We are deeply saddened to share news about the passing of a long-time beloved faculty colleague, Dr. Jon Wolff (pictured below), after a courageous battle with cancer.

Jon was a professor and division chief of genetics and metabolism in pediatrics for most of the 20 years he spent at the University of Wisconsin School of Medicine and Public Health. Prior to joining our faculty, he received his undergraduate education at Cornell University, received his M.D. from The Johns Hopkins School of Medicine, completed clinical training in Pediatrics and Medical Genetics at the University of California-San Diego, and was a post-doctoral fellow at the Agouron Institute. He remained an honorary volunteer faculty member here and was always a strong supporter of our master of genetic counselor studies degree program throughout his career.

As a physician-scientist, Jon made significant contributions to diagnostic and therapeutic knowledge that helped transform the standard-of-care for several different genetic conditions. As an advisory member for Wisconsin’s Newborn Screening Program he helped establish new programs for detecting metabolic disorders and genetic conditions. He also led medical genetics courses for medical students and genetic counseling students over several years, adding cutting-edge molecular content, bringing patients into classrooms as teachers, and greatly improving medical and molecular genetic education throughout various classroom, clinic, and laboratory settings.

Jon played an internationally recognized role in advancing gene therapy for liver, muscle, and brain disorders and in developing innovative techniques to transfer genetic material into cells, including novel vascular delivery of naked nucleic acids into tissues. Collaborating with other scientists he helped develop genetic vaccines. Jon was a highly productive physician-scientist, with well over 150 scientific publications and 80 patents, and he served on multiple committees and boards, including those at the NIH, the American Society of Gene Therapy, several journals, and the French Muscular Dystrophy Association (AFM). With colleagues he founded MirusBio Corporation in 1995. While there he led cutting edge work developing novel nanotechnologies to transfer siRNA into cells. The successful MirusBio therapeutics division he helped create was initially acquired by Roche and later by Arrowhead Research Corporation.

Although Jon was always well ahead of his time as a novel thought leader in gene therapy, throughout his career he also remained passionate about developing innovative and compassionate approaches to the care of people with genetic disorders and for the ethical use of genetic information and technology in health care in general. His patients trusted his tremendous expertise and devotion to their care, and remember his great sense of humor – he had a talent for bringing them joy, even in difficult circumstances. His passion to serve patients led to his creation of the charitable nonprofit foundation, Genetic Support Foundation, which has been dedicated to providing genetic counseling services for all since 2014, and where he has served as chairman of the board. When Jon became a patient himself he applied his incredible intellect and resourcefulness to explore innovative approaches to his own treatment, and in the process he became a powerful support resource for others with cancer.

Jon will be deeply missed by all of his departmental and SMPH colleagues, many of which were his former students and residents, former patients, and many close friends both in our local community and within the global genetics community.

University of Wisconsin researchers have an extensive history of making cutting edge discoveries in the field of molecular biology, and their works continue to inspire the technologies of today.

In the 1960s, UW biochemistry professor Har Gobind Khorana discovered how the genetic code translates to amino acids, the fundamental building blocks of the proteins which drive all human and animal physiology. In 1975, UW oncology professor Howard Temin discovered the reverse transcriptase enzyme, a breakthrough that challenged the central dogma by revealing that the genetic code could flow backward from RNA to DNA.

Former UW researcher and physician, Dr. Jon Wolff, is no exception to this track record of excellence. Wolff was well-known in the Madison area and beyond as an exceptional physician and forward thinker in the field of genetics, according to his colleagues. Wolff passed away in April 2020, but his legacy lives on in the spirit of his colleagues and the scientific work he left behind.

Dr. David Wargowsky, a physician who worked with Wolff at UW Health, recalled Wolff’s vast knowledge and intellect.

“[Wolff] was incredibly bright and knowledgeable — he knew about extremely rare metabolic conditions that others had never heard of,” Wargowsky said. “He was a wonderful resource for the hospital, his patients and the University.”

The Discovery that Changed Gene Therapy Forever

Wolff’s research into the delivery of naked DNA or RNA into the muscle cells of a living organism, is now more relevant than ever, as it spearheaded core principles that helped inspire the complex gene therapies we see today, from cancer treatment to chronic disease management. Also, Wolff’s work was foundational to the development of the mRNA COVID-19 vaccines.

According to the CDC, two of the three COVID-19 vaccines available in the U.S. are mRNA-based. These vaccines deliver a genetic blueprint in the form of mRNA, which instructs cells to produce proteins similar to those found in the SARS-CoV-2 virus, popularly known as the coronavirus. This trains the immune system to recognize the virus and helps vaccinated individuals elicit an immune response if they are infected.

The mRNA COVID-19 vaccines are the first of their kind, according to the CDC. Delivering genetic material to human cells unharmed and ready to be translated to proteins was once considered a daunting task, and at one point was thought by some to be impossible. In spite of this, Wolff and his team of researchers set out to advance this area of study.

One of Wolff’s most influential research articles, “Direct gene transfer into mouse muscle in vivo,” was the first to show that naked DNA or RNA could be delivered to the muscle cells of a living organism and expressed as proteins. According to the article, in the past, this had only been accomplished using viral vectors or other carrier methods which were less effective.

Dr. Gyula Acsadi, a co-author of the paper, said Wolff was originally interested in working on gene therapies for Parkinson’s disease, which spurred this proof-of-concept study. As stated in the conclusion of Wolff’s 1990 article, they foresaw that intramuscular injection of genes encoding viral antigens may provide an alternative approach to vaccine development. Acsadi said that at the time, this concept was completely new.

When asked if Wolff’s 1990 article was fundamental to COVID-19 vaccine development, Acsadi pointed to the sheer number of times it has been cited by researchers.

According to the National Institutes of Health’s PubMed database, the article stands at 634 citations, with seven coming from the past month alone.

“This paper has been cited many times by vaccine researchers … I assume that people recognize that this was a fundamental paper for this work because they are citing it almost constantly,” Acsadi said. “I feel that this discovery, especially showing that injection into muscle made such a strong response in terms of protein production, is important to today’s researchers.”

Though the manufacturers of the mRNA COVID-19 vaccines do not specifically reference Wolff’s work, Acsadi believes that Wolff’s influence on these technologies is strong. This is supported by citations of Wolff’s work in supporting research, which show that their work had direct influence on today’s therapies.

Wolff’s Work After the Discovery

After completing the 1990 article, which helped make Wolff famous among gene therapy researchers, he continued his academic work and went on to publish hundreds of articles. Many were published in prestigious scientific journals such as “Science” and “Nature.”

Acsadi said Wolff eventually went on to co-found Madison-based company Mirus Bio, which was heavily influenced by his research at UW. At Mirus, Wolff and his colleagues developed gene therapy transfection reagents, among other pharmaceutical products.

Acsadi said Wolff’s work with the company included development of new liposomal delivery systems, small hydrophobic packages which protect genetic material from degradation and are readily taken up by cells. A similar liposomal technology is a key component of the COVID-19 vaccines, which allows the mRNA to reach cells unscathed.

“His legacy is promoting DNA and RNA as therapeutic, and I think that [his work] has opened up the avenue of thinking outside of the box by using nonviral gene therapy techniques, which currently has proven clinical applications,” Acsadi said.

He said in the past most gene therapies relied on viral vectors for gene delivery, which is no longer a limitation to current therapies, in part due to Wolff’s work.

Today, ClinicalTrials.gov lists over 400 active gene therapy studies occurring all over the world. Many of these studies involve DNA and RNA based vaccines, which ostensibly were influenced in some way by Wolff’s work.

The experiment was so elementary, and the results so surprising, that researchers working with San Diego’s Vical Inc. couldn’t really believe what they were seeing. It all seemed too simple.

They had been injecting submicroscopic fatty globules containing DNA or RNA into mice to see what would happen. The idea was that the fat globules, called liposomes, would be taken up by cells. The cells would use the genetic material inside to make proteins they couldn’t otherwise make.

The researchers found moderate success with that, but the rigors of science demanded that the experiment have a “control” portion--injecting the raw DNA or RNA into the mice to show that the liposomes themselves were making it possible for the new genes to be incorporated into the cell’s processes.

It turned out the cells like the raw material even better and began making the new proteins for as long as six months.

“This was a big surprise, and that’s really what you’re looking for in this area,” said [Dr. Philip Louis Felgner (born 1950)], director of product development at Vical. Felgner worked on the experiment with [Dr. Jon Asher Wolff (born 1956)] and others at the University of Wisconsin at Madison.

Researchers spent several months longer trying to find flaws in their methods or their conclusions. The literature of science is littered with examples of experimental results that deserved the label of too good to be true, explained [Dr. Karl Yoder Hostetler (born 1939)], vice president for research and development at Vical.

“We didn’t want any fiascoes,” he said.

Vical hopes that the results of this checking and double-checking, reported in today’s issue of the journal Science, will convert the company from a bare-bones start-up to a major player in the ranks of San Diego’s biotechnology community.

The company, which was founded in 1987, hopes to find financing to more than double its scientific staff of 22 as a result of the study. It is talking with several large drug companies to see if any would like to buy into the follow-up studies on the new gene transfer method, said Vical President Wick Goodspeed.

Some familiar names in San Diego science and business have played a role in Vical. Among them:

[Dr. Karl Yoder Hostetler (born 1939)], who is on leave from his longtime post as professor of medicine in residence at UC San Diego. His specialties include investigating ways to use lipid chemistry to improve the effectiveness of drugs.

[Dr. Douglas Daniel Richman (born 1943)], a founder and scientific adviser to the firm. Richman is a professor in residence of medicine and pathology at UCSD, specializing in virology and clinical trials of AIDS treatments.

[Dr. Dennis A. Carson (born 1936) ], also a scientific adviser to the firm. Carson recently resigned as head of the division of clinical immunology at Scripps Clinic to become head of UCSD’s new institute for research on aging.

Timothy Wollaeger, chairman of the board. Wollaeger formerly was senior vice president in charge of finance and administration for Hybritech Inc., the monoclonal antibody firm whose success was capped in 1986 with its $485-million acquisition by Eli Lilly & Co.

Howard E. (Ted) Greene, a director of Vical. He formerly was chief executive officer for Hybritech. Greene and Wollaeger were the driving forces behind Biovest Partners, a venture capital firm that financed several San Diego biotech firms.

W. Larry Respess, a Vical director. A leader in biotech patent law, he formerly was general counsel of Gen-Probe and Hybritech.

Until now, the best combination of science and business for Vical has been the multi-year research contract it received last summer from Burroughs Wellcome Co. to develop new forms of AZT for AIDS therapy. The study is investigating the idea that encasing AZT in fat globules would make it more powerful within the body.

The gene-insertion technique reported in Science this week is being suggested as a way to cause the body to generate proteins that would block persistent viral infections, ranging from AIDS to herpes. It also is seen as having potential use as a way to trigger cells to immunize the body against diseases, researchers say.

Vical is calling the new method “gene therapeutics,” to distinguish it from the traditional goal of gene therapy, which uses viruses to insert missing genes into the genetic codes of people with genetic diseases.

The so-called retroviral method has proved difficult and slow, despite several years of intense effort by research groups around the country, including a group led by Dr. Theodore Friedmann at UCSD.

Because retroviruses insert their own genetic code into the cells of their host, the method is also expected to be problematic as a gene therapy technique--since some scientists worry that this could harm the patient irreversibly in some unforeseen way.

Inserting the genes themselves into muscle cells--without any retroviral carrier--avoids this stumbling block entirely, [Dr. Philip Louis Felgner (born 1950)] said. The genes do their work of producing proteins, called expression, but they don’t seem to affect the cell’s own genetic structure, he said.

“People have worked in the gene therapy area for years assuming that a rather complex viral delivery system would be required in order to get expression. And we have found that you can do it very simply,” Felgner said.

It was the slowness of the gene therapy field that led Felgner’s collaborator, of the University of Wisconsin, to decide less than two years ago to get out of it altogether, Wolff said in a telephone interview.

Wolff was an assistant professor and a researcher in Friedmann’s UCSD lab before going to Wisconsin as an assistant professor of pediatrics and medical genetics in 1988.

“I had pretty much planned to get out of the gene therapy field because I got discouraged with the retroviral approach. Scientifically, it wasn’t very challenging,” he said. “Everybody was doing the same thing, and nothing was working that well.”

The results of the research contract with Vical, begun in January, 1989, have rekindled his enthusiasm, [Dr. Jon Asher Wolff (born 1956)] said.

He believes that, in the end, genetic therapies will involve a variety of techniques, not just the Vical method. But he and [Dr. Philip Louis Felgner (born 1950)] acknowledge that they expect some resistance to their ideas from the traditional gene therapy community.

“You’re talking about somebody who has spent his life in this field, and who would like to make the real breakthroughs that are going to allow it to be used in patients with diseases,” Felgner said. “There’s quite a bit at stake.”

Other collaborators with Wolff and Felgner on the research were [Dr. Robert Wallace Malone (born 1959)] of Vical and Phillip Williams, Wang Chong, Gyula Acsadi and Agnes Jani in Wisconsin.

Vical is planning to try to patent the technique, even though it involves no novel or complex steps unfamiliar to molecular biologists. In essence, it involves preparing DNA or RNA with standard techniques and then injecting it in the conventional way into muscle.

“The reason why we have patent position is that it was such a total surprise. Some of those things are the best patents you can get,” Felgner said. “Nobody who was ‘skilled in the art’ would have ever thought that what we have accomplished here was even possible. Nobody would have even thought to do the experiment.”

ABSTRACT

The term "gene therapy" was coined to distinguish it from the Orwellian connotations of "human genetic engineering," which, in turn, was derived from the term "genetic engineering." Genetic engineering was first used at the Sixth International Congress of Genetics held in 1932 and was taken to mean "the application of genetic principles to animal and plant breeding." Once the basics of molecular genetics and gene transfer in bacteria were established in the 1960s, gene transfer into animals and humans using either viral vectors and/or genetically modified cultured cells became inevitable. Despite the early exposition of the concept of gene therapy, progress awaited the advent ofrecombinant D N A technology. The lack of trustworthy techniques did not stop m a n y researchers from attempting to transfer genes into cells in culture, animals, and humans. Viral genomes were used for the development of the first relatively efiicient methods for gene transfer into m a mamilian cells in culture. In the late 1970s, early transfection techniques were combined with selection systems for cultured cells and recombinant D N A technology. With the development of retroviral vectors in the early 1980s, the possibility of efficient gene transfer into mammalian cells for the purpose of gene therapy became widely accepted.

OVERVIEW SUMMARY

This article outlines the conceptual and experimental beginnings of gene therapy. Once the basics of molecular genetics became apparent, the idea for gene therapy arose independently in the minds of perspicacious investigators. Concurrently experimentalists attempted gene transfer into mammalian cells in culture and animals in anticipation of recombinant D N A technology. Unreliable practices in early studies opened the experimental results and the entire approach of gene therapy to question, even though many of the basic methods and theories would eventually be proven correct. It was not until transfection techniques and selection systems for cultured cells were coupled with the ability to manipulate recombinant D N A that substantive progress was made in gene transfer.

INTRODUCTION

"Victory has 1000 fathers; defeat is an orphan.'' (Count Ciano).

As THE ENTERPRISE OF GENE THERAPY enters an accelerating period of growth, it is timely to examine its earlier, more tentative phase of development. The rise of gene therapy has been compared to the growth of aeraonautics. Just as the ancient Greeks' fantasies about Ikaros' disastrous flight gave way to Da Vinci's fanciful, if unfiightworthy, drawings of flying machines, and later to the design of many strange and awkward contraptions, so too has the early history of gene therapy been marked by numerous dreamers, prophets of doom, naysayers, missteps, and fiascoes. In both cases, many experiments were attempted too early, before the necessary tools were available. Bu just as the invention ofthe Wright brothers' crade, but successful, early flying machines rapidly led to the development of supersonic jets and space travel, so, too, has the development ofrecombinant D N A technology brought us, in an astonishingly brief span of time, to the brink of curing hitherto untreatable diseases.

Gene therapy is not, as is generally thought, a new concept. The notion that genes can be used to treat human disease actually goes back several decades. M a n y of the pioneers of m o d em genetics were aware that their discoveries eventually would lead to medical applications. Perhaps this history is obscure because halting progress in the science of gene therapy and transfer was slow in coming. N o w that researchers and their patients have embarked on promising gene therapy trials, it m ay be useful to review the evolution of gene therapy's theoretical foundations and thereby gain insights into its medical, scientific, social, ethical, and economic milieu.

Reviews covering the later advances in gene therapy (Anderson, 1984; Friedmann, 1989; Miller, 1992; Mulligan, 1993) and in nonviral gene transfer methods (Feigner, 1990) have appeared and one of the early proponents for gene therapy has recently explored its history (Friedmann, 1992). This contribution, while stressing the early attempts at gene transfer in cell cultures and animals, seeks to place these studies in a wider historical context. Moreover, a historical perspective may be useful for evaluation of the "prior art" in patent applications.

THE BEGINNINGS OF GENETICS

Long before the discovery of the gene, our species practiced genetic manipulation. Since ancient times, animals and plants have been bred for specific intent. W e see the fraits of their labors when w e eat com, train a dog, race a horse, or smell a rose. Such selective breeding practices were the foreranners of the m o d e m science of genetics. The First Intemational Congress of Genetics was held in 1899 in London but was actually called the Intemational Conference of Hybridization and on the Cross-Breeding of Varieties (Crow, 1992). William Bateson proposed use of the term "genetics" at the Third Intemational Congress of Genetics in 1906. In the interim, Hugo de Vries and others had rediscovered Mendel's work, which originated the concept of the gene as a unit of heredity.

One of the first uses of the term "genetic engineering" was in a paper of that title presented at the Sixth Intemational Congress of Genetics held in 1932 in Ithaca, N e w York (Crow, 1992). Genetic engineering was defined as the application of genetic principles to animal and plant breeding. T o Marxist commentators, the term "genetic engineering'' was synonymous with "eugenics" in contrast to "social engineering" of U S S R public policy (Lederberg, 1973). The term "gene therapy" was adopted to distinguish itself from the ominous, germ-line perceptions ofthe term "human genetic engineering." While the human genome has previously been modified indirectly by human activity, more direct manipulations could usher in a new epoch.

THERAPEUTIC CUSTOMS

The concept of gene therapy is set in the context of pharmacological and surgical traditions. Gene therapy can be defined as the application of genetic principles to the treatment of human disease. Screening programs for phenylketonuria and Tay- Sachs disease and many conventional medical and surgical therapeutic approaches, such as liver transplantation, could fall under this definition because they were founded on an understanding of their genetics and biochemistry. More specifically, the term gene therapy unites pharmacotherapeutic with genetic principles, implicating the use of a polynucleotide to treat the disease state.

In 1878, Langley proposed the concept of the receptor substance, n o w known simply as the receptor (Goodman and Gilman, 1975). The hypothesis that interactions between a drag and its receptor are govemed by the law of mass action was further developed by A.J. Clark in the 1920s. The molecular understanding of protein function and enzyme action led to development or rational drag design based on targeting the receptor, or active site (Peratz, 1992). Antisense and ribozyme gene therapies represent an extension of this concept. Approaches that involve the addition or modification of genes are conceptually similar to protein replacement therapies, such as those for diabetes mellitus and the hemophilias, in that the naturally occurring macromolecules are administered as remedies. Transient expression of exogenous genes, e.g., the use of artificial m R N A (Wolff et al., 1990), would also be isologous to protein replacement therapies. Gene therapy with permanent incorporation of the therapeutic gene into the cell's chromosomes more closely resembles surgery, in which a tissue or organ is modified for the life span ofthe recipient.

MOLECULAR GENETICS

The discovery by Avery, MacLeod, and McCarthy that a gene could be transferred within nucleic acids is a critical point of reference (Avery et al., 1944). Their article begins with a prescient sentence:

• Biologists have long attempted by chemical means to induce in higher organisms predictable and specific changes which thereafter could be transmitted in series as hereditary characters.

The capacity for virases to transmit genes was first demonstrated in Salmonella (Zinder and Lederberg, 1952). The idea that viral genomes could become an abiding part of cell genomes (i.e., prophages) was discovered with bacteriophages (Lederberg, 1956; Lwoff, 1972) and then extended to animal virases. The ability of Rous sarcoma viras (RSV)-transformed cells in culture to produce new viras led to the idea that viral genes were responsible for the cells' transformation. Further studies of R S V infections in vitro led to the proviral hypothesis (Temin, 1976, 1971). Other studies demonstrated integration of SV40 viral D N A in SV40-transfomied cells (Sambrook et al., 1968).

The elucidation of the stracture of D N A and its parsimony with DNA's surmised function brought forth an upheaval on biology (Watson and Crick, 1953). The history of the subsequent discoveries of m R N A and the formulation of the central dogma (genetic information flows from D N A to R N A to pro tein) is chronicled in Horace Judson's seminal book. The Eighth Day of Creation (Judson, 1979).

SPECULATIONS ON GENE THERAPY

In the interval immediately preceeding the recombinant DNA era, key aspects of gene therapy were elaborated by forward-thinking investigators of the day. In N e w York on May, 1966 (Tatum, 1966), Edward Tatum predicted that viruses could be used to transduce genes.

• • • Finally, it can be anticipated that virases will be effectively used for man's benefit, in theoretical studies in somatic-cell genetics and possibly in genetic therapy . . W e can even be somewhat optimistic on the longrange possibility of therapy by the isolation or design, synthesis, and introduction of new genes into defective cells of particular organs.

Tatum speculated that because the basis of cancer is altered genes, "treatment could be achieved by modification and regulation of gene activities, or by means of gene repair or replacement." Indirect or ex vivo approaches toward gene therapy were also envisioned:

• • • Hence, it can be suggested that the first successful genetic engineering will be done with the patient's o w n cells, for example, liver cells, grown in cuhure. The desired new gene will be introduced, by directed mutation, from normal cells of another donor by transduction or by direct D N A transfer. The rare cell with the desired change will then be selected, grown into a mass culture, and re-implanted in the patient's liver. The efficiency of this process and its potentialities m a y be considerably improved by the synthesis of the desired gene according to the specifications ofthe genetic code and ofthe enzyme it determines, by in vitro enzymatic replication of this D N A , and by increasing the effectiveness of D N A uptake and integration by the recipient cells, as w e leam more about the factors and conditions affecting these processes.

Tatum was extremely confident that gene therapy would bevfeasible, with knowledge of the stracture and function of genes in hand.

One of us speculated about the possibility of gene therapy in an October 24, 1962 letter to Stanfield Rogers (Lederberg, personal communication):

• • • . it will only be a matter of time, and perhaps not a long time, before polynucleotide sequences can be grafted by chemical procedures onto a viras D N A .

These ideas were first published in a 1968 article (Lederberg, 1968b) that states,

• an attempt could then be made to ti-ansform liver cells of male offspring of haemophiliac ancestry by the introduction of carefully fractionated D N A carrying the normal alleles of the mutant haemophilia gene. This experiment would appear to be entirely analogous to the typical attempts at transforming bacterial forms. However, it is not clear whether one should regard this as a pure example of genetic engineering, since the practical outcome would probably be best achieved by influencing the nuclear constitution of somatic tissues rather than by direct tackling of the germ line. The precedent for this type of intervention would be the viras-mediated transduction of genetic characteristics that was also demonstrated in bacteria almost twenty years ago. The proposal, recently revived by Dr. S. Rogers, would require the discovery or artificial formation of cryptic virases to which specified genetic information relevant to the cure of a genetic disease has been grafted. These virases would then carry that information into the requisite cells of the host. Once the essential techniques for grafting segments of D N A from different sources onto that of a microbe have been perfected, experiments along these lines provide the most favourable opportunity to select those segments of D N A information which are needed. In this way it should not be extraordinarily difficult to obtain microbial D N A packets which are enriched with the gene, for example, for the synthesis of phenylalanine hydroxylase. One may of course argue that similar results could be achieved by the manipulation of tissue cells in culture as if they themselves were micro-organisms.

Arthur Komberg's successful replication of DNA in a test tube was widely reported in the popular press as the "creation of life in a test tube" and was viewed as an important milestone on the road to gene therapy (Lederberg, 1968).

Waclaw Szybalski (Szybalski, 1991, 1992), w h o performed one of the earliest mammalian gene transfer experiments (see below), stated at a presentation to the Poultry Breeder's Round Table,

• • • When presenting our data at seminars and symposia in 1962-1964, w e coined the terms 'gene surgery' and 'gene therapy' to stress the clinical potential of our work, but there was little interest in our results (except among poultry breeders), probably because at that time prokaryotes, D N A synthesis, and the genetic code was the center of attention.

By the late 1960s and early 1970s, gene therapy became the subject of an increasing number of articles and meetings. Sinsheimer raminated on the prospect for "designed genetic change" of mankind (Sinsheimer, 1969). At an autumn 1969 meeting, Aposhian (1970) advocated use of pseudovirases derived from the mouse polyoma viras and placed gene therapy in the pharmaceutical tradition.

• • • If one considers the purpose of a drag to be to restore the normal function of some particular process in the body, then D N A would be considered to be the ultimate drag.

In a 1970 Science article, B. Davis discussed human genetic engineering and explored the feasibility and ethics of somatic and germ cell alterations, cloning of humans, genetic modification of behavior, predetermination of sex, and selective reproduction (Davis, 1970). One of his major points was that "control of polygenic behavioral traits is much less likely than cure

of monogenic diseases." In 1971, a symposium on gene therapy was sponsored by the National Institute of Neurologic Disease and Stroke at the National Institutes of Health (NIH) and the Fogarty Intemational Center (Freese, 1971). The first session, entitled "Information Transfer by Mammalian Virases" included talks on recombinant SV40 virases by David Jackson and Paul Berg, and on R N A tumor virases by Howard Temin. Other sessions were entitled "Isolation of Altered Virases with Specific Genes," "Information Transfer by D N A , " "Mammalian Cellular Systems," and "Immunologic and Medical Aspects." Several other articles and meetings brought gene therapy more into the mainstream as well (Friedmann and Roblin, 1972; Morrow, 1976; Anderson, 1984).

Surely, there were other visionaries that an accounting of this early period could eventually reference. It is apparent that schemes for gene therapy occurted to many researchers, once the basics of molecular genetics were established. Despite these premonitions, Friedmann (1990) observed that

• • • It has not always been quite so obvious as it is now that gene therapy is a rational and logically consistent approach to the treatment of some forms of human disease, from both the medical and scientific perspectives. Until fairly recently, the concept of gene therapy has been criticized by a sizable portion of the molecular biologic community as being remote and even improbable, possibly even unnecessary.

In addition, several prominent scientists rejected the reductive view of D N A ' s central role in biology and therefore its implications for gene therapy (Bumet, 1971).

GENETIC MODIFICATION OF MAMMALIAN CELLS IN CULTURE

Several studies in the late 1950s and early 1960s revealed that cultured cells could take up radioactive D N A (Gartler, 1959, 1960; Sirotnak and Hutchinson, 1959; Azrin, 1961; Borenfreund and Bendich, I961;Kay, 1961; Mathias and Fischer, 1962; Schimizu et al., 1962; Rabotti, 1963; Hill and Huppert, 1970). Entry of the radioactive D N A into the nucleus of the cells was also reported. The ability for polynucleotide transfer to induce an effect was provided by viral studies. In the late 1950s and early 1960s, it was demonstrated that naked viral D N A or R N A is infective when applied to cells. This was first disclosed with plants and tobacco mosaic viras in 1956 and then reproduced with poliomyelitis, Semliki Forest encephalitis, influenza, and several other virases in mammalian cells. This prompted a Science review article that discussed the infectious disease implications of these studies (e.g., "their release from infected tissues and resistance to antibodies may explain some anomalous conditions") (Herriott, 1961). The infectious entity presumed to be polynucleotides was obtained by phenol extraction of the viras, was labile to nucleases, and was not neutralized by antibodies. For example, a phenol extract of polioviras yields R N A that produces plaque-forming polioviras when injected into embryonated eggs (Mountain and Alexander, 1959) or when applied to monkey kidney tissue culture (Klingler et al., 1959; Dubes and Klingler, 1961) mouse embryo cells (Weil, 1961), or HeLa cells (Alexander ef a/., 1958).

In studies of the effect of viral R N A concentration, solution composition, and temperature on infections of mengoviras encephalomyelitis in mouse fibroblasts (Colter et al., 1957; Colter and Ellem, 1961), hypertonic saline and sucrose solutions were found to increase the infectivity of the R N A . Exposure to hypertonic solutions increased the number of plaques in HeLa cells formed from polioviral R N A (Koch, 1960). Polioviral R N A uptake was also enhanced by high concentrations of magnesium sulfate (2 M ) (Holland, 1960). Dubes and Klinger (1961) reported higher efficiency of polioviral R N A plaque formation with the use of calcium depleted cells and "poorly soluble substances" such as C a H P 0 4 • 2 H 2 O and CrjOj, AI2O3, CaCOj, CaS04, Fe203, MgFj. A footnote reported that:

• • • Infection is also facilitated by the fine cloudy precipitate, very probably a calcium phosphate, formed when phosphate-buffered saline is made by mixing its ingredients before sufficient dilution with water.

Uptake of cellular or viral polynucleotides could also be improved by complexing with various proteins. A m o s found that the uptake of radiolabeled Escherichia coli R N A by cultured chick cells was enhanced by protamine (Amos, 1961). Other polycations, such as streptomycin, spermine, and spermidine, did not increase uptake but were observed to protect the R N A from RNase, causing precipitates to form in some formulations (Amos, 1961). Smull and Ludwig, in 1962, observed that calf thymus histone or protamine enhanced the infectivity of polioviral and Coxsackie B3 R N A in HeLa cells but did not protect the R N A from RNase digestion (Smull and Ludwig, 1962). Methylated albumin (a basic protein) protected D N A and R N A from nucleases and enhanced their uptake by HeLa cells (Cocito et al., 1962). Bensch and King's observations that L cells did not take up appreciable amounts of D N A but did phagocytose particles prompted them to incorporate D N A into particles (Bensch and King, 1961). Using the Feulgen stain, acridine orange, and radioactive D N A , they found that D N A complexed with charcoal or activated resin did not increase DNA's uptake. However, D N A incorporated into 0.5- to 50- p.m-sized gelatin particles entered the cytoplasm and nucleus of the L cells. Much later in 1975, Farber and co-workers concluded that cultured Chinese hamster lung cells took up more radiolabeled genomic D N A when complexed with polyomithine than with DEAE-dextran, 125 m M CaClj, latex spheres, spermine, polylysine, and polyarginine (Farber et al., 1975).

A potpourri of studies in the 1960s asserted changes in cellular phenotype by the transfer of nonviral genes. Bone martow cells in culture from a patient homozygous for sickle cell disease expressed the normal (J-globin polypeptide (per electrophoresis) when the cells were exposed to D N A from normal bone martOW cells (Azrin, 1961). The transfer of ribonucleoprotein (resistant to DNase but sensitive to RNase and trypsin digestion) from a normal bone marrow or spleen caused the expression of normal hemoglobin in sickle cell bone martow and reticulocytes as determined by electrophoresis, column chromatography, and peptide digests (Weisberger, 1962). Genetic susceptibility to mouse hepatitis viras infection was transferted between macrophages of two different mouse strains via a DNase-sensitive substance (Kantoch and Bang, 1962). The karyotype of chicken cells was modified by exposing them to cow D N A (Frederic and Corin-Frederic, 1962). Another study observed increased survival of irtadiated L cells if they were exposed to D N A from nonirradiated L cells (Djordjevic et al., 1962). Several negative results conceming D N A transfer were mentioned as well in the early 1960s. Mathias and Fisher (1962) attempted to transfer amethopterin resistance in mouse leukemic cells without success. The exposure of donor bone martow cells to naked and gelatin-complexed D N A from the host bone marrow cells in isotonic and hypertonic solutions did not affect the success of bone marrow transplantation in mice (Floersheim, 1962). The premise was that transfer of histocompatibility genes would induce immunotolerance.

The establishment of cell lines containing defined enzymatic defects and selectable systems proved extremely critical, ushering in the modem era of gene transfer. W . and E. Szybalski developed hypoxanthine-guanine phosphoribosyl transferase (HPRT)-deficient cell lines and the H A T selection media. D N A isolated from HPRT"^ cells was able to confer HAT-resistance to H P R T " cells. The D N A was transferred in a phosphate buffer containing spermine to bind the D N A and shield it from DNase activity (Szybalska and Szybalski, 1962). In Table 1 of their 1962 article, the Szybalskis showed a dose-dependent relationship between the amount of donor D N A and the number of transformants. Subsequent experiments indicated that the spermine contained 3 0 % CaClj; and that it could be replaced by CaClj when used with phosphate buffers. Precipitates were observed during these experiments. Later experiments with CaCl2 were reported in Table 2 of their paper in Proceedings of the 12th Annual Session National Poultry Breeder's Roundtable (Szybalski, 1963) in 1963:

• • • It was noticed diat only one of preparations of spermine was especially active in the transformation process. Addition of this spermine solution to the phosphate-buffered saline (PBS) resulted in clouding of the solution, both in the presence or absence of the transforming D N A and the cells. Since we found that the particular spermine preparation contained a high concentration of calcium, the precipitate was most probably the calcium phosphate.

Further chemical analysis indicated that the contaminant was in fact calcium (Szybalski, 1992). This finding was not widely disseminated because of the specialized nature of the publication in which it was reported, but Szybalski's eariy contribution to DNA-mediated gene transfer was recognized by Scangos and Ruddle in a 1981 review. Also, in 1962, Bradley, Roosa, and Law (1962) used 8-azaguanine selection to demonstrate the cellular uptake of naked, genomic D N A.

Notwithstanding these prior observations conceming calcium phosphate precipitation, most workers used DEAE-dextran to transfer foreign D N A into mammalian cells as a result of a report by Vaheri and Pagano in 1965 that showed increased transfer of polioviras R N A with DEAE-dextran (Vaheri and Pagano, 1965). This 1965 study compared the use of DEAEdextran to that of hypertonic magnesium solutions. Burnett and Hartington (1968) also used DEAE-dextran to transfer polyoma viras D N A but indicated that this was not successful with adenoviral D N A . Subsequently, several other studies promulgated the successful use of DEAE-dextran. McCutchan and Pagano (1968) transferred SV40 D N A , Warden and Thome (1968) transferted polyoma viras, and Nicolson and McAllister (1972) transferted adenoviras one. Hill and Hillova (1972) produced infectious R S V after noninfected cells were transfected using DEAE-dextran with D N A from R S V infected cells.

It was not until the detailed study of Graham and Van Der Eb (1972) on calcium phosphate-mediated transfection that this technique became widely used and accepted. They systematically investigated the use of calcium or magnesium ions for transfection and determined that co-precipitates of D N A , calcium, and phosphate were necessary for efficient transfection. They also varied several parameters, such as p H (6.9-7.4), incubation times, confluency of cells (60-90%), and adenoviral and carrier D N A concentration, in a systematic fashion. This and the reproducible, 50- to 100-fold increase in efficiency over DEAE-dextran had a major impact on the field and is widely cited as the primary source for the calcium phosphate transfection technique. After their initial publication in 1973, they modified the procedure further in 1974 (Graham et al., 1974).

DIRECT GENE TRANSFER INTO ANIMALS

Perhaps the earliest predecessor of a direct in vivo approach was the use of vaccines, which permanentiy modify the body's response to infection. Vaccination with attenuated virases may be viewed as a form of gene therapy, especially since the viral genomes may persist long term. The ease of administration, relative cheapness, and long-lasting effect of vaccines are ideal qualities to which proponents of direct gene therapy aspire.

Another notion for direct, in vivo therapy was treating bacterial infections by the injection of bacteriophages. Although, this therapeutic approach was discussed in Sinclair Lewis' novel Arrowsmith, there were several actual reports of its successful use in animals and humans (d'Herelle, 1926). The negative results of well-controlled studies (Boyd and Portnoy, 1944) and the ascent of antibiotics stopped further investigation. Interestingly, there have been some recent reports of its exploration in animals (Reynaud etal., 1992; Soothill, 1992).

The direct transfer of polynucleotides into tissues in vivo and in situ was attempted in the 1960s as well. Peritoneal malignant and normal cells in the peritoneum (Rieke, 1962) and tumors in situ (Rabotti, 1963) took up radioactive D N A , but the foreign D N A demonstrated no functional activity. Radioactive D N A injected intravenously or intraperitoneally in rodents were taken up by spleen and bone marrow cells (Hudnik-Plevnik et al., 1959; Hill, 1961).

J. Benoit et al. published beginning in 1956 that Pekin ducklings injected intraperitoneally with D N A extract from Khaki Campbell ducks exhibited characteristics of the Khaki Campbell ducks in terms of body and head size, and that these effects were passed onto their progeny (Benoit et al., 1960a,b). Much to the dismay of pate manufacturers and Chinese chefs who were expecting a culinary breakthrough, these results in ducks have never been reproduced.

The Benoit studies in ducks attracted enough attention to prompt several other investigators to attempt phenotypic modification by D N A transfer in other species. The injection of rat D N A from a pigmented rat into an albino rat did not produce any change in skin color (Perry and Walker, 1958). T w o other studies also indicated the inability to produce pigmentary studies in albino rodents by injection of D N A from a pigmented rodent (Beam and Kirby, 1961; Holoubek and Hnilica, 1961). In addition, the intraperitoneal injection of D N A from a normal rat did not cortect the hyperbilirabinemic state of the C N H strain (noted to be deficient in glucuronyl transferase) (Perry and Walker, 1958). In chickens, the injection of a Tyrode solution or D N A in the Tyrode solution into the bloodstream of byoung chicken embryos did not cause any morphological changes but did cause teratogenic malformations (Martinovitch et al., 1962). In mice, another group observed that the intraperitoneal injection of D N A from breast cancers of agouti C 3 H mice but not D N A from other organs caused cytological changes in the livers of white mice (Leuchtenberger et al., 1958). T wo different groups noted that injection of D N A from one mouse strain caused weak transplantation immunity against skin grafts but raised the possibility that contaminants may have been responsible for the effect (Haskova and Hrabesova, 1958; Medawar, 1958).

In summary, the Benoit studies prompted many attempts to research D N A uptake by vertebrate cells. However, the study made the entire field of gene transfer into cells of higher organisms somewhat suspect (W. Szybalski, personal communication).

Other studies explored the ability of D N A to transfer the neoplastic state. In the early 1950s, it was reported that new tumors formed after injection of D N A from tumor cells into normal mouse tissues (Stasney et al., 1950; Paschkis et al., 1955). One-third of rats injected subcutaneously with lymphosarcoma chromatin developed lymphosarcomas or leukemia, whereas one-third of rats injected intrahepatically developed hepatomas. A subsequent article observed a similar phenomenon but concluded that it was the result of contaminating cancer cells (Klein, 1952). The repeated, subcutaneous injections of herring sperm D N A caused duodenal adenocarcinoma in two mice but this was not repeatable with a different batch of D N A (Meek and Hewer, 1959). A subsequent study concluded that repeated, subcutaneous injections of herring sperm D N A caused intestinal carcinomas in cichlids (Stolk, 1960). The injection of Drosophila melanogaster D N A was mutagenic in Drosophila (Fahmy and Fahmy, 1961). Only many years later was the phenotype for neoplastic transformation reliably transferted from mammalian D N A into cells in culture (Shih et al., 1979; Cooper et al., 1980). Nonetheless, many studies in the late 1950s and early 1960s noted neoplastic transformation by viral polynucleotides in vivo, just as there was a flurry of reports at this time concerning the infectivity of viral polynucleotides. For example, phenol extracts of S E polyoma viras were able to cause infections and tumors in hamsters (DiMayorca et al., 1959). Also, phenol extracts of papillomatous tissue (Shope) of cottontail rabbits produced papillomas when injected into the skin of rabbits (Ito, 1960, 1961).

The maturation of plasmid expression vectors, reporter genes, and better in situ detection systems prompted more recent attempts at direct, in vivo gene transfer. In 1983, large liposomes containing the rat preproinsulin gene within a plasmid were injected intravenously into rats (Nicolau et al., 1983). The injections caused a — 3 0 % decrease in blood glucose and a — 5 0 % increase in blood insulin. In addition, calcium phosphate- precipitated polyoma viral D N A was injected into mouse liver and spleen along with hyaluronidase and collagenase (Dubensky et al., 1984). The investigators found polyoma D N A in the tissues and inferred that the viral D N A had to replicate. Similar studies were also done with polyoma D N A and proteoliposomes (Mannino and Gould-Fogerite, 1988). Calcium phosphate-precipitated plasmids containing chloramphenicol acetyltransferase (CAT), hepatitis B surface antigen, human growth hormone, or mouse preproinsulin genes were also injected intraperitoneally (Benvenisty and Reshef, 1986). The investigators observed some C A T activity, immunohistochemical staining for the hepatitis antigen, insulin R N A , and growth hormone R N A in livers injected with the respective plasmids. Substantial amounts of foreign gene expression has been observed in muscle injected with plasmid D N A (Wolff et al., 1990).

EARLY EXPLOITS AT HUMAN GENE THERAPY

Two early (perhaps too early) attempts at human gene therapy are briefly summarized. One represents a direct, in vivo, viral approach, while the other represents an indirect, ex vivo approach involving cell transplantation.

In the late 1960s, Stanfield Rogers injected the Shope papilloma viras into patients with arginase deficiency, based upon his studies that indicated that the viras contained an arginase gene. His initial observation was that rabbit skin tumors induced by the Shope papilloma viras contained high levels of arginase activity (Rogers, 1959, 1963). Because he did not find any arginase activity in normal rabbit skin, he concluded that the viras carried an arginase gene. H e also reported that the viras induced a viras-specific arginase in fibroblasts from a patient with arginase deficiency (Rogers et al., 1973). A biochemical assay demonstrated increased arginase activity, and in immunohistochemical stain with antisera specific against the viras-specific arginase distinguished the viras-induced arginase from native arginase. Administration of the viras to animals caused no harmful effect and reduced blood arginase levels. However, other workers found arginase activity in normal skin (Rothberg and van Scott, 1961; Orth et al., 1967). Rabbit liver arginase and Shope had similar kinetic and antigenic properties and papillomas induced by a carcinogen also contained arginase. They concluded that the Shope viras either induces arginase expression or leads to the preferential growth of cells with higher arginase activity. The final outcome to this controversy was that three siblings with arginase deficiency were injected with the Shope virus, without any effect on their arginine levels (Terheggenera/., 1975).

In a 1980 Nature report, M . Cline reported that D N A from a highly methotrexate-resistant Swiss 3T6 cell line (containing many copies of the D H F R gene) was transfected with calcium phosphate into mouse bone marrow cells (Cline et al., 1980). The donor bone martow cells had a different karyotype to distinguish them from the recipient bone marrow cells. The recipient mice were irtadiated and treated with methotrexate before being injected with the transfected cells. After transplantation, the recipient animals were reported to have an increased percentage of marrow cells with the donor karyotype and increased D H F R enzymatic activity. A similar study was published in Science that used recombinant D N A containing the herpes thymidine kinase (TK) gene in the pBR322 plasmid vector (Mercola etal., 1980). O n the basis of this experimental data, Cline et al. attempted to use the calcium phosphate method to transfect the p-globin gene into human bone martow cells, which were then transplanted into patients with thalassemia. Their clinical trial was criticized for both scientific and procedural reasons, and this led to Cline's censure by the NIH and by his university (Wade, 1980, 198Ia,b). A n indirect result was the N I H decision that all future human gene therapy tiials would have to be approved by the N I H Recombinant D N A Advisory Committee (RAC).

Unsound practices in these early studies in cell culture, animals, and humans made both the experimental results and the entire approach of gene therapy seem suspect, even though some of the basic concepts and approaches upon which the studies were based were eventually proven correct.

THE ADVENT OF RECOMBINANT DNA TECHNOLOGY

It was not until early transfection techniques and selection systems for cultured cells were combined with recombinant D N A technology that major progress was made in gene transfer. The isolation of a single gene enabled both greater efficiency and better documentation of its transfer.

Using UV-irtadiated herpes simplex viras, Munyon showed in 1971 that the T K gene from herpes simplex could rescue T K " cells in H A T media (Munyon et al., 1971). Later in the 1970s, several groups used total herpes simplex D N A and calcium phosphate transfection to transfer the herpes T K gene into T K " human cells (Bacchetti and Gaham, 1977; Maitland and McDougall, 1977; Wigler et al., 1977; Minson et al., 1978). Fragments of herpes simplex D N A generated by shearing or restriction enzyme digestion were calcium phosphate transfected into mouse T K " cells. Wigler et al. then provided definitive proof that the transformation occurted by transfer of the herpes T K gene (Wigler et al., 1978). Transformed TK"^ cell lines were shown to contain herpes T K activity by isoelectric focusing electrophoresis and the herpes T K gene by Southem blot analysis. Transfection oftotal cellular D N A from the transformed TK"^ cell line transferted T K activity to a TK-deficient cell.

Subsequently, other genes such as A P R T and human H P RT were transferted (Graf et al., 1979; Wigler et al., 1979; Willecke et al., 1979; Pellicer et al., 1980). A n advance was the transfer and selection for unlinked genes (Wigler et al., 1979). These studies demonstrated that any gene can be transferted into mammalian cells along with a selectable marker. By 1981, Scangos and Ruddle (1981) concluded:

• Thus, in the last two decades the field of D N A mediated gene transfer ( D M G T ) has progressed from relatively simple experiments in which the H A T selective system was developed and the feasibility of the technique was tested, through the development and refinement of the calciumphosphate technique, to experiments in which many selectable and non-selectable genes have been transferted into mammalian cells.

THE GENESIS OF VIRAL VECTORS

One of the earliest attempts to produce a viral vector was that by Rogers and Pfuderer (1968). These investigators enzymatically added a poly(A) sequence to tobacco mosaic viras R N A and reported that plants infected with this modified viras contained increased amounts of polylysine. Other workers experimented with purifying polyoma viral capsid proteins to form pseudovirions (Friedmann, 1971; Aposhian era/., 1972). In a series of studies. Berg and colleagues developed the first recombinant viral vector system based upon the papilloma simian viras (SV40). In 1972, X phage D N A and the E. coU galactose operon was ligated into D N A of S V 4 0 D N A (Jackson etal., 1972). In 1976, recombinant SV40 vectors with \ phage D N A were propagated in cultured monkey kidney cells (Goff and Berg, 1976). Hamer et al. similarly constracted an SV40 viras carrying an E. coli suppressor gene (Hamer et al., 1977). After the full-length double-stranded c D N A for rabbit globin m R N A was synthesized in vitro using reverse transcriptase and D N A polymerase I (Maniatis et al., 1976), it was exchanged with the major capsid protein, VPl of the SV40 viras. The recombinant vector expressed the rabbit P-globin protein in the infected cells (Mulligan et al., 1979). Other workers also found expression of P-globin protein from cells infected with SV40 recombinant virases carrying the P-globin gene (Hamer and Leder, 1979; Hamer e? a/., 1979; Mulligan era/., 1979).

The development of retroviral vectors independently by three different groups (Shimotohno and Temin, 1981; Tabin et al., 1982; Wei et al., 1981) has been reviewed previously (Miller and Rosman, 1989). These activities in viral vectors culminated in a meeting at the Banbury Center of the Cold Spring Harbor Laboratory in 1982 that generated much enthusiasm and interest in conducting further research toward human gene therapy (Cold Spring Harbor Laboratory, 1983). Helper-free packaging cell lines were subsequentiy constracted (reviewed in Miller, 1989). After several disease-related genes were transferted into various cells in culture, the possibility of efficient gene transfer into mammalian cells for the purpose of gene therapy became widely accepted.

CONCLUSION

Although the discovery of the central dogma of molecular genetics quickly led to the idea for gene therapy, advancement was hindered initially by several poorly designed studies. As the field has gained credibility in recent years, however, progress has accelerated. Central to this progress have been the discovery of basic genetic concepts in bacteria and bacteriophages and the elaboration of these concepts to mammalian cells, recombinant D N A technology, and mammalian gene transfer techniques, including viral vectors and physical-chemical methods.

The new field of gene therapy combines the advantages of pharmacology (namely, the ability to treat human disease with externally administered substances that have specific actions) and surgery (namely, the ability to alter a tissue or organ permanently). As such, gene therapy represents more than an extension of established medical practice; rather, it is an entirely new branch of medicine, one that could potentially revolutionize the way w e treat human disease.

ACKNOWLEDGMENTS

We thank Kirk Hogan, Jim Crow, and Waclaw Szybalski for helpful suggestions and Tim Lockie for help in preparing the manuscript. This article is derived from a book chapter in Gene Therapeutics (editor, Jon Wolff, Birkhauser, 1993).

This article is dedicated to the memory of Howard M . Temin whose work led us back to the success of gene therapy.

REFERENCES