Abstract
Herein, we performed a gene expression profiling in a cohort of 10 mild cognitive impairment (MCI), subdivided, according to the analysis of cerebrospinal fluid biomarkers, in prodromal Alzheimer’s disease (AD) and non-AD MCI, as compared with 27 AD patients and 24 controls, in order to detect early gene expression alterations. We observed a significant upregulation of insulin (INS) and INS Receptor (INSR) expression levels in AD both prodromal and fully symptomatic, as compared with controls, but not in MCI subjects. Our results suggest an early dysregulation of INS and INSR in AD pathogenesis and pave the way to a possible utility of these transcripts as peripheral biomarkers.
INTRODUCTION
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that slowly destroys memory and thinking skills. Currently, there is an ever-growing need to find biomarkers for the early diagnosis of AD as well as to predict disease progression, especially as the number of persons affected by this disease is nearing approximately 46 million worldwide [1]. At the histopathological level, AD is defined by the presence of senile plaques and neurofibrillary tangles. Current scientific evidence suggests that in preclinical AD, brain changes may begin years before symptoms, and the transition between normal cognition and full-blown dementia is represented by mild cognitive impairment (MCI) [3]. Although a few pathogenic mechanisms have been suggested, such as inflammation and oxidative damage, an extensive study of transcripts deregulated in AD pathogenesis is currently lacking. Moreover, the identification of early changes occurring in peripheral cells may be also useful to identify potential peripheral biomarkers to be included in the diagnostic routine. At present, the use of the cerebrospinal fluid (CSF) biomarkers amyloid-β (Aβ), tau, and phosphorylated tau (Ptau) allows the discrimination between MCI due to AD, i.e., prodromal AD, from MCI due to other causes (non-AD MCI) with a very high accuracy (see [4] for review), but implies an invasive procedure. Therefore, peripheral, easy to obtain, and possibly cheap, biomarkers, would be needed in clinical settings. In addition, they could be of help in the selection of patients for pharmacological trials with new drugs, or to monitor the biological effect of therapies [5]. Thus, comprehensive research is being performed to identify better and useful peripheral biomarkers for the early detection of AD with the hope that an effective therapeutic window will emerge.
In this study, we performed gene expression profiling of 84 candidate genes in a cohort of MCI, including prodromal AD (i.e., MCI with altered biomarkers in CSF) and non-MCI AD as compared with full-blown AD patients and controls, in order to detect early gene expression alterations.
POPULATION AND METHODS
Population
Ten MCI (five of which were diagnosed with prodromal AD), 27 AD, and 4 frontotemporal dementia (FTD) subjects were recruited at the Alzheimer Unit of the Fondazione Cà Granda, IRCCS Ospedale Maggiore Policlinico, University of Milan (Milan). All patients underwent a standard battery of examinations, including medical history, physical and neurological examination, screening laboratory tests, neurocognitive evaluation, and imaging. The Clinical Dementia Rating, the Mini-Mental State Examination (MMSE), the Frontal Assessment Battery, the Wisconsin Card Sorting Test, and the Tower of London test assessed cognitive dysfunctions. The presence of significant vascular brain damage was excluded (Hachinski Ischemic Score <4). Lumbar puncture was performed after one-night fasting. The diagnosis of non-AD MCI, prodromal AD, and AD was made according to current research criteria [6], whereas FTD patients were diagnosed according to current consensus criteria [7] and subsequent revision [8]. The control group consisted of 24 non-demented volunteers matched for ethnic background and age, without memory and psycho-behavioral dysfunctions (MMSE≥28). There were no significant differences in terms of mean age (years±SEM) between patients and controls (p = 0.1697). Informed consent to participate in this study was given by all subjects or their caregivers. The study protocol had been previously approved by the local Ethics Committee.
Characteristics of patients and controls are summarized in Table 1.
Characteristics of patients and controls
SEM, Standard Error of Mean.
CSF processing and biomarker determination
CSF samples were obtained in polypropylene tubes by lumbar puncture at the L4/L5 or L3/L4 interspace, centrifuged at 4°C at 2,000×g. The serum and CSF samples were removed and dispensed in aliquots of 400 μl into cryo-tubes. Specimens were stored at –80°C until use. Aβ42, tau, and P-tau CSF levels were determined with human specific ELISA kits (Innogenetics), as previously reported [9].
Total mRNA isolation from peripheral blood mononuclear cells (PBMCs)
For each subject, 14 ml of blood were collected in two BD VacutainerR CPTTM (1 ml NC, 2 ml Ficoll) as previously described [10]. From each tube, PBMCs were separated by gradient centrifugation and total RNA extracted with the single step acid phenol method, using Trizol (Invitrogen). RNA purity was measured by optical density and only samples with an OD 260/280 ratio ranging from 1.8 to 2 and an OD 260/230 of 1.8 or greater were used.
Screening of AD gene expression pathway by PCR array
RNA was retrotranscribed with RT2 First Strand Kit (SABiosciences), according to the instruction of the manufacturer. For Real Time PCR experiments, the Alzheimer’s disease Pathway PCR Array (PAHS-057Z, see Supplementary Material) was used and runs were performed in an Applied BioSystems StepOne Plus system. The array profiles the expression of 84 key genes involved in the onset, development, and progression of AD. These arrays included also five housekeeping genes (ACTB, B2M, G3PDH, HPRT1, and RPLP0) for the proper normalization of the data, mRNA reverse transcription control and a positive PCR control (Supplementary Material).
Validation of best hits by Taqman qRT-PCR assays
Total RNA was retrotranscribed by using SuperScript III First Strand (Invitrogen), according to the instruction of the manufacturer. For a quantitative estimate of mRNA levels, an ABI 7500 Fast Sequence Detector with dual-labelled TaqMan probes has been used. Specific TaqMan probes have been employed (details available in the Supplementary Material) and the relative amount of candidate genes mRNA determined by comparison with the housekeeping 18Sr and GADPH RNA probes (Hs99999901_s1, Hs99999905_m1, ThermoFisher Scientific) as previously described [11].
Statistical analysis
The SABiosciences PCR Array data analysis was based on ΔΔCt method with normalization of the raw data to housekeeping genes (using the software available at http://www.sabiosciences.com/pcarraydataanalysis.php). p values were calculated based on a Student’s t-test of the replicate 2∧ (- Delta Ct) values for each gene in the control and FTD groups. Best hits were chosen based on statistical significance (p < 0.050). Comparisons between INS and INSR gene expression levels in the different groups were performed using a two-sided nonparametric t test (GraphPad Prism, La Jolla, CA).
RESULTS
Gene expression profile of 84 key genes involved in the onset, development, and progression of AD was carried out in peripheral cells from 10 MCI (of which 5 with non-MCI AD and 5 with prodromal AD), 7 AD, 4 FTD patients, and 4 healthy controls. We observed a generalized upregulation of gene expression levels in AD patients as compared with controls (Fig. 1C) and FTD patients (Fig. 1D).
Heat maps: MCI (A), prodromal AD (B), AD (C), and FTD (D) versus controls. Data are expressed as fold change (fold difference) and each square represents a gene of the AD pathogenic pathway. Green indicates downregulation, red upregulation.
Significantly increased expression levels of insulin and insulin receptor (INS and INSR, respectively) was shown in both prodromal AD and AD patients as compared with controls (39.20 and 3.47-fold increase over controls in AD; 8.44 and 1.49 in prodromal AD; respectively, p < 0.05), but not in non-AD MCI subjects (5.38 and –1.68-fold regulation over controls, p > 0.05) and FTD patients (Fig. 1A-D). These results were then validated with Real Time PCR in 20 AD patients and 20 healthy subjects. Upregulation was confirmed for both genes: INS mean relative expression levels±SEM in AD versus controls: 1.556±0.2683 versus 0.4256±0.1273, p = 0.003; INSR in AD versus controls: 3.545±0.5422 versus 2.089±0.3583; p = 0.0372 (Fig. 2A). Stratifying according to gender, the gene expression upregulation was confirmed for INS gene: mean relative expression levels±SEM in female AD patients versus controls, 1.588±1.32 versus 0.44±0.18, p = 0,004; mean relative expression levels±SEM in male AD patients versus controls, 1.45±0.44 versus 0.39±0.18, p = 0.04 (Fig. 2B,C).
A) Histograms of INS (***p = 0.003) and INSR (*p = 0.0372) gene expression levels in AD patient compared to controls in the validation population (data represented as mean±SEM). B,C) Histograms of INS (in females ***p = 0.004 and in males *p = 0.04) and INSr (p > 0.05) gene expression levels in patients compared to controls stratifying according to the gender.
A trend toward increased expression levels (expressed as fold increase over controls) of the following genes was shown: Acetylcholinesterase (ACHE: 5.56 in non-AD MCI, 10.34 in prodromal and 20.5 in AD, respectively), gamma transducing activity polypeptide 1 (GNGT1: 3.3964 in MCI, 6.6036 in prodromal and 23.925 in AD patients, respectively), Low density lipoprotein receptor-related protein 8 (LRP8: 3.6451 only in AD patients), Neurotrophic tyrosine kinase, receptor, type 2 (NTRK2: 5.04 in MCI,13.45 in prodromal and 34.93 in AD patients, respectively), Serpin peptidase inhibitor 3 (SERPINA3: 6.5679 in MCI, 11.25 in prodromal and 63.38 in AD patients, respectively).
DISCUSSION
Herein, we showed that INS and INSR are overexpressed in PBMCs from patients with prodromal AD and increase with the progression of the disease to full blown dementia. Clinical AD begins with the early and mild dementia stage named MCI and within an average of 10 years gradually progresses to moderate and later to severe AD [1]. Due to the lack of sufficient knowledge on the mechanisms which initiate and drive the development of AD, patients affected by this disease are treated only with symptomatic therapies. The lack of early AD biomarkers, preferably from the periphery, greatly hinders the treatment of AD, which is usually introduced only in quite late stages. Moreover, biomarkers, which would help in monitoring the AD progression and response to therapies are also missing [2]. Therefore, there is an intensive, ongoing search for drugs and for biomarkers, preferably from the accessible tissues such as blood. In order to improve this field of research, we carried out a gene expression profile in peripheral cells of AD patients at different stage of disease.
We observed a generalized upregulation of gene expression levels that correlates with the disease progression. In particular, we observed, and then validated in a larger population, a statistically significant upregulation of INS and INSR gene expression levels. Stratifying according to gender, the upregulation was confirmed both in females and males for INS gene, suggesting that this effect is disease-specific and not conditioned by gender status. This interesting data, because the link between insulin deregulation pathway and AD pathogenesis have been an extremely productive area of current research [12]. Notably, our cohort of patients had normal plasma glucose levels (<126 mg/dl), suggesting that these results are not influenced by an impairment of insulin related mechanisms. Insulin is one of the key hormone regulators of metabolism throughout the body, through a variety of largely tissue-specific actions. For a long time, the brain was considered an insulin independent organ, but radioimmunoassays showed high levels of insulin in brain extracts [13]. Insulin in the brain is predominantly shuttled across the blood-brain barrier through specific insulin receptors or it can be actively transported into the brain via an endocytic-exocytic mechanism [14]. Moreover, insulin can be produced de novo in different brain regions such as frontal cortex and hippocampus [15]. Recent studies show that functional insulin signaling components in forebrain regions may exert a neuroprotective role in areas responsible for various functions of memory [16]. Given that, hyperinsulinemia and impaired glucose metabolism are associated with increased risk of AD [12], but it still unclear exactly how and at what stage this imbalance interact with the disease. The biological actions of insulin are mediated by the INSR gene, which encodes a member of the receptor tyrosine kinase family of proteins. The preproprotein is processed to generate alpha and beta subunits that form a heterotetrameric receptor. Binding of insulin to this receptor activates the insulin-signaling pathway, which regulates glucose uptake and release, as well as the synthesis and storage of carbohydrates, lipids, and protein.
According with our results, several recent studied demonstrated that the expression of INSR is significantly higher in red blood cells as well as in brain of late AD patients [17, 18]. This may reflect an overall upregulation of INSR expression caused by the relative insulin deficiency in the central nervous system. Moreover, we observed that this alteration already appears in the early stages of the disease, suggesting a possible future usefulness of INS and INSR expression levels as peripheral biomarkers for early AD. Nevertheless, further and larger cohort studies are required to confirm these preliminary data.
DISCLOSURE STATEMENT
Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/17-0861r1).
