INSIGHTS | PERSPECTIVES
sciencemag.org SCIENCEmuscle cells, which together facilitate the
transport function of these vessels ( 4 , 5 ).
The close proximity of HFSCs to ini-
tial lymphatic vessels raised the possibil-
ity that signaling between stem cells and
lymphatic endothelial cells might regulate
stem cell activity, lymphatic vessel func-
tion, or both. Gur-Cohen et al. found that
disrupting lymphatic vessel integrity in
mouse models promoted hair follicle pro-
liferation and hair regeneration. These
data suggested that signals from lymphatic
endothelial cells normally render HFSCs
quiescent (noncycling).
To explore whether the association be-
tween initial lymphatics and HFSCs was
temporally regulated, lymphatic vessel pat-
tern and function were assessed during the
cycle of hair growth and quiescence. These
analyses demonstrated that lymphatic
vessel proximity to the bulge region of
the follicle, where HFSCs reside, changed
during cyclical phases of hair growth and
quiescence: Initial lymphatic vessels were
closely associated with HFSCs during the
resting state and dissociated during the
proliferation and growth phase. Lymphatic
vessels also had reduced drainage capacity
during the proliferative phase of the hair
cycle. It will be intriguing in future studies
to understand whether the reduced drain-
age occurs as a result of a spatial change
in tissue architecture or a functional
change in lymphatic absorptive capacity.
To investigate the identity of factors
likely to be important for communication
between the HFSC niche and lymphatic
vessels, Gur-Cohen et al. profiled the
gene expression of quiescent and prolif-
erative HFSCs. Comparison of these data-
sets revealed a number of factors, some of
which had previously been implicated in
vascular remodeling, which were promi-
nent either during follicle quiescence—in
particular, angiopoietin-related protein 7
(ANGPTL7)—or during HFSC prolifera-
tion—particularly, ANGPTL4. Prolonga-
tion of ANGPTL7 expression through the
growth phase of the hair cycle maintained
the close association of lymphatic vessels
with the HFSC compartment, resulting in
maintenance of stem cell quiescence. Con-
versely, induced expression of ANGPTL4
during quiescence resulted in premature
dissociation of lymphatic vessels from the
niche and induction of stem cell prolifera-
tion. The expression of the Angptl4 gene
in mouse endothelial cells is important for
maintaining separation of the blood and
lymphatic vascular compartments in the
postnatal mouse intestine ( 6 ). This sug-
gests that ANGPTL4 provides a repulsion
signal to lymphatic endothelial cells that is
important in multiple tissue contexts.
Gur-Cohen et al. found that reducing
ANGPTL7 expression induced hyperplastic
follicles that entered the growth phase in
an asynchronous manner across the skin.
Moreover, lymphatic vessels were dilated
and exhibited reduced drainage capacity in
this scenario. Analysis of factors expressed
by lymphatic endothelial cells in different
phases of the hair cycle revealed that puta-
tive receptors for ANGPTL4 and ANGPTL7
are present on initial lymphatics. In addition,
WNT signaling inhibitors that restrict HFSC
activity and hair growth were produced by
lymphatic vessels during HFSC quiescence.
Together, these data suggest that fac-
tors secreted by stem cells directly affect
lymphatic vessel structure and/or func-
tion and vice versa (see the figure). It
will be interesting to determine whether
ANGPTL4 and ANGPTL7 have the same
impact on lymphatic vessels in other tis-
sues and to identify the receptors that
transduce ANGPTL signals in lymphatic
endothelial cells. It will also be fascinating
to understand the mechanisms by which
these factors regulate lymphatic vessel
absorptive capacity, particularly in the set-
tings of inflammation and disease. This
may result from active remodeling of cell
junctions. The possibility also exists that
mechanical signals, as a result of changes
in interstitial volume due to lymphatic ves-
sel remodeling, affect stem cell activity;
this will also be important to explore.
Additional recent studies have described
the intimate association of dermal lym-
phatic vessels with the HFSC niche ( 7 ,
8 ). Different mouse models were used in
these studies to investigate the impact of
increased or decreased lymphatic vessel
density on HFSC activity, but all concluded
that signaling between these two cellular
compartments is important for the coordi-
nation of hair growth in the skin. Further
studies exploring the link between lym-
phatic vessels and stem cell activity during
development and in disease could identify
new axes that can be targeted to enhance
tissue repair and regeneration. jREFERENCES AND NOTES- T. Tammela, K. Alitalo, Cell 140 , 460 (2010).
- S. Gur-Cohen et al., Science 366 , 121 (2019).
- S. Rafii, J. M. Butler, B.-S. Ding, Nature 529 , 316 (2016).
- T. Mäkinen et al., Genes Dev. 19 , 397 (2005).
- P. Baluk et al., J. Exp. Med. 204 , 2349 (2007).
- F. Bäckhed, P. A. Crawford, D. O’Donnell, J. I. Gordon,
Proc. Natl. Acad. Sci. U.S.A. 104 , 606 (2007). - D. Peña-Jimenez et al., EMBO J. 38 , e101688 (2019).
- S. Y. Yoon et al., PLOS ONE 14 , e0220341 (2019).
10.1126/science.aaz8780MOLECULAR BIOLOGYFolding
unpredicted
Unexpected topology
is the key to glutamate
receptor gating in
neurotransmission
By Jochen Schwenk and Bernd FaklerC
ornichon homologs (CNIHs) are a
family of highly conserved small
membrane proteins (140 to 160
amino acids in length) that serve as
shuttle(s) for the export of proteins
from the endoplasmic reticulum (ER)
and/or as auxiliary subunits that control
the building and gating of AMPA -type glu-
tamate receptors (AMPARs) in the brain.
According to predictions from databases
and algorithms, CNIHs exhibit a three-
transmembrane (TM) domain topology
with the protein’s amino terminus facing
the cytoplasm. On page 1259 of this issue,
Nakagawa pre-sents the cryo–electron mi-croscopy (cryo-EM) structure of a CNIH
protein in complex with a glutamate recep-
tor ( 1 ). The results prove database predic-
tions wrong and provide a structural key to
the diverse functions of the cornichon fam-
ily of proteins.
The founding member of this family, the
classical cornichon (CNIH1 in mammals),
was described ~20 years ago in the fruit
fly Drosophila melanogaster ( 2 ). It serves
as a cargo receptor required for efficient
ER export of the growth factor gurken ( 3 ).
Mechanistically, cornichon recognizes and
binds its cargo and transfers it to the secre-
tory pathway via recruitment into coat pro-
tein complex II (COPII)–coated vesicles ( 3 ).
This action was later detailed in yeast, where
the cornichon homolog Erv14 (ER-derived
vesicles protein 14) is able to accept a broadInstitute of Physiology II, Faculty of Medicine,
University of Freiburg, Freiburg, Germany.
Email: [email protected]Centre for Cancer Biology, SA Pathology and
University of South Australia, Adelaide, Australia.
Email: [email protected]
“...predictions from databases
and algorithms are good,
but rigorous experimental
data are better.”
1194 6 DECEMBER 2019 • VOL 366 ISSUE 6470
Published by AAASon December 12, 2019^http://science.sciencemag.org/Downloaded from