Advertisement
The Journal of Heart and Lung Transplantation
International Society for Heart and Lung Transplantation.

The human cytomegalovirus-specific and UL40-mediated imprint in the natural killer cell repertoire is associated with antibody-mediated rejection in lung transplant recipients

Open AccessPublished:October 21, 2022DOI:https://doi.org/10.1016/j.healun.2022.10.014

      Background

      CD16+ natural killer (NK-) cells play, together with donor-specific antibodies (DSA) and via antibody-dependent cellular cytotoxicity (ADCC), an important role in the pathogenesis of antibody-mediated rejection (ABMR) in lung-transplant recipients (LTRs). Cytotoxic CD16+NKG2C+ NK cells proliferate in response to human Cytomegalovirus (HCMV) infections via the presentation of HCMV-encoded and highly polymorphic UL40 peptides. In our study, we aimed to clarify whether infections with HCMV-strains carrying different UL40 peptide variants are associated with the shift of the NK cell repertoire and the development of ABMR in LTRs.

      Methods

      We included 30 DSA+ABMR+, 30 DSA+ABMR and 90 DSAABMR LTRs. In all patients, 1 episode of high-level HCMV-replication occurred. In all DSA+ABMR+ LTRs, HCMV-replication occurred prior to ABMR diagnosis. The association of HCMV UL40 variants with the expansion of CD16+ NK cell subsets and ABMR was assessed in NK cell proliferation and ADCC assays.

      Results

      Our study revealed that the VMAPRTLIL and VMTPRTLVL UL40 variants were significantly overrepresented in DSA+ABMR+ LTRs. Both peptides were associated with a pronounced proliferation of cytotoxic and proinflammatory CD16+NKG2C+ NK cells. The stimulation with both peptides led to a shift of the NK cell repertoire towards CD16+NKG2C+ NK cells, which was associated with strong ADCC responses after stimulation with endothelial cells and plasma from DSA+ABMR+ LTRs.

      Conclusions

      Distinct UL40 peptide variants of the infecting HCMV-strain are associated with the development of ABMR after lung transplantation, due to a shift towards a highly cytotoxic CD16+NKG2C+ NK cell population. These peptides are thus potential prognostic markers for ABMR.

      KEYWORDS

      Abbreviations:

      ABMR (antibody mediated rejection), ADCC (antibody-dependent cellular cytotoxicity), BAL (bronchoalveolar lavage), CLAD (chronic lung allograft dysfunction), CT (computed tomography), D(+) (donor (HCMV-seropositive)), DSA (Donor-specific antibody), FcγR (Fcγ-receptor), FEV (Forced Expiratory Volume), HCMV (human cytomegalovirus), HLA (human leukocyte antigen), IFNγ (interferon-γ), KTR (kidney transplant recipients), LTR (lung-transplant recipient), MFI (mean fluorescence intensity), MICA/B (MHC class I chain-related protein A/B), NK cell (natural killer cell), NPV (negative predicted value), PBMC (peripheral blood mononuclear cell), PCR (polymerase chain reaction), PPV (positive predictive value), R(-) (recipient (HCMV-seronegative)), RM ANOVA (Repeated measures analysis of variance), TNFα (tumor necrosis factor α)
      Natural Killer (NK) cells are innate lymphocytes, which are increasingly recognized as an important factor for allograft tolerance in lung-transplant recipients (LTRs).
      • Calabrese DR
      • Lanier LL
      • Greenland JR
      Natural killer cells in lung transplantation.
      NK cells respond to non-self-cells by the direct lysis and the production and secretion of proinflammatory effector molecules, such as interferon-γ (IFNγ) and tumor necrosis factor α (TNFα). NK cells are important effector cells in the antibody-mediated rejection (ABMR), which presents a significant challenge for long-term graft survival in LTRs.
      • Muduma G
      • Odeyemi I
      • Smith-Palmer J
      • Pollock RF
      Review of the clinical and economic burden of antibody-mediated rejection in renal transplant recipients.
      ,
      • Monteiro CMF
      • Walton DC
      • Cristiano Y
      • et al.
      Natural killer (NK) cells: the missing link between antibody-mediated rejection (ABMR) and chronic lung allograft dysfunction (CLAD)?.
      ABMR can arise from preexisting or de novo donor-specific antibodies (DSA), which bind to donor derived HLA, MICA/MICB or lung-parenchymal antigens.
      • Calabrese DR
      • Lanier LL
      • Greenland JR
      Natural killer cells in lung transplantation.
      ,
      • Hachem RR
      • Kamoun M
      • Budev MM
      • et al.
      Human leukocyte antigens antibodies after lung transplantation: primary results of the HALT study.
      DSA may bind to vascular endothelial cells in the allograft and mediate tissue injury via the activation of Fcγ-receptor (FcγR) expressing effector cells.
      • Muduma G
      • Odeyemi I
      • Smith-Palmer J
      • Pollock RF
      Review of the clinical and economic burden of antibody-mediated rejection in renal transplant recipients.
      ,
      • Rajalingam R
      The impact of HLA class I-specific killer cell immunoglobulin-like receptors on antibody-dependent natural killer cell-mediated cytotoxicity and organ allograft rejection.
      Especially mature and blood resident NK cells express high levels of the FcγR CD16a, which mediates the antibody-dependent activation of NK cells by a process termed antibody-dependent cellular cytotoxicity (ADCC).
      Human Cytomegalovirus (HCMV) may cause severe and potentially life-threatening infections in LTRs. HCMV-replication in the lung-allograft was recently associated with release of proinflammatory chemokines and increased risk of graft-rejection.
      • Streblow DN
      • Kreklywich C
      • Yin Q
      • et al.
      Cytomegalovirus-mediated upregulation of chemokine expression correlates with the acceleration of chronic rejection in rat heart transplants.
      ,
      • Streblow DN
      • Orloff SL
      • Nelson JA
      Acceleration of allograft failure by cytomegalovirus.
      While the direct effects of HCMV-infections on allograft tolerance was repeatedly demonstrated,
      • Toupance O
      • Bouedjoro-Camus MC
      • Carquin J
      • et al.
      Cytomegalovirus-related disease and risk of acute rejection in renal transplant recipients: a cohort study with case-control analyses.
      persistent infection with HCMV additionally leads to a characteristic imprint in the human NK cell repertoire.
      • Picarda G
      • Benedict CA
      Cytomegalovirus: shape-shifting the immune system.
      In response to HCMV-replication, a specialized subset of NKG2C+, and to a lower extent also NKG2A+ NK cells expand and persist in immunocompetent individuals and immunocompromised solid organ recipients.
      • Redondo-Pachon D
      • Crespo M
      • Yelamos J
      • et al.
      Adaptive NKG2C+ NK Cell response and the risk of cytomegalovirus infection in kidney transplant recipients.
      NKG2C and NKG2A, both together with CD94, bind to the nonclassical MHC class I molecule HLA-E, which is on the surface of HCMV-infected cells complexed and stabilized with the HCMV-encoded UL40 peptide.
      • Hammer Q
      • Ruckert T
      • Borst EM
      • et al.
      Peptide-specific recognition of human cytomegalovirus strains controls adaptive natural killer cells.
      The HCMV-UL40 peptide is highly polymorphic and UL40 variants bind with different binding affinities to HLA-E molecules, thereby affecting the level of NKG2C+ and NKG2A+ NK cell proliferation.
      • Hammer Q
      • Ruckert T
      • Borst EM
      • et al.
      Peptide-specific recognition of human cytomegalovirus strains controls adaptive natural killer cells.
      • Heatley SL
      • Pietra G
      • Lin J
      • et al.
      Polymorphism in human cytomegalovirus UL40 impacts on recognition of human leukocyte antigen-E (HLA-E) by natural killer cells.
      • Vales-Gomez M
      • Reyburn HT
      • Erskine RA
      • Lopez-Botet M
      • Strominger JL
      Kinetics and peptide dependency of the binding of the inhibitory NK receptor CD94/NKG2-A and the activating receptor CD94/NKG2-C to HLA-E.
      The expansion of NKG2C+ NK cells was so far mainly observed in HCMV-infected individuals and in response to HCMV-infected cells.
      • Guma M
      • Angulo A
      • Vilches C
      • Gomez-Lozano N
      • Malats N
      • Lopez-Botet M
      Imprint of human cytomegalovirus infection on the NK cell receptor repertoire.
      ,
      • Guma M
      • Budt M
      • Saez A
      • et al.
      Expansion of CD94/NKG2C+ NK cells in response to human cytomegalovirus-infected fibroblasts.
      After primary infection, HCMV establishes a life-long persistence, from which sporadic reactivations and re-infections occur. The consequential repeated stimulation of the immune system results in in a shift of the human NK cell repertoire, hallmarked by the accumulation of long-living NKG2C+ and NKG2A+ NK cells.
      • Picarda G
      • Benedict CA
      Cytomegalovirus: shape-shifting the immune system.
      ,
      • Hammer Q
      • Rückert T
      • Romagnani C
      Natural killer cell specificity for viral infections.
      The overall long-term consequences of such an HCMV-specific imprint in the NK cell repertoire for graft survival after lung transplantation, and in particular for ABMR are so far, however, unknown. The main aim of the present study was thus to reveal, whether there is an association between the HCMV-UL40 variants and the risk for ABMR and to which extent the HCMV-UL40 mediated expansion of CD16+NKG2C+ and CD16+NKG2A+ NK cells increase the risk for ABMR in LTRs.

      Material and methods

      Patients and samples

      In our study, 150 patients were included, 30 DSA positive (DSA+) patients with AMBR and 120 matched controls. The controls were matched to the ABMR group in regard of age at transplantation, gender, D/R serostatus, as well as time, compartment and duration of the episode of high-level HCMV replication using case-control matching (Supplementary Methods, Table 1).
      Table 1Characteristics of the Study Cohort
      Study cohort
      DSA+ABMR+ (N = 30)DSA+ABMR (N = 30)DSAABMR (N = 90)p-value
      Differences between groups were assessed with ANOVA or χ2-test.
      Female gender (%)N = 18 (40%)N = 18 (40%)N = 56 (40%)0.99
      Median age (range)52.0 (19.1 - 66.3)50.1 (18.5 - 66.4)53.2 (18.4 - 71.8)0.97
      D/R serostatus
       D+/R- (%)N=8 (26.7%)N=8 (26.7%)N=24 (26.7%)0.99
       D-/R+ (%)N=7 (23.3%)N=7 (23.3%)N=21 (23.3%)
       D+/R+ (%)N=15 (50.0%)N=15 (50.0%)N=45 (50.0%)
      HCMV-episode (median days)

      post-transplantation (range)


      265 (101-485)


      264 (97-484)


      241 (119-563)


      0.98
      Compartment of the HCMV Episode

      Blood (%)

      Lung (%)


      N=20 (66.6%)

      N=10 (13.3%)


      N=20 (66.6%)

      N=10 (33.3%)


      N=60 (66.6%)

      N=30 (33.3%)


      0.99
      Median Viral Load (copies/mL)

      during HCMV-episode (range)

      Blood

      Lung


      3.1 × 104 (2.6 × 103 - 4.9 × 104)

      2.3 × 104 (3.1 × 103 - 2.7 × 104)


      2.8 × 104 (2.6 × 103 - 4.9 × 104)

      1.9 × 104 (2.9 × 103 - 2.4 × 104)


      3.8 × 104 (2.4 × 103 - 4.6 × 104)

      1.8 × 104 (2.3 × 103 - 2.5 × 104)


      0.91
      ISHLT clinical ABMR grade
       Definite (%)2 (6.7%)0 (0%)0 (0%)
       Allograft dysfunction (FEV1% ≤80%) (%)2 (100%)
       Other causes excluded (%)2 (100%)
       Lung histology (%)2 (100%)
       Lung biopsy C4d (%)2 (100%)
       DSA (%)2 (100%)
       Probable (%)13 (43.3%)0 (0%)0 (0%)
       Allograft dysfunction (FEV1% ≤80%) (%)13 (100%)
       Other causes excluded (%)7 (53.8%)
       Lung histology (%)8 (61.5%)
       Lung biopsy C4d (%)11 (84.6%)
       DSA (%)13 (100%)
       Possible (%)15 (50.0%)0 (0%)0 (0%)
       Allograft dysfunction (FEV1% ≤80%) (%)15 (100%)
       Other causes excluded (%)0 (0%)
       Lung histology (%)3 (20%)
       Lung biopsy C4d (%)

      DSA (%)
      12 (80%)

      15 (100%)
      ISHLT sub-clinical ABMR grade
       Definite (%)0 (0%)0 (0%)0 (0%)
       Lung histology (%)

      Lung biopsy C4d (%)

      DSA (%)
       Probable (%)0 (0%)2 (13.4%)0 (0%)
       Lung histology (%)1 (50%)
       Lung biopsy C4d (%)1 (50%)
       DSA (%)2 (100%)
       Possible (%)0 (0%)28 (93.3%)0 (0%)
       Lung histology (%)0 (0%)
       Lung biopsy C4d (%)0 (0%)
       DSA (%)28 (100%)
      First detection of DSA (MFI>1,000) (median days) post-transplantation (range)305 (25-480)318 (32-450)0.91
      Clinical ABMR diagnosis (median days) post-transplantation (range)330 (189-550)
      DSA
      Pre-existing DSA (%)0 (0%)0 (0%)0 (0%)0.99
      De-novo HLA I-specific DSA (%)24 (80%)24 (80%)0 (0%)
      De-novo HLA II-specific DSA (%)30 (100%)30 (100%)0 (0%)
      De-novo HLA I-specific DSA, median MFI (range)4.2 × 104 (1.4 × 103 - 1.1 × 104)3.9 × 104 (1.3 × 103 - 9.3 × 103)0.89
      De-novo HLA II-specific DSA median MFI (range)2.1 × 104 (1.9 × 103 - 2.4 × 104)2.0 × 104 (2.1 × 103 - 2.7 × 104)
      Median absolute leukocyte counts (per mL blood) at diagnosis of ABMR or matched time point (range)4.7 × 103 (1.2 × 103 – 1.1 × 104)5.6 × 103 (3.4 × 103 – 1.4 × 104)5.8 × 103 (2.4 × 103 – 1.4 × 104)0.08
      Median absolute lymphocyte counts (per mL blood) at diagnosis of ABMR or matched time point (range)8.8 × 102 (1.9 × 102 – 1.9 × 103)1.2 × 103 (3 × 102 – 4.2 × 103)1.2 × 103 (4 × 102 – 4.7 × 103)0.07
      ABMR, antibody mediated rejection; C4d, complement component 4; D+, HCMV- seropositive donor; D-, HCMV- seronegative donor; DSA, donor-specific antibody; FEV1, forced expiratory volume in 1 second; MFI, mean fluorescence intensity; R+, HCMV- seropositive recipient; R-, HCMV- seronegative recipient.
      a Differences between groups were assessed with ANOVA or χ2-test.
      All patients were either HCMV-seropositive (R+) or received the organ from an HCMV-seropositive donor (D+/R-). The patients received induction therapy with 30mg Alemtuzumab (Berlex), and the maintenance therapy regimen included tacrolimus, corticosteroids and mycophenolate mofetil. HCMV-Immunoglobulin (Cytotect, 100 units/kg) was administered once weekly for 4 weeks posttransplantation and all patients received antiviral (Val-)Ganciclovir prophylaxis, R+ patients for 3, D+/R- patients for 12 months. All patients were followed-up by quantitative HCMV-PCR, weekly for 2 months, monthly to bimonthly for 1 year after transplantation and in larger intervals thereafter. HCMV-DNA levels of >1000 copies/ml plasma were preemptively treated with (Val-)Ganciclovir.
      The standard posttransplantation care was performed as recently described in detail.
      • Benazzo A
      • Worel N
      • Schwarz S
      • et al.
      Outcome of extracorporeal photopheresis as an add-on therapy for antibody-mediated rejection in lung transplant recipients.
      Surveillance bronchoscopy with a transbronchial biopsy and bronchoalveolar lavage, spirometry with body plethysmography were performed at 2 weeks and routinely thereafter. Computed tomography (CT) scan was performed each year and additionally in cases of lung function deterioration.
      • Benazzo A
      • Worel N
      • Schwarz S
      • et al.
      Outcome of extracorporeal photopheresis as an add-on therapy for antibody-mediated rejection in lung transplant recipients.
      ABMR was classified according to the most recent ISHLT consensus guidelines.
      • Levine DJ
      • Glanville AR
      • Aboyoun C
      • et al.
      Antibody-mediated rejection of the lung: a consensus report of the International Society for Heart and Lung Transplantation.
      LTRs were routinely tested for de novo DSA at 2 weeks and 1,2,3,6,9,12,18, and 24 months after transplantation, and in cases of lung function deterioration, using a bead array technique as recently described.
      • Benazzo A
      • Worel N
      • Schwarz S
      • et al.
      Outcome of extracorporeal photopheresis as an add-on therapy for antibody-mediated rejection in lung transplant recipients.
      Reactions with an MFI >1,000 were considered positive.

      NK cell assays

      Details of the antibody-dependent ABMR assays, NK cell proliferation assays and antibody-dependent ADCC assays are presented in detail in the supplementary material & methods section. For the ABMR assays, NK cells were cocultured together with the patient's plasma and human umbilical vein endothelial cells (HUVEC), followed by flow-cytometry analysis. For the NK cell proliferation assays, CFSE-stained NK cells were co-cultured together with UL40 peptide pulsed antigen-presenting RMA-S/HLA-E/LFA-3 cells, followed by analysis of proliferating and non-proliferating NK cell subsets. For the ADCC assays, NK cells were stimulated with UL40 peptides pulsed RMA-S/HLA-E/LFA-3 cells and proliferating and non-proliferating NK cells were subsequently cocultured either with Rituximab-coated Raji cells or with patient's plasma and HUVEC cells. NK cell activation was subsequently analyzed by flow-cytometry or LDH release assays.

      Statistical analysis

      The frequencies of the UL40 variants between the study cohorts were compared using the χ2-test. Repeated measures 1-way analysis of variance (RM ANOVA, with the Geisser-Greenhouse correction) and Tukey-post test (RM ANOVA: p < 0.05) were used to compare the means of proliferating and activated NK cell subsets between the cohorts. The time between the HCMV-episode and ABMR was compared between the UL40 variants using the Mantel-Cox test. The study was approved by the institutional review board of the Medical University of Vienna (EK-No.1687/2018).

      Results

      Characteristics of the study cohort

      The study cohort consisted of 30 patients who developed de novo DSA and clinical ABMR (DSA+ABMR+) within 2 years-post transplantation. In all DSA+ABMR+ LTRs, 1 episode of high-level HCMV-replication (>1000 copies/mL) occurred either in blood (N = 20) or in the bronchoalveolar lavage (BAL,N = 10), in median 130 (88-231) days prior to ABMR diagnosis. The median duration of these viral episodes was 14 (10-31) days.
      As matched controls 30 LTRs who developed de novo DSA but had no clinical, but sub-clinical ABMR (DSA+ABMR), and 90 patients who had neither DSA nor ABMR (DSAABMR) within 2 years posttransplantation were selected. In contrast to DSA+ABMR+ LTRs, DSA+ABMR and DSAABMR had no allograft dysfunction, as defined by a FEV1% ≤80% at baseline for ≥3 weeks, at the diagnosis of ABMR.
      From each patient and control, samples were available at 2 time points in the follow-up: The first plasma or BAL sample was available during the peak of the high-level HCMV-replication. The second sample was plasma, which was available after the diagnosis of ABMR in DSA+ABMR+ LTRs, before the patients underwent plasmapheresis or extracorporeal immunoadsorption for the elimination for DSA, or at a matched time point for DSA+ABMR or DSAABMR LTRs, respectively.

      NKG2C± NK cells are highly activated by DSA and HUVEC cells

      First, we aimed to analyze the contribution of CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells in the development of ABMR. Therefore, we first isolated CD56+CD16+ NK cells from 24 healthy, HCMV-seropositive blood donors and compared the CD16a expression levels between the NK cell subsets. As shown in Figure 1A, CD16+CD94+NKG2C+ showed significant higher CD16a expression levels, compared to CD16+CD94+NKG2A+ and CD16+CD94 NK cells. No different CD16a expression levels were however found between CD16+CD94+NKG2A+ and CD16+CD94 NK cells.
      Figure 1
      Figure 1Analysis of the NK cell responses of CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells, stimulated with plasma from DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs and endothelial HUVECs. (A) Expression analysis of CD16+ NK cells in CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells subsets. (B-G) Percent of activated NK cells (B,D,F) and expression levels of activated NK cells (C,E,G) of CD107a (B,C), IFNγ (D,E), or TNFα (F,G) positive CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells subsets. (A-G). The fold change was calculated using respective NK cell subsets, cultured with plasma of DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs, but without endothelial HUVEC. Data are shown as the mean fold changes of 24 biological replicates of 24 HCMV-seropositive healthy blood donors, ± SD. RM 1-way ANOVA (with the Geisser-Greenhouse correction) and Tukey-post Test was used for statistical comparison. ABMR: antibody-mediated rejection, DSA: donor-specific antibody, HUVEC: human umbilical vein endothelial cells, IFNγ: interferon-γ, TNFα: tumor necrosis factor α. p < 0.05 was considered significant. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.
      We then tested the isolated CD16+ NK cells in a flow-cytometry based in vitro antibody-dependent ABMR assay, using HUVECs and plasma samples from DSA+ABMR+ patients after the first ABMR diagnosis or from DSAABMR, DSA+ABMR at a comparable time point (Figure. S1). As shown in Figure 1B-G, stimulation with plasma samples from DSA+ABMR and DSA+ABMR+ LTRs lead to a significant activation of CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells, compared to DSAABMR LTRs. This was reflected by significantly higher percentages and expression levels of the cytotoxicity marker CD107a (Figure. 1B-C) as well as the pro-inflammatory cytokines IFNγ (Figure. 1D-E) and TNFα (Figure. 1F-G).
      We then also compared the antibody-dependent activation levels between CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells. When stimulated with plasma from DSA+ABMR or DSA+ABMR+ LTRs, CD16+CD94+NKG2C+ NK cells showed generally a significantly higher activation level for all tested markers compared to CD16+CD94+NKG2A+ and CD16+CD94 NK cells (all: p < 0.001,ANOVA and Dunn's Test). In contrast, we observed no significant difference between CD16+CD94+NKG2A+ and CD16+CD94 NK cells (all:ns, ANOVA and Dunn's Test). In summary, this shows that especially CD16+CD94+NKG2C+ cells are highly activated in DSA+ LTRs and thus may be the main CD16+ NK cell subset, which contributes to tissue injury in DSA+ LTRs during ABMR.

      HCMV-UL40 variants are associated with the development of ABMR

      As the expansion of CD16+CD94+NKG2C+, and to a lower extent also CD16+CD94+NKG2A+ NK cells depends on HCMV-encoded UL40 peptides,
      • Hammer Q
      • Ruckert T
      • Borst EM
      • et al.
      Peptide-specific recognition of human cytomegalovirus strains controls adaptive natural killer cells.
      we next analyzed the UL40 variants of the infecting HCMV-strains of the DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs by Sanger-sequencing. Overall, 5 different UL40 variants dominated in the study cohort: VMAPRTLIL (N = 48,32%), VMAPRTLLL (N = 38,25.3%), VMTPRTLIL (N = 26,17.3%), VMTPRTLVL (N = 12,8%) and VMAPRTLVL (N = 8,5.3%). Other UL40 variants were detected each only in 1 or 2 LTRs and were consequently summarized as “unusual” variants (N = 20,13.3%).
      We then compared the distribution of the UL40 variants between DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs and found a different pattern between DSA+ABMR+ LTRs and the controls (Figure 2, Figure S2). While DSAABMR and DSA+ABMR LTRs showed overall a similar UL40 variant diversity, in DSA+ABMR+ LTRs 2 variants, VMAPRTLIL (43.3%) and VMTPRTLVL (40%), dominated. Especially the VMTPRTLVL variant was solely found in patients with ABMR (PPV:1, NPV:0.4), while VMTPRTLIL as well as “unusual” variants occurred only in DSAABMR and DSA+ABMR LTRs (both: p < 0.0001, F-Test).
      Figure 2
      Figure 2Analysis of HCMV UL40 variants occurring in lung transplant recipients (LTRs) with or without both DSA and ABMR, respectively. Overall distribution of UL40 variants occurring in LTRs during 1 episode of high-level HCMV load (>1000 copies/mL) in BAL or in blood. Bars represent the relative frequency of UL40 variants. χ2 -test was used for statistical comparison between the study cohort. ABMR: antibody-mediated rejection, DSA: donor-specific antibody. p < 0.05 was considered significant. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.
      We then also compared the time between the occurrence of a high-viral load and ABMR between the UL40 variants. No significant differences were found (all: p > 0.05, Mantel-Cox test).

      HCMV-UL40 variants are associated with the shape of the NK cell repertoire

      We then tested to which extent the 5 major UL40 variants (VMAPRTLIL,VMAPRTLLL,VMTPRTLIL,VMTPRTLVL,VMAPRTLVL) were associated with a shift of the human CD16+ NK cell repertoire towards highly cytotoxic CD16+CD94+NKG2C+ NK cells. Therefore, we cocultured CD16+ NK cells with respective UL40-peptide pulsed antigen presenting cells or HCMV-infected HUVEC cells for 14 days. After 7 and 14 days, respectively, we assessed the proliferating and non-proliferating NK cell subsets by flow-cytometry (Figure S3). UL40 peptides (Figure 3) and HCMV-infected cells (Figure S4) elicited a significant proliferation of CD16+CD94+NKG2C+ and CD16+CD94+NKG2A+ NK cells. In contrast, the proliferation of CD16+CD94 NK cells dominated in response to antigen presenting cells without peptide or noninfected HUVEC (Figure 3, Figure S4).
      Figure 3
      Figure 3Analysis of the proliferating and total CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cell subsets, stimulated either with RMA-S/HLA-E/LFA-3 alone or with RMA-S/HLA-E/LFA-3 together with the VMAPRTLLL, VMTPRTLIL, VMAPRTLVL, VMTPRTLVL or VMAPRTLIL UL40 variant, respectively. Proliferating and total CD16+CD94+NKG2C+ (purple fraction), CD16+CD94+NKG2A+ (blue fraction) and CD16+CD94 (red fraction) NK cell subsets were assessed after 7 and 14 days of proliferation by flow-cytometry. Fractions represent the relative frequency of means of respective CD16+ NK cell subsets of 12 biological replicates of 12 HCMV-seropositive healthy blood donors. RM 1-way ANOVA (with the Geisser-Greenhouse correction) and Tukey-post Test was used for statistical comparison of the NK cell subsets after 14 days of coculture with baseline NK cell subsets. p < 0.05 was considered significant. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.
      We then compared the percentages of the proliferating CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cell subsets between the different UL40 peptides. We found that especially the VMAPRTLIL and VMTPRTLVL peptides were associated with the proliferation of CD16+NKG2C+ NK cells after 7 and 14 days of coculture, respectively. Compared to the VMAPRTLLL,VMTPRTLIL or VMAPRTLVL, stimulation with the VMAPRTLIL or VMTPRTLVL UL40 peptides resulted in a significantly higher percentage of CD16+CD94+NKG2C+ after 14 days of co-culture (all:p < 0.01, ANOVA and Dunn's Test).

      Proliferating NKG2C± NK cells are highly potent mediators of ADCC

      To test whether the proliferating and nonproliferating CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells are active and capable to elicit ADCC after 14 days of expansion, we tested the NK Cells in an in vitro flow-cytometry based ADCC assay, using Rituximab-coated Raji cells and the cytotoxicity marker CD107a. As shown in Figure 4A-D, all proliferating and nonproliferating CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells were capable of ADCC, as reflected by the Rituximab concentration-dependent increase of CD107a in all tested CD16+ NK cell subsets.
      Figure 4
      Figure 4Analysis of the ADCC and ABMR responses of CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cell subsets. (A-D) Analysis of the ADCC response in proliferating (A,C) and non-proliferating (B,D) CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cell subsets, stimulated either with the VMAPRTLLL, VMTPRTLIL, VMAPRTLVL, VMTPRTLVL or VMAPRTLIL UL40 variant, respectively, and tested against Rituximab-coated (0.2 - 0.0002 μg/ml) Raji cells. (A,C) Percent of activated, CD107a positive NK cells and (B,D) expression levels of activated NK cells of CD107a positive CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cells subsets. (E-F) Analysis of the ADCC response of NK cells, stimulated either with the VMAPRTLLL, VMTPRTLIL, VMAPRTLVL, VMTPRTLVL or VMAPRTLIL UL40 variant, respectively, and tested against endothelial HUVEC and plasma from DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs. (E) Absolute LDH release in response to NK cells, expanded for 14 days with respective UL40 peptides and tested against endothelial HUVEC and plasma from DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs. (F) Relative LDH release in response to NK cells, expanded for 14 days with respective UL40 peptides and tested against endothelial HUVEC and plasma from DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs. Fold change was calculated using NK cells, expanded without UL40 peptides. (A-F) Data are shown as the mean fold changes of 12 biological replicates of 12 HCMV-seropositive healthy blood donors, stimulated with each of the respective UL40 variants ± SD. RM 1-way ANOVA (with the Geisser-Greenhouse correction) and Tukey-post Test was used for statistical comparison. ABMR: antibody-mediated rejection, DSA: donor-specific antibody, LDH: Lactate dehydrogenase, w/o: without. p < 0.05 was considered significant. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.
      We then also compared the percentage of activated NK cells as well as the expression levels of CD107a between the proliferating and non-proliferating CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 subsets and found, that CD16+CD94+NKG2C+ showed a significantly higher degree of activation, compared to CD16+CD94+NKG2A+ and CD16+CD94 NK cells. No significant differences were however found between CD16+CD94+NKG2A+ and CD16+CD94 NK cells (Figure 4A-D).
      Next, we compared the activation level between proliferating and non-proliferating CD16+CD94+NKG2C+, CD16+CD94+NKG2A+ and CD16+CD94 NK cell subsets, respectively. All proliferating NK cell subsets showed generally a higher degree of activation, compared to non-proliferating cells (all:p < 0.01,ANOVA and Dunn's Test).

      HCMV UL40-mediated shape of the NK cell repertoire is associated with ABMR

      We then also tested whether the HCMV-UL40 mediated shift of the NK cell repertoire towards highly potent CD16+CD94+NKG2C+ NK cells result in an increased ADCC response. Therefore, we expanded NK cells with the UL40 peptides for 14 days and cocultured the NK cells then with HUVEC cells and the plasma samples of DSAABMR, DSA+ABMR and DSA+ABMR+ LTRs. As shown in Figure 4E only plasma from DSA+ABMR and DSA+ABMR+ LTRs, together with expanded NK cells, lead to a significant lysis of HUVEC cells, as reflected by a higher LDH release. Importantly, no significant differences were found in the lysis of HUVEC cells between plasma from DSA+ABMR and DSA+ABMR+ LTRs.
      We finally compared the level of lysis of HUVECs induced by NK cells, which were stimulated with the different UL40 peptides. As shown in Figure 4F, CD16+ NK cells, stimulated with the VMAPRTLIL and VMTPRTLVL variant lead to a significantly higher antibody-mediated lysis of HUVEC cells in the presence of patient plasma, compared to NK cells, stimulated with the VMAPRTLLL,VMTPRTLIL or VMAPRTLVL variant (all:p < 0.01,ANOVA and Dunn's Test).

      Discussion

      In the present study, we demonstrate that the specific UL40 variant of the infecting HCMV-strain and the resulting shaping of the human NK cell repertoire, especially towards CD16+NKG2C+ NK cells, play an important role in the antibody-mediated allograft injury during ABMR after lung transplantation.
      So far, it remained an open question why of all DSA+ LTRs only some developed ABMR, while other DSA+ patients show no evidence of humoral rejection over time. A recently published study in kidney transplant recipient (KTRs) found that the presence of DSA alone is not a risk factor for rejection or graft failure.
      • Parajuli S
      • Joachim E
      • Alagusundaramoorthy S
      • et al.
      Donor-specific antibodies in the absence of rejection are not a risk factor for allograft failure.
      In the present study, we revealed that all DSA+ patients, independent of whether they developed ABMR or not, show a comparable ADCC response against endothelial cells, when tested in the in vitro antibody-dependent ABMR assay. However, our study revealed that the UL40 peptide variant, present in the HCMV-strain infecting the patients, plays a significant role in the development of AMBR and was signifantly different between DSA+AMBR+ and DSA+AMBR patients. It has previously been shown that the highly polymorphic UL40 peptide occurs in different variants in HCMV-strains infecting LTRs.
      • Vietzen H
      • Rückert T
      • Hartenberger S
      • et al.
      Extent of cytomegalovirus replication in the human host depends on variations of the HLA-E/UL40 axis.
      In the present patient population, we identified in overall 18 different UL40 variants in the infecting HCMV-strains, detected during an episode of high-level virus replication in blood or lung. Of these however, 2 HCMV-encoded UL40 peptides, VMAPRTLIL and VMTPRTLVL, were especially prevalent, and the peptide VMTPRTLVL was even detected exclusively in DSA+ patients, developing ABMR in the follow-up. Our data are of special interest, as predictive factors for ABMR in de novo DSA+ LTRs are still scarce.
      • Kim MY
      • Brennan DC
      Therapies for chronic allograft rejection.
      In our patient population, the occurrence of the VMTPRTLVL variant, detected in an episode of virus replication was reliably associated with the subsequent occurrence of ABMR in LTRs.
      As a potential mechanism behind the association between the distinct UL40 variants and the development of ABMR, we revealed that the VMAPRTLIL and VMTPRTLVL variants are associated with a strong shift towards highly potent and proinflammatory CD16+NKG2C+ NK cells. This shift, demonstrated by NK cell proliferation assays, further results in an overall increase in the NK cell-mediated ADCC response, as revealed in the DSA-dependent ABMR assays. Our data thus demonstrate that HCMV-infections with VMAPRTLIL- and VMTPRTLVL-encoding strains lead to an increase in the proliferation of CD16+NKG2C+ NK cells, which then may mediate ABMR via an overall enhanced ADCC response.
      Different studies associated so far the occurrence of ABMR to the level of complement binding of the DSA in the single patients, as measured by C1q fixation or C3d-binding activity.
      • Nocera A
      • Tagliamacco A
      • Cioni M
      • et al.
      Kidney intragraft homing of De Novo donor-specific HLA antibodies is an essential step of antibody-mediated damage but not per se predictive of graft ;oss.
      • Sicard A
      • Ducreux S
      • Rabeyrin M
      • et al.
      Detection of C3d-binding donor-specific anti-HLA antibodies at diagnosis of humoral rejection predicts renal graft loss.
      • Loupy A
      • Lefaucheur C
      • Vernerey D
      • et al.
      Complement-binding anti-HLA antibodies and kidney-allograft survival.
      However, although the role of the complement in antibody-mediated allograft injury was earlier demonstrated also in LTRs,
      • Ali HA
      • Pavlisko EN
      • Snyder LD
      • Frank M
      • Palmer SM
      Complement system in lung transplantation.
      a recently published study indicated that it is the complement-independent ABMR, which represents the majority of humoral rejections in LTRs,
      • Aguilar PR
      • Carpenter D
      • Ritter J
      • et al.
      The role of C4d deposition in the diagnosis of antibody-mediated rejection after lung transplantation.
      that highlights the role of FcγR-expressing effector cells, such as NK cells. The overall importance of NK cells for the development of ABMR was also recently demonstrated in KTRs, in whom the level of intra-graft NK cell infiltrates and transcripts strongly predicted ABMR and graft survival.
      • Yazdani S
      • Callemeyn J
      • Gazut S
      • et al.
      Natural killer cell infiltration is discriminative for antibody-mediated rejection and predicts outcome after kidney transplantation.
      Another ex vivo study in KTRs demonstrated that especially ADCC-associated genes are highly upregulated in both, C4d-positive and negative ABMR patients,
      • Suviolahti E
      • Ge S
      • Nast CC
      • et al.
      Genes associated with antibody-dependent cell activation are overexpressed in renal biopsies from patients with antibody-mediated rejection.
      These studies are in line with our present data, and all together demonstrate an important role of the antibody-mediated activation of NK cells during humoral rejection.
      We showed that especially the shift towards the CD16+NKG2C+ NK cell population was important for ABMR. In our study, we demonstrated that CD16+NKG2C+ NK cells did not only provide cytotoxic potential but also secreted high levels of IFNγ and TNFα. These findings are of special interest, as recently published transcriptomic studies in KTRs with de novo DSA demonstrated that ABMR and allograft loss is hallmarked by IFNγ-inducible transcripts, such as the pro-inflammatory CXCL10 and CXCL11,
      • Aubert O
      • Loupy A
      • Hidalgo L
      • et al.
      Antibody-mediated rejection due to preexisting versus De Novo donor-specific antibodies in kidney allograft Recipients.
      ,
      • Venner JM
      • Hidalgo LG
      • Famulski KS
      • Chang J
      • Halloran PF
      The molecular landscape of antibody-mediated kidney transplant rejection: evidence for NK involvement through CD16a Fc receptors.
      It was consequently hypothesized that IFNγ may play a major role in the development and pathogenesis of ABMR. The effects of IFNγ include the upregulation of HLA and the activation of pro-inflammatory macrophages, both possibly facilitating the development of ABMR. TNFα is known as an important factor for the expression of adhesion molecules on endothelial cells, which facilitates the migration of lymphocytes into the allograft. It is thus reasonable that a strong CD16+NKG2C+ NK cell-mediated cytokine response plays an important role in the pathogenesis of ABMR, by inducing downstream donor-specific innate and adaptive immune responses.
      In summary, the present study reveals that distinct variants of the HCMV UL40 peptide present in the infecting HCMV-strain are associated with the following development of ABMR in LTRs, and that this is caused by a shift of the human NK cell repertoire towards a highly cytotoxic CD16+NKG2C+ NK cell population. Further studies are needed to evaluate whether the identification of the HCMV UL40 peptide may serve as a potential prognostic marker for ABMR and to further clarify the association of different HCMV-strains with the risk of allograft rejection and ABMR.

      Credit author statement

      Conceptualization: HV, EP; Methodology: HV, EP; Validation: HV; Formal analysis: HV; Investigation: HV; Resources: PJ, EP; Data Curation: HV, PJ; Writing Original Draft: HV, EP; Writing - Review & Editing: HV, EP; Visualization: HV; Supervision: EP.

      Disclosure statement

      The project was funded by the Startup Grant of the Research Platform Transplantation of the Medical University of Vienna.
      The authors declare that they have no conflict of interest.

      Appendix. Supplementary materials

      References

        • Calabrese DR
        • Lanier LL
        • Greenland JR
        Natural killer cells in lung transplantation.
        Thorax. 2018; 74: 397-404
        • Muduma G
        • Odeyemi I
        • Smith-Palmer J
        • Pollock RF
        Review of the clinical and economic burden of antibody-mediated rejection in renal transplant recipients.
        Adv Ther. 2016; 33: 345-356
        • Monteiro CMF
        • Walton DC
        • Cristiano Y
        • et al.
        Natural killer (NK) cells: the missing link between antibody-mediated rejection (ABMR) and chronic lung allograft dysfunction (CLAD)?.
        J Heart Lung Transplant. 2020; 39: S197
        • Hachem RR
        • Kamoun M
        • Budev MM
        • et al.
        Human leukocyte antigens antibodies after lung transplantation: primary results of the HALT study.
        Am J transplant. 2018; 18: 2285-2294
        • Rajalingam R
        The impact of HLA class I-specific killer cell immunoglobulin-like receptors on antibody-dependent natural killer cell-mediated cytotoxicity and organ allograft rejection.
        Front Immunol. 2016; 7: 585
        • Streblow DN
        • Kreklywich C
        • Yin Q
        • et al.
        Cytomegalovirus-mediated upregulation of chemokine expression correlates with the acceleration of chronic rejection in rat heart transplants.
        J Virol. 2003; 77: 2182-2194
        • Streblow DN
        • Orloff SL
        • Nelson JA
        Acceleration of allograft failure by cytomegalovirus.
        Curr Opin Immunol. 2007; 19: 577-582
        • Toupance O
        • Bouedjoro-Camus MC
        • Carquin J
        • et al.
        Cytomegalovirus-related disease and risk of acute rejection in renal transplant recipients: a cohort study with case-control analyses.
        Transplant Int. 2000; 13: 413-419
        • Picarda G
        • Benedict CA
        Cytomegalovirus: shape-shifting the immune system.
        J Immunol. 2018; 200: 3881-3889
        • Redondo-Pachon D
        • Crespo M
        • Yelamos J
        • et al.
        Adaptive NKG2C+ NK Cell response and the risk of cytomegalovirus infection in kidney transplant recipients.
        J Immunol. 2017; 198: 94-101
        • Hammer Q
        • Ruckert T
        • Borst EM
        • et al.
        Peptide-specific recognition of human cytomegalovirus strains controls adaptive natural killer cells.
        Nat Immunol. 2018; 19: 453-463
        • Heatley SL
        • Pietra G
        • Lin J
        • et al.
        Polymorphism in human cytomegalovirus UL40 impacts on recognition of human leukocyte antigen-E (HLA-E) by natural killer cells.
        J Biol Chem. 2013; 288: 8679-8690
        • Vales-Gomez M
        • Reyburn HT
        • Erskine RA
        • Lopez-Botet M
        • Strominger JL
        Kinetics and peptide dependency of the binding of the inhibitory NK receptor CD94/NKG2-A and the activating receptor CD94/NKG2-C to HLA-E.
        EMBO J. 1999; 18: 4250-4260
        • Guma M
        • Angulo A
        • Vilches C
        • Gomez-Lozano N
        • Malats N
        • Lopez-Botet M
        Imprint of human cytomegalovirus infection on the NK cell receptor repertoire.
        Blood. 2004; 104: 3664-3671
        • Guma M
        • Budt M
        • Saez A
        • et al.
        Expansion of CD94/NKG2C+ NK cells in response to human cytomegalovirus-infected fibroblasts.
        Blood. 2006; 107: 3624-3631
        • Hammer Q
        • Rückert T
        • Romagnani C
        Natural killer cell specificity for viral infections.
        Nat Immunol. 2018; 19: 800-808
        • Benazzo A
        • Worel N
        • Schwarz S
        • et al.
        Outcome of extracorporeal photopheresis as an add-on therapy for antibody-mediated rejection in lung transplant recipients.
        Transfusion Med Hemotherapy. 2020; 47: 205-213
        • Levine DJ
        • Glanville AR
        • Aboyoun C
        • et al.
        Antibody-mediated rejection of the lung: a consensus report of the International Society for Heart and Lung Transplantation.
        J Heart Lung Transplant. 2016; 35: 397-406
        • Parajuli S
        • Joachim E
        • Alagusundaramoorthy S
        • et al.
        Donor-specific antibodies in the absence of rejection are not a risk factor for allograft failure.
        Kidney Int Rep. 2019; 4: 1057-1065
        • Vietzen H
        • Rückert T
        • Hartenberger S
        • et al.
        Extent of cytomegalovirus replication in the human host depends on variations of the HLA-E/UL40 axis.
        mBio. 2021; 12: e02996-20
        • Kim MY
        • Brennan DC
        Therapies for chronic allograft rejection.
        Front Pharmacol. 2021; 12 (Available at:)
        • Nocera A
        • Tagliamacco A
        • Cioni M
        • et al.
        Kidney intragraft homing of De Novo donor-specific HLA antibodies is an essential step of antibody-mediated damage but not per se predictive of graft ;oss.
        Am J Transplant. 2017; 17: 692-702
        • Sicard A
        • Ducreux S
        • Rabeyrin M
        • et al.
        Detection of C3d-binding donor-specific anti-HLA antibodies at diagnosis of humoral rejection predicts renal graft loss.
        J Am Soc Nephrol. 2015; 26: 457-467
        • Loupy A
        • Lefaucheur C
        • Vernerey D
        • et al.
        Complement-binding anti-HLA antibodies and kidney-allograft survival.
        N Engl J Med. 2013; 369: 1215-1226
        • Ali HA
        • Pavlisko EN
        • Snyder LD
        • Frank M
        • Palmer SM
        Complement system in lung transplantation.
        Clin Transplant. 2018; 32: e13208
        • Aguilar PR
        • Carpenter D
        • Ritter J
        • et al.
        The role of C4d deposition in the diagnosis of antibody-mediated rejection after lung transplantation.
        Am J Transplant. 2018; 18: 936-944
        • Yazdani S
        • Callemeyn J
        • Gazut S
        • et al.
        Natural killer cell infiltration is discriminative for antibody-mediated rejection and predicts outcome after kidney transplantation.
        Kidney Int. 2019; 95: 188-198
        • Suviolahti E
        • Ge S
        • Nast CC
        • et al.
        Genes associated with antibody-dependent cell activation are overexpressed in renal biopsies from patients with antibody-mediated rejection.
        Transpl Immunol. 2015; 32: 9-17
        • Aubert O
        • Loupy A
        • Hidalgo L
        • et al.
        Antibody-mediated rejection due to preexisting versus De Novo donor-specific antibodies in kidney allograft Recipients.
        J Am Soc Nephrol. 2017; 28 (JASN): 1912-1923
        • Venner JM
        • Hidalgo LG
        • Famulski KS
        • Chang J
        • Halloran PF
        The molecular landscape of antibody-mediated kidney transplant rejection: evidence for NK involvement through CD16a Fc receptors.
        Am JTransplant. 2015; 15: 1336-1348