Detection of human neurotropic JCPyV DNA sequence in pediatric anaplastic xanthoastrocytoma

JC polyomavirus (JCPyV) is a ubiquitous human virus isolated in 1971 from the brain of a patient with Hodgkin disease (Padgett et al. 1971). It is the etiological agent of the progressive multifocal leukoencephalopathy (PML), a demyelinating disease of the brain, caused by lytic infection of oligodendrocytes upon viral reactivation (Pietropaolo et al. 2018). The viral genome comprises the early and late gene region and the non-coding control region (NCCR). The early region encodes for nonstructural proteins, large T antigen (LTAg), small t antigen (stAg), and T’135, T’136, and T’165 proteins involved in the regulation of the virus cycle and in cell transformation (Frisque et al. 1984; Khalili 2001). The N-terminal region of LTAg can interact with members of the retinoblastoma (Rb) protein family, whereas the C-terminal domain can bind p53 (Zheng et al. 2022). The interaction with these tumor suppressors induces progression of the cell cycle and is a major feature of the oncogenic properties (Del Valle et al. 2001). The late region encodes for the capsid proteins VP1, VP2, and VP3, for the agnoprotein (Frisque et al. 1984), and for two mature microRNAs (miRNAs), miR-J1-3p and miR-J1-5p, which can modulate viral replication by downregulating LTAg expression (Giovannelli et al. 2015). The NCCR, encompassing the origin of replication and transcription control sequences, has a hypervariable structure that contributes to neurotropism and neurovirulent properties of JCPyV (Pietropaolo et al. 2018). The archetype NCCR structure has been isolated in the kidney and urine from healthy subjects (Yogo et al. 1990) whereas the rearranged or “prototype” strain is frequently disease-associated (Frisque et al. 1984). JCPyV-mediated oncogenesis has been described both in in vitro and in vivo studies (White and Khalili 2005). Inoculation of JCPyV into rodents, owl, or squirrel monkeys produces central nervous system (CNS) tumors including medulloblastoma, astrocytoma, glioblastoma, and neuroblastoma (Khalili 2001; Del Valle and Khalili 2021). JCPyV DNA was also detected in human brain tumors such as glioblastomas, astrocytomas, and medulloblastomas (Ahye et al. 2020).

In 1998, JCPyV DNA with a Mad-4 type NCCR was detected in the brain tissue of a 9-year-old child with pleomorphic xanthoastrocytoma (PXA) (Boldorini et al. 1998). So far, no additional cases of JCPyV-positive PXA have been reported.

An 11-year-old child (Table 1) was admitted to the hospital in January 2022 with a right temporo-parietal mass on magnetic resonance scan images. Pathology slides revealed an anaplastic PXA (WHO grade III). Increased cellularity, mitotic index (7 mitoses/10 hpf), and necrotic foci were evident. Anaplastic features, such as big nucleolus and an abundant eosinophilic cytoplasm absorbed by a fibrillary stroma, were displayed by the cells, and some of these showed big and pleomorphic nuclei, nuclear pseudo inclusions, and foamy cytoplasm. Pseudo-papillary and fascicular growth areas and eosinophilic granular bodies were also observed.

Immunohistochemical analysis showed diffuse positivity for glial fibrillary acid protein and patchy positivity for synaptophysin; however, positivity for mutated BRAFV600E protein has also been demonstrated (Fig. 1).

Total DNA was extracted from the paraffin-embedded tissue sections using Quick-DNA FFPE Miniprep (Zymo Research, Irvine, CA) and was evaluated for its PCR suitability by amplifying the β-globin gene sequences. To detect the presence of JCPyV DNA, a quantitative PCR (qPCR), targeting LTAg DNA sequence, was performed. JCPyV DNA was further subjected to nested PCR (nPCR) using primers mapping sequences encoding the N- and C-terminal region of LTAg, the NCCR, and the VP1 gene (Flaegstad et al. 1991; Krynska et al. 1999). Positive PCR products were purified using miPCR purification kit (Metabion, Planegg, Germany) and sequenced in a dedicated facility (Bio-Fab Research, Roma, Italy). In addition, the expression of transcripts from JCPyV LTAg and VP1 genes was investigated. Total RNA was extracted using Quick-RNA Miniprep Plus Kit (Zymo Research, Irvine, CA), reverse transcribed by ZymoScript RT PreMix Kit (Zymo Research, Irvine, CA), and used for a PCR carried out with the same primers used on DNA. In order to investigate JCPyV miRNA expression, the reverse-transcribed RNA was used for PCR amplification of miR-J1-5p coding region (Giovannelli et al. 2015). DNA and cDNA were further analyzed for cellular p53 by PCR (Malekpour Afshar et al. 2016). JCPyV DNA was detected with a viral load value of 6.0 × 10⁴ gEq/mL. nPCR for LTAg region gave a positive result for the 5ʹ end, whereas sequences encoding the C-terminal region and the VP1 gene were undetectable. Only LTAg transcripts of 5ʹ were identified. The NCCR had the archetype A-B-C-D-E–F box arrangement. miRNAs’ investigation showed no miR-J1-5p expression. Moreover, p53 analysis showed a negative result at both DNA and RNA level as well as by immunohistochemical analysis.

Viral sequences have been detected in human brain tissue corroborating the hypothesis that JCPyV could be involved in the development of human brain tumors (Ahye et al. 2020). In our study, qPCR showed that PXA, a rare brain tumor with histopathological features resembling PML, harbored JCPyV DNA. To define JCPyV as infectious agent associated with brain cancer, JCPyV DNA positivity alone is not sufficiently specific to establish its etiological role. For this reason, in this study the expression of the LTAg and VP1 at the DNA and RNA level was investigated.

In 1998, Boldorini et al. identified JCPyV LTAg and VP1 DNA sequences, but not virus particles in the brain tissue of a 9-year-old boy with PXA (Boldorini et al. 1998). In our case, we detected DNA sequences and transcripts, corresponding to the 5ʹ end but not the 3ʹ end of the LTAg gene and VP1 was undetectable both at DNA and RNA level.

During viral replication, JCPyV displays an orderly gene expression cascade in which LTAg transcript is expressed first followed by the expression of the VP1 gene. Loss of the viral replication capacity is a common feature of virus-associated tumors. In this case, the hampered viral replication could explain the expression of LTAg but not VP1 gene.

Since NCCR and miRNA could represent independent modalities of regulating JCPyV replication at the transcriptional and post-transcriptional levels, in this study, NCCR architecture and miRNA expression were investigated.

We failed to detect p53 at the DNA, RNA, and protein level in our tumor specimen. As previously described, the C-terminal region of LTAg can interact with p53 (Zheng et al. 2022) although inactivation of p53 is not absolutely required for polyomaviruses to induce cancer as shown for MCPyV and MCC. In fact, the LTAg expressed in MCPyV-positive MCC lacks the p53 binding domain but retains the Rb domain (DeCaprio 2021), as observed in our case. Therefore, further studies are required to establish whether the genesis of PXA could depend on the transformation capacity of LTAg by Rb sequestration, as demonstrated in the MCC. Additional studies are required (i) to establish whether full-length or C-terminal truncated LTAg is expressed; (ii) to establish whether the truncation is the result of a deletion in the LTAg gene rather than nonsense mutation (as is the case in the LTAG gene of MCPyV-positive MCCs); and (iii) to explain the lack of p53 DNA detection in the tumor.

JCPyV detection in brain tumor tissue reinforces the idea that this virus could be involved in tumor development. However, further studied are warranted to define its role in oncogenesis of brain tumors.