Amyloid properties of the leader peptide of variant B
cystatin C: implications for Alzheimer and macular
degeneration
Ricardo Sant’Anna1
, Susanna Navarro1
, Salvador Ventura1
, Luminita Paraoan2
and Debora Foguel3
1 Institut de Biotecnologia i Biomedicina and Departament de Bioquımica i Biologia Molecular, Universitat Autonoma de Barcelona,
Bellaterra, Spain
2 Department of Eye and Vision Science, Institute of Ageing and Chronic Disease, University of Liverpool, UK
3 Instituto de Bioquımica Medica Leopoldo de Meis, Programa de Biologia Estrutural, Universidade Federal do Rio de Janeiro, Brazil
Correspondence
D. Foguel, Instituto de Bioquımica Medica
Leopoldo de Meis, Programa de Biologia
Estrutural, Universidade Federal do Rio de
Janeiro, Rio de Janeiro, 21941-901 Brazil
Fax: +552122708647
Tel: +552139386761
E-mail: foguel@bioqmed.ufrj.br
S. Ventura, Institut de Biotecnologia i
Biomedicina, Universitat Auto`noma de
Barcelona, Bellaterra L69 3BX, Spain
Fax: +34 93 581 2011
Tel: +34 93 586 8956
E-mail: salvador.ventura@uab.es
and
L. Paraoan, Department of Eye and Vision
Science, Institute of Ageing and Chronic
Disease, University of Liverpool, Liverpool
08193, UK
Fax: +44 151 795 8420
Tel: +44 151 794 9038
E-mail: lparaoan@liverpool.ac.uk
(Received 17 November 2015, revised 21
December 2015, accepted 29 December
2015, available online 26 February 2016)
doi:10.1002/1873-3468.12093
Edited by Jesus Avila
Variant B (VB) of cystatin C has a mutation in its signal peptide (A25T),
which interferes with its processing leading to reduced secretion and partial
retention in the vicinity of the mitochondria. There are genetic evidences of
the association of VB with Alzheimer’s disease (AD) and age-related macular
degeneration (AMD). Here, we investigated aggregation and amyloid propensities
of unprocessed VB combining computational and in vitro studies.
Aggregation predictors revealed the presence of four aggregation-prone
regions, with a strong one at the level of the signal peptide, which indeed
formed toxic aggregates and mature amyloid fibrils in solution. In light of
these results, we propose for the first time the role of the signal peptide in
pathogenesis of AD and AMD.
Keywords: Alzheimers’ disease and exudative age-related macular
degeneration; human cystatin C; protein aggregation
Highlights
• Variant B precursor of the cysteine proteinase inhibitor cystatin C has
been previously associated with increased risk of developing Alzheimer’s
disease (AD) and exudative age-related macular degeneration (AMD).
• We show here that the leader sequence of the variant B precursor cystatin
C presents high aggregation and amyloidogenic propensity.
• Retention of the leader sequence in the incompletely processed variant
can contribute to deleterious protein aggregation linked to pathogenesis
of AD and AMD.
Human cystatin C belongs to the cystatin superfamily
of reversible inhibitors of cysteine proteases encompassing
the papain and legumain families [1]. The mature
cystatin C is a monomeric protein composed of a
polypeptide chain of 120 amino acids (13 343 Da) [2,3].
The general fold of monomeric inhibitors of the cystatin
Abbreviations
AD, Alzheimer’s disease; AMD, age-related macular degeneration; APR, aggregation-prone regions; ATR-FTIR, attenuated total reflectanceFourier
transform infrared spectroscopy; BM, Bruch’s membrane; CR, Congo red; DLS, dynamic light scattering; HCCAA, hereditary cystatin
C amyloid angiopathy; RPE, retinal pigment epithelium; TEM, transmission electron microscopy; WT, wild-type.
644 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
superfamily has been defined by the crystal structure of
chicken cystatin. It is composed of a long helix running
across a five-stranded antiparallel b-sheet [3,4]
(Fig. 1A). Cystatin C is produced and found in most tissues
and body fluids with particular higher concentrations
in the cerebrospinal fluid, owing to expression by
the choroid plexus. Cystatin C is involved in many biological
functions, ranging from bone reabsorption, modulation
of inflammatory response, antiviral and
antibacterial properties, cell proliferation and growth,
tumor metastasis, and to astrocyte differentiation during
mouse brain development [5–9].
Wild-type (WT) cystatin C has been found in amyloid
deposits in the brain arteries of elderly people [10,11],
while the rare mutation L68Q causes hereditary cystatin C
amyloid angiopathy (HCCAA) [12,13], a condition in
which the patients present fatal cerebral hemorrhage in
early adulthood. In vitro, both WT and L68Q cystatins C
produce amyloid fibrils [12,14,15]. Interestingly, in vitro
and in vivo studies have shown that WT cystatin C binds
to soluble Ab protecting this peptide from oligomerization.
In addition, cystatin C has been colocalized with Ab
deposits in senile plaques and the vascular cell wall. Altogether,
these results suggest that mature cystatin C might
BA
C D
1-MAGPLRAPLLLLAILAVALAVSPATG-26
1 22
27-SSPGKPPRLVGGPMDASVEEEG-48
23 45
49-VRRALDFAVGEYNKASMDMYHSR-71
46 68
72-ALQVVRARKQIVAGVNYFLDVEL-94
69 91
95-GRTTCTKTQPNLDNCPFHDQPHL-117
92 114
118-KRKAFCSFQIYAVPWQGTMTLSK-140
115 120
141-STCQDA-146
Fig. 1. Tridimensional structure, primary sequence and in silico aggregation and amyloid propensity predictions for WT and variant B cystatin
C. (A) A monomer-stabilized human cystatin C with an engineered disulfide bond (PDB ID: 3GAX) [61] reveals the canonical cystatin fold,
based on the crystal structure of chicken cystatin [62]. In red are the regions pointed out by AGGRESCAN as the aggregation-prone regions in
the protein. (B) The amino acid sequence of the variant B precursor cystatin C with the aggregation-prone regions (APR) identified by
AGGRESCAN in bold and by Waltz underlined. The leader peptide is highlighted in yellow. Gatekeeper residues flanking APR of the signal
peptide are in blue. We are using two numerations, the orange numbering includes the signal peptide and contains 146 residues, the green
numeration corresponds to the processed cystatin C, which contains 120 residues. In (C), the aggregation propensity of both sequences
was predicted using AGGRESCAN (http://bioinf.uab.es/aggrescan/) using default parameters. (D) The same procedure was used to predict the
amyloid propensity by the algorithm Waltz (http://www.switchlab.org/bioinformatics/waltz). We call attention to the fact that continuous lines
(mature cystatin C) are superimposed to the dashed lines (immature cystatin C).
645FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
R. Sant’Anna et al. Amyloid properties of the leader peptide
be a neuroprotective protein in Alzheimer’s disease (AD)
acting as a modulator of amyloidosis [16–18].
In the eye, cystatin C is abundantly expressed by the
retinal pigment epithelium (RPE) [19–21], the
monolayer of cells essential for maintaining the homeostasis
of the neuroretina and the blood–retina barrier.
RPE dysfunction is central to the pathogenesis of agerelated
macular degeneration (AMD), with various cellular
processes contributing to the onset and progression
of the disease including: cellular debris
accumulation in drusen [22], changes to the underlying
Bruch’s membrane (BM) with and without breakage
of the blood/retina barrier [23], impaired clearing of
protein aggregates and/or damaged organelles [24,25].
The cystatin C variant type B has been associated as
a recessive allele with increased risk of AD and exudative
AMD [26–30]. This variant results from a SNP in
the cystatin C gene (CST3) signal/leader sequence
(1MAGPLRAPLLLLAILAVALAVSPAAG26)
(p.Ala25Thr due to a c.G73A substitution). The 26amino
acid signal sequence of precursor cystatin C is
essential for targeting the protein to the ER/Golgi
apparatus and processing through the secretory pathway
[31,32]. The A25T substitution decreases the efficiency
of cleavage of the signal peptide in the variant
B precursor resulting in a protein constituted of 146
amino acids that is abnormally processed in the cell
[33]. The variant B precursor cystatin C is processed
less efficiently through the secretory pathway resulting
in its decreased secretion to the extracellular space,
with some of the unprocessed molecules diverted from
the secretory pathway accumulating intracellularly in
association with the mitochondrial membrane network
[33,34]. The reduced secretion and subsequent
decreased concentration of mature cystatin C extracellularly
could explain the enhanced AD and AMD risk
observed in homozygous patients with variant B.
Here, due to its association with AD and AMD, we
investigate the aggregation and amyloid propensity of
the variant B, with emphasis on the signal peptide.
Using two well validated aggregation predictors,
namely AGGRESCAN, which was designed to predict
aggregation propensity regions in vivo, and Waltz,
which predicts specific amyloid formation regions
[35,36], we found on the immature form of the protein,
four short aggregation-prone sequences, one of them
being the leader peptide with very high aggregation/
amyloidogenic scores. With these predictions in hands,
we carried out in vitro experiments with the leader
peptide to corroborate its amyloidogenicity. Our data
unequivocally demonstrated that the leader peptide of
the variant B cystatin C aggregates instantaneously
upon aqueous dilution forming positive Thioflavin-T
(Th-T) and Congo red aggregates. ATR-FTIR, TEM
and DLS confirmed the presence of amyloid in solution.
The aggregates were shown to be very toxic to
neuroblastoma cells cultures. In light of these results,
we propose for the first time a role of the leader peptide
on the immature (unprocessed) form of variant B
cystatin C in its association with increased risk of AD
and AMD.
Materials and methods
Prediction of the peptide aggregation and
amyloid propensities
AGGRESCAN, Waltz, TANGO and Zipper DB Analysis –
The intrinsic aggregation and amyloid formation propensities
of mature and immature cystatin C were evaluated by
these four methods employing default settings, using as
input the primary sequence of either the mature protein or
the immature form of variant B cystatin C. Uniprot Accession
Number P01034 (CYTC_HUMAN) and the dbSNP
Classification code rs1064039 [Homo sapiens].
Peptides synthesis
The peptides were purchased from CASLO ApS c/o Scion
Denmark Technical Universit (Lyngby, Denmark). The
purity was higher than 98%. The peptide was weighted and
diluted in DMSO to obtain a stock solution of soluble peptide
with a concentration of 3 mM.
Peptides aggregation assays
Aggregation was performed at different concentrations as
stated in legends. Before each experiment, the stock solution
was diluted in PBS pH 7.4 and Th-T was added at 20 lM
(Fig. 2B) or 30 lM (Fig. 2A) final concentrations. The samples
were incubated at 25 °C with no agitation. Samples
were excited at 450 nm while emission was collected from
400 to 600 nm on a spectrofluorometer Jasco 8200. Intensity
at 482 nm was used to probe amyloid formation.
Congo red binding assay
The binding of CR to aggregated peptide was evaluated by
diluting fivefold the incubated peptide in CR, resulting in a
final CR concentration of 10 lM, and recording the absorbance
spectra of the mixture in the 380–670 nm range in a
Cary 400 spectrophotometer (Varian, Palo Alto, CA, USA ).
Dynamic light scattering measurements
Peptide solutions were let to aggregate at the conditions
stated above (100 lM) and measures were performed at 1 h
646 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
Amyloid properties of the leader peptide R. Sant’Anna et al.
and after 24 h of aggregation. DLS measurements were
performed at 25 °C in a Malvern Zetasizer Nano S90 (Malvern,
Worcestershire, UK). Each sample was measured tree
times; average distributions are presented.
Infrared spectroscopy
Attenuated total reflectance Fourier transform infrared
spectroscopy (ATR-FTIR) was used to determine the secondary
structure content of the aggregates formed at different
times during the aggregation kinetics. The experiments
were carried out using a Bruker Tensor 27 FTIR spectrometer
(Bruker Optics, Billerica, MA) with a Golden Gate
MKII ATR accessory. Each spectrum consists of 16 accumulations
measured at a resolution of 2 cmÀ1
in a wavelength
range between 1700 and 1600 cmÀ1
. Infrared spectra
were fitted through overlapping Gaussian curves, and the
amplitude, mass center, bandwidth at half of the maximum
amplitude and area for each Gaussian function were calculated
employing a nonlinear peak-fitting equation using the
PEAKFIT package (Systat Software, San Jose, CA, USA).
Transmission electron microscopy
Aggregated samples were diluted in Mili-Q water to 10 lM.
Five microliter of the solution was absorbed onto carboncoated
copper grids for 5 min and blotted to remove excess
material. Uranyl acetate (2% w/v) was used for negative
staining. Samples were dried on air for 5 min. Grids were
exhaustively scanned with a Hitachi H-7000 transmission
electron microscope operating at a voltage of 75 kV.
Neuroblastoma cell (N2a) culture and viability
assay
Neuroblastoma cells (N2a) were cultured in Dulbecco’s
modified Eagle’s medium supplemented with 10% fetal
bovine serum and 2% antibiotic (gentamicin) in a 5% CO2
atmosphere for 3 days and then transferred to a 96-well
plate and allowed to adhere for 24 h (5000 cells per well).
Following cell adhesion, aggregates were added to achieve
final concentration of 5 lM. Cells were challenged for 24 h,
and final cellular viability was measured by an MTT assay
as previously described [37].
Results and Discussion
Bioinformatics predicts that the signal peptide of
variant B cystatin C is aggregation prone and
amyloidogenic
AGGRESCAN is considered a reliable program to predict
the aggregation propensity of proteins in vivo [38] as
the aggregation score for each amino acid was derived
from aggregation assays carried out on a bacterial system
[35]. AGGRESCAN identifies aggregation-prone
regions (APR) as stretches of a given sequence at least
five amino acids long with high aggregation propensity
scores [39].
Recently, the pentapeptide 47
LQVVR51
from cystatin
C was demonstrated to form fibrils in vitro [40].
The same group by using the algorithm AMYLPRED
identified two other peptides from the core of cystatin
C with high aggregation propensity. These peptides,
namely 56
IVAGVNYFLD65
and 95
AFASFQIYAV104
Fig. 2. Monitoring aggregation of variant B cystatin C signal
peptide in vitro. Panel A: Increasing concentrations of the signal
peptide (5–100 lM, as indicated) were incubated in PBS (pH 7.4) at
25 °C without agitation (aggregation condition). After 1 h under this
condition, Th-T binding was evaluated by measuring its
fluorescence emission (Excitation = 450 nm and Emission = 450–
600 nm). [Th-T] = 20 lM. The inset in Panel A shows the spectra
of CR alone and in the presence of aggregates formed after 24 h
of incubation. Panel B displays the kinetics of aggregation at
different concentrations of the signal peptide (see the legend
inside this panel) as followed by the intensity of Th-T fluorescence
at 482. [Th-T] = 30 lM.
647FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
R. Sant’Anna et al. Amyloid properties of the leader peptide
(it contains a C97A mutation), were also amyloidogenic
upon incubation for several days at pH 5.5.
These peptides numberings are from the mature cystatin
C. In our Fig. 1B, green numbering [41].
Figure 1C shows the aggregation propensity scores
as predicted by AGGRESCAN for the immature (variant B)
and mature forms of cystatin C. As seen, according to
this algorithm, the mature protein contains three APRs
located at positions 54–58 (DFAVG), 83–92 (VAGVNYFLDV)
and 122–130 (FCSFQIYAV) (immature
form numbering, orange numbering on Fig. 1B). The
regions 83–89 and 122–130 in the immature form correspond
to residues 57–66 and 96–104 in the processed
protein (green numbering on Fig. 1B). Therefore, these
two APRs almost completely overlap with those previously
identified using AMYLPRED [42].
Interestingly, the signal peptide present in the immature
(uncleaved) protein (segment 9–22; LLLLAILAVALAVS)
presented even higher aggregation scores
than the three previous peptides identified by AGGRESCAN.
The primary sequence of precursor cystatin C,
including the signal peptide, is presented in Fig. 1B
and the aggregation-prone regions are highlighted.
APRs within the primary sequence of proteins are
flanked by gatekeeper residues such as proline and
charged amino acids, which modulate aggregation by
opposing nucleation of aggregates [43,44]. In the case
of the four stretches of cystatin C identified by AGGRESCAN,
we could observe the presence of these gatekeeper
residues. In particular, in the case of the signal
peptide, the sequence is flanked by two proline residues
at positions 8 and 23 (P-9LLLLAILAVALAVS22-P).
Recently, our group identified the
gatekeeper residues in the amyloidogenic protein
transthyretin. By replacing one of these residues in the
sequence of a peptide derived from TTR by a
hydrophobic amino acid (K35L), we could convert a
nonaggregating peptide into a highly aggregationprone
one [45].
Next, in order to investigate whether these APRs
would also exhibit high scores for forming amyloidspecific
aggregates, the Waltz program was utilized
(Fig. 1D). As seen, all four segments of cystatin C,
including the signal peptide (segment 10–20 as predicted
by this algorithm), presented propensity to form
amyloid assemblies. In this case, the other identified
regions with this property were 56–61 (AVGEYN),
84–92 (AGVNYFLDV) and 124–130 (SFQIIYAV)
(immature form numbering, orange numbering on
Fig. 1B). Both analyses were performed with the two
protein sequences, but as the only difference present
between them is the signal peptide, solid and dashed
lines appear superimposed.
We also used other algorithms such as TANGO
[46], Zipper DB [47] and AMYLPRED2 to analyze
the aggregation and amyloid propensity of the cystatin
C signal peptide. TANGO and AMYLPRED2 predicted
the 9–20 and 9–21 stretches to be aggregation
prone, respectively, whereas Zipper DB predicted the
sequence 9–17 to be highly amyloidogenic. Because the
methods we employed here for computational predictions
rely on very different principles, the prediction of
the signal peptide being highly aggregation prone and
amyloidogenic was considered to be very robust. The
intrinsic high hydrophobicity of signal peptides does
not always implies that they can form typical beta
sheet rich ordered amyloid fibrils, as cystatin C signal
peptide does. To strengthen this very important point,
we ran Waltz for 20 signal peptides of different human
proteins randomly chosen from Uniprot (data not
shown). Interestingly, only 2 out of 20 presented
regions that could drive amyloid formation (Waltz
negative: P30453, P30483, Q9TNN7, P04114, P25774,
P08185, P16870, P11597, P28325, Q9UBU2, Q9NR61,
P16444, P54803, P47871, Q6UWU2, P42262; Waltz
Positive: Q8NFZ8, P12111).
Variant B cystatin C signal peptide is able to
form amyloid fibrils in vitro
Having in silico evidence that the signal peptide presents
high aggregation propensity and amyloidogenicity,
we synthesized the peptide with the mutation
A25T (1MAGPLRAPLLLLAILAVALAVSPATG26)
(Fig. 1B) and incubated it at increasing concentrations
(5–100 lM) for 1 h at pH 7.4, 25 °C under stagnant
conditions. As shown in Fig. 2A, there was a proportional
increase in Thioflavin-T (Th-T) emission at
increasing concentrations of the peptide, suggesting
formation of amyloid-like aggregates in solution.
Congo red (CR) binding assays were also performed
to confirm the presence of amyloid-like aggregates
(Fig. 2A, inset).
In order to follow the aggregation kinetics of the
signal peptide, increasing concentrations of the peptide
were incubated with Th-T and its fluorescence emission
was collected over time (Fig. 2B). Notably, even
at very low peptide concentration such as 5 lM, maximum
emission of Th-T was attained in the first acquisition
points, with no further change in the
fluorescence signal of the probe even after 3 days
under aggregation conditions. This suggests that aggregation
of the signal peptide of cystatin C is very fast.
We tried several conditions to slow down the aggregation
process of the signal peptide such as low temperature
(4 and 8 °C) and addition of 5, 10 and 15%
648 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
Amyloid properties of the leader peptide R. Sant’Anna et al.
DMSO to the aggregating buffer without any success
(not shown).
In order to confirm the instantaneous aggregation of
the signal peptide, a suspension with 50 lM peptide
diluted into 30 lM Th-T was filtered through a 0.22
micrometer filter before measurements. The filter
became yellow due to the retention of the aggregates
bound to Th-T. The filtered solution, when let to
aggregate, was not able to bind additional Th-T since
all soluble peptide was converted into amyloid aggre-
1 h 24 h 72 hA
B
C
Fig. 3. Characterization of morphology, secondary structure content, size distribution and toxicity of the aggregates formed by the variant B
cystatin C signal peptide. (A) Signal peptide at 100 lM was left to aggregate under the conditions described in Fig. 2. At 1 h (left images),
24 h (middle images) and 72 h (right images) aliquots were withdrawn and imaged by TEM. Bars represent 100, 200 and 500 nm. (B)
Secondary structure changes that take place upon aggregation of the signal peptide analyzed by ATR-FTIR. The spectra displayed on the left
side are those of the soluble peptide (full line), and of the samples aggregated up to 1 h as follows: 5 min (dotted), 15 min (short dash),
40 min (dash) and 60 min (large dash). On the right are displayed the FTIR spectra of the samples aggregated for longer times as follows:
1 h (full), 16 h (dot), 24 h (short dash) and 72 h (dash-dot). Secondary structure quantification from the deconvolution of these spectra is
displayed in Table 1. (C) DLS was used to determine the size of the main species present at 1 h (dash-dot line) and 24 h (full line) under
aggregating conditions. A calibration with silicone particles was performed and is shown in dotted line. (D) Toxicity of the aggregates
formed at 1 and 24 h on neuroblastoma cells determined by MTT assay. In each case, 5 lM of aggregated material was added to the cells
in culture remaining in contact with them for 24 h before cell viability evaluation.
649FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
R. Sant’Anna et al. Amyloid properties of the leader peptide
gates that were retained in the filter (Fig. 2B, dashed
line).
Characterizing the morphology, secondary
structure and toxicity of the aggregates formed
by the signal peptide of variant B cystatin C
Next, we used TEM to study the morphology of the
species present at different times of aggregation
(Fig. 3A). Interestingly, although Th-T binding has
already leveled-off at 1 h under aggregation conditions,
the great majority of the aggregated material
presented an amorphous appearance and only few
amyloid fibrils were observed at this time (Fig. 3A,
left). The binding of Th-T to these amorphous species
suggests that they already present cross-b fold able to
accommodate this probe. These species are on-pathway
to fibril formation since after 72 h under aggregation
condition only mature fibrils were observed by
TEM (Fig. 3A, right). At 24 h, it was already possible
to see the presence of fibrils but small amorphous
aggregates still remained (Fig. 3A, middle).
In order to get insights into the secondary structural
changes that take place with the signal peptide upon
aggregation, we took advantage of ATR-FTIR (Fig. 3
B). Left panel shows the FTIR spectra of the peptide
dried from a stock solution with 100% DMSO (continuous
line) and those collected during the first hour
of aggregation. As seen, in DMSO the peptide presented
a unique and broad peak centered at
1660 cmÀ1
, which corresponds to disordered, randomcoiled
structures, which is a strong evidence that the
peptide is monomeric in DMSO [48–50].
At the initial times of aggregation (up to 1 h), the
spectra changed and a new peak at 1628 cmÀ1
appeared which was assigned to b-sheet structures. At
the same time, the peak related to random-coil structures
(1660 cmÀ1
) decreased but was still present. The
presence of b-sheet structures in these initial aggregates
explains why these species bind Th-T (as shown in
Fig. 2), but they are not fully organized, as seen by
TEM (Fig. 3A, left).
As aggregation proceeded for longer times (16, 24
and 72 h), the peak of b-sheet (1627 cmÀ1
) increased
even further, with a concomitant decrease of the peak
of random coil at 1660 cmÀ1
. Table 1 summarizes the
secondary structural changes that take place upon
aggregation of the signal peptide of cystatin C.
Next, to further confirm the difference in sizes of
the species formed at 1 and 24 h, DLS measurements
were performed (Fig. 3C). The main sizes of the two
more prevalent species present at 1 h were
0.194 nm Æ 0.06 nm and 904 Æ 32.14 nm. After 24 h
under aggregation conditions, there were two species
present, one with 947 Æ 134 nm and an extreme large
species with 3008 Æ 500 nm. Interestingly and corroborating
the FTIR data, the species with around
900 nm were still present at 24 h. Given the fact that
the aggregation pathways of aggregation-prone proteins
are distinct leading to the accumulation of a myriad
of aggregated species with different size and
morphologies, it is difficult to make a direct comparison
of the species sizes observed here with others.
However, initial fibrillar species from Ab peptide have
been shown to have sizes from 200 to 700 nm [51,52].
In another study [53], the formation of transthyretin
fibrillar oligomers of 1000 nm and mature fibrils with
mean size of 4000 nm was reported [54], in accordance
with our DLS data where early aggregating species
with 900 nm and mature fibrils with 3000 nm were
detected. We performed a calibration assay using a
standard solution of silicone particles and the expected
size was found (342 nm).
Next, neuroblastoma cell line N2a was used to probe
the cytotoxicity of the aggregates formed at 1 and 24 h
and cell viability was evaluated by MTT assays. Both
aggregates tested at 5 lM concentration were very cytotoxic,
killing ~ 40% of the cells (Fig. 3D).
Conclusions
The substitution A25T in the signal peptide of the
variant B precursor cystatin C compromises its cleavage
and/or processing precluding the efficient formation
of the mature form of the protein, which is
secreted from the cells to the extracellular milieu where
it exerts its biochemical effect as a cysteine protease
inhibitor. The variant leader presents a significant
stretch of sequence with amyloidogenic propensity,
which is even higher than the amyloidogenicsequences
in the sequence of mature protein. This is relevant in
the case when the leader is not cleaved thus imparting
Table 1. Secondary structure changes that take place upon
aggregation of variant B cystatin C signal peptide measured by
FTIR.
Time
Inter b-sheet (%) Disordered (%) b-sheet (%)
1623–1641 cmÀ1
1642–1657 cmÀ1
1674–1695 cmÀ1
0 – 100 –
5 min 42.98 37.44 19.57
15 min 42.07 39.56 18.36
40 min 41.64 37.23 21.11
1 h 41.78 38.42 19.79
16 h 64.02 35.98 –
24 h 83.55 16.43 –
72 h 86.22 13.76 –
650 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
Amyloid properties of the leader peptide R. Sant’Anna et al.
a higher amyloidogenic propensity to the full length
variant precursor compared with the mature, processed
protein. A model of the functional consequences for
trafficking and protein aggregation of this variant precursor
cystatin C is presented in Fig. 4. As previously
shown, in the extracellular space mature cystatin C
interacts with Ab diminishing its aggregation [17,18].
An inefficiently cleaved/processed variant B cystatin C
leads on one hand to a significantly reduced secretion
of the mature cystatin C as shown previously [33,34]
and, on the other, to the accumulation of full length
variant precursor protein containing the mutated,
highly amyloidogenic leader. On the contrary, the WT
signal peptide is efficiently excised from the normal
cystatin C protein during its processing and should be
completely cleaved. To our knowledge, there is no
description of the accumulation/aggregation of the
WT signal peptide inside RPE cells. In light of our
data, it is tempting to suggest that this variant might
aggregate into amyloid fibrils through the leader peptide,
since the other stretches with high aggregation
propensity are protected in the globular domain of the
protein. These aggregates composed of uncleaved/unprocessed
variant B could be formed either inside or
Fig. 4. Schematic representation of the effects of the A25T mutation on variant B precursor cystatin C cell trafficking and its possible
implications for protein aggregation. On the left, a RPE cell that produces WT cystatin C. Processing of cystatin C takes place normally in
the ER/Golgi network. The signal peptide is cleaved leading to the formation of the mature form of the protein, which is secreted from the
RPE cells. Outside the cell, cystatin C acts as inhibitor of cysteine proteases. Besides, it can interact with Ab modulating its aggregation
properties. On the right, a RPE cell that produces the variant B, which is abnormally processed accumulating in the periphery of the
mitochondria, probably as an aggregated material induced by the presence of the signal peptide. This variant is inefficiently secreted what
impairs its protective effect against Ab aggregation.
651FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
R. Sant’Anna et al. Amyloid properties of the leader peptide
outside the cells leading to a gain of toxic function. As
previously demonstrated [34] this variant is retained
close to mitochondria walls forming structures resistant
to protease digestion [34], a characteristic feature
of high molecular weight aggregates, including amyloid
fibrils. Upon aggregation of variant B, other important
intra- or extracellular protein components could be
incorporated into this aggregated material, possibly
depriving the cell/tissue from a specific function and
forming protein aggregates otherwise not present in
the cells. Additionally, lower levels of functional
mature cystatin C in the extracellular space could be
very deleterious, contributing to increased proteolytic
action at its extracellular sites of activity. Also, there
will be less available protein to interact with Ab peptide,
thus enhancing its aggregation, which may
explain the variant B cystatin C association with a
higher risk of developing AD.
As the amyloid deposits are the histological hallmark
of AD, drusen are the extracellular deposits found in
AMD. The presence of Ab peptide has been reported in
patients with severe forms of AMD [55–58]. The
involvement of protein aggregation in AMD pathogenesis
is suggested by the presence of various toxic amyloid
structures in drusen [56,59] and, indirectly, by the
decrease in the AMD risk obtained through anti-amyloid
treatment in a mouse model [60]. Variant B cystatin
C aggregation may therefore contribute to AMD pathogenesis.
In addition, if this mutation avoids the observed
partnership between cystatin C and Ab peptide it can
account for the higher incidence of AD in homozygous
patients. Further work is being carried out to investigate
amyloid deposits formed by this variant inside RPE
cells. More experimental approaches are necessary to
reveal in full the mechanism of action of variant B cystatin
C in these major degenerative diseases and we are
actively pursuing such studies. With the actual lack of
any in vitro aggregation experimental evidence to link
the higher incidence of AD and AMD with variant B
cystatin C, we believe that this study provides the
important first clues for further investigations.
Acknowledgements
RS was a recipient of a fellowship from the Brazilian
program Science without Borders – CAPES. We also
thank the funding agencies Fundacß~ao Carlos Chagas
Filho de Amparo a Pesquisa do Estado do Rio de
Janeiro (FAPERJ), Coordenacß~ao de Aperfeicßoamento
de Pessoal de Nıvel Superior (CAPES), Conselho
Nacional de Desenvolvimento Cientıfico e Tecnologico
(CNPq) and Ministerio de Economia y Competividad,
Spain [BFU2013-44763-P].
Conflict of interest
The authors declare no conflict of interest.
Author contribution
DF and LP conceived and supervised the study; RS,
SV, DF and LP designed experiments; RS and SN performed
experiments; LP and SV provided tools and
reagents; RS, DF, LP, SN and SV analyzed data; RS,
DF and LP wrote the manuscript; SV, LP and DF
made manuscript revisions.
References
1 Grubb A and L€ofberg H (1982) Human gamma-trace, a
basic microprotein: amino acid sequence and presence
in the adenohypophysis. Proc Natl Acad Sci USA 79,
3024–3027.
2 Abrahamson M, Alvarez-Fernandez M and Nathanson
C-M (2003) Cystatins. Biochem Soc Symp 70, 179–199.
3 Turk V and Bode W (1991) The cystatins: protein
inhibitors of cysteine proteinases. FEBS Lett 285, 213–
219.
4 Engh RA, Dieckmann T, Bode W, Auerswald EA,
Turk V, Huber R and Oschkinat H (1993)
Conformational Variability of Chicken Cystatin. J Mol
Biol 234, 1060–1069.
5 Tavera C, Leung-Tack J, Prevot D, Gensac MC,
Martinez J, Fulcrand P and Colle A (1992) Cystatin C
secretion by rat glomerular mesangial cells: autocrine
loop for in vitro growth-promoting activity. Biochem
Biophys Res Commun 182, 1082–1088.
6 Hasegawa A, Naruse M, Hitoshi S, Iwasaki Y,
Takebayashi H and Ikenaka K (2007) Regulation of glial
development by cystatin C. J Neurochem 100, 12–22.
7 Bobek LA and Levine MJ (1992) Cystatins–inhibitors
of cysteine proteinases. Crit Rev Oral Biol Med 3, 307–
332.
8 Warfel AH, Zucker-Franklin D, Frangione B and
Ghiso J (1987) Constitutive secretion of cystatin C
(gamma-trace) by monocytes and macrophages and its
downregulation after stimulation. J Exp Med 166,
1912–1917.
9 Lerner UH and Grubb A (2009) Human cystatin C, a
cysteine proteinase inhibitor, inhibits bone resorption
in vitro stimulated by parathyroid hormone and
parathyroid hormone-related peptide of malignancy.
J Bone Miner Res 7, 433–440.
10 Yamada M (2000) Cerebral amyloid angiopathy: an
overview. Neuropathology 20, 8–22.
11 Grubb AO (2000) Cystatin C–properties and use as
diagnostic marker. Adv Clin Chem 35, 63–99.
12 Palsdottir A, Snorradottir AO and Thorsteinsson L
(2006) Hereditary cystatin C amyloid angiopathy:
652 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
Amyloid properties of the leader peptide R. Sant’Anna et al.
genetic, clinical, and pathological aspects. Brain Pathol
16, 55–59.
13 Palsdottir A, Helgason A, Palsson S, Bjornsson HT,
Bragason BT, Gretarsdottir S, Thorsteinsdottir U,
Olafsson E and Stefansson K (2008) A drastic
reduction in the life span of cystatin C L68Q carriers
due to life-style changes during the last two centuries.
PLoS Genet 4, e1000099.
14 Nagai A, Kobayashi S, Shimode K, Imaoka K,
Umegae N, Fujihara S and Nakamura M (1998) No
mutations in cystatin C gene in cerebral amyloid
angiopathy with cystatin C deposition. Mol Chem
Neuropathol 33, 63–78.
15 van Duinen SG, Casta~no EM, Prelli F, Bots GT,
Luyendijk W and Frangione B (1987) Hereditary
cerebral hemorrhage with amyloidosis in patients of
Dutch origin is related to Alzheimer disease. Proc Natl
Acad Sci USA 84, 5991–5994.
16 Sastre M, Calero M, Pawlik M, Mathews PM, Kumar
A, Danilov V, Schmidt SD, Nixon RA, Frangione B
and Levy E (2004) Binding of cystatin C to Alzheimer’s
amyloid beta inhibits in vitro amyloid fibril formation.
Neurobiol Aging 25, 1033–1043.
17 Mi W, Pawlik M, Sastre M, Jung SS, Radvinsky DS,
Klein AM, Sommer J, Schmidt SD, Nixon RA,
Mathews PM et al. (2007) Cystatin C inhibits amyloidbeta
deposition in Alzheimer’s disease mouse models.
Nat Genet 39, 1440–1442.
18 Gauthier S, Kaur G, Mi W, Tizon B and Levy E (2011)
Protective mechanisms by cystatin C in
neurodegenerative diseases. Front Biosci (Schol Ed) 3,
541–554.
19 Paraoan L, Grierson I and Maden BE (2000) Analysis
of expressed sequence tags of retinal pigment
epithelium: cystatin C is an abundant transcript. Int J
Biochem Cell Biol 32, 417–426.
20 Ida H, Boylan SA, Weigel AL, Smit-McBride Z, Chao
A, Gao J, Buchoff P, Wistow G and Hjelmeland LM
(2004) EST analysis of mouse retina and RPE/choroid
cDNA libraries. Mol Vis 10, 439–444.
21 Wistow G, Bernstein SL, Wyatt MK, Fariss RN, Behal
A, Touchman JW, Bouffard G, Smith D and Peterson
K (2002) Expressed sequence tag analysis of human
RPE/choroid for the NEIBank Project: over 6000 nonredundant
transcripts, novel genes and splice variants.
Mol Vis 8, 205–220.
22 Anderson DH, Mullins RF, Hageman GS and Johnson
LV (2002) A role for local inflammation in the
formation of drusen in the aging eye. Am J Ophthalmol
134, 411–431.
23 Booij JC, Baas DC, Beisekeeva J, Gorgels TGMF and
Bergen AAB (2010) The dynamic nature of Bruch’s
membrane. Prog Retin Eye Res 29, 1–18.
24 Mitter SK, Rao HV, Qi X, Cai J, Sugrue A, Dunn
WA, Grant MB and Boulton ME (2012) Autophagy in
the retina: a potential role in age-related macular
degeneration. Adv Exp Med Biol 723, 83–90.
25 Kaarniranta K, Sinha D, Blasiak J, Kauppinen A,
Vereb Z, Salminen A, Boulton ME and Petrovski G
(2013) Autophagy and heterophagy dysregulation leads
to retinal pigment epithelium dysfunction and
development of age-related macular degeneration.
Autophagy 9, 973–984.
26 Hua Y, Zhao H, Lu X, Kong Y and Jin H (2012)
Meta-analysis of the cystatin C(CST3) gene G73A
polymorphism and susceptibility to Alzheimer’s disease.
Int J Neurosci 122, 431–438.
27 Butler JM, Sharif U, Ali M, McKibbin M, Thompson
JP, Gale R, Yang YC, Inglehearn C and Paraoan L
(2015) A missense variant in CST3 exerts a recessive
effect on susceptibility to age-related macular
degeneration resembling its association with
Alzheimer’s disease. Hum Genet 134, 705–715.
28 Paraoan L, Hiscott P, Gosden C and Grierson I (2010)
Cystatin C in macular and neuronal degenerations:
Implications for mechanism(s) of age-related macular
degeneration. Vision Res 50, 737–742.
29 Finckh U, von der Kammer H, Velden J, Michel T,
Andresen B, Deng A, Zhang J, M€uller-Thomsen T,
Zuchowski K, Menzer G et al. (2000) Genetic
association of a cystatin C gene polymorphism with lateonset
Alzheimer disease. Arch Neurol 57, 1579–1583.
30 Zurdel J, Finckh U, Menzer G, Nitsch RM and
Richard G (2002) CST3 genotype associated with
exudative age related macular degeneration. Br J
Ophthalmol 86, 214–219.
31 Paraoan L, White MR, Spiller DG, Grierson I and
Maden BE (2001) Precursor cystatin C in cultured
retinal pigment epithelium cells: evidence for processing
through the secretory pathway. Mol Membr Biol 18,
229–236.
32 Paraoan L, Grierson I and Maden BEH (2003) Fate of
cystatin C lacking the leader sequence in RPE cells.
Exp Eye Res 76, 753–756.
33 Benussi L, Ghidoni R, Steinhoff T, Alberici A, Villa A,
Mazzoli F, Nicosia F, Barbiero L, Broglio L, Feudatari
E et al. (2003) Alzheimer disease-associated cystatin C
variant undergoes impaired secretion. Neurobiol Dis 13,
15–21.
34 Paraoan L, Ratnayaka A, Spiller DG, Hiscott P, White
MRH and Grierson I (2004) Unexpected intracellular
localization of the AMD-associated cystatin C variant.
Traffic 5, 884–895.
35 Conchillo-Sole O, de Groot NS, Aviles FX, Vendrell J,
Daura X and Ventura S (2007) AGGRESCAN: a
server for the prediction and evaluation of “hot spots”
of aggregation in polypeptides. BMC Bioinformatics 8,
65.
36 Oliveberg M (2010) Waltz, an exciting new move in
amyloid prediction. Nat Methods 7, 187–188.
653FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
R. Sant’Anna et al. Amyloid properties of the leader peptide
37 Malmo C, Vilasi S, Iannuzzi C, Tacchi S, Cametti C,
Irace G and Sirangelo I (2006) Tetracycline inhibits
W7FW14F apomyoglobin fibril extension and keeps the
amyloid protein in a pre-fibrillar, highly cytotoxic state.
FASEB J 20, 346–347.
38 Belli M, Ramazzotti M and Chiti F (2011) Prediction
of amyloid aggregation in vivo. EMBO Rep 12, 657–
663.
39 Sanchez de Groot N, Pallares I, Aviles FX, Vendrell J
and Ventura S (2005) Prediction of “hot spots” of
aggregation in disease-linked polypeptides. BMC Struct
Biol 5, 18.
40 Tsiolaki PL, Hamodrakas SJ and Iconomidou VA
(2015) The pentapeptide LQVVR plays a pivotal role in
human cystatin C fibrillization. FEBS Lett 589, 159–164.
41 Tsiolaki PL, Louros NN, Hamodrakas SJ and
Iconomidou VA (2015) Exploring the “aggregationprone”
core of human Cystatin C: A structural study.
J Struct Biol 191, 272–280.
42 Tsolis AC, Papandreou NC, Iconomidou VA and
Hamodrakas SJ (2013) A consensus method for the
prediction of “aggregation-prone” peptides in globular
proteins. PLoS One 8, e54175.
43 Reumers J, Maurer-Stroh S, Schymkowitz J and
Rousseau F (2009) Protein sequences encode safeguards
against aggregation. Hum Mutat 30, 431–437.
44 Richardson JS and Richardson DC (2002) Natural betasheet
proteins use negative design to avoid edge-to-edge
aggregation. Proc Natl Acad Sci USA 99, 2754–2759.
45 Sant’Anna R, Braga C, Varej~ao N, Pimenta KM,
Gra~na-Montes R, Alves A, Cortines J, Cordeiro Y,
Ventura S and Foguel D (2014) The importance of a
gatekeeper residue on the aggregation of transthyretin:
implications for transthyretin-related amyloidoses. J
Biol Chem 289, 28324–28337.
46 Fernandez-Escamilla A-M, Rousseau F, Schymkowitz J
and Serrano L (2004) Prediction of sequence-dependent
and mutational effects on the aggregation of peptides
and proteins. Nat Biotechnol 22, 1302–1306.
47 Iwata K, Fujiwara T, Matsuki Y, Akutsu H, Takahashi
S, Naiki H and Goto Y (2006) 3D structure of amyloid
protofilaments of beta2-microglobulin fragment probed
by solid-state NMR. Proc Natl Acad Sci USA 103,
18119–18124.
48 Byler DM and Susi H (1986) Examination of the
secondary structure of proteins by deconvolved FTIR
spectra. Biopolymers 25, 469–487.
49 Barth A and Zscherp C (2002) What vibrations tell us
about proteins. Q Rev Biophys 35, 369–430.
50 Jackson M and Mantsch HH (1991) Protein secondary
structure from FT-IR spectroscopy: correlation with
dihedral angles from three-dimensional Ramachandran
plots. Can J Chem 69, 1639–1642.
51 Merz PA, Wisniewski HM, Somerville RA, Bobin SA,
Masters CL and Iqbal K (1983) Ultrastructural
morphology of amyloid fibrils from neuritic and
amyloid plaques. Acta Neuropathol 60, 113–124.
52 Bitan G, Kirkitadze MD, Lomakin A, Vollers SS,
Benedek GB and Teplow DB (2003) Amyloid beta protein
(Abeta) assembly: Abeta 40 and Abeta 42
oligomerize through distinct pathways. Proc Natl Acad
Sci USA 100, 330–335.
53 Ferreira N, Saraiva MJ and Almeida MR (2011)
Natural polyphenols inhibit different steps of the
process of transthyretin (TTR) amyloid fibril formation.
FEBS Lett 585, 2424–2430.
54 Ferreira N, Saraiva MJ and Almeida MR (2012)
Natural polyphenols as modulators of TTR
amyloidogenesis: in vitro and in vivo evidences towards
therapy. Amyloid 19 (Suppl 1), 39–42.
55 Dentchev T, Milam AH, Lee VM-Y, Trojanowski JQ
and Dunaief JL (2003) Amyloid-beta is found in drusen
from some age-related macular degeneration retinas,
but not in drusen from normal retinas. Mol Vis 9, 184–
190.
56 Luibl V, Isas JM, Kayed R, Glabe CG, Langen R and
Chen J (2006) Drusen deposits associated with aging
and age-related macular degeneration contain
nonfibrillar amyloid oligomers. J Clin Invest 116, 378–
385.
57 Anderson DH, Talaga KC, Rivest AJ, Barron E,
Hageman GS and Johnson LV (2004) Characterization
of beta amyloid assemblies in drusen: the deposits
associated with aging and age-related macular
degeneration. Exp Eye Res 78, 243–256.
58 Johnson LV, Leitner WP, Rivest AJ, Staples MK,
Radeke MJ and Anderson DH (2002) The Alzheimer’s
A beta -peptide is deposited at sites of complement
activation in pathologic deposits associated with aging
and age-related macular degeneration. Proc Natl Acad
Sci USA 99, 11830–11835.
59 Isas JM, Luibl V, Johnson LV, Kayed R, Wetzel R,
Glabe CG, Langen R and Chen J (2010) Soluble and
mature amyloid fibrils in drusen deposits. Invest
Ophthalmol Vis Sci 51, 1304–1310.
60 Ding J-D, Johnson LV, Herrmann R, Farsiu S, Smith
SG, Groelle M, Mace BE, Sullivan P, Jamison JA,
Kelly U et al. (2011) Anti-amyloid therapy protects
against retinal pigmented epithelium damage and vision
loss in a model of age-related macular degeneration.
Proc Natl Acad Sci USA 108, E279–E287.
61 Kolodziejczyk R, Michalska K, Hernandez-Santoyo A,
Wahlbom M, Grubb A and Jaskolski M (2010) Crystal
structure of human cystatin C stabilized against
amyloid formation. FEBS J 277, 1726–1737.
62 Bode W, Engh R, Musil D, Thiele U, Huber R,
Karshikov A, Brzin J, Kos J and Turk V (1988) The
2.0 A X-ray crystal structure of chicken egg white
cystatin and its possible mode of interaction with
cysteine proteinases. EMBO J 7, 2593–2599.
654 FEBS Letters 590 (2016) 644–654 ª 2016 Federation of European Biochemical Societies
Amyloid properties of the leader peptide R. Sant’Anna et al.