Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
1
Cryo-EM Structure of the Human Mitochondrial Translocase 1
TIM22 Complex 2
Liangbo Qi1, 5
, Qiang Wang1, 5
, Zeyuan Guan1, 5
, Yan Wu1, Jianbo Cao
2, Xing Zhang
3, 3
Chuangye Yan4, #
, and Ping Yin1, 6, #
4
5
1National Key Laboratory of Crop Genetic Improvement and National Centre of Plant Gene 6
Research, Huazhong Agricultural University, Wuhan 430070, China. 7
2Public Laboratory of Electron Microscopy, Huazhong Agricultural University, Wuhan, 8
China 9
3Department of Biophysics, and Department of Pathology of Sir Run Run Shaw Hospital, 10
Zhejiang University School of Medicine, Center of Cryo Electron Microscopy, Zhejiang 11
University, Hangzhou, China. 12
4Beijing Advanced Innovation Center for Structural Biology, Tsinghua-Peking Joint Center 13
for Life Sciences, School of Life Sciences, Tsinghua University, Beijing 100084, China 14
5These authors contributed equally to this work. 15
6Lead contact 16
17
#To whom correspondence should be addressed. E-mail: [email protected] to Ping 18
Yin or [email protected] to Chuangye Yan 19
20
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
2
Abstract 21
Mitochondria play vital functions in cellular metabolism, homeostasis, and apoptosis1-3
. 22
Most of the mitochondrial proteins are synthesized as precursors in the cytosol and 23
imported into mitochondria for folding or maturation4,5
. The translocase TIM22 24
complex is responsible for the import of multiple hydrophobic carrier proteins that are 25
then folded in the inner membrane of mitochondria6-8
. In mammalian cells, the TIM22 26
complex consists of at least six components, Tim22, Tim29, AGK, and three Tim 27
chaperones (Tim9, Tim10a and Tim10b)9-14
. Here, we report the cryo-EM structure of 28
the human translocase TIM22 complex at an overall resolution of 3.7 angstrom. The 29
core subunit, Tim22, contains four transmembrane helices, forming a partial pore that is 30
open to the lipid bilayer. Tim29 is a single transmembrane protein that provides an 31
N-terminal helix to stabilize Tim22 and a C-terminal intermembrane space (IMS) 32
domain to connect AGK and two TIM chaperone hexamers to maintain complex 33
integrity. One TIM hexamer comprises Tim9 and Tim10a in a 3:3 molar ratio, and the 34
other consists of two Tim9 units, three Tim10a units, and one Tim10b unit. The latter 35
hexamer faces the intramembrane region of Tim22, likely providing the dock to load the 36
precursors to the partial pore of Tim22. Our structure serves as a molecular basis for 37
the mechanistic understanding of TIM22 complex function. 38
39
40
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
3
Mitochondria are essential eukaryotic cellular organelles with multiple vital functions, 41
such as energetics, metabolism, and cellular signaling1,2
. These functions are performed by 42
more than 1,000 proteins3,5
. However, only a small set of these proteins are synthesized in the 43
mitochondria; most of the mitochondria functioning proteins (~99%) are encoded by nuclear 44
genes, and imported into the correct mitochondrial compartment by specific preprotein 45
translocase complexes3,5,15,16
. These complexes include the translocase of outer membrane 46
(the TOM complex)17-20
, the carrier translocase of inner membrane complex (the TIM22 47
complex)21-24
, the presequence translocase of the inner membrane (the TIM23 complex)4,19,25
, 48
the sorting and assembly machinery (the SAM complex)26,27
, and the mitochondrial import 49
complex (the MIM complex)28,29
, etc. These translocation machineries are crucial for 50
mitochondrial biogenesis, dynamics, and function3,4,30,31
. Their dysregulations are connected 51
to dozens of rare mitochondrial diseases and disorders, suggesting an involvement in disease 52
pathogenesis3,31-33
. 53
54
The carrier pathway is one major protein import pathway that is responsible for the 55
translocation and insertion of carrier proteins into the mitochondrial inner membrane4,5,34-38
. 56
The carrier precursors pass through the pore of the TOM complex, and are then transferred by 57
a hexameric small TIM chaperon complex (the Tim9/Tim10a complex is a main form) to the 58
TIM22 complex4,5
. Consequently, the TIM22 complex mediates the insertion and lateral 59
release of precursors into the inner membrane in a membrane potential-dependent manner39
. 60
In addition, the TIM22 complex also transports other multiple-transmembrane-segments 61
containing inner-membrane proteins, including members of the Tim17, Tim22, and Tim23 62
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
4
family40,41
. 63
64
In human, the TIM22 complex consists of at least six components: a hypothetic 65
channel-forming protein Tim229,21,24
; three chaperone components (Tim9, Tim10a and 66
Tim10b)10,42
; newly identified Tim2911,12
and acylglycerol kinase (AGK)13,14
(Fig. 1a). Tim22 67
and small chaperones are well conserved from yeast to mammals9, but Tim29 and AGK are 68
specific in metazoan11-14
. AGK participates in lipid biosynthesis43
, and mutations in AGK gene 69
have been associated with Sengers syndrome44-48
. Mutations in the TIMM8A gene (also 70
known as DDP1 encoding a small chaperone homologue TIM8A) cause deafness dystonia 71
syndrome49,50
. Recently, mutations in the TIM22 gene have been identified to cause 72
early-onset mitochondrial myopathy51
. 73
74
Despite advances in our understanding of the function and pathophysiology of the 75
TIM22 complex, structural characterization has been sparse. The limited structural 76
information of the TIM22 complex restricted to the crystal structures of Tim9/Tim10a 77
hexameric chaperone and homologues, and a nuclear magnetic resonance (NMR) analysis of 78
carrier precursor associated Tim9/Tim10a52-55
. Electron microscopy has been also applied to 79
structural studies of translocase complexes. The negatively stained images and the recent 80
cryoelectron microscopy (cryo-EM) structure of the TOM core complex at approximately 6.8 81
Å revealed pore-forming architecture56-59
. In contrast, only the low-resolution electron 82
micrographs of the yeast TIM22 complex revealed a two-pore like shape, which was 83
supposed to display a smaller pore diameter than the pores of the TOM core complex39
. 84
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
5
85
Here, we report the structure of the human TIM22 complex at resolutions of 3.7 Å for 86
the overall structure and 3.5 Å for the inter-membrane region, determined using 87
single-particle cryo-EM. The structure reveals the assembly and detailed structural 88
information of the complex, and provides an important framework for understanding the 89
function and mechanism of the TIM22 complex in carrier protein maturation. 90
91
We co-expressed all six components of the TIM22 complex in human embryonic kidney 92
(HEK) 293F cells. After Flag-tag affinity purification followed by gel filtration, the resulting 93
TIM22 complex displayed good solution (Fig. 1b). The apparent molecular weight was 94
approximately 440 kDa assessed by blue native PAGE (Extended Data Fig. 1), in line to 95
previous findings. Analysis of the purified complex by mass spectrometry (MS) confirmed the 96
presence of all the components of the TIM22 complex. Details of grid preparation, cryo-EM 97
data acquisition, and structural determination of the TIM22 complex can be found in the 98
Methods. Following the initial 2D classification, 3D classification and refinement of the 99
cryo-EM particle images yielded a final 3D EM reconstruction map at an overall 3.7 Å 100
resolution and 3.5 Å resolution for the inter-membrane region of out of 482,959 selected 101
particles (Fig. 1c, Extended Data Fig. 2 and Table 1), according to the gold-standard Fourier 102
shell correlation (FSC) 0.143 criterion (Extended Data Fig. 3). Atomic models were built into 103
the map for Tim22, Tim9, Tim10a, Tim10b, Tim29 and AGK (Fig. 1d and Extended Data 104
Table 2; examples of local densities are shown in Extended Data Fig. 4). The densities for the 105
Tim29 N-terminal helix were of lower resolutions, and poly-Ala was assigned to this region. 106
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
6
107
The overall structure of the TIM22 complex was approximately 100 Å in height and 160 108
Å in the longest dimension of width (Fig. 1e). In the structure, there was one Tim22, one 109
Tim29, one AGK, and two hexamer chaperones, Tim9/10a and Tim9/10a/10b, whose 110
stoichiometries were 3:3 and 2:3:1, respectively (Extended Data Fig. 5). Most of the 111
structures were located at the intermembrane space (IMS), including the N-terminus of Tim22 112
and the large extended C-terminus portion of Tim29 (Extended Data Fig. 5a). Four 113
transmembrane segments (TMs) of Tim22 together with a single TM of Tim29 constitute the 114
transmembrane element at the center of the TIM22 complex. Only one N-terminal helix of 115
Tim29 protrudes from the core transmembrane region and was oriented nearly parallel to the 116
plane of the membrane at the matrix side (Fig. 1e). The Tim9/10a/10b hexamer, similar to a 117
hub, was located at the center of TIM22 complex, and was encircled by the N-terminus of 118
Tim22, the middle portion of Tim29, AGK, and the Tim9/10a chaperone. Interestingly, the 119
Tim9/10a/10b hexamer was not perpendicular to the membrane, but rather approximately 45° 120
tilted (Fig. 1e). The Tim9/10a chaperone and AGK were located at the nearly-opposite side of 121
the Tim9/10a/10b hexamer, and AGK was anchored to the membrane via its portion of the 122
N-terminal helix and an additional membrane anchor helix loop helix (Fig. 1e). 123
124
Tim22 contains two helices (1 and 2) connected by an extended loop 1, and 4 125
transmembrane segments (TM 1-4) (Fig. 2a and Extended Data Fig. 5a). The 23 N-terminal 126
residues and the matrix loop (residues 94-118; connection between TM1 and TM2) had no 127
EM density, most likely due to their intrinsic flexibility. Two helices (1 and 2) protrude 128
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
7
toward the intermembrane space and interact with the Tim9/10a/10b hexamer. Helix 1 and 129
loop 1, similar to a hook, sinuously wind around the groove between the inner helices and the 130
outer helices of Tim9 and Tim10a. Helix 1 was nestled in a greasy pocket formed by Tim9 131
and Tim10a (Fig. 2b), and likely functioned as a plug to obstruct the hydrophobic carrier 132
precursor from wedging within the hexamer chaperone from this side (Extended Data Fig. 6a). 133
The recently identified disease-related mutation (Val33Leu) of Tim22 was located in the helix 134
151
. 135
136
Specific interactions comprise van der Waals contacts and one hydrogen bond. 137
Hydrophobic residues, Leu28, Leu29, Leu32, and Val33 interacted with residues Leu9, Leu13, 138
Met17, and Leu69 from helices of Tim9, and residues Phe14, Phe17, and Leu18 from the 139
inner helix of Tim10a. The aromatic ring of Tyr25 from Tim22 stacked against Phe14 and 140
Phe17 from Tim10a. Asn75 from Tim10a, was found to form a hydrogen bond with the main 141
chain of Arg40 from Tim22 (Fig. 2c). Reinforcing these interactions, one residue from helix 142
2 and four residues from the last turn of TM2 of Tim22 contacted residues of the outer helix 143
of Tim9. Glu67 from Tim22 interacted with Lys45 from Tim9 to form a salt bridge. The side 144
chains of Glu144 and Ser150 from Tim22 formed hydrogen bonds with Ser44 and the main 145
chain of Gly46 from Tim9. In addition, the main chains of Ser145 and Tyr146 from Tim22 146
accepted hydrogen bonds from Arg53 of Tim9 (Fig. 2c). 147
148
Notably, only one Tim22 was observed in the complex. The four TMs of Tim22 were 149
unlike a closed pore channel but constituted a lateral hydrophobic cave that was exposed to 150
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
8
the lipid bilayer (Fig. 2a), which is reminiscent of insertase60
. A disulfide bond formed 151
between Cys69 and Cys141 appeared to stabilize the conformations of TM1 and TM2 152
(Extended Data Fig. 6b, c), which was consistent with biochemical studies. Tim22 153
homologues from yeast to human share more than 40% similarity, and most of the interacting 154
residues and two cysteines are conserved (Extended Data Fig. 6c), suggesting that the Tim22 155
homologues exhibit similar folds9. 156
157
Tim29 was recently identified as a metazoan-specific subunit of the human TIM22 158
complex, which is required for the stability of the complex. Our structure corroborated this 159
observation, since. Tim29 exhibited an extended conformation comprising of a long 160
N-terminal helix 1 in the matrix, a single TM, an intermembrane space domain (IMS 161
Domain) and a C-terminal chaperone Tim9/10a recruiting motif (CRM) (Fig. 3a and Extended 162
Data Fig. 5a). One phospholipid, which was most likely to be phosphatidylethanolamine, was 163
observed at the IMS (Extended Data Fig. 7a, b). The helix 1 was oriented nearly parallel to 164
the plane of the membrane in the matrix and perpendicular to the TM of Tim29 (Fig. 3a, b), 165
which appeared to interact with TM3 of Tim22 and stabilize it. However, the lack of EM 166
density for the side chains of helix 1 prevented an unambiguous assessment of the details of 167
these molecular interactions. Interestingly, the single TM of Tim29 was positioned far from 168
the TMs of Tim22, of some extent, suggesting this TM was unable to strictly associate with 169
TMs of Tim22 (Fig. 3b). This arrangement suggests the TM is flexible for regulation of 170
precursor insertion. 171
172
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
9
The IMS domain of Tim29 interacts with the Tim9/10a/10b hexamer mainly through the 173
helix4, as illustrated by three polar residues Glu107, Gln111, and Arg119 hydrogen-bonding 174
with His37, Ser35, Arg31 from Tim10b, respectively (Fig. 3b, d). Additionally, the main 175
chains of Ser143 from Tim29 and Lys32 from Tim10a form a hydrogen bond (Fig. 3d). These 176
interactions robustly fix the orientation of the Tim9/10a/10b hexamer, in part due to the 177
presence of only one Tim10b in the heterohexamer (Fig. 3b and Extended Data Fig. 5a). 178
179
The CRM of Tim29 makes contact with the Tim9/10a chaperone via extensive 180
electrostatic interactions (Fig. 3b). The negatively charged residues of Asp183, Asp186, 181
Glu210 and Asp214 interact with several positively charged residues, Arg39 from Tim9, 182
including Arg53, Lys 45, Lys32, and Arg31 from Tim10a, respectively (Fig. 3c). Interestingly, 183
the external C-terminal helix, 9, has no interaction with the Tim9/10a chaperone. Given that 184
this portion is rich in negatively charged residues (Extended Data Fig. 7c, d), it might recruit 185
the other Tim9/Tim10a chaperone or interact with the TOM complex, which has been shown 186
to occur biochemically12
. 187
188
The Tim9/10a chaperone and the Tim9/10a/10b chaperone share almost identical 189
symmetric hexamer structures as that of the reported free Tim9/10a hexamer52
(Fig. 4 and 190
Extended Data Fig. 8a). Each individual subunit displays a helix-loop-helix fold and is 191
stabilized by two intramolecular disulfide bonds from a highly conserved “twin CX3C motif” 192
(Fig. 4b). The six inner helices (helix 1) collectively form a donut-shape with a 14~16 Å 193
diameter central hole. The outer helices (helix 2) act as tentacles that radiate from the central 194
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
10
region (Fig. 4a. b). Superimposing Tim9, Tim10a, and Tim10b revealed that the less 195
conserved connecting loops (C loops) exhibit distinct conformations (Fig. 4c), in line to the 196
observation that the C loop of Tim10b is primarily responsible for interaction with Tim29 (Fig. 197
3d). Furthermore, the C-terminus of the outer helix of Tim10b tilts towards the inner helix 198
(Fig. 4b), mainly due to its Pro72 residue (Fig. 4d). Thus, the major conformational difference 199
between these two chaperones appears to be the outer helix of Tim10b, while the diameter of 200
the Tim9/10a hole appears to be subtly narrowed (Extended Data Fig. 8b). A circular groove 201
was observed between the inner and outer helices, which was previously hypothesized to hold 202
unstructured carrier precursors52,55
. The twist outer helix of Tim10b disrupts the continuous 203
groove (Extended Data Fig. 8c, d), which might be involved in precursors unloading from the 204
Tim9/10a chaperone. The Tim9/10a chaperone and the Tim9/10a/10b hexamer were 205
selectively bound to the CRM and IMS domain of Tim29 as determined by Tim10b (Fig. 3c, 206
d). Tim10b interacted with the IMS domain of Tim29 and the DGK domain of AGK 207
(described below). These specific interactions prevented the binding of the Tim9/10a 208
chaperone in this position (Fig 3b). 209
210
AGK is a mitochondrial lipid kinase that converts monoacylglycerol (MAG) and 211
diacylglycerol (DAG) to lysophosphatidic acid (lyso-PA) and phosphatidic acid (PA), 212
respectively43
. Similar to the DAG kinase homologue, DgkB from Staphylococcus aureus 213
(SaDgkB), the AGK structure exhibits a typical two-domain fold61
(Fig. 5a and Extended 214
Data Fig. 5b). Notably, a predicted TM (helix 1) preceding domain 1 is partially embedded 215
in the membrane (Fig. 5a), and is consistent with previous findings13,14
. A protruded helix 9 216
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
11
and an ensuing loop of domain 2 were found to be anchored to the membrane via Trp225, 217
Tyr226, L227, L230, Phe237 and Phe238 (Fig. 5a), generating a positively-charged cavity 218
between AGK and the membrane to potentially facilitate the carrier precursor insertion (Fig. 219
5b). These membrane-anchoring structural features were found to be AGK-specific (Fig. 5c). 220
AGK interacts with both Tim29 and the Tim9/10a/10b hexamer, corroborating previous 221
studies that suggested AGK is a bona fide subunit of the TIM22 complex. Arg40 from helix 222
1 offers hydrogen bonds to the main chains of Tyr151 and Gln153 from Tim29 (Fig. 5d). 223
Gln52, Ala58, and Asp94 were found to hydrogen-bond with Lys45 from Tim10a, Arg62 from 224
Tim10b, and Arg39 from Tim9, respectively (Fig. 5e). 225
226
The structure of the TIM22 complex presented here not only offers the first molecular 227
level view of the interactions that facilitate the mammalian TIM22 complex assembly, but 228
also provides a framework for a mechanistic understanding of the import of carrier proteins 229
through the TIM22 complex. Typically, the carrier protein contains six TMs that are 230
assembled in three repeats, each comprising two TMs connected by a short matrix helix62-65
. 231
Both the amino and carboxy termini are oriented towards the intermembrane space63,64
. And, 232
the precursor protein is wrapped into the hydrophobic groove of the Tim9/10a chaperone55
. 233
Based on our structural analysis, we proposed a “bending-in” insertion model, wherein Tim22 234
and AGK generate a space-limited environment to recruit the two transmembrane helices of 235
the carrier proteins at once, and then bend them into the membrane (Extended Data Fig. 9). 236
After insertion of one repeat into the membrane, the two TM segments intermediate might be 237
transiently stabilized by the Tim22 until the next repeat inserts. This scenario the structures of 238
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
12
the TIM22 complex with the carrier precursor during an insertion cycle are necessitated to 239
fully understand the carrier translocation mechanism. 240
241
242
243
References 244
1 Harbauer, A. B., Zahedi, R. P., Sickmann, A., Pfanner, N. & Meisinger, C. The protein 245
import machinery of mitochondria-a regulatory hub in metabolism, stress, and disease. 246
Cell Metab 19, 357-372 (2014). 247
2 Kasahara, A. & Scorrano, L. Mitochondria: from cell death executioners to regulators 248
of cell differentiation. Trends Cell Biol 24, 761-770 (2014). 249
3 Pfanner, N., Warscheid, B. & Wiedemann, N. Mitochondrial proteins: from 250
biogenesis to functional networks. Nat Rev Mol Cell Biol 20, 267-284 (2019). 251
4 Chacinska, A., Koehler, C. M., Milenkovic, D., Lithgow, T. & Pfanner, N. Importing 252
mitochondrial proteins: machineries and mechanisms. Cell 138, 628-644 (2009). 253
5 Wiedemann, N. & Pfanner, N. Mitochondrial Machineries for Protein Import and 254
Assembly. Annu Rev Biochem 86, 685-714 (2017). 255
6 Rehling, P., Brandner, K. & Pfanner, N. Mitochondrial import and the twin-pore 256
translocase. Nat Rev Mol Cell Biol 5, 519-530 (2004). 257
7 Neupert, W. & Herrmann, J. M. Translocation of proteins into mitochondria. Annu 258
Rev Biochem 76, 723-749 (2007). 259
8 Bohnert, M., Pfanner, N. & van der Laan, M. Mitochondrial machineries for insertion 260
of membrane proteins. Curr Opin Struct Biol 33, 92-102 (2015). 261
9 Bauer, M. F. et al. The mitochondrial TIM22 preprotein translocase is highly 262
conserved throughout the eukaryotic kingdom. FEBS Letters 464, 41-47 (1999). 263
10 Muhlenbein, N., Hofmann, S., Rothbauer, U. & Bauer, M. F. Organization and 264
function of the small Tim complexes acting along the import pathway of metabolite 265
carriers into mammalian mitochondria. J Biol Chem 279, 13540-13546 (2004). 266
11 Callegari, S. et al. TIM29 is a subunit of the human carrier translocase required for 267
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
13
protein transport. FEBS Lett 590, 4147-4158 (2016). 268
12 Kang, Y. et al. Tim29 is a novel subunit of the human TIM22 translocase and is 269
involved in complex assembly and stability. Elife 5, e17463 (2016). 270
13 Kang, Y. et al. Sengers Syndrome-Associated Mitochondrial Acylglycerol Kinase Is a 271
Subunit of the Human TIM22 Protein Import Complex. Mol Cell 67, 457-470 (2017). 272
14 Vukotic, M. et al. Acylglycerol Kinase Mutated in Sengers Syndrome Is a Subunit of 273
the TIM22 Protein Translocase in Mitochondria. Mol Cell 67, 471-483 (2017). 274
15 Schleiff, E. & Becker, T. Common ground for protein translocation: access control for 275
mitochondria and chloroplasts. Nat Rev Mol Cell Biol 12, 48-59 (2011). 276
16 Schulz, C., Schendzielorz, A. & Rehling, P. Unlocking the presequence import 277
pathway. Trends Cell Biol 25, 265-275 (2015). 278
17 Kiebler, M. et al. Identification of a mitochondrial receptor complex required for 279
recognition and membrane insertion of precursor proteins. Nature 348, 610-616 280
(1990). 281
18 Hill, K. et al. Tom40 forms the hydrophilic channel of the mitochondrial import pore 282
for preproteins. Nature 395, 516-521 (1998). 283
19 Shiota, T. et al. Molecular architecture of the active mitochondrial protein gate. 284
Science 349, 1544-1548 (2015). 285
20 Mokranjac, D. & Neupert, W. Cell biology: Architecture of a protein entry gate. 286
Nature 528, 201-202 (2015). 287
21 Sirrenberg, C., Bauer, M. F., Guiard, B., Neupert, W. & Brunner, M. Import of carrier 288
proteins into the mitochondrial inner membrane mediated by Tim22. Nature 384, 289
582-585 (1996). 290
22 Koehler, C. M. et al. Import of mitochondrial carriers mediated by essential proteins 291
of the intermembrane space. Science 279, 369-673 (1998). 292
23 Sirrenberg, C. et al. Carrier protein import into mitochondria mediated by the 293
intermembrane proteins Tim10/Mrs11 and Tim12/Mrs5. Nature 391, 912-915 (1998). 294
24 Kovermann, P. et al. Tim22, the essential core of the mitochondrial protein insertion 295
complex, forms a voltage-activated and signal-gated channel. Mol Cell 9, 363-373 296
(2002). 297
25 Chacinska, A. et al. Mitochondrial presequence translocase: switching between TOM 298
tethering and motor recruitment involves Tim21 and Tim17. Cell 120, 817-829 299
(2005). 300
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
14
26 Wiedemann, N. et al. Machinery for protein sorting and assembly in the 301
mitochondrial outer membrane. Nature 424, 565-571 (2003). 302
27 Paschen, S. A. et al. Evolutionary conservation of biogenesis of beta-barrel 303
membrane proteins. Nature 426, 862-866 (2003). 304
28 Chacinska, A. et al. Essential role of Mia40 in import and assembly of mitochondrial 305
intermembrane space proteins. EMBO J 23, 3735-3746 (2004). 306
29 Naoe, M. et al. Identification of Tim40 that mediates protein sorting to the 307
mitochondrial intermembrane space. J Biol Chem 279, 47815-47821 (2004). 308
30 Opalinska, M. & Meisinger, C. Metabolic control via the mitochondrial protein 309
import machinery. Curr Opin Cell Biol 33, 42-48 (2015). 310
31 Kang, Y., Fielden, L. F. & Stojanovski, D. Mitochondrial protein transport in health 311
and disease. Semin Cell Dev Biol 76, 142-153 (2018). 312
32 Sokol, A. M., Sztolsztener, M. E., Wasilewski, M., Heinz, E. & Chacinska, A. 313
Mitochondrial protein translocases for survival and wellbeing. FEBS Lett 588, 314
2484-2495 (2014). 315
33 Martensson, C. U. & Becker, T. Acylglycerol Kinase: Mitochondrial Protein 316
Transport Meets Lipid Biosynthesis. Trends Cell Biol 27, 700-702 (2017). 317
34 Mokranjac, D. & Neupert, W. Thirty years of protein translocation into mitochondria: 318
unexpectedly complex and still puzzling. Biochimica et biophysica acta 1793, 33-41 319
(2009). 320
35 Wagner, K., Mick, D. U. & Rehling, P. Protein transport machineries for precursor 321
translocation across the inner mitochondrial membrane. Biochimica et biophysica 322
acta 1793, 52-59 (2009). 323
36 Webb, C. T. & Lithgow, T. Mitochondrial biogenesis: sorting mechanisms cooperate 324
in ABC transporter assembly. Current biology : CB 20, R564-567 (2010). 325
37 Marom, M., Azem, A. & Mokranjac, D. Understanding the molecular mechanism of 326
protein translocation across the mitochondrial inner membrane: still a long way to go. 327
Biochimica et biophysica acta 1808, 990-1001 (2011). 328
38 Becker, T., Bottinger, L. & Pfanner, N. Mitochondrial protein import: from transport 329
pathways to an integrated network. Trends Biochem Sci 37, 85-91 (2012). 330
39 Rehling, P. et al. Protein insertion into the mitochondrial inner membrane by a 331
twin-pore translocase. Science 299, 1747-1751 (2003). 332
40 Beverly, K. N., Sawaya, M. R., Schmid, E. & Koehler, C. M. The Tim8-Tim13 333
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
15
complex has multiple substrate binding sites and binds cooperatively to Tim23. J Mol 334
Biol 382, 1144-1156 (2008). 335
41 Curran, S. P., Leuenberger, D., Schmidt, E. & Koehler, C. M. The role of the 336
Tim8p-Tim13p complex in a conserved import pathway for mitochondrial polytopic 337
inner membrane proteins. J Cell Biol 158, 1017-1027 (2002). 338
42 Kerscher, O., Sepuri, N. B. & Jensen, R. E. Tim18p is a new component of the 339
Tim54p-Tim22p translocon in the mitochondrial inner membrane. Mol Biol Cell 11, 340
103-116 (2000). 341
43 Bektas, M. et al. A novel acylglycerol kinase that produces lysophosphatidic acid 342
modulates cross talk with EGFR in prostate cancer cells. J Cell Biol 169, 801-811 343
(2005). 344
44 Aldahmesh, M. A., Khan, A. O., Mohamed, J. Y., Alghamdi, M. H. & Alkuraya, F. S. 345
Identification of a truncation mutation of acylglycerol kinase (AGK) gene in a novel 346
autosomal recessive cataract locus. Hum Mutat 33, 960-962 (2012). 347
45 Haghighi, A. et al. Sengers syndrome: six novel AGK mutations in seven new 348
families and review of the phenotypic and mutational spectrum of 29 patients. 349
Orphanet J Rare Dis 9, 199 (2014). 350
46 Siriwardena, K. et al. Mitochondrial citrate synthase crystals: novel finding in 351
Sengers syndrome caused by acylglycerol kinase (AGK) mutations. Mol Genet Metab 352
108, 40-50 (2013). 353
47 Mayr, J. A. et al. Lack of the mitochondrial protein acylglycerol kinase causes 354
Sengers syndrome. Am J Hum Genet 90, 314-320 (2012). 355
48 Calvo, S. E. et al. Molecular diagnosis of infantile mitochondrial disease with 356
targeted next-generation sequencing. Sci Transl Med 4, 118ra110 (2012). 357
49 Koehler, C. M. et al. Human deafness dystonia syndrome is a mitochondrial disease. 358
Proc Natl Acad Sci U S A 96, 2141-2146 (1999). 359
50 Roesch, K., Curran, S. P., Tranebjaerg, L. & Koehler, C. M. Human deafness dystonia 360
syndrome is caused by a defect in assembly of the DDP1/TIMM8a-TIMM13 complex. 361
Hum Mol Genet 11, 477-486 (2002). 362
51 Pacheu-Grau, D. et al. Mutations of the mitochondrial carrier translocase channel 363
subunit TIM22 cause early-onset mitochondrial myopathy. Hum Mol Genet 27, 364
4135-4144 (2018). 365
52 Webb, C. T., Gorman, M. A., Lazarou, M., Ryan, M. T. & Gulbis, J. M. Crystal 366
structure of the mitochondrial chaperone TIM9.10 reveals a six-bladed 367
alpha-propeller. Mol Cell 21, 123-133 (2006). 368
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
16
53 Baker, M. J. et al. Structural and functional requirements for activity of the 369
Tim9-Tim10 complex in mitochondrial protein import. Mol Biol Cell 20, 769-779 370
(2009). 371
54 Endo, T., Yamano, K. & Kawano, S. Structural insight into the mitochondrial protein 372
import system. Biochimica et biophysica acta 1808, 955-970 (2011). 373
55 Weinhaupl, K. et al. Structural Basis of Membrane Protein Chaperoning through the 374
Mitochondrial Intermembrane Space. Cell 175, 1365-1379 (2018). 375
56 Ahting, U. et al. The TOM core complex: the general protein import pore of the outer 376
membrane of mitochondria. J Cell Biol 147, 959-968 (1999). 377
57 Kunkele, K. P. et al. The preprotein translocation channel of the outer membrane of 378
mitochondria. Cell 93, 1009-1019 (1998). 379
58 Model, K., Meisinger, C. & Kuhlbrandt, W. Cryo-electron microscopy structure of a 380
yeast mitochondrial preprotein translocase. J Mol Biol 383, 1049-1057 (2008). 381
59 Bausewein, T. et al. Cryo-EM Structure of the TOM Core Complex from Neurospora 382
crassa. Cell 170, 693-700 (2017). 383
60 Kumazaki, K. et al. Structural basis of Sec-independent membrane protein insertion 384
by YidC. Nature 509, 516-520 (2014). 385
61 Miller, D. J., Jerga, A., Rock, C. O. & White, S. W. Analysis of the Staphylococcus 386
aureus DgkB structure reveals a common catalytic mechanism for the soluble 387
diacylglycerol kinases. Structure 16, 1036-1046 (2008). 388
62 Endres, M., Neupert, W. & Brunner, M. Transport of the ADP/ATP carrier of 389
mitochondria from the TOM complex to the TIM22.54 complex. EMBO J 18, 390
3214-3221 (1999). 391
63 Nelson, D. R., Felix, C. M. & Swanson, J. M. Highly conserved charge-pair networks 392
in the mitochondrial carrier family. J Mol Biol 277, 285-308 (1998). 393
64 Nury, H. et al. Relations between structure and function of the mitochondrial 394
ADP/ATP carrier. Annu Rev Biochem 75, 713-741 (2006). 395
65 Pebay-Peyroula, E. et al. Structure of mitochondrial ADP/ATP carrier in complex 396
with carboxyatractyloside. Nature 426, 39-44 (2003). 397
398
Data availability 399
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
17
Cryo-EM map for the human TIM22 complex is available on the Electron Microscopy Data 400
Bank under accession number EMD-9958. Coordinate of the atomic structure have been 401
deposited in the Protein Data Bank under accession number 6KC9. All other data are 402
available from the corresponding authors upon reasonable request. 403
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
18
Main Figures 404
405
406
Fig. 1 | Cryo-EM structure of the human TIM22 complex. a, The schematic diagram for 407
each subunit of the TIM22 complex. b, A representative gel filtration chromatography of the 408
TIM22 complex. The peak fractions were pooled for cryo-EM study. c, A representative 409
electron micrograph of TIM22 complex. The typical particles are marked by yellow circles. d, 410
Cryo-EM density map of the TIM22 complex. Tim9, cyan; Tim10a, slate; Tim10b, magenta; 411
Tim22, orange; Tim29, yellow; AGK, chartreuse; the disc-shape micelle, gray. e, The 412
opposing side views of the TIM22 complex. The inner membrane (IM) is indicated by two 413
lines between the intermembrane space (IMS) and the matrix. All structure figures were 414
prepared using PyMol. 415
416
417
418
419
420
421
422
423
424
425
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
19
426
427
428
Fig. 2 | Structure of the Tim22 subunit. a, Two side views of Tim22. The transmembrane 429
(TM) segments are labeled. b, Tim22 contacts the Tim9/10a/10b chaperone mainly via the 430
N-terminal helix in the intermembrane space and the helices bundle at the inner membrane. 431
The two interfaces are highlighted by red and green box. c, Key residues mediating the 432
interactions between Tim22 subunit and Tim9/10a/10b chaperone are shown as sticks. 433
434
435
436
437
438
439
440
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
20
441
442
Fig. 3 | Structure of the Tim29 subunit. a, Tim29 subunit forms extensive contacts with 443
both Tim9/10a and Tim9/10a/10b chaperones via the C-terminal segments. The interacted 444
regions are indicated by pink and red box, respectively. b, The overall structure of the Tim29 445
subunit. The C-terminus of Tim29 is divided into two segments: IMS (intermembrane space) 446
domain and CRM (C-terminal recruiting motif). The secondary structural elements are labeled. 447
c, d, Key residues mediating the interactions between Tim29 subunit and Tim9/10a (c) or 448
Tim9/10a/10b (d) are shown as sticks. 449
450
451
452
453
454
455
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
21
456
457
Fig. 4 | Structure of the Tim9/10a/10b hexamer. a, Two perpendicular views of 458
Tim9/10a/10b hexamer. The Tim9/10a/10b hexamer forms a pore. b, Structural 459
superimposition of Tim9, Tim10a and Tim10b. Disulfide bonds formed from the signature 460
cysteines of twin CX3C motif are indicated. The C (center) loop is highlighted by green box. c, 461
Structural alignment of the C-loop of Tim9, Tim10a, and Tim10b. d, Sequence alignment of 462
Tim9, Tim10a and Tim10b. The conserved residues are colored red. Residues in the C loop 463
involved in the interaction with the Tim29 subunit are colored green. Cysteines in the twin 464
CX3C motif forming the disulfide bonds are shown with yellow numbers. Proline72 is colored 465
purple. The inner and outer helices are drawn above the sequences. 466
467
468
469
470
471
472
473
474
475
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
22
476
477
Fig. 5 | Structure of the AGK subunit. a, Two opposing views of AGK, which is composed 478
by domain 1 and domain 2. The first -helix (light pink) is embedded into the membrane. 479
Key residues bound to the membrane are shown as orange sticks. b, A positively charged 480
cavity of AGK is formed in between the first -helix and the anchor. c, Structural alignment 481
of AGK (Homo sapiens) and DgkB (Staphylococcus aureus, PDB code: 2VQ7). AGK 482
exhibits additional membrane anchor (a dashed circle) and the transmembrane d, e, Key 483
residues involved in the interactions between AGK and Tim29 (d) or Tim9/10a/10b chaperone 484
(e) are shown as sticks. 485
486
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
23
Methods 487
Transient expression of the human TIM22 complex 488
The codon-optimized full-length cDNAs for subunits of the human TIM22 complex were 489
synthesized by General Biosystems Company (Tim22, Uniport: Q9Y584; Tim29, Uniport: 490
Q9BSF4; Tim9, Uniport: Q9Y5J7; Tim10a, Uniport: P62072, Tim10b, Uniport: Q9Y5J6; 491
AGK, Uniport: Q53H12). TIM22, TIM29 and AGK were subcloned into the plasmid A; TIM9, 492
TIM10A and TIM10B were subcloned into the plasmid B using the pMlink vector1. Tim22 493
contains a N-terminal triple Flag tag. Expi293FTM
(Invitrogen) cells were cultured in SMM 494
293TI medium (Sino Biological Inc.) at 37 °C under 5% CO2 in a shaker and diluted into 2.0 495
× 106 cells ml
-1 with fresh medium when the cell density reached 3.5 × 10
6 ~ 4.0 × 10
6 cells 496
ml-1
for further transfection. For 1 liter cell culture, 1.4 mg plasmid A and 0.6 mg plasmid B 497
were pre-incubated with 4 mg linear polyethylenimines (PEIs) (Polysciences) in 50 ml fresh 498
medium for 20 min. The transfection was initiated by adding the mixture into the diluted cell 499
culture. Transfected cells were cultured for 48 hours before harvesting. 500
501
Mitochondria Preparation 502
The cells were harvested by centrifugation at 800 g for 20 min and washed with PBS, then 503
resuspended in the buffer A containing 10 mM Tris-HCl pH 7.5, 70 mM sucrose, 210 mM 504
mannitol, 1 mM EDTA, 1mg ml-1
BSA and 1mM PMSF2. The cells were disrupted using 505
dounce homogenizer (sigma) for 80 cycles on ice and the homogenate was centrifuged at 506
3,000 g for 10 min. The supernatant was further centrifuged at 20,000 g for 20 min to obtain 507
the crude mitochondrial pellet. The pellet was resuspended in buffer B containing10 mM Tris 508
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
24
pH 7.5, 250 mM sucrose, 60 mM KCl and 0.1 mM EDTA and centrifuged at 60,000 g at 4 °C 509
for 20 min in a discontinuous Percoll density gradient3. The clear mitochondria layer was 510
obtained carefully and diluted with buffer C containing 10 mM Tris-HCl pH 7.5, 70 mM 511
sucrose, 210 mM mannitol. The highly pure mitochondria were collected by centrifugation at 512
20,000 g for 30 min and stored in buffer C at -80 °C before use. 513
514
Purification of the TIM22 complex 515
The TIM22 complex were extracted from pure mitochondria by 1% LMNG (Anatrace), 0.25% 516
soybean lipids (Sigma) and 0.1 % CHS (Anatrace) in lysis buffer containing 25 mM HEPES 517
pH 7.4, 100 mM KOAc, 10 mM Mg(OAc)2, 0.1 mM EDTA, 10% glycerol, 1 mM PMSF, 4 518
µg ml-1
pepstatin A, 4 µg ml-1
aprotinin and 5 µg ml-1
leupeptin at 4 °C for 1.5 h4. The 519
extraction was centrifuged at 100,000 g for 30 min to remove the unsoluble component. The 520
supernatant was incubated with anti-Flag G1 affinity resin (Genscript) at 4 °C for 2 h and then 521
washed with 30 bed volumes of lysis buffer added 0.05% GDN (Anatrace). The protein was 522
eluted with lysis buffer added 0.05% GDN and 300 g ml-1
Flag peptide (Genscript). The 523
protein solution was concentrated with a 100-kDa cut-off centricon (Milipore) and further 524
purified by Superose-6 increase 10/300 column (GE Healthcare) using a buffer containing 50 525
mM Imidazole pH 6.0, 150 mM NaCl, 2 mM MgCl2 and 0.05% GDN5. The peak fractions 526
were pooled and concentrated to 7 mg ml-1
for further cryo-EM study. 527
528
Blue native-PAGE analysis 529
Blue native PAGE technique was used to determine native TIM22 complex mass as described 530
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
25
previously6. Chromatographically purified TIM22 complex sample was mixed with 10 × 531
loading buffer (0.1% (w/v) Ponceau S, 50% (w/v) glycerol) and subjected to 4%-16% blue 532
native PAGE mini gel (Invitrogen) for electrophoresis at 4 °C. The TIM22 complex was 533
transferred onto a PVDF membrane and detected by immunoblotting using Tim22 antibody 534
(cat number: 14927-1-AP, Proteintech). The size of the TIM22 complex was determined 535
based on the mobility pattern of a native protein molecular weight standard (cat number: 536
LC0725, Invitrogen). 537
538
Mass spectrometry analysis 539
The TIM22 complex proteins were separated by 1D SDS-PAGE, the gel bands of interest 540
were excised from the gel, reduced with 5 mM of dithiotreitol, and alkylated with 11 mM 541
iodoacetamide. In gel digestion was then carried out with sequencing grade modified trypsin 542
in 50 mM ammonium bicarbonate at 37 °C overnight. The peptides were extracted twice with 543
0.1% trifluoroacetic acid in 50% acetonitrile aqueous solution for 30 min. Extracts were then 544
centrifuged in a speedvac to reduce the volume. Tryptic peptides were redissolved in 20 l 0.1% 545
TFA and analyzed by LC-MS/MS. 546
547
Cryo-EM Data Acquisition 548
For cryo-EM sample preparation, 3.5 l aliquots of the recombinant human TIM22 complex 549
(7 mg ml-1
) were dropped onto glow discharged holey carbon grids (Quantifoil Au R1.2/1.3, 550
300 mesh), blotted with a Vitrobot Mark IV (ThemoFisher Scientific) using 5 s blotting time 551
with 100% humidity at 8 °C, and plunged into liquid ethane cooled by liquid nitrogen. The 552
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
26
sample was imaged on an FEI Titan Krios transmission electron microscope at 300 kV with a 553
magnification of 29,000 ×. Images were recorded by a Gatan K2 Summit direct electron 554
detector using the counting mode. Defocus values varied from -1.8 to -2.5 m. Each image 555
was dose fractionated to 32 frames with a total electron dose of 60 e- Å
-2 and a total exposure 556
time of 8.0 s. SerialEM7 was used for fully automated data collection. All stacks were motion 557
corrected using MotionCor28 with a binning factor of 1, resulting in a pixel size of 1.014 Å 558
and dose weighting was performed concurrently. The defocus values were estimated using 559
Gctf9. 560
561
Preliminary data processing 562
1,951 micrographs were used for initial data analysis. A small sets of particles for the TIM22 563
complex were auto-picked using reference-free autopicking method with 564
Laplacian-of-Gaussian in RELION3.010
. Good 2D averages were generated after 2D 565
classification and 552,779 particles were auto-picked with selected good 2D averages as 566
template. One round of 2D classification was performed, remaining 305,689 good particles. 567
Initial model was generated with stochastic gradient descent method in RELION10
using 568
about 1000 particles. Global 3D classification was performed with this initial model as 569
reference, generating one good class with 36.8% of input particles and three bad classes. 570
571
Data processing of the TIM22 complex 572
For the complete data set of the TIM22 complex, 2,401,464 particles were auto-picked from 573
14,105 micrographs. A guided multi-reference global classification procedure was applied to 574
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
27
the full dataset. One good and five bad references from preliminary data processing were used 575
as the initial references (Round 1). These six references were low-pass filtered to 40 Å. To 576
avoid the problem of discarding good particles, we simultaneously performed three parallel 577
multi-reference 3D global classifications. After the global classification, particles that belong 578
to the good classes from 24/27/30 iterations of the three parallel runs were separately used as 579
the input for follow-up local classification. Six 3D models from global classification were 580
used as references for the multi-reference local classification. Good particles from these 3X3 581
parallel runs were merged, and the duplicated particles were removed as described 582
previously11
. 1,107,389 particles (46.1% of the original input) remained for following 583
processing. 584
A second round (Round 2) of multi-reference local 3D classification was performed with 585
five references (from global classification, scale to pixel size 2.028 Å) using re-extracted and 586
re-centered 2x binned particles (pixel size: 2.028 Å). Particles from the good classes 587
(representing 45.4/47.8/50.2% of the input particles) were combined to yield 621,049 588
particles (representing 25.9% of the total original particles). 589
A third round (Round 3) local classification was performed using unbinned particles (pixel 590
size: 1.014 Å), focusing on the inter-membrane region. The largest class of 482,959 particles 591
(representing 77.8% of the input particles or 20.1% of the total original particles) yielded an 592
average resolution of 3.73 Å after auto-refinement. With application of a soft mask on the 593
inter-membrane region, these particles gave an average resolution of 3.53 Å for this region 594
after continuous auto-refinement. 595
In the 3.73Å overall and 3.53Å intermembrane region maps of the TIM22 complex, the 596
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
28
local resolution reaches 3.2-3.5 Å in the core regions. The angular distributions of the 597
particles used for the final reconstruction of the TIM22 complexes are reasonable, and the 598
refinement of the atomic coordinates did not suffer from severe over-fitting. The resulting EM 599
density maps display distinguishing features for the amino acid side chains in most regions. 600
Reported resolutions were calculated on the basis of the FSC 0.143 criterion, and the FSC 601
curves were corrected for the effects of a soft mask on the FSC curve using high-resolution 602
noise substitution12
. Prior to visualization, all density maps were corrected for the modulation 603
transfer function (MTF) of the detector, and then sharpened by applying a negative B-factor 604
that was estimated using automated procedures13
. Local resolution variations were estimated 605
using RELION10
. 606
607
Model building and refinement 608
We combined homology modeling and de-novo model building to generate the atomic 609
models. 610
Identification and docking of the two chaperones Tim9/10a and Tim9/10a/10b were 611
facilitated by the crystal structure of the human mitochondrial Tim9/10a hexametric complex 612
(PDB code: 2BSK). Tim10b was manually identified by comparing the sequence variance of 613
these three proteins and the cryo-EM density, and then manually build by COOT14
. 614
Identification of AGK protein was facilitated by the homology structure of a diacylglycerol 615
kinase DgkB from Staphylococcus aureus (PDB code: 2QV7). The atomic model of AGK 616
were generated by CHAINSAW15
and the backbone was manually adjusted using COOT14
. 617
After that, automated model rebuilding was performed with RosettaCM using the adjusted 618
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
29
model as the template and the experimental cryo-EM density as a guide16-18
. Then the 619
hydrogen atoms of the generated model were removed and model building was further 620
performed manually using COOT14
, The single trans-membrane helix and a membrane 621
anchoring motif was de-novo built using COOT14
. 622
The intermembrane space (IMS) domain of Tim29 was located under the Tim9/10a/10b 623
hexamer. The local resolution of IMS domain in cryo-EM map reaches around 3.2 Å, 624
allowing us de-novo building of these regions. The transmembrane helix and a C-terminal 625
fragment which recruits the Tim9/10a chaperone was also de-novo built by COOT14
. A 626
N-terminal helix of Tim29 was identified to be located in the matrix, adhering to the 627
membrane and stabilizing the four transmembrane helices Tim22. Due to limited resolution 628
(5~7 Å), only poly-ALA helix was built for the N-helix. 629
Tim22 is supposed to have four transmembrane helices19
. The cryo-EM density of the 630
membrane region is relatively lower than the soluble region, ranging from 5 Å to 3.2 Å. 631
Luckily, the local resolution of the loop between TM2 and TM3 is high enough, around 3.5 Å, 632
which allow us identify three bulky residues (Y146, R147, and W152) in the linker. The 633
density of the loop connecting TM3 and TM4 can also be clear visualized, which contacts 634
with N-helix of Tim29 in the mitochondrial matrix. A potential S-S bond between TM1 and 635
TM2 can be clearly visualized, which help us assign the sequence. A N-terminal plug bound 636
to Tim9/10a/10b hexamer was also manually built by COOT14
. 637
The final model of the TIM22 complex was refined against the overall cryo-EM maps 638
using PHENIX20
in real space with secondary structure restraints. Overfitting of the overall 639
model was monitored by refining the model in one of the two independent maps from the 640
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
30
gold-standard refinement approach, and testing the refined model against the other map21
. The 641
structures of the TIM22 complex were validated through examination of the Molprobity 642
scores and statistics of the Ramachandran plots. Molprobity scores were calculated as 643
described22
. 644
645
References 646
647
1 Lu, P. et al. Three-dimensional structure of human gamma-secretase. Nature 512, 648
166-170 (2014). 649
2 Clayton, D. A. & Shadel, G. S. Isolation of Mitochondria from Tissue Culture Cells. 650
Cold Spring Harbor Protocols 2014, 1109-1111 (2014). 651
3 Sims, N. R. & Anderson, M. F. Isolation of mitochondria from rat brain using Percoll 652
density gradient centrifugation. Nat Protoc 3, 1228-1239 (2008). 653
4 Rehling, P. et al. Protein insertion into the mitochondrial inner membrane by a 654
twin-pore translocase. Science 299, 1747-1751 (2003). 655
5 Chari, A. et al. ProteoPlex: stability optimization of macromolecular complexes by 656
sparse-matrix screening of chemical space. Nat Methods 12, 859-865 (2015). 657
6 Wittig, I., Braun, H. P. & Schagger, H. Blue native PAGE. Nat Protoc 1, 418-428 658
(2006). 659
7 Mastronarde, D. N. Automated electron microscope tomography using robust 660
prediction of specimen movements. J Struct Biol 152, 36-51 (2005). 661
8 Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for 662
improved cryo-electron microscopy. Nat Methods 14, 331-332 (2017). 663
9 Zhang, K. Gctf: Real-time CTF determination and correction. J Struct Biol 193, 1-12 664
(2016). 665
10 Kimanius, D., Forsberg, B. O., Scheres, S. H. & Lindahl, E. Accelerated cryo-EM 666
structure determination with parallelisation using GPUs in RELION-2. Elife 5, 667
e18722 (2016). 668
11 Yan, C., Wan, R., Bai, R., Huang, G. & Shi, Y. Structure of a yeast activated 669
spliceosome at 3.5 A resolution. Science 353, 904-911 (2016). 670
12 Chen, S. et al. High-resolution noise substitution to measure overfitting and validate 671
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
31
resolution in 3D structure determination by single particle electron cryomicroscopy. 672
Ultramicroscopy 135, 24-35 (2013). 673
13 Rosenthal, P. B. & Henderson, R. Optimal determination of particle orientation, 674
absolute hand, and contrast loss in single-particle electron cryomicroscopy. J Mol 675
Biol 333, 721-745 (2003). 676
14 Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta 677
Crystallogr D Biol Crystallogr 60, 2126-2132 (2004). 678
15 Stein, N. CHAINSAW: a program for mutating pdb files used as templates in 679
molecular replacement. J Appl Crystallogr 41, 641-643 (2008). 680
16 DiMaio, F., Tyka, M. D., Baker, M. L., Chiu, W. & Baker, D. Refinement of protein 681
structures into low-resolution density maps using rosetta. J Mol Biol 392, 181-190 682
(2009). 683
17 Song, Y. et al. High-resolution comparative modeling with RosettaCM. Structure 21, 684
1735-1742 (2013). 685
18 DiMaio, F. et al. Atomic-accuracy models from 4.5-A cryo-electron microscopy data 686
with density-guided iterative local refinement. Nat Methods 12, 361-365 (2015). 687
19 Wrobel, L., Sokol, A. M., Chojnacka, M. & Chacinska, A. The presence of disulfide 688
bonds reveals an evolutionarily conserved mechanism involved in mitochondrial 689
protein translocase assembly. Sci Rep 6, 27484 (2016). 690
20 Adams, P. D. et al. PHENIX: a comprehensive Python-based system for 691
macromolecular structure solution. Acta Crystallogr D Biol Crystallogr 66, 213-221 692
(2010). 693
21 Amunts, A. et al. Structure of the Yeast Mitochondrial Large Ribosomal Subunit. 694
Science 343, 1485-1489 (2014). 695
22 Davis, I. W. et al. MolProbity: all-atom contacts and structure validation for proteins 696
and nucleic acids. Nucleic Acids Res 35, W375-383 (2007). 697
698
699
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
32
ACKNOWLEDGMENTS 700
We thank Dr. Chang at Center of Cryo Electron Microscopy, Zhejiang University, for 701
assistance during data collection; and research associates at the Center for Protein Research 702
and Public Laboratory of Electron Microscopy, Huazhong Agricultural University, for 703
technical support. We thank the Tsinghua University Branch of China National Center for 704
Protein Sciences (Beijing) for providing the technical support on the Cryo-EM and 705
High-Performance Computation platforms. We thank Meng Han and protein chemistry 706
Facility at the Center for biomedical Analysis of Tsinghua University for sample mass-spec 707
analysis. We thank Ms. Wang Yan for model figure design. This work was supported by funds 708
from the Ministry of Science and Technology of China (2018YFA0507700), the National 709
Natural Science Foundation of China (31722017), the Fok Ying-Tong Education Foundation 710
(151021), and the Fundamental Research Funds for the Central Universities (2662017PY031). 711
This research was supported by Beijing Advanced Innovation Center for Structural Biology 712
(to Dr. Chuangye Yan). 713
AUTHOR CONTRIBUTIONS 714
P.Y. conceived the project. L.Q., Q.W., P.Y. designed all experiments. L.Q., Q.W., and Y.W. 715
performed the experiments. Q.W., Z.G., J.C., and X. Z collected the EM data. Z.G. and C. Y. 716
determined the structure. All authors analyzed the data and contributed to manuscript 717
preparation. P.Y wrote the manuscript. 718
719
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
33
720
Extended Data Fig. 1 | Blue native PAGE analysis of the TIM22 complex. Purified TIM22 721
complex sample was analyzed by Blue native PAGE. The complex bands were visualized by 722
Coomassie blue staining (left panel) and western blot (right panel). 723
724
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
34
725
726
Extended Data Fig. 2 | Flowchart for cryo-EM data processing of the human TIM22 727
complex. On the basis of the FSC value of 0.143, the final reconstruction has an average 728
resolution of 3.73 Å for the overall map, 3.53 Å for the inter-membrane region. Details are 729
presented in Methods. 730
731
732
733
734
735
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
35
736
737
Extended Data Fig. 3 | Cryo-EM analysis of the human TIM22 complex. a, The average 738
resolutions for the TIM22 complex are estimated to be 3.73 Å and 3.53 Å for the overall and 739
inter-membrane region, respectively. The resolutions are reported on the basis of the FSC 740
criterion of 0.143. b, Angular distribution of the particles used for reconstruction of the 741
TIM22 complex. Each cylinder represents one view and the height of the cylinder is 742
proportional to the number of particles for that view. c, The local resolutions are color-coded 743
for the TIM22 complex (left panel) and the inter-membrane region (right upper panel). The 744
highest resolution of the EM maps reaches 3.2 Å. d, The FSC curves of the final refined 745
models of the TIM22 complex versus the overall maps that it is refined against (black); of the 746
model refined in the first of the two independent maps used for the gold-standard FSC versus 747
that same map (red/purple); and of the model refined in the first of the two independent maps 748
versus the second independent map (green/blue). The generally similar appearances between 749
the purple and blue, red and green curves indicate that the refinement of the atomic 750
coordinates did not suffer from severe over-fitting. 751
752
753
754
755
756
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
36
757
758
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
37
Extended Data Fig. 4 | The electron microscopy maps for the subunits of the TIM22 759
complex. a-c, The electron microscopy maps for the Tim9/10a/10b hexamer: Overview of 760
chaperone Tim9/10a/10b , composed of Tim9, Tim10a and Tim10b in a ratio of 2:3:1 (a); A 761
close up view for representative Tim9, Tim10a and Tim10b (b); Three segments 762
(helix-loop-helix) of Tim9, Tim10a and Tim10b (c). d, The electron microscopy maps for 763
chaperone Tim9/10a. e, The 4 transmembrane segments (TM1-4), intermembrane space 764
segments (24-50) and loop (142-154) of Tim22. f, AGK -1 (the predicted TM), 225-238 (the 765
membrane-anchored region), 73-102, 113-139, 162-185, and 403-415 of AGK. g, The 766
transmembrane segments (58-74) and intermembrane space segments (79-117, 131-167, 767
168-190, 201-215) of Tim29. 768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
38
788
789
Extended Data Fig. 5 | Topology diagrams of the TIM22 complex. a, Secondary elements 790
are labeled. Tim29, yellow; Tim22, orange; AGK, chartreuse; Tim9/10a, slate; Tim9/10a/10b, 791
blue; Tim10b, magenta. b, The topology diagram of AGK. 1, light pink; anchor, orange; 792
domain 1 (2~7, 2 and ~) and domain 2 (8~10 and ~793
794
795
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
39
796
797
798
799
800
801
802
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
40
Extended Data Fig. 6 | The N-terminal helix and disulfide bond in Tim22. a, Cartoon and 803
surface model of Tim22 and Tim9/10a/10b. Plug indicates the first -helix of Tim22. b, 804
Disulfide bond between C69 and C141 of Tim22. c, The sequences (Uniprot code: Homo 805
sapiens, Q9Y584; Mus musculus, Q9CQ85; Xenopus laevis, Q5U4U5; Drosophila 806
melanogaster, Q8IN78; Saccharomyces cerevisiae, Q12328; Neurospora crassa, Q9C1E8) 807
are aligned using ClustalW. Residues involved in the interaction with Tim9 and Tim10a are 808
highlighted with cyan and slate balls. Cysteines forming disulfide bond are highlighted using 809
yellow boxes. The secondary structural elements are indicated above the sequences. 810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
41
833
834
Extended Data Fig. 7 | A phospholipid in the TIM22 complex and a new putative 835
chaperone recruiting interface of Tim29. a, A phospholipid, binding at the interface 836
between Tim29 and inner membrane. The aliphatic tails of the latter phospholipid may 837
interact with several hydrophobic residues (V125, L127, L129, W156 and F159). The 838
phosphate group likely to interact with several hydrophilic residues (Y134, and R162). b, The 839
electron microscopy maps for the phospholipid resembling to the structure of 840
Phosphatidylethanolamine (PDB code: PTY). c, d, Electrostatic surface of TIM29. A top view 841
(from IMS to matrix) of TIM22 complex is shown as colored surface or electrostatic surface 842
potential. The interface 1 is framed out using a slate circle. A putative interface for recruiting 843
another chaperone is framed out using a yellow dashed circle. Tim9/10a is omitted for clarity 844
in (D). AGK chartreuse; Tim22, orange; Tim9/10a/10b, blue; Tim9/10a, slate; Tim29, 845
electrostatic surface potential. 846
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
42
847
848
Extended Data Fig. 8 | Center pore comparison of Tim9/10a and Tim9/10a/10b hexamer. 849
a, Two perpendicular views of Tim9/10a/10b hexamer. b, Structural alignment of Tim9/10 850
and Tim9/10a/10b. Tim9, cyan; Tim10a, slate or peal blue; Tim10b, magenta. c, A structure 851
alignment of chaperone Tim9/10a/10b with free chaperone Tim9/10a (PDB code: 2BSK). 852
Tim9, cyan; Tim10a, blue; Tim10b, magenta; free chaperone Tim9/10a, wheat. d, A structure 853
alignment of chaperone Tim9/10a/10b with free chaperone Tim9/10a (PDB code: 2BSK) are 854
shown as colored surface: inner helices of Tim9/10a/10b, yellow; outer helices of 855
Tim9/10a/10b and free Tim9/10a, slate; Tim10b, magenta. 856
857
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
43
858
859
Extended Data Fig. 9 | Implication on carrier precursor insert into inner membrane. 860
Cartoon model of carrier precursor insertion into the membrane through The TIM22 complex. 861
The carrier precursor (red line, left panel) wraps around the Tim9/10a chaperone and then 862
inserts into the membrane (red stick, middle and right panel) through the cavity of AGK by a 863
“bending-in” mechanism. 864
865
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
44
Extended data Table 1 | Statistics of Cryo-EM data collection and refinement of the 866
human TIM22 complex. 867
868
Data collection
EM equipment FEI Titan Krios
Voltage (kV) 300
Detector Gatan K2
Pixel size (Å) 1.014
Electron dose (e-/Å2) 60
Defocus range (m) -1.8 ~ -2.5
Reconstruction
Software RELION 3.0
Number of used Particles 482,959
Accuracy of rotation (˚) 1.446 (1.386)
Accuracy of translation (Å) 0.852 (0.785)
Final Resolution (Å) 3.73 (3.53)
Model building
Software COOT/ROSETTA
Refinement
Software PHENIX
Map sharpening B-factor (Å2) -162.8
Average Fourier shell correlation 0.827
R-factor 0.335
Model composition
Protein residues 1599
Lipid 1
Validation
R.m.s deviations
Bonds length (Å) 0.012
Bonds Angle (˚) 0.908
Ramachandran plot statistics (%)
Preferred 91.22
Allowed 8.45
Outlier 0.33
Molprobity score 14.08
869
870
871
872
873
874
875
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint
45
Extended data Table 2 | Summary of model building for the human TIM22 complex. 876
877
Length Domain Copies PDB code Modeling1 Resolution (Å) Chain ID
Tim22 194 24-194 1 De novo 3.5-6.0 A
AGK 422 23-415 1 2QV7 HM 3.5-4.2 B
Tim29 260 21-244 1 De novo 3.2-5.2 C
Tim9
89 5-82 5 2BSK RD 3.2-4.0 D-F, K, L
Tim10a 90 2-80 6 2BSK RD 3.2-3.8 G-I, M-O
Tim10b 103 3-85 1 2BSK HM 3.2-3.8 J
878 1Abbrivations in the “Modeling” column are as follows. HM, homology modeling; RD, rigid 879
docking and refinement; De novo, de novo model building. 880
881
.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted December 9, 2019. ; https://doi.org/10.1101/869289doi: bioRxiv preprint