Deep-Sequence Identification and Role in Virus Replication of a JC Virus Quasispecies in Patients with Progressive Multifocal Leukoencephalopathy

Abstract

ABSTRACT

JC virus (JCV) is a DNA virus causing progressive multifocal leukoencephalopathy (PML) in immunodeficient patients. In the present study, 22 genetic quasispecies with more than 1.5% variant frequency were detected in JCV genomes from six clinical samples of PML by next-generation sequencing. A mutation from A to C at nucleotide (nt) 3495 in JCV Mad1 resulting in a V-to-G amino acid substitution at amino acid (aa) position 392 of the large T antigen (TAg) was identified in all six cases of PML at 3% to 19% variant frequencies. Transfection of JCV Mad1 DNA possessing the V392G substitution in TAg into IMR-32 and human embryonic kidney 293 (HEK293) cells resulted in dramatically decreased production of JCV-encoded proteins. The virus DNA copy number was also reduced in supernatants of the mutant virus-transfected cells. Transfection of the IMR-32 and HEK293 cells with a virus genome containing a revertant mutation recovered viral production and protein expression. Cotransfection with equal amounts of wild-type genome and mutated JCV genome did not reduce the expression of viral proteins or viral replication, suggesting that the mutation did not have any dominant-negative function. Finally, immunohistochemistry demonstrated that TAg was expressed in all six pathological samples in which the quasispecies were detected. In conclusion, the V392G amino acid substitution in TAg identified frequently in PML lesions has a function in suppressing JCV replication, but the frequency of the mutation was restricted and its role in PML lesions was limited.

IMPORTANCE

DNA viruses generally have lower mutation frequency than RNA viruses, and the detection of quasispecies in JCV has rarely been reported. In the present study, a next-generation sequencer identified a JCV quasispecies with an amino acid substitution in the T antigen in patients with PML. In vitro studies showed that the mutation strongly repressed the expression of JC viral proteins and reduced the viral replication. However, because the frequency of the mutation was low in each case, the total expression of virus proteins was sustained in vivo. Thus, JC virus replicates in PML lesions in the presence of a mutant virus which is able to repress virus replication.

INTRODUCTION

The JC virus (JCV) is a DNA virus of the human polyomavirus family (1–3). Serum antibodies to JCV are found in 51% to 76% of a healthy general population, indicating that JCV is ubiquitous in humans (4–7). The primary infection of JCV occurs asymptomatically during infancy, with latent infections occurring in the kidney, lymphocyte, spleen, and bone marrow. JCV spreads hematogenously to the central nervous system, or remains latent in the brain during primary viremia, and reactivates locally, causing progressive multifocal leukoencephalopathy (PML), a fatal demyelinating disease, in immunocompromised patients such as AIDS sufferers or patients receiving immunomodulatory therapies (8–10). In the PML lesion, remarkable demyelination is observed histologically in the white matter, accompanying enlarged oligodendroglial nuclei and bizarre astrocytes with nuclear atypia (11, 12). JCV is detected in the enlarged nucleus of oligodendroglia by immunohistochemistry (13, 14).

JCV encodes a large T antigen (TAg), a small t antigen, three virus capsid proteins (VP1 to -3), and an agnoprotein in its genome (15). JCV-encoded TAg plays an important role in virus replication that is similar to the role of the TAg encoded by other polyomaviruses (16, 17). TAg binds to the viral origin of replication through a DNA binding domain, and the nuclear localization signal in TAg recruits TAg to the nucleus (17). The helicase domain in TAg unwinds the DNA double helix and promotes viral DNA replication (18).

DNA viruses, including JCV, generally have fewer mutations than RNA viruses. JCV isolates from PML patients have unique mutations, deletions, and/or rearrangements (so called PML-type rearrangements) in the regulatory region; in addition, >90% of JCV isolates from PML patients carry point mutations in VP1 (19–23). Rearrangement of the regulatory region in polyomavirus has been detected frequently in patients with polyomavirus infection, including PML (10). The JCV genome with the archetype regulatory region undergoes insertion and deletion errors during replication, generating rearranged regulatory region genomes. Polyomaviruses with a rearranged regulatory region have increased levels of early gene expression and higher replication capacity, resulting in elimination of polyomaviruses with the archetype regulatory region from the PML lesion in immunodeficient hosts (21, 24). Mutations in the agnoprotein have been identified also in PML-variant JCV encephalopathy (25). Thus, mutation, deletion, and rearrangement are sometimes observed in JCV genomes isolated from patients with PML; however, the details concerning the frequency of quasispecies and mutations in the JCV genome in individuals have not been reported to date. In the present study, next-generation sequencing (NGS) was performed on pathological samples from PML lesions, which identified common quasispecies in the JCV genome among PML cases. In addition, the role of TAg in the replication of the common quasispecies was investigated.

RESULTS

JCV detection and full sequence determination in each PML case.

JCV DNA was quantified in the six PML samples. High numbers of JCV DNA ranging from 88 to 9,562 copies per cell were detected (Table 1). PCR and direct-sequence analysis revealed that the JCV genomes in all six PML cases had PML-type regulatory regions (Fig. 1A). NGS was performed on DNA extracted from pathological samples from the six PML cases. The full-length JCV genomes were constructed from the sequences of the regulatory region and the NGS contigs in each case. Phylogenetic tree analysis based on whole-genome sequences demonstrated that all six cases belonged to the B CY or the B MY cluster of JCV, each of which is a major cluster in Japan (Fig. 1A). NGS reads were mapped on each full-length genome. Finally, a minimum depth of 100 reads was completed in more than 95% of the JCV genome in all samples (Fig. 2A). Since low levels of coverage were observed at nucleotide (nt) 3830 to 4100 of the JCV genome in 5 cases, a fragment covering the region was amplified with a standard PCR and subjected to direct sequencing. The fragment was amplified at an expected size in all 6 cases (Fig. 2B), and no mutation or sequence variation was detected by direct sequencing.

TABLE 1.

List of patients and samples^a

Patient	Age (yr)	Sex	Background	No. of JCV copies/μg DNA	No. of beta-actin copies/μg DNA	No. of JCV copies/cell	Total no. of reads	No. of JCV reads	% of JCV reads
1	39	m	Hyper-IgM	3.60 × 108	6.18 × 105	1,165	2,859,446	27,039	0.95
2	59	f	Dermatomyositis	1.07 × 108	4.77 × 105	449	4,869,483	8,294	0.17
3	22	m	IgA def	2.47 × 109	1.54 × 106	3,803	7,360,833	271,119	3.68
4	78	m	DLBCL	1.55 × 109	5.56 × 106	558	12,985,497	190,136	1.46
5	63	m	CLL	2.62 × 109	5.48 × 105	9,562	6,456,731	221,856	3.44
6	62	m	AIDS	5.13 × 108	1.17 × 107	88	38,182,880	190,038	0.50

^aCLL, chronic lymphocytic leukemia; DLBCL, diffuse large B cell lymphoma; IgA def, IgA deficiency; f, female; m, male.

FIG 1.

Typing of the JCV genome in PML patients. (A) Recombination of the regulatory region of JCV in cases 1 to 6. (B) Phylogenetic tree of the whole genome of JCV. The samples of cases 1 to 6 are indicated in red. The genotypes of the JCV samples and the GenBank accession numbers of the reference strains are as follows: AB077870 (B SC), AB362351 (B 2E), AB077855 (B SC), AB048577 (B MY), AB092584 (B MX), AB118654 (B CY), AB113145 (B B3a), AB048557 (B B2), AB372037 (B B1-d), AB372038 (B B1-c), AB262402 (B B1-b), AB113144 (B B1-a), AB127013 (B Af2), AB038253 (C Af1), AB074575 (A EU-c), AB048563 (A EU-b), AB074580 (A EU-a), AB127349 (A EU-a), AB127348 (A EU-a), AB038254 (Tky-1), U61771 (Taiwan-3), and J02226 (Mad1).

FIG 2.

Coverage and variant frequencies in the JCV genome determined by NGS. (A) Coverage and variant frequencies. Coverages are shown in light blue, and variant frequencies are indicated by red. The transcripts encoded by JCV are shown by blue arrows at the bottom of the panels. (B) PCR amplification of the region with low coverage. A DNA fragment at JCV nt 3816 to 4199 of TAg was amplified in all six cases and electrophoresed. JCV region nt 1871 to 2254 of VP1 was also amplified as a positive control for JCV DNA. The amplified regions are indicated by green and red bars at bottom left of panel A. Predicted sizes of amplicons are indicated at the right. Agno, agnoprotein gene; Reg., regulatory region; P, positive control (JCV-Mad1 strain plasmid); N, negative control (no DNA).

Identification of JCV quasispecies from PML patients.

Mapping of the NGS reads to the full-length JCV genome identified 22 quasispecies which showed more than 1.5% variant frequency with respect to the JCV genome (Table 2 and Fig. 3A). Among the 22 quasispecies, A³⁴⁹⁵ to C³⁴⁹⁵ mutations were found in all six cases. G¹⁶⁶⁴ to C¹⁶⁶⁴ mutations were found in two cases, whereas the other 20 quasispecies were observed in a single case. The mutations of G²²⁶¹ to A²²⁶¹ and C²²⁷⁴ to T²²⁷⁴ result in amino acid substitutions from asparagine²⁶⁴ to aspartic acid²⁶⁴ and serine²⁶⁸ to cysteine²⁶⁸ in the JCV oligosaccharide-binding sites of VP1, respectively (23, 26). The variant frequency of A³⁴⁹⁵ to C³⁴⁹⁵ ranged from 3.29% to 19.68% in the six cases. The mutation of A to C at this position resulted in a single amino acid substitution from valine³⁹² to glycine³⁹² in TAg immediately upstream of the helicase domain (Fig. 3B). The valine at position 392 in the JCV TAg is highly conserved among the other 13 human polyomaviruses except KI polyomavirus (Fig. 3C).

TABLE 2.

Quasispecies in JCV genomes isolated from PML lesions^a

Coding	Position in J02226	Reference sequence	Variant	Variant frequency (%)
				Case 1	Case 2	Case 3	Case 4	Case 5	Case 6
No coding	14	G	t						5.06
No coding	61	T	acg					52.70
agno	281	T	g						1.62
agno	390	C	t		3.47
agno	403	G	t						2.76
agno	406	S	c	36.46
agno	429	A	g		3.65
agno	458	G	c		4.05
VP2	661	C	g		6.19
VP1	1664	G	c	29.58			22.79
VP1	1865	G	a	15.31
VP1	2261	G	a					7.86
VP1	2262	A	c				17.31
VP1	2268	A	c						11.81
VP1	2274	C	t						2.68
VP1	2317	A	c		3.73
VP1	2446	T	g						3.83
TAg	2804	G	c						13.06
TAg	3299	T	c		3.73
TAg	3495	A	c	3.39	3.76	3.29	3.73	3.46	19.68
TAg/ST	4534	C	g		5.06
No coding	5119	C	a						2.01

^aBold data indicate a common quasispecies among the PML cases. agno, agnoprotein.

FIG 3.

Map of variant frequencies in the JCV genome and TAg. (A) Map of the JCV genome. Variant frequencies are shown by blue rods in the upper graph. A quasispecies with a mutation of A to C at nt 3495 in JCV Mad1 is shown by the red arrow. Transcripts encoded by JCV are shown by light blue arrows at the bottom of the panel. (B) Map of TAg. The V392G mutation is upstream of the helicase domain. (C) Alignment of large T antigens encoded by human polyomaviruses. The red arrow indicates the JCV valine at position 392. The accession codes for the amino acid sequences used in the alignment, with the corresponding polyomavirus names in parentheses, are as follows: P03070 (SV40), P03071 (BKV), A3R4N4 (KI polyomavirus [KIV]), P03072 (JCV), A5HBG1 (WU virus [WUV]), AER35104 (Merkel cell polyomavirus [MCV]), BAO09095 (Trichodysplasia spinulosa polyomavirus [TSV]), ADE45440 (human polyomavirus 6 [HPyV6]), ADE45455 (HPyV7), ADV15633 (HPyV9), AFN43007 (HPyV10), AGL07668.1 (MW polyomavirus [MWV]), AFS65330 (MX polyomavirus [MXV]), AGH58117 (HPyV12), and AGC03170.1 (STL polyomavirus [STLPyV]).

Repression of JCV replication by the V392G mutation in TAg and its limited role.

To investigate the biological function of the V392G amino acid substitution in TAg, the wild-type JCV genome and the JCV genome with the V392G mutation and its revertant genome were each transfected into human neuroblastoma IMR-32 and human embryonic kidney 293 (HEK293) cells, and cell lysates were collected 4 days later. Immunoblot analysis revealed that expression of both early and late JCV proteins was strongly repressed in the cells transfected with the mutant JCV genome. The expression of these proteins completely recovered in cells transfected with the revertant JCV genome (Fig. 4A). Real-time PCR analysis revealed that the viral copy number was significantly decreased in the supernatant of IMR-32 and HEK293 cells transfected with the mutated JCV genome and that the copy numbers recovered in the cells transfected with revertant genome (Fig. 4A). It has been demonstrated that sequences of regulatory regions affect expression of early proteins encoded by polyomavirus (21, 24). Therefore, the JCV genome with the PML-type regulatory region from a patient was also transfected to IMR-32 cells in the presence or absence of the V392G amino acid substitution. Western blot analysis revealed that expression of both early and late JCV proteins was strongly repressed in the cells transfected with the V392G-mutated JCV genome with a regulatory region from case 6 (Fig. 4B). JCV copy numbers decreased in the culture supernatant of IMR-32 cells transfected with the V392G-mutated JCV genome with a regulatory region from case 6. These data suggest that the repression of TAg expression resulting from the V392G amino acid substitution was independent of sequences in the regulatory region. To evaluate the effect on TAg expression of the V392G amino acid substitution, TAg expression plasmids with or without the V392G mutation were transfected to IMR-32 cells. Western blot analysis using anti-TAg and anti-Flag antibodies demonstrated that expression of mutant TAg was much lower than that of wild-type TAg in transfected IMR-32 cells (Fig. 4C), suggesting that the V392G amino acid substitution directly caused low expression of TAg. Since NGS detected the mutant in 3% to 19% of reads in each clinical sample (Table 2), we investigated the effect of cotransfection with mutated and wild-type JCV genomes in various ratios. Western blot analysis revealed that transfection with a combination containing up to 50% of the mutated genome did not reduce expression of any JCV proteins whereas the mutant genome at a level of 100% repressed their expression (Fig. 4D). In addition, analysis of the viral copy numbers in the supernatant showed results comparable to those seen with the protein expression, indicating that contamination by the mutant genome at a level of less than 50% did not affect total JCV replication (Fig. 4D). Finally, expression of TAg was confirmed by immunohistochemistry in all the clinical samples of PML in which the quasispecies was detected (Fig. 4E), suggesting that the mutation of A to C at nt 3495 did not reduce expression of TAg and that the role of the mutation in the clinical samples was limited.

FIG 4.

The effects of the V392G mutation in the TAg of JCV on the expression of JC viral proteins and virus replication. (A) Transfection of JCV-Mad1 genome with and without a V392G mutation. Results of immunoblot analysis of the expression of the viral proteins in the JCV genome-transfected cells are shown (upper panels). JCV wild-type (WT) or V392G TAg mutant (Mut) or revertant (Rev) genome vectors were transfected into IMR-32 (left) and HEK293 (right) cells. The cell lysates were evaluated by immunoblotting using the indicated antibodies. The lower panels show results of real-time PCR assays for the detection of the JCV genome in the cultured supernatant. DNA was extracted from each cultured supernatant, and real-time PCR was performed. *, P = <0.001. (B) Transfection of JCV-Mad1 or JCV-case 6 regulatory region (RR) genome with and without V392G mutation in TAg. A JCV genome with a regulatory region from case 6 was transfected to IMR-32 cells in the presence or absence of the V392G amino acid substitution. Results of immunoblot analysis of the expression of the viral proteins in the JCV genome-transfected cells are shown (upper panels). The lower panel shows results of a real-time PCR assay for the detection of the JCV genome in the cultured supernatant. *, P = <0.001. (C) Transfection of TAg-expressing plasmid to IMR-32 cells. Equal amounts (200 ng per well) of pCXN2-Flag vector expressing wild-type TAg (pCXN2-Flag-JCV-TAg) or V392G mutant TAg (pCXN2-Flag-JCV-TAg-mut) were transfected into IMR-32 cells. TAg, Flag, and beta-actin were detected by immunoblotting. Duplicate experiments showed similar results. (D) Cotransfection with wild-type and mutated JCV vectors. JCV wild-type and mutated vectors were cotransfected into IMR-32 cells in various ratios. Cell lysates were collected and analyzed by immunoblotting (upper panels). DNA was extracted from each supernatant, and JC viral copy numbers were determined by real-time PCR (lower panels). (E) Histology of PML clinical samples from cases 3 (left) and 6 (right). Hematoxylin and eosin (HE) staining of PML lesions shows enlarged nuclei of the oligodendrocytes and atypical astrocytes in the demyelinated lesion. Positive signals for TAg and VP1 in JCV-infected cells are indicated by immunohistochemistry.

DISCUSSION

In the present study, NGS identified a JCV quasispecies with the amino acid substitution V392G in TAg in all 6 PML patients examined. Although it was difficult to detect a small population of variants in the host genome using a traditional approach such as PCR, NGS, owing to its depth, enabled us to detect a novel genomic variation (27). NGS has strongly supported the studies of viral genetic diversity, especially in RNA viruses (28, 29), whereas the detection by NGS of quasispecies in DNA virus has been reported less frequently (30–33). Using PCR analysis, the presence of VP1 quasispecies has been reported in polyomavirus BK (BKV) (34). In addition, the presence of quasispecies in the regulatory region of BKV has also been reported, with some of the quasispecies being associated with virus replication (35). JCV quasispecies have been reported in the regulatory region and in VP1 from urine samples using deep sequencing (36, 37). NGS analysis revealed that the JC viral population is often a complex mixture composed of multiple viral variants that contribute to the quasispecies in the cerebrospinal fluid (CSF) of PML patients (37). As far as we are aware, finding JCV quasispecies with a mutated TAg in the brain of patients with PML has not been reported previously.

The V392G mutation in JCV TAg was observed in all cases of PML examined in this study. This indicates that JCV replicates in PML lesions in the presence of a mutant virus which can repress virus replication. Results of a cotransfection experiment performed with wild-type and mutant JCV genomes demonstrated that the mutation does not function as a dominant negative. Intact TAg could fully recover the dysfunction induced by a small population of mutant TAg (Fig. 4D). Indeed, immunohistochemistry demonstrated that TAg was expressed in all six pathological samples in which the quasispecies was detected (Fig. 4E), implying that the role of the mutation in PML lesions was limited. Although the reason for the occurrence of such a common mutation which reduces virus replication is unknown, the frequency of the mutation is controlled by unknown factors, and the total expression of virus proteins is sustained in PML lesions. Generally, replication of polyomavirus is associated with the immune status of patients (7). Recently, immunoepitopes targeted by HLA-restricted T cells were identified in the early viral gene region of BK polyomavirus (38). Interestingly, a prediction of immunoepitopes (http://www.syfpeithi.de/bin/MHCServer.dll/EpitopePrediction.htm) identified the V392 in TAg of JCV as an immunodominant epitope with a top 20 score for HLA-A02 and -A24 nonamers, which are major HLA types in members of the Japanese population (39). It is possible that targeting by T cells might be associated with the occurrence of the common mutation in the JCV genome among patients with PML.

The mutation that we found causes an amino acid substitution of V392G in TAg of the quasispecies. After the entry of JCV into the nucleus of the host cell, TAg is transcribed and translated immediately as an early protein. TAg has DNA binding and helicase activities and stimulates transcription of late genes and suppresses that of the early genes (40). The amino acid substitution of V392G was located just upstream of the helicase domain in TAg (Fig. 3B). In our experiment performed in vitro, the V392G amino acid substitution dramatically decreased expression of both JCV-encoded early proteins and JCV-encoded late proteins (Fig. 4A and B), including TAg itself (Fig. 4C). The strong repressive function of the V392G substitution in TAg might be associated with the low expression of TAg, the ensuing production of other viral proteins, and the inhibition of viral replication. A recent study demonstrated that a single amino acid change in the C-terminal pocket of the TAg origin-binding domain impaired the role of TAg in DNA replication, suggesting that the conformational structure of TAg is crucial for its function (41). Because V392 in TAg is highly conserved among human polyomaviruses (Fig. 3C), the V392G substitution might cause a conformational change and low expression in TAg. Further studies on the structure of TAg will be required to clarify the role of the mutation in the structure, expression, and function of this protein. There are no drugs that can reduce polyomavirus replication efficiently in vivo (42–44). Since the inhibition of TAg function results in impairing virus replication (45–47), the single amino acid V392 might be a target for the inhibition of JCV replication.

MATERIALS AND METHODS

Samples.

Studies using human tissue were performed with the approval of the Institutional Review Board of the National Institute of Infectious Diseases (Approval No. 272). We examined frozen tissue from the brains of patients with PML. For NGS analysis, we examined brain samples from the six patients listed in Table 1. All patients were confirmed histologically to have PML and to be positive for JCV infection by PCR and using immunohistochemistry directed against JCV VP1. Since all specimens were obtained for pathological diagnosis, the patients did not receive any antiviral drug before the diagnosis of PML except for patient 6, who received antiretroviral therapy.

DNA extraction and real-time PCR.

DNA was extracted from unfixed frozen brain tissue using a DNeasy blood and tissue kit (Qiagen, Hilden, Germany). Real-time PCR for detecting JCV and human beta-actin DNA was performed using a standard TaqMan kit protocol (Applied Biosystems, Foster City, CA) on a Mx3005P system (Agilent Technologies, Santa Clara, CA) with previously reported probe and primer sets (48, 49). Amplicon sizes of JCV and beta-actin real-time PCR were 89 bp and 60 bp, respectively. For real-time PCR, the 20-μl PCR mixture included 1× QuantiTect probe PCR master mix (Qiagen), 0.3 μM TaqMan probe, 0.3 μM concentrations of forward and reverse primers, and approximately 100 ng of template DNA. The reaction mixtures were incubated at 95°C for 5 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The JCV DNA copy numbers per cell were calculated by dividing the JCV copy numbers by half of the beta-actin copy numbers, because each cell had two copies of the gene in two alleles (50).

PCR and sequencing.

The regulatory region of JCV was amplified by PCR as described previously (51). The fragments of nt 3830 to 4100 and nt 1871 to 2254 in the JCV genome were also amplified using JCV-Mad1-3816f (5′-TCCTCACACTTGGTTTCCAAG-3′)/JCV-Mad1-4199r (5′-TGCTTCTTTTGCTGTGTATACC-3′) and JCV-Mad1-1871f (5′-AAGCCAGTGCAGGGCACCAGCT-3′)/JCV-Mad1-2254r (5′-CATGCCACAGACATCAACAGCT-3′) primers. PCR amplification was carried out at 98°C for 30 s (one cycle), 98°C for 10 s, 55°C for 30 s, and 72°C for 30 s (35 cycles), and 72°C for 7 min (one cycle) with Phusion High-Fidelity PCR master mix (New England BioLabs, Ipswich, MA) using GeneAmp PCR System 9700 (Applied Biosystems). PCR products were purified using a QIAquick PCR purification kit (Qiagen), followed by direct sequencing with an ABI 3130 sequencer (Applied Biosystems) using a Big-Dye terminator ready reaction kit (Applied Biosystems) according to the manufacturer's instructions.

Deep sequencing by NGS.

The DNA library was prepared using a Genomic DNA Sample Prep kit (Illumina, San Diego, CA). DNA clusters were generated on a slide using a Cluster Generation kit (ver. 4) on an Illumina Cluster Station (Illumina) according to the manufacturer's instructions. The sequencing run was performed using Illumina Genome Analyzer IIx (GA IIx) with TruSeq SBS kit v5. Fluorescent images were analyzed using the Illumina RTA1.8/SCS2.8 base-calling pipeline to obtain FASTQ-formatted sequence data. Sequence reads were trimmed and assembled using VirusTAP pipeline software (https://gph.niid.go.jp/cgi-bin/virustap/index.cgi) (52). To identify the potential pathogens, sequencing reads were analyzed by MePIC2 (53). Sequence reads were analyzed using CLC Genomics Workbench (Qiagen).

Multiple alignment and phylogenetic tree analysis.

Nucleotide sequences of whole JCV genomes or TAg of 15 polyomaviruses were multiply aligned using Clustal W version 1.83 (54), and a phylogenetic tree was constructed using the neighbor-joining plot method with Genetyx software (Genetyx, Tokyo, Japan). In addition to our samples, previously reported JCV gene sequences were obtained from the GenBank database and used as reference sequences for comparisons.

Cell culture.

The IMR-32 human neuroblastoma cell line was purchased from the Health Science Research Resource Bank (Osaka, Japan). IMR-32 cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific, Rockford, IL) with 10% fetal bovine serum (FBS), 0.1 mM MEM nonessential amino acids (Thermo Fisher Scientific, Waltham, MA), and penicillin and streptomycin (Thermo Fisher Scientific). Human embryonic kidney 293 (HEK293) cells were maintained in DMEM with 5% FBS, penicillin, and streptomycin. Both cell lines were incubated in an atmosphere of 5% CO₂ at 37°C (55).

Antibodies.

A mouse monoclonal antibody to simian virus 40 (SV40) TAg (PAb416; Calbiochem, San Diego, CA) was previously confirmed to cross-react with JCV TAg (56). Rabbit polyclonal antibodies to JCV VP1, VP2/3, and agnoprotein were prepared as described previously (56, 57). A mouse monoclonal antibody to beta-actin and anti-Flag rabbit polyclonal antibody were obtained from Sigma-Aldrich (St. Louis, MO).

JCV genome recombination experiments.

The complete genome of JCV Mad1 strain (GenBank J02226), with an insertion of GGTC between nt positions 109 and 110, was subcloned into pUC19. Mutagenesis to construct the JCV Mad1 genome with a single A³⁴⁹⁵ to C³⁴⁹⁵ alteration, resulting in the TAg Val³⁹² being altered to Gly³⁹², was performed with a PrimeSTAR Mutagenesis Basal kit (TaKaRa Biotechnology, Tokyo, Japan) following the manufacturer's protocol. The sequences of the primers were as follows: forward (5′ to 3′), ATCCAGGCCCCCCCAGCCATATATTGCT; reverse (5′ to 3′), TGGGGGGGCCTGGATTCATTGCTT (the mutagenesis sites are underlined). After amplification, the plasmid was digested and self-ligated to construct a complete circular JCV Mad1 genome. A revertant JCV genome was also constructed from the mutated vector in the same way, and the sequences of the primers were as follows: forward (5′ to 3′), ATCCAGGCCACCCCAGCCATATATTGCT; reverse (5′ to 3′), TGGGGTGGCCTGGATTCATTGCTT. Both the mutated and revertant circular JCV Mad1 genomes were fully sequenced to confirm that all the sequences were identical to the wild-type JCV genome sequences except for the altered nucleotide. The JCV genome with a regulatory region from case 6 was generated by two-step recombination of the JCV Mad1 genome by the use of a KOD-mutagenesis kit (Toyobo, Osaka, Japan) using following primers: JCV-Mad1-36r (5′ to 3′, CCTTCCCTTTTTTTTATATATACAG) and JCV-niid11-53-i37-120f (5′ to 3′, TAGGGAGGAGCTGGCTAAAACTGGATGGCTGCCAGCCAAGCATGAGCTCATAATCACAAGTAAACAAAGCACAAGGGGAAGTGG) for the first step of deletion mutagenesis and JCV-niid-12-31-58r (5′ to 3′, AGTTTTAGCCAGCTCCTCCCTACCT) and JCV-niid-12-31-i59-193f (5′ to 3′, GGATGGCTGCCAGCCAAGCATGAGCTCATACCGTAAACAAAGCACAAGGGATGAGCTCATACCGTAAACAAAGCACAAGGGATGGCTGCCAGCCAAGCATGAGCTCATACCGTAAACAAAGCACAAGGGGAAGTG) for the second step of insertion mutagenesis. Recombination was confirmed by sequencing the plasmid DNA. pCXN2-Flag-JCV-TAg was used as a plasmid expressing Flag-tagged JCV TAg (58). Mutagenesis on pCXN2-Flag-JCV-TAg was performed in a manner similar to that described for the JCV Mad1 genome.

Transfection with viral genomes or plasmids.

Cells were seeded onto type I collagen-coated 24-well plates and transfected with 200 ng of the viral genomes or pCXN2-Flag-JCV-TAg using Attractene transfection reagent (Qiagen) according to the manufacturer's instructions. One day after transfection with viral genomes, transfected cells were transferred to type I collagen-coated 6-well plates and cultured for an additional 3 days. Cells transfected with pCXN2-Flag-JCV-TAg were collected 4 days after transfection.

Immunoblotting.

Cell lysates were prepared using M-PER mammalian protein extraction reagent (Thermo Fisher Scientific). After sonication and centrifugation, lysate proteins (total, 5 μl [originating from 12,500 cells in each sample]) were separated by SDS-polyacrylamide gel electrophoresis and examined by immunoblotting using anti-SV40 TAg (1:5,000), anti-VP1 (1:5,000), anti-VP2/3 (1:5,000), anti-agnoprotein (1:5,000), anti-Flag (1:5,000), and anti-beta-actin (1:5,000) antibodies. The immunopositive signals were visualized with horseradish peroxidase-conjugated secondary antibodies (Promega, Madison, WI) (1:10,000), Stable Peroxidase Solution and Luminol/Enhancer Solution (Thermo Fisher Scientific), and LAS-3000 (Fujifilm, Tokyo, Japan).

Detection of JCV DNA in culture supernatants.

Four days after transfection, the cell culture supernatants were filtered through a 0.22-μm-pore-size filter (Merck Millipore, Darmstadt, Germany). The filtered supernatants were treated with DNase (Thermo Fisher Scientific), which was then inactivated at 95°C for 5 min. DNA was extracted from the DNase-treated culture supernatants using a DNeasy blood and tissue kit (Qiagen).

Immunohistochemistry.

Immunohistochemistry was performed as described previously (58).

Statistics.

All the experiments performed in vitro were replicated at least three times. Statistical differences between experimental groups were analyzed using Student's t test.

Accession number(s).

GenBank accession numbers of JCV sequences from the clinical samples are LC164349 to LC164354. The JCV DNA sequence data analyzed by NGS in this study were deposited in the DNA Data Bank of Japan (DDBJ) (accession number DRA004832) (BioProject PRJDB4940).