Prevention by Diazepam of Adverse Effects of Maternal Restraint Stress on Postnatal Development and Learning in the Rat S. M. BARLOW, A. F. KNIGHT AND F. M. SULLIVAN Department of Pharmacology, Guy's Hospital Medical School, London SEI 9RT. England
ABSTRACT Rats on days 12-14 of pregnancy were treated with restraint stress alone (9h daily), restraint stress plus diazepam (1 mg/kg, twice daily), diazepam alone, or left as untreated controls. Postnatal development and behaviour was assessed on a wide-ranging battery of tests. Offspring of mothers subjected to restraint stress alone were significantly retarded on a number of developmental measures including growth, ear-opening, cliff avoidance response, auditory startle response and mid-air righting reflex. When adult these offspring also showed significantly impaired learning ability in a swimming maze. However, the rate of development and learning ability in the restraint plus diazepam or diazepam alone groups was comparable to or slightly advanced of that in untreated controls. It is concluded that concurrent administration of a low dose of the tranquiliser diazepam during restraint stress prevents the adverse postnatal effects of maternal restraint stress. Tranquilisers such as diazepam are frequently prescribed during pregnancy. Whilst diazepam is thought to increase the risk of cleft palate in the offspring of man (Safra and Oakley, '75; Saxen and Saxen, '75; Aarskog, '75), it is not known whether postnatal development may also be affected by prenatal exposure to diazepam. Diazepam and its active metabolites readily cross the placenta (Idanpaan-Heikkila et al., '71; Erkkola et al., '74) and there is some evidence that longterm, low-dose diazepam therapy during pregnancy may cause pronounced muscular hypotonia in the offspring, the "floppy infant syndrome" (Gillberg, '77; Speight, '77). Animal experiments on postnatal development following prenatal diazepam exposure have produced conflicting evidence. Dietary administration of diazepam throughout pregnancy in the mouse resulted in significant decreases in postnatal survival and offspring body weight (Guerriero and Fox, '77). Surviving offspring however showed greater activity and enhanced learning ability on a Y-maze task when adult in comparison with controls, though the differences did not quite achieve statistically significant levels (Fox et al., '77). In the rat, diazepam administered orally TERAMLOGY (1979) 19: 105-110.
throughout pregnancy has been shown to cause delayed sexual maturation, increased spontaneous and open field activity, and impaired acquisition of a conditioned avoidance response (Lyubimov et al., '74). These studies raise the possibility that relatively low doses of diazepam may have significant effects on postnatal development of the offspring. Since, in the human, diazepam is generally given to alleviate stress and anxiety, it was decided t o investigate the effects of prenatal exposure to diazepam in stressed animals. We have previously shown that maternal restraint stress during pregnancy in the rat causes significant delays in postnatal growth and development of the offspring (Barlow et al., '78). In the present experiments therefore, the postnatal effects of restraint stress alone on days 12-14 of pregnancy, restraint stress with concurrent diazepam treatment and diazepam treatment alone were investigated to see whether the diazepam reduced or increased the adverse effects of the stress. MATERIALS A N D METHODS
Animals Female albino rats of the Porton-Wistarl Received Feb. 23, "78.Accepted Sept. 6, '78
105
106
S. M. BARLOW, A. F. KNIGHT AND F. M. SULLIVAN
weighed and sexed. Each litter was then culled randomly within sex groups to 4 male and 4 female pups wherever possible. All surviving members of each litter were assessed on a battery of postnatal tests for physical, reflex and motor development and a t 70-75 days of age all male offspring were each given a 2 min open field trial on 3 successive days. The details of the above postnatal screening programme have been described previously (Barlow e t al., '78). At 95- 100 days of age 2 males were selected a t random from each litter and tested using a water-maze. The water-filled maze was in the shape of an E, the 3 short arms each measuring 32 cm and the long arm 90 cm in length, each arm being 18 cm wide. Rats were placed Maternal treatment singly a t the start position a t the end of the On day 1 of pregnancy animals were central short arm and had to swim round to an weighed and assigned randomly to one of the 4 exit ladder placed a t t h e end of 1of the 2 outer treatment groups, stress only, diazepam only, arms. The exit ladder was not visible from the stress plus diazepam, or untreated controls. In choice point where the central short arm the 2 groups subjected to stress, animals were joined the long arm. On the first day of trials, immobilised by restraint for 9h from 09.00- the exit ladder was placed in the right-hand 18.00 h on each of days 12-14 of pregnancy, as arm and on the second day in the left-hand described previously (Barlow et al., '78). In the arm. Each rat was given 6 trials a day, with a 2 groups given diazepam treatment, the drug 30 min interval between trials. An error was was administered orally a t a dose of 1 mg/kg, scored if the rat swam to the left from the twice daily a t 09.00 h and a t 13.00 h on days choice point when the exit was on the right 12-14 of pregnancy. Diazepam was suspended and vice-versa. The criterion for successful in 1%carboxy-methyl cellulose a t a concentra- learning of the maze was swimming the last 3 tion of 0.5 mg/ml, such that animals were of each daily series of 6 trials without making dosed with 0.2 ml of suspension/lOO g body any errors. Errors and time taken to complete weight. All animals were weighed a t 09.00 h each trial were noted. and a t 18.00 h on each of days 12-14 of pregStatistical analysis nancy. At 18.00 h on day 14 of pregnancy, each rat was briefly anaesthetised with ether and a Significant sex differences were found only 0.5 ml blood sample taken from the retro- in body weight. For all other postnatal assessorbital sinus for plasma corticosterone assay ments where both sexes were tested results by the microfluorometric method of Glick et from male and female offspring have therefore been pooled. Measurements of offspring al., '64). After blood sampling, all animals were body weight (means of litter means comreturned to their home cages and left un- pared), maternal body weight and plasma cordisturbed until day 21 of pregnancy. On day 2 1 ticosterone levels have been analysed using they were weighed and placed singly in mater- Student's t-test, two-tailed. All other assessnity cages with nesting material provided. ments have been analysed on a per pup basis The animals were then left undisturbed ex- using the Fisher Exact probability test (two cept for brief, thrice daily observation for re- tailed) unless otherwise stated. All mean values are shown -+ S.E.M. cording of the time of parturition.
Ola strain (purchased from Olac Ltd., Oxford, England) weighing 185-240 g were used. They were caged in groups of 3 and kept under reversed lighting (white light 19.00-07.00 h, red light 07.00-19.00 h) throughout the premating period, pregnancy and lactation, in an animal room maintained a t 21-24°C. Three weeks after arrival the females were mated with albino males of the Wistar strain (purchased from Charles River Ltd., Kent, England). One male was introduced into each cage of females in the morning and vaginal smears taken in the evening when the males were removed. The day of finding spermatozoa in the smear was designated day 1 of pregnancy.
Postnatal assessment Litters were reared by their own mothers and all postnatal assessments carried out blind. On the day of birth, designated day 1 of age, litters were inspected, the numbers of live and stillborn pups noted, and each pup
RESULTS
Maternal plasma corticosterone. The mean plasma corticosterone levels a t the end of the treatment period on day 14 of pregnancy are shown in table 1. Both restraint and restraint plus diazepam groups had significantly ele-
107
DIAZEPAM, STRESS AND POSTNATAL DEVELOPMENT TABLE 1
Maternal plasma corticosterone levels at the end of the third day of treatment (day 14) and body weight changes during pregnancy Treatment group
Control Diazepam Restraint Restraint + diazepam I
No. of animals
Plasma corticosterone fpg1100 ml)
Day 1
Day 12
Day 14
Day21
5 4 6 8
14.92 2.6 20.12 1.6 56.2t3.3 5 9 . 7 2 3.4
222%6 22527 22325 22325
25727 26721 25724 25724
274*7 28225 22425’
321*6 34026 297?4‘ 307t63
’ ’
Body weight ( g )
22026’
p < 0.001compared with control or diazepam p < 0.02 compared with control or diazepam. p < 0.01 compared with diazepam.
‘
vated levels in comparison with either control Static righting reflex. At both 3 and 4 days or diazepam alone groups. of age, the proportion of pups completing the Maternal body weight. Maternal body reflex within 15 sec was lowest in the reweight changes during pregnancy are shown straint alone group and highest in the rein table 1. At the start of the 3-day treatment straint plus diazepam group, with diazepam period on day 12, there was no significant dif- alone and control groups intermediate. The ference between treatment groups. By the end differences between the first 2 groups a t 3 and of treatment on day 14 however, restraint and 4 days of age, however, were not quite signifirestraint plus diazepam had lost weight and cant (p = 0.06). were significantly lighter than control or Cliff auoidance. The numbers of pups comdiazepam alone groups. By day 21, the group pleting the cliff avoidance response within 15 subjected to restraint alone was still signifi- sec a t 7 days of age are shown in table 2. Sigcantly lighter than control or diazepan alone nificantly fewer offspring in the restraint groups, but the group given diazepam during alone group showed the response compared restraint had gained more weight than the re- with the restraint plus diazepam group. straint alone group after treatment and were Auditory startle response. The numbers of not significantly different from the controls. pups showing this response a t 14 days of age Litter size and postnatal mortality. There are shown in table 2. The diazepam alone and were no significant differences among treat- restraint plus diazepam groups were signifiment groups in length of gestation (Mann- cantly advanced in comparison with controls. Whitney “U” test), numbers of live and still- The response was present in all pups by 17 born pups delivered, or postnatal mortality days in diazepam alone, by 19 days in restraint (Chi-square test). plus diazepam and by 20 days in controls and Body weight of offspring. The body weights restraint alone groups. of the male offspring are shown in figure 1. At Mid-air righting. The numbers of pups sucall ages, offspring in the group subjected to re- cessfully righting themselves on 3 successive straint alone were the lightest of the 4 treat- trials a t 20 days of age are shown in table 2. ment groups and the diazepam prevented the The restraint alone group was significantly postnatal weight reduction produced by the retarded in comparison with diazepam alone. stress. The body weights of female offspring The response was present in all pups by 20 (details not shown) followed an identical pat- days of age in diazepam alone, by 21 days in tern to that of the male offspring, the re- control and restraint plus diazepam and by 23 straint plus diazepam group being significant- days in restraint alone groups. ly heavier than the restraint along group a t 7 Rotarod. The numbers of pups completing (p < 0.025) and 21 (p < 0.05) days of age. the 1 min trial without any falls on the third Opening of the ears. Ear opening was signif- day of testing are shown in table 2. The icantly retarded in the restraint alone group offspring in the restraint plus diazepam group a t 4 days of age compared with any other made more falls from the rotarod than offgroup (table 2). By 5 days of age ears were spring in any other treatment group, the difopen in all pups except one from the restraint ference between restraint plus diazepam and alone group in which opening was complete by diazepam alone groups being significant on all 6 days of age. 3 days of testing (p < 0.05).
108
S. M. BARLOW, A. F. KNIGHT AND F. M. SULLIVAN
13
c -
170
I T
Kq(
45
Control
160
40
@ Diazepam 150
W Restraint
35 140
0
0
DAY 7
DAY 1
ON 42
DAY 21
* ~ ( 0 . 0 5compsed
€3 Restraint t Diazepam
with restraint + diazepam
Fig. 1 Body weight of male offspring at birth, 1, 3 and 6 weeks of age. Columns represent mean of litter means and vertical bars S.E.M.
TABLE 2
Sieificant effects on postnatal development and learning ability Treatment group ~~
~
Auditory startle (14 days)
(20days)
Mid-air righting
Rotarod (24 days)
Maze learning
30137 26/31 19/39'
24/37 25/31 21/35
12/33 22/39 19/33
27/32 28/28 26/33
21/30 23/28 18/29
7/15 4/13 1/15'
45/47
39/47
38/46'
43/45
20145 '
7/15
(95-100 days)
~
Control Diazepam Restraint Restraint + diazepam ~~~~~
Cliff avoidance (7 days)
Ears open (4 days)
~
' p < 0.01 compared with any other group.
p < 3p < 'p < p < p < 'p
ABSTRACT Rats on days 12-14 of pregnancy were treated with restraint stress alone (9h daily), restraint stress plus diazepam (1 mg/kg, twice daily), diazepam alone, or left as untreated controls. Postnatal development and behaviour was assessed on a wide-ranging battery of tests. Offspring of mothers subjected to restraint stress alone were significantly retarded on a number of developmental measures including growth, ear-opening, cliff avoidance response, auditory startle response and mid-air righting reflex. When adult these offspring also showed significantly impaired learning ability in a swimming maze. However, the rate of development and learning ability in the restraint plus diazepam or diazepam alone groups was comparable to or slightly advanced of that in untreated controls. It is concluded that concurrent administration of a low dose of the tranquiliser diazepam during restraint stress prevents the adverse postnatal effects of maternal restraint stress. Tranquilisers such as diazepam are frequently prescribed during pregnancy. Whilst diazepam is thought to increase the risk of cleft palate in the offspring of man (Safra and Oakley, '75; Saxen and Saxen, '75; Aarskog, '75), it is not known whether postnatal development may also be affected by prenatal exposure to diazepam. Diazepam and its active metabolites readily cross the placenta (Idanpaan-Heikkila et al., '71; Erkkola et al., '74) and there is some evidence that longterm, low-dose diazepam therapy during pregnancy may cause pronounced muscular hypotonia in the offspring, the "floppy infant syndrome" (Gillberg, '77; Speight, '77). Animal experiments on postnatal development following prenatal diazepam exposure have produced conflicting evidence. Dietary administration of diazepam throughout pregnancy in the mouse resulted in significant decreases in postnatal survival and offspring body weight (Guerriero and Fox, '77). Surviving offspring however showed greater activity and enhanced learning ability on a Y-maze task when adult in comparison with controls, though the differences did not quite achieve statistically significant levels (Fox et al., '77). In the rat, diazepam administered orally TERAMLOGY (1979) 19: 105-110.
throughout pregnancy has been shown to cause delayed sexual maturation, increased spontaneous and open field activity, and impaired acquisition of a conditioned avoidance response (Lyubimov et al., '74). These studies raise the possibility that relatively low doses of diazepam may have significant effects on postnatal development of the offspring. Since, in the human, diazepam is generally given to alleviate stress and anxiety, it was decided t o investigate the effects of prenatal exposure to diazepam in stressed animals. We have previously shown that maternal restraint stress during pregnancy in the rat causes significant delays in postnatal growth and development of the offspring (Barlow et al., '78). In the present experiments therefore, the postnatal effects of restraint stress alone on days 12-14 of pregnancy, restraint stress with concurrent diazepam treatment and diazepam treatment alone were investigated to see whether the diazepam reduced or increased the adverse effects of the stress. MATERIALS A N D METHODS
Animals Female albino rats of the Porton-Wistarl Received Feb. 23, "78.Accepted Sept. 6, '78
105
106
S. M. BARLOW, A. F. KNIGHT AND F. M. SULLIVAN
weighed and sexed. Each litter was then culled randomly within sex groups to 4 male and 4 female pups wherever possible. All surviving members of each litter were assessed on a battery of postnatal tests for physical, reflex and motor development and a t 70-75 days of age all male offspring were each given a 2 min open field trial on 3 successive days. The details of the above postnatal screening programme have been described previously (Barlow e t al., '78). At 95- 100 days of age 2 males were selected a t random from each litter and tested using a water-maze. The water-filled maze was in the shape of an E, the 3 short arms each measuring 32 cm and the long arm 90 cm in length, each arm being 18 cm wide. Rats were placed Maternal treatment singly a t the start position a t the end of the On day 1 of pregnancy animals were central short arm and had to swim round to an weighed and assigned randomly to one of the 4 exit ladder placed a t t h e end of 1of the 2 outer treatment groups, stress only, diazepam only, arms. The exit ladder was not visible from the stress plus diazepam, or untreated controls. In choice point where the central short arm the 2 groups subjected to stress, animals were joined the long arm. On the first day of trials, immobilised by restraint for 9h from 09.00- the exit ladder was placed in the right-hand 18.00 h on each of days 12-14 of pregnancy, as arm and on the second day in the left-hand described previously (Barlow et al., '78). In the arm. Each rat was given 6 trials a day, with a 2 groups given diazepam treatment, the drug 30 min interval between trials. An error was was administered orally a t a dose of 1 mg/kg, scored if the rat swam to the left from the twice daily a t 09.00 h and a t 13.00 h on days choice point when the exit was on the right 12-14 of pregnancy. Diazepam was suspended and vice-versa. The criterion for successful in 1%carboxy-methyl cellulose a t a concentra- learning of the maze was swimming the last 3 tion of 0.5 mg/ml, such that animals were of each daily series of 6 trials without making dosed with 0.2 ml of suspension/lOO g body any errors. Errors and time taken to complete weight. All animals were weighed a t 09.00 h each trial were noted. and a t 18.00 h on each of days 12-14 of pregStatistical analysis nancy. At 18.00 h on day 14 of pregnancy, each rat was briefly anaesthetised with ether and a Significant sex differences were found only 0.5 ml blood sample taken from the retro- in body weight. For all other postnatal assessorbital sinus for plasma corticosterone assay ments where both sexes were tested results by the microfluorometric method of Glick et from male and female offspring have therefore been pooled. Measurements of offspring al., '64). After blood sampling, all animals were body weight (means of litter means comreturned to their home cages and left un- pared), maternal body weight and plasma cordisturbed until day 21 of pregnancy. On day 2 1 ticosterone levels have been analysed using they were weighed and placed singly in mater- Student's t-test, two-tailed. All other assessnity cages with nesting material provided. ments have been analysed on a per pup basis The animals were then left undisturbed ex- using the Fisher Exact probability test (two cept for brief, thrice daily observation for re- tailed) unless otherwise stated. All mean values are shown -+ S.E.M. cording of the time of parturition.
Ola strain (purchased from Olac Ltd., Oxford, England) weighing 185-240 g were used. They were caged in groups of 3 and kept under reversed lighting (white light 19.00-07.00 h, red light 07.00-19.00 h) throughout the premating period, pregnancy and lactation, in an animal room maintained a t 21-24°C. Three weeks after arrival the females were mated with albino males of the Wistar strain (purchased from Charles River Ltd., Kent, England). One male was introduced into each cage of females in the morning and vaginal smears taken in the evening when the males were removed. The day of finding spermatozoa in the smear was designated day 1 of pregnancy.
Postnatal assessment Litters were reared by their own mothers and all postnatal assessments carried out blind. On the day of birth, designated day 1 of age, litters were inspected, the numbers of live and stillborn pups noted, and each pup
RESULTS
Maternal plasma corticosterone. The mean plasma corticosterone levels a t the end of the treatment period on day 14 of pregnancy are shown in table 1. Both restraint and restraint plus diazepam groups had significantly ele-
107
DIAZEPAM, STRESS AND POSTNATAL DEVELOPMENT TABLE 1
Maternal plasma corticosterone levels at the end of the third day of treatment (day 14) and body weight changes during pregnancy Treatment group
Control Diazepam Restraint Restraint + diazepam I
No. of animals
Plasma corticosterone fpg1100 ml)
Day 1
Day 12
Day 14
Day21
5 4 6 8
14.92 2.6 20.12 1.6 56.2t3.3 5 9 . 7 2 3.4
222%6 22527 22325 22325
25727 26721 25724 25724
274*7 28225 22425’
321*6 34026 297?4‘ 307t63
’ ’
Body weight ( g )
22026’
p < 0.001compared with control or diazepam p < 0.02 compared with control or diazepam. p < 0.01 compared with diazepam.
‘
vated levels in comparison with either control Static righting reflex. At both 3 and 4 days or diazepam alone groups. of age, the proportion of pups completing the Maternal body weight. Maternal body reflex within 15 sec was lowest in the reweight changes during pregnancy are shown straint alone group and highest in the rein table 1. At the start of the 3-day treatment straint plus diazepam group, with diazepam period on day 12, there was no significant dif- alone and control groups intermediate. The ference between treatment groups. By the end differences between the first 2 groups a t 3 and of treatment on day 14 however, restraint and 4 days of age, however, were not quite signifirestraint plus diazepam had lost weight and cant (p = 0.06). were significantly lighter than control or Cliff auoidance. The numbers of pups comdiazepam alone groups. By day 21, the group pleting the cliff avoidance response within 15 subjected to restraint alone was still signifi- sec a t 7 days of age are shown in table 2. Sigcantly lighter than control or diazepan alone nificantly fewer offspring in the restraint groups, but the group given diazepam during alone group showed the response compared restraint had gained more weight than the re- with the restraint plus diazepam group. straint alone group after treatment and were Auditory startle response. The numbers of not significantly different from the controls. pups showing this response a t 14 days of age Litter size and postnatal mortality. There are shown in table 2. The diazepam alone and were no significant differences among treat- restraint plus diazepam groups were signifiment groups in length of gestation (Mann- cantly advanced in comparison with controls. Whitney “U” test), numbers of live and still- The response was present in all pups by 17 born pups delivered, or postnatal mortality days in diazepam alone, by 19 days in restraint (Chi-square test). plus diazepam and by 20 days in controls and Body weight of offspring. The body weights restraint alone groups. of the male offspring are shown in figure 1. At Mid-air righting. The numbers of pups sucall ages, offspring in the group subjected to re- cessfully righting themselves on 3 successive straint alone were the lightest of the 4 treat- trials a t 20 days of age are shown in table 2. ment groups and the diazepam prevented the The restraint alone group was significantly postnatal weight reduction produced by the retarded in comparison with diazepam alone. stress. The body weights of female offspring The response was present in all pups by 20 (details not shown) followed an identical pat- days of age in diazepam alone, by 21 days in tern to that of the male offspring, the re- control and restraint plus diazepam and by 23 straint plus diazepam group being significant- days in restraint alone groups. ly heavier than the restraint along group a t 7 Rotarod. The numbers of pups completing (p < 0.025) and 21 (p < 0.05) days of age. the 1 min trial without any falls on the third Opening of the ears. Ear opening was signif- day of testing are shown in table 2. The icantly retarded in the restraint alone group offspring in the restraint plus diazepam group a t 4 days of age compared with any other made more falls from the rotarod than offgroup (table 2). By 5 days of age ears were spring in any other treatment group, the difopen in all pups except one from the restraint ference between restraint plus diazepam and alone group in which opening was complete by diazepam alone groups being significant on all 6 days of age. 3 days of testing (p < 0.05).
108
S. M. BARLOW, A. F. KNIGHT AND F. M. SULLIVAN
13
c -
170
I T
Kq(
45
Control
160
40
@ Diazepam 150
W Restraint
35 140
0
0
DAY 7
DAY 1
ON 42
DAY 21
* ~ ( 0 . 0 5compsed
€3 Restraint t Diazepam
with restraint + diazepam
Fig. 1 Body weight of male offspring at birth, 1, 3 and 6 weeks of age. Columns represent mean of litter means and vertical bars S.E.M.
TABLE 2
Sieificant effects on postnatal development and learning ability Treatment group ~~
~
Auditory startle (14 days)
(20days)
Mid-air righting
Rotarod (24 days)
Maze learning
30137 26/31 19/39'
24/37 25/31 21/35
12/33 22/39 19/33
27/32 28/28 26/33
21/30 23/28 18/29
7/15 4/13 1/15'
45/47
39/47
38/46'
43/45
20145 '
7/15
(95-100 days)
~
Control Diazepam Restraint Restraint + diazepam ~~~~~
Cliff avoidance (7 days)
Ears open (4 days)
~
' p < 0.01 compared with any other group.
p < 3p < 'p < p < p < 'p
