Reticuloendothelial Saturation Dictates the Development of RBC Resistance to Antibody-Mediated Clearance Following Incompatible Transfusion

Background: Immune-mediated destruction of red blood cells (RBCs) normally occurs following incompatible transfusion, however some RBCs may remain in circulation after initial cellular clearance despite the presence of RBC specific antibodies. Recent studies suggest that antibody-induced “antigen-modulation” may in part be responsible for this cellular resistance to antibody-mediated removal following the initial phase of RBC clearance. However, the factors that dictate whether cells undergo clearance or antigen modulation immediately following incompatible RBC transfusion remain unknown. As previous studies suggest that reticuloendothelial phagocytic capacity is sometimes surpassed upon engagement and removal of antibody-opsonized cells, we hypothesized that rapid change in the phagocytic burden of reticuloendothelial cells may contribute to the development of RBC resistance following incompatible RBC transfusion. Methods: Anti-Fy3 immunized or non-immunizedC57BL/6 recipient mice were first transfused with HOD.FVB RBCs expressing the HEL, OVA and Duffy chimeric antigen (HOD), followed by a second HOD.FVB RBC transfusion at various time intervals following the initial transfusion. Following transfusion, mice were evaluated at 10 min, 1h, 2h and 4h post-transfusion for HOD.FVB RBC survival, detectable HOD antigen, and RBC bound antibody by flow cytometric analysis. To examine the potential impact of RBC age on the sensitivity of HOD.FVB to antibody-induced clearance, HOD.FVB mice were injected with N-hydroxysulfosuccinimide (NHS) biotin 35 days prior to RBC isolation, to aid in differentiation of older from younger transfused RBCs, and then likewise transfused into anti-Fy3 immunized or non-immunizedC57BL/6 recipients and similarly evaluated for clearance, bound antibody and detectable antigen on biotin positive or negative transfused cells using flow cytometric analysis. Results: While HOD.FVB RBCs transfused into anti-Fy3-immunized C57BL/6 recipients displayed a rapid clearance following the initial transfusion, the degree of HOD.FVB RBC clearance significantly decreased following secondary transfusions that occurred within 2 hours of the initial transfusion. However, 2 days following the initial transfusion, recipients cleared HOD.FVB RBCs at the same rate as immunized recipients not previously exposed to HOD.FVB RBCs. Examination of the HOD antigen and bound antibody demonstrated that antigen levels likewise displayed the most significant decreases during the same time period in which RBC clearance decreased. Removal of HOD RBCs during the initial clearance phase did not appear to reflect preferential clearance of older RBCs, as transfused biotinylated and non-biotinylated RBCs harvested 1 day or 35 days following biotinylation displayed equal levels of antibody-induced clearance. Conclusion: Alterations in recipient clearance capacity following incompatible RBC exposure suggests that rapid phagocytic removal of RBCs may temporarily staturate the reticuloendothelial system immediately following RBC transfusion. However, antigen modulation appears to continue even while congestion of the reticuloendothelial system has halted RBC clearance. RBCs that escape the initial wave of clearance appear to become protected from antibody-mediated removal once the phagocytic capability of the recipient is restored, as the level of detectable antigen has dropped below that required for triggering phagocytic removal. The initial phase of clearance does not appear to reflect preferential clearance of older RBCs within the transfused units, but instead likely reflects stochastic removal of antibody bound RBCs, as RBC clearance occurred independent of RBC age. Disclosures No relevant conflicts of interest to declare.

Association between circulating antigen and parasite load in a model filarial system, Acanthocheilonema viteae in jirds

SUMMARYJirds (Meriones libycus) were infected with various numbers of Acanthocheilonema viteae L3 stage parasites. During the course of the ensuing 16 weeks, blood samples were collected at 2 weekly intervals and the amount of the major parasite excretory–secretory product (E–S 62) and antibodies directed against it measured. After 16 weeks, animals were sacrificed and the size of the mature worm burden established. In spite of interaction between E–S 62 and host antibody, a statistically significant relationship was found to exist between the amount of E–S 62 present in the bloodstream and the size of the parasite load. It is suggested that the detectable antigen level is more influenced by the size of the worm burden than the presence of antibody and that antibody is only likely to affect adversely antigen measurement in situations where the amount released is relatively low. Examples of this are early in infection and in low-level infections. These ideas are discussed in relation to the development and assessment of serological assays which attempt to predict parasite burden in human filarial infections.

IMMUNOLOGICAL SIMILARITIES BETWEEN PLASMA AND URINARY PROTEIN C INHIBITORS (PCIs) AND URINARY UROKINASE INHIBITOR (UKI)

Plasma PCIs have similar MW (∼ 57K), amino acid composition, and heparin dependence (Suzuki et al 1983, JBC 258:163) as urinary UKI (Stump et al 1986, JBC 261:12759). Urinary PCI of ∼ 50K MW has a similar heparin dependence and urokinase (UK) competes with activated protein C (APC) for this PCI (Geiger et al 1986, Circ. 74:11-234). For comparison, three forms of PCI, one from urine and two from plasma, were purified, and each exhibited heparin-dependent UK and APC inhibitory activity and formed heparin-dependent complexes with APC. The APC-PCI complexes were visible on immunoblots (nondenaturing gels) developed using: A) monoclonal anti-UKI + 125I-antimouse IgG; B) polyclonal anti-plasma PCI + 125I-plasma PCI; and C) monoclonal anti-protein C + 125I-protein C. The three forms of purified PCI were detected by methods A and B. Two new bands of APC-inhibitor complexes were seen upon incubation of plasma with APC in the presence of heparin, and the same pattern was visualized by methods A, B, and C. In the absence of heparin, only one APC-inhibitor band was visualized by methods A and B, but two bands were visualized by method C. Plasma immunodepleted of UKI by monoclonal anti-UKI-Sepharose showed no detectable antigen or complexes with APC as visualized by methods A and B. However, the UKI-depleted plasma contained components which formed a reduced amount of complexes with APC as visualized with protein C antibodies, i.e. method C. Heparin stimulates tenfold the PCI activity of normal plasma. Based on amidolytic assays of APC using S-2366, the UKI-depleted plasma was very deficient (< 15%) in heparin-dependent PCI activity, whereas the weak heparin-independent PCI activity was slightly reduced. This indicates that the majority of heparin-dependent PCI activity of plasma is immunologically.related to UKI. These studies suggest that the two slightly different forms of plasma PCI, the urinary UKI, and the urinary PCI are very similar if not identical proteins and that plasma may contain a minor heparin-independent PCI which is not immunologically related to these proteins.

Particle-counting immunoassay of a fetuin-like antigen in serum and cerebrospinal fluid.

A fetuin-like antigen was detected (smallest concentration detectable: 5 micrograms/L) by particle-counting immunoassay in 2% (13/641) of consecutive patients' sera but not in sera from 80 healthy blood donors, 40 neonates, or 40 pregnant women. The relation of the presence of detectable antigen to patients' diagnosis is not yet clear. However, in the group with cancer (154), it was found only in two of four patients with nephroblastoma and in three of five with tumors of tissue derived from the neurological crest: retinoblastoma (1/1), neuroblastoma (1/3), and medulloblastoma (1/1). Serum specimens from 422 patients with neurological disorders showed the antigen at a concentration greater than 5 micrograms/L in cases of neurosyphilis (5/11), peripheral neuropathy (12/38), Guillain-Barré syndrome (7/27), and multiple sclerosis (74/184). When we assayed 232 specimens of cerebrospinal fluid from the same neurological patients, we found the antigen in two cases of multiple sclerosis (6 and 15 micrograms/L) and in one case of Guillain-Barré syndrome (54 micrograms/L).

Evidence for a regulatory idiotypic network in the in vivo response to H-2 antigens.

Treatment of BALB/c mice with purified pig antiidiotype to 11-4.1 (anti-H-2Kk) monoclonal antibody has been found previously to induce the appearance of idiotype-bearing molecules (Id') in the serum of these mice, in the absence of detectable antigen-binding activity. In the present study we examined the effect of subsequent immunization of such antiidiotype-primed mice with the original H-2Kk antigen. Skin grafting of virgin BALB/c mice with BALB.K skin did not generate any detectable Id' antibodies when tested by enzyme-linked immunosorbent assay (ELISA). In contrast, grafting of antiidiotype-primed mice with BALB.K skin specifically boosted ther serum level of Id' molecules. Challenge of antiidiotype-primed mice with either B10.D2 or rat skin had no effect on the production of such Id' molecules. Absorption studies demonstrated that the majority of Id' molecules induced by H-2Kk antigenic stimulus and detected in ELISA are antigen-nonbinding molecules, thus indicating specific restimulation by the original H-2Kk antigen of nonbinding idiotype-positive B cell clones. The relevance of these findings to the existence of network interactions in the immune response to H-2 antigens is discussed.

Identification of a synaptic vesicle-specific membrane protein with a wide distribution in neuronal and neurosecretory tissue.

Two different monoclonal antibodies, characterized initially as binding synaptic terminal regions of rat brain, bind a 65,000-dalton protein, which is exposed on the outer surface of brain synaptic vesicles. Immunocytochemical experiments at the electron microscope level demonstrate that these antibodies bind the vesicles in many different types of nerve terminals. The antibodies have been used successfully to purify synaptic vesicles from crude brain homogenates by immunoprecipitation onto the surface of polyacrylamide beads. The profiles of the structures precipitated by these beads are almost exclusively vesicular, confirming the vesicle-specificity of the antibodies. In SDS gels, the antibodies bind a single protein of 65,000 daltons. The two antibodies are not identical, but compete for binding sites on this protein. Immune competition experiments also demonstrate that the antigenic components on the 65,000-dalton protein are widely distributed in neuronal and neural secretory tissues. Detectable antigen is not found in uninnervated tissue--blood cells and extrajunctional muscle. Low levels are found in nonneural secretory tissues; it is not certain whether this reflects the presence of low amounts of the antigen on all the exocytotic vesicles in these tissues or whether the antigen is found only in neuronal fibers within these tissues. The molecular weight and at least two antigenic determinants of the 65,000-dalton protein are highly conserved throughout vertebrate phylogeny. The two antibodies recognize a 65,000-dalton protein present in shark, amphibia, birds, and mammals. The highly conserved nature of the determinants on this protein and their specific localization on secretory vesicles of many different types suggest that this protein may be essential for the normal function of neuronal secretory vesicles.

Induction of idiotype-bearing, nuclease-specific helper T cells by in vivo treatment with anti-idiotype.

Treatment of BALB/c mice with purified pig anti-(BALB/c anti-nuclease) anti-idiotypic antibodies has been found to induce the appearance of idiotype-bearing immunoglobulins (Id') in the serum of these mice in the absence of detectable antigen binding activity. This phenomenon appeared to require T cells in the hosts because no Id' was detected in the serum of nude mice similarly treated. Furthermore, the spleens of BALB/c mice treated with anti-idiotype were found to contain helper T cells capable of providing help in an in vitro plaque-forming cell response to trinitrophenyl-nuclease equivalent to that provided by helper T cells from the spleens of nuclease-primed animals. Helper T cells from both anti-idiotype-treated and nuclease-treated animals were found to be antigen-specific and to be similarly susceptible to elimination by treatment with anti-idiotype plus complement. Therefore, treatment with both antigen and anti-idiotype appeared to prime similar populations of antigen-specific helper T cells, while having different effects on the induction of antibody. These findings are consistent with the network theory of receptor interactions in the immune response, and may provide a means for studying individual cell populations involved in such interactions.