# Environmental chemicals impair epigenetic suppression of the endogenous retrovirus HML-2 in human primordial germ cells, predisposing the next generation to malignancies through HML-2 reactivation

> **NIH NIH R01** · MASSACHUSETTS GENERAL HOSPITAL · 2024 · $620,940

## Abstract

Evidence supporting the epigenetic inheritance of diseases in mammals is accumulating, and the endogenous
retrovirus (ERV) may play as a heritable epigenetic vehicle. HML-2, the ERV species most recently integrated
into the human genome, is epigenetically silenced in normal cells, but our lab and others recently revealed that
HML-2 is transiently activated in human primordial germ cells (PGCs) but silenced as PGCs differentiate to
sex-specific germ cells. In contrast, HML-2 is often activated in common cancers (e.g., breast cancer, prostate
cancer, or melanomas), and two HML-2 viral proteins are presumed oncogenic, supporting a growing paradigm
that HML-2 reactivation may contribute to human carcinogenesis. We hypothesize that the epigenetic silencing
of HML-2 during the post-PGC development of human germ cells may be impaired by environmental toxicants.
The affected, incompletely silenced HML-2 loci may be vulnerable to reactivation, increasing the cancer risk in
the next generation. Our mechanistic reasoning of this hypothesis is that assembly of protein complexes
required for formation of heterochromatin – the main epigenetic machinery silencing ERVs – is known to be
disrupted by transcriptional activation in its vicinity. Therefore, transactivation and/or DNA demethylation by
toxic chemicals may impair epigenetic silencing of HML-2. HML-2 is strongly activated in pluripotent stem cell
(PSC)-derived PGC-Like Cells (PGCLCs). In the xenogeneic reconstituted testis (xrTestis) organoid culture
involving human PGCLCs and mouse embryonic testicular somatic cells, PGCLCs differentiate to gonocyte-
like cells (GnCLCs), in which all HML-2 loci are efficiently silenced. Our Specific Aim 1 will expose xrTestes to
toxicants (PFOA, dieldrin and benzo[a]pyrene), and the transcriptional and epigenetic states of all HML-2 loci
in GnCLCs will be examined by bulk and single-cell deep sequencing. Degrees of relaxation in epigenetic
silencing will be evaluated under pharmacologically induced, modest HML-2 reactivation using inhibitors of
DNA methyltransferase 1 or H3K9 methyltransferases. Also using deep sequencing, Specific Aim 2 will
determine whether impaired epigenetic silencing of HML-2 is repaired during the global epigenetic
reprogramming in human PSCs and PGCLCs. Epigenetic silencing of selected HML-2 loci will be specifically
relaxed by CRISPRa. Specific Aim 3 will create a humanized mouse model harboring the HML-2 locus
associating the PRODH gene in the mouse genome safe-harbor locus Rosa26 and examine its epigenetic
states and activation in mouse PGCs, gonocytes, and somatic cells muti-generationally after CRISPRa
activation of the HML-2 locus or gestational exposures to toxicants. Specific Aim4 will perform computational
analysis of RNA-seq data retrieved from databases to identify similarities and differences in the HML-2 loci
activated in various human cancers and PGCLCs. The outcome of this project may create a novel paradigm
focusing on the toxico...

## Key facts

- **NIH application ID:** 10890339
- **Project number:** 1R01ES035401-01A1
- **Recipient organization:** MASSACHUSETTS GENERAL HOSPITAL
- **Principal Investigator:** TOSHIHIRO SHIODA
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $620,940
- **Award type:** 1
- **Project period:** 2024-04-24 → 2029-02-28

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10890339

## Citation

> US National Institutes of Health, RePORTER application 10890339, Environmental chemicals impair epigenetic suppression of the endogenous retrovirus HML-2 in human primordial germ cells, predisposing the next generation to malignancies through HML-2 reactivation (1R01ES035401-01A1). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/10890339. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*