2024 年 100 巻 1 号 p. 1-14
In the late 1970s, crude interferon samples were found to exhibit anti-tumour activity. This discovery led to the interferon as a “magic drug” for cancer patients. Many groups, including those in Tokyo, Zürich, and San Francisco, attempted to identify human interferon cDNAs. Tadatsugu Taniguchi was the first to announce the cloning of human interferon-β cDNA in the December 1979 issue of Proc. Jpn. Acad. Ser. B. This was followed by the cloning of human interferon-α by a Zürich group and interferon-γ by a group in Genentech in San Francisco. Recombinant interferon proteins were produced on a large scale, and interferon-α was widely used to treat C-type hepatitis patients. The biological functions of interferons were quickly elucidated with the purified recombinant interferons. The molecular mechanisms underlying virus-induced interferon gene expression were also examined using cloned chromosomal genes. The background that led to interferon gene cloning and its impact on cytokine gene hunting is described herein.
About 70 years ago, a research group in Tokyo (Yasushi Nagano and Yasuhiko Kojima, The Institute of Infectious Disease, Tokyo University, renamed The Institute of Medical Science) found that tissue suspensions prepared from vaccinia-infected rabbit skin or testes contained substances (Facteur inhibiteur, inhibitory factor, or IF) that conferred the virus-resistance to the skin in vivo. Since Nagano, who had spent three years (1936–1939) at Pasteur Institute, was a Francophile, he published these findings in French in a Journal of the French Biological Society.1) Another group (Alick Isaacs and Jean Lindenmann, The National Institute for Medical Research, London) incubated chorioallantoic chicken membranes with heat-inactivated influenza virus in vitro. After 20–24 h, they collected culture supernatants and added them to fresh membrane fractions. They found that chorioallantoic membranes acquired resistance to the virus in a time-dependent manner.2),3) This inhibitory activity was species-specific; namely, the molecule produced by chicken cells exhibited higher activity in chicken cells than in human or mouse cells. It inhibited infections not only by the influenza virus but also by other viruses. Lindenmann, who had an affinity for physics, named the inhibitory molecule in the supernatant “interferon” based on terms used in physics (such as electrons, neutrons, and protons).
Many research groups then started to work on interferon and found that human lymphoid and fibroblast cells produced antigenically different interferons.4) They were called leukocyte and fibroblast interferons, respectively5) (Table 1). Mitogen-stimulated lymphoid cells were also found to produce interferon,6) which was antigenically different from leukocyte and fibroblast interferons and named immune interferon.5) Leukocyte and fibroblast interferons were resistant to treatments with pH 2 and SDS, and were designated as Type I interferons. In contrast, immune interferon was unstable under these conditions and, thus, was classified as a Type II interferon.5) The specific activities (the amount that reduces the infectivity of a virus to 50%) of interferons were estimated to be 108–109 units/mg.7)–9)
Name | Original name |
Type | Producer | Chr. | Biochemical properties |
Receptor |
---|---|---|---|---|---|---|
IFNα | Leukocyte IFN |
Type I | Virus-infected leukocyte |
9p21.3 intron-less |
Stable at pH 2.0 and SDS |
IFNAR1/ IFNAR2 |
IFNβ | Fibroblast IFN |
Type I | Virus-infected fibroblasts |
9p21.3 intron-less |
Stable at pH 2.0 and SDS |
IFNAR1/ IFNAR2 |
IFNγ | Immune IFN | Type II | Mitogen-stimulated T cells |
12q15 3 introns |
Unstable at pH 2.0 and SDS |
IFNGR1/ IFNGR2 |
After confirming in vivo anti-virus activity with mice,10) several research groups reported that interferons prevented the infection of hepatitis virus and herpes zoster virus in patients.11),12) Promising findings were also obtained for some cancer types,13) with interferons being referred to as a “magic drug” in the press (for example, March 31, 1980, TIME magazine: January 28, 1980, Newsweek). The interferons used in patients were obtained from the culture supernatants of Sendai virus-treated human leukocytes14) or poly(I)(C)-treated human skin fibroblasts.15) These samples were partially purified for clinical purposes. Nevertheless, based on their specific activities, the content of the interferon protein was 0.1%. Therefore, the possibility that the proposed effects of interferons, namely, anti-tumour, growth inhibitory, and immune suppressive activities, were due to other components could not be ruled out. In addition, interferon treatments were associated with substantial side effects, such as fever, thrombocytopenia, granulocytopenia, fatigue, and joint pain. These side effects were also attributed to impurities in the samples.
Life sciences were revolutionised in the 1970s. Biochemical analyses of phage DNA replication in Escherichia coli identified many enzymes (including DNA polymerase, terminal deoxynucleotidyl transferase, DNA ligase, and exonuclease) that synthesise DNA.16) These enzymes and restriction enzymes17) allowed to produce hybrid DNA composed of DNA fragments from different origins.18) Hybrid DNA was soon constructed with plasmid DNA, an extrachromosomal genetic element19) carrying a drug-resistant gene and replication origin. Following the transfection of chimeric DNA into Ca2+-treated E. coli,20) the introduced hybrid DNA was precisely amplified in the new host.21) This technique was soon applied to amplify specific DNA fragments from eukaryotes (Xenopus ribosomal DNA).22) Plasmids with different insert DNAs did not replicate in the same host E. coli due to their incompatibility, indicating that each colony grown in an agar plate carried separate inserts (clones).23)
Using reverse transcriptase (initially identified as RNA-dependent DNA polymerase) found in retrovirus particles,24),25) a method to prepare double-stranded cDNA (complementary DNA) from mRNA was developed.26) Since the pure mRNAs for rabbit globin, rabbit ovalbumin, and mouse immunoglobulin were easily obtained from reticulocytes, oviduct, and myeloma, their cDNAs were produced, connected to a plasmid, and amplified in E. coli.27)–29) At the same time, two groups developed a method to elucidate DNA sequences,30),31) and their protocol was quickly distributed worldwide.
After four years at New York University at the laboratory of Severo Ochoa, the 1959 Nobel Laureate, Charles Weissmann returned to Zürich in 1967 as the director of the Institute of Molecular Biology I, University of Zürich, Switzerland. His group was working on Qβ phage and the expression of the mouse β-globin gene. At the Gordon Conference in the summer of 1977, he met Peter Lengyel, a Professor at Yale University who had been a graduate student in Ochoa’s laboratory at the same time as Weissmann. Lengyel was working on mouse interferons at that time and discussed the interferon system, its physiological functions, and gene induction system with Weissmann.32),33) Weissmann decided to collaborate with Lengyel to clone the mouse interferon cDNA. They attempted to prepare mRNAs from virus-induced mouse cells and assay their interferon coding capacities by translating them in Xenopus oocytes. This project was challenging because there was no purified interferon or reliable specific antibodies against interferons. Tadatsugu Taniguchi, who had obtained a Ph.D. for his work on Qβ phage34),35) in Weissmann’s laboratory, started to work on this project. In March 1978, Weissmann founded the Biotech company Biogen with several renowned scientists, including two Nobel Laureates, Water Gilbert and Philip Sharp, and started the project on human interferon. He began collaborating with Kari Cantell, Helsinki, who produced interferons by infecting human leukocytes with the Sendai virus. Peter Curtis, who had been working on mouse globin precursor mRNAs in Weissmann’s laboratory, went to Helsinki to prepare RNAs from leukocytes stimulated by the Sendai virus at various intervals. At Zürich, he established a microinjection system into Xenopus oocytes and fractionated mRNAs using a sucrose density gradient to enrich interferon mRNAs. Curtis left Zürich in the middle of 1978 to work at the Wistar Institute, Philadelphia, and Taniguchi took over the project on human interferon. However, Taniguchi was soon after recruited by Prof. Masami Muramatsu to the Cancer Institute, Tokyo, to clone human interferon. Weissmann and Taniguchi agreed that a group in Zürich worked on the “leukocyte” interferon, while Taniguchi focused on the “fibroblast” interferon in Tokyo. Thus, I, Nagata, who had joined Weissmann’s group in November 1977 and worked on Qβ phage, was engaged to work on cloning the human “leukocyte” interferon from December 1978.
Figure 1 shows the procedure used to clone human interferon cDNAs. Regarding the leukocyte interferon, buffy coat cells from human blood were stimulated for 5 h with the Sendai virus. In contrast, Taniguchi stimulated the human skin fibroblast cell line DIP2 with Poly(I)(C) in the presence of cycloheximide to induce interferon genes. mRNAs were size-selected using a sucrose density gradient, and mRNAs in the 12S region were converted to dsDNAs. dsDNAs were inserted into the Pst I (Zürich) or Eco R1 (Tokyo) site of pBR322 via G-C or T-A tailing, respectively. Hybrid plasmids were then introduced into E. coli K12 χ1776 (EK2) in the P3 laboratory to produce 20,000 tetracycline-resistant (Zürich) or 3,600 ampicillin-resistant clones (Tokyo). Taniguchi first selected the clones carrying the virus-induced genes by hybridizing the clones with 32P-labeled cDNA enriched for the induced cDNA.36) The method to identify the clone carrying interferon cDNA was the “hybrid-selection” method designed by Weissmann (Fig. 2). Plasmid DNA was cleaved with Hind III, heat-denatured, and attached to a nitrocellulose filter to trap denatured single-stranded DNA. The filter was then hybridised with mRNA from interferon-producing cells. Interferon mRNA was expected to hybridise with the interferon cDNA trapped in nitrocellulose. mRNA was eluted from the filter by heating and then injected into Xenopus oocytes to assess its interferon-producing activity.
Preparation of the cDNA library. (A) Preparation of double-stranded (ds) DNA. The mRNA from poly(I)(C)-treated fibroblasts or Sendai virus-infected leukocytes was fractionated using a sucrose-density gradient. RNA at the 12S fraction containing IFN mRNA activity was reverse-transcribed to produce complementary DNA. mRNA was degraded by treatment with alkali, and the second strand was synthesised with DNA polymerase I. After removing the loop structure with S1 nuclease, dsDNA was tailed with dTMP or dCMP using terminal transferase. (B) Vector preparation. The plasmid pBR322 contains two drug-resistant genes (ampicillin and tetracycline), the replication origin, and several unique restriction enzyme recognition sites. The plasmid was digested with one of the unique restriction sites (Eco RI or Pst I) and tailed with dAMP or dGMP. (C and D) dT- or dC-tailed dsDNA was integrated into dA- or dC-tailed pBR322. The mixture of chimeric DNA was then introduced into E. coli and selected in agar plates containing antibiotics (ampicillin or tetracycline). In each colony, E. coli was expected to have plasmid DNA carrying different cDNAs; some were expected to be IFN cDNA.
Screening for a clone carrying IFN cDNA. (A) Plasmid DNA from E. coli clones was denatured and covalently attached to a nitrocellulose filter. The filter was hybridised with mRNA prepared from virus-infected or poly(I)(C)-treated cells expected to contain IFN mRNA. IFN cDNA should pick up the IFN mRNA. After washing, mRNA that bound to the filter was eluted. (B) mRNA was then micro-injected into Xenopus oocytes and incubated at room temperature overnight. IFN was expected to be secreted from oocytes into the medium. (C) The culture medium was then subjected to an interferon assay with human CCL23 cells and the Mengo virus. In Lane 3, 100 units of the IFN sample was serially diluted 3-fold from the top well and incubated with cells for 24 h. Cells were then challenged with the Mengo virus for 24 h and stained by Crystal violet. Lane 1, cells alone: Lane 2, cells not treated with IFN were challenged with the virus.
Taniguchi obtained a positive clone and reported his findings in the December 1979 issue of Proc. Jpn. Acad. Ser. B36) (accompanied paper). We in Zürich also had some positive clones in the hybrid-selection assay.37) However, we wanted to obtain more convincing data to support interferon cDNA being carried by these clones. At that time, Weissmann advised us to examine interferon activity in the extract of “positive” E. coli clones. He speculated that since cDNA was inserted at the Pst I site of the β-lactamase (ampicillin-resistant) gene, it may be transcribed and translated into protein. Since interferons exhibit very high specific activities, a trace amount of the interferon protein in E. coli may be detectable. We (Alan Hall, Hideharu Taira, Michel Streuli, and myself) followed his advice and detected interferon activity in the S100 extracts of some “positive” clones.37) Similar to authentic human interferons, interferon activity in E. coli was resistant to treatment with pH 2 and active on human cells but not on mouse cells. An anti-human leukocyte antibody neutralised this activity, whereas an anti-fibroblast antibody did not.
Soon after the cloning of cDNAs, their sequences were elucidated by Taniguchi38) and our group.39) The amino acid sequences derived from the DNA sequences were consistent with the N-terminal sequences of purified human interferons,40),41) confirming the authenticity of cDNAs. The amino acid sequences of human “leukocyte” and “fibroblast” interferons were similar, which suggested that the genes for these interferons were derived from the same ancestor.42) The international interferon nomenclature committee decided to call “leukocyte” and “fibroblast” interferons interferon-α (IFNα) and interferon-β (IFNβ), respectively.5) The “immune” interferon was named interferon-γ (IFNγ), the cDNA of which was later cloned by a group in Genentech.43) We soon noted that multiple copies of the intron-less human IFNα gene44)–48) were present on human chromosome 9 together with a single IFNβ gene.49) We now know thirteen active and four pseudogenes for IFNα and one functional IFNβ gene (Hugo Gene Nomenclature Committee: https://www.genenames.org/). Using the chromosomal genes of the IFNα and IFNβ genes, many research groups, including Taniguchi’s and Weissmann’s groups, investigated the mechanisms underlying the virus-induced expression of the IFN gene. Excellent reviews have been published on this subject.50)–52)
IFN cDNAs were engineered to produce their recombinant proteins in E. coli,53)–55) and their in vitro and in vivo biological functions, including anti-virus, cytotoxic, and immune regulatory activities, were confirmed.56),57) The safety of the recombinant IFNα in man was testified by Weissmann who injected 10 µg of the protein into his own vein. The homogeneous IFN samples had strong side effects,58) indicating that they were intrinsic properties of Type I IFN. The same receptor, a heterodimer IFNAR1/IFNAR2, was found to mediate the interferon activities of IFNα and IFNβ.59) On the other hand, IFNγ bound to a different receptor, a heterodimer of IFNGR1/IFNGR2.60),61) Although the molecular mechanisms by which Type I IFN exhibited anti-virus activity was unclear, it was used worldwide to treat patients with hepatitis C (HCV) until 2014. At that time, a more effective specific drug, inhibitors of HCV non-structural protein 5A (NS5A), was developed.58) Although IFN was not a magic drug for cancers,62) it had a strong efficacy against some types of cancers such as hairy cell leukaemia and chronic myeloid leukaemia (CML).63) In addition, the cloning of interferon genes had a huge impact on the cytokine field and many groups started to identify the cDNAs of various cytokines, including lymphokines, colony-stimulating factors, chemokines, and growth factors.64) Their receptors were then identified, and the signal transduction pathway mediated by these cytokine-their receptors was soon elucidated.65)
Contributed by Shigekazu NAGATA, M.J.A.; Edited by Shizuo AKIRA, M.J.A.
Correspondence should be addressed to: S. Nagata, Biochemistry & Immunology, WPI Immunology Frontier Research Center, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan (e-mail: snagata@ifrec.osaka-u.ac.jp).
This paper commemorates the 100th anniversary of this journal and introduces the following paper previously published in this journal. Taniguchi, T., Sakai, M., Fujii-Kuriyama, Y., Muramatsu, M., Kobayashi, S. and Sudo, T. (1979) Construction and identification of a bacterial plasmid containing the human fibroblast interferon gene sequence. Proc. Jpn. Acad. Ser. B 55 (9), 464–469 (https://doi.org/10.2183/pjab.55.464).
Shigekazu Nagata was born in Kanazawa in 1949, graduated from the University of Tokyo, Faculty of Science, and obtained his PhD in 1977 from the University of Tokyo. He did his post-doctoral research at the Institute of Molecular Biology I (Director: Prof. Charles Weissmann), University of Zürich. In 1982, he returned to the Institute of Medical Science, the University of Tokyo as an assistant professor. In 1987, he was appointed a head of the Molecular Biology Department of Osaka Bioscience Institute. In 1995, he joined the Osaka University Medical School as a professor in the Department of Genetics, and in 2007 moved to the Kyoto University Graduate School of Medicine. In 2015, he returned to Osaka University as a specially appointed professor at the Institute of Immunology Frontier Research Center. He received Emil von Behring Prize (Germany), the Robert Koch Award (Germany), Prix Lacassagne (France), Debrecen Award (Hungary), Japan Academy Prize, and Imperial Prize. The Japanese Government has recognised him as a Person of Cultural Merit. He is a member of the Japan Academy and an international member of the National Academy of the United States of America.