Rapid increase of ‘brain-type’ transferrin in cerebrospinal fluid after shunt surgery for idiopathic normal pressure hydrocephalus: a prognosis marker for cognitive recovery

Rapid increase of ‘brain-type’ transferrin in cerebrospinal fluid after shunt surgery for... Abstract Idiopathic normal pressure hydrocephalus (iNPH) is a dementia-inducing disorder. Primary cause of iNPH is speculated to be a reduction of cerebrospinal fluid (CSF) absorption, which secondarily induces hydrocephalus, compression of brain, and reduction of CSF production. Patients are treated by surgically inserting a shunt to deliver excess CSF to the abdominal cavity. The prognosis for cognitive improvement after shunt surgery has been difficult to predict. We therefore investigated various CSF proteins, hoping to find a biomarker predictive of cognitive performance one to two years after shunt surgery. CSF proteins of 34 iNPH and 15 non-iNPH patients were analysed by Western blotting, revealing two glycan isoforms of transferrin (Tf); ‘brain-type’ Tf with N-acetylglucosaminylated glycans and ‘serum-type’ Tf with α2, 6-sialylated glycans. Brain-type Tf levels decreased in iNPH but rapidly returned to normal levels within 1–3 months after shunt surgery. This change was positively correlated with recovery from dementia, per Mini-Mental State Examination and Frontal Assessment Battery scores at 11.8 ± 7.7 months post-operation, suggesting that brain-type Tf is a prognostic marker for recovery from dementia after shunt surgery for iNPH. Histochemical staining with anti-Tf antibody and an N-acetylglucosamine-binding lectin suggests that brain-type Tf is secreted from choroid plexus, CSF-producing tissue. cerebrospinal fluid, choroid plexus, dementia, shunt surgery, transferrin Cerebrospinal fluid (CSF) circulation has been extensively studied in recent years. It has been widely accepted that CSF is produced in the choroid plexus (1, 2), and absorbed through arachnoid granules of the superior sagittal sinus after flowing from ventricles to the subarachnoid space (3). More recent studies have revealed that a part of CSF is derived from the brain parenchymal interstitial fluid (4, 5). It has also been suggested that not only arachnoid granules but also the lymphatic system is involved in CSF absorption and regulation of CSF pressure (6–8). Pathophysiology of normal pressure hydrocephalus (NPH) is associated with a reduction in craniospinal compliance and disturbance in CSF outflow-absorption (9, 10). NPH of unknown aetiology (idiopathic normal pressure hydrocephalus; iNPH) has a prevalence of about 1.1% among elderly Japanese people (11) and 2.1% among the elderly Swedes (12), indicating that the disease is of clinical and economic significance in aging societies. Patients with iNPH show cognitive impairment, gait disturbance and urinary incontinence (13–15). Although an accurate incidence of the triad (all three symptoms) is not rigorously established in multi-centre prospective cohort studies in Japan, the triad was noted in 51–59% (11, 15). iNPH patients show normal CSF pressure, but they have enlarged brain ventricles or disproportionately enlarged subarachnoid-space hydrocephalus (DESH) (16), observable by computed tomography (CT) or magnetic resonance imaging (MRI). Symptomatic patients undergo a tap test, in which 30 ml of CSF is removed to assess partial relief of symptoms such as gait disturbance and dementia; e.g. improvement is estimated by the Timed Up and Go (TUG) test for gait disturbance and Mini-Mental State Examination (MMSE) or Frontal Assessment Battery (FAB) for dementia. The improvement of cognitive function is often much less than that of gait disturbance over short intervals of observation. Patients diagnosed with iNPH are treated with ventriculo-peritoneal (VP) or lumbar-peritoneal (LP) shunt surgery. Surgery offers symptomatic improvement, but efficacy varies among patients, i.e. improvement of gait disturbance (58–90%), dementia (29–80%), and urinary incontinence (20–82.5%) (11). Multi-centre prospective cohort studies of iNPH revealed that clinical improvement rates were 63–77% at post-operative month 12 (15, 16). With some other single-centre cohort studies, clinical improvement rates were 60–80% at post-operative years 5–7 (17–19). The tap test shows high specificity (100%) and a high positive predictive value (PPV; 100%), but has low sensitivity (at most 43%) and a low negative predictive value (NPV; at most 42%) (20), indicating that many patients curable by shunt surgery might be missed. To identify these patients, various molecules including neuropeptides in the CSF have been examined as biomarkers for diagnosing iNPH and predicting efficacy of shunt surgery (11). We also investigated biomarkers of iNPH in CSF and found that ‘brain-type’ transferrin (Tf) levels could be a marker (21). CSF contains two glycan-isoforms of Tf: one is ‘brain-type’ Tf, also designated as asialo-Tf or Tf-1, and the other is serum-type Tf, also designated as sialo-Tf or Tf-2. Brain-type Tf has biantennary agalacto-complex-type glycans with bisecting β1, 4-N-acetylglucosamine (GlcNAc) and core α1, 6-fucose. The ‘brain-type’ glycan was also carried by glycoproteins such as prostaglandin D synthase (PGDS) (22) and β-secretase (or beta-site amyloid precursor protein [APP] cleaving enzyme 1: BACE1) (23), suggesting that brain-type glycan would be a ‘tag’ for brain-derived glycoproteins. We found that anti-Tf antibody strongly stained choroid plexus, a CSF producing tissue, suggesting that the tissue produces Tf carrying the brain-type glycan. In contrast, serum-type Tf has a α2, 6-sialylated biantennary complex-type glycans identical to other glycan structures in serum, suggesting that serum-type Tf is derived from blood. It was noted that brain-type Tf, but not serum-type, decreased in iNPH, indicating that both isoforms showed different metabolism in CSF, and brain-type Tf is a diagnostic marker for iNPH (21). Its sensitivity, specificity, PPV, and NPV were 75.9%, 73.7%, 81.5%, and 66.7%, respectively (21). Alzheimer disease (AD) patients also show dementia and ventriculomegaly, and need to be distinguished from iNPH patients. CSF biomarkers for AD have been established based on the pathology of the disease, which is characterized by extracellular deposits of amyloid β (Aβ) peptides and intracellular accumulation of hyperphosphorylated-tau (p-tau) in the brain (24, 25). In the CSF of AD, the ratio of p-tau/Aβ42 (42-amino-acid form of Aβ) is diagnostic: p-tau increases whereas Aβ42 decreases, thus increasing the ratio in AD. Neurotoxic Aβ is generated from APP by sequential proteolytic cleavage by β-secretase and γ-secretase (26). Alternatively, APP is cleaved at the α-site within the Aβ sequence by α-secretases, a disintegrin and metalloprotease (ADAM) family proteases. Cleavage of APP at the α-site and β-site produces N-terminal soluble fragments of APP referred to as sAPPα and sAPPβ, respectively. APP has three kinds of alternatively spliced mRNA variants (25), APP695, APP751 and APP770 (27), of which APP695 is predominantly expressed in neurons (28). We previously found that brain vascular endothelial cells express APP770 and secrete sAPP770 into CSF. Levels of sAPP (sAPPα + sAPPβ) in CSF were elevated in AD whereas those of sAPP770 were not (29). Miyajima et al. reported that sAPPα, sAPPβ, sAPP and p-tau differentiated iNPH from AD, and that sAPPα and p-tau predicted cognitive improvement after shunt surgery (9). Nakajima et al. also reported that ratios of Aβ38/Aβ42 and Aβ42/p-tau in addition to p-tau levels predicted long-term improvement of cognitive function (30). Jeppsson et al. reported, however, that AD associated-biomarkers including Aβ and sAPP isoforms were not correlated with cognitive improvement after the surgery (31). Thus, the diagnostic value of these markers, especially in predicting post-operative cognitive function, is still controversial. In the present study we analyse levels of brain-type Tf, sAPP and sAPP770 after shunt surgery to examine whether these bio-markers are correlated with a long-term outcome of cognitive functions. Materials and Methods Ethics statement This study, including the process of securing informed consent, was approved by the Ethics Committees of Fukushima Medical University (approvals 2466 and 2478) and Juntendo University (approval 20039), which are guided by local policy, national laws, and the World Medical Association Declaration of Helsinki. Patients and CSF samples The patients were recruited at Juntendo University Hospital between 2008 and 2012. The iNPH patients were diagnosed according to clinical guidelines issued by the Japanese Society of Normal Pressure Hydrocephalus (32). Briefly, iNPH was suspected based on the clinical symptoms (gait disturbance, dementia and urinary incontinence) and ventriculomegaly, which was observed by CT or MRI. The frontal horn ratio (Evans index), which is defined as the maximal frontal horn ventricular width divided by the transverse inner diameter of the skull, signifies ventriculomegaly if it is > 0.3. The patients suspected of iNPH underwent a continuous CSF drainage test or a tap test. In the drainage test, CSF was withdrawn at a rate of 300–500 ml/day for 6 days and improvements of clinical symptoms were assessed. The tap test is a temporal removal of 30 ml of CSF by lumbar puncture. Tap test-negative (subtle improvement of symptoms) patients were classified as non-iNPH or control. Tap test-positive patients were treated by shunt surgery, either VP or LP shunt, using a Codman Hakim programmable valve with Siphonguard or Medtronic Strata system. CSF samples were withdrawn from the shunt valve and centrifuged to remove cells and debris, aliquoted, and stored in polypropylene tubes at −80°C. Patients showing improvement of their symptoms after surgery were definitively diagnosed with iNPH. The study identified 34 patients who fit the definitive criteria for iNPH, including 24 males and 10 females (74.6 ± 5.6-year-old). The non-iNPH control group consisted of 15 patients, including 10 males and 5 females (74.9 ± 6.2-year-old). Ten Alzheimer’s disease patients (5 men and 5 women) were excluded, per diagnostic criteria for AD by the NINCDS-ADRDA Work Group (33). Our understanding of Tf glycan isoforms advanced through an investigation of patients with hydranencephaly, a congenital abnormality characterized by substantial or complete absence of the cerebral hemispheres. Two female hydranencephaly patients were diagnosed at Fukushima Medical University in 2013 and 2015. Case 1 was 2 months old girl and Case 2 was 6 years old girl. Both cases were treated by VP shunt surgery for macrocephaly (Fig. 1). Fig. 1 View largeDownload slide A sagittal image of hydranencephaly brain on CT and protein analysis on CSF-like fluid of hydranencephaly. A sagittal image of hydranencephaly brain on CT (A). CSF-like fluid (0.5 μl) of hydranencephaly is analysed by SDS-PAGE followed by silver staining (B) or immunoblotting using an anti-Tf antibody (C). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Those of lipocalin-type prostaglandin D synthetase (L-PGDS) and transthyretin (TTR) are indicated with arrows. It is notable that CSF for immunoblotting was not treated with 2-mercaptoethanol, resulting in good separation of Tf isoforms (C). Fig. 1 View largeDownload slide A sagittal image of hydranencephaly brain on CT and protein analysis on CSF-like fluid of hydranencephaly. A sagittal image of hydranencephaly brain on CT (A). CSF-like fluid (0.5 μl) of hydranencephaly is analysed by SDS-PAGE followed by silver staining (B) or immunoblotting using an anti-Tf antibody (C). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Those of lipocalin-type prostaglandin D synthetase (L-PGDS) and transthyretin (TTR) are indicated with arrows. It is notable that CSF for immunoblotting was not treated with 2-mercaptoethanol, resulting in good separation of Tf isoforms (C). Histochemistry Formalin-fixed paraffin-embedded brain tissue containing the choroid plexus was sectioned at 5 μm on a microtome and mounted on glass slides. Deparaffinized sections were reacted with a rabbit anti-Tf antibody (A0061, Dako Ltd.) at room temperature, followed by staining with a goat anti-rabbit IgG conjugated with Alexa Fluor 488 (A11034, Thermo Fisher Scientific). A biotinylated recombinant PVL was provided by Medical and Biological Laboratories Co., Ltd. (Nagoya, Japan). The sections were reacted with a biotinylated Psathyrella velutina lectin (PVL) at room temperature, followed by staining with Cy3-streptoavidin (PA43001, GE Health Science). Sambucus sieboldiana agglutinin (SSA) lectin was provided by J-Oil Mills, Kanagawa, Japan. Analysis of CSF proteins For SDS polyacrylamide gel electrophoresis, the drainage fluid from hydranencephaly was dissolved in Laemmli sample buffer, boiled for 3 min and loaded on a gradient gel (5–20%) (SuperSep™ Ace; Wako Pure Chemical Industries, Osaka, Japan). After SDS-PAGE, protein bands were visualized with a Silver Stain II kit Wako (Wako Pure Chemical Industries). Tf isoforms, ‘brain-type’ and serum-type Tf, were analysed by immunoblotting. Briefly, CSF (0.5 μl) was dissolved in Laemmli sample buffer without 2-mercaptoethanol, boiled for 3 min and loaded on SDS-polyacrylamide gels. After SDS-PAGE, the proteins were transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA). The membrane was blocked in 3% skim milk, incubated sequentially with an anti-Tf antibody (Bethyl Laboratories, Montgomery, TX) and a horseradish peroxidase-labeled anti-goat IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) and developed using a SuperSignal West Dura Chemiluminescence Substrate Kit (Pierce Biotechnology, Rockford, IL). Signal intensities were quantified by chromato-scanning with a CS Analyzer (ATTO, Tokyo, Japan). Each western blot included 2–10 μg of ‘brain-type’ Tf to establish a calibration curve. Standard ‘brain-type’ Tf was purified from CSF according to Futakawa et al. (21). Antibodies against Lipocalin-type prostaglandin D synthase (L-PGDS) (PA1-46023, Thermo Fisher Scientific) and transthyretin (TTR) (ab9015, abcom) were used for detecting the antigens on a blot. Soluble forms of APP695 and APP770, sAPP and sAPP770, were quantified by a sandwich Enzyme-Linked ImmunoSorbent Assay (ELISA), a Human sAPP Total Assay Kit and APP770 Assay Kit, respectively (Immuno-Biological Laboratories Co., Ltd., Gunma, Japan). According to manufacturer’s instructions, CSF at 1:16 dilution was analysed in triplicate by ELISA. Evaluation of clinical grading scales Based on the Clinical Guidelines for iNPH (32, 34), we used neurological tests such as MMSE for dementia (0–30), FAB for frontal lobe function (0–18), and modified Rankin Scale (mRS) for neurological disturbance (0–6). We evaluated each clinical grading scale pre- and post-operatively (11.8 ± 7.7 months after surgery), and analysed its correlation with concentrations of ‘brain-type’ Tf or sAPP. Statistical analysis Data were assessed as parametric or non-parametric by either the Kolmogorov-Smirnov or Shapiro–Wilk method. We described parametric data as mean ± SD. Parametric data were analysed by the Student t-test and Welch test, and assessed for multiple comparisons using Dunnett’s test. We checked correlations between ‘brain-type’ Tf and clinical grading scales using Pearson correlation coefficients. We did not perform either sample size calculation or outlier detection. For each comparison, P-value < 0.05 was considered as statistically significant. Statistical analyses were performed with StatMate V for Win&Mac Hybrid software (ATMS, Co., Ltd., Tokyo). Results Transferrin in CSF-like fluid in hydranencephaly We previously reported that CSF contains two glycan-isoforms of Tf: one is ‘brain-type’ Tf and the other is serum-type. Serum-type Tf has α2, 6-sialic acid-terminated N-glycans which are identical to those of serum Tf, suggesting that serum-type Tf is derived from blood. Brain-type Tf has β1, 4-N-acetylglucosamine (GlcNAc)-terminated N-glycans with bisecting GlcNAc and core α1, 6-fucose, which are observed with brain-derived proteins such as PGDS (22) and β-secretase (or BACE1) (23), suggesting that the brain-type Tf is produced in the brain. To examine production of brain-type Tf in the brain or cerebrum, we analysed CSF or CSF-like fluid of hydranencephaly, malformation of the cerebrum, in which the cerebral hemispheres are absent due to bilateral occlusion of internal carotid arteries, whereas the midbrain, cerebellum, and other brain structures nourished by vertebral artery may remain (35, 36). On CT analysis, a hydranencephaly patient (Case 2) shows poorly developed cerebrum and intracranial space appears to be replaced with CSF-like fluid (Fig. 1A). The fluid and control CSF were analysed by SDS-PAGE followed by silver staining (Fig. 1B). Arrows indicate migration positions of purified serum proteins such as Tf (ca. 75 kDa), albumin (ca. 66 kDa), immunoglobulin heavy chain (ca. 48 kDa), and light chains (ca. 24 kDa). Bands with similar positions are detected in CSF of control and hydranencephaly patients. CSF additionally contains 22 and 14 kDa bands. These two bands were reacted with antibodies against L-PGDS and transthyretin (TTR), respectively, on immunoblotting (data not shown). The CSF band patterns of major proteins in the hydranencephaly patients are not similar to those in control, but the band intensities are more intense in the hydranecephaly patients than in control. On an immunoblot using anti-Tf antibody, brain-type Tf is markedly decreased in Case 1 and hardly detected in Case 2 (Fig. 1C), suggesting that brain-type Tf is mainly produced in the cerebrum. We examined binding specificity of lectins for detecting glycans of Tf isoforms. Tf isoforms in CSF were purified by immunoaffinity. The purified Tf fraction contains brain- and serum-type Tf (Fig. 2A and B). PVL lectin, a binder to GlcNAc (37), detects brain-type Tf but not serum-type in purified Tf fraction and crude CSF (Fig. 2C and D). The lectin also detects several glycoproteins in crude CSF (Fig. 2C). A major glycoprotein appears in immunoglobulin heavy chains, which are reported to have heterogenous glycans such as GlcNAc-, galactose-, and sialicacid-terminated N-glycans in the Fc portion (38). In contrast, SSA lectin, a binder to α2, 6sialic acid, detects serum-type Tf but not brain-type in purified Tf fraction and crude CSF. SSA also detected other glycoproteins in crude CSF. Fig. 2 View largeDownload slide Immunoblot and lectin blot analysis for purified Tf and crude CSF. Tf was purified from CSF by immunoaffinity column chromatography. Purified Tf (20 ng) and crude CSF (1 μl) were dissolved in laemmli sample buffer containing 2-mercaptoethanol, boiled for 3 min, and loaded on a SDS-polyacrylamide gel. The bands are visualized with a silver staining kit (A). Blots are probed with an anti-Tf antibody (B), recombinant PVL (C) or SSA (D). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Fig. 2 View largeDownload slide Immunoblot and lectin blot analysis for purified Tf and crude CSF. Tf was purified from CSF by immunoaffinity column chromatography. Purified Tf (20 ng) and crude CSF (1 μl) were dissolved in laemmli sample buffer containing 2-mercaptoethanol, boiled for 3 min, and loaded on a SDS-polyacrylamide gel. The bands are visualized with a silver staining kit (A). Blots are probed with an anti-Tf antibody (B), recombinant PVL (C) or SSA (D). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Histochemical analysis on choroid plexus We previously found that choroid plexus epithelial cells, which produce CSF, were strongly stained by anti-Tf antibody (21). To examine Tf isoforms produced by the choroid plexus, the tissue was stained with PVL or SSA lectin. First, we co-stained the choroid plexus with PVL lectin and Tf-antibody. Tf-antibody stains the choroid plexus epithelial cells, some parts of interstitial tissues, and blood clots in the capillary (Fig. 3A and B; left panels). High magnification shows preferential staining of basolateral portions of the epithelial cells with minimal staining of apical portions (Fig. 3A and B; right panels). PVL lectin also stains the basolateral portions of the epithelial cells, partially overlapping the Tf-antibody stain (Fig. 3C and D; left panels, black arrowhead). Other areas are scarcely stained with PVL lectin, suggesting that GlcNAc-terminated glycoproteins are localized in the epithelial cells. Some PVL lectin signals do not overlap those of anti-Tf antibody (Fig. 3C and D; right panels, black arrowhead), suggesting the presence of GlcNAc-terminated glycoproteins other than Tf. Next, we co-stained the choroid plexus with SSA lectin and anti-Tf antibody. SSA strongly stains parts of interstitial tissue and blood clots; the lectin signals substantially overlap strong signals of anti-Tf antibody (Fig. 3E–H; left panels). In contrast, the majority of lectin staining on basolateral portions of the epithelial cells do not overlap those of Tf antibody (Fig. 3E–H; right panels, black arrowhead), suggesting that the basolateral Tf signals are not due to the presence of serum Tf but possibly brain-type Tf. The epithelial cells are slightly stained with SSA, but the signals did not overlap those of Tf antibody. The cells may contain α2, 6-sialylated glycoproteins other than Tf. Fig. 3 View largeDownload slide Histochemical analysis of the choroid plexus. Illustration (A) and (E) are schemas of the choroid plexus: choroid plexus epithelial cells (epi), interstitial tissue (int), blood capillary (capi) and lateral ventricle (LV). The choroid plexus tissue is stained with an anti-Tf antibody (B, F), PVL lectin (C) or SSA lectin (G). The signals of anti-Tf antibody are merged with those of PVL (D) or SSA lectin (H). In each set of photograph, the high magnification of the left panel is shown in the right panel. Fig. 3 View largeDownload slide Histochemical analysis of the choroid plexus. Illustration (A) and (E) are schemas of the choroid plexus: choroid plexus epithelial cells (epi), interstitial tissue (int), blood capillary (capi) and lateral ventricle (LV). The choroid plexus tissue is stained with an anti-Tf antibody (B, F), PVL lectin (C) or SSA lectin (G). The signals of anti-Tf antibody are merged with those of PVL (D) or SSA lectin (H). In each set of photograph, the high magnification of the left panel is shown in the right panel. Changes of CSF biomarkers before and after CSF drainage and shunt surgery Previously we reported that brain-type Tf levels were decreased in the CSF of iNPH patients (21). Brain-type and serum-type Tf were estimated by Western blotting, and the ratio of serum-type/brain-type Tf was used as a diagnostic marker to minimize blot-to-blot variation of Tf signals. In the present study, we utilize only brain-type Tf as a marker for continuous drainage or post-operative specimens and do not use the ratio because accurate estimation of serum-type Tf is difficult due to surgical contamination of blood Tf, which is indistinguishable from serum-type Tf. Brain-type Tf was quantitatively estimated by Western blotting; the signals of standard ‘brain-type’ Tf was linear at least in a range of 2–10 ng (Fig. 4A and B). Each Western blot included 2–10 ng of standard and amount of ‘brain-type’ Tf was calculated. Standard ‘brain-type’ Tf was purified from CSF according to Futakawa et al. Brain-type Tf levels were analysed in CSF during continuous lumbar drainage to evaluate effect of CSF withdrawal. The levels increase during post-drainage days 1–3 (126 ± 11%, not significant) and 4–6 (143 ± 17%, P < 0.01) (Fig. 4C), indicating that the marker production begin to increase within a week after the drainage. The long-term recovery of brain-type Tf was analysed after shunt surgery. The levels are markedly increased during post-operative months 1–3 and then gradually decline during post-operative months 6–24 (Fig. 5). Before surgery, sAPP levels of iNPH were significantly lower than non-iNPH; 450.3 ± 187.6 ng/ml (n = 34) vs. 677.2 ± 430.5 ng/ml (n = 15) (P < 0.05). After shunt surgery, sAPP levels are significantly increased at 6 months and then gradually decline (Fig. 5). In contrast, we detect subtle changes in levels of sAPP770, which is derived from brain vascular endothelial cells. The results suggest that brain-type Tf and APP levels are rapidly normalized after shunt surgery, but with different time courses. Fig. 4 View largeDownload slide ‘Brain-type’ Tf levels during continuous lumbar CSF drainage for iNPH patients. Brain-type Tf in CSF (0.5 μl) is estimated by western blotting (A). Signals of purified standard ‘brain-type’ CSF (2–10 ng) give a linear calibration curve (B). CSF is continuously withdrawn from iNPH patients (n = 4) for 6 days. Brain-type Tf levels before drainage and post-drainage days 1–3 or 4–6 are indicated with box plots (C). Significant difference is shown by Dunnett’s multiple comparison analysis; n.s., not significant. Fig. 4 View largeDownload slide ‘Brain-type’ Tf levels during continuous lumbar CSF drainage for iNPH patients. Brain-type Tf in CSF (0.5 μl) is estimated by western blotting (A). Signals of purified standard ‘brain-type’ CSF (2–10 ng) give a linear calibration curve (B). CSF is continuously withdrawn from iNPH patients (n = 4) for 6 days. Brain-type Tf levels before drainage and post-drainage days 1–3 or 4–6 are indicated with box plots (C). Significant difference is shown by Dunnett’s multiple comparison analysis; n.s., not significant. Fig. 5 View largeDownload slide Changes of CSF biomarker levels after shunt surgery. CSF biomarker levels are quantified at pre-surgery (indicated as 100% at 0 month) and post-surgery (mean ± SE); brain-type Tf (closed circle), sAPP (grey circle), sAPP770 (open square). Significant difference is shown by Dunnett’s multiple comparison analysis and indicated with asterisks; P < 0.05 (*), P < 0.01 (**). Fig. 5 View largeDownload slide Changes of CSF biomarker levels after shunt surgery. CSF biomarker levels are quantified at pre-surgery (indicated as 100% at 0 month) and post-surgery (mean ± SE); brain-type Tf (closed circle), sAPP (grey circle), sAPP770 (open square). Significant difference is shown by Dunnett’s multiple comparison analysis and indicated with asterisks; P < 0.05 (*), P < 0.01 (**). Correlation between brain-type Tf levels and dementia scores after shunt surgery All neurological clinical grading scales were improved post-operatively, after 11.8 ± 7.7 months: pre- versus post-operative scores are as follow; MMSE, 21.4 ± 5.3 vs. 24.3 ± 4.7 (P = 0.022); FAB, 10.5 ± 3.3 vs. 12.3 ± 3.3 (P = 0.036); and mRS, 2.9 ± 0.8 vs. 2.1 ± 1.1 (P = 0.002), respectively. We analysed correlations between the cognitive scales and the biomarkers in CSF at various time points. Before shunt surgery, the levels of brain-type Tf show low (r = 0.351, P = 0.042) and moderate (r = 0.527, P = 0.002) correlation with MMSE and FAB scores, respectively (Fig. 6A and B). After post-operative month 3, increase of brain-type Tf strongly correlates with scores of MMSE (r = 0.697, P = 0.037) and FAB (r = 0.727, P = 0.041) (Fig. 6C and D; the left panel). The increase of brain-type Tf at post-operative month 12 is moderately correlated with MMSE (r = 0.549, P = 0.022) but not with FAB (r = 0.373, P = 0.154) (Fig. 6C and D; the right panel). The scores of mRS were not correlated with brain-type Tf levels before and after shunt surgery (data not shown). The results suggest that brain-type Tf levels at post-operative month 3 (and 12) would predict recovery from dementia of iNPH patients. Fig. 6 View largeDownload slide Correlation between brain-type Tf and cognitive scales. Correlation between ‘brain-type’ Tf levels and cognitive scales, MMSE (A) or FAB (B), are indicated. ‘Brain-type’ Tf levels are measured before surgery (A, B) and % increases at post-operative months 3 and 12 (C, D) are examined. Post-operative cognitive scales were examined at 11.8 ± 7.7 months after shunt surgery. Fig. 6 View largeDownload slide Correlation between brain-type Tf and cognitive scales. Correlation between ‘brain-type’ Tf levels and cognitive scales, MMSE (A) or FAB (B), are indicated. ‘Brain-type’ Tf levels are measured before surgery (A, B) and % increases at post-operative months 3 and 12 (C, D) are examined. Post-operative cognitive scales were examined at 11.8 ± 7.7 months after shunt surgery. Discussion In drainage fluid of hydranencephaly patients, major bands appear to be Tf, albumin, immunoglobulin heavy chain and light chain, which possibly diffuse into CSF from blood. Indeed, about 80% of CSF proteins are derived from blood (39). Additional bands at the positions of L-PGDS and TTR are more intense in CSF of hydranencephaly patients than the control. The observation is consistent with the fact that mRNAs of L-PGDS and TTR are highly expressed in leptomeninges and hypothalamus as well as cerebellum (40, 41), which are preserved in hydranencephaly patients. Brain-type Tf levels were markedly decreased in the drainage fluid of hydranencephaly patients, suggesting that brain-type Tf is derived from cerebrum. Histochemical analysis revealed that anti-Tf antibody stained choroid plexus tissue including choroid plexus epithelial cells, interstitial tissue, and blood clots in the capillaries, indicating ubiquitous presence of Tf in the tissue. The basolateral portions of epithelial cells were stained by the antibody, whereas apical portions were not. Tf in the apical portion may be rapidly secreted from the cells and its steady state levels may be very low, if any. The basolateral portions of epithelial cells were also stained with PVL lectin, a binder for GlcNAc-terminated glycans, and the staining was partially merged with anti-Tf antibody staining, suggesting the presence of brain-type Tf in choroid plexus epithelial cells. Some PVL signals are not merged with the antibody staining, suggesting the epithelial cells produced glycoproteins having GlcNAc-terminated glycans other than brain-type Tf. Based on these results we hypothesized that brain-type Tf is, at least partly, produced in choroid plexus epithelial cells and secreted together with CSF. To confirm the hypothesis, purification of Tf from choroid plexus and its glycan analysis will be needed in the future. Brain-type Tf levels are reduced in iNPH, in which decrease of CSF production is suspected (42), and increase after shunt surgery. It is notable that the marker levels begin to increase within a week by continuous drainage. In addition, brain-type Tf significantly increased during post-operative months 1–12, showing a peak at 3 months. The increment may involve several mechanisms such as increase of brain-type Tf secretion and/or reduced total volume of CSF. The reduced CSF volume could result in ‘apparent’increase of concentration of brain-type Tf. In this case, concentration of other proteins would be also increase. At post-operative month 3, concentration of brain-type Tf, sAPP and sAPP770 increased to different degrees: 220%, 145% and 110% respectively, suggesting that reduced volume of CSF is not the only reason for increase of brain-type Tf. Each protein has different origin, i.e. brain-type Tf from choroid plexus, sAPP from neurons (26), and sAPP770 from endothelial cells (29). Their secretion or excretion would be under different controlling mechanisms. CSF accumulates in the lateral ventricles of iNPH patients, compressing the brain parenchyma and impairing cognitive functions. The shunt surgery drains excess CSF, relieves the compression, and normalizes cognitive functions, although recovery from dementia often needs long-term intervention over a year. We speculated that dementia scales would correlate with levels of brain-type Tf. Indeed, levels of MMSE and FAB scales were correlated with those of brain-type Tf even before shunt surgery. After the surgery, dementia scores measured post-operatively at 11.8 ± 7.7 months were found to correlate with the increase of brain-type Tf levels at 3 (or 12) months, suggesting that the marker predicted recovery from dementia after shunt surgery. Soluble APP also significantly increased during post-operative months 6–12, but sAPP levels did not predict recovery from dementia. Consistent with our result, Jeppsson et al. reported that sAPP isoforms were not correlated with cognitive improvement after the surgery (31). Miyajima et al. reported, however, that MMSE scores at post-operative 6 months were correlated with sAPPα and phosphorylated-tau (9). Further analysis is required for establishing prognostic ability of sAPP isoforms. In conclusion, we demonstrate that brain-type Tf appears to emanate from the choroid plexus and would be an indicator for CSF production that predicts recovery from dementia after shunt surgery for iNPH patients. This favourable prediction would encourage patients and caregivers to persevere in their rehabilitation efforts. A limitation of the present study was to be a single-centre investigation with a limited number of patients. Acknowledgements We sincerely acknowledge our colleagues in the Hashimoto laboratories for valuable discussions. We acknowledge the Brain Bank for Aging Research, Tokyo Metropolitan Geriatric Hospital and Institute of Gerontology for providing brain tissues. Funding This work was partly supported by Japan Agency for Medical Research and Development (AMED) [grant numbers 16hm0102042h0001 and 17hm0102042h0002]; the Japan Science and Technology Agency [grant numbers AS221Z00232F, AS231Z01053, 241FT0255 and 149]; the Ministry of Education, Culture, Sports, Science, and Technology of Japan [a Grant-in-Aid for Scientific Research on Innovative Areas, grant number 23110002 (Deciphering sugar chain-based signals regulating integrative neuronal functions), a Grant-in-Aid for Scientific Research, grant number 23590367]; Fukushima Medical University [Grant for Strategic Research Promotion of FMU]; The General Insurance Association of Japan [a medical research grant on traffic accident]. Conflict of Interest None declared. References 1 Cushing H. ( 1914) Studies on the cerebro-spinal fluid: I. Introduction. J. Med. Res.  31, 1– 19 Google Scholar PubMed  2 Cutler R.W., Page L., Galicich J., Watters G.V. ( 1968) Formation and absorption of cerebrospinal fluid in man. Brain  91, 707– 720 Google Scholar CrossRef Search ADS PubMed  3 Cushing H. ( 1969) Selected Papers on Neurosurgery . Yale University Press, New Haven and London 4 Carare R.O., Hawkes C.A., Weller R.O. ( 2014) Afferent and efferent immunological pathways of the brain. Anatomy, function and failure. Brain Behav. Immun . 36, 9– 14 Google Scholar CrossRef Search ADS PubMed  5 Owler B.K., Pitham T., Wang D. ( 2010) Aquaporins: relevance to cerebrospinal fluid physiology and therapeutic potential in hydrocephalus. Cerebrospinal Fluid Res . 7, 15 Google Scholar CrossRef Search ADS PubMed  6 Brierley J.B. ( 1950) The penetration of particulate matter from the cerebrospinal fluid into the spinal ganglia, peripheral nerves, and perivascular spaces of the central nervous system. J. Neurol. Neurosurg. Psychiatry  13, 203– 215 Google Scholar CrossRef Search ADS PubMed  7 Miura M., Kato S., Lüdinghausen M. V. ( 1998) Lymphatic drainage of the cerebrospinal fluid from monkey spinal meninges with special reference to the distribution of the epidural lymphatics. Arch. Histol. Cytol . 61, 277– 286 Google Scholar CrossRef Search ADS PubMed  8 Miura M.U.T., Takahashi A. ( 2015) Latest knowledge of cerebrospinal fluid: production, circulation and absorption with specialreference to the collateral absorptive routes via the epidural lymphatic system. Spine Spinal Cord  28, 694– 703 9 Miyajima M., Nakajima M., Ogino I., Miyata H., Motoi Y., Arai H. ( 2013) Soluble amyloid precursor protein alpha in the cerebrospinal fluid as a diagnostic and prognostic biomarker for idiopathic normal pressure hydrocephalus. Eur. J. Neurol.  20, 236– 242 Google Scholar CrossRef Search ADS PubMed  10 Hamlat A., Adn M., Sid-ahmed S., Askar B., Pasqualini E. ( 2006) Theoretical considerations on the pathophysiology of normal pressure hydrocephalus (NPH) and NPH-related dementia. Med. Hypotheses  67, 115– 123 Google Scholar CrossRef Search ADS PubMed  11 Mori E., Ishikawa M., Kato T., Kazui H., Miyake H., Miyajima M., Nakajima M., Hashimoto M., Kuriyama N., Tokuda T., Ishii K., Kaijima M., Hirata Y., Saito M., Arai H., Japanese S., of Normal Pressure H. ( 2012) Guidelines for management of idiopathic normal pressure hydrocephalus: second edition. Neurol. Med. Chir (Tokyo)  52, 775– 809 Google Scholar CrossRef Search ADS PubMed  12 Jaraj D., Rabiei K., Marlow T., Jensen C., Skoog I., Wikkelso C. ( 2014) Prevalence of idiopathic normal-pressure hydrocephalus. Neurology  82, 1449– 1454 Google Scholar CrossRef Search ADS PubMed  13 Gallia G.L., Rigamonti D., Williams M.A. ( 2006) The diagnosis and treatment of idiopathic normal pressure hydrocephalus. Nat. Clin. Pract. Neurol.  2, 375– 381 Google Scholar CrossRef Search ADS PubMed  14 Relkin N., Marmarou A., Klinge P., Bergsneider M., Black P.M. ( 2005) Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery  57, S2-4– 16. Discussion ii–v Google Scholar CrossRef Search ADS   15 Kazui H., Miyajima M., Mori E., Ishikawa M., Investigators S. ( 2015) Lumboperitoneal shunt surgery for idiopathic normal pressure hydrocephalus (SINPHONI-2): an open-label randomised trial. Lancet Neurol . 14, 585– 594 Google Scholar CrossRef Search ADS PubMed  16 Hashimoto M., Ishikawa M., Mori E., Kuwana N. and Study of I.o.n.i. ( 2010) Diagnosis of idiopathic normal pressure hydrocephalus is supported by MRI-based scheme: a prospective cohort study. Cerebrospinal Fluid Res . 7, 18 Google Scholar CrossRef Search ADS PubMed  17 Meier U., Lemcke J. ( 2006) Clinical outcome of patients with idiopathic normal pressure hydrocephalus three years after shunt implantation. Acta Neurochir. Suppl . 96, 377– 380 Google Scholar CrossRef Search ADS PubMed  18 Pujari S., Kharkar S., Metellus P., Shuck J., Williams M.A., Rigamonti D. ( 2008) Normal pressure hydrocephalus: long-term outcome after shunt surgery. J. Neurol. Neurosurg. Psychiatry  79, 1282– 1286 Google Scholar CrossRef Search ADS PubMed  19 Golz L., Ruppert F.H., Meier U., Lemcke J. ( 2014) Outcome of modern shunt therapy in patients with idiopathic normal pressure hydrocephalus 6 years postoperatively. J. Neurosurg . 121, 771– 775 Google Scholar CrossRef Search ADS PubMed  20 Walchenbach R., Geiger E., Thomeer R.T., Vanneste J.A. ( 2002) The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J. Neurol. Neurosurg. Psychiatry  72, 503– 506 Google Scholar PubMed  21 Futakawa S., Nara K., Miyajima M., Kuno A., Ito H., Kaji H., Shirotani K., Honda T., Tohyama Y., Hoshi K., Hanzawa Y., Kitazume S., Imamaki R., Furukawa K., Tasaki K., Arai H., Yuasa T., Abe M., Arai H., Narimatsu H., Hashimoto Y. ( 2012) A unique N-glycan on human transferrin in CSF: a possible biomarker for iNPH. Neurobiol. Aging  33, 1807– 1815 Google Scholar CrossRef Search ADS PubMed  22 Hoffmann A., Nimtz M., Wurster U., Conradt H.S. ( 2002) Carbohydrate structures of beta-trace protein from human cerebrospinal fluid: evidence for “brain-type” N-glycosylation. J. Neurochem . 63, 2185– 2196 Google Scholar CrossRef Search ADS   23 Kizuka Y., Kitazume S., Fujinawa R., Saito T., Iwata N., Saido T.C., Nakano M., Yamaguchi Y., Hashimoto Y., Staufenbiel M., Hatsuta H., Murayama S., Manya H., Endo T., Taniguchi N. ( 2015) An aberrant sugar modification of BACE1 blocks its lysosomal targeting in Alzheimer’s disease. EMBO Mol. Med . 7, 175– 189 Google Scholar CrossRef Search ADS PubMed  24 Selkoe D.J. ( 2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev.  81, 741– 766 Google Scholar CrossRef Search ADS PubMed  25 Tanzi R.E., Bertram L. ( 2005) Twenty years of the Alzheimer’s disease amyloid hypothesis: a genetic perspective. Cell  120, 545– 555 Google Scholar CrossRef Search ADS PubMed  26 De Strooper B., Annaert W., Cupers P., Saftig P., Craessaerts K., Mumm J.S., Schroeter E.H., Schrijvers V., Wolfe M.S., Ray W.J., Goate A., Kopan R. ( 1999) A presenilin-1-dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature  398, 518– 522 Google Scholar CrossRef Search ADS PubMed  27 Tomita S., Kirino Y., Suzuki T. ( 1998) Cleavage of Alzheimer’s amyloid precursor protein (APP) by secretases occurs after O-glycosylation of APP in the protein secretory pathway. Identification of intracellular compartments in which APP cleavage occurs without using toxic agents that interfere with protein metabolism. J. Biol. Chem.  273, 6277– 6284 Google Scholar CrossRef Search ADS PubMed  28 Wertkin A.M., Turner R.S., Pleasure S.J., Golde T.E., Younkin S.G., Trojanowski J.Q., Lee V.M. ( 1993) Human neurons derived from a teratocarcinoma cell line express solely the 695-amino acid amyloid precursor protein and produce intracellular beta-amyloid or A4 peptides. Proc. Natl. Acad. Sci. U.S.A.  90, 9513– 9517 Google Scholar CrossRef Search ADS PubMed  29 Kitazume S., Yoshihisa A., Yamaki T., Oikawa M., Tachida Y., Ogawa K., Imamaki R., Hagiwara Y., Kinoshita N., Takeishi Y., Furukawa K., Tomita N., Arai H., Iwata N., Saido T., Yamamoto N., Taniguchi N. ( 2012) Soluble amyloid precursor protein 770 is released from inflamed endothelial cells and activated platelets: a novel biomarker for acute coronary syndrome. J. Biol. Chem.  287, 40817– 40825 Google Scholar CrossRef Search ADS PubMed  30 Nakajima M., Miyajima M., Ogino I., Akiba C., Sugano H., Hara T., Fusegi K., Karagiozov K., Arai H. ( 2015) Cerebrospinal fluid biomarkers for prognosis of long-term cognitive treatment outcomes in patients with idiopathic normal pressure hydrocephalus. J. Neurol. Sci . 357, 88– 95 Google Scholar CrossRef Search ADS PubMed  31 Jeppsson A., Holtta M., Zetterberg H., Blennow K., Wikkelso C., Tullberg M. ( 2016) Amyloid mis-metabolism in idiopathic normal pressure hydrocephalus. Fluids Barriers CNS  13, 13 Google Scholar CrossRef Search ADS PubMed  32 Ishikawa M. and Guideline Committe for Idiopathic Normal Pressure Hydrocephalus, J.S.o.N.P.H. ( 2004) Clinical guidelines for idiopathic normal pressure hydrocephalus. Neurol. Med. Chir (Tokyo)  44, 222– 223 Google Scholar CrossRef Search ADS PubMed  33 McKhann G., Drachman D., Folstein M., Katzman R., Price D., Stadlan E.M. ( 1984) Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology  34, 939– 944 Google Scholar CrossRef Search ADS PubMed  34 Mori K. ( 2001) Management of idiopathic normal-pressure hydrocephalus: a multiinstitutional study conducted in Japan. J. Neurosurg . 95, 970– 973 Google Scholar CrossRef Search ADS PubMed  35 Greene M.F., Benacerraf B., Crawford J.M. ( 1985) Hydranencephaly: US appearance during in utero evolution. Radiology  156, 779– 780 Google Scholar CrossRef Search ADS PubMed  36 Pavone P., Pratico A.D., Vitaliti G., Ruggieri M., Rizzo R., Parano E., Pavone L., Pero G., Falsaperla R. ( 2014) Hydranencephaly: cerebral spinal fluid instead of cerebral mantles. Ital. J. Pediatr.  40, 79 Google Scholar CrossRef Search ADS PubMed  37 Kochibe N., Matta K.L. ( 1989) Purification and properties of an N-acetylglucosamine-specific lectin from Psathyrella velutina mushroom. J. Biol. Chem . 264, 173– 177 Google Scholar PubMed  38 Takahashi N., Ishii I., Ishihara H., Mori M., Tejima S., Jefferis R., Endo S., Arata Y. ( 1987) Comparative structural study of the N-linked oligosaccharides of human normal and pathological immunoglobulin G. Biochemistry  26, 1137– 1144 Google Scholar CrossRef Search ADS PubMed  39 Reiber H., Peter J.B. ( 2001) Cerebrospinal fluid analysis: disease-related data patterns and evaluation programs. J. Neurol. Sci . 184, 101– 122 Google Scholar CrossRef Search ADS PubMed  40 Blay P., Nilsson C., Hansson S., Owman C., Aldred A.R., Schreiber G. ( 1994) An in vivo study of the effect of 5-HT and sympathetic nerves on transferrin and transthyretin mRNA expression in rat choroid plexus and meninges. Brain Res . 662, 148– 154 Google Scholar CrossRef Search ADS PubMed  41 Ohinata K., Takagi K., Biyajima K., Fujiwara Y., Fukumoto S., Eguchi N., Urade Y., Asakawa A., Fujimiya M., Inui A., Yoshikawa M. ( 2008) Central prostaglandin D(2) stimulates food intake via the neuropeptide Y system in mice. FEBS Lett . 582, 679– 684 Google Scholar CrossRef Search ADS PubMed  42 Silverberg G.D., Mayo M., Saul T., Rubenstein E., McGuire D. ( 2003) Alzheimer’s disease, normal-pressure hydrocephalus, and senescent changes in CSF circulatory physiology: a hypothesis. Lancet Neurol . 2, 506– 511 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations Aβ amyloid β AD Alzheimer disease APP amyloid precursor protein CSF cerebrospinal fluid CT computed tomography DESH disproportionately enlarged subarachnoid-space hydrocephalus ELISA enzyme-linked immunosorbent assay FAB Frontal Assessment Battery GlcNAc , N-acetylglucosamine iNPH idiopathic normal pressure hydrocephalus Ig immunoglobulin LP lumbar-peritoneal L-PGDS lipocalin-type prostaglandin D synthase MMSE Mini-Mental State Examination MRI magnetic resonance imaging mRS modified Rankin scale NPV negative predictive value PPV positive predictive value p-tau hyperphosphorylated-tau PVL Psathyrella velutina sAPP soluble protein fragments of amyloid precursor protein SSA Sambucus sieboldiana agglutinin Tf transferrin TTR transthyretin TUG Timed Up and Go VP ventriculo-peritoneal © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Biochemistry Oxford University Press

Rapid increase of ‘brain-type’ transferrin in cerebrospinal fluid after shunt surgery for idiopathic normal pressure hydrocephalus: a prognosis marker for cognitive recovery

Loading next page...
 
/lp/ou_press/rapid-increase-of-brain-type-transferrin-in-cerebrospinal-fluid-after-4ZoiMUYVK1
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
ISSN
0021-924X
eISSN
1756-2651
D.O.I.
10.1093/jb/mvy043
Publisher site
See Article on Publisher Site

Abstract

Abstract Idiopathic normal pressure hydrocephalus (iNPH) is a dementia-inducing disorder. Primary cause of iNPH is speculated to be a reduction of cerebrospinal fluid (CSF) absorption, which secondarily induces hydrocephalus, compression of brain, and reduction of CSF production. Patients are treated by surgically inserting a shunt to deliver excess CSF to the abdominal cavity. The prognosis for cognitive improvement after shunt surgery has been difficult to predict. We therefore investigated various CSF proteins, hoping to find a biomarker predictive of cognitive performance one to two years after shunt surgery. CSF proteins of 34 iNPH and 15 non-iNPH patients were analysed by Western blotting, revealing two glycan isoforms of transferrin (Tf); ‘brain-type’ Tf with N-acetylglucosaminylated glycans and ‘serum-type’ Tf with α2, 6-sialylated glycans. Brain-type Tf levels decreased in iNPH but rapidly returned to normal levels within 1–3 months after shunt surgery. This change was positively correlated with recovery from dementia, per Mini-Mental State Examination and Frontal Assessment Battery scores at 11.8 ± 7.7 months post-operation, suggesting that brain-type Tf is a prognostic marker for recovery from dementia after shunt surgery for iNPH. Histochemical staining with anti-Tf antibody and an N-acetylglucosamine-binding lectin suggests that brain-type Tf is secreted from choroid plexus, CSF-producing tissue. cerebrospinal fluid, choroid plexus, dementia, shunt surgery, transferrin Cerebrospinal fluid (CSF) circulation has been extensively studied in recent years. It has been widely accepted that CSF is produced in the choroid plexus (1, 2), and absorbed through arachnoid granules of the superior sagittal sinus after flowing from ventricles to the subarachnoid space (3). More recent studies have revealed that a part of CSF is derived from the brain parenchymal interstitial fluid (4, 5). It has also been suggested that not only arachnoid granules but also the lymphatic system is involved in CSF absorption and regulation of CSF pressure (6–8). Pathophysiology of normal pressure hydrocephalus (NPH) is associated with a reduction in craniospinal compliance and disturbance in CSF outflow-absorption (9, 10). NPH of unknown aetiology (idiopathic normal pressure hydrocephalus; iNPH) has a prevalence of about 1.1% among elderly Japanese people (11) and 2.1% among the elderly Swedes (12), indicating that the disease is of clinical and economic significance in aging societies. Patients with iNPH show cognitive impairment, gait disturbance and urinary incontinence (13–15). Although an accurate incidence of the triad (all three symptoms) is not rigorously established in multi-centre prospective cohort studies in Japan, the triad was noted in 51–59% (11, 15). iNPH patients show normal CSF pressure, but they have enlarged brain ventricles or disproportionately enlarged subarachnoid-space hydrocephalus (DESH) (16), observable by computed tomography (CT) or magnetic resonance imaging (MRI). Symptomatic patients undergo a tap test, in which 30 ml of CSF is removed to assess partial relief of symptoms such as gait disturbance and dementia; e.g. improvement is estimated by the Timed Up and Go (TUG) test for gait disturbance and Mini-Mental State Examination (MMSE) or Frontal Assessment Battery (FAB) for dementia. The improvement of cognitive function is often much less than that of gait disturbance over short intervals of observation. Patients diagnosed with iNPH are treated with ventriculo-peritoneal (VP) or lumbar-peritoneal (LP) shunt surgery. Surgery offers symptomatic improvement, but efficacy varies among patients, i.e. improvement of gait disturbance (58–90%), dementia (29–80%), and urinary incontinence (20–82.5%) (11). Multi-centre prospective cohort studies of iNPH revealed that clinical improvement rates were 63–77% at post-operative month 12 (15, 16). With some other single-centre cohort studies, clinical improvement rates were 60–80% at post-operative years 5–7 (17–19). The tap test shows high specificity (100%) and a high positive predictive value (PPV; 100%), but has low sensitivity (at most 43%) and a low negative predictive value (NPV; at most 42%) (20), indicating that many patients curable by shunt surgery might be missed. To identify these patients, various molecules including neuropeptides in the CSF have been examined as biomarkers for diagnosing iNPH and predicting efficacy of shunt surgery (11). We also investigated biomarkers of iNPH in CSF and found that ‘brain-type’ transferrin (Tf) levels could be a marker (21). CSF contains two glycan-isoforms of Tf: one is ‘brain-type’ Tf, also designated as asialo-Tf or Tf-1, and the other is serum-type Tf, also designated as sialo-Tf or Tf-2. Brain-type Tf has biantennary agalacto-complex-type glycans with bisecting β1, 4-N-acetylglucosamine (GlcNAc) and core α1, 6-fucose. The ‘brain-type’ glycan was also carried by glycoproteins such as prostaglandin D synthase (PGDS) (22) and β-secretase (or beta-site amyloid precursor protein [APP] cleaving enzyme 1: BACE1) (23), suggesting that brain-type glycan would be a ‘tag’ for brain-derived glycoproteins. We found that anti-Tf antibody strongly stained choroid plexus, a CSF producing tissue, suggesting that the tissue produces Tf carrying the brain-type glycan. In contrast, serum-type Tf has a α2, 6-sialylated biantennary complex-type glycans identical to other glycan structures in serum, suggesting that serum-type Tf is derived from blood. It was noted that brain-type Tf, but not serum-type, decreased in iNPH, indicating that both isoforms showed different metabolism in CSF, and brain-type Tf is a diagnostic marker for iNPH (21). Its sensitivity, specificity, PPV, and NPV were 75.9%, 73.7%, 81.5%, and 66.7%, respectively (21). Alzheimer disease (AD) patients also show dementia and ventriculomegaly, and need to be distinguished from iNPH patients. CSF biomarkers for AD have been established based on the pathology of the disease, which is characterized by extracellular deposits of amyloid β (Aβ) peptides and intracellular accumulation of hyperphosphorylated-tau (p-tau) in the brain (24, 25). In the CSF of AD, the ratio of p-tau/Aβ42 (42-amino-acid form of Aβ) is diagnostic: p-tau increases whereas Aβ42 decreases, thus increasing the ratio in AD. Neurotoxic Aβ is generated from APP by sequential proteolytic cleavage by β-secretase and γ-secretase (26). Alternatively, APP is cleaved at the α-site within the Aβ sequence by α-secretases, a disintegrin and metalloprotease (ADAM) family proteases. Cleavage of APP at the α-site and β-site produces N-terminal soluble fragments of APP referred to as sAPPα and sAPPβ, respectively. APP has three kinds of alternatively spliced mRNA variants (25), APP695, APP751 and APP770 (27), of which APP695 is predominantly expressed in neurons (28). We previously found that brain vascular endothelial cells express APP770 and secrete sAPP770 into CSF. Levels of sAPP (sAPPα + sAPPβ) in CSF were elevated in AD whereas those of sAPP770 were not (29). Miyajima et al. reported that sAPPα, sAPPβ, sAPP and p-tau differentiated iNPH from AD, and that sAPPα and p-tau predicted cognitive improvement after shunt surgery (9). Nakajima et al. also reported that ratios of Aβ38/Aβ42 and Aβ42/p-tau in addition to p-tau levels predicted long-term improvement of cognitive function (30). Jeppsson et al. reported, however, that AD associated-biomarkers including Aβ and sAPP isoforms were not correlated with cognitive improvement after the surgery (31). Thus, the diagnostic value of these markers, especially in predicting post-operative cognitive function, is still controversial. In the present study we analyse levels of brain-type Tf, sAPP and sAPP770 after shunt surgery to examine whether these bio-markers are correlated with a long-term outcome of cognitive functions. Materials and Methods Ethics statement This study, including the process of securing informed consent, was approved by the Ethics Committees of Fukushima Medical University (approvals 2466 and 2478) and Juntendo University (approval 20039), which are guided by local policy, national laws, and the World Medical Association Declaration of Helsinki. Patients and CSF samples The patients were recruited at Juntendo University Hospital between 2008 and 2012. The iNPH patients were diagnosed according to clinical guidelines issued by the Japanese Society of Normal Pressure Hydrocephalus (32). Briefly, iNPH was suspected based on the clinical symptoms (gait disturbance, dementia and urinary incontinence) and ventriculomegaly, which was observed by CT or MRI. The frontal horn ratio (Evans index), which is defined as the maximal frontal horn ventricular width divided by the transverse inner diameter of the skull, signifies ventriculomegaly if it is > 0.3. The patients suspected of iNPH underwent a continuous CSF drainage test or a tap test. In the drainage test, CSF was withdrawn at a rate of 300–500 ml/day for 6 days and improvements of clinical symptoms were assessed. The tap test is a temporal removal of 30 ml of CSF by lumbar puncture. Tap test-negative (subtle improvement of symptoms) patients were classified as non-iNPH or control. Tap test-positive patients were treated by shunt surgery, either VP or LP shunt, using a Codman Hakim programmable valve with Siphonguard or Medtronic Strata system. CSF samples were withdrawn from the shunt valve and centrifuged to remove cells and debris, aliquoted, and stored in polypropylene tubes at −80°C. Patients showing improvement of their symptoms after surgery were definitively diagnosed with iNPH. The study identified 34 patients who fit the definitive criteria for iNPH, including 24 males and 10 females (74.6 ± 5.6-year-old). The non-iNPH control group consisted of 15 patients, including 10 males and 5 females (74.9 ± 6.2-year-old). Ten Alzheimer’s disease patients (5 men and 5 women) were excluded, per diagnostic criteria for AD by the NINCDS-ADRDA Work Group (33). Our understanding of Tf glycan isoforms advanced through an investigation of patients with hydranencephaly, a congenital abnormality characterized by substantial or complete absence of the cerebral hemispheres. Two female hydranencephaly patients were diagnosed at Fukushima Medical University in 2013 and 2015. Case 1 was 2 months old girl and Case 2 was 6 years old girl. Both cases were treated by VP shunt surgery for macrocephaly (Fig. 1). Fig. 1 View largeDownload slide A sagittal image of hydranencephaly brain on CT and protein analysis on CSF-like fluid of hydranencephaly. A sagittal image of hydranencephaly brain on CT (A). CSF-like fluid (0.5 μl) of hydranencephaly is analysed by SDS-PAGE followed by silver staining (B) or immunoblotting using an anti-Tf antibody (C). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Those of lipocalin-type prostaglandin D synthetase (L-PGDS) and transthyretin (TTR) are indicated with arrows. It is notable that CSF for immunoblotting was not treated with 2-mercaptoethanol, resulting in good separation of Tf isoforms (C). Fig. 1 View largeDownload slide A sagittal image of hydranencephaly brain on CT and protein analysis on CSF-like fluid of hydranencephaly. A sagittal image of hydranencephaly brain on CT (A). CSF-like fluid (0.5 μl) of hydranencephaly is analysed by SDS-PAGE followed by silver staining (B) or immunoblotting using an anti-Tf antibody (C). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Those of lipocalin-type prostaglandin D synthetase (L-PGDS) and transthyretin (TTR) are indicated with arrows. It is notable that CSF for immunoblotting was not treated with 2-mercaptoethanol, resulting in good separation of Tf isoforms (C). Histochemistry Formalin-fixed paraffin-embedded brain tissue containing the choroid plexus was sectioned at 5 μm on a microtome and mounted on glass slides. Deparaffinized sections were reacted with a rabbit anti-Tf antibody (A0061, Dako Ltd.) at room temperature, followed by staining with a goat anti-rabbit IgG conjugated with Alexa Fluor 488 (A11034, Thermo Fisher Scientific). A biotinylated recombinant PVL was provided by Medical and Biological Laboratories Co., Ltd. (Nagoya, Japan). The sections were reacted with a biotinylated Psathyrella velutina lectin (PVL) at room temperature, followed by staining with Cy3-streptoavidin (PA43001, GE Health Science). Sambucus sieboldiana agglutinin (SSA) lectin was provided by J-Oil Mills, Kanagawa, Japan. Analysis of CSF proteins For SDS polyacrylamide gel electrophoresis, the drainage fluid from hydranencephaly was dissolved in Laemmli sample buffer, boiled for 3 min and loaded on a gradient gel (5–20%) (SuperSep™ Ace; Wako Pure Chemical Industries, Osaka, Japan). After SDS-PAGE, protein bands were visualized with a Silver Stain II kit Wako (Wako Pure Chemical Industries). Tf isoforms, ‘brain-type’ and serum-type Tf, were analysed by immunoblotting. Briefly, CSF (0.5 μl) was dissolved in Laemmli sample buffer without 2-mercaptoethanol, boiled for 3 min and loaded on SDS-polyacrylamide gels. After SDS-PAGE, the proteins were transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA). The membrane was blocked in 3% skim milk, incubated sequentially with an anti-Tf antibody (Bethyl Laboratories, Montgomery, TX) and a horseradish peroxidase-labeled anti-goat IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) and developed using a SuperSignal West Dura Chemiluminescence Substrate Kit (Pierce Biotechnology, Rockford, IL). Signal intensities were quantified by chromato-scanning with a CS Analyzer (ATTO, Tokyo, Japan). Each western blot included 2–10 μg of ‘brain-type’ Tf to establish a calibration curve. Standard ‘brain-type’ Tf was purified from CSF according to Futakawa et al. (21). Antibodies against Lipocalin-type prostaglandin D synthase (L-PGDS) (PA1-46023, Thermo Fisher Scientific) and transthyretin (TTR) (ab9015, abcom) were used for detecting the antigens on a blot. Soluble forms of APP695 and APP770, sAPP and sAPP770, were quantified by a sandwich Enzyme-Linked ImmunoSorbent Assay (ELISA), a Human sAPP Total Assay Kit and APP770 Assay Kit, respectively (Immuno-Biological Laboratories Co., Ltd., Gunma, Japan). According to manufacturer’s instructions, CSF at 1:16 dilution was analysed in triplicate by ELISA. Evaluation of clinical grading scales Based on the Clinical Guidelines for iNPH (32, 34), we used neurological tests such as MMSE for dementia (0–30), FAB for frontal lobe function (0–18), and modified Rankin Scale (mRS) for neurological disturbance (0–6). We evaluated each clinical grading scale pre- and post-operatively (11.8 ± 7.7 months after surgery), and analysed its correlation with concentrations of ‘brain-type’ Tf or sAPP. Statistical analysis Data were assessed as parametric or non-parametric by either the Kolmogorov-Smirnov or Shapiro–Wilk method. We described parametric data as mean ± SD. Parametric data were analysed by the Student t-test and Welch test, and assessed for multiple comparisons using Dunnett’s test. We checked correlations between ‘brain-type’ Tf and clinical grading scales using Pearson correlation coefficients. We did not perform either sample size calculation or outlier detection. For each comparison, P-value < 0.05 was considered as statistically significant. Statistical analyses were performed with StatMate V for Win&Mac Hybrid software (ATMS, Co., Ltd., Tokyo). Results Transferrin in CSF-like fluid in hydranencephaly We previously reported that CSF contains two glycan-isoforms of Tf: one is ‘brain-type’ Tf and the other is serum-type. Serum-type Tf has α2, 6-sialic acid-terminated N-glycans which are identical to those of serum Tf, suggesting that serum-type Tf is derived from blood. Brain-type Tf has β1, 4-N-acetylglucosamine (GlcNAc)-terminated N-glycans with bisecting GlcNAc and core α1, 6-fucose, which are observed with brain-derived proteins such as PGDS (22) and β-secretase (or BACE1) (23), suggesting that the brain-type Tf is produced in the brain. To examine production of brain-type Tf in the brain or cerebrum, we analysed CSF or CSF-like fluid of hydranencephaly, malformation of the cerebrum, in which the cerebral hemispheres are absent due to bilateral occlusion of internal carotid arteries, whereas the midbrain, cerebellum, and other brain structures nourished by vertebral artery may remain (35, 36). On CT analysis, a hydranencephaly patient (Case 2) shows poorly developed cerebrum and intracranial space appears to be replaced with CSF-like fluid (Fig. 1A). The fluid and control CSF were analysed by SDS-PAGE followed by silver staining (Fig. 1B). Arrows indicate migration positions of purified serum proteins such as Tf (ca. 75 kDa), albumin (ca. 66 kDa), immunoglobulin heavy chain (ca. 48 kDa), and light chains (ca. 24 kDa). Bands with similar positions are detected in CSF of control and hydranencephaly patients. CSF additionally contains 22 and 14 kDa bands. These two bands were reacted with antibodies against L-PGDS and transthyretin (TTR), respectively, on immunoblotting (data not shown). The CSF band patterns of major proteins in the hydranencephaly patients are not similar to those in control, but the band intensities are more intense in the hydranecephaly patients than in control. On an immunoblot using anti-Tf antibody, brain-type Tf is markedly decreased in Case 1 and hardly detected in Case 2 (Fig. 1C), suggesting that brain-type Tf is mainly produced in the cerebrum. We examined binding specificity of lectins for detecting glycans of Tf isoforms. Tf isoforms in CSF were purified by immunoaffinity. The purified Tf fraction contains brain- and serum-type Tf (Fig. 2A and B). PVL lectin, a binder to GlcNAc (37), detects brain-type Tf but not serum-type in purified Tf fraction and crude CSF (Fig. 2C and D). The lectin also detects several glycoproteins in crude CSF (Fig. 2C). A major glycoprotein appears in immunoglobulin heavy chains, which are reported to have heterogenous glycans such as GlcNAc-, galactose-, and sialicacid-terminated N-glycans in the Fc portion (38). In contrast, SSA lectin, a binder to α2, 6sialic acid, detects serum-type Tf but not brain-type in purified Tf fraction and crude CSF. SSA also detected other glycoproteins in crude CSF. Fig. 2 View largeDownload slide Immunoblot and lectin blot analysis for purified Tf and crude CSF. Tf was purified from CSF by immunoaffinity column chromatography. Purified Tf (20 ng) and crude CSF (1 μl) were dissolved in laemmli sample buffer containing 2-mercaptoethanol, boiled for 3 min, and loaded on a SDS-polyacrylamide gel. The bands are visualized with a silver staining kit (A). Blots are probed with an anti-Tf antibody (B), recombinant PVL (C) or SSA (D). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Fig. 2 View largeDownload slide Immunoblot and lectin blot analysis for purified Tf and crude CSF. Tf was purified from CSF by immunoaffinity column chromatography. Purified Tf (20 ng) and crude CSF (1 μl) were dissolved in laemmli sample buffer containing 2-mercaptoethanol, boiled for 3 min, and loaded on a SDS-polyacrylamide gel. The bands are visualized with a silver staining kit (A). Blots are probed with an anti-Tf antibody (B), recombinant PVL (C) or SSA (D). Migrating positions of purified serum transferrin (Tf), albumin (Alb), immunoglobulin heavy chain (Ig-H) and light chain (Ig-L) are indicated with arrowheads. Histochemical analysis on choroid plexus We previously found that choroid plexus epithelial cells, which produce CSF, were strongly stained by anti-Tf antibody (21). To examine Tf isoforms produced by the choroid plexus, the tissue was stained with PVL or SSA lectin. First, we co-stained the choroid plexus with PVL lectin and Tf-antibody. Tf-antibody stains the choroid plexus epithelial cells, some parts of interstitial tissues, and blood clots in the capillary (Fig. 3A and B; left panels). High magnification shows preferential staining of basolateral portions of the epithelial cells with minimal staining of apical portions (Fig. 3A and B; right panels). PVL lectin also stains the basolateral portions of the epithelial cells, partially overlapping the Tf-antibody stain (Fig. 3C and D; left panels, black arrowhead). Other areas are scarcely stained with PVL lectin, suggesting that GlcNAc-terminated glycoproteins are localized in the epithelial cells. Some PVL lectin signals do not overlap those of anti-Tf antibody (Fig. 3C and D; right panels, black arrowhead), suggesting the presence of GlcNAc-terminated glycoproteins other than Tf. Next, we co-stained the choroid plexus with SSA lectin and anti-Tf antibody. SSA strongly stains parts of interstitial tissue and blood clots; the lectin signals substantially overlap strong signals of anti-Tf antibody (Fig. 3E–H; left panels). In contrast, the majority of lectin staining on basolateral portions of the epithelial cells do not overlap those of Tf antibody (Fig. 3E–H; right panels, black arrowhead), suggesting that the basolateral Tf signals are not due to the presence of serum Tf but possibly brain-type Tf. The epithelial cells are slightly stained with SSA, but the signals did not overlap those of Tf antibody. The cells may contain α2, 6-sialylated glycoproteins other than Tf. Fig. 3 View largeDownload slide Histochemical analysis of the choroid plexus. Illustration (A) and (E) are schemas of the choroid plexus: choroid plexus epithelial cells (epi), interstitial tissue (int), blood capillary (capi) and lateral ventricle (LV). The choroid plexus tissue is stained with an anti-Tf antibody (B, F), PVL lectin (C) or SSA lectin (G). The signals of anti-Tf antibody are merged with those of PVL (D) or SSA lectin (H). In each set of photograph, the high magnification of the left panel is shown in the right panel. Fig. 3 View largeDownload slide Histochemical analysis of the choroid plexus. Illustration (A) and (E) are schemas of the choroid plexus: choroid plexus epithelial cells (epi), interstitial tissue (int), blood capillary (capi) and lateral ventricle (LV). The choroid plexus tissue is stained with an anti-Tf antibody (B, F), PVL lectin (C) or SSA lectin (G). The signals of anti-Tf antibody are merged with those of PVL (D) or SSA lectin (H). In each set of photograph, the high magnification of the left panel is shown in the right panel. Changes of CSF biomarkers before and after CSF drainage and shunt surgery Previously we reported that brain-type Tf levels were decreased in the CSF of iNPH patients (21). Brain-type and serum-type Tf were estimated by Western blotting, and the ratio of serum-type/brain-type Tf was used as a diagnostic marker to minimize blot-to-blot variation of Tf signals. In the present study, we utilize only brain-type Tf as a marker for continuous drainage or post-operative specimens and do not use the ratio because accurate estimation of serum-type Tf is difficult due to surgical contamination of blood Tf, which is indistinguishable from serum-type Tf. Brain-type Tf was quantitatively estimated by Western blotting; the signals of standard ‘brain-type’ Tf was linear at least in a range of 2–10 ng (Fig. 4A and B). Each Western blot included 2–10 ng of standard and amount of ‘brain-type’ Tf was calculated. Standard ‘brain-type’ Tf was purified from CSF according to Futakawa et al. Brain-type Tf levels were analysed in CSF during continuous lumbar drainage to evaluate effect of CSF withdrawal. The levels increase during post-drainage days 1–3 (126 ± 11%, not significant) and 4–6 (143 ± 17%, P < 0.01) (Fig. 4C), indicating that the marker production begin to increase within a week after the drainage. The long-term recovery of brain-type Tf was analysed after shunt surgery. The levels are markedly increased during post-operative months 1–3 and then gradually decline during post-operative months 6–24 (Fig. 5). Before surgery, sAPP levels of iNPH were significantly lower than non-iNPH; 450.3 ± 187.6 ng/ml (n = 34) vs. 677.2 ± 430.5 ng/ml (n = 15) (P < 0.05). After shunt surgery, sAPP levels are significantly increased at 6 months and then gradually decline (Fig. 5). In contrast, we detect subtle changes in levels of sAPP770, which is derived from brain vascular endothelial cells. The results suggest that brain-type Tf and APP levels are rapidly normalized after shunt surgery, but with different time courses. Fig. 4 View largeDownload slide ‘Brain-type’ Tf levels during continuous lumbar CSF drainage for iNPH patients. Brain-type Tf in CSF (0.5 μl) is estimated by western blotting (A). Signals of purified standard ‘brain-type’ CSF (2–10 ng) give a linear calibration curve (B). CSF is continuously withdrawn from iNPH patients (n = 4) for 6 days. Brain-type Tf levels before drainage and post-drainage days 1–3 or 4–6 are indicated with box plots (C). Significant difference is shown by Dunnett’s multiple comparison analysis; n.s., not significant. Fig. 4 View largeDownload slide ‘Brain-type’ Tf levels during continuous lumbar CSF drainage for iNPH patients. Brain-type Tf in CSF (0.5 μl) is estimated by western blotting (A). Signals of purified standard ‘brain-type’ CSF (2–10 ng) give a linear calibration curve (B). CSF is continuously withdrawn from iNPH patients (n = 4) for 6 days. Brain-type Tf levels before drainage and post-drainage days 1–3 or 4–6 are indicated with box plots (C). Significant difference is shown by Dunnett’s multiple comparison analysis; n.s., not significant. Fig. 5 View largeDownload slide Changes of CSF biomarker levels after shunt surgery. CSF biomarker levels are quantified at pre-surgery (indicated as 100% at 0 month) and post-surgery (mean ± SE); brain-type Tf (closed circle), sAPP (grey circle), sAPP770 (open square). Significant difference is shown by Dunnett’s multiple comparison analysis and indicated with asterisks; P < 0.05 (*), P < 0.01 (**). Fig. 5 View largeDownload slide Changes of CSF biomarker levels after shunt surgery. CSF biomarker levels are quantified at pre-surgery (indicated as 100% at 0 month) and post-surgery (mean ± SE); brain-type Tf (closed circle), sAPP (grey circle), sAPP770 (open square). Significant difference is shown by Dunnett’s multiple comparison analysis and indicated with asterisks; P < 0.05 (*), P < 0.01 (**). Correlation between brain-type Tf levels and dementia scores after shunt surgery All neurological clinical grading scales were improved post-operatively, after 11.8 ± 7.7 months: pre- versus post-operative scores are as follow; MMSE, 21.4 ± 5.3 vs. 24.3 ± 4.7 (P = 0.022); FAB, 10.5 ± 3.3 vs. 12.3 ± 3.3 (P = 0.036); and mRS, 2.9 ± 0.8 vs. 2.1 ± 1.1 (P = 0.002), respectively. We analysed correlations between the cognitive scales and the biomarkers in CSF at various time points. Before shunt surgery, the levels of brain-type Tf show low (r = 0.351, P = 0.042) and moderate (r = 0.527, P = 0.002) correlation with MMSE and FAB scores, respectively (Fig. 6A and B). After post-operative month 3, increase of brain-type Tf strongly correlates with scores of MMSE (r = 0.697, P = 0.037) and FAB (r = 0.727, P = 0.041) (Fig. 6C and D; the left panel). The increase of brain-type Tf at post-operative month 12 is moderately correlated with MMSE (r = 0.549, P = 0.022) but not with FAB (r = 0.373, P = 0.154) (Fig. 6C and D; the right panel). The scores of mRS were not correlated with brain-type Tf levels before and after shunt surgery (data not shown). The results suggest that brain-type Tf levels at post-operative month 3 (and 12) would predict recovery from dementia of iNPH patients. Fig. 6 View largeDownload slide Correlation between brain-type Tf and cognitive scales. Correlation between ‘brain-type’ Tf levels and cognitive scales, MMSE (A) or FAB (B), are indicated. ‘Brain-type’ Tf levels are measured before surgery (A, B) and % increases at post-operative months 3 and 12 (C, D) are examined. Post-operative cognitive scales were examined at 11.8 ± 7.7 months after shunt surgery. Fig. 6 View largeDownload slide Correlation between brain-type Tf and cognitive scales. Correlation between ‘brain-type’ Tf levels and cognitive scales, MMSE (A) or FAB (B), are indicated. ‘Brain-type’ Tf levels are measured before surgery (A, B) and % increases at post-operative months 3 and 12 (C, D) are examined. Post-operative cognitive scales were examined at 11.8 ± 7.7 months after shunt surgery. Discussion In drainage fluid of hydranencephaly patients, major bands appear to be Tf, albumin, immunoglobulin heavy chain and light chain, which possibly diffuse into CSF from blood. Indeed, about 80% of CSF proteins are derived from blood (39). Additional bands at the positions of L-PGDS and TTR are more intense in CSF of hydranencephaly patients than the control. The observation is consistent with the fact that mRNAs of L-PGDS and TTR are highly expressed in leptomeninges and hypothalamus as well as cerebellum (40, 41), which are preserved in hydranencephaly patients. Brain-type Tf levels were markedly decreased in the drainage fluid of hydranencephaly patients, suggesting that brain-type Tf is derived from cerebrum. Histochemical analysis revealed that anti-Tf antibody stained choroid plexus tissue including choroid plexus epithelial cells, interstitial tissue, and blood clots in the capillaries, indicating ubiquitous presence of Tf in the tissue. The basolateral portions of epithelial cells were stained by the antibody, whereas apical portions were not. Tf in the apical portion may be rapidly secreted from the cells and its steady state levels may be very low, if any. The basolateral portions of epithelial cells were also stained with PVL lectin, a binder for GlcNAc-terminated glycans, and the staining was partially merged with anti-Tf antibody staining, suggesting the presence of brain-type Tf in choroid plexus epithelial cells. Some PVL signals are not merged with the antibody staining, suggesting the epithelial cells produced glycoproteins having GlcNAc-terminated glycans other than brain-type Tf. Based on these results we hypothesized that brain-type Tf is, at least partly, produced in choroid plexus epithelial cells and secreted together with CSF. To confirm the hypothesis, purification of Tf from choroid plexus and its glycan analysis will be needed in the future. Brain-type Tf levels are reduced in iNPH, in which decrease of CSF production is suspected (42), and increase after shunt surgery. It is notable that the marker levels begin to increase within a week by continuous drainage. In addition, brain-type Tf significantly increased during post-operative months 1–12, showing a peak at 3 months. The increment may involve several mechanisms such as increase of brain-type Tf secretion and/or reduced total volume of CSF. The reduced CSF volume could result in ‘apparent’increase of concentration of brain-type Tf. In this case, concentration of other proteins would be also increase. At post-operative month 3, concentration of brain-type Tf, sAPP and sAPP770 increased to different degrees: 220%, 145% and 110% respectively, suggesting that reduced volume of CSF is not the only reason for increase of brain-type Tf. Each protein has different origin, i.e. brain-type Tf from choroid plexus, sAPP from neurons (26), and sAPP770 from endothelial cells (29). Their secretion or excretion would be under different controlling mechanisms. CSF accumulates in the lateral ventricles of iNPH patients, compressing the brain parenchyma and impairing cognitive functions. The shunt surgery drains excess CSF, relieves the compression, and normalizes cognitive functions, although recovery from dementia often needs long-term intervention over a year. We speculated that dementia scales would correlate with levels of brain-type Tf. Indeed, levels of MMSE and FAB scales were correlated with those of brain-type Tf even before shunt surgery. After the surgery, dementia scores measured post-operatively at 11.8 ± 7.7 months were found to correlate with the increase of brain-type Tf levels at 3 (or 12) months, suggesting that the marker predicted recovery from dementia after shunt surgery. Soluble APP also significantly increased during post-operative months 6–12, but sAPP levels did not predict recovery from dementia. Consistent with our result, Jeppsson et al. reported that sAPP isoforms were not correlated with cognitive improvement after the surgery (31). Miyajima et al. reported, however, that MMSE scores at post-operative 6 months were correlated with sAPPα and phosphorylated-tau (9). Further analysis is required for establishing prognostic ability of sAPP isoforms. In conclusion, we demonstrate that brain-type Tf appears to emanate from the choroid plexus and would be an indicator for CSF production that predicts recovery from dementia after shunt surgery for iNPH patients. This favourable prediction would encourage patients and caregivers to persevere in their rehabilitation efforts. A limitation of the present study was to be a single-centre investigation with a limited number of patients. Acknowledgements We sincerely acknowledge our colleagues in the Hashimoto laboratories for valuable discussions. We acknowledge the Brain Bank for Aging Research, Tokyo Metropolitan Geriatric Hospital and Institute of Gerontology for providing brain tissues. Funding This work was partly supported by Japan Agency for Medical Research and Development (AMED) [grant numbers 16hm0102042h0001 and 17hm0102042h0002]; the Japan Science and Technology Agency [grant numbers AS221Z00232F, AS231Z01053, 241FT0255 and 149]; the Ministry of Education, Culture, Sports, Science, and Technology of Japan [a Grant-in-Aid for Scientific Research on Innovative Areas, grant number 23110002 (Deciphering sugar chain-based signals regulating integrative neuronal functions), a Grant-in-Aid for Scientific Research, grant number 23590367]; Fukushima Medical University [Grant for Strategic Research Promotion of FMU]; The General Insurance Association of Japan [a medical research grant on traffic accident]. Conflict of Interest None declared. References 1 Cushing H. ( 1914) Studies on the cerebro-spinal fluid: I. Introduction. J. Med. Res.  31, 1– 19 Google Scholar PubMed  2 Cutler R.W., Page L., Galicich J., Watters G.V. ( 1968) Formation and absorption of cerebrospinal fluid in man. Brain  91, 707– 720 Google Scholar CrossRef Search ADS PubMed  3 Cushing H. ( 1969) Selected Papers on Neurosurgery . Yale University Press, New Haven and London 4 Carare R.O., Hawkes C.A., Weller R.O. ( 2014) Afferent and efferent immunological pathways of the brain. Anatomy, function and failure. Brain Behav. Immun . 36, 9– 14 Google Scholar CrossRef Search ADS PubMed  5 Owler B.K., Pitham T., Wang D. ( 2010) Aquaporins: relevance to cerebrospinal fluid physiology and therapeutic potential in hydrocephalus. Cerebrospinal Fluid Res . 7, 15 Google Scholar CrossRef Search ADS PubMed  6 Brierley J.B. ( 1950) The penetration of particulate matter from the cerebrospinal fluid into the spinal ganglia, peripheral nerves, and perivascular spaces of the central nervous system. J. Neurol. Neurosurg. Psychiatry  13, 203– 215 Google Scholar CrossRef Search ADS PubMed  7 Miura M., Kato S., Lüdinghausen M. V. ( 1998) Lymphatic drainage of the cerebrospinal fluid from monkey spinal meninges with special reference to the distribution of the epidural lymphatics. Arch. Histol. Cytol . 61, 277– 286 Google Scholar CrossRef Search ADS PubMed  8 Miura M.U.T., Takahashi A. ( 2015) Latest knowledge of cerebrospinal fluid: production, circulation and absorption with specialreference to the collateral absorptive routes via the epidural lymphatic system. Spine Spinal Cord  28, 694– 703 9 Miyajima M., Nakajima M., Ogino I., Miyata H., Motoi Y., Arai H. ( 2013) Soluble amyloid precursor protein alpha in the cerebrospinal fluid as a diagnostic and prognostic biomarker for idiopathic normal pressure hydrocephalus. Eur. J. Neurol.  20, 236– 242 Google Scholar CrossRef Search ADS PubMed  10 Hamlat A., Adn M., Sid-ahmed S., Askar B., Pasqualini E. ( 2006) Theoretical considerations on the pathophysiology of normal pressure hydrocephalus (NPH) and NPH-related dementia. Med. Hypotheses  67, 115– 123 Google Scholar CrossRef Search ADS PubMed  11 Mori E., Ishikawa M., Kato T., Kazui H., Miyake H., Miyajima M., Nakajima M., Hashimoto M., Kuriyama N., Tokuda T., Ishii K., Kaijima M., Hirata Y., Saito M., Arai H., Japanese S., of Normal Pressure H. ( 2012) Guidelines for management of idiopathic normal pressure hydrocephalus: second edition. Neurol. Med. Chir (Tokyo)  52, 775– 809 Google Scholar CrossRef Search ADS PubMed  12 Jaraj D., Rabiei K., Marlow T., Jensen C., Skoog I., Wikkelso C. ( 2014) Prevalence of idiopathic normal-pressure hydrocephalus. Neurology  82, 1449– 1454 Google Scholar CrossRef Search ADS PubMed  13 Gallia G.L., Rigamonti D., Williams M.A. ( 2006) The diagnosis and treatment of idiopathic normal pressure hydrocephalus. Nat. Clin. Pract. Neurol.  2, 375– 381 Google Scholar CrossRef Search ADS PubMed  14 Relkin N., Marmarou A., Klinge P., Bergsneider M., Black P.M. ( 2005) Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery  57, S2-4– 16. Discussion ii–v Google Scholar CrossRef Search ADS   15 Kazui H., Miyajima M., Mori E., Ishikawa M., Investigators S. ( 2015) Lumboperitoneal shunt surgery for idiopathic normal pressure hydrocephalus (SINPHONI-2): an open-label randomised trial. Lancet Neurol . 14, 585– 594 Google Scholar CrossRef Search ADS PubMed  16 Hashimoto M., Ishikawa M., Mori E., Kuwana N. and Study of I.o.n.i. ( 2010) Diagnosis of idiopathic normal pressure hydrocephalus is supported by MRI-based scheme: a prospective cohort study. Cerebrospinal Fluid Res . 7, 18 Google Scholar CrossRef Search ADS PubMed  17 Meier U., Lemcke J. ( 2006) Clinical outcome of patients with idiopathic normal pressure hydrocephalus three years after shunt implantation. Acta Neurochir. Suppl . 96, 377– 380 Google Scholar CrossRef Search ADS PubMed  18 Pujari S., Kharkar S., Metellus P., Shuck J., Williams M.A., Rigamonti D. ( 2008) Normal pressure hydrocephalus: long-term outcome after shunt surgery. J. Neurol. Neurosurg. Psychiatry  79, 1282– 1286 Google Scholar CrossRef Search ADS PubMed  19 Golz L., Ruppert F.H., Meier U., Lemcke J. ( 2014) Outcome of modern shunt therapy in patients with idiopathic normal pressure hydrocephalus 6 years postoperatively. J. Neurosurg . 121, 771– 775 Google Scholar CrossRef Search ADS PubMed  20 Walchenbach R., Geiger E., Thomeer R.T., Vanneste J.A. ( 2002) The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J. Neurol. Neurosurg. Psychiatry  72, 503– 506 Google Scholar PubMed  21 Futakawa S., Nara K., Miyajima M., Kuno A., Ito H., Kaji H., Shirotani K., Honda T., Tohyama Y., Hoshi K., Hanzawa Y., Kitazume S., Imamaki R., Furukawa K., Tasaki K., Arai H., Yuasa T., Abe M., Arai H., Narimatsu H., Hashimoto Y. ( 2012) A unique N-glycan on human transferrin in CSF: a possible biomarker for iNPH. Neurobiol. Aging  33, 1807– 1815 Google Scholar CrossRef Search ADS PubMed  22 Hoffmann A., Nimtz M., Wurster U., Conradt H.S. ( 2002) Carbohydrate structures of beta-trace protein from human cerebrospinal fluid: evidence for “brain-type” N-glycosylation. J. Neurochem . 63, 2185– 2196 Google Scholar CrossRef Search ADS   23 Kizuka Y., Kitazume S., Fujinawa R., Saito T., Iwata N., Saido T.C., Nakano M., Yamaguchi Y., Hashimoto Y., Staufenbiel M., Hatsuta H., Murayama S., Manya H., Endo T., Taniguchi N. ( 2015) An aberrant sugar modification of BACE1 blocks its lysosomal targeting in Alzheimer’s disease. EMBO Mol. Med . 7, 175– 189 Google Scholar CrossRef Search ADS PubMed  24 Selkoe D.J. ( 2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev.  81, 741– 766 Google Scholar CrossRef Search ADS PubMed  25 Tanzi R.E., Bertram L. ( 2005) Twenty years of the Alzheimer’s disease amyloid hypothesis: a genetic perspective. Cell  120, 545– 555 Google Scholar CrossRef Search ADS PubMed  26 De Strooper B., Annaert W., Cupers P., Saftig P., Craessaerts K., Mumm J.S., Schroeter E.H., Schrijvers V., Wolfe M.S., Ray W.J., Goate A., Kopan R. ( 1999) A presenilin-1-dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature  398, 518– 522 Google Scholar CrossRef Search ADS PubMed  27 Tomita S., Kirino Y., Suzuki T. ( 1998) Cleavage of Alzheimer’s amyloid precursor protein (APP) by secretases occurs after O-glycosylation of APP in the protein secretory pathway. Identification of intracellular compartments in which APP cleavage occurs without using toxic agents that interfere with protein metabolism. J. Biol. Chem.  273, 6277– 6284 Google Scholar CrossRef Search ADS PubMed  28 Wertkin A.M., Turner R.S., Pleasure S.J., Golde T.E., Younkin S.G., Trojanowski J.Q., Lee V.M. ( 1993) Human neurons derived from a teratocarcinoma cell line express solely the 695-amino acid amyloid precursor protein and produce intracellular beta-amyloid or A4 peptides. Proc. Natl. Acad. Sci. U.S.A.  90, 9513– 9517 Google Scholar CrossRef Search ADS PubMed  29 Kitazume S., Yoshihisa A., Yamaki T., Oikawa M., Tachida Y., Ogawa K., Imamaki R., Hagiwara Y., Kinoshita N., Takeishi Y., Furukawa K., Tomita N., Arai H., Iwata N., Saido T., Yamamoto N., Taniguchi N. ( 2012) Soluble amyloid precursor protein 770 is released from inflamed endothelial cells and activated platelets: a novel biomarker for acute coronary syndrome. J. Biol. Chem.  287, 40817– 40825 Google Scholar CrossRef Search ADS PubMed  30 Nakajima M., Miyajima M., Ogino I., Akiba C., Sugano H., Hara T., Fusegi K., Karagiozov K., Arai H. ( 2015) Cerebrospinal fluid biomarkers for prognosis of long-term cognitive treatment outcomes in patients with idiopathic normal pressure hydrocephalus. J. Neurol. Sci . 357, 88– 95 Google Scholar CrossRef Search ADS PubMed  31 Jeppsson A., Holtta M., Zetterberg H., Blennow K., Wikkelso C., Tullberg M. ( 2016) Amyloid mis-metabolism in idiopathic normal pressure hydrocephalus. Fluids Barriers CNS  13, 13 Google Scholar CrossRef Search ADS PubMed  32 Ishikawa M. and Guideline Committe for Idiopathic Normal Pressure Hydrocephalus, J.S.o.N.P.H. ( 2004) Clinical guidelines for idiopathic normal pressure hydrocephalus. Neurol. Med. Chir (Tokyo)  44, 222– 223 Google Scholar CrossRef Search ADS PubMed  33 McKhann G., Drachman D., Folstein M., Katzman R., Price D., Stadlan E.M. ( 1984) Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology  34, 939– 944 Google Scholar CrossRef Search ADS PubMed  34 Mori K. ( 2001) Management of idiopathic normal-pressure hydrocephalus: a multiinstitutional study conducted in Japan. J. Neurosurg . 95, 970– 973 Google Scholar CrossRef Search ADS PubMed  35 Greene M.F., Benacerraf B., Crawford J.M. ( 1985) Hydranencephaly: US appearance during in utero evolution. Radiology  156, 779– 780 Google Scholar CrossRef Search ADS PubMed  36 Pavone P., Pratico A.D., Vitaliti G., Ruggieri M., Rizzo R., Parano E., Pavone L., Pero G., Falsaperla R. ( 2014) Hydranencephaly: cerebral spinal fluid instead of cerebral mantles. Ital. J. Pediatr.  40, 79 Google Scholar CrossRef Search ADS PubMed  37 Kochibe N., Matta K.L. ( 1989) Purification and properties of an N-acetylglucosamine-specific lectin from Psathyrella velutina mushroom. J. Biol. Chem . 264, 173– 177 Google Scholar PubMed  38 Takahashi N., Ishii I., Ishihara H., Mori M., Tejima S., Jefferis R., Endo S., Arata Y. ( 1987) Comparative structural study of the N-linked oligosaccharides of human normal and pathological immunoglobulin G. Biochemistry  26, 1137– 1144 Google Scholar CrossRef Search ADS PubMed  39 Reiber H., Peter J.B. ( 2001) Cerebrospinal fluid analysis: disease-related data patterns and evaluation programs. J. Neurol. Sci . 184, 101– 122 Google Scholar CrossRef Search ADS PubMed  40 Blay P., Nilsson C., Hansson S., Owman C., Aldred A.R., Schreiber G. ( 1994) An in vivo study of the effect of 5-HT and sympathetic nerves on transferrin and transthyretin mRNA expression in rat choroid plexus and meninges. Brain Res . 662, 148– 154 Google Scholar CrossRef Search ADS PubMed  41 Ohinata K., Takagi K., Biyajima K., Fujiwara Y., Fukumoto S., Eguchi N., Urade Y., Asakawa A., Fujimiya M., Inui A., Yoshikawa M. ( 2008) Central prostaglandin D(2) stimulates food intake via the neuropeptide Y system in mice. FEBS Lett . 582, 679– 684 Google Scholar CrossRef Search ADS PubMed  42 Silverberg G.D., Mayo M., Saul T., Rubenstein E., McGuire D. ( 2003) Alzheimer’s disease, normal-pressure hydrocephalus, and senescent changes in CSF circulatory physiology: a hypothesis. Lancet Neurol . 2, 506– 511 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations Aβ amyloid β AD Alzheimer disease APP amyloid precursor protein CSF cerebrospinal fluid CT computed tomography DESH disproportionately enlarged subarachnoid-space hydrocephalus ELISA enzyme-linked immunosorbent assay FAB Frontal Assessment Battery GlcNAc , N-acetylglucosamine iNPH idiopathic normal pressure hydrocephalus Ig immunoglobulin LP lumbar-peritoneal L-PGDS lipocalin-type prostaglandin D synthase MMSE Mini-Mental State Examination MRI magnetic resonance imaging mRS modified Rankin scale NPV negative predictive value PPV positive predictive value p-tau hyperphosphorylated-tau PVL Psathyrella velutina sAPP soluble protein fragments of amyloid precursor protein SSA Sambucus sieboldiana agglutinin Tf transferrin TTR transthyretin TUG Timed Up and Go VP ventriculo-peritoneal © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The Journal of BiochemistryOxford University Press

Published: Apr 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off