SEARCH

SEARCH BY CITATION

Keywords:

  • Hippocampus;
  • learning;
  • memory;
  • metalloendoproteases;
  • mutant mice;
  • neuropeptide biosynthesis;
  • posttranslational processing

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

Endothelin-converting enzyme (ECE)-2 is a metalloprotease that possesses many properties consistent with it being a neuropeptide-processing enzyme. This protease is found primarily in neural tissues, with high levels of expression in midbrain, cerebellum, hypothalamus, frontal cortex and spinal cord and moderate levels in hippocampus and striatum. To evaluate its role in neural function, mice have been generated lacking this enzyme. Physical appearance, autonomic reflexes, motor co-ordination, balance, locomotor activity and spontaneous emotional responses appear normal in these knockout (KO) mice. However, these mutants display deficits in learning and memory as evidenced by marked impairment in the Morris water maze. Knockout mice are also deficient in object recognition memory where they show delays in discerning changes in object location and in recognizing the introduction of a novel object. In this study, perseveration appears to interfere with learning and memory. Finally, mutants are impaired in social transmission of food preference where they show poor short-term memory and perturbations in long-term memory; the latter can be ameliorated by reminder cues. As ECE-2 has been implicated in Alzheimer’s disease, the deficits in learning and memory in the KO mice may provide unique insights into processes that may contribute to this disease and possible other disorders of cognition.

Neuroendocrine peptides are intercellular messengers that play critical roles integrating nervous, reproductive, endocrine, cardiovascular, gastrointestinal, pulmonary and behavioral functions (Strand 1999). Peptides are synthesized as large precursor proteins that undergo multiple posttranslational processing steps to generate bioactive peptides. Typically, neuropeptide processing begins with endoproteolytic cleavage by prohormone convertases at the C-terminal side of mono- or dibasic residues (i.e. -KR, -RR, -RXXR-), considered ‘classical’ sites of cleavage (Docherty & Steiner 1982).

Some neuropeptides are generated by processing at ‘nonclassical’ sites. These peptides are found among bulk-purified peptides from neuroendocrine tissues and are detected in brains of mice lacking specific processing enzymes (Che et al. 2001; Mizuno et al. 1980; Sigafoos et al. 1993; Yang et al. 1985). Additionally, an examination of precursor sequences from some endogenous peptides further supports the idea that nonclassical-processing sites are used for generation of bioactive peptides (Liu et al. 2001; Perry et al. 1997; Vilim et al. 1999).

Members of the metalloprotease family have long been implicated in processing at nonclassical sites (Molineaux & Wilk 1991). Endothelin-converting enzyme (ECE)-1 and ECE-2 possess substrate specificity consistent with processing at nonclassical sites, and ECE-2 has additional properties consistent with those of a neuropeptide-processing enzyme. For instance, ECE-2 is optimally active at pH 5.5, is localized to intracellular secretory compartments (Davenport & Kuc 2000; Emoto & Yanagisawa 1995; Russell & Davenport 1999) and exhibits restricted neuroendocrine distribution (Nakagomi et al. 2000; Yanagisawa et al. 2000). Recently, a panel of 42 peptides was screened as substrates for ECE-2 (Mzhavia et al. 2003). Processing occurred for 10 of these peptides at acidic pH, a pH consistent with the milieu of secretory vesicles (Moore et al. 2002). Thus, ECE-2 may process a number of peptide intermediates, leading to a variety of endogenous ligands. Interestingly, studies characterizing the intracellular degradation of β-amyloid peptides have suggested a role for ECE-2 in Aβ clearance in vivo (Eckman et al. 2003). Thus, ECE-2 may play a role in neurodegenerative disorders such as Alzheimer’s disease (Eckman & Eckman 2005; Weeraratna et al. 2007).

Mice lacking active ECE-2 have been generated by replacing the exon encoding the conserved zinc-binding motif with a neo-cassette using homologous recombination (Yanagisawa et al. 2000). The mutants are viable and live a normal life span. As ECE-2 knockout (KO) mice are fertile and show no gross developmental defects, they are ideally suited for evaluation of the role of ECE-2 in neuropeptide processing and behavior. In this study, we have examined the behavioral phenotype of KO mice and find that they possess no gross behavioral abnormalities but are impaired on various tests of learning and memory.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

Animals

Adult naïve male and female wild type (WT) and KO mice (aged 12–20 weeks) were used in all experiments. Mice were propagated by heterozygous C57BL6/J-129/SvJ matings and genotyped by polymerase chain reaction (PCR) using the following primers: WT ECE-2 allele (sense) 5′-GCCATCTTACAGTAGAGGAG-3′ and (antisense) 5′-CTAGAATGGGCCCCTACCTT-3′ and KO allele: (sense) 5′-TATTCGGCTATGACTGGGCACAACAG-3′ and (antisense) 5′-TTCCACCATGATATTCGGCAAGCAGG-3′. Offspring were weaned after 21 days, segregated by sex and genotype and housed two to five mice per cage. Animals were maintained under a 14:10 h light : dark cycle in a humidity- and temperature-controlled room, with water and laboratory chow supplied ad libitum. All experiments were conducted in accordance with National Institutes of Health guidelines to minimize potential pain or discomfort and under approved protocols from the Institutional Animal Care and Use Committees at Mount Sinai School of Medicine and Duke University.

RNA isolation and real-time PCR

Mice were euthanized by carbon dioxide asphyxiation, and brain regions were dissected. Total RNA was isolated using the RNeasy® Mini kit (Qiagen, Valencia, CA, USA), and reverse transcription was performed using Superscript II® and random hexamer primers (Invitrogen, Carlsbad, CA, USA). The primer sequences for ECE-2 were (sense) 5′-CTTACTACCTTCCAAC-3′ and (antisense) 5′-GTCAGTGACTCATTCT-3′ and (sense) 5′-AATGAAATCGTCTTCC-3′ and (antisense) 5′-GTCAGTGACTCATTC-3′; primers for ECE-1 were (sense) 5′-ATCAGTGGAGTATGAC-3′ and (antisense) 5′-CCTTGATCATCGAAAG-3′; primers for neutral endopeptidase (NEP) were (sense) 5′-CGAAATCAGATAGTCT-3′ and (antisense) 5′-GTCTCCATCTTTATTG-3′; primers for β-actin (internal control) were (sense) 5′-ACCCACACTGTGCCCATCTA-3′ and (antisense) 5′-TCGGAACCGCTCGTTGCCA-3′. The ECE-2, ECE-1 and NEP PCR products were 150–200 bp, and they spanned the Zn+2-binding and catalytic domains. Real-time PCR was performed on a LightCycler II (Roche Applied Science, Indianapolis, IN, USA) using the QuantiTect® SYBR® Green PCR kit (Qiagen) where the initial activation step (95°C for 15 min) was followed by 45 cycles of PCR [95°C for 15 seconds; 50°C or 47°C (for NEP) for 25 seconds and 72°C for 10 seconds]. Detection of fluorimetric SYBR Green (i.e. amount of PCR product) was measured at the end of each elongation phase. The specificity of products was documented at the end of each run with a melting curve analysis. The threshold cycle (i.e. cycle number at which fluorescence corresponds to appearance of the amplified PCR product) was used to calculate the starting template amount in each sample. Expression of each gene was normalized to β-actin. Serial 10-fold dilutions (i.e. 100–0.0001 pg) of the murine ECE-2 plasmid were used to generate the standard curve to calculate the copy number. The relative expression of ECE-2, ECE-1 and NEP in different samples was determined by the 2−ΔΔCT method (Livak & Schmittgen 2001) using the maximum level of expression (i.e. midbrain for ECE-2, liver for ECE-1 and kidney for NEP as calibrators).

Behavioral assessments

Neurophysiological, emotional and sensorimotor gating screen

This screen involved an initial evaluation and examinations of various behaviors as described (Fukui et al. 2007; Pillai-Nair et al. 2005; Pogorelov et al. 2005; Ribar et al. 2000; Rodriguiz & Wetsel 2006). More detailed descriptions of all behavioral methods can be found in Appendix S1.

Morris water maze

This test was conducted as outlined (Rodriguiz & Wetsel 2006; Wolfer et al. 1999) in a 120-cm-diameter stainless steel tub filled with opaque water (25°C). A white metal platform (12 cm in diameter) was located approximately 1 cm below the surface of the water and 20 cm from the wall of the maze. Lighting was indirect (approximately 350 lux), and behavior was monitored with a video camera located 140 cm above the pool. Total swim distance, latency to locate the platform and swim velocity were calculated from tracking profiles created by Ethovision (Noldus Information Technology, Leesburg, VA, USA).

Object recognition

This test was conducted across 9 consecutive days (Mumby et al. 2002; Pogorelov et al. 2005). On days 1–3, the objects were arranged in a triangular configuration (Fig. 1a). On day 4, the location of object 2 was changed and this was maintained until day 7 (Fig. 1b). On days 8 and 9, object 3 was replaced with a novel object (Fig. 1c). Object exploration included all instances of sniffing, touching, climbing, rearing near, sitting on and time spent with the object. Additionally, on days 4–9, all instances and times sniffing or rearing in the old location of object 2 were scored. All tests were videotaped and coded with trained observers using Observer (Noldus Information Technology) who were blind to the genotype of the animals.

image

Figure 1. Schematic of procedure and object recognition memory for WT and KO mice. (a) Days 1–3: acclimatization. (b) Days 4–7: spatial change. (c) Days 8 and 9: novel object recognition. (d and e) Per cent time spent in object contact for WT (d) and KO (e) mice after object 2 had been moved to a new location. An omnibus rmanova found main effects of objects (F3,153 = 37.668, P < 0.001), and the objects by genotype (F3,153 = 41.226, P < 0.001) and the days by objects by genotype (F9,153 = 3.761, P < 0.007) interactions to be significant. Bonferroni-corrected pair-wise comparisons found that on day 4, WT mice increased exploration of object 2 relative to objects 1 and 3 and to the old location of object 2 (Ps < 0.038). This behavior was maintained until day 7 (Ps < 0.050) and the per cent time spent in the previous location of object 2 decreased significantly on days 5–7 relative to that on day 4 (Ps < 0.027). By comparison, per cent time spent with object 2 was not increased for KO mice. Instead, mutants spent more time exploring the old location of object 2 than objects 1–3 (Ps < 0.021). Exploration of object 2 by mutants increased compared with object 1 (Ps < 0.026) on days 5–7 and to object 3 on days 5 and 6 (Ps < 0.051). Unlike WT controls, ECE-2 mice increased exploration of the old location of object 2 on days 4–7 (Ps < 0.001) and spent less time exploring object 2 in its new location on days 4 and 5 (Ps < 0.020). (f and g) Per cent time in object contact for WT (f) and KO (g) mice after object 3 was substituted with a novel object. An omnibus rmanova showed main effects of objects (F3,51 = 103.112, P < 0.001), and the objects by genotype (F3,51 = 25.778, P < 0.001) and days by objects by genotype (F3,51 = 17.778, P < 0.001) interactions to be significant. Bonferroni comparisons showed that on day 8, WT mice increased exploration of the novel object relative to objects 1 and 2 (Ps < 0.001). However, by day 9, the per cent time spent exploring the novel object was not different than that for the two familiar objects. On both days, the per cent time spent in the old location was reduced relative to that for object 2 (Ps < 0.001). By comparison, KO mice failed to show any increase in the per cent time spent exploring the novel object on day 8, but by day 9, there was a marked increase in exploration of the novel object compared with the two familiar objects (Ps < 0.001). Nonetheless, mutants spent more time in the old location of object 2 (Ps < 0.011) on days 8 and 9 than WT controls. n = 9 WT and 10 KO mice. *P < 0.05, WT compared with KO animals; +P < 0.05, object 2 new location vs. its old location; #P < 0.05, novel object vs. objects 1 and 2 in its new location.

Download figure to PowerPoint

Social transmission of food preference

This test has been outlined (Kogan et al. 2000; Rodriguiz & Wetsel 2006) where learning and memory were evaluated in tester mice at 20 min and 24 h. The social transmission of food preference (STFP) test was run under two different protocols. In one, the demonstrator mouse remained in the home cage with the tester mice. In the other, the demonstrator was removed from the home cage following the initial 20-min interaction with the tester mice (e.g. after the 20-min test). All bowls were weighed before and after each test, and any spillage was recovered. Dietary preference was calculated as the difference in consumption between the two diets relative to total consumption.

Statistics

Statistical analyses were performed with the spss-11 statistical programs (SPSS Inc., Chicago, IL, USA), and all data were presented as means and standard errors of the mean. As no sex differences were detected in any behavioral test, data for males and females were combined. For the neurophysiological screen, genotype differences were examined with chi-squared test or t-test. For the STFP and object recognition tests, pre-exposure data were evaluated by univariate analysis of variance (anova). Prepulse-dependent inhibition was examined with repeated measures analysis of variance (rmanova), with prepulse intensity as the within-subjects factor and genotype as the between-subjects factor. Morris water maze data were analyzed by rmanova, with test day as the within-subjects effect and genotype as the between-subjects effect for hidden- and visible-platform testing. Probe test analyses by rmanova required two within-subjects effects: test day and maze quadrant and the between-subjects effect of genotype. For object preference, per cent object contact was examined with rmanova, with test day and object type or location as the within-subjects effects and genotype as the between-subjects effect. Finally, preference scores for STFP were examined with rmanova, with time (20 min and 24 h) as the within-subjects effect and genotype as the between-subjects effect. For anova and rmanova, Bonferroni-corrected pair-wise comparisons were used. In all cases, P < 0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

Distribution of ECE-2 suggests a neuroendocrine role

Expression of ECE-2 messenger RNA (mRNA) was examined in adult WT tissues by quantitative real-time PCR. Consistent with previous reports (Emoto & Yanagisawa 1995; Yanagisawa et al. 2000), ECE-2 exhibited primarily a neuroendocrine distribution, with very little expression in peripheral tissues (Table 1). Midbrain and cerebellum contained the highest levels in brain, followed by pituitary, hypothalamus and frontal cortex. Striatum and hippocampus had low levels of ECE-2 mRNA. High levels were also detected in spinal cord and adrenal medulla, with lower levels in ovaries and heart. By contrast, ECE-1 and NEP mRNAs were expressed at higher levels in peripheral tissues; comparable levels of ECE-1 and ECE-2 were found in the pituitary and adrenal medulla (Fig. 2). Together, the distribution pattern of ECE-2 expression in neuroendocrine tissues suggests a role for this enzyme in processing neuropeptides.

Table 1.  Expression of ECE-2 mRNA in brain and peripheral tissues
Tissue*Copy number/μl
  • *

    Expression of ECE-2 was normalized to that of β-actin and compared with a standard curve of serially diluted plasmids containing known concentrations of the ECE-2 complementary DNA.

  • n   =  3 mice.

Midbrain70 470 ± 6120
Cerebellum61 319 ± 6184
Spinal cord29 713 ± 1870
Hypothalamus27 683 ± 8894
Prefrontal cortex16 207 ± 971
Hippocampus3299 ± 559
Striatum3211 ± 612
Pituitary35 076 ± 18 235
Adrenal medulla17 875 ± 5937
Ovaries4385 ± 434
Testis972 ± 261
Heart1869 ± 806
Lung129 ± 129
Kidney115 ± 71
Muscle115 ± 71
Liver47 ± 27
image

Figure 2. Comparison of expression profiles of ECE-2, ECE-1 and NEP mRNA in various brain regions and peripheral tissues. The expression profiles are calculated as the ratio of enzyme to β-actin mRNA and normalized to the maximum expression in midbrain for ECE-2, in liver for ECE-1 and in kidney for NEP. Ad Med, adrenal medulla; CB, cerebellum; FC, frontal cortex; HIP, hippocampus; HYP, hypothalamus; MB, midbrain, Pit, pituitary; STR, striatum; SC, spinal cord. n = 3 mice.

Download figure to PowerPoint

Disruption of ECE-2 exerts no effects on gross behavior

No genotype differences was observed in assessments of physical appearance and sensory and motor functioning (Table S1). Responses in tail-flick, zero maze, tail suspension, transfer arousal and prepulse inhibition (PPI) were similar for WT and KO mice. Together, these results show that many behaviors are intact in the mutants.

KO mice show poor memory consolidation

Morris water maze

In the visible-platform test, no genotype differences emerged (Fig. 3a, Figs S1a and S2a). With the hidden-platform version of the test, KO animals were impaired (Fig. 3b). Wild type mice reduced their swim distances across test days, whereas mutants showed little change over time. Tracings of swim patterns on test days 1, 3 and 6 confirmed that KO mice swam further than WT controls (Fig. 3e). Moreover, KO animals swam in repetitive looping patterns. Swim time was also aberrant for KO animals (Fig. S1b). Analyses of swim distances for probe trials on days 4 and 6 confirmed that WT animals preferred the quadrant where the hidden platform had been located (Fig. 3c). By contrast, KO mice showed no preference for any quadrant on any probe trial, and by the last test day, mutants swam less far in the northeast (NE) and further in the southwest quadrant than WT controls (Fig. 3c,d). Swim tracings from each probe trial showed that WT mice preferentially searched in the NE quadrant on trials 4 and 6 (Fig. 3f). Knockout animals displayed broader search strategies and continued to circle through other quadrants on these trials. Swim times during probe trials reflected similar deficiencies in the mutants (Fig. S1c,d). Although swim velocities did not differ between genotypes (Fig. S2b–d), the swim distance and swim time results indicate that spatial learning and memory are impaired in KO mice.

image

Figure 3. Swim distances of WT and KO mice in the Morris water maze. (a) Swim distances for WT and KO mice during visible-platform testing. The rmanova showed significant main effects of test trial (F5,90 = 6.392, P < 0.001); the test trial by genotype interaction was not significant. Bonferroni-corrected pair-wise comparisons showed that on trials 3–6, all mice swam significantly shorter distances relative to trial 1 (Ps < 0.046). (b) Swim distances for WT and KO mice during 6 days of acquisition testing with the hidden platform. The rmanova within-subjects tests showed significant effects of test day (F5,100 = 20.397, P < 0.001) and a significant test day by genotype interaction (F5,100 = 6.527, P < 0.001). Bonferroni comparisons showed that although WT and KO animals did not differ on day 1 (P = 0.773), mutants had consistently longer swim distances on test days 2–6 (Ps < 0.048). In addition, WT mice showed reductions in swim distance on each subsequent test day compared with day 1 (Ps < 0.016), whereas no changes over days were observed for KO mice. (c and d) Swim distances during probe trials following acquisition testing on days 2, 4 and 6 for WT (c) and KO (d) mice. The rmanova showed significant within-subjects effects for probe trial (F2,120 = 6.441, P < 0.004) and quadrant (F3,120 = 32.515, P < 0.001) and a significant probe trial by quadrant by genotype interaction (F6,120 = 4.117, P < 0.006). Bonferroni comparisons failed to discern differences between WT and mutants on the first probe trial on day 2. However, WT mice swam further in the NE quadrant on days 4 and 6 relative to the other three quadrants (Ps < 0.031). By comparison, mutants failed to show any quadrant preferences on any of the three probe trials, and by the final probe trial, they swam shorter distances in the NE (P < 0.011) and longer distances in the SW quadrant (P < 0.047) than WT mice. (e) Tracings of representative swim patterns for WT and KO mice on test days 1, 3 and 6 when the hidden platform was located in the NE quadrant. (f) Tracings of representative swim patterns for WT and KO mice during probe trials on days 2, 4 and 6. The apex of the water maze in each tracing is north and the platform location is designated by an open square. The closed square represents where the mouse was located at the end of the trial. n = 10 mice/genotype for visible-platform trials and n = 11 mice/genotype for hidden platform and probe trials. *P < 0.05, WT compared with KO mice; #P < 0.05, NE compared with NW, SE or SW quadrants. NE, northeast; NW, northwest; SE, southeast; SW, southwest.

Download figure to PowerPoint

Object recognition testing

During acclimatization, no genotype differences were observed (Fig. S3a, left), and the per cent times spent contacting each object were similar (Fig. S3b,c). When object 2 was moved to a new location (day 4), animals spent more time contacting all objects (Fig. S3a, middle). Subsequently, the total time contacting objects decreased for WT but remained elevated for KO animals. Wild type mice increased their per cent time with the moved object 2 on days 4–7 (Fig. 1d) compared with the other two objects and time spent in the old location of object 2. Although mutants explored the displaced object, the per cent time spent in its old location was increased (Fig. 1e). By days 6 and 7, KO animals spent more time with object 2 than its old location. Nevertheless, exploration of the old location remained high. When object 3 was replaced with a novel object, WT animals increased exploration time for all objects (Fig. S3a, right); mutants continued to have high rates of object exploration. Hence, on day 8, WT mice enhanced their per cent exploration of the novel object (Fig. 1f), whereas KO mice reduced the time spent exploring the old location of object 2 and spent equal amounts of time with all three objects (Fig. 1g). On day 9, WT mice reduced their per cent time spent with the novel object (Fig. 1f), whereas mutants markedly increased its exploration (Fig. 1g). Collectively, these data indicate that KO mice are impaired in spatial learning and memory and they show perseveration to the ‘old’ location of object 2. This response may delay the ability of mutants to recognize the introduction of the novel object on day 8.

Social transmission of food preference

Further studies of learning and memory were instituted with the STFP test. When the demonstrator was housed continuously with tester animals, WT mice showed a preference for the familiar diet at 20 min and maintained this preference 24 h later (Fig. 4a). Knockout animals failed to develop a preference at 20 min but preferred the familiar diet at 24 h – similar to the WT controls. One reason for the result at 24 h may be the persistence of reminder cues provided by the demonstrator in the home cage. To evaluate this possibility, the demonstrator was removed at the time of the 20-min test and remained in a separate cage until the end of the study. Here, WT mice continued to show strong preferences for the familiar diet at 20 min and 24 h (Fig. 1b). Knockout mice again showed no preference at 20 min, but by 24 h, they displayed a greatly weakened preference for the familiar diet compared with WT controls. Additionally, regardless of demonstrator presence, no genotype differences were discerned in latencies to contact and eat the diets or in frequencies of bowl contacts at 20 min or 24 h in the presence (Fig. S4) or absence (Fig. S5) of the demonstrator. Together, these data suggest that acquisition of memory is slower in KO than in WT mice and that reminder cues (i.e. demonstrator animal) significantly improve memory consolidation.

image

Figure 4. Learning and memory in WT and KO mice in the STFP test. (a) When tester mice have access to demonstrator animals between the 20-min and the 24-h test, WT mice show strong preferences for the demonstrator diet at both the short- and the long-term intervals. Knockout animals fail to exhibit a preference for the demonstrator diet at 20 min but show a preference for this diet at 24 h, similar to the WT controls. (b) When access to demonstrator mice is restricted to 20 min before the short-interval test, WT mice show strong preferences for the familiar demonstrator diet at the 20-min and 24-h tests. Under these conditions, KO animals show reduced preference for the demonstrator diet at both the 20-min and the 24-h tests. An omnibus rmanova showed a main effect for test interval (F1,33 = 18.708, P < 0.001), and significant test interval by genotype (F1,33 = 6.319, P < 0.017) and test interval by genotype by paradigm (i.e. presence or absence of demonstrator) (F1,33 = 8.934, P < 0.005) interactions. Bonferroni-corrected pair-wise comparisons showed that when the demonstrator mouse was housed with the tester mice between tests (panel a), mutants had reduced preference for the familiar diet at 20 min relative to WT controls (P < 0.007) but not at the 24-h test. When the demonstrator mouse was removed immediately before testing (panel b), KO mice also showed reduced preference for the familiar diet at 20 min and 24 h relative to WT littermates (Ps < 0.030). Strikingly, preference for the familiar diet was lower for mutants at 24 h that had been housed without the demonstrator than those housed with the demonstrator (P < 0.036). n = 9 to 10 WT and 10 KO mice/test. *P < 0.05, WT compared with KO mice; #P < 0.05, preference at 20-min test compared with 24-h test; ^P < 0.05, preference at 24 h in presence or absence of demonstrator mice.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

Neuropeptides play important roles in normal brain function. In this study, ECE-2 mRNA is highly expressed in midbrain, cerebellum, hypothalamus and frontal cortex, with lower levels in striatum and hippocampus. The high expression of ECE-2 in cerebellum and midbrain suggests that mutants may be impaired in balance, motor co-ordination and/or motor learning (Houk & Wise 1995). However, KO mice failed to show conspicuous disturbances in these behaviors in the neurophysiological screen, rotorod and open field. Besides these responses, the cerebellum is implicated in encoding long-term memory and recognition memory (Weis et al. 2004). Hence, some of the defects in learning and memory with KO mice may be partially attributed to some undisclosed cerebellar dysfunction.

Besides cerebellum, we found ECE-2 to be present at moderate levels in hippocampus. An in situ hybridization study has shown that ECE-2 mRNA in rat hippocampus is restricted to the granule cell layer of the dentate gyrus (Nakagomi et al. 2000). In mice, ECE-2 mRNA expression in hippocampus is also confined to this cell layer (http://www.brain-map.org). Hence, in our biochemical experiment, the use of the entire hippocampus, rather than the dentate gyrus alone, probably underestimates the quantities of ECE-2 in this brain region.

The critical role of the hippocampus and related temporal lobe areas in learning and memory is well known (Eichenbaum & Cohen 2001). Loss of ECE-2 in KO hippocampus, along with the correlated anomalies in the Morris water maze, suggests that this gene may play a salient role in spatial learning and memory. In our studies, KO mice are impaired on the hidden-platform version of the task, whereas performance with the visible platform is indistinguishable from that of WT controls. These latter findings and those with the neurophysiological screen and PPI experiment indicate that sensorimotor functions are intact in KO animals. Nevertheless, it is noteworthy that rodents with hippocampal lesions (Eichenbaum et al. 1990) or genetic disruption of hippocampally expressed genes (Nakazawa et al. 2004; Pittenger et al. 2002) retain the ability to swim to a cued location but are unable to learn the location of the hidden platform. Additionally, these animals typically engage in longer swim paths with more near misses of the hidden platform than controls. Endothelin-converting enzyme-2 KO mice show a similar proclivity; however, they appear more impaired in the water maze than mere functional disruption of the dentate gyrus suggests (Niewoehner et al. 2007). Because ECE-2 is expressed in various areas of cerebral cortex and cerebellum (http://www.brain-map.org), it may be the case that ablation of Ece-2 in these additional brain regions may contribute to their impairment (D’Hooge & De Deyn 2001).

Besides water maze, mutants are also deficient in object recognition memory. They are slower to recognize shifts in object location and change from a familiar to a novel object. Interestingly, per cent exploration time for all objects is increased when the familiar object is moved or when a familiar object is exchanged for a novel one. This increase in per cent time with objects suggests that both genotypes recognize that the configuration or the appearance of objects has changed; however, because mutants spend increased time at the previous location for object 2, perseveration appears to interfere with their learning and memory. These findings are consistent with restricted expression of ECE-2 to the dentate gyrus as this hippocampal area is known to play a critical role in forming distinct representations of information in the environment, such as locations or comparisons between contexts (McHugh et al. 2007).

A possible contribution to the learning and memory impairment in KO animals may be as a result of perseveration. For instance, rats with fornix transections display similar deficiencies in the water maze and have difficulty shifting their search strategies to new locations when the hidden platform is moved to another quadrant (Whishaw & Tomie 1997). Although we did not change the position of the hidden platform because levels of WT and KO performance were not equivalent at the end of acquisition, responses to the moved object location suggest that the mutants have difficulty suppressing previous responses. Nevertheless, perseveration to the previous location suggests that rudimentary place or spatial learning can occur in KO animals, but their ability to solve new problems within the same test context is impaired. In this regard, extensive training of rodents with hippocampal lesions can be sufficient to overcome deficiencies in spatial learning and memory (Eichenbaum et al. 1990); however, when place cues surrounding the maze are moved or the starting location of the animals is varied, lesioned animals resort to perseveration. Thus, our results suggest that KO mice are capable of very simple learning in novel contexts, but once knowledge of context is established, they have difficulty integrating new information with pre-existing memories.

Mice were also tested in STFP, a test that relies on intact hippocampal functioning (Bunsey & Eichenbaum 1995). In ablation experiments, animals with hippocampal lesions prefer the familiar diet in the 20-min test, whereas this preference is lost at 24 h. This type of deficit has been attributed to impairments in consolidation and recall (Eichenbaum & Cohen 2001). ECE-2 KO mice are deficient in short-term memory, but long-term memory is relatively intact when the demonstrator remains with the tester mouse. Removal of the demonstrator results in greatly weakened long-term memory. These results suggest that KO animals are not deficient in formation of long-term memory per se, but processes underlying short-term and/or working memory may be perturbed, leading to delay in memory formation. When reminder cues (i.e. demonstrator animal) are provided, mutants show intact long-term memory, suggesting that consolidation is protracted. Interestingly, somewhat similar deficits are found in geriatric amnesiac syndromes in humans (Woods et al. 1982) and in rats with surgical lesions of the dentate gyrus (Geinsman et al. 1986). In these cases, there is a selective failure to retain new information while retrieval of older established memories is preserved. A loss of hippocampal integrity and neural plasticity may be partially due to slowed learning and delay of recall in these patients – although loss of plasticity in prefrontal cortex may contribute to cognitive decline (Laroche et al. 2000). Nevertheless, temporal aspects of memory consolidation may be detrimentally affected in geriatric dementia patients (Pomara et al. 2004). These findings are interesting in that ECE-1 and ECE-2 have been implicated in limiting the accumulation of β-amyloid peptides in brain, and levels of β-amyloid are increased in the KO brain (Eckman et al. 2003). Amyloid deposition in mice is known to impair memory consolidation and selectively reduce expression of early immediate genes implicated in synaptic plasticity (Dickey et al. 2004). A recent report has suggested that Alzheimer’s disease and the genes for β-amyloid catabolism and ECE-2 may be linked (Saido & Iwata 2006). Hence, elucidation of basic mechanisms underlying the deficits in learning and memory of ECE-2 KO mice may provide new insights into processes that contribute to Alzheimer’s disease and possible other disorders of cognitive function.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information
  • Bunsey, M. & Eichenbaum, H. (1995) Selective damage to the hippocampal region blocks long-term retention of a natural and nonspatial stimulus-stimulus association. Hippocampus 5, 546556.
  • Che, F.Y., Yan, L., Li, H., Mzhavia, N., Devi, L.A. & Fricker, L.D. (2001) Identification of peptides from brain and pituitary of Cpefat/Cpefat mice. Proc Natl Acad Sci U S A 98, 99719976.
  • Davenport, A.P. & Kuc, R.E. (2000) Cellular expression of isoforms of endothelin-converting enzyme-1 (ECE-1c, ECE-1b and ECE-1a) and endothelin-converting enzyme-2. J Cardiovasc Pharmacol 36, S12S14.
  • D’Hooge, R. & De Deyn, P.P. (2001) Applications of the Morris water maze in the study of learning and memory. Brain Res Rev 36, 6090.
  • Dickey, C.A., Gordon, M.N., Mason, J.E., Wilson, N.J., Diamond, D.M., Guzowski, J.F. & Morgan, D. (2004) Amyloid suppresses induction of genes critical for memory consolidation in APP + PS1 transgenic mice. J Neurochem 88, 434442.
  • Docherty, K. & Steiner, D.F. (1982) Post-translational proteolysis in polypeptide hormone biosynthesis. Annu Rev Physiol 44, 625638.
  • Eckman, E.A. & Eckman, C.B. (2005) A β-degrading enzymes: modulators of Alzheimer’s disease pathogenesis and targets for therapeutic intervention. Biochem Soc Trans 33, 11011105.
  • Eckman, E.A., Watson, M., Marlow, L., Sambamurti, K. & Eckman, C.B. (2003) Alzheimer’s disease β-amyloid peptide is increased in mice deficient in endothelin-converting enzyme. J Biol Chem 278, 20812084.
  • Eichenbaum, H. & Cohen, N.J. (2001) From Conditioning to Conscious Reflection: Memory Systems of the Brain. Oxford University Press, New York.
  • Eichenbaum, H., Stewart, C. & Morris, R.G.M. (1990) Hippocampal representation in spatial learning. J Neurosci 10, 35313542.
  • Emoto, N. & Yanagisawa, M. (1995) Endothelin-converting enzyme-2 is a membrane-bound, phosphoramidon-sensitive metalloprotease with acidic pH optimum. J Biol Chem 270, 1526215268.
  • Fukui, M., Rodriguiz, R.M., Zhou, J., Jiang, S.X., Phillips, L.E., Caron, M.G. & Wetsel, W.C. (2007) Vmat2 heterozygous mutant mice display a depressive-like phenotype. J Neurosci 27, 1052010529.
  • Geinsman, Y., De Toledo-Morrell, L. & Morrell, F. (1986) Loss of perforated synapses in the dentate gyrus: morphological substrate of memory deficit in aged rats. Proc Natl Acad Sci U S A 83, 30273031.
  • Houk, J.C. & Wise, S.P. (1995) Distributed modular architectures linking basal ganglia, cerebellum, and cerebral cortex: their role in planning and controlling action. Cereb Cortex 5, 95110.
  • Kogan, J.H., Frankland, P.W. & Silva, A.J. (2000) Long-term memory underlying hippocampus-dependent social recognition in mice. Hippocampus 10, 4756.
  • Laroche, S., Davis, S. & Jay, T.M. (2000) Plasticity at hippocampal prefrontal cortex synapses: dual roles in working memory and consolidation. Hippocampus 10, 438446.
  • Liu, Q., Guan, X.M., Martin, W.J., McDonald, T.P., Clements, M.K., Jiang, Q., Zeng, Z., Jacobson, M., Williams, D.L. Jr, Yu, H., Bomford, D., Figueroa, D., Mallee, J., Wang, R., Evans, J., Gould, R. & Austin, C.P. (2001) Identification and characterization of novel mammalian neuropeptide FF-like peptides that attenuate morphine-induced antinociception. J Biol Chem 276, 3696136969.
  • Livak, K.J. & Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods 25, 402408.
  • McHugh, T.J., Jones, M.W., Quinn, J.J., Balthasar, N., Coppari, R., Elmquist, J.K., Lowell, B.B., Fanselow, M.S., Wilson, M.A. & Tonegawa, S. (2007) Dentate gyrus NMDA receptors mediate rapid separation in the hippocampal network. Science 317, 9499.
  • Mizuno, K., Minamino, N., Kangawa, K. & Matsuo, H. (1980) A new endogenous opioid peptide from bovine adrenal medulla: isolation and amino acid sequence of a dodecapeptide (BAM-12P). Biochem Biophys Res Commun 95, 14821488.
  • Molineaux, C. & Wilk, S. (1991) Extracellular Processing of Peptides. CRC Press, Boca Raton, FL.
  • Moore, H.P., Andresen, J.M., Eaton, B.A., Grabe, M., Haugwitz, M., Wu, M.M. & Machen, T.E. (2002) Biosynthesis and secretion of pituitary hormones: dynamics and regulation. Arch Physiol Biochem 110, 1625.
  • Mumby, D.G., Gaskin, S., Glenn, M.J., Schramek, T.E. & Lehmann, H. (2002) Hippocampal damage and exploratory preferences in rats: memory for objects, places, and contexts. Learn Mem 9, 4957.
  • Mzhavia, N., Pan, H., Che, F.Y., Fricker, L.D. & Devi, L.A. (2003) Characterization of endothelin-converting enzyme-2. Implication for a role in the nonclassical processing of regulatory peptides. J Biol Chem 278, 1470414711.
  • Nakagomi, S., Kiryu-Seo, S. & Kiyama, H. (2000) Endothelin-converting enzymes and endothelin receptor B messenger RNAs are expressed in different neural cell species and these messenger RNAs are coordinately induced in neurons and astrocytes respectively following nerve injury. Neuroscience 101, 441449.
  • Nakazawa, K., McHugh, T.J., Wilson, M.A. & Tonegawa, S. (2004) NMDA receptors, place cells and hippocampal spatial memory. Nat Rev Neurosci 5, 361372.
  • Niewoehner, B., Single, F.N., Hvalby, Ø., Jensen, V., Zum Alten Borgloh, S.M., Seeburg, P.H., Rawlins, J.N.P., Sprengel, R. & Bannerman, D.M. (2007) Impaired spatial working memory but spared reference memory following functional loss of NMDA receptors in the dentate gyrus. Eur J Neurosci 25, 837846.
  • Perry, S.J., Yi-Kung Huang, E., Cronk, D., Bagust, J., Sharma, R., Walker, R.J., Wilson, S. & Burke, J.F. (1997) A human gene encoding morphine modulating peptides related to NPFF and FMRFamide. FEBS Lett 409, 426430.
  • Pillai-Nair, N., Panicker, A.K., Rodriguiz, R.M., Gilmore, K.L., Demyanenko, G.P., Huang, J.Z., Wetsel, W.C. & Maness, P.F. (2005) Neural cell adhesion molecule-secreting transgenic mice display abnormalities in GABAergic interneurons and alterations in behavior. J Neurosci 25, 46594671.
  • Pittenger, C., Huang, Y.Y., Paletzki, R.F., Boutchouladze, R., Scanlin, H., Vronskaya, S. & Kandel, E.R. (2002) Reversible inhibition of CREB/ATF transcription factors in region CA1 of the dorsal hippocampus disrupts hippocampus-dependent spatial memory. Neuron 34, 447462.
  • Pogorelov, V.M., Rodriguiz, R.M., Insco, M.L., Caron, M.G. & Wetsel, W.C. (2005) Novelty seeking and stereotypic activation of behavior in mice with disruption of the Dat1 gene. Neuropsychopharmacology 30, 18181831.
  • Pomara, N., Willoughby, L.M., Wesnes, K. & Sidtis, J.J. (2004) Increased anticholinergic challenge-induced memory impairment associated with APOE-ɛ4 allele in the elderly: a controlled pilot study. Neuropsychopharmacology 29, 403409.
  • Ribar, T., Rodriguiz, R.M., Khiroug, L., Wetsel, W.C., Augustine, G.J. & Means, A.R. (2000) Cerebellar defects in Ca2+/calmodulin kinase IV deficient mice. J Neurosci 20, RC107, 1 – 5.
  • Rodriguiz, R.M. & Wetsel, W.C. (2006) Assessments of cognitive deficits in mutant mice. In Levin, E. & Buccafusco, J.J. (eds), Animal Models of Cognitive Impairment. CRC Press, Boca Raton, FL, pp. 223282.
  • Russell, F.D. & Davenport, A.P. (1999) Evidence for intracellular endothelin-converting enzyme-2 expression in cultured human vascular endothelial cells. Circ Res 84, 891896.
  • Saido, T.C. & Iwata, N. (2006) Metabolism of amyloid β peptide in pathogenesis of Alzheimer’s disease: towards presymptomatic diagnosis, prevention, and therapy. Neurosci Res 54, 235253.
  • Sigafoos, J., Chestnut, W.G., Merrill, B.M., Taylor, L.C., Diliberto, E.J. Jr & Viveros, O.H. (1993) Novel peptides from adrenomedullary chromaffin vesicles. J Anat 183, 253264.
  • Strand, F.L. (1999) Neuropeptides: Regulators of Physiological Processes. MIT Press, Cambridge, MA.
  • Vilim, F.S., Aarnisalo, A.A., Nieminen, M.L., Lintunen, M., Karlstedt, K., Kontinen, V.K., Kalso, E., States, B., Panula, P. & Ziff, E. (1999) Gene for pain modulatory neuropeptide NPFF: induction in spinal cord by noxious stimuli. Mol Pharmacol 55, 804811.
  • Weeraratna, A.T., Kalehua, A., Deleon, I., Bertak, D., Maher, G., Wade, M.S., Lustig, A., Becker, K.G., Wood, W. III, Walker, D.G., Beach, T.G. & Taub, D.D. (2007) Alterations in immunological and neurological gene expression patterns in Alzheimer’s disease tissues. Exp Cell Res 313, 450461.
  • Weis, S., Klaver, P., Reul, J., Elger, C.E. & Fernandez, G. (2004) Temporal and cerebellar brain regions that support both declarative memory, formation, and retrieval. Cereb Cortex 14, 256267.
  • Whishaw, I.Q. & Tomie, J.-A. (1997) Perseveration on place reversals in spatial swimming pool tasks: further evidence for place learning in hippocampal rats. Hippocampus 7, 361370.
  • Wolfer, D.P., Stagljar-Bozicevic, M., Errington, M.L. & Lipp, H.-P. (1999) Spatial memory and learning in transgenic mice: fact or artifact? News Physiol Sci 13, 118123.
  • Woods, B.T., Schoene, W. & Kneisley, L. (1982) Are hippocampal lesions sufficient to cause lasting amnesia? J Neurol Neurosurg Psychiatry 45, 243247.
  • Yanagisawa, H., Hammer, R.E., Richardson, J.A., Emoto, N., Williams, S.C., Takeda, S., Clouthier, D.E. & Yanagisawa, M. (2000) Disruption of ECE-1 and ECE-2 reveals a role for endothelin-converting enzyme-2 in murine cardiac development. J Clin Invest 105, 13731382.
  • Yang, H.Y., Fratta, W., Majane, E.A. & Costa, E. (1985) Isolation, sequencing, synthesis, and pharmacological characterization of two brain neuropeptides that modulate the action of morphine. Proc Natl Acad Sci U S A 82, 77577761.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

We thank Rosalie Bateson, Sandeep Bhave, Liping Du, John Gilbert, Lien Nguyen, Lindsey Phillips, John Wilkins and Jiechun Zhou for their assistance in the behavioral tests at Duke. This work was funded in part through unrestricted funds to Dr W.C.W. and by National Institutes of Health grants (NS026880 and DA019521) to Dr L.A.D. This work was supported in part by the Mount Sinai Phenotyping Shared Resource Facility.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments
  8. Supporting Information

The following supplementary material is available for this article online from http://www.blackwell-synergy.com/doi/full/10.1111/j.1601-183X.2007.00365.x

Appendix S1: Behavioral assessments.

Table S1: Results from general behavioral assessments of WT and KO mice

Figure S1: Swim times of WT and KO mice in the Morris water maze.

Figure S2: Swim velocities of WT and KO mice in the Morris water maze.

Figure S3: Per cent time spent with objects for WT and KO mice.

Figure S4: Interactions with the familiar- and novel-flavored diets during the 20-min and 24-h tests by ECE-2 mice, with the demonstrator remaining in the home cage.

Figure S5: Interactions with the familiar- and novel-flavored diets during the 20-min and 24-h tests by ECE-2 mice, with the demonstrator removed from the home cage.

Please note: Blackwell Publishing is not responsible for the content or functionality of any supplementary materials supplied by the authors.

FilenameFormatSizeDescription
GBB_365_sm_FigureS1.tif2008KSupporting info item
GBB_365_sm_FigureS2.tif2012KSupporting info item
GBB_365_sm_FigureS3.tif2679KSupporting info item
GBB_365_sm_FigureS4.tif3027KSupporting info item
GBB_365_sm_FigureS5.tif3012KSupporting info item
GBB_365_sm_AppendixS1.doc81KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.