Constitutive AP2γ deficiency reduces postnatal hippocampal neurogenesis and induces behavioral deficits in juvenile mice that persist during adulthood

The transcription factor activating protein two gamma (AP2γ) is an important regulator of neurogenesis both during embryonic development as well as in the postnatal brain, but its role for neurophysiology and behavior at distinct postnatal periods is still unclear. In this work, we explored the neurogenic, behavioral, and functional impact of a constitutive AP2γ heterozygous deletion in mice from early postnatal development until adulthood. Constitutive AP2γ heterozygous deletion in mice caused a reduction of hippocampal transient amplifying progenitors (TAPs) in the postnatal brain, inducing significant impairments on hippocampal-dependent emotional- and cognitive-behavioral tasks including anxiety-like behavior and cognitive deficits, typically associated with an intact neurogenic activity. Moreover, AP2γ deficiency impairs dorsal hippocampus-to-prefrontal cortex functional connectivity. We observed a progressive and cumulative impact of constitutive AP2γ deficiency on the hippocampal glutamatergic neurogenic process, as well as alterations on limbic-cortical connectivity, together with impairments on emotional and cognitive behaviors from juvenile to adult periods. Collectively, the results herein presented demonstrate the importance of AP2γ in the generation of glutamatergic neurons in the postnatal brain and its impact on behavioral performance.

somatic and neurobiological paraments during the first 21 postnatal days (Supplementary Figure   139 2). Despite a variation in the eye-opening day, responsiveness in sensory-motor functions, 140 vestibular area-dependent tasks, and strength, as well somatic parameters were similar in WT 141 and AP2g +/mice. Furthermore, all analyzed parameters were within the previously described 142 range (Guerra-Gomes et al., 2020;Heyser, 2003). These observations suggest that constitutive 143 heterozygous deletion of AP2g has no impact on early postnatal development.

144
In juvenile mice (between PND 25-31), we performed the open-field (OF) test to assess 145 locomotor and anxiety-like behavior, tail suspension test (TST) and sucrose splash test (SST) to 146 assess behavioral despair and anhedonic-like behavior, and the object recognition test (ORT) to 147 assess memory (Figure 2A). In the OF, juvenile AP2g +/mice exhibited a lower distance traveled  Figure 2D). Notably, despite 154 no differences in the novel object location ( Figure 2E and F), AP2g +/mice displayed significant 155 deficits in the novel object recognition, as denoted by a decreased preference to explore the novel 156 object ( Figure 2G). tend to persist in adult mice, with no alteration in behavioral despair.
In the CFC, a behavior test described to be sensitive to changes in adult hippocampal 178 neurogenesis (Gu et al., 2012), mice were subjected to two distinct context tests, aimed to test 179 hippocampal-dependent memory, and a cue probe to assess the integrity of extrahippocampal 180 memory circuits ( Figure 3H) (Gu et al., 2012;Mateus-Pinheiro et al., 2017). In the context A,

181
AP2g +/mice exhibited reduced freezing behavior when exposed to a familiar context ( Figure 3I).

182
No alterations in the freezing behavior were observed neither in the context B ( Figure 3J) nor in 183 the cue probe ( Figure 3K). These observations suggest that AP2g +/mice exhibit deficits in 184 contextual hippocampal-related memory, and an intact associative non-hippocampal-dependent 185 memory when compared to WT littermates. Furthermore, experimental groups were also 186 subjected to the MWM test for evaluation of spatial memory ( Figure 4). In the reference memory 187 task, that relies on hippocampal function integrity (Cerqueira et al., 2007), AP2g +/and WT mice 188 exhibit similar performance to reach the hidden platform along the training days ( Figure 4A and 189 B). When the platform was changed to the opposite quadrant to assess behavior flexibility, which 190 relies not only in the hippocampal formation but also in prefrontal cortical areas (Hamilton and 191 Brigman, 2015), adult AP2g +/mice spent less time in the new quadrant than WT animals ( Figure   192 4C) suggesting that constitutive AP2g deficiency leads to impaired behavioral flexibility. Detailed 193 analysis of the strategies adopted to reach the escape platform (Antunes et al., 2020;Garthe et 194 al., 2009;Garthe and Kempermann, 2013;Mateus-Pinheiro et al., 2017;Ruediger et al., 2012) 195 revealed that AP2g +/mice delayed the switch from non-hippocampal dependent ("Block 1") to

225
The electrophysiological studies revealed that constitutive and heterozygous AP2g 226 deficiency led to two outcomes: first, a significant decrease of coherence between the dHip and 227 the mPFC indicating impairments in the ability of these regions to functionally interact; second, 228 this decrease in interregional coherence was accompanied by a diminished neuronal activity in a 229 wide range of frequencies in the mPFC, including in theta and beta frequencies, previously shown 230 to be critically related with behavioral outputs dependent on cortico-limbic networks (Colgin, 2011; Herein, we show that despite a normal early postnatal acquisition of neurodevelopmental milestones, constitutive and heterozygous deficiency of AP2g induces an anxiety-like state and 236 causes cognitive deficits in mice that persist from adolescence until adulthood. Our results

237
suggest that AP2g plays a crucial role for the proper development and maturation of neural circuits 238 implicated in emotional and cognitive functions.

239
Newly generated neurons are highly relevant to hippocampal functioning and 240 hippocampal-associated behaviors (Anacker and Hen, 2017;Christian et al., 2014;Fang et al., 241 2018;Gonçalves et al., 2016). Impairments in adult hippocampal neurogenesis precipitate the 242 emergence of depressive-and anxiety-like behaviors (Bessa et al., 2009;Hill et al., 2015;Mateus-243 Pinheiro et al., 2013b, 2013aRevest et al., 2009;Sahay and Hen, 2007). Here, we assessed the     1997;Schorle et al., 1996), we sought to understand whether constitutive and heterozygous 266 deficiency of AP2g could lead postnatally to functional and behavioral impairments. The 267 developmental milestones protocol showed no impact of the constitutive heterozygous deficiency 268 of AP2g in early postnatal neurodevelopment. Nevertheless, this deficiency in AP2g promotes emotional and cognitive behavior in later periods of life. At juvenile age, AP2g deficiency led to 270 the manifestation of anxiety-like behavior and significant impairments in recognition memory where adult AP2g +/mice displayed poor performances in hippocampal-dependent tasks (Garthe 274 et al., 2009;Garthe and Kempermann, 2013;Gu et al., 2012;Ruediger et al., 2012). Notably, 275 conditional deletion of AP2g in adulthood lead to a less evident effect on emotional behavior, 276 namely in anxiety-like behavior tested in the OF and EPM behavioral tests, when comparing with 277 the constitutive mice model herein presented (Mateus-Pinheiro et al., 2017). This result indicates 278 that constitutive deficiency of AP2g may exert a longitudinal cumulative impact leading to more 279 severe alterations in behavioral performance of mice, whereas in the conditional model during 280 adulthood, the AP2g deletion only occurs in a subset of newly formed neuroblasts.

281
Our results are consistent with previous publications were the suppression of the TAP's 282 regulator Tbr2 exerted both an anxiety-like phenotype during the juvenile period, and induced 283 cognitive deficits during early adulthood (Veerasammy et al., 2020). Moreover, Ngn2 is also 284 important for the modulation of cognitive behavior, namely in the rescue of cognitive function in 285 the T-Maze task. Interestingly, the regulation of TAPs' by AP2g seems to be also important for the preservation of cognitive performance as shown by its impact on hippocampal-dependent tasks.

287
Behavioral flexibility, a cognitive task that relies on the interaction of the hippocampal and 288 prefrontal cortical brain areas, was impaired in AP2g +/mice. Adult AP2g +/mice present significant 289 deficits of electrophysiological coherence between the dHip and the mPFC. In particular, 290 constitutive AP2g deficiency led to a decrease of the spectral coherence between the recorded 291 brain areas in a wide range of frequencies, previously associated to behavior outputs dependent 292 on cortico-limbic networks (Colgin, 2011;Fell and Axmacher, 2011;Oliveira et al., 2013;Sardinha 293 et al., 2017). The integrity of the hippocampus-to-PFC circuitry was described to be relevant for 294 example to the action of antidepressants, such as ketamine (Carreno et al., 2016), which promote 295 neurogenesis, suggesting that AP2g may be involved in conserving this neuronal circuit.

352
to detect a gradual progression in performance as for walking, surface righting, grasping, negative pad, and immediately after testing, the pups were returned to their home cage.
animal did not respond within 30 s, the test was ended. Mature response was achieved when the 361 pups were able to get right in less than 1 s for three consecutive days.

362
Negative geotaxis -PND1 to PND 14 -Pups were placed head down in a horizontal grid, 363 tilted 45º to the plane. The acquisition of a mature response was set when pups were able to head 364 up in less than 30 s for three consecutive days.

365
Air righting -PND8 to PND 21 -In this test, the pup was held upside down and released 366 from a height of approximately 13 cm from the soft padded surface and released. A mature 367 response was obtained when the animals landed on four paws for three consecutive days.

368
Cliff aversion -PND 1 to PND 14 -It evaluates the mouse pup's ability to turn and crawl 369 away when on the edge of a cliff. A mature response was achieved once the animal moved away 370 in less than 30 s for three consecutive days.

371
Postural reflex -PND 5 to PND 21 -Pups were placed in a small plastic box and gently 372 shaken up down and right. When the animals were able to maintain their original position in the 373 box by extending four paws, mature response was acquired.

374
Wire suspension -PND 5 to PND 21 -This test evaluates forelimb grasp and strength.

375
Pups were placed vertically to hold with their forepaws a 3 mm diameter metal wire suspended 5 376 cm above a soft foam pad. A mature response was achieved once the animal was able to grasp 377 the bar, holding it with four paws.

378
Grasping -PND 5 to PND 21 -The mouse pup forelimb was stimulated with a thin wire 379 to evaluate when the involuntary freeing reflex stopped. This reflex disappears with the 380 development of the nervous system, as so, the mature response achieved when the animal 381 grasped immediately and firmly the wire.

384
Ear twitch -PND 7 to PND 15 -In this test, the mouse pup ear was gently stimulated with 385 the tip of a cotton swab, three times. If the animal reacted, flattening the ear against the side of 386 the head for three consecutive days, the mature response was reached.