Rubisco activase remodels plant Rubisco via the large subunit N-terminus

The photosynthetic CO2 fixing enzyme ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) forms inhibited complexes with multiple sugar phosphates, including its substrate ribulose 1,5-bisphosphate. At least three classes of ATPases associated with diverse cellular activities (AAA+ proteins) termed Rubisco activases (Rcas) have evolved to remodel inhibited Rubisco complexes. The mechanism of green-type Rca found in higher plants has proved elusive, because until recently higher plant Rubiscos could not be expressed recombinantly. Towards identifying interaction sites between Rubisco and Rca, here we produce and characterize a suite of 33 Arabidopsis Rubisco mutants for their ability to be activated by Rca. We find that Rca activity is highly sensitive to truncations and mutations in the conserved N-terminus of the Rubisco large subunit. Both T5A and T7A substitutions cannot be activated by Rca, but present with increased carboxylation velocities. Our results are consistent with a model where Rca functions by transiently threading the Rubisco large subunit N-terminus through the axial pore of the AAA+ hexamer.

bisphosphate. At least three classes of ATPases associated with diverse cellular activities (AAA+ 23 proteins) termed Rubisco activases (Rcas) have evolved to remodel inhibited Rubisco complexes. The 24 mechanism of green-type Rca found in higher plants has proved elusive, because until recently higher 25 plant Rubiscos could not be expressed recombinantly. Towards identifying interaction sites between 26 Rubisco and Rca, here we produce and characterize a suite of 33 Arabidopsis Rubisco mutants for 27 their ability to be activated by Rca. We find that Rca activity is highly sensitive to truncations and 28 mutations in the conserved N-terminus of the Rubisco large subunit. Both T5A and T7A substitutions 29 cannot be activated by Rca, but present with increased carboxylation velocities. Our results are 30 consistent with a model where Rca functions by transiently threading the Rubisco large subunit N-31 terminus through the axial pore of the AAA+ hexamer. 32 33 34

INTRODUCTION 35
Virtually all carbon dioxide that enters the biological world does so via the Calvin Benson Bassham 36 cycle (1). Aerobic autotrophic organisms such as plants, algae and cyanobacteria all utilize this 37 somewhat suboptimal CO2-fixation process, which is dependent on catalysis by the slow and 38 promiscuous enzyme Rubisco. The enzyme binds the five carbon sugar Ribulose 1,5-bisphosphate 39 (RuBP), adds a carbon dioxide molecule and hydrolyzes the six carbon intermediate to form two 40 molecules of 3-phosphoglycerate (3PGA). In plants each active site only processes ~1-3 reactions per 41 second, and frequently oxygen gas is incorporated instead of CO2, which leads to production of the 42 toxic metabolite 2-phosphoglycolate. 43 To overcome these flux limitations, Rubisco is overexpressed to constitute up to 50% of the leaf 44 soluble protein and is believed to be the most abundant protein on earth (2,3). Recognition that the 45 enzyme catalyzes the rate limiting step have made its performance the heartpiece of multiple ongoing 46 crop improvement strategies (4,5). In addition to its slow speed and inaccuracy, the enzyme is also 47 susceptible to form dead-end inhibited complexes with several sugar phosphates that are present in its 48 environment (6,7). CO2 fixation ceases, unless the inhibitors are constantly removed. This action is 49 performed by a group of dedicated molecular chaperones that have been termed the Rubisco activases 50 (Rcas) (8-10). Three classes of Rca exist, and although all belong to the superfamily of AAA+ 51 proteins, their primary sequences and mechanisms are highly distinct, indicating convergent 52 evolution(11,12). Red-type Rca found in red-lineage phytoplankton and proteobacteria transiently 53 threads the C-terminus of the Rubisco large subunit through the axial pore of the AAA+ hexamer (13-54 15). In contrast the CbbQO-type Rca found in chemoautotrophic proteobacteria consists of a cup-55 shaped AAA+ hexamer (CbbQ6) bound to a single adaptor protein CbbO, which is essential for 56 Rubisco activation (9). During Rca function the hexamer remodels CbbO, which is bound to inhibited 57

Rubisco via a von Willebrand Factor A domain(16). 58
The detailed molecular mechanism by which inhibitory compounds are removed by higher plant Rca-59 mediated modelling of Rubisco's active site has long remained elusive (11,17). As functional Rca 60 could be produced recombinantly, a large volume of biochemical information has accumulated on Rca 61

A surface scan of higher plant Rubisco for Rca-interacting residues 76
We used the recently established E. coli plant Rubisco expression platform (25,26) to produce a series 77 of Arabidopsis Rubisco large subunits variants mutated in surface-localized residues in an effort to 78 discover additional regions important for protein-protein interactions. We first tested the βC-βD loop 79 mutations E94K and P89A as positive controls (Fig. 1A), as these substitutions had earlier been shown 80 to greatly perturb the ability of Spinach Rca to activate Chlamydomonas Rubisco (27). We then 81 assayed the fully activated holoenzyme (ECM) and the inhibited apo enzyme bound to the substrate 82 ribulose 1,5-bisphosphate (ER) in the presence and absence of the short (Rcaβ) isoform of 83 Arabidopsis Rca (Fig. 1B). Consistent with the Chlamydomonas-spinach result, the inhibited E94K 84 variant of Arabidopsis Rubisco remained non-functional in the presence of its cognate Rca, 85 reconfirming the importance of the N-terminal βC-βD loop in the interaction. The P89A variant, 86 however, was activated well in this system (Fig. 1C), suggesting that the βC-βD loop -Rca 87 interaction is less sensitive to mutation when using Arabidopsis proteins. 88 We next targeted a range of surface-localized Rubisco large subunit residues for mutagenesis ( Fig.  89 2A, Fig. S2). As we have found earlier, multiple positively charged residues on the face of the Rca 90 disc are important for its ability to activate Rubisco (21,28), and therefore the chosen mutations were 91 biased towards probing negatively charged surface residues. This included those located in a 92 negatively charged pocket at the dimer-dimer interface that has recently been implicated in the 93 binding of carboxysomal Rubisco linker proteins in prokaryotic green-type Rubiscos (Fig. 2B) (29,30). 94 We successfully purified 17 variants (Fig. S1), which were all able to carboxylate RuBP similarly to 95 wild-type (Fig. 2C, Fig. S3). Rca assays indicated that the different variants could still be activated 96 effectively indicating the chosen residues were not of critical importance to the Rubisco-Rca 97 interaction (Fig. 2C, Fig. S3). Only K14A showed a statistically significant 52% increase in its Rca-98 mediated activation rate, possibly reflecting a reduced stability of the inhibited complex. Clearly the 99 chosen single amino-acid substitutions were insufficient to disrupt the extensive protein-protein 100 interaction interface involved in Rubisco activation. However, attempts to produce combinations of 101 mutations were unsuccessful either due to insolubility or non-functionality for all tested cases. 102

103
The RbcL N-terminus is essential for Rca function 104 The red-type Rubisco activase CbbX transiently threads the RbcL C-terminus (13-15). However, the 105 C-terminus of green-type Rubisco large subunits is poorly conserved and is of variable length (31), 106 indicating a distinct mechanism for green-type Rca function. In contrast, while sequences at the N-107 terminus of red-type Rubisco large subunits differ between species, both length and sequence of the 108 N-terminus of higher-plant RbcL is essentially completely conserved (Fig. 3A). In available crystal 109 structures, residues 8-20 of the N-terminus are ordered only when the active site is in the closed 110 (ligand-bound) form (Fig. 3B). In the closed conformation, the N-terminus is positioned directly 111 above the 60's loop that co-ordinates P1 of the substrate, with F13, K14, G16 and K18 forming 112 interactions with multiple residues of the 60's loop (32). Coupled with evidence that residues 9-15 of 113 Rubisco from wheat are essential for functional carboxylation activity (33), the stringent conservation 114 of the first 8 residues suggested thus a tantalizing target for mutational analysis. 115 A Rubisco variant with the first seven amino-acids replaced by methionine (ΔN7) displayed 83 % of 116 wild-type carboxylation velocity (Fig. 3C). However, when the ER complex was formed, Rca was 117 unable to reactivate ΔN7 (Fig. 3C). This result was consistent with the notion that a higher plant Rca 118 hexamer engages the disordered N-terminus via its axial pore loops, followed by transient threading 119 leading to active site disruption and inhibitor release. 120

A dissection of the RbcL N-terminal binding motif 121
We then performed a detailed mutational analysis of the RbcL N-terminus, generating a series of 122 variants that, in the ECM form were all able to carboxylate at least as well as wild-type (Fig. 4). 123 Shortening the N-terminus by one or two amino acids (ΔN1, ΔN2) did not negatively affect Rca 124 function. In contrast, replacing the first four amino acids with methionine (ΔN3) or deleting residues 125 5-7 (ΔTET) almost completely eliminated the ability of Rca to activate Rubisco ( Fig 3A). 126 Lengthening the N-terminus by inserting two alanine residues upstream of residue 2 (M1insAA) 127 greatly reduced Rca function by ~64 %. Changing the register of the N-terminal sequence by inserting 128 a AAA sequence upstream of K8 (T7insAAA) in the wild-type or ΔTET variant (TET-AAA) also 129 eliminated function. These results indicated that Rca function was highly sensitive to both length and 130 identity of the N-terminus. 131 Next, we evaluated the effect of single amino acid substitutions in the N-terminal motif. Whereas E6A 132 and K8A substitutions were well tolerated, both T5A and T7A resulted in ~70% reductions in Rca 133 functionality. This findings indicates that the two threonine residues are likely to play an important 134 role in the threading process, possibly via specific interactions with residues in Rca's pore-loop 1 and 135 2 (20,21). We also further note that the observed 2 amino acid step interval would be consistent with 136 successive zipper-like interactions that have been described to be utilized for substrate engagement at 137 the central pore by multiple other AAA+ proteins (34-38). 138 Y20 forms a hydrogen bond to E60, a key catalytic residue that interacts with K334, which is 139 positioned at the apex of Loop 6, and thought to orient the CO2 molecule for gas addition (39). We 140 hypothesized that this interaction could act to disrupt the active site, when the N-terminus is displaced 141 by Rca-threading. The Y20F Rubisco mutant had a ~73% reduced carboxylation rate, but the 142 activation rate by Rca was increased by 39% (Fig. 4B). This result suggests that the Y20-E60 143 interaction is important for the integrity of the active site, and its loss facilitates disruption of the 144 inhibited complex. 145 Quite unexpectedly, we found that multiple N-terminal variants and E94K presented with 146 significantly enhanced carboxylation velocities (up to 53%) compared to the wild-type enzyme under 147 the conditions used in our spectrophotometric Rubisco assay (Fig. 4A,B). This suggests that reducing 148 the interactions of the N-terminus with the enzyme may result in faster Rubiscos. Interestingly, the 149 fast cyanobacterial Form IB enzyme from Synechococcus PCC6301 (40) possesses a truncated N-150 terminus (equivalent to ΔN6) (Fig. 3A). In a follow-up study, it will be important to use 14 C-CO2 151 fixation assays to accurately quantify the carboxylation kinetics (41) and the CO2/O2 specificity factor 152 of the fast N-terminal variants. 153

DISCUSSION 155
The availability of E. coli produced recombinant higher plant Rubiscos permitted us to rapidly 156 produce many variants and assay their capability to be engaged and activated by their cognate Rca 157 chaperone. Mutational analysis of the holoenzyme surface indicated that Rca compatibility was not 158 easily disrupted, with the tested variants remaining functional (Fig. 2). Arguably, inclusion of less 159 conservative substitutions such as charge switches, could have been more informative here. 160 In contrast, mutagenesis of the highly conserved N-terminus resulted in multiple variants that were 161 able to carboxylate RuBP, but could not be activated by Rca. The best described conserved 162 mechanism of numerous AAA+ ATPases concerns the translocation of a substrate peptide through the 163 central pore of the hexamer (42), and this is the modus operandi of the red-type Rca (13). Green-type 164 Rca pore-loops have been shown to be critical for Rubisco activation (20,21). In addition we have 165 long been aware of the RbcL βC-βD loop -Rca specificity helix H9 interaction (27,43,44). 166 Assuming an axial pore loop-RbcL N-terminal threading mechanism we can now further constrain 167 the positioning of an Rca hexameric model (20) in relation to an inhibited Rubisco structure (45). 168 Helix 9 elements of two adjacent Rca subunits can be placed in proximity to two RbcL βC-βD loops 169 that are located on two adjacent dimers (Fig. 5). In this configuration the N-terminal tail (missing 6 170 amino-acids in the structure used) is then accessible to the Rca pore. Transient threading would then 171 result in pulling residues 13 to 20 away from the large subunit body. Disruptions of the associated van 172 der Waal's and polar interactions (32), especially with the RbcL 60's loop, may be sufficient to 173 trigger active site opening and inhibitor release. The model is consistent with an exquisite structural snapshot of a prokaryotic carboxysome-associated 180 green-type Rca hexamer bound to cyanobacterial Rubisco that has been communicated in a concurrent 181 Biorxiv pre-print (46). In agreement with our findings, the study also reports that an N-terminal 9-182 amino acid truncation of the tobacco Rubisco large subunit abolishes tobacco Rca function. Green-183 type Rubisco activation is thus an ancient, conserved process that appears to precede the primary 184 endosymbiotic event (11). 185 186

Molecular biology 188
Plasmids pBAD33k-AtRbcLS, p11a-AtC60αβ/C20 and pCDFduet-AtR1/R2/Rx/B2 enabling 189 the production of Arabidopsis Rubisco in E. coli were a gift from Dr. Manajit Hayer-Hartl (25). To 190 achieve our final construct containing the large and small subunits of Rubisco, a 6x Histidine tag was 191 appended to the C-terminus of rbcS via the Quikchange protocol (Stratagene). Restriction free cloning 192 of the RBS-AtRbcLScHis cassette was utilized to insert the cassette into the multiple cloning site 1 of 193 the pRSFDuet™-1 plasmid (Novagen). To obtain single mutants, the Quikchange protocol was 194 applied to pRSFduet-AtRbcLScHis. Truncations of the N-terminus were performed by PCR 195 amplification of regions flanking the unwanted sequence. Linearized products were then 196 phosphorylated by T4 PNK (NEB) before end to end ligation. All primers used are listed in Table S1  197 and protein-encoding sequences were verified by DNA sequencing. 198 To obtain a vector encoding Arabidopsis thaliana Rca (AtRcaβ), the sequence corresponding 199 to amino acid residues 59 to 474 (Uniprot P10896) were amplified from a cDNA library of 200 Arabidopsis with BamHI and NotI restriction sites at the 5' and 3' end respectively. The sequence was 201 then inserted into the multiple cloning site of the pHue expression vector using the appropriate 202 restriction sites to yield the final construct pHueAthRcaβ. 203

Biochemical assays 222
Rubisco and Rubisco reactivation activities were measured and quantified as described (14), 223 using the spectrophotometric Rubisco assay (47). RuBP was synthesized enzymatically from ribose 5-224 phosphate (48)  Assays were conducted at 25°C using 0.5 µM Rubisco active sites (ER or ECM) and 2 µM wild type 255 activase protomers when appropriate. Error bars indicate the S.D. of at least three independent 256 experiments. The corresponding time course data is shown in Fig. S2. A significant difference from 257 the wild-type value is indicated by an asterisk (One-way ANOVA, posthoc Tukey test, p < 0.05). 258      Detailed views of surface residues substituted in the mutational analysis. Panels on the right depict the interaction network for S10 and Y20. Serine 10 hydrogen bonds to Glycine 67 of the small subunit while the hydroxyl group on the side chain of Tyrosine 20 hydrogen bonds to Glutamate 60. Mutants of these residues were generated to investigate whether these interactions were essential for the mechanism of the activase (Fig. 4B).