PERINATAL DELTA-9-TETRAHYDROCANNABINOL EXPOSURE DISRUPTS SOCIAL AND OPEN FIELD BEHAVIOR IN ADULT MALE RATS

Abstract

Marijuana is the most frequently used illegal drug among women of reproductive age, but little is known about the consequences of using marijuana during pregnancy. Delta-9-tetrahydrocannabinol (Δ⁹-THC), one of the active chemicals in marijuana, has been shown to cross the placental barrier easily. In this study, pregnant Long Evans rats were assigned to one of three treatment groups (Δ⁹-THC-exposed, vehicle control, and non-treated control) on day 1 of gestation. Drug exposure consisted of 2 mg/kg of natural Δ⁹-THC, administered twice daily by subcutaneous (s.c.) injection, from gestational day 1 through 22. Pups continued to receive drug exposure via s.c. injection from postnatal day 2 through 10. Male rats from each group were tested starting on postnatal day 90 in a battery of tests which included open field activity, active social interaction, and the forced-swim test. There were no significant differences in weight gained by dams or weight of offspring when compared to controls. Δ⁹-THC-exposed rats showed decreased time in the inner part of the open field and an increase in investigation time in the test of social interaction compared to both control groups. There were no differences among groups in the forced-swim test. Perinatal Δ⁹-THC exposure may result in increased susceptibility to anxious behavior and alter social functioning in adult offspring.

1. Introduction

Marijuana is a complex compound containing numerous active ingredients. One of the most prevalent psychoactive ingredients of marijuana is delta-9-tetrahydrocannabinol (Δ⁹-THC) which acts at cannabinoid (CB) receptors [35]. Two subtypes of cannabinoid receptors (CB₁ and CB₂) have been identified. The CB₁ receptors are found predominantly in the brain and mediate the subjective effects of marijuana [1]. The CB₂ receptors are found predominantly in the spleen and immune system but have recently been shown to be on microglia, astrocytes and some neuronal populations [15, 33, 50]. Importantly, developmental studies of rodents [6] and humans [34] suggest CB₁ receptors are present and active starting in early prenatal periods, with receptor density gradually but substantially increasing throughout development until adult levels are reached. CB receptor binding has been detected as early as gestational day (GD) 11 in rodent brains [7, 27] and week 14 of gestation in human brains [8, 34]. Endogenous cannabinoid activity at both CB₁ and CB₂ receptors has been proposed to have functions specific to development of the nervous system [14, 15, 27]. In particular, activation of CB1 receptors has been suggested to be involved in cell survival and differentiation [for a review, see 14, 27] and the CB₂ receptor has been proposed to be involved in cell death and neuroprotection [for a review, see 15].

Given the involvement of the endogenous cannabinoid system in development of the nervous system, it is not surprising that a variety of behavioral and cognitive abnormalities have been reported in children of women who self-reported marijuana use during pregnancy [12, 22, 23, 26, 45]. While changes in attention and cognitive function have been widely reported [19, 42, 45], socioemotional changes after developmental exposure to marijuana have been less frequently investigated. Nevertheless, increased fearfulness and shorter length of play have been reported in children after prenatal exposure to marijuana [12]. Marijuana-exposed children from the Maternal Health Practices and Child Development Project were reported to show impulsivity, hyperactivity, increased delinquency and increased externalizing problems [23] and an increase in prevalence childhood depressive and anxiety symptoms at ten years of age [26].

In order to control for the confounding factors in human studies of prenatal marijuana exposure, animal models have been used and are critical for isolating the effects of prenatal marijuana use from other confounds (for reviews, see [14, 27, 40]). As in the human literature, the research on animals has focused on cognitive deficits but there is some research on socioemotional changes. Exposure to hashish during the perinatal period has been shown to alter behavior in a social preference test in adult male but not adult female rats [39]. There is also evidence that perinatal exposure to Δ⁹-THC results in a sexually dimorphic change in the hypothalamic-pituitary-adrenal axis, with females showing an up-regulation and males showing a down-regulation of this system after exposure to Δ⁹-THC during development [46]. CB₁ receptor knockout mice show suppression of the endocrine and behavioral response to repeated stresses [18]. There are also reports of increased exploratory behavior in a plus-maze paradigm [46] and in the defensive withdrawal test [47] in Δ⁹-THC -exposed rats, suggesting less anxiety. Behavior in an open field is influenced by some, but not all, aspects of anxiety [44] with a decrease in locomotion, particularly in the inner portion of the open field, often interpreted as an increase in anxiety [44]. The findings on the effects of gestational exposure to cannabinoids on locomotor activity in rodents are mixed and may be both age- and sex-dependent. In adult rats, perinatal exposure to Δ⁹-THC has been shown to result in either decreased activity at long intervals (with stronger effects in males) [38] or increased activity in females [41, 46]. Perinatal exposure to hashish has been shown to have no significant effects on activity in adult rats [39] whereas prenatal exposure to anandamide, the endogenous ligand for the CB₁ receptor, results in decreased activity in both sexes [17].

Rodent models of marijuana exposure during development have been such that the exposure period either has been limited to gestation in the rat [e.g., 29, 37] or has been through the dam's milk during the early postnatal period [e.g., 39, 41]. The prenatal period in the rat does not include the period equivalent to the third trimester in humans [5]. The third trimester in humans or the postnatal period in the rat, which is equivalent to the third trimester, is a critical period for teratogenic effects on the brain [11]. While Δ⁹-THC crosses into the milk of a lactating animal, the amount that actually gets into the offspring is less than 1% of the dose [10, 30] (compared to almost complete transfer across the placental barrier [2, 4]) and is unlikely to mimic third trimester exposure in humans. Furthermore, the exposure of the lactating female rat to cannabinoids disrupts its maternal behavior and milk release which will impact pup development [9] in a manner different than a direct teratogenic effect of cannabinoid exposure on the offspring.

This study uses a pre- and early postnatal exposure period which closely mimics the period of central nervous system development that occurs during the three trimesters of human pregnancy [5]. Importantly, the drug administration during the postnatal period in the rat was directly to the pup to ensure that exposure of the pup to a moderate dose of Δ⁹-THC and that maternal behavior was not directly impacted. In this study, drug administration is standardized in two 2 mg/kg subcutaneous injections daily of natural Δ⁹-THC which mimics estimates of low to moderate daily exposure to this compound in humans and is comparable to the dose used in previous studies [41, 46]. The oral route of administration was avoided due to significant first pass metabolism and inconsistent absorption as well as the creation of new cannabinoids during digestion [35]. The daily dosage was divided into two injections to mimic the pattern of heavy human users of marijuana who report smoking multiple joints throughout the day [3], in a manner that would cause neuronal exposure to an elevated level of cannabinoids for a significant portion of every day.

Importantly, this study examines socioemotional behavior in the adult animal to determine any long-lasting or emerging effects due to Δ⁹-THC exposure across the period equivalent to all three trimesters in the human. Measures of socioemotional behavior included open field behavior, social interactions, and the behavioral despair test. Open field behavior is sensitive to both changes in overall activity and anxiety levels [44], with the activity in the inner field being particularly sensitive to anxiety [44]. Measurement of social interactions [36] can reflect changes in both anxiety and social behavior. The forced swim test has been used as an assay for antidepressant effects of drugs and is often considered a model of depression [43]. Given that adolescents exposed to marijuana during the prenatal period show high rates of depression [26], a test sensitive to anti-depressants was considered to be an important addition to the battery of tests to assess the socioemotional effects of perinatal administration of Δ⁹-THC in adult rats.

2. Methods

2.1. Subjects

Twenty-four female Long Evans rats were obtained from Harlan and bred in the laboratory after a 1 week period of adaptation. Breeding consisted of overnight housing of virgin females and proven stud males in breeder cages (4–5 females and 1 male in each cage). Pregnancy was determined the following morning by vaginal swabs that were smeared on culture slides and examined under microscope for the presence of sperm. Dams were randomly assigned to one of three groups: Δ⁹-THC -exposed (Δ⁹-THC), Non-treated control (NC), and Vehicle-injected control (IC). Each group consisted of 8 dams. Rats were housed in a colony with temperature and humidity controls, and were under a 12/12 light cycle. Food consumption was not limited. Pups were paw-marked for identification on PD 10 and weaned on PD 21 and housed in same-sex groups of 2–3. One male rat was randomly selected from each litter for testing resulting in a total of 8 rats per group. The selected offspring were tested in all behavioral tests. The pregnant dams were weighed daily from GD 1 to GD22, and offspring were weighed daily from PD 2 to PD 10, and on PD 21, 30, 60, and 90.

2.2. Drug Treatment

Drug treatment began on day 1 of gestation and consisted of 2 mg/kg injections of natural Δ⁹-THC (NIDA), administered twice daily. Δ⁹-THC was administered via subcutaneous injection in a vehicle of ethanol, Tween 80, and 0.9% saline, in a ratio of 1:1:18. Drug exposure continued from gestational day (GD) 1 to GD 22, and PD 2 to PD 10. IC rats received twice daily vehicle injections to control for the effects of the stress from drug administration and the vehicle. NC rats received no treatment. The injection volume for both the Δ⁹-THC and the vehicle was 1.06 ml/kg.

2.4. Behavioral Testing

Behavioral testing began on PD 90, when rats are considered fully developed adults [48]. The order of testing was as follows: observation of activity in an open field, observation of active social interactions, and behavior in the forced swim test.

2.4.1. Open Field Activity

Testing was as described in [25]. Each rat was habituated to a square box (40.6 × 40.6 × 40.6 cm) with high walls for three periods of 10 minutes daily for three days. The box was electronically divided into an inner square (10. 2 × 10.2 cm), and an outer area surrounding the inner square. Testing consisted of one ten-minute period in which movement was measured via a camera mounted above the open field and the Ethovision automated tracking program (Noldus). The box was wiped with alcohol before each testing session to minimize the scent of other rats. Measurements included distance and time spent in the inner and outer sections.

2.4.2. Active Social Interaction

Active social interactions were assessed as described in [31, 36]. After three daily ten minute habituation periods to a clear housing box, test rats were placed with an unfamiliar, non-experimental rat of the same size and sex. A new, clean housing box was used for each session. Each rat was videotaped for a 15 minute period. The tapes were scored for incidence of sniffing, dominant mounts, social grooming, chasing, pinning, biting, boxing, and avoidance. For each videotaped session, behaviors were scored every 5 seconds.

2.4.3 Forced Swim Test

The forced swim test was conducted as described in [43]. Each rat was placed in a tank of room temperature water for a habituation period of 15 minutes in order to induce a state of helplessness. Approximately 24 hours later, each rat was again placed in the water tank for a 5 minute period. The latency to immobility (judged as passively floating with movement only to keep one’s head above the water) was measured.

3. Results

3.1 Physical Data

Dams were weighed daily from gestational day (GD) 1 until GD 22. Analysis of body weights using a repeated measures analysis of variance (ANOVA) indicated no significant effect of group or interaction of group and gestational day. There was a main effect of gestational day (F(21,252) = 105.2, p < 0.001) indicating that all dams gained weight over pregnancy (see Table 1).

Table 1.

Mean body weights (in g) and standard error of the means (SEMs) for dams across gestational days (GD) and offspring across postnatal days (PD)

Groups	GD 5	GD 10	GD 21
Dams in NC group	258.1 ± 2.6	273.1 ± 4.1	360.5 ± 12.1
Dams in IC group	259.8 ± 7.6	275.6 ± 7.6	342.3 ± 11.8
Dams in THC group	253.5 ± 3.4	263.3 ± 4.9	333.2 ± 5.7
	PD 2	PD 10	PD 90
Pups in NC group	7.3 ± 0.2	19.8 ± 1.1	415.8 ± 15.0
Pups in IC group	7.1 ± 0.2	19.7 ± 0.8	385.8 ± 13.7
Pups in THC group	6.9 ± 0.2	19.9 ± 0.6	379.3 ± 7.6

Length of gestation did not differ between groups; all dams gave birth on gestational day 23 and there was no mortality among the pups. The mean weights of the experimental rats were analyzed over PD 2 through PD 10 and on PD 21, 30, 60, and 90. A repeated measures ANOVA indicated no significant differences among groups or interaction of group with postnatal day. There was a main effect of postnatal day (F(12,252) = 3,338.3, p < 0.001) indicating that all pups gained weight across postnatal day (see Table 1).

3.2. Behavioral Data

3.2.1. Open Field Activity

An ANOVA on time spent in the inner section showed a significant main effect of group [F(2,21) = 3.7, p < 0.05] and LSD post hoc tests showed that the Δ⁹-THC rats spend less time in the inner section compared to NC and IC control rats (p's < 0.05) (see Figure 1). An ANOVA on distance traveled in the inner section did not show a main effect but did show a trend towards a main effect of group [F(2,21) = 3.5, p = 0.051]. LSD post hoc tests showed that the Δ⁹-THC rats traveled less distance in the inner section compared to NC and IC controls (p's < 0.05) (see Table 2). Analyses of distance traveled or time spent in the outer section did not reveal a significant difference or a statistical trend towards a significant difference (p's > 0.10) among groups (see Table 2).

Figure 1. Open Field Activity: Time spent in inner section.

THC-exposed rats spent significantly less time in the inner area compared to NC and IC controls. Error bars represent SEMs.

Table 2.

Activity measures (means ± SEMs) in the open field

Groups	Distance in inner section (cm)	Time in outer section (sec)	Distance in outer section (cm)	Time ratio	Distance ratio
NC	537.9 ± 78.2	259.2 ± 7.2	3030.3 ± 175.5	0.15 ± 0.02	0.15 ± 0.02
IC	519.1 ± 87.5	247.0 ± 10.4	3089.6 ± 299.3	0.18 ± 0.03	0.15 ± 0.03
THC	287.1 ± 56.6	270.3 ± 7.7	2722.7 ± 185.4	0.09 ± 0.02	0.09 ± 0.02

In order to further explore these effects, a time ratio was constructed where the time spent in the inner section was divided by the sum of the time spent in the inner and outer sections of the open field. A distance ratio was constructed in a similar manner. These data are shown in Table 2. Planned comparisons revealed that the Δ⁹-THC rats had significantly smaller time and distance ratios compared to the control rats (p's < 0.05), suggesting that the effects on activity were specific to the inner section.

3.2.2 Active Social Interaction

To assess social interaction, investigation (sniffing) time was analyzed by an ANOVA. There was a main effect of group (F(2,21) = 4.2, p < 0.05); the Δ⁹-THC rats exhibited an increase in sniffing compared to both NC and IC rats (Tukey's post-hoc tests, p’s < 0.05) (See Figure 2). There were no incidences of social grooming, biting, or chasing. The incidence of dominant mounts, pinning, boxing, and avoidance were shown by only a few animals within each group and were too low to analyze (data shown in Table 3).

Figure 2. Active Social Interaction.

THC-exposed rats exhibited significantly more investigative sniffing compared to NC and IC controls. Error bars represent SEMs.

Table 3.

Incidence of behaviors during active social interactions (mean and SEMs)

Groups	Dominant mounts	Pinning	Boxing	Avoidance
NC	0.25 ± 0.25	1.88 ± 1.04	0	0.50 ± 0.50
IC	0	0.25 ± 0.25	0.25 ± 0.25	0.62 ± 0.32
THC	0	0.38 ± 0.26	0	0

3.2.3. Forced Swim Test

Δ⁹-THC rats did not differ significantly from NC and IC rats in latency to immobility in a five minute session 24 hours after inducing a state of helplessness. The average latencies to immobility and SEMs in seconds were 219.2 ± 23.3, 217.4 ± 11.6 and 223.8 ± 12.8 in the NC, IC and THC groups, respectively.

4. Discussion

The results of this study suggest that perinatal exposure to Δ⁹-THC causes alterations in the central nervous system that result in disruptions of social and emotional behavior in adult male Long Evans rats. The presence of these behavioral differences was observed on PD 90 and later, implying that they are long-lasting changes. Physical measurements of growth and length of gestation support the consensus of the majority of the existing literature that perinatal Δ⁹-THC exposure does not alter weight gained by dams or offspring, and does not shorten the length of gestation [20, 21, 37]. The few reports of shortened gestation period and reduced weight of human offspring due to prenatal marijuana use may be due to confounds such as additional drug use or environmental factors.

The behavioral results indicate specific effects on anxiety-like behavior and social interaction. In the test of open field activity, Δ⁹-THC-exposed rats spent significantly less time in the center area of the maze compared control rats. Distance traveled in the inner section of the field showed a trend towards significant effect by group. This type of behavior has been interpreted as anxious or fearful [25]. Ratios of time spent and distance traveled in the inner section divided by time and distance (respectively) in both the inner and outer sections suggest the decrease in behavior seen in Δ⁹-THC rats in the open field was specific to the inner section and not an overall disruption of psychomotor activity as reported elsewhere [38]. These results suggest increased susceptibility to anxiety-like behaviors in adulthood in the Δ⁹-THC -exposed rats.

The measurement of active social interactions revealed significant differences in investigative behavior (seen as sniffing in rats). Δ⁹-THC rats displayed a substantial increase in sniffing compared to controls. The increase in active social interactions could be interpreted as a decrease in anxiety [16] although this interpretation is inconsistent with our findings on open field activity described above. However, given the multi-faceted nature of anxiety, it is not unusual for different measures of anxiety to be inconsistent with each other [16, 44]. Another possibility is that the increased investigative behavior observed in this study is a precursor to aggression induced by the anxiety of exposure to a novel animal. This interpretation would be consistent with clinical reports of prenatal marijuana exposure correlating with behavioral disruptions such as shorter length of play, increased fearfulness, hyperactivity and increased delinquency and externalizing problems [12, 23]. However, Δ⁹-THC rats did not show increases in other aggressive behaviors (pinning, biting, chasing, or boxing) suggesting the effect of perinatal Δ⁹-THC on social interactions may be subtle. The specific nature of the change in social interactions induced by exposure to Δ⁹-THC during development could be further analyzed with more direct tests of aggression, play and sexual behaviors.

The results from the forced swim test suggest that perinatal Δ⁹-THC exposure is not correlated with increased susceptibility to depression or learned helplessness. The forced swim test was chosen for this study as a model of depression, since it has been shown to be sensitive to a variety of anti-depressants [32, 43]. A growing body of recent literature has suggested a relationship of endogenous cannabinoid activity to depressive and anxiety behaviors [28, 49, 51] and an increase in depressive and anxiety symptoms has been reported at 10 years of age [26] in a sample of children exposed to marijuana during gestation. It is possible that environmental stress and/or higher amounts of exposure to Δ⁹-THC are necessary to induce changes in behavior in the forced swim test. Another possibility is that the increased susceptibility to depression is corrected by adulthood, and therefore was not detected in our observations of adult rats.

The neural changes that mediate the effects of exposure to Δ⁹-THC during the perinatal period remain to be elucidated. Δ⁹-THC acts on both CB₁ and CB₂ receptors and both of these receptors are involved in development of the nervous system. Actions on the CB₁ receptors are likely to have complex effects on cell number, neuronal differentiation, and synaptic connections whereas actions on CB₂ receptors may have more specific effects on cell number [15, 27]. Repeated cannabinoid exposure in adults is associated with down regulation or desensitization of CB₁ receptors [24], and is reversible in fully developed brains; it is plausible that this may not be the case in the developing brain. Furthermore, given the interactions of the cannabinoid system with other systems that modulate development of the nervous system [27], it is likely that exposure to Δ⁹-THC during development may not only alter the endogenous cannabinoid system but other systems as well.

There are a number of limitations to this study. The first is the use of a single dose of Δ⁹-THC, since it is likely that the effects of perinatal exposure to cannabinoids are dependent upon dose. However, this study did use a dose that was within a moderate range [41, 46] and thus, the effects might reflect those that are typical. Another limitation is the testing of behavior at one time point. The ninety day time point was chosen because it is well past adolescence in a rat and considered to be a period where behavior is stable until aging effects occur [48]. Nevertheless, it might be revealing to study the time course of Δ⁹-THC effects across development both with respect to neural bases and behavioral mechanisms. It is also possible that the effects of exposure to Δ⁹-THC during early development on adult rats might either be ameliorated or exacerbated by the aging process or exposure to environmental stressors or enrichment. Examination of whether the Δ⁹-THC effects in rats can be ameliorated by various interventions would be an interesting avenue of research with important clinical implications.

Care must be taken when generalizing the current findings to other populations or measures. This study only examined behavior in males and the findings may differ in females. Many of the behavioral tests for rodents have been validated in males only. The performance in tests by females, particularly in tests of anxiety, does not necessarily reflect the same aspects of behavior as that in males [13]. This may explain some of the observed sexually dimorphic effects of perinatal Δ⁹-THC exposure [38, 40]. It is also important to note that this study only used two measures of anxiety in rats, namely open field behavior and the forced swim test. As already mentioned, anxiety is a multifaceted concept with multiple ways to measure it and each of those measures has various strengths and weaknesses (see [16, 44] for reviews]. It will be important to use other measures of anxiety, such as the elevated plus maze and potentiated startle test, to assess the extent to which the current findings can be generalized.

The results of this study are consistent with previous findings that the brain is altered by exposure to Δ⁹-THC during development and this alteration does not fully recover [14, 27]. Importantly, these findings are based on a rat model where the moderate exposure occurs throughout the period equivalent to all three human trimesters [5] and maternal behavior is not directly altered by the Δ⁹-THC exposure. Increased anxiety, as measured by decreased locomotion in the inner portion of an open field, and increased social investigation were found to be altered in adults that had been exposed to Δ⁹-THC during development, suggesting long-lasting changes in socioemotional behavior. Given that the cannabinoid system is involved in anxiety in a dose- and context-dependent manner [51], a permanent change in this system would impact anxiety and possibly social behaviors in adult rats. Clearly, more research needs to be conducted, particularly examining the impact of development exposure to Δ⁹-THC on cannabinoid receptors and endogenous cannabinoid activity and on other measures of anxiety and social behaviors. Long-term changes in socioemotional behaviors after exposure to cannabinoids during development have direct implications for behavioral and pharmacological treatments.