Crystal structure of calpain-3 penta-EF-hand (PEF) domain – a homodimerized PEF family member with calcium bound at the fifth EF-hand Sarathy K. Partha*, Ravikiran Ravulapalli†, John S. Allingham, Robert L. Campbell and Peter L. Davies Department of Biomedical and Molecular Sciences, Queen’s University, Kingston, Ontario, Canada

Keywords calcium coordination; calpain; musclespecific protease; muscular dystrophy; penta-EF-hand domain Correspondence P. L. Davies, Department of Biomedical and Molecular Sciences, Queen’s University, Kingston, Ontario, K7L 3N6 Canada Fax: +613-533-2497 Tel: +613-533-2983 E-mail: [email protected] Present addresses: *Department of Microbiology, Immunology and Infectious Diseases, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta T2N 4N1, Canada † Department of Molecular and Cellular Biology, University of Guelph, Guelph, Ontario N1G 2W1, Canada (Received 24 January 2014, revised 7 April 2014, accepted 15 May 2014) doi:10.1111/febs.12849

Calpains are Ca2+dependent intracellular cysteine proteases that cleave a wide range of protein substrates to help implement Ca2+ signaling in the cell. The major isoforms of this enzyme family, calpain-1 and calpain-2, are heterodimers of a large and a small subunit, with the main dimer interface being formed through their C-terminal penta-EF hand (PEF) domains. Calpain-3, or p94, is a skeletal muscle-specific isoform that is genetically linked to limb-girdle muscular dystrophy. Biophysical and modeling studies with the PEF domain of calpain-3 support the suggestion that full-length calpain-3 exists as a homodimer. Here, we report the crystallization of calpain-30 s PEF domain and its crystal structure in the presence of Ca2+, which provides evidence for the homodimer architecture of calpain-3 and supports the molecular model that places a protease core at either end of the elongated dimer. Unlike other calpain PEF domain structures, the calpain-3 PEF domain contains a Ca2+ bound at the EF5-hand used for homodimer association. Three of the four Ca2+-binding EF-hands of the PEF domains are concentrated near the protease core, and have the potential to radically change the local charge within the dimer during Ca2+ signaling. Examination of the homodimer interface shows that there would be steric clashes if the calpain-3 large subunit were to try to pair with a calpain small subunit. Database Structural data are available in the Protein Data Bank database under accession number 4OKH. Structured digital abstract  Calpain-3 and Calpain-3 bind by x-ray crystallography (View interaction)

Introduction Calpains constitute a widely dispersed family of Ca2+dependent cytosolic cysteine proteases. These complex enzymes are involved in a broad range of intracellular activities linked to Ca2+ signaling, which they mediate

through limited proteolysis of various target substrates [1–4]. Aberrations in calpain activity resulting from defects in Ca2+ homeostasis or mutations in a calpain gene can contribute to various disorders, including,

Abbreviations EF1, first EF-hand; EF2, second EF-hand; EF3, third EF-hand; EF4, fourth EF-hand; EF5, fifth EF-hand; IS1, insertion sequence 1; IS2, insertion sequence 2; L5, fifth EF-hand loop; MR, molecular replacement; NS, N-terminal region; PDB, Protein Data Bank; PEF, penta-EF hand; SAD, single-wavelength anomalous diffraction.

3138

FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

ischemic injury, Alzheimer’s disease, muscular dystrophy, cancer, and type 2 diabetes [5]. Within the calpain family, calpain-1 and calpain-2 are the two major isoforms, and have been well studied and characterized. They exist as heterodimers of a distinct large (80 kDa) and a common small (28 kDa) subunit [3] that both have modular architecture [6,7]. According to recent nomenclature [8,9], the large subunit comprises an anchor helix followed by the N-terminal enzyme core, composed of the protease core 1 domain and the protease core 2 domain, followed by a C2-like domain and a penta-EF hand (PEF)(L) domain. The small subunit is composed of a glycine-rich domain and a PEF(S) domain. In forming a heterodimer, the large and small subunits make numerous hydrophobic contacts, primarily through the fifth EF-hand (EF5) of the PEF domains, with contributions from the second EFhand (EF2) and the fourth EF-hand (EF4). Additional contacts are made, especially after Ca2+ activation, between the small subunit and the protease core. The presence of Ca2+ is an absolute requirement for calpain activity. Upon Ca2+ binding, calpain undergoes significant conformational changes that involve domain movements and reorganization of the catalytic cleft to cleave its substrates [10–12]. The protease core is well conserved across all members of the calpain family, whereas there are major variations in other domains. Some calpain isoforms completely lack a PEF domain (e.g. calpain-5 and calpain-6), and are unable to form a heterodimer with the small subunit. Other large-subunit isoforms that have a C-terminal PEF domain, such as calpain-3, might have the potential to form homodimers through this domain, rather than a heterodimer with the small subunit [13]. There is also the possibility of a large subunit forming heterodimers with other calpain large subunits [14]. The skeletal muscle-specific isoform, calpain-3, has been implicated in myofibrillogenesis and sarcomere remodeling [15]. Limb-girdle muscular dystrophy type 2A, a rare inherited disorder, is caused by inactivating mutations in the calpain-3 gene, and these have allowed the assignment of a key role for calpain-3 in skeletal muscle physiology [16]. The domain architecture of calpain-3 is similar to that of the large subunits of calpain-1 and calpain-2. Calpain-3 is larger, owing to the presence of two insertion sequences [insertion sequence 1 (IS1) and insertion sequence 2 (IS2)] and a longer N-terminal region (NS). The NS is 47 residues in length, and replaces the 18-residue anchor helix of calpain-1 and calpain-2. IS1 is present in the protease core 2 domain, and may serve as an internal propeptide that controls access to the active site [17]. IS2, FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

Structure of calpain-3 PEF domain homodimer

which binds to titin, is located between the C2-like domain and the PEF(L) domain [18]. Structure–function studies on calpain-3 have been hampered by its instability and rapid autolysis, which make it difficult to isolate this protein from tissues or purify it from recombinant sources [19–21]. However, the inactive C129S mutant of calpain-3 is produced as a stable protein that elutes as a dimer in size-exclusion chromatography [22]. PEF domains have the propensity to form either homodimers or heterodimers [23], and we have previously suggested that calpain-3 could form a homodimer through pairing of the C-terminal PEF domains. When the calpain-3 PEF domain is expressed as a recombinant protein in Escherichia coli, it forms a stable homodimer, much as the small-subunit PEF domain does [24]. To further investigate the nature of this homodimerization, we have solved the crystal structure of the calpain-3 PEF domain dimer in the presence of Ca2+. The structure supports computer models of the calpain-3 dimer with two active sites at opposite ends of the molecule [13]. Unexpectedly, the PEF domain binds Ca2+ at EF5, used for dimerization, and is the only member of the calpain family known to do so.

Results Crystallization and overall structure of the calpain-3 PEF domain Despite assay of a wide range of crystallization conditions, crystals of the calpain-3 PEF domain were found only at a Ca2+ concentration of 2 mM. Crystallization attempts at higher Ca2+ concentrations – up to 10 mM – resulted in either poor morphology or no crystals at all; this is unlike the crystals of the rat PEF (S) domain homodimer, which could be grown at 200 mM Ca2+ [24]. Also, although the calpain-3 PEF domain shares 31–57% sequence identity with other PEF domain structures (Fig. 1), attempts to solve the structure of the calpain-3 PEF domain by molecular replacement (MR) with the coordinates of these PEF domain proteins as search models were not successful. Ultimately, we were able to solve the structure by using the MR–single-wavelength anomalous diffraction (SAD) method of PHENIX [25], with datasets collected on La3+soaked crystals. The resulting electron density map showed three calpain-3 PEF domain chains in the asymmetric unit. Chains A and B form one homodimer unit (Fig. 2A), and chain C dimerizes with a symmetry-related molecule via a crystallographic two-fold axis. Although we could not identify Ca2+ ions in an anomalous difference map, as there was no detectable

3139

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

641 Calpain-3 CAPNS1 Calpain-2 ALG-2 Grancalcin Sorcin Calpain-13

700

670

680

710

720

730

EF1

740

750

KTHGFTLESCRSMIALMDTDGSGKLNLQE FHHLWNKI KAWQKIFKHYDTDQSGTINSYEMRN AVND.... KTDGFGIDTSRSMVAVMDSDTTGKLGFEE FKYLWNNI KKWQGIYKRFDTDRSGTIGSNELPG AFEA.... KSDGFSIETCKIMVDMLDEDGSGKLGLKE FYILWTKI QKYQKIYREIDVDRSGTMNSYEMRK ALEE.... TWTPFNPVTVRSIISMFDRENKAGVNFSE FTGVWKYI TDWQNVFRTYDRDNSGMIDKNELKQ ALSG.... TYSPFSLETCRIMIAMLDRDHTGKMGFNAFKELWAALNAWKENFMTVDQDGS GTVEHH ELRQ AIGL.... GYKPFNLETCRLMVSMLDRDMSGTMGFNEFKELWAVLNGWRQHFISF DTDRS GTVDPQ ELQKALTT.... PGDMFSLDECRSLVALMELKVNGRLDQEE FARLWKRL VHYQHVFQKVQTS.PGVLLSSDLWK AIENTDFL

760 Calpain-3 CAPNS1 Calpain-2 ALG-2 Grancalcin Sorcin Calpain-13

660

. . . . . . . . . . . . . . . . . . . . . . P G S S D Q E S E E Q Q Q F R N I F K QI A G . D D M E IC AD E L KK V L N T V V N K H K D L . . . . . . . . . . . . . . . . . . . M H Y S N I E A N E S E E E R Q F R K L F V QL A G . D D M E VS AT E L M N I L N K V V T R H P D L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G D G F R R L F A QL A G . E D A E IS AF E L Q T I L R R V L A K R E D I M A A Y S Y R P G P G G G P G P A A . . . . . . G A A L P . . D Q S F L W N V F Q RV D K D R S G V IS DN E L Q Q A L S . . . . N G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S V Y T Y F S AV A G . Q D G E VD AE E L Q R C L T Q S G I N G . . . . M A Y P G H P G A G G G Y Y P G G Y G G A P G G P A F P G Q T Q D P L Y G Y F A AV A G . Q D G Q ID AD E L Q R C L T Q S G I A G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M G S S H H H H H H S SG L V P R G S D ID AT Q L Q G L L N Q E L L T G P . .

690 Calpain-3 CAPNS1 Calpain-2 ALG-2 Grancalcin Sorcin Calpain-13

650

770

EF2

EF3 780

790

800

810

820

AGFHLNNQLYDIITMRYADKH.MNID FDS FICCF VRLEGMFRAFHAFDKDGD GI IKLNVLEW LQL TMYA. AGFHLNQHIYSMIIRRYSDET.GNMD FDN FISCL VRLDAMFRAFRSLDKNGT GQ IQVNIQEW LQL TMYS. AGFKLPCQLHQVIVARFADDE.LIID FDN FVRCL VRLEILFKIFKQLDPENTGT IQLDLISW LSF SVL.. FGYRLSDQFHD ILIRKFDRQGRGQ I AFDDFIQGCIVLQRLTDI FRRYDTDQD GW IQVSYEQY LSM VFSIV MGYRLSPQTLTTIVKRYSKNG..RIF FDD YVACC VKLRALTDFFKKRDHLQQGS ADFIYDDF LQG TMAI. MGFRLSPQAVN SIAKRYSTNG..K IT FDD YIACCVKLRALTDS FRRRDTAQQ GV VNFPYDDF IQC VMSV. RGIFISRELLHLVTLRYSDSV.GRVS FPS LVCFL MRLEAMAKTFRNLSKDGK G. LYLTEMEW MSL VMYN. ¡

¡

EF4

¡

¡

¡

EF5

Fig. 1. Sequence alignment of the PEF domains of calpain-3, calpain small subunit (CAPNS1), calpain-2, ALG-2, grancalcin, sorcin, and calpain-13. The EF-hand motifs are shown as solid bars, and the solid stars represent the loop regions. Blue: EF1. Red: EF2. Green: EF3. Purple: EF4. Cyan: EF5. Absolutely conserved residues are highlighted with a red background, and partially conserved residues are outlined by a blue box. Beneath the EF-hand regions, the residues that bind to Ca2+ via side chain oxygens (solid black boxes) and main chain oxygens (open black boxes) are indicated. Sequence identities between the calpain-3 PEF domain and the other sequences are as follows: rat calpain PEF(S) domain, 57%; rat calpain-2 PEF(L) domain, 48%; ALG-2, 30%; grancalcin, 35%; sorcin, 31%; and human calpain-13 PEF (L) domain, 32%.

anomalous signal for Ca2+ in the native dataset, peaks corresponding to seven La3+ ions were identified in an anomalous difference map calculated by use of the dataset collected on the La3+-soaked crystal and phases from the refined structure. Those La3+ ions were found at the first EF-hand (EF1), EF2 and EF5 for chains A and B, and at EF2 for chain C. For the other EF-hands, identification of bound Ca2+ ions was based on the electron density and the geometry of the oxygen ligands surrounding them. The electron density for chains A and B is clear and continuous, except for the absence of a few residues within the molecule (684–688 in chain A, and 683–688 in chain B), and for some of the N-terminal and C-terminal residues. Their Ca atoms superimpose with
for a root-mean-square deviation (RMSD) of 0.61 A 123 equivalent residues, indicating general similarity in their structures; however, major differences between chain A and chain B are observed in the EF5 region 3140

(Fig. 2B). In comparison, chain C shows more areas of missing density in the N-terminal region (residues 647– 664 and 682–689) and closer to the C-terminal region. Moreover, the RMSD values calculated during alignment of chains A and C, and chains B and C, were
and 1.6 A
for 122 and 131 equivalent Ca atoms, 1.0 A respectively (not shown). Nevertheless, the overall fold of chain C and its mode of dimerization with its symmetry-related molecule are similar to those for the dimer formed between chains A and B. Common to all three chains is the absence of the first 18–20 residues of the N-terminal region, which could be highly flexible. As in other PEF domain proteins, the EF-hands of the calpain-3 PEF domain are paired: EF1 pairs with EF2, and EF3 with EF4, and the lone EF5 of one subunit pairs up with EF5 of the other subunit (Fig. 2A). An overlay of the structures for chains A and B shows the greatest variation in the EF5 region (Fig. 2B). Comparing the structure of the calpain-3 PEF domain FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

A Chain B

Chain A

Chain B

Chain A N

N

C

C 684–688 N

684–688 N

C

B

L5 Fig. 2. Crystal structure of the calpain-3 PEF domain. (A) Stereoview of the homodimer of the calpain-3 PEF domain with chain A (orange), chain B (dark red), and bound Ca2+ (green spheres). Two disordered regions are drawn as dashed lines (labelled 684–688 for chain A). (B) Overlay of chains A (orange) and B (gray) of the calpain-3 PEF domain to show conformational differences at the EF5 region. (C) Overlay of chain A of the calpain-3 PEF domain (orange) with chain A of the rat calpain PEF(S) domain homodimer (light blue). (D) Stereoview of one protomer (chain B) of the calpain-3 PEF domain. The loop regions of the EFhands are colored blue (EF1), red (EF2), green (EF3), purple (EF4) and cyan (EF5) to match Fig. 1. LEF12 and LEF34 are the loops that connect the EF-hand pairs EF1– EF2 and EF3–EF4, respectively. LEF12 is partially disordered, as shown by the dotted line.

E5

F5 E5

C

C

N

D

N

EF1

EF2

α1

α3

EF5

α7

α8

α2 α4 LEF12

LEF34 α5

EF3

with the rat calpain small-subunit homodimer, we find
that the two A chains align with an RMSD of 1.1 A for 114 equivalent Ca atoms (Fig. 2C). Again, these structures tend to differ primarily in the EF5 region. Of the eight helices in each chain, a4 is shared between EF2 and EF3, and a7 is common to EF4 and EF5 (Fig. 2D). Superposition of the rat calpain small subunit, sorcin, grancalcin and ALG-2 on the calpain-3 PEF domain homodimer shows the overall similarity of their folds (Fig. 3). The most different structure is ALG-2, for which the a1–a4 helices have shifted relative to those for the other PEF domain structures. Dimeric structure In the calpain-3 PEF domain homodimer, the two EF5 modules form a four-helix bundle in which the FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

L5

F5

EF4

α6

EF1

EF2

α1

α3

EF5

α7

α8

α2

LEF34

α4 LEF12

α5 EF3

α6

EF4

two EF5 loops (L5s) align as a short antiparallel b-sheet, creating a structure that is common to other PEF domain family proteins (Fig. 4A) [23]. This inter
2 of the molecular surface area, face buries ~ 1750 A and involves the formation of numerous intimate hydrophobic contacts, many of which include residues that are highly conserved [26,27]. This association can be categorized as strong on the basis of thermal denaturation analysis, which showed a Tagg value of 75 °C in the presence of 2 mM Ca2+. Further examination of the dimer structure formed by the chain A and B protomers reveals a compact side and an expanded side (Fig. 4B). This asymmetry appears to arise from differences in the arrangement of the four-helix bundle at the dimer interface. Specifically, there is a shift of helices a3 (of chains A and B) and a7 (of chain B) that allows a few of the histidines 3141

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

α6

A

α6

α5

α5 α7

α7

α8

α6

B

α6

α5

α5 α7

α7

α8

α6

C

α8

α6

α5

α5 α7

α7

α8

α6

D

α8

α8

α6

α5

α5 α7

α8

α7

α8

of the Ni2+affinity tag to dock into the four-helix bundle region rather than extending into the bulk solvent (shown in yellow in Fig. 4D). These conformational differences between the monomers were not observed, for example, in the structure of the PEF(S) domain homodimer (Fig. 4C), and may account for our inability to solve the structure by MR. In contrast, the chain C protomer of the calpain-3 PEF domain crystal structure forms a symmetric dimer with a symmetryrelated chain C protomer through a crystallographic two-fold rotation axis. This protomer shows a less significant conformational difference than the PEF(S) domain homodimer – there is a shift of EF5 and 3142

Fig. 3. Structural similarity of PEF domain proteins. The calpain-3 PEF domain homodimer (chain A, orange; chain B, dark red) is superimposed on the homodimers of (A) the calpain PEF(S) domain (light blue, PDB code 1DVI [24]), (B) sorcin (yellow, PDB code 2GJY [50]), (C) grancalcin (green, PDB code 1K94 [34]), and (D) ALG-2 (pink, PDB code 2ZN9 [35]). The superposition was performed by aligning helices a5–a8 of chain A of the calpain-3 PEF domain with the corresponding helices of the A chains of the other proteins.

helix a8. None of the histidines of the Ni2+-affinity tag are visible for chain C. Ca2+ binding and conformational changes The calpain-3 PEF domain binds four Ca2+ ions per protomer through EF1, EF2, the third EF-hand (EF3), and EF5 (Fig. 2D). The Ca2+-binding residues are indicated in Fig. 1. The EF1 loop of the calpain-3 PEF domain has one residue fewer than the canonical EF-hand, but is similar in length to the EF1 loop of the calpain PEF(S) domain, sorcin, grancalcin, ALG-2 and the calpain-13 PEF(L) domain. The Ca2+ FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

A

Fig. 4. The dimer interface and subunit arrangement of the calpain-3 PEF domain. (A) Top view of the four-helix bundle that the two EF5 modules form at the dimer interface. Helices a7 and a8 and the L5s are highlighted (chain A, orange; chain B, dark red), and the rest of the structure is shown in light and dark gray. (B) The calpain-3 PEF domain homodimer is shown in cylinder representation to reveal the compact and expanded sides of the molecule. (C) The PEF(S) domain homodimer structure with its center of mass indicated by a gray sphere. The dashed lines in (B) and (C) connect the center of mass of the PEF(S) domain homodimer to equivalent points on the a4 helices of each protomer. (D) Two stereoviews of a ribbon representation of the calpain-3 PEF domain homodimer (chain A, light gray; chain B, dark gray, residues of the C-terminus of chain A represented as sticks) to highlight the Cterminal protrusion of chain A, which results in the splayed conformation of the dimer. The lower view is rotated by 180° as compared with the upper view.

N

N

N

D

α7

α7 α7

α7

N

N

N

N

α3

α3

C

C

α7

N

α7

1800

α7

N

α3

coordination sphere at EF1 involves the backbone carbonyl oxygen atoms of Ala662 and Glu667, the carboxylates of Asp665 and Glu672, and a water molecule (Fig. 5A). This pattern of coordination is similar to that observed in EF1 of the calpain PEF(S) domain, and suggests that this could be one of the high-affinity Ca2+-binding sites in the calpain-3 PEF domain. At EF2, Ca2+ is coordinated to the carboxylates of Asp705, Asp707, and Glu716, the side chain of Ser709, the backbone carbonyl oxygen of Lys711, and a water molecule (Fig. 5B). Glu667, whose backbone carbonyl is involved in coordinating the Ca2+ at EF1, forms hydrogen bonds via its side chain to the OH of Ser709 and to the water that is coordinated to the FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

N

C

B

α3

C

α7

α3

N

α3

N

C

Ca2+ at EF2. The electron density for Ca2+ binding at EF3 (Fig. 5C) was weaker, as shown by the elevated B-factors for the Ca2+ at this site for both chains A and B. Only five of the usual seven oxygen ligands are found to interact with the Ca2+ at EF3, and no La3+ ions were found to bind at EF3 in the LaCl3-soaked crystals. Despite the presence of conserved aspartates in the EF4 region, EF4 of the calpain-3 PEF domain does not show evidence of Ca2+ binding in any of the three protomers. Using EF4 of the calpain PEF(S) domain for comparison reveals that one of the ligands for Ca2+ coordination is His690 in the calpain-3 PEF domain rather than the canonical aspartate (Fig. 1).

3143

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

A

B K711

E667

E667

S709

D665

E672 A662

C

D707

E716

D705

D I806 D804

T741 S739

D800

D735

D737

D802

Fig. 5. Coordination of Ca2+ by EF-hands of the calpain-3 PEF domain. (A) Ca2+ coordination sphere at EF1 (Ca2+, green spheres; water, red spheres; side chain and main chain ligands rendered as sticks – carbon, yellow; oxygen, red; nitrogen, blue). Six ligands contribute to Ca2+ binding at this site, including the bidentate ligand E672. (B) At EF2, six amino acid ligands, including the bidentate E716 and one water molecule, coordinate the Ca2+. (C) At EF3, four protein ligands and one water molecule coordinate the Ca2+. (D) At EF5, four amino acid ligands, including the bidentate ligand D802 and two water ligands, contribute to Ca2+ coordination. For all panels, the electron density is from a 2mFo  DFc map contoured at 1.5r (0.31 e/
A3).

This histidine is in the region between a3 and a4 of the structure, for which there was no visible density in any of the chains. Moreover, the C-terminal end of protomer A protrudes towards EF4, and could thus potentially interfere with Ca2+ binding at this site. This suggests that EF4 of the calpain-3 PEF domain could be a lower-affinity Ca2+-binding site than in the PEF(S) domain EF4 site, or that Ca2+ binding might not occur at EF4. As compared with the canonical EF-hand structure, L5 in all PEF domains has a two-residue insertion. As a result, there is no conserved glutamate available to coordinate Ca2+ at the Z position, and L5 in general does not bind Ca2+. Indeed, the crystal structure of the Ca2+-bound calpain PEF(S) domain lacked Ca2+ at EF5 even at CaCl2 concentrations as high as 200 mM. It is generally thought that the role of EF5 in PEF domain proteins is not to bind Ca2+, but simply to stabilize the dimer through the residues at the interface. Surprisingly, the crystal structure of the calpain-3 PEF domain shows clear electron density in the L5 region that enabled us to model Ca2+ at this site. The 3144

anomalous difference map calculated from the LaCl3 dataset clearly indicates the presence of an La3+ ion at this location. The ligands involved in Ca2+ coordination are the side chain carboxylates of Asp800, Asp802, and Asp804, the backbone carbonyl oxygen of Ile806, and two water molecules (Fig. 5D). Despite the shift in register of glutamate at the Z position, the presence of one conserved and two nonconserved aspartates residues in L5 contributes to the Ca2+ binding at EF5 in the calpain-3 PEF domain. To our knowledge, this is this first observation of Ca2+ binding at EF5 within the calpain family. After EF1, we suggest that EF2 and EF5 are the high-affinity sites for Ca2+ in the calpain-3 PEF domain. This conclusion is based on our attempts to crystallize the calpain-3 PEF domain in the absence of Ca2+ but without EDTA present in the crystallization buffer. We found strong electron density consistent with the presence of Ca2+ ions at EF2 and EF5, suggesting that these, like EF1, are also high-affinity sites for Ca2+ (data not shown). Also, the Ca2+ bound to EF3 shows a relatively high B-factor (or partial occupancy). The source of Ca2+ could be the trace levels carried through purification steps, although extensive dialysis with storage buffer (10 mM Hepes and 2 mM dithiothreitol) was performed after purification.

Discussion Ca2+-binding EF-hand motifs are found in a large number of protein families, and perform a wide range of biological functions [28]. Typically these helix–loop– helix structures are paired. When the EF-hand-containing domains of calpains were first characterized, it became apparent that they had an odd number of these motifs (five), and that EF5 served as a dimerization motif. Thus, calpains are the founding members of a small family of proteins with PEF domains that include sorcin, ALG-2, peflin, and grancalcin [23]. Some PEF domains can form either homodimers or heterodimers. For example, the C-terminal PEF domain of the large subunit of calpain-1 and calpain-2 is heterodimerized with the small-subunit PEF domain, but, in the absence of the large subunit, the small subunit will form a tight homodimer through its PEF domain. Other calpain PEF domains appear to form exclusively homodimers or heterodimers [13]. Previous studies have suggested that the skeletal muscle tissue-specific calpain-3 exists as a homodimer. Evidence for self-association comes mainly from sizeexclusion analysis of an inactive mutant of calpain-3 (C129S) that elutes at Mr 180 000, which agrees with the calculated Mr of a calpain-3 homodimer [22]. In FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

addition, calpain-3 prepared from skeletal muscle tissue did not show any evidence of the presence of the calpain small subunit, although this muscle tissue does produce small subunit-containing calpains [29]. Furthermore, yeast two-hybrid assays on calpain-3 (active and inactive) did not show any binding of the small subunit to calpain-3 [18,30]. As the PEF domain of calpain-3 has an unpaired EF-hand, it was natural to suspect that this could act as the homodimer interface. Our previous studies on the isolated calpain-3 PEF domain expressed as a recombinant protein showed that it formed a stable homodimer [31] and did not form a heterodimer when coexpressed in E. coli with the calpain small subunit [13]. The present crystal structure confirms the ability of the calpain-3 PEF domain to form a stable homodimer, and provides direct structural support for the homodimerization of the full-length enzyme. To understand the structural basis of the calpain-3 PEF domain’s inability to heterodimerize with the calpain small subunit, shape complementarity (Sc) calculations were performed on these PEF domains [32]. Sc values were computed for a modeled heterodimer (see Experimental procedures) between the calpain-3 PEF and PEF(S) domains. The artificial heterodimer had an Sc of 0.38, whereas the heterodimer formed between the calpain-2 PEF(L) and PEF(S) domains had an Sc of 0.66. Moreover, Sc values for the homodimers of the calpain-3 PEF and PEF(S) domains were 0.73 and 0.75, respectively, indicative of much better shape complementarities. In addition, the modeled heterodimer of the calpain-3 PEF and PEF(S) domains showed several steric clashes (Fig. 6A) and a smaller buried surface area. We suggest that these steric factors, together with diminished shape complementarity, preclude the formation of a heterodimer of the calpain-3 PEF domain with the small subunit of calpain. On the basis of molecular modeling, the association of two full-length protomers could occur through the PEF domains without any steric interference from the rest of the enzyme’s structure, assuming that it folds in a manner similar to calpain-2 [13]. We have previously presented a model of the fulllength calpain-3 dimer [31]. If we superimpose the structure of the calpain-3 PEF domain on that model (Fig. 6B,C) we see that EF5 is very close to the location expected to be occupied by the NS of calpain-3. For calpain-1 and calpain-2, it has been suggested that binding of Ca2+ to EF2 promotes release of the anchor helix, owing to the change in electrostatic environment [7,33]. Similarly for calpain-3, we propose that binding of Ca2+ to both EF2 and EF5 would promote a change in conformation of the NS, and FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

Structure of calpain-3 PEF domain homodimer

A

B

C

NS NS

Fig. 6. Modeling of a full-length calpain-3 domain homodimer from the calpain-3 PEF domain dimer structure. (A) Chain A of the rat calpain PEF(S) domain homodimer is shown as a light blue surface, and is aligned with chain A (orange cartoon) of the calpain-3 homodimer. Chain B of the calpain-3 homodimer (dark red cartoon) clashes with the surface of chain A of the PEF(S) domain. (B) The full calpain dimer structure (light blue and green cartoon) superimposed on the calpain-3 PEF domain structure (orange and dark red cartoon). The Ca2+ ions are represented by spheres. The catalytic triad residues of calpain-2 are drawn as pink spheres to illustrate the location of the active sites for substrate binding relative to the dimer interface. (C) A close-up view of the dimer interface. The green and blue helices (labeled NS) represent the regions that would be occupied by the NS of calpain-3, showing its proximity to the Ca2+ bound at EF5.

would therefore accelerate its cleavage. The only other PEF domain protein that has been found to bind Ca2+ at EF5 is ALG-2, which does not bind Ca2+ at EF2 [34,35]. In summary, we have provided structural evidence for the homodimer organization of calpain-3 by

3145

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

crystallizing its PEF domain. The present structure shows a degree of flexibility in the dimer interface that has not been seen in other PEF domain proteins. In addition, Ca2+ binding at EF5 of the calpain-3 PEF domain is another distinct feature not observed in other calpain isoforms. It remains to be seen whether there is any functionality associated with Ca2+ binding at EF5, and whether this binding can be achieved in other PEF domain proteins by introducing aspartate into locations corresponding to that in EF5 of the calpain-3 PEF domain. Further structural investigations on calpain-3 domains should be attempted. For example, the structure of the protease core, which has one of the insertion sequences (IS1) peculiar to this isoform, will be needed to understand its function. Meanwhile, all structural information will be of significance for understanding the underlying disease caused by mutations in the calpain-3 gene.

well with the wavelength of the Cr X-ray source, we attempted SAD phasing on La3+-soaked crystals. The La3+soaked crystals (0.75 mM for 10–15 min) diffracted to
and appeared to have a strong anomalous signal. 2.9 A, The diffraction data for the LaCl3-soaked crystals were collected on a Rigaku Micromax-007 HF rotating anode X-ray generator mounted with a Cr target and an RAXIS IV++ image-plate detector equipped with a heliumflushed beam path cone to reduce atmospheric scatter of X-rays. Data for native crystals were collected at the 23ID-B beamline of the Advanced Photon Source (Argonne National Laboratory, Argonne, IL, USA) via remote access. The diffraction images were integrated and scaled with XDS [41,42]. Diffraction data for the native crystals (grown in the presence of 2 mM CaCl2) without soaking at higher Ca2+ concentrations were also collected, but did not show any binding of Ca2+ at EF3 (data not shown).

Structure solution and refinement

Experimental procedures Crystallization and data collection The calpain-3 PEF domain was expressed and purified as previously described [13]. The purified protein was concentrated to 15 mgmL1, and extensively dialyzed against 10 mM Hepes (pH 7.5) containing 10 mM dithiothreitol. Prior to crystallization, the protein sample was mixed with 2 mM calcium chloride and 10 mM dithiothreitol. Crystallization trials were performed by the hanging-drop vapor diffusion method with commercial broad screens from Qiagen (Toronto, Ontario, Canada). Promising results were obtained with Hepes (pH 7.5) as the crystallization buffer. After several rounds of optimization, diffraction-quality crystals were grown in 0.1 M Hepes (pH 7.5), 10% poly(ethylene glycol) 8000, and 8–10% ethylene glycol. Prior to flash-freezing, crystals were briefly soaked with cryoprotectant solution that contained mother liquor [0.1 M Hepes, pH 7.5, 10% poly (ethylene glycol) 8000, 8% ethylene glycol] supplemented with 20% glycerol and 200 mM calcium chloride. Attempts to solve the structure by MR with PHASER [36], with each of the structures listed in Fig. 1 as a search model, were not successful. Various data quality assessments showed no signs of twinning or pseudotranslational symmetry, and yet the partial solutions from the MR search had many breaks in the electron density map, and it proved impossible to find positions for chain C that did not clash with the neighboring subunits. Diffraction data were collected on our Cr target-equipped X-ray source to attempt sulfur-SAD phasing [37,38], but the anomalous signal for the seven methionines and four cysteines of the calpain-3 PEF domain protomer was too weak. Given that La3+ is a well-known isostere of Ca2+ [39,40] and the fact that the LIII absorption edge of La3+ matches

3146

The native crystals were isomorphous with the La3+-soaked crystals, so we combined the native and La3+-soak datasets to obtain initial phases with the single isomorphous replacement with anomalous scattering method [43]. This gave similar results to those obtained with the MR solutions, with ambiguous electron density and breaks in the electron density map. Finally, we were able to solve the structure by using the structure factors from the La3+-soaked crystal with the MR-SAD module of PHENIX [25]. The program 2+ AUTOMR was run with a protomer of the rat Ca bound small subunit [Protein Data Bank (PDB) code: 1ALV] as a search model. The resulting model with its phases and the anomalous data from La3+-soaked crystal were input together in the second step to run AUTOSOL. The heavy atom search module (Hyss) within AUTOSOL found seven La3+ sites. The phasing statistics and the density maps after the AUTOSOL run were encouraging, and suggestive of a true solution. Subsequent to this, model building was performed with the AUTOBUILD module of PHENIX, and this was followed by several rounds of refinement and manual model building until Rwork/Rfree converged to acceptable values (Table 1). The model obtained from MR-SAD was used as input for refinement against the native dataset. An initial rigid body refinement followed by several rounds of model building and refinement were performed with COOT 0.7 [44], PHENIX, and REFMAC5 [45]. PYMOL (PyMOL Molecular Graphics System, Version 1.5.0.4, Schr€ odinger, LLC) was used to prepare all molecular graphics images and to perform all structural alignment calculations (with default settings).

Thermal denaturation assay Thermal denaturation of the calpain-3 PEF domain was measured by differential static light scattering of the protein aggregates formed upon thermal denaturation. For

FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

Table 1. Diffraction data collection and refinement statistics. 3+

Data Space group Wavelength (
A) Cell dimensions a, b, c (
A) Resolution (
A) (outer shell limits in parentheses) No. of molecules/ asymmetric unit I/rI Rmeas CC1/2 [49] Completeness Redundancy Reflections, measured/unique Figure of merit Refinement Resolution (
A) (outer shell) Rwork/Rfree (%) No. of atoms protein/ligand/water B-factors (
A2), protein/ligand/water RMSD value Bond lengths (
A) Bond angles (°)

La -soaked derivative

Native

P41212 2.2909

P41212 0.9

106.70, 106.70, 100.99 19.6–2.9 (2.98–2.90)

107, 107, 96.7

3

3

24.3 (2.6) 0.093 (1.55) 1.00 (0.89) 93.7 (94.4) 15.1 (15.3) 350723/23197

18.0 (2.2) 0.142 (1.73) 1.00 (0.69) 100 (100) 14.3 (14.8) 305850/21268

47.85–2.45 (2.51–2.45)

0.3 –

–

47.85–2.45 (2.51–2.45) 0.198/0.269 (0.256/0.338) 4005/11/46

–

60.1/66.0/55

– –

0.012 1.52

–

Tagg measurements, the protein was used at 0.2 mgmL1 (final concentration) in 10 mM Hepes (pH 7.5), 150 mM NaCl, 2 mM calcium chloride, and 2 mM dithiothreitol. The solution was mixed thoroughly and transferred to a 96-well plate. The protein solution was covered with 50 lL of mineral oil, and the plate was then spun at 150 g for 1 min and placed in the heating chamber of the StarGazer [46–48]. The initial temperature of the chamber was set to 25 °C, and was then raised to 85 °C in increments of 0.5 °C min1. The program BIOFIT was used to extract the intensity of scattered light, which was plotted against the temperature. The midpoint of the curve obtained by nonlinear regression analysis of intensity versus temperature represents the Tagg of the protein, and the measurements were performed in triplicate.

Shape complementarity calculations Shape complementarity calculations were performed with the program SC [32]. The heterodimer between the calpain-3 PEF domain and the calpain-2 small subunit was prepared

FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

by aligning chain A of the small subunit with chain A of the calpain-3 PEF domain dimer, and then removing chain A of the calpain-3 PEF domain dimer. This alignment results in a heterodimer model composed of chain A (calpain-2 small subunit) and chain B of the calpain-3 PEF domain dimer. A second heterodimer was prepared by alignment of chain A of the calpain-2 small subunit and chain B of the calpain-3 PEF domain dimer. The calpain-2 PEF(S)–PEF(L) heterodimer was prepared from 3BOW [11] by removing other domains of the protein.

Acknowledgements We thank G. A. Senisterra from the Structural Genomics Consortium (Toronto) for his help with acquiring and interpreting thermal denaturation data with StarGazer, Zongchao Jia for sharing remote access to the synchrotron facilities at the Advanced Photon Source (Argonne National Laboratory), and S. Gauthier for technical assistance. This work was funded by a grant from the CIHR to P. L. Davies. P. L. Davies holds the Canada Research Chair in Protein Engineering.

Author contributions SKP and RR planned and performed experiments, analyzed data and wrote the paper. JSA, RLC and PLD planned experiments, analyzed data and wrote the paper.

References 1 Hata S, Sorimachi H & Suzuki K (2001) Structure and function of calpain superfamily. Seikagaku 73, 1129–1140. 2 Sorimachi H, Imajoh-Ohmi S, Emori Y, Kawasaki H, Ohno S, Minami Y & 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. 3 Goll DE, Thompson VF, Li H, Wei W & Cong J (2003) The calpain system. Physiol Rev 83, 731–801. 4 Franco SJ & Huttenlocher A (2005) Regulating cell migration: calpains make the cut. J Cell Sci 118, 3829–3838. 5 Huang Y & Wang KK (2001) The calpain family and human disease. Trends Mol Med 7, 355–362. 6 Hosfield CM, Elce JS, Davies PL & Jia Z (1999) Crystal structure of calpain reveals the structural basis for Ca(2+)-dependent protease activity and a novel mode of enzyme activation. EMBO J 18, 6880–6889. 7 Strobl S, Fernandez-Catalan C, Braun M, Huber R, Masumoto H, Nakagawa K, Irie A, Sorimachi H,

3147

S. K. Partha et al.

Structure of calpain-3 PEF domain homodimer

8 9 10

11

12

13

14

15

16

17

18

19

20

Bourenkow G, Bartunik H, et al. (2000) The crystal structure of calcium-free human m-calpain suggests an electrostatic switch mechanism for activation by calcium. Proc Natl Acad Sci USA 97, 588–592. Campbell RL & Davies PL (2012) Structure–function relationships in calpains. Biochem J 447, 335–351. Sorimachi H, Hata S & Ono Y (2011) Impact of genetic insights into calpain biology. J Biochem 150, 23–37. Moldoveanu T, Hosfield CM, Lim D, Elce JS, Jia Z & Davies PL (2002) A Ca(2+) switch aligns the active site of calpain. Cell 108, 649–660. Hanna RA, Campbell RL & Davies PL (2008) Calcium-bound structure of calpain and its mechanism of inhibition by calpastatin. Nature 456, 409–412. Moldoveanu T, Gehring K & Green DR (2008) Concerted multi-pronged attack by calpastatin to occlude the catalytic cleft of heterodimeric calpains. Nature 456, 404–408. Ravulapalli R, Campbell RL, Gauthier SY, DhePaganon S & Davies PL (2009) Distinguishing between calpain heterodimerization and homodimerization. FEBS J 276, 973–982. Hata S, Abe M, Suzuki H, Kitamura F, ToyamaSorimachi N, Abe K, Sakimura K & Sorimachi H (2010) Calpain 8/nCL-2 and calpain 9/nCL-4 constitute an active protease complex, G-calpain, involved in gastric mucosal defense. PLoS Genet 6, e1001040. Kramerova I, Kudryashova E, Tidball JG & Spencer MJ (2004) Null mutation of calpain 3 (p94) in mice causes abnormal sarcomere formation in vivo and in vitro. Hum Mol Genet 13, 1373–1388. Richard I, Broux O, Allamand V, Fougerousse F, Chiannilkulchai N, Bourg N, Brenguier L, Devaud C, Pasturaud P, Roudaut C, et al. (1995) Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 81, 27–40. Diaz BG, Moldoveanu T, Kuiper MJ, Campbell RL & Davies PL (2004) Insertion sequence 1 of musclespecific calpain, p94, acts as an internal propeptide. J Biol Chem 279, 27656–27666. Sorimachi H, Kinbara K, Kimura S, Takahashi M, Ishiura S, Sasagawa N, Sorimachi N, Shimada H, Tagawa K, Maruyama K, et al. (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. Sorimachi H, Toyama-Sorimachi N, Saido TC, Kawasaki H, Sugita H, Miyasaka M, Arahata K, Ishiura S & 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. Federici C, Eshdat Y, Richard I, Bertin B, Guillaume JL, Hattab M, Beckmann JS, Strosberg AD & Camoin

3148

21

22

23

24

25

26

27

28 29

30

31

32

33

L (1999) Purification and identification of two putative autolytic sites in human calpain 3 (p94) expressed in heterologous systems. Arch Biochem Biophys 363, 237–245. Branca D, Gugliucci A, Bano D, Brini M & Carafoli E (1999) Expression, partial purification and functional properties of the muscle-specific calpain isoform p94. Eur J Biochem 265, 839–846. Kinbara K, Ishiura S, Tomioka S, Sorimachi H, Jeong SY, Amano S, Kawasaki H, Kolmerer B, Kimura S, Labeit S, et al. (1998) Purification of native p94, a muscle-specific calpain, and characterization of its autolysis. Biochem J 335 (Pt 3), 589–596. Maki M, Kitaura Y, Satoh H, Ohkouchi S & Shibata H (2002) Structures, functions and molecular evolution of the penta-EF-hand Ca2+-binding proteins. Biochim Biophys Acta 1600, 51–60. Blanchard H, Grochulski P, Li Y, Arthur JS, Davies PL, Elce JS & 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. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Echols N, Headd JJ, Hung LW, Jain S, Kapral GJ, Grosse Kunstleve RW, et al. (2011) The Phenix software for automated determination of macromolecular structures. Methods 55, 94–106. Krissinel E & Henrick K (2007) Inference of macromolecular assemblies from crystalline state. J Mol Biol 372, 774–797. Blanchard H, Li Y, Cygler M, Kay CM, Simon J, Arthur C, Davies PL & Elce JS (1996) Ca(2+)-binding domain VI of rat calpain is a homodimer in solution: hydrodynamic, crystallization and preliminary X-ray diffraction studies. Protein Sci 5, 535–537. Lewit-Bentley A & Rety S (2000) EF-hand calciumbinding proteins. Curr Opin Struct Biol 10, 637–643. Farkas A, Tompa P & Friedrich P (2003) Revisiting ubiquity and tissue specificity of human calpains. Biol Chem 384, 945–949. Herasse M, Ono Y, Fougerousse F, Kimura E, Stockholm D, Beley C, Montarras D, Pinset C, Sorimachi H, Suzuki K, et al. (1999) Expression and functional characteristics of calpain 3 isoforms generated through tissue-specific transcriptional and posttranscriptional events. Mol Cell Biol 19, 4047–4055. Ravulapalli R, Diaz BG, Campbell RL & Davies PL (2005) Homodimerization of calpain 3 penta-EF-hand domain. Biochem J 388, 585–591. Lawrence MC & Colman PM (1993) Shape complementarity at protein/protein interfaces. J Mol Biol 234, 946–950. Nakagawa K, Masumoto H, Sorimachi H & Suzuki K (2001) Dissociation of m-calpain subunits occurs after autolysis of the N-terminus of the catalytic subunit,

FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

S. K. Partha et al.

34

35

36

37

38

39

40

41

and is not required for activation. J Biochem 130, 605–611. Jia J, Han Q, Borregaard N, Lollike K & Cygler M (2000) Crystal structure of human grancalcin, a member of the penta-EF-hand protein family. J Mol Biol 300, 1271–1281. Suzuki H, Kawasaki M, Inuzuka T, Okumura M, Kakiuchi T, Shibata H, Wakatsuki S & Maki M (2008) Structural basis for Ca2+-dependent formation of ALG-2/Alix peptide complex: Ca2+/EF3-driven arginine switch mechanism. Structure 16, 1562–1573. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC & Read RJ (2007) Phaser crystallographic software. J Appl Crystallogr 40, 658–674. Kitago Y, Watanabe N & Tanaka I (2005) Structure determination of a novel protein by sulfur SAD using chromium radiation in combination with a new crystalmounting method. Acta Crystallogr D Biol Crystallogr 61, 1013–1021. Watanabe N, Kitago Y, Tanaka I, Wang J, Gu Y, Zheng C & Fan H (2005) Comparison of phasing methods for sulfur-SAD using in-house chromium radiation: case studies for standard proteins and a 69 kDa protein. Acta Crystallogr D Biol Crystallogr 61, 1533–1540. Weis WI, Kahn R, Fourme R, Drickamer K & Hendrickson WA (1991) Structure of the calciumdependent lectin domain from a rat mannose-binding protein determined by MAD phasing. Science 254, 1608–1615. Shapiro L, Fannon AM, Kwong PD, Thompson A, Lehmann MS, Grubel G, Legrand JF, Als-Nielsen J, Colman DR & Hendrickson WA (1995) Structural basis of cell–cell adhesion by cadherins. Nature 374, 327–337. Kabsch W (2010). XDS. Acta Crystallogr D Biol Crystallogr 66, 125–132.

FEBS Journal 281 (2014) 3138–3149 ª 2014 FEBS

Structure of calpain-3 PEF domain homodimer

42 Kabsch W (2010). Integration, scaling, space-group assignment and post-refinement. Acta Crystallogr D Biol Crystallogr 66, 133–144. 43 Skubak P, Murshudov G & Pannu NS (2009) A multivariate likelihood SIRAS function for phasing and model refinement. Acta Crystallogr D Biol Crystallogr 65, 1051–1061. 44 Emsley P & Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60, 2126–2132. 45 Murshudov GN, Skubak P, Lebedev AA, Pannu NS, Steiner RA, Nicholls RA, Winn MD, Long F & Vagin AA (2011) REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr D67, 355–367. 46 Niesen FH, Berglund H & Vedadi M (2007) The use of differential scanning fluorimetry to detect ligand interactions that promote protein stability. Nat Protoc 2, 2212–2221. 47 Senisterra GA, Soo Hong B, Park HW & Vedadi M (2008) Application of high-throughput isothermal denaturation to assess protein stability and screen for ligands. J Biomol Screen 13, 337–342. 48 Vedadi M, Niesen FH, Allali-Hassani A, Fedorov OY, Finerty PJ Jr, Wasney GA, Yeung R, Arrowsmith C, Ball LJ, Berglund H, et al. (2006) Chemical screening methods to identify ligands that promote protein stability, protein crystallization, and structure determination. Proc Natl Acad Sci USA 103, 15835–15840. 49 Karplus PM & Diedrichs K (2012) Linking crystallographic model and data quality. Science 336, 1030–1033. 50 Ilari A, Johnson K, Nastopoulos V, Verzili D, Zamparelli C, Colotti G, Tsernoglou D & Chiancone E (2002) The crystal structure of the sorcin calcium binding domain provides a model of Ca2+-dependent processes in the full-length protein. J Mol Biol 317, 447–458.

3149