Dr. Chin-Yih Ou



MILLER: This is Dr. Bess Miller, and I'm here with Dr. Chin-Yih Ou. Today's date is November 17, 2017, and we are in Atlanta, Georgia, at the Centers for Disease Control and Prevention [CDC]. I am interviewing Dr. Ou as part of the oral history project, The Early Years of AIDS: CDC's Response to a Historic Epidemic. We are here to discuss your experience during the early years of CDC's work on AIDS, the Acquired Immunodeficiency Syndrome. Dr. Ou, do I have your permission to interview you and to record this interview?

OU: Sure.

MILLER: Dr. Ou, you have had a distinguished career in HIV [human immunodeficiency virus] laboratory science. When you first came to CDC in 1986, you took the lead in applying the newly emerging methodology of polymerase chain reaction, or PCR, to the diagnosis of HIV, really a revolutionary event. You 00:01:00went on to work on this methodology to allow the diagnosis of HIV in infants. Your work in the use of DNA [deoxyribonucleic acid] sequencing to look at the genetic diversity of HIV proved invaluable in identifying HIV transmission in the so-called Florida Dental investigation. Your work has been published in Science and the New England Journal of Medicine, among others, and you have been awarded the highest science award at CDC, the Charles C. Shepard Science Award. Thank you for agreeing to do this interview.

OU: Thank you for having me.

MILLER: Let's begin with your background. Would you tell me about where you grew up and your early family life?

OU: I grew up in Taiwan. My parents were actually from Mainland China. They fled away from the communists taking over China to Taiwan for (their) children, and 00:02:00actually I was the fifth. I was born in Taiwan and lived my life mostly in Taiwan until I came to the United States.

MILLER: Did you have scientists in your family? Do you remember what inspired you in terms of what you wanted to do in life?

OU: Yes. Actually, my parents were teachers. They were teachers of Chinese literature and things like that, and they taught in high school. Of our six siblings, two of them, well, three of them including myself, were scientists. Most of our siblings were educated in the United States for higher education. They got their Ph.D.s here, and one of my sisters actually got a degree in 00:03:00literature. She taught Russian language in Taiwan, and the last youngest sister, just like my parents, taught Chinese in high school.

MILLER: Where did you end up going to college?

OU: I studied medical technology in National Taiwan University, and later on I moved on to study microbiology and earn a master's degree there also in National Taiwan University. We had to have a mandatory army training for two years, so I was in the Taiwanese Army for two years before going back to graduate school. 00:04:00Actually, between those two, I also worked as a laboratory director in a small clinic hospital in Taiwan.

MILLER: What brought you to the United States?

OU: I went to Tennessee to the Oak Ridge National Laboratory to study molecular biology back in 1977. I actually stayed there until 1982 and then two years post-doc to 1984.

MILLER: That was a big jump. What inspired you to do that?

OU: As I said, I started my career as a laboratory scientist after I finished my army duty, but I found that it was kind of boring. I didn't want to go day in 00:05:00and day out just doing clinical laboratory work, so I decided to go back to graduate school.

MILLER: And the U.S. was the place to do that?

OU: Most people actually at that time came to the U.S. for higher education for their Ph.D. and so forth. As I said, four of our siblings came to the United States and one went over to Europe, so we were all over the place.

MILLER: At Oak Ridge for your Ph.D., what was your area of concentration? What was exciting to you or what was in vogue that you ended up studying there?

OU: I was working on murine leukemia virus; this is one of the retroviruses. At 00:06:00that time it was interesting to me because I just wanted to do something new. [When] I came to the graduate school in the beginning, I thought I wanted to do enzymology. After a three-month trial in the enzymology laboratory I found that it was way too difficult for me, so I wanted to move into something different. Virology and molecular biology at that time were just starting. I had to learn everything myself because things were new and not that many people knew how to do it. So I had to learn everything looking at the literature, the newly published literature, and try [to do it] myself. That actually helped me a lot 00:07:00later on in my CDC career, because I was doing retrovirology and HIV is one of the retroviruses. I learned all the basic techniques from cloning, sequencing, and everything, so I could quickly apply whatever I learned to HIV. The timing was just perfect.

MILLER: You moved to CDC in October of '86 to work in the Laboratory Investigations Branch in the Division of HIV-AIDS. First off you were tasked to establish a new microbiology molecular biology lab, and you eventually became chief. What was involved in setting up this molecular biology lab?

OU: Interesting you mention that. Actually, before I came to CDC I was thinking about going back to Taiwan, because at that point I had accepted a position 00:08:00there. I went through an interview, and then the CDC job came up. I happened to see that advertisement in Science, and my wife said, why don't you apply for it? She helped me actually send out quite a few applications besides CDC. I got only two replies. The CDC position was one of them, and I came over to talk to Dr. [Gerald G.] Schochetman, who actually came a year before me. He tried to also set up a new direction for the research, the application of work here in his new capacity. We talked, and he wanted me to look into the HIV genetic information in patients. Actually he wanted me to use a new technology that was at that time 00:09:00published by NIAID [National Institute of Allergy and Infectious Diseases] scientists. We were about to go for that, but right before I came to CDC to work, I happened to read a couple of papers. One of them was just published about the same time, and I was astonished by the new technology, PCR. I thought at that point, if I had had this technology available to me, I could have cut down my Ph.D. research by at least two or three years because it was so fantastic. With my background in clinical microbiology, I could see that that technique could be used right into the HIV field to diagnose HIV. Jerry was very 00:10:00nice. He gave me a new laboratory, and this laboratory had nothing in it at that point. I had to order the chairs, the chemicals, everything from scratch. It was a very fantastic experience for me. In the beginning, as I said, we decided to go for the technique that was just published by NIAID people.

MILLER: What technique was that?

OU: This was a technique for in vitro hybridization very similar to PCR, but it was looking into the cells to see whether they harbored HIV sequences, RNA [ribonucleic acid], and that kind of thing. Then you combined immunocytology to have a technique to also identify what kind of cells they are, T4 [lymphocyte] 00:11:00cells or macrophages. To me, I thought it was a very labor-intensive technique, and you had to see the cell morphology. It also happened that they told us that I had to go to their laboratory for six months to learn this thing. We decided that that was just way too long. So Jerry and I decided, sure, we'll just move into PCR right away.

MILLER: We're talking about trying to do a more efficient way of diagnosing HIV infection. What was the state of the art at that time before using PCR or moving 00:12:00into the direction that NIAID was moving?

OU: Certainly at that point, the way of diagnosing HIV infection was using antibody. If you have antibody to HIV, that means you're infected by the virus. But there are many other occasions [wherein] antibody presence doesn't mean that there's infection. For instance, in infants, the maternal antibody is present for at least 9 to 18 months in many infants, so you cannot do that [use the presence of antibody to diagnose HIV infection]. If you wanted to see the real presence of HIV, it had to be done through culture technique, and culturing viruses takes a long time. At that point you needed to have HIV culture laboratories. Then [it took] about four weeks before you see the virus 00:13:00replication using the reverse transcriptase activity to judge, and that's just way too difficult. Antigen detection is very insensitive, so I was thinking probably PCR technology is the way to go because you can amplify HIV sequences millions of times.

MILLER: Fabulous.

OU: Right.

MILLER: Also as you're coming in and setting up this new laboratory, what are your recollections of the role of the lab at CDC at that time? Was it a reference lab or a research lab? Were you paving new ground in moving into this lab?

OU: There were a lot of activities in terms of research, sort of an NIH 00:14:00[National Institutes of Health] type of operation. I think that was the reason Dr. Schochetman wanted me to move away from that kind of activity to do something that would be useful for CDC. I think that was a tremendous decision. You know, it is very hard to get a laboratory person to move away from basic research at that point.

MILLER: Did you struggle with that?

OU: No. As I said, I have a medical biology background, so I understand why you need this kind of thing. So, no, there was no hesitation on my part at all. The timing was just right.

MILLER: Can you tell us a little bit more about establishing this PCR technology 00:15:00in the lab? It sounds like you had some management responsibilities. Who did you recruit to get this up and running, and what was involved in getting up to speed on being able to use PCR technology?

OU: In the beginning, I was all by myself because I was the only person. There was another person who came to CDC with me and became my old friend, Clyde Hart. Dr. Hart together with me was trying to put PCR to work. It's not that difficult, you know? I actually got it to work in a couple of weeks once I decided to get on to it. I had to get primers and things like that. I ordered 00:16:00commercially. At that point, there was not the core facility in CDC yet, so we ordered primers from outside. I put things together and in two weeks, I started to see the beautiful results. At this point we had to use culture specimens that derived from HIV patients. Dr. Paul [M.] Feorino, who was in charge of the culture laboratory, provided me all those materials.

This is a great point: that when I got this thing to work, it's not just myself. It's also my colleagues in CDC and a great decision from Jerry Schochetman, who worked with me and provided me everything. So I began with a single-person operation, but quickly when I got it to work, he would give me technical 00:17:00support. At that point the very first two persons working with me were Sheila Mitchell and Jennifer Moore. They were great helping me to get things running, because in the beginning, back in the old days we didn't have a PCR machine. [There was] no commercial PCR machine at all, and you had to do everything by hand. I had to have two heat blocks: one heat block to denature DNA, and the other one for DNA polymerase to start the chain reaction. Then also there wasn't the heat-stable enzyme, DNA polymerase, so every single step I had to open up the tube, add the enzyme, close it, spin it, and then heat it up, doing the DNA synthesis, and then open it up again and add a new enzyme and so forth. When I 00:18:00would run 35 cycles, this was a tedious approach, and I developed carpal tunnel syndrome very quickly. I got a sore hand every day.

MILLER: From doing what?

OU: From just opening tubes, adding things, closing tubes, you know, snap it, put it into the--put it into the centrifuge to spin-- measure everything down there, and start the process again. You had to do this 35 times. Consider if you have 20 samples how many times you have to do this.

MILLER: Did you learn this from--who had been experimenting with this technology?

OU: No, we just read from the publications. There was nobody to learn from. You just looked into the publications and started from there. You just needed to try 00:19:00it a few times. As I said, I had basic training on DNA synthesis before, when I was in graduate school. I knew how to do DNA labeling and things like that with P32 [phosphorus 32], because we had to do Southern blot and all those things. In the beginning, we used culture cells. They actually will produce tons of DNA HIV copies in there, so it is not that difficult to get them to work. But once you move into patient specimens, it's another story. There are tons of human DNA in the background.

MILLER: When did you begin to do that? Once you were moving past Dr. Feorino's culture specimens, what did you start with, in terms of trying to look at 00:20:00patient specimens?

OU: Yes, that was our goal. Our goal was not just doing PCR using culture cells. We wanted to demonstrate that we could get this thing to work in the peripheral blood drawn from patients. The beautiful thing also was that in CDC we had those specimens archived away. These specimens were [in the] serum bank, or sample bank. Paul got part of them to do the culture, but the original samples [were] saved there, so we could go back to those original samples and make comparisons.

MILLER: These samples would be from various epidemiologic studies? Where were these samples drawn from that were archived at CDC?

OU: In the beginning, yes, they were all different kinds of specimens drawn from our epi [epidemiology] studies, so they were well kept. The other studies, specific studies, they were specific cohorts that were valuable to me, so it was very convenient. Once the laboratory showed that this thing will work, then we worked together with our epi people and got things done quickly. This is the great thing about CDC. I didn't need to go out and look for them or to start a new project to collect specimens, because sometimes it takes a few years just to collect those sequential or prospective samples and for CDC to go back to our serum bank or specimen bank to get them.

MILLER: I'm aware that there were a number of articles that you published on 00:22:00this, first in Science, is that right, using this technology?

OU: Yes.

MILLER: Then it became apparent that it could be useful for the diagnosis of HIV in infants. Can you tell us a little, you mentioned it before, what was the problem in diagnosing HIV in infants? Now we're into, I guess 1989, 1990, and [there were] quite a few cases in infants and young children.

OU: Yes. This is a very interesting and important question. Infants born to HIV-seropositive mothers would carry the mother's antibodies. The half-life of 00:23:00antibody normally is somewhere around, I think, 23 days. So the maternal antibody has to decay before you can say, oh, this infant is infected now. If they are not infected, the maternal antibody would decrease and disappear. If they are infected, then they will produce their own antibodies. You can see the persistence of antibodies because it's the infant's antibody. In the beginning, it was the maternal antibody in the infants. So if you want to treat the infected patients early, you need to know if they are HIV infected or not early, and maternal antibody is not the way to go. Then detection of viral antigens or 00:24:00viral nucleic acids by PCR appears to be the way to go.

Once we demonstrated that PCR can detect HIV directly from adults, we moved right into infants. I remember this particular meeting that Jerry Schochetman and I [had] with [Dr.] Harold [W.] Jaffe and [Dr.] Martha Rogers. We sat in the small room in the back of Building 6 to talk about this. Basically the lab people said, hey, now we have this new technology, we can solve the problem for infants. Martha was in charge of a study where the major sources of HIV-infected infant specimens were collected from New York. This was a beautiful combination 00:25:00of a new laboratory technique to solve the problems waiting to be solved for quite a few years.

MILLER: What was used at the time to try and diagnose the actual viral particles in infants? Was it nucleic acid?

OU: There weren't. You had to wait. Even very late, even in the last few years in the 2000's in Africa, most of the infants got diagnosed by antibody test. As I said, you had to wait a few months, nine months or even longer--the WHO [World Health Organization] recommendation actually is 12 months to 18 months later [to 00:26:00distinguish the infant antibody from the maternal antibody]. You had to see the disappearance or the persistence of HIV antibody in an infected infant, respectively.

MILLER: So you couldn't use culture.

OU: Yes, you can use culture, but it just takes a long time. You need a culture laboratory, which is expensive, and it takes somewhere around four weeks or longer to do. This kind of operation is impossible in resource-poor countries in Africa.

MILLER: Did you use blood directly? There was a dried blood-spot methodology?

OU: No, at that point we didn't use dried blood spots. As I mentioned earlier, Martha Rogers, who was in charge of the New York studies, had the specimens from 00:27:00infants stored, so we went right into it. That's how we made our speed very fast.

MILLER: How long would it take to do that procedure, given the blood specimens?

OU: The testing at that point was somewhere around 2-3 days, compared to the 4 weeks with culture. Today this thing can take only a few hours with the newly improved PCR technology. You mentioned dried blood spots. That provides a means to quickly get the specimen from infants' fingers or toes and then shipped to the central laboratory for testing.

MILLER: Can you tell us how that works? Do you actually use a type of paper and touch the blood spot to the paper?

OU: Yes.

MILLER: Was that [methodology] used for other things, or is that for the--

OU: That's the so-called PKU [phenylketonuria] studies back in the early days for the U.S. Even today DBS [dried blood spot] is still used for a lot of genetic diseases. You collect the blood there [onto filter paper], and HIV genetic information is very stable once the spot is dried.

MILLER: Wow.

OU: Yes, and it can be stored actually for years. Back in the old days, in the early '90s or late '80s, we received specimens. We were testing dried blood spots from Africa, and some samples received by our laboratory, we did get them [that were not in good shape]. You'd have mold growing if you didn't dry your spot completely, and so things started to grow [that is, contaminants]. Even 00:29:00there we could retrieve nucleic acids for testing. For sure, these days, the SOP, the standard operational procedures, require people to dry the spots, put it under an oven or dry at room temperature and make sure they are dry before you ship. Dried blood spots even today are being used for viral load determination. So it's quite an interesting substrate to collect patient specimens for a lot of things.

MILLER: I guess the heat does not kill the HIV virus?

OU: Actually, it kills. On the other hand, the virus will not be viable anymore, but the nucleic acids are there, and the nucleic acids are the things we are 00:30:00looking for. Even if there is degradation, breaking down of DNA, it is fine because we are only detecting a small portion of it.

MILLER: That must have been really exciting.

OU: Yes.

MILLER: Did you publish these? I know you did, but do you remember some of the journals you published in?

OU: In terms of DBS?

MILLER: In terms of using PCR to diagnose infants.

OU: Yes. This was the paper written by Martha Rogers, myself and a bunch of other people from New York. It was published in the New England Journal of Medicine. It's a very prestigious journal, and it is one of the two very first 00:31:00publications on the use of PCR for infant diagnosis. It's still being done today.

MILLER: You reached fame and fortune early in your career at CDC. What was that like? Was it stressful?

OU: At that point, actually, it wasn't stressful until a little bit later stage for other investigations. Then I thought it was a lot of fun, because I felt that I was a little kid come into a candy bar store and I could reach out for anything that I want. We were experimenting with something that other people [had] never done before. Also I have to say, I forgot to mention to you earlier, that we got great help from the company that invented PCR. I happened to read 00:32:00the publications published in '85 and '86 when I was coming to CDC to work, so I knew they had it. But once we started our work, we realized that doing it the way that we did, manually, was impossible. It was not good for clinical diagnostic work, because contamination can become a big problem. You are amplifying a whole bunch of virus sequences, and those sequences later on will become a contaminant for the new patient specimens. So I had to keep the laboratory clean and mop my own laboratory. I cleaned the handles and everywhere that I could think of. I would come in on weekends to clean the laboratory myself, because I didn't want other people to mop up my own floor. So I was very 00:33:00careful. Then we reached out to Cetus [Corporation] for help to see whether they have any new technology available in addition to their publications, and they were very helpful. John [J.] Sninsky and Shirley Kwok helped me a whole bunch. In terms of the PCR paper, the real final test on patient specimens, I actually carried them to Cetus to double check everything.

MILLER: So this was for the Science paper?

OU: Yes, and then the rest was done here in CDC. But on the other hand, as you mentioned, this is new, and we laboratory people always try to do new things. 00:34:00The way that we were doing [it], we had to use P32 to do detection.

MILLER: Explain what the implications of that are.

OU: [It's] too tedious and it's radioactivity, and then we had to expose the film in order to see the HIV sequences that we amplified. It takes 2-3 more hours for film exposure, and to us that is even too long. We wanted to cut it down even further. So at that point I also reached out to another company, GenPro in San Diego. I went over there to talk to them to see what kind of new technology they might have. Actually I knew, so I went to talk to them, and they were very helpful, too.

MILLER: What was the technology?

OU: It is acrydinium-labeled oligonucleotides (chemiluminescence) instead of P32-labeled (radioisotope). So I moved from radioactive to a non-radioactive probe, and it was done in a way that the machine would do the detection for us. At that point the machine was this big, huge machine. But people are very nice, you know. It is important for us people working in CDC to reach out, too. We don't need to rely on our own technology, because in many places we don't have [it]. The companies out there actually have new technology, and they love to work with us people in CDC. We come together and discuss our needs and their 00:36:00technology. We sign collaboration agreements, and we put things together and it worked just beautifully. That's something I really learned: to best use our CDC name to help us.

MILLER: It sounds like some of these things might have cost a little bit of money. Did you have the resources to travel and purchase equipment? What was that like? We're talking about the early '90s, late '80s.

OU: I think at that point at least--even now I think, compared to quite a lot of operations that I know in CDC, the AIDS [acquired immunodeficiency syndrome] program had really plentiful resources and money. I was lucky. Our organization 00:37:00had enough money, and we could travel, as you said, and talk to people and put things together. I was really lucky.

MILLER: Was there competition in this arena with NIH and some of their research?

OU: For sure. Naturally in the beginning, as I mentioned early on, I always wanted to get PCR to work even before I came to CDC. In the beginning Jerry Schochetman thought in vitro hybridization might be the way to help us out, but then I was thinking maybe PCR is the way to go, because I just happened to see 00:38:00the new publication then. At that point I really worried that somebody, as you said, in NIH or in Paris--those French people who cultured the first HIV virus, they would have known this. It didn't take me, you know? Somebody just [needed to] have a fresh Ph.D. or post-doc work to come up with this idea. So in the beginning I was really worried. I had an empty laboratory that I began with. I said, whenever the Science or Nature journal came out weekly, I would open it up and see if anybody [had] published it. So the pressure then was that, that I really worried somebody else did. But evidently, and this may be [important] for 00:39:00the younger generation, younger people, younger scientists, laboratory scientists, to see this as well. If you believe something, go ahead and do it. If you worry about it, that doesn't help you. It happened that those big scientists in NIH or in France, they had big projects ongoing. They had to worry about their projects. They didn't happen to see the paper. Not that we are brighter, it's just that I happened to see the paper.

MILLER: You're faster.

OU: Yes. And we got help from our people in Cetus who invented the PCR machine. You know, [Dr.] Kary Mullis got a Nobel Prize for that in 1993 a few years later. Yes, I was very lucky. Timing was just great. But I take credit that I 00:40:00didn't waste time. The day that we submitted our paper to Science and got accepted, I was so happy.

MILLER: It sounds like Dr. Schochetman gave you a lot of support as well.

OU: Oh, that is for sure.

MILLER: And the Office of the Director?

OU: Yes, yes. This is one thing I really am very grateful [for]. I'm a foreigner. At that point I wasn't even a U.S. citizen. I was working in this organization, and this is a huge place. I was a small potato and I cannot speak English well then, not now even, and they gave me all the support there. When they saw that this thing was working, they gave me everything: people, help, hands, and I was very grateful for that.

MILLER: You ended up doing some work looking at the genetic diversity of HIV. Tell us what's involved in that, and why was that important?

OU: First, my Ph.D. dissertation actually was on the genetic organization of murine retroviruses in the mouse genome. I was doing the sequencing, that kind of thing. Moving from that background to HIV was not difficult for me. HIV is known to mutate because it uses reversing transcriptase. It wasn't perfect; it 00:42:00makes a lot of errors. When HIV infected one person, even if only one [infection] originated from one single virus, eventually it is going to create a whole bunch of siblings, quasi species. The degree of diversity within one person changes over time, gets greater, greater and greater each year. If one person transmits the virus to a second person, it is going to do the same process, but will be similar to mine if I transmit a virus to that person and so forth. This became an interesting character[istic] for us to look at, to decide whether one person transmitted a virus to a second person.

I need to bring up Dr. [C. Robert] Horsburgh here. Bob Horsburgh [was] working with me very closely. One day I tried to test this idea. He gave me 20-some specimens from his Boston cohort collection. He told me, okay, I have one source patient, and here I gave you a whole bunch of 20-some people. One of them will be his partner. We know that he transmitted the virus to him [the partner]. I said, this is great. Once I got the result done, I called Bob and said, Bob, you told me that this one person infected another person among these 20-some people. But I said, this is not what I see. I see two persons. I worried why I found two 00:44:00persons. Bob said, Chin-Yih, you are right. Those two specimens were from the same person at two different time points. So I was very happy to see that.

Then this was the time that the Florida Dental study came about, so in the laboratory we were testing the idea. When [Dr. James W.] Jim Curran, the director then, mentioned to us that there was such a case, I thought, hmm, easy enough, we know how to solve this problem. I thought naively that we should be able to solve it in two weeks. We thought, if there's no connection, then I can prove there's no connection, and it's done.

MILLER: When we say that the virus [is] from the index patient, the person we 00:45:00know and a contact, the similarity wouldn't be 100%. There would be a percentage of similarity.

OU: Right, you cannot be 100%. Even in a single person, there will be different viruses, their siblings all together, quasi species, so [if it] moves to another person, it cannot be 100% [similar]. If it's 100% identical, that will mean you got a contamination, or these two viruses are from the same culture or something like that.

MILLER: Just to recap the so-called Florida Dental investigation: again, this was 1990, and the Florida Health Department reported through their AIDS surveillance about a woman of about 21 who had AIDS. She did not have any 00:46:00standard risk factors upon their studying her, but they identified the fact that she had had a dental procedure two years before her AIDS diagnosis. It was found that the dentist had AIDS. Then additional HIV-infected patients of the dentist came forward. They sent you the index patient and the dentist's specimens to see whether they were the same strain. If you hadn't been studying genetic diversity, what would have been the state of the art in terms of trying to link those specimens?

OU: Hmm. I couldn't think of one in terms of laboratory methods.

MILLER: Were others using this methodology, or was this an early use of the methodology?

OU: I think this is the earliest usage of it in terms of HIV, certainly the very first case I know of, yes.

MILLER: Can you tell us a little more about what your approach was, and how it turned out for you when you were looking at these [specimens]?

OU: Like I said, in the beginning I had [the] assumption that there might not be a direct linkage, but we had to prove it to show that the two sequences from these two persons were not the same. They diverged.

MILLER: Why was that your hypothesis?

OU: Actually I really don't know. My impression was that she might have got it 00:48:00from somebody else, but actually I didn't really know.

MILLER: But it had been shown that this [dental procedures] was a method of transmitting the virus.

OU: Right, right. But [the] genetic approach, DNA sequencing is the way to go. That's the way I look at it. Certainly, I was completely surprised. I shouldn't have said that, because I really didn't know what to expect, but I was surprised to see that they were so close. They were not identical, but they were close, very similar.

MILLER: How long did it take you to do this? There must have been a tremendous amount of pressure on you at that point.

OU: At that point I didn't know exactly what to expect, so I didn't really have--in the beginning I have to admit I didn't have that kind of pressure that I felt. Because we had done some of the groundwork already, it was very natural. 00:49:00We said, okay, we've got this case, interesting. We had to do it the way that [it] can be best done in the world. I said, contamination can be a problem, and we have to do things right. We received the specimens from Carol Ciesielski, who did the groundwork. I was trying to be very careful to make sure that if something happened, there [would not] be a legal problem. so right from the beginning I would segregate the work. We had two Ph.D. staff in our section, [so one person] was taking care of one specimen, and the other one took care of the 00:50:00other one. They were on two different floors. So this lady's sample, you worked it up there for the ten days, and [the other staff handling the other specimen] you did it down here. You two do not walk into each other's room at all, and that kind of thing, to segregate the work.

Later on when we got more samples from other patients as well as the local controls--because we had to bring in local controls--the specimens didn't even come into our laboratory directly. Charles [A.] Schable, who was in charge of the serology laboratory, would take care of the requisition work for me, so that people would not say, oh, Chin-Yih contaminated the sample. It would be somebody else handling the specimen. I would ask Charles, today we need the specimen, and he would request it for me. This custodial type of chain was very clearcut right 00:51:00from the beginning. That helped CDC's reputation later on. Certainly at the end Harold [Harold Jaffe] also decided that some of the specimens would be sent over to Scotland to be retested.

MILLER: Scotland--was that another lab that was proficient in doing this?

OU: Yes. That's what I thought, but it turned out to be the right choice. Actually, I don't know them much at all, but people recommended that this would be the right--

MILLER: In terms of doing DNA sequencing, or did they actually have experience and expertise in HIV?

OU: They had, but later on this story was written by the Government Accounting 00:52:00Office person, Mark Ron. Reading the article, the book, I realized that actually the Scottish laboratory failed to amplify DNA in their first trial. Then they had to modify their methods, and later on they got the result. At that point, [it's] interesting to talk about this, to me, I was under heavy pressure. I did my best to make sure the laboratory operations were correct. Things were segregated, we had different people doing several things. We checked the HLA [human leukocyte antigen] gene of the patient specimens to make sure they belonged to two different HLA types, meaning there were no missing specimens and that kind of thing. [For] some of the samples we had multiple samples. We 00:53:00double-checked, and I knew our operation was clean, nice and perfect. But what happens if this laboratory, which is supposed to verify our results, did something wrong? You see what I'm saying? When the result came back and I received the first result, I sat there for five minutes. I didn't know what to do. I wanted to make sure the results were the same as ours, and I checked one nucleotide after another. I was so happy when I saw the results were identical. I called up Harold and Jerry and told them they (the Scottish lab) had duplicated our results. They verified our results.

MILLER: This had, of course, huge implications from a policy point of view.

OU: Sure.

MILLER: That it would be possible to acquire HIV from a dentist, even though 00:54:00from an infection control assessment they were never able to really confirm a method of transmission--

OU: True.

MILLER: --a variety of procedures that could have punctured and so on.

OU: Yes. Even these days [when] I go to see my dentist, I open my mouth and every time I would think about this thing. Regardless, the dental operations here in this country became much better, cleaner and things like that. I also had seen my dentist in other places, and I always wondered, has this been done correctly in terms of your operation? You keep your place clean or what?

MILLER: You were really at the center of so many intense outbreaks and 00:55:00methodologies for diagnosis. Did you have a sense of being part of a team? We see lab, and we see epidemiology, the programmatic people or the ones doing operational research, and then the lab is somewhat separate at CDC. But it sounds like you were really in the mix. Can you comment on that?

OU: I think this is really a great thing about CDC. Certainly I was lucky. I moved into this lab/epi collaboration and produced wonderful results. Many of them influenced the initial policy, and so I was very lucky. I'm not quite sure 00:56:00whether other places in CDC are like that. Some of them may take a longer time. Because of PCR, that brought me to work with the epi people right away. [It was] not just infant diagnosis, Martha Rogers' work, Bob Horsburgh's work, [it was] also the work with hospital infections, like [Dr.] Janine Jason and [Dr.] Bruce [L.] Evatt, to look at the HIV in people, hemophiliacs as well as some other studies. Later on [it was] with [Dr.] Bruce [G.] Weniger in Thailand, to look at the HIV diversity in a bigger situation in this country, as well as in many 00:57:00other countries.

MILLER: Do you want to tell us a little more about the work with Bruce Weniger, who was director of the CDC Thai field office during the early '90s in Bangkok.

OU: Yes. That was really a beautiful study. I remember I went to Bangkok for the first time probably in 1990, I cannot recall exactly when, and I saw Bruce Weniger there the very first time. The thing I recall is that he said, hey, I have these specimens collected from different parts of Thailand. They were from 00:58:00persons infected with HIV through IV [intravenous] drug use or through social contacts. Can you take a look at them? That study was fantastic. I was also pretty surprised when the first set of results came out. As I said, they were done also by Claudia Bandea, one of the Ph.D. staff, and the other side, Chi-Cheng Lo. They were both involved in our Florida Dental studies. I think I forgot to mention their names, sorry for that. Also at that point I got new people, Marcia [L.] Kalish, who was very helpful to me. At that point we were also working on sequencing other HIV from different parts of the world, like 00:59:00Haiti and some African countries, and there was an NIH project that we helped work on.

MILLER: Can I just ask, what was the importance of that?

OU: That actually turned out to be interesting as well, with that particular study. We submitted our HIV sequences from different countries to the Los Alamos National HIV database and [Dr. Gerald] Gerry Myers, who helped us to determine the HIV Florida Dental cases. He was the director over there [Los Alamos HIV database]. He told me that at one point the data from CDC, from us, contributed 01:00:00to his database--quite a big chunk. From there [it] also included other people's sequences from other studies in many countries, and he was able to find that there are different HIV clades or subtypes, HIV-1 subtypes, around the world. This was a new concept: seeing different HIV-1 subtype clades in different countries. In some countries they may have two or three. The United States [in the] beginning was infected with subtype B. In Thailand while we were looking at them, we found that actually in Thailand there are two viruses in two different populations. They're basically segregated, and epidemiologically this is completely a new concept. When we found something we said--what? What is this? 01:01:00When I first saw the data coming from those two groups from my shop, I said, we must be wrong, let's go back and do everything over again.

MILLER: These two groups in Thailand--one was the IV drug-using population.

OU: Yes.

MILLER: The rise of HIV in Bangkok, as I recall, was dramatic. In the early '80s there wasn't much, and then in the late '80s it started to just shoot up in both the IV drug-using population and the heterosexual and homosexual cases. You had specimens both from the IV drug-using and then from--was it heterosexual or STD [sexually transmitted disease] clinics?

OU: Mm-hmm, yes. It was nice, it was just interesting to see them with two 01:02:00different viruses. One now is called subtype E and the other one subtype B, but at that point there wasn't an E subtype yet. We called them A and B.

MILLER: So that helped trace the origin? In other words, one might have thought that it was the same virus communicated through female sex workers or other ways the populations interact. What were the implications of this? We don't hear too much about it now. I guess there's emphasis on treatment, but what was your sense of the importance of it?

OU: At that point probably we could imagine that one virus is more transmissible than the other. It might replicate faster or something like that. I don't know 01:03:00whether that turned out to be true, though, but that was a logical first reaction to this kind of distribution. You might think, oh, maybe for vaccine preparation you need to designate this vaccine for this particular population with the other population, but it wasn't that way.

MILLER: Looking over CDC's laboratory strategy and emphasis, do you want to 01:04:00share some of your thoughts on what CDC focused on for the lab? Did you think we got it right, or could we have done better? It certainly worked out well for your particular area.

OU: True. There were quite a lot of resources in terms of financial support as well as technical people support for our HIV laboratory operation. In CDC, my career in CDC, I did notice that in other areas they might not be like us. I was fortunate to be in such a special unit that would allow us to do more, and hopefully it will be the same for other areas as well. I would think, as you 01:05:00alluded to, that the close collaboration between the lab people and the epi people is essential. I was in a particular situation that I moved into quickly, because [of] the leadership both from our laboratory part and also from the epi part, Harold Jaffe as well as Jim Curran, they were so supportive.

One thing I want to add is, I'm from an oriental culture, I see things a little bit differently from people like you. I was surprised actually--I told my wife during the Florida dental investigation, there were times I went home and I saw 01:06:00the national news. I was so stressed then. I looked at it and said, what happens if I missed something, and then I screw up the reputation of the entire organization here? Yet I told my wife that I never heard a doubt from Jerry Schochetman, from Harold Jaffe or from Jim Curran. They never asked me, hey, Chin-Yih, are you right? They never said anything like that. In this case I was completely grateful. In the oriental culture my boss would say, hey, did you screw up? They would have said that. In this case I was very appreciative that I was lucky enough to be working here in this organization. They gave me complete 01:07:00trust and gave me everything I needed to get the work done, and I can never forget them my whole life.

MILLER: You worked on AIDS a big chunk of your career, later on working for PEPFAR [President's Emergency Plan for AIDS Relief], the President's Emergency Plan. Did you ever worry about getting AIDS?

OU: Never. I never thought that I would get AIDS. Basically I believed whatever work that we, CDC, put out there. We told people you don't get AIDS by sitting on the toilet, you don't get AIDS through just kissing or that kind of thing. No, I was never afraid that I would get HIV at all.

MILLER: Was there concern early on among laboratory staff?

OU: Certainly, yes, if you are working in the HIV culture laboratories where there may have been a high concentration of viruses. Yes, there was concern, and you know that there were incidents like that before, especially in monkey virus. There was a concern, but personally I never had a concern. I'm happy working all my life. I have had 28 years at CDC, with the exception in the middle of six years. More than 20 years [was spent] in HIV. I never had any doubt at all.

MILLER: Thank you so much.

OU: I appreciate it.

15