Autotrophic carbon assimilation is the most important biosynthetic process in nature, which provides the primary building blocks for the construction of living organisms and produces the energy and materials essential for the development of human society (Berg, 2011). In recent years, there has been a growing interest in converting heterotrophic organisms into facultative autotrophs, which is expected to have a positive impact on the Earth's ecosystem. To date, only seven, or eight when incorporating crassulacean acid metabolism (CAM), natural carbon assimilation pathways, formed by spatiotemporal regulation, have been validated through either directly enzymatic and phenotypic tests or genomic analysis followed by experimental tests (Wurtzel et al., 2019). In 2010, Bar-Even et al. presented a pioneering framework for the systematic design of carbon assimilation pathways, outlining fundamental design principles and evaluation criteria for the artificial pathways (Bar-Even et al., 2010).

In 2016, a systematic approach was employed to design novel CO2 assimilation pathways by effectively combining natural carboxylases. This strategy resulted in the development of the in vitro validated crotonyl-CoA/ethylmalonyl-CoA/hydroxybutyryl-CoA (CETCH) cycle (Schwander et al., 2016), using enzymes from different species, which was further optimized (Miller et al., 2020) and adapted with other modules in subsequent studies (Scheffen et al., 2021). In 2018, Trudeau and colleagues addressed the challenge of recovering by-product glycolate (C2) caused by oxygenase activity of RuBisCO in the Calvin cycle (Trudeau et al., 2018). They designed a series of glycolate recovery modules, including the arabinose 5-phosphate (Ara5P) shunt and the tartronyl-CoA (TaCo) pathway, with 100% and 150% theoretical carbon yield, respectively. The Ara5P shunt was enabled by harnessing the pan-substrate activity of the natural aldolase. Subsequent studies successfully identified or rationally constructed the three key enzymes needed to implement the TaCo pathway (Scheffen et al., 2021). In the same year, Bar-Even advocated bolder pathway design rather than just picking the “low hanging fruit”, and published a series of novel carbon fixation modules, including the groundbreaking idea of using reverse 6-phosphogluconate dehydrogenase (Gnd) for carbon fixation (Bar-Even, 2018). Subsequently, the viability of this concept was validated through the implementation of the Gnd–Entner–Doudoroff (GED) cycle (Satanowski et al., 2020).

Artificial carbon assimilation pathways have progressed from system design to practice and application thanks to the increased efficiency of multi-omics sequencing, as well as the increasing abundance of database information, computational tools, and gene editing methods. This has enabled scientists to make great strides in utilizing reactions and enzymes, while also reducing the number of the design-build-test-learn cycle iteration. Through the course of research, our comprehension of pathways has been enhanced, leading to the iterative application and improvement of general screening principles and evaluation methods based on pathway structural characteristics, thermodynamic driving forces, enzyme kinetic parameters, pathway yields, and energy consumption levels. In summary, since 2010, stemming from the visionary advocacy of developing artificial carbon assimilation pathways by Bar-Even et al. to the publication of numerous innovative cases, artificial carbon assimilation has undergone a transformative trajectory. It has evolved from an emerging field to a central focus in research, carrying the significant responsibility of increased societal expectations.

The established natural CO2 assimilation pathway consists of core modules and enzyme components that not only facilitate the actualization of modified natural pathways, but also hold potential for the exploration and development of entirely new artificial pathways. At the same time, natural and artificial carbon assimilation pathways and their migration, modification, and application in different species have facilitated the construction of artificial autotrophic strains. In this review, we adopt a point-to-face approach to summarize the research context in the field of artificial carbon assimilation in an easily understandable way. Specifically, we achieve this by: (1) demonstrating the potential of breeding artificial pathways from natural pathways by examining the overlap and relationships between the natural reductive TCA cycle and its three sub-pathways; (2) showcasing the versatility and significant impact of efficient modules by reviewing a decade of research on the non-oxidative glycolytic pathway, from its proposal to its application; (3) highlighting the important value of natural enzyme libraries in the design and realization of artificial pathways by laying the foundation for the discovery and development of new enzyme functions; and (4) summarizing the research progress and development directions in the field by compiling studies on artificial autotrophic systems. Ultimately, our review aims to provide a comprehensive overview of the design and application of artificial carbon assimilation pathways, and to make a contribution towards the development and application of artificial carbon fixation pathways and non-natural autotrophic systems.