Young phosphorylation is functionally silent



Young phosphorylation is functionally silent
Young phosphorylation is functionally silent
Phosphorylation sites younger than 18 million years prevail in numbers, but not in function
n cell biology, important discoveries often
relate to mechanisms that are conserved
throughout evolution. Yet what makes
any species unique are specific molecular
changes accumulated during the course
of evolution. Thus, it is just as important
to identify differences between species as it
is to characterize their underlying similarities (1). On page 229 of this issue, Studer et
al. present such an evolutionary comparative
analysis, revealing similitudes and, remarkably, substantial differences in the phosphorylation landscapes of fungal species (2).
Phosphorylation is a reversible posttranslational modification of proteins that can
alter their structures and functions. This regulatory mechanism is ubiquitous throughout
eukaryotes, where it controls and amplifies
responses to intra- and extracellular signals.
Although the chemical process of phosphorylation is evolutionarily conserved, the
positions targeted for phosphorylation in
protein sequences are not necessarily so.
To characterize the evolution
of this regulatory mechanism,
Studer et al. identified phosphorEvolutionary phosphorylation insight
ylation events in the proteomes
(Top) Evolutionary tree of five representative fungal species for which Studer et al. characterized the phosphorylation (P)
of 18 fungal species, expanding
landscape (Sc, S. cerevisiae; Sb, S. bayanus; Ca, C. albicans; Kl, K. lactis; Sp, S. pombe). For reference, divergence times are
our view of this covalent mark
shown between humans and their LCA for different branches of the tree of life (14). (Bottom) Phosphorylation is not invariably
beyond the canonical model orassociated with function; many phosphosites are likely noisy, as reflected in their young age and lack of conservation.
ganisms (3–5). The work focuses
on fungi, but covers a broad evolutionary time period. The group
of closely related “SaccharomyAnimals
ces” fungi spans a divergence
time of ~18 million years, which
is comparable with the time sep69% of young P
arating humans from their last
2% of
common ancestor (LCA) with
old P
the apes (see the figure).
The authors analyzed 18 funSp
gal proteomes with mass spectrometry, allowing identification
180 Ma
65 Ma
340 Ma
731 Ma 650 Ma
525 Ma
18 Ma
of 73,340 phosphorylated sites
or “phosphosites” across the linFew old P , functional
Many young P , noisy
eage, with 3000 to 5000 sites per
Specifc noise
Nonspecifc noise
species. The subsequent comparphosphorylation
Randomly evolved
Random of-target efects
ison of these sites enabled the
kinase-target interactions
driven by mass action
Impacts ftness, selected
Weizmann Institute of Science, Department of
Structural Biology, Rehovot, 7610001, Israel.
*These authors contributed equally to this
work. Email: [email protected]
14 OCTOBER 2016 • VOL 354 ISSUE 6309
Published by AAAS
Downloaded from on October 13, 2016
lack of conservation appears to contradict
the textbook view that phosphorylation is
strictly controlled and regulates important
functions. Whereas certain phosphorylation
events do surely regulate function, many may
not. Edwin Krebs himself, who received the
1992 Nobel Prize with Edmond Fischer for
the characterization of “reversible protein
phosphorylation,” noted that there likely exists a degree of phosphorylation noise (8).
Noise—phosphorylation events not selected to carry out a specific function—can
provide a simple explanation for the weak
evolutionary conservation of phosphosites
(9). Mechanistically, the low degree of sequence specificity required for phosphorylation implies that new kinase recognition
motifs can frequently emerge by chance,
without having been selected for, and hence
need not be conserved. Kinase promiscuity
means that even noncanonical substrates
may be phosphorylated occasionally, so that
abundant proteins can yield subpopulations
detectable with mass spectrometry (10).
Evolutionary noise—mutations that are
not selected to carry out a specific func-
reconstruction of their probable evolutionary
history. Sites specific to a species or a group
of related species were inferred as having
emerged recently, whereas those conserved
throughout the lineage were inferred as being old and likely present in their LCA.
Studer et al. found that a mere ~2% of today’s phosphosites have been conserved since
the LCA, ~700 million years ago (Ma). To put
this in context, >73% of protein domains
have been retained in all these fungi since
their split from the LCA (6, 7), highlighting
that phosphorylation sites can be lost during
evolution much more rapidly than protein
domains are. Accordingly, the authors also
found phosphosites to be gained rapidly, with
69% being younger than 18 million years in
the baker’s yeast S. cerevisiae. In several control experiments, the authors ruled out data
coverage and condition-specific regulation
as being responsible for this large fraction
of young phosphosites. Moreover, using an
even more conservative approach that takes
adjacent sequence positions into account,
39% of phosphosites were still found to be
young and therefore not conserved. Such a
By Or Matalon,* Benjamin Dubreuil,*
Emmanuel D. Levy
M. Lynch et al., Proc. Natl. Acad. Sci. U.S.A. 111, 16990 (2014).
R. A. Studer et al., Science 354, 229 (2016).
C. S. Tan et al., Sci. Signal. 2, ra39 (2009).
J. Boekhorst et al., Genome Biol. 9, R144 (2008).
B. Bodenmiller et al., Nat. Biotechnol. 26, 1339 (2008).
M. E. Oates et al., Nucleic Acids Res. 43, D227 (2015).
73% is a conservative estimate, corresponding to the
domain superfamilies common to the 18 species, divided by
the total number of superfamilies in those species.
P. Cohen, Nat. Cell Biol. 4, E127 (2002).
C. R. Landry et al., Trends Genet. 25, 193 (2009).
E. D. Levy, S. W. Michnick, C. R. Landry, Philos. Trans. R. Soc.
London B Biol. Sci. 367, 2594 (2012).
G. I. Lang et al., Nature 500, 571 (2013).
C. Greenman et al., Nature 446, 153 (2007).
H. Steen et al., Proc. Natl. Acad. Sci. U.S.A. 102, 3948 (2005).
J. Rosindell, L. J. Harmon, PLOS Biol. 10, e1001406 (2012).
HIV particles (pink) bud from
the surface of a T lymphocyte
(blue) (colored transmission
electron micrograph shown).
Shock and kill with caution
Strategies to silence latent HIV infection should be explored
By Robert C. Gallo
ntiretroviral therapy has made lifelong suppression of human immunodeficiency virus (HIV) replication a
possibility for some patients. But with
the 2015 estimate of 36.7 million people infected worldwide, there is a great
need to explore other ways to address this
epidemic—from preventing new infections
by treating uninfected high-risk individuals,
to developing a vaccine, to targeting latent
HIV that hides in immune cells and persists
in patients. The idea of clearing latent infection has prompted strategies aimed at an
HIV-1 cure. Although this approach should
continue to be tested, other approaches,
including those that seek to permanently
suppress the latent virus, should also be explored. Different strategies may target different viral reservoirs, and may turn out to be
The existence of T cells harboring latent
HIV-1 provirus was first described in 1986 (1),
Institute of Human Virology, University of Maryland School
of Medicine, 725 West Lombard Street, Baltimore, MD 21201,
USA. Email: [email protected]
as was the demonstration that activation of
those T cells “reawakened” HIV-1 expression
(2). The importance of these latently infected
cells, however, came to the forefront following the work of several clinical investigators
(3) who showed that these cells persist long
after anti-HIV therapy and virus suppression,
and periodically release HIV-1, presumably in
response to a milieu favorable to T cell activation. The idea then arose that HIV-1 infection would be curable if latent virus could
be deliberately reactivated. This would lead
to T cell death, either directly from HIV-1
cytopathic effects or by cytotoxic T lymphocytes (CTLs). Concurrent anti-HIV therapy
would block new rounds of infection. This
idea has been called “shock and kill” or “kick
and kill,” and has spawned numerous studies
(4), clinical trials, and discussions at meetings. Part of this impetus was provided by an
enduring focus on one patient, the so-called
“Berlin patient” (5), who—through a series of
fortunate events—is the only known example
of complete HIV-1 cure. This individual was
HIV-1–positive, but also had leukemia, which
allowed physicians to treat him with total
body irradiation. Fortunately, he survived
this aggressive treatment, and his leukemia.
14 OCTOBER 2016 • VOL 354 ISSUE 6309
Published by AAAS
Downloaded from on October 13, 2016
tion—is in fact seen during evolutionary
adaptation. Whether in yeast adapting to a
new environment (11) or in cancer cells escaping growth-control mechanisms (12), rare
“driver” mutations arise in a background of
“passenger” mutations that have a negligible
impact on fitness. A common strategy to
identify driver mutations is to find recurrent
patterns across independent laboratory evolution experiments or cancer cell lines, such
as mutations falling in the same genes. But
to identify functional phosphosites, alternative approaches must be envisioned because
evolution cannot be replayed multiple times.
The work of Studer et al. opens new avenues
in this respect. They reveal that lineage-specific preferences in phosphosite context have
arisen across the fungal evolutionary tree.
For example, proteins in the baker’s yeast
lineage show a depletion in proline-based
phosphosite context and an increase in negatively charged context. They also found that
specific classes of proteins acquired phosphosites in a coordinated fashion during
specific time periods. These analyses suggest
that global properties of phosphoproteomes
are selected and therefore could be used to
predict functional phosphorylation events.
Equally important to predicting function will
be our ability to filter out noise, which will
require a more systematic consideration of
protein abundance (10) and phosphorylation
stoichiometry (13). More generally, the large
data set of ancient sites identified in this
work will make it possible to contrast structural and cellular properties of ancient and
young sites (such as structural environment,
presence of specific motifs, substrate interactions, expression, and localization) to discover new mechanisms and circuits involved
in functional versus noisy phosphorylation.
Evolutionary cell biology is still in its
early days (1). Thus, comparative proteomics
efforts will be increasingly important to
complement the postgenomic revolution,
elucidate molecular differences between cell
machineries across species, and fuel our understanding of life and its history. j
Young phosphorylation is functionally silent
Or Matalon, Benjamin Dubreuil and Emmanuel D. Levy (October
13, 2016)
Science 354 (6309), 176-177. [doi: 10.1126/science.aai8833]
This copy is for your personal, non-commercial use only.
Article Tools
Visit the online version of this article to access the personalization and
article tools:
Obtain information about reproducing this article:
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week
in December, by the American Association for the Advancement of Science, 1200 New York
Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for the
Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.
Downloaded from on October 13, 2016
Editor's Summary

Similar documents