Address correspondence and reprint requests to Dr. Kenji Yamamoto, Department of Pharmacology, Graduate School of Dental Science, Kyushu University Fukuoka 812-8582, Japan. E-mail: email@example.com
Cathepsin E is an endolysosomal aspartic proteinase predominantly expressed in cells of the immune system, but physiological functions of this protein in the brain remains unclear. In this study, we investigate the behavioral effect of disrupting the gene encoding cathepsin E in mice. We found that the cathepsin E-deficient (CatE−/−) mice were behaviorally normal when housed communally, but they became more aggressive compared with the wild-type littermates when housed individually in a single cage. The increased aggressive response of CatE−/− mice was reduced to the level comparable to that seen for CatE+/+ mice by pretreatment with an NK-1-specific antagonist. Consistent with this, the neurotransmitter substance P (SP) level in affective brain areas including amygdala, hypothalamus, and periaqueductal gray was significantly increased in CatE−/− mice compared with CatE+/+ mice, indicating that the increased aggressive behavior of CatE−/− mice by isolation housing followed by territorial challenge is mainly because of the enhanced SP/NK-1 receptor signaling system. Double immunofluorescence microscopy also revealed the co-localization of SP with synaptophysin but not with microtubule-associated protein-2. Our data thus indicate that cathepsin E is associated with the SP/NK-1 receptor signaling system and thereby regulates the aggressive response of the animals to stressors such as territorial challenge.
Previous in vitro studies have demonstrated that cathepsin E is able to degrade certain types of biologically active neuronal peptides, such as substance P (SP, neurokinin-1) and neurokinin A at weakly acidic pH, maximally at pH 5, in which the rate of their degradation were found to be several hundred-fold higher than that of the analogous lysosomal aspartic proteinase cathepsin D (Kageyama 1993). Furthermore, cathepsin E has been shown to process the precursors to neurotensin and related peptides much more rapidly and more specifically than do cathepsin D and other aspartic proteinases (Kageyama et al. 1995). These observations suggest that cathepsin E may be involved in regulation of the brain levels of these bioactive peptides. In the current study, to better understand the role of cathepsin E in brain, we have investigated the effect of disrupting the gene encoding this protein on the behavioral responses to major stressors such as territorial challenge. We herein report that cathepsin E deficiency leads to the increased aggressive response to territorial challenge in mice housed individually in a single cage. We found that this aggressive response was mainly because of the enhanced SP/NK-1 receptor signaling system by cathepsin deficiency. To address the mechanism underlying these consequences, we investigated the expression of SP in the affective brain areas of CatE−/− mice, including amygdala, hypoccampus, and periaqueductal gray (PAG), in comparison with that of the wild-type littermates. We also examined the cellular localization of SP in amygdala by double labeling immunofluorescence microscopy. Our results indicate that cathepsin E is associated at least in part with the regulation of aggressive responses of the animals to major stressors such as territorial challenge through the SP/NK-1 receptor signaling system.
Wild-type and CatE−/− male mice on C57BL/6 genetic background (8–11 weeks) were used as described previously (Yanagawa et al. 2007). All animals were housed in a temperature- and humidity-controlled environment with a 12-h light and dark cycle and free access to food and water under specific pathogen-free (SPF) conditions according to the guidelines of the Japanese Pharmacological Society. The animals and all experiments were approved by the Animal Research Committee of Graduate School of Dental Science, Kyushu University.
The behavior of mice was videotaped in an open field (55 cm × 65 cm × 50 cm) with lines dividing the floor area into 25 squares. Line crossing and the time spent in the nine central squares was counted in 15-min blocks from the videotapes. Locomotor activity was measured over a 24-h period, using automated activity boxes. For the anxiety activity assay, each genotype of mice was placed individually on an elevated plus-maze plate. The amount of time and frequency that the mice spent in the open arm was recorded. For the resident-intruder assay, male mice which had been reared in group cages (3–4 mice per cage) were isolated in single cages for 4 weeks. An isolated ‘intruder’ male mouse of either phenotype was introduced into the home cage of the isolated mouse of each genotype for 10 min. The aggressive responses of these mice, including the number of attacks, the latency to the first attack, and the amount of time that mice spent in vigorous combat, were measured. In some resident-intruder assays, the NK-1 receptor antagonist L-733,060 solved in 10% methyl cellulose solution was injected intraperitoneally into the two groups of animals after being housed individually in a single cage for 4 weeks, according to the method as described previously (Varty et al. 2003). At 30 min after injection with the NK-1 receptor antagonist, the behavior of mice was evaluated.
ELISA for SP
The amount of SP in the brain lysate of each genotype of mice was measured with the use of an ELISA kit (R&D Systems Inc., MN, USA). This assay is based on the competitive binding technique where SP in a sample competes with a fixed amount of horseradish peroxidase-labeled SP for sites on a mouse monoclonal antibody. In brief, under deep anesthesia with sodium pentobarbital (10 mg/100 g body weight, i.p.), animals were killed by decapitation and brains were removed rapidly and immersed in ice-cold phosphate-buffered saline (PBS), pH 7.4. Then the brain tissues were homogenized in PBS containing 0.1% Triton X-100 and 0.1% protease inhibitor cocktail (leupeptin, chymostatin, pepstatin A, antipain, and phenylsufonyl fluoride; 1 mg/mL for each) to give 25% homogenates. After sonication, the homogenates were centrifuged at 105 000 g for 30 min at 4°C. The resultant supernatant was subjected to ELISA assay according to the manufacturer’s instruction. Fifty microliters of sample or standard was added to each 96-well of microtiter plates, and then 50 μL of a mouse monoclonal antibody for SP. Then 50 μL of SP conjugated to horseradish peroxidase was added to the well. After incubation for 3 h at 22°C, the wells were washed four times with the wash buffer provided, and then incubated for 30 min at 22°C with 200 μL of substrate solution in dark. After terminating the reaction by the addition of 50 μL of 2 N sulfuric acid. The absorbance of each well at 450 nm was measured with a microtiter plate reader. It should be noted that the monoclonal antibody for SP employed cross-reacted exclusively with intact SP and that the cross-reactivity with its metabolites was very low (e.g., < 0.1% for SP1–7, < 0.1% for SP9–11).
Under deep anesthesia with sodium pentobarbital, mice were perfused through ascending aorta with 4% paraformaldehyde, 0.1% glutaraldehyde, and 0.2% picric acid in 0.1 M sodium phosphate buffer, pH 7.2 (PB), at 22°C. Serial coronal sections at 40 μm in thickness were cut with a vibratome microtome (Leica VT1000, Leica-Mycrosystems, Heiderberg, Germany) and processed for immunohistochemistry using the free-floating method. The sections were treated with rabbit polyclonal antibodies to SP (1 : 10 000 dilution, Immunostar, Inc., Hudson, WI, USA) and then biotinylated donkey anti-rabbit IgG (1 : 200 dilution, Jacson ImmunoResearch, West Grove, PA). Then the sections were rinsed in PBS and incubated with Vector ABC complex (1 : 200 dilution, Vector Laboratories, Burlingame, CA, USA) for 2 h. After several rinse in PBS, the sections were mounted and dehydrated with increasing gradients of ethanol and then coverslipped. Images of SP-like staining were taken by stereoscopic microscope (Olympus SZX12, Olympus, Tokyo, Japan).
For double labeling immunofluorescence microscopy, sections were pre-treated with 30% sucrose in PB and rapidly frozen-thawed using liquid nitrogen as described previously (Fukuda and Kosaka 2003;Shigematsu et al. 2006). The sections were then incubated overnight in PBS containing 1% bovine serum albumin and 0.1% Triton X-100 for 7 days with a mixture of rabbit polyclonal antibodies to SP (1 : 10 000 dilution, Immunostar, Inc., Hudson, WI, USA) and monoclonal antibody to MAP2 (1 : 250 dilution, Leinco Technologies, Inc, St. Louis, MI, USA) or monoclonal antibody to synaptophysin that was provided by Dr S.C. Fujita of Mitsubishi Kasei Institute of Life Science (1 : 1000 dilution). After several rinse in PBS, the sections were further incubated overnight with rhodamine red-conjugated donkey anti-rabbit IgG (1 : 100 dilution, Jackson ImmunoResearch, West Grove, PA, USA) or fluorescene isothicyanate (FITC)-conjugated anti-mouse IgG (1 : 100 dilution, Jackson ImmunoResearch). All steps were performed at 22°C, and all primary and secondary antibodies were diluted in PBS containing 1% bovine serum albumin and 0.1% Triton X-100. Permeation of each primary antibody into the deep part of sections was sufficient to rule out the possibility of false-negative staining in multiple immunolabelings. The sections were mounted in the anti-fading medium Vectashield, and then inspected by microscopy using a laser-scanning confocal system (Leica TCS, Leica-Mycrosystems, Germany). Control sections were prepared by omission of primary antibodies or by mismatching secondary antibodies. Either case provided only weak non-specific staining. Gray levels of SP-like immunoreactivity were measured using Image J software (v1.37r, NIH).
Quantitative data are presented as means ± SEM. In the experiments analyzing the effect of cathepsin E deficiency on the locomotor activity and the anxiety levels of animals, the results were compared between the two groups of mice with Mann–Whitney U-test. Results from the resident-intruder experiments were analyzed by Student’s t-test or one-way anova followed by post hoc comparison between each genotype of age-matched animals. P-values of < 0.05 were considered statistically significant.
Effect of cathepsin E gene disruption on mouse behaviors
CatE−/− mice showed no obvious phenotypes when raised under SPF conditions, and no gross abnormalities clearly related to genotype were noted at any age (Tsukuba et al. 2003). Terminal body weight or the weights of various tissues, including brain, kidney, heart, liver and spleen, did not differ between wild-type and CatE−/− mice. In addition, there were no significant differences in biochemical and immunological properties between age-matched wild-type and CatE−/− mice when reared under SPF conditions (Tsukuba et al. 2003). These include the numbers of neutrophils, lymphocytes, eosinophils, monocytes, erythrocytes, and platelets, and the serum concentrations of various immunoglobulins, such as IgG, IgM, and IgE, and cytokines such as interferon-γ and IL-2. CatE−/− mice also exhibit normal breeding behavior and have long-term survival rates comparable to those of their wild-type littermates when housed in group cages (3–4 mice per cage). However, our previous studies have demonstrated that CatE−/− mice spontaneously develop atopic dermatitis-like skin lesions when reared under conventional conditions (Tsukuba et al. 2003) and exhibit the increased susceptibility to bacterial infection accompanied by a marked decrease in killing of intracellular bacteria by macrophages (Tsukuba et al. 2006). Furthermore, the tumor growth and metastasis in CatE−/− mice bearing mouse B16 melanoma cells are more profound than that in wild-type littermates and the syngeneic mice over-expressing cathepsin E (Kawakubo et al. 2007). On the basis of these findings, we assume that certain kinds of inflammatory or nociceptive stimuli may cause altered cellular and behavioral responses for CatE−/− mice. To test this possibility, we first analyzed the locomotor activity of CatE−/− mice, as well as their wild-type littermates, fed ad libitum in group cages using an automated activity box, an apparatus which allows the recording of the movements and crossing of mice. As shown in Fig. 1a, the locomotor activity of CatE−/− mice was similar to that of the wild-type littermates. When anxiety was also evaluated on an elevated plus-maze plate, the amount of time and frequency spent in the open arm tended to decrease, but not statistically significant, in CatE−/− mice compared to CatE+/+ mice (Fig. 1b). Aggression was further assessed using the resident-intruder test, in which males of either genotype that had been isolated for 4 weeks in a single cage were exposed to each genotype of mice housed communally. Each genotype of age-matched male mice showed no obvious aggression when housed communally in group cages, as evidenced by the findings that both attack latency and incidence for CatE+/+ or CatE−/− mice were not observed during our experimental periods. Surprisingly, however, CatE−/− mice were found to be more aggressive compared with CatE+/+ mice when isolation-housed for 4 weeks in a single cage, as revealed by the resident-intruder assays, namely the attack latency was more intensely shortened in CatE−/− mice compared with CatE+/+ mice, and the total fighting scores and the duration of one fighting were more profound in CatE−/− mice compared with CatE+/+ mice (Fig. 1c).
Effect of the NK-1 receptor antagonist L-733,060 on the aggressive response of CatE−/− mice to territorial challenge
To test whether and to what extent the increased aggressive response to territorial challenge seen with CatE−/− mice subjected to isolation-housing for 4 weeks depends on the SP/NK-1 receptor signaling system, each genotype of isolation-housed mice was treated with the NK-1 receptor-specific antagonist L-733,060 for 30 min and then subjected to the resident-intruder test. The antagonist itself had no significant effect on the increased aggression of the isolation-housed CatE+/+ mice, as evaluated by attack latencies, attack incidence and duration of one fighting by (Fig. 2). However, this agent resulted in a dose-dependent reduction in the increased aggression of the isolation-housed CatE−/− mice, as evidenced by prolonged attack latencies and decreased total fighting scores, as well as reduced duration of one fighting. At 10 mg per kg, the antagonist markedly reduced the aggressive response of the isolation-housed CatE−/− mice to the level comparable to that seen in non-isolated CatE+/+ mice. We thus conclude that the increased aggressive response to territorial challenge seen in the isolation-housed CatE−/− mice totally depends on the SP/NK-1 receptor signaling system.
Elevated brain SP concentrations in CatE−/− mice
Recent studies with pharmacologic or genetic inactivation of NK-1 receptors have demonstrated that the SP/NK-1 receptor system plays a substantial role in the modulation of stress-related, affective and/or anxious behaviors and suggest that inhibition of this pathway may be a useful approach to the treatment of affective disorders (Saria 1999; Santarelli et al. 2002; Bilkei-Gorzo and Zimmer 2005). In addition, the SP content in the affective brain areas has been shown to be variable upon application of various stressful stimuli. SP is one of the most abundant neurokinin peptides in the mammalian CNS, which has the sequence of Arg-Pro-Lys-Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met-amide and is implicated in a variety of physiological and pathological processes through the SP/NK-1 receptor signaling system, such as stress regulation and affective- and anxiety-related behaviors. Furthermore, SP and the NK-1 receptor have been shown to be localized predominantly in brain areas known to be involved in the regulation of stress and anxiety responses, such as hypothalamus, amygdala, hippocampus, and nucleus accumbens (Cuello and Kanazawa 1978; Ljungdahl et al. 1978; Inagaki et al. 1982; Marchand et al. 1991; McLean et al. 1991; Nakaya et al. 1994). Therefore, we examined the effect of cathepsin E deficiency on SP content in the brain extract of wild-type and CatE−/− mice by an enzyme immunoassay. While the SP content in CatE+/+ mice was slightly increased between 6 and 24 weeks of age, that in the CatE−/− mice was significantly and age-dependently increased up to 12 weeks of age and then attained a plateau value (Fig. 3a). The mean SP levels were apparently higher in CatE−/− mice than in CatE+/+ mice. We further analyzed the SP content in the brain of either genotype at 12 weeks of age that had been isolated for 4 weeks in a single cage. The SP content in the brain of either genotype of mice when housed alone in a single cage was significantly increased (Fig. 3b). The ratio of SP levels between before and after isolation housing was similar between the wild-type and CatE−/− mice (1.59 vs. 1.72) (Table 1). However, the NK-1 receptor antagonist had no significant effect on the increased aggressive response of the isolation-housed CatE+/+ mice, whereas this agent strongly reduced the increased aggressive response of the isolation-housed CatE−/− mice. On the basis of these observations, it is assumed that the threshold for the aggressive response depending upon the SP/NK-1 receptor signaling system is above 737 pg of SP per brain. Given that the monoclonal antibody for SP employed cross-reacted exclusively with intact SP, the values determined by this ELISA assay was thought to represent the amounts of intact SP. Therefore, our results indicate that un-degraded SP markedly accumulates in the brain by the disruption of cathepsin E gene and is further increased by isolation housing, resulting in the enhancement of the SP/NK-1 receptor signaling system followed by the increased aggressive response to territorial challenge.
Table 1. Effect of isolation housing on SP levels in the brain extract of wild-type and CatE−/− mice
Isolation housing for 4 weeks
Each genotype of mice were housed individually in a single cage for 4 weeks, and then SP levels in the brain extract of each animals were determined by ELISA. Data are the mean ± SE of values from three independent experiments using three mice of each genotype (n =9). *p <0.001 for the values of the corresponding animals before isolation housing, ¶p <0.001 for indicated comparisons.
464.0 ± 42.8
736.6 ± 153.5 *
1148.8 ± 128.7
1979.5 ± 211.7 *
Localization of SP-like immunoreactivity
Given the predominant expression of SP and the NK1 receptor in brain areas known to be involved in the regulation of stress and anxiety responses, such as the hypothalamus, amygdale, hippocampus, and nucleus accumbens, we examined the immunohistochemical localization of SP-like peptides using coronal sections of brain tissues from each genotype of mice. No significant immunohistochemical differences were observed in the distribution of SP-like staining in the brain of CatE+/+ and CatE−/− mice (Fig. 4a). However, the intensity of SP-like immunoreactivity in areas of the hypothalamus, amygdala, and PAG was more profound in the CatE−/− mice compared with the CatE+/+ mice. These observations were further substantiated by a confocal microscopic study. The SP-like immunoreactivity in the amygdala, hypothalamus, and PAG of CatE−/− mice was significantly higher than that of CatE+/+ mice (Fig. 4b and c). To further assess the cellular localization of SP in these brain areas, we performed double immunostaining between SP and microtubule-associated protein-2 (MAP2), a marker for neuronal soma and dendrite, or synaptophysin, a marker for synaptic terminal, in the central amygdala. The amygdala is a key brain structure associated with emotional responses and thought to be a target of tachykinin receptor agonists for exerting psychopharmacological actions. As shown in Fig. 5, SP was significantly co-localized with synaptophysin, but not with MAP2, in the amygdala of both groups. The intensity of SP-like immunoreactivity in the synaptic terminal of central amygdala was shown to be markedly increased by cathepsin E deficiency. The results are consistent with previous studies indicating that, in the mammalian CNS, SP is mainly localized in axons and their terminals (Cuello and Kanazawa 1978; Ljungdahl et al. 1978; Inagaki et al. 1982). Upon various stressor stimuli, therefore, SP is likely released from the synaptic terminal of neurons and binds NK-1 receptor and thereby exerts its modulatory effects.
In the present study, we provide evidence for the first time that cathepsin E is associated with the regulation of the response of mice to certain stressors induced by isolation-housing followed by invasion of territory mainly through the SP/NK-1 receptor signaling system. Although no significant differences in the locomotor activity and the anxiety level, which were evaluated by the open-field test and the plus-maze test, respectively, were detected in the two groups of mice, we found that CatE−/− mice were much more aggressive than the wild-type littermates when housed individually for 4 weeks in a single cage (Fig. 1). The increased aggression of CatE−/− mice was markedly reduced by pre-treatment with the specific NK-1 receptor antagonist L-733,060 (Fig. 2). However, the pre-treatment of isolation-housed CatE+/+ mice with this agent did not exhibit significant reduction of the increased aggressive postures, implying that the increase in the observed aggression in CatE+/+ mice is unlikely to be directly mediated by the SP/NK-1 receptor signaling system.
In this context, we found that brain SP levels were age-dependently increased in the absence of cathepsin E and that the increased SP was more evident in the brain areas known to be involved in the regulation of stress, depression and associated anxiety compared with other brain areas (Fig. 4). Although the increase of SP levels in the affective brain areas was more profound in CatE−/− mice than in CatE+/+ mice, either genotype did not show obvious aggressive responses when housed communally under normal conditions. However, the isolation-housing stress induced the more aggressive response in CatE−/− mice compared with CatE+/+ mice. This may be related to further increase of the SP levels enhanced in the absence of cathepsin E upon stimulation of the isolation housing (Table 1). SP is known to be widely distributed throughout the mammalian CNS but is also present in the peripheral tissues including cells of the endocrine and immune systems and the gastrointestinal tract (Watling and Krause 1993) and to be generated by proteolytic processing of the precursor preprotachykinin-A. Since the monoclonal antibody for SP employed cross-reacted exclusively with intact SP, we tentatively conclude that the increased SP levels in the brain of CatE−/− mice is probably because of the impaired catabolism of intact SP. Previous studies have demonstrated that the increase in the SP content is induced by short-term restraint stress, subcutaneous saline injection, and sequential removal from the home-cage or social isolation (Lisoprawski et al. 1981; Rosen et al. 1992; Brodin et al. 1994). The enhanced SP release was also observed in the medial nucleus of amygdala in response to a wide variety of stressors, including elevated platform exposure, immobilization stress, mild footshock or exposure to an unfamiliar environment (Bannon et al. 1986; Elliott et al. 1986). Like SP, the NK-1 receptor is known to be highly expressed in brain areas known to be involved in the regulation of affective behavior and neurochemical responses to stress (De Felipe et al. 1998). Mice lacking the NK-1 receptor have been shown to be less aggressive than the wild-type littermates, but anxiety was similar between the two groups of mice (De Felipe et al. 1998). Therefore, the more enhanced SP content in these brain areas by cathepsin E deficiency is most likely to be associated with the more aggressive response to territorial challenge through the enhanced SP/NK-1 receptor signaling system.
Although it is well established that the SP/NK-1 receptor system modulates affective behaviors and neurochemical responses to stress (Mantyh et al. 1984), less is known about how the SP/NK-1 receptor signaling system is regulated. To date, several different proteases are reported to convert or degrade SP. These include neutral endopeptidase 24.11 (Bourne et al. 1989), angiotensin-converting enzyme (Persson et al. 1995), the cell surface neutral endopeptidase 24.11 (Matsas et al. 1983; Okamoto et al. 1994; Sturiale et al. 1999), dipeptidyl peptidase IV (Kenny et al. 1976), and a SP-specific endopeptidase found in human cerebrospinal fluid (Persson et al. 1995). On the other hand, the SP and NK-1 receptor complex is internalized and transported into the endosome, where SP is dissociated from the NK-1 receptor (Grady et al. 1995) and then degraded in endolysosomal peptidases such as cathepsin D (Benuck et al. 1977; Grady et al. 1995). Besides its extracellular degradation, therefore, SP may be degraded in the endosomal/lysosomal system. Despite many efforts to identify the protease(s) capable of inactivating SP in CNS, it remains unclear which and to what extent proteases are actually involved in the proteolytic cleavage of SP in vivo. Given the ability of cathepsin E to effectively degrade SP at weakly acidic pH (Kageyama 1993), the present results suggest that this enzyme is also a candidate for SP degrading proteases.
Double immunostaining revealed that SP was exclusively confined to synaptophysin-positive vesicles of neuronal synaptic terminals in both the wild-type and CatE−/− mice (Fig. 5). It is therefore possible that cathepsin E is associated with the regulation of SP content in neuronal synaptic terminals. At the present time, however, we could not confirm the co-localization of cathepsin E with SP or synaptophysin in these vesicles of both groups by immunohistochemistry under normal breeding conditions, because of a very low expression of cathepsin E in neurons. However, considering that cathepsin E is up-regulated and secreted by various cell types such as microglia and macrophages in response to various stimuli (Nakanishi et al. 1993, 1994, 1997; Amano et al. 1995; Sastradipura et al. 1998; Nishioku et al. 2002; Yanagawa et al. 2006) and that cathepsin E can hydrolyze certain peptides at neutral pH with rather restricted specificity (Athauda et al. 1991; Athauda and Takahashi 2002), it is possible that cathepsin E liberated at the synaptic terminal may degrade SP extracellularly. Meanwhile, it is also possible that cathepsin E may degrade SP in the endosome/lysosome system of postsynaptic neurons. However, this case is unlikely, because SP was found to accumulate extensively in synaptophysin-positive vesicles, but not MAP-positive compartments.
Finally, cathepsin E may control the SP/NK-1 receptor signaling system by activation of subcortical brain nuclei known to be involved in emotional, behavioral, autonomic and antinociceptive reactions through vagal afferent input from the stomach. Previous studies have revealed that solitary tract nucleus (Danzer et al. 2004; Appleyard et al. 2005) and parabrachial nucleus (Michl et al. 2001a,b), which project SP to amygdala (Yamano et al. 1988; Block et al. 1989; Riche et al. 1990), respond to challenge of the gastric mucosa through vagal afferent input. Synthesis of preprotachykinin-A in these nuclei are known to increase by various stimuli, and the projection and accumulation of SP is induced in several nuclei including amygdala and hypothalamus. Given that cathepsin E is most abundant in the gastric mucosa (Sakai et al. 1989) and that cathepsin E deficiency appears to increase the sensitivity to various gastric mucosal stimuli (unpublished), cathepsin E may be indirectly involved in the regulation of SP content in the brain through vagal afferent input from the stomach.
In conclusion, although the association of cathepsin E with the regulation of the SP/NK-1 receptor signaling system was demonstrated in this study, we did not establish a direct link between cathepsin E and SP. Therefore, the possibility that the accumulation of SP in the affective brain areas of CatE−/− mice may exhibit compensatory changes secondary to the loss of cathepsin E is still not ruled out. To add strength to the connection between cathepsin E and SP, additional experiments investigating the effects of conditional or brain-specific cathepsin E knockout or cathepsin E gene knockdown may be required. Studies along this line are in progress; nevertheless, the present results provide new insight into the functional diversity in brain.