A warm welcome for alternative CO₂ fixation pathways in microbial biotechnology

Biological CO2 fixation is a crucial process carried out by plants and a number of microorganisms, which can be harnessed for both agriculture and sustainable, biobased production of fuels and chemicals. Fixation of CO2 by plants enables the production of food, feed, fuels and chemicals. Additionally, fixation of CO2 by autotrophic microorganisms such as cyanobacteria and microalgae can be employed for converting CO2 into value-added products, such as commodity chemicals or fuels. However, a major challenge to fully realize sustainable autotrophic production of chemicals and fuels is the low growth rate, productivity and energy conversion efficiency of autotrophs. Excluding the energy loss photoautotrophs experience due to inefficient light-harvesting (Blankenship et al., 2011), another major energy loss occurs during the fixation of CO2 in both photoautotrophs and chemolithoautotrophs (Zhu et al., 2010). The large majority of autotrophs employ the relatively inefficient Calvin–Benson–Bassham cycle for CO2 fixation (Fig. 1A). After having a dominant role in nature for billions of years, in the coming years I expect synthetic biologists will replace the Calvin cycle in biotechnology with potentially better and more efficient alternatives. But what is the limiting factor of the Calvin cycle? The main issue is the key enzyme ribulose-bisphosphate carboxylase/oxygenase (RubisCO). RubisCO has a low catalytic rate, and therefore is generally expressed in high levels, which requires a considerable amount of cellular resources. This resource demand comes on top of the existing high ATP-demand of the Calvin cycle required to produce common metabolic precursors such as acetyl-CoA and pyruvate (Claassens et al., 2016). Furthermore, RubisCO has a major wasteful side-activity with O2 in atmospheric conditions. This side-activity results in the formation of 2-phoshpoglycoylate that has to be re-assimilated to the Calvin cycle through photorespiration pathways. These photorespiration pathways partly counteract CO2 fixation by releasing some CO2 and typically add ~40–50% extra NADPH and ATP to the costs of CO2 fixation (Bar-Even et al., 2010). In recent years, many ideas have been put forward and attempts have been made to increase the efficiency of the Calvin cycle including; improving catalytic properties of RubisCO by protein engineering (e.g. Parikh et al., 2006; Dur~ ao et al., 2015), introducing more efficient photorespiration pathways (e.g. Shih et al., 2014) and introducing CO2 concentrating mechanisms to increase the carboxylating activity of RubisCO (e.g. Bonacci et al., 2012; Kamennaya et al., 2015). Some of these attempts have slightly improved the performance of autotrophs employing the Calvin cycle. However, apart from fixing such inefficiencies of the Calvin cycle, future research should also seriously address the options to completely replace both the Calvin cycle and accompanying enzyme RubisCO by potentially more efficient alternatives. Fortunately, nature has evolved many carboxylases with more promising properties than RubisCO and more attractive novel carboxylases may be created by protein engineering (Erb, 2011). Some examples of attractive natural carboxylases are employed in alternative natural carbon fixation pathways, thus far five alternative autotrophic CO2 fixation pathways have been discovered (Berg, 2011). Moreover, attractive carboxylases can be embedded in synthetic CO2 fixation pathways, which have been extensively explored by computational analyses (e.g. Bar-Even et al., 2010; Volpers et al., 2016). Several of the alternative natural and synthetic CO2 fixation pathways have lower ATP costs than the Calvin cycle, however, some trade-offs exist. For example, the natural Wood–Ljungdahl pathway and the natural reductive tricarboxylic acid cycle have very low ATP costs. However, these pathways can have major drawbacks, as these pathways are mainly limited to anaerobic settings, due to oxygen-sensitive enzymes, and they require high CO2 concentrations to be thermodynamically feasible (Berg, 2011). Alternative to natural pathways, computational analysis identified several oxygen-tolerant Received 18 October, 2016; accepted 19 October, 2016. *For correspondence. E-mail nico.claassens@wur.nl; Tel. +31(0) 317481066; Fax +31 317 483829. Microbial Biotechnology (2017) 10(1), 31–34 doi:10.1111/1751-7915.1245