Research Article

Oxytocin signaling in the posterior hypothalamus prevents hyperphagic obesity in mice

RIKEN Center for Biosystems Dynamics Research, Japan
Laboratory of Molecular Biology, Department of Molecular and Cell Biology, Graduate School of Agricultural Science, Tohoku University, Japan
Division of Brain and Neurophysiology, Department of Physiology, Jichi Medical University, Japan
Department of Obesity and Inflammation Research, Fukushima Medical University, Japan
CREST, Japan Science and Technology Agency, Japan

Oct 25, 2022

https://doi.org/10.7554/eLife.75718

Version of Record

Accepted for publication after peer review and revision.

Download
Cite
Share
CommentOpen annotations (there are currently 0 annotations on this page).

Version of Record published: October 25, 2022 (This version)
Accepted: October 12, 2022
Preprint posted: December 1, 2021 (Go to version)
Received: November 20, 2021

1. Of interest
Polygenic risk scores for the prediction of common cancers in East Asians: A population-based prospective cohort study

Peh Joo Ho, Iain BeeHuat Tan ... Jingmei Li

Research Article Mar 27, 2023
Further reading

Article
Figures and data
Abstract
Editor's evaluation
Introduction
Results
Discussion
Materials and methods
Data availability
References
Decision letter
Author response
Article and author information
Metrics

Abstract

Decades of studies have revealed molecular and neural circuit bases for body weight homeostasis. Neural hormone oxytocin (Oxt) has received attention in this context because it is produced by neurons in the paraventricular hypothalamic nucleus (PVH), a known output center of hypothalamic regulation of appetite. Oxt has an anorexigenic effect, as shown in human studies, and can mediate satiety signals in rodents. However, the function of Oxt signaling in the physiological regulation of appetite has remained in question, because whole-body knockout (KO) of Oxt or Oxt receptor (Oxtr) has little effect on food intake. We herein show that acute conditional KO (cKO) of Oxt selectively in the adult PVH, but not in the supraoptic nucleus, markedly increases body weight and food intake, with an elevated level of plasma triglyceride and leptin. Intraperitoneal administration of Oxt rescues the hyperphagic phenotype of the PVH Oxt cKO model. Furthermore, we show that cKO of Oxtr selectively in the posterior hypothalamic regions, especially the arcuate hypothalamic nucleus, a primary center for appetite regulations, phenocopies hyperphagic obesity. Collectively, these data reveal that Oxt signaling in the arcuate nucleus suppresses excessive food intake.

Editor's evaluation

Inada and colleagues report results from a series of studies investigating the role of the neuropeptide oxytocin in regulating food intake and body weight. They used combinations of genetics and behavioral studies to demonstrate that oxytocin deletion results in increased food intake and body weight. They further show that deletion of the oxytocin receptor in the posterior hypothalamus causes a similar increase in food intake and body weight. Collectively, these studies support a role for the oxytocin system as a key regulator of energy balance.

https://doi.org/10.7554/eLife.75718.sa0

Introduction

Appetite is one of the strongest desires in animals. The consumption of nutritious foods is a primitive pleasure for animals because it is essential for survival. Yet, excessive food intake leads to obesity and increases the risk of disease. Understanding the neurobiological bases of appetite regulation is therefore an urgent issue, given that the body mass index of humans has increased dramatically over the last 40 years (NCD Risk Factor Collaboration, 2016).

Decades of studies in rodents have revealed molecular and neural circuit bases for body weight homeostasis (Andermann and Lowell, 2017; Sternson and Eiselt, 2017; Sutton et al., 2016). Classical studies with mechanical or electrical lesioning, as well as recent molecular or genetic dissections, both support the critical roles of appetite regulation by neurons in the arcuate hypothalamic nucleus (ARH), in particular, those expressing orexigenic agouti-related protein (Agrp) and anorexigenic pre-opiomelanocortin (Pomc) (Choi et al., 1999; Ollmann et al., 1997). These neurons receive both direct humoral inputs and neural inputs of interoception (Bai et al., 2019) to regulate food intake antagonistically at various timescales (Krashes et al., 2014; Sternson and Eiselt, 2017). The paraventricular hypothalamic nucleus (PVH) is one of the critical output structures of the primary appetite-regulating ARH neurons. Silencing PVH neurons phenocopies the overeating effect observed in the activation of Agrp neurons, whereas activating PVH neurons ameliorates the overeating caused by the acute activation of Agrp neurons (Atasoy et al., 2012; Garfield et al., 2015). Melanocortin-4 receptor (MC4R)-expressing neurons in the PVH are the key target of ARH Agrp and Pomc neurons. PVH MC4R neurons are activated by α-melanocyte-stimulating hormone provided by Pomc neurons and inhibited by GABAergic Agrp neurons, and are supposed to transmit signals to the downstream target regions in the midbrain and pons (Garfield et al., 2015; Stachniak et al., 2014; Sutton et al., 2016).

Although PVH MC4R neurons have been relatively well documented (Balthasar et al., 2005; Garfield et al., 2015), other PVH cell types may also mediate output signals to control feeding and energy expenditure (Sutton et al., 2016). However, little is known about the organization, cell types, and neurotransmitters by which appetite-regulating signals are conveyed to other brain regions. Neural hormone oxytocin (Oxt), which marks one of the major cell types in the PVH, has received attention in this context (Leng and Sabatier, 2017; Onaka and Takayanagi, 2019). The anorexigenic effect of Oxt has been shown in humans (Lawson et al., 2015; Thienel et al., 2016), and genetic variations of Oxt receptor (Oxtr) have been implicated as a risk factor of overeating (Çatli et al., 2021; Davis et al., 2017). In rodents, Oxt administration has been shown to suppress increases in food intake and body weight (Maejima et al., 2018). Pons-projecting Oxt neurons have been shown to be active following leptin administration (Blevins et al., 2004), and knockdown of Oxtr in the nucleus of the solitary tract has been reported to alter feeding patterns (Ong et al., 2017). In addition, Oxtr-expressing neurons in the ARH have been shown to evoke acute appetite suppression signals when opto- or chemo-genetically activated (Fenselau et al., 2017). Despite the importance of the Oxt-OxtR system in the food intake and homeostasis of body weight (Maejima et al., 2018), knockout (KO) and ablation studies still question such findings (Sutton et al., 2016; Worth and Luckman, 2021). For example, Oxt or Oxtr KO mice showed increased body weight at around 4 months of age (termed late-onset obesity), while their food intake was not different from that of wild-type mice (Camerino, 2009; Takayanagi et al., 2008). Diphtheria toxin-based genetic ablation of Oxt-expressing cells in adult mice increased the body weight of male mice with a high-fat diet, but not those with normal chow, and in both cases, food intake was unaffected (Wu et al., 2012b). To revisit the function and sites of action of Oxt signaling in the regulation of feeding, acute conditional KO (cKO) mouse models would be useful.

Here, we describe Oxt cKO phenotypes related to hyperphagic obesity. Our approach offers the following two advantages over previous studies: (i) the Oxt gene can be knocked out in adult mice, avoiding the influence of possible developmental and genetic compensations (El-Brolosy et al., 2019); and (ii) the manipulation can be restricted to the brain, or even to a single hypothalamic nucleus, providing a resolution that exceeds previous studies. Owing to these advantages, we show that Oxt cKO increases both body weight and food intake. The suppression of overeating and overweighting is predominantly regulated by Oxt neurons in the PVH, leaving Oxt neurons in the supraoptic nucleus (SO) with only a minor role. We further show that Oxtr-expressing neurons in the posterior part of the hypothalamus, especially the ARH, mediate the overeating-suppression signals generated by Oxt neurons.

Results

cKO of PVH Oxt increases body weight and food intake

To examine the necessity of Oxt for the regulation of food intake, we prepared recently validated Oxt^flox/flox mice (Inada et al., 2022). In this line, Cre expression deletes floxed exon 1 of the Oxt gene, resulting in the loss of transcription of Oxt mRNA (Inada et al., 2022). To perform the cKO in PVH Oxt neurons, we first crossed Oxt^flox/flox and Oxt KO (Oxt^–/–) mice and obtained Oxt^flox/– mice. Then, we injected AAV-Cre into the bilateral PVH of 8-week-old Oxt^flox/– male mice (Figure 1A and B). The number of neurons expressing Oxt, visualized by in situ hybridization (ISH), significantly decreased within 3 weeks after the AAV-Cre injection (Figure 1C). The body weight of Oxt^flox/– mice that received AAV-Cre injection started to deviate from the controls at around 3 weeks after the injection (Figure 1D). At 5 weeks after the injection, we compared the body weight of Oxt^flox/– mice that received AAV-Cre injection with the wild-type (Oxt^+/+), Oxt^–/–, and Oxt^flox/– mice that received vehicle injection. We found that AAV-Cre-injected Oxt^flox/– mice were heavier than those in the other groups (Figure 1E and F). Importantly, this increase in body weight was considered unlikely to be a reflection of late-onset obesity, as previously reported (Camerino, 2009; Takayanagi et al., 2008), because we did not find a significant difference between the wild-type and Oxt^–/– mice (Figure 1F). Next, we analyzed the relationship between the number of remaining Oxt+ neurons and body weight. We found that mice with a fewer number of remaining Oxt+ neurons showed a heavier body weight (Figure 1G). We also found an increase in food intake: the daily food intake of Oxt^flox/– mice that received AAV-Cre injection was significantly larger at >4 weeks after the injection (Figure 1H), and the total food intake during the 5 weeks after the injection was also larger in the mice that received AAV-Cre injection (Figure 1I). Of note, these effects are not due to the nonspecific toxicity of AAV-Cre injection per se, as AAV-Cre injection to wild-type mice did not alter body weight or daily food intake (Figure 1—figure supplement 1A–D). These results demonstrate that the cKO of Oxt evokes increases in both body weight and food intake.

Figure 1 with 1 supplement see all

Download asset Open asset

*Oxytocin (Oxt)* conditional knockout (cKO) in paraventricular hypothalamic nucleus (PVH) induces an increase in body weight and food intake.

(A) Schematic of the virus injection. *AAV-Cre* or vehicle was injected into the bilateral PVH of *Oxt^flox/–* male mice. (B) Representative coronal sections of the PVH from *Oxt^flox/–* mice received vehicle (left) or *AAV-Cre* (right) injection. Data were obtained at 5 weeks after the injection. Magenta and green represent *Oxt* and *Cre* in situ stainings, respectively. Blue, DAPI. Scale bar, 50 μm. (C) The number of remaining *Oxt*+ neurons in the PVH of mice that received *AAV-Cre* injection. **p<0.01, one-way ANOVA with post hoc Tukey’s HSD. N=5 each. (D) Time course of body weight after *AAV-Cre* or vehicle injection. N=6 each. (E) Representative photos of *Oxt^flox/–* mice that received either vehicle (left) or *AAV-Cre* injection (right). Five weeks after the injection. Scale bar, 5 cm. (F) Body weight of wild-type (*+/+*), *Oxt* KO (*–/–*), and *Oxt* cKO (*flox/–*) mice. The weight was measured at 5 weeks after injection of either vehicle or *AAV-Cre*. Note that this time point corresponds to 13 weeks of age. **p<0.01, one-way ANOVA with post hoc Tukey’s HSD. N=10, 7, 9, and 10 for *+/+*, *–/–*, *flox/–* vehicle, and *flox/–* Cre, respectively. (G) Relationship between the number of remaining *Oxt*+ neurons in the PVH and the body weight of *Oxt^flox/–* mice shown in (F). Magenta, exponential fit for the data from both Cre and vehicle. (H) Time course of daily food intake, defined as the average food intake in each week after *AAV-Cre* or vehicle injection. ***p<0.001, Student’s t-test with post hoc Bonferroni correction. N=6 each. (I) Cumulative food intake during the 5 weeks after the injection. ***p<0.001, Student’s t-test. N=6 each.

Error bars, standard error of mean (SEM).

In addition to the PVH, Oxt neurons are also clustered in the SO (Zhang et al., 2021). To examine whether Oxt neurons in the SO also play inhibitory roles on body weight and food intake, we injected AAV-Cre into the bilateral SO of Oxt^flox/– mice (Figure 2A and B). Similar to the PVH, the number of Oxt+ neurons in the SO significantly decreased at around 3 weeks after the injection (Figure 2C). Unlike the PVH, however, neither body weight nor food intake was significantly different compared with controls (Figure 2D–G), and no clear relationship was found between the number of the remaining Oxt+ neurons and body weight (Figure 2E). These results suggest that PVH Oxt neurons predominantly regulate food intake and body weight, and that SO Oxt neurons exert little influence.

Figure 2

Download asset Open asset

*Oxytocin (Oxt)* conditional knockout (cKO) in supraoptic nucleus (SO) has a negligible effect on food intake and body weight.

(A) Schematic of the virus injection. *AAV-Cre* or vehicle was injected into the bilateral SO of *Oxt^flox/–* male mice. (B) Representative coronal sections of left SO from *Oxt^flox/–* mice received vehicle (left) or *AAV-Cre* (right) injection. Five weeks after the injection. Magenta and green represent *Oxt* and *Cre* in situ stainings, respectively. Blue, DAPI. Scale bar, 50 μm. (C) The number of remaining *Oxt*+ neurons in the SO of mice that received *AAV-Cre* injection. **p<0.01, one-way ANOVA with post hoc Tukey’s HSD. N=5 each. (D) The body weight of *Oxt^flox/–* mice did not differ between vehicle or *AAV-Cre* (Student’s t-test). N=6 each. Data were obtained at 5 weeks after the injection. (E) Relationship between the number of remaining *Oxt*+ neurons in the SO and the body weight of *Oxt^flox/–* mice shown in (D). Magenta, exponential fit for the data from both Cre and vehicle. (F) The time course of daily food intake was not statistically different (Student’s t-test with post hoc Bonferroni correction). N=6 each. (G) Cumulative food intake during the 5 weeks after the injection. N=6 each.

Error bars, SEM.

Because of the minor role of SO Oxt neurons, we focused on the PVH Oxt neurons in the following experiments.

Weight of viscera and blood constituents

Increased food intake may influence not only body weight, but also the viscera and blood constituents. To examine these points, we collected internal organs and blood samples from non-fasted Oxt^flox/– mice that had received either AAV-Cre or vehicle injection into the bilateral PVH (Figure 3A). While the weight of the stomach was unchanged (Figure 3B), a significant increase was observed in the weight of the liver in Oxt^flox/– mice with AAV-Cre injection, likely because of the accumulation of fat in the liver (Figure 3C). We next measured the plasma concentration of glucose, triglyceride, and leptin. No significant differences in glucose levels were found (Figure 3D). In turn, the plasma concentrations of triglyceride and leptin were higher in Oxt^flox/– mice that had received AAV-Cre injection than in those that had received vehicle injection (Figure 3D). Of note, a prominent increase in plasma leptin was also reported in the late-onset obesity cases of 6-month-old Oxt KO mice (Camerino, 2009). Our data regarding Oxt cKO showed the plasma leptin phenotype in the earlier stage of 13-week-old mice. These results suggest that the cKO of Oxt affects the homeostasis of viscera and blood constituents.

Figure 3

Download asset Open asset

Weight of viscera and the blood constituents.

(A) Schematic of the virus injection. *AAV-Cre* or vehicle was injected into the bilateral paraventricular hypothalamic nucleus (PVH) of *Oxt^flox/–* male. Data were obtained at 5 weeks after the injection. (B) The weight of the stomach was not statistically different (p>0.5, Student’s t-test. N=9 and 10 for vehicle and Cre, respectively). (C) The weight of the liver was significantly heavier in *AAV-Cre*-injected mice (***p<0.001, Student’s t-test). (D) Plasma glucose (left), triglyceride (middle), and leptin (right) measured in the non-fasted *Oxt^flox/–* mice. **p<0.01, ***p<0.001, Student’s t-test. N=9 and 8 mice for vehicle and Cre, respectively.

Error bars, SEM.

Oxt supplementation partially rescues Oxt cKO

If Oxt cKO caused increases in body weight and food intake with higher plasma triglyceride and leptin, such effects might be mitigated by the external administration of Oxt. This hypothesis is also supported by the fact that intraperitoneal (ip) or intracerebroventricular injection of Oxt has been shown to reduce both body weight and food intake (Maejima et al., 2018). To examine this hypothesis, first, we injected AAV-Cre into the bilateral PVH of Oxt^flox/– mice, and from the next day of the injection, we conducted ip injection of Oxt (100 μL of 500 μM or 1 mM solution) once every 3 days (Figure 4A). Ip injection of the vehicle was used as control. At 5 weeks after the injection of AAV-Cre, Oxt-treated mice showed significantly reduced body weight, even though the number of remaining Oxt+ neurons was not significantly different (Figure 4B and C). We found that both daily food intake at 4–5 weeks and total food intake during the 5 weeks after the injection were also significantly reduced (Figure 4D and E). The reduction of body weight and food intake by our Oxt treatment paradigm was somehow specific to Oxt cKO mice, given that neither reduction of food intake nor body weight was observed in wild-type males (Figure 4—figure supplement 1A–D). No significant improvement in the blood samples was found: both plasma triglyceride and leptin tended to be reduced in Oxt-treated mice, but did not reach the level of statistical significance (Figure 4F). These results suggest that external administration of Oxt can rescue at least the hyperphagic obesity phenotype of Oxt cKO.

Figure 4 with 1 supplement see all

Download asset Open asset

Intraperitoneal (ip) injection of oxytocin (Oxt) partially rescues paraventricular hypothalamic nucleus (PVH) *Oxt* conditional knockout (cKO) phenotypes.

(A) Schematic of the experiments. *AAV-Cre* was injected into the bilateral PVH of *Oxt^flox/–* male mice. Data were obtained 5 weeks after the virus injection. Once every 3 days, the mice received ip injection of vehicle, 50.36 μg of Oxt (condition 1) or 100.72 μg of Oxt (condition 2) (see Materials and methods). (B) The number of remaining *Oxt+* neurons was not statistically different (p>0.9, one-way ANOVA). Cond, condition. N=5 each (same mice across panels **B–E**). (C) Ip injection of Oxt significantly decreased body weight. **p<0.01, one-way ANOVA with post hoc Tukey’s HSD. Cond, condition. (D) Time course of daily food intake. Asterisks (*) denote significant differences for vehicle versus condition 1 and vehicle versus condition 2 (p<0.05, Tukey’s HSD), and hashes (#) denote significant differences for vehicle versus condition 1, vehicle versus condition 2, and condition 1 versus condition 2 (p<0.05, Tukey’s HSD). (E) Cumulative food intake during the 5 weeks after the virus injection was decreased in the mice that received ip injection of Oxt (*p<0.05, one-way ANOVA with post hoc Tukey’s HSD). Cond, condition. (F) Plasma glucose (left), triglyceride (middle), and leptin (right) measured in the non-fasted *Oxt^flox/–* mice. Decreases in triglyceride and leptin on average were found in Oxt-treated mice but did not reach the level of statistical significance (p=0.314 and 0.065 for triglyceride and leptin, respectively, Student’s t-test). N=5 each.

Error bars, SEM.

In addition to the daily food intake that we have examined so far, previous studies showed that mice ate less within several hours after receiving ip injection of Oxt (Arletti et al., 1989; Maejima et al., 2011). To examine whether Oxt^flox/– mice that receive AAV-Cre injection similarly show reduced hourly food intake, we measured food intake after 6 hr of fasting (Figure 4—figure supplement 1E and F). After fasting, the mice received an ip injection of Oxt, and food was provided again (Figure 4—figure supplement 1F). Cumulative food intake was measured at 1, 3, and 5 hr after the placement of food (Figure 4—figure supplement 1F). Although the number of remaining Oxt+ neurons was comparable (Figure 4—figure supplement 1G), Oxt-injected mice ate less (Figure 4—figure supplement 1H). Taken together, ip injection of Oxt appears to reduce food intake on the scale of hours to days, thereby preventing the hyperphagic obesity induced by Oxt cKO in the PVH.

Oxtr-expressing cells in the ARH mediate appetite suppression

Having established the importance of Oxt to suppress hyperphagic obesity, we examined the site of action of Oxt signaling that mediates appetite suppression. To this end, we prepared Oxtr^flox/flox mice, in which the Oxtr gene can be knocked out under Cre expression (Takayanagi et al., 2005). Given that PVH Oxt neurons send dense projection to the hypothalamic nuclei (Yao et al., 2017; Zhang et al., 2021), and that OxtR expression is also found in the hypothalamus (Fenselau et al., 2017; Mitre et al., 2016; Newmaster et al., 2020), we suspected that a fraction of appetite suppression signals is mediated by the other nuclei of the hypothalamus. To test this possibility, we injected AAV-Cre (serotype 9) into the bilateral ‘anterior hypothalamus’, mainly aiming at nuclei such as the anteroventral periventricular nucleus, medial preoptic nucleus medial part (MPNm), MPN lateral part, and medial preoptic area (Figure 5A; see Materials and methods), and the ‘posterior hypothalamus’, containing nuclei such as the dorsomedial nucleus of the hypothalamus, ventromedial hypothalamic nucleus (VMH), lateral hypothalamic area (LHA), and ARH (Figure 5B). AAV-mediated Cre expression roughly covered these nuclei (Figure 5C and D). To examine if Cre expression reduced Oxtr expression, we visualized Oxtr mRNA using the RNAscope assay (see the Materials and methods) (Sato et al., 2020; Wang et al., 2012). As Oxtr expression was observed as a dot-like structure (Figure 5E and F), we counted the number of such RNAscope dots in each DAPI+ cell. In a negative control experiment utilizing Oxtr KO mice, we often detected one or two RNAscope dots in the DAPI+ cells (Figure 5—figure supplement 1A and B). Therefore, we regarded a cell with three or more dots as an Oxtr-expressing cell (Oxtr+; Figure 5G and H, and Figure 5—figure supplement 2A–C). We found that AAV-Cre injection successfully reduced the number of Oxtr+ cells in most of the targeted nuclei (Figure 5G and H). Oxtr cKO in the posterior but not anterior hypothalamus significantly increased body weight (Figure 5I and J). Similarly, a significant increase in food intake was observed in the mice that had received AAV-Cre injection into the posterior hypothalamus (Figure 5K and L).

Figure 5 with 2 supplements see all

Download asset Open asset

*Oxtr* conditional knockout (cKO) in the posterior hypothalamus induces increases in body weight and food intake.

(**A, B**) Schematic of the virus injection. *AAV-Cre* or vehicle was injected into the bilateral anterior or posterior hypothalamus (see Materials and methods) of *Oxtr^flox/flox* male mice. (**C, D**) Representative coronal section showing *Cre* mRNA (green). Blue, DAPI. Scale bar, 50 μm. (E) A representative coronal section showing anteroventral periventricular nucleus (AVPV) and medial preoptic nucleus medial part (MPNm) from a vehicle-injected mouse. *Oxtr* mRNA was visualized by RNAscope (magenta). Blue, DAPI. Scale bar, 30 μm. (F) Projection of a confocal stack in MPNm from a vehicle-injected mouse. Magenta, *Oxtr* mRNA. Blue, DAPI. Scale bar, 5 μm. (**G, H**) Fraction of DAPI+ cells expressing *Oxtr*. Cells showing three or more RNAscope dots were defined as *Oxtr*+ (Figure 5—figure supplement 2). Veh, vehicle. N=5 each. **p<0.01, ***p<0.001, Student’s t-test with Bonferroni correction. Decreases in the MPN lateral part (MPNl), medial preoptic area (MPO), and arcuate hypothalamic nucleus (ARH) on average were found in *AAV-Cre*-injected mice but did not reach the level of statistical significance in Student’s t-test with Bonferroni correction (p=0.045, 0.289, and 0.012, respectively). (**I, J**) Body weight measured at 5 weeks after the injection. **p<0.01, Student’s t-test. Anterior hypothalamus, N=5 and 6 for vehicle and Cre, respectively, and posterior hypothalamus, N=5 and 6 for vehicle and Cre, respectively. (**K, L**) Daily food intake measured at 5 weeks after the injection. ***p<0.001, Student’s t-test. Anterior hypothalamus, N=5 and 6 for vehicle and Cre, respectively, and posterior hypothalamus, N=5 and 6 for vehicle and Cre, respectively. Error bars, SEM.

We next aimed to pinpoint a specific nucleus in the posterior hypothalamus that could suppress hyperphagic obesity. To this end, we injected AAV-Cre (serotype 2) into the ARH or LHA (Figure 6A and B). This serotype of AAV-driven Cre expression was spatially localized (Figure 6C and D) compared with the AAV serotype 9 used in Figure 5. AAV-driven Cre expression reduced Oxtr expression in Oxtr^flox/flox mice (Figure 6E and F). Body weight, daily food intake, and cumulative food intake were significantly greater in the mice that received AAV-Cre injection into the ARH, whereas no significant difference was found in the mice that expressed Cre in the LHA (Figure 6G-L).

Figure 6

Download asset Open asset

*Oxytocin receptor (Oxtr)* expression in the arcuate hypothalamic nucleus (ARH) suppresses body weight and food intake.

(**A, B**) Schematic of the virus injection. *AAV-Cre* or vehicle was injected into the bilateral ARH or lateral hypothalamic area (LHA) of *Oxtr^flox/flox* male mice. (**C, D**) Representative coronal section showing *Cre* mRNA (green). Blue, DAPI. Scale bar, 50 μm. (**E, F**) Left, representative coronal section showing the ARH or LHA from a vehicle-injected mouse. *Oxtr* mRNA was visualized by RNAscope (magenta). Blue, DAPI. Scale bar, 5 μm. Right, fraction of DAPI+ cells expressing *Oxtr* in the ARH (E) or LHA (F) and ventromedial hypothalamic nucleus (VMH), a neighboring nucleus of the ARH and LHA. Cells showing three or more RNAscope dots were defined as *Oxtr*+. N=5 each. ***p<0.001, Student’s t-test with Bonferroni correction. (**G, H**) Body weight measured at 5 weeks after the injection. *p<0.05, Student’s t-test. N=5 each. (**I, J**) Daily food intake measured at 5 weeks after the injection. **p<0.01, Student’s t-test. N=5 each. (**K, L**) Cumulative food intake during the 5 weeks after the injection. *p<0.05, Student’s t-test. N=5 each.

Error bars, SEM.

Taken together, these results indicate that a fraction of the appetite suppression signals from Oxt neurons is mediated by Oxtr-expressing cells in the posterior hypothalamus, especially those in the ARH.

Discussion

Oxt cKO increased body weight and food intake

In this study, we performed cKO of the Oxt gene by injecting AAV-Cre, which enabled region-specific KO of Oxt. By this advantage, we showed that Oxt produced by PVH Oxt neurons contributes to the regulation of body weight and food intake, whereas that by SO Oxt neurons does not (Figures 1 and 2). These data extend the previous results that mechanical disruption of PVH in rats increased both body weight and food intake (Shor-Posner et al., 1985; Sims and Lorden, 1986). In contrast to the Oxt cKO phenotype (Figure 1H and I), whole-body Oxt KO mice showed a normal amount of food intake (Camerino, 2009), suggesting compensational mechanisms. For example, when a certain gene is knocked out, expression of the related gene(s) is enhanced to compensate for some of the KO phenotypes functionally (El-Brolosy et al., 2019; Ma et al., 2019). Transcriptomic analysis between Oxt KO and wild type may reveal a more complete picture of gene expression that can explain the compensational mechanisms in Oxt KO mice.

The phenotypic discrepancy between our data and diphtheria toxin-induced ablation of Oxt cells (Wu et al., 2012b) might be due to the loss of Oxt cells outside the PVH (maybe even outside the brain; Paiva et al., 2021) that somehow elicited appetite suppression, and therefore counterbalanced the overeating phenotype caused by the loss of Oxt in the PVH. Alternatively, neuropeptides or neurotransmitters other than Oxt expressed in the PVH Oxt neurons might have conveyed appetite-stimulating signals, which remained intact in our Oxt-selective cKO model, but were disrupted in the cell-based ablation, resulting in only the overeating phenotype to appear in our case. Regardless of the scenario, our data establish the necessity of Oxt in the PVH to suppress overeating and suggest the presence of a hormone-based output pathway of PVH appetite regulation signals, in addition to the well-established neural pathways mediated by MC4R neurons (Garfield et al., 2015; Stachniak et al., 2014; Sutton et al., 2016).

Downstream of Oxt neurons that mediate appetite suppression signals

After eating a sufficient amount of food, animals stop eating owing to the appetite suppression signals mediated in the brain. Several brain regions and cell types, such as Pomc-expressing neurons in the ARH (Ollmann et al., 1997; Sternson and Eiselt, 2017; Sutton et al., 2016), glutamatergic Oxtr-expressing neurons in the ARH (Fenselau et al., 2017), and calcitonin gene-related peptide expressing neurons in the parabrachial nucleus (Carter et al., 2013; Wu et al., 2012a), have been identified in this process. Our data showed that PVH Oxt neurons mediated appetite suppression signals. Previous studies have shown that both oxytocinergic neurites and Oxtr-expressing neurons can be found in various brain and spinal cord regions (Jurek and Neumann, 2018; Lefevre et al., 2021; Newmaster et al., 2020; Oti et al., 2021; Zhang et al., 2021). In the present study, by AAV-Cre-mediated cKO, we found that Oxtr expressed by neurons in the posterior hypothalamic regions, especially those in the ARH, mediates appetite suppression signals (Figures 5 and 6). Our data are, therefore, generally consistent with the view that Oxtr-expressing neurons in the ARH evoke satiety signaling (Fenselau et al., 2017; Maejima et al., 2014); however, we do not exclude the possibility that OxtR in the other parts of the posterior hypothalamus, such as the VMH (Leng et al., 2008; Viskaitis et al., 2017) and medulla (Ong et al., 2017), also contributes to appetite suppression. Collectively, we suggest that one of the output pathways of the PVH for body weight homeostasis is mediated by Oxt signaling-based modulation of other hypothalamic appetite regulation systems.

We also showed that ip administration of Oxt can mitigate the overeating phenotype caused by the PVH Oxt cKO model (Figure 4). Together with the importance of OxtR signaling in the ARH, one possibility is that the primary hypothalamic neurons that are located outside the blood-brain barrier (Yulyaningsih et al., 2017) directly receive ip-injected Oxt and transmit appetite suppression signals. Alternatively, OxtR-expressing neurons in the peripheral nervous system, such as those in the vagal sensory neurons transmitting intentional appetite suppression signals (Bai et al., 2019), may indirectly modify feeding. Future studies should further dissect the responsible cell types and physiological functions of OxtR signaling in the ARH. Recent advances in the real-time imaging of OxtR activities, for example, with a circularly permuted green fluorescent protein binding to OxtR (Ino et al., 2022; Qian et al., 2022), would be useful for delineating the circuit mechanism and spatiotemporal dynamics of the Oxt-mediated suppression of hyperphagic obesity, such as by pinpointing the site of Oxt release.

Materials and methods

Key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Strain, strain background (mouse, male)	Oxt KO	Inada et al., 2022	#CDB0204E
Strain, strain background (mouse, male)	Oxt cKO (floxed)	Inada et al., 2022	#CDB0116E
Strain, strain background (mouse, male)	Oxtr^flox/flox	Takayanagi et al., 2005
Recombinant DNA reagent	AAV9-hSyn-Cre	Addgene	RRID:Addgene_105555-AAV9
Recombinant DNA reagent	AAV2-CMV-Cre-GFP	University of North Carolina viral core		https://www.med.unc.edu/genetherapy/vectorcore/in- stock-aav-vectors/reporter-vectors/
Commercial assay or kit	RNAscope Multiplex Fluorescent Reagent Kit	Advance Cell Diagnostics	323110
Commercial assay or kit	RNAscope Mm-OXTR	Advance Cell Diagnostics	412171
Software, algorithm	Igor Pro	Wavemetrics	RRID: SCR_000325
Software, algorithm	ImageJ	NIH	RRID: SCR_003070

Animals

All experiments were conducted with virgin male mice. Animals were housed under a 12 hr light/12 hr dark cycle with ad libitum access to water and standard mouse pellets (MFG; Oriental Yeast, Shiga, Japan; 3.57 kcal/g). Wild-type C57BL/6J mice were purchased from Japan SLC (Hamamatsu, Japan). Oxt KO (Accession No. CDB0204E) and cKO (Accession No. CDB0116E) lines (listed at http://www2.clst.riken.jp/arg/mutant%20mice%20list.html) were generated and validated previously (Inada et al., 2022). The Oxtr^flox/flox mouse line has been described (Takayanagi et al., 2005). Oxtr KO mice were generated by injecting Cre mRNA into the Oxtr^flox/flox zygotes. We used only the mice that had the deletion allele without the flox allele by genotype PCR in the analysis. We confirmed the result of Figure 5—figure supplement 2 in a small number of Oxtr KO mice that had been generated by conventional crossing from Oxtr flox mice. All animal procedures followed the animal care guidelines approved by the Institutional Animal Care and Use Committee of the RIKEN Kobe branch.

We chose the Oxt flox/null model to increase the efficiency of cKO (Figures 1—4). If we had used the Oxt flox/flox mice for cKO, because of high Oxt gene expression levels, a small fraction of the flox alleles that do not experience recombination would easily mask the phenotypes. Because flox/null alone (without Cre) has no phenotype (Figure 1F), we could justify the use of the flox/null model. Regarding Oxtr cKO (Figures 5 and 6), we chose the flox/flox model because the haploinsufficiency gene effect of Oxtr has been reported, at least in the context of social behaviors (Sala et al., 2013).

Stereotactic viral injections

Request a detailed protocol

We obtained the AAV serotype 9 hSyn-Cre from Addgene (#105555; titer: 2.3×10¹³ genome particles/mL) and the AAV serotype 2 CMV-Cre-GFP from the University of North Carolina viral core (7.1×10¹² genome particles/mL). To target the AAV or saline (vehicle) into a specific brain region, stereotactic coordinates were defined for each brain region based on the Allen Mouse Brain Atlas (Lein et al., 2007). Mice were anesthetized with 65 mg/kg ketamine (Daiichi Sankyo, Tokyo, Japan) and 13 mg/kg xylazine (X1251; Sigma-Aldrich) via ip injection and head-fixed to stereotactic equipment (Narishige, Tokyo, Japan). The following coordinates were used (in mm from the bregma for anteroposterior [AP], mediolateral [ML], and dorsoventral [DV]): PVH, AP –0.8, ML 0.2, DV 4.5; SO, AP –0.7, ML 1.2, DV 5.5; LHA, AP –2.0, ML 1.2, DV 5.2; ARH, AP –2.0, ML 0.2, DV 5.8. We defined the anterior and posterior hypothalamus by the following coordinates: anterior, AP +0.2, ML 0.2, DV 5.2; posterior, AP –1.7, ML 1.0, DV 5.2. The injected volume of AAV was 200 nL at a speed of 50 nL/min. After viral injection, the animal was returned to the home cage. In Figure 4, 100 μL of Oxt (1910, Tocris) dissolved in saline (vehicle) at 500 μM or 1 mM was ip-injected once every 3 days from the next day of AAV-Cre injection.

Measurement of food intake

Request a detailed protocol

Food intake was measured by placing pre-weighted food pellets on the plate of a cage and reweighing them. In all the experiments, except those in Figure 4—figure supplement 1E–H, daily food intake was measured as follows: 200 g of food pellets were placed and food intake was measured once a week (weekly food intake). Daily food intake was calculated by dividing the weekly food intake by 7 (days) and reported with significance digits of 0.1 g. In Figure 4—figure supplement 1E–H, after 6 hr of fasting, 100 μL of Oxt dissolved in saline (vehicle) at 500 μM or 1 mM was ip-injected. Then, 80.0 g of food was placed and food intake was measured in units of 0.1 g after 1, 3, and 5 hr.

Fluorescent ISH

Request a detailed protocol

Fluorescent ISH was performed as previously described (Inada et al., 2022; Ishii et al., 2017). In brief, mice were anesthetized with sodium pentobarbital and perfused with PBS followed by 4% PFA in PBS. The brain was post-fixed with 4% PFA overnight. Twenty µm coronal brain sections were made using a cryostat (Leica). The following primer sets were used in this study: Cre forward, CCAAGAAGAAGAGGAAGGTGTC; Cre reverse, ATCCCCAGAAATGCCAGATTAC; Oxt forward, AAGGTCGGTCTGGGCCGGAGA; and Oxt reverse, TAAGCCAAGCAGGCAGCAAGC. Fluoromount (K024; Diagnostic BioSystems) was used as a mounting medium. Brain images were acquired using an Olympus BX53 microscope equipped with a ×10 (NA 0.4) objective lens. Cells were counted manually using the ImageJ Cell Counter plugin. In Figures 1C and 2C, cells were counted by an experimenter who was blind to the experimental conditions.

RNAscope assay

Request a detailed protocol

Oxtr mRNA was visualized by the RNAscope Multiplex Fluorescent Reagent Kit (323110; Advance Cell Diagnostics [ACD]) according to the manufacturer’s instructions. In brief, 20 µm coronal brain sections were made using a cryostat (Leica). A probe against Oxtr (Mm-OXTR, 412171, ACD) was hybridized in a HybEZ Oven (ACD) for 2 hr at 40°C. Then, the sections were treated with TSA-plus Cyanine 3 (NEL744001KT; Akoya Biosciences; 1:1500). Fluoromount (K024; Diagnostic BioSystems) was used as a mounting medium. Images subjected to the analysis were acquired using an Olympus BX53 microscope equipped with a ×10 (NA 0.40) or ×20 (NA 0.75) objective lens, as shown in Figures 5E, 6E and F. Figure 5F was obtained with a confocal microscope (LSM780, Zeiss) equipped with a ×63 oil-immersion objective lens (NA 1.40). RNAscope dots were counted manually using the ImageJ Cell Counter plugin.

Measurements of the weight of livers and stomachs

Request a detailed protocol

Mice were anesthetized with sodium pentobarbital. Livers and stomachs were obtained without perfusion and their weight was immediately measured. Before measurement, the stomach was gently pressed to eject the remaining contents.

Plasma measurements

Request a detailed protocol

Mice were fed ad libitum before blood sampling. Mice were anesthetized with isoflurane and blood was collected from the heart. EDTA was used to prevent blood coagulation. Plasma concentrations of glucose, triglycerides, and leptin were measured by the enzymatic method, HK-G6PDH, and ELISA, respectively, through a service provided by Oriental Yeast (Shiga, Japan).

Data analysis

Request a detailed protocol

All mean values are reported as the mean ± SEM. The statistical details of each experiment, including the statistical tests used, the exact value of n, and what n represents, are shown in each figure legend. The p-values are shown in each figure legend or panel; nonsignificant values are not noted. In Figures 1G and 2E, exponential fit was calculated by Igor (WaveMetrics).

Data availability

All data generated or analyzed during this study are included in the manuscript and the figure supplement.

References

1. Andermann ML
2. Lowell BB
(2017) Toward a wiring diagram understanding of appetite control
Neuron 95:757–778.
https://doi.org/10.1016/j.neuron.2017.06.014
- PubMed
- Google Scholar
(1989) Influence of oxytocin on feeding behavior in the rat
Peptides 10:89–93.
https://doi.org/10.1016/0196-9781(89)90082-x
- PubMed
- Google Scholar
1. Atasoy D
2. Betley JN
3. Su HH
4. Sternson SM
(2012) Deconstruction of a neural circuit for hunger
Nature 488:172–177.
https://doi.org/10.1038/nature11270
- PubMed
- Google Scholar
1. Bai L
2. Mesgarzadeh S
3. Ramesh KS
4. Huey EL
5. Liu Y
6. Gray LA
7. Aitken TJ
8. Chen Y
9. Beutler LR
10. Ahn JS
11. Madisen L
12. Zeng H
13. Krasnow MA
14. Knight ZA
(2019) Genetic identification of vagal sensory neurons that control feeding
Cell 179:1129–1143.
https://doi.org/10.1016/j.cell.2019.10.031
- PubMed
- Google Scholar
1. Balthasar N
2. Dalgaard LT
3. Lee CE
4. Yu J
5. Funahashi H
6. Williams T
7. Ferreira M
8. Tang V
9. McGovern RA
10. Kenny CD
11. Christiansen LM
12. Edelstein E
13. Choi B
14. Boss O
15. Aschkenasi C
16. Zhang C
17. Mountjoy K
18. Kishi T
19. Elmquist JK
20. Lowell BB
(2005) Divergence of melanocortin pathways in the control of food intake and energy expenditure
Cell 123:493–505.
https://doi.org/10.1016/j.cell.2005.08.035
- PubMed
- Google Scholar
(2004) Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling meal size
American Journal of Physiology. Regulatory, Integrative and Comparative Physiology 287:R87–R96.
https://doi.org/10.1152/ajpregu.00604.2003
- PubMed
- Google Scholar
1. Camerino C
(2009) Low sympathetic tone and obese phenotype in oxytocin-deficient mice
Obesity 17:980–984.
https://doi.org/10.1038/oby.2009.12
- PubMed
- Google Scholar
(2013) Genetic identification of a neural circuit that suppresses appetite
Nature 503:111–114.
https://doi.org/10.1038/nature12596
- PubMed
- Google Scholar
1. Çatli G
2. Acar S
3. Cingöz G
4. Rasulova K
5. Yarim AK
6. Uzun H
7. Küme T
8. Kızıldağ S
9. Dündar BN
10. Abacı A
(2021) Oxytocin receptor gene polymorphism and low serum oxytocin level are associated with hyperphagia and obesity in adolescents
International Journal of Obesity 45:2064–2073.
https://doi.org/10.1038/s41366-021-00876-5
- PubMed
- Google Scholar
1. Choi S
2. Sparks R
3. Clay M
4. Dallman MF
(1999) Rats with hypothalamic obesity are insensitive to central leptin injections
Endocrinology 140:4426–4433.
https://doi.org/10.1210/endo.140.10.7064
- PubMed
- Google Scholar
1. Davis C
2. Patte K
3. Zai C
4. Kennedy JL
(2017) Polymorphisms of the oxytocin receptor gene and overeating: the intermediary role of endophenotypic risk factors
Nutrition & Diabetes 7:e279.
https://doi.org/10.1038/nutd.2017.24
- PubMed
- Google Scholar
1. El-Brolosy MA
2. Kontarakis Z
3. Rossi A
4. Kuenne C
5. Günther S
6. Fukuda N
7. Kikhi K
8. Boezio GLM
9. Takacs CM
10. Lai S-L
11. Fukuda R
12. Gerri C
13. Giraldez AJ
14. Stainier DYR
(2019) Genetic compensation triggered by mutant mRNA degradation
Nature 568:193–197.
https://doi.org/10.1038/s41586-019-1064-z
- PubMed
- Google Scholar
1. Fenselau H
2. Campbell JN
3. Verstegen AMJ
4. Madara JC
5. Xu J
6. Shah BP
7. Resch JM
8. Yang Z
9. Mandelblat-Cerf Y
10. Livneh Y
11. Lowell BB
(2017) A rapidly acting glutamatergic ARC→PVH satiety circuit postsynaptically regulated by α-MSH
Nature Neuroscience 20:42–51.
https://doi.org/10.1038/nn.4442
- PubMed
- Google Scholar
1. Garfield AS
2. Li C
3. Madara JC
4. Shah BP
5. Webber E
6. Steger JS
7. Campbell JN
8. Gavrilova O
9. Lee CE
10. Olson DP
11. Elmquist JK
12. Tannous BA
13. Krashes MJ
14. Lowell BB
(2015) A neural basis for melanocortin-4 receptor-regulated appetite
Nature Neuroscience 18:863–871.
https://doi.org/10.1038/nn.4011
- PubMed
- Google Scholar
1. Inada K
2. Hagihara M
3. Tsujimoto K
4. Abe T
5. Konno A
6. Hirai H
7. Kiyonari H
8. Miyamichi K
(2022) Plasticity of neural connections underlying oxytocin-mediated parental behaviors of male mice
Neuron 110:2009–2023.
https://doi.org/10.1016/j.neuron.2022.03.033
- PubMed
- Google Scholar
1. Ino D
2. Tanaka Y
3. Hibino H
4. Nishiyama M
(2022) A fluorescent sensor for real-time measurement of extracellular oxytocin dynamics in the brain
Nature Methods 19:1286–1294.
https://doi.org/10.1038/s41592-022-01597-x
- PubMed
- Google Scholar
1. Ishii KK
2. Osakada T
3. Mori H
4. Miyasaka N
5. Yoshihara Y
6. Miyamichi K
7. Touhara K
(2017) A labeled-line neural circuit for pheromone-mediated sexual behaviors in mice
Neuron 95:123–137.
https://doi.org/10.1016/j.neuron.2017.05.038
- PubMed
- Google Scholar
1. Jurek B
2. Neumann ID
(2018) The oxytocin receptor: from intracellular signaling to behavior
Physiological Reviews 98:1805–1908.
https://doi.org/10.1152/physrev.00031.2017
- PubMed
- Google Scholar
1. Krashes MJ
2. Shah BP
3. Madara JC
4. Olson DP
5. Strochlic DE
6. Garfield AS
7. Vong L
8. Pei H
9. Watabe-Uchida M
10. Uchida N
11. Liberles SD
12. Lowell BB
(2014) An excitatory paraventricular nucleus to AGRP neuron circuit that drives hunger
Nature 507:238–242.
https://doi.org/10.1038/nature12956
- PubMed
- Google Scholar
(2015) Oxytocin reduces caloric intake in men
Obesity 23:950–956.
https://doi.org/10.1002/oby.21069
- PubMed
- Google Scholar
(2021) Oxytocinergic feedback circuitries: an anatomical basis for neuromodulation of social behaviors
Frontiers in Neural Circuits 15:688234.
https://doi.org/10.3389/fncir.2021.688234
- PubMed
- Google Scholar
1. Lein ES
2. Hawrylycz MJ
3. Ao N
4. Ayres M
5. Bensinger A
6. Bernard A
7. Boe AF
8. Boguski MS
9. Brockway KS
10. Byrnes EJ
11. Chen L
12. Chen L
13. Chen T-M
14. Chin MC
15. Chong J
16. Crook BE
17. Czaplinska A
18. Dang CN
19. Datta S
20. Dee NR
21. Desaki AL
22. Desta T
23. Diep E
24. Dolbeare TA
25. Donelan MJ
26. Dong H-W
27. Dougherty JG
28. Duncan BJ
29. Ebbert AJ
30. Eichele G
31. Estin LK
32. Faber C
33. Facer BA
34. Fields R
35. Fischer SR
36. Fliss TP
37. Frensley C
38. Gates SN
39. Glattfelder KJ
40. Halverson KR
41. Hart MR
42. Hohmann JG
43. Howell MP
44. Jeung DP
45. Johnson RA
46. Karr PT
47. Kawal R
48. Kidney JM
49. Knapik RH
50. Kuan CL
51. Lake JH
52. Laramee AR
53. Larsen KD
54. Lau C
55. Lemon TA
56. Liang AJ
57. Liu Y
58. Luong LT
59. Michaels J
60. Morgan JJ
61. Morgan RJ
62. Mortrud MT
63. Mosqueda NF
64. Ng LL
65. Ng R
66. Orta GJ
67. Overly CC
68. Pak TH
69. Parry SE
70. Pathak SD
71. Pearson OC
72. Puchalski RB
73. Riley ZL
74. Rockett HR
75. Rowland SA
76. Royall JJ
77. Ruiz MJ
78. Sarno NR
79. Schaffnit K
80. Shapovalova NV
81. Sivisay T
82. Slaughterbeck CR
83. Smith SC
84. Smith KA
85. Smith BI
86. Sodt AJ
87. Stewart NN
88. Stumpf K-R
89. Sunkin SM
90. Sutram M
91. Tam A
92. Teemer CD
93. Thaller C
94. Thompson CL
95. Varnam LR
96. Visel A
97. Whitlock RM
98. Wohnoutka PE
99. Wolkey CK
100. Wong VY
101. Wood M
102. Yaylaoglu MB
103. Young RC
104. Youngstrom BL
105. Yuan XF
106. Zhang B
107. Zwingman TA
108. Jones AR
(2007) Genome-Wide atlas of gene expression in the adult mouse brain
Nature 445:168–176.
https://doi.org/10.1038/nature05453
- PubMed
- Google Scholar
1. Leng G
2. Onaka T
3. Caquineau C
4. Sabatier N
5. Tobin VA
6. Takayanagi Y
(2008) Oxytocin and appetite
Progress in Brain Research 170:137–151.
https://doi.org/10.1016/S0079-6123(08)00413-5
- PubMed
- Google Scholar
1. Leng G
2. Sabatier N
(2017) Oxytocin-the sweet hormone?
Trends in Endocrinology and Metabolism 28:365–376.
https://doi.org/10.1016/j.tem.2017.02.007
- PubMed
- Google Scholar
1. Ma Z
2. Zhu P
3. Shi H
4. Guo L
5. Zhang Q
6. Chen Y
7. Chen S
8. Zhang Z
9. Peng J
10. Chen J
(2019) PTC-bearing mRNA elicits a genetic compensation response via upf3a and COMPASS components
Nature 568:259–263.
https://doi.org/10.1038/s41586-019-1057-y
- PubMed
- Google Scholar
1. Maejima Y
2. Iwasaki Y
3. Yamahara Y
4. Kodaira M
5. Sedbazar U
6. Yada T
(2011) Peripheral oxytocin treatment ameliorates obesity by reducing food intake and visceral fat mass
Aging 3:1169–1177.
https://doi.org/10.18632/aging.100408
- PubMed
- Google Scholar
1. Maejima Y
2. Sakuma K
3. Santoso P
4. Gantulga D
5. Katsurada K
6. Ueta Y
7. Hiraoka Y
8. Nishimori K
9. Tanaka S
10. Shimomura K
11. Yada T
(2014) Oxytocinergic circuit from paraventricular and supraoptic nuclei to arcuate POMC neurons in hypothalamus
FEBS Letters 588:4404–4412.
https://doi.org/10.1016/j.febslet.2014.10.010
- PubMed
- Google Scholar
(2018) The anorexigenic neural pathways of oxytocin and their clinical implication
Neuroendocrinology 107:91–104.
https://doi.org/10.1159/000489263
- PubMed
- Google Scholar
1. Mitre M
2. Marlin BJ
3. Schiavo JK
4. Morina E
5. Norden SE
6. Hackett TA
7. Aoki CJ
8. Chao MV
9. Froemke RC
(2016) A distributed network for social cognition enriched for oxytocin receptors
The Journal of Neuroscience 36:2517–2535.
https://doi.org/10.1523/JNEUROSCI.2409-15.2016
- PubMed
- Google Scholar
1. NCD Risk Factor Collaboration
(2016) Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled analysis of 1698 population-based measurement studies with 19·2 million participants
Lancet 387:1377–1396.
https://doi.org/10.1016/S0140-6736(16)30054-X
- PubMed
- Google Scholar
1. Newmaster KT
2. Nolan ZT
3. Chon U
4. Vanselow DJ
5. Weit AR
6. Tabbaa M
7. Hidema S
8. Nishimori K
9. Hammock EAD
10. Kim Y
(2020) Quantitative cellular-resolution map of the oxytocin receptor in postnatally developing mouse brains
Nature Communications 11:1885.
https://doi.org/10.1038/s41467-020-15659-1
- PubMed
- Google Scholar
1. Ollmann MM
2. Wilson BD
3. Yang YK
4. Kerns JA
5. Chen Y
6. Gantz I
7. Barsh GS
(1997) Antagonism of central melanocortin receptors in vitro and in vivo by agouti-related protein
Science 278:135–138.
https://doi.org/10.1126/science.278.5335.135
- PubMed
- Google Scholar
1. Onaka T
2. Takayanagi Y
(2019) Role of oxytocin in the control of stress and food intake
Journal of Neuroendocrinology 31:e12700.
https://doi.org/10.1111/jne.12700
- PubMed
- Google Scholar
(2017) Effects of endogenous oxytocin receptor signaling in nucleus tractus solitarius on satiation-mediated feeding and thermogenic control in male rats
Endocrinology 158:2826–2836.
https://doi.org/10.1210/en.2017-00200
- PubMed
- Google Scholar
1. Oti T
2. Satoh K
3. Uta D
4. Nagafuchi J
5. Tateishi S
6. Ueda R
7. Takanami K
8. Young LJ
9. Galione A
10. Morris JF
11. Sakamoto T
12. Sakamoto H
(2021) Oxytocin influences male sexual activity via non-synaptic axonal release in the spinal cord
Current Biology 31:103–114.
https://doi.org/10.1016/j.cub.2020.09.089
- PubMed
- Google Scholar
1. Paiva L
2. Lozic M
3. Allchorne A
4. Grinevich V
5. Ludwig M
(2021) Identification of peripheral oxytocin-expressing cells using systemically applied cell-type specific adeno-associated viral vector
Journal of Neuroendocrinology 33:e12970.
https://doi.org/10.1111/jne.12970
- PubMed
- Google Scholar
Preprint
1. Qian T
2. Wang H
3. Wang P
4. Geng L
5. Mei L
6. Osakada T
7. Tang Y
8. Kania A
9. Grinevich V
10. Stoop R
11. Lin D
12. Luo M
13. Li Y
(2022) Compartmental Neuropeptide Release Measured Using a New Oxytocin Sensor
bioRxiv.
https://doi.org/10.1101/2022.02.10.480016
- Google Scholar
1. Sala M
2. Braida D
3. Donzelli A
4. Martucci R
5. Busnelli M
6. Bulgheroni E
7. Rubino T
8. Parolaro D
9. Nishimori K
10. Chini B
(2013) Mice heterozygous for the oxytocin receptor gene (OXTR (+/-)) show impaired social behaviour but not increased aggression or cognitive inflexibility: evidence of a selective haploinsufficiency gene effect
Journal of Neuroendocrinology 25:107–118.
https://doi.org/10.1111/j.1365-2826.2012.02385.x
- PubMed
- Google Scholar
1. Sato K
2. Hamasaki Y
3. Fukui K
4. Ito K
5. Miyamichi K
6. Minami M
7. Amano T
(2020) Amygdalohippocampal area neurons that project to the preoptic area mediate infant-directed attack in male mice
The Journal of Neuroscience 40:3981–3994.
https://doi.org/10.1523/JNEUROSCI.0438-19.2020
- PubMed
- Google Scholar
(1985) Deficits in the control of food intake after hypothalamic paraventricular nucleus lesions
Physiology & Behavior 35:883–890.
https://doi.org/10.1016/0031-9384(85)90255-0
- PubMed
- Google Scholar
1. Sims JS
2. Lorden JF
(1986) Effect of paraventricular nucleus lesions on body weight, food intake and insulin levels
Behavioural Brain Research 22:265–281.
https://doi.org/10.1016/0166-4328(86)90071-9
- PubMed
- Google Scholar
(2014) Chemogenetic synaptic silencing of neural circuits localizes a hypothalamus→midbrain pathway for feeding behavior
Neuron 82:797–808.
https://doi.org/10.1016/j.neuron.2014.04.008
- PubMed
- Google Scholar
1. Sternson SM
2. Eiselt AK
(2017) Three pillars for the neural control of appetite
Annual Review of Physiology 79:401–423.
https://doi.org/10.1146/annurev-physiol-021115-104948
- PubMed
- Google Scholar
(2016) The role of PVH circuits in leptin action and energy balance
Annual Review of Physiology 78:207–221.
https://doi.org/10.1146/annurev-physiol-021115-105347
- PubMed
- Google Scholar
1. Takayanagi Y
2. Yoshida M
3. Bielsky IF
4. Ross HE
5. Kawamata M
6. Onaka T
7. Yanagisawa T
8. Kimura T
9. Matzuk MM
10. Young LJ
11. Nishimori K
(2005) Pervasive social deficits, but normal parturition, in oxytocin receptor-deficient mice
PNAS 102:16096–16101.
https://doi.org/10.1073/pnas.0505312102
- PubMed
- Google Scholar
(2008) Oxytocin receptor-deficient mice developed late-onset obesity
Neuroreport 19:951–955.
https://doi.org/10.1097/WNR.0b013e3283021ca9
- PubMed
- Google Scholar
1. Thienel M
2. Fritsche A
3. Heinrichs M
4. Peter A
5. Ewers M
6. Lehnert H
7. Born J
8. Hallschmid M
(2016) Oxytocin’s inhibitory effect on food intake is stronger in obese than normal-weight men
International Journal of Obesity 40:1707–1714.
https://doi.org/10.1038/ijo.2016.149
- PubMed
- Google Scholar
(2017) Modulation of SF1 neuron activity coordinately regulates both feeding behavior and associated emotional states
Cell Reports 21:3559–3572.
https://doi.org/10.1016/j.celrep.2017.11.089
- PubMed
- Google Scholar
1. Wang F
2. Flanagan J
3. Su N
4. Wang LC
5. Bui S
6. Nielson A
7. Wu X
8. Vo HT
9. Ma XJ
10. Luo Y
(2012) RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues
The Journal of Molecular Diagnostics 14:22–29.
https://doi.org/10.1016/j.jmoldx.2011.08.002
- PubMed
- Google Scholar
1. Worth AA
2. Luckman SM
(2021) Do oxytocin neurones affect feeding?
Journal of Neuroendocrinology 33:e13035.
https://doi.org/10.1111/jne.13035
- PubMed
- Google Scholar
(2012a) Deciphering a neuronal circuit that mediates appetite
Nature 483:594–597.
https://doi.org/10.1038/nature10899
- PubMed
- Google Scholar
1. Wu Z
2. Xu Y
3. Zhu Y
4. Sutton AK
5. Zhao R
6. Lowell BB
7. Olson DP
8. Tong Q
(2012b) An obligate role of oxytocin neurons in diet induced energy expenditure
PLOS ONE 7:e45167.
https://doi.org/10.1371/journal.pone.0045167
- PubMed
- Google Scholar
1. Yao S
2. Bergan J
3. Lanjuin A
4. Dulac C
(2017) Oxytocin signaling in the medial amygdala is required for sex discrimination of social cues
eLife 6:e31373.
https://doi.org/10.7554/eLife.31373
- PubMed
- Google Scholar
1. Yulyaningsih E
2. Rudenko IA
3. Valdearcos M
4. Dahlén E
5. Vagena E
6. Chan A
7. Alvarez-Buylla A
8. Vaisse C
9. Koliwad SK
10. Xu AW
(2017) Acute lesioning and rapid repair of hypothalamic neurons outside the blood-brain barrier
Cell Reports 19:2257–2271.
https://doi.org/10.1016/j.celrep.2017.05.060
- PubMed
- Google Scholar
1. Zhang B
2. Qiu L
3. Xiao W
4. Ni H
5. Chen L
6. Wang F
7. Mai W
8. Wu J
9. Bao A
10. Hu H
11. Gong H
12. Duan S
13. Li A
14. Gao Z
(2021) Reconstruction of the hypothalamo-neurohypophysial system and functional dissection of magnocellular oxytocin neurons in the brain
Neuron 109:331–346.
https://doi.org/10.1016/j.neuron.2020.10.032
- PubMed
- Google Scholar

Decision letter

Joel K Elmquist

Reviewing Editor; University of Texas Southwestern Medical Center, United States
Ma-Li Wong

Senior Editor; State University of New York Upstate Medical University, United States

Our editorial process produces two outputs: (i) public reviews designed to be posted alongside the preprint for the benefit of readers; (ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.

Decision letter after peer review:

Thank you for submitting your article "Oxytocin signaling in the posterior hypothalamus prevents hyperphagic obesity in mice" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Ma-Li Wong as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

The reviewers agree that several technical issues outlined below need to be addressed. In particular, the authors should address the sample sizes noted below.

In addition, more precise description of the anatomic details is needed throughout the manuscript.

Reviewer #1 (Recommendations for the authors):

(1) It is not clear the reason behind using the partial conditional KO mice, OTflox/-. Could you explain why flox/KO was used instead of flox/flox model? If so, these mice are "partial" conditional KO model instead of "bona fida" conditional KO, aren't they? The authors need to clarify this model.

(2) Regarding Figure 4, previous literature has shown that OT injection reduces food intake within an hour. Does OT injection into OT knockout in the PVH mice have the same effect? It is important as these type of experiments can point out that OT may be required for transcriptional changes and/or circuit re-wiring (and others) to regulate food intake.

(3) A large body of studies have shown that the ventrolateral side of the ventromedial hypothalamic nucleus (VMHvl) regulates food intake. In addition, recent optogenetic and chemogenetic studies have shown that the dorsomedial and center part of VMH (VMHdm/c) also can regulate food intake, although many genetic studies (e.g., knockout the important genes such as leptin receptors from the VMHdm/c) do not support this notion. The authors pinpointed that the phenotype in Figure 5 resulted from the ablation of OTR in the arcuate hypothalamic nucleus (ARH) and/or the lateral hypothalamic area (LHA). However, Cre-GFP expression was strongly observed throughout the VMH (Figure 5D). The authors may want to discuss/address the possibility that OTR in the VMH may be involved in OT-induced food suppression effects.

Reviewer #2 (Recommendations for the authors):

This is a straightforward and important study investigating the role of oxytocin signaling on appetite and body weight. By loss of function in adult mice, the authors show a clear role for oxytocin as a inhibitory molecule for overeating.

The major critiques of the study are

1. The sample sizes are typically too small (often n=4), especially in Figure 4 where the effect sizes are smaller.

2. The authors have not ruled out the possibility that viral overexpression of Cre in the PVH is responsible for the hyperphagia. Testing this in the OT-/- and in OT+/+ mice would be prudent. There is no information about the viral titer or surgical procedures, but I suspect high viral titers were sued, which could non-specifically reduced PVH function.

3. The fraction of neurons reported to be OXTR+ seems too high. The authors should determine the background rate of OXTR detection in OXTR-/- mice or in regions known not to expression OXTR and use this a threshold for OXTR expression vs noise e.g., lipofuscin can show up as a false positive in RNAscope and this is determined by measuring if the spot shows broadband fluorescence at other wavelengths.

Reviewer #3 (Recommendations for the authors):

The manuscript by Inada et al. examines the role of hypothalamic oxytocin (OT) signaling in feeding behavior. They demonstrate that conditional knockout of OT in the adult paraventricular hypothalamic nucleus (PVH) increases body weight through increases in food intake, and that conditional knockout of the OT receptor in the posterior hypothalamus has a similar effect. The authors therefore conclude that OT signaling in the posterior hypothalamus, presumably through oxytocin produced in the PVH, contributes to energy balance control.

Strengths:

There has been conflicting literature on the role of OT in feeding behavior. Although pharmacological and genetic approaches have suggested an anorexic effect of OT, knockout of OT or OT receptor has minimal effect on feeding. To address this apparent discrepancy, the authors use conditional knockout models to manipulate OT signaling. This allows not only temporal control of OT and OT receptor, but also allows investigation of signaling in different brain regions (versus, for example, whole body or organ). That the conditional knockout mice display hyperphagia and obesity begins to settle this conflict in the literature.

Weaknesses:

1) There is not a major conceptual advance. The data largely confirm what pharmacological and RNAi knockdown studies have previously demonstrated.

2) The finding that IP injection of OT partially rescues the phenotype of the KO mouse lacks rigor and proper controls.

• Only one dose of OT was used.

• The effect of IP OT in WT mice should be shown. If the IP OT reduces food intake in the WT mouse, then it could be an independent effect.

• Injection of OT within regions of the hypothalamus would provide more information about the localization of these effects.

• How do the authors reconcile the fact that there is no effect on feeding behavior but an effect on body weight, especially when the OT KO has a feeding phenotype?

3) There is little anatomical precision in the manipulation of OT receptors in the posterior hypothalamus." Understanding which of these brain regions (e.g. ARH, VMH, LHA, DMH, others?) is involved in mediating these effects would be very helpful. It is also a missed opportunity to not explore hindbrain OT receptor, as this would be a good positive control (previous literature).

4) Attempts to connect the OT KO data with the OT receptor KO data would greatly enhance this manuscript by delineating a circuit mechanism for central OT-induced anorexia. This could be achieved by giving central injections of OT into target regions of OT KO mice, for example.

Other comments:

1) Unless I am mistaken, the authors cite a biorxiv paper for the "previously-validated" mouse, but this has not been peer reviewed.

2) What dose of OT was used in Figure 4? Please provide in g.

3) How was food intake measured?

4) The authors say in the text that the changes in the weight of the viscera and the blood constituents were independent of body weight. However, the legend says that these measurements were taken 5 weeks post injection, and Figure 1 shows a clear BW phenotype at 5 weeks post-injection. Please clarify.

https://doi.org/10.7554/eLife.75718.sa1

Author response

Essential revisions:

The reviewers agree that several technical issues outlined below need to be addressed. In particular, the authors should address the sample sizes noted below.

In addition, more precise description of the anatomic details is needed throughout the manuscript.

Reviewer #1 (Recommendations for the authors):

(1) It is not clear the reason behind using the partial conditional KO mice, OTflox/-. Could you explain why flox/KO was used instead of flox/flox model? If so, these mice are "partial" conditional KO model instead of "bona fida" conditional KO, aren't they? The authors need to clarify this model.

We thank the reviewer for the positive evaluation of our main aims and conclusions, and for very constructive suggestions. We chose the Oxt flox/null model to increase the efficiency of Cre-mediated KO. If we had used Oxt flox/flox mice, which have very high Oxt gene expression levels, a small fraction of the flox alleles that do not experience recombination would easily mask the body weight and feeding phenotypes. Because flox/null alone (without Cre) has no phenotype (Figure 1F), we could justify the use of the flox/null model. In principle, this situation is comparable to analyzing homozygous mutants by using heterozygotes as a reference, which is common in the genetics of mice and flies. In addition, many influential papers have utilized flox/null models as conditional knockout (for example, Kwan et al. Genesis 39, 10, 2004; Pelosi et al., Plos One 10: e0136422, 2015). In the revised manuscript, we added additional explanations in the Materials and methods to clarify our thoughts (lines 312–318). Regarding Oxtr conditional knockout, we chose the flox/flox model because the haploinsufficiency gene effect of Oxtr has been reported, at least in the context of social behaviors (Sala et al. J Neuroendocrinology 25, 107, 2013).

(2) Regarding Figure 4, previous literature has shown that OT injection reduces food intake within an hour. Does OT injection into OT knockout in the PVH mice have the same effect? It is important as these type of experiments can point out that OT may be required for transcriptional changes and/or circuit re-wiring (and others) to regulate food intake.

We deeply appreciate this important comment. According to this suggestion, we conducted a new experiment to test the short-term effect of IP injection of Oxt. We found that Oxt cKO mice ate significantly less after 1, 3, and 5 hours of Oxt IP injection, consistent with previous literature using wild-type rats or mice. We reported these data in new Figure 4—figure supplement 1E–H.

(3) A large body of studies have shown that the ventrolateral side of the ventromedial hypothalamic nucleus (VMHvl) regulates food intake. In addition, recent optogenetic and chemogenetic studies have shown that the dorsomedial and center part of VMH (VMHdm/c) also can regulate food intake, although many genetic studies (e.g., knockout the important genes such as leptin receptors from the VMHdm/c) do not support this notion. The authors pinpointed that the phenotype in Figure 5 resulted from the ablation of OTR in the arcuate hypothalamic nucleus (ARH) and/or the lateral hypothalamic area (LHA). However, Cre-GFP expression was strongly observed throughout the VMH (Figure 5D). The authors may want to discuss/address the possibility that OTR in the VMH may be involved in OT-induced food suppression effects.

We are greatly thankful for this important comment. We have conducted additional experiments to remove the Oxtr gene with a high spatial resolution (now reported in new Figure 6). To achieve this, we injected serotype 2 of AAV-Cre into the ARH or LHA. As expected, Oxtr expression in the targeted nucleus was significantly reduced, while that in the VMH, a neighboring nucleus, was unaffected (Figure 6E and 6F). We also found an increase in food intake and body weight after injecting the AAV into the ARH, but not into the LHA. Still, these results do not rule out the possibility that VMH contributes to the suppression of hyperphagic obesity. Nevertheless, we appreciate the importance of VMH for feeding regulation and added a discussion regarding VMH in the revised manuscript (lines 278–280).

Reviewer #2 (Recommendations for the authors):

This is a straightforward and important study investigating the role of oxytocin signaling on appetite and body weight. By loss of function in adult mice, the authors show a clear role for oxytocin as a inhibitory molecule for overeating.

The critiques of the study are

1. The sample sizes are typically too small (often n=4), especially in Figure 4 where the effect sizes are smaller.

We are greatly thankful to this reviewer for the positive evaluation of our main aims and conclusions, as well as for the very constructive suggestions. As described in the revised manuscript, we have included additional cohorts of Oxt cKO (Figure 1) and IP injection of Oxt (Figure 4) to collect enough animals. In addition, in Figure 4, we added different doses of Oxt IP injection, which further strengthens our initial conclusion.

2. The authors have not ruled out the possibility that viral overexpression of Cre in the PVH is responsible for the hyperphagia. Testing this in the OT-/- and in OT+/+ mice would be prudent. There is no information about the viral titer or surgical procedures, but I suspect high viral titers were sued, which could non-specifically reduced PVH function.

We appreciate this important comment. According to this suggestion, we injected 200 nL of AAV serotype 9 hSyn-Cre (2.3 x 10¹³ genome particles/mL) into wild-type PVH. Neither body weight nor daily food intake measured at 5 weeks after the injection differed from the saline-injected control. This result is reported in new Figure 1—figure supplement 1, and the experimental conditions are described in the Materials and methods.

3. The fraction of neurons reported to be OXTR+ seems too high. The authors should determine the background rate of OXTR detection in OXTR-/- mice or in regions known not to expression OXTR and use this a threshold for OXTR expression vs noise e.g., lipofuscin can show up as a false positive in RNAscope and this is determined by measuring if the spot shows broadband fluorescence at other wavelengths.

We sincerely thank the reviewer for this important comment. We newly generated Oxtr^–/– mice by injecting Cre mRNA into Oxtr flox/flox zygotes to conduct negative control experiments of RNAscope assay. In agreement with the concern raised by the reviewer, 1 or 2 RNAscope dots were detected in approximately 25% of DAPI+ cells of Oxtr^–/– mice, as reported in new Figure 5—figure supplement 1. This means that Figure 5 of our initial manuscript contained a substantial fraction of pseudo-positive detection of Oxtr. Because we found that defining Oxtr+ cells as a DAPI+ cell with three or more dots provides a near-zero ratio (less than 5%) of pseudo-positive detection of Oxtr+ cells in Oxtr^–/– mice (Figure 5—figure supplement 2), we applied this new definition throughout the paper (Figures 5 and 6). Regardless of this revised definition, our conclusion remains the same, given that AAV-Cre injection into Oxtr^flox/flox mice significantly reduced the number of Oxtr+ cells (Figures 5 and 6).

Reviewer #3 (Recommendations for the authors):

The manuscript by Inada et al. examines the role of hypothalamic oxytocin (OT) signaling in feeding behavior. They demonstrate that conditional knockout of OT in the adult paraventricular hypothalamic nucleus (PVH) increases body weight through increases in food intake, and that conditional knockout of the OT receptor in the posterior hypothalamus has a similar effect. The authors therefore conclude that OT signaling in the posterior hypothalamus, presumably through oxytocin produced in the PVH, contributes to energy balance control.

Strengths:

There has been conflicting literature on the role of OT in feeding behavior. Although pharmacological and genetic approaches have suggested an anorexic effect of OT, knockout of OT or OT receptor has minimal effect on feeding. To address this apparent discrepancy, the authors use conditional knockout models to manipulate OT signaling. This allows not only temporal control of OT and OT receptor, but also allows investigation of signaling in different brain regions (versus, for example, whole body or organ). That the conditional knockout mice display hyperphagia and obesity begins to settle this conflict in the literature.

Weaknesses:

1) There is not a major conceptual advance. The data largely confirm what pharmacological and RNAi knockdown studies have previously demonstrated.

We are sincerely thankful to this reviewer for the positive evaluation of our main aims and conclusions, as well as for the very constructive suggestions. We agree with the view that the main aim of this study is to resolve the conflict between genetic mutant and pharmacological/knockdown studies. However, in the revised manuscript, we have provided additional evidence that OxtR signaling, specifically in the arcuate nucleus, is responsible for suppressing hyperphagia (as reported in new Figure 6). We think that the demonstration of such functional localization of Oxt-OxtR signaling can provide some biological advances, as it cannot be achieved easily by classical pharmacology.

2) The finding that IP injection of OT partially rescues the phenotype of the KO mouse lacks rigor and proper controls.

• Only one dose of OT was used.

• The effect of IP OT in WT mice should be shown. If the IP OT reduces food intake in the WT mouse, then it could be an independent effect.

• Injection of OT within regions of the hypothalamus would provide more information about the localization of these effects.

• How do the authors reconcile the fact that there is no effect on feeding behavior but an effect on body weight, especially when the OT KO has a feeding phenotype?

We are also thankful for this important comment. According to these suggestions, we have included an additional cohort of IP injection of Oxt into Oxt cKO mice to test different doses of Oxt, as reported in Figure 4 in the revised manuscript. Both 50.36 µg and 100.72 µg per injection significantly reduced body weight and food intake in Oxt cKO mice. Importantly, IP injection of the same dose of Oxt into wild-type mice did not affect body weight or food intake, as reported in new Figure 4—figure supplement 1A–D. With the greater number of animals in multiple conditions, the reduced feeding by IP injection of Oxt is more explicitly shown in the revised manuscript. Still, the plasma triglyceride and leptin concentrations were not fully rescued by IP injection of Oxt in Oxt cKO mice, suggesting that the Oxt provided via IP injection is somehow suboptimal in regard to its effect site in the body and/or brain.

Regarding the functional identification of the effective site of Oxt, we decided to conduct local cKO of Oxtr (see response to Reviewer3, point 3 below) instead of local infusion of Oxt into the target brain regions (such as ARH). This is because (i) it is generally difficult to evaluate the diffusion of a solution containing Oxt after local infusion, and (ii) animals need to be anesthetized repeatedly for local infusion of Oxt through a guide cannula, which by itself could affect their feeding patterns.

3) There is little anatomical precision in the manipulation of OT receptors in the posterior hypothalamus." Understanding which of these brain regions (e.g. ARH, VMH, LHA, DMH, others?) is involved in mediating these effects would be very helpful. It is also a missed opportunity to not explore hindbrain OT receptor, as this would be a good positive control (previous literature).

We appreciate this comment. Although precisely controlling Cre activity at the resolution of a single anatomical structure in the brain is not trivial, by using AAV serotype 2, which tends to spread less around the injection site, we conducted ARH- and LHA-specific loss-of-function of Oxtr (new Figure 6). We found that body weight, daily food intake, and cumulative food intake were significantly larger in the mice that received AAV-Cre injection into the ARH, whereas no significant difference was found in the mice that expressed Cre into the LHA. Although we do not exclude the contribution of Oxtr in other brain regions, these data suggest that the majority of effects given rise to by Oxtr cKO in the posterior hypothalamus can be explained by the reduction of Oxtr expression in the ARH.

4) Attempts to connect the OT KO data with the OT receptor KO data would greatly enhance this manuscript by delineating a circuit mechanism for central OT-induced anorexia. This could be achieved by giving central injections of OT into target regions of OT KO mice, for example.

We agree with this view that a circuit mechanism by which PVH Oxt neurons are linked to the posterior hypothalamus Oxtr-expressing neurons would strengthen the manuscript. However, it is highly challenging to pinpoint the site of Oxt release for preventing hyperphagia. Although our Oxtr cKO in the ARH (Figure 6) improved the spatial resolution of the site of Oxt action, this Oxtr signaling may be mediated by diffusion of Oxt from long range, such as cell bodies/dendrites in the PVH or Oxt-positive fibers elsewhere in the brain, rather than direct axonal projection from PVH Oxt neurons to the ARH. As IP-injected Oxt is also functional (Figure 4), in principle, it is possible that circulating Oxt in the bloodstream is responsible for the Oxt signaling in the ARH, presumably via a leaky blood–brain barrier or the tanycytic-mediated active transport of humoral signals. Future studies, including real-time imaging of Oxt ligand release by an Oxt sensor, would further reveal circuit mechanisms. In the revised manuscript, we have improved the spatial resolution of our Oxtr cKO (see our response #3-3). We have also mentioned that the suggested point is an important topic for future research in the Discussion (lines 293–295).

Other comments:

1) Unless I am mistaken, the authors cite a biorxiv paper for the "previously-validated" mouse, but this has not been peer reviewed.

We apologize that the citation was a preprint at the time of initial submission. During the revision, it was updated as follows.

Inada K et al., Neuron 110, 2009-2023.e5, 2022. https://pubmed.ncbi.nlm.nih.gov/35443152/.

We have also updated this citation in the revised manuscript.

2) What dose of OT was used in Figure 4? Please provide in g.

According to this comment, we added a description in Figure 4.

3) How was food intake measured?

We apologize that our description in the Materials and methods was not sufficient. In the revised manuscript, we have provided further details of the food intake measurements (lines 337–345).

4) The authors say in the text that the changes in the weight of the viscera and the blood constituents were independent of body weight. However, the legend says that these measurements were taken 5 weeks post injection, and Figure 1 shows a clear BW phenotype at 5 weeks post-injection. Please clarify.

We do not think that viscera and blood constituents are independent of body weight. We apologize that our initial description was unclear. We have clarified this in the revised manuscript (lines 155–158).

https://doi.org/10.7554/eLife.75718.sa2

Article and author information

Author details

Kengo Inada

RIKEN Center for Biosystems Dynamics Research, Kobe, Japan

Contribution
Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Writing – original draft, Project administration, Writing – review and editing

For correspondence
k.inada.repository@gmail.com

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-0859-4582
Kazoku Tsujimoto

RIKEN Center for Biosystems Dynamics Research, Kobe, Japan

Contribution
Formal analysis

Competing interests
No competing interests declared
Masahide Yoshida
1. Laboratory of Molecular Biology, Department of Molecular and Cell Biology, Graduate School of Agricultural Science, Tohoku University, Sendai, Japan
2. Division of Brain and Neurophysiology, Department of Physiology, Jichi Medical University, Shimotsuke, Japan
Contribution
Resources

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-1247-2294
Katsuhiko Nishimori
1. Laboratory of Molecular Biology, Department of Molecular and Cell Biology, Graduate School of Agricultural Science, Tohoku University, Sendai, Japan
2. Department of Obesity and Inflammation Research, Fukushima Medical University, Fukushima, Japan
Contribution
Resources

Competing interests
No competing interests declared
Kazunari Miyamichi
1. RIKEN Center for Biosystems Dynamics Research, Kobe, Japan
2. CREST, Japan Science and Technology Agency, Kawaguchi, Japan
Contribution
Conceptualization, Supervision, Funding acquisition, Writing – original draft, Project administration, Writing – review and editing

For correspondence
kazunari.miyamichi@riken.jp

Competing interests
No competing interests declared

Additional information
Lead Contact

"This ORCID iD identifies the author of this article:" 0000-0002-7807-8436

Funding

Japan Society for the Promotion of Science (KAKENHI 19J00403)

Kengo Inada

Japan Society for the Promotion of Science (KAKENHI 19K16303)

Kengo Inada

Japan Science and Technology Agency (CREST JPMJCR2021)

Kazunari Miyamichi

Japan Society for the Promotion of Science (KAKENHI 20K20589)

Kazunari Miyamichi

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We wish to thank Mitsue Hagihara for the extensive technical support, the Laboratory for Comprehensive Bioimaging for the microscopy services, and the members of the Miyamichi lab for their comments on an earlier version of this manuscript. This work was supported by the RIKEN Special Postdoctoral Researchers Program, a grant from the Kao Foundation for Arts and Sciences, and JSPS KAKENHI (19J00403 and 19K16303) to KI, and the JST CREST program (JPMJCR2021) and JSPS KAKENHI (20K20589) to KM.

Ethics

All animal procedures followed animal care guidelines approved by the Institutional Animal Care and Use Committee of the RIKEN Kobe branch (#A2017-15-12).

Senior Editor

Ma-Li Wong, State University of New York Upstate Medical University, United States

Reviewing Editor

Joel K Elmquist, University of Texas Southwestern Medical Center, United States

Publication history

Received: November 20, 2021
Preprint posted: December 1, 2021 (view preprint)
Accepted: October 12, 2022
Version of Record published: October 25, 2022 (version 1)

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.