A Free Radical-Generating System

Regulates Amyloid Oligomers: Involvement

of Cathepsin B

4 Patricia Llorentea,b, Henrike Kristena,b, Isabel Sastrea,b,c, Ana Toledano-Zaragozaa,

5 Jesu´s Aldudoa,b, Mar´ıa Recueroa,b,∗ and Mar´ıa J. Bullidoa,b,c,∗

6 a Centro de Biolog´ıa Molecular “Severo Ochoa” (CSIC-UAM), Universidad Auto´noma de Madrid, Madrid, Spain
7 bCentro de Investigacio´n Biome´dica en Red sobre Enfermedades Neurodegenerativas (CIBERNED),
98 Madrid, Spain
cInstituto de Investigacion Sanitaria “Hospital la Paz” (IdIPaz), Madrid, Spain

Accepted 30 September 2018

10 Abstract. Amyloid-β (Aβ), a major component of senile plaques, is generated via the proteolysis of amyloid-β protein
11 precursor (AβPP). This cleavage also produces AβPP fragment-derived oligomers which can be highly neurotoxic. AβPP
12 metabolism/processing is affected by many factors, one of which is oxidative stress (OS). Associated with aging, OS is
13 an important risk factor for Alzheimer’s disease. In addition, the protein degradation systems, especially those involving
14 cathepsins, are impaired in aging brains. Moreover, cathepsin B (CTSB) is a cysteine protease with potentially specific
15 roles in AβPP proteolysis (β-secretase activity) and Aβ clearance (Aβ degradative activity). The present work examines the
16 effect of OS and the involvement of CTSB in amyloid oligomer formation. The xanthine/xanthine oxidase (X-XOD) free
17 radical generating system induced the partial inhibition of CTSB activity, which was accompanied by an increase in large
18 amyloid oligomers. These were located throughout the cytosol and in endo-lysosomal vesicles. Cells treated with the CTSB
19 inhibitor CA-074Me also showed increased amyloid oligomer levels, whereas those subjected to OS in the presence of the
20 inhibitor showed no such increase. However, CTSB inhibition clearly modulated the AβPP metabolism/processing induced
21 by X-XOD, as revealed by the increase in intracellular AβPP and secreted α-secretase-cleaved soluble AβPP. The present
22 results suggest that CTSB participates in the changes of amyloid oligomer induced by mild OS.

23 Keywords: amyloid-β, amyloid-β protein precursor, cathepsin B, free radicals, oligomer, oxidative stress

24 INTRODUCTION

25 Alzheimer’s disease (AD) is a progressive neu-
26 rodegenerative disorder clinically characterized by
27 cognitive dysfunction and failing memory. The major
28 histochemical hallmarks of the disease are neurofib-
29 rillary tangles (NFTs) and plaques composed of
30 aggregated amyloid-β (Aβ) [1]. Aβ peptides are pro-
31 duced during the sequential proteolytic processing of
32 the amyloid-β protein precursor (AβPP) [2], a type

∗Correspondence to: Mar´ıa Recuero and Mar´ıa J. Bullido, Centro de Biolog´ıa Molecular “Severo Ochoa”, Universidad Auto´noma de Madrid, C/ Nicola´s Cabrera 1, 28049 Madrid, Spain. Tel.: (+34) 91 196 4674; Fax: (+34) 91 196 4420;
E-mails: [email protected] and [email protected].

1 transmembrane protein with its amino terminus in 33
the lumen/extracellular space and its carboxy termi- 34
nus in the cytosol. AβPP processing toward either 35
the amyloidogenic pathway (β- and γ-secretases), 36
and thus the generation of Aβ, or toward the non- 37
amyloidogenic pathway (α and γ-secretases), in 38
which no Aβ is produced [3], is directed by cellu- 39
lar levels of secretases, plus the trafficking of AβPP 40
to organelles that express these enzymes. 41
Aβ exists in two major peptide isoforms depend- 42
ing on its amino acids length, Aβ40 and Aβ42, with 43
the latter markedly more prone to accumulation and 44
aggregation [4]. Aβ42 firstly would form soluble 45
oligomers, which later would come to produce insol- 46
uble aggregates that could accumulate as amyloid 47

ISSN 1387-2877/18/$35.00 © 2018 – IOS Press and the authors. All rights reserved
48 fibrils. In AD brains, these go on to form senile
49 plaques [5]. Soluble Aβ oligomers formed during
50 the first stages of AD are believed to be particu-
51 larly toxic and responsible for early memory failure
52 [6]. Recently, intermediate and large oligomeric Aβ
53 assembly states have been associated with both aging
54 and AD [7]. AβPP-C-terminal fragments (CTFs)
55 generated by β-secretase can also form oligomers
56 [8, 9].
57 Cathepsins reside in the endo-lysosome system
58 [10] where β-secretase activity mainly occurs [11,
59 12]. Cathepsin B (CTSB) is a cysteine protease that
60 cleaves AβPP at the same site as β–secretase [13]
61 and also degrades the Aβ42 peptide, starting at its
62 C-terminal [14]. Enzyme and proenzyme forms of
63 CTSB have been identified in AD brains, and are
64 particularly common in early endosomes [15].
65 Many studies have reported OS markers to exist
66 in the brain and peripheral tissues of patients with
67 AD, and indicate them to be among the earli-
68 est indications of the disease [16, 17]. Aging is
69 an important risk factor for AD [18], and protein
70 degradation systems, especially those involving the
71 cathepsins, become impaired in aging brains [19, 20].
72 To study the involvement of OS in AD, our labora-
73 tory developed a human neuron model of mild OS
74 induced by the xanthine/xanthine oxidase (X-XOD)
75 free radical-generating system. This model allowed
76 us to investigate the events preceding cell death
77 [21], and the relationships between OS and AβPP
78 metabolism/processing [22]. Moreover, it was used
79 to show that OS influences AβPP processing, traf-
80 ficking, and catabolism via the ubiquitin-proteasome
81 system and autophagy-lysosomal pathway [23].
82 Defects in protein degradation might link aging
83 to neurodegeneration [24, 25]. In addition, CTSB,
84 with its potential function in AβPP proteolysis (β-
85 secretase activity) and in the clearance of Aβ (Aβ
86 degradative activity), plays likely an important role
87 in the formation of amyloid oligomers.
88 The hypothesis of the present work was that mild
89 OS regulates the formation of amyloid oligomers via
90 CTSB, and that this might be a major step in the
91 neurodegenerative process.

92 MATERIALS AND METHODS

93 Materials

94 SK-N-MC human neuroblastoma cells (HTB-10)
95 were obtained from the American Type Culture
96 Collection. Xanthine (X) was purchased from Sigma,

xanthine oxidase (XOD) from Roche, Ca-074Me (a 97
CTSB inhibitor) from Enzo and DAPT (a γ-secretase 98
inhibitor) from Sigma. Culture medium compo- 99
nents were purchased from Gibco Laboratories. 100
Other chemicals were purchased from Merck or 101
Sigma. 102
Cell culture and treatments 103
Cells were cultured in minimal essential medium 104
(MEM) supplemented with 10% fetal bovine serum 105
and 50 µg/ml gentamicin in a humidified 5% 106
CO2 atmosphere. Exponentially growing cells at 107
80–90% confluence were placed in culture dishes 108
with 6, 24, or 96 multiwells (M-6, M-24, or M- 109
96) one day before treatment with X-XOD and/or 110
Ca-074Me. 111
On the day of treatment, cells were placed in fresh 112
medium 1 h before addition of the compounds X 113
(10 µM) and XOD (50 mU/ml). The effects of their 114
addition were measured 24 or 48 h later. For Ca- 115
074Me treatments (concentrations are indicated in 116
each experiment), the inhibitor was added for 1 h and 117
the cell cultures then placed in fresh medium for the 118
remainder of the experiment (24 h). The concentra- 119
tion of DMSO (vehicle) in the cell culture was 0.01% 120
or lower. 121
Analysis of cell injury 122
The cell injury caused by exposure to OS was 123
determined using the 3-(4,5-dimethyl-thiazol-2-yl)- 124
2,5-diphenyl tetrazolium bromide (MTT) assay [26] 125
with minor modifications [21]. 126
CTSB activity assay 127
CTSB activity was assessed following the method 128
of Porter et al. [27] with modifications. Cells grown 129
in a 6-well plate and incubated under different treat- 130
ments were washed in PBS and lysed for 1 h at 131
4◦C with shaking in 200 µL of 50 mM sodium 132
acetate (pH 5.5), 0.1 M NaCl, 1 mM EDTA, and 133
0.2% Triton X-100 (lysis buffer). Lysates were 134
centrifuged at 13,000 g for 10 min and the super- 135
natant (clarified lysate) used to determine proteolytic 136
activity. For the activity assays, 50 µg of clarified 137
lysate were incubated at 37◦C for 30 min in lysis 138
buffer (100 µL) in the presence of 20 µM z-RR- 139
AMC (P-137; Enzo Life Sciences) (an appropriate 140
fluorogenic substrate). The fluorescence emitted 141
as a result of proteolysis was recorded using an 142

143 Infinite® 200 microplate reader (Tecan Trading AG)
144 (excitation/emission wavelengths 360/430 nm). Immunoprecipitation
191
Immunoprecipitation of cell lysates was performed
192
145 Antibodies in PBS containing 1% Triton X-100, employing the

Biotech.), rabbit anti-Rab7 (9367; Cell Signaling), mouse anti-LAMP1 (H4A3-C; DSHB Hybridoma Bank), mouse anti-CD222 (315902; BioLegend), and mouse anti-EEA1 (610457; BD Transduction Lab- oratories). α-tubulin levels were examined as an internal control (via the reaction with anti-α-tubulin [Sigma] in the same blots). Secondary antibodies for immunostaining included horseradish peroxidase- coupled antibodies (Vector), and antibodies labelled with Alexa Fluor 488 or Alexa Fluor 555 dye (Invit- rogen). The ability of each antibody to recognize different metabolites according to their epitope loca- tion in AβPP has been described [22].

Western blot analysis

Immunofluorescence imaging 199
Immunofluorescence assays were performed as 200
previously described [22]. Cells were examined using 201
an LSM 710 laser scanning confocal microscope 202
(Zeiss) coupled to a vertical M2 AxioImager (Zeiss) 203
equipped with a 63X/1.4 Plan-Apochromat oil objec- 204
tive lens, or using a Zeiss Axiovert 200 fluorescence 205
microscope equipped with a 63 X oil-immersion 206
objective. Pictures were taken with a Spot RT digital 207
camera (Diagnostic) using Zeiss ZEN 2010 imaging 208
system software. Images were processed using Adobe 209
Photoshop CS3. 210
Cell fractionation for endo-lysosome and cytosol 211

234 Statistical analysis

235 Results were expressed as means standard
236 errors, and analyzed using the Student t test. Signifi-
237 cance was set at p < 0.05.

Cathepsin B (CTSB) activity was assessed in SK-N-MC cells treated with the X-XOD free rad- ical generating system for 24 h [22]. To compare the effect of mild OS with the pharmacological inhibition of CTSB, cells were treated with CA- 074Me (a specific inhibitor of this protease). CTSB activity was quantified using a fluorometric assay. X-XOD-treated cells showed CTSB activity to be reduced by up to 58% (Fig. 1A) (p < 0.001 com- pared to control culture). CA-074Me reduced CTSB activity in a dose-dependent manner (Supplementary Fig. 1A)—by 97% (p < 0.001) at its highest concen- tration (5 µM), 71% (p = 0.0012) at an intermediate concentration (1 µM), and 48% (p = 0.0115) at the lowest concentration (0.2 µM). To confirm that the reduced activity was not due to any treatment-caused

 

Fig. 1. X-XOD reduces CTSB activity. SK-N-MC cells were treated with 10 µM X/50 mU/ml XOD (X-XOD). After incubation for 24 h, CTSB activity (A) and CTSB protein levels (B) were mea- sured. A) Proteolysis in the presence of the fluorogenic substrate z-RR-AMC was used to monitor CTSB activity. The graphs show the mean ( SEM) fluorescence values for each level of enzyme activity expressed (as a percentage of the control or vehicle value).
∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.01 (t-test, n = 3). B) Western
blotting was performed using the anti-CTSB antibody. The bands
correspond to the pro-enzyme (top) and enzyme (down) forms of CTSB. In the lower panel the band corresponds to α-tubulin. The data show the mean ( SEM) densitometry values (normal- ized against α-tubulin). Control or vehicle values were set at 1.
∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.01 (t-test, n = 3).

Fig. 2. X-XOD increases amyloid oligomers levels. SK-N-MC cells were treated (A) with 10 µM X/50 mU/ml XOD (X-XOD) and (B) with the indicated concentrations of CA-074Me. After incubation for 24 h, western blotting was performed using anti- N-terminal AβPP antibody (22C11) and anti-oligomer antibody (A11). Representative gels for X-XOD (A) and CA-074Me (B) treatments are shown. In the upper panel the band corresponds to high molecular weight amyloid oligomer, in the intermediate panel to hAβPP, and in the lower to α-tubulin. The data show the mean densitometry values (normalized against α-tubulin). Control or vehicle values were set at 1. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.01
(t-test, n = 4).
lessening of cell viability, MTT reduction assays were 256
performed. Supplementary Fig. 2 (A and B) shows 257
that none of the treatments reduced cell viability at 258
24 h of culture. Previous studies reported minimal cell 259
injury at this time [21]. 260
To deepen into the effect of these treatments on 261
CTSB, the active enzyme levels were examined by 262
western blotting in the same cell lysates, using a 263
CTSB-specific antibody that recognizes the enzyme 264
and its precursor (pro-enzyme). CTSB is synthe- 265
sized as a pre-proenzyme, following removal of the 266
signal peptide the inactive pro-enzyme undergoes fur- 267
ther modifications (37–42 kDa) including removal 268
of the pro region to result in the active enzyme 269
(25 kDa) [29]. X-XOD (Fig. 1B) induced a signif- 270
icant reduction in CTSB mature/active-form levels 271
(0.61 0.08-fold versus control; p = 0.098). The 272
CTSB inhibitor CA-074Me (Supplementary Fig. 1B) 273
also induced significant reductions [0.45 0.04-, 274
0.61 0.16- and 0.59 0.01-fold versus control for 275
5 (p < 0.001), 1 (p = 0.049) and 0.2 (p = 0.006) µM 276
CA-074Me, respectively]. 277
X-XOD increases amyloid oligomer levels 278
The effect of mild OS on soluble amyloid oligomer 279
levels was examined in cell lysates by western blot- 280
ting (Fig. 2) using the polyclonal antibody A11. 281
This antibody specifically detects soluble amyloid 282
assemblies distinct from fibrillar Aβ [30–32]. The 283

 

Fig. 3. X-XOD produces the intracellular accumulation of amyloid oligomers. SK-N-MC cells subjected to 10 µM X/50 mU/ml XOD (X-XOD) for 24 h were examined by double-label confocal microscopy. The representative panel shows immunofluorescence images for the anti-Aβ 6E10 (green) and anti-oligomer A11 (red) antibodies. The merged images show their overlay. For each cell culture, four microscopic fields from three independent cultures were analyzed. Original magnification: 63 . Scale bar: 10 µm. No staining was observed when the primary antibodies were omitted

effect of the CTSB inhibitor CA-074Me was also examined.
Figure 2A shows that the cells subjected to X- XOD treatment experienced a significant increase in a single band at 56 kDa (1.36 0.06-fold; p < 0.001), which also increased in the cells treated with CA- 074Me (1.32 0.07; p = 0.005 and 1.53 0.40-fold;
ns) at 5 µM and 1 µM, respectively (Fig. 2B).
Immunoprecipitation of this high molecular weight complex by the 6E10 antibody (an anti- Aβ, -βCTF and -holo AβPP [hAβPP] monoclonal antibody) but not by the C-terminal AβPP specific antibody (Supplementary Fig. 5) indicated it to be an Aβ oligomeric form derived from AβPP proteolysis. The possibility that this 56 kDa band, is derived from Aβ was reinforced by the observation that the inhi- bition of γ-secretase using DAPT led to a significant decrease in the levels of oligomers in the X-XOD treated cells (Supplementary Fig. 4).
Based on its electrophoretic mobility, this large amyloid oligomer most probably corresponds either to the 56 kDa soluble Aβ assembly known as Aβ*56 (detected in an AD transgenic rat brain model by Lesne et al. [33]) or to the more recently described oligomeric Aβ assemblies found in both ageing and AD brains [7].

Since the amyloid oligomer derives from AβPP 310
proteolytic processing, AβPP protein was analyzed 311
in the same cell lysates by western blotting using 312
the 22C11 antibody which recognizes full-length 313
AβPP (hAβPP) and AβPP fragments containing 314
the N-terminal part (sAβPP). Figure 2A shows 315
that treatment with X-XOD for 24 h led to a 316
3.85 1.07-fold increase in cellular AβPP levels 317
(p = 0.037). The above results indicate that, dur- 318
ing that period just before the transmission of cell 319
death signals, the X-XOD system increases intra- 320
cellular soluble amyloid oligomer and AβPP levels. 321
In contrast, the levels of AβPP protein (Fig. 2B) 322
were not significantly affected by CA-074Me 323
treatment. 324
Together, these results show that a decrease in 325
CTSB activity—via mild OS or the pharmacolog- 326
ical inhibition of CTSB—leads to increased levels 327
of amyloid oligomers, suggesting that the CTSB 328
pathway is a participant in the changes of amyloid 329
oligomer levels. 330
The intracellular location of amyloid oligomers 331
was then investigated by double immunofluores- 332
cence analysis using the A11 and 6E10 antibodies. 333
6E10-immunoreactive protein species are proba- 334
bly composed of hAβPP, βCTF, and Aβ/βCTF 335

 

Fig. 4. A11 positive oligomers localized in late endosomes. SK-N-MC cells subjected to 10 µM X/50 mU/ml XOD (X-XOD) for 24 h were examined by double-label confocal microscopy. The representative panel shows immunofluorescence images for the CD222 late endosome marker (green) and anti-oligomer A11 (red) antibodies. The merged images show their overlay. For each cell culture, four microscopic fields from two independent cultures were analyzed. Original magnification: 63×. Scale bar: 10 µm.
336 assemblies (oligomers) plus a small amount of Aβ X-XOD increases amyloid oligomer levels in
362
337 peptide. A11-positive staining reveals oligomeric partially purified cytosol and endo-lysosomal 363
338 forms but not monomers or fibrils [30]. vesicles 364
339 In control cells, 6E10-positive structures (Fig. 3,
340 green) appeared as small vesicles compatible with Amyloidogenic processing of AβPP takes place 365
341 endosomes/lysosomes as previously described [23]. mainly in the endo-lysosomes [12]. Based on the 366
342 In contrast, A11-stained structures appeared as above immunofluorescence results, it was decided 367
343 small points distributed throughout the cytoplasm to further explore the subcellular location of the 368
344 (cytosol/microsomes); colocalization with 6E10- oligomers that accumulate under OS conditions. For 369
345 positive structures was minimal. In the X-XOD this purpose, cells were fractionated following the 370
346 treated cells, the A11-stained structures distributed method of Avrahami [28] to separate out a fraction 371
347 throughout the cytoplasm (cytosol/microsomes) were enriched in lysosomes and late-endosomes (L/LE). 372
348 also present, but the number and size of structures Figure 5 shows that the CTSB enzyme was 373
349 positive for both antibodies increased. According to enriched in the L/LE fraction, appearing along- 374
350 previous data of the laboratory [23], the large yellow side the endosome and lysosome markers (Rab7 375
351 signals probably correspond to vesicles of the endo- and LAMP1). Large (56 kDa) and intermediate (a 376
352 lysosomal system, and contain amyloid oligomers less reproducible band of approximately 40 kDa) 377
353 and AβPP/βCTF/Aβ. molecular weight amyloid oligomers revealed by the 378
354 Double immunofluorescence assays of the A11 A11 antibody were detected in the partially purified 379
355 antibody with markers of the endolysosomal system cytosol (Cyt) fraction, along with the specific marker 380
356 were performed and revealed the presence of A11 for the early-endosome EEA1. However, the large 381
357 positive oligomers into CD222 positive organelles oligomer form (56 kDa) and AβPP were preferen- 382
358 (late endosomes) (Fig. 4). Furthermore, a significant tially located in the L/LE fraction. 383
359 increase in the size of the late endosomes, some of In the X-XOD treated cells, both the 56 kDa 384
360 which also contained A11 positive structures, was oligomer and the AβPP were clearly increased in both 385
361 observed in the X-XOD treated cells. fractions. The presence of A11-positive oligomers 386

Fig. 6. X-XOD and CA-074Me reduce CTSB activity. SK-N-MC cells were treated with 10 µM X/50 mU/ml XOD (X-XOD) in the presence or absence of 5 µM CA-074Me. After incubation for 24 h, CTSB activity was analyzed. Proteolysis in the presence of the fluorogenic substrate z-RR-AMC was used to monitor CTSB activity. The graph shows the mean ( SEM) fluorescence values for each activity level expressed as a percentage of the control value. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.01 (t-test, n = 4).

Fig. 5. X-XOD increases amyloid oligomer levels in partially puri- fied cytosol and endo-lysosomal vesicles. SK-N-MC cells were treated with 10 µM X/50 mU/ml XOD (X-XOD) for 24 h and then fractionated by differential centrifugation. Western blotting was performed using anti-oligomer (A11), anti-AβPP (22C11), anti-CTSB, anti-LAMP1, anti-Rab7, and anti-EEA1 antibodies. The protein quantity determined by a BCA assay was used as the loading control. A representative gel (of three independent experiments) is shown. Cyt, cytosol fraction; L/LE, lysosome-late endosome fraction.

in the Cyt and L/LE is consistent with the staining pattern obtained in the immunofluorescence analysis (Figs. 3 and 4).

Inhibition of CTSB modulates the AβPP metabolism/processing induced by X-XOD

Although the two treatments studied in this work (X-XOD system and the CTSB inhibitor CA-074Me) induced an increase of amyloid oligomers, their effect on AβPP levels was clearly different, since X-XOD increased hAβPP levels but CTSB inhibitor did not (Fig. 2). This suggested that X-XOD induced OS and the CTSB inhibitor modulate AβPP metabolism by different mechanisms. To deepen into the complexity

of AβPP metabolism modulation, we analyzed the 400
levels of amyloid oligomer and hAβPP in cells treated 401
with X-XOD in presence of CA-074Me. Firstly, the 402
quantification of CTSB enzymatic activity with both 403
treatments together (Fig. 6) revealed a complete inhi- 404
bition (up to 98%; p < 0.001 versus control culture), 405
whereas the results for X-XOD or CA-074Me alone 406
were similar to those described above (Fig. 1 and 407
Supplementary Fig.1). The MTT reduction assay was 408
performed to confirm that this reduced activity was 409
not due to a treatment-induced lessening of cell via- 410
bility (Supplementary Fig. 3). 411
The effect of mild OS in the presence of the 412
CTSB inhibitor on the levels of amyloid oligomers 413
and AβPP was then examined by western blotting, 414
using the A11 and 22C11 antibodies respectively. 415
Figure 7A shows that the increase in the large amy- 416
loid oligomer induced by X-XOD was attenuated, 417
but not significantly (p = 0.053) in the presence of 418
CA-074Me (0.65 0.15-fold over X-XOD alone). 419
The increase in AβPP protein induced by X-XOD 420
was, however, significantly enhanced in the presence 421
of CA-074Me (3.03 0.57-fold over X-XOD alone; 422
p = 0.0083). 423
The effect of both treatments together on sol- 424
uble AβPP (sAβPP) secretion was also examined 425
by western blotting using the 6E10 antibody, 426
which recognizes secreted AβPP derived from non- 427
amyloidogenic α-cleavage (sAβPPα). As shown in 428
Fig. 7B, an increase in sAβPPα at the culture medium 429
of cells treated with X-XOD compared to control cells 430

 

Fig. 7. Inhibition of CTSB modulates the AβPP metabolism/processing regulated by X-XOD. SK-N-MC cells were treated with 10 µM X/50 mU/ml XOD (X-XOD) in the presence or absence of 5 µM CA-074Me. After incubation for 24 h, cell cultures were examined A) intracellularly and B) extracellularly. A) Intracellular: Western blotting analysis was performed using anti-N-terminal AβPP antibody (22C11) and the anti-oligomer antibody A11. A representative gel is shown. In the upper panel, the band corresponds to high-molecular weight amyloid oligomer, in the intermediate to hAβPP, and in the lower panel to α-tubulin. The data (1) show the mean densitometry values (normalized against α-tubulin). The oligomer data for the representative gel are also shown (2). Values for X-XOD were set at 1. ∗∗∗p < 0.001
(t-test, n = 5). B) Extracellular: secreted AβPP in the culture medium, derived from non-amyloidogenic α-cleavage (sAβPPα), was analyzed
by western blotting with the 6E10 antibody. The data show the mean ( SEM) densitometry values. Values for control or X-XOD were set at 1. ∗p < 0.05, ∗∗∗p < 0.01 (t-test, n = 4).

(1.25 0.03-fold; p = 0.011) was observed, being in concordance with previously published results [23]. This increase was further enhanced in the presence of CA-074Me (1.69 0.04-fold over control, p < 0.001;
1.37 0.05-fold over X-XOD alone; p = 0.027).
These results indicate that the AβPP metabolism/ processing induced by mild OS is modulated by inhibiting CTSB, leading to an increase in intracellu- lar hAβPP and secreted sAβPPα.

DISCUSSION

Recently, large oligomeric assembly states of Aβ 454
have been associated with both ageing and AD [7]. 455
Numerous risk factors have been associated with 456
the onset of AD, being one of them the ageing- 457
associated OS [16, 17]. In addition to altering AβPP 458
metabolism, OS affects protein degradation path- 459
ways [39]. Delving into this topic, we had previously 460
reported the involvement of the ubiquitin-proteasome 461
and the lysosome protein degradation systems in the 462
regulation of AβPP metabolism/processing by the 463
X-XOD free radical-generating system [23]. 464

479 Cathepsins activity is regulated in different Regarding the intracellular location of the
531
480 ways—transcription, post-transcription/processing, oligomer, a punctate pattern of A11-positive struc-
532
481 translation, glycosylation, and trafficking, and also tures distributed throughout the cytoplasm was
533
482 via its binding to endogenous inhibitors [43–45]. detected both in control and X-XOD treated cells
534
483 In the present study, mild OS induced by the whereas this mild OS led to the appearance of
535
484 free radical generating system X-XOD produced a enlarged vesicles, compatible with endo-lysosomes,
536
485 reduction of CTSB activity and of CTSB protein positive for both A11 and 6E10; they therefore
537
486 levels, similarly to the pharmacological inhibition would contain amyloid oligomers, and probably
538
487 by CA-074Me. AβPP/βCTF/Aβ too. Actually, vesicles stained by

CA-074Me inhibition mechanisms are not com- pletely characterized, but there are indeed some reports [46] describing that CA-074Me binds to CTSB in a substrate-like mode, occupying the S’ sub- sites. The reduction of CTSB levels in the cells treated with CA-074Me may result from the inhibition of its autoproteolytic activity [43]. The regulation of CTSB activity by mild OS is likely to be more complex than its pharmacological inhibition, since X-XOD disrupts the autophagic flux in this cell model [23, 47], prob- ably affecting the trafficking of CTSB towards the endo-lysosomal vesicles.
Western blotting with the A11 antibody (which rec- ognizes oligomeric structures) showed both X-XOD and CA-074Me to affect the levels of soluble amy- loid oligomers, revealing a significantly increased large molecular weight soluble amyloid oligomer ( 56 kDa). This oligomer was formed by Aβ but not by βCTFs [8, 9, 33] since it was recognized by the 6E10 monoclonal antibody (which attaches to Aβ) but not by the AβPP C-terminal specific anti- body. Nevertheless, the immunoreactivity of A11 in

a CD222 specific antibody, corresponding to late 540
endosomes, were found to contain A11 positive 541
structures and to be significantly enlarged in X- 542
XOD treated cells. In summary, these experiments 543
revealed that A11 positive oligomers were dis- 544
tributed throughout the cell, both outside and inside 545
endolysosomal vesicles, being this last location 546
compatible with that described for the enzymes 547
and products of the amyloidogenic proteolysis of 548
AβPP [12, 52]. 549
The data obtained after subcellular fractionation of 550
the SK-N-MC cells revealed the presence of the 56 551
kDa oligomer and AβPP in the cell fraction enriched 552
in endo-lysosomal vesicles. Moreover, the present 553
results also showed CTSB to be mainly localized 554
in the endo-lysosome enriched fraction, which is in 555
line with data reported by other authors [43] and 556
reinforces the probability of its involvement in the 557
formation and/or degradation of amyloid oligomers. 558
Interestingly, previous studies in AD and Down’s 559
syndrome have revealed activated or enlarged endo- 560
somes containing soluble Aβ, prior to the deposition 561

different steps of AβPP metabolism induced by mild OS.

In the light of these results, we propose that the modulation of AβPP metabolism by mild OS is partly mediated by CTSB, but that the effects of OS most probably involve additional mechanisms, like its involvement on the ubiquitin proteasome pathway [23], or its general effect on the lysosomal function [23, 47] that we reported previously.
Although the mechanisms underlying the relations among Aβ/CTF aggregation, free radical damage and cell death remain unclear, recent reports support the involvement of CTSB in the induction of inflam- matory response by Aβ oligomers [54] or in the degradation of the key AD proteins [41].
In summary, the present results support the par- ticipation of CTSB in the regulation of AβPP metabolism by mild OS, although further research on its specific role and on the contribution of additional lysosomal activities in that regulation is guaranteed. This work also suggests that CTSB could serve as a therapeutic target for the treatment of AD, in line with previous studies in cellular and animal models of the disease [13, 55]. Besides, knowing the rela- tionship between CA-074 Me lysosomal function, OS, and AβPP metabolism/processing could help us to better under- stand the pathophysiology of the disease.

ACKNOWLEDGMENTS

This work was supported by the Ministerio de Ciencia e Innovacio´n (SAF2014-53954-R). Llorente P is supported via the Fundacio´n Severo Ochoa. The institutional grants of Fundacio´n Ramo´n Areces and Banco de Santander to the Centro de Biolog´ıa Molec- ular Severo Ochoa is gratefully acknowledged.
Authors’ disclosures available online (https:// www.j-alz.com/manuscript-disclosures/17-0159r4).

SUPPLEMENTARY MATERIAL

The supplementary material is available in the electronic version of this article: http://dx.doi.org/ 10.3233/jad-170159.

REFERENCES

[1] Blennow K, de Leon MJ, Zetterberg H (2006) Alzheimer’s disease. Lancet 368, 387-403.
[2] Glenner GG, Wong CW, Quaranta V, Eanes ED (1984) The amyloid deposits in Alzheimer’s disease: their nature and pathogenesis. Appl Pathol 2, 357-369.

[3] Haass C, Kaether C, Thinakaran G, Sisodia S (2012) Traf- 629
ficking and proteolytic processing of APP. Cold Spring Harb 630
Perspect Med 2, a006270. 631
[4] Jan A, Gokce O, Luthi-Carter R, Lashuel HA (2008) 632
The ratio of monomeric to aggregated forms of Abeta40 633
and Abeta42 is an important determinant of amyloid-beta 634
aggregation, fibrillogenesis, and toxicity. J Biol Chem 283, 635
28176-28189. 636
[5] Shah P, Lal N, Leung E, Traul DE, Gonzalo-Ruiz A, Geula 637
C (2010) Neuronal and axonal loss are selectively linked 638
to fibrillar amyloid-beta within plaques of the aged primate 639
cerebral cortex. Am J Pathol 177, 325-333. 640
[6] McLean CA, Cherny RA, Fraser FW, Fuller SJ, Smith MJ, 641
Beyreuther K, Bush AI, Masters CL (1999) Soluble pool of 642
Abeta amyloid as a determinant of severity of neurodegen- 643
eration in Alzheimer’s disease. Ann Neurol 46, 860-866. 644
[7] Baker-Nigh A, Vahedi S, Davis EG, Weintraub S, Bigio EH, 645
Klein WL, Geula C (2015) Neuronal amyloid-beta accu- 646
mulation within cholinergic basal forebrain in ageing and 647
Alzheimer’s disease. Brain 138, 1722-1737. 648
[8] Woltjer RL, Nghiem W, Maezawa I, Milatovic D, Vaisar 649
T, Montine KS, Montine TJ (2005) Role of glutathione 650
in intracellular amyloid-alpha precursor protein/carboxy- 651
terminal fragment aggregation and associated cytotoxicity. 652
J Neurochem 93, 1047-1056. 653
[9] Cheng F, Cappai R, Ciccotosto GD, Svensson G, Multhaup 654
G, Fransson LA, Mani K (2011) Suppression of amy- 655
loid beta A11 antibody immunoreactivity by vitamin 656
C: possible role of heparan sulfate oligosaccharides 657
derived from glypican-1 by ascorbate-induced, nitric 658
oxide (NO)-catalyzed degradation. J Biol Chem 286, 659
27559-27572. 660
[10] Nixon RA, Cataldo AM (1995) The endosomal-lysosomal 661
system of neurons: new roles. Trends Neurosci 18, 489-496. 662
[11] Vassar R, Bennett BD, Babu-Khan S, Kahn S, Mendiaz EA, 663
Denis P, Teplow DB, Ross S, Amarante P, Loeloff R, Luo 664
Y, Fisher S, Fuller J, Edenson S, Lile J, Jarosinski MA, 665
Biere AL, Curran E, Burgess T, Louis JC, Collins F, Tre- 666
anor J, Rogers G, Citron M (1999) Beta-secretase cleavage 667
of Alzheimer’s amyloid precursor protein by the transmem- 668
brane aspartic protease BACE. Science 286, 735-741. 669
[12] Soldano A, Hassan BA (2014) Beyond pathology: APP, 670
brain development and Alzheimer’s disease. Curr Opin 671
Neurobiol 27, 61-67. 672
[13] Hook V, Toneff T, Bogyo M, Greenbaum D, Medzihradszky 673
KF, Neveu J, Lane W, Hook G, Reisine T (2005) Inhibition 674
of cathepsin B reduces beta-amyloid production in regulated 675
secretory vesicles of neuronal chromaffin cells: evidence for 676
cathepsin B as a candidate beta-secretase of Alzheimer’s 677
disease. Biol Chem 386, 931-940. 678
[14] Mueller-Steiner S, Zhou Y, Arai H, Roberson ED, Sun 679
B, Chen J, Wang X, Yu G, Esposito L, Mucke L, Gan L 680
(2006) Antiamyloidogenic and neuroprotective functions of 681
cathepsin B: implications for Alzheimer’s disease. Neuron 682
51, 703-714. 683
[15] Cataldo AM, Barnett JL, Pieroni C, Nixon RA (1997) 684
Increased neuronal endocytosis and protease delivery to 685
early endosomes in sporadic Alzheimer’s disease: neu- 686
ropathologic evidence for a mechanism of increased 687
beta-amyloidogenesis. J Neurosci 17, 6142-6151. 688
[16] Nunomura A, Perry G, Aliev G, Hirai K, Takeda A, Balraj 689
EK, Jones PK, Ghanbari H, Wataya T, Shimohama S, Chiba 690
S, Atwood CS, Petersen RB, Smith MA (2001) Oxidative 691
damage is the earliest event in Alzheimer disease. J Neu- 692
ropathol Exp Neurol 60, 759-767. 693

694 [17] Bonda DJ, Wang X, Perry G, Nunomura A, Tabaton M, [34] Andrew RJ, Kellett KA, Thinakaran G, Hooper NM (2016) 759
695 Zhu X, Smith MA (2010) Oxidative stress in Alzheimer A Greek tragedy: the growing complexity of Alzheimer 760
696 disease: a possibility for prevention. Neuropharmacology amyloid precursor protein proteolysis. J Biol Chem 291, 761
697 59, 290-294. 19235-19244. 762
698 [18] Nunomura A, Moreira PI, Castellani RJ, Lee HG, Zhu X, [35] Viola KL, Klein WL (2015) Amyloid beta oligomers in 763
699 Smith MA, Perry G (2012) Oxidative damage to RNA in Alzheimer’s disease pathogenesis, treatment, and diagnosis. 764
700 aging and neurodegenerative disorders. Neurotox Res 22, Acta Neuropathol 129, 183-206. 765
701 231-248. [36] Benilova I, Karran E, De Strooper B (2012) The toxic Abeta 766
702 [19] Martinez-Vicente M, Sovak G, Cuervo AM (2005) Protein oligomer and Alzheimer’s disease: an emperor in need of 767
703 degradation and aging. Exp Gerontol 40, 622-633. clothes. Nat Neurosci 15, 349-357. 768
704 [20] Yamashima T (2013) Reconsider Alzheimer’s disease by the [37] Brody DL, Jiang H, Wildburger N, Esparza TJ (2017) 769
705 ‘calpain-cathepsin hypothesis’–a perspective review. Prog Non-canonical soluble amyloid-beta aggregates and plaque 770
706 Neurobiol 105, 1-23. buffering: controversies and future directions for target dis- 771
707 [21] Recuero M, Vicente MC, Martinez-Garcia A, Ramos MC, covery in Alzheimer’s disease. Alzheimers Res Ther 9, 62. 772
708 Carmona-Saez P, Sastre I, Aldudo J, Vilella E, Frank A, [38] Murakami K (2014) Conformation-specific antibodies to 773
709 Bullido MJ, Valdivieso F (2009) A free radical-generating target amyloid beta oligomers and their application to 774
710 system induces the cholesterol biosynthesis pathway: a role immunotherapy for Alzheimer’s disease. Biosci Biotechnol 775
711 in Alzheimer’s disease. Aging Cell 8, 128-139. Biochem 78, 1293-1305. 776
712 [22] Recuero M, Munoz T, Aldudo J, Subias M, Bullido [39] Kiffin R, Bandyopadhyay U, Cuervo AM (2006) Oxidative 777
713 MJ, Valdivieso F (2010) A free radical-generating sys- stress and autophagy. Antioxid Redox Signal 8, 152-162. 778
714 tem regulates APP metabolism/processing. FEBS Lett 584, [40] Nakanishi H (2003) Neuronal and microglial cathepsins in 779
715 4611-4618. aging and age-related diseases. Ageing Res Rev 2, 367-381. 780
716 [23] Recuero M, Munive VA, Sastre I, Aldudo J, Valdivieso F, [41] Cermak S, Kosicek M, Mladenovic-Djordjevic A, Smil- 781
717 Bullido MJ (2013) A free radical-generating system regu- janic K, Kanazir S, Hecimovic S (2016) Loss of cathepsin 782
718 lates AbetaPP metabolism/processing: involvement of the B and L leads to lysosomal dysfunction, NPC-like choles- 783
719 ubiquitin/proteasome and autophagy/lysosome pathways. terol sequestration and accumulation of the key Alzheimer’s 784
720 J Alzheimers Dis 34, 637-647. proteins. PLoS One 11, e0167428. 785
721 [24] Rubinsztein DC (2006) The roles of intracellular protein- [42] Cataldo AM, Nixon RA (1990) Enzymatically active lyso- 786
722 degradation pathways in neurodegeneration. Nature 443, somal proteases are associated with amyloid deposits in 787
723 780-786. Alzheimer brain. Proc Natl Acad SciUSA 87, 3861-3865. 788
724 [25] Douglas PM, Dillin A (2010) Protein homeostasis and aging [43] Ishidoh K, Kominami E (2002) Processing and activation 789
725 in neurodegeneration. J Cell Biol 190, 719-729. of lysosomal proteinases. Biol Chem 383, 1827-1831. 790
726 [26] Hansen MB, Nielsen SE, Berg K (1989) Re-examination [44] Mach L, Mort JS, Glossl J (1994) Maturation of human 791
727 and further development of a precise and rapid dye method procathepsin B. Proenzyme activation and proteolytic pro- 792
728 for measuring cell growth/cell kill. J Immunol Methods 119, cessing of the precursor to the mature proteinase, in vitro, 793
729 203-210. are primarily unimolecular processes. J Biol Chem 269, 794
730 [27] Porter K, Nallathambi J, Lin Y, Liton PB (2013) Lysosomal 13030-13035. 795
731 basification and decreased autophagic flux in oxidatively [45] Roberts R (2005) Lysosomal cysteine proteases: structure, 796
732 stressed trabecular meshwork cells: implications for glau- function and inhibition of cathepsins. Drug News Perspect 797
733 coma pathogenesis. Autophagy 9, 581-594. 18, 605-614. 798
734 [28] Avrahami L, Farfara D, Shaham-Kol M, Vassar R, Frenkel [46] Buttle DJ, Murata M, Knight CG, Barrett AJ (1992) CA074 799
735 D, Eldar-Finkelman H (2013) Inhibition of glycogen syn- methyl ester: a proinhibitor for intracellular cathepsin B. 800
736 thase kinase-3 ameliorates beta-amyloid pathology and Arch Biochem Biophys 299, 377-380. 801
737 restores lysosomal acidification and mammalian target of [47] Santana S, Sastre I, Recuero M, Bullido MJ, Aldudo J 802
738 rapamycin activity in the Alzheimer disease mouse model: (2013) Oxidative stress enhances neurodegeneration mark- 803
739 in vivo and in vitro studies. J Biol Chem 288, 1295-1306. ers induced by herpes simplex virus type 1 infection in 804
740 [29] Illy C, Quraishi O, Wang J, Purisima E, Vernet T, Mort JS human neuroblastoma cells. PLoS One 8, e75842. 805
741 (1997) Role of the occluding loop in cathepsin B activity. [48] Tambo K, Yamaguchi T, Kobayashi K, Terauchi E, Ichi I, 806
742 J Biol Chem 272, 1197-1202. Kojo S (2013) Racemization of the aspartic acid residue of 807
743 [30] Glabe CG (2008) Structural classification of toxic amyloid amyloid-beta peptide by a radical reaction. Biosci Biotech- 808
744 oligomers. J Biol Chem 283, 29639-29643. nol Biochem 77, 416-418. 809
745 [31] Kayed R, Head E, Thompson JL, McIntire TM, Milton [49] Inoue K, Hosaka D, Mochizuki N, Akatsu H, Tsutsumiuchi 810
746 SC, Cotman CW, Glabe CG (2003) Common structure of K, Hashizume Y, Matsukawa N, Yamamoto T, Toyo’oka 811
747 soluble amyloid oligomers implies common mechanism of T (2014) Simultaneous determination of post-translational 812
748 pathogenesis. Science 300, 486-489. racemization and isomerization of N-terminal amyloid beta 813
749 [32] Kayed R, Head E, Sarsoza F, Saing T, Cotman CW, Necula in Alzheimer’s brain tissues by covalent chiral deriva- 814
750 M, Margol L, Wu J, Breydo L, Thompson JL, Rasool S, tized ultraperformance liquid chromatography tandem mass 815
751 Gurlo T, Butler P, Glabe CG (2007) Fibril specific, confor- spectrometry. Anal Chem 86,797-804. 816
752 mation dependent antibodies recognize a generic epitope [50] Cheignon C, Tomas M, Bonnefont-Rousselot D, Faller P, 817
753 common to amyloid fibrils and fibrillar oligomers that is Hureau C, Collin F (2018) Oxidative stress and the amyloid 818
754 absent in prefibrillar oligomers. Mol Neurodegener 2, 18. beta peptide in Alzheimer’s disease. Redox Biol 14, 450- 819
755 [33] Lesne S, Koh MT, Kotilinek L, Kayed R, Glabe CG, Yang 464. 820
756 A, Gallagher M, Ashe KH (2006) A specific amyloid-beta [51] Cline EN, Bicca MA, Viola KL, Klein WL (2018) The 821
757 protein assembly in the brain impairs memory. Nature 440, amyloid-β oligomer hypothesis: beginning of the third 822
758 352-357. decade. J Alzheimers Dis 64, S567-S610. 823

824 [52] Whyte LS, Lau AA, Hemsley KM, Hopwood JJ, Sargeant [54] Taneo J, Adachi T, Yoshida A, Takayasu K, Takahara K, 833
825 TJ (2017) Endo-lysosomal and autophagic dysfunction: a Inaba K (2015) Amyloid beta oligomers induce interleukin- 834
826 driving factor in Alzheimer’s disease? J Neurochem 140, 1beta production in primary microglia in a cathepsin B- 835
827 703-717. and reactive oxygen species-dependent manner. Biochem 836
828 [53] Cataldo AM, Petanceska S, Terio NB, Peterhoff CM, Biophys Res Commun 458, 561-567. 837
829 Durham R, Mercken M, Mehta PD, Buxbaum J, Haroutu- [55] Cho K, Yoon SY, Choi JE, Kang HJ, Jang HY, Kim DH 838
830 nian V, Nixon RA (2004) Abeta localization in abnormal (2013) CA-074Me, a cathepsin B inhibitor, decreases APP 839
831 endosomes: association with earliest Abeta elevations in AD accumulation and protects primary rat cortical neurons 840
832 and Down syndrome. Neurobiol Aging 25, 1263-1272. treated with okadaic acid. Neurosci Lett 548, 222-227. 841