A multiplexed, indirect enzyme-linked immunoassay for the detection and differentiation of E. coli from other Enterobacteriaceae and P. aeruginosa from other glucose non-fermenters.

Gram-negative bacteria (GNB) are important causes of community (CA) and hospital (HA)- associated infections. Here we describe the development of an indirect ELISA (I-ELISA), which can be used to detect and differentiate the Enterobacteriaceae Escherichia coli, and glucose non-fermenter Pseudomonas aeruginosa from other GNB species. The I-ELISA utilizes six antibodies for bacterial speciation, which were grouped according to their bacterial targets; Enterobacteriaceae (SL-EntA and CH1810 mAb), Escherichia coli (SL-EcA and 6103-46 mAb), Pseudomonas aeruginosa (SL-PaA and SL-PaB). The six, anti-GNB antibodies were first screened against a panel of well-characterized clinical GNB isolates to optimize assay conditions and to determine individual antibody sensitivity and specificity. When tested against a diverse, blinded panel of 94 GNB clinical isolates, the I-ELISA exhibited the following sensitivity/specificity for each target: Enterobacteriaceae (94.4%/95%), E. coli (82.6%/88.7%), P. aeruginosa (83.3%/96%). An I-ELISA to detect and differentiate the most common GNB pathogens offers advantage in terms of simplicity over diagnostic tests currently used in most clinical settings.

Cytotoxic and proliferative allospecific T-cell clones contain perforin and mediate anti-CD3-induced cytotoxicity.

Some in vitro-generated allospecific T-cell clones can kill target cells bearing specific antigen, whereas others can only proliferate in response to that antigen. The mechanism of target lysis by clones that exhibit antigen-specific cytotoxicity is thought to involve the exocytosis of lytic granules, which contain the pore-forming protein perforin. Here, CD4+, CD8+, and CD4-8- T-cell clones, positive for CD3 and the alpha/beta T-cell receptor, were tested for their ability to lyse the mouse-anti-human CD3 hybridoma OKT3; this hybridoma has been shown to trigger the cytolytic mechanism in cytotoxic T cells regardless of their clonal specificity. We found that all in vitro-generated allospecific T-cell clones can efficiently lyse the OKT3 targets whether or not they can kill alloantigen-bearing lymphoblastoid B-cell line targets. Furthermore, all tested clones contained perforin. The OKT3 hybridoma was not lysed by perforin-negative, CD3+ leukemic T-cell lines or by CD3- NK clones. Thus, the presence of perforin in T-cell clones correlated with their ability to lyse OKT3 targets, but not with their ability to lyse alloantigen-bearing targets. These results demonstrate that T-cell clones that are nonlytic when activated by specific antigen nevertheless contain a complete lytic mechanism and also support the proposed central role in perforin in that mechanism.

Cytolytic effector function is present in resting peripheral T lymphocytes.

Antigen-specific cytotoxic killer lymphocytes (CTLs) represent one of the major effector functions of the immune system. It is well established that, as a consequence of TCR recognition of the antigen-bearing target cell, resting T lymphocytes develop into fully active antigen-specific CTLs. In contrast, natural killer (NK) cells are immediately lytic upon contact with an appropriate target cell. The lytic machinery of CTLs and NK cells is thought to include the contents of their cytoplasmic granules, in particular the pore-forming protein perforin. Here we report direct cytolytic activity by resting peripheral CD3+CD8+ T cells as a result of TCR-CD3 binding to the target cell; the murine OKT3 hybridoma (anti-human CD3) was used as a target. The cytotoxicity was more pronounced in the CD8+CD45RO+ population, which contains 'memory' T cells, than in the reciprocal CD8+CD45RA+ subset; CD8+CD4- mature thymocytes were non-cytotoxic. The cytolytic potential of these populations correlated with the presence or absence of perforin. The results demonstrate that the cytolytic machinery of T cells develops post-thymically and can be immediately triggered by TCR-CD3 stimulation.

Monoclonal antibodies detecting discrete epitopes of human perforin.

Perforin is a cytolytic protein of natural killer (NK) cells and cytotoxic T cells (CTL). Purified perforin has been shown to cause cell lysis and to form stable pores in the target cell membrane, but its relevance to cytolysis in vivo is not clear. The gene for human perforin has been cloned, but monoclonal antibodies (mabs) have not been available. In order to study further its role in cytotoxicity, we have generated mabs to different regions of human perforin. Four mabs were produced from mice immunized with hybrid proteins comprising E. coli TrpE protein at the N-terminus and different regions of human perforin at the C-terminus. These proteins were made using the pATH expression plasmids into which fragments of perforin cDNA were subcloned. Monoclonal antibody PA1 was made from a mouse immunized with a hybrid protein containing the C-terminal 240 amino acids (AA) of perforin, PE1 - the N-terminal 118 AA, and PB1 and PB2 - the central 199 AA. The three plasmid constructs contained non-overlapping cDNA segments which covered the entire sequence of perforin. All mabs reacted with the immunizing hybrid protein, but not with the other hybrid proteins, indicating that at least three epitopes are recognized by this set of mabs. All mabs immunoprecipitated a molecule of about 68 kd from lysates of metabolically-labelled cytolytic large granular lymphocytic leukemia cells, but not from control lysates of non-cytolytic promyelocytic U937 cells. These mabs should be of use in determining structure-function relationships for perforin.

Agonistic effects of anti-CD2 and anti-CD16 antibodies on human natural killer killing.

Two monoclonal antibodies (MoAb), 9-1 (anti-CD2) and 3G8 (anti-CD16), were previously shown to enhance the cytotoxic activity of human natural killer (NK) cells. The present study examined the effect of 9-1 and 3G8 with different effector and target cells to determine whether they activate NK cells through a common mechanism. Analysis of purified lymphocyte subpopulations demonstrated that the CD3+CD16+CD3- NK effector cell population is enhanced by both antibodies, while purified CD2+CD16-CD3+ T cells are not activated by either antibody. Although both antibodies enhance killing of K-562 and Daudi, killing of T-cell lines is enhanced by 9-1 and inhibited by 3G8. In contrast, killing of the promyelocytic cell line, U-937 is inhibited by 9-1 and enhanced by 3G8. On NK-susceptible cells the pattern of enhancement with 3G8, an IgG1 MoAb, is consistent with the pattern of target cell expression of an Fc receptor, FcR II, known to bind IgG1 antibodies. This suggests that 3G8 may cross-link effector and target cells through CD16 on the effectors and FcR II on these targets. This could activate NK killing by a mechanism similar to antibody-dependent cellular cytotoxicity reactions (ADCC) with the MoAb in the reverse orientation. The failure of 3G8 F(ab')2 fragments to enhance NK killing, further supports the reverse ADCC mechanism of enhancement by 3G8. The pattern of enhancement mediated by 9-1, an IgG3 MoAb, is not correlated with any target cell Fc-receptor known to bind IgG3 MoAb. The effect of 9-1 may result, instead, from its binding to the unique 9-1 epitope on the CD2 molecule involved in CD2-mediated T-cell activation, as previously described. Alternative mechanisms, including activation of NK killing by 9-1 mediated cross-linking of CD2 and CD16 on the effector cells, have also been discussed.

Total lymphoid irradiation prevents diabetes mellitus in the Bio-Breeding/Worcester (BB/W) rat.

Total lymphoid irradiation (TLI) at doses of 2200 rads or greater prevented diabetes in susceptible BB/W rats. Two of 29 (7%) treated rats became diabetic compared with 23 of 39 (59%) controls (P less than 0.001). TLI did not, however, prevent insulitis or thyroiditis in nondiabetic rats, nor did it restore the depressed concanavalin-A responsiveness of BB rat lymphocytes. T-lymphocyte subset proportions were the same in both groups. TLI was associated with significant radiation-related mortality, and nondiabetic TLI-treated rats weighed significantly less than controls. We conclude that TLI is effective in the prevention of BB rat diabetes. However, TLI fails to correct the subclinical immunologic abnormalities of the model and is associated with significant morbidity.