To the Editor

Autosomal recessive primary microcephaly (MCPH) is characterized by congenital microcephaly (>2–3 standard deviations [SD] below the mean for age and gender) and intellectual disability without additional syndromic features (Alcantara & O'Driscoll, 2014). Up to now, 18 genes have been linked to MCPH, all of them involved in the neurogenesis of radial glia cells as the primary progenitor cells of the developing cerebral cortex (Jayaraman, Bae, & Walsh, 2018). Most of these genes encode centrosomal proteins involved in centriole biogenesis (WDR62, CDK5RAP2, CASC5, ASPM, CENPJ, STIL, CEP135, CEP152, SASS6) (Jayaraman et al., 2018). Others are involved in DNA replication and repair, kinetochore function, transmembrane or intracellular transport, autophagy, or cell polarity (Jayaraman et al., 2018).

Barkovic et al. provide a comprehensive review on malformations of cortical development (MCD), a large heterogenous group of defects in cerebral cortex formation, resulting from dysfunctional neurogenesis, neuronal migration, and postmigrational development (Barkovich, Guerrini, Kuzniecky, Jackson, & Dobyns, 2012). MCPH was classified as group IA, representing a reduced size of the cerebral cortex due to reduced proliferation, generally without gross morphological abnormalities (Barkovich et al., 2012). Nonetheless, cases with structural changes have been reported, for example, cortical malformations and subcortical heterotopia in WDR62 patients and periventricular heterotopia in MCPH1 (Nicholas et al., 2010; Trimborn et al., 2004; Yu et al., 2010).

Several MCD are predominantly characterized by clusters of neurons unable to migrate to their proper position in the cortex, referred to as heterotopic gray matter brain malformation (HET) (Oegema, Barkovich, Mancini, Guerrini, & Dobyns, 2019). Periventricular nodular heterotopia (PNH) is the most common subtype, formed by nodules in the wall of the lateral ventricles (Oegema et al., 2019). Recently, a new classification for the less common subcortical heterotopias (SUBH) was introduced (Oegema et al., 2019). SUBH are considered a different disease entity, defined as heterotopic gray matter located in the white matter between the lateral ventricles and the cortex (Oegema et al., 2019).

Mutations in the Centrosomal Protein 135 gene (CEP135) are a very rare cause of primary microcephaly (MCPH8, OMIM 614673), since only three families have been reported without brain‐MRI, and detailed information on brain morphology have been lacking. Here, we report a patient presenting with epilepsy as new feature in CEP135 related disease and provide the first brain‐MRI images identifying subcortical heterotopia as underlying cerebral malformation.

The female patient is the only child of healthy unrelated German parents. Microcephaly had already been diagnosed prenatally. The girl was born spontaneously at term without complications after an otherwise uneventful pregnancy. Head circumference at birth was 31 cm (1.5 cm <third percentile, −3.6 SD) and 46.5 cm (2.5 cm <third percentile, −5 SD) at age 8 years. Her early motor and speech development were delayed, but she finished regular school and her overall IQ was 79 (WISCI‐IV at age 15 years). First complex‐partial seizures started at age 10 years, which initially were well‐controlled by Oxcarbazepine. At age 18 years, she suffered from generalized tonic–clonic‐seizures requiring hospitalization. Her head circumference was 48.5 cm (3.5 cm <third percentile, −6.5 SD), height 165.3 cm (33rd percentile), and weight 69.5 kg (87th percentile) at 18 years of age. Apart from a sloping forehead and a long thin nose, no dysmorphisms were apparent. Brain‐MRI displayed marked bilateral nodular heterotopia in the peritrigonal regions, defined as group 1a according to the most recent HET classification of Oegema et al. (Figure 1) (Oegema et al., 2019). No further structural abnormalities were apparent, especially gyral pattern and myelination were unremarkable.

Consecutively, whole exome sequencing was performed as previously described (Pergande et al., 2019) (see Supplementary Material). We identified a novel homozygous frameshift mutation (c.3211A>T; p.Lys1071*) in CEP135 (NM_025009.4). The mutation was confirmed by Sanger Sequencing, and both parents are heterozygous carriers. The variant has not been reported in any database and was classified as pathogenic (PVS1, PM2, PM3) according to the ACMG classification (Richards et al., 2015). In addition, two heterozygous variants in the FAT4 gene (NM_024582.4), known to cause autosomal‐recessive syndromic periventricular nodular heterotopia were apparent (Alders et al., 2014). But as both variants in FAT4 were inherited form the healthy mother and the reported FAT4‐phenotypes including the pattern of brain malformation and pathognomonic facial features did not match our patient, they were not considered disease causing. A missense variant c.12778G>A (p.Asp4260Asn) was detected in HUWE1 (NM_031407.6), variants in HUWE1 are known to cause dominant X‐linked intellectual disability of the Turner Type (OMIM: 309590). For HUWE1 patients multiple brain‐MRI reports are available, none of them depicting cortical brain malformations, and in animal models, no structural nervous system anomalies were reported (Moortgat et al., 2018; Vandewalle et al., 2013). Thus, we considered this variant to be a less probable cause of the subcortical nodular heterotopia compared to the CEP135 variant. Nonetheless, it may act as a disease modifier for the intellectual disability depending on the degree of X‐chromosome inactivation. No further likely pathogenic variants in genes previously associated with MCPH or MCD were identified, nor were convincing novel candidate‐genes found (see Supplementary Material). Still a modifying effect of the reported secondary variants on the phenotype together with other unknown genetic disease modifiers has to be considered.

Because we have a whole‐exome and not a whole genome sequencing, it is not possible to exclude the low probability of a digenic inheritance of deep intronic variants or structural variants in a second gene modifying the phenotype (Posey et al., 2017).

Homozygous truncating mutations in CEP135 have first been identified in two Pakistani families with primary microcephaly, learning disability, speech impairment, and sloping forehead (Farooq et al., 2016; Hussain et al., 2012). In addition, a two‐year‐old boy was reported presenting with primordial dwarfism, spasticity, and developmental delay. His dysmorphic features included microcephaly, scoliosis, hypotelorism, sloping forehead, small face, and broad nose (Shaheen et al., 2019). All of these cases presented severe microcephaly with a head circumference of −10 to −13 standard deviation (SD). Our patient has milder primary microcephaly (−6.5 SD) and mild learning disabilities.

Noteworthy in our case are the novel features: epilepsy and subcortical heterotopia group 1a associated with CEP135 mutations. CEP135 is involved in centrosomal microtubule assembly by serving as a linker protein directly connecting the central hub protein HsSAS‐6 to the outer microtubule (Lin et al., 2013). It consists of an N‐terminal microtubule interacting domain, a central CENPJ interacting domain, and a C‐terminal HsSAS‐6 interacting domain (Lin et al., 2013). Our stop mutation in CEP135 presumably results in a loss of the last sixth coiled‐coil domains and is predicted to be deleterious as it covers part of the region responsible for interaction with HsSAS‐6 (MCPH14, OMIM 616402). In addition, the transcript is predicted to undergo nonsense mediated decay (NMD) which would lead to a complete loss of function. The milder microcephaly and subcortical heterotopia could represent a hypomorphic allele manifestation in our case. The premature stop codon near the C‐terminus of CEP135 might only lead to partial NMD with residual activity in contrast to the previously complete loss‐of‐function mutations causing a more severe microcephaly.

Loss of CEP135 leads to disorganized interphase and multiple and fragmented centrosomes with disorganized microtubules (Hussain et al., 2012; Lin et al., 2013; Ohta et al., 2002). This disrupts cell division but also causes disordered neuronal cell polarity and basal body formation, which is essential for neuronal migration (Jana et al., 2018). CEP135 is not the very first MCPH gene being linked to neuronal migration defects. Interestingly, mutations of WDR62 (MCPH2, OMIM 604317) are reported in patients manifesting microcephaly and malformations of cerebral cortical architecture and subcortical heterotopia (Yu et al., 2010). Our patient presents microcephaly and solid nodular HET in the region of the peritrigonal optic pathway posterior to the deep gray nuclei (group 1a). The same pattern has been described by Oegema et al. in a patient with CENPJ mutations (MCPH6, OMIM 608393), which closely interacts with CEP135 in centrosome formation (Lin et al., 2013; Oegema et al., 2019; Tang, Fu, Wu, Hsu, & Tang, 2009). In addition, SUBH 1a was associated with TUBB (CDCBM6, OMIM 615771) coding for the microtubular subunit tubulin‐beta, and KATNB1 (LIS6, OMIM 616212) which forms the microtubule‐severing protein katanin together with KATNA1 and localizes to the centrosome (Breuss et al., 2012; Mishra‐Gorur et al., 2014; Oegema et al., 2019). All these genes are involved in the asymmetrical division of neuronal progenitor cells and disruption leads to impaired neuronal proliferation and migration (Breuss et al., 2012; Insolera, Bazzi, Shao, Anderson, & Shi, 2014; Mishra‐Gorur et al., 2014).

Here, we report the first patient with a homozygous CEP135 nonsense mutation and subcortical heterotopia; thus CEP135 mutations shall be added to the genetic differential diagnosis of subcortical nodular heterotopia of the peritrigonal region (SUBH 1a). Furthermore, our findings emphasize the effect of CEP135 related centrosomal disruption on neuronal migration in human brain development.

致编辑

常染色体隐性遗传性原发性小头畸形（MCPH）的特征是先天性小头畸形（> 2-3个标准差[SD]，低于年龄和性别平均值）和智力残疾，无其他症状特征（Alcantara＆O'Driscoll，2014）。到目前为止，已有18个基因与MCPH相关联，它们都参与了放射状胶质细胞的神经发生，这些神经胶质细胞是发育中的大脑皮层的主要祖细胞（Jayaraman，Bae，＆Walsh，2018）。这些基因中的大多数编码参与着生粒的中心体蛋白（WDR62，CDK5RAP2，CASC5，ASPM，CENPJ，STIL，CEP135，CEP152，SASS6）（Jayaraman等人，2018）。其他涉及DNA复制和修复，线粒体功能，跨膜或细胞内运输，自噬或细胞极性（Jayaraman等人，2018）。

Barkovic等。提供了关于皮质发育畸形（MCD）的全面综述，该皮质畸形是由功能异常的神经发生，神经元迁移和迁移后发育引起的大脑皮质形成中的一大类异质性缺陷（Barkovich，Guerrini，Kuzniecky，Jackson和Dobyns，2012年）。MCPH被归为IA组，代表由于增殖减少导致的大脑皮层大小减小，通常没有明显的形态异常（Barkovich等，2012）。尽管如此，已经报道了具有结构变化的病例，例如WDR62患者的皮质畸形和皮质下异位症以及MCPH1患者的脑室周围异位症（Nicholas等，2010 ; Trimborn等，2004 ; Yu等，2010）。

几种MCD的主要特征是无法迁移到皮层中适当位置的神经元簇，称为异位灰质脑畸形（HET）（Oegema，Barkovich，Mancini，Guerrini和Dobyns，2019年）。脑室周围结节性异位症（PNH）是最常见的亚型，由侧脑室壁中的结节形成（Oegema et al。，2019）。最近，引入了针对较不常见的皮层下异位症（SUBH）的新分类（Oegema等，2019）。SUBH被认为是一种不同的疾病实体，定义为位于侧脑室和皮层之间的白质中的异位灰质（Oegema et al。，2019）。

中心体蛋白135基因（CEP135）的突变是原发性小头畸形的一种非常罕见的病因（MCPH8，OMIM 614673），因为仅报道了三个家族没有进行脑MRI检查，并且缺乏关于脑形态学的详细信息。在这里，我们报告了一名癫痫患者，它是CEP135相关疾病的新特征，并提供了第一个脑部MRI图像，将皮质下异位症识别为潜在的脑畸形。

该女性患者是健康无亲属德国父母的唯一孩子。小头畸形已经在产前得到诊断。该女孩在正常怀孕后足月自然出生，没有并发症。出生时的头围在31岁（1.5厘米<第三百分点，−3.6 SD）和46.5厘米（2.5 cm <第三百分点，−5 SD）。她的早期运动和言语发展受到延迟，但她完成了正规学校教育，总体智商为79（15岁时为WISCI-IV）。首次复杂的部分性癫痫发作始于10岁，最初由奥卡西平控制得很好。18岁时，她患有全身性强直-阵挛性发作，需要住院治疗。她的头围为48.5厘米（3.5厘米<3个百分点，-6.5 SD），身高165.3厘米（33个百分点），体重为69。18岁时为5公斤（百分之87）。除了前额倾斜和鼻子细长而无畸形。脑MRI显示在三角区域周围有明显的双侧结节性异位症，根据Oegema等的最新HET分类将其定义为1a组。（图1）（Oegema等，2019）。没有发现进一步的结构异常，尤其是回旋型和髓鞘形成没有明显变化。

连续地，如前所述进行了整个外显子组测序（Pergande et al。，2019）（参见补充材料）。我们在CEP135（NM_025009.4）中鉴定了新的纯合子移码突变（c.3211A> T; p.Lys1071 * ）。通过Sanger测序证实了该突变，并且两个亲本均为杂合子携带者。该变体尚未在任何数据库中报告，并且根据ACMG分类（Richards等，2015）被分类为致病性（PVS1，PM2，PM3 ）。此外，在这两个变体的杂合FAT4基因（NM_024582.4），已知引起常染色体隐性综合征脑室周围结节性异位是明显的（桤木等人，2014）。但在这两个变种FAT4被继承的形式健康的母亲和报道FAT4-表型包括脑畸形和病征性面部特征的格局并没有我们的病人匹配，他们并不认为是引起疾病。错义变异体c.12778G> A（p.Asp4260Asn）中检测HUWE1（NM_031407.6），在变体HUWE1已知导致特纳类型（OMIM：309590）的显性X染色体连锁精神发育迟滞。对于HUWE1患者，可获得多种脑部MRI报告，但均未描述皮质大脑畸形，并且在动物模型中未报告结构神经系统异常（Moortgat等人，2018年; Vandewalle等人，2013）。因此，与CEP135变体相比，我们认为该变体是皮质下结节异位症的可能性较小。但是，根据X染色体失活的程度，它可能会成为智力障碍的疾病改良剂。没有发现以前与MCPH或MCD相关的基因的其他可能的致病变异，也没有发现令人信服的新候选基因（请参阅补充材料）。仍然需要考虑报告的次级变体对表型的修饰作用以及其他未知的遗传疾病修饰剂。

因为我们有一个完整的外显子组而不是一个完整的基因组测序，所以不可能排除第二个修饰表型的基因中深度内含子变体或结构变体的双基因遗传的可能性低（Posey等人，2017）。

CEP135的纯合截断突变首先在两个巴基斯坦小家庭中被发现，这些家庭有小头畸形，学习障碍，语言障碍和前额倾斜（Farooq等人，2016； Hussain等人，2012）。此外，据报道还有一个两岁男孩表现出原始侏儒症，痉挛和发育迟缓。他的畸形特征包括小头畸形，脊柱侧弯，视力减退，额头倾斜，小脸和宽鼻子（Shaheen等人，2019）。所有这些病例均表现为严重的小头畸形，头围为-10至-13标准偏差（SD）。我们的患者患有轻度原发性小头畸形（-6.5 SD）和轻度学习障碍。

在我们的案例中值得注意的是这些新特征：与CEP135突变相关的癫痫和皮质下异位症组1a 。CEP135通过作为连接蛋白直接连接中央枢纽蛋白HsSAS-6与外部微管而参与了中心体微管组装（Lin等人，2013）。它由一个N末端微管相互作用域，一个中央CENPJ相互作用域和一个C末端HsSAS-6相互作用域组成（Lin et al。，2013）。我们在CEP135中的终止突变大概会导致最后第六个螺旋线圈结构域的丢失，并且由于覆盖了与HsSAS-6（MCPH14，OMIM 616402）交互作用的部分区域，因此预计将是有害的。另外，该转录物预计会发生无义介导的衰变（NMD），这将导致功能的完全丧失。在我们的病例中，轻度的小头畸形和皮质下异位症可能代表亚型等位基因表现。与先前完全丧失功能的突变引起更严重的小头畸形形成对比，CEP135 C末端附近的提前终止密码子可能仅导致部分NMD残留活性。

CEP135的丢失导致相间的紊乱以及带有微管紊乱的多个中心体的碎片化（Hussain等人，2012； Lin等人，2013； Ohta等人，2002）。这破坏了细胞分裂，但也导致神经元细胞极性紊乱和基底体形成，这对于神经元迁移至关重要（Jana et al。，2018）。CEP135不是第一个与神经元迁移缺陷相关的MCPH基因。有趣的是，据报道WDR62（MCPH2，OMIM 604317）的突变在表现为小头畸形和大脑皮质结构和皮质下异位症畸形的患者中发生（Yu等，2010））。我们的患者在深灰色核（1a组）后方的三角光通路中表现出小头畸形和实性结节性HET。Oegema等人已经描述了相同的模式。具有CENPJ突变（MCPH6，OMIM 608393）的患者中，该突变与CEP135在中心体形成中密切相互作用（Lin等人，2013 ; Oegema等人，2019 ; Tang，Fu，Wu，Hsu，＆Tang，2009）。此外，SUBH 1A与相关联TUBB（CDCBM6，OMIM 615771），其编码的微小管状亚基微管蛋白β和KATNB1（LIS6，OMIM 616212），其与KATNA1一起形成微管切断蛋白katanin和定位于中心体（Breuss et al。，2012 ; Mishra-Gorur et al。，2014; Oegema et al。，2019）。所有这些基因都参与神经元祖细胞的不对称分裂，破坏导致神经元增殖和迁移受损（Breuss等，2012； Insolera，Bazzi，Shao，Anderson和Shi，2014； Mishra-Gorur等，2014）。

在这里，我们报道了第一例纯合CEP135无意义突变和皮层下异位症的患者。因此，CEP135突变应添加到三角周围区域（SUBH 1a）的皮层下结节性异位症的遗传鉴别诊断中。此外，我们的发现强调了CEP135相关的中心体破坏对人脑发育中神经元迁移的影响。