More on Nanotech
Sir Frederick
-----------------------------------------------------------------
http://www.newscientist.com/channel/life/mg19626241.600-the-camel-factor-nanobody-revolution.html
The camel factor: Nanobody revolution
03 October 2007
NewScientist.com news service
Henry Nicholls
IT IS not a problem university lecturers face all that often. In the 1980s, Raymond Hamers was confronted by a couple of bold
undergraduates complaining that the practical experiments for their course were boring and predictable. Could he find them something
more original to investigate?
Hamers, an immunologist then at the Free University of Brussels (VUB) in Belgium, remembered that he had half a litre of camel blood
sitting in a freezer. Although it was earmarked for research into sleeping sickness, he figured he could spare a little for the
students. "Why don't we see if we can purify camel antibodies?" he asked them.
The results flummoxed everyone. "We couldn't believe it," Hamers says. The pattern of antibodies extracted from the blood suggested
that, in addition to the standard type found in all vertebrates, the camel produced an entirely novel, simpler variety.
What started out as a student project soon turned into a major line of research for Hamers and his colleagues. At first they assumed
the smaller proteins were merely fragments of conventional antibodies. But fresh camel blood flown in from Kenya contained the same
novel antibodies. "We spent two years checking whether we were right because it was so blooming crazy," says Hammers.
“We couldn't believe it. We spent two years checking whether we were right because it was so blooming crazy”Unusual antibodies
The team went on to demonstrate that the new kind of antibody is found not only in the dromedary camel but also in other camelids -
the Bactrian camel, llama, alpaca, guanaco and vicuńa (Nature, vol 363, p 446).
Why camelids evolved this unique type of antibody in addition to the normal kind is still a mystery, says Serge Muyldermans, a
member of the original team who is now at the Flemish Institute of Biotechnology. Unpublished experiments indicate that camels
facing a range of different challenges to their immune systems do not favour one type over the other, he says.
However they evolved, the antibodies are far more than an immunological curiosity. From the start, Hamers knew he was onto something
special. While normal antibodies have huge medical potential, the sheer size of these bulky proteins is a problem for all sorts of
reasons, which is why many teams have been trying to create smaller versions.
What makes camel antibodies so special is not that they are somewhat smaller than conventional antibodies, but that their key
component - the variable portion that binds to other proteins - works fine all by itself. That means you can produce even smaller
proteins, dubbed nanobodies by Muyldermans, that can do almost everything normal antibodies do - and some things they cannot - yet
are just a tenth of their size.
First trial
Nearly two decades after that student project, the results of the first trial of a nanobody-based therapy are due out soon. And
before long we could be hearing much more about them thanks to their many advantages. Nanobodies can get to parts of the body and
parts of molecules that conventional antibodies cannot. They can also be attached to a toxin or other molecules without the whole
creation becoming impossibly large.
That's not all. Nanobodies are tougher than normal antibodies and could be swallowed to treat gut diseases without being digested,
for instance. They are so much easier and cheaper to make that it might be feasible to add them to consumer products, such as
anti-dandruff shampoos.
They can be engineered into plants and animals to produce all sorts of effects, from revealing the inner workings of cells to
altering metabolism. They might even be the ideal tool for creating miniature chemical laboratories.
Normal antibodies consist of Y-shaped proteins in which the structure of the tail stays constant while the tips of the arms vary
greatly, allowing different antibodies to bind to different targets. If a new virus invades our bodies, for instance, within days
our immune systems can generate antibodies that bind specifically to that virus and help destroy it.
Endless potential
Back in the mid-1970s, when scientists first worked out how to manufacture antibodies to order, there was great excitement. They
seemed to have endless potential as therapies for everything from infectious diseases to cancer. Conventional drugs consist of a
huge variety of small molecules that can have all sorts of unexpected side effects.
With antibodies, there is much less concern about toxicity, as our bodies produce them already. Their size can be an issue, though.
For instance, they are too large to be absorbed into the blood from the gut and thus have to be injected rather than swallowed.
Despite all the initial excitement, producing human-like antibodies that our immune system will accept proved to be rather tricky.
Only now, decades later, are antibodies really starting to deliver on their promise, with around 20 on the market and hundreds more
in the pipeline.
Even so, creating and manufacturing new therapeutic antibodies remains a challenging and expensive process, not least because the
molecules' size and complexity mean they can only be produced by mammalian cells. Hence the demand for smaller proteins that act
like antibodies.
Stripped-down antibodies
The simplest approach is to strip down normal antibodies. These consist of two heavy chains and two light chains, but only the ends
of the chains contain the variable regions that bind to other proteins (see Diagram). So why not chop off the variable regions, or
domains, and use these as "single-domain antibodies"?
This is just what Gregory Winter of the UK Medical Research Council's Laboratory of Molecular Biology in Cambridge did in the late
1980s. But his team ran into a serious obstacle. Since the heavy chain of human antibodies is usually bound to a light chain, the
proteins that form human single-domain antibodies are naturally sticky, and have a tendency to clump together and to bind to
proteins other than the target, making them useless. "We had to overcome this and find ways of giving them good biological
properties," says Winter.
While Winter works on solving the issues with human single-domain antibodies - several drugs based on them are being developed and
the company he co-founded was bought by GlaxoSmithKline earlier this year - other groups are using larger fragments of antibodies to
avoid the stickiness problem.
What makes the streamlined camelid antibodies so special, however, is that there is no stickiness problem with the single-domain
antibodies - nanobodies - derived from them. This is because they lack light chains altogether, so the variable domain at the end of
the heavy chain has evolved to work in isolation, rather than to stick to an adjoining light chain.
First find a camel
So creating nanobodies is relatively easy. The first step is to inject camels or llamas with whatever it is you want the nanobody to
bind to - a virus, say. A few days later you take some blood and identify the white blood cells that are making antibodies that bind
to the target, then extract the DNA that codes for the variable domain.
For many purposes, this is all that's needed. Nanobodies intended for injection into the human bloodstream may require some tweaking
to ensure they do not appear foreign and hence provoke an immune reaction, but they are already very similar to the variable domain
of human antibodies.
The big advantage of single-domain antibodies is that these proteins are simple enough to fold correctly when they are made in
genetically engineered bacteria or yeast. These methods of production are so much cheaper than using mammalian cells that nanobodies
could be used in a host of ways unthinkable for normal antibodies, such as in anti-dandruff shampoos.
Nanobodies are also pretty tough, partly thanks to extra internal bonds that reinforce their structure. "You can use them in very
harsh conditions where normal antibodies collapse, get digested or don't fold," says Muyldermans. That not only makes them easier to
store and transport than conventional antibodies, it means some nanobodies can survive a journey through the gut, raising the
prospect of nanobody pills for treating gut disorders such as inflammatory bowel disease or colon cancer.
A major killer
Another possibility is treating diarrhoea caused by rotavirus, a major killer of children in developing countries for which the only
current treatment is rehydration. Nanobodies that bind to rotaviruses have been shown to cut the death rate in mice. Children could
be dosed with live bacteria that churn out these nanobodies, making the treatment cheap.
There's more. Nanobodies open up a wealth of targets that larger antibodies cannot reach. They are small enough to wheedle their way
into the active sites of enzymes, deep clefts in receptors on the surface of viruses and bacteria, or into the heart of dense
tumours. It looks as if they might even penetrate the blood-brain barrier effectively enough for drug designers to think about
adapting them to treat conditions like Alzheimer's disease.
The small size of nanobodies and other single-domain antibodies does have one big disadvantage, though: "In practice, these things
get pissed out very fast, usually too fast to be of any therapeutic use," says Winter. This means nanobodies last just hours in the
bloodstream compared with up to three weeks for conventional antibodies.
The answer is to make them bigger by linking them to other molecules. While this can sacrifice some of their unique advantages,
these hybrid proteins are often exactly what's needed. Combining two identical nanobodies can make them bind more tightly to a
target, for instance.
Alternatively, two different nanobodies can be combined to create a protein capable of bringing together the target and a killer
cell from the patient's immune system (the same function is carried out by the "tail" of full-sized antibodies).
Hybrid proteins
For cancer treatments, a nanobody can be linked to an "effector molecule" that kills cells, such as a toxin, an enzyme or a
radioactive substance. In 2004, Muyldermans's team created a nanobody that binds to the surface of many tumour cells. They then
engineered bacteria to produce the nanobody linked to an enzyme called lactamase, which converts a harmless prodrug into a potent
killer of cells.
The idea was that the nanobody part would hold the enzyme close to the surface of the cancer cells (Cancer Research, vol 64, p
2853). "It was incredibly effective in animals," says Muyldermans.
For now, though, the company set up by the VUB in 2002 to develop nanobody-based healthcare products - Ablynx - is focusing on other
projects. The most advanced is an anti-clotting drug known as ALX-0081.
While there are already drugs on the market designed to reduce the likelihood of clotting in patients at risk of heart attacks or
strokes, many interfere with the clotting process everywhere in the body, which can result in bleeding. ALX-0081, however, consists
of two linked nanobodies that bind to a protein called von Willebrand factor, which is only involved in clotting where the blood is
moving fast. This means it should keep fast-flowing blood in arteries free of dangerous clots without affecting clotting elsewhere.
The signs are that ALX-0081 is going to breeze through its first trial, which began in March. Early results from healthy volunteers
suggest that it produces the desired effect without causing any serious side-effects, says Edwin Moses, head of Ablynx. The full
results should be released before the year is out.
Intrabodies
Ablynx is developing several other nanobody-based drugs, including one for rheumatoid arthritis and another designed to home in on a
target inside the brains of people with Alzheimer's. But there are plenty of other exciting ways in which these proteins could be
put to good use, says Hamers: "The applications are endless."
Before he retired a decade ago, Hamers explored the possibility of using nanobodies to study neural development. His idea was to
genetically engineer organisms so they produced nanobodies inside their cells. "Intrabodies" that targeted and inhibited specific
proteins would reveal the role of these proteins in laying down the nervous system.
This has now become a reality. "We can now express antibodies stably inside the cell to block or choke any protein," says Siyaram
Pandey, a biochemist at the University of Windsor, Ontario, in Canada.
Another group has shown that fluorescent nanobodies - dubbed chromabodies - can be used to light up specific proteins inside living
cells (Nature Methods, vol 3, p 887).
The applications of intrabodies are not limited to research. Since nanobodies are far simpler than full-blown conventional
antibodies, they can also be engineered into plants. Nanobodies that target specific enzymes have been shown to alter metabolic
pathways in potatoes (Nature Biotechnology, vol 21, p 77). This approach might also make it possible to endow crops with immunity to
specific pests or to create vegetables that combat gut infections when eaten.
Hamers's vision does not end there. Nanobodies that bind to different parts of a surface - on a crystal lattice, say - could be used
to align different enzymes into a carefully coordinated sequence. "This would be like creating a miniature chemistry laboratory
capable of performing a series of incredibly efficient reactions," he says.
For now, this is a futuristic vision. But who knows? If nanobodies and other single-domain antibodies realise their promise, one day
we could find these talented little proteins are everywhere.
Henry Nicholls is a science writer based in London, and author of Lonesome George: The life and loves of a conservation icon
(Palgrave Macmillan, 2006)
From issue 2624 of New Scientist magazine, 03 October 2007, page 50-53
Shark attack
Camels and their relatives are not the only vertebrates with an unusual, streamlined kind of antibody. In 1995, nurse sharks were
found to sport similarly slender antibodies for targeting invading viruses and bacteria.
A few years later, Stewart Nuttall, a microbiologist at CSIRO Health Sciences and Nutrition in Melbourne, Australia, decided to see
how widespread the phenomenon is. His team started by collecting blood samples from the spotted wobbegong, a flattish,
bottom-dwelling shark that can grow to over 3 metres long. They too had unusual antibodies similar to those in camelids (Molecular
Immunology, vol 38, p 313).
It now looks as though all sharks have them. "These antibodies are present in every shark species that we have tested, though for
obvious reasons we haven't gone near great whites," says Nuttall. "Genome sequencing initiatives suggest that the antibodies may
also be present in rays."
This suggests that sharks evolved these unusual antibodies soon after they diverged from other vertebrates more than 400 million
years ago. The similarity between the antibodies in sharks and camels is almost certainly an example of convergent evolution, with
natural selection arriving at the same solution on separate occasions.
Small proteins derived from shark antibodies are already being developed for various purposes: for example, to use in sensors
designed to detect cholera. Technically known as "shark new antigen receptor variable domains", they should of course be called
sharkbodies.