Human leucocyte antigen-defined microchimerism early post-transplant does not predict for stable lung allograft function

Authors

  • L. C. Rowntree,

    1. Department of Medicine, Monash University, Central Clinical School
    2. Department of Allergy, Immunology and Respiratory Medicine, The Alfred Hospital, Melbourne
    Search for more papers by this author
  • J. Bayliss,

    1. Department of Medicine, Monash University, Central Clinical School
    2. Department of Molecular Research and Development, Victorian Infectious Diseases Reference Laboratory, North Melbourne, Vic., Australia
    Search for more papers by this author
  • T. H. O. Nguyen,

    1. Department of Medicine, Monash University, Central Clinical School
    2. Department of Allergy, Immunology and Respiratory Medicine, The Alfred Hospital, Melbourne
    Search for more papers by this author
  • T. C. Kotsimbos,

    1. Department of Medicine, Monash University, Central Clinical School
    2. Department of Allergy, Immunology and Respiratory Medicine, The Alfred Hospital, Melbourne
    Search for more papers by this author
    • Equal senior authors.
  • N. A. Mifsud

    Corresponding author
    1. Department of Allergy, Immunology and Respiratory Medicine, The Alfred Hospital, Melbourne
    • Department of Medicine, Monash University, Central Clinical School
    Search for more papers by this author
    • Equal senior authors.

Correspondence: N. A. Mifsud, Departments of Medicine and Allergy, Immunology and Respiratory Medicine, Monash University, Level 7 The Alfred Centre, 99 Commercial Road, Melbourne, Vic. 3004, Australia.

E-mail: nicole.mifsud@monash.edu

Summary

Microchimerism is the presence of foreign cells in an individual below 1% of total cells, which can occur in the setting of solid organ transplantation. This study quantitated donor-derived cellular subsets longitudinally in human leucocyte antigen (HLA)-mismatched lung transplant recipients (LTR) during the first post-operative year and evaluated the pattern of peripheral microchimerism with clinical outcomes. Peripheral blood mononuclear cells (PBMC) isolated from non-HLA-B44 LTR who received HLA-B44 allografts were sorted flow cytometrically into three cellular subsets. Real-time quantitative polymerase chain reaction (q–PCR) demonstrated that donor-derived HLA-B44 microchimerism is a common phenomenon, observed in 61% of patients. The level of donor-derived cells varied across time and between LTR with frequencies of 38% in the B cells/monocytes subset, 56% in the T/NK cells subset and 11% in the dendritic cells (DC) subset. Observations highlighted that microchimerism was not necessarily associated with favourable clinical outcomes in the first year post-lung transplantation.

Introduction

Chimerism represents the presence of at least two genetically different DNA sources in the same individual, with foreign cell frequencies of fewer than 1% defining microchimerism and frequencies of greater than 1% representing macrochimerism [1-5]. The persistence of donor-derived leucocytes and their migration throughout a transplant recipient's body has been suggested to play an important role in the acceptance of the grafted organ [6, 7]. Previous work has suggested that induction of macrochimerism through haematopoietic stem cell transplantation alongside solid organ transplantation could lead to transplant tolerance [8-11]. While macrochimerism in human organ transplantation has been associated with operational tolerance and allograft acceptance [12, 13], questions still remain regarding the threshold level of chimerism needed to induce clinically significant control of alloimmunity [14]. The role of microchimerism in allograft tolerance induction or the contribution to adverse clinical outcomes, via alloreactivity, remains unclear in lung transplantation – in particular, whether microchimerism can induce prolonged allograft survival, or is it simply an epiphenomenon associated with the transfer of organ transplantation [2, 15]? Microchimerism from a clinical perspective has the potential to be a more sensitive measure of allograft acceptance than macrochimerism, although specificity may be compromised by focusing on very low levels of chimerism.

The quantitation of relatively low frequencies of donor-derived cells in the peripheral blood of transplant recipients requires a technical approach that is both sensitive and specific at the detection level. Nested polymerase chain reaction (PCR), quantitative PCR (q-PCR) and flow cytometric phenotyping techniques have been used individually to detect microchimerism successfully [5, 16, 17]. In addition, q-PCR established to discriminate between donor and host-derived cells using the sex determining region of the Y chromosome (SRY) generated reliable results, but was limited to gender-mismatched transplants [5, 17]. Donor cell discrimination focusing on human leucocyte antigen (HLA) offers a broader target due to the extreme genetic diversity or polymorphism displayed by HLA molecules within a population of unrelated individuals. We show that combining both flow cytometric cellular phenotyping and sorting followed by q-PCR for measurement of donor-specific HLA alloantigens facilitates exploration of microchimerism in solid organ transplantation.

Given the uncertainty regarding the relationship between microchimerism and its influence on immunological outcomes following transplantation, studies have investigated the role of donor-specific cellular subsets [16, 17]. We have reported previously in lung transplant recipients (LTR) that the frequency of microchimerism in both PBMC [excluding dendritic cells (DC)] and DC alone subsets were extremely heterogeneous, both within an individual patient and across the transplant cohort, and differed dramatically in their persistence in the first 12 months post-lung transplantation [17]. While this study was limited to 11 patients, as microchimerism was detected using the q-PCR of the sex-determining region of the Y chromosome (SRY) gene, the prolonged persistence of DC microchimerism was suggestive of an immunomodulatory role geared towards tolerance induction.

This study builds upon these findings by further dissection of peripheral blood mononuclear cell (PBMC) populations to determine whether particular subsets dominate the repertoire of microchimeric cells. Utilizing one of the most common HLA antigens, HLA-B44, in our population we were able to explore the extent of microchimerism in 20 LTR (non-HLA-B44) who had received an HLA-B44 allograft, which enabled a clear distinction between patient and donor-derived cells. In addition, we segregated specific populations of immune cells into subsets to quantitate the extent of donor-derived microchimerism exhibited. This dissection enabled evaluation of whether or not the pattern of peripheral microchimerism during the first 12 months post-lung transplant contributed to meaningful clinical outcomes, including acute cellular rejection (ACR), lung function, development of bronchiolitis obliterans syndrome (BOS; chronic rejection) and survival.

Materials and methods

Patient cohort, post-transplant clinical management and ethics

A retrospective cohort (2008–10) of 20 LTR was chosen for this study based on HLA selection criteria of a non-HLA-B44 recipient receiving an HLA-B44 lung allograft (age range 23–70 years; 11 males of 20 recipients). All LTR received standard triple immunosuppression (cyclosporin/tacrolimus, azathioprine/mycophenolate mofetil and corticosteroids). Patients identified pretransplant as being at high risk of developing renal dysfunction received induction therapy with the interleukin (IL)-2 receptor blocker basiliximab (20 mg intravenously on days 0 and 4) as a calcineurin inhibitor sparing agent. All LTR underwent routine surveillance bronchoscopy at 0·5, 1, 3, 6, 9 and 12 months post-transplant or if indicated clinically [18], at which time transbronchial biopsies were assessed for ACR according to standard histopathological criteria [19, 20]. Pulmonary function testing [spirometry-forced expiratory volume in one second (FEV1)] with chronic rejection/BOS was defined as sustained loss of pulmonary function with FEV1 falling below 80% of personal best [20, 21]. Patients and HLA-B44-positive healthy controls (n = 5) provided written consent and ethics approvals were granted by The Alfred Hospital (Victoria, Australia) and Monash University (Victoria, Australia).

Cellular subset phenotyping and isolation

PBMC collected at specific time-intervals of pretransplant and 1, 3, 6 and 12 months post-transplant (±4 weeks) were isolated by Ficoll (GE Healthcare, Uppsala, Sweden) density gradient centrifugation and cryopreserved at −180°C. Purification of cellular subsets was performed on cryopreserved PBMCs; thus, we cannot exclude the possibility that subset survival may be affected by this procedure although, in our hands, this is not the case when comparisons were made (data not shown). Phenotyping of thawed PBMC was performed using both monoclonal antibodies to identify T cells (CD3+; clone UCHT1; BD Pharmingen, Franklin Lakes, NJ, USA), natural killer (NK) cells (CD56+; clone NCAM16·2; BD Pharmingen) and NKT cells (CD3+CD56+). A blood DC enumeration kit (Miltenyi Biotec, Auburn, CA, USA) was used to identify B cells (CD19+), monocytes (CD14+) and DC subsets [myeloid DC1 (mDC1): BDCA-1+; plasmacytoid DC (pDC): BDCA-2+, myeloid DC2 (mDC2): BDCA-3+]. Cells were enumerated and sorted [fluorescence activated cell sorter (FACS)Aria; BD Biosciences] using sequential gating strategies into three specific populations: (i) B cells and monocytes, (ii) T cells, NK cells and NKT cells (T/NK cells) and (iii) DC (Fig. 1a). Given that microchimerism occurs in fewer than 1% of circulating foreign cells, pooled cellular subsets were examined to detect robust patterns and relatively low frequencies of HLA-B44 microchimerism. Each cellular subset was lysed using QuickExtract formalin-fixed, paraffin embedded (FFPE) DNA extraction solution, according to the manufacturer's instructions (Epicentre, San Diego, CA, USA).

Figure 1.

Phenotyping and quantitation of cellular subsets. (a) Peripheral blood mononuclear cells (PBMC) labelled with lineage markers (CD3, CD14, CD19, CD56, BDCA-1, BDCA-2 and BDCA-3) were sorted into three cellular subsets using a series of gating strategies. Lymphocytes were identified by forward-scatter (FSC) versus side-scatter (SSC). Cellular subset 1 comprised B cells (CD14, CD19+) and monocytes (CD14+, CD19), referred to hereafter as B cells/monocytes subset. Cellular subset 2 comprised T cells (CD3+, CD56), natural killer (NK) cells (CD3, CD56+) and NKT cells (CD3+, CD56+), referred to hereafter as T/NK cells subset. Cellular subset 3 comprised dendritic cells (DC) (1) and (2) (CD14, CD19, BDCA-1+, BCDA-2+, BCDA-3+), referred to hereafter as DC subset. (b) Group analysis for B cells/monocytes, T/NK cells and DC subsets between the lung transplant recipients (LTR) cohort (n = 18) and healthy donors (n = 5) was performed. Box-and-whisker graphs show median, quartiles and range of the percentage of live cells within each subset. Statistically significant differences are denoted (*P < 0·05; **P < 0·01).

q-PCR analysis of sorted cellular subsets

q-PCR analyses were performed on a 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA, USA) using RPP30 (housekeeping gene, internal positive control) and HLA-B44 to quantitate both total and donor-derived cell numbers. Primer pairs and probes were used at 100 nM and 50 nM for the RPP30 gene and 50 nM and 100 nM for the HLA-B44 gene, respectively, in a total volume of 9 μl with 1 μl of template DNA. The primer pair for the RPP30 gene was 5′-CGGACGGTCATGGGACTTC-3′ (forward) and 5′-CGCTCGCAGGTCCAAATG-3′ (reverse) and the probe 5'-VIC-ATGGCGGTGTTTGC-MGBNFQ-3′. The primer pair for the HLA-B44 gene was 5′-GGACCGGGAGACACAGATCT-3′ (forward) and 5′-CGGTGCGCAGGTTCTCTC-3′ (reverse) and the probe 5'-6FAM-AAGACCAACACACAGAC-MGBNFQ-3′. Standard curves generated using titration of RPP30 and HLA-B44-positive healthy control samples determined the dynamic linear range to be between 0·034 and 141·17 ng/μl. Known concentrations of HLA-B44 and non-HLA-B44 DNA were mixed in ratios ranging from 1:10 to 1:10 000 along with positive and negative controls. The threshold for detection was determined to be between 1:2000 and 1:10 000. All patient samples and standards were run in triplicate.

Exclusion of patients with pre-existing HLA-B44 microchimerism

The existence of microchimerism can occur through pregnancy, blood transfusion and allogeneic transplantation. To remove pre-existing microchimerism as a confounder we examined pre-transplant DNA samples for the presence of HLA-B44 in our patient cohort using q-PCR. Two patients [transplant (Tx) numbers 1 and 6] demonstrated pre-transplant HLA-B44 microchimerism in all three cellular subsets and were subsequently excluded from further analysis. The remaining 18 patients were negative for the presence of pre-transplant donor specific antibodies directed towards HLA-B44.

Statistics

Statistical analysis was performed using the Student's t-test (two-tailed distribution) and the Mann–Whitney U-test, where appropriate, using GraphPad Prism (GraphPad Software, San Diego, CA, USA). Statistical significance was assigned with a P-value of <0·05.

Results

Microchimerism is detectable in a variety of cellular populations

Quantitation of each of the three cellular subsets in 18 patients at five time-intervals (pretransplant and 1, 3, 6 and 12 months post-transplant) and five healthy controls was performed using flow cytometry. Cellular subset frequencies across the first year post-transplant in LTR were relatively stable at both individual and group levels; (i) 31·09 ± 13·24% [median ± standard deviation (s.d.)] for B cells/monocytes, (ii) 59·11 ± 17·61% for T/NK cells and (iii) 0·17 ± 0·65% for DC (Fig. 1b). In the B cells/monocytes subset, there was no significant difference between the percentage of live cells in the healthy controls and LTR while, in the T/NK cells subset, a significant difference between the percentage of live cells in the healthy controls and LTR at pretransplant, 1 and 6 months post-transplant was observed. Interestingly, significant differences between the percentage of live cells in the healthy controls and LTR were shown at all time-intervals in the DC subset. In addition, a significant difference was also determined between the percentage of live cells in DC in LTR between the pretransplant and 1 month post-transplant samples (P = 0·010, Mann–Whitney U-test) (Fig. 1b). Quantitation of cell numbers for each cellular subset was performed on purified sorted populations followed by q-PCR of the housekeeping gene RPP30 (sensitivity level of 0·034 ng/μl). Similar median cell numbers were observed for both B cells/monocytes and T/NK cells subsets, 12 829 ± 7689 and 10 537 ± 7010, respectively, while lower cell numbers were observed in the DC subset, 243 ± 2199 (Fig. 2).

Figure 2.

Total cell number quantitated by quantitative polymerase chain reaction (q-PCR) of a housekeeping gene. Cell number was determined using DNA quantity from q-PCR of the housekeeping gene RPP30 [cell number = DNA quantity × 3 (PCR constant) × 1000 (ng to pg) ÷ 6 (6 pg DNA per cell)]. Similar numbers of cells were observed in B cells/monocytes and T/natural killer (NK) cells subsets, median 12 829 ± 7689 (range 66–42 157) and median 10 537 ± 7010 (range 623–36 057), respectively. In comparison, reduced cell numbers were observed in the dendritic cells (DC) subset, median 243 ± 2199 (range 0–19 450).

Of the 61% of patients (11 of 18) that demonstrated HLA-B44 microchimerism, variability was observed across the three cellular subsets, with 38% (six of 16) in the B cells/monocytes subset, 56% (nine of 16) in the T/NK cells subset and 11% (two of 18) in the DC subset. While there was great variability in the magnitude and dynamics of HLA-B44 microchimerism observed in the first year post-lung transplantation in the B cells/monocytes and T/NK cells subsets, the majority of patients followed a similar pattern, with lower levels of HLA-B44 microchimerism being detected at 1 and 6 months post-transplant and maximum levels observed at 3 months post-transplant (Fig. 3a). This pattern was most prevalent in the B cells/monocytes subset, with 83% (five of six) of patients displaying this configuration. Similar fluctuations were observed in the T/NK cells subset, but HLA-B44 microchimerism levels did not adhere to the kinetic pattern to the same extent. Group data analysis showed that the average level of microchimerism in the B cells/monocytes and T/NK cells subsets were statistically significant (Student's t-test) at 3 months post-transplant compared to pretransplant (B cells/monocytes, P = 0·026; T/NK cells, P = 0·041) or 6 months post-transplant (B cells/monocytes, P = 0·029). At all other time-intervals, no statistically significance difference was observed. HLA-B44 microchimerism in the DC subset was observed in only two patients (Tx 2 and 3). Interestingly, DC subset chimerism observed in Tx 3 at 1 and 3 months post-transplant was at relatively high levels of 1·8% and 2·4%, respectively (Fig. 3a). In healthy individuals, the DC population represents 1% (pDC 0·37%, mDC1 0·60%, mDC2 0·03%) of total PBMC isolated from peripheral blood. Thus, the low level of detectable chimerism in the DC subset may be attributed to the low DC numbers isolated from patient PBMC, with the exceptions of Tx 2 and 3.

Figure 3.

Dynamics of human leucocyte antigen (HLA)-B44 microchimerism in lung transplant recipients (LTR). (a) Patient dynamics in the first 12 months post-transplant demonstrate the presence and magnitude of donor-derived cells following examination of HLA-B44 microchimerism in all three cellular subsets; B cells/monocytes, T/natural killer (NK) cells and dendritic cells (DC) subsets. Patients with no detectable HLA-B44 microchimerism were depicted as a zero value. (b) Individual analyses of patient levels of HLA-B44 microchimerism, pulmonary function (measured by % personal best) and acute cellular rejection (ACR) showed no correlation.

Microchimerism early post-lung transplantation does not necessarily predict stable allograft function

Stable pulmonary function, as measured by percentage of personal best, was observed in the majority of patients both in the presence [82%; (nine of 11, Tx 4, 5, 7, 8, 11, 12, 13, 14 and 18)] or absence [86%; (six of seven, Tx 9, 15, 16, 17, 19 and 20)] of HLA-B44 microchimerism (Fig. 3b). Of the 18 patients, three LTR showed a sustained and irreversible loss of lung function, indicating the onset of BOS (Tx 2, 3 and 10). Microchimerism was found in two of the three patients with BOS, suggesting that microchimerism does not necessarily protect against allograft function loss.

Three patients also experienced episodes of ACR within the first 6 months post-lung transplantation [ACR grade: Tx 2 (A1 at 6 months), Tx 12 (A2 at 1·5 months and A1 at 3 months) and Tx 13 (A1 at 1 month)]. Although microchimerism was identified in all three patients, it was also present in patients without ACR (Fig. 3b). Three patients displayed HLA-B44 macrochimerism. Interestingly, Tx 3 was found to have early DC macrochimerism, with donor cell numbers decreasing by 6 months post-transplant. This decrease in chimerism occurred with a sustained loss of lung function, which supports our previous study findings where DC chimerism persistence was suggestive of positive clinical outcomes [17]. Additionally, Tx 7 was observed to have persistent macrochimerism in both B cells/monocytes and T/NK cell subsets and experienced stable lung function during the first 12 months post-lung transplantation, while Tx 12 demonstrated marginal macrochimerism in the T/NK cells subset.

Discussion

This study utilized HLA mismatching successfully between transplant donor and recipient as a discriminator for foreign cell chimerism. A number of studies have determined that the immune mechanisms associated with the existence of macrochimerism are capable of inducing graft tolerance following human organ transplantation [14]. We sort to build upon this work by examining the potential of durable low-level chimerism, or microchimerism, to contribute to meaningful clinical outcomes in the first 12 months following lung transplantation. We investigated the dynamics of different cellular subsets in LTR and detected chimeric patterns in more than half the patient cohort. Our findings suggested that the presence of microchimerism in either B cells/monocytes and/or T/NK cells subsets did not necessarily associate with favourable clinical outcomes, but supported our previous study [17] demonstrating that loss of DC microchimerism was associated with a decline in physiological lung function. This current work, together with published reports investigating macrochimerism [12-14], suggest that a minimum threshold of donor-derived cells, likely to be above 1%, may be required for transplant tolerance induction.

Fluctuations in donor-derived cell numbers may result from differences in the mobilization rate of these cells from within the allograft into the peripheral blood of the recipient and also the rate of elimination by the host's immune system [22]. We have suggested previously that a rapid loss of microchimerism may be causative of either a vigorous alloresponse and/or a failure of donor cells to induce allograft tolerance that supresses immune reactivity [17]. If changes in microchimerism are driven solely by changes in allogeneic immune responses, little difference would be observed in the kinetics of the different cellular subsets. However, both this study and our previous publication demonstrated that dynamics or kinetics of microchimerism fluctuated over time and were discordant among individual LTR. Variation between donor-derived cellular subsets may reflect their individual capabilities to proliferate in a hostile environment by avoiding immune detection by host cells. In addition, the potent immunosuppressive therapies administered post-lung transplantation most probably influences the cellular subsets differently, with most drugs targeting T cell responses [23].

Chronic rejection, in the form of BOS, remains the major clinical barrier to long-term survival following lung transplantation, with 49% of recipients developing BOS by 5 years post-transplant [24]. Previous studies have found that chimerism, in the form of haematopoietic stem cell transplantation and/or solid organ transplantation, has the potential to induce transplant tolerance [13, 25]. Within this framework, microchimerism has been suggested to play an important role in graft rejection or tolerance, but there is no consensus regarding the relationship between microchimerism and the clinical outcomes in transplantation [1, 3, 4]. Indeed, we observed no relationship between the presence of microchimerism and clinical outcome measures of ACR, lung function and survival. This corroborates two previously published studies on microchimerism following lung transplantation, by Calhoun et al. [26] and Paantjens et al. [16], that did not detect a relationship between microchimerism levels and the clinical outcome of BOS.

Although observations in this study suggest that peripheral microchimerism does not correlate dramatically with clinical outcomes in our LTR cohort, minor influences could be detected using a larger cohort of patients. This interim study focused on a relatively small cohort of 20 non-HLA-B44 recipients with HLA-B44 allografts, but this could be extended to encompass other common HLA mismatches, including A1, A2, B7 and B8 [27]. If additional HLA mismatches were incorporated, microchimerism could be examined in approximately 80–90% of LTR transplanted in a given year at The Alfred Hospital. This would increase dramatically the statistical power of the study. In addition, given that patient lung function was measured for 24 months and that BOS generally develops 2–5 years post-transplantation, extension of the longitudinal study to 5 years would assist in clarifying the contribution of sustained microchimerism and the risk of developing BOS.

Using HLA-B44 mismatched q-PCR we have shown that peripheral blood microchimerism is a common phenomenon following lung transplantation, with differential dynamics between cellular subset populations observed in individual patients. Despite a limited cohort number, we found that there was no dramatic effect of microchimerism on lung transplant clinical outcomes in the first post-operative year.

Acknowledgements

The authors thank the generous support of Professor Greg Snell and all clinicians, nurses and allied health professionals associated with the Lung Transplant Service at The Alfred Hospital and all lung transplant patients involved with this study. Louise C. Rowntree is a recipient of a Monash University Jubilee Scholarship.

Disclosure

The authors have no financial conflicts of interest.

Ancillary