span, and increasing health span. For years, caloric restriction consumed geroscience because it was the only reliable method we’d found that could significantly extend longevity and delay the occurrence of age-related diseases.
In an experiment that is easily replicated in many rodent species, we restricted rats’ calories by 40 percent, and they lived about 40 percent longer than rats that were fed ad libitum—in other words, as much as they wanted. This was very exciting, so labs across the world began looking at why caloric restriction was increasing maximal life span. Collectively, we discovered that caloric restriction reduces age-related pathology, cancers, and other age-related diseases in rodents and slows down most physiological functions. So it not only increases life span but also increases health span. Still more exciting, there was good reason to believe that there would be similar effects in humans. But as a doctor with a specialty in treating type 2 diabetes, I know how difficult it is for people to cut calories. I tell all my patients to lose weight, but less than 3 percent of them are able to do it. If caloric restriction is as effective for people as it is for rats, we will need to isolate the mechanism that makes restriction beneficial and develop drugs or other treatments that don’t involve cutting so many calories—calorie restriction mimetics. No matter how hard they try, most people will not be able to consistently cut calories by 40 percent by eating less than they normally would at each meal. For those who manage to do it, we always say, “I don’t know if they’ll live longer, but it will probably feel that way.”
While our research certainly made it look like restricting calories was directly related to less disease and longer lives, I wasn’t convinced that it was the restriction itself that was producing these dramatic results. Initially, I thought that eating less prevents not aging itself but obesity and that the lack of obesity triggers the protective mechanisms that delay the onset of disease and extend life. My line of thinking was that rats that live outdoors run many miles a day to look for food, so they use a lot of calories. If we put them in a cage where they can eat as much as they want—and certainly more than they would if they lived outdoors—but can’t run around, the experiment is in some part about obesity versus leanness, not necessarily about eating a weight-maintaining diet versus caloric restriction. It was time to look at fat from a new perspective, so we set up a study to do just that.
PEPTIDE OR PROTEIN?
Peptides and proteins are both sequences of amino acids, but peptides are smaller. They are generally defined as molecules that consist of up to fifty amino acids, while proteins consist of more than fifty amino acids. Peptides also tend to be less well defined in structure than proteins, which can be organized into complex arrangements.
We used to think that fat tissue was just storage of excess fat, but we’ve discovered that fat—a.k.a. adipose tissue—has a biology and secretes several hormones and peptides. One of the hormones, called leptin, tells the brain when the body has had enough to eat, and years back, many researchers thought that leptin would become a treatment for weight loss. But while it’s true that the more fat tissue we have, the more leptin we have, our brain receptors stop responding past a certain point, becoming resistant to the effect of this hormone. Instead of receiving a signal telling us that we have had enough to eat, the signal is blocked and we feel like we haven’t had enough. In our own leptin experiments involving rats, we could not elicit in old animals the same effects we elicited in young animals. While caloric restriction generally seemed to reverse aging, even calorically restricted old animals did not respond to leptin. So advanced age seems to be a leptin-resistant state, and we lost interest in leptin as a potential gerotherapeutic supplement that would mimic caloric restriction’s effects.
With another hormone, adiponectin, the opposite happens. The more adipose tissue we have, the less adiponectin we have. But considering that this hormone is generally good for all aspects of our metabolism, lowers insulin resistance and inflammation, and has a number of other positive properties, this is not a case where less is more. In our Longevity Genes Project, we had found that centenarians have high adiponectin even though they are not restricting calories. Instead, some of them have a helpful mutation in the adiponectin gene that keeps their levels of this good hormone high and acts as a protective mechanism. For those of us without the beneficial mutation, a little bit of fat works in our favor, but having too much intra-abdominal fat (visceral fat) is at the core of the metabolic decline that occurs with age. Abdominal obesity—think older people who are thin but have big bellies—is the marker for the onset of diseases. This is accepted now, but when I first introduced to the field of gerontology the idea that the biology of adipose tissue might be the mechanism underlying the apparent success of caloric restriction, my theory was viciously attacked.
Being a researcher requires thick skin and an open mind. Things are not always what they appear to be, and when I’m tempted to jump to a conclusion, I recall a story I heard about a photograph that was taken of Pope John Paul II at the wheel of a limousine. As the story goes, the pope had given a sermon at a cathedral in Yonkers, New York, in the mid-’90s, and afterward, a Cadillac limousine was waiting to take him to the airport. When the pope saw his driver, he said, “My son, when I was in Poland, I drove all the time. I really want to drive a Cadillac.”
So the driver, not knowing how to say no to the pope, got into the back seat, and the pope headed for JFK. On the way there, he was pulled over for speeding, and the officer asked to see his driver’s license.
“I’m sorry, son, I don’t have a license,” the pope said. “But we’re only going a short distance to the airport, so I’m sure we’ll be fine.”
Not sure what to do, the officer called his captain. “I stopped a limo for speeding, and it’s somebody really important,” he said.
“Who is it?”
“I don’t know.”
“You’re telling me it’s somebody really important, but you don’t know who?”
“Well,” the officer said, “it has to be somebody really important because the pope is his driver.”
Had the officer peered into the back seat and seen the driver in his chauffeur uniform, he would have reached a different conclusion. What’s obvious is often misleading, so researchers make a habit of peering into the back seat. This is what we had in mind when we constructed a major study to test my theory that the “success” of caloric restriction was actually due to having only a small amount of visceral fat. We could have chalked that success up to caloric restriction, but we decided to look into the back seat by performing a new experiment.
We started with four groups of rats—one group of young rats and three groups of old rats. Of the three groups of old rats, we restricted the calories of one group and let the other groups eat as much as they wanted. By the end of the experiment, the old rats with restricted calories weighed the same amount as the young rats, but the two groups of old rats that ate as much as they wanted were obese, weighing two hundred grams more than the rats in the other groups.
Then we did something dramatic with the two groups of obese rats that required extensive practice and experience. In one group, we performed a surgery where we removed visible deposits of visceral fat from inside the abdomens. In the other group, we removed the same amount of surface (subcutaneous) fat as we’d removed in the form of visceral fat from the other group. In the young group and the older group that had been calorically restricted, we made an incision in their abdominal walls but sutured the wounds without removing fat for the sake of control (known as a sham operation).
We measured all the rats’ insulin levels and insulin’s action regarding driving glucose out of the bloodstream and into the muscle. One of the hallmarks of the metabolic decline that comes with aging is a decline in sensitivity to the effects of insulin, the hormone we need to store sugar molecules. Insulin is manufactured in the beta cells of the pancreas, and its levels become elevated to overcome the decline in insulin sensitivity. Those of us who can’t secrete enough insulin become type 2 diabetics. Obesity accelerates this process because it increases resistance to insulin, so obese people can become diabetic at earlier ages.
The gold standard for determining whole-body insulin sensitivity is the “insulin clamp” test, which my Einstein mentor and colleague Luciano Rosetti and I were the first researchers to apply to rodents and to aging. I had performed insulin clamp studies in humans during medical school at the Technion in Israel, where I worked with my first mentor, Eddy Karnieli, and during fellowships at Yale with leading diabetes researcher Ralph DeFronzo. For the studies, insulin is administered, but glucose concentration is “clamped” so that the study participants’ normal blood sugar levels are maintained—they do not go up or down. The more glucose you need to administer to hold glucose levels steady, the more sensitive the body is to insulin. The less glucose you need to administer to maintain the levels, the more resistant the body is to insulin. For the same levels of insulin, much less glucose needs to be administered for elderly people than for young people, because elderly people are generally in an insulin-resistant state. In our studies, when aging animals had their subcutaneous fat removed, they had insulin resistance typical of advanced age. In contrast, old caloric-restricted animals had the same insulin sensitivity as young animals. The group that I was holding my breath about was the group of old rats that had had visceral fat removed. If our theory was correct, their insulin sensitivity would be similar to that of the young and the old caloric-restricted animals, and to our delight, they were. All we had to do to achieve a healthy metabolism in rats that ate as much as they wanted and were also obese was to take out their visceral fat!
We knew we were onto something, and we wanted to see what would happen if we did a similar experiment with a model of diabetic rats. A breed of lab rats known as Zucker fatty rats begins developing diabetes at two months, and by the time they’re five months old, 100 percent of them will be diabetic. We removed our rats’ visceral fat at two months of age, and by the time they were five months old, only 20 percent of them had diabetes. It looked like a victory until we hit the end of month six, by which time all the rats in the study were diabetic. At first, we were disappointed, but then we realized that the rats’ visceral fat had grown back and that that’s why they became diabetic. Further study revealed that they would become diabetic when 60 percent of the visceral fat had grown back. As long as they had less than 40 percent, though, they did not become diabetic, so the experiment ended up proving exactly what we had theorized.
But the ultimate study was the one we did with more than 150 rats that were divided into three groups. One group was fed as much as they wanted. The second group had their calories restricted. And the third group ate as much as they wanted, but we removed their visceral fat at the beginning of the experiment. We also performed a sham operation on the first two groups, surgically opening and closing them without removing visceral fat.
The caloric restriction did what we expected it to do, and the restricted group lived about 40 percent longer than the group with no caloric restriction. The group that had had visceral fat removed didn’t live quite as long, but they did live about 20 percent longer than the control. So with the visceral fat removed, their maximum life span was close to that of the caloric-restricted animals. What this means is that nutrients themselves, or the time they are administered, are playing a role in aging, but taking out the visceral fat has significant effects on longevity.
While we won’t be doing this surgical procedure on people, we may be able to develop a less invasive treatment or drug that does the same thing. I have been helping a company that developed a method to melt intra-abdominal fat that has demonstrated a significant metabolic improvement in monkeys, and now they’re starting human trials. So that’s good news on the horizon, and I also have some wonderful news about subcutaneous fat. We’re learning that having a little bit of fat under the skin is a good thing. Not just because it can act as a protective barrier against viruses, germs, and other substances that manage to penetrate the skin but also because it produces the “good” peptides and fat hormones like adiponectin, which as I mentioned earlier is found at high levels in centenarians.
Around the same time as our rat study, researchers Richard Weindruch, Joseph Kemnitz, and Rozalyn Anderson at the University of Wisconsin and Donald Ingram, Julie Mattison, and Rafael de Cabo at the National Institute on Aging in Baltimore studied the effects of caloric restriction on primates. The experiments were different, but in both cases, the animals that were permitted to eat as much as they wanted became obese from excess calories and lack of exercise, which is also what happens with people. Both studies showed that a variety of age-related diseases, including diabetes and cardiovascular disease, were delayed in the monkeys with restricted diets. These are the same diseases that are delayed when you fight obesity. But only the University of Wisconsin study showed a significant increase in life span.
At one point, I was one of the reviewers on the Wisconsin program, and when we visited the research facility at the university, I wasn’t the only one who noticed that over the past year or so, the animals being fed restricted calories had come to weigh almost as much as those that were permitted to eat as much as they wanted. “There’s only one explanation here,” I said, “and that is that somebody sabotaged the experiment.” The other reviewers agreed, and we wondered what the study’s researchers would say about this.
When we arrived for the review panel the next day, one of the researchers explained that a caretaker had felt sorry for the animals with restricted diets and gave them almost as much food as she gave the control group. I didn’t want our animal care workers to do the same thing in any of our studies, so when I got back to Einstein, I attended their next meeting and told them what had happened in Wisconsin. I explained that while the concerned caretaker thought she was helping the animals on restricted diets by feeding them more, if they had gotten fewer calories, they would have had healthier and longer lives.
Since caloric restriction looked like it was extending health span and life span, the research community was curious to learn how it affected levels of growth hormones, sex hormones, thyroid hormones, insulin levels, and cortisol levels. As it turned out, animal models whose levels of hormones were maintained at caloric-restricted levels in isolation did not realize extended life spans. So far, the only decrease that’s known to make a difference in longevity is the decrease in growth hormones. (More about this in chapter 5.) As for levels of cortisol, which is a stress hormone, they went up because restricting the diet by 40 percent causes stress, but when cortisol is administered to old animals, it actually shortens their life spans.
From my standpoint, the study about caloric restriction that’s most relevant to humans is one that showed how genetics change the effects of caloric restriction. Jim Nelson, a colleague from San Antonio, gathered other colleagues to this fascinating study in which he bred two mice that were genetically very different. Eventually, the mice produced male and female offspring with forty-one genetically distinct backgrounds. All the mice were either fed ad libitum or fed restricted diets, but unexpectedly, only about half of the caloric-restricted animals lived longer than the ad libitum mice, and the other half lived shorter. This means that genetic background is very important and that caloric restriction cannot be universally applied. Whether caloric restriction will lead to a longer life in humans also depends on our genetic backgrounds. And how many calories we should restrict ourselves to may also depend on our DNA. Among other limitations of this study, it’s possible that the researchers would have seen more longevity if the calorie restriction had been less severe than 40 percent. If they had cut 20 percent of the mice’s calories, I think that significantly more than half of the calorie-restricted mice would have outlived the ad libitum mice.
It’s possible that caloric restriction produces the most benefits for the obese and does harm to people who are of average weight or leaner. Yet there are thousands of people who are not obese who religiously follow a diet of calorie restriction with optimum nutrition (CRON). These people, popularly referred to as CRONies, eat a low-protein, low-fat, high-plant diet and consume only about nineteen hundred calories a day. I once gave a talk about caloric restriction and adipose tissues at the University of Cambridge, and during the luncheon that followed, I found myself surrounded by a group of CRONies. I recognized them because they all looked emaciated. They each took out a packed lunch of greens—just greens—and a small kitchen scale and proceeded to weigh the lettuce before sprinkling it with a bit of vinegar. None of them looked healthy and vibrant, and the younger ones looked like they might be prematurely aging. One young man in particular adamantly defended caloric restriction, and I gently tried to broaden his perspective.
“You know, one of the things you’re doing is exchanging reproduction for longevity,” I said.
“No, that’s not true,” he said. “We can have reproduction.”
The others nodded.
“Do you know your testosterone level?” I asked him.
“Two hundred.”
“Well, that’s very low. How do you think you’ll have reproduction?”
“Because I can stop this at any time.”
“No, because when you have low testosterone, you don’t have desire. You’re not going to have the desire to stop. I have longevity because my genes are in my kids already. My DNA may live forever, but you’re only going to have your life span. Maybe you think it’ll be long, but not if you’re killing yourself.”
I don’t run into CRONies as often as I used to, so maybe they are adopting healthier approaches to eating.
One of the biggest surprises we’ve had regarding aging in the past few decades is that aging begins in utero. This hypothesis by David Barker, a British epidemiologist, proposed that slow intrauterine growth, low birth weight, and premature birth have a relationship with the origins of diseases that typically show up in middle age, including hypertension, coronary heart disease, and type 2 diabetes.
Barker’s theory is based on observations he made of people who had been born in an area of the Netherlands near the end of World War II, when the food supply was extremely low. The women who were pregnant at that time gave birth to small babies, and when those children were very young, they already had a high incidence of age-related diseases, including diabetes, hypertension, and kidney failure. Barker’s hypothesis was that they had protected themselves while in utero to survive with minimal calories but that when they were born and had enough to eat, that survival strategy had an adverse effect. Somehow, getting a normal amount of nutrition was actually harming them.
In a study at Einstein, a team of obstetricians, Francine Einstein and Hye Jung Heo, and geneticist John Greally Reid Thompson and I showed that young rats that were small for their age had an epigenetic methylation profile that was similar to those of old rats, linking the mechanism of aging as a result of small birth and epigenetic mechanisms. In separate human studies, they also showed that babies who are born small have different epigenetic methylation patterns than average-size babies. These studies and others indicate that some aging depends on the conditions that exist in utero and continues throughout stages of life. So scratches on the CD start the aging process before we’re even born.
So far, we’ve discovered three primary traits that SuperAgers have in common. Not all SuperAgers have all three, but these three phenomena most profoundly slow aging—and offer hope for the rest of us.
1. High levels of good cholesterol
HDL—high-density lipoprotein—cholesterol protects from heart attacks and dementia.
2. Unusually low levels of the growth hormone IGF-1
This protein produced in the liver helps to grow tissue. Low IGF-1 helps us shift our energy from growth to survival.
3. Unusually high levels of some MDPs
These proteins originate in the mitochondria and have many varieties—some of which are found only in SuperAgers—that build resilience against the stresses of aging.
Along with giving us some great leads to follow, our research with SuperAgers has made it clear that we need to follow exceptional longevity over years and decades. Though we have learned a lot with our Longevity Genes Project and made important observations and connections, studying people at a certain age provides only a fragment of the information we can get by following them throughout life. With that in mind, we launched a new study called LonGenity, for which we recruited Ashkenazi Jews whose parents had lived to be ninety-five or older and who were not already in our LGP project. This group is called OPEL—offspring of parents with exceptional longevity—while our control group is called OPUS—offspring of parents with usual survival. We have nearly fourteen hundred people in the study, and we are doing a vast amount of testing on them every year, including extensive neurocognitive tests, MRIs of their brains, and CT scans of their coronary arteries. We’re trying to capture detailed information about their health as they grow old and, of course, find the relationship between their health and the longevity genes. The long-term objectives are to identify genes that contribute to exceptional longevity in humans and assess associations among these genes, age-related diseases, and longevity.
To date, our research results have been encouraging and enthusiastically received by the medical research community. Among the findings, the team has learned that longevity is: