The Calvin-Benson Cycle serves as the primary carbon-fixation pathway in nearly all photosynthetic organisms.

carbon fixation. Although many efforts have been made to enhance RuBisCO’s carbon fixing efficiency,

significant advancements are still needed. In this study we examine the current understanding of RuBisCO’s

catalytic mechanism, its complex oxygenase activity, evolutionary trade-offs, and new developments in

RuBisCO engineering.

The Calvin-Benson Cycle is the most dominant carbon-fixation mechanism in almost all photosynthetic

organisms form prokaryotes, to protists such as algae, to green plants. As the enzyme that catalyses the

immobilization/fixation of atmospheric CO

Bisphosphate Carboxylase/Oxygenase (RuBisCO) is the most crucial and the rate-limiting component of the

elusive oxygenase activity, evolutionary trade-offs, and the to highlight the progress in RuBisCO engineering.

As the primary enzyme for carbon fixation across all domains of life, the core tertiary structure of RuBisCO is

highly conserved between vastly different phylogenetic clades. The key catalytic component of RuBisCO is its

large core subunit, also denoted as RbcL [3]. A single RbcL monomer consists of a N-terminal domain, which

promotes dimerization, and a TIM Barrel fold situated at the C terminal (Figure.1). This classic TIM Barrel

structure consists of an inner β-sheet-based cylindrical structure with a surrounding ring of alpha-helices. The

key catalytic residues are situated inside the β-barrel, with a divalent magnesium ion coordinated at the centre,

forming the active site of RuBisCO. A lid-loop domain flanking the β-Barrel (also called loop-6) is an integral

component of the active site and gates the entrance of the substrate RuBP [4]. When loop 6 adopts a closed

conformation, the complete RuBisCO active site is formed. Of those highly conserved catalytic residues, a single

modified side chain called carbamoylysine [1]. The moiety functions as a general base during the catalysis

carried out by RuBisCO, which will be described later in the article.

Figure.1 (A) Ribbon-band crystallographic structure of spinach RuBisCO. The catalytic TIM Barrel is coloured

in olive green, whilst the N-terminal domain is in light green. A Mg2+ ion (light green) and 6C-intermediate

mimic molecule (CABP) bind to the centre of the TIM barrel. (B) A closer look at the active site of RbcL. Note

that the strategic carbamoylysine (KCX 201) is situated right below the Mg2+ ion. Crucial residues are coloured

orange, whilst the lid-loop region (loop 6) is coloured light orange. The crystallographic structures were

originally downloaded from the PDB RCSB database (PDB ID: 8RUC) and then processed and coloured by

modern plants, 8 RbcS oligomerizes with 4 pairs of RbcL2 dimers, forming a hexadecamer denotated as

(RbcL)8(RbcS)8. Other forms of RuBisCO, including forms II and III, form RbcL2 homo-oligomers with

different numbers of RbcL2 involved, but lacking the RbcS subunit [6]. The actual role of RuBisCO small

subunit is still under investigation; some evidence suggest that the RbcS is responsible for the correct folding

and assembly of RbcL, whilst others delineate a regulatory role of RbcS in RuBisCO’s catalysis [7].

Figure.2 Different forms of functional RuBisCO units in different organisms. In modern green plants and green

algae, Form I RuBisCO (Left) is dominant, and it exists as a hetero octamer formed by RbcL and RbcS subunits.

In many prokaryotes, only a dimer of RbcL is present (middle). Other forms of oligomeric RuBisCO include the

chloroplasts. The expression of RuBisCO chaperones is crucial for the heterologous production of RuBisCO in

model organisms such as Escherichia coli, which facilitates the further analysis of its functionality owing to the

short replication cycle of prokaryotes [9], compared to the green plants. Apart from the chaperones that assist

the folding process of RuBisCO, a special RuBisCO chaperone called RuBisCO Activase (RCA) is responsible

for removing an inhibitory RuBP molecule bound to the inactive RuBisCO [10], and the chemical modification

of the lysine residue into carbamoylysine. These actions facilitate the formation of the RuBisCO-Mg2+ complex,

making the RuBisCO enzyme ready for catalysis.

This study is intended to comprehensively review the current scientific perceptions and developments regarding

the carboxylase activity of Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) for greater carbon

capture. A systematic literature search was used for the study, aimed at the identification, analysis, and synthesis

of information from peer-reviewed, scientific literature.

A comprehensive and wide search of literature was performed on major academic databases such as, but not

limited to, PubMed, Scopus, Web of Science, and Google Scholar. The search strategy made use of keyword

combinations and phrases associated with RuBisCO and carbon fixation such as: "RuBisCO," "carbon capture,"

"carboxylase activity," "oxygenase activity," "photorespiration," "RuBisCO engineering," "site-directed

mutagenesis," "directed evolution," "rational design," "molecular grafting," "ancient RuBisCO," "carbon dioxide

photorespiration.

specificity.

mechanisms or synthetic carbon fixation pathways.

efficiency in plants or other photosynthetic organisms.

Studies were excluded if they: focused solely on the non-photosynthetic carbon metabolism or had no close

design).

parameters in engineered systems.

The extracted information was then critically analysed, categorised, and synthesised to extract key points,

common challenges, promising strategies and emerging trends in RuBisCO engineering and carbon capture

itself or its surrounding biochemical environment. The discussion progresses from classical protein engineering

techniques to advanced computational design, molecular grafting, and the implementation of sophisticated

carbon concentrating mechanisms and synthetic pathways. The review concludes by proposing a combinatorial

strategy to integrate these diverse approaches for synergistic effects, aiming to provide a roadmap for future

research in boosting carbon capture.

The catalytic mechanism of RuBisCO demonstrates the multi-step nature of the entire process (Figure.3). The

mechanism can be divided into four major stages [11]: First, the activation of RuBisCO, which is catalysed by

the chaperone RCA, as we have mentioned earlier. Second, the generation of the 6C enediolate from RuBP,

which is essential for both the carboxylation and the oxygenation reaction. This procedure is initiated by the

Figure.3 The detailed catalytic mechanism of RuBisCO. An Mg

bound enediolate intermediate is required by

both the carboxylation and the oxygenation reaction. Note that the exact detail of the oxygenation reaction

remains enigmatic (see text). (Source: Prywes et al., 2023) [2].

Figure.4 The strong positive correlation between Kcat-CO

can be explained by

RuBisCO’s catalytic mechanism. This indicates a relatively constant Sc/o. Some clades (such as Red algae) may

have slightly greater Sc/o compared to others. (Source: Flamholz et al., 2019) [17].

also be represented mathematically by the partition coefficient, or τ , which is the gradient of carboxylase activity

against the oxygenase activity of the RuBisCO enzyme (Gutteridge et al., 1984):

first high-resolution crystallographic structure of RuBisCO was solved by Andersson [24] in the mid-90s. Some

mutations target the amino acid residues at or adjacent to the active site, whilst many others concentrate on a

dynamic loops structure termed as loop-6 (See Figure. 1). Though the rationale of many of these mutations were

inferred from the characterized catalytic mechanism of RuBisCO, many ended up compromising the enzymatic

activity. Some site-directions, though, resulted in a modest improvement in either the catalytic rate (Kcat-CO

or specificity (Sc/o) [19].

Whilst these rational, site-directed mutations could only explore a limited portion of the sequence space (all the

possible permutations and combinations of amino acid residues in a polypeptide of give length) near the wild-

type, the “irrational” approach of directed-evolution (DE) allows for a wider discovery of the potentially optimal

organism for directed evolution. The selective pressure can be readily introduced via an enzyme named as

phosphoribulokinase (prk) (Figure.6). The enzyme phosphorylates a common metabolite – Ribulose-5-

phosphate (Ru5P) in the pentose phosphate pathway (PPP) of the bacteria, generating Ribulose-bisphosphate

(RuBP), which cannot be catabolized by other enzymes in E. coli thereby creating a metabolic dead-end. The

uncontrolled accumulation of RuBP imposes a toxic effect on the bacteria, which can only be alleviated by the

newly introduced enzyme RuBisCO, which uses RuBP as the substrate. It was believed that, after rounds of this

selection, E. coli lineages with enhanced RuBisCO functionality will emerge [19]. Variant strategies that involve

the perturbation of the glycolytic enzymes were also designed, which makes RuBP the only viable source of

energy [2].

Figure.6 Escherichia coli as the chassis for directed evolution. A toxic pathway (brown) containing the enzyme

PRK can be introduced into the metabolic network. If RuBP cannot be efficiently removed by RuBisCO (blue

arrow), its toxic accumulation will hamper the growth of the bacteria, thus exerting a selective pressure.

Knocking out the glycolytic enzymes (red crosses) can further enhance this selective pressure by making PRK-

RuBisCO the only energy acquisition pathway. (The idea was derived from Cai et al., 2014, the diagram was

illustrated by Biorender.)

in the de novo design of the enzyme luciferase [28], with kinetic parameters comparable to that of the wild type,

and a significant improvement in substrate selectivity is also observed. If the same principle could be applied to

the enzyme RuBisCO in near future, we could expect new RuBisCO variants with improved Kcat-CO

been engineered. A more recent, open-source generative AI model developed by the same research group named

as RF diffusion [27] further illuminates the potential of de novo protein design and active-site modelling.

protein tyrosine-phosphatases [29]. The activity of RuBisCO is also determined by the dynamic loop termed as

loop-6 (see previous sections), which controls the entry of substrates and forms a portion of the active-site. If

generative AIs such as RF diffusion can be deployed to re-engineer the loop-6, an enhancement in RuBisCO’s

specificity towards CO

discovering the superlative RuBisCO variants, due to its much wider exploration of the sequence space compared

to site-directed mutations and directed-evolutions (Figure.7).

Figure.7 Visualizing the concept of sequence space. The central dark green point is the original sequence.

Different improvement strategies have different extent of coverage to the entire sequence space. AI-aided design

holds the greatest development potential owing to its wide sequence space coverage. (The diagram was

illustrated via Biorender.)

alternative solution, since it appears not requiring any of the complex and time-consuming molecular engineering

in the previous methods. However, many of the initial attempts in RuBisCO molecular grafting ended up with

fiasco, owing to the misfolding of newly introduced red algal RuBisCO in plant’s chloroplasts [30]. This is

mainly due to the incompatibility of the chaperone system between red algae and plants. It was highly pitiful

that, in a study carried out in 2018 [31], the RuBisCO from red algae Griffithsia monilis (GmRuBisCO) failed

to function normally when heterologous expressed in tobacco plants. Owing to its spectacular catalytic efficiency,

the successful grafting of GmRuBisCO would have great significance to the lant growth. This failure presents

as a setback for this methodology.

It was later discovered that the RuBisCO variant from purple bacterium Rhodobacter sphaeroides is less

susceptible to the incompatibility of chaperone system [32]. This variant (RsRuBisCO) could achieve a stable

(329A and 332V) on the loop-6 of GmRuBisCO to the one of RsRuBisCO, the chimera still exhibits normal

expression and folding, whilst the kcat-co2 parameter increased by 60%, efficiency of CO

-fixation increased

by 22% and a 7% increase in Sc/o was observed. This presents as a major progression in RuBisCO engineering,

since the tobacco plants transformed with the engineered chimeric RsRuBisCO exhibits a twofold growth rate

compared to the plant that expresses wild-type RsRuBisCO [20].

Apart from the utilization of existing RuBisCO variants, another very innovative approach emerged recently is

to use phylogenetic study to elucidate the amino acid sequence of an ancient specie’s RuBisCO, which functioned