Correlation between amount of virus with altered nucleotide sequence and the monkey test for acceptability of oral poliovirus vaccine.

Production of live attenuated oral poliomyelitis vaccine (OPV) requires rigorous neurovirulence safety testing of each vaccine lot, currently carried out in monkeys. It has been reported that a change from 472-U to 472-C in the type 3 OPV RNA is associated with an increased histologic lesion score produced upon intraspinal inoculation of the mutant virus in monkeys. We have developed a method, based on polymerase chain reaction, for measuring the relative abundance of these mutant sequences directly in vaccine preparations and used this method to evaluate the proportion of 472-C in 40 different lots of type 3 OPV. Six vaccine lots that had failed the intraspinal monkey neurovirulence test contained a higher proportion of 472-C than all other lots that had passed this test. OPV type 3 virus containing 472-C was rapidly selected during serial passages in African green monkey kidney cells that are used for manufacturing of the vaccine. We have also found that the wild-type poliovirus type 3 strain Leon/37, from which the vaccine strain was originally derived, contained a mixture of 472-U and 472-C sequences. No other mutations in OPV type 3 RNA have been detected by similar assays at position 2034, also associated with attenuation, or at several other positions reported to be altered in some vaccine preparations. Our results suggest that molecular diagnostics may provide a supplement or a potential alternative to animal testing of live attenuated vaccines.

Quantitative analysis of MDR1 (multidrug resistance) gene expression in human tumors by polymerase chain reaction.

The resistance of tumor cells to chemotherapeutic drugs is a major obstacle to successful cancer chemotherapy. In human cells, expression of the MDR1 gene, encoding a transmembrane efflux pump (P-glycoprotein), leads to decreased intracellular accumulation and resistance to a variety of lipophilic drugs (multidrug resistance; MDR). The levels of MDR in cell lines selected in vitro have been shown to correlate with the steady-state levels of MDR1 mRNA and P-glycoprotein. In cells with a severalfold increase in cellular drug resistance, MDR1 expression levels are close to the limits of detection by conventional assays. MDR1 expression has been frequently observed in human tumors after chemotherapy and in some but not all types of clinically refractory tumors untreated with chemotherapeutic drugs. We have devised a highly sensitive, specific, and quantitative protocol for measuring the levels of MDR1 mRNA in clinical samples, based on the polymerase chain reaction. We have used this assay to measure MDR1 gene expression in MDR cell lines and greater than 300 normal tissues, tumor-derived cell lines, and clinical specimens of untreated tumors of the types in which MDR1 expression was rarely observed by standard assays. Low levels of MDR1 expression were found by polymerase chain reaction in most solid tumors and leukemias tested. The frequency of samples without detectable MDR1 expression varied among different types of tumors; MDR1-negative samples were most common among tumor types known to be relatively responsive to chemotherapy.

Structure and expression of the human MDR (P-glycoprotein) gene family.

The human MDR (P-glycoprotein) gene family is known to include two members, MDR1 and MDR2. The product of the MDR1 gene, which is responsible for resistance to different cytotoxic drugs (multidrug resistance), appears to serve as an energy-dependent efflux pump for various lipophilic compounds. The function of the MDR2 gene remains unknown. We have examined the structure of the human MDR gene family by Southern hybridization of DNA from different multidrug-resistant cell lines with subfragments of MDR1 cDNA and by cloning and sequencing of genomic fragments. We have found no evidence for any other cross-hybridizing MDR genes. The sequence of two exons of the MDR2 gene was determined from genomic clones. Hybridization with single-exon probes showed that the human MDR1 gene is closely related to two genes in mouse and hamster DNA, whereas MDR2 corresponds to one rodent gene. The human MDR locus was mapped by field-inversion gel electrophoresis, and both MDR genes were found to be linked within 330 kilobases. The expression patterns of the human MDR genes were examined by enzymatic amplification of cDNA. In multidrug-resistant cell lines, increased expression of MDR1 mRNA was paralleled by a smaller increase in the levels of MDR2 mRNA. In normal human tissues, MDR2 was coexpressed with MDR1 in the liver, kidney, adrenal gland, and spleen. MDR1 expression was also detected in colon, lung, stomach, esophagus, muscle, breast, and bladder.