Harald  zur Hausen (2014) - Infections Linked to Human Cancers: Mechanisms and Synergisms

It is really a great pleasure for me as well to be here today. Let me start out with a more general remark in the beginning. The interpretation that human cancer is linked to an imbalance between oncogenes and tumour suppressor genes really found enormous fascination in the scientific public and also among, of course, those people who invented this. The involvement of infectious agents in human cancer are basically a disturbing factor in this beautiful picture. And in a way to such an extent that even some of the protagonists in this field don’t even like so much to discuss this aspect. This in spite of the fact that today we know that about slightly more than 20% of human cancers are linked to infections, not only virus infections but also bacteria and parasitic infections. That we are presently able to prevent or are in the process of preventing 2 major human cancers, hepatitis B linked liver cancer and cervical cancer by vaccination. And that even the 20 tons of HeLa cells which have been, supposedly, been produced during a long period of cultivation, of decades of cultivation. That they as a result of a continued expression of 2 specific viral oncogenes namely A6, A7 genes of human papillomavirus type 18, so far - if you switch them off, the HeLa cells wouldn’t be any more malignant cells. Now what I try to do today is just to very briefly go into some aspects of mechanism of carcinogenesis. But since each of these topics would require probably another lecture, I’ll just leave it as that what it indicated here. We can basically differentiate between 2 types of interactions of infectious agents in carcinogenesis. Namely what we call 'direct carcinogenesis' where the presence and, in many instances, expression of specific viral components is required for the tumorigenic growth of the respective types of cells. Frequently accompanied by the integration of viral DNA to chromosomal DNA. There may be not a human example but clearly an animal one. An insertional gene activation or suppression of specific cellular genes. And even this is a novel and quite interesting aspect which probably deserves much more attention, namely the continued episomal presence of viral nucleic acid. And the suppression and activation of cellular genes, for instance by viral micro RNase, under those circumstances. Now besides this we have a number of indirect interactions. Where a number of agents are able, either by the induction of immunosuppression. In this case, Francoise Barre-Sinoussi discussed it very intensively with you, by immunosuppression leading to the activation of latent tumour virus genomes. For instance resulting in Kaposi sarcomas due to the human herpes virus type 8. Or B cell lymphomas due to Epstein-Barr virus. The induction of oxygen and nitrogen radicals by some of the infectious events, linked really to inflammatory events. Another interesting and very poorly investigated aspect is the amplification of other latent tumour virus DNA by co-infecting the same cell by other agents like herpes viruses, adenoviruses, vaccine viruses, pox viruses. The induction of mutations and/or translocations, also of epigenetic modifications. And the prevention of apoptosis, the self-programmed cell death. Well I initially had planned to discuss this in more detail today. But I decided in the end to change my topic a little bit and to come mainly to perspectives which came up, particularly in recent periods of times. Now we can at the beginning state that none of the infectious agents is inducing cancer as a direct consequence of infection. Wherever this is happening, and there is a rare exception: This person inherited already some genetic modifications prior to the development of the infection and subsequently of the tumour. So even some of these infectious agents emerged as necessary factors for the respective form of cancer. As for instance indicated in the case of human papillomavirus and cervical cancer. They uniformly require additional epigenetic modifications, commonly in host cell genes. And really this requirement of additional modifications is an explanation for the varying latency periods between individual agents, between primary infection and cancer development. And in host cell genes these modifications are either inherited or most frequently acquired during lifetime. They may, however, also occur as in the rare case of a specific human polyomavirus, a merkel polyomavirus, within in the genome of the infecting agent. So there are a number of varieties possible here. And this really brings me directly to a topic which maybe is a slightly unorthodox and probably also somewhat anti-dogmatic view of some of the developments in human cancer and in chronic diseases. I have shown this slide at a couple of different times, I think initially even in Stockholm just for the first time, as epidemiological evidences that still some cancers exist which have not yet been linked to infections, but where the epidemiology may point to an infectious event in the background of this type of cancer. Now cancers occurring at increased frequency under immunosuppression, I am not going to dwell on this anymore. There are still a couple of cancers like kidney cancer, thyroid cancer which are increased under immunosuppression. They have not yet been linked to infectious events but clearly others are, which are highly increased under those conditions. Cancers with reduced incidence under immunosuppression, an extremely interesting topic. And I will come, if the time permits it, at the end back to this question. Cancers influenced by other basically nontumorigenic infections. This concerns mainly childhood cancers occurring in the first years of life. Again I will quit this today, although it is an extremely interesting topic. And I will mainly concentrate on nutritional risk factors possibly linked to infection. And here cancer of the colon sticks out since quite a long time. There are a large number of reports. Prospective epidemiological studies which very uniformly, with maybe 2 or 3 exceptions only, ascribe a higher incidence rate of approximately 20 to 30% mainly of colon cancer. It’s usually in epidemiological studies combined with rectal cancer, although this may not be fully justified. Linking it to the higher rate of consumption of red or processed meat, air dried, smoked meat and so on, in particular of beef. I am coming to this point specifically. Countries with the highest rate of red meat consumption are listed here, commonly reveal a very high rate of colon cancer and colorectal cancer. If you look into the incident data for this type of cancer, established in the world cancer report, you can see that here for instance in 4 European countries - and it accounts probably, it's true for most other European countries as well In other parts of the world it’s slightly different. Specifically in the United States, the rate is going down right now. Probably due to effective screening procedures, colonoscopies which are being conducted, leading to the discovery of precursor lesions and the removal of the precursor lesions. And thus to the reduction in the cancer rate. In Japan and Korea the rate is rapidly increasing, 20 years after the Second World War and 20 years after the Korean War. India, a very interesting example as well, has a very low rate of this type of cancer. Now since more than 30 years ago, when Sugimura in Tokyo discovered that broiling, roasting, grilling, frying of meat leads to the formation of chemical carcinogens in the preparatory process of consumption, forming heterocyclic amines, heterocyclic aromatic hydrocarbons and so on, a couple of others. And this seemed to explain fully the story that here are chemical carcinogens being produced which at least in animal models are also carcinogenic, but specifically for the colon. And they’re usually applied in concentrations 1,000 to 10,000 or higher than under these natural conditions of preparing this food. But clearly they are carcinogens and this would very well fit into the story. And I believe that the large majority of epidemiologists is even up to today deeply convinced that this is a major factor in the so called western diet as one of the risk factors for the development of this type of cancer. The story would fit very nicely if there wouldn't be some exceptions from this rule. First of all the production of, the preparation of poultry and fish by the same types of procedure, grilling, barbecue and frying and so on leads to the formation of the same chemical carcinogens. Sometimes even at higher concentrations than in a well broiled or barbecued steak under those conditions. And for fish even it has been claimed that it has a somewhat protective effect for colon cancer. But there are a few other points which are quite interesting because if you look into the global epidemiology of colon cancer, colorectal cancer. You can see that there are parts of the world, northern America, Europe, this part of Russia with very high incidence. Australia by the way belongs to them, particularly the white population in South Africa is effected to a large degree, Argentina - those are the so called high risk countries for this type of cancer. But there are a few which are quite unusual because Mongolia for instance, India and Bolivia because the rates there consistently has been shown to be very low, of this type of cancer. Now Mongolia is a particularly striking and interesting example. This is shown here. There are a couple of studies which are very well conducted, demonstrating that the rate of colon cancer is very low, the mortality is very low. There are a few other parts in south East Asia, a few islands in particular, with low rate. But if you take Taiwan, Japan, South Korea, Brunei, Singapore, they have very high rates again of the same types of cancer. So if it's red meat in particular, what is the red meat consumption in Mongolia? Well, Mongolia is supposed to have the highest red meat consumption in the world in comparison to all other countries. Particularly barbecued, grilled meat and also air dried meat. So that is an interesting point. So what kind of meat is being consumed in Mongolia? And I found this type of statistics here to say it’s mainly Yak meat. It’s a different species of cattle. It’s not our common dairy cattle. It's mutton and goat, camel meat, horse meat and so on. As I said before it’s mainly consumed this way. And it’s interesting, in recent years there’s an introduction also of dairy products from the west to Mongolia. And also to some degree some dairy meat is being, dairy cattle meat is being introduced in the same country. So there are some differences. And this made me into a kind of a specialist in cattle by now. Because these are the common, the ox or the common ancestor of those types of cattle which we call basically now 'the dairy cattle' globally. The Zebus are clearly a different species, very distantly related. Also they are still able to cross breed with those here. Water buffalos don’t seem to play a role in this whole discussion. But Yaks are clearly important in this respect. They are mainly kept in the mountainous areas and also in the plains of Mongolia, and also of Tibet and parts of China. In Africa for instance it is interesting. Watusi cattle with these enormously long horns which have, obviously, as it has been shown by genetic studies and the dark area here indicates this, a very high mixture of Zebu blood in them, cross breeding with Zebu. Here in Asia the Zebu play a significant role. In Bolivia, and this is Bolivia here, indeed there is mainly Zebu being maintained at the same time. So what it led us to suspect is that really the risk factor in red meat seems to be particularly our dairy cattle, our common dairy cattle. And if you look into the geographic distribution - this is another type of showing this picture. By the way, in China we have also high mixture of Zebu cattle, the so called yellow cattle in China, which is mainly maintained. But of course also pork meat is mainly consumed. It has an intermediate rate in comparison to some of the other high risk areas. By the way Bolivia is indicated in the wrong place but it is clearly Bolivia. Now again I have shown this slide a couple of years ago, probably the first time in Stockholm. And this is really intriguing us since quite a long time. Namely that we do know a couple of human pathogenic viruses. Well in many incidences they are pathogenic in humans but consistently, persistently maintained and even excreted, like the polyomavirus BK and JC, some adenoviruses, to a limited degree also Epstein-Barr virus and high-risk human papillomavirus. But they clearly produce tumours under certain circumstances. But let’s say if you transfer those agents to other species in which they cannot replicate. where they are replication-deficient, they are very actively producing tumours under those conditions. And there’s in all monkeys even by the JC virus, gliomas have been induced. Now this really triggered the consideration: Is it possible that some of the animals with which we live in our close proximity, that they do have similar agents? Not pathogenic, in those animals in which they can replicate, but potentially pathogenic in other species, in humans, where they cannot replicate. Is it possible that the transmission occurs either by consumption of the product or by close contact that this is happening? Let’s go back for a moment to the situation in Japan and Korea because it’s particularly striking, this enormous increase. If you take these 2 curves away, 20 years after the Second World War, 20 years after the Korean War, they overlap completely. Virtually overlapping. Within this period of time enormous quantities of beef and dairy products, but also to some degree pork products, have been imported. But in particular from the United States to those countries. And the dietary customs changed a lot. Many of you have been in Japan I am sure and have experienced it yourself. That previously, in previous years the sashimi was a rare fish. Now you frequently get it offered as rare beef. And the sukiyaki also rare beef. The Shabu Shabu meat fondue, where you dip briefly rare slices of beef into boiling water and after a minute or so, when you take it out, the outside is well done, the inner side is still rare. As my host told me, this is the delicious part of this really. In Korea the Yuk Hwe became quite popular tartar, nicely decorated with egg yolk and vegetables. But anyway these types of customs are not only spreading in these 2 countries. But they are also spreading at present to Vietnam, to Thailand and to some of the other south East Asian countries which still have a relatively low rate of colon cancer. It will be interesting to see what is going to happen there in the future. Now for virologists air dried beef or air dried meat is the best preservative for keeping viruses in an infectious state. The Bünderfleisch in Switzerland, the Bresaola in Italy, the Kobe beef in Japan and the Biltong in South Africa are particularly interesting because it was in the past mainly consumed by the white population, the pieces of raw beef, air dried, but now recently became also very popular amongst the black population in South Africa as well. So this lead at our place to suspicions that there might exist a species-specific bovine factor causally involved in human colorectal cancer. And according to our interpretation this factor might represent a relatively thermo-resistant virus, not necessarily thermo-resistant but possibly, present in meals of raw and undercooked beef. It may lead to latent infections in the intestinal tract of those hosts. Its potential carcinogenic function, as we know it from most of the other tumorigenic agents in humans for instance, is suppressed by specific cellular proteins, cellular factors. And that preceding concomitant or subsequent mutation events, specifically those mediated by the chemical carcinogens arising in the preparatory steps for beef consumption. Or also chronic inflammatory events. In recent years specifically streptococcus bovis, streptococcus infantium as it is also called. Fusobacterium nucleatum have been relatively often reported that they may play a role here. But these are clearly involved in inflammatory events. They lead to the induction of oxygen radicals under those conditions and may act synergistically in the development of colon cancer. This is basically was the model. These are the 2 types of bacteria, the fusobacteria and the streptococci which I show here. But let me just come to the model which came up from these types of studies. Namely if you take this as the lifespan of humans, of let’s say 80 years. That we have arbitrarily an exposure to these putative infectious agents due to the consumption of raw or undercooked red meat, beef specifically apparently. This leads to transient or latent infection by putative carcinogenic viruses. But that on the other hand the consumption of cooked and cured red meat, due to the induction of mutations in host cell DNA by the chemical or the biological carcinogens like fusobacterium and streptococci, would lead to acquired genetic modifications. Again the arrows here arbitrarily introduced. Inherited genetic modifications would play quite a similar role here. Leading eventually to the formation of polyps and to subsequent development of tumours. Now can we approach this question experimentally? Yes I think we can and we did. Because we analysed a total of 120 sera from healthy cattle. All 5 year old healthy cows from the veterinary faculty of the University of Leipzig. We purified virus-like particles by initially purifying DNase- and RNase-resistant particles, extracted subsequently the sequence of DNA from these preparations. And thus far we came up with 18 novel agents which are present in the blood of these healthy cattle. All of them apparently single-stranded circular DNase, belonging to 4 different groups of so far only episomally persistent genomes. And isolation of some of these DNase also originated from milk samples. Obviously the presence in the blood also leads to the excretion into milk and probably also into dairy products. Similar to most other known single-stranded DNA viruses, we anticipate these agents will be relatively heat resistant. Although we do not have a direct proof of this because we don’t have really a working infectivity system so far established. This shows 5 of these isolates for instance with quite a number of variation in the size between here something like 1,000 nucleotides up to 2,900 nucleotides here on the upper right side. So there is a substantial variation among them, a wide variability. Now the analysis of one of these isolates. We have only done it with one of these isolates so far, for the presence of its DNA in several colon cancer cell lines yielded only negative results. But, and this is the surprising result of this whole story, we found evidence first of all for transmission of these agents to human cells, at least the kind of abortive cycle in human cells, and also links to non-malignant chronic neurological disorder. I am not willing to talk at this moment about this because it’s really in the process of intensive investigation at our place. But clearly there is some kind of a very interesting relationship becoming apparent which for us was clearly unexpected. We can conclude at this stage, for this part, that the blood of healthy dairy cattle contains infections with several specific single-stranded circular DNA molecules. Obviously encapsidated in a DNase and RNase resistant protein code. The DNA of at least of some of these agents can be demonstrated in commercially available pasteurised milk, an interesting point here, but we don’t know whether it’s still infectious or not. Those which may be human pathogenic are probably species-specific, persisting in European/Asian dairy cattle, probably not, although we have no proof of this either, not in Yak and Zebu cattle. It’s only derived from the epidemiological studies. And at least some of these newly isolated agents can be successfully transmitted to human cells. And are occasionally found in specific human sera and to specific human disorders. Now preliminary evidence points to the potential role of these agents in neurological diseases. Somewhat surprising was that under those conditions. These are the people who work on this aspect. Ethel-Michelle de Villiers who is here, who is really the main driver of the analysis and also of the DNA sequencing. The studies which have been conducted here. And Hermann Muller who provided us with the cattle sera and a couple of other devoted people who collaborate in the studies. And here are our involuntary collaborators. The cattle from Leipzig. Exactly the cows from which the blood has been obtained. Now let me just conclude by making a few additional statements. And that is interesting in terms of other types of cancers which have been linked to an increased risk after red meat consumption, and that is in particular breast cancer. There are a number of interesting studies which come up. And this morning even I discovered, there are more recently which came up along the same line: that red meat consumption is also a risk factor for breast cancer. There are some studies implying also for lung cancer and even for a couple of other cancers. But particular here the data are much more controversial. Now if you look into breast cancer, we have an interesting model. Breast cancer in humans is one of the few cancers where immunosuppression exerts a protective effect. In Aids patients and in renal or organ allograft recipients, the breast cancer incidence is reduced by about 15% in comparison to non-immunosuppressed controls. Now we have an animal model here, discovered originally by John Bittner in 1938 in the United States, the mouse mammary tumour virus. Because this virus is transmitted from the mothers feeding the new born babies via the milk of the respective mothers. And it passes through the intestines and settles down in the lymphocytes of the Peyer's patches, where it infects B and T cells leading to what is called super antigen induction. The cells virtually swell, diluted into the blood stream, produce large quantities of the virus. And infact the quantity of the virus is the responsible factor for reaching sufficient concentrations in the mammary gland. And here settling down in these regulating very specific pathways within, mainly within the wnt pathway under those circumstances, and to become activated by female hormones, particularly in the course of lactation and producing again virus particles to a large degree. Now if you immunosuppress these mice, the tumour incidence is reduced because the superantigen-producing cells are suppressed under those conditions. So the viral production is going down. And this lowers the risk for the subsequent events. Is the same happening in human breast cancer cells? We clearly don’t know at this stage, at this point. It deserves attention. Probably it might be much more worthwhile to look into potential carriers of the lymphatic system for such agents than looking specifically, only or exclusively into breast tissue itself. Now if you compare the geographic incidence of breast cancer and colon cancer: it is basically quite similar. It’s not absolutely identical but quite similar in many countries. Again here for instance Mongolia, Bolivia and also to some degree India stick out with a comparatively low rate of this type of cancer. But there are also some other interesting differences obvious. Bolivia for instance, where the Zebu cattle are quite prevalent is one of the countries where the rate of milk consumption is supposed to be the lowest one in the world. I have no chance to test this, I took it from this publication here. But if you look into the rate for colon cancer again which I showed before for Japan, Korea and India, there are some differences because colon cancer rose rapidly in this period of time. On the other hand breast cancer in Japan, also in Korea. rose only slightly during this period of time. In India it’s going up as well. Whereas colon cancer is basically remaining low under those circumstances. So if there is a factor involved, a similar factor as in colon cancer, we would suspect that it is not an identical factor but probably something which may occur from similar sources as colon cancer as well. Now it’s also interesting and these are my last 3 slides, I still have one minute, to look into the rate of breast cancer in the urban areas of Mongolia and in the rural areas of the same country. Here the rate of this type of cancer remained practically constant, in the urban areas it’s gradually increasing. I mentioned before that dairy products and some meat of dairy cattle have been introduced recently. They do not sustain very well in Mongolia due to the cold temperatures and they cannot be maintained there, because winter temperatures of minus 50 degrees are not tolerated by these animals. This is my very last slide. It shows that more than 20 epidemiological studies also point to an increased risk in butchers, slaughter house workers for lung and oropharyngeal cancer. This occupation group is regularly exposed to aerosols originating from handling slaughtered animals. The oropharynx and the lung should be the prime organs of exposure. Although initially suspected to originate from the heavy smoking habits of this group, even controlling for this, it didn’t modify the picture. So in a way that’s a new area emerging at least in our view. An area which points to potential role of episomally persisting agents which in most cases escape the deep sequencing essays of individual tumours. First of all in the process of isolating chromosomal DNA, and secondly in the bioinformatics filtering process subsequently which have in most cases, only very recently a few people pay more attention to it, have not been identified under those circumstances. Although we find for instance in colon cancer in almost 100% positivity for TT virus, another group of single-stranded agents which are quite common in humans. But we need to be open for those types of developments. They are different from our conventional models which we established for interaction of infections with cancer. But they may be even more worthwhile to study them intensively. Thank you very much for your attention. Applause.

Harald zur Hausen (2014)

Infections Linked to Human Cancers: Mechanisms and Synergisms

Harald zur Hausen (2014)

Infections Linked to Human Cancers: Mechanisms and Synergisms

Abstract

Slightly more than 20% of the global cancer incidence is presently being linked to viral, bacterial, or parasitic infections. The mechanisms by which these agents mediate malignant transformation differ substantially. Some contribute directly, frequently integrating their genome into chromosomal DNA, others may persist episomally. Persistence of the respective DNA is required to maintain the malignant phenotype of the affected cells. Indirect contributions may involve immunosuppression, induction of reactive oxygen radicals, amplification of latent tumor virus DNA, induction of mutations and translocations, and prevention of apoptosis. Even in case of direct contributions to carcinogenesis additional modifications of host cell genes are required prior to onset of malignant growth. The required number of these genetic or epigenetic changes determines the latency period between primary infections and malignant proliferation. In most cases it varies between 10 and 60 years.
Interactions of potentially carcinogenic infections with other mutagenic factors (e.g. chemical carcinogens, irradiation, hormones, inflammations, specific virus infections) emerges more as rule than exception. Nevertheless, the identification of potentially carcinogenic infectious agents permitted novel approaches in cancer prevention and identification of persons at risk for specific cancers. The emerging picture of the role of infections in human carcinogenesis, however, results in great difficulties to apply criteria outlined by Koch, later by Hill and others, to define causality.

Recommended Readings:
zur Hausen H., de Villiers, E.-M.: Prenatal Infections with Subsequent Immune Tolerance Could Explain the Epidemiology of Common Childhood Cancers. World Cancer Report 2014, pages 261-265, IARC Lyon
zur Hausen, H. de Villiers, E.M. Cancer «causation» by infection – individual contributions and synergistic networks. Seminars in Oncology, in print.

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT


Related Content