- Email: [email protected]
Estrus cycle stage modifies the presentation of stress-induced startle suppression in female Sprague–Dawley rats
Available online at www.sciencedirect.com
Physiology & Behavior 93 (2008) 1019 – 1023 www.elsevier.com/locate/phb
Estrus cycle stage modifies the presentation of stress-induced startle suppression in female Sprague–Dawley rats Kevin D. Beck ⁎, Xilu Jiao, Tara P. Cominski, Richard J. Servatius Neurobehavioral Research Laboratory (129), Veterans Affairs New Jersey Health Care System, East Orange, NJ 07018, United States Stress and Motivated Behavior Institute, Department of Neurology and Neurosciences, New Jersey Medical School – UMDNJ, Newark, NJ 07103, United States Received 10 July 2007; received in revised form 12 December 2007; accepted 11 January 2008
Abstract Tailshock stress causes transient reductions in startle reactivity, associative learning and open field activity in female rats in an ovarian hormone dependent manner. Others have shown estrogen modulation of associative learning by testing across the estrus cycle and pharmacological manipulations. Here we tested whether stress-induced suppression of startle reactivity can be attributed to circulating ovarian hormones. Female rats were tracked across the estrus cycle and subjected to the stressor (2 h periodic tailshock) the morning of diestrus, proestrus, estrus, or metestrus. Startle reactivity was tested 2 h following the cessation of the tailshock. Using a multi-stimulus protocol, we determined there were differences in startle sensitivity and responsivity. Following stressor exposure, estrus females exhibited reduced startle responsivity. In contrast, diestrus females exhibited increased sensitivity to the lowest acoustic stimulus. The results are discussed with respect to ovarian hormone regulation of the immune system and sensory reactivity during and following trauma that may lead to different abnormal behaviors in females in the wake of traumatic stress. Published by Elsevier Inc. Keywords: Female; Stress; Ovarian hormones; Estrus cycle; Startle
Because of the recent increase in both the number of women serving in the military and the number of troops returning from service in Iraq and Afghanistan with post-traumatic stress disorder (PTSD), it is necessary at this time to delineate sex differences in response to traumatic stress. For instance, one of the hallmark symptoms of PTSD is an exaggerated startle response [3]. However, this is not always evident in women. Medina and colleagues [12] reported that a population of abused women with PTSD has blunted startle responses. This suggests that stress-induced changes in sensory reactivity may not be as homogenous as thought in the past, at least in women. Much of the neurobiological bases of abnormal behavior following stressor exposure in males and females have been documented in rat models. For instance, following a period of
⁎ Corresponding author. Neurobehavioral Research Laboratory (129), VA NJ Health Care System, 385 Tremont Ave, East Orange, NJ 07018, United States. Tel.:+1 973 676 1000x3682; fax: +1 973 395 7114. E-mail address: [email protected] (K.D. Beck). 0031-9384/$ - see front matter. Published by Elsevier Inc. doi:10.1016/j.physbeh.2008.01.012
inescapable tailshock stress, males and females exhibit differing capabilities of associative learning, as exemplified by the acquisition of a conditional eyeblink response [5,19,20]. Stressed males learn the sound-blink association quicker than nonstressed males, but similarly stressed females acquire the conditional eyeblink response slower than nonstressed females. In addition to the effects on associative learning, this particular type of stressor also affects startle reactivity differentially across sex. Acoustic startle responses are suppressed for a period of time in female rats [6], whereas males exhibit enhanced startle responses (albeit several days later)[15,16]. We have established that an important element of stressinduced startle suppression in the female rats is the presence of ovarian hormones [6]. Both repeated tailshock exposure and systemic injections of interleukin-1β suppress the magnitudes of elicited startle responses in female rats, unless they are ovariectomized [6,7]. This reduction in startle reactivity appears to be solely through a reduction in motor responsivity because we have not observed any decrements in the sensitivity of the rats to startle in response to different acoustic intensities [6].
1020
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
This study was designed to further explore the role of ovarian hormones in the modification of startle reactivity in female rats following exposure to a traumatic stressor. Here, we attempted to discern whether differences in stress-induced startle suppression are evident across different stages of the estrus cycle. Shors and colleagues [17,19] have shown this type of stressor can differentially affect the acquisition of a conditional eyeblink response, depending on the stage of estrus during which the classical conditioning occurred. However, it should be noted that those researchers did not observe any changes in sensory reactivity following stress (assessed prior to conditioning). Thus, although they observed that rats in proestrus were most affected by stressor exposure (and those in diestrus were least affected), the effects of stress on learning through the ovarian hormones could be different than what occurs to startle reactivity following stressor exposure. Others have suggested progesterone metabolites may serve to dampen anxiety-induced increases in startle reactivity [18]. Estradiol levels rise during the later part of diestrus and peak during proestrus. Progesterone levels rise sharply a few hours later and initiate the beginning of estrus. Hence, we may expect stress-induced startle suppression to be more evident during either during the hormone rise in proestrus or the fall during estrus, if estradiol and/or progesterone are necessary to cause a blunting of startle responsivity. Moreover, we should not expect to observe a suppression of startle responsivity during metestrus or diestrus, when the ovarian hormones are circulating in much lower quantities.
tailshock has been described previously [6,16]. Briefly, it involves administering 40, 3-s, 2-mA electrical shocks through an electrode clipped 1–2 cm below the base of the tail of rats restrained in Plexiglas restrainers (Harvard Apparatus). This stimulus has been shown before to be sufficient to cause changes in startle responding in both male and female Sprague– Dawley rats. Two hours following the termination of stressor presentation, all rats were tested for startle reactivity. The startle procedures used are based on the methods of Blumenthal [9] that utilize multiple acoustic stimulus intensities in determining differences in the number of startles elicited at different intensities (sensitivity) in combination with the magnitudes of those responses (responsivity). LabView software was used to produce 8 white-noise acoustic stimuli, each of an intensity of 92, 96, and 102 dB (5 ms rise/fall, 100 ms duration) for a total of 24 trials. The electrical signals produced from the accelerometer during displacements occurred at a rate of 1 kHz, with A/D conversion into the computer for analysis by S-plus software (Insightful Corp). For each startle stimulus presentation, a response threshold for whole body response was computed as the average rectified activity 200 ms prior to stimulus onset plus 4 times the standard deviation of that rectified activity [16]. Response amplitudes, the maximum rectified activity within 125 ms after stimulus onset, were only recorded when poststimulus activity exceeded the response threshold. For trials in which activity did not reach this criterion “not available” was recorded, for all others, this calculated value was corrected by each rat's body weight (taken the morning of startle testing).
1. Methods 2. Results 1.1. Subjects Sixty-eight 3-month old female Sprague–Dawley rats obtained from Charles River (Wilmington, MA) served as subjects. All rats were allowed at least 10 days to acclimate to the facility and the 12:12 h light:dark cycle (lights on 0700). All the rats were well-handled over the following 10–14 days due to the vaginal smearing procedures. 1.2. Procedures All rats had vaginal smears to determine estrus cycle stage. This involved sampling the cells of the vaginal canal with sterile saline using a glass pipette. The recovered solution containing cells were placed on microscope slides, stained with Evan's blue, and dried. Subsequent histological analysis occurred by viewing the dried slides under medium-power microscope (Wild Heerbrugg). Cell descriptions, as described by Sharp and LaRegina [23], were used to classify rats as being in diestrus, proestrus, estrus, or metestrus. Once cycling patterns were observed, we administered a 2 h tailshock protocol to rats staged that morning. The number of rats determined to be in each estrus stage and assigned to control or stress condition were as follows: diestrus – control (11), diestrus – stress (9), proestrus – control (7), proestrus – stress (10), estrus – control (6), estrus – stress (5), metestrus – control (11), and metestrus – stress (9). For the stress manipulation, the
Startle sensitivity and startle responsivity were measured in differently staged female rats following exposure to tailshock. The comparisons between number of startles elicited and the magnitudes of the elicited startles were conducted using a 4 (Estrus Stage) × 2 (Stress) × 3 (Stimulus Intensity) mixed analysis of variance (ANOVA) with Fisher's LSD post-hoc tests (p b .05). The number of elicited startle responses was different depending of the intensity of the acoustic stimuli. As shown in Fig. 1, as intensity increased, the number of startle responses detected increased. This was confirmed by a significant main effect of Stimulus Intensity, F (2, 120)= 244.3, p b .001. However, an additional Estrus Stage × Stress × Stimulus Intensity interaction, F (6, 120)= 2.3, p b .05, suggested this pattern was not the same across groups. Post-hoc tests revealed a significant difference in the number of elicited startle responses to the lowest acoustic stimuli following stress exposure but only during diestrus. The diestrus females tested following stressor exposure had more startle responses than their same-stage control counterparts. For the other dependent measure, startle magnitudes, the calculated values were log-transformed in order to achieve the normality requirements of the ANOVA. As shown in Fig. 2, startle magnitudes were greater for the higher stimulus intensities, as evidenced by a significant main effect of Stimulus Intensity, F (2, 116) = 262.8, p b .001. In addition, both a Stress × Stimulus Intensity, F (2, 116) = 3.7, p b .05, and Estrus Stage × Stress × Stimulus Intensity interaction were significant, F (6, 116) = 2.2,
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
1021
p b .05. Post-hoc tests revealed a significant difference in startle magnitude due to stress exposure evident in response to the lowest intensity acoustic stimulus during estrus. When startle testing was conducted in estrus females following stressor exposure, the magnitude of the elicited startle responses was significantly lower than their same-stage control counterparts. 3. Discussion The current results further our understanding of the endocrine modulation of startle suppression by showing female rats at certain stages of the estrus cycle appear to be more or less likely to exhibit changes in startle sensitivity and responsivity. Startle suppression, reflected as blunted startle responsivity, following stressor exposure was observed. When stress occurred during estrus, startle suppression was most evident 2 h later during testing. In contrast, stressor exposure during diestrus was associated with an increase in startle sensitivity. Specifically, those rats startled more often to the lowest intensity stimulus, without exhibiting any significant differences in responsivity from their same-stage controls. This suggests that stress is causing rather different effects upon startle reactivity in females depending on the phase of the estrus cycle. The importance of this finding is that we now have evidence that the effect a painful stressor can have upon sensory reactivity can be quite different depending on Fig. 2. Female rats at 4 stages of the estrus cycle were either subjected to tailshock stress (shaded bars) or served as controls (open bars). Startle testing occurred 2 h following the termination of stressor exposure. Shown are the mean startle magnitudes of the startle responses elicited at each of 3 different stimulus intensities (in arbitrary units) for each condition for each estrus cycle group. An asterisk (⁎) represents a significant difference between stress and control groups at a particular stimulus intensity (p b .05).
Fig. 1. Female rats at 4 stages of the estrus cycle were either subjected to tailshock stress (shaded bars) or served as controls (open bars). Startle testing occurred 2 h following the termination of stressor exposure. Shown are the mean percentages of startle responses elicited at each of 3 different stimulus intensities for each condition for each estrus cycle group. An asterisk (⁎) represents a significant difference between stress and control groups at a particular stimulus intensity (p b .05).
the endocrine state of the female at the time of the trauma. This could explain, in part, why some women with PTSD develop exaggerated startle responses [11] while others develop suppressed startle reactivity [12]. Others have recently reported suppressed startle responses in male rats following a combination of injection and predatorexposure stressors [1,2]. In that model, both GABA and glucocorticoid receptors have been implicated as mechanisms for the reduction in startle reactivity [1,2]. We have not observed startle suppression in male rats exposed to the tailshock protocol described herein [6]; in fact, if it occurs daily for over 3 consecutive days, the startle becomes exaggerated in the male rats while still exhibiting characteristics of suppression in the similarly stressed female rats [4]. Still, mechanisms proposed by Adamec and colleagues [1,2] could be quite relevant for our female model, now that we have identified phases where startle sensitivity can be increased and startle responsivity decreased following stressor exposure. Our research thus far has focused on early immune cascade mechanisms involved in the response to tailshock that may affect neural circuitry involved with dampening sensory reactivity processes (such as IL-1β) [4–7], but the mechanisms proposed in the male predator-stress model can be highly influenced by circulating ovarian hormones as well. For instance, RU486 was
1022
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
shown to block the suppression of the startle response elicited by predator exposure and a subsequent injection stress [1]. The authors proposed a glucocorticoid receptor mechanism, and the use of RU 486 to block the stress-induced suppressed startle, because the added injection-stress procedure increased corticosterone release above that of predator stress exposure alone [2]. We can also consider the other actions of RU 486, namely as a progesterone receptor antagonist. Core body temperature is increased by IL-1β for a longer period of time during proestrus (compared to diestrus) apparently because of the actions of progesterone [13] through hypothalamic cyclooxygenase [14]. Still, progesterone is elevated at the end of proestrus and decreases through the onset of estrus. Recalling that IL-1β causes startle suppression in intact females [7], progesterone receptor antagonism by RU 486 may block the stress-induced startle suppression because a progesterone-proinflammatory cytokine interaction is being blocked. This effect may be more prominent in females because of the added presence of estradiol. Constant estradiol treatment leads to a blunted HPA-axis response when stimulated with IL-1β, compared to OVX animals [21,22]. Thus, the antiinflammatory actions of corticosterone are reduced and, presumably, there is a reduced amount of central corticotrophinreleasing hormone (CRH) stimulated by peripheral IL-1β actions upon the vagus nerve. As expected, others have shown the combination of estradiol and progesterone will reduce the magnitude of CRH-enhanced startle [18]. With diestrus being the estrus phase with the least circulating estradiol and progesterone, it is not surprising in the current study that we saw an apparent increase in startle sensitivity and no reduction in startle responsivity during diestrus. Others have shown progesterone is the main cause of changes in the degree CRH can enhance startle reactivity when administered i.c.v. [18]. This is not likely through a direct action of progesterone upon its receptor, but through one of its metabolites (allopregnanolone) increasing GABA activity and acting as an endogenous anxiolytic [8,10,18]. With exposure to a traumatic stressor, it seems as though physical reactions occur that both enhance and dampen sensory reactivity, and it may be differences in this balance that determine the specific direction startle reactivity changes following stressor exposure. For example, the increase in startle sensitivity during diestrus could be due to more or greater excitatory processes (such as CRH) being recruited because of the lower levels of estradiol and progesterone. However, during estrus, when progesterone is reducing from its peak, it could be that the inhibitory mechanisms (described above) outweigh those supporting excitation and enhanced reactivity. Obviously, hormone manipulation studies need to occur in order to definitely determine the precise timing of the endocrine actions involved in regulating stress-induced startle suppression in female rats. The literature appears to support our initial impressions from the current study that progesterone is a likely candidate mechanism. Our future work is going to examine hormone replacement combinations with relation to tailshock stress and IL-1β treatment to discern if the suppressed startle from each of these treatments appears to be regulated by the same ovarian hormonal mechanisms. Understanding how ovarian
hormones influence startle reactivity (and sensory reactivity in general) following traumatic stress is important for the further development of stress models that take into account individual differences in stress reactivity that lead to different abnormal behaviors. Acknowledgment This research was supported by the Department of Veterans Affairs Merit Review research program to KDB and program support through the Stress and Motivated Behavior Institute (RJS). The described experimentation was conducted with approval by the VANJHCS Institutional Animal Care and Use and Research and Development Committees. Tara Cominski is currently a doctoral student in Neuroscience at UMDNJ – Robert Wood Johnson Medical School Piscataway, NJ. We also acknowledge the technical support provided by Toni Marie Dispenziere and Paul William Ong. References [1] Adamec R, Muir C, Grimes M, Pearcey K. Involvement of noradrenergic and corticoid receptors in the consolidation of the lasting anxiogenic effects of predator stress. Behav Brain Res 2007;179(2):192–207. [2] Adamec R, Strasser K, Blundell J, Burton P, McKay DW. Protein synthesis and the mechanisms of lasting change in anxiety induced by severe stress. Behav Brain Res 2006;167(2):270–86. [3] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th-TR ed. Washington, DC: American Psychiatric Association; 2000. [4] Beck KD, Brennan FX, Servatius RJ. Effects of stress on nonassociative learning processes in male and female rats. Integr Physiol Behav Sci 2002;37(2):128–39. [5] Beck KD, Servatius RJ. Stress and cytokine effects on learning: what does sex have to do with it? Integr Physiol Behav Sci 2003;38(3):179–88. [6] Beck KD, Servatius RJ. Stress-induced reductions of sensory reactivity in female rats depend on ovarian hormones and the application of a painful stressor. Horm Behav 2005;47(5):532–9. [7] Beck KD, Servatius RJ. Interleukin-1beta as a mechanism for stress-induced startle suppression in females. Ann NY Acad Sci 2006;1071:534–7. [8] Bitran D, Purdy RH, Kellogg CK. Anxiolytic effect of progesterone is associated with increases in cortical allopregnanolone and GABAA receptor function. Pharmacol Biochem Behav 1993;45(2):423–8. [9] Blumenthal TD, Berg WK. Stimulus rise time, intensity, and bandwidth effects on acoustic startle amplitude and probability. Psychophysiology 1986;23(6):635–41. [10] Callachan H, Cottrell GA, Hather NY, Lambert JJ, Nooney JM, Peters JA. Modulation of the GABAA receptor by progesterone metabolites. Proc R Soc Lond B Biol Sci 1987;231(1264):359–69. [11] Fullerton CS, Ursano RJ, Epstein RS, Crowley B, Vance K, Kao TC, et al. Gender differences in posttraumatic stress disorder after motor vehicle accidents. Am J Psychiatry 2001;158(9):1486–91. [12] Medina AM, Mejia VY, Schell AM, Dawson ME, Margolin G. Startle reactivity and PTSD symptoms in a community sample of women. Psychiatry Res 2001;101(2):157–69. [13] Mouihate A, Chen X, Pittman QJ. Interleukin-1beta fever in rats: gender difference and estrous cycle influence. Am J Physiol 1998;275(5 Pt 2): R1450–4. [14] Mouihate A, Pittman QJ. Neuroimmune response to endogenous and exogenous pyrogens is differently modulated by sex steroids. Endocrinology 2003;144(6):2454–60. [15] Servatius RJ, Ottenweller JE, Bergen MT, Soldan S, Natelson BH. Persistent stress-induced sensitization of adrenocortical and startle responses. Physiol Behav 1994;56(5):945–54.
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023 [16] Servatius RJ, Ottenweller JE, Natelson BH. Delayed startle sensitization distinguishes rats exposed to one or three stress sessions: further evidence toward an animal model of PTSD. Biol Psychiatry 1995;38(8):539–46. [17] Shors TJ, Lewczyk C, Pacynski M, Mathew PR, Pickett J. Stages of estrous mediate the stress-induced impairment of associative learning in the female rat. Neuroreport 1998;9(3):419–23. [18] Toufexis DJ, Davis C, Hammond A, Davis M. Progesterone attenuates corticotropin-releasing factor-enhanced but not fear-potentiated startle via the activity of its neuroactive metabolite, allopregnanolone. J Neurosci 2004;24(45):10280–7. [19] Wood GE, Beylin AV, Shors TJ. The contribution of adrenal and reproductive hormones to the opposing effects of stress on trace conditioning in males versus females. Behav Neurosci 2001;115(1):175–87.
1023
[20] Wood GE, Shors TJ. Stress facilitates classical conditioning in males, but impairs classical conditioning in females through activational effects of ovarian hormones. Proc Natl Acad Sci U S A 1998;95(7):4066–71. [21] Xia-Zhang L, Xiao E, Ferin M. A 5-day estradiol therapy, in amounts reproducing concentrations of the early-mid follicular phase, prevents the activation of the hypothalamo-pituitary-adrenal axis by interleukin-1 alpha in the ovariectomized rhesus monkey. J Neuroendocrinol 1995;7(5): 387–92. [22] Xiao E, Xia-Zhang L, Ferin M, Wardlaw SL. Differential effects of estradiol on the adrenocorticotropin responses to interleukin-6 and interleukin-1 in the monkey. Endocrinology 2001;142(7):2736–41. [23] Sharp PE, La Regina MC. The laboratory rat. CRC Press, 1998.
Physiology & Behavior 93 (2008) 1019 – 1023 www.elsevier.com/locate/phb
Estrus cycle stage modifies the presentation of stress-induced startle suppression in female Sprague–Dawley rats Kevin D. Beck ⁎, Xilu Jiao, Tara P. Cominski, Richard J. Servatius Neurobehavioral Research Laboratory (129), Veterans Affairs New Jersey Health Care System, East Orange, NJ 07018, United States Stress and Motivated Behavior Institute, Department of Neurology and Neurosciences, New Jersey Medical School – UMDNJ, Newark, NJ 07103, United States Received 10 July 2007; received in revised form 12 December 2007; accepted 11 January 2008
Abstract Tailshock stress causes transient reductions in startle reactivity, associative learning and open field activity in female rats in an ovarian hormone dependent manner. Others have shown estrogen modulation of associative learning by testing across the estrus cycle and pharmacological manipulations. Here we tested whether stress-induced suppression of startle reactivity can be attributed to circulating ovarian hormones. Female rats were tracked across the estrus cycle and subjected to the stressor (2 h periodic tailshock) the morning of diestrus, proestrus, estrus, or metestrus. Startle reactivity was tested 2 h following the cessation of the tailshock. Using a multi-stimulus protocol, we determined there were differences in startle sensitivity and responsivity. Following stressor exposure, estrus females exhibited reduced startle responsivity. In contrast, diestrus females exhibited increased sensitivity to the lowest acoustic stimulus. The results are discussed with respect to ovarian hormone regulation of the immune system and sensory reactivity during and following trauma that may lead to different abnormal behaviors in females in the wake of traumatic stress. Published by Elsevier Inc. Keywords: Female; Stress; Ovarian hormones; Estrus cycle; Startle
Because of the recent increase in both the number of women serving in the military and the number of troops returning from service in Iraq and Afghanistan with post-traumatic stress disorder (PTSD), it is necessary at this time to delineate sex differences in response to traumatic stress. For instance, one of the hallmark symptoms of PTSD is an exaggerated startle response [3]. However, this is not always evident in women. Medina and colleagues [12] reported that a population of abused women with PTSD has blunted startle responses. This suggests that stress-induced changes in sensory reactivity may not be as homogenous as thought in the past, at least in women. Much of the neurobiological bases of abnormal behavior following stressor exposure in males and females have been documented in rat models. For instance, following a period of
⁎ Corresponding author. Neurobehavioral Research Laboratory (129), VA NJ Health Care System, 385 Tremont Ave, East Orange, NJ 07018, United States. Tel.:+1 973 676 1000x3682; fax: +1 973 395 7114. E-mail address: [email protected] (K.D. Beck). 0031-9384/$ - see front matter. Published by Elsevier Inc. doi:10.1016/j.physbeh.2008.01.012
inescapable tailshock stress, males and females exhibit differing capabilities of associative learning, as exemplified by the acquisition of a conditional eyeblink response [5,19,20]. Stressed males learn the sound-blink association quicker than nonstressed males, but similarly stressed females acquire the conditional eyeblink response slower than nonstressed females. In addition to the effects on associative learning, this particular type of stressor also affects startle reactivity differentially across sex. Acoustic startle responses are suppressed for a period of time in female rats [6], whereas males exhibit enhanced startle responses (albeit several days later)[15,16]. We have established that an important element of stressinduced startle suppression in the female rats is the presence of ovarian hormones [6]. Both repeated tailshock exposure and systemic injections of interleukin-1β suppress the magnitudes of elicited startle responses in female rats, unless they are ovariectomized [6,7]. This reduction in startle reactivity appears to be solely through a reduction in motor responsivity because we have not observed any decrements in the sensitivity of the rats to startle in response to different acoustic intensities [6].
1020
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
This study was designed to further explore the role of ovarian hormones in the modification of startle reactivity in female rats following exposure to a traumatic stressor. Here, we attempted to discern whether differences in stress-induced startle suppression are evident across different stages of the estrus cycle. Shors and colleagues [17,19] have shown this type of stressor can differentially affect the acquisition of a conditional eyeblink response, depending on the stage of estrus during which the classical conditioning occurred. However, it should be noted that those researchers did not observe any changes in sensory reactivity following stress (assessed prior to conditioning). Thus, although they observed that rats in proestrus were most affected by stressor exposure (and those in diestrus were least affected), the effects of stress on learning through the ovarian hormones could be different than what occurs to startle reactivity following stressor exposure. Others have suggested progesterone metabolites may serve to dampen anxiety-induced increases in startle reactivity [18]. Estradiol levels rise during the later part of diestrus and peak during proestrus. Progesterone levels rise sharply a few hours later and initiate the beginning of estrus. Hence, we may expect stress-induced startle suppression to be more evident during either during the hormone rise in proestrus or the fall during estrus, if estradiol and/or progesterone are necessary to cause a blunting of startle responsivity. Moreover, we should not expect to observe a suppression of startle responsivity during metestrus or diestrus, when the ovarian hormones are circulating in much lower quantities.
tailshock has been described previously [6,16]. Briefly, it involves administering 40, 3-s, 2-mA electrical shocks through an electrode clipped 1–2 cm below the base of the tail of rats restrained in Plexiglas restrainers (Harvard Apparatus). This stimulus has been shown before to be sufficient to cause changes in startle responding in both male and female Sprague– Dawley rats. Two hours following the termination of stressor presentation, all rats were tested for startle reactivity. The startle procedures used are based on the methods of Blumenthal [9] that utilize multiple acoustic stimulus intensities in determining differences in the number of startles elicited at different intensities (sensitivity) in combination with the magnitudes of those responses (responsivity). LabView software was used to produce 8 white-noise acoustic stimuli, each of an intensity of 92, 96, and 102 dB (5 ms rise/fall, 100 ms duration) for a total of 24 trials. The electrical signals produced from the accelerometer during displacements occurred at a rate of 1 kHz, with A/D conversion into the computer for analysis by S-plus software (Insightful Corp). For each startle stimulus presentation, a response threshold for whole body response was computed as the average rectified activity 200 ms prior to stimulus onset plus 4 times the standard deviation of that rectified activity [16]. Response amplitudes, the maximum rectified activity within 125 ms after stimulus onset, were only recorded when poststimulus activity exceeded the response threshold. For trials in which activity did not reach this criterion “not available” was recorded, for all others, this calculated value was corrected by each rat's body weight (taken the morning of startle testing).
1. Methods 2. Results 1.1. Subjects Sixty-eight 3-month old female Sprague–Dawley rats obtained from Charles River (Wilmington, MA) served as subjects. All rats were allowed at least 10 days to acclimate to the facility and the 12:12 h light:dark cycle (lights on 0700). All the rats were well-handled over the following 10–14 days due to the vaginal smearing procedures. 1.2. Procedures All rats had vaginal smears to determine estrus cycle stage. This involved sampling the cells of the vaginal canal with sterile saline using a glass pipette. The recovered solution containing cells were placed on microscope slides, stained with Evan's blue, and dried. Subsequent histological analysis occurred by viewing the dried slides under medium-power microscope (Wild Heerbrugg). Cell descriptions, as described by Sharp and LaRegina [23], were used to classify rats as being in diestrus, proestrus, estrus, or metestrus. Once cycling patterns were observed, we administered a 2 h tailshock protocol to rats staged that morning. The number of rats determined to be in each estrus stage and assigned to control or stress condition were as follows: diestrus – control (11), diestrus – stress (9), proestrus – control (7), proestrus – stress (10), estrus – control (6), estrus – stress (5), metestrus – control (11), and metestrus – stress (9). For the stress manipulation, the
Startle sensitivity and startle responsivity were measured in differently staged female rats following exposure to tailshock. The comparisons between number of startles elicited and the magnitudes of the elicited startles were conducted using a 4 (Estrus Stage) × 2 (Stress) × 3 (Stimulus Intensity) mixed analysis of variance (ANOVA) with Fisher's LSD post-hoc tests (p b .05). The number of elicited startle responses was different depending of the intensity of the acoustic stimuli. As shown in Fig. 1, as intensity increased, the number of startle responses detected increased. This was confirmed by a significant main effect of Stimulus Intensity, F (2, 120)= 244.3, p b .001. However, an additional Estrus Stage × Stress × Stimulus Intensity interaction, F (6, 120)= 2.3, p b .05, suggested this pattern was not the same across groups. Post-hoc tests revealed a significant difference in the number of elicited startle responses to the lowest acoustic stimuli following stress exposure but only during diestrus. The diestrus females tested following stressor exposure had more startle responses than their same-stage control counterparts. For the other dependent measure, startle magnitudes, the calculated values were log-transformed in order to achieve the normality requirements of the ANOVA. As shown in Fig. 2, startle magnitudes were greater for the higher stimulus intensities, as evidenced by a significant main effect of Stimulus Intensity, F (2, 116) = 262.8, p b .001. In addition, both a Stress × Stimulus Intensity, F (2, 116) = 3.7, p b .05, and Estrus Stage × Stress × Stimulus Intensity interaction were significant, F (6, 116) = 2.2,
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
1021
p b .05. Post-hoc tests revealed a significant difference in startle magnitude due to stress exposure evident in response to the lowest intensity acoustic stimulus during estrus. When startle testing was conducted in estrus females following stressor exposure, the magnitude of the elicited startle responses was significantly lower than their same-stage control counterparts. 3. Discussion The current results further our understanding of the endocrine modulation of startle suppression by showing female rats at certain stages of the estrus cycle appear to be more or less likely to exhibit changes in startle sensitivity and responsivity. Startle suppression, reflected as blunted startle responsivity, following stressor exposure was observed. When stress occurred during estrus, startle suppression was most evident 2 h later during testing. In contrast, stressor exposure during diestrus was associated with an increase in startle sensitivity. Specifically, those rats startled more often to the lowest intensity stimulus, without exhibiting any significant differences in responsivity from their same-stage controls. This suggests that stress is causing rather different effects upon startle reactivity in females depending on the phase of the estrus cycle. The importance of this finding is that we now have evidence that the effect a painful stressor can have upon sensory reactivity can be quite different depending on Fig. 2. Female rats at 4 stages of the estrus cycle were either subjected to tailshock stress (shaded bars) or served as controls (open bars). Startle testing occurred 2 h following the termination of stressor exposure. Shown are the mean startle magnitudes of the startle responses elicited at each of 3 different stimulus intensities (in arbitrary units) for each condition for each estrus cycle group. An asterisk (⁎) represents a significant difference between stress and control groups at a particular stimulus intensity (p b .05).
Fig. 1. Female rats at 4 stages of the estrus cycle were either subjected to tailshock stress (shaded bars) or served as controls (open bars). Startle testing occurred 2 h following the termination of stressor exposure. Shown are the mean percentages of startle responses elicited at each of 3 different stimulus intensities for each condition for each estrus cycle group. An asterisk (⁎) represents a significant difference between stress and control groups at a particular stimulus intensity (p b .05).
the endocrine state of the female at the time of the trauma. This could explain, in part, why some women with PTSD develop exaggerated startle responses [11] while others develop suppressed startle reactivity [12]. Others have recently reported suppressed startle responses in male rats following a combination of injection and predatorexposure stressors [1,2]. In that model, both GABA and glucocorticoid receptors have been implicated as mechanisms for the reduction in startle reactivity [1,2]. We have not observed startle suppression in male rats exposed to the tailshock protocol described herein [6]; in fact, if it occurs daily for over 3 consecutive days, the startle becomes exaggerated in the male rats while still exhibiting characteristics of suppression in the similarly stressed female rats [4]. Still, mechanisms proposed by Adamec and colleagues [1,2] could be quite relevant for our female model, now that we have identified phases where startle sensitivity can be increased and startle responsivity decreased following stressor exposure. Our research thus far has focused on early immune cascade mechanisms involved in the response to tailshock that may affect neural circuitry involved with dampening sensory reactivity processes (such as IL-1β) [4–7], but the mechanisms proposed in the male predator-stress model can be highly influenced by circulating ovarian hormones as well. For instance, RU486 was
1022
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023
shown to block the suppression of the startle response elicited by predator exposure and a subsequent injection stress [1]. The authors proposed a glucocorticoid receptor mechanism, and the use of RU 486 to block the stress-induced suppressed startle, because the added injection-stress procedure increased corticosterone release above that of predator stress exposure alone [2]. We can also consider the other actions of RU 486, namely as a progesterone receptor antagonist. Core body temperature is increased by IL-1β for a longer period of time during proestrus (compared to diestrus) apparently because of the actions of progesterone [13] through hypothalamic cyclooxygenase [14]. Still, progesterone is elevated at the end of proestrus and decreases through the onset of estrus. Recalling that IL-1β causes startle suppression in intact females [7], progesterone receptor antagonism by RU 486 may block the stress-induced startle suppression because a progesterone-proinflammatory cytokine interaction is being blocked. This effect may be more prominent in females because of the added presence of estradiol. Constant estradiol treatment leads to a blunted HPA-axis response when stimulated with IL-1β, compared to OVX animals [21,22]. Thus, the antiinflammatory actions of corticosterone are reduced and, presumably, there is a reduced amount of central corticotrophinreleasing hormone (CRH) stimulated by peripheral IL-1β actions upon the vagus nerve. As expected, others have shown the combination of estradiol and progesterone will reduce the magnitude of CRH-enhanced startle [18]. With diestrus being the estrus phase with the least circulating estradiol and progesterone, it is not surprising in the current study that we saw an apparent increase in startle sensitivity and no reduction in startle responsivity during diestrus. Others have shown progesterone is the main cause of changes in the degree CRH can enhance startle reactivity when administered i.c.v. [18]. This is not likely through a direct action of progesterone upon its receptor, but through one of its metabolites (allopregnanolone) increasing GABA activity and acting as an endogenous anxiolytic [8,10,18]. With exposure to a traumatic stressor, it seems as though physical reactions occur that both enhance and dampen sensory reactivity, and it may be differences in this balance that determine the specific direction startle reactivity changes following stressor exposure. For example, the increase in startle sensitivity during diestrus could be due to more or greater excitatory processes (such as CRH) being recruited because of the lower levels of estradiol and progesterone. However, during estrus, when progesterone is reducing from its peak, it could be that the inhibitory mechanisms (described above) outweigh those supporting excitation and enhanced reactivity. Obviously, hormone manipulation studies need to occur in order to definitely determine the precise timing of the endocrine actions involved in regulating stress-induced startle suppression in female rats. The literature appears to support our initial impressions from the current study that progesterone is a likely candidate mechanism. Our future work is going to examine hormone replacement combinations with relation to tailshock stress and IL-1β treatment to discern if the suppressed startle from each of these treatments appears to be regulated by the same ovarian hormonal mechanisms. Understanding how ovarian
hormones influence startle reactivity (and sensory reactivity in general) following traumatic stress is important for the further development of stress models that take into account individual differences in stress reactivity that lead to different abnormal behaviors. Acknowledgment This research was supported by the Department of Veterans Affairs Merit Review research program to KDB and program support through the Stress and Motivated Behavior Institute (RJS). The described experimentation was conducted with approval by the VANJHCS Institutional Animal Care and Use and Research and Development Committees. Tara Cominski is currently a doctoral student in Neuroscience at UMDNJ – Robert Wood Johnson Medical School Piscataway, NJ. We also acknowledge the technical support provided by Toni Marie Dispenziere and Paul William Ong. References [1] Adamec R, Muir C, Grimes M, Pearcey K. Involvement of noradrenergic and corticoid receptors in the consolidation of the lasting anxiogenic effects of predator stress. Behav Brain Res 2007;179(2):192–207. [2] Adamec R, Strasser K, Blundell J, Burton P, McKay DW. Protein synthesis and the mechanisms of lasting change in anxiety induced by severe stress. Behav Brain Res 2006;167(2):270–86. [3] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th-TR ed. Washington, DC: American Psychiatric Association; 2000. [4] Beck KD, Brennan FX, Servatius RJ. Effects of stress on nonassociative learning processes in male and female rats. Integr Physiol Behav Sci 2002;37(2):128–39. [5] Beck KD, Servatius RJ. Stress and cytokine effects on learning: what does sex have to do with it? Integr Physiol Behav Sci 2003;38(3):179–88. [6] Beck KD, Servatius RJ. Stress-induced reductions of sensory reactivity in female rats depend on ovarian hormones and the application of a painful stressor. Horm Behav 2005;47(5):532–9. [7] Beck KD, Servatius RJ. Interleukin-1beta as a mechanism for stress-induced startle suppression in females. Ann NY Acad Sci 2006;1071:534–7. [8] Bitran D, Purdy RH, Kellogg CK. Anxiolytic effect of progesterone is associated with increases in cortical allopregnanolone and GABAA receptor function. Pharmacol Biochem Behav 1993;45(2):423–8. [9] Blumenthal TD, Berg WK. Stimulus rise time, intensity, and bandwidth effects on acoustic startle amplitude and probability. Psychophysiology 1986;23(6):635–41. [10] Callachan H, Cottrell GA, Hather NY, Lambert JJ, Nooney JM, Peters JA. Modulation of the GABAA receptor by progesterone metabolites. Proc R Soc Lond B Biol Sci 1987;231(1264):359–69. [11] Fullerton CS, Ursano RJ, Epstein RS, Crowley B, Vance K, Kao TC, et al. Gender differences in posttraumatic stress disorder after motor vehicle accidents. Am J Psychiatry 2001;158(9):1486–91. [12] Medina AM, Mejia VY, Schell AM, Dawson ME, Margolin G. Startle reactivity and PTSD symptoms in a community sample of women. Psychiatry Res 2001;101(2):157–69. [13] Mouihate A, Chen X, Pittman QJ. Interleukin-1beta fever in rats: gender difference and estrous cycle influence. Am J Physiol 1998;275(5 Pt 2): R1450–4. [14] Mouihate A, Pittman QJ. Neuroimmune response to endogenous and exogenous pyrogens is differently modulated by sex steroids. Endocrinology 2003;144(6):2454–60. [15] Servatius RJ, Ottenweller JE, Bergen MT, Soldan S, Natelson BH. Persistent stress-induced sensitization of adrenocortical and startle responses. Physiol Behav 1994;56(5):945–54.
K.D. Beck et al. / Physiology & Behavior 93 (2008) 1019–1023 [16] Servatius RJ, Ottenweller JE, Natelson BH. Delayed startle sensitization distinguishes rats exposed to one or three stress sessions: further evidence toward an animal model of PTSD. Biol Psychiatry 1995;38(8):539–46. [17] Shors TJ, Lewczyk C, Pacynski M, Mathew PR, Pickett J. Stages of estrous mediate the stress-induced impairment of associative learning in the female rat. Neuroreport 1998;9(3):419–23. [18] Toufexis DJ, Davis C, Hammond A, Davis M. Progesterone attenuates corticotropin-releasing factor-enhanced but not fear-potentiated startle via the activity of its neuroactive metabolite, allopregnanolone. J Neurosci 2004;24(45):10280–7. [19] Wood GE, Beylin AV, Shors TJ. The contribution of adrenal and reproductive hormones to the opposing effects of stress on trace conditioning in males versus females. Behav Neurosci 2001;115(1):175–87.
1023
[20] Wood GE, Shors TJ. Stress facilitates classical conditioning in males, but impairs classical conditioning in females through activational effects of ovarian hormones. Proc Natl Acad Sci U S A 1998;95(7):4066–71. [21] Xia-Zhang L, Xiao E, Ferin M. A 5-day estradiol therapy, in amounts reproducing concentrations of the early-mid follicular phase, prevents the activation of the hypothalamo-pituitary-adrenal axis by interleukin-1 alpha in the ovariectomized rhesus monkey. J Neuroendocrinol 1995;7(5): 387–92. [22] Xiao E, Xia-Zhang L, Ferin M, Wardlaw SL. Differential effects of estradiol on the adrenocorticotropin responses to interleukin-6 and interleukin-1 in the monkey. Endocrinology 2001;142(7):2736–41. [23] Sharp PE, La Regina MC. The laboratory rat. CRC Press, 1998.