Activation of the acute phase response and complement C3 in patients with IgA nephropathy

Recently we showed systemic complement activation in patients with immunoglobulin A (IgA) nephropathy (measured by "activated C3" [actC3], in other words, neoantigens developing on breakdown products after C3 activation) and reported that plasma levels of actC3 can indicate disease activity and renal outcome. In this study we investigated whether plasma C3a and C-reactive protein (CRP), which require tests that are more readily available, have a similar diagnostic and predictive value. CRP was measured using a highly sensitive enzyme-linked immunosorbent assay and C3a using a specific immunoassay. CRP and C3a levels were significantly higher in 56 patients with IgA nephropathy as compared with 55 healthy controls. C3a levels in IgA nephropathy patients were also significantly increased in comparison with 42 patients with hypertension or nonimmune renal diseases. Neither C3a nor CRP levels correlated with those of actC3 in IgA nephropathy patients. We also compared 10 IgA nephropathy patients with stable, normal renal function with eight IgA nephropathy patients progressing from normal to impaired renal function during mean follow-ups of 7.1 and 5.1 years, respectively. Mean CRP but not C3a levels during the observation period were significantly higher in IgA nephropathy patients with disease progression than in those with stable renal function.

CONCLUSION

Systemic complement activation can be detected by measurement of plasma C3a in IgA nephropathy, but C3a levels cannot substitute for actC3 in predicting renal prognosis. Subclinical induction of the acute phase response is also present in patients with progressive IgA nephropathy, but again its prognostic value is limited. Repeated determinations performed over prolonged time courses may possibly improve the prognostic value of CRP levels.

Ligand-induced phosphorylation of anaphylatoxin receptors C3aR and C5aR is mediated by "G protein-coupled receptor kinases

Continuous stimulation of anaphylatoxin receptors C3aR and C5aR with their cognate ligands engenders, within minutes, diminished responsiveness of these receptors. We tested the hypothesis that agonist-induced desensitization involves C3aR and C5aR phosphorylation by G protein-coupled receptor kinases (GRK). When expressed in rat basophilic leukemia cells and exposed to C3a, the C3aR underwent rapid (t(1/2) approximately 15 s), dose-dependent (EC50 approximately 10 nM) and reversible phosphorylation by a kinase refractory to the effects of PKC inhibitors. Phosphoamino acid analysis revealed that the C3aR is phosphorylated on serine and threonine, but not on tyrosine residues. Overexpression of GRK2, GRK3, GRK5 or GRK6 together with C3aR in COS-7 cells enhanced the C3a-induced C3aR phosphorylation 1.5 - 1.9-fold (p < 0.05), but each kinase reduced ligand-stimulated phospholipase C activity differently. Conversely, antibody-mediated inhibition of endogenous GRK2 and GRK3 significantly inhibited C3aR phosphorylation in permeabilized cells. GRK overexpression in cells which co-expressed C5aR and were exposed to C5a resulted in the hyperphosphorylation of the C5aR. These findings are of physiological relevance, since we observed anaphylatoxin-induced phosphorylation of C3aR and C5aR endogenously expressed in human mast cells (HMC-1) which contain significant intracellular levels of GRK2 and GRK3.

Expression of the anaphylatoxin C5a receptor in non-myeloid cells

C5, a 74 amino acid peptide cleaved from the complement protein C5, represents the most potent anaphylatoxin and possesses inflammatory as well as immunomodulatory activities. In the past, expression of the receptor for the anaphylatoxin C5a (C5aR) has been thought to be restricted to cells of myeloid origin. However, recent evidence suggests that the C5aR is constitutively expressed in non-myeloid cells including epithelial, endothelial and smooth muscle cells in the human liver and lung. These findings are contrasted by results from our laboratory which demonstrated that in the normal human liver and lung C5aR expression is detectable exclusively in macrophages and macrophage-derived cells (Kupffer cells). Interestingly, we found evidence that C5aR expression may be inducible in epithelial cells as C5aR mRNA was observed in vivo in human keratinocytes of the inflamed but not of the normal skin. Herein we review the work of our laboratory and others on the expression of the C5aR in various human non-myeloid cells types. A better understanding of the expression patterns of this important anaphylatoxin receptor may provide new insights in the pathophysiological role of C5a in vivo.

Specific detection by flow cytometry of histidine-tagged ligands bound to their receptors using a tag-specific monoclonal antibody

Engineering proteins to contain a histidine (His)-tag has proved to be very useful for the purification and analyses of these molecules. In the present study, we demonstrate that the binding of His-tagged ligands to their receptors may be visualised by flow cytometry making use of a selected monoclonal antibody (mAb) against the His-tag. Employing this method, a recombinant C3a (rC3a) anaphylatoxin with a His-tag at its N-terminus could be shown to bind to C3a receptor (C3aR)-expressing RBL-2H3 transfectants with a half-maximal effective concentration (EC50) of about 3 nM which is well within the range of published affinity constants. Binding of a recombinant interleukin-8 (rIL-8) molecule with a C-terminal His-tag to RBL-2H3 cells which stably express the IL-8 receptors CXCR1 or CXCR2 could also be demonstrated using the tag-specific mAb. Furthermore, aminoterminally tagged C5a molecules of rat or human origin could be shown to bind to the human C5a receptor (C5aR). However, the fluorescence signal of the binding of rat rC5a to the human C5aR was distinctly higher over a wide range of ligand concentrations than the signal of human rC5a binding although both ligands were equally potent in the induction of chemotaxis in C5aR-expressing cells. Thus, the tag-specific mAb was able to interfere with the binding of human but not rat rC5a to the human C5aR. This observation is in agreement with the hypothesis of a two binding site model for the interaction of human C5a with its receptor whereas a different binding mode may apply for rat C5a. Our data demonstrate that the selected His-tag specific mAb may be a valuable tool for the visualisation of the binding of recombinant ligands to their receptors and may also provide useful information on the specific binding properties of the ligands.

C5a suppresses the production of IL-12 by IFN-gamma-primed and lipopolysaccharide-challenged human monocytes

IL-12 is a key mediator of the immune response, skewing T lymphocytes toward a type 1 cytokine pattern. Priming with IFN-gamma or GM-CSF is required for expression of IL-12p70 by cells in which IL-12 is inducible by bacterial products such as LPS. We here show for the first time that the production of bioactive IL-12 by human monocytes can be significantly suppressed by C5a if applied to IFN-gamma-primed monocytes before LPS stimulation. There was a dose-dependent suppression by IL-12 (p70) on the levels of intracellular cytokine production and cytokine secretion. mRNA studies consistently showed a reduction of IL-12p40 and IL-12p35 expression by stimulation in the presence of C5a. The results of several different experimental approaches suggest that IL-12 down-regulation was not due to endogenous IL-10, IL-4, or PGE2 production induced by C5a. Moreover, stimulation of IFN-gamma-primed monocytes with C5a did not lead to a down-regulation of the CD14 Ag, which is an LPS receptor. These findings show that the anaphylatoxin C5a has the capacity to directly interact with the complex regulation of IL-12.

Evaluation of C3a receptor expression on human leucocytes by the use of novel monoclonal antibodies

Varying results have been published in the past regarding the reactivity of different leucocyte subpopulations, including neutrophils, monocytes and B lymphocytes, to the anaphylatoxin C3a and its degradation product C3a(desArg). To better characterize the cellular distribution of C3a receptor (C3aR) expression, monoclonal antibodies against two different epitopes on the third extracellular domain of the human C3aR were generated. Quantification of C3aR as compared with C5aR densities was performed on peripheral blood leucocytes by quantitative indirect immunofluorescence. Eosinophils and basophils expressed similar numbers of C3aR and C5aR molecules/cell. On eosinophils 10 700+/-4500 (mean+/-SD) C3aR and 14 700+/-4100 C5aR were found, whereas basophils carried 8100+/-2100 C3aR and 13 500+/-3800 C5aR. Monocytes expressed approximately six times more C5aR than C3aR molecules on their surface (6000+/-2500 C3aR versus 34 100+/-9300 C5aR molecules) whereas on neutrophils, the expression of C5aR was more than 20 times higher than the expression of C3aR (3100+/-1000 C3aR versus 63 500+/-12 200 C5aR). No C3aR expression was detectable on peripheral blood-derived B lymphocytes and on tonsillar B cells before and after stimulation with interleukin-2/Staphylococcus aureus Cowan strain I. Our findings correspond well with the paucity of data on C3a-induced functional activities in monocytes and neutrophils and suggest that eosinophilic and basophilic granulocytes represent the primary effector cells in the peripheral blood which can be stimulated by C3a.

C3a(desArg) does not bind to and signal through the human C3a receptor

Contradictory results have been published in the past regarding the functional responses of different cell types to the anaphylatoxin C3a and its natural catabolite C3a(desArg). To elucidate the interaction of the C3a receptor (C3aR) with its ligand(s) we studied the binding of human recombinant C3a (rC3a) and rC3a(desArg) to RBL-2H3 transfectants which express the C3aR. As the addition of 11 aminoterminal amino acids did not alter the functional activity of the recombinant C3a as compared to serum-derived C3a the specific binding of rC3a and rC3a(desArg) to the transfectants could be determined by flow cytometry using a monoclonal antibody (mab) against their N-terminal histidine tag. Recombinant C3a bound to the C3aR with a half maximal concentration of about 3 nM whereas no evidence for a binding of rC3a(desArg) could be obtained. Furthermore, rC3a(desArg) did not signal through the C3aR. Neither the release of lysosomal N-acetyl-beta-D-glucosaminidase nor the directional migration of C3aR-expressing RBL-2H3 transfectants could be detected in response to rC3a(desArg) whereas rC3a was highly active in both assays. Our data demonstrate a defined ligand specificity of the C3aR for the anaphylatoxin C3a. Its natural catabolite C3a(desArg), however, does not signal through the C3aR. Modulating effects of C3a(desArg) on the synthesis of cytokines in human monocytes and B lymphocytes may therefore be induced by receptor-independent mechanisms while their in vivo relevance remains as yet undefined.

C5a receptor and interleukin-6 are expressed in tissue macrophages and stimulated keratinocytes but not in pulmonary and intestinal epithelial cells

The anaphylatoxin derived from the fifth component of the human complement system (C5a) mediates its effects by binding to a single high-affinity receptor (C5aR/CD88), the expression of which has been traditionally thought to be restricted to granulocytes, monocytes, macrophages (Mphi), and cell lines of myeloid origin. Recent immunohistochemical data suggested that human bronchial and alveolar cells express C5aR as well. To reexamine the tissue distribution of human C5aR expression, transcription of the C5aR gene was investigated in normal and pathologically affected human lung (bronchopneumonia, tuberculosis), large intestine (acute appendicitis, Crohn's disease), and skin (pyogenic granuloma, lichen planus) using in situ hybridization. In contrast to previous evidence, C5aR mRNA could not be detected in pulmonary or intestinal epithelial cells, whereas keratinocytes in inflamed but not in normal skin revealed detectable levels of C5aR transcripts. Additionally, it could be documented that only migrating Mphi express C5aR mRNA, whereas sessile Mphi in normal tissues and epithelioid/multinucleated Mphi found in granulomatous lesions do not. Because C5a has been demonstrated to upregulate the expression of interleukin (IL)-6 in human monocytes, we also studied IL-6 gene transcription in parallel to the C5aR. IL-6 mRNA was detectable in many tissue Mphi. Surprisingly, a tight co-expression of C5aR and IL-6 mRNA was observed in keratinocytes from lesions of pyogenic granuloma and lichen planus. These results point to an as yet unknown role for C5a in the pathogenesis of skin disorders beyond its well-defined function as a chemoattractant and activator of leukocytes.

Anaphylatoxin C3a but not C3a(desArg) is a chemotaxin for the mouse macrophage cell line J774

Varying results have been published regarding the functional reactivity of different cell types, including human monocytes, to the anaphylatoxin C3a and its degradation product C3a(desArg). To further delineate the functions of C3a and C3a(desArg) on this cell type we used the murine macrophage (Mø) cell line J774A.1 which is known to respond to the anaphylatoxin C5a. J774 cells specifically bound fluorescein isothiocyanate-labeled recombinant human C3a (rC3a). The cells migrated along rC3a concentration gradients in a dose-dependent manner with an optimal concentration of about 3 nM (rC5a:7 nM) and a half-maximal effective concentration (EC50) of about 1.2 nM (rC5a: 2 nM). The degradation product rC3a(desArg) was devoid of chemotactic activity. mRNA for the recently cloned G protein-coupled mouse high-affinity C3a receptor (C3aR) was detected in J774 cells, suggesting that this receptor represents the binding site for C3a on J774 Mø. In support of the specific nature of C3a-stimulated cellular mobility, RBL-2H3 transfectants expressing the human C3aR were also shown to migrate along gradients of rC3a (optimal concentration about 8 nM; EC50 about 3.5 nM) whereas rC3a(desArg) was again inactive. In summary, our findings demonstrate for the first time a specific, receptor-mediated chemoattraction of cells of the monocytic lineage to the anaphylatoxin C3a which may contribute to the accumulation of Mø at sites of inflammation.

A novel ELISA for the evaluation of the classical pathway of complement

Assessment of the overall function of the classical pathway of complement is traditionally performed by the hemolytic titration assay CH50. In the present study, we established a novel method for the quantitation of complement activity by measuring the deposition of C1q, C4, C3 and C9 on solid-phase IgM by an enzyme-linked immunosorbent assay (ELISA). Using the CH50 method as the reference, C9 deposition values displayed a sensitivity of 96.3% and a specificity of 99.4% in sera from patients with a variety of diseases. For C3, the sensitivity was 91.3% and the specificity 100%, for C4, the values were 95% and 100%, and for C1q the corresponding values were 52.9% and 98.9%. A close correlation was found between CH50 values below 30 U/ml and the deposition of C9 (r = 0.92), C3 (r = 0.91) and C4 (r = 0.92). In two patients with postinfectious glomerulonephritis normal C4 and C1q deposition was accompanied by decreased C3 and C9 deposition reflecting complement activation predominantly through the alternative pathway. In contrast, in two patients with complete C2 deficiency the deposition of C3 and C9 was undetectable together with normal C4 deposition values. Furthermore, in two patients with hereditary C1-inhibitor deficiency distinctly increased C1q deposition was accompanied by decreased C4 deposition values. In conclusion, the determination of complement deposition by ELISA represents a novel, quantitative method for the evaluation of complement activity. The measurement of C9 deposition alone or in combination with further complement proteins makes this ELISA a valuable tool for assessing the degree and level of complement consumption as well as localizing the missing protein in the case of complement deficiencies.

The human mast cell line HMC-1 binds and responds to C3a but not C3a(desArg)

Controversial results have been published in the past regarding the functional reactivity of different cell types to the anaphylatoxin C3a and its degradation product C3a(desArg). To understand better the effects of C3a and C3a(desArg) on human mast cells, the authors performed binding experiments and calcium mobilization studies on the human mast cell line HMC-1 which has been shown previously to express C3a binding sites. For this purpose, functionally active, recombinant C3a (rC3a) was constructed with an 11 amino acid peptide attached to the N-terminus of the molecule. Using a monoclonal antibody (MoAb) against this tag, binding of rC3a to HMC-1 cells could be demonstrated by flow cytometry. Its binding was specific as it could be blocked with serum-derived C3a. In contrast, no binding of rC3a(desArg) to HMC-1 cells was detectable. Recombinant C3a led to a transient mobilization of intracellular calcium [Ca2+]i in HMC-1 which was inhibitable by the C3a-specific MoAb K13/16. No increase of [Ca2+]i was detected when the cells were treated with C3a(desArg). The authors found C3a receptor (C3aR)-specific mRNA in HMC-1 cells indicating that this receptor represents the binding site for C3a on these cells. These results demonstrate a specific binding for C3a but not for C3a(desArg) on cells of the human mast cell line HMC-1. As a consequence, functional activity was restricted to C3a with C3a(desArg) being completely inactive. Therefore, the data strongly suggest that the recently cloned high affinity C3aR which is assumed to represent the binding site for the anaphylatoxin on HMC-1 cells is unresponsive to C3a(desArg).

Blood- and skin-derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient release of calcium via a pertussis toxin-sensitive signal transduction pathway

Controversial results have been published in the past regarding the functional reactivity of monocytes (Mo) and macrophages (M phi) to the anaphylatoxin C3a and its degradation product C3a(desArg). In this study we performed binding and calcium mobilization experiments with recombinant human C3a (rC3a) and rC3a(desArg). Blood Mo displayed non-inhibitable binding of FITC-labeled rC3a (rC3aFITC) but responded to rC3a with a transient release of the intracellular calcium concentration ([Ca2+]i), whereas rC3a(desArg) was completely inactive. In contrast, binding of rC3aFITC to eosinophilic granulocytes and the mast cell line HMC-1 which have been shown previously to express C3a binding sites could be blocked by a monoclonal anti-C3a antibody. The rC3a-induced [Ca2+]i release in blood Mo was pertussis toxin (PTX)-sensitive suggesting the involvement of G-proteins in the signal transduction pathway. Skin-derived Mo/M phi reacted similarly to blood Mo as no specific binding of rC3aFITC to these cells could be demonstrated, whereas an intracellular release of calcium ions in response to the anaphylatoxin was observed. Homologous desensitization to rC3a but not heterologous desensitization to rC5a was detected in further experiments. The functional effect of C3a, but not the unspecific binding of rC3aFITC to blood Mo and skin-derived Mo/M phi could be blocked by the monoclonal anti-C3a antibody. These results suggest the expression of the recently cloned G-protein-coupled receptor for C3a on human blood Mo and skin-derived Mo/M phi. However, the total number of specific C3a binding sites on these cells is distinctly lower as compared to eosinophilic granulocytes and cells of the mast cell line HMC-1. The small number of C3a receptors on Mo/M phi may be masked by a pronounced non-inhibitable binding of rC3aFITC. This binding, however, may contribute to the recently described biological effects of C3a(desArg) on Mo.

Molecular cloning and expression of the functional rat C5a receptor

The C5a receptor belongs to the superfamily of G-protein coupled receptors with seven transmembrane segments. In this study we report on the cloning of the rat C5a receptor (ratC5aR). We used a hybridization probe produced by PCR utilizing degenerate primers which corresponded to conserved parts of the human, canine and murine C5a receptor nucleotide sequences and to the published partial amino acid sequence of the rat C5a receptor to screen a rat macrophage cDNA library. We found two overlapping clones containing an open reading frame of 1056 bp, a 3'untranslated region of 683 bp and a 5'untranslated region of 27 bp. The overall nucleotide acid sequence identity, compared to the murine, human and canine C5a receptor sequences, was 85.8, 70.5 and 68.9%, respectively. The greatest diversity exists in the putative extracellular domains, especially in the aminoterminal domain which is assumed to be involved in ligand binding. An N-glycosylation site is present within the N-terminal sequence at residue 6. One of the cDNA containing the 5'untranslated region, the coding sequence and part of the 3'untranslated region was cloned into an eucaryotic expression vector and stably transfected into the rat basophilic leukemia cell line RBL-2H3. Expression of the rat C5a receptor on the surface of these cells could be demonstrated by flow cytometric analysis using FITC-labeled recombinant rat C5a (rrC5a). By measuring the release of calcium from intracellular stores after stimulation with rrC5a it could further be shown that the receptor is functionally coupled. Receptor binding assays showed that rrC5a specifically binds to the ratC5aR with a KD of 0.91 +/- 0.36 and to the human C5a receptor (huC5aR) with a KD of 7.19 +/- 1.56. The determined KD for binding of human C5a (huC5a) to the huCSaR was 2.16 +/- 0.65. No binding of huC5a to the ratC5aR could be observed although high concentrations of this ligand (> 60 nM) promoted chemotaxis of RBL cells transfected with the huC5aR.

Chemoattractant receptors for interleukin-8 and C5a: expression on peripheral blood leukocytes and differential regulation on HL-60 and AML-193 cells by vitamin D3 and all-trans retinoic acid

Two homologous high-affinity receptors for the chemoattractant interleukin-8, IL-8RA and IL-8RB, and one for the chemoattractant C5a (C5aR) have been cloned. These membrane proteins are members of the rhodopsin superfamily of G-protein coupled seven-transmembrane segment receptors. New monoclonal antibodies (mAb) directed against the deduced N-terminal sequences of the IL-8RA (mAb SE2) and IL-8RB (mAb HC2) were generated to determine the IL-8R expression on human blood leukocytes and two human myeloid cell lines. The C5aR expression was detected using the mAb W17/1. Approximately 107,000 C5aR, 55,000 IL-8RA, and 25,000 IL-8RB molecules per cell could be detected on human granulocytes by flow cytometric analysis. On peripheral blood monocytes, 42,000 C5aR molecules/cell and 3000 IL-8RB molecules/cell were expressed. However, we were unable to quantitate IL-8RA expression, which was detectable but below 2500 molecules per cell and thus outside the standard range for the quantitation of receptor molecules by flow cytometry. On AML-193 cells, only the IL-8RB was constitutively expressed, whereas on HL-60 cells, we could not detect expression of any of the three receptors. Vitamin D3 (250 ng/ml, 7 days), which has been shown to induce differentiation of AML-193 and HL-60 cells into the monocytic phenotype, led to an up-regulation of IL-8RB and C5aR in both cell lines in the absence of any expression of IL-8RA. In contrast, all-trans retinoic acid (0.1 microM, 7 days), which induces differentiation into the granulocytic phenotype, led to an up-regulation of IL-8RB in AML-193 cells and to an expression of IL-8RB and C5aR in HL-60 cells. Again, neither cell line expressed IL-8RA. These findings suggest that regulation of IL-8RA expression differs from that of its IL-8RB homolog and may be a late event in leukocyte maturation.

Activated complement C3: a potentially novel predictor of progressive IgA nephropathy

In the search for a serologic marker of disease activity, we measured concentrations of activated C3 (actC3, that is, neoantigens developing after C3 activation on breakdown products), C4-C3 complexes and soluble C5b-9 (sC5b-9) in one or two plasma samples from adult patients with IgA nephropathy (IgAN, N = 50) or Henoch-Schönlein purpura (HSP, N = 4). As controls, 20 patients with non-immune renal disease, but comparable age, degree of proteinuria, renal dysfunction and prevalence of hypertension were studied. Compared to controls, actC3 levels were elevated in 30% of the patients with IgAN and one of the HSP patients. C4-C3 complexes were elevated in only 8% of the IgAN patients, and sC5b-9 levels were within the control range in all IgAN and HSP patients. In IgAN patients with elevated actC3 levels, proteinuria and hematuria were more pronounced than in those with normal levels. Elevated plasma concentrations of actC3 at the first presentation correlated with subsequent deterioration of renal function both in patients with initially normal and already impaired renal function (r = -0.56, N = 44, P = 0.003). The five IgAN patients with elevated actC3 on both occasions of obtaining plasma showed the most rapid loss of renal function. We conclude that mainly alternative pathway complement activation can be demonstrated in patients with IgAN and HSP. In IgAN patients the presence of complement activation is associated with more severe renal disease. Further studies are warranted to examine the clinical usefulness of actC3 as a predictor of the subsequent course of IgAN.

Nucleotide and corrected amino acid sequence of the functional recombinant rat anaphylatoxin C5a

For bacterial expression of rat anaphylatoxin C5a, the cDNA was amplified by reverse transcriptase-polymerase chain reaction (RT-PCR) using rat liver RNA and degenerate primers designed according to the published amino acid sequence [1]. Surprisingly, the amino acid sequence deduced from cDNA differed at positions 55 (N for K), 63 (K for H), 67 (E for N), 68 (S for E) and 69 (H for S) from the published sequence. The overall amino acid composition, however, was unchanged because these 5 amino acids were located at different positions compared to the published sequence. As a consequence, the proposed N-glycosylation site was absent, suggesting O-glycosylation of the mature molecule. Recombinant rat C5a with a 6 histidine tag at the N-terminus was expressed in bacteria, purified and renatured. The peptide was as potent as recombinant human C5a in eliciting lysosomal enzyme release from human granulocytes.

C5a receptors are detectable on mast cells in normal human skin and in psoriatic plaques but not in weal and flare reactions or in uticaria pigmentosa by immunohistochemistry

The expression of the receptor for the anaphylatoxin C5a on mast cells was studied with three monoclonal antibodies directed to the N-terminal domain of the C5a receptor. Human skin was investigated by immunohistology applied to sequential 2 micron sections of acrylate-embedded tissues. All anti-C5a receptor antibodies stained c-kit+ or tryptase+ cells which were metachromatic after toluidine blue staining in normal human skin. The binding of anti-C5a receptor antibodies was inhibitable by a peptide representing the first 31 amino acids of the C5a receptor. A similar expression of C5a receptors was found on mast cells in chronic psoriatic plaques. However, C5a receptors were not detectable on mast cells in weal and flare reactions or in lesional skin of uticaria pigmentosa. These findings suggest that (1) anti-C5a receptor antibodies directed to the N-terminal domain of the receptor are suitable tools for the identification of mast cells in acrylate-embedded sections of human skin, (2) mast cell activation in weal and flare reactions results in C5a receptor downregulation or receptor blockade and (3) mast cells in urticaria pigmentosa lack a typical marker of normal human skin mast cells.

CD88 antibodies specifically bind to C5aR on dermal CD117+ and CD14+ cells and react with a desmosomal antigen in human skin

The expression of the C5aR (CD88) on human epidermal and dermal cells was studied with five anti-C5aR mAb directed to the N-terminal domain of the receptor. All mAb bound to suspended dermal CD117+ mast cells and to dermal CD14+ cells. The binding to CD14+ and CD117+ cells could be blocked by rC5a and by peptide EX-1 representing amino acid residues 1-31 of the C5aR. In acetone-fixed frozen or in paraformaldehyde-fixed, paraffin-embedded tissue, we detected a binding of the Abs to dermal perivascular cells and, additionally, to keratinocytes and dermal epithelial cells that could be blocked by EX-1. Immunoelectromicroscopy revealed a binding of anti-C5aR mAb to desmosomal regions in human epidermis. However, the following results indicate that CD88 mAb cross-react with epithelium in a specific way: 1) the binding to suspended epidermal cells and to the epidermal cell line HaCat could be blocked by EX-1 but not by rC5a; 2) FITC-labeled C5a bound to CD117+ and to CD14+ cells but not to epidermal cells; 3) C5a led to transient calcium fluxes in CD14+ and CD117+ dermal but not in epidermal cells; 4) C5aR mRNA was detectable by reverse transcription PCR in granulocytes but not in keratinocytes or in HaCat. Our results show that CD88 mAb are good tools for the investigation of the C5aR on hemopoietic cells. Results with epithelial cells should be considered with caution, as the binding of CD88 mAb that were raised to a synthetic peptide sequence may be due to a cross-reactivity.

CD88 antibodies specifically bind to C5aR on dermal CD117+ and CD14+ cells and react with a desmosomal antigen in human skin

The expression of the C5aR (CD88) on human epidermal and dermal cells was studied with five anti-C5aR mAb directed to the N-terminal domain of the receptor. All mAb bound to suspended dermal CD117+ mast cells and to dermal CD14+ cells. The binding to CD14+ and CD117+ cells could be blocked by rC5a and by peptide EX-1 representing amino acid residues 1-31 of the C5aR. In acetone-fixed frozen or in paraformaldehyde-fixed, paraffin-embedded tissue, we detected a binding of the Abs to dermal perivascular cells and, additionally, to keratinocytes and dermal epithelial cells that could be blocked by EX-1. Immunoelectromicroscopy revealed a binding of anti-C5aR mAb to desmosomal regions in human epidermis. However, the following results indicate that CD88 mAb cross-react with epithelium in a specific way: 1) the binding to suspended epidermal cells and to the epidermal cell line HaCat could be blocked by EX-1 but not by rC5a; 2) FITC-labeled C5a bound to CD117+ and to CD14+ cells but not to epidermal cells; 3) C5a led to transient calcium fluxes in CD14+ and CD117+ dermal but not in epidermal cells; 4) C5aR mRNA was detectable by reverse transcription PCR in granulocytes but not in keratinocytes or in HaCat. Our results show that CD88 mAb are good tools for the investigation of the C5aR on hemopoietic cells. Results with epithelial cells should be considered with caution, as the binding of CD88 mAb that were raised to a synthetic peptide sequence may be due to a cross-reactivity.

The human mast cell line HMC-1 expresses C5a receptors and responds to C5a but not to C5a(desArg)

The expression of the receptor for the anaphylatoxin C5a (C5aR, CD88) on the human mast cell line HMC-1 was studied with four anti-C5aR monoclonal antibodies directed to the N-terminal domain of the receptor. All antibodies bound to the human mast cell line HMC-1. The binding could be blocked by recombinant C5a and by peptide EX-1 representing amino residues 1-31 on the N-terminal domain of the C5aR. In addition, FITC-labelled C5a bound to HMC-1, and this binding could be blocked by unlabelled C5a or C5aR antibodies. C5aR-specific mRNA was detected in HMC-1 cells by RT-PCR which confirmed the expression of the C5aR gene made by these cells. Lymphocyte-conditioned medium, interferon-gamma or phorbol esters which have been shown to induce a down-regulation of C5aR on myeloid cells did not influence the expression of C5aR on HMC-1. C5a led to a transient mobilization of intracellular calcium in HMC-1 which could be inhibited by pre incubation of C5a with a C5a-specific antibody. In contrast to findings with granulocytes, HMC-1 did not respond to C5a(desArg), confirming previous findings with human skin mast cells. The findings show that (i) although HMC-1 differ from granulocytes in their responsiveness to C5a(desArg), they express similar C5aR and (ii) HMC-1 cells resemble skin mast cells in the expression and function of C5aR and may therefore serve as a model in future studies addressing the biology of this anaphylatoxin receptor on skin mast cells.

A novel ELISA for the assessment of classical pathway of complement activation in vivo by measurement of C4-C3 complexes

Measurements of complement split products by enzyme-linked immunosorbent assays (ELISA) are well established for the assessment of in vivo complement activation. We have combined two monoclonal antibodies (mAb) with specificities for C3b/iC3b/C3dg (mAb I3/15) and C4/C4b/C4d (mAb M4d2), respectively, in a sandwich ELISA to quantitate C4-C3 complexes as an indicator of complement activation. Serum incubated with heat aggregated IgG (HAG) was used as a standard and the C4-C3 levels expressed as microgram equivalent HAG/ml (microgram HAG-equ/ml). Normal values of C4-C3 complexes in plasma (EDTA) of healthy probands (n = 11) were 6.3 micrograms HAG-equ/ml +/- 1.5 (mean +/- 1 standard deviation (SD), with a range from 3.6 to 9.1). In patients with systemic lupus erythematosus (SLE, n = 23) C4-C3 values were clearly elevated (48.8 micrograms HAG-equ/ml +/- 52.9, range 7.5-184.7) as compared to samples from patients with idiopathic hypertension (IDH, n = 10) (6.5 micrograms HAG-equ/ml +/- 1.7, range 4.1-9.4). For SLE patients C4-C3 levels significantly correlated with values for C3b/iC3b/C3d (r = 0.69, p < 0.001) and C3 containing immune complexes (r = 0.68, p < 0.001), but not with the C4d fragment (r = 0.26). C4-C3 levels of 96% of the studied SLE patients were increased more than 2 SD above the normal mean as compared to 74% of C4d and activated C3 values, respectively. Serum treated with zymosan as an activator of the alternative pathway of complement did not exhibit higher C4-C3 values. These results demonstrate that the quantitation of in vivo generated C4-C3 complexes by ELISA provide a novel, sensitive parameter for classical pathway of complement activation.

Stimulation of glycogen phosphorylase in rat hepatocytes via prostanoid release from Kupffer cells by recombinant rat anaphylatoxin C5a but not by native human C5a in hepatocyte/Kupffer cell co-cultures

Human anaphylatoxin C3a had previously been shown to increase glycogenolysis in perfused rat liver and prostanoid formation in rat liver macrophages. Surprisingly, human C5a, which in other systems elicited stronger responses than C3a, did not increase glycogenolysis in perfused rat liver. Species incompatibilities within the experimental system had been supposed to be the reason. The current study supports this hypothesis: (1) In rat liver macrophages that had been maintained in primary culture for 72 h recombinant rat anaphylatoxin C5a in concentrations between 0.1 and 10 micrograms/ml increased the formation of thromboxane A2, prostaglandin D2, E2 and F2 alpha 6- to 12-fold over basal within 10 min. In contrast, human anaphylatoxin C5a did not increase prostanoid formation in rat Kupffer cells. (2) The increase in prostanoid formation by recombinant rat C5a was specific. It was inhibited by a neutralizing monoclonal antibody. (3) In co-cultures of rat hepatocytes and rat Kupffer cells but not in hepatocyte mono-cultures recombinant rat C5a increased glycogen phosphorylase activity 3-fold over basal. This effect was inhibited by incubation of the co-cultures with 500 microM acetylsalicyclic acid. Thus, C5a generated either locally in the liver or systemically e.g. in the course of sepsis, may increase hepatic glycogenolysis by a prostanoid-mediated intercellular communication between Kupffer cells and hepatocytes.

Classical pathway of complement activation in mammalian kidneys

Two monoclonal antibodies (mAb M4d2 and M4d3) specific for the alpha 2-fragment (C4d), and one antibody (mAb M4c3) specific for the gamma-chain of human complement protein C4, have been tested for cross-reactivity against mammalian complement. These mAb have also been found to react with C4 from guinea-pig (mAb M4d2 and M4c3) as well as from cattle, baboon and rhesus monkey (mAb M4d3 and M4c3) in an activation ELISA. Since reactivity of mAb M4d2 and M4c3 included guinea-pig complement, the specific recognition of mammalian C4 could be confirmed with sera from C4-deficient (def') guinea-pigs. mAb M4d2 or M4d3, but not mAb M4c3, stained glomerular deposits within renal tissue sections from pig, cattle and guinea-pig. In the case of mAb M4d2, specificity of that staining could also be demonstrated in kidney specimens from C4-def' guinea-pigs. It can be concluded that, as in humans, the C4d fragment is also present in mammalian glomeruli. Compared with normal guinea-pigs, the C4-def' and C2-def' animals showed markedly increased glomerular deposits of IgM. It appears that glomerular deposition of complement C4d in mammals: (1) indicates activation via the classical pathway; (2) represents a general phenomenon of renal homeostasis; and (3) seems to be involved in the physiological clearance of immune complexes.

Expression of a functional chimeric Ig-MHC class II protein

We have generated a chimeric protein molecule composed of the alpha- and beta-chains of the MHC class II I-E molecule fused to antibody V regions derived from anti-human CD4 mAb MT310. Expression vectors were constructed containing the functional, rearranged gene segments coding for the V region domains of the antibody H and L chains in place of the first domains of the complete structural genes of the I-E alpha- and beta-chains, respectively. Cells transfected with both hybrid genes expressed a stable protein product on the cell surface. The chimeric molecule exhibited the idiotype of the antibody MT310 as shown by binding to the anti-idiotypic mAb 20-46. A protein of the anticipated molecular mass was immunoprecipitated with anti-mouse IgG antiserum. Furthermore, human soluble CD4 did bind to the transfected cell line, demonstrating that the chimeric protein possessed the binding capacity of the original mAb. Thus, the hybrid molecule retained: 1) the properties of a MHC class II protein with regard to correct chain assembly and transport to the cell surface; as well as 2) the Ag binding capacity of the antibody genes used. The generation of hybrid MHC class II molecules with highly specific, non-MHC-restricted binding capacities will be useful for studying MHC class II-mediated effector functions such as selection of the T cell repertoire in thymus of transgenic mice.

Expression of a functional chimeric Ig-MHC class II protein

We have generated a chimeric protein molecule composed of the alpha- and beta-chains of the MHC class II I-E molecule fused to antibody V regions derived from anti-human CD4 mAb MT310. Expression vectors were constructed containing the functional, rearranged gene segments coding for the V region domains of the antibody H and L chains in place of the first domains of the complete structural genes of the I-E alpha- and beta-chains, respectively. Cells transfected with both hybrid genes expressed a stable protein product on the cell surface. The chimeric molecule exhibited the idiotype of the antibody MT310 as shown by binding to the anti-idiotypic mAb 20-46. A protein of the anticipated molecular mass was immunoprecipitated with anti-mouse IgG antiserum. Furthermore, human soluble CD4 did bind to the transfected cell line, demonstrating that the chimeric protein possessed the binding capacity of the original mAb. Thus, the hybrid molecule retained: 1) the properties of a MHC class II protein with regard to correct chain assembly and transport to the cell surface; as well as 2) the Ag binding capacity of the antibody genes used. The generation of hybrid MHC class II molecules with highly specific, non-MHC-restricted binding capacities will be useful for studying MHC class II-mediated effector functions such as selection of the T cell repertoire in thymus of transgenic mice.

Classical pathway of complement activation in normal and diseased human glomeruli

Monoclonal antibodies (mAb) reactive against complement components involved in the classical activation pathway were applied in an indirect immunoperoxidase technique for the histological study of normal and diseased human renal tissues. Prominent staining with antibodies against the C4d fragment was seen in all glomeruli and some renal arteriolar walls. The C4d staining was mesangial with light microscopy, whereas the subendothelial site of the glomerular basement membrane (GBM) also appeared to be positive in immunoelectron microscopy. In similar localization, albeit with distinctly weaker intensity, IgM and C4 binding protein (C4bp) were detected. In kidney biopsies from patients with various types of glomerulonephritis, C4d reactive antibodies stained the glomerular structures in a strong, diffuse or granular pattern in contrast to the more segmental distribution and weaker staining intensity in normal kidney specimens. Increased amounts of C4d, occasionally also of C4b, were paralleled in diseased kidney tissues by distinct deposits of IgM and/or IgG in the presence of C4bp. This study suggests that the C4d fragment in normal human glomeruli is indicative of a continuous, local complement activation via the classical pathway induced by the physiological deposition of IgM-containing immune complexes.

Deposition of complement activation products on plastic-adsorbed immunoglobulins. A simple ELISA technique for the detection of defined complement deficiencies

The activation of complement components in human serum has been studied using immunoglobulins adsorbed to microtiter plates. The sequential deposition of complement fragments was detected by a series of mono- and polyclonal antibodies in an indirect enzyme-linked immunosorbent assay (ELISA). Antibodies against C1q, C1s, C4b/d, C3b/d, factor B, C5b-9 membrane attack complex (MAC), the regulatory complement proteins C4 binding protein (C4bp) and properdin were reactive. Several lines of evidence suggest that complement activation was via the classical pathway: (1) complement activation was highly isotype-restricted with regard to the adsorbed Igs (human IgG1 and IgG3 as well as mouse IgM, IgG2a and IgG2b isotypes are strong activators in contrast to human IgG2, IgG4, IgA and mouse IgG1); (2) Ca2+ depletion, heat treatment (56 degrees C for 45 min), incubation with 0.5 M KSCN or heat-aggregated immunoglobulins (aggIgG) abrogated serum activity; (3) complement deficient sera (C1q def', C2 def', C6 def' human sera; C2 def', C4 def' guinea pig sera) showed impaired deposition of the complement components that follow the missing component in the cascade of activation. In a clinical study sera from patients with systemic lupus erythematosus (SLE) were investigated in order to measure the effect of hypocomplementemia due to complement consumption. The results obtained suggest that this new and simple assay is well suited for (1) the detection of various inherited complement deficiencies, (2) the semiquantitative evaluation of sera with decreased complement levels, (3) a more detailed study of complement components bound to a solid phase.

Complement activation in human lymphoid germinal centres

The presence of complement activation products has been studied in morphologically normal human lymphatic tissue from tonsil, spleen and lymph node. Newly established monoclonal antibodies (mAbs) with reactivity against the C4 cleavage fragments C4a, C4b, C4c and C4d were applied on cyrostat sections in the indirect immunoperoxidase staining technique. Irrespective of organ type, C4d activation product could be detected in germinal centres of all secondary lymphoid follicles. To substantiate this finding, the complete sequence of complement activation products was investigated by a series of mono- and polyclonal antibodies to the complement proteins C1, C2, C3, factor B, C5, C9 to C5b-9 neoantigens and to the regulatory complement proteins C4 binding protein (C4bp), factor I, factor H and properdin. Similar to C4d, all secondary follicles exhibited a strong staining reaction for C3d antigens restricted to germinal centres. At the same site, albeit with distinctly weaker intensity, components of the membrane attack complex (MAC) C5b-9 were found. The simultaneous deposition of C1, C4b and C4bp in certain germinal centres indicates that complement activation is induced via the classical pathway. Concomitant deposition of IgM suggests IgM-antigen complexes that have been trapped on follicular dendritic cells (FDC) during normal immune response as the most likely candidates for activators of the classical pathway. Our data demonstrate that human lymphoid germinal centres as important sites of immune regulation closely interrelate with the complete cascade of complement-activation products, including the membrane attack complex (MAC).

Biosynthesis of complement C4 messenger RNA in normal human kidney

Complementary DNA (cDNA) probes were used to investigate the extrahepatic production of the major histocompatibility complex (MHC)-linked complement components C4, factor B and C2 in various normal human tissues. The presence of the corresponding messenger RNA (mRNA) was tested by Northern blot analysis. Complement C4 mRNA was found in liver, and with high intensity also in normal kidneys. In contrast, no C2 mRNA and only very low amounts of factor B mRNA could be detected in the kidney. Slot blot hybridization was performed to quantitate the amount of C4 mRNA, and the intensity of C4 mRNA hybridization in the kidney samples was about 25% compared with liver RNA. C4-specific transcripts were not present in isolated glomeruli but in the renal interstitium. Other human tissues, such as tonsil, spleen, thymus, brain, lung and peripheral mononuclear cells, contained no C4 mRNA. Low amounts of C4 mRNA were found in colon, thyroid gland, lymph node and breast carcinoma. The results obtained with lung, where C2 mRNA was found but no C4 mRNA, further indicate an independent, tissue-specific regulation of the class III gene expression. The results, showing that the complement C4 genes are transcribed very efficiently in normal human kidney, suggest a direct role of complement C4 in renal pathogenesis.

Human complement factor H: expression of an additional truncated gene product of 43 kDa in human liver

The human complement factor H is an important factor in the control of the alternative pathway and also induces the stimulation of B cells and macrophages in vitro. Using a human factor H cDNA clone as probe, two factor H-specific transcripts of 4.4 and 1.8 kb were detected in four human livers. Both mRNAs were found independently of the expression of an acute phase marker SAA, in these livers, indicating that their presence is not linked to an acute phase state. The shorter transcript was cloned in two cDNA plasmids H-19 and H-20, lacking only seven amino acids of the N-terminus of the mature factor H protein. The deduced protein sequence showed that this protein is identical to the N-terminal portion of the large classical factor H of 150 kDa mol. mass. Parallel to the finding that the N-terminal sequence of factor H is expressed by two distinct mRNA species, evidence is presented that the C-terminal sequence is also contained on two different transcripts, the common 4.4-kb mRNA and an additional 1.0-kb mRNA. A novel, short form of human complement factor H of 43 kDa was detected in human sera which represents the translation product of the 1.8-kb factor H-specific mRNA detected in human liver. Five distinct epitopes detected with 6 monoclonal antibodies are present on both the 150-kDa factor H protein and the truncated 43-kDa molecule. We conclude that additional H-specific mRNAs are found in human liver and that at least one of them is translated yielding a truncated form of factor H.

Detection of both isotypes of complement C4, C4A and C4B, in normal human glomeruli

Monoclonal antibodies reactive against the complement C4A and C4B isotypic components were used in an immunoperoxidase technique for the histological study of normal human renal tissue. Prominent staining with both antibodies was seen in the mesangial areas of all normal kidney sections investigated. Occasional staining of arteriolar walls of the same tissues, however, was also observed. In contrast, no mesangial staining was seen using monoclonal antibodies reactive against other 'early' complement components, such as C1q and C3. Specificity of the glomerular staining with the anti-C4 reagents was demonstrated in two patients possessing only the C4A serum component but lacking genetically the C4B locus products. As would be predicted, glomerular staining with the anti-C4A reagent, but not anti-C4B, was clearly demonstrable. It is concluded that both isotypes of complement C4 are present in normal human glomeruli and thus might be operative for normal mesangial function.