The FEBS Journal 2023 Hounjet Iron Responsive Element of Divalent
The FEBS Journal 2023 Hounjet Iron Responsive Element of Divalent
The FEBS Journal 2023 Hounjet Iron Responsive Element of Divalent
Introduction
Notch signalling constitutes a highly conserved pathway through Adam10 protease at the S2-site (NEXT). Subse-
in tissue homeostasis that occurs during development quent cleavage at the S3 site by c-secretase releases the
and in adult tissues [1]. Notch receptors are transmem- Notch intracellular domain (Nicd) that translocates to
brane transcriptional regulators that are activated by the nucleus. Together with the DNA binding protein
ligand-induced cleavage. In the absence of ligands, Notch RBP-jK, it activates Notch downstream target genes [2].
receptors are in a proteolysis-resistant state. Upon ligand Aberrant Notch signalling is observed in many
binding, the receptor unfolds and undergoes cleavage human malignancies, including T-ALL, breast ,
Abbreviations
cAMP, cyclic-AMP; Dll4, delta-like 4; Dmt1, Divalent metal transporter 1; DOX, doxycycline; Eea1, early endosomal marker 1; GSI, c-
secretase inhibitor; Ire, iron-responsive element; Lamp1, lysosomal-associated membrane protein 1; Lc3b-II, microtubule-associated proteins
1A/1B light chain 3B-II; MEF, mouse embryonic fibroblasts; Msf, myosin skeletal FAST; Myf5, myogenic factor 5; MyoD, myoblast
determination protein 1; MyoG, myogenin; NAC, n-acetyl cysteine; NEXT, S2-cleaved Notch1 receptor; Nicd, Notch intracellular domain;
Rab5, Ras-related protein 5; ROS, reactive oxygen species; Slc11A2, solute carrier family 11, member 2; TfR1, transferrin receptor 1; TMIC,
S1-cleaved Notch1 receptor; U9, screening cell line.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 1
Federation of European Biochemical Societies.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
2 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
Fig. 1. Loss of Dmt1 inhibits ligand-independent Notch signalling. (A) Notch1 activity in a screening cell line (U9) upon DEGF-Notch1-L1594P
expression and DMSO, GSI or chloroquine (CQ) treatment for 24 h measured by luciferase counts [one-way ANOVA (Tukey comparison),
***P < 0.001, significantly compared with control]. (B) Immunoblot analysis of (cleaved) Notch1 (Myc), Val1744 and b-Actin (loading control)
in U9 cells treated with DMSO, GSI, chloroquine (CQ) or Bafilomycin A1 (BAFA1) for 24 h. Notch reporter for c-secretase-dependent Notch
signalling due to inhibition of Nicd1 formation, as shown by the appearance of S2-cleaved Notch1 (Next). (C) shRNA screen in Adam10/
17/-deficient MEFs expressing active ligand-independent Notch1 and Notch reporter. RFP-positive and GFP-null expressing cells were
sorted, and HT-barcode sequencing revealed the Slc11a2 gene, encoding Dmt1 as a novel regulator of Notch signalling. (D) Representation
of genes that were included in the shRNA screen. Nonsignificant (grey dot) and significant (red dot) genes are shown according to the aver-
age number of reads from the screen. (E) Pathway analysis of the shRNA screen reporting the number of genes included in each of the
pathways. GSI, c-secretase-inhibitor dibenzazepine. Data are representative of three independent experiments and values are expressed in
mean SD.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 3
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
Fig. 2. Loss of Dmt1 results in diminished endogenous Notch1 signalling, which is rescued by overexpression of Nicd. (A) mRNA
expression levels of Notch target genes, Hey1 and Hey2, in Dmt1 WT and KO MEFs stimulated with Dll4 and treated with DMSO or GSI
for 24 h, measured by qRT-PCR. Csnk2a2 mRNA expression was employed as a housekeeping control [one-way ANOVA (Tukey compari-
son), *P < 0.05, **P < 0.01, ***P < 0.001, significantly compared with Dmt1 WT cells stimulated with Dll4]. (B) Fluorescent ectopic expres-
sion of Nicd1-mCherryGFP in Dmt1 WT and KO cells upon doxycycline (DOX)-induction. Cells were counterstained with DAPI. Scale bar:
10 lm. GSI, c-secretase inhibitor dibenzazepine. Data are representative of three independent experiments, and values are expressed in
mean SD.
Fig. 3. Loss of Dmt1 results in increased Notch1 cleavage and disturbed intracellular trafficking. (A) Immunoblot analysis of Notch1 (C-20),
Val1744, Myc (JAGGED2) and Vinculin (loading control) protein levels in Dmt1 WT and KO MEFs treated with DMSO or GSI for 24 h. (B)
Flow cytometry analysis of extracellular Notch1 receptor expression at the plasma membrane (unpermeabilized) and total expression (per-
meabilized) in Dmt1 WT and KO MEFs (Student t-test, *P < 0.05, ***P < 0.001). (C) Fluorescently labelled dextran uptake within 1 and 24 h
in Dmt1 WT and KO MEFs, pretreated with or without chloroquine for 24 h, measured by flow cytometry (one-way ANOVA [Tukey compari-
son], ns, nonsignificant, *P < 0.05, ***P < 0.001, significantly compared with untreated Dmt1 WT MEFs after 1 h of dextran uptake). (D)
Blue/yellow fluorescence ratio of Dmt1 WT and KO MEFs pretreated with or without chloroquine for 24 h, after 24 h of Lysosensor-labelled
dextran uptake [one-way ANOVA (Tukey comparison), ns, nonsignificant, **P < 0.01, significantly compared with untreated Dmt1 WT
MEFs]. (E) Immunoblot analysis of Eea1, Rab5, Lamp1, Lc3b-II and b-Actin (loading control) protein levels in Dmt1 WT and KO MEFs treated
with or without chloroquine for 24 h. (F) Representative electron microscopy images of Dmt1 WT and KO MEFs. Dmt1 KO MEFs show
damaged intracellular vesicles, displaying nonintact membranes (red asterisks), missing membranes (red hashtags) and formation of isolation
membranes (phagophores, red arrows). (G) Cellular Fe2+ probe fluorescence in Dmt1 WT and KO MEFs measured by flow cytometry (Stu-
dent t-test, **P < 0.01). (H) Cytoplasmic ROS levels measured by flow cytometry in Dmt1 WT and KO MEFs (pretreated with 10 mM of
NAC; one-way ANOVA [Tukey comparison], **P < 0.01). CQ, chloroquine; Eea1, Early endosome antigen-1; GSI, c-secretase-inhibitor diben-
zazepine; Lamp1, lysosomal-associated membrane protein 1; Lc3b-II, membrane-associated microtubule-associated protein 1 light chain 3-II;
MFI, median fluorescent intensity; NAC, N-acetyl-L-cysteine. Data are representative of three independent experiments, and values are
expressed in mean SD.
4 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
full-length, S1-furin-cleaved (Tmic) and S2-cleaved Because we observed high levels of Val1744-cleaved
(Next) Notch1 receptor levels were increased in Dmt1 Nicd1 without target gene activation in Dmt1 KO
KO MEFs. Total and cell surface staining for Notch1 cells, we investigated whether Dmt1 loss perturbs
demonstrated a twofold increase in the Notch1 receptor Notch1 receptor trafficking. We observed a threefold
levels in Dmt1 KO MEFs, albeit without a change in increase in the uptake of fluorescently labelled dextran
the surface/total ratio (Fig. 3B). within 24 h in Dmt1 KO MEFs compared with wild-
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 5
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
type cells, which was not affected by blocking autop- We also used immunostaining and ectopic Dmt1 iso-
hagy with chloroquine (Fig. 3C). We utilized dextran form expression to investigate whether the differences
as a pH sensor to test whether intracellular pH in Dmt1 isoform localization could explain the oppo-
changes correlated perturbed Notch activity, using a site effects of Dmt1b isoforms on Notch activity
dextran-emitting yellow fluorescence in acidic and blue (Fig. 4A–C). While Dmt1b-ire was expressed in small
fluorescence in neutral environments. In Dmt1 KO cytoplasmic speckles and showed (peri-) nuclear accu-
MEFs, an absolute increase in yellow and blue fluores- mulation, Dmt1b+ire overexpression marked large
cence was observed compared with the uptake in cytoplasmic speckles (Fig. 4C) that strongly over-
Dmt1 WT MEFs (Fig. 3C, Fig.S2F); however, the lapped with the lysosomal marker Lamp1. Together,
ratio remained constant (Fig. 3D). Chloroquine neu- these data show that different Dmt1 isoforms show
tralized the pH (increase in blue fluorescence) in both different localization, which may be indicative of the
Dmt1 wild-type and KO cells. different roles in Notch trafficking.
Furthermore, compared with Dmt1 WT MEFs,
Dmt1 KO MEFs also exhibited increased levels of the
Dmt1 controls Notch1-mediated cell fate
early endosomal marker 1 (Eea1), Rab5, the lyso-
decisions
somal-associated membrane protein 1 (Lamp1) and
the autophagosome marker Lc3b-II (Atg8) (Fig. 3E). Because Dmt1 isoforms strongly and oppositely affect
In addition, we observed a decreased autophagic flux Notch signalling, we investigated whether Dmt1 was
in Dmt1 KO MEFs, evident by the decreased accumu- necessary and sufficient for Notch-mediated cell fate
lation of Lc3b-II in the presence of chloroquine. In decisions in the Notch-controlled tissues such as mus-
line with this, electron microscopic analysis of Dmt1 cle, nervous system and intestinal epithelium.
KO MEFs revealed damaged intracellular vesicles with First, we created Dmt1+ire or Dmt1-ire knockdown
disrupted and punctured membranes (Fig. 3F). These in murine C2C12 myoblasts. C2C12 cells express both
data suggest that loss of Dmt1 compromises the integ- Dmt1b-ire and Dmt1b+ire isoforms (Fig. S4A). C2C12
rity of endo/lysosomal membranes, disrupting the cells differentiate into myotubes upon serum starva-
autophagy/lysophagy catabolic process and trafficking tion, and differentiation is blocked by ectopic Notch1
of Notch receptors. activation and accelerated by Notch inhibition [28].
To understand how Dmt1 loss perturbs vesicle integ- When treated with GSI, C2C12 cells displayed acceler-
rity, we measured cytoplasmic Fe2+ and ROS levels as ated myotube formation (Fig. 5A) with a concomitant
the cytotoxic by-products of iron-catalysed reactions increase in myogenic gene expression and downregula-
upon deregulated metal transport. Dmt1 KO MEFs tion of Hey1 (Fig. 5B,C). Dmt1-ire knockdown
showed increased cellular levels of Fe2+ (Fig. 3G) and (~ 79%, Fig. 5D) in C2C12 cells resulted in increased
accumulation of cytoplasmic ROS, which could be myotube formation (Fig. 5A) and premature and
reverted by the antioxidant n-acetyl cysteine (NAC) increased expression of the muscle differentiation
(Fig. 3H). Collectively, these data suggest that Dmt1 markers, Msf (myosin fast) (Fig. 5B), Myf5, MyoD
deficiency may cause vesicle membrane damage due to and MyoG (Fig. 5C) [29]. Consistent with the loss of
abnormal iron (Fe2+) accumulation and metabolism, Notch1, Dmt1-ire knockdown decreased Hey1 mRNA
leading to oxidative stress. after 4 days of C2C12 differentiation (Fig. 4C). In
contrast, Dmt1+ire knockdown (~ 76%, Fig. 5D) sup-
pressed myotube formation and muscle cell differentia-
Dmt1 iron response element controls Notch1
tion, with reduced MyoD, MyoG, Myf5 and Msf
signalling differentially
expression (Fig. 5A–C). In line with decreased myo-
Dmt1 occurs in different isoforms, with and without a 3- blast differentiation, Dmt1+ire knockdown showed
primed iron-responsive element; Dmt1+ire and Dmt1- enhanced Notch activity through increased Hey1
ire, respectively. To investigate how these isoforms regu- expression. The data show that Dmt1 isoform expres-
late Notch signalling, we stably transduced U9 Notch sion is necessary and sufficient to control Notch-
reporter cells (which only express Dmt1b (Fig. S3A)) mediated muscle cell differentiation in C2C12 cells.
with Dmt1-ire and Dmt1+ire shRNAs (Fig. S3B). Silenc- Next, we investigated the role of Dmt1 in Notch-
ing of Dmt1-ire reduced Notch reporter activity dependent neuronal differentiation [30]. The neuronal
(Fig. S3C) and Hey1 and Hes1 expression (Fig. S3D). differentiation of Neuro2A cells [which only express
Conversely, silencing of the Dmt1+ire isoform had the Dmt1b isoforms (Fig. S4B)] is induced by serum deple-
opposite effect, causing increased Notch reporter activ- tion and cAMP treatment (Fig. 6A). As expected, GSI
ity and target gene expression (Fig. S3C,D). treatment enhanced Neuro2A differentiation, as
6 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
Fig. 4. Ectopic expression of Dmt1b+ire localizes to the lysosomes. (A) Dmt1b mRNA expression levels in U9 cells with stable
overexpression of empty vector (EV), Dmt1b-ire (1B-) or Dmt1b+ire (1B+). Csnk2a2 mRNA expression was used as a housekeeping control.
(B) Immunoblot analysis of HA, FLAG and b-Actin (loading control) protein levels in U9 cells with stable overexpression of empty vector
(EV), Dmt1b-ire or Dmt1b+ire treated with DMSO or GSI for 24 h. (C) Immunofluorescence co-staining for HA- or FLAG-tagged Dmt1b iso-
forms (green) and Lamp1 (red) in U9 cells with stable overexpression of empty vector (EV) or the different Dmt1b isoforms. Cells were
counterstained with DAPI. Scale bar: 10 lm. GSI, c-secretase inhibitor dibenzazepine; Lamp1, lysosomal-associated membrane protein 1.
Data are representative of three independent experiments, and values are expressed in mean SD.
measured by the increased number and length of neur- Dmt1 isoforms control cell fates in intestinal
ites (Fig. 6B,C). Interestingly, under noninducing basal organoids oppositely
conditions, Dmt1-ire knockdown (~ 90%, Fig. 6D)
Notch signalling is essential for intestinal cell renewal
resulted in an increased number of neurites that were
and differentiation [31–33]. To study Dmt1 function in
further enhanced upon cAMP stimulation and compa-
intestinal homeostasis, we perturbed Dmt1 isoform
rable with GSI treatment. Silencing of the Dmt1+ire
expression in mouse intestinal organoids using shRNA
isoforms (~ 65%, Fig. 6D) showed no significant
knockdown. Wild-type intestinal organoids expressing
reduction in neuronal differentiation. To test whether
both Dmt1a and Dmt1b isoforms (Fig. 8A, Fig. S4C)
the increased neurite formation in Dmt1-ire knock-
showed a mixed population of immature (sphere-like)
down could be suppressed by ectopic Notch1, we
and mature (crypt-like) organoids (Fig. 8B). Macro-
transduced Neuro2A shDmt1-ire cells with lentiviral
scopically, Dmt1-ire knockdown (~ 52%) resulted in
Nicd1-GFP (Fig. 7A). Nicd-overexpression (GFP+) in
mature crypt-villus-containing mini-guts, while knock-
Dmt1-ire knockdown Neuro2A cells strongly impaired
down of Dmt1+ire (~ 63%) resulted in sphere-like
neurite formation (Fig. 7B) and neurite length
organoids (Fig. 8B–D). Moreover, shDmt1+ire orga-
(Fig. 7C) under both basal and inducing conditions
noids showed a significantly higher replating capacity
compared with GFPneg shDmt1-ire Neuro2A cells.
than both shScr and shDmt1-ire knockdown orga-
These findings demonstrate that constitutive Nicd1
noids, consistent with their more immature phenotype
expression is sufficient for blocking the accelerated
(Fig. 8E). Typically, both shScr and shDmt1-ire intes-
neuronal differentiation induced by the loss of Dmt1-
tinal organoids displayed columnar organization,
ire expression and that Dmt1 isoforms function
microvilli and goblet cells: features lacking in
upstream of Nicd production.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 7
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
Fig. 5. Dmt1 controls Notch1-mediated myoblast differentiation. (A) Bright-field representative images of C2C12 cells differentiated for
6 days. Scale bar: 100 lm. (B) Dmt1-ire and Dmt1+ire mRNA expression in C2C12 cells with stable knockdown of scrambled, Dmt1-ire or
Dmt1+ire analysed by qRT-PCR [one-way ANOVA (Tukey comparison), **P < 0.01, ***P < 0.001, significantly compared with scrambled
control]. Csnk2a2 mRNA expression was used as a housekeeping control. (C) Immunoblot analysis of Msf and lamin A (loading control) pro-
tein levels in undifferentiated (Day 0) and differentiated (Days 1–6) C2C12 cells with stable knockdown of scrambled (treated with GSI),
Dmt1-ire or Dmt1+ire. (D) qRT-PCR for the mRNA expression of differentiation markers MyoD, MyoG, Myf5 and Notch target gene Hey1 at
4 days postdifferentiation initiation in C2C12 cells with knockdown of scrambled treated with DMSO or GSI, Dmt1-ire or Dmt1+ire. Csnk2a2
mRNA expression was used as a housekeeping control [one-way ANOVA (Tukey comparison), **P < 0.01, ***P < 0.001, significantly com-
pared with the DMSO-treated scrambled control]. GSI, c-secretase-inhibitor dibenzazepine; Msf, myosin skeletal FAST; Myf5, myogenic fac-
tor 5; MyoD, myoblast determination protein 1; MyoG, myogenin. Data are representative of three independent experiments, and values are
expressed in mean SD.
Dmt1+ire knockdown organoids (Figs 8F and 9A). expression of Trop2 is associated with worse outcomes
Silencing of Dmt1-ire led to reduced expression of in cancer patients, we examined whether Dmt1 also
Notch targets, Hes1 and c-Myc (Fig. 9B). In contrast, regulates Notch signalling in human cancer cells.
in immature Dmt1+ire knockdown organoids, we Human colorectal adenocarcinoma (LS174T) cells only
observed increased expression of the Wnt/b-catenin express Dmt1b isoforms (Fig. S4D) and retain the
stem cell marker, Ascl2 (Fig. 9B) and fetal endodermal ability to differentiate into secretory cells, including
stem/progenitor cell genes, Spp1, Trop2 and Cnx43 goblet cells, upon treatment with GSI [34,35]. In
(Fig. 9C). These results confirm that isoform-specific Dmt1-ire knockdown (~83%) LS174T cells (Fig. 10A),
loss of Dmt1 results in either the induction (shDmt1- we observed an increased number of goblet cells
ire) or the suppression (shDmt1+ire) of differentiation (Fig. 10B), elevated Muc5ac mRNA (Fig. 10C) and a
through Notch signalling. decrease in the expression of the intestinal stem cell
marker Olfm4 (Fig. 10D) and Notch target genes,
Hes1 and Hes4 (Fig. 10E). The phenotype of Dmt1-ire
Dmt1 controls Notch activity in human colorectal
LS174T cells strongly resembled those after treatment
cancer cells and Apc/ organoids
with GSI. Moreover, immunoblot analysis revealed
Because Notch deregulation is frequently observed in increased Val1744-cleaved Notch1 upon Dmt1-ire
leukaemia and many solid cancers and the increased silencing (Fig. 10F). We confirmed these findings using
8 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
Fig. 6. Dmt1 regulates Notch-mediated neuronal differentiation. (A) Bright-field representative images of basal (left, cAMP) and differenti-
ated (right, +cAMP) Neuro2A cells with knockdown of scrambled (treated with DMSO or GSI), Dmt1-ire, or Dmt1+ire. Scale bar: 50 lm. (B,
C) Quantification of the number of neurites per cell and neurite length under basal and differentiated conditions (one-way ANOVA [Tukey
comparison], **P < 0.01, ***P < 0.001, significantly compared with the DMSO-treated scrambled control). (D) Dmt1-ire and Dmt1+ire
mRNA expression in Neuro2A cells with stable knockdown of scrambled, Dmt1-ire or Dmt1+ire measured by qRT-PCR. Csnk2a2 mRNA
expression was used as a housekeeping control (one-way ANOVA (Tukey comparison), ns, nonsignificant, **P < 0.01, ***P < 0.001, signifi-
cantly compared with scrambled control).
an independent shRNA, targeting a different region of by increasing fetal and stem cell identity and blocking epi-
Dmt1-ire (#2) (Fig. 10A,E). Taken together, the Dmt1- thelial differentiation.
ire isoform sustains high Notch activity in colorectal Lastly, we investigated the clinical relevance of
cancer cells. Dmt1 isoform expression in a cohort of 113 colorectal
Because most human colorectal cancers have perturba- peritoneal cancer metastases (RSEM) [36]. We selected
tion of the Wnt/b-catenin pathway, we utilized murine two transcripts [ENST00000262052.9_2 (Dmt1+ire)
Apc/-derived intestinal organoids to model colorectal and ENST00000644495.1_1 (Dmt1-ire)] in GENCODE
adenoma. Apc/ organoids showed a sphere-like pheno- that were expressed throughout the cohort (Fig. S5B).
type (Fig. 11A), low expression of intestinal differentia- Interestingly, there is no direct correlation between the
tion markers (Fig. S5A), and increased Trop2 and Hes1 expression of the two isoforms in this dataset
expression (Fig. 11B), analogous to wild-type organoids (R = 0.093, P = 0.328, Fig. S5C). When we analysed
with knockdown of Dmt1+ire. In line with this, both the ratio of Dmt1-ire/Dmt1+ire expression, we identi-
Dmt1a+ire and Dmt1b+ire expression was diminished in fied a significant correlation with expression of the
Apc/ compared with Apc+/+ intestinal organoids Notch target gene Hes4. Furthermore, when using
(Fig. 11C). Altogether, decreased Dmt1+ire expression in Hes4 as a proxy for Notch activity, the Dmt1+ire and
intestinal organoids results in altered intestinal cell fates Dmt1-ire isoform expression in this dataset shows
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 9
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
opposite correlations (Fig. 11D). For Dmt1-ire, a sig- In line with our findings, loss of Apc in colorectal
nificant positive association can be seen only in BRAF cancer results in the accumulation of iron resulting
mutated CRC (R = 0.401 P = 0.017). In addition, sig- from increased expression of Tfr1 and Dmt1 through
nificant positive correlations were also found for other Wnt pathway activation [37]. Furthermore, inhibitors
canonical Notch target genes, including Hey1 and of Dmt1 selectively target cancer stem cells, which are
Hey2 with Dmt1 isoform ratio (Dmt1-ire/+ire) addicted to iron, inducing ferroptosis by blocking lyso-
(Table S2). Together, these data further support the somal iron translocation [38]. Finally, Dmt1 is a prom-
relevance of distinct Dmt1 isoforms and link its ising reporter protein for tracking human neural
expression to Notch activation in cancer patients. progenitor/stem cells by tracking manganese [39], and
iron is essential for human pluripotent stem cells [40].
We observed increased Trop2 expression in Dmt1+ire
Discussion
knockdown organoids and Apc/ organoids. The
We identified Dmt1 as an essential novel regulator of findings are consistent with studies in colorectal cancer
Notch signalling and demonstrated that Dmt1 showing that Trop2 expression is associated with high
isoforms are opposite regulators of Notch-mediated Notch activity, a more malignant phenotype and worse
self-renewal and cell fate. Interestingly, our study high- outcomes [41,42]. Thus, our data show an important
lights the responsiveness of Dmt1 expression to iron, role for Dmt1 isoforms and iron in stem cell mainte-
through the presence or absence of the iron response nance and demonstrate that these functions involve
element, as a critical factor in Dmt1 regulation. There- Notch activity regulation.
fore, the Dmt1-ire isoform is necessary to sustain Interestingly, we observed high Val1744-cleaved
Notch signalling, while Dmt1 iron-sensitive isoforms Nicd1 while Notch target genes were decreased in Dmt1
are potent suppressors of Notch activity. Furthermore, KO MEFs and upon silencing of Dmt1-ire in LS174T
the specific modulation of different isoforms leads to cells. We previously showed that blocking the vesicular
opposite effects on Notch-dependent cellular fate in trafficking in cells using chloroquine, or V-ATPase inhi-
several cell types and in organoid mini-guts where bition also led to high levels of Val1744-cleaved Nicd1
Dmt1 controls the balance between cell renewal and in Notch1-mutated T-ALL cells without Notch target
differentiation. gene activation [12]. These findings emphasize that
10 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
Fig. 8. Dmt1 isoforms regulate mouse intestinal organoid differentiation and stemness. (A) Primary mouse intestinal organoids were
generated from Villin-Cre mice. A mixed population of immature and mature organoids was obtained after stem cell enrichment. (B) Repre-
sentative images of mouse intestinal organoids generated after viral transduction with shScrambled (shScr), shDmt1-ire, and shDmt1+ire at
the fourth replate. Scale bar: 50 lm. (C) Dmt1-ire and Dmt1+ire mRNA expression in mouse intestinal organoids with stable knockdown of
Scrambled (Scr), Dmt1-ire or Dmt1+ire analysed by qRT-PCR [one-way ANOVA (Tukey comparison), **P < 0.01, ***P < 0.001, significantly
compared with scrambled control]. Csnk2a2 mRNA expression was used as a housekeeping control. (D) Mature (left) and immature (right)
organoid counts in all replates combined from two donor mice [one-way ANOVA (Tukey comparison), **P < 0.01, ***P < 0.001]. (E) Total
number of organoids in final replate [one-way ANOVA (Tukey comparison), *P < 0.05, **P < 0.01]. (F) Representative electron microscopic
images of mouse intestinal organoids transduced with hairpins against Scr, Dmt1-ire and Dmt1+ire. shScr (mature and immature) and
shDmt1-ire organoids showed columnar organization (top panel) and well-established microvilli (bottom panel), both of which were lost in
shDmt1+ire organoids. Scale bar: 5 lm. The experiment was performed using intestinal stem cells obtained from two individual donor mice,
and values are expressed in mean SD.
Val1744-Notch1 cleavage is not sufficient for Notch1 transcriptional activity and caution against the use of
activity and that subsequent intracellular events are rate Val1744 as a single biomarker for Notch1 activity.
limiting [12,43]. Since Nicd1 rescues the phenotype of While ectopic Nicd1 expression rescued the Notch sig-
Dmt1 KO MEFs, Dmt1 acts upstream or parallel to the nalling defects in Dmt1 KO and Neuro2A shDmt1-ire,
NICD/RBP-jk transcriptional regulation. Therefore, overexpression of a Notch1 gain-of-function T-ALL
our work demonstrates that c-secretase cleavage and mutant in Adam10/17/ deficient cells was not suffi-
Val1744-Nicd1 expression are insufficient for Notch1 cient to restore Notch activity upon Dmt1-ire silencing.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 11
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
This observation is consistent with the identification of expression upon loss of Dmt1, we propose that Dmt1
the shRNA targeting the Dmt1-ire isoform in our screen. function involves Notch regulation in vesicles and
Consistent with the increase in Notch1 surface trafficking.
12 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
Fig. 9. Dmt1 isoforms regulate Notch signalling and fetal transcript programs in mouse intestinal organoids. (A) Toluidine blue and electron
microscopy was used to detect goblet cells (black arrows) in primary mouse intestinal organoids with stable knockdown of Scr, Dmt1-ire or
Dmt1+ire. Scale bar: 5 lm. The experiment was performed using stem cells derived from two individual donor mice, unless stated other-
wise, and values are expressed in mean SD. (B) mRNA expression levels of Notch target genes, Hes1 and c-Myc, and Wnt target gene
Ascl2 in shScr, shDmt1-ire and shDmt1+ire mouse intestinal organoids measured by qRT-PCR. Csnk2a2 mRNA expression was used as a
housekeeping control [one-way ANOVA (Tukey comparison), *P < 0.05, ***P < 0.001, significantly compared with shScr organoids]. (C) Fetal
transcript program genes Spp1, Trop2 and Cnx43 mRNA expression in shScr, shDmt1-ire and shDMt1+ire mouse intestinal organoids, mea-
sured by qRT-PCR. Csnk2a2 mRNA expression was employed as a housekeeping control [one-way ANOVA (Tukey comparison), *P < 0.05,
**P < 0.01, ***P < 0.001, significantly compared with shScr organoids]. The experiment was performed using intestinal stem cells obtained
from two individual donor mice, and values are expressed in mean SD.
Fig. 10. Dmt1 regulates Notch signalling and differentiation in human colorectal cancer cells. (A) qRT-PCR for Dmt1-IRE and Dmt1+IRE
mRNA expression in LS174T cells with stable knockdown of scrambled (shSCR) or Dmt1-IRE (sh-IRE) [one-way ANOVA (Tukey comparison),
ns, nonsignificant, **P < 0.01, ***P < 0.001, significantly compared with scrambled control]. An independent hairpin targeting another
sequence of Dmt1-IRE was used as a control (sh-IRE#2). GAPDH mRNA expression was used as a housekeeping control. (B) Periodic Acid–
Schiff (PAS) staining in LS174T cells treated with DMSO or GSI for 7 days or LS174T cells with stable knockdown of SCRAMBLED (SCR)
or Dmt1-ire. Scale bar: 50 lm. (C) qRT-PCR for mRNA levels of MUC5AC in shSCR cells treated with DMSO or GSI for 7 days and shDmt1-
IRE cells. GAPDH was used as a housekeeping control [one-way ANOVA (Tukey comparison), **P < 0.01, significantly compared with the
DMSO control]. (D) OLFM4 mRNA expression in LS174T cells with knockdown of Dmt1-IRE, measured by qRT-PCR. GAPDH was employed
as a housekeeping control (Student t-test, ***P < 0.001, significantly compared with the scrambled control). (E) qRT-PCR for Notch target
gene (HES1, HES4) mRNA expression in LS174T cells with knockdown of SCR (treated with DMSO or GSI for 7 days) or Dmt1-IRE (and
independent hairpin sh-IRE #2). GAPDH was used as a housekeeping control [one-way ANOVA (Tukey comparison), **P < 0.01,
***P < 0.001, significantly compared with the scrambled control]. (F) Immunoblot analysis of protein levels of Notch1 (C-20), Val1744 and
lamin A (loading control) in LS174T cells with knockdown of SCR (treated with DMSO or GSI for 7 days) or Dmt1-IRE. GSI, c-secretase
inhibitor dibenzazepine; NEXT, Notch1 extracellular truncation; TMIC, transmembrane/intracellular fragment. Data are representative of three
independent experiments, and values are expressed in mean SD.
Dmt1 loss disrupted vesicular trafficking, increased Eea1-positive endosomes, surface proteins are trans-
endocytosis, increased expression of endo- and lyso- ported directly from the cell surface to the nucleus
somal markers, and increased basal autophagic flux. In [44]. The increased levels of Eea1 upon loss of Dmt1
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 13
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
Fig. 11. Dmt1 isoform expression correlates with Notch target gene expression in peritoneal CRC metastases. (A) Representative bright-
field images of APC+/+ and APC/ mouse intestinal organoids. Scale bar: 50 lm. (B) Hes1 and Trop2 mRNA expression in APC+/+ and
APC/ mouse intestinal organoids, measured by qRT-PCR. Csnk2a2 mRNA expression was utilized as a housekeeping control (Student t-
test, **P < 0.01, ***P < 0.001, significantly compared with APC+/+ organoids). (C) qRT-PCR for mRNA levels of Dmt1 isoforms in Apc+/+ and
Apc/ intestinal organoids. Csnk2a2 mRNA expression was utilized as a housekeeping control. (D) Correlations between Dmt1+ire (R = -
0.533, P = 1.17e-09), the ratio of Dmt1-ire/+ire (R = 0.409, P = 7.05e-06) and Dmt1-ire (R = 0.210, P = 0.026), and HES4 (Notch target gene)
expression (log2) in tumour colon metastases [cohort of 113 colorectal peritoneal cancer metastases (RSEM)]. Data are representative of
three independent experiments, and intestinal stem cells were derived from three individual donor mice per condition (Apc/ and Apc+/+).
Values are expressed in mean SD.
may disrupt full-length and S2-cleaved Notch trans- Dmt1 KO MEFs due to the accumulation and disba-
port and activation. In addition, loss of Dmt1 resulted lance of cytoplasmic and vesicular iron, causing oxida-
in damaged intracellular vesicles with disrupted/miss- tive stress that results in lysosomal membrane
ing membranes and the formation of isolation mem- peroxidation [46]. Lysosomal damage is sensed by
branes, which is characteristic of the formation of cathepsins which trigger lysophagy, the selective autop-
phagophores, an early step in autophagy/lysophagy hagy of lysosomes, which correlates with the elevated
[45]. Lysosomal membrane damage may occur in basal autophagy flux in Dmt1 KO MEFs [45]. These
14 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
findings support observations that blocking vesicle by Adam10 [7,58] and c-secretase activity is modulated
function inhibits Notch-dependent T-ALL growth and by zinc and copper [59]. Further research is needed to
leads to ROS accumulation [12]. Consistently, chloro- establish if and how Notch activity and iron/divalent
quine was less effective in raising the intravesicular pH metal signalling are hardwired.
in Dmt1 KO MEFs, reflecting perturbed trafficking The ubiquitous expression of Dmt1b in many tissues
and a lysosome defect. The c-secretase complex, which during development and in adult tissues [22] may sug-
is present on the plasma membrane, endocytic com- gest a general role of Dmt1 in regulating stem cell
partments and lysosomes [11,47], shows more efficient renewal and cell fate. Our study primarily focused on
S3-cleavage in endocytic compartments, which have a the role of Dmt1 on Notch1 activity. Mammalian cells
lower pH [8,48]. Together with the finding that defects encode for four Notch receptors, and whether Dmt1
in vATP-ase reduce Notch signalling [49], these data isoforms also control these remains unknown. Further
support the notion that perturbations in vesicular traf- research is needed to identify how Dmt1 isoform
ficking and pH caused by loss of Dmt1 result in an expression and iron transport are regulated in tissues
accumulation of inactive Notch/Nicd in vesicles. and their relation to Notch transcriptional activity.
Previous studies have shown that intracellular ROS Although we provide evidence for the critical role
levels regulate Notch signalling in airway basal stem cells played by Dmt1 isoforms in Notch activation cascade,
through Nrf2-dependent Notch signalling [50]. These the precise identity of the vesicles through which Dmt1
observations raise the possibility that ROS levels can regulates Notch cell fates requires further study.
directly inactivate the Notch pathway by reducing NICD Finally, we show that Dmt1 isoform expression in
degradation that could contribute to observed pheno- colorectal cancers may have clinical relevance and
types. Further experiments will be needed to dissect the implication and correlate with expression of canonical
direct and indirect roles of ROS on the Notch pathway. Notch target genes. It has been previously shown that
In this study, we report that Dmt1-ire loss phenoco- tumour expression of DMT1 is elevated in CRC
pies Notch is a loss of function phenotype in muscle, patients [60]; however, Dmt1 isoform expression is not
neurons, intestinal organoids and human cancer tis- assessed. Therefore, further validation in different
sues. However, Dmt1 knockout mice are viable while CRC clinical subtypes is necessary but the role of
Notch1 knockouts are embryonic lethal. This suggests Notch in CRC is supported by literature and the in
that Dmt1 is not important for all tissues that require vitro models presented here. Together, these findings
Notch signalling. Another explanation could be warrant a further investigation into the role of Dmt1
that Dmt1 knockouts do not express any Dmt1 iso- as a biomarker and therapeutic target in cancer. In
form. Dmt1-ire specific isoform knockout mice (retain- addition, our data imply that pharmacological inhibi-
ing Dmt1+ire expression) could address this question. tion of Dmt1-ire isoforms may be employed to curtail
Our findings warrant further investigation into the reg- Notch signalling in cancers. At the same time,
ulation of Notch activity by Dmt1 in specific tissues. Dmt1+ire inhibitors might augment Notch signalling
Others have also shown that Dmt1-ire is expressed in to specify cell fates or promote tissue regeneration.
early, sorting and recycling endosomes [51–53]. Our study To conclude, we identified Dmt1 as a novel regula-
highlighted that Dmt1b-ire is expressed in intracellular tor of Notch activity, which, in an isoform-specific
vesicles with minor co-localization to lysosomes, while manner, determines the fate of the Notch signalling
Dmt1b+ire is almost exclusively located in the lysosomes. cascade between c-secretase cleavage and downstream
Therefore, we speculate that Dmt1b-ire promotes Notch transcriptional activation. Furthermore, our data dem-
intracellular trafficking and receptor processing in early, onstrate that the current paradigm of c-secretase cleav-
sorting and recycling endosomes to promote Notch activ- age equating Notch activation is incomplete. We
ity [54], whereas Dmt1+ire isoform promotes lysosomal propose that Dmt1-containing intracellular vesicles
degradation to turn off Notch activity. If and how iron is represent a new and essential rate-limiting step
controlling Dmt1 isoform localization, expression and through which Notch receptors traverse to control cell
activity requires further study. fate in normal tissues and diseases.
A relation between iron and Notch signalling has
been reported before. For example, iron drives the
Materials and methods
proliferation of Notch-induced T-ALL in mice [55]
and breast cancer cells upregulate Dmt1 to import iron
Compounds
[56]. Beyond iron, Dmt1 transports other divalent
metals, including Zn2+, Co2+, Cu2+ and Mn2+ [57]. Cells were treated with dimethyl sulfoxide (DMSO), 0.2 lM
Importantly, zinc is required for Notch-S2 shedding of c-secretase inhibitor dibenzazepine (DBZ) (Syncom,
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 15
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
16 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
inhibited after c-secretase inhibition (GSI) (Fig. S1A,B) post-transduction, single-cell suspensions were generated and
and is referred to as screening cell line (U9). expanded to obtain single-cell-derived clones.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 17
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
18 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
temperature. After fixation, cells were incubated with peri- Notch1 receptor flow cytometry
odic acid solution (Sigma, cat. #3951) for 5 min at room
temperature. Next, cells were rinsed with MilliQ and Notch1 receptor availability at the cell surface was analysed
stained with Schiff reagent (Sigma, cat. #3952) for 15 min by flow cytometry. Dmt1 WT and KO MEFs were fixed in
at room temperature. Finally, cells were rinsed with tap 4% PFA in 1XPBS and stained with a PE-labelled mouse
water, dehydrated using 80% ethanol and mounted on anti-Notch1 antibody targeting the extracellular domain of
glass slides using DPX mounting medium. Notch1 (1 : 40, BioLegend, San Diego, CA, USA; cat.
#352105) for 15 min at 4 °C. For total Notch1 receptor
expression, Dmt1 WT and KO MEFs were permeabilized
Immunoblotting with saponin-based permeabilization reagent (Thermo
Fisher Scientific, cat. #C10418) before staining. A
Cell lysates were prepared in 1xLaemlli loading buffer. Pro- PE-labelled mouse IgG1j was used as isotype control (Bio-
teins were separated on Tris–HCL SDS/PAGE gels and trans- Legend). After staining, cells were analysed using a FACS-
ferred onto PVDF membranes. Membranes were blocked in CantoII cytometer with bd facsdivaBD FACSDIVA 6.1.1
5% dried skimmed milk (Marvel) and 0.05% Tween20 software. Using FLowV10.1, doublets and cellular debris
in 1XTBS. Protein detection was performed with subsequent were excluded. Mean fluorescent intensity (MFI) was deter-
primary antibodies: rabbit anti-cleaved Notch1 (Val 1744, mined and normalized to the control to obtain the fold
D3B8) (Cell Signalling, Danvers, MA, USA; cat. #4147S, change in extracellular Notch1 receptor expression.
1 : 1000), rabbit anti-lamin A (C-term) (Sigma, cat. #L1293,
1 : 1000), mouse anti-actin clone C4 (MP Biomedicals, Santa
Ana, CA, USA; cat. # 691001, 1 : 20 000), mouse anti-Myc DMT1 isoform expression in CRC peritoneal
(9e10) (3 mgmL1, 1 : 1000), mouse anti-skeletal myosin metastases
FAST (Sigma, cat. #M4276, 1 : 1000), rabbit anti-Notch1
RNA-Seq data from human colorectal cancer patients were
(C-20) (Santa Cruz, Dallas, TX, USA; cat. #sc-6014-R,
described [36] deposited and retrieved from GEO
1 : 1000), rabbit anti-LC3B (MBL, WOBURN, MA, USA;
(GSE190609). Fastq files were mapped to HG19/GRCh37
cat. #PM036, 1 : 1000), mouse anti-Rab5A (Cell Signalling,
using the STAR [73] algorithm, with GENCODE-
cat. #46449, 1 : 1000), rabbit anti-EEA1 (Abcam, cat.
v32_GRCh37 as annotation source. BAM files were subse-
#ab2900, 1 : 1000), mouse anti-vinculin (Sigma, cat.
quently passed to RSEM [74] for isoform quantification
#V9131, 1 : 5000), mouse anti-FLAG M2 (Sigma, cat.
using GENCODE-v32_GRCh37 as annotation source. The
#F3165, 1 : 1000), rabbit anti-HA (Sigma, cat. #H6908,
results were converted into a dataset in the R2: genomics
1 : 1000), rabbit anti-Dmt1 (Novus Biologicals, Centennial,
analysis and visualization platform (http://r2.amc.nl). R2
CO 80112, USA; cat. #NBP2-30045, 1 : 1000), mouse anti-
was used for subsequent analyses and visualizations.
transferrin receptor (Invitrogen, cat. #13-6800, 1 : 1000)
and rabbit anti-LAMP1 (Abcam, cat. #ab24170, 1 : 1000).
Secondary antibodies used were anti-mouse (Cell Signal- Iron (Fe2+) and cytoplasmic ROS flow
ling, cat. #7076S, 1 : 5000) or anti-rabbit IgG-horseradish cytometry assay
peroxidase (Cell Signalling, cat. #7074S, 1 : 5000). Amer-
sham ECL Prime Western Blotting Detection Reagent (GE Iron (Fe2+) was detected in Dmt1 WT and KO MEFs with
Healthcare, Chicago, IL, USA) was used for visualization the fluorescent imaging probe BioTracker 575 Red Fe2+
as described by the manufacturer. Dye (Sigma, cat. #SCT030) according to the manufac-
turer’s protocol. In short, cell culture media was removed,
and cells were washed with 1xHBSS and subsequently
Quantitative RT-PCR stained with 5 lM BioTracker 575 Red Fe2+ dye for 1 h at
37 °C. Next, cells were washed with 1xHBSS analysed by
Total RNA was isolated using NucleoSpin RNA flow cytometry. Cytoplasmic ROS levels were analysed
(Macherey-Nagel, D€ uren, Germany) from cells according by incubating Dmt1 WT and KO MEFs [pretreated with
to the manufacturer’s protocol. cDNA was obtained using 10 mM of N-Acetylcysteine (NAC, Sigma Aldrich) for
iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA) 20 min at 37 °C in complete medium] with 5 lM of Cell-
followed by SYBR-green-based reverse transcription quan- ROX Deep Red Reagent (Thermo Fisher, cat. # C10422)
titative PCR (qRT-PCR) using SensiMix SYBR high-ROX in serum-free medium at 37 °C for 30 min. Fe2+ and cyto-
kit (GC Biotech, Waddinxveen, The Netherlands). mRNA plasmic ROS levels were analysed by a FACSCantoII cyt-
expression was analysed using of forward and reverse ometer with BD FACSDIVA 6.1.1 software. FlowJo V10.1 was
primers (Tables S4 and S5). In addition, cycle threshold used to: exclude doublets and cellular debris and analyse
(Ct) values were analysed with CFX Connect Real Time the mean fluorescent intensity (MFI). MFI was normalized
System (Bio-Rad) and Csnk2a2 (mouse) and GAPDH to the DMSO control to obtain the fold change in cyto-
(human) were used as housekeeping genes. plasmic ROS levels.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 19
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
Intestinal organoid culture and lentiviral electron microscope equipped with an Eagle 4kx4k CCD
transduction camera (Thermo Fisher Scientific).
20 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
manuscript and designed the figures under the supervi- 10 Tagami S, Okochi M, Yanagida K, Ikuta A, Fukumori
sion of AG and MV. AG, MV, FR and KK contrib- A, Matsumoto N, Ishizuka-Katsura Y, Nakayama T,
uted to the design and implementation of the research. Itoh N, Jiang J et al. (2008) Regulation of Notch
OK, DZ and JK performed the expression analysis on signaling by dynamic changes in the precision of S3
peritoneal metastases. All authors discussed the results cleavage of Notch-1. Mol Cell Biol 28, 165–176.
and commented on the manuscript. 11 Sannerud R, Esselens C, Ejsmont P, Mattera R, Rochin
L, Tharkeshwar AK, de Baets G, de Wever V, Habets
R, Baert V et al. (2016) Restricted location of PSEN2/
Peer review c-secretase determines substrate specificity and
generates an intracellular Ab Pool. Cell 166, 193–208.
The peer review history for this article is available at
12 Hounjet J, Habets R, Schaaf MB, Hendrickx TC,
https://www.webofscience.com/api/gateway/wos/peer-
Barbeau LM, Yahyanejad S, Rouschop KM, Groot AJ
review/10.1111/febs.16946.
& Vooijs M (2019) The anti-malarial drug chloroquine
sensitizes oncogenic NOTCH1 driven human T-ALL to
Data availability statement c-secretase inhibition. Oncogene 38, 5457–5468.
13 Kobia F, Duchi S, Deflorian G & Vaccari T (2014)
The data that support the findings of this study Pharmacologic inhibition of vacuolar H+ ATPase
are available from the corresponding author (marc. reduces physiologic and oncogenic Notch signaling.
[email protected]) upon reasonable request. Mol Oncol 8, 207–220.
14 Schneider M, Troost T, Grawe F, Martinez-Arias A &
Klein T (2013) Activation of Notch in lgd mutant cells
References requires the fusion of late endosomes with the
1 Siebel C & Lendahl U (2017) Notch signaling in lysosome. J Cell Sci 126, 645–656.
development, tissue homeostasis, and disease. Physiol 15 Sethi N, Yan Y, Quek D, Schupbach T & Kang Y
Rev 97, 1235–1294. (2010) Rabconnectin-3 is a functional regulator of
2 Kopan R & Ilagan MXG (2009) The canonical notch mammalian Notch signaling. J Biol Chem 285, 34757–
signaling pathway: unfolding the activation mechanism. 34764.
Cell 137, 216–233. 16 Vaccari T, Duchi S, Cortese K, Tacchetti C & Bilder D
3 Aster JC, Pear WS & Blacklow SC (2017) The varied (2010) The vacuolar ATPase is required for
roles of notch in cancer. Annu Rev Pathol 12, 245–275. physiological as well as pathological activation of the
4 Groot AJ, Cobzaru C, Weber S, Saftig P, Blobel CP, Notch receptor. Development 137, 1825–1832.
Kopan R, Vooijs M & Franzke CW (2013) Epidermal 17 Fleming MD, Trenor CC, Su MA, Foernzler D, Beier
ADAM17 is dispensable for Notch activation. J Invest DR, Dietrich WF & Andrews NC (1997) Microcytic
Dermatol 133, 2286–2288. anaemia mice have a mutation in Nramp2, a candidate
5 Groot AJ, Habets R, Yahyanejad S, Hodin CM, Reiss iron transporter gene. Nat Genet 16, 383–386.
K, Saftig P, Theys J & Vooijs M (2014) Regulated 18 Su MA, Trenor CC III, Fleming JC, Fleming MD &
proteolysis of NOTCH2 and NOTCH3 receptors by Andrews NC (1998) The G185R mutation disrupts
ADAM10 and presenilins. Mol Cell Biol 34, 2822– function of the iron transporter Nramp2. Blood 92,
2832. 2157–2163.
6 Sulis ML, Saftig P & Ferrando A (2011) Redundancy 19 Touret N, Martin-Orozco N, Paroutis P, Furuya W,
and specificity of the metalloprotease system mediating Lam-Yuk-Tseung S, Forbes J, Gros P & Grinstein S
oncogenic NOTCH1 activation in T-ALL. Leukemia 25, (2004) Molecular and cellular mechanisms underlying
1564–1569. iron transport deficiency in microcytic anemia. Blood
7 van Tetering G, van Diest P, Verlaan I, van der Wall E, 104, 1526–1533.
Kopan R & Vooijs M (2009) Metalloprotease 20 Fleming MD, Romano MA, Su MA, Garrick LM,
ADAM10 is required for Notch1 site 2 cleavage. J Biol Garrick MD & Andrews NC (1998) Nramp2 is mutated
Chem 284, 31018–31027. in the anemic Belgrade (b) rat: evidence of a role for
8 Pasternak SH, Bagshaw RD, Guiral M, Zhang S, Nramp2 in endosomal iron transport. Proc Natl Acad
Ackerley CA, Pak BJ, Callahan JW & Mahuran DJ Sci USA 95, 1148–1153.
(2003) Presenilin-1, nicastrin, amyloid precursor protein, 21 Lee PL, Gelbart T, West C, Halloran C & Beutler E
and c-secretase activity are co-localized in the lysosomal (1998) The human Nramp2 gene: characterization of
membrane. J Biol Chem 278, 26687–26694. the gene structure, alternative splicing, promoter region
9 Kaether C, Haass C & Steiner H (2006) Assembly, and polymorphisms. Blood Cells Mol Dis 24, 199–215.
trafficking and function of c-secretase. Neurodegener 22 Hubert N & Hentze MW (2002) Previously
Dis 3, 275–283. uncharacterized isoforms of divalent metal transporter
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 21
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
(DMT)-1: implications for regulation and cellular 35 Milano J, McKay J, Dagenais C, Foster-Brown L,
function. Proc Natl Acad Sci USA 99, 12345–12350. Pognan F, Gadient R, Jacobs RT, Zacco A, Greenberg B
23 Mackenzie B, Takanaga H, Hubert N, Rolfs A & & Ciaccio PJ (2004) Modulation of Notch processing by
Hediger MA (2007) Functional properties of multiple c-secretase inhibitors causes intestinal goblet cell
isoforms of human divalent metal-ion transporter 1 metaplasia and induction of genes known to specify gut
(DMT1). Biochem J 403, 59–69. secretory lineage differentiation. Toxicol Sci 82, 341–358.
24 Garrick MD, Kuo H-C, Vargas F, Singleton S, Zhao 36 Laoukili J, Constantinides A, Wassenaar ECE, Elias
L, Smith JJ, Paradkar P, Roth JA & Garrick LM SG, Raats DAE, van Schelven SJ, van Wettum J,
(2006) Comparison of mammalian cell lines expressing Volckmann R, Koster J, Huitema ADR et al. (2022)
distinct isoforms of divalent metal transporter 1 in a Peritoneal metastases from colorectal cancer belong to
tetracycline-regulated fashion. Biochem J 398, 539– consensus molecular subtype 4 and are sensitised to
546. oxaliplatin by inhibiting reducing capacity. Br J Cancer
25 Bozkulak EC & Weinmaster G (2009) Selective use of 126, 1824–1833.
ADAM10 and ADAM17 in activation of Notch1 37 Radulescu S, Brookes MJ, Salgueiro P, Ridgway RA,
signaling. Mol Cell Biol 29, 5679–5695. McGhee E, Anderson K, Ford SJ, Stones DH, Iqbal
26 Hur JY, Teranishi Y, Kihara T, Yamamoto NG, Inoue TH, Tselepis C et al. (2012) Luminal iron levels govern
M, Hosia W, Hashimoto M, Winblad B, Frykman S & intestinal tumorigenesis after Apc loss in vivo. Cell Rep
Tjernberg LO (2012) Identification of novel c-secretase- 2, 270–282.
associated proteins in detergent-resistant membranes 38 Turcu AL, Versini A, Khene N, Gaillet C, Ca~ neque T,
from brain. J Biol Chem 287, 11991–12005. M€ uller S & Rodriguez R (2020) DMT1 inhibitors kill
27 Wu ZQ, Li D, Huang Y, Chen XP, Huang W, Liu CF, cancer stem cells by blocking lysosomal iron
Zhao HQ, Xu RX, Cheng M, Schachner M et al. translocation. Chemistry 26, 7369–7373.
(2017) Caspr controls the temporal specification of 39 Lewis C, Graves S, Cai W, Nickles R, Meyerand M &
neural progenitor cells through Notch signaling in the Suzuki M (2014) DMT1, a novel PET/MR reporter
developing mouse cerebral cortex. Cereb Cortex 27, protein for neural stem cell tracking. J Nucl Med 55
1369–1385. (Suppl 1), 60.
28 Nofziger D, Miyamoto A, Lyons KM & Weinmaster G 40 Han Z, Yu Y, Xu J, Bao Z, Xu Z, Hu J, Yu M,
(1999) Notch signaling imposes two distinct blocks in Bamba D, Ma W, Ding F et al. (2019) Iron
the differentiation of C2C12 myoblasts. Development homeostasis determines fate of human pluripotent stem
126, 1689–1702. cells via glycerophospholipids-epigenetic circuit.
29 Berkes CA & Tapscott SJ, eds. (2005) MyoD and the Stem Cells 37, 489–503.
transcriptional control of myogenesis. Semin Cell Dev
41 Svec tastna M, Janeckova L, Hrckulak D,
J, S
Biol 16, 585–595. Vojtechova M, Onhajzer J, Krız V, Galuskova K,
30 Franklin J, Berechid B, Cutting F, Presente A,
Sloncov a E, Kubovciak J et al. (2022) TROP2
Chambers C, Foltz DR, Ferreira A & Nye JS (1999) represents a negative prognostic factor in colorectal
Autonomous and non-autonomous regulation of adenocarcinoma and its expression is associated with
mammalian neurite development by Notch1 and features of epithelial–mesenchymal transition and
Delta1. Curr Biol 9, 1448–1457. invasiveness. Cancer 14, 4137.
31 van Es JH, Van Gijn ME, Riccio O, Van Den Born M, 42 Ohmachi T, Tanaka F, Mimori K, Inoue H, Yanaga K
Vooijs M, Begthel H et al. (2005) Notch/c-secretase & Mori M (2006) Clinical significance of TROP2
inhibition turns proliferative cells in intestinal crypts expression in colorectal cancer. Clin Cancer Res 12,
and adenomas into goblet cells. Nature 435, 959–963. 3057–3063.
32 Vooijs M, Ong C-T, Hadland B, Huppert S, Liu Z, 43 Vooijs M, Schroeter EH, Pan Y, Blandford M &
Korving J, van den Born M, Stappenbeck T, Wu Y, Kopan R (2004) Ectodomain shedding and
Clevers H et al. (2007) Mapping the consequence of intramembrane cleavage of mammalian Notch proteins
Notch1 proteolysis in vivo with NIP-CRE. Development are not regulated through oligomerization. J Biol Chem
134, 535–544. 279, 50864–50873.
33 Vooijs M, Liu Z & Kopan R (2011) Notch: architect, 44 Chaumet A, Wright GD, Seet SH, Tham KM, Gounko
landscaper, and guardian of the intestine. NV & Bard F (2015) Nuclear envelope-associated
Gastroenterology 141, 448–459. endosomes deliver surface proteins to the nucleus. Nat
34 Yang Y, Zhu R, Bai J, Zhang X, Tian Y, Li X, Peng Commun 6, 8218.
Z, He Y, Chen L, Ji Q et al. (2011) Numb modulates 45 Papadopoulos C & Meyer H (2017) Detection and
intestinal epithelial cells toward goblet cell phenotype clearance of damaged lysosomes by the endo-lysosomal
by inhibiting the Notch signaling pathway. Exp Cell damage response and lysophagy. Curr Biol 27, R1330–
Res 317, 1640–1648. R1341.
22 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
J. Hounjet et al. Dmt1 controls Notch-mediated cell fates
46 Kurz T, Terman A, Gustafsson B & Brunk UT (2008) 59 Gerber H, Wu F, Dimitrov M, Osuna GMG &
Lysosomes in iron metabolism, ageing and apoptosis. Fraering PC (2017) Zinc and copper differentially
Histochem Cell Biol 129, 389–406. modulate amyloid precursor protein processing by c-
47 Small SA & Gandy S (2006) Sorting through the cell secretase and amyloid-b peptide production. J Biol
biology of Alzheimer’s disease: intracellular pathways Chem 292, 3751–3767.
to pathogenesis. Neuron 52, 15–31. 60 Xue X, Ramakrishnan SK, Weisz K, Triner D, Xie L,
48 Marshansky V & Futai M (2008) The V-type H+- Attili D, Pant A, Gy} orffy B, Zhan M, Carter-Su C
ATPase in vesicular trafficking: targeting, regulation et al. (2016) Iron uptake via DMT1 integrates cell cycle
and function. Curr Opin Cell Biol 20, 415–426. with JAK-STAT3 signaling to promote colorectal
49 Yan Y, Denef N & Sch€ upbach T (2009) The vacuolar tumorigenesis. Cell Metab 24, 447–461.
proton pump, V-ATPase, is required for Notch 61 Moffat J, Grueneberg DA, Yang X, Kim SY, Kloepfer
signaling and endosomal trafficking in drosophila. Dev AM, Hinkle G, Piqani B, Eisenhaure TM, Luo B,
Cell 17, 387–402. Grenier JK et al. (2006) A lentiviral RNAi library for
50 Paul MK, Bisht B, Darmawan DO, Chiou R, Ha VL, human and mouse genes applied to an arrayed viral
Wallace WD, Chon AT, Hegab AE, Grogan T, high-content screen. Cell 124, 1283–1298.
Elashoff DA et al. (2014) Dynamic changes in 62 Sarbassov DD, Guertin DA, Ali SM & Sabatini DM
intracellular ROS levels regulate airway basal stem cell (2005) Phosphorylation and regulation of Akt/PKB by
homeostasis through Nrf2-dependent Notch signaling. the rictor-mTOR complex. Science 307, 1098–1101.
Cell Stem Cell 15, 199–214. 63 Barger CJ, Branick C, Chee L & Karpf AR (2019) Pan-
51 Tabuchi M, Tanaka N, Nishida-Kitayama J, Ohno H cancer analyses reveal genomic features of FOXM1
& Kishi F (2002) Alternative splicing regulates the overexpression in cancer. Cancer 11, 251.
subcellular localization of divalent metal transporter 1 64 Ong C-T, Cheng H-T, Chang L-W, Ohtsuka T,
isoforms. Mol Biol Cell 13, 4371–4387. Kageyama R, Stormo GD & Kopan R (2006) Target
52 Lam-Yuk-Tseung S & Gros P (2006) Distinct targeting selectivity of vertebrate Notch proteins collaboration
and recycling properties of two isoforms of the iron between discrete domains and CSL-binding site
transporter DMT1 (NRAMP2, Slc11A2). Biochemistry architecture determines activation probability. J Biol
45, 2294–2301. Chem 281, 5106–5119.
53 Tabuchi M, Yoshimori T, Yamaguchi K, Yoshida T & 65 Shahi P, Seethammagari MR, Valdez JM, Xin L &
Kishi F (2000) Human NRAMP2/DMT1, which Spencer DM (2011) Wnt and Notch pathways have
mediates iron transport across endosomal membranes, interrelated opposing roles on prostate progenitor cell
is localized to late endosomes and lysosomes in HEp-2 proliferation and differentiation. Stem Cells 29, 678–688.
cells. J Biol Chem 275, 22220–22228. 66 Garrick MD, Zhao L, Roth JA, Jiang H, Feng J, Foot
54 Yamamoto S, Charng W-L & Bellen HJ (2010) NJ, Dalton H, Kumar S & Garrick LM (2012) Isoform
Endocytosis and intracellular trafficking of Notch and specific regulation of divalent metal (ion) transporter
its ligands. Curr Top Dev Biol 92, 165–200. (DMT1) by proteasomal degradation. Biometals 25,
55 Khwaja SS, Liu H, Tong C, Jin F, Pear WS, Van 787–793.
Deursen J & Bram RJ (2010) HIV-1 rev–binding 67 Dull T, Zufferey R, Kelly M, Mandel R, Nguyen M,
protein accelerates cellular uptake of iron to drive Trono D & Naldini L (1998) A third-generation
Notch-induced T cell leukemogenesis in mice. J Clin lentivirus vector with a conditional packaging system. J
Invest 120, 2537–2548. Virol 72, 8463–8471.
56 Jiang XP, Elliott RL & Head JF (2010) Manipulation 68 Stewart SA, Dykxhoorn DM, Palliser D, Mizuno H,
of iron transporter genes results in the suppression of Yu EY, An DS, Sabatini DM, Chen IS, Hahn WC,
human and mouse mammary adenocarcinomas. Sharp PA et al. (2003) Lentivirus-delivered stable gene
Anticancer Res 30, 759–765. silencing by RNAi in primary cells. RNA 9, 493–501.
57 Shawki A, Anthony SR, Nose Y, Engevik MA, 69 Rose PP, Hanna SL, Spiridigliozzi A, Wannissorn N,
€
Niespodzany EJ, Barrientos T, Ohrvik H, Worrell RT, Beiting DP, Ross SR, Hardy RW, Bambina SA, Heise
Thiele DJ & Mackenzie B (2015) Intestinal DMT1 is MT & Cherry S (2011) Natural resistance-associated
critical for iron absorption in the mouse but is not macrophage protein is a cellular receptor for sindbis
required for the absorption of copper or manganese. Am virus in both insect and mammalian hosts. Cell Host
J Physiol Gastrointest Liver Physiol 309, G635–G647. Microbe 10, 97–104.
58 Mumm JS, Schroeter EH, Saxena MT, Griesemer A, 70 Fiddes IT, Lodewijk GA, Mooring M, Bosworth CM,
Tian X, Pan DJ, Ray WJ & Kopan R (2000) A ligand- Ewing AD, Mantalas GL, Novak AM, van den Bout
induced extracellular cleavage regulates gamma- A, Bishara A, Rosenkrantz JL et al. (2018) Human-
secretase-like proteolytic activation of Notch1. Mol Cell specific NOTCH2NL genes affect Notch signaling and
5, 197–206. cortical neurogenesis. Cell 173, 1356–1369.e22.
The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of 23
Federation of European Biochemical Societies.
17424658, 0, Downloaded from https://febs.onlinelibrary.wiley.com/doi/10.1111/febs.16946, Wiley Online Library on [15/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dmt1 controls Notch-mediated cell fates J. Hounjet et al.
71 Foot NJ, Dalton HE, Shearwin-Whyatt LM, Dorstyn of Lgr5+ intestinal stem cells and their progeny. Nat
L, Tan S-S, Yang B & Kumar S (2008) Regulation of Methods 11, 106–112.
the divalent metal ion transporter DMT1 and iron
homeostasis by a ubiquitin-dependent mechanism
involving Ndfips and WWP2. Blood 112, 4268–4275.
Supporting information
72 Langen RC, Schols AM, Kelders MC, Wouters EF & Additional supporting information may be found
Janssen-Heininger YM (2003) Enhanced myogenic online in the Supporting Information section at the end
differentiation by extracellular matrix is regulated at the of the article.
early stages of myogenesis. In Vitro Cell Dev Biol Anim
39, 163–169. Table S1. Pathway analysis shRNA screen.
73 Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski
Table S2. Correlation analysis of log2 ratio Dmt-Ire/
C, Jha S, Batut P, Chaisson M & Gingeras TR (2013)
+Ire and Notch target gene expression.
STAR: ultrafast universal RNA-seq aligner.
Table S3. Targeting sequences of shRNAs.
Bioinformatics 29, 15–21.
Table S4. Mouse qRT-PCR primer sequences.
74 Li B & Dewey CN (2011) RSEM: accurate transcript
Table S5. Human qRT-PCR primer sequences.
quantification from RNA-Seq data with or without a
reference genome. BMC Bioinformatics 12, 323.
Fig. S1. Loss of Dmt1 results in decreased Notch
75 Barker N, van Es JH, Kuipers J, Kujala P, van den signalling.
Born M, Cozijnsen M, Haegebarth A, Korving J, Fig. S2. Ectopic activation of Notch1 rescues loss of
Begthel H, Peters PJ et al. (2007) Identification of stem Notch signalling in Dmt1 KO MEFs.
cells in small intestine and colon by marker gene Lgr5. Fig. S3. Dmt1 controls Notch1 signalling in an iso-
Nature 449, 1003–1007. form-dependent manner.
76 Ladang A, Rapino F, Heukamp LC, Tharun L, Fig. S4. Dmt1 isoform expression in cell lines and
Shostak K, Hermand D, Delaunay S, Klevernic I, Jiang mouse intestinal organoids.
Z, Jacques N et al. (2015) Elp3 drives Wnt-dependent Fig. S5. Loss of APC in intestinal organoids results in
tumor initiation and regeneration in the intestine. J Exp diminished differentiation correlating with Dmt1 in
Med 212, 2057–2075. human colorectal cancer.
77 Yin X, Farin HF, van Es JH, Clevers H, Langer R &
Karp JM (2014) Niche-independent high-purity cultures
24 The FEBS Journal (2023) ª 2023 The Authors. The FEBS Journal published by John Wiley & Sons Ltd on behalf of
Federation of European Biochemical Societies.