Skip to main content
Genetics in Medicine Open logoLink to Genetics in Medicine Open
. 2025 Mar 20;3:103425. doi: 10.1016/j.gimo.2025.103425

Autosomal dominant HK1-related neurodevelopmental disorder with visual defects and brain anomalies (NEDVIBA): An emerging mitochondrial disorder

Bobby G Ng 1, Erik A Eklund 2, Jill A Rosenfeld 3,4, Abdallah F Elias 5, Aya Abu-El-Haija 6, Celine Bris 7, Magalie Barth 8, Jong-Hee Chae 9,10, Murim Choi 11, Holly A Dubbs 12, Carl Fratter 13, Nicola Foulds 14, Candace Gamble 15, Ralitza H Gavrilova 16, Jaclyn Haven 5, Trevor L Hoffman 17, Jill V Hunter 18, Austin Larson 19, Timothy Edward Lotze 4, Pilar Magoulas 4,20, Emily C Magness 4,20, Debra M Bootin 21, Eric D Marsh 12, Victoria Nesbitt 22, Matthew T Pastore 23, Joanna Poulton 24, Shamima Rahman 25,26, Fernando Scaglia 4,20,27, Chaya Murali 4,20, Jennifer Posey 4, Joshua Rotenberg 28, Betsy Schmalz 23, Deepali N Shinde 29, Zöe Powis 30, Rivka Sukenik-Halevy 31,32, Kristen V Truxal 23, Tami Uster 33, Matheus Vernet Machado Bressan Wilke 34, Erik Klee 34, Hyewon Woo 35, Donald Younkin 12, Jianhua Zhao 36, Jorge Granadillo 37, Seema Lalani 4,20, David Chitayat 33, Wendy K Chung 38,, Hudson H Freeze 1,, Volkan Okur 39,
PMCID: PMC12135434  PMID: 40469904

Abstract

Purpose

Hexokinase 1 (HK1) encodes a ubiquitously expressed hexokinase, which is responsible for the first step of glycolysis, phosphorylation of glucose to glucose-6-phosphate. Both autosomal recessive and dominant variants in this gene have previously been shown to cause human disease, and presently, there are clinical data available for 27 individuals with the monoallelic neurodevelopmental disorder with visual defects and brain anomalies. Delineation of the entire phenotypic spectrum and genotype-phenotype relations will aid in management and counseling decisions.

Methods

We present molecular and clinical data on 22 additional individuals with heterozygous, mostly de novo, variants in HK1. We also reviewed data from the published literature.

Results

The clinical manifestations of neurodevelopmental disorder with visual defects and brain anomalies include varying degrees of intellectual disability/developmental delay, hypotonia, epileptic encephalopathy, visual deficits, a Leigh syndrome spectrum pattern on brain magnetic resonance imaging, and elevated lactate in blood and cerebrospinal fluid, suggesting mitochondrial dysfunction. Based on severity, individuals can be classified into mild, moderate, severe, or lethal forms. In terms of genotype-phenotype correlation, we find that all individuals carrying a missense variant at the threonine 457 residue have severe clinical features.

Conclusion

HK1 should be included in mitochondrial disorder gene sequencing panels.

Keywords: Hexokinase, HK1, Leigh syndrome spectrum, Mitochondrial disorder, NEDVIBA

Introduction

In the glycolytic pathway, glucose is converted to pyruvate beginning with the conversion of glucose to glucose-6-phosphate (G6P), catalyzed by hexokinase (HK; adenosine triphosphate [ATP]:D-hexose 6-phosphate transferase, EC 2.7.1.1).1 HK exists as 4 different isozymes,2 encoded by 4 different genes. The ubiquitously expressed HK1 (HGNC:4922), with high expression in the brain, retina, and skeletal muscle, starts with an N-terminal hydrophobic sequence that connects it to porin in the outer membrane of the mitochondria.3 Mammalian HK1 bound to mitochondria forms tetramers, and the HK1/porin interaction facilitates the exchange of ATP/adenosine diphosphate (ADP) with the mitochondrial matrix and thus is part of the oxidative phosphorylation cascade.4,5 HK1 is composed of an N-terminal regulatory portion joined to the C-terminal catalytic region via a short α-helix. The high sequence similarity between the N- and C-terminal halves probably resulted from a duplication/fusion event of a primordial hexokinase gene.6 The enzyme product, G6P, binds to both halves and exerts either allosteric (N-terminal) or direct (C-terminal) inhibition of the enzyme, important for glycolytic feedback inhibition and regulation.6 Furthermore, HK1 is distinct from other HK isozymes in that inorganic phosphate can abolish inhibition by G6P.7

Neuronal development is intricately linked to glucose metabolism. As neurons transition from an immature to a mature state, there is a metabolic shift from anaerobic glycolysis to mitochondrial oxidative phosphorylation.8 Within neurons, HK1 serves as a critical pacemaker, modulating the rate of glycolysis and linking glycolysis with mitochondrial oxidative phosphorylation.5 Variants in HK1 have been associated with a variety of clinical phenotypes. Biallelic variants (compound heterozygous or homozygous) have been implicated in forms of hemolytic anemia9 (OMIM 235700) and hereditary motor and sensory polyneuropathy, type Russe (CMT4G, OMIM 605285),10 whereas autosomal dominant nonsyndromic retinitis pigmentosa 7911 (OMIM 617460) and the recently described neurodevelopmental disorder with visual defects and brain anomalies (NEDVIBA, OMIM 618547) are caused by heterozygous variants.12 Recently, heterozygous de novo regulatory variants within the 42-bp conserved region encompassed by a regulatory element in intron 2 have also been reported to cause congenital hyperinsulinism by turning on hexokinase expression in the pancreatic beta cells.13 Although a recurrent single missense variant p.(Glu874Lys) (NC_000010.11:g.69398758G>A; NM_000188.3:c.2539G>A) has been reported to cause retinitis pigmentosa 79 in many individuals, 27 individuals with a total of 6 heterozygous missense variants causing NEDVIBA have been published.12,14, 15, 16 Here, we present clinical and molecular data on 22 additional individuals with heterozygous, mostly de novo, variants in HK1 with a clinical phenotype similar to the previously published cases including mitochondrial dysfunction and suggest a genotype/phenotype symptom severity correlation for certain variants.

Materials and Methods

All individuals and their biological parents underwent exome sequencing or genome sequencing testing in a diagnostic clinical laboratory. Parental relationships for de novo variant status were determined from exome sequencing or genome sequencing data unless otherwise stated in Supplementary Table 1. The only inclusion criterion for this study was the presence of a rare heterozygous variant in HK1 that would explain the presenting phenotype and exclusion of other potential genetic causes. Retrospective clinical and molecular data were collected in a clinical report form at the participating institutions. Primary fibroblasts were obtained for 2 affected individuals (P2 and P3) grown from a skin biopsy obtained by the participants’ physicians. Fibroblasts were used for measuring HK1 enzyme activity as previously described.17

We classified the severity of the commonly observed major findings that most affect quality of life. Lethal designation was used for anyone who died of their disease. Severe findings are defined by any one of the following: severe developmental delay (DD) (unable to walk independently if older than 2 years of age and unable to speak if older than 3 years of age), profound intellectual disability (ID), chronic respiratory insufficiency, and intractable seizures/epileptic encephalopathy. Moderate findings are defined by any one of the following: at least 1 major structural brain abnormality (cerebral atrophy/microcephaly, cerebellar atrophy, basal ganglia lesions, and agenesis/hypoplasia of corpus callosum), increased cerebrospinal fluid (CSF) and/or serum/plasma lactate, pronounced hypotonia and/or hypertonia, medically responsive seizures, gastrostomy-tube requirement for nutrition, and visual issues of optic atrophy, retinitis pigmentosa, or cortical visual impairment. Individuals without any severe or moderate findings were classified to have mild disease manifestation.

Results

Participants

An international cohort of 22 individuals with rare heterozygous variants in HK1 were aggregated via personal communication and using GeneMatcher.18 In addition to the sporadic occurrences, familial recurrence is also observed in 5 siblings from 2 families. All but 4 individuals had de novo variants, whereas the inheritance was unknown in 4 individuals because of a lack of biological parental samples.

A summary of the demographic, genetic, and medical data along with the disease severity classifications are presented in Supplementary Table 1, and the heatmap summary of main clinical features by variants are shown in Figure 1.

Figure 1.

Figure 1

Heatmap representation of the clinical findings by variant residues across all reported individuals in this study. Red: present; green: absent; blank: NA or NR. ∗At least 1 of the following: cerebral atrophy, cerebellar atrophy, basal ganglia lesions, and corpus callosum agenesis/hypoplasia with or without accompanying ventriculomegaly. CSF, cerebrospinal fluid; CVI, cortical visual impairment; DD, developmental delay; FTT, failure to thrive; G-tube, gastrostomy tube; ID, intellectual disability; NA, not applicable; NR, not reported; OA, optic atrophy; OSA, obstructive sleep apnea; RP, retinitis pigmentosa.

Variants

A total of 10 different genetic variants in HK1 (NC_000010.11:NM_000188.3: NP_000179.2) were identified, of which g.69382591C>T:c.1370C>T:p.(Thr457Met) constitutes 8/22 (36%), g.69382555C>T:c.1334C>T:p.(Ser445Leu) 3/22 (14%), g.69382591C>A:c.1370C>A:p.(Thr457Arg) 3/22 (14%), and g.69380070G>A:c.1240G>A:p.(Gly414Arg) 2/22 (9%) (Supplementary Table 1). The other variants are only present in 1 individual each. Previously unpublished missense variants were p.(Thr457Arg), g.69398614C>A:c.2395C>A:p.(Gln799Lys), g.69398620C>T:c.2401C>T:p.(Arg801Trp), g.69398629C>T:c.2410C>T:p.(Leu804Phe), g.69398667T>G:c.2448T>G:p.(Ser816Arg), and g.69401059C>T:c.2678C>T:p.(Ser893Phe). Variant classifications are according to the American College of Medical Genetics and Genomics guidelines19 and recommendations from ClinGen Sequence Variant Interpretation Working Group (https://clinicalgenome.org/working-groups/sequence-variant-interpretation/). The 2-dimensional locations of all reported variants to date for all HK1-related phenotypes are depicted in Figure 2, and the 3-dimensional proximities of variants reported in individuals with NEDVIBA are shown in Figure 3. All variants reported in this study are absent from population databases gnomAD v4.0.0 (https://gnomad.broadinstitute.org/), TOPMed Freeze 10 (https://bravobeta.sph.umich.edu), All of Us (https://databrowser.researchallofus.org/variants), and UK Biobank (https://www.ukbiobank.ac.uk/) as of July 5, 2024. In silico predictions using 4 computational algorithms (CADD v1.6, REVEL, BayesDel, and AlphaMissense) are provided in Supplementary Table 1.20, 21, 22, 23, 24

Figure 2.

Figure 2

Distribution of HK1 variants on 2D protein schematic that have been reported in individuals affected with different HK1-related phenotypes. The frequencies of reported variants are provided as a representative lollipop plot with the number of cases provided within respective lollipops. Red: NEDVIBA [AD; OMIM 618547]; maroon: neuropathy, hereditary motor, and sensory, Russe type [AR; OMIM 605285]; blue: nonsyndromic retinitis pigmentosa 79 [AD; OMIM 617460]; green: anemia, congenital, nonspherocytic hemolytic, 5, hexokinase deficient [AR; OMIM 235700]; black: congenital hyperinsulinism [AD; PMID: 36333503]. 2D, 2-dimensional; HK, hexokinase; LoF, loss of function; NEDVIBA, neurodevelopmental disorder with visual defects and brain anomalies; PBD, porin binding domain.

Figure 3.

Figure 3

The schematic of hexokinase’s role in the glycolytic pathway (A) and superimposition of the certain NEDVIBA variants on the 3D structure of HK1 (B). HK1 adopts a bi-lobed architecture consisting of 2 globular domains connected by a rigid linker (a-helix). The C-terminal domain has an active site that converts glucose and adenosine triphosphate (ATP) into glucose-6-phosphate (G6P) and adenosine diphosphate (ADP). The N-terminal domain lacks catalytic activity but can bind G6P and serves a regulatory role. The mutations are clustered in 3 regions of the protein corresponding to key sites in the N-terminal domain, the linker region, and the C-terminal domain. p.Gly414, p.Lys418, and p.Ser445 are proximal to the binding site for G6P in the N-terminal domain, suggesting that mutations of these residues may affect the binding or regulatory function of G6P. p.Thr457 is located at the interface between the N-terminal domain and the linker region, which suggests that mutations here could change the structural orientation of the N- and C-terminal domains. p.Arg801, p.Leu804, and p.Ser816 are also clustered together near the linker region, suggesting that variants affecting these residues may disrupt the connection between the N- and C-terminal halves. The p.Ser893 residue is proximal to the binding sites of ATP and glucose, which hints that variants affecting this residue may alter catalytic activity. However, biochemical assays performed on patient cells with the p.(Lys418Glu), p.(Ser445Leu), and p.(Thr457Met) variants have thus far showed that there is no decrease in hexokinase activity. Mapping of the variants onto available atomic models suggests that the variants might induce structural changes to the protein that may have important functional consequences. 3D, 3-dimensional; HK, hexokinase; NEDVIBA, neurodevelopmental disorder with visual defects and brain anomalies.

Genotype-phenotype correlation

As has previously been described, pathogenic heterozygous missense variants in HK1 are associated with a neurodevelopmental phenotype with a wide spectrum of symptoms, sharing features of a mitochondrial disorder including lactic acidosis and basal ganglia lesions on brain magnetic resonance imaging (MRI).12,15 A summary of the major symptom frequency is presented in Figure 1.

Variants at p.Thr457

Eight individuals carry the most frequent p.(Thr457Met) variant, all de novo, and 4 of them (4/6, 6%) were deceased. Three individuals died before or around 2 years of age, and 1 individual died at around age 8 years. All individuals had severe disease manifestations. Structural major brain anomalies (8/8, 100%), severe DD/ID (8/8, 100%), hypotonia/spasticity (8/8, 100%), increased CSF (range 4-9 mmol/L) lactate (4/4, 100%), non-ambulatory (5/5, 100%), microcephaly/small head (5/6, 83%), increased serum lactate (range 3.5-11.5 mmol/L) (3/4, 75%), gastrostomy-tube requirement (5/8, 63%), intractable seizures/epileptic encephalopathy (4/8, 50%), significant visual impairment (4/8, 50%), and chronic respiratory insufficiency (4/8, 50%) were the most commonly observed clinical features. Ketogenic diet was tried on 2 individuals with epileptic encephalopathy with no benefit.

Three individuals carry a novel missense variant p.(Thr457Arg) with clinical findings similar to individuals with the p.(Thr457Met) variant. All 3 individuals (1 full sibling and 2 paternal half-siblings from an egg donor) died before the age of 5 years. All 3 individuals had severe DD, basal ganglia lesions, elevated serum/plasma lactate, chronic respiratory insufficiency, myoclonic jerks, and hypotonia; 2 individuals had a gastrostomy-tube; and the individual who died at age 5 years was nonambulatory. One individual also had cortical visual impairment. The inheritance of the p.(Thr457Arg) variant was unknown in all siblings.

Variants at p.Ser445

Three individuals carry a de novo p.(Ser445Leu) variant, of which 1 was deceased in adulthood. Of the remaining 2 individuals, 1 had severe, and 1 had moderate disease manifestations.

Variants at p.Gly414

Two individuals carry a de novo p.(Gly414Arg) variant. One individual was deceased in adulthood, and the other had moderate disease manifestations. One individual carries a de novo g.69380071G>A: c.1241G>A:p.(Gly414Glu) variant and had severe disease manifestations.

Novel variants within the hexokinase large subdomain 2

Five individuals each had a novel heterozygous variant within the hexokinase large subdomain 2 in the C-terminal half of HK1. All 5 individuals were alive at the time of reporting. All but 1 individual had an established de novo variant, whereas the inheritance was unknown in 1 individual. Three individuals manifest a NEDVIBA clinical phenotype. Individuals with the p.(Leu804Phe) and p.(Ser816Arg) variants had severe disease manifestations with respiratory insufficiency and basal ganglia lesions in both individuals along with other clinical findings compatible with NEDVIBA, including elevated serum/plasma lactate and optic atrophy. The individual with the p.(Ser893Phe) variant had moderate disease manifestations with basal ganglia lesions and elevated CSF lactate and other clinical findings compatible with NEDVIBA. Individuals with the p.(Gln799Lys) and p.(Arg801Trp) variants have nonspecific neurodevelopmental phenotypes. Additionally, the inheritance of the p.(Gln799Lys) variant is unknown, and the clinical phenotype might have been confounded by the prenatal substance exposure and traumatic delivery.

Hexokinase activity results

There was no decrease in hexokinase activity measured in fibroblasts obtained from 2 individuals with the p.(Thr457Met) variant compared with controls (data not shown).

Discussion

Although the widespread utilization of next-generation sequencing studies unraveled molecular diagnoses of many genetic disorders, we also increasingly realize that there could be multiple disease associations for a single gene, eg, LMNA, based on different molecular mechanisms. Biallelic variants in HK1 resulting in loss of hexokinase activity are a well-known cause of hemolytic anemia, and a homozygous founder variant was reported to cause a peripheral neuropathy in a Roma population (Figure 2).9,10 Hemolytic anemia is associated with decreased HK activity in blood, thus causing insufficient glycolytic activity in erythrocytes.25 A neurodevelopmental disorder in individuals with heterozygous variants in HK1 has so far been described in 27 individuals.12,14, 15, 16 In this study, we add clinical and molecular genetic data for another 22 individuals, making a total of 49 reported individuals. In the reported 49 patients, missense variants at 3 amino acid residues (p.Gly414, p.Ser445, and p.Thr457) account for 86% (42/49) of the variants, suggesting that these are mutation hotspots, and only variants affecting these residues are currently being classified as likely pathogenic or pathogenic according to the American College of Medical Genetics and Genomics guidelines modified by ClinGen recommendations. Although the remaining variants are currently of uncertain significance, they are plausible candidates owing to the NEDVIBA spectrum phenotypes in affected individuals and lack of other molecular causes after next-generation sequencing techniques.

There is a wide clinical spectrum across affected individuals, and we delineated 4 major phenotypic groups: mild, moderate, severe, and lethal. Previously described patients also fit well into these groups. At the moderate end, individuals experience symptoms such as chorioretinal dystrophy, cerebellar ataxia, and increased CSF lactate, whereas at the most severe end, individuals experience minimal motor and cognitive development, intractable seizures, chronic respiratory insufficiency, and severely shortened lifespan. Because over 20 individuals so far have been described with a missense variant at the p.Thr457 residue, we can now predict a severe clinical course for variants affecting this residue with greater certainty. Individuals with other variants seem to be less severely affected compared with the p.Thr457 variants; however, severe disease manifestations and death in adulthood were also observed in some individuals with other variants such as p.(Ser445Leu), p.(Gly414Arg), and p.(Leu804Phe). The unbiased distribution of the clinical findings across individuals with different variants is also comparable to our classification system (Figure 1).

All but 1 of the previously reported variants were located in the hexokinase large subdomain 1 (p.(Gly414Glu), p.(Gly414Arg), p.(Lys418Glu), and p.(Ser445Leu)) toward the end of HK1’s N-terminal half and within the connecting alpha helix (p.(Thr457Met)) (Figure 2). In addition to individuals carrying the same variants or novel variants affecting these residues, we also report novel heterozygous missense p.(Gln799Lys), p.(Arg801Trp), p.(Leu804Phe), p.(Ser816Arg), and p.(Ser893Phe) variants located in the C-terminal hexokinase large subdomain 2 of HK1, where the p.(Asp657Asn) variant previously reported by Poole et al15 is also located (Figure 2). Because the C-terminal half of HK1 is hypothesized to be duplicated from the primordial HK1 composed of only the N-terminal half, we compared the amino acid sequences of both large subdomains. Interestingly, the p.(Ser893Phe) variant in the hexokinase large subdomain 2 corresponds to the p.(Ser445Leu) variant in the hexokinase large subdomain 1. When excluding the loss-of-function variants associated with hemolytic anemia, only another recurrent missense variant (p.(Glu847Lys)) in hexokinase large subdomain 2 has been reported to cause an autosomal dominant disorder of nonsyndromic retinitis pigmentosa (Figure 2).11,26

As previously reported in red blood cells of individuals carrying the p.(Lys418Glu) and p.(Ser445Leu) variants,12 we also did not identify any significant decrease in hexokinase activity between primary fibroblasts from 2 unrelated controls and 2 affected individuals (P2 and P3) with the recurrent p.(Thr457Met) variant. Of note, hexokinase activity measurements are primarily established to detect decreased hexokinase activity levels in red blood cells to diagnose HK1-related hemolytic anemia, and whether these assays are sensitive enough to detect other changes, such as constitutional activation, is not known. Although the exact mechanism of disease remains unclear for the variants associated with NEDVIBA, mapping of the variants onto the predicted 3-dimensional structure of HK1 provides some clues (Figure 3). The mutated residues are located near the surface of the protein and are partially buried. One shared feature is substitution of a residue with a smaller side chain to a residue with a bulkier side chain. Consequently, these variants may distort the local structure of the protein around the mutation sites. For the mutation hotspots located around the linker region between the N- and C-terminal halves, structural distortion of this region could cause a functional decoupling between the N- and C-terminal halves. In contrast, variants near the ligand binding sites may alter the affinity for glucose, G6P, ATP, ADP, or inorganic phosphate. Although studies thus far indicate that the activity of mutated HK1s are not decreased in peripheral blood and fibroblast samples, further work will be necessary to assess the functional consequences of the variants in individuals with NEDVIBA.

Similar to the reports by Okur et al12 in 1 individual with the p.(Thr457Met) variant, Poole et al15 in 3 individuals with the p.(Ser445Leu) and p.(Thr457Met) variants, and Wortmann et al16 in 12 individuals with p.(Ser445Leu) and p.(Thr457Met) variants, we also report increased CSF lactate in all individuals (7/7, 100%) in whom CSF lactate was measured (Figure 1). Furthermore, the serum lactate levels in our cohort of severely affected individuals were also increased in 75% (8/12), whereas only 1 of 3 individuals in the Poole et al15 cohort showed lactic acidemia. Individual serum lactate levels were not reported by Wortmann et al16 but noted to be generally within reference ranges. Thus, an increased CSF lactate seems to be a more sensitive biomarker for NEDVIBA than serum lactate in individuals with an HK1 variant, and HK1 should be included in mitochondrial disorder gene sequencing panels. In parallel with the increased CSF lactate, several patients showed MRI findings resembling a Leigh syndrome spectrum pattern with symmetrical lesions in the basal ganglia and brainstem.27 Additionally, findings such as cerebral atrophy, agenesis/hypoplasia of corpus callosum, ventriculomegaly, periventricular cysts, and hyporotation of hippocampi that are more commonly detected in individuals with pyruvate dehydrogenase complex deficiency (PDCD) have also been reported in individuals in this and previous studies.12,15,16 The brain MRIs of individuals were not collected in this study. Detailed analysis of MRI findings of individuals at different ages were recently reported by Wortmann et al.16

Okur et al12 previously postulated that the molecular mechanism(s) of this dominant form of HK1-related NEDVIBA could involve disruption of the G6P binding domain of HK1, thus disabling autoregulation of HK1 activity and causing constitutive phosphorylation of glucose. Indeed, such disinhibition would promote continuous production of pyruvate, with further fermentation of lactate leading to lactic acidosis and hyperalaninemia, as seen in PDCD.15 Additionally, 2 individuals in our study were also reported to have low CSF glucose levels and low CSF/serum glucose ratios, a biomarker of GLUT1 deficiency, as similarly reported by Poole et al15 and Wortmann et al.16 It appears that HK1-related NEDVIBA shares some phenotypic, imaging, and biochemical features with PDCD and GLUT1 deficiency disorders, albeit with a different molecular pathogenesis, leading to a blended phenotype of both disorders.

Multiple therapeutic approaches were used in individuals with HK1-related NEDVIBA, mainly for treating seizures. Probably owing to the overlapping clinical and biochemical characteristics to PDCD and GLUT1 deficiency, ketogenic diet trials were commonly tried but were not effective in 2 individuals carrying the p.(Thr457Met) variant in this study and 5 of 7 individuals carrying the p.(Thr457Met) variant in previous studies.15,16 Antiepileptic drugs, such as clonazepam and oxcarbazepine, have also been used in some individuals with variable response. More data are needed to understand the effectiveness of cofactors, such as thiamine, vitamin B6, and co-enzyme Q10, and whether some agents (ie, sodium-channel blockers and barbiturates) should be avoided in individuals with HK1-related NEDVIBA.

Two individuals in the Okur et al12 study, 2 individuals in the Wortmann et al16 study, and 3 individuals in this study were born into the same families, respectively, and they all had a variant at the p.Thr457 residue. Although the inheritance is unknown for the 3 individuals in this study, they only share the biological father, who is unlikely to be heterozygous given the severity of disease manifestations. This familial recurrence might be due to a very low-level germline and/or confined gonadal mosaicism in the parents. Furthermore, although it is currently not possible to determine on which parental allele these variants have arisen, whether they confer evolutionary advantages to sperm cells might also shed light on their recurrence in unrelated individuals.28

We also provide the in silico prediction scores of the reported variants for 4 common algorithms (Supplementary Table 1). Although CADD v1.6 scores are congruent for all pathogenic and likely pathogenic variants, the most common pathogenic p.(Thr457Met) variant has the lowest REVEL, BayesDel, and AlphaMissense scores. The p.(Thr457Arg) variant also has lower REVEL and BayesDel scores compared with other pathogenic and likely pathogenic variants. Future studies are needed to determine the most accurate computational algorithm for classification of novel variants in HK1.

In summary, we present clinical and molecular genetic data on 22 individuals with HK1-related NEDVIBA, bringing the total number of reported cases to 49. We describe a genotype-phenotype correlation for recurrent variants. In particular, the p.Thr457 variants are consistently associated with a severe disease course that can be lethal. We report 6 novel variants in addition to strengthening the evidence to support pathogenicity of the previously described variants p.(Thr457Met), p.(Ser445Leu), p.(Gly414Arg), and p.(Gly414Glu).

Data Availability

Reported variants in these individuals have been submitted to ClinVar under the following accession numbers: SCV005619967 (p.(Thr457Met)) for individuals 1, 3, 5, 6, and 8; SCV003835364.1 (p.(Thr457Met)) for individuals 2 and 7; SCV001444501.3 (p.(Thr457Met)) for individual 4; SCV005619968 (p.(Thr457Arg)) for individuals 9 to 11; SCV005619969 (p.(Ser445Leu)) for individuals 12 to 14; SCV005619970 (p.(Gly414Arg)) for individuals 15 and 16; SCV005619971 (p.(Gly414Glu)) for individual 17; SCV005619972 (p.(Gln799Lys)) for individual 18; SCV005619973 (p.(Arg801Trp)) for individual 19; SCV005619974 (p.(Leu804Phe)) for individual 20; SCV005619975 (p.(Ser816Arg)) for individual 21; and SCV005619976 (p.(Ser893Phe)) for individual 22.

Conflict of Interest

The Department of Molecular and Human Genetics at Baylor College of Medicine receives revenue from clinical genetic testing conducted at Baylor Genetics Laboratories. Authors Zöe Powis and Deepali N. Shinde were former employees of Ambry Genetics and current employers of Quest Diagnostics. All other authors declare no conflicts of interest.

Acknowledgments

The authors thank the patients and their families for their contributions.

Funding

B.G.N. and H.H.F. are supported by The Rocket Fund and R01DK099551. W.K.C. is supported by National Institute of Child Health and Human Development (NICHD) P50HD109879. This research was supported by the National Institute of Health Research Great Ormond Street Hospital Biomedical Research Centre. The views expressed are those of the authors and not necessarily those of the National Health Service, the National Institute of Health Research, or the Department of Health. C.F., V.N., and J.P. were funded by the UK National Health Service Highly Specialized Service for Rare Mitochondrial Disorders of Adults and Children, and J.P. thanks the Lily Foundation. This study was also supported in part by The National Human Genome Research Institute (NHGRI) UM1 HG006542.

Author Contributions

Conceptualization: V.O., H.H.F., W.K.C., B.G.N., E.A.E; Data Curation: V.O., B.G.N., E.A.E., J.A.R.; Formal Analysis: B.G.N., E.A.E., J.A.R., F.E.A., A.A.-E.-H., C.B., M.B., J.-H.C., M.C., H.A.D., C.F., N.F., C.G., R.H.G., J.H., T.L.H., J.V.H., A.L., T.E.L., P.M., E.M., D.B., E.D.M., V.N., M.T.P., J.P., S.R., F.S., C.M., J.P., J.R., B.S., D.N.S., Z.P., R.S.-H., K.V.T., T.U., M.V.M.B.W., E.K.K., H.W., D.Y., J.Z., J.G., S.R.L., D.C., W.K.C., H.H.F., V.O.; Funding Acquisition: B.G.N., H.H.F., W.K.C., C.F., V.N., J.P., J.P.; Investigation: V.O., B.G.N., E.A.E.; Project Administration: V.O., B.N.H., J.A.R.; Resources: B.G.N., H.H.F., E.A.E., D.C., T.U.; Supervision: H.H.F., W.K.C.; Visualization: V.O., B.G.N., E.A.E.; Writing-original draft: B.G.N., E.A.E., V.O., H.H.F., W.K.C.; Writing-review and editing: B.G.N., E.A.E., J.A.R., F.E.A., A.A.-E.-H., C.B., M.B., J.-H.C., M.C., H.A.D., C.F., N.F., C.G., R.H.G., J.H., T.L.H., J.V.H., A.L., T.E.L., P.M., E.M., D.B., E.D.M., V.N., M.T.P., J.P., S.R., F.S., C.M., J.P., J.R., B.S., D.N.S., Z.P., R.S.-H., K.V.T., T.U., M.V.M.B.W., E.K.K., H.W., D.Y., J.Z., J.G., S.R.L., D.C., W.K.C., H.H.F., V.O.

ORCIDs

Jong-Hee Chae: http://orcid.org/0000-0002-9162-0138

David Chitayat: http://orcid.org/0000-0003-1944-7294

Murim Choi: http://orcid.org/0000-0002-9195-1455

Wendy K. Chung: http://orcid.org/0000-0003-3438-5685

Carl Fratter: http://orcid.org/0000-0001-7125-5391

Bobby G. Ng: http://orcid.org/0000-0003-0649-848X

Volkan Okur: http://orcid.org/0000-0001-6461-0957

Jennifer Posey: http://orcid.org/0000-0003-4814-6765

Jill A. Rosenfeld: http://orcid.org/0000-0001-5664-7987

Hyewon Woo: http://orcid.org/0000-0002-1407-3441

Jianhua Zhao: http://orcid.org/0000-0002-4997-766X

Ethics Declaration

Families were provided written consent under approved study protocols at Sanford Burnham Prebys Medical Discovery Institute (Institutional Review Board 2014-038-22) or Columbia University Medical Center (Institutional Review Board #AAAA5719), or publication consent was provided through the patient’s physician. All legal guardians or patients provided written informed consent for publication.

Footnotes

This article was invited and the Article Publishing Charge (APC) was waived.

Bobby G. Ng and Erik A. Eklund have contributed equally to this work.

Wendy K. Chung, Hudson H. Freeze, and Volkan Okur have contributed equally to this work.

Additional Information

The online version of this article (https://doi.org/10.1016/j.gimo.2025.103425) contains supplemental material, which is available to authorized users.

Contributor Information

Wendy K. Chung, Email: wendy.chung@childrens.harvard.edu.

Hudson H. Freeze, Email: hudson@sbpdiscovery.org.

Volkan Okur, Email: vokur@nygenome.org.

Additional Information

Supplementary Table 1

Clinical and molecular genetics data of individuals with heterozygous HK1 variants

mmc1.xlsx (43.1KB, xlsx)

References

  • 1.Wilson J.E. Hexokinases. Rev Physiol Biochem Pharmacol. 1995;126:65–198. doi: 10.1007/BFb0049776. [DOI] [PubMed] [Google Scholar]
  • 2.Katzen H.M., Schimke R.T. Multiple forms of hexokinase in the rat: tissue distribution, age dependency, and properties. Proc Natl Acad Sci U S A. 1965/10/1965;54(4):1218–1225. doi: 10.1073/pnas.54.4.1218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Wilson J.E. Isozymes of mammalian hexokinase: structure, subcellular localization and metabolic function. J Exp Biol. 2003;206(12):2049–2057. doi: 10.1242/jeb.00241. [DOI] [PubMed] [Google Scholar]
  • 4.Xie G., Wilson J.E. Tetrameric structure of mitochondrially bound rat brain hexokinase: a crosslinking study. Arch Biochem Biophys. 1990;276(1):285–293. doi: 10.1016/0003-9861(90)90040-6. [DOI] [PubMed] [Google Scholar]
  • 5.Cesar MdC., Wilson J.E. Further studies on the coupling of mitochondrially bound hexokinase to intramitochondrially compartmented ATP, generated by oxidative phosphorylation. Arch Biochem Biophys. 1998;350(1):109–117. doi: 10.1006/abbi.1997.0497. [DOI] [PubMed] [Google Scholar]
  • 6.Aleshin A.E., Kirby C., Liu X., et al. Crystal structures of mutant monomeric hexokinase I reveal multiple ADP binding sites and conformational changes relevant to allosteric regulation. J Mol Biol. 2000;296(4):1001–1015. doi: 10.1006/jmbi.1999.3494. [DOI] [PubMed] [Google Scholar]
  • 7.Solheim L.P., Fromm H.J. Effect of inorganic phosphate on the reverse reaction of bovine brain hexokinase. Biochemistry. 1983;22(9):2234–2239. doi: 10.1021/bi00278a027. [DOI] [PubMed] [Google Scholar]
  • 8.Traxler L., Lagerwall J., Eichhorner S., Stefanoni D., D’Alessandro A., Mertens J. Metabolism navigates neural cell fate in development, aging and neurodegeneration. Dis Model Mech. 2021;14(8) doi: 10.1242/dmm.048993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.van Wijk R., Rijksen G., Huizinga E.G., Nieuwenhuis H.K., van Solinge W.W. HK Utrecht: missense mutation in the active site of human hexokinase associated with hexokinase deficiency and severe nonspherocytic hemolytic anemia. Blood. 2003;101(1):345–347. doi: 10.1182/blood-2002-06-1851. [DOI] [PubMed] [Google Scholar]
  • 10.Šafka Brožková D., Haberlová J., Mazanec R., Laštůvková J., Seeman P. HSMNR belongs to the most frequent types of hereditary neuropathy in the Czech Republic and is twice more frequent than HMSNL. Clin Genet. 2016;90(2):161–165. doi: 10.1111/cge.12745. [DOI] [PubMed] [Google Scholar]
  • 11.Sullivan L.S., Koboldt D.C., Bowne S.J., et al. A dominant mutation in hexokinase 1 (HK1) causes retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2014;55(11):7147–7158. doi: 10.1167/iovs.14-15419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Okur V., Cho M.T., van Wijk R., et al. De novo variants in HK1 associated with neurodevelopmental abnormalities and visual impairment. Eur J Hum Genet. 2019;27(7):1081–1089. doi: 10.1038/s41431-019-0366-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Wakeling M.N., Owens N.D.L., Hopkinson J.R., et al. Non-coding variants disrupting a tissue-specific regulatory element in HK1 cause congenital hyperinsulinism. Nat Genet. 2022;54(11):1615–1620. doi: 10.1038/s41588-022-01204-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Peretz R.H., Zein W.M., Hufnagel R.B., et al. A de novo hexokinase 1 (HK1) variant presenting as Boucher-Neuhäuser syndrome. Am J Med Genet A. 2023;191(2):624–629. doi: 10.1002/ajmg.a.63045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Poole R.L., Badonyi M., Cozens A., et al. Expanding the neurodevelopmental phenotype associated with HK1 de novo heterozygous missense variants. Eur J Med Genet. 2023;66(3) doi: 10.1016/j.ejmg.2023.104696. [DOI] [PubMed] [Google Scholar]
  • 16.Wortmann S.B., Feichtinger R.G., Abela L., et al. Clinical, neuroimaging, and metabolic footprint of the neurodevelopmental disorder caused by monoallelic HK1 variants. Neurol Genet. 2024;10(2) doi: 10.1212/NXG.0000000000200146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.McFarlane C.R., Murray J.W. A sensitive coupled enzyme assay for measuring kinase and ATPase kinetics using ADP-specific hexokinase. Bio Protoc. 2020;10(9) doi: 10.21769/BioProtoc.3599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Sobreira N., Schiettecatte F., Valle D., Hamosh A. GeneMatcher: a matching tool for connecting investigators with an interest in the same gene. Hum Mutat. 2015;36(10):928–930. doi: 10.1002/humu.22844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Richards S., Aziz N., Bale S., et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405–424. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ioannidis N.M., Rothstein J.H., Pejaver V., et al. REVEL: an ensemble method for predicting the pathogenicity of rare missense variants. Am J Hum Genet. 2016;99(4):877–885. doi: 10.1016/j.ajhg.2016.08.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Feng B.J. PERCH: a unified framework for disease gene prioritization. Hum Mutat. 2017;38(3):243–251. doi: 10.1002/humu.23158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Rentzsch P., Witten D., Cooper G.M., Shendure J., Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res. 2019;47(D1):D886–D894. doi: 10.1093/nar/gky1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Pejaver V., Byrne A.B., Feng B.-J., et al. Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria. Am J Hum Genet. 2022;109(12):2163–2177. doi: 10.1016/j.ajhg.2022.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Cheng J., Novati G., Pan J., et al. Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science. 2023;381(6664):eadg7492. doi: 10.1126/science.adg7492. eadg7492. [DOI] [PubMed] [Google Scholar]
  • 25.Kanno H., Murakami K., Hariyama Y., Ishikawa K., Miwa S., Fujii H. Homozygous intragenic deletion of type I hexokinase gene causes lethal hemolytic anemia of the affected fetus. Blood. 2002;100(5):1930. doi: 10.1182/blood-2002-05-1599. 1930. [DOI] [PubMed] [Google Scholar]
  • 26.Wang F., Wang Y., Zhang B., et al. A missense mutation in HK1 leads to autosomal dominant retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2014;55(11):7159–7164. doi: 10.1167/iovs.14-15520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.McCormick E.M., Keller K., Taylor J.P., et al. Expert panel curation of 113 primary mitochondrial disease genes for the Leigh syndrome spectrum. Ann Neurol. 2023;94(4):696–712. doi: 10.1002/ana.26716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Goriely A., Wilkie A.O.M. Paternal age effect mutations and selfish spermatogonial selection: causes and consequences for human disease. Am J Hum Genet. 2012;90(2):175–200. doi: 10.1016/j.ajhg.2011.12.017. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Table 1

Clinical and molecular genetics data of individuals with heterozygous HK1 variants

mmc1.xlsx (43.1KB, xlsx)

Data Availability Statement

Reported variants in these individuals have been submitted to ClinVar under the following accession numbers: SCV005619967 (p.(Thr457Met)) for individuals 1, 3, 5, 6, and 8; SCV003835364.1 (p.(Thr457Met)) for individuals 2 and 7; SCV001444501.3 (p.(Thr457Met)) for individual 4; SCV005619968 (p.(Thr457Arg)) for individuals 9 to 11; SCV005619969 (p.(Ser445Leu)) for individuals 12 to 14; SCV005619970 (p.(Gly414Arg)) for individuals 15 and 16; SCV005619971 (p.(Gly414Glu)) for individual 17; SCV005619972 (p.(Gln799Lys)) for individual 18; SCV005619973 (p.(Arg801Trp)) for individual 19; SCV005619974 (p.(Leu804Phe)) for individual 20; SCV005619975 (p.(Ser816Arg)) for individual 21; and SCV005619976 (p.(Ser893Phe)) for individual 22.


Articles from Genetics in Medicine Open are provided here courtesy of Elsevier

RESOURCES