Re-Engineering the Human Immune System
Swine Flu. Spanish Flu. SARS. Almost every year, it seems, there is a new virus to watch out for. Roughly thirty thousand Americans die annually from a new flu strain — meaning roughly one flu fatality for every two victims of car accidents — and there is always the possibility that we will do battle with a much deadlier strain of flu virus, such as the one (cousin to the current swine flu) that killed 50 million people in 1918.
Currently, our bodies’ responses are, almost literally, catch as can. The immune system has two major components. Innate immunity responds first, but its responses are generic, its repertoire built-in and its memory nonexistent. On its own, it would not be enough. To deal with chronic infection and to develop responses targeted to specific pathogens the body also relies on a second “acquired immune system” that regulates and amplifies the responses of the inbuilt system, but also allows the body to cope with new challenges. Much of its action turns on production of antibodies, each of which is individually tailored to the physical chemistry of a particular alien invader. In the best case, the immune system creates an antibody that is a perfect match to some potential threat, and, more than that, the acquired immune system maintains a memory of that antibody, better preparing the body for future invasions from the same pathogen. Ideally, the antibody in question will bind to — and ultimately neutralize or even kill — the potentially threatening organisms.
Alas, at least for now, the process of manufacturing potent antibodies depends heavily on chance, and a type of lymphocyte known as B cells. In principle, B cells have the capacity to recombine to form a nearly infinite variety of antibodies: roughly 65 different “V regions” in the genome can combine with roughly 25 “D regions” and 6 “J regions,” which further undergo random mutations. In practice, getting the right antibody depends on getting the right combination at the right time. Which combinations emerge at any given moment in any given individual is a function of two things: the repertoire of antibody molecules a given organism has already generated, and a random interplay of combination and mutation that is much like natural selection itself — new B cells that are effective in locking onto enemy pathogens persist and spread; those that do a poor job tend to die off.
Unfortunately, there is no guarantee that this system will work. In any given individual there may be no extant antibody that is sufficiently close. If there is a hole in a given individual’s repertoire, that individual may never develop an adequate antibody. Even if there is an adequate starting point, the immune system still may fail to generate a proper antibody. The most useful mutations may or may not emerge, in part because the whole system is governed by a second type of immune cell known as the T cell. The job of T cells is to recognize small fragments of viral proteins as peptides and then help the B cells produce antibodies. Like B cells, T cells also have a recombinative system, generating billions of different receptors, only a few of which will recognize a given viral antigen. In effect, two separate systems must independently identify the same pathogen in order for the whole thing to work. At its best, the system is remarkably powerful — a single exposure to a pathogen can elicit a protective antibody that lasts a lifetime; people who were exposed to Spanish flu in 1918 still retain relevant antibodies today, 91 years later. (See Resources) But the system can be hit-or-miss. That same Spanish flu claimed 50 million lives, and there is no assurance that any given person will be able to generate the antibodies they need, even if they are vaccinated.
For now, the best way to supplement the body’s own defenses is through vaccines, but vaccines are far from a panacea. Each vaccine must be prepared in advance, few vaccines provide full protection to everybody, and despite popular misconception, even fewer last a lifetime. For example, smallpox vaccinations were lifelong, but tetanus vaccines generally last 5-10 years. There is still no vaccine for HIV infection. And when it comes to bacteria like tuberculosis, current vaccines are almost entirely ineffective. What’s more, the whole process is achingly indirect. Vaccines work by first stimulating B cells and T cells in order to induce production of antibodies. They don’t directly produce the needed antibodies. Rather, they try (not always successfully) to get the body to generate its own antibodies. In turn, stimulation of T cells requires yet another set of cells — called dendritic cells — and the presence of a diverse set of molecules called the major histocompatibility complex, creating still further complexity in generating an immune response.
Our best hope may be to cut out the middleman. Rather than merely hoping that the vaccine will indirectly lead to the antibody an individual needs, imagine if we could genetically engineer these antibodies and make them available as needed. Call it immunity-on-demand.
At first blush, the idea might seem farfetched. But there’s a good chance this system, or something like it, will actually be in place within decades. For starters, as mentioned above, every T cell and B cell expresses a unique receptor that recognizes a very small piece of a foreign structure from viruses or bacteria, such as proteins. Advances in recent genetic technology have made it possible to reprogram B cells, directly or through stem cells, to produce antibodies against parts of viral or bacterial proteins. Similarly, a new clonal army of T cells that are genetically engineered to recognize parts of a virus or bacteria would help the B cells produce potent antibodies against soft spots of these viruses and other pathogens that would otherwise neutralize or kill them.
Already scientists at Caltech, headed by Nobel laureate David Baltimore, have engineered stem cells that can be programmed into B cells, which produce potent antibodies against HIV. Meanwhile, cancer researcher Steven Rosenberg at NIH has been engineering clonal T cells capable of recognizing tumors and transferring these cells to patients with a skin cancer called melanoma. His work has shown promising results in clinical trials. Together, these results could lay the groundwork for a new future, in which relevant antibodies and T cell receptors are directly downloaded, rather than indirectly induced.
Of course, many challenges remain. The first is to be able to better understand the pathogens themselves: each has an Achilles’ heel, but we’ve yet to find a fully systematic way of finding any given pathogen’s weakness, a prerequisite for any system of immunity on demand. It will also be important to develop structural models to artificially create the antibodies and T cell receptors that can recognize these regions. Eventually, as computational power continues to grow and as our structural biology knowledge increases, we may be able to design artificial vaccines completely in silico. For now, this is more dream than reality.
The real obstacle, however, is not the creation or the manufacture of protective antibodies against pathogens, but the delivery of those antibodies or cells into the body. Currently the only way to deliver antibodies into the body is difficult and unreliable. One needs to isolate stem or immune cells (B and T cells) from each individual patient and then custom-tailor the receptors for their genetic backgrounds, a process that is far too expensive to implement on a mass scale. Stem cells, nonetheless, do offer real promise. Already it seems plausible that in the future, bioengineers could create new stem cells from your blood cells and freeze them until needed. If there were to be a deadly new virus, bioprogrammers could design the potential immune receptors and genetically engineer and introduce them into your stored stem cells, which can then be injected into your blood. Eventually it may even be possible to deliver the immune receptor genes directly into your body, where they would target the stem cells and reprogram them.
At first blush, the idea of immunity-on-demand might seem farfetched.
All this is, of course, a delicate proposition. In some ways, an overactive immune system is as much of a risk as an underactive one: more than a million people worldwide a year die from collateral damage, like septic shock after bacterial infection, and inflammations that may ultimately induce chronic illness such as heart disease and perhaps even cancer. Coping with the immune system’s excesses will require advances in understanding the precise mechanisms of immune regulation. This fine-tuning of the immune response could also have the bonus effect of preventing autoimmune diseases.
We are not sure when this will all happen, but there’s a good chance it will, and perhaps the only question is when. There was a great leap forward in medicine when sterilization techniques were first implemented. Here’s to the hope that the fruits of information technology can underwrite a second, even bigger leap.
Derya Unutmaz is an Associate Professor of Microbiology and Pathology at N.Y.U. School of Medicine. His current research is focused on understanding the function of human immune system.
Gary Marcus is an author and a Professor of Psychology at NYU. His most recent book is Kluge: The Haphazard Construction of the Human Mind.