- Email: [email protected]
An assessment of behavioral aging in the Mongolian gerbil
ExperimentalGerontology,Vol. 32, No. 6, pp. 707-717, 1997 Publishedby ElsevierScienceInc. Printedin the USA. All rights reserved 0531-5565/97$0.00 + .00 ELSEVIER
PII S0531-5565(97)00055-7
AN ASSESSMENT OF BEHAVIORAL AGING IN THE MONGOLIAN GERBIL
EDWARD L. SPANGLER,l JOHN HENGEMIHLE,1 GRANT BLANK,2 DOREY L. SPEER,1 STACIE BRZOZOWSKI,2 NAMISHA PATEL,1 AND DONALD K. INGRAMl ~Molecular Physiology and Genetics Section, Nathan W. Shock Laboratories, Gerontology Research Center, National Institute on Aging, NIH, Baltimore, Maryland 21224 and 2Department of Psychology, Towson University, Towson, Maryland 21252
Abstract--Male Mongolian gerbils (Meriones unguiculatus) 14-54 months old (n = 77) were evaluated in a battery of psychomotor (open field, locomotor, and runwheel activity, rotorod performance) and learning (one-way active avoidance in a straight runway and in 14-unit T-maze performance) tests. Body weight and seizure activity were also monitored. According to linear regression analysis, runwheel activity decreased with age; and the number of errors in the 14-unit T-maze increased as a function of age (ps < 0.05). None of the other behavioral measures or body weight were significantly correlated with age. This gerbil strain (Tumblebrook Farms; West Brookfield, MA) tended to be very prone to seizures with 64% of the gerbils experiencing at least one seizure while being tested. Seizures tended to occur when the gerbil was exposed to a novel situation (e.g., initial weighing, placement on the rotorod). An age-related decline in some aspects of psychomotor and learning performance was observed, suggesting the gerbil as an additional mammalian model of aging. The high incidence of seizure activity presented a complicating and confounding variable to the interpretation of the results of the behavioral tests used in the present study. Interventions to control seizure activity (e.g., systematic, controlled breeding; adaptation to apparati) in this model will likely increase its viability as a mammalian model of aging. Published by
Elsevier Science Inc. Key Words: animal model, epilepsy, motor skills, learning, memory INTRODUCTION THE GERBIL (Meriones unguiculatus), described extensively by Thiessen and Yahr (1977), has b e e n suggested as a model of m a m m a l i a n aging (see Cheal, 1986; Hazzard et al., 1991, for review). In a series of studies, Cheal (1987a, b, 1995) evaluated age-related behavioral changes
Correspondence to: Edward L. Spangler, Gerontology Research Center, 4940 Eastern Avenue, Box 11, Baltimore, MD 21224 (Received 20 December 1996; Accepted 16 June 1997) 707
708
EL. SPANGLER et al.
in a large group of male and female gerbils longitudinally while performing comparisons with cross-sectional/samples of the same age (median age = 36 months; maximum lifespan = 54 months). This rodent was observed to remain active across its lifespan with little change in activity scores, clinging or climbing behavior (Cheal, 1986), although latency to climb down from a pole declined significantly and less rearing was noted with age. Age-related decrements in the ability of the gerbil to learn an eight-arm radial maze have been reported (Carney et al., 1991). Young (three-month) gerbils made significantly fewer arm reentries when patrolling a radial arm maze compared to 15- to 18-month-old gerbils, and performance in the older gerbils was significantly improved when N-tert-butyl-a-phenylnitrone (PBN), a spin-trap agent, was administered chronically. Administration of PBN also improved performance in ischemic gerbils that had been given three-minute bilateral carotid artery occlusion. Stanley et al. (1993) reported that aged gerbils given a bilateral carotid artery occlusion and then treated with PBN performed better in learning the radial arm maze than untreated old gerbils. These studies indicate the relative benefits of free radical scavenger treatment in a gerbil ischemia model as well as in aged gerbils. The Mongolian gerbil has been reported to be seizure prone. While the susceptibility or propensity to seizure activity makes the gerbil a potentially valuable model of epilepsy (Cheal 1986), this represents a potential confound to the interpretation of behavioral results. Cheal (1987b) indicated that before two months of age gerbils do not show seizure activity during handling, experimental testing, or when a novel stimulus is presented. After the age of two months, seizures occurred during behavioral testing, but adaptation/habituation attenuated the incidence of seizures to suggest that handling or exposure to a novel apparatus prior to testing had a beneficial effect. Cheal (1995) has also reported an age-related increase in seizure activity. We have developed a battery of behavioral tests (Spangler et al., 1994) that have consistently demonstrated age-related changes in a number of strains of rats and mice. The objective in the present study was to determine whether administration of this battery of tests to different ages of gerbils would reveal a similar pattern of age-related changes in performance. MATERIALS AND METHODS
Subjects Experimentally naive male Mongolian gerbils (Meriones unguiculatus; Tumblebrook Farms; West Brookfield, MA) between the ages of 12 and 52 months (n = 77) were shipped from Cortex Pharmaceuticals (Irvine, CA) to the Gerontology Research Center (GRC). All of the gerbils were housed either singly or in groups of up to four animals in large plastic cages located in a movable metal rack and containing corn cob bedding. The grouping of animals remained identical to that maintained prior to arrival at the GRC. Food and filtered water were provided ad libitum. The vivarium was kept at a constant temperature of 23°C, and had a 12-h light/12-h dark cycle. The gerbils were allowed to acclimate to the vivarium for one month prior to commencement of behavioral testing. All behavioral experiments were conducted during the light cycle (lights on 0600 h/off 1800 h), and for each test the gerbils were transported to the testing rooms 20 min prior to the start of the experiment. Body weight measurements were taken weekly. A record was maintained of seizure activity observed during weekly weighing and during behavioral testing (described below) in the following tests: rotorod, straight runway training and 14-unit T-maze testing. Subjects were removed from rotorod testing if a seizure occurred during testing. Seizures noted during straight
709
BEHAVIORAL AGING IN GERBILS TABLE 1. BEHAVIORAL TEST SCHEDULE
Behavioral test (n) 24-h locomotor activity (66) 1 week runwheel activity (74) 3-min rotorod (40) 10-min square open field (73) Shock-motivatedone-wayactive avoidance in straight runway (37) Shock-motivatedacquisition performancein 14-unit T-maze (25)
Dependent measure(s) Total activity count (photocell beams broken) m traveled/day Total number of falls Distance traveled (cm) Percent correct shock avoidances Errors, runtime, shock frequency, and duration
runway testing resulted in gerbils being removed from this test and from 14-unit T-maze testing, while gerbils observed with a seizure during 14-unit T-maze training were also removed from that test. Observation of seizure activity in the other behavioral measures was either not possible, or not feasible. Behavioral apparatus and procedure Table 1 describes the battery of behavioral tests, and the dependent measures of interest for each test. Testing in a Morris-type water maze had been planned, but this test was not utilized due to the extreme stress that the gerbils exhibited during pilot testing. Nor were gerbils tested on an inclined screen described previously (Spangler et al., 1994). In pilot tests the gerbils jumped from the inclined screen each time they were placed onto the wire mesh. A brief description of each behavioral apparatus and the behavioral procedures that were utilized follows. Locomotor activity. Locomotor activity of each gerbil was measured by removing the gerbil from its home cage and placing it into a square open field that consisted of a wire mesh grid (1.5 cm square openings; grid 40 cm 2) that was surrounded by opaque white plastic walls (17 cm high). The entire open field was elevated by four legs (14 cm high) that rested on the surface of a rectangular testing platform (112 × 104 cm) painted white that was surrounded by white curtains (183 cm long) and was elevated 60 cm from the floor of the room. A video camera mounted 125 cm above the open field and connected to a Videomex tracking system (Columbus Instruments; Columbus, OH) was initiated five seconds after the gerbil was placed into the center of the open field. Software on a personal computer (PC) that operated the tracking system was used to score the total number of beam interruptions by the gerbil during a 24-h test period. Overhead fluorescent lights provided adequate contrast for the tracking system. The test area was surrounded by five speakers connected to a stereo system that was used to mask background noise. Runwheel activity. Measurement of voluntary wheel running was performed using an automated, PC-driven system (Dataquest 5.0; Data Sciences International, St. Paul, MN) that was connected simultaneously to runwheels (Nalge, Rochester, NY). Basically, each wheel unit consisted of a stainless steel cage (50 × 26.8 × 36.4 cm) with a contiguous runwheel attachment (wheel circumference -- 105.4 cm; width = 8.9 cm). The cage had a total floor area of 929 cm 2, and a free floor area of 516 cm 2 with the wheel in place. Complete wheel revolutions of the activity wheel were counted with a magnetic switch connected to the PC that recorded the data for later analysis. The number of revolutions was recorded every hour, and the count was transformed into the mean daily distance traveled.
710
E.L. SPANGLER et al.
Rotorod. Described in detail previously (Spangler et al., 1994), this task required the gerbil to remain on a scored plastic rotating cylinder (47 cm diam) that contained metal dividers (48 cm high) to prevent attempts to move across the cylinder and escape. Located in a plastic frame housing, the rotorod was rotated by a chain link pulley system attached to a motor that rotated the rod at 4 rpm. The top of the rotatable cylinder was located 56 cm above the base of the frame. The entire base area was covered with foam rubber padding to cushion falls and prevent injury to the animal. For the test, each gerbil was removed gently from the home cage by grasping the base of its tail, and placing the gerbil on top of and parallel to the rotating rod while simultaneously initiating a clock. The clock was stopped and a fall was recorded each time the gerbil fell from the rod. The gerbil was then returned to the rod, and the clock reinitiated. This procedure was followed each time the gerbil fell from the rod until three minutes had elapsed, at which time the gerbil was returned to its home cage. Square open field. The open field consisted of a white plastic square (40 crn/side; 17 cm high) with a wire mesh floor. The open field was elevated 14 cm above a white rectangular table (112 X 104 cm). A video camera located 125 cm above the open field arena was connected to an image analyzing system (Videomex; Columbus Instruments; Columbus, OH) that monitored movement in the arena, and stored the data for subsequent analysis. The test arena was surrounded by four speakers that provided music to mask any background noise. For the open field test, each gerbil was removed gently from the home cage, and was placed into the center of the arena. Five seconds later the image analysis system was activated, and locomotor activity in the arena measured for the next 10 min. After each test, the gerbil was returned to its home cage, and the arena was cleaned with a 95% ethanol solution to mask any odors. Straight runway. A straight runway (2 m long), described in detail elsewhere (Spangler et al., 1986), was used as a pretraining avoidance task. The runway was constructed of clear plastic surrounded by gray walls to minimize external stimuli, and had a stainless steel diagonally oriented grid floor that was wired to receive scrambled shock from a Coulbourn Instruments (Lehigh Valley, PA) E13-08 grid floor shocker. Black plastic boxes with a guillotine door and a retractable rear wall with a metal handle could be placed over the grid floor to serve interchangeably as start and goal boxes. For training, each gerbil was removed gently from its home cage and placed into one of the black plastic boxes, and the guillotine door closed. The box was then placed into the start area of the runway over the grid floor, and the guillotine door was raised. The gerbil was forced from the box into the runway by gently pushing the back wall forward until the gerbil was expelled into the runway. The guillotine door of the box was then closed, and the timer was started. If the gerbil failed to move down the runway to the goal box within 10 s, scrambled foot shock (0.3 mA) was initiated until the gerbil entered the box at the opposite end of the runway. The training criteria for each gerbil was 13 avoidances (no shock) out of 15 trials to a maximum of 40 trials. If a gerbil received 60 s of shock during an individual trial, it was removed to the goal box. Three trials of 60 s of shock resulted in the gerbil being removed from further testing. All gerbils that met the criteria began testing in the 14-unit T-maze on the following day. 14-Unit T-maze. The 14-unit T-maze (2 X 2 m; Fig. 1), described in detail previously (Spangler et al., 1986), was constructed of clear plastic and was surrounded by gray walls to minimize external stimuli. A stainless steel diagonally oriented grid floor was wired to receive
BEHAVIORAL AGING IN GERBILS
71 1
__l
L1
-]
FIG. 1. Arrows denote correct pathway from Start (S) to Goal (G) Box in the 14-unit T-maze. Solid lines ( - - ) indicate placement of operational guillotine doors while dashed lines (. . . . ) indicate placement of nonfunctional guillotine doors. G - - g o a l area; S - - S t a r t area; - - guillotine door; - - false guillotine door.
scrambled shock from a Coulboum Instruments E13-08 grid floor shocker. The maze was divided into sections by four guillotine doors that could be raised or lowered. Nonfunctional guillotine doors were located opposite these operational doors to prevent use o f a door as a cue. Operational guillotine doors were also located at the start and goal areas. The criterion for the maze training procedure was similar to the pretraining procedure criterion Each gerbil had to move from the start area past the first operational guillotine door within 10 s to avoid shock, or escape past the door once the shock (0.3 mA) was initiated. After the gerbil passed under the opened guillotine door, it was closed to prevent backtracking, and the 10-s timer was reset. The same procedure was followed as the gerbil made its way through the maze and past each operational guillotine door. If the shock administered to a gerbil on an individual trial exceeded 300 s, the trial was terminated and the gerbil was removed and placed into one of the goal boxes until the next trial. Once a gerbil received three 300-s trials, it was dropped from further testing. Each gerbil received 20 trials in the 14-unit T-maze over two days; five trials during a morning session and five trials during an afternoon session on each day. A two-minute intertrial interval separated each trial. During this intertrial interval, a pulley system was used to hoist up the maze to clean the grid floor with ethanol to prevent odor cues, and the data for the trial was recorded. Data for each trial was transferred from a microprocessor to a PC to allow for later analysis o f errors, and runtime. Shock frequency and duration were recorded manually from an event recorder and a timer after each trial.
712
ILL. S P A N G L E R et al.
180 160 140 ÷
+t-
120
+
++
..E 100 >, "0 0 ¢'n
80
+
~
÷
÷
60 40 200 10
I
I
i
I
I
t
I
I
15
20
25
30
35
40
45
50
55
Age (mo)
FIG. 2. Scatterplot showing mean body weight (five weighings) and regression of age onto body weight.
Statistical analysis. Linear regression analyses were conducted by regressing age on each dependent measure of performance, body weight, incidence, and number of seizures. The number of gerbils included in behavioral tests and subsequently included in statistical analysis declined across the course of the experiments as deaths occurred, morbidity was observed, or the animal was removed from an experiment due to the occurrence of a seizure or was removed from the test for failing to meet a criterion (e.g., straight runway and 14-unit T-maze testing). RESULTS
Body weight Figure 2 provides a scatterplot of the mean body weight for five weighings across the age groups, and the regression of age onto body weight. No relationship between body weight and age was observed, r(77) = - 0 . 1 5 , NS. Between 12 and 26 months body weights generally ranged between 80 and 120 g, as shown in Fig. 2. Only seven gerbils were older than 26 months; thus, it was difficult to interpret or generalize about body weight at older ages.
Seizure activity Age was not observed to be related to the incidence (i.e., at least one) of seizures, r(77) = - 0 . 0 7 , nor was there a relationship observed between number of seizures and age, r(77) =
BEHAVIORAL AGING IN GERBILS
713
-0.04, NS. At least one seizure was observed during testing or weighing in 47 of 77 (64%) of the gerbils in the study: Seizures were more likely to occur during exposure to a novel event. During the first weighing, for instance, 16 of 77 (21%) of the gerbils experienced a seizure. By the third weighing, 3 of 65 (5%) of the gerbils experienced a seizure. Stress of the test also appeared to play a role, as a single exposure to the rotorod resulted in 33 of 73 (44%) gerbils tested experiencing a seizure, while 10 of 49 (20%) seized in straight runway training and 7 of 32 (22%) in the 14-unit T-maze.
Behavioral tests Scatterplots for dependent measures of behavioral performance and the results of the linear regression analyses are displayed in Figs. 3 and 4. Log transformations were performed for the total 24-h activity and wheel revolution measures due to heterogeneity of variance among subjects within age groups. Total activity measured over a 24-h period was not significantly lower in older gerbils, r(64) = 0.18, NS, but older gerbils were less active as measured by the number of wheel revolutions during one week in an automated rnnwheel, r(72) = -0.47, p < 0.001. Neither the distance traveled in a 10-min square open-field test, r(71) = 0.09, NS, nor the number of falls from a rotorod during a three-minute exposure, r(38) = 0.27, NS, declined significantly with age. Thus, in tests of psychomotor performance as displayed in Fig. 3, only activity in a runwheel was significantly lower in old gerbils. Variability in the number of falls from the rotorod, at least in young animals, may have been influenced by the tendency of some young gerbils to jump from the moving rod. The percent of active avoidance responses in the straight runway showed a modest but nonsignificant decline with age, r(37) = -0.26. An age-related decline in performance in the 14-unit T-maze was observed (Fig. 4), but only for errors, r(23) = 0.45, p < 0.05. However, there was marked individual variability in acquisition of the maze, as depicted in the scatterplot for errors. No significant effects were observed for the runtime, r(23) = 0.10, shock frequency, r(23) = 0.18, or shock duration, r(23) = 0.03, measures when age was regressed onto these measures. DISCUSSION Evidence of behavioral aging in the Mongolian gerbil (Meriones unguiculatus) was observed in tests of psychomotor performance and learning and memory in a cross-sectional study. Specifically, age was associated with reduced runwheel activity measured over a one-week testing interval and with increased errors in the acquisition of a shock-motivated 14-unit T-maze without effects on other aspects of maze performance (i.e., runtime, shock frequency and duration). No differences were observed in spontaneous 24-h locomotor activity, the distance traveled in a 10-min square open field test, the number of falls from a rotating rod, or in the number of successful avoidances in one-way active avoidance in a straight runway. The high incidence and number of seizures observed at all age levels, however, complicates the interpretation of the results obtained in the present investigation. Moreover, the inability to obtain adequate numbers of gerbils beyond 26 months of age makes the interpretation of nonsignificant age correlations difficult. Pilot data with several additional tests contraindicated their usage as part of the battery, or required adaptation. Pilot testing of six young gerbils with an inclined screen resulted in the
A
10
,"
*
15
I
I
.
I
20
25
o
.
•
~
I
I
",,
I
,°
÷
30
I
°
I
÷
35
I
I
40
I
I
45
I
""
I
I
50
o
I
,
55
n-
90
£
t,l..
o
IJ_
E3
U ¢-.
0
5--
10--
15--
20--
25--
30--
35
D
C
0
2000 --
4000 --
6000 --
8000 --
10000 - -
10
°
15
I
I
°
I
I
20
°*
°
25
I
I
°
30
I
I
35
I
I
f
40
I
I
J
J
45
I
I
50
I
I
55
FIG. 3. Scatterplots depicting individual gerbil performance on psychomotor tests, and the regression of age on these variables. Displayed are (A) distance traveled in a 24-h test of locomotor activity; (B) mean daily wheel revolutions during one week in an activity wheel; (C) distance traveled during 10 min in a square open field; and (D) number of falls from a rotating rod during a 3-min interval.
B
0
2-
_~
o
3--
1 -
O
t~
8_
t~
o
I--
2-
3-
<
O
4--
?,
o
5--
12000
14 12 10 8uJ
6-4-
1:
2-
120
100-
80eo(1
40-
20 0 80 70 60-
~
5o 40
Q ~
30.
2010 0 5
t ~'
3
.
I 10
20
30
40
Age (mo)
FIG. 4. Scatterplots displaying individual gerbil performance on dependent measures of 14-unit T-maze performance and the regression of age on these variables.
7 ]6
E,L, S P A N G L E R er a/.
gerbils immediately jumping off when placed on the screen. Likewise, when placed on a smaller version of the rotorod, used for testing mice in our laboratory, the gerbils jumped as soon as they were placed on the rod. The inclined screen was dropped from the battery, and rotorod testing was completed but with a larger and more elevated version of the rotorod used to test rats in our laboratory. In some young gerbils this tendency to jump was also observed with the larger rotorod. Pilot testing with a version of the Morris water maze used to test mice in our laboratory contraindicated usage of this test because all of the gerbils tested had to be rescued from the tank. Results of these pilot tests suggest that for the gerbil, a small desert rodent, swimming may not have sufficient "survival value" compared to running, digging, or burrowing. On the other hand, the high level of spontaneous activity and excellent visual acuity in the gerbil (Wilkinson, 1984) may make it too suited to escape by jumping for some standard laboratory tests (e.g., inclined screen) to be useful for assessment. In agreement with Cheal (1987a), we observed that handling reduced the number of subjects experiencing seizures. In our study, the percent of subjects experiencing seizures during weekly body weight measurements declined from 21% to 5% from the first to the third weighing while Cheal (1987a) had reported a decrease from 3 0 - 4 0 % on the first day of handling to 2% by the fifth day. Unlike Cheal (1995), the proportion of gerbils in our study experiencing seizures did not increase with age. In the Cheal study, the gerbils were followed longitudinally and were separated into groups based upon either a control condition in which the animals were exposed to a small indoor environment for 1 h a month or an environmental enrichment experience consisting of 1 h a month in an outdoor arena from 1 to 28 months of age. In that study, the environmentally enriched gerbils had a lower incidence of seizures than controls from 4 to 31 months of age. However, after 31 months of age the incidence of seizure activity was high in both groups. In our cross-sectional study, neither the proportion of gerbils experiencing at least one seizure nor the total number of seizures observed was correlated with age. The inability to monitor seizure activity in a number of tasks (square open field, runwheel activity, locomotor activity) and the low number of aged gerbils greater than 26 months of age may have influenced the results in our study. The number of gerbils dropped from the rotorod, straight runway and 14-unit T-maze testing also raises serious questions regarding the utility of these types of test for studies of behavioral aging in the gerbil. Adaptation of the gerbil to these apparati may reduce the incidence of seizures and, thus, the loss of subjects, but may also further complicate the interpretation of results. That is, learning may occur with exposure to the rotorod. Additional adaptation in shock avoidance in the straight runway by increasing the number of days and trials to criterion may increase the number of subjects available for testing in the 14-unit T-maze without confounding the interpretation of results. Thus, this phenomena will require further evaluation to determine the effects of a variety of influences including the amount and type of handling, stress of the tests utilized, and the overall seizure activity of the strain utilized. Increasing the number and variety of animal models available for aging research has been a goal of gerontologists and the National Institute on Aging (Hazzard et al., 1991). Based on the results observed in this study, we conclude that the gerbil: a) is an appropriate model for epilepsy, as suggested by Cheal (1986); and b) might be a suitable model for age-related learning and memory deficits, in agreement with tests of learning and memory tests in a radial arm maze (Carney et al., 1991; Stanley et al., 1993) if care is taken to reduce the incidence of seizures. Moreover, the gerbil would be an appropriate model because it remains active and inquisitive through old age (Cheal, 1986), does not gain significant body weight over its lifespan, and appears relatively coordinated (rotorod) and without physical impairment through-
BEHAVIORALAGINGIN GERBILS
717
out its lifespan. However, while some of the traditional age-related confounds (decline in activity levels, increased body fat, less agile, and mobile) are lessened in this model, the issue of seizure activity needs to be controlled and/or addressed more clearly (e.g., adaptation to experimenter, apparatus, and possibly the task). Additionally, the research of others suggests the gerbil may make an excellent model of ischemia, due to the incomplete Circle of Willis, and for treatment of this condition (e.g., free radical scavengers) (Phillis and Clough-Helfman, 1990; Carney et al., 1991). Rederiving gerbil colonies may serve to reduce the incidence of seizure activity in individual colonies and, thus, may make the use of this animal as a model of behavioral aging, particularly learning and memory, more feasible. In summary, based on our previous findings in a variety of laboratory strains of rats and mice, we anticipated that the gerbil would further confirm the age-related differences observed in the battery of psychomotor and learning and memory tests. Our findings only partially confirmed our hypothesis. Failure to observe differences in a number of psychomotor tests and in the runtime and shock measures in the 14-unit T-maze may be attributable to the increased exploratory activity and inquisitiveness in the gerbil (Cheal, 1995). Future studies of aging in the gerbil should utilize a greater age distribution with more animals at or beyond the median life span ( > 3 6 months; Cheal, 1995). Acknowledgments--The Gerontology Research Center, NIH, is fully accredited by the American Association for Accreditation of Laboratory Animal Care. We gratefully acknowledge the donation of the gerbils by Cortex Pharmaceuticals, and the editorial assistance of Gloria Dunnigan.
REFERENCES CARNEY, J.M., STARKE-REED, P.E., OLIVER, C.N., LANDUM, R.W., CHENG, M.S., and WU, J.F. Reversal of age-related increase in brain protein oxidation, decrease in enzyme activity, and loss in temporal and spatial memory by chronic administration of the spin-trapping compound N-tert-butyl-a-phenylnitrone. Proc. Natl. Acad. Sci. USA, 88, 3633-3636, 1991. CHEAL, M.L. The gerbil: A unique model for research on aging. Exp. Aging Res. 13, 3-21, 1986. CHEAL, M.L. Adult development: Plasticity of stable behavior. Exp. Aging Res. 13, 29-36, 1987a. CHEAL, M.L. Lifespan environmental influences on species typical behavior of Meriones unguiculatus. In: Evolution of Longevity in Animals, Woodhead, A.D. and Thompson, K.H. (Editors), pp. 145-159, Plenum, New York, 1987b. CHEAL, M. Multiple factors modulating courtship, habituation, attention, and other motivated and motivating behaviors. In: Biological Perspectives on Motivated and Cognitive Activities. Wong, R. (Editor), Ablex Press. Norwood, NJ, 1995. HAZARD, D.G., WARNER, H.R., and FINCH, C.E. National Institute on Aging, NIH workshop on alternative animal models for research on aging. Exp. GerontoL 26, 411-439, 1991. PHILLIS, J.W. and CLOUGH-HELFMAN, C. Protection from cerebral ischemic injury in gerbils with the spin trap agent N-tert-butyl-a-phenylnitrone (PBN). Neurosci. Lett. 116, 315-319, 1990. SPANGLER, E.L., RIGBY, P., and INGRAM, D.K. Scopolamine impairs learning performance of rats in a 14-unit T-maze. Pharmacol. Biochem. Behav., 25, 673-679, 1986. SPANGLER, E.L., WAGGLE, K.S., HENGEMIHLE, J., ROBERTS, D., HESS, B., and INGRAM, D.K. Behavioral assessment of aging in male Fischer 344 and Brown Norway rat strains and their F~ hybrid. Neurobiol. Aging 15, 319-328, 1994. STANLEY, J.A., BULL, C.O., MOON, S.L., and ORTIZ, A.A. N-tert-Butyl-a-Phenyl-Nitrone (BPN), a free radical scavenger, facilitates learning by elderly gerbils following bilateral carotid occlusion (BCO). Soc. Neurosci. Abstr. 1653, 1993. THIESSEN, D. and YAHR, P. The Gerbil in Behavioral Investigations. University of Texas Press; Austin, TX, 1977. WILKINSON, F. The development of visual acuity in the Mongolian gerbil (Meriones unguiculatus). Behav. Brain Res. 83-94, 1984. ZHANG, J-R., ANDRUS, P.K., and HALL, E.D. Age-related regional changes in hydroxyl radical stress and antioxidant in gerbil brain. Soc. Neurosci. Abstr. 19: 458; 1993.
PII S0531-5565(97)00055-7
AN ASSESSMENT OF BEHAVIORAL AGING IN THE MONGOLIAN GERBIL
EDWARD L. SPANGLER,l JOHN HENGEMIHLE,1 GRANT BLANK,2 DOREY L. SPEER,1 STACIE BRZOZOWSKI,2 NAMISHA PATEL,1 AND DONALD K. INGRAMl ~Molecular Physiology and Genetics Section, Nathan W. Shock Laboratories, Gerontology Research Center, National Institute on Aging, NIH, Baltimore, Maryland 21224 and 2Department of Psychology, Towson University, Towson, Maryland 21252
Abstract--Male Mongolian gerbils (Meriones unguiculatus) 14-54 months old (n = 77) were evaluated in a battery of psychomotor (open field, locomotor, and runwheel activity, rotorod performance) and learning (one-way active avoidance in a straight runway and in 14-unit T-maze performance) tests. Body weight and seizure activity were also monitored. According to linear regression analysis, runwheel activity decreased with age; and the number of errors in the 14-unit T-maze increased as a function of age (ps < 0.05). None of the other behavioral measures or body weight were significantly correlated with age. This gerbil strain (Tumblebrook Farms; West Brookfield, MA) tended to be very prone to seizures with 64% of the gerbils experiencing at least one seizure while being tested. Seizures tended to occur when the gerbil was exposed to a novel situation (e.g., initial weighing, placement on the rotorod). An age-related decline in some aspects of psychomotor and learning performance was observed, suggesting the gerbil as an additional mammalian model of aging. The high incidence of seizure activity presented a complicating and confounding variable to the interpretation of the results of the behavioral tests used in the present study. Interventions to control seizure activity (e.g., systematic, controlled breeding; adaptation to apparati) in this model will likely increase its viability as a mammalian model of aging. Published by
Elsevier Science Inc. Key Words: animal model, epilepsy, motor skills, learning, memory INTRODUCTION THE GERBIL (Meriones unguiculatus), described extensively by Thiessen and Yahr (1977), has b e e n suggested as a model of m a m m a l i a n aging (see Cheal, 1986; Hazzard et al., 1991, for review). In a series of studies, Cheal (1987a, b, 1995) evaluated age-related behavioral changes
Correspondence to: Edward L. Spangler, Gerontology Research Center, 4940 Eastern Avenue, Box 11, Baltimore, MD 21224 (Received 20 December 1996; Accepted 16 June 1997) 707
708
EL. SPANGLER et al.
in a large group of male and female gerbils longitudinally while performing comparisons with cross-sectional/samples of the same age (median age = 36 months; maximum lifespan = 54 months). This rodent was observed to remain active across its lifespan with little change in activity scores, clinging or climbing behavior (Cheal, 1986), although latency to climb down from a pole declined significantly and less rearing was noted with age. Age-related decrements in the ability of the gerbil to learn an eight-arm radial maze have been reported (Carney et al., 1991). Young (three-month) gerbils made significantly fewer arm reentries when patrolling a radial arm maze compared to 15- to 18-month-old gerbils, and performance in the older gerbils was significantly improved when N-tert-butyl-a-phenylnitrone (PBN), a spin-trap agent, was administered chronically. Administration of PBN also improved performance in ischemic gerbils that had been given three-minute bilateral carotid artery occlusion. Stanley et al. (1993) reported that aged gerbils given a bilateral carotid artery occlusion and then treated with PBN performed better in learning the radial arm maze than untreated old gerbils. These studies indicate the relative benefits of free radical scavenger treatment in a gerbil ischemia model as well as in aged gerbils. The Mongolian gerbil has been reported to be seizure prone. While the susceptibility or propensity to seizure activity makes the gerbil a potentially valuable model of epilepsy (Cheal 1986), this represents a potential confound to the interpretation of behavioral results. Cheal (1987b) indicated that before two months of age gerbils do not show seizure activity during handling, experimental testing, or when a novel stimulus is presented. After the age of two months, seizures occurred during behavioral testing, but adaptation/habituation attenuated the incidence of seizures to suggest that handling or exposure to a novel apparatus prior to testing had a beneficial effect. Cheal (1995) has also reported an age-related increase in seizure activity. We have developed a battery of behavioral tests (Spangler et al., 1994) that have consistently demonstrated age-related changes in a number of strains of rats and mice. The objective in the present study was to determine whether administration of this battery of tests to different ages of gerbils would reveal a similar pattern of age-related changes in performance. MATERIALS AND METHODS
Subjects Experimentally naive male Mongolian gerbils (Meriones unguiculatus; Tumblebrook Farms; West Brookfield, MA) between the ages of 12 and 52 months (n = 77) were shipped from Cortex Pharmaceuticals (Irvine, CA) to the Gerontology Research Center (GRC). All of the gerbils were housed either singly or in groups of up to four animals in large plastic cages located in a movable metal rack and containing corn cob bedding. The grouping of animals remained identical to that maintained prior to arrival at the GRC. Food and filtered water were provided ad libitum. The vivarium was kept at a constant temperature of 23°C, and had a 12-h light/12-h dark cycle. The gerbils were allowed to acclimate to the vivarium for one month prior to commencement of behavioral testing. All behavioral experiments were conducted during the light cycle (lights on 0600 h/off 1800 h), and for each test the gerbils were transported to the testing rooms 20 min prior to the start of the experiment. Body weight measurements were taken weekly. A record was maintained of seizure activity observed during weekly weighing and during behavioral testing (described below) in the following tests: rotorod, straight runway training and 14-unit T-maze testing. Subjects were removed from rotorod testing if a seizure occurred during testing. Seizures noted during straight
709
BEHAVIORAL AGING IN GERBILS TABLE 1. BEHAVIORAL TEST SCHEDULE
Behavioral test (n) 24-h locomotor activity (66) 1 week runwheel activity (74) 3-min rotorod (40) 10-min square open field (73) Shock-motivatedone-wayactive avoidance in straight runway (37) Shock-motivatedacquisition performancein 14-unit T-maze (25)
Dependent measure(s) Total activity count (photocell beams broken) m traveled/day Total number of falls Distance traveled (cm) Percent correct shock avoidances Errors, runtime, shock frequency, and duration
runway testing resulted in gerbils being removed from this test and from 14-unit T-maze testing, while gerbils observed with a seizure during 14-unit T-maze training were also removed from that test. Observation of seizure activity in the other behavioral measures was either not possible, or not feasible. Behavioral apparatus and procedure Table 1 describes the battery of behavioral tests, and the dependent measures of interest for each test. Testing in a Morris-type water maze had been planned, but this test was not utilized due to the extreme stress that the gerbils exhibited during pilot testing. Nor were gerbils tested on an inclined screen described previously (Spangler et al., 1994). In pilot tests the gerbils jumped from the inclined screen each time they were placed onto the wire mesh. A brief description of each behavioral apparatus and the behavioral procedures that were utilized follows. Locomotor activity. Locomotor activity of each gerbil was measured by removing the gerbil from its home cage and placing it into a square open field that consisted of a wire mesh grid (1.5 cm square openings; grid 40 cm 2) that was surrounded by opaque white plastic walls (17 cm high). The entire open field was elevated by four legs (14 cm high) that rested on the surface of a rectangular testing platform (112 × 104 cm) painted white that was surrounded by white curtains (183 cm long) and was elevated 60 cm from the floor of the room. A video camera mounted 125 cm above the open field and connected to a Videomex tracking system (Columbus Instruments; Columbus, OH) was initiated five seconds after the gerbil was placed into the center of the open field. Software on a personal computer (PC) that operated the tracking system was used to score the total number of beam interruptions by the gerbil during a 24-h test period. Overhead fluorescent lights provided adequate contrast for the tracking system. The test area was surrounded by five speakers connected to a stereo system that was used to mask background noise. Runwheel activity. Measurement of voluntary wheel running was performed using an automated, PC-driven system (Dataquest 5.0; Data Sciences International, St. Paul, MN) that was connected simultaneously to runwheels (Nalge, Rochester, NY). Basically, each wheel unit consisted of a stainless steel cage (50 × 26.8 × 36.4 cm) with a contiguous runwheel attachment (wheel circumference -- 105.4 cm; width = 8.9 cm). The cage had a total floor area of 929 cm 2, and a free floor area of 516 cm 2 with the wheel in place. Complete wheel revolutions of the activity wheel were counted with a magnetic switch connected to the PC that recorded the data for later analysis. The number of revolutions was recorded every hour, and the count was transformed into the mean daily distance traveled.
710
E.L. SPANGLER et al.
Rotorod. Described in detail previously (Spangler et al., 1994), this task required the gerbil to remain on a scored plastic rotating cylinder (47 cm diam) that contained metal dividers (48 cm high) to prevent attempts to move across the cylinder and escape. Located in a plastic frame housing, the rotorod was rotated by a chain link pulley system attached to a motor that rotated the rod at 4 rpm. The top of the rotatable cylinder was located 56 cm above the base of the frame. The entire base area was covered with foam rubber padding to cushion falls and prevent injury to the animal. For the test, each gerbil was removed gently from the home cage by grasping the base of its tail, and placing the gerbil on top of and parallel to the rotating rod while simultaneously initiating a clock. The clock was stopped and a fall was recorded each time the gerbil fell from the rod. The gerbil was then returned to the rod, and the clock reinitiated. This procedure was followed each time the gerbil fell from the rod until three minutes had elapsed, at which time the gerbil was returned to its home cage. Square open field. The open field consisted of a white plastic square (40 crn/side; 17 cm high) with a wire mesh floor. The open field was elevated 14 cm above a white rectangular table (112 X 104 cm). A video camera located 125 cm above the open field arena was connected to an image analyzing system (Videomex; Columbus Instruments; Columbus, OH) that monitored movement in the arena, and stored the data for subsequent analysis. The test arena was surrounded by four speakers that provided music to mask any background noise. For the open field test, each gerbil was removed gently from the home cage, and was placed into the center of the arena. Five seconds later the image analysis system was activated, and locomotor activity in the arena measured for the next 10 min. After each test, the gerbil was returned to its home cage, and the arena was cleaned with a 95% ethanol solution to mask any odors. Straight runway. A straight runway (2 m long), described in detail elsewhere (Spangler et al., 1986), was used as a pretraining avoidance task. The runway was constructed of clear plastic surrounded by gray walls to minimize external stimuli, and had a stainless steel diagonally oriented grid floor that was wired to receive scrambled shock from a Coulbourn Instruments (Lehigh Valley, PA) E13-08 grid floor shocker. Black plastic boxes with a guillotine door and a retractable rear wall with a metal handle could be placed over the grid floor to serve interchangeably as start and goal boxes. For training, each gerbil was removed gently from its home cage and placed into one of the black plastic boxes, and the guillotine door closed. The box was then placed into the start area of the runway over the grid floor, and the guillotine door was raised. The gerbil was forced from the box into the runway by gently pushing the back wall forward until the gerbil was expelled into the runway. The guillotine door of the box was then closed, and the timer was started. If the gerbil failed to move down the runway to the goal box within 10 s, scrambled foot shock (0.3 mA) was initiated until the gerbil entered the box at the opposite end of the runway. The training criteria for each gerbil was 13 avoidances (no shock) out of 15 trials to a maximum of 40 trials. If a gerbil received 60 s of shock during an individual trial, it was removed to the goal box. Three trials of 60 s of shock resulted in the gerbil being removed from further testing. All gerbils that met the criteria began testing in the 14-unit T-maze on the following day. 14-Unit T-maze. The 14-unit T-maze (2 X 2 m; Fig. 1), described in detail previously (Spangler et al., 1986), was constructed of clear plastic and was surrounded by gray walls to minimize external stimuli. A stainless steel diagonally oriented grid floor was wired to receive
BEHAVIORAL AGING IN GERBILS
71 1
__l
L1
-]
FIG. 1. Arrows denote correct pathway from Start (S) to Goal (G) Box in the 14-unit T-maze. Solid lines ( - - ) indicate placement of operational guillotine doors while dashed lines (. . . . ) indicate placement of nonfunctional guillotine doors. G - - g o a l area; S - - S t a r t area; - - guillotine door; - - false guillotine door.
scrambled shock from a Coulboum Instruments E13-08 grid floor shocker. The maze was divided into sections by four guillotine doors that could be raised or lowered. Nonfunctional guillotine doors were located opposite these operational doors to prevent use o f a door as a cue. Operational guillotine doors were also located at the start and goal areas. The criterion for the maze training procedure was similar to the pretraining procedure criterion Each gerbil had to move from the start area past the first operational guillotine door within 10 s to avoid shock, or escape past the door once the shock (0.3 mA) was initiated. After the gerbil passed under the opened guillotine door, it was closed to prevent backtracking, and the 10-s timer was reset. The same procedure was followed as the gerbil made its way through the maze and past each operational guillotine door. If the shock administered to a gerbil on an individual trial exceeded 300 s, the trial was terminated and the gerbil was removed and placed into one of the goal boxes until the next trial. Once a gerbil received three 300-s trials, it was dropped from further testing. Each gerbil received 20 trials in the 14-unit T-maze over two days; five trials during a morning session and five trials during an afternoon session on each day. A two-minute intertrial interval separated each trial. During this intertrial interval, a pulley system was used to hoist up the maze to clean the grid floor with ethanol to prevent odor cues, and the data for the trial was recorded. Data for each trial was transferred from a microprocessor to a PC to allow for later analysis o f errors, and runtime. Shock frequency and duration were recorded manually from an event recorder and a timer after each trial.
712
ILL. S P A N G L E R et al.
180 160 140 ÷
+t-
120
+
++
..E 100 >, "0 0 ¢'n
80
+
~
÷
÷
60 40 200 10
I
I
i
I
I
t
I
I
15
20
25
30
35
40
45
50
55
Age (mo)
FIG. 2. Scatterplot showing mean body weight (five weighings) and regression of age onto body weight.
Statistical analysis. Linear regression analyses were conducted by regressing age on each dependent measure of performance, body weight, incidence, and number of seizures. The number of gerbils included in behavioral tests and subsequently included in statistical analysis declined across the course of the experiments as deaths occurred, morbidity was observed, or the animal was removed from an experiment due to the occurrence of a seizure or was removed from the test for failing to meet a criterion (e.g., straight runway and 14-unit T-maze testing). RESULTS
Body weight Figure 2 provides a scatterplot of the mean body weight for five weighings across the age groups, and the regression of age onto body weight. No relationship between body weight and age was observed, r(77) = - 0 . 1 5 , NS. Between 12 and 26 months body weights generally ranged between 80 and 120 g, as shown in Fig. 2. Only seven gerbils were older than 26 months; thus, it was difficult to interpret or generalize about body weight at older ages.
Seizure activity Age was not observed to be related to the incidence (i.e., at least one) of seizures, r(77) = - 0 . 0 7 , nor was there a relationship observed between number of seizures and age, r(77) =
BEHAVIORAL AGING IN GERBILS
713
-0.04, NS. At least one seizure was observed during testing or weighing in 47 of 77 (64%) of the gerbils in the study: Seizures were more likely to occur during exposure to a novel event. During the first weighing, for instance, 16 of 77 (21%) of the gerbils experienced a seizure. By the third weighing, 3 of 65 (5%) of the gerbils experienced a seizure. Stress of the test also appeared to play a role, as a single exposure to the rotorod resulted in 33 of 73 (44%) gerbils tested experiencing a seizure, while 10 of 49 (20%) seized in straight runway training and 7 of 32 (22%) in the 14-unit T-maze.
Behavioral tests Scatterplots for dependent measures of behavioral performance and the results of the linear regression analyses are displayed in Figs. 3 and 4. Log transformations were performed for the total 24-h activity and wheel revolution measures due to heterogeneity of variance among subjects within age groups. Total activity measured over a 24-h period was not significantly lower in older gerbils, r(64) = 0.18, NS, but older gerbils were less active as measured by the number of wheel revolutions during one week in an automated rnnwheel, r(72) = -0.47, p < 0.001. Neither the distance traveled in a 10-min square open-field test, r(71) = 0.09, NS, nor the number of falls from a rotorod during a three-minute exposure, r(38) = 0.27, NS, declined significantly with age. Thus, in tests of psychomotor performance as displayed in Fig. 3, only activity in a runwheel was significantly lower in old gerbils. Variability in the number of falls from the rotorod, at least in young animals, may have been influenced by the tendency of some young gerbils to jump from the moving rod. The percent of active avoidance responses in the straight runway showed a modest but nonsignificant decline with age, r(37) = -0.26. An age-related decline in performance in the 14-unit T-maze was observed (Fig. 4), but only for errors, r(23) = 0.45, p < 0.05. However, there was marked individual variability in acquisition of the maze, as depicted in the scatterplot for errors. No significant effects were observed for the runtime, r(23) = 0.10, shock frequency, r(23) = 0.18, or shock duration, r(23) = 0.03, measures when age was regressed onto these measures. DISCUSSION Evidence of behavioral aging in the Mongolian gerbil (Meriones unguiculatus) was observed in tests of psychomotor performance and learning and memory in a cross-sectional study. Specifically, age was associated with reduced runwheel activity measured over a one-week testing interval and with increased errors in the acquisition of a shock-motivated 14-unit T-maze without effects on other aspects of maze performance (i.e., runtime, shock frequency and duration). No differences were observed in spontaneous 24-h locomotor activity, the distance traveled in a 10-min square open field test, the number of falls from a rotating rod, or in the number of successful avoidances in one-way active avoidance in a straight runway. The high incidence and number of seizures observed at all age levels, however, complicates the interpretation of the results obtained in the present investigation. Moreover, the inability to obtain adequate numbers of gerbils beyond 26 months of age makes the interpretation of nonsignificant age correlations difficult. Pilot data with several additional tests contraindicated their usage as part of the battery, or required adaptation. Pilot testing of six young gerbils with an inclined screen resulted in the
A
10
,"
*
15
I
I
.
I
20
25
o
.
•
~
I
I
",,
I
,°
÷
30
I
°
I
÷
35
I
I
40
I
I
45
I
""
I
I
50
o
I
,
55
n-
90
£
t,l..
o
IJ_
E3
U ¢-.
0
5--
10--
15--
20--
25--
30--
35
D
C
0
2000 --
4000 --
6000 --
8000 --
10000 - -
10
°
15
I
I
°
I
I
20
°*
°
25
I
I
°
30
I
I
35
I
I
f
40
I
I
J
J
45
I
I
50
I
I
55
FIG. 3. Scatterplots depicting individual gerbil performance on psychomotor tests, and the regression of age on these variables. Displayed are (A) distance traveled in a 24-h test of locomotor activity; (B) mean daily wheel revolutions during one week in an activity wheel; (C) distance traveled during 10 min in a square open field; and (D) number of falls from a rotating rod during a 3-min interval.
B
0
2-
_~
o
3--
1 -
O
t~
8_
t~
o
I--
2-
3-
<
O
4--
?,
o
5--
12000
14 12 10 8uJ
6-4-
1:
2-
120
100-
80eo(1
40-
20 0 80 70 60-
~
5o 40
Q ~
30.
2010 0 5
t ~'
3
.
I 10
20
30
40
Age (mo)
FIG. 4. Scatterplots displaying individual gerbil performance on dependent measures of 14-unit T-maze performance and the regression of age on these variables.
7 ]6
E,L, S P A N G L E R er a/.
gerbils immediately jumping off when placed on the screen. Likewise, when placed on a smaller version of the rotorod, used for testing mice in our laboratory, the gerbils jumped as soon as they were placed on the rod. The inclined screen was dropped from the battery, and rotorod testing was completed but with a larger and more elevated version of the rotorod used to test rats in our laboratory. In some young gerbils this tendency to jump was also observed with the larger rotorod. Pilot testing with a version of the Morris water maze used to test mice in our laboratory contraindicated usage of this test because all of the gerbils tested had to be rescued from the tank. Results of these pilot tests suggest that for the gerbil, a small desert rodent, swimming may not have sufficient "survival value" compared to running, digging, or burrowing. On the other hand, the high level of spontaneous activity and excellent visual acuity in the gerbil (Wilkinson, 1984) may make it too suited to escape by jumping for some standard laboratory tests (e.g., inclined screen) to be useful for assessment. In agreement with Cheal (1987a), we observed that handling reduced the number of subjects experiencing seizures. In our study, the percent of subjects experiencing seizures during weekly body weight measurements declined from 21% to 5% from the first to the third weighing while Cheal (1987a) had reported a decrease from 3 0 - 4 0 % on the first day of handling to 2% by the fifth day. Unlike Cheal (1995), the proportion of gerbils in our study experiencing seizures did not increase with age. In the Cheal study, the gerbils were followed longitudinally and were separated into groups based upon either a control condition in which the animals were exposed to a small indoor environment for 1 h a month or an environmental enrichment experience consisting of 1 h a month in an outdoor arena from 1 to 28 months of age. In that study, the environmentally enriched gerbils had a lower incidence of seizures than controls from 4 to 31 months of age. However, after 31 months of age the incidence of seizure activity was high in both groups. In our cross-sectional study, neither the proportion of gerbils experiencing at least one seizure nor the total number of seizures observed was correlated with age. The inability to monitor seizure activity in a number of tasks (square open field, runwheel activity, locomotor activity) and the low number of aged gerbils greater than 26 months of age may have influenced the results in our study. The number of gerbils dropped from the rotorod, straight runway and 14-unit T-maze testing also raises serious questions regarding the utility of these types of test for studies of behavioral aging in the gerbil. Adaptation of the gerbil to these apparati may reduce the incidence of seizures and, thus, the loss of subjects, but may also further complicate the interpretation of results. That is, learning may occur with exposure to the rotorod. Additional adaptation in shock avoidance in the straight runway by increasing the number of days and trials to criterion may increase the number of subjects available for testing in the 14-unit T-maze without confounding the interpretation of results. Thus, this phenomena will require further evaluation to determine the effects of a variety of influences including the amount and type of handling, stress of the tests utilized, and the overall seizure activity of the strain utilized. Increasing the number and variety of animal models available for aging research has been a goal of gerontologists and the National Institute on Aging (Hazzard et al., 1991). Based on the results observed in this study, we conclude that the gerbil: a) is an appropriate model for epilepsy, as suggested by Cheal (1986); and b) might be a suitable model for age-related learning and memory deficits, in agreement with tests of learning and memory tests in a radial arm maze (Carney et al., 1991; Stanley et al., 1993) if care is taken to reduce the incidence of seizures. Moreover, the gerbil would be an appropriate model because it remains active and inquisitive through old age (Cheal, 1986), does not gain significant body weight over its lifespan, and appears relatively coordinated (rotorod) and without physical impairment through-
BEHAVIORALAGINGIN GERBILS
717
out its lifespan. However, while some of the traditional age-related confounds (decline in activity levels, increased body fat, less agile, and mobile) are lessened in this model, the issue of seizure activity needs to be controlled and/or addressed more clearly (e.g., adaptation to experimenter, apparatus, and possibly the task). Additionally, the research of others suggests the gerbil may make an excellent model of ischemia, due to the incomplete Circle of Willis, and for treatment of this condition (e.g., free radical scavengers) (Phillis and Clough-Helfman, 1990; Carney et al., 1991). Rederiving gerbil colonies may serve to reduce the incidence of seizure activity in individual colonies and, thus, may make the use of this animal as a model of behavioral aging, particularly learning and memory, more feasible. In summary, based on our previous findings in a variety of laboratory strains of rats and mice, we anticipated that the gerbil would further confirm the age-related differences observed in the battery of psychomotor and learning and memory tests. Our findings only partially confirmed our hypothesis. Failure to observe differences in a number of psychomotor tests and in the runtime and shock measures in the 14-unit T-maze may be attributable to the increased exploratory activity and inquisitiveness in the gerbil (Cheal, 1995). Future studies of aging in the gerbil should utilize a greater age distribution with more animals at or beyond the median life span ( > 3 6 months; Cheal, 1995). Acknowledgments--The Gerontology Research Center, NIH, is fully accredited by the American Association for Accreditation of Laboratory Animal Care. We gratefully acknowledge the donation of the gerbils by Cortex Pharmaceuticals, and the editorial assistance of Gloria Dunnigan.
REFERENCES CARNEY, J.M., STARKE-REED, P.E., OLIVER, C.N., LANDUM, R.W., CHENG, M.S., and WU, J.F. Reversal of age-related increase in brain protein oxidation, decrease in enzyme activity, and loss in temporal and spatial memory by chronic administration of the spin-trapping compound N-tert-butyl-a-phenylnitrone. Proc. Natl. Acad. Sci. USA, 88, 3633-3636, 1991. CHEAL, M.L. The gerbil: A unique model for research on aging. Exp. Aging Res. 13, 3-21, 1986. CHEAL, M.L. Adult development: Plasticity of stable behavior. Exp. Aging Res. 13, 29-36, 1987a. CHEAL, M.L. Lifespan environmental influences on species typical behavior of Meriones unguiculatus. In: Evolution of Longevity in Animals, Woodhead, A.D. and Thompson, K.H. (Editors), pp. 145-159, Plenum, New York, 1987b. CHEAL, M. Multiple factors modulating courtship, habituation, attention, and other motivated and motivating behaviors. In: Biological Perspectives on Motivated and Cognitive Activities. Wong, R. (Editor), Ablex Press. Norwood, NJ, 1995. HAZARD, D.G., WARNER, H.R., and FINCH, C.E. National Institute on Aging, NIH workshop on alternative animal models for research on aging. Exp. GerontoL 26, 411-439, 1991. PHILLIS, J.W. and CLOUGH-HELFMAN, C. Protection from cerebral ischemic injury in gerbils with the spin trap agent N-tert-butyl-a-phenylnitrone (PBN). Neurosci. Lett. 116, 315-319, 1990. SPANGLER, E.L., RIGBY, P., and INGRAM, D.K. Scopolamine impairs learning performance of rats in a 14-unit T-maze. Pharmacol. Biochem. Behav., 25, 673-679, 1986. SPANGLER, E.L., WAGGLE, K.S., HENGEMIHLE, J., ROBERTS, D., HESS, B., and INGRAM, D.K. Behavioral assessment of aging in male Fischer 344 and Brown Norway rat strains and their F~ hybrid. Neurobiol. Aging 15, 319-328, 1994. STANLEY, J.A., BULL, C.O., MOON, S.L., and ORTIZ, A.A. N-tert-Butyl-a-Phenyl-Nitrone (BPN), a free radical scavenger, facilitates learning by elderly gerbils following bilateral carotid occlusion (BCO). Soc. Neurosci. Abstr. 1653, 1993. THIESSEN, D. and YAHR, P. The Gerbil in Behavioral Investigations. University of Texas Press; Austin, TX, 1977. WILKINSON, F. The development of visual acuity in the Mongolian gerbil (Meriones unguiculatus). Behav. Brain Res. 83-94, 1984. ZHANG, J-R., ANDRUS, P.K., and HALL, E.D. Age-related regional changes in hydroxyl radical stress and antioxidant in gerbil brain. Soc. Neurosci. Abstr. 19: 458; 1993.