The KDET Motif in the Intracellular Domain of the Cell Adhesion Molecule L1 Interacts with Several Nuclear, Cytoplasmic, and Mitochondrial Proteins Essential for Neuronal Functions

Abnormal functions of the cell adhesion molecule L1 are linked to several neural diseases. Proteolytic L1 fragments were reported to interact with nuclear and mitochondrial proteins to regulate events in the developing and the adult nervous system. Recently, we identified a 55 kDa L1 fragment (L1-55) that interacts with methyl CpG binding protein 2 (MeCP2) and heterochromatin protein 1 (HP1) via the KDET motif. We now show that L1-55 also interacts with histone H1.4 (HistH1e) via this motif. Moreover, we show that this motif binds to NADH dehydrogenase ubiquinone flavoprotein 2 (NDUFV2), splicing factor proline/glutamine-rich (SFPQ), the non-POU domain containing octamer-binding protein (NonO), paraspeckle component 1 (PSPC1), WD-repeat protein 5 (WDR5), heat shock cognate protein 71 kDa (Hsc70), and synaptotagmin 1 (SYT1). Furthermore, applications of HistH1e, NDUFV2, SFPQ, NonO, PSPC1, WDR5, Hsc70, or SYT1 siRNAs or a cell-penetrating KDET-carrying peptide decrease L1-dependent neurite outgrowth and the survival of cultured neurons. These findings indicate that L1’s KDET motif binds to an unexpectedly large number of molecules that are essential for nervous system-related functions, such as neurite outgrowth and neuronal survival. In summary, L1 interacts with cytoplasmic, nuclear and mitochondrial proteins to regulate development and, in adults, the formation, maintenance, and flexibility of neural functions.


Introduction
The cell adhesion molecule L1 is involved in the regulation of multiple and diverse neural functions, such as proliferation, survival, and migration of neural cells; neuritogenesis, axonal outgrowth, fasciculation, and guidance of axons, as well as myelination and synaptogenesis during nervous system development (for reviews, see [1][2][3][4]). In adulthood, L1 modulates synaptic plasticity, learning, and memory (for a review, see [4]) and contributes to regeneration after injury by promoting axonal regrowth and remyelination (for references, see [5]). In humans, mutations in L1 are associated with neurological and psychiatric disorders, such as fetal alcohol syndrome, Hirschsprung's disease, schizophrenia, Alzheimer's disease, and L1 syndrome, which comprises a spectrum of mild to severe congenital X chromosome-linked developmental disorders [3,[6][7][8][9]. In mice, L1 deficiency can lead to severe malformations and malfunctions of the nervous system [10][11][12].
The beneficial functions of L1 in the nervous system not only depend on homo-and heterophilic interactions (for references, see [4,13]), but also on the proteolytic cleavage of L1 (for references, see, for instance, [14]). L1 consists of an extracellular N-terminal part comprising six immunoglobulin-like (Ig) domains and five fibronectin type III (FNIII)
Since recombinant SFPQ, NonO, PSPC1, and NDUFV2 bind to L1-ICD in ELISA [17,28], we performed competition ELISA with the KDET peptide to analyze whether binding of these proteins to L1-ICD is mediated by KDET. As a negative control, a peptide with a disrupted KDET motif was used. Disruption of the motif was achieved by the highly conserved substitution of K, D, E, and T by Q, N, Q, and S, respectively. We preferred this QNQS control peptide with conservative amino acid substitutions over a control peptide with a scrambled KDET sequence, because non-conservative scrambling could affect the primary structure much more than conservative substitutions, e.g., when K are exchanged by D, E, or T, when D or E is exchanged by K or T, and when T is exchanged by K, D, or E. In parallel, we tested whether L1-ICD binds to WDR5, TOP1, HistH1e, Hsc70, SYT1, ERα, RXRβ, PPARγ, AR, or VDR via the KDET motif. The interaction of L1-ICD with Nup93 and impβ1 was not analyzed, because no appropriate recombinant Nup93 and impβ1 proteins were available. The binding of L1-ICD to recombinant SFPQ, NonO, PSPC1 WDR5, NDUFV2, Hsc70, SYT1, HistH1e, and HP1α was reduced by the KDET peptide, but not by the negative control QNQS peptide, while the binding of L1 to TOP1, ERα, RXRβ, PPARγ, VDR, and AR was not affected by either peptide (Figure 2a). ELISA with substrate-coated recombinant HistH1e, WDR5, Hsc70, SYT1, or AR and different L1-ICD concentrations showed concentration-dependent binding of L1-ICD to WDR5, HistH1e, Hsc70, SYT1, and AR ( Figure 2b-f), indicating that the binding of WDR5, HistH1e, Hsc70, SYT1, and AR to L1-ICD is specific and suggesting that not only AR but also other nuclear receptors, e.g., ERα, RXRβ, PPARγ, and VDR bind to L1-ICD. For negative control, CHL1-ICD did not bind to WDR5, HistH1e, SYT1, and AR (Figure 2b-eSince Hsc70 is known to bind to CHL1-ICD via the HPD motif [30], ELISA with Hsc70 and CHL1 was not performed. The results indicate that the binding of SFPQ, NonO, PSPC1 WDR5, NDUFV2, Hsc70, SYT1, and HistH1e to L1-ICD is mediated by the KDET motif. AR (Figure 2b-f), indicating that the binding of WDR5, HistH1e, Hsc70, SYT1, and AR to L1-ICD is specific and suggesting that not only AR but also other nuclear receptors, e.g., ERα, RXRβ, PPARγ, and VDR bind to L1-ICD. For negative control, CHL1-ICD did not bind to WDR5, HistH1e, SYT1, and AR (Figure 2b-eSince Hsc70 is known to bind to CHL1-ICD via the HPD motif [30], ELISA with Hsc70 and CHL1 was not performed. The results indicate that the binding of SFPQ, NonO, PSPC1 WDR5, NDUFV2, Hsc70, SYT1, and HistH1e to L1-ICD is mediated by the KDET motif.  . L1-ICD binds to several binding partners via its KDET motif. Recombinant L1-binding partners were substrate-coated and incubated with a constant L1-ICD concentration in the absence or presence of the KDET peptide or QNQS control peptide (a) or with increasing L1-ICD concentrations (b-f). Binding was determined by ELISA using mouse L1 antibody C-2 and horseradish peroxidaseconjugated secondary antibodies. The mean values ± SD from three independent experiments carried out in triplicate are shown for the binding relative to control (values in the absence of peptides set to 100%). *** p < 0.005, **** p < 0.001; one-way ANOVA with Bonferroni s multiple comparison test.

The KDET Motif Is Essential for L1-Dependent Neurite Outgrowth and Neuronal Survival
Next, we investigated whether the KDET-mediated interaction of L1 with its binding partners plays a role in L1-dependent neurite outgrowth. Since treatment with L1 antibody 557 increases neurite outgrowth [19,21,22], we applied this antibody to cerebellar and cortical neurons in order to augment L1-dependent neurite outgrowth and to determine whether binding of L1 to its binding partners via the KDET motif is involved in this L1-stimulated enhancement of neurite outgrowth. Treatment with this antibody promoted neurite outgrowth in comparison to non-treated neurons (Figure 4a,b). This enhanced In cultured cortical neurons, L1 interacts with several binding partners via its KDET motif. Cultured cortical neurons were treated with vehicle, tat-KDET peptide, or tat-QNQS control peptide, followed by treatment without (a) and with (b) L1 antibody 557 and proximity ligation with a L1 antibody and an antibody against MeCP2, NDUFV2, SFPQ, NonO, PSPC1, WDR5, TOP1, HistH1, Nup93, Hsc70, SYT1, ERα, or PPARγ. Mean values + SD from two independent experiments are shown for the average numbers of red dots per cell relative to control (values of treatment with vehicle set to 100%) (**** p < 0.001; one-way ANOVA with Bonferroni s multiple comparison test).

The KDET Motif Is Essential for L1-Dependent Neurite Outgrowth and Neuronal Survival
Next, we investigated whether the KDET-mediated interaction of L1 with its binding partners plays a role in L1-dependent neurite outgrowth. Since treatment with L1 antibody Int. J. Mol. Sci. 2023, 24, 932 7 of 17 557 increases neurite outgrowth [19,21,22], we applied this antibody to cerebellar and cortical neurons in order to augment L1-dependent neurite outgrowth and to determine whether binding of L1 to its binding partners via the KDET motif is involved in this L1stimulated enhancement of neurite outgrowth. Treatment with this antibody promoted neurite outgrowth in comparison to non-treated neurons (Figure 4a,b). This enhanced neurite outgrowth was reduced in cerebellar and cortical neurons by treatment with the cell-penetrating tat-KDET peptide in a concentration-dependent manner (Figure 4a,b). The peptide did not affect neurite outgrowth from non-stimulated neurons (Figure 4a,b). Treatment of cerebellar and cortical neurons with the tat-KDET peptide, but not the tat-QNQS peptide, inhibited L1-induced neurite outgrowth from cerebellar and cortical neurons (Figure 4c,d). These results indicate that L1-mediated neurite outgrowth depends on the KDET-mediated interaction of L1 with its binding partners. Since treatment with antibody 557 also protects neurons against hydrogen peroxide-induced oxidative stress [31,32], we analyzed whether the KDET-mediated interaction of L1 and its binding partner is involved in neuronal survival. The hydrogen peroxide treatment increased cell death, which was reduced in the presence of antibody 557 ( Figure  4e). The antibody 557-mediated cell survival of cerebellar neurons was reduced by the tat-KDET peptide in a concentration-dependent manner (Figure 4e). The peptide did not affect hydrogen peroxide-induced cell death in the absence of L1 antibody 557 (Figure 4e). These results indicate that L1-mediated neuronal survival also depends on the KDET-mediated interaction of L1 with its binding partners. . The KDET-mediated interaction of L1 with its binding partners is essential for L1-dependent neurite outgrowth and neuronal cell survival. Cerebellar (a,c) and cortical (b,d) neurons were treated with 0, 0.5, 1, 2, 5, 10, 20, 50, or 100 µg/mL tat-KDET peptide (a,b), with 0 or 50 µg/mL tat-KDET peptide (P WT ) or with 0 or 50 µg/mL tat-QNQS peptide (P mut ) (c,d). Neurons were then treated without or with antibody 557. Mean values + SEM from three independent experiments are shown for total neurite lengths (**** p < 0.0001 relative to stimulated neurons not treated with tat-KDET peptide, § § § § p < 0.0001 relative to non-stimulated neurons not treated with tat-KDET peptide or tat-QNQS peptide; one-way ANOVA with Dunn's multiple comparison test). (e) Cerebellar neurons were first treated with 0, 2, 10, 20, 50, or 100 µg/mL tat-KDET peptide and then treated without or with antibody 557 in the absence or presence of H2O2. Mean values + SEM from three independent experiments are shown for the relative numbers of dead cells (**** p < 0.0001 relative to stimulated neurons not treated with tat-KDET peptide in the presence of H2O2, § § § § p < 0.0001 relative to non-stimulated neurons not treated with tat-KDET peptide in the presence of H2O2, #### p < 0.0001 relative to unstimulated neurons not treated with tat-KDET peptide in the absence of H2O2; one-way ANOVA with Dunn´s multiple comparison test).

Reduction of HistH1e, NDUFV2, SFPQ, NonO, WDR5, Hsc70, or SYT1 Expression Decreases L1-Dependent Neurite Outgrowth
To analyze whether the L1 binding partners HistH1e, NDUFV2, SFPQ, NonO, PSPC1, WDR5, Hsc70, or SYT1, which bind to KDET, are involved in L1-dependent neurite outgrowth, neurite outgrowth was determined after transfection of cortical neurons with . The KDET-mediated interaction of L1 with its binding partners is essential for L1-dependent neurite outgrowth and neuronal cell survival. Cerebellar (a,c) and cortical (b,d) neurons were treated with 0, 0.5, 1, 2, 5, 10, 20, 50, or 100 µg/mL tat-KDET peptide (a,b), with 0 or 50 µg/mL tat-KDET peptide (P WT ) or with 0 or 50 µg/mL tat-QNQS peptide (P mut ) (c,d). Neurons were then treated without or with antibody 557. Mean values + SEM from three independent experiments are shown for total neurite lengths (**** p < 0.0001 relative to stimulated neurons not treated with tat-KDET peptide, § § § § p < 0.0001 relative to non-stimulated neurons not treated with tat-KDET peptide or tat-QNQS peptide; one-way ANOVA with Dunn's multiple comparison test). (e) Cerebellar neurons were first treated with 0, 2, 10, 20, 50, or 100 µg/mL tat-KDET peptide and then treated without or with antibody 557 in the absence or presence of H 2 O 2 . Mean values + SEM from three independent experiments are shown for the relative numbers of dead cells (**** p < 0.0001 relative to stimulated neurons not treated with tat-KDET peptide in the presence of H 2 O 2 , § § § § p < 0.0001 relative to non-stimulated neurons not treated with tat-KDET peptide in the presence of H 2 O 2 , #### p < 0.0001 relative to unstimulated neurons not treated with tat-KDET peptide in the absence of H 2 O 2 ; one-way ANOVA with Dunn s multiple comparison test).

Discussion
In this study, we showed that HistH1e, NDUFV2, WDR5, SFPQ, NonO, PSPC1, Hsc70, and SYT1 interact with L1's KDET motif. L1's interactions with these and other proteins, namely TOP1, ERα, RXRβ, PPARγ, VDR, and AR, were shown by ELISA and proximity ligation, which detects L1-interacting proteins at a distance of less than 40. Proximity ligation may lead to the identification of the multiplicity of L1's binding partners. Yet, the predominant goal of the proximity ligation assay was to show that the molecules of interest should be detected in a cellular context as being close to each other. We did not aim at finding new L1 interaction partners.
The interaction between L1 and HistH1e was reduced by the γ-secretase inhibitor DAPT. Since the generation of L1-55 by γ-secretase and the interactions of L1-55 with its binding partners MeCP2 and HP1 has been shown to be reduced by DAPT [26,27], we conclude that HistH1e interacts with L1-55. Co-immunoprecipitation verified this notion. The interaction of L1 with NDUFV2, WDR5, SFPQ, NonO, PSPC1, Hsc70, and SYT1 was not affected by DAPT, indicating that these KDET-binding proteins do not interact with L1-55, but associate with other known or yet unidentified L1 fragments. Previously [17] and in this study, we showed that NDUFV2 interacts with L1-70. SFPQ and NonO have also been reported to associate with L1-70 [28].
HistH1e interacts with HP1α, HP1β, and HP1γ [33] and competes with MeCP2 for common chromatin binding sites [34]. It is thus conceivable that the interplay between L1-55, MeCP2, HP1, and HistH1e could regulate L1-, MeCP2-, HP1-, and HistH1e-dependent functions during development and in adulthood. L1-55 interacts with HistH1e, MeCP2, and HP1 isoforms via its KDET motif. The interaction between L1 and HistH1e is affected by the tat-KDET peptide in non-stimulated neurons, as seen for the interaction of L1 with MeCP2 [27], while the interaction of L1 with NDUFV2, WDR5, SFPQ, NonO, PSPC1, Hsc70, or SYT1 is affected by the tat-KDET peptide in stimulated neurons, as seen for the L1/HP1 interactions [27]. It is conceivable that the tat-KDET peptide could not bind to HP1, NDUFV2, WDR5, SFPQ, NonO, PSPC1, Hsc70, or SYT1 under basal conditions and thus could not compete with the binding of L1 to these binding partners in non-stimulated neurons. Only after stimulation, the peptide binds to HP1, NDUFV2, WDR5, SFPQ, NonO, PSPC1, Hsc70, or SYT1 and interfere with the binding of L1 to these proteins. However, it is noteworthy that the peptide binds to MeCP2 and HistH1e in non-stimulated neurons to reduce the binding of MeCP2 and HistH1e to L1.
KDET is present in the L1-ICD sequences of mammalian, amphibian, avian, reptilian, and fish species, suggesting that this motif, being conserved in evolution, plays a crucial role in regulating L1 functions. Here, we show that the KDET motif is essential for the regulation of L1-dependent neural functions, such as neurite outgrowth and neuronal survival. Disturbance of the interaction between L1 and its binding partners via the KDET motif prevents neurite outgrowth and neuronal protection against oxidative stress. Moreover, reduction of NDUFV2, WDR5, SFPQ, NonO, Hsc70, and SYT1 expression, and thus reduction of the level of the interactions between L1 and these KDET-binding proteins, inhibits L1-dependent neurite outgrowth. Of note, the reduction of PSPC1 expression does not affect L1-dependent neurite outgrowth, indicating that KDET-mediated interaction of L1 with PSPC1 is not required for L1-dependent neurite outgrowth. SYT1, which functions as calcium sensors in vesicular trafficking and exocytosis of neurotransmitters and hormones, is involved in the regulation of neurite outgrowth [35][36][37] and it is linked to a rare neurodevelopmental disorder known as SYT1-associated neurodevelopmental disorder or Baker-Gordon syndrome [38]. Besides other symptoms, patients show intellectual disability, poor speech, and delayed development of walking. Similar symptoms are observed in patients with L1 syndrome.
In utero suppression of NDUFV2 expression inhibits migration of cortical neurons and impairs the actin and tubulin cytoskeleton in cortical neurons [39]. NDUFV2 is associated with schizophrenia, bipolar disorder, and Parkinson's disease (for references, see [39]).
WDR5 plays a crucial role in regulating neuronal gene expression and neurite outgrowth and contributes to the development of X chromosome-linked mental retardation [40].
Mutations in HistH1e cause a rare neurodevelopmental disorder known as Rahman syndrome, which is characterized in addition to other symptoms by intellectual disability and autism spectrum disorder [41].
SFPQ is a key regulator of long neuronal gene expression and thus plays an important role in the pathogenesis of amyotrophic lateral sclerosis, frontotemporal lobar degeneration, and autism spectrum disorder, which are often caused by a dysregulation of the expression of long genes [42]. In addition, SFPQ regulates the expression of mRNAs essential for axon viability and is required for the axonal trafficking of these mRNAs [43].
NonO, which forms heterodimers with SFPQ, is associated with an intellectual disability syndrome, including macrocephaly and a thickened corpus callosum [44].
A Hsc70 inhibitor and Hsc70 siRNA have been reported to decrease neurite outgrowth induced by a neuritogenic reagent [45].
Based on these findings, we propose that the concomitant or consecutive interactions of L1 with its binding partners via the KDET motif are not only required for regulating L1-dependent cellular functions, such as neurite outgrowth, but also for regulating nervous system development and functions, such as synaptic plasticity, learning and memory, as well as behavior. It is noteworthy, in this context, that the identified L1 binding partners do not display obvious similarities in function but appear to mostly subserve different functional properties. Our study thus expands the spectrum of L1 activities. It is also noteworthy that different fragments of L1, as exemplified by L1-55 and L1-70 ( Figure 6), interact with different binding partners via the KDET motif. An explanation for these findings could be that the accessibility of the different binding partners to different fragments is distinct in different compartments and under different metabolic conditions. We propose that other factors determine the binding of different fragments to different molecules. Finally, it is conceivable that the disturbance of the interaction of L1 with its binding partners via the KDET motif contributes to the pathogenesis of L1-associated diseases. SFPQ is a key regulator of long neuronal gene expression and thus plays an important role in the pathogenesis of amyotrophic lateral sclerosis, frontotemporal lobar degeneration, and autism spectrum disorder, which are often caused by a dysregulation of the expression of long genes [42]. In addition, SFPQ regulates the expression of mRNAs essential for axon viability and is required for the axonal trafficking of these mRNAs [43].
NonO, which forms heterodimers with SFPQ, is associated with an intellectual disability syndrome, including macrocephaly and a thickened corpus callosum [44].
A Hsc70 inhibitor and Hsc70 siRNA have been reported to decrease neurite outgrowth induced by a neuritogenic reagent [45].
Based on these findings, we propose that the concomitant or consecutive interactions of L1 with its binding partners via the KDET motif are not only required for regulating L1-dependent cellular functions, such as neurite outgrowth, but also for regulating nervous system development and functions, such as synaptic plasticity, learning and memory, as well as behavior. It is noteworthy, in this context, that the identified L1 binding partners do not display obvious similarities in function but appear to mostly subserve different functional properties. Our study thus expands the spectrum of L1 activities. It is also noteworthy that different fragments of L1, as exemplified by L1-55 and L1-70 ( Figure 6), interact with different binding partners via the KDET motif. An explanation for these findings could be that the accessibility of the different binding partners to different fragments is distinct in different compartments and under different metabolic conditions. We propose that other factors determine the binding of different fragments to different molecules. Finally, it is conceivable that the disturbance of the interaction of L1 with its binding partners via the KDET motif contributes to the pathogenesis of L1-associated diseases.  proteases, the β-site of an amyloid precursor protein cleaving enzyme and ɣ-secretase generates L1-55, which comprises amino acids of the TMD and the entire ICD with the KDET motif.

Animals
Mice were bred and maintained at the Universitätsklinikum Hamburg-Eppendorf at 25 °C on a 12 h light/12 h dark cycle with ad libitum access to food and water. C57BL/6J males and females or L1-deficient males [10] and wild-type male littermates were used for all experiments. All animal experiments were conducted in accordance with the German and European Community laws on the protection of experimental animals and approved by the local authorities of the State of Hamburg (animal permit numbers ORG 1022). The manuscript was prepared following the ARRIVE guidelines for animal research [46].

Animals
Mice were bred and maintained at the Universitätsklinikum Hamburg-Eppendorf at 25 • C on a 12 h light/12 h dark cycle with ad libitum access to food and water. C57BL/6J males and females or L1-deficient males [10] and wild-type male littermates were used for all experiments. All animal experiments were conducted in accordance with the German and European Community laws on the protection of experimental animals and approved by the local authorities of the State of Hamburg (animal permit numbers ORG 1022). The manuscript was prepared following the ARRIVE guidelines for animal research [46].

Reagents and Antibodies
The following antibodies were from Santa Cruz Biotechnology ( proteases, the β-site of an amyloid precursor protein cleaving enzyme and ɣ-secretase generates L1-55, which comprises amino acids of the TMD and the entire ICD with the KDET motif.

Animals
Mice were bred and maintained at the Universitätsklinikum Hamburg-Eppendorf at 25 °C on a 12 h light/12 h dark cycle with ad libitum access to food and water. C57BL/6J males and females or L1-deficient males [10] and wild-type male littermates were used for all experiments. All animal experiments were conducted in accordance with the German and European Community laws on the protection of experimental animals and approved by the local authorities of the State of Hamburg (animal permit numbers ORG 1022). The manuscript was prepared following the ARRIVE guidelines for animal research [46].

ELISA
For ELISA, 25 µL of 10 or 20 µg/mL recombinant proteins were incubated overnight at 4 • C in 384-well microtiter plates with a high-binding surface (Corning, Tewksbury, MA, USA). All of the following steps were performed at room temperature. Wells were washed with Dulbecco s phosphate-buffered saline with MgCl 2 and CaCl 2 (Sigma-Aldrich, Taufkirchen, Germany) (PBS), treated with blocking solution (2% essentially fatty acid-free bovine serum albumin in PBS) for 2 h, washed again with PBS containing 0.005% Tween 20 (PBST), and incubated with increasing concentrations of recombinant His-tagged L1-ICD or CHL1-ICD as ligands for 1 h under gentle agitation. For competition ELISA, 2.5 µM L1-ICD was preincubated for 1 h without or with a 5-fold molar excess of L1 peptides KDET or QNQS. The mixtures were then incubated with substrate-coated recombinant proteins. After washing two times with PBS and three times with PBST, L1 antibody C-2 (1:500) or CHL1 antibody C-18 (1:200) in blocking solution were applied for 1 h, followed by two washes with PBS and three washes with PBST, and incubation with horseradish peroxidase-coupled anti-mouse antibody (diluted 1:2000 in blocking solution) for 1 h. Wells were washed again with PBST, and 1 mg/mL ortho-phenylenediamine dihydrochloride (Thermo Fisher Scientific) was used for the detection of bound L1-ICD or CHL1-ICD. The reaction was terminated by the addition of 25 µL of 2.5 M sulfuric acid. Absorbance was measured at 492 nm with an ELISA reader (µQuant; BioTek, Bad Friedrichshall, Germany).

Cultures of Cerebellar and Cortical Neurons
Cerebellar neurons were prepared from the cerebella of 6-to 8-day-old mice. Cerebella were incubated with 10 mg/mL trypsin and 0.5 mg/mL DNase I (Sigma-Aldrich) in Hanks' balanced salt solution (HBSS) for 15 min at 37 • C, washed with HBSS, mechanically dissociated, and centrifuged at 100× g for 15 min. Cells were then diluted in Neurobasal A medium (Thermo Fisher Scientific), supplemented with 2 mM L-glutamine (Thermo Fisher Scientific), 4 nM L-thyroxine (Sigma-Aldrich), 0.1 mg/mL BSA (Sigma-Aldrich), 12.5 µg/mL insulin (Sigma-Aldrich), 30 nM sodium selenite (Sigma-Aldrich), 100 µg/mL transferrin, 0.1 mg/mL streptomycin, and 100 U/mL penicillin (Thermo Fisher Scientific). For the proximity ligation assay, cells were seeded onto poly-L-lysine-coated 12 mm coverslips in a 24-well plate at a density of 2.5 × 10 5 cells per well. For the determination of neurite outgrowth, cells were seeded at a density of 5 × 10 4 cells per well of a 48-well plate coated with poly-L-lysine (Sigma-Aldrich). For neuronal survival analysis, cells were seeded at a density of 1.25 × 10 5 cells per well of a 48-well plate coated with poly-L-lysine.
For the culturing of cortical neurons, cerebral cortices were dissected from 15.5-to 16.5-day-old embryos and incubated in 0.025% trypsin (Sigma-Aldrich) in HBSS at 37 • C for 30 min. The cortices were then incubated in HBSS containing 1% BSA (Sigma-Aldrich) and 1% trypsin inhibitor (T-6522, Sigma-Aldrich) at 37 • C for 5 min. After washing in HBSS, the tissue was mechanically dissociated, and the dissociated cells were cultured in Neurobasal medium (Thermo Fisher Scientific) supplemented with 1% B-27 (Thermo Fisher Scientific), 2 mM L-glutamine (Thermo Fisher Scientific), 100 U/mL penicillin (Thermo Fisher Scientific), and 100 µg/mL streptomycin (Thermo Fisher Scientific). For the proximity ligation assay and immunostaining, cells were seeded onto poly-L-lysine-coated 12 mm coverslips in a 24-well plate at a density of 2.5 × 10 5 cells per well. For the determination of neurite outgrowth, cells were seeded at a density of 5 × 10 4 cells or 1 × 10 6 per well of a 48-well plate coated with poly-L-lysine (Sigma-Aldrich). For Western blot analysis, cells were seeded onto poly-L-lysine-coated 12-well plates at a density of 1.5 × 10 6 cells per well.
For transfection of cortical neurons, cortical neurons were seeded onto poly-L-lysinecoated coverslips in 24-well plates (proximity ligation and immunostaining) or onto 12-well plates (Western blot analysis) and 48-well plates (neurite outgrowth) and maintained for 2 h before transfection with 1 µL (48-well), 2 µL (24-well), or 4 µL (12-well) of 10 µM siRNA and 1-4 µL FuGENE transfection reagent per well. L1 antibody 557 (50 µg/mL) was added to the cultures 24 h after transfection (for proximity ligation and cell survival) or 2 h after transfection (neurite outgrowth), and 48 h after transfection, the cells were analyzed for neurite outgrowth and cell survival, or used for immunostaining, Western blot analysis, or proximity ligation assay.

Proximity Ligation Assay and Immunostaining with Cerebellar and Cortical Neurons
Cultures were fixed for 15 min at room temperature in 4% formaldehyde, washed with PBS, and used for immunostaining or subjected to a proximity ligation assay using Duolink PLA products according to the manufacturer's protocol (Sigma-Aldrich; Duolink PLA technology) with minor modifications. For the proximity ligation assay, cells were incubated with 1% Triton X-100 in PBS for 30 min, washed once with PBS, blocked with Duolink Blocking solution for 30 min, and incubated for 24 h at 4 • C with mouse L1 antibody C-2 and goat or rabbit antibodies against TOP1, SYT, SFPQ, WDR5, hnRNP A isoforms, Nup93, PSPC1, NonO, NDUFV2, HistH1e, ERα, RXR, PPARγ, AR, or VDR, or with rabbit L1 antibody 12399 and mouse antibodies against impβ1 or hnRNP A, all diluted 1:10 in Duolink Antibody Diluent. Cells were washed two times using Duolink Wash Buffer A and incubated with a mixture of secondary antibodies conjugated with oligonucleotides (Duolink PLA Anti-Rabbit or Anti-Goat Probe MINUS and Duolink Anti-Mouse PLA Probe PLUS). The proximity ligation reaction was then performed according to the manufacturer's protocol using the Duolink In Situ Detection Reagent RED. Thereafter, the coverslips were incubated with 5 µg DAPI/mL in PBS for 15 min, washed twice with PBS, and mounted in Immuno-Mount (Thermo Fisher Scientific). Ten images per condition were taken using an Olympus F1000 confocal microscope and analyzed using ImageJ software (ImageJ version 1.53q; https://imagej.nih.gov/ij/index.html; RRID:SCR_003070, 30 March 2022). Numbers of red dots and numbers of DAPI-stained nuclei were determined using ImageJ, and the number of dots per image was divided by the number of nuclei per image to determine the average number of dots per nucleus. The average values were determined from 10 images per condition.