The production of IgG HLA antibodies after lung transplantation (LTx) is considered to be a major risk factor for the development of chronic rejection, represented by the bronchiolitis obliterans syndrome (BOS). It has recently been observed that elevated levels of IgM HLA antibodies also correlates with the development of chronic rejection in heart and kidney transplantation. This study investigates the relationship between IgM and IgG antibodies against HLA and MICA after lung transplantation. Serum was collected from 49 patients once prior to transplantation and monthly for up to 1 year after lung transplantation was analyzed by Luminex to detect IgM and IgG antibodies against HLA and MICA. The presence of either IgM or IgG HLA and/or MICA antibodies prior to or after transplantation was not related to survival, gender, primary disease, or the development of BOS. Additionally, the production of IgG alloantibodies was not preceded by an increase in levels of IgM, and IgM levels were not followed by an increase in IgG. Under current immune suppressive regimen, although the presence of IgM antibodies does not correlate with BOS after LTx, IgMhigh IgGlow HLA class I antibody titers were observed more in patients with BOS compared to patients without BOS.
The bronchiolitis obliterans syndrome (BOS) represents chronic rejection that accounts for the majority of mortality after lung transplantation (LTx). Almost 50% of lung transplant recipients develop BOS within five years after LTx [
The presence of IgG anti-HLA in patient sera reactive with antigens present on the donor lung prior to transplantation is a contraindication for transplantation [
The goal of this study was to determine the relationship between IgM HLA antibodies after lung transplantation and the development of BOS. Additionally, we examined whether a correlation existed between IgM and IgG HLA antibodies after lung transplantation to determine if the isotype switch is inhibited by the immune suppressive regimen of tacrolimus/mycophenolate mofetil.
A total of 49 LTx patients transplanted between September 2003 and March 2008 at the Heart Lung Centre in Utrecht, The Netherlands, who exhibited a greater than three months survival, were included in this study. Eleven patients developed BOS during followup. BOS was defined as an irreversible decline in FEV1 of more than 20% compared to the postoperative baseline in the absence of infection or other etiology [
The study design was approved by the medical ethical committee. Informed consent was obtained from each patient. Patients donated blood every month during the first year after transplantation and once every three months in the following years.
Sera of 49 patients with known HLA type were screened for the presence of anti-HLA and anti-MICA antibodies prior to transplantation and then longitudinally, with an average of 20.7 months (range 10–29 months), after transplantation. A total of 382 samples were analyzed for IgG, and 477 samples were analyzed for IgM antibodies against HLA using Luminex beads according to the manufacturer’s protocol (LABScreen Mixed, LSM12, One Lambda). For the IgM assay, the IgG detecting antibody was replaced with an IgM detecting antibody (R-phycoerythrin-conjugated AffiniPure F(ab) Fragment Donkey anti-human IgM obtained from Jackson ImmunoResearch). Three hundred eighty-two samples obtained from 49 patients were analyzed for both IgG and IgM.
The IgM tests were validated by measurement of 30 unimmunized males and measurements of samples obtained from kidney transplantation patients known to have high titers of IgM using both ELISA and Luminex. Optimal serum dilutions, conjugate dilutions, and incubation times were determined.
Dithiothreitol (DTT) was used to deplete serum IgM as previously described [
Positivity for IgG anti-HLA or -MICA was defined using the default settings of the software (Fusion) provided by the manufacturer (One Lambda). For IgM-positive HLA and MICA antibodies, a threshold was defined. Thirty healthy HLA-unimmunized males were analyzed to determine background levels. For each bead, the average background, including the range and standard deviation, was determined in these healthy controls. A patient was considered to have positive antibodies when the mean fluorescent intensity (MFI) (corrected for the negative bead) was >5 times the average background MFI of the unimmunized males.
Titers of IgM and IgG antibodies were defined as IgMhigh/IgMlow when MFI values were above or below the overall average of all beads in thirty healthy HLA-unimmunized males or IgGhigh/IgGlow when MFI levels were above or below the average background levels of all beads for IgG as defined by the manufacturer.
For each patient, it was determined which beads contained at least 1 antigen from a mismatched donor HLA. These beads were considered to possess possible donor-specific antigens (DSA). Beads that were negative for mismatched donor HLA antigens were identified as third-party antigens (TPA).
A
This study involved 49 patients whose details are displayed in Table
Patient characteristics.
BOS | Non-BOS | |
---|---|---|
Total number | 11 | 38 |
Deceased | 4 (36%) | 1 (3%) |
Gender | ||
Male | 3 (27%) | 22 (58%) |
Female | 8 (73%) | 16 (42%) |
Average age (years; SD) | 45 ( | 42 ( |
Primary disease | ||
CF | 3 (27%) | 19 (50%) |
EMF | 5 (46%) | 9 (24%) |
FD | 3 (27%) | 10 (26%) |
Type of graft | ||
Single | 2 (18%) | 4 (11%) |
Bilateral | 9 (82%) | 34 (89%) |
Average ischemic time (minutes; SD) | ||
Single | 216 (±70) | 209 (±43) |
Bilateral | 317 (±51) | 322 (±72) |
HLA mismatches (average; SD) | ||
Class I | 3.6 (±0.7) | 3.1 (±0.7) |
Class II | 1.7 (±0.5) | 1.7 (±0.5) |
BOS onset (months; SD) | 22.5 (±13.9) | ND |
BOS grade | ||
I | 3 (27%) | ND |
II | 2 (18%) | |
III | 6 (55%) |
BOS: bronchiolitis obliterans syndrome, SD: standard deviation, CF: cystic fibrosis, EMF: emphysema, and FD: fibrotic disease.
We obtained serum from 41 patients prior to transplantation. Twelve of these patients tested positive for IgM anti-HLA class-I and/or class-II, and 5 patients tested positive for IgG HLA class I or II antibodies, indicating that IgM antibodies against HLA occurred more frequently in our study population compared to IgG antibodies. Additionally, antibodies against MICA were more frequently of the IgM isotype (
Average mean fluorescent intensity (MFI) of IgM isotype HLA class I, class II and MICA antibodies over time (a quarter is equal to 4 months). Forty-one patients were analyzed once prior to LTx, and 49 patients were measured longitudinally after transplantation resulting in a total of 477 samples that were analyzed by Luminex (LABScreen Mixed, LSM12, One Lambda, Calif, USA), with an average of 9.7 samples (range 5–10 samples) per patient. Shown are results of Luminex analysis with beads containing HLA class I (closed dot) class II (closed square), or MICA antigens (triangle with dotted line). These data were averaged from all LTx patients per quarter after lung transplantation.
A total of 341 serum samples taken after LTx were analyzed for both HLA and MICA antibodies of IgM and IgG isotype, with a total of 12,958 bead-serum combinations being measured. The MFI values of beads containing either HLA class I, HLA class II, or MICA antigens are displayed in Figure
Correlation plots of IgM and IgG HLA class I (a), HLA class II (b), and MICA (c) antibodies after LTx. Grey lines represent the average background as detected in 27 unimmunized males (IgM) or as defined by the manufacturer (IgG). The background range of the individual beads is displayed by the grey area.
To investigate whether the lack of positivity for IgG anti-HLA antibodies may result from the physical inhibition by IgM anti-HLA antibody binding to the beads, 10 sera samples exhibiting elevated levels of MFI specific for IgM anti-HLA class I antibodies were selected and treated with DTT to disrupt IgM. One sample exhibiting high IgG MFI levels and low expression of IgM was used. IgM anti-HLA was not detected following incubation with DTT, and MFI levels of IgG anti-HLA did not increase. The IgG anti-HLA control sample still exhibited elevated MFI values following DTT treatment, indicating that IgG binding was not impaired by DTT (data not shown). Given these findings, the lack of IgG anti-HLA detection in serum samples with IgM anti-HLA antibodies is not the result of inhibition by antibodies of the IgM isotype.
In samples containing elevated levels of IgM or IgG antibodies against HLA, it was investigated whether high IgM was followed in time by high IgG or whether high IgG was preceded by high IgM. To this end, beads were identified recognized by patient sera and the other sera collected from the same patients at other time points were analyzed for reactivity against the same beads. Figure
Examples of no correlation between levels of IgM and IgG anti-HLA. Elevated levels of IgM are not followed in time by high levels of IgG levels (a), and increased levels of IgG are not preceded by high levels of IgM (b). Figures are indicative of 1 bead obtained from 1 patient in time.
We next investigated if clinical characteristics were related to IgM or IgG anti-HLA antibodies present prior to lung transplantation. There was no difference in the gender of patients that tested positive for IgM or IgG isotype HLA antibodies prior to transplantation. The percentage of males expressing IgM antibodies against MICA prior to transplantation was slightly higher than that of females (
Analysis of antibody titers after LTx indicated no relationship to the development of BOS (data not shown), as no rise or fall in antibody titers was observed prior to BOS diagnosis. The presence of IgM or IgG antibodies against HLA class I, HLA class II, and/or MICA prior to or after lung transplantation is not related to the development of BOS. Donor-specific antibodies may better correlate with development of BOS, so we designated DSA for beads with possible donor-specific antigens and TPA beads lacking these antigens. Our results indicated no differences between patients with BOS and without BOS for IgM anti-HLA, although a trend was observed in patients lacking IgG anti-HLA, as they developed BOS with a higher frequency (
Differences in HLA class I antibody isotype profiles in BOS and non-BOS patients.
Mean fluorescent intensities of HLA class I antibodies. (a) All HLA class I antibodies of both IgG and IgM isotype in BOS and non-BOS patients. (b) Possible DSA HLA class I antibodies of both IgG and IgM isotype in BOS and non-BOS patients. Error bar depicts the standard error of mean.
In this study, we demonstrate that the presence of IgM HLA and MICA antibodies prior to or after lung transplantation is not related to the development of BOS. We also show that the course of IgM antibody titers is stable after lung transplantation. After LTx, there is no correlation between IgM and IgG antibodies, as high IgM is not followed by high IgG and high IgG is not preceded by IgM. Patients diagnosed with BOS, however, do exhibit elevated HLA class I IgM antibody titers and low IgG titers compared to patients without BOS.
IgM antibodies against HLA have been considered to be clinically irrelevant, but recent findings challenge this concept. In heart transplantation, it was shown that non-HLA antibodies of IgM isotype were correlated with a reduction in graft survival [
One study detected a correlation between MICA antibodies and development of rejection after kidney transplantation. However, a number of other studies of heart or kidney transplantation did not find such a relation [
High levels of IgM antibodies in serum may physically inhibit the binding of IgG antibodies to the Luminex beads [
Although titers of HLA class I IgMhigh IgGlow antibodies are elevated in patients with BOS, this cannot be used as a possible marker for the development of BOS because some patients without BOS also have antibody titers exhibiting elevated IgM and low IgG. Additionally, although patients without BOS express higher HLA IgG antibody levels compared to patients with BOS, some considerations have to be placed. First, HLA IgG antibody titers are low and may not be capable of activating the complement cascade. Second, it is unknown if IgG HLA antibodies are of the IgG1, IgG3 subclasses able to fix complement, or of the IgG2 and IgG4 subclasses. If HLA antibodies are sequestered to the transplanted lung rendering them undetectable in BOS patients, they could fix the complement system and hence contribute to the development of BOS after lung transplantation [
In conclusion, lung transplant patients develop IgM HLA and MICA antibodies prior to and after transplantation. The presence of these antibodies and their titers are not related to chronic rejection under the current immune suppressive regime, and the isotype switch is inhibited. Patients diagnosed with BOS, however, exhibit elevated levels of IgM and reduced IgG HLA expression compared to patients without BOS. HLA antibodies are thought to be involved in chronic graft rejection as a result of their complement fixing ability as observed by C4D deposition within the graft. Mechanistically, our data could imply that IgM HLA antibodies are relevant to the pathogenesis of BOS because they are able to fix complement more efficiently than IgG.
The authors would like to thank One Lambda for kindly providing the Luminex kits LABScreen Mixed.