Inhibition of Cxcr4 chemokine receptor signaling improves habituation learning in a zebrafish model of neurofibromatosis.

INTRODUCTION

Neurofibromatosis type 1 (NF1) is an autosomal dominant, neurogenetic disorder caused by loss-of-function mutations in the gene neurofibromin 1 ( NF1 ). NF1 is a comparatively common single-gene disorder estimated to affect 1 in 2000-3000 live births ( Evans et al., 2010 ; Huson et al., 1989 ; Uusitalo et al., 2015 ). NF1 is associated with a wide range of symptoms involving multiple organ systems. Neurological features, including peripheral nerve tumors, brain tumors and cognitive dysfunction are major clinical concerns ( Cimino and Gutmann, 2018 ). Up to 80% of children with NF1 experience cognitive or behavioral deficits that impact social and academic development ( Hyman et al., 2005 ; Torres Nupan et al., 2017 ). Despite increased attention on the role of cognitive function in quality of life for patients with NF1 ( Krab et al., 2009 ; Varni et al., 2019 ), the molecular basis of NF1-associated cognitive dysfunction remains unclear ( Miller and Halloran, 2022 ).

NF1 encodes neurofibromin, a multifunctional GTPase-activating protein (GAP) that inhibits Ras signaling ( Basu et al., 1992 ; Dasgupta et al., 2005 ; DeClue et al., 1991 ; Endo et al., 2013 ; Johannessen et al., 2005 ; Marchuk et al., 1991 ) and activates cAMP signaling ( Anastasaki and Gutmann, 2014 ; Guo et al., 1997 ; Hannan et al., 2006 ). cAMP is a second messenger regulated by G protein-coupled receptor (GPCR) modulation of adenylyl cyclase (AC) activity. Dysregulated GPCR signaling is seen in neurofibromin-deficient human and mouse tissue, including increased CXCR4 chemokine receptor signaling associated with tumorigenesis ( Karaosmanoglu et al., 2018 ; Mo et al., 2013 ; Warrington et al., 2007 ) and reduced dopamine signaling associated with cognitive function ( Brown et al., 2010a , 2011 ; Diggs-Andrews et al., 2013 ). Ras and cAMP signaling also have been implicated in the regulation of neurofibromin-dependent cognitive function. Studies in both fruit flies and zebrafish show that loss of neurofibromin results in learning deficits that are rescued by enhancing the cAMP signaling pathway, whereas memory deficits are rescued by inhibiting the Ras signaling pathway ( Ho et al., 2007 ; Wolman et al., 2014 ). Therefore, the role of neurofibromin might differ in neural circuits controlling distinct cognitive functions. The mechanism by which neurofibromin regulates cAMP signaling is unclear, as Ras-independent ( Brown et al., 2010b , 2012 ; Guo et al., 1997 ), Ras-dependent ( Anastasaki and Gutmann, 2014 ; Walker et al., 2006 ), and both Ras-independent and dependent pathways ( Hannan et al., 2006 ) have been proposed.

Translation of research from animal models into treatments for NF1-associated cognitive dysfunction has seen limited success. Clinical trials for patients with NF1 using lovastatin or simvastatin to reduce Ras signaling have demonstrated limited effectiveness in improving learning or other measures of cognitive function ( Bearden et al., 2016 ; Krab et al., 2008 ; Mainberger et al., 2013 ; Payne et al., 2016 ; Stivaros et al., 2018 ; van der Vaart et al., 2013 ). In a more recent clinical trial, regulation of Ras signaling through mitogen-activated protein kinase kinase (MEK) inhibitor (MEKi) treatment showed stable but not improved group-level performance on working memory and visual learning/memory tasks ( Walsh et al., 2021 ). Identifying additional drug targets that improve learning in Nf1 mutant animals and characterizing their mechanisms of action will provide new opportunities to develop treatments for cognitive dysfunction in patients with NF1.

In this study, using a zebrafish model of NF1 ( Shin et al., 2012 ), we took two approaches to gain insight into the molecular mechanisms by which neurofibromin deficiency leads to behavioral and cognitive deficits. The first, a small-molecule drug screen to identify regulators of habituation learning, revealed AMD3100 (hereafter referred to as plerixafor), a modulator of the Cxcl12-Cxcr4 and Cxcl12-Ackr3 chemokine signaling axes, improves habituation deficits in nf1 mutants. The second, RNAseq analysis to identify changes in gene expression with loss of neurofibromin, showed that chemokine receptor cxcr4b is upregulated in nf1 mutants. Cxcr4 signaling is known to activate Ras-Raf-MEK-ERK signaling and inhibit cAMP-PKA signaling ( Busillo and Benovic, 2007 ). Therefore, we further investigated the effects of pharmacological inhibition of Cxcr4 on these signaling pathways. Using a whole-brain mitogen-activated protein kinase mapping (MAP-mapping) approach, we demonstrated that nf1 mutant larvae have region-specific increases and decreases in phosphorylated-ERK1/2 (pERK). However, plerixafor treatment produced limited changes in brain pERK levels in wild-type or nf1 mutant larvae. In contrast, using an enzyme-linked immunosorbent assay (ELISA) and whole-brain immunolabeling, we showed that nf1 mutant larvae have decreased whole-body cAMP and whole-brain phosphorylated-PKA-substrate (pPKA-substrate) levels that are increased by plerixafor treatment in nf1 heterozygous mutant larvae. Overall, we have identified Cxcr4 signaling as a potential regulator of neurofibromin-dependent habituation learning and a modulator of cAMP-PKA signaling in nf1 mutant larval zebrafish.

RESULTS

Small-molecule screening reveals regulators of acoustically evoked behaviors in nfl mutant larvae

A zebrafish model of NF1 harboring null alleles in the NF1 orthologs nf1a and nf1b has previously been developed using a zinc finger nuclease strategy ( Shin et al., 2012 ). We used these nf1 mutants in an unbiased small-molecule screen to identify potential drug targets that regulate neurofibromin-dependent behaviors. Previous work has shown that zebrafish nf1 mutants have defects in prepulse inhibition (PPI) and habituation of the acoustic startle response, two measures of sensory filtering that demonstrate the ability of zebrafish larvae to adapt their behavior based on previous sensory experience ( Randlett et al., 2019 ; Shin et al., 2012 ; Wolman et al., 2014 ). We measured the effects pharmacological manipulation of cellular signaling has on PPI and habituation learning in nf1 mutant larvae.

In larval zebrafish, acoustic stimuli evoke robust startle responses consisting of ‘C-shape’ bends with distinct kinematic properties ( Burgess and Granato, 2007a ; Jain et al., 2018 ; Kimmel et al., 1974 ). Initiation of a short-latency C-bend (SLC) drives rapid escape behavior ( Burgess and Granato, 2007a ; Hecker et al., 2020 ; Jain et al., 2018 ; Marsden et al., 2018 ; Wolman et al., 2015 ). Our behavioral assay was designed to measure baseline startle initiation, PPI and habituation of the zebrafish SLC response in nf1 mutant larvae. A total of 60 acoustic stimuli were delivered with varying intensities and interstimulus intervals (ISIs) ( Fig. 1A ). Motor responses of individual larvae at 5 days post fertilization (dpf) were recorded with a high-speed camera and response kinematics were analyzed using automated tracking software. The low-intensity stimulus was designed to evoke SLCs in response to ∼25% of stimuli ( Fig. S1A ). The high-intensity stimulus was designed to evoke SLCs in response to ∼75% of stimuli ( Fig. S1A ). PPI was assessed by comparing SLC initiation in response to individual high-intensity stimuli versus high-intensity stimuli that were preceded by low-intensity stimuli ( Fig. 1A ). Habituation was assessed by comparing SLC initiation in response to high-intensity stimuli separated by 1.5 s ISIs instead of 20 s ISIs ( Fig. 1A ).

Pharmacological screening of 1134 small-molecule compounds with known biological targets was completed over the course of 60 experimental days. Pair mating nf1a +/− ; nf1b −/− and nf1a −/− ; nf1b +/− adults ( Shin et al., 2012 ) yielded a mix of nf1 mutant embryos with predicted equal ratios of nf1a +/− ;nf1b +/− , nf1a +/− ;nf1b −/− , nf1a −/− ;nf1b +/− and nf1a −/− ;nf1b −/− mutant alleles. To determine baseline behavioral responsiveness in non-treated nf1 mutant larvae, we quantified the baseline initiation of SLC responses and total motor reactions to low-intensity and high-intensity stimuli ( Fig. S1A,B ). To examine neurofibromin-dependent behaviors in non-treated nf1 mutant larvae, we quantified habituation to stimuli 41-50, habituation to stimuli 51-60 and PPI ( Fig. 1B ). Data from two groups of 32 non-treated nf1 mutant larvae were averaged on each experimental day and used as comparison to identify compounds that significantly regulated acoustically evoked behaviors.

To test the specific behavioral effects of compounds, we first analyzed drugs that altered baseline initiation of total reactions to low or high-intensity stimuli in nf1 mutant larvae. Forty-nine compounds increased baseline initiation of total reactions to low-intensity stimuli ( Fig. S1C ; Table S1 ), and 87 compounds decreased baseline initiation of total turns to high-intensity stimuli ( Fig. S1C ; Table S1 ). We then performed enrichment analysis to identify classes of compounds with similar bioactivity by using biological targets defined by the compound library manufacturer (Selleck, https://www.selleckchem.com/screening-libraries.html ). Compounds that increased baseline initiation of total reactions to low-intensity stimuli were enriched for retinoid receptor targets ( Table 1 ). Compounds that decreased baseline initiation of total turns to high-intensity stimuli were enriched for estrogen and progestogen receptor, calcium channel and 5-alpha reductase targets ( Table 1 ). Although biologically interesting, we did not pursue compounds that altered baseline initiation of total turns in nf1 mutant larvae and, instead, focused on compounds that specifically regulated habitation or PPI. Of those, 22 compounds improved habituation to stimuli 41-50 ( Fig. 1C ; Table S1 ) and 32 compounds improved habituation to stimuli 51-60 ( Fig. 1C ; Table S1 ). Compounds that improved habituation in nf1 mutant larvae were enriched for adrenergic and 5-hydroxytryptamine (5-HT) receptor targets ( Table 1 ). Additionally, 41 compounds improved PPI ( Fig. 1C ; Table S1 ). Compounds that improved PPI were enriched for 5-HT receptor, phosphodiesterase inhibitor and topoisomerase inhibitor targets ( Table 1 ).

Adrenergic and 5-HT signaling pathways are known to regulate habituation learning in wild-type animals ( Cassel, 2010 ; Kocher and Straumann, 2025 ; Wolman et al., 2011 ). We were interested in identifying pathways specific to loss of neurofibromin. Therefore, of the drugs that specifically improve habituation and/or PPI without affecting baseline, we focused on plerixafor, a modulator of chemokine-receptor signaling. Plerixafor modulates the CXCL12-CXCR4 and CXCL12-ACKR3 signaling axes through antagonism of CXCR4 or allosteric agonism of ACKR3 (formerly known as CXCR7) ( Hatse et al., 2002 ; Jørgensen et al., 2021 ; Kalatskaya et al., 2009 ; Naumann et al., 2010 ; Sarma et al., 2023 ; Singh et al., 2013 ). CXCR4 is an inhibitory G-protein-coupled (G i ) chemokine receptor and its inhibition has been proposed as a treatment for optic glioma and malignant peripheral nerve sheath tumors (MPNSTs) in NF1 patients ( Mo et al., 2013 ; Warrington et al., 2007 ). ACKR3 is an atypical chemokine receptor that can scavenge CXCL12 ligand or signal through recruitment of β-arrestin and kinase assembly ( Gritsina et al., 2023 ; Rajagopal et al., 2010 ; Ray et al., 2012 ). However, the effects of CXCL12-CXCR4 and CXCL12-ACKR3 signaling on NF1-associated cognitive dysfunction have not been studied. Therefore, we further characterized the effects of plerixafor on habituation in wild-type and nf1 mutant larvae.

Cxcr4 signaling regulates habituation in nfl mutant larvae

Previous studies of habituation of the SLC response in nf1 mutant larvae have only identified deficits in nf1 double homozygous mutants ( nf1a −/− ;nf1b −/− ) ( Shin et al., 2012 ; Wolman et al., 2014 ). Because the human disease is autosomal dominant and manifests in heterozygous individuals, we established an assay sensitive to habituation deficits in heterozygous nf1 mutants. We modified our original behavioral assay by removing the PPI phase and increasing the ISI during the pre-habituation (30 s) and the habituation (3 s) phases ( Fig. 1D ). We tested non-treated wild-type and nf1 mutant larvae in this modified assay, obtaining an average habituation value from groups of larvae during a total of eight experimental days ( Fig. 1D ). Larvae were genotyped post hoc for nf1a and nf1b mutant alleles ( Shin et al., 2012 ) and, in total, 148-213 larvae were tested per genotype. One-way ANOVA showed a statistically significant difference between genotypes [ F (4,35)=23.62, P <0.001]. Adjusting for multiple comparisons between genotypes, habituation was significantly higher in wild-type larvae compared to each of the nf1 mutant groups. Additionally, habituation was significantly lower in nf1a −/− ;nf1b −/− mutant larvae compared to each of the three nf1 heterozygous ( nf1a +/− ;nf1b +/− , nf1a +/− ;nf1b −/− , nf1a −/− ;nf1b +/− ) mutants ( Fig. 1D ). Using this modified behavioral assay, we were then able to test if pharmacological modulation of the Cxcl12-Cxcr4 and Cxcl12-Ackr3 signaling axes signaling axis could improve habituation in both nf1 double homozygous and nf1 heterozygous mutant larvae.

We tested the effect of Cxcr4 and Ackr3 receptor signaling on habituation by treating larvae with plerixafor and plerixafor 8HCl, two separate formulations of AMD3100 that inhibit Cxcr4 and activate Ackr3 receptor signaling. Using both formulations ( Fig. 1E ; Fig. S1D ), we found significant differences between groups. Plerixafor treatment ( Fig. 1E ) improved habituation in all nf1 mutant genotypes without affecting wild-type habituation – restoring habituation to near wild-type levels in nf1a +/− ;nf1b +/− , nf1a +/− ;nf1b −/− and nf1a −/− ;nf1b −/− mutant larvae. Treatment with plerixafor 8HCl ( Fig. S1D ) significantly improved habituation in nf1a −/− ;nf1b −/− and nf1a +/− ;nf1b −/− mutant larvae. Since plerixafor is known to modulate both Cxcr4 and Ackr3 receptor signaling with varying effects on downstream signaling ( Bonecchi and Graham, 2016 ; Jørgensen et al., 2021 ; Lu et al., 2024 ; Tian et al., 2019 ), we next sought to differentiate between the two receptor signaling pathways. We tested the effect of Ackr3 activation on habituation by treating larvae with VUF11207 , an ACKR3 agonist that induces recruitment of β-arrestin to ACKR3 without altering CXCR4 G-protein activity ( Sarma et al., 2023 ; Wijtmans et al., 2012 ). We found that VUF11207 did not significantly alter habituation in wild-type or nf1 mutant larvae ( Fig. S2A ), supporting the conclusion that plerixafor is acting through Cxcr4 and not Ackr3 signaling to improve habituation. Overall, these pharmacological results support the conclusion that Cxcr4 signaling regulates habituation in nf1 mutant larvae.

Inhibition of Cxcr4 does not affect general motor performance at doses that improve habituation

To test the behavioral specificity of Cxcr4 inhibition at different treatment doses, we analyzed the effect of plerixafor on general motor performance by quantifying initiation of total turns to pre-habituation high-intensity stimuli. At the 3 µM dose for plerixafor and 300 nM dose for plerixafor 8HCl, treatment did not significantly affect baseline initiation of total turns to high-intensity stimuli in wild-type larvae ( Fig. 2A ; Fig. S3A ). Therefore, we selected the 3 µM dose of plerixafor as a treatment that selectivity improves habituation in nf1 mutant larvae ( Fig. 1E ) without affecting baseline initiation of turns ( Fig. 2A ). All subsequent experiments were conducted comparing non-treated and 3 µM plerixafor-treated larvae.

In a previous report on nf1 mutant larvae, nf1a −/− ;nf1b −/− mutants displayed motor deficits in addition to their habituation learning deficits, with kinematically weaker SLC responses as measured by decreased head turn angle, maximum angular velocity, and total distance traveled ( Shin et al., 2012 ). To test if Cxcr4 inhibition affects the kinematic properties of turning behavior, we analyzed turn kinematics in non-treated and 3 µM plerixafor-treated larvae ( Fig. 2B-D ). Consistent with the previous report ( Shin et al., 2012 ), we observed kinematically weaker SLC responses for turn angle, maximum angular velocity and total distance moved in nf1a −/− ;nf1b −/− larvae. Additionally, we found a decrease in total distance moved for each of the nf1 heterozygous mutant larvae and a decrease in turn angle in nf1a +/− ;nf1b −/− larvae. However, these motor deficits were not improved by inhibiting Cxcr4. Therefore, these results support a conclusion that Cxcr4 inhibition specifically regulates habituation learning in nf1 mutants without altering general motor performance.

Gene expression analysis reveals cxcr4b is upregulated in nf1 double homozygous mutants

We used a second approach, RNAseq analysis, to identify additional or overlapping molecular targets influenced by loss of neurofibromin. RNA was isolated from whole larvae in wild-type and putative nf1 double homozygous mutant larvae identified by hyperpigmentation phenotype (see Materials and Methods). Differential gene expression analysis followed by KEGG pathway enrichment analysis identified multiple metabolic pathways enriched for genes upregulated in nf1 double homozygous mutants ( Table 2 ). Focusing on Cxcr4 signaling, differential expression revealed upregulation of cxcr4b at 5 dpf in nf1 double homozygous mutants compared to wild-type larvae (fold change=1.52, adjusted P -value<0.001). The increased expression of cxcr4b in nf1 double homozygous mutant larvae alongside the small molecule drug screen results lead us to investigate the relationship between Cxcr4 and neurofibromin signaling.

Cxcr4 signaling has limited effects on canonical Ras-Raf-MEK-ERK signaling in the brain

CXCR4 is a G i protein-coupled chemokine receptor specific for the ligand stromal-derived-factor-1 (CXCL12, also known as SDF1). Through Gα i protein activation, the CXCR4 receptor is known to increase Ras signaling and inhibit cAMP signaling ( Busillo and Benovic, 2007 ). Neurofibromin is also known to regulate these signaling pathways but in the opposing directions – by inhibiting Ras signaling and increasing cAMP signaling. Therefore, in neurofibromin-deficient larvae, we predicted Cxcr4 inhibition might decrease overactive Ras signaling and/or increase suppressed cAMP signaling.

We first tested where overactive Ras signaling is present in the brains of nf1 mutant larvae. The canonical Ras signaling pathway involves a kinase cascade that phosphorylates the mitogen-activated protein kinase (MAPK) family members MAPK1 and MAPK3 [also known as extracellular signal-regulated kinases 2 and 1 (ERK2 and ERK1), respectively] ( Lavoie et al., 2020 ). Phosphorylation of ERK1/2 can be detected and quantified in larval zebrafish brains using immunolabeling and the open-source Z-BRAIN reference atlas ( Randlett et al., 2015 ). This technique, named MAP-mapping, is often used to detect changes in neuronal activity as calcium influx activates ERK1/2 signaling ( Randlett et al., 2015 ; Rosen et al., 1994 ; Thomas and Huganir, 2004 ). Here, differences in phosphorylation of ERK1/2 between experimental groups may result from direct loss of neurofibromin altering Ras-Raf-MEK-ERK signaling or the indirect effects of altered neuronal activity in nf1 mutants. We fixed freely swimming larvae at 6 dpf, immunostained for total ERK1/2 (tERK) and phosphorylated ERK1/2 (pERK), and then quantified and mapped images from wild-type, nf1a +/− ;nf1b +/− and nf1a −/− ;nf1b −/− larvae. Comparing nf1a +/− ;nf1b +/− larvae to wild type ( Fig. 3A ) and nf1a −/− ;nf1b −/− larvae to wild type ( Fig. 3B ), we found region-specific increases and decreases in levels of pERK. We focused on regions of interest associated with acoustic startle that displayed increased amounts of pERK ( Table 3 ), an indicator that loss of neurofibromin may be contributing to overactivation of Ras in these regions. Acoustic startle responses in zebrafish are triggered by activation of reticulospinal Mauthner neurons in the hindbrain, which integrate acoustic and vibrational inputs from the statoacoustic ganglia and the lateral line and send output signals to motor neurons to induce muscle contraction ( Eaton et al., 1977 ; Hecker et al., 2020 ; Liu and Fetcho, 1999 ; Wolman et al., 2015 ). Additionally, the octavolateralis nuclei and torus semicircularis relay auditory information to the thalamus. Some auditory information from the torus semicircularis and thalamus is also sent back to the hindbrain in zebrafish ( Vanwalleghem et al., 2017 ) and could, potentially, provide top-down regulation of hindbrain activity to regulate acoustic startle habituation. Compared to wild type, nf1a +/− ;nf1b +/− and nf1a −/− ;nf1b −/− larvae showed increased amounts of pERK in the torus semicircularis and hindbrain, indicating that Ras signaling may be dysregulated in these regions. Therefore, we next asked if Cxcr4 inhibition decreases pERK in these brain regions.

To investigate whether Cxcr4 inhibition modulates Ras signaling in brain regions associated with acoustic startle, we performed MAP-mapping on wild-type ( Fig. 3C ), nf1a +/− ;nf1b +/− ( Fig. 3D ) and nf1a −/− ;nf1b −/− ( Fig. 3E ) larvae comparing non-treated to plerixafor-treated groups within each genotype ( Table 4 ). Overall, plerixafor treatment had limited effects on pERK levels in both wild-type and nf1 mutant larvae. We did find decreased levels of pERK in the torus semicircularis of plerixafor-treated wild-type and nf1a −/− ;nf1b −/− larvae compared to non-treated larvae. However, these differences were unilateral and not observed in nf1a +/− ;nf1b +/− larvae ( Fig. 3D ). Additionally, plerixafor treatment had very limited effects on pERK in the hindbrain ( Table 4 ), where Mauthner neuron circuitry drives acoustic startle responses. Interestingly, the olfactory bulb showed decreased levels of pERK in both wild-type and nf1 mutant larvae treated with plerixafor ( Fig. 3C-E ; Table 4 ). This is consistent with studies of larval zebrafish and adult mice showing Cxcr4 mRNA is highly expressed in the olfactory bulb ( Förster et al., 2017 ; Stumm et al., 2002 ). However, this result is unlikely to be related to regulation of acoustic startle responses. Taken together, these results support the conclusion that Cxcr4 signaling has a limited role in modulating canonical Ras-Raf-MEK-ERK signaling in the larval zebrafish brain.

Cxcr4 signaling increases downregulated cAMP-PKA signaling in nf1 heterozygous mutants

Previous studies indicate that neurofibromin-deficient mice, fruit flies and zebrafish have reduced cAMP levels ( Brown et al., 2010b ; Ho et al., 2007 ; Tong et al., 2002 ; Wolman et al., 2014 ). Consistent with these results, we found decreased cAMP levels in nf1 mutants compared to wild-type larvae when using enzyme-linked immunosorbent assay (ELISA) to quantify cAMP ( Fig. 4A ). We collected tissue for ELISA analyses from 5 dpf wild-type and nf1 mutant larvae separated into nf1 heterozygous mutant and nf1 double homozygous mutant groups by hyperpigmentation phenotype. We found that cAMP levels were significantly lower in nf1 heterozygous mutants compared to wild-type larvae and lower in nf1 double homozygous mutants compared to both nf1 heterozygous mutants and wild-type larvae ( Fig. 4A ). Next, we asked if Cxcr4 inhibition increased cAMP levels in nf1 mutant larvae. We found that plerixafor treatment increased cAMP levels in wild-type and nf1 heterozygous mutant larvae but not in nf1 double homozygous mutant larvae ( Fig. 4A ). This finding suggests that CXCR4 inhibition of cAMP has a neurofibromin-dependent component, as plerixafor does not significantly increase cAMP levels in the complete absence of neurofibromin. Taken together, these results support the conclusion that all nf1 mutants have suppressed cAMP levels but that Cxcr4 inhibition raises cAMP levels in nf1 heterozygous mutant larvae.

To regulate habituation of the SLC response, Cxcr4 signaling would need to modulate cAMP in neural circuits associated with acoustic startle. Therefore, we asked if Cxcr4 inhibition increases cAMP signaling in the brain. The cAMP effector protein kinase A (PKA) phosphorylates protein targets to regulate diverse cellular processes ( Taskén and Aandahl, 2004 ). Using an antibody that recognizes substrates phosphorylated by PKA (pPKA-substrates) ( Schmitt and Stork, 2002 ), we performed whole-brain immunolabeling on 5 dpf wild-type and nf1 mutant larval brains. nf1 heterozygous mutant and nf1 double homozygous mutant larvae were separated into groups by hyperpigmentation phenotype. To enable the pPKA-substrate antibody to penetrate the brain, we dissected brain tissue from larvae and used a staining protocol that prevented us from registering images to the Z-BRAIN reference atlas. Instead, we created average fluorescence summation projections ( Fig. 4B ) to quantify pPKA-substrate fluorescence in whole brains ( Fig. 4C ). We found that nf1 heterozygous mutants have significantly lower pPKA-substrate fluorescence than wild-type larvae and that nf1 double homozygous mutants have significantly lower pPKA-substrate fluorescence than wild-type and nf1 heterozygous mutants. With plerixafor treatment, we found significantly increased pPKA-substrate fluorescence in nf1 heterozygous mutant larvae only. These results are consistent with the whole-body cAMP measurements in that all nf1 mutants have suppressed pPKA levels and Cxcr4 inhibition raises pPKA levels in nf1 heterozygous mutant larvae. Overall, this supports the conclusion that downregulated cAMP-PKA signaling in neurofibromin-deficiency can be rescued by Cxcr4 inhibition in nf1 heterozygous mutants, which model the autosomal dominant human disease.

DISCUSSION

Translation of research in animal models into treatments for NF1-associated cognitive dysfunction has not yet been successful, in part due to the complicated, multi-functional roles that neurofibromin plays in the distinct neural circuits and intracellular signaling cascades that regulate cognitive function. Here, we identify Cxcr4 chemokine receptor signaling as a regulator of habituation, a sensory filtering behavior and attention-based learning process similar to the cognitive behavior processes that are impaired in NF1 patients. While Cxcr4 signaling can regulate both the Ras and cAMP signaling pathways, we find that Cxcr4 predominantly affects cAMP-PKA signaling in nf1 mutant larval zebrafish. Previous work showed that CXCR4 expression is elevated in malignant peripheral nerve sheath tumors (MPNSTs), and CXCR4 signaling was suggested as a target for treating neurofibromin-deficient tumorigenesis ( Mo et al., 2013 ; Warrington et al., 2007 ). However, roles in NF1 cognitive dysfunction have not been explored. Our results provide evidence that Cxcr4 signaling may also regulate neurofibromin-dependent cognitive function and, thus, is a potential therapeutic target for NF1 cognitive deficits.

To identify molecular pathways that regulate neurofibromin-dependent behaviors, we chose to measure PPI and habituation of the SLC response in larval zebrafish. PPI and habituation learning deficits have previously been identified in neurofibromin-deficient mice and homozygous mutant zebrafish ( Li et al., 2005 ; Randlett et al., 2019 ; Shin et al., 2012 ; Wolman et al., 2014 ). To develop a model that more closely resembles the heterozygous loss-of-function mutations observed in patients with NF1, we modified our behavioral assay to increase the interstimulus interval between acoustic stimuli during the habituation phase. This modification revealed behavioral deficits in nf1 heterozygous mutants with at least one functioning allele of nf1 , thus allowing us to better model the human disease.

Our behavioral drug screening experiments implicate Cxcr4 signaling as a modifier of behavioral deficits in nf1 mutants. By testing multiple doses of two formulations of the Cxcr4 inhibitor plerixafor in nf1 mutants, we found robust increases in habituation at doses that did not affect baseline reactions to high-intensity stimuli, and that did not alter the kinematic properties of turning behavior. These findings suggest that Cxcr4 signaling specifically regulates habituation in nf1 mutants without altering general motor performance. An important caveat to pharmacological approaches is the possibility for off-target effects, and indeed plerixafor, first thought to be a specific Cxcr4 inhibitor, can also stimulate Ackr3 (formerly Cxcr7) at higher concentrations. Our data showing that the specific Ackr3 agonist VUF11207 did not rescue habituation in nf1 mutants supports the idea that the action of plerixafor in nf1 mutants is via Cxcr4 inhibition.

NF1 and Cxcr4 signaling both affect the Ras-Raf-MEK-ERK and cAMP-PKA signaling pathways, but in opposite directions; NF1 decreases Ras activity and increases cAMP activity, while Cxcr4 increases Ras activity and suppresses cAMP activity. Thus, the finding that Cxcr4 inhibition rescues nf1 mutant behavioral defects is, perhaps, not surprising but prompts the question of the molecular mechanisms by which Cxcr4 signaling influences behavior in nf1 mutants. Our finding that cAMP and PKA signaling were reduced in nf1 mutant larvae, while increases in phosphorylation of ERK1/2 were limited and region-specific, suggests that cAMP signaling is the primary pathway influencing habituation behavior. These results are consistent with past work that investigated the molecular basis for learning and memory deficits in nf1 mutant zebrafish and Drosophila , and found that learning defects were rescued by manipulations that increase cAMP levels, while memory defects were rescued by inhibition of Ras signaling ( Brown et al., 2010b ; Ho et al., 2007 ; Tong et al., 2002 ; Wolman et al., 2014 ).

Our finding that Cxcr4 inhibition increased whole-body cAMP levels and brain pPKA-substrate immunofluorescence in nf1 heterozygous but not homozygous larvae suggests that a component of the Cxcr4 effect on cAMP and pPKA is neurofibromin-dependent. One potential explanation for this result is a negative feedback mechanism in which cAMP-bound PKA regulatory subunits interact with Gα i and enhance Gα i signaling ( Stefan et al., 2011 ) ( Fig. 4D ). High cAMP levels inhibit further adenylyl cyclase activity by potentiating G i protein-coupled receptors, such as Cxcr4. Therefore, the greater effect of Cxcr4 inhibition on cAMP-PKA signaling in nf1 heterozygous compared to nf1 homozygous larvae may be due to the higher baseline cAMP levels in heterozygous animals generated by the presence of some functioning neurofibromin. Surprisingly, we found that plerixafor treatment in wild-type larvae did not have a measurable effect on pPKA immunofluorescence in the brain. This finding may reflect a key role for neurofibromin in maintaining signaling pathway homeostasis. Under wild-type conditions and normal neurofibromin levels, the effects of plerixafor treatment on pPKA substrates may be transient and not detectable with the whole-brain immunohistochemistry method. Cxcr4 and neurofibromin both influence multiple downstream signal transduction pathways with extensive possibilities for cross-talk and feedback within pathways. This complexity underscores the need for continued research into molecular mechanisms of NF1 disease.

It is interesting that total loss of neurofibromin in nf1 double homozygous mutants does not lead to more global increases of pERK levels, as nf1a and nf1b are broadly expressed in the larval zebrafish brain ( Padmanabhan et al., 2009 ). One possible explanation is that phosphorylation of ERK1/2 could be regulated by altered neuronal calcium signaling in addition to neurofibromin-dependent increases in Ras signaling nf1 mutants ( Rosen et al., 1994 ; Thomas and Huganir, 2004 ). Dysregulation of intracellular calcium in neurons could potentially diminish the hyperactivation of Ras caused by loss of neurofibromin. Previous work in mice shows that loss of neurofibromin can lead to increased gamma aminobutyric acid (GABA) release in neural circuits ( Cui et al., 2008 ; Shilyansky et al., 2010 ), which could inhibit calcium influx and thus inhibit the activation of Ras in affected brain regions.

Through MAP-mapping analysis, we did identify one brain region with increased levels of pERK that could, potentially, regulate acoustically evoked startle habituation – the torus semicircularis ( Fig. 3A ; Table 3 ). In larval zebrafish, the torus semicircularis relays auditory information to the thalamus but also signals back to the hindbrain ( Vanwalleghem et al., 2017 ), where it could, potentially, provide top-down regulation of hindbrain activity to regulate acoustic startle habituation. However, pERK in this region is not reduced in nf1 heterozygous mutants by Cxcr4 inhibition ( Fig. 3D ; Table 4 ). Therefore, Cxcr4 signaling is unlikely to regulate habituation through modulation of the canonical Ras-Raf-MEK-ERK signaling pathway. This is consistent with a previous report that showed MEK inhibition with U0126 does not improve short-term habituation of the SLC response in larval zebrafish ( Wolman et al., 2014 ). However, neurofibromin has also been shown to regulate cellular signaling through non-canonical Ras pathways ( Anastasaki and Gutmann, 2014 ). Ras can activate atypical protein kinase C zeta (PKCζ) to regulate cAMP homeostasis. If neurofibromin regulates cAMP through a Ras-PKCζ pathway in zebrafish larvae, it would not be detected with the pERK MAP-mapping technique used in this study. Therefore, we cannot rule out a potential role for non-canonical Ras signaling in the dysregulation of cAMP homeostasis we observed in nf1 mutants.

Other molecular targets that regulate neurofibromin-dependent behaviors

It is important to consider if the biological actions of the target compounds identified by us are specific to loss of neurofibromin or if they affect behavior more generally. A small-molecule screen measuring short-term habituation of the SLC response has previously been performed in wild-type larvae ( Wolman et al., 2011 ). Target biological pathways identified in both our nf1 mutant screen and the previous wild-type screen, including the 5-HT receptor pathway, may not be specific to loss of neurofibromin ( Table 1 ; Wolman et al., 2011 ). Therefore, the effects of modulating serotonin signaling on habituation learning may not be specific to loss of neurofibromin. However, neurofibromin has been shown to modulate 5-HT 6 receptor constitutive activity in mouse neurons ( Deraredj Nadim et al., 2016 ). Therefore, the relationship between neurofibromin and 5-HT receptor activity warrants further consideration. With respect to the compounds that target adrenergic receptors identified in our screen, the alpha-2 adrenergic receptor agonists are known to have sedative effects in humans and zebrafish ( Giovannitti et al., 2015 ; Ruuskanen et al., 2005 ). Therefore, the effects of these compounds also may not be specific to habituation.

The compounds identified in our screen that improved PPI in nf1 mutants were enriched for those targeting biological pathways that include 5-HT receptors, phosphodiesterase inhibitors and topoisomerase inhibitors ( Table 1 ). Modulation of 5-HT receptor signaling has previously been shown to regulate PPI in wild-type animals ( Fletcher et al., 2001 ; Rigdon and Weatherspoon, 1992 ; Sipes and Geyer, 1994 ). Therefore, the effects of modulating serotonin signaling on PPI may not be specific to loss of neurofibromin. Phosphodiesterase (PDE) inhibitors increase cAMP signaling by preventing the degradation of cAMP. Interestingly, PDE inhibitors have previously been shown to improve habituation in nf1 double homozygous mutants ( Wolman et al., 2014 ) but their effects on PPI were unknown. Although compounds that improved habituation in our screen were not enriched for PDE inhibitors, we did identify rolipram and doxofylline as two PDE inhibitors that improve habituation in nf1 mutants ( Table S1 ). Topoisomerases bind to and cut the DNA phosphate backbone and are used in medicine as antibiotics and chemotherapy agents. It is unclear how topoisomerase inhibitors might regulate PPI in nf1 mutants. However, topoisomerase inhibitors can activate epigenetically silenced ubiquitin protein ligase alleles in neurons associated with Angelman syndrome ( Huang et al., 2011 ), and mouse models for Angelman syndrome display deficits in acoustic startle and PPI ( Huang et al., 2013 ). Further research will be required to understand how these effects might relate to neurofibromin.

Overall, we identify Cxcr4 chemokine receptor signaling as a regulator of habituation and modulator of cAMP-PKA signaling in nf1 mutant larval zebrafish, suggesting that Cxcr4 signaling may regulate neurofibromin-dependent cognitive function in addition to its roles in tumorigenesis. Clinically, plerixafor is known to mobilize stem cells from the bone marrow to the blood stream and has originally been approved by the United States Food and Drug Administration (FDA) in 2008 for autologous transplantation in patients with Non-Hodgkin's Lymphoma or multiple myeloma ( De Clercq, 2019 ). Since then, research with plerixafor has highlighted its potential as a treatment option for various cancers ( De Clercq, 2019 ), including the treatment of NF1-associated malignant peripheral nerve sheath tumors ( Mo et al., 2013 ). Our work here highlights the potential for more preclinical investigation into the relationship between CXCL12-CXCR4 signaling and neurofibromin-dependent cognitive functions. Further, if CXCR4 inhibitor treatment advances toward clinical trials to treat tumors in patients with neurofibromatosis type 1, cognitive function could be considered as an additional clinical outcome.

MATERIALS AND METHODS

Zebrafish maintenance and breeding

Zebrafish ( Danio rerio ) larvae used in this study were generated from crosses of wild-type (Tüpfel long fin background) adults or nf1 mutant adults carrying a combination of nf1a Δ5 ( nf1a − ) and nf1b +10 ( nf1b − ) mutant alleles ( Shin et al., 2012 ). Crossing nf1a +/− ; nf1b −/− and nf1a −/− ; nf1b +/− adults yielded embryos with nf1a +/− ; nf1b +/− , nf1a +/− ; nf1b −/− , nf1a −/− ; nf1b +/− or nf1a −/− ; nf1b −/− mutant alleles. Genotyping of nf1a Δ5 and nf1b +10 alleles was performed as previously described ( Shin et al., 2012 ). In indicated experiments, nf1 mutant larvae were classified by the presence of a hyperpigmentation phenotype in nf1a −/− ; nf1b −/− larvae. The resulting two groups of nf1 mutant larvae are putative nf1 heterozygous ( nf1a +/− ; nf1b +/− , nf1a +/− ; nf1b −/− , nf1a −/− ; nf1b +/− ) and putative nf1 double homozygous ( nf1a −/− ; nf1b −/− ) mutants. This method of identification was validated by post hoc genotyping (100% positive predictive value and 93% negative predictive value for nf1a −/− ; nf1b −/− ; n =94).

Embryos and larvae were raised at 29°C in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4 adjusted to pH 6.8–6.9 with NaHCO 3 ) on a 14/10 h light/dark cycle. E3 medium was changed at 48 and 96 h post fertilization (hpf). Before behavior testing, larvae were maintained on a white light box (800 μW cm −2 ) for at least 1 h before being transferred to the testing arena. All experiments were conducted between 5- and 6-days post fertilization (dpf), which is before sex determination. All animals in these studies were handled in accordance with the National Institutes of Health Guide for the care and use of laboratory animals, and the University of Wisconsin Institutional Animal Care and Use Committee (IACUC). These studies were approved by the University of Wisconsin IACUC (protocol L005704).

Behavioral recording and tracking

Behavioral responses to acoustically evoked startles were recorded as previously described ( Burgess and Granato, 2007a ; Wolman et al., 2011 ). Briefly, larvae housed individually in acrylic grids were exposed to acoustic/vibrational stimuli with a small vibration exciter (4810; Brüel and Kjaer, Norcross, GA, USA) controlled by a digital–analog card (PCI-6221; National Instruments, Austin, TX, USA). Larvae were illuminated from above with LED lights (MCWHL5 6500K LED, powered by LEDD1B driver, Thorlabs, NJ, USA) and below with infrared lights (IR Illuminator CM-IR200B, C&M Vision Technologies, Houston, TX, USA). Larval movement was recorded using MotionPro Y4 high-speed video cameras (Integrated Design Tools, Pasadena, CA, USA) with 50 mm macro lenses (Sigma Corporation of America, Ronkonkoma, NY, USA) at 1000 frames per second at either 512×512 or 1024×1024 pixel resolution.

Video images were analyzed for movement initiations and kinematics with the open-source Flote (v-2.0; https://www.nichd.nih.gov/about/org/dir/affinity-groups/AMHD/burgess/software/flote-downloads ) software package as previously described ( Burgess and Granato, 2007a , b ; Wolman et al., 2011 ). Larvae that failed to initiate any movement to 60% or more of stimuli were excluded. Data from individual larvae were matched to genotype information using a custom R script that is available upon request.

Drug screen behavioral assay

For the drug screen behavioral assay, nf1 mutant larvae ( nf1a +/− ; nf1b +/− , nf1a +/− ; nf1b −/− , nf1a −/− ; nf1b +/− , or nf1a −/− ; nf1b −/− mutant alleles, expected ratio 25% each) were tested at 5 dpf. A total of 60 acoustic stimuli were delivered with varying intensities and interstimulus intervals (ISIs) to assess baseline startle initiation, prepulse inhibition (PPI), and habituation ( Fig. 1A ). During the PPI phase (stimuli 1-30), a series of low-intensity, high-intensity and paired low-high-intensity acoustic stimuli were delivered, separated by 20 s ISIs, with each stimulus repeated ten times. The paired low-high-intensity stimuli were separated by 400 ms. During the proceeding habituation phase (stimuli 31-60), 30 high-intensity acoustic stimuli were delivered, separated by 1.5 s ISIs.

Baseline startle initiation probability was assessed by measuring initiations (total turns or SLCs, separately) to both low- and high-intensity stimuli during the PPI phase (stimuli 1, 2, 4, 5, 7, 8, 10, 11, 13, 14, 16, 17, 19, 20, 22, 23, 25, 26, 28, 29). The number of response initiations divided by the total analyzed was reported as the turn or SLC initiation percentage.

Habituation was assessed by calculating two ratios of SLC response initiation. The first was a ratio of SLC initiation in response to habituation phase stimuli 41-50 over SLC initiation in response to the first 10 high-intensity stimuli (PPI phase stimuli 2, 5, 8, 11, 14, 17, 20, 23, 26, 29). The second was a ratio of SLC initiation in response to stimuli 51-60 over SLC initiation in response to the first 10 high-intensity stimuli (PPI phase stimuli 2, 5, 8, 11, 14, 17, 20, 23, 26, 29). The habituation percentage is reported as 1 minus these ratios.

PPI was assessed by calculating the ratio of SLC initiations in response to paired low-high-intensity stimuli (PPI phase stimuli 3, 6, 9, 12, 15, 18, 21, 24, 27, 30) over SLC initiations in response to unpaired high-intensity stimuli (PPI phase stimuli 2, 5, 8, 11, 14, 17, 20, 23, 26, 29). The PPI percentage is reported as 1 minus this ratio.

Habituation behavioral assay

The habituation behavioral assay was designed to be sensitive to more modest habituation deficits than the drug screen behavioral assay. Wild-type or nf1 mutant larvae (genotyped post hoc ) were tested in this assay at 5 dpf. A total of 50 acoustic stimuli were delivered with varying intensities and ISIs to assess baseline startle initiation and habituation ( Fig. 2A ). During the pre-habituation phase (stimuli 1-20), ten low-intensity stimuli were delivered, followed by ten high-intensity stimuli, all separated by 30 s ISIs. During the proceeding habituation phase (stimuli 21-50), 30 high-intensity stimuli were delivered, separated by 3 s ISIs.

Baseline startle initiation probability was assessed by measuring initiations (total turns or SLCs, separately) to both low and high-intensity stimuli during the pre-habituation phase (stimuli 1-20). The number of response initiations divided by the total analyzed was reported as the startle initiation percentage.

Habituation was assessed by calculating the ratio of SLC initiations in response to the final ten high-intensity stimuli (habituation phase stimuli 41-50) over SLC initiations in response to the first ten high-intensity stimuli (pre-habituation phase stimuli 11-20). The habituation percentage is reported as 1 minus this ratio.

Pharmacology

For the small-molecule screen, 1134 bioactive compounds from an FDA-approved small-molecule library (Selleck) were tested on nf1 mutant larvae. Compounds were maintained at −80°C dissolved in DMSO and were administered to larvae at a final concentration of 1 µM or 10 µM per compound (in 1% DMSO in E3 medium) for a treatment duration of 1 h before behavioral testing. Each compound was administered to 32 larvae (mix of nf1 mutant genotypes). Two groups of 32 larvae (mix of nf1 mutant genotypes) were left untreated on each experimental day. Behavioral testing was completed over the course of 60 experimental days.

For follow-up habituation experiments, two formulations of the Cxcr4 inhibitor AMD3100 (plerixafor 8HCl and plerixafor; Selleck) were tested on wild-type and nf1 mutant larvae (genotyped post hoc ). Both formulations were dissolved in water and administered to larvae in a series of doses for a treatment duration of 1 h before behavioral testing. For plerixafor 8HCl, the final concentrations in E3 medium were 300 nM, 1 µM, and 3 µM. For plerixafor, the final concentrations in E3 medium were 1 µM, 3 µM, and 10 µM. For remaining experiments (measures of ERK1/2, cAMP, and PKA signaling), the 3 µM plerixafor dose was used as treatment for a duration of 1 h.

RNA sequencing and differential gene expression analysis

RNA was isolated from groups of 20 whole-larvae at 5 dpf using TRIzol Reagent (Invitrogen) following the manufacturer's instructions. Eight biological replicates for wild-type (Tüpfel long fin background) and putative nf1 double mutants were processed for RNA sequencing. RNA sequencing was carried out by the Biotechnology Center Gene Expression Center at the University of Wisconsin-Madison on an Illumina HiSeq 2500 platform. Transcriptomic data processing adhered to ENCODE Consortium guidelines and best practices for RNA-Seq ( ENCODE Project Consortium, 2016 ). Raw strand-specific paired-end (2×150 bp) Illumina reads were adapter-trimmed using Skewer (v0.1.123; Jiang et al., 2014 ; https://sourceforge.net/projects/skewer ). Trimmed reads were aligned to the Danio rerio GRCz11 reference genome (NCBI Assembly: GCA_000002035.4) using STAR (v2.5.3a; Dobin et al., 2013 ; https://github.com/alexdobin/STAR ), a splice-aware aligner, with gene annotation from Ensembl release 102 ( https://useast.ensembl.org/index.html ). Gene- and isoform-level expression quantification was performed with RSEM (v1.3.1; Li and Dewey, 2011 ; https://github.com/deweylab/RSEM ), which estimates transcript abundance using expectation-maximization. For downstream differential gene expression analysis, expected read counts from RSEM were rounded to the nearest integer and imported into edgeR (v3.28.0; Robinson et al., 2010 ; https://bioconductor.org/packages/release/bioc/html/edgeR.html ), run within the R statistical environment (v3.6.2; https://www.r-project.org ) Library size normalization across samples was performed using the trimmed mean of M-values (TMM) method ( Robinson and Oshlack, 2010 ). Genes were retained for statistical testing if they exceeded a count-per-million (CPM) threshold in a minimum number of samples, defined by the number of biological replicates per condition, with a minimum read count of ten. This independent filtering step reduced noise and increased statistical power. Differential expression testing was conducted using negative binomial generalized linear models. Resulting P -values were adjusted for multiple testing using the Benjamini–Hochberg false discovery rate (FDR) procedure, with a significance threshold of 5% ( Reiner et al., 2003 ). The validity of the multiple testing correction was assessed by visual inspection of the distribution of unadjusted P -values. KEGG pathway analysis was performed as described by Subhash and Kanduri (2016) .

Immunohistochemistry and confocal imaging

Immunostaining and whole-brain imaging for total-ERK1/2 (tERK) and phosphorylated-ERK1/2 (pERK) was performed as previously described ( Randlett et al., 2015 ). Larvae at 6 dpf were fixed in 4% paraformaldehyde (diluted to 4% w/v in PBS from 16% w/v in 0.1 M phosphate buffer, 0.25% v/v Triton X-100 pH 7.4) overnight at 4°C. The next day, larvae were incubated in bleaching solution (3% v/v hydrogen peroxide, 1% w/v KOH, in water) for 30 min to remove pigment. Larvae were then incubated in 150 mM Tris HCl pH 9.0 for 15 min at 70°C. Larvae were permeabilized in 0.05% Trypsin-EDTA on ice for 45 min. Larvae were then incubated for 1 h in block solution (1% w/v bovine serum albumin, 2% v/v normal goat serum, 0.25% v/v Triton X-100, 1% v/v DMSO, in PBS pH 7.4). Larvae were then incubated in primary antibodies at 1:300 in incubation buffer (IB; 1% w/v bovine serum albumin, 0.25% v/v Triton X-100, 1% v/v DMSO, in PBS pH 7.4) overnight at 4°C. The next day, larvae were incubated in fluorescently conjugated secondary antibodies at 1:500 in IB overnight at 4°C and stored in Vectashield mounting medium (Vector Laboratories) until mounting for imaging. Primary antibodies included phospho-ERK1/2 antibody (Cell Signaling, #4370) and total-ERK1/2 antibody (Cell Signaling, #4696). Secondary antibodies included Alexa Fluor 488-conjugated goat anti-mouse IgG1 and Alexa Fluor 594-conjugated goat anti-rabbit IgG (Thermo Fisher Scientific, #A-21121, #A-11008) Whole-brain image stacks were acquired on an Olympus FLUOVIEW confocal laser scanning microscope (FV1000) using a 20× oil immersion objective and Fluoview software (FV10-ASW 4.2; https://evidentscientific.com/en/downloads ). Images were acquired at a x/y/z resolution of 0.795/0.795/2 μM. The entire brain was captured by imaging in two sections and stitching images using the Pairwise Stitching plugin in FIJI ( Preibisch et al., 2009 ). Pixel saturation was adjusted to 0.1% of the most intense pixels for each sample using a custom FIJI macro available upon request.

Whole-brain image stacks for total-ERK1/2 (tERK) and phosphorylated-ERK1/2 (pERK) were registered to the open-source Z-BRAIN atlas using the Jefferis lab graphical user interface for the Computational Morphometry Toolkit (CMTK) hosted on GitHub. Parameters used included: -awr 0102 -X 52 -C 8 -G 80 -R 3 -A ‘--accuracy 0.4’ -W ‘--accuracy 1.6’. Images were downsampled and smoothed using the PrepareStacksForMAPMapping FIJI macro. MAP-maps were created using the MakeTheMAPMap MATLAB script. Z-BRAIN analysis of MAP-maps was generated using the ZbrainAnalysisOfMAPMaps. FIJI and MATLAB scripts available from the open-source zebrafish atlas Z-BRAIN ( https://zebrafishexplorer.zib.de/home/ ) hosted by the Zuse Institut Berlin (ZIB), Berlin, Germany.

Immunostaining and whole-brain imaging was performed on dissected brains for phosphorylated-PKA substrates (pPKA-substrates). Larvae at 5 dpf were fixed in 4% paraformaldehyde (diluted to 4% w/v in PBS from 16% w/v in 0.1 M phosphate buffer, 5% w/v sucrose pH 7.4) for 1 h at room temperature. Brains were dissected from the head using fine tungsten dissecting needles. Brains were permeabilized in collagenase (0.1% w/v in PBS) for 1 h and blocked for 1 h at room temperature in incubation buffer (IB; 0.2% w/v bovine serum albumin, 2% v/v normal goat serum, 0.8% v/v Triton X-100, 1% v/v DMSO, in PBS pH 7.4). Brains were incubated in primary antibodies at 1:300 in IB overnight at 4°C. Brains were incubated in fluorescently conjugated secondary antibodies at 1:500 in IB overnight at 4°C and stored in Vectashield mounting medium (Vector Laboratories) until mounting for imaging. Primary antibodies against phospho-(Ser/Thr) PKA substrate (Cell Signaling, #9621) and against total-ERK1/2 (Cell Signaling, #4696) were used. Secondary antibodies included Alexa Fluor 488-conjugated goat anti-mouse IgG1 and Alexa Fluor 594-conjugated goat anti-rabbit IgG (Thermo Fisher Scientific, #A-21121, #A-11008). Whole-brain image stacks were acquired on an Olympus Fluoview confocal laser scanning microscope (FV1000) using a 20× oil immersion objective and Fluoview software (FV10-ASW 4.2; https://evidentscientific.com/en/downloads ). Images were acquired at a x/y/z resolution of 0.795/0.795/2 μM.

Whole-brain image stacks for pPKA-substrates were processed using the open-source software FIJI (v-1.53c) ( Schindelin et al., 2012 ). Whole-brain pPKA-substrates fluorescence was measured by drawing a region of interest (ROI) around the entire brain from Z-stack average intensity projections. Background fluorescence was measured by drawing a ROI away from sample in the same projection. The total corrected fluorescence was quantified using the following calculation: integrated density−(area of ROI×mean fluorescence of the background). Fluorescence was reported as total corrected fluorescence per total brain area.

cAMP enzyme-linked immunosorbent assay (ELISA)

For cAMP quantification, zebrafish larvae at 5 dpf were snap-frozen in liquid nitrogen. Tissue from 15 larvae per group was lysed by homogenizing in 200 µl of 0.1 M HCl using a hand-held homogenizer. Samples were spun at ∼10,000 g for 15 min at 4°C. The supernatant was transferred to a new microcentrifuge tube and diluted 1:1 in 0.1 M HCl. cAMP concentration was quantified using a direct cAMP ELISA kit (Enzo Life Science, ADI900066) following the acetylated version of the manufacturer's instructions. Three technical replicates were averaged for each sample reported.

Statistical analysis

Descriptive statistics (mean, standard deviation, standard error) and z -score were calculated in the software environments RStudio (v-1.3; https://posit.co/download/rstudio-desktop/ ), GraphPad Prism (v-9.1.2, GraphPad Software Incorporated; https://www.graphpad.com ), and MATLAB (v-9.6.0, R2019A, The MathWorks Inc.; https://www.mathworks.com/products/matlab.html ). The z -scores for the effects of individual compounds in the small-molecule screen were calculated with the following formula: z -score=(mean treated value − mean non-treated control value)/(standard deviation non-treated control value). The z -scores were identified as significant by setting a critical value based on the one-tailed 99% confidence interval. Enrichment analysis for target compounds was performed using Fisher's exact test. Assumptions of normality were tested by Shapiro-Wilk's test. One-way ANOVA, two-way ANOVA, and Tukey's and Dunnet's multiple comparisons testing were performed with a significance level of 0.05. Gene expression analysis was performed using R (v-3.6.2; https://www.r-project.org ). Data are presented as the mean±standard error of the mean (±s.e.m.). Sample sizes and statistical tests for specific experiments are identified in the figures and figure legends.

Supplementary Material

Acknowledgements

The authors thank Dr Marc Wolman for technical advice on the small-molecule compound behavioral screen, Dr Owen Randlett (MeLiS Institute, Lyon, France) for technical advice on the MAP-mapping procedure, Dr Changjiu Zhao (University of Wisconsin-Madison, Madison, WI, USA) for technical assistance with ELISA experiments, Riley Kelley, Morgan Wetzel, Hollis Howe, Liza Rick, Hannah Olson, Jessa Snower and Blanche Danecek for technical assistance (University of Wisconsin-Madison, Madison, WI, USA), Dr Michael Granato (University of Pennsylvania, Philadelphia, PA, USA) for the nf1 mutant fish lines. We thank the University of Wisconsin-Madison Biotechnology Center: Gene Expression Center (RRID:SCR_017757) for RNA library preparation; DNA Sequencing Facility (RRID:SCR_017759) for sequencing; and Bioinformatics Resource Center (RRID:SCR_017799) for analysis services.

Footnotes

Author contributions

Conceptualization: A.H.M., M.C.H.; Formal analysis: A.H.M., Y.Y., N.S., M.E.B.; Funding acquisition: M.C.H.; Investigation: A.H.M., Y.Y., N.S., J.M.; Resources: M.C.H.; Supervision: M.C.H.; Validation: M.E.B.; Visualization: A.H.M.; Writing – original draft: A.H.M.; Writing – review & editing: A.H.M., M.E.B., M.C.H.

Funding

This work was supported by the National Institutes of Health [T32GM007507 to A.H.M., 1R01NS086934 to M.C.H., R56NS132890 to M.C.H.], University of Wisconsin NF1 foundation funds (administered by Charles Konsitzke), University of Wisconsin Hilldale Undergraduate Research Award to N.S., and U.S. Department of Defense Congressionally Directed Medical Research Program [W81XWH-16-1-0091 to M.A. Wolman]. Open Access funding provided by University of Wisconsin. Deposited in PMC for immediate release.

Data and resource availability

Larval zebrafish behavior tracking was implemented using the open-source Flote (v-2.0) software package hosted by the Burgess lab at the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) ( https://www.nichd.nih.gov/research/atNICHD/Investigators/burgess/software ). MAP-mapping experiments were conducted using the open-source Z-BRAIN atlas hosted by the Engert Lab and Zuse Institut Berlin ( https://zebrafishexplorer.zib.de/home/ ). Analysis scripts and viewer function for Z-Brain were implemented following ( Randlett et al., 2015 ) available on GitHub ( https://github.com/owenrandlett/Z-Brain ). RNA sequencing data generated in this study have been deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) repository and are accessible through accession number GSE317922 ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE317922 ). All other relevant data and details of resources can be found within the article and its supplementary information .

References

Anastasaki , C. and Gutmann , D. H. ( 2014 ). Neuronal NF1/RAS regulation of cyclic AMP requires atypical PKC activation . Hum. Mol. Genet. 23 , 6712 - 6721 . 10.1093/hmg/ddu389 25070947 PMC4245041

Basu , T. N. , Gutmann , D. H. , Fletcher , J. A. , Glover , T. W. , Collins , F. S. and Downward , J. ( 1992 ). Aberrant regulation of ras proteins in malignant tumour cells from type 1 neurofibromatosis patients . Nature 356 , 713 - 715 . 10.1038/356713a0 1570015

Bearden , C. E. , Hellemann , G. S. , Rosser , T. , Montojo , C. , Jonas , R. , Enrique , N. , Pacheco , L. , Hussain , S. A. , Wu , J. Y. , Ho , J. S. et al. ( 2016 ). A randomized placebo-controlled lovastatin trial for neurobehavioral function in neurofibromatosis I . Ann. Clin. Transl. Neurol 3 , 266 - 279 . 10.1002/acn3.288 27081657 PMC4818747

Bonecchi , R. and Graham , G. J. ( 2016 ). Atypical chemokine receptors and their roles in the resolution of the inflammatory response . Front. Immunol. 7 , 224 . 10.3389/fimmu.2016.00224 27375622 PMC4901034

Brown , J. A. , Emnett , R. J. , White , C. R. , Yuede , C. M. , Conyers , S. B. , O'Malley , K. L. , Wozniak , D. F. and Gutmann , D. H. ( 2010a ). Reduced striatal dopamine underlies the attention system dysfunction in neurofibromatosis-1 mutant mice . Hum. Mol. Genet. 19 , 4515 - 4528 . 10.1093/hmg/ddq382 20826448 PMC2957316

Brown , J. A. , Gianino , S. M. and Gutmann , D. H. ( 2010b ). Defective cAMP generation underlies the sensitivity of CNS neurons to neurofibromatosis-1 heterozygosity . J. Neurosci. 30 , 5579 - 5589 . 10.1523/JNEUROSCI.3994-09.2010 20410111 PMC2864934

Brown , J. A. , Xu , J. , Diggs-Andrews , K. A. , Wozniak , D. F. , Mach , R. H. and Gutmann , D. H. ( 2011 ). PET imaging for attention deficit preclinical drug testing in neurofibromatosis-1 mice . Exp. Neurol. 232 , 333 - 338 . 10.1016/j.expneurol.2011.09.005 21963652 PMC3202049

Brown , J. A. , Diggs-Andrews , K. A. , Gianino , S. M. and Gutmann , D. H. ( 2012 ). Neurofibromatosis-1 heterozygosity impairs CNS neuronal morphology in a cAMP/PKA/ROCK-dependent manner . Mol. Cell. Neurosci. 49 , 13 - 22 . 10.1016/j.mcn.2011.08.008 21903164 PMC3237958

Burgess , H. A. and Granato , M. ( 2007a ). Sensorimotor gating in larval zebrafish . J. Neurosci. 27 , 4984 - 4994 . 10.1523/JNEUROSCI.0615-07.2007 17475807 PMC6672105

Burgess , H. A. and Granato , M. ( 2007b ). Modulation of locomotor activity in larval zebrafish during light adaptation . J. Exp. Biol. 210 , 2526 - 2539 . 10.1242/jeb.003939 17601957

Busillo , J. M. and Benovic , J. L. ( 2007 ). Regulation of CXCR4 signaling . Biochim. Biophys. Acta 1768 , 952 - 963 . 10.1016/j.bbamem.2006.11.002 17169327 PMC1952230

Cassel , J.-C. ( 2010 ). Experimental Studies on the Role(s) of Serotonin in Learning and Memory Functions . In Handbook of the Behavioral Neurolobiology of Serotonin ( ed. C. P. Müller and B. L. Jacobs ), pp. 429 - 447 . Elsevier .

Cimino , P. J. and Gutmann , D. H. ( 2018 ). Neurofibromatosis type 1 . Handb. Clin. Neurol. 148 , 799 - 811 . 10.1016/B978-0-444-64076-5.00051-X 29478615

Cui , Y. , Costa , R. M. , Murphy , G. G. , Elgersma , Y. , Zhu , Y. , Gutmann , D. H. , Parada , L. F. , Mody , I. and Silva , A. J. ( 2008 ). Neurofibromin regulation of ERK signaling modulates GABA release and learning . Cell 135 , 549 - 560 . 10.1016/j.cell.2008.09.060 18984165 PMC2673196

Dasgupta , B. , Yi , Y. , Chen , D. Y. , Weber , J. D. and Gutmann , D. H. ( 2005 ). Proteomic analysis reveals hyperactivation of the mammalian target of rapamycin pathway in neurofibromatosis 1-associated human and mouse brain tumors . Cancer Res. 65 , 2755 - 2760 . 10.1158/0008-5472.CAN-04-4058 15805275

De Clercq , E. ( 2019 ). Mozobil ® (Plerixafor, AMD3100), 10 years after its approval by the US Food and Drug Administration . Antivir. Chem. Chemother. 27 , 2040206619829382 . 10.1177/2040206619829382 30776910 PMC6379795

Declue , J. E. , Cohen , B. D. and Lowy , D. R. ( 1991 ). Identification and characterization of the neurofibromatosis type 1 protein product . Proc. Natl. Acad. Sci. USA 88 , 9914 - 9918 . 10.1073/pnas.88.22.9914 1946460 PMC52837

Deraredj Nadim , W. , Chaumont-Dubel , S. , Madouri , F. , Cobret , L. , De Tauzia , M.-L. , Zajdel , P. , Bénédetti , H. , Marin , P. and Morisset-Lopez , S. ( 2016 ). Physical interaction between neurofibromin and serotonin 5-HT6 receptor promotes receptor constitutive activity . Proc. Natl. Acad. Sci. USA 113 , 12310 - 12315 . 10.1073/pnas.1600914113 27791021 PMC5087017

Diggs-Andrews , K. A. , Tokuda , K. , Izumi , Y. , Zorumski , C. F. , Wozniak , D. F. and Gutmann , D. H. ( 2013 ). Dopamine deficiency underlies learning deficits in neurofibromatosis-1 mice . Ann. Neurol. 73 , 309 - 315 . 10.1002/ana.23793 23225063 PMC3608728

Dobin , A. , Davis , C. A. , Schlesinger , F. , Drenkow , J. , Zaleski , C. , Jha , S. , Batut , P. , Chaisson , M. and Gingeras , T. R. ( 2013 ). STAR: ultrafast universal RNA-seq aligner . Bioinformatics 29 , 15 - 21 . 10.1093/bioinformatics/bts635 23104886 PMC3530905

Eaton , R. C. , Bombardieri , R. A. and Meyer , D. L. ( 1977 ). The Mauthner-initiated startle response in teleost fish . J. Exp. Biol. 66 , 65 - 81 . 10.1242/jeb.66.1.65 870603

Endo , M. , Yamamoto , H. , Setsu , N. , Kohashi , K. , Takahashi , Y. , Ishii , T. , Iida , K.-I. , Matsumoto , Y. , Hakozaki , M. , Aoki , M. et al. ( 2013 ). Prognostic significance of AKT/mTOR and MAPK pathways and antitumor effect of mTOR inhibitor in NF1-related and sporadic malignant peripheral nerve sheath tumors . Clin. Cancer Res. 19 , 450 - 461 . 10.1158/1078-0432.CCR-12-1067 23209032

Evans , D. G. , Howard , E. , Giblin , C. , Clancy , T. , Spencer , H. , Huson , S. M. and Lalloo , F. ( 2010 ). Birth incidence and prevalence of tumor-prone syndromes: estimates from a UK family genetic register service . Am. J. Med. Genet. A 152A , 327 - 332 . 10.1002/ajmg.a.33139 20082463

Fletcher , P. J. , Selhi , Z. F. , Azampanah , A. and Sills , T. L. ( 2001 ). Reduced brain serotonin activity disrupts prepulse inhibition of the acoustic startle reflex. Effects of 5,7-dihydroxytryptamine and p-chlorophenylalanine . Neuropsychopharmacology 24 , 399 - 409 . 10.1016/S0893-133X(00)00215-3 11182535

Förster , D. , Arnold-Ammer , I. , Laurell , E. , Barker , A. J. , Fernandes , A. M. , Finger-Baier , K. , Filosa , A. , Helmbrecht , T. O. , Kölsch , Y. , Kühn , E. et al. ( 2017 ). Genetic targeting and anatomical registration of neuronal populations in the zebrafish brain with a new set of BAC transgenic tools . Sci. Rep. 7 , 5230 . 10.1038/s41598-017-04657-x 28701772 PMC5507991

Giovannitti , J. A. , Thoms , S. M. and Crawford , J. J. ( 2015 ). Alpha-2 adrenergic receptor agonists: a review of current clinical applications . Anesth. Prog. 62 , 31 - 38 . 10.2344/0003-3006-62.1.31 25849473 PMC4389556

Gritsina , G. , Fong , K.-W. , Lu , X. , Lin , Z. , Xie , W. , Agarwal , S. , Lin , D. , Schiltz , G. E. , Beltran , H. , Corey , E. et al. ( 2023 ). Chemokine receptor CXCR7 activates Aurora Kinase A and promotes neuroendocrine prostate cancer growth . J. Clin. Investig. 133 , e166248 . 10.1172/JCI166248 37347559 PMC10378179

Guo , H.-F. , The , I. , Hannan , F. , Bernards , A. and Zhong , Y. ( 1997 ). Requirement of Drosophila NF1 for activation of adenylyl cyclase by PACAP38-like neuropeptides . Science 276 , 795 - 798 . 10.1126/science.276.5313.795 9115204

Hannan , F. , Ho , I. , Tong , J. J. , Zhu , Y. , Nurnberg , P. and Zhong , Y. ( 2006 ). Effect of neurofibromatosis type I mutations on a novel pathway for adenylyl cyclase activation requiring neurofibromin and Ras . Hum. Mol. Genet. 15 , 1087 - 1098 . 10.1093/hmg/ddl023 16513807 PMC1866217

Hatse , S. , Princen , K. , Bridger , G. , De Clercq , E. and Schols , D. ( 2002 ). Chemokine receptor inhibition by AMD3100 is strictly confined to CXCR4 . FEBS Lett. 527 , 255 - 262 . 10.1016/S0014-5793(02)03143-5 12220670

Hecker , A. , Schulze , W. , Oster , J. , Richter , D. O. and Schuster , S. ( 2020 ). Removing a single neuron in a vertebrate brain forever abolishes an essential behavior . Proc. Natl. Acad. Sci. USA 117 , 3254 - 3260 . 10.1073/pnas.1918578117 32001507 PMC7022180

Ho , I. S. , Hannan , F. , Guo , H.-F. , Hakker , I. and Zhong , Y. ( 2007 ). Distinct functional domains of neurofibromatosis type 1 regulate immediate versus long-term memory formation . J. Neurosci. 27 , 6852 - 6857 . 10.1523/JNEUROSCI.0933-07.2007 17581973 PMC6672704

Huang , H.-S. , Allen , J. A. , Mabb , A. M. , King , I. F. , Miriyala , J. , Taylor-Blake , B. , Sciaky , N. , Dutton , J. W. , Lee , H.-M. , Chen , X. et al. ( 2011 ). Topoisomerase inhibitors unsilence the dormant allele of Ube3a in neurons . Nature 481 , 185 - 189 . 10.1038/nature10726 22190039 PMC3257422

Huang , H.-S. , Burns , A. J. , Nonneman , R. J. , Baker , L. K. , Riddick , N. V. , Nikolova , V. D. , Riday , T. T. , Yashiro , K. , Philpot , B. D. and Moy , S. S. ( 2013 ). Behavioral deficits in an Angelman syndrome model: effects of genetic background and age . Behav. Brain Res. 243 , 79 - 90 . 10.1016/j.bbr.2012.12.052 23295389 PMC3629944

Huson , S. M. , Compston , D. A. and Harper , P. S. ( 1989 ). A genetic study of von Recklinghausen neurofibromatosis in south east Wales. II. Guidelines for genetic counselling . J. Med. Genet. 26 , 712 - 721 . 10.1136/jmg.26.11.712 2511319 PMC1015741

Hyman , S. L. , Shores , A. and North , K. N. ( 2005 ). The nature and frequency of cognitive deficits in children with neurofibromatosis type 1 . Neurology 65 , 1037 - 1044 . 10.1212/01.wnl.0000179303.72345.ce 16217056

Jain , R. A. , Wolman , M. A. , Marsden , K. C. , Nelson , J. C. , Shoenhard , H. , Echeverry , F. A. , Szi , C. , Bell , H. , Skinner , J. , Cobbs , E. N. et al. ( 2018 ). A forward genetic screen in zebrafish identifies the G-protein-coupled receptor CaSR as a modulator of sensorimotor decision making . Curr. Biol. 28 , 1357 - 1369.e5 . 10.1016/j.cub.2018.03.025 29681477 PMC5940496

Jiang , H. , Lei , R. , Ding , S.-W. and Zhu , S. ( 2014 ). Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads . BMC Bioinformatics 15 , 182 . 10.1186/1471-2105-15-182 24925680 PMC4074385

Johannessen , C. M. , Reczek , E. E. , James , M. F. , Brems , H. , Legius , E. and Cichowski , K. ( 2005 ). The NF1 tumor suppressor critically regulates TSC2 and mTOR . Proc. Natl. Acad. Sci. USA 102 , 8573 - 8578 . 10.1073/pnas.0503224102 15937108 PMC1142482

Jørgensen , A. S. , Daugvilaite , V. , De Filippo , K. , Berg , C. , Mavri , M. , Benned-Jensen , T. , Juzenaite , G. , Hjortø , G. , Rankin , S. , Våbenø , J. et al. ( 2021 ). Biased action of the CXCR4-targeting drug plerixafor is essential for its superior hematopoietic stem cell mobilization . Commun. Biol. 4 , 569 . 10.1038/s42003-021-02070-9 33980979 PMC8115334

Kalatskaya , I. , Berchiche , Y. A. , Gravel , S. , Limberg , B. J. , Rosenbaum , J. S. and Heveker , N. ( 2009 ). AMD3100 is a CXCR7 ligand with allosteric agonist properties . Mol. Pharmacol. 75 , 1240 - 1247 . 10.1124/mol.108.053389 19255243

Karaosmanoglu , B. , Kocaefe , Ç. Y. , Söylemezoğlu , F. , Anlar , B. , Varan , A. , Vargel , İ. and Ayter , S. ( 2018 ). Heightened CXCR4 and CXCL12 expression in NF1-associated neurofibromas . Childs Nerv. Syst. 34 , 877 - 882 . 10.1007/s00381-018-3745-6 29455242

Kimmel , C. B. , Patterson , J. and Kimmel , R. O. ( 1974 ). The development and behavioral characteristics of the startle response in the zebra fish . Dev. Psychobiol. 7 , 47 - 60 . 10.1002/dev.420070109 4812270

Köcher , L. and Straumann , D. ( 2025 ). Habituation learning: insights from zebrafish larvae . Front. Mol. Neurosci. 18 , 1697688 . 10.3389/fnmol.2025.1697688 41323234 PMC12660209

Krab , L. C. , De Goede-Bolder , A. , Aarsen , F. K. , Pluijm , S. M. F. , Bouman , M. J. , Van Der Geest , J. N. , Lequin , M. , Catsman , C. E. , Arts , W. F. M. , Kushner , S. A. et al. ( 2008 ). Effect of simvastatin on cognitive functioning in children with neurofibromatosis type 1: a randomized controlled trial . JAMA 300 , 287 - 294 . 10.1001/jama.300.3.287 18632543 PMC2664742

Krab , L. C. , Oostenbrink , R. , De Goede-Bolder , A. , Aarsen , F. K. , Elgersma , Y. and Moll , H. A. ( 2009 ). Health-related quality of life in children with neurofibromatosis type 1: contribution of demographic factors, disease-related factors, and behavior . J. Pediatr. 154 , 420 - 425 , 425.e1 . 10.1016/j.jpeds.2008.08.045 18950800

Lavoie , H. , Gagnon , J. and Therrien , M. ( 2020 ). ERK signalling: a master regulator of cell behaviour, life and fate . Nat. Rev. Mol. Cell Biol. 21 , 607 - 632 . 10.1038/s41580-020-0255-7 32576977

Li , B. and Dewey , C. N. ( 2011 ). RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome . BMC Bioinformatics 12 , 323 . 10.1186/1471-2105-12-323 21816040 PMC3163565

Li , W. , Cui , Y. , Kushner , S. A. , Brown , R. A. M. , Jentsch , J. D. , Frankland , P. W. , Cannon , T. D. and Silva , A. J. ( 2005 ). The HMG-CoA reductase inhibitor lovastatin reverses the learning and attention deficits in a mouse model of neurofibromatosis type 1 . Curr. Biol. 15 , 1961 - 1967 . 10.1016/j.cub.2005.09.043 16271875

Liu , K. S. and Fetcho , J. R. ( 1999 ). Laser ablations reveal functional relationships of segmental hindbrain neurons in zebrafish . Neuron 23 , 325 - 335 . 10.1016/S0896-6273(00)80783-7 10399938

Lu , L. , Li , J. , Jiang , X. and Bai , R. ( 2024 ). CXCR4/CXCL12 axis: “old” pathway as “novel” target for anti-inflammatory drug discovery . Med. Res. Rev. 44 , 1189 - 1220 . 10.1002/med.22011 38178560

Mainberger , F. , Jung , N. H. , Zenker , M. , Wahlländer , U. , Freudenberg , L. , Langer , S. , Berweck , S. , Winkler , T. , Straube , A. , Heinen , F. et al. ( 2013 ). Lovastatin improves impaired synaptic plasticity and phasic alertness in patients with neurofibromatosis type 1 . BMC Neurol. 13 , 131 . 10.1186/1471-2377-13-131 24088225 PMC4015838

Marchuk , D. A. , Saulino , A. M. , Tavakkol , R. , Swaroop , M. , Wallace , M. R. , Andersen , L. B. , Mitchell , A. L. , Gutmann , D. H. , Boguski , M. and Collins , F. S. ( 1991 ). cDNA cloning of the type 1 neurofibromatosis gene: complete sequence of the NF1 gene product . Genomics 11 , 931 - 940 . 10.1016/0888-7543(91)90017-9 1783401

Marsden , K. C. , Jain , R. A. , Wolman , M. A. , Echeverry , F. A. , Nelson , J. C. , Hayer , K. E. , Miltenberg , B. , Pereda , A. E. and Granato , M. ( 2018 ). A Cyfip2-dependent excitatory interneuron pathway establishes the innate startle threshold . Cell Rep. 23 , 878 - 887 . 10.1016/j.celrep.2018.03.095 29669291 PMC6642828

Miller , A. H. and Halloran , M. C. ( 2022 ). Mechanistic insights from animal models of neurofibromatosis type 1 cognitive impairment . Dis. Model. Mech. 15 , dmm049422 . 10.1242/dmm.049422 36037004 PMC9459395

Mo , W. , Chen , J. , Patel , A. , Zhang , L. , Chau , V. , Li , Y. , Cho , W. , Lim , K. , Xu , J. , Lazar , A. J. et al. ( 2013 ). CXCR4/CXCL12 mediate autocrine cell- cycle progression in NF1-associated malignant peripheral nerve sheath tumors . Cell 152 , 1077 - 1090 . 10.1016/j.cell.2013.01.053 23434321 PMC3594500

Naumann , U. , Cameroni , E. , Pruenster , M. , Mahabaleshwar , H. , Raz , E. , Zerwes , H.-G. , Rot , A. and Thelen , M. ( 2010 ). CXCR7 Functions as a Scavenger for CXCL12 and CXCL11 . PLoS ONE 5 , e9175 . 10.1371/journal.pone.0009175 20161793 PMC2820091

Padmanabhan , A. , Lee , J.-S. , Ismat , F. A. , Lu , M. M. , Lawson , N. D. , Kanki , J. P. , Look , A. T. and Epstein , J. A. ( 2009 ). Cardiac and vascular functions of the zebrafish orthologues of the type I neurofibromatosis gene NFI . Proc. Natl. Acad. Sci. USA 106 , 22305 - 22310 . 10.1073/pnas.0901932106 19966217 PMC2799742

Payne , J. M. , Barton , B. , Ullrich , N. J. , Cantor , A. , Hearps , S. J. C. , Cutter , G. , Rosser , T. , Walsh , K. S. , Gioia , G. A. , Wolters , P. L. et al. ( 2016 ). Randomized placebo-controlled study of lovastatin in children with neurofibromatosis type 1 . Neurology 87 , 2575 - 2584 . 10.1212/WNL.0000000000003435 27956565 PMC5207004

Preibisch , S. , Saalfeld , S. and Tomancak , P. ( 2009 ). Globally optimal stitching of tiled 3D microscopic image acquisitions . Bioinformatics 25 , 1463 - 1465 . 10.1093/bioinformatics/btp184 19346324 PMC2682522

Rajagopal , S. , Kim , J. , Ahn , S. , Craig , S. , Lam , C. M. , Gerard , N. P. , Gerard , C. and Lefkowitz , R. J. ( 2010 ). β-arrestin- but not G protein-mediated signaling by the “decoy” receptor CXCR7 . Proc. Natl. Acad. Sci. USA 107 , 628 - 632 . 10.1073/pnas.0912852107 20018651 PMC2818968

Randlett , O. , Wee , C. L. , Naumann , E. A. , Nnaemeka , O. , Schoppik , D. , Fitzgerald , J. E. , Portugues , R. , Lacoste , A. M. B. , Riegler , C. , Engert , F. et al. ( 2015 ). Whole-brain activity mapping onto a zebrafish brain atlas . Nat. Methods 12 , 1039 - 1046 . 10.1038/nmeth.3581 26778924 PMC4710481

Randlett , O. , Haesemeyer , M. , Forkin , G. , Shoenhard , H. , Schier , A. F. , Engert , F. and Granato , M. ( 2019 ). Distributed plasticity drives visual habituation learning in larval zebrafish . Curr. Biol. 29 , 1337 - 1345.e4 . 10.1016/j.cub.2019.02.039 30955936 PMC6545104

Ray , P. , Mihalko , L. A. , Coggins , N. L. , Moudgil , P. , Ehrlich , A. , Luker , K. E. and Luker , G. D. ( 2012 ). Carboxy-terminus of CXCR7 regulates receptor localization and function . Int. J. Biochem. Cell Biol. 44 , 669 - 678 . 10.1016/j.biocel.2012.01.007 22300987 PMC3294180

Reiner , A. , Yekutieli , D. and Benjamini , Y. ( 2003 ). Identifying differentially expressed genes usingfalse discovery rate controlling procedures . Bioinformatics 19 , 368 - 375 . 10.1093/bioinformatics/btf877 12584122

Rigdon , G. C. and Weatherspoon , J. K. ( 1992 ). 5-Hydroxytryptamine 1a receptor agonists block prepulse inhibition of acoustic startle reflex . J. Pharmacol. Exp. Ther. 263 , 486 - 493 . 10.1016/S0022-3565(25)10375-3 1432685

Robinson , M. D. and Oshlack , A. ( 2010 ). A scaling normalization method for differential expression analysis of RNA-seq data . Genome Biol. 11 , R25 . 10.1186/gb-2010-11-3-r25 20196867 PMC2864565

Robinson , M. D. , McCarthy , D. J. and Smyth , G. K. ( 2010 ). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data . Bioinformatics 26 , 139 - 140 . 10.1093/bioinformatics/btp616 19910308 PMC2796818

Rosen , L. B. , Ginty , D. D. , Weber , M. J. and Greenberg , M. E. ( 1994 ). Membrane depolarization and calcium influx stimulate MEK and MAP kinase via activation of Ras . Neuron 12 , 1207 - 1221 . 10.1016/0896-6273(94)90438-3 8011335

Ruuskanen , J. O. , Peitsaro , N. , Kaslin , J. V. M. , Panula , P. and Scheinin , M. ( 2005 ). Expression and function of alpha-adrenoceptors in zebrafish: drug effects, mRNA and receptor distributions . J. Neurochem. 94 , 1559 - 1569 . 10.1111/j.1471-4159.2005.03305.x 16000146

Sarma , P. , Carino , C. M. C. , Seetharama , D. , Pandey , S. , Dwivedi-Agnihotri , H. , Rui , X. , Cao , Y. , Kawakami , K. , Kumari , P. , Chen , Y.-C. et al. ( 2023 ). Molecular insights into intrinsic transducer-coupling bias in the CXCR4-CXCR7 system . Nat. Commun. 14 , 4808 . 10.1038/s41467-023-40482-9 37558722 PMC10412580

Schindelin , J. , Arganda-Carreras , I. , Frise , E. , Kaynig , V. , Longair , M. , Pietzsch , T. , Preibisch , S. , Rueden , C. , Saalfeld , S. , Schmid , B. et al. ( 2012 ). Fiji: an open-source platform for biological-image analysis . Nat. Methods 9 , 676 - 682 . 10.1038/nmeth.2019 22743772 PMC3855844

Schmitt , J. M. and Stork , P. J. S. ( 2002 ). PKA phosphorylation of Src mediates cAMP's inhibition of cell growth via Rap1 . Mol. Cell 9 , 85 - 94 . 10.1016/S1097-2765(01)00432-4 11804588

Shilyansky , C. , Karlsgodt , K. H. , Cummings , D. M. , Sidiropoulou , K. , Hardt , M. , James , A. S. , Ehninger , D. , Bearden , C. E. , Poirazi , P. , Jentsch , J. D. et al. ( 2010 ). Neurofibromin regulates corticostriatal inhibitory networks during working memory performance . Proc. Natl. Acad. Sci. USA 107 , 13141 - 13146 . 10.1073/pnas.1004829107 20624961 PMC2919968

Shin , J. , Padmanabhan , A. , de Groh , E. D. , Lee , J.-S. , Haidar , S. , Dahlberg , S. , Guo , F. , He , S. , Wolman , M. A. , Granato , M. et al. ( 2012 ). Zebrafish neurofibromatosis type 1 genes have redundant functions in tumorigenesis and embryonic development . Dis. Model. Mech. 5 , 881 - 894 . 10.1242/dmm.009779 22773753 PMC3484870

Singh , A. K. , Arya , R. K. , Trivedi , A. K. , Sanyal , S. , Baral , R. , Dormond , O. , Briscoe , D. M. and Datta , D. ( 2013 ). Chemokine receptor trio: CXCR3, CXCR4 and CXCR7 crosstalk via CXCL11 and CXCL12 . Cytokine Growth Factor. Rev. 24 , 41 - 49 . 10.1016/j.cytogfr.2012.08.007 22989616 PMC4172454

Sipes , T. A. and Geyer , M. A. ( 1994 ). Multiple serotonin receptor subtypes modulate prepulse inhibition of the startle response in rats . Neuropharmacology 33 , 441 - 448 . 10.1016/0028-3908(94)90074-4 7984282

Stefan , E. , Malleshaiah , M. K. , Breton , B. , Ear , P. H. , Bachmann , V. , Beyermann , M. , Bouvier , M. and Michnick , S. W. ( 2011 ). PKA regulatory subunits mediate synergy among conserved G-protein-coupled receptor cascades . Nat. Commun. 2 , 598 . 10.1038/ncomms1605 22186894 PMC3247815

Stivaros , S. , Garg , S. , Tziraki , M. , Cai , Y. , Thomas , O. , Mellor , J. , Morris , A. A. , Jim , C. , Szumanska-Ryt , K. , Parkes , L. M. et al. ( 2018 ). Randomised controlled trial of simvastatin treatment for autism in young children with neurofibromatosis type 1 (SANTA) . Mol. Autism 9 , 12 . 10.1186/s13229-018-0190-z 29484149 PMC5824534

Stumm , R. K. , Rummel , J. , Junker , V. , Culmsee , C. , Pfeiffer , M. , Krieglstein , J. , Höllt , V. and Schulz , S. ( 2002 ). A dual role for the SDF-1/CXCR4 chemokine receptor system in adult brain: isoform-selective regulation of SDF-1 expression modulates CXCR4-dependent neuronal plasticity and cerebral leukocyte recruitment after focal ischemia . J. Neurosci. 22 , 5865 - 5878 . 10.1523/JNEUROSCI.22-14-05865.2002 12122049 PMC6757949

Subhash , S. and Kanduri , C. ( 2016 ). GeneSCF: a real-time based functional enrichment tool with support for multiple organisms . BMC Bioinformatics 17 , 365 . 10.1186/s12859-016-1250-z 27618934 PMC5020511

Taskén , K. and Aandahl , E. M. ( 2004 ). Localized effects of cAMP mediated by distinct routes of protein kinase A . Physiol. Rev. 84 , 137 - 167 . 10.1152/physrev.00021.2003 14715913

Thomas , G. M. and Huganir , R. L. ( 2004 ). MAPK cascade signalling and synaptic plasticity . Nat. Rev. Neurosci. 5 , 173 - 183 . 10.1038/nrn1346 14976517

Tian , X. , Xie , G. , Xiao , H. , Ding , F. , Bao , W. and Zhang , M. ( 2019 ). CXCR4 knockdown prevents inflammatory cytokine expression in macrophages by suppressing activation of MAPK and NF-κB signaling pathways . Cell Biosci. 9 , 55 . 10.1186/s13578-019-0315-x 31304005 PMC6607528

Tong , J. , Hannan , F. , Zhu , Y. , Bernards , A. and Zhong , Y. ( 2002 ). Neurofibromin regulates G protein-stimulated adenylyl cyclase activity . Nat. Neurosci. 5 , 95 - 96 . 10.1038/nn792 11788835

Torres Nupan , M. M. , Velez Van Meerbeke , A. , López Cabra , C. A. and Herrera Gomez , P. M. ( 2017 ). Cognitive and behavioral disorders in children with neurofibromatosis type 1 . Front. Pediatr. 5 , 227 . 10.3389/fped.2017.00227 29164079 PMC5670111

Uusitalo , E. , Leppävirta , J. , Koffert , A. , Suominen , S. , Vahtera , J. , Vahlberg , T. , Pöyhönen , M. , Peltonen , J. and Peltonen , S. ( 2015 ). Incidence and mortality of neurofibromatosis: a total population study in Finland . J. Invest. Dermatol. 135 , 904 - 906 . 10.1038/jid.2014.465 25354145

van der Vaart , T. , Plasschaert , E. , Rietman , A. B. , Renard , M. , Oostenbrink , R. , Vogels , A. , De Wit , M.-C. Y. , Descheemaeker , M.-J. , Vergouwe , Y. , Catsman-Berrevoets , C. E. et al. ( 2013 ). Simvastatin for cognitive deficits and behavioural problems in patients with neurofibromatosis type 1 (NF1-SIMCODA): a randomised, placebo-controlled trial . Lancet Neurol. 12 , 1076 - 1083 . 10.1016/S1474-4422(13)70227-8 24090588

Vanwalleghem , G. , Heap , L. A. and Scott , E. K. ( 2017 ). A profile of auditory-responsive neurons in the larval zebrafish brain . J. Comp. Neurol. 525 , 3031 - 3043 . 10.1002/cne.24258 28599354

Varni , J. W. , Nutakki , K. and Swigonski , N. L. ( 2019 ). Pain, skin sensations symptoms, and cognitive functioning predictors of health-related quality of life in pediatric patients with Neurofibromatosis Type 1 . Qual. Life Res. 28 , 1047 - 1052 . 10.1007/s11136-018-2055-5 30465317

Walker , J. A. , Tchoudakova , A. V. , McKenney , P. T. , Brill , S. , Wu , D. , Cowley , G. S. , Hariharan , I. K. and Bernards , A. ( 2006 ). Reduced growth of Drosophila neurofibromatosis 1 mutants reflects a non-cell-autonomous requirement for GTPase-Activating Protein activity in larval neurons . Genes Dev. 20 , 3311 - 3323 . 10.1101/gad.1466806 17114577 PMC1686607

Walsh , K. S. , Wolters , P. L. , Widemann , B. C. , Del Castillo , A. , Sady , M. D. , Inker , T. , Roderick , M. C. , Martin , S. , Toledo-Tamula , M. A. , Struemph , K. et al. ( 2021 ). Impact of MEK inhibitor therapy on neurocognitive functioning in NF1 . Neurol. Genet. 7 , e616 . 10.1212/NXG.0000000000000616 34377779 PMC8351286

Warrington , N. M. , Woerner , B. M. , Daginakatte , G. C. , Dasgupta , B. , Perry , A. , Gutmann , D. H. and Rubin , J. B. ( 2007 ). Spatiotemporal differences in CXCL12 expression and cyclic AMP underlie the unique pattern of optic glioma growth in neurofibromatosis type 1 . Cancer Res. 67 , 8588 - 8595 . 10.1158/0008-5472.CAN-06-2220 17875698

Wijtmans , M. , Maussang , D. , Sirci , F. , Scholten , D. J. , Canals , M. , Mujić-Delić , A. , Chong , M. , Chatalic , K. L. S. , Custers , H. , Janssen , E. et al. ( 2012 ). Synthesis, modeling and functional activity of substituted styrene-amides as small-molecule CXCR7 agonists . Eur. J. Med. Chem. 51 , 184 - 192 . 10.1016/j.ejmech.2012.02.041 22424612

Wolman , M. A. , Jain , R. A. , Liss , L. and Granato , M. ( 2011 ). Chemical modulation of memory formation in larval zebrafish . Proc. Natl. Acad. Sci. USA 108 , 15468 - 15473 . 10.1073/pnas.1107156108 21876167 PMC3174630

Wolman , M. A. , de Groh , E. D. , McBride , S. M. , Jongens , T. A. , Granato , M. and Epstein , J. A. ( 2014 ). Modulation of cAMP and ras signaling pathways improves distinct behavioral deficits in a zebrafish model of neurofibromatosis type 1 . Cell Rep. 8 , 1265 - 1270 . 10.1016/j.celrep.2014.07.054 25176649 PMC5850931

Wolman , M. A. , Jain , R. A. , Marsden , K. C. , Bell , H. , Skinner , J. , Hayer , K. E. , Hogenesch , J. B. and Granato , M. ( 2015 ). A genome-wide screen identifies PAPP-AA-mediated IGFR signaling as a novel regulator of habituation learning . Neuron 85 , 1200 - 1211 . 10.1016/j.neuron.2015.02.025 25754827 PMC4368495