Signals that communicate a cell’s metabolic state allow organisms to adapt to the suboptimal conditions they often encounter in the environment. If a particular nutrient is lacking, the organism reduces the energy it expends on growth and reproduction, instead redirecting its resources to maintain basic cellular functions.

Cyanobacteria were the driving force behind the oxygenation of Earth’s atmosphere, the ancestor of the chloroplast, and remain an important group of primary producers today. These ‘blue-green’ bacteria carry out photosynthesis, capturing solar light energy and converting it into the chemical energy that sustains life. They also perform carbon fixation, reducing carbon dioxide into organic carbon compounds. These processes are both required to sustain cyanobacterial growth, so when the inputs of these pathways – like light – are unavailable, many metabolic changes occur that necessitate physiological adaptations.

Here, I demonstrate a major mechanism by which the cyanobacterium Synechococcus elongatus senses and responds to stresses. I find that the stringent response, a conserved bacterial stress response, is induced in response to darkness, photosynthetic inhibition, and nutrient starvation. The second messengers of this response, the phosphorylated nucleotides ppGpp and pppGpp, reprogram gene expression and induce physiological responses that can be either general or nutrient-specific. I have also begun investigating the roles of polyphosphate, a compound that accumulates in response to stress and may help cells adapt to challenging conditions, in Synechococcus.

Cyanobacteria like Synechococcus are an interesting subject for study of the stringent response due to their unique metabolism and physiology. While the logic of the stringent response – indeed, many stress responses – is similar across many bacteria, the specific mechanisms behind them often differ. Learning more about how the same stress response mechanism has been adapted by diverse organisms will continue to increase our understanding of how bacteria interact with their environments.

CRISPR technology has revolutionized the way biology is conducted. In a few years, scientists have broken through barriers that have hamstrung the field for decades, thanks to the conceptually simple ability to target specific sequences in an organism’s genome. That ability alone has led to enormous leaps in our understanding of complex traits, biological pathways, disease, and evolution. Additionally, genome editing with CRISPR has ushered in a new age of therapeutics and genetic engineering. As new applications of CRISPR technology emerge, we are beginning to push the boundaries of what it is capable of, and new modalities of CRISPR are required to overcome new challenges. For example, despite its reputation of being the “flagship” CRISPR molecule, the S. pyogenes Cas9 (a.k.a. SpCas9), is still too large to be genetically encoded and delivered by adeno-associated viruses (AAVs). As AAVs are one of the predominant delivery mechanisms for gene therapies, precluding SpCas9 from its repertoire is a significant deficiency.

SpCas9 has been studied extensively to date. However, certain aspects of the molecule and its mechanisms are still unknown and evade our understanding even as we expand our gene-editing toolkit to include new Cas9s, new CRISPR systems, and new platforms. One of the questions associated with SpCas9 is its multi-domain architecture, which is atypically complex and large for bacterial proteins. While the general molecular mechanism of Cas9 is understood thanks to structural and kinetics studies, the evolution of the overall protein and its domains are less understood. This is especially interesting considering how little sequence identity Cas9s share among orthologs while still possessing a similar three-dimensional architecture.

Additionally, SpCas9’s combination of stability, precision, and versatility makes it a prime candidate for protein engineering. Researchers have tried adding new functions to SpCas9, improving its performance, and even making it more suitable for certain use cases using rational design principles. However, making the protein smaller has been an under-utilized concept with limited success. Minimizing SpCas9 while still retaining its DNA targeting function has two main functions: a) understand the essentiality of its domains for DNA targeting, and b) develop a novel protein scaffold that is smaller and more feasible for AAV and other forms of cellular delivery. In Chapter 2 of this thesis, I discuss a project to probe the amino acid deletion landscape of SpCas9, in an effort to biochemically characterize its domains, and also to arrive at a minimal DNA-binding protein module.

CRISPR-Cas9 has undoubtedly changed the biotechnology world by advancing genetic engineering and breaking through technical barriers that have plagued the field for decades. Simultaneously, the maturation of several other synthetic biology tools has also converged into a new era of biotechnology. Genome editing, sequencing, bioinformatics, gene synthesis, etc. have brought upon a revolution of new bioengineered products to the market. Leading the charge are genetically engineered microorganisms, or GEMs. GEMs are being developed for the purpose of making biofuels, commodity chemicals, materials, food and food additives, pesticides, and fertilizers. Many of these products have promise as more efficient, eco-friendly, and overall more beneficial alternatives to conventional industries.

However, adoption and deployment of these technologies are not as simple as building them. Understandably, bioengineered products require regulatory oversight to ensure safety and efficacy. In the U.S., biotechnology products are regulated by a tri-agency framework consisting of the Food and Drug Administration (FDA), U.S. Department of Agriculture (USDA), and the Environmental Protection Agency (EPA). These agencies have distinct roles and evaluation criteria when assessing new biotech products for market approval. However, one of the key issues with the current regulatory landscape in the U.S. is its complicated and circuitous nature that often delay and/or drive up costs of product development. To maximize the impact of all the innovation and benefit of scientific advances happening in the laboratory, regulation needs to be more streamlined without compromising safety.

In Chapter 3 of this thesis, I describe the current state of GEMs in the U.S., from the development and regulation perspectives. I discuss the product areas in which GEMs are emerging as feasible and scalable alternatives to conventional industrial processes, such as fuels, food, and agriculture. I also attempt to explain the current tri-agency regulatory framework as set by the FDA, USDA, and EPA. Finally, I discuss ways in which both GEM regulators and product developers must work hand in hand to enact large-scale solutions to major modern problems like climate change and food insecurity.

Photosynthesis encompasses the set of reactions that convert light into stored chemical energy. Photosynthetic organisms are the foundation of life, as they catalyze the assimilation of inorganic carbon (Ci) into the biosphere and are the ultimate source of nutrition on Earth.

Cyanobacteria are an ancient group of oxygenic photosynthetic prokaryotes and remain important primary producers today. The chloroplasts of all land plants can trace their ancestry to cyanobacteria. In many ways, such as the light reactions of photosynthesis that capture solar energy, the protein machinery of cyanobacteria is a representative, simplified form of that found in chloroplasts. In other ways, the cyanobacterial machinery is more complex as in the dark reactions that ultimately store solar energy as chemical energy by fixing CO2. To overcome inherent limitations in the principal CO2-fixing enzyme rubisco, cyanobacteria possess an elaborate integrated physiological strategy to increase the local CO2 concentration near rubisco and facilitate on-pathway catalysis. This physiological strategy, known as a CO2-concentrating mechanism (CCM), relies on energy-coupled Ci-transporters and encapsulation of rubisco in proteinaceous organelles known as carboxysomes. While some species of land plants possess CCMs, none resemble nor are nearly as productive as those found in cyanobacteria. Therefore, cyanobacteria are an ideal platform to study the basic workings of photosynthetic machinery and potentially provide insight on engineering strategies that may improve photosynthetic productivity in land plants of societal interest.

Here, we investigated the structure and function of protein complexes from cyanobacteria and other prokaryotes to elucidate their roles in photosynthesis and CCMs. First, we determined the structure of the NAD(P)H dehydrogenase-like complex, a key component in the light reactions, and develop putative models for its catalytic mechanism. Next, we discovered a new type of Ci-transporter we named the DABs accumulate bicarbonate (DAB) complex and characterize the complex through a variety of genetic, biochemical, and biophysical approaches. Finally, we determined the structure of simplified carboxysome shells that provide insight into the architecturally principles and assembly of native shell components.

A fundamental goal of protein biochemistry is to determine the sequence-function relationship, but the vastness of sequence space makes comprehensive evaluation of this landscape difficult. Advances in DNA synthesis and sequencing now allow researchers to assess the functional impact of thousands of amino acid substitutions in a single experiment, however, the quality and diversity of these mutations controls the breadth of knowledge gained by these emerging methods. Comprehensive and programmable protein mutagenesis is critical for understanding structure-function relationships and improving protein function. However, current techniques enabling comprehensive protein mutagenesis are based on PCR and require in vitro reactions involving specialized protocols and reagents. This has complicated efforts to rapidly and reliably produce desired comprehensive protein libraries. Here we demonstrate that plasmid recombineering is a simple and robust in vivo method for the generation of protein mutants for both comprehensive library generation as well as programmable targeting of sequence space. Using the fluorescent protein iLOV as a model target, we build a complete mutagenesis library and find it to be specific and comprehensive, detecting 99.8% of our intended mutations. We then develop a thermostability screen and utilize our comprehensive mutation data to rapidly construct a targeted and multiplexed library that identifies significantly improved variants, thus demonstrating rapid protein engineering in a simple protocol.

Beyond simple amino acid substitutions, protein topology is also well-established as a key mechanism by which large, complex multi-domain proteins evolve highly specialized functions. While rationally constructed protein deletions have long been essential to elucidating biochemical properties, current techniques are insufficient for a comprehensive approach. Here we develop a method for constructing fitness landscapes for even the largest and most complex proteins, comprehensively surveying functional deletions in the RNA-guided DNA binding protein dCas9, the foundation for powerful genome editing and modifying technologies. CRISPR proteins are highly complex with numerous distinct domains responsible for activities such as guide RNA binding, DNA recognition, DNA unwinding, specificity sensing and ultimately the cleavage of each DNA strand. We exploit the fitness landscape to revert functionality and step backward in domain evolution, comprehensively minimizing dCas9 and screening for an essential function. We demonstrate the power of this technique by revealing the minimal RNA guided DNA binding module at 64% of the full CRISPR-Cas9 platform, providing many new opportunities for fusions and delivery. This exploration also uncovers evidence for a DNA unwinding mechanism in a domain heretofore viewed as dispensable in Cas9. These results highlight the power of comprehensive protein deletions to clearly elucidate the boundaries of a central function.

Together, amino acid substitution and topological mutation (encompassing deletions, insertions, and circular permutations) comprise all possible genetic protein modifications. This work has served to develop simple and robust methods, which remain programmable and comprehensive, for both substitution and topological mutagenesis. The construction of high-quality protein libraries is a foundational step for applications in the fundamentals of protein biochemistry, disease prediction, and protein engineering. Ultimately, understanding the general principles of protein sequence-function landscapes - enabled by massively parallel experimentation - will allow computational methods to synergize with programmable mutagenesis and vastly improve the search for novel fitness variants.

Many bacteria employ a protein organelle, the carboxysome, to concentrate carbon dioxide

and catalyze the initial fixation reaction. Only 10 genes from Halothiobacillus

neapolitanus are sufficient for heterologous expression of carboxysomes in Escherichia coli,

opening the door to mechanistic analysis of the assembly process of this 200 MDa+

complex. One of these genes, csoS2, produces two highly repetitive intrinsically-disordered

protein isoforms and has been shown to be indispensable in carboxysome assembly.

Detailed functional characterization of csoS2, however, has been hindered by the lack of

understanding of how this single gene yields expression of two gene products. In this work,

we set out to develop a deeper understanding of CsoS2's biogenesis and its function in α-

carboxysome assembly. Using tandem mass spectrometry and biochemical assays, we have

revealed that -1 programmed ribosomal frameshifting (- 1 PRF) is responsible for the

generation of a truncated protein with C-terminus translated from the -1 frame, CsoS2A, in

addition to the full-length protein, CsoS2B. We show for the first time that CsoS2B can be

independently produced by mutations of -1 PRF elements and only CsoS2B is necessary for

the assembly of H. neapolitanus carboxysomes in native and heterologous hosts. With the

knowledge of the identity of CsoS2 isoforms, we next investigate the ability of individual

CsoS2 domains to participate in protein-protein interaction with RuBisCO, the primary

enzymatic component of the carboxysome. Here, we demonstrate that the 259-aa N-

terminal domain of CsoS2 multivalently binds RuBisCO, potentially via its short

amphiphatic helices. Finally, based on our findings, we propose a hypothetical model that

describes the formation and maturation of the α-carboxysome. This work illustrates, for

the first time, the simultaneous involvement of cotranslational regulation and an

intrinsically-disordered protein in the assembly of a prokaryotic organelle.

All life on Earth relies on biological carbon fixation, the process by which organisms convert inorganic carbon, primarily in the form of carbon dioxide (CO2), into longer chain compounds to fuel cellular processes. To enhance the efficiency of CO2 fixation, certain types of bacteria, specifically cyanobacteria and some proteobacteria, evolved specialized proteinaceous microcompartments called carboxysomes. Carboxysomes encapsulate the enzymes carbonic anhydrase and Rubisco inside a polyhedral layer of shell proteins. This molecular architecture serves to concentrate CO2 around Rubisco, allowing it to operate at its maximum catalytic rate.

Correct carboxysome assembly is essential to the survival of the organism in CO2 concentrations found in today’s atmosphere (~0.04%). In the ⍺-carboxysome lineage, the disordered scaffold protein CsoS2 links Rubisco and shell proteins, and is absolutely required for carboxysome formation and cell growth at ambient CO2 levels. This work examines how the sequence of CsoS2 scales from a disordered amino acid chain to directing the ordered self-assembly of thousands of proteins. It investigates how cells utilize specific chemistries, such as redox reactions, to assist in this assembly pathway. The result of this molecular design and coordinated construction is to build a carboxysome with a precise permeability, yet this permeability has never been measured. Results presented here address these fundamental questions.

I interrogated highly conserved and repetitive residues in CsoS2 to determine their role in carboxysome assembly. Through in vivo mutagenesis and in vitro biochemical assays I discovered that the residues VTG and Y are necessary for carboxysome assembly, and bind weakly yet multivalently to shell proteins. Conserved cysteine doublets, which hinted at a role for redox in assembly, showed no effect when mutated in vivo, but displayed biochemical phenotypes in vitro. In a major step towards reconstituting carboxysomes in vitro, I demonstrated formation of carboxysome-like phase-separated condensates with Rubisco, CsoS2, and shell, thereby showing that key carboxysome proteins can self-associate in a cell-free environment.

Once assembled correctly, the carboxysome establishes a permeability barrier and selectivity filter, allowing entry of essential metabolites such as ribulose bisphosphate and bicarbonate while restricting leakage of CO2. To measure carboxysome permeability, we developed two parallel methods, one based on a bulk plate assay and one on single-particle microscopy. Both methods utilized the redox sensitive reporter protein roGFP to simultaneously measure both the permeability of reducing agents and the internal carboxysome redox environment. Data from both approaches revealed that purified carboxysomes were permeable to the reducing agent TCEP, which reduced encapsulated roGFP over time.

Carboxysomes are the bacterial domain’s solution to the problem of capturing dilute CO2 from air and water, concentrating it, and converting it into sugars. Carboxysome functionality depends on the robust self-assembly of thousands of proteins, establishment of a specific internal chemical environment, and control over metabolite permeability. Insights from this work augment our understanding of these processes, and will aid future efforts to engineer carboxysomes into alternative organisms or cell-free systems for enhanced biological carbon capture.

Ribulose Bisphosphate Carboxylase/Oxygenase (Rubisco) is the central enzyme of the Calvin-Benson-Bassham (CBB) cycle and the most abundant enzyme on Earth. Nearly all carbon enters the biosphere via Rubisco carboxylation, yet no Rubisco has a maximum carboxylation rate above 15 /s and all Rubiscos catalyze a competing oxygenation of their five-carbon substrate. Evolution seems to have overcome this problem by selecting for CO2 concentrating mechanisms (CCMs) that place Rubisco in compartments where CO2 is highly concentrated. Distinct families of CCMs are found among bacteria, algae and plants, but they all function by elevating CO2 to promote carboxylation and inhibit oxygenation by Rubisco.

Cyanobacteria are the ancestors of all green photosynthetic lineages, including plant chloroplasts, and have a CCM that requires a large (> 200 MDa) proteinaceous organelle called the carboxysome. Carboxysomes are composed of 10,000 proteins and encapsulate 2000 Rubisco active sites. Although present-day atmosphere is rich in O2 (21%) and CO2-poor (0.04%), cyanobacteria use the CCM to grow robustly in air. Mutations to the CCM abrogate air growth and are only rescued by markedly elevated CO2. By contrast, few plant species have CCMs. Land plants with CCMs, such as maize, are particularly productive, but no plant CCMs resemble those in bacteria. These considerations raise the prospect that engineering plants to express bacterial CCMs might improve their growth. Motivated by these questions, I use complementary experimental, informatic and modeling approaches to study the evolution and function of the bacterial CCM.

I first perform a meta-analysis of the kinetic properties of 300 distinct Rubiscos and show that carboxylation and oxygenation are inextricably linked: increasing carboxylation efficiency entails an equal increase to oxygenation efficiency, suggesting that oxygenation cannot be avoided by mutating Rubisco itself. Second, by mathematical modeling I show that the 20 known components of the bacterial CCM are sufficient to concentrate CO2 and promote fast carboxylation in silico. As pH has wide-ranging effects on inorganic carbon chemistry, I also demonstrate that pH must be considered for CCM models to produce plausible results matching physiological measurements of cyanobacteria.

To confirm that the CCM is formed of a small number of genes, I designed a high-throughput genetic screen in the chemotrophic bacterium H. neapolitanus. This screen identified 20 genes in 3 operons directly involved in the H. neapolitanus CCM, including a novel class of inorganic carbon transporters I term "DABs." To test whether these genes are sufficient for CCM function, I developed a Rubisco-dependent E. coli strain, CCMB1, whose growth reports on CCM function in vivo. CCMB1 only grows in minimal media when Rubisco is expressed, and only under elevated CO2 (> 20x ambient). I found that expression of the H. neapolitanus CCM genes permits CCMB1 growth in ambient air, thereby establishing a facile system for testing the contributions of individual genes to the functioning of the bacterial CCM.

Membrane protein structure determination, as well as functional characterization, has significantly lagged behind similar investigations of soluble proteins. Membrane proteins are typically expressed at low levels, must be solubilized from the membrane in detergents, and even when purified often heterogenous. Thus, membrane proteins present many unique challenges. To overcome these challenges we have selected a model membrane protein, AqpZ, of the aquaporin family of water channels. Such a model system can be used not only to investigate the structure-function properties of this class of channels, but also to elucidate fundamental conventions which can be applied to all membrane proteins. We have expressed, purified, and solved the x-ray crystal structure of AqpZ. To determine the mechanism of water selectivity a conserved region of the channel was engineered to allow conduction of other substrates and functional analysis shows the basis to be chemical rather than steric in nature. Furthermore, aquaporins were characterized with the use of mercurial compounds, and using AqpZ we have also elucidated a steric inhibition mechanism by mercurials. From these and other studies, it became apparaent that protein expression was a fundamental barrier to working with membrane protein so we therefore set out to develop new expression systems and protocols targeted towards membrane proteins. We have adapted a reconstituted cell-free protein expression system that can express proteins at the milligram level and shown these proteins can be solubilized and purified to homogeneity. Finally, with an eye towards structural genomics we have introduced a series of protocols to streamline structure determination. As a test case the structure of the monotopic membrane protein CcmG was solved using these protocols with a minium amount of inputed work. Thus we have demonstrated, from both a single protein and a structural genomics level, a unified approach for moving towards membrane protein structure determination.

For many decades, it was thought that the phenomenon of cellular compartmentalization was exclusive to eukaryotes. Advances in cell imaging techniques have revealed that the presence of subcellular compartments is widespread throughout prokaryotes and that there exists a great diversity of organelles in bacteria and archaea. Recently, a new class of prokaryotic organelle has been discovered – the protein-bounded prokaryotic nanocompartments, also called encapsulins. Encapsulins are among the simplest of the prokaryotic organelles, often comprised of a two gene system; a shell-encoding gene and a cargo-encoding gene. Despite their simplicity, little is known about the biochemistry and physiological function of the encapsulins.

Here we investigate the structure and function of the encapsulin nanocompartments from various prokaryotes. First, we demonstrate that the existence of prokaryotic nanocompartments extends beyond the sequence homologs of the previously characterized encapsulins and that there exist additional, evolutionarily distinct families of encapsulins with unique cargo systems. We have characterized one of these new encapsulin families, which we name Family 2A, from the cyanobacteria Synechococcus elongatus PCC 7942. This encapsulin system hosts a cysteine desulfurase cargo enzyme that is directed to the interior of the nanocompartment via an N-terminal targeting sequence. We have determined the structure of the Family 2A encapsulin using cryo- electron microscopy to 2.2 Å resolution. This structure has yielded insights into the cargo- binding site within the organelle and the potential size and charge selectivity of the compartment. Additionally, we have characterized the cysteine desulfurase activity of the Family 2A encapsulin and show an increase in cargo enzyme activity upon encapsulation.

Finally, we examine the biochemistry and physiological role of a previously discovered encapsulin from Mycobacterium tuberculosis (Mtb). We demonstrate that the Mtb encapsulin is stable and active under the harsh conditions of the phagolysosome – the ecological niche we implicate the Mtb encapsulin is localized. Furthermore, we show Mtb encapsulin plays a functional role in Mtb’s resistance to oxidative stress.

Taken together, our work provides critical insight as to how the structure of the prokaryotic nanocompartments informs their function in vivo and show that this emerging class of protein-bounded organelle represents a new paradigm within the field of prokaryotic cell biology.

The sinking of the Titanic in April 1912 took the lives of 68 percent of the people aboard. Who survived? It was women and children who had a higher probability of being saved, not men. Likewise, people traveling in first class had a better chance of survival than those in second and third class. British passengers were more likely to perish than members of other nations. This extreme event represents a rare case of a well-documented life and death situation where social norms were enforced. This paper shows that economic analysis can account for human behavior in such situations.