The physiological relevance of naturally occurring IgM-ALA remains to be elucidated. These autoantibodies are present from birth and increase in diverse inflammatory states that are both infectious and noninfectious. Clinical observations showing significantly less acute allograft rejections in recipients having high IgM-ALA levels, led us to investigate whether IgM-ALA could have a functional role in attenuating T cell mediated inflammatory responses. In pursuit of this hypothesis, we did studies using IgM purified from the serum of normal individuals, patients with end stage renal disease, and HIV-1 infection. All preparations of IgM immunoprecipitated certain receptors e.g., CD3, CD4, CCR5, and CXCR4 from whole cell lysates but failed to immunoprecipitate IL-2R and HLA Ags. In physiological doses IgM down-regulated CD4, CD2 and CD86 but not CD8 and CD28, inhibited T cell proliferation, decreased production of certain proinflammatory cytokines e.g., TNF-α, IL-13 and IL-2, but not IFN- γ, IL-1β, GM-CSF, IL-6 and IL-8 and inhibited leukocyte chemotaxis. These inhibitory effects were more pronounced when using IgM from patients with high levels of IgM-ALA and these inhibitory effects were significantly reduced after using IgM preabsorbed with leukocytes. IgM-ALA binding to leukocytes was found to be highly specific, as <10% of IgM secreting B cell clones had IgM-ALA specificity with some clones having specificity for either T cells or monocytes. These findings support the concept that IgM-ALA provides an innate mechanism to regulate T cell mediated inflammatory responses.

Repertoires of naturally occurring self-reactive Abs of different isotypes have been intensively studied during the last three decades (1, 2, 3, 4). These Abs, which are produced in the absence of deliberate immunization or exposure to foreign Ags, are detectable at birth, but the full repertoire develops by early childhood and is maintained throughout life (5, 6, 7, 8). Data in mice clearly show that these naturally occurring autoantibodies are produced by a subset of B cells (i.e., B1 cells), which produce Abs independently of T cell help (5, 6, 9). Additionally, such naturally occurring Abs are encoded by minimally or nonmutated germline genes and hence are characteristically polyreactive with low binding affinity and, therefore, differ from disease-producing autoantibodies in that the latter are predominantly of the IgG isotype and bind with high affinity and specificity to the autoantigen (1, 10, 11).

We have been interested in the naturally occurring anti-leukocyte autoantibodies which are predominantly of the IgM isotype and these Abs were initially discovered because of their binding reactivity to lymphocytes (13, 14). B lymphocytes producing IgM anti-leukocyte Abs can be found in the umbilical cord blood as is the case with other B cell secreting, naturally occurring autoantibodies (12). Such Abs will be referred to as IgM-anti-leukocyte autoantibodies (IgM-ALA).3 These IgM-ALA are present at low levels in normal individuals and increase during inflammatory disorders and various infections including a subset of HIV-1 patients and end stage renal dialysis patients (ESRD) (13, 14, 15, 16, 17, 18, 19). Previous studies have demonstrated that IgM-ALA are a heterogeneous group of several Abs that are reactive to autologous and allogeneic cells and have specificities for different, largely undefined membrane receptors including receptors with phospholipids and glycolipids (20, 21, 22, 23, 24, 25, 26). These Abs are not cytolytic in the presence of complement at 37°C but are cytolytic at colder temperatures, i.e., 18–20°C (13, 14, 15). Human kidney and heart transplants performed in the subset (approx. 30%) of patients having high levels of IgM-ALA have a lower incidence of acute rejections and of less severity, thus permitting better graft survival (27, 28, 29, 30, 31). The protective effect of IgM-ALA in allograft transplantation together with the observation that IgM-ALA nonspecifically increase in various inflammatory and infective states led us to hypothesize whether the increased production of IgM-ALA has a role in inhibiting T cell activation and attenuating the inflammatory response by binding to certain cell membrane receptors important in the inflammatory process i.e., receptors that activate T lymphocytes and receptors that are important for chemotaxis.

To investigate this possibility, we performed in vitro studies using IgM purified from normal, HIV, and ESRD serum. We show that IgM preparations from normal, HIV-1 and ESRD individuals’ immunoprecipitate CD3, CD4, CCR5, and CXCR4 (but not HLA and CD25) from whole cell lysates of lymphoid cell lines. There was however, a marked individual variation in levels of IgM with specificity to a single receptor (as determined by Western blots) and this variation was present in normals as well as disease states. The data show that polyclonal preparations of IgM with high binding reactivity to CD3, CD4 and chemokine receptors inhibits T cell activation and proliferation in response to alloantigens and anti-CD3 (OKT3) as well as inhibits binding of chemokines to their receptors.

Briefly, in this assay Ficoll-Hypaque isolated mono-nuclear cells are initially digested with pronase to remove receptors that bind to the Fc domain of Abs (e.g., CD16 that binds to Fc of IgG) and then reacted with recipients sera for 30 min at 4°C, washed two times at 4°C, and then interacted with FITC-labeled goat anti-human IgM (Fc specific) and PE anti-CD3 before quantitating IgM binding to T and non-T cells with two-color flow cytometry (32). However, cell lines and neutrophils were not predigested with pronase as the enzyme reduced their viability. In these experiments, 2 × 105 cells were interacted with 25 μl of undiluted sera. Quantity of serum IgM bound to T lymphocytes (IgM-ALA) is classified as 1) “no” if there is <20 mean channel fluorescence (MCF) shift of IgM binding to T cells when compared with T cells before adding serum, 2) “low” if the MCF shift varies between 20 and 200 channels, and 3) “high” when the MCF shift is >200 channels. The levels of serum IgM-ALA binding to allogeneic donor T lymphocytes was similar to binding on autologous T cells.

IgM was purified by size exclusion column chromatography (Sephacryl S-300 HR) from heat-inactivated sera (56°C) of normal individuals, HIV-1 infected and ESRD patients. IgM was not isolated by dialyzing sera in water or by ammonium chloride precipitation as both these techniques yielded IgM with impaired functional activity. Contaminating IgG was removed from the IgM preparation by exposure to protein G-agarose (Sigma-Aldrich). We used these procedures to remove low m.w. substances (e.g., chemokines) and IgG Abs that could affect the interpretation of our data. Serum protein electrophoresis and immunoelectrophoresis revealed that these IgM preparations contained IgG (<1%), albumin (<3%), and other proteins (<1%). We did not affinity purify these IgM Abs as such procedures e.g., binding of IgM to mannan binding protein or binding of IgM to agarose coupled with goat anti-human IgM yielded 10–15% of the starting IgM and has the potential of depleting certain IgM subsets. Instead, in several experiments we used as controls pooled IgG, IgA, albumin and α 2 macroglobulins to determine whether our observations could be explained by some of these minor contaminants. We did not detect CCL5 (RANTES) or CXCL12 (SDF-1α) in these IgM preparations using Western blot or ELISA techniques. All IgM preparations were absorbed with mouse, goat and rabbit IgG (bound to agarose beads) to remove IgM with rheumatoid factor activity that could interfere with the functional activity of animal IgG used either in cell cultures (e.g., anti-CD3) or for receptor detection using flow cytometry.

In these studies we compare ESRD or HIV IgM to normal IgM. ESRD IgM was obtained from sera with high IgM anti-lymphocyte activity. All IgM preparations were used in concentrations of 5–30 μg/L × 106 cells/ml in functional assays. In some experiments, IgM from either normal, HIV, or ESRD individuals were pooled, while in other experiments, single individual IgM was used. IgM, obtained from a patient with Waldenstrom macroglobulinemia and having a monoclonal IgM-ALA Ab that bound to an undefined leukocyte receptor was used as our negative control. This Waldenstrom IgM had ≈25% nonmonoclonal IgM as well.

IgM (2.5 ml) at 0.2 mg/ml in RPMI 1640 was absorbed twice, each time for 45 min at 37°C in 5% CO2 with a pool of 280 × 106 cells (washed two times) consisting of equal numbers of Jurkat, SupT-1, Daudi, and U937 cells (suspended in 2 ml of medium). We used these cell lines as these cells express most of the leukocyte membrane receptors including CD3, CD4, and chemokine receptors. The IgM was centrifuged at the end of 45 min to remove cells, and the absorbed IgM was quantitated using ELISA techniques. In the absorption technique, 50–60% of IgM was lost. Absorbed IgM had <15% residual binding activity to Jurkat cells, U937 cells, lymphocytes, neutrophils, or cultured endothelial cells as determined by flow cytometry. Absorbed IgM was re-run through the column (Sephacryl S-300 HR) to remove cytokines that may have accompanied the large number of leukocytes.

Monomeric IgM was made from the pentameric form in 200 nM Tris, 150 mM NaCl, and 1 mM EDTA (pH 8.0), and by reduction with 5 mM DTT for 2 h at room temperature. Subsequent alkalinization was performed for 1 h on ice with 12 mM iodoacetamide. IgM monomers were isolated from any remaining pentameric forms by column chromatography (Superdex-200) equilibrated with PBS. Purity of monomeric IgM was confirmed with SDS-PAGE Western blots under reducing and nonreducing conditions. With flow cytometry, one observed <20% reduction in binding of monomeric IgM to lymphocytes when compared with the pentameric form.

All the cell lines were obtained from the American Type Culture Collection Program at National Institutes of Health. SupT-1 and Jurkat cell lines are human CD4 and CD3 positive T cell leukemia lines constitutively expressing high levels of CXCR4. Hut78 is a human T cell line that secretes IL-2 and constitutively expresses IL-2R, CD-8, CCR5 and CXCR4. U937 is a human monocytoid leukemia monocyte cell line constitutively expressing CD4, CXCR4, CCR5, CCR4, CCR3, CCR2B and CCR1, but not CD3. Daudi is a B cell lymphoma cell line expressing HLA-class II, CXCR4 and CCR5 but not CD3, HLA-class I and no membrane Ig.

Anti-CCR5 murine IgG monoclonals (clones 2D7, CTC-5, 45502, 45523, and 45549) and anti-CXCR4 (CTC-5, 12G5, 44708, 44717, and 44716) were obtained from R&D Systems. Clone 4G10, a murine IgG monoclonal that binds to the N-terminal region of CXCR4 was a gift from Dr. Chris Broder (33). The following Abs were used as primary Abs in the Western blot procedure: polyclonal rabbit IgG Abs to CD3 (Santa Cruz Biotechnology), CD4 (R&D Systems), or CXCR4 (Biochain); monoclonal mouse IgG Abs to CCR5 (clone CTC, N-terminal) and CXCR4 (clone 4G10 N-terminal).

Cell lines (80 × 106) were incubated for 30 min at 4°C with 10 ml of 100 mM (NH4)2SO4, 20 mM Tris-HC1 (pH 7.5) containing 10% glycerol, 1% Cymal-5 (Anatrace) and 1 tab minicomplete (Roche) to solubilize membrane receptors with minimal denaturation (34). Because different cell lines vary in their expression of the different receptors, we used lysates of SupT-1 and Jurkat cells to immunoprecipitate CD3 and CD4, Daudi B cells for CXCR4, which is mostly nonglycosylated, Daudi and U937 cells for CCR5. IgM/receptor complexes were formed by interacting 100 μl of cell lysate (containing the equivalent of 50 × 106 cells) with 100 μg of IgM for 6–12 h. The mixture of IgM/cell lysate was then interacted for 1–2 h with 50 μl of washed agarose beads containing covalently bound goat IgG anti-human IgM (Sigma-Aldrich). The agarose bead with bound IgM/receptor complexes was washed three times (700 rpm) with Tris buffer containing 0.1% fat-free milk protein and 0.01% Cymal-5 and two times with only Tris buffer (pH 6.8). The washed beads with IgM/receptor complexes were then interacted with Laemmili buffer containing 4% 2-ME and incubated at 57°C for 30 min to dissociate receptors under minimal reducing conditions so as to prevent alteration in receptor conformation especially since recognition of chemokine receptors by primary Abs is highly dependent on receptor conformation. Supernatants were then electrophoresed in SDS-PAGE and transferred onto nitrocellulose and then probed with primary IgG Abs specific for the receptor in question. It was not unusual for the secondary HRP-conjugated Ab to bind to extra protein bands of both the heavy and light chains of goat IgG (that were denatured and disassociated from the beads) as well as the light chains of IgM. To circumvent this problem, we used HRP-conjugated secondary Abs that were specific for nondenatured IgG (eBioscience). As negative controls, the Western blot procedure was performed with supernatant from beads that were interacted with IgM but with no cell membrane lysate or with beads and lysate but with no IgM so as to identify presence of nonspecific bands. As a positive control the membrane lysate without beads or IgM was interacted with Laemmli buffer under similar conditions and then electrophoresed in SDS-PAGE. Binding of IgM to CD4 and chemokines was also determined by using IgM (75 μg) to immunoprecipitate 1 μg of recombinant soluble CD4, CXCL12 (SDF-1α), or CCL5 (RANTES) using the same immunoprecipitation and Western blot techniques as for cell lysates. The recombinant CD4 protein is glycosylated and comprises the full length of the extracellular domain (aa 1–370).

B cells were obtained from human umbilical cord blood by initially isolating the leukocytes with histopaque and then subjecting the cells to the B cell negative isolation magnetic bead kit (Dynal Biotech). Purified B cells (95% pure) were dispensed at 5,000 cells per well (96-well plate) in RPMI 1640 containing 10% FCS. 50,000 irradiated (3000 rad) human PBL were added to each well as feeder cells. B cells were then infected with EBV containing supernatant from cell line B95–8 (American Type Culture Collection). After 4 wk in culture, culture supernatants were examined for IgM using an ELISA technique. Ninety percent of wells had IgM detected by the ELISA technique, with ∼18% of culture wells at concentrations of 300-1000 ng/ml (high IgM). Each of these high IgM containing supernatants were examined for IgM-ALA with a pool of leukocyte cell lines i.e., Jurkat, SupT-1, Daudi, and U937. Similarly, supernatants were examined for IgG and IgM anti-CD4 reactivity using an ELISA technique (see below). Of note, the wells at this stage did not have monoclonal IgM secreting B cell clones.

We used a previously described technique (35); 100 μl containing 250 ng of recombinant, full length extracellular domain of CD4 (Progenics) were added to a 96-well nunc-immulon plate and the ELISA plate was prepared using the Immuno-Tek ELISA construction system (Zepto Metrix). A total of 200–300 ng of IgM from serum, culture supernatants, or purified IgM was added to the well and incubated at room temperature for 1 h. The secondary Ab used was a 1/4000 dilution of HRP goat anti-human IgM or IgG (Southern Biotechnology Associates).

In these studies, we used Ficoll-Hypaque isolated PBMC (PBL). Briefly, 0.15 × 106 PBL in 0.15 ml RPMI 1640 containing 10% FCS were cocultured (in triplicate) with similar number of cells from another individual known to have different HLA-class 1 and DR Ags. Cells were cocultured in flat-bottom BD Falcon 96-well tissue culture plates. After 5–6 days in culture, [3H]thymidine was added to cells in each well, and cells were harvested 12–18 h later. Different doses (5–30 μg) of IgM were added to each well on day 0 and day 1 of the culture period. For day 0 cocultures, cells from one individual was initially incubated with IgM at 37°C for 30 min in 5% CO2 before adding cells from the second individual.

OKT3 (0.01–1.0 μg/ml) or PHA (5 μg/ml) plus IL-2 (40 u/ml) was added to Ficoll-Hypaque isolated PBL (1 × 106cells/ml) in RPMI 1640 and 10% FCS. Cells were cultured in triplicate in a flat-bottom 96-well plate (0.3 × 106 cells/well) for 3–4 days at 37°C in 5% CO2 and cell proliferation was quantitated as in the MLR assay. The effect of IgM was evaluated by initially pretreating cells with IgM at 37°C for 30 min in 5% CO2 before adding OKT3.

Cytokines in PBL culture supernatants were assayed either by an ELISA technique or in a semiquantitative manner using the Ray Bio human cytokine array no. 3 kit (Ray Biotech), which consists of a membrane array containing 42 different primary murine Abs, each specific for a cytokine. One milliliter of supernatant is incubated for 2 h with the membrane, which is then washed and re-incubated for 1 h with a mixture of 42 primary Abs. After re-washing, the membrane is incubated with an HRP-conjugated secondary Ab. Cytokine positive spots are detected on an x-ray film and quantitated with a densitometer. IgM-induced changes in cytokine levels, as detected by the Ray Bio cytokine array kit, were validated and quantitated using ELISA techniques. The IgM-induced changes in cytokine levels were similar whether supernatants were assayed on days 1, 2, or 3 of the MLR culture. Supernatant cytokine levels were maximal on day 5 of MLR. Additionally, there was no reduction in cytokine levels on day 5 supernatants even when quantitating for those cytokines that were produced in the first 24–48 h of culture.

Certain cytokines, e.g., IL-2 and IFN-γ, could not be detected by the Ray Biotech kit either because these secreted cytokines avidly bind to cell receptors (e.g., IL-2) or are secreted at very low levels in the MLR culture. To circumvent this problem we evaluated by two-color flow cytometry the percentage of activated T cells expressing intracellular cytokine. Briefly, PBL were either activated in an MLR or by immobilized anti-CD3 and cells were assayed for intracellular cytokine on day 4 of culture. Brefeldin (10 μg/ml) was added 12–14 h before terminating the culture. Murine mAbs, specific for the above cytokines, and effective for intracytoplasmic staining after cell fixation, was obtained from eBioscience. Maximum percentage of T cells with intracytoplasmic cytokines was observed at day 3 or day 4 of culture.

Studies on phosphorylation of Zap-70 were performed with freshly obtained PBL and phosphorylation was quantitated at 0, 2, 5 and 10 min (early stage) or at 16 h (late stage). In these studies, cells (0.6 × 106/0.6 ml) were initially incubated with or without IgM (final concentration,. 5–15 μg/ml) for 30–45 min at 37°C and were then activated with immobilized anti-CD3 (OKT3) (l μg of Ab in a well of a 48 well plate). Cells were then incubated for the required time at 37°C in RPMI 1640 medium with HEPES buffer and no FCS for the “early stage” experiments and in the same medium with 5% FCS and in 5% CO2 for the “late stage” experiments. Phosphorylation of Zap-70 was evaluated in the absence (to determine background phosphorylation) or presence of immobilized anti-CD3. PBL activated for the desired length of time were immediately chilled in ice for 10 min before fixing and permeabilization. Cells were then stained with Abs for the phosphorylated signaling protein or for the total signaling proteins and Ab binding to the signaling protein was quantitated by flow cytometry.

This assay was performed using the 24-well Costar transwell tissue culture inserts with 5 micron polycarbonate filters. 0.15 × 106 cells in 0.15 ml RPMI 1640 with 0.5% human albumin were added to the upper transwell. Thirty minutes later, 100 ng of SDF-1α or RANTES or MIP-1α were added to the bottom well containing 0.6 ml of the same medium as in the upper well. The chemotaxis assay was performed at 37°C for 2 h for activated PBL, 4 h for Jurkat cells and 12 h for Hut78 cells. Cells migrating to the bottom well were enumerated by flow cytometry. As a control for chemotaxis, four-fold chemokine was added to the upper transwell in the presence or absence of chemokine in the bottom well. The effect of IgM on chemotaxis was evaluated by incubating IgM (5–30 μg) with 0.15 × 106 cells at 37°C for 30 min before adding cells to the upper transwell.

Cells (0.5 × 106 in 0.5 ml) obtained from T cell lines (Hut 78 and Jurkat E-6) or monocytoid cell line (U937) or PBL activated for 3 days with PHA plus IL-2 were incubated with or without IgM (1–30 μg/L ×106 cells/ml) in PBS buffer containing CaCL2 at 37°C for 45 min, and without a wash step, cells were re-incubated at 37°C for 45 min with biotin-labeled cytokine (50 ng). Cells were then washed in the cold and stained with PE-streptavidin and chemokine binding was quantitated by flow cytometry.

Cells (0.2 × 106/0.2 ml) expressing CXCR4 were pretreated with medium (PBS containing CaCl2 and 2% FCS) or medium with 3 μg of normal IgM (pre-absorbed with mouse IgG) at 37°C for 30 min, not washed and then cells were interacted with CXCL12 (25–100 μg) at 37°C for 30 min. Cells were then washed at 4°C and incubated with FITC-labeled anti-CXCR4 (12G5, murine IgG mAb) at 4°C for 45 min.

The human investigation committee (HIC) at our institution has approved this study and informed consent was obtained from all subjects.

Before a human kidney transplant, sera from the recipient is routinely reacted with the potential donor’s lymphocytes to exclude the presence of IgG anti-HLA Abs, which can give rise to hyperacute or accelerated rejection, and to determine whether the recipient has IgM reactive to donor lymphocytes. Table Idepicts the incidence of biopsy proven acute rejections in kidney allograft recipients having different quantities of IgM-ALA reactive to their donor lymphocytes. None of these recipients were empirically treated for a rejection without a biopsy. Recipients with high levels of IgM anti-donor lymphocyte (IgM-ALA) reactivity had significantly less rejection episodes when compared with the other subset of patients. All recipients were on similar anti-rejection medications, i.e., prednisone, cyclosporin A or Prograf and Mycophenolate (Cell Cept). These differences in rejection rates among the three groups of recipients did not corelate with age, sex, race, level of HLA Ag mismatch and propensity to develop anti-HLA Abs pretransplant.

Table I.

Kidney rejection correlated with recipient’s serum IgM binding to donor T cells

No IgM-ALA (MCFa <20)Low IgM-ALA (MCF 21–200)High IgM-ALA (MCF >200)
No. of patients 65 22 21 
Percentage of acute rejection 32 32 9.5b 
Percentage of graft loss at 1 yr 20 9.1 0b 
No IgM-ALA (MCFa <20)Low IgM-ALA (MCF 21–200)High IgM-ALA (MCF >200)
No. of patients 65 22 21 
Percentage of acute rejection 32 32 9.5b 
Percentage of graft loss at 1 yr 20 9.1 0b 
a

MCF, Increase in mean channel fluorescence of anti-IgM staining after addition of serum.

b

p < 0.02 comparing high IgM-ALA to low and no IgM-ALA groups.

These data, together with previously published data, would indicate that IgM-ALA may have a role in inhibiting or attenuating the inflammatory response mediated by acute rejections (27, 28, 29, 30, 31). Presence of high IgM-ALA did not corelate with age, sex, race, renal disease causing ESRD or pretransplant exposure to CMV and EBV viruses. Similarly, there was no difference in total IgM serum levels among the three groups of recipients. These IgM anti-lymphocyte autoantibodies had reactivity to T, B, and NK cells, and also CD14 positive macrophages and CD15 positive neutrophils (data not shown). Prior studies also revealed reactivity of IgM to endothelial cells obtained from human kidneys and the myeloid K562 cell line lacking HLA class I and II receptors (36).

These in vivo observations of decreased rejections and the in vitro finding that IgM bound to T cells and other leukocytes and endothelial cells prompted us to examine whether IgM mediated these beneficial effects by inhibiting T cell activation and proliferation as well as by inhibiting the inflammatory process mediated by chemokines and chemokine receptors. Binding of IgM to chemokine receptors appeared to be an attractive possibility and could explain the widespread binding of IgM to the different leukocyte and endothelial cells.

Certain normal, ESRD or HIV sera have minimal or no IgM-ALA reactivity when evaluated by IgM binding to leukocytes (Table I) or to immobilized recombinant soluble CD4. In these studies, soluble CD4 was also used to evaluate IgM-ALA reactivity as IgM immuno precipitated CD4 from whole cell lysates as well as soluble recombinant CD4 (data presented in Fig. 3). We therefore wanted to determine whether purified IgM, isolated from such sera, had similarly low IgM-ALA reactivity. Surprisingly, purified IgM, obtained from sera with low IgM-ALA reactivity, had substantial more IgM binding to leukocytes and immobilized CD4, i.e., high IgM-ALA reactivity (Fig. 1, A and B). This avid binding to leukocytes and immobilized CD4 with purified IgM, was not observed with IgG purified from the same sera. Furthermore, purified IgM, when combined with autologous sera, maintained this enhanced IgM-ALA reactivity indicating that the observed lack of IgM-ALA reactivity with normal serum is not due to a serum factor that inhibits IgM binding (Fig. 1, A and B). However, absorbing purified IgM with leukocytes removed IgM-ALA reactivity (Fig. 1,C). Taken together, these observations would suggest that the apparent lack of IgM-ALA reactivity in certain sera is a result of IgM-ALA binding to leukocyte receptors, which are then shed in an IgM/receptor complex in the serum. The IgM purification process dissociates IgM from the receptor complex thus enhancing IgM-ALA reactivity. Hence, presence of high IgM-ALA reactivity in certain sera, as exemplified for lymphocyte binding in Table I and for CD4 binding in Fig. 1 B (see HIV no. 12, no. 50), can best be explained by increased IgM-ALA production, which tilts the balance in favor of more noncomplexed IgM-ALA in serum.

FIGURE 1.

IgM-ALA reactivity in purified IgM and autologous serum. A, Representative data from five different IgM preparations, showing significantly more IgM binding to T cells with purified IgM than with autologous serum. Binding of purified IgM is not decreased when combined with autologous serum. In these experiments, sera or purified IgM containing 6 μg of IgM was added to 100,000 PBL preactivated with PHA plus IL-2. B, IgM anti-CD4 reactivities in individual purified IgM and corresponding autologous sera from normal (N) and HIV-1 patients using an ELISA technique (see Materials and Methods). Sera containing 300 ng of IgM was compared with 300 ng of purified IgM. Note that there is significantly more IgM anti-CD4 reactivity in purified IgM. Also note that some HIV sera, e.g., nos. 12 and 50, have high IgM anti-CD4. Absorbing purified IgM with leukocyte cell lines (C) removes IgM-ALA reactivity. 300 μg of purified normal IgM was absorbed ×2, each time with a mixture of 250 × 106 cell lines (Jurkat, SupT-1, U937, Daudi and RAJI). Data are representative of four different IgM preparations.

FIGURE 1.

IgM-ALA reactivity in purified IgM and autologous serum. A, Representative data from five different IgM preparations, showing significantly more IgM binding to T cells with purified IgM than with autologous serum. Binding of purified IgM is not decreased when combined with autologous serum. In these experiments, sera or purified IgM containing 6 μg of IgM was added to 100,000 PBL preactivated with PHA plus IL-2. B, IgM anti-CD4 reactivities in individual purified IgM and corresponding autologous sera from normal (N) and HIV-1 patients using an ELISA technique (see Materials and Methods). Sera containing 300 ng of IgM was compared with 300 ng of purified IgM. Note that there is significantly more IgM anti-CD4 reactivity in purified IgM. Also note that some HIV sera, e.g., nos. 12 and 50, have high IgM anti-CD4. Absorbing purified IgM with leukocyte cell lines (C) removes IgM-ALA reactivity. 300 μg of purified normal IgM was absorbed ×2, each time with a mixture of 250 × 106 cell lines (Jurkat, SupT-1, U937, Daudi and RAJI). Data are representative of four different IgM preparations.

Close modal

Because IgM is a pentameric molecule and is encoded by nonmutated germline genes, it became necessary to exclude the possibility of nonspecific binding by IgM to carbohydrate moeties on leukocyte receptors. This possibility was examined by evaluating the binding of IgM from a panel of IgM secreting human B cell clones derived from human umbilical cord blood. Of the 79 supernatants containing IgM (concentration, 300-1100 ng/ml), only 8 had IgM secreting clones with IgM-ALA reactivity. Clones secreting IgM-ALA had the following binding specificities to cell lines: 2 clones with only anti-T cell (Jurkat, SupT-1) reactivity, 2 clones with only anti-monocyte (U937) reactivity and 4 clones with reactivity to all cell lines i.e., T,B and U937 (monocytoid) cells (Fig. 2,A). Two of these 8 wells also had IgM that bound to soluble CD4 as evaluated by an ELISA technique (Fig. 2 B). All 8 clones secreting IgM-ALA did not bind to IgG coated latex beads (data not shown). These findings i.e., presence of IgM-ALA reactivity in 10% of IgM secreting clones (and not the majority of IgM secreting clones) and binding of IgM-ALA to CD4 and specific cell lines (i.e., T cells or monocytes) would strongly argue against nonspecific IgM binding to leukocyte receptors and provide in addition evidence indicating that IgM-ALA consists of different Abs having different specificities e.g., IgM with specificity for CD4 and another Ab with specificity for a receptor on T or monocyte cells. Furthermore, binding of IgM to all cell lines i.e., T cells, Daudi B cells and U937 can best be explained either by polyspecificity of certain IgM autoantibodies or by binding of IgM-ALA to a leukocyte receptor that is common to all cell lines e.g., CD45.

FIGURE 2.

Presence of IgM-ALA in supernatants from umbilical cord B cells and in cord blood serum. A and B, IgM-ALA reactivity in IgM containing supernatants from B cell clones activated with the EBV virus. Supernatants contained between 300 and 1000 ng of IgM/ml. A, IgM-ALA reactivity that was detectable in 8 of 79 supernatants. Here, 75 μl supernatants were interacted with 100,000 cells containing a mixture of cell lines, i.e., Jurkat, SupT-1, Daudi, and U937. Note that the supernatants have IgM-ALA that is specific for either T or non-T cells or all four cell lines. Subsequent studies revealed that IgM anti-non-T cell had binding reactivity to only U937 cells and not to Daudi cells. B, Data on IgM anti-CD4 reactivity using an ELISA technique. 50 μl supernatant containing 300 ng of IgM/ml was added to each well (see Materials and Methods). Note that only 2 of 79 IgM containing supernatants had IgM anti-CD4 reactivity and both these supernatants also had IgM with binding to leucocytes. C, IgM-ALA in cord blood serum. Serum (125 μl) from cord blood was added to a mixture of cells lines as in A. Note that cord serum no. 7 had IgM-ALA that bound to all four cell lines.

FIGURE 2.

Presence of IgM-ALA in supernatants from umbilical cord B cells and in cord blood serum. A and B, IgM-ALA reactivity in IgM containing supernatants from B cell clones activated with the EBV virus. Supernatants contained between 300 and 1000 ng of IgM/ml. A, IgM-ALA reactivity that was detectable in 8 of 79 supernatants. Here, 75 μl supernatants were interacted with 100,000 cells containing a mixture of cell lines, i.e., Jurkat, SupT-1, Daudi, and U937. Note that the supernatants have IgM-ALA that is specific for either T or non-T cells or all four cell lines. Subsequent studies revealed that IgM anti-non-T cell had binding reactivity to only U937 cells and not to Daudi cells. B, Data on IgM anti-CD4 reactivity using an ELISA technique. 50 μl supernatant containing 300 ng of IgM/ml was added to each well (see Materials and Methods). Note that only 2 of 79 IgM containing supernatants had IgM anti-CD4 reactivity and both these supernatants also had IgM with binding to leucocytes. C, IgM-ALA in cord blood serum. Serum (125 μl) from cord blood was added to a mixture of cells lines as in A. Note that cord serum no. 7 had IgM-ALA that bound to all four cell lines.

Close modal

The production of IgM-ALA by umbilical cord B cells provided more evidence to show that IgM-ALA belong to the repertoire of naturally occurring IgM Abs. We therefore evaluated umbilical cord serum to determine the quantity of IgM using an ELISA technique (Zepto Metrix, NY) as well as to determine whether IgM-ALA was present in serum. Of the 7 sera analyzed, IgM was detectable in 6 sera at concentrations of 30–40 μg/ml while in one serum (cord blood no. 7) the IgM was present at 130 μg/ml. All cord sera had IgM-ALA (Fig. 2 C). Such data indicate that IgM-ALA is present in normal sera at birth.

Previous studies demonstrating the presence of naturally occurring IgG autoantibodies, with reactivity to intracellular proteins, prompted studies to examine whether IgG-ALA are also present in newborn sera, normal sera and disease sera (37). In these studies, presence of IgG-ALA was evaluated by binding of IgG from B cell clone supernatants or sera to cell lines. No IgG-ALA was detected in 96 umbilical cord B cell clone supernatants, in 12 cord sera and in 27 normal adult sera. However six of 135 ESRD sera had IgG-ALA that bound to either B cells or T cells. All six of these ESRD IgG-ALA were autoantibodies when tested against autologous leukocytes but none of these sera had IgG anti-HLA reactivity when tested with immobilized HLA Ags (class I and II) using the single HLA Ag flow beads (One Lambda). None of 19 HIV-1 sera had IgG-ALA. These studies clearly demonstrate that IgG-ALA are not present in newborn sera and in normal sera but occasionally present in ESRD sera. Further studies are needed to evaluate whether ESRD IgG-ALA are encoded by minimal or nonmutated germline genes, which are a characteristic feature of naturally occurring autoantibodies.

In the studies to be described below, we analyzed the effect of IgM-ALA on T cells and the chemokine receptor by using purified polyclonal IgM rather than serum and controlled for nonspecific IgM binding to leukocyte receptors by (i) using purified IgM obtained from different individual donors that included normals and patients with ESRD or HIV-1, and (ii) using IgM from a patient who was diagnosed with Waldenstrom’s macroglobulinaemia and had a monoclonal IgM with binding reactivity to an undefined leukocyte receptor. We could not identify any purified normal IgM as a negative control as all the purified normal IgM from >30 normals had IgM-ALA reactivity.

Based on our observations on allograft rejection, it became important to initially determine by immuno-precipitation assays whether IgM bound to receptors important in T cell activation as well as receptors involved in leukocyte chemotaxis. As seen in Fig. 3, purified normal, ESRD and HIV IgM immunoprecipitated CD3, CD4, CCR5 and CXCR4 receptors. However, significantly more of these receptors were immunoprecipitated by HIV and ESRD IgM from whole cell lysates. Of note, IgM from several different normal, HIV, and ESRD individuals failed to immunoprecipitate IL2-R, HLA-A, and HLA-DR receptors (data not shown). Interestingly, even though normal IgM immunoprecipitated low levels of CD4 from the cell lysates (see Fig. 3,A), one clearly observed that the same dose of normal IgM immunoprecipitated significantly more CD4 when using the soluble recombinant CD4 (extracellular domain). Differences in CD4 conformation and epitope availability between recombinant and native CD4 as previously suggested, could explain these observations (38). However, binding of IgM to membrane expressed CD4 was ascertained by using B cell clones (e.g., 4G4) secreting IgM with reactivity to recombinant CD4 (ELISA). As can be seen from Fig. 3,C, IgM from clone 4G4 had significant binding to CD4+ T cells when compared with CD4 T cells (MCF 71.5 vs 16.6). Similar data could not be obtained with polyclonal IgM as binding of the multiple different anti-receptor Abs present in the IgM preparations obscured the increased binding of IgM anti-CD4 to CD4+ T cells. Taken together, data from Figs. 1,B, 2,B, and 3, A and C, clearly demonstrate that IgM binds to membrane expressed CD4 in a specific manner. Normal or ESRD IgM failed to immunoprecipitate chemokines i.e., recombinant CCL5 and CXCL12 (data not shown).

FIGURE 3.

Immunoprecipitation experiments to show binding of IgM to CD3, CD4, CCR5, and CXCR4. Identical quantities of individual (labeled no. 1, 2, etc.) or pooled (labeled P) IgM from normal (labeled N), HIV (labeled H), ESRD (labeled E), or Waldenstrom (labeled W) were used to immunoprecipitate leukocyte receptors from equal amounts of whole cell lysates or recombinant soluble CD4. Jurkat and SupT-1 cell lysates were used for CD3/CD4, Daudi B cell lysates for CXCR4, and Daudi/U937 lysates for CCR5. As controls, Western blots were performed with cell lysates in the absence of agarose beads (to control for binding of primary Ab to leukocyte receptor and to determine receptor size (labeled Ly). In another control, agarose beads without IgM were added to lysate to determine whether the leukocyte receptor nonspecifically bound to the bead (B plus Ly). Note that several fold more receptors were immunoprecipitated by ESRD and HIV IgM when compared with normal IgM (B). Also note that CXCR4 in Daudi is in a nonglycosylated form. IgM preferentially immunoprecipitated the monomeric form of CXCR4 or CCR5 receptor (as depicted in these figures) but certain normal and patient IgM also immunoprecipitated lower levels of the dimeric or other low m.w. isoforms of these receptors as previously described (3459 ) (data not shown). The primary polyclonal Ab detected all these receptor isoforms in the lysates we used. In A, recombinant, glycosylated, solubilized CD4 (Sol-CD4), comprising the full length extracellular domain was immunoprecipitated by IgM. For controls, Western blots were performed with soluble CD4 in absence of IgM and agarose beads (labeled CD4) and with Sol-CD4 added to agarose beads in absence of IgM (labeled CD4+ B). C, Significantly increased binding of IgM from B cell clone 4G4 to CD4+ T cells when compared with CD4 T cells (MCF 71.6 vs 16.6)). Clone 4G4 secreted IgM with anti-CD4 reactivity detected by ELISA. Note that no increased binding was observed on CD4+ T cells using a B cell clone (IE12) secreting IgM without anti-CD4 reactivity (MCF 22.5 vs 22.3). Both clones have IgM anti-leukocyte binding. In these experiments 200,000 PBL were initially incubated (4°C) with identical quantities of IgM (1 μg), washed and then simultaneously stained with PE anti-CD3, FITC anti-CD4 and Cy5 anti-human IgM. These data are representative data obtained from two clones with anti-CD4 reactivity and 8 clones without anti-CD4 reactivity.

FIGURE 3.

Immunoprecipitation experiments to show binding of IgM to CD3, CD4, CCR5, and CXCR4. Identical quantities of individual (labeled no. 1, 2, etc.) or pooled (labeled P) IgM from normal (labeled N), HIV (labeled H), ESRD (labeled E), or Waldenstrom (labeled W) were used to immunoprecipitate leukocyte receptors from equal amounts of whole cell lysates or recombinant soluble CD4. Jurkat and SupT-1 cell lysates were used for CD3/CD4, Daudi B cell lysates for CXCR4, and Daudi/U937 lysates for CCR5. As controls, Western blots were performed with cell lysates in the absence of agarose beads (to control for binding of primary Ab to leukocyte receptor and to determine receptor size (labeled Ly). In another control, agarose beads without IgM were added to lysate to determine whether the leukocyte receptor nonspecifically bound to the bead (B plus Ly). Note that several fold more receptors were immunoprecipitated by ESRD and HIV IgM when compared with normal IgM (B). Also note that CXCR4 in Daudi is in a nonglycosylated form. IgM preferentially immunoprecipitated the monomeric form of CXCR4 or CCR5 receptor (as depicted in these figures) but certain normal and patient IgM also immunoprecipitated lower levels of the dimeric or other low m.w. isoforms of these receptors as previously described (3459 ) (data not shown). The primary polyclonal Ab detected all these receptor isoforms in the lysates we used. In A, recombinant, glycosylated, solubilized CD4 (Sol-CD4), comprising the full length extracellular domain was immunoprecipitated by IgM. For controls, Western blots were performed with soluble CD4 in absence of IgM and agarose beads (labeled CD4) and with Sol-CD4 added to agarose beads in absence of IgM (labeled CD4+ B). C, Significantly increased binding of IgM from B cell clone 4G4 to CD4+ T cells when compared with CD4 T cells (MCF 71.6 vs 16.6)). Clone 4G4 secreted IgM with anti-CD4 reactivity detected by ELISA. Note that no increased binding was observed on CD4+ T cells using a B cell clone (IE12) secreting IgM without anti-CD4 reactivity (MCF 22.5 vs 22.3). Both clones have IgM anti-leukocyte binding. In these experiments 200,000 PBL were initially incubated (4°C) with identical quantities of IgM (1 μg), washed and then simultaneously stained with PE anti-CD3, FITC anti-CD4 and Cy5 anti-human IgM. These data are representative data obtained from two clones with anti-CD4 reactivity and 8 clones without anti-CD4 reactivity.

Close modal

Because IgM immunoprecipitated CD3 and CD4, it became important to determine whether purified IgM inhibited T cell activation, proliferation and cytokine production. Data in Fig. 4 indicate that purified IgM, especially ESRD IgM, inhibits T cell proliferation in response to alloantigens and anti-CD3 only when IgM was added to the cultures within the first 24 h of initiating the cultures. Certain individual normal IgM (e.g., N no. 1), unlike the other 5 normals, significantly inhibited the proliferative response in the allogeneic MLR (see Fig. 4,B) indicating that the repertoire of IgM-ALA can vary in their functional activity, even among normals. However, in general, ESRD individual IgM was more inhibitory when compared with HIV IgM ALA. The pentameric form of IgM is not essential for the inhibitory effect on MLR as monomeric IgM was as effective (Fig. 4,B). This inhibitory effect of ESRD IgM on proliferation was not associated with increased cell death or apoptosis as determined by flow cytometry using propidium and anti-annexin reagents. The inhibitory effect of IgM-ALA on cell proliferation was not mediated by any contaminating IgG in the IgM preparation as pooled normal, ESRD, or HIV IgG failed to inhibit proliferation. The inhibitory effect of IgM-ALA on MLR-induced proliferation was most likely mediated by binding of IgM to CD3 (OKT3) and/or CD4 as additional experiments with IgG murine monoclonals clearly showed that Abs against CD3 or CD4 but not CXCR4 (12G5) and CCR5 (2D7) inhibited MLR-induced T cell proliferation (see Fig. 4 B). Prior studies have shown that anti-CD3, even though mitogenic to T lymphocytes, will inhibit proliferation of T cells in an MLR (39).

FIGURE 4.

Effect of purified IgM on T cell proliferation - Effect of pooled or individual normal, ESRD and HIV IgM on T cell proliferation induced by anti-CD3 (A) and MLR (B). 15 μg of IgM (including monomeric IgM) were used for 1 × 106 cells/ml in these experiments and added at the time cells were cocultured. Individual (labeled no. 1, 6, etc.) IgM (4.5 μg/0.3 ml), pooled IgG (2 μg/0.3 ml), and murine mAbs (2 μg/0.3 ml) to CD3 (OKT3), CXCR4 (12G5), and CCR5 (2D7) were used at the initiation of the MLR. Values for T cell proliferation assays are mean ± SD of experiments performed in triplicate.

FIGURE 4.

Effect of purified IgM on T cell proliferation - Effect of pooled or individual normal, ESRD and HIV IgM on T cell proliferation induced by anti-CD3 (A) and MLR (B). 15 μg of IgM (including monomeric IgM) were used for 1 × 106 cells/ml in these experiments and added at the time cells were cocultured. Individual (labeled no. 1, 6, etc.) IgM (4.5 μg/0.3 ml), pooled IgG (2 μg/0.3 ml), and murine mAbs (2 μg/0.3 ml) to CD3 (OKT3), CXCR4 (12G5), and CCR5 (2D7) were used at the initiation of the MLR. Values for T cell proliferation assays are mean ± SD of experiments performed in triplicate.

Close modal

Further studies were performed on supernatants from MLR cultures to determine whether the anti-proliferative effects of IgM-ALA were associated with a decrease in cytokine production. The Array III kit detected a significant increase in the basal secretion of IL-6, IL-8, IL-13, TNF-α, GM-CSF, MCP-1, MIG, MDC, TARC, and GRO when PBL were activated in an MLR assay. However, the addition of both normal and patient IgM at the initiation of the MLR culture, had no inhibitory effect on MLR-induced production of IL-6, IL-8, MCP-1, MIG, GRO, GM-CSF, IL-1β and TGF-β but significantly decreased the quantity of certain other cytokines e.g., TNF-α, IL-13, MDC and TARC in the MLR supernatants (see Fig. 5,A). This decrease in cytokine level (e.g., TNF-α) in the presence of IgM was not a result of IgM neutralizing the secreted TNF-α, as the addition of IgM directly to the control MLR supernatants, obtained on day 5, did not decrease the level of TNF-α detected by two assay systems (i.e., Array III and an ELISA technique), thus indicating that IgM specifically inhibits the secretion or production of certain cytokines. Because the Array III kit could not detect IL-2 and IFN-γ, in MLR supernatants, we resorted to determine the effect of IgM on intracellular expression of these cytokines. Both normal and ESRD IgM caused a major decrease in the percentage of cells expressing intracytoplasmic IL-2, but had no effect on IFN-γ expressing cells (see Fig. 5 B). Under the culture conditions, there was no detectable IL-17 or IL-10 in the supernatants or cytoplasm. Nonetheless, IgM did not increase production of these two cytokines. The combined data, with the different cytokine assay techniques, clearly demonstrate that normal and patient IgM predominantly inhibits the production of certain cytokines involved in T cell proliferation (e.g., IL-2) and certain proinflammatory cytokines and chemokines (e.g., TNF-α, IL-13, MDC and TARC). Conversely, IgM did not significantly inhibit other proinflammatory cytokines (e.g., IFN-γ, IL-6, GM-CSF, IL-1β and IL-8).

FIGURE 5.

Effect of purified IgM on cytokine production and phosphorylation of Zap-70. A, Pooled normal, ESRD, and HIV IgM but not Waldenstrom IgM significantly inhibit the increase in TNF- α and IL-13 but not that of IL-6, IL-8, MIG, and MCP-1 produced in response to alloantigen activation of T cells. Supernatants were obtained from day 5 MLR cultures stimulated in the presence or absence of pooled IgM (15 μg/ml), added on day 0. Dot plots (B) depicting the effect of IgM on intracytoplasmic IL-2 and IFN-γ in PBL activated for 48 h in an MLR. 15 μg of IgM were used for 1 × 106 cell/ml in these experiments. These are representative data of six experiments with different IgM. C, Flow cytometry histograms of intracellular phos-Zap-70 present in PBL. Note that background phos-Zap 70 (shaded gray) increases after incubating cells with immobilized anti-CD3 for 16 h (MCF increased from 29 to 54). PBL were preincubated with individual IgM at 37°C for 30 min before the overnight incubation with anti-CD3. Note that ESRD and normal IgM inhibited anti-CD3 activated phos-Zap-70 in ∼40–50% of cells, i.e., phos-Zap-70 could not be inhibited in all cells. These are representative data of five experiments with different IgM. D, The effect of individual IgM on anti-CD3 activated (overnight) phos-Zap-70. Note that normal, ESRD, and HIV, but not Waldenstrom IgM inhibited the increase in phos-Zap-70 without inhibiting total Zap-70. In this experiment, anti-CD3 increased MCF of background phos-Zap-70 from 146 to 276 (MCF of negative control was <10). C and D, MCF represents the mean of phos-Zap-70 present in both cell populations, i.e., cells where phos-Zap-70 was inhibited as well as noninhibited. MCF values are obtained from 10,000 gated events.

FIGURE 5.

Effect of purified IgM on cytokine production and phosphorylation of Zap-70. A, Pooled normal, ESRD, and HIV IgM but not Waldenstrom IgM significantly inhibit the increase in TNF- α and IL-13 but not that of IL-6, IL-8, MIG, and MCP-1 produced in response to alloantigen activation of T cells. Supernatants were obtained from day 5 MLR cultures stimulated in the presence or absence of pooled IgM (15 μg/ml), added on day 0. Dot plots (B) depicting the effect of IgM on intracytoplasmic IL-2 and IFN-γ in PBL activated for 48 h in an MLR. 15 μg of IgM were used for 1 × 106 cell/ml in these experiments. These are representative data of six experiments with different IgM. C, Flow cytometry histograms of intracellular phos-Zap-70 present in PBL. Note that background phos-Zap 70 (shaded gray) increases after incubating cells with immobilized anti-CD3 for 16 h (MCF increased from 29 to 54). PBL were preincubated with individual IgM at 37°C for 30 min before the overnight incubation with anti-CD3. Note that ESRD and normal IgM inhibited anti-CD3 activated phos-Zap-70 in ∼40–50% of cells, i.e., phos-Zap-70 could not be inhibited in all cells. These are representative data of five experiments with different IgM. D, The effect of individual IgM on anti-CD3 activated (overnight) phos-Zap-70. Note that normal, ESRD, and HIV, but not Waldenstrom IgM inhibited the increase in phos-Zap-70 without inhibiting total Zap-70. In this experiment, anti-CD3 increased MCF of background phos-Zap-70 from 146 to 276 (MCF of negative control was <10). C and D, MCF represents the mean of phos-Zap-70 present in both cell populations, i.e., cells where phos-Zap-70 was inhibited as well as noninhibited. MCF values are obtained from 10,000 gated events.

Close modal

Because prior studies have shown that TCR/CD3 receptor activation leads to increased intracellular phosphorylation of Zap-70, we evaluated whether IgM, after binding to CD3 and/or CD4, will inhibit anti-CD3 mediated phosphorylation of Zap-70. Phosphorylated Zap-70 and total Zap-70 were quantitated by flow cytometry at 0, 2, and 5 min and at 16 h after immobilized anti-CD3 activation of freshly isolated human PBL. IgM had a mild inhibitory effect on anti-CD3-induced phosphorylation of Zap-70 at 2 min. However, the majority of individual normal, ESRD and HIV IgM (but not Waldenstrom) inhibited both background phosphorylation of Zap-70 and the increase in Zap-70 phosphorylation induced by anti-CD3 at 16 h in 40–50% of T cells (see Fig. 5,C). Importantly, IgM mediated inhibition of phos-Zap-70 was not associated with inhibition of total Zap-70 indicating that IgM inhibits proximal intracellular signaling (see Fig. 5 D). Pooled normal IgG had no effect on Zap-70 phosphorylation (data not shown).

The inhibitory effect of purified normal and ESRD IgM on T cell proliferation and TNF-α production in response to alloantigen stimulation (MLR) prompted us to evaluate whether the inhibitory effects were also mediated by IgM down-regulation of receptors involved in T cell activation. In these studies, cell receptor expression was quantitated by flow cytometry using PBL activated either in an MLR or with PHA and IL-2. IgM was added to the cultures either at the initiation of the 4–5 day cultures or after T cells were activated. As exemplified in Fig. 6,A, both normal and ESRD IgM, but not Waldenstrom IgM, when incubated for 4 h at 37°C with 5 day activated T lymphocytes, down-regulated CD2 and CD4 on T lymphocytes. Similar data were also obtained with HIV IgM (data not shown). Importantly, only ESRD and HIV IgM, but not normal IgM down-regulated CD86 (B7-2) on CD14 macrophages activated in an MLR for 24 h (Fig. 6,C). However, normal and ESRD IgM did not change receptor expression (including up-regulation) of CD3, CD8, intracellular CD152, CD154, PD-1, CD28, CD40L, HLA-AB and HLA-DR on lymphocytes as well as CD80 (B7-1), CD40, and ICAM-1 on macrophages (data shown for CD8 and CD28 in (Fig. 6 A). Pooled normal IgG failed to downmodulate any of these receptors (data not shown).

FIGURE 6.

Effect of purified IgM on costimulatory molecules and CD4+, CD25+, Foxp3+, regulatory T cells. A, Individual normal and ESRD IgM, but not murine IgG anti-CD3 (OKT3) inhibits CD2 and CD4 expression on PBL activated in MLR for 4 days. Data represent percent change in MCF in presence of Abs when compared with MCF in medium control. Day 4-activated cells were incubated for 16 h at 37°C in medium (control) or with ESRD or normal IgM, 10 μg/ml (obtained from a single individual) or soluble OKT3 (1 μg/ml). Note that IgM had a marked inhibitory effect on CD4 and CD2 receptors but not CD8 and CD28. Data are from eight separate experiments with different IgM and values represent mean ± SD. B, IgM inhibits both surface and intracytoplasmic CD4. Cells were initially labeled with PE anti-CD4 to quantitate surface CD4 and then permeabilized and relabeled with PE anti-CD4 to quantitate both surface and intracytoplasmic CD4. Normal IgM was added at initiation of MLR assay. Data in this figure is a representative example from one of five individuals. C, The effect of individual IgM on membrane CD86 expressed by CD14 positive macrophages present in PBL activated in a 24 h MLR assay. Note that HIV and ESRD IgM, but not normal IgM, inhibit CD86. D, The effect of IgM on CD4+ CD25(highly +) Foxp3+ regulatory T cells in MLR-activated (day 5) PBL (upper panels). Pooled ESRD IgM (15 μg/ml) was added at initiation of MLR. There was no significant change in the percentage of this subpopulation or in intracytoplasmic Foxp3 in the presence of IgM. This is a representative example of three separate experiments with three different individual IgM. In all panels, MCF values are obtained from 10,000 gated events.

FIGURE 6.

Effect of purified IgM on costimulatory molecules and CD4+, CD25+, Foxp3+, regulatory T cells. A, Individual normal and ESRD IgM, but not murine IgG anti-CD3 (OKT3) inhibits CD2 and CD4 expression on PBL activated in MLR for 4 days. Data represent percent change in MCF in presence of Abs when compared with MCF in medium control. Day 4-activated cells were incubated for 16 h at 37°C in medium (control) or with ESRD or normal IgM, 10 μg/ml (obtained from a single individual) or soluble OKT3 (1 μg/ml). Note that IgM had a marked inhibitory effect on CD4 and CD2 receptors but not CD8 and CD28. Data are from eight separate experiments with different IgM and values represent mean ± SD. B, IgM inhibits both surface and intracytoplasmic CD4. Cells were initially labeled with PE anti-CD4 to quantitate surface CD4 and then permeabilized and relabeled with PE anti-CD4 to quantitate both surface and intracytoplasmic CD4. Normal IgM was added at initiation of MLR assay. Data in this figure is a representative example from one of five individuals. C, The effect of individual IgM on membrane CD86 expressed by CD14 positive macrophages present in PBL activated in a 24 h MLR assay. Note that HIV and ESRD IgM, but not normal IgM, inhibit CD86. D, The effect of IgM on CD4+ CD25(highly +) Foxp3+ regulatory T cells in MLR-activated (day 5) PBL (upper panels). Pooled ESRD IgM (15 μg/ml) was added at initiation of MLR. There was no significant change in the percentage of this subpopulation or in intracytoplasmic Foxp3 in the presence of IgM. This is a representative example of three separate experiments with three different individual IgM. In all panels, MCF values are obtained from 10,000 gated events.

Close modal

Receptor down-regulation of CD2 and CD4 did not appear to be a result of receptor internalization as down-regulation of surface receptor expression was also accompanied by intracellular down-regulation of CD4 (see Fig. 6 B). No down-regulation of CD2 and CD4 was observed by incubating MLR activated cells with IgM for 4–6 h at 4°C, indicating that the observed down-regulation of CD4 at 37°C was an active process and not due to IgM inhibiting the binding of the murine monoclonal used to detect CD4. Additionally, down-regulation of CD2 and CD4 occurred in the presence of monomeric IgM but the extent of down-regulation was ∼50% when compared with pentameric IgM (data not shown).

Finally, we wanted to determine whether the inhibitory effect of IgM on T cell proliferation and cytokine production was secondary to IgM mediated increase in naturally arising regulatory T cells (i.e., CD4+, Foxp3+ cells expressing high levels of CD25). Representative Data in Fig. 6 D show that ESRD IgM, when added at the initiation of an MLR culture, did not increase the CD4+ regulator T cell subpopulation expressing high levels of CD25 and had no effect on total IL-2R levels and intracytoplasmic Foxp3 levels even though IgM significantly decreased CD4 receptors on T cells (MCF of CD4 changed from 429 to 79).

Thus far, the data indicate that purified polyclonal IgM from normals, HIV and ESRD patient’s immunoprecipitate CD3 and CD4, down modulate CD4 and CD2 and inhibit T cell activation (Zap-70) and proliferation. These inhibitory effects were more pronounced when using ESRD or HIV IgM. Furthermore ESRD IgM down modulated CD86, a costimulatory molecule, present on macrophages. Of particular interest, even though there was some variability in the functional effect of individual IgM e.g., on receptor down modulation or MLR, the different IgM preparations had similar functional effects i.e., predictably down modulating certain receptors e.g., CD4 and CD2 but not other receptors e.g., IL2-R, HLA, CD8, CD3 and CD28. Similarly, the different IgM preparations had a similar effect in inhibiting only certain cytokines, particularly TNF-α and IL-2, and not other proinflammatory cytokines.

Because IgM immunoprecipitated CXCR4 and CCR5 from cells (Fig. 3), it became important to determine whether purified IgM inhibited binding of chemokine to these receptors as well as inhibited chemotaxis. Data in Fig. 7 A clearly demonstrate that both normal and ESRD IgM inhibited to a similar degree binding of biotin-labeled CCL3 (MIP-Iα) to CCR5 and CXCL12 to CXCR4 present on a cell line (U937) and on PBL activated for 3 days with PHA and IL-2. IgM inhibited chemokine binding in a dose dependent manner as exemplified for binding of CCL3 to U937 cells and CXCL12 to activated PBL. Incubating cells with IgM and/or chemokine at 37°C or 4°C did not change the magnitude of the inhibitory effect of IgM on chemokine binding thus indicating that the IgM mediated inhibitory effect was not due to IgM-induced internalization of the receptor at 37°C. Furthermore, the lack of chemokine binding in the presence of IgM was not due to IgM forming a complex with the chemokine as we failed to detect IgM binding to the chemokine using immunoprecipitation techniques and an ELISA technique with immobilized chemokine. Waldenstrom IgM and pooled human IgG had no inhibitory effect on chemokine binding.

FIGURE 7.

Effect of purified IgM on chemokine binding, chemotaxis, and CXCL12-induced CXCR4 down-regulation. A, Normal and ESRD IgM have a similar inhibitory effect on binding of biotin-labeled CXCL12 and CCL3 to cells. Cells (0.5 × 106 in 0.5 ml) obtained from monocytoid cell line (U937) or PBL activated for 3 days with PHA plus IL-2 were incubated with or without IgM (1–30 μg/L ×106 cells/ml) in PBS buffer containing CaCL2 at 37°C for 45 min, and without a wash step, cells were re-incubated at 37°C for 45 min with biotin-labeled cytokine (50 ng). Cells were then washed in the cold and stained with PE-streptavidin. These are representative examples from seven different experiments. In these experiments, MCF of CXCL12 binding to activated PBL was 193, whereas MCF of CCL3 binding to U937 was 600. MCF for negative controls was <10. ESRD IgM is more inhibitory to chemotaxis and chemokinesis when compared with normal IgM (B). Hut 78 or Jurkat cells were incubated with or without IgM (obtained from different individuals) at 37°C for 30 min before adding the cells, without a wash step, to the upper chamber of the chemotaxis Transwells. CXCL12 was added to the bottom wells. Data labeled media indicate baseline chemokinesis in the absence of CXCL12. Addition of CXCL12 to both the upper and lower wells completely inhibited chemotaxis (data not shown). These are representative examples from five separate experiments. Purified normal IgM (C) inhibits CXCL12-induced down-regulation of CXCR4. Details of assay are in Materials and Methods. 0.2 × 106 Jurkat and RAJI cells were either pretreated with CXCL12 (50 μg) or initially with purified normal IgM (3 μg) followed 30 min later with CXCL12 and then incubated at 37°C for 30 min before quantitating membrane CXCR4. In all panels, flow cytometry data were obtained from at least 10,000 events.

FIGURE 7.

Effect of purified IgM on chemokine binding, chemotaxis, and CXCL12-induced CXCR4 down-regulation. A, Normal and ESRD IgM have a similar inhibitory effect on binding of biotin-labeled CXCL12 and CCL3 to cells. Cells (0.5 × 106 in 0.5 ml) obtained from monocytoid cell line (U937) or PBL activated for 3 days with PHA plus IL-2 were incubated with or without IgM (1–30 μg/L ×106 cells/ml) in PBS buffer containing CaCL2 at 37°C for 45 min, and without a wash step, cells were re-incubated at 37°C for 45 min with biotin-labeled cytokine (50 ng). Cells were then washed in the cold and stained with PE-streptavidin. These are representative examples from seven different experiments. In these experiments, MCF of CXCL12 binding to activated PBL was 193, whereas MCF of CCL3 binding to U937 was 600. MCF for negative controls was <10. ESRD IgM is more inhibitory to chemotaxis and chemokinesis when compared with normal IgM (B). Hut 78 or Jurkat cells were incubated with or without IgM (obtained from different individuals) at 37°C for 30 min before adding the cells, without a wash step, to the upper chamber of the chemotaxis Transwells. CXCL12 was added to the bottom wells. Data labeled media indicate baseline chemokinesis in the absence of CXCL12. Addition of CXCL12 to both the upper and lower wells completely inhibited chemotaxis (data not shown). These are representative examples from five separate experiments. Purified normal IgM (C) inhibits CXCL12-induced down-regulation of CXCR4. Details of assay are in Materials and Methods. 0.2 × 106 Jurkat and RAJI cells were either pretreated with CXCL12 (50 μg) or initially with purified normal IgM (3 μg) followed 30 min later with CXCL12 and then incubated at 37°C for 30 min before quantitating membrane CXCR4. In all panels, flow cytometry data were obtained from at least 10,000 events.

Close modal

IgM in addition inhibited chemotaxis of activated PBL and T cell lines in response to CXCL12. However ESRD IgM had a significantly more pronounced inhibitory effect on chemotaxis as depicted in Fig. 7,B, even though both normal and ESRD IgM had a similar inhibitory effect on the binding of CXCL12 to the CXCR4 receptor. These differences in inhibitory effects on chemotaxis with the T cell lines were not due to increased apoptosis or cell death as evaluated by flow cytometry using propridium and anti-annexin and would suggest that ESRD IgM in addition inhibits chemotaxis through effects on other cell receptors (e.g., adhesion molecules or integrins) and/or intracellular activation pathways that are involved in both chemokinesis and chemotaxis activity. Data in Fig. 7 B show that both normal and ESRD IgM also has an inhibitory effect on chemokinesis of cells.

Because IgM inhibited binding of chemokine to their receptor, it became important to determine whether purified IgM also inhibited chemokine-induced receptor down-regulation as has previously been described (40). In these studies, cell lines expressing high levels of CXCR4 (Jurkat T cell, RAJI B cell) were used. Data in Fig. 7 C clearly show that CXCL12 markedly down-regulates CXCR4 expression on the cell membrane secondary to receptor internalization. However, pretreatment of cells with IgM inhibits the CXCL12-induced down-regulation of CXCR4. Waldenstrom IgM had no inhibitory effect on CXCL12-induced down-regulation of CXCR4 (data not shown). Similar studies could not be done with cell lines expressing the CCR5 receptor as IgM inhibited binding of all the murine monoclonal anti-CCR5 Abs that were needed to detect membrane expression of CCR5.

These studies indicate that purified IgM binds to chemokine receptors and through this mechanism inhibits chemokine binding, chemokine-induced chemotaxis, and chemokine-induced receptor down-regulation. Finally, we failed to demonstrate binding of IgM to chemokines using both immunoprecipitation and ELISA techniques.

The current studies clearly show that IgM-ALA are naturally occurring as these Abs are present in all umbilical cord serum (Fig. 2,C) and umbilical cord B cell clones secrete IgM-ALA. Furthermore the evidence indicates that IgM-ALA binding to leukocyte receptors has specificity. First, several different individual purified IgM, whether from normal or diseased sera, bound to the same receptors, inhibited production of certain specific cytokines and down-regulated only certain cell receptors. Secondly, the data clearly show that IgM from certain individuals binds to CXCR4 but not CCR5 while IgM from other individuals bind to CCR5 and not CXCR4 (Fig. 3,B). These latter findings strongly argue against nonspecific binding of IgM to membrane receptors. One would have expected purified IgM from different individuals as well as monoclonal Waldenstrom IgM to randomly bind to different receptors and to randomly down-regulate different receptors if there was nonspecific binding of polyreactive IgM to receptors. However, specificity of IgM binding is best exemplified by 1) analysis of IgM-ALA from B cell clone supernatants with 10% of IgM secreting clones having specificity for leukocytes, and with some IgM binding to either monocytes or T cells (Fig. 2,A); 2) studies with purified IgM from individual HIV-1 sera clearly showing that IgM from certain individuals will only bind to CXCR4 and not CCR5 while IgM from other HIV individuals will bind to CCR5 and not CXCR4 (Fig. 3); and 3) lack of IgM-ALA binding to immobilized IgG.

The occasional presence of IgG-ALA in disease sera e.g., ESRD, but not in normal sera, and normal umbilical cord sera would argue against these Abs as being naturally occurring. IgG-ALA, unlike IgM ALA, are cytolytic in the presence of complement (at 37°C) or FcIgG receptor bearing cells (ADCC) (41). Hence there is a possibility that such autoantibodies are pathogenic (e.g., cause lymphopenia) and under normal conditions inhibitory mechanisms could prevent B cells from secreting IgG autoantibodies. Finally, it is unclear whether IgG-ALA are encoded by nonmutated germline genes.

There has been very little progress in understanding the biological role of naturally occurring IgM anti-leukocyte autoantibodies (IgM-ALA) even though these autoantibodies were first identified ∼35 years ago (13). A major difficulty has been the lack of information regarding the identity of the different leukocyte receptors which bind to these IgM-ALA. However clues on the potential identity of some of these receptors became apparent when we attempted to explain certain clinical observations in human kidney and heart transplants. Allograft recipients with high levels of IgM-ALA were found to have significantly better allograft survival and fewer rejections even during the precyclosporin era (27, 28, 29, 30, 31). Recognizing that rejection in an allograft is a T cell mediated inflammatory process, we hypothesized that IgM-ALA mediated these inhibitory effects by binding to leukocyte receptors involved in T cell activation and leukocyte chemotaxis. We felt that such an hypothesis could also explain the increase in IgM-ALA in various inflammatory and infective states, perhaps to attenuate the inflammatory response (13, 14, 15, 16, 17, 18, 19).

Data in the present investigations show that IgM-ALA immunoprecipitates CD3, CD4, CXCR4 and CCR5 but not IL-2R and HLA. No attempt has been made in the current studies to elucidate the identity of the entire repertoire of leukocyte receptors that bind to IgM-ALA, which have been known to be heterogenous and to have binding reactivity to receptors of different molecular sizes, including glycoproteins, phospholipids and glycolipids (20, 21, 22, 23, 24, 25, 26). However, our data demonstrate that IgM-ALA 1) will inhibit T cell activation and proliferation in response to alloantigens and the mitogenic effect of anti-CD3; 2) down-regulates costimulatory molecules, e.g., CD2 and CD4 on T cells and CD86 on macrophages; and in addition 3) IgM-ALA inhibits chemotaxis induced by CXCL12 or CCL3. These observations support the concept that IgM-ALA could provide an innate mechanism to regulate a T cell mediated inflammatory response.

Data in Fig. 1, A and B, clearly demonstrate that purified IgM has substantially more IgM-ALA reactivity especially when compared with nonpurified IgM present in autologous sera. A potential explanation for this observation is that most of the IgM-ALA in certain sera, especially normal sera, exists in an IgM/receptor complex which forms when the IgM binds to the receptor on the cell membrane. The purification process dissociates the IgM-ALA from the receptor complex. Support for such a concept comes from two observations 1) purified IgM obtained from normal sera or from sera with an apparent lack of IgM-ALA reactivity has high binding of IgM to leukocytes and this enhancement in IgM-ALA reactivity is not inhibited even when purified IgM is combined with autologous sera, and 2) IgM-ALA reactivity in purified IgM is removed when purified IgM is preabsorbed with leukocytes. Hence, the observed enhanced IgM-ALA reactivity of certain sera obtained from ESRD or HIV patients (see Table I and Fig. 1 B) can best be explained on the basis of increased production of IgM with anti-leukocyte reactivity thus resulting in an increase in the level of uncomplexed IgM-ALA in such sera.

Interestingly, the data on T cell activation and proliferation indicate that purified IgM from normal, HIV and ESRD sera have similar effects on 1) CD2 and CD4 down-regulation; 2) inhibition of anti-CD3-induced phosphorylation of Zap-70; and 3) inhibition in production of IL-2 and certain proinflammatory cytokines (e.g., TNF-α) in response to alloantigen (MLR) stimulation. These observations raise two basic questions. Firstly, why does normal purified IgM down-regulate CD2 and CD4, inhibit Zap-70 phosphorylation and inhibit cytokine production? Secondly, how can one explain the pronounced inhibitory effects of ESRD and HIV IgM on MLR and anti-CD3-induced proliferation of T cells (see Fig. 4) considering that normal, HIV and ESRD IgM have a similar down regulating effect on CD2 and CD4 as well as on Zap-70 phosphorylation? A potential explanation centers on the physiological role of IgM-ALA. We postulate that under normal conditions, there is production of low levels of IgM-ALA primarily to attenuate T cell activation and the inflammatory response especially to autologous Ags and peptides. However, when confronted with excess T cell activation e.g., by nonself Ags or an acute or chronic inflammatory state, there is a significant increase in production of IgM-ALA and in addition there is recruitment and expansion of other B-1 cell clones that produce IgM with reactivity to other cell receptors as evidenced by ESRD (but not normal) IgM’s effect on CD86 (B7.2) to further regulate excess T cell activation, proliferation, and production of certain proinflammatory cytokines. In support of such a concept are observations in mice that are only deficient in IgM. Such mice exhibit an increased tendency to develop autoimmune diseases (42, 43). Additional support for such a conclusion also derives from studies of other investigators who clearly showed that impaired T cell activation and proliferation of T cells seen in ESRD patients were associated with a defect in macrophage activation of T cells. Macrophages from ESRD patients could not activate purified T lymphocytes from normal individuals and ESRD macrophages were found to have significantly decreased expression of CD86 (B7-2) (44).

Particularly intriguing is the observation that IgM-ALA down-regulates certain specific receptors (e.g., CD4 and CD2 on T cells and CD86 on macrophages) as well as certain cytokines (e.g., IL-2, TNF- α and IL-13). Perhaps such an innate mechanism evolved to inhibit receptors and cytokines that play a more dominant role in certain inflammatory processes. In support of such a concept is the observation that an Ab to a single proinflammatory cytokine (i.e., anti-TNF-α) significantly improves the acute inflammatory manifestations in patients with rheumatoid arthritis or Crohn’s disease as well as the inflammatory response triggered by an allograft rejection in humans (45, 46).

Currently, our studies have not addressed the mechanism by which IgM-ALA down-regulates CD2, CD4 and CD86 receptors or how IgM-ALA attenuates T cell activation and proliferation. Several potential mechanisms may be operative. IgM, by binding to CD3 and CD4, could directly inhibit T cell activation in an MLR. Murine IgG monoclonals specific for CD3 and CD4 have been shown to inhibit activation and proliferation of human T cells in response to alloantigens (MLR) (39, 47). Secondly, IgM by inhibiting IL-2 production could also inhibit T cell proliferation. Finally, recent studies provide indirect evidence to show that chemokine receptors can also influence or inhibit intracellular signaling involved in T cell activation (48, 49). It is therefore possible that IgM-ALA could also operate via such a mechanism. Clearly studies using monoclonal IgM-ALA with specificities for different receptors should help unravel some of the mechanisms leading to inhibition of T cell activation and proliferation in presence of IgM-ALA.

Several mechanisms could also be operative in IgM-ALA mediated inhibition of chemotaxis. An obvious mechanism is the inhibition of chemokine binding to the receptor. IgM also inhibited chemokinesis, which could result from IgM inactivating intracellular signaling or down-regulating receptors involved in chemokinesis and/or chemotaxis.

Finally, and of particular interest, is the observation that the repertoire of IgM-ALA can vary in normals as well as in patients. This is best exemplified with IgM-ALA obtained from HIV-I patients, where in certain patients IgM-ALA bound predominantly to CXCR4 and minimally to CCR5 and vice-versa. Similarly IgM from a normal individual (labeled no. 1) had a marked anti-proliferative effect in the allogeneic MLR (Fig. 4 B) even though the majority of other individual IgM obtained from normals, had no or minimal inhibitory effect on the MLR. In general, the majority of individual ESRD IgM was more inhibitory than HIV-1 IgM in the various assay systems. Normal IgM, in general, was the least inhibitory. Such variability may have functional consequences. For example, it is possible that individuals with IgM-ALA that fails to inhibit the production of TNF-α could be predisposed to a more acute and serious inflammatory response after exposure to a foreign Ag.

The maintenance of immune system homeostasis is essential for protecting the host against excess inflammation and autoimmunity (50, 51). In this regard, T cells that have escaped central deletion or silencing mechanisms are subjected to peripheral control mechanisms. Homeostasis maintenance by different subsets of natural and adaptive regulatory T cells (CD4+ or CD8+), NK-T cells and dendritic cells have gained particular attention for understanding peripheral regulation of self reactive T cells (Fig. 8) (52, 53, 54, 55, 56). Other observations would also indicate that TH2 inhibitory cytokines and certain costimulatory molecules (e.g., CD152, PDL1, and PDL2) serve as additional peripheral control mechanisms that inhibit T cell mediated immune response (56, 57, 58). Our studies would suggest yet another innate immune regulatory mechanism that involves a noncytolytic (at 37°C) naturally occurring IgM anti-leukocyte autoantibody to maintain immune homeostasis and protect against autoimmunity (Fig. 8). Potential mechanisms for IgM-ALA include inhibition of Th cell activation with decreased production of certain specific proinflammatory cytokines, e.g., TNF-α, IL-2. Importantly the current data would also indicate that IgM-ALA controls the magnitude of the inflammatory response by not broadly inhibiting all proinflammatory cytokines (possibly to avert excess immunosuppression) and secondly by regulating leukocyte chemotaxis. Furthermore, it would appear that IgM-ALA may not inhibit the function of CD8+ T cells as CD8 is not downmodulated. Studies are in progress to determine whether IgM-ALA inhibits CD8+ regulator T cells and CTLs. Additionally, the marked individual variation in the repertoire of IgM with specificity to the different leukocyte receptors that we observed in both normal and disease states, may play a role in contributing to the differences in the vigor or character of the inflammatory response among different individuals when exposed to the same inciting agent.

FIGURE 8.

Proposed mechanisms by which IgM-ALA regulates T cell-mediated inflammation, A, A schematic diagram of proposed mechanisms by which IgM-ALA regulates an inflammatory process initiated by tissue injury, e.g., secondary to alloantigen or virus. IgM-ALA binds to CD3/CD4 resulting in down-regulation of costimulatory molecules CD4 and CD2 on T cells and CD86 on APC and inhibition of cell activation, proliferation of T cells, and inhibition of certain specific proinflammatory cytokines, e.g.TNF-α, IL-2, IL-13. IgM-ALA also blocks chemokine receptors which results in decreased chemotaxis of inflammatory cells to the injured tissue. B, Some of the known innate and adaptive immune mechanisms that regulate T cell activation and inflammation following an inciting agent, e.g., HIV virus. Some of the viral induced T cell clones recognize autoantigens and are also inhibited by regulatory mechanisms to prevent autoimmune disease. Known mechanisms to avert autoimmunity are derived from knockout murine models including IgM-deficient mice. The proposed role of IgM-ALA in regulating activated T cells reactive to viral Ags and in inhibiting autoimmune T cells is also depicted. For brevity, all known regulatory mechanisms are not depicted in this figure.

FIGURE 8.

Proposed mechanisms by which IgM-ALA regulates T cell-mediated inflammation, A, A schematic diagram of proposed mechanisms by which IgM-ALA regulates an inflammatory process initiated by tissue injury, e.g., secondary to alloantigen or virus. IgM-ALA binds to CD3/CD4 resulting in down-regulation of costimulatory molecules CD4 and CD2 on T cells and CD86 on APC and inhibition of cell activation, proliferation of T cells, and inhibition of certain specific proinflammatory cytokines, e.g.TNF-α, IL-2, IL-13. IgM-ALA also blocks chemokine receptors which results in decreased chemotaxis of inflammatory cells to the injured tissue. B, Some of the known innate and adaptive immune mechanisms that regulate T cell activation and inflammation following an inciting agent, e.g., HIV virus. Some of the viral induced T cell clones recognize autoantigens and are also inhibited by regulatory mechanisms to prevent autoimmune disease. Known mechanisms to avert autoimmunity are derived from knockout murine models including IgM-deficient mice. The proposed role of IgM-ALA in regulating activated T cells reactive to viral Ags and in inhibiting autoimmune T cells is also depicted. For brevity, all known regulatory mechanisms are not depicted in this figure.

Close modal

We thank the Ryan White HIV Clinic at University of Virginia for identifying patients willing to donate their blood, and Rhonda S. Richardson for her expert assistance in preparing this manuscript.

Patents are pending on IgM anti-leukocyte antibodies.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institutes of Health, National Institute of Allergy and Infectious Disease Grant 5R21AI052740-02 (to P.I.L.), funds from the Nephrology Division Research Development Funds, University of Virginia Research Development funds, and the Fillipo Research Funds.

3

Abbreviations used in this paper: IgM-ALA, IgM anti-leukocyte autoantibodies; ESRD, end-stage renal disease patients; MCF, mean (geometric) channel fluorescence.

1
Nakamura, M., S. E. Burastero, Y. Ueki, J. W. Larrick, A. L. Notkins, P. Casali.
1988
. Probing the normal and autoimmune B cell repertoire with Epstein-Barr virus: frequency of B cells producing monoreactive high affinity autoantibodies in patients with Hashimoto’s disease and systemic lupus erythematosus.
J. Immunol.
141
:
4165
-4172.
2
Mouthon, L., S. Lacroix-Desmazes, A. Nobrega, C. Barreau, A. Coutinho, M. D. Kazatchkine.
1996
. The self reactive antibody repertoire of normal human serum IgM is acquired in early childhood and remains conserved throughout life.
Scand. J. Immunol.
44
:
243
-251.
3
Dighiero, G., B. Guilbert, J. P. Fermand, P. Lymberi, F. Danon, S. Avrameas.
1983
. Thirty-six human monoclonal immunoglobulins with antibody activity against cytoskeleton proteins, thyroglobulin, and native DNA: immunologic studies and clinical correlations.
Blood
62
:
264
-270.
4
Steele, E. J., A. J. Cunningham.
1978
. High proportion of Ig producing cells making autoantibody in normal mice.
Nature
274
:
483
-486.
5
Hardy, R. R., K. Hayakawa.
1986
. Development and physiology of Ly-1 B and its human homolog, Leu-1 B.
Immunol. Rev.
93
:
53
-79.
6
Clarke, S. H., L. W. Arnold.
1998
. B-1 cell development: evidence for an uncommitted immunoglobulin (Ig)M+ B cell precursor in B-1 cell differentiation.
J. Exp. Med.
187
:
1325
-1334.
7
Ailus, K., T. Palosuo.
1995
. IgM class autoantibodies in human cord blood.
J. Reprod. Immunol.
29
:
61
-67.
8
Barbouche, R., M. Forveille, A. Fischer, S. Amraveas, A. Durandy.
1992
. Spontaneous IgM autoantibody production In-vitro by B lymphocytes of normal human neonates.
Scand. J. Immunol.
35
:
659
-667.
9
Fagarasan, S., T. Honjo.
2000
. T-independent immune response: new aspects of B cell biology.
Science
290
:
89
-92.
10
Lacroix-Desmazes, S., S. V. Kaveri, L. Mouthon, A. Ayouba, E. Melanchere, A. Coutinho, M. D. Kazatchkine.
1998
. Self-reactive antibodies (natural autoantibodies) in healthy individuals.
J. Immunol. Methods
216
:
117
-137.
11
Brezinschek, H. P., S. J. Foster, R. I. Brenzinschek, T. Dorner, R. Domiati-Saad, P. E. Lipsky.
1997
. Analysis of human VH gene repertoire: differential effects of selection and somatic hypermutation on human peripheral CD5+/IgM and CD5/IgM+ B cells.
J. Clin. Invest.
99
:
2488
-2501.
12
Cervenak, J., K. Kiss, F. Uher.
1999
. Partial characterization of two lymphocyte-specific natural autoantibodies isolated from newborn mice.
Acta Microbiol. Immunol.
46
:
53
-62.
13
Terasaki, P. I., V. D. Mottironi, E. V. Barnett.
1970
. Cytotoxins in disease.
N. Engl. J. Med.
283
:
724
-727.
14
Winfield, J. B., R. J. Winchester, P. Wernet, S. M. Fu, H. G. Kunkle.
1975
. Nature of cold-reactive antibodies to lymphocyte surface.
Artritis Rheum.
18
:
1
-8.
15
Lobo, P. L..
1981
. Nature of autolymphocytotoxins present in renal hemodialysis patients: their possible role in controlling alloantibody formation.
Transplantation
32
:
233
-237.
16
Lobo, P. I., P. M. Surratt.
1979
. Studies on the autoantibody to lymphocytes in sarcoidosis.
J. Clin. Lab. Immunol.
1
:
283
-288.
17
Gilbreath, M. J., K. Pavanand, R. P. MacDermott, R. A. Wells, M. A. Ussery.
1983
. Characterization of cold reactive lymphocytotoxic antibodies in malaria.
Clin. Exp. Immunol.
51
:
232
-238.
18
Dorsett, B., W. Cronin, V. Chuma, H. I. Iochim.
1985
. Anti-lymphocyte antibodies in patients with the acquired immune deficiency syndrome.
Am. J. Med.
78
:
621
-626.
19
Williams, R. C., H. Masur, T. J. Spira.
1984
. Lymphocyte reactive antibodies in acquired immune deficiency syndrome.
J. Clin. Immunol.
4
:
118
-123.
20
Griggi, T., R. Bauer, T. Garofalo, S. Kukel, L. Lenti, A. P. Massetti, C. Muller, M. Sorice, G. M. Pointieri.
1994
. Autoantibodies against ganglioside GM3 represents a portion of antilymphocytic antibodies in AIDS patients.
Scand. J. Immunol.
40
:
77
-82.
21
Sorice, M., L. Lenti, R. Misasi, M. A. DiPrima, T. Griggi, L. Cignarella, V. Vullo, C. Masala.
1991
. Anti-glycosphingolipid antibodies in HIV infection.
AIDS
5
:
345
-346.
22
Strickler, R. B., T. M. McHugh, D. J. Moody, W. J. W. Morrow, D. P. Stites, M. A. Shuman, J. A. Levy.
1987
. An AIDS-related cytotoxic autoantibody reacts with a specific antigen on stimulated CD4 cells.
Nature
327
:
710
-713.
23
Warren, R. Q., E. A. Johnson, R. P. Donnelly, M. F. Lavia, K. Y. Tsang.
1988
. Specificity of anti-lymphocyte antibodies in sera from patients with AIDS-related complex (ARC) and healthy homosexuals.
Clin. Exp. Immunol.
73
:
168
-173.
24
Minota, S. W..
1987
. Identification of three major target molecules of IgM antilymphocytic autoantibodies in SLE.
J. Immunol.
139
:
3644
-3651.
25
Mimura, T., P. Fernsten, W. Jarjour, J. B. Winfield.
1994
. Autoantibodies specific for different isoforms of CD45 in SLE.
J. Exp. Med.
172
:
653
-656.
26
Stimmler, M. M., F. P. J. Quismorio, W. G. McGehee, T. Boylen, O. P. Sharma.
1989
. Anticardiolipin antibodies in acquired immunodeficiency syndrome.
Arch. Intern. Med.
149
:
1833
-1835.
27
Klouda, P. T., M. Jeannet.
1976
. Cold and warm antibodies and graft survival in kidney allograft recipients.
Lancet
1
:
876
-878.
28
Lobo, P. I., L. Rudolf, F. B. Westervelt.
1977
. Enhanced kidney allograft survival across a positive crossmatch (Cx) arising from B-cell specific and cold reactive antibodies.
Proc. Dial. Transpl. Forum.
7
:
4
-6.
29
Iwaki, Y., P. I. Terasaki, M. S. Park, R. Billing.
1978
. Enhancement of human kidney allografts by cold B-lymphocyte cytotoxins.
Lancet
1
:
1228
-1229.
30
Kerman, R. H., B. Susskind, I. Buyse.
1999
. Flow cytometry-detected IgG is not a contraindication to renal transplantation: IgM may be beneficial to outcome.
Transplantation
68
:
1855
-1858.
31
Przybylowski, P., M. Balogna, B. Radovancevic, O. H. Frazier, B. Susskind, S. C. Van Buren Katz, B. D. Kahan, R. Kerman.
1999
. The role of flow cytometry-detected IgG and IgM anti-donor antibodies in cardiac allograft recipients.
Transplantation
67
:
258
-262.
32
Lobo, P. I., C. E. Spencer, W. C. Stevenson, C. McCullough, T. L. Pruett.
1995
. The use of pronase-digested human leukocytes to improve specificity of the flow cytometric crossmatch.
Transpl. Int.
8
:
472
-480.
33
Chabot, D. J., H. Chen, D. S. Dimitrov, C. C. Broder.
2000
. N-linked glycosylation of CXCR4 masks coreceptor function for CCR5-dependent human immunodeficiency virus type 1 isolates.
J. Virol.
74
:
4404
-4413.
34
Mirzabekov, T., N. Bannert, M. Farzan, W. Hofmann, P. Kolchinsky, L. Wu, R. Wyatt, J. Sodroski.
1999
. Enhanced expression, native purification, and characterization of CCR5, a principal HIV-1 coreceptor.
J. Biol. Chem.
274
:
28745
-28750.
35
Burastero, S. E., D. Gafffi, L. Lopalco, G. Tambussi, B. Borgonovo, C. De Santis, C. Abecasis, P. Robbioni, A. Gasparri, A. Lazzarin, et al
1996
. Autoantibodies to CD4 in HIV type 1-exposed seronegative individuals.
AIDS Res. Hum. Retroviruses
12
:
273
-280.
36
Lobo, P. I., F. B. Westervelt, L. Rudolf, C. White.
1980
. Cold lymphocytotoxins: an important cause of acute tubular necrosis (ATN) occurring immediately post-transplant.
Lancet
2
:
879
-882.
37
Bernaman, A., T. Ternynck, S. Avrameas.
1992
. Natural mouse IgG reacts with self antigens including molecules involved in the immune response.
Eur. J. Immunol.
22
:
625
-633.
38
Lynch, G. W., M. Dearden, A. J. Sloane, I. Humphrey-Smith, A. L. Cunningham.
1996
. Analysis of recombinant and native CD4 by one-and two-dimensional gel electrophroesis.
Electrophoresis
17
:
227
-234.
39
Rao, P. E., W. Newman, E. Meyer, B. Fahmy, M. A. Talle, V. A. Fanning.
1992
. Identification of anti-CD3 antibodies that do not modulate antigen but induce mitogenesis and block the mixed lymphocyte reaction.
Hum. Immunol.
33
:
275
-283.
40
Signoret, N., J. Oldridge, A. Pelchen-Matthews, P. J. Klasse, T. Tran, L. F. Brass, M. M. Rosenkilde, T. W. Schwartz, W. Homes, W. Dallas, et al
1997
. Phorbol esters and SDF-1 induce rapid endocytosis and down modulation of the chemokine receptor CXCR4.
J. Cell Biol.
139
:
651
-664.
41
Winfield, J. B., T. Mimura.
1992
. Pathogenetic significance of anti-lymphocyte autoantibodies in systemic lupus erythematosus.
Clin. Immunol. Immunopathol.
63
:
13
-16.
42
Boes, M., T. Schmidt, K. Linkemann, B. C. Beaudette, A. Marshak-Rothstein, J. Z. Chen.
2000
. Accelerated development of IgG autoantibodies and autoimmune disease in the absence of secreted IgM.
Proc. Natl. Acad. Sci. USA
97
:
1184
-1189.
43
Ehrenstein, M. R., H. T. Cook, M. S. Neuberger.
2000
. Deficiency in serum immunoglobulin (Ig)M predisposes to development of IgG autoantibodies.
J. Exp. Med.
191
:
1253
-1258.
44
Girndt, M., M. Sester, U. Kau, H. Kohler.
2001
. Molecular aspects of T- and B-cell function in uremia.
Kidney Int.
78
: (Suppl.: 78):
S206
-S211.
45
McDevitt, H., S. Munson, R. Ettinger, A. Wu.
2002
. Multiple roles for tumor necrosis factor-α and lymphotoxin αβ in immunity and autoimmunity.
Arthritis Res.
4
: (Suppl. 3l):
S141
-S152.
46
Pascher, A., J. Klupp, J. M. Langreher, P. Neuhas.
2005
. Anti-TNF α therapy for acute rejection in intestinal transplantation.
Transplant Proc.
37
:
1635
-1636.
47
Pullar, C. E., P. J. Morris, K. J. Wood.
2003
. Altered proximal T-cell receptor signaling events in mouse CD4+ T cells in the presence of anti-CD4 monoclonal antibodies: evidence for reduced phosphorylation of Zap-70 and LAT.
Scand. J. Immunol.
57
:
333
-341.
48
Peacock, J. W., F. R. Jirik.
1999
. TCR activation inhibits chemotaxis toward stromal cell-derived factor-1: evidence for reciprocal regulation between CXCR4 and the TCR.
J. Immunol.
162
:
215
-223.
49
Ticchioni, M., C. Charvet, N. Noraz, L. Lamy, M. Steinberg, A. Bernard, M. Deckert.
2002
. Signaling through ZAP-70 is required for CXCL12-mediated T-cell transendothelial migration.
Blood
99
:
3111
-3118.
50
Freitas, A. A., B. Rocha.
2000
. Population biology of lymphocytes: the flight for survival.
Annu. Rev. Immunol.
18
:
83
-111.
51
Jiang, H., L. Chess.
2006
. Regulation of immune responses by T cells.
N. Eng. J. Med.
354
:
1166
-1176.
52
Munn, D. H., A. L. Mellor.
2003
. Macrophages and the regulation of self-reactive T cells.
Curr. Pharm. Des.
9
:
257
-264.
53
Sakaguchi, S., N. Sakaguchi, J. Shimizu, S. Yamazaki, T. Sakihama, M. Itoh, Y. Kuniyasu, T. Nomura, M. Toda, T. Takahashi.
2001
. Immunologic tolerance maintained by CD25+ CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance.
Immunol. Rev.
182
:
18
-32.
54
Shevach, E. M..
2002
. CD4+ CD25+ suppressor T cells: more questions than answers.
Nat. Rev. Immunol.
2
:
389
-400.
55
Ludewig, B., P. Krebs, T. Junt, G. Bocharov.
2003
. Dendritic cell homeostasis in the regulation of self-reactivity.
Curr. Pharm. Des.
9
:
221
-231.
56
Nelson, B. H..
2004
. IL-2, regulatory T cells, and tolerance.
J. Immunol.
172
:
3983
-3988.
57
Judge, T. A., Z. Wu, X. G. Zheng, A. H. Sharpe, M. H. Sayegh, L. A. Turka.
1999
. The role of CD80, CD86, and CTLA4 in alloimmune responses and the induction of long-term allograft survival.
J. Immunol.
162
:
1947
-1951.
58
Sayegh, M. H., L. A. Turka.
1998
. The role of T-cell costimulatory activation pathways in transplant rejection.
N. Engl. J. Med.
338
:
1813
-1821.
59
Lapham, C. K., T. Romantseva, E. Petricoin, L. R. King, J. Manischewitz, M. B. Zaitseva, H. Golding.
2002
. CXCR4 heterogeneity in primary cells: possible role of ubiquitination.
J. Leukocyte Biol.
72
:
1206
-1214.