Early Candidate Urine Biomarkers for Detecting Alzheimer’s Disease Before Amyloid-β Plaque Deposition in an APP (swe)/PSEN1 dE9 Transgenic Mouse Model

Abstract

Alzheimer’s disease (AD) is an incurable age-associated neurodegenerative disorder that is characterized by irreversible progressive cognitive deficits and extensive brain damage. The identification of candidate biomarkers before amyloid-β plaque deposition occurs is therefore of great importance for the early intervention of AD. Urine, which is not regulated by homeostatic mechanisms, theoretically accumulates changes associated with AD earlier than cerebrospinal fluid and blood. In this study, an APP (swe)/PSEN1dE9 transgenic mouse model was used to identify candidate biomarkers for early AD. Urine samples were collected from 4-, 6-, and 8-month-old transgenic mouse models, and the urinary proteomes were profiled using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). The levels of 29 proteins differed significantly between wild type and 4-month-old mice, which had not started to accumulate amyloid-β plaques. Among these proteins, 13 have been associated with the mechanisms of AD, while 9 have been suggested as AD biomarkers. Our results indicated that urine proteins enable detection of AD before amyloid-β plaque deposition, which may present an opportunity for intervention.

Keywords

Alzheimer’s disease early diagnosis urine proteome

INTRODUCTION

Alzheimer’s disease (AD) is a chronic age-associated neurodegenerative disorder associated with cognitive impairment and progressive dementia [1]. As the pathological process leading to AD begins decades before clinical symptoms, finding clues in the early stages of AD, especially before amyloid-β (Aβ) plaque deposition, is urgent in today’s AD research. Most candidate AD biomarkers came from cerebrospinal fluid (CSF) and blood studies; for example, tau levels increase in the CSF of most AD patients, and stable miRNAs in human serum are potentially valuable novel biomarkers for the diagnosis of AD. However, Aβ deposition can appear in AD patients, even in the stage 1 of preclinical AD, and can be detected by positron emission tomography (PET) amyloid imaging or CSF Aβ levels [1].

Biomarkers are measurable changes associated with disease. Unlike blood, which is stable because of homeostasis mechanisms, urine can accumulate many kinds of changes; some of these changes are associated with disease and will become biomarkers [2, 3]. In addition, urine is a resource for biomarker discovery that can be easily and non-invasively collected from AD patients. As a sensitive biomarker source, urine therefore might reflect pathological changes, especially in the early stages of disease [4]. However, whether time-course analyses of urine proteins at different AD phases can reveal sensitive biomarker clues before Aβ deposition is unclear, as urinary proteins are easily affected by complicated factors such as medicine and diet, especially in human samples. Using animal models is therefore the most efficient way to establish an association with disease, as animal models can reduce the influence of genetic and environmental factors on the urine proteome to the greatest degree [2]. In addition, animal models can represent the early stages of disease without clinical symptoms as it is impossible to clinically collect such samples [5].

Several studies have applied proteomic [6] and metabolomic [7, 8] analyses to characterize candidate biomarkers in mouse models. However, these studies used either plasma or brain tissues, and analyses of these materials were not sensitive enough for the early diagnosis of AD (before Aβ plaque deposition). Other studies have illustrated urine markers using samples from AD patients [9]. However, all these studies were conducted after Aβ deposition had appeared in brain tissue and cognitive impairment or movement disorder symptoms had occurred. Recently, it was reported that high-performance plasma Aβ biomarkers would predict brain Aβ burden, which demonstrate the potential clinical utility of plasma biomarkers in predicting brain Aβ burden at an individual level [10]. CSF has a greater homeostatic priority than blood and changes less than blood. Urine may thus be a better choice for the detection of earlier biomarkers, even for brain diseases [11]. In addition, the limited studies about urine-based biomarkers of brain diseases suggest that urine is an ideal potential source for biomarkers in brain diseases [12]. For example, urine biomarkers in an astrocytoma rat model can reflect early astrocytoma changes before magnetic resonance imaging (MRI) and thus gives potential clues for the early clinical diagnosis of astrocytoma patients [13].

Amyloid precursor protein (APP) (swe)/PSEN1^dE9 transgenic mice, overexpressing mutant APP and PS1 (APP/PS1), have been widely used as a model of AD to elucidate the pathogenic processes of the disease and to investigate candidate biomarkers [14]. Deposition of Aβ in the hippocampus of 6-month-old mice occasionally occurred [15, 16], and a mass of Aβ had been deposited in the hippocampus of 8-month-old mice [17, 18]. Compared with wild-type mice, the 8-month-old transgenic mice were mainly impaired in the foreground fear conditioning test and in plasticity to acquire strategies to swim toward a cued escape platform [19]. The 12-month-old transgenic mice showed spatial learning deficits as well as long-term contextual memory deficits and large deposits of aggregated insoluble Aβ [20]. The cognitive abnormality of APP (swe)/PSEN1^dE9 transgenic mice appeared at 4 months, when Aβ plaques had not yet deposited [21]. Potential candidate urinary biomarkers that appear before brain pathology can thus be identified by using 4-month-old APP (swe)/PSEN1^dE9 transgenic mice. Figure 1 presents the characteristics of behavioral and pathological profiles of different stages of the APP (swe)/PSEN1^dE9 transgenic mouse model.

Fig. 1.

Characteristics of behavioral and pathological profiles of different stages of APP (swe)/PSEN1^dE9 transgenic mice. Figure 1 presented the characteristics of behavioral and pathological profiles of different stages of the APP (swe)/PSEN1^dE9 transgenic mouse model. In 4- month-old APP (swe)/PSEN1^dE9 transgenic mice, the cognitive abnormality of APP (swe)/PSEN1^dE9 transgenic mice appeared at 4 months, when Aβ plaques had not yet deposited. Deposition of Aβ in the hippocampus of 6-month-old mice occasionally occurred, and a mass of Aβ had been deposited in the hippocampus of 8-month-old mice.

In this study, the APP (swe)/PSEN1^dE9 transgenic mouse model was used to identify candidate early AD biomarkers. To identify changes in the urinary proteome during AD development, we collected and analyzed urine samples of APP (swe)/PSEN1^dE9 transgenic mice at three growth stages (4-, 6-, and 8-month-old). The workflow of the proteomic analysis in this study is shown in Fig. 2. By using label-free and TMT-labeling proteomics analysis, we analyzed differential urine biomarkers from different stages of AD.

Fig. 2.

Workflow of urinary proteomics in APP (swe)/PSEN1^dE9 transgenic mouse model. Urine samples were collected from the control and transgenic groups. The proteins were analyzed using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) identification. Differential proteins were analyzed by DAVID and IPA.

MATERIALS AND METHODS

Experimental animals

APP (swe)/PSEN1^dE9 transgenic mice (n = 3, each month) and C57BL/6NCrl mice as the wild-type mice (n = 3, each month) were used in this study. The APP (swe)/PSEN1^dE9 transgenic mice carry the human APPswe (Swedish mutations K594N/M595L) and presenilin-1 with the exon-9 deletion (PS1-dE9) under control of the mouse prion protein promoter and produce increasing amounts of amyloid deposits in the brain. All animal protocols governing the experiments in this study were approved by the Institute of Basic Medical Sciences Animal Ethics Committee, Peking Union Medical College (Approved ID: ACUC-A02-2014-008). All animals were maintained with a standard laboratory diet at a controlled indoor temperature (22±1°C) and humidity (65 ∼ 70%) and with a 12-h light-dark cycle. The study was performed according to the guidelines developed by the Institutional Animal Care and Use Committee of Peking Union Medical College. All efforts were made to minimize suffering.

Urinary protein extraction

Urine samples from 4-, 6-, and 8-month-old mice in transgenic and wide-type control groups were collected in metabolic cages. During urine collection, all rats were given free access to water without food to avoid contamination.

Urine was centrifuged at 2,000×g for 15 min at 4°C. After the cell debris had been removed, the supernatant was centrifuged at 12,000×g for 15 min at 4°C. Three volumes of acetone were added after the pellets had been removed, and precipitation was allowed to occur at 4°C. After the supernatant was removed, the pellets were resuspended in lysis buffer (8 M urea, 2 M thiourea, 25 mM dithiothreitol (DTT) and 50 mM Tris). Protein concentrations were measured using the Bradford method.

In-gel protein digestion

Proteins were digested with trypsin (Trypsin Gold, Mass Spec Grade, Promega, Fitchburg, WI, USA) using in-gel protein digestion [22]. Protein samples were separated by SDS-PAGE. The gel was rinsed three times, 5 min each, in pure water and stained for 30 min in Coomassie brilliant blue at room temperature, after which the staining solution was discarded. The gel was destained for 1 h in pure water at room temperature. When destaining was complete, the solution was discarded. A clean razor blade was used to cut the protein bands from the gel. The gel slices were then placed into a 0.5-ml microcentrifuge tube that had been prewashed twice with 50% acetonitrile (ACN)/0.1% trifluoroacetic acid (TFA). The gel slices were destained twice with 0.2 ml of 100 mM NH₄HCO₃/50% ACN for 45 min each at 37°C to remove the Coomassie brilliant blue stain. The gel slices were dehydrated for 5 min at room temperature in 100μl of 100% ACN. At this point, the gel slices were much smaller than their original size and whitish or opaque in appearance. The gel slices were dried in a Speed Vac^® for 10 min at room temperature to remove the ACN. Trypsin Gold was resuspended at 1μg/μl in 50 mM acetic acid and then diluted in 40 mM NH₄HCO₃/10% ACN to 20μg/ml. The gel slices were preincubated in a minimal volume (10–20μl) of the trypsin solution at room temperature (do not exceed 30°C) for 1 h. The slices rehydrated during this time. If the gel slices appeared white or opaque after 1 h, an additional 10–20μl of trypsin was added, and they were incubated for another hour at room temperature. Sufficient digestion buffer (40 mM NH₄HCO₃/10% ACN) to completely cover the gel slices was added, and the tubes were capped tightly to avoid evaporation. The samples were then incubated overnight at 37°C. Next, the gel slice digests were incubated with 150μl of pure water for 10 min with frequent vortex mixing. The liquid was removed and saved in a new microcentrifuge tube. The gel slice digests were extracted twice with 50μl of 50% ACN/5% TFA (with mixing) for 60 min each time at room temperature. All extracts were pooled and dried in a Speed Vac^® at room temperature for 2–4 h. The extracted peptides were purified and concentrated using ZipTip^® pipette tips (Millipore Corporation, Cat# ZTC18S096, Darmstadt, Germany). The peptides eluted from the ZipTip^® tips were then ready for mass spectrometric analysis.

Filter-aided protein digestion

Proteins were digested with trypsin (Trypsin Gold, Mass Spec Grade, Promega, Fitchburg, WI, USA) using filter-aided sample preparation methods [23]. After proteins were loaded into a 10-kDa filter unit (Pall, Port Washington, New York, USA), UA buffer (8 M urea in 0.1 M Tris-HCl, pH 8.5) and NH₄HCO₃ (25 mM) were added successively, and the tube was centrifuged at 12,000×g for 30 min at 18°C. Proteins were denatured by incubation with 20 mM DTT at 37°C for 1 h and then alkylated with 50 mM iodoacetamide (IAA) for 45 min in the dark. After the samples had been centrifuged with UA twice and NH₄HCO₃ four times, the proteins were redissolved in an NH₄HCO₃ solution and digested with trypsin (1:50) at 37°C overnight. The tryptic peptides were desalted using Oasis HLB cartridges (Waters, Milford, Massachusetts, USA), and the desalted peptides were dried by vacuum evaporation (Thermo Fisher Scientific, Bremen, Germany).

Peptide tandem mass tag (TMT) labeling

The filter-aided digested peptides were solubilized in 100 mM tetraethylammonium bromide (TEAB) and labeled with 6-plex Tandem Mass Tag Label Reagents provided by Thermo Fisher Scientific (Pierce, Rockford, IL, USA), which had been equilibrated to room temperature immediately before use. Then, 41μl of anhydrous acetonitrile was added to each tube, and the reagent was allowed to dissolve for 5 min with occasional vortexing. The samples were briefly centrifuged to gather the solution, and 20μl of the TMT Label Reagent was added. The reaction was then incubated for 2 h at room temperature. After the peptides were labeled with isobaric tags, they were mixed at a 1:1:1:1:1:1 ratio based on the amount of total peptide, which was determined by running equal volumes of labeled samples through liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) and comparing the total signal intensities of all peptides. Finally, samples from the wild-type group (n = 3) and AD group (n = 3) were analyzed by two-dimensional LC-MS/MS (2DLC-MS/MS).

HPLC separation

The TMT-labeled samples were fractionated using a high-pH reversed-phase liquid chromatography (RPLC) column from Waters (4.6 mm×250 mm, Xbridge C18, 3μm) and loaded onto the column in buffer A1 (H₂O, pH = 10). The elution gradient progressed from 5–25% buffer B1 (90% ACN, pH = 10; flow rate = 1 ml/min) over 60 min. The eluted peptides were collected at one fraction per minute. The 60 fractions were dried, resuspended in 0.1% formic acid and pooled into 30 samples by combining fractions 1 and 31, 2 and 32, and so on. The odd-numbered fractions were chosen for further analysis. A total of 15 fractions from urinary peptide mixtures were analyzed by LC-MS/MS.

LC-MS/MS analysis

The in-gel digested peptides and TMT-labeled peptides were analyzed with a reverse-phase C18 self-packed capillary LC column (75μm×100 mm). The elution gradient progressed from 5–30% buffer B2 (0.1% formic acid, 99.9% ACN; flow rate = 0.3μL/min) over 40 min. A Triple TOF 5600 mass spectrometer was used to analyze the peptides eluted from LC, and each fraction was run twice. MS data were acquired using the high-sensitivity mode with the following parameters: 30 data-dependent MS/MS scans per full scan, full scans acquired at a resolution of 40,000, MS/MS scans acquired at a resolution of 20,000, rolling collision energy, charge state screening (including precursors with a charge state of +2 to +4), dynamic exclusion (exclusion duration 15 s), an MS/MS scan range of 100–1800 m/z, and a scan time of 100 ms.

Data analysis

The MS/MS data were analyzed with Mascot software (version 2.4.1, Matrix Science, London, UK), and proteins were identified by comparing the peptide spectra against the Swissprot_2014_07 databases (taxonomy: mus). Trypsin was selected as the digestion enzyme, up to two missed cleavage sites were allowed, and carbamidomethylation of a cysteine was defined as a fixed modification. The precursor ion mass tolerance and the fragment ion mass tolerance were 0.05 Da. The protein identification Mascot results were validated by using Scaffold Proteome Software (version Scaffold_4.3.3, Proteome Software Inc., Portland, OR). Peptide identification was accepted at a false discovery rate (FDR) of less than 1.0% at the protein level and if the sample produced at least 2 unique peptides. The spectra counts of each protein were normalized by the total spectra counts of identified proteins. The Scaffold Q+ was used for Label-Based Quantification (TMT, iTRAQ, SILAC, etc.) of peptides and proteins. Normalized reporter ion intensities were used to calculate the relative protein abundance. The protein ratios were then quantified by the median of the transformed reporter ion intensity ratios [24]. Label-free quantification was performed using Progenesis LC-MS software (version 4.1, Nonlinear, UK) as described previously [25]. The abundance of each protein was normalized by the median abundance of the common identified proteins.

Statistical analysis

All statistical analysis was performed with the Statistical Package for Social Studies software (SPSS, version 16, IBM). Comparisons between independent groups were conducted using one-way ANOVA followed by post hoc analysis with the least significant difference (LSD) test. The log2 (log10) transformation was not applied before we analyze these normalized data. Statistically significant proteins were defined as P-values less than 0.05, and fold-change in 4/6/8-month-old (APP) (swe)/PSEN1^dE9 transgenic mice was ge;1.5 or ≤0.67.

RESULTS AND DISCUSSION

Urinary proteome changes in APP (swe)/PSEN1^dE9 transgenic mice

The urinary proteins of 4-month-old mice were digested by an in-gel method and quantified by a label-free method. A total number of 443 protein groups were identified in the urinary proteome at an FDR <1%, and each generated at least 2 unique peptides. The levels of 29 urinary proteins were significantly different between the AD and wild-type groups (Table 1); among these proteins, sulfhydryl oxidase 1 levels were different in both the 50–80 kDa and 30–50 kDa lanes, and deoxyribonuclease-1 levels were different in both the 15–30 kDa and <15 kDa lanes. In these 29 differential proteins, 15 had been reported to be associated with AD. Specifically, 13 had been reported to be associated with the pathology mechanisms of AD, while 9 were identified as direct AD biomarkers (Table 1).

Table 1

Details of differential urinary proteins of 4-month-old mice

MW	Protein name	UniProt	Human UniProt	p value	Fold change	Confidence score	Normalized abundance						Pathology and mechanism	Bio-markers
							WT-1	WT-2	WT-3	AD-1	AD-2	AD-3
>85 kDa	Cadherin-1	P09803	P12830	0.022	1.87	218.7	2619939	3821894	4545058	7387696	5702665	7448193	[43, 85]	–
	Keratin, type II cytoskeletal 2 oral	Q3UV17	Q01546	0.041	0.56	181.9	95619	152124	160984	52841	108167	67233	–	–
50–85 kDa	Ceruloplasmin	Q61147	P00450	0.018	3.19	142.8	1405128	700799	1034296	4233275	2979048	2817111	[32, 33]	[31]
	Alpha-amylase1	P00687	P04745	0.037	2.98	1271.6	16862562	16394995	16747254	27733747	36333676	85064392	[44]	–
	Ectonucleotide pyrophosphatase	Q9R1E6	Q13822	0.038	2.89	438.4	6410477	4227746	3261439	14447745	15522911	10137661	–	[41]
	AnnexinA11	P97384	P50995	0.035	2.54	124.5	112847	385493	224516	851694	561465	420834	[39]	[40]
	Angiotensinogen	P11859	P01019	0.036	2.53	340.3	1832470	6008159	4047815	8130284	9777242	12155633	[38]	[37]
	Kallikrein-1	P15947	P06870	0.028	2.55	161.4	13993666	15481983	8200572	32915114	24952991	38216076	[26]	[27]
	Pantetheinase	Q9Z0K8	O95497	0.010	2.25	329.7	7264591	9291049	8941437	18429885	12901960	26056688	–	–
	Sulfhydryloxidase1	Q8BND5	O00391	0.029	1.93	444.1	4216654	7066885	5327850	12135722	10500098	9505950	–	–
	Sphingomyelin phosphodiesterase	Q04519	P17405	0.004	1.71	138.9	789408	934631	809095	1433994	1281043	1625436	[29 , 91]	–
	N-Acetylmuramoyl-L-alanine amidase	Q8VCS0	Q96PD5	0.001	1.63	237.3	4431595	4089554	4317879	6991718	6218791	7761766	–	–
	Galactocerebrosidase	P54818	P54803	0.002	1.51	296.1	23253378	21116228	25380610	39727782	31492109	33948684	[28]	[29]
30–50 kDa	Carbonic anhydrase 2	P00920	P00918	0.050	7.64	139.9	1026487	137540	1196867	2483019	8183767	7363671	[46]	–
	ATP-binding cassette sub-family A member 13	Q5SSE9	Q86UQ4	0.003	3.97	186.0	162885	413064	239944	1326933	1045477	870531	–	–
	Protein LEG1 homolog	Q8C6C9	Q6P5S2	0.046	3.17	331.5	617434	5666756	2910064	10253781	9689764	9179997	–	–
	Cathepsin B	P10605	P07858	0.029	3.13	338.5	7139961	1114522	1317986	9277618	8644600	12074430	[36]	[14]
	Acid ceramidase	Q9WV54	Q13510	0.031	0.42	264.6	39183681	22168743	30757070	11403664	10956829	16671159	[47]	–
	Sulfhydryl oxidase 1	Q8BND5	O00391	0.005	0.40	271.4	6515045	6236645	3506111	2174726	1897719	2494845	–	–
	Eosinophil cationic protein 1	P97426	no	0.031	0.39	167.5	251235145	103629026	273458396	112170010	66736059	65435575	–	–
	Fibronectin	P11276	P02751	0.017	0.36	191.7	6815535	4157474	3089723	1594554	2037884	1485064	[35]	[34]
	Prostaglandin-H2 D-isomerase	O09114	P41222	0.018	0.27	196.1	116338435	118916422	35968640	25071521	28557837	19099588	–	–
	Dipeptidyl peptidase 2	Q9ET22	Q9UHL4	0.017	0.23	329.3	36130176	10831853	12769425	4129725	4626310	4946023	–	–
	Lipoprotein lipase	P11152	P06858	0.025	0.18	157.9	12700373	4762033	4358911	1006139	706457	2137022	[48]	–
	Sialate O-acetylesterase	P70665	Q9HAT2	0.041	0.17	150.5	5186756	1470459	629510	242721	323999	669184	–	–
	Alpha-1-antitrypsin 1-4	Q00897	no	0.029	0.09	386.0	47453084	7866154	10014171	1352876	1413734	3276721	–	–
15–30 kDa	Deoxyribonuclease-1	P49183	P24855	0.002	2.46	177.0	6661598	5715618	7105781	11646345	22351861	13973828	–	–
	Ig kappa chain C region	P01837	P01834	0.014	1.80	103.9	8822671	11713287	12329617	21590819	21713190	15916672	–	[42]
<15 kDa	Deoxyribonuclease-1	P49183	P24855	0.026	2.47	146.4	1360757	728821	1080584	1869443	3272391	2687073	–	–
	Secretoglobin family 2B member 20	Q9JI02	no	0.026	0.56	323.6	15981747	29749962	27032579	15560357	9828085	15548114	–	–
	Ig kappa chain V-V region L7 (Fragment)	P01642	no	0.021	0.20	130.7	893023	1868961	641920	208754	175548	309125	–	–

Fig. 3.

Urinary proteins with significantly different levels in the 4-, 6- and 8-month-old APP (swe)/PSEN1^dE9 transgenic mice and wild-type mice. Figure 3 presented differential proteins in 4-, 6-, and 8-month-old APP (swe)/PSEN1^dE9 transgenic mice. There were 29, 33, and 86 differential proteins in the 4-, 6-, and 8-month-old group, respectively.

Seven proteins relate to the development of AD as well as biomarkers of AD. 1) Kallikrein-1 (KLK1) levels were different in 4-, 6-, and 8-month-old groups than the wild-type group, and the kallikrein-kinin system mediates inflammation in AD in vivo. Kallikrein-6, one member of the kallikrein family, was reported as a biomarker of AD and to have increased levels in CSF and serum and decreased levels in tissue [27]. 2) Galactocerebrosidase (GALC) levels were different in 4- and 8-month-old groups than the wild-type group, and a deficiency in galactosylceramidase potentially contributed to neurodegenerative disease [28]. Ceramide has been suggested to participate in the neuronal cell death that leads to AD. GALC, as a gene connected to ceramide metabolism, was upregulated in the brain tissue of AD patients, making it an attractive candidate for diagnostic purposes and for intervening in neurodegenerative processes [29]. 3) Ceruloplasmin (CERU) levels were higher in both 4- and 8-month-old groups than in the control group, with the same trend of increased levels found in the serum of AD patients [30]. The levels of CSF diagnostic markers, such as Aβ₄₂, tau, and phospho-tau, were correlated with lower plasma copper and CERU levels in patients with AD [31]. The ratio of CERU concentrations measured by enzymatic methods (eCP) to those measured by immunological methods (iCP), eCP/iCP, reflects the high specificity of AD patients as well as a decreased risk of having AD [32]. CERU had less ferroxidase activity in AD patients than wild-type patients, which contributed to the development of AD [33]. (4) Fibronectin (FINC) levels were different in 4- and 8-month-old groups than in the wild-type group and were reported to be a novel biomarker for AD from blood [34]. In addition, FINC levels were significantly lower in plasma in mild cognitive impairment (MCI) patients than healthy patients, which provides further insight into the biological pathways and processes that underpin the pathophysiology and progression of MCI and AD [35]. 5) Cathepsin B (CATB) levels were different in 4- and 8-month-old groups than the wild-type group. CATB has been previously reported to be upregulated in brain tissues from APP/PS1 transgenic mice, and its levels changed in the same direction relative to healthy mice as they changed in the serum of AD patients relative to control patients, making CATB a potential biomarker of AD [14]. In addition, CATB produces brain pyroglutamate Aβ, which represents a potential AD therapeutic [36]. 6) Angiotensinogen (ANGT) was upregulated in 4-month-old groups and has exhibited increased levels in the CSF of AD patients [37]. As a component of the renin-ANGT system (RAS), ANGT is helpful for AD processes [38]. 7) Annexin A11 (ANX11) was upregulated in 4-month-old groups. Annexin A1 was reported to be expressed strongly in the microglia of AD patients, which aids the surveillance of microglia and the maintenance of brain homeostasis by using Annexin A1-dependent mechanisms [39], and Annexin A5 was reported to be a biomarker of AD and present at increased levels in the plasma of AD patients [40]. The protein-normalized abundances and spectral counts of all seven differential proteins were greater in every mouse in the high-abundance group than those in the low-abundance group.

Two other differential proteins have been mentioned as potential AD biomarkers. 1) Ectonucleotide pyrophosphatase/phosphodiesterase family member 2 (ENPP2) levels were higher in the CSF of AD patients than the CSF of healthy controls. ENPP2 can be used to specifically discriminate AD from Lewy body dementia, making it a candidate AD biomarker [41]. 2) The Ig kappa chain C (IGKC) region was reported to be a potential serum biomarker of AD at the early stage [42].

Six differential proteins are associated with the mechanism of AD pathology. 1) CDH1, the gene of cadherin-1, is an AD risk gene that has significant association in Caribbean Hispanic individuals [43]. 2) Alpha-amylase 1 (AMY1) reflects M3 muscarinic agonist activity in AD [44]. 3) ASMase, the gene of sphingomyelin phosphodiesterase, which is connected to ceramide metabolism, was upregulated in the brain tissue of AD patients, making it an attractive candidate for both diagnostic purposes and intervening in neurodegenerative processes [29]. Significant correlations were observed between brain sphingomyelin phosphodiesterase levels and the levels of Aβ and phosphorylated tau protein, which contribute to AD disease pathogenesis [45]. 4) Plasma carbonic anhydrase 2 (CAH2) was elevated in AD patients and may thus play a role in the pathogenesis of AD [46]. 5) Acid ceramidase (ASAH1) levels were elevated in AD brain, suggesting that acid ceramidase might play a role in controlling neuronal apoptosis and that acid ceramidase-mediated signaling pathways might be involved in the molecular mechanism of AD [47]. 6) Lipoprotein lipase (LIPL) is associated with neurite pathology, and its levels were reduced in the dentate gyrus of AD brains [48]. Some differential proteins have not been reported in research papers and may provide clues to candidates involved in the AD pathology mechanism.

The urinary proteins of 6-month-old mice were digested by a filter-aided method and quantified by a TMT-labeling method. At the protein level, 310 protein groups were identified in the urinary proteome at an FDR <1%, and each generated at least 2 unique peptides. The levels of 33 urinary proteins were significantly different between the AD and wild-type groups. Eight of these 33 differential proteins had been reported to be associated with AD. To be specific, 6 were associated with the pathology mechanisms of AD, while 3 had been directly referred to be AD biomarkers (Table 2).

Table 2

Details of differential urinary proteins of 6-month-old mice

Protein name	UniProt	Human UniProt	p value	Fold change	Spectral count	MW	Normalized ratio						Pathology and mechanism	Biomarker
							WT-1	WT-2	WT-3	AD-1	AD-2	AD-3
Major urinary protein 1	P11588	NO	0.004	10.92	1657	21 kDa	0.45	0.95	1.6	8.45	13.9	10.4	–	–
Neogenin	P97798	Q92859	0.005	7.22	3	38 kDa	1	1.45	0.1	4.8	7.75	5.85	[50]	–
CMRF35-like molecule 2	Q8K249	Q496F6	0.004	5.31	3	22 kDa	0.5	3	3.4	12.45	9.7	14.5	–	–
Major urinary protein 5	P11591	no	0.011	3.79	326	21 kDa	1.05	4	3	12.45	7.55	10.5	–	–
WAP four-disulfide core domain protein 12	Q9JHY3	Q8WWY7	0.029	2.94	5	10 kDa	0.9	0.95	0.65	3.25	2.5	1.6	–	–
Keratin, type I cytoskeletal 19	P19001	P08727	0.014	2.41	9	45 kDa	1	1.2	1.25	3.35	2.05	2.9	–	–
Cadherin-15	P33146	P55291	0.015	2.06	3	86 kDa	1	0.6	0.75	1.95	1.4	1.5	–	–
Charged multivesicular body protein 2a	Q9DB34	O43633	0.027	1.88	4	25 kDa	1	0.8	0.65	1.35	1.35	1.9	[51]	–
Cadherin-11	P55288	P55287	0.024	1.87	2	88 kDa	1	0.7	0.55	1.65	1.25	1.3	–	–
Peroxisomal carnitine	Q9DC50	Q9UKG9	0.030	1.86	2	70 kDa	1	1.05	0.85	2.25	1.75	1.4	[52]	–
Desmoplakin	E9Q557	P15924	0.041	1.80	27	333 kDa	0.8	0.85	0.8	1.3	1.2	1.9	–	–
Pancreatic secretory granule membrane major glycoprotein GP2	Q9D733	P55259	0.048	1.79	3	59 kDa	1	0.8	0.55	1.55	1.05	1.6	–	–
Alpha-1-antitrypsin 1-4	Q00897	no	0.022	1.75	6	46 kDa	0.9	0.8	0.9	1.45	1.85	1.25	–	–
Heat shock cognate 71 kDa protein	P63017	P11142	0.020	1.67	7	71 kDa	0.9	0.9	0.95	1.45	1.3	1.85	[53]	–
Kynurenine/alpha-aminoadipate aminotransferase, mitochondrial	Q9WVM8	Q8N5Z0	0.002	1.63	2	48 kDa	1	1.15	0.85	1.65	1.6	1.65	–	–
Semaphorin-4A	Q62178	Q9H3S1	0.002	1.62	5	83 kDa	0.85	0.7	0.7	1.2	1.15	1.3	–	–
Serine protease inhibitor A3M	Q03734	no	0.006	1.53	3	47 kDa	0.9	0.95	0.8	1.2	1.45	1.4	–	–
Leukemia inhibitory factor receptor	P42703	P42702	0.044	1.51	6	122 kDa	0.9	0.9	1.05	1.45	1.15	1.7	[54]	–
Kallikrein-1	P15947	P06870	0.028	0.61	79	29 kDa	1	1.5	1.35	0.8	0.75	0.8	[26]	[27]
Prominin-1	O54990	O43490	0.027	0.60	2	97 kDa	1	0.75	0.75	0.55	0.4	0.55	–	–
Glycerophosphodiester phosphodiesterase domain-containing protein 3	Q99LY2	Q7L5L3	0.039	0.59	7	38 kDa	0.9	1.35	1.4	0.65	0.7	0.8	–	–
Ig kappa chain C region	P01837	P01834	0.003	0.57	25	12 kDa	1	0.85	0.8	0.5	0.5	0.5	–	[42]
Solute carrier family 12 member 3	P59158	P55017	0.030	0.55	3	111 kDa	1	0.9	1.3	0.5	0.75	0.5	–	–
Meprin A subunit beta	Q61847	Q16820	0.029	0.54	6	79 kDa	1.1	1.05	0.9	0.3	0.7	0.65	–	–
Solute carrier family 23 member 1	Q9Z2J0	Q9UHI7	0.014	0.54	9	66 kDa	1	0.7	0.8	0.4	0.45	0.5	–	–
Sushi domain-containing protein 2	Q9DBX3	no	0.001	0.54	3	90 kDa	0.9	1	0.9	0.5	0.45	0.55	–	–
Villin-1	Q62468	P09327	0.011	0.52	8	93 kDa	1	0.7	0.8	0.4	0.45	0.45	–	–
Ezrin	P26040	P15311	0.027	0.51	11	69 kDa	0.9	0.55	0.7	0.35	0.4	0.35	–	–
Chloride intracellular channel protein 4	Q9QYB1	Q9Y696	0.008	0.50	3	29 kDa	1	0.7	0.9	0.45	0.45	0.4	–	–
Na(+)/H(+) exchange regulatory cofactor NHE-RF3	Q9JIL4	Q5T2W1	0.013	0.48	4	56 kDa	1	0.8	0.7	0.3	0.45	0.45	–	–
Ras-related protein Rab-10	P61027	P61026	0.041	0.43	2	23 kDa	1	1.05	1.65	0.4	0.75	0.45	–	[49]
Na(+)/H(+) exchange regulatory cofactor NHE-RF1	P70441	O14745	0.011	0.42	13	22 kDa	1	0.65	0.75	0.3	0.35	0.35	–	–
Ig kappa chain V-V region L7 (Fragment)	P01642	no	0.005	0.38	10	13 kDa	1	0.8	0.7	0.35	0.3	0.3	–	–

Three differential proteins (KLK1, IGKC, and RAB10) have been suggested as AD biomarkers. The target gene RAB10, which is regulated by miRNA-369-3p, miRNA-30e-5p, miRNA-30e-3p, and miRNA-655, may play key roles in the progression and development of AD and may be a biomarker of AD [49].

Six differential proteins are associated with the mechanism of AD pathology. 1) Neogenin (NEO1) expresses weak NEO1 immunoreactivity in a small subset of amyloid plaques of AD brains [50]. 2) Charged multivesicular body protein 2a (CHM2A) levels were elevated in the 6-month-old group. Mutations in charged multivesicular body protein 2B (CHMP2B) can give rise to FTLD-UPS [51]. 3) Peroxisomal carnitine (OCTC) levels were higher in the 6-month-old group than the wild-type group, and new therapies acting on peroxisomal metabolism might be developed to prevent cognitive decline and other age-related neurological disorders [52]. 4) Heat shock cognate 71 kDa protein (HSP7C) was upregulated in 6-month-old groups, while its gene level was significantly lower across the three brain regions in AD patients than in controls patients, suggesting their participation in AD pathogenesis [53]. 5) Cooperative expression of leukemia inhibitory factor (LIF) and its receptor (LIFR) in AD hippocampus may indicate a role for LIF in neuronal damage or repair at these sites [54].

The urinary proteins of 8-month-old mice were digested by an in-gel method and quantified by a label-free method. At the protein level, 273 proteins were identified in the urinary proteome at an FDR <1%, and each generated at least 2 unique peptides. The levels of 86 urinary proteins were significantly different between the AD and wild-type groups. Thirty-one of these 86 differential proteins have been associated with AD. Specifically, 26 are associated with the mechanism of AD pathology, while 19 have been suggested as candidate AD biomarkers (Table 3).

Table 3

Details of differential urinary proteins of 8-month-old mice

MW range	Protein name	UniProt	Human UniProt	p value	Fold change	Spectral count						Pathology and mechanism	Biomarker
						WT1	WT2	WT3	AD1	AD2	AD3
>100 kDa	Angiotensin-converting enzyme	P09470	P12821	0.0092	0.63	27	28	27	20	13	19	[92]	–
	Aminopeptidase N	P97449	P15144	0.025	2.17	6	4	8	15	10	14	[93]	–
	Uromodulin	Q91X17	P07911	0.027	1.73	95	98	165	186	208	227	–	–
	Lysosome-associated membrane glycoprotein 2	P17047	P13473	0.031	0.29	3	2	2	1	0	1	[94]	[95]
	Podocalyxin	Q9R0M4	O00592	0.042	∞	0	0	0	2	1	4	–	–
65–100 kDa	Beta-glucuronidase	P12265	P08236	0.0001	0.00	5	4	6	0	0	0	[96]	–
	Carboxylesterase 1C	P23953	no	0.00029	0.67	17	18	17	12	12	11	–	–
	Biotinidase	Q8CIF4	P43251	0.0034	0.00	2	2	3	0	0	0	–	–
	Galactocerebrosidase	P54818	P54803	0.011	0.00	1	2	2	0	0	0	[28]	[29]
	Pancreatic alpha-amylase	P00688	no	0.016	∞	0	0	0	6	4	10	–	–
	Uromodulin	Q91X17	P07911	0.016	2.36	39	33	43	75	81	115	–	–
	Complement factor D	P03953	P00746	0.016	5.67	1	0	2	8	4	5	–	–
	Tyrosine-protein kinase receptor UF	Q00993	P30530	0.018	6.00	0	0	1	2	2	2	–	–
	Isoform 2 of V-type proton ATPase catalytic subunit A	P50516	P38606	0.02	0.00	1	1	2	0	0	0	–	[55]
	Pantetheinase	Q9Z0K8	O95497	0.021	0.67	2	2	2	2	1	1	–	–
	Ceruloplasmin	Q61147	P00450	0.036	3.00	1	1	0	2	2	2	[32, 33]	[31]
	Alpha-N-acetylglucosaminidase	O88325	P54802	0.041	0.33	8	10	12	2	1	7	–	–
	Ig mu chain C region	P01872	no	0.049	∞	0	0	0	3	2	7	–	–
50–65 kDa	Glycosylation-dependent cell adhesion molecule 1	Q02596	no	0.0002	28.00	0	1	0	8	10	10	–	–
	Afamin	O89020	P43652	0.00049	9.50	1	1	0	6	7	6	–	[89]
	Protein Fcgbp	E9Q0B5	no	0.0005	4.00	2	4	2	10	12	10	–	–
	Alpha-2-macroglobulin	Q61838	no	0.00076	0.00	11	8	9	0	0	0	–	–
	N-acetylgalactosamine-6-sulfatase	Q571E4	P34059	0.00096	0.00	11	15	11	0	0	0	–	–
	Lysosomal Pro-X carboxypeptidase	Q7TMR0	P42785	0.0013	0.00	8	7	5	0	0	0	–	–
	Beta-hexosaminidase subunit beta	P20060	P07686	0.0021	0.02	15	12	20	0	0	1	–	–
	Pancreatic alpha-amylase	P00688	no	0.0024	1.64	57	48	42	82	79	80	–	–
	Aminopeptidase N	P97449	P15144	0.0025	∞	0	0	0	4	5	6	[93]	[97]
	Growth arrest-specific protein 1	Q01721	P54826	0.0026	13.00	1	0	0	4	4	5	[98]	–
	Desmocollin-2	P55292	Q02487	0.0027	∞	0	0	0	2	3	2	–	–
	Acidic mammalian chitinase	Q91XA9	Q9BZP6	0.0029	33.00	1	0	0	14	11	8	–	–
	Beta-2-glycoprotein 1	Q01339	P02749	0.0038	15.00	1	0	0	5	4	6	–	–
	Major urinary protein 20	Q5FW60	NO	0.0047	0.58	11	11	9	6	6	6	–	–
	Serotransferrin	Q921I1	P02787	0.0062	2.53	18	12	6	32	30	29	[99]	[100]
	Acyloxyacyl hydrolase	O35298	P28039	0.0073	0.00	3	2	4	0	0	0	–	–
	Alpha-N-acetylgalactosaminidase	Q9QWR8	P17050	0.0096	0.00	2	2	1	0	0	0	–	–
	Lysosomal thioesterase PPT2	O35448	Q9UMR5	0.0098	0.00	1	2	1	0	0	0	–	–
	Protein LEG1 homolog	Q8C6C9	Q6P5S2	0.011	0.17	7	12	10	1	3	1	–	–
	Protein Ighg2b	A0A075B5P3	no	0.011	∞	0	0	0	7	3	6	–	–
	Acid sphingomyelinase-like phosphodiesterase 3a	P70158	Q92484	0.011	0.35	7	5	5	3	1	2	–	–
	Kallikrein-1	P15947	P06870	0.012	2.18	2	4	5	8	7	9	[26]	[27]
	Lactadherin	P21956	Q08431	0.012	3.80	1	3	1	8	6	5	[90]	–
	Alpha-1-antitrypsicn 1-5	Q00898	P01009	0.015	0.42	22	17	14	7	8	7	–	[101]
	Protein 9530053A07Rik	E9PVG8	Q9Y6R7	0.015	0.02	26	11	24	0	1	0	–	–
	Sushi domain-containing protein 2	Q9DBX3	Q9UGT4	0.016	0.20	2	4	4	0	1	1	–	–
	Polymeric immunoglobulin receptor	O70570	P01833	0.017	1.84	8	10	13	16	21	20	–	–
	Protein Ighg2c	F6TQW2	no	0.018	4.00	0	1	2	4	5	3	–	–
	Carboxylesterase 1C	P23953	no	0.018	3.33	1	2	0	4	3	3	–	–
	Tyrosine-protein kinase receptor UF	Q00993	P30530	0.018	∞	0	0	0	1	2	1	–	–
	Hemopexin	Q91X72	P02790	0.019	4.44	4	5	0	11	11	18	[102]	[103]
	Tripeptidyl-peptidase 1	O89023	O14773	0.019	0.00	1	3	3	0	0	0	–	–
	Napsin-A	O09043	O96009	0.02	1.94	12	11	11	17	23	26	[104]	–
	Complement C3	P01027	P01024	0.02	∞	0	0	0	6	2	4	[60, 105]	[59]
	Galactocerebrosidase	P54818	P54803	0.021	0.36	3	6	5	1	2	2	[28]	[29]-
	Matrix-remodeling-associated protein 8	Q9DBV4	Q9BRK3	0.023	10.00	0	1	0	5	3	2	–	–
	CD44 antigen	P15379	P16070	0.027	1.67	2	2	2	3	4	3	[62]	–
	Major urinary protein 2	P11589	no	0.031	0.72	63	70	53	43	49	42	–	–
	Actin, aortic smooth muscle	P62737	P62736	0.031	3.50	0	1	1	2	2	3	–	–
	Meprin A subunit alpha	P28825	Q16819	0.032	1.35	17	17	17	19	25	25	[106]	–
	Leucine-rich HEV glycoprotein	Q91XL1	P02750	0.034	0.67	15	13	11	8	9	9	–	–
	Fibronectin	P11276	P02751	0.038	3.40	3	2	0	6	7	4	[35]	[34]
	Alpha-1-acid glycoprotein 1	Q60590	P02763	0.039	1.78	4	3	2	6	5	5	–	–
	MCG15829, isoform CRA_a	Q3KQQ2	no	0.04	0.57	8	11	9	5	4	7	–	–
	Ceruloplasmin	Q61147	P00450	0.044	8.00	0	1	0	2	2	4	[32, 33]	[30]
	N-acetylglucosamine-6-sulfatase	Q8BFR4	P15586	0.048	0.33	3	6	3	1	1	2	–	–
30–50 kDa	Mammalian ependymin-related protein 1	Q99M71	Q9UM22	0.0001	0.00	2	2	2	0	0	0	–	–
	N-acyl-aromatic-L-amino acid amidohydrolase (carboxylate-forming)	Q91XE4	Q96HD9	0.00017	0.00	7	6	6	0	0	0	–	–
	Pancreatic alpha-amylase	P00688	no	0.00022	7.64	7	2	2	28	29	27	–	–
	Protein Ighg2c	F6TQW2	no	0.00024	2.58	3	4	5	10	11	10	–	–
	Alpha-2-macroglobulin	Q61838	no	0.00026	0.05	21	21	17	1	1	1	–	–
	Glycosylation-dependent cell adhesion molecule 1	Q02596	no	0.0005	∞	0	0	0	2	1	2	–	–
	Prothrombin	P19221	P00734	0.00073	1.69	4	4	5	7	8	7	[107 –109]	–
	Cluster of Lysosomal thioesterase PPT2	O35448	Q9UMR5	0.0015	0.00	3	2	3	0	0	0	–	–
	Urokinase-type plasminogen activator	P06869	P00749	0.0023	2.19	6	8	7	17	13	16	[63]	–
	Nidogen-2	O88322	Q14112	0.0027	∞	0	0	0	2	1	1	–	–
	Prostatic spermine-binding protein	P15501	no	0.0034	0.00	3	3	5	0	0	0	–	–
	dorant-binding protein 2a	Q8K1H9	no	0.0052	0.11	12	7	8	1	1	1	–	–
	Deoxyribonuclease-1	P49183	P24855	0.0053	0.61	22	24	24	14	17	12	–	–
	Complement factor D	P03953	P00746	0.0062	2.45	20	41	28	78	78	62	–	–
	UPF0762 protein C6orf58 homolog	Q8C6C9	Q6P5S2	0.0095	0.00	2	2	1	0	0	0	–	–
	Major urinary protein 20	Q5FW60	no	0.013	0.49	16	17	20	9	11	6	–	–
	L-lactate dehydrogenase B chain	P16125	P07195	0.016	0.00	1	2	1	0	0	0	–	–
	Procollagen C-endopeptidase enhancer 1	Q61398	Q15113	0.016	4.40	1	2	2	8	9	5	–	–
	Leucine-rich HEV glycoprotein	Q91XL1	P02750	0.017	0.50	6	5	7	4	3	2	–	–
	Hemoglobin subunit alpha	P01942	P69905	0.019	0.00	2	5	3	0	0	0	–	–
	Hemoglobin subunit beta-1	P02088	P68871	0.022	0.00	2	6	6	0	0	0	–	–
	Lysosomal protective protein	P16675	P10619	0.022	0.00	1	3	3	0	0	0	[110]	–
	Uromodulin	Q91X17	P07911	0.022	2.72	11	8	6	21	17	30	–	–
	Prostate stem cell antigen	P57096	O43653	0.028	6.00	0	0	1	2	1	3	[111]	–
	Cathepsin B	P10605	P07858	0.03	0.00	3	1	2	0	0	0	[36]	[14]
	Meprin A subunit alpha	P28825	Q16819	0.03	1.68	7	7	11	14	12	16	[106]	–
	MCG140531	A2AV72	no	0.033	0.65	10	8	8	5	6	6	–	–
	Ig kappa chain V19-17	P01633	P06312	0.038	0.33	1	1	1	1	0	0	–	–
	Cathepsin Z	Q9WUU7	Q9UBR2	0.044	0.18	4	2	5	0	0	2	[112]	–
20–30 kDa	Odorant-binding protein 2a	Q8K1H9	no	0.004	0.00	2	2	1	0	0	0	–	–
	Lymphocyte antigen 6D	P35459	Q14210	0.0048	7.00	1	0	0	3	2	2	–	–
	Napsin-A	O09043	O96009	0.0068	4.00	1	1	0	3	3	2	[104]	–
	Complement factor D	P03953	P00746	0.0078	5.50	1	2	3	8	14	11	–	–
	Kallikrein-1	P15947	P06870	0.008	∞	0	0	0	5	7	10	[26]	[27]
	Serum albumin	P07724	P02768	0.01	2.83	2	1	3	5	5	7	[56]	[55]
	Pro-epidermal growth factor	P01132	P01133	0.033	14.00	1	0	0	5	7	2	–	–
<20 kDa	Major urinary protein 20	Q5FW60	no	0.0012	0.23	20	15	17	5	2	5	–	–
	Sulfated glycoprotein 1	Q61207	P07602	0.003	0.44	16	16	13	8	6	6	–	[41]
	Clusterin	Q06890	P10909	0.0034	10.00	0	1	0	3	3	4	[57, 113]	[114]
	Lysozyme C-2	P08905	P61626	0.0049	∞	0	0	0	4	3	2	–	–
	Uromodulin	Q91X17	P07911	0.0083	4.75	1	2	1	5	7	7	–	–
	Complement factor D	P03953	P00746	0.011	3.53	6	7	6	16	23	28	–	–
	Cystatin-C	P21460	P01034	0.011	∞	0	0	0	2	1	1	–	[61, 115]
	Ubiquitin-60S ribosomal protein L40	P62984	P62987	0.011	∞	0	0	0	2	1	1	–	–
	Cluster of MCG116526	D3YYY1	no	0.029	0.24	10	6	5	2	2	1	–	–
	Fibronectin	P11276	P02751	0.038	5.50	0	1	1	4	2	5	[35]	[34]
	Protein Col6a3	E9PWQ3	P12111	0.038	∞	0	0	0	3	1	1	–	–
	Pancreatic alpha-amylase	P00688	no	0.046	5.00	0	1	1	5	2	3	–	–
	Plasminogen	P20918	P00747	0.047	4.33	0	2	1	5	5	3	[116]	[117]

Thirteen proteins have been previously reported to be related to the development of AD as well as biomarkers of AD, while 5 of them (KLK1, GALC, CERU, FINC, and CATB) had been already noted in 4-month-old groups. For example: 1) serum albumin (ALBU) was differentially expressed between brains of AD mouse models and those of healthy mice [55], and serum levels of albumin-Aβ complexes were lower in AD patients than healthy controls, which is very useful to apply in monitoring the progression of AD [56]. 2) Clusterin (CLUS) levels were higher in AD patients than controls, especially in regions with most abundant in Aβ, facilitating the development of AD [57]. As a heterodimeric glycoprotein, clusterin was more abundant in the CSF of some neurodegenerative disease patients than that of healthy controls and could thus serve as a potential biomarker to differentiate Parkinson’s disease from dementia with Lewy bodies [58]. 3) CSF complement C3 (CO3) is a staging biomarker in AD [59]. In addition, the dysregulation of neuron-glia interactions through NFκB/ C3/C3aR signaling may contribute to synaptic dysfunction in AD [60]. Nineteen differential proteins have been mentioned as potential AD biomarkers. For example: 1) sulfated glycoprotein 1 (SAP) is a novel CSF biomarker for neurodegeneration and has been validated by a high-throughput multiplexed targeted proteomic assay [41]. 2) Cystatin-C (CYTC) is a novel CSF biomarker for staging early AD [61]. Twenty-six differential proteins are associated with the mechanism of AD pathology. For example: 1) CD44 antigen (CD44) gene expression was greater in lymphocytes derived from AD patients than in those from healthy controls, suggesting that it plays roles in the peripheral immune response during the development of AD [62]. 2) The rs2227564 polymorphism of the urokinase-type plasminogen activator (UROK) gene increased the risk of AD [63]. Other differential proteins are annotated in Table 3.

Functional analysis of differential urine proteins in APP (swe)/PSEN1^dE9 transgenic mouse

Functional annotation of differential urinary proteins was performed using DAVID [64]. The differential proteins at three time points were classified to be involved with certain biological processes (Fig. 4), molecular components, and molecular functions (Supplementary Figs. 1 and 2).

Fig. 4.

Biological processes of differential proteins in different stages of AD development. A) The biological process of differential proteins on 4-month-old APP (swe)/PSEN1^dE9 transgenic mice with p value < 0.05. B) The biological process of differential proteins on 6-month-old APP (swe)/PSEN1^dE9 transgenic mice with p value < 0.05. C) The biological process of differential proteins on 8-month-old APP (swe)/PSEN1^dE9 transgenic mice with p value < 0.05.

In 4-month-old mice, differential urine proteins involved in lipid metabolism and lipid catabolism were enriched. As lipid metabolism is a fundamental process for brain development and function, aberrant lipid metabolism was not surprisingly common in an animal model of AD that relates to AD pathogenesis [65 –67]. For example, AD-induced perturbation of niche fatty acid metabolism can suppress the homeostatic and regenerative functions of neural stem cells, supporting the mechanism of AD pathogenesis [65]. Abnormal lipid metabolism influences Aβ metabolism and deposition in both brain parenchyma and vasculature, as well as tau hyperphosphorylation and aggregation, which then likely triggers a series of downstream catalytic events that eventually affect the progression of AD pathogenesis [66]. More importantly, lipid rafts from human cerebral cortex are associated with the pathogenesis of early AD, as β-secretase/AβPP (amyloid-β protein precursor) interactions and lipid raft microviscosity are strongly and positively correlated in AD frontal and entorhinal cortices, indicating that the aberrant lipid metabolism had already occurred in the early stage of AD [68]. Decidualization and calcium-dependent cell-cell adhesion via plasma membrane cell adhesion molecules, which are not reported in AD studies, may play roles in the mechanism of AD pathology. Recent studies demonstrated that using a triple receptor agonist (TA), which activates GIP-1, GIP and glucagon receptors at the same time, reduces the total amount of Aβ in an APP/PS1 transgenic mouse model through a 2-month TA treatment [69]. However, this research used 6-month-old transgenic mice for a 2-month treatment, and Aβ had already appeared. We thus suppose that a more effective treatment can be achieved when the 4-month-old transgenic mice are used for therapy. The top ranked process of 4-month-old APP (swe)/PSEN1^dE9 transgenic mice was lipid metabolic process, and the Galactocerebrosidase, Acid ceramidase, Ectonucleotidepyroph, Prostaglandin-H2 D-isomerase, Lipoprotein lipase were included in this process.

The differential proteins in 6-month-old mice were involved in transport processes, while axonal transport defects play an important role in the pathogenesis of AD [70]. Other biological processes such as microvillus assembly, the regulation of cell shape and protein localization to the plasma membrane may contribute to the formation of Aβ plaque deposits in the brain of AD patients, which provides potential clues for the studying the mechanism of AD. The top ranked process of 6-month-old APP (swe)/PSEN1^dE9 transgenic mice was microvillus assembly, and the Na(+)/H(+) exchange regulatory cofactor NHE-RF1, Villin-1, and Ezrin were included in this process.

A larger number of differential urine proteins were identified in 8-month-old mice and indicated other AD-related biological processes. Proteolysis processes were overrepresented and reported to be associated with AβPP proteolysis, which is consistent with the pathology of AD [71]. In addition, the urinary proteins that were related to fibrinolysis may be factors contributing to AD [72]. What is more, lysosome organization accumulated at Aβ in mouse models of AD, which gave rise to potential therapies of AD patients [73]. Lastly, angiogenesis processes were reported to contribute to the pathogenesis of AD, as they are a common feature of Aβ plaques [74, 75], and the levels of some angiogenesis factors were elevated in the plasma of AD patients [76]. The top ranked process of 8-month-old APP (swe)/PSEN1^dE9 transgenic mice was proteolysis, the Kallikrein-1, Tripeptidyl-peptidase 1, Plasminogen, Lysosomal Pro-X carboxypeptidase, Urokinase-type plasminogen activator, Aminopeptidase N, Napsin-A, Cathepsin B, Angiotensin-converting enzyme, Meprin A subunit alpha, Cathepsin Z, Procollagen C-endopeptidase enhancer 1, Complement factor D, Lysosomal protective protein, and Prothrombin were included in this process.

It was reported that mitochondrial dysfunction has been revealed in many human AD studies [77]. In our study, there were 29, 33 and 86 differential urinary proteins identified in the 4-, 6-, and 8-month-old group, respectively, and some of them were also reported with mitochondrial dysfunction. For example: 1) Ceruloplasmin was reported to be associated with the development of AD as well as biomarkers of AD, and the mitochondrial energy metabolism was impaired in aceruloplasminemia brains [78]. 2) ASMase, the gene of sphingomyelin phosphodiesterase, was reported to regulate mitochondrial function in alcoholic liver disease [79]. 3) Heat shock cognate 71 kDa protein (HSP7C) was reported to differentially expressed in the mitochondria of Parkinson’s disease [80]. 4) Clusterin was reported to be related to the development of AD as well as biomarkers of AD, and it was reported to inhibit mitochondrial apoptosis as well [81]. 5) The decreased expression level of gene complement C3 was reported to be mediated by mitochondrial DNA variants [82].

In summary, in the biological process category, metabolic and carbohydrate metabolic processes were overrepresented in 4- and 8-month-old AD mice. The differential urine proteins in 4- and 8-month-old mice indicated an enrichment of metabolic process, while the three subtypes of AD can be distinguished by metabolic profiling [83]. Considering that the biological processes whose activities changed in this experiment did not extensively overlap at these three time points, this result indicates that only three different stages might occur, implying that different therapeutic strategies may be appropriate for different stages. If biomarkers can be found in human as early as was found for the 4-month-old in this model, we may have an early intervention window in which disease development can be stopped. A new paper published recently showed that an FDA-approved drug, memantine, currently used only for alleviating the symptoms of moderate-to-severe AD, might be used to prevent or slow the progression of the disease if used before symptoms appear [84].

In the cellular component category (Supplementary Fig. 1), most differential proteins came from the extracellular exosome, the extracellular space, the anchored component of membranes, and blood microparticles, whereas a small number of differential proteins were derived from organelles such as the lysosome and the Golgi apparatus. In the molecular function category (Supplementary Fig. 2), hydrolase activity was overrepresented in 4- and 8-month-old mice; calcium ion binding was overrepresented in 4-month-old mice; protein complex binding, sodium ion transmembrane transporter activity and protein binding were overrepresented in 6-month-old mice; and peptidase, endopeptidase, serine-type peptidase activities were overrepresented in 8-month-old mice.

IPA of differential proteins in AD development

To identify the differential proteins with mostly affected cellular death and survival, ingenuity pathway analysis (IPA) was applied to cell death and survival networks (Fig. 5). Cadherin-1, which was reported to be associated with AD [43], is a core component of these networks in 4-month-old mice. CP, ENPP2, AGT, SMPD1, and PLA2G7 were upregulated and play roles in AD [29 , 86], whereas sialate O-acetylesterase was downregulated in 4-month-old mice. In 6-month-old mice, NEO1, CROT, GP2, SEMA4A, and LIFR were present in higher levels in transgenic mice than the control group, and NEO1 as well as CROT were associated with the pathology and mechanism of AD [50, 87]. In contrast, MEP1B, CLIC4, and PDZK1 levels were lower in transgenic mice than the control group. FINC was the core component of the network in 8-month-old mice and plays roles in AD [88]. In addition, AFM, MFGE8, C3, CD44, and PLAU were upregulated in transgenic mice and are potential biomarkers or play roles in the development of AD [60 , 90], while VNN1, HEXB, and CTSZ were downregulated in transgenic mice when compared with control groups.

Fig. 5.

IPA determined that the changed proteins are involved in the cell death and survival network. A) Cell death and survival network of differential proteins in 4-month-old APP (swe)/PSEN1^dE9 transgenic mice. B) Cell death and survival network of differential proteins in 6-month-old APP (swe)/PSEN1^dE9 transgenic mice. (C) Cell death and survival network of differential proteins in 8-month-old APP (swe)/PSEN1^dE9 transgenic mice. Red indicated upregulated genes, and green indicated downregulated genes in the present study.

In this preliminary study, only a small number of APP (swe)/PSEN1^dE9 transgenic mice were used. As there are various kinds of AD transgenic mice models, it would be more comprehensive to use some other AD transgenic mice models to reflect the early changes of AD. Using exactly same proteomic approaches at different AD stage would be better. In future studies, clinical urine samples of AD patients will be used to validate the differential proteins identified in this study.

CONCLUSION

Our results indicated that urine proteins enable AD early detection before Aβ plaque deposition, which may provide an opportunity for intervention.

Footnotes

ACKNOWLEDGMENTS

This research was supported by the National Key Research and Development Program of China (2016YFC1306300, 2018YFC0910202), Beijing Natural Science Foundation (7172076), Beijing cooperative construction project (110651103), Beijing Normal University (11100704), Peking Union Medical College Hospital (2016-2.27), the National Natural Science Foundation of China (NSFC #81271239, #81771205, #91632113), the Natural Science Foundation and Major Basic Research Program of Shanghai (16JC1420500, 16JC1420502), and the CAMS Innovation Fund for Medical Sciences (CIFMS #2017-I2M-3-008). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Tan

, Yu

, Tan

(2014) Biomarkers for preclinical Alzheimer’s disease. J Alzheimers Dis 42, 1051–1069.

Gao

(2014) Opinion: Are urinary biomarkers from clinical studies biomarkers of disease or biomarkers of medicine? MOJ Proteom Bioinform 1, 00028.

Gao

(2013) Urine-an untapped goldmine for biomarker discovery? Sci China Life Sci 56, 1145–1146.

, Guo

, Gao

(2017) Dynamic changes of urine pro-teome in a Walker 256 tumor-bearing rat model. Cancer Med 6, 2713–2722.

Zhao

, Li

, Shao

, Yin

, Gao

(2014) Dynamic changes of urinary proteins in a focal segmental glomeru-losclerosis rat model. Proteome Sci 12, 42.

Shevchenko

, Wetterhall

, Bergquist

, Hoglund

, Andersson

, Kultima

(2012) Longitudinal characterization of the brain proteomes for the tg2576 amyloid mouse model using shotgun based mass spectrometry. J Proteome Res 11, 6159–6174.

Peng

, Guo

, Xia

, Zhou

, Yang

, Westaway

, Wishart

, Li

(2014) Development of isotope labeling liquid chromatography mass spectrometry for mouse urine metabolomics: Quantitative metabolomic study of trans-genic mice related to Alzheimer’s disease. J Proteome Res 13, 4457–4469.

, Browne

, Liu

, Angel

, Ho

, Chan

(2012) Metabonomic profiling of TASTPM transgenic Alzheimer’s disease mouse model. J Proteome Res 11, 5903–5913.

Youn

, Park

, Han

, Kim

(2011) Urine neural thread protein measurements in Alzheimer disease. J Am Med Dir Assoc 12, 372–376.

10.

Nakamura

, Kaneko

, Villemagne

, Kato

, Doecke

, Dore

, Fowler

, Li

, Martins

, Rowe

, Tomita

, Matsuzaki

, Ishii

, Arahata

, Iwamoto

, Ito

, Tanaka

, Masters

, Yanagisawa

(2018) High performance plasma amyloid-beta biomarkers for Alzheimer’s disease. Nature 554, 249–254.

11.

, Gao

(2016) Should we search for early brain disease biomarkers in urine. Author J 1, 00003.

12.

, Gao

(2015) Urinary biomarkers ofbrain diseases. Genomics Proteomics Bioinformatics 13, 345–354.

13.

, Zhang

, An

, Yin

, Gao

(2018) Early candidate biomarkers found from urine of astrocytoma rat before changes in MRI. Sci China Life Sci 61, 982–987.

14.

Sun

, Rong

, Lu

, Peng

, Li

, Xu

, Wang

(2015) Translational study of Alzheimer’s disease (AD) biomarkers from brain tissues in AβPP/PS1 mice and serum of AD patients. J Alzheimers Dis 45, 269–282.

15.

Laursen

, Mork

, Plath

, Kristiansen

, Bastlund

(2013) Cholinergic degeneration is associated with increased plaque deposition and cognitive impairment in APPswe/PS1dE9 mice. Behav Brain Res 240, 146–152.

16.

Kilgore

, Miller

, Fass

, Hennig

, Haggarty

, Sweatt

, Rumbaugh

(2010) Inhibitors of class 1 histone deacetylases reverse contextual memory deficits in a mouse model of Alzheimer’s disease. Neuropsychophar-macology 35, 870–880.

17.

Holcomb

, Gordon

, McGowan

, Yu

, Benkovic

, Jantzen

, Wright

, Saad

, Mueller

, Morgan

, Sanders

, Zehr

, O’Campo

, Hardy

, Prada

, Eck-man

, Younkin

, Hsiao

, Duff

(1998) Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor protein and presenilin 1 transgenes. Nat Med 4, 97–100.

18.

Garcia-Alloza

, Robbins

, Zhang-Nunes

, Pur-cell

, Betensky

, Raju

, Prada

, Greenberg

, Bacskai

, Frosch

(2006) Characterization of amyloid deposition in the APPswe/PS1dE9 mouse model of Alzheimer disease. Neurobiol Dis 24, 516–524.

19.

Janus

, Flores

, Xu

, Borchelt

(2015) Behavioral abnormalities in APPSwe/PS1dE9 mouse model of AD-like pathology: Comparative analysis across multiple behavioral domains. Neurobiol Aging 36, 2519–2532.

20.

Zhang

, Hao

, Liu

, Zhang

, Lei

, Su

, Miao

, Li

(2011) Soluble Abeta levels correlate with cognitive deficits in the 12-month-old APPswe/PS1dE9 mouse model of Alzheimer’s disease. Behav Brain Res 222, 342–350.

21.

Bonardi

, de Pulford

, Jennings

, Pardon

(2011) A detailed analysis of the early context extinction deficits seen in APPswe/PS1dE9 female mice and their relevance to preclinical Alzheimer’s disease. Behav Brain Res 222, 89–97.

22.

Speicher

, Kolbas

, Harper

, Speicher

(2000) Systematic analysis of peptide recoveries from in-gel digestions for protein identifications in proteome studies. JBiomol Tech 11, 74–86.

23.

Wisniewski

, Zougman

, Nagaraj

, Mann

(2009) Universal sample preparation method for proteome analysis. Nat Methods 6, 359–362.

24.

Twigt

, Bezstarosti

, Demmers

, Lindemans

, Laven

, Steegers-Theunissen

(2015) Preconception folic acid use influences the follicle fluid proteome. Eur J Clin Invest 45, 833–841.

25.

Stoop

, Singh

, Stingl

, Martin

, Khademi

, Olsson

, Hintzen

, Luider

(2013) Effects of natal-izumab treatment on the cerebrospinal fluid proteome of multiple sclerosis patients. J Proteome Res 12,1101–1107.

26.

Viel

, Buck

(2011) Kallikrein-Kinin system mediated inflammation in Alzheimer’s disease in vivo. Curr Alzheimer Res 8, 59–66.

27.

Diamandis

, Yousef

, Petraki

, Soosaipillai

(2000) Human kallikrein 6 as a biomarker of Alzheimer’s disease. Clin Biochem 33, 663–667.

28.

Marshall

, Bongarzone

(2016) Beyond Krabbe’s disease: The potential contribution of galactosylcerami-dase deficiency to neuronal vulnerability in late-onset synucleinopathies. J Neurosci Res 94, 1328–1332.

29.

Filippov

, Song

, Zhang

, Vinters

, Tung

, Kirsch

, Yang

, Duerksen-Hughes

(2012) Increased ceramide in brains with Alzheimer’s and other neurodegenerative diseases. J Alzheimers Dis 29, 537–547.

30.

Park

, Lee

, Park

(2014) Elevated serum copper and ceruloplasmin levels in Alzheimer’s disease. Asia Pac Psychiatry 6, 38–45.

31.

Kessler

, Pajonk

, Meisser

, Schneider-Axmann

, Hoffmann

, Supprian

, Herrmann

, Obeid

, Multhaup

, Falkai

, Bayer

(2006) Cerebrospinal fluid diagnostic markers correlate with lower plasma copper and ceruloplasmin in patients with Alzheimer’s disease. J Neural Transm (Vienna) 113, 1763–1769.

32.

Siotto

, Simonelli

, Pasqualetti

, Mariani

, Ca

, Bucossi

, Ventriglia

, Molinario

, Ant

, Rongioletti

, Rossini

, Squitti

(2016) Association between serum ceruloplasmin specific activity and risk of Alzheimer’s disease. J Alzheimers Dis 50, 1181–1189.

33.

Bush

(2013) The metal theory of Alzheimer’s disease. J Alzheimers Dis 33, S277–281.

34.

Long

, Pan

, Ifeachor

, Belshaw

, Li

(2016) Discovery of novel biomarkers for Alzheimer’s disease from blood. Dis Markers 2016, 4250480.

35.

Muenchhoff

, Poljak

, Song

, Raftery

, Brodaty

, Duncan

, McEvoy

, Attia

, Schofield

, Sachdev

(2015) Plasma protein profiling of mild cognitive impairment and Alzheimer’s disease across two independent cohorts. J Alzheimers Dis 43, 1355–1373.

36.

Hook

, Yu

, Toneff

, Kindy

, Hook

(2014) Brain pyroglutamate amyloid-beta is produced by cathepsin B and is reduced by the cysteine protease inhibitor E64d, representing a potential Alzheimer’s disease therapeutic. J Alzheimers Dis 41, 129–149.

37.

Mateos

, Ismail

, Gil-Bea

, Leoni

, Winblad

, Björkhem

, Cedazo-Mínguez

(2011) Upregulation of brain renin angiotensin system by 27-hydroxycholesterol in Alzheimer’s disease. J Alzheimers Dis 24, 669–679.

38.

Savaskan

(2005) The role of the brain renin-angiotensin system in neurodegenerative disorders. Curr Alzheimer Res 2, 29–35.

39.

McArthur

, Cristante

, Paterno

, Christian

, Roncar-oli

, Gillies

, Solito

(2010) Annexin A1: A central player in the anti-inflammatory and neuroprotective role of microglia. J Immunol 185, 6317–6328.

40.

Yamaguchi

, Kokai

, Imai

S-I

, Utsumi

, Matsumoto

, Honda

, Mizue

, Momma

, Maeda

, Toyomasu

, Ito

, Kobayashi

, Hashimoto

, Saito

, Sohma

(2010) Investigation of annexin A5 as a biomarker for Alzheimer’s disease using neuronal cell culture and mouse model. J Neurosci Res 88, 2682–92.

41.

Heywood

, Galimberti

, Bliss

, Sirka

, Paterson

, Magdalinou

, Carecchio

, Reid

, Heslegrave

, Fenoglio

, Scarpini

, Schott

, Fox

, Hardy

, Bhatia

, Heales

, Sebire

, Zetterberg

, Mills

(2015) Identification of novel CSF biomarkers for neurodegeneration and their validation by a high-throughput multiplexed targeted proteomic assay. Mol Neurodegener 10, 64.

42.

Shen

, Chen

, Yang

, Chen

, Liao

, Li

, Ying

, Tian

, Liu

, Ni

(2016) Redox proteomic profiling of specifically carbonylated proteins in the serum of triple transgenic Alzheimer’s disease mice. Int J Mol Sci 17, 469.

43.

Shang

, Lv

, Zhang

, Duan

, Wang

, Li

, Liu

, Ruijie

, Jiang

(2015) Genome-wide haplotype association study identify TNFRSF1A, CASP7, LRP1B, CDH1 and TG genes associated with Alzheimer’s disease in Caribbean Hispanic individuals. Oncotarget 6, 4250442514.

44.

Sramek

, Cutler

, Hurley

, Seifert

(1995) The utility of salivary amylase as an evaluation of M3 muscarinic agonist activity in Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry 19, 85–91.

45.

, Huang

, Li

, Gong

, Schuchman

(2010) Deregulation of sphingolipid metabolism in Alzheimer’s disease. Neurobiol Aging 31, 398–408.

46.

Jang

, Yun

, Ahn

, Song

, Jo

, Kim

, Park

, Han

, Koh

(2010) Plasma carbonic anhydrase II protein is elevated in Alzheimer’s disease. J Alzheimers Dis 21, 939–945.

47.

Huang

, Tanimukai

, Liu

, Iqbal

, Grundke-Iqbal

, Gong

(2004) Elevation of the level and activity of acid ceramidase in Alzheimer’s disease brain. Eur J Neurosci 20, 3489–3497.

48.

Gong

, Dong

, Rostad

, Marcovina

, Albers

, Brunzell

, Vuletic

(2013) Lipoprotein lipase (LPL) is associated with neurite pathology and its levels are markedly reduced in the dentate gyrus of Alzheimer’s disease brains. J Histochem Cytochem 61, 857–868.

49.

Zhao

, Tan

, Sheng

, Li

(2016) Identification of biomarkers associated with Alzheimer’s disease by bioin-formatics analysis. Am J Alzheimers Dis Other Demen 31, 163–168.

50.

Satoh

, Tabunoki

, Ishida

, Saito

, Arima

(2013) Accumulation of a repulsive axonal guidance molecule RGMa in amyloid plaques: A possible hallmark of regen-erative failure in Alzheimer’s disease brains. Neuropathol Appl Neurobiol 39, 109–120.

51.

Goedert

, Ghetti

, Spillantini

(2012) Frontotem-poral dementia: Implications for understanding Alzheimer disease. Cold Spring Harb Perspect Med 2, a006254.

52.

Lizard

, Rouaud

, Demarquoy

, Cherkaoui-Malki

, Iuliano

(2012) Potential roles of peroxisomes in Alzheimer’s disease and in dementia of the Alzheimer’s type. J Alzheimers Dis 29, 241–254.

53.

Silva

, Furuya

, Braga

, Rasmussen

, Labio

, Bertolucci

, Chen

, Turecki

, Mechawar

, Payao

, Mill

, Smith

(2014) Analysis of HSPA8 and HSPA9 mRNA expression and promoter methylation in the brain and blood of Alzheimer’s disease patients. J Alzheimers Dis 38, 165–170.

54.

Soilu-Hanninen

, Broberg

, Roytta

, Mattila

, Rinne

, Hukkanen

(2010)ExpressionofLIFandLIFreceptor beta in Alzheimer’s and Parkinson’s diseases. Acta Neurol Scand 121, 44–50.

55.

, Zhao

, Pan

, Wang

, Cui

, Shi

, Wang

, Yin

, Zhou

, Yang

(2015) Proteomic analysis of protein expression throughout disease progression in a mouse model of Alzheimer’s disease. J Alzheimers Dis 47,915–926.

56.

Yamamoto

, Shimada

, Koh

, Ataka

, Miki

(2014) Serum levels of albumin-amyloid beta complexes are decreased in Alzheimer’s disease. Geriatr Gerontol Int 14,716–723.

57.

Miners

, Clarke

, Love

(2017) Clusterin levels are increased in Alzheimer’s disease and influence the regional distribution of Abeta. Brain Pathol 27, 305–313.

58.

Prikrylova Vranova

, Henykova

, Mares

, Kaiserova

, Mensikova

, Vastik

, Hlustik

, Zapletalova

, Strnad

, Stejskal

, Kanovsky

(2016) Clusterin CSF levels in differential diagnosis of neurodegenerative disorders. J Neurol Sci 361, 117–121.

59.

, Watts

, Tailor

, Nguyen

, Howell

, Lee

, Seyfried

, Gearing

, Hales

, Levey

, Lah

, Lee

, Alzheimer’s Disease Neuroimaging Initiative (2016) CSF complement 3 and factor H are staging biomarkers in Alzheimer’s disease. Acta Neuropathol Commun 4, 14.

60.

Lian

, Yang

, Cole

, Sun

, Chiang

, Fowler

, Shim

, Rodriguez-Rivera

, Taglialatela

, Jankowsky

, Lu

, Zheng

(2015) NF-kappaB-activated astroglial release of complement C3 compromises neuronal morphology and function associated with Alzheimer’s disease. Neuron 85, 101–115.

61.

Perrin

, Craig-Schapiro

, Malone

, Shah

, Gilmore

, Davis

, Roe

, Peskind

, Li

, Galasko

, Clark

, Quinn

, Kaye

, Morris

, Holtz-man

, Townsend

, Fagan

(2011) Identification and validation of novel cerebrospinal fluid biomark-ers for staging early Alzheimer’s disease. PLoS One 6, e16032.

62.

Uberti

, Cenini

, Bonini

, Barcikowska

, Styczyn-ska

, Szybinska

, Memo

(2010) Increased CD44 gene expression in lymphocytes derived from Alzheimer disease patients. Neurodegener Dis 7, 143–147.

63.

, Jiang

, Wang

, Zhang

(2013) Meta-analysis of the association between urokinase-plasminogen activator gene rs2227564 polymorphism and Alzheimer’s disease. Am J Alzheimers Dis Other Demen 28, 517–523.

64.

Liu

, Cao

(2016) Characteristics and significance of the pre-metastatic niche. Cancer Cell 30, 668–681.

65.

Hamilton

, Dufresne

, Joppe

, Petryszyn

, Aumont

, Calon

, Barnabe-Heider

, Furtos

, Parent

, Chaurand

, Fernandes

(2015) Aberrant Lipid metabolism in the forebrain niche suppresses adult neural stem cell proliferation in an animal model of Alzheimer’s disease. Cell Stem Cell 17, 397–411.

66.

Liu

, Zhang

(2014) Lipid metabolism in Alzheimer’s disease. Neurosci Bull 30, 331–345.

67.

Giannopoulos

, Joshi

, Pratico

(2014) Novel lipid signaling pathways in Alzheimer’s disease pathogenesis. Biochem Pharmacol 88, 560–564.

68.

Díaz

, Fabelo

, Martín

, Ferrer

, Gómez

, Marín

(2015) Biophysical alterations in lipid rafts from human cerebral cortex associate with increased BACE1/AβPP interaction in early stages of Alzheimer’s disease. J Alzheimers Dis 43, 1185–1198.

69.

Tai

, Liu

, Li

, Holscher

(2017) Neuroprotective effects of a triple GLP-1/GIP/glucagon receptor agonist in the APP/PS1 transgenic mouse model of Alzheimer’s disease. Brain Res 1678, 64–74.

70.

Wang

, Tan

, Yu

(2015) Axonal transport defects in Alzheimer’s disease. Mol Neurobiol 51, 1309–1321.

71.

Qian

, Shen

, Wang

(2016) The distinct role of ADAM17 in APP proteolysis and microglial activation related to Alzheimer’s disease. Cell Mol Neurobiol 36, 471–482.

72.

Cortes-Canteli

, Paul

, Norris

, Bronstein

, Ahn

, Zamolodchikov

, Bhuvanendran

, Fenz

, Strickland

(2010) Fibrinogen and beta-amyloid association alters thrombosis and fibrinolysis: A possible contributing factor to Alzheimer’s disease. Neuron 66, 695–709.

73.

Gowrishankar

, Yuan

, Wu

, Schrag

, Paradise

, Grutzendler

, De Camilli

, Ferguson

(2015) Massive accumulation of luminal protease-deficient axonal lyso-somes at Alzheimer’s disease amyloid plaques. Proc Natl Acad Sci USA 112, E3699–3708.

74.

Jefferies

, Price

, Biron

, Fenninger

, Pfeifer

, Dickstein

(2013) Adjusting the compass: New insights into the role of angiogenesis in Alzheimer’s disease. Alzheimers Res Ther 5, 64.

75.

Vagnucci

Jr , Li

(2003) Alzheimer’s disease and angiogenesis. Lancet 361, 605–608.

76.

Qin

, Jia

, Wang

, Zuo

, Wu

, Zhou

, Li

, Min

, Wei

, Tang

, Xing

, Dong

, Wang

, Gao

, Li

, Jia

(2015) Elevated plasma angiogenesis factors in Alzheimer’s disease. J Alzheimers Dis 45, 245–252.

77.

Lynn

, Wang

, Markesbery

, Lovell

(2010) Quantitative changes in the mitochondrial proteome from subjects with mild cognitive impairment, early stage, and late stage Alzheimer’s disease. J Alzheimers Dis 19, 325–339.

78.

Miyajima

, Kono

, Takahashi

, Sugimoto

(2002) Increased Lipid Peroxidation and Mitochondrial Dysfunction in Aceruloplasminemia Brains. Blood Cells Mol Dis 29, 433–438.

79.

Marí

, Morales

, Colell

, Garcia-Ruiz

, Fernández-Checa

(2014) Mitochondrial cholesterol accumulation in alcoholic liver disease: Role of ASMase and endoplas-mic reticulum stress. Redox Biol 3, 100–108.

80.

Burte

, De Girolamo

, Hargreaves

, Billett

(2011) Alterations in the mitochondrial proteome of neuroblastoma cells in response to complex 1 inhibition. J Proteome Res 10, 1974–1986.

81.

Trougakos

, Lourda

, Antonelou

, Kletsas

, Gorgoulis

, Papassideri

, Zou

, Margaritis

, Boothman

, Gonos

(2009) Intracellular clus-terin inhibits mitochondrial apoptosis by suppressing p53-activating stress signals and stabilizing the cytoso-lic Ku70-Bax protein complex. Clin Cancer Res 15, 48–59.

82.

Lukiw

, Kenney

, Chwa

, Atilano

, Pavlis

, Falatoonzadeh

, Ramirez

, Malik

, Hsu

, Woo

, Soe

, Nesburn

, Boyer

, Kuppermann

, Jazwinski

, Miceli

, Wallace

, Udar

(2013) Mitochondrial DNA variants mediate energy production and expression levels for CFH, C3 and EFEMP1 genes: Implications for age-related macular degeneration. PLoS One 8, e54339.

83.

Bredesen

(2015) Metabolic profiling distinguishes three subtypes of Alzheimer’s disease. Aging 7, 595–600.

84.

Kodis

, Choi

, Swanson

, Ferreira

, Bloom

(2018) N-methyl-D-aspartate receptor-mediated calcium influx connects amyloid-β oligomers to ectopic neuronal cell cycle reentry in Alzheimer’s disease. Alzheimers Dementia 14, 1302–1312.

85.

Seong

, Yuan

, Arikkath

(2015) Cadherins and catenins in dendrite and synapse morphogenesis. Cell Adh Migr 9, 202–213.

86.

Bacchetti

, Vignini

, Giulietti

, Nanetti

, Provinciali

, Luzzi

, Mazzanti

, Ferretti

(2015) Higher levels of oxidized low density lipoproteins in Alzheimer’s disease patients: Roles for platelet activating factor acetyl hydrolase and paraoxonase-1. J Alzheimers Dis 46, 179–186.

87.

Stephan

, Matthews

, Ma

, Muniz

, Hunter

, Davis

, McKeith

, Foster

, Ince

, Brayne

(2012) Alzheimer and vascular neuropathological changes associated with different cognitive States in a non-demented sample. J Alzheimers Dis 29, 309–318.

88.

Howard

, Pilkington

(1992) Fibronectin staining detects micro-organisms in aged and Alzheimer’s disease brain. Neuroreport 3, 615–618.

89.

Kitamura

, Usami

, Ichihara

, Kida

, Satoh

, Tomi-moto

, Murata

, Oikawa

(2017) Plasma protein profiling for potential biomarkers in the early diagnosis of Alzheimer’s disease. Neurol Res 39, 231–238.

90.

Boddaert

, Kinugawa

, Lambert

, Boukhtouche

, Zoll

, Merval

, Blanc-Brude

, Mann

, Berr

, Vilar

, Garabedian

, Journiac

, Charue

, Silvestre

, Duyckaerts

, Amouyel

, Mariani

, Tedgui

, Mallat

(2007) Evidence of a role for lactadherin in Alzheimer’s disease. Am J Pathol 170, 921–929.

91.

Alessenko

, Bugrova

, Dudnik

(2004) Connection of lipid peroxide oxidation with the sphingomyelin pathway in the development of Alzheimer’s disease. Biochem Soc Trans 32, 144–146.

92.

Rygiel

(2016) Can angiotensin-converting enzyme inhibitors impact cognitive decline in early stages of Alzheimer’s disease? An overview of research evidence in the elderly patient population. J Postgrad Med 62, 242–248.

93.

Gard

, Fidalgo

, Lotter

, Richardson

, Farina

, Rusted

, Tabet

(2017) Changes of renin-angiotensin system-related aminopeptidases in early stage Alzheimer’s disease. Exp Gerontol 89, 1–7.

94.

Kaur

, Pawlik

, Gandy

, Ehrlich

, Smiley

, Levy

(2017) Lysosomal dysfunction in the brain of a mouse model with intraneuronal accumulation of carboxyl terminal fragments of the amyloid precursor protein. Mol Psychiatry 22, 981–989.

95.

Armstrong

, Mattsson

, Appelqvist

, Janefjord

, Sandin

, Agholme

, Olsson

, Svensson

, Blennow

, Zetterberg

, Kagedal

(2014) Lysosomal network proteins as potential novel CSF biomarkers for Alzheimer’s disease. Neuromolecular Med 16, 150–160.

96.

McGeer

, McGeer

, Akiyama

, Harrop

(1989) Cortical glutaminase, beta-glucuronidase and glucose utilization in Alzheimer’s disease. Can J Neurol Sci 16, 511–515.

97.

Puertas Mdel

, Martínez-Martos

, Cobo

, Lorite

, Sandalio

, Palomeque

, Torres

, Carrera-González

, Mayas

, Ramírez-Expósito

(2013) Plasma renin-angiotensin system-regulating aminopepti-dase activities are modified in early stage Alzheimer’s disease and show gender differences but are not related to apolipoprotein E genotype. Exp Gerontol 48, 557–564.

98.

Chapuis

, Vingtdeux

, Campagne

, Davies

, Maram-baud

(2011) Growth arrest-specific 1 binds to and controls the maturation and processing of the amyloid-beta precursor protein. Hum Mol Genet 20, 2026–2036.

99.

Hare

, Doecke

, Faux

, Rembach

, Volitakis

, Fowler

, Grimm

, Doble

, Cherny

, Masters

, Bush

, Roberts

(2015) Decreased plasma iron in Alzheimer’s disease is due to transferrin desaturation. ACS ChemNeurosci 6, 398–402.

100.

Haldar

, Beveridge

, Wong

, Singh

, Galimberti

, Borroni

, Zhu

, Blevins

, Greenlee

, Perry

, Mukhopadhyay

, Schmotzer

, Singh

(2013) A low-molecular-weight ferroxidase is increased in the CSF of sCJD cases: CSF ferroxidase and transferrin as diagnostic biomarkers for sCJD. Antioxid Redox Signal 19, 1662–1675.

101.

Lieberman

, Schleissner

, Tachiki

, Kling

(1995) Serum alpha 1-antichymotrypsin level as a marker for Alzheimer-type dementia. Neurobiol Aging 16, 747–753.

102.

Morello

, Tonoli

, Logrand

, Fiorito

, Fagoonee

, co

, Silengo

, Vercelli

, Altruda

, Tolosano

(2009) Haemopexin affects iron distribution and ferritin expression in mouse brain. J Cell Mol Med 13,4192–4204.

103.

Manral

, Sharma

, Hariprasad

, Chandralekha , Tripathi

, Srinivasan

(2012) Can apolipoproteins and complement factors be biomarkers of Alzheimer’s disease? Curr Alzheimer Res 9, 935–943.

104.

Grüninger-Leitch

, Schlatter

, Küng

, Nelböck

, Döbeli

(2002) Substrate and inhibitor profile of BACE (beta-secretase) and comparison with other mammalian aspartic proteases. J Biol Chem 277, 4687–4693.

105.

Daborg

, Andreasson

, Pekna

, Lautner

, Hanse

, Minthon

, Blennow

, Hansson

, Zetterberg

(2012) Cerebrospinal fluid levels of complement proteins C3, C4 and CR1 in Alzheimer’s disease. J Neural Transm (Vienna) 119, 789–797.

106.

Bien

, Jefferson

, Causevic

, Jumpertz

, Munter

, Multhaup

, Weggen

, Becker-Pauly

, Pietrzik

(2012) The metalloprotease meprin beta generates amino terminal-truncated amyloid beta peptide species. J Biol Chem 287,33304–33313.

107.

Akiyama

HIK

, Kondo

, McGeer

(1992) Thrombin accumulation in brains of patients with Alzheimer’s disease. Neurosci Lett 146, 152–154.

108.

Arai

TMJ

, Klegeris

, Guo

, McGeer

(2006) Thrombin and prothrombin are expressed by neurons and glial cells and accumulate in neurofibrillary tangles in Alzheimer disease brain. J Neuropathol Exp Neurol 65, 19–25.

109.

Lewczuk

PWJ

, Lange

, Jahn

, Reiber

, Ehrenreich

(1999) Prothrombin concentration in the cerebrospinal fluid is not altered in Alzheimer’s disease. Neurochem Res 24, 1531–1534.

110.

Butler

, Hwang

, Estick

, Nishiyama

, Kumar

, Baveghems

, Young-Oxendine

, Wisniewski

, Charalambides

, Bahr

(2011) Protective effects of positive lysosomal modulation in Alzheimer’s disease transgenic mouse models. PLoS One 6, e20501.

111.

Jensen

, Arvaniti

, Mikkelsen

, Michalski

, Pinborg

, Hartig

, Thomsen

(2015) Prostate stem cell antigen interacts with nicotinic acetylcholine receptors and is affected in Alzheimer’s disease. Neurobiol Aging 36, 1629–1638.

112.

Hafner

, Glavan

, Obermajer

, Zivin

, Schliebs

, Kos

(2013) Neuroprotective role of gamma-enolase in microglia in a mouse model of Alzheimer’s disease is regulated by cathepsin X. Aging Cell 12, 604–614.

113.

Vishnu

, Modi

, Sharma

, Mohanty

, Goyal

, Lal

, Khandelwal

, Mittal

, Prabhakar

(2016) Role of plasma clusterin in Alzheimer’s disease-a pilot study in a tertiary hospital in Northern India. PLoS One 11, e0166369.

114.

Liang

, Russell

, Mitra

, Chung

, Hye

, Bazenet

, Lovestone

, Pike

, Ward

(2015) Glycosyla-tion of human plasma clusterin yields a novel candidate biomarker of Alzheimer’s disease. J Proteome Res 14, 5063–5076.

115.

Choi

, Hou

, Choe

, Lee

(2013) Targeted human cerebrospinal fluid proteomics for the validation of multiple Alzheimer’s disease biomarker candidates. J Chromatogr B Analyt Technol Biomed Life Sci 930, 129–135.

116.

Hanzel

, Iulita

, Eyjolfsdottir

, Hjorth

, Schultzberg

, Eriksdotter

, Cuello

(2014) Analysis of matrix metallo-proteases and the plasminogen system in mild cognitive impairment and Alzheimer’s disease cere-brospinal fluid. J Alzheimers Dis 40, 667–678.

117.

, Lee

, Song

, Park

, Kim

(2014) Plasminogen activator inhibitor-1 as an early potential diagnostic marker for Alzheimer’s disease. Exp Gerontol 60, 87–91.

Early Candidate Urine Biomarkers for Detecting Alzheimer’s Disease Before Amyloid-β Plaque Deposition in an APP (swe)/PSEN1 dE9 Transgenic Mouse Model

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Experimental animals

Urinary protein extraction

In-gel protein digestion

Filter-aided protein digestion

Peptide tandem mass tag (TMT) labeling

HPLC separation

LC-MS/MS analysis

Data analysis

Statistical analysis

RESULTS AND DISCUSSION

Urinary proteome changes in APP (swe)/PSEN1dE9 transgenic mice

Functional analysis of differential urine proteins in APP (swe)/PSEN1dE9 transgenic mouse

IPA of differential proteins in AD development

CONCLUSION

Footnotes

ACKNOWLEDGMENTS

References

Urinary proteome changes in APP (swe)/PSEN1^dE9 transgenic mice

Functional analysis of differential urine proteins in APP (swe)/PSEN1^dE9 transgenic mouse