Introduction
Stress is a physical, mental, andemotional response to an external stimulus that could cause adaptationto the changes. Today, the study of stress and its complications isimportant because acute and chronic stresses are among the most importantmental health problems in human society (
Derijk et al., 2008;
Ehteram et al., 2017;
Asalgoo et al., 2015). Stress also plays an importantrole in many health problems, including asthma, cancer, diabetes,and endocrine disorders. These diseases place large financial burdenson healthcare systems every year (
Segerstrom and Miller, 2004;
Salleh, 2008).
Stress responses, the coordinatedresponses that increase one’s chances of survival, include changesin behavior, autonomic system performance, and the secretion of severalhormones such as adrenocorticotropin, cortisol, corticosterone, andcatecholamine. One of the important neuroendocrine system responsesto stress is activation of the hypothalamic–pituitary–adrenal(HPA) axis (
Derijk et al., 2008;
Pourhashemi et al., 2016). A stressful stimulus causes the release of corticotropin-releasinghormone (CRH) and vasopressin (VP) from the hypothalamus paraventricularnucleus (PVN). CRH and VP stimulate the secretion of adrenocorticotropichormone (ACTH) from the anterior pituitary to the blood, which resultsin the production and secretion of glucocorticoids from the adrenalcortex (
Ghobadi et al., 2016; McEwen et al., 2016;
Hassantashet al., 2017). Through their receptors located in theperipheral target organs, glucocorticoids cause metabolic changessuch as mobilizing energy to maintain muscle and brain function, increasingblood flow and cerebral glucose utilization, increasing cardiac output,increasing the respiratory rate, and distributing the blood flow toincrease substrate and energy transmission to the brain and muscles(
McEwen, 2007; McEwenet al., 2016;
Erfani et al., 2017). CRH and ACTH act on the hypothalamus neurons to play a major rolein the development of anorexia nervosa in stress (
Connan et al., 2007). Thus, stresscan affect the hormonal and neurotransmitter systems. Many studieshave shown that psychological and physiologic stress can induce sexualdysfunction and infertility (
Kennedyet al., 1999;
Baldwin,2001). There is a functional interaction between thegonads and the adrenal axes that is in large part due to the interactiveeffects of sex hormones and glucocorticoids, demonstrating that severaldisease states connected to stress are sex-dependent. Testosteronecan act and interact with different aspects of basal and stress HPAfunction. The basal ACTH release is regulated by testosterone-dependenteffects on arginine vasopressin synthesis and corticosterone-dependenteffects on CRH synthesis in the PVN of the hypothalamus. In contrast,testosterone and corticosterone interact on stress-induced ACTH releaseand drive to the PVN motor neurons. In males, stress significantlyreduces levels of gonadotropin-releasing hormone (GnRH), folliclestimulating hormone (FSH), luteinizing hormone (LH), and testosterone(
Arun et al., 2016;
Rai et al., 2004).
It has been shown that response tostress involves sexual dimorphism, which means that men and womenrespond to stress differently. On the other hand, studies in humansand animals have indicated sexual differences in terms of sensitivityto the stress response. These sexual differences may be related toenvironmental, cultural, social, genetic, and sex hormone factors(
Curtis et al., 2006;
McEwen and Milner, 2007; McEwen etal., 2016). Prolonged stress reduces the male sex hormone level, whichcan cause a rapid response to new situations. Chronic stress, whichkeeps the body in a state of constant threat, is a major factor inthe decline of testosterone levels. It has been shown that testosteroneand estrogen exert reliable inhibitory and stimulatory effects, respectively,on HPA axis activity (
Williamson etal., 2005).
Lund etal. (2006) showed that different metabolites of testosterone,such as 5a-dihydrotestosteroneand its 3b-diol metabolite, arecapable of acting locally to suppress stress-induced levels of PVNFos mRNA and plasma ACTH and corticosterone. Studies also showed thatdifferent kinds of stress, like restraint, electrical shock, cold,and sleep deprivation, could decrease testosterone levels (
Hardy et al., 2005;
Hari Priya and Reddy, 2012). However,studies of the relationship between stress and sex hormones in rodentsshowed conflicting results. Although several studies have investigatedthe effects of testosterone during stress, studies that compare theeffect of the intraventricular (ICV) and intraperitoneal (IP) administrationof testosterone on metabolic changes during electric foot shock stressare lacking. Therefore, the aim of this study was to investigate theeffects of the ICV and IP injection of testosterone on plasma cortisolconcentration and eating behaviors such as anorexia, water and foodintake, weight change, and brain/adrenal gland volume during acuteand chronic stress in male gonadectomized Naval Medical Research Institute(NMRI) mice.
Material and methods
Animals
Male NMRI mice weighing 30±5g each were kept in groups of four per cage in 12-h light/dark conditionsat 22–24°C with food and water provided ad libitum except during the experimental time. The animalswere housed in the standard animal room for 1 week for adaptationbefore the start of the study. The animals were randomly divided intocontrol and experimental groups (n = 8/each). Food and water intake was recorded for each animal atspecific hours every day using the amount of food and water left ineach cage. Animal experiments were conducted in accordance with theGuidelines of the National Institute of Health for the Care and Useof Laboratory Animals and approved by the local ethical committee(The Baqiyatallah University of Medical Sciences Committee on theUse and Care of Animals, 87/381).
Drugs
The following drugs were used throughoutthe experiments: testosterone (Abu-Rayhan-Iran), ketamine hydrochloride(Sigma–Aldrich, USA), and diazepam (Sigma–Aldrich). Testosteronewas dissolved in sesame oil and injected ICV (0.01, 0.1, 0.05 mg/mouse) 5 min or IP (0.01, 0.1, 0.05 mg/kg)30 min before the stress induction. The ketamine hydrochloride anddiazepam hydrochloride were dissolved in sterile saline.
Cannulation
For cannulation, the animals werefirst anesthetized with ketamine 50–75 mg/kg and diazepam 5–7mg/kg and the surgical area was shaved. The animals were placed ina stereotaxic apparatus and a small incision was made in the scalpto expose the skull. Using bregma and lambda as landmarks, the skullwas leveled in the coronal and sagittal planes with one or two guidecannulas (gauge no. 23; World Precision Instruments) implanted intothe skull above the lateral ventricles utilizing the
Paxinos and Franklin (2001) atlas(for the lateral ventricles, AP= + 0.9 mm; ML= ±2 mm; DV= 3mm) and fixed with dental acrylic cement. The animals were given 7days to recover after the surgery. A dental needle head no. 30 (Alibaba;INTR), polyethylene tubes, and 10–mL Hamilton syringes wereused for the injections. The unilateral (left lateral ventricle) administrationof different doses of testosterone (0.1, 0.01, and 0.05 mg/mouse) was conducted daily for 5 min priorto the stress induction. The brain injection was gradual and lasted60 s, during which time the animals were free to move around.
Gonadectomy
First, the animal was anesthetizedby an IP injection of ketamine hydrochloride 70 mg/kg and xylazine10 mg/kg (
Mohammadian et al., 2017;
Sadeghi-Gharajehdaghi et al., 2017). Each animal was placed in dorsal recumbency, the hair over thecaudal abdomen was shaved, and the surgical area was disinfected with70% alcohol followed by a 2% chlorhexidine solution. Next, a midlineincision approximately 5–7 mm long was made in the scrotum toexpose the tunica. One testis was pushed out of the tunica and thevas deferens and spermatic blood vessels were cauterized. Finally,the testes were removed and the incision was sutured.
Animal group
The animals were randomly dividedinto nine groups (
n = 8 each).The control group animals received no treatment. The experimentalgroup animals were gonadectomized and divided into two acute and chronicstress groups. Testosterone was injected IP or ICV (left side) intothe experimental group animals. Testosterone was injected IP (0.01,0.1, 0.5 mg/kg) 30 min or injected ICV (0.01, 0.1, 0.5 mg/mouse) 5 min before the induction of chronicor acute stress. After the testosterone administration, to inducestress, the animals were placed into a communication box consistingof nine separate parts (50 × 16 × 16 cm) with Plexiglaswalls and tiny holes with a 2-mm diameter to allow olfactory and auditorycommunication between the mice. Steel bars (with a 4-mm diameter)were placed on the floor of the instrument at 1.3 cm apart throughwhich the electric shock is transmitted to the animal’s soles(
Dalooei et al., 2016;
Amouei et al., 2016). Shock duration and intensity were controlled by a computer connectedto the communication box (60 mV, 10 Hz, for 60 s). The electric shockwas induced randomly for 9–13 h. To allow the animals to adaptto the environment, they were transferred to the test room 60 minbefore the stress induction and remained there 30 min before and 30min after the stress induction. In acute stress, the animals receivedfoot shocks for 1 day; in chronic stress, the animals received footshocks for 4 consecutive days. The control animals were placed inthe device for 60 min without receiving any shocks. After the stressinduction, the animals were returned to their cages. Anorexia (theelapsed time between mouse replacement in the home cage and the beginningof food intake was calculated as a delay of eating or anorexia), waterand food intake, and weight were measured every day at a specifictime. The control group received sesame oil (testosterone solvent).After completing the tests on day 4, the animals were anesthetizedusing high doses of ketamine and their brains and adrenal glands wereremoved and kept in 4% formalin solution for fixation. Sixty dayslater, the brains and adrenal glands were removed from the formalin.The weights and volumes of the brains and adrenal glands were measuredby mercury immersion.
Cortisol concentration
One day before the start and on thelast day of the experiment, blood samples were collected from allanimals in all groups from their retro-orbital sinus (0.5 mL of bloodin 0.5 mL of 3% ethylenediaminetetraacetic acid). The blood was thencentrifuged at 3000 rpm for 5 min at 4°C and the serum was collectedfor cortisol detection. The serum was collected and frozen at -20°C and the cortisol concentrationswere determined using an enzyme-linked immunosorbent assay kit (CortisolELISA kit 4164; DRG Instruments GmbH, Germany). Briefly, serum sampleswere added to 96-well plates containing biotinylated primary antibodyand then incubated at 37°C for 45 min. Thereafter, the plateswere washed and horseradish peroxidase–conjugated streptavidinsolution was added to the wells and incubated for an additional 30min at 37°C. The 3,30,5,50-tetramethylbenzidine substrate wasadded, the plates were incubated for an additional 15 min at 37°C,and stop solution was added to the wells to terminate the reaction.Cortisol concentrations were determined using a standard curve.
Data analysis
The data are expressed as mean±S.E.M.To analyze the data, one-way analysis of variance followed by Tukeytest were used. Values of p<0.05were considered statistically significant.
Results
Effects of testosterone administration on plasma cortisol levelsduring acute and chronic stress induction
The results showed that acute andchronic stress significantly increased plasma cortisol concentrationscompared to the control group. In both acute and chronic stress, theIP administration of testosterone has no significant effect on plasmacortisol concentrations. However, the ICV injection of testosteronein acute and chronic stress significantly decreased cortisol concentrations(Figs. 1A and 1B).
Effect of testosterone administration on anorexia during acuteand chronic stress
After stress induction, the experimentalgroups were returned to their holding cages and the duration of anorexia(delay of eating) was measured. The results obtained on the firstday were taken as 100 and as a point of reference for measurementsmade on subsequent days (percentage). As shown in Fig. 2A, acute stresssignificantly decreased anorexia time compared to the control group.The IP (0.1 and 0.05 mg/kg) and ICV (all doses) injections of testosteronein acute stress significantly increased anorexia time compared tothe stress group. This increase was higher when testosterone was injectedICV. Anorexia time in chronic stress also decreased compared to thecontrol group. Only the IP administration of testosterone 0.05 mg/kgsignificantly increased the anorexia time compared to the chronicstress group, whereas the ICV administration of testosterone at alldoses significantly increased anorexia time compared to the chronicstress group (Fig. 2B).
Effect of testosterone administration on food intake afteracute and chronic stress induction
The animals were returned to theircages after stress induction and the food intake was measured during24 h for 4 consecutive days. The results obtained on the first daywere taken as 100 and as a point of reference for measurements madeon subsequent days (percentage). The results showed that both acuteand chronic stress significantly reduced food intake in gonadectomizedmice. However, the effect of chronic stress on food intake reductionwas lower than that of acute stress. In both acute and chronic stress,IP injections of testosterone (0.01 and 0.05 mg/kg in acute stressand 0.1 and 0.05 mg/kg in chronic stress) increased the food intakecompared to the stress group. However, in the chronic stress experiments,the IP injection of testosterone at 0.01 mg/kg significantly decreasedthe food intake compared to the stress group (Figs. 3A, 3B). In theacute stress experiments, the ICV injection of testosterone 0.01 and0.05 µg/mouse increased the food intake compared to the stressgroup, but the injection of 0.1 µg/mouse decreased the foodintake compared to the stress group. In chronic stress, the ICV injectionof testosterone at 0.1 and 0.01 µg/mouse reduced the food intake,while no significant changes were observed at 0.05 µg/mousecompared to the stress group.
Effect of testosterone administration on water intake afteracute and chronic stress induction
As shown in Figs. 4A and 4B, bothacute and chronic stress reduced the water intake compared to thecontrol group. The IP administration of testosterone at 0.05 and 0.1mg/kg in acute stress significantly reduced water intake comparedto the stress group. However, the ICV administration of testosteroneat all doses (0.1, 0.05, 0.01 µg/mouse) significantly increasedwater intake (Fig. 4A). In the chronic stress experiments, the IPadministration of testosterone had no effect on water intake comparedto the stress group; however, the ICV administration of testosteronesignificantly reduced water intake compared to the stress group (Fig.4B).
Effect of testosterone administration on weight changes afteracute and chronic stress induction
The animals were weighed every daybefore the stress induction and their weight changes were measuredover 4 days. Figure 5 shows the effect of different doses of testosteroneon the animals’ weight. Acute stress significantly increasedthe animals’ weights compared to the control group (Fig. 5A),while chronic stress had no effect on bodyweight compared to the controlgroup (Fig. 5B). On the other hand, in the acute stress experiments,the IP administration of testosterone at 0.1 and 0.01 mg/kg significantlyreduced the animals’ weights compared to the stress group (Fig.5A). However, the ICV administration of testosterone had no effecton bodyweight compared to the stress group. As shown in Fig. 5, theIP injection of testosterone at 0.01 mg/kg significantly reduced theweights compared to the stress and control groups. However, the ICVinjection of testosterone had no effect on animal weights.
Effect of IP and ICV injection of testosterone on brain/adrenalgland weight after acute and chronic stress
As shown in Figs. 6A and 6B, acutestress caused a significant decrease in brain and adrenal gland weightscompared to the control group. However, chronic stress had no significanteffect on brain and adrenal gland weights compared to the controlgroup. In both acute and chronic stress, the IP and ICV injectionsof testosterone 0.01 mg/kg significantly increased the ratio of brainto adrenal gland weight compared to the stress and control groups.
Discussion
In this study, acute and chronicstress increases plasma cortisol levels in male mice. The IP injectionof testosterone could not reduce the plasma cortisol concentrations.However, the ICV injection of testosterone reduced plasma cortisolconcentrations. Stress can cause increased HPA axis activity, andan increased level of glucocorticoid hormones is one of the most importantsigns of stress (
McEwen, 2007). In the present study, these mechanisms may increase plasma concentrationsof cortisol in animals after stress induction. Several studies showedthat androgens inhibit stress-stimulated ACTH and cortisol concentrationsin animals (
Handa et al., 1994;
Papadopoulos and Wardlaw, 2000). Testosterone (through several mechanisms such as decreased CRH,increased glucocorticoid receptor concentrations, decreased argininevasopressin [AVP], and suppression of cortisol secretion through itsmetabolite 3 beta androstanediol) reduced cortisol concentrations(
Lund et al., 2004;
Rubinow et al., 2005). It is possiblethat the ICV injection of testosterone reduces cortisol concentrationsvia these mechanisms.
Our results showed that chronic stresscauses slight weight gain in animals, but this increase was not significantand probably due to stress-induced overeating. The IP injection oftestosterone 0.01 mg/kg reduced the animals’ weights in thestress experiments. However, the ICV administration of testosteronehad no effect on the animals’ weights in the chronic stresscondition. Overeating caused by stress is common in human societiesand may lead to metabolic disorders such as obesity and diabetes (
Mikolajczyk et al., 2009). It isbelieved that the presence of high plasma and brain cortisol concentrationsleads to high brain sensitivity that is reflected in increased nutritionalintake and the tendency to the use special foods, increase sugar andfat cravings, and even increase fat stores. This is one mechanismby which stress increases fat and leads to metabolic dysfunction.Chronic stress, together with high glucocorticoid concentrations,usually reduces bodyweight gain in rats; in contrast, in stressedor depressed humans, chronic stress induces either elevated comfortfood intake and bodyweight gain or reduced intake and bodyweight loss.Comfort food ingestion, which produces obesity, decreases corticotropin-releasingfactor (CRF) mRNA in the rat hypothalamus (
Dallman et al., 2003). Also, ghrelinis an important molecular mediator that leads to weight gain. Ghrelinis an appetizer peptide that is synthesized mainly in the stomachand observed in the blood of healthy individuals, but its circulatingconcentrations have been found to increase following stress (
Chuang and Zigman, 2010). Ghrelinplays a role in food intake, weight gain, and obesity in rodents.Since chronic stress can lead to exposure to high cortisol levelsin the brain and body and has direct and indirect effects on rewardsystems, the combination of these factors (high cortisol and morecalories) can increase the amount of abdominal fat and leads to obesity(
Adam and Epel, 2007).
Studies have shown that chronic stressreduces food intake in animals. Several studies have shown that stresscan inhibit bodyweight gain and food intake in rodents (
Krahn et al., 1990;
Dallman et al., 1992). This reductionin food intake may be due to stress adaptation in animals. Acute stresscan reduce food intake 1 day after stress in male mice. In the currentstudy, the IP injection of testosterone had no significant abilityto inhibit the effects of chronic stress on food intake, whereas theICV injection of testosterone 0.05 µg/mouse increased the foodintake. The HPA axis plays an important role in appetite regulation.The CRF is responsible for the anorexia in stress. The injection ofCRF into the brain reportedly reduces appetite (
Majzoub, 2006). However, there aresexual differences in this regard, and some studies have shown thatfood consumption, especially in women, increases during stress (
Habhab et al., 2009). Testosteronereceptors are involved in the control of food intake. For instance,androgen levels are involved in food intake and metabolism in women,which results in weight gain (
Cotrufoet al., 2000; Hill, 2010). The arcuate nucleus playsa pivotal role in the control of food intake by opposing orexigenicand anorexigenic neuronal circuits. The anorexigenic neurons expresscocaine- and amphetamine-regulated transcript (CART) and pro-opiomelanocortin(POMC). The activation of POMC/CART neurons signals to the downstreamneuronal pathways that suppress food intake.
Hill (2010) showed that androgenscould affect POMC gene expression and that androgen receptors regulatethe transcription of target genes by interacting with DNA responseelements.
In the present study, the effectsof stress on the delay to eating were also investigated. Chronic stressincreases the delay to eating in the stress group. As mentioned above,CRF has a significant effect on appetite suppression and is considereda strong neuropeptide appetite inhibitor. Also, glutamate drives theHPA axis stress responses through excitatory signaling via ionotropicglutamate receptor signaling. Moreover, glutamate innervation of thePVN undergoes neuroplastic alteration under chronic stress and maybe involved in the sensitization of HPA axis responses (
Evanson and Herman, 2015). Chronicstress may reduce the regulation of or changes in CRF receptor distributionin the hypothalamus and decrease the effect of CRF on the inductionof anorexia during the chronic stress. The IP administration of testosteroneat three doses reduced the delay to eating and the ICV injection oftestosterone 0.1 or 0.05 µg/mouse increased the delay to eatingtime.
Our results show that acute stresscan lead to reduced water intake in animals, a finding that is inconsistentwith previous studies (
Osanloo etal., 2015). The mechanisms involved in thirst are activatedduring stress and lead to the thirst sensation. The HPA axis playsan important role in adaptation to environmental stressors throughthe secretion of CRF and VP, which regulate ACTH release from theanterior pituitary and are involved in regulating the body’swater intake. During stress stimulation, the simultaneous secretionof these two neurohormones from the PVN exacerbates the sensationof thirst and increases water intake (
McEwen, 2012). In this study, the IP testosterone administrationcould not inhibit the effects of acute stress on reducing water intake,but the ICV injection of testosterone at three doses can indeed inhibitthe effects of acute stress and result in increased water intake.We showed here that chronic stress also leads to a reduced water intakein male mice, while the ICV and IP injections of testosterone couldnot reduce water intake. Testosterone has a suppressive effect onstress-induced AVP biosynthesis in the median eminence (
Viau and Meaney, 1996).
We also studied the effects of stresson brain/adrenal gland volume ratio. The hippocampus is a part ofthe brain involved in learning and memory (
Meftahi et al., 2014;
Eslamizade et al., 2015;
Meftahi et al., 2015). Several studieshave shown that the volume and weight of any part of the brain suchas the hippocampus are reduced after chronic stress, demonstratingstress-induced dendrite atrophy and shrinkage, loss of dendrite spinesin neuronal populations, and increased volume and weight of the outerpart of the adrenal gland (
Listonet al., 2006;
Ulrich-Laiet al., 2006;
Lee etal., 2009;
Motahariet al., 2016).
Our results show that chronic stresscan decrease the brain to adrenal gland volume ratio. On the otherhand, the brain to adrenal gland weight ratio was also reduced instresses animals. The IP and ICV administration of testosterone inhibitedthe damaging effects of stress; however, both increased brain/adrenalgland volume. These results showed that androgen receptors in thebrain play an important role in inhibiting the damage caused by chronicstress.
ACTH and corticosterone responsesto acute stress in male rats are increased by gonadectomy and is reversedby 5α-dihydrotestosterone (the reduced nonaromatizable formof testosterone), a finding that is consistent with our results (
Handa et al., 1994;
Viau and Meaney, 1996). This showsthat the suppressive effect of testosterone in males on stress HPAfunction is mediated by an androgen receptor–mediated effectunder stress conditions. Acute exposure to stress elevates glutamaterelease in many parts of the brain such as limbic system, prefrontalcortex, hippocampus, and amygdala. If stress becomes chronic (morethan a few hours or over long periods of time more than once a day),glucocorticoid levels increase severely and lead to stimulation ofthe glutamate system, which damages different parts of the nervoussystem including the cerebral cortex and hippocampus (
Venero and Borrell, 1999;
Yuen et al., 2012). For this reason,reductions in brain size and weight after a period of chronic stressis predictable. Studies have shown that chronic stress increases adrenalgland weight by hyperplasia of the outer zona fasciculate and hypertrophyof the inner zona fasciculata and medulla. Stress-induced increasesin adrenal medullary size and/or catecholamine content is frequentlyobserved after other types of stress as well. Furthermore, followingstress there is a generalized increase in medullary function, whichsuggests that medullary hypertrophy may be a general consequence ofchronic stress (
Miyamoto et al., 1999;
Ulrich-Lai et al., 2006).
Wrightet al. (1991) have shown that androgens can increasethe volume, neuron number, and synapses of developing rat superiorcervical ganglion.
Hammond et al.(2001) also indicated that physiologic levels of testosteronecould protect against serum deprivation–mediated neuronal apoptosisby interacting with androgen receptors. One of the proposed mechanismsis that androgens increase hippocampal neurogenesis by enhancing cellsurvival with no significant effects on cell proliferation in thedentate gyrus of male rodents (
Spritzerand Galea, 2007). Thus, these findings are consistentwith the present results demonstrating brain/adrenal volume decreasesin stress.
Conclusions
The results of this study showedthat the ICV and IP administration of testosterone in gonadectomizedmale mice can play an important role in the modulation of hormonaland metabolic effects caused by acute and chronic stress, includingwater and food intake, bodyweight changes, anorexia, and plasma cortisollevel and the ratio of brain weight and volume to adrenal gland weightand volume during the induction of acute and chronic stress.
Higher Education Press and Springer-Verlag GmbHGermany, part of Springer Nature
PDF
(490KB)
