Popular legend has it that when Willie Sutton was asked why he robbed banks, he said “Because that’s where the money is.” I have sometimes been asked why I have worked together with researchers in Europe and Asia. Why travel thousands of miles for a collaboration when there are experts next door? The short answer is “Because those are the collaborations that worked.” But that raises the following questions: What is a fruitful collaboration? How can one make it happen?
Scientific collaborations are driven by need or opportunity. My interest in finding pain genes, and my need to find the right patients, arose from watching my father spend the last years of his life sedated by the opiate medications that were used—unsuccessfully—in an attempt to treat the neuropathic pain that he experienced as a result of nerve damage from diabetes. As in many individuals with neuropathic pain, the medications produced a dulling of my father’s sensorium, with only minimal relief from his discomfort. As I tried to think of ways to alleviate my father’s pain, I was struck again and again by the remarkable fact that some patients with neuropathy experience excruciating pain while others describe mild tingling or electric-like sensations that are not severe and do not require medication. At this time I was already working on sodium channels. By understanding the differences in the sodium channels in these patients, I hoped that I might be able to silence the molecular machinery that generates their pain.
To find out whether patients with painful peripheral neuropathy might have mutations in their sodium channel genes, I needed access to two groups of people: patients with peripheral neuropathy and minimal or no pain, and patients with neuropathy and severe pain. By comparing the two groups, I hoped we could find a culprit gene causing or predisposing certain individuals to pain, or a protective gene that holds pain back.
Diabetes is of course quite prevalent in our society, and, because nerve damage occurs commonly as a complication of diabetes, patients with diabetic neuropathy are not hard to find. I asked colleagues throughout the United States—in general medicine, in endocrinology, and in neurology—if they could identify patients in these two groups, so that I could determine, from their DNA, whether there were mutations in sodium channel genes that correlated with the degree of pain. My vision, of a joint, focused effort with all of the contributors included as full partners in the work, was met with mixed responses including some enthusiasm. But enthusiasm, when present, was not accompanied by action. A common response was, “I’d like to help, but I’m too busy.” A well-regarded diabetes expert signed on to a joint project but did not respond to drafts of the research plan. Another potential collaborator, a neurologist, agreed to send DNA, but, despite reminders, it never arrived.
So, what makes a collaboration succeed? Common purpose, complementary strengths, focus, and a bit of luck. I saw the first three up close as a graduate student in 1967 and 1968 at the Marine Biological Laboratory in Woods Hole, Massachusetts. Although the genes for sodium channels had not been discovered and I did not work on pain at that time, Woods Hole glowed with the theme of teamwork, a theme that continued in our conference room—the place where we planned experiments and which we called the War Room—after I moved to Yale in 1986. Decades after my time in Woods Hole, the search for a pain gene began.
Nestled within a small peninsula jutting out from the base of Cape Cod, the village of Woods Hole was the departure point for the ferry to Martha’s Vineyard. But more importantly, as the home of the Woods Hole Oceanographic Institution and the Marine Biological Laboratory, it was an oasis for scientists. The Marine Biological Laboratory was a unique institution. In the 1960s and 1970s biological researchers from around the world flocked to MBL each summer, drawn by a uniquely collaborative ethos and the availability of a rich ensemble of sea-dwelling species—squid, sea slugs, and even lobsters—whose large nerve cells and relatively simple brain circuitry provided useful experimental models. Several hundred investigators converged each June from leading institutions on both sides of the Atlantic and crammed themselves into small laboratories in the four buildings that made up MBL, within view of the ocean. There were very few organizational snafus, no administrative meetings, and not much paperwork. An investigator merely signed up, arranged for his or her grant to pay the appropriate fees to secure laboratory space, loaded equipment onto a rental truck or an airplane, and moved in.
Informality was the rule at Woods Hole. Lunches were eaten on Stony Beach, a small stretch of sand less than half of the size of a football field. Here, a towel occupied by graduate students might overlap a blanket belonging to a Nobel Prize winner, and a soft drink or a beer might be shared in both directions. My wife and I spent our first summer as a married couple in Woods Hole, and our marriage was celebrated at a party that also honored another newlywed, Sir John Eccles. Seminars as well as formal lectures provided a geyser of information, and the “Tuesday Night Fights” offered a public forum for the exchange of views, often violently different, between the giants of neurophysiology. In this open arena, notable neuroscientists like Stephen Kuffler of Harvard, Harry Grundfest of Columbia, and Mike Bennett of Albert Einstein tore each other’s ideas to shreds, before adjourning for a beer at the local bar, the Cap’n Kidd. Equally important were chance encounters in MBL’s hallways or at its tennis court. Investigators working at the MBL knew that they had to return to their home institutions in late August or early September, and it was understood by all that, for a collaboration to succeed, it had to advance within a short time window. Years later physiologist Ann Stuart summed up the Wood Hole ethos as “let’s just do it.” And it worked.
I moved from Stanford University to Yale in 1986. At Yale I established a research center as a tripartite partnership between Yale University, the Department of Veterans Affairs, and the Paralyzed Veterans of America. Our overall goal was to capitalize on the “molecular revolution” to deliver new therapies for disorders of the nervous system. By design, the research center was multidisciplinary, bringing together cell and molecular biologists, neurophysiologists, ion channel biophysicists, pharmacologists, optical imaging experts, pain researchers, and clinicians in a concerted, focused effort. A key to our progress was seamless collaboration.
Next to my office, within our research center, was a room that I designated as the War Room. In reality, the War Room was a small conference room with a whiteboard on both walls and half a dozen chairs surrounding an oblong table. It was here that my colleagues and I jointly developed research proposals, strategized about experiments, reviewed data, and discussed progress, problems, and prospects. Although our wit did not match that of the Tuesday Night Fights at Woods Hole, our passion and bluntness did: “That will never work!” “You need to think that through more carefully!” “We’ll never complete this experiment unless you come up with a better method!” An outsider, listening from the hall adjacent to the War Room, might have thought that the voices from within it represented adversaries. But we were not opponents. Candid discussion, to-the-point criticism, and a joint effort to wrestle challenges to the ground propelled our research and ultimately provided a template for our international collaborations.
Soon after we published our 2004 paper on erythromelalgia (Cummins, Dib-Hajj, and Waxman 2004) we began two collaborations, one with a clinical researcher in China, Yong Yang, the second with a researcher in the Netherlands, Joost Drenth. Mutations of NaV1.7 in people with the man on fire syndrome would, we hoped, provide “experiments of nature” that would help us to understand how NaV1.7 causes pain. To hone in on the channel and how it works, we wanted to study as many mutations as possible. Both of these highly skilled physicians were receiving referrals of patients with inherited erythromelalgia.
Because patients from throughout China were sent to dermatologist Yong Yang in Beijing for diagnosis and treatment, he saw very rare diseases. Among these were kindreds with familial disorders involving the skin, including painful disorders. In late 2004, shortly after we published our initial paper on the functional profiles of erythromelalgia mutations, Yong Yang sent us a clinical history of a fifteen-year-old boy admitted to the Peking University First Hospital and asked if we were interested in working together to understand that child’s disease. The boy suffered from debilitating burning pain in the feet, and there was a family history that included a similar disorder in a brother. Several weeks later Yong sent the coordinates of the previously undescribed mutation in these two children. The mutation, L858F, substituted a phenylalanine for a leucine at position 858 within the NaV1.7 channel. This mutation was of special interest to us because we suspected, from earlier experiments on another amino acid substitution, L858H, at the same site within the channel (Cummins, Dib-Hajj, and Waxman 2004), that the presence of a leucine at position 858 was crucial for proper functioning of the NaV1.7 channel.
Yong had the patients. The patients carried the mutations. We had the expertise to find out how the mutations worked. We agreed to work together, with Yong sending the coordinates of new NaV1.7 mutations so that we could do the functional profiling in my laboratory. As part of this collaboration Yong spent some months on a sabbatical working directly with us. During this visit his family joined him for a week, and I gave them a tour of New Haven. I bought ice cream cones at a shop next to the Yale Art Gallery and ended up with as much chocolate on my shirt as Yong’s five-year-old son had on his.
There was more to this collaboration. Yong wanted Chongyang Han, an energetic graduate student who had worked with him in Beijing, to learn how to physiologically analyze mutant NaV1.7 channels, and our laboratory was to be the training ground. Chongyang traveled to New Haven from Beijing and spent seven months working in my laboratory. His analysis of the L858F mutation showed us that the substitution of this amino acid enhanced channel activation. This work also showed that, in some patients, a founder mutation (in this case in the patient’s father) can act to launch a new family with erythromelalgia. We published this work together with Yang in 2006 (Han et al. 2006), and Chongyang returned to China to complete his PhD at the Chinese Academy of Medical Sciences. But this was not the end of this story. Chongyang returned to our laboratory in 2008 as a postdoctoral fellow. His initial plan was to return to China after a few years working with me. But this was not how it turned out. Chongyang and his wife decided to remain in America. He was promoted to a higher rank at Yale, associate research scientist, in 2011. After several very productive years in my laboratory Chongyang applied for a green card establishing his permanent residence status, and it was easy for me to write a letter highlighting his unique expertise as an ion channel researcher. He grew into a facile channel biophysicist, and he continues to be an important part of our research team.
Shortly after our 2004 paper (Cummins, Dib-Hajj, and Waxman 2004) appeared, a Dutch medical investigator, Joost Drenth, published two articles describing additional erythromelalgia mutations of NaV1.7 (Drenth et al. 2005; Michiels et al. 2005). An internist with a razor-sharp intellect, Drenth had trained in genetics in France and was best known for important discoveries—including identification of the gene for polycystic liver disease—in which he was unraveling the genetic basis for disorders of the liver, kidneys, and pancreas. More relevant to our interests, he was an erythromelalgia expert. As a medical student, Drenth had published papers on the diagnostic classification of erythromelalgia and on the role of certain medications in triggering the onset of erythromelalgia (Drenth 1989; Drenth and Michiels 1990). His linkage analysis (Drenth et al. 2001) had pointed to the site, on a specific chromosome, that housed the gene for erythromelalgia. Even though he was head of the Department of Gastrointestinal and Liver Diseases, a far cry from disorders involving neurons, Drenth’s clinic at the Radboud University Medical Center in the Netherlands attracted patients from throughout Europe with the man on fire syndrome.
Because of Drenth’s experience with erythromelalgia, and his expertise in medical genetics, we wanted to partner with him. He shared our interest in working together, and soon after we established contact in 2006, he made his first trip to New Haven. Over the ensuing years, working in tandem on a shoestring budget, we discussed interesting cases from throughout Europe as they were referred to Drenth, and we performed the functional profiling of the mutations in New Haven. Our partnership yielded a series of informative studies (Ahn et al. 2010; Cheng et al. 2011; Choi et al. 2011; Choi et al. 2010; Estacion et al. 2008; Estacion et al. 2011; Han, Hoeijmakers, Ahn, et al. 2012; Han, Hoeijmakers, Liu, et al. 2012), each teaching us something about NaV1.7 and the ways it can go awry. One joint study allowed us to tease out the contribution to erythromelalgia of a process called slow inactivation, whereby a long-term silencing due to previous activity builds up to slowly inhibit the function of NaV1.7 sodium channels. A NaV1.7 mutation from another of Drenth’s patients provided clues about why the pain of inherited erythromelalgia usually begins in infancy or early childhood but waits until later in some families. Working together, we learned about the role of alternative splicing, or the use of alternative gene subparts, as the NaV1.7 gene is transcribed during different stages of development. Each mutation taught us something about how NaV1.7 works, and about how its abnormal activity leads to pain. And, in each instance, the mutations that Drenth sent were accompanied by sage comments. Nearly a decade after his first visit, Joost Drenth continues to be a valued collaborator and a friend.
In mid-2010, I received an email from two neurologists, Catharina (Karin) Faber and Ingemar Merkies from Maastricht University in the Netherlands. They had an impressive record of assembling cohorts of patients for clinical studies, and they were experts on peripheral nerve disease. They wrote as follows:
Dear Professor Waxman,
We are two neurologists from The Netherlands (Maastricht University Medical Centre) with particular interest in peripheral neuropathies.
We have read your papers that helped us in creating a genetic hypothesis. In the last 3 years we have succeeded in creating a comprehensive database with > 60 patients with neuropathy, containing data ranging from aspects like pain and autonomic symptoms, skin biopsy findings (intraepidermal nerve fibre density [IENFD]), temperature threshold testing, to disability and quality of life assessments.
Some of these patients are believed to have (new) mutations in the SCN9A gene. … Based on these findings, we ask whether you would be interested in interchanging ideas and perhaps come to a scientific joined venture. … We would be very honoured if we could meet and we are certainly willing to fly over to your institute to discuss things. Tentatively, these are some dates that we could come over: July 16th–July 20th, September 25th–Sept 29th, October 18th–October 20th, October 25th–October 28th.
We thank you beforehand for your time and hope that you would be interested in meeting us.
Sincerely yours,
Catharina Faber / Ingemar Merkies
Every scientist studying a rare disease hopes that it will hold lessons about more common disorders. Now Karin Faber and Ingemar Merkies were suggesting a springboard, from which to extrapolate from a rare genetic disorder, inherited erythromelalgia, to a common disease, painful neuropathy. It took me only a few minutes to reply:
Dear Drs. Faber and Merkies,
It was good to hear from you. We would like to collaborate with you. Working together, I think we can come up with a significant story, spanning from the clinical to the molecular and biophysical, on the relationship of NaV1.7 mutations and neuropathies.
With respect to your questions:
Yes, we would be very interested in exchanging ideas and shaping a collaboration. Our research center possesses unique resources for the profiling and functional characterization of mutant NaV1.7 channels. Each mutation’s analysis is labor-intensive, so we prioritize the mutations in order of interest. Do you have any mutations in patients with i) a definitive and well-characterized small fiber neuropathy (e.g. in terms of IENFD) and ii) a strong family history, with the mutation segregating with disease (if we know that all affected patients within a family house the mutation, it provides strong evidence that the mutation is linked to disease). If you could send us the history etc. and the mutation from such families, we could move those mutations to the front of the line and get our physiologists and biophysicists to work on them now, in advance of your visit, so that we might have data waiting for you.
Finally, yes, we would love to have you visit. … Yale University is located in New Haven … a relatively European city (for the U.S.!) and a pleasant place to visit this time of year.
I look forward to hearing from you, and to meeting you,
Steve Waxman
It was clear to all of us, even prior to July 2010 when Faber and Merkies made their first visit to New Haven, that our two groups, working together, were poised to make exciting progress. Our research team, which at that time included Sulayman Dib-Hajj, Mark Estacion, Chongyang Han, Jin Choi, Xiaoyang Cheng, Hye-Sook Ahn, and Jianying Huang, was highly talented. In Maastricht, Faber and Merkies had established a peripheral neuropathy clinic that attracted patients from throughout Europe. Faber, Merkies, and I recognized that we had the makings of a fruitful collaboration. But my team worked in New Haven, while Faber and Merkies lived and worked in the Netherlands three thousand miles away. Would it work?
Faber, Merkies, and I spent the first part of their 2010 visit talking about logistics. We had the manpower to make things work. Rather than looking for additional funds, we decided to move ahead using existing resources. Patients would be studied clinically in Maastricht, and then their mutations would be functionally analyzed at Yale. It was perhaps fortuitous that all of us worked on a “let’s do it now!” time clock. Remarkably, by late August 2010 we reached, together with our institutions in Maastricht and New Haven, a streamlined agreement which established the official basis for our collaboration.
In 2010, the term “translational research”—research that linked the laboratory bench and the clinic—was evoking excitement throughout the biomedical community. The Maastricht–Yale partnership, like our transoceanic collaborations with Drenth and Yong Yang, was translational and brought together multiple parts—on the European end, highly effective accrual of human subjects, quantitative clinical evaluation, assessment of peripheral nerve pathology, computerized sensory testing, human electrophysiology, and clinical genetics; in New Haven, molecular and cell biology, ion channel biophysics, cellular neurophysiology, computer modeling, and pharmacology. The results of this collaboration exceeded our expectations. Faber, Merkies, and their co-workers in Maastricht made a major contribution even at the outset of this work, providing a series of twenty-eight “gold standard” patients in whom a diagnosis of small-fiber neuropathy had been confirmed, both by demonstrating nerve damage by analysis of skin biopsies, and by computerized testing which could detect subtle deficits in sensation (Faber et al. 2012). Their clinical acumen echoed throughout the project.
As we moved forward, we stayed in close touch by email and phone, the only ground rule being that we would try to not wake each other in the middle of the night. On at least a few occasions, I violated this rule, with calls to a sleeping Netherlands announcing, for example, “We have it! The mutation in patient XX destabilizes the closed state of the NaV1.7 channel!” Only a science addict could appreciate such a call. Candor, reminiscent of the Tuesday Night Fights at Woods Hole and echoing debates in our War Room at Yale, punctuated our progress. In commenting on each others’ ideas, none of us was hesitant to drive a point home. Emails and telephone discussions incorporated friendly queries about families or intermittent quips about Ingemar Merkies’s golf game (“It’s just physics with a little ball,” he would insist). But other than this, our conversations included few niceties. Summaries of progress and drafts of articles bounced from one side of the ocean to the other filled with comments such as “how do you know?,” “that doesn’t make sense,” or “provide a strong rationale or TAKE THIS OUT.” Sometimes a comment was accompanied by “NO!!” We made up for our lack of social graces when we met face-to-face, over good food, cappuccino, and fine wine.
Within three years, from this initial group of patients and additional patients that followed, we identified a string of mutations in NaV1.7 linked to small fiber neuropathy, each instructive in its own way. A series of papers (Ahn et al. 2013; Estacion et al. 2011; Faber et al. 2012; Han, Hoeijmakers, Ahn, et al. 2012; Han, Hoeijmakers, Liu, et al. 2012; Persson et al. 2013; Lauria et al. 2012; Hoeijmakers et al. 2012; Estacion et al. 2015) emanated from these discussions. In this work, we made the leap from inherited erythromelalgia, a rare disease, to a common disorder and showed that mutations of NaV1.7 play a major role in painful neuropathy.
Neither Yong Yang nor Drenth nor the Maastricht group nor my team would have made so much progress alone. We did it well, and quickly, and together. So, what makes a collaboration work? Did we have to cross borders? As I wrote this book I posed that question to my Dutch colleagues. Drenth cited our complementary strengths, “We all need each other.” And he said that he appreciated “constantly being pushed.” I don’t think anybody else has ever thanked me for being a nag! Karin Faber answered with a Dutch saying: “Trust arrives by foot, but leaves on a horse.” This means, she explained, that trust is precious but fragile: It can easily evaporate. And indeed, we steered clear of horses as we built these collaborations. Only a little money was involved, and we shared what there was. There were no committees. We said what we thought, rapidly reached consensus or agreed to disagree, and moved ahead quickly. I’m reminded of animated discussions on Woods Hole’s Stony Beach where scientists from both sides of the globe could watch the sun set, plan an experiment, and return to the laboratory with a simple sense of “let’s just do it.”