Efforts have been made to overcome the intrinsic shortcomings of photosynthetic microorganisms by rewiring metabolism related to carbon fixation and photosynthesis. While efforts to improve RuBisCO have not been met with much success, current research has shifted focus towards rerouting metabolism to improve overall photosynthetic efficiency by focusing on key aspects of the Calvin-Benson cycle or the photosynthetic electron transport chain (PETC). One strategy used to harness the excess energy being lost by the PETC in cyanobacteria involved overexpressing the protein OmcS (Meng et al., 2021). This strategy coupled the excess electrons from the PETC to NADH production and was shown to increase intracellular ATP and NADH allowing for a fourfold improvement of D-lactate production in the cyanobacterium Synechococcus elongatus UTEX 2973 (hereon 2973) (Meng et al., 2021).

An inherent drawback of RuBisCO is its promiscuous nature, when RuBisCO undergoes oxygenase activity a costly side pathway known as photorespiration occurs where the oxygenase product is recycled back into usable metabolism consuming energy and losing CO₂ in the process. As much as 30% of energy produced by photosynthesis has been observed to be lost through photorespiration in plants (Hagemann and Bauwe, 2016). Rewiring or preventing photorespiration represents a promising way to improve the overall efficiency of carbon fixation in photosynthetic organisms. Efforts to rewire photorespiration generally involve deleting energetically costly steps, circumventing steps where CO₂ is lost, and rerouting metabolites towards central carbon metabolism (Hagemann and Bauwe, 2016). One of the more ambitious efforts to ameliorate the cost of photorespiration was the expression of a synthetic carbon capture pathway to serve as both a photorespiratory bypass and as a supplement to the Calvin-Benson cycle, this was shown to be a viable use of synthetic biology to counteract the costly natural photorespiration pathway (Shih et al., 2014).

It should be noted that photorespiration is not the sole pathway responsible for carbon inefficiencies, many metabolic processes include steps where CO₂ is lost to the environment. An important way to engineer microorganisms for sustainability involves carbon conservation, focusing on rerouting metabolism to circumvent decarboxylation reactions (François et al., 2020). Of the more notable strategies is the non-oxidative glycolysis pathway (NOG) which has been shown to function in Escherichia coli and which can effectively conserve all carbon associated with sugar catabolism to acetyl-CoA (Bogorad et al., 2013). While carbon conservation is a powerful methodology for engineering metabolism, the field is still in its infancy and further work is required to evaluate the industrial viability of many carbon conservation strategies. Additionally, de novo carbon fixation pathways, which will be addressed later in this review, are currently being developed and may prove to be a better methodology for the development of sustainable production hosts.

In contrast to research centering on canonical CO₂ fixation, investigations into de-novo CO₂ fixation pathways have been explored and theorized in recent years as more efficient alternatives to traditional RuBisCO based CO₂ assimilation. These pathways may provide advantages in chemical production hosts by offering insight into carboxylation reactions that could work in tandem with RuBisCO. The expression of formate dehydrogenase in the cyanobacterium, Anabaena sp. PCC 7120, was shown to successfully increase intracellular formate concentration, representing an alternative to the photo-reduction of CO₂ and can act to supplement natural carbon fixation pathways (Ihara et al., 2013).

Many of these de novo CO₂ fixation pathways have had limited success when installed into model organisms such as E. coli and yeast and it has yet to be shown if these pathways can function effectively in photosynthetic hosts. One pathway of note that has been shown to work in E. coli is the reductive glycine pathway, hereafter RGP (Tashiro et al., 2018). This pathway leverages the native glycine cleavage system in the reverse direction to combine one equivalent of CO₂ with 5,10-methylenetetrahydrofolate that has been produced from formate to produce pyruvate. This method allows E. coli to directly assimilate CO₂ into central metabolic pathways and is a more efficient method for CO₂ fixation than traditional RuBisCO (Bar-Even et al., 2013). This inorganic carbon can then be leveraged for biochemical synthesis. Additional work has also recently shown that the expression of formate dehydrogenase confers further renewable characteristics to strains harboring the RGP by removing the need for glucose supplementation (Bang et al., 2020). Other notable CO₂ fixation pathways include the crotonyl-CoA/ethylmalonyl-CoA/hydroxybutyryl-CoA (CETCH) cycle and the tartronyl-CoA (TaCo) pathways (Scheffen et al., 2021). While the RGP has proven to be a viable carbon fixation pathway that was shown to function in E. coli, the growth exhibited by this CO₂ fixing E. coli is slower than its traditional heterotrophic phenotype (Tashiro et al., 2018; Bang et al., 2020). Advancements in modeling and metabolomics may allow for an increase in the creation of de-novo carbon fixation pathways that may prove to be both more efficient than traditional pathways and capable of functioning in a wider array of chemical production hosts.

Another approach to increase chemical production capacity is to supplement CO₂ with carbohydrates as an auxiliary carbon source for the Calvin-Benson cycle, thereby making the organism photomixotrophic. By re-engineering glucose catabolism to direct carbon flux into the Calvin-Benson cycle, more ribulose-1,5-bisphosphate can be supplied to RuBisCO, accelerating CO₂ fixation. This ultimately results in faster growth and production of downstream targets, as well as a six-fold increase in titer once metabolism was rewired to accommodate photomixotrophy (Kanno et al., 2017). The addition of a heterotrophic mode also allows for CO₂ fixation in darkness, resulting in a 24 h production period under natural diurnal conditions. Glucose can be readily obtained from the acid hydrolysis of agricultural waste products such as corn stover, in conjunction with other sugars: xylose, arabinose, and galacturonic acid (Mourtzinis et al., 2016). By installing catabolic pathways for these non-glucose sugars, these agricultural waste products can be used more efficiently. Xylose, the second most abundant sugar in corn stover lysate, has successfully been used to achieve photomixotrophic production of 2,3-butanediol in light and dark conditions with significant improvements in growth and product titer over the equivalent photoautotrophic organism (McEwen et al., 2016). A similar strategy has also recently been used to improve the production of 3-hydroxypropionic acid by 4.1 fold in cyanobacteria through the installation of a xylose photomixotrophic module along with other modifications to help assimilate the additional carbon source (Yao et al., 2022). It should also be noted that the cyanobacterium Synechocystis sp. PCC 6803 (hereafter 6803) natively possesses photomixotrophic machinery to assimilate glucose. A recent study was able to improve these photomixotrophic conditions through the installation of the NOG pathway along with targeted gene knockouts to increase the intracellular concentration of acetyl-CoA, thereby improving the growth phenotype of 6803 (Song et al., 2021).

Most research done in this field focuses on using just a handful of species that have traditionally been used as model organisms to study the mechanics of photosynthesis. With increasing interest in using photosynthetic organisms for industrial production there have been efforts to uncover new species that are faster growing and more receptive to engineering. In the realm of cyanobacterial chemical production, species like 2973 and S. elongatus PCC 11802 (hereon 11802) have risen in popularity as they are faster growing than the traditional S. elongatus PCC 7942 (hereon 7942) and have been shown to produce higher titers of target chemical products under certain circumstances (Yu et al., 2015; Sengupta et al., 2020). Additionally, these fast-growing organisms are providing inspiration for how to better engineer existing model organisms. The fast-growing cyanobacterium 2973 has relatively little differences genetically when compared to the model cyanobacterium 7942 (Yu et al., 2015). However, a notable difference in 2973 is an increase in the expression levels of PSI, cytochrome b₆f, and plastocyanin on a per cell basis which improves the downstream flux of electrons from PSII, which helps the faster growing cyanobacteria to better utilize photosynthetic energy (Ungerer et al., 2018). The discovery of new fast-growing cyanobacteria may enhance our understanding of photosynthesis and characterizing the differences between these new species with current model organisms.

In cyanobacteria, traditional genomic modifications are a labor-intensive task and limiting in nature due to the polyploidal nature of these organisms and the need for antibiotic resistance markers (Griese et al., 2011). The current methodology for genomic integration involves constructing a plasmid with an antibiotic selection marker in a plasmid host such as E. coli. After introduction of this plasmid to cyanobacteria, several rounds of antibiotic screening are required to ensure complete genome segregation (Golden et al., 1987). This process generally limits the number of modifications that can be performed in a single strain due to the physiological constraints of expressing multiple different antibiotic resistance genes.

The overall task of metabolic engineering in cyanobacteria has been made dramatically more efficient thanks to the advent of CRISPR gene editing which allows for markerless edits (Behler et al., 2018). However, the protein Cas9 is toxic to a number of cyanobacteria species (Wendt et al., 2016). Researchers have recently uncovered other endonucleases that are similarly capable of CRISPR gene editing. The main endonuclease of interest is Cpf1 which, while similar to Cas9, is better tolerated by photosynthetic hosts (Ungerer and Pakrasi, 2016). As the body of research grows around Cpf1, more engineering strategies will be made available in the realm of photosynthetic chemical production and should offer a boon towards the viability of these organisms to begin replacing their non-CO₂ fixing brethren in the realm of biochemical production (Bishé et al., 2019; Niu et al., 2019). Additionally, having the ability to perform markerless genomic modifications unlocks the potential to engineer these microorganisms far more ambitiously than what was previously possible.

Other work on CRISPR technologies in cyanobacteria includes the use of CRISPR inhibition (CRISPRi) by using dead Cas9 (dCas9) (Qi et al., 2013). While the endonuclease activity of the intact Cas9 protein seems to be toxic to these production hosts, dCas9 is able to function in the same manner in photosynthetic hosts as it is able to in heterotrophic hosts such as E. coli (Santos et al., 2021). While the use of dCas9 may not be as broadly useful as traditional CRISPR, dCas9 has been shown to be invaluable in certain chemical production applications where more traditional gene knockouts would otherwise be toxic.