Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Accumulating evidence indicates that the genetic background of mice contributes to their behavioral profile and may interact with the effects of drugs and genetic manipulations. For example, several reports describe strain differences in measures of prepulse inhibition (PPI) and startle responding in mice (Dulawa and Geyer, 1996; Bullock et al., 1997; Logue et al., 1997; Paylor and Crawley, 1997). PPI is a cross-modal phenomenon wherein the startle response is reduced when the startling stimulus is preceded by a low intensity prepulse (Graham, 1975; Hoffman and Ison, 1980). Although it has been clearly established that dopamine (DA) agonists such as amphetamine and apomorphine reduce PPI in rats (Mansbach et al., 1988; Peng et al., 1990; Swerdlow et al., 1991), relatively few reports have characterized the effects of DA agonists on PPI in mice (Dulawa and Geyer, 1996; Curzon and Decker, 1998; Ralph et al., 1999). Given the marked phenotypic differences in both PPI and startle responding between mouse strains, characterizing more than one strain of mice may reveal differential effects or sensitivities to DA agonists such as amphetamine.

In an effort to describe locomotor activity, many studies report on the overall level of arousal as quantified by some measure of the amount of locomotor activity, e.g., number of beam breaks or distance traveled. In addition, some investigators report the number of times an animal enters a specific area of interest or significance, e.g., the center of the enclosure, which rodents usually avoid. These methods, however, do not adequately quantify the spatio-temporal patterns of locomotor activity and thus fail to assess the structure or sequential organization of mouse motor behavior. The spatial scaling exponent, d, provides a quantitative measure of the pattern of motor behavior that is independent of the amount of locomotor activity. As in rats (Paulus and Geyer, 1991a, 1993; Paulus et al., 1998), measures of the patterns of locomotor behavior exhibited by different strains of mice cannot be predicted by the strain differences in the amount of locomotor activity (Paulus and Geyer, 1991a). These findings indicate that measures of both the amount and pattern of locomotor activity provide independent information about the behavior of mice in a novel environment (Paulus et al., 1999). Activation of the DA system with drugs such as amphetamine is known to affect the amount of locomotor activity in mice, but the effects of amphetamine on patterns of locomotor activity in mice are not known.

C57BL/6J and 129 substrains are commonly used to create mutant lines of mice. These background strains of mice differ in measures of PPI and locomotor activity (Dulawa and Geyer, 1996; Bullock et al., 1997; Logue et al., 1997; Paylor and Crawley, 1997; Paulus et al., 1999). Without characterizing the background lines, it is difficult to conclude that observed phenotypes in knockout mice are due to a specific mutation rather than to genetic contributions from one of its parental lines (e.g., Kelly et al., 1998). We have reported on both D2 (created from 129SvEv and C57BL/6J strains) and D3 (created from 129SvJ and C57BL/6J strains) receptor (/) mice, where the D3 (/) mice had disrupted PPI after amphetamine treatment, but the D2 (/) mice were insensitive to the effects of amphetamine (Ralph et al., 1999). It is unclear, however, how amphetamine would affect PPI or the patterns of motor behavior in the C57BL/6J, 129SvEv, or 129SvJ mice.

Here we characterized the C57BL/6J, 129S6 (formerly 129SvEv), and 129X1 (formerly 129SvJ) strains in a cross-modal PPI session and an unconditioned locomotor activity paradigm. Although different sensory modalities have been used to elicit startle responses in mice, only acoustic prepulses have been used to assess PPI in mice (Dulawa and Geyer, 1996; Bullock et al., 1997; Logue et al., 1997; Paylor and Crawley, 1997; Curzon and Decker, 1998; Ralph et al., 1999). Early-onset hearing loss in mice can affect both acoustic startle responding and PPI elicited by acoustic prepulses (Parham and Willott, 1988; Willott et al., 1994; Zheng et al., 1999). Therefore, we included a nonacoustic prepulse-startle stimulus pairing (i.e., light prepulse with an airpuff startle stimulus) to further characterize PPI in mice. In light of the demonstrated role of DA in the modulation of both PPI and locomotor behavior and the known propensity for differences in response to pharmacological manipulations between strains of mice (Heyser et al., 1997; Miner, 1997; Schlussman et al., 1998; Homanics et al., 1999), we also conducted dose-response studies using the indirect DA agonist amphetamine.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Forty male C57BL/6J and 32 129X1 mice (formerly known as the 129SvJ) (Jackson Labs, Bar Harbor, ME), and 36 129S6 mice (formerly known as the 129SvEv) (Taconic Labs, Germantown, NY) were maintained in an Association for Assessment and Accreditation of Laboratory Animal Care-approved animal facility at the University of California, San Diego. This facility meets all Federal and State requirements for animal care. Mice from each strain were group housed in a climate-controlled animal colony with a reversed day/night cycle (lights on at 7:00 PM, off at 7:00 AM). All behavioral testing started at approximately 8 to 9 weeks of age and occurred between 9:00 AM and 5:00 PM. Food (Harlan Teklab, Madison, WI) and water were available throughout the experiments, except during behavioral testing.

Drugs. d-Amphetamine sulfate was obtained from Sigma (St. Louis, MO) and was dissolved in 0.9% saline. Free-base drug weights were used in all drug calculations. Injections of 1.0, 3.0, or 10.0 mg/kg amphetamine or saline were given i.p. immediately before behavioral testing at a volume of 5 ml/kg of body weight.

Apparatus. Startle reactivity was measured using four startle chambers (SR-LAB, San Diego Instruments, San Diego, CA). Each chamber consisted of a clear nonrestrictive Plexiglas cylinder resting on a platform inside a ventilated box. A high-frequency loudspeaker inside the chamber produced both a continuous background noise of 65 db and the various acoustic stimuli. Vibrations of the Plexiglas cylinder caused by the whole-body startle response of the animal were transduced into analog signals by a piezoelectric unit attached to the platform. These signals were then digitized and stored by a computer. Sixty-five readings were taken at 1-ms intervals, starting at stimulus onset, and the average amplitude was used to determine the acoustic startle response. Sound levels in db(A) SPL were measured as described previously (Dulawa et al., 1997). The SR-LAB calibration unit was used routinely to ensure consistent stabilimeter sensitivity between test chambers and over time (Geyer and Swerdlow, 1998).

Prepulse Inhibition Session. All PPI test sessions consisted of startle trials (PULSE-ALONE, PUFF), prepulse trials (PREPULSE + PULSE, LIGHT + PUFF), and no-stimulus trials (NOSTIM). The PULSE-ALONE trial consisted of a 40-ms 120-db pulse of broadband noise. Acoustic PPI was measured by PREPULSE + PULSE trials that consisted of a 20-ms noise prepulse, 100-ms delay, then a 40-ms 120-db startle pulse (120-ms onset to onset interval). The acoustic prepulse intensities were 4, 8, and 16 db above the 65-db background noise (i.e., 69, 73, and 81 db). The NOSTIM trial consisted of background noise only. The acoustic section of the test session began and ended with five presentations of the PULSE-ALONE trial; in between, each acoustic or NOSTIM trial type was presented 10 times in a pseudorandom order. Light PPI and airpuff startle reactivity were assessed during the same test session, immediately after the acoustic trial types were completed. PUFF startle trials consisted of a 40-ms 20-psi airpuff and light PPI trials consisted of a 20-ms light prepulse, 100-ms delay, then a 40-ms 20-psi airpuff. As with the acoustic PPI and startle trials, there were five presentations of the PUFF trial, followed by 10 presentations each of PUFF and LIGHT + PUFF trial types in pseudorandom order, and concluding with five presentations of the PUFF trial. There was an average of 15 s (range: 12-30 s) between trials. After the mice were placed in the startle chambers, a 65-db background noise level was presented for a 10-min acclimation period and continued throughout the test session. Each animal was always tested in the same startle chamber.

Locomotor Pattern Testing. After the PPI session, mice were placed immediately into the VT enclosure, approximately 30 min after injection. Each mouse was placed in the bottom left corner of each enclosure at the start of the test session. The movements of the mice were tracked for 30 min, with data being stored in three 10-min blocks. Two categories of measures were obtained. First, the amount of locomotor activity was assessed as the number of entries made in different areas of the open field, i.e., the wall area, corner area, or center area (see Geyer et al., 1986). Second, the geometric patterns of locomotor activity were quantified by the spatial scaling exponent, d, as described in detail elsewhere (Paulus et al., 1999). Briefly, the spatial scaling exponent, d, measures the degree to which consecutive movements are along a straight line (d  1), are characterized by meandering patterns (d  1.5), or include many directional changes (d  2).

Statistical Analyses. In PPI and locomotor activity experiments, strain and drug treatment were between-subjects variables. For brevity, PPI data were collapsed across prepulse intensity, and locomotor activity measures were collapsed across block of time in the amphetamine dose response studies. Analyses of variance (ANOVAs) were used to compare means, and Tukey's tests were used for post hoc analysis on between-subjects factors.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Prepulse Inhibition and Startle Reactivity. All three strains of mice were characterized in a baseline startle session. As previously reported (Dulawa and Geyer, 1996; Bullock et al., 1997; Logue et al., 1997; Paylor and Crawley, 1997), strain differences were found in the amount of acoustic PPI (F_2,113 = 41.7, P < 0.001). The C57BL/6J mice had lower acoustic PPI (P < 0.01) than either 129 substrain, while the 129S6 mice had lower acoustic PPI than the 129X1 mice (Fig. 1A). In contrast to the acoustic PPI findings, the C57BL/6J strain had higher levels of light PPI than either the 129S6 or 129X1 mice (P < 0.01) (Fig. 1A). There were also differences between the strains of mice in acoustic (F_2,113 = 64.6, P < 0.001) and airpuff (F_2,113 = 24.4, P < 0.001) startle responding. Both acoustic and airpuff startle reactivity were higher in C57BL/6J than in 129S6 or 129X1 mice (P < 0.01). In addition, acoustic, but not airpuff, startle reactivity was higher in 129S6 than in 129X1 mice (P < 0.01) (Fig. 1B).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Baseline characterization of PPI (A) and startle reactivity (B) in C57BL/6J, 129S6, and 129X1 mice. Acoustic prepulse intensities are above the 65-db background noise. Startle reactivity was measured after presentation of either an acoustic or airpuff startle stimulus. Measures of PPI represent mean percentage of PPI ± S.E.M., and startle values (arbitrary units) represent mean startle magnitude ± S.E.M. **P < 0.01 compared with C57BL/6J mice, P < 0.01 compared with 129S6 mice. n =8 to 10 mice per strain.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Acoustic (A-C) and light (D-F) PPI after treatment with amphetamine (doses: 1.0, 3.0, and 10.0 mg/kg or saline, i.p., at a volume of 5 ml/kg of body weight) in C57BL/6J, 129S6, and 129X1 mice. Values represent mean percentage of PPI ± S.E.M. *P < 0.05, **P < 0.01 compared with saline control. n = 8 to 10 mice per strain.

Locomotor Activity and Patterns. Figure 3 shows the differential patterns of motor activity for each strain of mice in the amphetamine dose-response study. Previous reports have demonstrated differences in the amount of activity and patterns of motor behavior between the C57BL/6J, 129S6, and 129X1 mice (Paulus et al., 1999), prompting the a priori hypothesis of strain differences in the effects of amphetamine. Therefore, we analyzed the vehicle-treated mice to identify strain differences in baseline behavior and analyzed each strain separately to determine the strain-specific effects of amphetamine.

View larger version (60K): [in this window] [in a new window]   Fig. 3.   Patterns of motor behavior in individual C57BL/6J, 129S6, and 129X1 mice after treatment with amphetamine (AMPH) (doses: 1.0, 3.0, and 10.0 mg/kg or saline, i.p., at a volume of 5 ml/kg of body weight). Patterns were reconstructed using the (x, y, t) coordinates of one mouse per treatment group for the first 10 min of motor behavior sampled. Sample patterns were selected based on data closest to the means for both the total amount of activity and average spatial d for each strain and drug dose tested.

Strain Differences after Vehicle Treatment. As shown in Fig. 3, the vehicle-treated C57BL/6J and 129X1 mice exhibited a mixture of straight, meandering, and circumscribed movements, while the 129S6 mice were less active and predominantly moved along the perimeter of the enclosure (Fig. 3). These differences were confirmed quantitatively using total entries and the spatial scaling exponent, d, a measure that quantifies the degree to which sequences of movements are straight (d  1) or circumscribed (d  2) (Paulus and Geyer, 1991b). Significant strain differences were found in total entries (F_2,23 = 27.7, P < 0.01), where the C57BL/6J mice were more active than either the 129S6 or 129X1 mice (P < 0.01) (Fig. 4A). The three strains of mice also differed in the number of center entries (F_2,23 = 21.5, P < 0.01), with the C57BL/6J mice moving into the center more often than either 129 substrain (P < 0.01) (Fig. 4B). Patterns of motor behavior also differed between strains (F_2,23 = 20.3, P < 0.01), as the 129S6 mice made more circumscribed movements (higher spatial d values) than either the 129X1 or C57BL/6J mice (P < 0.01) (Fig. 4C).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Measures of total entries (A), center entries (B), and spatial d (C) in vehicle-treated C57BL/6J, 129S6, and 129X1 mice. Values represent mean ± S.E.M. n = 8 to 10 mice per group.

Effects of Amphetamine on Locomotor Activity. In the C57BL/6J mice, amphetamine significantly altered both the total amount of activity (F_3,35 = 13.8, P < 0.001) and the number of center entries (F_3,35 = 10.5, P < 0.001) (Fig. 5). These effects were attributable primarily to the 10-mg/kg amphetamine dose decreasing both total activity and entries into the center of the open field (P < 0.01). Amphetamine also significantly affected patterns of motor activity as measured by the spatial scaling exponent d (F_3,36 = 26.2, P < 0.001). In contrast to lower doses that did not affect or slightly decreased d, 10 mg/kg amphetamine significantly increased d. Thus, whereas lower doses of amphetamine did not change the patterns of locomotor activity, the high dose resulted in more circumscribed movement patterns, a finding consistent with the emergence of focal stereotypies. In contrast to the C57BL/6J mice, amphetamine treatment did not significantly change locomotor activity, center entries, or spatial d in the 129S6 mice (Fig. 5). In the 129X1 strain of mice, amphetamine altered both the amount of activity (F_3,31 = 3.6, P < 0.05) and patterns of motor behavior (F_3,31 = 4.3, P < 0.01) (Fig. 5). These effects were due mainly to both the 1- and 3-mg/kg doses of amphetamine increasing the amount of activity and 10 mg/kg amphetamine producing more local bouts of locomotor behavior.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Measures of total entries (A), center entries (B), and spatial d (C) after treatment with amphetamine (doses: 1.0, 3.0, and 10.0 mg/kg or saline, i.p., at a volume of 5 ml/kg of body weight) in C57BL/6J, 129S6, and 129X1 mice. Values represent mean ± S.E.M. n = 8 to 10 mice per group.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study yielded three main findings. First, both C57 and 129 strains showed intra- and cross-modal PPI. Second, the degree to which amphetamine disrupted PPI differed across C57 and 129 strains. Third, the effects of amphetamine on locomotor activity, entries into the center area of the open field, and patterns of locomotor behavior differed significantly across strains. In addition to confirming that these mouse strains display intramodal acoustic PPI (Bullock et al., 1997; Logue et al., 1997; Paylor and Crawley, 1997), we also found that they display light PPI in a prepulse-pulse combination without an explicit acoustic component. The DA agonist amphetamine disrupted acoustic PPI in all three strains tested, and light PPI in the C57BL/6J and 129S6 mice. Amphetamine also affected the amount and pattern of motor activity in C57BL/6J and 129X1 mice, but had no significant effect on unconditioned motor activity in the 129S6 mice. Given that amphetamine releases presynaptic DA, these findings are consistent with the role of DA in the modulation of PPI and motor patterns in mice, and they highlight the differential sensitivity of these strains to the effects of amphetamine. They also emphasize that the "baseline" phenotype, when highly strain-dependent, may be as important for behavioral measures as the "generic" effect of the pharmacological agent.

Although many studies show that DA agonists disrupt PPI in rats, few reports have confirmed similar effects in mice (see Introduction). Here, amphetamine significantly disrupted acoustic PPI in each of the mouse strains tested, lending further support to the hypothesis that activation of DA systems disrupts PPI in rodents. In previous studies, amphetamine failed to disrupt PPI in D2 receptor knockout mice derived from a combination of the C57BL/6J and 129S6 strains (Ralph et al., 1999), supporting the hypothesis that D2 receptors are critical for the modulation of PPI. Some investigators, however, have shown that background strains may contribute significantly to observed phenotypes and can complicate interpretations (Kelly et al., 1998). The present finding that amphetamine disrupted PPI in both background strains involved in the D2 knockout mice supports the hypothesis that the absence of an amphetamine effect on PPI in these mice is attributable to the targeted mutation rather than to a background strain.

PPI is a multi-modal phenomenon, where the prepulse and startle stimuli can be presented in either the same or different sensory modalities. In previous reports of mouse PPI, acoustic prepulses have been combined with either acoustic or airpuff startle stimuli (as reviewed above), but no previous report in mice has described PPI in the absence of an explicit acoustic component. The present cross-modal PPI test used a light prepulse preceding a tactile airpuff startle stimulus. Using this cross-modal test, all three strains displayed light PPI. Although some strain differences were found in the initial characterization of light PPI, these strain differences were not seen in the subsequent amphetamine studies in vehicle-treated mice and thus may not be reliable. Amphetamine significantly disrupted light PPI in both the C57BL/6J and 129S6 mice, but there was a trend toward an increase in light PPI in the 129X1 mice, suggesting there are differential sensitivities to amphetamine between the strains. In the current studies, we used the same prepulse-to-startle stimulus interval (100 ms) in all conditions; more robust levels of light PPI might be obtained in each of the strains by optimizing the parameters used to generate cross-modal PPI. By expanding this nonacoustic PPI paradigm, future studies may avoid the experimental complication of hearing loss in mice, a phenomenon that has been reported in numerous strains (Zheng et al., 1999). For example, C57BL/6J mice, a common background strain in many mutant mouse lines, develop an age-related hearing loss that affects startle responding (Parham and Willott, 1988). By using a light prepulse paired with tactile startle stimuli, PPI may be measured even if hearing loss has already begun, thereby increasing the window of testing opportunities.

Previous studies indicate that amphetamine reduces acoustic startle reactivity in mice (Dulawa and Geyer, 1996; Ralph et al., 1999). Here we report similar findings for acoustic startle reactivity and extend it to include decreases in airpuff startle reactivity, although there were differential effects between strains of mice. In the C57BL/6J and 129S6 strains, both 3 and 10 mg/kg amphetamine significantly lowered acoustic startle reactivity, but the 129X1 strain was unaffected by any dose of amphetamine. This lack of effect in the 129X1 mice may have been due to a "floor effect", as the low level of acoustic startle reactivity was maintained regardless of the dose of amphetamine tested. Amphetamine also lowered airpuff startle reactivity, but only in the 129 substrains, where both the 3- and 10-mg/kg doses of amphetamine significantly lowered airpuff startle reactivity. Thus, there were different sensitivities to the effects of amphetamine depending on the modality of the startle stimuli and the strain of mouse. Similarly, rat strains exhibit differences in the effects of another DA agonist, apomorphine (Mansbach et al., 1988; Davis et al., 1990; Rigdon, 1990; Varty and Higgins, 1994; Swerdlow et al., 2000). These data provide further evidence for a dissociation between the mechanisms that control PPI and startle reactivity in both mice and rats treated with amphetamine (Swerdlow et al., 1990; Dulawa and Geyer, 1996; Curzon and Decker, 1998; Ralph et al., 1999).

We have previously described differences in both the amount and patterns of motor behavior between several strains of mice (Paulus et al., 1999). Consistent with this report, we found similar strain differences in motor phenotypes in our vehicle-treated C57BL/6J, 129S6, and 129X1 mice. C57BL/6J mice continued to explore the enclosure during the entire test session. The 129X1 mice initially explored the enclosure, briefly moving into the center, while the 129S6 mice sampled only a small area of the test environment, rarely entering the center of the test area. Interestingly, the vehicle-treated C57BL/6J mice appeared to have slower rates of habituation in measures of total entries, center entries, and spatial d than the other two strains. Both the 129S6 and 129X1 mice showed marked decreases in activity and increases in spatial d over the test session, while the C57BL/6J mice showed only slight, although significant, changes in these values. As with the PPI and startle findings, the significant differences in both locomotor activity and patterns across strains observed here support the general idea that some, but not all, dimensions of the behavioral phenotype are strain-dependent.

When the mice were challenged with amphetamine, there were strain-dependent responses with regard to both amount of activity and patterns of motor behavior. In the C57BL/6J mice, 3 mg/kg amphetamine produced both increases in activity and straighter motor patterns, wherein the mice moved predominantly around the perimeter of the test area. The 129X1 mice were more sensitive to the effects of amphetamine, as the low dose increased activity, even into the center area, and produced straighter sequences of movements, while the middle dose increased activity around the edge of the enclosure and decreased center entries. In both the C57BL/6J and 129X1 mice, 10 mg/kg amphetamine significantly decreased activity, including into the center, and the mice displayed bouts of both straight and circumscribed motor patterns. Thus, amphetamine, which exerts its behavioral effects by blocking the DA transporter and increasing levels of extracellular DA, produced hyperactivity and straighter patterns of motor behavior in both the C57BL/6J and 129X1 mice, although at different doses. While amphetamine affected locomotor behavior of both C57BL/6J and 129X1 mice, there were no significant effects found in the 129S6 strain of mice. Amphetamine did have significant effects, however, in both cross-modal PPI and startle reactivity in the 129S6 mice, demonstrating that amphetamine was exerting some behavioral effects in this strain. These findings suggest that considerable differences may exist in the motor pathways in the 129S6 mice compared with the other two inbred strains. Thus, the 129S6 strain is a good candidate for studies of PPI and startle reactivity, but it may prove to be a poor candidate for locomotor activity studies. There have been several reports of strain differences between inbred lines of mice, both in behavioral phenotypes and in response to pharmacological manipulations (see Introduction). Taken together, these findings further emphasize the importance of selecting an appropriate strain of mice for behavioral and pharmacological characterization experiments and, furthermore, for the creation of mutant and transgenic mouse lines.

The present results establish that three inbred strains of mice display cross-modal as well as intramodal PPI, and that activation of DA systems via amphetamine decreases PPI and startle reactivity in a dose-, sensory modality-, and strain-specific manner. Furthermore, both the amount of locomotor activity and pattern of motor sequences are altered differentially after treatment with amphetamine in C57BL/6J and 129X1 mice, but not in 129S6 mice. While these experiments are consistent with the role of DA in the modulation of PPI and locomotor patterns in mice, they also highlight the importance of selecting an appropriate strain of mice for behavioral, pharmacological, and genetic studies.