The mammalian Slitrk family of proteins consists of six transmembrane proteins with two leucine-rich repeat domains (Slitrk1, Slitrk2, Slitrk3, Slitrk4, Slitrk5, Slitrk6; Aruga and Mikoshiba, 2003; Aruga et al., 2003). They are predominantly and differentially expressed in both immature and mature neural tissues in humans and mice (Aruga et al., 2003; Aruga and Mikoshiba, 2003; Beaubien and Cloutier, 2009; Stillman et al., 2009). They possess the ability to control neurite development (Aruga and Mikoshiba, 2003; Abelson et al., 2005; Katayama et al., 2009) and enhance synapse formation (Takahashi et al., 2012; Tekin et al., 2013; Yim et al., 2013; Beaubien et al., 2016; Kang et al., 2016; Schroeder et al., 2018; Biselli et al., 2021; El Chehadeh et al., 2022; Hatayama et al., 2022). However, these two abilities differ qualitatively and quantitatively among the six members (Aruga and Mikoshiba, 2003; Takahashi et al., 2012; Yim et al., 2013). In this review article, we focus on SLITRK1 and its involvement in the brain functions of health and disease, primarily in view of review articles dealing with the Slitrk family (Proenca et al., 2011; Ko, 2012; Won et al., 2019).

Among the human SLITRK genes, SLITRK1 was the first to be identified as a candidate genetic factor for neuropsychiatric disorders. In 2005, Abelson et al. identified a frameshift mutation (SLITRK1 L422fs) in an individual with Tourette’s syndrome (TS; Figure 1A; Table 1) and in a cohort of TS patients, inv.(13; q31.1; q33.1) and variants in the 3′ untranslated region (SLITRK1 var321) were enriched (Abelson et al., 2005). TS is diagnosed by the sustained presence of both vocal and motor tics, and represent the more severe end of the spectrum of tic disorders (Fernandez et al., 2018). Following the initial report, numerous subsequent studies surveyed SLITRK1 variants in independent TS patient groups. Many studies failed to identify the L422fs or var321 (Deng et al., 2006; Keen-Kim et al., 2006; Verkerk et al., 2006; Chou et al., 2007; Fabbrini et al., 2007; Orth et al., 2007; Pasquini et al., 2008; Scharf et al., 2008; Zimprich et al., 2008; Yasmeen et al., 2013). However, some studies have supported the association between SLITRK1 and TS (Miranda et al., 2009; Karagiannidis et al., 2012; Inai et al., 2015; Alexander et al., 2016). Overall, these results indicate the involvement of SLITRK1 in TS etiology in a small fraction of patients.

Conversely, SLITRK1 missense mutations have been reported in trichotillomania (R584K, S593G; Zuchner et al., 2006) and obsessive–compulsive disorders (OCD; N400I, T418S; Figure 1A; Table 1; Ozomaro et al., 2013). Trichotillomania is a chronic behavioral disorder characterized by the recurrent pulling of one’s own hair, leading to hair loss (Grant and Chamberlain, 2016). Obsessive–compulsive disorder is characterized by the presence of obsessions and/or compulsions (Stein et al., 2019). Obsessions are repetitive and persistent thoughts, images, impulses, or urges that are intrusive and unwanted, and are commonly associated with anxiety. Compulsions are repetitive behaviors or mental acts that the individual feels driven to perform in response to an obsession according to rigid rules, or to achieve a sense of “completeness” (Stein et al., 2019). Family and treatment studies have indicated that TS, trichotillomania, and OCD comprise a larger spectrum of conditions (obsessive–compulsive spectrum disorder or obsessive–compulsive and related disorders, hereafter OCRD; Ferrao et al., 2009; Leckman et al., 2010; Stein et al., 2019). In pharmacotherapy for OCRD, monoaminergic neurotransmission is targeted. TS responds to haloperidol (a dopamine D2 receptor antagonist) and clonidine (an adrenergic α2 receptor agonist; Fernandez et al., 2018). Selective serotonin reuptake inhibitors (SSRI) and clomipramines (serotonin–noradrenaline [NA] reuptake inhibitors) are used as first-line drugs for treating OCD (Stein et al., 2019). Clomipramine has some benefits in trichotillomania treatment as well (Hoffman et al., 2021).

In addition to OCRD, a functionally damaging mutation (A444S) was significantly enriched in a bipolar disorder (BPD) patient cohort (Figure 1A; Table 1; Hatayama et al., 2022). Furthermore, a genome-wide association study identified a single nucleotide polymorphism ca 780 kb upstream of SLITRK1, associated with schizophrenia (Bansal et al., 2018). Taken together, SLITRK1 appears to be a risk factor for OCRD and other neuropsychiatric diseases. Summarizing the SLITRK1 mutations that have been implicated in neuropsychiatric disorders, missense mutations were clustered in the second leucine-rich repeat domains of this protein (Figure 1A). This may indicate the pathological significance of the second LRR domain.

However, many missense mutations have been identified in SLITRK1 from not only disease cases, but also control subjects upon whole exome sequencing studies for bipolar disorder, schizophrenia, and epilepsy¹. This indicates that functional variations of SLITRK1 exist even in healthy individuals as well. In agreement with this idea, the S330A mutation (5 × 10⁻⁵ – 3 × 10⁻³ for both cases and controls) affected SLITRK1 function (Figure 1A; Table 1; Hatayama et al., 2022). S330A is a revertant of the A330S mutation that was acquired in SLITRK1 during the evolution of Homo sapiens from the common ancestor of Homo neanderthalensis (Table 1).

The phenotypes of Slitrk1 KO mice (Katayama et al., 2010; Hatayama et al., 2022) provided clues to consider the role of Slitrk1 at individual levels (Figure 2A). Slitrk1 KO male mice showed lower body weight at as early as P3 during development. The body weight difference stabilizes at one point (compared to control) and again becomes apparent after weaning, to later stages with a 9–14% lower body weight than that of WT mice at the same stage. Female KO mice showed a transiently lower body weight only at P14 (−17%). Isolation-induced ultrasonic vocalization calls were weak at P4 (males) and P7 (both sexes), and the calling rates were lower in females at P7 and P10. Thus, neurodevelopmental phenotypes exist in the neonatal stage in a sex-dependent manner.

Slitrk1 KO mice consistently exhibited decreased locomotor activity, which was manifested at 5 weeks of age (Katayama et al., 2010; Hatayama et al., 2022). The behavioral phenotype at the adult stage involves anxiety-and depression-like phenotypes (Katayama et al., 2010). The anxiety phenotype is characterized by reduced stay time in the open-field apparatus, reduced open-arm stay time, and enhanced freezing responses in fear conditioning tests (Katayama et al., 2010). Depression-like behavior is characterized by increased immobile time in the forced swimming and tail-suspension test (Katayama et al., 2010).

Other behavioral features, such as stereotypy, tremor, seizure, and abnormal repetitive behaviors, were not observed in the timed video recordings of Slitrk1 KO mice (Katayama et al., 2010). None of the responses exhibited abnormalities in the marble burying behavior, Morris water maze, or prepulse inhibition tests (Katayama et al., 2010). Innate reflexes, such as righting, pivoting, rooting, geotaxis, bar holding, grasping, visual place response, auditory startle, and tactile startle, appeared timely during development (Hatayama et al., 2022). Taken together, the neonatal body weight loss, neonatal vocalization abnormalities, and anxiety-like and depression-like behaviors in the adolescent and adult stages feature neurodevelopmental phenotypes in Slitrk1 KO mice (Figure 2A).

As many antidepressants and anxiolytics target proteins that control monoamine dynamics, monoamine abundance in Slitrk1 KO mice have been examined (Katayama et al., 2010; Hatayama et al., 2022).

In adult male mice, noradrenaline (NA) and its metabolite, 3-methoxy-4-hydroxyphenylglycol (MHPG), tended to be higher in the three brain regions of the preferential cortex, nucleus accumbens, and stratum in Slitrk1 KO mice. The NA content was significantly higher in the prefrontal cortex (PFC), and the MHPG in the nucleus accumbens, than in wild-type mice (Figure 2A). Further, the serotonin metabolite (5-hydroxyindole acetic acid) content was higher in the nucleus accumbens, and choline content was lower in the Slitrk1 KO striatum (Katayama et al., 2010). Administration of clonidine, an a2-adrenergic agonist that is frequently used to treat patients with Tourette’s syndrome and OCRD, attenuated the anxiety-like behavior of Slitrk1 KO mice (Figure 2A; Katayama et al., 2010), suggesting NA dysregulation to be associated with anxiety-like behaviors.

At the neonatal stage (P7), NA levels were higher in the PFC of male Slitrk1 KO PFC (Figure 2A), but the NA levels were comparable in the female PFC (Hatayama et al., 2022). Furthermore, the levels of the NA metabolite, MHPG decreased in both sexes (Hatayama et al., 2022).

In terms of monoaminergic fiber morphology, NA fiber density was increased two-fold in the P7 PFC of both male and female Slitrk1 KO mice (Figures 2C,D; Hatayama et al., 2022). The size of the LC increased at the same stage (Figure 2D; Hatayama et al., 2022). However, excessive NA innervation in the PFC was limited to the neonatal period and was unclear in the adolescent and adult stages (Hatayama et al., 2022). Therefore, Slitrk1 plays a role in the suppression of noradrenergic projections during the neonatal stage. However, the varicosities in serotonergic fibers are enlarged only in the male Slitrk1 KO PFC (Figure 2D), which is proposed to be due to the actions of excessive NA through the α2-heteroreceptor based on the results of clonidine treatment for neonates (Hatayama et al., 2022).

NA-dynamics are sexually dimorphic (Ngun et al., 2011), and this seems to be associated with the sex differences in vesicular monoamine transporter (VMAT2) function, where female mice possess greater striatal VMAT2 levels/activity (Dluzen and McDermott, 2008), as well as the sex-dependent role of glucocorticoid receptors in the noradrenergic system (Chmielarz et al., 2013). Furthermore, catechol O-methyl transferase (COMT), an NA and dopamine-metabolizing enzyme, exhibits sexual dimorphism (Gogos et al., 1998). Therefore, the role of Slitrk1 in NA fiber suppression is proposed to be primarily responsible for other sexually dimorphic phenotypes (excess NA and serotonergic varicosity enlargement).

Slitrk1 overexpression in the developing somatosensory cortex reduces NA fibers (Hatayama et al., 2022), supporting the idea that Slitrk1 suppresses NA fiber overgrowth in the neonatal cortex.

When LC cells were cultured from Slitrk1 KO newborn mice, they showed higher and lower neurite complexity, respectively, in the proximal and distal neurites, (Hatayama et al., 2022), indicating the cell-autonomous function of Slitrk1 in controlling LC projections. Conversely, addition of the Slitrk1 ECD domain to the culture medium inhibited neurite development similar to those of Slitrk1 KO-derived neurites (Figure 1C NA neurite growth; Hatayama et al., 2022), indicating the non-autonomous function of Slitrk1. The suppression of NA projections by Slitrk1 in the PFC seems valid because Slitrk1 is strongly expressed in the frontal cortex during early postnatal development (Figures 2B,C), and Slitrk1 ECD is produced in the brain by α-secretase-mediated cleavage (Kajiwara et al., 2009).

Besides the Slitrk1 ECD, Sema3a expression is reduced in the Slitrk1 KO PFC of mice in the search for deregulated neurite development-controlling proteins (Figure 1C NA neurite growth; Hatayama et al., 2022). In the LC culture, Sema3a proteins inhibited neurite growth in wild-type LC neurons, but not in Slitrk1 KO-derived LC neurons (Hatayama et al., 2022), indicating that Sema3a also acts as a suppressive signal for NA neuron development and that the cell-autonomous functions of Slitrk1 involves mediating Sema3a signals (Figure 1C NA neurite growth). Thus, there are two possible contact-points between Sema3a signaling and Slitrk1 functionality. One is Slitrk1-mediated enhancement of Sema3a expression, and the other is Slitrk1-mediated Sema3a signaling facilitation.

The molecular basis for Sema3a signaling facilitation has been hypothesized that Slitrk1 deprives the Sema3a/Nrp1/L1CAM complex of L1CAM and affects the signaling efficacy of Sema3a (Figure 1C Endocytosis). This is because L1CAM, a Slitrk1-binding protein, serves as an NRP1 (neuropilin1, a Sema3a receptor)-associated signal transducing transmembrane protein (Castellani et al., 2000; Bechara et al., 2008), and L1CAM increases Sema3a receptor endocytosis (Castellani et al., 2004). Furthermore, because Sema3a acts as both a receptor and ligand in the bidirectional regulation of Sema3a signaling (Jongbloets and Pasterkamp, 2014), the reduction of Sema3a expression in the Slitrk1 KO PFC could be affected in the reverse direction. Although this hypothesis needs further validation, the functional linkages among Slitrk1, L1CAM, and Sema3a signaling seem to uncover a novel molecular function of SLITRK family proteins.

Consequently, both secreted Slitrk1 ECD and Sema3a are thought to be suppressive of NA fiber growth downstream of Slitrk1 in the PFC. Both cell-autonomous (actions of Slitrk1 proteins in LC neurons) and non-cell-autonomous (actions of Slitrk1 proteins in PFC neurons) are presumed to explain this biological function.

During postnatal cortical development, the NA content transiently increases at the neonatal stage (NA surge; Figure 2A; mouse, P5–P7; rat, P0–P3; Levitt and Moore, 1979). During an NA surge, NA neurites extend rapidly in the PFC. Tangential NA fibers increases in the surface region (prospective layer I), where Slitrk1 is strongly expressed (Figures 2B,C). Such studies on Slitrk1 raised a possibility of the NA surge being the causal factor for some psychiatric disorders. Supporting this idea, many pharmacological or genetic studies have already shown the neuroplastic role of NA signaling in some neural circuits (Saboory et al., 2020).

Overall, sufficient evidence seems to support the role of NA signaling in neuroplasticity. Although excessive neonatal NA, as found in Slitrk1 KO exerts some neuroplastic roles, whether excessive neonatal NA signaling has pathological meaning depends on future studies. To clarify this point, the following points are important. First, the mechanism for NA fiber development during the neonatal NA surge should be clarified. It is probable that NA fiber growth in the cerebral cortex occurs in conjunction with corticogenesis. However, we still may not have sufficient results for the control of neonatal NA fiber development, as suggested by the Slitrk1 KO study (Hatayama et al., 2022). Additional molecular cues or signaling mechanisms for modulating NA fibers in the cortex require further studies. Second, excessive neonatal NA signaling should be modeled in experimental animals. Adaptation of optogenetic or chemogenetic methods for the neonatal animals in a less stressful environment would facilitate the analysis. Moreover, given the modular organization of the LC output (Chandler et al., 2019), the selective modulation of the modules would be more informative. Third, the significance of α2 hetero-receptor-mediated plasticity can be addressed by combining appropriate monoamine sensors (Feng et al., 2019; Wan et al., 2021). The interaction between NA signaling and serotonin signaling likely occurs during the neonatal period, as suggested by the Slitrk1 KO phenotype. Serotonergic signaling is critical for neonatal development and neuroplasticity (Brummelte et al., 2017). Fourth, the effects of the neonatal NA surge on disease-associated neural circuits should be clarified. For OCRD, recent studies have proposed specific neural circuits that could mediate cognitive and affective processing defects in patients with OCD. Globally, cortico-striatothalamocortical circuits could be involved in OCD (Stein et al., 2019). Cortico-striatothalamocortical circuits comprise of parallel and partly segregated circuits involved in sensorimotor, cognitive, affective, and motivational processes. Specifically, the cortico-striatothalamocortical circuits include the dorsomedial PFC and ventromedial PFC, which are partly related to the mouse medial PFC. Recent chemogenetic and optogenetic studies have shown the medial PFC to be a hub for both the depression-like and anxiety-like behavior-associated neural circuits (Xia and Kheirbek, 2020; Biselli et al., 2021). Fifth, the entity of NA-mediated neuroplasticity can be clarified using appropriate models. These could be persistent changes in the neuronal circuit structure or altered epigenetic signatures that cause long-lasting changes in gene expression of functional molecules.

LC-NA dysfunction has been implicated in many disorders, including depression, anxiety, attention-deficit hyperactivity disorder, post-traumatic stress disorder, Alzheimer’s disease, and Parkinson’s disease (Poe et al., 2020). However, NA dysfunction is not a known specific cause of any symptomatology or disease process (Poe et al., 2020). Studies on SLITRK1 in this article may have provided us with genetic clues to consider the pathophysiological significance of NA surges and the neuroplastic role of NA.

All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.

Many ideas in this manuscript was obtained through studies supported by Kakenhi funds (22H02722, 20K06927, 19H03327, and 16K07057).

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.