Rationale for haploinsufficiency correction therapy in neurofibromatosis type 1
,
, ...
Abstract
Neurofibromatosis type 1 (NF1) is a genetic disorder with a wide range of manifestations and severity. Currently, the few available NF1 treatments target specific manifestations, with no available therapies targeted to correct the underlying driver of all NF1 manifestations. Evidence supports that haploinsufficiency in NF1 caused by a decreased amount of wild-type (WT) neurofibromin in all NF1+/- cells directly causes or facilitates a range of NF1 manifestations. Consequently, NF1 haploinsufficiency correction therapy (NF1-HCT) represents a potentially effective approach to treat some NF1 manifestations. NF1-HCT would normalize the level of WT neurofibromin in all NF1-haploinsufficient cells, including those integral to the NF1 phenotype such as Schwann cells (SCs), melanocytes, neurons, bone cells, and cells of the tumor microenvironment. This would correct altered cellular signaling pathways and, in turn, restore normal function to cells with a retained WT allele. NF1-HCT will not restore WT neurofibromin in NF1-/- cells; however, by restoring function in the surrounding NF1+/- microenvironment cells, NF1-HCT is predicted to have a beneficial effect on NF1-/- cells. NF1-HCT is expected to have a clinical effect in some NF1 manifestations, as follows: (i) prevention, or delay of onset, of potential manifestations; and (ii) reversal, or halting/slowing progression, of established manifestations. This review describes the rationale for NF1-HCT, including specific NF1 considerations (e.g., NF1 clinical phenotype, neurofibromin function/regulation, NF1 mutational spectrum, genotype-phenotype correlation, and the impact of haploinsufficiency in NF1), HCT in other haploinsufficient diseases, potential NF1-HCT drug treatment strategies, and the potential advantages/challenges of NF1-HCT.
Keywords
INTRODUCTION
NF1 is a rare autosomal dominant genetic disorder (birth incidence of ~1:2500-3000) whose underlying genetic abnormality is the loss of normal function in one of the two NF1 alleles[1-3]. In ~50% of cases in North America and Europe, NF1 is inherited, while the remaining half represent a spontaneous new mutation[2]. NF1 impacts about 120,000 Americans (2.5 M worldwide) and is therefore classified by the U.S. Food and Drug Administration (FDA) as an orphan disease.
Typical management comprises careful symptom monitoring, ideally in a specialized center. In some situations, targeted treatment of certain manifestations may be required, including surgical removal of tumors, spine instrumentation for scoliosis, carboplatin/vincristine for optic pathway gliomas (OPGs), MEK inhibitors (MEKi) for a subset of pediatric plexiform neurofibromas (pNFs), and/or stimulant prescriptions for attention deficit hyperactivity disorder (ADHD); however, none of these symptomatic therapies treat the underlying driver of NF1. Given the broad range of NF1 manifestations with significant morbidity, including cognitive and social dysfunction (CD/SD), cutaneous neurofibromas (cNFs) and pNFs, and bone dysplasia, most NF1 individuals experience a decreased quality of life. NF1 individuals also have an increased risk of developing chronic pain, depression and life-threatening complications such as cardiovascular disease and malignancies. For example, the lifetime risk of breast cancer for NF1 individuals is three times that of the general population, and amongst NF1 females under 40 years of age, the risk is eleven times that of age-matched non-NF1 women[4]. Consequently, there is great urgency to develop systemic NF1 treatments that can prevent or significantly delay progression across a range of NF1-related manifestations.
For NF1 individuals (excluding NF1-mosaicism), every cell initially has one normal and one abnormal NF1 allele (NF1+/- cells). The former encodes WT neurofibromin, while the latter does not. This results in insufficient WT neurofibromin to maintain normal cell function, termed haploinsufficiency. The abnormal cell function resulting from haploinsufficiency is critical in NF1 pathogenesis, with evidence supporting that it causes certain NF1 manifestations and permits or accelerates others[5-17]. Some NF1-associated manifestations arise in cells that lack a normal NF1 allele (NF1-/- cells), and NF1-HCT will not restore WT neurofibromin levels in these cells. However, because of the close interplay between NF1-/- and NF1+/- cells, restoration of normal function in NF1+/- cells by NF1-HCT is predicted to have a beneficial effect on NF1-/- cells and, in turn, prevent or delay the onset of manifestations that arise in NF1-/- cells. Consequently, NF1-HCT represents a potential NF1 treatment. By normalizing the level of WT neurofibromin in all NF1-haploinsufficient cells, NF1-HCT is expected to normalize cell function and consequently reverse, prevent or delay the development of NF1 manifestations [Figure 1]. This review describes in detail the rationale and specific considerations for NF1-HCT, with an emphasis on the potential benefits of a small molecule NF1-HCT approach, and how this compares to (i) currently available NF1 treatments and (ii) NF1 gene-based therapeutic strategies (e.g., NF1 gene replacement, and NF1 gene editing), which represent another potential, albeit more challenging, approach to correct levels of WT neurofibromin.
Figure 1. Small molecule therapy to correct NF1-haploinsufficiency. In the normal cell (A), two normal NF1 alleles provide a normal level of WT neurofibromin. By contrast, in the NF1+/- haploinsufficient cell (B), the abnormal NF1 allele does not contribute any WT neurofibromin, resulting in NF1 protein haploinsufficiency, and a disease state. In the treated NF1+/- cell (C), a drug increases transcription by the normal NF1 allele (red arrow), resulting in increased WT neurofibromin (purple arrows) and correction of haploinsufficiency. Increasing transcription is just one of several methods to increase neurofibromin expression. NF1: Neurofibromatosis type 1; WT: wild-type.
CONSIDERATIONS OF THE NF1-HCT APPROACH
NF1 clinical phenotype
Successful treatment of NF1 requires understanding its clinical manifestations. In particular, any potential NF1 therapeutic strategy must consider: (i) NF1 has a broad range of manifestations; and (ii) manifestations can vary markedly across NF1 individuals.
While the clinical hallmarks of NF1, namely café au lait macules (CALMs), Lisch nodules (iris hamartoma), and cNFs, occur in almost all NF1 individuals, additional manifestations occur in a significant percentage, including pNFs (35%-50%), malignant peripheral nerve sheath tumors (MPNSTs: 8%-13%), CD/SD (up to 80% of children), bone abnormalities such as scoliosis (true prevalence is unknown - most studies quote figures in the range of 10%-36%), OPGs (15%-20% of pediatric patients), cardiovascular abnormalities, hypertension, and various other malignancies that contribute to a lifetime risk of non-NF1 cancers that is ~2 times the general population[4,18-23]. Disease severity and associated morbidity vary dramatically between NF1 individuals, wherein some may have a few cNFs and CALMs, while others may have thousands of cNFs, large pNFs impinging on normal structures, prominent CD/SD, chronic and severe pain/itchiness, and marked skeletal abnormalities and disfigurement[24-27]. Among family members harboring the same NF1 mutation, concordance for certain manifestations (e.g., pigmentary changes, numbers of cNFs, and CDs) is high amongst monozygotic twins but decreases with increasing familial separation, indicating that for these manifestations, the germline mutation plays a primary role but that other factors, such as genes unlinked to the NF1 locus, are important modifiers of the phenotype[28]. Other manifestations show discordance even amongst monozygotic twins (e.g., pNFs and malignancies), supporting the influence of non-heritable factors such as stochastic somatic mutation/deletion of the WT NF1 allele or environmental factors[29].
Mosaicism in NF1
It is notable that there are no differences in clinical presentation between familial and sporadic cases of NF1. In contrast, those individuals who present with localized manifestations, mosaic NF1, display an attenuated natural history. Mosaic NF1 has been calculated to represent approximately 5% of NF1 cases; however, the true prevalence is unknown[30-32]. Mosaic NF1 arises when the initial NF1 mutation occurs in the postzygotic state, resulting in a mixture of NF1+/- (low WT neurofibromin level), and NF1+/+ (normal WT neurofibromin level) cells. In NF1 mosaics, NF1-HCT would require careful dosing and patient monitoring to correct deficits in NF1+/- cells without adversely affecting NF1+/+ cells (i.e., if excess WT neurofibromin were to be detrimental to the cell). If the therapeutic window is too narrow, this milder NF1 category may not be treatable with NF1-HCT.
Normal expression and function of the NF1 gene product neurofibromin
Crucial to the development of a system-wide NF1 therapy is an understanding of neurofibromin’s functions within different cell types and how dysfunction results in various manifestations. The NF1 gene, located at the 17q11.2 locus, is one of the largest human genes (~280 kb)[30,33,34]. NF1 contains three embedded genes
NF1 encodes neurofibromin, which is a GTPase activating protein (GAP) that negatively regulates Ras. Neurofibromin contains the following domains (starting at the N-terminus): cysteine-serine-rich domain, tubulin-binding domain, GAP-related domain (GRD), Sec14-like domain, pleckstrin homology domain, and the C-terminal domain. The three-dimensional structure and the RAS-GAP activity of the GRD have been well-characterized[42-45], whereas the exact role of the other domains is not as well understood. Neurofibromin exists as six main isoforms as a result of alternative splicing. The two most abundant isoforms are (i) isoform 1 (nomenclature of UniProt), which contains 2818 amino acids; has a predicted molecular weight of 327 kDa; and contains no alternatively included exons, and (ii) isoform 2, which contains a 21-amino-acid insertion that is encoded by exon 23a and located within the GRD[46,47]. Notably, the RAS-GAP activity of isoform 1 is 10-fold higher than that of isoform 2[47,48].
Neurofibromin is widely expressed, with varying degrees of expression, and varying isoform ratios, in different tissues and developmental stages[41,49]. From embryonic day 11 onwards, most murine tissues demonstrate high levels of neurofibromin, whereas postnatally neurofibromin levels drop significantly in most terminally differentiated tissues, apart from certain cell types such as Schwann cells, neurons, and adrenal medulla cells[49,50]. In adult rats, isoform 1 is the predominant isoform in CNS neurons, whereas isoform 2 is the predominant form in Schwann cells, adrenal medullary cells and ovary[41]. Neurofibromin is primarily localized in the cytoplasm; however, binding of SPRED1 to the GRD facilitates neurofibromin translocation to the plasma membrane, which is critical as it allows neurofibromin to interact with membrane-bound Ras[51,52].
Neurofibromin exists as an obligate high-affinity, pseudo-symmetric dimer, that is ~620 -kDa, and ~32 nm along the long axis[53]. The dimer exists in closed and open states, with the closed dimer being the predominant form. Dimers in the closed state have both protomers in a self-inhibited, Zn-stabilized state that prevents Ras binding by the GRD. Dimers in the open state have one protomer in a closed conformation and the other in an open conformation that allows Ras binding by the GRD[54].
Neurofibromin plays critical cellular regulatory roles, particularly in neural crest-derived cells, as evidenced by the fact that (i) although Nf1+/- mice do not develop typical NF1 symptoms such as neurofibromas, they have accelerated rates of tumorigenesis (composed of tumors typically seen in older WT mice), and shortened lifespans, compared to Nf1+/+ mice, while mice harboring two abnormal Nf1 alleles die in utero by embryonic day 14 due to cardiac abnormalities[55,56], (ii) monoallelic NF1 abnormalities result in a diverse NF1 phenotype that includes manifestations arising in neural crest-derived cells, and (iii) NF1 abnormalities are present in various malignancies in non-NF1 individuals, including tumors that arise in neural crest-derived cells (e.g., melanoma) and tumors that arise in non-neural crest-derived cells (e.g., acute myeloid leukemia and various carcinomas)[18].
RAS-dependent functions of neurofibromin
The best characterized neurofibromin function is its critical role as a RAS-GTPase activating protein (RAS-GAP), wherein it negatively regulates RAS by increasing the intrinsic hydrolysis of RAS-bound GTP by a factor of 105, resulting in rapid conversion of active RAS-GTP to inactive RAS-GDP[43-45]. RAS is a proto-oncogene and a key regulator of at least eleven intracellular signaling pathways, including those involved in cell differentiation and homeostasis[51,57]. Tight control of RAS activity is essential for normal cellular homeostasis. This is underscored by the fact that oncogenic RAS is present in ~25% of human cancers, where it drives tumor initiation and maintenance[58]. Furthermore, altered RAS function underlies the broad group of developmental disorders known as “RASopathies”[59,60], which have a wide range of phenotypes and include NF1. A variety of NF1 manifestations result from increased RAS signaling secondary to the loss of WT neurofibromin. These include NF1-specific tumors such as pNFs, cNFs, OPGs, and MPNSTs. In addition, RAS plays a neurofibromin-dependent role in cognitive function, as demonstrated in Nf1+/- mice displaying spatial learning and attention deficits modeling those seen in NF1 individuals[13,16]. When RAS hyperfunction is normalized in these mice, these deficits are corrected, and deficits in long-term potentiation and GABA-mediated inhibition are reversed[15,16].
Amongst non-NF1 individuals, various sporadic cancers contain NF1 abnormalities. These include cases confirmed to have increased RAS-GTP without RAS mutations, supporting the notion that decreased neurofibromin promotes sporadic tumor development through an upregulated RAS pathway[18,61]. For example, nearly all mammary carcinomas in a breast carcinoma-prone conditional mouse model (CMM) contained Nf1 gene deletions and increased RAS-GTP, and amongst human breast tumors in The Cancer Genome Atlas, 23% have hemizygous NF1 deletions and 4% have NF1 mutations[62]. Sporadic glioblastoma multiforme (GBM) has NF1 inactivating mutations/deletions in ~23% of cases, while ~14% of cutaneous melanomas have NF1 mutations[18]. Restoring neurofibromin in neurofibromin-deficient sporadic tumors may represent an effective therapeutic approach, particularly given the efficacy of blocking neurofibromin effectors in these tumors in vitro and in vivo. For example, MEKi therapy shows efficacy in neurofibromin-deficient sporadic GBM cell lines[63] and Nf1-deficient acute myeloid leukemias in cell culture and mouse xenografts[64]. MEKi treatment also restores sensitivity to EGFR inhibitors in neurofibromin-deficient lung cancer cells and mouse xenografts[65]. There are limited reports of MEKi therapy targeting NF1-deficient sporadic tumors in clinical practice. In one case report, MEKi (trametinib) administration resulted in a partial response of advanced NF1-deficient melanoma[66]. In a separate study assessing the efficacy of genomics-guided treatment in GBM, trametinib was commenced in a patient with progressed NF1-deficient GBM, resulting in a time to progression of > 665 days[67]. Two prospective trials are currently investigating MEK-inhibition in sporadic NF1 mutant tumors (NCT02645149, NCT02465060)[66].
Functions of neurofibromin beyond RAS
Neurofibromin also interacts with and regulates other cellular proteins and pathways in a RAS-independent manner, with dysregulation of these likely underlying some NF1 manifestations. Learning deficits in Nf1+/- mice result from enhanced neuronal inhibition, which, at least in part, appears to be caused by RAS-independent mechanisms. For example, neurofibromin interacts with hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1), and decreased neurofibromin results in attenuation of the HCN1-mediated incoming channel current (Ih) in GABAergic inhibitory interneurons. The resultant hyperexcitability in these interneurons strongly contributes to this neuronal inhibition in neurofibromin-deficient mice. Stimulating the HCN current with lamotrigine corrects the learning deficits in these mice. The HCN1-neurofibromin interaction may be RAS-independent, given that (i) RAS knock-in mice do not have altered Ih; (ii) Ih is not affected by MEKi treatment; and (iii) HCN channels lack ERK consensus sites[68].
Neurofibromin positively regulates cAMP in mice[69] and Drosophila[70,71]. In the latter, it occurs through two pathways: (1) a Ras-dependent pathway that is associated with long-term memory formation and (2) a Ras-independent pathway that is essential for associative learning and short-term memory formation[69,72,73]. Neurofibromin regulation of cAMP is also essential for somatic growth, which has been reported as Ras-dependent in some studies[74] and Ras-independent in others[75,76]. For example, mice with conditional Nf1 knockout in the central nervous system (CNS) (BLBP-Cre; Nf1flox/flox) are smaller than controls due to reduced growth hormone secretion secondary to disruption of hypothalamic Ras-independent neurofibromin regulation of cAMP[76]. Importantly, these findings provide insight into the mechanisms underlying the reduced stature present in NF1 individuals (20%-30% of NF1 adults have a height below the 3rd centile)[77,78].
In the OPG CMM (Nf1-/- CNS glial cells; all other cells Nf1+/-), mice have decreased striatal dopamine, decreased dopaminergic neuron integrity, and deficits in spatial memory, exploratory behavior, and attention. OPG mice are rescued from CDs using treatments that elevate dopamine but not those inhibiting RAS or increasing cAMP. This suggests a RAS-independent role of neurofibromin in regulating CNS dopamine homeostasis, and alteration of this neurofibromin-dopamine axis may contribute to NF1 CDs[79,80]. Neurofibromin has been proposed to regulate neuronal differentiation by (i) direct complex formation with CRMP-2 (Collapsin response mediator protein-2), thus directly blocking access to kinases (Rho, Cdk5 and GSK-3β) that phosphorylate CRMP-2, and (ii) suppressing RAS activation of the same kinases[81]. CRMP-2 is also a key player in NF1-associated pain in rat models through its regulation of CaV2.2 and NaV1.7 channels[82]. In a GBM invasiveness study, the neurofibromin leucine-rich domain inhibited GBM invasion but failed to hydrolyze RAS-GTP, suggesting that its anti-invasive function is RAS-independent[83]. Neurofibromin can promote sensitivity to apoptosis mediated by RAS-dependent, RAS-independent and cAMP-independent pathways[84]. In conclusion, neurofibromin appears to play a RAS-independent regulatory role in a variety of processes, including cognition, axonal growth, pain perception, dopamine homeostasis, tumor invasiveness, apoptosis, and body growth in some models.
NF1 manifestations where neurofibromin’s role is unclear
It is not clear how neurofibromin contributes to certain NF1 manifestations. NF1 individuals tend to have larger heads and 24% have macrocephaly (head circumference > 2 SD above the mean). This is suspected to result from increased brain size, although the exact mechanism is unknown[77]. NF1 individuals often experience motor deficits of unknown cause, including difficulties with coordination, decreased muscle tone, strength, and easy fatigability. Originally these were felt to be due to neurocognitive deficits, yet more recent data suggest a possible primary myopathic process[85,86]. NF1 individuals are prone to headaches of unknown causes. Amongst 50 NF1 children-adolescents, 62% experienced headaches (14% in controls), and 54% experienced migraines (12% in controls)[87]. NF1 individuals have an increased incidence of sleep disturbance of uncertain cause[88,89]. In some studies, NF1 individuals have decreased Vitamin D levels of uncertain cause[90-92]; however, Vitamin D levels are not decreased in other studies[93,94].
Implications for NF1-HCT
Given that neurofibromin regulates many unrelated effectors that are associated with a variety of NF1 manifestations, any therapy targeting just one effector is expected to correct only a small subset of manifestations. For example, drugs targeting specific RAS effectors (e.g., MEK) are unlikely to correct RAS-independent CDs mediated by dopamine, cAMP and HCN1, or manifestations arising from other (non-MAPK) RAS pathways. Similarly, by only targeting RAS-independent effectors, ongoing RAS hyperactivity will continue to promote tumor development. However, an NF1-HCT that normalizes neurofibromin levels in all NF1+/- cells is expected to correct all neurofibromin-regulated pathways and benefit the wide range of NF1 manifestations - including both RAS-dependent and RAS-independent mechanisms. Separately, NF1-HCT may also benefit non-NF1 cancer patients with neurofibromin-deficient tumors.
Regulation of neurofibromin
Development of an NF1-HCT requires understanding the mechanisms regulating neurofibromin levels, as each may be targeted to increase WT neurofibromin. Regulation of NF1 expression occurs at multiple levels, including transcriptional control, RNA processing, mRNA transport, miRNA regulation, protein targeting and protein degradation[95,96]. Functional sites in the NF1 promotor bind transcription regulators such as CRE, SP1, and RUNX1, and mutations in these sites significantly decrease NF1 transcription and neurofibromin levels[97,98]. A CRISPRa NF1 transcriptional regulator that increases NF1 RNA expression in immortalized NF1+/- SCs has also been developed (Infixion Bioscience, unpublished data). Also, miRNAs repress NF1 mRNA in a variety of cell types in vivo, including SCs (miR-27)[99] and neurons (miR-128, miR-103, and miR-107)[96]. Finally, proteasomal neurofibromin degradation is a sensitive regulatory process involving protein kinase C and Cullin 3 E3 ligase. Degradation occurs within 5 minutes following growth factor stimulation, resulting in RAS activation. Subsequently, neurofibromin levels normalize within 30 minutes after growth factor removal[100-102].
NF1 mutational spectrum
NF1 individuals have a wide range of NF1 gene abnormalities that must be considered in any treatment approach. In NF1, the underlying abnormality is within the NF1 gene on one of the two alleles. In ~95% of cases, this is an intragenic mutation, while in ~5% of cases, there is a complete deletion of NF1 (microdeletion) and adjacent genes[103]. The result in both cases is a decrease in WT neurofibromin.
NF1 intragenic mutations
The number of documented unique NF1 pathogenic variants is extensive. As of 2018, the University of Alabama analyzed 8400 unrelated NF1 individuals using a comprehensive NF1 mutation analysis, of which more than 2800 different germline pathogenic variants were identified, with only 31 present in ≥ 0.5% of unrelated individuals[104]. Most are small genetic changes including single-base substitutions, insertions or deletions[105], that result in splicing (27%), frameshift (26%), nonsense (21%) and missense (16%) pathogenic variants[103]. Approximately 80% of pathogenic intragenic variants predict truncated neurofibromin from generated frameshifts and premature termination codons (PTCs); however, negligible abnormal protein is expected as PTC-containing mRNA typically undergoes nonsense-mediated decay (NMD)[106] [Figure 2]. The remaining ~20% of intragenic mutations result in full-length neurofibromin with decreased function. In either genetic setting, the result is NF1-haploinsufficiency.
Figure 2. NF1 abnormalities predict absence or presence of mutant neurofibromin. In NF1-HCT, increased transcription of the abnormal NF1 allele will result in either (i) no/negligible mutant neurofibromin: NF1 microdeletion (~5% of cases), and intragenic mutations that result in nonsense-mediated decay (~75% of cases), or (ii) an increase in mutant neurofibromin: intragenic mutations producing full-length neurofibromin (~20% of cases). Note: a small number of NF1 nonsense and frameshift variants may produce neurofibromin due to escape from nonsense-mediated decay. NF1: Neurofibromatosis type 1; NF1-HCT: NF1 haploinsufficiency correction therapy.
While decreased WT neurofibromin underlies the NF1 phenotype (i.e. the phenotype results from the lack of WT neurofibromin), it has been suggested that non-truncating mutations may, in rare cases, result in a dominant-negative effect (i.e. certain manifestations may be directly due to the mutant neurofibromin protein). Supporting the latter are reports of NF1 patient subsets that have (i) specific phenotypes, and (ii) rates of non-truncating NF1 mutations that are higher than the baseline rate of ~20% in the NF1 population. This suggests that the full-length and presumably more stable mutant protein may result in a dominant-negative effect that potentiates the specific phenotype. For example, one group reported that (i) in their literature review, almost all NF1 individuals with Neurofibromatosis-Noonan Syndrome (NFNS) and pulmonary stenosis (PS) had non-truncating mutations (8/9); and (ii) the majority of an additional cohort of NF1 individuals with PS had non-truncating mutations (8/11). Consequently, the authors suggested that amongst the ~1% of NF1 individuals with PS, some may have a non-truncated mutant neurofibromin with a dominant-negative cardiac effect[107]. NF1 individuals with spinal neurofibromatosis represent another NF1 patient subset with a specific phenotype and an increased incidence of non-truncating mutations (43%). Affected individuals have bilateral neurofibromas involving all spinal roots, but a lower incidence of typical NF1 features: CALMs - 67%; freckling - 18%; cNFs - 31%[108]. The potential benefit of NF1-HCT in these populations will require care and study during any potential future human clinical trials.
NF1 microdeletions
In NF1 microdeletion cases, there is no neurofibromin produced from the affected allele. Microdeletion cases are separated into four types (type-1, type-2, type-3 and atypical) based on the amount of genetic material deleted along with the NF1 gene. The most frequent are type-1 (~75% of cases), in which 1.4 Mb of DNA is missing, including NF1 (~0.35 Mb) and 17 additional genes producing 13 proteins (CRLF3, ATAD5, TEFM, ADAP2, RNF135, OMG, EVI2B, EVI2A, RAB11FIP4, COPRS, UTP6, SUZ12, LRRC37B) and four miRNAs (MIR4733, MIR193A, MIR365B, MIR4725). Type-1 microdeletions often result in more severe manifestations than intragenic NF1 mutations[35]. In the remaining 25%, smaller amounts of DNA, and fewer co-deleted genes, are impacted in type-2 and type-3 cases, while atypical-type cases are heterogeneous in terms of size and the number of genes lost, with variable impact on phenotype across these three types. While there is evidence suggesting that co-deleted genes play a key role in disease severity, the exact impact of many co-deleted genes on NF1 manifestations is not well understood[35]. One key co-deleted gene is SUZ12 which encodes a subunit of the polycomb repressive complex 2. The latter catalyzes histone H3 lysine 27 methylation to mediate the epigenetic silencing of target genes[109].
Implications for NF1-HCT
In all categories of NF1 germline abnormality (i.e., truncating and non-truncating mutations, and microdeletions), there is decreased WT neurofibromin. Consequently, the NF1-HCT approach of increasing WT neurofibromin from the WT allele is expected to be effective in all cases. However, it is important to consider the effects of upregulating the abnormal allele [Table 1 and Figure 2]. Amongst the ~75% of cases with intragenic mutations predicting a truncated protein, upregulation of the abnormal allele is not expected to have untoward consequences as the mRNA will undergo NMD. Conversely, amongst the ~20% of cases with intragenic non-truncating mutations, NF1-HCT is expected to increase levels of both WT and mutant neurofibromin. If rare mutant proteins have a dominant-negative effect[107], then increasing their levels could have adverse effects. As such, genetic and clinical screening may be required to stratify candidates into optimal NF1-HCT approaches, taking into account potential dominant-negative effects. One potential dominant-negative mechanism represents dimerization of mutant and WT neurofibromin[53,54], with resultant degradation of WT neurofibromin at the faster rate of the mutant[110]. In this scenario, an NF1-HCT approach that decreases neurofibromin degradation may be preferable to one that increases NF1 transcription. Interestingly, some degree of benefit may result from upregulating the mutant NF1 allele. In particular, when 29 Nf1 variant cDNAs were transfected into NF1-/- HEK293 cells, many corrected RAS activity to at least some degree[111]. Upregulating NF1 expression in the ~5% of cases caused by microdeletion will not result in any abnormal neurofibromin as the abnormal allele lacks an NF1 gene. As such, NF1-HCT should correct NF1 manifestations without adverse effects due to a mutant protein. However, as NF1-HCT is not expected to upregulate genes that are co-deleted with NF1, those manifestations attributable to co-deleted genes, such as SUZ12, are not expected to improve.
Predicted NF1-HCT effect amongst different NF1 germline abnormalities and mosaic cases
| NF1 germline pathogenic variant | Cases (%) | Predicted protein | Level of abnormal neurofibromin | Level of normal neurofibromin | Comment | ||
| Before treatment | After treatment | Before treatment | After treatment | ||||
| Frameshift | 26 | Truncated* | Absent* | Absent* | Insufficient | Adequate | Upregulating alleles that predict truncated proteins are not expected to have a pathogenic effect |
| Nonsense | 21 | Truncated | Absent | Absent | Insufficient | Adequate | |
| Splicing | 27 | Truncated | Absent* | Absent* | Insufficient | Adequate | |
| Shorter protein | Present | Increased | Insufficient | Adequate | Increased levels of abnormal shorter or full-length neurofibromin could have untoward effects, but only if the variant causes a pathogenic gain of function. May also have a beneficial effect | ||
| Missense | 16 | Full-length | Present | Increased | Insufficient | Adequate | |
| Gene deletion | 5 | No protein | Absent | Absent | Insufficient | Adequate | Symptoms solely due to co-deleted genes are not expected to be corrected |
| Mosaicism | See note | Depends on gene abnormality | Insufficient in affected cells. Adequate in normal cells | Adequate in affected cells. Increased in normal cells | The effect of increasing WT neurofibromin in normal cells is not certain | ||
NF1 genotype-phenotype correlation
Compounding the challenges associated with developing an NF1 therapy that addresses the wide range of gene abnormalities and manifestations is the absence of a genotype-phenotype correlation in ~90% of cases. Amongst the ~10% of cases where an association has been documented, the majority have been in the ~5% of cases caused by NF1 microdeletions. In these cases, the phenotype is typically more severe and includes both new and exaggerated NF1 manifestations such as dysmorphic facial features, hypertelorism, intellectual disability, cardiovascular abnormalities, childhood overgrowth, a greater tumor burden and an increased risk of MPNST[35,112]. It is unknown precisely how much of the microdeletion phenotype is a direct consequence of NF1-haploinsufficiency versus that of co-deleted genes. Of the ~95% of NF1 cases caused by intragenic mutations, there are six reported genotype-phenotype correlations, which together represent another 4.8% (383 of 8000 unrelated probands analyzed) of NF1 individuals [Table 2][104,113-117].
Genotype-phenotype associations in NF1 patients bearing small pathogenic variants
| # | Genotype | Mutation type | % of NF1 cases | Total in 8000 unrelated probands | Associated Phenotype (key features listed) | Publication |
| 1 | p.Met992del | Single aa loss | 0.9 | 74 | Mild. CALMs, SFF. No evNFs | [115,116] |
| 2 | p.Arg1809 | Missense | 1.2 | 99 | Mild. CALMs, Lisch nodules. No evNFs. 25% NLF. 50% CDs/learning disabilities | [117] |
| 3 | codons 844-848 | Missense | 0.8 | 67 | More severe. Major superficial pNFs, OPGs, sNFs, skeletal abnormalities, malignancies | [104] |
| 4 | p.Met1149 | Missense | 0.4 | 34 | Mild. CALMs, SFF, CDs, NLF, low prevalence PS | [114] |
| 5 | p.Arg1276 | Missense | 0.7 | 57 | More severe. sNFs, NLF, hpCCA | [114] |
| 6 | p.Lys1423 | Missense | 0.7 | 52 | More severe. NLF, hpCCA. No evpNFs | [114] |
The inability to predict phenotype in most (~90%) NF1 individuals makes genetic counseling and tailored clinical management difficult. For example, it is virtually impossible to determine which pre-symptomatic NF1 individuals might benefit from certain targeted therapies. This is further complicated when therapies are largely beneficial when administered pre-symptomatically (e.g., potential therapies to prevent scoliosis) and have possible adverse side effects, thus raising a risk-to-benefit dilemma. By contrast, NF1-HCT is expected to have a preventative or symptom-delaying effect in all NF1 individuals. Consequently, NF1-HCT with a demonstrated safety profile is expected to be indicated in the pre-symptomatic stages of most NF1 individuals, with potential exclusion of certain cases (e.g., mosaicism, and cases with potentially dominant-negative mutant neurofibromin).
Haploinsufficiency and complete loss of function in NF1
Amongst the NF1 manifestations with known molecular pathogenesis, there are two primary mechanisms by which an NF1 gene abnormality results in the clinical phenotype: (1) complete loss of neurofibromin function (cLOF) and (2) haploinsufficiency. In order to develop a global NF1 therapy, a fundamental understanding of both, including the interaction between them, is required.
Impact of complete loss of function in NF1
cLOF, occurring when both NF1 alleles are abnormal, is a key initiator of certain NF1 manifestations. NF1 is a tumor suppressor gene (TSG); however, its effects extend well beyond tumor suppression. As a TSG, it follows the Knudson two-hit hypothesis for tumor development[118]. The “first hit” represents the germline abnormality present in one NF1 gene allele from conception. While this first hit is insufficient to cause certain NF1 manifestations (e.g., tumors), it results in haploinsufficiency which can directly cause other NF1 manifestations (e.g., cognitive deficits). The “second hit” refers to a pathogenic change in the remaining WT allele, thus resulting in cLOF. When certain cells acquire a second-hit, manifestations such as cNFs and pNFs (cLOF in SCs), CALMs (cLOF in melanocytes), pseudarthroses (cLOF in mesenchymal precursor cells), and gastrointestinal stromal tumors can develop[32]. Importantly, the development of certain cLOF-associated NF1 manifestations in mouse models either requires or is facilitated by a microenvironment of haploinsufficient cells[7,8,17].
Evidence supporting stand-alone effect of haploinsufficiency in NF1
Despite the multitude of different germline NF1 pathogenic variants, the ultimate result is the same: all cells in NF1 individuals, excluding mosaics, begin with one abnormal NF1 allele[119]. The other allele produces WT neurofibromin, but this is insufficient to maintain normal cellular function [Figure 1]. This is termed haploinsufficiency, and evidence supports that this alone (i.e., without a ‘second hit’) can result in certain NF1 manifestations, such as CD/SD, osteopenia, osteoporosis, short stature, macrocephaly, and vasculopathy[5,10,11,32,78,120-122].
CD/SD occurs in most NF1 individuals (65%-80%) and represents a source of morbidity starting in childhood. The high prevalence and generalized nature support that NF1-haploinsufficiency alone is the driver. This is further supported by Nf1+/- haploinsufficient mice demonstrating a range of correctable cognitive and social deficits mirroring those seen in NF1 individuals [Table 3]. The correctability of these deficits in adult mice underscores their labile nature and further supports NF1-HCT as a strategy to prevent and reverse NF1-related CD/SD.
Nf1 mouse models demonstrating manifestations caused directly by Nf1-haploinsufficiency
| System modeled | Mouse genotype | Nf1+/-cells | Induction | Manifestation | Treatment | Publication |
| Bone mineral density | Nf1+/- | All | Ovariectomy | Excess bone mass loss | Not performed | [5,6] |
| LysMCre;Nf1flox/+ | MPCs only | |||||
| Bone - impaired fracture healing | Nf1+/- | All | Distal tibial open fracture | Significantly higher proportion with non-union | Combined rhBMP and bisphosphonates decreased non-union rate | [125,126] |
| Blood vessels - stenosis | Nf1+/- | All | Endothelial injury | Excess neointimal proliferation and stenosis | Gleevec & rosuvastatin were preventative | [10,11] |
| LysMCre;Nf1flox/+ | MPCs only | |||||
| Nf1+/- with WT BM | All cells except BM | No excess neointimal proliferation | WT BM was preventative | [11] | ||
| Blood vessels -aneurysm | Nf1+/- | All | Angiotensin II infusion | Excess aortic aneurysm formation | Simvastatin & apocynin were preventative | [12] |
| LysMCre;Nf1flox/+ | MPCs only | |||||
| Spatial learning | Nf1+/- | All | None | Spatial learning deficits | Reversed with additional training, Picrotoxin, farnesyl transferase inhibitor & Lovastatin | [13-16] |
| Attention deficit | Nf1+/- | All | None | Attention deficits and ADHD-type behavior | Reversed with Lovastatin | [16] |
| Social learning | Nf1+/- | All | None | Selective social behavioral deficits | Reversed with Pak1 inhibitor | [133] |
Decreased bone mineral density, which underlies osteopenia and osteoporosis, is detected in up to 50% of NF1 individuals at an early age[78]. An increased incidence of fractures has also been reported[78]. The frequent occurrence and generalized (i.e., non-focal) involvement of the skeleton by NF1-related osteopenia and osteoporosis supports the underlying driver being NF1-haploinsufficiency[121]. Moreover, Nf1+/- mice have increased numbers of osteoclasts (bone resorptive cells)[5], and Nf1+/- osteoprogenitors form fewer osteoblasts (bone-forming cells)[123].
NF1-related vasculopathy, affecting up to 6.4% of NF1 individuals[122], is associated with excess mortality, especially in younger NF1 individuals[127-129]. The most common lesions include aneurysms or stenoses of aortic, renal, carotid and cerebral arteries[128,129]. Several mouse models provide compelling evidence that haploinsufficiency alone causes NF1-related vasculopathy. In Nf1+/- mice; in WT mice transplanted with
While all NF1+/- cells have reduced WT neurofibromin, the degree of reduction can be highly variable across NF1 individuals. For example, analysis of neurofibromin levels in fibroblasts from 11 NF1 individuals with different germline NF1 mutations identified two distinct groups: one with < 25% of reductions in neurofibromin and the other with > 70%[130]. The findings were suspected to reflect NF1 allelic imbalance[131,132]. All cases had high RAS activity, indicating that even small decreases in neurofibromin can significantly dysregulate RAS. However, dopamine levels in neural progenitor cells derived from these fibroblasts were variably decreased and linearly correlated with neurofibromin levels[130]. The findings suggest that near-complete restoration of neurofibromin may be required to correct RAS-dependent manifestations, while small increases in neurofibromin may improve some RAS-independent manifestations (e.g. CDs secondary to lowered dopamine).
In summary, the high prevalence, generalized involvement, and mouse study findings support that NF1-haploinsufficiency alone is responsible for a significant portion of NF1-related morbidity (CD/SD, osteopenia and osteoporosis), mortality (vasculopathy) and altered growth (short stature and macrocephaly). Moreover, in mouse models, many of these manifestations are indeed preventable or reversible [Table 3]. The findings support NF1-HCT as a strategy to prevent/reverse many NF1-related manifestations.
Haploinsufficient background cells facilitate a cLOF phenotype in NF1
Haploinsufficiency (NF1+/-) and cLOF (NF1-/-) are interrelated in NF1, with both required for the development of certain manifestations. In the Krox20-cre CMM, Nf1-/- SCs only develop into pNFs if they are within a haploinsufficient microenvironment[7], of which Nf1+/- mast cells are a key component[134]. In the same CMM, correction of the haploinsufficient background by transplantation of WT BM prevents pNF development[135]. In the GFAP-cre CMM, where astrocytes have cLOF, OPGs develop in all mice with a haploinsufficient background, but in no mice with a WT background[136]. In the Prss56-cre, Plp-cre, and Hoxb7-cre CMMs, mice with Nf1-/- SCs develop neurofibromas in both haploinsufficient and WT backgrounds; however, symptoms develop sooner and/or survival times are shorter in mice with haploinsufficient backgrounds [Table 4][8,9,137]. The variable reliance on a haploinsufficient background seen in these neural crest (NC)-SC axis Nf1 CMMs likely stems from the fact that they knockout Nf1 within different NC-derived cell populations[137]. Nonetheless, all CMMs show that a haploinsufficient background promotes tumor formation by Nf1-/- SCs. Interestingly, when Plp-cre CMM mice with pNFs were allowed to age, MPNSTs developed in 10% of mice with an Nf1+/+ (normal) microenvironment but in none of the mice with an Nf1+/− (haploinsufficient) microenvironment. The authors proposed a model wherein some NF1 haploinsufficient immune cells in the tumor microenvironment (i.e., NF1+/− mast cells, macrophages) promote pNF development, whereas others (i.e., NF1+/− T cells) delay pNF progression to MPNST[138].
Beneficial effect of WT background in Nf1 CMMs
| Cre-linked gene promoter | Nf1-/-cell | Tumor type | Results based on Nf1 in background cells | Beneficial effect of WT (+/+) background | Publication | |
| +/- | +/+ | |||||
| GFAP | Astrocyte | OPG | Tumors in all mice | No tumors | Prevented tumors | [17] |
| Krox20 | NC-SC axis | pNF | Tumors in all mice | No tumors | Prevented tumors | [7] |
| Plp | NC-SC axis | pNF | 50% survival (months) | Increased median survival by 6 months (1.8-fold increase) | [8] | |
| 8 | 14 | |||||
| Hoxb7 | NC-SC axis | pNF & cNF | 50% survival (months) | Increased median survival by 11.5 months (1.7-fold increase) | [9] | |
| 17.5 | 29 | |||||
| Prss56 | Boundary cap cells | pNF & cNF | Symptom onset (months) | Increased mean time of symptom onset by 3 months (1.3-fold increase) | [137] | |
| 10 | 13 | |||||
Implications for NF1-HCT
The essential role of NF1-haploinsufficiency in the disease phenotype, whether acting singlehandedly or in concert with cLOF, makes NF1-HCT a promising strategy, with the potential to prevent, reverse or delay virtually all NF1 manifestations. Moreover, as the relative number of NF1-/- cells is expected to be much less in NF1 individuals (accumulate throughout a lifetime) than in the CMMs outlined here (all produced during embryogenesis), haploinsufficiency correction may show a greater protective effect in NF1 individuals than in CMMs. Also, because baseline neurofibromin levels are highly variable across NF1 individuals, clinical monitoring with a pharmacodynamic biomarker may be valuable to guide dosing to achieve optimal neurofibromin levels. Given that one small study showed increased pNF progression to MPNST in aged mice with WT backgrounds, NF1-HCT animal studies and clinical trials will need to further study if NF1-HCT risks inducing malignant transformation of NF1-related tumors.
Increased NF1 gene copy phenotype
While NF1 is caused by a loss of function in NF1, there are 29 reported individuals possessing three NF1 alleles due to microduplication of NF1 and additional genes within and surrounding the NF1 locus[35]. Developmental delay and intellectual disability are frequently present in affected individuals, whereas NF1-specific manifestations are typically absent. This raises the possibility of an adverse effect of excess WT neurofibromin; however, given that elevated neurofibromin levels have not been confirmed and other genes are also duplicated, the role each element plays in this rare phenotype is unknown[35,139-141]. In the absence of regulatory feedback loops, an extra NF1 allele would be expected to decrease RAS-GTP levels, the exact significance of which is difficult to predict. However, mouse models with germline Ras deletions (in contrast to oncogenic mutations) and normal phenotypes may offer some insight. In particular, mice with the following Ras isoform deletions are viable and develop normally without phenotypic manifestations: N-ras+/-, N-ras-/-[142], H-ras+/-, H-ras-/-[143], and K-ras+/-[144]. Mice with certain combinations of Ras isoform deletions are also viable: N-ras-/-/H-ras-/- and N-ras+/-/K-ras+/-[145]. The findings imply significant functional overlap amongst RAS isoforms, suggesting that decreased RAS-GTP secondary to excess neurofibromin may be functionally tolerated to some degree. However, given the potential for adverse effects secondary to neurofibromin overcorrection, the incorporation of a biomarker to guide NF1-HCT dosing should prove beneficial, if not essential.
HAPLOINSUFFICIENCY IN OTHER DISEASES
Haploinsufficiency is not unique to NF1, with over 660 genes known to cause a broad range of human diseases due to haploinsufficiency[146,147]. These include cancer and tumorigenesis (e.g. familial adenomatous polyposis), mental retardation (e.g. Deletion 22q11.2 syndrome), neurological disorders (e.g. epilepsy in tuberous sclerosis complex and Dravet syndrome), growth retardation, and immunodeficiency[147]. Haploinsufficiency is also an important contributor to certain sporadic cancers[148].
Two haploinsufficiency syndromes demonstrate similarities with NF1, including the potential for HCT. Glucose transporter-1 deficiency syndrome (Glut1DS), caused by monoallelic SLC2A1 gene mutations, results in decreased Glut1 protein levels and impaired brain glucose transport. Patients present with infantile drug-resistant seizures and developmental delay. Early intervention with a ketogenic diet, which provides a non-glucose source of brain fuel, reduces disease severity; however, some symptoms persist and long-term treatment is challenging[149,150]. Next, monoallelic SCN1A mutations result in decreased voltage-gated sodium channel 1.1 (NaV1.1) levels. Individuals present with epilepsy syndromes ranging from mild to severe (e.g. Dravet syndrome)[151]. Similar to NF1, SLC2A1 and SCN1A undergo multiple distinct pathogenic mutations, and affected individuals display a wide range of severity. Importantly, HCT improves the phenotype in Slc2a1+/- and Scn1a+/- CMMs (see next section).
HAPLOINSUFFICIENCY CORRECTION AS A TREATMENT APPROACH
Evidence supporting HCT has been reported in NF1 and other haploinsufficiency disease models, as illustrated in the following examples: (1) Introducing WT neurofibromin into NF1+/- fibroblasts restores function and normalizes pERK without signs of toxicity when neurofibromin levels far exceed normal[119], (2) PAX6-related aniridia is a haploinsufficient panocular condition with substantial visual impairment. In a representative CMM (Pax6Sey-Neu/+), small molecule administration (MEKi) increased PAK6 expression, corrected PAK6-haploinsufficiency, and significantly improved the phenotype, including increased retinal function and vision[152], (3) In a Glut1DS CMM (Slc2a1+/-), HCT (systemic AAV9-associated Glut1 cDNA) given pre-symptomatically raised cerebral Glut1 and CSF glucose levels, decreased seizure activity, and improved motor performance. Similarly-treated adult mice showed no benefit, indicating the importance of early intervention[153]. A small molecule approach to Glut1DS-HCT has also been suggested[150], (4) In Dravet syndrome, a sodium channelopathy caused in most cases by SCN1A-haploinsufficiency, patients present in infancy with therapy-resistant severe epilepsy[151,154]. Delivery of a dCas9-based system that increases Scn1a transcription resulted in increased NaV1.1 channel protein in Scn1a+/- (but not Scn1a+/+) neurons and rescue of membrane excitability and action potential firing. AAV9-associated intraventricular delivery of this system in Scn1a+/- mice attenuated hyperthermia-induced seizures[154], (5) CRISPR activation of the normal allele of Sim1- and Mc4r-haploinsufficient neurons in mice corrected haploinsufficiency and the obesity phenotype[146], (6) Angelman Syndrome is a severe developmental disorder caused by an abnormal maternally-derived UBE3A gene. Using a high-content screen, a small molecule (Topotecan) was identified that could ‘unsilence’ the paternally-derived Ube3a gene (purportedly by reducing transcription of a regulatory antisense RNA), resulting in increased paternally-derived Ube3a transcription and restoration of UBE3A brain levels in mice[155], and (7) A separate high-throughput screen identified diverse small molecule classes, including epigenetic agents, that upregulate transcription of PARK2, which is linked to haploinsufficient familial forms of Parkinson’s disease[156].
NF1-HCT USING SMALL MOLECULES
Small-molecule drugs are organic compounds with low molecular weight (most are < 500 Daltons)[157,158] that are capable of modulating biochemical processes to treat or prevent diseases[159,160]. Small molecule drugs represent approximately 80%-90% of the marketed therapeutics[161] and include common therapeutics such as aspirin (180 Daltons). Small molecule drugs are distinct from the category of biological products that include vaccines, blood products, tissues, cells, gene therapies (e.g., gene replacement, CRISPR gene editing), and recombinant proteins (enzymes and antibodies)[161].
The benefits of small molecules in drug discovery include their well-defined structures, relative ease of manufacturing, oral administration, mostly non-immunogenic profiles, and potential to cross the blood-brain barrier. Candidate small molecules may be discovered through screening compound libraries, including repurposing drug libraries where drug toxicity and safety are already established, allowing for expedited clinical trial evaluation[161] and an easier regulatory approval path.
The known NF1 pathobiology allows for target-based drug discovery, wherein small molecules may be screened based on their ability to raise neurofibromin levels via different mechanisms (e.g., increased NF1 transcription, decreased miRNA repression of NF1 mRNA, and decreased neurofibromin degradation by the proteasome). This approach enables lower development cost, greater ease of structure-activity relationship development, and a faster pathway to clinical trials compared to phenotypic-based approaches[162].
Small molecules have previously been utilized to treat genetic diseases. Cystic fibrosis (CF) is a lethal inherited genetic disease caused by mutations in the gene encoding the CF transmembrane conductance regulator (CFTR) protein. Using combinations of small molecules that decrease abnormal CFTR degradation and increase its effectiveness can confer significant clinical benefits for CF patients[163]. The integrated stress response is activated in the brains of Down syndrome patients and mice, resulting in translation reprogramming. Small molecule inhibition of a branch of the integrated stress response reverses the translation changes, and rescues deficits in long-term memory and synaptic plasticity in Down syndrome mice[164].
Small molecules that alter gene expression indirectly are widely utilized to treat disease. For instance, 10% of FDA-approved cancer drugs target nuclear hormone receptors, which function as transcription factors. For example, prostate cancer, which is hormone-driven and mediated by androgen receptors, responds to various small molecules that downregulate androgen receptor signaling and in turn alter target gene transcription[165]. The anti-rejection drug Tacrolimus causes pronounced immunosuppression by indirectly decreasing IL2 transcription in T-cells. This occurs via binding to immunophilin, which results in calcineurin inhibition and subsequent inactivation of the IL2 transcription factor[166].
TIMING OF NF1-HCT
The primary aim of NF1-HCT is to prevent and delay the onset/progression of NF1 manifestations by normalizing neurofibromin levels, ideally pre-symptomatically, and continuing this protective therapy throughout the NF1 person’s life. To determine the optimum age for NF1-HCT commencement, two main factors must be considered: (i) the temporal and spatial distribution of neurofibromin, and (ii) the onset period of various NF1 manifestations. The importance of establishing the temporal and spatial requirements of HCT, and replenishing protein at an early age, has been noted in other haploinsufficiency disorders[150].
Widespread neurofibromin expression begins during embryogenesis
Whole-body mouse Nf1 mRNA levels are low-undetectable on embryonic days E8-E10 (levels not measured prior to E8) and rise five-fold on E11 to a level that is maintained until birth (E11-E16)[167,168]. Neurofibromin is essentially expressed by all tissues from E11-E16, with fluctuations in some tissues[49,167]. Neurofibromin drops to low-undetectable levels in most tissues upon reaching terminal development, except for the nervous system and adrenal medulla, where it remains enriched[169].
Timeline of NF1 manifestation presentation
NF1 manifestations can be broadly grouped into three age ranges during which they present[1,23,170], although it is not clear if and when somatic inactivation of the normal allele occurs with respect to key manifestations associated with neural crest cells:
1. Gestation-adolescence: pNFs (believed to be congenital), CALMs, juvenile myelomonocytic leukemia (JMML), OPG (most develop before age 6), and skeletal abnormalities (includes congenital lesions such as sphenoid wing and/or long bone dysplasia).
2. Early childhood-adulthood: speech, language and learning issues, and CD/SD.
3. Late childhood-adulthood: cNFs (become detectable in teenage years and increase in number/size with age), MPNST (typically after age 30 but can occur in teenage years), cardiovascular disease, and non-NF1 cancers.
Summary of timing of NF1-HCT
Given that some NF1 manifestations are congenital and widespread neurofibromin expression begins in mice on E11, NF1-HCT would ideally commence at the corresponding period in human gestation (week 4-5)[171]. If intrauterine diagnosis and drug delivery are not possible, then NF1-HCT should start promptly following diagnosis, as significant manifestations begin early in life. Nonetheless, starting NF1-HCT at later ages is still expected to be beneficial in delaying the progression of established manifestations and preventing those that typically occur later. Moreover, NF1-HCT may ameliorate/reverse established manifestations that have an element of plasticity (e.g. CDs) [Figure 3].
Figure 3. Timeline of neurofibromin expression, NF1 manifestations, and timing of NF1-HCT. Blue boxes: neurofibromin expression. Green boxes: three NF1-manifestation groups based on the age of occurrence. NF1-HCT would ideally commence at weeks 4-5 of gestation when neurofibromin is widely expressed at high levels (blue arrow). Otherwise, NF1-HCT should begin as early as possible after birth (green arrow). Even if started later in life (yellow arrows), a benefit is expected given the lifelong increase in NF1 manifestations and non-NF1 tumors, and the potential to reverse established manifestations that have an element of plasticity. SWLBD: Sphenoid wing and/or long bone dysplasia. NF1: Neurofibromatosis type 1; NF1-HCT: NF1 haploinsufficiency correction therapy.
SMALL MOLECULE NF1-HCT VERSUS OTHER NF1 TREATMENTS
While the potential exists for NF1 gene replacement therapy by cDNA delivery, this currently has significant challenges. Traditional gene replacement therapy would require full-length NF1 cDNA (8.5 kb) delivery; however, this far exceeds the maximum size deliverable by standard adeno-associated virus (AAV) vector approaches (4.7 kb)[161]. Alternative large gene delivery approaches have been explored (e.g., oversized gene, and multiple vector systems); however, these have limitations and require careful design and consideration of the target tissues[172,173]. Another potential approach is to deliver only the GRD portion of the NF1 cDNA (~1.1 kb)[174]; however, non-GRD neurofibromin functions are not expected to be restored. The varied distribution and function of neurofibromin isoforms pose another challenge[175], as this complexity will be difficult to replicate with a single cDNA. In the case of gene editing strategies, a personalized approach would be required to specifically edit each of the thousands of unique NF1 mutations[32]. Also, system-wide delivery of any gene therapy to multiple organs, including the brain, would be challenging. Finally, stopping gene therapy would be difficult in the event of adverse effects. By contrast, a small molecule NF1-HCT would (i) be easy to administer, (ii) have systemic distribution, (iii) cross the blood-brain barrier, (iv) utilize the normal NF1 allele, thus ensuring correct neurofibromin isoform ratios and full function, and (v) can be quickly stopped in the event of adverse effects.
NF1 therapies blocking RAS signaling pathways, such as the MEKi Selumetinib, have shown partial shrinkage of NF1 plexiform tumors[176]. However, as neurofibromin’s regulatory role is not limited to one signaling pathway, any therapy targeting a specific downstream effector will likely result in only partial NF1-phenotype resolution, as opposed to therapies that restore neurofibromin levels and thus correct all dysregulated pathways. This necessity for a system-wide therapy to correct haploinsufficiency in NF1 and in neurofibromin-deficient tumor microenvironments has been echoed by other authors[18,119].
Despite the limitations associated with drugs targeting single downstream effectors, these solutions do hold the potential for treating at least a subset of manifestations. As such, a tailored multitherapeutic approach that includes these approaches and NF1-HCT may be beneficial. One scenario includes lifelong NF1-HCT to correct haploinsufficiency in all NF1+/- cells, with supplemental therapies targeting specific manifestations driven by NF1-/- cells (e.g., pNF), as required. Such an approach is expected to provide a lifelong baseline protective state by correcting haploinsufficiency; a broader range of effectiveness due to targeted treatment of specific manifestations that may not be as amenable to NF1-HCT alone; and fewer side effects due to the potential use of smaller drug doses.
ADVANTAGES OF SMALL MOLECULE NF1-HCT
The potential advantages of a small-molecule NF1-HCT approach have been described in detail previously in this review and are summarized here.
1. Treats the underlying cause of NF1 - decreased neurofibromin - using the body’s innate ability to make this protein.
2. Prevention, delay and/or reversion of at least a subset, and potentially the majority, of NF1 manifestations.
3. Affects manifestations caused by haploinsufficiency alone, and those initiated by cLOF that require a haploinsufficient background.
4. Benefit not restricted to a single pathway.
5. Unaffected by lack of genotype-phenotype correlations.
6. Benefit seen regardless of the NF1 gene abnormality, avoiding the complexity of thousands of unique NF1 mutations.
7. Small molecule drugs have an established drug development path compared to traditional gene therapies.
8. Avoids challenges of gene editing/replacement therapy.
9. Dosage can be tailored using a biomarker, and therapy can be halted in the event of adverse effects.
10. Proof of concept for NF1-HCT shown in NF1 and other haploinsufficient genetic disorders.
11. Possible extension to non-NF1 individuals with NF1-haploinsufficient tumors.
12. May provide a drug development platform for 660+ haploinsufficient disorders.
CHALLENGES FOR NF1-HCT
Identifying challenges associated with NF1-HCT is important to optimize and adapt research efforts, as well as prepare for clinical trial approaches. These challenges, along with strategies to overcome them, are listed here.
Establishing a safe and effective level of cellular neurofibromin
As with any novel therapy, a primary challenge is to establish a safe therapeutic window resulting in a beneficial effect while avoiding under- and over-treatment. To achieve this, the use of a pharmacodynamic biomarker will be helpful, and monitoring the clinical response will also be important to assess the impact on phenotype. Once an appropriate dosage is determined for an NF1 individual, this will ideally be consistent and require infrequent adjustment.
Possible reversion of NF1-haploinsufficiency potentially protective features
NF1-haploinsufficiency may be associated with protection from certain processes as follows:
Diabetes: NF1 individuals have a reduced rate of type 2 diabetes (HR: 0.27) and a statistically non-significant reduction in the rate of type 1 diabetes (HR: 0.58)[177]. As NF1-HCT may remove this protective effect, fasting glucose and HbA1c should be monitored to determine if NF1-HCT increases glucose levels.
Progression to MPNST: Given the proposed model wherein the NF1+/− microenvironment may have a protective effect in delaying malignant transformation of pNFs[138], careful evaluation, first in animal models and then in clinical trials, will be required to determine if NF1-HCT potentiates malignant transformation in benign tumors.
Drug delivery to a wide variety of NF1-relevant cell types (e.g., SCs, myeloid cells, neurons)
Resolution is similar to comparable small molecule drug treatment development efforts. This will require an understanding of the NF1 regulatory process differences across key cell types and validation of any potential therapy accordingly.
Possible incomplete resolution in microdeletion cases
Manifestations arising from codeleted genes are not expected to be corrected by NF1-HCT. However, NF1-HCT is still expected to correct the typical NF1 manifestations in this group. Careful follow-up of this subgroup will be instructive to determine the extent of benefit from NF1-HCT.
Adverse effects from upregulation of the abnormal NF1 allele
In cases of missense variants with dominant-negative effects, upregulation of the mutant allele could theoretically worsen clinical manifestations. Therefore, genetic and clinical screening will be required and clinical trials should exclude cases with potential dominant-negative variants. Additionally, careful monitoring of all NF1 individuals receiving therapy will be required to assess for untoward effects.
Off-target effects
Any potential small molecule NF1-HCT may have off-target effects that will need to be evaluated. This is not unique to NF1-HCT, but is applicable to any new small molecule therapy. Given that the desired neurofibromin increase is minimal (nominal 2X increase) and the required drug dose is therefore not expected to be substantial to achieve benefit, this alone may minimize significant off-target effects. The following strategies can also be employed to minimize off-target effects: (i) medicinal chemistry to improve drug specificity; (ii) drug combinations to reduce dosage; and (iii) preference for repurposed drugs with limited off-target effects.
NF1 mosaicism
In NF1 mosaics, careful drug dosage will be required to obtain a beneficial neurofibromin increase in affected cells while not reaching potentially harmful neurofibromin levels in unaffected cells. This may be challenging and may limit the use of NF1-HCT in this group.
CONCLUSION
NF1 has a broad range of manifestations and severity, resulting in decreased quality of life and increased mortality in most patients. Despite the need for a system-wide curative NF1 therapy, none is currently available. Haploinsufficiency is integral to NF1 manifestation development, and HCT has been shown to be potentially useful in NF1 and other haploinsufficient genetic disorders. Consequently, it is predicted that systemic small molecule NF1-HCT has the potential to prevent, delay and/or alleviate a range of NF1 manifestations. In addition, NF1-HCT potentially may be extended to non-NF1 patients with tumors harboring decreased neurofibromin levels, and to demonstrate a drug development platform model for identifying small molecules to treat other haploinsufficient conditions.
DECLARATIONS
AcknowledgmentsWe thank Dr. Robert Kesterson PhD (UAB) and Dr. Kaleb Yohay MD PhD (NYU) for their careful review of the manuscript.
Authors’ contributionsConcept and design of manuscript: Sarnoff H, Croston GE, Frost M
Wrote the manuscript: Frost M
Substantial contribution to the manuscript: Serra E, Viskochil D, Korf BR, Croston GE, Mattson-Hoss MK, Sarnoff H
Availability of data and materialsNot applicable.
Financial support and sponsorshipNone.
Conflicts of interestSarnoff H, Frost M, Croston GE, and Mattson-Hoss MK are core members of Infixion Bioscience Inc. Viskochil D and Korf BR are scientific advisors/consultants for Infixion Bioscience Inc.
Serra E is a scientific advisor to and wet-lab collaborator with Infixion Bioscience Inc.
Viskochil D is an International Consulting Investigator for Alexion/AstraZeneca, and a member of the Speaker Bureau for SpringWorks Therapeutics.
Korf BR is a medical advisor to SpringWorks Therapeutics and GenomeMedical.
Ethical approval and consent to participateNot applicable.
Consent for publicationNot applicable.
Copyright© The Author(s) 2022.
REFERENCES
1. Friedman JM. Neurofibromatosis 1. In: Adam M, Ardinger H, et al., editors. Gene reviews. Seattle: University of Washington; 2019. p. 1-45. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1109/ [Last accessed on 21 Oct 2022].
2. Evans DG, Howard E, Giblin C, et al. Birth incidence and prevalence of tumor-prone syndromes: estimates from a UK family genetic register service. Am J Med Genet A 2010;152A:327-32.
3. Lammert M, Friedman JM, Kluwe L, Mautner VF. Prevalence of neurofibromatosis 1 in German children at elementary school enrollment. Arch Dermatol 2005;141:71-4.
4. Uusitalo E, Rantanen M, Kallionpää RA, et al. Distinctive cancer associations in patients with neurofibromatosis type 1. J Clin Oncol 2016;34:1978-86.
5. Yang FC, Chen S, Robling AG, et al. Hyperactivation of p21ras and PI3K cooperate to alter murine and human neurofibromatosis type 1-haploinsufficient osteoclast functions. J Clin Invest 2006;116:2880-91.
6. Rhodes SD, Yang H, Dong R, et al. Nf1 haploinsufficiency alters myeloid lineage commitment and function, leading to deranged skeletal homeostasis. J Bone Miner Res 2015;30:1840-51.
7. Zhu Y, Ghosh P, Charnay P, Burns DK, Parada LF. Neurofibromas in NF1: Schwann cell origin and role of tumor environment. Science 2002;296:920-2.
8. Liao CP, Booker RC, Brosseau JP, et al. Contributions of inflammation and tumor microenvironment to neurofibroma tumorigenesis. J Clin Invest 2018;128:2848-61.
9. Chen Z, Mo J, Brosseau JP, et al. Spatiotemporal loss of NF1 in Schwann Cell Lineage Leads to Different Types of Cutaneous Neurofibroma Susceptible to Modification by the Hippo Pathway. Cancer Discov 2019;9:114-29.
10. Stansfield BK, Bessler WK, Mali R, et al. Heterozygous inactivation of the Nf1 gene in myeloid cells enhances neointima formation via a rosuvastatin-sensitive cellular pathway. Hum Mol Genet 2013;22:977-88.
11. Lasater EA, Bessler WK, Mead LE, et al. Nf1+/- mice have increased neointima formation via hyperactivation of a Gleevec sensitive molecular pathway. Hum Mol Genet 2008;17:2336-44.
12. Li F, Downing BD, Smiley LC, et al. Neurofibromin-deficient myeloid cells are critical mediators of aneurysm formation in vivo. Circulation 2014;129:1213-24.
13. Silva AJ, Frankland PW, Marowitz Z, et al. A mouse model for the learning and memory deficits associated with neurofibromatosis type I. Nat Genet 1997;15:281-4.
14. Cui Y, Costa RM, Murphy GG, et al. Neurofibromin regulation of ERK signaling modulates GABA release and learning. Cell 2008;135:549-60.
15. Costa RM, Federov NB, Kogan JH, et al. Mechanism for the learning deficits in a mouse model of neurofibromatosis type 1. Nature 2002;415:526-30.
16. Li W, Cui Y, Kushner SA, et al. The HMG-CoA reductase inhibitor lovastatin reverses the learning and attention deficits in a mouse model of neurofibromatosis type 1. Curr Biol 2005;15:1961-7.
17. Bajenaru ML, Zhu Y, Hedrick NM, Donahoe J, Parada LF, Gutmann DH. Astrocyte-specific inactivation of the neurofibromatosis 1 gene (NF1) is insufficient for astrocytoma formation. Mol Cell Biol 2002;22:5100-13.
18. Yap YS, McPherson JR, Ong CK, et al. The NF1 gene revisited - from bench to bedside. Oncotarget 2014;5:5873-92.
19. Hyman SL, Shores A, North KN. The nature and frequency of cognitive deficits in children with neurofibromatosis type 1. Neurology 2005;65:1037-44.
20. Akbarnia BA, Gabriel KR, Beckman E, Chalk D. Prevalence of scoliosis in neurofibromatosis. Spine 1992;17:S244-8.
21. Jaremko JL, MacMahon PJ, Torriani M, et al. Whole-body MRI in neurofibromatosis: incidental findings and prevalence of scoliosis. Skeletal Radiol 2012;41:917-23.
22. Holt JF. 1977 Edward B. D. Neuhauser lecture: neurofibromatosis in children. AJR Am J Roentgenol 1978;130:615-39.
23. Miller DT, Freedenberg D, Schorry E, Ullrich NJ, Viskochil D, Korf BR. Council on Genetics. Health supervision for children with neurofibromatosis type 1. Pediatrics 2019;143:e20190660.
24. Peltonen S, Pöyhönen M. Clinical diagnosis and atypical forms of NF1. In: Upadhyaya M, Cooper DN, editors. Neurofibromatosis type 1: molecular and cellular biology. Oxford: Bios Scientific; 2012. p. 17-30.
25. Mautner VF. Value of whole body MRI in patients with NF1. In: Upadhyaya M, Cooper DN, editors. Neurofibromatosis type 1: molecular and cellular biology. Oxford: Bios Scientific; 2012. p. 83-91.
27. Ferner RE. Clinical aspects of neurofibromatosis 1. In: Upadhyaya M, Cooper DN, editors. Neurofibromatosis type 1: from genotype to phenotype. Oxford: Bios Scientific; 1998. p. 21-38.
28. Easton DF, Ponder MA, Huson SM, Ponder BA. An analysis of variation in expression of neurofibromatosis (NF) type 1 (NF1): evidence for modifying genes. Am J Hum Genet 1993;53:305-13.
29. Rieley MB, Stevenson DA, Viskochil DH, Tinkle BT, Martin LJ, Schorry EK. Variable expression of neurofibromatosis 1 in monozygotic twins. Am J Med Genet A 2011;155A:478-85.
30. Messiaen L, Xie J. NF1 germline and somatic mosaicism. In: Upadhyaya M, Cooper D, editors. Neurofibromatosis type 1: molecular and cellular biology. Berlin Heidelberg: Springer-Verlag; 2012. p. 151-72.
31. García-Romero MT, Parkin P, Lara-Corrales I. Mosaic neurofibromatosis type 1: a systematic review. Pediatr Dermatol 2016;33:9-17.
32. Walker JA, Upadhyaya M. Emerging therapeutic targets for neurofibromatosis type 1. Expert Opin Ther Targets 2018;22:419-37.
33. Piovesan A, Antonaros F, Vitale L, Strippoli P, Pelleri MC, Caracausi M. Human protein-coding genes and gene feature statistics in 2019. BMC Res Notes 2019;12:315.
34. NCBI. Available from: https://www.ncbi.nlm.nih.gov/books/NBK7273/ [Last accessed on 21 Oct 2022].
35. Kehrer-Sawatzki H, Mautner VF, Cooper DN. Emerging genotype-phenotype relationships in patients with large NF1 deletions. Hum Genet 2017;136:349-76.
36. Danglot G, Régnier V, Fauvet D, Vassal G, Kujas M, Bernheim A. Neurofibromatosis 1 (NF1) mRNAs expressed in the central nervous system are differentially spliced in the 5’ part of the gene. Hum Mol Genet 1995;4:915-20.
37. Geist RT, Gutmann DH. Expression of a developmentally-regulated neuron-specific isoform of the neurofibromatosis 1 (NF1) gene. Neuroscience Letters 1996;211:85-8.
38. Kaufmann D, Müller R, Kenner O, et al. The N-terminal splice product NF1-10a-2 of the NF1 gene codes for a transmembrane segment. Biochem Biophys Res Commun 2002;294:496-503.
39. Nishi T, Lee PSY, Oka K, et al. Differential expression of two types of the neurofibromatosis type 1 ( NF1) gene transcripts related to neuronal differentiation. Oncogene 1991;1:1555-9.
40. Suzuki Y, Suzuki H, Kayama T, Yoshimoto T, Shibahara S. Brain tumors predominantly express the neurofibromatosis type 1 gene transcripts containing the 63 base insert in the region coding for GTPase activating protein-related domain. Biochem Biophys Res Commun 1991;181:955-61.
41. Gutmann D, Geist R, Wright D, Snider W. Expression of the neurofibromatosis 1 (NF1) isoforms in developing and adult rat tissues. Cell Growth Differ 1995;6:315-23.
42. Scheffzek K, Ahmadian MR, Wiesmüller L, et al. Structural analysis of the GAP-related domain from neurofibromin and its implications. EMBO J 1998;17:4313-27.
43. Ballester R, Marchuk D, Boguski M, et al. The NF1 locus encodes a protein functionally related to mammalian GAP and yeast IRA proteins. Cell 1990;63:851-9.
44. Martin GA, Viskoohil D, Bollag G, et al. The GAP-related domain of the neurofibromatosis type 1 gene product interacts with ras p21. Cell 1990;63:843-9.
45. Bos JL, Rehmann H, Wittinghofer A. GEFs and GAPs: critical elements in the control of small G proteins. Cell 2007;129:865-77.
46. Marchuk DA, Saulino AM, Tavakkol R, et al. cDNA cloning of the type 1 neurofibromatosis gene: complete sequence of the NF1 gene product. Genomics 1991;11:931-40.
47. Andersen LB, Ballester R, Marchuk DA, et al. A conserved alternative splice in the von Recklinghausen neurofibromatosis (NF1) gene produces two neurofibromin isoforms, both of which have GTPase-activating protein activity. Mol Cell Biol 1993;13:487-95.
48. Yunoue S, Tokuo H, Fukunaga K, et al. Neurofibromatosis type I tumor suppressor neurofibromin regulates neuronal differentiation via its GTPase-activating protein function toward Ras. J Biol Chem 2003;278:26958-69.
49. Daston MM, Ratner N. Neurofibromin, a predominantly neuronal GTPase activating protein in the adult, is ubiquitously expressed during development. Dev Dyn 1992;195:216-26.
50. Daston MM, Scrable H, Nordlund M, Sturbaum AK, Nissen LM, Ratner N. The protein product of the neurofibromatosis type 1 gene is expressed at highest abundance in neurons, Schwann cells, and oligodendrocytes. Neuron 1992;8:415-28.
51. Simanshu DK, Nissley DV, McCormick F. RAS proteins and their regulators in human disease. Cell 2017;170:17-33.
52. Stowe IB, Mercado EL, Stowe TR, et al. A shared molecular mechanism underlies the human rasopathies Legius syndrome and Neurofibromatosis-1. Genes Dev 2012;26:1421-6.
53. Sherekar M, Han SW, Ghirlando R, et al. Biochemical and structural analyses reveal that the tumor suppressor neurofibromin (NF1) forms a high-affinity dimer. J Biol Chem 2020;295:1105-19.
54. Naschberger A, Baradaran R, Rupp B, Carroni M. The structure of neurofibromin isoform 2 reveals different functional states. Nature 2021;599:315-9.
55. Brannan CI, Perkins AS, Vogel KS, et al. Targeted disruption of the neurofibromatosis type-1 gene leads to developmental abnormalities in heart and various neural crest-derived tissues. Genes Dev 1994;8:1019-29.
56. Jacks T, Shih TS, Schmitt EM, Bronson RT, Bernards A, Weinberg RA. Tumour predisposition in mice heterozygous for a targeted mutation in Nf1. Nat Genet 1994;7:353-61.
57. Wee P, Wang Z. Epidermal growth factor receptor cell proliferation signaling pathways. Cancers 2017;9:52.
58. Hobbs GA, Der CJ, Rossman KL. RAS isoforms and mutations in cancer at a glance. J Cell Sci 2016;129:1287-92.
60. Jindal GA, Goyal Y, Burdine RD, Rauen KA, Shvartsman SY. RASopathies: unraveling mechanisms with animal models. Dis Model Mech 2015;8:769-82.
61. Philpott C, Tovell H, Frayling IM, Cooper DN, Upadhyaya M. The NF1 somatic mutational landscape in sporadic human cancers. Hum Genomics 2017;11:13.
62. Wallace MD, Pfefferle AD, Shen L, et al. Comparative oncogenomics implicates the neurofibromin 1 gene (NF1) as a breast cancer driver. Genetics 2012;192:385-96.
63. See WL, Tan IL, Mukherjee J, Nicolaides T, Pieper RO. Sensitivity of glioblastomas to clinically available MEK inhibitors is defined by neurofibromin 1 deficiency. Cancer Res 2012;72:3350-9.
64. Lauchle JO, Kim D, Le DT, et al. Response and resistance to MEK inhibition in leukaemias initiated by hyperactive Ras. Nature 2009;461:411-4.
65. de Bruin EC, Cowell C, Warne PH, et al. Reduced NF1 expression confers resistance to EGFR inhibition in lung cancer. Cancer Discov 2014;4:606-19.
66. Py C, Christinat Y, Kreutzfeldt M, McKee TA, Dietrich PY, Tsantoulis P. Response of NF1-Mutated Melanoma to an MEK Inhibitor. JCO Precis Oncol 2018;2:1-11.
67. Byron SA, Tran NL, Halperin RF, et al. Prospective feasibility trial for genomics-informed treatment in recurrent and progressive glioblastoma. Clin Cancer Res 2018;24:295-305.
68. Omrani A, van der Vaart T, Mientjes E, et al. HCN channels are a novel therapeutic target for cognitive dysfunction in Neurofibromatosis type 1. Mol Psychiatry 2015;20:1311-21.
69. Tong J, Hannan F, Zhu Y, Bernards A, Zhong Y. Neurofibromin regulates G protein-stimulated adenylyl cyclase activity. Nat Neurosci 2002;5:95-6.
70. Guo HF, The I, Hannan F, Bernards A, Zhong Y. Requirement of Drosophila NF1 for activation of adenylyl cyclase by PACAP38-like neuropeptides. Science 1997;276:795-8.
71. Guo HF, Tong J, Hannan F, Luo L, Zhong Y. A neurofibromatosis-1-regulated pathway is required for learning in Drosophila. Nature 2000;403:895-8.
72. Hannan F, Ho I, Tong JJ, Zhu Y, Nurnberg P, Zhong Y. Effect of neurofibromatosis type I mutations on a novel pathway for adenylyl cyclase activation requiring neurofibromin and Ras. Hum Mol Genet 2006;15:1087-98.
73. Ho IS, Hannan F, Guo HF, Hakker I, Zhong Y. Distinct functional domains of neurofibromatosis type 1 regulate immediate versus long-term memory formation. J Neurosci 2007;27:6852-7.
74. Walker JA, Tchoudakova AV, McKenney PT, et al. Reduced growth of Drosophila neurofibromatosis 1 mutants reflects a non-cell-autonomous requirement for GTPase-Activating Protein activity in larval neurons. Genes Dev 2006;20:3311-23.
75. The I, Hannigan GE, Cowley GS, et al. Rescue of a Drosophila NF1 mutant phenotype by protein kinase A. Science 1997;276:791-4.
76. Hegedus B, Yeh TH, Lee DY, Emnett RJ, Li J, Gutmann DH. Neurofibromin regulates somatic growth through the hypothalamic-pituitary axis. Hum Mol Genet 2008;17:2956-66.
77. Szudek J, Birch P, Friedman JM. Growth in North American white children with neurofibromatosis 1 (NF1). J Med Genet 2000;37:933-8.
78. Elefteriou F, Kolanczyk M, Schindeler A, et al. Skeletal abnormalities in neurofibromatosis type 1: approaches to therapeutic options. Am J Med Genet A 2009;149A:2327-38.
79. Brown JA, Emnett RJ, White CR, et al. Reduced striatal dopamine underlies the attention system dysfunction in neurofibromatosis-1 mutant mice. Hum Mol Genet 2010;19:4515-28.
80. Diggs-Andrews KA, Tokuda K, Izumi Y, Zorumski CF, Wozniak DF, Gutmann DH. Dopamine deficiency underlies learning deficits in neurofibromatosis-1 mice. Ann Neurol 2013;73:309-15.
81. Patrakitkomjorn S, Kobayashi D, Morikawa T, et al. Neurofibromatosis type 1 (NF1) tumor suppressor, neurofibromin, regulates the neuronal differentiation of PC12 cells via its associating protein, CRMP-2. J Biol Chem 2008;283:9399-413.
82. Bellampalli SS, Khanna R. Towards a neurobiological understanding of pain in neurofibromatosis type 1: mechanisms and implications for treatment. Pain 2019;160:1007-18.
83. Fadhlullah SFB, Halim NBA, Yeo JYT, et al. Pathogenic mutations in neurofibromin identifies a leucine-rich domain regulating glioma cell invasiveness. Oncogene 2019;38:5367-80.
84. Shapira S, Barkan B, Friedman E, Kloog Y, Stein R. The tumor suppressor neurofibromin confers sensitivity to apoptosis by Ras-dependent and Ras-independent pathways. Cell Death Differ 2007;14:895-906.
85. Summers MA, Quinlan KG, Payne JM, Little DG, North KN, Schindeler A. Skeletal muscle and motor deficits in neurofibromatosis type 1. J Musculoskelet Neuronal Interact 2015;15:161-70.
86. Cornett KM, North KN, Rose KJ, Burns J. Muscle weakness in children with neurofibromatosis type 1. Dev Med Child Neurol 2015;57:733-6.
87. Pinho RS, Fusão EF, Paschoal JKSF, et al. Migraine is frequent in children and adolescents with neurofibromatosis type 1. Pediatr Int 2014;56:865-7.
88. Licis AK, Vallorani A, Gao F, et al. Prevalence of sleep disturbances in children with neurofibromatosis type 1. J Child Neurol 2013;28:1400-5.
89. Leschziner GD, Golding JF, Ferner RE. Sleep disturbance as part of the neurofibromatosis type 1 phenotype in adults. Am J Med Genet A 2013;161A:1319-22.
90. Lammert M, Friedman JM, Roth HJ, et al. Vitamin D deficiency associated with number of neurofibromas in neurofibromatosis 1. J Med Genet 2006;43:810-3.
91. Tucker T, Schnabel C, Hartmann M, et al. Bone health and fracture rate in individuals with neurofibromatosis 1 (NF1). J Med Genet 2009;46:259-65.
92. Stevenson DA, Viskochil DH, Carey JC, et al. Pediatric 25-hydroxyvitamin D concentrations in neurofibromatosis type 1. J Pediatr Endocrinol Metab 2011;24:169-74.
93. Souza Mario Bueno L, Rosset C, Aguiar E, et al. Vitamin D status and VDR genotype in NF1 patients: a case-control study from Southern Brazil. Int J Endocrinol 2015;2015:402838.
94. Hockett CW, Eelloo J, Huson SM, et al. Vitamin D status and muscle function in children with neurofibromatosis type 1 (NF1). J Musculoskelet Neuronal Interact 2013;13:111-9.
95. Jouhilahti EM, Peltonen S, Heape AM, Peltonen J. The pathoetiology of neurofibromatosis 1. Am J Pathol 2011;178:1932-9.
97. Zou MX, Butcher DT, Sadikovic B, Groves TC, Yee SP, Rodenhiser DI. Characterization of functional elements in the neurofibromatosis (NF1) proximal promoter region. Oncogene 2004;23:330-9.
98. Yang G, Khalaf W, van de Locht L, et al. Transcriptional repression of the Neurofibromatosis-1 tumor suppressor by the t(8;21) fusion protein. Mol Cell Biol 2005;25:5869-79.
99. Lu H, Liu P, Pang Q. MiR-27a-3p/miR-27b-3p promotes neurofibromatosis type 1 via targeting of NF1. J Mol Neurosci 2021;71:2353-63.
100. Cichowski K, Santiago S, Jardim M, Johnson BW, Jacks T. Dynamic regulation of the Ras pathway via proteolysis of the NF1 tumor suppressor. Genes Dev 2003;17:449-54.
101. McGillicuddy LT, Fromm JA, Hollstein PE, et al. Proteasomal and genetic inactivation of the NF1 tumor suppressor in gliomagenesis. Cancer Cell 2009;16:44-54.
102. Hollstein PE, Cichowski K. Identifying the Ubiquitin Ligase complex that regulates the NF1 tumor suppressor and Ras. Cancer Discov 2013;3:880-93.
103. Messiaen L, Wimmer K. NF1 mutational spectrum. In: Kaufmann D, editor. Neurofibromatoses. Basel: KARGER; 2008. p. 63-77.
104. Koczkowska M, Chen Y, Callens T, et al. Genotype-phenotype correlation in NF1: evidence for a more severe phenotype associated with missense mutations affecting NF1 codons 844-848. Am J Hum Genet 2018;102:69-87.
105. Valero MC, Martín Y, Hernández-Imaz E, et al. A highly sensitive genetic protocol to detect NF1 mutations. J Mol Diagn 2011;13:113-22.
106. Kurosaki T, Popp MW, Maquat LE. Quality and quantity control of gene expression by nonsense-mediated mRNA decay. Nat Rev Mol Cell Biol 2019;20:406-20.
107. Ben-Shachar S, Constantini S, Hallevi H, et al. Increased rate of missense/in-frame mutations in individuals with NF1-related pulmonary stenosis: a novel genotype-phenotype correlation. Eur J Hum Genet 2013;21:535-9.
108. Ruggieri M, Polizzi A, Spalice A, et al. The natural history of spinal neurofibromatosis: a critical review of clinical and genetic features. Clin Genet 2015;87:401-10.
109. Martin MC, Zeng G, Yu J, Schiltz GE. Small molecule approaches for targeting the polycomb repressive complex 2 (PRC2) in cancer. J Med Chem 2020;63:15344-70.
110. Young L, Huang Z, Han S, et al. Severe phenotype-linked NF1 substitution mutations in codons 844-848 of neurofibromin act in a dominant negative manner by destabilizing wild-type protein. In: NF Conference Book. Children’s Tumor Foundation; 2020. p. 23. Available from: https://www.ctf.org/images/uploads/documents/2020_NF_Conference_Book.pdf [Last accessed on 21 Oct 2022].
111. Long A, Liu H, Liu J, et al. Analysis of patient-specific NF1 variants leads to functional insights for Ras signaling that can impact personalized medicine. Hum Mutat 2022;43:30-41.
112. Pasmant E, Sabbagh A, Spurlock G, et al. Members of the NF France Network. NF1 microdeletions in neurofibromatosis type 1: from genotype to phenotype. Hum Mutat 2010;31:E1506-18.
113. Barrea C, Vaessen S, Bulk S, Harvengt J, Misson JP. Phenotype-genotype correlation in children with neurofibromatosis type 1. Neuropediatrics 2018;49:180-4.
114. Koczkowska M, Callens T, Chen Y, et al. Clinical spectrum of individuals with pathogenic NF1 missense variants affecting p.Met1149, p.Arg1276, and p.Lys1423: genotype-phenotype study in neurofibromatosis type 1. Hum Mutat 2020;41:299-315.
115. Upadhyaya M, Huson SM, Davies M, et al. An absence of cutaneous neurofibromas associated with a 3-bp inframe deletion in exon 17 of the NF1 gene (c.2970-2972 delAAT): evidence of a clinically significant NF1 genotype-phenotype correlation. Am J Hum Genet 2007;80:140-51.
116. Koczkowska M, Callens T, Gomes A, et al. Expanding the clinical phenotype of individuals with a 3-bp in-frame deletion of the NF1 gene (c.2970_2972del): an update of genotype-phenotype correlation. Genet Med 2019;21:867-76.
117. Rojnueangnit K, Xie J, Gomes A, et al. High incidence of noonan syndrome features including short stature and pulmonic stenosis in patients carrying NF1 missense mutations affecting p.Arg1809: genotype-phenotype correlation. Hum Mutat 2015;36:1052-63.
118. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A 1971;68:820-3.
119. Mellert K, Lechner S, Lüdeke M, et al. Restoring functional neurofibromin by protein transduction. Sci Rep 2018;8:6171.
120. Lasater EA, Li F, Bessler WK, et al. Genetic and cellular evidence of vascular inflammation in neurofibromin-deficient mice and humans. J Clin Invest 2010;120:859-70.
121. Stevenson DA, Elefteriou F. Molecular basis of bone abnormalities in NF1. In: Upadhyaya M, Cooper DN, editors. Molecular and cellular biology of neurofibromatosis type 1. Heidelberg: Springer Verlag; 2012. p. 327-40. Available from: https://link.springer.com/book/10.1007/978-3-642-32864-0 [Last accessed on 21 Oct 2022].
122. Stansfield BK, Ingram DA, Conway SJ, Friedman JM. Molecular basis of cardiovascular abnormalities in NF1. In: Upadhyaya M, Cooper DN, editors. Molecular and cellular biology of neurofibromatosis type 1. Heidelberg: Springer Verlag; 2012. p. 353-66. Available from: https://link.springer.com/book/10.1007/978-3-642-32864-0 [Last accessed on 21 Oct 2022].
123. Yu X, Chen S, Potter OL, et al. Neurofibromin and its inactivation of Ras are prerequisites for osteoblast functioning. Bone 2005;36:793-802.
124. Stevenson DA, Schwarz EL, Viskochil DH, et al. Evidence of increased bone resorption in neurofibromatosis type 1 using urinary pyridinium crosslink analysis. Pediatr Res 2008;63:697-701.
125. Schindeler A, Morse A, Harry L, et al. Models of tibial fracture healing in normal and Nf1-deficient mice. J Orthop Res 2008;26:1053-60.
126. Schindeler A, Birke O, Yu NY, et al. Distal tibial fracture repair in a neurofibromatosis type 1-deficient mouse treated with recombinant bone morphogenetic protein and a bisphosphonate. J Bone Joint Surg Br 2011;93:1134-9.
127. Rasmussen SA, Yang Q, Friedman JM. Mortality in neurofibromatosis 1: an analysis using U.S. death certificates. Am J Hum Genet 2001;68:1110-8.
128. Friedman JM, Arbiser J, Epstein JA, et al. Cardiovascular disease in neurofibromatosis 1: report of the NF1 Cardiovascular Task Force. Genet Med 2002;4:105-11.
129. Oderich GS, Sullivan TM, Bower TC, et al. Vascular abnormalities in patients with neurofibromatosis syndrome type I: clinical spectrum, management, and results. J Vasc Surg 2007;46:475-84.
130. Anastasaki C, Woo AS, Messiaen LM, Gutmann DH. Elucidating the impact of neurofibromatosis-1 germline mutations on neurofibromin function and dopamine-based learning. Hum Mol Genet 2015;24:3518-28.
131. Jentarra GM, Rice SG, Olfers S, Rajan C, Saffen DM, Narayanan V. Skewed allele-specific expression of the NF1 gene in normal subjects: a possible mechanism for phenotypic variability in neurofibromatosis type 1. J Child Neurol 2012;27:695-702.
132. Hoffmeyer S, Assum G, Griesser J, Kaufmann D, Nürnberg P, Krone W. On unequal allelic expression of the neurofibromin gene in neurofibromatosis type 1. Hum Mol Genet 1995;4:1267-72.
133. Molosh AI, Johnson PL, Spence JP, et al. Social learning and amygdala disruptions in Nf1 mice are rescued by blocking p21-activated kinase. Nat Neurosci 2014;17:1583-90.
134. Yang FC, Ingram DA, Chen S, et al. Neurofibromin-deficient Schwann cells secrete a potent migratory stimulus for Nf1+/- mast cells. J Clin Invest 2003;112:1851-61.
135. Yang FC, Ingram DA, Chen S, et al. Nf1-dependent tumors require a microenvironment containing Nf1+/- and c-kit-dependent bone marrow. Cell 2008;135:437-48.
136. Bajenaru ML, Hernandez MR, Perry A, et al. Optic nerve glioma in mice requires astrocyte Nf1 gene inactivation and Nf1 brain heterozygosity. Cancer Res 2003;63:8573-7.
137. Radomska KJ, Coulpier F, Gresset A, et al. Cellular origin, tumor progression, and pathogenic mechanisms of cutaneous neurofibromas revealed by mice with Nf1 knockout in boundary cap cells. Cancer Discov 2019;9:130-47.
138. Brosseau JP, Liao CP, Wang Y, et al. NF1 heterozygosity fosters de novo tumorigenesis but impairs malignant transformation. Nat Commun 2018;9:5014.
139. Kehrer-Sawatzki H, Bengesser K, Callens T, et al. Identification of large NF1 duplications reciprocal to NAHR-mediated type-1 NF1 deletions. Hum Mutat 2014;35:1469-75.
140. Moles KJ, Gowans GC, Gedela S, et al. NF1 microduplications: identification of seven nonrelated individuals provides further characterization of the phenotype. Genet Med 2012;14:508-14.
141. Grisart B, Rack K, Vidrequin S, et al. NF1 microduplication first clinical report: association with mild mental retardation, early onset of baldness and dental enamel hypoplasia? Eur J Hum Genet 2008;16:305-11.
142. Umanoff H, Edelmann W, Pellicer A, Kucherlapati R. The murine N-ras gene is not essential for growth and development. Proc Natl Acad Sci U S A 1995;92:1709-13.
143. Esteban LM, Vicario-Abejón C, Fernández-Salguero P, et al. Targeted genomic disruption of H-ras and N-ras, individually or in combination, reveals the dispensability of both loci for mouse growth and development. Mol Cell Biol 2001;21:1444-52.
144. Koera K, Nakamura K, Nakao K, et al. K-ras is essential for the development of the mouse embryo. Oncogene 1997;15:1151-9.
145. Nakamura K, Ichise H, Nakao K, et al. Partial functional overlap of the three ras genes in mouse embryonic development. Oncogene 2008;27:2961-8.
146. Matharu N, Rattanasopha S, Tamura S, et al. CRISPR-mediated activation of a promoter or enhancer rescues obesity caused by haploinsufficiency. Science 2019;363:eaau0629.
147. Dang VT, Kassahn KS, Marcos AE, Ragan MA. Identification of human haploinsufficient genes and their genomic proximity to segmental duplications. Eur J Hum Genet 2008;16:1350-7.
148. Delaney JR, Patel CB, Willis KM, et al. Haploinsufficiency networks identify targetable patterns of allelic deficiency in low mutation ovarian cancer. Nat Commun 2017;8:14423.
149. Leen WG, Klepper J, Verbeek MM, et al. Glucose transporter-1 deficiency syndrome: the expanding clinical and genetic spectrum of a treatable disorder. Brain 2010;133:655-70.
150. Tang M, Park SH, De Vivo DC, Monani UR. Therapeutic strategies for glucose transporter 1 deficiency syndrome. Ann Clin Transl Neurol 2019;6:1923-32.
151. Brunklaus A, Zuberi SM. Dravet syndrome--from epileptic encephalopathy to channelopathy. Epilepsia 2014;55:979-84.
152. Rabiee B, Anwar KN, Shen X, et al. Gene dosage manipulation alleviates manifestations of hereditary PAX6 haploinsufficiency in mice. Sci Transl Med 2020;12:eaaz4894.
153. Tang M, Gao G, Rueda CB, et al. Brain microvasculature defects and Glut1 deficiency syndrome averted by early repletion of the glucose transporter-1 protein. Nat Commun 2017;8:14152.
154. Colasante G, Lignani G, Brusco S, et al. dCas9-based scn1a gene activation restores inhibitory interneuron excitability and attenuates seizures in dravet syndrome mice. Mol Ther 2020;28:235-53.
155. Huang HS, Allen JA, Mabb AM, et al. Topoisomerase inhibitors unsilence the dormant allele of Ube3a in neurons. Nature 2011;481:185-9.
156. Hasson SA, Fogel AI, Wang C, et al. Chemogenomic profiling of endogenous PARK2 expression using a genome-edited coincidence reporter. ACS Chem Biol 2015;10:1188-97.
157. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 2001;46:3-26.
158. Leeson PD, Springthorpe B. The influence of drug-like concepts on decision-making in medicinal chemistry. Nat Rev Drug Discov 2007;6:881-90.
159. Li Q, Kang C. Mechanisms of action for small molecules revealed by structural biology in drug discovery. Int J Mol Sci 2020;21:5262.
161. Sun W, Zheng W, Simeonov A. Drug discovery and development for rare genetic disorders. Am J Med Genet A 2017;173:2307-22.
163. Middleton PG, Mall MA, Dřevínek P, et al. VX17-445-102 Study Group. Elexacaftor-tezacaftor-ivacaftor for cystic fibrosis with a single phe508del allele. N Engl J Med 2019;381:1809-19.
164. Zhu PJ, Khatiwada S, Cui Y, et al. Activation of the ISR mediates the behavioral and neurophysiological abnormalities in Down syndrome. Science 2019;366:843-9.
165. Porter BA, Ortiz MA, Bratslavsky G, Kotula L. Structure and function of the nuclear receptor superfamily and current targeted therapies of prostate cancer. Cancers 2019;11:1852.
166. Zheng XS, Chan T-F, Zhou HH. Genetic and genomic approaches to identify and study the targets of bioactive small molecules. Chem Biol 2004;11:609-18.
167. Huynh DP, Nechiporuk T, Pulst SM. Differential expression and tissue distribution of type I and type II neurofibromins during mouse fetal development. Dev Biol 1994;161:538-51.
168. Gutmann DH, Cole JL, Collins FS. Chapter 33 Expression of the neurofibromatosis type 1 (NF1) gene during mouse embryonic development. Gene expression in the central nervous system. Elsevier; 1995. p. 327-35.
169. Sherman L, Daston M, Ratner N. Neurofibromin: distribution, cell biology and role in neurofibromatosis type 1. In: Upadhyaya M, Cooper D, editors. Neurofibromatosis type 1 from genotype to phenotype. Oxford: Bios scientific; 1998. p. 113-26.
170. Ratner N, Miller SJ. A RASopathy gene commonly mutated in cancer: the neurofibromatosis type 1 tumour suppressor. Nat Rev Cancer 2015;15:290-301.
172. Patel A, Zhao J, Duan D, Lai Y. Design of AAV vectors for delivery of large or multiple transgenes. Methods Mol Biol 2019;1950:19-33.
173. Trapani I. Adeno-associated viral vectors as a tool for large gene delivery to the retina. Genes 2019;10:287.
174. Bai RY, Esposito D, Tam AJ, et al. Feasibility of using NF1-GRD and AAV for gene replacement therapy in NF1-associated tumors. Gene Ther 2019;26:277-86.
176. Gross AM, Wolters PL, Dombi E, et al. Selumetinib in children with inoperable plexiform neurofibromas. N Engl J Med 2020;382:1430-42.
Cite This Article
Open Access
, ... Herb SarnoffHow to Cite
Frost, M.; Serra E.; Viskochil D.; Korf B. R.; Mattson-Hoss M. K.; Croston G. E.; Sarnoff H. Rationale for haploinsufficiency correction therapy in neurofibromatosis type 1. J. Transl. Genet. Genom. 2022, 6, 403-28. http://dx.doi.org/10.20517/jtgg.2022.14
Download Citation
Export Citation File:
Type of Import
Tips on Downloading Citation
Citation Manager File Format
Type of Import
Direct Import: When the Direct Import option is selected (the default state), a dialogue box will give you the option to Save or Open the downloaded citation data. Choosing Open will either launch your citation manager or give you a choice of applications with which to use the metadata. The Save option saves the file locally for later use.
Indirect Import: When the Indirect Import option is selected, the metadata is displayed and may be copied and pasted as needed.
About This Article
Copyright
Comments
Comments must be written in English. Spam, offensive content, impersonation, and private information will not be permitted. If any comment is reported and identified as inappropriate content by OAE staff, the comment will be removed without notice. If you have any queries or need any help, please contact us at support@oaepublish.com.