Polyomavirus BK-Specific Cellular Immune Response to VP1 and Large T-Antigen in Kidney Transplant Recipients
2007; Elsevier BV; Volume: 7; Issue: 5 Linguagem: Inglês
10.1111/j.1600-6143.2007.01754.x
ISSN1600-6143
AutoresS. Binggeli, Adrian Egli, Stefan Schaub, Isabelle Binet, Michael Mayr, J. Steiger, Hans H. Hirsch,
Tópico(s)Parvovirus B19 Infection Studies
ResumoAmerican Journal of TransplantationVolume 7, Issue 5 p. 1131-1139 Free Access Polyomavirus BK-Specific Cellular Immune Response to VP1 and Large T-Antigen in Kidney Transplant Recipients S. Binggeli, S. Binggeli Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, SwitzerlandSearch for more papers by this authorA. Egli, A. Egli Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, SwitzerlandSearch for more papers by this authorS. Schaub, S. Schaub Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorI. Binet, I. Binet Division of Nephrology, Kantonsspital St. Gallen, St. Gallen, SwitzerlandSearch for more papers by this authorM. Mayr, M. Mayr Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorJ. Steiger, J. Steiger Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorH. H. Hirsch, Corresponding Author H. H. Hirsch Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, Switzerland Infectious Diseases & Hospital Epidemiology, University Hospital Basel, Basel, Switzerland * Corresponding author: Hans H. Hirsch, hans.hirsch@unibas.chSearch for more papers by this author S. Binggeli, S. Binggeli Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, SwitzerlandSearch for more papers by this authorA. Egli, A. Egli Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, SwitzerlandSearch for more papers by this authorS. Schaub, S. Schaub Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorI. Binet, I. Binet Division of Nephrology, Kantonsspital St. Gallen, St. Gallen, SwitzerlandSearch for more papers by this authorM. Mayr, M. Mayr Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorJ. Steiger, J. Steiger Division of Transplantation Immunology & Nephrology, University Hospital Basel, Basel, SwitzerlandSearch for more papers by this authorH. H. Hirsch, Corresponding Author H. H. Hirsch Transplantation Virology, Institute for Medical Microbiology, University of Basel, Basel, Switzerland Infectious Diseases & Hospital Epidemiology, University Hospital Basel, Basel, Switzerland * Corresponding author: Hans H. Hirsch, hans.hirsch@unibas.chSearch for more papers by this author First published: 23 April 2007 https://doi.org/10.1111/j.1600-6143.2007.01754.xCitations: 150AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Polyomavirus BK (BKV) is the primary cause of polyomavirus-associated nephropathy (PVAN) in kidney transplant (KT) recipients. Using ELISpot assays, we compared the frequency of interferon-γ (IFN-γ) secreting peripheral blood mononuclear cells (PBMC) after stimulation with overlapping peptide pools covering BKV large T-antigen (LT) and VP1 capsid proteins (VP1). In 10 healthy donors, LT and VP1 responses were low with median 24 (range 15–95) and 25 (7–113) spot-forming units/106 PBMC (SFU), respectively. In 42 KT patients with current or recent plasma BKV loads, median LT and VP1 responses of 29 (0–524) and 114 (0–1432) SFU were detected, respectively. In KT patients with decreasing or past plasma BKV loads, significantly higher median BKV-specific IFN-γ responses were detected compared to KT patients with increasing or persisting BKV loads [LT: 78 (8–524) vs. 22 (0–120) SFU, p = 0.003; VP1: 285 (45–1432) vs. 53 (0–423) SFU, p = 0.001, respectively]. VP1-specific IFN-γ responses were higher and more likely to involve CD4+ T cells, while CD8+ T cells were more frequently directed against LT. Stimulation with JCV-specific VP1 and LT peptides indicated only low-level cross-recognition. The data suggest that control of BKV replication is correlated with differentiated expansion of BKV-specific cellular immune responses. Introduction Polyomavirus-associated nephropathy (PVAN) is currently the most challenging infectious cause of kidney transplant (KT) failure (1) affecting 1–10% of patients with graft loss in >50% (2-6). The human polyomavirus type 1, called BK virus (BKV) (7), is the primary etiologic agent, although few cases have been attributed to the closely related JC virus (JCV), the primary cause of progressive multifocal leucoencephalopathy (8, 9). BKV seroprevalence rates increase during childhood exceeding 90% in adults worldwide (10). BKV persists in a nonreplicative latent stage in the renourinary tract, and reactivation and low-level replication is seen intermittently in 5% of nonimmunosuppressed individuals at comparatively low levels of 107 cp/mL occurs in 20–60% of transplant and other immunocompromised patients, and may be followed by viremia in KT patients at risk for PVAN (13, 14). Although the risk factors of PVAN are not unequivocally defined and may include partially complementary determinants of patient, graft and virus (3), impaired BKV-specific antiviral immune control is viewed as key factor (15). Different studies have implicated BKV-seropositive donors, seronegative recipients, HLA-mismatching, HLA-C7 negativity of donor or recipient, intensity of maintenance immunosuppression and antirejection treatment (15-19). Accordingly, decreasing of immunosuppressive drugs may be followed by clearing of BKV replication in pre-emptive settings (20) as well as in cases with histologically defined PVAN (13, 21-23). In such patients, BKV-specific T-cell responses become increasingly detectable among peripheral blood mononuclear cells (PBMC) (17, 24, 25). In a pilot study of five patients, we observed that declining BKV loads in plasma were associated with increasing cellular immune responses against BKV early gene large T-antigen (LT) and late VP1 capsid protein (VP1) in PBMC of KT patients (26). We sought to further evaluate this observation and to compare it with responses in BKV-seropositive healthy donors (HD). Study Participants and Methods Study participants HD were 39 years old (median; range 28–53, seven males, three females) without BKV or JCV in plasma or urine. KT patients (median age 54 years, range 21–65) attending outpatient nephrology clinic during March 3–October 10, 2006, were enrolled with written informed consent (Ethics Committee of the Basel Cantons 235/06) if there was evidence of present or past plasma BKV viremia (for summary, see Table 1). Where indicated, immunosuppression was either increased, decreased or remained unchanged in the preceding 8 weeks of analysis. Cidofovir was not used in any patient. Leflunomide had been used in the one patient who subsequently suffered renal allograft loss [1/42 (2.3%)]. All study participants were seropositive for BKV and JCV as determined by testing of 1:400 diluted plasma by ELISA format for IgG against BKV- or JCV-VP1 virus-like particles purified from Sf9 insect cells. Table 1. BKV-replication and -specific cellular immune response in KT patients Patient tested Age (years) Sex (f/m) S-Creat conc (μmol/L) Weeks post-TX IS Urine BKV-load (cp/mL) Plasma BKV- load (cp/mL) Plasma ΔBKV-load (Δlog cp/mL) Plasma ΔBKV- load (weeks) BKV-LT (SFU/Mio PBMC) BKV-VP1 (SFU/Mio PBMC) Group 1 increasing 206 64 f 147 450 unc 1.00E + 08 3.00E + 02 1.41 7 0 8 (n = 5) 214 40 f 88 54 unc 3.00E + 02 2.06E + 03 0.84 2 4 19 175 61 m 144 23 unc 1.00E + 08 1.84E + 06 2.94 11 16 0 91 36 m 136 21 inc 2.62E + 07 1.93E + 07 2.07 10 25 56 134 62 f 150 163 inc 1.03E + 04 3.00E + 02 0.82 1 37 84 Mean 53 133 142 4.52E + 07 4.23E + 06 1.62 6 16 33 SD 13 26 182 5.11E + 07 8.47E + 06 0.90 5 15 35 Median 61 144 54 2.62E + 07 2.06E + 03 1.41 7 16 19 Range 36–64 88–150 21–40 3.00E + 02–1.00E + 08 3.00E + 02–1.93E + 07 0.82–2.94 1–11 0–37 0–84 Stable (n = 17) 160 54 m 217 22 unc 2.74E + 07 2.39E + 03 1.94 102 0 12 8 51 m 166 298 unc n.a. 2.12E + 03 −1.46 8 3 7 4 64 m n.a. 56 unc n.a. 2.77E + 05 −0.27 41 8 120 153 54 m 235 15 dec 1.00E + 08 9.91E + 04 0.32 8 9 92 112 57 f 156 n.a. unc 3.00E + 02 4.38E + 05 −0.05 24 12 6 203 52 m 139 27 unc 1.00E + 08 6.75E + 03 0.34 15 16 28 62 72 m 126 21 unc 1.00E + 08 3.53E + 04 −0.50 11 19 44 145 70 f 320 218 unc 5.43E + 04 2.25E + 03 0.59 24 19 49 202 52 m 133 23 unc 1.00E + 08 5.14E + 03 0.46 11 24 31 196 52 m 132 13 unc 3.06E + 07 1.49E + 04 0.00 1 28 12 88 36 m 136 11 dec n.a. 1.63E + 05 −1.38 3 37 156 66 72 m 106 26 unc 1.00E + 08 6.55E + 05 1.76 16 40 243 116 26 m 105 332 unc 2.14E + 06 3.68E + 03 −0.41 16 45 423 97 54 m 111 20 unc 8.31E + 07 1.12E + 04 0.23 8 49 227 237 46 f 130 313 unc 3.43E + 03 2.65E + 04 1.95 6 49 157 73 72 m 107 35 inc 7.15E + 07 7.28E + 04 −0.81 25 65 259 172 21 m 192 105 dec 3.10E + 07 6.96E + 03 −0.06 34 120 256 Mean 53 157 96 5.33E + 07 1.07E + 05 0.16 21 32 125 SD 15 58 120 4.38E + 07 1.85E + 05 1.01 24 29 121 Median 54 135 27 5.13E + 07 1.49E + 04 0.00 15 24 92 Range 21–72 88–320 11–332 3.00E + 02–1.00E + 08 2.12E + 03–6.55E + 05 −1.46–1.95 1–102 0–120 6–423 Total group 1 (n = 22) Mean 53 151 107 unc 5.12E + 07 1.04E + 06 0.49 17 28 104 SD 14 53 133 unc 4.45E + 07 4.10E + 06 1.15 22 27 114 Median 54 136 27 unc 3.10E + 07 1.31E + 04 0.33 11 22 53 Range 21–72 105–320 11–40 unc 3.00E + 02–1.00E + 08 3.00E + 02–1.93E + 07 −1.46–2.94 1–102 0–120 0–423 Group 2 decreasing 164 54 m 257 27 dec 7.25E + 04 1.66E + 03 −2.10 19 8 72 (n = 13) 208 64 f 171 462 unc 1.85E + 04 3.00E + 02 −1.411 5 11 45 167 54 m 150 36 unc 1.70E + 05 3.00E + 02 −2.84 29 15 101 54 37 f 103 152 dec 1.04E + 04 3.00E + 02 −4.34 20 29 107 183 61 m n.a. 38 dec n.a. 2.92E + 05 −1.661 4 29 223 187 61 m 155 44 dec 6.22E + 05 6.44E + 03 −3.33 10 69 293 100 54 m 117 28 unc n.a. 3.00E + 02 −2.42 8 80 105 82 59 m 206 21 unc 2.17E + 07 1.93E + 04 −2.20 12 93 293 25 54 f 172 376 unc 5.40E + 03 3.00E + 02 −1.311 10 109 373 126 51 f 137 71 unc 2.96E + 06 2.43E + 02 −2.03 30 121 317 186 61 m 163 42 unc 3.49E + 05 4.43E + 03 −3.49 8 268 840 182 61 m 202 36 dec 1.00E + 08 1.57E + 05 −1.941 3 465 1432 49 37 f 104 140 dec 8.99E + 06 1.12E + 04 −2.77 8 524 392 Mean 55 161 113 1.23E + 07 3.80E + 04 −2.45 13 140 353 SD 9 45 143 2.98E + 07 8.75E + 04 0.88 9 172 386 Median 54 159 42 3.49E + 05 1.66E + 03 −2.20 10 80 293 Range 37–64 103–257 21–462 5.4E + 03–1.00E + 08 2.43E + 02–2.92E + 05 −4.34–1.31 3–30 8–524 45–1432 After PVAN (n = 7) 119 26 m 116 356 (dec)/unc 1.57E + 05 3.00E + 02 −3.602 28 20 147 141 66 m n.a. n.a. no n.a. 3.00E + 02 −4.692 73 20 47 38 38 f 113 46 (dec)/unc 3.00E + 02 3.00E + 02 −4.322 22 24 53 16 54 f 203 349 (dec)/unc 7.55E + 03 3.00E + 02 −2.412 9 75 276 139 41 m 217 340 (dec)/unc 3.00E + 02 3.00E + 02 −3.322 19 93 412 129 62 f 141 141 (dec)/unc 4.50E + 03 3.00E + 02 −4.432 33 97 373 55 37 f 105 156 (dec)/unc 3.70E + 03 3.00E + 02 −4.332 24 400 816 Mean 46 149 231 2.89E + 04 3.00E + 02 −3.88 30 104 303 SD 14 49 134 6.31E + 04 0.00E + 00 0.81 21 135 269 Median 41 129 248 4.10E + 03 3.00E + 02 −4.32 24 75 276 Range 26–66 105–217 46–356 3.00E + 02–1.57E + 04 3.00E + 02–3.00E + 02 2.41–4.70 9–73 20–400 47–816 Total group 2 (n = 20) Mean 52 157 151 7.95E + 06 2.48E + 04 −2.95 19 128 336 SD 12 45 148 2.44E + 07 7.19E + 04 1.09 16 157 343 Median 54 153 71 7.25E + 04 3.00E + 02 −2.80 16 78 285 Range 26–66 103–257 21–462 3.00E + 02–1.00E + 08 2.43E + 02–2.92E + 05 −4.34–4.70 3–73 8–524 45–1432 Total (n = 42) Mean 52 154 128 3.08E + 07 5.59E + 05 −1.04 18 76 214 SD 13 49 140 1.42E + 07 2.98E + 06 2.11 19 120 274 Median 54 141 45 1.38E + 06 4.06E + 03 −1.06 11 29 114 Range 21–72 88–320 11–462 3.00E + 02–1.00E + 08 2.43E + 02–1.93E + 07 −4.70–2.94 1–102 0–524 0–1432 f = female; m = male; S-Creat = serum-creatinine; unc = unchanged; dec = decreased within the preceding 8 weeks; inc = increased within the preceding 8 weeks; no = no immunosuppression; (dec)/unc = decreased after diagnosis; unchanged with the preceding 8 weeks; cp/mL = copies per milliliter urine or plasma; LT = BKV-large T-antigen; VP1 = BKV-VP1 capsid protein; SFU/Mio = spot-forming units per million PBMC; n.a. = not available. 1Decreasing >2log10 in previous episode. 2Maximum decrease in previous episode. According to our previous analysis of plasma BKV load dynamics in KT patients taking into account sampling density, individual fluctuation and the variation coefficient of 29.6 (13, 22), we considered plasma BKV load changes of >1.5log10 significant for the purpose of this study. KT patients were divided into two groups according to the changes in plasma BKV load at the time of PBMC testing for BKV-specific interferon-γ (IFN-γ) production. Group 1 included all KT patients with increasing plasma BKV loads (n = 5) or persisting BKV loads with past BKV loads of >104 cp/mL and 1.5log10 decline (n = 13) or past BKV viremia (n = 7) at the time of PBMC testing (Table 1), the latter with histologically confirmed PVAN at median 20 months earlier (range 5–75 months). Collection of blood cells and plasma Human PBMC of BKV-seropositive HD and KT patients were isolated using CPT™ tubes (Becton Dickinson, Allschwil, Switzerland). The cells were washed, counted and either used directly for IFN-γ ELISpot assay (ESA) or for antigen stimulation and 9-day culture, or cryopreserved in 10% DMSO/90% FCS (SIGMA, Buchs, Switzerland). PBMC for ESA were diluted in RPMI/5% human serum/1% glutamax/1% penicillin streptomycin (R5AB, SIGMA). Patient plasma and urine were used to detect BKV- and JCV-DNA by real-time PCR. Quantitative PCR for BKV- and JCV-DNA detection BKV- and JCV-DNA were isolated with QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) and quantified by real-time PCR (TaqMan/7700, Stratagene Mx4000 or BioRad iCycler) as described previously (27). The following primers were used for BKV-LT amplification: BKV forward; AGC AGG CAA GGG TTC TAT TAC TAAAT (26 mer), BKV reverse; GAA GCA ACA GCA GAT TCT CAA CA (23mer), Fam-Tamra-labelled BKV probe; AAG ACC CTA AAG ACT TTC CCT CTG ATC TACACC AGT TT (38 mer), JCV forward; CTA AAC ACA GCT TGA CTG AGG AAT G (25mer), JCV reverse; CAT TTA ATG AGA AGT GGG ATG AAG AC (26mer), Fam-Tamra-labelled JCV probe; TAG AGT GTT GGG ATC CTG TGT TTT CAT CAT CAC T (34 mer). Primers and probes were obtained from Eurogentech (Geneva, Switzerland). The linear range of the real-time PCR is 102–108 cp/assay and the limit of detection was 300 cp/mL. BKV and JCV overlapping peptide pools Overlapping peptide pools spanning large LT and VP1 coding sequences from the BKV Dunlop strain and from the JCV Mad 1 strain were used for PBMC stimulation. The pools consisted of 180 (BKV-LT), 88 (BKV-VP1), 170 (JCV-LT) and 86 (JCV-VP1) peptides of 15 amino acids (aa) in length overlapping by 11aa (Eurogentech) and dissolved in DMSO (SIGMA). ESA for IFN-γdetection BKV- and JCV-specific cellular immune responses were determined by measuring IFN-γ upon stimulation of PBMC. For ESA, 96-well multiscreen filter plates (MSIPN, Millipore, Volketswil, Switzerland) were coated with 100 μL (10 μg/mL) coating IFN-γ antibody (1-D1K, Mabtech, Hamburg, Germany) and incubated overnight at 4°C. PBMC were seeded in triplicate at 2.5 × 105/well with or without peptides (2 μg/mL) and at 2.5 × 104/well with Staphylococcus enterotoxin B (SEB; SIGMA, 1 μg/mL) as positive control, and incubated overnight at 37°C/5% CO2. Plates were washed five times with PBS/0.05% TWEEN 20 (SIGMA) and incubated for 2 h at room temperature with 100 μL (1μg/mL) biotinylated detection IFN-γ antibody (7-B6-1 biotin, Mabtech), washed five times and incubated for 1 h at room temperature with 100 μL (1μg/mL) Streptavidin ALP (Mabtech). Spots were developed by adding 100 μL BCIP/NBT (Calbiochem, Luzern, Switzerland) for 10 min. IFN-γ spots were counted by using an ELISpot reader (AID, Büron, Switzerland). The number of spot-forming units/106 PBMC (SFU) per well was calculated from triplicates after subtractions of negative control. Nine-day expansion of PBMC PBMC (2 × 106/mL) were distributed into 24-well plates and incubated overnight at 37°C/5% CO2. Nonadherent cells were recovered, washed and adjusted to 2 × 106 cells/mL. Adherent monocytes were detached by scraping/pipetting, adjusted to 2 × 106 cells/mL and stimulated with 5 μg/mL BKV-LT and -VP1 for 2 h at 37°C/5% CO2. Monocytes were washed and 105 cells/mL were added to 2 × 106 of nonadherent cells. After 9 days at 37°C/5% CO2, ESA was performed as described above using 105 cells for peptide restimulation or 2.5 × 104 for SEB per well. Intracellular cytokine staining and flow cytometry BKV-specific IFN-γ production was quantified using intracellular cytokine staining and flow cytometry as published by Sester et al. (28) after stimulation with BKV-LT or -VP1 peptide pools (2 μg/mL, Eurogentech). Medium alone served as negative control and stimulation with SEB (1 μg/mL, SIGMA) as positive control. At least 30 000 CD3+ cells were acquired and analysed on a FACS-Canto (Becton Dickinson). The frequency of BKV-specific cellular immune responses was determined for each antigen and expressed as percent of IFNg+ cells in CD3+CD4+ or CD3+CD8+ gated lymphocyte populations, respectively. Statistical methods Nonparametric statistical tests were performed using the SPSS (Version-14) to account for sample size and non-normal distribution. For comparison of the study groups, two-sided Mann–Whitney U-Test was used for calculating the p value. The correlation between paired samples was evaluated by the two-sided Wilcoxon signed rank sum test. Data were expressed either as mean ± standard deviation (SD) or as median and range. Differences with p values <0.05 were considered as statistically significant. For better comparison, box plots are shown with median and interquartile range over data scatters, with outliers above or below the whiskers. Results BKV-specific cellular immune response in HD PBMC from BKV-seropositive HD were stimulated using peptide pools covering the entire BKV large LT and VP1, respectively. IFN-γ secretion was detected by ESA. The median spot-forming units (SFU)/106 PBMC (SFU) were 24 (range 15–95) for LT peptide pools and 25 (range 7–113) for VP1 peptide pools (Figure 1). No statistical difference was observed between LT and VP1 induced responses in HD (p = 0.626). Figure 1Open in figure viewerPowerPoint IFN-γ response of PBMC after stimulation with BKV-LT and -VP1 peptides by ESA. PBMC of HD and KT patients were stimulated with BKV-LT and -VP1 overlapping peptide pools and IFN-γ secretion was quantified by ESA. Box plots show median and interquartile range superimposed over scatter (Wilcoxon, two-sided). BKV-specific cellular immune response in KT patients We tested PBMC from BKV-seropositive KT patients with ongoing or past BKV viremia who attended outpatient nephrology clinics between March and October 2006 (Table 1). Stimulation with BKV-LT and -VP1 peptides elicited 29 [median, range (0–524)] and 114 [median, range (0–1432)] SFU, respectively (Figure 1). Overall, the SFU cellular immune responses to BKV-VP1 antigens were higher than the LT-specific response (p = 0.005) which was also true at the individual level for 38 of the 42 KT patients (91%). When KT patients were grouped according to the level of plasma BKV load, no significant differences were observed between patients with BKV loads 105 cp/mL (Figure 2). Figure 2Open in figure viewerPowerPoint IFN-γ response of PBMC of KT patients with low or high plasma BKV load. KT patients were divided into groups of 105 cp/mL and compared for SFU by ESA obtained after stimulation with BKV-LT and -VP1 peptides. To identify the contribution of CD4+ and CD8+ T cells in PBMC, we performed intracellular cytokine staining and flow cytometry after direct stimulation with BKV-LT or -VP1 peptides. In 21 KT patients (13 with >100 IFN-γ SFU/106 PBMC to VP1 peptides), we found IFN-γ positive cells at frequencies above 0.01%. Although the overall responses were low, mean IFN-γ responses to BKV-VP1 tended to be higher for CD4+ T cells than for CD8+ T cells (p = 0.038), whereas the difference did not reach significance for BKV-LT peptides (p = 0.304) (Figure 3). However, CD8+ T cells were more likely to respond to BKV-LT than to -VP1 peptides (p = 0.033). Figure 3Open in figure viewerPowerPoint Intracellular IFN-γ response in CD4+ T cells and CD8+ T cells of KT patients after stimulation with BKV-LT and -VP1 peptides. PBMC of KT patients were stimulated with indicated peptides and gated for CD4+CD3+ and CD8+CD3+ lymphocytes and analysed for intracellular IFN-γ production using flow cytometry (Wilcoxon, two-sided). To investigate the association of plasma BKV load dynamics and immune responses, we divided KT patients in two groups: group 1 consisted of 22 KT patients with increasing (n = 5) or persistent plasma BKV loads (n = 17) at the time of PBMC testing. Group 2 consisted of 20 KT patients with plasma BKV loads decreasing on an average >2log10 cp/mL (n = 13) or with past PVAN (n = 7) at the time of PBMC testing (Table 1, see Section 'Study Participants and Methods'). Although the number of IFN-γ SFU were scattered in both groups for both antigens, BKV-LT-specific responses in group 1 were significantly lower than in group 2 [median 22 (0–120) vs. 78 (8–524); p = 0.003]. Similarly, BKV-VP1-specific responses in group 1 patients were significantly lower than in group 2 [median 53 (0–423) vs. 285 (45–1432) SFU/106 PBMC; p = 0.001 (Figure 4)]. Comparison with HD did not allow identifying a cut-off for KT patients with a protective cellular immune response. However, 69 SFU/106 PBMC for BKV-LT antigens, but not for VP1 antigens, identified significantly more KT-patients of group 2 with decreasing plasma BKV loads than of group 1 with increasing plasma BKV loads (12 of 20 vs. 1 of 22, respectively; p = 0.019, Fisher's exact test, two-sided). Figure 4Open in figure viewerPowerPoint IFN-γresponse of PBMC and plasma BKV load dynamics in KT patients. PBMC of KT patients with increasing or persisting BKV loads (group 1) or with decreasing (mean >2log10) or past plasma BKV load (group 2) were stimulated with indicated peptides and IFN-γ production was measured by ESA (Wilcoxon, two-sided). BKV-specific cell expansion by short-term culture To examine the ability of T cells able to expand in response to BKV-specific antigens in vitro (after wash out of immunosuppression), we stimulated PBMC of 24 KT patients (12 from each group) with autologous activated mononuclear cells primed with BKV-LT or -VP1 peptides. After 9 days of in vitro culture, cells were restimulated with BKV-LT or -VP1 peptides and tested for IFN-γ production by ESA. In most cultures, we observed a significant expansion of BKV-LT- and -VP1-specific responses (p = 0.002 and 0.033, respectively) compared to the SFU obtained prior to expansion (Figure 5). When we compared the SFU obtained after in vitro expansion of T cells from KT patients grouped according to plasma BKV load dynamics (group 1, inc/hi; group 2, dec), BKV-LT-specific expansion yielded significantly higher SFU in KT patients with declining plasma BKV loads (group 2; p = 0.013) than in group 1 KT patients with increasing or high plasma BKV loads (group 1; p = 0.064). In contrast, no significant difference was observed for expanding BKV-VP1-specific responses of both groups (p = 0.084 and 0.286, respectively). Figure 5Open in figure viewerPowerPoint IFN-γ responses of PBMC after direct stimulation or after 9-day culture and restimulation. PBMC of KT patients were stimulated with BKV-LT or -VP1 peptides and cultured for 9 days. Cells were restimulated with indicated BKV peptides and IFN-γ SFU were quantified by ESA. The SFU responses post-culture were compared with preculture responses obtained by direct stimulation (Wilcoxon, two-sided). Cross-stimulation with JCV peptide pools As BKV and JCV share 75% DNA sequence homology, we wondered, if JCV-LT and -VP1 overlapping peptide pools were able to elicit a comparable IFN-γ response as BKV-LT and -VP1 peptides. When PBMC from 40 KT patients were stimulated directly with JCV-LT and -VP1 peptides, mean 49 (±60; median 27, range 0–255) (±59) SFU and mean 118 (±169; median 47, range 0–868) SFU/106 PBMC were found, respectively. The JCV-peptide mediated response was significantly lower when compared to the corresponding BKV-mediated response (p = 0.008 and p < 0.001, respectively). When JCV-LT or -VP1 peptides were used to restimulate T cells after BKV-specific 9-day expansion in vitro, the overall BKV-VP1 restimulation responses were significantly higher compared to JCV-VP1-specific responses (p = 0.016), but few individual exceptions of higher JCV responses were noted (Figure 6). In contrast, no significant difference was observed between BKV-LT and JCV-LT restimulation responses (p = 0.398) (Figure 6). Figure 6Open in figure viewerPowerPoint Cross-stimulation with homologous JCV-LT and -VP1 peptides. IFN-γ responses to BKV- and JCV-LT (top) and -VP1 (bottom) were compared after 9-day culture with BKV peptides (top panel: restimulation with JCV-LT, p = 0.398; bottom: restimulation with JCV-VP1; lower, p = 0.016; Wilcoxon, two-sided). Discussion A failing balance between BKV replication and BKV-specific cellular immune functions has been suspected as the common denominator of PVAN pathogenesis (15). Likely, this balance can be perturbed at different points of the patient, graft and virus interaction (3) which may account for controversial results on some of the implicated risk factors such as type of immunosuppression, BKV serostatus of donor and recipient, HLA mismatches or antirejection therapy. However, plasma BKV load is now widely accepted as surrogate marker of this failing balance and the risk of PVAN in KT patients (2, 13, 20). Also, plasma BKV loads not only provided first estimates of cytopathic damage (2, 13, 20), but also indicated the average efficacy of reduced immunosuppression as the current key treatment (22). Accordingly, clearance of plasma BKV viremia in patients with 105–107 cp/mL requires 7–11 weeks, once the immune system has started to curtail BKV replication (22). Despite the versatility of BKV loads in clinical screening and monitoring, individual courses are often difficult to predict. Instead, BKV-specific cellular immune responses in PBMC have been suggested as a more direct measure of this failing balance complementing BKV loads in KT patients (15, 17). The results of the present study indicate that BKV-seropositive KT patients with increasing or persisting BKV loads show significantly lower IFN-γ SFU for both, BKV-LT and -VP1 peptides than KT patients with mean 2.45log10 decreasing or past plasma BKV loads. Thereby, the dynamics of plasma BKV replication are mirrored by emerging BKV-specific cellular immune responses in PBMC. Previous reports suggested that in some KT patients, VP1-specific cellular immune responses were preferentially detectable in patients with high plasma BKV loads of >105 cp/mL (25). Although BKV-specific cellular immune responses are detectable in some of these patients, we observed no significant differences between patients with plasma BKV loads above or below 105 cp/mL. Our results suggest that replication dynamics, not replication levels correlated with emerging BKV-specific cellular immune responses. The increasing responses to defined BKV early gene (LT-) and late gene (VP1-) antigens reported here extend previous findings in paediatric KT patients using whole virus preparations (17) or VP1-based responses with peptide pools (25) or tetramer staining limited to single HLA-0201 binding epitopes in adults (24, 29). Our data further suggest that BKV-LT and -VP1 responses seem to differ in several respects. First, the median number of IFN-γ SFU to BKV-VP1 was significantly higher than to BKV-LT (p = 0.005) which was also true at the individual level for 38 (91%) of the 42 KT patients tested. In line with our previous observation (30), VP1 responses seem to be a more sensitive measure of cellular immune responses in PBMC which can be detected at an earlier time point. Preliminary data on five KT patients with serial testing support this notion which requires further study in a prospective fashion. Second, flow cytometry indicated that higher IFN-γ frequencies to BKV-VP1 largely resulted from CD4+ T-cells as compared to CD8+ T cells (p = 0.038). While part of the CD4+ predominance is likely to reflect the preferential presentation and recognition of 15mers peptides in th
Referência(s)