Trisomic dose of several chromosome 21 genes perturbs haematopoietic stem and progenitor cell differentiation in Down's syndrome.

Children with Down's syndrome (DS) have 20-50-fold higher incidence of all leukaemias (lymphoid and myeloid), for reasons not understood. As incidence of many solid tumours is much lower in DS, we speculated that disturbed early haematopoietic differentiation could be the cause of increased leukaemia risk. If a common mechanism is behind the risk of both major leukaemia types, it would have to arise before the bifurcation to myeloid and lymphoid lineages. Using the transchromosomic system (mouse embryonic stem cells (ESCs)) bearing an extra human chromosome 21 (HSA21)) we analyzed the early stages of haematopoietic commitment (mesodermal colony formation) in vitro. We observed that trisomy 21 (T21) causes increased production of haemogenic endothelial cells, haematopoietic stem cell precursors and increased colony forming potential, with significantly increased immature progenitors. Transchromosomic colonies showed increased expression of Gata-2, c-Kit and Tie-2. A panel of partial T21 ESCs allowed us to assign these effects to HSA21 sub-regions, mapped by 3.5 kbp-resolution tiling arrays. The Gata-2 increase on one side, and c-Kit and Tie-2 increases on the other, could be attributed to two different, non-overlapping HSA21 regions. Using human-specific small interfering RNA silencing, we could demonstrate that an extra copy of RUNX1, but not ETS-2 or ERG, causes an increase in Tie-2/c-Kit levels. Finally, we detected significantly increased levels of RUNX1, C-KIT and PU.1 in human foetal livers with T21. We conclude that overdose of more than one HSA21 gene contributes to the disturbance of early haematopoiesis in DS, and that one of the contributors is RUNX1. As the observed T21-driven hyperproduction of multipotential immature precursors precedes the bifurcation to lymphoid and myeloid lineages, we speculate that this could create conditions of increased chance for acquisition of pre-leukaemogenic rearrangements/mutations in both lymphoid and myeloid lineages during foetal haematopoiesis, contributing to the increased risk of both leukaemia types in DS.

Functional dichotomy in natural killer cell signaling: Vav1-dependent and -independent mechanisms.

The product of the protooncogene Vav1 participates in multiple signaling pathways and is a critical regulator of antigen-receptor signaling in B and T lymphocytes, but its role during in vivo natural killer (NK) cell differentiation is not known. Here we have studied NK cell development in Vav1-/- mice and found that, in contrast to T and NK-T cells, the absolute numbers of phenotypically mature NK cells were not reduced. Vav1-/- mice produced normal amounts of interferon (IFN)-gamma in response to Listeria monocytogenes and controlled early infection but showed reduced tumor clearance in vivo. In vitro stimulation of surface receptors in Vav1-/- NK cells resulted in normal IFN-gamma production but reduced tumor cell lysis. Vav1 was found to control activation of extracellular signal-regulated kinases and exocytosis of cytotoxic granules. In contrast, conjugate formation appeared to be only mildly affected, and calcium mobilization was normal in Vav1-/- NK cells. These results highlight fundamental differences between proximal signaling events in T and NK cells and suggest a functional dichotomy for Vav1 in NK cells: a role in cytotoxicity but not for IFN-gamma production.

Dissecting NK cell development using a novel alymphoid mouse model: investigating the role of the c-abl proto-oncogene in murine NK cell differentiation.

NK lymphocytes participate in both innate and adaptive immunity by their prompt secretion of cytokines including IFN-gamma, which activates macrophages, and by their ability to lyse virally infected cells and tumor cells without prior sensitization. Although these characteristics of NK cells are well documented, little is known about the genetic program that orchestrates NK development or about the signaling pathways that trigger NK effector functions. By crossing NK-deficient common gamma-chain (gammac) and recombinase activating gene (RAG)-2 mutant mice, we have generated a novel alymphoid (B-, T-, and NK-) mouse strain (RAG2/gammac) suitable for NK complementation in vivo. The role of the c-abl proto-oncogene in murine NK cell differentiation has been addressed in hemopoietic chimeras generated using RAG2/gammac mice reconstituted with c-abl-/- fetal liver cells. The phenotypically mature NK cells that developed in the absence of c-abl were capable of lysing tumor targets, recognizing "missing self," and performing Ab-dependent cellular cytotoxicity. Taken together, these results exclude any essential role for c-abl in murine NK cell differentiation in vivo. The RAG2/gammac model thereby provides a novel approach to establish a genetic map of NK cell development.

Regulation of immune complex-mediated complement activation by autoantibodies (F-42) isolated from sera of patients with systemic lupus erythematosus.

Sera of some patients with systematic lupus erythematosus (SLE) contain IgG autoantibodies (F-42) which have been shown to stabilize the cell bound classical pathway C3 convertase of complement, C42. C42 is susceptible to inactivation by the plasma protein C4bp while stabilized C42 is relatively resistant to C4bp. The present study demonstrates that F-42 by itself does not induce activation of the classical pathway in vitro but that it is able to modulate the immune complex-induced consumption of C2 and C3 in whole serum. Incubation of incremental concentrations of F-42 with normal human serum (NHS) for 30 min at 30 degrees C did not result in detectable consumption of C1q, C4, C2 and C3. However, when soluble immune aggregates or immune complexes were incubated in NHS together with 100 u/ml of F-42, a significant increase in consumption of C3 was seen as compared to the reaction mixture containing immune complexes or F-42 alone. In addition the presence of F-42 during the immune complex mediated consumption of complement was associated with relative protection of C2 consumption. (Fab)'2 and Fab' fragments of F-42 behaved as intact F-42, except that their activities on a molar basis were less than that of intact F-42. The results presented in this paper suggest that F-42 may play a regulatory role in the immune complex-mediated consumption at C2 and C3 in vivo.

Ratios of T lymphocyte subpopulations predict survival of cadaveric renal allografts in adult patients on low dose corticosteroid therapy.

Peripheral blood T lymphocyte subpopulations were monitored in 45 consecutive adult recipients of cadaveric renal allografts by using monoclonal antibodies and flow cytometrie. All patients were treated with low dose corticosteroids and azathioprine. In 37 patients pre-transplant OKT4/OKT8 ratios were available. Six of 26 patients (23%) with pre-transplant OKT4/OKT8 ratios greater than 1.6 and seven of 11 patients (64%) with pre-transplant OKT4/OKT8 ratio less than or equal to 1.6 lost their graft due to rejection within 6 months. The difference in transplant survival between patients with pre-transplant OKT4/OKT8 ratios greater than 1.6 and less than or equal to 1.6i is just significant (P = 0 . 049 Fishers test). No correlation was found between post-transplant values of individual lymphocyte subpopulations or OKT4/OKT8 ratios and the incidence of subsequent rejection episodes. Forty out of 45 patients suffered one or more rejection episodes which were treated by raising the dosage of prednisone. In 24 of these patients the rejection episode was reversed, leading to a transplant survival of at least 6 months. In these 24 patients the OKT4/OKT8 ratio was greater than 1.6 for at least 3 days before the institution of any rejection treatments. Sixteen patients lost their graft due to rejection within 6 months after transplantation. In 11 of these 16 patients OKT4/OKT8 ratios less than or equal to 1.6 preceded the institution of all rejection treatments for at least 3 days, while in three patients the OKT4/OKT8 ratio was greater than 1.6 before the first rejection episode but this ratio was less than or equal to 1.6 before subsequent rejection episodes. Thus, OKT4/OKT8 ratios greater than 1.i6 correlated with reversible rejection episodes and OKT4/OKT8 ratios less than or equal to 1.6 correlated with irreversible rejection (P less than 0 . 001).

Increased IgM and IgM immune complex-like material in the circulation of renal transplant recipients with primary cytomegalovirus infections.

Twelve of 60 consecutive adult recipients of cadaver kidney transplants had increased polyethyleneglycol (PEG) precipitable IgM immune complex-like material in their circulation in the first 4 months after transplantation. All 10 recipients with primary CMV and two of four with secondary CMV infections had significant elevations in PEG precipitable IgM that coincided with rises in their CMV antibody titres. Ultracentrifuge analysis demonstrated two peaks of PEG precipitable material with sedimentation rates of about 20S and 40S. Total IgM immunoglobulin levels also were increased in transplant recipients with CMV infections, but this was less specific and occurred in patients without CMV infections. The Clq binding assay, which is more sensitive for IgG than IgM containing complexes, was positive in only three of 10 patients with primary CMV and none of four with secondary CMV. Granular deposits of IgM, but not IgG, were detected in the glomeruli of six of seven transplants biopsied during CMV infection. The PEG-IgM assay was not influenced by rejection or prednisone therapy. Thus, transplant patients, who develop primary CMV infections, produce elevated levels of circulating IgM and IgM immune complex-like material. These findings may help to differentiate CMV infection from transplant rejection as well as to increase our understanding of the special pathogenic properties of CMV in transplant recipients.

Influence of aggregate size on the binding and activation of the first component of human complement by soluble IgG aggregates.

The interaction between small aggregates of human IgG and the first component of human complement was studied. Stabilized soluble IgG aggregates of restricted size were prepared by heat aggregation of human IgG, followed by sucrose-density ultracentrifugation. Human C1 was isolated in its precursor form by euglobulin precipitation, followed by gel filtration and immunoadsorption. A C1 preparation was obtained of which more than 90% was still in its unactivated form. Soluble aggregates containing 20, 10 or 5 molecules IgG, and monomeric IgG were tested for their ability to bind and to activate C1. The binding of C1 was determined by C1 consumption, whereas the activation of C1 was measured as the increased ability of the C1 preparation to consume purified human C4 after the incubation with the aggregates. The three aggregates tested and monomeric IgG were all able to bind and to activate C1, but the efficiency of both processes markedly increased with increasing aggregate-size. Furthermore, it was found that all four preparations activated an appreciable amount of C1 at concentrations that did not result in any detectable C1 fixation. These results confirm earlier suggestion that C1 can be activated during a short, transient binding to small aggregates or immune complexes that have a low avidity for C1, after which the activated form, C1, is released into the medium.

Circulating and tissue-bound immune complexes in allergic vasculitis: relationship between immunoglobulin class and clinical features.

Sixty-two patients with allergic vasculitis and histological evidence of leukocytoclastic vasculitis were examined to determine whether clinical features, such as course and degree of systemic involvement, could be related to the class of the immunoglobulins in immune complexes (IC) in both the circulation and the vessel wall. Circulating IC were detected with the 125I-C1q-binding assay (C1q-BA), an anti-IgA inhibition binding assay (a-IgA-Inh BA) and the conglutinin-binding assay (Con-BA). Stabilized heat-aggregated IgG, IgA and IgM were used to determine the immunoglobulin class specificity of these assays. With the C1q-BA only aggregated IgG were detected, with the a-IgA-Inh Ba only aggregates containing IgA. With the Con-BA IgG, IgA or IgM in reactive aggregates were identified with class-specific antibodies. For patients with acute cutaneous vasculitis all assays were negative. The a-IgA-Inh BA was frequently positive in sera of patients with chronic cutaneous and acute systemic vasculitis; in the latter group conglutinin-binding IC of the IgG, IgA and IgM class were also detected. Levels in the C1q-BA were high for patients with chronic systemic vasculitis. Comparison of the results of the IC assays with the immunofluorescence studies of the cutaneous vessel walls of the same patients showed agreement between the results of the C1q-BA and deposition of IgG and the results of the a-IgA-Inh BA and IgA deposition. The class of immunoglobulin in conglutinin-binding IC did not correspond as well with the immunoglobulins in vessel walls. This study shows that certain clinical features of allergic vasculitis are related to the composition of the IC in the circulation and in the vessel wall.

Stabilization of the classical pathway C3 convertase C42, by a factor F-42, isolated from serum of patients with systemic lupus erythematosus.

Sera from sixteen patients with SLE were investigated for the presence of a factor which would conserve convertase activity on preformed EAClgp 4hu2hu for 30 min at 30 degrees in EDTA. Although such a factor could not be detected readily in the sera, chromatography on DE-52 cellulose yielded fractions appearing as three peaks in one patient and as two peaks in a second patient. These peaks were capable of conserving C42 activity and were designated as F-42. Purification of F-42 from the second peak eluting between 4 and 7 mS on DE-52 was obtained by SP-C50, S-300 and QAE-A50 chromatography. F-42 exhibited charge heterogeneity upon SP-C50 chromatography. On polyacrylamide gel electrophoresis the final material migrated as one band, which coincided with the position of F-42 activity upon eluation from a parallel gel. F-42 had an apparent molecular weight of 150,000 and reacted with anti-IgG in Ouchterlony analysis. Sepharose-bound anti-IgG was capable of neutralizing F-42 activity. The purified material was shown to prolong the half-life (T 1/2) of performed cell-bound C42 in GVB-EDTA at 30 degrees from 5 to 80 min.

The influence of phagocyte function on glomerular localization of aggregated IgM in rats.

The blood clearance and hepatic localization of aggregated human IgM (AIgM) were studied in relation to the glomerular localization patterns in rats. AIgM was cleared rapidly from the circulation by sinusoidal cells in the liver, monomeric IgM (MIgM) was cleared less rapidly. Localization of AIgM within the glomerulus was dependent on the level in the blood. Low doses led only to a mesangial localization, whereas high doeses resulted in a localization along the capillary walls. Immunoelectron microscopy showed that in the glomerular capillaries the AIgM was localized within endothelial cells or subendothelially, and did not occur within the glomerular basement membrane. Inhibition of the hepatic uptake of AIgM by colloidal carbon resulted in increased levels of circulating AIgM and prolonged the deposition of AIgM in the glomerular capillaries.