A little more than three years ago a medical team from Berlin published the results of a unique experiment that astonished HIV researchers. The German group had taken bone marrow—the source of the body’s immune cells—from an anonymous donor whose genetic inheritance made him or her naturally resistant to HIV. Then the researchers transplanted the cells into a man with leukemia who had been HIV-positive for more than 10 years. Although treatment of the patient’s leukemia was the rationale for the bone marrow transplant therapy, the group also hoped that the transplant would provide enough HIV-resistant cells to control the man’s infection. The therapy exceeded the team’s expectations. Instead of just decreasing the amount of HIV in the patient’s blood, the transplant wiped out all detectable traces of the virus from his body, including in multiple tissues where it could have lain dormant. The German researchers were so surprised by the spectacularly positive results that they waited nearly two years before publishing their data.

The news seemed too good to be true. And yet five years after undergoing his initial treatment, the so-called Berlin patient (who later disclosed his identity as Timothy Ray Brown of California) still shows no signs of harboring the AIDS virus—despite not taking HIV-fighting, antiretroviral drugs for all this time. Of the more than 60 million people who have been infected with HIV over the past few decades, Brown is so far the only individual who appears to have well-documented eradication of the infection.

The approach cannot be applied widely for many reasons, not the least of which is that the patient’s own immune system has to be destroyed as a first step—a very risky undertaking. But the unexpected success has inspired researchers around the world to see if they can find safer and less expensive ways to give patients a new, HIV-resistant immune system like the one Brown received. Such a feat would allow doctors to essentially slam the door on HIV, so that it could no longer spread from cell to cell in the body. Eventually the modified immune system would also be likely to clear any remaining HIV from various hiding places in the body. Rather than following in the footsteps of previous therapies that merely suppress the virus, a new approach that mimics the Berlin treatment would—if successful—eliminate the virus and potentially cure the disease.

In fact, the two of us and our colleagues think we might have an easier way to give HIV patients an immune system like the one underlying the Berlin patient’s successful treatment. The procedure has shown prom­ise in the laboratory, and we are now carrying out early clinical trials in a small number of HIV-infected people. We have much work ahead of us and cannot be assured that the therapy will be effective, but the Berlin patient’s continuing freedom from HIV and our own preliminary results make us feel the treatment we are developing could well be life-changing for millions of people infected with HIV today.

Fine-tuning the Immune System
Our approach to engineering an immune system to fight HIV builds on research that addressed two related challenges. Scientists needed to figure out how to turbocharge the immune system against HIV. And they needed a way to keep HIV from being able to enter its favored cells, CD4+ cells, also known as helper T cells. These particular T cells serve as the quarterbacks of the immune response by coordinating the interaction among many different types of immune cells. When HIV first infects a helper T cell, the virus does not cause any real harm. Then later, when the immune cell is activated to fight an ongoing infection, it instead spits out more HIV copies. Even more unfortunate, HIV eventually kills these coordinating cells as well, depleting the immune system’s ability to fight many other infections. In this way, HIV selectively eliminates the immune system’s best-trained players. As they decline, so, too, does the body’s ability to fight infections, until AIDS—the end stage marked by deadly infections—sets in.

Figuring out how to boost the immune system, let alone protect helper T cells, has not been easy. When news of the Berlin patient surfaced, however, progress had already been made on both fronts, albeit in separate lines of research.

For years scientists who study cancer, as well as those who investigate viral infections, have searched for ways to pump up the immune system—such as by taking T cells from a patient, exposing them to substances that cause them to both multiply and become more active against either cancer or viral infections, and then returning the juiced-up cells to the patient’s body. The two of us joined the effort 20 years ago, when Levine came to work with June at what is now called the Walter Reed National Military Medical Center in Bethesda, Md. Building on the work of others—notably Philip Greenberg and Stanley Riddell of the Fred Hutchinson Cancer Research Center in Seattle and Malcolm Brenner and Cliona Rooney, now at the Baylor College of Medicine in Houston—we began experiments to improve methods for growing T cells outside the body. At that time, T cells from a donor could be cultured in the laboratory only by using complex cocktails of chemical messengers or by extracting from the donor’s blood yet another type of cell, called dendritic cells, that normally instructs T cells to mature and multiply dramatically.

We thought we could simplify the process by creating artificial dendritic cells. Starting with tiny magnetic beads, slightly smaller than T cells, we attached to their surface two proteins that mimicked the molecules on dendritic cells. When mixed with T cells in laboratory flasks, the beads proved to be very efficient at their appointed task. By replenishing the beads every two weeks or so, we could keep a colony of active T cells multiplying happily for more than two months and increasing their numbers by a trillionfold.

When we began testing this approach using blood samples taken from HIV-positive volunteers, we discovered, much to our surprise, that the T cells we produced turned out to have a significant—albeit temporary—ability to deflect HIV’s advances. We published our results in June 1996 while still not knowing why our magnetic bead method for growing T cells would boost their resistance to infection with HIV. But later that year an important clue emerged that would ultimately help explain the mystery.

A Doorway to Infection
At the same time as we were developing our system for growing T cells, other researchers discovered a key flaw in HIV’s method of attack. Very early on in the AIDS epidemic, investigators had identified a small number of individuals who appeared to be highly resistant to infection with HIV despite having been exposed to the virus multiple times. Toward the end of 1996, in a scientific publishing frenzy, several laboratories reported that a particular protein, known as CCR5, which sits on the surface of helper T cells and certain other cells, acts like a doorway, allowing HIV to gain entry. Furthermore, researchers showed that people who naturally lacked the protein did not become infected [see “In Search of AIDS-Resistance Genes,” by Stephen J. O’Brien and Michael Dean; Scientific American, September 1997].

The absence of the doorway results from the deletion of 32 nucleotides (the A, T, C and G letters of the DNA alphabet) in the gene that codes for the cell-surface protein. The deletion results in a shortened CCR5 protein that is unable to make its way to the cell’s surface. About 1 percent of Caucasians have inherited two copies of this defective gene, dubbed CCR5-Delta32, making their cells highly resistant to HIV infection. The mutation is rare in
Native Americans, Asians and Africans, however. Apart from their genetic peculiarity, affected individuals appear to be heal­thy, although they may be more vulnerable to West Nile virus.

People who have inherited just one copy of the CCR5-Delta32 gene are still susceptible to HIV—but it takes longer on average for them to progress from initial infection to the later stages of disease. Investigators have shown that natural chemical messengers called beta-chemokines can block a normal CCR5 receptor—making it unavailable to HIV. Indeed, blocking the CCR5 receptor is the basis for an entire class of anti-HIV medications. Unfortunately, it is tough to keep all the CCR5 receptors on all cells that bear it continuously coated with enough of the drug so that HIV cannot gain entry to any of them. In addition, HIV can mutate to avoid the blockade, and these slightly altered viruses can still use the CCR5 doorway to get into T cells.

The discovery of CCR5’s role in HIV infection helped to explain why our artificially grown T cells proved resistant. Some­how the activation of the T cells by the beads caused the cells to shut down their production of CCR5 proteins. Without a working doorway, HIV was unable to enter the cells.

At that point, we wondered if we could use the CCR5 discovery, together with our newly refined method of growing T cells, to create a novel treatment for HIV. This idea led to a collaboration with Kristen Hege and Dale Ando, both then at the San Francisco–based biotechnology company Cell Genesys, to take an early step: conducting human clinical trials to ascertain the safety of T cells that had been genetically modified to seek out and attack HIV-infected cells—T cells that had also been expanded using our magnetic bead techniques. The cells proved to be safe all right and survived for years after infusion. The specific genetic modification we studied had only a modest effect, however, on HIV replication in patients. Cell Genesys eventually shut down the effort.

Engineering an HIV-Resistant Cell
By 2004, a few years after the two of us had moved to the University of Pennsylvania, Ando came to visit us in our new digs and to propose a second experiment. His new employer, Sangamo BioSciences, had recently developed a technique for cutting the DNA strands of genes in carefully selected places. This method was fundamentally different from and far more efficient than other approaches because it was able to target a specific gene sequence for editing. Previously researchers had no good way to control which genes, or sections of genes, were changed.

The Sangamo technology that Ando was talking about depends on two types of proteins to delete a section of a gene that is already in place. The first type are zinc finger proteins, which are naturally occurring molecules that bind to DNA during gene transcription, the process in which the information in the DNA molecule is converted into an RNA molecule needed for the synthesis of an encoded protein. Humans produce approximately 2,500 different zinc finger proteins, and each one binds to a different, specific nucleotide sequence on the DNA molecule.

Over a period of years scientists worked out a way to design and artificially construct zinc finger proteins able to latch onto any particular DNA sequence of interest—such as, for example, a section of the CCR5 gene. Ando proposed that Sangamo create a customized set of DNA scissors first by creating zinc finger proteins that would attach to either end of a sequence that we wanted to delete. Then to each of these proteins, company scientists would add a second protein, an enzyme called a nuclease, able to cut DNA strands in two. The zinc finger part of this complex would identify the sections of the DNA to cut, and the nuclease would snip the genetic material. By developing the right pairs of zinc fingers, Sangamo could target just the particular section of the CCR5 gene that we were interested in—without accidentally damaging other genes.

Once these designer zinc finger nucleases had bound to the DNA sequence in question, the cell’s own repair machinery would take over. This machinery would recognize the break and rejoin the severed pieces of DNA, chewing up a few nucleotides or adding some extra ones in the process. Thus, the repair process itself would help to further ensure that the slit gene would be unable to give rise to a working copy of the CCR5 protein.

After Ando finished his proposal and left our lab, one of us (June, who is usually highly optimistic) turned to the other and said, “Yeah right, like that’s gonna work!” But it was worth a try. Beyond being very specific for the CCR5 deletion, the zinc finger system was appealing because the proteins need only a short time to function and leave no residual trace in the cell.

Hopes Bolstered by Berlin Patient
We had already received permission from the FDA and the National Institutes of Health to start safety studies in humans when news broke about the apparently successful treatment of the Berlin patient—giving us more reason to think that infusing T cells with mutated CCR5 genes into patients could deal a significant blow to the HIV in their bodies. In particular, Gero Hütter and his colleagues reported that they had been able to conduct what was, perhaps, a once-in-a-lifetime experiment. One of their patients, who had been HIV-positive for more than 10 years and was doing well on antiviral drugs, developed acute myeloid leukemia, which was unrelated to his HIV infection. He underwent chemotherapy, but the cancer came back. Without a bone marrow transplant, in which the immune system of one person (including all the T cells) is essentially re-created in another person, he would die.

Hütter searched the European databases of potential bone marrow donors, looking for an individual who would match his patient’s HLA markers, a group of proteins (the human leukocyte antigens) that the immune system uses to distinguish its own tissues from that of other creatures. Matching a transplant recipient’s HLA type is vital to keep the transplanted cells from viewing the new host as foreign and attacking its tissues (a condition known as graft versus host disease) and to prevent rejection by any residual components of the patient’s previous immune system.

Hütter did not stop there, however. He hoped to find someone with the right HLA markers whose cells also naturally carried two copies of the CCR5-Delta32 mutation. A bone marrow transplant from such a person might conceivably provide an HIV-positive recipient with a new immune system that was resistant to virus that continued to persist.

Amazingly, after Hütter searched the databases and tested genes from more than 60 potential donors, he found a candidate who fit the bill. (The search was complicated by the fact that the HLA region varies so much from individual to individual and the HLA genes and the CCR5 gene are on different chromosomes.) This discovery was a lucky break considering that so few people have the CCR5-Delta32 mutation in both copies of their CCR5 gene in the first place. Fortunately, the Berlin patient also had a very common HLA pattern. (To give an idea of just how rare this combination was, researchers across the globe have tried to replicate the German experiment and have yet to find any individuals with the right set of HLA markers and CCR5 mutations.)

In the end, the Berlin patient needed two bone marrow transplants from the donor to cure his leukemia. Strikingly, more than five years after the transplant operation and in the continued absence of antiretroviral drug therapy, physicians have been unable to detect any HIV in his blood, liver, gut, brain, lymph tissues or plasma, using the most sensitive molecular tests available. No one knows whether HIV was truly eradicated from every tissue in the Berlin patient’s body, achieving what is known as a “sterilizing cure,” because HIV can insert its genes in the chromosomes of various cells [see “Can HIV Be Cured?” by Mario Stevenson; Scientific American, November 2008], allowing it to lie dormant for many years. Also unknown is whether total destruction of all HIV in his body is necessary if his immune system is now capable of dispatching any infection that might reemerge, meaning he is “functionally cured.” At any rate, the patient no longer has to take antiretroviral drugs and is free of detectable virus. (Of course, he still has to take medication to maintain the health of his bone marrow transplant.)

Unfortunately, the German experiment may prove to be the sole example of a bone marrow cure for HIV for years to come. Not only is the right combination of HLA and genetic mutations in donor and recipient extremely rare, this particular approach is very expensive (bone marrow transplants incur minimum costs of $250,000 at our hospital), requires an intense regimen of chemotherapy, a risky bone marrow transplant and a lifelong regimen of antirejection drugs. In effect, the Berlin patient has traded one set of problems—HIV infection (and leukemia)—for another—being a transplant recipient. Most people who are able to lead more or less healthy, productive lives on anti-HIV drugs—albeit with significant side effects and lifetime costs—would hesitate to make a similar trade. Of course, because the Berlin patient had developed a deadly leukemia, he had no choice.

Although we were buoyed by the Berlin findings, we also knew that the CCR5 deletion in the donated immune system might not have been the only reason for the patient’s apparently HIV-free state. Perhaps the patient’s reservoir of dormant HIV particles was drained during the years of treatment with anti­retroviral drugs. Or perhaps the patient had no residual HIV left after his original immune system was destroyed in preparation for the transplant. Or perhaps the one instance of life-threatening graft versus host disease that the Berlin patient suffered during his treatment also destroyed any remaining HIV-infected cells before the reaction was brought under control with medication. (No HLA match is ever 100 percent perfect—except among identical twins.) Still, the CCR5 deletion remained the most likely explanation for the transplant’s success, and so we eagerly plowed on with our own experiments.

Clinical Trials Are Under Way
When news of the berlin patient came out, Sangamo had, as promised, developed a set of zinc finger nucleases that targeted a spot near the key 32-nucleotide sequence of the CCR5 gene. (Because the goal was to disable CCR5, it did not matter if we reproduced the naturally occurring genetic mutation exactly as long as the resulting protein stopped functioning.) With Elena Perez, then a postdoctoral fellow in the lab, we had shown that the HIV infection itself could, ironically, aid the process of reshaping the immune system to become more resistant to the virus. Our laboratory experiments demonstrated that even when T cells whose CCR5 genes had been disabled by zinc finger nucleases were initially present at low frequency in cultures, the altered cells were able to replenish and stabilize the T cell population after exposure to HIV; in contrast, nonedited T cells that still contained CCR5 receptors were destroyed by HIV. In other words, HIV killed the vulnerable T cells, leaving behind more and more of the CCR5-deficient T cells, which are exactly the cells that are resistant to HIV and can thus do their job as immune cells and provide protection from infections.

Our preliminary results in a safety trial in people have also been encouraging. Under the guidance of Pablo Tebas, the physician leading our trial in Philadelphia, the first patient received his CCR5-modified T cell reinfusion in the summer of 2009. Since then, we have treated an 11 additional HIV-positive volunteers in a study sponsored by the nih. Sangamo is conducting a similar study on the West Coast. Although these safety studies by their very nature are not designed to prove whether a treatment is effective, we have observed that the number of helper T cells measured in ongoing blood tests has increased from baseline in all patients to date, a sign that the treatment probably is protecting T cells. In addition, helper T cells that lack a functioning CCR5 receptor have been detected in the lymphatic tissue of the intestines and in the blood. (These cells could have derived only from the reimplanted cells that were modified by the zinc finger nucleases.)

The next step is to test the ability of the newly altered immune cells to fight off the HIV particles that are already present in the body. We are employing a well-accepted, if nonetheless daunting, strategy to do so. Under close monitoring by study physicians, we plan to stop our volunteers’ anti-HIV medications to see what happens. When we did this for 12 weeks with one of our treated subjects, who had inherited a single CCR5-Delta32 gene (thus giving him a slight natural advantage), we found no evidence of the virus in his blood or lymph tissues at the conclusion of the three-month interruption of antiviral medications. The more recently treated patients are in the midst of their postinfusion regimen and follow-up, with completion of these visits over the next year. Additional clinical trials to test the efficacy of this novel technology are planned. If we are ultimately successful, the zinc finger nuclease approach could be significantly less expensive than either the rare CCR5-deficient bone marrow transplant or a lifetime of anti-HIV drug therapy.

Only a few years ago the idea of developing safe, effective and less expensive therapies that offer long-term, drug-free control of HIV was a vision that few of us even dared to dream. Even if our custom-designed zinc finger nucleases are not a cure, we believe they could be the closest anyone has come to locking out HIV in 30 years. 

This article was published in print as "Blocking HIV's Attack."