Introduction
RNA polymerase III (POLR3) transcribes genes encoding small, non-coding RNAs including tRNAs, 5S RNA, 7SK RNA, and U6 small nuclear RNA, which are involved in the regulation of transcription, RNA processing, and translation [
1].
Disease-causing variants in genes coding for POLR3 subunits were first discovered in patients with hypomyelinating leukodystrophy. They are located in
POLR3A [
2] and
POLR3B [
1,
3], which encode the largest and second largest subunits of POLR3 forming the catalytic centre of the enzyme, as well as in
POLR1C [
4], a gene encoding a shared POLR1 and POLR3 subunit. The resulting 4H leukodystrophy (hypomyelination, hypodontia, hypogonadotropic hypogonadism) is characterized by hypomyelination in combination with early cerebellar and subsequent pyramidal signs (usually mild) and variable non-neurological manifestations, namely dental and endocrine features as well as myopia [
5]. Ataxia is the predominant clinical finding in 4H leukodystrophy. Dystonia is an additional, common, and initially under-recognized feature in 4H leukodystrophy [
6], but not prominent at disease onset, and basal ganglia abnormalities as a potential correlate of dystonia have not been reported in 4H leukodystrophy. Clinical manifestations and hypomyelination in 4H leukodystrophy are more severe in patients with variants in
POLR3A and
POLR1C than in patients with variants in
POLR3B [
7,
8]; hypomyelination, however, is not obligatory, and manifestation without hypomyelination occurs in patients with variants in
POLR3A or
POLR3B [
9].
During the last years,
POLR3A variants without predominant ataxia have been reported: A striatal manifestation with predominant dystonia and MR involvement of putamen, caudate and red nucleus due to a homozygous founder variant in intron 13 was reported for three patients from two families with a Roma background [
10]. In addition, biallelic
POLR3A variants have been recognized as a cause of hereditary spastic ataxia [
11,
12].
In order to characterize the striatal variant of POLR3A-related disease, we reviewed clinical, genetic, and MRI findings of nine patients with POLR3A variants and striatal changes.
Patients and methods
We retrospectively identified nine patients from eight families with biallelic
POLR3A variants and striatal changes on MRI through the patient database at the Center for Childhood White Matter Disorders Amsterdam. Patients were referred to the Center for Childhood White Matter Disorders Amsterdam after identification of
POLR3A variants, but without typical presentation for 4H leukodystrophy, for diagnostic evaluation. In all patients,
POLR3A variants were identified by diagnostic whole exome sequencing, performed at different centres. Segregation analysis established their biallelic occurrence in all patients except patient 6, of whom only one parent was available for testing and carried one of the patient’s two variants. No other variants were found explaining the movement disorder. NIW saw patients 1, 3, 7–9; IKM saw patient 5, AB patient 2, GB and RH patient 6, AR-P patient 4. Records were reviewed for clinical presentation and are summarized in Table
1; for case histories, see
supplemental material.
Table 1Main clinical characteristics. BEAR brainstem evoked acoustic responses, F female, M male, mo month, n/a not applicable, yrs. years
Gender | F | F | F | M | F | M | M | M | F |
Current age | 2 yrs | 2 yrs | 21 mo | 7 yrs | 7 yrs | 6 yrs | 22 yrs | 26 yrs | 29 yrs |
Affected siblings | No | No | No | No | No | No | Yes, patient 8 | Yes, patient 7 | No |
Consanguinity | No | No | No | No | No | No | Yes | Yes | No |
Age at onset | 2 months | First days of life | First days of life | First year of life | Second year of life | 13 months | 4 years | Second year of life | 14 months |
Signs/symptoms at onset | No smiling, failure to thrive | No crying, absent visual contact | Abnormal movements, restlessness | Global developmental delay, poor facial expression | Mild motor delay | Abnormal gait | Psychomotor retardation | Mild delay in language acquisition | Abnormal gait, prone to falling |
Age at last examination | 17 mo | 2 yrs. | 16 mo | 5 yrs | 5 yrs | 4 yrs. | 21 yrs | 26 yrs | 29 yrs |
Axial hypotonia | Yes | Yes, if relaxed. Severe opisthotonus in agitated episodes | Severe | No | No | Mild | No | No | No |
Head balance | Suboptimal | Suboptimal at 2 mo, lost at 4 mo | Suboptimal | Suboptimal at three mo | Normal | Normal | Normal | Normal | Normal |
Ataxia | No intentional movements | No intentional movements | No | Yes | Mild gait ataxia | Yes | No | No | Head titubation |
Pyramidal signs | No | Yes | No | Yes | No | Yes | No | No | Mild pyramidal signs (legs) |
Extrapyramidal signs | Yes, choreic movements and opisthotonus | Yes, choreic movements | Yes, choreic movements and opisthotonus | Yes, dystonia and bradykinesia | No | Yes, dystonia | Severe, compatible with Parkinsonism; no tremor | Severe, compatible with Parkinsonism; in addition (rubral) tremor (3/s), increasing with action | Yes, (rubral) tremor (3/s), increasing with action; mild posturing |
Eye movements | Saccadic pursuit | No fixation | Short periods of fixation | Normal | Normal | Saccadic pursuits and hypometric saccades | Saccadic pursuit | Saccadic pursuit and hypometric saccades | Saccadic pursuit |
Highest motor achievement | Some head balance | Some head balance | Some head balance, tries to reach for objects | Walks with posterior walker, manages to walk 20 steps without support | Walking without support | Walking without support | Walking without support | Walking without support | Walking without support (age 14 mo), wheelchair dependent from age 12 yrs |
Swallowing problems | Mild | Severe | Severe (nasogastric tube) | Yes, especially with liquids (prone to aspiration) | No | No | No | Severe dysphagia | Yes |
Speech and language | None | None | None | Delayed (uses about 20 words, difficult to understand) | Mild delay in language development, at age 4 yrs., stutter | Speech delay, dysarthria | Severe dysarthria | Language deterioration from age 5 yrs., now anarthric | Severe dysarthria |
Cognition | Severe global delay | Severe global delay assumed | Severe global delay | Not formally tested but seems normal | Mild learning difficulties | Not formally tested but seems normal | Learning disability | Learning disability | Normal |
Epilepsy | Yes, myoclonic jerks from age 15 mo | No | No | No | No | No | No | No | No |
Dentition | Abnormal (lack of maxillary incisors) | Abnormal (delayed dentition, deciduous molars first teeth to erupt) | Abnormal (lack of maxillary incisors) | Abnormal (delayed eruption of maxillary incisors) | Normal | Normal | Normal | Abnormal (first teeth to erupt maxillary incisors, 2 persisting decidual teeth) | Abnormal (molars first to erupt, incisors erupted at 4y of age) |
Puberty development | n/a | n/a | n/a | n/a | n/a | n/a | Normal | Normal | Normal |
Growth | Failure to thrive | Failure to thrive | Failure to thrive | Failure to thrive | Normal | n/a | Very low weight due to inadequate intake | Very low weight due to inadequate intake | Normal |
Head circumference | Normal | Normal | Normal | Normal | Normal | Normal | Normal | Normal | Normal |
Myopia | Not tested | Not tested | Not tested | No | Mild myopia | No | No | No | Mild myopia |
Hearing loss | Not tested | Not tested | Abnormal BEAR | BEAR normal | Not tested, clinically normal | Not tested, clinically normal | Not tested, clinically normal | Not tested, clinically normal | Not tested, clinically normal |
Other | n/a | Laboured breathing | Prone to respiratory tract infections; bacterial meningitis | n/a | n/a | n/a | n/a | n/a | n/a |
The patients’ 18 cranial MRI scans (age at examination 0.5–29 years, mean 9.1 years, median 4.8 years) were systematically reviewed in consensus by a pediatric neuroradiologist (IH) and pediatric neurologist (NIW). Axial T2-weighted (T2w) and T1-weighted (T1w) images were available for all MRI scans, sagittal T1w images for all but the follow-up MRI in patient 6 (sagittal 3D-T2w), diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) for at least one MRI in all patients (13/18 MRIs). MRI was assessed for presence and extent of T2w grey and white matter changes, in particular for involvement of deep grey matter and brainstem tracts, and for corresponding T1-signal changes. DWI and ADC-maps were inspected for restricted diffusion, namely hyperintensity on DWI and corresponding low signal on ADC (below 60 × 10-5 mm2/s), or increased diffusion with high signal on ADC (above 100–110 × 10-5 mm2/s, for basal ganglia and white matter, respectively [
13‐
15]).
T2 gradient echo and susceptibility-weighted images, available for patient 9 and first MRIs of patients 1, 4, 5, and 6 (field strength 1.5 (2) or 3 Tesla(3)), were checked for hypointensities due to calcifications and/or blood degradation products; the cerebral CT scan available for patient 8 was checked for hyperdensities. Spinal MRIs were available for patients 1, 5, and 6.
For comparison of involvement of cerebellar peduncles and/or striatum in typical 4H leukodystrophy, we additionally reviewed 40 MRIs of 36 patients with 4H leukodystrophy and imaging between 2.8 and 40 years previously published [
7].
Discussion
We present nine patients with biallelic variants in POLR3A carrying at least one of two intronic variants (c.1771-6C > G or c.1771-7C > G), with predominantly extrapyramidal manifestations and characteristic MR changes of striatum, superior, and, often, inferior cerebellar peduncle. Their neurological presentation differs from classic 4H leukodystrophy, the initially described presentation of POLR3A variants. It also differs from a subform of spastic ataxia and the rare Wiedemann-Rautenstrauch syndrome (OMIM#264090), which have only recently been associated with POLR3A. Clinical presentation of these nine patients forms a continuum between a severe, extrapyramidal movement disorder with early onset at one end and juvenile parkinsonism with onset in childhood at the other end of the spectrum. Interestingly, six of the nine patients had abnormal dentition comparable with the abnormal dentition seen in 4H leukodystrophy. There was no evidence for endocrine involvement, although only three patients were old enough to exclude delayed puberty due to hypogonadotropic hypogonadism. Severe myopia, which occurs very frequently in children with 4H leukodystrophy and especially POLR3B variants, was not present.
Although our patients share variants with the spastic ataxia cohort (c.1771-7C > G; homozygous in several patients [
12]), none was clinically classified as spastic ataxia. Interestingly, the original description of the patients homozygous for this variant also mentions extrapyramidal features and early onset of disease. And, while striatal changes are not mentioned, FLAIR-hyperintensity along the superior cerebellar peduncles was noted in almost all patients with the c.1909 + 22G > A variant, but, interestingly, not in the two patients homozygous for the c.1771-7C > G variant [
12]. They might thus also be classified as the striatal variant of
POLR3A-associated disease. One variant seen in our cohort, c.1048 + 5G > T, has also been found in spastic ataxia [
12] and Wiedemann-Rautenstrauch syndrome [
16]. However, our patient did not have the intrauterine and marked postnatal growth retardation, lipodystrophy, or distinctive facies characteristic of the progeroid syndrome of Wiedemann-Rautenstrauch [
17].
The two brothers (patients 7 and 8) with the previously described, homozygous c.1771-6C > G variant [
10], had a presentation similar to that of the three initially described patients [
10] with onset in childhood and severe dysarthria, hypokinesia, and rigidity. A prominent, slow resting, and acting tremor, also sometimes called rubral tremor, in addition to severe dysarthria was seen in the older brother and in patient 9, who carried the c.1771-6C > G variant in combination with a missense variant.
The c.1771-6C > G variant was also described in one patient said to have spastic ataxia but also with dystonia [
10,
11], without detailed MRI information. In a publication on atypical radiological findings in 4H leukodystrophy, one patient also carried this variant in heterozygous form, and, in retrospect, his last MRI showed small caudate and putamen with elevated T2 signal in addition to the signal abnormalities in the posterior limb of the internal capsule, fitting with his prominent extrapyramidal symptoms [
9]. Recently, a young child with the c.1771-6C > G variant in trans with a frameshift variant has been published; the MRI shows the typical basal ganglia involvement described in this work, although this was not recognized as abnormal [
18].
Among the other six patients, clinical manifestation varied despite sharing the c.1771-7C > G variant on one allele and the three youngest patients (patients 1–3) were much more severely affected than patients 4–6. The c.1048 + 5G > T variant on the second allele in one severely affected patient has been predicted to cause a frameshift with premature stop of translation [
16]. It can be classified as a loss-of-function variant, similar to the variant found in patient 2 (c.4025-1G > A). The c.1771-7C > G variant itself has been shown to lead to two aberrant transcripts in addition to the normal cDNA, interpreted as activating a leaky splice site with both wild-type and aberrant transcripts [
12]. Similar results were obtained for the c.1771-6C > G variant, with skipping of exon 14 and a premature termination of a part of the transcripts, with the shorter transcript being subject to nonsense-mediated decay [
10].
Findings at brain imaging reflect the prominent extrapyramidal movement disorder: T2-hyperintensity and atrophy of the striatum were present in all patients, either at first imaging or on follow-up. In one case, a small striatum preceded T2-hyperintensity. A normal striatum was not seen after onset of extrapyramidal movement disorder. T2-hyperintensity was discrete and relatively inconspicuous compared with striatal injury, e.g. in glutaric aciduria type 1 or ischemia. Atrophy varied between mild and severe, e.g. in patients 1 and 7 (Figs.
2 and
5), similar to the initially described patients with striatal injury and homozygous c.1771-6C > G variant [
10]. Extrapyramidal signs can also develop in classic 4H leukodystrophy [
6], with visually normal basal ganglia on brain MRI.
The second characteristic MRI feature was involvement of SCP, which was present in all our patients. This included the dentate nucleus and/or its hilus as the starting point of the efferent neurons of SCP in six patients and the red nucleus as a relay station in two of these. ICP was additionally involved in six patients. Involvement of SCP in patients with
POLR3A variants has previously been reported for the spastic-ataxia cohort [
12] and in four of eight atypical patients [
19]. It is also depicted in a report of a patient with hypomyelination and a previously unreported homozygous variant of c.2423G > A in exon 18 (Fig.
2a in [
20]). Involvement of the red nucleus as a relay station of SCP was reported for the three patients homozygous for the c.1771-6C > G variant [
10]. The symmetric, anterior mesencephalic T2-hyperintensity also reported might rather correspond to the superior mesencephalic course of SCP than the proposed intraparenchymal course of the oculomotor nerve [
10].
Changes in SCP, dentate and red nuclei, and ICP were not clearly associated with striatal injury since they preceded striatal injury in two patients and were subsequent in one. Moreover, their decrease and disappearance in two patients suggest that they are a potentially transient phenomenon. While SCP involvement in the spastic-ataxia cohort was thought to represent the structural correlate of the cerebellar manifestation [
12], contribution of SCP and ICP to the clinical picture in our patients is difficult to pinpoint. This is due to the predominantly extrapyramidal movement disorder, the infrequent ataxia, and the unchanged presentation in those patients with decreasing or resolving changes, although the prominent tremor in patients 8 and 9 is certainly compatible with the involvement of the striatum and the dentate outflow tract [
21].
Compared with classic 4H leukodystrophy, infratentorial involvement was practically inverted in patients with the striatal variant of
POLR3A-related disease: While MCP was normal in the striatal variant, T2-hyperintensity of MCP has been noted in reported cases of 4H leukodystrophy [
22‐
24] and was present in 29 of the 36 the previously reported patients with classic 4H leukodystrophy [
7] re-reviewed for comparison. SCP was involved in only two patients with classic 4H leukodystrophy, but in all patients with the striatal variant, ICP in none of the patients with classic 4H leukodystrophy and in six of nine patients with the striatal variant
POLR3A-related disease. Moreover, T2-hyperintensity of cerebellar white matter with relatively T2-hypointense dentate nucleus and early cerebellar atrophy are common in 4H leukodystrophy [
25], while cerebellar signal changes in patients with the striatal variant were restricted to dentate area, and cerebellar atrophy was absent. In addition, none of the 36 patients with classical 4H leukodystrophy re-reviewed for comparison had striatal T2-hyperintensity.
In 4H leukodystrophy, diffuse hypomyelination is a core finding, commonly with some myelination of the visual tract, the pyramidal tract in the posterior limb of the internal capsule, and the anterolateral thalamus [
5,
25]. In contrast, myelination delay and white matter changes in our patients preferentially involved the optic radiation and pyramidal tracts, and none had frank hypomyelination. Thinning of corpus callosum, another common feature of 4H leukodystrophy, though somewhat less common in patients with carrying
POLR3A variants [
5], was only present in one of the nine patients with the striatal variant.
In conclusion, we present nine patients with biallelic variants in POLR3A carrying at least one of two intronic variants and prominent extrapyramidal involvement. MRI is characterized by striatal injury, involvement of SCP and commonly of ICP, and, variably, irregular myelination of pyramidal and visual tracts. Although our study is limited by the relatively small number of patients, clinical manifestation and MRI differ from 4H leukodystrophy and are consistent with a distinct, striatal variant of POLR3A-related disease.
Recognition of the characteristic MRI pattern, including awareness of the potentially relatively mild T2-hyperintensity and atrophy of the striatum, should trigger genetic testing for POLR3A in patients with unexplained extrapyramidal movement disorders.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.