calpain: transitioning from the use of the protease core to the
TRANSCRIPT
CALPAIN: TRANSITIONING FROM THE USE OF THE
PROTEASE CORE TO THE FULL-LENGTH ENZYME FOR THE
DEVELOPMENT OF SPECIFIC SUBSTRATES AND INHIBITORS
by
JACQUELINE CHRISTINE KELLY
A thesis submitted to the Department of Biochemistry
In conformity with the requirements for
the degree of Master of Science
Queen’s University
Kingston, Ontario, Canada
(September, 2008)
Copyright © Jacqueline Christine Kelly, 2008
ii
Abstract
Calpains are a family of calcium-dependent cysteine proteases involved in
intracellular signaling. They participate in many normal cellular processes such as cell
motility and apoptosis but when over-activated they contribute to diseases ranging from
ischemic injury to neurodegenerative disorders.
The major calpain isoforms µ- and m- are large heterodimeric enzymes that are
subject to autoproteolysis and aggregation when activated by Ca2+. To avoid these
complications the protease core (domains I and II) has been used to screen inhibitors and
design substrates. Using the protease core of µ-calpain, I showed that the superior calpain
substrate, PLFMER, is cut at the intended scissile bond between F and M. Alanine
substitutions at each position optimized the sequence to PLFAAR, which has a 2.3-fold
higher turnover rate. The set of substrates derived from this study provided a tool for
profiling the activity of calpain isoforms. One disadvantage of the protease core is that it
is less active than the whole enzyme. This was even more apparent with the protease
cores of the tissue-specific calpains 3, 8, 9 and 15 such that it prevented their use in
substrate and inhibitor screening.
The recently solved crystal structure of calcium-bound full-length m-calpain has
revealed additional sites for the interaction of substrates and inhibitors in the unprimed
side of the catalytic cleft provided by domain III. To sample these sites, it is necessary to
prevent the full-length calpain from aggregating and precipitating upon calcium binding.
I have developed a method here that uses portions of calpastatin (CAST), the natural
iii
endogenous inhibitor and stabilizer of calpain, to keep the enzyme soluble. By artificially
connecting those portions of calpastatin that bind to calpain domains IV and VI, it is
possible to stabilize the enzyme without blocking its active site. Of the three constructs
made, 1C-2A, 2C-3A, and 3C-4A, the 3C-4A peptide was shown to completely inhibit
aggregation of m-calpain at a 1:1 molar ratio, as monitored by turbidity. This mechanism
of stabilization will permit the use of full-length calpains for the development of specific
substrates and inhibitors.
iv
Acknowledgements
I would first and foremost like to acknowledge my research supervisor, Dr. Peter L.
Davies from whom I have gained a great deal. Not only through his guidance in my research but
also through his enthusiasm and love for the lab, have I appreciated the opportunity to be a part of
his team. He has an incredible dedication to scientific research as well as the professional and
personal development of the students and employees around him. It has been a great honour to
have spent my graduate studies under his guidance and my future successes in research will be in
a large part attributed to his mentorship.
In my personal life, I would like to thank my loving husband, Shane Kelly, who has
expressed his unwavering support and encouragement for my personal and academic decisions
during the past 6 years of our marriage. I would also like to acknowledge my parents, Ron and
Shirley Charles, as well as my brothers, Scott and Brian, for their love and support throughout my
life.
My immense appreciation goes out to the members, past and present, of the Davies lab
that have made the last 2 years of my life both enjoyable and memorable: Rachel Hanna,
Ravikiran Ravulapalli, Dr. Andy Scotter, Dr. Laurie Graham, Adam Middleton, Chris Garnham,
Nelson Lin, Sally Yu, Kyra Nabeta, Jin Qian, Dr. Yee-Foong Mok, Jordan Chou and Kristin Low.
I would like to extend special acknowledgement to:
• The technical expertise and patience from Sherry Gauthier that helped facilitate the transition
to my graduate studies.
• The early guidance of Dr. Dominic Cuerrier in my research project and additional assistance
even after pursuing his career.
v
• The patience of Dr. Rob Campbell in teaching me crystallography and editing my work.
Lastly, I would like to acknowledge Dr. Darrell Goll, the ‘father of calpain’, who passed away
mid July of this year. I had the privilege of meeting Dr. Goll at a calpain conference and know
that his enthusiasm and dedication to scientific research will live on through his written and
spoken words.
vi
Table of Contents
Abstract............................................................................................................................................ii
Acknowledgements......................................................................................................................... iv
Table of Contents............................................................................................................................vi
List of Figures................................................................................................................................. ix
List of Tables ................................................................................................................................... x
List of Abbreviations ......................................................................................................................xi
Chapter 1 General Introduction ....................................................................................................... 1
1.1 Calcium Signaling.................................................................................................................. 1
1.2 Calpains ................................................................................................................................. 2
1.3 Calpain Family in Disease ..................................................................................................... 2
1.4 Calpain Structure ................................................................................................................... 5
1.5 The Mechanism of Activation of Calpain by Calcium .......................................................... 6
1.6 Tissue-Specific Calpains...................................................................................................... 10
1.7 Active-site Specificity and Substrate/Inhibitor Design........................................................ 12
1.8 Active-Site Nomenclature.................................................................................................... 13
1.9 Autolysis and Aggregation .................................................................................................. 14
1.10 Calpastatin ......................................................................................................................... 15
Chapter 2 Relative Activity of Calpains 3, 8, 9 and 15 and Inhibitor Profiling of Calpain 8........ 18
2.1 Preface ................................................................................................................................. 18
2.2 Abstract ................................................................................................................................ 18
2.3 Introduction.......................................................................................................................... 19
2.4 Materials and Methods......................................................................................................... 20
2.4.1 Materials ....................................................................................................................... 21
2.4.2 Mutagenesis .................................................................................................................. 21
2.4.3 Relative Activity Measurements on Protease Cores ..................................................... 21
2.4.4 Inhibitor Profiling ......................................................................................................... 22
2.5 Results and Discussion ........................................................................................................ 22
2.5.1 Mini-calpains 3, 8 and 9 are Less Active than µI-II ..................................................... 23
2.5.2 Mutation of Mini-calpain 8 Increases its Activity ........................................................ 25
2.5.3 Mini-calpain 8 has Preferences for Leu, Ile and Val at the S2 Subsite......................... 25
vii
Chapter 3 Profiling of calpain activity with a series of FRET-based Substrates ........................... 33
3.1 Preface ................................................................................................................................. 33
3.2 Abstract ................................................................................................................................ 34
3.3 Introduction.......................................................................................................................... 34
3.4 Materials and Methods......................................................................................................... 36
3.4.1 Materials ....................................................................................................................... 36
3.4.2 Determination of the Cleavage Site within the PLFAER Substrate.............................. 37
3.4.3 Effect of Individual Residue Substitutions on Cleavage Kinetics using µI-II, µ- and m-
calpain.................................................................................................................................... 37
3.5 Results and Discussion ........................................................................................................ 39
3.5.1 Determination of the Substrate Cleavage Site .............................................................. 39
3.5.2 Efficacy of Ala and Met at the P1΄ Position ................................................................. 42
3.5.3 Effect of Individual Residue Substitutions on Cleavage Kinetics ................................ 42
Chapter 4 The Stabilization of Calcium-bound Calpain by Calpastatin Peptides.......................... 48
4.1 Preface ................................................................................................................................. 48
4.2 Abstract ................................................................................................................................ 49
4.3 Introduction.......................................................................................................................... 50
4.4 Materials and Methods......................................................................................................... 53
4.4.1 CAST C-A Peptide Cloning.......................................................................................... 53
4.4.2 Protein Expression and Purification.............................................................................. 54
4.4.3 CAST C-A Peptide Amino Acid Analysis.................................................................... 54
4.4.4 Monitoring Aggregation of mC105S by Turbidity ....................................................... 55
4.4.5 The Effect of CAST C-A Peptides on m-Calpain Activity........................................... 55
4.4.6 The Effect of CAST C-A Peptides on m-Calpain Autolysis......................................... 55
4.5 Results.................................................................................................................................. 56
4.5.1 Purification and Validation of the C-A Peptides........................................................... 56
4.5.2 CAST C-A Peptides Inhibit the Ca2+-Induced Insoluble Aggregation of mC105S ...... 58
4.5.3 CAST C-A Peptides Stabilize m-Calpain in the Presence of Ca2+ while Keeping the
Active Site Clear .................................................................................................................... 62
4.6 Discussion............................................................................................................................ 65
Chapter 5 General Discussion........................................................................................................ 69
5.1 Is This the End of the Road for Mini-calpains? ................................................................... 69
viii
5.2 Full-Length Calpains, the Way of the Future ...................................................................... 71
5.3 Applications of Allosteric Inhibitor Design......................................................................... 74
5.4 Summary and Conclusions .................................................................................................. 76
ix
List of Figures
Figure 1.1. Mammalian calpain gene family. .................................................................................. 3
Figure 1.2. The catalytic triad of papain, Ca2+-free m-calpain and µI-II. ........................................ 7
Figure 1.3. Calcium-induced conformational change of m-calpain................................................. 9
Figure 1.4. Domain structure of CAST and its interaction with calpain........................................ 16
Figure 2.1. Inactive m-calpain (mC105S) hydrolysis activity assay of mini-calpains 1 (A), 8 (B),
3 (C), 9 (D) and 15 (E)......................................................................................................... 24
Figure 2.2. The effect of the stability of α7-helix on Trp106 (purple) in mI-II (A), µI-II (B), m-
calpain (C) and mini-calpain 8 (D).. .................................................................................... 26
Figure 2.3. Autolysis assay of mini-calpain 8................................................................................ 27
Figure 2.4. Structure of the peptidomimetic epoxide library.. ....................................................... 29
Figure 2.5. SDS-PAGE of mC105S hydrolysis by mini-calpain 8. ............................................... 30
Figure 2.6. Profiling the S2 subsite of mini-calpain 8 with the natural amino acids. .................... 31
Figure 2.7. Profiling the S2 subsite of mini-calpain 8 with the non-natural amino acids.............. 32
Figure 3.1. NanoESI-MS spectrum of the cleavage products from the hydrolysis of (EDANS)-
EPLFMERK-(DABCYL) by µI-II....................................................................................... 40
Figure 3.2. Relative initial rates of the hydrolysis of the substituted FRET substrate series......... 45
Figure 4.1. The proposed mechanism of calpain stabilization using CAST.. ................................ 52
Figure 4.2. Purification of CAST 1C-2A....................................................................................... 57
Figure 4.3. Monitoring calpain aggregation by turbidity............................................................... 60
Figure 4.4. Monitoring the effect of the CAST C-A peptides on m-calpain activity and inhibition
as measured with a FRET-based assay.. .............................................................................. 63
Figure 4.5. Autolysis of m-calpain in the presence of the CAST C-A peptides monitored by SDS-
PAGE. .................................................................................................................................. 64
x
List of Tables
Table 3.1. Potential and observed FRET substrate hydrolysis products........................................ 41
Table 3.2. Preference for each amino acid at P1'-P3' positions of the hexapeptide substrate. ...... 43
Table 4.1. Validation of the CAST C-A amino acid analysis. ...................................................... 59
xi
List of Abbreviations
AD, Alzheimer’s disease
βME, 2-mercaptoethanol;
CAST, calpastatin;
COP, coatomer protein;
DABCYL, 4-((4-(dimethylamino)phenyl)azo)benzoic acid;
DMSO, dimethyl sulfoxide;
DTT, dithiothreitol;
E-64, trans-epoxysuccinyl-L-leucylamido (4-guanidino)butane;
EDANS, 5-[(2-aminoethyl) amino]naphthalene-1-sulfonic acid;
EDTA, ethylenediaminetetraacetic acid;
FRET, fluorescent resonance energy transfer;
HEPES, 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid;
HPLC, high performance liquid chromatography;
IFE, inner filter effect;
LGMD2A, limb girdle muscular dystrophy type 2A;
MES, 2-(N-morpholino)ethanesulfonic acid;
MOPS, 3-(N-morpholino)propanesulfonic acid;
MS, mass spectrometry;
PEF, penta-EF-hand;
SDS, sodium docecyl sulfate;
SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis;
Tris, tris(hydroxymethyl) aminomethane;
1
Chapter 1
General Introduction
1.1 Calcium Signaling
The role of calcium in cell signaling was serendipitously discovered over a
century ago when Sidney Ringer was examining the contraction of isolated rat hearts [1].
The hearts would contract well when they were suspended in a saline medium in which
he used tap water. However, he realized that when distilled water was substituted, the
heartbeats became very weak and shortly thereafter ceased. He determined that it was the
calcium in the tap water that was important in contraction of the heart muscle.
Calcium is now recognized as the universal information carrier through the entire
life cycle of a cell and its homeostasis is imperative to normal cell function, as both
calcium depletion and overload leads to cell death via apoptosis or necrosis [2,3,4,5].
Another important feature of calcium is its self-regulation, as it is able to control its own
influx, thereby effecting downstream events [6,7]. Calcium has also been shown to act as
a primary messenger through activation of extracellular calcium-sensing receptors in the
parathyroid and thyroid glands, kidney, osteoblasts, gastrointestinal mucosa, and recently
in the nervous system [8].
The most important and best-characterized calcium-binding proteins are the EF-
hand family of proteins. These proteins bind calcium, inducing a conformational change
which then relays the signal to other target proteins [9]. A classic example of this family
2
is calmodulin. Some of the calpains, the focus of this thesis, have penta-EF-hand (PEF)
domains, in addition to the protease core with two non-EF-hand calcium-binding sites.
These two sites act in concert with the PEF domains to cause a conformational change in
calpain that leads to proteolytic cleavage of target substrates.
1.2 Calpains
The name calpain stems from calcium ‘cal’, for its calcium binding characteristic,
and papain ‘pain’, the classic, much-studied plant thiol protease [10]. Although calpain
was named after papain, it actually has very low sequence identity to it and other cysteine
proteases, and therefore has been distinguished as a CLAN CA, family C2, class of
cysteine protease [11]. Calpain 1 (a.k.a. µ-calpain) was first discovered in rat brain in
1964 by Guroff [12], followed by the purification and partial characterization of calpain 2
(a.k.a. m-calpain) in 1976 [13]. Since that time, a total of 14 isoforms of calpain have
been discovered in mammals (Fig. 1.1). The ubiquitous µ- and m-calpains are known to
form heterodimers with the small subunit, calpain 4, and are named according to their
calcium requirement (3-50 µM for µ-calpain and 400-800 µM for m-calpain) [14].
1.3 Calpain Family in Disease
Calpains are involved in many cellular processes including, but not limited to,
cytoskeletal remodeling, cell motility [15], cell cycle progression [16], as well as
apoptosis and necrosis [17]. Knockout studies have revealed the importance of calpain in
3
1, 2, 8, 9, 11,12,13,14
Small Subunit
3
5, 6
7
10
15
Atypical
100
1, 2, 8, 9, 11,12,13,14
Small Subunit
3
5, 6
7
10
15
Atypical
1, 2, 8, 9, 11,12,13,14
Small Subunit
3
5, 6
7
10
15
Atypical
100
Figure 1.1. Mammalian calpain gene family. The typical calpains 1, 2, 3, 8, 9, 11, 12, 13 and
14 are composed of an N-terminal variable region (red), protease core domains I (blue) and II
(cyan), a C2-like domain III (green) and PEF domain IV (yellow). Calpains 1 and 2 form a
heterodimer with the small subunit via the PEF domains IV and VI (orange). Calpains 5, 6, 7, 10
and 15 have domain deletions and additions, classifying them as atypical. Tissue-specific
isoforms (3, 8, 9, 11 and 12) are shown in bold. The length of 100 amino acids is shown by the
bar below DI of the typical calpains.
4
mice, as the absence of m-calpain [18] and the small subunit [19,20] are embryonically
lethal. Surprisingly, µ-calpain knockout mice have normal long-term potentiation [21],
but do lack proper platelet function [22]. The fact that µ-calpain knockouts are not
embryonically lethal has lead to the hypothesis that µ- and m-calpains are functionally
redundant and m-calpain may take the place of µ-calpain. However, more recently,
calpain 1 has been localized to the intermembrane space of the mitochondria [23,24].
Thus, the functions of calpains 1 and 2 might be distinct because of compartmentation,
with calpain 1 being principally involved in apoptosis. Calpain 3 knockout mice have
been reported to have abnormal sarcomere formation [25] and calpain 10 knockout mice
are viable but have not yet been characterized [26].
Calpains have been implicated in many pathological conditions, where calcium
homeostasis is disrupted, leading to the mis-regulation of calpain. Chronic
neurodegenerative disorders including Alzheimer’s [27,28], Parkinson’s [29] and
Huntington’s [30] as well as ischemic brain, heart and kidney injuries [31] and traumatic
brain and spinal cord injury are such conditions where calpain activity is up-regulated by
a sustained increase in intracellular calcium. This results in both specific and non-specific
cleavage of proteins involved in these processes.
Although Alzheimer’s disease is far from completely understood, two hallmark
events have been shown to occur, the formation of amyloid plaques and neurofibrillary
tangles [32]. The most well-known calpain-specific event is the truncation of p35, the
regulatory subunit and activator of the serine/threonine kinase Cdk5, by calpain to p25
5
[33]. p25 is more resistant to degradation than p35 and therefore accumulates in the
brains of patients with AD and constitutively activates and mislocalizes Cdk5 [34]. The
p25/Cdk5 complex has also been shown to display increased and altered tau
phosphorylation, leading to reduced association of tau with microtubules and the
formation of neurofibrillary tangles [34].
Since the implications of calpain in disease are extremely diverse and complex, it
is difficult to determine whether its role is causal or coincidental. One disease where
calpain is known to have a causal affect is limb girdle muscular dystrophy type 2A
(LGMD2A), where mutations in the skeletal muscle-specific calpain 3 are known to
cause this muscle wasting disease. Although our understanding of the roles of calpain in
disease has come a long way in the past 40 years, a tremendous amount remains to be
determined.
1.4 Calpain Structure
The domain structure of the mammalian calpain gene family is shown in Fig. 1.1.
The ubiquitous µ- and m-calpains have distinct large subunits, but share a common small
subunit. The large subunits are composed of an N-terminal anchor helix that interacts
with domain VI (DVI) of the small subunit in the absence of calcium (Fig.1.2A). This
helix is amphipathic in µ-calpain and has been found to act as a mitochondrial targeting
sequence [35]. This sequence is followed by the protease core, domains I and II, where
the catalytic triad (Cys105, His262 and Asn286 in m-calpain) resides at the interface
6
[36]. DIII is a C2-like domain with an 8-stranded anti-parallel β-sandwich [37,38].
Following DIII is the PEF domain IV which has 5 EF-hand motifs, 4 of which bind
calcium and the 5th forms the heterodimeric interface with the small subunit via pairing
with the 5th EF-hand of DVI [37,39,40]. The small subunit shared by µ- and m-calpains is
composed of two domains; a flexible, Gly-rich DV of unknown function and a PEF DVI,
similar to DIV of µ- and m-calpains.
With respect to protein structure, the calpain family can be classified as either
typical or atypical (Fig. 1.1). The typical calpains (3, 8, 9, 11, 12, 13 and 14) have a
domain structure similar to that of µ- and m-calpains, and are classified by the presence
of a PEF domain IV. Although they had a PEF DIV, they do not necessarily
heterodimerize with the small subunit. Calpain 3, for example, appears to form a
homodimer of the large subunit, again through the 5th EF-hand of DIV [41]. On the other
hand, the atypical calpains (5, 6, 7, 10 and 15) are classified based on domain additions or
deletions to the typical subfamily, specifically, the deletion of the PEF domain IV (Fig.
1.1).
1.5 The Mechanism of Activation of Calpain by Calcium
In papain, the active-site Cys and His are within 3.7 Å of each other, a distance
that allows the His to deprotonate the Cys-SH, increasing its nucleophilicity (Fig. 1.2A).
It was shown clearly from the crystallographic structure of m-calpain [37,42] that this
equivalent inter-atomic distance in the calcium-free form is ~10.5 Å (Fig. 1.2B). This
7
A
B C
10.64 Å
His262
Ser105
Asn286
3.72 Å
Cys115
His272
Asn296
3.79 Å
Cys25
His159
Asp158
A
B C
10.64 Å
His262
Ser105
Asn286
3.72 Å
Cys115
His272
Asn296
3.79 Å
Cys25
His159
Asp158
Figure 1.2. The catalytic triad of papain, Ca2+
-free m-calpain and µI-II. (A) The active site
Cys25-S of papain is 3.79 Å away from His159-Nδ, demonstrating an aligned active site. (B) The
unaligned active site of m-calpain. In the absence of calcium, Ser105-O is 10.64 Å away from
His262-Nδ. (C) Calcium binding to calpain (µI-II shown here) causes a conformational change
that aligns the active site by bringing Cys115-S within 3.72 Å of His272-Nδ. The protein
backbone is coloured as shown in Fig. 1.1 [43].
8
distance is much too far for acid-base catalysis to occur between these residues and
provides an explanation for the inactivity of calpain in the absence of calcium [37].
Therefore, it was proposed that the major structural change upon calcium binding must be
the alignment of the active site. Subsequently, the protease core of µ-calpain (µI-II) was
shown to bind two calcium ions, one in each of DI and DII [44]. These bind
cooperatively to peptide loops, which in turn cause conformational changes to the
protease core that align Cys115 and His272 of the catalytic triad within 3.7 Å of each
other, as seen in papain (Fig. 1.2C). Calpain was thought to bind one or more calcium
ions to acidic loops on DIII [45] but this was not seen in the recent structure of calcium-
bound inactive m-calpain [46]. Up to four calcium ions bind to PEF domain VI in the
small subunit [47,48], and by inference the same number were thought to bind to the
homologous PEF domain DIV in the large subunit [42,44]. Conformational changes in
DVI of the small subunit are very slight, with an 18° change in the angle between the first
two α-helices of the first EF-hand (EF-1) exposing a hydrophobic surface [47,48].
Recent structural data from the co-crystallization of CAST and inactive calcium-
bound m-calpain have supported these prior claims (Fig. 1.3B) [46]. The predominant
structural changes upon calcium binding do not occur within each domain, but in the
relative positions of the domains to each other, resulting in a more compact protein. The
changes in DI and DII are exactly the same as shown in the structure of µI-II, to align the
active site [44]. In addition, the two PEF domains (IV and VI) are shifted up towards the
active site, which displaces the N-terminal anchor helix [46]. As mentioned above, there
9
+ Ca2+
- Ca2+
+ Ca2+
- Ca2+
Figure 1.3. Calcium-induced conformational change of m-calpain. (A) In the absence of
calcium, the N-terminal anchor helix interacts with DVI of the small subunit, and calpain is in an
extended conformation. (B) Calpain binds two calcium ions in the protease core, one per domain,
as well as four in each of the PEF domains DIV and DVI. Upon calcium binding, the N-terminal
anchor helix is released from DVI, there is a rotation of DI relative to DII, and DIV and DVI are
pushed closer to the protease core, resulting in an overall compaction of the enzyme. This is a
dynamic process, as the addition of excess EDTA pushes this reaction back to the Ca2+-free form.
The protein backbone is coloured as shown in Fig. 1.1. Calcium ions are illustrated as grey
spheres [37,46].
10
is exposure of hydrophobic pockets in DIV and DVI on binding calcium. These provide
binding sites for the amphipathic helices in the conserved A and C sub-domains of CAST
while the conserved B sub-domain binds to the newly formed catalytic cleft. These
structural data have given scientists in the calpain field answers that have been long
sought after, the conformational changes that occur in the entire enzyme upon calcium
binding, as well as the mode of interaction and inhibition of calpain by its specific
inhibitor, CAST.
1.6 Tissue-Specific Calpains
The calpain family members can also be classified based on their expression, as
either ubiquitous or tissue-specific. For example, the tissue-specific calpains 3, 8, 9, 11
and 12 are predominantly expressed in skeletal muscle, stomach, digestive tract, testes
and hair follicles, respectively [49].
Calpain 3 (a.k.a. p94) has a similar domain structure to µ- and m-calpains (54 and
51% sequence identity respectively) [50], with the exception of two unique insertion
sequences, IS1, in DII, and IS2, situated between DIII and DIV. It is extremely unstable
when recombinantly expressed and undergoes rapid autolysis, which is avoided when it is
expressed with the deletion of IS1 and IS2 [51]. This instability has also been shown
during isolation of calpain 3 from skeletal muscle, while it remains stable in intact
muscle. This property has been attributed to its interaction with connectin through IS2
[52]. As mentioned in section 1.3, mutations in the calpain 3 gene are known to cause
limb girdle muscular dystrophy type 2A (LGMD2A) [53,54]. These mutations result in
11
calpain 3 deficiency that eventually leads to LGMD2A, although the exact mechanism by
which this occurs is not known [55].
The stomach-specific calpain 8 (a.k.a. nCL-2) has been shown to be localized in
pit cells of the stomach as well as goblet cells in the upper half of the villi in the
duodenum [56]. A novel function for calpain 8 has been identified in membrane
trafficking in mucus cells through its interaction with, and proteolysis of, β-COP of the
COP1 coatomer complex [56]. The digestive tract-specific calpain 9 (a.k.a. nCL-4) is also
located in goblet cells, although in the lower half of the villi [56]. Unlike most calpains,
calpain 9 is down-regulated in human gastric cancer tissue [56,57]. Consistent with this
result, mouse NIH3T3 fibroblast cells with antisense-induced calpain 9 deficiency,
underwent cellular transformation and tumorigenesis [58].
The function and pathological roles of the tissue-specific calpains 11 and 12
remain elusive. Calpain 11 is expressed predominantly in the testis and has been
postulated to be involved in the regulation of signal transduction events of cytoskeletal
remodeling during meiosis, spermiogenesis and sperm function, based on its tissue
location and temporal distribution [59]. Calpain 12 is highly expressed in the cortex of
the hair follicle, but its function remains to be determined [60]. Due to the tissue-
specificity of these calpains, they are promising targets for gene disruption and the
elucidation of their function may be pharmacologically significant.
12
1.7 Active-site Specificity and Substrate/Inhibitor Design
The development of calpain- and isoform-specific inhibitors is not only important
for potential pharmacological treatments, but also for a greater understanding of the
physiological roles of the calpain family members. The main problem associated with the
development of calpain inhibitors is the cross-reactivity with other cysteine proteases,
which share some common features in their active sites. The only known specific
inhibitor of calpain is CAST, which interacts with heterodimeric calpains through the
PEF domains IV and VI as well as the catalytic active-site cleft [46]. Although CAST has
an extremely specific interaction with calpain, it is postulated only to inhibit the
heterodimeric calpains, and then only after they have been activated by calcium. Even the
individual inhibitory domains of CAST are too large at 10-15k Da to introduce into the
cell as drugs and as intrinsically unstructured proteins they are extremely susceptible to
proteolysis.
An obvious target for inhibitor development is the active site. Small
peptidomimetic molecules could potentially be developed that take advantage of small
differences in preferences between each calpain family member. The crystal structure of
µI-II has provided a means for the development of specific active-site inhibitors, as it is
stable in the presence of calcium, thereby capturing an aligned catalytic triad [44].
Structures of µI-II in complex with inhibitors have suggested ways in which more
specific calpain inhibitors can be produced [61]. The two archetypal cysteine protease
inhibitors, E-64 and leupeptin, were shown to interact with only one side (unprimed) of
13
the active-site cleft of µI-II [43] and further studies revealed a way to design inhibitors
that interact with both sides of the active site to provide increased specificity [62].
Calpains cleave extended regions in their target substrates but do not have
absolute sequence specificity, unlike proteases such as trypsin or chymotrypsin.
Nevertheless, calpains have preferences at each position of the active-site [63]. A library-
based approach has previously been used to determine these preferences at positions
flanking both sides of the active site to produce optimal substrate and inhibitor sequences
for µI-II [63,64,65]. This method can be extended to other calpain isoforms, in order to
better understand their physiological function and develop isoform-specific therapeutic
agents.
1.8 Active-Site Nomenclature
The catalytic cleft of calpain contains an active cysteine, which undergoes
covalent catalysis with its target substrates and, in some cases, its inhibitors. Flanking the
cysteine are sites on both sides of the cleft, which can accommodate neighbouring
residues of the substrate. For the purposes of clarity and consistency, Schechter and
Berger proposed a nomenclature for the subsites of the catalytic cleft of proteases (S) as
well as the positions of the sequences that interact with it (P) [66]. In the case of cysteine
proteases, the region that is N-terminal to the active-site cysteine is designated as the
unprimed side of the active site, with each subsite labelled S1-S3. The primed side of the
active-site is the region that is C-terminal to the cysteine, and aptly named the S1’-S3’
14
subsites. Similarly, the amino acid residues of target substrates have been labelled
according to the subsite they interact with, P3-P3’. This nomenclature will be used
throughout the thesis.
1.9 Autolysis and Aggregation
As previously mentioned in section 1.5, calpain binds calcium in the protease core
as well as domains IV and VI, thereby activating the enzyme through a structural change.
Upon calcium binding, two hallmark events occur that make calpain difficult to work
with: autolysis and aggregation.
The events of autolytic cleavage of calpain are well documented and begin with
the release of the N-terminal anchor helix, effectively lowering the calcium requirement
of m-calpain from 400-800 µM to 50-150 µM [67]. The calcium requirement of µ-calpain
is not lowered until the autoproteolytic release of DV of the small subunit, when it is
decreased from 3-50 µM to 0.5-2.0 µM [68]. Further autolysis occurs in flexible loops
between DII and DIII, which releases a calcium-dependent weakly active protease core,
as well as between DIII and DIV.
Calpain is unstable in the presence of calcium, even when it is inactivated by the
mutation of the active-site cysteine to a serine. Previous structural studies have shown
that calcium binding causes a conformational change in which hydrophobic patches are
unveiled in domains IV and VI, providing possible sites for aggregation [69]. The
mechanism by which aggregation occurs is very poorly understood and more insight
15
would be advantageous for crystallographic studies, as well as inhibitor and substrate
specificity. This instability of calpain is known to be prevented by its endogenous
inhibitor, CAST [70].
1.10 Calpastatin
Calpastatin (CAST) was first discovered in the mid 1970s when an unexpected
lack of calpain activity was detected in porcine skeletal muscle [13]. This activity was not
restored until calpain was isolated out of solution by precipitation [71]. The reason for
this inactivity was the presence of a trypsin-labile, heat-resistant factor [71], later named
CAST by Takashi Murachi in 1979 (see ref. [72]). It is a multidomain [73,74,75,76],
unstructured protein [77] and the only known natural endogenous inhibitor that is specific
for calpain [73].
CAST has at least 9 different isoforms due to the use of different promoters
[78,79,80] as well as alternative splicing mechanisms [81,82]. They are each unstructured
in aqueous solution and are typically composed of four homologous inhibitory domains
(I-IV) and an N-terminal L (or XL in the case of one isoform) domain that has no
inhibitory activity (Fig. 1.4A) [75]. It was later determined that in each domain, there are
three conserved sub-domains: A, B and C [83,84]. Sub-domains A and C are known to
form amphipathic α-helices upon the recognition of calpain in its calcium-bound form
and interact with DIV of calpain and DVI of the small subunit, respectively [46,85,86].
This interaction is facilitated by the calcium-induced conformational change of calpain
16
1 2 3 4L
A
A
B
CA
B
CA
B
C
B
A C
B
A C
B
A C
B
A C
B
A CB A CB A CB A CB
190 191 323 324 435 436 572 573 713
1 2 3 4L
A
A
B
CA
B
CA
B
C
B
A C
B
A C
B
A C
B
A CA
B
CA
B
CA
B
C
B
A C
B
A C
B
A C
B
A C
B
A C
B
A C
B
A CB A CB A CB A CBA CB A CB A CB A CB
190 191 323 324 435 436 572 573 713
Figure 1.4. Domain structure of CAST and its interaction with calpain. (A) Rat CAST is
composed of an N-terminal L domain, followed by four inhibitory domains that each have A
(turquoise), B (orange) and C (yellow) subdomains. Residue numbers marking the beginning of
the A subdomain of each inhibitory domain are written below the bar diagram. (B) Graphical
representation of CAST binding to four calpain molecules. The A (turquoise) and C (yellow)
subdomains interact with DIV and DVI, respectively, and the B (orange) subdomain interacts
with the calpain active site [87]. Calpain domains are coloured as in Fig. 1.1.
17
and resulting open conformation of EF-1 of both domains, exposing hydrophobic
binding pockets for sub-domains A and C [46,69]. Domains I-IV were discovered to each
bind calpain simultaneously, thereby providing inhibitory activity for up to four calpain
molecules at once (Fig. 1.4B) [88,89]. There was some controversy over where the B
sub-domain interacted [44,69,90], but with recent structural findings it was shown to be
located in the active-site of calpain, where it exerts its inhibitory activity [46].
Since CAST interacts with the PEF domains in both the large and small subunits,
it is expected to inhibit only those calpains that, first have a PEF domain IV, and
secondly form a heterodimer with the small subunit. Currently, µ- and m-calpains are the
only known typical isoforms that are inhibited by CAST. Since information is lacking on
the other typical members of the calpain family as to whether or not they homo- or
heterodimerize, it remains to be determined which of these would be expected to bind
CAST. According to this classification, the atypical calpains (5, 6, 7, 10 and 15) would
not interact with or be inhibited by CAST, as they lack a PEF domain IV (Fig. 1.1).
The two largest obstacles in working with calpain have been its aggregation and
autoproteolytic properties, as outlined in section 1.9. CAST seems to solve this problem
by trapping calpain in a stable calcium-bound state, making it an excellent candidate to
stabilize active, full-length calpain for in vitro experiments such as crystallography,
inhibitor and substrate profiling and whole-cell screening. In order to do this, CAST
would have to be re-routed out of the active site, to preserve calpain activity, which will
be the focus of chapter 4 of this thesis.
18
Chapter 2
Relative Activity of Calpains 3, 8, 9 and 15 and Inhibitor Profiling of
Calpain 8
2.1 Preface
This chapter will be combined with additional information on the structure of mini-
calpain 8 solved by the Structural Genomics Consortium in Toronto, Ontario.
Jacqueline Kelly was responsible for the mutagenesis of mini-calpain 8, activity assays of
the mini-calpains as well as the profiling of the S2 position of mini-calpain 8. The
experiments were performed under the supervision and guidance of Dr. Peter L. Davies
with insight from Dr. Dominic Cuerrier. The protease cores were expressed and purified
by Dr. Tara Davis at the Structural Genomics Consortium in Toronto, Ontario.
2.2 Abstract
Calpains are calcium-dependent proteases that carry out limited proteolysis on
many substrates in response to increases in intracellular calcium levels. Since various
calpain isoforms are involved in pathological conditions such as Alzheimer`s disease,
Type 2 diabetes, cancer, and LGMD2A, the development of specific small-molecule
inhibitors is of pharmacological significance. The protease core of calpain 1 (µI-II) has
been used to screen substrate and inhibitor libraries without the complications of
autoproteolysis. Here we have explored the extension of this approach to a number of
19
tissue-specific calpains. The relative activities of several calpain cores were compared by
looking at the time course for their digestion of inactivated m-calpain. One of the more
active of these cores was that of stomach-specific calpain 8. Its inhibitor specificity was
profiled with a synthetic peptidomimetic library of epoxide-based compounds that target
the active site. The library probed the P2 position with both (R,R) and (S,S)
stereochemistry of natural and non-natural amino acids. Calpain inhibition was analyzed
by the extent of proteolysis of inactive m-calpain on an SDS-PAGE gel by the protease
core of calpain 8. The stomach-specific calpain had similar inhibitor specificity at the P2
position as µI-II, particularly with the natural amino acids. However, the observed amino
acid preferences for calpain 8 were weaker than for µ-calpain, which might reflect the
overall lower activity of the calpain 8 core.
2.3 Introduction
Calpains are a family of cysteine proteases that are activated by an increase in
intracellular calcium levels. The human genome consists of 14 calpain isoforms that can
be classified according to their expression, either ubiquitous (calpains 1, 2, 5, 7, 10, 13,
14 and 15) or tissue-specific (calpains 3, 8, 9, 11 and 12) [14,49,91]. Although the
ubiquitous calpains 1 and 2 (aka. µ- and m-calpains, respectively) have been well
characterized to date, much is to be learned of the remaining calpain isoforms. It is
known that calpain 3 is expressed in skeletal muscle and is mutated in LGMD2A [53,54]
and calpains 8 and 9 are predominantly localized to the stomach [92] and digestive tract
[93], respectively. It has been proposed that, in the presence of calcium, calpain 8 cleaves
20
βCOP in pit cells and plays an important role in membrane trafficking of mucus cells
[56,92]. Also, calpain 9 has been demonstrated to be down-regulated in gastric cancer
[57,94] and has a role in the suppression of tumorigenesis [57,58,94].
The calpain gene family can also be broken down into typical, those with a
similar domain structure to µ- and m-calpains, or atypical. Calpain 15 is classified as
atypical with a distinctive domain structure and undetermined function and protease
activity [49]. It is composed of N-terminal zinc finger repeats, followed by a conserved
protease core domain (DI-II) and a C-terminal SOL homology domain [95].
To help better understand the physiological roles of these enzymes and to aid in
potential drug discovery, it would be useful to have isoform-specific small-molecule
inhibitors. In an attempt to develop such inhibitors, the active-site specificity must be
characterized. This has previously been done for the protease cores of µ-calpain (µI-II)
and a G203A mutant of the m-calpain protease core (mI-II G203A), where a
peptidomimetic epoxide library of inhibitors was used to profile their active-site
preferences [64]. The use of mini-calpains in biochemical characterization offers many
advantages, including their ease of expression, resistance to autoproteolysis, and simpler
crystallization compared to the full-length calpains. Here we have examined the protease
cores of several calpain isoforms for their relative activity and profiled the active site of
the tissue-specific calpain 8 using an epoxide library of inhibitors.
2.4 Materials and Methods
21
2.4.1 Materials
Inactive rat m-calpain (mC105S) and the protease core of rat µ-calpain (µI-II)
were expressed in E. coli and purified as previously described [37,44]. The triple-wide
gel electrophoresis apparatus was obtained from C.B.S. Scientific (Solana Beach, CA).
The positional-scanning epoxide libraries were synthesized as described elsewhere [96],
and used from 50 mM stocks in 100 % DMSO. The structures and names of the non-
natural amino acids used are provided in the supplementary table to Cuerrier et al. 2007
[64]. The protease cores of calpains 3, 8, 9 and 15 were provided by the Structural
Genomics Consortium (Toronto, ON).
2.4.2 Mutagenesis
Site-directed mutagenesis using the QuikChange kit (Stratagene, CA) was
performed on mini-calpain 8 to mutate Gly203 to Ala. The primers used for the mutation
were 5' CCACAGTGGAGGCCTTTGAGGATTTCAC 3' (sense) and 5'
GTGAAATCCTCAAAGGCCTCCACTGTGG 3' (anti-sense) with the Ala mutation
(underlined in bold) that introduced a StuI cut site AGG/CCT for screening clones.
2.4.3 Relative Activity Measurements on Protease Cores
A solution containing 1 mg/mL of inactive m-calpain (mC105S) and 1 mg/mL of
calpain protease core (either calpain 3, 8 wild type, 9 or 15) was made in reducing buffer
(50 mM HEPES-NaOH pH 7.8 and 10 mM DTT). Equal amounts of this solution and
reducing buffer containing 100 mM CaCl2 and 5 % DMSO were used to initiate
22
digestion. The reaction proceeded at 30 °C and was quenched with 100 mM EDTA and
3X SDS-PAGE sample buffer (150 mM Tris-HCl (pH 6.8), 300 mM DTT, 6 % SDS, 0.3
% bromophenol blue, and 30 % glycerol) at 0, 0.5, 1, 3.5, 5, 8.5 and 25 h after initiation.
A control reaction was run where equal amounts of reducing buffer were added in the
absence of calcium. The samples were run on an SDS-PAGE gel to compare the extent
and rate of digestion of the 80 kDa large subunit of mC105S.
2.4.4 Inhibitor Profiling
Digestions were set up as above with the addition of 0.24 mM of the library of
epoxide inhibitors with 19 natural and 42 non-natural amino acids at the P2 position with
(R,R) or (S,S) stereochemistry. The reaction was incubated at 30 °C for 25 h and
quenched with 100 mM EDTA and 3X SDS-PAGE sample buffer. A triple-wide gel
apparatus was used to run up to 65 samples on a 12 % polyacrylamide gel. Densitometry
analysis was done on the 80 kDa band using the free open source Image J software
(http://rsb.info.nih.gov/ij/). The background intensity above each band was subtracted
from every lane. The reactions were compared by assigning to each a preference equal to
its fold-over average inhibition, the average being 1.0.
2.5 Results and Discussion
23
2.5.1 Mini-calpains 3, 8 and 9 are Less Active than µI-II
Inactive m-calpain (mC105S) was used as a substrate for the protease cores to
determine the activity of mini-calpains 3, 8, 9 and 15 relative to µI-II (Fig. 2.1). For
mini-calpain 8, there are three protein bands present at zero time (Fig. 2.1B). Bands (i)
and (vi) are the large and small subunits, respectively, of the substrate, and band (iv) is
that of the mini-calpain enzyme. After 1 h, three digestion products appear. Band (ii),
just below the 80 kDa band, contains domains I-III. A domain I-II fragment of the
mC105S substrate appears just above that of mini-calpain (band iii). The other, just above
the 21 kDa small subunit band, is domain IV- the other cleavage product (band v). These
digestion products increase in intensity during the course of the digestion and do not
appear in the absence of calcium (control data shown in Fig. 2.3). The extent of digestion
of the substrate by mini-calpain 8 after 25 h is roughly half that seen after just 0.5 h when
µI-II is used as the enzyme (Fig. 2.1A). The digests produced by mini-calpains 3 and 9
are qualitatively the same as that produced by mini-calpain 8 but they proceed at about
one third of the rate seen with mini-calpain 8 (Fig. 2.1C and 2.1D, respectively). Indeed,
the domain I-II digestion product is barely visible even after 25 h of digestion. In this
assay, mini-calpain 15 was shown to be inactive over a 25 h period (Fig. 2.1E). The
activity of mini-calpain 15 has not yet been determined and it may be that this isoform
does not function as an enzyme, mC205S has no recognition sites for it, or that it requires
the whole protein for activity [49].
24
B0.5 1 3.50 5 8.5 250 250.5
A
(I)(II)
(III)(IV)
(V)
(VI)
0.5 1 3.50 5 8.5 25
C
0.5 1 3.50 5 8.5 25 0.5 1 3.50 5 8.5 25
D E
80K
38K
21K
B0.5 1 3.50 5 8.5 250 250.5
A
(I)(II)
(III)(IV)
(V)
(VI)
0.5 1 3.50 5 8.5 25
C
0.5 1 3.50 5 8.5 25 0.5 1 3.50 5 8.5 25
D E
80K
38K
21K
Figure 2.1. Inactive m-calpain (mC105S) hydrolysis activity assay of mini-calpains 1 (A), 8
(B), 3 (C), 9 (D) and 15 (E). This assay monitors the digestion of the 80 kDa large subunit of
mC105S (band i) by mini-calpain (band iv) into its cleavage products: Domains I-III (band ii), I-
II (band iii), and IV (band v). Band vi represents the small subunit. Each digestion was done at 30
°C in the presence of 50 mM CaCl2 and 0.5 mg/mL of each mini-calpain and mC105S. Sample
(10 µL) was added to 10 µL of SDS-PAGE loading buffer and loaded on a 12 % polyacrylamide
gel. The time of digestion is listed in hours above each lane and the calcium controls are shown in
Fig. 2.3.
25
2.5.2 Mutation of Mini-calpain 8 Increases its Activity
In the protease core of rat m-calpain, G203 destabilizes part of helix α7 in the absence of
the supporting domains III, IV and VI (small subunit) causing Trp 106 to swing into the
active site, resulting in a decrease in activity (Fig. 2.2A) [97]. In fact, its activity is only 1
% of µI-II, which has an Ala at position 203, and Trp116 pointed away from the active
site (Fig. 2.2B). The recent structure of calcium-bound full-length m-calpain displays the
stabilization of this helix by the presence of DIII, DIV, and DVI, even with a Gly at this
position (Fig. 2.2C). In an attempt to increase the activity of mini-calpain 8, the Gly at
position 203 in the conserved helix α7 was mutated to an Ala. The activity was increased
about two-fold, a much lower effect that was seen in mI-II (Fig. 2.3). A closer look at the
structure (PDB entry 2NQA) revealed that the Trp at residue 106 was not protruding into
the active site in the presence of the destabilized helix α7, as was seen with mI-II (Fig.
2.2D). This supports the smaller difference in activity upon mutation of G203 to an Ala
seen with mini-calpain 8 than with mI-II.
2.5.3 Mini-calpain 8 has Preferences for Leu, Ile and Val at the S2 Subsite
Of the tissue-specific mini-calpains, the core of calpain 8 was the only one with
sufficient activity to attempt inhibitor profiling. Specificity at the P2 site was determined
by profiling the digestion of mC105S by mini-calpain 8 with a peptidomimetic library of
compounds, containing a fixed amino acid at the P2 position and an equimolar mix of
natural residues at the P3 and P4 positions with both (R,R) and (S,S) stereochemistry at
26
A B
DC
A B
DC
Figure 2.2. The effect of the stability of α7-helix on Trp106 (purple) in mI-II (A), µI-II (B),
m-calpain (C) and mini-calpain 8 (D). Leupeptin is shown in the active site of all 4 structures,
Trp106 is coloured purple and position 203 is coloured yellow. (A) In the absence of supporting
DIII-DVI, G203 of mI-II destablizes α7-helix, causing Trp106 to swing into the S2 pocket of the
active site, decreasing its activity [97]. (B) Position 203 of µI-II is occupied by a helix-stabilizing
Ala and Trp116 is positioned away from the active site. (C) In full-length m-calpain, the presence
of DIII-DVI stabilize α7-helix even with G203, causing Trp106 to face away from the active site.
(D) The protease core of calpain 8 has a Gly at this conserved position that disrupts α7-helix but
does not cause Trp106 to swing directly in the active site, it is pushed downwards instead. The
PDB entries are 1mdw, 1tl9, 1df0 and 2nqa, for the x-ray crystallographic structures of mI-II, µI-
II, m-calpain and mini-calpain 8, respectively. The protein backbone is coloured as shown in Fig.
1.1.
27
0.5+ 0.5-
A B C
0 0.5 5 25 0 0.5 5 250.5+ 0.5-
A B C
0 0.5 5 25 0 0.5 5 25
Figure 2.3. Autolysis assay of mini-calpain 8. The time points are listed in hours and the
autolysis fragments are the same as shown in Fig. 2.1. (A) The control hydrolysis of mC105S (0.5
mg/mL) by µI-II (0.5 mg/mL) after 30 min with (+) and without (-) 50 mM CaCl2 at 30 °C. (B
and C) The hydrolysis from 0 to 25 h for 0.5 mg/mL wild type (WT) (B) and mutant (G203A)
mini-calpain 8 (C) at 30 °C. Sample (10 µL) was added to 10 µL of SDS-PAGE loading buffer
and loaded on a 12 % polyacrylamide gel
28
the epoxide warhead (Fig. 2.4). SDS-PAGE shows differential digestion of the mC105S
80 kDa band by mini-calpain 8 into its cleavage products, as described in section 2.4.1
(Fig. 2.5). Using the library with (S,S) stereochemistry, a preference was demonstrated at
the P2 site of the inhibitor for the branched aliphatic natural amino acids Leu, Ile and Val,
and also a slight preference for Trp. This preference was similar to that observed with µI-
II (Fig. 2.6). The library with (R,R) stereochemistry showed similar preferences as the
(S,S) library, although the difference is much smaller (Fig. 2.6). The non-natural amino
acid profiles showed similar preferences for numbers 2, 8, 16, 18, 19, 23, 27 and 42 for
the (S,S) library as previously determined for µI-II and mI-II [98], but failed to pull out
preferences for the (R,R) library (Fig. 2.7). This demonstrates that mini-calpain 8 is
calpain-like, as it exhibits similar preferences at the P2 site to both µI-II and mI-II. The
preferences were much lower for mini-calpain 8 than for µI-II and mI-II, which may
reflect its lower activity.
In summary, the relative activity of the protease cores of calpains 3, 8, 9 were
determined to be very low and statistical analysis was therefore not pursued. The low
activity of the isoforms as well as the difficulty in profiling mini-calpain 8 stresses the
need for the isolation of the full-length enzymes for inhibitor and substrate profiling. The
whole enzymes are expected to have a higher activity as well as provide additional
preferences at the unprimed side of the active site by the close association of domain III.
29
NH
NH
X
O
O
OO
O
NH
O
H N O Et
m ix
m ix
P2
P3
P4(S,S)
or (R,R)
NH
NH
X
O
O
OO
O
NH
O
H N O Et
m ix
m ix
P2
P3
P4(S,S)
or (R,R)
Figure 2.4. Structure of the peptidomimetic epoxide library. A fixed amino acid is at the P2
scanned position (X) and a mix of natural amino acids at the P3 and P4 sites. Natural amino acids
except methionine and cysteine, with the addition of norleucine as well as the non-natural amino
acids were scanned at the P2 position. The library contained both (S,S) and (R,R) stereochemistry
at the epoxide warhead to increase the potential for specificity [64].
30
PF
RHA
L NE
Q ST V
YG
(+)
(-)D I K W
N.L AL
DN
EP
FQ
GR
HS
I(+)
K
(-)T
YVN.L
WLep
11
4
13
51
9 10
3
12 14
(-) (+) lep6
15 16
2 7 8
2322 27
20
29
3017
25 26
19
28 38 (-)
21
39 40
18 31 32
24 (+)
3733 35
41 42
34 36
PF
RHA
L NE
Q ST V
YG
(+)
(-)D I K W
N.L AL
DN
EP
FQ
GR
HS
I(+)
K
(-)T
YVN.L
WLep
11
4
13
51
9 10
3
12 14
(-) (+) lep6
15 16
2 7 8
2322 27
20
29
3017
25 26
19
28 38 (-)
21
39 40
18 31 32
24 (+)
3733 35
41 42
34 36
Figure 2.5. SDS-PAGE of mC105S hydrolysis by mini-calpain 8. The hydrolysis of 0.5
mg/mL mC105S by 0.5 mg/mL mini-calpain 8 was done in the presence of a peptidomimetic
library of epoxide inhibitors (0.24 mM) probing the S2 subsite of mini-calpain 8 with natural
(top) and non-natural (bottom) amino acids with (R,R) (coloured red) and (S,S) stereochemistry
at the epoxide warhead [(R,R) stereochemistry not shown for the non-natural amino acid gel].
The bands are as described in Fig. 2.1. SDS-PAGE loading buffer (5 µL) was added to 5 µL
sample and loaded onto 12 % polyacrylamide gel. The gel was stained with Coomassie Blue. (+)
no inhibitor; (-) no calcium; N.L norleucine; Lep leupeptin.
31
0.0
0.5
1.0
1.5
2.0
2.5
A D E F G H I K L N P Q R S T V W Y N.L
Amino Acid
Res
idu
e P
refe
ren
ce
0.0
0.5
1.0
1.5
2.0
2.5
3.0
A D E F G H I K L N P Q R S T V W Y N.L
Amino Acid
Res
idu
e P
refe
ren
ce
A
B
0.0
0.5
1.0
1.5
2.0
2.5
A D E F G H I K L N P Q R S T V W Y N.L
Amino Acid
Res
idu
e P
refe
ren
ce
0.0
0.5
1.0
1.5
2.0
2.5
3.0
A D E F G H I K L N P Q R S T V W Y N.L
Amino Acid
Res
idu
e P
refe
ren
ce
A
B
Figure 2.6. Profiling the S2 subsite of mini-calpain 8 with the natural amino acids. Inhibitor
profiles of the P2 (R,R) (A) and (S,S) (B) natural amino acid library with mini-calpain 8 (top)
compared to µI-II (bottom left) and mI-II G203A (bottom right) [64]. Residue preference = fold-
over average inhibition where average is equal to 1.0 (black line). Error bars correspond to +/-
one standard deviation based on three separate experiments.
32
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Amino Acid
Res
idu
e P
refe
ren
ce
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Amino Acid
Res
idu
e P
refe
ren
ce
A
B
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Amino Acid
Res
idu
e P
refe
ren
ce
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Amino Acid
Res
idu
e P
refe
ren
ce
A
B
Figure 2.7. Profiling the S2 subsite of mini-calpain 8 with the non-natural amino acids.
Inhibitor profiles of the P2 (R,R) (A) and (S,S) (B) non-natural amino acid library with mini-
calpain 8 (top) compared to µI-II (bottom left) and mI-II G203A (bottom right) [64]. Residue
preference = fold-over average inhibition where average is equal to 1.0 (black line). Error bars
correspond to +/- one standard deviation based on three separate experiments.
33
Chapter 3
Profiling of calpain activity with a series of FRET-based Substrates
3.1 Preface
This chapter has been submitted to FEBS letters.
Kelly, J.C., Cuerrier, D., Campbell, R.L., and Davies, P. (2008). Profiling of calpain
activity with a series of FRET-based substrates. FEBS Letters. Submitted.
Jacqueline Kelly was responsible for completing the digestion and mass spectrometry
analysis to confirm the cleavage site of PLFMER. Dr. Dominic Cuerrier profiled the
alanine scanning variants of PLFAER with µI-II, which was repeated by Jacqueline who
extended this approach to the full-length µ- and m-calpains. The experiments were
performed under the guidance of Dr. Peter L. Davies with advice from Dr. Robert L.
Campbell and Dr. Dominic Cuerrier. Most of the alanine variants were synthesized by the
Alberta Peptide Institute (Edmonton, Alberta) and the others were synthesized in-house.
Mass spectrometry was performed by Dr. YiMin She (Queen’s University, Kingston,
Ontario). The manuscript was written by both Jacqueline Kelly and Dominic Cuerrier
with guidance from Dr. Peter L. Davies with editorial input from Dr. Robert L. Campbell.
34
3.2 Abstract
Calpains are intracellular proteases that selectively cleave proteins in response to
calcium signals. Although calpains cut many different sequences, residue preferences
within peptide substrates were recently determined and incorporated into a superior
FRET-based substrate (PLFAER). Here we show PLFAER is cleaved by calpain at the
intended F-A scissile bond. Sequential replacement of individual residues by alanine
reduced activity except with PLFAAR, which is cleaved 2.3 times faster than PLFAER.
The rates of hydrolysis of the alanine-substituted substrates were used to compare calpain
substrate preferences. These profiles for full-length µ- and m-calpains were similar and
closely matched that of the protease core of calpain 1.
3.3 Introduction
Calpains are intracellular cysteine proteases that are activated by calcium to make
discrete cuts in a wide variety of protein substrates as directed by calcium signaling.
Physiological roles for these select cleavages include the promotion of cell motility [15],
cell division [16], granule secretion [99] and apoptosis [17]. Many of the well-
documented calpain substrates are cytoskeletal proteins such as MAP2 [100], tau [101],
talin [102], spectrin, integrin [103] and vimentin [14]. Compilations of their cut sites
reveal very little sequence specificity other than a preference for a hydrophobic residue
(particularly Leu) in the P2 position [63,104,105]. However, it is also true that little
attention has been paid to the issue of how quickly individual sites are cleaved. Thus the
consensus sequence compiled by Tompa et al. includes all documented protein cleavage
35
sites irrespective of whether they are cut rapidly or slowly [100]. As with any protease,
cleavage of folded proteins is limited to discrete unstructured sections that generally
occur in loops, linker regions and termini. This might account for the abundance of
proline in the consensus sequence, since this amino acid disrupts secondary structure.
In view of these limitations, we earlier examined the sequence specificity of
calpain by experimentation [63] using the degenerate peptide library method pioneered
by Turk et al. (2001) [65]. This showed that there were preferred residues at the P1' (Met
or Ala), P2' (Glu) and P3' (Arg) positions in calpain peptide substrates. The incorporation
of this primed side sequence into a second degenerate peptide library led to the
identification of optimal residues at the unprimed positions. These were primarily
aliphatic and aromatic residues, including Leu at P2. The combination of preferences
gave rise to a hexapeptide Pro-Leu-Phe-Met/Ala-Glu-Arg sequence that was incorporated
into a FRET substrate with an N-terminal EDANS donor group and a C-terminal
DABCYL quencher. When this was compared to other hexapeptide FRET substrates, it
was found to be cleaved by the protease core of µ-calpain eight times more rapidly than
the consensus cleavage sequence Pro-Leu-Lys-Ser-Pro-Pro and 18 times more rapidly
than the spectrin cleavage sequence Glu-Val-Tyr-Gly-Met-Met [63]. The superiority of
PLFAER as a calpain substrate was confirmed by Polster et al. who compared it to other
known calpain substrate sequences [106]. This group used the increased specificity of
PLFAER to produce a biomarker derivative that has improved cell permeability. Through
slight modifications, they also reduced its digestibility by trypsin and chymotrypsin.
36
In the present study, we set out to further characterize this kinetically optimized
substrate. We validated the preferred substrate sequence by confirming its cleavage at the
intended scissile bond between Phe and Met. Confirmation of the optimal fit of each
residue was performed by replacing them in turn with alanine, resulting in the discovery
that the sequence Pro-Leu-Phe-Ala-Ala-Arg is an even better substrate for the calpain
protease core (µI-II). Lastly, we used this set of FRET peptides to show that µ- and m-
calpain have similar substrate preferences and that the protease core of µ-calpain has only
slightly different substrate preferences to those of the whole enzyme.
3.4 Materials and Methods
3.4.1 Materials
The FRET substrates, (EDANS)-EPLFAERK-(DABCYL), (EDANS)-
EPLAAERK-(DABCYL), (EDANS)-EPAFAERK-(DABCYL), (EDANS)-EALFAERK-
(DABCYL), (EDANS)-EPLFMERK-(DABCYL), and (EDANS)-EPLFGERK-
(DABCYL) were synthesized by the Alberta Peptide Institute (Edmonton, Alberta). The
FRET substrates (EDANS)-EPLFAARK-(DABCYL) and (EDANS)-EPLFAEAK-
(DABCYL) were synthesized in-house. Bold type indicates the residue substituted into
the original (EDANS)-EPLFAERK-(DABCYL). µ-Calpain was purchased from
Calbiochem (cat# 208712). The recombinant protease core (µI-II), and m-calpain were
prepared as described previously [37,44].
37
3.4.2 Determination of the Cleavage Site within the PLFAER Substrate
The substrate (EDANS)-EPLFMERK-(DABCYL) was incubated at a
concentration of 50 µM with 3.2 µM µI-II in cleavage buffer (50 mM MOPS pH 7.6, 100
mM NaCl, 10 mM CaCl2, 0.1% β-ME) and allowed to react to completion. Similarly, an
undigested control reaction in Ca2+-free conditions was prepared in 50 µM (EDANS)-
EPLFMERK-(DABCYL) and 3.2 µM µI-II in control buffer (50 mM MOPS pH 7.6, 100
mM NaCl, 1 mM EDTA and 0.1% β-ME). Digestion was monitored by a Perkin Elmer
LS50B luminescence spectrometer with excitation and emission wavelengths of 335 nm
and 500 nm, respectively. The reaction was allowed to go to completion, until no further
increase in fluorescence was seen, and the sample was flash frozen with ethanol/dry ice
and stored at -20 °C. Both samples were analysed on an Applied Biosystems/MDS Sciex
QSTAR XL QqTOF mass spectrometer with a nano-ESI source.
3.4.3 Effect of Individual Residue Substitutions on Cleavage Kinetics using µI-II, µ-
and m-calpain
The rate of hydrolysis of the FRET substrates by µI-II was monitored by fluorescence in
a 1 mL cuvette as described above. Readings were obtained at 1 s intervals. The cuvette
contained 2, 5 or 10 µM substrate, 0.71 µM µI-II in 50mM Tris-HCl pH 7.6 and 10 mM
DTT and the reaction was initiated by 5 mM CaCl2 in a final volume of 1 mL. The
amount of substrate digested per relative fluorescence unit (RFU) was calculated for each
individual substrate by using a control that was digested to completion with papain.
Briefly, substrate (2 µM) in control buffer was analyzed in the fluorimeter with and
38
without 0.35 µM papain and the change in fluorescence per µMole substrate was
calculated (Equation 1).
Fluorescence yield per µM = (RFUdigested – RFUundigested)/2µM Equation 1
These values were then used to calculate initial rates of hydrolysis by calpain and
were normalized relative to the initial rate of hydrolysis of PLFAER at the same substrate
concentration (all reactions were performed in the linear portion of the Michaelis-Menten
curve, where [S] << Km). A correction factor for the inner filter effect was also
incorporated as shown below. This experiment was repeated with µ- and m-calpain with
the exception that 0.13 µM enzyme was used for both and 1 mM CaCl2 was used with µ-
calpain.
The inner filter effect for the 1 mL cuvette in the Perkin Elmer LS50B
luminescence spectrometer was corrected for by determining the change in fluorescence
caused by 0.5 µM EDANS FRET donor at each substrate concentration as described
previously by Liu et al., 1999 [107] . Fluorescence was monitored at excitation and
emission wavelengths of 335 nm and 500 nm respectively for PLFAER at 0, 0.5, 1, 2, 5,
8 and 10 µM in buffer 3. EDANS (0.5 µM) was then added and the change in
fluorescence was determined for each substrate concentration. A correction factor was
calculated for each individual substrate and applied to the rates of hydrolysis (Equation
2).
39
Correction = RFUEDANS (at each substrate concentration) Equation 2
RFUEDANS (in the absence of substrate)
3.5 Results and Discussion
3.5.1 Determination of the Substrate Cleavage Site
The FRET substrate (EDANS)-EPLFMERK-(DABCYL) was hydrolyzed by the
calpain protease core, µI-II, until there was no further increase in fluorescence. The
digestion products were analyzed by MALDI-MS together with an undigested control
(Fig. 3.1). The control demonstrated a peak at m/z 775 corresponding to the uncleaved
FRET substrate with a +2 charge. There were no detectable signals at m/z values
corresponding to potential cleavage products. In the digested sample, there was no sign
of the substrate peak at m/z 775 (+2 charge), which indicates that the digestion had gone
to completion. Instead, a primary cleavage site of the substrate between F and M was
demonstrated by peaks at m/z 753 (EDANS-EPLF, +1), 377 (EDANS-EPLF, +2), 408
(MERK-DABCYL, +2) and 272 (MERK-DABCYL, +3) (Fig. 3.1). A minor peak with
m/z 342 corresponds to ERK-DABCYL, representing a secondary cleavage site between
M and E at the P1' and P2' positions, respectively. The absence of a peak corresponding
to (EDANS)-EPLFM identifies this cleavage as secondary to the P1-P1' hydrolysis. The
[M+H]+ values of the observed peaks correspond well with the theoretical values with
similar systematic errors (Table 3.1). The primary cleavage site revealed through mass
40
400 600 800 1000m/z, amu
0
200
400
600
800
1000
1200
1322 407.693
272.132
342.176
256.135
753.312
377.164
Figure 3.1. NanoESI-MS spectrum of the cleavage products from the hydrolysis of
(EDANS)-EPLFMERK-(DABCYL) by µI-II. Peaks at m/z of 753.3 and 377.2 represent the
(EDANS)-EPLF product with a +1 and +2 charge respectively. Mass to charge ratios of 407.7
and 272.1 indicate the c-terminal end of the substrate (MERK-(DABCYL)) with a +2 and +3
charge respectively. A secondary cleavage is also shown between the P1' Met and P2' Glu at a
m/z of 342.2
41
1548.713(EDANS)-EPLFMERK-(DABCYL)
1171.608PLFMERK-(DABCYL)
1169.512(EDANS)-EPLFMER
1074.555LFMERK-(DABCYL)
1013.411(EDANS)-EPLFME
961.471FMERK-(DABCYL)
884.368(EDANS)-EPLFM
0.011814.391 (+2, +3)814.403MERK-(DABCYL)
0.008753.320 (+1, +2)753.328(EDANS)-EPLF
0.01683.352 (+2)683.362ERK-(DABCYL)
606.259(EDANS)-EPL
554.320RK-(DABCYL)
493.175(EDANS)-EP
398.219K-(DABCYL)
396.122(EDANS)-E
∆[M+H]+Observed [M+H]+
(ion state)Theoretical
[M+H]+Potential Hydrolysis Product
1548.713(EDANS)-EPLFMERK-(DABCYL)
1171.608PLFMERK-(DABCYL)
1169.512(EDANS)-EPLFMER
1074.555LFMERK-(DABCYL)
1013.411(EDANS)-EPLFME
961.471FMERK-(DABCYL)
884.368(EDANS)-EPLFM
0.011814.391 (+2, +3)814.403MERK-(DABCYL)
0.008753.320 (+1, +2)753.328(EDANS)-EPLF
0.01683.352 (+2)683.362ERK-(DABCYL)
606.259(EDANS)-EPL
554.320RK-(DABCYL)
493.175(EDANS)-EP
398.219K-(DABCYL)
396.122(EDANS)-E
∆[M+H]+Observed [M+H]+
(ion state)Theoretical
[M+H]+Potential Hydrolysis Product
Table 3.1. Potential and observed FRET substrate hydrolysis products. The FRET substrate
(EDANS)-EPLFMERK-(DABCYL) was fully digested by µI-II in the presence of 10 mM CaCl2
and analysed by Nano.ESI-MS. All product masses are represented in their mono-protonated
form ([M+H]+ and the systematic error between the observed and theoretical [M+H]+ values is
shown (∆[M+H]+).
42
spectrometry between F and M supports the inferred position of the scissile bond
obtained through substrate profiling [63].
3.5.2 Efficacy of Ala and Met at the P1΄ Position
The peptide sequence PLFMER is rapidly cleaved by µI-II at the F-M bond, but
the extent to which each residue contributes to substrate recognition and binding in the
context of the whole peptide is unknown. Although it is assumed that each side-chain
makes a positive contribution to the reaction based on previous research [63], this claim
has been tested here. We initially looked at the preference between Met and Ala at the P1'
position, where these two residues were strongly favoured in the study by Cuerrier et al.,
2005 [63]. Although Met had been selected over Ala in a ratio of 4:3, in the context of the
whole peptide (Table 3.2), it was found to have only a slightly higher (though not
statistically significant) rate of hydrolysis compared to PLFAER (Fig. 3.2). The latter
substrate is preferred over the former for routine assays because Met is susceptible to
oxidation. PLFMER, however, was used in the mass spectrometry analysis of the
preferred cleavage site because the mass differences in the expected cleavage products
were more distinct with Met in the P1' position.
3.5.3 Effect of Individual Residue Substitutions on Cleavage Kinetics
To evaluate the contribution of specific residues in the other positions of the
optimal sequence, a series of alanine-substituted FRET substrates was synthesized. For
the P1' position already represented by alanine, a glycine substitution was made. The rate
43
0.91+/-0.061.94+/-0.312.84+/-0.35A
Selectivity RatioAmino Acid
P3’P2’P1’
0.47+/-0.160.67+/-0.300.33+/-0.04G
2.91+/-0.770.91+/-0.162.39+/-0.02R
0.35+/-0.292.13+/-0.570.74+/-0.13E
1.56+/-0.312.34+/-0.244.05+/-0.47M
0.91+/-0.061.94+/-0.312.84+/-0.35A
Selectivity RatioAmino Acid
P3’P2’P1’
0.47+/-0.160.67+/-0.300.33+/-0.04G
2.91+/-0.770.91+/-0.162.39+/-0.02R
0.35+/-0.292.13+/-0.570.74+/-0.13E
1.56+/-0.312.34+/-0.244.05+/-0.47M
Table 3.2. Preference for each amino acid at P1'-P3' positions of the hexapeptide substrate.
The abundance of Met (M), Ala (A), Glu (E), Arg (R) and Gly (G) at the P1’, P2’ and P3’
positions of the hexapeptide substrate, using a substrate library approach, shown as a selectivity
ratio. The selectivity ratio is defined as the ratio of the abundance of the residue observed at the
position to the abundance of the residue expected in the absence of any specificity (For more
information on the methods see [63]).
44
of the hydrolysis of these substrates by µI-II, µ- and m-calpain was determined and
compared to the unsubstituted PLFAER substrate. The PLFAER sequence was designed
from data obtained using µI-II [63]. With this protease construct, alanine-substitutions of
the P2-Leu and the P1-Phe had the greatest detrimental effect on substrate hydrolysis,
with corresponding substrates demonstrating 4 % and 8 % of the rate of hydrolysis of the
unsubstituted substrate, respectively (Fig. 3.2A). This is consistent with these two
subsites interacting with a large hydrophobic surface area in the substrate binding pocket,
as observed from the P2-Leu P1-Phe sequence of inhibitor SNJ-1945 in complex with µI-
II [62], and highlights the importance of these two positions on cleavage kinetics. The
P3'-Arg to Ala substitution reduced the rate of hydrolysis to 20 %, demonstrating the
importance of this relatively distal primed-side position. Substitution of the P1'-Ala by
Gly showed a similar detrimental effect (28 %), possibly by increasing the
conformational flexibility of the peptide. Alanine-substitution of the P3-Pro only had a
minor impact (73 %). Nevertheless, the presence of the P3-P1' and the P3' residues of the
unsubstituted substrate were all preferred over their Ala or Gly replacements, confirming
the positive effect of the side-chains.
Interestingly, the P2'-Glu to Ala substitution improved the cleavage of the
substrate (2.3-fold), suggesting that Ala is preferred at this position in the context of the
whole peptide. In retrospect this result is not in conflict with the original specificity
profiles [63], as the P2' position demonstrated selectivity for Ala and Met in addition to
Glu (Table 3.2). The reason Glu was incorporated at the P2' position was because the
45
A
B
Preference
C
0 1 2 3Preference
PLFAER ---PLAAER ---
PAFAER ---ALFAER ---
PLFMER ---PLFGER ---PLFAEA ---PLFAAR ---
0 1 2Preference
PLFAER ---
PLAAER ---PAFAER ---ALFAER ---
PLFMER ---PLFGER ---PLFAEA ---PLFAAR ---
0 1 2
PLFAER ---PLAAER ---PAFAER ---ALFAER ---
PLFMER ---
PLFGER ---
PLFAEA ---PLFAAR ---
A
B
Preference
C
0 1 2 3Preference
PLFAER ---PLAAER ---
PAFAER ---ALFAER ---
PLFMER ---PLFGER ---PLFAEA ---PLFAAR ---
0 1 2Preference
PLFAER ---
PLAAER ---PAFAER ---ALFAER ---
PLFMER ---PLFGER ---PLFAEA ---PLFAAR ---
0 1 2
PLFAER ---PLAAER ---PAFAER ---ALFAER ---
PLFMER ---
PLFGER ---
PLFAEA ---PLFAAR ---
Figure 3.2. Relative initial rates of the hydrolysis of the substituted FRET substrate series
by µI-II (A), µ-calpain (B) and m-calpain (C) normalized relative to PLFAER (preference
=1.0). Each substrate was analyzed in duplicate for µ-calpain, and triplicate for µI-II and m-
calpain at 2, 5 and 10 µM. The internal peptide sequence of the substrates are shown with the
substituted amino acid underlined, red and in bold. The peptide with the internal sequence
PLFAER corresponds to the unsubstituted substrate. Error bars correspond to +/- one standard
deviation.
46
presence of Ala and Met in the second cycle was misinterpreted as carry over from the
first cycle of the Edman degradation reaction, possibly due to incomplete coupling or
cleavage [63]. Therefore, contrary to initial expectations, the sequence PLFAAR is
cleaved more rapidly, and corresponds to a better calpain substrate than the sequence
PLFAER.
Using the panel of alanine-substituted sequences, the substrate cleavage profiles
for µ- and m-calpain were found to be very similar (Fig. 3.2). Moreover, they are
qualitatively similar to that of µI-II with a few distinctions. The P3-Pro to Ala
substitution appears to have slightly improved rates of hydrolysis by the full-length
calpains (1.4- and 1.1-fold for µ- and m-calpain respectively compared to 0.73-fold with
µI-II). This may be attributed to the added rigidity provided by the proline, which may
force the Glu and EDANS fluorophore into a less favourable position within domain III
that flanks the active site. Since the active site of µI-II is open on the unprimed side due
to the lack of domain III, it is less restrictive at this position. Also, the P1'-Ala to Gly
substitution is not as detrimental in the full-length enzyme as was shown with µI-II (1.2-
and 0.71-fold for µ- and m-calpain respectively, compared to 0.28-fold with µI-II). It has
been suggested that the active-site conformation of µI-II is more labile than in full-length
calpain because it lacks the support of the other domains. Indeed, the crystallization of
calpain protease cores is facilitated by having covalently bound inhibitors in the active
site cleft to make stabilizing contacts with the flanking domains I and II [43,62,108].
Therefore, the full-length enzyme is likely to tolerate the loss of a side chain in P1' better
47
than the less stable protease core µI-II.
The P2'-Glu to Ala substitution was found to produce a higher rate of hydrolysis
compared to the original substrate with both µ- and m-calpain as well as with µI-II. Thus,
PLFAAR is a superior substrate sequence for the whole enzyme of both major calpain
isoforms as well as the protease core. Now that we have improved the PLFAER substrate
that already showed superiority to other synthesized substrates, PLFAAR can be
stabilized as described by Polster et al. and tested as a biomarker for calpain activity in
cells [106].
In summary, we have demonstrated by mass spectrometry that the FRET
substrate, (EDANS)-EPLFMERK-(DABCYL), is cleaved by µI-II between the F and M
residues. Alanine-scanning variants were used to validate the importance of the P3-Pro,
P2-Leu, P1-Phe and P3'-Arg. Also, an improved sequence, PLFAAR, was determined to
have a 2.3-fold increase in initial rates of hydrolysis by µI-II relative to the original
PLFAER sequence. Full-length µ- and m-calpain were profiled with these alanine-
scanning variants, for comparison with the protease core profile. The profiling shows that
the two major calpains have very similar substrate preferences. In addition, with two
minor exceptions, these substrate preferences are retained in the protease core that lacks
the support of domains II, IV, V and VI. Profiling with these FRET substrates can
potentially be extended to the other calpain isoforms. These non-classical calpains may
exhibit different active-site specificities, which can be used in the development of
isoform-specific calpain inhibitors.
48
Chapter 4
The Stabilization of Calcium-bound Calpain by Calpastatin Peptides
4.1 Preface
This chapter will form the basis of a manuscript on the stabilization of calcium-bound
calpain.
Kelly, J., Hanna, R., Davies, P.L. (2008). The stabilization of calcium-bound calpain by
calpastatin peptides.
Jacqueline Kelly was responsible for cloning and purification of the CAST C-A peptides
as well as the enzyme and gel autolysis assays. The turbidity assays were done with
assistance from Rachel Hanna. The concentrations of the CAST C-A peptides were
determined by amino acid analysis performed by The Advanced Protein Technology
Centre (Hospital for Sick Children, Toronto, ON). The experiments were performed
under the supervision and guidance of Dr. Peter L. Davies with helpful input from Rachel
Hanna. The chapter was written by Jacqueline Kelly with insight from Dr. Peter L.
Davies.
49
4.2 Abstract
Calpains are non-lysosomal cysteine proteases activated by intracellular influxes
of calcium. Over-activation of the calpains by misregulated calcium levels contributes to
many pathological conditions and is one reason to develop pharmacological inhibitors of
these enzymes. To avoid problems with autolysis and calcium-induced precipitation that
hamper work with the full-length enzymes, this has previously been pursued with the
protease cores of calpain. However, they are much less active than the full-length
enzymes and are missing a domain that contributes to inhibitor specificity. Here we have
explored the use of segments of the natural inhibitor, calpastatin, to prevent calcium-
induced calpain precipitation. Peptides that span the three different, contiguous C and A
sub-domains of calpastatin, while omitting the inhibitory B sub-domain, were made and
mixed with calpain at different ratios. All three C-A peptides inhibited precipitation at a
2.5:1 molar ratio of peptide to enzyme as monitored by optical density at 320 nm. Peptide
3C-4A was the most effective of the three and prevented the onset of precipitation even at
a 1:1 molar ratio. In addition, the peptides did not affect the activity of calpain or its
inhibition by leupeptin. Although the peptides are cleaved by active m-calpain, they
continue to protect the enzyme from aggregation and precipitation. This method increases
the opportunities for using full-length calpains in in vitro experiments such as X-ray
crystallography and active-site profiling of substrates and inhibitors.
50
4.3 Introduction
Calpains are a family of cysteine proteases that perform limited proteolysis of
their target substrates in response to increases in intracellular calcium levels [14]. Since
calpains become unregulated and over-activated in many disorders such as reperfusion
injury following stroke and heart attack, Parkinson’s disease and Duchenne muscular
dystrophy, they are key targets for the development of small-molecule inhibitors [109].
One obstacle to the use of calpains in drug development has been their instability and
tendency to aggregate when their cofactor, calcium, is present. This problem has partly
been resolved with the use of the protease core (DI-II) of calpain, which is stable and
active in the presence of calcium [44]. The protease core of µ-calpain (µI-II) has been
used to determine optimal inhibitor and substrate sequences at the active site [64,63].
Unfortunately, the protease cores are much less active than the full-length calpains. For
example, µI-II is only 2-3 % as active as the whole enzyme [97]. They also lack the
additional specificity provided by DIII at the unprimed side of the active site. Therefore,
the stabilization of full-length calpains would allow one to work with the whole enzymes
in the presence of calcium and may accelerate the development of calpain-specific
substrates and inhibitors.
Calpastatin is the only known inhibitor that is specific for the heterodimeric calpains. It
can interact with up to four calpain molecules at one time [87] because it contains four
homologous tandemly arranged inhibitory domains (CAST1-4). Each CAST domain
interacts with calcium-bound calpain through three well conserved sub-domains: A, B
51
and C (Fig. 4.1A) [75,83,84,110] . Recently, calcium-bound m-calpain was co-
crystallized in complex with CAST4 (Fig. 4.1B) [46]. The X-ray crystal structure of the
complex revealed an aligned active site where sub-domain B of CAST bound, escaping
cleavage by looping out around the active-site cysteine, and exerting its inhibitory affect
by blocking access to substrates. This structure also confirmed the presence of small
hydrophobic patches exposed upon calcium binding in DIV of the large subunit and DVI
of the small subunit. CAST sub-domains A and C form amphipathic α-helices and make
hydrophobic contacts with these regions in DIV and DVI, respectively [46,85,86,111].
It has been proposed by other researchers that the instability of calcium-bound
calpain may involve the dissociation of the small subunit, leading to aggregation
[112,70]. If so, it is possible that CAST helps stabilize calpain because it physically links
the two subunits together. It is also possible that the exposed hydrophobic patches formed
during the activation of calpain by calcium might promote aggregation when they are not
bound and covered by CAST [46]. In the present study, we propose a mechanism by
which calpain can be stabilized in the presence of calcium using a portion of its natural
inhibitor while keeping the active site clear (Fig. 4.1C). The inter-domain regions of
calpastatin that link the A and C sub-domains of CAST from one domain to the next (1C-
2A, 2C-3A and 3C-4A) were recombinantly produced to provide a physical link between
the PEF domains (DIV and DVI) (Fig. 4.1A). Although the dissociation constants
between calpain and each CAST domain have previously been determined by surface
plasmon resonance, the C-A linkers between each domain are hybrid molecules and have
52
B CAB CAB CAB CA
1 2 3 4L
A
C
A
CN
B
C
A
N
C
B C
86 64 98
B CA B CAB CA B CAB CA B CAB CA B CA
1 2 3 4L
A
C
A
CN C
A
CN C
A
CN
B
C
A
N
C
B
C
A
N
C
B C
86 64 98
Figure 4.1. The proposed mechanism of calpain stabilization using CAST. (A) The domain
structure of CAST including an N-terminal L domain followed by 4 homologous domains each
consisting of A, B and C sub-domains. The regions highlighted with a red rectangle are the C-A
peptides that were cloned to stabilize calpain and the number of amino acids for each peptide is
written below. (B) The x-ray crystallographic structure of CAST bound to m-calpain. CAST is
shown in purple with cylindrical helices and labelled sub-domains. The dashed line corresponds
to floppy regions not resolved in the x-ray structure [46] (C) A model of the proposed mechanism
of stabilization by the CAST C-A peptides. The active site that is normally occupied by the B
sub-domain of CAST is free in this model. N- and C-termini of CAST are shown.
53
not been assessed in this way [87]. Therefore, each C-A peptide was tested for its degree
of calpain stabilization.
4.4 Materials and Methods
4.4.1 CAST C-A Peptide Cloning
DNAs coding for the CAST C-A peptides, 1C-2A, 2C-3A and 3C-4A, (Fig. 4.1)
were amplified by polymerase chain reactions (PCR) using the following primers:
CATTGCATATGATGGGAATCGACCATGCTATAG (sense 1C-2A);
TCGATCTCGAGGTCTGGCTGCCGGGTGC (anti-sense 1C-2A);
CATTGCATATGTGCCTGAGTGAGTCAGAGC (sense 2C-3A);
TCGATCTCGAGTTCTGGATCTTCTTTCCTTGTC (anti-sense 2C-3A);
CATTGCATATGGAACAACTTCCACCCTTAAGCG (sense 3C-4A);
TCGATCTCGAGTGGTTTGTTCTCATCTGGATCAG (anti-sense 3C-4A). The DNAs
coding for CAST 2C-3A and the pET24a vector (Novagen) were digested with NdeI and
XhoI and directly ligated such that the expressed protein has a C-terminal His tag. The
DNAs coding for CAST 1C-2A and 3C-4A were TA cloned into Top 10 cells
(Invitrogen), released from the intermediate vector by digestion with NdeI and XhoI, gel
purified and ligated into the NdeI and XhoI cut sites of the pET24a vector (Novagen).
54
4.4.2 Protein Expression and Purification
Recombinant calpains, both active m-calpain (m80k/21k) as well as inactive
(C105S 80k/21k) m-calpain, with their N-terminally truncated small subunit were
expressed and purified as previously described [37,113]. CAST C-A plasmids were
transfected into Escherichia coli BL21 (DE3) cells (Novagen) by electroporation. A
single colony was used to inoculate LB media (25 mL) with kanamycin selection for
growth overnight before starting a 1 L culture with LB/kanamycin broth. The cells were
grown to an OD 595 nm of 1.0 at 37 °C. CAST C-A expression was initiated with 0.4 mM
IPTG for 3 h. The cells were resuspended in lysis buffer (25 mM Tris-HCl pH 7.6, 5 mM
EDTA, 5 % glycerol, 10 mM β-mercaptoethanol, 100 µM PMSF) and lysed via
sonication. The supernatant was heated to 85 °C for 15 min and the insoluble particulates
of denatured protein were removed by centrifugation. The soluble fraction was loaded on
a 5-mL Ni2+-chelating agarose resin column (Qiagen). The column was washed with two
column volumes of 25 mM Tris-HCl pH 7.6, 2 % glycerol, 500 mM NaCl buffer and
eluted with 250 mM imidizole in the same buffer. The elution fractions containing the
CAST C-A peptides were identified by SDS-PAGE, concentrated and flash frozen.
4.4.3 CAST C-A Peptide Amino Acid Analysis
The CAST peptides have unusual amino acid compositions and two of the three
peptides do not absorb at 280 nm. For these reasons, the concentrations of each peptide
were determined using amino acid analysis. The Advanced Protein Technology Centre
(Hospital for Sick Children, Toronto, ON) performed the analyses.
55
4.4.4 Monitoring Aggregation of mC105S by Turbidity
The rate and extent of protein aggregation were monitored by optical density at
320 nm in a 96-well non-binding surface plate (Corning). The wells contained 4 µM
mC105S with a 1-, 1.25-, 2.5- and 5-times molar ratio of CAST C-A peptides (4, 5, 10,
and 20 µM, respectively) in 50 mM Tris-HCl pH 7.6, 10 mM DTT. Lastly, CaCl2 was
added to a final concentration of 4 mM. The plate was shaken in a Powerwave XS
Microplate spectrophotometer (Bio-Tek Instruments) for 10 s prior to each reading taken
at intervals over 2 h at room temperature. Controls lacking calcium and CAST C-A were
also done as described above.
4.4.5 The Effect of CAST C-A Peptides on m-Calpain Activity
The hydrolysis of 10 µM (EDANS)-EPLFAER-(DABCYL) by 0.2 µM m-calpain
was monitored on a Perkin Elmer LS50B luminescence spectrometer in the presence of 2
µM CAST C-A peptide in buffer containing 50 mM Tris-HCl pH 7.6, 10 mM DTT, and 4
mM CaCl2. The excitation and emission wavelengths were 335 and 500 nm, respectively.
A calpain control in the absence of CAST C-A was also done. The digestion was repeated
as described above with the addition of 0.5 mM leupeptin 3 min after initiation of the
reaction.
4.4.6 The Effect of CAST C-A Peptides on m-Calpain Autolysis
Autolysis of m-calpain in the presence of calcium with or without leupeptin was
monitored by SDS–PAGE. A solution of 9 µM m-calpain in autolysis buffer (50 mM
56
HEPES, 100 mM NaCl, and 20 mM DTT) and 25 µM CAST C-A was added to an equal
volume of calcium buffer (autolysis buffer + 10 mM CaCl2) and incubated at room
temperature. Aliquots (20 µL) were taken at 0, 1.5, 3, 5, 10, 15, and 30 min and added to
20 µL 3X SDS-PAGE sample buffer (150 mM Tris-HCl pH 6.8, 300 mM DTT, 6 %
SDS, 0.3 % bromophenol blue, and 30 % glycerol). The samples were heated at 95 °C
and 10 µL was loaded on a Tris-tricine SDS 12 % polyacrylamide gel.
4.5 Results
4.5.1 Purification and Validation of the C-A Peptides
The first step in the purification of the CAST C-A peptides was the heat denaturation of
most E. coli proteins (Fig. 4.2A, lanes 1, 2). This step was very efficient as the C-A
peptides lack secondary structure and are extremely heat stable, while most other
contaminants denatured, aggregated and were removed by centrifugation. The soluble
portion following heat treatment was subject to a Ni2+-affinity column, since the C-A
peptides were cloned with a C-terminal His-tag (Fig. 4.2B). Most of the remaining
contaminants were removed in the flow through and column washes (Fig. 4.2B, lanes 2-
4) resulting in the elution of the CAST 1C-2A peptides (Fig. 4.2B, lanes 5-9). The
expression and purification from 1 L of culture, yielded 20 mg, 22 mg, and 10 mg of
CAST 1C-2A, 2C-3A and 3C-4A, respectively, and the concentrated samples of each C-
A peptide were shown to be >95 % pure based on SDS-PAGE (Fig. 4.2C). The expected
molecular weights for CAST 1C-2A, 2C-3A, and 3C-4A were 9535, 7169, and 10823,
57
M 2 3 4
M 2 3 4 5 6 7 8 9
20
27
6666
272014
6.5
66
2720
14
4335
4335
kDa
kDa
kDa
A B1 2
C
6.5
M 2 3 4
M 2 3 4 5 6 7 8 9
20
27
6666
272014
6.5
66
2720
14
4335
4335
kDa
kDa
kDa
A B1 2
C
6.5
Figure 4.2. Purification of CAST 1C-2A. Lane M refers to the molecular weight markers. For
(A) and (B), 10 µL of sample was added to 10 µL of SDS-PAGE loading buffer and 10 µL was
run on a 12 % polyacrylamide SDS-PAGE gel. (A) Lane 1, cell lysate; Lane 2, supernatant of
heated sample. (B) Ni2+-affinity Column. Lane 2, flow through of Ni2+ column; Lanes 3 and 4,
Ni2+ column washes; Lanes 5-9, Ni2+ column eluted fractions. (C) Concentrated peptides (1 µg).
Lane 2, CAST 1C-2A; Lane 3, CAST 2C-3A; Lane 4, CAST 3C-4A.
58
respectively, and were measured to be about 11000, 9000, and 14000, respectively by
SDS-PAGE.
The concentrated C-A peptides were sent for amino acid analysis to the Advanced
Protein Technology Centre (Hospital for Sick Children, Toronto, Ontario). The number
of each amino acid in the peptides was calculated using the DNA sequences and
compared to the observed values based on the amino acid analysis (Table 4.1). This was
done to validate the concentration obtained by this method and as a check on purity. The
amino acid analyses of Ser, Met, and Lys were shown to be outliers based on the
expected values and were not included in the calculated concentration. The His-affinity
tag is represented in the amino acid analysis as six His residues.
4.5.2 CAST C-A Peptides Inhibit the Ca2+
-Induced Insoluble Aggregation of
mC105S
To test the effect of the CAST C-A peptides on the stability of calpain without the
added complication of autolysis, the inactive m-calpain that has the active-site cysteine
mutated to a serine (mC105S) was used. The time of onset and rate of insoluble
aggregation of mC105S were monitored by turbidity at 320 nm (Fig. 4.3). In the absence
of calcium and CAST, there was no increase in OD 320 nm during the 200 min of the
experiment (Fig. 4.3A). In the presence of calcium, aggregation occurred that was visible
to the eye and was plotted as an increase in turbidity. The onset was rapid and turbidity
reached a peak value after 15 - 20 min. The reversibility of this aggregation was shown
59
Expected Observed Expected Observed Expected ObservedAla 7 7.0 5 5.3 9 8.8
Asx 7 6.6 4 3.9 15 15.5
Glx 13 13.4 8 10.5 12 12.6
Ser 10 8.3 5 4.2 10 8.0
Gly 6 6.1 3 3.6 3 3.4
His 7 6.7 6 5.8 7 8.3
Arg 3 3.2 2 2.2 3 3.9
Thr 6 5.3 3 2.8 3 3.2
Pro 4 4.3 5 5.2 12 12.2
Tyr 0 0.2 1 1.2 0 0.1
Val 4 4.8 4 4.5 3 3.5
Met 3 1.4 2 1.2 1 0.3
Ile 2 1.7 2 1.9 1 0.8
Leu 5 5.2 6 5.6 13 11.8
Phe 1 0.8 1 0.8 3 2.1
Lys 7 3.5 6 2.0 3 2.1
Cys 1 0.1 1 0.0 0 N/A
Trp 0 N/A 0 N/A 0 N/A
Total 86 78.7 64 60.9 98 96.4
Amino Acid
1C-2A 2C-3A 3C-4A
Table 4.1. Validation of the CAST C-A amino acid analysis. The number of each amino acid
in the three C-A peptides as expected from the DNA sequence and observed from the amino acid
analysis.
60
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 20 40 60 80 100 120
Time (min)
Op
tic
al
De
ns
ity
(3
20
nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 50 100 150 200
Time (min)O
pti
cal D
ensi
try
(320
nm
)
Calcium Control
No Calcium Control
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
A
B C
D E
+ EDTA
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 20 40 60 80 100 120
Time (min)
Op
tic
al
De
ns
ity
(3
20
nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 50 100 150 200
Time (min)O
pti
cal D
ensi
try
(320
nm
)
Calcium Control
No Calcium Control
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80
Time (min)
Optical D
ensity (
320nm
)
Control
CAST 1C-2A
CAST 2C-3A
CAST 3C-4A
A
B C
D E
+ EDTA
Figure 4.3. Monitoring calpain aggregation by turbidity. (A) Aggregation of 4 µM mC105S in
the presence ( ) and absence ( ) of 4 mM CaCl2. Excess EDTA (8 mM) was added to the
calcium control 75 min after the initiation of the reaction. (B–E) Turbidity monitored in the
presence of a 5:1 (B), 2.5:1 (C), 1.25:1 (D) and 1:1 (E) molar ratio of CAST C-A to mC105S. In
each case, reactions with CAST 1C-2A ( ), 2C-3A ( ) and 3C-4A ( ) as well as a no-CAST
control ( ) were performed.
61
by the sharp return to baseline absorbance following the addition of excess EDTA (final
concentration 8 mM) at 80 min. After EDTA was added, there was no further increase in
turbidity during the remaining 120 min of the experiment.
All three CAST C-A peptides inhibited the calcium-induced aggregation of mC105S at
5:1 and 2.5:1 molar ratios of CAST to mC105S (Fig. 4.3B and C, respectively). In these
experiments, the calcium control consistently reached a maximum turbidity after 15-20
min. The gradual decline in OD 320 nm thereafter can be attributed to the formation of
fewer but larger aggregates resulting in a net decrease in the amount of light scattering.
This is consistent with the greater fluctuations seen in the turbidity values towards the
end of the experiment.
It was not until a 1.25:1 molar ratio of CAST to mC105S was used that some
aggregation occurred with the CAST 1C-2A and 2C-3A peptides. However, these
calpastatin fragments caused a substantial delay in the onset of precipitation of ~100 and
~60 min, respectively. In addition, the amount of turbidity seen with CAST 1C-2A and
2C-3A only reached 8 % and 36 % of the control reaction, respectively. At a 1.25:1 molar
ratio of peptide to enzyme, CAST 3C-4A completely inhibited the formation of
aggregates (Fig. 4.3D). This trend continued when the molar ratio of CAST to mC105S
was further reduced to 1:1. With CAST 1C-2A the delay in the onset of measurable
aggregation fell to ~50 min and the final optical density relative to the control reaction
rose to ~68 % (Fig. 4.3E). With CAST 2C-3A, the delay was determined to be ~30 min
with the final OD 320 nm reaching ~80 % of the control reaction. Surprisingly, even with a
62
1:1 ratio, CAST 3C-4A inhibited aggregation as monitored by turbidity for at least 200
min.
4.5.3 CAST C-A Peptides Stabilize m-Calpain in the Presence of Ca2+
while Keeping
the Active Site Clear
To determine the effect of the CAST C-A peptides on the specific activity of calpain, a
FRET-based assay using the substrate, (EDANS)-EPLFAER-(DABCYL), was performed
(Fig. 4.4). The plot of relative fluorescence units as a function of time of reaction traced a
rectangular hyperbola that was approaching a plateau at 12 min. The presence of any of
the three CAST C-A peptides did not affect the rate at which calpain
cleaved the FRET substrate, as the slope of the reaction curves were superimposable over
the control reaction (Fig. 4.4). In addition, leupeptin quickly and completely inhibited the
enzyme, further demonstrating that the C-A peptides do not block the active site or cause
a conformational change that would disrupt the alignment of the catalytic triad.
Therefore, we have shown that the peptides inhibit aggregation and precipitation, while
keeping the active site clear to interact with target substrates as well as active-site
inhibitors.
Since the CAST peptides are unstructured, they are expected to be proteolytic
targets of calpain. Therefore, the extent of digestion of the C-A peptides as well as their
effect on autolysis of calpain was monitored by SDS-PAGE (Fig. 4.5). The no-CAST
control reaction demonstrated rapid autolysis in the presence of calcium (Fig. 4.5A).
63
0
50
100
150
200
250
300
350
400
450
500
0 2 4 6 8 10 12
Time (min)
RFU
Control
1c-2a
2c-3a
3c-4a
Control + Lep
1c-2a + Lep
2c-3a + Lep
3c-4a + Lep
+ Lep
0
50
100
150
200
250
300
350
400
450
500
0 2 4 6 8 10 12
Time (min)
RFU
Control
1c-2a
2c-3a
3c-4a
Control + Lep
1c-2a + Lep
2c-3a + Lep
3c-4a + Lep
+ Lep
Figure 4.4. Monitoring the effect of the CAST C-A peptides on m-calpain activity and
inhibition as measured with a FRET-based assay. The activity of m-calpain (0.2 µM) in the
absence of CAST ( ) was compared to its activity in the presence of a 10:1 molar ratio of CAST
1C-2A ( ), 2C-3A ( ) and 3C-4A ( ) to m-calpain and 4 mM CaCl2. Unfilled symbols
represent reactions that were inhibited with the addition of 0.5 mM leupeptin (Lep) ~3 min after
initiation.
64
0 1.5 3 5 10 15 30 0 1.5 3 5 10 15 30
E F
1.5 3 5 10 15 30 0 1.5 3 5 10 15 30G H
0
DI-IV
DI-II
DVI
A0 1.5 3 5 1015 30
6643
272014
6.5
35
kDa
B0 1.5 3 5 10 15 30
1.5 3 5 10 15 30 0 1.5 3 5 10 15 300
DI-III
DIV
C D
0 1.5 3 5 10 15 30 0 1.5 3 5 10 15 30
E F
1.5 3 5 10 15 30 0 1.5 3 5 10 15 30G H
0
DI-IV
DI-II
DVI
A0 1.5 3 5 1015 30
6643
272014
6.5
35
kDa
B0 1.5 3 5 10 15 30
1.5 3 5 10 15 30 0 1.5 3 5 10 15 300
DI-III
DIV
C D
Figure 4.5. Autolysis of m-calpain in the presence of the CAST C-A peptides monitored by
SDS-PAGE. The time points listed are in minutes. The zero time lanes for (B), (D) and (G) were
run on a separate gel. The autolysis fragments of m-calpain (DI-IV, DI-III, DI-II and DVI) and
bands corresponding to the C-A peptides are labelled with a star (red, green, and blue for 1C-2A,
2C-3A, and 3C-4A, respectively). Autolysis of 9 µM m-calpain was monitored in the presence of
4 mM CaCl2 and 25 µM CAST 1C-2A (B), CAST 2C-3A (C) and CAST 3C-4A (D). The effect
of 0.5 mM leupeptin on autolysis was monitored in the presence of CAST 1C-2A (F), CAST 2C-
3A (G) and CAST 3C-4A (H). Control reactions were done in the absence of CAST C-A (A and
E).
65
In particular, some protease core (domains I and II) was released within the first few
minutes of autolysis and there was accumulation of DI-III and DIV as the reaction
progressed. This autolysis was not noticeably inhibited by the CAST C-A peptides as
judged by the decrease in the amount of 80k subunit with time and accumulation of
autolysis products (Fig. 4.5B-D). The CAST C-A peptides were themselves cleaved
within 1.5 min of the start of the reaction as shown by their disappearance from the gel.
Given that the peptides are rapidly cleaved by calpain, the addition of an active-site
calpain inhibitor (leupeptin) reduced the extent of proteolysis (Fig. 4.5F-H). There was
no decrease in the staining of the 80k large calpain subunit during the 30 min incubation
and no appearance of the autolysis products seen in Fig. 4.5A-D. CAST 1C-2A was the
most resistant to proteolysis, as there was very little diminution of the band
corresponding to the whole peptide (~10 kDa) and appearance of CAST degradation
products over time (Fig. 4.5F). In the case of CAST 2C-3A, there seems to be an initial
cleavage in the first minute of the assay that does not undergo further proteolysis in the
remaining 60 min. This is evident since no increase in intensity of the bands
corresponding to the cut peptides is shown (Fig. 4.5G). The CAST 3C-4A peptide
undergoes slow but prolonged digestion, as shown by the disappearance of the band at
~14 kDa.
4.6 Discussion
Autolysis coupled with aggregation and precipitation of calpain in the presence of
calcium have been major obstacles to the characterization of full-length calpains. To
66
circumvent this instability, some research progress was made using the soluble protease
cores. Unfortunately, the protease cores are known to be much less active than the whole
enzyme and, because of the absence of domain III, they lack specificity beyond the P3
positions of substrates and inhibitors. In this study, we have used the inter-domain
regions of CAST (1C-2A, 2C-3A, and 3C-4A), a calpain-specific inhibitor that is known
to stabilize full-length calcium-bound calpain. This region normally resides between
calpain molecules but was modeled to be long enough to span from DIV to DVI of
calpain. It was our hypothesis that this would provide a link from DIV of the large
subunit to DVI of the small subunit, covering hydrophobic pockets that are potential
points of aggregation. In addition, if subunit dissociation actually occurs coincident with
aggregation, these peptides may add stability to the interaction of the PEF domains by
non-covalently cross-linking them.
While maintaining the calpain stabilization of full-length CAST, the C-A peptides
of CAST were used to avoid interference with small molecule inhibitors at the active site
of calpain. This was shown in two ways: the C-A peptides did not affect m-calpain
activity in the presence of calcium and did not interfere with the inhibition of calpain by
leupeptin. Although the CAST C-A peptides prevented calpain precipitation and did not
affect calpain activity, the flexible linker regions were shown to undergo proteolysis.
These regions are vital to the prevention of calpain precipitation, as the individual C and
A peptides are much less effective than the C-A linked peptides (Hanna et al.,
67
unpublished results). Therefore, their vulnerability to proteolysis should be reduced by
sequence modification.
CAST 3C-4A was determined to be the best of the three peptides at preventing the
precipitation of inactive m-calpain but when assayed with the active isoform, it was
found to be the most readily proteolysed. This shows perhaps that the 3C-4A peptide has
a stronger interaction with calpain (a lower KD) than the remaining peptides.
Unfortunately, the linker region is much longer and more vulnerable to cleavage. As well
as being the longest peptide, CAST 3C-4A has the most Leu (13 compared to 5 and 6 for
CAST 1C-2A and 2C-3A, respectively). Leu is extremely favoured at the unprimed side
of the calpain active site, where S3, S2, and S1 demonstrated a high preference for this
amino acid, using a substrate library-based approach [63]. Since CAST 3C-4A was the
most effective at preventing calpain precipitation, it is the obvious candidate for
modification to protect it from digestion.
There are several ways to reduce the vulnerability of the C-A peptides to
proteolysis. The 3C-4A peptide could be further optimized by mutation of the Leu in the
linker region to less favourable residues such as Pro or Ser [63]. The extended, flexible
linker regions could be replaced with a rigid secondary structural element, such as a
helix. This would increase the resistance of the peptides to proteolysis, as calpain prefers
extended peptide regions. Since CAST 3C-4A was determined to be the best stabilizer of
calpain and CAST 1C-2A was the most resistant to proteolysis, the linker region of the
former could be replaced with the latter. This modified peptide could then be tested for its
68
ability to prevent precipitation by turbidity. It may be that length of the 3C-4A linker is
optimal for preventing aggregation and the 1C-2A linker is too short. In addition to the
improvement of CAST 3C-4A, any other combination of the C and A peptides could be
tested for its effectiveness at preventing calpain precipitation while being protected from
digestion.
In conclusion, we have developed a way of stabilizing calpain in the presence of
calcium while avoiding interference with the active site by using a portion of calpain’s
natural inhibitor, CAST. The presence of the CAST peptides affects neither calpain
activity nor inhibition and opens the doors for co-crystallization of full-length calpains
bound to active-site inhibitors. This will aid in the structure-based design of active-site
inhibitors that take advantage of added preferences provided by DIII in the unprimed
positions.
69
Chapter 5
General Discussion
5.1 Is This the End of the Road for Mini-calpains?
The use of the calpain protease cores has facilitated significant advances in the
characterization of the active site of µ- and m-calpains. Since the full-length calpains
undergo autoproteolysis and aggregation in the presence of calcium, the protease cores
have been invaluable in launching the structure-based design of calpain inhibitors. The
identification of the first four µI-II / inhibitor structures led to insights about structural
preferences at the active site of calpain [62,64]. For example, the structure of µI-II bound
to SNJ 1945 revealed that the primed-side extension of the inhibitor caused a loop
displacement that exposed new sites for interaction by calpain inhibitors [62]. This was
further demonstrated in the recent structure of µI-II with an α-ketoamide inhibitor that
extended into and revealed novel aromatic stacking interactions with the Trp on the
primed side of the cleft [61]. This compound is being used as a template for the addition
of functional groups onto the inhibitor that target the primed side to increase its
specificity and potency. In addition to crystallographic analysis, the protease core
approach opened the doors to active site profiling with substrate and inhibitor libraries.
Since the protease cores of calpain resist autoproteolysis, a peptide-library
approach pioneered by Turk et al. was used in the development of a calpain substrate,
PLFAER [63]. This was shown to be kinetically superior to substrates built from α-
70
spectrin or consensus sequences obtained from substrate alignments [63,65]. It also
revealed sequence preferences at the primed side of the active site. A further
improvement of this substrate to PLFAAR revealed a 2.3-fold higher turnover rate when
tested with µI-II (see chapter 4).
Sequencing of the human genome has increased the number of known calpain
isoforms to 14, most of which have not yet been characterized at the protein level. Some
of these isoforms are implicated in diseases such as gastric cancer (calpain 9) and type II
diabetes (calpain 10), but their active-site specificity has not been determined. This is
partly due to their low expression levels, making isolation from tissue difficult. The
Structural Genomics Consortium (Toronto, Ontario) initiated a project to express and
purify all human isoforms. Following our success with the protease cores of µ- and m-
calpains they were able to produce protease cores of human calpains 1, 3, 8, 9 and 15,
and determine the crystallographic structures of three: mini-calpains 1, 8 and 9 (PDB
entries 2ARY, 2NQA and 1ZIV, respectively). They found that the protease cores
crystallized much better with an inhibitor bound to the active site, thereby stabilizing the
structure.
Although the protease cores have revealed structural preferences at the active site
of µ- and m-calpains, there are two main drawbacks to their use. First, they are known to
be much less active than the parent enzyme. For example, µI-II is only 2-3 % as active as
µ-calpain [97]. This difficulty was compounded when the activities measured with the
calpain 3, 8, and 9 protease cores proved much lower than that of µI-II. For this reason, it
71
was difficult to determine inhibitor specificity even at the S2 position of mini-calpain 8.
For mini-calpain 8, some of the decrease in activity can be credited to the reduced
stability of the protease core. Specifically, helix α7 in DI is stable in the full-length
enzyme but collapses when a Gly is present at position 203 within the helix. This
hypothesis was supported by an increase in activity of mI-II when the Gly at this position
was mutated to an Ala [97]. Similarly, the activity of mini-calpain 8 was increased,
although to a lesser extent, by the same mutation (G203A).
The second drawback associated with mini-calpains is some loss in specificity
provided by DIII, one of the supportive domains of the large subunit, which affects the
unprimed side of the catalytic cleft. Cuerrier et al. determined that full-length m-calpain
produced preferences that were slightly different from mI-II at the unprimed positions,
especially those further from the active-site cysteine (≥P3) [64]. Preliminary research on
the non-classical calpain isoforms suggests that although the use of the µ- and m-calpain
protease cores has been useful in the probing of substrates and inhibitors and for
structure-based inhibitor design, the use of the protease cores may not be transferable to
the other calpain isoforms. The different full-length calpain isoforms may demonstrate
greater sequence specificity than their protease cores when compared to other cysteine
proteases as well as to the other calpain isoforms.
5.2 Full-Length Calpains, the Way of the Future
Given the limitations of protease cores, full-length calpains could be more useful
for the development of calpain- and isoform-specific substrates and inhibitors. The
72
problems with this approach are that the full-length enzymes autolyze, aggregate, and
precipitate upon calcium binding. Previous attempts at crystallizing inactive (C105S)
full-length m-calpain in the presence of calcium have been difficult, due to precipitation
and proposed subunit dissociation [70]. These problems were circumvented by CAST
binding as demonstrated by the recent co-crystal structure of calcium-bound m-calpain
and CAST4 [46]. Although CAST is known to stabilize calpain, it also inhibits it by
occupying the active site. In this thesis, we have demonstrated a way of using the
stabilizing property of CAST while avoiding its inhibitory potential.
Although the CAST C-A peptides inhibited aggregation of inactive calpain, they
are readily cleaved by the active enzyme. Their unstructured, extended conformation
renders them susceptible to proteolysis. This vulnerability may be reduced by the
mutation of residues that are known to be favoured in the active site of calpain. The
linkers could be scanned for sequences that match the optimal PLFAER and PLFAAR
cleavage sequences at two or more sites, especially ones that have Leu, Ile and Val in the
S2 subsite. For example, one of the prominent autoproteolysis sites in calpain between
domains III and IV is RVFSEK [114]. This matches at two positions with a further three
conservative substitutions. It would be relatively easy to spoil this cleavage site by
mutation of VF to hydrophilic residues. The addition of Pro in the peptides may also
provide protection from proteolysis as cleavage is restricted to the C-terminal side of that
amino acid. The insertion of secondary structural elements that are resistant to calpain-
mediated cleavage would also decrease digestion of the CAST C-A peptides.
73
The applications of this method of calpain stabilization are numerous, including
substrate and inhibitor specificity determination at the unprimed side of calpain where
DIII helps form the S3 and S4 and outer unprimed sub-sites. So long as the CAST C-A
peptides are protected from calpain digestion; the library-based approaches described
previously can be extended to the stabilized full-length calpain. Following the
identification of an optimal substrate sequence, the method developed by Polster et al.
with PLFAER could be used to produce cell-permeable biomarkers of calpain over-
activation in cells [106].
In addition, structure-based design of calpain inhibitors with stabilized full-length
calpain that target both the active site as well as allosteric positions can now be pursued.
Compounds that have been shown to extend past subsites covered by the protease core
alone, such as WR13 (R,R) and WR18 (S,S) would be valuable targets for the co-
crystallization with the stabilized whole enzyme [64]. These inhibitors have an N-
terminal Trp where no density appeared in the inhibitor-bound µI-II structure. This can
be attributed to the lack of DIII, which would provide a site of interaction to hold the
residue in place in the whole enzyme. In addition, the N-terminal region of SNJ 1945 was
crystallized in two conformations, suggesting flexibility that may be restricted by DIII in
the full-length enzyme [62].
These methodologies can also be extended to other calpain isoforms if they can be
produced as active enzymes. Unfortunately, these isoforms are extremely difficult to
produce and purify. Calpain 3 is a case in point where the insertion sequences seem to
74
predispose the enzyme to autoproteolysis [115]. Our laboratory tried to express just the
protease core of calpain 3 (containing insertion sequence 1) but it was subject to
inexorable intramolecular autoproteolysis even in the absence of exogenous calcium
[116]. Because of these problems, it may be useful to produce pseudo-full-length
calpains. That is, the protease cores of these isoforms could potentially be inserted into
DIII-DVI of m-calpain. In this way, m-calpain would act as a surrogate stabilizer for
profiling and structure-based inhibitor design of the non-classical calpain isoforms. This
may lead to the development of isoform-specific calpain substrates and inhibitors that
will help clarify their cellular roles.
5.3 Applications of Allosteric Inhibitor Design
Much of this thesis has focused on the determination of the active-site specificity
of calpain isoforms. To date, the search for calpain-specific inhibitors targeting the active
site has been relatively unsuccessful, mostly due to cross-reactivity with other cysteine
proteases. The stabilization of full-length calpain by the CAST peptides proposed in
chapter 4 of this thesis may provide the necessary advantage to begin inhibitor screening
projects. In particular, the development of inhibitors that inactivate calpain allosterically
would be of significant pharmacological interest as they would diversify the available
compounds and be less likely to cross-react with other proteases.
Examples of inhibitors that bind distal to the active site include the novel caspase-
1, -3 and -7 inhibitors revealed by disulfide trapping, a fragment-based drug discovery
approach [117,118]. These compounds interact with a deep cavity at the dimerization
75
interface 15 Å away from the active site and lock the enzymes in an inactive zymogen-
like conformation. Similarly, recent structural data have revealed an allosteric inhibitor of
the caspase 2 isoform that covalently binds to the dimer interface resulting in slight
structural changes that promote a zymogen-like conformation, including a displacement
of 1 Å of the active-site cysteine [119].
The development of calpain-specific allosteric inhibitors has been minimal to
date. Two compounds, PD150606 and PD151746, are examples of such inhibitors that
target the calcium-binding PEF domains of calpain [120]. PD150606 was determined to
be an allosteric inhibitor as it did not block or compete with substrate binding at the
active site [120]. X-ray crystallographic analysis demonstrated that PD150606 interacts
with the same hydrophobic patch in DVI of calpain as the CAST C peptide [69]. Both
inhibitors were specific for µ- and m-calpains over other proteases [120].
It is important to point out that in this thesis neither the C peptide alone nor the C-
A peptides have been shown to alter calpain activity. Two separate groups have published
contradictory evidence that interaction at the hydrophobic pocket occupied by the CAST
C peptide affects the activity of calpain. As mentioned above, Wang et al. have proposed
that PD150606 inhibits calpain through interaction at this site [120]. In addition, Tompa
et al. have postulated that the CAST C peptide acts as an activator of calpain [121]. It is
difficult to resolve these hypotheses as they were neither evident in our results nor have
they been reported elsewhere. With respect to the PD150606 inhibitor, it was crystallized
with DVI of calpain alone. In order to verify that the inhibitor does not act on another
76
allosteric site within calpain, it should be crystallized with the full-length enzyme. In
addition, a FRET assay can be done with the protease core of calpain to determine
whether it can inhibit at a site within DI and DII.
The isoform-specific caspase allosteric inhibitors described above demonstrate the
importance of developing compounds that have an indirect mechanism of action through
structural rearrangements. With the limited specificity achieved thus far with active-site
inhibitors of calpain, probing of allosteric compounds may reduce cross-reactivity with
other cysteine proteases. Since calpains are activated by calcium binding to the protease
core domains DI and DII as well as the PEF domains DIV and DVI, targeting these sites
in the calcium-free form might prevent calcium binding and therefore calpain activation.
With the recent stabilization of full-length calcium-bound calpain by CAST C-A,
a screening approach can be attempted to isolate compounds and/or proteins that interact
with calpain allosterically. This may lead to the development of calpain- and isoform-
specific allosteric inhibitors for pharmacological drug design as well as possible calpain
binding partners in an attempt to clarify their cellular roles.
5.4 Summary and Conclusions
1. The mini-calpains 3, 8 and 9 were determined to have much lower activities than µI-II
when using inactive m-calpain (mC105S) as a substrate. In the same experiment,
mini-calpain 15 could not digest mC105S, highlighting the inactivity of this isoform.
77
2. The activity of mini-calpain 8 was improved by the mutation of Gly203 to an Ala,
which is proposed to stabilize a helix in the protease core and prevent Trp106 from
interrupting the active site.
3. Mini-calpain 8 was shown to have similar preferences as µI-II and mI-II for small
aliphatic amino acids at the P2 position.
4. The site of calpain-mediated cleavage of the FRET substrate (EDANS)-EPLFMER-
(DABCYL) between the F and M residues was confirmed using mass spectrometry.
5. The FRET substrate, (EDANS)-EPLFAERK-(DABCYL) was further optimized to
(EDANS)-EPLFAARK-(DABCYL), which was turned-over by µ-calpain 2.3-fold
more rapidly.
6. Peptides engineered from the endogenous inhibitor of calpain, CAST, were shown to
inhibit the aggregation of full-length m-calpain in the presence of calcium. The
peptides were cloned from the C sub-domain of one CAST domain to the A sub-
domain of the next, producing three peptides (CAST 1C-2A, 2C-3A and 3C-4A).
CAST 3C-4A was revealed through turbidity analysis to inhibit calpain aggregation at
a 1:1 molar ratio of peptide to enzyme.
7. The CAST C-A peptides did not effect either the activity or autolysis of calpain but
were shown to be cleaved by calpain in the absence of an active-site inhibitor.
78
References
1. Ringer,S. (1883). A further Contribution regarding the influence of the different Constituents of the Blood on the Contraction of the Heart. J. Physiol 4, 29-42.
2. Verkhratsky,A. and Toescu, E. C. (2003). Endoplasmic reticulum Ca(2+) homeostasis and neuronal death. J. Cell Mol. Med. 7, 351-361.
3. Paschen,W. and Mengesdorf, T. (2005). Endoplasmic reticulum stress response and neurodegeneration. Cell Calcium 38, 409-415.
4. Leist,M. and Nicotera, P. (1998). Apoptosis versus necrosis: the shape of neuronal cell death. Results Probl. Cell Differ. 24, 105-135.
5. Berliocchi,L., Bano, D., and Nicotera, P. (2005). Ca2+ signals and death programmes in neurons. Philos. Trans. R. Soc. Lond B Biol. Sci. 360, 2255-2258.
6. Brehm,P. and Eckert, R. (1978). Calcium entry leads to inactivation of calcium channel in Paramecium. Science 202, 1203-1206.
7. Marban,E. and Tsien, R. W. (1982). Enhancement of calcium current during digitalis inotropy in mammalian heart: positive feed-back regulation by intracellular calcium? J. Physiol 329, 589-614.
8. Spitzer,N.C. (2008). Calcium: first messenger. Nat. Neurosci. 11, 243-244. 9. Carafoli,E. (2005). Calcium--a universal carrier of biological signals. Delivered on 3 July
2003 at the Special FEBS Meeting in Brussels. FEBS J. 272, 1073-1089.
79
10. Murachi,T., Tanaka, K., Hatanaka, M., and Murakami, T. (1980). Intracellular Ca2+-dependent protease (calpain) and its high-molecular-weight endogenous inhibitor (calpastatin). Adv. Enzyme Regul. 19, 407-424.
11. Barret,A.J., Rawlings, N. D., and Woessner, J. F. (1998). Hanbook of Proteolytic Enzymes.
12. GUROFF,G. (1964). A neutral, calciu-activated proteinase from the soluble fraction of rat brain. J. Biol. Chem. 239, 149-155.
13. Dayton,W.R., Goll, D. E., Zeece, M. G., Robson, R. M., and Reville, W. J. (1976). A Ca2+-activated protease possibly involved in myofibrillar protein turnover. Purification from porcine muscle. Biochemistry 15, 2150-2158.
14. Goll,D.E., Thompson, V. F., Li, H., Wei, W., and Cong, J. (2003). The calpain system. Physiol Rev. 83, 731-801.
15. Glading,A., Lauffenburger, D. A., and Wells, A. (2002). Cutting to the chase: calpain proteases in cell motility. Trends Cell Biol. 12, 46-54.
16. Janossy,J., Ubezio, P., Apati, A., Magocsi, M., Tompa, P., and Friedrich, P. (2004). Calpain as a multi-site regulator of cell cycle. Biochem. Pharmacol. 67, 1513-1521.
17. Neumar,R.W., Xu, Y. A., Gada, H., Guttmann, R. P., and Siman, R. (2003). Cross-talk between calpain and caspase proteolytic systems during neuronal apoptosis. J.
Biol. Chem. 278, 14162-14167. 18. Dutt,P., Croall, D. E., Arthur, J. S., Veyra, T. D., Williams, K., Elce, J. S., and Greer, P.
A. (2006). m-Calpain is required for preimplantation embryonic development in mice. BMC. Dev. Biol. 6, 3-
19. Zimmerman,U.J., Boring, L., Pak, J. H., Mukerjee, N., and Wang, K. K. (2000). The calpain small subunit gene is essential: its inactivation results in embryonic lethality. IUBMB. Life 50, 63-68.
20. Arthur,J.S., Elce, J. S., Hegadorn, C., Williams, K., and Greer, P. A. (2000). Disruption of the murine calpain small subunit gene, Capn4: calpain is essential for embryonic development but not for cell growth and division. Mol. Cell Biol. 20, 4474-4481.
21. Grammer,M., Kuchay, S., Chishti, A., and Baudry, M. (2005). Lack of phenotype for LTP and fear conditioning learning in calpain 1 knock-out mice. Neurobiol.
Learn. Mem. 84, 222-227. 22. Azam,M., Andrabi, S. S., Sahr, K. E., Kamath, L., Kuliopulos, A., and Chishti, A. H.
(2001). Disruption of the mouse mu-calpain gene reveals an essential role in platelet function. Mol. Cell Biol. 21, 2213-2220.
23. Badugu,R., Garcia, M., Bondada, V., Joshi, A., and Geddes, J. W. (2008). N terminus of calpain 1 is a mitochondrial targeting sequence. J. Biol. Chem. 283, 3409-3417.
24. Garcia,M., Bondada, V., and Geddes, J. W. (2005). Mitochondrial localization of mu-calpain. Biochem. Biophys. Res. Commun. 338, 1241-1247.
25. Kramerova,I., Kudryashova, E., Tidball, J. G., and Spencer, M. J. (2004). Null mutation of calpain 3 (p94) in mice causes abnormal sarcomere formation in vivo and in vitro. Hum. Mol. Genet. 13, 1373-1388.
26. Bevers,M.B. and Neumar, R. W. (2008). Mechanistic role of calpains in postischemic neurodegeneration. J. Cereb. Blood Flow Metab 28, 655-673.
80
27. Saito,K., Elce, J. S., Hamos, J. E., and Nixon, R. A. (1993). Widespread activation of calcium-activated neutral proteinase (calpain) in the brain in Alzheimer disease: a potential molecular basis for neuronal degeneration. Proc. Natl. Acad. Sci. U. S.
A 90, 2628-2632. 28. Nixon,R.A., Saito, K. I., Grynspan, F., Griffin, W. R., Katayama, S., Honda, T., Mohan,
P. S., Shea, T. B., and Beermann, M. (1994). Calcium-activated neutral proteinase (calpain) system in aging and Alzheimer's disease. Ann. N. Y. Acad.
Sci. 747, 77-91. 29. Crocker,S.J., Smith, P. D., Jackson-Lewis, V., Lamba, W. R., Hayley, S. P., Grimm, E.,
Callaghan, S. M., Slack, R. S., Melloni, E., Przedborski, S., Robertson, G. S., Anisman, H., Merali, Z., and Park, D. S. (2003). Inhibition of calpains prevents neuronal and behavioral deficits in an MPTP mouse model of Parkinson's disease. J. Neurosci. 23, 4081-4091.
30. Gafni,J., Hermel, E., Young, J. E., Wellington, C. L., Hayden, M. R., and Ellerby, L. M. (2004). Inhibition of calpain cleavage of huntingtin reduces toxicity: accumulation of calpain/caspase fragments in the nucleus. J. Biol. Chem. 279, 20211-20220.
31. Vanderklish,P.W. and Bahr, B. A. (2000). The pathogenic activation of calpain: a marker and mediator of cellular toxicity and disease states. Int. J. Exp. Pathol. 81, 323-339.
32. Camins,A., Verdaguer, E., Folch, J., and Pallas, M. (2006). Involvement of calpain activation in neurodegenerative processes. CNS. Drug Rev. 12, 135-148.
33. Lee,M.S., Kwon, Y. T., Li, M., Peng, J., Friedlander, R. M., and Tsai, L. H. (2000). Neurotoxicity induces cleavage of p35 to p25 by calpain. Nature 405, 360-364.
34. Patrick,G.N., Zukerberg, L., Nikolic, M., de la, Monte S., Dikkes, P., and Tsai, L. H. (1999). Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Nature 402, 615-622.
35. Badugu,R., Garcia, M., Bondada, V., Joshi, A., and Geddes, J. W. (2008). N terminus of calpain 1 is a mitochondrial targeting sequence. J. Biol. Chem. 283, 3409-3417.
36. Berti,P.J. and Storer, A. C. (1995). Alignment/phylogeny of the papain superfamily of cysteine proteases. J. Mol. Biol. 246, 273-283.
37. Hosfield,C.M., Ye, Q., Arthur, J. S., Hegadorn, C., Croall, D. E., Elce, J. S., and Jia, Z. (1999). Crystallization and X-ray crystallographic analysis of m-calpain, a Ca2+-dependent protease. Acta Crystallogr. D. Biol. Crystallogr. 55, 1484-1486.
38. Sutton,R.B., Davletov, B. A., Berghuis, A. M., Sudhof, T. C., and Sprang, S. R. (1995). Structure of the first C2 domain of synaptotagmin I: a novel Ca2+/phospholipid-binding fold. Cell 80, 929-938.
39. Blanchard,H., Grochulski, P., Li, Y., Arthur, J. S., Davies, P. L., Elce, J. S., and Cygler, M. (1997). Structure of a calpain Ca(2+)-binding domain reveals a novel EF-hand and Ca(2+)-induced conformational changes. Nat. Struct. Biol. 4, 532-538.
40. Lin,G.D., Chattopadhyay, D., Maki, M., Wang, K. K., Carson, M., Jin, L., Yuen, P. W., Takano, E., Hatanaka, M., DeLucas, L. J., and Narayana, S. V. (1997). Crystal structure of calcium bound domain VI of calpain at 1.9 A resolution and its role in enzyme assembly, regulation, and inhibitor binding. Nat. Struct. Biol. 4, 539-547.
81
41. Ravulapalli,R., Diaz, B. G., Campbell, R. L., and Davies, P. L. (2005). Homodimerization of calpain 3 penta-EF-hand domain. Biochem. J. 388, 585-591.
42. Strobl,S., Fernandez-Catalan, C., Braun, M., Huber, R., Masumoto, H., Nakagawa, K., Irie, A., Sorimachi, H., Bourenkow, G., Bartunik, H., Suzuki, K., and Bode, W. (2000). The crystal structure of calcium-free human m-calpain suggests an electrostatic switch mechanism for activation by calcium. Proc. Natl. Acad. Sci.
U. S. A 97, 588-592. 43. Moldoveanu,T., Campbell, R. L., Cuerrier, D., and Davies, P. L. (2004). Crystal
structures of calpain-E64 and -leupeptin inhibitor complexes reveal mobile loops gating the active site. J. Mol. Biol. 343, 1313-1326.
44. Moldoveanu,T., Hosfield, C. M., Lim, D., Elce, J. S., Jia, Z., and Davies, P. L. (2002). A Ca(2+) switch aligns the active site of calpain. Cell 108, 649-660.
45. Tompa,P., Emori, Y., Sorimachi, H., Suzuki, K., and Friedrich, P. (2001). Domain III of calpain is a ca2+-regulated phospholipid-binding domain. Biochem. Biophys.
Res. Commun. 280, 1333-1339. 46. Hanna,R., Campbell, R. L., and Davies, P. L (2008). Calcium-bound structure of calpain
and its mechanism of inhibition by the endogenous inhibitor, calpastatin. Nature.
In Press
47. Blanchard,H., Grochulski, P., Li, Y., Arthur, J. S., Davies, P. L., Elce, J. S., and Cygler, M. (1997). Structure of a calpain Ca(2+)-binding domain reveals a novel EF-hand and Ca(2+)-induced conformational changes. Nat. Struct. Biol. 4, 532-538.
48. Lin,G.D., Chattopadhyay, D., Maki, M., Wang, K. K., Carson, M., Jin, L., Yuen, P. W., Takano, E., Hatanaka, M., DeLucas, L. J., and Narayana, S. V. (1997). Crystal structure of calcium bound domain VI of calpain at 1.9 A resolution and its role in enzyme assembly, regulation, and inhibitor binding. Nat. Struct. Biol. 4, 539-547.
49. Suzuki,K., Hata, S., Kawabata, Y., and Sorimachi, H. (2004). Structure, activation, and biology of calpain. Diabetes 53 Suppl 1, S12-S18.
50. Garcia Diaz,B.E., Gauthier, S., and Davies, P. L. (2006). Ca2+ dependency of calpain 3 (p94) activation. Biochemistry 45, 3714-3722.
51. Sorimachi,H., Toyama-Sorimachi, N., Saido, T. C., Kawasaki, H., Sugita, H., Miyasaka, M., Arahata, K., Ishiura, S., and Suzuki, K. (1993). Muscle-specific calpain, p94, is degraded by autolysis immediately after translation, resulting in disappearance from muscle. J. Biol. Chem. 268, 10593-10605.
52. Sorimachi,H., Kinbara, K., Kimura, S., Takahashi, M., Ishiura, S., Sasagawa, N., Sorimachi, N., Shimada, H., Tagawa, K., Maruyama, K., and . (1995). Muscle-specific calpain, p94, responsible for limb girdle muscular dystrophy type 2A, associates with connectin through IS2, a p94-specific sequence. J. Biol. Chem.
270, 31158-31162. 53. Sorimachi,H., Imajoh-Ohmi, S., Emori, Y., Kawasaki, H., Ohno, S., Minami, Y., and
Suzuki, K. (1989). Molecular cloning of a novel mammalian calcium-dependent protease distinct from both m- and mu-types. Specific expression of the mRNA in skeletal muscle. J. Biol. Chem. 264, 20106-20111.
54. Richard,I., Broux, O., Allamand, V., Fougerousse, F., Chiannilkulchai, N., Bourg, N., Brenguier, L., Devaud, C., Pasturaud, P., Roudaut, C., and . (1995). Mutations in
82
the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 81, 27-40.
55. Sorimachi, H. and Beckmann, J. S. 2002. Defects of non-lysosomal proteolysis: calpain 3 deficiency.In: Structural and molecular basis of skeletal muscle diseases, edited by Karpati, G. Basel, Switzerland Neuropath Press, 148-153.
56. Hata,S., Koyama, S., Kawahara, H., Doi, N., Maeda, T., Toyama-Sorimachi, N., Abe, K., Suzuki, K., and Sorimachi, H. (2005). Stomach-specific Calpain, nCL-2. Localizes in Mucus Cells and Proteolyzes the Beta-Subunit of Coatomer Complex, Beta-COP. The Journal of Biological Chemistry 281, 11214-11224.
57. Yoshikawa,Y., Mukai, H., Hino, F., Asada, K., and Kato, I. (2000). Isolation of two novel genes, down-regulated in gastric cancer. Jpn. J. Cancer Res. 91, 459-463.
58. Liu,K., Li, L., and Cohen, S. N. (2000). Antisense RNA-mediated deficiency of the calpain protease, nCL-4, in NIH3T3 cells is associated with neoplastic transformation and tumorigenesis. J. Biol. Chem. 275, 31093-31098.
59. Dear,T.N., Moller, A., and Boehm, T. (1999). CAPN11: A calpain with high mRNA levels in testis and located on chromosome 6. Genomics 59, 243-247.
60. Dear,T.N., Meier, N. T., Hunn, M., and Boehm, T. (2000). Gene structure, chromosomal localization, and expression pattern of Capn12, a new member of the calpain large subunit gene family. Genomics 68, 152-160.
61. Qian,J., Cuerrier, D., Davies, P. L., Li, Z., Powers, J. C., and Campbell, R. L. (2008). Cocrystal Structures of Primed Side-Extending alpha-Ketoamide Inhibitors Reveal Novel Calpain-Inhibitor Aromatic Interactions. J. Med. Chem.
62. Cuerrier,D., Moldoveanu, T., Inoue, J., Davies, P. L., and Campbell, R. L. (2006). Calpain inhibition by alpha-ketoamide and cyclic hemiacetal inhibitors revealed by X-ray crystallography. Biochemistry 45, 7446-7452.
63. Cuerrier,D., Moldoveanu, T., and Davies, P. L. (2005). Determination of peptide substrate specificity for mu-calpain by a peptide library-based approach: the importance of primed side interactions. J. Biol. Chem. 280, 40632-40641.
64. Cuerrier,D., Moldoveanu, T., Campbell, R. L., Kelly, J., Yoruk, B., Verhelst, S. H., Greenbaum, D., Bogyo, M., and Davies, P. L. (2007). Development of calpain-specific inactivators by screening of positional scanning epoxide libraries. J.
Biol. Chem. 282, 9600-9611. 65. Turk,B.E., Huang, L. L., Piro, E. T., and Cantley, L. C. (2001). Determination of protease
cleavage site motifs using mixture-based oriented peptide libraries. Nat.
Biotechnol. 19, 661-667. 66. Schechter,I. and Berger, A. (1967). On the size of the active site in proteases. I. Papain.
Biochem. Biophys. Res. Commun. 27, 157-162. 67. Li,H., Thompson, V. F., and Goll, D. E. (2004). Effects of autolysis on properties of mu-
and m-calpain. Biochim. Biophys. Acta 1691, 91-103. 68. Zimmerman,U.J. and Schlaepfer, W. W. (1991). Two-stage autolysis of the catalytic
subunit initiates activation of calpain I. Biochim. Biophys. Acta 1078, 192-198. 69. Todd,B., Moore, D., Deivanayagam, C. C., Lin, G. D., Chattopadhyay, D., Maki, M.,
Wang, K. K., and Narayana, S. V. (2003). A structural model for the inhibition of calpain by calpastatin: crystal structures of the native domain VI of calpain and its complexes with calpastatin peptide and a small molecule inhibitor. J. Mol.
Biol. 328, 131-146.
83
70. Pal,G.P., Elce, J. S., and Jia, Z. (2001). Dissociation and aggregation of calpain in the presence of calcium. J. Biol. Chem. 276, 47233-47238.
71. Okitani,A., Goll, D. E., Stromer, M. H., and Robson, R. M. (1976). Intracellular inhibitor of Ca2+-activated protease involved in myofibrillar protein turnover. Federation
Proceedings 35, 1746- 72. Murachi,T. (1989). Intracellular regulatory system involving calpain and calpastatin.
Biochem. Int. 18, 263-294. 73. Emori,Y., Kawasaki, H., Imajoh, S., Imahori, K., and Suzuki, K. (1987). Endogenous
inhibitor for calcium-dependent cysteine protease contains four internal repeats that could be responsible for its multiple reactive sites. Proc. Natl. Acad. Sci. U.
S. A 84, 3590-3594. 74. Emori,Y., Kawasaki, H., Imajoh, S., Minami, Y., and Suzuki, K. (1988). All four
repeating domains of the endogenous inhibitor for calcium-dependent protease independently retain inhibitory activity. Expression of the cDNA fragments in Escherichia coli. J. Biol. Chem. 263, 2364-2370.
75. Maki,M., Takano, E., Mori, H., Sato, A., Murachi, T., and Hatanaka, M. (1987). All four internally repetitive domains of pig calpastatin possess inhibitory activities against calpains I and II. FEBS Lett. 223, 174-180.
76. Takano, E., Masatoshi, M., and Yuen, P. 1999. Structure of calpastatin and its inhibitory control of calpain.In: Calpain: pharmacology and toxicology of calcium-
dependent protease, edited by Wang, K. K. W and Yuen, P. W. Philadelphia 25-50.
77. Bokor,M., Csizmok, V., Kovacs, D., Banki, P., Friedrich, P., Tompa, P., and Tompa, K. (2005). NMR relaxation studies on the hydrate layer of intrinsically unstructured proteins. Biophys. J. 88, 2030-2037.
78. Imajoh,S., Kawasaki, H., Emori, Y., and Suzuki, K. (1987). Calcium-activated neutral protease inhibitor from rabbit erythrocytes lacks the N-terminal region of the liver inhibitor but retains three inhibitory units. Biochem. Biophys. Res. Commun.
146, 630-637. 79. Wang,L.F., Wei, S. G., Miao, S. Y., Liu, Q. Y., and Koide, S. S. (1994). Calpastatin gene
in human testis. Biochem. Mol. Biol. Int. 33, 245-251. 80. Cong,M., Thompson, V. F., Goll, D. E., and Antin, P. B. (1998). The bovine calpastatin
gene promoter and a new N-terminal region of the protein are targets for cAMP-dependent protein kinase activity. J. Biol. Chem. 273, 660-666.
81. Lee,W.J., Ma, H., Takano, E., Yang, H. Q., Hatanaka, M., and Maki, M. (1992). Molecular diversity in amino-terminal domains of human calpastatin by exon skipping. J. Biol. Chem. 267, 8437-8442.
82. Takano,J., Kawamura, T., Murase, M., Hitomi, K., and Maki, M. (1999). Structure of mouse calpastatin isoforms: implications of species-common and species-specific alternative splicing. Biochem. Biophys. Res. Commun. 260, 339-345.
83. Maki,M., Takano, E., Osawa, T., Ooi, T., Murachi, T., and Hatanaka, M. (1988). Analysis of structure-function relationship of pig calpastatin by expression of mutated cDNAs in Escherichia coli. J. Biol. Chem. 263, 10254-10261.
84. Uemori,T., Shimojo, T., Asada, K., Asano, T., Kimizuka, F., Kato, I., Maki, M., Hatanaka, M., Murachi, T., Hanzawa, H., and . (1990). Characterization of a
84
functional domain of human calpastatin. Biochem. Biophys. Res. Commun. 166, 1485-1493.
85. Yang,H.Q., Ma, H., Takano, E., Hatanaka, M., and Maki, M. (1994). Analysis of calcium-dependent interaction between amino-terminal conserved region of calpastatin functional domain and calmodulin-like domain of mu-calpain large subunit. J. Biol. Chem. 269, 18977-18984.
86. Takano,E., Ma, H., Yang, H. Q., Maki, M., and Hatanaka, M. (1995). Preference of calcium-dependent interactions between calmodulin-like domains of calpain and calpastatin subdomains. FEBS Lett. 362, 93-97.
87. Hanna,R.A., Garcia-Diaz, B. E., and Davies, P. L. (2007). Calpastatin simultaneously binds four calpains with different kinetic constants. FEBS Lett. 581, 2894-2898.
88. Kawasaki,H., Emori, Y., Imajoh-Ohmi, S., Minami, Y., and Suzuki, K. (1989). Identification and characterization of inhibitory sequences in four repeating domains of the endogenous inhibitor for calcium-dependent protease. J. Biochem.
106, 274-281. 89. Maki,M., Takano, E., Osawa, T., Ooi, T., Murachi, T., and Hatanaka, M. (1988).
Analysis of structure-function relationship of pig calpastatin by expression of mutated cDNAs in Escherichia coli. J. Biol. Chem. 263, 10254-10261.
90. Croall,D.E. and McGrody, K. S. (1994). Domain structure of calpain: mapping the binding site for calpastatin. Biochemistry 33, 13223-13230.
91. Sorimachi,H. and Suzuki, K. (2001). The structure of calpain. J. Biochem. 129, 653-664. 92. Sorimachi,H., Ishiura, S., and Suzuki, K. (1993). A novel tissue-specific calpain species
expressed predominantly in the stomach comprises two alternative splicing products with and without Ca(2+)-binding domain. J. Biol. Chem. 268, 19476-19482.
93. Lee,H.J., Sorimachi, H., Jeong, S. Y., Ishiura, S., and Suzuki, K. (1998). Molecular cloning and characterization of a novel tissue-specific calpain predominantly expressed in the digestive tract. Biol. Chem. 379, 175-183.
94. Lee,H.J., Tomioka, S., Kinbara, K., Masumoto, H., Jeong, S. Y., Sorimachi, H., Ishiura, S., and Suzuki, K. (1999). Characterization of a human digestive tract-specific calpain, nCL-4, expressed in the baculovirus system. Arch. Biochem. Biophys.
362, 22-31. 95. Kamei,M., Webb, G. C., Heydon, K., Hendry, I. A., Young, I. G., and Campbell, H. D.
(2000). Solh, the mouse homologue of the Drosophila melanogaster small optic lobes gene: organization, chromosomal mapping, and localization of gene product to the olfactory bulb. Genomics 64, 82-89.
96. Greenbaum,D., Medzihradszky, K. F., Burlingame, A., and Bogyo, M. (2000). Epoxide electrophiles as activity-dependent cysteine protease profiling and discovery tools. Chem. Biol. 7, 569-581.
97. Moldoveanu,T., Hosfield, C. M., Lim, D., Jia, Z., and Davies, P. L. (2003). Calpain silencing by a reversible intrinsic mechanism. Nat. Struct. Biol. 10, 371-378.
98. Cuerrier,D., Moldoveanu, T., and Davies, P. L. (2005). Determination of peptide substrate specificity for mu-calpain by a peptide library-based approach: the importance of primed side interactions. J. Biol. Chem. 280, 40632-40641.
85
99. Croce,K., Flaumenhaft, R., Rivers, M., Furie, B., Furie, B. C., Herman, I. M., and Potter, D. A. (1999). Inhibition of calpain blocks platelet secretion, aggregation, and spreading. J. Biol. Chem. 274, 36321-36327.
100. Tompa,P., Buzder-Lantos, P., Tantos, A., Farkas, A., Szilagyi, A., Banoczi, Z., Hudecz, F., and Friedrich, P. (2004). On the sequential determinants of calpain cleavage. J. Biol. Chem. 279, 20775-20785.
101. Johnson,G.V., Jope, R. S., and Binder, L. I. (1989). Proteolysis of tau by calpain. Biochem. Biophys. Res. Commun. 163, 1505-1511.
102. Hayashi,M., Suzuki, H., Kawashima, S., Saido, T. C., and Inomata, M. (1999). The behavior of calpain-generated N- and C-terminal fragments of talin in integrin-mediated signaling pathways. Arch. Biochem. Biophys. 371, 133-141.
103. Du,X., Saido, T. C., Tsubuki, S., Indig, F. E., Williams, M. J., and Ginsberg, M. H. (1995). Calpain cleavage of the cytoplasmic domain of the integrin beta 3 subunit. J. Biol. Chem. 270, 26146-26151.
104. Sasaki,T., Kikuchi, T., Yumoto, N., Yoshimura, N., and Murachi, T. (1984). Comparative specificity and kinetic studies on porcine calpain I and calpain II with naturally occurring peptides and synthetic fluorogenic substrates. J. Biol. Chem. 259, 12489-12494.
105. Sasaki,T., Kikuchi, T., Fukui, I., and Murachi, T. (1986). Inactivation of calpain I and calpain II by specificity-oriented tripeptidyl chloromethyl ketones. J. Biochem.
99, 173-179. 106. Polster,B.M., Arze, R., Lyttle, M. H., Nicholls, D. G., and Hudson, D. (2007). Solid
Phase Synthesis of Dual Labeled Peptides: Development of Cell Permeable Calpain Specific Substrates. International Journal of Peptide Research and
Therapeutics 13, 83-91. 107. Liu,Y., Kati, W, Chen, C, Tripathi, R, Molla, A, and Kohlbrenner, W (1999). Use of a
Fluorescence Plate Reader for Measuring Kinetic Parameters with Inner Filter Effect Correction. Analytical Biochemistry 267, 331-335.
108. Davis,T.L., Walker, J. R., Finerty, P. J., Jr., Mackenzie, F., Newman, E. M., and Dhe-Paganon, S. (2007). The crystal structures of human calpains 1 and 9 imply diverse mechanisms of action and auto-inhibition. J. Mol. Biol. 366, 216-229.
109. Saez,M.E., Ramirez-Lorca, R., Moron, F. J., and Ruiz, A. (2006). The therapeutic potential of the calpain family: new aspects. Drug Discov. Today 11, 917-923.
110. Cottin,P., Vidalenc, P. L., and Ducastaing, A. (1981). Ca2+-dependent association between a Ca2+-activated neutral proteinase (CaANP) and its specific inhibitor. FEBS Lett. 136, 221-224.
111. Mucsi,Z., Hudecz, F., Hollosi, M., Tompa, P., and Friedrich, P. (2003). Binding-induced folding transitions in calpastatin subdomains A and C. Protein Sci. 12, 2327-2336.
112. Yoshizawa,T., Sorimachi, H., Tomioka, S., Ishiura, S., and Suzuki, K. (1995). Calpain dissociates into subunits in the presence of calcium ions. Biochem. Biophys. Res.
Commun. 208, 376-383. 113. Elce,J.S., Hegadorn, C., Gauthier, S., Vince, J. W., and Davies, P. L. (1995).
Recombinant calpain II: improved expression systems and production of a C105A active-site mutant for crystallography. Protein Eng 8, 843-848.
86
114. Crawford,C., Brown, N. R., and Willis, A. C. (1993). Studies of the active site of m-calpain and the interaction with calpastatin. Biochem. J. 296 ( Pt 1), 135-142.
115. Sorimachi,H., Toyama-Sorimachi, N., Saido, T. C., Kawasaki, H., Sugita, H., Miyasaka, M., Arahata, K., Ishiura, S., and Suzuki, K. (1993). Muscle-specific calpain, p94, is degraded by autolysis immediately after translation, resulting in disappearance from muscle. J. Biol. Chem. 268, 10593-10605.
116. Garcia Diaz,B.E., Gauthier, S., and Davies, P. L. (2006). Ca2+ dependency of calpain 3 (p94) activation. Biochemistry 45, 3714-3722.
117. Hardy,J.A., Lam, J., Nguyen, J. T., O'Brien, T., and Wells, J. A. (2004). Discovery of an allosteric site in the caspases. Proc. Natl. Acad. Sci. U. S. A 101, 12461-12466.
118. Scheer,J.M., Romanowski, M. J., and Wells, J. A. (2006). A common allosteric site and mechanism in caspases. Proc. Natl. Acad. Sci. U. S. A 103, 7595-7600.
119. Schweizer,A., Roschitzki-Voser, H., Amstutz, P., Briand, C., Gulotti-Georgieva, M., Prenosil, E., Binz, H. K., Capitani, G., Baici, A., Pluckthun, A., and Grutter, M. G. (2007). Inhibition of caspase-2 by a designed ankyrin repeat protein: specificity, structure, and inhibition mechanism. Structure. 15, 625-636.
120. Wang,K.K., Nath, R., Posner, A., Raser, K. J., Buroker-Kilgore, M., Hajimohammadreza, I., Probert, A. W., Jr., Marcoux, F. W., Ye, Q., Takano, E., Hatanaka, M., Maki, M., Caner, H., Collins, J. L., Fergus, A., Lee, K. S., Lunney, E. A., Hays, S. J., and Yuen, P. (1996). An alpha-mercaptoacrylic acid derivative is a selective nonpeptide cell-permeable calpain inhibitor and is neuroprotective. Proc. Natl.
Acad. Sci. U. S. A 93, 6687-6692. 121. Tompa,P., Mucsi, Z., Orosz, G., and Friedrich, P. (2002). Calpastatin subdomains A and
C are activators of calpain. J. Biol. Chem. 277, 9022-9026.