Abstract
The term frontotemporal lobar degeneration (FTLD) defines a group of heterogeneous conditions histologically characterized by neuronal degeneration, inclusions of various proteins, and synaptic loss. However, the molecular mechanisms contributing to these alterations are still unknown. As the Rho-GTPase family member Cell division cycle 42 (Cdc42) plays a key role in the regulation of actin cytoskeleton dynamics and spine formation, we investigated whether Cdc42 protein levels were altered in the disease. Cdc42 was increased in the frontal cortex of FTLD patients compared to age-matched controls, but also in Alzheimer’s disease (AD) patients included in the data-set. On the other hand, the pool of circulating Cdc42 in the plasma was altered in FTLD but not in AD patients. Interestingly, the stratification of the FTLD patients according to the different clinical variants showed a specific decrease of Cdc42 expression in the behavioral subgroup. This data support a role of Cdc42 in FTLD and specifically in the behavioral variant.
INTRODUCTION
Frontotemporal lobar degeneration (FTLD) is the second most common form of early-onset dementia after Alzheimer’s disease (AD), with a prevalence of 15-22/100,00 individuals [1–3]. The term FTLD defines a broad group of disorders all characterized by the degeneration of the frontal and anterior temporal lobes but very heterogeneous in terms of clinical and histological pathology [4]. The major clinical subtype of FTLD is called behavioral variant of frontotemporal dementia (bvFTD). It is characterized by progressive behavioral changes, such as apathy, anxiety, disinhibition, compulsive/ritualistic behavior, and hyperorality [5]. Another typical subtype of FTLD is the language variant, with the semantic and non-fluent variant of primary progressive aphasia (PPA) [4, 6].
From a histopathological point of view, the autoptic brain examination of FTLD patients often reveals neuronal inclusions containing TDP-43 and tau [7–10]. However, the correlation between abnormal protein accumulation and presentation of the disease is rather weak. Genetic factors are the only cause of FTLD so far identified and up to 50% of the patients have a family history of the disease [11–13]. Until now, 7 disease-causing genes have been discovered including MAPT, GRN, and C9ORF72, the most frequent genetic determinants [14–18].
In recent years, several promising biomarkers linked to neurodegenerative disorders have been identified thanks to genomic and proteomic studies. However, there is still an urgent need for new biomarkers. The current lack of unanimously accepted biomarkers for FTLD delays the achievement of a correct diagnosis and impedes appropriate patient stratification during clinical trial recruitment. This is a key point in FTLD due to the high heterogeneity and the poor correlation between clinical phenotype and underlying molecular pathology [19]. Presently, the only useful biomarker for patients’ stratification is progranulin: the level of circulating progranulin is a useful marker for early identification of GRN-related FTLD [20–23]. The decreased ratio of total tau and amyloid-β42 (Aβ42) in the cerebrospinal fluid (CSF) seems to be the most sensitive and specific biomarker able to differentially diagnose FTLD from AD [24]. One of the pathophysiological hallmarks of neurodegeneration, highly conserved between different diseases, is the alteration of the synaptic machinery. Alterations of synaptic functions have been correlated to several psychiatric and neurological brain disorders, including bipolar disorder, autism spectrum disorder, schizophrenia, and AD [25–28]. Among the factors regulating the activity and the maintenance of synaptic plasticity, the Rho-GTPase family has an important role in regulating the dynamics of the F-actin-rich cytoskeleton present in the spines [29–31]. Cell division cycle 42 (Cdc42) is one of the members of the family, which regulates the organization, polarity, and growth of the actin cytoskeleton [32] and promotes spine formation and maturation [33]. Cdc42 has been shown to contribute to signaling pathways involved with schizophrenia [34], mental retardation [35], bipolar disorder [36], and Huntington’s disease [37]. In the context of these studies, we evaluated Cdc42 protein levels in the frontal cortex and in the plasma of AD and FTLD patients as well as age-matched non-demented controls.
MATERIALS AND METHODS
Subjects
Brain samples from AD (n = 24) and FTLD (n = 4) patients, and age-matched non-demented controls (n = 11) were obtained from the Biobank of the IRCCS Foundation – Carlo Besta Neurological Institute and from the Brain Bank of the Department of Pathology at Indiana University School of Medicine. The neuropathological diagnosis was performed according to international guidelines for the assessment of AD [38] and FTLD [39].
For the plasma study, we included AD (n = 115), and FTLD (n = 101) patients, and age- and sex-matched cognitively healthy controls (CTRL, n = 104). The patients underwent clinical and neurological examination at the MAC Memory Center of the IRCCS Centro San Giovanni di Dio-Fatebenefratelli, Brescia, Italy. Clinical diagnosis of AD or FTLD was made according to international guidelines [4, 41]. Biological samples were collected and stored in the Biobank of the IRCCS Centro San Giovanni di Dio-Fatebenefratelli, Brescia, Italy, after obtaining informed consent, as approved by the local ethics committee (approval No. 26/2014). The study was also approved by the local ethics committee (approval No. 03/2015). Plasma was isolated according to standard procedures.
The demographic characteristics of the patients in the study are shown in Tables 1 and 2 and the Braak stage of the AD brain samples is shown in Table 3.
Demographic characteristic and Cdc42 levels in the three groups included in the brain study
Demographic characteristic and Cdc42 levels in the three groups included in the plasma study
Braak stage and known co-pathologies of the brain samples in the AD study group
Biochemical analyses
Brain samples from frontal cortex of AD, FTLD patients, and age-matched non-demented controls were homogenized in PBS1X and centrifuged at 1,500×g for 15 min. Supernatants were collected and stored at –80°C. The total proteins amount was measured by BCA Protein Assay kit (Pierce).
Blood samples from AD, FTLD patients, and age-matched non-demented controls were kept at 4°C for at least 20 min and then centrifuged for 5 min (4°C, 1,000×g). Plasma was collected and centrifuged 5 min (4°C, 1,000×g) after the addition of the protease inhibitors. Plasma samples were aliquoted and stored at –80°C.
Cdc42 levels were measured in plasma and brain homogenates, in duplicate, using a commercially available ELISA kit (Human Cell Division Cycle Protein 42, WUXI DONGLIN SCI&TECH DEVOLPMENT CO., LTD, Wuxi, Jiangsu, PRC).
Statistical analysis
The Kolmogorov-Smirnov test was performed in all continuous variables to define the presence of normality. The non-parametric two-tailed Mann Whitney test or Kruskal-Wallis with post hoc Dunn’s test was used to evaluate differences between the study groups. Data were analyzed using Prism 6 (GraphPad Software) and displayed with box and whiskers (minimum to maximum). Statistical significant differences are reported as *p < 0.05 and **p < 0.01.
RESULTS
Cdc42 levels increased in the brains of patients affected by AD and FTLD
To explore the possible role of Cdc42 in dementia, protein levels were measured in brain homogenates from AD and FTLD patients, and from age-matched non-demented controls. We observed that Cdc42 levels significantly increased both in AD and FTLD brains, as compared to CTRL (Fig. 1; Kruskal-Wallis: CTRL versus AD versus FTLD, p = 0.0006; post hoc Dunn’s test: CTRL versus FTLD, p < 0.01; CTRL versus AD, p < 0.01; AD versus FTLD, p > 0.05).
Cdc42 in brains of CTRL subjects, AD and FTLD patients. Cdc42 levels (ng/mg of protein) in brain homogenates from CTRL subjects (n = 11), AD (n = 24), and FTLD patients (n = 4). Brain Cdc42 levels were increased in AD and FTLD groups compared to CTRL.
Cdc42 levels decreased in the plasma of patients affected by FTLD
We then evaluated the circulating Cdc42 protein levels in the plasma of patients affected by AD, FTLD, and age- and sex-matched healthy controls. We found that Cdc42 significantly decreased in the plasma of FTLD patients compared to CTRL and to AD patients (Fig. 2; Kruskal-Wallis: CTRL versus AD versus FTLD, p = 0.0048; post hoc Dunn’s test: CTRL versus FTLD, p = 0.0189; AD versus FTLD, p = 0.0094; CTRL versus AD, p > 0.9).
Plasma Cdc42 protein levels in CTRL subjects, AD, and FTLD patient. Cdc42 levels (ng/ml) were analyzed in plasma of the three studied groups: CTRL subjects (n = 104), AD (n = 115), and FTLD patients (n = 101). Plasma Cdc42 levels were decreased in FTLD group compared to CTRL and to AD patients.
Cdc42 levels were decreased in the plasma of patients with behavioral variant of FTD
We analyzed Cdc42 levels in the following FTLD subgroups: bvFTD (n = 50), PPA (n = 31), and unspecified FTLD subgroup (n = 20). We found that the levels of Cdc42 protein specifically decreased in the plasma of bvFTD patients compared to CTRL and to AD patients (Fig. 3; Kruskal-Wallis: CTRL versus AD versus bvFTD versus PPA versus unspecified FTLD, p = 0.0185; post hoc Dunn’s test: CTRL versus bvFTD, p = 0.0458; bvFTD versus AD, p = 0.0266; CTRL versus AD, p > 0.9; CTRL versus PPA, p > 0.9; CTRL versus unspecified FTLD, p > 0.9; PPA versus AD, p > 0.9; AD versus unspecified FTLD, p > 0.9; bvFTD versus PPA, p > 0.9; bvFTD versus unspecified FTLD, p > 0.9; PPA versus unspecified, p > 0.9).
Quantification of plasma Cdc42 protein levels in the subgroups of FTLD patients. Cdc42 levels (ng/ml) were analyzed in plasma of CTRL subjects (n = 104), AD patients (n = 115), bvFTD (n = 50), PPA (n = 31), and unspecified FTLD (n = 20). Plasma Cdc42 levels were decreased in bvFTD group compared to CTRL subjects and to AD patients.
DISCUSSION
Cdc42 is one of the members of the Rho-GTPase family. By acting as a molecular switch from an inactive GDP-bound form to an active GTP-bound form [42], it is able to regulate a variety of biological responses such as cell polarity, cytoskeleton remodeling, proliferation, migration, adhesion membrane trafficking, and transportation [32, 43]. The controlled regulation of these processes by Cdc42 is fundamental for normal cell function. In mice, it has been shown that Cdc42 loss or overexpression can alter phenotypes and pathways in different tissues such as heart, liver, pancreas, eye, skin, nervous system, blood, and immune system [44]. Its deregulation has been associated with several neurological and psychiatric disorders such as schizophrenia [34], mental retardation [35], bipolar disorder [36], and Huntington’s disease [37].
Cdc42 has an important role in the regulation of the organization, polarity, and growth of the actin cytoskeleton [32] fundamental for spine formation and maturation [33]. Alterations in this highly regulated pathway have been observed in several neurological and psychiatric disorders including bipolar disorder, autism, and schizophrenia [26, 27]. In FTLD patients, the extensive synaptic and dendritic loss observed [45, 46] is correlated with the cognitive impairment, as already reported for AD patients [47–51]. In the present study, we investigated the role of Cdc42 in neurodegenerative dementias (i.e., AD and FTLD).
In brain homogenates, we found an increase of Cdc42 both in AD and in FTLD patients. One potential explanation could be related to a synaptic compensation process, which has been previously described in AD [52, 53]. In autoptic AD brain samples, an increase in the expression levels of MAP2 and synaptophysin was described at Braak stages 3 and 4. Synaptophysin then decreased at Braak stages 5 and 6 [52]. According to this hypothesis, the brain would make use of the residual plastic capacity to try to counteract the synaptic deficit. This could in part explain the increased demand in the expression of proteins involved in cytoskeleton remodeling as Cdc42. As a synergistic mechanism, Cdc42 might also exert neurotrophic function as constitutively active Cdc42 was shown to promote filopodial dynamics which were similar to the effects of brain derived neurotrophic factor [54]. Further studies are clearly necessary to better understand Cdc42 brain dynamics.
Pointing toward a compartmentalized effect of Cdc42 in the context of AD, plasma levels of the protein were unchanged in AD patients compared to age-matched controls. On the other hand, Cdc42 levels significantly increased in FTLD brain samples and decreased in the plasma compared to age-matched controls. Interestingly, we showed that the alteration of the protein is present in the subgroup of the bvFTD but not in the PPA and in the unspecified FTLD. This indicates a specific role of Cdc42 in the prediction of the behavioral variant, which is the most common clinical subtype of FTLD. This variant is characterized by progressive behavioral changes, such as disinhibition, apathy/inertia, loss of sympathy/empathy, perseverative, stereotyped and compulsive/ritualistic behavior, and hyperorality [5]. These symptoms are similar to those observed in other non-neurodegenerative psychiatric disorders such as bipolar disorder and schizophrenia [55]. It is tempting to speculate that the underlying mechanism of the behavioral component involves Cdc42 alteration. Supporting this hypothesis, subjects with schizophrenia exhibited altered Cdc42-pathway-related transcriptional alteration in cortical layer 3 neurons [56].
Our results suggest the involvement of Cdc42 in FTLD, particularly in bvFTD subtype. The identification of this specific biomarker to predict the underlying pathology represents a potential clinically relevant tool to be further validated.
