Lewis/Life - On-line Resource Articles

Technology Makes Additional Inroads in the Clinic and in the Field

Past looks at PCR applications in GEN have highlighted the techniqueís amazing versatility. From identifying dead royals, to spotting protozoans in water supplies and bacteria in food, to nailing moose poachers and cataloging maggot species in corpses, PCR has been just about everywhere.

Now scientists are using PCR to track HIV infection and to select the most genetically fit cows and corn. For example, with recent rapid drug approvals and people staying healthy longer with HIV, the AIDS picture is finally beginning to brighten. And PCR is becoming a key player in tracking disease progression and in evaluating new antivirals.

PCR can measure viral burden, which is a more direct measure of infection than detecting an antibody response or symptom onset. PCR has been critical in evaluating a new drug, Pharmacia & UpJohn Inc.ís (London U.K., and Kalamazoo, MI) reverse transcriptase inhibitor delavirdine mesylate (Rescriptor). The company recently reported preliminary results of clinical trials.

The multicenter trials, begun in April 1994 in North America, Europe and Australia, are tracking 1,900 HIV-positive individuals. One trial assigned participants to two sub-groups, one receiving delavirdine and ddI, and the other ddI alone. The second trial includes subgroups receiving zidovudine and delavirdine or zidovudine alone.

"We used RNA PCR of plasma to detect free circulating virus and also DNA PCR of sorted peripheral blood mononuclear cells (PBMC) to look for integration of the viral genome," says Donald Batts, M.D., clinical research manager at Pharmacia & UpJohn. "Then we correlated those PCR measurements of viral burden with all the previously known surrogate markers of outcomeñCD4 cells, p24 antigen and quantitative viral culture. We correlated all those surrogates, PCR and non PCR, with clinical outcome," he adds.

Effective Combinations

If a PCR-based measurement of viral burden predicts disease progressionñas the preliminary results of the clinical trial suggestñpatients could receive effective drug combinations earlier in the game. "This would save time, lives, shorten duration of clinical trials, get more drugs to market quicker, and would provide more reassurance to physician and patient that the person is on effective therapy," Dr. Batts says. In the past, a physician would often wait until a patientís CD4 count dropped to begin therapy.

PCR to detect viral burden is becoming ever more powerful. "We are refining the PCR assay all the time. At first we got it down to detecting 20,000, then 2,000, now 40-200 copies per cc (of plasma). If you have viremia below the detection level of the assay, almost always you will be a long term survivor," Dr. Batts says.

So far PCR has indicated a passing grade for delavirdine. "The data show that many patients were able to achieve a sustained drop of viral burden of at least 68 percent as a result of anti-retroviral therapy. All patients who achieved a sustained reduction in viral burden were less likely to progress to AIDS-defining events or deaths," according to William Freimuth, M.D., Ph.D., medical monitor for the trials at the company.

At the University of South Florida in St. Petersburg, Robert Nelson Jr., M.D., and colleagues are using PCR to detect HIV in newborns whose mothers are HIV positive. Studies show that more than 90% of HIV-infected children acquired the virus before or just after birth, yet early detection, necessary to begin early treatment, is complicated by lack of symptoms in babies and the immaturity of their immune responses. But even though symptoms and antibodies may not be present in infected infants, HIV may be there. Enter PCR.

The researchers evaluated 286 HIV-exposed infants in a blinded fashion, following them from one to four weeks of age until at least 18 months of age, when antibody tests are meaningful. They used an antibody test, evaluation of symptoms, two direct tests (p24 antigen detection and HIV culture), and PCR utilizing primers against two viral sequences. PCR detected 95% of the infected babies, and ruled out 100% of those not infected. Detection accuracy increased to 100% if the babies were 5 weeks of age or older.

PCR is superior to direct viral culture both in time and cost, taking 3 to 10 days for $175, compared to 3 to 4 weeks and $300 to $600 for viral culture. "PCR has been considered an investigational test. We think these data show itís time PCR be recognized as a standard clinical test for HIV detection," says Dr. Nelson.

Bull Market

Many assisted reproductive technologies that help humans become parents have evolved from work on barnyard residents. Ironically, the StockMarks for Cattle kit for tracing cattle paternity, introduced by Perkin-Elmerís Applied Biosystems Division (Foster City, CA), is the reverseñit adapts PCR-based paternity testing in humans to cattle.

The StockMarks kit fluorescently labels and detects 12 bovine microsatellite sequences. The kit uti-process, an ABI373 DNA Sequencer, and GENESCAN and Genotyper software. The kit is a first in two waysñit is the first commercially available PCR kit for DNA testing in cattle, and it is the companyís first foray into the agricultural market.

Genotyping cattle is used to trace lineages and construct pedigrees, and to evaluate specific traits of agricultural import. These include milk production and composition, leanness, fertility and ease of birthing. "In cattle breeding, it is important to keep accurate ancestry records. Board associations require verification tests to prove a calf comes from a particular sire and dam. Serotyping and HLA testing have been done for the past 10 to 15 years, and now weíve moved on to PCR," says Stephen Bates, Ph.D., group leader for agriculture applications in the Applied Biosystems Division.

Some 400,000 cattle undergo conventional blood tests to determine lineage each year. In the cattle industry, a bull with valuable characteristics is identified by the quality of his offspring, and then his semen is frozen and widely distributed to artificially inseminate cows. But it can take 5 to 7 years for a bullís progeny to be mature enough to evaluate their traits. The cost of determining a single bullís fitness can be $50,000.

StockMarks will speed the time it takes breeders to conduct the artificial selection that has been the basis of agriculture since its beginnings 10,000 or more years ago. Users can specify which genes to detect, and then use the information to set up matings. For example, a Brahman bull from southeast Asia has genes conferring tolerance to heat that would be valuable in a Holstein cow in Texas with high milk yield, but poor heat tolerance.

StockMarks is only the beginning of PCR-based cattle testing. An ongoing bovine genome project is identifying quantitative trait loci, or QTLs. Such polygenic traits typically confer continuously varying phenotypes. With multiple genes, each with multiple variants to track, the need for an automated detection process will mount. "The time is ripe for DNA testing to move into agriculture, particularly for cattle. DNA analysis is far more accurate, easily automated, cheaper and less labor-intensive," says Dr. Bates. PCR-detection of valuable traits for breeding purposes may soon be applied to thoroughbred horses and purebred dogs, he adds.

Predicting Corn Yield

Whether considering cows or corn, it helps to know which individuals are most likely to yield valuable offspring. Researchers at Dekalb Genetics Corp. (Dekalb, IL) are also using PCR for this purpose. The company recently received a patent on the application of DNA marker technology to predicting corn yield. The patent covers several methods for this application, including restriction fragment length polymorphisms, random amplified polymorphic DNA, microsatellite analysis and use of other PCR-derived markers.

A corn breeder with luck can coax three corn generations a year to grow in a greenhouse, or two in the field, in warmer climates. Evaluating yield in these conventional settings takes months, and because many genes contribute to this characteristic, the task is complex, with many plants not of the desired phenotype. Moving in a molecular marker direction will allow breeders to evaluate germplasm rather than full-grown plants, saving much time. "Using computer analysis, breeders can predict yield without actually growing the plant. Only the germplasm with the highest predicted yield is then selected for evaluation in the field," says Bruce P. Bickner chairman and CEO at DeKalb. Breeders will be "able to screen a far greater pool of germplasm than they could ever grow in the field," he adds.

By Ricki Lewis, Ph.D.

Back