*For correspondence:
michael-stout@ouhsc.edu
Competing interests: The
authors declare that no
competing interests exist.
Funding:
See page 21
Received: 03 June 2020
Accepted: 07 December 2020
Published: 08 December 2020
Reviewing editor: Rochelle
Buffenstein, Calico Life Sciences,
LLC, United States
Copyright Mann et al. This
article is distributed under the
terms of the
Creative Commons
Attribution License,
which
permits unrestricted use and
redistribution provided that the
original author and source are
credited.
Health benefits attributed to 17a-
estradiol, a lifespan-extending compound,
are mediated through estrogen
receptor a
Shivani N Mann
1,2,3
, Niran Hadad
4
, Molly Nelson Holte
5
, Alicia R Rothman
1
,
Roshini Sathiaseelan
1
, Samim Ali Mondal
1
, Martin-Paul Agbaga
3,6,7
,
Archana Unnikrishnan
3,8
, Malayannan Subramaniam
5
, John Hawse
5
,
Derek M Huffman
9
, Willard M Freeman
2,10,11
, Michael B Stout
1,2,3
*
1
Department of Nutritional Sciences, University of Oklahoma Health Sciences
Center, Oklahoma City, United States;
2
Oklahoma Center for Geroscience,
University of Oklahoma Health Sciences Center, Oklahoma City, United States;
3
Harold Hamm Diabetes Center, University of Oklahoma Health Sciences Center,
Oklahoma City, United States;
4
The Jackson Laboratory, Bar Harbor, United States;
5
Department of Biochemistry and Molecular Biology, Mayo Clinic, Rochester, United
States;
6
Department of Cell Biology, University of Oklahoma Health Sciences
Center, Oklahoma City, United States;
7
Dean McGee Eye Institute, University of
Oklahoma Health Sciences Center, Oklahoma City, United States;
8
Department of
Biochemistry and Molecular Biology, University of Oklahoma Health Sciences
Center, Oklahoma City, United States;
9
Department of Molecular Pharmacology,
Albert Einstein College of Medicine, New York, United States;
10
Genes & Human
Disease Research Program, Oklahoma Medical Research Foundation, Oklahoma
City, United States;
11
Oklahoma City Veterans Affairs Medical Center, Oklahoma
City, United States
Abstract Metabolic dysfunction underlies several chronic diseases, many of which are
exacerbated by obesity. Dietary interventions can reverse metabolic declines and slow aging,
although compliance issues remain paramount. 17a-estradiol treatment improves metabolic
parameters and slows aging in male mice. The mechanisms by which 17a-estradiol elicits these
benefits remain unresolved. Herein, we show that 17a-estradiol elicits similar genomic binding and
transcriptional activation through estrogen receptor a (ERa) to that of 17b-estradiol. In addition,
we show that the ablation of ERa completely attenuates the beneficial metabolic effects of 17a-E2
in male mice. Our findings suggest that 17a-E2 may act through the liver and hypothalamus to
improve metabolic parameters in male mice. Lastly, we also determined that 17a-E2 improves
metabolic parameters in male rats, thereby proving that the beneficial effects of 17a-E2 are not
limited to mice. Collectively, these studies suggest ERa may be a drug target for mitigating chronic
diseases in male mammals.
Introduction
Aging is the leading risk factor for most chronic diseases, many of which are associated with declines
in metabolic homeostasis (Lo
´
pez-Otı´n et al., 2013). Metabolic detriments associated with advancing
age are further exacerbated by obesity (
Villareal et al., 2005; Waters et al., 2013), which has risen
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 1 of 30
RESEARCH ARTICLE
substantially in the older population (>65 years) over the past several decades (Flegal et al., 2010;
Flegal et al., 2016). Moreover, obesity in mid-life has been shown to accelerate aging mechanisms
and induce phenotypes more commonly observed in older mammals (
Bischof and Park, 2015;
Horvath et al., 2014; Nevalainen et al., 2017; Yang et al., 2009; Whitmer et al., 2005a;
Whitmer et al., 2005b; Dye et al., 2017). These observations have led many to postulate that obe-
sity may represent a mild progeria syndrome (
Salvestrini et al., 2019; Tzanetakou et al., 2012;
Pe
´
rez et al., 2016; Tchkonia et al., 2010; Stout et al., 2017a). Although it is well established that
dietary interventions, including calorie restriction, can reverse obesity-related metabolic sequelae,
many of these strategies are not well tolerated in older patients due to concomitant comorbidities
(
Villareal et al., 2005; Jensen et al., 2014). Compliance issues across all age groups also remain a
paramount hurdle due to calorie restriction adversely affecting mood, thermoregulation, and muscu-
loskeletal mass (
Dirks and Leeuwenburgh, 2006). These adverse health outcomes demonstrate the
need for pharmacological approaches aimed at curtailing metabolic perturbations associated with
obesity and aging.
17a-estradiol (17a-E2) is one of the more recently studied compounds to demonstrate efficacy
for beneficially modulating obesity- and age-related health outcomes. The NIA Interventions Testing
Program (ITP) found that long-term administration of 17a-E2 extends median lifespan of male mice
in a dose-dependent manner (
Strong et al., 2016; Harrison et al., 2014). Our group has been
exploring potential mechanisms by which 17a-E2 may improve healthspan and extend lifespan in a
sex-specific manner. We have found that 17a-E2 administration reduces calorie intake and regional
adiposity in combination with significant improvements in a multitude of systemic metabolic parame-
ters in both middle-aged obese and old male mice without inducing deleterious effects (
Stout et al.,
2017b
; Steyn et al., 2018; Miller, 2020). Other groups have also determined that lifelong adminis-
tration of 17a-E2 beneficially modulates metabolic outcomes, including glucose tolerance, mTORC2
signaling, and hepatic amino acid composition and markers of urea cycling, which were reported to
be dependent upon the presence of endogenous androgens (
Garratt et al., 2017; Garratt et al.,
2018
). Additionally, multiple lifespan extending compounds, including 17a-E2, exhibit similar modifi-
cations in liver function (
Tyshkovskiy et al., 2019). In all, recent studies by several independent labo-
ratories strongly indicate that the lifespan-extending effects of 17a-E2 are at least associated with, if
not dependent on, metabolic improvements.
Despite the mounting evidence demonstrating that 17a-E2 improves numerous health parame-
ters, the signaling mechanism(s) and primary tissues through which 17a-E2 elicits these benefits
remain unknown. Although 17a-E2 is a naturally occurring enantiomer to 17b-estradiol (17b-E2), it
has been postulated that 17a-E2 signals through a novel uncharacterized receptor (
Toran-Aller-
and, 2005
; Toran-Allerand et al., 2002; Toran-Allerand et al., 2005; Green and Simpkins, 2000)
as opposed to classical estrogen receptors a (ERa) and b (ERb), which is due to 17a-E2 having signif-
icantly reduced binding affinity for ERa and ERb as compared to 17b-E2 (
Edwards and McGUIRE,
1980
; Korenman, 1969; Littlefield et al., 1990; Anstead et al., 1997). For this reason, 17a-E2 is
often referred to as a non-feminizing estrogen (
Green and Simpkins, 2000; Engler-Chiurazzi et al.,
2017
; Kaur et al., 2015). A few studies have suggested that a novel but uncharacterized estrogen
receptor, termed ER-X, may mediate 17a-E2 actions in the brain (
Toran-Allerand, 2005 ; Toran-
Allerand et al., 2002
; Toran-Allerand et al., 2005; Green and Simpkins, 2000), although more
recent studies supporting this hypothesis are lacking in the literature. Similarly, no reports to date
have directly tested whether the doses of 17a-E2 shown to improve healthspan and lifespan in mice
are mediated through ERa and/or ERb.
There is a multitude of data in the diabetes and metabolism literature demonstrating that ERa is
a regulator of systemic metabolic parameters. Although most of these studies have historically been
performed in female mammals, more recent studies have demonstrated that ERa also plays a critical
role in modulating metabolism in male mammals. For instance, Allard and colleagues recently dem-
onstrated that genomic actions of ERa regulate systemic glucose homeostasis in mice of both sexes
and insulin production and release in males (
Allard et al., 2019). Other studies have also determined
that hepatic steatosis and insulin sensitivity, and therefore the control of gluconeogenesis, are regu-
lated through FOXO1 in an ERa-dependent manner in male mice (
Yan et al., 2019 ). Furthermore,
hepatocyte-specific deletion of ERa was sufficient to abrogate similar estrogen-mediated metabolic
benefits (Guillaume et al., 2019; Qiu et al., 2017; Meda et al., 2020). Given that several reports
have linked the administration of 17a-E2 to improvements in metabolic homeostasis, we
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 2 of 30
Research article Medicine
hypothesized that 17a-E2 signals through ERa to modulate hepatic function and systemic metabo-
lism, thereby potentially contributing to the lifespan-extending effects of 17a-E2.
The work outlined in this report sought to determine if ERa is the primary receptor by 17a-E2 sig-
nals and modulates health parameters in mice. We initially determined that 17a-E2 and 17b-E2 elicit
similar genomic actions through ERa. Given that no studies to date have tested the potential role of
ERa in modulating 17a-E2-mediated effects in vivo, we treated obese wild type (WT) and ERa
knockout (ERa KO) littermate mice with 17a-E2 to determine if the ablation of ERa could attenuate
17a-E2-induced benefits on adiposity, metabolic homeostasis, and hepatic function. We found that
the ablation of ERa completely attenuated all beneficial metabolic effects of 17a-E2. Follow-up stud-
ies in male WT rats undergoing hyperinsulinemic-euglycemic clamps revealed that 17a-E2 modulates
hepatic insulin sensitivity following acute exposure. Given the established connection between the
hypothalamus and liver in the modulation of hepatic insulin sensitivity (
Ko
¨
nner et al., 2007;
Ruud et al., 2017; Pocai et al., 2005a; Pocai et al., 2005b; Dodd et al., 2018), coupled with our
data demonstrating ERa-dependency of 17a-E2 actions on metabolic parameters, we speculate that
17a-E2 acts through ERa in the liver and/or hypothalamus to improve metabolic homeostasis in
male mammals.
Results
17a-E2 and 17b-E2 similarly modulate genomic binding and
transcriptional activit y of ERa
Ligand-mediated ERa dimerization leads to nuclear translocation and transcriptional activity. Previ-
ous work has shown that 17a-E2 and 17b-E2 can bind to ERa with different affinities (
Edwards and
McGUIRE, 1980
; Korenman, 1969; Littlefield et al., 1990; Anstead et al., 1997), yet potential dif-
ferences in resultant genomic binding and transcriptional activity between the two ligands remains
unexplored. We assessed ERa DNA binding and transcriptional induction following exposure to 17b-
E2 (10 nM) or 17a-E2 (10 nM or 100 nM) in U2OS cells that stably express ERa following doxycycline
induction. We chose to use these cells because they do not endogenously express any form of ERa
or ERb and have been extensively utilized to elucidate the effects of ERa and ERb agonists and
antagonists on gene expression (
Monroe et al., 2003; Monroe et al., 2005). ChIP-sequencing
revealed peaks of ERa genomic binding in all conditions, that when compared, are qualitatively simi-
lar across treatments (
Figure 1A). Statistically significant differences in ERa binding were deter-
mined by negative binomial regression with a Wald’s pairwise post-hoc comparison (false discovery
rate correction, FDR < 0.05). A total of 21,443 peaks were found to have a significant pairwise post-
hoc comparison between vehicle and 17a-E2 and/or 17b-E2 treated cells. No statistically significant
differences between 17a-E2 and 17b-E2-treated groups were observed. 17a-E2 and 17b-E2 not only
induced ERa binding at the same genomic locations but also to similar magnitudes. Comparing the
levels of increased or decreased ERa binding (as compared to vehicle control) between treatments
demonstrates the consistency of ERa genomic binding regardless of the agonist (
Figure 1B,
Supplementary file 1). The degree of increased or decreased ERa binding was highly similar
between 10 nM 17a-E2 and 10 nM 17b-E2, (Pearson’s r = 0.95, p<0.001) and 100 nM 17a-E2 and 10
nM 17b-E2 (Pearson’s r = 0.96, p<0.001). As expected, ERa-binding sites were enriched for estrogen
response elements (ERE), estrogen-related receptor beta (Esrrb), and estrogen-related receptor
alpha (Erra). Other common motifs found within ER elements, including steroidogenic factor-1 (SF1)
(
Lin et al., 2007), and motif elements of known interacting partners, including retinoid acid recep-
tor:retinoid X receptor (RAR:RXR) (
Lee et al., 1998), were also enriched (Figure 1C). In addition, we
observed enrichment of androgen response elements (ARE) in ERa peaks (
Figure 1C). Of particular
relevance, many of the top enriched motifs identified contained the ERE consensus sequence
TTGAC (
Supplementary file 2). Following motif enrichment, we performed pairwise differential
motif enrichment across all groups to determine if a specific agonist or agonist concentration caused
a differential enrichment of any motifs, as would be suggestive of differential genomic binding. No
differential motif binding was observed across treatment groups indicating that both 17a-E2 and
17b-E2 cause ERa to bind to the same types of genomic elements (Hypergeometric test,
FDR < 0.05).
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 3 of 30
Research article Medicine
í
í
3&
3&
ĮQ0
ĮQ0
ȕQ0
VHKLFOH(W2+
D
Vehicle
(W2+
Į
Q0
Į
Q0
ȕ
Q0
1
3
í í
Coverage
'LIIHUHQWLDOO\([SUHVVHG*HQHV
A
E
í
Vehicle
(W2+
Į
Q0
Į
Q0
ȕ
Q0
Vehicle
(W2+
Į
Q0
Į
Q0
ȕ
Q0
Coverage
Į
Q0
Į
Q0
ȕ
Q0
Vehicle
(W2+
0HDQ&HQWHUHG([SUHVVLRQ
C
TBP3
Nur77
NLP7
PR
RAR:RXR
$5(
SF1
(Ura
(VUUE
T
*
A
C
C
*
T
A
C
T
*
A
A
C
T
*
A
C
T
*
*
A
C
T
*
A
T
C
T
*
C
A
*
T
A
C
T
A
C
*
T
C
*
A
A
C
T
*
C
A
T
*
A
*
C
T
A
*
T
C
C
*
T
A
C
T
*
A
C
T
A
*
A
C
T
*
C
*
A
T
A
T
*
C
C
T
*
A
A
*
T
C
T
*
C
A
T
C
*
A
C
T
*
A
A
C
T
*
C
A
T
*
A
C
*
T
A
T
*
C
*
T
C
A
T
A
C
*
C
A
T
*
A
*
C
T
T
A
C
*
*
T
C
A
*
T
A
C
T
A
*
C
A
*
C
T
*
A
C
T
A
T
C
*
T
C
*
A
T
*
C
A
C
T
*
A
C
A
T
*
C
T
A
*
C
A
*
T
A
*
T
C
C
*
T
A
A
T
*
C
A
T
*
C
T
A
C
*
*
C
A
T
T
C
A
*
*
T
C
A
*
A
T
C
*
T
A
C
C
T
*
A
C
T
A
*
T
C
*
A
C
*
T
A
A
T
*
C
C
*
T
A
A
T
C
*
C
*
A
T
T
A
*
C
*
C
A
T
A
T
C
*
*
C
A
T
A
*
C
T
*
A
T
C
*
A
C
T
A
*
C
T
Vehicle
(W2+
Į
Q0
Į
Q0
ȕ
Q0
T
*
A
C
C
T
*
A
C
T
A
*
T
C
*
A
C
T
*
A
A
T
*
C
C
*
T
A
A
C
T
*
*
C
A
T
*
T
A
C
*
C
A
T
A
T
C
*
*
C
A
T
A
*
C
T
*
A
T
C
*
A
C
T
C
T
A
*
C
T
A
*
*
T
A
C
A
*
T
C
*
A
T
C
*
A
C
T
*
A
C
T
T
A
*
C
T
C
A
*
T
C
A
*
*
A
C
T
A
C
*
T
C
*
T
A
A
C
T
*
A
C
T
*
A
C
T
*
A
*
T
C
C
*
T
A
*
T
C
A
A
C
*
T
C
T
A
*
C
*
T
A
A
*
T
C
*
T
A
C
A
C
*
T
A
C
*
T
A
C
*
T
*
T
C
A
*
T
A
C
T
*
A
C
*
A
C
T
(5(
í
í
í
í
í
ORJSYDOXH
6HTXHQFHVZLWKPRWLI
Coverage
1RUPDOL]HG5HDG&RXQWV
Coverage
1RUPDOL]HG5HDG&RXQWV
17ĮQ0
17ĮQ0
)ROGDYV9HK
)ROGEYV9HK
B
&DOOHG3HDNV
3HDN&RYHUDJH
Figure 1. 17a-E2 and 17b-E2 elicit similar genomic binding and transcriptional profiles through ERa. (A) Heatmap representing normalized genome-
wide DNA binding by ERa via ChIP sequencing analyses centered according to peak summits for each treatment group and compared to each other
group. (B) Differential binding was identified between vehicle and 17a-E2 or 17b-E2 treatment groups, but no differences were identified between 17a -
E2 and 17b-E2-treated groups (negative binomial regression, followed by Wald test for pairwise comparisons, FDR < 0.05). Fold change in binding
relative to vehicle control was compared between 17a-E2 treatments and 17b-E2. (C) Motif enrichment analysis, filtered for mammalian and non-
overlapping motif groups, showing the top 10 non-redundant enriched sequence motifs across treatment groups (hypergeometric test, FDR < 0.05), (D)
PCA plot of transcriptional profiles by RNA sequencing analyses demonstrating clustering of 17a-E2 and 17b-E2 treatment groups together, opposite
from Vehicle-treated group along the first principle component, and (E) Heatmap representing differentially expressed genes (negative binomial
regression, followed by Wald test for pairwise comparisons, FDR < 0.05) by RNA sequencing analyses (left) and ERa binding patterns within the gene
body ±5 kb flanking regions of these genes via ChIP sequencing (right). Significant differential pairwise expression was observed only between 17a-E2
and 17b-E2 treatment groups and vehicle-treated control. These studies utilized U2OS-ERa cells treated with low dose (10 nM) 17a-E2, high dose (100
nM) 17a-E2, 17b-E2 (10 nM), or vehicle (EtOH). n = 3/group.
Figure 1 continued on next page
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 4 of 30
Research article Medicine
Next, we examined potential differences in transcriptional responses between treatment groups
using RNA-sequencing. Principle component analysis based on the entire transcriptome revealed
that all samples exposed to either 17a-E2 or 17b-E2 clustered together, whereas vehicle-treated
cells remained distinctly separated from treated cells on the first principle component, which
explains the majority of the variance in transcription (70.6%) (
Figure 1D). These data suggest that
treatment vs vehicle is the primary covariate explaining variance in transcriptional profiles, not the
specific agonist. Next, differential expression was assessed between all groups using a negative
binomial regression model with a Wald pairwise post-hoc test. No genes were found to be differen-
tially regulated (FDR < 0.05) between the estrogen treatments. Yet, compared to vehicle-treated
cells, treatment of U2OS cells with either 10 nM or 100 nM 17a-E2 or 10 nM 17b-E2 resulted in
nearly identical gene suppression and activation signatures (
Figure 1E, left). Additionally, both 17a-
E2 or 17b-E2 treatment conditions resulted in higher ERa DNA binding affinity to gene bodies of
these differentially expressed transcripts compared to vehicle treatment, and no differences were
observed between 17 a-E2 and 17b-E2 conditions (
Figure 1E, right) (negative binomial regression
with Wald pairwise post-hoc). These findings led us to postulate that 17a-E2 may be the signaling
through ERa to modulate health parameters in male mice. As such, we subsequently sought to
determine if the ablation of ERa in vivo would mitigate the effects of 17a-E2.
ERa ablation attenuates 17a-E2-mediated benefits on metabolic
parameters in male mice in vivo
To induce obesity and metabolic perturbations in male mice, we administered high-fat diet (HFD) for
several months prior to initiating 17a-E2 treatment. Control mice remained on HFD, whereas 17a-
E2-treated mice were switched to an identical HFD containing 17a -E2. Almost immediately after
17a-E2 treatment began, male WT mice displayed significant reductions in mass (
Figure 2A–B) and
adiposity (
Figure 2C–D). This is aligned with our previous reports demonstrating that 17a-E2 admin-
istration quickly reduces body mass and adiposity (
Stout et al., 2017b; Steyn et al., 2018;
Miller, 2020), which we have linked to hypothalamic regulation of anorexigenic signaling pathways
(
Steyn et al., 2018). Indeed, male WT mice in the current study also displayed robust declines in cal-
orie consumption during the first 4 weeks of treatment (
Figure 2E). Conversely, all these benefits
were completely abolished in male mice lacking ERa (ERa KO), thereby confirming that 17a-E2
definitively acts through ERa to modulate feeding behaviors, mass, and adiposity in male mice.
Given the close association between adiposity and metabolic homeostasis, coupled with our previ-
ous work demonstrating the ability of 17a-E2 to improve metabolic parameters (Stout et al.,
2017b
; Steyn et al., 2018), we also assessed several metabolic variables in these studies. Similar to
the mass and adiposity data described above, male WT mice receiving 17a-E2 displayed significant
improvements in fasting insulin (
Figure 3B), HbA1C (Figure 3C), and glucose tolerance (Figure 3D–
E
, Figure 3—figure supplement 1), whereas male ERa KO mice receiving 17a-E2 failed to recapitu-
late these findings. Interestingly, despite the masses of the male WT 17a-E2 treatment group being
nearly 15 grams greater than those of the male WT chow-fed controls, glucose tolerance was essen-
tially identical between these groups, thereby indicating that 17a-E2 restores metabolic flexibility in
the presence of obesity in male mice (
Figure 3D–E, Figure 3—figure supplement 1). We also evalu-
ated the effects of 17a-E2 on metabolic parameters in female WT and ERa KO mice provided a stan-
dard chow diet. In contrast to the males, we chose not to subject female WT and ERa KO mice to
HFD because female ERa KO mice spontaneously develop obesity due to the ablation of ERa
(Manrique et al., 2012; Vidal et al., 1999). Given that the female ERa KO mice are already in a chal-
lenged state, HFD would further exacerbate mass and adiposity differences between ERa KO and
WT female mice. We found that 17a-E2 failed to elicit improvements in mass, adiposity, calorie con-
sumption, or metabolic parameters in female mice of either genotype (
Figure 3—figure supple-
ment 2). The positive effects of 17a-E2 in male mice led us to speculate that the liver may play a key
role in modulating 17a-E2-mediated effects on systemic metabolic homeostasis. Importantly, several
Figure 1 continued
The online version of this article includes the following figure supplement(s) for figure 1:
Figure supplement 1. 17a-E2 and 17b-E2 elicit similar ERa binding profile.
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 5 of 30
Research article Medicine
W
T
H
F
D
WT
H
FD+17
7
K
O
H
FD
K
O
HFD+17
17
W
T
HFD
W
T HFD+1
7
7
K
O
H
FD
K
O
H
F
D+
1
7
7
**
*
**
**
** **
*
*
*
**
**
**
**
**
**
**
*
*
**
**
**
**
*
Figure 2. ERa is required for 17a-E2 to reduce mass, adiposity, and calorie intake in male mice. (A) Percent change in mass (mean ± SEM, two-way
repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005 between WT HFD and WT HFD+17a), (B) Mass at baseline (week 0; solid)
and week 14 (striped) (mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005), (C) Percent change in fat
mass (mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005), (D) Fat mass at baseline (week 0; solid) and
week 14 (striped) (mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005), and (E) Average daily calorie
intake per week in WT and ERa KO mice provided 45% HFD (TestDiet 58V8)±17a-E2 (14.4ppm) (mean ± SEM, two-way repeated measures ANOVA with
Figure 2 continued on next page
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 6 of 30
Research article Medicine
studies have implicated hepatic ERa in the regulation of glucose homeostasis, insulin sensitivity, and
crosstalk with hypothalamic neurons that modulate metabolism and feeding behavior (
Meda et al.,
2020
; Torre et al., 2017; Barros and Gustafsson, 2011).
Figure 2 continued
Holm-Sidak post-hoc; *p<0.05, **p<0.005). Age-matched, male WT, chow-fed (TestDiet 58YP) mice were also evaluated as a normal-weight reference
group and the corresponding means are depicted as dashed gray lines. n = 10 (WT HFD), 10 (WT HFD+17a), 9 (KO HFD), 10 (KO HFD+17a), 12–15 (WT
Chow).
WT H
F
D
W
T
HFD+17
7
KO
H
FD
KO HFD+17
7
atbitrary units
WT HFD
WT HFD+17
7
KO H
F
D
K
O HFD+1
7
7
mg/dl
WT HFD
WT HFD+17
7
KO H
F
D
K
O HFD+1
7
7
ng/ml
W
T
H
F
D
W
T
H
F
D
+
17
17
K
O
H
F
D
KO HFD+17
7
3.5
4.0
4.5
5.0
5.5
6.0
HbA1c
%
**
*
*
*
*
**
* *
*
Figure 3. 17a-E2 reverses obesity-related metabolic dysfunction in male WT, but not ERa KO, mice. (A) Fasting glucose (mean ± SEM, two-way
repeated measures ANOVA), (B) Fasting insulin (mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005), and
(C) glycosylated hemoglobin (HbA1c) at baseline (week 0; solid) and week 14 (striped) in WT and ERa KO mice provided 45% HFD (TestDiet 58V8)±17a-
E2 (14.4 ppm) (mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; **p<0.005). (D) Glucose tolerance testing (GTT; 1 mg/kg)
(mean ± SEM, two-way repeated measures ANOVA with Holm-Sidak post-hoc; *p<0.05 between WT HFD and WT HFD+17a), and (E) GTT AUC during
week 10 of the study (mean ± SEM, two-way ANOVA with Holm-Sidak post-hoc; *p<0.05). Age-matched, male WT, chow-fed (TestDiet 58YP) mice were
also evaluated as a normal-weight reference group and the corresponding means are depicted as dashed gray lines. n = 9–10 (WT HFD), 8–10 (WT
HFD+17a), 9–10 (KO HFD), 8–10 (KO HFD+17a), 12–15 (WT Chow).
The online version of this article includes the following figure supplement(s) for figure 3:
Figure supplement 1. 17a-E2 reverses obesity-related metabolic dysfunction in male WT, but not ERa KO, mice.
Figure supplement 2. 17a-E2 fails to alter metabolic parameters in WT or ERa KO female mice.
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 7 of 30
Research article Medicine
17a-E2 improves liver disease pathology in an ERa-dependent manner
in male mice
We previously reported that 17a-E2 alters hepatic lipid deposition and DNA damage responses in
male mice through unknown mechanisms (
Stout et al., 2017b). In the current study, we sought to
determine if these findings are mediated through ERa. We found that 17a-E2 significantly reduced
liver mass and steatosis in male WT, but not ERa KO mice, as evidenced by reductions in oil-red-O
positivity, fatty acid content, and triglyceride accumulation (
Figure 4, Figure 4—figure supplement
1). These observations were accompanied by significant alterations in gene expression associated
with de novo lipogenesis (fatty acid synthase [Fasn]) and b-oxidation (peroxisome proliferator-acti-
vated receptor alpha [Ppara]; sterol regulatory element binding transcription factor 1 [Srebf1]) (
Fig-
ure 4—figure supplement 1
). These findings are similar to previous reports showing that 17b-E2
acts through ERa to modulate the expression and activity of genes that regulate hepatic lipid
metabolism (
Della Torre et al., 2016; Stubbins et al., 2012; Zhang et al., 2013). Interestingly,
despite seeing overall reductions in hepatic fatty acid content with 17a-E2 treatment in male WT
mice (
Figure 4C), we also observed elevations in specific fatty acids in these mice as compared to
WT HFD controls. Notably, arachidonic acid (AA, 20:4n6) and docosahexaenoic acid (DHA, 22:6n3),
both of which are precursors for eicosanoid, resolvin, and protectin production (
Szefel et al., 2015;
Kohli and Levy, 2009), were found to be increased by 17a-E2 treatment in male WT mice (Fig-
ure 4—figure supplement 2
). Our findings are aligned with a previous report by Garratt et al. show-
ing that 17a-E2 increases AA and DHA in liver (
Garratt et al., 2018). None of the 17 a-E2-mediated
changes in fatty acid profiles were observed in male ERa KO mice receiving 17a-E2. In response to
the elevations in AA and DHA with 17a-E2 treatment, we also assessed circulating eicosanoids. We
found that 17a-E2 treatment also mildly altered several circulating eicosanoid concentrations in male
WT mice (
Supplementary file 3). Many of these have been linked to changes in inflammatory signal-
ing (
Kiss et al., 2010; Gilroy et al., 2016), although the role they are playing in 17a-E2-mediated
effects of on metabolism and/or aging remain unclear.
Due to the association between obesity-related hepatic steatosis and the onset of fibrosis, we
assessed collagen deposition by trichrome staining and found that 17a-E2 reduced this in male WT,
but not ERa KO, mice (
Figure 5A). We also observed significant suppression of several transcripts
associated with liver fibrosis in male WT mice receiving 17a-E2, including collagen type 1 alpha 1
WT HFD
KO HFD
WT 17α HFD
KO 17α HFD
WT Chow
*
** **
WT HFD WT HFD+17α
KO HFD+17α KO HFD
WT Chow
Figure 4. 17a-E2 reverses obesity-related hepatic steatosis in an ERa-dependent manner in male mice. (A) Liver mass (mean ± SEM, two-way ANOVA
with Holm-Sidak post-hoc; *p<0.05), (B) Representative liver oil-red-O staining, (C) Liver fatty acids (mean ± SEM, two-way ANOVA with Holm-Sidak
post-hoc; **p<0.005), and (D) Liver triglycerides in WT and ERa KO mice provided 45% HFD (TestDiet 58V8)±17a-E2 (14.4ppm) for 14 weeks
(mean ± SEM, two-way ANOVA with Holm-Sidak post-hoc; **p<0.005). Age-matched, male WT, chow-fed (TestDiet 58YP) mice were also evaluated as a
normal-weight reference group and the corresponding means are depicted as dashed gray lines. n = 4–10 (WT HFD), 4–9 (WT HFD+17a), 4–9 (KO
HFD), 4–10 (KO HFD+17a), 4–15 (WT Chow).
The online version of this article includes the following figure supplement(s) for figure 4:
Figure supplement 1. 17a-E2 alters markers of lipid and glucose homeostasis predominantly through ERa in male mice.
Figure supplement 2. 17a-E2 alters the hepatic fatty acid profile in male WT, but not ERa KO, mice.
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 8 of 30
Research article Medicine
(Col1a1) (Hayashi et al., 2014; Lua et al., 2016), cyclin-dependent kinase inhibitor 1A (Cdkn1a)
(
Crary and Albrecht, 1998; Yang et al., 2020), matrix metallopeptidase 1 (Mmp1)
(
Lichtinghagen et al., 2003), matrix metallopeptidase 12 (Mmp12) (Madala et al., 2010), monocyte
chemoattractant protein 1 (Ccl2) (Glass et al., 2018; Baeck et al., 2012), C-X-C motif chemokine
ligand 1 (Cxcl1) (Yang et al., 2017), growth differentiation factor 15 (Gdf15) ( Koo et al., 2018), and
Col1a1
C
d
kn
1
a
Mmp1
Mmp1
2
Ccl2
Cxcl1
Gdf
1
5
T
n
fr
s
f
1a
D
B
E
p-AKT
AKT
p-FOXO1
FOXO1
GAPDH
C
WT
HFD
WT
HFD+17
KO
HFD
KO
HFD+17
WT
Chow
*
*
*
*
*
*
**
*
**
**
* * * *
** **
WT HFD WT HFD+17α
KO HFD+17α
KO HFD
WT Chow
Figure 5. 17a-E2 reverses obesity-related liver fibrosis and insulin resistance in an ERa-dependent manner in male mice. (A) Representative liver
Masson’s trichrome staining for collagen and (B) Liver transcriptional markers of fibrosis in WT and ERa KO mice provided 45% HFD (TestDiet 58V8)±
17a-E2 (14.4ppm) for 14 weeks (box plots depict total range of fold changes in gene expression with mean shown as a horizontal black line, Benjamini–
Hochberg multiple testing correction, two-way ANOVA with Holm-Sidak post-hoc; *p<0.05, **p<0.005). (C) Schematic of in vivo insulin stimulation
(2mU/g) in fasting mice, (D) Representative liver immunoblots, and (E) Quantification of phospho/total (p/t) AKT (pS473) and FOXO1 (pS256) in WT and
ERa KO mice provided 60% HFD (TestDiet 58Y1)±17a-E2 (14.4ppm) for 12 weeks (mean ± SEM, Benjamini–Hochberg multiple testing correction, two-
way ANOVA with Holm-Sidak post-hoc; **p<0.005). Age-matched, male WT, chow-fed (TestDiet 58YP) mice were also evaluated as a normal-weight
reference group and the corresponding means are depicted as dashed gray lines. n = 7–10 (WT HFD), 8–9 (WT HFD+17a), 7–10 (KO HFD), 10 (KO HFD
+17a), 7–11 (WT Chow).
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 9 of 30
Research article Medicine
TNF receptor superfamily member 1A (Tnfrsf1a) (Grattagliano et al., 2019; Figure 5B). Transcripts
shown to be predicative of hepatic insulin resistance (follistatin [Fst], inhibin subunit beta E [Inhbe],
insulin receptor substrate 2 [Irs2]) (
Tao et al., 2018; Parks et al., 2015) and gluconeogenic plasticity
(phosphoenolpyruvate carboxykinase 1 [Pck1], pyruvate kinase [Pkm]) (
Xiong et al., 2011) were also
beneficially modulated by 17a-E2 in male WT mice (
Figure 4—figure supplement 1 ). To confirm
that 17a-E2 improves hepatic insulin sensitivity, we also evaluated phosphorylation status of AKT
and FOXO1 in livers from male WT and ERa KO mice following the administration of an insulin bolus
(
Figure 5C). We found dramatic improvements in phosphorylated AKT (pS473) and FOXO1 (pS256)
in male WT mice treated with 17a-E2 (
Figure 5D–E), whereas these benefits were not observed in
male ERa KO mice. Our findings are aligned with previous reports demonstrating that hepatic ERa
plays a critical role in regulating insulin sensitivity in the liver of male mice (
Yan et al., 2019 ;
Guillaume et al., 2019; Qiu et al., 2017; Zhu et al., 2014). Collectively, these findings suggest that
the liver is highly responsive to 17a-E2 and that hepatic ERa is likely the signaling mechanism by
which 17a-E2 prevents and/or reverses steatosis, fibrosis, and insulin resistance.
Despite our findings demonstrating that 17a-E2 reduces calorie intake and improves liver disease
parameters in male mice in an ERa-dependent manner, it has historically been unclear if the benefits
attributed to 17a-E2 occur primarily due to long-term reductions in calorie intake. Moreover, it
remains unclear if 17a-E2 acts in a tissue-specific manner and if these observations would also occur
in other mammalian species. To address these questions, we subsequently evaluated the effects of
acute 17a-E2 administration during hyperinsulinemic-euglycemic clamps in male WT outbred rats.
These experiments allowed us to evaluate tissue-specific insulin-sensitivity following acute 17a-E2
exposure, thereby circumventing long-term effects of the compound including reductions in calorie
intake.
Acute 17a-E2 administration improves hepatic insulin sensitivity in male
rats
The hyperinsulinemic-euglycemic clamp is the gold-standard for assessing insulin action in vivo
(
Ayala et al., 2010). Animals are fasted overnight prior to receiving a constant infusion of insulin and
a variable infusion of [3-
3
H] glucose to maintain a euglycemia throughout the clamping period. Blood
samples are frequently obtained to assess glucose concentration and adjust glucose infusion rates
(GIRs) to maintain euglycemia, thereby allowing the calculation of insulin sensitivity to be done. Our
first set of experiments in male rats sought to determine if acute peripheral infusions of 17a-E2 mod-
ulates metabolic parameters during hyperinsulinemic-euglycemic clamps (
Figure 6A). We found that
acute peripheral administration of 17a-E2 significantly increased systemic insulin responsiveness as
compared to vehicle controls, which is indicated by increased GIRs (
Figure 6B). These studies also
determined that peripheral 17a-E2 administration robustly suppressed hepatic gluconeogenesis as
compared to vehicle controls (R
a
; Figure 6C–D), whereas glucose disposal rates (R
d
; Figure 6E)
were essentially identical between groups under clamped conditions. These data indicate that 17a-
E2 beneficially modulates metabolic parameters independent of reductions in calorie intake and adi-
posity. Furthermore, these findings strongly suggest that the liver is a primary site where 17a-E2
acts to improve metabolic homeostasis due to gluconeogenesis being tightly controlled by hormonal
actions on hepatocytes (
Zhang et al., 2018). However, it also well established that the hypothalamus
can directly modulate gluconeogenesis in the liver through hepatic innervation (
Timper and Bru
¨
ning,
2017
); therefore, we sought to determine if acute intracerebroventricular (ICV) delivery of 17a-E2
(
Figure 6F) could modulate metabolic parameters similarly to that observed during peripheral 17a-
E2 administration. Interestingly, we found that central administration of 17a-E2 essentially phe-
nocopied the effects of peripheral 17a-E2 infusion with regard to GIRs and suppression of hepatic
gluconeogenesis (
Figure 6G–I). These findings suggest that 17a-E2 likely acts through hypothalamic
neurons to regulate hepatic gluconeogenesis. Indeed, agouti-related peptide/neuropeptide Y
(AgRP/NPY) and pro-opiomelanocortin (Pomc) neurons are known to regulate hepatic glucose pro-
duction (
Ko
¨
nner et al., 2007; Ruud et al., 2017; Pocai et al., 2005a; Pocai et al., 2005b;
Dodd et al., 2018) and both neuronal populations express ERa (Smith et al., 2013; Skinner and
Herbison, 1997
; Xu et al., 2011; Acosta-Martinez et al., 2007; Stincic et al., 2018; Kelly and Røn-
nekleiv, 2015; Smith et al., 2014). Collectively, the hyperinsulinemic-euglycemic clamp studies
revealed that 17a-E2 definitively modulates metabolic homeostasis in an acute manner and suggests
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 10 of 30
Research article Medicine
that the liver and hypothalamus are two primary sites of action for the regulation of metabolic
parameters by 17a-E2.
Discussion
17a-E2 has recently been found to increase median lifespan in male mice through uncharacterized
mechanisms (
Strong et al., 2016; Harrison et al., 2014). We and others have shown that metabolic
improvements by 17a-E2 may underlie the lifespan extending effects. In these studies, we sought to
determine the role of ERa in 17a-E2-mediated transcriptional effects in vitro and metabolic effects
in vivo. Although previous studies have shown that 17a-E2 has limited binding affinity for ERa, it
remains unclear if 17a-E2 can induce transcriptional and physiological alterations in this manner.
Given the close association between metabolic improvements and ERa activity, we hypothesized
that 17a-E2 signals through ERa to elicit beneficial health outcomes. In these studies, we utilized
U2OS cells stably expressing ERa and ERa global knockout mice to assess the involvement of this
receptor in mediating 17a-E2 effects. Results from these studies demonstrate that ERa plays a piv-
otal role in 17a-E2-mediated effects on genomic activity and metabolism. Moreover, these data sug-
gest that ERa may be a target for the treatment of aging and chronic diseases in males.
Given the similarities between the metabolic benefits observed in vivo with 17a-E2 treatment and
the established body of literature linking ERa activity to systemic metabolic regulation (
Barros and
Gustafsson, 2011
), we utilized a well-established cell line model to globally assess the ERa cistrome
and transcriptome following 17a-E2 and 17b-E2 treatment. We found that, regardless of dose, 17a-
E2 and 17b-E2 elicited the same pattern of ERa genomic binding loci and these loci shared the
same DNA motif enrichments. Additionally, activation and suppression of gene expression were sim-
ilar with both 17a-E2 and 17b-E2 exposure and were independent of dosage. This provides strong
evidence that 17a-E2 is signaling through ERa to elicit beneficial outcomes, which is contrary to
Figure 6. Acute delivery of 17a-E2 improves hepatic insulin sensitivity. (A) Schematic of peripheral 17a-E2 infusions (or vehicle) during hyperinsulinemic-
euglycemic clamps, (B) glucose infusion rates (GIR) (mean ± SEM, unpaired Student’s t-test; **p<0.005), (C) rate of glucose appearance (R
a
; hepatic
glucose production) (mean ± SEM, unpaired Student’s t-test on Clamp; **p<0.005), (D) % suppression of hepatic glucose production (mean ± SEM,
unpaired Student’s t-test; **p<0.005), and (E) rate of glucose disappearance (R
d
; peripheral glucose disposal) in 6 month old, male, FBN-F1 hybrid rats
(mean ± SEM, unpaired Student’s t-test on Clamp). (F) Schematic of ICV (central) 17a-E2 infusions (or vehicle) during hyperinsulinemic-euglycemic
clamps, (G) GIR (mean ± SEM, unpaired Student’s t-test; *p<0.05), (H) R
a
(mean ± SEM, unpaired Student’s t-test on Clamp; **p<0.005), (I) %
suppression glucose production (mean ± SEM, unpaired Student’s t-test; *p<0.05), and (J) R
d
in 6-month-old, male, FBN-F1 hybrid rats (mean ± SEM,
unpaired Student’s t-test on Clamp). n = 5–9 (Veh.), 7–8 (17a).
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 11 of 30
Research article Medicine
what other reports have suggested (Harrison et al., 2014; Garratt et al., 2017; Garratt et al.,
2018
; Toran-Allerand et al., 2002; Toran-Allerand et al., 2005; Toran-Allerand, 2005). Toran-
Allerand et al. reported that 17a-E2 signals through a novel receptor in the brain, which they termed
ER-X (
Toran-Allerand et al., 2002; Toran-Allerand, 2005). Although our findings appear to dispute
this notion, several reports have shown that ERa exists and functions as multiple alternatively spliced
variants (
Flouriot et al., 1998; Wang et al., 2005; Taylor et al., 2010; Zhang et al., 2016;
Lin et al., 2013). Therefore, we speculate that ER-X may have been a truncated,
alternatively spliced, form of ERa, which nonetheless causes the same genomic and transcriptomic
effects. These findings led us to investigate how ERa may modulate 17a-E2-induced benefits in vivo
using ERa global KO mice.
In alignment with our previous reports (
Stout et al., 2017b; Steyn et al., 2018), 17a-E2 reduced
calorie intake, body mass, adiposity, and obesity-related metabolic perturbations in male WT mice.
Conversely, 17 a-E2 failed to elicit these beneficial effects in ERa KO mice, further supporting our
hypothesis that ERa is the receptor by which 17a-E2 signals to induce beneficial metabolic out-
comes. These observations are similar to how ERa is known to mediate the actions of endogenous
estrogens on metabolic parameters in females (
Barros and Gustafsson, 2011). In particular, 17b-E2
acts through ERa to regulate systemic insulin sensitivity, lipid distribution, thermogenesis, and hypo-
thalamic anorexigenic pathways (
Barros and Gustafsson, 2011; Stincic et al., 2018; Lo
´
pez and
Tena-Sempere, 2015
). The loss of endogenous estrogen action due to menopause in humans or
ovariectomy (OVX) in rodents eliminates these beneficial effects and elicits metabolic perturbations
(
Stefanska et al., 2015). Moreover, OVX following sexual maturation has also been shown to reduce
lifespan in female mice (
Benedusi et al., 2015), indicating that endogenous estrogens regulate life-
span in females; which we surmise is at least partially mediated through ERa.
In the current study, 17a-E2 failed to induce beneficial metabolic effects in female mice of either
genotype, which we postulate is due to endogenous 17b-E2 saturating ERa in female WT mice,
thereby limiting the potential benefits of 17a-E2 treatment. This interpretation is supported by our
recent report showing that OVX renders WT female mice responsive to the beneficial effects of 17a-
E2 on adiposity and bone mass (
Mann et al., 2020), both of which are regulated by ERa activity
(
Heine et al., 2000; Khosla and Monroe, 2018). In males, very few studies have evaluated the role
of ERa in metabolism, although a few recent reports have suggested that ERa plays tissue-specific
roles, particularly in the liver, by regulating glucoregulatory pathways (
Allard et al., 2019;
Yan et al., 2019; Guillaume et al., 2019; Qiu et al., 2017; Meda et al., 2020; Zhu et al., 2014).
These studies, coupled with our current findings, led us to speculate that 17a-E2 may be signaling
through ERa in the liver to reverse metabolic disease and potentially extend healthspan and/or life-
span in males.
The liver is a major regulator of systemic metabolic homeostasis. Obesity and advancing age
often promote a variety of liver conditions, including steatosis, fibrosis, and insulin resistance; all of
which are associated with hallmarks of aging (
Hunt et al., 2019), including cellular senescence
(
Ogrodnik et al., 2017), epigenetic alterations (Horvath et al., 2014), and dysregulated nutrient-
sensing (
Lo
´
pez-Otı´n et al., 2013). We have previously shown that 17a-E2 can reduce hepatic steato-
sis, hepatic insulin resistance, and hepatocyte DNA damage through unknown mechanisms
(
Stout et al., 2017b). In the present study, we sought to determine if these findings are mediated
through ERa. We found that 17a-E2 dramatically reduced liver mass and lipid content. As expected,
these observations were not seen in ERa KO mice, providing further support for the hypothesis that
17a-E2 regulates systemic metabolic parameters through ERa. Interestingly, our findings suggest
that 17a-E2 suppresses de novo lipogenesis and increases b-oxidation, predominantly in an ERa-
dependent manner. This is aligned with previous reports showing that 17b-E2 can modulate hepatic
lipid dynamics through both genomic and non-genomic actions (Pedram et al., 2016), leading to
altered expression of rate limiting enzymes that control de novo lipogenesis (
Zhang et al., 2013)
and b-oxidation (
Camporez et al., 2013). Reports have also shown that 17b-E2 can increase triglyc-
eride export, thereby decreasing hepatic lipid deposition (
Zhu et al., 2013). Although we did not
directly assess cholesterol profiles in these studies, we speculate that 17a-E2 may partially reduce
hepatic steatosis by increasing VLDL synthesis and/or triglyceride incorporation into VLDL. Addi-
tional studies will be needed to confirm how 17a-E2 alters hepatic lipoprotein dynamics.
Hepatic steatosis promotes liver fibrosis, which exacerbates hepatic insulin resistance (
Kim et al.,
2015
). Endogenous estrogens and hormone replacement therapies in post-menopausal women
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 12 of 30
Research article Medicine
have been shown to serve a protective role on liver function (Stubbins et al., 2012;
Camporez et al., 2013; Zhu et al., 2013; Rossi et al., 2004; Brussaard et al., 1997). Additionally,
male humans are at a higher risk of developing hepatic steatosis and fibrosis as compared to age-
matched females (
Guy and Peters, 2013; GBD 2017 Cirrhosis Collaborators and Collaborators,
2020
). In addition to reducing hepatic lipid deposition in male WT mice in the current studies, 17a-
E2 also dramatically suppressed transcriptional and histological markers of hepatic fibrosis in an
ERa-dependent manner. We also determined that 17a-E2 improved hepatic insulin sensitivity in
male WT, but not ERa KO, mice. Several transcriptional markers associated with hepatic insulin resis-
tance were suppressed in male WT mice receiving 17a-E2, whereas this suppression was almost
entirely absent in ERa KO mice. Subsequent studies employing insulin stimulation prior to euthanasia
also revealed that 17a-E2 robustly increased liver AKT and FOXO1 phosphorylation in male WT
mice, indicating a reversal of obesity-related hepatic insulin resistance and increased control of glu-
coneogenesis. These findings clearly demonstrate that 17a-E2 modulates hepatic insulin sensitivity
in an ERa-dependent manner. These observations are aligned with previous reports showing that
17b-E2 acts through ERa to improve glucoregulation (
Meda et al., 2020; Zhu et al., 2013). This pro-
vides further support that 17a-E2 is eliciting metabolic improvements through ERa that are specific
to the liver. Therefore, hepatic ERa may be a promising target for the development of therapeutics
to alleviate metabolic disease in males. Future studies utilizing cell-type-specific ablation of ERa in
the liver will be needed to unravel these possibilities.
Despite the robust effects of 17a-E2 on liver function, it remained unclear if 17a-E2 directly mod-
ulates hepatic insulin sensitivity or if these benefits were a secondary response to prolonged reduc-
tions in calorie intake, adiposity, and lipid redistribution. To test this, we performed
hyperinsulinemic-euglycemic clamps, in conjunction with acute infusions of 17a-E2 in male WT rats.
We found that peripheral infusions of 17a-E2 improved hepatic insulin sensitivity almost immedi-
ately, as evidenced by a greater suppression of hepatic glucose production in rats receiving 17a-E2
as compared to vehicle controls. Additionally, we did not observe improvements in glucose disposal,
thereby indicating that 17a-E2 does not acutely increase systemic insulin-stimulated glucose uptake.
This observation is aligned with recent literature demonstrating limited involvement of ERa in skele-
tal muscle insulin sensitivity (
In
˜
igo et al., 2020). These data indicate that 17a-E2 primarily alters sys-
temic metabolic homeostasis through the modulation of hepatic gluconeogenesis, which is known to
account for 76–87% of glucose production in the body (
Cherrington et al., 1994). Although these
studies are suggestive of direct actions in the liver, 17a-E2 also has the ability to cross the blood
brain barrier and elicit responses in the hypothalamus (
Steyn et al., 2018). Given that the hypothala-
mus can regulate hepatic glucose production (
Ko
¨
nner et al., 2007; Ruud et al., 2017; Pocai et al.,
2005a
; Pocai et al., 2005b; Dodd et al., 2018; Timper and Bru
¨
ning, 2017), we also evaluated sys-
temic insulin sensitivity following central administration of 17a-E2. These experiments essentially
phenocopied the results of the peripheral 17a-E2 infusions, suggesting that the suppression of
hepatic gluconeogenesis by 17a-E2 is at least partially mediated by the hypothalamus. A multitude
of studies have shown that the arcuate nucleus (ARC) of the hypothalamus plays a critical role in the
regulation of hepatic gluconeogenesis through autonomic regulation and vagus nerve activity
(
Ruud et al., 2017; Pocai et al., 2005a ; Zhang et al., 2018; Brandt et al., 2018). Multiple neuronal
populations within the ARC are known to be involved in the regulation of hepatic glucose produc-
tion, including Pomc (
Dodd et al., 2018) and AgRP/NPY (Ko
¨
nner et al., 2007; Ruud et al., 2017;
Pocai et al., 2005a; Pocai et al., 2005b). Similarly, we have previously shown that the effects of
17a-E2 on calorie intake and adiposity are dependent upon functional Pomc neurons, thereby pro-
viding evidence that 17a-E2 can act through the hypothalamus to mediate systemic metabolic
parameters (Steyn et al., 2018). However, in the absence of functional Pomc neurons, 17a-E2 was
still able to reduce fasting glucose and insulin, suggesting that 17a-E2 modulates peripheral metab-
olism through multiple mechanisms, which may include alternative neuronal populations. Given that
both Pomc (
Xu et al., 2011) and AgRP/NPY (Smith et al., 2013; Skinner and Herbison, 1997;
Acosta-Martinez et al., 2007; Stincic et al., 2018; Kelly and Rønnekleiv, 2015; Smith et al., 2014;
Sar et al., 1990 ) neurons express ERa, regulate systemic metabolic parameters, and modulate feed-
ing circuitry in a coordinated counter-regulatory fashion, it remains unclear whether 17a-E2 is alter-
ing hepatic and systemic metabolic parameters through Pomc and/or AgRP/NPY neurons.
Interestingly, a recent report from Debarba et al. demonstrated that 17a-E2 increased hypothalamic
ERa expression in the ARC (
Debarba, 2020), which further suggests that 17a-E2 signals through
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 13 of 30
Research article Medicine
ERa in the hypothalamus. Future studies utilizing hypothalamic cell-type-specific ERa KO models will
be needed to disentangle which populations of neurons are required for 17a-E2 to control food
intake, peripheral glucose homeostasis, and insulin sensitivity.
Collectively, our findings strongly suggest that 17a-E2 acts through ERa in the liver and/or hypo-
thalamus to modulate metabolic parameters. However, our findings are in contrast to other reports
suggesting that 17a-E2 elicits health benefits by modulating androgen metabolism (
Garratt et al.,
2017
; Garratt et al., 2018; Garratt and Stout, 2018). Garratt et al. reported that responsiveness to
17a-E2 was significantly attenuated in castrated male mice (
Garratt et al., 2018), which the authors
proposed may indicate 17a-E2 acts as a 5a-reductase inhibitor (
Schriefers et al., 1991) to prevent
the conversion of testosterone into dihydrotestosterone (DHT). 17a-E2 is known to be a mild 5a-
reductase inhibitor that is prescribed as a topical treatment for androgenetic alopecia (
Orfanos and
Vogels, 1980
). 5a-Reductase inhibition could conceivably elicit beneficial metabolic effects by either
reducing the concentration of DHT, which has been shown to decrease adiposity (
Move
´
rare-
Skrtic et al., 2006
; Bolduc et al., 2004), or by promoting greater aromatization of testosterone to
17b-E2 (
Veldhuis et al., 2009), which has been linked to improvements in metabolic parameters
(
Rubinow, 2017). If true, this would imply that the benefits of 17a-E2 are occurring in an indirect
manner. However, the dose of 17 a-E2 used in the vast majority of these studies, does not induce
dramatic feminization of the sex hormone profiles in male mice (
Stout et al., 2017b), which leads us
to speculate that 17a-E2 is acting in a direct manner through ERa rather than indirectly through
androgen modulation. Furthermore, studies in male rodents (
Livingstone et al., 2015;
Dowman et al., 2013) and humans Wei et al., 2019 demonstrate that 5a-reductase inhibition or
deficiency increases insulin resistance and hepatic steatosis and fibrosis, which are contradictory to
the effects of 17a-E2 treatment in all of our studies utilizing male mice (
Stout et al., 2017b;
Steyn et al., 2018; Miller, 2020; Sidhom et al., 2020). Despite these contrasting observations, the
studies by Garratt et al. do provide important insights into the interconnected and underappreciated
relationship between androgen- and estrogen-signaling pathways and their roles in metabolism and
aging. For instance, several recent reports have demonstrated interactions between the androgen
receptor (AR) and ERa (
D’Amato et al., 2016; Panet-Raymond et al., 2000; Peters et al., 2009),
which suggests that modulation of one may affect function of the other. Additional factors to con-
sider when comparing and contrasting our studies from those of Garratt et al. are differences in the
length of study, age, and obesity status of the mice, and counterregulatory and/or compensatory
effects of castration. Notably, it is plausible that 17a-E2 could be inducing metabolic benefits and
lifespan-extending effects through several distinct mechanisms, including direct actions through
ERa, suppression of DHT production, and/or aromatization of testosterone. Future studies will be
needed to discern the potentially interdependent nature of 17a-E2 actions on ERa and androgen
metabolism in metabolic improvement and lifespan extension.
There are a few notable caveats to our studies. First, we utilized constitutive global ERa KO mice,
which have been shown to display varying degrees of compensatory ERb activity due to the absence
of ERa during development (
Sa
´
nchez-Criado et al., 2012; Rosenfeld et al., 1998). However, if com-
pensatory ERb expression was playing a role in our study, we likely would not see a complete attenu-
ation of 17a-E2-mediated effects. As such, the results of our studies clearly indicate that ERa is the
primary receptor by which 17a-E2 signals. Another potential concern of the model is that ERa KO
mice are known to have elevated endogenous testosterone levels (
Gould et al., 2007), although the
studies by Garratt et al. would suggest that higher testosterone levels could potentially render the
male mice more responsive to 17a-E2, whereas we observed the opposite. Future studies utilizing
inducible Cre models to knockdown ERa post-sexual development may be considered if it is deter-
mined that Cre induction and subsequent ERa ablation is consistent throughout multiple organ sys-
tems, which has been shown to be inconsistent in other reports (
Murray et al., 2012). Despite these
minor concerns related to the model, the use of the constitutive global ERa KO was undoubtedly
the best option for these studies. However, it must also be noted that female mice present a greater
phenotypic response than males to the ablation of ERa, thereby exacerbating obesity and metabolic
dysfunction which makes comparisons to female WT mice as well as their male littermates problem-
atic (Manrique et al., 2012 ; Vidal et al., 1999). For this reason, we chose not to provide HFD to
female mice in these studies. Regardless, 17a-E2 still failed to elicit beneficial responses in female
mice of either genotype (WT or ERa KO), which is aligned with previous reports demonstrating a
lack of effect of 17a-E2 in intact females (Garratt et al., 2017; Garratt et al., 2018). We also
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 14 of 30
Research article Medicine
recently reported that OVX renders female mice responsive to several of the benefits conferred by
17a-E2 treatment in male mice (
Mann et al., 2020). These observations, coupled with the genomic
data presented herein, support the hypothesis that endogenous 17b-E2 actions on ERa diminishes
potential benefits of 17a-E2 in intact female mice. However, future studies will be needed to defini-
tively determine if female mice subjected to diet-induced obesity will display responsiveness to 17a-
E2 once severe metabolic dysfunction has emerged. Lastly, the current studies were relatively short
in duration and it remains unclear if metabolic improvements with 17a-E2 treatment are required for
the lifespan extension effects of the compound. Although several other studies have evaluated the
long-term effects of 17a-E2 (
Strong et al., 2016; Harrison et al., 2014; Garratt et al., 2017;
Garratt et al., 2018), a shorter treatment duration was effective for testing our hypothesis in these
studies. Similarly, given the close relationship between metabolic homeostasis, sex hormones, and
longevity (
Lo
´
pez-Otı´n et al., 2013; Barros and Gustafsson, 2011), we surmise that future studies
evaluating the effects of ERa on male lifespan in the presence or absence of 17a-E2 will be needed.
Although our current report does not provide direct evidence that ERa modulates the lifespan
extending effects of 17a-E2, it does provide insight into the involvement of hepatic and/or hypotha-
lamic ERa on 17a-E2-mediated metabolic effects in male mice.
In summary, the data presented herein are the first to show that 17a-E2 and 17b-E2 induce nearly
identical ERa chromatin association patterns and transcriptional activity. Moreover, we demonstrate
that the metabolic benefits of 17a-E2 in male mice are ERa-dependent. We also provide evidence
that strongly suggests 17a-E2 acts through the liver and hypothalamus to regulate metabolic
homeostasis in male mice. These effects were mirrored by studies in male WT rats receiving 17a-E2,
indicating that 17a-E2 can modulate metabolism almost instantaneously and that these effects are
not limited to a single mammalian species. Future studies will be needed to confirm that 17a-E2 acts
predominantly through ERa in a cell-type-specific manner in the liver and hypothalamus to modulate
systemic metabolic homeostasis. It is also imperative that we determine if ERa exclusively modulates
the lifespan-extending effects of 17a-E2 in male mice. Another potential avenue of investigation that
remains unresolved is whether 17a-E2 acts through ERa in a genomic or non-genomic manner to
modulate health parameters. Potential interactions between androgen and estrogen signaling must
also be considered when evaluating the effects of 17a-E2 on metabolism and lifespan. These studies
will provide additional insight into mechanisms of metabolic improvement and lifespan extension by
17a-E2. Our studies provide critical insight into the molecular mechanisms by which 17a-E2 elicits
metabolic benefits in males, which were previously unknown and may underlie its lifespan-extending
effects.
Materials and methods
Key resources table
Reagent type (species)
or resource Designation Source or reference Identifiers Additional information
Genetic reagent
(M. musculus)
B6N(Cg)-Esr1tm4.2Ksk/J The Jackson
Laboratory
Stock No:026176;
RRID:
IMSR_JAX:026176
ERa (Esr1) KO mice
Cell line
(Homo sapien)
U2OS Cells ATCC HTB-96;
RRID:
CVCL_0042
PMID:15802376
PMID:14505348
Antibody anti-FLAG M2
(Mouse monoclonal)
Sigma-Aldrich F1804 IP: 1 uL per pull-down
(1 mg/mL)
Commercial
assay or kit
Protein G Dynabeads Applied Biosystems/
Thermofisher Scientific
10003D IP: 30 uL per IP
Chemical compound,
drug
17a-E2 Steraloids, Inc E0870-000
Chemical compound,
drug
Novolin R 100 U/ml Novolin 2mU/g
Other (diet) Chow; TestDiet 58YP TestDiet TestDiet 58YP
Other (diet) HFD; TestDiet 58V8 TestDiet TestDiet 58V8 HFD 45% by kcal
Other (diet) HFD; TestDiet 58Y1 TestDiet TestDiet 58Y1 HFD 60% by kcal
Continued on next page
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 15 of 30
Research article Medicine
Continued
Reagent type (species)
or resource Designation Source or reference Identifiers Additional information
Commercial assay
or kit
Mouse Ultrasensitive
Insulin ELISA
ALPCO Cat# 80-INSMSU-E01;
RRID:
AB_2792981
Commercial assay
or kit
Free Glycerol Agent Sigma-Aldrich Sigma F6428
Commercial assay
or kit
Triglyceride Reagent Sigma-Aldrich Sigma F6428
Commercial assay
or kit
Glycerol Standard Sigma-Aldrich Sigma G1394
Antibody anti-pS473 AKT
(Rabbit polyclonal)
Abcam Cat# ab81283;
RRID:
AB_2224551
WB: (1:3000)
Antibody Anti-pan-AKT
(Rabbit polyclonal)
Abcam Cat# ab179463;
RRID:
AB_2810977
WB (1:10000)
Antibody Anti-pS256 FOX01
(Rabbit polyclonal)
Abcam Cat# ab131339;
RRID:
AB_11159015
WB (1:1000)
Antibody Anti-FOX01a
(Rabbit polyclonal)
Abcam Cat# ab52857;
RRID:
AB_869817
WB (1:1000)
Antibody Anti-GAPDH
(Rabbit polyclonal)
Abcam Cat# ab9485;
RRID:
AB_307275
WB (1:2500)
Antibody Anti-Rabbit IgG,
IRDye 800 CW
LI-COR Cat# 926–32211;
RRID:
AB_621843
WB (1:15000)
Commercial assay
or kit
TaqMan Gene
Expression Master Mix
Applied Biosystems/
Thermofisher Scientific
4369542
Sequenced-based
reagent
qPCR primer Mmp1 Integrated DNA
Technologies
Mm.PT.58.42286812
Ref Seq: NM_008607(1)
Exon 5–6
Sequenced-based
reagent
qPCR primer Mmp12 Integrated DNA
Technologies
Mm.PT.58.31615472
Ref Seq: NM_008605(1)
Exon 8–9
Sequenced-based
reagent
qPCR primer Ccl2 Integrated DNA
Technologies
Mm.PT.58.42151692
Ref Seq: NM_011333(1)
Exon 1–3
Sequenced-based
reagent
qPCR primer Srebf1 Integrated DNA
Technologies
Mm.PT.58.8508227
Ref Seq: NM_011480(1)
Exon 1–2
Sequenced-based
reagent
qPCR primer Pck1 Integrated DNA
Technologies
Mm.PT.58.11992693
Ref Seq: NM_011044(1)
Exon 3–4
Sequenced-based
reagent
qPCR primer Cdkn1a Integrated DNA
Technologies
Mm.PT.58.17125846
Ref Seq: NM_007669(1)
Exon 2–3
Sequenced-based
reagent
qPCR primer Ppara Integrated DNA
Technologies
Mm.PT.58.9374886
Ref Seq: NM_001113418(2)
Exon 8–9
Sequenced-based
reagent
qPCR primer Cxcl1 Integrated DNA
Technologies
Mm.PT.58.42076891
Ref Seq: NM_008176(1)
Exon 2–4
Sequenced-based
reagent
qPCR primer Col1a1 Integrated DNA
Technologies
Mm.PT.58.7562513
Ref Seq: M_007742(1)
Exon 1–2
Sequenced-based
reagent
qPCR primer Tnfrsf1a Integrated DNA
Technologies
Mm.PT.58.28810479
Ref Seq: NM_011609(1)
Exon 5–7
Software, algorithm SigmaPlot 12.5 Systat Software RRID:
SCR_003210 statistical analyses
Software, algorithm ImageJ ImageJ RRID:
SCR_003070 histological quantification
Software, algorithm Image Studio LI-COR RRID:
SCR_015795 western blot quantification
Software, algorithm RStudio GenomicAlignments
DiffBind
DESeq2
GenomicRanges
RRID:
SCR_000432 Peak Calling
Differential expression
Differential binding
Software, algorithm Bowtie2
MACS2
Bedtools
Samtools
Picard-tools
Trimmomatic
Bowtie2
MACS2
Bedtools
Samtools
Picard-tools
Trimmomatic
Alignment, Peak Calling,
trimming, duplicate
identification
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 16 of 30
Research article Medicine
U2OS cells
U2OS osteosarcoma cells stably expressing flag-tagged ERa (U2OS-ERa) under the control of doxy-
cycline (dox)-inducible promoter (
Monroe et al., 2003) were utilized for the studies described here.
U2OS cells were originally purchased from ATTC and were authenticated using IDEXX BioAnalytics
(Westbrook, ME). Cells were also regularly checked for mycoplasma contamination using a PCR-
based mycoplasma detection kit from SouthernBiotech (Birmingham, AL) and were confirmed to be
negative. U2OS-ERa cells were cultured in phenol-free aMEM medium supplemented with 10%
HyCloneTM charcoal/dextran stripped FBS (GE Healthcare Life Sciences, Pittsburgh, PA), 1% antibi-
otic/antimycotic, 5 mg/L blasticidin S, and 500 mg/L zeocin in a humidified 37˚C incubator with 5%
CO
2
. Cells were plated in 12-well plates in the presence of doxycycline to induce ERa expression.
The following day, cells were treated for 24 hr with 17b-E2 (10 nM) or 17a-E2 (10 nM and 100 nM)
(Steraloids, Newport, RI) in charcoal-stripped FBS-containing media.
ChIP-sequencing
To evaluate patterns of ERa binding agonized by 17a-E2 vs 17b-E2, we performed ChIP-Sequencing.
U2OS-ERa cells were harvested 24 hr post-treatment and chromatin immunoprecipitation was per-
formed as previously described (
Reese et al., 2018; Nelson et al., 2006). Briefly, ERa was immuno-
precipitated overnight at 4˚C using 10 mg of Flag antibody (clone M2, Sigma-Aldrich, St. Louis, MO).
Complexes bound to the antibody were captured with protein G Dynabeads (Thermo Fisher Scien-
tific, Waltham, MA), extensively washed, and reverse cross-linked at 65˚C overnight. DNA isolation
was performed by phenol/chloroform extraction and was used for ChIP-sequencing library prepara-
tion. Libraries were sequenced using paired-end 100 bp reads on the Illumina HiSeq 4000
(GSE151039). Reads were aligned to the human genome (hg19,
https://genome.ucsc.edu/cgi-bin/
hgGateway
) using bowtie2 (Langmead and Salzberg, 2012) and duplicated reads were flagged
with Picard-tools (
http://broadinstitute.github.io/picard/). ERa-binding peaks were called using
MACS2 (
Zhang et al., 2008) with recommended settings. Peak genomic location, breadth of cover-
age, and peak summit location were determined using MACS2. NarrowPeak files containing peak
information were used to determine differential ERa binding. First, peaks were centralized around
the summit and 250 bp flanking regions were added to the summit location to generate equal 500
bp regions across all experimental groups. Peak files were then used to extract read counts from the
aligned de-duplicated BAM file using samtools (
Li et al., 2009), read counts were then normalized
to total library sequencing depth. To determine differential binding, the R package diffbind was uti-
lized (
Ross-Innes et al., 2012). Normalized read counts were log2 transformed and normalized
across all experimental groups. Differential binding between treatment groups was determined
using negative binomial regression models utilized in the R package DESeq2, statistical significance
for pairwise comparisons between experimental groups was determined using Wald test. To account
for multiple comparisons, we used Benjamini-Hochberg multiple testing correction (False-discovery
rate, FDR). Motif analysis was performed using HOMER with standard settings to identified motifs.
Peak regions called for each treatment group were analyzed to identify enriched motifs relative to
the entire genome. For pairwise differential motif enrichment or depletion between experimental
groups, we utilized a hypergeometric test with the number of sequences with a motif from each
group and total number of peaks as total sample size. Motifs that appeared in less than five sequen-
ces between both pairwise test groups were removed. Benjamini-Hochberg multiple testing correc-
tion was utilized to control for multiple testing (FDR < 0.05).
RNA-sequencing
U2OS-ERa cells were harvested 24 hr post-treatment and RNA was extracted using Trizol and
DNase cleanup. RNA libraries were prepared with Illumina’s TrueSeq RNA-seq library prep accord-
ing to manufacturer protocol. Libraries were sequenced with 150 bp paired-end reads on the Illu-
mina 4000 platform (Illumina, San Diego, CA) (GSE151039). Sequence quality control was performed
with fastQC, Paired reads were trimmed using trimmomatic, and were aligned to the hg19 genome
using STAR (
Dobin et al., 2013). Differential expression was determined using previously described
methods (Hadad et al., 2019). In brief, gene counts were determined with the R package Genomi-
cAlignments ‘summarizeOverlap’ function. Gene counts were then transformed using regularized log
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 17 of 30
Research article Medicine
transformation and normalized relative to library size using the DESeq2 (Love et al., 2014) R pack-
age. Differential expression was determined using negative binomial generalized linear model using
counts ~ treatment model. We performed pairwise differential expression between all experimental
groups using Wald test. All comparisons were corrected for multiple testing using Benjamini-Hoch-
berg multiple testing correction method. Differential expression significance threshold was set to
FDR corrected p<0.05.
Animal study 1
To determine if ERa is the primary receptor by which 17a-E2 signals to elicit metabolic benefits in
vivo, we utilized male global ERa KO and WT littermate mice. Mice were acquired from Dr. Kenneth
Korach (National Institute of Environmental Health Sciences [NIEHS]) and were also bred at OUHSC
by pairing ERa heterozygous KO mice (JAX; strain #026176). Mice were fed a 45% high-fat diet
(HFD) (TestDiet 58V8, 35.5% CHO, 18.3% PRO, 45.7% FAT) from TestDiet (Richmond, IN) for 4
months prior to study initiation to induce obesity and metabolic perturbations. Additionally, age-
matched, male WT, chow-fed mice were maintained on TestDiet 58YP (66.6% CHO, 20.4% PRO,
13.0% FAT) throughout the entire study as a healthy-weight reference group. Mice were individually
housed with ISO cotton pad bedding, cardboard enrichment tubes, and nestlets at 22 ± 0.5˚C on a
12:12 hr light-dark cycle. Unless otherwise noted, all mice had ad libitum access to food and water
throughout the experimental timeframe. At the conclusion of the fattening period, all mice (age: 6–8
months) receiving HFD were randomized within genotype by age, body mass, fat mass, calorie
intake, fasting glucose, fasting insulin, and glycosylated hemoglobin (HbA1C) into HFD or HFD+17a
(TestDiet 58V8 + 17a-E2, 14.4ppm; Steraloids, Newport, RI) treatment groups for a 14-week inter-
vention. Body mass and calorie intake were assessed daily for the first 4 weeks, followed by body
mass and body composition (EchoMRI, Houston, TX) on a weekly basis. At 10 weeks post-treatment,
mice were fasted for 5–6 hr and fasting glucose, fasting insulin, HbA1C, and glucose tolerance were
assessed. At the conclusion of the study (14 weeks post treatment), mice were euthanized with iso-
flurane in the fasted state (5–6 hr). Blood was collected into EDTA-lined tubes by cardiac puncture,
and plasma was collected and frozen. Tissues were excised, weighed, flash frozen, and stored at
80˚C unless otherwise noted. Small sections of liver were fixed in 4% paraformaldehyde in prepara-
tion for paraffin- or cryo-embedding for future analyses. All animal procedures were reviewed and
approved by the Institutional Animal Care and Use Committee at OUHSC.
Animal study 2
Although previous studies [
Strong et al., 2016; Harrison et al., 2014; Garratt et al., 2017;
Garratt et al., 2018] have demonstrated minimal effects of 17a-E2 in female mice, we thought it
prudent to determine if the ablation of ERa would alter female responsiveness to 17a-E2. Female
WT and ERa KO mice were acquired from Dr. Kenneth Korach (National Institute of Environmental
Health Sciences [NIEHS]). Female mice were maintained on Chow TestDiet 58YP (66.6% CHO, 20.4%
PRO, 13.0% FAT) and were not subject to HFD feeding due to ERa KO female mice naturally dis-
playing an obesity phenotype. Mice were individually housed with ISO cotton pad bedding, card-
board enrichment tubes, and nestlets at 22 ± 0.5˚C on a 12:12 hr light-dark cycle. Unless otherwise
noted, all mice had ad libitum access to food and water throughout the experimental timeframe. At
age 9–11 months, female mice were randomized within genotype by age, body mass, fat mass, calo-
rie intake, fasting glucose, fasting insulin, and glycosylated hemoglobin (HbA1C) into Chow or Chow
+17a-E2 (TestDiet 58YP + 17a-E2,14.4ppm; Steraloids, Newport, RI) treatment groups. The study
was terminated following a 4-week intervention due to a lack of responsiveness to 17a-E2. At the
conclusion of the study, mice were euthanized with isoflurane in the fasted state (5–6 hr). Blood was
collected into EDTA-lined tubes by cardiac puncture, and plasma was collected and frozen. Tissues
were excised, weighed, flash frozen, and stored at 80˚C. All animal procedures were reviewed and
approved by the Institutional Animal Care and Use Committee at OUHSC.
Animal study 3
To assess insulin sensitivity within the liver, an additional cohort of ERa KO and WT littermate mice
were bred from mice acquired from Jackson Laboratory (JAX; strain #026176), which were gener-
ated from identical founder strains in the laboratory of Dr. Korach at NIEHS. Male ERa KO and WT
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 18 of 30
Research article Medicine
mice were fed a 60% high-fat diet (HFD; TestDiet 58Y1, 20.3% CHO, 18.1% PRO, 61.6% FAT) for 4
months prior to study initiation to induce obesity and metabolic perturbations. Additionally, as was
done in Animal Study 1, age-matched, male WT, chow-fed mice were maintained on TestDiet 58YP
(66.6% CHO, 20.4% PRO, 13.0% FAT) throughout the entire study as a healthy-weight reference
group. Mice were group housed with corncob bedding, cardboard enrichment tubes, and nestlets at
22 ± 0.5˚C on a 12:12 hr light-dark cycle. Unless otherwise noted, all mice had ad libitum access to
food and water throughout the experimental timeframe. At the conclusion of the fatting period, all
mice (age: 6 months) receiving HFD were randomized within genotype by body mass, fat mass, calo-
rie intake, fasting glucose, and fasting insulin into HFD or HFD+17a (TestDiet 58Y1 + 17a-E2,
14.4ppm; Steraloids, Newport, RI) treatment groups for a 12-week intervention. Prior to being
euthanized, mice were fasted (5–6 hr) and IP injected with insulin (Novolin R 100 U/ml; 2mU/g) to
assess insulin activity and sensitivity in tissue as previously described (
Lu et al., 2012). Each mouse
was euthanized with isoflurane 15 min following their insulin injection. Blood was collected into
EDTA-lined tubes by cardiac puncture, and plasma was collected and frozen. Tissues were excised,
weighed, flash frozen, and stored at 80˚C unless otherwise noted. All animal procedures were
reviewed and approved by the Institutional Animal Care and Use Committee at OUHSC.
Animal study 4
Hyperinsulinemic-euglycemic clamp experiments, the gold-standard for assessing insulin sensitivity,
were performed in male rats to determine if 17a-E2 can acutely modulate insulin sensitivity and glu-
cose homeostasis. FBN-F1 hybrid male rats were acclimated to the animal facilities within the Ein-
stein Nathan Shock Center for 2 weeks prior to undergoing surgeries in preparation for
hyperinsulinemic-euglycemic clamp studies. Rats were fed Purina 5001 (58.0% CHO, 28.5% PRO,
13.5% FAT) and were individually housed with corncob bedding at 22 ± 0.5˚C on a 14:10 hr light-
dark cycle with ad libitum access to food and water. All surgeries were conducted under 2% isoflur-
ane. For clamp studies incorporating central infusions, rats underwent two surgical procedures. First,
stereotactic placement of a steel-guide cannula (Plastics One, Roanoke, VA) reaching the 3rd ventri-
cle was performed. The implant was secured in place with dental cement and animals were treated
with analgesic as needed. Approximately 14 days later, animals underwent a second surgical proce-
dure to place indwelling catheters into the right internal jugular vein and the left carotid artery,
which was also performed for animals undergoing only peripheral clamp studies. Hyperinsulinemic-
euglycemic clamp studies incorporating peripheral 17a-E2 infusions were performed as previously
described (
Einstein et al., 2010). For studies employing peripheral infusions of 17a-E2, 17a-E2 was
diluted in sterile saline to a final concentration of 30 ng/ml. Beginning at t = 0 min animals received a
primed-continuous infusion of saline or 30 ng/ml 17a-E2 provided as a 3 mg bolus at a rate of 20 ul/
min over 5 min, followed by a continuous infusion at a rate of 0.06 ml/hr over 235 min (9.4 ng/hr) for
a maintenance dose of 7 mg (total dose 10 mg). Hyperinsulinemic-euglycemic clamp studies with
intracerebroventricular (ICV) infusions were performed as previously described (
Huffman et al.,
2016a
). 17a-E2 powder (Steraloids, Newport, RI) was dissolved in DMSO at a concentration of 10
mg/ml and stored at 80˚C. For ICV infusions, 17a-E2 was diluted in artificial cerebral spinal fluid
(ACSF) to a final concentration of 2 ng/ml. Beginning at t = 0 min, animals received a primed-continu-
ous ICV infusion of ACSF (Veh.) or 17a-E2 (17a) provided as a 15 ng bolus at a rate of 1 ml/min over
7.5 min, followed by a continuous infusion of 56.5 ng at a rate of 0.08 ml/hr over 6 hr (9.4 ng/hr) and
a total dose of 71.5 ng. All animal procedures were reviewed and approved by the Institutional Ani-
mal Care and Use Committee at the Einstein College of Medicine.
In vivo metabolic analyses in mice
To evaluate the effects of 17a-E2 on metabolic parameters in vivo, we performed several assess-
ments of glucose homeostasis. Unless otherwise noted, all experiments requiring fasting conditions
were performed in the afternoon, 5–6 hr following the removal of food at the beginning of the light-
cycle as outlined elsewhere (
Ayala et al., 2010). To ensure fasting conditions, mice were transferred
to clean cages containing ISO cotton padding and clean cardboard enrichment tubes. Non-terminal
blood was collected via tail snip. Fasting glucose was evaluated using a Bayer Breeze 2 Blood Glu-
cose Monitoring System (Bayer Global, Leverkusen, Germany). Fasting insulin was evaluated using a
Mouse Ultrasensitive Insulin ELISA from Alpco (Salem, NH). HbA1c was assessed by A1C-Now
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 19 of 30
Research article Medicine
Monitoring kits (Bayer, Whippany, NJ). Glucose tolerance tests were performed following a 5 hr fast
using an intraperitoneal filtered dextrose injection of 1 g/kg body mass (
Huffman et al., 2016b).
Blood glucose was measured immediately pre-injection (time 0) and at 15, 30, 60, 90, and 120 min
post-injection.
Liver histology
To evaluate the effects of 17a-E2 treatment on lipid accumulation and fibrosis, we evaluated fixed
liver tissue. Tissues were fixed in 4% PFA for 24 hr, cryo-embedding samples were transferred to
30% sucrose for 72 hr and embedded in OCT, paraffin-embedding samples were transferred to 1X
PBS for 48 hr, then to 70% ethanol until embedding. Liver oil-red-O and Masson’s trichrome staining
were performed by the Oklahoma Medical Research Foundation Imaging Core Facility using previ-
ously reported methodology (
Leonard et al., 2018; Mehlem et al., 2013). Oil-red-O (ORO, Sigma-
Aldrich, St. Louis, MO) and H and E counterstaining were performed on cryo-embedded tissues, and
were imaged within 6 hr of staining. Red lipid stain was blindly quantified from 10 images per animal
using ImageJ software and presented as a lipid to total tissue ratio. Masson’s trichrome staining was
performed on paraffin embedded liver tissue and was used for qualitative purposes. In brief, slides
were stained with Weigert’s Iron Hematoxylin (Sigma-Aldrich, St. Louis, MO), washed, stained with
Biebrich Scarlet-Acid Fusion (Sigma-Aldrich, St. Louis, MO), washed, stained with Phosphomolybdic
Acid-Phosphotunsctic Acid, and then stained with Aniline Blue (Sigma-Aldrich, St. Louis, MO).
Liver triglycerides
We evaluated the effects of 17a-E2 treatment on triglyceride accumulation in the liver. Liver samples
(~100 mg) were homogenized on ice for 60 s in 10X (v/w) Cell Signaling Lysis Buffer (Cell Signaling,
Danvers, MA) with protease and phosphatase inhibitors (Boston BioProducts, Boston, MA). Total
lipid was extracted from the homogenate using the Folch method with a 2:1 chloroform-methanol
mixture (
Folch et al., 1957). Lipid was dried down using a nitrogen drier at room temperature, and
resuspended in 100 ml of 3:1:1 tert-butyl alcohol-methanol-Triton X-100 solution. Final triglyceride
concentrations were determined using a spectrophotometric assay with a 4:1 Free Glycerol Agent/
Triglyceride Agent solution (Sigma Triglyceride and Free-Glycerol reagents, St. Louis, MO) as previ-
ously described (
Stout et al., 2011).
Liver fatty acids
We evaluated the effects of 17a-E2 on hepatic fatty acid content. Liver samples (~50 mg) were
homogenized and on ice for 60 s in 10X (v/w) Cell Signaling Lysis Buffer (Cell Signaling, Danvers,
MA) with protease and phosphatase inhibitors (Boston BioProducts, Boston, MA). Total lipid was
extracted using a modified Bligh and Dyer method (
Bligh and Dyer, 1959 ) (Sigma-Aldrich, St. Louis,
MO). Of 15:0 and 17:0 internal standards, 50 nmol were added and acid hydrolysis/methanolysis
was done to generate fatty acid methyl esters (FAMEs) (
Agbaga et al., 2018). FAMEs were identi-
fied as previously described by GC-MS (
Agbaga et al., 2018). A 6890N gas chromatograph with
flame ionization detector (GC-FID) (Agilent Technologies) was used to quantify FAMEs (
Yu et al.,
2012
). Standards 15:0 and 17:0 were used to compare and determine sample concentrations. Data
is represented as the relative mole percent of each fatty acid.
Plasma eicosanoids
We evaluated the effects of 17a-E2 treatment on circulating eicosanoids (
Supplementary file 3).
Plasma eicosanoid analyses were performed by the UCSD Lipidomics Core as described previously
(
Quehenberger et al., 2010). Eicosanoids were isolated from plasma, extracted, separated using
liquid chromatography, and analyzed with mass spectrometry (MDS SCIEX 4000 Q Trap; Applied
Biosystems, Foster City, CA) (
Quehenberger et al., 2010).
Real-time PCR
To evaluate alterations in gene expression following 17a-E2 treatment, we performed qPCR for
genes related to fibrosis, lipid metabolism, insulin resistance, and glucose metabolism in the liver.
Total RNA was extracted using Trizol (Life Technologies, Carlsbad, CA) and was reverse transcribed
to cDNA with the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City,
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 20 of 30
Research article Medicine
CA). Real-time PCR was performed in a QuantStudio 12K Flex Real Time PCR System (Thermofisher
Scientific, Waltham, MA) using TaqMan Gene Expression Master Mix (Applied Biosystems/Thermo-
fisher Scientific, Waltham, MA) and predesigned gene expression assays with FAM probes from Inte-
grated DNA Technologies (Skokie, Illinois). Target gene expression was expressed as 2
DDCT
by the
comparative CT method (
Livak and Schmittgen, 2001) and normalized to the expression of TATA-
Box Binding Protein (TBP) in liver.
Western blotting
To determine if 17a-E2 altered hepatic insulin sensitivity, we evaluated phosphorylation status of
AKT and FOXO1 following an insulin bolus. Liver was homogenized in RIPA Buffer (Cell Signaling,
Danvers, MA) with protease and phosphatase inhibitors (Boston Bioproducts, Boston, MA). Total
protein was quantified using BCA Protein Assay Reagent Kit (Pierce, Rockford, IL). Proteins were
separated on an Any kD Criterion TGX Stain-Free Protein Gel (Biorad, Hercules, CA) at 75V for 150
min in Running Buffer (Cell Signaling, Danvers, MA) and transferred to a 0.2 mm pore size nitrocellu-
lose membrane, (Biorad, Hercules, CA) at 75V for 90 min on ice. Primary antibodies used were
pS256 FOX01 (Abcam ab131339, 1:1000), FOX01a (Abcam ab52857, 1:1000), pS473 AKT (Abcam
ab81283, 1:3000), pan-AKT (Abcam ab179463, 1:10000), GAPDH (Abcam ab9485, 1:2500). Primary
antibody detection was performed with IRDye 800CW Infrared Rabbit (LI-COR Biotechnology, Lin-
coln, NE) at 1:15,000 concentration. GAPDH was diluted in 5% dry milk (Cell Signaling, Danvers,
MA), all other antibodies were diluted in 5% BSA (Cell Signaling, Danvers, MA). Blot imaging was
done on Odyssey Fc Imaging System (LI-COR Biotechnology, Lincoln, NE) with a two-minute expo-
sure time at 800l , and protein detection and quantification were performed using Image Studio
Software (LI-COR Biotechnology, Lincoln, NE).
Statistical analyses
Results are presented as mean ± SEM unless otherwise stated with p values less than 0.05 consid-
ered to be significant unless otherwise specified. Analyses of differences between groups were per-
formed by two-way ANOVA, two-way repeated measures ANOVA, or Student’s t-test where
appropriate using SigmaPlot 12.5 Software. A Benjamini-Hochberg multiple testing correction was
applied to the F test result to correct for the number of transcripts, proteins, and fatty acids
analyzed.
Acknowledgements
We thank Dr. Kenneth Korach at the National Institute of Environmental Health Sciences for provid-
ing ERa KO and WT littermate mice. We also thank Dr. Lora Bailey-Downs, Richard Brush, and
Michael Sullivan for technical support. This work was supported by the National Institutes of Health
(R00 AG51661 and R01 AG069742 to MBS, T32 AG052363 to SNM, R01 EY030513 to M-PA, and
R01 AG059430 to WMF), Veterans Affairs (I01B 003906 to WMF) and pilot research funding from
the Harold Hamm Diabetes Center (MBS and SNM), Einstein Nathan Shock Center (P30 AG038072)
of Excellence in the Basic Biology of Aging (MBS), and OUHSC Lipidomics Core (P30 EY012190).
Additional information
Funding
Funder Grant reference number Author
National Institutes of Health R00 AG51661 Michael B Stout
Harold Hamm Diabetes Center Pilot Research Funding Shivani N Mann
Michael B Stout
National Institutes of Health R01 AG069742 Michael B Stout
National Institutes of Health R01 AG059430 Willard M Freeman
Veterans Affairs Oklahoma City I01BX003906 Willard M Freeman
University of Oklahoma Health P30 EY012190 Martin-Paul Agbaga
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 21 of 30
Research article Medicine
Sciences Center
National Institutes of Health R01 EY030513 Martin-Paul Agbaga
National Institutes of Health T32 AG052363 Shivani N Mann
Einstein Nathan Shock Center P30 AG038072 Michael B Stout
The funders had no role in study design, data collection and interpretation, or the
decision to submit the work for publication.
Author contributions
Shivani N Mann, Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodol-
ogy, Writing - original draft, Project administration, Writing - review and editing; Niran Hadad, Soft-
ware, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - review and
editing; Molly Nelson Holte, Formal analysis, Validation, Methodology; Alicia R Rothman, Formal
analysis, Validation, Writing - review and editing; Roshini Sathiaseelan, Formal analysis, Investigation,
Methodology; Samim Ali Mondal, Formal analysis, Investigation, Writing - review and editing; Mar-
tin-Paul Agbaga, Archana Unnikrishnan, Formal analysis, Investigation, Methodology, Writing -
review and editing; Malayannan Subramaniam, Formal analysis, Supervision, Investigation, Method-
ology, Writing - review and editing; John Hawse, Formal analysis, Supervision, Validation, Investiga-
tion, Methodology, Writing - review and editing; Derek M Huffman, Data curation, Formal analysis,
Validation, Investigation, Visualization, Methodology, Writing - review and editing; Willard M Free-
man, Data curation, Software, Formal analysis, Supervision, Investigation, Visualization, Writing -
review and editing; Michael B Stout, Conceptualization, Resources, Data curation, Formal analysis,
Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing -
original draft, Project administration, Writing - review and editing
Author ORCIDs
Willard M Freeman http://orcid.org/0000-0001-7027-999X
Michael B Stout https://orcid.org/0000-0002-9996-9123
Ethics
Animal experimentation: This study was performed in strict accordance with the recommendations
in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of
the animals were handled according to approved institutional animal care and use committee
(IACUC) protocols (#19-063-SEAHI) of the University of Oklahoma Health Science Center.
Decision letter and Author response
Decision letter https://doi.org/10.7554/eLife.59616.sa1
Author response https://doi.org/10.7554/eLife.59616.sa2
Additional files
Supplementary files
.
Supplementary file 1. Pairwise statistical comparisons of ERa binding. Negative binomial regres-
sion Wald post-hoc comparison test, FDR < 0.05. n = 3/group.
.
Supplementary file 2. ERa binding motif analysis. Motif analysis was performed using HOMER with
standard settings with the significance threshold set to FDR corrected p<0.05. Peak regions called
for each treatment group were analyzed to identify enriched motifs relative to the entire genome.
Hypergeometric test was used to test enrichment. Only motifs with FDR corrected p<0.05 were
reported as significant. For pairwise differential motif enrichment or depletion across experimental
groups, we utilized the hypergeometric test by using the number of sequences with motif from each
group and total number of peaks as total sample size. Motifs that appear in less than five sequences
between both test groups were removed. Benjamini-Hochberg multiple testing correction was uti-
lized to control for false discovery rate (FDR < 0.05).
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 22 of 30
Research article Medicine
.
Supplementary file 3. Circulating eicosanoid levels (pmol/ml). 17a-E2 mildly alters the circulating
eicosanoid profile in obese middle-aged male mice. WT mice were provided 45% HFD (TestDiet
58V8)±17a-E2 (14.4ppm) for 14 weeks. Age-matched, male WT, chow-fed (TestDiet 58YP) mice
were also evaluated as a normal-weight reference group. All data are presented as mean ± SEM and
were analyzed by Student’s t-test with the WT Chow group being excluded from statistical compari-
sons. n = 5–7.
.
Transparent reporting form
Data availability
Sequencing data has been deposited in GEO under accession code GSE151039.
The following dataset was generated:
Author(s) Year Dataset title Dataset URL
Database and
Identifier
Stout M, Hawse J,
Freeman W, Hadad
N, Mann S
2020 Assessment of transcriptional ERa
activity following exposure to 17a-
E2 and 17b-E2
https://www.ncbi.nlm.
nih.gov/geo/query/acc.
cgi?acc=GSE151039
NCBI Gene
Expression Omnibus,
GSE151039
References
Acosta-Martinez M
, Horton T, Levine JE. 2007. Estrogen receptors in neuropeptide Y neurons: at the crossroads
of feeding and reproduction. Trends in Endocrinology & Metabolism 18:48–50.
DOI: https://doi.org/10.1016/j.
tem.2006.12.001
, PMID: 17174101
Agbaga MP, Merriman DK, Brush RS, Lydic TA, Conley SM, Naash MI, Jackson S, Woods AS, Reid GE, Busik JV,
Anderson RE. 2018. Differential composition of DHA and very-long-chain PUFAs in rod and cone
photoreceptors. Journal of Lipid Research 59:1586–1596.
DOI: https://doi.org/10.1194/jlr.M082495, PMID: 2
9986998
Allard C, Morford JJ, Xu B, Salwen B, Xu W, Desmoulins L, Zsombok A, Kim JK, Levin ER, Mauvais-Jarvis F. 2019.
Loss of nuclear and membrane estrogen Receptor-a differentially impairs insulin secretion and action in male
and female mice. Diabetes 68:490–501.
DOI: https://doi.org/10.2337/db18-0293, PMID: 30305367
Anstead GM, Carlson KE, Katzenellenbogen JA. 1997. The estradiol pharmacophore: ligand structure-estrogen
receptor binding affinity relationships and a model for the receptor binding site. Steroids 62:268–303.
DOI: https://doi.org/10.1016/S0039-128X(96)00242-5, PMID: 9071738
Ayala JE, Samuel VT, Morton GJ, Obici S, Croniger CM, Shulman GI, Wasserman DH, McGuinness OP, NIH
Mouse Metabolic Phenotyping Center Consortium. 2010. Standard operating procedures for describing and
performing metabolic tests of glucose homeostasis in mice. Disease Models & Mechanisms 3:525–534.
DOI: https://doi.org/10.1242/dmm.006239, PMID: 20713647
Baeck C, Wehr A, Karlmark KR, Heymann F, Vucur M, Gassler N, Huss S, Klussmann S, Eulberg D, Luedde T,
Trautwein C, Tacke F. 2012. Pharmacological inhibition of the chemokine CCL2 (MCP-1) diminishes liver
macrophage infiltration and steatohepatitis in chronic hepatic injury. Gut 61:416–426.
DOI: https://doi.org/10.
1136/gutjnl-2011-300304
, PMID: 21813474
Barros RP, Gustafsson JA
˚
. 2011. Estrogen receptors and the metabolic network. Cell Metabolism 14:289–299.
DOI: https://doi.org/10.1016/j.cmet.2011.08.005, PMID: 21907136
Benedusi V, Martini E, Kallikourdis M, Villa A, Meda C, Maggi A. 2015. Ovariectomy shortens the life span of
female mice. Oncotarget 6:10801–10811.
DOI: https://doi.org/10.18632/oncotarget.2984, PMID: 25719423
Bischof GN, Park DC. 2015. Obesity and aging: consequences for cognition, brain structure, and brain function.
Psychosomatic Medicine 77:697–709.
DOI: https://doi.org/10.1097/PSY.0000000000000212, PMID: 26107577
Bligh EG, Dyer WJ. 1959. A rapid method of total lipid extraction and purification. Canadian Journal of
Biochemistry and Physiology 37:911–917.
DOI: https://doi.org/10.1139/o59-099, PMID: 13671378
Bolduc C, Larose M, Yoshioka M, Ye P, Belleau P, Labrie C, Morissette J, Raymond V, Labrie F, St-Amand J.
2004. Effects of dihydrotestosterone on adipose tissue measured by serial analysis of gene expression. Journal
of Molecular Endocrinology 33:429–444.
DOI: https://doi.org/10.1677/jme.1.01503, PMID: 15525599
Brandt C, Nolte H, Henschke S, Engstro
¨
m Ruud L, Awazawa M, Morgan DA, Gabel P, Sprenger HG, Hess ME,
Gu
¨
nther S, Langer T, Rahmouni K, Fenselau H, Kru
¨
ger M, Bru
¨
ning JC. 2018. Food perception primes hepatic ER
homeostasis via Melanocortin-Dependent control of mTOR activation. Cell 175:1321–1335.
DOI: https://doi.
org/10.1016/j.cell.2018.10.015
, PMID: 30445039
Brussaard HE, Gevers Leuven JA, Fro
¨
lich M, Kluft C, Krans HM. 1997. Short-term oestrogen replacement therapy
improves insulin resistance, lipids and fibrinolysis in postmenopausal women with NIDDM. Diabetologia 40:
843–849.
DOI: https://doi.org/10.1007/s001250050758, PMID: 9243107
Camporez JP, Jornayvaz FR, Lee HY, Kanda S, Guigni BA, Kahn M, Samuel VT, Carvalho CR, Petersen KF,
Jurczak MJ, Shulman GI. 2013. Cellular mechanism by which estradiol protects female ovariectomized mice
from high-fat diet-induced hepatic and muscle insulin resistance. Endocrinology 154:1021–1028.
DOI: https://
doi.org/10.1210/en.2012-1989
, PMID: 23364948
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 23 of 30
Research article Medicine
Cherrington AD, Wasserman DH, McGinness OP. 1994. Renal contribution to glucose production after a brief
fast: fact or fancy? The Journal of Clinical Investigation 93:2303.
DOI: https://doi.org/10.1172/JCI117232,
PMID: 8200961
Crary GS, Albrecht JH. 1998. Expression of cyclin-dependent kinase inhibitor p21 in human liver. Hepatology 28:
738–743.
DOI: https://doi.org/10.1002/hep.510280320, PMID: 9731566
D’Amato NC, Gordon MA, Babbs B, Spoelstra NS, Carson Butterfield KT, Torkko KC, Phan VT, Barton VN,
Rogers TJ, Sartorius CA, Elias A, Gertz J, Jacobsen BM, Richer JK. 2016. Cooperative dynamics of AR and ER
activity in breast Cancer. Molecular Cancer Research : MCR 14:1054–1067.
DOI: https://doi.org/10.1158/1541-
7786.MCR-16-0167
, PMID: 27565181
Debarba LK. 2020. Sex hormones underlying 17a-estradiol effects on neuroinflammation. bioRxiv . DOI: https://
doi.org/10.1101/2020.05.26.117689
Della Torre S, Mitro N, Fontana R, Gomaraschi M, Favari E, Recordati C, Lolli F, Quagliarini F, Meda C, Ohlsson
C, Crestani M, Uhlenhaut NH, Calabresi L, Maggi A. 2016. An essential role for liver era in coupling hepatic
metabolism to the reproductive cycle. Cell Reports 15:360–371.
DOI: https://doi.org/10.1016/j.celrep.2016.03.
019
, PMID: 27050513
Dirks AJ, Leeuwenburgh C. 2006. Caloric restriction in humans: potential pitfalls and health concerns.
Mechanisms of Ageing and Development 127:1–7.
DOI: https://doi.org/10.1016/j.mad.2005.09.001,
PMID: 16226298
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. 2013. STAR:
ultrafast universal RNA-seq aligner. Bioinformatics 29:15–21.
DOI: https://doi.org/10.1093/bioinformatics/
bts635
, PMID: 23104886
Dodd GT, Michael NJ, Lee-Young RS, Mangiafico SP, Pryor JT, Munder AC, Simonds SE, Bru
¨
ning JC, Zhang ZY,
Cowley MA, Andrikopoulos S, Horvath TL, Spanswick D, Tiganis T. 2018. Insulin regulates POMC neuronal
plasticity to control glucose metabolism. eLife 7:e38704.
DOI: https://doi.org/10.7554/eLife.38704,
PMID: 30230471
Dowman JK, Hopkins LJ, Reynolds GM, Armstrong MJ, Nasiri M, Nikolaou N, van Houten EL, Visser JA, Morgan
SA, Lavery GG, Oprescu A, Hu
¨
bscher SG, Newsome PN, Tomlinson JW. 2013. Loss of 5a -reductase type 1
accelerates the development of hepatic steatosis but protects against hepatocellular carcinoma in male mice.
Endocrinology 154:4536–4547.
DOI: https://doi.org/10.1210/en.2013-1592, PMID: 24080367
Dye L, Boyle NB, Champ C, Lawton C. 2017. The relationship between obesity and cognitive health and decline.
The Proceedings of the Nutrition Society 76:443–454.
DOI: https://doi.org/10.1017/S0029665117002014,
PMID: 28889822
Edwards DP, McGUIRE WL. 1980. 17a-estradiol is a biologically active estrogen in human breast Cancer cells in
tissue culture*. Endocrinology 107:884–891.
DOI: https://doi.org/10.1210/endo-107-4-884
Einstein FH, Huffman DM, Fishman S, Jerschow E, Heo HJ, Atzmon G, Schechter C, Barzilai N, Muzumdar RH.
2010. Aging per se increases the susceptibility to free fatty Acid-Induced insulin resistance. The Journals of
Gerontology Series A: Biological Sciences and Medical Sciences 65A:800–808.
DOI: https://doi.org/10.1093/
gerona/glq078
Engler-Chiurazzi EB, Covey DF, Simpkins JW. 2017. A novel mechanism of non-feminizing estrogens in
neuroprotection. Experimental Gerontology 94:99–102.
DOI: https://doi.org/10.1016/j.exger.2016.10.013,
PMID: 27818250
Flegal KM, Carroll MD, Ogden CL, Curtin LR. 2010. Prevalence and trends in obesity among US adults, 1999-
2008. Jama 303:235–241.
DOI: https://doi.org/10.1001/jama.2009.2014, PMID: 20071471
Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. 2016. Trends in obesity among adults in the
united states, 2005 to 2014. Jama 315:2284–2291.
DOI: https://doi.org/10.1001/jama.2016.6458,
PMID: 27272580
Flouriot G, Griffin C, Kenealy M, Sonntag-Buck V, Gannon F. 1998. Differentially expressed messenger RNA
isoforms of the human estrogen receptor-alpha gene are generated by alternative splicing and promoter
usage. Molecular Endocrinology 12:1939–1954.
DOI: https://doi.org/10.1210/mend.12.12.0209, PMID: 984
9967
Folch J, Lees M, Sloane Stanley GH. 1957. A simple method for the isolation and purification of total lipides from
animal tissues. The Journal of Biological Chemistry 226:497–509.
PMID: 13428781
Garratt M, Bower B, Garcia GG, Miller RA. 2017. Sex differences in lifespan extension with acarbose and 17-a
estradiol: gonadal hormones underlie male-specific improvements in glucose tolerance and mTORC2 signaling.
Aging Cell 16:1256–1266.
DOI: https://doi.org/10.1111/acel.12656, PMID: 28834262
Garratt M, Lagerborg KA, Tsai YM, Galecki A, Jain M, Miller RA. 2018. Male lifespan extension with 17-a
estradiol is linked to a sex-specific metabolomic response modulated by gonadal hormones in mice. Aging Cell
17:e12786.
DOI: https://doi.org/10.1111/acel.12786, PMID: 29806096
Garratt M, Stout MB. 2018. Hormone actions controlling sex-specific life-extension. Aging 10:293–294.
DOI: https://doi.org/10.18632/aging.101396, PMID: 29514132
GBD 2017 Cirrhosis Collaborators, Collaborators G. 2020. The global, regional, and national burden of cirrhosis
by cause in 195 countries and territories, 1990-2017: a systematic analysis for the global burden of disease
study 2017. The Lancet. Gastroenterology & Hepatology 5:245–266.
DOI: https://doi.org/10.1016/S2468-1253
(19)30349-8
, PMID: 31981519
Gilroy DW, Edin ML, De Maeyer RP, Bystrom J, Newson J, Lih FB, Stables M, Zeldin DC, Bishop-Bailey D. 2016.
CYP450-derived oxylipins mediate inflammatory resolution. PNAS 113:E3240–E3249.
DOI: https://doi.org/10.
1073/pnas.1521453113
, PMID: 27226306
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 24 of 30
Research article Medicine
Glass O, Henao R, Patel K, Guy CD, Gruss HJ, Syn WK, Moylan CA, Streilein R, Hall R, Mae Diehl A, Abdelmalek
MF. 2018. Serum Interleukin-8, osteopontin, and monocyte chemoattractant protein 1 are associated with
hepatic fibrosis in patients with nonalcoholic fatty liver disease. Hepatology Communications 2:1344–1355.
DOI: https://doi.org/10.1002/hep4.1237, PMID: 30411081
Gould ML, Hurst PR, Nicholson HD. 2007. The effects of oestrogen receptors alpha and beta on testicular cell
number and steroidogenesis in mice. Reproduction 134:271–279.
DOI: https://doi.org/10.1530/REP-07-0025,
PMID: 17660237
Grattagliano I, Montezinho LP, Oliveira PJ, Fru
¨
hbeck G, Go´ mez-Ambrosi J, Montecucco F, Carbone F,
Wieckowski MR, Wang DQ, Portincasa P. 2019. Targeting mitochondria to oppose the progression of
nonalcoholic fatty liver disease. Biochemical Pharmacology 160:34–45.
DOI: https://doi.org/10.1016/j.bcp.
2018.11.020
, PMID: 30508523
Green PS, Simpkins JW. 2000. Estrogens and estrogen-like non-feminizing compounds their role in the
prevention and treatment of alzheimer’s disease. Annals of the New York Academy of Sciences 924:93–98.
DOI: https://doi.org/10.1111/j.1749-6632.2000.tb05566.x, PMID: 11193809
Guillaume M, Riant E, Fabre A, Raymond-Letron I, Buscato M, Davezac M, Tramunt B, Montagner A, Smati S,
Zahreddine R, Palierne G, Valera MC, Guillou H, Lenfant F, Unsicker K, Metivier R, Fontaine C, Arnal JF, Gourdy
P. 2019. Selective liver estrogen receptor a modulation prevents steatosis, Diabetes, and Obesity Through the
Anorectic Growth Differentiation Factor 15 Hepatokine in Mice. Hepatology Communications 3:908–924.
DOI: https://doi.org/10.1002/hep4.1363, PMID: 31304450
Guy J, Peters MG. 2013. Liver disease in women: the influence of gender on epidemiology, natural history, and
patient outcomes. Gastroenterologia Y HepatologiA 9:633–639.
Hadad N, Masser DR, Blanco-Berdugo L, Stanford DR, Freeman WM. 2019. Early-life DNA methylation profiles
are indicative of age-related transcriptome changes. Epigenetics & Chromatin 12:58.
DOI: https://doi.org/10.
1186/s13072-019-0306-5
, PMID: 31594536
Harrison DE, Strong R, Allison DB, Ames BN, Astle CM, Atamna H, Fernandez E, Flurkey K, Javors MA, Nadon
NL, Nelson JF, Pletcher S, Simpkins JW, Smith D, Wilkinson JE, Miller RA. 2014. Acarbose, 17-a-estradiol, and
nordihydroguaiaretic acid extend mouse lifespan preferentially in males. Aging Cell 13:273–282.
DOI: https://
doi.org/10.1111/acel.12170
, PMID: 24245565
Hayashi M, Nomoto S, Hishida M, Inokawa Y, Kanda M, Okamura Y, Nishikawa Y, Tanaka C, Kobayashi D,
Yamada S, Nakayama G, Fujii T, Sugimoto H, Koike M, Fujiwara M, Takeda S, Kodera Y. 2014. Identification of
the collagen type 1 a 1 gene (COL1A1) as a candidate survival-related factor associated with hepatocellular
carcinoma. BMC Cancer 14:108.
DOI: https://doi.org/10.1186/1471-2407-14-108, PMID: 24552139
Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS. 2000. Increased adipose tissue in male and female
estrogen receptor-alpha knockout mice. PNAS 97:12729–12734.
DOI: https://doi.org/10.1073/pnas.97.23.
12729
, PMID: 11070086
Horvath S, Erhart W, Brosch M, Ammerpohl O, von Scho
¨
nfels W, Ahrens M, Heits N, Bell JT, Tsai PC, Spector
TD, Deloukas P, Siebert R, Sipos B, Becker T, Ro
¨
cken C, Schafmayer C, Hampe J. 2014. Obesity accelerates
epigenetic aging of human liver. PNAS 111:15538–15543.
DOI: https://doi.org/10.1073/pnas.1412759111,
PMID: 25313081
Huffman DM, Farias Quipildor G, Mao K, Zhang X, Wan J, Apontes P, Cohen P, Barzilai N. 2016a. Central
insulin-like growth factor-1 (IGF-1) restores whole-body insulin action in a model of age-related insulin
resistance and IGF-1 decline. Aging Cell 15:181–186.
DOI: https://doi.org/10.1111/acel.12415, PMID: 2653486
9
Huffman DM, Justice JN, Stout MB, Kirkland JL, Barzilai N, Austad SN. 2016b. Evaluating health span in
preclinical models of aging and disease: guidelines, challenges, and opportunities for geroscience. The Journals
of Gerontology Series A: Biological Sciences and Medical Sciences 71:1395–1406.
DOI: https://doi.org/10.
1093/gerona/glw106
Hunt NJ, Kang SWS, Lockwood GP, Le Couteur DG, Cogger VC. 2019. Hallmarks of aging in the liver.
Computational and Structural Biotechnology Journal 17:1151–1161.
DOI: https://doi.org/10.1016/j.csbj.2019.
07.021
, PMID: 31462971
In
˜
igo MR, Amorese AJ, Tarpey MD, Balestrieri NP, Jones KG, Patteson DJ, Jackson KC, Torres MJ, Lin CT, Smith
CD, Heden TD, McMillin SL, Weyrauch LA, Stanley EC, Schmidt CA, Kilburg-Basnyat BB, Reece SW, Psaltis CE,
Leinwand LA, Funai K, et al. 2020. Estrogen receptor-a in female skeletal muscle is not required for regulation
of muscle insulin sensitivity and mitochondrial regulation. Molecular Metabolism 34:1–15.
DOI: https://doi.org/
10.1016/j.molmet.2019.12.010
, PMID: 32180550
Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, Hu FB, Hubbard VS, Jakicic JM, Kushner
RF, Loria CM, Millen BE, Nonas CA, Pi-Sunyer FX, Stevens J, Stevens VJ, Wadden TA, Wolfe BM, Yanovski SZ,
Jordan HS, et al. 2014. AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a
report of the American college of cardiology/American heart association task force on practice guidelines and
the obesity society. Circulation 129:S102–S138.
DOI: https://doi.org/10.1161/01.cir.0000437739.71477.ee,
PMID: 24222017
Kaur SP, Bansal S, Chopra K. 2015. 17a-estradiol: a candidate neuroserm and non-feminizing estrogen for
postmenopausal neuronal complications. Steroids 96:7–15.
DOI: https://doi.org/10.1016/j.steroids.2015.01.
004
, PMID: 25595449
Kelly MJ, Rønnekleiv OK. 2015. Minireview: neural signaling of estradiol in the hypothalamus. Molecular
Endocrinology 29:645–657.
DOI: https://doi.org/10.1210/me.2014-1397, PMID: 25751314
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 25 of 30
Research article Medicine
Khosla S, Monroe DG. 2018. Regulation of bone metabolism by sex steroids. Cold Spring Harbor Perspectives in
Medicine 8:a031211.
DOI: https://doi.org/10.1101/cshperspect.a031211, PMID: 28710257
Kim IH, Kisseleva T, Brenner DA. 2015. Aging and liver disease. Current Opinion in Gastroenterology 31:184–
191.
DOI: https://doi.org/10.1097/MOG.0000000000000176, PMID: 25850346
Kiss L, Schu
¨
tte H, Padberg W, Weissmann N, Mayer K, Gessler T, Voswinckel R, Seeger W, Grimminger F. 2010.
Epoxyeicosatrienoates are the dominant eicosanoids in human lungs upon microbial challenge. European
Respiratory Journal 36:1088–1098.
DOI: https://doi.org/10.1183/09031936.00000309, PMID: 20378604
Kohli P, Levy BD. 2009. Resolvins and protectins: mediating solutions to inflammation. British Journal of
Pharmacology 158:960–971.
DOI: https://doi.org/10.1111/j.1476-5381.2009.00290.x, PMID: 19594757
Ko
¨
nner AC, Janoschek R, Plum L, Jordan SD, Rother E, Ma X, Xu C, Enriori P, Hampel B, Barsh GS, Kahn CR,
Cowley MA, Ashcroft FM, Bru
¨
ning JC. 2007. Insulin action in AgRP-expressing neurons is required for
suppression of hepatic glucose production. Cell Metabolism 5:438–449.
DOI: https://doi.org/10.1016/j.cmet.
2007.05.004
, PMID: 17550779
Koo BK, Um SH, Seo DS, Joo SK, Bae JM, Park JH, Chang MS, Kim JH, Lee J, Jeong WI, Kim W. 2018. Growth
differentiation factor 15 predicts advanced fibrosis in biopsy-proven non-alcoholic fatty liver disease. Liver
International 38:695–705.
DOI: https://doi.org/10.1111/liv.13587, PMID: 28898507
Korenman SG. 1969. Comparative binding affinity of estrogens and its relation to estrogenic potency. Steroids
13:163–177.
DOI: https://doi.org/10.1016/0039-128X(69)90004-X, PMID: 5773887
Langmead B, Salzberg SL. 2012. Fast gapped-read alignment with bowtie 2. Nature Methods 9:357–359.
DOI: https://doi.org/10.1038/nmeth.1923, PMID: 22388286
Lee SK, Choi HS, Song MR, Lee MO, Lee JW. 1998. Estrogen receptor, a common interaction partner for a
subset of nuclear receptors. Molecular Endocrinology 12:1184–1192.
DOI: https://doi.org/10.1210/mend.12.8.
0146
, PMID: 9717844
Leonard AK, Loughran EA, Klymenko Y, Liu Y, Kim O, Asem M, McAbee K, Ravosa MJ, Stack MS. 2018. Methods
for the visualization and analysis of extracellular matrix protein structure and degradation. Methods in Cell
Biology 143:79–95.
DOI: https://doi.org/10.1016/bs.mcb.2017.08.005, PMID: 29310793
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, 1000 Genome
Project Data Processing Subgroup. 2009. The sequence alignment/Map format and SAMtools. Bioinformatics
25:2078–2079.
DOI: https://doi.org/10.1093/bioinformatics/btp352, PMID: 19505943
Lichtinghagen R, Bahr MJ, Wehmeier M, MichelsI D, Haberkorn CI, Arndt B, Flemming P, Manns MP, Boeker
KHW. 2003. Expression and coordinated regulation of matrix metalloproteinases in chronic hepatitis C and
hepatitis C virus-induced liver cirrhosis. Clinical Science 105:373–382.
DOI: https://doi.org/10.1042/
CS20030098
Lin CY, Vega VB, Thomsen JS, Zhang T, Kong SL, Xie M, Chiu KP, Lipovich L, Barnett DH, Stossi F, Yeo A,
George J, Kuznetsov VA, Lee YK, Charn TH, Palanisamy N, Miller LD, Cheung E, Katzenellenbogen BS, Ruan Y,
et al. 2007. Whole-genome cartography of estrogen receptor alpha binding sites. PLOS Genetics 3:e87.
DOI: https://doi.org/10.1371/journal.pgen.0030087, PMID: 17542648
Lin AH, Li RW, Ho EY, Leung GP, Leung SW, Vanhoutte PM, Man RY. 2013. Differential ligand binding affinities
of human estrogen Receptor-a isoforms. PLOS ONE 8:e63199.
DOI: https://doi.org/10.1371/journal.pone.
0063199
, PMID: 23646196
Littlefield BA, Gurpide E, Markiewicz L, McKinley B, Hochberg RB. 1990. A simple and sensitive microtiter plate
estrogen bioassay based on stimulation of alkaline phosphatase in Ishikawa cells: estrogenic action of Delta
5
adrenal steroids. Endocrinology 127:2757–2762. DOI: https://doi.org/10.1210/endo-127-6-2757, PMID: 224
9627
Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using real-time quantitative PCR and
the 2(-Delta delta C(T)) Method. Methods 25:402–408.
DOI: https://doi.org/10.1006/meth.2001.1262,
PMID: 11846609
Livingstone DE, Barat P, Di Rollo EM, Rees GA, Weldin BA, Rog-Zielinska EA, MacFarlane DP, Walker BR,
Andrew R. 2015. 5a-Reductase type 1 deficiency or inhibition predisposes to insulin resistance, hepatic
Steatosis, and liver fibrosis in rodents. Diabetes 64:447–458.
DOI: https://doi.org/10.2337/db14-0249,
PMID: 25239636
Lo
´
pez M, Tena-Sempere M. 2015. Estrogens and the control of energy homeostasis: a brain perspective. Trends
in Endocrinology & Metabolism 26:411–421.
DOI: https://doi.org/10.1016/j.tem.2015.06.003, PMID: 26126705
Lo
´
pez-Otı´n C, Blasco MA, Partridge L, Serrano M, Kroemer G. 2013. The hallmarks of aging. Cell 153:1194–
1217.
DOI: https://doi.org/10.1016/j.cell.2013.05.039, PMID: 23746838
Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with
DESeq2. Genome Biology 15:550.
DOI: https://doi.org/10.1186/s13059-014-0550-8, PMID: 25516281
Lu M, Wan M, Leavens KF, Chu Q, Monks BR, Fernandez S, Ahima RS, Ueki K, Kahn CR, Birnbaum MJ. 2012.
Insulin regulates liver metabolism in vivo in the absence of hepatic akt and Foxo1. Nature Medicine 18:388–
395.
DOI: https://doi.org/10.1038/nm.2686, PMID: 22344295
Lua I, Li Y, Zagory JA, Wang KS, French SW, Se´ vigny J, Asahina K. 2016. Characterization of hepatic stellate cells,
portal fibroblasts, and mesothelial cells in normal and fibrotic livers. Journal of Hepatology 64:1137–1146.
DOI: https://doi.org/10.1016/j.jhep.2016.01.010, PMID: 26806818
Madala SK, Pesce JT, Ramalingam TR, Wilson MS, Minnicozzi S, Cheever AW, Thompson RW, Mentink-Kane
MM, Wynn TA. 2010. Matrix metalloproteinase 12-deficiency augments extracellular matrix degrading
metalloproteinases and attenuates IL-13-dependent fibrosis. The Journal of Immunology 184:3955–3963.
DOI: https://doi.org/10.4049/jimmunol.0903008, PMID: 20181883
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 26 of 30
Research article Medicine
Mann SN, Pitel KS, Nelson-Holte MH, Iwaniec UT, Turner RT, Sathiaseelan R, Kirkland JL, Schneider A, Morris KT,
Malayannan S, Hawse JR, Stout MB. 2020. 17a-estradiol prevents ovariectomy-mediated obesity and bone loss.
Experimental Gerontology 142:111113.
DOI: https://doi.org/10.1016/j.exger.2020.111113, PMID: 33065227
Manrique C, Lastra G, Habibi J, Mugerfeld I, Garro M, Sowers JR. 2012. Loss of estrogen receptor a signaling
leads to insulin resistance and obesity in young and adult female mice. Cardiorenal Medicine 2:200–210.
DOI: https://doi.org/10.1159/000339563, PMID: 22969776
Meda C, Barone M, Mitro N, Lolli F, Pedretti S, Caruso D, Maggi A, Della Torre S. 2020. Hepatic era accounts for
sex differences in the ability to cope with an excess of dietary lipids. Molecular Metabolism 32:97–108.
DOI: https://doi.org/10.1016/j.molmet.2019.12.009, PMID: 32029233
Mehlem A, Hagberg CE, Muhl L, Eriksson U, Falkevall A. 2013. Imaging of neutral lipids by oil red O for
analyzing the metabolic status in health and disease. Nature Protocols 8:1149–1154.
DOI: https://doi.org/10.
1038/nprot.2013.055
, PMID: 23702831
Miller BF. 2020. Short-term calorie restriction and 17a-estradiol administration elicit divergent effects on
proteostatic processes and protein content in metabolically active tissues. The Journals of Gerontology: Series
A 75 :849–857.
DOI: https://doi.org/10.1093/gerona/glz113
Monroe DG, Getz BJ, Johnsen SA, Riggs BL, Khosla S, Spelsberg TC. 2003. Estrogen receptor isoform-specific
regulation of endogenous gene expression in human osteoblastic cell lines expressing either ERalpha or
ERbeta. Journal of Cellular Biochemistry 90:315–326.
DOI: https://doi.org/10.1002/jcb.10633, PMID: 14505348
Monroe DG, Secreto FJ, Subramaniam M, Getz BJ, Khosla S, Spelsberg TC. 2005. Estrogen receptor alpha and
beta heterodimers exert unique effects on estrogen- and tamoxifen-dependent gene expression in human
U2OS osteosarcoma cells. Molecular Endocrinology 19:1555–1568.
DOI: https://doi.org/10.1210/me.2004-
0381
, PMID: 15802376
Move
´
rare-Skrtic S, Venken K, Andersson N, Lindberg MK, Svensson J, Swanson C, Vanderschueren D, Oscarsson
J, Gustafsson JA, Ohlsson C. 2006. Dihydrotestosterone treatment results in obesity and altered lipid
metabolism in orchidectomized mice. Obesity 14:662–672.
DOI: https://doi.org/10.1038/oby.2006.75,
PMID: 16741268
Murray SA, Eppig JT, Smedley D, Simpson EM, Rosenthal N. 2012. Beyond knockouts: cre resources for
conditional mutagenesis. Mammalian Genome 23:587–599.
DOI: https://doi.org/10.1007/s00335-012-9430-2,
PMID: 22926223
Nelson JD, Denisenko O, Sova P, Bomsztyk K. 2006. Fast chromatin immunoprecipitation assay. Nucleic Acids
Research 34:e2.
DOI: https://doi.org/10.1093/nar/gnj004, PMID: 16397291
Nevalainen T, Kananen L, Marttila S, Jylha
¨
va
¨
J, Mononen N, Ka
¨
ho
¨
nen M, Raitakari OT, Hervonen A, Jylha
¨
M,
Lehtima
¨
ki T, Hurme M. 2017. Obesity accelerates epigenetic aging in middle-aged but not in elderly
individuals. Clinical Epigenetics 9:20.
DOI: https://doi.org/10.1186/s13148-016-0301-7, PMID: 28289477
Ogrodnik M, Miwa S, Tchkonia T, Tiniakos D, Wilson CL, Lahat A, Day CP, Burt A, Palmer A, Anstee QM,
Grellscheid SN, Hoeijmakers JHJ, Barnhoorn S, Mann DA, Bird TG, Vermeij WP, Kirkland JL, Passos JF, von
Zglinicki T, Jurk D. 2017. Cellular senescence drives age-dependent hepatic steatosis. Nature Communications
8:15691.
DOI: https://doi.org/10.1038/ncomms15691, PMID: 28608850
Orfanos CE, Vogels L. 1980. [Local therapy of androgenetic alopecia with 17 alpha-estradiol A controlled,
randomized double-blind study (author’s transl)]. Dermatologica 161:124–132.
PMID: 7398983
Panet-Raymond V, Gottlieb B, Beitel LK, Pinsky L, Trifiro MA. 2000. Interactions between androgen and
estrogen receptors and the effects on their transactivational properties. Molecular and Cellular Endocrinology
167:139–150.
DOI: https://doi.org/10.1016/S0303-7207(00)00279-3, PMID: 11000528
Parks BW, Sallam T, Mehrabian M, Psychogios N, Hui ST, Norheim F, Castellani LW, Rau CD, Pan C, Phun J,
Zhou Z, Yang WP, Neuhaus I, Gargalovic PS, Kirchgessner TG, Graham M, Lee R, Tontonoz P, Gerszten RE,
Hevener AL, et al. 2015. Genetic architecture of insulin resistance in the mouse. Cell Metabolism 21:334–347.
DOI: https://doi.org/10.1016/j.cmet.2015.01.002, PMID: 25651185
Pedram A, Razandi M, Blumberg B, Levin ER. 2016. Membrane and nuclear estrogen receptor a collaborate to
suppress adipogenesis but not triglyceride content. The FASEB Journal 30:230–240.
DOI: https://doi.org/10.
1096/fj.15-274878
, PMID: 26373802
Pe
´
rez LM, Pareja-Galeano H, Sanchis-Gomar F, Emanuele E, Lucia A, Ga´ lvez BG. 2016. ’Adipaging’: ageing and
obesity share biological hallmarks related to a dysfunctional adipose tissue. The Journal of Physiology 594:
3187–3207.
DOI: https://doi.org/10.1113/JP271691, PMID: 26926488
Peters AA, Buchanan G, Ricciardelli C, Bianco-Miotto T, Centenera MM, Harris JM, Jindal S, Segara D, Jia L,
Moore NL, Henshall SM, Birrell SN, Coetzee GA, Sutherland RL, Butler LM, Tilley WD. 2009. Androgen
receptor inhibits estrogen receptor-alpha activity and is prognostic in breast Cancer. Cancer Research 69:
6131–6140.
DOI: https://doi.org/10.1158/0008-5472.CAN-09-0452, PMID: 19638585
Pocai A, Obici S, Schwartz GJ, Rossetti L. 2005a. A brain-liver circuit regulates glucose homeostasis. Cell
Metabolism 1:53–61.
DOI: https://doi.org/10.1016/j.cmet.2004.11.001, PMID: 16054044
Pocai A, Lam TK, Gutierrez-Juarez R, Obici S, Schwartz GJ, Bryan J, Aguilar-Bryan L, Rossetti L. 2005b.
Hypothalamic K(ATP) channels control hepatic glucose production. Nature 434:1026–1031.
DOI: https://doi.
org/10.1038/nature03439
, PMID: 15846348
Qiu S, Vazquez JT, Boulger E, Liu H, Xue P, Hussain MA, Wolfe A. 2017. Hepatic estrogen receptor a is critical
for regulation of gluconeogenesis and lipid metabolism in males. Scientific Reports 7:1661.
DOI: https://doi.
org/10.1038/s41598-017-01937-4
, PMID: 28490809
Quehenberger O, Armando AM, Brown AH, Milne SB, Myers DS, Merrill AH, Bandyopadhyay S, Jones KN, Kelly
S, Shaner RL, Sullards CM, Wang E, Murphy RC, Barkley RM, Leiker TJ, Raetz CR, Guan Z, Laird GM, Six DA,
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 27 of 30
Research article Medicine
Russell DW, et al. 2010. Lipidomics reveals a remarkable diversity of lipids in human plasma. Journal of Lipid
Research 51:3299–3305.
DOI: https://doi.org/10.1194/jlr.M009449, PMID: 20671299
Reese JM, Bruinsma ES, Nelson AW, Chernukhin I, Carroll JS, Li Y, Subramaniam M, Suman VJ, Negron V,
Monroe DG, Ingle JN, Goetz MP, Hawse JR. 2018. ERb-mediated induction of cystatins results in suppression
of tgfb signaling and inhibition of triple-negative breast Cancer metastasis. PNAS 115:E9580–E9589.
DOI: https://doi.org/10.1073/pnas.1807751115, PMID: 30257941
Rosenfeld CS, Ganjam VK, Taylor JA, Yuan X, Stiehr JR, Hardy MP, Lubahn DB. 1998. Transcription and
translation of estrogen receptor-beta in the male reproductive tract of estrogen receptor-alpha knock-out and
wild-type mice. Endocrinology 139:2982–2987.
DOI: https://doi.org/10.1210/endo.139.6.6028, PMID: 9607809
Ross-Innes CS, Stark R, Teschendorff AE, Holmes KA, Ali HR, Dunning MJ, Brown GD, Gojis O, Ellis IO, Green
AR, Ali S, Chin SF, Palmieri C, Caldas C, Carroll JS. 2012. Differential oestrogen receptor binding is associated
with clinical outcome in breast cancer. Nature 481:389–393.
DOI: https://doi.org/10.1038/nature10730,
PMID: 22217937
Rossi R, Origliani G, Modena MG. 2004. Transdermal 17-beta-estradiol and risk of developing type 2 diabetes in
a population of healthy, nonobese postmenopausal women. Diabetes Care 27:645–649.
DOI: https://doi.org/
10.2337/diacare.27.3.645
, PMID: 14988279
Rubinow KB. 2017. Estrogens and body weight regulation in men. Advances in Experimental Medicine and
Biology 1043:285–313.
DOI: https://doi.org/10.1007/978-3-319-70178-3_14, PMID: 29224100
Ruud J, Steculorum SM, Bru
¨
ning JC. 2017. Neuronal control of peripheral insulin sensitivity and glucose
metabolism. Nature Communications 8:15259.
DOI: https://doi.org/10.1038/ncomms15259, PMID: 28469281
Salvestrini V, Sell C, Lorenzini A. 2019. Obesity may accelerate the aging process. Frontiers in Endocrinology 10:
266.
DOI: https://doi.org/10.3389/fendo.2019.00266, PMID: 31130916
Sa
´
nchez-Criado JE, Trudgen K, Milla´ n Y, Blanco A, Monterde J, Garrido-Gracia JC, Gordon A, Aguilar R, de Las
Mulas JM, Ko C. 2012. Estrogen receptor (ESR) 2 partially offsets the absence of ESR1 in gonadotropes of
pituitary-specific Esr1 knockout female mice. Reproduction 143:549–558.
DOI: https://doi.org/10.1530/REP-11-
0214
, PMID: 22367588
Sar M, Sahu A, Crowley WR, Kalra SP. 1990. Localization of neuropeptide-Y immunoreactivity in estradiol-
concentrating cells in the hypothalamus. Endocrinology 127:2752–2756.
DOI: https://doi.org/10.1210/endo-
127-6-2752
, PMID: 2249626
Schriefers H, Wright MC, Rozman T, Hevert F. 1991. Inhibition of testosterone metabolism by 17-alpha-estradiol
in rat liver slices]. Arzneimittel-Forschung 41:1186–1189.
PMID: 1810267
Sidhom S, Schneider A, Fang Y, McFadden S, Darcy J, Sathiaseelan R, Palmer AK, Steyn FJ, Grillari J, Kopchick
JJ, Bartke A, Siddiqi S, Masternak MM, Stout MB. 2020. 17a-estradiol modulates IGF1 and hepatic gene
expression in a Sex-Specific manner. The Journals of Gerontology: Series A 28:glaa215.
DOI: https://doi.org/
10.1093/gerona/glaa215
Skinner DC, Herbison AE. 1997. Effects of photoperiod on estrogen receptor, tyrosine hydroxylase,
neuropeptide Y, and beta-endorphin immunoreactivity in the ewe hypothalamus. Endocrinology 138:2585–
2595.
DOI: https://doi.org/10.1210/endo.138.6.5208, PMID: 9165052
Smith AW, Bosch MA, Wagner EJ, Rønnekleiv OK, Kelly MJ. 2013. The membrane estrogen receptor ligand STX
rapidly enhances GABAergic signaling in NPY/AgRP neurons: role in mediating the anorexigenic effects of 17b-
estradiol. American Journal of Physiology-Endocrinology and Metabolism 305:E632–E640.
DOI: https://doi.
org/10.1152/ajpendo.00281.2013
, PMID: 23820624
Smith AW, Rønnekleiv OK, Kelly MJ. 2014. Gq-mER signaling has opposite effects on hypothalamic orexigenic
and anorexigenic neurons. Steroids 81:31–35.
DOI: https://doi.org/10.1016/j.steroids.2013.11.007, PMID: 2426
9736
Stefanska A, Bergmann K, Sypniewska G. 2015. Metabolic syndrome and menopause: pathophysiology, clinical
and diagnostic significance. Advances in Clinical Chemistry 72:1–75.
DOI: https://doi.org/10.1016/bs.acc.2015.
07.001
, PMID: 26471080
Steyn FJ, Ngo ST, Chen VP, Bailey-Downs LC, Xie TY, Ghadami M, Brimijoin S, Freeman WM, Rubinstein M, Low
MJ, Stout MB. 2018. 17a-estradiol acts through hypothalamic pro-opiomelanocortin expressing neurons to
reduce feeding behavior. Aging Cell 17:e12703.
DOI: https://doi.org/10.1111/acel.12703
Stincic TL, Rønnekleiv OK, Kelly MJ. 2018. Diverse actions of estradiol on anorexigenic and orexigenic
hypothalamic arcuate neurons. Hormones and Behavior 104:146–155.
DOI: https://doi.org/10.1016/j.yhbeh.
2018.04.001
, PMID: 29626486
Stout MB, Liu LF, Belury MA. 2011. Hepatic steatosis by dietary-conjugated linoleic acid is accompanied by
accumulation of diacylglycerol and increased membrane-associated protein kinase C e in mice. Molecular
Nutrition & Food Research 55:1010–1017.
DOI: https://doi.org/10.1002/mnfr.201000413, PMID: 21480517
Stout MB, Justice JN, Nicklas BJ, Kirkland JL. 2017a. Physiological aging: links among adipose tissue
dysfunction, Diabetes, and Frailty. Physiology 32:9–19.
DOI: https://doi.org/10.1152/physiol.00012.2016,
PMID: 27927801
Stout MB, Steyn FJ, Jurczak MJ, Camporez J-PG, Zhu Y, Hawse JR, Jurk D, Palmer AK, Xu M, Pirtskhalava T,
Evans GL, de Souza Santos R, Frank AP, White TA, Monroe DG, Singh RJ, Casaclang-Verzosa G, Miller JD,
Clegg DJ, LeBrasseur NK, et al. 2017b. 17a-estradiol alleviates Age-related metabolic and inflammatory
dysfunction in male mice without inducing feminization. The Journals of Gerontology Series A: Biological
Sciences and Medical Sciences 72:3–15.
DOI: https://doi.org/10.1093/gerona/glv309
Strong R, Miller RA, Antebi A, Astle CM, Bogue M, Denzel MS, Fernandez E, Flurkey K, Hamilton KL, Lamming
DW, Javors MA, de Magalha
˜
es JP, Martinez PA, McCord JM, Miller BF, Mu
¨
ller M, Nelson JF, Ndukum J,
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 28 of 30
Research article Medicine
Rainger GE, Richardson A, et al. 2016. Longer lifespan in male mice treated with a weakly estrogenic agonist,
an antioxidant, an a-glucosidase inhibitor or a Nrf2-inducer. Aging Cell 15:872–884.
DOI: https://doi.org/10.
1111/acel.12496
, PMID: 27312235
Stubbins RE, Holcomb VB, Hong J, Nu
´
n
˜
ez NP. 2012. Estrogen modulates abdominal adiposity and protects
female mice from obesity and impaired glucose tolerance. European Journal of Nutrition 51:861–870.
DOI: https://doi.org/10.1007/s00394-011-0266-4, PMID: 22042005
Szefel J, Kruszewski WJ, Sobczak E. 2015. Factors influencing the eicosanoids synthesis in vivo. BioMed Research
International 2015:690692.
DOI: https://doi.org/10.1155/2015/690692, PMID: 25861641
Tao R, Wang C, Sto
¨
hr O, Qiu W, Hu Y, Miao J, Dong XC, Leng S, Stefater M, Stylopoulos N, Lin L, Copps KD,
White MF. 2018. Inactivating hepatic follistatin alleviates hyperglycemia. Nature Medicine 24:1058–1069.
DOI: https://doi.org/10.1038/s41591-018-0048-0, PMID: 29867232
Taylor SE, Martin-Hirsch PL, Martin FL. 2010. Oestrogen receptor splice variants in the pathogenesis of disease.
Cancer Letters 288:133–148.
DOI: https://doi.org/10.1016/j.canlet.2009.06.017, PMID: 19608332
Tchkonia T, Morbeck DE, Von Zglinicki T, Van Deursen J, Lustgarten J, Scrable H, Khosla S, Jensen MD, Kirkland
JL. 2010. Fat tissue, aging, and cellular senescence. Aging Cell 9:667–684.
DOI: https://doi.org/10.1111/j.1474-
9726.2010.00608.x
, PMID: 20701600
Timper K, Bru
¨
ning JC. 2017. Hypothalamic circuits regulating appetite and energy homeostasis: pathways to
obesity. Disease Models & Mechanisms 10:679–689.
DOI: https://doi.org/10.1242/dmm.026609, PMID: 285
92656
Toran-Allerand CD, Guan X, MacLusky NJ, Horvath TL, Diano S, Singh M, Connolly ES, Nethrapalli IS, Tinnikov
AA. 2002. ER-X: a novel, plasma membrane-associated, putative estrogen receptor that is regulated during
development and after ischemic brain injury. The Journal of Neuroscience 22:8391–8401.
DOI: https://doi.org/
10.1523/JNEUROSCI.22-19-08391.2002
, PMID: 12351713
Toran-Allerand CD. 2005. Estrogen and the brain: beyond ER-alpha, ER-beta, and 17beta-estradiol. Annals of
the New York Academy of Sciences 1052:136–144.
DOI: https://doi.org/10.1196/annals.1347.009,
PMID: 16024756
Toran-Allerand CD, Tinnikov AA, Singh RJ, Nethrapalli IS. 2005. 17alpha-estradiol: a brain-active estrogen?
Endocrinology 146:3843–3850.
DOI: https://doi.org/10.1210/en.2004-1616, PMID: 15947006
Torre D, Lolli F, Ciana P, Maggi A. 2017. Sexual dimorphism and estrogen action in mouse liver. Advances in
Experimental Medicine and Biology 1043:141–151.
DOI: https://doi.org/10.1007/978-3-319-70178-3_8,
PMID: 29224094
Tyshkovskiy A, Bozaykut P, Borodinova AA, Gerashchenko MV, Ables GP, Garratt M, Khaitovich P, Clish CB,
Miller RA, Gladyshev VN. 2019. Identification and application of gene expression signatures associated with
lifespan extension. Cell Metabolism 30:573–593.
DOI: https://doi.org/10.1016/j.cmet.2019.06.018,
PMID: 31353263
Tzanetakou IP, Katsilambros NL, Benetos A, Mikhailidis DP, Perrea DN. 2012. "Is obesity linked to aging?":
adipose tissue and the role of telomeres. Ageing Research Reviews 11:220–229.
DOI: https://doi.org/10.1016/j.
arr.2011.12.003
, PMID: 22186032
Veldhuis JD, Mielke KL, Cosma M, Soares-Welch C, Paulo R, Miles JM, Bowers CY. 2009. Aromatase and 5alpha-
reductase inhibition during an exogenous testosterone clamp unveils selective sex steroid modulation of
somatostatin and growth hormone secretagogue actions in healthy older men. The Journal of Clinical
Endocrinology & Metabolism 94:973–981.
DOI: https://doi.org/10.1210/jc.2008-2108, PMID: 19088159
Vidal O, Lindberg M, Sa
¨
vendahl L, Lubahn DB, Ritzen EM, Gustafsson JA, Ohlsson C. 1999. Disproportional body
growth in female estrogen receptor-alpha-inactivated mice. Biochemical and Biophysical Research
Communications 265:569–571.
DOI: https://doi.org/10.1006/bbrc.1999.1711, PMID: 10558910
Villareal DT, Apovian CM, Kushner RF, Klein S, American Society for Nutrition, NAASO, The Obesity Society.
2005. Obesity in older adults: technical review and position statement of the american society for nutrition and
NAASO, the obesity society. Obesity Research 13:1849–1863.
DOI: https://doi.org/10.1038/oby.2005.228,
PMID: 16339115
Wang Z, Zhang X, Shen P, Loggie BW, Chang Y, Deuel TF. 2005. Identification, cloning, and expression of
human estrogen receptor-alpha36, a novel variant of human estrogen receptor-alpha66. Biochemical and
Biophysical Research Communications 336:1023–1027.
DOI: https://doi.org/10.1016/j.bbrc.2005.08.226,
PMID: 16165085
Waters DL, Ward AL, Villareal DT. 2013. Weight loss in obese adults 65years and older: a review of the
controversy. Experimental Gerontology 48:1054–1061.
DOI: https://doi.org/10.1016/j.exger.2013.02.005,
PMID: 23403042
Wei L, Lai EC, Kao-Yang YH, Walker BR, MacDonald TM, Andrew R. 2019. Incidence of type 2 diabetes mellitus
in men receiving steroid 5alpha-reductase inhibitors: population based cohort study. BMJ 365:l1204.
DOI: https://doi.org/10.1136/bmj.l1204, PMID: 30971393
Whitmer RA, Gunderson EP, Barrett-Connor E, Quesenberry CP, Yaffe K. 2005a. Obesity in middle age and
future risk of dementia: a 27 year longitudinal population based study. BMJ 330:1360.
DOI: https://doi.org/10.
1136/bmj.38446.466238.E0
, PMID: 15863436
Whitmer RA, Sidney S, Selby J, Johnston SC, Yaffe K. 2005b. Midlife cardiovascular risk factors and risk of
dementia in late life. Neurology 64:277–281.
DOI: https://doi.org/10.1212/01.WNL.0000149519.47454.F2,
PMID: 15668425
Xiong Y, Lei QY, Zhao S, Guan KL. 2011. Regulation of glycolysis and gluconeogenesis by acetylation of PKM
and PEPCK. Cold Spring Harbor Symposia on Quantitative Biology 285–289.
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 29 of 30
Research article Medicine
Xu Y, Nedungadi TP, Zhu L, Sobhani N, Irani BG, Davis KE, Zhang X, Zou F, Gent LM, Hahner LD, Khan SA, Elias
CF, Elmquist JK, Clegg DJ. 2011. Distinct hypothalamic neurons mediate estrogenic effects on energy
homeostasis and reproduction. Cell Metabolism 14:453–465.
DOI: https://doi.org/10.1016/j.cmet.2011.08.009,
PMID: 21982706
Yan H, Yang W, Zhou F, Li X, Pan Q, Shen Z, Han G, Newell-Fugate A, Tian Y, Majeti R, Liu W, Xu Y, Wu C,
Allred K, Allred C, Sun Y, Guo S. 2019. Estrogen improves insulin sensitivity and suppresses gluconeogenesis
via the transcription factor Foxo1. Diabetes 68:291–304.
DOI: https://doi.org/10.2337/db18-0638, PMID: 304
87265
Yang H, Youm YH, Vandanmagsar B, Rood J, Kumar KG, Butler AA, Dixit VD. 2009. Obesity accelerates thymic
aging. Blood 114:3803–3812.
DOI: https://doi.org/10.1182/blood-2009-03-213595, PMID: 19721009
Yang L, Miura K, Zhang B, Matsushita H, Yang YM, Liang S, Song J, Roh YS, Seki E. 2017. TRIF differentially
regulates hepatic steatosis and inflammation/Fibrosis in mice. Cellular and Molecular Gastroenterology and
Hepatology 3:469–483.
DOI: https://doi.org/10.1016/j.jcmgh.2016.12.004, PMID: 28462384
Yang L, Fu WL, Zhu Y, Wang XG. 2020. Tb4 suppresses lincRNA-p21-mediated hepatic apoptosis and fibrosis by
inhibiting PI3K-AKT-NF-kB pathway. Gene 758:144946.
DOI: https://doi.org/10.1016/j.gene.2020.144946,
PMID: 32649978
Yu M, Benham A, Logan S, Brush RS, Mandal MN, Anderson RE, Agbaga MP. 2012. ELOVL4 protein
preferentially elongates 20:5n3 to very long chain PUFAs over 20:4n6 and 22:6n3. Journal of Lipid Research 53:
494–504.
DOI: https://doi.org/10.1194/jlr.M021386, PMID: 22158834
Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, Liu
XS. 2008. Model-based analysis of ChIP-Seq (MACS). Genome Biology 9:R137.
DOI: https://doi.org/10.1186/
gb-2008-9-9-r137
, PMID: 18798982
Zhang H, Liu Y, Wang L, Li Z, Zhang H, Wu J, Rahman N, Guo Y, Li D, Li N, Huhtaniemi I, Tsang SY, Gao GF, Li X.
2013. Differential effects of estrogen/androgen on the prevention of nonalcoholic fatty liver disease in the male
rat. Journal of Lipid Research 54:345–357.
DOI: https://doi.org/10.1194/jlr.M028969, PMID: 23175777
Zhang J, Ren J, Wei J, Chong CC, Yang D, He Y, Chen GG, Lai PB. 2016. Alternative splicing of estrogen
receptor alpha in hepatocellular carcinoma. BMC Cancer 16:926.
DOI: https://doi.org/10.1186/s12885-016-
2928-3
, PMID: 27899088
Zhang X, Yang S, Chen J, Su Z. 2018. Unraveling the regulation of hepatic gluconeogenesis. Frontiers in
Endocrinology 9:802.
DOI: https://doi.org/10.3389/fendo.2018.00802, PMID: 30733709
Zhu L, Brown WC, Cai Q, Krust A, Chambon P, McGuinness OP, Stafford JM. 2013. Estrogen treatment after
ovariectomy protects against fatty liver and may improve pathway-selective insulin resistance. Diabetes 62:424–
434.
DOI: https://doi.org/10.2337/db11-1718, PMID: 22966069
Zhu L, Martinez MN, Emfinger CH, Palmisano BT, Stafford JM. 2014. Estrogen signaling prevents diet-induced
hepatic insulin resistance in male mice with obesity. American Journal of Physiology-Endocrinology and
Metabolism 306:E1188–E1197.
DOI: https://doi.org/10.1152/ajpendo.00579.2013, PMID: 24691030
Mann et al. eLife 2020;9:e59616. DOI: https://doi.org/10.7554/eLife.59616 30 of 30
Research article Medicine
