Making Movies in DNA

Last time we reminded ourselves of one of the ways in which cancer is odd but, of course, underpinning not just cancers but all the peculiarities of life is DNA. The enduring wonder is how something so basically simple – just four slightly different chemical groups (OK, they are bases!) – can form the genetic material (the instruction book, if you like) for all life on earth. The answer, as almost everyone knows these days, is that there’s an awful lot of it in every cell – meaning that the four bases (A, C, G & T) have an essentially infinite coding capacity.

That doesn’t make it any the less wonderful but it does carry a huge implication: if something you can squeeze into a single cell can carry limitless information it must be the most powerful of all storage systems.

A picture’s worth a thousand words

We looked at the storage power of DNA a few months ago (in “How Does DNA Do It?”) and noted that its storage density is 1000 times that of flash memories, that it’s fairly easy to scan text and transform the pixels into genetic code and that, as an example, someone has already put Shakespeare’s sonnets into DNA form.

Now Seth Shipman, George Church and colleagues at Harvard have taken the field several steps forward by capturing black and white images and a short movie in DNA. Moreover they’ve managed to get these ‘DNA recordings’ taken up by living cells from which they could subsequently recover the images.

Crumbs! How did they do it?

First they used essentially the text method to encode images of a human hand: assign the four bases (A, C, G & T) to four pixel colours (this gives a grayscale image: colours can be acquired by using groups of bases for each pixel). These DNA sequences were then introduced into bacteria (specifically E. coli) by electroporation (an electrical pulse briefly opens pores in the cell membrane).

The cells treat this foreign DNA as though it was from an invading virus and switch on their CRISPR system (summarized in “Re-writing the Manual of Life”). This takes short pieces of viral DNA and inserts them into the cell’s own genome in the form of ‘spacers’ (the point being that the stored sequences confer ‘adaptive immunity’: the cell has an immunological memory so it is primed to respond effectively if it’s infected again by that viral pathogen).

In this case, however, the cells have been fooled: the ‘spacers’ generated carry encoded pictures, rather than viral signatures.

Because spacers are short it’s obvious that you’ll need lots of them to carry the information in a photo. To keep track when it comes to reassembling the picture, each DNA fragment was tagged with a barcode (and fortunately we explained cellular barcoding in “A Word From The Nerds”).

Once incorporated in the bugs the information was maintained over many bacterial generations (48 in fact) and is recoverable by high-throughput sequencing and reconstruction of the patterns using the barcodes.

And the movie bit?

Simple. In principle they used the same methods to encode sequential frames.

Pictures in DNA.

Top: Using triplets of bases to encode 21 pixel colours. Images of a human hand (top) and a horse (bottom) were captured. For the movie they used freeze frames taken in 1872 by the English photographer Eadweard Muybridge. These showed that, for a fraction of a second, a galloping horse lifts all four hooves off the ground. Seemingly this won a return for the sometime California governor, Leland Stanford (he of university-founding fame) who had put a wager on geegees doing just that. From Shipman et al., 2017. You can see the movie here.

Getting the picture clear

To recap, in case you’re wondering if this is some scientific April Fools’ prank. What Church & Co. did is scan pictures and transform pixel density into the genetic code (i.e. sequences of the four bases A, C, G & T). They then made DNA carrying these sequences, persuaded bacteria to take up the DNA and incorporate it into their own genomes and, after growing many generations of the bugs, extracted their DNA, sequenced it and reconstructed the original images. By scanning sequential frames this can be extended to movies.

It’s not science fiction – but it is pretty amazing. With a droll turn of phrase Seth Shipman said “We want to turn cells into historians” and the work does have significant implications in showing something of the scope of biological memory systems.

Won’t be long before the trendy, instead of birthday presents of electronic family photo albums, are giving small tubes of bugs!

References

Shipman, S.L., Nivala, J., Macklis, J.D. & Church, G.M. (2017). CRISPR–Cas encoding of a digital movie into the genomes of a population of living bacteria. Nature 547, 345–349.

Advertisements

What Took You So Long?

A long, long time ago – 25 years to be precise – I was lucky enough to work for a few months at The University of New England in Armidale, up on the Northern Tablelands of New South Wales. And jolly wonderful it was too. You could see grazing kangaroos from my lab window and I got to play grade cricket! To anyone who’ll listen I can still describe in vivid detail the scoring of my first run in Oz. We’d won the toss and … (that’s quite enough cricket, Ed).

Equally wonderful is the fact that, in part courtesy of The University of Queensland, I’m going again to Oz – this time to do what I didn’t manage then: visit all the major cities. We begin in Brisbane this week giving a lecture in the U of Q’s Global Leadership series (yes really!), explaining the biology of cancer to an audience of largely non-scientists – at least I hope I’ve got the right brief! We end up in Perth in May having, in between if I can stick the pace, given a variety of talks and seminars to the general public, to schools and to cancer research institutes in Sydney, Melbourne and Adelaide. How good is that? Being invited to warble on about one of your favorite subjects whilst touring Oz? Wow!

What’s new?

All of which makes you think a bit about Father Time and what has happened in the interim. Answer quite a lot, of course. Collapse of communism, collapse and resurgence of Australian cricket (that’s your last warning, Ed) and so on but we’re supposed to inform and enthuse about cancer here so how’s that faired, particularly in Australia? Well, in the year I first followed Captain Cook (watch it, Ed) onto the shore of Botany Bay about 60,000 Australians were diagnosed with cancers of one sort or another and some 30,000 died from these diseases. At that time one in three men and one in four women would be directly affected by cancer in the first 75 years of life.

A Cook

Alastair Cook

And now? This time round the estimated numbers are 128,000 and over 43,000 with one in two men/one in three women discovering they have cancer by time they’re 85. All told, cancer accounts for about three in ten Australian deaths – much the same contribution as heart disease. To add to the gloom the numbers are going up not down so the prediction is 150,000 new cancer cases in 2020.

Not a lot and no surprise

Well, you may be thinking, no change there then – or even I told you so. After all, I’m forever in these pieces elaborating on current cancer stories holding forth about how slow is the progress of science: one step forward, two back, more of a shuffle than a step really, and so on. Or as Martin Schwartz more eloquently puts it, describing science as the art of productive stupidity – being ignorant by choice. This follows almost inevitably from the nature of research because working on what we don’t understand puts us in the awkward position of being ignorant. As Schwartz has it, one of the beautiful things about science is that it allows us to bumble along, getting it wrong time after time, and feel perfectly fine as long as we learn something each time. That’s why I keep telling you to ignore the “great breakthough” newspaper headline dribble – that’s just the hacks trying anything to persuade their editors to give them space to promote themselves.

But wait a mo.

All that sounds consistent with the signs that things in Oz have been going backwards at a rate of knots over the last 20-odd years. But hang on. As ever, bare stats can be a bit misleading (remember what Disraeli said). Thus although around 19,000 more people die each year from cancer than 30 years ago, this is due mainly to population growth and aging – Australian life expectancy has gone up by over four years since 1990 (it’s now 82). The death rate from cancers has fallen by more than 16% and the survival rate for many common cancers has increased by 30 per cent in the past two decades. So that’s great: terrific ad for living in Oz and something of a triumph for medical science.

A sunny side in Oz?

What’s more you can put a positive twist on even the gloomy side of the picture by noting that, if indeed there’s strength in unity, Australia’s trends are much the same as everyone else’s in what we like to call the developed world. Well sort of but there’s a serious negative for Australia Fair, as you might put it, something that sticks out like a sore thumb (or an itchy mole) when you glance at the stats. Between 1980 and 2010 the incidence of skin cancer has shot up in Australia by around 60%. The most common type is non-melanoma skin cancer – usually treatable as it generally doesn’t spread around the body. The nasty version is malignant melanoma – which does metastasize, although is essentially curable if caught before some of its cells escape from the primary site. And the really bad news is that it is now the third most common cancer in Australians and in those aged 15-44 years it is the most common cancer. In 2012, over 12,000 Australians were diagnosed with melanoma and it killed over 1,600. This disease is usually set off by ultraviolet light from sunlight (or sunbeds) damaging DNA (i.e. causing mutations) and you will not have missed the allusion to the fact that people with fair skin (or blue or green eyes/red or blond hair) are most at risk.So the current Oz figures are a bit of a blow to Richie Benaud’s campaign of which I made great play in Slip-Slop-Slap Is Not Enough.

220px-Melanoma_vs_normal_mole_ABCD_rule_NCI_Visuals_Online

ABCD rule illustration: On the left side from top to bottom: melanomas showing asymmetry, a border that is uneven, ragged, or notched, coloring of different shades of brown, black, or tan and diameter that had changed in size. The normal moles on the right side do not have abnormal characteristics (no asymmetry, even border, even color, no change in diameter).

Meanwhile in the lab?

It’s pretty sobering for me to reflect that it was only a few years before I went to Oz that the first human cancer gene (oncogene) was discovered. That was RAS, detected in human cancer cells in 1982 by Geoffrey Cooper at Harvard, Mariano Barbacid and Stuart Aaronson at the NIH, Robert Weinberg at MIT and Michael Wigler at Cold Spring Harbor Laboratory. Between then and 2003 several hundred more cancer genes were identified in a huge frenzy of molecular stamp collecting. Then came the human genome sequencing project and in its wake analysis of tumours on a scale and level of detail that is almost stupefying and would have been unimaginable before 2003. To appreciate the mountain of cancer data that has been assembled over that period, screen the literature data base for research papers that have ‘RAS’ in the title: that is, contain significant info relating to that gene. Answer: 76,000. That’s seventy-six thousand separate pieces of research that have made it through all the peer review and editorial machinery to see the light of day in print. And RAS, massive player though it is, is not the biggest. Do the same check for a gene called P53 and the number is: over 145,000!!

Confused? The plot so far …

First up we noted that the cancer burden in Oz has got a lot heavier over the last 25 years, then we reminded you that advances in science are of the snail-like variety – so you shouldn’t be surprised when things seem to go backwards. But, flipping to the other hand, we trotted out another set of figures saying things have actually got much better (life expectancy and cancer survival rates have steadily climbed). Though, switching hands again, melanoma’s gone through the roof. However, going back to the first hand, if we can still locate it, we noted the massive explosion in the facts mountain of cancer biology for which the blue touch paper was only lit about 25 years ago.

And your parliamentary candidate is …

What with all this sleight-of-hand, flip-flopping and U-turning, it occurs to me that I’m shaping up rather well as a prospective politician. I’m quite taken with the idea, especially as if I stood as an MP in my own constituency I’d be up against Andrew Lansley who, as you’ve probably forgotten, was once upon a time Secretary of State for Health. Being a virtuous and helpful soul, when Betrayed by Nature came out I sent him a copy as a gift, a freebie, – figuring that, as a career civil servant and politician who’d become responsible for the nation’s health, he might find it useful to read a basic primer on something that was killing 150,000 UK citizens every year. Thoughtful, you’d say? Indeed. Did I expect to find him on my doorstep next day gushing gratitude and thirsting for more knowledge? Maybe not, even though he only lives round the corner and we have actually met in the dim past. But at least one might have received a note – a one line email, perhaps – from his PA, who can scarcely be too busy to be polite. But no. Nothing. Zippo. So I came up with a brief sentence that summarised my take on this example of voter wooing, or indeed plain good manners, but I can’t remember it now – for the best perhaps. What is it the Bible says about getting narked? Something along the lines of “whoever says, ‘You fool!’ shall be liable to the hell of fire.”

So thank heavens we’ve side-stepped that but nevertheless, Andrew, it really would be a joy to give you a bloody nose – electorally speaking, of course – so let’s just give those credentials one more buffing. We started by lowering your expectations of science with the reminder that things proceed at a snail’s pace {you do realise that common analogy is very unfair on snails? Scientists have shown they can bowl along at a metre an hour (yippee, we do discover things!) – not much slower than your average supermarket trolley-pusher, but here’s the thing. Snail’s pace means they can get round the garden in one night. That’s the whole of their world covered in one go – without mechanical assistance!! Not so slow after all, eh?}. But the flip side is that the genomic era has already seen the development of a number of drugs that are effective against malignant melanoma. They’re not perfect but at least they take us a step further in dealing with this cancer once it has spread around the body.

And the message?

(That’s quite enough politics, Ed). OK. Let’s abandon a promising career and go back to being a scientist with a typically punchy summary. Australia’s wonderful but when it comes to cancer it’s not much different to any other rich country (not really a flip that, just a statement of fact). Folk are living longer so, of course, more of us will ‘get’ cancer but we seem to think that longevity buys us more time to smoke, booze, burn ourselves pink and eat crappy food. Medical science is doing wonders in detection and treatment: at nearly $400 million a year on cancer research, almost a quarter of all health research expenditure in Australia, it jolly well should. But if we don’t do more to help ourselves the cancer burden is going to overwhelm health resources not just ‘down under’ but all over.

Reference

Schwartz, M.A. (2008). The importance of stupidity in scientific research. J Cell Sci 2008 121:1771; doi:10.1242/jcs.033340

 

Fancy that?

Seeing as they started 28 years ago we can hardly blame members of the Harvard School of Public Health for publishing the results of their labours in tracking 120,000 people, asking them every few years what they’ve eaten and seeing what happened to them (a ‘prospective’ study). About one in five of the subjects died while this was going on but the message to emerge was that eating red meat contributes to cardiovascular disease, cancer and diabetes. The diabetes is non-insulin-dependent diabetes mellitus (NIDDM) or adult-onset diabetes – about 90% of diabetes cases. The cancers weren’t specified, although the evidence for a dietary link is generally strongest for colon carcinoma. The risk is a little higher for processed red meat than unprocessed.

How much?

Massive, if you mean the amount of data they accumulated from such a huge sample size followed over many years. If you mean on a plate, their standard serving size was 85 grams (3 ounces) for unprocessed beef, pork or lamb) and 2 slices of bacon or a hot dog for processed red meat. One of those a day and your risk of dying from heart disease is increased by about 20 per cent and from cancer by about 10 per cent – and the risks are similar for men and women. Just to be clear, that is a daily consumption – and the authors very honestly acknowledge that ‘measurement errors inherent in dietary assessments were inevitable’. They also mentioned that one or two things other than steak can contribute to our demise.

Are we any wiser?

If you recall from Rasher Than I Thought? the risk of pancreatic cancer is increased by just under 20 per cent if you eat 50 grams of processed meat every day. This report suggests that a limit of 1.5 ounces (42 grams) a day of red meat (one large steak a week) could prevent around one in 10 early deaths. So does it tell us anything new? Not really. Was it worth doing? Yes, because it adds more solid data to that summarized in Are You Ready To Order?

And the message?

Unchanged. Do some exercise and eat a balanced diet – just in case you’ve forgotten, that means limit the amount of red meat (try fish, poultry, etc.), stick with the ‘good carbs’ (vegetables, fruits, whole grains, etc.), cut out the ‘bad’ (sugar – see Biting the Bitter Bullet), eat fishy fats not sat. fats and, to end on a technical note, don’t pig out.

 References

Pan A, Sun Q, Bernstein AM; et al. Red meat consumption and mortality: results from 2 prospective cohort studies [published online March 12, 2012]. Arch Intern Med. doi:10.1001/archinternmed.2011.2287.

Pan A, Sun Q, Bernstein AM; et al. Red meat consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis. Am J Clin Nutr. 2011;94(4):1088-1096.