SCIENCE science.orgRecently, total viral RNA was extracted
from ocean water in a harbor in China.
The researchers were able to identify more
than 4500 new viruses. Zayed et al. cast
an even wider net to identify RNA viruses
from water samples collected during the
Tara Oceans expeditions. The Tara Oceans
expeditions sailed 125,000 km across the
global ocean, sampling 210 sites through
all oceanic basins at depths down to 1000
m to get a three-dimensional picture of the
microbial diversity and ecology ( 7 ). Zayed
et al. mined 771 metatranscriptomes—the
sequences from total RNA—from different
depths at 121 different locations. Using inno-
vative bioinformatics strategies to identify
distant homologs of the RNA-directed RNA
polymerase (RdRp)—a protein found only
in orthornavirans RNA viruses—the authors
doubled the number of orthornaviran phyla
from 5 to 10. From there, they reconstructed
a robust phylogenetic tree that revealed new
insights into the evolution of RNA viruses.
Most known RNA viruses infect eukary-
otes, with very few infecting bacteria and
none infecting archaea (see the figure).
However, retroelements have been found in
both eukaryotes and prokaryotes, which in-
clude bacteria and archaea. Retroelements
are RNA genetic elements that can move to
new locations in the genome, which they
do through an RNA intermediate similar
to orthornavirans. This behavior suggests
that there is an early origin of the RdRp
( 8 ). Zayed et al. discovered a globally dis-
tributed phylum, “Taraviricota,” that pro-
vides the missing link for the evolutionary
origins of RNA viruses with regard to retro-
elements. They propose that retroelements
and Taraviricota viruses share a common
ancestor. This ancestor could be a capsid-
less RNA replicon, as opposed to viruses,
which have an outer shell called a capsid.
The presence of retroelements in all do-
mains of life but the absence of RNA viruses
in archaea suggests an important path of
evolution resulting from the separation of
cellular life from the last universal cellular
ancestor (LUCA). All cellular life shares a set
of universal genes hypothesized to have been
inherited from the LUCA, which was likely
a complex community of organisms that
shared features of both bacteria and archaea
( 9 ). It is hypothesized that the LUCA’s virome
was a complex assemblage of viruses that in-
cluded both DNA and RNA viruses, indicat-
ing that viruses had several points of origin
before the LUCA ( 9 ).
The divergence of cellular life has played
a major role in RNA virus evolution. When
cellular life evolved to include a nucleus and
an endomembrane system—both distinct
features of eukaryotes—it created a barrier
for DNA virus replication while creating a
favorable niche for RNA virus replication
( 10 , 11 ). Studying viral evolution highlights
the importance of the coevolution of viruses
with their hosts and the implications it has
in understanding the evolution of life. Viral
evolution is more complicated than simply
tracking the presence and absence of viruses
that infect the three distinct domains. In all,
there are five major branches of viruses that
are distinguished by the nucleic acid they
use for their genomes: double-stranded
DNA (dsDNA), single-stranded DNA (ss-
DNA), double-stranded RNA (dsRNA), and
positive- and negative-sense single-stranded
RNA (+ssRNA and –ssRNA, respectively),
which indicate polarity in respect to mes-
senger RNA ( 12 ). Orthornavirans, the group
of RNA viruses investigated by Zayed et al.,
include dsRNA, +ssRNA, and –ssRNA, sug-
gesting that all three share a common evo-
lutionary origin pre-LUCA.
It is difficult for multiple reasons to track
the inheritance of genes between lineages of
viruses. Gene acquisition is not linear, and
viruses can acquire genes from cellular hosts
and other viruses ( 11 ). Additionally, viral ge-
nomes have high mutation rates that lead
to rapid evolution ( 13 ). To further compli-
cate matters, viruses do not share universal
genes with highly conserved sequences and
functions, such as the ribosomal RNA genes
found in cellular life that inform the tree of
life ( 14 ). Even though many groups of viruses
share genes, this does not necessarily trans-
late to a common ancestor. There are genes
referred to as viral hallmark genes that are
shared among two or more branches of vi-
ruses, which is the case for the RdRp of or-
thornavirans. Studies such as that by Zayed
et al. create connections between viral and
cellular worlds, allowing for the possibility
of a fully integrated tree of life and a more
complete understanding of the origins and
evolution of all life. jREFERENCES AND NOTES- A. A. Zayed et al., Science 376 , 156 (2022).
- C. A. Suttle, Nat. Rev. Microbiol. 5 , 801 (2007).
- L. F. Jover, T. C. Effler, A. Buchan, S. W. Wilhelm, J. S. Weitz,
Nat. Rev. Microbiol. 12 , 519 (2014). - J. R. Brum et al., Science 348 , 1261498 (2015).
- A. C. Gregory et al., Cell 177 , 1109 (2019).
- S. Roux et al., bioRxiv 10.1101/053090 (2016).
- S. Sunagawa et al., Nat. Rev. Microbiol. 18 , 428 (2020).
- S. T. de Farias et al., Front. Genet. 8 , 125 (2017).
- E. V. Koonin, Nat. Rev. Microbiol. 1 , 127 (2003).
- M. Krupovic, V. V. Dolja, E. V. Koonin, Nat. Rev. Microbiol.
18 , 661 (2020). - E. V. Koonin et al., Microbiol. Mol. Biol. Rev. 84 , e00061
(2020). - A. E. Gorbalenya et al., Nat. Microbiol. 5 , 668 (2020).
- S. Duffy, L. A. Shackelton, E. C. Holmes, Nat. Rev. Genet.
9 , 267 (2008). - L. A. Hug et al., Nat. Microbiol. 1 , 16048 (2016).
10.1126/science.abo5590Department of Marine Biology, Texas A&M University at
Galveston, Galveston, TX, USA. Email: [email protected]
MEDICINEUnlocking the
secrets to
Janus kinase
activation
The full-length structure of
a Janus kinase provides in-
sights for drug development
By R oss L. Levine^1 and Stevan R. Hubbard^2M
embers of the Janus family of non-
receptor tyrosine kinases (JAK1,
JAK2, JAK3, and TYK2) transmit
a diversity of ligand-mediated
signals, from cytokines and hor-
mones, resulting in activation of
downstream signaling pathways and al-
terations in gene expression. They have im-
portant roles in key physiologic functions,
including hematopoiesis and immune ef-
fector function. Additionally, aberrant ac-
tivation of JAK signaling plays a critical
role in various disease states, including
autoimmune disorders and various malig-
nancies. This has led to the development of
small-molecule JAK inhibitors, which pro-
vide therapeutic benefit to patients with in-
flammatory diseases, including rheumatoid
arthritis ( 1 ). However, new approaches are
needed to inhibit JAK signaling in other dis-
eases. On page 163 of this issue, Glassman
et al. ( 2 ) present a full-length JAK structure,
which provides a structural roadmap for
understanding the regulatory mechanisms
that govern JAK activity and the promise of
new therapeutic approaches.
Activating point mutations and fusion
events that involve JAKs have been identi-
fied in different human cancers. The most
common oncogenic events target JAK2, most
frequently through Val^617 Phe (V617F) sub-
stitution in most patients with myeloprolifer-
ative neoplasms (MPNs) ( 3 ). First-generation
JAK inhibitors show JAK2 inhibitory efficacy
in the MPNs myelofibrosis and polycythemia
vera ( 4 ). Although JAK2 inhibitors can im-
prove disease parameters in MPN patients,(^1) Human Oncology and Pathogenesis Program,
Memorial Sloan Kettering Cancer Center, New York,
NY, USA.^2 Department of Biochemistry and Molecular
Pharmacology, New York University Grossman School of
Medicine, New York, NY, USA. Email: [email protected];
[email protected]
8 APRIL 2022 • VOL 376 ISSUE 6589 139