The Carter Lab at NIH: A Model of Inclusive Excellence in Biomedical Research

Abstract

In the 1980s and early 1990s, Dr. Barrie Carter served as the chief of the Laboratory of Molecular and Cellular Biology in the National Institute of Diabetes and Digestive and Kidney Diseases at the National Institutes of Health. During that time, his group performed seminal work in adeno-associated virus (AAV) type 2 (AAV2) biology, including creating one of the first infectious clones of AAV2 and some of the first packaged AAV2 vectors. This work contributed substantially to the development of AAVs as gene therapy vectors. Part of the success of the group was due to Dr. Carter's ability to attract and manage a diverse team of talented individuals who synergized into a collaborative group that was more than the sum of its parts. This review describes some of the promising practices employed by the Carter group, which allowed such a diverse group to function so well. These practices included promoting a culture of co-mentoring, open communication, and respectful questioning.

Introduction

The leadership of the National Institutes of Health (NIH) frequently asserts that diverse teams of talented individuals are better able to tackle complex biomedical problems than homogeneous groups.¹ This review presents a case study of how this works in the real world. I was a postdoctoral fellow in the Molecular Biology Section, Dr. Barrie Carter's immediate research group within the Laboratory of Molecular and Cellular Biology, in the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), at the NIH, from 1988 until his departure from NIH, in 1992, to become vice president of a biotechnology company. Barrie also served as the chief of the Laboratory of Molecular and Cellular Biology (department chair-equivalent).

During that time, he had a diverse group of trainees and staff, including multiple African American and Hispanic trainees at postdoctoral and predoctoral levels (Fig. 1). The laboratory took advantage of NIDDK's partnership with the NIH Minority Access to Research Careers (MARC) program to recruit black and Hispanic summer students, who were able to work side by side with role models. The group also performed groundbreaking, and often dogma-challenging, work in adeno-associated virus (AAV) type 2 (AAV2) biology, including the creation of plasmid pAV2 (Fig. 2), one of the first infectious AAV2 clones,² and some of the first packaged AAV2 vectors.^3,4 My assertion is that the diversity of the group and the able management of this diversity by Dr. Carter set the stage for the scientific discoveries.

Figure 1.

Photograph of members of the Laboratory of Molecular and Cellular Biology (circa 1992). Members of Barrie Carter's immediate research group shown are Rikki Solow (first row, second from left), Linda Hamilton (office manager, first row, center), Roland Owens (second row, third from left), Sandra Afione (third row, far left), Sirkka Kyöstiö-Moore (third row, third from the left), Marcus Wallace (fourth row, far left) and Matthew Weitzman (fourth row, second from the left). Drs. Carter and Flotte are not shown.

Figure 2.

Schematic of the construction of plasmid pHIVrep (after Antoni et al.⁷). The AvaI fragment of the AAV2 infectious clone pAV2, which contains most of Rep protein coding sequence was blunt-end ligated into the HindIII-BamHI fragment of pBennCat,⁸ which contains the HIV-LTR, to form pBenn HIVrep. The SalI fragment of the AAV vector, pJDT95,³ which contains part of the AAV2 rep gene and the downstream ITR, was then ligated to SalI-digested pBenn HIVrep, to create pHIVrep. AAV, adeno-associated virus; ITR, inverted terminal repeat.

The diversity of the group was a reflection of Barrie's multidimensional approach to research. In addition to placing a high value on basic research, Barrie was always looking for practical applications. As a result, the group had basic researchers and clinically oriented researchers. Following Barrie's example, the trainees and staff were also on the lookout for practical applications of our work.

During my time with the Carter group, there were postdoctoral fellows from the United States, Argentina, Finland, and the United Kingdom. It is a tribute to the training environment that four of the scientists from that group continue to be active contributors to gene therapy research (Terence Flotte, Sandra Afione, Sirkka Kyöstiö-Moore, and Matthew Weitzman). Additional scientists went on to successful biomedical research careers in the United States and Israel.

More traditional research groups, with persons from similar backgrounds, have the advantage that group members frequently do not have to explain why things are done in a particular way. However, this lack of challenge to the status quo can stifle innovation. Barrie established an environment of respectful questioning by bringing in people from different research and personal backgrounds, and by encouraging co-mentoring and collaboration within the group.

Because group members came from different backgrounds, there was the expectation that questions would be asked about routine procedures. Having a group of talented researchers from diverse backgrounds, who respectfully challenged each other to explain things, made for a very innovative group. I will discuss some of the promising practices exemplified by the Carter group and describe how these practices benefitted the research.

Promising Practices and their Consequences

One excellent practice of the Carter group was to get to know people before bringing them into the research group. When I applied for a postdoctoral fellowship in his group, he invited me to visit the laboratory. We spoke one-on-one for about 3 h straight. He got to know my background, my research experience, my motivations, and my philosophy of research. I also got to speak with postdoctoral fellows in the laboratory, in Barrie's absence, to discuss potential projects and the laboratory environment. I could tell from these discussions that the postdoctoral fellows considered themselves to be co-mentors, with Barrie.

Selecting postdoctoral fellows who valued mentoring, as well as diversity and inclusion, promoted an environment of collaboration within the group. The greatest practical advantage of this environment is that every trainee got off to a flying start by being co-mentored by Barrie and at least one senior postdoctoral fellow. I personally benefitted from co-mentoring by James Trempe and Nor Chejanovsky, who were coauthors on my first article from the Carter group.⁵ I then served as a co-mentor, and later the primary mentor, for Matthew Weitzman and Sirkka Kyöstiö-Moore.

The second practical benefit of a mentoring-rich environment is that projects rarely went unfinished. Knowledge and techniques were shared freely within the group, so there was always someone who could complete work left behind by a trainee who left for another position. In other cases, postdoctoral fellows were allowed to take their projects with them when they established their own research groups. Barrie once told me that creating one's own competition should be viewed as a sign of success for a principal investigator, rather than being something to be feared.

The work of Hong and Page⁶ suggests that diverse groups work best when their diversity does not inhibit communication. One of Barrie Carter's great talents as a leader is establishing and actively keeping open lines of communication. He did not simply have an open-door policy, several times per week he would walk through the laboratory and chat with each one of us. This not only provided Barrie with real-time progress reports, it gave others in the laboratory a chance to hear what each one of us was doing.

Because of the pre-established spirit of cooperation, there was no need for us to keep secrets from each other. Quite the contrary, if one of us would mention to Barrie that we needed something, it was a frequent occurrence that another person in the laboratory would overhear the conversation and volunteer to help. In other cases, one of us would hear about a novel reagent or process, which could be applied to our own project. Weekly group meetings were also very positive events. Because we had a vested interest in each other's success, questioning at our formal group meetings was generally focused on problem solving, rather than one-upmanship.

One example of how these philosophies and practices positively influenced the research was how the group handled its discovery that there could be robust AAV2 rep gene expression from the long terminal repeat promoter of HIV-1 (HIV-LTR), in the absence of the HIV-1 Tat protein, in human 293 cells.⁷ Barrie applied for and received funds from the NIH Intramural AIDS Targeted Antiviral Program (IATAP), to explore the possibility that AAV2 Rep proteins might interfere with HIV-1 replication, in the same way that they interfere with the replication of AAV2's helper viruses.

James Trempe and Irving Miller, both of the Carter group, created plasmid pHIVrep (Fig. 2), which contained the AAV2 rep gene driven by the HIV-LTR.⁷ Beth Ann Antoni took over this project and would be the primary author on the publication that resulted from this line of research.⁷ The HIV-LTR is relatively inactive in the absence of a transactivator protein.⁸ However, the group found that there was robust expression of the AAV2 Rep78 protein when this plasmid was transfected into one of the standard human cell lines used by the laboratory, 293 cells.^7,9

This was a serendipitous observation. This cell type was in common use in the Carter Lab, because they are immortalized by integrated copies of the adenovirus e1a and e1b genes.⁹ The presence of these adenovirus genes allows AAV2 replication in the presence of a defective adenovirus,¹⁰ reducing contamination of AAV2 preparations with replication-competent adenovirus, as well as reducing health risks to staff from adenovirus exposure.

Around the same time, it was reported by another research group that products of the adenovirus e1a gene could activate the HIV-LTR.¹¹ The Carter Lab had developed two antibodies against AAV2 Rep proteins,^12,13 which allowed us to estimate the relative concentrations of Rep proteins, using western blots, even in crude nuclear extracts of 293 cells transfected with plasmids expressing the wild-type or mutated versions of the AAV2 rep gene from the HIV-LTR.^5,7

These results were all discussed openly within the laboratory. As a group, we reached the conclusion that this HIV-Rep system, in 293 cells, might facilitate both biochemical analyses of wild-type and mutated Rep proteins, and the examination of the impact of Rep proteins on AAV2 and cellular gene expression in live human cells. The goal was to correlate in vitro biochemical and biophysical properties of the Rep proteins with specific functions in living human cells. Others in the laboratory had been developing systems to generate sufficient quantities of Rep proteins to perform biochemical analyses. In particular, Nor Chejanovsky had success using a baculovirus expression system, but this system did not allow in vivo studies in human cells.⁵

In 1989, reports from two independent laboratories indicated that AAV2 Rep68 and/or Rep78 proteins could bind the hairpin form, but not the linear form, of the AAV2 inverted terminal repeat (ITRs) DNA.^14,15 Immediately recognizing the importance of this work, Barrie suggested that I learn the electrophoretic mobility shift DNA-binding assay used in the reports, and then use it to assess the functionality of the Rep proteins made in the baculovirus system. The project was to see if the baculovirus-produced protein similarly discriminated between the linear and hairpin forms of the ITR.

In addition, Nor had a mutated version of the baculovirus-produced Rep protein, which contained the same amino acid substitution (K340H) as a mutant AAV2 genome, which is dominant-negative for AAV2 DNA replication.¹⁶ We were curious to know if the mutated Rep protein could also bind the hairpin form of the ITR. The same mutation was incorporated into the HIV-Rep plasmid and nuclear extracts of transfected 293 cells were also tested for their DNA-binding activity.⁵

The two previously reported DNA-binding assays differed in the way the target DNA was generated and labeled, and in the way binding to linear ITR DNA was assessed. Ashktorab and Srivastava looked directly for binding activity in their Rep protein-containing nuclear extracts to full-length linear ITR DNA, but there was no guarantee that the linear and hairpin DNA had exactly the same specific activity.¹⁴ Im and Muzyczka showed that 50-fold excess of unlabeled full-length linear ITR DNA failed to disrupt binding to radiolabeled hairpin DNA, but they only used truncated linear DNA for direct binding assays.¹⁵

Because of the Carter Lab's tradition of questioning the way routine processes were done, instead of simply following one of the published procedures, we developed an alternative method for generating the target DNA. Our method produced radiolabeled full-length linear and hairpin ITRs, with exactly the same specific activity, by end-labeling full-length linear ITR DNA, then using part of the resulting labeled DNA to generate hairpin ITR.⁵ We found what appeared to be weak, but detectable binding of Rep78 to full-length radiolabeled linear AAV2 ITR DNA.⁵ Because a small percentage of the full-length linear ITR DNA could have spontaneously converted to the hairpin form, during storage at 4°C, we could not say definitively that this result indicated binding to linear ITR DNA.

Later work by three laboratories, including my own, and one of the laboratories that originally reported that Rep68 and Rep78 did not bind linear DNA, conclusively demonstrated binding to linear DNA, in the AAV ITR and within multiple sites in the human genome, including a site within the preferential integration locus for AAV2 DNA on human chromosome 19.^17–22 As suggested by our 1991 publication,⁵ the Rep68 and Rep78 proteins produced in eukaryotic cells (including the baculovirus expression system in insect cells or the HIV-Rep system in human 293 cells) have a much higher affinity for the hairpin AAV2 ITR than for any linear DNA tested.^20,21

A complicating factor is that Rep68 and Rep78 appear to bind DNA through a combination of DNA sequence-specific contacts and less-specific contacts, possibly mediated by the phosphate backbone of sequences adjacent to the specific binding motif.^{14,15,18,23–25} As a result, binding is not stable if the DNA molecule is too short, even if it contains the complete (∼16 bp) specific binding sequence.¹⁸ It is of interest that there was much less difference in affinity between a 57 bp linear ITR DNA segment and the hairpin ITR, if a maltose binding protein-Rep68 fusion protein was used.¹⁸

In parallel, the Carter group, and later my group, with Sirkka Kyöstiö-Moore as the lead investigator, used the HIV-Rep expression system to examine the roles of the Rep proteins in AAV2 gene expression.^26,27 Because of the laboratory's tradition of co-mentoring and collaboration, HIV-Rep plasmids created for my DNA-binding assays²⁴ were made freely available for Sirkka's gene expression studies.^26,27 Additional plasmids created by Sirkka for her gene expression studies, such as a plasmid that primarily produced the Rep68 protein,^26,27 were shared with me to expand my DNA-binding analyses.²⁴ Rep proteins produced in the HIV-Rep system were also used by my group to examine DNA helicase activities of mutated and wild-type Rep proteins.²⁸ One novel result of this line of research is that two mutated Rep proteins were identified that were dominant-negative for helicase activity, suggesting that the active form is multimeric.²⁸ The rapid development of these lines of research would not have been possible without the diverse, inclusive, and collaborative environment established by Dr. Carter.

Conclusion

I would assert that anyone who is serious about innovation, diversity, and inclusion should promote an environment of co-mentoring, respectful questioning, and open communication. The Carter Lab at NIH provides a fine example.

Author Disclosure

R.A.O. is a coinventor on two patents involving AAV vectors. To the extent that this work will increase the value of those patents, he has a competing interest.

Funding Information

This research was supported by the Intramural Research Program of the National Institutes of Health.