BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

Contents lists available at ScienceDirect

BBA - Molecular and Cell Biology of Lipids

journal homepage: www.elsevier.com/locate/bbalip

Review

Adipose tissue as a target for second-generation (atypical) antipsychotics: A T

molecular view
⁎
Vitor Ferreira, Diana Grajales, Ángela M. Valverde
Instituto de Investigaciones Biomédicas Alberto Sols (CSIC-UAM), Madrid, Spain
CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), ISCIII, Madrid, Spain

A R T I C LE I N FO A B S T R A C T

Keywords: Schizophrenia is a neuropsychiatric disorder that chronically affects 21 million people worldwide. Second-
Adipose tissue generation antipsychotics (SGAs) are the cornerstone in the management of schizophrenia. However, despite
Antipsychotics their efficacy in counteracting both positive and negative symptomatology of schizophrenia, recent clinical
Schizophrenia observations have described an increase in the prevalence of metabolic disturbances in patients treated with
Lipid metabolism
SGAs, including abnormal weight gain, hyperglycemia and dyslipidemia. While the molecular mechanisms re-
Adipocyte differentiation
sponsible for these side-effects remain poorly understood, increasing evidence points to a link between SGAs and
Thermogenesis
Browning adipose tissue depots of white, brown and beige adipocytes. In this review, we survey the present knowledge in
this area, with a particular focus on the molecular aspects of adipocyte biology including differentiation, lipid
metabolism, thermogenic function and the browning/beiging process.

1. Introduction sufficiently, restoring a productive life and allowing the integration of

patients into society [3]. According to current guidelines, antipsychotic
1.1. Etiology and pathophysiology of schizophrenia and its treatment with agents are the first line of treatment in schizophrenia [4,5]. Because
antipsychotic agents discontinuation of treatment is associated with an exponential risk of
relapse as compared with maintenance therapy, patients generally
Schizophrenia is a chronic, severe mental disorder that affects about continue the same treatment that was effective in the acute phase for as
21 million people worldwide. The disease typically manifests in late long as it is well tolerated.
adolescence/early adulthood and usually involves positive symptoms Since the introduction of chlorpromazine in 1952, the first-genera-
that reflect an excess or distortion of normal functions, resulting in tion antipsychotics (FGAs) have changed psychiatric care dramatically,
behavior problems such as delusions, hallucinations, trouble thinking allowing many patients with debilitating and severe mental illnesses
and concentrating; and/or negative symptoms related to withdrawal or (schizophrenia, bipolar mania and acute agitation, among other con-
lack of normal cognitive functions, including apathy, avolition, alogia ditions) to reintegrate into society. FGAs predominantly counteract the
and anhedonia [1,2]. Often, the symptoms are associated with psy- positive symptoms of schizophrenia through mechanisms that remain
chotic episodes that disrupt the stability and quality of life of patients, unknown. The most accepted hypothesis of FGAs action relates to the
denying them a normal life in society. Fortunately, pharmacological dopaminergic theory of schizophrenia, which posits that the positive
interventions are effective in suppressing the symptomatology symptomatology is caused by the increased subcortical release of

Abbreviations: SGAs, second-generation (atypical) antipsychotics; FGAs, first-generation (typical) antipsychotics; D2, dopamine-2; L-DOPA, l-3,4-dihydrox-
yphenylalanine; WAT, white adipose tissue; TG, triglyceride; BAT, brown adipose tissue; UCP-1, uncoupling protein-1; SREBP, sterol regulatory element-binding
protein; FAS, fatty acid synthase; C/EBP, CCAAT/enhancer binding protein; SCD1, stearoyl-CoA desaturase-1; ASC, adipose-derived stem cells; CNS, central nervous
system; Insig, insulin-induced gene; SCAP, SREBP cleavage-activating protein; ER, endoplasmic reticulum; PPAR-γ, peroxisome proliferator-activated receptor
gamma; LPL, lipoprotein lipase; PKC, protein kinase C; MSCs, mesenchymal stem cells; PLIN, perilipin; ATGL, adipose tissue triglyceride lipase; FFA, free fatty acids;
NF-κB, nuclear factor-κB; TNF, tumor necrosis factor; IL, interleukin; AMPK, 5′ adenosine monophosphate-activated protein kinase; HSL, hormone-sensitive lipase;
ACC, acetyl-CoA carboxylase; IBMX, 3-isobutyl-1-methylxanthine; TSPO, translocator protein; GLUT4, glucose transporter-4; BMI, body mass-index; S100B, calcium
binding-protein B; ERK, extracellular signal-regulated kinase; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1 alpha; AP2, adipocyte protein
2
⁎
Corresponding author at: Instituto de Investigaciones Biomédicas Alberto Sols (CSIC-UAM), Madrid, Spain.
E-mail address: [email protected] (Á.M. Valverde).

https://doi.org/10.1016/j.bbalip.2019.158534
Received 22 August 2019; Received in revised form 18 October 2019; Accepted 23 October 2019
Available online 29 October 2019
1388-1981/ © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

dopamine and the subsequent enhanced activation of dopamine-2 (D2) subcutaneous (underneath the skin) depots, with important ontogenetic
receptors [6,7], likely derived from disturbances in the cortical and metabolic differences between the two [16]. Under a state of po-
pathway through the nucleus accumbens. Conversely, the negative sitive energy balance, adipose tissue expands via hypertrophy of ex-
symptoms appear to be caused by blunted dopaminergic signaling isting adipose cells and via hyperplasia, with de novo formation of
through reduced dopamine-1 receptor activity in the prefrontal cortex adipocytes. Beyond its classical role as an energy storage and release
and diminished nucleus caudatus activity [7–9]. Alterations in the ex- unit, WAT also functions to protect other organs (liver and muscle)
pression and activity of dopamine-3 receptors have also been associated from lipid-associated toxicity (lipotoxicity) [17] and is a key player in
with the negative symptoms of schizophrenia [10]. Some studies have endocrine signaling [18].
corroborated the dopaminergic theory of schizophrenia, showing that Brown adipose tissue (BAT) is much less abundant than WAT and is
D2 antagonists such as FGAs act on different dopaminergic pathways located mainly in subcutaneous interscapular regions. BAT is char-
(mesolimbic, mesocortical, nigrostriatal and tuberoinfundibular) to acterized by the presence of small multiocular adipocytes that were
control schizophrenia symptomatology. By contrast, D2 agonists in- initially perceived as skeletal muscle-like lineage cells, but more re-
cluding l-3,4-dihydroxyphenylalanine (L-DOPA), cocaine and amphe- cently their adipocyte origin has been suggested [19,20]. BAT is
tamines, trigger psychomimetic effects in individuals that are not abundant in mitochondria enriched in uncoupled protein-1 (UCP-1) in
schizophrenic per se. Nevertheless, FGAs are associated with extra- the inner membrane, which functions to dissipate chemical energy in
pyramidal side-effects such as dyskinesia and dystonia [11]. In an at- the form of heat by uncoupling fuel oxidation from ATP synthesis
tempt to counteract this, a variety of new agents were investigated in [21–23]. In humans, BAT was previously believed to exist only in
the 1990s, leading to the approval of second-generation (atypical) an- newborn infants, but it has recently been identified in adults in the
tipsychotics (SGAs), which are now the mainstay for patients with lower neck area [24–26], where its functions and characteristics are
schizophrenia and other psychotic disorders [4]. Beyond the interaction currently under extensive study.
of SGAs with dopaminergic receptors, as for FGAs, they also have the Beige adipocytes are mainly present in subcutaneous white depots
ability to block serotonergic (5-HT2A and 5-HT2C) receptors, with a and in restricted amounts of visceral WAT [27]. Beige adipocytes are
higher affinity than that for D2 receptors [8,11], suggesting that al- considered brown-like thermogenic adipose cells with a similar multi-
terations in serotoninergic pathways could also play a role in schizo- locular lipid droplet morphology and UCP-1 expression, and are de-
phrenia development. More importantly, the neurological side-effects veloped from the so-called beiging or browning of WAT. This differ-
associated with FGAs are not as evident in patients on SGAs, which can entiation phenomenon is induced by cold stress or agonists that can
counteract both positive and negative symptoms of the disease [4]. mimic this effect, such as β3-adrenergic agonists [28,29]. The origin of
However, clinical observations have revealed a variety of discrete me- beige adipocytes is, however, not completely understood and several
tabolic dysfunctions in a relevant proportion of patients on SGAs, such processes seem to be involved; for instance, they can be derived from a
as abnormal body weight gain, hyperglycemia and dyslipidemia beige progenitor lineage [19,27,30], but can also transdifferentiate
[12–14]. These side-effects (Table 1) suggest that the use of SGAs has from mature white adipocytes [31,32], or even differentiate from other
indirect effects on different peripheral tissues and systems, including fat origins [16].
depots, liver and immune cells. As an endocrine organ, adipose tissue actively participates in in-
flammatory processes by producing and secreting a wide variety of
bioactive peptides including cytokines and the so-called adipokines
1.2. Adipose tissue: a hub for energy balance and endocrine signaling
such as leptin and adiponectin. These peptides have both local and
distant actions related to the modulation of lipid and glucose metabo-
Abnormal body weight gain is a major side-effect of therapy with
lism and energy balance [16].
SGAs [11] and is associated with an increase in fat (adipose) depots
Given the evident metabolic dysregulation in some patients on
[3,14,15]. Adipose tissue is a highly specialized organ regulating en-
SGAs, a better understanding of the effects of these drugs on adipocytes
ergy homeostasis and metabolism, and is comprised mainly of three
will be important for elucidating the molecular mechanisms underlying
general classes of adipocytes in mammals – white, brown and beige
these dysfunctions in patients. Accordingly, this review will examine
adipose cells – which have distinct developmental origins, morpholo-
recent advances in our understanding of the impact of SGAs on the
gies, lipid droplet distribution, mitochondrial networks and gene ex-
mechanisms of adipocyte differentiation and function.
pression patterns [16].
Functionally, white adipose tissue (WAT) is the predominant store
of surplus energy in the body, in the form of triglycerides (TGs), and 2. Effect of second-generation antipsychotics on white adipose
contains adipocytes with a large unilocular lipid droplet and few mi- tissue
tochondria. WAT accounts for 5–50% of the total body weight in hu-
mans and includes both visceral (within the abdominal cavity) and WAT is organized into discrete depots – mainly visceral and

Table 1
List of first generation and second-generation antipsychotics mentioned in this review and their side-effect profiles.
Antipsychotics mentioned in this Review Extrapyramidal side effects in Humans Metabolic side effects in Humans

Sedation Cognitive impairment Tardive dyskinesia Diabetes Increased Lipids/dislipidemia Hyperprolactinemia

Chlorpromazine ++ ++ ++ +++ +++ +

Haloperidol + 0 ++ 0/+ 0/+ ++
Pimozide 0/+ + +++ 0/+ 0/+ ++
Clozapine +++ +++ 0 +++ ++ +
Olanzapine +/++ ++ 0/+ +++ +++ +
Aripiprazole 0/+ 0 0/+ 0/+ 0/+ 0
Quetiapine ++ +/++ 0/+ ++ ++ 0
Risperidone + 0 0/+ + + +++
Ziprasidone + 0 0/+ 0/+ 0/+ +
Blonanserin 0/+ 0 + 0/+ + +

0: none; 0/+ sporadic; +: occasionally; ++: recurrent; +++ very often.

2
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

Fig. 1. Impact of SGAs in white adipocytes. A. Scheme of the molecular effects of SGAs in white adipocytes focusing on lipid-related proteins and enlargement of the
lipid droplets. B. List of the receptors through which SGAs can impact white adipocytes.

subcutaneous – that are differentially associated with insulin resistance gain and lipid abnormalities [41,42], SGAs function in a cell-autono-
and the risk of metabolic syndrome. While the accumulation of visceral mous manner by increasing lipogenesis.
WAT is considered deleterious, due to its promotion of inflammation, The insulin-induced gene (Insig) family proteins are important ne-
subcutaneous WAT seems to be protective against the development of gative regulators of SREBP function and lipogenesis. Insigs are SREBP-
metabolic diseases in mice and humans [33,34]. chaperone proteins that form a complex with SREBP and SREBP clea-
It has been described that patients with schizophrenia have sig- vage-activating protein (SCAP) in the endoplasmic reticulum (ER).
nificantly more visceral WAT than healthy individuals [15], and this When sterol levels are low, SREBPs are activated by proteolytic clea-
could explain their susceptibility to develop insulin resistance and more vage and the N-terminal domain is released from the ER and translo-
severe metabolic dysfunctions such as type 2 diabetes mellitus [35]. As cates to the nucleus where it acts as a transcription factor for multiple
mentioned, SGAs therapy induces an increase of both subcutaneous and lipid biosynthesis genes [43]. The Insig family is composed of two
visceral fat in some patients [15], and also alters the mechanisms of isoforms: Insig-1, a target of nuclear SREBPs whose mRNA levels are
differentiation and response to physiological stimulus, as described directly associated with their presence in the nucleus, and Insig-2,
below. which is negatively regulated by insulin in a SREBP-independent
manner [36]. In sterol-replete conditions, Insig-2 can retain the SCAP/
2.1. Effect of second-generation antipsychotics on the differentiation of SREBP complex in the ER [44]. Of interest, a relationship between
white adipocytes SREBP-induced lipid biosynthesis, Insig-2 blockade and SGA-associated
weight gain was described by Chen et al., showing that clozapine
A possible explanation for weight gain associated with SGAs is their (10 μM) significantly suppresses Insig-2 expression during adipogenesis
positive influence on the differentiation of preadipocytes to mature of human ASCs, at both early and induction (late) phases. Moreover,
adipocytes. In relation to adipocyte differentiation, the sterol regulatory after treatment, the levels of Insig-2 negatively correlated with the
element-binding protein (SREBP) family of transcription factors, con- expression of SREBP-1 and lipid-biosynthesis genes. Of note, the au-
stituted by the SREBP-1a and -1c isoforms (encoded by the SREBF1 thors found that on the third day of the differentiation, all of the SGAs
gene) and SREBP-2 (encoded by SREBF2) [36], play a major regulatory tested – clozapine (10 μM), olanzapine (1 μM) and risperidone (0.4 μM)
role. Yang et al. showed that treatment of differentiating murine 3T3-L1 – decreased Insig-2 expression; however, only clozapine significantly
preadipocytes with the SGA olanzapine (10 or 50 μM for 24 h) increases suppressed Insig-2 at day 7 of differentiation and, consequently, SREBP-
TG accumulation and activates SREBP-1, accompanied by the over- 1 activity was increased [38]. Moreover, even though olanzapine
expression of fatty acid synthase (FAS), an enzymatic complex of long- treatment did not maintain the suppression of Insig-2 expression on the
chain fatty acid synthesis [36]. They also found that SREBP-1, but not last day of differentiation, it increased SREBP-1 expression. Conversely,
the transcription factor CCAAT/enhancer binding protein α (C/EBP-α), Insig-2 overexpression during human ASC differentiation results in a
is overexpressed and activated in olanzapine-stimulated 3T3-L1 cells. suppression of SREBP-1 expression, leading to the inhibition of SGA-
Similarly, olanzapine (100 μM for 3 h) increases lipogenesis and reduces induced lipid biosynthesis [38]. This study implicates Insig-2 in the
lipolysis [37]. Likewise, clozapine (10 μM) induces stearoyl-CoA desa- pathogenesis of the metabolic dysfunctions developed by patients under
turase-1 (SCD1) and SREBP-1 expression levels at both early (day 3) SGAs treatment. Despite the fact that the interactions between the two
and late (day 7) stages of human adipose-derived stem cells (ASC) Insig isoforms are related to SGA-induced metabolic syndrome, Insig-1
differentiation [38]. Overall, these studies show that some SGAs in- is reported not to correlate with these metabolic side effects per se [45]
crease the expression of SREBP-1 and its downstream lipogenic targets, (Fig. 1A).
although to different levels. Indeed, it has been suggested that the ef- In mammalian cells, peroxisome proliferator-activated receptor
fects of SGAs on lipid droplet formation (clozapine > olanzapine > gamma (PPAR-γ) is critical not only for adipogenic differentiation, but
risperidone) depend on the nuclear translocation of SREBP-1 and the also for the maintenance of the mature adipocyte phenotype [46,47]
subsequent modulation of adipogenesis [38]. Moreover, in the human and, together with C/EBPs, is considered a master regulator of adipo-
hepatic HepG2 cell line, treatment with olanzapine (10 and 50 μM for genesis [48]. Sertie et al. reported the impact of SGAs, particularly
24 h) increases SREBP-1 response element activity in a dose-dependent clozapine (used at 20–30 μM) and olanzapine (40–100 μM), in the en-
manner [39]. However, these effects are not so evident in other hepatic hancement of both early (day 3) C/EBP-β and PPAR-γ expression, and
cell types treated with olanzapine, haloperidol or mirtazapine (25 μM late (day 14) PPAR-γ and lipoprotein lipase (LPL) levels during adipo-
for 24 h), as reported by Raeder and co-workers [40]. Given the genic differentiation of human ASCs [49]. Consistent with this,
aforementioned results in adipocyte culture, we speculate that in ad- Hemmrich et al. differentiated human ASCs in the presence of clozapine
dition to acting on the central nervous system (CNS) leading to weight (5 and 50 μM) for the first 5 days and then in its absence for 9 additional

3
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

days (14 days of differentiation in total), finding increased activity of 2.2. Second-generation antipsychotics modify the endocrine function of
glycerol 3-phophatase dehydrogenase, another key marker of adipo- white adipocytes
genic differentiation. They also found that clozapine increases the
percentage of differentiated cells when compared with untreated con- Adipose tissue is a main player in systemic metabolism and in-
trols [50]. These results provide a plausible explanation as to why flammation, or so-called immunometabolism [16,58], and produces
clozapine treatment is associated with a higher risk of weight gain in and releases a plethora of biomolecules with paracrine, autocrine and
relation to other SGAs [51,52]. Likewise, other agents, particularly endocrine functions associated with the balance between energy ex-
pimozide (10 μM), is reported to promote adipogenesis of 3T3-L1 pre- penditure and intake, regulation of lipid and glucose homeostasis, and
adipocytes up to day 6 of differentiation by inhibiting fatty acid binding also inflammatory processes [16].
protein 4 and upregulating PPAR-γ protein levels of the cells [53]. Si- Adiponectin is the most abundant of the adipokines and is also one
milarly, treatment of human ASCs for 3 days with clozapine, olanza- of the most relevant molecules secreted by mature adipocytes, and it is
pine, quetiapine or risperidone (all used at 250 nM) increases total lipid negatively regulated in obesity, insulin resistance and metabolic syn-
production, an effect blocked by the addition of a protein kinase C drome [59]. Adiponectin modulates pathways related to carbohydrate
(PKC)-β inhibitor [51]. At the molecular level, this study shows that and lipid metabolism, and also vascular processes, due to its anti-in-
treatment of ASCs with SGAs, particularly clozapine, promotes the flammatory, anti-atherogenic and insulin sensitizing properties [60].
translocation of PKC-β to the plasma membrane and its activation. Several studies have investigated the adiponectin response to SGAs.
Importantly, adipogenesis of skeletal muscle-derived stem cells is also Human ASCs differentiated in the presence of clozapine (100 ng/ml),
boosted by SGAs, suggesting that the weight gain associated with SGAs quetiapine (50 ng/ml) or aripiprazole (100 ng/ml) show elevated adi-
also involves the commitment of adipogenic features of other stem cell ponectin expression [61] and, similarly, olanzapine (10 μM) treatment
pools [54]. induces early adiponectin expression in differentiating 3T3-L1 cells
The increase in body weight associated with SGAs treatment can [39]. In another study, short (2 days in mature adipocytes) or long
also result from alterations in adipocyte lipid storage (hypertrophy). (10 days, including during differentiation) clozapine (10–30 μM) treat-
For instance, differentiation of rat ASCs in the presence of clozapine ment in 3T3-L1 adipocytes was found to decrease adiponectin secretion
(10 μM) increases the formation of lipid droplets, whereas the addition without altering its mRNA levels [62], and 10 days exposure to blo-
of olanzapine (1 μM) or risperidone (0.4 μM) induces only a slight en- nanserin (0.01–0.1 μM) also reduced adiponectin mRNA levels. Thus,
hancement at day 7 [38]. Likewise, in adipocytes differentiated from this opposite effect on adiponectin expression/secretion is intriguing
human mesenchymal stem cells (MSCs), olanzapine treatment and may be dependent on the adipocyte origin or the sensitivity of the
(5–100 μM) up to 12 days increases lipid accumulation in a dose-de- cell-based systems to SGAs.
pendent manner [52]. Further proteomic studies revealed that olanza- In contrast to adiponectin, leptin is a secreted adipokine that is
pine upregulates the levels of perilipin-4 (PLIN4) and other enzymes positively related to the amount of body fat, and modulates food intake
related to lipid metabolism including FAS. Indeed, MSCs differentiated and energy expenditure [63]; accordingly, leptin is typically elevated in
in the presence of olanzapine show enlarged lipid droplets coated with the serum of obese patients [64]. Tsubai et al. reported that exposure of
PLIN1, 2 and 4. However, only PLIN2 protein levels are upregulated by 3T3-L1 adipocytes to clozapine (10–30 μM) for 2 or 10 days decreases
olanzapine at early stages of differentiation and decrease from day 12, leptin mRNA levels and also its secretion into the culture medium [62].
whereas PLIN4 and PLIN1 increase at later stages. While it is known Of interest, this work tested the role of serotonergic 5-HT2c and hista-
that PLIN proteins are regulated by PPAR-γ, the authors failed to find minergic H1 receptors on the impact of clozapine in leptin expression
alterations in its mRNA levels by olanzapine. However, C/EBP-α mRNA and secretion, as they are ubiquitously expressed in the brain and
was decreased at day 9–16 in the presence of olanzapine, which could peripheral tissues [65,66], and play a role in alterations/disorders of
downregulate its target protein fatty acid translocase/CD36 in late eating patterns that lead to dysregulated lipid metabolism [67–69]. The
stages of differentiation [55]. Nonetheless, the relevance of this authors used histamine (H1 receptor agonist) and diphenhydramine (H1
downregulation is controversial as the deletion of fatty acid translo- receptor antagonist) as well as serotonin (5-HT2c receptor agonist) and
case/CD36 in mice was found to confer protection from high-fat diet- SB242084 (5-HT2c receptor antagonist). Short-term treatment of 3T3-
induced adipose tissue deposition [56]. Of note, other genes encoding L1 cells with histamine or diphenhydramine failed to modify leptin
relevant proteins of lipid metabolism such as SREBP-1c, FAS, LPL, secretion, suggesting that this process is not mediated through H1 re-
leptin and adiponectin, were unaltered by the treatment. The afore- ceptors. However, treatment with serotonin decreased leptin secretion
mentioned study also showed that the expression of adipose tissue tri- when compared to control-treated cells, but no synergistic or additive
glyceride lipase (ATGL) is increased by olanzapine. In adipocytes, ATGL effect was found when it was combined with clozapine (30 μM). In-
is localized on large PLIN1-positive lipid droplets and is liberated upon terestingly, SB242084 also decreased leptin secretion in 3T3-L1 cells
PLIN1 phosphorylation, leading the authors to postulate that ATGL and further decreased secretion when combined with clozapine (30 and
expression increases in olanzapine-treated MSCs concomitant with the 50 μM) in a dose-dependent manner. Nevertheless, this relationship
accumulation of lipid droplets (Fig. 1, panel A). Moreover, changes in seems to be more complex because serotonin failed to reverse this effect
gene expression and protein levels of PLIN family members by olan- by increasing leptin secretion (Fig. 1, panel B) [62]. By contrast, in the
zapine might suggest a possible downstream mechanism for the in- study performed in human ASCs, treatment with SGAs increased leptin
creased adiposity in patients undergoing treatment with this drug [55]. expression [61]. Thus, more studies are required to fully understand the
Dyslipidemia is another metabolic side-effect of SGAs therapy and is mechanisms governing the interactions between SGAs and serotonergic
directly related to the ability of adipocytes to sequester free fatty acids receptors in adipocyte hypertrophy and adipokine secretion.
(FFA). In this regard, a study with human adipocytes showed that Adipocytes secrete cytokines with pro-inflammatory properties,
clozapine, olanzapine, quetiapine or risperidone (all tested at 50 or which are the main drivers of the chronic low-grade inflammation as-
100 μM) failed to inhibit fatty acid transport, concluding that this is sociated with obesity-related metabolic abnormalities [70]. In the
likely not the mechanism for dyslipidemia observed in patients treated aforementioned study of Sarvari et al. [58], in vitro treatment of human
with SGAs [57]. Despite all these achievements, further studies are ASCs with olanzapine, ziprasidone, clozapine, quetiapine, aripiprazole,
required for a better understanding of the impact of SGAs on the early risperidone or haloperidol induced the expression of the transcription
and late events of adipogenesis, as well as on the hypertrophy of the factor nuclear factor-κB (NF-κB), a key component of the inflammatory
adipose cells. response, and this was accompanied by an increase in the expression of
the pro-inflammatory cytokines tumor necrosis factor (TNF)-α, inter-
leukin (IL)-1β, IL-8 and MCP-1 and the release of IL-8 and MCP-1 into

4
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

the culture medium. These results suggest that antipsychotic treatments described that female Sprague-Dawley rats treated with olanzapine
might “prime” patients for developing a low-grade chronic pro-in- (2 mg/kg twice a day by oral gavage) for 2 weeks show increases in
flammatory state. Indeed, exposure of 3T3-L1 adipocytes to clozapine total body weight gain, which is mainly due to an increase in liver and
(30 μM) for 10 days increases MCP-1 and IL-6 expression [62]. This WAT weight [85]. Moreover, the animals develop hyperlipidemia, hy-
increase in the production of pro-inflammatory molecules in turn pro- perglycemia, hyperinsulinemia, insulin resistance and present elevated
motes insulin resistance and exacerbates metabolic dysfunction serum IL-6 levels together with tissue chromium depletion. Interest-
[71,72]. Of note, adiponectin can block local pro-inflammatory cyto- ingly, daily supplementation of chromium during olanzapine treatment
kine production by antagonizing toll-like receptors (TLRs) [73] and produces a milder phenotype and supplementation with AICAR, an 5′
inhibiting NF-κB [74,75]. Thus, the hypoadiponectinemia associated adenosine monophosphate-activated protein kinase (AMPK) activator,
with hyperleptinemia following SGAs treatments is accompanied by the ameliorates olanzapine-induced hyperglycemia and hyperlipidemia,
increased expression and secretion of pro-inflammatory cytokines suggesting that low chromium and AMPK activity are related to olan-
[61,62]. The altered pattern of pro-inflammatory cytokine secretion in zapine-induced metabolic dysfunction [85]. Interestingly, female rats
adipocytes and the “shotgun” affinity of SGAs for several receptors in- are reported to be more susceptible than male rats to the metabolic
cluding histaminergic H1, serotonergic 5-HT2c, and adrenergic re- damaging effects of SGAs [81,86]. Albaugh et al. reported that an in-
ceptors [35], is likely the cause of the development of insulin resistance creasing dosing regimen of olanzapine mixed in cookie dough (4 mg/kg
and type 2 diabetes mellitus associated to the treatment with theses from day 0–6; 8 mg/kg from day 7–20 and 12 mg/kg from day 21–29)
drugs. increases the body weight of female, but not male, Wistar and Sprague-
Dawley rats [81]. Comparable results were previously reported by
2.3. Second-generation antipsychotics modulate insulin sensitivity and Pouzet and co-workers in female Wistar rats treated with 5 and 20 mg/
glucose uptake in white adipocytes kg of olanzapine via oral administration for 21 days [86]. Furthermore,
Goudie et al. showed that female Wistar rats injected intraperitoneally
In addition to the control of whole body energetic balance, adipose with olanzapine (4 mg/kg) for 19 days increased their body weight
tissue also regulates glucose and lipid metabolism [16,58,76]. Glucose [83]. Of note, it has been described that the weight gain is very fast at
transport resulting from activation of insulin signaling is highly re- the beginning of the treatment and is reversible once treatment is ended
levant because it provides both fatty acids and glycerol for TG synthesis [83,86]. In this respect, Albaugh et al. provided evidence for the as-
[77,78]. Vestri et al. compared the effects of clozapine, olanzapine, sociation between body weight gain and hyperphagia in female rats
risperidone and quetiapine (1–500 μM) with the FGAs butyrophenone receiving olanzapine self-administered via cookie dough, starting from
and trifluoperazine for insulin-induced glucose transport and lipogen- the first 24 h of treatment [81]. In the same study, olanzapine con-
esis/lipolysis in 3T3-L1 cells and rat primary adipocytes. In both sys- sumption led to an increase in leptin levels and adiposity, and induced
tems, olanzapine and clozapine at concentrations as low as 5 μM mild insulin resistance from day 12 of treatment, suggesting that
strongly reduced insulin-induced glucose transport, whereas the FGAs olanzapine-induced weight gain could be a secondary effect of hyper-
failed to modify this response. Moreover, all of the antipsychotics phagia related to leptin resistance (hyperleptinemia). By contrast, in a
(tested at 100 μM) increased basal and insulin-induced glucose oxida- study performed in female Sprague-Dawley rats injected in-
tion rates [37]. Interestingly, another study showed that 3T3-L1 cells tramuscularly with olanzapine (100 mg/kg) for 14 days in a 4-injection
treated with haloperidol, quetiapine or clozapine at 10 μM, but not treatment, Horska et al. reported body weight gain together with a
1 μM, present reduced glucose uptake without alterations in insulin significant increase in the amount of visceral white fat, but without
sensitivity or in Akt/protein kinase B activation [79]. In the same line, hyperphagia [87]. Furthermore, in addition to increases in both adip-
Robinson et al. found that therapeutic concentrations of olanzapine osity and adipocyte size, Tan et al. reported that olanzapine treatment
(7–350 nM, see Table 2) failed to alter basal and insulin-induced glu- (ranging from 0.003 mg/ml to 0.03 mg/ml) in the drinking water
cose transport in 3T3-L1 cells [80]. These results suggest that the effects causes morphological alterations in subcutaneous WAT of female
of SGAs on glucose transport are dose-dependent. Moreover, FGAs did Wistar rats, an effect related to an increase in the number of un-
not affect lipolysis in response to isoproterenol or glucose uptake in differentiated adipose cells in this depot that was detected as early as
response to insulin, which might explain why conventional therapies the third day of treatment in a dose- and time-dependent manner, but
are less associated with secondary metabolic side effects [37]. independently of the body weight gain [87,88].
In contrast to what is found in female animals treated with SGAs,
the body weight increase is more difficult to mimic in males. Minet-
2.4. White adipose tissue and antipsychotic effects in animal models:
Ringuet et al. showed that male Sprague-Dawley rats fed with an SGAs-
dysregulation of lipid metabolism
supplemented diet for 5 weeks (1 mg/kg haloperidol or olanzapine and
10 mg/kg ziprasidone) did not gain body weight but did show increased
Animal models provide a translational platform to decipher the
adiposity in subcutaneous and visceral adipose depots [86]. A deeper
molecular basis of the metabolic side effects associated with SGAs.
analysis of adipocytes isolated from these treated animals showed no
Several studies have reported that clozapine and olanzapine are re-
alterations in basal lipolytic activity; however, olanzapine reduced the
sponsible for the higher weight gain associated with deficits in glucose
lipolytic effects triggered by the β-adrenergic agonist isoprenaline or
and lipid metabolism in rodents [81–84]. For example, Yang et al.
the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX).
Moreover, the anti-lipolytic actions of insulin, the adenosine analogue
Table 2
phenylisopropyladenosine, or the α2-adrenergic agonist UK14304,
Therapeutic reference ranges of different SGAs in accordance with the steady-
state plasma concentrations in patients under treatment.
were unaltered in olanzapine-treated primary adipocytes. Overall, the
results of this study point to a reduction of β-adrenergic receptor sen-
Second-generation Therapeutic reference Therapeutic reference sitivity by olanzapine, which might explain its negative impact on lipid
antipsychotics range (ng/ml) range (nM)
metabolism. However, the fact that IBMX-induced lipolysis was also
Clozapine 350–500 1073–1533 impaired suggests that the molecular mechanisms might be more
Olanzapine 20–40 64–128 complex than the blockade of β-adrenergic receptors by SGAs [89]. On
Aripiprazole 150–210 335–468 the other hand, olanzapine, haloperidol and ziprasidone are moderate
Quetiapine 50–500 130–1304
α1 and α2-adrenergic receptor antagonists [90], but since α1-adre-
Risperidone 20–60 49–146
Ziprasidone 50–130 121–315 nergic receptors are not involved in lipolysis in rats [91], it is unlikely
that they are implicated in the mechanism of action of these drugs.

5
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

Moreover, in the study by Minet-Ringuet et al., the fact that SGAs also play a relevant role in appetite and food intake patterns, since
treatment failed to counteract the inhibition of lipolysis induced by α2- olanzapine might activate the melanin-concentrating hormone system
adrenergic receptor agonists [89] discounts these receptors in the me- (feeding-initiation system) in the lateral hypothalamus that has pro-
chanism by which SGAs disrupt lipid mobilization. jections into the nucleus accumbens [42] enhancing food intake. Re-
Lipid homeostasis is the balance between lipogenesis and lipolysis. garding the CNS-mediated effects of SGAs that result in a pro-in-
Interestingly, male Sprague-Dawley animals treated with olanzapine flammatory profile, Guesdon et al. described an effect of olanzapine
present an increased expression of FAS and a decreased expression of (1 mg/kg) administration in male Wistar rats for 13 days via implanted
hormone-sensitive lipase (HSL) in adipocytes, thus favoring lipid mini-pumps by moderately increasing the mRNA levels of melanin-
synthesis over lipolysis, and providing a possible explanation for the concentrating hormone receptor in the nucleus accumbens shell [41].
increase in adipose depots in rats [89]. By contrast, Victoriano et al. Over time, this could lead to increased food intake and, consequently,
reported that male Sprague-Dawley rats treated with olanzapine (2 mg/ in adipose tissue deposition that can per se recruit immune cells to the
kg) or haloperidol (1 mg/kg) mixed in the food for 46 days showed no fat depots, thereby triggering inflammatory processes. However, as
changes in the levels of lipogenic markers (FAS, and acetyl-CoA car- mentioned above, the link between peripheral and central effects fol-
boxylase (ACC)) and lipolytic enzymes (LPL and HSL) in WAT [92]. lowing SGAs treatment remains poorly understood and needs to be
Clearly, more preclinical studies are needed to fully understand the investigated in future studies.
molecular basis of the impact of SGAs in lipid metabolism, including
gender intrinsic susceptibility, to be translated to humans. 2.4.2. In vivo insulin-related disturbances in WAT by second-generation
antipsychotics
2.4.1. Effect of second-generation antipsychotics on hormone/cytokine Several animal studies have reported a direct influence of SGAs in
expression and secretion in WAT the response of WAT to insulin. In this line, Cui et al. reported that the
As stated earlier, adipose tissue is now recognized as a main player increased body weight in female C57Bl/6 female mice treated with
in the development of systemic inflammation and, consequently, of olanzapine (3 mg/kg/day) mixed in the food for 2, 4 or 8 weeks as-
insulin resistance and metabolic dysfunction [93,94]. While the me- sociates with hyperinsulinemia and insulin resistance [99]. These
chanisms through which SGAs influence systemic and/or local in- findings were corroborated by Coccurello et al. in the same animal
flammation remain unclear, it has been hypothesized that adipose model [100], and also by Hou et al. in female C57Bl/6 mice receiving
tissue is responsible for the inflammatory response induced by SGAs, olanzapine (6 mg/kg) via oral gavage for 7 weeks [101]. Calevro et al.
which in turn is the main cause of the metabolic dysfunctions associated hypothesize that the insulin resistance in treated animals is closely re-
with antipsychotic therapies [93]. By treating female Balb/c mice and lated to inflammation since olanzapine increases macrophage infiltra-
Sprague-Dawley rats intraperitoneally with olanzapine (10 mg/kg) for tion and pro-inflammatory cytokine expression in WAT, particularly IL-
8 weeks, Li and co-workers found an increase in the levels of circulating 6 [93]. Also, the effect of SGAs on the Wnt signaling pathway seems to
pro-inflammatory cytokines TNF-α, IL-6, IL-8 and IL-1β, which corre- be relevant for the alterations of glucose metabolism and insulin sen-
lated with their elevated mRNA expression in WAT [95]. Similarly, sitivity in adipose tissue. In this regard, Li and coworkers found that, in
administration of olanzapine (10 mg/kg) via osmotic mini-pumps in addition to an increase in insulin levels during fasting, the expression of
male Sprague-Dawley rats increases the levels of IL-6 and F4/80 im- TFC7L2, a key effector of the Wnt pathway, was increased in the WAT,
munostaining in WAT samples, which positively correlated with the liver and skeletal muscle of male C57Bl/6 male treated with olanzapine
levels of translocator protein (TSPO) – a target for radiotracers puta- (4 mg/kg/day) via oral gavage for 8 weeks. The addition of metformin,
tively indicating microgliosis in clinical neuroimaging studies – an anti-hyperglycemic drug, effectively blocked the changes in insulin
whereas no changes were found in the abundance of IL-1β, IL-4, IL-5, plasma levels and TFC7L2 expression, suggesting a potential correlation
IFN-γ or TNF-α [93]. Using oral administration of olanzapine (1 mg/kg between olanzapine-induced insulin dysfunctions and TCF7L2 [102].
olanzapine, 3 times daily) in female Sprague-Dawley rats, Zhang et al. Also, at the molecular level, the final stage of the insulin-signaling
reported an enhancement in monocyte infiltration in WAT in parallel cascade in adipose tissue is the translocation of glucose transporter-4
with body weight gain and an increase in adipocytes size [96]. More- (GLUT4) to the plasma membrane. However, hyperinsulinemic-eu-
over, the authors found a high correlation between adipocytes size and glycemic clamp studies performed in male Sprague-Dawley rats treated
macrophage infiltration in WAT, and this was accompanied by an up- with olanzapine mixed in cookie dough with an increasing dose re-
regulation of TNF-α, IL-1β and IL-6. Interestingly, the inflammatory gimen (day 0–6: 4 mg/kg/day; day 7–13: 8 mg/kg/day and day 14–29:
response in the WAT of SGAs-treated animals also might have a gender- 12 mg/kg/day) failed to find alterations in adipose tissue glucose up-
dependent component. For example, Davey et al. tested the effects of take although glucose uptake was impaired in skeletal muscle [103].
olanzapine (2 or 4 mg/kg) administered intraperitoneally twice daily in Besides the general direct relationship of insulin resistance and
male and female Sprague-Dawley rats, finding that female rats gained SGAs, a study in obese male C57Bl/6 mice found that olanzapine
more body weight than males, an effect associated with hyperphagia treatment (2 mg/kg/day) via oral gavage, once daily for 4 weeks, in-
[97]. Also, female, but not male, animals showed an up-regulation of IL- creased insulin sensitivity by lowering glucose and insulin plasma le-
6 in WAT and elevated plasma levels of IL-8 and IL-1β. However, vels, an effect related to autophagy by potentiation of lysosomal func-
adiposity and macrophage infiltration was increased by olanzapine tion in adipocytes [104]. Thus, further research will provide new
treatment in both genders, together with alterations in the gut micro- insights to fully understand the effect of these therapies in vivo in order
biota [97]. The aforementioned study of Victoriano et al. performed in to unravel their clinical relevance.
male Sprague-Dawley rats treated with olanzapine or haloperidol mixed
in the food also described WAT inflammation that manifested as ele- 2.5. Direct and indirect effects of second-generation antipsychotics in
vated TNF-α mRNA levels and infiltration of CD68-positive cells [92]. human WAT
Since SGAs treatment leads to NF-κB overexpression in human ASCs
[61], and the relationship between inflammation and insulin resistance Human studies are indispensable to translate data on the mechan-
is well documented [98], a shift to a more pro-inflammatory profile isms of action of SGAs identified in cellular or animal models, since the
might explain some of the metabolic dysfunctions described in adipose latter do not fully recapitulate the disease in patients. Regarding the
tissue after SGAs treatment. Accordingly, NF-κB could be a potential effect of SGAs in adipokine expression and insulin resistance, the levels
target to prevent and/or mitigate olanzapine-induced insulin resistance of the pro-inflammatory adipokine resistin are known to be elevated in
due to WAT inflammation [95]. patients with schizophrenia under stable therapy with clozapine
It is important to mention that the action of SGAs in the CNS might [105,106], and correlate with circulating IL-1Ra, TNF-α and C-reactive

6
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

protein [107]. In line with these results, Sapra and co-workers com- ghrelin, with no correlation to changes in adiposity [120]. Overall,
pared body weight and several inflammatory indicators between a these studies illustrate the direct effect of SGAs on the organism in-
group of 8 non-diabetic men with schizophrenia under treatment for at dependently of alterations in body weight or adiposity [117].
least 6 months with SGAs (independently of the chosen agent) and age- Because schizophrenia can manifest in early adulthood, some stu-
and body mass-index (BMI)-matched healthy men, finding that adipo- dies have focused on younger cohorts of patients. A study on SGAs-
nectin plasma levels were lower and C-reactive protein levels were naive 6–18-year-old patients diagnosed with disruptive behavior dis-
higher in the former after an overnight fast, which associated with in- orders and treated randomly with aripiprazole, olanzapine or risper-
creased insulin resistance [108]. Likewise, in an open-label prospective idone (commonly used in children) for 12 weeks showed that olanza-
single-center study with 113 patients treated either with risperidone pine leads to a higher weight gain (4.12%) when compared with
(54 patients with an average dosage of 4.4 mg/day) or olanzapine (59 aripiprazole (1.66%) or risperidone (1.18%), in association with an
patients with an average dose of 17.4 mg/day), body weight gain and increase in subcutaneous fat [119]. Also, patients treated with olanza-
the prevalence of metabolic syndrome were significantly greater in the pine or aripiprazole show a reduction in insulin-stimulated glucose
olanzapine group than in the risperidone group. Also, whereas adipo- uptake (29.34% and 30.26%, respectively), indicating insulin re-
nectin levels significantly increased in the risperidone group over time, sistance. The increase in body weight and reduced insulin sensitivity in
they significantly decreased in olanzapine-treated patients. By contrast, the first 12 weeks of treatment might be responsible for the future
no significant differences were found between the groups for fasting cardiometabolic morbidity and mortality associated with SGAs therapy
glucose, insulin levels and insulin resistance, suggesting that the effect in the young population [121].
of olanzapine on adiponectin levels precedes dysfunctions in whole- The recent introduction of co-therapies has ameliorated some of the
body glucose metabolism [109]. In a similar type of study, Richards SGAs-induced metabolic side-effects. For instance, in a cohort of 30
et al. [110] examined the effects of olanzapine and other SGAs on the patients with schizophrenia under stable treatment with olanzapine,
levels of adiponectin in patients with schizophrenia versus matched Taveira and co-workers tested the effect of co-administration of na-
healthy controls, finding olanzapine-associated hypoadiponectinemia proxen, an opioid receptor antagonist. This combinatorial therapy de-
with a specific decrease of the high molecular weight forms of the creased body fat mass and increased free-fat mass, conferring protection
protein. However, the study failed to find alterations in adiponectin against weight gain. This study does not prove that olanzapine-asso-
expression or in multimer composition in primary adipocytes isolated ciated weight gain is induced by activation of the opioid system since
from subcutaneous WAT and treated ex vivo with olanzapine (10 ng/ml) by itself this system has the ability to improve body weight control, but
for up to 7 days. These results point to the notion that alterations in suggests that blockade of opioid receptors could counteract the dysre-
adiponectin expression and secretion might occur progressively. gulation of lipid metabolism associated with SGAs treatment [122].
Beyond the specific effects of SGAs on metabolic parameters, dis- The molecular impact of SGAs therapy for human adipose tissue is
ease-specific changes should also be considered. In a study performed still poorly understood and many contradictory reports have appeared
with 9 medication-free non-diabetic patients and matched controls, in the literature. A possible reason for these discrepancies is the mis-
Cohn et al. provided evidence for schizophrenia-related insulin re- match of the control groups used in clinical studies (age, BMI, sex and
sistance with an inadequate compensation in insulin secretion [111]. smoking status, among other parameters). Moreover, the effect of SGAs
However, this was not associated with a significant loss of adiponectin is clearly different between healthy volunteers and patients with schi-
levels, as reported in another study of medication-free patients with zophrenia or other psychotic disturbances, and so it is possible that
schizophrenia [112]. It is known that adiponectin levels can be influ- disease-specific alterations are overlooked when using healthy subjects.
enced by gender, as testosterone modulates adiponectin expression and Of interest, alterations in adiposity and increased BMI observed in in-
the secretion of multimers in vitro [113]. Therefore, indirect and direct dividuals undergoing SGAs treatments are closely related to cardio-
drug effects, inflammatory phenomena and/or hypoadiponectinemia metabolic comorbidity and mortality. Accordingly, further studies are
can also be considered as potential mechanisms for the metabolic dis- required to understand the direct (or indirect) impact of these agents in
turbances induced by SGAs [108]. Likewise, plasma levels of calcium human adipose tissue with regard to differentiation, gene expression,
binding-protein B (S100B) are increased in female, but not male, pa- metabolic routes and secretory patterns.
tients treated with clozapine (125–900 mg/day) and positively corre-
late with BMI, pointing to a possible link between this protein and in- 3. Effect of second generation antipsychotics in brown adipose
creased adiposity [114]. It should be mentioned that despite the belief tissue
for astrocytes as the only cells that express and secrete S100B, adipo-
cytes have also been shown to secrete S100B in a process that is ne- BAT is the main player in adaptive thermogenesis, which functions
gatively controlled by insulin [115]. S100B inhibits adenylate cyclase to maintain the core body temperature in cold environments. In con-
by activating D2 receptors and enhancing the extracellular signal- trast to WAT, BAT is defined by the expression of the UCP-1, which
regulated kinases (ERK) in astrocytes [116] and, accordingly, the burns metabolic substrates and dissipates energy in the form of heat
S100B/D2-receptor complex is a potential molecular target of the SGAs [21–23]. Despite their different functions, however, the differentiation
[114]. of white and brown adipocytes is controlled by similar transcription
Healthy volunteers are used in clinical trial studies because they are cascades [123], although brown adipogenesis requires the expression of
normally naive to the tested pharmacological treatments and can pro- additional transcription factors such as the zinc protein PRDM16, which
vide information on the of the treatment per se, independently of pa- forms a complex with C/EBP-β and activates PPAR-γ2, resulting in the
thology. In healthy volunteers, a single dose of olanzapine (10 mg) expression of BAT-specific markers such as UCP-1, peroxisome pro-
elevates fasting glucose levels in the first 4.25 h after administration, liferator-activated receptor gamma coactivator 1 alpha (PGC-1α) and
without altering body weight, and also decreases serum cortisol and Cidea [124].
FFA levels [117]. Since cortisol activates hepatic-sensitive LPL and
ATGL in adipocytes [118], its decrease could explain the reduction in 3.1. The impact of second-generation antipsychotics on BAT thermogenesis
FFA levels. A decrease in fasting and postprandial FFA concentrations
was also found in healthy volunteers that received olanzapine (10 mg/ BAT was recognized for many years to be functionally important for
day) over 8 days, which was associated with increased nocturnal adi- cold acclimatization in small mammals, such as mice and rats, and in
ponectin levels independently of BMI alterations [119]. Moreover, newborn humans. Recent studies using 18F-fluorodeoxyglucose positron
15 days treatment of olanzapine (10 mg/day) in healthy men elevates emission tomography combined with computed tomography have
the levels of adiponectin, leptin and TNF-α, and decreases those of proved the existence of discrete areas of metabolically-active BAT in

7
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

adult humans [24,125], which is functionally controlled by both ca- suggesting that olanzapine induces both BAT morphological alterations
techolamines and insulin [126]. Whereas healthy brown adipocyte and deficiency in its thermogenic function [132]. In addition to these
differentiation increases energy expenditure and contributes to the re- studies, it should be noted that a direct link between hyperphagia and
duction of weight gain [127], diminished brown adipogenesis is related BAT thermogenesis has been recently identified. After a meal, an in-
to obesity and insulin resistance [128]. crease in the gut hormone secretin in circulation activates BAT ther-
In an attempt to unravel the involvement of brown adipocytes in mogenesis by binding to its receptor in brown adipocytes thereby sti-
SGAs-induced weight gain, Oh et al. differentiated a murine brown mulating lipolysis, which is sensed in the brain and promotes satiation
adipocyte cell line in the presence of clozapine (40 μM), quetiapine [133]. Whether this regulatory mechanism is affected by SGA treat-
(30 μM) or ziprasidone (10 μM). At day 8 of differentiation, clozapine ments remains to be elucidated.
inhibited almost completely the expression of PRDM16, UCP-1 and The demonstration of active BAT metabolism in adult humans
Cidea. Quetiapine also reduced the expression of these genes, but its [24–26,125,134] has led to the hypothesis that heat production in BAT
effect was less robust. Moreover, ziprasidone treatment inhibited PPAR- could contribute to emotional hyperthermia. Indeed, mild psycholo-
γ expression at the initiation, but not at the end, of the differentiation gical stress has recently been shown to activate BAT thermogenesis in
process. These findings correlated with the Oil Red-O staining of lipids, adult humans [135]. Related to this, Blessing et al. conducted a study in
which showed an almost complete inhibition of brown adipogenesis by male Sprague-Dawley rats treated subcutaneously or intraperitoneally
clozapine, moderate inhibition by quetiapine, and no inhibition by zi- (catheter) with the FGA chlorpromazine (0.1–5 mg/kg) or with the SGA
prasidone [129]. In parallel to the thermogenic program, it is well clozapine (30 μg–2 mg/kg) or risperidone (6.25 μg–1 mg/kg) and ex-
known in rodents that brown adipocytes differentiate at the end of fetal posed to an intruder rat. The objective of the study was to understand
life via an adipogenic program related to lipid synthesis and the ex- the effect of the antipsychotic agents on emotional hyperthermia, which
pression of lipogenic enzymes, resulting in a multilocular fat droplets activates BAT thermogenesis and tail artery constriction. All of the
phenotype [130]. In the aforementioned study, clozapine reduced the antipsychotics tested strongly reduced the intruder-elicited BAT ther-
expression levels of genes encoding ACC, SCD1, GLUT4, adipocyte mogenesis and artery tail vasocontriction, diminishing the emotional
protein 2 (AP2) and CD36, but not FAS. In contrast to the response of hyperthermia in a dose-dependent manner [136]. Moreover, all of the
clozapine, quetiapine increased ACC and FAS levels, whereas ziprasi- doses required to elicit an effect on the thermogenic capacity of BAT
done treatment did not modulate lipogenic gene expression [129] were lower than those that impact cardiovascular parameters induced
(Fig. 2, panel A). Of interest, the effects of SGAs in brown adipocytes by open field stress [137], acoustic startle response [138] or pre-pulse
differentiation correlate with their ability to induce weight gain in inhibition [139–141], indicating a higher sensitivity of BAT to these
patients (clozapine > quetiapine > ziprasidone). Overall, these results drugs. Even though the D2 antagonist chlorpromazine could inhibit the
support the hypothesis that inhibition of brown adipogenesis may be a thermoregulatory actions in a manner similar to clozapine or risper-
mechanism by which SGAs induce weight gain as a side-effect [129]. idone, this does not necessarily mean that the mechanism of action is
Contrasting with these findings, Ota et al. reported that male Sprague- dependent on D2 receptor blockade, as a previous study with the FGA
Dawley rats treated subcutaneously with risperidone (0.1 mg/kg/day) haloperidol, another potent D2 antagonist, failed to show an acute ef-
for 21 days present hypothermia without changes in adipogenic, lipo- fect on resting body temperature when administered subcutaneously to
genenic or thermogenic gene expression programs in BAT [131]. male Sprague-Dawley rats in one single dose (up to 3 mg/kg) [142].
As mentioned above, hyperphagia is believed to be the main cause Blessing et al. also tested the selective and potent D2 antagonist ra-
of weight gain induced by short-term SGAs treatments. However, in clopride, finding that it did not reduce intruder-elicited BAT thermo-
long-term treatment, when food intake is normalized, a reduction in genesis [136]. By contrast, the hypothermic action of the D2 agonist
energy expenditure due to diminished thermogenesis and locomotor apomorphine was counteracted by haloperidol in a dose-dependent
activity is likely responsible for body weight increase. Zhang and co- manner [142]. Similarly, a low dose of the D2 antagonist spiperone
workers found in female Sprague-Dawley rats that a 34-day treatment ablated apomorphine-induced hypothermia as well as quinpirole (D2
with olanzapine (1 mg/kg, 3 times daily) mixed in cookie dough sig- agonist)-mediated inhibition of BAT thermogenesis. Moreover, spi-
nificantly reduces BAT temperature detected at 45–150 min post- perone diminished the tail artery vasoconstriction induced by clozapine
treatment. This decrease was associated with reductions in UCP-1 and [143]. All of these findings suggest that perhaps the stimulation, and
PGC-1α levels and a diminished abundance of brown adipocytes, not the blockade, of dopamine D2 receptors in the CNS reduces body

Fig. 2. Impact of SGAs in brown adipogenesis and

BAT thermogenic function. A. Scheme of the mole-
cular cell autonomous effects of SGAs in brown adi-
pocytes, focusing on their effects in lipid-related and
thermogenic proteins, as well as the inhibition of
brown adipogenesis. B. List of the receptors in the
CNS through which SGAs.

8
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

temperature [136,143]. Additionally, pharmacological studies demon- different sub-regions of the lateral hypothalamus that induce inhibitory
strated that 5-HT1A agonists and 5-HT2A antagonists reduce BAT ther- responses, constraining sympathetic nerve activity [155], although it
mogenesis and cutaneous vasoconstriction thereby both contributing to has yet to be tested in BAT. In this scenario, the response to olanzapine
hypothermia [144,145], and also agonists of α2-adrenergic signaling would mediate inhibitory actions at the spinal-cord level, but more
reduce BAT thermogenesis [146]. Conversely, activation of H1 re- work is needed to fully understand the involvement of hypothalamus-
ceptors in histamine-producing neurons increases body temperature in BAT axis in the side effects induced by SGAs [150]. Moreover, S100B
association with stimulation of the ascending arousal system [147]. Due and its chaperone calsyntenin 3β have been shown to positively influ-
to their shotgun binding profile, clozapine and risperidone impact all of ence sympathetic innervation of BAT, while their deficiency predis-
the aforementioned families of receptors, motivating the hypothesis poses mice to diet-induced obesity [156]. As aforementioned, S100B is
that perhaps the influence of these SGAs on BAT thermogenic function increased in plasma of female patients under clozapine treatment
is a consequence of synergistic interactions among them [136] (Fig. 2, [114]; however, it should be mentioned that an elevation in circulating
panel B). These observations are still preliminary and further in- S100B does not necessarily mean a positive effect in the functional
vestigation is needed to understand how antipsychotics impact on BAT innervation of BAT, since this adipokine can activate innate immune
thermogenesis. cells by interaction with receptors for advanced glycation end-products
BAT activation may be critical for the gender differences found in [157], which might result in defective insulin sensitivity and thermo-
weight gain upon SGAs treatment. This hypothesis was tested by Ferno genic function in BAT [158]. Further studies are required to address the
et al. in male Sprague-Dawley rats treated with two intramuscular in- impact of the higher plasma levels of S100B upon SGA treatment.
jections of commercially available olanzapine, pamoate depot for- Nevertheless, other reports describe different effects of SGAs on BAT
mulation (ZypAdheras®, 100 mg/kg) at days 0 and 9, and sacrificed at gene expression. Treatment of female C57Bl/6 J mice with risperidone
day 17 after the first injection [148]. The dose was previously used by for 3 weeks (4 mg/kg/day, intraperitoneal) increases food intake and
the authors to induce hyperphagia in female Sprague-Dawley rats body weight, and is associated with reduced locomotor activity in the
[149]. A second group of animals fed a high-fat diet was also treated. first 2 days after the first injection, and without alterations in the core
The authors found that olanzapine transiently increases food intake in body temperature at this time period [159]. Yet, at the end of the third
the days following its administration, with a prolonged effect in the week, the body temperature of animals receiving risperidone increased,
group fed a standard diet. However, despite the induction of temporary and this was associated with lower locomotor activity during the dark
hyperphagia, the olanzapine-treated groups (independently of diet) phase. Also, expression of UCP-1 in BAT and UCP-3 in gastrocnemius
presented a decrease in body weight, suggesting reduced feed effi- was elevated by risperidone, whereas the mRNA encoding orexin A was
ciency. Yet, the treatment failed to modify the mRNA levels of UCP-1 decreased in the hypothalamus. These results suggest that risperidone-
and PGC-1α in the BAT of the experimental groups, likely excluding an induced weight gain in mice is a consequence of increased energy in-
effect mediated by increased BAT activity. The authors also treated take and reduced activity, whereas the higher body temperature may be
another cohort of animals with a single higher dose of the same for- a result of diet-induced thermogenesis and elevated UCP-1 and UCP-3
mulation of olanzapine (150, 200 or 250 mg/kg) and sacrificed them at together with reduced hypothalamic orexin A expression [159]. By
day 13 post-injection. This acute treatment also led to a transient in- contrast, using male Sprague-Dawley rats treated with olanzapine
crease in food intake accompanied by a decrease in cumulative weight (1 mg/kg/day) in food for 6 weeks, Minet-Ringuet and co-workers re-
gain which, contrary to the previous treatment, could be explained by ported no alterations in mitochondrial thermogenesis in the BAT of
an up-regulation of UCP-1 and PGC-1α in the BAT of olanzapine-treated olanzapine-treated animals. Even in the presence of guanosine dipho-
animals indicating increased energy expenditure [148]. This is contrary sphate, which inhibits UCP-1, the respiratory rates at different mem-
to what has been reported in female animals [149,150]. For example, brane potentials showed no alterations in proton conductance after the
Skrede et al. described that treatment of female Sprague-Dawley rats treatment [160]. Considering the apparent contradictory results in
with olanzapine (6 mg/kg) twice daily by oral gavage for 13 days de- some of these studies, we cannot discard that these outcomes might be
creases both UCP-1 and PGC-1α in BAT and increases weight gain. Of related, at least in part, to the different binding affinities, doses and/or
note, when the animals were treated with aripiprazole in the same administration routes of the SGAs tested.
experimental set-up, only PGC-1α expression decreased [151]. In an- Co-therapies have also been tested to counteract the side effects of
other study [150], female Sprague-Dawley rats treated orally with SGAs in BAT function. Specifically, and since the antagonism of these
olanzapine (6 mg/kg/day) for 24 days displayed a transient increase in drugs to histaminergic receptors was identified as a main contributor to
food intake together with increased body weight gain along the treat- body weight alterations, the H1 receptor agonist and H3 receptor an-
ment, as compared with non-treated controls. Likewise, in pair-fed rats tagonist betahistine was tested in female Sprague-Dawley rats as a co-
receiving olanzapine, body weight increased along the treatment as therapy together with olanzapine (3 mg/kg/day in cookie dough) for
compared with vehicle-treated rats, in association with decreased en- 3 weeks after 23 days of olanzapine treatment and then followed by 19
ergy expenditure, a reduction in BAT temperature and a decreased days of washout [161]. Betahistine co-therapy reduced (45%) the body
expression of UCP-1 protein. Interestingly, an increase in FOS protein weight gain induced by olanzapine and counteracted the olanzapine-
was found in spinal cord neurons projecting to discrete sites in the induced increases in H1 receptor protein levels and AMPKα phosphor-
brainstem and hypothalamus, suggesting their excitatory activation, ylation in the hypothalamus, as well as the decrease in UCP-1 and PGC-
and some of these neurons, specifically those located at the perifornical 1α in BAT. Importantly, this work led to the hypothesis that olanzapine-
region of the lateral hypothalamus, were positive for orexin A [150], a induced AMPK activation in the hypothalamus mediates weight im-
key neuropeptide in BAT-directed neurons in the lateral hypothalamus balances by diminishing BAT thermogenesis through the hypothalamic
[152]. In a previous study, the activation of these neurons upon olan- H1 receptor-AMPK pathway [161]. These results align with the effects
zapine treatment was associated with hyperphagia, likely due to their of specific genetic activation of hypothalamic AMPK in the ven-
projections to the areas associated with hunger in the cerebral cortex tromedial nucleus of the hypothalamus, which counteracts the central
[153]. Additionally, the same perifornical orexin A-positive neurons in effect of the thyroid hormone triiodothyronine in BAT activation [162].
the anterior cingulated cortex project axon collaterals to sites that are Another tested co-therapy is the nonselective β-adrenergic-blocker
related to food intake and thermogenesis [154], corroborating that the propranolol, which can mitigate risperidone (0.75 mg/kg/day, oral
activation of these neurons by olanzapine provides a possible ex- gavage for 4 weeks)-induced alterations in BAT in C57BL/6 J females by
planation for SGA-mediated changes in both feeding and energy ex- increasing UCP-1 and PGC-1α levels [163].
penditure, thereby leading to weight gain. Another possible factor re-
lated to the decrease in thermogenesis by SGAs is the stimulation of

9
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

3.2. The impact of second-generation antipsychotics on BAT cytokine haloperidol (all used at 50 ng/ml) or aripiprazole (100 ng/ml) failed to
expression alter UCP-1 expression [169]. Of interest, elevations of peripheral ser-
otonin levels in obese mice have been described to inhibit beiging, as
Beyond the impact of SGAs on BAT adipogenesis, lipogenesis and well as the sensitivity of both beige and brown cells to thermogenic
function, the aforementioned study by Oh et al. [129] showed that stimulus, in a cell autonomous manner [172,173]. Also, increased levels
differentiation of murine brown pre-adipocytes in the presence of clo- of peripheral serotonin [174] and polymorphisms in the gene encoding
zapine also modulates the expression of adipokines by reducing resistin, tryptophan hydroxylase 1, which catalyzes the rate-limiting step of
leptin and adiponectin. The same outcome was observed for resistin and serotonin production outside the CNS, are associated with obesity
leptin levels in quetiapine-treated cells, whereas ziprasidone was un- [175]. The importance of serotonin signaling was demonstrated by
able to affect adipokine expression. Again, the effect of these SGAs in Kristof and co-workers, who co-treated adipocytes [55] with exogenous
leptin expression reflects their ability to induce weight gain [129]. Of serotonin and clozapine during differentiation, and found a partial re-
interest, leptin synthesis and secretion is modulated by insulin [164] duction of browning caused by the lowered expression of brown-related
and it has been reported that SGAs treatment interferes with insulin genes. Of interest, the serotonergic receptor HTR2A was found up-
signaling [165], although it has not yet been reported in BAT. In such regulated in clozapine-treated adipocytes [169], and it has recently
case, the impact of SGAs on the insulin response in BAT might po- been described that HTR2A activation results in Gq-mediated signaling
tentiate the reduction of leptin levels [129]. Additionally, the role of capable of abolishing browning in mice and humans [176]. In fact, Gq
UCP-1 in augmenting the anorexigenic effect of leptin should also be protein expression negatively correlates with UCP-1 levels in the WAT
considered [166], suggesting that in synergy with the direct suppres- in mice. Given this, and considering that clozapine antagonizes several
sion of leptin expression, clozapine also reduces leptin actions through serotonergic receptors with specific high affinity for HTR2A [177], the
inhibition of UCP-1 expression [129]. Independently of these studies, alteration in Gq signaling induced by serotonin can, at least in part,
some reports suggest that clozapine increases the serum levels of this explain the unexpected results of Kristof and co-workers. Of interest,
adipokine in patients with schizophrenia [167]. Of interest, the work of clozapine is a well-known agonist of H4 receptors [178,179], which are
Zhang et al. discussed above [130] in rats treated with olanzapine highly expressed in adipocytes. In a recent study where these receptors
showed increased infiltration of macrophages in BAT, which was ac- were knocked-down in subcutaneous WAT, cold-induced browning and
companied by up-regulation of the inflammatory cytokines TNF-α, IL- lipolysis were abolished. By contrast, when 4-methylhistamine, a se-
1β and IL-6, suggesting that olanzapine can trigger peripheral in- lective H4 receptor agonist, was adjacently injected in subcutaneous
flammation. Moreover, these inflammatory cytokines are also up- WAT, browning was increased through activation of p38/MAPK and
regulated in the hypothalamus [96], an effect possibly related to the ERK1/2/MAPK signaling pathways, together with an acceleration in
undermining of thermogenesis, as previously described outwith the metabolic rate and tolerance to hypothermia [180]. Considering these
context of SGAs treatment [168]. results, the increased browning in ASCs treated with clozapine de-
scribed by Kristof et al. could be a consequence, at least in part, of H4
4. Effect of second generation antipsychotics on beige adipose receptor activation. Nevertheless, more studies are required to under-
tissue stand the precise impact of SGAs in beige adipocytes and the browning
process. In this sense, the study by Kristof et al. [169] is an excellent
Beige adipose tissue is found within the white depots, most notably start. Moreover, it highlights specific signaling pathways that might
in subcutaneous depots. Like BAT, beige adipocytes are characterized affect the differentiation of the adipocytes by clozapine, with special
by high amounts of mitochondria and by their ability to dissipate en- importance to serotonergic signaling, pointing to a possible co-ther-
ergy by thermogenesis due to the considerable levels of UCP-1 [16]. apeutic approach to ameliorate the adverse effects of these drugs in the
Kristof et al. showed that ex vivo treatment of human ASCs isolated from pool of beige adipocytes.
20 to 65-year-old healthy volunteers (BMI < 29.9) with clozapine
(100 ng/ml) during differentiation resulted in a 10-fold increase of their 5. Concluding remarks
expression levels of UCP-1, pointing to a beiging/browning process of
these precursors [169]. The up-regulation of UCP-1 mRNA levels was The study of adipose tissue depots as both metabolically-active tis-
associated with increased expression of other browning marker genes, sues and endocrine organs that orchestrate peripheral insulin action,
including TBX1. Indeed, the addition of clozapine to differentiating inflammation and energy expenditure is a relatively new field. Much
ASCs resulted in a 1.5-fold increase in the number of beige cells. This effort is being directed to understand the relevance of adipose tissues as
pattern of browning-related gene expression was up-regulated even triggers of metabolic dysfunctions and/or possible therapeutic targets
when the cells were treated only in the last 2 or 4 days of the differ- for metabolic pathologies including those associated with SGAs treat-
entiation process, suggesting that clozapine can promote a white-to- ment. In this review, we have attempted to illustrate the modulation of
beige shift in adipocyte cell fate in the latter stages of differentiation. adipocyte fate and their metabolic and endocrine functions by SGAs,
However, in spite of the increase in browning, the cells were unable to both in cell-based systems and in pre-clinical and clinical studies. These
activate basal thermogenesis unless they were stimulated with β-adre- drugs can modulate white adipocyte differentiation, resulting in in-
nergic agonists. Independently of this, clozapine added for 12 h to creased lipid accumulation and adiposity that is related to the clinical
terminally-differentiated adipocytes from ASCs failed to alter their ex- manifestations of patients under SGAs therapy in an apparent gender-
pression profile of beiging-related genes. This suggests that there is a dependent manner. Also, SGAs exacerbate the chronic low-grade in-
critical stage of the differentiation process at which clozapine commits flammatory processes of WAT by augmenting the expression and se-
mesenchymal adipocyte progenitors to beige adipocytes. In addition, cretion of pro-inflammatory cytokines and the recruitment of immune
the lipid droplets found in the differentiated-clozapine treated cells cells, which is associated with altered secretory patterns of adipokines
were smaller and presented a multiocular distribution, which was as- correlating with obesity, insulin resistance and hyperphagia. These
sociated with increased levels of mitochondrial DNA, but with dimin- processes are clearly main contributors to the evident metabolic dys-
ished sensitivity to cAMP stimulation [169]. In this regard, defective functions under SGAs therapy. In addition, SGAs treatment impacts
cAMP signaling has been found in the brain of humans [170] and mice brown adipogenesis and BAT homeostasis through their ability to
[171] under SGAs treatment. By contrast, and as mentioned above, modulate the thermogenic differentiation program by controlling the
treatment of white adipocytes with olanzapine during differentiation expression of UCP-1, also in a gender-dependent manner. Similarly,
augments lipid accumulation. Moreover, differentiation of human ASCs browning of white adipocytes is altered by SGAs, particularly clozapine.
in the presence of olanzapine, ziprasidone, risperidone, quetiapine, Only one study so far has examined this particular adipose depot, but it

10
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

could be a crucial factor in the secondary side-effects during SGAs schizophrenia, J. Clin. Psychiatry 62 (Suppl. 7) (2001) 32–37.
treatments. Another important but less studied theme is how the CNS- [15] P. Manu, et al., Weight gain and obesity in schizophrenia: epidemiology, patho-
biology, and management, Acta Psychiatr. Scand. 132 (2) (2015) 97–108, https://
adipose axis is affected by SGAs. The central outputs of this axis are doi.org/10.1111/acps.12445.
major coordinators of peripheral adipose tissue function and home- [16] L. Luo, M. Liu, Adipose tissue in control of metabolism, J. Endocrinol. 231 (3)
ostasis by supplying neurotransmitters and, in this regard, alterations in (2016) R77–R99, https://doi.org/10.1530/JOE-16-0211.
[17] C.Y. Tan, A. Vidal-Puig, Adipose tissue expandability: the metabolic problems of
the neuronal circuits that control energy expenditure or food intake obesity may arise from the inability to become more obese, Biochem. Soc. Trans.
boost the peripheral disturbances in synergy with specific cell autono- 36 (Pt 5) (2008) 935–940, https://doi.org/10.1042/BST0360935.
mous effects of the SGAs. Finally, and not covered in this work, meta- [18] S. Kajimura, Adipose tissue in 2016: advances in the understanding of adipose
tissue biology, Nat. Rev. Endocrinol. 13 (2) (2017) 69–70, https://doi.org/10.
bolic deficits in patients under SGAs treatment might be due to the 1038/nrendo.2016.211.
defective cross-talk between fat and liver/skeletal muscle, which are [19] P. Seale, et al., PRDM16 controls a brown fat/skeletal muscle switch, Nature 454
key tissues for de novo lipogenesis or glucose uptake, respectively. (7207) (2008) 961–967, https://doi.org/10.1038/nature07182.
[20] C. Lepper, C.M. Fan, Inducible lineage tracing of Pax7-descendant cells reveals
In conclusion, despite the great strides made in our understanding of
embryonic origin of adult satellite cells, Genesis 48 (7) (2010) 424–436, https://
the metabolic consequences of the treatment with SGAs, further in- doi.org/10.1002/dvg.20630.
vestigations are required to fully address the impact of SGAs in the [21] W. Aherne, D. Hull, Brown adipose tissue and heat production in the newborn
different adipose tissue depots of the organism and the interactome infant, J. Pathol. Bacteriol. 91 (1) (1966) 223–234, https://doi.org/10.1002/path.
1700910126.
between them either by inflammatory molecules, adipokines and/or [22] B. Cannon, J. Nedergaard, Brown adipose tissue: function and physiological sig-
other activator or inhibitory compounds, as well as the connections nificance, Physiol. Rev. 84 (1) (2004) 277–359, https://doi.org/10.1152/physrev.
between the CNS and fat. 00015.2003.
[23] A. Fedorenko, P.V. Lishko, Y. Kirichok, Mechanism of fatty-acid-dependent UCP1
uncoupling in brown fat mitochondria, Cell 151 (2) (2012) 400–413, https://doi.
Transparency document org/10.1016/j.cell.2012.09.010.
[24] J. Nedergaard, T. Bengtsson, B. Cannon, Unexpected evidence for active brown
adipose tissue in adult humans, Am. J. Physiol. Endocrinol. Metab. 293 (2) (2007)
The Transparency document associated this article can be found, in E444–E452, https://doi.org/10.1152/ajpendo.00691.2006.
online version. [25] A.M. Cypess, et al., Identification and importance of brown adipose tissue in adult
humans, N. Engl. J. Med. 360 (15) (2009) 1509–1517, https://doi.org/10.1056/
NEJMoa0810780.
Acknowledgements [26] M.C. Zingaretti, et al., The presence of UCP1 demonstrates that metabolically
active adipose tissue in the neck of adult humans truly represents brown adipose
This work was funded by H2020 Marie Skłodowska-Curie tissue, FASEB J. 23 (9) (2009) 3113–3120, https://doi.org/10.1096/fj.09-133546.
[27] J. Wu, et al., Beige adipocytes are a distinct type of thermogenic fat cell in mouse
ActionsITN-TREATMENT (Grant Agreement 721236, European
and human, Cell 150 (2) (2012) 366–376, https://doi.org/10.1016/j.cell.2012.05.
Commission). We also acknowledge grants RTI2018-094052-B-100 016.
(MICINN/FEDER, Spain), S2017/BMD-3684 MOIR2-CM (Comunidad [28] P. Young, J.R. Arch, M. Ashwell, Brown adipose tissue in the parametrial fat pad of
de Madrid, Spain) and CIBERdem (ISCIII, Spain). the mouse, FEBS Lett. 167 (1) (1984) 10–14.
[29] M. Harms, P. Seale, Brown and beige fat: development, function and therapeutic
potential, Nat. Med. 19 (10) (2013) 1252–1263, https://doi.org/10.1038/nm.
Declaration of competing interest 3361.
[30] N. Petrovic, et al., Chronic peroxisome proliferator-activated receptor gamma
(PPARgamma) activation of epididymally derived white adipocyte cultures reveals
The authors declare they have no conflicts of interest to be declared. a population of thermogenically competent, UCP1-containing adipocytes mole-
cularly distinct from classic brown adipocytes, J. Biol. Chem. 285 (10) (2010)
References 7153–7164, https://doi.org/10.1074/jbc.M109.053942.
[31] J. Himms-Hagen, et al., Multilocular fat cells in WAT of CL-316243-treated rats
derive directly from white adipocytes, Am. J. Physiol. Cell. Physiol. 279 (3) (2000)
[1] W.G. Rosen, et al., Positive and negative symptoms in schizophrenia, Psychiatry C670–C681, https://doi.org/10.1152/ajpcell.2000.279.3.C670.
Res. 13 (4) (1984) 277–284, https://doi.org/10.1016/0165-1781(84)90075-1. [32] G. Barbatelli, et al., The emergence of cold-induced brown adipocytes in mouse
[2] G. Carra, et al., Positive and negative symptoms in schizophrenia: a longitudinal white fat depots is determined predominantly by white to brown adipocyte
analysis using latent variable structural equation modelling, Schizophr. Res. 204 transdifferentiation, Am. J. Physiol. Endocrinol. Metab. 298 (6) (2010)
(2019) 58–64, https://doi.org/10.1016/j.schres.2018.08.018. E1244–E1253, https://doi.org/10.1152/ajpendo.00600.2009.
[3] W.W. Fleischhacker, et al., Schizophrenia—time to commit to policy change, [33] M.T. Foster, et al., Subcutaneous adipose tissue transplantation in diet-induced
Schizophr. Bull. 40 (Suppl. 3) (2014) S165–S194, https://doi.org/10.1093/ obese mice attenuates metabolic dysregulation while removal exacerbates it,
schbul/sbu006. Physiol. Rep. 1 (2) (2013), https://doi.org/10.1002/phy2.15.
[4] K.R. Patel, et al., Schizophrenia: overview and treatment options, P T. 39 (9) [34] U. Smith, B.B. Kahn, Adipose tissue regulates insulin sensitivity: role of adipo-
(2014) 638–645. genesis, de novo lipogenesis and novel lipids, J. Intern. Med. 280 (5) (2016)
[5] G. Remington, et al., Guidelines for the pharmacotherapy of schizophrenia in 465–475, https://doi.org/10.1111/joim.12540.
adults, Can. J. Psychiatry 62 (9) (2017) 604–616, https://doi.org/10.1177/ [35] F.C. Starrenburg, J.P. Bogers, How can antipsychotics cause diabetes mellitus?
0706743717720448. Insights based on receptor-binding profiles, humoral factors and transporter pro-
[6] R. Brisch, et al., The role of dopamine in schizophrenia from a neurobiological and teins, Eur Psychiatry 24 (3) (2009) 164–170, https://doi.org/10.1016/j.eurpsy.
evolutionary perspective: old fashioned, but still in vogue, Front. Psychiatry 5 2009.01.001.
(2014) 47, https://doi.org/10.3389/fpsyt.2014.00047. [36] J. Ye, R.A. DeBose-Boyd, Regulation of cholesterol and fatty acid synthesis, Cold
[7] L.H. Shen, M.H. Liao, Y.C. Tseng, Recent advances in imaging of dopaminergic Spring Harb. Perspect. Biol. 3 (7) (2011), https://doi.org/10.1101/cshperspect.
neurons for evaluation of neuropsychiatric disorders, J. Biomed. Biotechnol. 2012 a004754.
(2012) 259349, https://doi.org/10.1155/2012/259349. [37] H.S. Vestri, et al., Atypical antipsychotic drugs directly impair insulin action in
[8] O.D. Howes, R.M. Murray, Schizophrenia: an integrated sociodevelopmental- adipocytes: effects on glucose transport, lipogenesis, and antilipolysis,
cognitive model, Lancet 383 (9929) (2014) 1677–1687, https://doi.org/10.1016/ Neuropsychopharmacology 32 (4) (2007) 765–772, https://doi.org/10.1038/sj.
S0140-6736(13)62036-X. npp.1301142.
[9] P. O’Donnell, A.A. Grace, Dysfunctions in multiple interrelated systems as the [38] C.C. Chen, et al., Overexpression of Insig-2 inhibits atypical antipsychotic-induced
neurobiological bases of schizophrenic symptom clusters, Schizophr. Bull. 24 (2) adipogenic differentiation and lipid biosynthesis in adipose-derived stem cells, Sci.
(1998) 267–283, https://doi.org/10.1093/oxfordjournals.schbul.a033325. Rep. 7 (1) (2017) 10901, https://doi.org/10.1038/s41598-017-11323-9.
[10] E.H. Simpson, et al., Selective overexpression of dopamine D3 receptors in the [39] L.H. Yang, et al., Olanzapine induces SREBP-1-related adipogenesis in 3T3-L1
striatum disrupts motivation but not cognition, Biol. Psychiatry 76 (10) (2014) cells, Pharmacol. Res. 56 (3) (2007) 202–208, https://doi.org/10.1016/j.phrs.
823–831, https://doi.org/10.1016/j.biopsych.2013.11.023. 2007.05.007.
[11] J. Lally, J.H. MacCabe, Antipsychotic medication in schizophrenia: a review, Br. [40] M.B. Raeder, et al., SREBP activation by antipsychotic- and antidepressant-drugs
Med. Bull. 114 (1) (2015) 169–179, https://doi.org/10.1093/bmb/ldv017. in cultured human liver cells: relevance for metabolic side-effects? Mol .Cell.
[12] D. Cohen, et al., Hyperglycemia and diabetes in patients with schizophrenia or Biochem. 289 (1–2) (2006) 167–173, https://doi.org/10.1007/s11010-006-
schizoaffective disorders, Diabetes Care 29 (4) (2006) 786–791, https://doi.org/ 9160-4.
10.2337/diacare.29.04.06.dc05-1261. [41] B. Guesdon, R.G. Denis, D. Richard, Additive effects of olanzapine and melanin-
[13] D.E. Casey, Dyslipidemia and atypical antipsychotic drugs, J. Clin. Psychiatry 65 concentrating hormone agonism on energy balance, Behav. Brain Res. 207 (1)
(Suppl. 18) (2004) 27–35. (2010) 14–20, https://doi.org/10.1016/j.bbr.2009.09.032.
[14] I. Kurzthaler, W.W. Fleischhacker, The clinical implications of weight gain in [42] D. Georgescu, et al., The hypothalamic neuropeptide melanin-concentrating

11
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

hormone acts in the nucleus accumbens to modulate feeding behavior and forced- 403–411.
swim performance, J. Neurosci. 25 (11) (2005) 2933–2940, https://doi.org/10. [70] K. Makki, P. Froguel, I. Wolowczuk, Adipose tissue in obesity-related inflamma-
1523/JNEUROSCI.1714-04.2005. tion and insulin resistance: cells, cytokines, and chemokines, ISRN Inflamm. 2013
[43] A.K. Walker, et al., A conserved SREBP-1/phosphatidylcholine feedback circuit (2013) 139239, https://doi.org/10.1155/2013/139239.
regulates lipogenesis in metazoans, Cell 147 (4) (2011) 840–852, https://doi.org/ [71] A.S. Ryan, et al., Plasma adiponectin and leptin levels, body composition, and
10.1016/j.cell.2011.09.045. glucose utilization in adult women with wide ranges of age and obesity, Diabetes
[44] D. Yabe, M.S. Brown, J.L. Goldstein, Insig-2, a second endoplasmic reticulum Care 26 (8) (2003) 2383–2388, https://doi.org/10.2337/diacare.26.8.2383.
protein that binds SCAP and blocks export of sterol regulatory element-binding [72] T. Sugai, et al., Dysregulation of adipocytokines related to second-generation
proteins, Proc. Natl. Acad. Sci. U. S. A. 99 (20) (2002) 12753–12758, https://doi. antipsychotics in normal fasting glucose patients with schizophrenia, J. Clin.
org/10.1073/pnas.162488899. Psychopharmacol. 32 (3) (2012) 390–393, https://doi.org/10.1097/JCP.
[45] Y.J. Liou, et al., Gene-gene interactions of the INSIG1 and INSIG2 in metabolic 0b013e3182524393.
syndrome in schizophrenic patients treated with atypical antipsychotics, [73] G. Fantuzzi, Adiponectin in inflammatory and immune-mediated diseases,
Pharmacogenomics J. 12 (1) (2012) 54–61, https://doi.org/10.1038/tpj.2010.74. Cytokine 64 (1) (2013) 1–10, https://doi.org/10.1016/j.cyto.2013.06.317.
[46] T. Imai, et al., Peroxisome proliferator-activated receptor gamma is required in [74] K. Robinson, J. Prins, B. Venkatesh, Clinical review: adiponectin biology and its
mature white and brown adipocytes for their survival in the mouse, Proc. Natl. role in inflammation and critical illness, Crit. Care 15 (2) (2011) 221, https://doi.
Acad. Sci. U.S.A. 101 (13) (2004) 4543–4547, https://doi.org/10.1073/pnas. org/10.1186/cc10021.
0400356101. [75] F.S. Lira, et al., Both adiponectin and interleukin-10 inhibit LPS-induced activation
[47] P. Tontonoz, B.M. Spiegelman, Fat and beyond: the diverse biology of of the NF-κB pathway in 3T3-L1 adipocytes, Cytokine 57 (1) (2012) 98–106,
PPARgamma, Annu. Rev. Biochem. 77 (2008) 289–312, https://doi.org/10.1146/ https://doi.org/10.1016/j.cyto.2011.10.001.
annurev.biochem.77.061307.091829. [76] A. Gastaldelli, M. Gaggini, R.A. DeFronzo, Role of adipose tissue insulin resistance
[48] D. Moseti, A. Regassa, W.K. Kim, Molecular regulation of adipogenesis and po- in the natural history of type 2 diabetes: results from the San Antonio metabolism
tential anti-adipogenic bioactive molecules, Int. J. Mol. Sci. 17 (1) (2016), https:// study, Diabetes 66 (4) (2017) 815–822, https://doi.org/10.2337/db16-1167.
doi.org/10.3390/ijms17010124. [77] P.D. Harvey, C.R. Bowie, Ziprasidone: efficacy, tolerability, and emerging data on
[49] A.L. Sertie, et al., Effects of antipsychotics with different weight gain liabilities on wide-ranging effectiveness, Expert. Opin. Pharmacother. 6 (2) (2005) 337–346,
human in vitro models of adipose tissue differentiation and metabolism, Prog. https://doi.org/10.1517/14656566.6.2.337.
Neuropsychopharmacol. Biol. Psychiatry. 35 (8) (2011) 1884–1890, https://doi. [78] S.J. Hunter, W.T. Garvey, Insulin action and insulin resistance: diseases involving
org/10.1016/j.pnpbp.2011.07.017. defects in insulin receptors, signal transduction, and the glucose transport effector
[50] K. Hemmrich, et al., Clozapine enhances differentiation of adipocyte progenitor system, Am. J. Med. 105 (4) (1998) 331–345.
cells, Mol. Psychiatry 11 (11) (2006) 980–981, https://doi.org/10.1038/sj.mp. [79] G.C. Smith, et al., Atypical antipsychotic drugs induce derangements in glucose
4001892. homeostasis by acutely increasing glucagon secretion and hepatic glucose output
[51] J.W. Newcomer, Second-generation (atypical) antipsychotics and metabolic ef- in the rat, Diabetologia 51 (12) (2008) 2309–2317, https://doi.org/10.1007/
fects: a comprehensive literature review, CNS Drugs 19 (Suppl. 1) (2005) 1–93, s00125-008-1152-3.
https://doi.org/10.2165/00023210-200519001-00001. [80] K.A. Robinson, S.Z. Yacoub Wasef, M.G. Buse, At therapeutic concentrations,
[52] D.B. Allison, et al., Antipsychotic-induced weight gain: a comprehensive research olanzapine does not affect basal or insulin-stimulated glucose transport in 3T3-L1
synthesis, Am. J. Psychiatry 156 (11) (1999) 1686–1696, https://doi.org/10. adipocytes, Prog. Neuro-Psychopharmacol. Biol. Psychiatry 30 (1) (2006) 93–98,
1176/ajp.156.11.1686. https://doi.org/10.1016/j.pnpbp.2005.06.008.
[53] Y. Wang, et al., Pimozide, a novel fatty acid binding protein 4 inhibitor, promotes [81] V.L. Albaugh, et al., Hormonal and metabolic effects of olanzapine and clozapine
adipogenesis of 3T3-L1 cells by activating PPARgamma, ACS Chem. Neurosci. 6 related to body weight in rodents, Obesity (Silver Spring) 14 (1) (2006) 36–51,
(2) (2015) 211–218, https://doi.org/10.1021/cn5002107. https://doi.org/10.1038/oby.2006.6.
[54] C. Pavan, et al., Weight gain related to treatment with atypical antipsychotics is [82] A.I. Zugno, et al., Energy metabolism, leptin, and biochemical parameters are
due to activation of PKC-beta, Pharmacogenomics J. 10 (5) (2010) 408–417, altered in rats subjected to the chronic administration of olanzapine, Braz. J.
https://doi.org/10.1038/tpj.2009.67. Psychiatry 34 (2) (2012) 168–175.
[55] S. Nimura, et al., Olanzapine promotes the accumulation of lipid droplets and the [83] A.J. Goudie, J.A. Smith, J.C. Halford, Characterization of olanzapine-induced
expression of multiple perilipins in human adipocytes, Biochem. Biophys. Res. weight gain in rats, J. Psychopharmacol. 16 (4) (2002) 291–296, https://doi.org/
Commun. 467 (4) (2015) 906–912, https://doi.org/10.1016/j.bbrc.2015.10.045. 10.1177/026988110201600402.
[56] L. Qiao, et al., Transcriptional regulation of fatty acid translocase/CD36 expres- [84] A.M. Volpato, A.I. Zugno, J. Quevedo, Recent evidence and potential mechanisms
sion by CCAAT/enhancer-binding protein alpha, J. Biol. Chem. 283 (14) (2008) underlying weight gain and insulin resistance due to atypical antipsychotics, Braz.
8788–8795, https://doi.org/10.1074/jbc.M800055200. J. Psychiatry 35 (3) (2013) 295–304, https://doi.org/10.1590/1516-4446-2012-
[57] N. Saini, et al., Fatty acid transport protein-2 inhibitor Grassofermata/CB5 pro- 1052.
tects cells against lipid accumulation and toxicity, Biochem. Biophys. Res. [85] C.P. Yang, et al., Olanzapine induced dysmetabolic changes involving tissue
Commun. 465 (3) (2015) 534–541, https://doi.org/10.1016/j.bbrc.2015.08.055. chromium mobilization in female rats, Int. J. Mol. Sci. 20 (3) (2019), https://doi.
[58] G.S. Hotamisligil, Inflammation, metaflammation and immunometabolic dis- org/10.3390/ijms20030640.
orders, Nature 542 (7640) (2017) 177–185, https://doi.org/10.1038/ [86] B. Pouzet, et al., Chronic treatment with antipsychotics in rats as a model for
nature21363. antipsychotic-induced weight gain in human, Pharmacol. Biochem. Behav. 75 (1)
[59] T. Yamauchi, T. Kadowaki, Physiological and pathophysiological roles of adipo- (2003) 133–140.
nectin and adiponectin receptors in the integrated regulation of metabolic and [87] K. Horska, et al., Olanzapine-depot administration induces time-dependent
cardiovascular diseases, Int. J. Obes. 32 (Suppl. 7) (2008) S13–S18, https://doi. changes in adipose tissue endocrine function in rats, Psychoneuroendocrinology
org/10.1038/ijo.2008.233. 73 (2016) 177–185, https://doi.org/10.1016/j.psyneuen.2016.07.218.
[60] M.E. Trujillo, P.E. Scherer, Adiponectin–journey from an adipocyte secretory [88] W. Tan, H. Fan, P.H. Yu, Induction of subcutaneous adipose proliferation by
protein to biomarker of the metabolic syndrome, J. Intern. Med. 257 (2) (2005) olanzapine in rodents, Prog. Neuro-Psychopharmacol. Biol. Psychiatry 34 (6)
167–175, https://doi.org/10.1111/j.1365-2796.2004.01426.x. (2010) 1098–1103, https://doi.org/10.1016/j.pnpbp.2010.06.002.
[61] A.K. Sarvari, et al., Atypical antipsychotics induce both proinflammatory and [89] J. Minet-Ringuet, et al., Alterations of lipid metabolism and gene expression in rat
adipogenic gene expression in human adipocytes in vitro, Biochem. Biophys. Res. adipocytes during chronic olanzapine treatment, Mol. Psychiatry 12 (6) (2007)
Commun. 450 (4) (2014) 1383–1389, https://doi.org/10.1016/j.bbrc.2014.07. 562–571, https://doi.org/10.1038/sj.mp.4001948.
005. [90] G.E. Duncan, S. Zorn, J.A. Lieberman, Mechanisms of typical and atypical anti-
[62] T. Tsubai, et al., Effects of clozapine on adipokine secretions/productions and lipid psychotic drug action in relation to dopamine and NMDA receptor hypofunction
droplets in 3T3-L1 adipocytes, J. Pharmacol. Sci. 133 (2) (2017) 79–87, https:// hypotheses of schizophrenia, Mol. Psychiatry 4 (5) (1999) 418–428.
doi.org/10.1016/j.jphs.2017.01.004. [91] M. Lafontan, et al., Adrenergic receptors and fat cells: differential recruitment by
[63] J.M. Friedman, J.L. Halaas, Leptin and the regulation of body weight in mammals, physiological amines and homologous regulation, Obes. Res. 3 (Suppl. 4) (1995)
Nature 395 (6704) (1998) 763–770, https://doi.org/10.1038/27376. 507S–514S.
[64] S.B. Heymsfield, et al., Recombinant leptin for weight loss in obese and lean [92] M. Victoriano, et al., Olanzapine-induced accumulation of adipose tissue is asso-
adults: a randomized, controlled, dose-escalation trial, JAMA 282 (16) (1999) ciated with an inflammatory state, Brain Res. 1350 (2010) 167–175, https://doi.
1568–1575. org/10.1016/j.brainres.2010.05.060.
[65] L.K. Heisler, et al., Activation of central melanocortin pathways by fenfluramine, [93] A. Calevro, et al., Effects of chronic antipsychotic drug exposure on the expression
Science 297 (5581) (2002) 609–611, https://doi.org/10.1126/science.1072327. of translocator protein and inflammatory markers in rat adipose tissue,
[66] M. Yamashita, et al., Expression cloning of a cDNA encoding the bovine histamine Psychoneuroendocrinology 95 (2018) 28–33, https://doi.org/10.1016/j.
H1 receptor, Proc. Natl. Acad. Sci. U. S. A. 88 (24) (1991) 11515–11519, https:// psyneuen.2018.05.021.
doi.org/10.1073/pnas.88.24.11515. [94] V. Rotter, I. Nagaev, U. Smith, Interleukin-6 (IL-6) induces insulin resistance in
[67] M. He, C. Deng, X.F. Huang, The role of hypothalamic H1 receptor antagonism in 3T3-L1 adipocytes and is, like IL-8 and tumor necrosis factor-alpha, overexpressed
antipsychotic-induced weight gain, CNS Drugs 27 (6) (2013) 423–434, https:// in human fat cells from insulin-resistant subjects, J. Biol. Chem. 278 (46) (2003)
doi.org/10.1007/s40263-013-0062-1. 45777–45784, https://doi.org/10.1074/jbc.M301977200.
[68] L.H. Tecott, et al., Eating disorder and epilepsy in mice lacking 5-HT2c serotonin [95] H. Li, et al., Chronic olanzapine administration causes metabolic syndrome
receptors, Nature 374 (6522) (1995) 542–546, https://doi.org/10.1038/ through inflammatory cytokines in rodent models of insulin resistance, Sci. Rep. 9
374542a0. (1) (2019) 1582, https://doi.org/10.1038/s41598-018-36930-y.
[69] T. Sakata, H. Yoshimatsu, M. Kurokawa, Hypothalamic neuronal histamine: im- [96] Q. Zhang, et al., Effects of olanzapine on the elevation of macrophage infiltration
plications of its homeostatic control of energy metabolism, Nutrition 13 (5) (1997) and pro-inflammatory cytokine expression in female rats, J. Psychopharmacol. 28

12
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

(12) (2014) 1161–1169, https://doi.org/10.1177/0269881114555250. 788–796, https://doi.org/10.1001/jamapsychiatry.2018.1088.

[97] K.J. Davey, et al., Gender-dependent consequences of chronic olanzapine in the [122] T.H. Taveira, et al., The effect of naltrexone on body fat mass in olanzapine-treated
rat: effects on body weight, inflammatory, metabolic and microbiota parameters, schizophrenic or schizoaffective patients: a randomized double-blind placebo-
Psychopharmacology 221 (1) (2012) 155–169, https://doi.org/10.1007/s00213- controlled pilot study, J. Psychopharmacol. 28 (4) (2014) 395–400, https://doi.
011-2555-2. org/10.1177/0269881113509904.
[98] S.E. Shoelson, J. Lee, A.B. Goldfine, Inflammation and insulin resistance, J. Clin. [123] S. Kajimura, P. Seale, B.M. Spiegelman, Transcriptional control of brown fat de-
Invest. 116 (7) (2006) 1793–1801, https://doi.org/10.1172/JCI29069. velopment, Cell Metab. 11 (4) (2010) 257–262, https://doi.org/10.1016/j.cmet.
[99] D. Cui, et al., Macrophage migration inhibitory factor mediates metabolic dys- 2010.03.005.
function induced by atypical antipsychotic therapy, J. Clin. Invest. 128 (11) [124] S. Kajimura, et al., Initiation of myoblast to brown fat switch by a PRDM16-C/EBP-
(2018) 4997–5007, https://doi.org/10.1172/JCI93090. beta transcriptional complex, Nature 460 (7259) (2009) 1154–1158, https://doi.
[100] R. Coccurello, et al., Olanzapine (LY170053, 2-methyl-4-(4-methyl-1-piperazinyl)- org/10.1038/nature08262.
10H-thieno[2,3-b][1,5] benzodiazepine), but not the novel atypical antipsychotic [125] W.D. van Marken Lichtenbelt, et al., Cold-activated brown adipose tissue in
ST2472 (9-piperazin-1-ylpyrrolo[2,1-b][1,3]benzothiazepine), chronic adminis- healthy men, N. Engl. J. Med. 360 (15) (2009) 1500–1508, https://doi.org/10.
tration induces weight gain, hyperphagia, and metabolic dysregulation in mice, J. 1056/NEJMoa0808718.
Pharmacol. Exp. Ther. 326 (3) (2008) 905–911, https://doi.org/10.1124/jpet. [126] J. Orava, et al., Different metabolic responses of human brown adipose tissue to
108.137240. activation by cold and insulin, Cell Metab. 14 (2) (2011) 272–279, https://doi.
[101] P.H. Hou, et al., Long-term administration of olanzapine induces adiposity and org/10.1016/j.cmet.2011.06.012.
increases hepatic fatty acid desaturation protein in female C57BL/6J mice, Iran J. [127] Y.H. Tseng, et al., New role of bone morphogenetic protein 7 in brown adipo-
Basic Med. Sci. 21 (5) (2018) 495–501, https://doi.org/10.22038/IJBMS.2018. genesis and energy expenditure, Nature 454 (7207) (2008) 1000–1004, https://
22759.5780. doi.org/10.1038/nature07221.
[102] R. Li, et al., The Wnt signaling pathway effector TCF7L2 mediates olanzapine- [128] X. Yang, S. Enerback, U. Smith, Reduced expression of FOXC2 and brown adipo-
induced weight gain and insulin resistance, Front. Pharmacol. 9 (2018) 379, genic genes in human subjects with insulin resistance, Obes. Res. 11 (10) (2003)
https://doi.org/10.3389/fphar.2018.00379. 1182–1191, https://doi.org/10.1038/oby.2003.163.
[103] V.L. Albaugh, et al., Olanzapine promotes fat accumulation in male rats by de- [129] J.E. Oh, et al., Inhibition of mouse brown adipocyte differentiation by second-
creasing physical activity, repartitioning energy and increasing adipose tissue li- generation antipsychotics, Exp. Mol. Med. 44 (9) (2012) 545–553, https://doi.
pogenesis while impairing lipolysis, Mol. Psychiatry 16 (5) (2011) 569–581, org/10.3858/emm.2012.44.9.062.
https://doi.org/10.1038/mp.2010.33. [130] A.M. Valverde, M. Benito, The brown adipose cell: a unique model for under-
[104] X. Zhang, et al., Regulation of obesity-associated metabolic disturbance by the standing the molecular mechanism of insulin resistance, Mini Rev. Med. Chem. 5
antipsychotic drug olanzapine: role of the autophagy-lysosome pathway, Biochem. (3) (2005) 269–278.
Pharmacol. 158 (2018) 114–125, https://doi.org/10.1016/j.bcp.2018.10.001. [131] M. Ota, et al., Resistance to excessive bodyweight gain in risperidone-injected rats,
[105] A.M. Kunnari, et al., The expression of human resistin in different leucocyte Clin. Exp. Pharmacol. Physiol. 32 (4) (2005) 279–287, https://doi.org/10.1111/j.
lineages is modulated by LPS and TNFalpha, Regul. Pept. 157 (1–3) (2009) 57–63, 1440-1681.2005.04184.x.
https://doi.org/10.1016/j.regpep.2009.05.002. [132] Q. Zhang, et al., Olanzapine reduced brown adipose tissue thermogenesis and
[106] D.R. Schwartz, M.A. Lazar, Human resistin: found in translation from mouse to locomotor activity in female rats, Prog. Neuro-Psychopharmacol. Biol. Psychiatry
man, Trends Endocrinol. Metab. 22 (7) (2011) 259–265, https://doi.org/10.1016/ 51 (2014) 172–180, https://doi.org/10.1016/j.pnpbp.2014.02.003.
j.tem.2011.03.005. [133] Y. Li, et al., Secretin-activated brown fat mediates prandial thermogenesis to in-
[107] J.P. Klemettila, et al., Resistin as an inflammatory marker in patients with schi- duce satiation, Cell 175 (6) (2018) 1561–1574 (e12), https://doi.org/10.1016/j.
zophrenia treated with clozapine, Nord. J. Psychiatry 71 (2) (2017) 89–95, cell.2018.10.016.
https://doi.org/10.1080/08039488.2016.1230649. [134] K.A. Virtanen, et al., Functional brown adipose tissue in healthy adults, N. Engl. J.
[108] M. Sapra, et al., Adiposity-independent hypoadiponectinemia as a potential Med. 360 (15) (2009) 1518–1525, https://doi.org/10.1056/NEJMoa0808949.
marker of insulin resistance and inflammation in schizophrenia patients treated [135] L.J. Robinson, et al., Brown adipose tissue activation as measured by infrared
with second generation antipsychotics, Schizophr. Res. 174 (1–3) (2016) 132–136, thermography by mild anticipatory psychological stress in lean healthy females,
https://doi.org/10.1016/j.schres.2016.04.051. Exp. Physiol. 101 (4) (2016) 549–557, https://doi.org/10.1113/EP085642.
[109] M. Wampers, et al., Differential effects of olanzapine and risperidone on plasma [136] W.W. Blessing, et al., Clozapine, chlorpromazine and risperidone dose-depen-
adiponectin levels over time: results from a 3-month prospective open-label study, dently reduce emotional hyperthermia, a biological marker of salience,
Eur. Neuropsychopharmacol. 22 (1) (2012) 17–26, https://doi.org/10.1016/j. Psychopharmacology 234 (21) (2017) 3259–3269, https://doi.org/10.1007/
euroneuro.2011.03.010. s00213-017-4710-x.
[110] A.A. Richards, et al., Olanzapine treatment is associated with reduced high mo- [137] M. van den Buuse, Acute effects of antipsychotic drugs on cardiovascular re-
lecular weight adiponectin in serum: a potential mechanism for olanzapine-in- sponses to stress, Eur. J. Pharmacol. 464 (1) (2003) 55–62.
duced insulin resistance in patients with schizophrenia, J. Clin. Psychopharmacol. [138] N.R. Swerdlow, et al., Effects of spiperone, raclopride, SCH 23390 and clozapine
26 (3) (2006) 232–237, https://doi.org/10.1097/01.jcp.0000218404.64619.52. on apomorphine inhibition of sensorimotor gating of the startle response in the
[111] T.A. Cohn, et al., Insulin resistance and adiponectin levels in drug-free patients rat, J. Pharmacol. Exp. Ther. 256 (2) (1991) 530–536.
with schizophrenia: a preliminary report, Can. J. Psychiatry 51 (6) (2006) [139] M.A. Geyer, et al., Pharmacological studies of prepulse inhibition models of sen-
382–386, https://doi.org/10.1177/070674370605100608. sorimotor gating deficits in schizophrenia: a decade in review,
[112] L.J. van Nimwegen, et al., Hepatic insulin resistance in antipsychotic naive schi- Psychopharmacology 156 (2–3) (2001) 117–154.
zophrenic patients: stable isotope studies of glucose metabolism, J. Clin. [140] J. Kollias, R.W. Bullard, The influence of chlorpromazine on physical and chemical
Endocrinol. Metab. 93 (2) (2008) 572–577, https://doi.org/10.1210/jc.2007- mechanisms of temperature regulation in the rat, J. Pharmacol. Exp. Ther. 145
1167. (1964) 373–381.
[113] A. Xu, et al., Testosterone selectively reduces the high molecular weight form of [141] N.R. Swerdlow, et al., Seroquel, clozapine and chlorpromazine restore sensor-
adiponectin by inhibiting its secretion from adipocytes, J. Biol. Chem. 280 (18) imotor gating in ketamine-treated rats, Psychopharmacology 140 (1) (1998)
(2005) 18073–18080, https://doi.org/10.1074/jbc.M414231200. 75–80.
[114] K. O'Connell, J. Thakore, K.K. Dev, Levels of S100B are raised in female patients [142] R.E. Chipkin, Effects of D1 and D2 antagonists on basal and apomorphine de-
with schizophrenia, BMC Psychiatry 13 (2013) 146, https://doi.org/10.1186/ creased body temperature in mice and rats, Pharmacol. Biochem. Behav. 30 (3)
1471-244X-13-146. (1988) 683–686.
[115] J. Steiner, et al., S100B serum levels in schizophrenia are presumably related to [143] W.W. Blessing, Y. Ootsuka, Activation of dopamine D2 receptors in the CNS in-
visceral obesity and insulin resistance, Cardiovasc. Psychiatry Neurol. (2010) hibits sympathetic cutaneous vasomotor alerting responses (SCVARs), con-
480707, https://doi.org/10.1155/2010/480707. tributing to clozapine’s SCVAR-inhibiting action, Prog. Neuro-Psychopharmacol.
[116] Y. Liu, D.C. Buck, K.A. Neve, Novel interaction of the dopamine D2 receptor and Biol. Psychiatry 31 (2) (2007) 328–336, https://doi.org/10.1016/j.pnpbp.2006.
the Ca2+ binding protein S100B: role in D2 receptor function, Mol. Pharmacol. 74 09.005.
(2) (2008) 371–378, https://doi.org/10.1124/mol.108.044925. [144] S.F. Morrison, C.J. Madden, Central nervous system regulation of brown adipose
[117] M.K. Hahn, et al., Acute effects of single-dose olanzapine on metabolic, endocrine, tissue, Compr. Physiol. 4 (4) (2014) 1677–1713, https://doi.org/10.1002/cphy.
and inflammatory markers in healthy controls, J. Clin. Psychopharmacol. 33 (6) c140013.
(2013) 740–746, https://doi.org/10.1097/JCP.0b013e31829e8333. [145] W. Blessing, R. McAllen, M. McKinley, Control of the cutaneous circulation by the
[118] J.S. Samra, et al., Effects of physiological hypercortisolemia on the regulation of central nervous system, Compr. Physiol. 6 (3) (2016) 1161–1197, https://doi.org/
lipolysis in subcutaneous adipose tissue, J. Clin. Endocrinol. Metab. 83 (2) (1998) 10.1002/cphy.c150034.
626–631, https://doi.org/10.1210/jcem.83.2.4547. [146] C.J. Madden., et al., alpha2 Adrenergic receptor-mediated inhibition of thermo-
[119] S. Vidarsdottir, et al., Orally disintegrating and oral standard olanzapine tablets genesis, J. Neurosci. 33 (5) (2013) 2017–2028, https://doi.org/10.1523/
similarly elevate the homeostasis model assessment of insulin resistance index and JNEUROSCI.4701-12.2013.
plasma triglyceride levels in 12 healthy men: a randomized crossover study, J. [147] B. Lkhagvasuren, T. Oka, The histaminergic system is involved in psychological
Clin. Psychiatry 71 (9) (2010) 1205–1211, https://doi.org/10.4088/JCP. stress-induced hyperthermia in rats, Physiol. Rep. 5 (8) (2017), https://doi.org/
08m04654yel. 10.14814/phy2.13204.
[120] R.J. Fountaine, et al., Increased food intake and energy expenditure following [148] J. Ferno, et al., Olanzapine depot exposure in male rats: dose-dependent lipogenic
administration of olanzapine to healthy men, Obesity (Silver Spring) 18 (8) (2010) effects without concomitant weight gain, Eur. Neuropsychopharmacol. 25 (6)
1646–1651, https://doi.org/10.1038/oby.2010.6. (2015) 923–932, https://doi.org/10.1016/j.euroneuro.2015.03.002.
[121] G.E. Nicol, et al., Metabolic effects of antipsychotics on adiposity and insulin [149] S. Skrede, et al., Olanzapine depot formulation in rat: a step forward in modelling
sensitivity in youths: a randomized clinical trial, JAMA Psychiatry 75 (8) (2018) antipsychotic-induced metabolic adverse effects, Int. J. Neuropsychopharmacol.

13
V. Ferreira, et al. BBA - Molecular and Cell Biology of Lipids 1865 (2020) 158534

17 (1) (2014) 91–104, https://doi.org/10.1017/S1461145713000862. proliferation and endothelial differentiation through the activation of the PI3K/
[150] A. Stefanidis, et al., The role of thermogenesis in antipsychotic drug-induced Akt/mTOR and AMPK pathways, Int. J. Mol. Med. 37 (5) (2016) 1299–1309,
weight gain, Obesity (Silver Spring) 17 (1) (2009) 16–24, https://doi.org/10. https://doi.org/10.3892/ijmm.2016.2542.
1038/oby.2008.468. [166] Y. Okamatsu-Ogura, et al., Possible involvement of uncoupling protein 1 in ap-
[151] S. Skrede, et al., Olanzapine, but not aripiprazole, weight-independently elevates petite control by leptin, Exp. Biol. Med. (Maywood) 236 (11) (2011) 1274–1281,
serum triglycerides and activates lipogenic gene expression in female rats, Int. J. https://doi.org/10.1258/ebm.2011.011143.
Neuropsychopharmacol. 15 (2) (2012) 163–179, https://doi.org/10.1017/ [167] P. Monteleone, et al., Pronounced early increase in circulating leptin predicts a
S1461145711001271. lower weight gain during clozapine treatment, J. Clin. Psychopharmacol. 22 (4)
[152] B.J. Oldfield, et al., The neurochemical characterisation of hypothalamic pathways (2002) 424–426.
projecting polysynaptically to brown adipose tissue in the rat, Neuroscience 110 [168] A.P. Arruda, et al., Low-grade hypothalamic inflammation leads to defective
(3) (2002) 515–526. thermogenesis, insulin resistance, and impaired insulin secretion, Endocrinology
[153] J. Fadel, M. Bubser, A.Y. Deutch, Differential activation of orexin neurons by 152 (4) (2011) 1314–1326, https://doi.org/10.1210/en.2010-0659.
antipsychotic drugs associated with weight gain, J. Neurosci. 22 (15) (2002) [169] E. Kristof, et al., Clozapine modifies the differentiation program of human adi-
6742–6746 (DOI:20026632). pocytes inducing browning, Transl. Psychiatry 6 (11) (2016) e963, https://doi.
[154] B.J. Oldfield, et al., Lateral hypothalamic ‘command neurons’ with axonal pro- org/10.1038/tp.2016.230.
jections to regions involved in both feeding and thermogenesis, Eur. J. Neurosci. [170] I. Elman, et al., Effects of risperidone on the peripheral noradrenegic system in
25 (8) (2007) 2404–2412, https://doi.org/10.1111/j.1460-9568.2007.05429.x. patients with schizophrenia: a comparison with clozapine and placebo,
[155] G.V. Allen, D.F. Cechetto, Functional and anatomical organization of cardiovas- Neuropsychopharmacology 27 (2) (2002) 293–300, https://doi.org/10.1016/
cular pressor and depressor sites in the lateral hypothalamic area: I. Descending S0893-133X(02)00314-7.
projections, J. Comp. Neurol. 315 (3) (1992) 313–332, https://doi.org/10.1002/ [171] Y. Dwivedi, H.S. Rizavi, G.N. Pandey, Differential effects of haloperidol and clo-
cne.903150307. zapine on [(3)H]cAMP binding, protein kinase A (PKA) activity, and mRNA and
[156] X. Zeng, et al., Innervation of thermogenic adipose tissue via a calsyntenin 3beta- protein expression of selective regulatory and catalytic subunit isoforms of PKA in
S100b axis, Nature 569 (7755) (2019) 229–235, https://doi.org/10.1038/s41586- rat brain, J. Pharmacol. Exp. Ther. 301 (1) (2002) 197–209, https://doi.org/10.
019-1156-9. 1124/jpet.301.1.197.
[157] R. Donato, S100: a multigenic family of calcium-modulated proteins of the EF- [172] J.D. Crane, et al., Inhibiting peripheral serotonin synthesis reduces obesity and
hand type with intracellular and extracellular functional roles, Int. J. Biochem. metabolic dysfunction by promoting brown adipose tissue thermogenesis, Nat.
Cell Biol. 33 (7) (2001) 637–668, https://doi.org/10.1016/s1357-2725(01) Med. 21 (2) (2015) 166–172, https://doi.org/10.1038/nm.3766.
00046-2. [173] C.M. Oh, et al., Regulation of systemic energy homeostasis by serotonin in adipose
[158] F. Villarroya, et al., Inflammation of brown/beige adipose tissues in obesity and tissues, Nat. Commun. 6 (2015) 6794, https://doi.org/10.1038/ncomms7794.
metabolic disease, J. Intern. Med. 284 (5) (2018) 492–504, https://doi.org/10. [174] H.J. Kim, et al., Metabolomic analysis of livers and serum from high-fat diet in-
1111/joim.12803. duced obese mice, J. Proteome Res. 10 (2) (2011) 722–731, https://doi.org/10.
[159] M.B. Cope, et al., Risperidone alters food intake, core body temperature, and lo- 1021/pr100892r.
comotor activity in mice, Physiol. Behav. 96 (3) (2009) 457–463, https://doi.org/ [175] S.H. Kwak, et al., Association of variations in TPH1 and HTR2B with gestational
10.1016/j.physbeh.2008.11.011. weight gain and measures of obesity, Obesity (Silver Spring) 20 (1) (2012)
[160] J. Minet-Ringuet, et al., Long term treatment with olanzapine mixed with the food 233–238, https://doi.org/10.1038/oby.2011.253.
in male rats induces body fat deposition with no increase in body weight and no [176] K. Klepac, et al., The Gq signalling pathway inhibits brown and beige adipose
thermogenic alteration, Appetite 46 (3) (2006) 254–262, https://doi.org/10. tissue, Nat. Commun. 7 (2016) 10895, https://doi.org/10.1038/ncomms10895.
1016/j.appet.2006.01.008. [177] C.U. Correll, From receptor pharmacology to improved outcomes: individualising
[161] J. Lian, et al., Preventing olanzapine-induced weight gain using betahistine: a the selection, dosing, and switching of antipsychotics, Eur. Psychiatry 25 (Suppl.
study in a rat model with chronic olanzapine treatment, PLoS One 9 (8) (2014) 2) (2010) S12–S21, https://doi.org/10.1016/S0924-9338(10)71701-6.
e104160, https://doi.org/10.1371/journal.pone.0104160. [178] K.F. Buckland, T.J. Williams, D.M. Conroy, Histamine induces cytoskeletal
[162] N. Martinez-Sanchez, et al., Hypothalamic AMPK-ER stress-JNK1 axis mediates the changes in human eosinophils via the H(4) receptor, Br. J. Pharmacol. 140 (6)
central actions of thyroid hormones on energy balance, Cell Metab. 26 (1) (2017) (2003) 1117–1127, https://doi.org/10.1038/sj.bjp.0705530.
212–229, https://doi.org/10.1016/j.cmet.2017.06.014 (e12). [179] H.D. Lim, et al., Evaluation of histamine H1-, H2-, and H3-receptor ligands at the
[163] K.J. Motyl, et al., Propranolol attenuates risperidone-induced trabecular bone loss human histamine H4 receptor: identification of 4-methylhistamine as the first
in female mice, Endocrinology 156 (7) (2015) 2374–2383, https://doi.org/10. potent and selective H4 receptor agonist, J. Pharmacol. Exp. Ther. 314 (3) (2005)
1210/en.2015-1099. 1310–1321, https://doi.org/10.1124/jpet.105.087965.
[164] R. Saladin, et al., Transient increase in obese gene expression after food intake or [180] Y.X. Zhao, et al., Stimulation of histamine H4 receptor participates in cold-induced
insulin administration, Nature 377 (6549) (1995) 527–529, https://doi.org/10. browning of subcutaneous white adipose tissue, Am. J. Physiol. Endocrinol.
1038/377527a0. Metab. (2019), https://doi.org/10.1152/ajpendo.00131.2019.
[165] J. Cui, et al., Macrophage migration inhibitory factor promotes cardiac stem cell

14