We report the detection of insulin-like antigens in a large range of species utilizing a modified ELISA plate assay and Western blotting. We tested the leaves or aerial parts of species of Rhodophyta (red alga), Bryophyta (mosses), Psilophyta (whisk ferns), Lycopodophyta (club mosses), Sphenopsida (horsetails), gymnosperms, and angiosperms, including monocots and dicots. We also studied species of fungi and a cyanobacterium, Spirulina maxima. The wide distribution of insulin-like antigens, which in some cases present the same electrophoretic mobility as bovine insulin, together with results recently published by us on the amino acid sequence of an insulin isolated from the seed coat of jack bean (Canavalia ensiformis) and from the developing fruits of cowpea (Vigna unguiculata), suggests that pathways depending on this hormone have been conserved through evolution.

Refractoriness to random-donor platelets as a result of alloimmunization remains a major problem in long-term platelet transfusion therapy despite the use of HLA-matched platelets. We have therefore studied the use of two methods for detection of platelet associated IgG as platelet crossmatch tests for the selection of platelet donors. These methods use radiolabelled staphylococcal protein A ( 1251-SPA) and peroxidase anti-peroxidase (PAP), respectively.One hundred and ten crossmatch tests using I2%SPA were performed retrospectively in 18 alloimmunized patients. The results indicated that the predictive value of a positive or a negative test was 8 7%; the sensitivity was 73% and the specificity was 95%. Results with the PAP test were similar. The HLA types were known for 48 donor-recipient pairs. With few exceptions, there was a correlation between the results of the platelet crossmatch tests and the effectiveness of platelet transfusion regardless of the degree of HLA match. These results indicate that platelet crossmatch tests may be valuable even when closely HLA matched donors are not available. A large-scale prospective study is warranted, particularly in highly immunized patients. Alloimmunization and the resultant refractoriness to random donor platelet transfusions occur in patients who receive repeated transfusions of platelets from random donors. Estimates of the frequency of alloimmunization range from about 35% to 100% (Dutcher et al, 1981; Klein & Blajchman, 1982). Alloimmunized patients frequently respond to platelet transfusions from donors matched for HLA-A and -B antigens (Yankee et al, 1969, 1973; Lohnnann et al, 19 74) or serologically crossreactive antigens (Svejgaard & Kissmeyer-Nielsen, 1968; Mittal & Terasaki, 1974; Duquesnoy et al, 1977a; Tomasulo, 1978).

Summary. Refractoriness to random‐donor platelets as a result of alloimmunization remains a major problem in long‐term platelet transfusion therapy despite the use of HLA‐matched platelets. We have therefore studied the use of two methods for detection of platelet associated IgG as platelet crossmatch tests for the selection of platelet donors. These methods use radiolabelled staphylococcal protein A (125I‐SPA) and peroxidase anti‐peroxidase (PAP), respectively. One hundred and ten crossmatch tests using 125I‐SPA were performed retrospectively in 18 alloimmunized patients. The results indicated that the predictive value of a positive or a negative test was 87%; the sensitivity was 73% and the specificity was 95%. Results with the PAP test were similar. The HLA types were known for 48 donor‐recipient pairs. With few exceptions, there was a correlation between the results of the platelet crossmatch tests and the effectiveness of platelet transfusion regardless of the degree of HLA match. These results indicate that platelet crossmatch tests may be valuable even when closely HLA matched donors are not available. A large‐scale prospective study is warranted, particularly in highly immunized patients.

An immunohistochemical procedure for the detection of immunoglobulin G adherent to platelets is described. The peroxidase anti-peroxidase method is used to detect antibody activity directed against platelets from normal donors in the sera from 305 individuals. These subjects were divided into three groups: group 1, patients referred for tissue typing; group 2, healthy normal females; group 3, healthy normal males. In group 1, 28% of the sera were found to be positive; in most of these a history of prior transfusions was obtained. In group 2, 7.4% were found to be positive, most having previous pregnancies. Only 1% were found to be positive in group 3, and no reason for presensitization was found. Results from the indirect immunofluorescence technique served as a control and as a means to compare the sensitivity. Under the conditions chosen, the peroxidase anti-peroxidase test was two to eight times more sensitive than the immunofluorescence technique. Specificity of the peroxidase anti-peroxidase technique was demonstrated using a monospecific anti-PLA1 antiserum. It is concluded that the peroxidase anti-peroxidase slide technique may be a useful tool in the study of platelet-related immunophenomena.

White blood cells from 22 patients with leukemia and lymphoma were studied for the presence of terminal deoxynucleotidyl transferase with a peroxidase-antiperoxidase method. The enzyme was detected in leukemic cells of 5 patients with acute lymphoblastic leukemia and 1 patient with chronic myelogenous leukemia, whereas 16 patients with different forms of leukemia or lymphoma were negative for this enzyme. Comparative studies using a biochemical and an indirect immunofluorescence assay revealed complete concordance between these three methods.