The authors have declared that no competing interests exist.
Conceived and designed the experiments: MER HKC. Performed the experiments: MER EJdC. Analyzed the data: MER HKC. Contributed reagents/materials/analysis tools: MER HJL HKC. Wrote the paper: HKC.
Oxytocin (Oxt) acting through its single receptor subtype, the Oxtr, is important for the coordination of physiology and behavior associated with parturition and maternal care. Knockout mouse models have been helpful in exploring the contributions of Oxt to maternal behavior, including total body
In females, the neuropeptide oxytocin (Oxt) helps to coordinate uterine contractions and milk ejection (for review see
In addition to its importance to maternal behavior, Oxt in the postpartum period may impact anxiety and depression. One of the changes thought to be necessary for normal maternal behavior is decreased anxiety during the postpartum period
While elegant pharmacological work has provided insight into the neural regulation of maternal behavior and postpartum anxiety-like and depression-like behaviors
Unfortunately, there have been conflicting reports regarding the effects of Oxtr disruption on maternal behavior in these two lines of Oxtr −/− mice. Like Oxt −/− mice, Oxtr −/− females appear to have normal parturition and an inability to milk eject. In the Nishimori line of Oxtr −/− mice, virgin and postpartum females display longer latencies to retrieve pups and postpartum females display less time crouching over pups
Previous research has also found that Oxtr −/− mice have deficits in social recognition
Female siblings (19–20 days old) from heterozygous breeding pairs were weaned at 21 days and group housed until at least 8 weeks of age. Animals were maintained on a 12∶12 light-dark cycle (lights on 0200h) and food and water were available
Virgin Oxtr +/+ mice were thelectomized to eliminate milk ejection, and Oxtr −/− mice received a sham surgery, as they already cannot milk eject. Our choice to have Oxtr −/− mice only receive a sham surgery was guided by previous research indicating that suckled and non-suckled rat dams display similar levels of maternal behaviors and have similar c-fos expression in brain regions associated with the regulation of maternal behavior
At least 1 week post surgery, Oxtr +/+ (n = 19) and Oxtr −/− mice (n = 26) were group housed with wet nurses, as neither group of experimental mice were able to milk eject. Wet nurses were female C57BL/6J or females from our Flox/Flox line of mice, which are a mixture of C57BL/6J and 129/Sv. Wet nurse dams and experimental mice were paired for at least 1-week post surgery. 3 days prior to mating, male bedding was added to the females' cages to induce the Whitten effect, where male odors induce estrus and synchronize estrus cycles among females
Approximately 1 week prior to the expected parturition date, individual experimental Oxtr +/+ and Oxtr −/− mice and their wet nurses were rearranged so that experimental animals and the wet nurses had the same projected parturition date. If necessary, Oxtr +/+ and Oxtr −/− mice were paired with an alternative wet nurse prior to parturition to ensure that the wet nurse experienced parturition on the same day or the day prior to Oxtr +/+ or Oxtr −/− mice. If an experimental dam gave birth prior to its paired wet nurse it was removed from the study, as the pups were unable to survive due to the lack of milk ejection in the experimental animals. On PND0, defined as the day of parturition, all wet nurse and experimental litters were collectively culled to a total of 4 pups. No attempt was made to discriminate between offspring from experimental versus wet nurse dams. Prior to testing on PND1, dams were checked for pups. Those dams that did have pups remaining were then tested for maternal care. If no pups remained it was noted whether pups were abandoned or cannibalized. Prior to all behavioral testing, animals were acclimated to the testing space for 1 hour, and unless otherwise indicated, tests were performed 1 hour after lights out under dim red light illumination.
Oxtr +/+ (n = 9) and Oxtr −/− (n = 7) females were tested for maternal care on PND1-3. The wet nurse was removed for all portions of the maternal care testing, and the maximum amount of time the wet nurse was separated from the pups was approximately 20 minutes. Pup retrievals and maternal behaviors were recorded by first removing all pups from the Oxtr +/+ or Oxtr −/− dams for 5 minutes. Removed pups were placed into a cage with their wet nurse. After 5 minutes, the pups were scattered in the home cage opposite of the dam, and the dam's behavior was then videotaped for an additional 20 minutes and later quantified for maternal behaviors.
Pup retrievals were quantified in the first 5 minutes following the return of the pups to the cage, and included the latency to retrieve the first pup and the latency to retrieve all of the pups. Latency to retrieve all pups was determined by the amount of time the dam took to return all 4 pups to the nest. If not all pups were retrieved within 5 minutes of being returned to the home cage, a latency score of 300 s was recorded and maternal behaviors were not scored. If all the pups were retrieved, 10 minutes of maternal care was scored. Behaviors scored were: time on/off the nest, nest building, sniffing pups, licking pups, nursing position, self-grooming, digging, and non-social behaviors (eating, sleeping, and exploratory behaviors). All behaviors were coded using Noldus Observer 9.0 (Leesburg, VA) by an observer blind to the genotypes. Following testing, the wet nurse was returned to the home cage. Across the 3 days of testing there was a maximum of 15 minutes of searching behavior and 30 minutes of maternal care recorded for each animal. The duration of time the Oxtr +/+ and Oxtr −/− dams spent engaging in all the behaviors measured was summed, and maternal care and pup retrieval were analyzed within each day using a one-way ANOVA with genotype as the main factor. A repeated measures analysis was not performed because only some animals retrieved pups on PND1-3. Thus, we did not score maternal behaviors for all 3 days for all of the animals. A Fisher's exact test was used to compare pup mortality between Oxtr +/+ and Oxtr −/− mice. In all statistical tests a p-value of ≤0.05 was considered statistically significant.
Maternal aggression was tested in Oxtr +/+ and Oxtr −/− dams each day from PND4-6 1 hour after lights out under dim red light. Testing was completed on PND4-6 so it would not interfere with the maternal behavior testing, as we did not want to inadvertently cause an increase in pup abandonment. Wet nurse dams were removed for the duration of testing, and pups remained in their home cage. An intruder Balb/c (adult, gonad-intact) (The Jackson Laboratory, Bar Harbor, ME) or C57BL/6J (adult, gonad-intact) male mouse between the age of 2 and 6 months was added to the home cage and the animals were recorded. If no attack against the intruder occurred during the 5-minute test, a latency of 300 s was recorded. When an attack occurred, the dam's behavior was scored for an additional 2 minutes after the time of the initial attack. If an attack occurred, the behaviors scored were: tail rattle, fleeing, attack behavior (lunge-bite), aggression (pushing, chasing, or contact with the intruder), self-grooming, and non-social behavior (eating, sleeping, and exploratory behaviors). All behaviors were scored using Noldus Observer 9.0 (Leesburg, VA) by an observer blind to the genotypes. No statistical analyses were performed for maternal aggression, as too few experimental animals displayed aggressive behavior.
Measurements of postpartum anxiety-like and depression-like behaviors were conducted on a separate group of Oxtr +/+ (n = 8) and Oxtr −/− (n = 9) dams on PND4-6. These animals underwent surgery as described above. PND4-6 were selected for testing because this was after maternal behavior had been established and we were less likely to facilitate pup abandonment. All Oxtr +/+ and Oxtr −/− dams were paired with a wet nurse for the duration of testing, and were tested regardless of whether or not they had abandoned their pups. An elevated plus maze test was performed on PND4 under ∼100 lux lighting using the mouse maze as previously described
On PND5, experimental mice were tested in an open field. The open field arena was made out of Plexiglas and measured 45.5×45.5×30 cm
On PND6 Oxtr +/+ and Oxtr −/− mice were recorded in a forced swim test. They were placed into a 19 cm in diameter cylindrical tank filled to 24 cm with room temperature (∼21°C) water for 10 minutes, and observed for any signs of distress. Following testing, dams were returned to a clean cage with their wet nurse and pups. Forced swim was later scored using the Noldus Observer 9.0 (Leesburg, VA) as previously described
10 out of 15 Oxtr −/− dams had abandoned their litters by PND1, with all pups cannibalized or found dead in the cages (See
67% of Oxtr −/− females cannibalized or abandoned their pups by PND1 compared to 20% of Oxtr +/+ females. *two-tailed p = 0.021.
There were no genotypic differences in the latency to retrieve the first pup across the 3 days of testing.
Day 1 | ||
Behavior | Oxtr +/+ (mean ± SEM) (n = 9) | Oxtr −/− (mean ± SEM) (n = 7) |
Time on nest | 200.36±47.59 | 260.51±18.41 |
Time off nest | 396.93±46.63 | 338.72±18.15 |
Duration Nest building | 68.11±35.45 | 67.21±23.54 |
Duration Sniffing pup | 12.85±5.44 | 9.93±8.60 |
Duration Licking pups | 3.14±1.48 | 5.52±3.48 |
Duration Retrieving pups | 11.13±9.35 | 1.69±1.67 |
Duration Self-grooming | 11.13±3.67 | 16.93±3.86 |
Duration Non-social | 490.61±35.92 | 497.38±30.41 |
Day 2 | ||
Behavior | Oxtr +/+ (mean ± SEM) (n = 9) | Oxtr −/− (mean ± SEM) (n = 7) |
Time on nest | 257.80±58.58 | 243.34±48.13 |
Time off nest | 341.69±58.59 | 356.09±48.08 |
Duration Nest building | 82.04±33.40 | 89.46±48.96 |
Duration Sniffing pup | 11.21±2.42 | 8.068±3.22 |
Duration Licking pups | 6.92±2.03 | 20.60±9.60 |
Duration Retrieving pups | 15.83±9.19 | 5.10±2.77 |
Duration Self-grooming | 24.90±8.20 | 30.14±8.10 |
Duration Non-social | 454.28±33.00 | 445.29±47.90 |
Day 3 | ||
Behavior | Oxtr +/+ (mean ± SEM) (n = 9) | Oxtr −/− (mean ± SEM) (n = 7) |
Time on nest | 225.20±54.99 | 191.94±39.27 |
Time off nest | 374.23±54.95 | 407.48±38.23 |
Duration Nest building | 106.81±35.50 | 41.62±13.68 |
Duration Sniffing pup | 8.23±3.09 | 5.46±2.92 |
Duration Licking pups | 6.95±6.42 | 13.19±8.28 |
Duration Retrieving pups | 9.57±9.57 | 9.13±8.40 |
Duration Self-grooming | 10.67±4.32 | 3.2±0.80 |
Duration Non-social | 469.48±35.30 | 498.94±25.26 |
Oxtr −/− dams did not display increased anxiety-like and depression-like behaviors during the postpartum period compared to Oxtr +/+ dams. In the elevated plus maze, there was no genotypic difference between the percent time spent in the open and closed arms, F = 1,15 = 2.081, p = 0.170 (See
Oxtr +/+ (mean ± SEM) (n = 8) | Oxtr −/− (mean ± SEM) (n = 9) | |
Time in inner arena (%) | 27.3±7.0 | 15.9±3.1 |
Time in outer arena (%) | 72.7±7.0 | 84.1±3.1 |
Number of entries inner arena | 135.7±29.4 | 73.6±11.9 |
Total distance traveled (cm) | 5667.6±387.3 | 5060.3±556.6 |
Time in closed arms (%) | 80.5±4.9 | 64.5±9.3 |
Time in open arms (%) | 19.5±4.9 | 35.5±9.3 |
Swim (%) | 34.1±2.7 | 48.3±7.8 |
In the current study we used Oxtr −/− female mice to assess the importance of Oxt signaling in maternal behavior, anxiety-like, and depression-like behaviors in the postpartum period. Oxtr −/− females have increased pup abandonment, but those who initiated maternal behavior did not differ from wildtype thelectomized controls. These findings are consistent with past research that indicates that Oxt is important for the initiation, but not the maintenance of maternal behavior
Past research using Oxtr −/− and Oxtr FB/FB mice also support the hypothesis that Oxt is important to the initiation of maternal behavior. Previous studies have reported deficits in maternal behavior, as measured by longer latencies to retrieve pups, in Oxt −/− and Oxtr −/− mice
Given the design of Experiment 1, there are some potential limitations that should be considered. Based on observations that Oxtr −/− pups have reduced ultrasonic vocalizations compared to Oxtr +/+ pups
Our finding of 67% of Oxtr −/− females showing pup abandonment is higher than the percentages reported in Oxtr FB/FB mice (only 40%–60% pup abandonment in their first litters)
In Experiment 2, we found that Oxtr −/− dams exhibit no differences in postpartum anxiety-like or depression-like behaviors compared to Oxtr +/+ females. While there was a nearly statistically significant difference in the open field test, with Oxtr +/+ dams entering the inner arena more frequently than Oxtr −/− dams, we think that this is unlikely to be due to genotypic differences in anxiety-like behavior since no other behavioral measures indicated an anxiety-like phenotype. Also, the increased crossings do not appear to be the result of overall increased locomotion in the Oxtr +/+ females, as there was no genotypic difference in the total distance travelled. The lack of genotypic differences in the measures of anxiety was surprising given previous research indicating the anxiolytic effects of Oxt in the postpartum period
Postpartum mood and anxiety disorders in humans are associated with impaired maternal care and a lack of attachment for the child
In summary, we have demonstrated that an absence of Oxt signaling results in increased pup abandonment in Oxtr −/− dams. However, in Oxtr −/− dams that did not abandon their pups, maternal behavior is normal. Also, in Oxtr −/− dams that have undergone parturition, there appear to be no profound effects on measures of anxiety-like or depressive-like behavior. Our research is consistent with the hypothesis that Oxt lowers the threshold for the initiation of maternal behavior, but that once the program is initiated Oxt is not critical to its maintenance.
We thank Devinne Dietz for her assistance with the behavioral studies, Shannah Witchey for her husbandry and genotyping work, and W.S. Young, III for genetically engineering the Oxtr −/− mice and providing us with the original breeding pairs. We also thank W.S. Young, III and members of the Caldwell laboratory for comments on this manuscript as well as the excellent care provided to our animals by the members of the vivarium staff at KSU.