Very preterm-born children may suffer from significant deficits in executive function, processing speed, and intelligence (Brydges et al., 2018). The risk for neurodevelopmental disorders such as attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), and anxiety are increased two to four times in preterm children, as compared to term-born controls (Rogers et al., 2018). Altered GABAergic signaling has been implicated in the pathogenesis of ASD (Gaetz et al., 2014; Tanifuji et al., 2017), ADHD (Naaijen et al., 2017), as well as anxiety (Nuss, 2015). Pharmacological antagonization and agonization of γ-aminobutyric acid (GABA) receptors in the neonatal animal model of the mouse during postnatal brain development corroborate an important role of GABA and their receptors in programming neurobehavioral phenotypes in adulthood (Salari and Amani, 2017).

While human brain growth reaches its highest velocity at birth (Watson et al., 2006), the so-called “brain growth spurts” in rodents peak at the seventh postnatal day (Semple et al., 2013). Postnatal rat pups may therefore serve as a model for the human stage of brain development, corresponding to the last trimester of pregnancy. Both in humans and rodents, the hippocampus undergoes developmental changes close to birth (Semple et al., 2013). During the formation of the dentate gyrus (DG) neural progenitor cells (NPC) are generated and form a proliferative zone that remains active during postnatal stages, becoming the site of adult hippocampal neurogenesis called the subgranular zone (SGZ; Paridaen and Huttner, 2014). Supplemented new neurons form highly complex neural circuits, supporting a role for hippocampal neurogenesis in memory, learning, and behavior (Deng et al., 2010; Anacker and Hen, 2017).

The last trimester of a human pregnancy and early postnatal life involves a period of brain development with neuronal organization and maturation, such as neurogenesis, migration, dendritogenesis, synaptogenesis, and plasticity of developing neurons. These processes are regulated by neurotransmitters such as GABA and glutamate, which remain at risk of disruption after preterm birth (Malik et al., 2013). In humans, the phase of rapid brain growth starts at gestational week 28, and peaks at the time of birth. In the rat model, this phase occurs from postnatal day (P)4 to P11 and peaks at P7 (Dobbing and Sands, 1979; Semple et al., 2013) which makes early postnatal pups useful as model organisms in studies of human neuronal development, corresponding roughly to the last trimester of pregnancy. These neurodevelopmental processes are highly vulnerable and clinically relevant, as they can be affected by oxidative stress (hyperoxia, hypoxia) or various necessary medical interventions (Malik et al., 2013; Steinhorn et al., 2015; Duerden et al., 2016; Isokawa, 2016). In addition to neurotransmission and developmentally mediated excitatory-inhibitory transition of GABA action during the perinatal period (Ben-Ari, 2018), GABA and its receptors assume multiple essential functions during fetal and postnatal brain development (Cellot and Cherubini, 2013; Wu and Sun, 2015; Tang et al., 2021). In the rat, the main generation of hippocampal granular cells starts around birth and peaks during the first postnatal week (Altman and Bayer, 1990). In line with this, GABAergic transmission changes from excitatory to inhibitory also during the end of the first postnatal week (Rivera et al., 1999; Khirug et al., 2005). Neurobehavioral disorders, including autism, are more common in survivors of preterm birth and have been associated with decreased GABA concentrations, underscoring the importance of in vivo examination of GABA changes during early postnatal life in preterm infants (Ream and Lehwald, 2018; Basu et al., 2021). Peerboom and Wierenga (2021) postulated that the postnatal shift from depolarizing to hyperpolarizing GABA is a pivotal event in brain development and its timing affects brain function throughout life. Altered timing of the postnatal GABA shift is associated with several neurodevelopmental disorders (Schulte et al., 2018). In addition, preterm birth itself, as well as pharmacologic drugs used in the preterm infant, influence GABA receptor associated pathways (Shaw et al., 2015; Steinhorn et al., 2015). Extremely and very preterm infants showed reduced GABA concentrations in the brain measured by magnetic resonance imaging (Kwon et al., 2014; Basu et al., 2020). The developmental timeline of the GABAergic system becomes more relevant in preterm infants whose developing brains are prematurely exposed to extra uterine stress, and predisposed to neurological disorders, perhaps in part as sequela of early derangement in GABAergic systems.

Despite the postnatal developmental differences in GABAergic signaling, there are many similarities in the generalized course of neural maturation in early development and adulthood (Song et al., 2012). A complex interaction of the intrinsic programs of neuronal stem cells (NSCs) and their progressively produced progenitors (NPC) regulates neurogenesis, which is orchestrated by intrinsic pathways and extracellular signaling molecules (Faigle and Song, 2013; Bjornsson et al., 2015). As shown schematically in Figure 5, hippocampal neurogenesis originates from NPC and leads to granule cell neurons, which go through different stages leads to granule cell neurons after progressing through the stages of NSC/type-1 cells, NPC/type-2a cells, neuroblast/type-2b cells, immature-mitotic neuron/type-3 cells as well as postmitotic-immature and mature granular neurons (Kempermann et al., 2004). NSCs are the shared NPC of both neurons and astrocytes and therefore express the astrocytic markers glial fibrillary acidic protein (GFAP) and Scl1a3 as well as the neuronal marker nestin (DeCarolis et al., 2013; Berg et al., 2018; Vieira et al., 2018). NSCs express Sox2 to maintain their multipotency and proliferation capacity (Mercurio et al., 2019). Sox2 represses the expression of NeuroD1 and therefore prevents the cells’ progression in neurogenesis, preserving their self-renewal capacity (Kuwabara et al., 2009). A key regulator of the mainstay of NSCs after their transition to asymmetric neurogenic division is the Notch target gene Hes5 (Lugert et al., 2010). By directly changing the expression of genes associated with self-renewal and differentiation [e.g., Sox2 (Wen et al., 2008), Ngn2, and NeuroD1 (Scardigli et al., 2003; Shimojo et al., 2008)], Pax6 is essential to regulating the NPCs’ proliferation (Maekawa et al., 2005). NSCs give rise to intermediate progenitor cells, which express as a specific marker Tbr2. Tbr2 labels type-2 cells, while Tbr1 is expressed by immature granule neurons (Englund et al., 2005; Hodge et al., 2008; Nicola et al., 2015). Additionally, type-2a cells express the proneural markers Ascl1 (also known as Mash1) and Ngn2 (Amador-Arjona et al., 2015; Pérez-Domínguez et al., 2018). At the late stage of typ-2a cells, Ngn2 is downregulated, whilst Tbr2 expression in typ-2b cells persists (Roybon et al., 2009). Type-2b cells start to express NeuroD1, a crucial transcription factor for neurogenesis during hippocampal development that marks the transition from amplifying progenitor to neuroblast (Kuwabara et al., 2009). NeuroD1 is necessary for further survival and maturation of neurons (Pérez-Domínguez et al., 2018). Type-3 cells are less proliferative and may migrate and exit the cell cycle before full maturation into granule neurons (Nicola et al., 2015). NeuroD2 starts to be expressed just after NeuroD1 and continues to be highly expressed in postmitotic mature neurons (Roybon et al., 2009). Postmitotic neurons following further differentiation express as specific markers NeuN, Tbr1 (Englund et al., 2005) and NeuroD2, of which the latter is necessary for cell cycle regulation and survival of neurons (Olson et al., 2001; Wilke et al., 2012). Granule cell maturation necessarily depends on Prox1 expression, which starts with type-2b cells and is maintained further in differentiation (Lavado et al., 2010). Different types of signals, including glutamatergic and GABAergic signals from local neural networks, mediate these complex neurogenic processes.

The neurotransmitter GABA, in tandem with the neurotransmitter glutamate and its receptors, is essential for balancing excitation and inhibition and subject to various changes during pre- and postnatal development (Takesian and Hensch, 2013). The ionotropic GABA_A receptor and the metabotropic GABA_B receptor are two main subtypes of GABA receptors. Berg et al. (2013) postulated that in addition to the progression of neural stem cells into functionally integrated mature neurons, cell cycle regulation and cell differentiation might be part of the functions of neurotransmitters. Various neurotransmitter receptors are expressed on different neuronal cell types (Pocock and Kettenmann, 2007). The functional roles for GABA during adult neurogenesis are already well described in part (Ge et al., 2006; Giachino et al., 2014; Song et al., 2016; Catavero et al., 2018), but it is poorly understood which steps in hippocampal neurogenesis at the time of brain growth spurts are vulnerable to disturbance of GABAergic signaling. Since the referenced evidence suggests that modulation of the GABA receptors during early life induces behavioral abnormalities in later life, we hypothesized that neonatal pharmacological blockade of GABA_A and GABA_B receptors during excitatory to inhibitory switch of GABA signaling may alter neuronal proliferation, differentiation, and maturation of hippocampal neurons in newborn rat pups.

In the present study, we investigated the effects of a GABA_A and GABA_B receptor antagonist on postnatal hippocampal neurogenesis in newborn rats, during the developmental depolarizing-to-hyperpolarizing switch in response to the neurotransmitter GABA (Ben-Ari, 2018). Our data suggest that antagonizing GABA_B receptor activity impaired the progression of neural progenitors to mitotic immature neurons, reduced the expression of neurotrophins, and increased the proliferation capacity. In contrast, antagonizing GABA_A receptors did not affect postnatal hippocampal neurogenesis at a cellular level.

GABA is one of the most abundant neurotransmitters in the central nervous system. GABAergic neurons are the main source of GABA; additionally, synthesis occurs in glial cells (Angulo et al., 2008; Héja et al., 2012). The subtypes of GABA_B receptors, GABA_B1a/b and GABA_B2, were shown to be located both pre- and postsynaptically, indicating a more modulatory function in the developing brain (López-Bendito et al., 2004; Evenseth et al., 2020). Various experimental models suggest that GABA_B receptor antagonists modulate hippocampal-linked learning, memory, and behavior (Cryan and Kaupmann, 2005; Heaney and Kinney, 2016).

During neuronal lineage development, intermediate progenitor cells have been described to successively express a cascade of transcription factors starting with Pax6 expression in type-1 and type-2a cells, and further involving Ngn2, Tbr2, NeuroD1, Tbr1, and Prox1, which is expressed from the stage of type-2b cells onwards (see schematic overview in Figure 5; Gonçalves et al., 2016; Pérez-Domínguez et al., 2018; Hevner, 2019). First, GABA_B receptor blockade reduced the total Nestin+ and proliferating Nestin+ cells in the DG, depending on concentration. Nestin is widely used as a marker for proliferating neuroepithelial and progenitor cells as well as some astrocytes in the hippocampus (Von Bohlen und Halbach, 2011; Wilhelmsson et al., 2019). Further, Nestin is essential for the survival and self-renewal of NSC (Park et al., 2010). Quiescent type-1 NSC and active NSC differ in frequency of proliferation, while both cell types share the characteristic expression of the glial cell marker GFAP and neural progenitor marker Nestin (Lugert et al., 2010). Neither GFAP nor the astrocytic markers SCL1a3/GLAST were impaired in gene expression by the antagonization of the GABA_B receptor. Sox2-expressing cells in the SGZ generate differentiated and identical cells, indicating their multipotent properties and importance for self-renewal (Suh et al., 2007). Similar to Sox2, Hes5-positive cells specifically inhabit the SGZ, with the property of NPC in the neurogenic niche (Lugert et al., 2010). Both transcription factors of type-1 cells, Sox2 and Hes5, were not affected by the blockade of GABA. This is in line with the thesis that type-1 NSCs are not affected as a whole, while the proliferating subpopulation of type-1 NSC depends on GABAergic signaling.

It is interesting that Pax6, classifying newly born cells from the differentiation state in SGZ and regulating self-renewal of NSC (Maekawa et al., 2005), was significantly reduced by the performed GABA_B receptor blockade. The impact on Nestin+ NSC and Pax6 expression suggests that GABA_B receptor-mediated GABA signals may be central in determining the neuronal fate and proliferation of NSCs. NSCs self-renew to generate asymmetrically a new daughter NSC and a daughter NSC with properties of intermediate NPC (type-2 cells; Gonçalves et al., 2016). The maintenance of pluripotency in NPC by Sox2 is well-defined (Favaro et al., 2009). Sox2 binds to the regulatory regions of a number of genes involved in neuronal differentiation (Lodato et al., 2013), including proneural genes such as Ascl1 (Vasconcelos and Castro, 2014) and NeuroD1 (Kuwabara et al., 2009). The substitution of Sox2 by neurogenic signals at the regulatory promotor regions is required for activation of proneural genes and therefore downregulation of Sox2 a precondition for further neuronal differentiation (Kuwabara et al., 2009). GABA antagonists did not interfere with the level of Sox2 expression in the postnatal DG.

One step further in neuronal differentiation, changes caused by GABA receptor blockade were detectable. Mediators of intermediate progenitor cells, Ascl1, NeuroD1 as well as proliferating NeuroD1+ cells were significantly decreased under GABA_B receptor blockade. Ascl1 is a proneural transcription factor and is expressed in mitotic NPC (Uda et al., 2007). The transcription factor NeuroD1, which may act as a neuronal determination gene, is expressed in the middle and late stages of NPC (Von Bohlen und Halbach, 2011). Interestingly, Ascl1 is able to play two opposing roles. On the one hand, it promotes proliferation and on the other supports cell-cycle exit and differentiation (Andersen et al., 2014). Ascl1 enables the transcription of multiple proneuronal target genes (Vasconcelos and Castro, 2014). A decreased expression of Ascl1 after GABA_B receptor blockade might explain the observed delay in NPC differentiation. This does not necessarily lead to decreasing proliferation per se. Although the proportion of mitotic maturing neurons (NeuroD1+/PCNA+ cells) was reduced, the overall proliferation capacity (PCNA+ cells, CycD2) was induced. This could be due to Ascl1’s interaction with Notch target genes in neural precursor cells via downstream mediators (Andersen et al., 2014), similar to Hes5 (Lugert et al., 2010). Hes proteins activated via the Notch signaling pathway act as repressors of proneural transcription factors. Due to their self-regulation and short half-lives, the cellular expression of Hes proteins is not constant and thus leads to fluctuating expression of downstream target genes, such as Ngn2 and Ascl1 (Lugert et al., 2010; Vasconcelos and Castro, 2014). Expressions that can be changed in this way could promote proliferation, while steady conditions lead to the progression of differentiation.

Current measurement technology at a certain point is not sufficiently capable of detecting oscillating expressions of these mediators. However, our findings support the hypothesis that NeuroD1, which is directly regulated by Pax6 (Thakurela et al., 2016), and Ascl1 are significantly reduced in expression, while proliferation is increased. Typically, NeuroD1 expression peaks in early neuroblasts/type-2b and type-3 cells. Diminished NeuroD1 expression seems to be partly responsible for the stagnation of differentiation and progression of neurogenesis (Gao et al., 2009).

A possible explanation for the observed general boost in proliferation capacity beyond the transcriptional regulation of hippocampal neurogenesis might be GABAergic signaling itself. GABAergic signals through GABA receptors as well as the loss of GABA_A and GABA_B receptor subtypes were described to increase proliferation of NSC in the adult DG (Song et al., 2012; Giachino et al., 2014). GABA_B receptor function is associated with suppressed proliferation of adult hippocampal neurogenesis (Felice et al., 2012). The impact of GABA_B receptor blockade already in postnatal development with excitatory GABA on NPC associate transcription factors suggests that GABA_B receptor-mediated signals may be important for initiating transcription of proneural mediators and generate mitotic neuroblasts.

Nevertheless, mitotic immature neuroblasts/type-2b cells express Tbr2 (Hodge et al., 2012). Before the neuroblasts exit the cell cycle to become mature granular neurons, they start expressing Tbr1, Prox1, and NeuroD2 (Gonçalves et al., 2016). Downregulation of Tbr2 is the precondition for NPC to differentiate into postmitotic immature neurons and this coincides with upregulation of Tbr1, which is strongly associated with new mature neurons (Hodge et al., 2012). Late neuroblasts/type-3 cells and postmitotic granular neurons express the transcription factor Prox1 (Li et al., 2009). NeuroD2 induces a neural phenotype, while its possible involvement in terminal differentiation is still being investigated (Ravanpay et al., 2010). Considering the crucial tasks of these transcription factors for final neurogenic termination, a significant downregulation of gene expression would suggest a reduction in the number of mature neurons. Controversially, GABA_B receptor antagonization sustainably reduced the expression of relevant transcription factors (Prox1, Tbr1, NeuroD2), but did not affect the NeuN+ neurons in DG.

Prox1 was applied as a dentate granule neuron lineage marker (Rubin and Kessaris, 2013). Several studies (Lavado et al., 2010; Iwano et al., 2012) suggested Prox1 to act pleiotropically and modulate multiple targets as well as signaling pathways possibly through dysregulation of Notch signaling. Postnatal deletion as well as a conditional knockout of Prox1 in mice results in decreased counts of NPC subtypes (type-2a/b cells) and NSC/ type-1 cells (Lavado et al., 2010). Further, overexpression of NeuroD1, but not of either Ngn2 or NeuroD2, was able to drive the exclusive production of new neurons in the adult hippocampus (Richetin et al., 2015). A milieu dominated by the differentiation-driving effects of NeuroD1 might overlay the stage-specific action of NeuroD2. Particularly since NeuroD2 is required for the survival of hippocampal neurons (Olson et al., 2001), one might expect apoptosis-associated mediators to increase; however, in our study neither the expression profiles of apoptosis-associated mediators, effector caspase 3 (Casp3), and apoptosis-inducing factor (AIF), nor the total DAPI+ cell counts showed changes after switching off GABA receptor signaling compared to the control animals (see Supplementary Figures 1A,B).

The temporal delay in the differentiation of the NPC in our model of GABA_B receptor blockade is in line with a hypothesis formulated by Lugert et al. (2012). Instead of the amplification of type-2a intermediate progenitors as seen previously, though, they proposed that divisions of neural stem cells and early neuroblasts drive DG neurogenesis (Lugert et al., 2012). Mitotic NSCs in the DG provide neuronal progenitor expansion and initiate neurogenic differentiation, a process requiring more than 3 weeks under homeostatic conditions. After the initiation of neurogenesis, cells remain at the immature NPC developing stage for some weeks. In this phase, neuroblasts adapt to physiological or pathophysiological stimuli that may affect their maturation and differentiation. The NPC population reacts to specific noxae and, if required, rapidly initiates the production of many neurons (Dranovsky et al., 2011; Lugert et al., 2012). The hypothesis that NPCs remain in an immature state for a long period and retain their ability to expand neuronal cells and continue neurogenesis in order to quickly provide missing neurons when needed could support our observation of an unchanged population of mature neurons after GABA receptor blockade despite massive impairment of intermediate neurons. A hypothetical subsequent effect on the proportion of mature neurons resulting from a reduced differentiation capacity of maturing intermediary neurons did not occur at time P15 (see Supplementary Figure 2).

The described impact of transcription factors associated with mature granular neurons suggests that GABA_B receptor-mediated GABA signals might be important for the maintenance of the NPC cell pool by the initiation of pleiotropic interactions. Since glial markers were not influenced by GABA_A or GABA_B receptor antagonization, only the neural line seems to be affected. Interestingly GABA_A receptor blockade did not lead to any changes similar to those caused by GABA_B receptor blockade. Instead, it resulted in an overexpression of Tbr2 and Tbr1. There seem to be essential differences regarding the influence of GABA and its mediation via the respective receptors. Considering the GABA receptor density, a significant increase of GABA_A receptor occurs between P10 and P20, while the GABA_B receptor density peaks between P0 and P10 at P5 (Behuet et al., 2019). Due to the animal model we used, an effect of GABA blockade via the GABA_B receptor seems more probable. GABA-mediated effects via the GABA_A receptor require further investigation.

Intrinsic factors including neurotransmitters (Catavero et al., 2018), hormones (Triviño-Paredes et al., 2016), and neurotrophic factors (Lee et al., 2002) influence hippocampal neurogenesis. Neurotrophic factors, such as NT-3, NGF, and BDNF, are key players in the stimulation of NSC proliferation and differentiation (Ding et al., 2013). Interestingly, GABA_B receptor peak coincides with the time course of synaptogenesis, which is regulated by neurotrophins such as BDNF (Gaiarsa and Porcher, 2013). Coordinated effects of GABA and neurotrophins are important for the long–term survival of newborn neurons (Vilar and Mira, 2016). Waterhouse et al. (2012) suggested a reciprocal interaction between neurotransmitters and neurotrophic factors and showed that BDNF promotes progenitor cell differentiation and maturation by enhancing the release of GABA in the SGZ, resulting in increased differentiation and maturation of progenitor cells. Glial cells and GABAergic neurons are not sources for the secretion of neurotrophins themselves. BDNF is provided by pyramidal neurons or mossy fibers of hippocampal granular cells (Danzer and Mcnamara, 2004). BDNF is a crucial regulator of synapse development (Cohen-Cory et al., 2010) and shapes the development of neuronal circuits, as well as the construction of inhibitory connections throughout life (Gottmann et al., 2009). Similar findings were reported for NGF and NT-3, both involved in several critical processes in the developing brain (Shimazu et al., 2006; Frielingsdorf et al., 2007). We found that the inactivation of metabotropic GABA_B receptors downregulates the transcription of BDNF, NGF, and NT-3. Currently, studies on the effects of GABA_B receptor blockade on the neurotrophic response are limited. Overexpression of BDNF in hippocampal cultures results in an earlier maturation of inhibitory synapses by inducing both presynaptic and postsynaptic structural and functional modifications to enhance GABAergic transmission (Vicario-Abejón et al., 1998). Badurek et al. (2020) show that early disruption of Trkb signaling, the receptor for BDNF, from immature mouse hippocampal dentate granule cells affects the integration and maturation of newly formed DGCs in the hippocampal circuitry, and drives a premature shift from depolarizing to hyperpolarizing GABAergic actions. Conditional NT-3 knockout in mice goes along with an impairment of neuronal differentiation (Shimazu et al., 2006). Intracerebroventricular treatment with NGF increases hippocampal neurogenesis and enhances the survival of new neurons in young and aged rats (Frielingsdorf et al., 2007). Likewise, following GABA_B receptor activation and regulated secretion of BDNF increases the number of postsynaptic GABA_A receptor subunits (Kuczewski et al., 2011). This implies complex interactions between GABA_B receptor activation and neurotrophins to contribute to the functional and structural maturation of the developing hippocampus. A decreased neurotrophic response following GABA_B receptor antagonization appears to be consistent with the stagnant maturation of NPC. Neurotrophins may have important roles in hippocampal neurogenesis and developmental plasticity (Porcher et al., 2018). Imbalance in neurotransmission seems to lead to the reduction of neurotrophin transcription; therefore, further investigations on the precise mechanisms of the relationship between neurotrophic growth factors and GABAergic transmission are required.

Taken together, GABA influences the development and functioning of the GABAergic network mainly via GABA_B receptor during the first 10 postnatal days. As a result of the limited signal transduction of GABA via the GABA_B receptor, impairment of neurogenesis advanced, which did not affect the final termination to mature neurons, and diminished neurotrophin signals. This could imply that although antagonization of GABA_B receptors recruited quiescent cells to the active proliferative stem cell pool, progression towards terminal differentiation was not induced. This is in line with Giachino et al.’s (2014) proposal that GABA_B receptor activity controls the number of proliferating NPC in the adult hippocampus. In sum, GABA and the GABA_B receptors’ transmission affect neural stem cell and progenitor cell proliferation in the developing hippocampus.

The involvement of GABA receptors in neurogenesis was linked to the modulation of cognitive processes, like memory formation, executive function, learning, and intelligence. Imbalances during GABAergic transmission in neuronal circuits, such as drugs or oxygen, can affect these vulnerable phases of brain development and disrupt homeostatic control (Vertkin et al., 2015; Friedman and Kahen, 2019). Failure in neuronal homeostasis has been linked to pathophysiological mechanisms of various brain disorders (Marín, 2012; Kim and Yoon, 2017), like autism, hyperactivity, inattention, social and emotional incompetence, which are also associated with preterm birth (Fatemi et al., 2009; Arpino et al., 2010; Hashemi et al., 2017). For a deeper understanding of postnatal brain development and improving the understanding of GABA signaling in correspondence to preterm birth, further analysis will be required to establish which cues regulate the various stages of neurogenesis beyond NSC expansion, NPC proliferation, and maturation as well as the survival of newly formed neurons. These may open new therapeutic strategies to alleviate behavioral impairments and neurological disorders, perhaps in part as sequela of early derangement in GABAergic systems.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by LAGeSo, Berlin, Germany, approval number G-0075/18.

TScheuer and SE conceived the ideas. TScheuer designed the experiments. SE, CG, and TScheuer executed the experiments. CG and SE wrote the first draft of the manuscript and approved the final draft. TScheuer and SE performed the animal studies. CG performed and analyzed immunohistological staining and qPCR. TSchmitz and CB revised the manuscript and contributed to the critical discussion. All authors have contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.