Download Subduction undone

RESEARCH NEWS & VIEWS
previously been shown to result in the removal
of cells by a process known as extrusion, followed by the death of the extruded cells4,5. This
process of regulated epithelial-cell extrusion is
mediated by the activation of Piezo1 (ref. 4).
In addition to Piezo1’s role in controlling cell
extrusion, the work by Gudipaty and colleagues reveals that this protein also controls
cell division in the epithelium.
How might Piezo1 activation cause both
cell division in response to cell stretching
and cell extrusion in response to cellular
overcrowding? To answer this, Gudipaty and
colleagues analysed and compared the downstream signalling pathways through which
Piezo1 controls cell division and extrusion. A
Piezo1-mediated increase in the intracellular
concentration of Ca2+ was sufficient to trigger
cell division, but not cell extrusion. Therefore, other Piezo1-mediated changes must be
required for cell extrusion to occur.
Piezo-family ion channels are thought to
be activated by an increase in tension in the
lipid membrane in which they are inserted3.
But how can the same protein be activated in
response to both cell stretching and cellular
overcrowding, which probably have opposite
effects on physical tension in tissues and their
constituent cells? Piezo1 is a large transmembrane protein that might respond to changes
in membrane curvature that occur because
of alterations in cell shape6. It is possible that,
rather than responding directly to an increase
in membrane tension, Piezo1 responds to
alterations in cell shape and membrane curvature that occur as a consequence of changes
in tissue tension.
Gudipaty and colleagues also provide
evidence that cell stretching and associated
changes in tissue tension can control not only
the activity, but also the subcellular localization
of Piezo1 in epithelial cells. For Piezo1 to control intracellular Ca2+ concentration, it needs
to localize to organelles or structures that are
involved in calcium regulation, such as the
endoplasmic reticulum or the cell membrane.
The authors present evidence that Piezo1 is
present on the cell membrane and endoplasmic reticulum in cells that form evenly spaced
or sparsely populated epithelium, whereas
Piezo1 exists in aggregated structures in the
cytoplasm in cells that form an overcrowded
epithelium in which extrusion occurs. Therefore, cell stretching and cellular overcrowding
probably control Piezo1 activity not only by
directly activating the ion channel through
changes in membrane tension and curvature,
but also by regulating the subcellular location
of the protein.
How cell stretching and cellular overcrowding control Piezo1 subcellular localization is
unknown, but it is conceivable that associated
changes in cell-membrane tension might be
involved. High membrane tension is thought
to trigger exocytosis, the process of vesiclemediated transport of material to the cell
membrane and out of the cell, whereas low
membrane tension promotes endocytosis, the
vesicle-mediated transport of material into the
cell7. Cell stretching and cellular overcrowding
might control Piezo1 subcellular localization
by regulating cell-membrane tension, leading to changes in the rate of exocytosis versus
endocytosis and, consequently, to changes in
vesicle-mediated transport of Piezo1 within
the cell.
The work by Gudipaty and colleagues
illuminates an intriguing mechanosensitive
feedback loop between tissue tension and cell
division, through which the integrity of epithelial-cell layers is maintained. Whether and how
this feedback loop operates in other tissues
experiencing different rates of cell division,
and whether it contributes to the healing
process for an injured epithelial layer, remain
important questions for future research. ■
Carl-Philipp Heisenberg is at the
Institute of Science and Technology Austria,
Klosterneuburg 3400, Austria.
e-mail: [email protected]
1.
2.
3.
4.
5.
6.
7.
Gudipaty, S. A. et al. Nature 543, 118–121 (2017).
Ingber, D. E. FASEB J. 20, 811–827 (2006).
Coste, B. et al. Science 330, 55–60 (2010).
Eisenhoffer, G. T. et al. Nature 484, 546–549 (2012).
Marinari, E. et al. Nature 484, 542–545 (2012).
Lewis, A. H. & Grandl, J. eLife 4, e12088 (2015).
Dai, J. & Sheetz, M. P. Cold Spring Harb. Symp.
Quant. Biol. 60, 567–571 (1995).
This article was published online on 15 February 2017.
GEO SC I EN C E
Subduction undone
Rocks are subjected to increased pressure as they are buried during subduction.
Contrary to general belief, a study suggests that peak pressures recorded in
subducted rocks might not reflect their maximum burial depths.
K I P V. H O D G E S
I
n the 1960s, the concept of plate tectonics
revolutionized the field of geoscience.
For most of the following two decades,
conventional wisdom in the geosciences
held that Earth’s continental crust does not
subduct into the mantle at convergent plate
boundaries because continents are much less
dense than the underlying mantle. This inference was challenged in dramatic fashion by
the discovery of ‘ultrahigh-pressure’ (UHP)
mineral assemblages in exposed continental rocks in the western Alps1 and the Scandinavian Caledonides 2. Since then, UHP
assemblages have been documented in
many mountain systems3 — some of the best
examples are found in the Tso Morari region
of the northwest Indian Himalaya4–6 (Fig. 1).
Almost all metamorphic petrologists have
interpreted these assemblages as evidence of
the subduction of continental rocks deep into
the mantle. But this interpretation begs the
question of how UHP rocks that are deeply
buried are subsequently returned to the surface (exhumed), and for this many possible
mechanisms have been proposed3. Writing in
Nature Geoscience, Yamato and Brun7 present
a mechanical analysis that questions both our
assumptions about what the burial depths
of UHP assemblages represent and their
geodynamic implications.
During subduction, rocks undergo mineral
transformations that record changes in temperature and pressure. The pressure acting
on subducted rocks is usually assumed to be
directly proportional to the thickness of the
overlying rock column8. A convenient approximation is that pressure increases by 0.027 gigapascals for every kilometre of depth, such that
rocks subjected to 2.7 GPa of pressure have
mantle depths of about 100 km. Subducted
rocks are therefore expected to attain peak
pressure at their maximum burial depth.
This type of calculation assumes that stresses
in Earth are lithostatic, that is, uniform in all
directions. But in tectonically active regions,
this assumption cannot be strictly correct
because deformational structures common
in these regions — such as folds and faults
— indicate the presence of shear stresses that
require non-uniformity. The question, however, is how large these variations in stress
(differential stresses) can be.
Most geoscientists would argue that
differential stresses are limited to about
1–2 GPa in Earth’s upper crust9. At deeper
levels, where temperatures are high enough
for ductile behaviour to be common, rocks of
similar composition would be expected to be
too weak to support large differential stresses10.
But some researchers have suggested11,12 that
differential stresses deep in subduction zones
could be high enough to produce pressures
of up to twice those expected for the lithostatic-stress condition. And yet the notion
of such large ‘overpressures’ seems inconsistent not only with our general knowledge of
rock-deformation mechanisms deep inside
Earth13, but also with petrological evidence14,15
for extreme decompression after UHP rocks
reach their maximum burial depth. This
4 4 | NAT U R E | VO L 5 4 3 | 2 M A RC H 2 0 1 7
ǟ ƐƎƏƗ !,(++- 4 +(2'#12 (,(3#"Ʀ /13 .$ /1(-%#1 341#ƥ ++ 1(%'32 1#2#15#"ƥ
ɥ
ɥ
ɥ
ɥ
ɥ
ɥ ɥ
ɥ
ɥ ɥ
ɥ
NEWS & VIEWS RESEARCH
MARNIE FORSTER
distributed UHP mineral assemblages — as
opposed to localized occurrences in rapidly
crystallized frictional melts — seems unlikely.
Tests of models such as that of Yamato and
Brun will require detailed laboratory and
geological field studies designed to achieve an
evidence-based understanding of the rheology
and tectonics of UHP terrains. ■
Kip V. Hodges is in the School of Earth and
Space Exploration, Arizona State University,
Tempe, 85287-6004 Arizona, USA.
e-mail: [email protected]
Figure 1 | Ultrahigh-pressure rocks in Tso Morari. In this outcrop of continental crustal rocks in the
Tso Morari region of the northwest Indian Himalaya, the darker rocks contain minerals indicative of
ultrahigh-pressure (UHP) conditions during subduction. Yamato and Brun7 suggest that peak pressures
recorded in UHP rocks worldwide reflect a change in tectonic stresses, rather than burial depth.
decompression is usually associated with the
early stages of exhumation.
Yamato and Brun note that temperature–
pressure data from UHP rocks worldwide
suggest that the peak pressure recorded in
these rocks is directly proportional to the drop
in pressure during the first stage of decompression. Accepting the hypothesis of large
differential stresses deep in subduction-zone
environments, the authors describe a simple
physical model that can explain this relationship. They propose that the decompression
evident in petrological data is actually caused
by a rapid switch in the “stress state” of the
rocks — from compression during burial to
extension at the onset of exhumation — rather
than extreme uplift of the rocks towards the
surface. If correct, this means that peak pressures are recorded in UHP rocks at the onset
of extension, rather than when the rocks are at
their maximum burial depth.
If large overpressures do occur in the
continental subduction zones at which UHP
mineral assemblages form, the power of
petrological data to elucidate tectonic processes in these environments could be severely
limited. However, sceptics might point to
potential issues with both the overpressure
hypothesis and Yamato and Brun’s mechanism
for decompression.
First, extremely rapid decompression,
regardless of the depth at which it occurs,
should leave a significant petrological signature that, so far, has not been confirmed in
UHP rocks. Second, the mechanical analysis
that led to the authors’ model is based on the
assumption of frictional rock behaviour deep
in continental subduction zones. Thus far,
although there is evidence for transient frictional behaviour in these environments16, the
idea that such behaviour could persist long
enough to result in the development of widely
1. Chopin, C. Contrib. Mineral. Petrol. 86, 107–118
(1984).
2. Smith, D. C. Nature 310, 641–644 (1984).
3. Hacker, B. R., Gerya, T. V. & Gilotti, J. A. Elements 9,
289–293 (2013).
4. de Sigoyer, J., Guillot, S. & Dick, P. Tectonics 23,
TC3003 (2004).
5. Sachan, H. K. et al. Eur. J. Mineral. 16, 235–240
(2004).
6. Wilke, F. D. H., O’Brien, P. J., Schmidt, A. &
Ziemann, M. A. Lithos 231, 77–91 (2015).
7. Yamato, P. & Brun, J. P. Nature Geosci. 10, 46–50
(2017).
8. Cammarano, F. Geophys. Res. Lett. 40,
4834–4838 (2013).
9. Burov, E. B. Mar. Petrol. Geol. 28, 1402–1443
(2011).
10. Brace, W. F. & Kohlstedt, D. L. J. Geophys. Res. 85,
6248–6252 (1980).
11. Gerya, T. J. Metamorphic Geol. 33, 785–800
(2015).
12. Reuber, G., Kaus, B. J. P., Schmalholz, S. M.
& White, R. W. Geology 44, 343–346 (2016).
13. Burov, E. et al. Tectonophysics 631, 212–250
(2014).
14. Chopin, C. Earth Planet. Sci. Lett. 212, 1–14 (2003).
15. Ernst, W. G. Lithos 92, 321–335 (2006).
16. Austrheim, H. & Boundy, T. M. Science 265, 82–83
(1994).
CA R D I OVAS C U L A R DI S E AS E
Commonality
with cancer
Ageing is associated with an increased risk of cardiovascular disease caused by the
rupture of inflamed cholesterol plaques in arteries. It emerges that this might be
partly due to genetic mutations that cause cancerous changes in white blood cells.
A L A N R . TA L L & R O S S L . L E V I N E
A
geing is a prominent risk factor for
a condition called atherosclerosis,
in which cholesterol accumulates in
arteries as plaques. When plaques become
inflamed, they can rupture or erode, leading
to blood clots that occlude the arteries and
cause heart attacks and strokes. One possible
driver is clonal haematopoiesis — a phenomenon in which mutations arise in blood-forming haematopoietic stem cells (HSCs) during
ageing, and promote the proliferation of
blood-cell populations bearing these
mutations at the expense of wild-type blood
lineages. But how this phenomenon might
drive atherosclerosis has been unclear.
Writing in Science, Fuster et al.1 outline a
pathway by which deficiency in the gene Tet2
causes accelerated atherosclerosis through
clonal haematopoiesis in mice.
In 2014, a study2 revealed that mutations
that cause clonal haematopoiesis occur in more
than 10% of people over the age of 70. The
most common mutations were in genes, such
as TET2, that encode proteins that modulate
the addition or removal of molecular modifications to DNA to alter gene expression, or in
2 M A RC H 2 0 1 7 | VO L 5 4 3 | NAT U R E | 4 5
ǟ ƐƎƏƗ !,(++- 4 +(2'#12 (,(3#"Ʀ /13 .$ /1(-%#1 341#ƥ ++ 1(%'32 1#2#15#"ƥ
ɥ
ɥ
ɥ
ɥ
ɥ
ɥ ɥ
ɥ
ɥ ɥ
ɥ