Making super-sensitive antibodies with help from camels and sharks

What do camels and sharks have in common? A remarkable, binary immune system that quells viruses and other infectious nasties with an efficiency that the human immune system cannot match, according to CSIRO molecular immunologist Dr Peter Hudson.

Hudson told the annual Lorne Conference on Protein Structure and Function on Phillip Island that new technologies that allow researchers to fast-forward the evolution of shark and camel antibodies promise to lead to super-sensitive new diagnostic reagents for medicine, and antibody-derived therapeutic molecules for intractable infectious diseases.

For all the promise of monoclonal antibody (mAb) therapeutics, and design advances that have 'humanised' them for safe clinical use Hudson said relatively few have been commercialised - although some, like the rheumatoid arthritis remedies Humira (Abbott) and Remicade (Centacor) have been very successful.

A major obstacle to antibody therapeutics, Hudson said, is that human and other mammalian antibodies - camels excepted - are concave in shape at their business end, and have stubby, finger-like loops adapted to grasping relatively smooth, curved surfaces on antigens - they also mould efficiently to smaller antigens.

But their stubby "fingers" find it very difficult to insert themselves and get a grip on deep clefts or pockets on antigenic surfaces, that the influenza and AIDS viruses and others have evolved to conceal the active sites of key enzymes against antibody attack.

Without access to these active sites, antibodies cannot efficiently protect the body against infection.

Hudson said sharks and camels also make these types of concave-ended antibodies, but both have independently evolved a second immune system that produces unique, smaller antibodies with long, finger-like loops projecting from their active domains, that can plumb the depths of antigenic canyons and pockets.

Sharks are a very ancient and successful groups that pre-date the dinosaurs, and Hudson said it is possible that this type of long-loop antibody, called an IgNAR antibody, may be ancestral to the mammalian immunoglobulin (Ig) antibodies.

Camels also have both types, and may have independently re-invented IgNAR antibodies after ancestral mammals dispensed with them.

Sharks injected with familiar antigens will preferentially make IgNAR antibodies instead of classical Ig antibodies - "Some big sharks around are making incredibly sophisticated little molecules that give them very powerful immune surveillance," said Hudson.

The surfaces of shark IgNAR active domains, unlike mammalian antibodies, are highly hydrophilic, and are stabilised by sulfide bonds that give them exceptional stability and longevity in serum - a very desirable property for many antibody therapeutics, Hudson said.

Hudson's team has sequenced the peptide sequences of the variable, active domains of IgNAR antibodies of sharks and camels, and found they are very similar to neural cell-adhesion molecules (NCAMs) in humans, which also have long "sticky" loops that allow them to glue neurons together.

Human NCAMs are already tolerated by the human system, so they can be used as molecular scaffolds on which to mount the active domains of selected shark or camel IgNAR antibodies for therapeutic use - avoiding the violent anaphylactic reactions that prevented early, mouse-derived monoclonal antibodies being used in humans.

Hudson's team has studied the structure of the IgNAR active domains, and is experimentally creating longer, more complex structures containing "sticky", sulfur-rich cysteine residues - antibody mimetics that will be even more potent and selective than native IgNARs.

They are also experimenting with multi-headed "hydra" molecules with trimeric or even tetrameric structures in place of the usual two-headed IgNAR structures.

These antibody mimetics can be progressively "moulded" to target antigens by the rapid-evolution ribosome-display technology patented by Australian designer-protein company Evogenix, with which Hudson's group is working, or by phage display antibody technology

In both cases, the gene encoding a selected antibody is copied by a 'sloppy' viral polymerase enzyme that introduces random errors into the gene's DNA code, creating thousands of subtle variations on the original antibody theme.

The gene is expressed to produce the antibody, which remains linked to the genetic 'recipe'.

Antibodies that bind even more avidly to the target antigen are purified, with gene 'recipes' still attached, for further rounds of mutation and antibody selection.

Eventually the antibody makes a tight, high-affinity fit with the antigen, and is ready for testing as a potential diagnostic or therapeutic.

Hudson said the mammalian antibody repertoire is limited to around 100 million different antibodies. Phage and ribosome display technology, with added peptides, increases the range of unique antibody conformations by six orders of magnitude, creating effectively limitless possibilities to design antibodies to fit any protein or peptide antigen in nature.

His CSIRO team, working with Latrobe University researchers, has used the technology to evolve a "spectacularly sensitive" antibody to the anthrax bacterium's lethal factor protein, that could incorporated in a bio-electronic sensor to provide instant warning of terrorist biowarfare attacks with the deadly microbe.