Physiology & Behavior, Vol. 48, pp. 429--433. Pergamon Press plc, 1990. Printed in the U.S.A.
0031-9384/90 $3.00 + .00
Behavior of Streptozotocin-Diabetic Mice in Tests of Exploration, Locomotion, Anxiety, Depression and Aggression L. A. H I L A K I V I - C L A R K E , K. M. W O Z N I A K , M. J. D U R C A N A N D M. L I N N O I L A
Laboratory o f Clinical Studies, DICBR, National Institute on Alcohol Abuse and Alcoholism Room 3C102, Bethesda, M D 20892 R e c e i v e d 10 April 1990
HILAKIVI-CLARKE, L. A., K. M. WOZNIAK, M. J. DURCAN AND M. LINNOILA. Behavior of streptozotocin-diabetic mice in tests of exploration, locomotion, anxiety, depression and aggression. PHYSIOL BEHAV 48(3) 429-433, 1990.--The present study examined behavior of streptozotocin-diabetic mice in Porsolt's swim test, a putative animal model of depression, in the holeboard test of exploration and locomotor activity, in the plus maze test of anxiety, and in the resident-intruder paradigm of aggression. Two weeks after an IP injection of 200 mg/kg streptozotocin, which caused a 20% weight loss and increased fluid consumption and urination, male NIH Swiss mice were found to show lengthened duration of immobility in the swim test. One week of insulin treatment (0.1 IU/g/day) partially antagonized this change. The locomotor activity scores in the streptozotocin-treated mice were lower in the holeboard but higher in the plus maze than in the controls; therefore, the lengthened immobility was not likely to be due to a general motor impairment, No significant changes in the time spent in social interaction or aggressive behavior were found in the streptozotocintreated mice. The results indicate that streptozotocin-treated mice show lengthened immobility in the swim test. Streptozotocin
Diabetes
Porsolt's swim test
Holeboard
DIABETES is associated with alterations in central monoaminergic systems (5, 15, 25), changes which are also linked to depression (9, 13, 26). Therefore, a functional link between the proposed connection with diabetes and depression (14,23) may exist at the level of the brain monoamine systems. Few experiments have been conducted in animal models to study relationships between diabetes and depressive behavior. An often-used model to study diabetes in rodents takes advantage of the direct toxic effects produced by streptozotocin on the beta cells of the pancreatic islets. It has been reported that streptozotocindiabetic rats did not show impaired operant escape responding (helpless behavior) (16), but they showed enhanced reactivity to an exposure to a cold environment or hypobaric hypoxia (3). In the present study, behavior of streptozotocin-treated mice was examined in Porsolt's swim test. This test has been developed to predict efficacy of antidepressant drugs (22). In Porsolt's swim test an animal is put into a cylinder containing water, and the time it spends floating motionless is used as an index of depressive behavior. Antidepressants generally shorten the period of immobility in the water (6,22). Although Porsolt's swim test paradigm probably does not induce "depression," it has been shown to serve as a probe of an animal's behavioral state before the test (8, 10, 27). To investigate specificity of the behavioral changes caused by streptozotocin, animals were also tested in a holeboard test of exploration and locomotor activity (7), the plus maze test of anxiety (13,21), and the resident-intruder paradigm of aggression (20). A reduction in activity in all the tests was expected to be
Plus maze
Resident-intruder test
Mice
seen, if immobility in the swim test was reflective of the animals physical health, rather than a functional relationship between "depressive' '-type behavior and diabetes. The behavioral changes seen in the swim test behavior of streptozotocin-diabetic mice, following insulin treatment, were also investigated. METHOD
Animals Male NIH Swiss mice, weighing approximately 22 g at the beginning of the experiments, were housed in groups of 10 per cage and maintained on a 12-hr light:12-hr dark cycle (lights on 6:00 a.m.), room temperature +22-24°C, and allowed ad lib access to food and water.
Induction of Diabetes Diabetes was experimentally induced, as described by Massol
et al. (17), by a single intraperitoneal injection of 200 mg/kg streptozotocin (STZ) (Sigma Chemicals, St. Louis, MO). Streptozotocin was dissolved in 0.05 M citrate buffer and immediately given in a volume of 10 ml/kg body weight.
Blood Glucose Determinations Blood glucose determinations were made by drawing blood from the tail tip of individual mice. A drop of blood was placed directly onto a Glucostik reagent strip where it underwent a timed
429
430
reaction following which glucose measurements were made (mg/ 100 ml) using a Glucose II reflectance photometer (Ames Division, Miles Laboratories Inc., Yellow Springs, OH). The effect of insulin on blood glucose levels was also measured. In this experiment, the streptozotocin and control mice (number of animals per group = 11-14), 19 days after the streptozotocin injection, were given 0 or 0.1 IU/g insulin purified from bovine pancreas (Sigma, St. Louis, MO) 60 rain before the determination of blood glucose levels. A few drops of 0.05 M citric acid were added to the solutions. The injection volume was 10 ml/kg body weight.
HILAKIVI-CLARKE E T AL.
diameter 21 cm) containing 8 cm of water maintained at about 25°C for 10 min. The 10-min period included a 2-min acclimation period at the beginning of the test, immediately followed by an 8-min test. Half of the animals were observed during testing, and the behavior of the other half was recorded using a camera and a videocassette recorder. A mouse was judged to be immobile when it was floating making only those movements necessary to keep its head above the water. The time spent immobile was scored using a keyboard linked to PDP-microcomputer running SKED-11 software or by a stopwatch. Effect o f Insulin Treatment on Swim Test Behavior
Behavioral Tests
The behavioral tests were carried out in a dimly lit room which was maintained at about 25°C. The tests were done between 11:00 a.m. and 5:00 p.m. Holeboard test. Ten days after the streptozotocin injection (day 10), 19 streptozotocin mice and 19 control mice were studied in the holeboard. The holeboard apparatus was made of Plexiglas (40 x 40 × 30 cm) and had four holes 3 cm in diameter equally spaced in the floor. Infrared photocells in the walls of the box and directly beneath each hole provided automated measures of locomotor activity (number of beam interruptions), the number of exploratory head-dips made and the duration of head-dipping. The holeboard testing involved placing a mouse in the center of the floor and allowing it to explore for 5 min. Plus maze. The plus maze was made of Plexiglas and consisted of two open arms 30 x 5 cm and two closed arms 30 × 5 cm with 14.5-cm high side walls. The arms extended from a central platform, and the floor of the arms was made of black Plexiglas. The apparatus was mounted on a Plexiglas base, raising it 38.5 cm above the floor. Immediately after the end of the holeboard test, each mouse was placed in the center of the plus maze facing an open arm. During the 5-rain test the number of entries into each type of arm and the time spent in each arm were scored using a keyboard interfaced with a PDP-11 microcomputer running SKED-11 software (State Systems, Kalamazoo, MI). A mouse was taken to have entered an arm when its four legs were on the arm. In addition to the total number of arm entries, the number of entries into the open arms expressed as a percentage of total arm entries and the time spent on the open arms expressed as a percentage of the time spent on both the open and closed arms were measured. These two measures are used as indices of anxiety. Resident-intruder test. On day 14, 13 streptozotocin-treated and 17 control mice were housed individually for 7 days. They were then confronted in their home cage with a group-housed male intruder (no previous treatment). The body weights of most of the intruders were matched with that of the resident; however, some of lightest resident streptozotocin-mice were mixed with a slightly heavier intruder and therefore also some controls were mixed with a heavier intruder. During the 7.5-min test period, an observer monitored the behavior of the resident with a keyboard interfaced with a PDP-microcomputer. The behaviors recorded were the number and duration of social investigation (sniffing, following, grooming) and aggression (lateral threat, tail rattle, biting, fighting). Swim test. Thirteen days after the streptozotocin administration, 12 streptozotocin-treated and 12 control mice, not tested previously in the holeboard or plus maze, were studied in Porsolt's swim test. Each mouse was placed in a plastic cylinder (height 17 cm,
Starting one week after the streptozotocin injection, 14 diabetic mice were treated IP daily (at 4:00 p.m.), for 7 days, with 0.1 IU/g/day insulin. Insulin was dissolved in a slightly acid citrate buffer (pH 6), and the injection volume was 5 ml/kg body weight. Simultaneously with the insulin treatment, fifteen streptozotocin and 15 control mice received daily citrate buffer injections. On day 8 of the insulin injections, the mice were given insulin or saline at 8:00 a.m., and six hours later the swim test was carried out, as described above. Statistics
The body weights and the data obtained in the behavioral tests were analysed using one- and two-way analysis of variance. Between-group comparisons were made using Fisher's least significant difference test. Since the data for the time spent in aggressive behavior and blood glucose levels were not normally distributed, they were analysed using the Kruskal-Wallis test. Between-group comparisons were made using Dunn's procedure. Correlations between the data of the swim test and body weights were analysed using the Spearman rank correlation coefficient. RESULTS The presence of diabetes was confirmed when at least 48 hr from the streptozotocin treatment had passed, as described in the Method section. More than 95% of the streptozotocin-treated animals had blood glucose levels higher than 250 mg/dl (this value was used as a criterion of diabetes), whereas the mean ( -+ SEM) blood glucose level in the control animals was 159.3+2.9 mg/dl. The upper limit for blood glucose concentration possible to detect by the Glucose II reflectance photometer was 400; 60% of the streptozotocin animals had values higher than that. It was noted that the streptozotocin-treated mice consumed more water than the controls, and also urinated more. No systematic measurements, however, were carried out. Weight Gain
At the beginning of the experiments the control mice (mean-+ SEM: 21.9-+0.1 g, n = 137) and the mice which were to receive streptozotocin (22.0-+0.1 g, n = 129) had similar body weights. Ten to 13 days after the streptozotocin or saline injection, the controls weighed 24.1 -+O. 1 g and the streptozotocin-treated mice 20.5 -+ 0.2 g. Thus, a statistically significant weight x time interaction occurred between the groups, F(1,516)_+164.8, p
0031-9384/90 $3.00 + .00
Behavior of Streptozotocin-Diabetic Mice in Tests of Exploration, Locomotion, Anxiety, Depression and Aggression L. A. H I L A K I V I - C L A R K E , K. M. W O Z N I A K , M. J. D U R C A N A N D M. L I N N O I L A
Laboratory o f Clinical Studies, DICBR, National Institute on Alcohol Abuse and Alcoholism Room 3C102, Bethesda, M D 20892 R e c e i v e d 10 April 1990
HILAKIVI-CLARKE, L. A., K. M. WOZNIAK, M. J. DURCAN AND M. LINNOILA. Behavior of streptozotocin-diabetic mice in tests of exploration, locomotion, anxiety, depression and aggression. PHYSIOL BEHAV 48(3) 429-433, 1990.--The present study examined behavior of streptozotocin-diabetic mice in Porsolt's swim test, a putative animal model of depression, in the holeboard test of exploration and locomotor activity, in the plus maze test of anxiety, and in the resident-intruder paradigm of aggression. Two weeks after an IP injection of 200 mg/kg streptozotocin, which caused a 20% weight loss and increased fluid consumption and urination, male NIH Swiss mice were found to show lengthened duration of immobility in the swim test. One week of insulin treatment (0.1 IU/g/day) partially antagonized this change. The locomotor activity scores in the streptozotocin-treated mice were lower in the holeboard but higher in the plus maze than in the controls; therefore, the lengthened immobility was not likely to be due to a general motor impairment, No significant changes in the time spent in social interaction or aggressive behavior were found in the streptozotocintreated mice. The results indicate that streptozotocin-treated mice show lengthened immobility in the swim test. Streptozotocin
Diabetes
Porsolt's swim test
Holeboard
DIABETES is associated with alterations in central monoaminergic systems (5, 15, 25), changes which are also linked to depression (9, 13, 26). Therefore, a functional link between the proposed connection with diabetes and depression (14,23) may exist at the level of the brain monoamine systems. Few experiments have been conducted in animal models to study relationships between diabetes and depressive behavior. An often-used model to study diabetes in rodents takes advantage of the direct toxic effects produced by streptozotocin on the beta cells of the pancreatic islets. It has been reported that streptozotocindiabetic rats did not show impaired operant escape responding (helpless behavior) (16), but they showed enhanced reactivity to an exposure to a cold environment or hypobaric hypoxia (3). In the present study, behavior of streptozotocin-treated mice was examined in Porsolt's swim test. This test has been developed to predict efficacy of antidepressant drugs (22). In Porsolt's swim test an animal is put into a cylinder containing water, and the time it spends floating motionless is used as an index of depressive behavior. Antidepressants generally shorten the period of immobility in the water (6,22). Although Porsolt's swim test paradigm probably does not induce "depression," it has been shown to serve as a probe of an animal's behavioral state before the test (8, 10, 27). To investigate specificity of the behavioral changes caused by streptozotocin, animals were also tested in a holeboard test of exploration and locomotor activity (7), the plus maze test of anxiety (13,21), and the resident-intruder paradigm of aggression (20). A reduction in activity in all the tests was expected to be
Plus maze
Resident-intruder test
Mice
seen, if immobility in the swim test was reflective of the animals physical health, rather than a functional relationship between "depressive' '-type behavior and diabetes. The behavioral changes seen in the swim test behavior of streptozotocin-diabetic mice, following insulin treatment, were also investigated. METHOD
Animals Male NIH Swiss mice, weighing approximately 22 g at the beginning of the experiments, were housed in groups of 10 per cage and maintained on a 12-hr light:12-hr dark cycle (lights on 6:00 a.m.), room temperature +22-24°C, and allowed ad lib access to food and water.
Induction of Diabetes Diabetes was experimentally induced, as described by Massol
et al. (17), by a single intraperitoneal injection of 200 mg/kg streptozotocin (STZ) (Sigma Chemicals, St. Louis, MO). Streptozotocin was dissolved in 0.05 M citrate buffer and immediately given in a volume of 10 ml/kg body weight.
Blood Glucose Determinations Blood glucose determinations were made by drawing blood from the tail tip of individual mice. A drop of blood was placed directly onto a Glucostik reagent strip where it underwent a timed
429
430
reaction following which glucose measurements were made (mg/ 100 ml) using a Glucose II reflectance photometer (Ames Division, Miles Laboratories Inc., Yellow Springs, OH). The effect of insulin on blood glucose levels was also measured. In this experiment, the streptozotocin and control mice (number of animals per group = 11-14), 19 days after the streptozotocin injection, were given 0 or 0.1 IU/g insulin purified from bovine pancreas (Sigma, St. Louis, MO) 60 rain before the determination of blood glucose levels. A few drops of 0.05 M citric acid were added to the solutions. The injection volume was 10 ml/kg body weight.
HILAKIVI-CLARKE E T AL.
diameter 21 cm) containing 8 cm of water maintained at about 25°C for 10 min. The 10-min period included a 2-min acclimation period at the beginning of the test, immediately followed by an 8-min test. Half of the animals were observed during testing, and the behavior of the other half was recorded using a camera and a videocassette recorder. A mouse was judged to be immobile when it was floating making only those movements necessary to keep its head above the water. The time spent immobile was scored using a keyboard linked to PDP-microcomputer running SKED-11 software or by a stopwatch. Effect o f Insulin Treatment on Swim Test Behavior
Behavioral Tests
The behavioral tests were carried out in a dimly lit room which was maintained at about 25°C. The tests were done between 11:00 a.m. and 5:00 p.m. Holeboard test. Ten days after the streptozotocin injection (day 10), 19 streptozotocin mice and 19 control mice were studied in the holeboard. The holeboard apparatus was made of Plexiglas (40 x 40 × 30 cm) and had four holes 3 cm in diameter equally spaced in the floor. Infrared photocells in the walls of the box and directly beneath each hole provided automated measures of locomotor activity (number of beam interruptions), the number of exploratory head-dips made and the duration of head-dipping. The holeboard testing involved placing a mouse in the center of the floor and allowing it to explore for 5 min. Plus maze. The plus maze was made of Plexiglas and consisted of two open arms 30 x 5 cm and two closed arms 30 × 5 cm with 14.5-cm high side walls. The arms extended from a central platform, and the floor of the arms was made of black Plexiglas. The apparatus was mounted on a Plexiglas base, raising it 38.5 cm above the floor. Immediately after the end of the holeboard test, each mouse was placed in the center of the plus maze facing an open arm. During the 5-rain test the number of entries into each type of arm and the time spent in each arm were scored using a keyboard interfaced with a PDP-11 microcomputer running SKED-11 software (State Systems, Kalamazoo, MI). A mouse was taken to have entered an arm when its four legs were on the arm. In addition to the total number of arm entries, the number of entries into the open arms expressed as a percentage of total arm entries and the time spent on the open arms expressed as a percentage of the time spent on both the open and closed arms were measured. These two measures are used as indices of anxiety. Resident-intruder test. On day 14, 13 streptozotocin-treated and 17 control mice were housed individually for 7 days. They were then confronted in their home cage with a group-housed male intruder (no previous treatment). The body weights of most of the intruders were matched with that of the resident; however, some of lightest resident streptozotocin-mice were mixed with a slightly heavier intruder and therefore also some controls were mixed with a heavier intruder. During the 7.5-min test period, an observer monitored the behavior of the resident with a keyboard interfaced with a PDP-microcomputer. The behaviors recorded were the number and duration of social investigation (sniffing, following, grooming) and aggression (lateral threat, tail rattle, biting, fighting). Swim test. Thirteen days after the streptozotocin administration, 12 streptozotocin-treated and 12 control mice, not tested previously in the holeboard or plus maze, were studied in Porsolt's swim test. Each mouse was placed in a plastic cylinder (height 17 cm,
Starting one week after the streptozotocin injection, 14 diabetic mice were treated IP daily (at 4:00 p.m.), for 7 days, with 0.1 IU/g/day insulin. Insulin was dissolved in a slightly acid citrate buffer (pH 6), and the injection volume was 5 ml/kg body weight. Simultaneously with the insulin treatment, fifteen streptozotocin and 15 control mice received daily citrate buffer injections. On day 8 of the insulin injections, the mice were given insulin or saline at 8:00 a.m., and six hours later the swim test was carried out, as described above. Statistics
The body weights and the data obtained in the behavioral tests were analysed using one- and two-way analysis of variance. Between-group comparisons were made using Fisher's least significant difference test. Since the data for the time spent in aggressive behavior and blood glucose levels were not normally distributed, they were analysed using the Kruskal-Wallis test. Between-group comparisons were made using Dunn's procedure. Correlations between the data of the swim test and body weights were analysed using the Spearman rank correlation coefficient. RESULTS The presence of diabetes was confirmed when at least 48 hr from the streptozotocin treatment had passed, as described in the Method section. More than 95% of the streptozotocin-treated animals had blood glucose levels higher than 250 mg/dl (this value was used as a criterion of diabetes), whereas the mean ( -+ SEM) blood glucose level in the control animals was 159.3+2.9 mg/dl. The upper limit for blood glucose concentration possible to detect by the Glucose II reflectance photometer was 400; 60% of the streptozotocin animals had values higher than that. It was noted that the streptozotocin-treated mice consumed more water than the controls, and also urinated more. No systematic measurements, however, were carried out. Weight Gain
At the beginning of the experiments the control mice (mean-+ SEM: 21.9-+0.1 g, n = 137) and the mice which were to receive streptozotocin (22.0-+0.1 g, n = 129) had similar body weights. Ten to 13 days after the streptozotocin or saline injection, the controls weighed 24.1 -+O. 1 g and the streptozotocin-treated mice 20.5 -+ 0.2 g. Thus, a statistically significant weight x time interaction occurred between the groups, F(1,516)_+164.8, p
