SCIENCE science.orgboundary conditions—in this case, provided
by tissue geometry—can instruct cells to self-
organize deterministically into domains that
resemble the crypt-villus axis (see the figure).
An instructive role for tissue boundaries
in cell patterning has long-standing roots
in developmental biology, and recent efforts
have begun applying these concepts to engi-
neering organoids and other developmental
processes in vitro ( 3 ). For example, the ini-
tial geometry of mammary epithelial tubules
determines branching patterns ( 2 ), and the
geometry of human pluripotent stem cell
colonies influences their self-organization
into structures that resemble the neural tube
( 4 ). How is the geometry of the tissue bound-
ary mechanistically linked to patterning?
Broadly, these effects can be attributed to the
dynamic reciprocity that exists between a cell
and the mechanical and molecular compo-
nents of its microenvironment—for example,
the ECM ( 5 ).
A cell’s ability to sense the microenviron-
ment can occur through myriad mechanisms,
including through its physical deformation.
In intestinal organoids, this appears to be
mediated by the mechanosensitive transcrip-
tion factor YAP. Preceding the up-regulation
of activity of ISC markers in geometrically
constrained organoids, regions of high cur-
vature develop increased cell crowding and
cytoplasmic (inactive) YAP, relative to neigh-
boring flat regions with nuclear (active) YAP.
Suppression of crypt identity occurs in flat re-
gions of active YAP. Subsequently, patterning
of the crypt domain begins in the curved ends
where some cells then activate YAP and the
Notch ligand delta-like 1 (DLL1), which trig-
gers Notch signaling in neighboring cells that
retain inactive YAP. This symmetry-breaking
lateral inhibition event culminates in the
differentiation of Paneth cells that maintain
ISCs locally. A similar mechanism acts sto-
chastically in traditional organoid culture
to determine the site of crypt formation ( 6 ),
where curvature is subsequently generated
through differential actomyosin contraction
and luminal pressure ( 7 ).
Although regions with villus-like cell com-
position exist in organoids, the morphology
of these domains does not resemble villi in
vivo, and thus, their function cannot be accu-
rately modeled. Having identified engineer-
ing principles for deterministic patterning
of crypt-like domains, Gjorevski et al. turn to
the patterning of villus domains. They seed
cells onto engineered hydrogels that mimic
the surface topography of the small intestine,
with villus-shaped pillars located above well-
shaped crypts. Cells on these topographically
rich surfaces self-organized into both crypts
and villi, further supporting the idea that
intestinal epithelia can innately pattern in
a manner dependent on the tissue geometry
and independent of additional cues, such as
signals derived from stromal cells that are
absent in these epithelial cell-only organoids.
The findings of Gjorevski et al. raise the
question of how well intestinal organoid
morphogenesis recapitulates intestinal mor-
phogenesis during development—a point of
recent debate ( 8 , 9 ). For example, during nor-
mal development, villi emerge before crypts,
suggesting that the location of villi may de-
termine the location of crypts. Indeed, the
morphogen gradients that emerge during
folding of the epithelium shape signaling
gradients that localize ISC progenitors to the
intervillus spaces that later give rise to crypts
( 10 , 11 ). These same signaling gradients are
also important for homeostasis in the adult
( 12 ). How are these two determinants of cell
state, one geometry-based and the other sig-
naling-based, normally integrated during de-
velopment, homeostasis, and regeneration?
Whether these are redundant mechanisms or
they work in concert to ensure that cell iden-
tity and tissue structure mutually reinforce
one another is unclear.
What is now clear, however, is that both
geometry and epithelial-autonomous sig-
naling pathways are sufficient to pattern
cell identity in the small intestine. For
example, previous work established that
intestinal organoid monolayers grown in
the absence of any surface contours seem
to display zones with crypt and villus
identity that are likely determined by the
same bone morphogenetic protein (BMP)
and Wnt signaling axes that set up these
domains in vivo ( 13 , 14 ). Therefore, future
studies should investigate whether and
how these mechanisms work in concert
to ensure that tissue form, and thus tissue
function, are robust outputs of develop-
ment and homeostasis. jREFERENCES AND NOTES- N. Gjorevski et al., Science 375 , eaaw9021 (2021).
- C. M. Nelson et al., Science 314 , 298 (2006).
- I. Martyn, Z. J. Gartner, Dev. Biol. 474 , 62 (2021).
- E. Karzbrun et al., Nature 599 , 268 (2021).
- M. J. Bissell et al., J. Theor. Biol. 99 , 31 (1982).
- D. Serra et al., Nature 569 , 66 (2019).
- Q. Yang et al., Nat. Cell Biol. 23 , 733 (2021).
- J. Guiu, K. B. Jensen, Cell. Mol. Gastroenterol. Hepatol.
13 , 1 (2021). - S. Sugimoto, T. Sato, Cell. Mol. Gastroenterol. Hepatol.
13 , 195 (2021). - A. E. Shyer et al., Cell 161 , 569 (2015).
- J. Guiu et al., Nature 570 , 107 (2019).
- N. McCarthy et al., Cell Stem Cell 26 , 391 (2020).
- C. A. Thorne et al., Dev. Cell 44 , 624 (2018).
- C. Pérez-González et al., Nat. Cell Biol. 23 , 745 (2021).
ACKNOWLEDGMENTS
Z.J.G. is an equity holder in Scribe Biosciences and
Provenance Bio and is an advisor for Serotiny.
10.1126/science.ab n3054MICROBIOLOGYArchaeal
nitrification
without oxygen
The single-cell organism
can self-produce oxygen for
ammonia oxidation
By Willm Martens-Habbena^1 and Wei Qin^2A
mmonia-oxidizing archaea (AOA)
constitute up to 30% of the microbial
plankton in the oceans and play a
key role in the marine nitrogen and
carbon cycle. Together with nitrite-
oxidizing bacteria (NOB), they oxi-
dize ammonia to nitrate—the predominant
inorganic nitrogen source in the sea. AOA
and NOB are thought to rely on molecular
oxygen (O 2 ), but recent studies revealed
their presence in strictly anoxic marine oxy-
gen minimum zones, challenging this para-
digm (1–3). On page 97 of this issue, Kraft
et al. ( 4 ) demonstrate that the marine AOA
Nitrosopumilus maritimus, under anoxic
conditions, produces O 2 for ammonia oxida-
tion by itself while simultaneously reducing
nitrite to nitrous oxide (N 2 O) and dinitrogen
(N 2 ). These results provide a possible expla-
nation for the presence of AOA in marine
oxygen minimum zones, where they may
have an important role in nitrogen loss. The
findings may further have implications for
the evolution of the nitrogen cycle on Earth.
The discovery of ammonia oxidation
within the domain Archaea has substantially
changed the understanding of the global
nitrogen cycle ( 5 ). Phylogenomic and mo-
lecular dating analyses using different clock
models have suggested that ammonia oxida-
tion in the Archaea first arose at least a bil-
lion years ago ( 6 , 7 ). Modern AOA are among
the most abundant and ecologically success-
ful microbial groups, showing extraordinary
adaptations to low nutrient and energy
fluxes in the deep oceans ( 8 , 9 ). They also
possess the most energy-efficient aerobic
pathway for carbon fixation ( 10 ). As predom-
inant ammonia oxidizers in the oceans, AOA
are considered a primary source of ocean-(^1) Department of Microbiology and Cell Science, University
of Florida, Institute for Food and Agricultural Sciences,
Fort Lauderdale Research and Education Center, Davie,
FL 33314, USA.^2 Department of Microbiology and Plant
Biology, University of Oklahoma, Norman, OK 73019, USA.
Email: [email protected]
Department of Pharmaceutical Chemistry, University of
California San Francisco, CA, USA. Email: [email protected]
7 JANUARY 2022 • VOL 375 ISSUE 6576 27