Academic Journal
View in EDS
PDF Full Text
HTML Full Text

An Insight of RuBisCO Evolution through a Multilevel Approach

Bibliographic Details
Title: An Insight of RuBisCO Evolution through a Multilevel Approach
Authors: Vladimir Camel, Gaston Zolla
Source: Biomolecules, Vol 11, Iss 12, p 1761 (2021)
Publisher Information: MDPI AG, 2021.
Publication Year: 2021
Collection: LCC:Microbiology
Subject Terms: Bio3D, structural dynamics, structural flexibility, cross-correlation dynamics, Microbiology, QR1-502
More Details: RuBisCO is the most abundant enzyme on earth; it regulates the organic carbon cycle in the biosphere. Studying its structural evolution will help to develop new strategies of genetic improvement in order to increase food production and mitigate CO2 emissions. In the present work, we evaluate how the evolution of sequence and structure among isoforms I, II and III of RuBisCO defines their intrinsic flexibility and residue-residue interactions. To do this, we used a multilevel approach based on phylogenetic inferences, multiple sequence alignment, normal mode analysis, and molecular dynamics. Our results show that the three isoforms exhibit greater fluctuation in the loop between αB and βC, and also present a positive correlation with loop 6, an important region for enzymatic activity because it regulates RuBisCO conformational states. Likewise, an increase in the flexibility of the loop structure between αB and βC, as well as Lys330 (form II) and Lys322 (form III) of loop 6, is important to increase photosynthetic efficiency. Thus, the cross-correlation dynamics analysis showed changes in the direction of movement of the secondary structures in the three isoforms. Finally, key amino acid residues related to the flexibility of the RuBisCO structure were indicated, providing important information for its enzymatic engineering.
Document Type: article
File Description: electronic resource
Language: English
ISSN: 2218-273X
Relation: https://www.mdpi.com/2218-273X/11/12/1761; https://doaj.org/toc/2218-273X
DOI: 10.3390/biom11121761
Access URL: https://doaj.org/article/1ad7b1963aad401f9654ffd8bcb08e10
Accession Number: edsdoj.1ad7b1963aad401f9654ffd8bcb08e10
Database: Directory of Open Access Journals
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHjPtM4BHU3ZchRwgzYmadcigk49r9CVlbU7V5F6lgH7WwHbkykE6L6bDIRm22el6MfsAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDFVrw9-ozve4mRAiigIBEICBmpk2luWOy2c2dyqfBaYLQ-spyPfkX-shY-xMuIuScR70fXGTARWAsp6EW9yvN98jCe0VB2x1Ai8i7xZAhGTScY9dPBRYbUiPB3xx6Oit8paiPvPf_h5o12hc34o0zbst7cOUEw-35Yn1Uxhxi4wTJFS49osQqTBlHg3-EoQBBYZOsqlfi5_f-IWX2LeY44z6lFvE6ae1ZGt7OQE=
Text:
  Availability: 1
  Value: <anid>AN0154345237;[fdu3]01dec.21;2021Dec29.06:31;v2.2.500</anid> <title id="AN0154345237-1">An Insight of RuBisCO Evolution through a Multilevel Approach </title> <p>RuBisCO is the most abundant enzyme on earth; it regulates the organic carbon cycle in the biosphere. Studying its structural evolution will help to develop new strategies of genetic improvement in order to increase food production and mitigate CO<sub>2</sub> emissions. In the present work, we evaluate how the evolution of sequence and structure among isoforms I, II and III of RuBisCO defines their intrinsic flexibility and residue-residue interactions. To do this, we used a multilevel approach based on phylogenetic inferences, multiple sequence alignment, normal mode analysis, and molecular dynamics. Our results show that the three isoforms exhibit greater fluctuation in the loop between αB and βC, and also present a positive correlation with loop 6, an important region for enzymatic activity because it regulates RuBisCO conformational states. Likewise, an increase in the flexibility of the loop structure between αB and βC, as well as Lys330 (form II) and Lys322 (form III) of loop 6, is important to increase photosynthetic efficiency. Thus, the cross-correlation dynamics analysis showed changes in the direction of movement of the secondary structures in the three isoforms. Finally, key amino acid residues related to the flexibility of the RuBisCO structure were indicated, providing important information for its enzymatic engineering.</p> <p>Keywords: Bio3D; structural dynamics; structural flexibility; cross-correlation dynamics</p> <hd id="AN0154345237-2">1. Introduction</hd> <p>RuBisCO (ribulose-1,5-bisphosphate carboxylase oxygenase) is the most abundant enzyme in nature and plays essential functions in the entry of carbon into the biosphere and in photorespiration processes [[<reflink idref="bib1" id="ref1">1</reflink>]]. It is found in most autotrophic organisms such as bacteria, archaea and eukarya (algae, higher plants) [[<reflink idref="bib2" id="ref2">2</reflink>]]. Evolutionary studies in RuBisCO have allowed its classification into four isoforms (I, II, III and IV) [[<reflink idref="bib3" id="ref3">3</reflink>]]. Isoform I is the predominant enzyme in nature and is found in cyanobacteria, green algae and in higher and lower plants. It is a holoenzyme consisting of eight large (RbcL) and eight small (RbcS) subunits [[<reflink idref="bib5" id="ref4">5</reflink>]]. The isoform II enzyme is present in bacteria and is composed only of large-type subunit multimers [(L2)x], and appears to be less efficient in cleaving CO<subs>2</subs> and O<subs>2</subs> [[<reflink idref="bib4" id="ref5">4</reflink>], [<reflink idref="bib6" id="ref6">6</reflink>]]. Isoform II has a distinct physiological role, and it is used primarily to allow the Calvin–Benson–Bassham pathway to balance the cell redox potential [[<reflink idref="bib7" id="ref7">7</reflink>]]. Isoform III is found in archaeas and consists of a toroid-shaped pentagonal decamer composed of L subunits [[<reflink idref="bib9" id="ref8">9</reflink>]]. In addition, the enzyme shows extreme thermostability with high carboxylase activity at high temperatures [[<reflink idref="bib10" id="ref9">10</reflink>]] and exceeds the RuBisCO activity of spinach by 20 times, but it is not efficient at room temperature [[<reflink idref="bib12" id="ref10">12</reflink>]]. Moreover, it is not affected by the presence of oxygen [[<reflink idref="bib9" id="ref11">9</reflink>], [<reflink idref="bib13" id="ref12">13</reflink>]]. Isoform IV includes proteins similar to RuBisCO (RLP) but does not use CO<subs>2</subs> as the main source of carbon [[<reflink idref="bib15" id="ref13">15</reflink>]]. Despite the variability of the amino acid sequences within the different RuBisCO isoforms [[<reflink idref="bib5" id="ref14">5</reflink>], [<reflink idref="bib16" id="ref15">16</reflink>]], the key residues of the active site, catalytic chemistry and activation processes are conserved, and this supports the concept that there is a conserved set of residues that are critical for folding and maintaining the general structure of the enzyme [[<reflink idref="bib15" id="ref16">15</reflink>], [<reflink idref="bib17" id="ref17">17</reflink>]]. However, it is possible that proteins within the same isoform may have different enzymatic and kinetic properties. For example, phylogenetic studies show that the sequence of RbcL from <emph>Arabidopsis thaliana</emph> (5IU0) is different from <emph>Oryza sativa</emph> (1WDD) despite exhibiting high structural similarity. On the other hand, the amino acid sequences of isoform III in <emph>Methanococcoides burtonii</emph> (5MAC) is closer to isoform II. Likewise, the RuBisCO of <emph>Nostoc</emph> sp. (6KKM) and <emph>Synechococcus elongatus</emph> (6SMH) is isoform I, but their sequences are similar to isoform III. Due to these differences, it is necessary to understand the relationship between the structure and function of the RuBisCO enzyme in order to understand the role of the residues directly involved in catalysis. Furthermore, Nishitani et al. [[<reflink idref="bib10" id="ref18">10</reflink>]] showed that mutations (SP5-V330T) in the RbcL 3A12<sups>WT</sups> protein of <emph>Thermococcus kodakarensis</emph> increased the flexibility of the α-helix 6 and loop 6 regions, being important to increase the photosynthetic efficiency of the enzyme at room temperature. Likewise, the closure of the active site implies movements of loop 6 and flexible elements of the N-terminal domain of the adjacent subunit in the dimer [[<reflink idref="bib18" id="ref19">18</reflink>]]. Currently, the two states, open and closed, of the RuBisCO enzyme are quite well-defined structurally, but the details of the closing mechanism are still unknown [[<reflink idref="bib19" id="ref20">19</reflink>]]. Therefore, it is necessary to study the influence between the structure, the amino acid composition and the flexibility of the RuBisCO structures.</p> <p>On the other hand, RuBisCO is a widely studied enzyme. Consequently, the PDB (Protein Data Bank) repository has several RbcL structures [[<reflink idref="bib20" id="ref21">20</reflink>]], which are useful to understand the evolution of the different RuBisCO isoforms. In this sense, the Bio3D [[<reflink idref="bib21" id="ref22">21</reflink>]] and Prody packages emerged as computational tools that help to better understand the relationship between the structure, dynamics and function of sets of evolutionarily related proteins [[<reflink idref="bib23" id="ref23">23</reflink>]].</p> <p>Consequently, in the present work, we evaluate how the evolution of sequence and structure among isoforms I, II and III of RuBisCO defines their intrinsic flexibility and residue-residue interaction.</p> <hd id="AN0154345237-3">2. Materials and Methods</hd> <p></p> <hd id="AN0154345237-4">2.1. Classification of RuBisCO Isoforms</hd> <p>RuBisCO protein codes (1RLC, 4RUB, 4HHH, 1GK8, 4LF1 and 3A12) were used to search for homologous structures using BLAST [[<reflink idref="bib25" id="ref24">25</reflink>]]; then, the different RuBisCO structures determined by crystallography were downloaded from RCSB PDB [[<reflink idref="bib20" id="ref25">20</reflink>]] using the Bio3D package [[<reflink idref="bib21" id="ref26">21</reflink>]]. A sequence identity threshold of 70% was used according to Kalenkiewicz et al. [[<reflink idref="bib26" id="ref27">26</reflink>]] to isolate structures of isoforms I, II and III. In this way 64 crystalline structures of the RbcL subunit of RuBisCO were downloaded, but redundant structures with missing amino acid residues were removed. These criteria allowed us to select 46 unique from wild-type RbcL and mutants in proteobacteria and archaea (Table S1 and Figure S1). The alignment of these amino acid sequences was performed with the MUSCLE algorithm [[<reflink idref="bib27" id="ref28">27</reflink>]]. All conformations were structurally superimposed on each other by least-squares fitting of the Cartesian coordinates of C-α atoms equivalent to the C-terminal domain, since this region was found to be the most structurally invariant. Principal component analysis (PCA) was used to evaluate the relationships between conformer sets of overlapping structures, as it is very useful for evaluating the distributions of experimental structures and comparing them with the conformations obtained through molecular dynamics (MD) simulations (Figure S1) [[<reflink idref="bib21" id="ref29">21</reflink>], [<reflink idref="bib26" id="ref30">26</reflink>]].</p> <hd id="AN0154345237-5">2.2. RuBisCO Structure Selection and Phylogenetic Analysis</hd> <p>Based on the component analysis, RbcL<sups>Wt</sups> structures and mutants from model organisms were selected which had a resolution ≤ 2.7 Å [[<reflink idref="bib28" id="ref31">28</reflink>]], which did not have missing amino acid residues, and for which the crystallized structure was ≥ 95% of the total protein (RbcL) in its three isoforms (I, II, III). Thus, 137 RuBisCO protein sequences, including the sequences provided by Kacar et al. [[<reflink idref="bib7" id="ref32">7</reflink>]] and the selected RuBisCO sequences, were used to build a phylogenetic tree according to Kacar et al. [[<reflink idref="bib7" id="ref33">7</reflink>]] with the PhyloBot web service [[<reflink idref="bib30" id="ref34">30</reflink>]]. RuBisCO orthologs were identified by the NCBI BLAST tool [[<reflink idref="bib25" id="ref35">25</reflink>]]. Then, multiple sequence alignments were inferred by the MSAProbs [[<reflink idref="bib31" id="ref36">31</reflink>]] and MUSCLE [[<reflink idref="bib27" id="ref37">27</reflink>]] algorithms with their default settings. Maximum likelihood (ML) phylogenetic inference was estimated using the PROTCATWAG model [[<reflink idref="bib32" id="ref38">32</reflink>]] in the RAxML web service [[<reflink idref="bib34" id="ref39">34</reflink>]]. Subsequently, ML phylogeny files were exported to the PhyML website [[<reflink idref="bib35" id="ref40">35</reflink>]] in order to calculate statistical support for branches as approximate likelihood ratios and the sequence from the group IV family as the outgroup to root the tree [[<reflink idref="bib7" id="ref41">7</reflink>]]. Finally, the phylogeny plot was developed with Mega6 software [[<reflink idref="bib36" id="ref42">36</reflink>]].</p> <hd id="AN0154345237-6">2.3. Normal Mode Analysis</hd> <p>Normal mode analysis (NMA) is a simple method to predict and characterize the internal dynamics of proteins, where slow low-frequency movements are often of functional importance [[<reflink idref="bib21" id="ref43">21</reflink>]]. NMA analyses were developed in the Bio3D package, where simultaneous analysis of a large set of structures is easily performed through the implementation of ensemble normal mode analysis (eNMA) [[<reflink idref="bib21" id="ref44">21</reflink>]], allowing the rapid characterization and comparison of flexibility across homologous structures. eNMA allows the prediction and identification of different flexibility patterns between different protein isoforms that are available at PDB [[<reflink idref="bib21" id="ref45">21</reflink>]]. In this way, high resolution crystallographic structures of the RbcL subunits of RuBisCO were selected: 6 structures of isoform I (PDB code: 1WDD, 4RUB, 5IU0, 1IWA, 1GK8 and 6FTL), 3 structures of isoform II (PDB code: 4LF1<sups>WT</sups>, 5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) and 3 structures of isoform III (PDB code: 3A12<sups>WT</sups>, 3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups>). As input, the set of pdbs structure aligned with MUSCLE software was provided [[<reflink idref="bib27" id="ref46">27</reflink>]]. Then, an efficient model based on C-alpha was used to enable the modes to be calculated quickly. Aligned eigenvectors and mode fluctuations were obtained as results for all RbcL structures.</p> <hd id="AN0154345237-7">2.4. Molecular Dynamics</hd> <p>The simulation models were built based on the high-resolution crystallographic structures of the RbcL subunits of RuBisCO; 6 structures of the isoform I were selected (PDB code: 1WDD, 4RUB, 5IU0, 1IWA, 1GK8 and 6FTL), as well as 3 structures of isoform II (PDB code: 4LF1<sups>WT</sups>, 5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) and 3 structures of isoform III (PDB code: 3A12<sups>WT</sups>, 3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups>). Before running MD simulations, proteins were treated. First, water molecules and monomers were eliminated. Moreover, missing amino acid residues from all structures were completed with MODELLER software version 10.0 (Accelerys, San Diego, CA, USA) [[<reflink idref="bib37" id="ref47">37</reflink>]]. The selected models satisfied spatial constraints such as bond lengths, bond angles, dihedral angles, and interactions between unbound residues. Models' stereochemical quality was assessed with Ramachandran graphs generated on the MolProbity server [[<reflink idref="bib38" id="ref48">38</reflink>]], and fold quality was determined by Verify3D [[<reflink idref="bib39" id="ref49">39</reflink>]].</p> <p>Molecular dynamics simulation was performed using the Groningen Machine for Chemical Simulations GROMACS version 2020 [[<reflink idref="bib41" id="ref50">41</reflink>]]. The PDB2GMX module was used to generate the topology that had information about the unbound parameters (types of atoms and charges) and bound parameters (bonds, angles and dihedrals) within the simulation. The CHARMM36 force field [[<reflink idref="bib42" id="ref51">42</reflink>]] was used for the simulations of all RuBisCO systems following similar studies [[<reflink idref="bib14" id="ref52">14</reflink>], [<reflink idref="bib43" id="ref53">43</reflink>]]. Periodic boundary conditions (PBC) were applied in all directions of a cube box with a 10 Å lateral size. The systems were solvated with the TIP3P water model [[<reflink idref="bib36" id="ref54">36</reflink>]]. Na<sups>+</sups> ions were added to neutralize the system, as in previous studies [[<reflink idref="bib45" id="ref55">45</reflink>]]. To minimize energy in all systems, the algorithm of descending steps was used with 50,000 steps and with a search for energy less than 1000 kcal/mol. We used the isothermal-isobaric set with two equilibrium phases to simulate a system at cellular physiological conditions. The first equilibrium phase was done in the NVT ensemble at a constant temperature of 300 K with a Berendsen thermostat. The second equilibrium phase was done in the NPT ensemble at a pressure of 1 bar for 2 ns with the Parrinello–Rahman barostat. The simulation was carried out for 50 ns with integration steps of 2 fs under constant pressure and temperature conditions with the leapfrog integration algorithm. The LINCS algorithm was used to constrain all bonds during equilibrium [[<reflink idref="bib47" id="ref56">47</reflink>]], and the Ewald particle mesh algorithm was used for long-range ionic interactions.</p> <p>In the 50-ns MD simulation, 5000 trajectories were obtained. The analysis of the output structures was performed by the following GROMACS commands: gmx_mpi rmsd to calculate root mean square deviation (RMSD) values; gmx_mpi rmsf to calculate root mean square fluctuation (RMSF) values; and gmx_mpi gyrate to calculate the radius of gyration. Finally, PCA and DCCM analyses were carried out with the Bio3D package, following Yu and Dalby's [[<reflink idref="bib48" id="ref57">48</reflink>]] recommendations. Conversion of the trajectory from XTC to DCD format was done with the CatDCD plugins of VMD software [[<reflink idref="bib49" id="ref58">49</reflink>]], and Pymol was used for image editing [[<reflink idref="bib50" id="ref59">50</reflink>]].</p> <hd id="AN0154345237-8">2.5. Stability and Flexibility Analysis</hd> <p>RMSD was used to measure the deviations of the protein backbone from its original structural conformation to its final structural conformation. When the stationary phase of the RMSD curve is reached, the protein is in equilibrium [[<reflink idref="bib51" id="ref60">51</reflink>]]. On the other hand, RMSF was used to measure the average individual residue flexibility during MD simulation. RMSF can indicate structurally which amino acids in a protein are more important in molecular motion [[<reflink idref="bib51" id="ref61">51</reflink>]]. RMSD and RMSF were performed using built-in protocols from GROMACS [[<reflink idref="bib41" id="ref62">41</reflink>]] and Bio3D [[<reflink idref="bib21" id="ref63">21</reflink>]].</p> <hd id="AN0154345237-9">2.6. Principal Component Analysis</hd> <p>Principal component analysis (PCA) was performed using the Bio3D package [[<reflink idref="bib21" id="ref64">21</reflink>]] implemented in the R-Project and ProDy software [[<reflink idref="bib23" id="ref65">23</reflink>]]. The PCA was carried out on Cα atoms during the last 40 ns of the trajectories [[<reflink idref="bib52" id="ref66">52</reflink>]]. The correlated movements of the whole protein can be represented by the eigenvectors and eigenvalues. The eigenvectors, also called principal components (PC), gave the direction of the coordinated movement of the atoms, and the eigenvalues represented the magnitude of the movement along the corresponding eigenvectors [[<reflink idref="bib53" id="ref67">53</reflink>]]. Thus, PC1 and PC2 were computed, because they contributed more significantly to the PCA analysis [[<reflink idref="bib54" id="ref68">54</reflink>]].</p> <p>Briefly, the PCA was based on the diagonalization of the covariance matrix, <emph>C</emph>, with elements <emph>Cij</emph> calculated from the aligned and overlapping Cartesian coordinates, <emph>r</emph>, of equivalent Cα atoms [[<reflink idref="bib21" id="ref69">21</reflink>]]:</p> <p> <ephtml> <math display="block" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>C</mi><mrow><mi>i</mi><mi>j</mi></mrow></msub><mo>=</mo><mo>⟨</mo><msub><mi>r</mi><mi>i</mi></msub><mo>−</mo><mo>⟨</mo><msub><mi>r</mi><mi>i</mi></msub><mo>⟩</mo><mo /><mo>.</mo><mo /><mrow><msub><mi>r</mi><mi>j</mi></msub><mo>−</mo><mo>⟨</mo><msub><mi>r</mi><mi>j</mi></msub><mo>⟩</mo></mrow><mo>⟩</mo></mrow></semantics></math> </ephtml> </p> <p>where <emph>r<subs>i</subs></emph> and <emph>r<subs>j</subs></emph> are cartesian coordinates of the <emph>i</emph>th and <emph>j</emph>th Cα atoms, and ⟨<emph>r<subs>i</subs></emph>⟩ and ⟨<emph>r<subs>j</subs></emph>⟩ represent the average time over all configurations derived from the molecular dynamics simulation. The analysis was limited to Cα atoms because they were less disturbed by statistical noise and offers a meaningful characterization of essential spatial movements [[<reflink idref="bib55" id="ref70">55</reflink>]].</p> <hd id="AN0154345237-10">2.7. Dynamic Cross-Correlation Matrices (DCCM)</hd> <p>To have a better understanding of the dynamics of the three RuBisCO isoforms, cross-correlation analysis (DCCM) was used to evaluate the motions (shifts) of alpha (Cα) carbon atoms in the <emph>MD</emph> simulations of all systems [[<reflink idref="bib56" id="ref71">56</reflink>]]. Additionally, it provides useful information regarding the mutation effect on protein dynamics by analyzing how atomic shifts were correlated [[<reflink idref="bib57" id="ref72">57</reflink>]], and it was constructed using the Bio3D package from R-Project [[<reflink idref="bib21" id="ref73">21</reflink>]].</p> <p>The DCCM map is a 3D matrix annotation that displays time-related information for protein residues. Time-dependent data based on residuals can be analyzed using visual pattern recognition. The DCCM map shows the correlations of amino acid movements, and was calculated according to Ichiye and Karplus' [[<reflink idref="bib59" id="ref74">59</reflink>]] equation:</p> <p> <ephtml> <math display="block" xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>C</mi><mrow><mi>i</mi><mi>j</mi></mrow></msub><mo>=</mo><mfrac><mrow><mrow><mo>Δ</mo><msub><mi>r</mi><mi>i</mi></msub><mo /><mi>x</mi><mo /><mo>Δ</mo><msub><mi>r</mi><mi>j</mi></msub></mrow></mrow><mrow><msup><mrow><mrow><mo>⟨</mo><mo>Δ</mo><msub><mi>r</mi><mi>i</mi></msub><msup><mrow /><mn>2</mn></msup><mo>⟩</mo><mo>⟨</mo><mo>Δ</mo><msub><mi>r</mi><mi>j</mi></msub><msup><mrow /><mn>2</mn></msup><mo>⟩</mo></mrow></mrow><mrow><mn>1</mn><mo>/</mo><mn>2</mn></mrow></msup></mrow></mfrac></mrow></semantics></math> </ephtml> </p> <p>where <emph>Δr<subs>i</subs></emph> and <emph>Δr<subs>j</subs></emph> are the displacements from the mean position of the <emph>i</emph>-th and <emph>j</emph>-th atoms with respect to time. The angle brackets "⟨⟩" represent the average time over the entire trajectory. <emph>C<subs>ij</subs></emph> values ranged from −1 to +1; a positive value represented a positively correlated movement between residues <emph>i</emph> and <emph>j</emph>, while a negative value implied a negatively correlated movement between residues <emph>i</emph> and <emph>j</emph> [[<reflink idref="bib60" id="ref75">60</reflink>]].</p> <hd id="AN0154345237-11">3. Results</hd> <p></p> <hd id="AN0154345237-12">3.1. RuBisCO Forms Classification</hd> <p>From the RCSB protein database, 64 crystal structures of the RbcL subunit of RuBisCO were downloaded. A total of 18 structures were not considered due to a lack of coordinates, leaving 46 RuBisCO complexes between wild-types and mutants (Figure 1). Next, principal component analysis (PCA) was carried out. 73% of the total variance of the atomic fluctuations was captured along the first principal component (PC), while the second and third dimensions were necessary to capture 83.2 and 88.2, respectively (Figure 1).</p> <p>The PCA structure shows three conformational clusters of RuBisCO (Figure 1). The largest cluster (in green) corresponds to 32 proteins (1BXN, 5OYA, 6FTL, 5MZ2, 5NV3, 1IWA, 1BWV, 5WSK, 2V6A, 1UW9, 2V68, 2VDH, 2VDI, 1GK8, 2V69, 4HHIH, 4RUB0, 4MKV, 1IR1, 1WDD, 3ZXW, 1RSC, 1RBL, 7JFO, 6URA, 1SVD, 1RLC, 1RLD, 1EJ7, 6SMH and 6KKM) and involves most RuBisCO structures from higher plants, green algae, blue-green algae, cyanobacteria, diatoms, and proteobacteria, including the 6URA structure of the bacteria <emph>Promineofilum breve</emph>, which is a benchmark to understand the evolution of RuBisCO Form I (Table S1). The second cluster (in red) includes 10 proteins (5MAC, 4LF1, 5HQM, 5KOZ, 5HQL, 5HJY, 5HAT, 5HAN, 5HAO and 5HJX), and they are mainly proteobacteria that present RuBisCO Form II, with the exception of 5MAC, which is found in <emph>Methanococcoides burtonii</emph> (archaea) and represents form II/III (Figure 1). Finally, in the third cluster (in blue), there are 4 proteins (3KDO, 3A13, 3WQP and 3A12), which corresponded to archaea (Figure 1).</p> <p>Based on PCA analysis, RbcL wild-type structures and mutants from model organisms (Figure 1) were selected with a resolution ≤2.7 Å, with crystallized structure ≥95% and without any lost amino acid residues. This criterium allowed 6 wild-type structures to be selected from the largest cluster (in green: 1WDD, 5IU0, 4RUB, 1GK8, 6FTL and 1IWA). From the second cluster (in red) 1 wild-type structure was selected (4LF1<sups>WT</sups>) as well as 2 mutants (5HJX<sups>A47V</sups> and 5HAN<sups>S59F</sups>), and from the third cluster (in blue) 1 wild-type (3A12<sups>WT</sups>) and 2 mutants (3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups>) were selected.</p> <p>To classify the 12 selected structures according to their evolutionary groups, a maximum likelihood (ML) phylogenetic tree was made with 137 amino acid sequences of RuBisCO RbcL. The phylogeny shows that the different RuBisCO isoforms (I, II, III and IV) share a common evolutionary ancestor. The RbcL sequences of isoform I have subgroups IA, IB, and IC/D, which includes higher plants, cyanobacteria, green algae, red algae, and some bacteria. Consequently, the RuBisCO structures of the <emph>Arabidopsis thaliana</emph> (5IU0), <emph>Nicotiana tabacum</emph> (4RUB), <emph>Oriza sativa</emph> (1WDD) and <emph>Chlamydomonas reinhardtii</emph> (1GK8) species corresponded to IB subgroup (Figure 2), while the <emph>Galdieria partita</emph> (1IWA) and <emph>Skeletonema marinoi</emph> (6FTL) are related to the IC/D subgroup (Figure 2). The RbcL sequences of isoform II include proteobacteria and some eukaryotic alveolates; also within this clade are <emph>Rhodopseudomonas palustris</emph>, including members of this species which are wild-types (4LF1<sups>WT</sups>) and mutants (5HJX<sups>A47V</sups> and 5HAN<sups>S59F</sups>) (Figure 2). Other wild-type (3A12<sups>WT</sups>) and mutant (3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups>) structures belong to <emph>Thermococcus kodakarensis</emph>; its lineage arches and its RuBisCO structure are form III (Figure 2).</p> <p>According to multiple sequence alignment analysis, the ends of the N-terminal and C-terminal regions showed greater variation in amino acid sequences. In Figure 3, the conserved active site residues are shown with an asterisk in the alignment. The secondary structures that maintain the amino acid residues involved in the catalysis were: αB (E72), α0 (N144), a loop that connects β1 and α1 (K202; K204), a catalytic motif located between α1 and α2 (G223; D225; F226; K228; D230; E231), β5 (H324) and loop 6 (K366) (Figure 3). Likewise, the secondary structures that are conserved and involved in the union of the phosphate groups in C1 and C5 of RuBP were: a loop between αB and βC (T77), β5 (R325), β6 (H358), β6, a loop that connects β7 and α7 (S411; G413) and a loop connecting β8 and α8 (G436; G437) (Figure 3).</p> <p>The catalytic loop 6 in Figure 3 is characterized as a conserved and flexible sequence because it interacts with the tail in the C-terminal region to close on the catalytic pocket when it binds to RuBP. This tail then opens to allow product release. On the other hand, the CD loop is located in the N-terminal domain and approaches the opening of the active site from the opposite direction to loop 6; furthermore, it is packed against loop 6. The observed differences among RbcL sequences and between species is reflected in the molecular complexity of RbcL isoforms (Figure 3).</p> <hd id="AN0154345237-13">3.2. Stability and Flexibility Evaluation of RuBisCO Forms</hd> <p>Protein function depends on its structure and dynamics and can be altered by mutations. Consequently, it is necessary to understand the intrinsic structural flexibility of the observed differences in multiple sequence alignment. Thus, the flexibility of the RuBisCO forms was evaluated by normal mode analysis (NMA) and molecular dynamics (MD). In Figure 4, the eNMA shows the consensus fluctuations are highlighted and reveal a conserved pattern among species and RuBisCO forms (Figure 4a,b). The three isoforms show a greater fluctuation in the N-terminal domain spanning the amino acid residues (51–68) between the secondary elements αB and βC (Figure 4a), which are functionally relevant for RbcL. Likewise, RuBisCO form III presents greater fluctuation (≥3 Å) with respect to form I and II (Figure 4a,b). Conversely, the catalytic domain of the α/β barrel subunit (150–444) was more stable in all structures evaluated.</p> <p>To evaluate the conformational changes of RuBisCO isoforms, we performed MD analysis for 50 ns over time. The mean square deviation (RMSD) was used to evaluate the conformational stability of the protein during the simulations. The mean square fluctuation (RMSF) was useful to identify rigidity and flexibility among RuBisCO forms. An RMSF value greater than 0.3 nm was considered as high fluctuation [[<reflink idref="bib62" id="ref76">62</reflink>]]. In Figure 5a, RSMD showed great variability in form I, where 4RUB (~0.36 nm ± 0.00268), 1GK8 (~0.34 nm ± 0.0025), 6FTL (~0.37 nm ± 0.0027) and 1IWA (~0.36 nm ± 0.0032) presented the lowest values (Table 1). Contrary to this, 1WDD (~0.52 nm ± 0.004) and 5IU0 (~0.46 nm ± 0.0033) showed very high RMSD values.</p> <p>The RMSF of form I showed greater fluctuations in the N-terminal and C-terminal tails because the first and last amino acid residues form a loop-shaped structure. Likewise, the six systems showed greater flexibility in the loop located between αB and βC (≥0.3 nm) of the N-terminal domain spanning the following amino acid residues: 1WDD (Thr68−Ser76), 4RUB (Gly64−Thr75), 5IU0 (Trp66−Thr75), 1IWA (Ala73−Ala86), 1GK8 (Val69−Thr75), and 6FTL (Ser71−Thr80); see Table 1. On the other hand, the active site of RuBisCO (α/β barrel) was stable (RMSF ≤ 0.3 nm) because it is located in the TIM barrel domain that allows a protein to be slightly rigid (Figure 5b).</p> <p>On the other hand, the analysis of form II included the 4LF1<sups>WT</sups> system and two mutants, 5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>. The results showed that 4LF1<sups>WT</sups> (~0.24 nm ± 0.001) and the mutant 5HAN<sups>S59F</sups> (~0.28 nm ± 0.002) had lower RMSD values and greater stability during the trajectories. Moreover, there was an increase in the RMSD value of the 5HJX<sups>A47V</sups> mutant (~0.34 nm ± 0.002), and this can be attributed to the presence of valine 47 (A47V) in the αB region at the N-terminal end (Table 1). On the other hand, the 4LF1<sups>WT</sups> structure and mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) had similar flexibility in most amino acid residues. Among the systems analyzed in form II, the greatest fluctuation was found in the loop between αB and βC (4LF1<sups>WT</sups> (Gly53−Asp63 residues), 5HAN<sups>S59F</sups> (Val56−Thr65 residues) and 5HJX<sups>A47V</sups> (Thr54−Asp63 residues)), and other relevant fluctuations (≥0.3 nm) were: 4LF1<sups>WT</sups> (Val201−Phe202, Pro458−Ala461 residues), 5HAN<sups>S59F</sups> (Pro458−Ala461 residues) and 5HJX<sups>A47V</sups> (Lys330−Met331, Pro458−Ala461 residues). In conclusion, the 5HAN<sups>S59F</sups> mutant analysis allowed key residues that reduced (Gly53−Glu57 residues, Val201−Phe202 residues) and increased (Phe59, Asp63−Phe64 residues) the flexibility by ≥0.1 nm to be identified (Table 1). Changes were found at loop 6 and in the loop between αB and βC, which is a critical region for gaseous substrate binding after RuBP enolization has been completed [[<reflink idref="bib18" id="ref77">18</reflink>], [<reflink idref="bib63" id="ref78">63</reflink>]]. Moreover, the comparison between 4LF1<sups>WT</sups> and 5HJX<sups>A47V</sups> allowed the identification of key residues that increased (Gly35, Lys330−Met331) and reduced (Val201−Phe202) RuBisCO form II fluctuation by more than 0.1 nm (Table 1), which can be attributed to A47V mutation and which has an effect on RuBisCO catalytic activity [[<reflink idref="bib63" id="ref79">63</reflink>]].</p> <p>The RMSD analysis of form III showed that the 3A12<sups>WT</sups> protein (~0.18 nm ± 0.001) was more stable than 3KDO<sups>SP6</sups> (~0.28 nm ± 0.001) and 3WQP<sups>T289D</sups> (~0.22 nm ± 0.001); see Table 1. The 3KDO<sups>SP6</sups> mutant showed regions where RMSF differed markedly from WT (Figure 5e). The residues Trp55-Tyr62 (loop that connects αB and βC) and Asn347 (loop that connects α6 and β7) showed, on average, a higher RMSF than WT (Table 1). However, the region that exhibited a lower RMSF in the 3KDO<sups>SP6</sups> mutant with respect to WT was the αF region (Ala286); see Table 1. Moreover, the analysis of the 3WQP<sups>T289D</sups> mutant showed a fluctuation greater than 0.1 nm in residues 60–61 (loop connecting αB and βC) and 322 (loop 6); see Table 1. Likewise, amino acid 322 is involved in direct interaction with the ligand CAP 2-carboxyarabinitol-1,5-diphosphate (C<subs>6</subs>H<subs>14</subs>O<subs>13</subs>P<subs>2</subs>). Thus, loop 6 is a region that plays a critical role in improving the enzymatic activity in the 3WQP<sups>T289D</sups> mutant. Finally, our results of NMA and RMSF are in agreement because they were able to identify similar regions of greater flexibility in RuBisCO isoforms, where the loop between αB and βC presented greater flexibility.</p> <hd id="AN0154345237-14">3.3. Principal Component Analysis (PCA)</hd> <p>To obtain information on the conformational states of RbcL form I (5IU0, 1IW1, 1GK8, 1WDD, 4RUB and 6FTL), the PCA of the Cα atoms was carried out. The first two PCs (PC 1/2) were taken in account. Figure 6 indicates the variance in the conformational distributions of proteins, where the display of continuous color points (from blue to white and to red) highlights periodic jumps between structural conformations. Moreover, the PC 1/2 of the MD trajectories were quite varied for the six systems, showing differences in the movement and stability of RuBisCO form I (Figure 6). In 5IU0, 1IWA, 1GK8 and 1WDD systems, there was greater correlated movement along the first two components, with a percentage of 85.5%, 80.4%, 76.7% and 75.6%, respectively, while in the 4RUB and 6FTL systems, the PC values were 70.6% and 70.5%, respectively (Figure 6). On the other hand, the PC 1/2 for 5IU0, 1GK8, 1IWA and 6FTL systems clearly shows the thermodynamically distinct periodic jumps (Figure 6), where most of the blue and red dots were assembled and distributed in opposite regions; therefore, proteins were in a relatively stable state in the system (Figure 6). Thus, the RbcL structures of <emph>Oryza sativa</emph> (1WDD) and <emph>Nicotiana tabacum</emph> (4RUB) showed a uniform distribution, overlapping PC subspace where there were not energy barriers, because most dots were in a scattered state. PCA analysis could suggest that the IB substructure of RbcL may undergo a periodic change in its conformation to reorient its domains (N-terminal and α/β barrel).</p> <p>Regarding RuBisCO form II, PCA analysis allowed information on the conformational states of 4LF1<sups>WT</sups> and two mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) to be obtained. The PC 1/2 of 4LF1<sups>WT</sups>, 5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups> was 62.66%, 50.91% and 61.79%, respectively (Figure 7a–c). The scatter distribution of red and blue dots represents two different stable conformational states of the protein. The 4LF1<sups>WT</sups> system was revealed to be more stable than the mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>). Finally, the scatter plot of the 5HJX<sups>A47V</sups> mutant (Figure 7c) showed the most unstable state of <emph>R. palustris</emph>. This is in agreement with RMSD results (Figure 5c), where 5HJX<sups>A47V</sups> demonstrated more flexibility (~0.1 nm) than WT. Thus, the more dispersed conformational state was produced by repressor mutations (S59F and A47V).</p> <p>Figure 7 shows PCA analysis of <emph>T. kodakarensis</emph> with 1 wild-type (3A12<sups>WT</sups>) and 2 mutants (3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups>). The first two eigenvectors captured most of the variance. The PCs (PC1/2) of the three systems contributed 54.42%, 64.11% and 56.96%, respectively (Figure 7d–f). Likewise, the analysis of the variance in the conformational distribution of 3KDO<sups>SP6</sups> and 3WQP<sups>T289D</sups> shows that mutants were energetically more stable than the WT system (Figure 7d–f). This analysis suggests that the WT may undergo a periodic change in its conformation to reorient its N-terminal or C-terminal domain. Differentiated grouping can be energy expensive; however, it can provide a control mechanism in the photosynthetic activity of RuBisCO.</p> <hd id="AN0154345237-15">3.4. Dynamic Cross-Correlation Matrix (DCCM)</hd> <p>One structural transition that is essential for the carboxylation of the 2,3-ene-diol(ate) intermediate is the closure of the active site of loop 6 in the large subunit and the concomitant movement of loop connecting αB and βC at the N-terminal end to stabilize the catalytic of loop 6 conformation [[<reflink idref="bib18" id="ref80">18</reflink>], [<reflink idref="bib63" id="ref81">63</reflink>]]. Therefore, DCCM was performed to probe the conformational ensemble of these zones. Thus, similar correlation patterns were observed in 1WDD, 4RUB, 5IU0, 1GK8 and 6FTL (Figure 8a–d). This may provide insights into a conserved mechanism among <emph>Chlamydomonas</emph>, <emph>Skeletonema</emph>, <emph>Arabidopsis</emph>, tobacco and rice, since there was a correlated movement of residues 60–80 (part of αB region and the loop connecting αB and βC) at the N-terminal end with secondary structures as α4, β5, αF, βF, α5, β6 and loop 6, which are located between residues 270–345 in the C-terminal end (Figure 8a–d). This correlation is important in RuBisCO since it connects the region with the greatest flexibility (loop connecting αB and βC) with the key substrate-binding residues H294, R295, H327 and K334. Moreover, the βC, loop CD, βD and αC structures had a strong negative correlation with α4, β5, αF, βF, α5, β6 and loop 6 structures. This suggests highly synchronized movements of the RuBisCO structure.</p> <p>On the other hand, <emph>Galdieria partita</emph> (1IWA) had a different correlation pattern than other isoform I proteins (Figure 8e). Thus, 1IWA presented a strong negative correlation (≥−0.5) at residues 25–220 (secondary structures α-2, βB, αB, βC, loop CD, βD, α-1, αC, α0, βE, αD and αE) with respect to amino acid residues located at position 350–493 (α6, β7, α7, β8, α8, αG and αH) in the blue dotted line rectangle. However, a positive correlation was observed between the loop that connects αB and βC with the structures β4, α4, β5, αF, βF, α5, β6 and loop 6 (black rectangle); see Figure 8e.</p> <p>The DCCM plots for RuBisCO Form II showed that residues 53–63 exhibited anticorrelated movement with the structures β4, α4, β5, αF, βF and α5 (Ala255–Gly317). Moreover, the most flexible region (residues 53–63) located at the loop connecting αB and βC moves in the same direction (positive correlation > 0.5) as β6, loop 6 and α6 structures that are located between residues Gly326-Ala341 at the C-terminal domain (Figure 8a–c). This is because the movement of the N-terminal domain (the loop that connects αB and βC) towards the active site is important, since it is a key step in the catalytic mechanism of RuBisCO that involves CO<subs>2</subs> addition. Furthermore, the comparison between 5HAN<sups>S59F</sups> (Figure 9b) and LF1<sups>WT</sups> (Figure 9a) showed minor anticorrelations (blue dashed line box). Therefore, S59F mutation is the most flexible region with a direct effect on loop 6 residues. Regarding the 5HJX<sups>A47V</sups> mutant (Figure 9c), no significant changes were found in relation to 4LF1<sups>WT</sups> (Figure 9a).</p> <p>The DCCM analysis of the loop connecting αB and βC in RuBisCO form III demonstrated an anticorrelated movement (<−0.5) of Pro60-Ala72 residues with β6 and loop 6 structures. However, Ser50-Tyr59 residues presented a positive correlation (>0.5) with respect to β6 and loop 6 structures (Figure 9). On the other hand, correlated movements were significantly reduced in 3WQP<sups>T289D</sups> with respect to 3A12<sups>WT</sups> (blue dotted line rectangle). This region comprises a positive correlation between αE, β1 and loop structures that connect β1 and α1 with the β6 and loop 6 structures. However, the positive correlation movements of these regions decreased in the 3WQP<sups>T289D</sups> mutant (Figure 9f), since loop 6 (Lys322) mutation affects the correlation between specific residues, making them more flexible. Finally, the 3KDO<sups>SP6</sups> mutant (Figure 9e) did not show visible changes in the residue-residue correlation patterns when compared to 3A12<sups>WT</sups> (Figure 9d).</p> <hd id="AN0154345237-16">4. Discussion</hd> <p>RuBisCO plays a key role in carbon fixation on Earth. This enzyme possibly evolved during the Archean Eon [[<reflink idref="bib65" id="ref82">65</reflink>]] from an ancestral non-carbon fixing enzyme long before the appearance of the Calvin–Benson–Bassham cycle [[<reflink idref="bib3" id="ref83">3</reflink>], [<reflink idref="bib66" id="ref84">66</reflink>]]. Thus, our phylogenetic analysis (Figure 2) allows the different RuBisCO isoforms to be identified, supporting the idea that photosynthetic RuBisCO (form I, II and III) and RubisCO-like protein (RLP, Form IV) evolved from the same ancestral protein [[<reflink idref="bib67" id="ref85">67</reflink>]]. This would have allowed photosynthetic organisms to adopt different strategies to improve CO<subs>2</subs> specificity (as in the <emph>Galdieria partita</emph> case) [[<reflink idref="bib69" id="ref86">69</reflink>]], increase intracellular CO<subs>2</subs> concentration through a mechanism of carbon concentration [[<reflink idref="bib70" id="ref87">70</reflink>]], or inhabit ecological niches that have low levels of O<subs>2</subs>/CO<subs>2</subs>, such as in methanogenic organisms (as in the <emph>Methanococcoides burtonii</emph> case) [[<reflink idref="bib66" id="ref88">66</reflink>]].</p> <p>The RbcL subunit is common among all isoforms (Figure 1). Currently, there are more than 64 RbcL structures in the PDB database (Figure 1, Table S1). Our PCA results using the Bio3D package [[<reflink idref="bib21" id="ref89">21</reflink>]] to identify three clusters with different structural flexibilities (Figure 1). This also allowed the identification of the key amino acid residues, several structural characteristics, and conformational changes that are critical for folding and catalytic activity (Figure 3). The largest cluster corresponded to Isoform I, which included higher plants, green algae, blue-green algae, cyanobacteria, diatoms, and proteobacteria (Figure 1). The emergence of form I complexes through the incorporation of small subunits represents a transitional key that is little understood in RuBisCO evolution [[<reflink idref="bib19" id="ref90">19</reflink>]]. However, the 6URA structure of <emph>Promineofilum breve</emph> (a bacteria) has a structural flexibility that allowed grouping of isoform I (Figure 1), taking into account that 6URA does not present small RbcS subunits [[<reflink idref="bib71" id="ref91">71</reflink>]] and also presents deletions in secondary structural elements such as loop-CD and the loop that connects α8-αG, making it a reference point to advance our understanding of isoform I evolution. In the second cluster (Figure 1), proteobacteria of RuBisCO Form II were mostly reported, except for the methanogenic archaea <emph>Methanococcoides burtonii</emph> (5MAC), which presents an isoform II/III [[<reflink idref="bib5" id="ref92">5</reflink>], [<reflink idref="bib65" id="ref93">65</reflink>]]. It also shows a unique insert of 26–30 amino acids between the α6 and β7 secondary structures at the bottom of the βα-barrel [[<reflink idref="bib72" id="ref94">72</reflink>]]. Methanogens like <emph>Methanococcoides burtonii</emph> (5MAC) are strictly anaerobic and cannot survive in the presence of oxygen [[<reflink idref="bib73" id="ref95">73</reflink>]]. Thus, their RuBisCO are not under selection pressure to mitigate the competitive binding of O<subs>2</subs> over CO<subs>2</subs>.</p> <p>Despite having only 30% amino acid identity, the multiple sequence alignment analysis among RuBisCO isoforms showed large changes in the N-terminal and C-terminal regions (Figure 3). Moreover, the structures retain the residues involved in substrate binding and catalytic activity (Figure 3), supporting the idea that they are critical to folding and maintaining the overall structure and function of the photosynthetic RuBisCO [[<reflink idref="bib1" id="ref96">1</reflink>], [<reflink idref="bib66" id="ref97">66</reflink>]]. Moreover, with the molecular dynamics analysis, it was possible to sample the transition of the RuBisCO conformation in 50 ns, and the RMSD results were consistent with previous studies [[<reflink idref="bib14" id="ref98">14</reflink>], [<reflink idref="bib45" id="ref99">45</reflink>]]. Our results added more theoretical evidence on the structural movements of RuBisCO, helping to understand the structure flexibility and how this could affect the synchronization of the residues and the closing mechanism, which are still unknown [[<reflink idref="bib19" id="ref100">19</reflink>]]. Thus, the NMA and RMSF results revealed similar flexibility patterns in the RbcL structures (Figure 4 and Figure 5), showing the distribution of temperature factors (B-factors) and the fact that loops which are more flexible during catalysis are also more flexible in crystallized structures [[<reflink idref="bib74" id="ref101">74</reflink>], [<reflink idref="bib76" id="ref102">76</reflink>]]. Consequently, our results indicate that secondary structures such as the loop connecting αB and βC (~64–85) and the tails in the N-terminal and C-terminal region show greater fluctuations in the three isoforms (Figure 5), and the structural movements are related to the structural changes and activities of RuBisCO during the transition from its open to closed state [[<reflink idref="bib74" id="ref103">74</reflink>], [<reflink idref="bib77" id="ref104">77</reflink>]]. Thus, closed-state carboxylation of RuBisCO is more likely when the substrate is attached, whereas fluctuations in larger tails can also cause structural changes and deactivations of RuBisCO.</p> <p>According to DCCM analysis of isoform I, the structure of 1WDD, 4RUB, 5IU0, 1GK8 and 6FTL preserved the direction of the movements in the time of residues ~64–85 (part of the αB region and the connecting loop between αB and βC) and ~270–345 residues (related to the secondary structures α4, β5, αF, βF, α5, β6 and loop 6); see Figure 8. This correlation is important for RuBisCO activity, since it connects the region of greatest flexibility (connecting loop between αB and βC) with the key residues (H294, R295, H327 and K334) that bind the substrate (Figure 4), where the two states (open and closed) of RuBisCO are distinguished by the degree of accessibility of the solvent to the active site [[<reflink idref="bib74" id="ref105">74</reflink>]]. The closed state is associated with substrates and inhibitors (CABP 2-carboxyarabinitol 1,5-bisphosphate) that are attached to the active site. This would be achieved through a movement of loop 6 (residues 331–338) and the connecting loop between αB and βC (residues ~64–85); see Table 1. In addition, the N-terminal and C-terminal loops function like latches that hold loop ~64–85 and loop 6 in their closed positions with an extremely slow release of CABP [[<reflink idref="bib74" id="ref106">74</reflink>]]. Moreover, RuBisCO studies carried out on spinach and wheat were able to show that residues ~8 to 20 at the N-terminal end are only ordered when the active site is closed [[<reflink idref="bib78" id="ref107">78</reflink>]]. In the closed conformation, the N-terminal end (Phe13, Lys14, Gly16 and Lys18) is placed directly on the connecting loop between αB and βC (~64–85 residues), which coordinates the P1 site of the substrate [[<reflink idref="bib78" id="ref108">78</reflink>]]. In contrast, the open state is associated with weak products being attached (metal ions that are catalytically inert). In the open state, loop 6 of the α/β barrel is away from the active site, and the C-terminal end of the large subunit is disordered, so the active site is open.</p> <p>On the other hand, <emph>Galdieria partita</emph> is a thermophilic red alga with a high specificity, and this alga showed a marked difference with respect to the RuBisCO of higher plants [[<reflink idref="bib79" id="ref109">79</reflink>]]. Our results are in agreement with Watanabe et al. [[<reflink idref="bib79" id="ref110">79</reflink>]] because the movements of the RbcL structure of <emph>G. partita</emph> presented a different residue-residue correlation pattern than IB and IC/D structures (Figure 8e). Although <emph>G. partita</emph> has more residues in the N-terminal and C-terminal regions (Figure 3), this phenomenon could play an important role in the structural movement of the RuBisCO enzyme, as indicated by some studies [[<reflink idref="bib78" id="ref111">78</reflink>]]. Likewise, the RMSF analysis of 1IWA (<emph>G. partita</emph>) showed the greatest flexibility between residues ~Trp73–Ala86 (the connecting loop between αB and βC). Moreover, it is necessary to point that Thr74 in the RbcL sequence of <emph>G. partita</emph> is an evolutionarily conserved amino acid that binds the phosphate groups of RuBP. Thus, the Thr74 residue can alter the packing in the C-terminal end [[<reflink idref="bib79" id="ref112">79</reflink>]]. Likewise, when RuBisCO is in the open state, the hydrogen bond breaks between P1 and Thr74, as well as between Thr76 and Trp470 [[<reflink idref="bib79" id="ref113">79</reflink>]]. Thr74 is stabilized by the carbonyl of the Thr76 backbone, while the stabilization of the Thr76 and Trp470 side chains occurs in the closed state of loop 6 [[<reflink idref="bib79" id="ref114">79</reflink>]]. Moreover, the work of Satagopan et al. [[<reflink idref="bib63" id="ref115">63</reflink>]] on <emph>R. rubrum</emph> (RuBisCO isoform II) is in agreement with Watanabe et al. [[<reflink idref="bib79" id="ref116">79</reflink>]], because the link between the αB and βC loop with loop 6 was also identified; this link allows the conformation of the catalytic loop to be stabilized. Satagopan et al. [[<reflink idref="bib63" id="ref117">63</reflink>]] used mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) in <emph>R. rubrum</emph> where CO<subs>2</subs> was the only carbon source; their results showed a similar biological growth between mutants and a decrease with respect to WT [[<reflink idref="bib63" id="ref118">63</reflink>], [<reflink idref="bib80" id="ref119">80</reflink>]]. Thus, our PCA results of DM indicated that mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) would have undergone a change from an unstable intermediate conformation (4LF1<sups>WT</sups>) to an unstable one, showing a greater variation in the conformational distribution of the RuBisCO (Figure 7b,c), being consistent with the RMSD results where 5HJX<sups>A47V</sups> was more flexible (~0.1 nm) than WT (Figure 5). Consequently, the fluctuation in the 5HJX<sups>A47V</sups> mutant was increased by more than 0.1 nm in the key residues (Lys330–Met331) that are located in loop 6 (Table 1). Some experimental studies showed that the mutations (M331L and M331A) affected the growth of the purple photosynthetic bacterium <emph>R. palustris</emph>.</p> <p>Thus, the residue Met331 and its interactions seem to be specific and critical for the addition of CO<subs>2</subs> to the intermediate 2,3-enediol(ate) derived from RuBP [[<reflink idref="bib63" id="ref120">63</reflink>], [<reflink idref="bib80" id="ref121">80</reflink>]]. On the contrary, strains with double mutations (A47V/M331A and S59F/M331A) in <emph>R. palustris</emph> did show growth. Thus, growth inhibition induced by M331A was suppressed by the substitution of A47V and S59F [[<reflink idref="bib63" id="ref122">63</reflink>]]. To explain this phenomenon, Satagopan et al. [[<reflink idref="bib80" id="ref123">80</reflink>]] evaluated the movements and crystallized structures of WT and mutants, where the α carbon of Ala47 (structure αB) was ∼15 Å away from Met331 (loop 6) [[<reflink idref="bib80" id="ref124">80</reflink>]]. Furthermore, our results indicate that Ala47 and Met331 move in the same direction (positive correlation > 0.5; Figure 9a–c). Likewise, the side chain is found in a hydrophobic environment within 4 Å of Ala70 and Val72 [[<reflink idref="bib80" id="ref125">80</reflink>]]. Substitution with a Val (A47V) would push the αB structure towards the active site, and thus Lys330 and Met331 showed greater flexibility (Table 1; Figure 5). Glu49 located in the αB structure works to stabilize Lys330 and helps to close loop 6 during catalysis, and Thr54 binds phosphate P1 of the substrate [[<reflink idref="bib78" id="ref126">78</reflink>], [<reflink idref="bib81" id="ref127">81</reflink>]].</p> <p> <emph>Thermococcus kodakarensis</emph> is a hyperthermophilic archaea. Its optimal growth is at 85 °C, and it can exceed the activity of RuBisCO spinach by 20 times [[<reflink idref="bib12" id="ref128">12</reflink>], [<reflink idref="bib83" id="ref129">83</reflink>]], but at room temperature its activity is only one-eighth [[<reflink idref="bib12" id="ref130">12</reflink>]]. For this reason, it is necessary to develop new <emph>T. kodakarensis</emph> strains with good photosynthetic performance at room temperatures. In this sense, 3WQP<sups>T289D</sups> [[<reflink idref="bib12" id="ref131">12</reflink>]] and 3KDO<sups>SP6</sups> [[<reflink idref="bib10" id="ref132">10</reflink>]] mutants were developed. Their results show an increase in the carboxylase activity of 24% and 31%, respectively. In the 3KDO<sups>SP6</sups> mutant, residues of the α6 region were replaced by 11 amino acid residues from spinach (E326-L336 "ERDITLGFVDL"). Consequently, Nishitani et al. [[<reflink idref="bib10" id="ref133">10</reflink>]] evaluated the flexibility between 3A12<sups>WT</sups> and 3KDO<sups>SP6</sups>, showing that there is an increase in temperature factors (B-Factors Å2) in the secondary structure α6. Moreover, our ANM results showed a greater fluctuation in the loop between αB and βC and in the loop that connects α6 and β7 (His341-Ala361) with respect to isoforms I and II. Likewise, the mutants were more energetically stable than WT (Figure 7e,f), suggesting that WT may undergo a periodic change in its conformation to reorient its N-terminal or C-terminal domain, since the differentiated clustering of RuBisCO conformational distributions can be energetically expensive [[<reflink idref="bib84" id="ref134">84</reflink>]]. However, it can also provide a control mechanism in the photosynthetic activity of RuBisCO.</p> <p>In addition, 3KDO<sups>SP6</sups> showed greater flexibility of ~0.15 nm in the loop that connects αB and βC (Figure 5f) and the 3WQP<sups>T289D</sups> mutant in the residue Lys322 (a residue of loop 6 which is catalytically critical) (Figure 5f, Table 1). From this analysis, the changes between αF and βF in 3WQP<sups>T289D</sups> can influence loop 6′s movement (Table 1). Therefore, an increase in the loop flexibility between αB and βC, Lys322 or residues in the vicinity of the catalytic center is important to increase the catalytic activity of RuBisCO from <emph>T. kodakarensis</emph> at room temperature. Finally, our MD simulation results corroborate very well the work of Satagopan et al. [[<reflink idref="bib63" id="ref135">63</reflink>]] and Fujihashi et al. [[<reflink idref="bib12" id="ref136">12</reflink>]]. Likewise, the regions of greater flexibility and active sites exhibit highly correlated or anticorrelated movements between the different isoforms (Figure 8 and Figure 9), building a dynamic correlation network where information signal are transmitted [[<reflink idref="bib48" id="ref137">48</reflink>]]. Therefore, it is necessary to build next-generation computational tools, where a PAN-DM approach will allow us to integrate very complex information on molecular structure, dynamics and evolution.</p> <hd id="AN0154345237-17">5. Conclusions</hd> <p>Our PCA showed a wide range of the conformational space of the RuBisCO crystal structures, allowing the identification of different isoforms. Likewise, phylogenetic analysis supports the idea that RuBisCO evolved from the same ancestral enzyme, conserving the residues involved in substrate binding and catalytic activity. On the other hand, molecular dynamics analyses were able to sample the transition of RuBisCO conformation in 50 ns. Thus, the NMA and RMSF results revealed similar flexibility patterns. The places where the secondary structures loop between αB and βC as well as the tails in the N-terminal and C-terminal regions show greater fluctuations among the three isoforms, and their movements are possibly related to the structural changes and functional activities of the RuBisCO enzyme during the transition from its open to closed state. On the other hand, the DCCM results indicate that there are changes in the movement direction of the secondary structures of the three isoforms. However, movements in the same direction are preserved in loop 6 and the connecting loop between αB and βC. This correlation is important for enzymatic activity and to stabilize the conformation of the catalytic loop. In isoform I, the 1WDD, 4RUB, 5IU0, 1GK8 and 6FTL structures showed a positive correlation between the residue movement of ~64–85 (part of the αB region and the connecting loop between αB and βC) and residues ~270–345 (secondary structures α4, β5, αF, βF, α5, β6 and loop 6). On the other hand, 1IWA (<emph>Galdieria partita</emph>) had a marked difference in the direction of structural movements with respect to the others' isoform I. This could have a key role, and it is required to deepen its study with mutants. On the other hand, the PCA results of <emph>R. rubrum</emph> (RuBisCO form II) indicated that mutants (5HAN<sups>S59F</sups> and 5HJX<sups>A47V</sups>) would have undergone a change from an unstable intermediate conformation (4LF1<sups>WT</sups>) to an unstable one, showing a greater variation in the conformational distribution of RuBisCO. Consequently, the 5HJX<sups>A47V</sups> mutant allowed the fluctuation of the key residues Lys330–Met331 in loop 6 to be increased by more than 0.1 nm. Regarding isoform III, the mutants (3KDO<sups>SP6</sups>, 3WQP<sups>T289D</sups>) of the hyperthermophilic archaea <emph>Thermococcus kodakarensis</emph> were more energetically stable than WT, suggesting that the WT may undergo a periodic change in its conformation to reorient its N-terminal or C-terminal domain, a control mechanism in the photosynthetic activity of RuBisCO. 3KDO<sups>SP6</sups> showed a greater flexibility of ~0.15 nm in the loop between αB and βC and Lys322 in the 3WQP<sups>T289D</sups> mutant (a catalytic residue critical in loop 6) with respect to 3A12<sups>WT</sups>. Thus, an increase in the loop flexibility between αB and βC, Lys322 or residues in the vicinity of the catalytic center is important to increase the photosynthetic efficiency of RuBisCO from <emph>T. kodakarensis</emph> at room temperature. Finally, our results added more evidence regarding the structural movements of RuBisCO, helping to understand the details of the synchronization and the closing mechanism that are still unknown.</p> <hd id="AN0154345237-18">Figures and Table</hd> <p>Graph: Figure 1 Principal component analysis (PCA) of RuBisCO RbcL isoforms. Green represents isoform I, red represents isoform II and isoform III is in blue.</p> <p>Graph: Figure 2 Maximum likelihood phylogenetic analysis of the RuBisCO RbcL protein family. Four forms of RuBisCO were classified along an evolutionary trajectory from the most recent common ancestor. Ancestral sequences were tagged according to their position in the RuBisCO subfamilies. Phylogenetic positions of the selected Wt and mutant proteins were marked with a black rhombus.</p> <p>Graph: Figure 3 Multiple sequence alignment based on the structure of three isoforms of RuBisCO. Isoform I: O. sativa (1WDD), N. tabacum (4RUB), A. thaliana (5IU0), C. reinhardtii (1GK8), G. partita (1IWA) and S. marinoi (6FTL). Isoform II: 1 wild-type (4LF1WT) and 2 mutants (5HJXA47V and 5HANS59F) of R. palustris. Isoform III: 1 wild-type (3A12WT) and 2 mutants (3KDOSP6 and 3WQPT289D) of T. kodakarensis. Amino acid residue numbers are shown at the top of the sequence. Secondary structural elements, such as α-helices (bars) and β-strands (arrows), are shown in the figure. With an asterisk is depicted, it indicates mechanically important active site residues. Red background shading represents identical amino acids, blue shading designates similar amino acids while white shading indicates no similarity.</p> <p>Graph: Figure 4 Normal mode analysis. (a) Consensus fluctuations of RuBisCO forms I, II and III. α-helices are in black and β-strands in gray; (b) Consensus fluctuations among species. In Form I are O. sativa (1WDD), N. tabacum (4RUB), A. thaliana (5IU0), C. reinhardtii (1GK8), G. partita (1IWA) and S. marinoi (6FTL). In form II are R. palustris with 1 wild-type (4LF1WT) and 2 mutants (5HJXA47V and 5HANS59F), and in form III are T. kodakarensis with 1 wild- type (3A12 WT) and 2 mutants (3KDOSP6 and 3WQPT289D); (c) Monomeric structure of RbcL. The monomer is divided into two domains: the N-terminal domain and the C-terminal domain. Loop αB-βC, loop CD, loop 6 and α6 are indicated. The different colors indicate Form I (green), II (red), and III (blue).</p> <p>Graph: Figure 5 RMSD and RMSF profiles of RuBisCO forms I, II and III. (a) 50 ns RMSD of form I; (b) RuBisCO form I RMSF; (c) 50 ns RMSD of form II; (d) RuBisCO form II RMSF; (e) 50 ns RMSD of form III; (f) RuBisCO form III RMSF. RMSD was used to measure the deviations of the protein backbone from its original structural conformation to its final structural conformation. C-alpha atoms were used to calculate RMSF.</p> <p>Graph: Figure 6 Principal component analysis of RuBisCO isoform I. (a) O. sativa (1WDD); (b) N. tabacum (4RUB); (c) A. thaliana (5IU0); (d) G. partita (1IWA); (e) C. reinhardtii (1GK8) and (f) S. marinoi (6FTL).</p> <p>Graph: Figure 7 Principal component analysis of RuBisCO isoforms II (R. palustris) and III (T. kodakarensis). (a) 4LF1WT (R. palustris); (b) 5HANS59F (R. palustris mutant); (c) 5HJXA47V (R. palustris mutant); (d) 3A12WT (T. kodakarensis); (e) 3KDOSP6 (T. kodakarensis mutant); and (f) 3WQPT289D (T. kodakarensis mutant).</p> <p>Graph: Figure 8 Cross-correlation analysis (DCCM) of RuBisCO form I. (a) O. sativa (1WDD); (b) N. tabacum (4RUB); (c) A. thaliana (5IU0); (d) G. partita (1IWA); (e) C. reinhardtii (1GK8) and (f) S. marinoi (6FTL). The color scale ranges from pink (for values ranging from −1 to −0.5) to white (−0.5 to 0.5) and to cyan (0.5 to 1).</p> <p>Graph: Figure 9 Cross-correlation analysis (DCCM) of RuBisCO form II (R. palustris) and III (T. kodakarensis). (a) 4LF1WT (R. palustris); (b) 5HANS59F (R. palustris mutant); (c) 5HJXA47V (R. palustris mutant); (d) 3A12WT (T. kodakarensis); (e) 3KDOSP6 (T. kodakarensis mutant); and (f) 3WQPT289D (T. kodakarensis mutant). The color scale ranges from pink (for values ranging from −1 to −0.5) to white (−0.5 to 0.5) and to cyan (0.5 to 1).</p> <p>Table 1 Average and standard error of RMSD in 12 RbcL structures. Regions were chosen according to their residues with the highest mean RMSF (≥0.3 nm).</p> <p> <ephtml> <table><thead><tr><th align="center" style="border-top:solid thin;border-bottom:solid thin">Form</th><th align="center" style="border-top:solid thin;border-bottom:solid thin">Protein</th><th align="center" style="border-top:solid thin;border-bottom:solid thin">RMSD</th><th align="center" style="border-top:solid thin;border-bottom:solid thin">Region ≥0.3<break />of RMSF</th><th align="center" style="border-top:solid thin;border-bottom:solid thin">Sequence</th><th align="center" style="border-top:solid thin;border-bottom:solid thin">Structures</th></tr></thead><tbody><tr><td rowspan="6" align="center" valign="middle" style="border-bottom:solid thin">I</td><td align="center" valign="middle" style="border-bottom:solid thin">1WDD<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.52 ± 0.004</td><td align="center" valign="middle" style="border-bottom:solid thin">68–76</td><td align="center" valign="middle" style="border-bottom:solid thin">TVWTDGLTS</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">4RUB<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.36 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">64–75; 125;<break />209–211</td><td align="center" valign="middle" style="border-bottom:solid thin">GTWTTVWTDGLT; F; QPF</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; α0; Loop connecting β2 and α2</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">5IU0<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.46 ± 0.003</td><td align="center" valign="middle" style="border-bottom:solid thin">22; 66–75</td><td align="center" valign="middle" style="border-bottom:solid thin">L; WTTVWTDGLT</td><td align="center" valign="middle" style="border-bottom:solid thin">N-terminal tail; <break />Loop connecting αB and βC</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">1IWA<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.44 ± 0.003</td><td align="center" valign="middle" style="border-bottom:solid thin">55–56; 73–86; 482</td><td align="center" valign="middle" style="border-bottom:solid thin">PG; WTVVWTDLLTAA; T</td><td align="center" valign="middle" style="border-bottom:solid thin">βB; Loop connecting αB and βC; C-terminal tail</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">1GK8<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.34 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">69–75</td><td align="center" valign="middle" style="border-bottom:solid thin">VWTDGLT</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">6FTL<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.37 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">71–80; 211–212</td><td align="center" valign="middle" style="border-bottom:solid thin">TVVWTDLLTA; NS</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; <break />Loop connecting β2 and α2</td></tr><tr><td rowspan="3" align="center" valign="middle" style="border-bottom:solid thin">II</td><td align="center" valign="middle" style="border-bottom:solid thin">4LF1<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.24 ± 0.001</td><td align="center" valign="middle" style="border-bottom:solid thin">53–63; 201–202</td><td align="center" valign="middle" style="border-bottom:solid thin">GTNVEVSTTDD; VF</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; <break />Loop connecting β2 and α2</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">5HAN<sup>S59F</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.28 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">56–65</td><td align="center" valign="middle" style="border-bottom:solid thin">VEVFTTDDFT</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">5HJX<sup>A47V</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.34 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">54–63; 330–331</td><td align="center" valign="middle" style="border-bottom:solid thin">TNVEVSTTDD; KM</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; Loop 6</td></tr><tr><td rowspan="3" align="center" valign="middle" style="border-bottom:solid thin">III</td><td align="center" valign="middle" style="border-bottom:solid thin">3A12<sup>WT</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.18 ± 0.001</td><td align="center" valign="middle" style="border-bottom:solid thin">58–59; 286</td><td align="center" valign="middle" style="border-bottom:solid thin">LY; A</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; αF</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">3KDO<sup>SP6</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.28 ± 0.002</td><td align="center" valign="middle" style="border-bottom:solid thin">55–62; 347</td><td align="center" valign="middle" style="border-bottom:solid thin">WTTLYPWY; N</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; <break />Loop connecting α6 and β7</td></tr><tr><td align="center" valign="middle" style="border-bottom:solid thin">3WQP<sup>T289D</sup></td><td align="center" valign="middle" style="border-bottom:solid thin">0.22 ± 0.001</td><td align="center" valign="middle" style="border-bottom:solid thin">57–63; 322</td><td align="center" valign="middle" style="border-bottom:solid thin">TLYPWYE; K</td><td align="center" valign="middle" style="border-bottom:solid thin">Loop connecting αB and βC; Loop 6</td></tr></tbody></table> </ephtml> </p> <hd id="AN0154345237-19">Author Contributions</hd> <p>Conceptualization, V.C.; data curation, V.C.; formal analysis, V.C.; funding acquisition, G.Z.; investigation, V.C. and G.Z.; methodology, V.C.; project administration, V.C. and G.Z.; software, V.C.; supervision, G.Z.; validation, V.C.; visualization, V.C.; writing—original draft, V.C. and G.Z.; writing—review & editing, V.C. and G.Z. All authors have read and agreed to the published version of the manuscript.</p> <hd id="AN0154345237-20">Funding</hd> <p>This research was funded by PROCIENCIA grant numbers 177-2015-FONDECYT and 159-2018-FONDECYT-BM-IADT-AV.</p> <hd id="AN0154345237-21">Institutional Review Board Statement</hd> <p>This study did not involve humans or animals.</p> <hd id="AN0154345237-22">Informed Consent Statement</hd> <p>Not applicable.</p> <hd id="AN0154345237-23">Data Availability Statement</hd> <p>The datasets generated and/or analyzed during the current study, are available on request from the corresponding author.</p> <hd id="AN0154345237-24">Conflicts of Interest</hd> <p>The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.</p> <hd id="AN0154345237-25">Acknowledgments</hd> <p>The authors want to thank: Emilio Paredes Solis for helping to install the Gromacs software, Georcki Ropon-Palacios for sharing his knowledge in computational biophysics and Haoran Yu for sharing codes for molecular dynamics. We also thank two anonymous reviewers for providing constructive and helpful comments on the manuscript.</p> <hd id="AN0154345237-26">Supplementary Materials</hd> <p>The following are available online at https://<ulink href="http://www.mdpi.com/article/10.3390/biom11121761/s1,">www.mdpi.com/article/10.3390/biom11121761/s1,</ulink> Table S1: 46 structures of isoforms I, II and III from wild-type RbcL and mutants in proteobacteria and archaea by X-ray crystallography, Figure S1: Analysis Pipeline of RuBisCO using Bio3d package.</p> <ref id="AN0154345237-27"> <title> Footnotes </title> <blist> <bibl id="bib1" idref="ref1" type="bt">1</bibl> <bibtext> Publisher's Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.</bibtext> </blist> </ref> <ref id="AN0154345237-28"> <title> References </title> <blist> <bibtext> Andersson I., Backlund A. Structure and Function of Rubisco. Plant Phys. Biochem. 2008; 46: 275-291. 10.1016/j.plaphy.2008.01.001</bibtext> </blist> <blist> <bibl id="bib2" idref="ref2" type="bt">2</bibl> <bibtext> Tabita F.R., Hanson T.E., Li H., Satagopan S., Singh J., Chan S. Function, Structure, and Evolution of the RubisCO-Like Proteins and Their RubisCO Homologs. Microbiol. Mol. Biol. Rev. 2007; 71: 576-599. 10.1128/MMBR.00015-07. 18063718</bibtext> </blist> <blist> <bibl id="bib3" idref="ref3" type="bt">3</bibl> <bibtext> Erb T., Zarzycki J. A Short History of RubisCO: The Rise and Fall (?) of Nature's Predominant CO2 Fixing Enzyme. Plant Biotechnol. 2018; 49: 100-107. 10.1016/j.copbio.2017.07.017</bibtext> </blist> <blist> <bibl id="bib4" idref="ref5" type="bt">4</bibl> <bibtext> Stec B. Structural Mechanism of RuBisCO Activation by Carbamylation of the Active Site Lysine. Proc. Natl. Acad. Sci. USA. 2012; 109: 18785-18790. 10.1073/pnas.1210754109. 23112176</bibtext> </blist> <blist> <bibl id="bib5" idref="ref4" type="bt">5</bibl> <bibtext> Liu D., Chettiyan R., Ramya S., Mueller-cajar O. Surveying the Expanding Prokaryotic Rubisco Multiverse. FEMS Microbiol. Lett. 2017: 1-10. 10.1093/femsle/fnx156. 28854711</bibtext> </blist> <blist> <bibl id="bib6" idref="ref6" type="bt">6</bibl> <bibtext> Kitano K., Maeda N., Fukui T., Atomi H., Imanaka T., Miki K. Crystal Structure of a Novel-Type Archaeal Rubisco with Pentagonal Symmetry. Structure. 2001; 9: 473-481. 10.1016/S0969-2126(01)00608-6</bibtext> </blist> <blist> <bibl id="bib7" idref="ref7" type="bt">7</bibl> <bibtext> Kacar B., Hanson-smith V., Adam Z., Boekelheide N. Constraining the Timing of the Great Oxidation Event within the Rubisco Phylogenetic Tree. Geobiology. 2017; No. May: 628-640. 10.1111/gbi.12243</bibtext> </blist> <blist> <bibl id="bib8" type="bt">8</bibl> <bibtext> Mueller-cajar O., Morell M., Whitney S.M. Directed Evolution of Rubisco in Escherichia Coli Reveals a Specificity-Determining Hydrogen Bond in the Form II Enzyme. Biochemistry. 2007; 46: 14067-14074. 10.1021/bi700820a</bibtext> </blist> <blist> <bibl id="bib9" idref="ref8" type="bt">9</bibl> <bibtext> Maeda N., Kitano K., Fukui T., Ezaki S., Atomi H., Miki K., Imanaka T. Ribulose Bisphosphate Carboxylase/Oxygenase from the Hyperthermophilic Archaeon Pyrococcus Kodakaraensis KOD1 Is Composed Solely of Large Subunits and Forms a Pentagonal Structure. J. Mol Biol. 1999; 293: 57-66. 10.1006/jmbi.1999.3145</bibtext> </blist> <blist> <bibtext> Nishitani Y., Yoshida S., Fujihashi M., Kitagawa K., Doi T., Atomi H., Imanaka T., Miki K. Structure-Based Catalytic Optimization of a Type III Rubisco from a Hyperthermophile. J. Biol. Chem. 2010; 285: 39339-39347. 10.1074/jbc.M110.147587</bibtext> </blist> <blist> <bibtext> Yoshida S., Atomi H., Imanaka T. Engineering of a Type III Rubisco from a Hyperthermophilic Archaeon in Order to Enhance Catalytic Performance in Mesophilic Host Cells. Appl. Environ. Microbiol. 2007; 73: 6254-6261. 10.1128/AEM.00044-07</bibtext> </blist> <blist> <bibtext> Fujihashi M., Nishitani Y., Kiriyama T., Aono R., Sato T., Takai T., Tagashira K., Fukuda W., Atomi H., Imanaka T. Mutation Design of a Thermophilic Rubisco Based on Three-Dimensional Structure Enhances Its Activity at Ambient Temperature. Prot. Struct. Funct. Bioinform. 2016; 84: 1339-1346. 10.1002/prot.25080. 27273261</bibtext> </blist> <blist> <bibtext> Yu H., Dalby P.A. Coupled Molecular Dynamics Mediate Long- and Short-Range Epistasis between Mutations That Affect Stability and Aggregation Kinetics. Proc. Natl. Acad. Sci. USA. 2018; 115: E11043-E11052. 10.1073/pnas.1810324115</bibtext> </blist> <blist> <bibtext> Faulkner M., Szabó I., Weetman S.L., Sicard F., Huber R.G., Bond P.J., Rosta E., Liu L.N. Molecular Simulations Unravel the Molecular Principles That Mediate Selective Permeability of Carboxysome Shell Protein. Sci. Rep. 2020; 101. 10.1038/s41598-020-74536-5</bibtext> </blist> <blist> <bibtext> Tabita F.R., Hanson T.E., Satagopan S., Witte B.H., Kreel N.E. Phylogenetic and Evolutionary Relationships of RubisCO and the RubisCO-like Proteins and the Functional Lessons Provided by Diverse Molecular Forms. R. Soc. 2008: 2629-2640. 10.1098/rstb.2008.0023. 18487131</bibtext> </blist> <blist> <bibtext> Tabita F.R., Satagopan S., Hanson T.E., Kreel N.E., Scott S.S. Distinct Form I, II, III, and IV Rubisco Proteins from the Three Kingdoms of Life Provide Clues about Rubisco Evolution and Structure/Function Relationships. J. Exp. Botany. 2008; 59: 1515-1524. 10.1093/jxb/erm361</bibtext> </blist> <blist> <bibtext> Li H., Sawaya M.R., Tabita F.R., Eisenberg D. Crystal Structure of a RuBisCO-like Protein from the Green Sulfur Bacterium Chlorobium Tepidum. Structure. 2005; 13: 779-789. 10.1016/j.str.2005.02.017. 15893668</bibtext> </blist> <blist> <bibtext> Duff A.P., Andrews T.J., Curmi P.M.G. The Transition between the Open and Closed States of Rubisco Is Triggered by the Inter-Phosphate Distance of the Bound Bisphosphate. J. Mol. Biol. 2000; 298: 903-916. 10.1006/jmbi.2000.3724</bibtext> </blist> <blist> <bibtext> Genkov T., Spreitzer R.J. Highly Conserved Small Subunit Residues Influence Rubisco Large Subunit Catalysis. J. Biol. Chem. 2009; 284: 30105-30112. 10.1074/jbc.M109.044081</bibtext> </blist> <blist> <bibtext> Berman H.M., Westbrook J., Feng Z., Gilliland G., Bhat T.N., Weissing H., Shindyalov I., Bourne P. The Protein Data Bank. Nucleic Acids Res. 2000; 28: 235-242. 10.1093/nar/28.1.235</bibtext> </blist> <blist> <bibtext> Skjaerven L., Yao X.Q., Scarabelli G., Grant B.J. Integrating Protein Structural Dynamics and Evolutionary Analysis with Bio3D. BMC Bioinform. 2014; 15: 1-11. 10.1186/s12859-014-0399-6. 25491031</bibtext> </blist> <blist> <bibtext> Grant B.J., Skjærven L., Yao X.Q. The Bio3D Packages for Structural Bioinformatics. Prot. Sci. 2021; 30: 20-30. 10.1002/pro.3923. 32734663</bibtext> </blist> <blist> <bibtext> Bakan A., Meireles L.M., Bahar I. ProDy: Protein Dynamics Inferred from Theory and Experiments. Bioinformatics. 2011; 27: 1575-1577. 10.1093/bioinformatics/btr168. 21471012</bibtext> </blist> <blist> <bibtext> Bakan A., Dutta A., Mao W., Liu Y., Chennubhotla C., Lezon T.R., Bahar I. Structural Bioinformatics Evol and ProDy for Bridging Protein Sequence Evolution and Structural Dynamics. Bioinformatics. 2014; 30: 2681-2683. 10.1093/bioinformatics/btu336. 24849577</bibtext> </blist> <blist> <bibtext> Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J. Basic Local Alignment Search Tool. J. Mol. Biol. 1990; 215: 403-410. 10.1016/S0022-2836(05)80360-2</bibtext> </blist> <blist> <bibtext> Kalenkiewicz A., Grant B.J., Yang C.Y. Enrichment of Druggable Conformations from Apo Protein Structures Using Cosolvent-Accelerated Molecular Dynamics. Biology. 2015; 4: 344-366. 10.3390/biology4020344</bibtext> </blist> <blist> <bibtext> Edgar R.C. MUSCLE: A Multiple Sequence Alignment Method with Reduced Time and Space Complexity. BMC Bioinform. 2004; 5: 1-19. 10.1186/1471-2105-5-113. 15318951</bibtext> </blist> <blist> <bibtext> Blow D.. Outline of Crytallography for Biologists1st ed.; Oxford University Press: New York, NY, USA. 2002</bibtext> </blist> <blist> <bibtext> Yu-Feng H.. Study of Mining Protein Structural Properties and Its Application; National Taiwan University: Taipei, Taiwan. 2007</bibtext> </blist> <blist> <bibtext> Hanson-Smith V., Johnson A. PhyloBot: A Web Portal for Automated Phylogenetics, Ancestral Sequence Reconstruction, and Exploration of Mutational Trajectories. PLoS Comput. Biol. 2016; 12: 1-10. 10.1371/journal.pcbi.1004976. 27472806</bibtext> </blist> <blist> <bibtext> Liu Y., Schmidt B., Maskell D.L. MSAProbs: Multiple Sequence Alignment Based on Pair Hidden Markov Models and Partition Function Posterior Probabilities. Bioinformatics. 2010; 26: 1958-1964. 10.1093/bioinformatics/btq338</bibtext> </blist> <blist> <bibtext> Le S.Q., Gascuel O. An Improved General Amino Acid Replacement Matrix. Mol. Biol. Evolution. 2008; 25: 1307-1320. 10.1093/molbev/msn067</bibtext> </blist> <blist> <bibtext> Lartillot N., Philippe H. A Bayesian Mixture Model for across-Site Heterogeneities in the Amino-Acid Replacement Process. Mol. Biol. Evol. 2004; 21: 1095-1109. 10.1093/molbev/msh112</bibtext> </blist> <blist> <bibtext> Kozlov A.M., Darriba D., Flouri T., Morel B., Stamatakis A. RAxML-NG: A Fast, Scalable and User-Friendly Tool for Maximum Likelihood Phylogenetic Inference. Bioinformatics. 2019; 35: 4453-4455. 10.1093/bioinformatics/btz305. 31070718</bibtext> </blist> <blist> <bibtext> Guindon S., Dufayard J., Lefort V., Anisimova M., Hordijk W., Gascuel O. New Algorithms and Methods to Estimate Maximim-Likelihood Phylogenies Assessing the Performance of PhyML 3.0. Syst. Biol. 2010; 59: 307-321. 10.1093/sysbio/syq010. 20525638</bibtext> </blist> <blist> <bibtext> Tamura K., Stecher G., Peterson D., Filipski A., Kumar S. MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Mol. Biol. Evol. 2013; 30: 2725-2729. 10.1093/molbev/mst197</bibtext> </blist> <blist> <bibtext> Webb B., Sali A. Comparative Protein Structure Modeling Using Modeller. Curr. Protoc. Bioinform. 2016; 54: 1-37. 10.1002/cpbi.3</bibtext> </blist> <blist> <bibtext> Williams C.J., Headd J.J., Moriarty N.W., Prisant M.G., Videau L.L., Deis L.N., Verma V., Keedy D.A., Hintze B.J., Chen V.B. MolProbity: More and Better Reference Data for Improved All-Atom Structure Validation. Prot. Sci. 2018; 27: 293-315. 10.1002/pro.3330. 29067766</bibtext> </blist> <blist> <bibtext> Luthy R., Bowie J., Eisenberg D. Assessment of Protein Models with Three-Dimensional Profiles. Nature. 1992; 359: 83-85. 10.1038/356083a0. 1538787</bibtext> </blist> <blist> <bibtext> Bowie J., Luthy R., Eisenberg D. A Method to Identify Protein Sequences That Fold into a Known Three-Dimensional Stucture. Science. 1991; 253: 164-169. 10.1126/science.1853201</bibtext> </blist> <blist> <bibtext> Abraham M.J., Murtola T., Schulz R., Páll S., Smith J.C., Hess B., Lindah E. Gromacs: High Performance Molecular Simulations through Multi-Level Parallelism from Laptops to Supercomputers. SoftwareX. 2015; 1–2: 19-25. 10.1016/j.softx.2015.06.001</bibtext> </blist> <blist> <bibtext> Huang J., MacKerell A. CHARMM36 All-Atom Additive Protein Force Field: Validation Based on Comparison to NMR Data. J. Comput. Chem. 2013; 30: 2135-2145. 10.1002/jcc.23354</bibtext> </blist> <blist> <bibtext> Ahrari S., Khosravi F., Osouli A., Sakhteman A., Nematollahi A., Ghasemi Y., Savardashtaki A. MARK4 Protein Can Explore the Active-like Conformations in Its Non-Phosphorylated State. Sci. Rep. 2019; 9: 1-14. 10.1038/s41598-019-49337-0. 31506531</bibtext> </blist> <blist> <bibtext> Salmas R.E., Unlu A., Yurtsever M., Noskov S.Y., Durdagi S. In Silico Investigation of PARP-1 Catalytic Domains in Holo and Apo States for the Design of High-Affinity PARP-1 Inhibitors. J. Enzym. Inhib. Med. Chem. 2016; 31: 112-120. 10.3109/14756366.2015.1005011</bibtext> </blist> <blist> <bibtext> Guinot A.D.M. Structural Studies of Different Form I Rubiscos Using Molecular Dynamics Simulations. Doctoral Dissertation; Imperial College London: London, UK. 2016. 10.25560/51422</bibtext> </blist> <blist> <bibtext> Siqueira A.S., Lima A.R.J., Dall'Agnol L.T., de Azevedo J.S.N., da Silva Gonçalves Vianez J.L., Gonçalves E.C. Comparative Modeling and Molecular Dynamics Suggest High Carboxylase Activity of the Cyanobium Sp. CACIAM14 RbcL Protein. J. Mol. Model. 2016; 223. 10.1007/s00894-016-2943-y</bibtext> </blist> <blist> <bibtext> Hess B., Bekker H., Berendsen H.J.C., Fraaije J.G.E.M. LINCS: A Linear Constraint Solver for Molecular Simulations. J. Comput. Chem. 1997; 18: 1463-1472. 10.1002/(SICI)1096-987X(199709)18:12<1463:AID-JCC4>3.0.CO;2-H</bibtext> </blist> <blist> <bibtext> Yu H., Dalby P.A. A Beginner's Guide to Molecular Dynamics Simulations and the Identification of Cross-Correlation Networks for Enzyme Engineering. Methods Enzymol. 2020; 643: 15-49. 10.1016/bs.mie.2020.04.020. 32896280</bibtext> </blist> <blist> <bibtext> Humphrey W., Dalke A., Schulten K. VMD: Visual Molecular Dynamics. J. Mol. Grap. 1996; 14: 33-38. 10.1016/0263-7855(96)00018-5</bibtext> </blist> <blist> <bibtext> Schrodinger L. The PyMOL Molecular Graphics System, Version 1.3r1. 2010Available online: https://<ulink href="http://www.mdpi.com/1422-0067/21/19/7166/htm(accessed">www.mdpi.com/1422-0067/21/19/7166/htm(accessed</ulink> on 23 September 2021)</bibtext> </blist> <blist> <bibtext> Martínez L. Automatic Identification of Mobile and Rigid Substructures in Molecular Dynamics Simulations and Fractional Structural Fluctuation Analysis. PLoS ONE. 2015; 10: 1-10. 10.1371/journal.pone.0119264</bibtext> </blist> <blist> <bibtext> Lesgidou N., Eliopoulos E., Goulielmos G., Vlassi M. Insights on the Alteration of Functionality of a Tyrosine Kinase 2 Variant: A Molecular Dynamics Study. Bioinformatics. 2018; 34: i781-i786. 10.1093/bioinformatics/bty556</bibtext> </blist> <blist> <bibtext> Hong L., Ying M., Zheng C.-J., Jin W.-Y., Liu W.-S., Wang R.-L. Exploring the Effect of D61G Mutation on SHP2 Cause Gain of Function Activity by a Molecular Dynamics Study. J. Biomol. Struct. Dyn. 2018; 36: 3856-3868. 10.1080/07391102.2017.1402709</bibtext> </blist> <blist> <bibtext> Rajapaksha H., Pandithavidana D., Dahanayake J. Demystifying Chronic Kidney Disease of Unknown Etiology (CKDu): Computational Interaction Analysis of Pesticides and Metabolites with Vital Renal Enzymes. Biomolecules. 2021; 11261. 10.3390/biom11020261</bibtext> </blist> <blist> <bibtext> Zalewski M., Kmiecik S., Kolinski M. Molecular Dynamics Scoring of Protein–Peptide Models Derived from Coarse-Grained Docking. Moleculaes. 2021; 263293. 10.3390/molecules26113293</bibtext> </blist> <blist> <bibtext> Bahar I., Atilgan A.R., Erman B. Direct Evaluation of Thermal Fluctuations in Proteins Using a Single-Parameter Harmonic Potential. Fold. Design. 1997; 2: 173-181. 10.1016/S1359-0278(97)00024-2</bibtext> </blist> <blist> <bibtext> Wang R.R., Ma Y., Du S., Li W.Y., Sun Y.Z., Zhou H., Wang R.L. Exploring the Reason for Increased Activity of SHP2 Caused by D61Y Mutation through Molecular Dynamics. Comput. Biol. Chem. 2019; 78: 133-143. 10.1016/j.compbiolchem.2018.10.013</bibtext> </blist> <blist> <bibtext> Li W.Y., Wei H.Y., Sun Y.Z., Zhou H., Ma Y., Wang R.L. Exploring the Effect of E76K Mutation on SHP2 Cause Gain-of-Function Activity by a Molecular Dynamics Study. J. Cell. Biochem. 2018; 119: 9941-9956. 10.1002/jcb.27316</bibtext> </blist> <blist> <bibtext> Ichiye T., Karplus M. Collective Motions in Proteins: A Covariance Analysis of Atomic Fluctuations in Molecular Dynamics and Normal Mode Simulations. Proteins Struct. Funct. Bioinform. 1991; 11: 205-217. 10.1002/prot.340110305</bibtext> </blist> <blist> <bibtext> Liu W.S., Wang R.R., Sun Y.Z., Li W.Y., Li H.L., Liu C.L., Ma Y., Wang R.L. Exploring the Effect of Inhibitor AKB-9778 on VE-PTP by Molecular Docking and Molecular Dynamics Simulation. J. Cell. Biochem. 2019; 120: 17015-17029. 10.1002/jcb.28963. 31125141</bibtext> </blist> <blist> <bibtext> Sun Y.Z., Chen X.B., Wang R.R., Li W.Y., Ma Y. Exploring the Effect of N308D Mutation on Protein Tyrosine Phosphatase-2 Cause Gain-of-Function Activity by a Molecular Dynamics Study. J. Cell. Biochem. 2019; 120: 5949-5961. 10.1002/jcb.27883</bibtext> </blist> <blist> <bibtext> Selvaraj C., Omer A., Singh P., Singh S. Molecular Insights of Protein Contour Recognition with Ligand Pharmacophoric Sites through Combinatorial Library Design and MD Simulation in Validating HTLV-1 PR Inhibitors. Mol. BioSyst. 2015; 11: 178-189. 10.1039/C4MB00486H. 25335799</bibtext> </blist> <blist> <bibtext> Satagopan S., North J.A., Arbing M.A., Varaljay V.A., Haines S.N., Wildenthal J.A., Byerly K.M., Shin A., Tabita F.R. Structural Perturbations of Rhodopseudomonas Palustris Form II RuBisCO Mutant Enzymes That Affect CO2 Fixation. Biochemistry. 2019; 58: 3880-3892. 10.1021/acs.biochem.9b00617</bibtext> </blist> <blist> <bibtext> Ashida H., Saito Y., Kojima C., Kobayashi K., Ogasawara N., Yokota A. A Functional Link between RuBisCO-like Protein of Bacillus and Photosynthetic RuBisCO. Science. 2003; 302: 86-290. 10.1126/science.1086997. 14551435</bibtext> </blist> <blist> <bibtext> Iñiguez C., Capó-Bauçà S., Niinemets Ü., Stoll H., Aguiló-Nicolau P., Galmés J. Evolutionary Trends in RuBisCO Kinetics and Their Co-Evolution with CO2 Concentrating Mechanisms. Plant J. 2020; 101: 897-918. 10.1111/tpj.14643</bibtext> </blist> <blist> <bibtext> Poudel S., Pike D.H., Raanan H., Mancini J.A., Nanda V., Rickaby R.E.M., Falkowski P.G. Biophysical Analysis of the Structural Evolution of Substrate Specificity in RuBisCO. Proc. Natl. Acad. Sci. USA. 2020; 117: 30451-30457. 10.1073/pnas.2018939117. 33199597</bibtext> </blist> <blist> <bibtext> Saito Y., Ashida H., Sakiyama T., de Marsac N.T., Danchin A., Sekowska A., Yokota A. Structural and Functional Similarities between a Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase (RuBisCO)-like Protein from Bacillus Subtilis and Photosynthetic RuBisCO. J. Biol. Chem. 2009; 284: 13256-13264. 10.1074/jbc.M807095200. 19279009</bibtext> </blist> <blist> <bibtext> Ashida H., Saito Y., Nakano T., Tandeau De Marsac N., Sekowska A., Danchin A., Yokota A. RuBisCO-like Proteins as the Enolase Enzyme in the Methionine Salvage Pathway: Functional and Evolutionary Relationships between RuBisCO-like Proteins and Photosynthetic RuBisCO. J. Exp. Botany. 2008; 59: 1543-1554. 10.1093/jxb/ern104</bibtext> </blist> <blist> <bibtext> Sugawara H., Yamamoto H., Shibata N., Inoue T., Okada S., Miyake C., Yokota A., Yasushi K. Crystal Structure of Carboxylase Reaction-Oriented Ribulose 1,5- Bisphosphate Carboxylase/Oxygenase from a Thermophilic Red Alga, Galdieria Partita. J. Biol. Chem. 1999; 274: 15655-15661. 10.1074/jbc.274.22.15655</bibtext> </blist> <blist> <bibtext> Wang Y., Stessman D., Spalding M. The CO2 Concentrating Mechanism and Photosynthetic Carbon Assimilation in Limiting CO2: How Chlamydomonas Works against the Gradient. Plant J. 2015; 82: 429-448. 10.1111/tpj.12829</bibtext> </blist> <blist> <bibtext> Banda D.M., Pereira J.H., Liu A.K., Orr D.J., Hammel M., He C., Parry M.A.J., Carmo-Silva E., Adams P.D., Banfield J.F. Novel Bacterial Clade Reveals Origin of Form I Rubisco. Nat. Plants. 2020; 6: 1158-1166. 10.1038/s41477-020-00762-4</bibtext> </blist> <blist> <bibtext> Alonso H., Blayney M.J., Beck J.L., Whitney S.M. Substrate-Induced Assembly of Methanococcoides Burtonii D-Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase Dimers into Decamers. J. Biol. Chem. 2009; 284: 33876-33882. 10.1074/jbc.M109.050989</bibtext> </blist> <blist> <bibtext> Gunn L.H., Valegard K., Andersson I. A Unique Structural Domain in Methanococcoides Burtonii Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase (Rubisco) Acts as a Small Subunit Mimic. J. Biol. Chem. 2017; 292: 6838-6850. 10.1074/jbc.M116.767145</bibtext> </blist> <blist> <bibtext> Schreuder H.A., Knight S., Curmi P.M.G., Andersson I., Cascio D., Branden C.I., Eisenberg D. Formation of the Active Site of Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase by a Disorder-Order Transition from the Unactivated to the Activated Form. Proc. Natl. Acad. Sci. USA. 1993; 90: 9968-9972. 10.1073/pnas.90.21.9968. 8234342</bibtext> </blist> <blist> <bibtext> Seno Y., Go N. Deoxymyoglobin Studied by the Conformational Normal Mode Analysis. I. Dynamics of Globin and the Heme-Globin Interaction. J. Mol. Biol. 1990; 216: 95-109. 10.1016/S0022-2836(05)80063-4</bibtext> </blist> <blist> <bibtext> Levitt M., Sander C., Stern P.S. Protein Normal-Mode Dynamics: Trypsin Inhibitor, Crambin, Ribonuclease and Lysozyme. J. Mol. Biol. 1985; 181: 423-447. 10.1016/0022-2836(85)90230-X</bibtext> </blist> <blist> <bibtext> Schloss J.V. Comparative Affinities of the Epimeric Reaction-Intermediate Analogs 2- and 4-Carboxy-D-Arabinitol 1,5-Bisphosphate for Spinach Ribulose 1,5-Bisphosphate Carboxylase. J. Biol. Chem. 1988; 263: 4145-4150. 10.1016/S0021-9258(18)68901-X</bibtext> </blist> <blist> <bibtext> Ng J., Guo Z., Mueller-Cajar O. Rubisco Activase Requires Residues in the Large Subunit N Terminus to Remodel Inhibited Plant Rubisco. J. Biol. Chem. 2020; 295: 16427-16435. 10.1074/jbc.RA120.015759</bibtext> </blist> <blist> <bibtext> Watanabe H., Enomoto T., Tanaka S. Ab Initio Study of Molecular Interactions in Higher Plant and Galdieria Partita Rubiscos with the Fragment Molecular Orbital Method. Biochem. Biophys. Res. Commun. 2007; 361: 367-372. 10.1016/j.bbrc.2007.07.004</bibtext> </blist> <blist> <bibtext> Satagopan S., Chan S., Perry L.J., Tabita F.R. Structure-Function Studies with the Unique Hexameric Form II Ribulose-1,5-Bisphosphate Carboxylase/Oxygenase (Rubisco) from Rhodopseudomonas Palustris. J. Biol. Chem. 2014; 289: 21433-21450. 10.1074/jbc.M114.578625. 24942737</bibtext> </blist> <blist> <bibtext> Mueller-Cajar O. The Diverse AAA+ Machines That Repair Inhibited Rubisco Active Sites. Front. Mol. Biosci. 2017; 431. 10.3389/fmolb.2017.00031</bibtext> </blist> <blist> <bibtext> Taylor T., Andersson I. Structural Transitions during Activation and Ligand Binding in Hexadecameric Rubisco Inferred from the Crystal Structure of the Activated Unliganded Spinach Enzyme. Nat. Struct. Biol. 1996; 3: 95-101. 10.1038/nsb0196-95</bibtext> </blist> <blist> <bibtext> Atomi H., Fukui T., Kanai T., Morikawa M., Imanaka T. Description of Thermococcus Kodakaraensis Sp. Nov., a Well Studied Hyperthermophilic Archaeon Previously Reported as Pyrococcus Sp. KOD1. Archaea. 2004; 1: 263-267. 10.1155/2004/204953</bibtext> </blist> <blist> <bibtext> Anwar M., Choi S. Structure-Activity Relationship in TLR4 Mutations: Atomistic Molecular Dynamics Simulations and Residue Interaction Network Analysis. Sci. Rep. 2017; 7: 1-14. 10.1038/srep43807</bibtext> </blist> </ref> <aug> <p>By Vladimir Camel and Gaston Zolla</p> <p>Reported by Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib10" firstref="ref9"></nolink> <nolink nlid="nl2" bibid="bib12" firstref="ref10"></nolink> <nolink nlid="nl3" bibid="bib13" firstref="ref12"></nolink> <nolink nlid="nl4" bibid="bib15" firstref="ref13"></nolink> <nolink nlid="nl5" bibid="bib16" firstref="ref15"></nolink> <nolink nlid="nl6" bibid="bib17" firstref="ref17"></nolink> <nolink nlid="nl7" bibid="bib18" firstref="ref19"></nolink> <nolink nlid="nl8" bibid="bib19" firstref="ref20"></nolink> <nolink nlid="nl9" bibid="bib20" firstref="ref21"></nolink> <nolink nlid="nl10" bibid="bib21" firstref="ref22"></nolink> <nolink nlid="nl11" bibid="bib23" firstref="ref23"></nolink> <nolink nlid="nl12" bibid="bib25" firstref="ref24"></nolink> <nolink nlid="nl13" bibid="bib26" firstref="ref27"></nolink> <nolink nlid="nl14" bibid="bib27" firstref="ref28"></nolink> <nolink nlid="nl15" bibid="bib28" firstref="ref31"></nolink> <nolink nlid="nl16" bibid="bib30" firstref="ref34"></nolink> <nolink nlid="nl17" bibid="bib31" firstref="ref36"></nolink> <nolink nlid="nl18" bibid="bib32" firstref="ref38"></nolink> <nolink nlid="nl19" bibid="bib34" firstref="ref39"></nolink> <nolink nlid="nl20" bibid="bib35" firstref="ref40"></nolink> <nolink nlid="nl21" bibid="bib36" firstref="ref42"></nolink> <nolink nlid="nl22" bibid="bib37" firstref="ref47"></nolink> <nolink nlid="nl23" bibid="bib38" firstref="ref48"></nolink> <nolink nlid="nl24" bibid="bib39" firstref="ref49"></nolink> <nolink nlid="nl25" bibid="bib41" firstref="ref50"></nolink> <nolink nlid="nl26" bibid="bib42" firstref="ref51"></nolink> <nolink nlid="nl27" bibid="bib14" firstref="ref52"></nolink> <nolink nlid="nl28" bibid="bib43" firstref="ref53"></nolink> <nolink nlid="nl29" bibid="bib45" firstref="ref55"></nolink> <nolink nlid="nl30" bibid="bib47" firstref="ref56"></nolink> <nolink nlid="nl31" bibid="bib48" firstref="ref57"></nolink> <nolink nlid="nl32" bibid="bib49" firstref="ref58"></nolink> <nolink nlid="nl33" bibid="bib50" firstref="ref59"></nolink> <nolink nlid="nl34" bibid="bib51" firstref="ref60"></nolink> <nolink nlid="nl35" bibid="bib52" firstref="ref66"></nolink> <nolink nlid="nl36" bibid="bib53" firstref="ref67"></nolink> <nolink nlid="nl37" bibid="bib54" firstref="ref68"></nolink> <nolink nlid="nl38" bibid="bib55" firstref="ref70"></nolink> <nolink nlid="nl39" bibid="bib56" firstref="ref71"></nolink> <nolink nlid="nl40" bibid="bib57" firstref="ref72"></nolink> <nolink nlid="nl41" bibid="bib59" firstref="ref74"></nolink> <nolink nlid="nl42" bibid="bib60" firstref="ref75"></nolink> <nolink nlid="nl43" bibid="bib62" firstref="ref76"></nolink> <nolink nlid="nl44" bibid="bib63" firstref="ref78"></nolink> <nolink nlid="nl45" bibid="bib65" firstref="ref82"></nolink> <nolink nlid="nl46" bibid="bib66" firstref="ref84"></nolink> <nolink nlid="nl47" bibid="bib67" firstref="ref85"></nolink> <nolink nlid="nl48" bibid="bib69" firstref="ref86"></nolink> <nolink nlid="nl49" bibid="bib70" firstref="ref87"></nolink> <nolink nlid="nl50" bibid="bib71" firstref="ref91"></nolink> <nolink nlid="nl51" bibid="bib72" firstref="ref94"></nolink> <nolink nlid="nl52" bibid="bib73" firstref="ref95"></nolink> <nolink nlid="nl53" bibid="bib74" firstref="ref101"></nolink> <nolink nlid="nl54" bibid="bib76" firstref="ref102"></nolink> <nolink nlid="nl55" bibid="bib77" firstref="ref104"></nolink> <nolink nlid="nl56" bibid="bib78" firstref="ref107"></nolink> <nolink nlid="nl57" bibid="bib79" firstref="ref109"></nolink> <nolink nlid="nl58" bibid="bib80" firstref="ref119"></nolink> <nolink nlid="nl59" bibid="bib81" firstref="ref127"></nolink> <nolink nlid="nl60" bibid="bib83" firstref="ref129"></nolink> <nolink nlid="nl61" bibid="bib84" firstref="ref134"></nolink>
CustomLinks:
  – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsdoj&genre=article&issn=2218273X&ISBN=&volume=11&issue=12&date=20211101&spage=1761&pages=1761-1761&title=Biomolecules&atitle=An%20Insight%20of%20RuBisCO%20Evolution%20through%20a%20Multilevel%20Approach&aulast=Vladimir%20Camel&id=DOI:10.3390/biom11121761
    Name: Full Text Finder (for New FTF UI) (s8985755)
    Category: fullText
    Text: Find It @ SCU Libraries
    MouseOverText: Find It @ SCU Libraries
  – Url: https://doaj.org/article/1ad7b1963aad401f9654ffd8bcb08e10
    Name: EDS - DOAJ (s8985755)
    Category: fullText
    Text: View record from DOAJ
    MouseOverText: View record from DOAJ
Header DbId: edsdoj
DbLabel: Directory of Open Access Journals
An: edsdoj.1ad7b1963aad401f9654ffd8bcb08e10
RelevancyScore: 907
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 906.677795410156
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: An Insight of RuBisCO Evolution through a Multilevel Approach
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Vladimir+Camel%22">Vladimir Camel</searchLink><br /><searchLink fieldCode="AR" term="%22Gaston+Zolla%22">Gaston Zolla</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Biomolecules, Vol 11, Iss 12, p 1761 (2021)
– Name: Publisher
  Label: Publisher Information
  Group: PubInfo
  Data: MDPI AG, 2021.
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2021
– Name: Subset
  Label: Collection
  Group: HoldingsInfo
  Data: LCC:Microbiology
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Bio3D%22">Bio3D</searchLink><br /><searchLink fieldCode="DE" term="%22structural+dynamics%22">structural dynamics</searchLink><br /><searchLink fieldCode="DE" term="%22structural+flexibility%22">structural flexibility</searchLink><br /><searchLink fieldCode="DE" term="%22cross-correlation+dynamics%22">cross-correlation dynamics</searchLink><br /><searchLink fieldCode="DE" term="%22Microbiology%22">Microbiology</searchLink><br /><searchLink fieldCode="DE" term="%22QR1-502%22">QR1-502</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: RuBisCO is the most abundant enzyme on earth; it regulates the organic carbon cycle in the biosphere. Studying its structural evolution will help to develop new strategies of genetic improvement in order to increase food production and mitigate CO2 emissions. In the present work, we evaluate how the evolution of sequence and structure among isoforms I, II and III of RuBisCO defines their intrinsic flexibility and residue-residue interactions. To do this, we used a multilevel approach based on phylogenetic inferences, multiple sequence alignment, normal mode analysis, and molecular dynamics. Our results show that the three isoforms exhibit greater fluctuation in the loop between αB and βC, and also present a positive correlation with loop 6, an important region for enzymatic activity because it regulates RuBisCO conformational states. Likewise, an increase in the flexibility of the loop structure between αB and βC, as well as Lys330 (form II) and Lys322 (form III) of loop 6, is important to increase photosynthetic efficiency. Thus, the cross-correlation dynamics analysis showed changes in the direction of movement of the secondary structures in the three isoforms. Finally, key amino acid residues related to the flexibility of the RuBisCO structure were indicated, providing important information for its enzymatic engineering.
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: article
– Name: Format
  Label: File Description
  Group: SrcInfo
  Data: electronic resource
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 2218-273X
– Name: NoteTitleSource
  Label: Relation
  Group: SrcInfo
  Data: https://www.mdpi.com/2218-273X/11/12/1761; https://doaj.org/toc/2218-273X
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.3390/biom11121761
– Name: URL
  Label: Access URL
  Group: URL
  Data: <link linkTarget="URL" linkTerm="https://doaj.org/article/1ad7b1963aad401f9654ffd8bcb08e10" linkWindow="_blank">https://doaj.org/article/1ad7b1963aad401f9654ffd8bcb08e10</link>
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsdoj.1ad7b1963aad401f9654ffd8bcb08e10
PLink https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsdoj&AN=edsdoj.1ad7b1963aad401f9654ffd8bcb08e10
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/biom11121761
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 1
        StartPage: 1761
    Subjects:
      – SubjectFull: Bio3D
        Type: general
      – SubjectFull: structural dynamics
        Type: general
      – SubjectFull: structural flexibility
        Type: general
      – SubjectFull: cross-correlation dynamics
        Type: general
      – SubjectFull: Microbiology
        Type: general
      – SubjectFull: QR1-502
        Type: general
    Titles:
      – TitleFull: An Insight of RuBisCO Evolution through a Multilevel Approach
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Vladimir Camel
      – PersonEntity:
          Name:
            NameFull: Gaston Zolla
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 11
              Type: published
              Y: 2021
          Identifiers:
            – Type: issn-print
              Value: 2218273X
          Numbering:
            – Type: volume
              Value: 11
            – Type: issue
              Value: 12
          Titles:
            – TitleFull: Biomolecules
              Type: main
ResultId 1