introducing viruses and cancer

laura n. hindle & david j. blackbourn

at the spring 2013 sgm conference, viruses and cancer will be emphasised with a symposium entitled ‘viruses and cancer: causes to cures’. this symposium will be complemented by the pre-eminent sgm prize medal lecture, to be given by harald zur hausen. professor zur hausen was awarded the 2008 nobel prize in physiology or medicine ‘for his discovery of human papillomaviruses causing cervical cancer’. in turn, his work led to the development of human papillomavirus (hpv) vaccines that are expected to reduce the global burden of cancer of the cervix.

over one in three people in the uk will develop cancer during their lifetime. yet far fewer of us are aware of the association of viruses with cancer development. globally, the latest figures indicate that of the 12.7 million cases of cancer occurring annually, 16% are attributable to infectious agents. in developing countries, this frequency rises to almost 1 in 4. the majority of these cancer-inducing (oncogenic) infectious agents are viruses, the subject of the present article. rob newton provides more insight into the incidence of virus-associated cancers in the following article.

in considering the global burden of human cancer, these figures are significant for two reasons. first, these cancers are potentially preventable either by vaccinating people against the oncogenic agents, preventing their infection, or by treating the infections before cancer develops. second, the study of oncogenic viruses can and has provided a route to understanding the biology of cancer cells.

discovering cancer

although cancer was less common before the 20th century, it is certainly not a new disease. the paleoanthropologist louis leakey discovered the first known case of homonid cancer in an archaeological jawbone, which was suggestive of burkitt’s lymphoma occurring at least 300,000 years ago (see timeline).

the first contemporary documentation of cancer can be traced back thousands of years to the ancient egyptians in 3000–1500 bc. in several egyptian medical manuscripts, such as the edwin smith papyrus, a description of cancer can be found.

the concept of behavioural and environmental risk factors influencing cancer development dates back hundreds of years. in 1714, bernardino ramazzini noted that nuns developed less cervical cancer (but more breast cancer) compared to the general population. the importance of this observation is reflected in the fact that cervical cancer is now recognised as a sexually transmitted disease, being caused by hpv, a sexually transmitted virus. ramazzini’s work is one of the first great epidemiologic studies and demonstrates the value of epidemiology in helping to identify risk factors in cancer. subsequently, percival pott described in 1755 an occupational cancer of the scrotum in chimney sweeps, caused by their chronic exposure to soot.

in 1838, german pathologist johannes muller demonstrated that cancer is made up of cells, while his student, rudolph virchow (1821–1902), concluded that all cells, including cancer cells, are derived from other cells. these ideas founded the modern cellular theory of cancer.

viruses and cancer timeline. images: skull, digital vision/thinkstock; ramazzini, pott, smith papyrus & gallo, public domain; müller & virchow, us national library of medicine; rous, nypl/science source/science photo library; epstein, prof. sir anthony epstein cbe frs, wolfson college, oxford; zur hausen, armin kübelbeck; chang & moore, i. atherton, sgm; houghton, © university of alberta

discovering tumour viruses

the origins of virology date back to the end of the 19th century, with beijerinck studying tobacco mosaic disease in plants and loeffler and frosch working on the aetiology of foot-and-mouth disease. around a decade later, the theory that viruses might cause certain cancers began to emerge, but it took many years for this idea to become widely accepted. ellerman and bang published observations on the viral transmission of leukaemia in chickens in 1908 and, in 1911, peyton rous reported on sarcoma in chickens being caused by what later became known as rous sarcoma virus (rsv). he found that he could transfer the sarcoma to other chickens by their inoculation with a cell-free filtrate of the tumour. his work was widely discredited at that time, but in the decades that followed more tumour viruses were discovered (see timeline). this accumulating evidence began to change opinions and, in 1966, rous was awarded the nobel prize in physiology or medicine ‘for his discovery of tumour-inducing viruses’, pioneering work he began over 50 years earlier.

the discovery of rsv led to the identification by harold varmus and michael bishop of the src gene that is responsible for rsv-induced tumours. they showed that src was a derivative of a normal cellular gene. this discovery of a cellular gene with cancer-inducing potential (or proto-oncogene) was noteworthy because it indicated that some tumours can develop by the aberrant expression of cellular genes responsible for controlling the mechanisms by which cells either divide or die. expression of the oncogene (a mutant of a proto-oncogene) then permits cell division to proceed unchecked. for this discovery, varmus and bishop were awarded the 1989 nobel prize in physiology or medicine ‘for their discovery of the cellular origin of retroviral oncogenes’. rsv is a retrovirus, which means that it encodes an enzyme activity to reverse transcribe its rna genome into a ‘proviral’ dna copy that integrates in to the host chromosomal dna. it was complicit in the award of yet another nobel prize in physiology or medicine to howard temin and david baltimore (with renato dulbecco) in 1975 for their discovery of this enzyme activity: reverse transcriptase.

human tumour viruses

several types of viruses are now known to be associated with the cause of human cancers. indeed, according to harald zur hausen, viruses confer a risk of developing cancer second only to smoking tobacco. the mechanisms by which they do so are beyond the scope of this article, but some of the accompanying articles address this topic. the known oncogenic human viruses are listed in table 1 and described briefly below.

ebv

the first human tumour virus was discovered in 1964 when tony epstein and yvonne barr visualised herpesvirus- like particles in cells from endemic (i.e. occurring in individuals of african
origin) burkitt’s lymphoma – the same tumour recognised by louis leakey in 1932. this virus was later found to be a novel herpesvirus and was named epstein–barr virus (ebv). the link between ebv and burkitt’s lymphoma was met with scepticism initially, as ebv is insufficient to cause lymphomas: approximately 90% of the population is infected with ebv. however, overwhelming evidence now supports the causal association of ebv with endemic burkitt’s lymphoma, and ebv is now also associated with hodgkin lymphoma and nasopharyngeal carcinoma (npc), amongst others. indeed, harald zur hausen’s work contributed to the recognition of the role of ebv in burkitt’s lymphoma and npc. graham taylor and james turner discuss ebv in more detail in the accompanying article (p. 34).

since the discovery of ebv, strong evidence for five other viruses being associated with the development of human cancers has accumulated, and a sixth virus, merkel cell polyomavirus (mcpyv), is likely to be added to the list (see table 1), equating to seven oncogenic human viruses.

hbv & hcv

hepatitis b virus (hbv) was discovered in 1968 and linked to the development of hepatocellular carcinoma (hcc) in 1975. subsequently, an hbv vaccine was developed to protect against hbv infection and was implemented in 1980. it is still in use today and was the first vaccine to prevent the development of a specific human cancer. the early studies correlating reduced hbv infection and decreased hcc incidence in taiwan provided the essential epidemiological link between the virus and the tumour.

in 1989, michael houghton and his team discovered hepatitis c virus (hcv) during their studies of non-a, non-b hepatitis. the world heath organization estimates that there are more than 170 million people chronically infected with hcv who are at risk of developing liver cirrhosis and hcc. jane mckeating and colin howard discuss hbv and hcv in their accompanying article.

table 1. the seven human oncogenic viruses and the major malignancies with which they are associated.

virus	malignant disease
epstein–barr virus	burkitt’s lymphoma; nasopharyngeal carcinoma; hodgkin lymphoma
hepatitis b virus	liver cancer
human papillomavirus	cervical cancer
human t-cell leukaemia virus type 1	adult t-cell leukaemia
hepatitis c virus	liver cancer
kaposi’s sarcoma-associated herpesvirus	kaposi’s sarcoma; primary effusion lymphoma
merkel cell polyomavirus*	merkel cell carcinoma

*a causal association between merkel cell polyomavirus and merkel cell carcinoma has yet to be formally established.

hpv

in the early 1980s, harald zur hausen identified novel strains of hpv dna in cervical cancer biopsies. thus, in 1983 his lab isolated hpv type 16 (hpv-16) in ~50% of the biopsies and in 1984 he reported hpv-18 in ~20%. the importance of these viruses in the cause of cervical cancer was demonstrated by the frequent clonal integration of their genomes in the tumour cells, suggesting the viruses were driving the proliferation of the malignant cells. moreover, two hpv genes, e6 and e7, were always expressed in the tumours and in pre-malignant lesions, suggesting that these viral proteins were responsible for the malignancy. in 2008, the year
of professor zur hausen’s nobel prize, anti-hpv vaccines were introduced into the uk for school-age girls with the aim of preventing their infection with the so-called high-risk strains of the virus, thereby reducing the incidence of cancer of the cervix. sally roberts and jo parish provide further information on hpv in their accompanying article (p. 26).

htlv-1

since rsv, the first oncogenic virus discovered, is a retrovirus, the potential of retroviruses to cause cancer in humans has been of great interest and under intense investigation. the identification of human t-cell leukaemia virus type 1 (htlv-1) in cutaneous t-cell lymphoma in 1980 by robert gallo and colleagues provided one opportunity to analyse the role of this retrovirus in human disease development. it is now recognised as the aetiological agent of adult t-cell leukaemia (atl). this virus is endemic in certain geographic regions, such as southern japan and central africa. however, as with infection with so many oncogenic viruses, only a proportion (1–4%) of infected individuals will develop atl and the onset of disease can take decades. htlv-1 can also cause a non-malignant, progressive neurological disease called htlv-1-associated myelopathy (ham) or tropical spastic paraparesis (tsp). htlv-1 transforms cells into a malignant state through the action of the tax protein. tax has a multitude of activities, including being able to transform human t-cells by transactivating various cellular promoters, such as those of cytokines and cytokine receptors, leading to signalling cascade activation.

kshv

kaposi’s sarcoma (ks) was first described in elderly mediterranean men in 1872 by moritz kaposi. the lesions are unusually angioproliferative and inflammatory in nature. it was originally a quite rare cancer, also seen in certain individuals living in africa and in organ transplant recipients. however, its appearance in young gay men during the 1980s coincided with the beginning of the aids pandemic and it is now an aids-defining illness. the epidemiology of ks in hiv-infected people, studied largely by valerie beral, led many to search for a causal agent that was hypothesised to be a virus. in 1994, the husband and wife team of yuan chang and patrick moore used a new technique (representational difference analysis) to identify new dna fragments in ks tissue compared to healthy tissue of the same patient. these fragments belonged to a new virus that yuan and patrick called kaposi’s sarcoma-associated herpesvirus (kshv). fifteen years after its discovery, kshv was formally acknowledged as the causal agent of ks and a rare lymphoma called primary effusion lymphoma. of the 84 genes encoded by kshv, a subset [e.g. latency-associated nuclear antigen (lana) and viral cyclin, as well as virus-encoded micrornas] are expressed during latency drive cell proliferation and inhibit apoptosis. kshv lytic proteins (e.g. a viral cytokine and viral chemokines) also contribute to the unique pathogenesis of ks.

mcpyv

undeterred by their considerable success in discovering kshv, yuan and patrick went on to discover yet another virus, merkel cell polyomavirus (mcpyv), this time by a technique they devised called digital transcriptome subtraction (dts). this virus is very likely to be the cause of merkel cell carcinoma, an aggressive cutaneous malignancy that arises from neuroendocrine mechanoreceptors (merkel cells) in the basal layer of the epidermis. as indicated by its name, mcpyv belongs to the polyomaviridae family, members of which have a strong track record in causing tumours, though not necessarily in humans.

in recognition of their considerable contributions to tumour virology, by their discovery of two novel viruses and their contributions to understanding the biology of them, yuan and patrick were awarded the sgm marjorie stephenson prize in 2012, for which their lecture at the spring conference in dublin was entitled ‘old themes and new variations in human tumour virology’ (a video of this lecture is available on the sgm youtube channel at http://bit.ly/14dctsv).

tumour viruses do not always cause cancer

a common theme of human tumour viruses is that they are not rare, yet they confer a significant risk for cancer development. this begs the question: why doesn’t everyone infected with such viruses develop cancer? first, there are checkpoints built into the mechanism controlling cell division, and overcoming these checkpoints to turn a healthy cell into a malignant transformed cancer cell is not a one-step event. alfred knudsen hypothesised in 1971 that one cancer in particular, childhood retinoblastoma, could evolve in as few as two steps. in general it is recognised that multiple (perhaps 4–6) steps need to be overcome in order to generate a transformed cell. various factors (host genetics, diet, lifestyle, etc.) contribute to each step. second, the immune system plays an important role in preventing virus-mediated tumorigenesis. since our immune system is highly adapted to distinguish self from non-self, and viruses represent non-self, the immune system can in many cases recognise viral proteins expressed in transformed cells and eliminate those cells. when the immune system is compromised, surveillance is
reduced and the risk of cancer is increased. thus, cancer development is multistep and multifactorial. tumour viruses can contribute one of the steps in the pathway.

the future

understanding virus-associated cancer has made an important impact on improving human health. the implementation of an hbv vaccine dramatically reduced infection with this virus, and concomitantly hbv-associated hcc.

another success is likely to be the efficacy of hpv vaccines in reducing the human burden of cancer of the cervix. new generation therapeutics inhibiting hcv replication and undergoing clinical trials may likewise reduce the burden of hcc associated with this virus. hopefully these success stories will continue.

but what’s next? are there other human viruses yet to be discovered that can therefore be targeted to provide hope in the treatment and prevention of their respective tumours? with current and rapidly developing molecular tools, we may not need to wait too long to find
out.

laura n. hindle & david j. blackbourn

university of birmingham, cancer research uk institute for cancer studies, edgbaston, birmingham b15 2tt; email [email protected]; [email protected]

further reading

www.nobelprize.org/nobel_prizes/medicine/www.who.int/csr/disease/hepatitis/en/index.html

de martel, c. & others (2012). global burden of cancers attributable to infections in 2008: a review and synthetic analysis. lancet oncol. 13, 607.

knudson, a.g., jr (1971). mutation and cancer: statistical study of retinoblastoma. proc natl acad sci u s a 68, 820.

mesri, e.a., cesarman, e. & boshoff, c. (2010). kaposi’s sarcoma and its associated herpesvirus. nat rev cancer 10, 707.

moore, p.s. & chang, y. (2010). why do viruses cause cancer? highlights of the first century of human tumour virology. nat rev cancer 10, 878.

mukherjee, s. (2011). the emperor of all maladies: a biography of cancer. london: harpercollins.

rott, r. & siddell, s. (1998). one hundred years of animal virology. j gen virol 79, 2871.

figures

viruses. istockphoto / thinkstock

viruses and cancer timeline. images: skull, digital vision/thinkstock; ramazzini, pott, smith papyrus & gallo, public domain; müller & virchow, us national library of medicine; rous, nypl/science source/science photo library; epstein, prof. sir anthony epstein cbe frs, wolfson college, oxford; zur hausen, armin kübelbeck; chang & moore, i. atherton, sgm; houghton, © university of alberta

kaposi’s sarcoma skin plaques on the skin of an aids patient. national cancer institute

merkel cell carcinoma on the eyelid of an 80-year-old patient. paul nghiem md phd; www.merkelcell.org