Keywords
Hereditary Sensory and Motor Neuropathy - Charcot-Marie-Tooth Disease - Early Growth
Response Protein 2
INTRODUCTION
Charcot-Marie-Tooth disease (CMT), a group of hereditary sensory-motor neuropathies,
stands as one of the most prevalent inherited neurological disorders worldwide, with
an estimated prevalence ranging from 9.7 to 82.3 in 100 thousand individuals, depending
on ethnic background and ascertainment method.[1]
[2]
[3]
[4]
[5] This group of disorders commonly manifests with slowly progressive distal muscle
weakness and a variable degree of sensory involvement, with phenotypes that vary in
severity and age of onset.[6]
[7] In addition, recent advances have unveiled the increasing genetic complexity underlying
this group of diseases, with more than 100 genes implicated in the pathogenesis of
CMT, each contributing to distinct subtypes and even overlapping with other complex
neurological syndromes.[8]
[9]
[10]
In 1998, Warner et al. hypothesized that the early growth response 2 (EGR2) gene may act as a transcription factor affecting late myelin genes, and that human
myelinopathies of the peripheral nervous system may result from disease-causing variants
in this gene.[11] Since the initial description, numerous variants of the EGR2 gene have been identified, with the majority situated within the zinc-finger (ZNF)
DNA-binding domain. These variants exhibit diverse inheritance patterns, including
autosomal dominant, autosomal recessive, and de novo mutations in sporadic cases.[7]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Most studies on CMT subtypes show that EGR2 accounts for less than 1% of cases.[20]
[21]
[22]
[23]
[24] Despite being a rare cause of CMT, the variable phenotypic presentation and presence
of additional features, such as sensorineural hearing-loss, scoliosis, and cranial
nerve involvement, can make the diagnosis challenging.[19]
[25]
In Central and South America, information regarding the epidemiology and clinical
characteristics of EGR2-related CMT is limited.[21]
[26] The present study conducted a retrospective analysis of the clinical and ancillary
data from four individuals with confirmed molecular diagnosis of EGR2-related CMT and compared these findings with previously reported cases.
METHODS
Patients
We retrospectively analyzed data from four individuals with confirmed molecular diagnosis
of EGR2-related CMT, recorded in our Neuromuscular Outpatient Registry. All individuals received
their diagnosis between 2010 and 2024.
The inclusion criteria were:
-
A clinical diagnosis of CMT established through clinical symptoms and neurophysiological
findings;
-
Confirmation of pathogenic or likely pathogenic variants in the EGR2 gene via genetic testing; and
-
Availability of detailed clinical records and ancillary test results.
The exclusion criterion was patients with a clinical diagnosis of CMT who did not
have genetic confirmation of pathogenic or likely pathogenic EGR2 variants.
The current study received approval from the Ethics Committee of Rede SARAH de Hospitais
de Reabilitação in the Brazilian Federal District of Brasília, in accordance with
Brazilian ethical regulations (CAAE 42388815.2.0000.0022).
Clinical assessment
Patients were categorized based on the age of onset, motor nerve conduction velocity
(MNCV) of the ulnar nerve, and, when available, nerve biopsy findings. The age of
onset was determined based on the age of the first clinical symptom attributed to
CMT.
The terms congenital hypomyelinating neuropathy (CHN) and Dejerine-Sottas neuropathy
(DSN) have been preserved due to their frequent usage in the current medical literature.
However, it is important to acknowledge that these terms represent facets of a wider
phenotypic spectrum of hereditary neuropathies rather than distinct clinical entities.
The CMT Neuropathy Score version 2 (CMTNSv2) was applied to assess the severity of
CMT in three subjects.[27] The score was directly administered to Subject 2, while scores for Subjects 1 and
3 were retrospectively estimated from their clinical records. This scoring system
categorizes patients into mild (CMTNS ≤ 10), moderate (CMTNS 11–20), or severe (CMTNS > 20)
CMT. However, it is essential to note that this scale has not yet been validated in
Portuguese, necessitating the use of a free translation for this analysis.
Neurophysiology
All patients underwent neurophysiological evaluations conducted according to standardized
protocols. These assessments included sensory and motor nerve conduction velocities
measurements and needle electromyography. Subjects were categorized based on the motor
nerve conduction velocity (MNCV) of the ulnar nerve into 3 groups: 1) demyelinating
(MNCV ≤ 35 m/s); 2) intermediate (MNCV ranging from 36–45 m/s); and 3) axonal (MNCV > 45
m/s). Reference values for nerve conduction studies are provided in Supplementary Material Table S1 (available at https://www.arquivosdeneuropsiquiatria.org/wp-content/uploads/2025/01/Supplementary-Material-Table-S1.xlsx).
Next-generation sequencing and analyses
Genomic DNA was extracted from peripheral blood samples using a standard protocol.
For initial screening of DNA samples from all patients diagnosed with demyelinating
CMT, multiplex ligation-dependent probe amplification (MLPA) was used to detect duplications
or deletions in the PMP22 gene.
In suspected cases of demyelinating CMT without male-to-male inheritance or sporadic
cases lacking PMP22 duplication (PMP22dup), Sanger sequencing was performed to analyze the GJB1 gene. Intermediate and axonal CMT cases, as well as those negative for PMP22dup and pathogenic or likely pathogenic GJB1 variants, were further assessed using a genetic panel covering 72 genes (Supplementary Material Table S2–available at https://www.arquivosdeneuropsiquiatria.org/wp-content/uploads/2025/01/Supplementary-Material-Table-S2.xlsx).
Next-generation sequencing (NGS) was conducted using the SureSelect QXT and QXT HS2
Library Prep Kit (Agilent Technologies, Santa Clara, CA, USA) on an Illumina MiSeq
sequencer (Illumina, Inc., San Diego, CA, USA). The sequencing data were processed
with the Illumina Pipeline (Illumina), using the hg19 genome as a reference. Bioinformatic
analysis included base calling, read alignment, variant identification, and variant
annotation through the Varstation v.3.0 software.[28]
To further investigate the inheritance patterns of identified variants, we performed
Sanger sequencing on the parents of subject 2 and the grandparents of subject 1 to
detect the c.1142G > A (p.Arg381His) variant. However, due to logistical challenges
and difficulties in coordinating appointments, testing for the c.1066G > A (p.Glu356Lys)
variant in subject 3 and the c.1234G > C (p.Glu412Gln) variant in subject 4 was not
feasible in their parents, as the patients and their families live at considerable
distances from the hospital facilities.
All identified variants were classified according to the American College of Medical
Genetics (ACMG) standards and guidelines. Cases with pathogenic and likely pathogenic
variants were considered genetically confirmed.
RESULTS
Clinical picture and ancillary tests
Four patients with EGR2-related hereditary neuropathy were thoroughly evaluated. A detailed pedigree chart
was created to illustrate the inheritance patterns and document the family histories
([Figure 1]). Clinical features, nerve conduction studies, and genetic analyses of the subjects
are summarized in [Table 1].
Figure 1 Heredogram illustrating inheritance patterns and family histories in early growth
response 2 (EGR2)-related neuropathy.
Table 1
Clinical characteristics of patients with EGR2-related Charcot-Marie-Tooth disease
Characteristics
|
Subject 1
|
Subject 2
|
Subject 3
|
Subject 4
|
Sex
|
Male
|
Male
|
Female
|
Female
|
Age at onset (years)
|
2
|
30
|
0
|
11
|
Delivery route
|
Normal
|
Normal
|
C-section
|
Normal
|
Age at first evaluation (years)
|
6
|
40
|
3
|
21
|
Family history
|
Autosomal dominant
|
Autosomal dominant
|
Sporadic
|
Sporadic
|
Scoliosis
|
+
|
−
|
+
|
−
|
Foot deformity (pes cavus)
|
+
|
+
|
+
|
+
|
Brainstem auditory evoked potential
|
Normal
|
Normal
|
Abnormal
|
Normal
|
Visual evoked potential
|
Normal
|
Normal
|
N/A
|
N/A
|
Pulmonary function
|
Individual dependent on mechanical ventilation/tracheostomized
|
Mild restrictive ventilatory defect (FVC 80%/FEV1 67%)
|
N/A
|
N/A
|
Dysphagia
|
+
|
−
|
−
|
−
|
Ulnar MNCV (m/s)
|
10.1
|
28.3
|
6.9
|
43
|
Ulnar CMAP (mV)
|
0.3
|
2.2
|
0.5
|
2.7
|
NGS-CMT Panel
|
c.1142G > A (p.Arg381His)
|
c.1142G > A (p.Arg381His)
|
c.1066G > A (p.Glu356Lys)
|
c.1234G > C (p.Glu412Gln)
|
CMTNS - version 2
|
33 (severe)
|
14 (Moderate)
|
32 (severe)
|
N/A
|
Phenotype
|
Familial CMT1D
|
Familial CMT1D
|
Sporadic CHN case
|
Sporadic CMT1D case
|
Abbreviations: CMAP, compound motor action potential; CMTNS, Charcot-Marie-Tooth Neuropathy
Score; EGR2, early growth response 2; MNCV, motor nerve conduction velocity; NGS, next-generation
sequencing.
Subject 1 was first evaluated in our service at age 6 years, due to a history of frequent
trips and falls after starting to walk independently at 14 months. Over time, he developed
dyspnea, hypophonia, scoliosis, and finger contractures, eventually requiring walking
aids due to foot drop. Vocal cord paralysis was considered, but an otorhinolaryngologist
did not evaluate the patient. At the age of 13, he suffered a cardiac arrest during
pneumonia treatment, leading to permanent ventilatory assistance, loss of independent
walking, and placement of a gastrostomy. He died at the age of 18 due to complications
from surgery to repair a chronic pneumothorax. Despite being classified as CMT1D,
this patient's disease course was marked by severe respiratory compromise, requiring
permanent invasive ventilatory support after a cardiac arrest. Postmortem nerve biopsy
revealed features of CMT, including a marked reduction in myelin fiber density, Schwann
cell abnormalities, severe fiber loss, collagen pockets, Büngner bands, rare regenerative
groups, and occasional macrophages and mast cells ([Figure 2]).
Figure 2 Electron microscopy of the sural nerve biopsy. (A) Region demonstrating loss of myelinated fibers (arrowhead indicates a thinly myelinated
fiber). (B) Axon (asterisk) surrounded by concentric layers of Schwann cell laminae (arrowhead)
enveloping bundles of collagen fibers (empty collagen pockets indicated by arrow).
Subject 2, 46-year-old man and the father of subject 1, presented with a decade-long
history of distal leg numbness, ankle twisting, and foot deformity. Physical examination
revealed pes cavus and mild weakness of the lower limbs, along with absent ankle reflexes
and reduced pinprick and vibration sensitivity. Ancillary tests indicated vitamin
B12 deficiency and mild restrictive lung disease.
Subject 3, a 15-year-old female, exhibited hypotonia at birth, delayed motor milestones,
and recurrent pneumonia episodes. Speech therapy evaluation indicated frequent choking,
cyanosis, and respiratory distress during meals, primarily in the later stages, due
to muscular fatigue without coordination disorders. Subsequently, she developed scoliosis
and progressive dysphagia requiring gastrostomy. The patient presented respiratory
symptoms, including episodes of orthopnea and dyspnea in some daily tasks. Although
a diagnosis of obstructive sleep apnea was reported from an external service, the
polysomnography was unavailable for analysis. The chest computed tomography (CT) revealed
significant scoliosis of the thoracic spine with leftward convexity, accompanied by
rib cage deformity and reduced left lung volume. Linear atelectasis was observed in
the posterior lung bases. The esophagram showed multiple gastroesophageal reflux episodes
extending to the thoracic esophagus's upper third, with rapid clearance. Additional
tests indicated partial bilateral involvement of auditory evoked potentials. There
was no family history of neuromuscular disorders, and her parents were non-consanguineous
and asymptomatic. The patient was classified as CHN.
Subject 4, a 21-year-old female, presented at age 12 with a history of stumbling,
leg tingling, and pes cavus, accompanied by urinary urgency. At her last appointment,
the patient was using ankle-foot orthoses as a mobility aid. There was no family history
of neuromuscular disorders, and her parents were non-consanguineous and asymptomatic.
CMTNSv2
The CMTNSv2 was evaluated in three patients, with findings detailed in [Table 1] and Supplementary Material Table S3 (available at https://www.arquivosdeneuropsiquiatria.org/wp-content/uploads/2025/01/Supplementary-Material-Table-S3.xlsx). Subjects 1 and 3 were classified as having severe CMT (CMTNSv2 > 20), while subject
2 was classified as having moderate CMT (CMTNSv2: 11–20).
Neurophysiology
Nerve conduction studies are consistent with CMT1 in 3 subjects, with ulnar nerve
MNCV ≤ 35 m/s. Furthermore, very slow MNCV (≤ 5 m/s) were observed in individuals
1 and 3. Subject 4 presented MNCV in the intermediate range (36–45 m/s) ([Table 2]).
Table 2
Nerve conduction study findings in subjects with EGR2-related Charcot-Marie-Tooth disease
Subject
|
Age (years)
|
Motor nerves
|
CMAP (mV)
|
Distal latency (ms)
|
MNCV (m/s)
|
Sensory nerves
|
SNAP (µV)
|
Subject 1
|
7
|
Median
|
N/A
|
N/A
|
N/A
|
Radial
|
NR
|
Ulnar
|
0.3
|
N/A
|
10.1
|
Ulnar
|
N/A
|
Peroneal
|
N/A
|
N/A
|
N/A
|
Sural
|
NR
|
Tibial
|
N/A
|
N/A
|
N/A
|
|
|
Subject 2
|
41
|
Median
|
0.73
|
17.1
|
29.1
|
Radial
|
NR
|
Ulnar
|
1.3
|
12.3
|
28.3
|
Ulnar
|
NR
|
Peroneal
|
1.77
|
5.44
|
N/A
|
Sural
|
NR
|
Tibial
|
N/A
|
N/A
|
N/A
|
|
|
Subject 3
|
3
|
Median
|
0.2
|
33.6
|
6.8
|
Radial
|
NR
|
Ulnar
|
0.5
|
32.2
|
6.9
|
Ulnar
|
NR
|
Peroneal
|
NR
|
NR
|
NR
|
Sural
|
NR
|
Tibial
|
NR
|
NR
|
NR
|
|
|
Subject 4
|
22
|
Median
|
4.4
|
12.24
|
45
|
Radial
|
N/A
|
Ulnar
|
2.7
|
10.63
|
43
|
Ulnar
|
18.1
|
Peroneal
|
NR
|
NR
|
NR
|
Sural
|
NR
|
Tibial
|
NR
|
NR
|
NR
|
|
|
Abbreviations: CMAP, compound muscle action potential; EGR2, early growth response 2; MNCV, motor nerve conduction velocity; SNAP, sensory nerve
action potential; NR, non-recordable; N/A, not available.
Next-generation sequencing and analyses
Panel-based next-generation sequencing (NGS) identified pathogenic or likely pathogenic
mutations in all four subjects, revealing three different variants ([Table 3] and [Figure 3]).
Figure 3 Schematic representation of the EGR2 gene and the identified variants.
Table 3
Genetic variants identified in subjects with EGR2-related Charcot-Marie-Tooth disease
Subject
|
Phenotype
|
SeqRef
|
Variant
|
State
|
ACMG criteria
|
Classification
|
Reference
|
1
|
CMT1D
|
NM_000399.5
|
c.1142G > A (p.Arg381His)
|
Heterozygous
|
PM1
PM2 PM5
PP5
PS2
PS3
|
Pathogenic
|
30
|
2
|
CMT1D
|
NM_000399.5
|
c.1142G > A (p.Arg381His)
|
Heterozygous
|
PM1
PM2 PM5
PP5
PS2
PS3
|
Pathogenic
|
30
|
3
|
CHN
|
NM_000399.5
|
c.1066G > A (p.Glu356Lys)
|
Heterozygous
|
PM1
PM2 PM5 PP5
PS2
|
Likely
pathogenic
|
33
|
4
|
CMT1D
|
NM_000399.5
|
c.1234G > C (p.Glu412Gln)
|
Heterozygous
|
PM1
PM2 PM5
|
Likely pathogenic
|
This work
|
Abbreviations: ACMG, American College of Medical Genetics; EGR2, early growth response 2; PM, pathogenic moderate; PP, pathogenic supporting; PS,
pathogenic strong.
The first variant, c.1142G > A (p.Arg381His), was identified in a heterozygous state
in subjects 1 and 2, who were part of a previous study on the frequency of genetic
subtypes of CMT published by our group.[21] Sanger sequencing did not detect this variant in the parents of subject 2, suggesting
a de novo mutation. It is not reported in gnomAD exomes and genomes. Most algorithms developed
to predict the effect of missense changes on protein structure and function suggest
that this variant is likely to be disruptive. Assessment of experimental evidence
suggests that this variant results in abnormal protein function.[29] This variant has been previously described in medical literature and is associated
with different phenotypes of hereditary neuropathies.[21]
[25]
[30]
[31] It is classified as pathogenic.
The second variant, c.1066G > A (p.Glu356Lys), was found in a heterozygous state in
subject 3. Although this variant is not reported in gnomAD exomes and genomes, it
is classified as likely pathogenic according to the ACMG standards and guidelines.
It was previously identified in a Chinese cohort of CMT patients without a detailed
phenotype description.[32] It has also been reported in patients with chronic lymphocytic leukemia (CLL) and
associated with a shorter time to treatment and poor overall survival, but without
clinical diagnosis of CMT.[33] Predictive algorithms have conflicting results on this variant's potential for disruption,
and functional studies have yet to confirm these predictions.
The third and novel variant, c.1234G > C (p.Glu412Gln), was detected in a heterozygous
state in subject 4. This variant is also unreported in gnomAD exomes and genomes and
is classified as likely pathogenic according to the ACMG standards and guidelines.
Predictive algorithms have conflicting results regarding this variant's potential
to be disruptive. Functional studies have not yet confirmed these predictions. This
specific variant has not been described in the medical literature, but similar amino
acid changes at the same codon (c.1235A > G (p.Glu412Gly) and c.1234G > A (p.Glu412Lys))
have been reported.[32]
[34]
[35]
DISCUSSION
Our study provides valuable insights into the clinical and genetic spectrum of EGR2-related CMT disease in a Brazilian cohort, addressing a significant gap in epidemiological
data for this condition in Central and South America. Notably, only two studies have
previously reported on the frequency of genetic subtypes of CMT in Brazilian clinical
populations, underscoring the limited data on EGR2-related CMT in the region.[21]
[26] We identified a novel variant, c.1234G > C (p.Glu412Gln), within the ZNF domain
of EGR2, expanding the known mutational spectrum of this gene. This discovery, along with
the significant clinical features observed, such as respiratory complications and
dysphagia, underscores the complex nature of EGR2-related CMT.
Most patients with EGR2-related CMT have mutations in the ZNF DNA-binding domain.[13]
[14]
[15]
[16]
[17]
[18]
[19] The EGR2 gene encodes a C2H2-type zinc-finger transcription factor that regulates the expression of genes involved
in myelin formation and maintenance, including myelin-associated glycoprotein (MAG), myelin basic protein (MBP), myelin protein zero (MPZ), gap junction beta 1 (GJB1), peripheral myelin protein (PMP22), and periaxin (PRX).[12]
[13] The variation in clinical severity seen with disease-causing variants in the ZNF
domain appears to correlate with the level of residual DNA binding.[12]
To date, only two variants outside the three-ZNF DNA-binding domain have been identified,
both located in the R1 domain of EGR2 and following an autosomal recessive inheritance pattern.[11]
[14] In those cases, the proposed mechanism is related to the inhibition of the interaction
of EGR2 with NAB corepressors, which are necessary for the myelination process.[11]
[14]
In our study, all identified pathogenic and likely pathogenic variants were within
the three ZNF regions ([Table 3]). The discovery of the novel likely pathogenic variant c.1234G > C (p.Glu412Gln)
in the ZNF3 domain of EGR2 expands our understanding of the mutational spectrum underlying EGR2-related CMT. However, additional functional studies are required to elucidate the
precise molecular consequences of this variant.
Phenotypic variability is observed among patients with EGR2-related CMT, highlighting the heterogeneous nature of this condition.[7]
[14]
[15]
[16]
[17]
[18] Our data corroborate previous findings, with three individuals classified as CMT1D
and one as CHN. This variability underscores the importance of considering both clinical
and genetic factors in disease classification and prognosis, emphasizing the need
for individualized therapeutic approaches.
Subjects 1 and 2 exhibited restrictive patterns on pulmonary function tests, indicating
respiratory muscle weakness—a recognized complication in CMT.[21]
[25]
[36] Notably, subject 1's disease course was unusually severe for CMT1D, progressing
to respiratory failure and requiring permanent ventilatory support following a cardiac
arrest during an episode of pneumonia. This atypical severity may be related to the
EGR2 variant c.1142G > A (p.Arg381His), which has been associated with severe CMT phenotypes,
including respiratory complications.[25] In contrast, subject 2, his father, who carries the same mutation, presented with
a milder phenotype, limited to distal leg weakness and mild restrictive pulmonary
changes. This intrafamilial variability suggests the influence of additional genetic,
epigenetic, or environmental modifiers in EGR2-related neuropathies. Similarly, the respiratory symptoms and recurrent pneumonia
in subject 3 highlight the importance of proactive respiratory monitoring and management
in these patients, given the broad phenotypic spectrum even within families.
Dysphagia, another significant clinical feature observed in subjects 1 and 3, is likely
due to oropharyngeal muscle weakness, a recognized complication of certain CMT subtypes.[37] Bulbar weakness increases the risk of aspiration and nutritional deficits, contributing
to the overall disease burden. Multidisciplinary care is essential to enhance quality
of life, prevent life-threatening respiratory complications, and minimize the impact
of dysphagia and bulbar involvement in patients with EGR2-related CMT.
Peripheral nerve biopsies in EGR2-related neuropathies commonly show a combination of demyelinating and axonal degeneration.[18]
[34]
[35] Most important findings include thinning of myelin sheaths, frequent onion bulb
formations, and axonal loss with regenerative clusters. Schwann cell morphological
abnormalities are also prominent, reflecting EGR2's critical role in Schwann cell differentiation. Additionally, some patients may
present with a predominantly axonal phenotype, characterized by significant axonal
degeneration with less pronounced demyelination, highlighting the variability in pathological
manifestations of EGR2-related neuropathies.[34]
[35]
Our study has limitations, including a small sample size and retrospective design
that must be considered. Larger, prospective studies are essential to validate our
findings and further explore genotype-phenotype correlations. Furthermore, functional
characterization of the novel variant identified will be crucial to understand its
pathogenicity and molecular mechanisms.
Future research should focus on conducting functional studies to determine the impact
of the novel and other identified variants on EGR2 function. Specifically, transactivation assays could evaluate the transcriptional
activity of these variants, while electrophoretic mobility shift assays (EMSA) would
assess their DNA-binding capabilities.[12]
[18]
[38] Developing animal models to investigate the in vivo effects of EGR2 mutations could provide insights into their role in myelination processes. Furthermore,
research on effective management strategies for respiratory and swallowing complications
in EGR2-related CMT patients is also necessary. Collaborative efforts are needed to gather
more extensive data on EGR2-related CMT in diverse populations, particularly in underrepresented regions like
Central and South America.
In conclusion, our findings highlight the heterogeneity in disease severity among
individuals with EGR2-related CMT, emphasizing the need for a multidisciplinary approach
to patient care. Early genetic diagnosis and comprehensive clinical assessment are
critical for managing this condition's diverse phenotypic spectrum. By advancing our
understanding of EGR2-related CMT, we can improve outcomes for affected patients and their families.
Bibliographical Record
Eduardo Boiteux Uchôa Cavalcanti, Savana Camilla de Lima Santos, Christian Marques
Couto, Galeno Vieira Rocha, Maria Cristina Del Negro Barroso Freitas, Osvaldo José
Moreira do Nascimento. Genetic and clinical spectrum of early growth response 2-related
Charcot-Marie-Tooth disease in a Brazilian cohort. Arq Neuropsiquiatr 2025; 83: s00451806820.
DOI: 10.1055/s-0045-1806820