8
Cutting Free of the Cellular Spider Webs
Our cells—and thus our bodies—are progressively damaged by protein-derived junk that gathers over the years in the space between cells. Alzheimer’s disease is perhaps the best-known condition associated with this, but there are others that are equally fatal. However, there is a way forward for medical science and our health: recent and very promising research demonstrates that science can turn our own immune systems against this dangerous material.
In the previous chapter, I talked about the junk that accumulates inside our cells with age—how it contributes to the biological aging process, and what can be done to get rid of it. In this chapter, the focus is on the garbage that accumulates on the outside of our cells and tissues, enmeshing them in webs of damaged proteins, impairing their function, and contributing to aging and age-related disease.
Most of the junk that we’ll be discussing is amyloid of one kind or another. When I say “amyloid,” of course, almost everyone thinks of beta-amyloid protein (also called “amyloid beta”), which accumulates as the waxy “senile plaques” that cluster around the brain cells of people with Alzheimer’s disease. But many other, less-well-known diseases (amyloidoses) are also rooted in abnormal protein aggregates of this type. Most amyloids are cell-snaring chains of molecules that begin their existence as healthy proteins already present naturally in our blood, or in the fluid bathing our brains. A wide range of proteins can become amyloids under the wrong circumstances, including immunoglobulin light chain, a key component of the antibodies in your immune system; the protein transthyretin, which is responsible for carrying around thyroid hormones in your blood; and a small protein (islet amyloid polypeptide, or IAPP—also referred to as amylin) that helps your body regulate its blood sugar levels in association with insulin.
What turns these proteins into snares that squeeze the life out of cells and organs is how they are folded. Misfolded proteins are just what they sound like: proteins that have become twisted out of their proper configuration in ways that cause them to undergo toxic interactions with each other, or with other constituents of the cell. The ones that cause amyloid diseases contain sites within their structure that, if exposed, readily stick to other proteins of the same sort, causing them to link together one after another in a sinister, self-assembling daisy chain. These sticky sites are normally kept safely tucked away within the complex folding of the protein’s three-dimensional architecture, precisely to prevent such interactions from happening. Misfolding exposes such sites, initiating the spinning of a cell-choking web.
Many of the amyloid diseases result from the victims having faulty genes that produce defective versions of these proteins. In some such disorders, the underlying mutation introduces fatal flaws into the structure of the protein itself, causing it to open up at inappropriate locations, exposing the critical “sticky” site in its structure. Others involve mutations in enzymes that normally chop up the protein into functional units as it emerges from the cell’s protein-assembly machinery. These mutations cause the enzymes to cut too close to the critical site, again unleashing it from the restraining influence of the rest of the protein’s normal conformation. Another route to congenital amyloidosis is errors in “chaperone” proteins whose job it is to direct the emerging (and potentially amyloidogenic) protein to assume a safe, nonamyloidogenic final shape.
But in addition to these inherited protein-misfolding diseases, there are also universal amyloidoses—ones that are the result not of mutations, but of the fundamental vulnerabilities that proteins face in the course of their critical jobs in the molecular maelstrom of cellular biochemistry. With free radicals, sugars (sugars? Oh yes. See Chapter 9) and vibration constantly acting on them, proteins are bound to get bent out of shape now and again in ways that open them up to becoming the seed of an amyloid fibril. Once one such protein is formed, it can sometimes twist other proteins out of shape as it grabs at them, exposing another site and forming the nucleus of an ever-expanding fibril chain. One example of this happening in fast-forward is seen in people with kidney failure, when the body ceases to pass out beta-2-microglobulin in the urine. Beta-2-microglobulin is normally a perfectly safe protein that helps the body to distinguish its own cells as “self” from “nonself” cells of bacteria or other organisms. But without regular excretion, levels of this protein begin to climb to abnormal levels, and they eventually reach so high a concentration that they start to spontaneously glom together, forming amyloid deposits.
Indeed, Cambridge professor Chris Dobson, who has spent his academic life looking into protein misfolding diseases, says that “conditions could be found in which seemingly any protein could form amyloid fibrils [emphasis mine]…although the propensity to form such structures under given circumstances can vary greatly from one protein to another.”1 Over time, these fibrils build up to potentially pathological levels, coiling around our cells and organs, choking them off like so much bindweed.
Mind-Forg’d Manacles
Most researchers now believe that the horrors of Alzheimer’s disease can mostly be traced to abnormal processing of an otherwise healthy molecule called amyloid precursor protein (APP). Everyone’s brain produces APP, and it is required for some essential function in our bodies. Ironically, in fact, properly processed APP actually appears to be required for many of the key activities of healthy neurons, such as their ability to rewire themselves in response to new learning and to grow out the branching “electric cables” (neurites) that allow them to talk with one another.
When things go right, APP is produced in the main body of the cell and then sent for further processing to alpha-secretase,2 a type of enzyme called an endoprotease. The result is the creation of two molecules, one of which remains in the membrane of the neuron’s neurites, while the other is released into the fluid inside the cell. APP cannot form the evil beta-amyloid when it is processed by alpha-secretase. After this, one of the fragments is chopped further, by a distinct enzyme named gamma-secretase.
APP only become dangerous when, instead of being trimmed down by alpha-secretase, it is mistakenly cut up by a different, but related, enzyme called beta-secretase.3 Beta-secretase, like APP, is not a villain: it has a proper place in the cellular “factory,” as part of another, distinct cellular assembly line from the one that handles APP. In that assembly line, beta-secretase makes essential trims in the structure of other proteins that bear some molecular resemblance to APP itself. But if beta-secretase performs the same action on APP, it chops it in the wrong place. This distorts the protein’s shape and creates a molecule with a totally different action within the cell.
It’s as if beta-secretase were an overly helpful laborer who, while crossing the factory floor on the way back from lunch, had seen some APP lying on a stopped assembly line and mistaken it for a part that he or she normally works on. Seeing no alpha-secretase around, and presuming to know what the half-finished product needs, beta-secretase steps in to do alpha-secretase a favor by taking care of a little bit of its workload. After giving it a few whacks of its molecular hammers, beta-secretase tosses the APP fragment—now subtly misshapen—back onto the line, where it eventually reaches gamma-secretase. And because gamma-secretase is a busy enzyme, it’s too caught up in its work to notice the change, and proceeds to splice and dice the distorted APP fragment just as it would if alpha-secretase had made the proper modifications. Beta-amyloid is the product of this mistaken sequence—the sequential cleavage by beta-and gamma-secretase rather than by alpha and gamma.
When processed appropriately, the middle APP component (between the sites of cleavage of alpha-and gamma-secretase) assumes a shape similar to a stretched-out coiled spring—a conformation called an alpha helix. But thanks to beta-secretase’s molecular meddling (and gamma-secretase’s unwitting cooperation), this fragment loses its normal shape—and, just as would happen if you were accidentally to cut into a tightly drawn spring with a pair of wire cutters, the improper slicing of APP causes the fragment to jump backwards on itself, creating a shape more like a bent hairpin (a beta-sheet) that gives beta-amyloid the fatal molecular stickiness that characterizes amyloid proteins.
Once released by gamma-secretase, individual fragments (monomers) of beta-amyloid initially float free in the brain. But they soon come into contact with other monomers, and their “stickiness” causes them to glom together into larger—but at this stage still free-floating—units called oligomers. These fibrous strands, in turn, get stuck to one another to form still longer strands, which eventually grow so large and complex that they can’t stay dissolved in the brain’s fluids, but precipitate out into the spaces between neurons to form the notorious plaques. Under a microscope, these mind-snaring webs can be seen extending to the neurons’ caretaking support staff (the glial cells), and down to the neurites (the wiring system that I mentioned earlier).
Some people produce unusually large amounts of beta-amyloid because they have inherited mutations that either cause their bodies to produce too much of APP itself (thus increasing the odds that the problematic enzymes will come across molecules of it and mistakenly hew their structure), or else encode defective secretase enzymes that are not so good at doing their selective jobs as the more common varieties. But because everyone has both APP and the enzymes that can sometimes turn it into beta-amyloid, we all produce beta-amyloid, and given a constant output of the stuff, some fraction of that amyloid precursor is bound to get snipped in the wrong way now and again. Once that happens, it’s only a matter of time before enough of it builds up to form Alzheimer’s-type plaques—and indeed, all of us have at least some plaque in our brains by the time we reach late middle age.
Thus, like other aging damage, beta-amyloid plaques simply accumulate over time, and it’s reasonable to think that neurological impairment occurs when a critical threshold is reached. This is probably why most cases of Alzheimer’s disease are not inherited, but instead occur sporadically in the population: the underlying biochemistry is just part of the kind of organisms we are, living in the kind of universe that we do. Lifestyle risk factors and most genetic predispositions merely determine how early on in our lives the process begins to impair our intellects and identities.4 This is also why, apart from a very small number of inherited, early-onset cases, almost no one in early middle age or younger gets Alzheimer’s…and why the prevalence of the disease doubles every five years beyond age sixty-five, so that victims pile up with age like the grains of rice on the Emperor’s chess-board in the old fable. Our brains are slowly being enmeshed in beta-amyloid plaques—it’s just a matter of when we reach the threshold beyond which our brains can’t keep up sufficient function to carry on the lives and identities that we have spent so many years creating. Barring some radical new therapy, each and every one of us will be struck down by Alzheimer’s dementia if something else doesn’t kill us first.
“O Captive, Bound and Double-Ironed…”
Beta-amyloid also causes brain damage and death in many people who never develop Alzheimer’s disease. This is because, in addition to building up in the neurons of the brain, beta-amyloid also clings to the interior surfaces of its blood vessels. The resulting condition, called cerebral amyloid angiopathy (CAA), is a crusting-up of these pipelines of oxygen and nutrients, weakening them and reducing their ability to flex in response to the surging flow of the pulse. This leaves them vulnerable to bursting open in a bleeding stroke.
CAA is certainly more common in people with Alzheimer’s (about a quarter of all patients have it as a complication), but as we age it becomes an increasingly serious issue in people not struck by the latter disease. Just 5 percent of us have CAA in our seventies, but after the age of ninety over half of us are suffering from the disease, and it is responsible for about 15 percent of all bleeding strokes in people over the age of sixty.
But there’s more: beta-amyloid is just one mangled protein among many. Less well known, and less recognized as causes of death and disability, are a variety of other age-related amyloidoses that also don’t seem to be related to an inherently malformed protein, but to the healthy version being damaged in the rough-and-tumble of its biochemical environment. Senile cardiac amyloidosis is one example. As you might guess, this disorder is most clearly characterized by amyloid fibrils building up in the heart, although it damages the lung, liver, and kidneys, too. This buildup interferes with the regular beating of the heart, and can cause heart failure. The fibrils are made up of transthyretin—the rickshaw driver for thyroid hormones that I mentioned earlier on—and while it can arise from a mutated version of the protein, it can also result (at a slower rate) from damage to the form that most of us carry.
As the name implies, senile cardiac amyloidosis is a strongly age-related disease—first showing up in people over the age of seventy, and found at pathological levels in about a quarter of people over ninety. As in the case of Alzheimer’s disease, if we all lived long enough without something else killing us first, each of us would wind up with the lives squeezed from our hearts by this form of aging damage. The disease is known to be a common contributing cause of death in the “oldest old,” such that about half of people over ninety years old have diagnosable senile cardiac amyloidosis at autopsy.
Much earlier on in life, nearly everyone gets some degree of amyloidosis of the aorta, the main blood vessel leading out of the heart. Two different proteins are involved, one of which builds up in the innermost layer of the aorta in an astounding 97 percent of people over the age of fifty, while the other accumulates deeper into the middle of the vessel wall in about a third of these cases. This amyloidosis is not currently recognized as a cause of specific pathology or death, but again it seems that this is only a matter of stepping over a fatal threshold that we don’t reach in a normal lifespan today because we die of other things first.
Add them all up, and amyloid deposits of any of several misfolded proteins in the heart are significant contributors to death in the elderly, causing abnormal heartbeat, weakening of the muscle of the heart, “blackouts” of the electrical activity that keeps it beating, and heart failure.
And those are just deposits of the cardiovascular system. Every one of us becomes riddled with microscopic amyloid deposits across multiple tissues in the body by the time we hit our eighties. Its toll isn’t widely appreciated because the very old are autopsied so rarely, and you just can’t see the deposits without opening a body up. This lack of curiosity about death in the very old of today is just another example of our routine acceptance of the massive toll of aging processes in people who have enjoyed only—yes, only—a few score years of life.
Moreover, while the evidence is still preliminary, amyloidosis of one organ system or another appears to become an increasingly critical factor in the snuffing-out of people at the extremes of the current “natural” longevity range. Some of this evidence comes from Japan, where the presence of a few centenarian “hotspots” has made it an exception to the widespread pattern of voluntary ignorance about what kills the oldest old. An autopsy study carried out at Aichi Medical Center in Japan from 1989 to 1995 found brain-wide CAA in 16 of their 19 centenarian patients.5 Unfortunately, this study was restricted to the central nervous system, so it did not provide any information on what other amyloid diseases might have riddled these long-lived humans’ bodies or to what extent such diseases may have contributed to their deaths.6
Even more suggestive is the early evidence coming out of the important effort by the newly launched Supercentenarian Research Foundation (SRF)7 to autopsy as many “supercentenarians” (the extremely rare people who live beyond the age of 110) as can be identified and convinced to donate their remains to science after their deaths. Of the six who have thus far been examined, four were felled by some form of amyloid disease (the other two deaths were cancer victims).8
Again, we don’t yet know what the pathological consequences of many of these deposits may be, but it seems awfully likely that they are doing us harm—so by the engineer’s definition, they’re aging damage, since they’re not found in the young. Hence, you can bet your life that I want to clean them up along with the ones that have already been exposed as culprits in specific age-related diseases.
Alzheimer’s, Amyloids, Aging
In a perverse way, then, the fact that Alzheimer’s is such a widespread and obviously terrible disease has aided the cause of general anti-aging biomedicine. The attention to this specific age-related curse—together with the widespread professional and public belief that amyloid beta deposits are a major factor in its development and progression—has driven scientists to attack this particular form of amyloid as a therapeutic target in its own right, and that work has opened up the strong possibility that a similar strategy can form the basis of foreseeable therapies for amyloid-type extracellular damage generally. As with other cases that we’ve discussed in previous chapters, the existence of a recognized disease that is caused by aging damage has “legitimized” research into ways to clean it up—and, as this research bears fruits in new cures for these diseases, anti-aging biotech will be able to hitch a ride to develop treatments for aging damage itself.
Alzheimer’s is an especially good example of this phenomenon, because it is both so utterly fearsome and so common in our parents and grandparents (in contrast to rare and rapidly fatal disorders like the mitochondriopathies or the lysosomal storage diseases). As the sheer number of Alzheimer’s victims explodes as the population’s biological age creeps upward, victims’ families and loved ones have organized politically. There are now thousands of people in the United States and elsewhere who are demanding—and getting—enormous government investments of intellectual and financial capital into the quest for a cure. (Indeed, for better and for worse, Alzheimer’s research now consumes over half of the National Institute on Aging’s budget.)
Once the view that beta-amyloid was the key to the disease became dominant, Alzheimer’s specialists (possibly because they were not biogerontologists?) began thinking about this particular form of extracellular junk along the same damage-reversal lines that underlie the engineering approach to age-related damage generally. All that currently available treatments for Alzheimer’s disease can do is improve the symptoms of the disease: sadly, no existing therapy has the power to check the ongoing degeneration of the brain itself (see sidebar, “Alzheimer’s treatment today”). This does mean that users of these drugs are better off at any given point than they would be without them, but the underlying progression of the disease continues unabated with every passing day. Functionally, even modern Alzheimer’s drugs perform the way ibuprofen and antidepressant medications do for diabetics—in providing merely superficial relief from the nerve pain that often accompanies the disease. While the pain may indeed diminish, diabetics’ nerves themselves continue to be savaged by the “caramelization” chemistry of that disease. (See Chapter 9 for SENS’s main answer to late-onset diabetes.)
ALZHEIMER’S TREATMENT TODAY
Currently, the most widely used treatments for Alzheimer’s disease are the cholinesterase inhibitor drugs, such as donepezil (Aricept), rivastigmine (Exelon), and the herb galantamine (Reminyl/Razadyne), which attempt to bolster brain functioning by boosting levels of some of the signalling molecules involved in some of the flagging aspects of memory. Of course, such a treatment is purely palliative, with no effect on the underlying disease process. A recent study9 suggests that these drugs are even less effective than is implied by their inability to prevent the brain-ravaging effects of the disease: It appears that while the drugs boost scores on standardized tests of some aspects of brain function, they have no effects on the kind of real-world functionality whose loss forces families to institutionalize victims.
There was hope, for a while, that a more recently introduced drug called memantine (Namenda) would at least slow down the progress of Alzheimer’s disease, by shielding neurons against the damaging effects of another signalling molecule (glutamate) that can kill brain cells when present in excess. A recent trial10 suggests that this isn’t so. The study compared people who had already been on memantine in a six-month placebo-controlled trial, and who were then allowed to continue taking it for an additional six months, to people in the same trial who had originally been taking the placebo but who were given the real deal for those subsequent six months. If memantine were really slowing down the underlying disease process, you’d expect that people who started on the drug earlier would have been in better shape than those who’d had to wait through the first six months on the sugar pills, because they would not have been suffering the full effects of brain degeneration for the first six months and thus would have more intact brains later on. But instead, it was found that the patients who started taking the drug later quickly caught up, in terms of their improvement over their baseline condition, with those who had been getting it all along. This suggests that me mantine’s effects are only on the immediate symptoms of the disease.
That’s good news, in a sense, for those who start taking me mantine at a more advanced disease state, because it means that they haven’t lost anything by waiting to get started. However, the bad news is that no one taking memantine can expect that it will actually stop their minds slowly dying from the tangled mess in their brains.
It’s actually not clear that blocking glutamate’s effects on neurons would be an entirely good thing in any case. As with so many things in the finely tuned network of metabolic pathways, glutamate is a molecule with two faces. While it can stimulate brain cells to death when it’s present in excess, it’s also a key chemical signalling molecule in the brain, required for the normal storage and retrieval of memories. This suggests the possibility that memantine may cause problems in the laying down of new memories, even as it preserves the brain cells that store the old ones. There is no direct evidence of such an effect yet, but it still isn’t clear that the drug helps much, either: While this trial did seem to show a benefit from memantine, there were no statistically significant benefits compared to sugar pills in the two other major trials of the drug.
In any case, even a drug that could slow down the rate at which brain cells are lost would be unable to prevent—let alone reverse—the degeneration of the brain, since the underlying damage that has already occurred is left unrepaired.
As the evidence supporting a central role for beta-amyloid in the development and pathology of Alzheimer’s built up, a new hope emerged. Scientists began talking seriously about the idea that, by making beta-amyloid itself the target for new medical interventions, they would be able to develop new treatments that would treat the disease instead of merely providing crutches to a crippled—and rapidly deteriorating—mind. Once researchers had the tools they needed, in the form of engineered mice whose brains produced variations on human beta-amyloid that led to the formation of brain plaques and dysfunctions of brain and memory, they could start work on testing therapies that would target beta-amyloid directly.
New Targets, Old Rifles
But simply believing that that beta-amyloid is the main villain in the Alzheimer’s story doesn’t tell you what to do about it. Thus, it’s no surprise that academic labs and pharmaceutical companies around the world have been working on quite a variety of anti-amyloid strategies, each hoping to make a Nobel-quality breakthrough or market a blockbuster new drug with a desperate—and ominously, inexorably expanding—“target market.”
Predictably, however, when scientists began thinking about how to tackle the beta-amyloid plaque problem, many of them first turned to classically preventative strategies typical of the old-style gerontological approach to aging. Recall that beta-amyloid, like other amyloid proteins, is formed from an essentially healthy protein—amyloid precursor protein (APP). That protein is found in long strands woven through brain cell membranes, and while its exact function is unknown, it’s at least harmless as long as it remains intact and in place. But occasionally, the beta-secretase enzyme mistakenly latches onto the APP protein and chops it at an unintended place; gamma-secretase innocently follows suit, not recognising the fatal flaw in the misprocessed APP; and beta-amyloid—with its exposed, sticky binding sites—is released to wreak its havoc in the brain.
With this in mind, one of the first ideas for an anti-amyloid therapy was to create drugs that would dampen down the activity of these enzymes, thereby cutting down the production of beta-amyloid. This in turn would reduce plaque formation, and thus either slow down or prevent the emergence of the disease.
First out of the gate was a drug that interfered with gamma-secretase activity. Animal studies showed that even a single dose of the drug could reduce levels of soluble, pre-plaque beta-amyloid in both the brain and the plasma, and it was duly moved through the development pipeline into the “Phase II” trials that are designed to give preliminary evidence of a drug’s efficacy and safety in a moderate-sized population of people with the disease.
That was in 2001. I’m writing this in early 2007, and to date there has been total silence about the results of the trials of this first gamma-secretase inhibitor. We may never know what happened, but we may be able to guess. Even as the drug was being tried in humans, greater understanding of the role of gamma-secretase in the body emerged. A question that had long hung over the enzyme-inhibition approach was what, exactly, the enzyme was supposed to be doing in the body.
Many harmful mutations lurk in isolated pockets of the human family, but we all have gamma-secretase in our brains—and, as I discussed in Chapter 3, evolution does not design us to suffer horrible diseases. Although gamma-secretase has the unfortunate long-term side effect of beta-amyloid production, scientists always had in the back of their minds the acknowledgement that it had to serve some useful purpose, too. And, sure enough, researchers discovered while the trial was still progressing that gamma-secretase operates on several proteins in the body—including Notch receptor 1 (NOTCH1), a protein with critical functions. By this I mean really critical: They include the activation of stem cells that renew damaged muscle tissue, the growth of new blood vessels, and the maturation of some kinds of immune cells.
So, what happens to these important biological processes, when you start interfering with an enzyme that’s needed to keep them going? Animal models, using either a “knockout” of the gamma-secretase gene or an alternative gamma-secretase inhibitor drug, showed that dampening down the enzyme clearly prevented immune cells from developing in both the bone and the thymus, reducing the number of these cells and causing pathology in the gut. It seems highly plausible that the reason for the stony silence on the human trials is a similar profile of side effects.
However, some researchers are still chasing after therapies based on the same basic strategy. In 2002, researchers at Eli Lilly presented animal data showing that LY450139, the code name for a new gamma-secretase inhibitor, lowered beta-amyloid levels in Alzheimer’s mice without interfering with NOTCH1. As this chapter was being drafted in 2006, the results of an early human trial in seventy Alzheimer’s patients came in, showing that the drug reduced levels of beta-amyloid by 38 percent without apparently causing any serious side effects. By April 2006, the company had partnered with several universities and hospitals and was gearing up to perform a larger clinical trial to see if it could actually affect the disease. Elan Pharmaceuticals is also continuing to research a gamma-secretase inhibitor.
But deactivating NOTCH1 is far from the only potential concern with these drugs. Remember, gamma-secretase is also an essential partner in the normal, non-amyloidogenic metabolism of APP into products that appear to be essential to the functioning of neurons. It’s unlikely that you can get away with ratcheting its normal activity down by force with no negative impact on the very brain function that researchers are desperately trying to preserve. It may just take longer than the brief six weeks over which the new drug’s first safety test was performed. The new trials are just starting to recruit patients at this writing; we’ll see how they fare in the clinic, and keep our fingers crossed for the people taking them.
Other scientists are pursuing a slightly different version of the same basic strategy. Some are developing drugs that inhibit beta-secretase, or that turn up the activity of alpha-secretase, with whose normal processing of APP beta-secretase interferes. Because such drugs are still in the early stages of development, we don’t yet know what their side effects will be, but again it seems unlikely that the activity of an enzyme produced normally throughout the body—and especially in the brain—can be altered without cost. For instance, one specter that already looms over the beta-secretase inhibitors (based on animal studies) is that they may make users more vulnerable to some psychological disorders. While very young animals with their beta-secretase genes knocked out seem to be physiologically more or less normal, they are timid and don’t like to explore their environment, and seem to run through serotonin (the chemical messenger whose metabolism is modulated by drugs like Prozac) abnormally quickly.
Freeing the Prisoners…or Letting Loose the Inmates?
Other scientists are pursuing an alternative approach that is, superficially, more in line with the engineering principles that I have advocated. Namely: to ignore the formation of beta-amyloid itself, and instead focus in on the process whereby beta-amyloid becomes aggregated into neuron-enmeshing plaques. A surprisingly large number of drugs and even herbal concentrates—from extracts of the spice turmeric (used in curries) to custom drugs with the highly marketable moniker beta-breakers—either interfere with the glomming-together of beta-amyloid into plaque fibrils, or even break existing aggregates apart…in a test tube. Most of these compounds never worked out once they got beyond the petri dish and into a living, breathing organism, but a few have been shown to reduce the plaque burden in animals genetically engineered to produce large amounts of beta-amyloid, and a few of those are now going into clinical trials in people with Alzheimer’s disease.
Here, however, we once again run into the problem of over-reliance on our hypotheses about what biochemical processes “cause” the disease. In the other amyloidoses, the connection between fibril and pathology is pretty obvious. Indeed, you can dramatically extend life expectancy in patients with several kinds of amyloidosis by “simply” replacing the amyloid-strangled organ with a transplanted one free of fibrils.
But the reality is that, despite a decade of intense research into the “amyloid hypothesis” of Alzheimer’s disease, we still don’t understand the mysterious metabolic underpinnings of the disease. Even among the majority of researchers who are convinced that beta-amyloid is the key to Alzheimer’s, the consensus around the detailed mechanistic role of the protein in the disease is as shallow as it is wide. Controversy continues to rage about what exactly links it—the protein itself, and/or the plaques that form from it—to the decay of brain and body that we see in its victims. Thus, the premise that beta-amyloid plaques cause Alzheimer’s—or link the underlying metabolic defect(s) to disease—is still not enough in itself to tell you what should be done about them.
If anything, the balance of evidence is that it may not be the plaques themselves that impair neurological functioning the most, but the soluble beta-amyloid oligomers: short chains, made up of just a few single beta-amyloid molecules (“monomers”) linked together in the same way that plaque fibrils are, but whose small size allows them to remain dissolved in the fluid bathing the cells of the brain rather than precipitating out into deposits. In cultured cells and in experimental animals, beta-amyloid oligomers derived from human nerve cells clearly disrupt neuronal function and interfere with normal memory, in ways that are not observed with either the beta-amyloid plaques that they form or the lone “links” of beta-amyloid monomers of which they are composed. After being microinjected with human-derived oligomers, rats become confused and forgetful. Beta-amyloid monomers don’t have the same effect, and while giving animals or brain cell cultures chemicals that clear all forms of beta-amyloid (oligomers and monomers) out of the fluid bathing the neurons prevents the negative effects of exposure to a mixture of oligomers and monomers, an enzyme that selectively degrades free beta-amyloid monomers while leaving the oligomers intact provides no protection.
Likewise, the rats’ memory function is largely recovered a day after introduction of the oligomers, when they have been cleared out by the animals’ natural protective systems. This again suggests that the oligomers, rather than the plaques that they form, are the guilty parties in cognitive function. And indeed, preliminary studies suggest that the oligomers act as the molecular equivalent of dust in the eyes of neurons, interfering with their ability to receive signals from other neurons and pass the signal on to their internal machinery. Adding to the uncertainty, one group have reported a mouse model in which abundant plaques form but no neurological deficits result.
Another series of experiments strongly suggests that the plaques are the result, rather than the cause, of the widespread death of neurons in the Alzheimer’s brain.11 This study looked in detail at the location of both soluble amyloid species and plaques in the brains of people who had died of the disease. Soluble amyloid was found accumulated in still-intact cells within their lysosomes (the cellular garbage disposal units discussed extensively in the last chapter). Areas with very high levels of amyloid showed evidence of the rupturing of neurons, with beta-amyloid and the lysosomes’ digestive enzymes dispersed outward from a central locus in a pattern that suggests nothing so much as a bomb blast. And wherever plaques were found, the researchers also found the remnants of a destroyed neuron’s cell nucleus in the debris.
The strong suggestion was that the cells had been trying to dispose of the toxic amyloid oligomers by dumping them into the lysosome. Remember, again from the last chapter, that the Alzheimer’s brain shows clear evidence of dysfunction of these organelles. Moreover, a lot of the beta-amyloid found inside brain cells is actually clustered in and around lysosomes, and the aggregates themselves are in fact naturally taken up and degraded in microglia (the immune cells of the brain), but at a rate that is too slow to keep up with plaque formation.12
This suggests that the burden of amyloid may eventually overcome the capacity of lysosomes to dispose of it, leading to the death-spiral of dysfunction outlined previously—and eventually to the death and rupture of the cell, during which the beta-amyloid and lysosomal enzymes spew forth from the dying neuron, creating plaque deposits like so much slag from a bombed-out building.
However, the innocence of plaques in Alzheimer’s disease can’t be asserted any more confidently than their guilt. The beta-breakers that have been shown to be effective in animal models (as opposed to just test tubes) do restore memory function as they break up the plaques. And again, a mere glance at the snarling webs of amyloid-enmeshed cells of the Alzheimer’s brain defies the observer to accept the notion that the plaques are harmless.
One theory that may reconcile these conflicting conclusions is the idea that, at least in the short term, the plaques’ most damaging effect on the brain is to act as reservoirs for the beta-amyloid oligomers. You may have done simple experiments involving solutions in junior high-school science class, in which a substance is dissolved at very high concentrations in a glass of water. Eventually, levels reach so high a concentration that the water can’t hold any more, and a crystal precipitates out of the solution.
But the teacher may have explained to you—or even demonstrated—that the crystal is not a static entity, but exists in a state of “dynamic equilibrium,” with some dissolved material continually precipitating out onto its surface to build it up even as some of its existing surface molecules are continually dissolving out into solution. The volume of crystal will remain constant at a given solution concentration, as the rate of dissolution out of the crystal remains equal to the rate of precipitation into it. But of course, if you add more water or dissolved material into the solution, the equilibrium shifts accordingly.
The same thing, more or less, may be happening with amyloid plaques. As the concentration of beta-amyloid monomers and oligomers increases, they aggregate into plaques, which keeps the level of dissolved oligomers lower than it might otherwise be, thus effectively reducing their potential toxicity to local neurons. But when the level of oligomers dissolved into the fluid is reduced, the aggregated oligomers dissolve back into solution, maintaining their signal-jamming influence in a toxic steady state.
This potentially creates a real dilemma for therapies designed to deal with beta-amyloid. Using drugs like beta-breakers to prevent amyloid fibrils from forming, or to break existing plaques apart, would free up beta-amyloid oligomers that would otherwise have been sequestered in the plaque mass—which would actually expose neurons to more oligomeric interference than just leaving the plaques alone or even letting them build up.
In fact, both things would happen at once—and this could well have parallels in the Alzheimer’s brain. It seems very likely to me that beta-amyloid plaques play—or eventually would play—a similar role in the brain of the Alzheimer’s patient to what, say, transthyretin deposits do in the hearts of people struck by senile cardiac amyloidosis: you can’t look at the mess of a victim’s brain and not suspect that the plaques have been choking neurons to death. While the evidence is strong that any intervention which results in an increase in neuronal exposure to soluble oligomers of beta-amyloid can be expected to damage the brain, simply leaving the plaques to grow larger and larger as the diseased neurons continue to produce beta-amyloid (and/or to rupture and spill unprocessed beta-amyloid out of their lysosomes) surely sets the brain up for an even greater problem further down the road.
These intriguing theoretical questions are the kind that excites the curious minds of scientists; moreover, answering those questions seems to many such scientists to be the natural way of identifying and developing treatment for this terrible disease. As for other kinds of aging damage, however, I believe that this assumption is flawed. Let me say it again: We do not need to understand in detail how aging damage accumulates, or by what mechanism it wreaks its havoc, in order to undo that damage. However exciting an intellectual challenge it may be to sort out the exact pathway that leads from the fatal nips and tucks on APP, to beta-amyloid formation, to plaques, lysosomal dysfunction, cognitive impairment, and neuronal cell death, the bottom line in terms of the biomedical challenge is that we have here a material that is clearly accumulating and altering the composition of our aging and diseased bodies. When I see that, I say: it must go.
You can probably guess, by now, what sort of anti-amyloid therapy I prefer. While the remaining concerns about the exact role of plaques and oligomers in the disease process make simply breaking apart the aggregated beta-amyloid a potentially risky strategy, that doesn’t rule out a solution based on removing the plaques in whole cloth. That would eliminate the source of the problem, no matter what step in the formation, metabolism, or aggregation of the constituent material is in fact the key to its toxicity. Such an intervention would be classic anti-aging engineering as I’ve envisaged it—if it could be done.
Fortunately, we should know very soon. A potential solution is already undergoing clinical trials.
“Immune” from Plaque: The Beta-amyloid Vaccine
As I mentioned earlier, researchers found evidence some time ago that microglial cells—the immune cells of the brain—slowly eat up and digest away beta-amyloid deposits from nerve cells. Unfortunately, it was clear that the rate of clearance was not nearly high enough to keep up with the pace of deposition in Alzheimer’s patients. But researchers guessed that this natural defense mechanism could be stimulated to greater throughput. The obvious way to do this would be as we do for other targets of the immune system: with a vaccine.
The vision: inject patients with beta-amyloid, and the silent sentinels of the immune system would be roused up in defense, seeing the protein as a foreign invader. The same forces that your body marshals against chicken pox or influenza would be mobilized for an all-out war on the brain-choking protein, churning out antibodies specific to it and inducing the brain’s microglial cells to go on a search-and-destroy mission against brain beta-amyloid.
This was an even more exciting idea than it might at first appear, because an anti-beta-amyloid vaccine would be expected to have an even greater impact against Alzheimer’s disease than had the vaccines used against earlier epidemics. Such vaccines had allowed us nearly to wipe out diphtheria, polio, and measles in developed countries by preventing new cases from emerging. The promise of a vaccine targeting beta-amyloid was that it would actually cure all but the most advanced cases of the disease. Armies of activated microglia, their appetites for beta-amyloid whetted, would actually consume and thereby remove the existing beta-amyloid deposits. Once those choking fibrils were removed, the brain’s normal structure—and thus, function—would be restored.13
Critically, this approach should work no matter what theory of the link between beta-amyloid and memory loss turned out to be correct. (It would not be the whole solution if other hallmarks of Alzheimer’s, such as the intracellular neurofibrillary tangles, were also contributing to disease progression—but SENS incorporates solutions to those other aspects too, as you’ve seen.) The vaccine cleared out the plaques—and also, it shortly turned out, the more soluble oligomers, even inside the nerve cell itself14—not by breaking them apart into their constituent elements, but by causing immune cells to internalize and digest them.15 Because no soluble beta-amyloid is released by such an approach, there is no risk of doing new damage as levels of the soluble form rise: beta-amyloid just goes away, clearing neurons of its malign influence no matter what its exact mode of toxicity might be.
It turned out to be a relatively easy matter to test this concept in animals engineered to develop a version of Alzheimer’s disease. While vaccines against viruses must be carefully modified to ensure that they are close enough to the real thing to set off the immune system’s alarms, but yet different enough that they don’t actually infect people with the disease, no grand feat of molecular engineering was used in the first test of the concept: mice were simply injected with aggregated human beta-amyloid.
It worked smashingly well from the outset. Plaques quickly regressed from the mice’s brains. The swelling and dysfunction of the neurites (the branches that transport chemical messengers from one nerve cell to the next) faded away, leaving healthy, functional units. The dense, inflammatory overgrowth of supporting cells around the neurons retreated.16 And memory function—evidenced by the animals’ abilities to find hidden platforms in flooded mazes—became more like that of younger, healthy animals.17,18
The company coordinating this work, Elan Pharmaceuticals, obtained results in many different models of engineered mice, each with a different mutation in the processing of APP. Best of all, the treatment appeared to be quite safe. Contrary to what had been feared, the immune attack on the beta-amyloid around the animals’ brain cells did not cause collateral damage to the fine network of supporting cells onto which the plaques were glued. Likewise, some had feared that the immune attack on beta-amyloid would be so aggressive that it would punch holes in the protective shield that protects the brain from the toxins in the bloodstream, triggering a flood of foreign substances into the brain from the rest of the body; but little evidence of such an effect was found. The FDA was so impressed with the results that it quickly gave Elan the nod to move their vaccine—code-named AN-1792—into placebo-controlled clinical trials.
Nearly 400 patient volunteers were recruited, of whom 300 received the amyloid vaccine and 72 were given injections of saline solution as a placebo control. Their baseline condition was determined, their mental state and functionality were assessed on a battery of neuropsychiatric and clinical tests, and a regimen of periodic injections was instituted.
Twelve months later, disaster struck.
A Fire in the Brain
Just a few months after this trial had begun, some patients started exhibiting serious side effects. Of more than 300 patients recruited from 28 clinical centers across Europe and North America, about one in 15 developed meningoencephalitis, a life-threatening swelling of the brain, apparently as a result of an overreaction of the immune system inside the brain itself.19
As soon as the side effect was discovered, the trial was halted, sending researchers scrambling to try to figure out what had gone wrong. The problem came as a complete shock. The vaccine had been tested in mice with a wide range of genetic abnormalities, each leading to Alzheimer’s-like plaque formation from a different defect in the synthesis or metabolism of APP, and no such side effect had been observed—this despite the fact that scientists had been much more aggressive in their treatment protocols with mice than they would dare to be with human patients.
How such a crisis could occur, after such careful preclinical testing, has been documented extensively in the media and the academic literature, and I think it’s too much of a digression to describe here. The important point is that researchers rapidly homed in on the problems with the first vaccine—and, as we shall see, on how those problems can be overcome.
Snatching the Ore from the Smelter’s Ashes
The first trial proved catastrophic for some patients, and it was terminated before science could assess the full range of effects of the vaccine—positive and negative. But despite—and in a way, because of—the constraints imposed by this serious adverse reaction, researchers were desperate to squeeze every available bit of data that they could from the trial, to make up for its human and financial costs.20,21,22 Through careful sifting of the data, scientists managed to collect some preliminary information suggesting that, despite the horrors of inflamed brains in a few patients, immunization with beta-amyloid fundamentally does work as a therapy in humans. And when combined with further animal studies, the findings also suggesed ways to avoid this side effect (and others) in future vaccines—some of which are either under development, or even in clinical trials, already.
In the immediate aftermath of the study’s sudden termination, information about the patients was still scattered throughout the twenty-eight independent clinical centers at which the patients had received treatment. Preliminary assessments of the readily available information were not promising: researchers saw little evidence of improvement in the memory or other cognitive functions of people in the trial. But as the dust settled and researchers began to collect and analyze volunteers’ full medical records, a new pattern emerged.23 A given vaccine (any vaccine) does not do the same to all its recipients: some people create a stronger immune response to it than others. By separating out those participants whose blood work showed that they had mounted a substantial antibody response to the injected beta-amyloid by the time the study was shut down (fifty-nine volunteers) from the rest (who had not), scientists were able to show that those who had responded well to the vaccine had, apparently, fared better.
This finding took time to emerge because it was initially obscured by the statistics. When the researchers looked individually at each of the cognitive tests that volunteers had been administered, they could find no differences that passed the test of statistical significance. However, an integrated analysis of the whole battery of tests suggested that people whose immune systems had responded to the challenge suffered less decline during the study period than had people administered the placebo—a difference that was only beginning to become clear at the twelve-month mark when the trial was halted. Of particular interest was the fact that the emerging difference was most apparent for a composite score on the memory tests. And most suggestively, there did seem to be a sort of “dose-response” effect, with the greatest improvements in scores on overall memory, immediate and delayed memory, a nine-component memory score, and possibly a test of “executive function” (that is, higher-order brain functions involved in governing ourselves in a goal-directed fashion in the beginning and over time) in those whose antibody responses were strongest.
These results were all the more impressive considering the likelihood that most of the active responders were suffering from low-level brain inflammation and “ministrokes” triggered by the vaccine—a possibility suggested not only by autopsy findings and animal studies, but by the fact that headaches and confusion were reported as side effects so much more often in subjects who had received vaccination than in those who had received the placebo. Indeed, studies in breeds of Alzheimer’s mice that had beta-amyloid deposits on their vasculature, and were thus vulnerable to micro-hemorrhages in response to the vaccine, still showed some cognitive improvements, despite the direct damage their brains suffered from the tiny bleeds. If your brain function is being preserved in spite of a quiet, chronic attack on it by your own immune system, the implication is that something else about your underlying clinical condition has improved even more. A vaccine that would clear out beta-amyloid (as this one appears to do) without causing the inflammatory side effects would therefore be expected to yield a much more robust improvement in the workings of the mind.
Another interesting finding came when researchers looked at the levels of the protein tau, which is the major constituent of neurofibrillary tangles, in the fluids bathing the central nervous system (the cerebrospinal fluid, or CSF). Even though they only had about ten subjects in each group with good enough data to compare, the medical teams did observe that the vaccine responders had lower levels of tau than did the patients receiving placebo. While it’s very indirect, this might be a sign of reduced rates of brain cell death, since high levels of tau in these fluids are associated with the death of neurons in people with the disease.
What about evidence on the intended effects of the vaccine—its ability to actually clear out beta-amyloid plaques? Unfortunately, it’s not yet routine for scientists to look into the brain to see the plaque burden in people’s brains before they’ve died, although new imaging techniques to do just that have been developed and are being tested for accuracy now. What can easily be done is to analyze levels of beta-amyloid in the cerebrospinal fluid. In animal studies, vaccination against beta-amyloid almost always results in an increase in CSF levels of beta-amyloid—a finding usually interpreted as a sign that the vaccine is helping the animals to clear the stuff out of their brains and then transport it elsewhere in the body for disposal. This effect was not seen in the small number of trial volunteers for whom scientists had samples that they could compare. But we have another, more direct source of evidence on the subject: the three vaccine responders who died over the course of the trial. While one must again be cautious in reading too much into the results observed in just three people’s brains, autopsies of these volunteers by independent groups did show a dramatic reduction in levels of plaques in key regions of each of them as compared with control subjects.
Moreover, pathologists examining these brains found microglia in close proximity to many of the plaques that remained, suggesting that the vaccine was working as scientists had always hoped: the activated immune system had successfully mobilized microglia to clear out the deposits of beta-amyloid. This conclusion was further bolstered by a study—only completed after the end of the human trial—showing that old Caribbean green monkeys given beta-amyloid vaccination exhibited huge (66 percent) reductions in beta-amyloid levels in the brain, and a complete absence of plaques, and that the dense tangling-up of neurons’ supporting glial cells usually seen in human Alzheimer’s patients was considerably reduced.24 This is an important piece of supporting evidence, because these monkeys develop some Alzheimer’s-type pathology naturally as they age, and are much closer relatives to us than any mouse. The researchers who reported this finding are currently working to develop ways to assess the monkeys’ cognitive function.
The Next Beta-amyloid Vaccine
All of this tantalizing, albeit inconclusive, information has once again erected the banner of beta-amyloid vaccination as a true cure (or at least a major component of a cure) for Alzheimer’s disease. We’ve learned enough about both the potential of the vaccine and the reasons for the deadly brain inflammation to develop a variety of new approaches to the basic strategy, one or more of which will almost certainly prove effective without putting its recipients at risk.
The key to designing a safe vaccine is, of course, to avoid enraging the immune cells that attacked the original subjects’ brains in the first trial even as the microglia were loyally cleaning out the amyloid plaques. There are several ways that this might be done, and each of them now enjoys support in animal models of the disease—supporting their efficacy while providing evidence that they will not have a deadly immune side effect.
One relatively straightforward approach that has already entered clinical trials is something called passive vaccination. Unlike a conventional active vaccine, in which the patient receives the offending agent itself (in this case, beta-amyloid) in order to signal the immune system to produce its own antibodies in response, passive vaccination involves directly providing the very antibodies that are desired, bringing out the immune response that the same antibodies elicit when they are produced by the body. The advantage of this approach is that it would allow scientists to choose which antibodies would be circulating throughout a patient’s body. These antibodies might be chosen, or even custom-made, to activate the type of response that sends the microglia out to break down and digest the amyloid deposits25 without the risk of eliciting the undesired antivasculature response.
The main disadvantage of this approach is that it would not induce the kind of semipermanent immune vigilance we’ve come to expect from vaccination against diseases like mumps or polio, but would instead require patients to receive regular reinjections of the antibodies to keep their amyloid-fighting supplies topped up. But even this apparent inconvenience has an upside: because the recipient’s immune system isn’t put into a long-term state of vigilance against beta-amyloid, physicians can stop treatment of an individual patient—or even stop an entire trial—at any time in case of side effects, without the fear that the body will remain in a destabilized autoimmune-like state of the kind elicited by the original vaccine.
Another approach is to manufacture an active vaccine composed of only part of the beta-amyloid molecule. When scientists looked at the immune response to active vaccination with the whole beta-amyloid molecule, they found that a mixture of different antibodies was being produced. Only a small number of those antibodies were of a type that harms the vasculature—and, significantly, these antibodies were only observed in humans administered the original vaccine, and not in the mice or monkeys that received the same vaccine and had not suffered the tragic assault on the brain. These antibodies were sensitized to only one segment of the total beta-amyloid protein, located in the middle of the molecule. By contrast, in mice, monkeys, and humans, most of the antibodies were of types that would mobilize the immune system against a completely different segment of the beta-amyloid molecule, located on its far end.
Since it’s reasonably clear that the antivasculature response is not only unnecessary (since it isn’t seen in the animal models with active or passive vaccination, yet these procedures dramatically clear out amyloid and improve memory) but is extremely harmful (due to its role in brain inflammation), it seems highly likely that a vaccine based on the above principles will work fine and be free of the risk of brain inflammation so long as it reacts only to this key subsection of the protein. Such a vaccine would induce the desired immune response on an active basis, without sending the body’s immune system on a misguided mission against the very brain in whose defense it was aroused.
To date, several vaccines have already been developed based on this principle. They work by binding the key segment of the beta-amyloid protein to other proteins or antigens, or by combining it with appropriate immune stimulants, or by joining several such segments together into a sort of beta-amyloid molecular pincushion, all in an effort to maximize the antibody response to the vaccine without initiating the overreaction. These vaccines have all been shown to reduce levels of beta-amyloid, and often of plaques—some of them using very convenient delivery systems, such as transdermal patches similar to the ones widely used for nicotine treatment, or else the kind of nasal spray now used for quick decongestant relief of stuffed-up noses.
And even this has not exhausted scientific creativity. As this book was in preparation, for instance, researchers at the University of Texas’ Southwestern Medical Center reported that injecting animals with the DNA for the most toxic form of the beta-amyloid proteins, smuggled under the skin inside tiny gold microparticles, led to production of the protein and to the body responding with a rigorous and apparently safe antibody response. The result: after receiving eleven injections over the course of several months, the Alzheimer’s mice enjoyed a 60 to 77.5 percent reduction of plaque burden.26
As I mentioned, the most well-studied passive vaccine is already in mid-stage clinical trials orchestrated by Elan Pharmaceuticals. And while this vaccine was the first out of the gate, the race to develop a clinically effective and safe vaccine to defeat beta-amyloid is very much on. Indeed, the question today seems to be not so much whether vaccination against amyloid will work, but which of these many ingenious strategies will ultimately prove to be most effective and safe. Surveying the progress made so far, we can say with a high degree of confidence that we should soon be able to harness the ancient powers of the immune system to cut our way through the mind-stealing webs of beta-amyloid, redeeming the captive minds that they bind.
Amyloid Vaccination: Beyond Alzheimer’s
I’ve spent so much time discussing the Alzheimer’s beta-amyloid vaccine that you may well have forgotten how I got started: by noting that beta-amyloid is just one famous example of a class of extracellular aggregating proteins (“amyloids”) associated with aging and with age-related diseases. And no, I’m not repeating the mistake of the National Institute on Aging in this regard, which is to throw nearly half of all of its resources every year into Alzheimer’s research, at the expense of its real mandate, which is (or ought to be) to find ways to treat aging itself rather than one particular age-related disease. Instead, I’ve been guiding you through the vaccine’s progress simply because it is in such an advanced state of clinical development. There’s every reason to think that we will be able to exploit the same sort of immunological strategy to tackle most of the other age-related disorders caused by coatings of extracellular junk.
Next to beta-amyloid vaccination for Alzheimer’s disease, the immunological amyloid-buster in the most advanced state of development is a vaccine for systemic AL amyloidosis, also known as primary amyloidosis. I briefly mentioned this form of amyloidosis early on in this chapter. This is the most common form of amyloidosis in the United States and some other industrialized countries—it strikes two to three thousand Americans annually. It’s the result of overproduction by a specific cell type, plasma cells, of immunoglobulin light chain (L—thus “AL,” for “Amyloidosis Light-chain”), a component of a class of antibodies.
However, I’m not going to spend much time on AL amyloidosis here, because it’s not an age-related amyloidosis and the amyloid involved may differ in important ways from ones that are age-related. The main difference is that it’s laid down so extremely fast that it may not have the same degree of problematic cross-linking as age-related amyloids. The main thing I do want to tell you about AL amyloidosis concerns an immunotherapy protocol for it that shows promising signs of being transferable to age-related amyloidoses.
The existing therapies for AL are decidedly inadequate. Until recently, the standard intervention was a regimen of high-dose chemotherapy designed to kill off the originating plasma cells, often combined with bone marrow transplants to replace some of the other blood cells that are destroyed in the process. More recently, a new chemotherapeutic drug called I-DOX has been used, after the serendipitous discovery that it could accelerate removal of systemic AL amyloid plaques through an as-yet unknown mechanism that apparently does not involve suppressing plasma cells. But even this new therapy tends to cause blood disorders, and both these treatments are only helpful for people with soft-tissue deposits, extending no benefit to the more serious cases involving the heart or kidneys. They’re also not terribly successful at saving lives, with a mortality rate of about 40 percent.
At the turn of the millennium, scientists in Dr. Alan Solomon’s lab at the Human Immunology and Cancer Program at the University of Tennessee Graduate School of Medicine developed a new animal model for AL, created by simply injecting mice with human light-chain amyloid extracted from the livers or spleens of patients who had died of the disease. The AL quickly began forming amyloid “tumors” in the animals, the size of which varied with the amount of material the animals were administered: at higher doses, the amyloid masses on their backs grew so large that they could be felt by hand.27,28
As they explored the effects of the disease on the animals, Dr. Solomon’s group demonstrated that antibodies against a segment of the unique “beta-sheet” conformation of the light-chain amyloid fibrils were partially breaking down the deposits, making them more susceptible to immune attack. When they took amyloid tumors out of mice, subjected them to such antibodies and then returned the tumors into the mice, the animals cleared them out twice as quickly as new tumors of similar size. Clearance was also speeded when the original, unaggregated amyloid extracts were pretreated with antibodies before being injected. Even immune-deficient mice cleared out the deposits more quickly when they were given antibodies along with the amyloid extracts.29 One such antibody—an immunoglobulin G1 antibody that they named 11-1F4—was found to have the strongest “homing instinct,” rapidly converging on tumors composed of either of the two major classes of human light-chain amyloid fibrils, whether in a test tube or in mice bearing them. Just as important, the antibody was found to target the amyloid specifically, without infiltrating tissues anywhere else in the body or samples of human tissue. And it also worked on pre-formed AL amyloid.
A Jack-of-All-Amyloids?
As I mentioned, the reason I’m telling you about 11-1F4 is not because of its effects on AL amyloidosis. The one disadvantage of Solomon’s model of systemic AL amyloidosis was that it’s an imprecise model of the human disease, with little spread of the deposits to major organs. Noting this, his group sought ways of testing it against other amyloid diseases, such as amyloid protein A amyloidosis (AA—also termed “inflammatory” or “secondary” amyloidosis), the most common amyloid disorder outside the United States.30 AA has the advantage of being easy to induce as a full-blown disorder in many strains of mice: you simply inject them with chemicals like silver nitrate that induce a strong inflammatory response, resulting in overproduction of serum amyloid A (SAA), a protein produced by the liver during times of inflammation. Like immunoglobulin light chain, SAA is only partially degraded by the body’s macrophages, resulting in the release of sticky, half-digested SAA fragments that tend to clump up and accumulate in the kidney and liver. This is the genesis of the disease in both mice and humans, so the amyloid disease that results when mice are given these inflammatory chemicals very closely mimics its human counterpart.
Surprisingly, the University of Tennessee team found that the 11-1F4 antibody also reacted to AA amyloids from mice—and that it cleared them out of affected mice.31,32 In fact, it was very nearly as effective against AA deposits as it was against the original AL target, with the average organ amyloid burden dropping by over three-quarters in liver and spleen alike! This may be because the molecular architecture underlying the different fibrils’ stickiness is similar, leading to a similar antigenic profile. It may also be related to the long-established fact that extracts of AL amyloid deposits accelerate the development of AA in response to inflammation in mice. If so, maybe the aggregating properties of the different amyloids are such that they can interact, with one serving as a sort of crystallizing center around which other amyloids gather. Think of a small deposit of cooked-on food on the surface of a pot. If the stain isn’t immediately scrubbed out, it becomes increasingly stubbornly imbedded on its surface by future cooking, and then begins to catch food particles from subsequent meals, slowly expanding into a larger and larger stain that is ever more difficult to remove.
Even more remarkably, the University of Tennessee team then went on to show that 11-1F4 could also react with, and remove, amyloids based on transthyretin (TTR), the thyroid-hormone transporting protein whose aggregation causes senile cardiac amyloidosis, and which is a cause of death in so many of the oldest humans among us today. If passive immunization with 11-1F4 can reverse the course of both of these major forms of amyloidosis, we are well on our way to clearing out some of the most important sources of extracellular junk in the population at large—as well, potentially, as other forms of amyloidosis that are less common causes of death only because our lives are currently so brief.
This is exciting research, and the next step for the antibody is clearly to test it out in humans with the corresponding amyloidoses. For this to be done, the antibody must first be “humanized.” Remember, while the light-chain amyloids that the antibodies remove are derived from humans, the agent that’s doing the removal is a mouse antibody, which would probably not interact well or safely with the human immune system. To overcome this problem, Dr. Solomon’s team “chimerized” the antibody, combining its antigen-seeking “business end” with a human “handle.” The resulting antibody still recognized both light-chain amyloid and amyloid protein A aggregates, and even cleared them out of mice just as the original vaccine had done. The results are so promising that the National Cancer Institute’s Drug Development Group has arranged for large-scale pharmaceutical production of the new antibody, with the intention of moving it into preliminary human clinical trials.
Open Possibilities
There is every reason to believe that this kind of immune-based, vaccination approach to amyloids, demonstrated in animal models of Alzheimer’s disease and three human amyloidoses (and now in clinical trials for the former), will also work in other cases of cellular bindweed. Take, for example, amylin, or “islet amyloid polypeptide,” whose amyloid-inducing properties I briefly mentioned early on in this chapter. Amylin aggregates accumulate on insulin-producing beta cells in the pancreases of nearly all people with type 2 (late-onset, non-insulin-dependent) diabetes. Either the aggregates or the soluble oligomers of which they’re composed appear to play some role in the gradual dying-off of beta cells that occurs as the disease progresses,33 leading to the failure of the body to produce enough insulin to keep up with the incessant surges of sugar that accompany every meal.
No one has yet tried to develop a vaccine to remove these deposits, but the feasibility of such an approach is suggested by the fact that amylin fibrils have been identified inside macrophages harvested from areas adjacent to the amylin deposits, where amylin accumulates without being fully degraded. Moreover, amylin fibrils are engulfed by and accumulate within macrophages exposed to them under test-tube conditions.34 All of this suggests that that the immune system mounts an attack against this form of extracellular junk just as it does against amyloid beta and the junk responsible for secondary amyloidosis—in which case, there is every reason to think that this attack could be strengthened with a vaccine similar to those currently in the pipeline for those other amyloidoses. The therapeutic promise of such an approach would be even greater if it were combined with a souping-up of the macrophages’ lysosomes with enzymes that are more able to digest the amylin fibrils—a job that cries out for the use of the LysoSENS approach that I discussed in the last chapter.
Other forms of amyloidosis could also fall before an infusion of targeted antibodies or other vaccines. And while the focus for drug development today is on treatments for specific amyloid-based diseases, this same research can be bootstrapped into the SENS agenda. Once it has proven its efficacy in Alzheimer’s disease, senile cardiac amyloidosis, and type 2 diabetes, the spinoff technology will allow for the rapid development of vaccines for more obscure amyloid deposits that today may go nearly unnoticed except in people with a hundred candles or more to illuminate their birthday cakes.
The fact that these therapies have moved so quickly from the laboratory into clinical trials (remember, results in mice with the first beta-amyloid vaccine were reported in 1999, and it was in clinical trials by 2001) suggests that we will be able to move even more rapidly in the future, when the first anti-amyloid vaccines have passed through clinical trials and have been successfully used in doctors’ offices all over the world.
Eventually, I envisage a protocol to keep our bodies clear of extracellular junk in which we might take a regular sequence of anti-amyloid vaccines, not unlike the standardized series now given in regular succession over the course of our childhood. The timing and frequency of administration of a given vaccine would depend on how quickly its target builds up to levels that impair function: we would get a “booster shot” of some every few years, while others would be administered only a few times in each century of a greatly expanded lifespan. Each time we took one of these vaccines, our cells and organs would once again live and function free of a specific species of molecular bindweeds, returning them to the literally unbound potential of youth.