A new technique, the PCR-flow assay is described that has allowed for the serial identification and quantitation of discrete mononuclear cell subsets of donor (or recipient) bone marrow derived cells in cadaver kidney transplant recipients infused postoperatively with donor vertebral body bone marrow cells. With fixed permeabilized cells in flow cytometry the amplification power of the polymerase chain reaction (PCR), using fluorescent-labeled primers to identify single copy HLA class II DRbeta1 genes of either donor or recipient origin, is combined with multi-color fluorochrome-labeled CD epitope-specific monoclonal antibodies. The details of the methodology are described; these support the utility of the assay. Initial observations were made on the chimeric makeup of the peripheral blood as well as iliac crest bone marrow between six months and one year posttransplantation in recipients serially followed weekly and then monthly, concomitantly compared with a control group of stable kidney transplant recipients using similar therapeutic protocols, who did not receive cadaver bone marrow. Several findings are of note. In 14 recipients of two bone marrow infusions totalling a mean of 6.29+/-2.18x10(10) cells, donor CD34 positive (+) (immature) cells were fourteen times as numerous in peripheral blood six months postoperatively as in six recipients given half as many bone marrow cells in one infusion (averaging 3.02+/-0.5x10(10)). These donor CD34+ cells unexpectedly averaged 36+/-7% of the total (donor plus recipient) CD34+ subset counted. Moreover, iliac crest bone marrow aspirates contained an average of thirteen times this number of CD34+ cells than in the peripheral blood, supporting the notion of engraftment. Of additional interest, between six months and one year posttransplant although no donor cells could be detected in peripheral blood of the controls there was an identifiable presence of donor CD34+ cells in their iliac crest bone marrow, albeit 10-fold less than the marrow-infused patients. In the clinical follow-up, although there were three unrelated mortalities, there were no additional kidney losses with current serum creatinine concentrations averaging 1.3+/-0.06 mg/dl. In conclusion, the PCR-flow assay presents the possibility of identifying discrete subsets of donor or recipient cells that may have an immunoregulatory function.