Tristan: Why did you choose Leiden and the Netherlands to continue your career, and how did you meet Professor Alex van der Eb?

Dr. Frank Graham: It is quite a story. In 1968 and 1969, I was finishing my PhD at the Ontario Cancer Institute. My wife, Silvia Bacchetti, who is also a scientist, and I wanted to relocate to Europe for a postdoctoral position. Marvin Gold at OCI had worked alongside the Dutch scientist Adrian de Waard and put us in touch. Adrian passed my CV to Professor J.A. Cohen at the University of Leiden, who arranged a position for me in Physiological Chemistry and helped Silvia join a group at TNO in Delft, led by Professor Arthur Rorsch. Sadly, Professor Cohen passed away before we arrived, so we never had the opportunity to thank him in person. I later learned I would work with Alex van der Eb. We first met in Munich in the winter of 1969 at a scientific meeting. Lex, as we called him, had just completed a postdoc at Caltech and was setting up his lab in Leiden. He had the space and funding for me to join, so we moved in the spring of 1970.

Tristan: Did your research interests already align, or did the project evolve once you arrived?

Dr. Frank Graham: My background was in cell culture, cell biology, and biochemistry, not virology, and certainly not adenoviruses. Once it was settled that I would join an adenovirus group, I read Lex’s papers and a 1968 review by Maurice Green, which summarized almost everything known about adenoviruses at the time. About thirty human adenoviruses have been identified and grouped by serotype. Some were highly tumourigenic in rodents, some intermediate, and some were not tumourigenic. There seemed to be a link between tumourigenicity and DNA composition, with oncogenic viruses being AT-rich and the non-oncogenic ones GC-rich. That turned out to be a red herring, but it shaped my early thinking. I received support from the National Cancer Institute of Canada and proposed a postdoctoral plan to transform cells using adenovirus DNA, fractionate the genome, and identify which parts could transform cells. In hindsight, it was naive because there was no proven way to transform mammalian cells with DNA or to fractionate viral genomes in the way I imagined. Even so, the fellowship was awarded, and I began.

Tristan: Did you manage to do what you proposed?

Dr. Frank Graham: In the end, yes, but it took a lot of work. I learned to grow adenovirus, perform plaque assays, and purify viral DNA, with help from Lex and our technician, Marie Alice Salgado. I first tried the DEAE dextran method to introduce viral DNA into HeLa cells. It worked, but the efficiency was low and unreliable, which was insufficient to transform cells reproducibly. After more than a year with little to show, Lex asked the key question. How did I know how much DNA was getting into cells? I decided to measure uptake directly with radioactive DNA. Before I began, I revisited the bacterial transformation literature and noticed the role of divalent cations such as calcium. Our media already contained low levels, so I asked what would happen if I increased them. Adding calcium chloride at concentrations above 100 millimolar resulted in a nearly 100-fold increase in DNA uptake and a similar rise in infectivity in plaque assays. Magnesium did not help. When I excluded DEAE dextran, the effect persisted. At higher calcium, the DNA solutions turned cloudy. That was Calcium Phosphate forming, and the DNA co-precipitated with it. This became the Calcium Phosphate Transfection, a widely used method for transfection that has been in use for many years. It allowed us to reliably transform rodent kidney cells and then proceed to mapping. We demonstrated that sheared adenoviral DNA remained capable of transformation, even at concentrations of five to ten percent of the genome. By separating half molecules, we mapped the transforming activity to the left half, and, with Herbert Heyneker at TNO, we narrowed it down to about one percent from the left end, the region known as Early Region 1.

Tristan: Under what conditions was HEK293 first isolated and then transfected? Did you have biosafety cabinets or single-use materials? Did you use antibiotics?

Dr. Frank Graham: HEK293 would not exist without the Calcium Phosphate Transfection. Many human cells are permissive for adenovirus replication, so an intact genome kills the cells. By fragmenting the DNA to remove infectivity, we could transform permissive cells. Syrian hamster cells worked first. We then aimed to use human embryonic kidney cells to propagate E1-defective viruses and later grow E1-deleted vectors. Human cells were significantly more difficult to transform than rodent cells. Across many dishes, I found only two transformed colonies. By then, I could recognise the distinctive morphology of transformed cells. Of the two colonies, I coaxed one into a line. I named it 293 3.1, from dish 3.1 in experiment 293. The name was later shortened to 293. Our cultural conditions were basic. No laminar flow cabinets. We worked in still hoods and used Bunsen burners to sterilise pipettes and caps. We started with reusable glassware and then transitioned to single-use items, including flasks, dishes, and pipettes. We used antibiotics, and mould contamination during long plaque incubations could be difficult to avoid. I maintained 293 continuously from the initial transformation for almost six hundred days.

Tristan: HEK293 also contains the adenovirus pIX gene. Has that influenced the biology or use of the cells?

Dr. Frank Graham: pIX sits within E1 but is not expressed in 293 cells, so it is unlikely to affect their biology. HEK293 contains viral sequences only from the left end of the genome.

Tristan: At the time, did you foresee how widely HEK293 would be used?

Dr. Frank Graham: I expected 293 to be useful for work on adenoviruses, including the creation and growth of viruses with E1 mutations and E1-deleted vectors. I did not foresee how efficiently the cells would take up DNA and produce protein, or how well they would serve as hosts for other vectors, such as AAV and lentiviruses. That is why HEK293 became the workhorse for viral vector production.

Tristan: Will HEK293 remain central, or might other cell lines replace it for viral vector manufacture?

Dr. Frank Graham: It is fifty years since 293 was created. Other human cell lines that contain and express adenovirus E1 genes have been developed, for example, PER.C6, yet they have not replaced 293 for many applications. As Yogi Berra said, it is difficult to make predictions, especially about the future.

Tristan: How important have collaborations been in your work?

Dr. Frank Graham: Essential. The most important was with Lex van der Eb, whose support and mentoring were vital. I worked with Willy Russell and Jim Smiley on the initial characterizations of 293 cells, and with Jim Williams on host range adenoviruses that grow in 293 cells but not in other human cells due to E1 defects. Later, at McMaster University, I collaborated with Lud Prevec for many years on methods for generating adenovirus vectors and their application in gene transfer and vaccine development.

Tristan: Any fond memories from your time in Leiden and the Netherlands, away from the lab?

Dr. Frank Graham: Many. Adrian de Waard met us at Schiphol when we arrived. We rented his house in Noordwijkerhout for six months while he was on sabbatical. That spring, we looked out over fields of tulips in full bloom. Hans van Ormondt took us to dinner on our first night. Our closest friendship was with Herbert Heyneker, a colleague of my wife’s at TNO. With his help, we moved to a flat on a farm at the southern edge of Wassenaar, very convenient for both Leiden and Delft, and very gezellig. We spent a week at a sailing school in Friesland, where we learned to sail in Dutch, purchased a Dutch dinghy called a Stern, and sailed it in Holland and on trips in the Mediterranean off the coast of Italy.

Tristan: How do you see the development of the viral vector field and cell and gene therapy?

Dr. Frank Graham: I have been retired for more than twenty years, so I would not claim to be fully up to date. Even so, the progress with AAV and lentiviral vectors in correcting genetic diseases has been impressive. I am hopeful that technologies such as CRISPR will bring further advances in treating genetic and other diseases. I always thought adenovirus vectors could be valuable as vaccines. That has been realised in some cases, and I think there is still more potential to explore.

Dr. Frank Graham: Non‑profits like NecstGen can act as facilitators and intermediaries between academia and industry. Bringing academic discoveries to patients requires resources for patents, clinical trials, and market access, which are often beyond university means. Non‑profits can provide access to facilities and technologies that many labs cannot afford, help identify promising intellectual property, and support its transfer to companies that can develop it further. In a way, they can play a role similar to an agent who smooths the path from one home to the next.

Tristan: Looking back, what are your most significant milestones?

Dr. Frank Graham: First, the Calcium Phosphate Transfection, because it made so much else possible. Second, demonstrating that linear DNA fragments can transform cells and mapping the adenovirus-transforming genes on the genome. Third, the creation of HEK293, which enabled the use of adenovirus and other viral vectors for vaccines and gene therapy.

Dr. Graham’s breakthroughs in Leiden helped shape modern genetic engineering. HEK293 remains central to viral vector research and manufacturing, and we acknowledge its role across the field. Leiden was, and is, a great place to do science. With sincere thanks to Dr. Frank Graham for the interview and for the technology that continues to power Cell and Gene Therapy.