Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins

Article

Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins Graphical Abstract

Authors Jessica J. Chaston, Callum Smits, David Araga˜o, ..., Ricardo A. Bernal, Daniela Stock, Alastair G. Stewart

Correspondence [email protected] (D.S.), [email protected] (A.G.S.)

In Brief Thermophiles survive under wide ranges of extreme conditions. Sulfolobales have evolved a modular system of chaperonins that rescue different classes of substrate proteins under different conditions. Structural analyses of three different complexes sheds new light on substrate specificity and dynamics as well as on the assembly and evolution of chaperonins.

Highlights d

Identification of three distinct TF55 chaperonin complexes in Sulfolobus solfataricus

d

Structural analyses of these suggests different kinetics and substrate specificities

d

Geometry of complexes sheds new light on chaperonin assembly and evolution

Chaston et al., 2016, Structure 24, 1–11 March 1, 2016 ª2016 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.str.2015.12.016

Accession Numbers 4XCD 4XCG 4XCI

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Structure

Article Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins Jessica J. Chaston,1,2 Callum Smits,1,2 David Araga˜o,3 Andrew S.W. Wong,4,5 Bilal Ahsan,4 Sara Sandin,4,5 Sudheer K. Molugu,6 Sanjay K. Molugu,6 Ricardo A. Bernal,6 Daniela Stock,1,2,* and Alastair G. Stewart1,7,* 1Molecular,

Structural and Computational Biology Division, The Victor Chang Cardiac Research Institute, Darlinghurst, NSW 2010, Australia of Medicine, The University of New South Wales, Sydney, NSW 2052, Australia 3Australian Synchrotron, Clayton, VIC 3168, Australia 4School of Biological Sciences, Nanyang Technological University, Singapore 637551 5NTU Institute of Structural Biology, Nanyang Technological University, Singapore 637551 6Department of Chemistry, University of Texas at El Paso, El Paso, TX 79968, USA 7School of Molecular Bioscience, The University of Sydney, Sydney, NSW 2006, Australia *Correspondence: [email protected] (D.S.), [email protected] (A.G.S.) http://dx.doi.org/10.1016/j.str.2015.12.016 2Faculty

SUMMARY

Chaperonins are essential biological complexes assisting protein folding in all kingdoms of life. Whereas homooligomeric bacterial GroEL binds hydrophobic substrates non-specifically, the heterooligomeric eukaryotic CCT binds specifically to distinct classes of substrates. Sulfolobales, which survive in a wide range of temperatures, have evolved three different chaperonin subunits (a, b, g) that form three distinct complexes tailored for different substrate classes at cold, normal, and elevated temperatures. The larger octadecameric b complexes cater for substrates under heat stress, whereas smaller hexadecameric ab complexes prevail under normal conditions. The cold-shock complex contains all three subunits, consistent with greater substrate specificity. Structural analysis using crystallography and electron microscopy reveals the geometry of these complexes and shows a novel arrangement of the a and b subunits in the hexadecamer enabling incorporation of the g subunit.

INTRODUCTION Chaperonins are ubiquitous protein complexes that assist protein folding and prevent detrimental aggregation of un- or misfolded proteins (Lopez et al., 2015; Saibil, 2013; Skjaerven et al., 2015). They are typically up-regulated under conditions of cellular stress, including heat, and are therefore considered to be heat-shock proteins. Multiple copies of one or more subunits form molecular cages of around 1 MDa that sequester misfolded protein substrates in their inner cavity. Allosteric binding and hydrolysis of ATP is believed to create ratchet-like movements within the cage (Bigotti and Clarke, 2008; Fei et al., 2013; Horovitz and Willison, 2005; Saibil et al., 2013). These movements are thought to trigger changes in the substrate-binding surfaces, thereby helping to overcome local energy minima and encouraging proper folding of the substrate. Type I chaper-

onins are made of 14 copies of a 60-kDa protein arranged in 72point symmetry (D7) (Figure 1A) (Braig et al., 1994; Horwich et al., 1990). These occur in eubacteria (GroEL) and within the mitochondrial matrix (cpn60/HSP60). Type II chaperonins occur in archaea (thermosome, TF55) and in the evolutionary-related eukaryotic cytosol (CCT, TRiC) (Horwich et al., 2007; Lopez et al., 2015; Skjaerven et al., 2015). Some archaeal chaperonins are constructed from a single type of 60-kDa subunit, but the majority consist of two paralogous subunits, a and b (Bigotti and Clarke, 2008; Ditzel et al., 1998). These are arranged alternately in a hexadecameric complex of two eight-membered rings (Figure 1B) and are closely related to the eight paralogous eukaryotic CCT subunits, which are also arranged in pairs of eightmembered rings (Figure 1C) (Cong et al., 2012; Dekker et al., 2011; Kalisman et al., 2013; Munoz et al., 2011). Chaperonins are known to function in a highly cooperative manner mediated by the subunit geometry within the complexes and this differs in type I and type II chaperonins (Bigotti and Clarke, 2008; Lopez et al., 2015; Reissmann et al., 2012; Skjaerven et al., 2015; Yebenes et al., 2011). Thermophilic Factor 55 (TF55, rosettasome) is a type II chaperonin occurring in Sulfolobus (Trent et al., 1991). Unlike most other archaea, which encode two related type II chaperonin subunits arranged in complexes of 42-point symmetry (D4), Sulfolobus encodes three paralogous TF55 subunits, a, b, and g that form a series of different complexes (Figures 1D–1F). These complexes have been analyzed and discussed over the past two decades, but their precise structure, stoichiometry, and subunit arrangement remained controversial (Ellis et al., 1998; Kapatai et al., 2006; Marco et al., 1994; Trent et al., 1991). The Sulfolobus b subunit was the first type II chaperonin subunit to be described because of its abundance under heat shock (Trent et al., 1990). It was found that its up-regulation allows Sulfolobus to survive extreme temperatures, up to 92 C. Initially, 8and 9-fold symmetrical particles were discussed (Trent et al., 1991). Subsequent electron microscopy (EM) studies using material derived from heat-shocked cells confirmed 9-fold symmetry (Knapp et al., 1994; Marco et al., 1994). However, when the a subunit was identified it was concluded that Sufolobus TF55 is most likely constructed from a and b subunits in the same manner as observed in other archaeal chaperonins showing 8-fold symmetry (Kagawa et al., 1995; Kapatai et al., 2006).

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 1

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Figure 1. Chaperonin Architecture Across Taxonomic Systems (A–C) Type I chaperonins (GroEL, HSP60, cpn60) consist of 14 copies of the same 60-kDa subunit arranged in a cylindrical particle with two seven-membered rings arranged back to back (A). Type II chaperonins (archaea, thermosomes; eukaryotic cytosol, CCT) generally consist of two eight-membered rings; thermosomes typically have two evolutionary-related (paralogous) subunits that are arranged alternately within one ring (B), eukaryotic CCT/TRiC has eight paralogous subunits per ring (C). (D–F) Sulfolobales have three paralogous subunits that, as we show here, can form three types of complexes: two species of nine-membered rings (either homomeric all-b or heterotrimeric abg) arranged in pairs to form octadecameric complexes (D and F), or hexadecameric ab complexes (E).

Following discovery of the g subunit in the Sulfolobus genome (Archibald et al., 1999), analysis of Sulfolobus shibatae cultures revealed that the expression levels and relative amounts of subunits vary with temperature (Kagawa et al., 2003). Thus, the third subunit, g, is only present under cold-shock conditions (60 C– 75 C), while a and b are present at normal growth temperatures (75 C–79 C). Under heat-shock conditions (>86 C) the b subunit predominates. Here we show that, like the S. shibatae TF55 subunits, the three corresponding Sulfolobus solfataricus subunits have remarkably different properties in terms of thermal stability and their propensity to form complexes. We show that an all-b complex is formed under heat shock and prevails at extreme temperatures, whereas a mixed ab complex predominates at normal growth temperatures, and a mixed abg complex only forms at low temperatures. Determination of the structures of the three complexes using a combination of X-ray crystallography and electron cryomicroscopy (cryo-EM) defines the subunit arrangements within each complex and suggests that the different complexes have different substrate specificities and folding chamber sizes. Different inter-ring interactions point to different cooperativity networks within these complexes compared with GroEL and thermosomes, and provide new insight into the evolution of chaperonins. Although little is known about in vivo substrates

of archaeal chaperonins, our X-ray data show a potential substrate interaction between domains of adjacent subunits. RESULTS Characterization of TF55 Complexes and Subunits Using ion-exchange and size-exclusion chromatography, we purified two different TF55 complexes from S. solfataricus cells grown at 80 C. One complex contains only b subunits (termed prepA), whereas the other contains both a and b subunits (termed prepB). The g subunit was undetectable in either preparation via MALDI-TOF analysis of in-gel tryptic digests. To establish the role of the g subunit, we cloned, expressed, and purified all three subunits from Escherichia coli and analyzed their thermal stability and propensity to form complexes (Figure S2). Recombinant expression levels of the three subunits in E. coli vary significantly, preventing co-expression of subunits. Expression levels for the a subunit are low, whereas the b and g subunits express well. The b subunit forms inclusion bodies that fold upon heat treatment of the crude lysate and subsequently remain folded and in solution (Figure S1). The thermostability of individual subunits was assessed by thermal melt analysis (Figure S2A). These confirm that the g subunit is heat labile and unfolds above 75 C, just below the optimal

2 Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Figure 2. S. solfataricus TF55 All-b Chaperonin Structure (A) Cartoon representation of the biological unit. b subunits are shown in blue and nucleotides in yellow. (B) Cartoon representation of an individual b subunit with domains labeled: SL, sensor loop; Ct, C terminus. See also Figures S1, S2, S3, S7, and Table S1.

growth temperature range of S. solfataricus. The a and b subunits are substantially more heat resistant, both starting to unfold at around 94 C. This is consistent with observations made for S. shibatae subunits and, together with varying transcription levels determined by Northern analysis of S. shibatae cells grown at different temperatures (Kagawa et al., 2003) (Figure S2B), explains the varying in vivo subunit ratios observed in Sulfolobus at different temperatures (Figure S2C). Native gel electrophoresis combined with western blot analysis shows that complex formation is triggered by addition of ATP. However, the propensity of subunits to form complexes differs significantly as follows. The b subunit forms complexes by itself, whereas the a and g subunits do not (Figure S2D). The a and b subunits together form complexes, as do all three subunits together. The g subunit requires the presence of both the a and b subunits to form a complex. To exclude the possibility that all high-molecular weight bands on the native gel correspond to all-b complexes, we combined untagged wild-type b subunit and ab complexes with tagged recombinant subunits. A mixed ab complex is formed by adding recombinant a subunits to wild-type b subunits (Figure S2D), but no mixed ag or bg complexes are formed. Interestingly, adding recombinant g subunit to wild-type ab complex results in an abg heterotrimeric complex. This suggests that the g subunit is either able to insert itself into the complex, or that a dynamic equilibrium of complex formation and disassembly exists as previously suggested (Quaite-Randall et al., 1995). Furthermore, when incorporated into the complex, the g subunit still degrades above 75 C (Figure S2E). X-Ray Structure Determination of Complexes The All-b TF55 Chaperonin Structure The crystal structure of the recombinant all-b complex was determined to 3.8-A˚ resolution by molecular replacement using the closely related Acidianus tengchongensis type II chaperonin structure, 84% sequence identity (Huo et al., 2010), as a search model. The crystallographic symmetry (P63 with six monomers per asymmetric unit, Figure S3A) builds up an octadecameric chaperonin complex of 92-point symmetry (D9) (Figure 2A),

consistent with that seen in the A. tengchongensis chaperonin structure that crystallized in a different space group, C2, with nine subunits (one ring) per asymmetric unit. Despite the different crystal symmetries, the two structures are remarkably similar (root mean square deviation [rmsd] 0.92 A˚) and show an open conformation with subunits in register across the equator. The complex has a maximum internal diameter of 78 A˚ (Figure 2A), suggesting that it is limited to binding proteins with a volume of around 250 nm3, corresponding to a molecular weight of 200 kDa. The ab TF55 Chaperonin Structures Mixed ab subunit containing PrepB provided crystals of space group I422, with one heterodimer per asymmetric unit consistent with a Thermoplasma acidophilum ‘‘thermosome-like’’ structure (Ditzel et al., 1998) of 42-point symmetry (Figures 3A and 3B). Two separate crystal forms were obtained from the same preparation using the same crystallization conditions. One has a long c axis (304.7 A˚), whereas the other has a shorter c axis (248.1 A˚), and we refer to them as ‘‘ablongC’’ and ‘‘abshortC,’’ respectively. The structure of each crystal form was solved independently by molecular replacement using the coordinates of the S. solfataricus b subunit described above (Figure 2B) and the A. tengchongensis a subunit, 83% sequence identity (Zhang et al., 2013), as individual search models. Although the resolution of the X-ray data for the ablongC crystal form extends to 3.0 A˚ in the best direction, the data are highly anisotropic (Table 1). This is reflected in the electron density for the apical domains being so poor that unequivocal assignment of the residues in this region was not possible. Although the main chain of the b subunit could be traced for all three domains, the electron density for the a subunit’s apical domain (Figure 2B) disappears into noise and was therefore omitted from the final model (see Table 1 for missing residues). However, the electron density for the equatorial and intermediate domains is remarkably clear and the a and b subunits could be identified using small differences between the sequences of the subunits (Figures S6A and S6B). In addition, a platinum derivative confirmed that we assigned the subunits correctly (Figure S6C). The data for the abshortC crystal form extend to only 3.75-A˚ resolution, but is less anisotropic and the

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 3

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Figure 3. S. solfataricus TF55 ab Chaperonin Crystal Structures (A and B) Cartoon representation of the biological unit of the ablongC (A) and abshortC (B) hexadecameric structures. b subunits are shown in blue, a subunits in green and nucleotide in yellow. (C and D) Silhouette of individual subunits with respect to the closed hexadecameric structure (PDB: 1A6D, darker gray) and open hexadecameric structure (PDB: 3IZH, lighter gray). The subunits were superposed to the biological complex (Figure S4B), rather than individual subunits (Figure S4A), so that the relative position within the complex as well as conformational changes can be assessed. See also Figures S3, S6, S7, and Table S1.

electron density allowed unequivocal tracing of all three domains of either subunit with the omission of only few residues (Table 1). Density corresponding to ADP is clear in the b subunit of the abshortC crystal form, whereas no density for nucleotide is detectable in the a subunit, nor in any subunit of the ablongC crystal form. In both crystal forms the crystal symmetry builds up a hexadecameric chaperonin complex, which in each case is constructed from two rings containing four a subunits and four b subunits. The subunits alternate within a single ring as seen in the T. acidophilum ab thermosome structure (Ditzel et al., 1998). However, most interestingly, the inter-ring arrangements of the a and b subunits are different in between the two species; the a and b subunits in the T. acidophilum structure form homodimers across the equator, whereas in the Sulfolobus TF55 structure the rings are rotated by 45 , thus leading to ab heterodimers across the equator in a staggered arrangement (Figures 1B, 1E, 3A, and 3B). The ablongC crystal structure most likely adopts a near closed conformation when compared with other chaperonin structures (Figure 3C) (Ditzel et al., 1998; Douglas et al., 2011). However, in the abshortC structure, which clearly shows the apical domains, the b subunit is in a closed conformation, whereas the a subunit

adopts an extremely extended open conformation that has not been seen in any previous chaperonin structure (Figures 3D and S4). The molecular arrangement within the ablongC and abshortC crystal lattices accounts for the large c axis difference observed between them (Figures 3, S3B, and S3C). In the ablongC crystal, the asymmetric unit is arranged such that one a subunit is in contact with the outside of the other (Figure 4A). Conversely, in the abshortC crystal, the a subunit is highly extended and intercalates with the a subunit of an adjacent complex (Figure 4B). This leads to a shorter distance between asymmetric units (27 A˚) and hence a shorter crystallographic c axis (57 A˚, twice the distance since there are two asymmetric units per unit cell along the c axis). The maximum internal diameter of the ab complexes is smaller than that of the all-b complex (65 A˚). This suggests that the ab complex would be limited to binding proteins with a maximum volume of around 145 nm3 or 120 kDa. Detailed analysis of the crystal contacts between the a subunits in the abshortC crystal form highlights an interesting feature; the N-terminal part of the helical protrusion of the a subunit (Figure 2B) has unwound to create an extended b-sheet with the sensor loop in the equatorial domains of an adjacent complex reminiscent of the structural conversion seen in prion formation (Prusiner, 2012) (Figure 4C).

4 Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Table 1. Data Collection and Refinement Statistics Data Collection Crystal

all-b: 4XCD

ablongC: 4XCI

abshortC: 4XCG

Space group

P63

I422

I422

Cell parameters (a, b, c) Wavelength (A˚) Resolution range (A˚)a

163.4, 163.4, 327.3

148.4, 148.4, 304.7

145.8, 145.8, 248.1

0.95370

0.95370

0.95370

65.38–3.78 (3.91–3.78)a

48.83–3.00 (3.15–3.00)

47.69–3.73 (4.17–3.73)

High resolution cutoff for h, k, l (A˚)b

3.78, 3.78, 5.37

3.00, 3.00, 4.65

3.73

Unique reflectionsa

48,892 (4,186)

34,477 (4,522)

14,246 (3,926)

Multiplicitya

6.4 (6.1)

14.7 (15.0)

19.2 (19.0)

Data completenessa

99.4 (93.6)

100.0 (100.0)

99.6 (98.6)

Mn(I) half-set correlation CC(1/2)a

0.997 (0.378)

0.998 (0.408)

0.984 (0.475)

I/s(I)a

5.2 (0.9)

6.8 (0.6)

6.9 (2.2)

Wilson B value (A˚2)

95

71

92

Solvent content (%)

64

65

55

Refinement Statistics Resolution range (A˚)a

64.93–3.79 (3.89–3.79)

41.91–3.00 (3.12–3.00)

47.60–3.74 (4.03–3.74)

Number of reflections Rwork/Rfree

34,788/3,645

24,929/1,273

14,206/718

Data completeness (%)a

71 (18)

73 (20)

100 (99)

Atoms (protein/ligands)

22,955/162

11,776/0

15,026/27

Rwork (%)a

24.9 (33.5)

22.8 (32.7)

22.7 (25.9)

Rfree (%)a

27.9 (35.8)

26.7 (40.1)

28.5 (32.8)

Rmsd bond length (A˚)

0.006

0.004

0.003

Rmsd bond angle ( ) Mean B value (A˚2)

0.978

0.761

0.642

178

95

121

Ramachandran plot (%) (favored/ additional/disallowed)c

96.6/3.4/0.0

94.9/4.6/0.5

94.7/5.1/0.1

Maximum likelihood coordinate error

0.37

0.39

0.54

Missing residues

T7 tag and linker: 1–29, 255–256, 541–557

a chain: 1–16, 166–174, 203–394, 531–559 b chain: 1–28, 251–282, 296–298, 303–306, 318–319, 339–355, 536–557

a chain: 1–18, 251–267, 532–559 b chain: 1–29, 99–101, 232–235, 254–276, 318–321, 536–557

Twin operator/fraction

K, H, -L/0.499

None

None

a

Data for the outermost shell are given in parentheses. As suggested by Aimless (Evans and Murshudov, 2013). c As defined by MolProbity (Chen et al., 2010). b

Electron Microscopy Complexes were analyzed by negative stain and cryo-EM (Figure S5). As expected, classification of all-b complexes shows a homogeneous distribution of nine-membered rings in top views (Figure S5). Classification of wild-type ab complexes shows predominantly eight-membered rings with a small fraction (5%–6%) of nine-membered rings, presumably consisting of all-b complexes (Figure S5). Addition of recombinant g subunit to this sample converts the eightmembered rings into nine-membered rings (Figure S5). This not only corroborates the observation, made by native gel analysis, that the g subunits can insert themselves into preformed ab complexes, but also confirms that the ninemembered rings observed in images of recombinant abg complexes represent indeed the ternary complex rather than an all-b complex.

A 16-A˚ resolution reconstruction of the unstained ternary complex made from recombinant a, b, and g subunits was obtained via cryo-EM and shows an octadecameric complex consisting of two nine-membered rings of alternating a, b, and g subunits with 32-point symmetry (D3) (Figures 5 and S5). Attempts were made to obtain higher-resolution data using a direct electron detector; however, data converged at 16-A˚ resolution indicating that the sample limited resolution rather than the detector. In order to identify the correct symmetry of the complex, we used the multi-model refinement algorithm (multirefine) of the EMAN software suite (Ludtke, 2010). The dataset was refined using models without symmetry as well as all possible symmetries for a cylindrical octadecamer up to D9. The resulting D3 subset converged into a substantially betterresolved map compared with the others, indicating that the symmetry of the complex is D3 (Figure S5). At the current

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 5

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

the 2-fold axis (Figures 1D, 5B, and 6). The octadecameric all-b crystal structure fits well into the EM density indicating that this map shows the complex in a near open conformation. DISCUSSION

Figure 4. Crystal Contacts between the a Subunits of the abshortC Crystal Structure (A and B) Silhouettes of the a subunits show how the different contacts cause a large change in the c axis dimension between the ablongC (A) and the ablongC (B). (C) Cartoon representation of the major crystal contact of the abshortC structure; the helical protrusion (HP) has been partially unwound into a b-strand to extend the b-sheet formed by the sensor loop (SL) and the C terminus (Ct) of a and b subunits of an adjacent complex. See also Movie S1.

resolution it is not possible to establish the exact sequence of subunits in a ring to show whether a follows b follows g or a follows g follows b. However we suggest that the subunits can be assigned to the EM density according to the following considerations: (1) the hexadecameric TF55 structures have alternating a and b subunits that form heterodimers across the equator. Assuming that as many contacts as possible remain preserved, the only way an octadecameric complex can be assembled with an additional subunit, g, is for g to form a homodimer across the equator. (2) Due to the chiral nature of proteins, the 32-point symmetry of the 18-mer dictates an arrangement of one homodimer sitting on the 2-fold axis that is surrounded by two heterodimers on either side generated by

Chaperonin Evolution: Jack of All Trades versus Specialists Cellular stress impairs protein folding, leading to cytotoxic protein aggregation (Gregersen et al., 2006; Tam et al., 2006, 2009). All kingdoms of life therefore employ chaperonins for protein quality control and rescue, yet with very different degrees of complexity. Bacterial type I chaperonin complexes (GroEL) consist of homooligomeric heptadecamers that bind to virtually all unfolded proteins via unspecific hydrophobic interactions (Saibil, 2013). Since both average protein size and genomes are larger in archaea and eukaryotes, these organisms have evolved larger chaperonins (hexa- or octadecamers) consisting of two to eight different subunits that originated from gene duplication (Figure 1). These paralogous subunits recognize their substrates specifically via a combination of hydrophilic and hydrophobic interactions (Joachimiak et al., 2014). Extremophiles are under particular pressure to protect their protein repertoire under a very wide range of conditions. Whereas the eight paralogous subunits in eukaryotes have likely evolved by duplications that occurred early and only once, the two to five different subunits found in different species of archaea most likely originated from various independent and lineage-specific duplications for adaptation to extreme environments (Archibald et al., 2001; Bigotti and Clarke, 2008). In contrast to eukaryotes, where the eight paralogous subunits constitute one eight-membered ring, the three paralogous TF55 chaperonin subunits found in Sulfolobales exhibit considerable plasticity with regards to the complexes they can form. This is achieved by a combination of different expression levels and thermostability of subunits, which leads to formation of three distinct complexes with different properties suitable to rescue different classes of substrates under different forms of stress. TF55 Complex Formation and Geometry In this study we used S. solfataricus cells grown at 80 C, which is slightly above normal growth temperatures (75 C–79 C) and significantly above the thermostability range for subunit g. We isolated intact TF55 complexes consisting of hexadecameric ab complexes. EM analysis of this preparation showed that 5%–6% of particles consist of 9-fold symmetrical, most likely all-b complexes (Figure S5). This suggests that ab complexes are the most stable and predominate species at normal growth temperatures. A previous study using heat-shocked (85 C) S. shibatae cells and similar purification protocols had isolated intact, exclusively 9-fold symmetrical TF55 complexes (Schoehn et al., 2000). Thus, while the mixed ab complexes found at normal growth temperatures are similar to thermosomes of other archaea, the octadecameric all-b complexes appear to be unique to Sulfolobales and prevail under heat-shock conditions. We confirm that the b subunit forms homooligomers readily when ATP is added (Kagawa et al., 2003), whereas the a subunit forms homooligomers only to a limited extent and the g subunit does not oligomerize on its own into larger complexes. This

6 Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Figure 5. Cryo-EM Reconstruction of the abg Complex (A) 3D reconstruction of the recombinant abg complex to 16-A˚ resolution in side and top views (gray density). One set of three symmetry-unrelated subunits (labeled 1,2,3) is highlighted in a red box. (B) Rigid body docking of the symmetrical all-b 18-mer structure into the EM density. The a and b equivalent subunits are colored light or dark cyan (indicating they cannot be assigned) and the g subunit across the 2-fold symmetry axes is colored in red. See also Figures S5.

varies somewhat to chaperonins of other species. For example, the a subunits from T. acidophilum and A. tengchongensis form stable complexes of either 8- or 9-fold symmetry (Waldmann et al., 1995; Zhang et al., 2013). A common theme for thermostability and the propensity to oligomerize seems to be the extreme N- and C-termini of the subunits (Luo and Robb, 2011; Zhang et al., 2013). While the b subunit has an extended N terminus, the a subunit has an extended C terminus, but the g subunit has neither (Figure S6A), and this might explain its lack of selfoligomerization. The structure of the chaperonin subunit from the psychrophilic Antarctic methanogen Methanococcoides burtonii is the only example of a monomeric chaperonin structure and the construct used for structural analysis lacks its termini (Pilak et al., 2011). Here we have also demonstrated that as soon as a subunits are added to b subunits, mixed ab complexes are formed. This implies that the a subunit must have a greater affinity for the b subunit than the b subunit has for itself. Similarly, the g subunit must have the highest affinity to bind to the a subunit on one side and to the b subunit on the other, since addition of the g subunit to preformed ab complexes results in incorporation of the g subunit. Interestingly the TF55 ab complex geometry differs to that found previously in the ab thermosome from T. acidophilum (Ditzel et al., 1998). In the T. acidophilum thermosome, subunits homodimerize between the rings of the complex (Figures 1B and 6A), whereas in the TF55 complex the subunits heterodimerize between the rings (Figures 1E and 6B). This concept has interesting implications for the abg complex. When two ab rings form a complex, they can be arranged either to form homodimeric interactions (a-a or b-b) between the subunits across the equator (as seen in the T. acidophilum thermosome), or heterodimeric (a-b) interactions (as seen in the TF55 ab complex). This would be of little consequence provided that the number of different subunits was two. However, when moving to a three-subunit system, the chiral nature of the rings dictates that not all subunits can homodimerize. Instead only one subunit can homodimerize and the others must heterodimerize (Figures 1D, 4C, and 6C). Thus, the homodimeric a-a and b-b interactions between subunits in opposite rings in the T. acidophilum thermosome would not easily accommodate

the incorporation of the g subunit, whereas the ab heterodimeric interactions in the TF55 complex does. The same considerations apply to the eukaryotic CCT complex: the chiral nature of the subunits dictates an antiparallel arrangement of subunits between rings with two homodimers (subunits b and z) along the 2-fold symmetry axis, with the other subunits arranged as heterodimers clockwise in one ring and anti-clockwise in the other (Figure 6D). Conformations of Subunits and Complexes Superposition of individual subunits from the PDB on their structurally conserved equatorial domains gives an indication of the conformational space that the subunits can occupy (Figure S4). Conformational changes fall into two classes; first there is the helical protrusion, which is the most flexible part of the structure that moves along a trajectory reminiscent of tentacles fishing for prey. The other movement is around a different axis at the interface of the intermediate and equatorial domains (Booth et al., 2008; Clare et al., 2008; Zhang et al., 2010). Comparison of all archaeal chaperonin subunit structures indicates that the a subunit of the abshortC structure forms the most extended of all conformations seen so far and the b subunit the most closed one (Figure S4). Opening and closing of the substrate chamber also involves relative movement of subunits, and this is best investigated by superimposing subunits in the context of the intact complexes (Figures 3C, and S4). This confirms that the ablongC structure is in an almost closed conformation, whereas the abshortC complex is made of extreme open a subunits and closed b subunits. It is interesting to note that, although the ab complex was purified from source, without the addition of nucleotide, differences in nucleotide binding exist between the two crystal forms. Only the b subunit of the abshortC crystal form contains ADP in the nucleotide-binding site (Figure S7B), whereas all other subunits are nucleotide free. This implies that the two subunits have different nucleotide affinities and provides structural validation of previous biochemical analysis of the T. acidophilum thermosome (Bigotti and Clarke, 2005). Here, eight molecules of ATP were seen to hydrolyze in a rapid first phase followed by a slow phase where a further eight molecules are hydrolyzed.

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 7

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Figure 6. Mercator Projections of Chaperonin Complexes (A–D) Planar projections (in analogy to Cong et al., 2012) highlight the differences in subunit arrangement within chaperonin complexes. (A) Thermosome (a, green and b, blue), (B) ablongC (a, green and b, blue), (C) abg model (a/b, light/dark cyan; g, red), and (D) CCT (a yellow; b, red; g, white; d, orange; 3, green; z, black; h, indigo; and q, purple) (PDB: 4V94). Two-fold symmetry axes are indicated. Note the different locations within the T. acidophilum and S. solfataricus structures, causing the differences in subunit arrangement.

the protrusion were suggested to participate in both substrate recognition and in modulating the conformation of the protrusion for substrate release upon lid closure. Our crystal structure may show this substrate recognition region in complex with a surrogate substrate, that being the hydrophobic sensor loop of the open a subunit. This region in itself has been implicated in substrate recognition (Skjaerven et al., 2015; Yebenes et al., 2011).

Whether the abshortC structure represents a physiologically relevant conformation or not is unclear. However, the conformation of the extended a subunit shows some interesting features that might point to possible interactions with substrates. The helical protrusion is partially unwound forming a b strand that contributes to an extended b sheet formed by the sensor loop of another a subunit and the C terminus of the neighboring b subunit (Figure 4C). These observations confirm that the secondary structure of the helical protrusion is weak and prone to unfolding, but in this case might also mimic a substrate that can be recognized by the sensor loop of another a subunit or vice versa. A similar conformation of the helical protrusion could be seen in an NMR-derived model of the apical domain of subunit CCT3 (Joachimiak et al., 2014). Here the helical protrusion has unwound to form a similar b strand. Residues at the base of

Allosteric Networks and Subunit Cooperativity within Complexes Chaperonins Are Intricate Molecular Machines GroEL has been shown to have negative cooperativity with one chamber preferentially closed by GroES and the other open, giving rise to ‘‘bullet’’-shaped structures (Xu et al., 1997). Subunit interactions in GroEL across rings are out of register (i.e. one subunit in one ring contacts the gap in between two subunits in the other) and these interactions are likely to confer the negative cooperativity between rings (Lopez et al., 2015; Skjaerven et al., 2015; Yebenes et al., 2011). The reaction cycle and cooperativity between subunits and rings in type II chaperonins are more complex and less clear. In contrast to type I chaperonins, subunits across the equator are in register suggesting different modes of crosstalk. However, in type II chaperonins with more than one subunit there are further differences: with two different subunits the subunits can either homodimerize across the equator (as in the T. acidophilum thermosome) or heterodimerize (as in the Sulfolobus TF55 ab complex). As mentioned earlier, different subunits have been shown to have different nucleotide affinities and show different kinetics in ATP turnover (Bigotti and Clarke, 2008; Cong et al., 2012; Reissmann et al., 2012). The different geometries of subunits will therefore likely influence the kinetics of ATP hydrolysis and substrate refolding. This is enhanced in complexes with more than two subunits

8 Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

and the antiparallel arrangement of subunits in the TF55 abg complex, and even more so in CCT will lead to very complex kinetics that is likely to be important for substrate-specific refolding (Cong et al., 2012; Lopez et al., 2015). EXPERIMENTAL PROCEDURES Purification of Wild-Type TF55 Complexes from S. Solfataricus Cells Sulfolobus solfataricus (DSM1617) cell paste was purchased from the University of Georgia fermentation facility. Cells were grown at 80 C (pH 3.7) and harvested at an OD600 of 1.2. The cell paste was kept at 80 C until use. TF55 complexes were prepared using ion-exchange and size-exclusion chromatography as described in Supplemental Information. Recombinant Cloning and Expression of TF55 Subunits in E. coli The genes of the a, b, and g subunits were amplified from S. solfataricus genomic DNA and cloned into E. coli expression vectors as described in Supplemental Information. Initial attempts to co-express the subunits (with the a subunit in MCS1 and the b subunit in MCS2 of a pETDUET expression vector) failed because of different expression levels and properties of the subunits. Expression of the a subunit is low (1.5 mg/l culture), whereas the b and g subunits express at much higher levels (14 and 12 mg/l culture, respectively). The b subunit forms insoluble inclusion bodies, whereas the g subunit is soluble. Heat-shock treatment of E. coli crude lysate containing insoluble b subunit denatures most E. coli proteins, whereas the b subunit becomes soluble (Figure S1). Once refolded by heat shock, the b subunit remains stable and in solution. All three subunits were individually purified to homogeneity as monomers (Figure S1) and mixed in equimolar amounts for complex formation (Figure S1) as described in Supplemental Information. Crystallization of ab and All-b Complexes and Structure Determinations All-b Complex Both recombinant and native all-b complexes (PrepA) crystallized. Crystals of recombinant protein were easier to reproduce and therefore used for structure determination. Fractions from the Superose 6 column were concentrated to 15 mg/ml and subjected to crystallization trials. Large single crystals (0.15– 0.25 mm) grew in Linbro plates with a 2-ml protein solution and a 2-ml reservoir mixed in hanging drops over a 500-ml reservoir containing 2 M sodium formate. These crystals were cryo-protected with 4 M trimethylamine N-oxide (TMAO) (pH 8.0) before being plunge frozen in liquid N2. A native dataset was collected at beamline MX2 at the Australian synchrotron. Data were processed using XDS (Kabsch, 2010) and scaled using Aimless (Evans and Murshudov, 2013). Crystals appear to belong to space group P6322. However, after careful analysis using Phenix xtriage (Adams et al., 2010) and reprocessing of the data (Collaborative Computational Project Number 4, 1994), it became apparent that the crystals are twinned and instead belong to spacegroup P63, with a twinning fraction close to 50% and a twinning operator of K, H, -L. The structure was subsequently solved by molecular replacement with Phaser (McCoy et al., 2007) using the A. tengchongensis chaperonin coordinates (PDB: 3KO1, chain A, 84% sequence identity) as the search model, finding six monomers in the asymmetric unit. This model was iteratively refined using Phenix/Rosetta (DiMaio et al., 2013) and refmac5 (Vagin et al., 2004), with manual model building using coot (Emsley and Cowtan, 2004) and final R factors of Rwork=24.37% and Rfree=27.93% (Table 1). The nucleotide bound to the A. tengchongensis structure was initially omitted from the model, but strong difference density (peaks >5s for all six sites) appeared in the known nucleotide-binding site during refinement. This difference density was interpreted to be ADP (as in the A. tengchongensis structure) and refined well (Table 1). ab Complex Wild-type ab complex (PrepB) crystallized more readily than recombinant ab complex and was therefore used for structure determination. Fractions from the Superose 6 column were concentrated and subjected to crystallization trials. Large single crystals were produced using hanging drop Linbro plates with 1 ml protein solution at 3.3 mg/ml and 1 ml mother liquor consisting of 0.1 M Tris (pH 7.9), 6.25% PEG 2000, 10% 2-propanol and 40 mM SrCl2. Crystals were cryo-protected by gradual addition of 4 M TMAO (pH 8.0) to

the drop over 20 min, before being plunge frozen in liquid N2. Datasets were collected at beamline MX2 at the Australian synchrotron. Data were processed to 3.0 and 3.75 A˚ (ablongC and abshortC, respectively) using XDS and scaled using Aimless (merging two datasets taken from the same crystal at different positions in the case of the abshortC crystal) in space group I422. The data from the ablongC crystal form is severely anisotropic, and an anisotropy-corrected dataset from the Diffraction Anisotropy Server (Strong et al., 2006) was used truncating the c axis to 5.37 A˚. The structure of each crystal form was solved independently by molecular replacement with Phaser using the coordinates of the S. solfataricus b subunit described above and the A. tengchongensis a subunit (PDB: 3J1B, chain A, sequence identity 83%) as individual search models, finding a single monomer of each in the asymmetric unit. These models were iteratively refined using Phenix/Rosetta and refmac5, with manual model building using coot and omit maps with final Rwork and Rfree of 22.83% and 26.66% for ablongC and 24.74% and 29.20 for abshortC (Table 1). In both crystal forms the equatorial domain is very well ordered, and side chains are easily visible for the majority of residues (Figures S3D and S6B). However, the density corresponding to the apical domain is much weaker and the model presented may contain side chain errors in this region. Moreover, in the ablongC crystal form the apical domain of the a subunit was not visible even after the final stages of refinement and model building. The ablongC model was therefore truncated as shown in Table 1. Cryo-EM Sample Preparation and Imaging (abg Complex) Samples for cryo-EM were applied onto grids using a 3-ml droplet of the abg complex (at 2 mg/ml). The drop was deposited onto a (400-mesh) copper grid with holey carbon film (Quantifoil R2/2) that was glow discharged for 30 s. Excess protein solution was blotted away using Whatman #1 filter paper and then rapidly plunged into liquid ethane to flash freeze it. Single-particle EM data were collected on a JEOL 3200FS transmission electron microscope (JEOL) equipped with a US4000 CCD detector (Gatan). The images were recorded at 300 keV at liquid nitrogen temperature with a nominal magnification of 360,000 (calibrated magnification, 369,000) corresponding to a pixel size of 2.174 A˚. Particle Selection, Two-Dimensional Class Averaging, and Refinement Particle selection was performed using the e2boxer.py component of the EMAN2 software package (Ludtke, 2010). From an initial 32,240 particles in the dataset, 16,534 particles were used in the final reconstruction. The defocus for each of the CCD frames was calculated using the program CTFFIND3 to assess drift, astigmatism and used for CTF correction. The EMAN program CTFIT was used for flipping the phases in each of the micrographs. An initial model was generated with startcsym program imposing a C3 symmetry that was then refined for eight iterations using refine algorithm of EMAN. Later, the multirefine program was used to sort the particles in the dataset. The highest quality particles were extracted from the dataset and refined to a final resolution of 16 A˚. ACCESSION NUMBERS The accession numbers for the TF55 all-b, TF55 abshortC, and TF55 ablongC crystal structures reported in this paper are PDB: 4XCD, 4XCG, and 4XCI, respectively. The accession number for the abg EM reconstruction reported in this paper is EMDB: EMD-6291. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, seven figures, one table and one movie and can be found with this article online at http://dx.doi.org/10.1016/j.str.2015.12.016. AUTHOR CONTRIBUTIONS D.S. and A.G.S. designed and performed experiments, analyzed data and wrote manuscript, J.J.C. performed the majority of the biochemical

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 9

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

experiments, C.S. refined X-ray structures and processed EM data, D.A. collected X-ray data, A.W., B.A., S.S., S.K.M., and R.A.B. collected and processed EM data.

DiMaio, F., Echols, N., Headd, J.J., Terwilliger, T.C., Adams, P.D., and Baker, D. (2013). Improved low-resolution crystallographic refinement with Phenix and Rosetta. Nat. Methods 10, 1102–1104.

ACKNOWLEDGMENTS

Ditzel, L., Lowe, J., Stock, D., Stetter, K.O., Huber, H., Huber, R., and Steinbacher, S. (1998). Crystal structure of the thermosome, the archaeal chaperonin and homolog of CCT. Cell 93, 125–138.

We would like to thank Murray Stewart, MRC Laboratory of Molecular Biology, Cambridge, UK for help with data processing and structure determination and Joel Mackay and Ana Silva, University of Sydney, Australia for collecting the dataset for the ablongC structure. Mei Ling, NTU Singapore is acknowledged for help with electron microscopy. The Australian Synchrotron MX2 beamline staff is acknowledged for support during data collection. This work was funded by the Australian Research Council Discovery Project Grant 1093649. AS was supported by a National Health and Medical Research Council Fellowship 1090408. Part of the electron microscopy work was funded by both the Welch Foundation Award AH-1649 and NSF-MRI Award DBI-0923437 to R.A.B.. A.W., S.S., and B.A. are supported by the Singapore Ministry of Education Academic Research Fund Tier 3 (MOE2012-T3-1-001). A.R. and S.S. are supported by Nanyang Structural Biology Institute, NTU, Singapore. Received: June 2, 2015 Revised: December 14, 2015 Accepted: December 16, 2015 Published: February 4, 2016 REFERENCES

Douglas, N.R., Reissmann, S., Zhang, J., Chen, B., Jakana, J., Kumar, R., Chiu, W., and Frydman, J. (2011). Dual action of ATP hydrolysis couples lid closure to substrate release into the group II chaperonin chamber. Cell 144, 240–252. Ellis, M.J., Knapp, S., Koeck, P.J.B., Fakoor-Biniaz, Z., Ladenstein, R., and Hebert, H. (1998). Two-dimensional crystallization of the chaperonin TF55 from the hyperthermophilic archaeon Sulfolobus solfataricus. J. Struct. Biol. 123, 30–36. Emsley, P., and Cowtan, K. (2004). Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132. Evans, P.R., and Murshudov, G.N. (2013). How good are my data and what is the resolution? Acta Crystallogr. D Biol. Crystallogr. 69, 1204–1214. Fei, X., Yang, D., LaRonde-LeBlanc, N., and Lorimer, G.H. (2013). Crystal structure of a GroEL-ADP complex in the relaxed allosteric state at 2.7 A resolution. Proc. Natl. Acad. Sci. USA 110, E2958–E2966. Gregersen, N., Bross, P., Vang, S., and Christensen, J.H. (2006). Protein misfolding and human disease. Annu. Rev. Genomics Hum. Genet. 7, 103–124. Horovitz, A., and Willison, K.R. (2005). Allosteric regulation of chaperonins. Curr. Opin. Struct. Biol. 15, 646–651. Horwich, A.L., Neupert, W., and Hartl, F.U. (1990). Protein-catalysed protein folding. Trends Biotechnol. 8, 126–131.

Adams, P.D., Afonine, P.V., Bunkoczi, G., Chen, V.B., Davis, I.W., Echols, N., Headd, J.J., Hung, L.W., Kapral, G.J., Grosse-Kunstleve, R.W., et al. (2010). PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213–221.

Horwich, A.L., Fenton, W.A., Chapman, E., and Farr, G.W. (2007). Two families of chaperonin: physiology and mechanism. Annu. Rev. Cell Dev. Biol. 23, 115–145.

Archibald, J.M., Logsdon, J.M., and Doolittle, W.F. (1999). Recurrent paralogy in the evolution of archaeal chaperonins. Curr. Biol. 9, 1053–1056.

Huo, Y., Hu, Z., Zhang, K., Wang, L., Zhai, Y., Zhou, Q., Lander, G., Zhu, J., He, Y., Pang, X., et al. (2010). Crystal structure of group II chaperonin in the open state. Structure 18, 1270–1279.

Archibald, J.M., Blouin, C., and Doolittle, W.F. (2001). Gene duplication and the evolution of group II chaperonins: implications for structure and function. J. Struct. Biol. 135, 157–169. Bigotti, M.G., and Clarke, A.R. (2005). Cooperativity in the thermosome. J. Mol. Biol. 348, 13–26. Bigotti, M.G., and Clarke, A.R. (2008). Chaperonins: the hunt for the Group II mechanism. Arch. Biochem. Biophys. 474, 331–339. Booth, C.R., Meyer, A.S., Cong, Y., Topf, M., Sali, A., Ludtke, S.J., Chiu, W., and Frydman, J. (2008). Mechanism of lid closure in the eukaryotic chaperonin TRiC/CCT. Nat. Struct. Mol. Biol. 15, 746–753. Braig, K., Otwinowski, Z., Hegde, R., Boisvert, D.C., Joachimiak, A., Horwich, A.L., and Sigler, P.B. (1994). The crystal structure of the bacterial chaperonin GroEL at 2.8 A. Nature 371, 578–586. Chen, V.B., Arendall, W.B., Headd, J.J., Keedy, D.A., Immormino, R.M., Kapral, G.J., Murray, L.W., Richardson, J.S., and Richardson, D.C. (2010). MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12–21. Clare, D.K., Stagg, S., Quispe, J., Farr, G.W., Horwich, A.L., and Saibil, H.R. (2008). Multiple states of a nucleotide-bound group 2 chaperonin. Structure 16, 528–534. Collaborative Computational Project Number 4 (1994). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50, 760–763.

Joachimiak, L.A., Walzthoeni, T., Liu, C.W., Aebersold, R., and Frydman, J. (2014). The structural basis of substrate recognition by the eukaryotic chaperonin TRiC/CCT. Cell 159, 1042–1055. Kabsch, W. (2010). Xds. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132. Kagawa, H.K., Osipiuk, J., Maltsev, N., Overbeek, R., Quaite-Randall, E., Joachimiak, A., and Trent, J.D. (1995). The 60 kDa heat shock proteins in the hyperthermophilic archaeon Sulfolobus shibatae. J. Mol. Biol. 253, 712–725. Kagawa, H.K., Yaoi, T., Brocchieri, L., McMillan, R.A., Alton, T., and Trent, J.D. (2003). The composition, structure and stability of a group II chaperonin are temperature regulated in a hyperthermophilic archaeon. Mol. Microbiol. 48, 143–156. Kalisman, N., Schroder, G.F., and Levitt, M. (2013). The crystal structures of the eukaryotic chaperonin CCT reveal its functional partitioning. Structure 21, 540–549. Kapatai, G., Large, A., Benesch, J.L., Robinson, C.V., Carrascosa, J.L., Valpuesta, J.M., Gowrinathan, P., and Lund, P.A. (2006). All three chaperonin genes in the archaeon Haloferax volcanii are individually dispensable. Mol. Microbiol. 61, 1583–1597. Knapp, S., Schmidt-Krey, I., Hebert, H., Bergman, T., Jornvall, H., and Ladenstein, R. (1994). The molecular chaperonin TF55 from the Thermophilic archaeon Sulfolobus solfataricus. A biochemical and structural characterization. J. Mol. Biol. 242, 397–407.

Cong, Y., Schroder, G.F., Meyer, A.S., Jakana, J., Ma, B., Dougherty, M.T., Schmid, M.F., Reissmann, S., Levitt, M., Ludtke, S.L., et al. (2012). Symmetry-free cryo-EM structures of the chaperonin TRiC along its ATPase-driven conformational cycle. EMBO J. 31, 720–730.

Lopez, T., Dalton, K., and Frydman, J. (2015). The mechanism and function of group II chaperonins. J. Mol. Biol. 427, 2919–2930.

Dekker, C., Roe, S.M., McCormack, E.A., Beuron, F., Pearl, L.H., and Willison, K.R. (2011). The crystal structure of yeast CCT reveals intrinsic asymmetry of eukaryotic cytosolic chaperonins. EMBO J. 30, 3078–3090.

Luo, H., and Robb, F.T. (2011). A modulator domain controlling thermal stability in the Group II chaperonins of Archaea. Arch. Biochem. Biophys. 512, 111–118.

Ludtke, S.J. (2010). 3-D structures of macromolecules using single-particle analysis in EMAN. Methods Mol. Biol. 673, 157–173.

10 Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved

Please cite this article in press as: Chaston et al., Structural and Functional Insights into the Evolution and Stress Adaptation of Type II Chaperonins, Structure (2016), http://dx.doi.org/10.1016/j.str.2015.12.016

Marco, S., Urena, D., Carrascosa, J.L., Waldmann, T., Peters, J., Hegerl, R., Pfeifer, G., Sack-Kongehl, H., and Baumeister, W. (1994). The molecular chaperone TF55. Assessment of symmetry. FEBS Lett. 341, 152–155. McCoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., Winn, M.D., Storoni, L.C., and Read, R.J. (2007). Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674. Munoz, I.G., Yebenes, H., Zhou, M., Mesa, P., Serna, M., Park, A.Y., BragadoNilsson, E., Beloso, A., de Carcer, G., Malumbres, M., et al. (2011). Crystal structure of the open conformation of the mammalian chaperonin CCT in complex with tubulin. Nat. Struct. Mol. Biol. 18, 14–19. Pilak, O., Harrop, S.J., Siddiqui, K.S., Chong, K., De Francisci, D., Burg, D., Williams, T.J., Cavicchioli, R., and Curmi, P.M. (2011). Chaperonins from an Antarctic archaeon are predominantly monomeric: crystal structure of an open state monomer. Environ. Microbiol. 13, 2232–2249. Prusiner, S.B. (2012). Cell biology. A unifying role for prions in neurodegenerative diseases. Science 336, 1511–1513. Quaite-Randall, E., Trent, J.D., Josephs, R., and Joachimiak, A. (1995). Conformational cycle of the archaeosome, a TCP1-like chaperonin from Sulfolobus shibatae. J. Biol. Chem. 270, 28818–28823. Reissmann, S., Joachimiak, L.A., Chen, B., Meyer, A.S., Nguyen, A., and Frydman, J. (2012). A gradient of ATP affinities generates an asymmetric power stroke driving the chaperonin TRIC/CCT folding cycle. Cell Rep. 2, 866–877. Saibil, H. (2013). Chaperone machines for protein folding, unfolding and disaggregation. Nat. Rev. Mol. Cell Biol. 14, 630–642. Saibil, H.R., Fenton, W.A., Clare, D.K., and Horwich, A.L. (2013). Structure and allostery of the chaperonin GroEL. J. Mol. Biol. 425, 1476–1487. Schoehn, G., Quaite-Randall, E., Jimenez, J.L., Joachimiak, A., and Saibil, H.R. (2000). Three conformations of an archaeal chaperonin, TF55 from Sulfolobus shibatae. J. Mol. Biol. 296, 813–819. Skjaerven, L., Cuellar, J., Martinez, A., and Valpuesta, J.M. (2015). Dynamics, flexibility, and allostery in molecular chaperonins. FEBS Lett. 589, 2522–2532. Strong, M., Sawaya, M.R., Wang, S., Phillips, M., Cascio, D., and Eisenberg, D. (2006). Toward the structural genomics of complexes: crystal structure of a

PE/PPE protein complex from Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 103, 8060–8065. Tam, S., Geller, R., Spiess, C., and Frydman, J. (2006). The chaperonin TRiC controls polyglutamine aggregation and toxicity through subunit-specific interactions. Nat. Cell Biol. 8, 1155–1162. Tam, S., Spiess, C., Auyeung, W., Joachimiak, L., Chen, B., Poirier, M.A., and Frydman, J. (2009). The chaperonin TRiC blocks a huntingtin sequence element that promotes the conformational switch to aggregation. Nat. Struct. Mol. Biol. 16, 1279–1285. Trent, J.D., Osipiuk, J., and Pinkau, T. (1990). Acquired thermotolerance and heat shock in the extremely thermophilic archaebacterium Sulfolobus sp. strain B12. J. Bacteriol. 172, 1478–1484. Trent, J.D., Nimmesgern, E., Wall, J.S., Hartl, F.U., and Horwich, A.L. (1991). A molecular chaperone from a thermophilic archaebacterium is related to the eukaryotic protein t-complex polypeptide-1. Nature 354, 490–493. Vagin, A.A., Steiner, R.A., Lebedev, A.A., Potterton, L., McNicholas, S., Long, F., and Murshudov, G.N. (2004). REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for its use. Acta Crystallogr. D Biol. Crystallogr. 60, 2184–2195. Waldmann, T., Nitsch, M., Klumpp, M., and Baumeister, W. (1995). Expression of an archaeal chaperonin in E. coli: formation of homo- (alpha, beta) and hetero-oligomeric (alpha+beta) thermosome complexes. FEBS Lett. 376, 67–73. Xu, Z., Horwich, A.L., and Sigler, P.B. (1997). The crystal structure of the asymmetric GroEL-GroES-(ADP)7 chaperonin complex. Nature 388, 741–750. Yebenes, H., Mesa, P., Munoz, I.G., Montoya, G., and Valpuesta, J.M. (2011). Chaperonins: two rings for folding. Trends Biochem. Sci. 36, 424–432. Zhang, J., Baker, M.L., Schroder, G.F., Douglas, N.R., Reissmann, S., Jakana, J., Dougherty, M., Fu, C.J., Levitt, M., Ludtke, S.J., et al. (2010). Mechanism of folding chamber closure in a group II chaperonin. Nature 463, 379–383. Zhang, K., Wang, L., Liu, Y., Chan, K.Y., Pang, X., Schulten, K., Dong, Z., and Sun, F. (2013). Flexible interwoven termini determine the thermal stability of thermosomes. Protein Cell 4, 432–444.

Structure 24, 1–11, March 1, 2016 ª2016 Elsevier Ltd All rights reserved 11