Luck Favors the Prepared

The success of a young scientist illustrates the valuable role mentors can play in establishing a research career

By Andreas von Bubnoff

The child of a professor of German literature at Brigham Young University (BYU) in Provo, Utah, Brandon Keele knew he wanted a career similar to his father’s. “I always wanted to be a professor, which is funny,” says the 38 year old, referring to the fact that his current position as head of the viral evolution and genomics core at the National Cancer Institute (NCI) in Frederick, Maryland, does not involve teaching. Instead, it’s all about research. “It turns out that I am a fairly good researcher,” he says, “so I thought I would stick with what has proven to be successful.”

Last year, Keele moved to the NCI after a successful six-year postdoctoral stint in the laboratories of Beatrice Hahn and her husband George M. Shaw, both professors at the University of Alabama in Birmingham. His time there was a success by anyone’s estimation. He was first author of several seminal papers, including a 2008 paper that showed for the first time that in most cases, productive HIV infection is the result of transmission of a single founder virus (Proc. Natl. Acad. Sci. 105, 7552, 2008). Other papers he was first author of include a 2006 paper suggesting that the origin of HIV lies in an area in southeastern Cameroon (Science 313, 523, 2006), and a paper published just last year that showed that chimpanzees in Gombe National Park in Tanzania are more likely to die if they are infected with simian immunodeficiency virus (SIV), challenging the view that natural SIV infections don’t cause disease (Nature 460, 515, 2009).

While hard work, passion, good hands, and a portion of luck all played a role in Keele’s success, his career shows how important it can be for a young researcher to choose the right mentors and lab for their postdoctoral work. “I was in the right environment with very supportive mentors,” Keele says.

The makings of a researcher

Keele first got excited about immunology when taking classes as an undergraduate at BYU. “I was interested in what causes human suffering,” he says. “I kind of have this soft spot for human problems and I think I really wanted to help the world. I was interested in human diseases, what caused disease, why vaccines worked or didn’t work.”

In 1997, still an undergraduate at BYU, Keele got his first molecular biology experience as an intern in the lab of Greg Burton, at the time an associate professor in the department of microbiology. “[Brandon] just came to me and said I would like to do something in the lab and work on a project with you,” remembers Burton.

Keele says he always wanted to eventually get a PhD degree, but after graduating from BYU, he initially wanted to first obtain a master’s degree and wait until his wife, whom he had met at BYU, was ready to move on to another university with him. But Burton convinced him to do a PhD in his lab. “I tried to twist his arm to consider a doctoral program because he [was] an exceptionally gifted student,” says Burton, in whose lab Keele started his PhD research in 1999. “Once I had everything running [in Burton’s lab], it seemed like a waste of time to start over at another school,” Keele says. Keele was studying how HIV accumulates on the surface of so-called follicular dendritic cells (FDCs), which can be found in secondary lymphoid tissue such as lymph nodes. A few years earlier, Burton had shown in mice that HIV stays infectious at the surface of FDCs, suggesting that FDCs can act as a reservoir, keeping infectious HIV particles trapped on their surface (Nature 377, 740, 1995).

Keele’s work in mice then showed that HIV on FDCs remains infectious for at least nine months and that FDCs keep HIV particles infectious by holding them in complexes with antibodies on their surface (J. Immunol. 166, 690, 2001; J. Immunol. 168, 2408, 2002). He also sequenced HIV particles isolated from different cells and tissues of infected patients, including FDCs, and concluded that the HIV particles from FDCs were older because their sequences changed less over time than virus in other tissues of the patient. This was further evidence that FDCs keep a reservoir of non-replicating, but infectious HIV particles.

From immunology to viral evolution

It was several years before this HIV sequencing study would eventually get published (J. Virol. 82, 5548, 2008), but working with viral sequences immediately got Keele interested in viral evolution, which he decided he wanted to study in his postdoctoral work. “Sequencing and analyzing the changes that occurred in viruses revealed to me a powerful tool in understanding the virus, how it evolves, and potentially how we can inhibit it,” he says.

After receiving his PhD in 2003, Keele began to look for postdoc positions with researchers who studied viral evolution. This led him to Hahn, who was studying the origin of HIV. At the time, Hahn’s group had published high-profile studies that suggested that HIV originated from SIV found in chimpanzees in Africa (Nature 397, 436, 1999;Science 295, 465, 2002), although the exact location of the precursor of HIV-1 remained to be found. Keele says he was a little worried that some people might find it odd that he did both his undergraduate and graduate work at the same university, and when he gave a talk at Hahn’s lab he felt it went “really poorly,” he remembers. “They were just very critical and really wanted to know all the ins and outs of the data. I initially thought they wouldn’t hire me.” But Hahn felt differently. “I talked to him and I immediately liked him,” Hahn recalls, adding that she didn’t even look much at Keele’s publication record.

As his first project, Hahn asked Keele to work with Mario Santiago, a graduate student in her lab, to test the hypothesis put forward in a 1999 book called “The River” by journalist Edward Hooper that the HIV pandemic originated from SIV-infected cells from chimpanzees used to prepare or grow oral polio vaccine in Kisangani in the Democratic Republic of Congo. Santiago taught Keele how to isolate and analyze SIV RNA from fecal samples from chimpanzees. “We had just gotten some fecal samples from Kisangani from chimps that were sampled directly around where [Hooper] said to look,” Keele says. “The idea was to ask what kind of virus they have.” In a paper that appeared in Nature, with Keele as the third author (Michael Worobey of the University of Arizona in Tucson was the first author), they reported that the SIV in these animals was too different from HIV for Hooper’s hypothesis to be true, Keele says (Nature 428, 820, 2004).

This meant that the true SIV precursor for the pandemic HIV-1 group M, which makes up the majority of HIV infections, still remained to be found. “With the help of many great collaborators who spent countless hours in the forest looking for piles of poop,” Keele says, he and Santiago kept searching for the origins of HIV in other areas. It seemed an almost impossible task. “We originally thought that maybe those viruses that are most closely related to HIV-1 are not there anymore,” Santiago says. “Chimpanzees are hunted like crazy in that area of the world.”

But to their surprise Keele found an SIV variant in chimpanzee fecal samples from southeast Cameroon that was more similar to HIV-1 group M than any other known SIV variants, suggesting that SIV from there gave rise to much of the HIV pandemic. “We thought it was really lucky to find something that clustered very closely with HIV-1,” says Santiago.

The resulting study, published in 2006 in Science with Keele as first author, gave Keele his first big break, attracting calls from many journalists (Science 313, 523, 2006). “This was really big,” Keele remembers. “It was new for me to have so much attention. We had the BBC talking to us, we had lots of interviews, we had lots of stuff going on, lots of newsprint, lots of fun.” Lots of fun, that is, except for one thing. When it was time to toast the success with champagne, Keele wouldn’t drink because he is a Mormon, Hahn recalls. “He will not touch alcohol, not even to toast when you have the best papers,” she says.

With hard work comes luck

Discoveries like the one in Cameroon were the result of a combination of hard work and luck, Keele says. “It has to be some luck because it’s not every day that you get a Nature or a Science paper, but it’s also a lot of hard work,” he says. “Luck favors the prepared.”

Luck played a role because Keele was the one who analyzed the fecal sample that contained the relevant SIV strain, Hahn remembers. The sample came from a place called Lobeke in Cameroon. Because Keele analyzed it, the sample came to be known as “LBK,” for Lobeke and for “Lucky Brandon Keele,” says Santiago. “Luck did play a role,” Keele says, but adds that since this was his project, “I analyzed all the samples, so I would have come to it sooner or later.”

Isolating and sequencing RNA from fecal samples is no easy task, Keele says. One might even call it a shitty job. “It’s a lot of hard work because it’s isolating RNA from shit,” he says. “The fecal samples stink to high heaven and RNA doesn’t do well in fecal samples.” One important thing that made the 2006 paper possible was Keele’s knack for improving experimental approaches or adopting them for new applications. “He started thinking out of the box to try to improve the methods that we started out with,” remembers Santiago. Keele, Hahn says, figured out a quick way to test if fecal samples contain SIV-specific antibodies. “I came up with a convenient way to get antibodies out in a fairly high-throughput manner,” Keele says. This saved time and effort because it made it possible to focus only on sequencing SIV from fecal samples that were positive for antibodies.

In addition, Santiago says, Keele used genetic tools in a way that made it possible to identify individual chimpanzees in wild living populations that were not habituated, or used to being around humans. This way, researchers could determine how many different individual chimpanzees the fecal samples came from and how prevalent SIV infection really was among the chimpanzees in a given area. It was the first time anyone combined the genetic tools with antibody and viral RNA detection to accurately determine the prevalence of SIV infection in different ape populations, Keele says.

The lab next door

After his success in studying the origins of HIV, Keele began looking to do something different for a second postdoctoral position. It turns out that he didn’t have to go far. Shaw’s lab, where Keele ended up, is next door to Hahn’s. “Our labs are almost joined at the hip,” says Shaw.

Working with Shaw, Keele again adapted an existing technology for use in a new context. He was the first to use single genome amplification (SGA) to characterize HIV samples from acutely HIV-infected individuals. In this approach, HIV samples are diluted so much that they likely contain just one copy of the HIV genome, which can then be amplified and sequenced. This makes it possible to determine the proportion of different HIV variants in an infected person.

The SGA project also showed Keele’s willingness to look for expertise outside the lab when necessary, Shaw says. “Brandon went to John Coffin’s laboratory at NCI Frederick and brought the single genome amplification technology back to Birmingham, modified it, and then rapidly exploited that technology,” Shaw says, adding that this was just one of several key things Keele was involved in “that allowed us to realize the potential of SGA and sequencing,” in acutely infected people.

The resulting study analyzed env sequences of people infected with HIV for just a few weeks to show that the majority of clinically productive HIV infections are the result of a single transmitted founder virus (Proc. Natl. Acad. Sci. 105, 7552, 2008). Other studies had already suggested that there might be a genetic bottleneck that limits the number of transmitted founder viruses that eventually cause productive infection to just one HIV variant out of the many in the donor, but the 2008 study for the first time revealed the exact sequence of the actual founder virus and quantified this genetic bottleneck, showing that in the majority of transmissions there is just one transmitted founder virus. “What we really tried to do was to define in a quantitative way how many viruses initiated infection,” Keele says. “For the first time ever we were able to say look, this patient was infected with this exact nucleotide for nucleotide virus and we can see the evolution away from this virus and we can time all of this accordingly.”

Cynthia Derdeyn, the first author of a study in 2004 that suggested that a genetic bottleneck exists in HIV transmission, says that prior to Keele’s 2008 study, “I don’t think the field appreciated that one or a few differences in the viruses present in very early infection could represent adaptation to immune selection and thus were not the actual founder virus.”

“I think most regard [the 2008 PNAS paper as] a seminal paper in the field,” says Bart Haynes, a co-author of the paper and director of the Center for HIV/AIDS Vaccine Immunology (CHAVI), which provided many of the clinical blood plasma samples and some of the funding for the research. Later, Keele also used SGA to show that transmission of just one transmitted founder virus in most cases can be recapitulated in rhesus macaques that are rectally infected with low doses of SIV (J. Exp. Med. 206, 1117, 2009).

Even while Keele was working with Shaw he kept analyzing fecal samples for Hahn as part of a longitudinal study that followed wild, but habituated, chimpanzees in Gombe National Park in Tanzania. Keele determined which chimps became newly infected with SIV using the genetic tools he had developed to identify the individual chimpanzees over the years. The fact that they were habituated also helped in their identification. “There were a lot of other people involved but [Keele] was the one who carried the project almost from the minute he walked into the lab,” Hahn says. “He kept track of these chimps, he analyzed the chimp samples as they came in, he knew who became newly infected, he analyzed the viruses, etc.” The study, which followed the chimpanzees for over nine years, appeared last year with Keele as co-first author (Nature 460, 515, 2009). It showed that the SIV-infected chimpanzees were 10-16 times more likely to die than uninfected ones, challenging the view that all natural SIV infections are non-pathogenic. Derdeyn, who was not connected to the study, calls it “a paradigm shift in our thinking of ‘nonpathogenic’ natural SIV infections of nonhuman primate species.”

Many reasons for success

Keele’s time in Birmingham “would go down in anyone’s book as successful,” he says, attributing this success in large part to having Hahn and Shaw as mentors. “They facilitate young people being successful,” Keele says. Santiago, who has worked for both Hahn and Shaw and is now an assistant professor of medicine at the University of Colorado in Denver, agrees. “They are just fantastic mentors,” he says.

“In addition to my excellent mentors,” Keele adds, “I was fortunate enough to be associated with great collaborators. My research has benefited significantly from the generosity and kindness of many in the AIDS research community.”

Keele says he aimed quite high when he applied for his first postdoctoral position, but believes that doing so is possible because hiring a postdoctoral researcher carries little risk for a principal investigator. “I think most people ought to shoot really high for their postdoc,” he says. “A postdoc is a very easy hire for most people because you are only under a one year contract and the price is fairly low.”

The size of the lab also plays a role, he adds. “If I were to speak to graduate students throughout America, I would say to find somebody who does excellent science but who doesn’t have an enormously large lab, because the enormously large labs [are] how a postdoc or a graduate student can get lost,” says Keele.

It’s not just that Keele found great mentors in Hahn and Shaw; they found a great postdoc in Keele, who was not only passionate about science, Hahn says, but also unusually productive and reliable. “Whenever you would see him in the hallway or on the weekends or whenever, he’d say, ‘Oh, have you seen this?’ and he would show me a new piece of data,” Hahn says. “He was a quick learner and a great experimentalist. He is passionate about science, that’s why he is successful.”

Keele was also a good collaborator, Shaw adds. “One of his greatest strengths would be his ability to work very effectively with many different people.” Hahn agrees, and says this has contributed to his success. “He gets the most out of people and gets them to do what he wants,” she says. “I think everything goes better and the science moves faster when everyone is sharing data and is working together,” Keele says.

Last year, Keele moved to the NCI in Frederick to start his first independent position, which was hard to do after so much success, he says. “If you can’t prove your independence in a certain amount of time, you are written off as a lifetime postdoc,” he adds.

In Frederick, Keele is not working with any students because NCI is not affiliated with a university, but he has two technical employees and works with an animal facility on the main NCI campus in Bethesda. One advantage of being at NCI, he says, is that he doesn’t have to apply for grants. “It’s great being here because we can just focus on the science. We spend much less time worrying about other things.” This way, he also doesn’t compete with his former advisors for grant money and instead actually collaborates. “I like working here because I am able to collaborate with lots of people,” Keele says, adding that he is now collaborating with over 20 labs.

For his own studies, Keele decided to focus on studying virus transmission in more detail in rhesus macaques to better understand it and also to recapitulate HIV transmission in an animal model. With Chris Miller, a professor in the School of Veterinary Medicine at the University of California at Davis, he has been using SGA to study transmission routes other then rectal transmission in rhesus macaques. This has shown that intravaginal infection of different animals with the same dose of SIV leads to a more variable number of transmitted founder viruses than rectal transmission (J. Virol. 84, 7083, 2010). Keele and Miller are also using SGA to determine the number of transmitted founder viruses in male rhesus macaques whose penises were exposed to SIVmac251.

Keele also wants to understand where along the transmission route the genetic botteneck is located. In collaboration with Jake Estes and Jeffrey Lifson at NCI Frederick, Keele is using a laser to isolate SIV-infected cells from histological sections of the cervix of animals vaginally infected with SIV. They then sequence the integrated proviral SIV from these cells. The goals are to find if there is a specific location where the transmitted founder viruses first take hold in an infected animal, and to see if the location of the transmitted viruses in the infected animals plays any role in how well they replicate and cause systemic infection.

Keele is as busy as ever. “I still work very hard, and I have a hundred things to do as we speak.” But he doesn’t talk much science at home. “I like to shut off and kick the soccer ball around with my kids,” says Keele, who is married with three kids and likes hiking, bike riding, skiing, and photography.

And compared with his postdoc days, he doesn’t work every weekend anymore. But he still keeps a stack of Progresso soup cans in his desk drawer so he can grab a quick lunch. And he still thinks it’s possible that maybe one day he will follow in his father’s footsteps and become a professor. “Maybe one day when I’m older I’ll put on a tweed jacket and stand up in front of budding scientists and tell them how it is.”