LA JOLLA, Calif. -- In five years, the test tube will be mightierthan the mouse for generating monoclonal antibodies, accordingto a molecular biology team at Scripps Research Institute here,headed by the institute's president, Richard A. Lerner.

Today's issue of Science carries a two-page "perspective" titled"Antibodies Without Immunization," which describes the team'sstrategy and tactics to genetically engineer monoclonals fromscratch rather than from mice, as they are now.

"Hopefully, within five years we will have developed geneticsystems in E. coli that mimic the immune system to such anextent that there will be no need to immunize animals,"molecular biologist Carlos F. Barbas III, a principal co-author ofthe Scripps paper, told BioWorld.

What drives this "major goal" is not placating the animal rightsactivists, Barbas explained, "but not having to rely on theanimal rather than the researcher deciding which antigenicepitopes on a protein it (the animal) is going to react to. Wewant to control these processes so we can develop highlypotent therapeutic and catalytic antibodies."

Lerner is not alone in this ambitious endeavor. His main -- andperhaps only -- rival is protein engineer Greg Winter of theMedical Research Council in Cambridge, England.

Two months ago, in the Journal of Molecular Biology, Winterpublished a paper titled "By-passing Immunization." And onAug. 15 he published in the Journal of Biological Chemistry a"minireview" subtitled "Mimicking the strategy of the immunesystem." These articles updated research toward a totally test-tube monoclonal.

The trans-Atlantic contenders pursue parallel procedures intheir in vitro race to synthesize antibodies.

Lerner told BioWorld that one of his lab's as-yet-unpublishedachievements was to create HIV-neutralizing monoclonals thatbind the AIDS virus. To demonstrate the concept, he and hiscolleagues took blood from one of their staff members who hadreceived anti-tetanus shots years ago and isolated his toxoid-targeting antibodies. Then they randomized a portion of thegene sequence that encoded one of the antibody'shypervariable heavy-chain regions -- which actually makecontact with antigens -- and re-selected from the resultinglibrary active sites capable of binding to an array of quitedifferent antigens.

"It's the variability in these loops, of which there are six on anantibody," Barbas explained, "which allows the immune systemto recognize any sort of foreign invader by simply varyingthese loops."

Using this methodology, Barbas said, "we've been able to derivefairly good affinity antibodies, which bind a number ofdifferent haptens as well as some proteins."

So have Greg Winters and his colleagues in England.

A simplified account of how both are tackling their mammal-free antibody goes like this:

1. Choose regions on the antibody loops involved in antigenrecognition, and randomize them over a specific area, usingerror-prone polymerase chain reaction (PCR) amplification tomimic somatic mutation, and obtain a wide diversity ofsequences.

2. Make libraries of more than 100 million antibody sequences.This is an order of magnitude greater than the 10 millionantibodies a mouse can clone.

3. Using a filamentous bacterial virus, or phage, to mimic theimmune system's antibody-generating B cells, geneticallyengineer these phages to produce an antibody on their surfacecoats -- a different one on each phage.

4. Inside their envelopes, the viral DNA replicates its uniqueantibody in quantity.

5. Expose and bind the panoply of antibody-studded phages toantigens of interest.

6. Infect E. coli host cells with these antigen-antibodycomplexes, and let them multiply.

Patrick C. Kung, a founding father of hybridoma technology,called Lerner's approach "a major theoretical advance inimmunology." Kung, chief scientific officer of T Cell Sciences Inc.of Cambridge, Mass., in the mid-1980s developed OKT3, thefirst therapeutic monoclonal antibody to enter clinical practice.

"Based on Dr. Lerner's studies, progress to date and the planthey outlined," Kung said, "it is just a matter of time untilscientists will be able to duplicate the entire process ofantibody production in vitro." Meanwhile, Kung sees someoutstanding problems:

-- The need for a purified antigen, while in an immunizedanimal even whole cells can be used as immunogens;

-- the still-unknown immunogenicity of synthetic antibodiesintended for therapeutic treatment.

"Depending on cost and other factors," Kung concluded,"laboratory rodents will have an important role in producingantibodies for many years to come."

-- David N. Leff Science Editor

(c) 1997 American Health Consultants. All rights reserved.