Targeted disruption of the orphanin FQ/nociceptin gene increases stress susceptibility and impairs stress adaptation in mice

Abstract

The neuropeptide orphanin FQ (also known as nociceptin; OFQ/N) has been implicated in modulating stress-related behavior. OFQ/N was demonstrated to reverse stress-induced analgesia and possess anxiolytic-like activity after central administration. To further study physiological functions of OFQ/N, we have generated OFQ/N-deficient mice by targeted disruption of the OFQ/N gene. Homozygous mice display increased anxiety-like behavior when exposed to a novel and threatening environment. OFQ/N-null mice show elevated basal pain threshold but develop normal stress-induced analgesia. Interestingly, these mice show impaired adaptation to repeated stress when compared with wild-type mice, whereas their performance in spatial learning remained unaffected. Basal and poststress plasma corticosterone levels were found to be elevated in OFQ/N-deficient animals. Thus, OFQ/N appears to be crucially involved in the neurobiological regulation of stress-coping behavior and fear.

Physiological responses to stress include changes in behavior, sensory processing, and endocrine and metabolic homeostasis that are positively or negatively regulated by a multitude of neuronal pathways (1–7). An increased vulnerability to stress is discussed as a major contributing factor in human psychiatric disorders, such as anxiety and depression (8). At the hormonal level, these diseases often are accompanied by an overactivity of the hypothalamic-pituitary-adrenal (HPA) system (9, 10). However, the physiological basis for this dysregulation remains unclear. The recently discovered neuropeptide OFQ/N (11, 12) appears to alleviate behavioral and sensory responses to stress, such as fear responses (13) or analgesia (14). Further studies on the functions of OFQ/N in the neuronal processing of stress are hampered by the unavailability of a selective and high-affinity antagonist. Therefore, we took a genetic approach and generated OFQ/N-deficient mice that were analyzed for phenotypical differences in stress-related responses. The absence of OFQ/N increases stress-related variables of behavior and sensory processing, such as anxiety and nociceptive threshold, in genetically engineered mice. Mice lacking OFQ/N show elevated glucocorticoid levels, indicating a chronic activation of the HPA system that might contribute to the observed phenotypic changes. In addition, an important function of OFQ/N for stress adaptation was discovered, because OFQ/N-deficient mice failed to habituate after repeated exposure to stressful stimuli. These results suggest that the OFQ/N system may have important functions in the neural circuitry of stress processing.

MATERIALS AND METHODS

Gene Targeting and Animals. Exon 2 of the murine OFQ/N gene was cloned as a 6-kilobase HindIII fragment from a 129/Sv-derived genomic library in λFIX II (Stratagene) by using the rat OFQ/N cDNA as probe (15). The targeting vector was constructed in pUC19. After ApaI digestion of the cloned genomic DNA, 3′ overhangs were blunted with Klenow DNA polymerase, and a neomycin-resistance cassette was inserted, thus disrupting the ORF within the OFQ/N-coding sequence. A herpes simplex virus thymidine kinase cassette was cloned 3′ of the targeting construct as a negative selection marker. The targeting vector was linearized with NotI and electroporated into E14 embryonic stem cells (originally derived from 129 Ola mice) as described (16). After selection with G418 and gancyclovir, positive clones were identified by Southern blotting of SphI-digested genomic DNA by using a 900-bp SphI–HindIII fragment (5′ probe). Homologous recombination occurred in eight of 192 screened clones (targeting frequency ≈4%). Two independent embryonic stem cell clones, E14/OFQ/N-125 and E14/OFQ/N-190, were microinjected into C57BL/6-derived blastocysts, and both gave rise to chimeric offspring. Four chimeric males (two from each embryonic stem cell clone) transmitted the mutation on breeding with C57BL/6 female mice. Nonlitter F₁ and F₂ mice were intercrossed to obtain F₂ and F₃ mice, respectively, on a 129/Ola × C57BL/6 hybrid background. Experiments were performed with F₂ and F₃ hybrids obtained from heterozygous breeding pairs. Homozygous mutant mice were mated to investigate potential effects of the mutation on fertility and parental care.

Unless indicated differently, male mice (10–16 weeks old) from each litter were group-housed (4–6 animals per cage) under controlled conditions (temperature 20 ± 2°C; relative humidity 50–60%; 12-h light-dark cycle, lights on 7:00 a.m.). Single-housed mice were isolated for at least 2 weeks. Animals had free access to food and water. All animal experiments had been approved by a local ethics committee and were done in accordance to German federal regulations and guidelines on animal experimentation.

Immunohistochemistry.

Animals were anaesthetized and perfused with saline followed by 4% paraformaldehyde/0.2% picric acid in 0.1 M phosphate buffer, pH 6.9. Brains and peripheral organs were removed and postfixed. Immunocytochemical staining was performed with 40-μm cryostat sections by using a rabbit polyclonal anti-OFQ/N antiserum as described (17). Positive labeling was visualized by immunofluorescence after staining with Cy3-coupled streptavidin (excitation: 568 nm, emission: 570–630 nm band-pass filter).

Biochemical Analyses.

Receptor binding was done on mouse brain membranes as described by using ¹²⁵I-labeled [Tyr¹⁴]OFQ/N as a radioligand (18). For determination of plasma corticosterone, trunk blood was collected after decapitation in aprotinin/EDTA-coated tubes between 7:00 and 8:00 a.m. (basal) or 5 min after a 10-min exposure of naive mice to the elevated plus-maze (mild stress), respectively. Plasma corticosterone was measured by RIA (DRG, Marburg, Germany).

Anxiety-Like Behavior. The open field consisted of four adjacent activity chambers (each 50 × 50 × 50 cm) monitored by an automated video motility system (Video-Mot II, TSE, Bad Homburg, Germany). Locomotor activity was recorded over 10 min. The imaginary central zone was defined as a 15 × 15 cm square in the middle of each observation area. Illumination in the central zone was 150 lux.

The plus maze consisted of two open (30 × 5 cm) and two wall-enclosed arms (30 × 5 × 15 cm) connected by a central platform (5 × 5 cm). Light intensity on the open arms was 150 lux. The apparatus was elevated 75 cm above the floor. Behavioral testing was started by placing a mouse in the central area facing a closed arm in which the animal usually enters first. Exploratory behavior was monitored by the same video motility system as for the open field over a period of 5 min. Numbers of entries into open arms, time and distance traveled in open and closed arms, general activity, and latency until the first open-arm entry were recorded and quantified automatically. Entries were defined as the body center of an animal entering a new zone.

The light–dark box was divided into a lit compartment (30 × 20 × 25 cm, illumination 150 lux) and a dark compartment (15 × 20 × 25 cm) connected by a 4-cm tunnel. The experiment was started by placing the animal in the dark compartment. Numbers of transitions, total time spent in the lit compartment, and latency until the first exit were quantified and recorded on video during 5 min.

Pain Perception and Stress-Induced Analgesia.

The tail-flick test was performed by using an electronically controlled tail-flick analgesiameter (Ugo Basile, Comerio-Varese, Italy) according to standard procedures. In all tests, the cutoff time of the IR radiant heat source (50 W, setting 17) was fixed to 21 sec to prevent tissue damage. Stress was induced by forced swimming for 10 min in 18°C water. Mice were dried with a cotton towel before testing.

Spatial Learning and Memory.

Mice were trained to find a submerged platform (7-cm diameter, 1 cm below surface) in a circular pool (diameter, 80 cm; height, 30 cm) filled with milky water (depth, 20 cm; 20 ± 1°C) (19, 20). External visual cues were placed around the pool to facilitate navigation of the animals. Each mouse was placed in the water facing the wall of the pool in a fixed starting position randomly chosen from four and allowed to swim for 1 min to reach the platform. Mice failing to complete the task were placed on the platform manually. All mice were allowed to rest there for 20 sec. Mice performed four consecutive trials per session with three sessions per day (2-h interval between sessions) over a 3-day training period. The time to reach the target was measured (escape latency) and the swim path and speed for each mouse were recorded automatically. Directly after the last trial on day 3, the platform was removed and each mouse had to swim freely for 1 min. The time each mouse spent in the different quadrants of the pool and the swim path were recorded.

Data Analysis.

Differences between genotypes for biochemical parameters and signs of anxiety-like behavior were analyzed by ANOVA and post-hoc Scheffe’s test. MANOVA was used to detect correlations across the different tests of anxiety-like behavior. In paradigms testing stress-induced analgesia and spatial learning results were analyzed by MANOVA. Poststress analgesia values of the two genotypes were compared by univariate F test and planned contrast.

RESULTS

Generation of OFQ/N^−/− Mice. By using homologous recombination in embryonic stem cells, positively targeted embryonic stem clones were obtained, from which two were selected for injection into blastocysts to generate chimeric offspring (Fig. 1A). Mating of heterozygous F₁ mice produced homozygous offspring in a Mendelian fashion (Fig. 1B). Homozygous and heterozygous mice appeared healthy and grew normally. Preliminary examination of sensory and motor functions revealed no differences between genotypes. No gross anatomical differences could be detected in brain or other organs by examination of serial sections. Homozygous and heterozygous mice grew and bred normally and cared for their offspring.

Figure 1.

Disruption of the murine orphanin FQ/nociceptin gene and analysis of expression. (A) Knockout strategy. Restriction maps of the wild-type locus (Top), the targeting construct (Middle), and the mutant gene locus (Bottom) are shown. neo, neomycin resistance; HSV-tk, herpes simplex virus thymidine kinase; restriction sites: A, ApaI; H, HindIII; Sp, SphI. Horizontal bars show SphI restriction fragments of wild-type and mutant alleles detected by the 5′ probe. (B) Southern blot analysis of SphI-digested genomic DNA from mouse-tail biopsies hybridized with a 900-bp HindIII–SphI-fragment as 5′ probe. ctrl., control DNA from untransfected E14 embryonic stem cells; genotypes: +/+, wild type; +/−, heterozygous; −/−, homozygous. (C) Immunocytochemical analysis of OFQ/N expression in OFQ/N^+/+ and OFQ/N^−/− mouse brains. Coronal sections at the level of the hypothalamus are shown. Intense staining can be observed in OFQ/N^+/+ mice in the ventromedial hypothalamic nucleus, amygdala, and habenular nucleus. (Bar = 1 mm.)

Analysis of OFQ/N-like immunoreactivity in brain slices showed intense labeling of neuronal structures in wild-type but not in knockout animals (Fig. 1C). As reported (17), prominent staining was observed in hypothalamus, thalamus, amygdala, and brainstem nuclei in brain sections of wild-type animals. OFQ/N^−/− mice were devoid of any OFQ/N-like immunoreactivity. Expression levels and affinity of the OFQ/N receptor were found unchanged in wild-type and knockout mice as measured by binding on brain membranes (Table 1).

Table 1.

OFQ/N receptor binding and plasma corticosterone

Genotype	OFQ/N receptor binding				Plasma corticosterone
	Kd, pM	n	Bmax, fmol/mg protein	n	Basal, ng/ml	n	Mild stress, ng/ml	n
OFQ/N+/+	108 ± 38	4	251 ± 81	4	23.6 ± 4.9	12	195 ± 8.2	10
OFQ/N−/−	114 ± 45	4	272 ± 64	4	42.3 ± 2.2*	14	257 ± 14.0*	14

Values are mean ± SEM. ∗, P < 0.02 OFQ/N^−/− versus wild type. K_d, dissociation constant; B_max, maximal receptor-binding capacity. 

OFQ/N^−/− Mice Display Increased Anxiety-Like Behavior. As a consequence of exposure to stress, animals and humans often display behavioral symptoms of anxiety-like behavior. OFQ/N and its receptor are expressed in brain areas involved in the processing of stress responses and anxiety-related behavior, such as the hypothalamus, hippocampus, amygdala, bed nucleus of the stria terminalis, and locus coeruleus (15, 21, 22). We have shown recently that intracerebroventricular administration of OFQ/N produces anxiolytic-like effects in rats and mice (13). To investigate whether a complete lack of OFQ/N would affect emotional reactivity, we tested anxiety-like behavior of OFQ/N^−/− mice and their heterozygous and wild-type littermates in three different paradigms that have been validated for the detection of innate fear in rodents (23–25). These tests (open field, elevated plus maze, light–dark box) are based on the natural aversion of rodents toward novel or unprotected, and thus potentially dangerous, areas. Exposure to the experimental environment is sufficient to induce mild stress and fear at the same time (26). All behavioral experiments also were conducted in parallel with the respective parental strains (129/Ola and C57BL/6) to control for effects of genetic background on the mutant phenotype.

In the open-field test, OFQ/N^−/− mice displayed reduced locomotor activity and spent less time in the center of the observation area than did heterozygous and wild-type mice (Fig. 2A). A reduction in exploratory activity and time spent in the unprotected center of the open field can be interpreted as an increased level of anxiety (27). Also, in the elevated plus maze test, OFQ/N^−/− mice displayed high levels of anxiety-like behavior. Homozygous mice spent significantly less time exploring the open arms of the maze, made fewer entries into the open arms, and showed an increased latency until first entry of an open arm (Fig. 2B). Although OFQ/N-deficient mice were hypoactive in the open field, the distance traveled in the closed arms of the plus maze did not differ between genotypes (OFQ/N^+/+, 581 ± 40.1 cm; OFQ/N^+/−, 589 ± 61.4; OFQ/N^−/−, 506 ± 40.1 cm). Thus, it is unlikely that the observed effects in the open field can be explained by a general hypoactivity of OFQ/N^−/− mice. In the light–dark box, the fear-inducing stimulus mainly consists of the bright light that illuminates one of the compartments. OFQ/N^−/− mice spent more time in the dark compartment and showed significantly reduced numbers of crossings into the light as compared with OFQ/N^+/− and OFQ/N^+/+ animals. Also, the latency until first exit from the dark compartment was greater in OFQ/N^−/− mice than with wild-type mice (Fig. 2C). Significant differences between OFQ/N^−/− mice and their wild-type and heterozygous littermates were found across all tests of anxiety-like behavior, when the two comparable measures of time spent in the more illuminated part of the test apparatus (F_2,43 = 23.50; P < 0.001) and number of entries into the more illuminated part of the test apparatus (F_2,43 = 30.99; P < 0.001) were compared.

Figure 2.

Anxiety-like behavior in mutant mice lacking OFQ/N (−/−) and their heterozygous (+/−) and wild-type (+/+) littermates. (A) Open-field test. Numbers of entries in the central zone, time spent in the central zone, and total distance traveled during a 10-min test trial are depicted. OFQ/N^+/+, n = 12; OFQ/N^+/−, n = 12; OFQ/N^−/−, n = 20. (B) Plus maze assay. (Left) Entries into open arms (percent of all entries into new zones) and time spent in open arms during a 5-min trial are shown. (Right) Latency until first entry into an open arm is depicted. OFQ/N^+/+, n = 12; OFQ/N^+/−, n = 12; OFQ/N^−/−, n = 20. (C) Light-dark box. (Left) Scores for entries into the illuminated compartment and percentage of time spent in there during the 5-min trial period are shown. (Right) Latency until first exit from the dark compartment is depicted. OFQ/N^+/+, n = 12; OFQ/N^+/−, n = 12; OFQ/N^−/−, n = 20. All data are mean ± SEM. ∗, P < 0.05; ∗∗, P < 0.01.

OFQ/N^−/− mice thus displayed significantly higher levels of anxiety-like behavior in all three behavioral assays than their wild-type littermates. Because we have selectively disrupted the OFQ/N gene, these effects can be attributed to a lack of OFQ/N and the adaptive consequences resulting thereof. Heterozygous mice were statistically indistinguishable from wild-type mice in all paradigms tested. The parental strains, C57BL/6 and 129 Ola, displayed differences in anxiety-like behavior that had been reported before (27). However, no consistent correlation of the mutant or wild-type phenotype with a specific parental strain could be detected across all three tests of anxiety-like behavior (data not shown). Therefore, our results cannot be simply interpreted by the influence of a predominant genetic background.

Nociceptive Processing in OFQ/N^−/− Mice.

Stimulated by the high degree of homology between OFQ/N and its receptor with the endogenous opioid peptides and their corresponding receptors, a lot of attention has been focused on a possible role of OFQ/N in nociception (28). When we analyzed pain perception in the tail-flick test, OFQ/N^−/− mice showed elevated basal nociceptive thresholds as compared with heterozygous and wild-type littermates (Fig. 3A). OFQ/N^−/− mice developed a normal dynamic and temporal pattern of stress-induced analgesia (SIA) after forced swimming (Fig. 3B). However, their tail-flick latency values were consistently elevated as compared with wild-type littermates. Thus, the nociceptive system in OFQ/N^−/− mice appears to function properly under both resting and challenged conditions.

Figure 3.

Nociceptive responses in mutant mice lacking OFQ/N (−/−), their heterozygous (+/−) and wild-type (+/+) littermates determined by the tail-flick test. (A) Basal tail-flick latencies in group-housed and single-housed (isolated) mice of the different genotypes. OFQ/N^+/+ (group), n = 15; OFQ/N^+/+ (isolated), n = 10; OFQ/N^+/− (group), n = 12; OFQ/N^−/− (group), n = 16; OFQ/N^−/− (isolated), n = 12. All data are mean ± SEM. ∗∗, P < 0.01 (group-housed OFQ/N^−/− mice versus all other groups). (B) Temporal pattern of stress-induced analgesia after first (Left) and third (Right) exposure to the stressor (one stress session per day). Mice were stressed by 10 min of forced swimming, and tail-flick latencies were measured 5 min before (basal), immediately after (stress), and at indicated intervals. ○, OFQ/N^+/+, n = 15; ■, OFQ/N^−/−, n = 16. (Left) MANOVA revealed significant main effect of treatment (F_3,27 = 122.94; P < 0.001) and group (F_1,29 = 23.47; P < 0.001). Univariate F test indicated that OFQ/N^−/− mice had higher baseline and poststress values (P < 0.05). Compared with basal values, tail-flick latencies were increased at 0 and 15 min after stress (P < 0.05). (Right) MANOVA revealed significant main effect of treatment (F_3,25 = 12.50; P < 0.001) and group (F_1,27 = 136.89; P < 0.001) and interaction effect of treatment by group (F_3,25 = 6.13; P < 0.003). Tests with contrast indicate an increase in tail-flick latencies in OFQ/N^−/− mice at 0 and 15 min poststress (P < 0.05).

OFQ/N^−/− Mice Fail to Adapt to Repeated Stress.

One reason for the elevated basal pain thresholds in OFQ/N^−/− mice could be an increased sensitivity to handling during the test procedure, thus precipitating SIA. In fact, it is very difficult to assess basal pain thresholds without producing some stress to the animal at the same time. Another possible explanation might be an increased level of social stress in group-housed OFQ/N^−/− mice because of natural hierarchical fighting. To distinguish between these two potential sources of SIA, we tested pain sensitivity of OFQ/N^−/− and OFQ/N^+/+ mice that had been isolated beforehand for more than 2 weeks. Interestingly, this procedure reduced basal pain thresholds of OFQ/N^−/− mice to a level indistinguishable from wild-type mice but had no effect on wild-type mice, suggesting an increased stress level in group-housed OFQ/N-deficient mice (Fig. 3A). Furthermore, repeated exposure to the swim-stress paradigm failed to produce adaptation in OFQ/N^−/− mice, whereas wild-type mice developed almost complete tolerance to the stressor after three trials (Fig. 3B). These results are in agreement with a recent report demonstrating that intracerebroventricular administration of an OFQ/N antiserum could reverse tolerance to electroacupuncture-induced analgesia (29).

Spatial Learning and Memory in OFQ/N^−/− Mice.

The development of tolerance to SIA has been reported to involve associative learning processes (30, 31). To investigate the possibility that the observed impairment of stress adaptation could be a consequence of cognitive deficits caused by the mutation, we examined spatial learning and memory performances in OFQ/N-deficient mice. Using the Morris water maze task, which can be assumed to impose a similar level of repeated stress as the swim-stress procedure to the animals, we could not detect any difference between mutant and wild-type animals (Fig. 4). Both groups learned to find the hidden platform and retained this memory equally well. Search strategy and swim speed were similar in both genotypes (data not shown). When the platform was removed after the final trial, wild-type and mutant mice also displayed similar search times in the target quadrant (data not shown). These data indicate that spatial cognition is normal in mutant mice. However, because the water-maze paradigm is analyzing only a subset of learning and memory types, we cannot exclude other cognitive deficits produced by the mutation that could ultimately influence their response to repeated stress.

Figure 4.

Spatial learning in mutant mice lacking OFQ/N (−/−) and their wild-type (+/+) littermates in the Morris water maze paradigm. Data are expressed as mean latency ± SEM for each session (each session consisting of four individual trials). ○, OFQ/N^+/+, n = 11; ■, OFQ/N^−/−, n = 19. No significant difference between genotypes was detected.

Elevated Plasma Corticosterone Levels in OFQ/N^−/− Mice.

At the biochemical level, the increased stress susceptibility of OFQ/N^−/− mice is reflected by elevated basal as well as poststress plasma corticosterone levels as compared with wild-type littermates (Table 1). These findings suggest that the OFQ/N system might provide an inhibitory input to the HPA system, which constitutes the main integrator of neural processing of stress.

DISCUSSION

In the present study, we have analyzed phenotypic changes in mutant mice lacking the neuropeptide OFQ/N. OFQ/N-deficient mice display behavioral, sensory, and endocrine symptoms of an increased stress susceptibility. In addition, adaptive responses to repeated stress were significantly impaired in the mutant animals. No gene-dosage effect could be detected, because OFQ/N^+/− mice were indistinguishable from wild-type animals. Because administration of OFQ/N to normal animals can attenuate behavioral responses to stress, our results obtained with homozygous mutant mice support the view that OFQ/N is an integral constituent of the neuronal systems regulating physiological responses to stress (13, 32).

Repeated confrontation of an individual with a stressor that turns out to be not life-threatening normally leads to a gradual decline of the stress response, known as adaptation (33–35). However, OFQ/N^−/− mice failed to produce stress adaptation, suggesting that OFQ/N might be required for the development of adequate coping strategies to repeated stress. Such a maladaptation could account for a tonic induction of SIA and would also explain the increased anxiety and elevated glucocorticoid levels of mutant mice. A potential source of chronic stress could be the social environment produced by group housing of the animals. Alternatively, we cannot exclude that the absence of OFQ/N could affect the social dynamics of the mutant mice in a way that would indirectly influence their stress-coping behavior. However, our observations of maladaptation to repeated physical stressors such as swim stress indicate a more direct involvement of the OFQ/N system in stress processing. Other adaptive neural processes that also depend on adequate responding to novel or aversive and thus stressful situations are, for example, memory formation and addiction (36–38). Our observations of impaired stress adaptation in OFQ/N-deficient mice are suggestive of an involvement of OFQ/N in processes of neural plasticity. Indeed, it has been reported that intrahippocampal administration of OFQ/N can impair spatial learning in rats (39).

Manifestation of tolerance to some forms of SIA is accompanied by cross-tolerance to morphine, indicating that similar neuronal mechanisms might be involved (34). A number of reports also have examined the consequences of concurrent stress during the development of tolerance to morphine-induced analgesia. It was shown that stress (40) as well as administration of stress-related hormones such as corticotropin (41) or glucocorticoids (42–44) could attenuate opiate tolerance development. Because OFQ/N-deficient mice show chronically elevated plasma levels of corticosterone, their failure to adapt to repeated stress could be produced by a similar mechanism.

Recently, mice devoid of the ORL1 receptor (the biochemical target of OFQ/N) were generated. However, analysis of their phenotype did not demonstrate changes in basal nociception or anxiety-related behavior (45, 46) but revealed an improvement in spatial attention and memory (46, 47). It is interesting to note that mice lacking the ORL1 receptor failed to develop morphine tolerance (48), indicating that the OFQ/N system is involved in adaptive processes after chronic stimulation of the opiate system. The observation that a receptor knockout produces a different phenotype than the inactivation of the corresponding neuropeptide/ligand precursor protein could indicate the presence of additional receptor subtypes or related ligand molecules. The OFQ/N precursor protein has been reported to encode two other bioactive neuropeptides, nocistatin (49) and OFQ/N II (50). It could thus be that the biological activities of both nocistatin and OFQ/N II, which are not mediated by the ORL1 receptor, might contribute to the different phenotype observed in ORL1-deficient mice as compared with OFQ/N^−/− mice. Also, because the receptor knockout and the ligand knockout were generated in different embryonic stem cell lines, an influence of the divergent genetic background on the expressed phenotypes cannot be excluded.

It has been proposed that OFQ/N might function as an anti-opioid in pain processing because it is able to reverse opioid-mediated SIA after intracerebroventricular administration (14). However, a complete lack of OFQ/N does not impair nociceptive processing. If one assumes that in nature the sensation of fear and the likelihood of painful injury often coincide, e.g., during confrontation with a natural predator, the opioids and OFQ/N can rather be viewed as synergistic neurotransmitter systems acting to reduce stress vulnerability. Both suppression of excessive fear and pain are vital for the individual to cope with a potentially hazardous situation and maintain its ability to react adequately. It is therefore not surprising that both neuropeptide systems display such a high degree of structural and evolutionary similarity. This hypothesis could be tested once a selective OFQ/N antagonist will be available.

In conclusion, we have shown that OFQ/N-deficient mice display an impairment of behavioral, sensory and endocrine responses to acute and repeated stress. This evidence strongly suggests that OFQ/N is an important regulator in the neurobiological processing of stress responses, functionally opposing the stress-promoting effects of the HPA system. OFQ/N-deficient mice could provide a useful model for the development of novel therapeutic strategies in the treatment of psychiatric disorders that are accompanied by an increased stress susceptibility.

ABBREVIATIONS

OFQ/N

orphanin FQ/nociceptin

HPA

hypothalamic–pituitary–adrenal

SIA

stress-induced analgesia