Taking a Swiss Army Knife to Cancer

Murder is easy. You just need a weapon and a victim. And, I guess the police would add, opportunity. I hasten to point out that’s an observational note rather than an autobiographical aside. It’s relevant here because treating cancer is intentional homicide on a grand scale – the slaughter of millions of tumour cells. For individuals we cannot say whether the perfect murder is possible – how would we know – but on the mass scale history shows that even the most efficient machines for genocide have been, fortunately, less than perfect. In other words through incredible adaptability, ingenuity, determination and sheer will power, some folk will survive even the most extreme efforts of their fellow men to exterminate them. Cancers tend to mirror their carriers. With only rare exceptions, whatever we throw at them in attempts to eliminate their unwelcome presence, some of the little blighters will dodge the bullets, take a deep breath and start reproducing again. ‘What doesn’t kill you makes you stronger’ as the American chanteuse Kelly Clarkson has it – though to be fair I think she pinched the line from Nietzsche.

Multiple whammys

For cancer cells dodging extinction requires adaptability: using the flexibility of the human genetic code enshrined in DNA to change the pattern of gene expression or to develop new mutations that short-circuit the effects of drugs – finding many different ways to render useless drugs that worked initially. You might draw a parallel with the idea, correct as it turned out, of the prisoners who staged The Great Escape from Stalag Luft III in World War II that if they initially dug three tunnels simultaneously (Tom, Dick and Harry) the guards might find one but they’d probably not find all three.

An approach increasingly permeating cancer therapy is how to target several escape routes at once – can we at least give the tumour cell a serious headache in the hope that while it’s grappling with a molecular carpet bombing it might be more likely to drop dead. One way of doing this is simply to administer drug combinations and this has met with some success. However, for the most part, agents are not specific for tumour cells and their actions on normal cells give rise to the major problem of side effects.

Step forward Yongjun Liu and colleagues from Shandong University, Jinnan, People’s Republic of China and the sophisticated world of chemistry with efforts to fire a broad-shot that combines different ways of killing tumour cells with at least some degree of specific targeting.

Making the bullets

These chemists are clever chaps but, taken one step at a time, what they’ve done to make a very promising agent is simple. The game is molecular Lego – making a series of separate bits then hooking them together. The trendy name is click chemistry, a term coined in 1998 by Barry Sharpless and colleagues at The Scripps Research Institute, to describe reactions in which large, pre-formed molecules are linked to make even more complex multi-functional structures. You could describe proteins as a product of ‘click chemistry’ as cells join amino acid units to make huge chains – but you wouldn’t as it’s better to keep the name for synthetic reactions that make novel modules.

It might help to recall some school chemistry:

acid + base = salt + water (e.g., HCl + NaOH = NaCl + H2O)

Click chemistry is the same idea but the reactants are large molecules, rather than atoms of hydrogen, chlorine and sodium.

Anti-freeze to anti-cancer in a couple of clicks

The starting point here is remarkably familiar – it’s antifreeze, a chemical added to cooling systems to lower the freezing point of water (e.g., in motor engines). Antifreeze is ethylene glycol (two linked atoms of carbon with hydrogens: HO-CH2-CH2-OH): make a string of these molecules and you have a polymer – poly-ethylene glycol (PEG).

For click chemists it’s easy to tag things on to biologicial molecules, including PEG and most proteins. This study used biotin – a vitamin that works like a molecular glue by sticking strongly to another small molecule called avidin, found in egg white. Avidin can therefore be used to fish for anything tagged with biotin – it simply hooks two biotins together. The protein used here is an antibody that binds to a signaling molecule (VEGFR) present on the surface of most tumour cells and blood vessels. VEGFR helps tumour growth by providing a new blood supply – an effect blocked when the antibody binds to it.

Sounds familiar?

If chains of carbon atoms decorated with hydrogens seem familiar, so they should. They’re fats (the saturated fats you get in cream and butter are very similar to the chains of PEG). As anyone who’s done the washing up knows, fats and water don’t get on (which is why we have detergents). Put them in water and fats huddle together in blobs called micelles – sacs of fat. This gives them a useful property: if you mix something else in the water – a drug for instance – and then add PEG and separate the micelles that form, you’ve got drug trapped in a kind of carrier bag. Often called nanoparticles, these small, molecular bubbles made by chemists are packets of drug ready to be delivered.

Micelle Blog picA sac of poly-ethylene glycol (PEG) with entrapped drug (red dots) tagged in three different ways (Liu et al., 2014).

Addressing the parcel

To turn PEG into a parcel two chemical tricks are needed. The first is to tag PEG with biotin. Now the nanoparticles will pick up VEGFR antibody labeled with avidin – and the antibody label can target the micelles to tumour cells and blood vessels.

Exploding the package

The second trick is the addition of another polymer (a chain of histidine amino acids) that triggers the disassembly of the nanoparticles when they find themselves close to or inside tumour cells – a more acidic environment than the circulation.

Seeing the results

The final twist is to include another modified PEG – this with a chemical group that binds gadolinium when it’s added to the water. Gadolinium is an ion (Gd3+) which shows up brightly in MRI scans – the idea being to highlight where the nanoparticles end up after injection into animals.

Does it work?

These multicomponent nanoparticles resemble a Swiss Army knife – all sorts of gadgets sticking out all over the place: PEG to make sacs that contain a drug, biotin hooked to VEGFR antibody to home in on tumour cells, an acidity sensor so the thing falls apart and releases its content on arrival and a contrast enhancer that shows up where this is happening in an MRI scan.

Injected into mice with liver tumours, these multi-functional nanoparticles do indeed home in on the tumours and their surroundings and drastically reduce tumour growth when they carry the drug sorafenib. Sorafenib is the only agent that has been shown to affect liver cancers, although its effects are brief. Compared to sorafenib alone, these new nanoparticles are about three times more potent – presumably because of their targeted delivery.

Where are we?

This wonderfully clever chemistry will not cure liver cancer. A good result when it reaches human trials would be six months remission by comparison with the current average of two months from treatment with sorafenib alone. But what it does show is that hitting cancers hard in multiple ways at least slows them down. We can only hope that more potent drugs and further ingenuity will progressively extend this capacity. The end is not in sight but brilliant technical advances such as that from Yongjun Liu’s lab may be spotlighting the way ahead.


Yongjun Liu et al., (2014). Multifunctional pH-sensitive polymeric nanoparticles for theranostics evaluated experimentally in cancer. Nanoscale 6, 3231-3242.


Spray Painting Cancer

I’m pretty certain that anyone reading this will be fully aware that one of the biggest problems in cancer is spotting the blighters. We have, of course, X-ray detection (as in mammography), CTs and MRI scans, all so familiar we need not bother to define them, and there’s also a variety of sampling methods for specific cancers (e.g., the Pap test for cervical cancer). But, useful though all these are, the plain fact of the matter is that none are ideal and in particular the pictures created by imaging methods are very limited in sensitivity. Put another way, they won’t pick something up until it is quite large – a centimeter in diameter – meaning that the abnormal growth is already quite advanced.

Cunning Chemistry

Needless to say, much inspiration and perspiration is being applied to this matter and what has been really exciting over the last ten years or so is the way very smart chemists are collaborating with clinicians to come up with new ways of looking at the problem. One of these clever tactics is being developed in the University of Tokyo using a different type of imaging ‘reporter’ that signals its presence by fluorescing. Fluorescence occurs when a molecule absorbs light and becomes ‘excited’ before relaxing back to its ‘ground state’ by giving off a photon. Fluorescent molecules (fluorophores) are much used in biology because the background signal is often very low so the high signal-to-noise ratio gives excellent sensitivity.

Spray Paint scheme

The cell-surface enzyme GGT converts the small molecule  gGlu-HMRG to a fluorescent form (HMRG) that is then taken up by the cell. GGT is only found on tumor cells so they light up and normal cells do not

Fortunately we don’t need to know how the chemists did it – merely to say that Yasuteru Urano and his colleagues came up with a small molecule (called gGlu-HMRG for short) that does not give off light until a small fragment is chopped off its end, whereupon it changes shape: this flips the switch that turns on fluorescence. The cutting step needs an enzyme that is found on the surface of various cancer cells but not in normal tissue (GGT for short).

Joining Forces

To show that there was real mileage in their idea they followed the time-honored blue-print of cancer research, showing first that it works on tumor cells grown in the lab (and, equally important, that it doesn’t highlight normal cells), before moving to mouse models of ovarian tumors. The later is where chemists meet clinicians because an endoscope is required (quite a small one) – a flexible tube for looking inside the body – devices now so sophisticated that they can incorporate a fluorescence camera.

In the final synthetic step the cunning chemists formulated a spray-on version of their probe molecule so that it can be dispensed during endoscopy or surgery – a bit like an underarm deodorant. Now it’s easy: find suspect tissue, give it a squirt of gGlu-HMRG, wait a few minutes and see if it lights up. The answer is, of course, that in their ovarian cancer model the spray-on graffiti lights up within 10 minutes of sticking to a tumor cell and can detect clumps of cells as small as 1 millimeter in diameter – a terrific advance in terms of sensitivity. The brief time taken for the signal to be visible after the probe has been applied means that within the same procedure it could be used to guide surgeons in removing small tumor masses.

The Tokyo system is not the only one under development. My colleague Andre Neves at the Cambridge Cancer Centre, another of these fiendishly clever chemists, is working on a parallel line using different fluorophores that can be topically applied to the lining of the intestine. The goal here is, of course, the early detection of colon tumors. Yet other approaches use molecules that accumulate preferentially in tumor cells and respond to light in the near-infrared region of the spectrum (800 nm to 2500 nm wavelength, compared to just under 500 nm for gGlu-HMRG), giving an even better signal-to-noise ratio.

This is, as Mr. Churchill might have pointed out, not even the beginning of the end of this story. But it is one more small and innovative step forward. Not all cancers even of the same type will be detectable by a given probe because they vary so much in the genes they express but the ingenuity of the chemists gives hope that a substantial panel of ever more sensitive reporters will emerge. It is also true that endoscopy is unlikely to gain widespread popularity as a routine screening method. However, these advances, moving us to detection at ever earlier stages may become very powerful as a follow-up test, combined with the capacity for simultaneous treatment, when tumor cells have been detected in more comfortable screens, for example as circulating cells in small blood samples, an immensely exciting prospect to which we will return in a later episode.


Urano, Y., Masayo Sakabe, Nobuyuki Kosaka, Mikako Ogawa, Makoto Mitsunaga, Daisuke Asanuma, Mako Kamiya, Matthew R. Young, Tetsuo Nagano, Peter L. Choyke, and Kobayashi, H. (2011). Rapid Cancer Detection by Topically Spraying a γ-Glutamyltranspeptidase–Activated Fluorescent Probe. Science Translational Medicine 3, 110ra119.


Shi, C. (2012). Comment on “Rapid Cancer Detection by Topically Spraying a γ-Glutamyltranspeptidase–Activated Fluorescent Probe. Science Translational Medicine 4, 121le1.