Aim Reactivation of latent BK polyomavirus (BKV) infection is relatively common following renal transplantation and BKV-associated nephropathy has emerged as a significant complication. JC polyomavirus (JCV) reactivation is less well studied. The aim of the study was to determine reactivation patterns for these polyomaviruses in renal transplant recipients using an in-house quantitative real-time multiplex PCR assay and IgG serological assays using recombinant BK and JC virus-like particles.
Methods Retrospective analysis of urine and plasma samples collected from 30 renal transplant patients from February 2004 to May 2005 at Leeds Teaching Hospitals NHS Trust. Samples were collected at 5 days and thereafter at 1, 3, 6 and 12 months post-transplantation.
Results Eight patients (26.7%) were positive for BK viruria; three of these patients submitted plasma samples and two had BK viraemia. Five patients (16.7%) were positive for JC viruria. A corresponding rise in BKV and JCV antibody titres was seen in association with high levels of viruria.
Conclusions Different patterns of reactivation were observed: BK viruria was detected after 3–6 months, and JC viruria was observed as early as 5 days post-transplantation. One patient had biopsy-proven BKV nephropathy. No dual infections were seen. In order to ensure better graft survival, early diagnosis of these polyomaviruses is desirable.
- BK polyomavirus
- haemagglutination inhibition
- JC polyomavirus
- polyomavirus-associated nephropathy
- real-time PCR
Statistics from Altmetric.com
- BK polyomavirus
- haemagglutination inhibition
- JC polyomavirus
- polyomavirus-associated nephropathy
- real-time PCR
BK polyomavirus (BKV) and JC polyomavirus (JCV) are both ubiquitous, causing sub-clinical infection within the first 10–15 years of life. After primary infection, BKV establishes latency within renal tissues and JCV in circulating B lymphocytes.1–4
BK polyomavirus-associated nephropathy (PVAN) has emerged as a major cause of kidney graft failure (not due to rejection) in renal transplant patients (RTPs). JCV is the causative agent for the neurological disease progressive multifocal leucoencephalopathy, which occurs primarily in AIDS patients.5 JCV has been identified in kidney biopsy tissue and urine by immunohistochemistry and PCR, respectively, from a subset of RTPs with tubulointerstitial nephritis.6–9 However, its role as a cause of PVAN remains to be defined.
The introduction of potent immunosuppressive drugs in the past decade has made reactivation of latent infection by BKV and JCV relatively common.10 More recently steroid-sparing immunosuppressive regimens have been described following renal transplantation.11–13 It is unclear whether these will affect risk of polyomavirus reactivation and subsequent disease.
Interdisciplinary analyses carried out by an independent panel of experts and The American Society of Transplantation suggest RTPs with progressive infection who have high urine DNA levels (>1×107 copies/ml) are more likely to develop detectable DNA in plasma (>1×104 copies/ml) and prolonged viraemia, which precedes clinically overt nephropathy.14 15 Once diagnosis has been made by kidney biopsy, prognosis is poor, with 30–50% of patients developing progressive renal dysfunction and, ultimately, allograft loss. In order to have better graft survival, early diagnosis of BKV is desirable. PVAN has been treated by reducing immunosuppression, treatment with cidofovir or leflunomide, or intravenous immunoglobulin. However, optimal management has yet to be established.16–20
The aim of the study was to determine reactivation patterns for BKV and JCV in RTPs, using an in-house developed quantitative real-time multiplex PCR assay. We also verified the performance of serological assays based on recombinant virus-like particles.
Primers and hydrolysis probes
A comparison of complete BKV (types I, III and IV) and JCV genomes within GenBank (http://www.ncbi.nlm.nih.gov) was performed using Bionumerics. A conserved region within the large T antigen was identified as an appropriate target, which distinguished BKV from JCV (table 1). Primer and probe design was undertaken and verified using NetPrimer from Premier Biosoft International software (http://www.premierbiosoft.com), and homology to other viruses or host genome sequences was checked using BLAST on the National Centre for Biotechnology database (http://blast.ncbi.nlm.nih.gov/Blast.cgi).
Plasmid DNA standards
PCR products for plasmid preparation were produced from positive BKV and JCV urine samples, and identified by electron microscopy and PCR. Plasmids were constructed by ligation of the PCR product in to the vector pCRBlunt. The plasmid DNA was extracted; the concentration (μg/ml) was determined using a spectrophotometer, and the copy number was calculated.
This study was approved by Leeds (East) Research Ethics Committee (reference no. 03/209). We undertook retrospective analysis of 30 new RTPs from February 2004 to May 2005 at Leeds Teaching Hospitals NHS Trust, who met inclusion criteria of three urine samples being collected at 5 days and thereafter at 1, 3, 6 and 12 months post-transplantation. Plasma samples were not routinely collected but were available for eight RTPs. Patients received induction with tacrolimus and mycophenolate mofetil, a single dose of intravenous corticosteroids intraoperatively and two doses of basiliximab on days 0 and 4. At 3 months, patients were randomised to continue with either tacrolimus (group I) or sirolimus (group II).
Extraction and amplification
DNA was extracted using the Corbett X-tractor Gene (Qiagen, Crawley, UK) and real-time multiplex PCR was performed on the Stratagene MX3000P (Agilent, South Queensferry, UK). To achieve absolute quantification of BKV and JCV DNA, standard curves were constructed using 10-fold dilution series of each plasmid from 1×108 to 1×103 copies/μl. A 25 μl reaction mix was prepared with Brilliant Multiplex QPCR Master Mix (Agilent), primers at 0.3 μM, BKV and JCV probes at 0.2 μM and 0.1 μM, respectively, and 5 μl plasmid or sample extract. The thermal profile reaction conditions were 95°C for 10 min followed by 45 cycles of 95°C for 15 s, 55°C for 1 min and 72°C for 30 s.
Haemagglutination inhibition and enzyme immunoassay serology
A haemagglutination inhibition (HAI) assay was performed in 96-well plates to determine IgG antibody titres, using recombinant BK and JC VP1 virus-like particles, as described previously.21 22 Antibody testing was also performed by enzyme immunoassay (EIA) using a virus-like particle concentration of 0.1 μg/ml in phosphate-buffered saline.23
In view of the small numbers, data were analysed using Fisher's exact test, exact confidence interval (CI) and the Mann–Whitney test. Analysis was carried out using SPSS v15 (SPSS, Chicago, Illinois, USA). Missing samples were treated as negative, and no formal missing value analysis was attempted.
There was no cross-reactivity seen between BKV and JCV PCR assays using high plasmid concentrations (data not shown). This quantitative real-time multiplex PCR assay was optimised to achieve between 90% and 100% efficiency using 10-fold dilution series of BKV (figure 1) and JCV plasmids, with sensitivities achieved of approximately 500 copies/ml of sample. This was verified in the Quality Control for Molecular Diagnostics 2009 JC and BK virus DNA EQA Programme (data not shown).
One hundred and seven urine samples were submitted from 30 patients and 35 plasma samples were available from eight of these.
Of the 30 RTPs analysed eight (26.7%) were positive for BK viruria, five (16.7%) were positive for JC viruria, and 17 (56.7%) were negative. Results comparing the different groups are shown in table 2.
BKV DNA was detected in one or more urine samples at a median of 3–6 months after transplantation (range 5 days to 12 months). Viral loads ranged from 1×102 to 1×1010 copies/ml. Plasma samples were collected from three of these eight patients. Two patients (no 70 and 72) with viruria levels >1×107 copies/ml also had BK viraemia, with viral loads of 1×102 and 1×104 copies/ml. Patient no. 70 exhibited a 16-fold rise in antibody titre at 3 months, and this then declined; chronic renal damage was seen on renal biopsy at 6 months post-transplant, but this was not suggestive of PVAN. Patient no. 72 had viruria levels >1×1010 copies/ml and viraemia >1×104 copies/ml, with a corresponding 500-fold rise in antibody titre by EIA. Biopsies taken for histopathology at 3 and 6 months post-transplant showed acute tubular necrosis but no evidence of rejection or PVAN. Patient no. 74 had an eightfold rise in antibody titre; however, viruria was at a relatively low level (2.5×104 copies/ml) and no viraemia was observed. Only patient no. 6 had biopsy-proven BKV nephropathy by histopathology at 12 months, and this correlated with the high viral load in the urine at 6 and 12 months post-transplant (table 3 and figure 1); there were no plasma samples available for determination of viraemia or serology. Patient no. 2 had BK viruria at 12 months post-transplant, but had experienced severe rejection at 3 months post-transplant. Results for BKV positive patients are summarised in tables 3 and 4.
JC viruria was detected at a median of 5 days (range 5 days to 3 months). One patient (no. 69) had viruria levels >1×107 copies/ml and had plasma referred; the plasma was negative for JC viraemia but showed an eightfold rise in antibody titre. Plasma samples were not available for the other JCV-positive cases. None of the five patients had any rejection episodes. From the patient notes there was no difference seen in the serum creatinine levels between BK/JC viruria positive or negative groups at 5 days, or 1, 3, 6 or 12 months post-transplant. Results for JCV-positive patients are summarised in tables 4 and 5.
There was a significant difference seen in median time to first BK or JC viruria (p=0.03). Although there was a preponderance of female cases in the BKV-positive group, this sex distribution did not differ significantly for viruria (p=0.07). There was no change seen in antibody titres from BKV-negative and JCV-negative patients who had plasma samples collected (data not shown).
We used a quantitative real-time multiplex PCR assay for the detection of BKV and JCV to describe the patterns of reactivation in RTPs. Other real-time quantitative PCR assays have been described that investigated large T antigen, VP1 or VP2 genes, but they have used different PCR platforms. Those assays detected BKV DNA,18 24–26 26–32 JCV DNA28 29 31 32 or BKV and JCV DNA,33 but required a separate assay to differentiate viruses.34
No dual BKV and JCV infections were observed. From this cohort of 30 RTPs, different patterns of BK and JC virus reactivation were observed; where JC viruria was detected earlier (median 5 days) than BK (median 3–6 months). This finding has not been described previously, but a larger study would be required to confirm this disassociation.
In other studies, BK viruria has been detected between 3.6 and 4 months by decoy cells and PCR; this is similar to our findings.35 36 BK viraemia was detected in two of the three RTPs where viruria was observed. BK viraemia was observed at 1 and 6 months post-transplantation in comparison to means ranging from 5.75 to 11.5 months in other studies.24 35 37–39 Early or post-transplant BKV infection is thought to be of donor origin.40 41 We were not able to establish the origin of BKV in this study. It is difficult to confirm the exact timing of viraemia after viruria, as plasma samples were not collected regularly from patients. There was only one case of biopsy-proven BKV nephropathy at 12 months, and this patient had high levels of viruria at 6 and 12 months. In other studies, BKV nephropathy has been detected at means ranging from 6 to 12.8 months, with BK viruria levels between 107 and 1010 copies/ml and viraemia levels between 103 and 107 copies/ml. In these patients nephropathy resolved and inclusions disappeared, with the resolution of viraemia and decreased viruria levels.35 42–45 If after histological diagnosis the appropriate intervention is not carried out, then 45% of cases will lose allograft within 6 months.46 The one JCV-positive RTP who had corresponding plasma samples with viruria levels >1×107 copies/ml did not result in viraemia or nephropathy; this is similar to findings by Randhawa et al.47
A corresponding rise in BKV or JCV antibody titres was seen in association with increasing levels of viruria. All RTPs had antibody titres to BKV and JCV but showed no significant rise when BKV/JCV PCR was negative, suggesting past infection. However, there was a greater rise seen in EIA antibody titres in comparison with HAI titres, suggesting that the EIA is more sensitive and can detect a greater number of epitopes than the HAI. This confirms the findings of Hamilton et al.5 There was no difference seen in the serum creatinine levels between BK/JC viruria-positive or viruria-negative groups, although an association between viraemia and high creatinine levels has been demonstrated previously.48
To conclude, from this study of 30 RTPs, 13 patients developed polyomavirus infection with only one case of nephropathy. We describe a quantitative real-time multiplex PCR assay that is sensitive and can be used for monitoring RTPs, allowing the early diagnosis of polyomavirus infections. Although not always associated with renal disease, this may guide the need for renal biopsy and interpretation.
Real-time quantitative multiplex BK polyomavirus and JC polyomavirus PCR assay can be used to monitor infections in renal transplant patients.
Polyomavirus-associated nephropathy is associated with high viruria levels (>1×107 copies/ml) and lower levels of viraemia (>1×104 copies/ml).
BK polyomavirus reactivation occurs 3 months post-transplantation whereas JC polyomavirus viruria is earlier.
The authors wish to thank the Renal Unit, St James's University Hospital, Leeds Teaching Hospitals NHS Trust, Leeds, UK for financial support, the Virology Department at Leeds General Infirmary for technical support, and Wendy Knowles from Health Protection Agency (Centre for Infections, Colindale) for the JCV-positive material.
Funding St James’s University Hospital Renal Transplant research fund.
Competing interests None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Leeds (East) Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.