An Unusual Presentation of CLN3-Associated Batten Disease With Classic Histopathologic and Ultrastructural Findings

To the Editor:

Neuronal ceroid lipofuscinoses (NCLs) are a broad class of inherited lysosomal storage disorders. Known mutations in at least 13 different genes can result in NCL with variable ages of onset, symptoms, and pathologic findings (1). Generally, these patients experience cognitive and motor decline, seizures, visual impairment, and premature death. Pathologically, NCL patients display heterogeneous histologic abnormalities, but consistently exhibit neuronal loss, reactive gliosis, and lysosomal accumulation of autofluorescent storage material or lipopigment (2, 3).

Juvenile-onset NCL has been classically referred to as Batten disease. It is an autosomal recessive condition that is usually caused by mutations in the ceroid-lipofuscinosis, neuronal 3 gene (CLN3). CLN3 encodes battenin, a ubiquitously expressed transmembrane protein of unknown function that is associated with cellular homeostasis and neuronal survival (4). The initial clinical symptom of CLN3-associated NCL is central vision loss, which is usually detected between 4 and 9 years of age. Seizures typically begin early in the second decade of life and affected individuals rarely live beyond their mid-20s (3).

Here, we describe the neuropathologic findings in a woman with CLN3-associated NCL who had no visual complaints until she failed a school vision screening in first grade. When evaluated by an ophthalmologist, her visual acuity was found to be 20/400 OU and she was diagnosed with cone-rod dystrophy. Her vision progressively worsened over the next 24 months and at age 8 the diagnosis was changed to retinitis pigmentosa. She remained well otherwise until age 23 when she experienced 2 generalized seizures. Brain magnetic resonance imaging (MRI) at that time revealed significant, generalized cortical atrophy of unknown etiology. She was started on levetiracetam which was successful in controlling the seizures. The following year, she began to have difficulty choosing her words and also experienced some emotional volatility and short-term memory loss. Further evaluation by an inherited eye disease specialist raised the possibility that her vision loss and cortical atrophy could be related to each other. Clinically focused genetic testing revealed a homozygous 966 bp deletion in the gene CLN3, the most common disease-causing variant reported (3, 5). The patient’s CNS symptoms progressed steadily until her death at 27 years of age.

An autopsy limited to the brain was conducted. Relevant gross findings included a reduced brain weight of 950 g (normal = 1200 g for an adult female), generalized cortical atrophy (Fig. 1A), and a mildly thin cortical ribbon consistent with significant neuronal loss (6). Tan coloration of CNS white matter, pallor in the substantia nigra, and significant bilateral atrophy of the lateral geniculate nuclei were also noted (Fig. 1B, C).

FIGURE 1.

Gross autopsy findings of a patient with CLN3-associated NCL. (A) Generalized cortical atrophy due to neuronal loss (brain weight 950 g). (B) Coronal brain section at the level of the thalamus and posterior hippocampus with mild cortical atrophy and highly atrophic lateral geniculate nuclei (arrowheads). (C) Midbrain section demonstrating abnormal pallor in the substantia nigra (arrows). Scale bar: 1 cm.

The brain was fixed in 10% neutral buffered formalin, cut coronally with extensive sampling, routinely processed, paraffin-embedded, and sectioned for light microscopy. Sections were stained with hematoxylin and eosin (H&E), Luxol fast blue with H&E (LFB/H&E), and periodic acid-Schiff (PAS). Immunohistochemistry (IHC) for glial fibrillary acidic protein (GFAP) was also performed. Microscopic examination of H&E sections revealed widespread CNS involvement and many neurons throughout the brain were filled with intracytoplasmic lipopigment (Fig. 2A, D). This granular inclusion material was found to be PAS- and LFB-positive (Fig. 2B, C, E, F). In addition, diffuse reactive gliosis was detected with the GFAP immunostain (Fig. 2G).

FIGURE 2.

Brain histopathology and ultrastructure of a patient with CLN3-associated NCL. (A) Cingulate gyrus with neuronal loss and vacuolization in the cortex, hematoxylin and eosin (H&E) (200× magnification). (B) Periodic acid-Schiff (PAS)-positive storage material and (C) Luxol fast blue (LFB)-positive storage material in neuronal cytoplasm in the cingulate gyrus (black arrows; 400×). (D) Hippocampus, H&E (200×). (E) PAS-positive storage material and (F) LFB-positive storage material in pyramidal neuronal cytoplasm in the hippocampus (black arrows; 400×). (G) GFAP IHC highlighting reactive astrogliosis in the cortex of the cingulate gyrus (100×). (H–K) Ultrastructural findings in the hippocampus include (H, I) electron dense granular material (yellow arrowheads; 2000× and 6000×, respectively) and (J, K) characteristic fingerprint profiles (yellow arrowheads; 12 000×).

Every brain region sampled showed neurons containing LFB- and PAS-positive intracytoplasmic lipopigment including neocortex (frontal, temporal, parietal, and occipital/calcarine cortex), cingulate gyrus, deep gray matter (basal ganglia, thalamus, and amygdala), bilateral hippocampi, brainstem, and cerebellum. Regarding brainstem sections, inclusions were identified in the substantia nigra (SN), locus ceruleus, medullary olives, and brainstem cranial nerve nuclei. Neurons of the SN contained cytoplasmic lipopigment along with neuromelanin in a background of reactive gliosis and extraneuronal neuromelanin. These findings suggest a component of neurodegeneration, although it is possible that there could have also been delayed pigmentation of the SN due to chronic disease. Cerebellar neuronal inclusions were seen in many surviving Purkinje cells, granule cells, and dentate nucleus neurons. Other than highly significant neuron loss and atrophy of the lateral geniculate nucleus bilaterally, every other sampled section showed a similar level of mild to moderate, diffuse neuron dropout and gliosis.

Formalin-fixed hippocampal tissue was also subsequently post-fixed in 2.5% buffered isosmolar glutaraldehyde, embedded in epon, and sectioned for transmission electron microscopy. Ultrastructural analysis revealed neurons containing cytoplasmic electron dense granular inclusion material (Fig. 2H, I) as well as fingerprint profiles (Fig. 2J, K), which are the predominant type of abnormal intraneuronal deposits seen in juvenile-onset NCL (2, 6). Curvilinear bodies and rectilinear profiles were also seen (data not shown).

Isolated central vision loss is almost always the presenting symptom for juvenile-onset NCL whether it is caused by mutations in CLN3, as in this case, or by hypomorphic genotypes in genes that normally cause infantile or late-infantile forms of the disease. The unusual feature of the present case is the 18-year delay between the first visual symptom and the first extraocular symptom. Although sometimes a protracted course of a degenerative disease can be explained by a milder genotype (7), our patient was homozygous for a 966 bp deletion. Historically referred to as a 1.02 kb deletion, this variant has been found on one or both alleles of 85% of patients with CLN3-associated NCL and leads to a truncated, nonfunctional protein (3, 8).

Although this case is unusual, it is not unique. Indeed, just 2 years after the CLN3 gene was discovered, Jarvela et al. (3) reported the clinical and MRI findings in 36 Finnish patients who harbored the 1.02 kb deletion on one or both alleles. Among these 36 patients was a male homozygote who could still walk and speak clearly at age 31 and 2 homozygous sisters with more than 8 years difference in their lifespan. These authors proposed the existence of one or more modifying genes as the explanation for these clinical differences, which may be contributing to the unusual presentation in our current case. Although it will be challenging to identify genetic modifiers for such a rare disease, the milder course in our patient, as well as those described by Jarvela and colleagues, does raise the possibility of a therapeutic target that could meaningfully extend the lives of patients with CLN3-associated NCL in the future.

COMPETING INTERESTS

The authors have no duality or conflicts of interest to declare.