Interview with Dr. Robert Gallo

This is the third oral history interview with Dr. Robert Gallo of the National Cancer Institute concerning the history of AIDS at NIH. The interviewers are Dr. Victoria A. Harden, Director, NIH Historical Office, and Dennis Rodrigues, program analyst, NIH Historical Office. The interview takes place on 8 June 1995 in Dr. Gallo's laboratory in Building 37, Room 6A11.

Harden: Dr. Gallo, it has been some time since we conducted the earlier interviews. We will try to pick up where we left off last time. We were talking about your laboratory's research on the basic molecular biology of HIV. I am interested in some of the spin-offs of the research. I know that you found one or two new human herpesviruses which might be cofactors in AIDS. I would like you to describe the whole idea of cofactors and whether these new herpesviruses are now viewed as cofactors in HIV infection.

Gallo: Just talking generally, the spin-off in terms of the herpesvirus actually did not come out of molecular biology; it came more out of virology/biology experiments with a design, in fact, a plan, to discover new herpesviruses. As usual, the general idea bears fruit, but it is never for the reasons that you predicted. I will come to that in just a second and then we will talk about cofactors.

But, in terms of molecular biology spin-offs, I think the biggest ones are the discovery of new genes within HIV, within the genome of HIV, that have rather novel molecular mechanisms of action, that have, in fact, contributed to molecular biology. Some of those new genes have also become targets for people trying to interfere with HIV replication. Some of them are genes, the tat gene being an example, essential to HIV replication.

Let us turn to the issue that you wanted to talk about, the cofactors. Many people have the idea that if you do not always get a disease from a microbe that causes the disease, then either the microbe does not cause the disease, or something else is required as a cofactor. But this is not true. A microbe may cause disease in a very small percentage of people. That is usually the case. In fact, the usual case is that it causes no disease at all. The determination of whether somebody gets disease or not from a microbe depends on a very large number of factors.

The number one factor is, of course, the nature of the microbe. Some are efficient in causing disease, some are not.

Second, it sometimes depends on the host. Most of the time it depends on genetic factors in the host. Sometimes it depends on chance events–the dose–and if that seems trite, there are experimental examples that actually prove that in a number of systems.

Staying with retroviruses, it is clear that in nature–I probably used this example in earlier discussions, but I will use it again–when a cat gets infected by the feline leukemia retrovirus, it usually does not get leukemia; it is usually carrying the virus without leukemia occurring. My belief is that if cats lived long enough, let us say, for 100 years, the majority would get leukemia. There is a chance of genetic events occurring due to the integration of the provirus that eventually leads to leukemia. But it is known that if you inoculate the right dose in a young enough kitten, you get leukemia all the time. This is typical. The same is true with chicken leukemia retrovirus. If you inoculate newborn chicks with a proper dose, most will get leukemia. But, in nature, when chickens get infected as adults, it is unusual for them to get leukemia. So, it sometimes depends on the age of the individual, or on the age of the organism, or on the dose of the microbe. These are chance events. People often do not appreciate that.

The third is that sometimes it depends on the genetics of the host and how it handles a given microbe.

The determinants, for example, of HIV, are still unknown, but it could be that there is so much virus variation that different variants may have different virulence. But all the other factors I have just listed may be important.

Now, in my mind, when we talk about cofactors, we have to give a definition. Most people, when they think of cofactors, start to believe that you mean something that is absolutely required to get the disease. I would divide them into categories. I would say that there are essential and non-essential cofactors or, maybe it would be better, to call something a cofactor if it is truly required. Other things are just catalysts or promoters of the probability of getting disease and of the probability of the disease being more rather than less vicious, or more acute rather than it taking a long time to develop the disease. Let us go directly to HIV.

Many people have proposed cofactors for AIDS without data, and sometimes without even ideas; just claiming that there must be cofactors. But there is no evidence for any cofactor. Can HIV alone produce AIDS? I believe so.

Have we ever argued for a possible cofactor? Yes, with the qualification I just told you, that it is obvious that some things will promote progression and some things will inhibit progression. One of those things may be the genetics of me versus you. We can say dose is a factor that can lead to progression, or lack of it, and at a greater or lesser rate. But we have argued for certain herpesviruses as possibly being a factor in promoting AIDS progression. Several groups have argued for cytomegalovirus because it does do things and it does activate more HIV in some subtle settings.

In the middle of the 1980s, we became aware that the lymphomas that were associated with HIV infection were perhaps one-third of the time EBV-positive. Epstein-Barr virus, as you know, can immortalize some B cells and, when you have EBV-positive lymphomas, generally they are the kind of lymphomas that, more or less... If they do not require EBV, EBV makes the probability of getting a lymphoma much greater, because the cell cannot die easily. It is immortalized. Other genetic events are needed to develop the lymphoma, but the immortalization of the cell is perhaps a key factor that makes it probable that it will be an EBV-containing cell that is the one that will become a lymphoma.

What about the two-thirds of lymphomas in HIV-infected persons that were not EBV-positive? We wondered if there were herpesviruses yet to be discovered. We looked in the B-cell lymphomas of patients with AIDS who were negative for EBV and we discovered the first new herpesvirus in 25 years, and the first herpesvirus of man that targeted predominantly the T cell.

We had a new herpesvirus, but it was not involved in the lymphoma, at least not as far as anybody knows, even today. We even misnamed it. We called it HBLV, because we found it in a B-cell lymphoma. Then we studied it more intensively and determined that it primarily infected T cells, not B cells, which was an unexpected finding. We learned that it killed T cells when it replicated. Then we learned that it infected natural killer cells and, when it did so, it made those cells attack other natural killer cells. We learned that it could infect the same cell as HIV and activate HIV expression. Next we learned that it infected CD8 cells and activated the gene for CD4, the only known biological agent I am aware of that activates the gene for CD4. Now, the CD4+/CD8+ cells could be targets for HIV.

It was at that stage we proposed that the herpesvirus might be a cofactor for progression of AIDS. It was then that I started to be careful of the use of these words and called it a “catalyst” for progression, that is, a nonessential cofactor, but something that makes disease progression go faster and also makes it more probable that immune deficiency will develop.

We put that idea out and it got a little bit of a reception by [Dr. Larry] Corey in Seattle, and by [Dr. Donald] Don Carrigan at Wisconsin. But then [Dr. Harold] Jaffe published a paper, the data of which we already had in hand. I think that paper–by Jaffe and his colleagues at the CDC [Centers for Disease Control and Prevention]–was not a sophisticated look at the problem. Namely, they said, “Look, everybody has antibody, so how can it be a factor in progression?” That is like saying a cytokine like TNF [tumor necrosis factor] is not important in disease pathogenesis because everybody has it. The question is, if 90 percent of the human population has it, they also have EBV, but EBV can cause Burkitt's lymphoma under certain settings. The question is, does it get activated in an immune-suppressed individual?

We put the problem aside for a while because we did not have a quantitative assay to measure the amount of herpesvirus; only this antibody that indicated a previous exposure to the virus, which everybody showed. We argued, however, when we presented it, that we needed to have a quantitative assay for virus in blood and the amount of human herpesvirus-6 [HHV-6] DNA in lymphocytes circulating around.

At this time we learned of Carrigan's work. He reported in a few clinical papers, that sometimes in immune-suppressed people, following transplantations, he saw an enormous amount of human herpesvirus-6 replication and that he believed it was responsible for some of the bone marrow abnormalities in such people. He showed a lot of virus in bone marrow. Second, he pointed out and emphasized that interstitial pneumonia is the cause of death in 10 percent of the deaths of HIV-positive people. No one knows the cause of that interstitial pneumonia, and he found the lungs of those who died loaded with human HHV-6. He presented at our laboratory meeting that he thought it was very likely that HHV-6 was the cause of those deaths.

Meanwhile, before this, Japanese workers had shown that HHV-6 was the cause of roseola infantum, also known as exanthem subitum, a disease of infants, with fever and rash, but usually with not much more.

So now what is new? I have discussed with my colleague [Dr.] Paolo Russo that the only way we are going to get any proof of this, or get stronger support, is if we get a specific inhibitor that does not inhibit HIV, inhibits HHV-6, and as far as we know does not inhibit anything else, is relatively non-toxic, and then show that patients do better rather than worse.

Another way of doing it would be to find an animal model not infected with a parallel virus to HHV-6 which can be infected by SIV [simian immunodeficiency virus]. SIV can induce some immune deficiency–not the acute sort–but there are monkeys in which SIV induces nothing, there are monkeys in which some strains of SIV induce an acute AIDS, and there are monkeys where some strains of SIV induce a disease more similar to the human disease, where it takes time.

We used such a system. This is not published data. With the SIV alone, there was a little immune deficiency, and with the HHV-6 alone nothing, but with the two together they got it. I think we have proven the point with that rhesus virus and that we can publish that soon. So, I believe human herpesvirus-6 is a factor in AIDS progression.

Harden: Thank you. I would like to turn, for a moment then, to another...

Gallo: But wait. One last sentence about this, and to jump away from it, it will be interesting. There are other people who have suggested things [as cofactors], like Montagnier suggested mycoplasma. He has absolutely no evidence for it, nor it appears has he made any attempt to get evidence, and that has been going on for six or seven years. That is where we need a little more critiquing in the field. You cannot just keep talking about what could be, or what is theoretically possible. Mycoplasma can do things in culture, but so can anything; so can acid and certain things you are exposed to many times. If you put a lot of bacteria in a culture, it will kill the culture. This is not meaningful. You need to do the epidemiological studies, because the mycoplasma is not ubiquitous, like the herpesvirus, so you could certainly do a prospective study and prove it one way or the other. Basically there was not any more reason to suggest it than that. It is a common tissue culture contaminant and it is a common contaminant of an immune-suppressed person. But no one has ever shown that these things have any immunosuppressive activity in vivo, and the effects in vitro are very mixed. There is no epidemiological link and there is nothing else, so the idea sits there.

I do not even know right now what are the other claims for cofactors. I do not believe there are any that I can think of. Some people argued for the Kaposi's sarcoma drugs, the etoposide, but the epidemiologists tell me none of those data hold up.

Harden: Let us move then from herpes to Kaposi's sarcoma, which you have also worked on extensively. Maybe you can begin by briefly describing how Kaposi's differs from other tumors. Apparently it has characteristics suggestive of an infection. People have called it “perhaps” an immunological disease. Then I would like you to tell us how AIDS related Kaposi's sarcoma differs from other Kaposi's.

Gallo: The latter is easy enough. AIDS-related Kaposi's sarcoma differs from the others in not too many ways, except that many clinicians say it is one of the most aggressive forms. Obviously the link with HIV infection and maleness is a peculiar factor among all Kaposi's sarcomas, but even more so in HIV-Kaposi's and the strange story of gay men. They have much, much more Kaposi's. That is what is special about it. That is easy to answer.

The first question about the nature of Kaposi's sarcoma, what it is, and all the interesting things about it requires a longer answer. I find myself, if I am talking to a general audience, saying that if you start with the emotional side, it is bad enough to have a tumor, but when you have to watch it disfigure you and grow on your face and body, it adds an extra dimension of horror. This sarcoma has an unusual aspect that way.

continued on Page 02

Back To Top | Photography Credits