Distinct amyloid precursor protein processing machineries of the olfactory system

Accepted Manuscript Distinct amyloid precursor protein processing machineries of the olfactory system Jae Yeon Kim, Ameer Rasheed, Seung-Jun Yoo, So Yeun Kim, Bongki Cho, Gowoon Son, Seong-Woon Yu, Keun-A. Chang, Yoo-Hun Suh, Cheil Moon PII:

S0006-291X(17)32138-1

DOI:

10.1016/j.bbrc.2017.10.153

Reference:

YBBRC 38770

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 23 October 2017 Accepted Date: 28 October 2017

Please cite this article as: J.Y. Kim, A. Rasheed, S.-J. Yoo, S.Y. Kim, B. Cho, G. Son, S.-W. Yu, K.A. Chang, Y.-H. Suh, C. Moon, Distinct amyloid precursor protein processing machineries of the olfactory system, Biochemical and Biophysical Research Communications (2017), doi: 10.1016/ j.bbrc.2017.10.153. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Distinct Amyloid Precursor Protein Processing Machineries of the Olfactory System

Jae Yeon Kim1, Ameer Rasheed1, Seung-Jun Yoo1,2, So Yeun Kim1,2, Bongki Cho1,2, Gowoon Son1,

1

RI PT

Seong-Woon Yu1, Keun-A Chang3, Yoo-Hun Suh3, Cheil Moon1,2*

Department of Brain and Cognitive Sciences, Graduate School, Daegu Gyeungbuk Institute of

Science and Technology, Daegu, Korea; 2Convergence Research Advanced Centre for Olfaction,

M AN U

School of Medicine, Gachon Medical School, Incheon, Korea.

SC

Daegu Gyeungbuk Institute of Science and Technology, Daegu, Korea, 3Department of Pharmacology,

*

Correspondence: Cheil Moon, Department of Brain and Cognitive Sciences, Graduate School, Daegu

Gyeungbuk Institute of Science and Technology, 333, Techno Jung-Ang Daero, Hyeonpung-Myeon, Dalseong-Gun, Daegu, 711-873, Korea. Tel: +82-53-785-1040; Fax: +82-53-785-6109; E-mail:

AC C

EP

TE D

[email protected]

1

ACCEPTED MANUSCRIPT Abstract

Processing of amyloid precursor protein (APP) occurs through sequential cleavages first by βsecretase and then by the γ-secretase complex. However, abnormal processing of APP leads to excessive production of β-amyloid (Aβ) in the central nervous system (CNS), an event which is

RI PT

regarded as a primary cause of Alzheimer’s disease (AD). In particular, gene mutations of the γsecretase complex—which contains presenilin 1 or 2 as the catalytic core—could trigger marked Aβ accumulation.

Olfactory dysfunction usually occurs before the onset of typical AD-related symptoms (eg, memory loss or muscle retardation), suggesting that the olfactory system may be one of the most

SC

vulnerable regions to AD. To date however, little is known about why the olfactory system is affected so early by AD prior to other regions. Thus, we examined the distribution of secretases and levels of

M AN U

APP processing in the olfactory system under either healthy or pathological conditions. Here, we show that the olfactory system has distinct APP processing machineries. In particular, we identified higher expressions levels and activity of γ-secretase in the olfactory epithelium (OE) than other regions of the brain. Moreover, APP c-terminal fragments (CTF) are markedly detected. During AD progression, we note increased expression of presenilin2 of γ- secretases in the OE, not in the OB, and show that neurotoxic Aβ*56 accumulates more quickly in the OE.

TE D

Taken together, these results suggest that the olfactory system has distinct APP processing machineries under healthy and pathological conditions. This finding may provide a crucial understanding of the unique APP-processing mechanisms in the olfactory system, and further

EP

highlights the correlation between olfactory deficits and AD symptoms.

AC C

Keyword: Alzheimer’s Disease(AD), Amyloid Precursor Protein(APP), olfactory system, Olfactory Epithelium(OE), γ-Secretase, presenilin.

2

ACCEPTED MANUSCRIPT Introduction Alzheimer's disease (AD) is a neurodegenerative disorder characterized by progressive memory loss and cognitive decline. AD pathology is attributed to abnormal processing of amyloid precursor protein (APP) and the accumulation of β-amyloid (Aβ) peptides [1]. APP proteolysis occurs in sequence by a number of enzymes (ie, α-, β-, γ-, and η- secretases). α-Secretase is known to cut the

RI PT

Aβ domain, and the resulting cleaved fragments have been shown to induce a neuro-protective effect [2]. Aβ is generated following β- and γ-secretase-mediated cleavage. γ-Secretase is composed of catalytic subunits (ie, presenilin1 (PS1) or presenilin2 (PS2)) and regulatory subunits (ie, nicastrin, presenilin enhcancer2 (Pen2) and Aph1) [3,4]. Under certain physiological conditions, presenilin1 and

SC

2 undergo endo-proteolytic cleavage for γ-secretase enzymatic activity [5,6,7]. Recently identified ηsecretase has the matrix metalloproteinase-24 (MMP24) as catalytic core and generates c-terminal fragment η (CTFη) proteins which are known to be accumulated in the brain of patients with AD [8].

M AN U

Presenilins of γ-secretase are reported as only gene mutations of secretases in AD patients. For example, 179 PSEN1 (presenilin 1 gene locus) and 14 PSEN2 gene mutations result in early-onset familial AD [9]. In the case of BACE1, the level itself is related to late-onset sporadic AD [6]. However, the relationship between each secretases and pathological symptoms is still unclear. One of the representative and initial symptoms of AD is olfactory decline [10]. The olfactory system consists of olfactory epitheilum (OE) as the peripheral region directly connected to the central

TE D

nervous system (CNS) and olfactory bulb (OB) as the part of the brain. Since OE is exposed directly to the external enviornment, OE might have discriminative properties from the brain. In addtion, it has been suggested that OE may be more affected regions during AD progression and hence OE could be useful in the diagnosis of AD onset [11,12,13,14,15]. Nevertheless, the precise molecular mechanisms

EP

underlying these observations is still largely unknown. We hypothesize that the olfactory system may have machineries and APP-processing

AC C

pathways that are unique from other regions of the brain, and these differences are likely to reflect olfactory loss during AD progression. In the current study, we first screened the relative distribution of secretases and APP fragments within the olfactory system. Furthermore, we also measured changes in APP machineries at early pathologocal stages using the Tg2576 mice.

3

ACCEPTED MANUSCRIPT Materials and Methods Animals

RI PT

Heterozygous Tg2576 mice, which express a human amyloid-β precursor protein (APP) variant linked to AD, were used as developed and described previously[16]. We obtained these mice from Taconic (USA). Age-matched non-transgenic littermates were served as the wild-type. All animal experiments were approved and conducted in accordance with guidelines of Ethic Committee of Seoul National University’ (IACUC No. SNU-091208-1).

RT-PCR

M AN U

SC

Real-time PCR reaction samples were prepared using cDNA and primers via QuantiTech® SYBR Green PCR Kit (cat#204141) with a final volume of 20 µl. The reaction was run in a real-time PCR machine (Rotor Gene Q, Qiagen, Germany) with following parameters: 95°C for 30 sec, 59°C for 30 sec and 72°C for 30 sec (35 cycles). Analysis and calculation were made through the ∆∆Ct method. Cyclophilin A was used for normalization and determination of the ∆Ct values. Further results were relatively quantified as suggested previously [17] using software (Rotor Gene Q series software). Data represent mean ± SEM and statistical analysis was performed using student’s t-test (two-tailed). Values *P< 0.05, **P < 0.01, and ***P < 0.001 were considered statistically significant. Western blot

EP

TE D

For Western blotting analysis, samples were homogenized in RIPA buffer (ThermoScientifc, USA) containing 1% of protease/phosphatase inhibitor cocktail (HaltTM, ThermoScientifc, USA), and APP CTF was detected following the published protocol [18]. Primary antibodies, including 6E10 (Covance, SIG-39320, Princeton, NJ, USA), APP C-terminal (Millipore, AB5352, Billerica, MA, USA), GAPDH (Millipore, MAB374, Billerica, MA, USA), Presenilin1 (Millipore, MAB5232) and Presenilin2 (Cell Signaling, 2192) were used for this study.

Primary culture of olfactory sensory neurons (OSNs)

AC C

Primary cultures were prepared as previously described with a few modifications[19]. Precursor cells from nasal turbinates of 0.5 day-old rat pup (Sprague–Dawley rat) were cultured at a density of 2 × 106 cells/mL on tissue culture dishes (Falcon, Lincoln Park, NJ, USA) coated with 25 µg/mL laminin (BD Bioscience, San Diego, CA, USA) in modified Eagle's medium containing d-valine (MDV, Welgen Inc., Worcester, MA, USA). On 2 days in vitro every day thereafter, cells were incubated with MDV containing 15% dialyzed fetal bovine serum (Gibco, Rockville, MD, USA), gentamicin, kanamycin and 2.5 ng/mL nerve growth factor (NGF). Two days prior to experiment, the culture medium is changed to NGF free medium.

Immunocytochemistry (ICC) At DIV3, olfactory receptor neurons (ORNs) are fixed by 4% paraformaldehyde (PFA) for 15 minutes and permeated with 0.05% PBS-T (Triton X-100) for 5 minutes. By using the 4% normal donkey

4

ACCEPTED MANUSCRIPT serum in PBS, ORNs were blocked for 1 hour. The primary and secondary antibodieswere incubated overnight at 4°C. At each step, ORNS were washed three times by PBS (5 minutes each).

RI PT

Results The olfactory system has APP processing machineries that are unique from other brain regions. To examine the distribution of secretases in the olfactory system, we monitored mRNA levels of the catalytic cores of secretases in the OE and OB (Fig.1A). There were no differences in the expression levels of ADAM10 and BACE1 between the OE and the OB, whereas levels of MMP24

SC

expression were lower in the OE than in the OB. Interestingly, both presenilin1 and 2 were more highly expressed in the OE than in the OB. Furthermore, CTF protein, the product of presenilin1 and

M AN U

2, was also expressed in the OE (Fig.1B). Since presenilin activation is necessary to cleave and generate CTF proteins of its own [20], levels of presenilin CTF proteins have been regarded as an indirect marker for presenilin activation. Using the dissociated culture of ORNs, we confirmed that presenilin1 and 2 were located in the soma and neurites of the neuron specific tubulin (NST)-positive neurons (Fig.1C). Our results indicate that secretases have distinct distributions within the olfactory system, and γ-secretase, in particular, presenilin2 is highly expressed and active in the OE.

TE D

Since APP fragments are generated by secretases, distinct distribution of secretases may be closely related to various APP fragments produced in the olfactory system. α-, η- and γ-secretases are known to produce C83 (15 kDa), CTF- η (30 kDa) and AICD (10 kDa), respectively [8], and all these CTFs can be detected using an antibody specific to the c-terminus of APP [21]. Using an antibody

EP

specific to the c-terminus of APP, we observed noticeably different expression patterns of APP fragments in the OE compared with other regions of the central nervous system. Moreover, unknown

AC C

modifications of native APP form near the 70 kDa were observed only in the OE. The 30 kDa band— likely produced by η-secretase—was observed in the OE, and the amount of 30 kDa protein production was less than in other brain regions. Interestingly, CTF fragments (eg C99, C83 and AICD), were more abundant in the OE (Fig. 1D). Taken together, the olfactory system may have different APP processing that appear to be consistent with the distributions of secretases.

The OE displays unregulated APP processing under pathological conditions Next, we examined the expression patterns of APP-processing machineries as well as APP processing in the olfactory system under pathological conditions. We checked mRNA expression levels of ADAM10, BACE1, MMP24, presenilin1 and 2 in the OE and OB of 10 month-old WT and 5

ACCEPTED MANUSCRIPT Tg2576 mice (Fig. 2A,B). 10 month-old Tg2576 mice overexpressing the hAPP variant linked to AD show olfactory dysfunction [22]. While there was no significant difference between expression levels of ADAM10 and BACE1, MMP24, presenilin1 and 2 levels were significantly increased in the OE of Tg2576 mice. In the OB of Tg2576 mice, the expression levels of ADAM10, MMP24 and presenilin1 were increased more than twice, wherease the expression levels of BACE1 and presenilin2 were not

RI PT

changed. This observation indicates that, the OE and OB under pathological conditions express different machineries for APP processing, that is, the changes in the expression of α- and γ-secretases. In addition, we examined the production of APP fragments in the OE and other brain regions of 10-month old WT and Tg2576 mice. 10-month old Tg2576 mice showed dramatic differences in APP

SC

processing in the OE. Interestingly, a secondary maturation form of APP (a band of near 150 kDa) was only observed in the OE of 6-month old Tg2576 mice (data not shown) but not in the OE of 10-

M AN U

month old Tg2576 mice (Fig. 2C). The expression levels of the 30 kDa (CTF- η) fragment was decreased about 27% and the expression level of the 15 kDa (C99) protein was dramatically increased 2 fold in the OE. Crude multi-bands smaller than 15 kDa were also observed in the OE of 10-month old Tg2576 mice (Fig. 2C). Of particular interest, pathological APP fragments resulting from unusual APP processing in the OE were also observed. We detected Aβ*56(12-mer, ~56 kDa) in the OE, OB and cortex of 10-month old WT and Tg2576 mice (Fig. 2B). Because Aβ*56 is neurotoxic without

TE D

forming the neurite plaques or neurofibrillary tangles, increased accumulation of Aβ*56 have been observed in the initial stages of AD [23,24]. Here we have shown that the amount of toxic Aβ*56 was significantly increased (about 1.5 fold) in the OE of Tg2576 mice. On the other hand, no such change between WT and Tg2576 mice was found in the OB. These changes indicate that unusual APP

AC C

EP

processing may induce the accumulation of neurotoxic Aβ*56 in the OE during AD development.

Discussion

In the current study, we found that the olfactory system has unique APP-related machineries and APP processing under both normal and pathological conditions. This study suggests that unusual secretases expression in the OE could mostly contribute to abnormal APP processing, leading to olfactory dysfunction. Therefore, we expect that the properties of the olfactory system itself, which are different from other brain regions, might provide insight into why the olfactory system is so vulnerable to AD. Predominant expression of presenilin1 and 2 in OE is very intriguing, since it has been recently proposed that γ-secretase may also be involved in notch signaling in addition to APP processing and 6

ACCEPTED MANUSCRIPT subsequent AD pathogenesis [25]. Notch signaling is an important pathway for normal development and neuronal survival of the OE and the high expression of γ-secretase under healthy conditions may be mainly involved in inducing a pro neuro-protective environment [26]. However, it could lead to the amyloidogenic condition like AD similarly to our study, if the increased expression of γ-secretase coincides with APP. To date, it is still unknown how the balance of neuro-protective vs pathological

RI PT

activities of γ-secretase are regulated and dysregulated may result in pathological conditions. BACE1, a part of ß-secretase, is known to be a negative regulator in the AD model [27,28], and studies have shown no differences between the expression levels of BACE1 mRNA in the OE of WT and TG mice (Fig2A). Of interest to this group, we observed previously that elevated BACE1 activity is concentrated in the axonal terminal of the olfactory sensory neuron, and can leads to Aβ

SC

accumulation and cell death [29]. This inconsistency can be explained by the distinction between the expression site and activity site of BACE1. Here, we used the OE organ as a whole and measured the

M AN U

mRNA levels expressed in cell body. Therefore, BACE1 might be more closely related to its protein activity rather than expression itself under pathological conditions.

Although γ-secretase is also necessary to process the APP and generate the Aβ fragment, no specific roles or activation mechanisms of the γ-secretase in the olfactory system have been reported. Here, we showed, for the first time, that γ-secretase activity in the OE is higher than in other brain regions. And

presenilin1 of γ-secretase is more expressed in both the OE and OB under pathological

TE D

condition, while the expression of presenilin2 is signlificantly increased only in the OE, not in the OB. Further studies may help better elucidate the correlations between BACE1, presenilin1 and presenilin2 of γ-secretase in the olfactory system during AD progression. It has been identified many toxic Aβ oligomers in AD patients [23]. Of those, Aβ*56(12-mer), mainly found in the cortex of the early stages of AD, is closely related into cognitive impairments

EP

[23,24]. However, we have shown a markedly increased Aβ*56 levels in the OE of 6-month old Tg2576 mice which have no cognitive defecit [29], and the Aβ*56 was also increased in the OE of 10

AC C

month-old Tg2576 mice and this was more markedly increased than the cortex. Considering our hypothesis that the olfactory system and the cortex may have different machineries for APP processing, we could think a toxic Aβ*56 level might be sufficient to explain the neurotoxicity and the vulnerability in the OE of AD mice. The most challenging task in AD research may be to discover additional therapies and ideally a cure, but also of importance is the improved diagnosis of AD before memory loss or cognitive decline appears. A number of biomarkers for disease screening in the preclinical stages have been suggested, including cerebrospinal fluid tau and amyloid beta. However, the use of these biomarkers is associated with some practical limitations (eg, invasiveness and high expense). Our study may enlighten the possibility for disease screening in the preclinical stages, since expressions of presenilin1 and 2 are increased exclusively in the OE under pathological conditions. It can be detected 7

ACCEPTED MANUSCRIPT in a part of the periphery nervous system (PNS) from presymptomatic stage of AD, based on our observation using Tg2576 mice [29]. Due to direct exposure of the PNS olfactory system to the external environment, monitoring of presenilins expression in the PNS olfactory system could be a novel and feasible biomarker for the diagnosis of AD. Furthermore, we can prevent at least olfactory dysfunction in advance by developing a drug to inhibit the activity of presenilin1 and 2 of the OE,

RI PT

although we should find the correlations between activation of presenilins in the OE and onset of AD in the CNS.

In summary, we have suggested a feasible explanation for the long unknown mechanism of olfactory dysfunction in AD. Our findings imply that APP-processing machineries and APP processing in the OE are different from other regions of the brain. Furthermore, the dysregulation of

SC

γ-secretase components in OE can be closely related to vulnerability of the OE in early stage of AD

AC C

EP

TE D

M AN U

progression.

8

ACCEPTED MANUSCRIPT Figure Legends Figure 1. Unique APP processing machineries in the olfactory system. (A) Analyses of ADAM10 (α-Secretase), BACE1 (β-Secretase), MMP24 (η-Secretase) and presenilin1 and 2(γ-Secretase) mRNA expression in the OE compared to the OB of 2-month old C57BL/6 mice. mRNA expression levels were determined by qRT-PCR and are presented as fold

RI PT

increase relative to OE. Data represent mean ± SEM and statistical analyses were performed using student’s paired t-test. Values *P< 0.05, **P < 0.01, were considered statistically significant.

(B) Tissue extracts of 2-month old C57BL/6 mice were analyzed by western blot for PS1 and 2. Antibodies for PS1 and 2 target epitopes at the C-terminals of PS1 and 2. Results are shown for one

SC

representative experiment of 3 performed.

(C) Localization of presenilin1 and 2 in the OE. Scale bar = 50μm. Immunocytochemistry of olfactory receptor neuron primary culture. At DIV3, ORNs are fixed with 4% PFA. NST is a neuron-specific

M AN U

class Ⅲ beta tubulin. Scale bar = 100µm. DIV3 ORNs are immature stage. The white dotted box is enlarged. Scale bar = 10μm.

(D) Unique pattern of APP expression and its fragments in the OE compared with CNS tissues extracts from ~ 2 month-old C57BL/6 normal mice following probing with c-terminus-specific antibodies. Results are shown for one representative experiment of three performed. OE: olfactory

TE D

epithelium, OB: olfactory bulb, H: hippocampus, Cort: cortex.

Figure2. Unregulated APP processing in the olfactory epithelium under pathological conditions

EP

(A). mRNA expression levels of ADAM10 (α-secretase), BACE1 (β-secretase), MMP24 (η-secretase), presenilin1 and 2(γ-secretase) in the OE of 10-month old WT and Tg2576 mice. mRNA expression

AC C

levels were determined using qRT-PCR and are presented as fold increase relative to OE. Data represent mean ± SEM and statistical analysis was performed using student’s paired t-test. Values *P< 0.05, **P < 0.01, were considered statistically significant. (B). mRNA expression levels of ADAM10 (α-secretase), BACE1 (β-secretase) and MMP24 (ηsecretase), presenilin1 and 2(γ-secretase) in the OB of 10-month old WT and Tg2576 mice. (C) For the detection of Aβ*56, A11 antibody was used for immunoprecipitation followed by blotting with the 6E10 antibody. (D) Expression pattern of APP and its fragments in OE compared with various brain regions extracts from 10-month old Tg2576 mice. Tissues were lysed using RIPA buffer followed by probing with anti-c-terminal antibody. Results are shown for one representative experiment of three performed.

9

ACCEPTED MANUSCRIPT Figure 3. Working hypothesis illustrating possible APP-processing machineries and olfactory system-related implications. Under healthy conditions, there are high levels of presenilin 1 and 2 expression and activity in OE. Under pathological conditions, OE have distinct increased expression of presenilin1 and 2(γ-secretase)

RI PT

and it could help explain the accumulation of neurotoxic Aβ*56 in the OE.

SC

Acknowledgements: This work was supported by the National Research Foundation of Korean

AC C

EP

TE D

M AN U

(NRF-2015M3A9E2028884)

10

ACCEPTED MANUSCRIPT References

AC C

EP

TE D

M AN U

SC

RI PT

[1] G. McKhann, D. Drachman, M. Folstein, R. Katzman, D. Price, E.M. Stadlan, Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease, Neurology 34 (1984) 939-944. [2] E.R. Vardy, A.J. Catto, N.M. Hooper, Proteolytic mechanisms in amyloid-beta metabolism: therapeutic implications for Alzheimer's disease, Trends Mol Med 11 (2005) 464-472. [3] A. Fukumori, R. Fluhrer, H. Steiner, C. Haass, Three-amino acid spacing of presenilin endoproteolysis suggests a general stepwise cleavage of gamma-secretase-mediated intramembrane proteolysis, J Neurosci 30 (2010) 7853-7862. [4] H. Steiner, R. Fluhrer, C. Haass, Intramembrane proteolysis by gamma-secretase, J Biol Chem 283 (2008) 29627-29631. [5] B. De Strooper, W. Annaert, P. Cupers, P. Saftig, K. Craessaerts, J.S. Mumm, E.H. Schroeter, V. Schrijvers, M.S. Wolfe, W.J. Ray, A. Goate, R. Kopan, A presenilin-1-dependent gammasecretase-like protease mediates release of Notch intracellular domain, Nature 398 (1999) 518-522. [6] R.J. O'Brien, P.C. Wong, Amyloid precursor protein processing and Alzheimer's disease, Annu Rev Neurosci 34 (2011) 185-204. [7] N. Takasugi, T. Tomita, I. Hayashi, M. Tsuruoka, M. Niimura, Y. Takahashi, G. Thinakaran, T. Iwatsubo, The role of presenilin cofactors in the gamma-secretase complex, Nature 422 (2003) 438-441. [8] M. Willem, S. Tahirovic, M.A. Busche, S.V. Ovsepian, M. Chafai, S. Kootar, D. Hornburg, L.D. Evans, S. Moore, A. Daria, H. Hampel, V. Muller, C. Giudici, B. Nuscher, A. Wenninger-Weinzierl, E. Kremmer, M.T. Heneka, D.R. Thal, V. Giedraitis, L. Lannfelt, U. Muller, F.J. Livesey, F. Meissner, J. Herms, A. Konnerth, H. Marie, C. Haass, eta-Secretase processing of APP inhibits neuronal activity in the hippocampus, Nature 526 (2015) 443-447. [9] T. Toda, Y. Noda, G. Ito, M. Maeda, T. Shimizu, Presenilin-2 mutation causes early amyloid accumulation and memory impairment in a transgenic mouse model of Alzheimer's disease, J Biomed Biotechnol 2011 (2011) 617974. [10] R.L. Doty, The olfactory system and its disorders, Semin Neurol 29 (2009) 74-81. [11] S.E. Arnold, B.T. Hyman, J. Flory, A.R. Damasio, G.W. Van Hoesen, The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer's disease, Cereb Cortex 1 (1991) 103-116. [12] R.L. Buckner, J. Sepulcre, T. Talukdar, F.M. Krienen, H. Liu, T. Hedden, J.R. Andrews-Hanna, R.A. Sperling, K.A. Johnson, Cortical hubs revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer's disease, J Neurosci 29 (2009) 1860-1873. [13] O. Guerin, S. Andrieu, S.M. Schneider, F. Cortes, C. Cantet, S. Gillette-Guyonnet, B. Vellas, Characteristics of Alzheimer's disease patients with a rapid weight loss during a six-year follow-up, Clin Nutr 28 (2009) 141-146. [14] W.W. Seeley, R.K. Crawford, J. Zhou, B.L. Miller, M.D. Greicius, Neurodegenerative diseases target large-scale human brain networks, Neuron 62 (2009) 42-52. [15] D.W. Wesson, E. Levy, R.A. Nixon, D.A. Wilson, Olfactory dysfunction correlates with amyloidbeta burden in an Alzheimer's disease mouse model, J Neurosci 30 (2010) 505-514. [16] K. Hsiao, P. Chapman, S. Nilsen, C. Eckman, Y. Harigaya, S. Younkin, F. Yang, G. Cole, Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice, Science 274 (1996) 99-102. [17] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method, Methods 25 (2001) 402-408. 11

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[18] M.S. Garcia-Ayllon, I. Lopez-Font, C.P. Boix, J. Fortea, R. Sanchez-Valle, A. Lleo, J.L. Molinuevo, H. Zetterberg, K. Blennow, J. Saez-Valero, C-terminal fragments of the amyloid precursor protein in cerebrospinal fluid as potential biomarkers for Alzheimer disease, Sci Rep 7 (2017) 2477. [19] G.V. Ronnett, L.D. Hester, S.H. Snyder, Primary culture of neonatal rat olfactory neurons, J Neurosci 11 (1991) 1243-1255. [20] M.S. Wolfe, Toward the structure of presenilin/gamma-secretase and presenilin homologs, Biochim Biophys Acta 1828 (2013) 2886-2897. [21] Z. Muresan, V. Muresan, A phosphorylated, carboxy-terminal fragment of beta-amyloid precursor protein localizes to the splicing factor compartment, Hum Mol Genet 13 (2004) 475-488. [22] M.A. Westerman, D. Cooper-Blacketer, A. Mariash, L. Kotilinek, T. Kawarabayashi, L.H. Younkin, G.A. Carlson, S.G. Younkin, K.H. Ashe, The relationship between Abeta and memory in the Tg2576 mouse model of Alzheimer's disease, J Neurosci 22 (2002) 1858-1867. [23] I. Benilova, E. Karran, B. De Strooper, The toxic Abeta oligomer and Alzheimer's disease: an emperor in need of clothes, Nat Neurosci 15 (2012) 349-357. [24] S. Lesne, M.T. Koh, L. Kotilinek, R. Kayed, C.G. Glabe, A. Yang, M. Gallagher, K.H. Ashe, A specific amyloid-beta protein assembly in the brain impairs memory, Nature 440 (2006) 352-357. [25] X. Zhang, Y. Li, H. Xu, Y.W. Zhang, The gamma-secretase complex: from structure to function, Front Cell Neurosci 8 (2014) 427. [26] Y. Orita, H. Nagatsuka, H. Tsujigiwa, J. Yoshinobu, Y. Maeda, M. Kakiuchi, S. Orita, A. Takeuchi, Y. Takeda, K. Fukushima, N. Nagai, K. Nishizaki, Expression of Notch1 and Hes5 in the developing olfactory epithelium, Acta Otolaryngol 126 (2006) 498-502. [27] R.M. Holsinger, J.S. Lee, A. Boyd, C.L. Masters, S.J. Collins, CSF BACE1 activity is increased in CJD and Alzheimer disease versus [corrected] other dementias, Neurology 67 (2006) 710-712. [28] C. Schubert, Alzheimer disease: BACE1 branches out, Nat Med 12 (2006) 1123. [29] S.J. Yoo, J.H. Lee, S.Y. Kim, G. Son, J.Y. Kim, B. Cho, S.W. Yu, K.A. Chang, Y.H. Suh, C. Moon, Differential spatial expression of peripheral olfactory neuron-derived BACE1 induces olfactory impairment by region-specific accumulation of beta-amyloid oligomer, Cell Death Dis 8 (2017) e2977.

12

ACCEPTED MANUSCRIPT Figure 1. The olfactory system has different APP processing machineries.

A

0.0

0.5

3.0

1.5

0.0

0.0

OE

OB

40

TE D EP

◄ CTF of PS1 ◄ CTF of PS2 ◄ ß-actin

AC C

OE OB 15 25

PS1/NST/DAPI

C

50μm

PS2/NST/DAPI

B

OE

OB

OE

SC

n.s

M AN U

0.5

1.0

MMP24 mRNA

1.0

50μm

OB

*

**

1.0

PS2 mRNA

**

n.s

γ-Secretase

RI PT

η-Secretase

PS1 mRNA

β-Secretase

BACE1 mRNA

ADAM10 mRNA

α-Secretase

0.5

0.0

1.0

0.5

0.0

OE

OB

D

OE

OB

CNS OE OB H CTX ◄ mature APP ◄ immature APP

130 100

70 35 25

◄ native APP

15

◄ CTF-η * * ◄ C99 ◄ C83

10

◄ AICD

40

◄ ß-actin

25

*

ACCEPTED MANUSCRIPT Figure 2. Unregulated APP processing in the olfactory epithelium under the pathological condition.

A

1.0

0

B

WT

TG

WT

RI PT 1.0

TG

WT

β-Secretase

4

1.0

**

PS1 mRNA

2

n.s

1.5

MMP24 mRNA

4

2.0

4

2 0

0

WT

TG

WT

TG

TG

WT

TG

4

*

2

0

WT

1.0

γ-Secretase

6

AC C

*

BACE1 mRNA

ADAM10 mRNA

6

1.5

TG

η-Secretase

EP

α-Secretase

TG

1.5

TE D

WT

*

PS2 mRNA

2

1.5

*

2.0

2.0

SC

n.s

γ-Secretase

PS2 mRNA

1.0

4

*

2.0

PS1 mRNA

n.s

6

M AN U

1.5

η-Secretase

MMP24 mRNA

2.0

β-Secretase

BACE1 mRNA

ADAM10 mRNA

α-Secretase

2

n.s

0

WT

TG

WT

TG

D

SC IB : α-c term

M AN U

40

◄ Aβ56 ◄ ß-actin

TE D

72

OE

130 100

EP

OE OB CTX WT TG WT TG WT TG

AC C

IP : A11 IB : 6E10

C

RI PT

ACCEPTED MANUSCRIPT Figure 2. Unregulated APP processing in the olfactory epithelium under the pathological condition.

70 35 25

OB

H

CTX

WT TG WT TG WT TG WT TG

◄ mAPP ◄ imAPP ◄ native APP ◄ CTF-η *

15

◄ C99

10

◄ AICD

40

◄ ß-actin

APP fragments

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT Figure 3. Model illustrating possible APP processing machineries in the olfactory system.

α,β-secretase : OE≒OB η-secretase : OE < OB γ-secretase : OE > OB

Aβ

γ-secretase: OE > OB

ACCEPTED MANUSCRIPT

Highlights

 Under healthy condition, active γ-secretase is highly expressed in the olfactory system.

RI PT

 The amyloid precursor protein(APP) c-terminal fragments(CTFs) are markedly detected in the olfactory epithelium(OE).  During Alzheimer disease(AD) progression, the expressions of presenilin2(γ-secretases) are increased in the OE, not in the olfactory bulb(OB), in turn neurotoxic Aβ*56 is accumulated more quickly in the OE.

AC C

EP

TE D

M AN U

SC

 Distinct APP processing machineries in the OE can be a feasible biomarker for the early diagnosis of AD onset.