Heir of the Dog

I’ve probably in the past owned up to causing generations of students to do that raised eyebrow thing, familiar to all parents of teenagers, that, far more pointedly than words, says ‘The old boy’s finally lost it.’ Indeed I may well have a bit of a causative repertoire but one that unfailingly works is revealing that, even after a life in science, I still get ‘Wow’ moments every couple of months or so when I read or hear of some new discovery, method or insight that brings home yet again the wonder of Nature – or has you asking ‘Why didn’t I think of that?’ (The response to that one’s easy, by the way, so please don’t write in).

A common question

The most recent of these jaw-dropping events relates to a question often asked about cancer: ‘Can you catch it from someone else?’ In other words, can cancers be passed from one person to another by infection, much as happens with ’flu? The answer’s ‘No’ but, as usual in this field, even the firmest statement can do with a little explanation. The first point is that the ‘No’ is true even for 20% or so of cancers that are actually started by microbial infection – what you might call ‘bugs’ – bacteria, fungi, and viruses. One such, the bacterium Helicobacter pylori, can cause stomach ulcers that may lead to cancer. Those even smaller bugbears, viruses (typically one one-hundredth the size of a bacterium), are responsible for much of the cervical and liver cancer burden world-wide. Oh, and there’s a little, single-cell parasite (Trichomonas vaginalis), the most common non-viral, sexually transmitted infection in the world that, in men, can cause prostate cancer. But these infections are not cancers even though they may be an underlying cause – bacteria through prolonged inflammation and effects on the immune system and viruses by making proteins that affect how cells behave. Only when these perturbations cause genetic damage – i.e. DNA mutations – do you have a cancer. Which is why the answer to the original question is ‘No.’

There’s always one

Well, two in this case – and, given that we’re talking about cancer, you won’t be surprised that there are some oddities. They’re not exceptions to the ‘No’ answer because they occur in other animals – not in humans – but, in each, tumour cells are directly transferred from one creature to another – so it is cancer by infection. One such contagious tumour occurs in the Tasmanian devil. It’s transmitted by biting, an activity popular with these little chaps, and it gives rise to a particularly virulent facial tumor, eventually fatal because it prevents eating. To counter the probability that Tasmanian devils will become extinct in their native habitat, a number of Australian sanctuaries have breeding programmes aimed at setting up a disease-free colony on Kangaroo Island, South Australia.


Tas D






Tasmanian devils – cancer-free – Lone Pine Koala Sanctuary, Brisbane

A very similar condition in dogs known as canine transmissible venereal tumour (CTVT: also called Sticker’s sarcoma), mainly affects the external genitalia. First spotted in the nineteenth century by a Russian vet, it too is spread either by licking or biting and also through coitus. Dogs with CTVT can now be found on five continents and, from DNA analysis, we’ve known for some time that – remarkably – all their cancers are descended from a single, original tumour cell that appeared many years ago. They’re like one of those cell lines grown in labs all over the world, except they’ve been going far longer than any lab – with man’s best friend doing the cultivating.

So what is new?

Elizabeth Murchison and colleagues at The Wellcome Trust Sanger Institute, Cambridge have just produced the first whole-genome sequences of two of these tumours – from Australia and Brazil (an Aboriginal camp dog and a purebred American cocker spaniel). These confirmed that all CTVTs descend from a single ancestor who, they estimated, was trotting around about 11,000 years ago. The last common relative of the two dogs whose tumours were sequenced lived about 500 years ago, before his descendants went walkies to different continents.

And the ‘Wow’?

We already had a pretty good idea of how CTVTs have been handed down. In this paper the really amazing bit came in the detail. The authors estimated roughly how many mutations were present in each tumour. Answer: a staggering 1.9 million. And it’s staggering partly because it’s only slightly less than a change every 1,000 units (bases) in dog DNA but it’s truly awesome when you note that it’s several hundred times more than you find in most human cancers. We’re getting used to the idea of thousands or tens of thousands of mutations turning up in human cancer cells with associated gross disruptions of individual chromosomes. But these canine cancers display genetic mayhem on a massive scale – perhaps best visualized by comparing their chromosomes with those of a normal dog using a method that labels each with a different colour. A glance at the two pictures tells the story: all the cancer chromosomes from one of the tumour-bearing dogs (on the right) have been shuffled as if in some molecular card game. The full range of colours can still be seen, but of the normal pattern of 39 pairs of identical segments of DNA (left) there is no sign.

Two dogs chromos

Dog chromosomes. Left: normal; right: CTVT

(from Murchison, E.P. et al. (2014) Science 343, 437-440)

It seems incredible that cells can survive such a shattering of their genetic material – a state called ‘genetic instability’ because, once DNA damage sets in, mutations usually continue to accumulate. These cancers are uniquely bizarre, however, because although their genomes have been blown to smithereens, not only do the cells survive but they’ve continued suspended in this surreal state for centuries. They’re genetically stable – it really is the cellular equivalent of balancing an elephant on a pin.

‘Wow’ Indeed – but so what?

So like me you’ve been blown away by these discoveries but you may be asking, apart from the excitement, what’s in it for us humans? Well, there’s one other very strange thing about these dog cancers. Infected animals do indeed develop the most unpleasant, large tumours – but most of them are eventually rejected by the host dog. That is, its immune system gets to work to eliminate them – and after that the dog is immune to further infection. We are only just beginning to find ways of boosting the human immune system so that it can attack cancers and maybe, just maybe, we can extract from the stable chaos of the CTVT genome the secret of how they provoke rejection – and maybe that will guide human treatments.


Murchison, E.P. et al. (2014). Transmissible Dog Cancer Genome Reveals the Origin and History of an Ancient Cell Lineage. Science 343, 437-440.