INSIGHTS | PERSPECTIVES
By Belén Gómez-González
and Andrés Aguilera
The integrity of DNA is constantly threat- ened by the molecules that are endog- enously generated by cell metabolism, the most common by-product being reactive oxygen species (ROS) (1). This is particularly relevant during DNA
replication, as highlighted by the overrepresentation of replication-associated spontaneous mutations in cancer (2). In a constantly
changing environment, cell survival requires
a fine-tuned control of replication. Because
DNA replication in eukaryotic cells initiates
from multiple replication origins, it can be
regulated at both the frequency of origin initiation and the rate of replication fork progression. The in vitro reconstitution of the
eukaryotic replisome with purified proteins
has recently boosted our knowledge of both
processes in eukaryotes (3, 4). However, we
still lack a full understanding of the mechanisms and implications of fork rate modulation in vivo. Notably, on page 797 of this issue,
Somyajit et al. (5) show that the reduction of
replication fork speed by low levels of ROS
is a major mechanism to mitigate their negative impact on DNA replication and genome
integrity, in a process that may be critical for
tumor cell survival.
Somyajit et al. report that forks slowed
down in human cells that were treated with
low-dose hydroxyurea (HU), an inhibitor of
the ribonucleotide reductase (RNR). RNR is
an essential enzyme in all living organisms
that catalyzes the production of deoxyribo-nucleotide triphosphates (dNTPs), the components that make up DNA. Low levels and
imbalanced dNTPs challenge genome integ-
Centro Andaluz de Biología Molecular y Medicina
Regenerativa, Universidad de Sevilla, Avenida Américo
Vespucio 24, 41092 Seville, Spain. E-mail: email@example.com
Fork speed modulation
imbalance to safeguard
Hemisphere warm spell that began abruptly
about 14,600 years ago (8). Menounos et al.
hypothesize that this large warming, evident
in records from the North Atlantic to the
North Pacific (9), lifted the snow line high on
the ice sheet. In response, the Cordilleran ice
sheet thinned rapidly and became dissected
by mountains into a “labyrinth of valley
glaciers,” setting the stage for the alpine readvances that they dated.
More provocatively, the authors suggest
that this ice loss may have contributed to
subsequent Northern Hemisphere cooling
through freshwater inputs to the ocean that
disrupted its overturning circulation, and
to meltwater pulse-1A, an abrupt jump in
sea level during the Bølling transition with
unknown origin (10). The Cordilleran ice
sheet was too small to account for much of
the meltwater pulse-1A sea level rise, but
Menounos et al.’s Bølling-driven thinning
scenario might explain a fifth of the event.
A presumably analogous response from the
Laurentide ice sheet would push this estimate much higher.
The new ages reported by Menounos et
al. help to better constrain the Cordilleran
ice sheet history after the last ice age. The
data show how mountainous topography
can modulate ice sheet responses to climate change, which may have important
implications for understanding past, and
future, deglaciation in parts of Greenland
and Antarctica and offer modelers a useful test case. But questions remain regarding the precise timing of Cordilleran ice
sheet collapse and its connections to past
sea level and abrupt climate changes. In
particular, it remains unclear whether the
Cordilleran ice sheet played an appreciable
role in meltwater pulse-1A and whether the
freshwater input was large enough to usher
in a climate reversal during the last glacial
termination. Future studies may provide a
clearer answer by applying
a similar dating approach
as that of Menounos et al.
directly to ice sheet deposits rather than to younger
features that only give a
limit on its disappearance.
In doing so, they could constrain the vertical changes
of the ice sheet through
time (11). Continuous sediment records from ocean
archives in the northeastern Pacific may also help to
better constrain the timing
of the ice sheet’s demise
(12) and the climate forcing
that drove it. j
Cordilleran ice sheet
As Menounos et al. show, the
Cordilleran ice sheet melted earlier
than anticipated. Small glaciers
formed at high elevations.
Laurentide ice sheet
Retreat of the Laurentide ice
sheet was much more uniform
because of the fatness of the
10 km 1000 km
Last glacial maximum
20,000 years ago
14,500 years ago
Glacial readvance 1
14,000 years ago
Glacial retreat and fnal
11,000 years ago
722 10 NOVEMBER 2017 • VOL 358 ISSUE 6364
1. J. J. Clague, in Quaternary Geology
of Canada and Greenland, R. J.
Fulton, Ed. (Geological Survey of
Canada, 1989), vol. 1.
2. A. S. Dyke, in Developments in
Quaternary Sciences (Elsevier,
2004), vol. 2, part B, pp. 373–424.
3. M. W. Pedersen et al. , Nature 537,
4. R.F.Ivanovicetal., Geophys.Res.
Lett. 44, 383 (2017).
5. B.Menounos et al., Science358,
6. M. D. Kurz, Geochim. Cosmochim.
Acta. 50, 2855 (1986).
7. G. Balco, Quat. Sci. Rev. 30, 3
8. P. M. Grootes et al., Nature 366, 552
9. S. K. Praetorius, A. C. Mix, Science
345, 444 (2014).
10. J. Liu et al ., Nat. Geosci. 9, 130
11. E. J. Brook et al., Geology 24, 207
12. A. D. Wickert et al., Nature 502, 668
How ice sheets melt
At the end of the last glaciation, ice sheets across North America began
to melt. The melting patterns depended on whether the ice sheets
rested on flat or mountainous terrain.