INSIGHTS | PERSPECTIVES
GRAPHIC: KELLIE HOLOSKI/SCIENCEscience.org SCIENCEnate immune responses, triggering nucleic
acid–sensing pattern recognition receptors
and interferon signaling, much like infec-
tion by an exogenous virus. Depending
on the type of transposon and cell con-
text, these innate immune responses
may be mediated by DNA damage, by
the activity of retrotransposon-encoded
reverse transcriptase, or by the presence
of self-hybridizing double-stranded RNAs
(dsRNAs). The resulting so-called “viral
mimicry” can curtail tumor growth through
cell autonomous mechanisms (inducing
cell cycle arrest or limiting protein transla-
tion) or by prompting clearance of the cells
by the immune system. Thus, in the early
stages of transformation, epigenetic activa-
tion of transposons may prevent or restrain
tumor development ( 9 ). In fully developed
cancers, transposon activity may promote
genome evolution and the emergence of
tumor cell clones with particular charac-
teristics. Additionally, malignancies could
be at a “tipping point,” at which enhanc-
ing transposition is unsustainable or limits
tumor growth under particular conditions.
Epigenetic agents to enhance transposon
expression, for example, may enhance
responses to immunotherapies or create
molecular dependencies ( 10 – 12 ). To the
extent that transposons can be leveraged
to incite DNA damage, this strategy may
sensitize cancers to genotoxic agents or to
DNA repair inhibitors ( 8 ).
Beyond cancer, transposon-induced in-
nate immune signaling or DNA damage is
implicated in aging and various diseases.
In aging, reduced expression of the 3 9 exo-
nuclease TREX1 (three-prime repair exo-
nuclease 1) and up-regulated activity of
LINE-1 ORF2p together may promote ac-
cumulation of cytoplasmic cDNA, inducing
senescence-associated interferon signal-
ing ( 13 ). Mutations compromising TREX1
in Aicardi-Goutières syndrome (AGS) may
have an analogous interaction with trans-
posons that fuels pathologic interferon
signaling and encephalopathy ( 14 ). This
hypothesis led to ongoing clinical trials
of reverse transcriptase inhibitors in AGS.
Reverse transcriptases are one of several
protein types encoded by the “noncoding”
genome, and others may be relevant to
disease. For example, ERV expression has
been associated with the neurodegenera-
tive disease amyotrophic lateral sclerosis,and expression of an ERV-encoded env
protein in mice causes DNA damage and
motor neuron degeneration through un-
known mechanisms ( 15 ). The hypothesis
that retroelement expression is pathogenic
is now being pursued for several neurode-
generative diseases.
There are a host of technical needs to
better understand the role of transposons
in disease. Interspersed repeats are vari-
able and exist in high copy numbers in the
genome, and they are pervasively incorpo-
rated in pre-mRNAs and long noncoding
RNAs (lncRNAs). Therefore, finding DNA
insertion variants or differentially ex-
pressed loci in sequencing datasets is chal-
lenging. Reference genomes and genotyp-
ing tools are needed that capture heritable
variants in active transposons. Single-cell
sequencing approaches are needed to in-
dicate de novo insertions and associated
genome rearrangements, to reveal expres-
sion of specific transposon loci, and to pro-
file repeat-containing cDNA and dsRNA
species. Reagents to study this biology in
diseased tissues and in clinical specimens
will also be important. Most nucleic acid
probes and primers designed against a re-
petitive sequence will indiscriminately hy-
bridize with unintended targets that exist
throughout the genome. If overexpression
of a transposable element is technically
straightforward to recreate in experimen-
tal models and sufficient to cause disease,
tools are needed to ascertain whether it is
a necessary contributor. Removing retroel-
ements from a disease model is challeng-
ing, especially because they are often spe-
cies-specific and exist among many highly
homologous sequences. Models that use
human tissue, locus-specific genetic ap-
proaches, and pharmacologic agents that
precisely target retroelements are needed
for functional studies and to set the stage
for our bench-to-bedside aspirations. jR EFERENCES AND NOTES- H. H. Kazazian Jr. et al., Nature 332 , 164 (1988).
- C. R. Beck et al., Cell 141 , 1159 (2010).
- G. D. Evrony et al., eLife 5 , e12966 (2016).
- N. Rodić et al., Am. J. Pathol. 184 , 1280 (2014).
- B. Tiwari et al., Genes Dev. 34 , 1439 (2020).
- E. C. Scott et al., Genome Res. 26 , 745 (2016).
- B. Rodriguez-Martin et al., Nat. Genet. 52 , 306 (2020).
- D. Ardeljan et al., Nat. Struct. Mol. Biol. 27 , 168 (2020).
- Y. Zhao et al., Nat. Immunol. 22 , 1219 (2021).
- D. Roulois et al., Cell 162 , 961 (2015).
- K. B. Chiappinelli et al., Cell 162 , 974 ( 2 0 1 5 ).
- P. Mehdipour et al., Nature 588 , 169 (2020).
- M. De Cecco et al., Nature 566 , 73 (2019).
- C. A. Thomas et al., Cell Stem Cell 21 , 319 (2017).
1 5. W. L i et al., Sci. Transl. Med. 7 , 307ra153 (2015).
ACKNOWLEDGMENTS
K. H.B. is supported by the National Institutes of Health
(R01GM130680 and R01CA240816). She is a Scientific
Advisory Board member of Transposon Therapeutics and the
scientific cofounder of Oncolinea Pharmaceuticals.
10.1126/science.abl7399PRRsAltered expression of nearby genesInnate immune responses and DNA damageCytosolic
DNADNA damage
and mutationSelf-hybridizing
TE RNARNAInnate
immune
signalingTE RNA dsRNATranslation of
TE-encoded
proteinReverse
transcriptaseRibonucleic
protein complexRetrotranspositioncDNATENucleus CytosolCellular effects of transposable element dysregulation
Transposable element (TE) activation can directly regulate gene expression when the locus encodes
a regional promoter or enhancer (top). TE dysregulation can also have cellular effects (bottom); for example,
innate immune pattern recognition receptors (PRRs) can be activated by DNA damage or cDNA produced
by retroelement proteins, or by self-hybridizing double-stranded RNA (dsRNA).
354 22 APRIL 2022 • VOL 376 ISSUE 6591