Gender Dependent Evaluation of Autism like Behavior in Mice Exposed to Prenatal Zinc Deficiency

Abstract

Zinc deficiency has recently been linked to the etiology of autism spectrum disorders (ASD) as environmental risk factor. With an estimated 17% of the world population being at risk of zinc deficiency, especially zinc deficiency during pregnancy might be a common occurrence, also in industrialized nations. On molecular level, zinc deficiency has been shown to affect a signaling pathway at glutamatergic synapses that has previously been identified through genetic mutations in ASD patients, the Neurexin-Neuroligin-Shank pathway, via altering zinc binding Shank family members. In particular, prenatal zinc deficient but not acute zinc deficient animals have been reported to display autism like behavior in some behavioral tests. However, a full behavioral analysis of a possible autism like behavior has been lacking so far. Here, we performed an extensive behavioral phenotyping of mice born from mothers with mild zinc deficiency during all trimesters of pregnancy. Prenatal zinc deficient animals were investigated as adults and gender differences were assessed. Our results show that prenatal zinc deficient mice display increased anxiety, deficits in nest building and various social interaction paradigm, as well as mild alterations in ultrasonic vocalizations. A gender specific analysis revealed only few sex specific differences. Taken together, given that similar behavioral abnormalities as reported here are frequently observed in ASD mouse models, we conclude that prenatal zinc deficient animals even without specific genetic susceptibility for ASD, already show some features of ASD like behavior.

Keywords: ASD; ZIP; Zn; Zn2+; ZnT; brain; plasticity.

Figure 1

Generation and general health of prenatal zinc deficient (PZD) mice. (A) After 8 weeks of zinc deficient diet including 3 weeks of pregnancy, a significant reduction in blood zinc levels was found in zinc deficient mothers (n = 3 per group) using AAS. (B) Eight weeks of zinc deficient diet including 3 weeks of pregnancy did not significantly alter Na⁺, Mg²⁺, K⁺, and Fe²⁺ levels. However, the blood Cu²⁺ concentration was significantly (p = 0.009) increased (n = 3 per group; measured by ICP-MS). (C) Maternal zinc deficiency reduces the level of zinc in the brain of pups. Using the zinc specific fluorophore Zinpyr1, a significant reduction in signal intensity in brains from pups (post-natal Day PD2, n = 3) from zinc deficient mothers can be seen. Sections of control and PZD pups were imaged and quantified by measuring at least three optic fields of view and controls were set to 100%. (D) Blood zinc levels of adult PZD mice vs. Controls (n = 10 each). At the time-point of testing, no significant difference in blood zinc level was detected using AAS. (E) PZD mice displayed significant differences regarding their body weight compared to controls. N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. ^*p ≤ 0.05; ^**p ≤ 0.01.

Figure 2

PZD mice show increased anxiety. Prenatal zinc deficiency affects anxiety related behaviors in the open field test (A–F) and elevated plus maze (EPM) (G–N). (A) PZD mice spent significantly less time in the center of the open field arena and displayed increased time in the boarder zone (B). Gender specific analysis (small inserts) shows a clear trend toward this behavior for both male and female PZD mice without significant gender specific effect. (C) PZD mice show a reduced number of entries in the center zone seen as clear trend (#). However, PZD mice are not impaired in the locomotor parameters of the open field test. (D–F) PZD mice walked the same distance, traveled with a speed similar to control mice, and showed no differences in the number of ambulations (F) compared to controls. (A–F) N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. No gender specific effects were observed. (G) In the EPM, PZD mice show a significant difference to control mice regarding the time spent in the open arms of the maze. (H) The entries into the closed arms, and the entries into the open arms were not significantly different in PZD mice compared to controls. (J) The total number of entries was not significantly different in PZD mice. Regarding anxiety like behavior, no gender specific effect was observed. (K–M) In the EPM, PZD mice displayed impairments in the locomotor parameters. (K) PZD mice walked less a distance, and (L) traveled with less speed compared to control mice. (M) No significant differences were detected in the mean number of ambulations. (N) Exemplary images showing the tracked path of a CTRL and PZD mouse in the EPM. (G–N) N = 32 CTRL animals and n = 36 PZD mice were analyzed. ^*p ≤ 0.05.

Figure 3

PZD animals show impaired nest building. (A) The nest quality score was assessed 24 h after introduction of a nesting material in the home cage according to a 5 point rating scale. PZD mice show deficits in nest building behavior. N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. (B) Exemplary images showing the built nest after 24 h after introduction of nesting material from CTRL (upper image) and PZD (two lower images) mice. ^*p ≤ 0.05.

Figure 4

PZD animals show altered behavior during the three chamber test, olfactory habituation test, and reciprocal social interactions. (A,B) No innate side preference was detected in CTRL and PZD mice in the habituation phase given that no differences were detected in the time spent in a chamber the number of transitions (B). (C) Exemplary tracking path of a CTRL and PZD mouse in the test for social novelty. (D–F) Normal sociability was found in PZD mice. Both, male and female PZD mice spent significantly more time in the chamber containing a stranger mouse vs. the empty wire cage (D) and sniffing the wire cage containing a stranger mouse (E). (F) No significant difference in the number of transition was detected in this phase of the test. (A–F) No gender specific effects were detected. (G–I) In the second part of the test however, PZD male show a significant preference for stranger 2 in terms of time spent in the chamber with stranger 2 (G), and the time spent sniffing at stranger 2 (H). In contrast, female PZD mice do not display a significant preference for stranger 2 in terms of time spent in the chamber with stranger 2 (G). (H) The time spent sniffing at stranger 2 was not altered in PZD female mice. (I) No significant difference in the number of transition was detected in this phase of the test. (A–I) N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. (J,K) PZD mice were exposed to distilled water and four different odors consecutively presented three times for 2 min. The sniffing behavior of the animals in response to non-social odors: almond and banana extract, and social odors: animals of the same sex from the strains C3H/HeNRj (Social C1) and C57BL/6jRi6 (Social B1) was recorded and analyzed. The mean of the cumulative time the animals spent sniffing at the presented odor is presented. Abnormal sniffing times compared to controls can be seen regarding social odors in PZD males (J). (J,K) N = 10 males and 10 females (CTRL) and n = 11 males and 11 females (PZD) were analyzed. (L–O) Reciprocal social interactions of male—male and female—female pairings were evaluated for 4 min. (L) Example chronogram representing all the labeled events of the data shown in (B) comparing female control and PZD mice over the 4 min of experimentation. Rows represent individual mice. Single events of contact are shown as lines. (M) The average time of contact between mice was quantified. No significant differences between CTRL and PZD were detected. (N) Oral—oral contact of mice was quantified and the average time spent in contact is shown. A significantly reduced contact time is seen specifically in PZD female mice compared to female controls and PZD male mice. (O) Increased aggression of PZD mice can be observed quantifying the average time of the resident chasing the intruder that is significantly increased for PZD mice without gender specific effect. (L–O) N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. ^*p ≤ 0.05; ^**p ≤ 0.01; ^***p ≤ 0.001.

Figure 5

PZD mice show altered vocalization. Reciprocal social interactions of male—male and female—female pairings were evaluated for 4 min. (A,B) Ultrasonic vocalizations were recorded during the first 2 min of the test for reciprocal social interactions. (A) The latency to call was significantly increased in PZD animals. Gender specific analysis shows no gender specific effect. (B) The total number of calls was not reduced. N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. ^*p ≤ 0.05.

Figure 6

PZD mice show impaired motor learning. (A–C) Motor learning task using a Rotarod Mice were tested with two training trials per day. (C) The latency to fall was significantly different between CTRL and PZD mice. N = 22 CTRL animals and n = 24 PZD mice were analyzed. ^*p ≤ 0.05.

Figure 7

PZD mice show only minor differences in spontaneous alternation and repetitive behavior. (A–C) Spontaneous alternation behavior in the Y-maze labyrinth, a hippocampus dependent task of spatial working memory. (A) No significant differences in the latency to leave the starting arm were observed. (B) A trend (#) toward a decrease in the percentage of alternation during the 5 min test session in the Y maze labyrinth was detected in PZD mice. (C) The mean number of entries was not altered. (A–C) No gender specific effects were observed in the Y-maze test. (D) PZD mice show no significant differences in the time spent self-grooming independent of gender. N = 24 CTRL animals (12 males and 12 females) and n = 25 PZD mice (12 males and 13 females) were analyzed. (E,F) Repetitive marble burying behavior was measured in PZD mice after a 30 min test session. PZD females buried significantly less marbles in comparison to control mice. N = 20 CTRL animals (10 males and 10 females) and n = 22 PZD mice (11 males and 11 females) were analyzed. ^*p ≤ 0.05.