Evolution and Synthetic Biology

Introduction

Evolutionary observations have often served as an inspiration for biological design. Decoding of the central dogma of life at a molecular level and understanding of cellular biochemistry have been elegantly used to engineer various synthetic biology applications including building genetic circuits in vitro and in cells^1,2, building synthetic translational systems^3–5, and metabolic engineering in cells to biosynthesize and even bio-produce complex high value molecules^6–8. Here, we review three broad areas of synthetic biology that are inspired by evolutionary observations: (i) combinatorial approaches towards cell-based biomolecular evolution, (ii) engineering interdependencies to establish microbial consortia, and (iii) synthetic immunology. In each of the areas we will highlight the evolutionary premise that was central towards designing these platforms. These are only a subset of the examples where evolution and natural phenomena directly or indirectly serve as a powerful source of inspiration in shaping synthetic biology.

Combinatorial approach towards cell-based biomolecular evolution

Combinatorial approaches for the evolution of biomolecules including catalytic RNAs, peptides, β-lactamase, and triose-phosphate isomerase, demonstrated that the principles of Darwinian evolution can be applied to evolve biomolecules by linking biomolecular diversity to selection processes^9–16. Many application-driven in vitro evolution technologies subsequently began to arise^17–19. In vitro evolution quickly became a powerful and versatile tool used in myriad ways such as in the development of de novo enzyme function^20–22, the improvement of enzyme function in harsh environments^23,24, high throughput compartmentalized catalysis^25–29, and protein engineering³⁰. Early work on in vitro evolution experiments, along with emerging insights into biomolecular evolution in natural systems (e.g., antibody evolution and maturation in adaptive immune system) and advances in molecular biology, launched a new era of biomolecular directed evolution by linking evolution to cell survival. In this section we focus on bacteria, yeast, and mammalian cell-based directed evolution platforms that use the concept of linking evolution to cell survival as an underlying principle. We will not discuss virus-based or Lenski-like continuous evolution methods that are reviewed in depth elsewhere^31–35.

One powerful example of biomolecular evolution by linking evolutionary outcomes to cell survival was the directed evolution of orthogonal tRNA/tRNA synthetase pairs in E. coli³⁶ to expand the genetic code^37–40. In these experiments tRNA/tRNA synthetase pairs were evolved to recognize a unique genetic code and incorporate a defined non-canonical amino acid at a defined position in a protein of interest. This system has been developed and utilized abundantly to recombinantly express proteins encoding non-canonical amino acids^41–45 in bacteria⁴⁶, yeast^47–51, plants^52,53, mammalian cells^54–61, Drosophila melanogaster⁶², Caenorhabditis elegans⁶³, and mice^64–68. This technology has been used for translational synthetic biology applications including antibody drug conjugates⁶⁹, other protein biologics^70,71, cell and virus-based vaccines⁷⁰, and even applied in organisms with synthetic genomes and contracted genetic codes^72,73. Another experiment linking evolution to survival was used to evolve and develop genomic base editors. D. Liu and coworkers evolved a tRNA adenosine deaminase to utilize DNA substrates when fused to dCas9, allowing for specific base pair changes at a desired genomic locus⁷⁴. Several base editors are now in preclinical and clinical trials^75,76. Other in vivo evolution experiments in E. coli where evolution is linked to cell survival utilize error-prone polymerase I⁷⁷, mutagenesis plasmids⁷⁸, random mutagenesis using mutator strains^79,80, processive protein chimera (T7 polymerase + cytidine deaminase) constructs⁸¹, CRISPR guided DNA pol⁸², retroelement-based genome editing and evolution⁸³ and DNA shuffling⁸⁴.

The principle of linking biomolecular evolution to survival has also been extended to eukaryotic cell-based directed evolution experiments. Some of the pioneering studies demonstrating the use of sophisticated synthetic biology tools for directed evolution in yeast were performed by Cornish and coworkers where they utilized reiterative recombination to sequentially assemble DNA constructs at a defined locus for the in vivo assembly of libraries of multigene pathways^85,86. Analogous to the highly error prone DNA pol I system⁷⁷ developed in E. coli, C. Liu and coworkers developed Orthorep, a continuous evolution platform which utilizes a highly error prone orthogonal DNA polymerase to evolve genes of interest⁸⁷. This system has been used for applications such as enzyme evolution⁸⁸, fragment antibody evolution⁸⁹, and optimizing cis, cis-muconic acid production⁹⁰, an important bioplastic precursor, in yeast. Several groups have also translated the processive protein chimera and CRISPR guided DNA Pol technologies developed in E. coli, to S. cerevisiae^91,92,90. Other synthetic biology tools developed for in vivo directed evolution in S. cerevisiae include utilization of retrotransposon Ty1 for continuous mutagenesis of genes and pathways⁹³ and mutagenic homologous recombination for targeted mutagenesis⁹⁴. In vivo yeast-based selection platforms have also been utilized to engineer synthetic chromosomes^95–97, as well as to study and evolve translationally important viral enzymes^89,98,99. In a recent example from our lab, Ornelas et al. developed a modular, phenotypic yeast-based complementation platform (YeRC0M) for molecular characterization and directed evolution of viral RNA capping enzyme to identify attenuation mutations in these essential viral enzymes⁹⁹.

An example of linking evolution to survival in mammalian cells naturally exists in the adaptive immune system, where B cell selection occurs through combinatorial libraries of immunoglobulins followed by selection and evolution of antigen binding cells. There are a few examples of harnessing these mechanisms to evolve antibodies¹⁰⁰ and fluorescent proteins¹⁰¹ in B cells. Evolution experiments have also been performed in mammalian cell lines to evolve ACE2 variants as decoy proteins that bind more tightly to the SARS-CoV-2 spike protein than the endogenous ACE2 receptor¹⁰². Our immune system uses similar decoy protein strategies¹⁰³. Mammalian cell-based platforms have also been developed for linking biomolecular evolution to cell survival. Few examples include studies where AID has been fused to dCas9 to evolve the target of the cancer therapeutic bortezomib, PSMB5¹⁰⁴ or to diversify genomes¹⁰⁵. Similarly, a T7 polymerase-driven continuous editing (TRACE) system was developed which utilizes an AID-T7 polymerase fusion to evolve MEK1 by linking evolution to survival¹⁰⁶ (concept from⁸¹).

Metabolic crosstalk as a basis of microbial communities

Microbial communities engage in complex behaviors that are often distinct from those in monoculture¹⁰⁷. Key features of several natural symbiotic microbial communities of the marine, gut, root and food source microenvironments are microbial crosstalk through metabolic interdependencies, communication networks and spatial organization^108,109. Synthetic biologists have leveraged this premise to build synthetic microbial communities and demonstrate their coevolution through amino acid or signaling molecule exchanges^{107,110–119}. There are several reviews on this area of research^120–122. Here we will cover a few examples of how evolutionary observations of cell-cell communications beyond amino acid or signaling molecule exchanges have been used to expand the scope of interdependencies in synthetic microbial communities.

Nature has expanded the breadth and scope of ligand-receptor pairs in higher eukaryotes through the evolution of crosstalk mediated by G protein-coupled receptors (GPCRs) and their corresponding ligands, which include proteins, oligopeptides, biogenic amines, and lipids. These molecular interactions modulate diverse cellular and physiological responses such as growth, death, migration, muscle contraction, neurotransmission, and secretion^123,124. Inspired by these observations, Cornish and co-workers identified 32 peptide-GPCR pairs which were then used to construct a language for cell-to-cell communication in yeast^125,126. Through the principle of evolution linked to survival, it is anticipated that such GPCR-peptide interfaces could be significantly expanded further by using genome mining and directed evolution approaches^127,128.

A remarkable example of bioenergetically driven microbial consortia is the evolution of modern-day eukaryotic organelles such as mitochondria and chloroplasts, which are believed to have evolved from once free-living life forms that were established as endosymbionts inside a host cell (cell-in-cell system). Extensive metabolite exchange has been observed in naturally existing endosymbiotic systems¹²⁹. A frontier of synthetic biology is to engineer such endosymbiotic chimeric life forms that are sustained and driven by complex metabolite interdependencies (we term this area of research as directed endosymbiosis). There have been several efforts since the 1930s to build cell-within-cell systems^130–142. However more recently, directed endosymbiosis has resulted in engineering cell-within-cell systems where the endosymbiont sustains the growth of the host cell under selection conditions and vice versa. Directed endosymbiosis has been established between yeast cells (host) and engineered E. coli^143,144 and cyanobacteria¹⁴⁵ (endosymbionts) to develop yeast/bacteria chimera where endosymbiotic bacteria are necessary to support the bioenergetic functions of the host cells. One such example is the engineering of artificial photosynthetic lifeforms composed of cyanobacterial endosymbionts inside of the yeast cells, where the cyanobacteria perform chloroplast like functions for the yeast cell.¹⁴⁵ Other efforts to build cell-in-cell systems include B. subtilis in macrophage^146,147 and E. coli inside HeLa cells¹⁴⁸.

Metabolite driven exchanges also play an important role in organisms like Anabaena spp. that display multicellular behavior.¹⁴⁹ Remarkably, laboratory evolution experiments have also demonstrated that multicellular behaviors can be evolved in some extant unicellular organisms. For example, of experimental evolution, selection experiments with unicellular yeast, Saccharomyces cerevisiae, have resulted in generating snowflake variants of Saccharomyces cerevisiae^150,151 that demonstrate multicellular behaviors. Similarly multicellular behaviors has also been evolved in algal strains of Chlamydomonas reinhardtii.¹⁵² In these studies, surprisingly the transition occurred in a rapid manner by subjecting organisms to conditions which favor clustered phenotypes or division of labor¹⁵⁰. Multicellular crosstalk and consortia are central to the development and functioning of animal tissues, such as in the immune system. The principles of multicellular eukaryotic cell consortia also see translational applications in the case of tissue regeneration^{117,153–155} and synthetic immunology.

Synthetic immunology

In this section, we will discuss a few examples of how our molecular understanding of the adaptive immune system can be combined with synthetic biology to develop modern-day therapeutics (synthetic immunology). We will mainly discuss examples of mammalian cell-based therapeutics, and will not cover bacteria, virus and phage based therapies which are discussed in depth elsewhere^156–158.

Previously adaptive immune system in animals has been effectively used to evolve catalytic and therapeutic antibodies¹⁵⁹. More recently, our molecular understanding of the adaptive immune system has been effectively coupled to advances in synthetic biology to engineer adaptive immune cells (B and T cells) as therapeutics. Particularly, engineered immune cells expressing chimeric receptors (CRs) expressed in mammalian cells have been widely developed as therapeutic modules. CRs are engineered protein fusions between a receptor-binding domain of a targeted protein and a domain that initiates a signaling cascade within the effector cells. CRs were initially reported for EGFR reception linking EGFR signal to insulin production¹⁶⁰ and have since been expanded for affecting signaling in various cell lines and tissues^160–163. A leading example of this approach is T cells expressing Chimeric antigen receptors (CARs)^164–167. The CAR-T cells merge the binding recognition ability of B cells with the neutralizing potential of T cells. The development and identification of the receptor binding domain of CARs targeting specific antigens has often benefited from a variety of directed evolution platforms, including display technologies^168–172. Engineered T cells expressing CARs (CAR-T cell therapy) are now used to treat blood lymphoma^173–176 as well as to combat solid tumors^177–181. In addition to this, synthetic circuits have also been engineered in T cells to instruct T cells to sequentially activate multiple cellular programs such as proliferation and antitumor activity to drive synergistic therapeutic responses¹⁸². CAR-independent gene circuits that induced IL-2 secretion in tumor tissue specifically have also been engineered. These are suggested to minimize the toxicity of the CAR T therapies¹⁸³.

Similar to T-cell based therapies, engineered B-cell-based therapies have also been tested. B cell engineering followed by their interactions within cellular consortia of adaptive immune cells has been repurposed to potentially develop novel therapeutic strategies. For example, B cell technologies have been engineered in non-human cells to develop libraries of human antibodies and select those libraries in vivo^184–187. In one example, Lin and coworkers utilized alternative RNA splicing to engineer antibody libraries by fusion of IgG segments to IgM segments that were displayed and selected for on the surface of chicken DT40 cells¹⁸⁸. Similar display technologies are undergoing engineering in human cells; for example, Moffett and coworkers developed a method to directly fuse the Ig light chain to the Ig heavy chain using a 60-a.a. linker sequence in an effort to overcome the challenge of combined antibody expression in the evolution of Fab domains¹⁸⁹. Engineered B cells expressing evolved anti-HIV antibodies have also been tested as potential therapeutics^190,191. Similarly, plasma cells have also been engineered for potential applications in enzyme replacement therapies¹⁹².

Summary and Conclusions

In this review we have highlighted how evolutionary observations have directly or indirectly shaped modern-day synthetic biology. We first discussed how the principle of linking biomolecular diversity to cell survival was used to evolve biomolecules with defined functions and properties. Next, we discussed how evolutionary observations of cell-cell communications beyond amino acid or signaling molecule exchanges have inspired the engineering of synthetic microbial communities and artificial endosymbiosis. Lastly, we discussed how the molecular understanding of the adaptive immune system can be combined with synthetic biology to engineer cell-based therapeutics. We believe that these are only a subset of the examples where evolution has inspired synthetic biology. There are a wide range of other fascinating evolutionary phenomena¹⁹³ such as transition from abiotic information storage to cells and life-forms, origin of viruses, transition from unicellular cells to multicellularity, evolution of tissues and neural networks, evolution of sexual population, evolution of cell differentiation amongst others, that can be potentially leveraged to build new synthetic biology platforms. In addition to this, we believe that the growing involvement of machine learning and modern-day modeling techniques will play a crucial role in synthetic biology advances. As always, advancements made until today will serve as steps towards the greater leaps of tomorrow and we think evolutionary observations and the lessons learned from the molecular understanding and modeling of evolution and natural phenomena will play a central role in synthetic biology of the future.

Figure 1:

Combinatorial approaches for the evolution of biomolecules in bacteria, yeast, and mammalian cells. Methods for diversifying biomolecules highlighted in this review include orthogonal tRNA/tRNA synthetase pairs, base editors, error-prone DNA polymerases, T7 RNA polymerase - cytidine deaminase fusions, DNA polymerase - CRISPR Cas9 fusions, and synthetic genomes. All methods outlined here follow the approach of mutate, screen, and select.

Figure 2:

Synthetic microbial communities based on natural observation. A) Naturally occurring microbial consortia exist in all habitats and display distinct characteristics from mono-culture organisms. These communities are largely defined by obligate metabolite exchange and demonstrate division of labor in which growth and bioproduction in one species each carry metabolic burden. B) Mating of fungal species is mediated by secretion of peptide hormones, which bind a GPCR and affect downstream processes which allow mating. Peptide/GPCR pairs are not conserved between all fungi. C) Eukaryotic cells contain metabolically essential organelles such as mitochondria and chloroplast. These organelles are thought to have evolved from endosymbiotic bacteria which propagated inside host cells through syntrophy and/or parasitism. D) The principles of naturally occurring microbial communities have been leveraged by synthetic biologists for a number of purposes. Mee et al., have engineered a 14-member consortium defined by amino acid auxotrophs. Synthetic consortia have also been used as the basis of circuits which can be used to control populations of different species in the consortium. Through the principle of division of labor, consortia can be used to more efficiently synthesize natural products and biofuels which place excessive burden on mono-culture microbes. E) Orthogonal peptide-GPCR pairs from many fungal taxa have been expressed in mutant S. cerevisiae to engineer communication networks based on cell survival. F) Model bacteria have been engineered to express genes associated with modern-day endosymbionts and organelles. These mutant bacteria were fused with mutant S. cerevisiae, where they fulfill an essential bioenergetic role from within the yeast cytosol in the manner of an organelle.

Figure 3:

Modalities of tumor elimination by cellular consortia. CAR T cells act as signaling agents in cellular consortia by means of cytokine release, which stimulates proliferation of T and B cells during the adaptive immune response. Engineered B cells are also used in treating tumors by means of their specific antibody secretion tagging of tumor tissue for elimination by neutrophils and macrophages within the cellular consortia of both the innate and adaptive immune responses. Bacteria-based therapeutics are also used as anti-cancer modalities by means of recruiting cellular consortia by antigen secretion, antigen display on their surfaces, metabolite secretion in the TME, as well as direct toxin delivery for tumor elimination. All methods directly benefit from biomolecular evolution of various receptors (i.e., CARs, antibodies).

Data availability

No data were used for the research described in this article