Journal Logo

Original Basic Science—General

Intragraft Antiviral-Specific Gene Expression as a Distinctive Transcriptional Signature for Studies in Polyomavirus-Associated Nephropathy

Sigdel, Tara K. PhD; Bestard, Oriol MD, PhD; Salomonis, Nathan PhD; Hsieh, Szu-Chuan MS; Torras, Joan MD, PhD; Naesens, Maarten MD, PhD; Tran, Tim Q. MS; Roedder, Silke PhD; Sarwal, Minnie M. MD, PhD

Author Information
doi: 10.1097/TP.0000000000001214

Polyomavirus-associated nephropathy (PVAN) remains an important opportunistic infection after renal transplantation. Asymptomatic viremia may be observed in 10% to 30% of transplant recipients, and 4% to 10% may develop PVAN with allograft loss occurring in approximately 50% of cases.1-4 Even though immunosuppressive condition is the main cause for viral reactivation, not all latent infections in kidney transplant patients lead to the development PVAN. This suggests that although the increased burden of immunosuppression appears to be crucial for viral reactivation, individual immune-susceptibility to viral infection and type of intragraft inflammation are likely to be involved in the pathogenesis of PVAN.5

BK viremia replication can rapidly be monitored by nucleic acid testing (quantitative polymerase chain reaction [qPCR]) analysis of urine or plasma samples. The diagnosis of PVAN requires histological assessment showing typical viral cytopathic changes in tubular epithelial cells and a positive immunostaining against the LT antigen of simian polyomavirus (SV40). Importantly, accompanying these lesions, there is an important tubulointerstitial inflammatory cell infiltrate, which is indistinguishable from typical histological patterns of that observed in acute T cell–mediated rejection (TCMR).6,7 This is of clinical relevance because the therapeutic approaches are opposed; PVAN treatment fundamentally focuses on a progressive reduction of immunosuppression with eventual adjuvant medication, whereas TCMR requires of additional rescue immunosuppressive therapy.8 Therefore, an accurate recognition of the PVAN-associated molecular fingerprint could help to better discriminate these 2 processes and provide new insight of the pathogenic mechanisms of the disease.

Gene expression studies in biopsy samples of PVAN and TCMR focusing at specific genes by PCR analysis have reported a similar overexpressed transcripts associated with similar T cell activation and costimulation pathways, such as IFN-[Latin Small Letter Gamma], perforin, CXCR3, or CD40/CD40L, respectively.8,9 Likewise, in the urine, increased inflammatory cytokines similar to TCMR have also been shown in PVAN patients.10,11 However, the evaluation using high-throughput microarray analysis of PVAN/TCMR tissue allograft samples to significantly enrich the genomic picture of these pathological features is scarce. Recently, Lubetzky et al12 investigated the genomics of PVAN, mainly in whole blood and in a reduced number of tissue allograft samples by microarray analysis. Authors reported a significantly increased pathogenesis-based transcript activity of cytotoxic T cells and natural killer cells in PVAN resembling TCMR, suggesting the involvement of adaptive and innate immunity in both settings.

With the aim of obtaining a deeper understanding of the main molecular mechanisms taking place during the inflammatory process both in PVAN and TCMR, we used a high throughput microarray analysis of kidney allograft biopsies. Here, we report that although both pathological features overlap with a relevant number of inflammatory and cytotoxic pathogenesis-related transcripts, PVAN does also show differentially upregulated gene transcripts related to immune response to organisms and particularly to viral infection. Importantly, significantly overexpressed genes and gene products (proteins) in PVAN patients were further confirmed by qPCR and immunohistochemistry (IHC).

MATERIALS AND METHODS

Patients and Biopsies

The study comprised 168 posttransplant renal allograft biopsies (TCMR = 26, PVAN = 10, normal functioning graft without subclinical rejection or other injury (STA) = 73, and interstitial fibrosis (IF)/tubular atrophy (TA) =59) from 168 unique pediatric and adolescent kidney allograft recipients (1 to 21 years of age) (Table 1). All patients received an immunosuppressive regimen consisting of a combination of tacrolimus (Prograf, Astellas Pharma), mycophenolate mofetil (Cellcept, Hoffman-La Roche), and daclizumab (Zenapax, Hoffman-La Roche) or thymoglobulin (Sanofi) induction. Some patients received a steroid-avoidance regimen, whereas others received a steroid-based immunosuppressive regimen, as previously described.13 Clinical and histological demographic characteristics were collected for all the biopsies. A subset of biopsies with PVAN, TCMR, and STA phenotypes, matched for major clinical variables, such as recipient and donor age, % living donor kidneys, time posttransplant, immunosuppression usage, which was used as selected case-controls for more stringent analysis for PVAN biology (Table 1). Diagnosis of TCMR and IF/TA was made by biopsy histology Banff classification.14 All IF/TA samples showed Banff scores grade II or higher, without showing any other specific accompanying lesions or AR. PVAN was defined as positivity of polyomavirus PCR in peripheral blood, together with a positive SV40 stain in the concomitant renal allograft biopsy according to Banff criteria.15 A small number of patients had BK DNA replication but no evidence of PVAN on biopsy; these patients are categorized as BK viremia in blood only (BKVB) and have been included in the global gene expression first phase analysis in Table 1 and Figure 1. The ethics committee of Stanford University Medical School and UCSF Medical Center approved the study. All patients/guardians provided informed consent to participate in the research, in full adherence to the Declaration of Helsinki. The clinical and research activities being reported are consistent with the Principles of the Declaration of Istanbul as outlined in the Declaration of Istanbul on Organ Trafficking and Transplant Tourism.

TABLE 1
TABLE 1:
Relevant clinical demographics
FIGURE 1
FIGURE 1:
A, Global gene expression pattern of samples classified as PVAN were largely similar to TCMR samples when compared with normal kidney biopsies, BKVB, and IF/TA. B, Gene expression clustering demonstrates TCMR-specific gene signature. C, Gene expression clustering demonstrates PVAN-specific gene signature compared to normal (STA). A few genes (n = 16) were also differentially expressed in the IF/TA patients compared with PVAN but this analysis is not as clean as many patients with PVAN also have associated IF/TA changes in the graft.

RNA Extraction, Quality Control, Amplification, and Microarray Hybridization

Needle biopsies were collected at the time of biopsy procedure and immediately submerged in RNAlater (Qiagen, Valencia, CA) and stored at −80°C until use. For the gene expression analysis purpose, total RNA was extracted and hybridized to Affymetrix GeneChip Human Genome U133 Plus 2.0 Arrays as previously described in Naesens et al.16 For processing and normalization of the scanned images, dChip 2006 software was used, with perfect match/mismatch difference modeling and invariant set normalization.17

qPCR Verification

A total of 250 ng total RNA was first reversed transcribed into complementary DNA using Superscript II (Invitrogen, Life Technologies, Foster City, CA) according to the manufacturer's protocol in the Eppendorf vapo protect thermal cycler (Eppendorf, Hauppauge, NY). From each sample 1.56 ng complementary DNA were amplified in a target-specific amplification step for 4 genes (RPS15, complement factor D [CFD], lactotransferrin [LTF], nitric oxide synthase interacting protein [NOSIP]) TaqMan PreAmp Master Mix (cat. no. 4488593; Life Technologies and TaqMan Primers and Probes (RPS15—cat. no. Hs01358643_g1; CFD—cat. no. Hs00157263_m1; LTF—cat. no. Hs00914334_m1; NOSIP—cat. no. Hs00211028_m1; Life Technologies) for a total of 18 amplification cycles. Quantitative polymerase chain reaction reactions were performed in the BioMark qPCR system using 18S gene (primer and probe cat. no. Hs03003631_g1; Life Technologies) as a housekeeping gene and Human XpressRef Universal Total RNA (Qiagen Inc, Valencia, CA, cat. no. 338112) as a reference RNA for a total of 40 cycles (Fluidigm, San Francisco, CA). Resulting chip data was initially analyzed for QC using the BioMark Analysis Software Version 2.0 (Fluidigm) and Ct values were exported into Excel.

Immunohistochemistry of Kidney Allograft Biopsies

Nine representative cases of kidney allograft biopsies of PVAN (n = 3), TCMR (n = 3), and STA (n = 3) were stained for 2 overexpressed transcripts in PVAN kidney biopsies. To determine the extent of renal damage and classify them into the 3 different phenotypes (PVAN, TCMR, and STA), all renal biopsies were analyzed by 2 blinded pathologists. The biopsies were probed with rabbit polyclonal to LTF (Abcam, Cambridge, MA), rabbit polyclonal to IFN-inducible transmembrane 1 (IFITM1) (Abcam) and SV40 (Becton Dickinson, Madrid, Spain). Immune staining in formalin-fixed, paraffin-embedded tissues was performed as described previously.18-20 As positive controls for LTF, IFITM1, and SV40 human tonsil, liver carcinoma, and kidney allograft paraffin-embedded tissues were used as positive controls, respectively. To quantify LTF and IFITM1 expression, a semiquantitative score from 0 to 3 in the different compartments of the kidney (glomeruli, vessels, tubuli, and interstitium) was used.

Data Processing and Analysis

Differentially expressed genes between biopsy groups were identified using an empirical Bayes moderated t test using a Benjamini-Hochberg–adjusted adjustment in the program AltAnalyze version 2.0.8.21 In all comparisons, an FDR-adjusted P less than 0.05 was used for filtering. After identifying the differentially expressed genes, these probe sets were analyzed using GO-Elite version 1.2.6 with all available default annotation resources21-24 to identify enriched biological pathways, ontologies and gene sets. To evaluate immune cell infiltration, immune cell-type-specific markers were computationally inferred using a new marker identification algorithm (LineageProfiler) applied to 2 large published microarray studies (GSE22886, GSE15907) to be used by GO-Elite. Principal component analysis, expression clustering (hierarchical or hierarchical ordered partitioning and collapsing hybrid), pathway filtering, and visualization were also performed in AltAnalyze using the default parameters. The raw data sets for the 168 biopsies included are deposited at the Gene Expression Omnibus under GSE72925.

RESULTS

Demographics

During the time frame period of the study, 168 unique pediatric kidney allograft biopsies indicated either for cause or protocol, were initially included. Subsequently, 55 of the 168 sample-set with matched demographics were included for the gene expression analysis. Ten (~18%) patients were found to have Polyomavirus-associated nephropathy (PVAN) (70% stage C and 30% stage B),15 15 of 55 (~27%) pure TCMR, and 30 of 55 (~55%) were considered as patients with STA because no abnormalities were observed in their biopsies. As shown in Table 1, there were no differences regarding main clinical demographic characteristics, such as donor and recipient age and sex, type of maintenance and induction immunosuppression, number of previous transplants, and type of transplant. All biopsies were performed during the first 24 months after transplantation, either for protocol (at 6 or 24 months) or for cause because of allograft dysfunction or presence of BK DNA replication in peripheral blood. At the time of assessment, allograft function was not different between the 3 groups. Only the acute inflammatory Banff scores in the tubuli and interstitium of renal allograft compartments were significantly higher among PVAN and TCMR as compared with STA patients. The mean serum creatinine value of selected 15 TCMR patients was significantly higher (1.6 mg/dL) than the mean serum creatinine value of selected 30 STA patients (0.85 mg/dL) (P < 0.001). The relatively lower serum creatinine seen in AR in this cohort is a function of this being a pediatric cohort. As per definition, only PVAN patients showed positivity for SV40 immunostaining and BK DNA replication in peripheral blood. Five patients with BKVB were included in the cohort of normal biopsies.

Unsupervised Hierarchical Clustering of All Kidney Allograft Biopsies

First, unsupervised hierarchical clustering analysis was performed among all 168 posttransplant renal allograft biopsies with the aim of having a broad gene expression picture using the 500 top ranked probe sets based on an FDR-adjusted f test P value (P < 0.05 and 2-fold increase expression). This analysis indicated that the global gene expression pattern of samples classified as PVAN were largely similar to TCMR samples as both compared to normal kidney biopsies (Figure 1). The 5 biopsies with BKVB aligned with normal biopsies and were not studies as a separate phenotype. The top 500 genes were enriched in immune system process (P = 3.38E-21) and immune response (P = 3.38E-21).

Distinctive Gene Set Expression for PVAN and TCMR as Compared With STA in Well-Matched Kidney Transplant Patients

Next, to have the most comparable study population, we selected for each PVAN (n = 10) and TCMR (n = 15) patients, the best demographically matched STA patient for this analysis (n = 30). With an FDR-adjusted criterion of P values less than 0.05, a total of 4047 probe sets showed that TCMR-specific regulation of 2483 probe sets were significantly upregulated, and 1564 were significantly downregulated. With the same FDR-adjusted criterion of P values less than 0.05, a total of 11 594 probe sets showed that PVAN-specific regulation of 5450 probe sets were significantly upregulated, and 6144 were significantly downregulated. Among upregulated genes, 241 probe sets were common in both TCMR and PVAN, and among downregulated genes, 332 probe sets were common in both TCMR and PVAN tissues.

Using PCA and taking into account the 150 top ranked probe sets based on an FDR-adjusted P value (P < 0.05 and 2-fold increase expression), a significant clustering of genes was observed among TCMR and STA as well as PVAN and STA. PVAN and TCMR samples clustered together when all 3 sample types were used in the PCA plot (Figure S1, SDC,http://links.lww.com/TP/B269).

With the aim of identifying critical PVAN injury-specific genes and in TCMR, we increased stringency of the specificity to a P value less than 0.01 with greater than 2-fold increased expression that resulted in 209 unique genes increased in PVAN (Table S1, SDC,http://links.lww.com/TP/B269) and 252 unique genes increased in TCMR as compared with STA kidney biopsies. The TCMR-specific probe sets that were not significant in PVAN and STA individuals were basically involved in CTLA4 (CD3G, CD28, CD3E, PIK3CG, PIK3R5, CD86, CD8A) and T cell receptor (NFATC2, CD8A, LCP2, ITK) signaling in cytotoxic T lymphocytes as well as related to cellular movement (CCL4, CCR5, CCR7, FAM65B) of T lymphocytes. The 115 unique probe sets most significantly enriched in PVAN and not in TCMR and STA patients were mainly involved in DNA replication and RNA binding (BST2, EIF3G, F13A1, FGFR1, HSPD1, IFITM1, LTF, RPS15, NOSIP, and RARRES3), assembly of RNA polymerase (POLR2I, TAF10), and pathogen recognition receptors (C3, C1QA, C3AR1, and CFD). As indicated by both PCA and hierarchical ordered partitioning and collapsing hybrid gene clusterings of these PVAN-specific genes, the most optimal separation between the 30 STA, 15 TCMR, and 10 PVAN samples was achieved, relative to any of the prior gene sets (Figures 1B and C).

To determine the functional relevance of genes enriched in PVAN but not TCMR, we performed a comprehensive pathway/gene set analysis using the software GO-Elite. This analysis showed enrichment of 2 distinct set of pathways. Immune system–related pathways, such as complement cascade, TCR signaling, innate immune system, and adaptive immune system were specific to TCMR-associated genes, whereas PVAN-associated genes were enriched with DNA replication pathways, such as messenger RNA processing, ribosomal scanning, viral messenger RNA translation, and so on, and metabolic pathways, such as urea cycle, gluconeogenesis, tricarboxylic acid cycle, and so on, demonstrating 2 distinct molecular events occurring at the time of TCMR and PVAN (Figure 2).

FIGURE 2
FIGURE 2:
A comprehensive pathway/gene set analysis using the software GO-Elite was performed to identified molecular pathways enriched in PVAN compared with TCMR. This resulted in enrichment of 2 distinct set of pathways. Immune system–related pathways such as complement cascade, TCR signaling, innate immune system, adaptive immune system were specific to TCMR, whereas PVAN-associated genes were enriched with DNA replication pathways such as mRNA processing, ribosomal scanning, viral mRNA translation, and so on, and metabolic pathways, such as urea cycle, gluconeogenesis, tricarboxylic acid cycle, and so on, were specific to PVAN demonstrating 2 distinct molecular events occurring at the time of TCMR and PVAN. mRNA, messenger RNA.

qPCR Validation of Selected Genes

To validate PVAN-specific gene expression data, 4 genes (LTF, CFD, RPS15, and NOSIP) were selected for qPCR validation. As illustrated in Figure 3, overexpression of all 4 genes in PVAN versus STA and also in PVAN versus TCMR was confirmed. In an independent set of PVAN (n = 15), AR (n = 18), and STA (n = 18), fold increase in gene expression for: (i) LTF in PVAN was significant when compared with AR (P = 0.04) and STA (P = 0.02), (ii) CFD in PVAN was significant when compared to AR (P = 0.05) and STA (P = 0.05), (iii) 40 ribosomal protein S15 (RPS15) in PVAN was significant when compared with AR (P = 0.002) and STA (P = 0.02), (iv) NOSIP in PVAN was significant when compared with AR (P = 0.004) and STA (P = 0.007).

FIGURE 3
FIGURE 3:
qPCR validations of PVAN-specific genes. To validate PVAN-specific gene expression data 4 genes (LTF, CFD, RPS15, and NOSIP) were selected for qPCR validation. Overexpression of all 4 genes in PVAN versus STA and also in PVAN versus TCMR was confirmed. Gene expression for: (A) LTF in PVAN was significant when compared to AR (P = 0.04) and STA (P = 0.02), (B) CFD in PVAN was significant when compared with AR (P = 0.05) and STA (P = 0.05), (C) 40 ribosomal protein S15 (RPS15) in PVAN was significant when compared with AR (P = 0.002) and STA (P = 0.02), (D) NOSIP in PVAN was significant when compared to AR (P = 0.004) and STA (P = 0.007). The first and third quartiles are at the ends of the box, the median is indicated with a horizontal line in the interior of the box, and the maximum and minimum are at the ends of the whiskers.

Immunohistochemistry Analysis of PVAN-Associated Gene Products

Using IHC analysis, we assessed the expression of 2 gene transcripts (LTF and IFITM1) at the protein level. These transcripts were highly upregulated in kidney biopsies with PVAN as compared with TCMR and STA and have been previously reported to have antiviral properties, LTF24,25 and IFITM1 (92,30) (Figure 4A). As shown in Figure 4B, LTF and IFITM1 expressions were significantly higher in both tubuloepithelial cells and within the mononuclear cellular infiltrates in PVAN patients as compared with STA and TCMR kidney transplant recipients.

FIGURE 4
FIGURE 4:
Using IHC validation of PVAN-specific expression of LTF and IFITM1 in kidney biopsies with PVAN. We assessed the expression of 2 gene transcripts (LTF and IFITM-1) at the protein level. These transcripts were highly upregulated in kidney biopsies with PVAN as compared with TCMR and STA. A, Three representatives phenotypes from 3 representative transplant patient biopsies evaluated for the different protein stains. TCMR and STA are shown at 10× magnification and the PVAN samples are shown at 40× magnification. Colocalization of IFITM-1 with BK viral inclusions are marked with yellow arrows in the PVAN patient. LTF also localizes in proximity to the SV40 and IFITM1 stains in the renal tubule. B, Semiquantitative analysis of protein expression at TEC and mononuclear cells of patients with PVAN, TCMR, and STA. TEC, tubuloepithelial cells.

DISCUSSION

The advent of PVAN is still a major concern in kidney transplantation, because it accounts for the main cause of allograft loss. This is explained, in great part, to the rather poor understanding of the dominant mechanisms of the disease. Although it is widely accepted, the fact that recognizing PVAN as early as possible is a key factor to increase the likelihood of success5 still remains unclear, which is the best treatment approach to follow. In this study, using high-throughput microarrays analysis, we demonstrate that the inflammatory process occurring both in PVAN and TCMR, either due to a protective antiviral or an allogeneic immune response, respectively, merges in a remarkably similar transcriptional gene pattern, essentially associated with effector immune pathways of both adaptive and innate immunities.

As reported earlier by other researchers,8,12 PVAN and TCMR share important similarities at the gene expression level, further helping to explain the difficulty of differentiating both types of diseases. In this regard, Mannon et al8 evaluated target genes by qPCR in kidney allograft biopsies from recipients with PVAN, TCMR, or patients with stable allograft function and showed that despite the significantly high resemblance of gene expression between PVAN and TCMR, transcription of certain molecules associated with graft fibrosis and markers of epithelial-mesenchymal transformation were significantly higher in PVAN specimens, suggesting a higher profibrogenic transcriptional profile in PVAN than that in TCMR patients. The higher tissue chronicity in the PVAN patient samples may also explain some of the observed differences between TCMR and PVAN relating to glucose and protein metabolism genes. Furthermore, in a recent study, Lubetzky and coworkers12 evaluated gene expression in the whole blood as well as few kidney tissue allograft samples in PVAN and TCMR kidney transplant recipients. These studies have shown high levels of proinflammatory molecules in both settings, such as the interferon γ-induced chemokines CXCL9 and CXCL10,11 as well as other Th-2–induced cytokines, such as sIL-1RA, IL-3, IL-6, and sIL-6R, being particularly high in patients with high BK DNA replication.25

Our study validated 4 PVAN-specific genes by qPCR and further confirmed using IHC. This illustrates the potential functional relevance of some of the particular transcripts uniquely expressed in PVAN biopsies as compared with rejecting patients. First, LTF, a member of the transferrin family found in mucous epithelial cells and secondary granules of polymorphonuclear neutrophils was highly differentially expressed among PVAN patients. Its expression in PVAN biopsies was significantly higher both within cellular infiltrates and in tubuloepithelial cells as compared with biopsies with TCMR and STA. In fact, LTF has been shown to play a key role in the defense against various pathogenic microorganisms by inhibiting different enveloped26,27 and naked28,29 viruses in different virus cell systems. In addition, in experimental animal models, it has been shown that oral administration of LTF or peptides thereof is effective in reducing bacterial infections and inflammation in the urinary tract, possibly through transfer of LTF or its peptides to the site of infection via renal secretion.30 Of note, LTF treatment has been evaluated and shown to prevent early steps of BK virus infection in vitro, most likely through the interaction with BK viral capsidic structures.31 This finding is of relevance, because the high expression of LTF transcripts in kidney allografts infected by BK, might be the expression of a physiological protective response of the host against the virus to overcome the viral infection. Observation of increased expression of genes associated with tricarboxylic acid cycle, urea cycle, and gluconeogenesis is biologically plausible. Even though there is no report of increased metabolism in BKV infection, alteration of metabolism especially glucose metabolism and tricarboxylic acid cycle has been observed in case of cytomegalovirus infection.32 It is intuitive to assume that successful replication of virus in the infected cells requires an environment that is suitable for increased supply of nutrient, energy, and macromolecular synthesis, which is reflected in the upregulated gene expression of genes related with metabolic pathways. Complement factor D was also upregulated in PVAN, which is a serine protease that cleaves C3b-bound factor B, resulting in the generation of Bb and formation of the alternative pathway C3 convertase (C3bBb) a key system for immune surveillance and homeostasis.33 Increased expression of CFD and other complement genes shows activation of complement system at the time of viral infection. There is increased generation of nitric oxide (NO) due to viral infection, which is harmful.34 Nitric oxide synthase interacting protein is one of the genes upregulated in PVAN, negatively regulates NO production by inducing translocation of NOS1 and NOS3 to actin cytoskeleton and inhibiting their enzymatic activities.35 Furthermore, another highly upregulated gene observed within PVAN patients as compared with TCMR and stable individuals was the IFITM1 transcript. Interestingly, IFITM proteins are a family of ubiquitously expressed restriction factors that mediate potent IFN-induced antiviral activity by inhibiting viral entry, particularly the step of membrane fusion.36-38 Nevertheless, although IFITM1 antiviral activity has been well characterized against RNA viruses,37,39,40 it has been shown to induce the opposite effect, that is to enhance infection of several DNA viruses. Indeed, it has been demonstrated that the antiviral activity of IFITM proteins is likely mediated by preventing endosome fusion and viral entry into the cytosol,36-38 as well as inhibit viral entry by preventing escape from the endocytic pathway particularly among DNA virus.41 Therefore, overproduction of IFITM1 molecules in PVAN patients could be hypothesized to represent a signal of persistent DNA replication, illustrating the aggressive nature of BKV infection in kidney transplantation. Indeed the colocalization of IFITM1 in close proximity to LTF in tubular cells showing BK viral inclusions further corroborate these previously reported data (Figure 4).

In summary, even though PVAN and TCMR kidney allografts share great similarities on gene perturbation, particular PVAN-specific transcripts are differentially expressed, some of them encoding for molecules with well-known antiviral properties. Further tracking such effector molecules in the context of BK virus infection may lead to the discovery of novel potential therapeutic targets that may eventually overcome the development and persistence of BKV infection after kidney transplantation.

ACKNOWLEDGMENTS

The authors would like to acknowledge Anyou Wang and Mark Nguyen for manuscript preparation and Cristian Varela, Nuria Bolaños and Dr. Benjamin Torrejón from the SCT at UB for the methodical help with all the IHC experiments.

REFERENCES

1. Dharnidharka VR, Cherikh WS, Abbott KC. An OPTN analysis of national registry data on treatment of BK virus allograft nephropathy in the United States. Transplantation. 2009;87:1019–1026.
2. Hirsch HH, Randhawa PPractice ASTIDCo. BK polyomavirus in solid organ transplantation. Am J Transplant. 2013;13:179–188.
3. Ramos E, Drachenberg CB, Wali R, et al. The decade of polyomavirus BK-associated nephropathy: state of affairs. Transplantation. 2009;87:621–630.
4. Hirsch HH, Knowles W, Dickenmann M, et al. Prospective study of polyomavirus type BK replication and nephropathy in renal-transplant recipients. N J Engl Med. 2002;347:488–496.
5. Babel N, Volk HD, Reinke P. BK polyomavirus infection and nephropathy: the virus-immune system interplay. Nat Rev Nephrol. 2011;7:399–406.
6. Drachenberg CB, Papadimitriou JC, Hirsch HH, et al. Histological patterns of polyomavirus nephropathy: correlation with graft outcome and viral load. Am J Transplant. 2004;4:2082–2092.
7. Hirsch HH, Randhawa PPractice ASTIDCo. BK virus in solid organ transplant recipients. Am J Transplant. 2009;9:S136–146.
8. Mannon RB, Hoffmann SC, Kampen RL, et al. Molecular evaluation of BK polyomavirus nephropathy. Am J Transplant. 2005;5:2883–2893.
9. Girmanova E, Brabcova I, Klema J, et al. Molecular networks involved in the immune control of BK polyomavirus. Clin Dev Immunol. 2012;2012:972102.
10. Dadhania D, Snopkowski C, Ding R, et al. Validation of noninvasive diagnosis of BK virus nephropathy and identification of prognostic biomarkers. Transplantation. 2010;90:189–197.
11. Jackson JA, Kim EJ, Begley B, et al. Urinary chemokines CXCL9 and CXCL10 are noninvasive markers of renal allograft rejection and BK viral infection. Am J Transplant. 2011;11:2228–2234.
12. Lubetzky M, Bao Y, O Broin P, et al. Genomics of BK viremia in kidney transplant recipients. Transplantation. 2014;97:451–456.
13. Li L, Chang A, Naesens M, et al. Steroid-free immunosuppression since 1999: 129 pediatric renal transplants with sustained graft and patient benefits. Am J Transplant. 2009;9:1362–1372.
14. Sis B, Mengel M, Haas M, et al. Banff '09 meeting report: antibody mediated graft deterioration and implementation of Banff working groups. Am J Transplant. 2010;10:464–471.
15. Sar A, Worawichawong S, Benediktsson H, et al. Interobserver agreement for Polyomavirus nephropathy grading in renal allografts using the working proposal from the 10th Banff Conference on Allograft Pathology. Hum Pathol. 2011;42:2018–2024.
16. Naesens M, Khatri P, Li L, et al. Progressive histological damage in renal allografts is associated with expression of innate and adaptive immunity genes. Kidney Int. 2011;80:1364–1376.
17. Li C, Hung Wong W. Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol. 2001;2: RESEARCH0032.
18. Ieni A, Barresi V, Branca G, et al. Immunoexpression of lactoferrin in bone metastases and corresponding primary carcinomas. Oncology letters. 2013;5:1536–1540.
19. Parra-Herran CE, Yuan L, Nucci MR, et al. Targeted development of specific biomarkers of endometrial stromal cell differentiation using bioinformatics: the IFITM1 model. Mod Pathol. 2014;27:569–579.
20. Schmid H, Burg M, Kretzler M, et al. BK virus associated nephropathy in native kidneys of a heart allograft recipient. Am J Transplant. 2005;5:1562–1568.
21. Emig D, Salomonis N, Baumbach J, et al. AltAnalyze and DomainGraph: analyzing and visualizing exon expression data. Nucleic Acids Res. 2010;38:W755–762.
22. Dahlquist KD, Salomonis N, Vranizan K, et al. GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways. Nat Genet. 2002;31:19–20.
23. Salomonis N, Hanspers K, Zambon AC, et al. GenMAPP 2: new features and resources for pathway analysis. BMC Bioinformatics. 2007;8:217.
24. Zambon AC, Gaj S, Ho I, et al. GO-Elite: a flexible solution for pathway and ontology over-representation. Bioinformatics. 2012;28:2209–2210.
25. Sadeghi M, Daniel V, Schnitzler P, et al. Urinary proinflammatory cytokine response in renal transplant recipients with polyomavirus BK viruria. Transplantation. 2009;88:1109–1116.
26. Andersen JH, Jenssen H, Gutteberg TJ. Lactoferrin and lactoferricin inhibit Herpes simplex 1 and 2 infection and exhibit synergy when combined with acyclovir. Antiviral Res. 2003;58:209–215.
27. Andersen JH, Jenssen H, Sandvik K, et al. Anti-HSV activity of lactoferrin and lactoferricin is dependent on the presence of heparan sulphate at the cell surface. J Med Virol. 2004;74:262–271.
28. Drobni P, Naslund J, Evander M. Lactoferrin inhibits human papillomavirus binding and uptake in vitro. Antiviral Res. 2004;64:63–68.
29. Pietrantoni A, Di Biase AM, Tinari A, et al. Bovine lactoferrin inhibits adenovirus infection by interacting with viral structural polypeptides. Antimicrob Agents Chemother. 2003;47:2688–2691.
30. Haversen LA, Engberg I, Baltzer L, et al. Human lactoferrin and peptides derived from a surface-exposed helical region reduce experimental Escherichia coli urinary tract infection in mice. Infect Immun. 2000;68:5816–5823.
31. Longhi G, Pietropaolo V, Mischitelli M, et al. Lactoferrin inhibits early steps of human BK polyomavirus infection. Antiviral Res. 2006;72:145–152.
32. Yu Y, Clippinger AJ, Alwine JC. Viral effects on metabolism: changes in glucose and glutamine utilization during human cytomegalovirus infection. Trends Microbiol. 2011;19:360–367.
33. Ricklin D, Hajishengallis G, Yang K, et al. Complement: a key system for immune surveillance and homeostasis. Nat Immunol. 2010;11:785–797.
34. Akaike T, Maeda H. Nitric oxide and virus infection. Immunology. 2000;101:300–308.
35. Dedio J, Konig P, Wohlfart P, et al. NOSIP, a novel modulator of endothelial nitric oxide synthase activity. FASEB J. 2001;15:79–89.
36. Diamond MS, Farzan M. The broad-spectrum antiviral functions of IFIT and IFITM proteins. Nat Rev Immunol. 2013;13:46–57.
37. Brass AL, Huang IC, Benita Y, et al. The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus. Cell. 2009;139:1243–1254.
38. Perreira JM, Chin CR, Feeley EM, et al. IFITMs restrict the replication of multiple pathogenic viruses. J Mol Biol. 2013;425:4937–4955.
39. Huang IC, Bailey CC, Weyer JL, et al. Distinct patterns of IFITM-mediated restriction of filoviruses, SARS coronavirus, and influenza A virus. PLoS Pathog. 2011;7:e1001258.
40. Raychoudhuri A, Shrivastava S, Steele R, et al. ISG56 and IFITM1 proteins inhibit hepatitis C virus replication. J Virol. 2011;85:12881–12889.
41. Warren CJ, Griffin LM, Little AS, et al. The antiviral restriction factors IFITM1, 2 and 3 do not inhibit infection of human papillomavirus, cytomegalovirus and adenovirus. PLoS One. 2014;9:e96579.

Supplemental Digital Content

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.