Research Article: A Segmental Copy Number Loss of the SFMBT1 Gene Is a Genetic Risk for Shunt-Responsive, Idiopathic Normal Pressure Hydrocephalus (iNPH): A Case-Control Study

Date Published: November 18, 2016

Publisher: Public Library of Science

Author(s): Hidenori Sato, Yoshimi Takahashi, Luna Kimihira, Chifumi Iseki, Hajime Kato, Yuya Suzuki, Ryosuke Igari, Hiroyasu Sato, Shingo Koyama, Shigeki Arawaka, Toru Kawanami, Masakazu Miyajima, Naoyuki Samejima, Shinya Sato, Masahiro Kameda, Shinya Yamada, Daisuke Kita, Mitsunobu Kaijima, Isao Date, Yukihiko Sonoda, Takamasa Kayama, Nobumasa Kuwana, Hajime Arai, Takeo Kato, Francesc Palau.


Little is known about genetic risk factors for idiopathic normal pressure hydrocephalus (iNPH). We examined whether a copy number loss in intron 2 of the SFMBT1 gene could be a genetic risk for shunt-responsive, definite iNPH. Quantitative and digital PCR analyses revealed that 26.0% of shunt-responsive definite iNPH patients (n = 50) had such a genetic change, as compared with 4.2% of the healthy elderly (n = 191) (OR = 7.94, 95%CI: 2.82–23.79, p = 1.8 x 10−5) and 6.3% of patients with Parkinson’s disease (n = 32) (OR = 5.18, 95%CI: 1.1–50.8, p = 0.038). The present study demonstrates that a copy number loss within intron 2 of the SFMBT1 gene may be a genetic risk factor for shunt-responsive definite iNPH.

Partial Text

Idiopathic normal pressure hydrocephalus (iNPH) is a disease of the elderly characterized by ventricular enlargement of the brain and symptoms of gait disturbance, cognitive impairment, and urinary incontinence [1,2]. iNPH is clinically important as a treatable gait disturbance and/or preventable dementia by shunt operation [3]. It usually occurs with a sporadic onset; however, genetic factors are suggested to be involved in the pathogenesis of iNPH because of the presence of familial onset (two or more patients with NPH in a single family) of the disease, the clinical and brain MRI features of which are indistinguishable from those of sporadic iNPH [4–8]. The familial aggregation of iNPH has also been reported [7]. Previously, we made a whole-genome analysis for copy number variations (CNV) in community-dwelling Japanese elderly and found that a segmental copy number loss in the 12-kb region within intron 2 of the SFMBT1 (Scm-like with four MBT domains protein 1) gene was observed in 4 of 8 subjects (50%) with possible iNPH and AVIM (asymptomatic ventriculomegaly with features of iNPH on MRI) [9]; AVIM seems to be a pre-symptomatic state of iNPH [10,11]. On the other hand, the frequency of such a genetic change was very low in healthy controls (only one of the 110 controls: 0.9%) [9]. Even now, however, it remains undetermined whether patients with definite iNPH, whose neurological symptoms are improved after shunt operation, show such a change in the gene. The aim of the present study was to clarify whether a segmental copy number loss in intron 2 of the SFMBT1 gene would also be a genetic risk factor for shunt-responsive, definite iNPH.

The diagnosis of iNPH was made in accordance with the Japanese Guidelines for Management of iNPH [3]. The guidelines show three levels of diagnostic certainty of iNPH: “possible,” “probable,” and “definite” iNPH. The diagnosis of definite iNPH was made only when one or more symptoms of the iNPH triad were improved by shunt operation. We used 50 patients with definite iNPH (33 men and 17 women; mean age of 78.2 years, ranging from 65 to 87) diagnosed in accordance with the Japanese Guidelines; in brief, patients were 65 years of age or older, had one or more symptoms of the iNPH triad, and showed ventricular enlargement on brain MRI (Evans index > 0.3) with no cerebrospinal fluid (CSF) abnormalities upon examination. All were shunt-responsive with an improvement of one point or more (favorable outcome) on the modified Rankin Scale and/or iNPH Grading Scale with shunt operation [12,13]. There was no family history of iNPH in any patients. In addition, most of the patients showed the MRI findings of DESH (Disproportionately Enlarged Subarachnoid-space Hydrocephalus): narrowing of the subarachnoid space and cortical sulci at the high convexity of the cerebrum and widening of the Sylvian fissures [14]. Because the subarachnoid space and cortical sulci of the cerebrum become enlarged in Alzheimer’s disease due to atrophy of the cerebral cortex, the DESH findings seems to be useful to differentiate iNPH from Alzheimer’s disease [3]. Other diseases causing gait disturbance and/or cognitive impairment were excluded by extensive examinations, including neurological, neuroimaging, and CSF examinations. Eight were inpatients of our hospital, which is located in the northern part of Japan; the others (n = 42) were from 6 hospitals in the central, southern, or northernmost parts of Japan. As a control (control-1), 99 healthy Japanese elderly were used, all of whom were 70 years of age [9]. Neurological examination or brain MRI found no abnormalities in these subjects. We also employed another population of control subjects (control-2) (n = 92), aged 65 years or older, who had no apparent neurological symptoms [15]. In addition, 32 patients with Parkinson’s disease (PD) were used as a disease control. The diagnostic and inclusion criteria for PD have been described elsewhere [16].

As a first step, qPCR was used to examine DNA from 8 inpatients with shunt-responsive definite iNPH in our hospital to determine the copy number in the region of intron 2 of the SFMBT1 gene. The results showed that five of the 8 patients had a heterogeneous copy number loss in the region of intron 2 of the SFMBT1 gene. Such genetic changes were observed in 5 of the 99 healthy elderly who had no neurological symptoms or brain MRI abnormalities. Next, to test whether the results could be replicated in another population of iNPH, we collected DNA samples of 42 patients with definite iNPH from other (central, southern, and northernmost) areas of Japan and examined the copy number of the region in the SFMBT1 gene by using digital PCR. The same 99 samples were used as a healthy control. The results showed that the copy number loss in the region of the gene was found in 8 of the 42 patients (OR = 4.37, 95%CI: 1.17–18.22, p = 0.021) (Tables 1 and 2). To check the possibility that the control group of healthy elderly we used might not represent the general population of healthy people, we used another independent population of healthy subjects as control-2 (n = 92). The frequency of the copy number loss of the gene was 3.3% in control-2; thus, the above results were again replicated (OR = 6.86, 95%CI: 1.53–42.52, p = 0.004) (Tables 1 and 2). To combine and analyze all of the samples together (50 definite iNPH patients vs. 191 controls), the OR was 7.94 (95%CI: 2.82–23.79), and the p value was 1.8 x 10−5 (Tables 1 and 2, Fig 2).

The present study examined the SFMBT1 gene in patients with shunt-responsive definite iNPH, healthy elderly and PD patients, and found that a segmental copy number loss in intron 2 of the gene was more frequently observed in the iNPH patients than in the healthy elderly and PD patients. Therefore, the alteration in the gene may be a genetic risk for definite iNPH.

The present study demonstrates that a copy number loss in intron 2 of the SFMBT1 gene may be a genetic risk for shunt-responsive definite iNPH. Further studies of SFMBT1 will contribute to the elucidation of the molecular basis of iNPH.