Prof. James C. Liao, President of Academia Sinica, Taiwan, Professor and Chair of the Department of Chemical and Biomolecular Engineering at the University of California, Los Angeles

Introduction

Following the Industrial Revolution in the 18th century, fossil fuels such as coal, petroleum, and natural gas have been heavily exploited and used as primary energy sources. The combustion of these fuels releases significant amounts of CO₂, leading to a rapid increase in atmospheric CO₂ concentration over recent decades. Additionally, the growing world population has led to increased agricultural activities, which emit CO₂ and other greenhouse gases, such as methane.

The accumulation of greenhouse gases has caused global warming and climate change, which could lead to catastrophic consequences if not addressed. Despite efforts to reduce greenhouse gas emissions, progress has been insufficient. Currently, global CO₂ emissions primarily from fossil fuel combustion, agriculture, and land use changes, total approximate 40 billion tons per year and are still increasing. The CO₂ level in the atmosphere is greater than 420 parts per million (ppm), which is 100 ppm more than the level in 1960.

Biology offers a potential solution for absorbing greenhouse gases. Plants, algae, and cyanobacteria have evolved to utilize carbon dioxide, and certain microorganisms, such as methanotrophs, can use methane for growth. Biological organisms naturally assimilate approximately 440 billion tons of CO₂ from the atmosphere each year, which is more than ten times the amount of CO₂ produced by human activities. However, about half of this assimilated CO₂ is rapidly respired back into the atmosphere, and the remaining portion is slowly released as CO₂ or methane.

Before the Industrial Revolution, Earth’s natural biological systems effectively balanced the carbon cycle, keeping atmospheric CO₂ levels in check. However, the rapid increase in human-caused CO₂ emissions, driven by fossil fuel combustion, agricultural expansion, and land-use changes, has overwhelmed this natural balance.

Despite this challenge, biological systems still offer one of the few scalable solutions for mitigating atmospheric greenhouse gases. By boosting the rate at which plants and microorganisms capture carbon dioxide, we could convert it into forms that benefit agriculture or replace fossil fuels. Such enhanced carbon assimilation could significantly reduce human-caused greenhouse gas emissions by improving agricultural efficiency and reducing reliance on fossil fuels. To achieve this vision, the first step is to enhance CO₂ fixation by plants and cyanobacteria, as well as increase methane or methanol assimilation by microorganisms.

Nature has evolved several molecular systems to fix one-carbon (C1) compounds such as CO₂, methane, and methanol. However, such systems are insufficient to deal with the increased challenge posed by the increased carbon emission due to human activities. Thus, new mechanisms to boost C1 fixation need to be designed and evolved. Here we discuss a few examples of how biological systems are re-designed to boost C1 fixation using plants and microorganisms as examples.

I.               Recycling the photorespiration product

The majority of CO₂ on earth is fixed by an enzyme called RuBisCO, the most abundant protein on earth. Through billions of years’ evolution, an equilibrium between activity and specificity has been reached. Consequently, RuBisCO cannot distinguish been CO₂ and O₂, and possesses both a carboxylase and an oxygenase activity. The oxygenase activity typically accounts for 20-30 percent of RuBisCO’s activity in C3 plants. This activity produces glycolate, which is converted into glycerate in peroxisomes and CO₂ in mitochondria (Maier et al., 2012) accompanied by ammonia release. This process, called photorespiration, is thought to be one of the factors limiting agricultural productivity and CO₂ fixation on earth. In the past decade or so, several novel pathways have been devised to recycle the photorespiration product to a productive pathway.

The first approach is through complete oxidation of the photorespiration product, glycolate, to CO₂ in chloroplasts, followed by re-assimilation of CO₂. Native plants degrade glycolate in peroxisome and mitochondria. If glycolate can be degraded completely to CO₂ in chloroplasts, CO₂ can be re-assimilated by RuBisCO directly without ammonia release.

This concept was implemented (Maier et al., 2012) by introducing two heterologous enzymes, a malate synthase (MS) from pumpkin and a catalase (CAT) from E. coli, and overexpressing an endogenous enzyme, peroxisomal glycolate oxidase (GO) in Arabidopsis chloroplasts (Fig. 1a). With these modifications, glycolate is converted to CO₂ in chloroplasts and re-assimilated by RuBisCo. The resulting transgenic Arabidopsis showed an increase in rosette size and quantity, as well as greater biomass. Nevertheless, incorporating the three aforementioned genes into tobacco plants did not enhance their biomass in greenhouse experiments (South et al., 2019). Interestingly, when glycolate oxidase (GO), which produces H₂O₂, is substituted with glycolate dehydrogenase (GDH) from Chlamydomonas (Fig. 1a), which produces NADH, the transgenic tobacco demonstrated higher carbon assimilation rates, enhanced resistance to photorespiration stress, and a significant increase in biomass during field tests (South et al., 2019). This approach also prevented yield loss at high temperatures (Cavanagh et al., 2022).

Another implementation of this concept is a pathway called GOC (Shen et al., 2019), which consists of a rice glycolate oxidase (OsGLO3), an oxalate oxidase (OsOXO3), and a catalase (OsCATC) (Fig.1a). These enzymes were expressed in rice chloroplasts, and completely converted glycolate to CO₂ via oxalate. The GOC pathway in rice led to as much as 22% improvement in photosynthesis; however, field tests showed that the yields fluctuated with the season (Shen et al., 2019).

Instead of fully oxidizing glycolate, a synthetic pathway derived from E. coli can partially oxidize two glycolate molecules into one CO₂ and one glycerate (Fig. 1b). Both of these products can be assimilated in the chloroplast through the Calvin-Benson-Bassham (CBB) pathway. This synthetic pathway retains 75% of the carbon from two glyoxylate molecules to produce glycerate, which is then recycled back into the CBB cycle. The remaining carbon is released as CO₂, which can theoretically be reassimilated by RuBisCO. Expression of these genes in Arabidopsis chloroplasts (Kebeish et al., 2007) led to increased growth rates and biomass yields. This pathway also proved beneficial for some crop plants such as Camelina sativa (Dalal et al., 2015) and potato (Nölke et al., 2014) under greenhouse and growth chamber conditions.

II.              Synthetic CO₂ fixing pathways

In addition to recycling photorespiration products from RuBisCo, other efforts aim to increase CO₂ fixation by devising synthetic CO₂ fixation pathways. These include the Malyl-CoA glycerate (MCG) cycle (Yu et al., 2018), and the tartronyl-CoA (TaCo) pathway (Trudeau et al., 2018), the CETCH pathway (Schwander et al., 2016) and the rPS-MCG cycle (Luo et al., 2022). Except for the first one, others have not been implemented in organisms, and still in the stage of in vitro development.

The MCG cycle (Yu et al, 2018) utilizes an abundant oxygen-insensitive carboxylase, phosphoenolphyruvate carboxylase (PPC), and forms a cycle with 3-, 4-, and 2- carbon intermediates (Fig. 1c). The additional carboxylase can increase CO₂ fixation rate, and the product, acetyl-coA is a key metabolite in all cells. Thus, the advantage of this cycle is that it is compatible with existing organisms. This cycle was implemented in cyanobacteria Synechococcus elongatus PCC7942 (Yu et al., 2018) and showed increased CO₂ fixation and acetyl-CoA production.

Since RuBisCO exhibits a limitation in distinguishing between oxygen and CO₂, other carboxylases have been used, including PPC mentioned above, crotonyl-CoA carboxylase/reductase (CCR) (Schwander et al., 2016) and a new-to-nature glycolyl-CoA carboxylase (GCC) (Scheffen et al., 2021). To utilize CCR as the carbon-fixing enzyme, a synthetic cycle, CETCH (Schwander et al., 2016), has been designed and tested in vitro. Another synthetic cycle rPS-MCG (Luo et al., 2022), which utilizes both CCR and PEPC has also been demonstrated in vitro with optical controls for cofactor regeneration. Both cycles have shown in vitro CO₂ fixation activities comparable to plants. Another synthetic pathway, the TaCO pathway, utilized a rationally designed and then evolved GCC to carboxylaste glycoly-CoA to make glycerate, which is then assimilated in the CBB pathway (Scheffen et al., 2021).

III.            Deleting a cyanobacterial CBB brake to boost carbon fixation

Similar to plants, cyanobacteria utilize solar energy to split water and produce NADPH and ATP to support carbon fixation via the CBB cycle. Nevertheless, the cellular energy status is dynamic because of light fluctuation and various stress conditions. Controlling utilization of energy and reducing equivalents by carbon fixation is pivotal for cell growth and survival. Sophisticated regulatory mechanisms have been reported to control the CBB cycle in cyanobacteria. The post-translational oscillator KaiABC modulate circadian gene expressions, based on phosphorylation states of the core clock regulators, to control various processes including photosynthesis and nitrogen fixation (Ishiura et al., 1998). Additionally, the thioredoxin-modulated protein CP12 regulates some CBB enzyme activities based on the redox state through protein complex formation (Lucius et al., 2022). In contrast, regulation of carbon fixation in response to energy fluctuations remains unclear. We recently discovered that S. elongatus phosphoketolases (SeXPK) plays an important role in dynamic regulation of carbon fixation flux (Fig. 2a) (Lu et al., 2023).

We found that under light conditions, S. elongatus PCC7942 contains high levels of ATP, which inhibits SeXPK though a unique ATP binding domain (Fig. 2a), which exists between two SeXPK subunits in the dodecamer (Lu et al., 2023). However, during the transition from light to dark conditions, the ATP level drops within a minute and XPK is released from ATP-inhibition. During this time, SeXPK becomes active and divert the CBB cycle intermediate Xu5P to C2 and C3 compounds (Fig. 2b). This action drains the CBB cycle intermediates and shuts down carbon fixation immediately. When the cell is exposed to light again, its ATP level rapidly returns to normal, and inhibits XPK activity to allow carbon fixation. Therefore, XPK serves as a metabolic “brake” to control carbon fixation under fluctuating light conditions (Fig. 2b).

Interestingly, when XPK is deleted, S. elongatus exhibited higher CO₂ fixation under fluctuating light conditions. This fluctuating light condition can also be created in high density cultures by self-shading. The DSeXPK mutant showed 60% higher CO₂ fixation in high density shaking culture (OD730=4-7). Unexpectedly, the additional carbon fixation all resulted in sucrose and secreted into the medium (Fig. 2b) (Lu et al., 2023). This result creates a possibility to use S. elongatus for sucrose production after optimizing productivity. Since cyanobacteria grows faster than plants, the system may provide an additional avenue for agricultural production.

IV.           Evolution engineering for methanol assimilation

Methanol is the only C1 compound that exists in the liquid form under ambient conditions. It can be derived from either CO₂ or methane, both of which are potent greenhouse gases. Therefore, effective methanol utilization may open the door for recycling C1 greenhouse gases to useful products. Thus, developing an industrial organism that can assimilate methanol as the sole carbon source for growth is essential for such an application. Compared to glucose utilization, methanol utilization requires only three additional enzymes (methanol dehydrogenase, Hps, Psi) to complete the pathway. Thus, it was deceivingly straightforward to engineer E. coli, a glucose-utilizing industrial microorganism, to grow on methanol as the sole carbon source. However, several groups have attempted to achieve this goal, but have met with surprising difficulties. We have developed a strategy to evolve an E. coli strain to grow on methanol as a carbon source, and identified the mechanistic difficulties involved.

The first step was to create an E. coli strain whose growth on xylose was dependent on methanol (Fig. 3a) (Chen et al., 2018). This is done by deleting two essential genes rpiAB, such that the xylose metabolic product ribulose 5-phosphate (Ru5P) cannot be further metabolized. This strain was then rescued by introducing three heterologous genes, mdh, hps, psi which can convert methanol and Ru5P to fructose 6-phosphate, which then enters the normal glycolytic pathway. This strain can grow on xylose only in the presence of methanol, and is called a methanol auxotroph.

The second step was to evolve this methanol auxotroph to wane down its dependence on xylose by expressing rpiAB. This step turns out to be the most challenging. After several trials, we discovered that the first step of methanol metabolism produces formaldehyde, which is highly toxic because it causes DNA-protein cross-linking (DPC) (Chen et al., 2020). During cell growth, formaldehyde accumulates and diffuses out of the cell to affect neighboring cells. So even if some beneficial mutations emerge to allow cell growth, formaldehyde accumulation at the stationary phase eventually killed the cell. Thus, this cell is not evolvable. Native methylotrophs have evolved a formaldehyde detoxification process that oxides this compound to CO₂, but it is counter productive if we try to utilize this compound for growth. Furthermore, since this compound can freely diffuse through the cell membrane to poison the neighboring cells, beneficial mutations are difficult to be selected or enriched in a culture. We used both rational design and evolution techniques and successfully developed a synthetic methylotrophic E. coli strain (Chen et al., 2020).

To further improve the synthetic methylotrophic E. coli, we developed a copy number tuning technique. This technique is based on an unexpected finding that a particular sequence on the ddp operon was able to intervene the replication of Bacterial Artificial Chromosome (BAC) and cause it to form tandem repeats of concatemers with a wide-range of copy number distribution. We used this technique to dynamically tune the expression level of enzymes that are involved in formaldehyde consumption. The introduction of this ddp-BAC containing the formaldehyde consumption genes (RHTTP) was able to facilitate the cell to overcome the evolutionary block of DPC, even though at the cost of higher protein burden. After the cell continue to evolve, the cell was able to accumulate beneficial mutations that were able to avoid DPC more effectively, and tuned down the copy number of ddp-BAC::RHTTP by an IS insertion of the recA gene. Thus, this dynamic copy number tuning technique provides an expediency to overcome an evolutionary block and allow the accumulation of other beneficial mutations to occur at a much lower frequency.

After several rounds of evolution, we have developed a synthetic methylotrophic E. coli strain with a doubling time of 3.5 hours, faster than native model methylotrophs, Methylorubrum extorquens AM1 (T_d~4hr) and Bacillus methanolicus at 37°C (T_d~5hr).

Conclusion

Biology has provided one of the best approaches at scales to assimilate C1 greenhouse gases. However, native organisms did not evolve to solve the problems caused by the anthropogenic release of C1 greenhouse gases. Artificial design and evolution must be used to accelerate this process. Here we showed a few examples to increase C1 assimilation by redesigning the carbon fixation cycle in plants and microorganisms. In many cases, human design can only provide the biochemical capability to enable new biochemical pathways. It must be followed by laboratory evolution for fine-tuning to fit organismal growth demand. The examples discussed here provide some enabling tools for further investigations. These tools can be pivotal in developing sustainable solutions for reducing greenhouse gas concentrations and mitigating climate change.

References

1.     Maier A, Fahnenstich H, von Caemmerer S, Engqvist MK, Weber AP, Flügge UI, Maurino VG. Transgenic Introduction of a Glycolate Oxidative Cycle into A. thaliana Chloroplasts Leads to Growth Improvement. Front Plant Sci. 2012 Feb 28;3:38. doi: 10.3389/fpls.2012.00038. PMID: 22639647; PMCID: PMC3355595.

2.     South PF, Cavanagh AP, Liu HW, Ort DR. Synthetic glycolate metabolism pathways stimulate crop growth and productivity in the field. Science. 2019 Jan 4;363(6422):eaat9077. doi: 10.1126/science.aat9077. Epub 2019 Jan 3. Erratum in: Science. 2019 Aug 2;365(6452):eaay8818. doi: 10.1126/science.aay8818. PMID: 30606819; PMCID: PMC7745124.

3.     Cavanagh AP, South PF, Bernacchi CJ, Ort DR. Alternative pathway to photorespiration protects growth and productivity at elevated temperatures in a model crop. Plant Biotechnol J. 2022 Apr;20(4):711-721. doi: 10.1111/pbi.13750. Epub 2021 Nov 24. PMID: 34786804; PMCID: PMC8989507.

4.     Shen BR, Wang LM, Lin XL, Yao Z, Xu HW, Zhu CH, Teng HY, Cui LL, Liu EE, Zhang JJ, He ZH, Peng XX. Engineering a New Chloroplastic Photorespiratory Bypass to Increase Photosynthetic Efficiency and Productivity in Rice. Mol Plant. 2019 Feb 4;12(2):199-214. doi: 10.1016/j.molp.2018.11.013. Epub 2019 Jan 10. PMID: 30639120.

5.     Kebeish R, Niessen M, Thiruveedhi K, Bari R, Hirsch HJ, Rosenkranz R, Stäbler N, Schönfeld B, Kreuzaler F, Peterhänsel C. Chloroplastic photorespiratory bypass increases photosynthesis and biomass production in Arabidopsis thaliana. Nat Biotechnol. 2007 May;25(5):593-9. doi: 10.1038/nbt1299. Epub 2007 Apr 15. PMID: 17435746.

6.     Dalal J, Lopez H, Vasani NB, Hu Z, Swift JE, Yalamanchili R, Dvora M, Lin X, Xie D, Qu R, Sederoff HW. A photorespiratory bypass increases plant growth and seed yield in biofuel crop Camelina sativa. Biotechnol Biofuels. 2015 Oct 29;8:175. doi: 10.1186/s13068-015-0357-1. PMID: 26516348; PMCID: PMC4625952.

7.     Nölke G, Houdelet M, Kreuzaler F, Peterhänsel C, Schillberg S. The expression of a recombinant glycolate dehydrogenase polyprotein in potato (Solanum tuberosum) plastids strongly enhances photosynthesis and tuber yield. Plant Biotechnol J. 2014 Aug;12(6):734-42. doi: 10.1111/pbi.12178. Epub 2014 Mar 7. PMID: 24605946.

8.     Yu H, Li X, Duchoud F, Chuang DS, Liao JC. Augmenting the Calvin-Benson-Bassham cycle by a synthetic malyl-CoA-glycerate carbon fixation pathway. Nat Commun. 2018 May 22;9(1):2008. doi: 10.1038/s41467-018-04417-z. PMID: 29789614; PMCID: PMC5964204.

9.     Trudeau DL, Edlich-Muth C, Zarzycki J, Scheffen M, Goldsmith M, Khersonsky O, Avizemer Z, Fleishman SJ, Cotton CAR, Erb TJ, Tawfik DS, Bar-Even A. Design and in vitro realization of carbon-conserving photorespiration. Proc Natl Acad Sci U S A. 2018 Dec 4;115(49):E11455-E11464. doi: 10.1073/pnas.1812605115. Epub 2018 Nov 20. PMID: 30459276; PMCID: PMC6298115.

10.  Schwander T, Schada von Borzyskowski L, Burgener S, Cortina NS, Erb TJ. A synthetic pathway for the fixation of carbon dioxide in vitro. Science. 2016 Nov 18;354(6314):900-904. doi: 10.1126/science.aah5237. PMID: 27856910; PMCID: PMC5892708.

11.  Luo, S., Lin, P.P., Nieh, L.Y. et al. A cell-free self-replenishing CO2-fixing system. Nat Catal 5, 154–162 (2022). https://doi.org/10.1038/s41929-022-00746-x

12.  Scheffen, M., Marchal, D.G., Beneyton, T. et al. A new-to-nature carboxylation module to improve natural and synthetic CO2 fixation. Nat Catal 4, 105–115 (2021). https://doi.org/10.1038/s41929-020-00557-y

13.  Ishiura M, Kutsuna S, Aoki S, Iwasaki H, Andersson CR, Tanabe A, Golden SS, Johnson CH, Kondo T. Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria. Science. 1998 Sep 4;281(5382):1519-23. doi: 10.1126/science.281.5382.1519. PMID: 9727980.

14.  Lucius S, Theune M, Arrivault S, Hildebrandt S, Mullineaux CW, Gutekunst K, Hagemann M. CP12 fine-tunes the Calvin-Benson cycle and carbohydrate metabolism in cyanobacteria. Front Plant Sci. 2022 Oct 11;13:1028794. doi: 10.3389/fpls.2022.1028794. PMID: 36330266; PMCID: PMC9623430.

15.  Lu KJ, Chang CW, Wang CH, Chen FY, Huang IY, Huang PH, Yang CH, Wu HY, Wu WJ, Hsu KC, Ho MC, Tsai MD, Liao JC. An ATP-sensitive phosphoketolase regulates carbon fixation in cyanobacteria. Nat Metab. 2023 Jul;5(7):1111-1126. doi: 10.1038/s42255-023-00831-w. Epub 2023 Jun 22. Erratum in: Nat Metab. 2023 Oct;5(10):1836. doi: 10.1038/s42255-023-00906-8. PMID: 37349485; PMCID: PMC10365998.

16.  Chen CT, Chen FY, Bogorad IW, Wu TY, Zhang R, Lee AS, Liao JC. Synthetic methanol auxotrophy of Escherichia coli for methanol-dependent growth and production. Metab Eng. 2018 Sep;49:257-266. doi: 10.1016/j.ymben.2018.08.010. Epub 2018 Aug 30. Erratum in: Metab Eng. 2019 Jul;54:292. doi: 10.1016/j.ymben.2019.04.006. PMID: 30172686.

17.  Chen FY, Jung HW, Tsuei CY, Liao JC. Converting Escherichia coli to a Synthetic Methylotroph Growing Solely on Methanol. Cell. 2020 Aug 20;182(4):933-946.e14. doi: 10.1016/j.cell.2020.07.010. Epub 2020 Aug 10. PMID: 32780992.