616 11 FEBRUARY 2022 • VOL 375 ISSUE 6581 science.org SCIENCE
compartment in which the spicule forms,
yields spicules with morphologies defined
by this modified compartment ( 12 ). More
recently, the shape of larval spicules in vitro
has been engineered by directing the assem-
bly of cultured cells that then generate the
three-dimensional environment in which the
spicule forms ( 13 ).
It has been widely proposed that biominer-
alization by way of an amorphous precursor
phase—which has no preferred geometry—
allows the mineral to be molded into com-
plex shapes by templating. However, as seen
in both synthetic systems ( 14 ) and organisms
such as coccoliths ( 15 ), it is not a pre-requi-
site. With the ease of forming crystals that
far exceed the length scale of the template,
and fairly isotropic crystal morphologies, cal-
cite is a perfect construction material for the
stereom, where it can fill a template of any
shape without correspondence between the
micrometer-scale morphology of the crystal
and its crystal lattice.
The ultimate question of what determines
the ultrastructure remains open, however.
Given that lipids can assemble into G-, D-,
and P-TPMS in vitro, and similar assem-
blies of soft matter are present in butterfly
wing scales, it is possible that the calcite
grows within a sponge-like membrane that
templates the calcite. That case is strength-
ened by the presence of both P and D mor-
phologies in echinoderms. However, current
knowledge does not explain how those struc-
tures can form with huge lattice parameters
(>10 mm). They may be the result of dynamic
and coordinated cellular processes, com-
monly assumed to control morphologies in
living systems. Alternatively, they may result
from growth within a porous matrix, analo-
gous to the origin of those structures in syn-
thetic soft materials. j
REFERENCES AND NOTES
- G. Donnay, D. L. Pawson, Science 166 , 1147 (1969).
- T. Yang et al., Science 375 , 647 (2022).
- S. T. Hyde, S. Schröder-Turk, Interface Focus 2 , 529
(2012). - S. Hyde et al., The Language of Shape (Elsevier Science,
1996). - B. D. Wilts, K. Michielsen, J. Kuipers, H. De Raedt, D. G.
Stavenga, Proc. Biol. Sci. 279 , 2524 (2012). - H. T. Ghiradella, M. W. Butler, J. R. Soc. Interface 6 (suppl.
2), S243 (2009). - L. Han et al., IUCrJ 7 , 228 (2020).
- K. M. Towe, Science 157 , 1048 (1967).
- J. Seto et al., Proc. Natl. Acad. Sci. U.S.A. 109 , 3699
(2012). - D. M. Raup, in Physiology of Echinodermata, R. A.
Boolootian, Ed. (Interscience, 1966), pp. 379–395. - A. Jantschke, I. Pinkas, A. Schertel, L. Addadi, S. Weiner,
Acta Biomater. 102 , 427 (2020). - K. Okazaki, Embryologia (Nagoya) 7 , 21 (1962).
- C. H. Wu, A. Park, D. Joester, J. Am. Chem. Soc. 133 , 1658
(2011). - N. B. J. Hetherington et al., Adv. Funct. Mater. 21 , 948
(2011). - Y. Kadan, F. Tollervey, N. Varsano, J. Mahamid, A. Gal,
Proc. Natl. Acad. Sci. U.S.A. 118 , e2025670118 (2021).
10.1126/science.abn2717
By Pavitra Muralidhar1,2 and Carl Veller1,2I
n many species, sex is determined by
the presence or absence of a particular
chromosome. In mammals, for example,
males usually carry a Y and an X chro-
mosome and females usually carry two
X chromosomes. The sex-specific chro-
mosome is often degenerate—small in size
and gene-poor. Until recently, biologists
have relied on a classical model to explain
the evolution of degenerate sex chromo-
somes, involving sex differences in selection
and the principle that selection acts less ef-
ficiently in regions of the genome that do
not recombine in gametogenesis ( 1 , 2 ). How-
ever, this model has received little concrete
empirical support, and recent observations
have called into question its generality ( 3 ,
4 ). On page 663 of this issue, Lenormand
and Roze ( 5 ) propose a new model for the
evolution of degenerate sex chromosomes,
based on the evolution of gene regulation.
Theories of sex chromosome evolution
start with a pair of chromosomes that differ
only in the genetic variant that they carry at
a single, sex-determining locus. For simplic-
ity, consider one variant to be dominantly
male-determining, defining a “proto-Y chro-
mosome” (see the figure). (Alternatively, the
variant could be dominantly female-deter-
mining, defining a ZW-female and ZZ-male
system, as in birds and butterflies.) Genetic
exchange between the proto-Y and proto-X
chromosomes, through recombination dur-
ing sperm formation, homogenizes them.
For differences between the proto-Y and
proto-X to evolve, recombination must be
shut down across a region that includes the
sex-determining locus.
The classical theory invokes sexually an-
tagonistic selection to explain this initial
recombination shutdown ( 3 , 4 ). Sexual an-
tagonism occurs when males and females
have different fitness optima for some
traits—genetic variants that influence these
traits and are expressed in both sexes can
be beneficial in one sex and costly in theother. According to the classical theory,
first, a male-beneficial variant appears on
a chromosome with the male-determining
gene, forming a combination that jointly
triggers male development and increases
male fitness. To preserve this mutually
beneficial arrangement, recombination is
shut down along a segment of the proto-
Y that contains the male-beneficial and
male-determining genes, and the segment
subsequently spreads to high frequency. Re-
combination shutdown can be achieved in
several ways, the simplest of which is physi-
cal inversion of the chromosomal segment.
Because selection acts inefficiently in ge-
nomic regions that do not recombine, dele-
terious mutations subsequently accumulate
in various genes within the nonrecombin-
ing segment of the proto-Y ( 1 , 2 ). To mini-
mize their negative effects, expression of
mutated proto-Y genes is then reduced. This
silencing of the proto-Y creates an imbal-
ance in gene expression between XY males
and XX females, selecting for mechanisms
that correct this imbalance by increasing
proto-X expression in males or decreasing
proto-X expression in females—a phenom-
enon called dosage compensation.
This “textbook” theory has received rela-
tively little empirical support ( 3 , 4 ). A key
uncertainty revolves around the causal role
of sexual antagonism in the initial shut-
down of recombination, which has been
demonstrated in some taxa ( 6 , 7 ) but ap-
pears to be absent in others ( 8 , 9 ). The am-
biguous importance of sexual antagonism,
together with the discovery of cases where
dosage compensation precedes degenera-
tion ( 10 , 11 ), has led to growing skepticism
about the generality of the classical theory.
Lenormand and Roze propose a radically
different theory for sex chromosome evolu-
tion, with gene regulation playing the star-
ring role. In the simplest version of their
model, the expression of each of the many
genes on the proto-sex chromosomes is con-
trolled by one “cis-regulator” on the same
chromosome as the gene and two “trans-
regulators” on distinct chromosomes. Each
cis-regulator controls only the copy of the
gene to which it is physically linked, whereas
the trans-regulators—one male-specific andGENETICSA new model of sex
chromosome evolution
A new model, based on the evolution of gene regulation,
challenges the classical theory
INSIGHTS | PERSPECTIVES
(^1) Center for Population Biology, University of California,
Davis, CA, USA.^2 Department of Evolution and Ecology,
University of California, Davis, CA, USA.
Email: [email protected]; [email protected]