Acyl-Alkyl-Phosphatidlycholines are Decreased in Saliva of Patients with Alzheimer’s Disease as Identified by Targeted Metabolomics

Abstract

Diagnosis of Alzheimer’s disease (AD) is still a challenge. Salivary analysis could produce an easily accessible and inexpensive possibility to study metabolic changes in AD. In the present pilot study, we show for the first time using targeted metabolomics that acyl-alkyl phosphatidylcholines (PCae C34:1-2; PCae C36:1-2-3; PCaeC38:1c3; PCae C40:2-3) are significantly reduced in saliva of AD patients (n = 25) compared to healthy controls (n = 25). Saliva levels of PCae C36Λ1-2-3) were also decreased in patients with mild cognitive impairment (n = 25). No changes were seen for saliva diacyl-phosphatidylcholines, lyso-acyl-phosphatidylcholines, and sphinogomyelins. These data suggest specific lipid changes in the saliva of AD patients, thus salivary measures could establish new biomarkers. However, these preliminary results have to be established in larger scale studies.

Keywords

Alzheimer’s disease biomarkers diagnosis metabolomics phosphatidylcholines saliva

INTRODUCTION

The life expectancy of humans has increased within the last 100 years. As age is the main risk factor for Alzheimer’s disease (AD), the number of patients suffering from AD will dramatically increase within the next 50 years so that about 80 million AD patients can be expected worldwide by 2050. These enormously high numbers of presumed AD patients call for further establishment of reliable diagnostic surrogate markers for diagnosing and monitoring disease and therapy progression. A valid and easily accessible diagnostic procedure should be the basis for treatment. Definitive diagnosis of AD requires both a clinical diagnosis of the disease and postmortem detection of AD pathologies. A probable diagnosis of AD can be established with a confidence of approximately >90% based on clinical criteria including medical history, physical examination, laboratory tests, neuroimaging, and neuropsychological evaluation. A promising area of research for biochemical diagnosis of AD is the analysis of cerebrospinal fluid (CSF), where measurement of amyloid-β (Aβ₄₂), total tau, and phospho-tau-181 can distinguish AD patients from healthy subjects with high specificity and sensitivity [1 –3]. Unfortunately, the collection of CSF is an invasive procedure. Thus, there is a considerable need for discovering biomarkers in other body fluids. There is strong evidence that metabolomics can discover putative biomarkers in blood [4 –6], but so far none of the potential blood biomarkers described in AD have proven to be useful in clinical practice. To date, it seems likely that the combination of several biomarkers derived from blood will successfully define a patient-specific biomarker signature [1].

Early, fast, and inexpensive diagnosis from body fluids using modern ultrasensitive methods will be extremely important in the future for the differentiation of AD from other forms of dementia and to measure therapeutic effects on disease progression. To establish biomarkers in other human body fluids, such as urine or saliva would facilitate the collection of samples. Saliva is a human fluid which is easily accessible, but is not yet investigated as a source for biomarkers in AD. The saliva proteome contains approximately 2,300 proteins and 27% of them are identical to plasma proteins [7, 8]. Several saliva proteins have been identified as biomarkers for testing diseases like cancer, HIV, autism, and fibromyalgia [7]. A very recent paper claims that saliva Aβ₄₂ can distinguish AD from controls [9]. A few very recent pilot studies reported on the discovery of potential biomarkers in saliva of AD patients using untargeted ¹H NMR based metabolomics [10, 11]. Dame et al. [8] characterized the human saliva metabolome and identified 308 salivary metabolites.

The aim of the present study was to apply targeted metabolomics to characterize the metabolic phenotype in saliva samples of healthy controls and AD patients. Targeted metabolomics is the quantitative or semi-quantitative measurement of defined groups of chemically characterized and biochemically annotated metabolites. We will analyze 163 metabolites, including phosphatidylcholines (PCs) or lysophosphatidylcholines (lysoPCs), and sphingomyelins (SM) using the Absolute IDQ ^® p150 Kit (BIOCRATES Life Sciences AG) that have been demonstrated to produce useful signatures of AD [12]. Thus, this pilot study will reveal whether there is a specific lipid change in AD and mild cognitive impairment (MCI), and whether saliva analysis has the potential to detect and establish further biomarkers for dementia.

MATERIAL AND METHODS

Patients

Cognitively healthy subjects and patients suffering from AD were recruited at Hall/Tirol State Hospital, Austria. In this study 75 subjects were included in total, 25 healthy controls, 25 AD patients, and 25 patients with MCI. The diagnostic procedure including laboratory testing as well as thorough neuropsychological assessments have been described by us in detail elsewhere [4 , 12–14]. A panel including a neurologist, psychiatrist, and neuropsychologist examined all clinical and diagnostic features. All study participants were also assessed by magnetic resonance imaging using a 1.5 Tesla Siemens Symphony MRI scanner with a T1-weighted FLASH 3D sequence. The medial temporal lobe atrophy (MTA) score is based on a visual estimation of the volume of the medial temporal lobe [15]. The MTA score ranges from 0 (no atrophy) to 4 (severe atrophy). The Clinical Dementia Rating Scale (CDR) was obtained through semistructured interviews of patients and informants, and cognitive functioning is rated in 6 domains of functioning [16]: memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care. Each domain is rated on a 5-point scale of functioning as follows: 0, no impairment; 0.5, questionable impairment; 1, mild impairment; 2, moderate impairment; and 3, severe impairment (personal care is scored on a 4-point scale without a 0.5 rating available). Exclusion criteria for healthy subjects and patients suffering from AD included other psychiatric or neurological diseases or diseases including cancer, vascular diseases, or other diseases with clinically significant hepatic, renal, pulmonary, metabolic, or endocrine disturbances, and inflammation. Especially excluded were participants with significant small vessel cerebrovascular disease as well as large vessel strokes. The study was approved by the Ethics Committee of the Medical University of Innsbruck and was performed in accordance with the Helsinki Declaration. All subjects gave written informed consent.

Collection of Saliva

Collection was done in the early morning and all subjects were asked to refrain from eating, drinking, smoking, or using oral hygiene prior to saliva collection. Saliva samples were obtained by spitting into a 15 ml sterile falcon tube for 2 min. The average volume collected was 1–2 ml. The samples were sent to the laboratory at room temperature, processed within 3-4 hours, and frozen at –80°C until analysis. At the day of analysis, saliva was thawed, centrifuged at 14,000 g for 5 min and then analyzed.

Targeted Metabolomic Analysis

The endogenous metabolites were analyzed with a targeted quantitative and quality-controlled metabolomics approach using the Absolute IDQ ^® p150 Kit (BIOCRATES Life Science AG, Innsbruck, Austria). This validated assay allows the comprehensive identification and quantification of 163 endogenous metabolites, including 40 acylcarnitines (Cx:y), hydroxylacylcarnitines (C(OH)x:y), and dicarboxylacylcarnitines (Cx:y-DC), hexose (H1), 14 amino acids, 15 sphingomyelins (SM x:y) and hydroxysphingomyelins (SM(OH) x:y), 77 phosphatidylcholines (PC aa = diacyl x:y, PC ae = acyl-alkyl x:y), and 15 lysophosphatidylcholines (lysoPC a x:y). A targeted analysis for these metabolites is performed by applying flow injection analysis-tandem mass spectrometry (FIA-MS/MS) acquisition with multiple reaction monitoring (MRM) in positive and negative electrospray ionization mode. Quantification is achieved with internal standards. The Absolute IDQ ^® p150 Kit, a 96-well plate format assay, was performed as described in the manufacturer’s instructions and as reported by us previously [12]. In short, 10μL of sample mixture was pipetted onto filter spots suspended in the wells of a 96-well filter plate. The filter plate was fixed on top of a deep-well plate serving as a receiving plate for the extract later on, i.e., a combi-plate structure. After drying under a nitrogen stream, 50μl of a 5% phenylisothiocyanate solution was added to enable derivatization of amino acids. After 20 min of shaking and nitrogen drying, 300μL of 5 mM ammonium acetate in methanol was added to the wells. After 30 min of incubation, the combi-plate was centrifuged to move the extracts into the lower receiving deep-well plate, which was then detached from the upper filter plate. After adding another 300μL of 5 mM ammonium acetate in methanol to the extracts and briefly shaking, the plate was placed in the autosampler of the FIA-MS/MS system for analysis. The FIA-MS/MS system consisted of a Knauer K-1001 LC pump (Knauer, Berlin, Germany), a CTC-PAL HTS9 autosampler (CTC Analytics AG, Zwingen, Switzerland), and a QTrap 3200 mass spectrometer (Sciex, Toronto, Canada). The injection volume was 30μl. The flow rate was set to 30μl/min. Metabolite concentrations (μM) were automatically calculated by the MetIDQ software package part of the Absolute IDQ ^® p150 Kit.

Statistical Analysis

Statistical analysis was performed by One Way ANOVA with a subsequent Dunnett or Tukey posthoc test, where p < 0.05 was considered significant.

RESULTS

Seventy-five individuals were included in this pilot study, 25 controls, 25 AD patients, and 25 MCI patients. Patient characteristics are shown in Table 1. Age, sex, education, and Geriatric Depression Scale scores were not statistically different between all 3 groups. AD patients had a significantly lower Mini-Mental State Examination score (Table 1). CDR and MTA scores were significantly higher in MCI and AD patients (Table 1). No changes were observed in plasma cholesterol, triglyerides and statin use (Table 1). Out of the 163 targeted metabolites, only 59 were clearly detectable with concentration levels above 100 nM in control saliva. Out of these 59, only 9 metabolites showed concentrations above 1μM in control saliva.

Table 1

Characteristics of study participants

	Controls	MCI	AD
n	25	25	25
Sex (Female/Male)	16/9	16/9	17/8
Age (y)	74.8±4.4	75.9±8.8	80.4±7.2
Education (y)	11.6±2.8	12.2±3.0	10.2±3.0
MMSE	28.1±1.8	26.6±2.7	17.7±5.7^***
GDS	2.3±0.5	2.7±0.7	3.2±0.9
CDR	0±0	0.5±0.0^***	2.0±0.6^***
MTA score	0.5±0.5	1.3±0.5^***	2.9±0.7^***
Cholinesterase inhibitors	0	0	21/25
Cholesterol (mg/dL)	207±37	186±26	212±37
Triglyceride (mg/dL)	161±89	140±75	156±78
Statin Use	4/25	4/25	5/25

Data are expressed as mean±standard deviation; n, number of participants; AD, Alzheimer’s disease; CDR, Clinical Dementia Rating; GDS, Geriatric Depression Scale; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; MTA score, medial temporal lobe atrophy score. Statistical analysis was performed by One Way ANOVA with a Tukey HSD posthoc Test against controls, where p < 0.05 represents significance (^***p < 0.001).

The analysis of saliva acyl-alkyl-phosphatidylcholines (PCae) is given in Table 2. There was a clear decrease of many PCae metabolites in MCI and AD patients compared to controls, but only PCae C36:2 and PCae C36:3 were significant. However, combining selected saliva PCae members, AD patients could be significantly differentiated from healthy controls: PCae C34:(1 + 2) p = 0.008; PCae C36:(1 + 2 + 3) p = 0.0011; PCae C38:(1 + 3) p = 0.009; and PCae C40:(2 + 3) p = 0.011. Interestingly, saliva PCae C36:(1 + 2 + 3) levels also differentiated MCI patients from controls (p = 0.04). When all those above-mentioned lipid metabolites were combined, AD could be differentiated from controls with a high significance of p < 0.0001 and MCI from controls with a significance of p = 0.01.

Table 2

Quantitative analysis of saliva acyl-alkyl-Phosphatidylcholines (PC-ae) in healthy controls (control) and patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI)

	Control	MCI	AD
30:0	207±54	361±80	131±23
32:1	244±70	154±47	94±34 (p = 0.08)
34:0	263±75	198±62	100±33
34:1	1163±374	697±194	427±130 (p = 0.07)
34:2	808±246	547±153	288±94 (p = 0.07)
34:3	212±67	162±56	70±21
36:1	715±196	441±116	327±69 (p = 0.08)
36:2	752±233	428±133	251 ± 64 (p = 0.049)
36:3	312±99	171±49	93 ± 24 (p = 0.03)
36:4	255±92	168±57	83±25
38:1	128±35	77±20	54±16 (p = 0.07)
38:2	136±41	80±20	65±19
38:3	142±43	77±18	59±14 (p = 0.07)
38:4	212±74	124±36	86±25
38:5	205±64	119±43	82±22
40:0	132±20	113±16	117±27
40:2	152±47	84±24	60±16 (p = 0.08)
40:3	105±29	60±16	46±15 (p = 0.08)
42:2	143±49	65±17	58±15
42:3	135±42	90±25	58±15
34:(1 + 2)	985±223	622±123	358 ± 80 (p = 0.008)
36:(1 + 2 + 3)	593±108	346 ± 62 (p = 0.04)	224 ± 34 (p = 0.0011)
38:(1 + 3)	135±27	77±13 (p = 0.06)	57 ± 10 (p = 0.009)
40:(2 + 3)	128±27	72±14 (p = 0.07)	53 ± 11 (p = 0.011)
SUM all^§	475±65	287 ± 37 (p = 0.01)	178 ± 22 (p < 0.0001)

Values are given as mean±SEM nM; the number of patients was n = 25 per group. Statistical analysis was performed by ANOVA with a subsequent Dunnett posthoc test (p < 0.05 is defined as significant). ^§SUM all = PCae C34(1 + 2) + PCae C36(1 + 2 + 3) + PCae C38(1 + 3) + PCae C40(2 + 3).

The analysis of saliva PC aa metabolites is given in Table 3 and shows a not-significant decrease tendency between controls, MCI, and AD patients. The analysis of saliva lysoPC metabolites is given in Table 4 and shows no changes between controls and patients with MCI and AD, but a non-significant decrease in lyso-PCa C18:1. The analysis of saliva sphingomyelins and hydroxylated-sphingomyelins is given in Table 5. None of the metabolites were altered in MCI and AD patients compared to controls.

Table 3

Quantitative analysis of saliva diacyl-Phosphatidylcholines (PC-aa) in healthy controls (Co) and patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI)

	Healthy Controls (Co)	MCI	AD
28:1	111±2	86±19	50±10 (p = 0.08)
30:0	711±192	804±246	414±125
32:0	1051±246	962±294	449±106
32:1	314±102	217±59	223±109
34:1	2730±791	1490±321	1562±519
34:2	1496±429	1030±230	962±305
34:3	125±38	123±28	90±27
36:0	156±33	104±30	94±24
36:1	1391±380	962±227	733±216
36:2	1826±525	1138±265	978±293
36:3	802±235	534±120	465±138
36:4	395±111	204±47	240±57
38:0	112±27	106±28	60±13
38:3	271±73	178±47	141±30
38:4	240±63	119±26	187±53
38:5	103±34	51±13	65±17
38:6	113±32	66±17	47±11 (p = 0.06)

Values are given as mean±SEM nM; the number of patients was n = 25 per group. Statistical analysis was performed by ANOVA with a subsequent Dunnett posthoc test. None of the analytes were significantly different.

Table 4

Quantitative analysis of saliva Lyso-acyl-Phosphatidylcholines (lyso-PC-a) in healthy controls (Co) and patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI)

	Healthy Controls (Co)	MCI	AD
C16:0	305±82	266±94	194±55
C18:0	328±118	284±92	163±49
C18:1	700±212	398±101	286±63 (p = 0.06)
C18:2	224±67	191±68	223±68
C20:3	354±57	247±55	304±32

Table 5

Quantitative analysis of saliva sphingomyelins (SM) and hydroxylated sphingomyelins (SM(OH)) in healthy controls (Co) and patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI)

	Healthy Controls (Co)	MCI	AD
SM C16:0	7371±1899	6279±1891	3569±904
SM C16:1	241±63	187±55	125±37
SM C18:0	659±183	586±154	419±129
SM C18:1	134±43	103±25	64±18
SM C24:0	1334±322	1189±274	929±301
SM C24:1	2102±628	1707±455	1132±257
SM C26:0	292±78	391±110	239±66
SM C26:1	154±41	152±43	148±66
SM(OH) C16:1	222±63	156±35	131±35
SM(OH) C22:1	274±74	249±62	162±27
SM(OH) C22:2	154±43	110±32	87±23
SM(OH) C24:1	171±41	278±60	155±41

DISCUSSION

Saliva is the fluid that bathes the mouth and oral cavity and is made and secreted from the salivary glands (parotid, submandibular, sublingual, and minor salivary glands). Oral fluid further contains non-salivary components (gingival crevicular fluid, nasal and bronchial secretions, serum and blood derivatives from wounds, desquamated epithelial linings, food components, and micro-organisms in the oral cavity [7, 17]. Whole saliva is composed of water, peptides, and proteins, including hormones and enzymes, sugars, lipids, and electrolytes. An average person generates between 0.75–1.5 L of saliva per day [8]. In the present study, we show for the first time that concentrations of acyl-alkyl phospatidylcholines are significantly reduced in saliva of AD and MCI patients. Thus, saliva is a very powerful and easily accessible human fluid for diagnostic purposes; however, unlike CSF and plasma, composition of saliva rapidly changes in response to biological stimuli.

There are only very few publications dealing with saliva in diagnosis of AD. So far only 6 publications are found in PubMed when entering “saliva [Title] and Alzheimer”. In 2008, Li et al. [18] showed the relationship between saliva MHPG and mental health in an elderly general population. In the same year, Boston et al. [19] developed a simple laboratory test for AD by measuring acetylcholinesterase in saliva. In 2010, Bermejo-Pareja et al. [20] reported for the first time that saliva Aβ₄₂ could become a potential biomarker, which was recently reproduced by Lee et al. [9]. All AD cases secreted more than double of Aβ₄₂ in saliva as controls [9]. However, so far this has not entered clinical routine. It is expected that a neurodegenerative condition, such as AD involving widespread damage to the brain, will involve membrane dysfunction.

The analysis of the saliva metabolome is a very recent and very potent sensitive technique. Already in 2013, Tsuruoka et al. [21] showed that arginine and tyrosine were significantly different between dementia patients and controls using CE-TOF-MS. A very recent paper [10] used ¹H NMR metabolomics and identified 22 salivary metabolites, including galactose, imidazole, acetone, creatine, and 5-aminopentanoate, that were useful for distinguishing between AD, MCI, and healthy patients. In another study using liquid chromatography-mass spectrometry for untargeted metabolomics, saliva from 256 patients with AD and 218 age-matched healthy controls was analyzed [22]. They identified six metabolites (sphinganine-1-phosphate, ornithine, phenylacetic acid, inosine, 3-dehydrocarnitine, and hypoxanthine) that were significantly different in AD patients compared to controls. In the present study, we show now for the first time using targeted metabolomics that acyl-alkyl phosphatidylcholines were significantly decreased in saliva of AD and MCI patients. Our data show that single lipid metabolites only give tendency for being decreased. However, by combining selected lipid metabolites (PCae C34, C36, C38, and C40), AD patients could be significantly differentiated from healthy controls, Further, the sum of PCae C36:1-2-3 also differentiated MCI patients from controls. More importantly the sum of those selected lipids (PCae C34:1-2; PCae C36:1-2-3; PCaeC38:1–3; PCae C40:2-3) highly significantly differentiated MCI and AD patients from healthy controls. This is a novel finding, but supports previous studies showing profound metabolic alterations in the saliva of AD.

Indeed, multiple studies have found reductions in various classes of phospholipids including phosphatidylcholines, phosphatidylinositols, or phosphatidylethanolamines in human plasma of AD patients compared with healthy controls [6 , 23–26]. Phospholipids are the principal membrane-forming lipid family and influence many complex cell processes including trafficking or modulation of membrane proteins. The changes in lipid metabolism presume that lipid perturbations are involved in AD pathology. In this study, we cannot provide an explanation why lipids are altered in saliva of AD and MCI patients, but we may speculate that the decreased saliva lipid metabolites are caused by 1) enhanced enzymatic degradation of lipids in saliva, 2) a reduced secretion of lipids from the salivary glands, or 3) a markedly increased lipid turnover in saliva and salivary glands. We provide evidence that lipid disturbances are prominent in AD and MCI and that an altered saliva turnover reflects dementia. Definitely, lipid changes could be potential biomarkers to identify AD patients; however, in future studies salivary measures have to be compared with blood and CSF as the main fluid compartments of interest for biomarker research. Much more work is necessary, especially in analyzing the enzymes which affect the lipid metabolism and recent evidence also suggests that the microbiome may be affected in AD and thus may indirectly influence lipid metabolisms in saliva.

There are several limits in this study. 1) This is a pilot study with a small number of participants. 2) The applied analysis and processing technique is different to previous studies, which may limit the comparison of our findings to previous studies. However, we collected saliva samples following a standardized protocol. We have collected saliva early in the morning and thus can exclude any diurnal rhythms. However, we cannot exclude diurnal metabolomic alterations of lipids, e.g., acetate, propionate or isobutyrate or urea display altered concentrations depending on the collection. 3) Although we excluded exogenous factors, such as eating, drinking, or oral hygiene, we cannot exclude that medications, small oral bleeding, or other factors may influence the analysis.

Taken together, we show that acyl-alkyl phosphatidylcholines are decreased in saliva of AD and MCI patients. The alteration was statistically highly significant and may give rise for being useful as a diagnostic marker, especially the sum of (PCae C34:1-2; PCae C36:1-2-3; PCaeC38:1–3; PCae C40:2-3).

Footnotes

ACKNOWLEDGMENTS

This study was supported by Österreichische Nationalbank Jubiläumsfonds (Nr. 15887). We thank Karin Albrecht and Monika Greil for assistance with saliva processing.

Authors’ disclosures available online ().

References

Humpel

(2011) Identifying and validating biomarkers for Alzheimer’s disease. Trends Biotechnol 29, 26–32.

Blennow

(2005) CSF biomarkers for Alzheimer’s disease: Use in early diagnosis and evaluation of drug treatment. Expert Rev Mol Diagn 5, 661–672.

Blennow

, Hampel

, Weiner

, Zetterberg

(2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6, 131–144.

Klavins

, Koal

, Dallmann

, Marksteiner

, Kemmler

, Humpel

(2015) The ratio of phosphatidylcholines to lysophosphatidylcholines in plasma differentiates healthy controls from patients with Alzheimer’s disease and mild cognitive impairment. Alzheimers Dement (Amst) 1, 295–302.

Mapstone

, Cheema

, Fiandaca

, Zhong

, Mhyre

, MacArthur

, Hall

, Fisher

, Peterson

, Haley

, Nazar

, Rich

, Berlau

, Peltz

, Tan

, Kawas

, Federoff

(2014) Plasma phospholipids identify antecedent memory impairment in older adults. Nat Med 20, 415–418.

Wilkin

, Trushina

(2018) Application of metabolomics in Alzheimer’s disease. Front Neurol 8, 719.

Loo

, Yan

, Ramachandran

, Wong

(2010) Comparative human salivary and plasma proteomes. J Dent Res 89, 1016–1023.

Dame

, Aziat

, Mandal

, Krishnanurthy

, Bouatra

, Borzouie

, Guo

, Sajed

, Deng

, Lin

, Liu

, Dong

, Wishart

(2015) The human saliva metabolome. Metabolomics 11, 1864–1883.

Lee

, Gio

, Kennedy

, McGeer

(2017) A method for diagnosing Alzheimer’s disease based on salivary amyloid-beta protein 42 levels. J Alzheimers Dis 55, 1175–1182.

10.

Yilmaz

, Geddes

, Han

, Bahado-Singh

, Wilson

, Imam

, Maddens

, Graham

(2017) Diagnostic biomarkers of Alzheimer’s disease as identified in saliva using 1H NMR-based metabolomics. J Alzheimers Dis 58, 355–359.

11.

Figueira

(2016) NMR analysis of the human saliva metabolome distinguishes dementia patients from matched controls. Mol Biosyst 12, 2562–2571.

12.

Oberacher

, Arnhard

, Linhart

, Diwo

, Marksteiner

, Humpel

(2017) Targeted metabolomic analysis of soluble lysates from platelets of patients with mild cognitive impairment and Alzheimer’s disease compared to healthy controls: Is PC aeC40:4 a promising diagnostic tool? J Alzheimers Dis 57, 493–504.

13.

Hochstrasser

, Ehrlich

, Marksteiner

, Sperner-Unterweger

, Humpel

(2012) Matrix metalloproteinase-2 and epidermal growth factor are decreased in platelets of Alzheimer patients. Curr Alzheimer Res 9, 982–989.

14.

Marksteiner

, Humpel

(2013) Platelet-derived secreted amyloid-precursor protein-β as a marker for diagnosing Alzheimer’s disease. Curr Neurovasc Res 10, 297–30.

15.

Scheltens

, Leys

, Barkhof

, Huglo

, Weinstein

, Vermersch

, Kuiper

, Steinling

, Wolters

, Valk

(1992) Atrophy of medial temporal lobes on MRI in ‘probable’ Alzheimer’s disease and normal ageing: Diagnostic value and neuropsychological correlates. J Neurol Neurosurg Psychiatry 55, 967–972.

16.

Hughes

, Berg

, Danziger

, Coben

, Martin

(1982) A new clinical scale for the staging of dementia. Br J Psychiatry 140, 566–572.

17.

Kaufman

, Lambster

(2002) The diagnostic application of saliva - a review. Crit Rev Oral Biol Med 13, 197–212.

18.

, Watanabe

, Kunitake

, Sugataka

, Muraoka

, Kojima

, Kawashima

, Yamada

(2008) Relationship between saliva level of 3-methoxy-4-hydroxyphenylglycol and mental health in the elderly general population. Psychiatry Clin Neurosci 62, 562–567.

19.

Boston

, Gopalkaje

, Manning

, Middleton

, Loxley

(2008) Developing a simple laboratory test for Alzheimer’s disease: Measuring acetylcholinesterase in saliva - a pilot study. Int J Geriatr Psychiatry 23, 439–440.

20.

Bermejo-Pareja

, Antequera

, Vargas

, Molina

, Carro

(2010) Saliva levels of Abeta1-42 as potential biomarker of Alzheimer’s disease: A pilot study. BMC Neurol 10, 108.

21.

Tsuruoka

, Hara

, Hirayama

, Sugimoto

, Soga

, Shankle

, Tomita

(2013) Capillary electrophoresis-mass spectrometry-based metabolome analysis of serum and saliva from neurodegenerative dementia patients. Electrophoresis 34, 2865–2872.

22.

Liang

, Liu

, Zhang

, Jiang

, Xing

, Zhang

A-H

(2015) Metabolomics-based screening of salivary biomarkers for early diagnosis of Alzheimer’s disease. RSC Adv 5, 96074–96079.

23.

Kosicek

, Hecimovic

(2013) Phospholipids and Alzheimer’s disease: Alterations, mechanisms and potential biomarkers. Int J Mol Sci 14, 1310–1322.

24.

Trushina

, Dutta

, Persson

, Mielke

, Petersen

(2013) Identification of altered metabolic pathways in plasma and CSF in mild cognitive impairment and Alzheimer’s disease using metabolomics. PLoS One 8, e63644.

25.

Whiley

, Sen

, Heaton

, Proitsi

, García-Gómez

, Leung

, Smith

, Thambisetty

, Kloszewska

, Mecocci

, Soininen

, Tsolaki

, Vellas

, Lovestone

, Legido-Quigley

(2014) AddNeuroMed Consortium. Evidence of altered phosphatidylcholine metabolism in Alzheimer’s disease. Neurobiol Aging 35, 271–278.

26.

Wong

, Braidy

, Poljak

, Pickford

, Thambisetty

, Sachdev

(2017) Dysregulation of lipids in Alzheimer’s disease and their role as potential biomarkers. Alzheimers Dement 13, 810–827.