ADVANCING PHOTOSYNTHETICALLY DRIVEN MICROBIAL CO-COCULTURES FOR SUSTAINABLE BIOPRODUCTION By Lisa Yun A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Biochemistry and Molecular Biology – Doctor of Philosophy 2024 ABSTRACT As carbon emissions reach record levels annually, the allure of cyanobacterial biotechnology becomes increasingly desirable. The technology to receive a variety of renewable chemicals in exchange for CO2 and light is available, but many advancements are required before it can be commercially applicable. The focus of this dissertation is to improve the potential of cyanobacterial co-culture systems for photosynthetic bioproduction. The co-culture approach offers a degree of flexibility in output products that is difficult to match using other metabolic engineering approaches, and the overall sucrose production from cyanobacteria is promising relative to other carbohydrate feedstocks. Nonetheless, at this time, any attempts to scale cyanobacterial co-culture to industrially relevant volumes are likely to encounter various challenges. My dissertation research focused on two potential weaknesses of cyanobacterial co- culture. In Chapter 2, I describe my research to maximize the output of sucrose by modulating the cultivation conditions for sucrose-secreting strains of Synechococcus elongatus PCC 7942. Maximization of heterotrophic yield in co-culture is directly linked to the amount of fixed carbon that can be supplied by the cyanobacterial partner, and I interrogated a number of different illumination conditions and methods for providing inorganic CO2 to cyanobacterial strains. In Chapter 3, I describe my efforts to rationally design co-cultures for higher resilience and productivity through engineering in a system for physical attachment between the heterotroph and cyanobacterial partner. Specifically, I expanded upon prior work on a cyanobacterial surface display system with the long-range goal of installing extracellular binding domains on the surface of S. elongatus that could be used to mediate direct physical attachments with other proteins displayed on the surface of an engineered heterotrophic partner. Finally, in Chapter 4, I review the outcomes and conclusions of my research program and reflect upon both my experience and future prospects for the topic areas central to my dissertation. ACKNOWLEDGEMENTS I would like to take this opportunity to express my appreciation and gratitude to everyone who has supported me up to and throughout my PhD. There are too many to name individually, and for that, I am blessed. First and foremost, I would like to thank my advisor, Dr. Daniel C. Ducat, who has been unfailingly supportive academically and personally. Your compassion has been the foundation of my PhD experience, which I could not have weathered the storms of without. Your patience has not gone without notice, and I am deeply grateful for it. Thank you for guiding me through the twists and turns of my research, and helping me see the possibilities where I could not before. I am forever grateful for your mentorship and belief in me. The members of the Ducat Lab have been a joy to work with. I thank postdoctoral researchers Dr. María Santos-Merino and Dr. Jonathan Sakkos for sharing their wealth of knowledge with me. Dr. Sreeahila Retnadhas, Dr. Amit Singh, and Dr. Joshua MacCready have answered my questions, large and small. Graduate students Rees Rillema, M.S. and Emmanuel Kokarakis, M.S. have been comrades-in-arms; there is nothing like the solidarity that comes from working together in the odd hours of the day. Heartfelt appreciation is extended to my committee members: Dr. Robert Hausinger, Dr. Michaela TerAvest, Dr. Gregory Bonito, and Dr. Beronda Montgomery. Your guidance and support have been instrumental in my academic growth, and I deeply value your contributions to my success. Special thanks to Dr. TerAvest for introducing me to Michigan State University during my undergraduate years. I am grateful to the departments and programs at the University that have fostered my academic and personal development. In the Department of Biochemistry and Molecular Biology, Dr. David Arnosti fosters an environment of safety and support for all graduate students. I am grateful for the opportunity you provided me to begin an outreach program with the Greer Outreach Center; it was a learning and healing experience to introduce young minds to the joys of science. Your compassion and curiosity remind me to keep mine alive. The Plant Biotechnology for Health and Sustainability training program, helmed by Dr. Robert Last and Dr. Jyothi Kumar, afforded me the opportunity to undertake an industrial internship. Jessica Lawrence, who supports both the department and program, helped me navigate the intricacies of the University. There are many to thank within the collaborative environment of the Plant Research Laboratory as well. The iii leadership and staff that kept us steadfast during the uncertainty of the pandemic, the camaraderie of my friends and colleagues in other labs kept me going, and the invisible work performed by Linda Danhof, Melissa Borrusch, Heather Sharick, Adam Goetschy, and Jacob Hicks kept the place running smoothly. I am very lucky to have worked here. Graduate school is not for the faint of heart, and I am forever grateful for those who have supported me personally through the highs and lows of the journey. To the many graduate students who I have befriended in my time: it has been an honor to have walked through the crucible with you. Deanna, I value your perspective; thank you for helping keep my head on straight. Diego, I am deeply grateful for your companionship, and I look forward to its continuance. Hannah, Bianca, Aiko, and Emily – thank you for your friendship and support; some hard things can be made easier with good company. I sincerely wish you all the best in your individual pursuits. I would also like to thank those who have supported me before and during graduate school. Ron – you began my scientific career, and I could not be where I am today without you. I am forever indebted to your wisdom, kindness, and mentorship; thank you for believing me. Alec, you are my biggest hype man. Thank you for your love and energy; it will always inspire me. Frankie, Virginia, Nicole, Angela, and Erick – thank you for being present in my times of need and for your loving support. Mr. McCullough, thank you for providing a space to explore my academic aptitude; your classes were the framework of critical thinking and writing that so much of my life today rests upon. You encouraged me to reach beyond what I thought was possible – of what I thought I was capable – and I would not have pursued a PhD or met the wonderful people in my journey without you igniting the first spark. Finally, I would like to acknowledge my families, found and inherited. I am grateful to my parents for raising me with the qualities of hard work and perseverance, which have served as a foundation in my every endeavor. I give boundless thanks to my partner, Cort, for support and solidarity in the pursuits of love, life, and one of our biggest achievements thus far. And finally, to Cassy, the kitten I adopted a month before the pandemic who has welcomed me home after every long experiment and has kept me company throughout every all-nighter. iv TABLE OF CONTENTS CHAPTER 1: INTRODUCTION TO SUCROSE IN CYANOBACTERIA .................................1 REFERENCES ................................................................................................................. 25 CHAPTER 2: ENHANCEMENT OF CYANOBACTERIAL SUCROSE BIOSYNTHESIS THROUGH CULTIVATION METHODS................................................................................... 36 REFERENCES ................................................................................................................. 52 APPENDIX ....................................................................................................................... 56 CHAPTER 3: UTILIZATION OF ENDOGENOUS AND HETEROLOGOUS OUTER- MEMBRANE PROTEINS FOR CYANOBACTERIAL SURFACE DISPLAY ........................ 61 REFERENCES ................................................................................................................. 74 APPENDIX ....................................................................................................................... 76 CHAPTER 4: CONCLUDING REMARKS AND FUTURE DIRECTIONS ............................. 79 REFERENCES ................................................................................................................. 87 v CHAPTER 1: INTRODUCTION TO SUCROSE IN CYANOBACTERIA This chapter was adapted from text originally published in: Santos-Merino, M., Yun, L. and Ducat, D.C. (2023) Cyanobacteria as cell factories for the photosynthetic production of sucrose. Front. Microbiol, 14, p.1126032. Copyright, © 2024 Frontiers Media — Reproduced with permission. 1 Sucrose in cyanobacteria Cyanobacteria are ancient photosynthetic prokaryotes that date back as far as 3.5 billion years ago, and have since significantly diversified, enabling them to colonize terrestrial, aquatic, and benthic habitats (Gaysina et al., 2019). Through the development of mechanisms to withstand harsh conditions including UV rays, salt concentrations, temperatures, and irradiances, cyanobacteria have not only survived but thrived in the face of the unique challenges of each habitat, becoming arguably one of the most important and abundant taxa in existence (Rosic, 2021; Crockford et al., 2023). Their remarkable success led to the Great Oxidation Event approximately 2.5 billion years ago, causing a rapid and significant rise in atmospheric oxygen levels (Crockford et al., 2023). This shift from a reduced to an oxidized state transformed biogeochemical cycles, facilitating in the evolution of complex oxygen-dependent life (Pathak et al., 2021). Apart from adapting to ecological niches, cyanobacteria have also forged various symbiotic relationships with prokaryotes and eukaryotes. By providing essential services like carbon and nitrogen fixation, cyanobacteria are key players with associated and neighboring organisms, even becoming so integral that they are the origin of chloroplasts (Falcón et al., 2010). The advantageous combination of oxygenic photosynthesis and ecological, biological, bioproductive, and symbiotic flexibility position cyanobacteria as attractive candidates for diverse biotechnological applications. Efforts are underway to enhance the synthesis of primary and secondary metabolites (e.g., carbohydrates, fatty acids, amino acids, antioxidants, phycobiliproteins, bioplastics, biofuels), and to explore their use as a carbon source for bioproductive heterotrophs in engineered microbial communities (Zhang et al., 2021; Bakku and Rakwal, 2022; Kariyazono et al., 2022). Moreover, there is interest in employing cyanobacteria for tasks like wastewater treatment, carbon capture and sequestration, enhancing agricultural output, and more (Kroumov et al., 2021; Singh et al., 2021; Bakku and Rakwal, 2022). Despite their ancestral lineage, cyanobacteria remain a subject of ongoing study as they continue to contribute to the planetary ecosystem and a growing circular economy. Biological applications of sucrose Like in all photosynthetic organisms, sucrose [α-d-glucopyranosyl (1→2) β-d- fructofuranoside] metabolism is an integral component connected to central carbon metabolism in cyanobacteria. Sucrose is utilized by a number of cyanobacterial species as a protective metabolite, or so called “compatible solute,” that can be hyperaccumulated in the cell without ill effect. Unlike 2 other compatible solutes like glucoglycerol (GG) or glycine betaine (GB), sucrose is believed to have direct interactions with macromolecules (Hagemann, 2011). The arrangement of hydroxyl groups allow disaccharides like sucrose and trehalose to interact with the water-soluble portion of glycerolipids, enabling them to replace some water in the membrane hydration shells without altering the water structure, as observed with other compatible solutes (Potts et al., 2005; Hagemann, 2011). It is hypothesized that sucrose was the first compatible solute to emerge in cyanobacteria; the capacity for sucrose biosynthesis appears to be widespread, as shown by the presence of at least one sucrose-synthesizing enzyme found in nearly all known genome sequences (Blank, 2013). Studies on numerous cyanobacterial strains have revealed at least 60 species accumulate sucrose under high-salt conditions, with a consistent pattern of strains that accumulate sucrose as the primary compatible solute to have relatively low salt tolerances (e.g., Synechoccocus elongatus PCC 7942, Microcystis aeruginosa PCC 7806, etc) (Hagemann, 2011). While sucrose is considered to be somewhat weak as an osmoprotectant, conferring protection up to only 50-100% that of seawater salinity in such strains, it has been found in minor or transient portions of the total compatible solute pool in halotolerant species that accumulate other solutes as primary osmoprotectants (Reed and Stewart, 1985; Hagemann, 2011). In these cases, sucrose can be utilized as a secondary osmoprotectant to enhance salt tolerance, and is employed in combating general stresses such as desiccation, heat, and cold. In addition to its accumulation in the cytoplasm for osmotic stress, sucrose also concentrates in the extracellular matrix during drought conditions to create a protective glass-like shell to reduce water loss and molecular diffusion in preparation for dormancy (Waditee-Sirisattha and Kageyama, 2022). Elevated temperatures induce sucrose biosynthesis, likely due to its superior ability to bolster membrane stability at higher temperatures compared to GG or GB (Hincha and Hagemann, 2004). Sucrose biosynthesis is crucial for cyanobacterial acclimation to environmental stresses, but its biological utility extends beyond the producing organism. As a sugar, sucrose is a versatile carbon source for cells that cannot produce their own. For example, Anabaena sp. PCC 7120, a diazotrophic filamentous cyanobacterium partitions photosynthetic vegetative cells apart from heterocysts performing oxygen-sensitive nitrogen fixation, all while facilitating carbon and nitrogen exchange between the two cell types (Herrero et al., 2016). Cyanobacteria also supply sucrose and other sugars to heterotrophs, fostering 3 multitrophic associations. A notable example are cyanolichens, which are composite organisms consisting of obligately symbiotic fungi and cyanobacteria (Rikkinen, 2015). While each lichen is unique in species composition, generally, the cyanobacteria provide fixed carbon and/or nitrogen to the fungi in exchange for minerals, nutrients, secondary metabolites, CO2, and protection from the external environment (Rikkinen, 2017). An engineered lichen consisting of a sucrose-secreting S. elongatus PCC 7942 co-cultured with either Rhodotorula glutinis or Cryptococcus curvatus exhibited enhanced lipid yields from cyanobacteria compared to when they were grown axenically, indicating that the benefits of co-cultivation exceed the nutritional value of mutualistic exchange (Li et al., 2017). Likewise, symbiotic relationships with plants, mosses, diatoms, and prokaryotes have been identified, further highlighting the adaptable and advantageous nature of cyanobacterial primary production (Rai et al., 2002). Enzymes responsible for sucrose biosynthesis The primary enzymes involved in sucrose biosynthesis are sucrose phosphate synthase (SPS) and sucrose phosphate phosphatase (SPP). Briefly, the glucosyltransferase domain (GTD) of SPS transfers a glycosyl group from NDP-glucose to fructose-6-phosphate to produce sucrose- 6-phosphate; SPP then hydrolyzes the intermediate product with the phosphohydrolyase domain (PHD) to create sucrose. Cyanobacterial SPS is the rate-limiting enzyme in sucrose biosynthesis; it most often accepts UDP-glucose, but can also utilize GDP-, TDP-, and ADP-glucose, a substrate flexibility not observed in plant SPS proteins (Porchia and Salerno, 1996; Curatti et al., 1998). Additionally, there are two types of cyanobacterial SPS proteins: one which contains only the SPS domain, and another that is fused with a SPP domain (Santos-Merino et al., 2023). The SPS-SPP fusion proteins are similar to those found in plants, but cyanobacterial SPP domains often exhibit little or no SPP activity, raising questions regarding their function in such instances (Lunn et al., 2000). It is hypothesized that the SPP domain may bind to newly formed sucrose-6-phosphate from SPS for metabolite channeling (Fieulaine et al., 2005) or function as a regulator (Albi et al., 2016). Unidomainal SPS is the most prevalent form, and SPP is encoded separately to perform the reversible dephosphorylation of sucrose-6-phosphate (Santos-Merino et al., 2023). Despite sharing little sequence similarity, these proteins exhibit three conserved motifs in the active site of the phosphohydrolase domain, which are structurally consistent across photosynthetic organisms (Fieulaine et al., 2005). However, some cyanobacterial strains code for the bidomainal SPS and no separate SPP; in these cases, SPP-like proteins, which do not contain the motifs that define SPP, 4 may be utilized instead (Santos-Merino et al., 2023). Interestingly, they are also they are present in strains with SPP, prompting inquiries to their role in such cases (Santos-Merino et al., 2023). There is significantly less research regarding the function of cyanobacterial SPP-like proteins relative to SPP, and their functions remain unknown. Cyanobacterial sucrose catabolism pathways While sucrose is required to acclimate to environmental stresses, it must be degraded once it is no longer required to mitigate carbon and energy loss (Baran et al., 2017). Invertases are a large and diverse group of sucrose-cleaving enzymes, and are the most broadly encoded pathway for cyanobacterial sucrose degradation in which this enzyme performs the irreversible hydrolysis of sucrose into glucose and fructose. In some cases, such as in Synechocystis sp. PCC 6803, invertases are the only enzyme responsible for in vivo degradation of sucrose (Kirsch et al., 2018). Only alkaline/neutral invertases (A/N-invertases; preferring pH of 6.5 to 8) are found in cyanobacteria, and they have a high specificity for the α-1,2-glycosidic linkage of sucrose (Vargas et al., 2003; Vargas and Salerno, 2010; Xie et al., 2016; Wan et al., 2018; Liang et al., 2020). In the filamentous Anabaena sp. PCC 7120, A/N-invertases are encoded by the genes invA and invB (Page-Sharp et al., 1999; Vargas et al., 2003). InvA regulates the carbon flux from vegetative cells to heterocysts (López-Igual et al., 2010), whereas InvB coordinates sucrose and glycogen metabolism exclusively in heterocysts (Vargas et al., 2011). Sucrose synthase (SuS) is a glucosyltransferase that performs the reversible hydrolysis of sucrose to U/ADP-glucose and fructose, though it is believed to primarily degrade sucrose in vivo (Porchia and Salerno, 1996; Curatti et al., 2002). SuS is not ubiquitous among cyanobacteria, and is most often found in heterocyst-forming strains, though there have been reports of its presence in unicellular cyanobacteria (Koleman et al., 2012; Hagemann, 2013). It appears to play a vital role in nitrogen fixation by controlling the flux of carbon from vegetative cells and possibly producing precursors for cell wall synthesis (Porchia et al., 1999; Curatti et al., 2000, 2002, 2006, 2008; Cumino et al., 2007). In addition to regulating sucrose degradation, SuS may also be involved in recycling sucrose into glycogen as it as it generates U/ADP-glucose that can be reverted to glucose-1-phosphate by UDP-glucose pyrophosphorylase (UGP), a building block of glycogen (Cumino et al., 2002, 2007). The rarest sucrose-degrading enzyme in cyanobacteria is amylosucrase, a glucosyltransferase that hydrolyzes the glycosidic bond to produce glucose and fructose; the 5 resulting glucose is then used to form maltooligosaccharides (Potocki de Montalk et al., 2000). It was first discovered in Synechococcus sp. PCC 7002 in the same transcriptional unit with sps, spp, and fructokinase (Perez-Cenci and Salerno, 2014). Engineering cyanobacteria to produce sucrose Innovations in biotechnology have taken advantage of aquatic photosynthetic organisms’ ability to create valuable products (e.g., lipids, antioxidants, pigments) to cope with environmental stressors (Chen et al., 2017; Morone et al., 2019). As a bioproduct naturally synthesized at high levels by some species of cyanobacteria in response to salt stress, sucrose has garnered attention for its potential as an alternative carbohydrate feedstock for higher-value goods (Hays and Ducat, 2015; Zhang et al., 2021). Sucrose generated by cyanobacteria could offer a number of advantages relative to plant-based feedstock crops, including potentially higher photosynthetic efficiencies and reduced requirements for potable water or arable land. Here, I review recent strategies employed to make cyanobacterial bioproduction of sucrose more productive and affordable. Engineered heterologous transporters for sucrose export As discussed above, cyanobacteria can accumulate osmoprotectants up to hundreds-of- millimolar concentrations when exposed to hypersaline conditions (e.g., sucrose, trehalose, GG) (Hagemann, 2011; Klähn and Hagemann, 2011). For instance, under moderate salt stress (200 mM NaCl), the common freshwater model cyanobacterium S. elongatus PCC 7942 accumulates nearly 300 mM intracellular sucrose, representing a significant portion of the cell biomass (Suzuki et al., 2010). Although this degree of metabolite production presents an industrial and agricultural opportunity, cytosolic volume constrains how much sucrose can be accumulated: the costs associated with cyanobacterial cell recovery, lysis, and processing would likely exceed the economic value of the commodity products contained in the cytosol (Prabha et al., 2022). Therefore, secreting sugars into the supernatant for collection has been proposed as a more financially viable strategy. For this purpose, cyanobacteria have been engineered to express heterologous transporters capable of exporting lactate and hexoses (Niederholtmeyer et al., 2010; Angermayr et al., 2012). Similarly, S. elongatus PCC 7942 was originally engineered to export sucrose by expressing sucrose permease (cscB) from Escherichia coli ATCC 700927 (Ducat et al., 2012), and multiple cyanobacterial species have since been similarly modified by different research teams (Table 1.1). In its native host, CscB is a sucrose/proton symporter that typically operates by 6 utilizing the free energy of the proton gradient to import both molecules (Vadyvaloo et al., 2006). By contrast, during periods of cyanobacterial sucrose synthesis, internal sucrose concentrations greatly exceed external levels causing reversal of chemical gradients and driving sucrose export through CscB instead. CscB-expressing, sucrose-exporting S. elongatus PCC 7942 strains can secrete up to 80% of photosynthetically fixed carbon as sucrose, diverting these resources away from the accumulation of cellular biomass (Ducat et al., 2012). Although efforts to scale-up cyanobacterial sucrose production have not yet come to fruition (e.g., Proterro; Aikens and Turner, 2013), it has been estimated that such cyanobacterial strains have the potential to produce comparable amounts of sugar to traditional plant-based carbohydrate feedstocks. Realizing the promise of cyanobacterial sucrose is likely to require efforts to address problems of cyanobacterial/microalgal cultivation (beyond the scope of my dissertation, but see (Su et al., 2017; Khan et al., 2018) as well as strategies to maximize bioproduction rates. Increasing metabolic flux to sucrose pathways Published strategies for improving rates of cyanobacterial sucrose productivity generally fall into two related strategies: increasing carbon flux towards the synthesis of sucrose through the upregulation of relevant biosynthetic activities, or by reducing the loss of carbon to competing pathways or sucrose reuptake. Perhaps the most straightforward approach for improving sucrose titers has been the overexpression of genes related to sucrose biosynthesis. Several studies have now found that flux leading to sucrose production can be most impacted by increasing the activity of SPS (Du et al., 2013; Duan et al., 2016; Lin et al., 2020a), which is largely intuitive given that this enzyme catalyzes a commitment step to sucrose biosynthesis. Significant increases in sucrose production can be found in strains overexpressing SPS, even without allowing for sucrose export. First reported in Synechocystis sp. PCC 6803, a strain engineered to overexpress its native SPS (SPS6803) accumulated nearly twice as much intracellular sucrose than its wild-type counterpart (Du et al., 2013). Likewise, when the native SPS in S. elongatus PCC 7942 was overexpressed, internal sucrose concentrations were 93% higher than in wild-type (Duan et al., 2016). In addition, pairing SPS overexpression with sucrose export further increases total sucrose yields. When SPS7942 and CscB were co-overexpressed in S. elongatus PCC 7942, there was a 74% increase in sucrose compared to the CscB-only strain (Table 1.1) (Duan et al., 2016), yet the nature of the SPS homolog that is overexpressed can strongly influence the degree to which sucrose production is improved. SPS7942 is bidomainal and bifunctional (i.e., possessing active GTD and PHD domains), 7 in contrast to SPS6803 which is also bidomainal, but has a non-functional PHD domain and is regulated distinctly from SPS7942 (Curatti et al., 1998; Lunn et al., 1999; Gibson et al., 2002). However, the partial functionality of SPS6803 does not mean it is less effective, as heterologous co- overexpression of SPS6803 and CscB in S. elongatus PCC 7942 increases sucrose production relative to overexpression of the native SPS7942 (Abramson et al., 2016; Dan et al., 2022) (Table 1.1). It is curious that SPS6803 is a more effective enzyme for rerouting carbon flux towards sucrose bioproduction, given that it lacks a functional SPP domain (S. elongatus PCC 7942 encodes other endogenous SPP proteins in the examples above), so it is possible that this observation is related either to the manner in which salt-ions can regulate the function of some SPS domains (Liang et al., 2020), or to other unknown functions for SPP and/or SPP-like domains other than sucrose-6- phosphate phosphatase activity. While the overexpression of SPS has yielded substantial improvements, this strategy has not been equally successful with other proteins in the sucrose biosynthetic pathway. Overexpression of SPP from Synechocystis sp. PCC 6803 (SPP6803) either had no effect on, or decreased sucrose productivity in S. elongatus PCC 7942 or Synechococcus elongatus UTEX 2973 (Du et al., 2013; Lin et al., 2020a). Similarly, overexpression of UGP, the protein producing UDP- glucose as a substrate for SPS, led to less sucrose secretion (Ducat et al., 2012; Du et al., 2013). Only when these three enzymes were overexpressed simultaneously (i.e., SPS6803, SPP6803, and UGP) were sucrose levels increased in comparison with SPS6803-only strain, albeit marginally (Du et al., 2013) (Table 1.1). Another successful approach for improving the flux of carbon to sucrose biosynthesis is to accelerate the rate of the upstream carbon supply from the Calvin Benson cycle. Multiple strains of cyanobacteria have been engineered to secrete sucrose through the heterologous expression of cscB, but the highest yields to-date have been obtained from strains with a more rapid metabolism and higher light tolerance relative to classic laboratory models (Table 1.1). S. elongatus UTEX 2973 is a recently re-characterized species that is 99.99% identical to S. elongatus PCC 7942, but has a doubling time of ~2 h (compared to ~5–9 h for S. elongatus PCC 7942), and is more tolerant of high-light and high temperature conditions (Kratz and Meyers, 1955; Yu et al., 2015; Adomako et al., 2022). Expression of cscB in S. elongatus UTEX 2973 led to the development of strains with relatively high sucrose productivities (Song et al., 2016; Lin et al., 2020a). A high sucrose titer was originally reported in such strains when exposed to 150 mM NaCl, reaching approximately 8 80 mg L-1 (Song et al., 2016). Lin et al. also created a S. elongatus UTEX 2973-cscB strain, and observed an even greater sucrose titer at 8 g L-1 at 150 mM NaCl, averaging out to 1.9 g L-1 day-1, over 2-fold higher than the productivities of S. elongatus PCC 7942, representing the highest sucrose titer published thus far (Lin et al., 2020a), and illustrating the potential benefits of utilizing fast-growing strains that can reach higher densities. Somewhat surprisingly, activation of the sucrose export pathway itself has been reported to increase the overall photosynthetic flux in some cyanobacterial strains. In S. elongatus PCC 7942, when sucrose synthesis pathways are placed under inducible promoters, a variety of enhancements in features related to photosynthesis have been reported in the hours following activation of the pathway (Ducat et al., 2012; Abramson et al., 2016; Santos-Merino et al., 2021; Singh et al., 2022). The quantum efficiency of photosystem II, rate of oxygen evolution, relative rate of electron flux through the photosynthetic electron transport chain, oxidation status of the photosystem, and rate of carbon fixation are all increased (Ducat et al., 2012; Abramson et al., 2016; Santos-Merino et al., 2021). The latter observation is correlated with an increase in Rubisco abundance that was revealed by proteomic analysis >24 h following induction of sucrose export, and a concomitant increase in carboxysome number (Singh et al., 2022). While the mechanisms underlying these changes in photosynthetic performance are not well understood, it has been hypothesized that they arise from a relaxation in “sink limitations” on photosynthesis that can occur when the downstream consumption of products of photosynthesis (e.g., ATP, NADPH, Calvin Benson cycle outputs) is insufficient to keep up with the supply (Santos‐Merino et al., 2021). Stated differently, when carbon fixation is not the rate-limiting step of cell metabolism (e.g., under enriched CO2 atmospheres commonly used in laboratory conditions), the expression of a heterologous pathway may act as an additional “sink” and bypass downstream limitations of cell growth and division. While this remains a speculative possibility, the relaxation of acceptor-side limitations on photosystem I suggests that sucrose secretion pathways (or other heterologous metabolic sinks) may utilize “excess” light energy that might otherwise be lost to photosynthetic inefficiencies under certain conditions (Abramson et al., 2016; Santos-Merino et al., 2021). Uncovering the mechanisms underlying the photosynthetic phenotypes coupled to sucrose export might allow even greater improvements in photosynthesis and/or sucrose bioproduction. Reducing metabolic flux to competing pathways The alternative strategy to boost sucrose production is to improve the pool of sucrose or 9 sucrose precursors by reducing flux to pathways that compete with sucrose biosynthesis for either substrates or total carbon pools. A straightforward example is to eliminate the dominant route for sucrose breakdown, such as the Inv proteins that are a dominant route of sucrose hydrolysis in many cyanobacterial models. In a recent report, inactivation of the Synechocystis sp. PCC 6803 invertase increased accumulated sucrose by 10-fold in both salt and salt-free conditions (Kirsch et al., 2018). These results were of higher magnitude, but similar trajectory to reports in other cyanobacteria, such as in sucrose-exporting S. elongatus PCC 7942 where a ΔinvA background exhibited a 15% increase in extracellular sucrose (Ducat et al., 2012). Glycogen is a storage molecule of cyanobacteria that is a significant alternative carbon sink, yet inhibiting glycogen synthesis has yielded variable results on sucrose secretion. For example, knockout of the two glycogen synthase genes (glgA-I and glgA-II) of Synechococcus sp. PCC 7002 led to an accumulation of three times more sucrose than wild-type under hypersaline conditions (Xu et al., 2013) (Table 1.1). However, when another glycogen synthesis gene, ADP- glucose pyrophosphorylase (glgC), was downregulated in sucrose-secreting S. elongatus PCC 7942, there was only a minor or insignificant increase in sucrose (Qiao et al., 2018). GlgP is responsible for hydrolyzing glycosidic bonds in glycogen to release glucose-1-phosphate, so it was theorized that increasing GlgP activity would mobilize carbon from the glycogen pool for sucrose biosynthesis. However, when GlgP was overexpressed in sucrose-secreting strains of S. elongatus PCC 7942 with its native SPS, there were no changes in glycogen content and a decrease in sucrose was observed (Ducat et al., 2012; Dan et al., 2022), while heterologous overexpression of both SPS6803 and GlgP reduced glycogen content while increasing sucrose secretion by 2.4-fold (Dan et al., 2022). The variability in sucrose production of glycogen-deficient strains might be related to the pleiotropic cellular deficiencies of these strains, including reduced growth, reduced O2 evolution and consumption, abnormal pigmentation, and light sensitivity (Suzuki et al., 2010; Ducat et al., 2012; Gründel et al., 2012; Xu et al., 2013; Qiao et al., 2018). These phenotypes align with a potential broader role for glycogen beyond carbon storage, which may include buffering against periods of starvation, oxygenic stress, high-light stress, salt stress, or diurnal/transient changes in light availability (Luan et al., 2019; Shinde et al., 2020). Given the increasing recognition of regulatory roles of glycogen, more nuanced strategies may be required to regulate the flux of carbon towards glycogen synthesis in order to reliably improve sucrose bioproduction (Huang et al., 2016). 10 Table 1.1. Productivity and genetic modifications of sucrose-producing cyanobacteria. Over- expressedb* Down-regulatedc Maximum Productivityd Salt for SPS Sucrose promoter Ref. − − − 24 mol 10-17 cellse glgA-I, glgA-II 71 mol 10-17 cellse 1 M NaCl 1 M NaCl − − (Xu et al., 2013) Speciesa Syn7002 Syn6803 Syn7942 Syn7942 cscB, sps6803, spp6803, ugp sps6803, spp6803, ugp cscB cscB, sps7942 − − cscB, sps6803 Syn7942 cscB, sps6803, glgP cscB, sps7942, glgC Syn7942 Syn6803 − − − ggpS, ggtCD 0.69 mg L-1 h-1 f 400 mM NaCl PpetE ggpS 3.1 mg L-1 h-1 f 600 mM NaCl PpetE − − 3.6 mg gDW-1 h-1 200 mM NaCl 6.2 mg gDW-1 h-1 200 mM NaCl Synpcc7942_1125 5.9 mg L-1 OD730 -1 h-1 e 150 mM NaCl manR (Synpcc7942_1404) 6.7 mg L-1 OD730 -1 h-1 e 150 mM NaCl 5.6 mg L-1 h-1 6.9 mg L-1 h-1 − − (Du et al., 2013) (Duan et al., 2016) (Qiao et al., 2019) (Dan et al., 2022) − Ptrc − − PcpcB PcpcB 8 mg L-1 h-1 150 mM NaCl Ptrc (Qiao et al., 2018) cscB, sps6803, invA (sll0626), spp6803, ugp ggpS, ggtCD 10.1 mg L-1 h-1 200 mM NaCl PpetE (Kirsch et al., 2018) Syn7942 cscB − 10.4 mg L-1 h-1 150 mM NaCl − (Löwe et al., 2017) 11 Table 1.1. (cont’d) Syn7942 cscB Syn7942 cscB Syn7942 cscB, sps6803 Syn2973 cscB Syn7942 cscB cscB cscB, sps6803, Syn2973 spp6803 cscB − − − − − − − 11 mg L-1 h-1 150 mM NaCl 16.7 mg L-1 h-1 106 mM NaCl − − (Li et al., 2022) (Hays et al., 2017) 30 mg L-1 h-1 − Ptrc (Abramson et al., 2016) 24.6 mg L-1 h-1 150 mM KCl 35.5 mg L-1 h-1 150 mM NaCl 28 mg L-1 h-1 150 mM NaCl invA, glgC 36.1 mg L-1 h-1 150 mM NaCl 22.2 mg L-1 h-1 47.2 mg L-1 h-1 − − 79.2 mg L-1 h-1 150 mM NaCl − (Song et al., 2016) (Ducat et al., 2012) − − − − Ptrc1O, induced Ptrc1O, (Lin et al., 2020a) uninduced a Syn7002, Synechococcus sp. PCC 7002; Syn7942, Synechococcus elongatus PCC 7942; Syn2973, Synechococcus elongatus UTEX 2973; Syn6803, Synechocystis sp. PCC 6803. b cscB, sucrose permease; glgC, ADP-glucose pyrophosphorylase; glgP, glycogen phosphorylase; rpaB, regulator of phycobilisome- associated B; spp, sucrose phosphate phosphatase; sps, sucrose phosphate synthase; ugp, UDP-glucose pyrophosphorylase. c Genes are down-regulated or knocked out; ggpS, glucosylglycerol (GG)-phosphate synthase; ggtCD, GG transport system permease; glgA-I/glgA-II, glycogen synthase; glgC, ADP-glucose pyrophosphorylase; invA, invertase; manR (Synpcc7942_1404), manganese sensing response regulator; Synpcc7942_1125, histidine-containing phosphotransfer. d Approximated extracellular sucrose values provided or calculated from titers. e Calculated from intracellular sucrose yields. f Calculated from intracellular and extracellular (total) sucrose yields. * Subscripts in sps and spp indicate the strain that it comes from (i.e., 6803 for Synechocystis sp. PCC 6803). 12 In some cyanobacterial strains that utilize other compatible solutes as the dominant metabolite for osmoprotection, synthesis of these osmoprotectant compounds may compete with sucrose biosynthesis. One example is GG, the primary solute utilized by moderately halotolerant cyanobacteria such as Synechocystis sp. PCC 6803 (Klähn and Hagemann, 2011). When GG- phosphate synthase (GgpS), the enzyme that generates a GG precursor, was knocked out in Synechocystis sp. PCC 6803, increased flux of carbon to sucrose production was reported (Du et al., 2013; Kirsch et al., 2019; Thiel et al., 2019) (Table 1.1). A GgpS variant incapable of generating GG under salt stress instead accumulated nearly 1.5-fold more sucrose than wild-type cells, although these engineered strains also exhibited growth inhibition at lower salt concentrations that would be well tolerated by wild-type lines (Du et al., 2013). While most studies focus on restricting metabolic pathways that consume cellular carbon resources, downregulation of processes that compete for reducing equivalents may be an alternative approach to engineering strains with high-sucrose productivity. Flavodiiron proteins are part of cyanobacterial photoprotective systems that are engaged during periods of redox stress (e.g., fluctuating light) and can direct electrons from an over-reduced photosynthetic electron transport chain to oxygen (Allahverdiyeva et al., 2015). The flavodiiron-catalyzed reaction is essentially a water–water cycle that dissipates potential energy from reducing equivalents generated from photosynthetic light reactions, but this reaction appears to be important for preventing photodamage under dynamic light conditions (Allahverdiyeva et al., 2013). Knockout of flavodiiron genes flv1 and flv3 in S. elongatus PCC 7942 could boost sucrose production in a cscB/sps6803 expressing background (Santos-Merino et al., 2021). Furthermore, activation of sucrose secretion pathways could partially compensate for the loss of Flv1/Flv3 under transient light changes, further suggesting that heterologous metabolic sinks may have some ability to utilize “overpotential” on the photosynthetic electron transport chain (Santos-Merino et al., 2021). Altering regulatory networks to increase sucrose synthesis Sucrose biosynthesis is a natural component of many cyanobacterial adaptive responses, so a deeper understanding of the regulatory networks that control this process could allow researchers to manipulate sucrose production in the absence of abiotic stressors. In this context, a couple of studies have reported promising improvements in sucrose secretion rates by altering cyanobacterial two-component regulatory proteins, although the specific mechanisms remain uncertain. In a screen of all two-component regulatory factors in S. elongatus PCC 7942, Qiao and 13 colleagues identified genes indirectly linked to sucrose productivity, glycogen accumulation, and photosynthetic activity (Qiao et al., 2019). The partial deletion of ManR, a protein that plays a regulatory role in Mn2+ uptake (Ogawa et al., 2002; Yamaguchi et al., 2002; Zorina et al., 2016), increased sucrose by 60%, a complete knockout of Synpcc7942_1125 increased sucrose by 41% (Qiao et al., 2019) (Table 1.1). In a separate study, overexpression of the two-component protein regulator of phycobilisome assembly B (rpaB) reproduced a growth-arrest phenotype in S. elongatus PCC 7942 (Moronta-Barrios et al., 2013), and increased sucrose secretion in a cscB- expressing background (Abramson et al., 2018) (Table 1.1). It was suggested that the growth arrest restricted carbon flux to many downstream pathways that might otherwise compete for sucrose biosynthesis, though a more specific alteration in regulatory processes controlling sucrose synthesis could not be excluded. While the number of studies is still limited, two-component signaling pathways have so far proven to be a promising strategy to improve sucrose yields, though our mechanistic understanding for these phenotypes is far from complete. Applications of sucrose production in co-culture While the biotechnological focus for cyanobacteria has predominantly been upon direct synthesis of high-value products (Ducat et al., 2012; Knoot et al., 2018), there is growing interest in utilizing sugar-producing cyanobacteria for indirect bioproduction. This approach involves the use of carbohydrate-secreting cyanobacteria that support the growth of co-cultivated heterotrophic microbes. Co-cultures become “one-pot” reactions where cyanobacteria specialize in photosynthetic metabolism to supply carbon to a heterotroph, which in turn performs the metabolic labor of converting the carbon to higher value goods or services (Hays and Ducat, 2015; Ortiz- Reyes and Anex, 2022). In this section, I will cover the modular nature of synthetic microbial consortia designed using sucrose-secreting cyanobacteria, their applications, and their future opportunities and challenges. Potential advantages of modular microbial platforms Microbial bioproduction is now a well-recognized approach that harnesses metabolic diversity for synthesis of valuable chemicals (e.g., polymers, fuels, pharmaceuticals) as an alternative to traditional environmentally unsustainable processes (Tsuge et al., 2016; Wendisch et al., 2016; Liu and Nielsen, 2019; Zhong, 2020; Wu et al., 2021). Multiple decades of sustained investments in microbial research, prospecting, and genetic engineering have yielded a wealth of bacterial strains optimized to generate specific bioproducts. In some cases, efficient bioproduction 14 Reference (Hays et al., Table 1.2. Sucrose-based autotroph-heterotroph co-cultures and their products. Sucrose Strain Heterotroph Strain Speciesa Genotypeb* Maximum productivityc Induction Speciesd Genotypee Productf Maximum productivityg Syn7942 cscB 400 mg L-1 d-1 106 mM B. subtilis − α-amylase not quantified NaCl E. coli W PHB 0.04 mg L-1 d-1 2017)† phaCAB ΔcscR Syn7942 cscB 34.2 mg L-1 d-1 Syn7942 cscB 27.4 mg L-1 d-1 Syn7942 cscB 0.5 mg L-1 d-1 N/A; physical encapsulation 150 mM NaCl 170 mM NaCl A. vinelandii ΔnifL PHB 8 mg L-1 d-1 A. vinelandii ΔnifL PHB 3.8% DW d-1 H. boliviensis − PHB 28.3 mg L-1 d-1 Syn7942 cscB 102 mg L-1 d-1 150 mM P. putida cscRABY NaCl EM178 ΔnasT PHA 2.3 mg L-1 d-1 Syn7942 cscB 250 mg L-1 d-1 150 mM P. putida NaCl EM178 cscAB PHA 23.8 mg L-1 d-1 Syn7942 cscB 108 mg L-1 d-1** 150 mM P. putida cscRABY NaCl EM178 ΔnasT PHA 42.1 mg L-1 d-1 (Smith and Francis, 2017)‡ (Smith and Francis, 2016)† (Weiss et al., 2017)‡ (Hobmeier et al., 2020)† (Löwe et al., 2017) (Kratzl et al., 2022) Syn7942 cscB 45 mg L-1 d-1 100 mM NaCl R. glutinis − DW 24.8 mg L-1 DW-1 d-1 (Li et al., 2017)† TFA 1.2 mg L-1 d-1 15 Table 1.2. (cont’d) Syn2973 cscB 96 mg L-1 d-1 150 mM E. coli cscA, cscB, NaCl BL21(DE3) cscK, mcr 3-HP 9.8 mg L-1 d-1 (Ma et al., 2022)† Therm PKUAC Syn7942 cscB 10 mg L-1 d-1 cscB 18.1 mg L-1 d-1 150 mM E. coli efe ethylene 0.74 mg L-1 d-1 NaCl BL21(DE3) ispS isoprene 0.03 mg L-1 d-1 vioABCDE, cscABK violacein 4.5 mg L-1 d-1 CupH16 sps6803, spp6803, scrY 18.1 mg L-1 d-1 0.3% arabinose E. coli W ΔcscR crtEBIY, Syn2973 cscB 0.7 mg L-1 d-1** Y. lipolytica 150 mM NaCl P. putida KT2440 Syn7942 cscB 263.5 mg L-1 d-1 150 mM NaCl V. natriegens 16 cscABK β-carotene 4.8 mg L-1 d-1 ΔcscR carB, carRP β-carotene 325 mg L-1 d-1 sfp, bpsA indigoidine 1.9 g L-1 d-1 tyr tal budABC melanin 1.56 mg L-1 d-1 p-coumaric acid 2,3- butanediol 8.75 mg L-1 d-1 60 mg L-1 d-1 ldh lactate 100 mg L-1 d-1 (Cui et al., 2022) (Nangle et al., 2020) (Zhao et al., 2022)‡ (Li et al., 2022)† Table 1.2. (cont’d) Syn7942 cscB, sh3l 108 mg L-1 d-1 50 mM NaCl P. putida S12 cscA, hmfH, sh3d FDCA ~100% in 4 d (Lin et al., 2020b)†‡ Syn7942 cscB 240 mg L-1 d-1 100 mM P. putida cscRABY, NaCl EM173 dnt 2,4-DNT degradation 22.7 mg L-1 d-1 (Fedeson et al., PHA 5.1 mg L-1 d-1 2020)‡ a CupH16, Cupriavidus necator H16; Syn7942, Synechococcus elongatus PCC 7942; Syn2973, Synechococcus elongatus UTEX 2973; Syn6803, Synechocystis sp. PCC 6803; ThermPKUAC, Thermosynechococcus elongatus PKUAC-SCTE542. b cscB, sucrose permease; scrY, sucrose porin; sh3l, SH3 ligand; sps, sucrose phosphate synthase; spp, sucrose phosphate phosphatase. c Approximated values from axenic cultivations in conditions most similar to co-culture conditions. d A. vinelandii, Azotobacter vinelandii AV3; B. subtilis, Bacillus subtilis 168; E. coli, Escherichia coli; G. sulfurreducens, Geobacter sulfurrenducens PCA; H. boliviensis, Halomonas boliviensis; P. putida, Pseudomonas putida; R. glutinis, Rhodotorula glutinis; S. onedensis, Shewanella onedensis MR-1; Y. lipolytica, Yarrowia lipolytica CLIB138; V. natriegens, Vibrio natriegens. e bpsA, non-ribosomal peptide synthetase; budABC, 2,3-butanediol gene cluster; carB, phytoene dehydrogenase; carRP, bifunctional lycopene cyclase/phytoene synthase; crtEBIY, β-carotene biosynthesis cassette; cscA, sucrose hydrolase; cscB, sucrose permease; cscK, fructokinase; cscR, sucrose operon repressor; cscY, sucrose porin; dnt, dinitrotoluence degradation gene cluster; efe, ethylene- forming protein; frdABCD, operon encoding fumarate reductase; glk, glucokinase; gtfA, sucrose phosphorylase; hmfH, HMF/furfural oxidoreductase; Inv, cscA invertase gene with an N-terminal pelB leader sequence; ispS, isoprene synthase; ldh, D-lactate dehydrogenase; mcr, malonyl-CoA reductase; mgsA, methylglyoxal synthase; narG, napA, and narZ, nitrate reductases; nasT, nitrate response regulator; nifL, negative regulator of nitrogen fixation; Parvi, synthetic Baeyer–Villiger monooxygenase; pflB, pyruvate formate-lyase B; phaCAB, polyhydroxybutyrate synthesis operon; sfp, phosphopantetheinyl transferase; sh3d, SH3 domain; tal, tyrosine ammonia lyase; tyr, tyrosinase; vioABCDE, violacein biosynthesis cassette. Unless otherwise denoted by “Δ,” genes are heterologously expressed. f 3-HP, 3-hydroxypropionic acid; DNT, dinitrotoluene; DW, cyanobacterial biomass dry weight; FDCA, 2,5-furandicarboxylic acid; PHA, polyhydroxyalkanoate; PHB, polyhydroxybutyrate; TFA, cyanobacterial total fatty acids. g Approximated values provided or calculated from titers. † Enhanced photoautotroph growth in co-culture. ‡ Implemented spatial control of co-culture. * Subscript in sps and spp indicates the strain that it comes from (i.e., 6803 for Synechocystis sp. PCC 6803). ** Values from axenic cultivation prior to the introduction of heterotroph. *** Maximum power output reported for the four-species consortium. 17 of a target compound can be achieved by expressing relevant metabolic pathways in different microbial species. But there are also many examples where heterologously expressed metabolic pathways perform poorly due to other physiological features of a microbe that make it a non- optimal chassis (Calero and Nikel, 2019). For this reason, it is often non-trivial to re-engineer cyanobacterial metabolism for direct synthesis of a desired compound, which may stubbornly resist efforts to improve product titer (Savakis and Hellingwerf, 2015; Nagarajan et al., 2017; Lin and Pakrasi, 2019). A modular approach for multi-species product synthesis offers the capacity to leverage species with the most compatible physiology and desirable endogenous pathways for a given biochemical transformation, thus bypassing metabolic limitations of one biological chassis. At least in theory, each member of a synthetic microbial consortium can be conceptualized as a “module” selected specifically to perform functions well-suited with organism’s abilities. In this context, cyanobacteria-heterotroph co-cultures can be rationally designed to retain the advantages of cyanobacterial metabolism (i.e., use of light/ CO2 inputs, efficient carbon fixation) and paired with other microbes that have demonstrated efficiency in transforming simple carbohydrates into a desired end product. Additionally, because the co-culture output can be changed by swapping the “heterotrophic module” (i.e., organism), some steps to optimize synthesis for one product (e.g., improving cyanobacterial sucrose production) may be transferable to achieve enhanced synthesis across many distinct cyanobacteria-heterotroph pairings. In practice, sucrose-secreting cyanobacteria have already been used as the basis for engineered microbial communities with numerous heterotrophic species and for a variety of end products (Table 1.2), although a number of improvements will be required to make these co-cultures feasible for scaled application. Cyanobacterial co-culture as a flexible platform for value-added products At the time of this writing, the most common metabolic output reported from cyanobacteria-heterotroph co-cultures are polyhydroxyalkanoates (PHAs), a class of biological polymers with comparative qualities to petroleum-based plastics. PHAs have the advantage of being both compatible in blends with commonly used petroleum-based polymers while also exhibiting superior biodegradation properties (Boey et al., 2021; Mezzina et al., 2021). Additionally, some heterotrophic microbes utilize PHAs as an intracellular storage polymer and under stress, conditions can naturally hyperaccumulate PHAs in excess of 80% of their dry cell mass (Leong et al., 2014; Lee et al., 2021), making these compounds an ideal test case for the 18 division of labor between metabolic specialists, as outlined above. Polyhydroxybutyrate (PHB) is a PHA polymer that has been produced in cyanobacterial co-culture with three different heterotrophic species: Azotobacter vinelandii, Halomonas boliviensis, and E. coli W (Hays et al., 2017; Smith and Francis, 2017; Weiss et al., 2017) (Table 1.2). PHB is a natural storage polymer for both A. vinelandii and H. boliviensis, while heterologous expression of the phaCAB operon in E. coli will confer PHB synthesis capability. The most productive co-cultures reported included a heterotrophic partner species that was naturally capable of PHB synthesis. Notably, the co- cultivation of S. elongatus PCC 7942 cscB with H. boliviensis was extended over 6 months with no organic carbon input, demonstrating that these synthetic consortia can be stable and productive over long time periods (Weiss et al., 2017). Pseudomonas putida is a model organism that naturally accumulates medium chain length PHAs (mcl-PHAs) granules in response to starvation, primarily under low-nitrogen and high- carbon conditions (Hoffmann and Rehm, 2004). While sucrose is not naturally consumed by P. putida, expression of heterologous sucrose transporters and sucrose hydrolyzing enzymes allows it to grow on sucrose as the sole carbon source (Sabri et al., 2013; Löwe et al., 2020), a strategy that has been used to enable other microbial species without native pathways to consume cyanobacterially secreted sucrose (Sabri et al., 2013; Hobmeier et al., 2020; Zhang et al., 2020). Indeed, initial reports demonstrated that P. putida expressing cscAB was capable of growing solely on sucrose provided by S. elongatus PCC 7942 and accumulated PHA in co-culture, though sucrose utilization was incomplete and productivities were modest (Löwe et al., 2017; Fedeson et al., 2020). Additional expression of a sucrose porin (cscY) and a sucrose operon repressor (cscR) further improved sucrose utilization (Löwe et al., 2020), while further optimization of the nitrogen- deficiency response pathway (Hobmeier et al., 2020) and culture conditions could boost PHA titer further (Kratzl et al., 2022) (Table 1.2). Other co-culture products include the metabolites ethylene, isoprene, 3-hydroxypropionic acid (3-HP), and 2,3-butanediol (Table 1.2), which are compounds in a broader class of industrially relevant precursors widely used for chemical synthesis (e.g., diols, organic acids, gaseous alkenes) (Cui et al., 2022; Li et al., 2022; Ma et al., 2022). In most of these reports, the heterotrophic microbe utilized were E. coli substrains, although the rapidly growing halophile Vibrio natriegens was able to produce a relatively high amount of 2,3-butanediol in co-culture (Li et al., 2022). Interestingly, co-cultures of S. elongatus PCC 7942 and P. putida designed to convert 5- 19 hydroxymethylfurfural to 2,5-furandicarboxylic acid (FDCA), a common precursor molecule, exhibited higher efficiency when the two species were engineered to display complementary surface proteins (Lin et al., 2020b). The authors suggest that physical binding between the two species could improve metabolic exchange (Lin et al., 2020b), an intriguing strategy that may be valuable to develop further. Beyond commodity products, several higher-value chemicals expand the metabolic repertoire of cyanobacteria-heterotroph co-cultures. The pigment industry routinely uses a number of compounds that generate significant environmental hazards when chemically synthesized (Pereira and Alves, 2012). Biosynthetic pathways for pigment derivatives (e.g., indigoidine for the popular pigment, indigo) are being explored for more environmentally conscious pigment synthesis (Celedón and Díaz, 2021). Recently, co-cultures have been reported for the synthesis of indigoidine using the heterotroph P. putida, β-carotene with E. coli or the yeast Yarrowia lipolytica, and violacein by E. coli (Nangle et al., 2020; Zhao et al., 2022). Although most cyanobacteria-heterotroph co-cultures make use of the model laboratory strain S. elongatus PCC 7942, Zhao and colleagues used a sucrose-secreting variant of the fast-growing and high-light tolerant relative, S. elongatus UTEX 2973, in their co-culture experiments to produce indigoidine and β-carotene (Zhao et al., 2022). The cosmetic p-coumaric acid, is another higher value compound useful for its antioxidant and antimicrobial properties (Boz, 2015; Boo, 2019). The biosynthetic pathway for p-coumaric acid was introduced into V. natriegens and co-cultures of these engineered strains with S. elongatus PCC 7942 allowed for photosynthetically driven p- coumaric acid production (Li et al., 2022). Other recent reports provide further evidence of the flexibility of this cyanobacterial co-cultivation system (Table 1.2), including bioproduction of fatty acids (Li et al., 2022), ε-caprolactone (Tóth et al., 2022), lactate (Li et al., 2022), and secreted enzymes (Hays et al., 2017). Additionally, some products can be used to feed downstream bacteria and develop more complex systems. A four-species consortium utilized lactate-consuming Shewanella oneidensis to generate electricity and acetate, in which the latter was consumed by Geobacter sulfurreducens to produce CO2 for S. elongatus PCC 7942 (Zhu et al., 2022). Co-culture as a platform to study microbial communities Phototrophs and heterotrophs are often metabolically intertwined in natural contexts (Morris, 2015; Henry et al., 2016). For example, many marine Prochlorococcus species secrete organic carbon to neighboring heterotrophic partners that perform functions in detoxifying reactive 20 oxygen species present in the open ocean (Morris et al., 2011; Braakman et al., 2017). It has been hypothesized that the natural export of sugars from Prochlorococcus and other cyanobacteria may prime them to engage with surrounding heterotrophs via cross-feeding, and potentially “outsource” the metabolic burden of synthesizing some nutritional requirements to other organisms (Werner et al., 2014; Henry et al., 2016; Braakman et al., 2017). Natural microbial communities and symbiotic relationships usually develop over evolutionary time scales and may exhibit numerous and complex cross-feeding patterns and other self-stabilizing interactions (Konopka et al., 2015). Yet, these important dynamics can be challenging to study due to the difficulty of disentangling specific mechanisms from the complex interaction networks (Ponomarova and Patil, 2015). The fact that many natural symbioses also have cyanobacterial partners that exchange fixed carbon for other microbial partner(s) has led some groups to explore synthetic cyanobacteria/heterotroph co- cultures as a possible “bottom-up” system to gain insight into complex microbial consortia. Synthetic phototroph-heterotroph microbial consortia may represent a complementary system to study natural consortia in parallel, as they present a platform for interrogating microbial interactions that is relatively simple, genetically tractable, and experimentally tunable (De Roy et al., 2014; Song et al., 2015) (Table 1.3). One intriguing phenomenon that recurs across several synthetic cyanobacteria/ heterotroph co-cultures is an increase in the vigor or productivity of one or both partners relative to axenic controls. For instance, cyanobacterial growth was enhanced in mixed culture with several heterotrophic species (Hays et al., 2017; Li et al., 2017; Hobmeier et al., 2020; Ma et al., 2022), although the partner species were evolutionarily “naïve” to one another. Similarly, heterotrophic productivity in co-culture can be significantly higher than can be attributed to the cyanobacterially secreted sucrose (Hays et al., 2017; Cui et al., 2022; Ma et al., 2022). Conversely, when cyanobacteria are allowed to overpopulate a synthetic co-culture, heterotrophic partners may exhibit reduced viability (Hays et al., 2017). It is highly likely that some of these effects arise due to unprogrammed metabolic interactions and emergent behaviors of division of labor (Rafieenia et al., 2022), such as the generation of damaging reactive oxygen species (Hays et al., 2017; Ma et al., 2022). Our current understanding of the emergent properties of mixed microbial communities is limited and cannot fully explain observed phenomena. Preliminary analyses and multi-omics approaches have been used to predict hidden interactions within synthetic consortia, providing insight on areas of cooperation and competition that could be validated and exploited to design 21 Table 1.3. Synthetic cyanobacteria-heterotroph microbial consortia used as a platform to study microbial interactions. Sucrose straina Genotypeb Heterotroph strainc E. coli K-12 Genotyped AA Notes Reference knockouts Utilizes metabolic modeling and experimental Syn7942 cscB E. coli W − validation to predict co-cultivation outcomes (Zuñiga et al., 2020) Y. lipolytica SUC2 and identify optimizable parameters. B. subtilis − A. vinelandii ΔnifL Develops tripartite consortium with carbon- E. coli K-12 MG1655 C. glutamicum B. subtilis 168 cscABK − − providing S. elongatus PCC 7942 and nitrogen- providing A. vinelandii to support a third microbe. Performed computational analyses to identify bottlenecks to improve cultivation conditions. Syn7942 cscB, sps6803 (Carruthers, 2020) Syn2973 cscB E. coli BL21(DE3) Syn7942 cscB E. coli MG1655 Utilizes transcriptomic, proteomic, and metabolomic analyses to reveal differentially cscABK, mcr regulated pathways during co-cultivation to (Ma et al., 2022) identify optimizable parameters to improve stability and 3-hydroxypropionic productivity. Spatially separates subpopulations with cscABK encapsulation to impart species stability while (Wang et al., 2022) still allowing the transport of small molecules. 22 Table 1.3 (cont’d) Syn7942 Syn7942 cscB, sps6803 cscB, sps6803 E. coli W ΔcscR E. coli W ΔcscR Utilizes individual-based modeling in spatial context to predict colony fitness. Integrates quorum sensing modules for cross- species communication. (Sakkos et al., 2022) (Kokarakis et al., 2022) a Syn7942, Synechococcus elongatus PCC 7942; Syn2973, Synechococcus elongatus UTEX 2973. b AA knockouts, multiple one-way amino acid auxotrophs were generated; cscB, sucrose permease; sps6803, sucrose phosphate synthase from Synechocystis sp. PCC 6803. c A. vinelandii, Azotobacter vinelandii AZBB163; B. subtilis, Bacillus subtilis 168; C. glutamicum, Corynebacterium glutamicum 13032; E. coli, Escherichia coli, Y. lipolytica, Yarrowia lipolytica Po1g. dcscABK, sucrose utilization operon; cscR, sucrose operon repressor; mcr, malonyl-CoA reductase; nifL, negative regulator of nitrogen fixation; SUC2, cassette for internal and external invertases. 23 more robust co-cultures (Carruthers, 2020; Zuñiga et al., 2020; Ma et al., 2022). Synthetic co- cultures also present a simpler set of variables in comparison to natural communities which may be more amenable to simulations, such as agent-based modeling, for predicting emergent behaviors in a population (Sakkos et al., 2023). Finally, additional layers of metabolic exchange can be designed into the synthetic co-culture system to experimentally probe and validate hypotheses of interspecies exchange. A notable example in this regard is multiple groups’ use of the diazotroph, A. vinelandii, to fix atmospheric nitrogen and secrete ammonia, effectively creating a carbon-for-nitrogen exchange in co-culture with sucrose-secreting cyanobacteria (Smith and Francis, 2017; Carruthers, 2020). Taken together, the computational, systems, and genetic toolkits available for synthetic microbial consortia may lead to important insights on the dynamics of microbial exchange that would be difficult to probe in natural microbiomes. Organizational structure of dissertation In my dissertation, I focused upon improving the potential of the aforementioned co-culture system of photosynthetic bioproduction. The co-culture approach offers a degree of flexibility in output products that is difficult to match using other metabolic engineering approaches, and the overall sucrose production from cyanobacteria is promising relative to terrestrial plant carbohydrate feedstocks. Nonetheless, at this time, any attempts to scale cyanobacterial co-culture to industrially relevant volumes would surely run into a number of problems. My dissertation research focused on two potential weaknesses of cyanobacterial co-culture. In Chapter 2, I describe my research to maximize the output of sucrose by modulating the cultivation conditions for sucrose-secreting strains of S. elongatus. Maximizing heterotrophic yield in co-culture is directly related to the amount of fixed carbon that can be supplied by the cyanobacterial partner, and I interrogated a number of different conditions of illumination as well as ways to supply inorganic CO2 to cyanobacterial cultures. In Chapter 3, I describe my efforts to rationally design co-cultures for higher resilience and productivity through engineering in a system for physical attachment between the heterotroph and cyanobacterial partner. Specifically, I expanded upon prior work on a cyanobacterial surface display system with the long-range goal of installing extracellular binding domains on the surface of S. elongatus that could be used to mediate direct physical attachments with other proteins displayed on the surface of an engineered heterotrophic partner. Finally, in Chapter 4, I review the outcomes and conclusions of my research program and reflect upon both my experience and future prospects for the topic areas central to my dissertation. 24 REFERENCES Abramson, B. W., Kachel, B., Kramer, D. M., and Ducat, D. C. (2016). Increased photochemical efficiency in cyanobacteria via an engineered sucrose sink. Plant Cell Physiol. 57, 2451– 2460. doi: 10.1093/pcp/pcw169 Abramson, B. W., Lensmire, J., Lin, Y. T., Jennings, E., and Ducat, D. C. (2018). Redirecting carbon to bioproduction via a growth arrest switch in a sucrose-secreting cyanobacterium. Algal Res. 33, 248–255. doi: 10.1016/j.algal.2018.05.013 Adomako, M., Ernst, D., Simkovsky, R., Chao, Y.-Y., Wang, J., Fang, M., et al. (2022). comparative genomics of Synechococcus elongatus explains the phenotypic diversity of the strains. mBio 13, e00862-22. doi: 10.1128/mbio.00862-22 Aikens, J. and Turner, R.J., PROTERRO Inc (2013). Method of producing a fermentable sugar. U.S. Patent 8,597,914. Albi, T., Ruiz, M. T., Reyes, P. de los, Valverde, F., and Romero, J. M. (2016). Characterization of the sucrose phosphate phosphatase (SPP) isoforms from Arabidopsis thaliana and role of in dimerization. PLOS ONE 11, e0166308. doi: the S6PPc domain 10.1371/journal.pone.0166308 Allahverdiyeva, Y., Isojärvi, J., Zhang, P., and Aro, E.-M. (2015). Cyanobacterial oxygenic flavodiiron proteins. Life 5, 716–743. doi: is protected by photosynthesis 10.3390/life5010716 Allahverdiyeva, Y., Mustila, H., Ermakova, M., Bersanini, L., Richaud, P., Ajlani, G., et al. (2013). Flavodiiron proteins Flv1 and Flv3 enable cyanobacterial growth and photosynthesis under fluctuating light. Proc. Natl. Acad. Sci. 110, 4111–4116. doi: 10.1073/pnas.1221194110 Angermayr, S. A., Paszota, M., and Hellingwerf, K. J. (2012). Engineering a cyanobacterial cell factory for production of lactic acid. Appl. Environ. Microbiol. 78, 7098–7106. doi: 10.1128/AEM.01587-12 Bakku, R. K., and Rakwal, R. (2022). “Applications of cyanobacterial compounds in the energy, health, value-added product, and agricultural sectors: A perspective,” in Cyanobacterial Physiology, eds. H. Kageyama and R. Waditee-Sirisattha (Academic Press), 149–164. doi: 10.1016/B978-0-323-96106-6.00009-5 Baran, R., Lau, R., Bowen, B. P., Diamond, S., Jose, N., Garcia-Pichel, F., et al. (2017). Extensive turnover of compatible solutes in cyanobacteria revealed by deuterium oxide (D2O) stable isotope probing. ACS Chem. Biol. 12, 674–681. doi: 10.1021/acschembio.6b00890 Blank, C. E. (2013). Phylogenetic distribution of compatible solute synthesis genes support a freshwater origin for cyanobacteria. J. Phycol. 49, 880–895. doi: 10.1111/jpy.12098 Boey, J. Y., Mohamad, L., Khok, Y. S., Tay, G. S., and Baidurah, S. (2021). A review of the applications and biodegradation of polyhydroxyalkanoates and poly(lactic acid) and its composites. Polymers 13, 1544. doi: 10.3390/polym13101544 Boo, Y. C. (2019). p-Coumaric acid as an active ingredient in cosmetics: a review focusing on its antimelanogenic effects. Antioxidants 8, 275. doi: 10.3390/antiox8080275 25 Boz, H. (2015). p-Coumaric acid in cereals: presence, antioxidant and antimicrobial effects. Int. J. Food Sci. Technol. 50, 2323–2328. doi: 10.1111/ijfs.12898 Braakman, R., Follows, M. J., and Chisholm, S. W. (2017). Metabolic evolution and the self- organization of ecosystems. Proc. Natl. Acad. Sci. U. S. A. 114, E3091–E3100. doi: 10.1073/pnas.1619573114 Calero, P., and Nikel, P. I. (2019). Chasing bacterial chassis for metabolic engineering: a perspective review from classical to non‐traditional microorganisms. Microb. Biotechnol. 12, 98–124. doi: 10.1111/1751-7915.13292 Carruthers, D. (2020). Engineering modular synthetic microbial consortia for sustainable bioproduction from CO2. University of Michigan, Horace H. Rackham School of Graduate Studies. Celedón, R. S., and Díaz, L. B. (2021). Natural pigments of bacterial origin and their possible biomedical applications. Microorganisms 9, 739. doi: 10.3390/microorganisms9040739 Chen, B., Wan, C., Mehmood, M. A., Chang, J.-S., Bai, F., and Zhao, X. (2017). Manipulating environmental stresses and stress tolerance of microalgae for enhanced production of lipids and value-added products–a review. Bioresour. Technol. 244, 1198–1206. doi: 10.1016/j.biortech.2017.05.170 Crockford, P. W., Bar On, Y. M., Ward, L. M., Milo, R., and Halevy, I. (2023). The geologic doi: productivity. Curr. Biol. 4741-4750.e5. 33, history 10.1016/j.cub.2023.09.040 primary of Cui, Y., Rasul, F., Jiang, Y., Zhong, Y., Zhang, S., Boruta, T., et al. (2022). Construction of an artifcial consortium of Escherichia coli and cyanobacteria for clean indirect production of volatile platform hydrocarbons from CO2. Front. Microbiol. 13, 965968. doi: 10.3389/fmicb.2022.965968 Cumino, A. C., Marcozzi, C., Barreiro, R., and Salerno, G. L. (2007). Carbon cycling in Anabaena sp. PCC 7120. sucrose synthesis in the heterocysts and possible role in nitrogen fixation. Plant Physiol. 143, 1385–1397. doi: 10.1104/pp.106.091736 Cumino, A., Curatti, L., Giarrocco, L., and Salerno, G. L. (2002). Sucrose metabolism: Anabaena sucrose-phosphate synthase and sucrose-phosphate phosphatase define minimal functional domains shuffled during evolution. FEBS Lett. 517, 19–23. doi: 10.1016/S0014- 5793(02)02516-4 Curatti, L., Flores, E., and Salerno, G. (2002). Sucrose is involved in the diazotrophic metabolism of the heterocyst-forming cyanobacterium Anabaena sp. FEBS Lett. 513, 175–178. doi: 10.1016/S0014-5793(02)02283-4 Curatti, L., Folco, E., Desplats, P., Abratti, G., Limones, V., Herrera-Estrella, L., et al. (1998b). Sucrose-phosphate synthase from Synechocystis sp. strain PCC 6803: identification of the spsa gene and characterization of the enzyme expressed in Escherichia coli. J. Bacteriol. 180, 6776–6779. doi: 10.1128/jb.180.24.6776-6779.1998 Curatti, L., Giarrocco, L. E., Cumino, A. C., and Salerno, G. L. (2008). Sucrose synthase is involved in the conversion of sucrose to polysaccharides in filamentous nitrogen-fixing 26 cyanobacteria. Planta 228, 617–625. doi: 10.1007/s00425-008-0764-7 Curatti, L., Giarrocco, L., and Salerno, G. L. (2006). Sucrose synthase and RuBisCo expression is similarly regulated by the nitrogen source in the nitrogen-fixing cyanobacterium Anabaena sp. Planta 223, 891–900. doi: 10.1007/s00425-005-0142-7 Curatti, L., Porchia, A. C., Herrera-Estrella, L., and Salerno, G. L. (2000). A prokaryotic sucrose synthase gene (susA) isolated from a filamentous nitrogen-fixing cyanobacterium encodes a protein similar to those of plants. Planta 211, 729–735. doi: 10.1007/s004250000343 Dan, Y., Sun, J., Zhang, S., Wu, Y., Mao, S., Luan, G., et al. (2022). Manipulating the expression of glycogen phosphorylase in Synechococcus elongatus PCC 7942 to mobilize glycogen synthesis. Front. Bioeng. Biotechnol. 10, 925311. doi: storage 10.3389/fbioe.2022.925311 sucrose for De Roy, K., Marzorati, M., Van den Abbeele, P., Van de Wiele, T., and Boon, N. (2014). Synthetic microbial ecosystems: an exciting tool to understand and apply microbial communities. Environ. Microbiol. 16, 1472–1481. doi: 10.1111/1462-2920.12343 Du, W., Liang, F., Duan, Y., Tan, X., and Lu, X. (2013). Exploring the photosynthetic production doi: cyanobacteria. Metab. Eng. sucrose 17–25. 19, of capacity by 10.1016/j.ymben.2013.05.001 Duan, Y., Luo, Q., Liang, F., and Lu, X. (2016). Sucrose secreted by the engineered cyanobacterium and its fermentability. J. Ocean Univ. China 15, 890–896. doi: 10.1007/s11802-016-3007-8 Ducat, D. C., Avelar-Rivas, J. A., Way, J. C., and Silvera, P. A. (2012). Rerouting carbon flux to enhance photosynthetic productivity. Appl. Environ. Microbiol. 78, 2660–2668. doi: 10.1128/AEM.07901-11 Falcón, L. I., Magallón, S., and Castillo, A. (2010). Dating the cyanobacterial ancestor of the chloroplast. ISME J. 4, 777–783. doi: 10.1038/ismej.2010.2 Fedeson, D. T., Saake, P., Calero, P., Nikel, P. I., and Ducat, D. C. (2020). Biotransformation of 2,4-dinitrotoluene in a phototrophic co-culture of engineered Synechococcus elongatus and Pseudomonas putida. Microb. Biotechnol. 13, 997–1011. doi: 10.1111/1751-7915.13544 Fieulaine, S., Lunn, J. E., Borel, F., and Ferrer, J.-L. (2005). The structure of a cyanobacterial sucrose-phosphatase reveals the sugar tongs that release free sucrose in the cell. Plant Cell 17, 2049–2058. doi: 10.1105/tpc.105.031229 Gaysina, L. A., Saraf, A., and Singh, P. (2019). “Cyanobacteria in diverse habitats,” in Cyanobacteria, eds. A. K. Mishra, D. N. Tiwari, and A. N. Rai (Academic Press), 1–28. doi: 10.1016/B978-0-12-814667-5.00001-5 Gibson, R. P., Turkenburg, J. P., Charnock, S. J., Lloyd, R., and Davies, G. J. (2002). Insights into trehalose synthesis provided by the structure of the retaining glucosyltransferase OtsA. Chem. Biol. 9, 1337–46. doi: 10.1016/s1074-5521(02)00292-2 Gründel, M., Scheunemann, R., Lockau, W., and Zilliges, Y. (2012). Impaired glycogen synthesis causes metabolic overflow reactions and affects stress responses in the cyanobacterium Synechocystis sp. PCC 6803. Microbiology 158, 3032–3043. doi: 10.1099/mic.0.062950-0 27 Hagemann, M. (2011). Molecular biology of cyanobacterial salt acclimation. FEMS Microbiol. Rev. 35, 87–123. doi: 10.1111/j.1574-6976.2010.00234.x Hagemann, M. (2013). “Genomics of salt acclimation: Synthesis of compatible solutes among cyanobacteria,” in Advances in Botanical Research, eds. F. Chauvat and C. Cassier- Chauvat (Academic Press), 27–55. doi: 10.1016/B978-0-12-394313-2.00002-0 Hays, S. G., and Ducat, D. C. (2015). Engineering cyanobacteria as photosynthetic feedstock factories. Photosynth. Res. 123, 285–295. doi: 10.1007/s11120-014-9980-0 Hays, S. G., Yan, L. L. W., Silver, P. A., and Ducat, D. C. (2017). Synthetic photosynthetic consortia define interactions leading to robustness and photoproduction. J. Biol. Eng. 11, 1-4. doi: 10.1186/s13036-017-0048-5 Henry, C. S., Bernstein, H. C., Weisenhorn, P., Taylor, R. C., Lee, J., Zucker, J., et al. (2016). Microbial community metabolic modeling: a community data‐driven network reconstruction. J. Cell. Physiol. 231, 2339–2345. doi: 10.1002/jcp.25428 Herrero, A., Stavans, J., and Flores, E. (2016). The multicellular nature of filamentous heterocyst- forming cyanobacteria. FEMS Microbiol. Rev. 40, 831–854. doi: 10.1093/femsre/fuw029 Hincha, D. K., and Hagemann, M. (2004). Stabilization of model membranes during drying by compatible solutes involved in the stress tolerance of plants and microorganisms. Biochem. J. 383, 277–283. doi: 10.1042/BJ20040746 Hobmeier, K., Lowe, H., Liefeldt, S., Kremling, A., and Pfluger-Grau, K. (2020). A nitrate-blind P. putida strain boosts PHA production in a synthetic mixed culture. Front. Bioeng. Biotechnol. 8, 486. doi: 10.3389/fbioe.2020.00486 Hoffmann, N., and Rehm, B. H. A. (2004). Regulation of polyhydroxyalkanoate biosynthesis in Pseudomonas putida and Pseudomonas aeruginosa. FEMS Microbiol. Lett. 237, 1–7. doi: 10.1111/j.1574-6968.2004.tb09671.x Huang, C.-H., Shen, C. R., Li, H., Sung, L.-Y., Wu, M.-Y., and Hu, Y.-C. (2016). CRISPR interference (CRISPRi) for gene regulation and succinate production in cyanobacterium S. elongatus PCC 7942. Microb. Cell Fact. 15, 1–11. doi: 10.1186/s12934-016-0595-3 Kariyazono, R., Ito, S., and Osanai, T. (2022). “Carbon metabolism of great biotechnological interest: Metabolic engineering and synthetic biology of cyanobacteria,” in Cyanobacterial Physiology, eds. H. Kageyama and R. Waditee-Sirisattha (Academic Press), 189–200. doi: 10.1016/B978-0-323-96106-6.00003-4 Khan, M., Shin, J., and Kim, J. (2018). The promising future of microalgae: current status, challenges, and optimization of a sustainable and renewable industry for biofuels, feed, and other products. Microb. Cell Fact., 17, 1-21. doi: 10.1186/s12934-018-0879-x Kirsch, F., Klähn, S., and Hagemann, M. (2019). Salt-regulated accumulation of the compatible solutes sucrose and glucosylglycerol in cyanobacteria and its biotechnological potential. Front. Microbiol. 10, 482569. doi: 10.3389/fmicb.2019.02139 Kirsch, F., Luo, Q., Lu, X., and Hagemann, M. (2018). Inactivation of invertase enhances sucrose production in the cyanobacterium Synechocystis sp. PCC 6803. Microbiology 164, 1220– 1228. doi: 10.1099/mic.0.000708 28 Klähn, S., and Hagemann, M. (2011). Compatible solute biosynthesis in cyanobacteria. Environ. Microbiol. 13, 551–562. doi: 10.1111/j.1462-2920.2010.02366.x Knoot, C. J., Ungerer, J., Wangikar, P. P., and Pakrasi, H. B. (2018). Cyanobacteria: promising biocatalysts for sustainable chemical production. J. Biol. Chem. 293, 5044–5052. doi: 10.1074/jbc.R117.815886 Konopka, A., Lindemann, S., and Fredrickson, J. (2015). Dynamics in microbial communities: ISME J. 9, 1488–1495. doi: identify principles. to Unraveling mechanisms 10.1038/ismej.2014.251 Kratz, W. A., and Meyers, J. (1955). Nutrition and Growth of Several Blue-Green Algae. Am. J. Bot. 42, 282–287. doi: 10.2307/2438564 Kratzl, F., Kremling, A., and Pflüger-Grau, K. (2022). Streamlining of a synthetic co-culture towards an individually controllable one-pot process for polyhydroxyalkanoate production from light and CO2. Eng. Life Sci. 21, e2100156. doi: 10.1002/elsc.202100156 Kroumov, A. D., Zaharieva, M. M., Scheufele, F. B., Balabanova, V., and Najdenski, H. (2021). “Engineering challenges of carbon dioxide capture and sequestration by cyanobacteria,” in Ecophysiology and Biochemistry of Cyanobacteria, ed. R. P. Rastogi (Singapore: Springer Nature), 351–372. doi: 10.1007/978-981-16-4873-1_16 Lee, J., Park, H. J., Moon, M., Lee, J.-S., and Min, K. (2021). Recent progress and challenges in microbial polyhydroxybutyrate (PHB) production from CO2 as a sustainable feedstock: A doi: state-of-the-art 10.1016/j.biortech.2021.125616 Bioresour. Technol. 125616. review. 339, Leong, Y. K., Show, P. L., Ooi, C. W., Ling, T. C., and Lan, J. C.-W. (2014). Current trends in polyhydroxyalkanoates (PHAs) biosynthesis: Insights from the recombinant Escherichia coli. J. Biotechnol. 180, 52–65. doi: 10.1016/j.jbiotec.2014.03.020 Li, C., Wang, R., Wang, J., Liu, L., Li, H., Zheng, H., et al. (2022). A highly compatible phototrophic community for carbon‐negative biosynthesis. Angew. Chem. Int. Ed. 62, e202215013. doi: 10.1002/anie.202215013 Li, T., Li, C. T., Butler, K., Hays, S. G., Guarnieri, M. T., Oyler, G. A., et al. (2017). Mimicking lichens: Incorporation of yeast strains together with sucrose-secreting cyanobacteria improves survival, growth, ROS removal, and lipid production in a stable mutualistic co- culture production platform. Biotechnol. Biofuels 10, 1-11. doi: 10.1186/s13068-017-0736- x Liang, Y., Zhang, M., Wang, M., Zhang, W., Qiao, C., Luo, Q., et al. (2020). Freshwater cyanobacterium Synechococcus elongatus PCC 7942 adapts to an environment with salt stress via ion-induced enzymatic balance of compatible solutes. Appl. Environ. Microbiol. 86, 2904–2923. doi: 10.1128/AEM Lin, P. C., Zhang, F., and Pakrasi, H. B. (2020a). Enhanced production of sucrose in the fast- growing cyanobacterium Synechococcus elongatus UTEX 2973. Sci. Rep. 10, 390. doi: 10.1038/s41598-019-57319-5 Lin, P. C., and Pakrasi, H. B. (2019). Engineering cyanobacteria for production of terpenoids. 29 Planta 249, 145–154. doi: 10.1007/s00425-018-3047-y Lin, T. Y., Wen, R. C., Shen, C. R., and Tsai, S. L. (2020b). Biotransformation of 5- Hydroxymethylfurfural to 2,5-Furandicarboxylic Acid by a Syntrophic Consortium of Engineered Synechococcus elongatus and Pseudomonas putida. Biotechnol J 15, e1900357. doi: 10.1002/biot.201900357 Liu, Y., and Nielsen, J. (2019). Recent trends in metabolic engineering of microbial chemical factories. Curr. Opin. Biotechnol. 60, 188–197. doi: 10.1016/j.copbio.2019.05.010 López-Igual, R., Flores, E., and Herrero, A. (2010). Inactivation of a heterocyst-specific invertase indicates a principal role of sucrose catabolism in heterocysts of Anabaena sp. J. Bacteriol. 192, 5526–5533. doi: 10.1128/jb.00776-10 Löwe, H., Hobmeier, K., Moos, M., Kremling, A., and Pfluger-Grau, K. (2017). Photoautotrophic production of polyhydroxyalkanoates in a synthetic mixed culture of Synechococcus elongatus cscB and Pseudomonas putida cscAB. Biotechnol. Biofuels 10, 190. doi: 10.1186/s13068-017-0875-0 Löwe, H., Sinner, P., Kremling, A., and Pflüger‐Grau, K. (2020). Engineering sucrose metabolism in Pseudomonas putida highlights the importance of porins. Microb. Biotechnol. 13, 97– 106. doi: 10.1111/1751-7915.13283 Luan, G., Zhang, S., Wang, M., and Lu, X. (2019). Progress and perspective on cyanobacterial doi: engineering. Biotechnol. Adv. 771–786. 37, glycogen metabolism 10.1016/j.biotechadv.2019.04.005 Lunn, J. E., Ashton, A. R., Hatch, M. D., and Heldt, H. W. (2000). Purification, molecular cloning, and sequence analysis of sucrose-6F-phosphate phosphohydrolase from plants. Proc. Natl. Acad. Sci. 97, 12914–12919. doi: 10.1073/pnas.230430197 Lunn, J. E., Price, G. D., and Furbank, R. T. (1999). Cloning and expression of a prokaryotic sucrose-phosphate synthase gene from the cyanobacterium Synechocystis sp. PCC 6803. Plant Mol. Biol. 40, 297–305. doi: 10.1023/a:1006130802706 Ma, J., Guo, T., Ren, M., Chen, L., Song, X., and Zhang, W. (2022). Cross-feeding between cyanobacterium Synechococcus and Escherichia coli in an artificial autotrophic- heterotrophic coculture system revealed by integrated omics analysis. Biotechnol. Biofuels Bioprod. 15, 69. doi: 10.1186/s13068-022-02163-5 Mezzina, M. P., Manoli, M. T., Prieto, M. A., and Nikel, P. I. (2021). Engineering native and tailored synthetic pathways polyhydroxyalkanoates. Biotechnol. J. 16, 2000165. doi: 10.1002/biot.202000165 in Pseudomonas putida the production of for Morone, J., Alfeus, A., Vasconcelos, V., and Martins, R. (2019). Revealing the potential of cyanobacteria in cosmetics and cosmeceuticals — A new bioactive approach. Algal Res. 41, 101541. doi: 10.1016/j.algal.2019.101541 Moronta-Barrios, F., Espinosa, J., and Contreras, A. (2013). Negative control of cell size in the cyanobacterium Synechococcus elongatus PCC 7942 by the essential response regulator RpaB. FEBS Lett. 587, 504–509. doi: 10.1016/j.febslet.2013.01.023 Morris, J. J. (2015). Black Queen evolution: The role of leakiness in structuring microbial 30 communities. Trends Genet. 31, 475–482. doi: 10.1016/j.tig.2015.05.004 Morris, J. J., Johnson, Z. I., Szul, M. J., Keller, M., and Zinser, E. R. (2011). Dependence of the cyanobacterium Prochlorococcus on hydrogen peroxide scavenging microbes for growth at the ocean’s surface. PloS One 6, e16805. doi: 10.1371/journal.pone.0016805 Nagarajan, D., Lee, D.-J., Kondo, A., and Chang, J.-S. (2017). Recent insights into biohydrogen production by microalgae–From biophotolysis to dark fermentation. Bioresour. Technol. 227, 373–387. doi: j.biortech.2016.12.104 Nangle, S. N., Ziesack, M., Buckley, S., Trivedi, D., Loh, D. M., Nocera, D. G., et al. (2020). Valorization of CO2 through lithoautotrophic production of sustainable chemicals in Cupriavidus necator. Metab Eng 62, 207–220. doi: 10.1016/j.ymben.2020.09.002 Niederholtmeyer, H., Wolfstadter, B. T., Savage, D. F., Silver, P. A., and Way, J. C. (2010). Engineering cyanobacteria to synthesize and export hydrophilic products. Appl Env. Microbiol 76, 3462–6. doi: 10.1128/AEM.00202-10 Ogawa, T., Bao, D. H., Katoh, H., Shibata, M., Pakrasi, H. B., and Bhattacharyya-Pakrasi, M. (2002). A two-component signal transduction pathway regulates manganese homeostasis in Synechocystis 6803, a photosynthetic organism. J. Biol. Chem. 277, 28981–28986. doi: 10.1074/jbc.M204175200 Ortiz-Reyes, E., and Anex, R. P. (2022). Economic and environmental performance of non- cellulosic fermentable carbohydrates production for biofuels and chemicals. J. Clean. Prod. 353, 131526. doi: 10.1016/j.jclepro.2022.131526 Page-Sharp, M., Behm, C. A., and Smith, G. D. (1999). Involvement of the compatible solutes trehalose and sucrose in the response to salt stress of a cyanobacterial Scytonema species isolated from desert soils. Biochim. Biophys. Acta BBA - Gen. Subj. 1472, 519–528. doi: 10.1016/S0304-4165(99)00155-5 Pathak, J., Singh, P. R., Sinha, R. P., and Rastogi, R. P. (2021). “Evolution and distribution of cyanobacteria,” in Ecophysiology and Biochemistry of Cyanobacteria, ed. R. P. Rastogi (Singapore: Springer Nature), 1–30. doi: 10.1007/978-981-16-4873-1_1 Pereira, L., and Alves, M. (2012). “Dyes—Environmental impact and remediation,” in Environmental Protection Strategies for Sustainable Development, eds. A. Malik and E. Grohmann (Dordrecht: Springer Netherlands), 111–162. doi: 10.1007/978-94-007-1591- 2_4 Perez-Cenci, M., and Salerno, G. L. (2014). Functional characterization of Synechococcus amylosucrase and fructokinase encoding genes discovers two novel actors on the stage of cyanobacterial doi: 10.1016/j.plantsci.2014.04.003 sucrose metabolism. 95–102. Plant 224, Sci. Ponomarova, O., and Patil, K. R. (2015). Metabolic interactions in microbial communities: the Gordian knot. Curr. Opin. Microbiol. 27, 37–44. doi: untangling 10.1016/j.mib.2015.06.014 Porchia, A. C., Curatti, L., and Salerno, G. L. (1999). Sucrose metabolism in cyanobacteria: sucrose synthase from Anabaena sp. strain PCC 7119 is remarkably different from the plant 31 enzymes with respect to substrate affinity and amino-terminal sequence. Planta 210, 34– 40. doi: 10.1007/s004250050651 Porchia, A. C., and Salerno, G. L. (1996). Sucrose biosynthesis in a prokaryotic organism: presence of two sucrose-phosphate synthases in Anabaena with remarkable differences compared with the plant enzymes. Proc. Natl. Acad. Sci. 93, 13600–13604. doi: 10.1073/pnas.93.24.13600 Potocki de Montalk, G., Remaud-Simeon, M., Willemot, R.-M., Sarçabal, P., Planchot, V., and Monsan, P. (2000). Amylosucrase from Neisseria polysaccharea: novel catalytic properties. FEBS Lett. 471, 219–223. doi: 10.1016/S0014-5793(00)01406-X Potts, M., Slaughter, S. M., Hunneke, F.-U., Garst, J. F., and Helm, R. F. (2005). Desiccation tolerance of prokaryotes: Application of principles to human cells. Integr. Comp. Biol. 45, 800–809. doi: 10.1093/icb/45.5.800 Prabha, S., Vijay, A. K., Paul, R. R., and George, B. (2022). Cyanobacterial biorefinery: Towards economic feasibility through the maximum valorization of biomass. Sci. Total Environ. 814, 152795. doi: 10.1016/j.scitotenv.2021.152795 Qiao, C., Duan, Y., Zhang, M., Hagemann, M., Luo, Q., and Lu, X. (2018). Effects of reduced and enhanced glycogen pools on salt-induced sucrose production in a sucrose-secreting strain of Synechococcus elongatus PCC 7942. Appl. Environ. Microbiol. 84. doi: 10.1128/AEM.02023-17 Qiao, C., Zhang, M., Luo, Q., and Lu, X. (2019). Identification of two two‐component signal transduction mutants with enhanced sucrose biosynthesis in Synechococcus elongatus PCC 7942. J. Basic Microbiol. 59, 465–476. doi: 10.1002/jobm.201800676 Rafieenia, R., Atkinson, E., and Ledesma-Amaro, R. (2022). Division of labor for substrate utilization in natural and synthetic microbial communities. Curr. Opin. Biotechnol. 75, 102706. doi: 10.1016/j.copbio.2022.102706 Rai, A. N., Bergman, B., and Rasmussen, U. (2002). Cyanobacteria in symbiosis. Springer. Reed, R. H., and Stewart, W. D. P. (1985). Osmotic adjustment and organic solute accumulation in unicellular cyanobacteria from freshwater and marine habitats. Mar. Biol. 88, 1–9. doi: 10.1007/BF00393037 Rikkinen, J. (2015). Cyanolichens. Biodivers. Conserv. 24, 973–993. doi: 10.1007/s10531-015- 0906-8 Rikkinen, J. (2017). “Symbiotic Cyanobacteria in lichens,” in Algal and Cyanobacteria Symbioses, (World Scientific), 147–167. doi: 10.1142/9781786340580_0005 Rosic, N. (2021). “Molecular mechanisms of stress tolerance in cyanobacteria,” in Ecophysiology and Biochemistry of Cyanobacteria, ed. R. P. Rastogi (Singapore: Springer Nature), 131– 153. doi: 10.1007/978-981-16-4873-1_7 Sabri, S., Nielsen, L. K., and Vickers, C. E. (2013). Molecular control of sucrose utilization in Escherichia coli W, an efficient sucrose-utilizing strain. Appl. Environ. Microbiol. 79, 478–487. doi: 10.1128/AEM.02544-12 32 Sakkos, J. K., Santos-Merino, M., Kokarakis, E. J., Li, B., Fuentes-Cabrera, M., Zuliani, P., et al. (2023). Predicting partner fitness based on spatial structuring in a light-driven microbial community. PLOS Comput. Biol. 19, e1011045. doi: 10.1371/journal.pcbi.1011045 Santos‐Merino, M., Singh, A. K., and Ducat, D. C. (2021). "Sink engineering in photosynthetic microbes," in Cyanobacteria Biotechnology, eds. J. Nielsen, S. Lee, G. Stephanopoulos, and P. Hudson (Wiley-VCH), 171–209. doi: 10.1002/9783527824908.ch6 Santos-Merino, M., Torrado, A., Davis, G. A., Röttig, A., Bibby, T. S., Kramer, D. M., et al. (2021). Improved photosynthetic capacity and photosystem I oxidation via heterologous metabolism engineering in cyanobacteria. Proc. Natl. Acad. Sci. 118, e2021523118. doi: 10.1073/pnas.2021523118 Santos-Merino, M., Yun, L., and Ducat, D. C. (2023). Cyanobacteria as cell factories for the sucrose. Front. Microbiol. 14, 1126032. doi: photosynthetic production of 10.3389/fmicb.2023.1126032 Savakis, P., and Hellingwerf, K. J. (2015). Engineering cyanobacteria for direct biofuel production from CO2. Curr. Opin. Biotechnol. 33, 8–14. doi: 10.1016/j.copbio.2014.09.007 Shinde, S., Zhang, X., Singapuri, S. P., Kalra, I., Liu, X., Morgan-Kiss, R. M., et al. (2020). Glycogen metabolism supports photosynthesis start through the oxidative pentose phosphate pathway cyanobacteria. Plant Physiol. 182, 507–517. doi: 10.1104/pp.19.01184 in Singh, A. K., Santos-Merino, M., Sakkos, J. K., Walker, B. J., and Ducat, D. C. (2022). Rubisco regulation in response to altered carbon status in the cyanobacterium Synechococcus elongatus PCC 7942. Plant Physiol. 189, 874–888. doi: 10.1093/plphys/kiac065 Singh, S., Kumar, V., Shekar, S., Kapoor, D., Bhatia, D., Dhanjal, D. S., et al. (2021). “Phycoremediation of wastewater,” in Ecophysiology and Biochemistry of Cyanobacteria, ed. R. P. Rastogi (Singapore: Springer Nature), 269–289. doi: 10.1007/978-981-16-4873- 1_13 Smith, M. J., and Francis, M. B. (2017). Improving metabolite production in microbial co-cultures using a spatially constrained hydrogel. Biotechnol Bioeng 114, 1195–1200. doi: 10.1002/bit.26235 Song, H. S., Renslow, R. S., Fredrickson, J. K., and Lindemann, S. R. (2015). Integrating ecological and engineering concepts of resilience in microbial communities. Front. Microbiol. 6. doi: 10.3389/fmicb.2015.01298 Song, K., Tan, X., Liang, Y., and Lu, X. (2016). The potential of Synechococcus elongatus UTEX 2973 for sugar feedstock production. Appl. Microbiol. Biotechnol. 100, 7865–7875. doi: 10.1007/s00253-016-7510-z Su, Y., Song, K., Zhang, P., Su, Y., Cheng, J., and Chen, X. (2017). Progress of microalgae biofuel’s commercialization. Renew. Sustain. Energy Rev. 74, 402–411. doi: 10.1016/j.rser.2016.12.078 Suzuki, E., Ohkawa, H., Moriya, K., Matsubara, T., Nagaike, Y., Iwasaki, I., et al. (2010). Carbohydrate metabolism in mutants of the cyanobacterium Synechococcus elongatus 33 PCC 7942 defective in glycogen synthesis. Appl Env. Microbiol 76, 3153–9. doi: 10.1128/AEM.00397-08 Thiel, K., Patrikainen, P., Nagy, C., Fitzpatrick, D., Pope, N., Aro, E.-M., et al. (2019). Redirecting photosynthetic electron flux in the cyanobacterium Synechocystis sp. PCC 6803 by the deletion of flavodiiron protein Flv3. Microb. Cell Factories 18, 189. doi: 10.1186/s12934- 019-1238-2 Tóth, G. S., Siitonen, V., Nikkanen, L., Sovic, L., Kallio, P., Kourist, R., et al. (2022). Photosynthetically produced sucrose by immobilized Synechocystis sp. PCC 6803 drives biotransformation in E. coli. Biotechnol. Biofuels Bioprod. 15, 146. doi: 10.1186/s13068- 022-02248-1 Tsuge, Y., Kawaguchi, H., Sasaki, K., and Kondo, A. (2016). Engineering cell factories for producing building block chemicals for bio-polymer synthesis. Microb. Cell Factories 15, 1–12. doi: 10.1186/s12934-016-0411-0 Vadyvaloo, V., Smirnova, I. N., Kasho, V. N., and Kaback, H. R. (2006). Conservation of residues involved in sugar/H+ symport by the sucrose permease of Escherichia coli relative to lactose permease. J Mol Biol 358, 1051–9. doi: 10.1016/j.jmb.2006.02.050 Vargas, W. A., Nishi, C. N., Giarrocco, L. E., and Salerno, G. L. (2011). Differential roles of alkaline/neutral invertases in Nostoc sp. PCC 7120: Inv-B isoform is essential for diazotrophic growth. Planta 233, 153–162. doi: 10.1007/s00425-010-1288-5 Vargas, W. A., and Salerno, G. L. (2010). The Cinderella story of sucrose hydrolysis: Alkaline/neutral invertases, from cyanobacteria to unforeseen roles in plant cytosol and organelles. Plant Sci. 178, 1–8. doi: 10.1016/j.plantsci.2009.09.015 Vargas, W., Cumino, A., and Salerno, G. L. (2003). Cyanobacterial alkaline/neutral invertases. Origin of sucrose hydrolysis in the plant cytosol? Planta 216, 951–960. doi: 10.1007/s00425-002-0943-x Waditee-Sirisattha, R., and Kageyama, H. (2022). “Extremophilic cyanobacteria,” in Cyanobacterial Physiology, eds. H. Kageyama and R. Waditee-Sirisattha (Academic Press), 85–99. doi: 10.1016/B978-0-323-96106-6.00012-5 Wan, H., Wu, L., Yang, Y., Zhou, G., and Ruan, Y.-L. (2018). Evolution of sucrose metabolism: The dichotomy of invertases and beyond. Trends Plant Sci. 23, 163–177. doi: 10.1016/j.tplants.2017.11.001 Weiss, T. L., Young, E. J., and Ducat, D. C. (2017). A synthetic, light-driven consortium of cyanobacteria and heterotrophic bacteria enables stable polyhydroxybutyrate production. Metab. Eng. 44, 236–245. doi: 10.1016/j.ymben.2017.10.009 Wendisch, V. F., Jorge, J. M., Pérez-García, F., and Sgobba, E. (2016). Updates on industrial production of amino acids using Corynebacterium glutamicum. World J. Microbiol. Biotechnol. 32, 1–10. doi: 10.1007/s11274-016-2060-1 Werner, G. D. A., Strassmann, J. E., Ivens, A. B. F., Engelmoer, D. J. P., Verbruggen, E., Queller, D. C., et al. (2014). Evolution of microbial markets. Proc. Natl. Acad. Sci. U. S. A. 111, 1237–1244. doi: 10.1073/pnas.1315980111 34 Wu, S., Snajdrova, R., Moore, J. C., Baldenius, K., and Bornscheuer, U. T. (2021). Biocatalysis: enzymatic synthesis for industrial applications. Angew. Chem. Int. Ed. 60, 88–119. doi: 10.1002/anie.202006648 Xie, J., Cai, K., Hu, H.X., Jiang, Y.L., Yang, F., Hu, P.F., Cao, D.D., Li, W.F., Chen, Y. and Zhou, C.Z (2016). Structural analysis of the catalytic mechanism and substrate specificity of Anabaena alkaline invertase InvA reveals a novel glucosidase. J. Biol. Chem. 291, 25667– 25677. doi: 10.1074/jbc.M116.759290 Xu, Y., Tiago Guerra, L., Li, Z., Ludwig, M., Charles Dismukes, G., and Bryant, D. A. (2013). Altered carbohydrate metabolism in glycogen synthase mutants of Synechococcus sp. strain PCC 7002: Cell factories for soluble sugars. Metab. Eng. 16, 56–67. doi: 10.1016/j.ymben.2012.12.002 Yamaguchi, K., Suzuki, I., Yamamoto, H., Lyukevich, A., Bodrova, I., Los, D. A., et al. (2002). A two-component Mn2+-sensing system negatively regulates expression of the mntCAB operon in Synechocystis. Plant Cell 14, 2901–2913. doi: 10.1105/tpc.006262 Yu, J., Liberton, M., Cliften, P. F., Head, R. D., Jacobs, J. M., Smith, R. D., et al. (2015). Synechococcus elongatus UTEX 2973, a fast-growing cyanobacterial chassis for biosynthesis using light and CO2. Sci. Rep. 5, 8132. doi: 10.1038/srep08132 Zhang, L., Chen, L., Diao, J., Song, X., Shi, M., and Zhang, W. (2020). Construction and analysis of an artificial consortium based on the fast-growing cyanobacterium Synechococcus elongatus UTEX 2973 to produce the platform chemical 3-hydroxypropionic acid from CO2. Biotechnol. Biofuels 13. doi: 10.1186/s13068-020-01720-0 Zhang, S., Sun, H., Sun, J., Luo, Q., Luan, G., and Lu, X. (2021). “Engineering cyanobacteria cell factories for photosynthetic production of sucrose,” in Ecophysiology and Biochemistry of Cyanobacteria, ed. R. P. Rastogi (Singapore: Springer Nature), 373–399. doi: 10.1007/978-981-16-4873-1_17 Zhao, R., Sengupta, A., Whelan, R., Pinkerton, T., Menasalvas, J., Eng, T., et al. (2022). Photobiological production of high-value pigments via compartmentalized co-cultures using Ca-alginate hydrogels. Sci. Rep. 12, 22163. doi: 10.1038/s41598-022-26437-y Zhong, C. (2020). Industrial-scale production and applications of bacterial cellulose. Front. Bioeng. Biotechnol. 8, 605374. doi: 10.3389/fbioe.2020.605374 Zhu, H., Xu, L., Luan, G., Zhan, T., Kang, Z., Li, C., et al. (2022). A miniaturized bionic ocean- battery mimicking the structure of marine microbial ecosystems. Nat. Commun. 13, 5608. doi: 10.1038/s41467-022-33358-x Zorina, A., Sinetova, M. A., Kupriyanova, E. V., Mironov, K. S., Molkova, I., Nazarenko, L. V., et al. (2016). Synechocystis mutants defective in manganese uptake regulatory system, ManSR, are hypersensitive to strong light. Photosynth. Res. 130, 11–17. doi: 10.1007/s11120-015-0214-x Zuñiga, C., Li, T., Guarnieri, M. T., Jenkins, J. P., Li, C. T., Bingol, K., et al. (2020). Synthetic microbial communities of heterotrophs and phototrophs facilitate sustainable growth. Nat. Commun. 11. doi: 10.1038/s41467-020-17612-8 35 CHAPTER 2: ENHANCEMENT OF CYANOBACTERIAL SUCROSE BIOSYNTHESIS THROUGH CULTIVATION METHODS This chapter was adapted from text originally published in: Yun, L. Zegarac, R., and Ducat, D.C. (2024) Impact of irradiance and inorganic carbon availability on heterologous sucrose production in Synechococcus elongatus PCC 7942. Front. Plant Sci. 15: 1378573. doi: 10.3389/fpls.2024.1378573 Copyright, © 2024 Frontiers Media — Reproduced with permission. 36 Introduction There is a need for sustainable carbohydrate feedstocks as we increasingly look to biotechnology and green chemistry approaches for energy and other commodity chemicals. Sucrose is one significant feedstock for bio-ethanol (Balat and Balat, 2009) and is a promising building block for many value-added chemicals (Peters et al., 2010). However, sucrose is typically produced by plant crop species (e.g., sugar beet and sugar cane) (Peters et al., 2010), giving rise to ethical concerns about diverting arable land and potable water from food production to fuel and chemicals (Graham-Rowe, 2011; Lang et al., 2017). Cellulosic biomass is another promising alternative source of fermentable carbohydrates that is under active study, yet the high costs of converting lignocellulosic materials into simple carbohydrates suitable for fermentation is currently prohibitive (Graham-Rowe, 2011; Ziolkowska, 2014). For these reasons, an alternative source for sucrose that circumvents the controversies and complexities of plants is desirable. Cyanobacteria are a diverse phylum of photosynthetic prokaryotes that are under consideration as an alternative carbohydrate feedstock species due to their ability to hyperaccumulate soluble sugars under certain growth conditions (Hays and Ducat, 2015). Moreover, cyanobacteria are less likely to compete for land and water resources with food crops, are faster growing, relatively easier to genetically engineer, and have a higher photosynthetic efficiency than land plant sources of carbohydrate feedstocks (Parmar et al., 2011; Satta et al., 2023). One promising approach for cyanobacterial sucrose production involves the heterologous expression of sucrose permease (cscB), which was originally reported to allow up to ~80% of fixed carbon to be exported in the form of sucrose when expressed in the model cyanobacterium, Synechococcus elongatus PCC 7942 (Ducat et al., 2012). The high productivity of such strains is in part due to an emergent property whereby photosynthetic activity is increased: exporting sucrose increases the quantum efficiency of photosystem II, rate of photosystem II oxygen evolution, and carbon fixation rate (Ducat et al., 2012; Abramson et al., 2016; Santos-Merino et al., 2021, 2023; Wang et al., 2023). Additional genetic modifications have been employed to increase sucrose export by increasing carbon flux to the sucrose synthesis pathway and/or decreasing flux to potential competing pathways, with mixed success (Santos-Merino et al., 2023). For example, multiple groups have shown that sucrose productivity can be enhanced by artificially over- expressing sucrose phosphate synthase (sps), an enzyme involved in a rate-limiting step of sucrose synthesis, in the background of sucrose exporting lines (Ducat et al., 2012; Duan et al., 2016; Hays 37 et al., 2017; Qiao et al., 2018). Similar strategies to enable sucrose export have been used across a variety of cyanobacterial species; cscB has been expressed in S. elongatus UTEX 2973 (Lin et al., 2020; Zhang et al., 2020), Synechococcus sp. PCC 7002 (Xu et al., 2013; Han et al., 2023), and Synechocystis sp. PCC 6803 (Du et al., 2013; Thiel et al., 2019). Different cyanobacterial species exhibit varying rates of sucrose productivity (Santos-Merino et al., 2023), though it is often uncertain if inter-species differences are due to genetics, distinct conditions utilized for laboratory growth, metabolic requirements, tolerances for environmental stresses, or inherent photosynthetic capacities (Billis et al., 2014; Ungerer et al., 2018b; Cassier-Chauvat et al., 2021; Adomako et al., 2022). For instance, S. elongatus PCC 7942 and S. elongatus UTEX 2973 share 99.9% genome similarity, yet the latter has significantly faster growth rates, higher light and temperature tolerances, and has been reported to have a significantly higher sucrose productivity (Yu et al., 2015; Ungerer et al., 2018b; Lin et al., 2020; Adomako et al., 2022). A handful of single nucleotide polymorphisms appear to significantly contribute to enhanced tolerance of light and heat in S. elongatus UTEX 2973 (Ungerer et al., 2018b), and these strains also exhibit substantially faster growth rates (2.5 hour doubling time compared to 7-8 hours). Therefore, it is possible that the increased sucrose productivity of engineered S. elongatus UTEX 2973 strains may be due to an inherently improved photosynthetic capacity. Alternatively, other factors like higher routine temperature, light growth conditions or altered partitioning of fixed carbon may contribute to the productivity of S. elongatus UTEX 2973 (Mueller et al., 2017; Hendry et al., 2019). In this study, we set out to systematically assess how different laboratory equipment and growth conditions impact the efficacy of sucrose production, with a particular focus on light and inorganic carbon availability as two critical factors that may impact sucrose secretion rates, and that could contribute (in part or in full) to the differences in sucrose productivity observed across different laboratories and cyanobacterial species. The interest in exploring cyanobacteria as an alternative carbohydrate feedstock has led to many recent reports documenting sucrose production, yet, as we recently reviewed (Santos-Merino et al., 2023), there can be considerable variation in productivity between different publications, even when the species and genetic modifications are identical. We explored a variety of commercial and custom-built cyanobacterial growth chambers, and report cultivation conditions and bioproduction values in standardized units to facilitate cross- comparisons between laboratories. We find a direct relationship between irradiance used for 38 cyanobacterial cultivation and sucrose productivity, especially when inorganic carbon availability was not limiting. Our results may be valuable for the broader community of scientists working with different strains of cyanobacteria engineered as potential alternative carbohydrate feedstocks, and especially point to limitations imposed on productivity caused by inefficient CO2 delivery methods. Additionally, our results have implications for the degree of variation in sucrose productivity of related cyanobacterial species, which may be significantly influenced by cultivation conditions rather than solely due to inherent metabolic and physiological limitations of slower-growing species. Results Assessing high-light tolerance in S. elongatus We first selected a strain of S. elongatus that has been engineered for sucrose secretion in a salt-independent manner. Briefly, hyperaccumulation of cytosolic sucrose is a common strategy for osmotic stress protection in many freshwater cyanobacterial species (Klähn and Hagemann, 2011), where this compatible solute can act as a counter osmoticum while also conferring other cellular protective effects. A rate-limiting step of sucrose biosynthesis occurs at the enzyme sucrose phosphate synthase (SPS), that catalyzes the condensation of NDP-glucose with fructose 6-phosphate to form sucrose 6-phosphate. SPS enzymes are frequently encoded as a bidomainal protein that also contains sucrose 6-phosphate phosphatase (SPP) activity that dephosphorylates this intermediate to sucrose. We and others have found that expression of sucrose permease (cscB), a symporter of sucrose and protons, will lead to the export of cytosolic sucrose to the medium (Ducat et al., 2012; Santos-Merino et al., 2023). In S. elongatus strains with inducible copies of both sps and cscB, sucrose synthesis and export can be achieved in the absence of external osmotic pressure (Abramson et al., 2016, 2018; Lin et al., 2020; Dan et al., 2022). We therefore selected this strain to minimize the requirement for the addition of salt to the growth medium. We first used the MC-1000 multi-cultivator to evaluate the impact of varied light intensity (150, 250, 500, 1000, 2000 µmol photons m-2 s-1) on growth and sucrose production of S. elongatus PCC 7942 cscB/sps strains and to broadly assess light tolerance. The growth of wild-type (WT) strains cultured with 150 and 1000 µmol photons m-2 s-1, and non-exporting cscB/sps with 1000 µmol photons m-2 s-1 grew at similar rates, reaching a final density of ~2.2 OD750 (Figure 2.1A, left). Sucrose-exporting strains, however, can partition a large fraction of their fixed carbon to secreted sucrose, leading to less cellular growth and biomass under certain conditions (Ducat et 39 Figure 2.1. Growth and sucrose titers of MC-1000 cultures. Growth and sucrose yields of WT and cscB/sps-expressing S. elongatus grown in the MC-1000 under various light intensities. (A) The OD750 of non-sucrose exporting (left) and exporting (center and right) strains, and (B) sucrose production of cultures started at low cell density (~0.3 OD750). Brackets in the legend denote significant differences (p < 0.05) at 72 h by one-way ANOVA in (A) all cultures and (B) induced cultures. (C) The OD750 and (D) sucrose production of cultures started at high cell densities (~1.5 and 2 OD750). Error bars denote standard deviation. 40 al., 2012) (see also, Figure S2.3). The highest OD750 amongst sucrose exporters were from cultures grown with moderate irradiances of 250 and 500 µmol photons m-2 s-1 (Figure 2.1A, right), though differences were not statistically significant from other exporting conditions (p > 0.05). For sucrose-secreting lines in the MC-1000, there was a direct correlation between illumination and both cell growth and sucrose production up to 500 µmol photons m-2 s-1; light intensities beyond this threshold led to a decline in productivity (Figure 2.1A, B). Sucrose production over 72 hours peaked at 0.56 g L-1 for 250 µmol photons m-2 s-1 and 0.68 g L-1 for 500 µmol photons m-2 s-1 (p < 0.001 against all other light conditions) (Figure 2.1B). The relationship between irradiance and growth, however, is abolished when strains are not induced to export sucrose, and growth rates were relatively unaffected by illumination between 150 and 1000 µmol photons m-2 s-1 (Figure 2.1A, left). These data suggest that high light may indeed be harmful, and the increased photosynthetic capacity provided by a heterologous sink (i.e., sucrose export) is not protective against light stress. However, it appears that S. elongatus PCC 7942 can remain resilient when not imposed with the additional burden of producing and secreting sucrose. To evaluate if the sucrose productivity in these experiments was limited by a low-starting density, we increased the starting inoculation density. Since the effects of self-shading within higher density cultures lead to an effectively lower irradiation under the same outside illumination, we utilized brighter lights for our high-density starter cultures of 1.5 and 2.0 OD750, the upper range of OD750 the MC-1000 was able to support. To select the irradiances for these dense cultures, we revisited the most successful light intensity in our previous experiment. Cultures of 0.3 OD750 exposed to 500 µmol photons m-2 s-1 light receive ~300 µmol photons m-2 s-1 light when measured 1 cm from the surface (Figure S2.4). Application of 2000 and 2500 µmol photons m-2 s-1 to cultures of 1.5 and 2.0 OD750, respectively, would achieve similar internal light intensities (Figure S2.4). We observed limited growth within the first 24 h of these cultures, but the density of the culture plateaued or declined in the following day to values statistically similar to starting turbidities (p > 0.05) (Figure 2.1C). Cultures grown with 2000 µmol photons m-2 s-1 produced 0.23 g L-1 sucrose within the first day, and had a final sucrose content of 0.35 g L-1 (p > 0.4); cultures grown with 2500 µmol photons m-2 s-1 produced 0.22 and 0.28 g L-1 sucrose at 24 and 48 h (p = 1.0), respectively (Figure 2.1D). Notably, the level of sucrose production from high density cultures was only marginally improved (~15%) at 48 h relative to cultures seeded at a lower density grown with 2000 µmol photons m-2 s-1 (p = 0.002) (Figure 2.1B, D). 41 Optimizing high cell density cultivation with increased irradiance and CO2 Figure 2.2. Growth and sucrose titers of HDC cultures grown with different media. Growth and sucrose production of induced and uninduced cscB/sps-expressing S. elongatus grown in the CellDEG HDC cultivator system at moderate light intensity (~250 μmol photons m-2 s-1). The OD750 and sucrose production, respectively, of cultures grown in (A, B) BG-11 medium, and (C, D) CD medium. (E) Total fixed carbon for induced strains in BG-11 and CD media as shown by sucrose and dry cell weight accumulated per day. Error bars denote standard deviation. Asterisks denote significant differences between induction conditions at the indicated timepoint (p < 0.005) 42 Figure 2.2. (cont’d) by independent samples t-test. Daggers denote significant differences in induced cultures between (A, B) and (B, C) conditions (p < 0.001) by independent samples t-test. Curiously, both the cellular growth rate and sucrose productivity were relatively modest in the MC-1000 in comparison to values reported for similar strains by our group and others (Santos- Merino et al., 2023), suggesting that these cultures might be limited by another environmental variable (Figure 2.1A, B). Although 3% CO2 was bubbled into the cultivation tubes for these experiments, the method of delivery (sparging via a large-bore needle) may not be the most efficient compared to other methods for enriching inorganic carbon. Consistent with a possible inorganic carbon limitation, sucrose productivity was slightly, but significantly, enhanced in MC- 1000 cultures supplemented with 10 mM HCO3 - (Figure 2.1B) relative to cultures without bicarbonate when grown with 500, 1000, and 2000 µmol photons m-2 s-1 (p = 0.038, 0.020, 0.028, respectively) (Figure S2.5). However, we were unable to overcome any potential limitation of inorganic carbon in dense cultures through the supplementation of additional HCO3 - (Figure S2.6). To more directly evaluate the hypothesis that carbon limitation was limiting for growth and sucrose productivity in the MC-1000, we utilized the High-Density Cultivation (HDC) system (CellDEG GmbH, Germany). These cultivation vessels are designed to deliver a saturating level of inorganic carbon through a permeable membrane with a large surface area (Lippi et al., 2018). Using an identical growth medium (BG-11), a significant enhancement of cell growth and sucrose production was observed in the CellDeg HDC system relative to the MC-1000 with comparable illumination (250 μmol photons m-2 s-1). Strains that were not induced to export sucrose grew to 5.3 OD750 over 72 h while producing a marginal amount of sucrose (0.02 g L-1). Strains induced to secrete sucrose grew to 4.6 OD750 and produced 1.0 g L-1 sucrose (Figure 2.2A, B). Curiously, under these conditions of high inorganic carbon availability, we observed little, but significant, difference in growth rate in the sucrose-exporting strains despite a significant rate of carbon export in the form of sucrose (Figure 2.2A, B, E). Since nutrients in BG-11 medium can become limiting at higher density, we next evaluated an enriched medium (CD) designed to support high-density cultures, such as those achievable in the HDC system (Dienst et al., 2020). When cultures were grown in CD medium, non-exporting strains reached 11.1 OD750 over the course of 5 days (Figure 2.2C), while also secreting a disproportionally high level of sucrose for an uninduced culture (Figure 2.2D). Strains induced to export sucrose were significantly stunted in growth after 120 h (3 OD750; Figure 2.2C) and secreted 43 sucrose at a higher rate. Increasing light and CO2 availability to enhance sucrose synthesis Figure 2.3. Growth and sucrose titers of HDC cultures grown with varying irradiances. Growth and sucrose production of induced and uninduced cscB/sps-expressing S. elongatus grown in the HDC system with moderately high and high-light intensity (MH and H, respectively) and CD medium. The OD750 and supernatant sucrose concentration of cultures grown with (A, B) 500 μmol photons m-2 s-1 for 0-120 h, and (C, D) 500 μmol photons m-2 s-1 for 0-48 h and 1000 μmol photons m-2 s-1 for 48-120 h. Error bars denote standard deviation. Asterisks denote significant differences between induction conditions at the indicated timepoint (p < 0.001) by independent 44 Figure 2.3 (cont’d) samples t-test. Daggers denote significant differences in induced cultures between (A, B) and (C, D) conditions (p < 0.005) by independent samples t-test. Environmental conditions enriched with nutrients and inorganic carbon supported substantially higher cyanobacterial production with increased irradiation. Cultivating cscB/sps strains with high CO2 (HDC) and with 500 μmol photons m-2 s-1, non-exporting and sucrose- exporting strains reached OD750 of 15.8 and 5.2, respectively (Figure 2.3A). Furthermore, strains induced to secrete sucrose accumulated up to 3.3 g L-1 over the course of 5 days, while uninduced cultures accumulated 1.15 g L-1 (Figure 2.3B). Further enhancements in bioproduction could be achieved by further increasing light intensity to 1000 μmol photons m-2 s-1 at 48 h when cultures were more turbid: sucrose-secreting strains reached an OD750 of 6.0 and sucrose yields of 3.8 g L- 1 (Figure 2.3C, D). Uninduced strains under these growth conditions achieved 18.4 OD750 and secreted a surprisingly high amount of sucrose, reaching 2.3 g/L sucrose at the end of 120 h. Discussion Considerable variation in the productivity of cyanobacteria engineered to secrete sucrose has been reported in the literature (Table 1.1), and this variability has significant implications for the viability of biotechnological applications. Obviously, light is a crucial component of photosynthetically driven cultivation, yet it is also well-established that over-irradiation of cyanobacteria can lead to cell damage and/or death if surplus light energy is not safely dissipated (Montgomery, 2014). S. elongatus PCC 7942 has been often described in the literature as a strain that prefers a relatively low-light, and many groups routinely cultivate this strain ≤100 µmol photons m-2 s-1. Our results show that light and CO2 availability strongly impact the bioproductivity of sucrose-secreting cyanobacterial strains, independent of the genetic background. Under conditions of high nutrient and CO2 availability, cell biomass and sucrose secretion rates are positively correlated with irradiation, up to 1000 μmol photons m-2 s-1 (Figures 2, 3): a light intensity regarded as photodamaging for S. elognatus in some literature. The light tolerance we observe is consistent with recent publications reporting that the maximum growth rate of S. elongatus PCC 7942 is achieved with 400 µmol photons m-2 s-1 (Takatani et al., 2015; Ungerer et al., 2018a). Further complication of S. elongatus’s light tolerance is that self-shading in dense cultures or chambers with long optical cross-sections can rapidly attenuate effective light availability (Andersson et al., 2019). Our results suggest that differences in cyanobacterial cultivation conditions likely can account for a significant portion of variability (up to 3- to 10-fold) 45 in reported sucrose production values, even when solely accounting for strains grown in common, commercially available photobioreactors. We observe that the mechanism of CO2 delivery strongly impacts the bioproduction potential for sucrose-secreting cyanobacteria at higher light intensities. In the MC-1000 cultivator, CO2 is delivered via a bubble column-style sparging system, and we observed indications of carbon limitation despite continual sparging with 3% CO2 gas. In the MC-1000, a positive correlation for production was observed with increasing illumination up to ≤ 500 μmol photons m-2 s-1 (Figure 2.1A, B). Interestingly, even the highest light conditions were not lethal (Figure 2.1A; 2000 µmol photos m-2 s-1 is roughly equivalent to full, midday sunlight), although S. elongatus PCC 7942 is frequently described in the literature as a low-light tolerant strain (e.g., Ungerer et al., 2018b; Takatani et al., 2022; Walker and Pakrasi, 2022). Furthermore, the final OD750 and sucrose productivities of strains cultivated in the MC-1000 system were substantially lower than values reported by our group and others with the same growth medium in other bioreactor designs (Abramson et al., 2016, 2018; Qiao et al., 2018). One possible explanation may be that inorganic carbon availability could be limiting, despite the continual sparging of 3% CO2 delivered via a stream of CO2 bubbles through a submerged needle design (Markou et al., 2014). When cultures were grown under conditions with greater inorganic carbon and/or nutrient availability, increasing illumination resulted in an even greater boost in productivity. The HDC system uses cultivators with a gas-permeable bottom inserted into a carrier plate that exposes it to the CO2 headspace created by a HCO3 - buffer reservoir; the device can generate up to 8% CO2 under the conditions outlined in this work (CellDEG GmbH, 2020). Using the HDC system, uninduced cscB/sps strains achieved an OD750 of 5.3, while induced strains reached an OD750 of 4.6 and sucrose yield of 1.0 g L-1 when grown in BG-11 medium (Figure 2A, B). For sucrose- exporting strains, this amounts to 245% higher OD750 and 188% more sucrose than cultures grown in the MC-1000 with the same medium and irradiance (250 µmol photons m-2 s-1). Importantly, under conditions of saturating inorganic carbon, we do not observe a “cost” for sucrose secretion, as indicated by the comparable levels of cellular biomass even when substantial sucrose is secreted (Figure 2.2E). This suggests that growth does not have to be sacrificed when inorganic carbon is not a limiting factor, assuming that the fraction of fixed carbon diverted to sucrose production does not exceed a threshold value. By comparison, when grown in CD medium, the OD750 of non- exporting strains rose to an average of 11.1, but the OD750 of sucrose-secreting strains dropped to 46 ~3.0 at the end of the 120 h experimental period (Figure 2.2C). We attribute this loss of cell biomass in cultures induced to secrete sucrose to the higher osmotic pressure of CD medium. At ~140 mOsm, CD medium has an osmolarity that is approximately 3-fold higher than that of BG- 11 alone and can provide osmotic pressure equivalent to BG-11 supplemented with ~50 mM NaCl. The strain in this study produces sucrose utilizing a heterologous salt-independent SPS; however, increased salt or osmotic stress engages the endogenous salt-dependent sucrose production pathways and has been shown to lead to a higher diversion of fixed carbon towards sucrose biosynthesis (Ducat et al., 2012; Kirsch et al., 2019; Liang et al., 2020). Sucrose exporters grown in CD medium produced 139% more sucrose than their BG-11 counterparts by 72 hours, and a total of 1.8 g L-1 sucrose by 120 h (Figure 2.2D). Non-exporting strains in CD medium produce more sucrose than in BG-11 as well; this can also be explained by the medium exerting osmotic stress, and thus activating the native SPS. To potentially combine the productivity gains of cultivating with high light with those of high CO2 availability, we required a separate light source capable of reaching intensities beyond those of most commercial photobioreactors. For this purpose, we created a custom LED light source in-house consisting of four COB LEDs used for growth of photosynthetic organisms; when placed above the HDC and shaker, there is generally an equitable distribution of light between the six cultivators. By cultivating strains in the HDC system with CD medium and 250 µmol photons m-2 s-1, growth and sucrose yield increased in both induced and uninduced cscB/sps strains. By doubling irradiance to 500 µmol photons m-2 s-1, OD750 and sucrose yield increased by 173% and 182%, respectively, in cscB/sps expressing strains (Figure 2.3A, B). Non-sucrose exporting strains underwent a 142% and 251% increase in growth and sucrose, respectively. Finally, a further increase in culture productivity could be achieved by both increasing light intensity to 1000 µmol photons m-2 s-1 while taking advantage of the self-shading of higher density cultures. By doubling the light intensity of cultures that were grown with 500 µmol photons m-2 s-1 at 48 h (average of 2.3 OD750), cell growth increased by 17% (p < 0.001) and sucrose productivity increased by 15% (p < 0.005) (Figure 2.3C, D). The highest sucrose productivity we report was reached under these conditions, at 47.8 mg L-1 h-1. Surprisingly, sucrose bioproduction was nearly as strong in strains not induced to express cscB/sps (highest productivity observed at 45.3 mg L-1 h-1), likely due to a combination of very high cell density (OD750 up to 18.3) and induction of endogenous sucrose production via native osmotic-responsive pathways (Figure 2.3A, B). This latter observation may 47 suggest an alternative route for high sucrose production from late-phase, dense cultures with only basal levels of activation of the sucrose export pathway. Conclusions In this study we assessed how irradiance, inorganic carbon availability, and cultivation method affect the growth and sucrose production of S. elongatus PCC 7942. We found that increasing light intensity up to 1000 µmol photons m-2 s-1 has a positive effect on sucrose production, and higher irradiances can be tolerated if cultures grow into sufficient turbidity. By adjusting light regimes and increasing CO2 availability, S. elongatus cscB/sps reached a productivity of 47.8 mg sucrose L-1 h-1, which is comparable to or exceeds other reported strains, including some metabolic engineering interventions that can negatively affect cell health. Our survey underscores the importance of cultivation conditions on sucrose productivity and may partially explain the variability observed across different institutions using similar strains. These results may contribute insights into future optimization of cyanobacterial bioproduction efforts. Methods and Materials Strains and cultivation conditions The base strain of cyanobacterium we utilize for our sucrose measurements is constructed as reported in Abramson et al., 2016. Briefly, this strain contains a copy of sucrose permease (cscB) under an IPTG-inducible Ptrc promoter installed into Neutral Site 3. An additional copy of sucrose phosphate synthase (sps) cloned from Synechocystis sp. PCC 6803 is also expressed under Ptrc and is integrated into the genome at Neutral Site 2. Inducible sps allows us to bypass the requirement for external salt for sucrose production that is in many sucrose-secreting strains. For results from WT strains, we use the ATCC strain of S. elongatus PCC 7942 (ATCC #33912). S. elongatus PCC 7942 strains were routinely grown in BG-11 medium (C3061, Sigma- Aldrich, USA) supplemented with 1 g L-1 HEPES and set to pH 8.3 with NaOH to maintain or expand cultures. Cultures were incubated in a Multitron Infors HT incubator at 32°C with 2% CO2, 150 rpm shaking, and ~150 μmol m-2 s-1 of continuous light from Sylvania 15 W Gro-Lux fluorescent bulbs. To standardize the growth history of strains between different experiments, all cultures were back-diluted 1:10 into BG-11 for three days prior to initiating any of the reported experiments. When needed, 1 mM IPTG was added to induce the expression of cscB and sps genes at time = 0 h. Chloramphenicol (Cm; 25 μg mL-1) and kanamycin (Kn; 50 μg mL-1) were used to maintain cscB and sps, respectively. Antibiotics were not added on the final day of back-dilution 48 to minimize confounding effects. Examining light tolerance and sucrose production in MC-1000 multi-cultivators The MC-1000 multi-cultivator system (Photon System Instruments, Czech Republic) was used to examine the light tolerance of S. elongatus; the device emits cool white light at varying light intensities (150, 250, 500, 1000, 2000, or 2500 μmol photons m-2 s-1). MC-1000 cultures were kept at 32°C and provided with 3% CO2. Overnight cultures were back-diluted to an OD750 of ~0.3 into MC-1000 vessels with fresh BG-11 medium supplemented with 10 mM HCO3 -, and induced with 1 mM IPTG as needed for low-starting density experiments. Samples were taken at 0, 24, 48, and 72 h for OD750 and sucrose measurements. For the experiments specifically evaluating high-light with high-starting density cultivation in the MC-1000 system, overnight cultures (prepared as outlined in section 2.1) were pelleted in a benchtop centrifuge. The supernatant was removed, and the cell pellet was resuspended with fresh BG-11 medium supplemented with 10 mM HCO3 - to an OD750 of ~1.5 or 2.0 (provided 2000 and 2500 μmol photons m-2 s-1, respectively). An additional 10 mM HCO3 - was added at 24 h for vessels that were provided bicarbonate daily. Samples were taken at 0, 24, and 48 h for OD750 measurements and sucrose quantification. Sucrose production in high-density multi-cultivators The High-Density Cultivation (HDC) 6.10B starter kit (CellDEG GmbH, Germany) was used for high-CO2 and high-density cultivation. The kit consists of a bottom base with a buffer reservoir and 10 mL cultivators with a gas permeable bottom membrane. At the start of the experiments, the reservoir was filled with 200 mL of a 3 M KHCO3/ 3 M K2CO3 solution to provide CO2. Cultures prepared in BG-11 were centrifuged for 10 min at 3700 xg and 22°C before removing the supernatant and resuspending to an OD750 of ~0.3 with CD medium. CD medium is designed to sustain extended periods of high-density cultivation of photosynthetic microbes, and has higher concentrations of macronutrients (N, P, and K) and micronutrients (Mg, S, Ca, Na, Cl, Fe, Zn, and Mo) relative to BG-11. Due to its enrichment, CD medium has a high osmolarity (~140 mOsm) that is approximately three times greater than of BG-11 (~40 mOsm); the exact composition is available on protocols.io (dx.doi.org/10.17504/protocols.io.2bxgapn). Relevant cultures were induced with 1 mM IPTG before transferring 12 mL to the HDC system cultivators. Moderate-light HDC experiments were incubated in a Multitron incubator (Infors HT, Switzerland) providing ~150 rpm shaking and ~250 μmol photons m-2 s-1 from Sylvania 15 W 49 Gro-Lux fluorescent bulbs. High-light HDC experiments were incubated in a growth chamber (AL-36L4, Percival Scientific, USA) with either ~500 μmol photons m-2 s-1 for 0-120 h, or ~500 μmol photons m-2 s-1 for 0-48 h and ~1000 μmol photons m-2 s-1 for 48-120 h from a light array fabricated in-house consisting of four equidistant chip-on-board (COB) LED lights (BXRA- 56C5300-H, Bridgelux, USA) (Figure S2.1). Both incubators provided 2% ambient CO2, constant illumination, and ~150 rpm shaking. Samples were taken at 0, 24, 48, 72, 96, and 120 h to measure OD750 and to collect supernatant for sucrose quantification. Measuring light penetration through culture depth The following experimental configuration was used to measure the relationship between surface irradiance, light penetration, and culture turbidity: a glass graduated cylinder with a clear bottom was clamped onto a ring stand. A Submersible Spherical Micro Quantum Sensor (US- SQS/L; WALZ, Germany) connected to a LI-250A light meter (LI-COR, USA) was clamped above and in-center of the graduated cylinder. A single COB LED from the in-house lighting device was placed below and center of the graduated cylinder. With this setup, 5 cm of liquid (BG- 11 or cell culture at the indicated OD750) fill the graduated cylinder, and the sensor measures the light that penetrates through the liquid when 0, 1, 2, or 3 cm away from the light interface of the culture (0 cm). A black, opaque sheet was wrapped around the graduated cylinder to cover all but the bottom 4 cm of the culture to block ambient light. To set the desired irradiance (150, 500, 1000, 2000, 3000, or 4000 µmol photons m-2 s-1), the graduated cylinder was filled with 5 cm of BG-11, and the sensor was placed flush to the bottom of the glass to measure the incoming light at 0 cm. WT cultures were prepared as outlined in Section 2.1, pelleted, and supernatants removed. Pellets were resuspended with fresh BG-11 to create a concentrate that was diluted with BG-11 to 0.3, 1.0, 2.0, 5.0, and 10.0 OD750. Five mL of each culture filled the graduated cylinder and penetrating irradiances at 1, 2, and 3 cm were recorded at applicable light intensities. For this and setting the irradiance, the average function of the LI-250A was used in which the device averages the values recorded over a 15-second interval. Sucrose quantification Samples collected at the time points specified above were pelleted in a microcentrifuge; the supernatants were saved and stored at -20°C until the end of the experiment. Sucrose quantification was performed using the Sucrose/D-Glucose Assay Kit (K-SUCGL; Megazyme, USA). 50 Dry cell weight determination To determine the dry cell weight of S. elongatus PCC 7942 strains, cultivation conditions were recreated as outlined in Section 2.2 for low-starting density experiments in the MC-1000, and Section 2.3 for moderate-light experiments in the HDC. Samples were collected at a variety of culture densities (OD750) in order to generate a broad distribution for standard curves correlating OD750 with dry cell biomass (Figure S2.2). The OD750 of collected samples was determined with a visible spectrophotometer (Genesys 20; Thermo Fisher Scientific, USA), washed with Milli-Q water twice, and resuspended into the desired OD750 with Milli-Q water. Three to 20 mL of the prepared samples were pelleted in a Falcon tube and most of the supernatant removed; the remaining water was used to create a concentrated resuspension to transfer to a pre-weighed 3 mL glass test tube. Milli-Q water was used as needed to rinse the Falcon tube and added to the glass test tube. Samples were dried in a 90°C oven for 24 h, or until the mass remained stable. The mass of the empty glass test tube was subtracted from the final weight, and the resulting value was divided by the volume used to determine the dry cell weight (g L-1) for corresponding OD750. Python was used to create a linear regression model and derive the function parameters, which was then used to convert OD750 into dry cell weight. ACKNOWLEDGEMENTS We thank the Ducat Lab for their helpful comments on the manuscript, the Kerfeld Lab for the use of equipment, and the staff of the Plant Research Laboratory for their technical assistance. The research reported in this manuscript was primarily supported by the National Science Foundation and the Division of Molecular and Cellular Bioscience (Grant: 1845463) to D.C.D. Fellowship and training support provided by a fellowship from the Plant Biotechnology for Health and Sustainability Training Program at Michigan State University (Grant: NIH T32-GM110523). Additional institutional and facilities support from the Office of Science, Office of Basic Energy Sciences of the U.S. Department of Energy (DOE) under grant DE-FG02-91ER2002. 51 REFERENCES Abramson, B. W., Kachel, B., Kramer, D. M., and Ducat, D. C. (2016). Increased photochemical efficiency in cyanobacteria via an engineered sucrose sink. Plant Cell Physiol. 57, 2451– 2460. doi: 10.1093/pcp/pcw169 Abramson, B. W., Lensmire, J., Lin, Y. T., Jennings, E., and Ducat, D. C. (2018). Redirecting carbon to bioproduction via a growth arrest switch in a sucrose-secreting cyanobacterium. Algal Res. 33, 248–255. doi: 10.1016/j.algal.2018.05.013 Adomako, M., Ernst, D., Simkovsky, R., Chao, Y.-Y., Wang, J., Fang, M., et al. (2022). Comparative Genomics of Synechococcus elongatus Explains the Phenotypic Diversity of the Strains. mBio 13, e00862-22. doi: 10.1128/mbio.00862-22 Andersson, B., Shen, C., Cantrell, M., Dandy, D. S., and Peers, G. (2019). The Fluctuating Cell- Specific Light Environment and Its Effects on Cyanobacterial Physiology. Plant Physiol. 181, 547–564. doi: 10.1104/pp.19.00480 Balat, M., and Balat, H. (2009). Recent trends in global production and utilization of bio-ethanol fuel. Appl. Energy 86, 2273–2282. doi: 10.1016/j.apenergy.2009.03.015 Billis, K., Billini, M., Tripp, H. J., Kyrpides, N. C., and Mavromatis, K. (2014). Comparative Transcriptomics between Synechococcus PCC 7942 and Synechocystis PCC 6803 Provide Insights doi: into Mechanisms 10.1371/journal.pone.0109738 of Stress Acclimation. PLoS ONE 9. Cassier-Chauvat, C., Blanc-Garin, V., and Chauvat, F. (2021). Genetic, Genomics, and Responses to Stresses in Cyanobacteria: Biotechnological Implications. Genes 12, 500. doi: 10.3390/genes12040500 CellDEG GmbH (2020). High-Density Cultivation System HDC 6.10B User Manual. (Berlin, Germany: CellDeg GmbH). Dan, Y., Sun, J., Zhang, S., Wu, Y., Mao, S., Luan, G., et al. (2022). Manipulating the Expression of Glycogen Phosphorylase in Synechococcus elongatus PCC 7942 to Mobilize Glycogen Storage doi: for 10.3389/fbioe.2022.925311 Synthesis. Front. Bioeng. Biotechnol. Sucrose 10. Dienst, D., Wichmann, J., Mantovani, O., Rodrigues, J. S., and Lindberg, P. (2020). High density cultivation for efficient sesquiterpenoid biosynthesis in Synechocystis sp. PCC 6803. Sci. Rep. 10, 5932. doi: 10.1038/s41598-020-62681-w Du, W., Liang, F., Duan, Y., Tan, X., and Lu, X. (2013). Exploring the photosynthetic production doi: cyanobacteria. Metab. Eng. sucrose 17–25. 19, of capacity by 10.1016/j.ymben.2013.05.001 Duan, Y., Luo, Q., Liang, F., and Lu, X. (2016). Sucrose secreted by the engineered cyanobacterium and its fermentability. J. Ocean Univ. China 15, 890–896. doi: 10.1007/s11802-016-3007-8 Ducat, D. C., Avelar-Rivas, J. A., Way, J. C., and Silvera, P. A. (2012). Rerouting carbon flux to enhance photosynthetic productivity. Appl. Environ. Microbiol. 78, 2660–2668. doi: 10.1128/AEM.07901-11 52 Graham-Rowe, D. (2011). Agriculture: Beyond food versus fuel. Nature 474, S6–S8. doi: 10.1038/474S06a Han, X., Wang, W., and Lu, X. (2023). Engineering Synechococcus sp. PCC 7002 for sustainable production of sucrose. Algal Res. 74, 103212. doi: 10.1016/j.algal.2023.103212 Hays, S. G., and Ducat, D. C. (2015). Engineering cyanobacteria as photosynthetic feedstock factories. Photosynth. Res. 123, 285–295. doi: 10.1007/s11120-014-9980-0 Hays, S. G., Yan, L. L. W., Silver, P. A., and Ducat, D. C. (2017). Synthetic photosynthetic consortia define interactions leading to robustness and photoproduction. J. Biol. Eng. 11. doi: 10.1186/s13036-017-0048-5 Hendry, J. I., Gopalakrishnan, S., Ungerer, J., Pakrasi, H. B., Tang, Y. J., and Maranas, C. D. (2019). Genome-Scale Fluxome of Synechococcus elongatus UTEX 2973 Using Transient 13C-Labeling Data. Plant Physiol. 179, 761–769. doi: 10.1104/pp.18.01357 Kirsch, F., Klähn, S., and Hagemann, M. (2019). Salt-Regulated Accumulation of the Compatible Solutes Sucrose and Glucosylglycerol in Cyanobacteria and Its Biotechnological Potential. Front. Microbiol. 10. doi: 10.3389/fmicb.2019.02139 Klähn, S., and Hagemann, M. (2011). Compatible solute biosynthesis in cyanobacteria. Environ. Microbiol. 13, 551–562. doi: 10.1111/j.1462-2920.2010.02366.x Lang, T., Schoen, V., Hashem, K., McDonald, L., Parker, J., and Savelyeva, A. (2017). “The Environmental, Social, and Market Sustainability of Sugar,” in Advances in Food Security and Sustainability, ed. D. Barling (Elsevier), 115–136. doi: 10.1016/bs.af2s.2017.09.002 Liang, Y., Zhang, M., Wang, M., Zhang, W., Qiao, C., Luo, Q., et al. (2020). Freshwater Cyanobacterium Synechococcus elongatus PCC 7942 Adapts to an Environment with Salt Stress via Ion-Induced Enzymatic Balance of Compatible Solutes Downloaded from. Aemasmorg 1 Appl. Environ. Microbiol. 86, 2904–2923. doi: 10.1128/AEM Lin, P. C., Zhang, F., and Pakrasi, H. B. (2020). Enhanced production of sucrose in the fast- growing cyanobacterium Synechococcus elongatus UTEX 2973. Sci. Rep. 10. doi: 10.1038/s41598-019-57319-5 Lippi, L., Bähr, L., Wüstenberg, A., Wilde, A., and Steuer, R. (2018). Exploring the potential of high-density cultivation of cyanobacteria for the production of cyanophycin. Algal Res. 31, 363–366. doi: 10.1016/j.algal.2018.02.028 Markou, G., Vandamme, D., and Muylaert, K. (2014). Microalgal and cyanobacterial cultivation: The supply of nutrients. Water Res. 65, 186–202. doi: 10.1016/j.watres.2014.07.025 Montgomery, B. L. (2014). The Regulation of Light Sensing and Light-Harvesting Impacts the Use of Cyanobacteria as Biotechnology Platforms. Front. Bioeng. Biotechnol. 2. doi: 10.3389/fbioe.2014.00022. Mueller, T. J., Ungerer, J. L., Pakrasi, H. B., and Maranas, C. D. (2017). Identifying the Metabolic Differences of a Fast-Growth Phenotype in Synechococcus UTEX 2973. Sci. Rep. 7, 41569. doi: 10.1038/srep41569 Parmar, A., Singh, N. K., Pandey, A., Gnansounou, E., and Madamwar, D. (2011). Cyanobacteria 53 and microalgae: A positive prospect for biofuels. Bioresour. Technol. 102, 10163–10172. doi: 10.1016/j.biortech.2011.08.030 Peters, S., Rose, T., and Moser, M. (2010). “Sucrose: A Prospering and Sustainable Organic Raw Material,” in Carbohydrates in Sustainable Development I, eds. A. P. Rauter, P. Vogel, and Y. Queneau (Berlin, Heidelberg: Springer), 1–23. doi: 10.1007/128_2010_58 Qiao, C., Duan, Y., Zhang, M., Hagemann, M., Luo, Q., and Lu, X. (2018). Effects of reduced and enhanced glycogen pools on salt-induced sucrose production in a sucrose-secreting strain of Synechococcus elongatus PCC 7942. Appl. Environ. Microbiol. 84. doi: 10.1128/AEM.02023-17 Santos-Merino, M., Torrado, A., Davis, G. A., Röttig, A., Bibby, T. S., Kramer, D. M., et al. (2021). Improved photosynthetic capacity and photosystem I oxidation via heterologous metabolism engineering in cyanobacteria. Proc. Natl. Acad. Sci. 118, e2021523118. doi: 10.1073/pnas.2021523118 Santos-Merino, M., Yun, L., and Ducat, D. C. (2023). Cyanobacteria as cell factories for the sucrose. Front. Microbiol. 14, 1126032. doi: photosynthetic production of 10.3389/fmicb.2023.1126032 Satta, A., Esquirol, L., and Ebert, B. E. (2023). Current Metabolic Engineering Strategies for in Cyanobacteria. Microorganisms 11, 455. doi: Photosynthetic Bioproduction 10.3390/microorganisms11020455 Takatani, N., Uenosono, M., Hara, Y., Yamakawa, H., Fujita, Y., and Omata, T. (2022). Chlorophyll and Pheophytin Dephytylating Enzymes Required for Efficient Repair of PSII in Synechococcus elongatus PCC 7942. Plant Cell Physiol. 63, 410–420. doi: 10.1093/pcp/pcac006 Takatani, N., Use, K., Kato, A., Ikeda, K., Kojima, K., Aichi, M., et al. (2015). Essential Role of Acyl-ACP Synthetase in Acclimation of the Cyanobacterium Synechococcus elongatus Strain PCC 7942 to High-Light Conditions. Plant Cell Physiol. 56, 1608–1615. doi: 10.1093/pcp/pcv086 Thiel, K., Patrikainen, P., Nagy, C., Fitzpatrick, D., Pope, N., Aro, E.-M., et al. (2019). Redirecting photosynthetic electron flux in the cyanobacterium Synechocystis sp. PCC 6803 by the deletion of flavodiiron protein Flv3. Microb. Cell Factories 18, 189. doi: 10.1186/s12934- 019-1238-2 Ungerer, J., Lin, P.-C., Chen, H.-Y., and Pakrasi, H. B. (2018a). Adjustments to Photosystem Stoichiometry and Electron Transfer Proteins Are Key to the Remarkably Fast Growth of the Cyanobacterium Synechococcus elongatus UTEX 2973. mBio 9, 10.1128/mbio.02327- 17. doi: 10.1128/mbio.02327-17 Ungerer, J., Wendt, K. E., Hendry, J. I., Maranas, C. D., and Pakrasi, H. B. (2018b). Comparative genomics reveals the molecular determinants of rapid growth of the cyanobacterium Synechococcus elongatus UTEX 2973. Proc. Natl. Acad. Sci. 115, E11761–E11770. doi: 10.1073/pnas.1814912115 Walker, P. L., and Pakrasi, H. B. (2022). A Ubiquitously Conserved Cyanobacterial Protein Phosphatase Essential for High Light Tolerance in a Fast-Growing Cyanobacterium. 54 Microbiol. Spectr. 10, e01008-22. doi: 10.1128/spectrum.01008-22 Wang, B., Zuniga, C., Guarnieri, M. T., Zengler, K., Betenbaugh, M., and Young, J. D. (2023). Metabolic engineering of Synechococcus elongatus 7942 for enhanced sucrose biosynthesis. Metab. Eng. 80, 12–24. doi: 10.1016/j.ymben.2023.09.002 Xu, Y., Tiago Guerra, L., Li, Z., Ludwig, M., Charles Dismukes, G., and Bryant, D. A. (2013). Altered carbohydrate metabolism in glycogen synthase mutants of Synechococcus sp. strain PCC 7002: Cell factories for soluble sugars. Metab. Eng. 16, 56–67. doi: 10.1016/j.ymben.2012.12.002 Yu, J., Liberton, M., Cliften, P. F., Head, R. D., Jacobs, J. M., Smith, R. D., et al. (2015). Synechococcus elongatus UTEX 2973, a fast-growing cyanobacterial chassis for biosynthesis using light and CO2. Sci. Rep. 5, 8132. doi: 10.1038/srep08132 Zhang, L., Chen, L., Diao, J., Song, X., Shi, M., and Zhang, W. (2020). Construction and analysis of an artificial consortium based on fast-growing cyanobacterium Synechococcus elongatus UTEX 2973 to produce platform chemical 3-hydroxypropionic acid from CO2. doi: 10.21203/rs.2.22486/v1 Ziolkowska, J. R. (2014). Prospective technologies, feedstocks and market innovations for ethanol the US. Biotechnol. Rep. 4, 94–98. doi: in and biodiesel production 10.1016/j.btre.2014.09.001 55 APPENDIX ADDITIONAL SUPPORTING DATA FIGURES Figure S2.1. Custom lighting array consisting of four COB LED lights (BXRA-56C5300-H, Bridgelux, USA) mounted on an aluminum panel two inches apart and connected to a power supply (LS35-24, TDK-Lambda, Japan). The panel was attached to a ring clamp, and the clamp fastened to a ring stand. Light intensity can be set by adjusting the distance of the panel to the HDC, or by modulating the voltage of the power supply. 56 Figure S2.2. Relationship between dry cell weight (DCW, g/L) and OD750 in different growth conditions and cultivators. (A) WT, and uninduced and induced cscB/sps cultured in the MC-1000 under various light intensities. (B) Uninduced and induced cscB/sps grown in the CellDEG HDC system with 250 μmol photons m-2 s-1 and 2% CO2. 57 Figure S2.3. Total fixed carbon as shown by sucrose and dry cell weight accumulated per day in cultures grown in the MC-1000 under varying irradiances. Error bars denote standard deviation. 58 Figure S2.4. Light perceived in the center of a culture at varying OD750, distance from the light interface, and irradiance. Measurements are averaged values from a 15-second interval. 59 Figure S2.5. Growth and sucrose production of WT and cscB/sps-expressing S. elongatus grown -. (A) OD750 and (B) sucrose in the MC-1000 under various light intensities and 0 mM HCO3 production of cultures. Error bars denote standard deviation. Figure S2.6. (A) Growth as OD750 and (B) sucrose production of cscB/sps-expressing S. elongatus of cultures started at high cell densities (1.5 and 2 OD750). Error bars denote standard deviation. 60 CHAPTER 3: UTILIZATION OF ENDOGENOUS AND HETEROLOGOUS OUTER-MEMBRANE PROTEINS FOR CYANOBACTERIAL SURFACE DISPLAY This chapter was adapted from text originally published in: Yun, L. Sakkos, J.K., and Ducat, D.C. (2024) Population-level heterogeneity complicates utilization of Synechococcus elongatus PCC 7942 surface display platforms. microPub. Biol. doi: 10.17912/micropub.biology.001097 Copyright, © 2024 microPublications Biology — Reproduced with permission. 61 Introduction Surface display technologies offer a number of useful cellular properties that can be utilized for both academic research and biotechnological applications. However, these advancements have primarily been developed for only a few model heterotrophic microbes, leaving photosynthetic microbes, including cyanobacteria, with an underdeveloped molecular toolkit for surface display (Samuelson et al., 2002; Kondo and Ueda, 2004; Han et al., 2018). The ability to engineer cellular surface properties and to customize the binding of biologically derived and inorganic materials to the outermost cell membrane leaflet holds the potential to introduce innovative strategies to overcome the challenges of large-scale cultivation of photosynthetic microbes. For example, biomass recovery and dewatering represent a major barrier to achieving economically competitive cultivation of microalgae and cyanobacteria, and surface display could offer new options for inexpensive harvesting (Vandamme et al., 2013; Mathimani and Mallick, 2018) . In the current study, I set out to broaden the applicability of recently developed surface display systems for cyanobacteria to accelerate the potential of this important class of biotechnology chassis species (Santos-Merino et al., 2019, 2023). Surface display of engineered epitope tags was reported in the model cyanobacterium, Synechococcus elongatus PCC 7942, to mediate binding to both biotic and abiotic targets (Chungjatupornchai and Fa-Aroonsawat, 2008; Chungjatupornchai and Fa-aroonsawat, 2009; Chungjatupornchai et al., 2011; Ferri et al., 2015; Fedeson and Ducat, 2017). The primary strategy of these studies was to encode an epitope within an extracellular-facing loop of the endogenous outer-membrane porin protein, SomA. Initial designs showed some reactivity of engineered epitope protein tags (e.g., FLAG tag), but accessibility of the tags to appropriate antibodies was greatly improved by the genetic deletion of other factors likely occluding the cell surface, such as the O-antigen (OAg, wzt) and the putative surface layer (S-layer, slpA) proteins. Deletion strains (ΔwztΔslpA) exhibit more accessible tags that interacted with antibodies targeting this specific epitope (Fig. 1A) and a similar improved penetrance of immunostaining was observed in cells treated with chelating agents known to strip away cell surface components (Fedeson and Ducat, 2017). To expand upon this toolkit, I evaluated multiple variants of the SomA outer-membrane porin and the heterologous adhesin, intimin, in a range of genetic backgrounds of S. elongatus PCC 7942. Importantly, we sought to improve the utility of this surface display system by removing the requirement for specialized mediating antibodies for binding and to allow for direct 62 adhesion. Towards this aim, we focused on two strategies: 1) inserting a SpyTag peptide for the covalent binding of its protein partner, SpyCatcher, into SomA at sites previously shown to allow for extracellular epitope tagging (Zakeri et al., 2012; Fedeson and Ducat, 2017), and 2) heterologous expression of intimin proteins that form an extended surface spike-like projection for the specific purpose of attaching to other cells such as Escherichia coli, and which have been modified with protein binding domains to mediate intercellular binding (Figure 3.1; Glass & Riedel-Kruse, 2018; Robledo et al., 2022). In each approach we attempted multiple designs while varying the peptide linker length between domains, linker type (i.e., flexible vs. rigid), different binding motifs, and expressed these constructs in a wild-type (WT) or ΔwztΔslpA background of S. elongatus PCC 7942 to control for any impacts of surface-occluding factors, though this material will be relegated to the appendix of this chapter due to its preliminary nature. Figure 3.1. Schematic of background strains and engineered outer-membrane proteins. Diagram of WT and ΔwztΔslpA strains with outer-membrane occlusions and proteins (SomA or intimin) with engineered binding domains expressed on the surface of S. elongatus PCC 7942. Surface domains were detected by immunofluorescence or protein co-localization, as indicated. The table (right) summarizes the linker and epitope/peptide modifications for each outer -membrane protein construct displayed in this figure. Results Use of endogenous SomA for surface display We evaluated expression, localization, and accessibility of the different surface display constructs using immunofluorescence and soluble reporter proteins fused to a compatible protein interaction domain. For surface display domains inserted in an extracellular facing loop of SomA (Figure 3.1, constructs 1 and 2), we first probed with α-FLAG antibodies in fixed cell populations. Immunostaining indicated a rise in the proportion of labeled cells following induced expression of the tagged SomA (Figure 3.2A). Furthermore, our results generally reinforced our prior 63 Figure 3.2. Immunofluorescence of fixed S. elongatus PCC 7942 cells with SomA. Background strains with constructs 1 and 2 labeled with (A.) α-FLAG and (B.) α-SpyTag antibodies in cells bearing SomA-SpyTag-FLAG fusions. All constructs appeared to localize to the surface of S. elongatus PCC 7942 when induced, albeit within a subset of the cell population that was dependent on strain background and the primary antibody used. All scale bars indicate 5 μm. observations that the engineered epitope tag was more accessible on the surface of the ΔwztΔslpA mutant background, relative to paired controls expressing the same construct in a WT background (Figure 3.2). However, significant intercellular variation of immunolabeling was observed and a proportion of cells in the induced cultures stained poorly or did not exhibit fluorescence at higher levels than the uninduced controls. Generally, constructs with longer linker sequences (e.g., GSAGSAAGSGEF) between introduced domains appeared to stain poorly with α-FLAG 64 antibodies relative to constructs with shorter, flexible linker sequences (Figure 3.2A). Hypothetically, longer linker sequences may increase the distance between the epitope and the cell surface, but an extended linker may instead impose difficulty in translocating to the outer- membrane. In our best-performing SomA designs (i.e., GGGSx3 linker sequences in a ΔwztΔslpA background), the majority of fixed cells exhibited some staining localized to the cell periphery, although considerable variation in staining intensity was observed (Figure 3.2A). Due to the heterogenous staining, concerns regarding the integrity of the introduced SpyTag-FLAG epitope were raised, and we therefore conducted complementary immunostaining against SpyTag to validate if this epitope was localized to the cell surface. An extended incubation with α-SpyTag showed staining patterns similar to that of α-FLAG but with higher penetrance and average relative fluorescence intensity (Figure 3.2B). The majority of cells exhibited clear (albeit sometimes weak) staining with the α-SpyTag, with the exception of construct 2 in the WT background (Figure 3.2B). An overnight incubation with α-FLAG yielded a small handful of cells with puncta-like fluorescence on the cell periphery, whereas a 36-hour incubation with α-SpyTag reveals a spotty but more comprehensive labeling of the outer membrane (Figure 3.2B). Figure 3.3. SDS-PAGE of purified SpyTag-GFP and SpyCatcher-mNG. SpyTag (32.5 kDa; lane 1) and SpyCatcher-mNG (43.2 kDa; lane 2) after 10 and 120 minutes of in vitro binding (lanes 3 and 4, respectively). Nearly all of the SpyCatcher and SpyTag proteins are competent to form covalent bonds, as exhibited under the longer incubation periods. 65 To evaluate if the neighboring SpyTag domains were functionally active and accessible in the SomA fusions, we incubated cells in the presence of soluble SpyCatcher-mNeonGreen (mNG) reporter proteins. The majority of purified SpyCatcher-mNG formed covalent bonds with purified compatible SpyTag-GFP protein within a 10-minute in vitro incubation; the reaction was driven to near-completion in 2 hours (Figure 3.3). When live cells bearing modified SomA-SpyTag constructs had their surfaces stripped with EDTA and incubated with the soluble SpyCatcher-mNG for 3 hours, a limited number of cells displayed a peripheral fluorescence signal consistent with successful binding between the SpyTag and SpyCatcher pair (Figure 3.4); this staining pattern was never observed in negative control experiments using S. elongatus PCC 7942 lacking an integrated SomA-SpyTag gene (Figure 3.4). However, the vast majority of cells did not exhibit any SpyCatcher-mNG staining. Furthermore, there was no consistent trend in the proportion of cells exhibiting SpyCatcher-mNG attachment when considering induction of gene expression, various linker combinations, or the parental background in which SomA-SpyTag variants were expressed (Figure 3.4). Nonetheless, the small fraction (<<1%) of labeled cells often displayed a clear localization pattern that could not be dismissed as a staining artifact (Figure 3.4; red arrowheads). Figure 3.4. Protein co-localization of live S. elongatus PCC 7942 cells with SomA. Background strains with constructs 1 and 2 stripped with EDTA to remove surface-associated proteins and polysaccharides. Living cells were incubated with purified SpyCatcher-mNG and a small fraction of cells (generally <1%) exhibited localization patterns consistent with uniform binding of the reporter to SomA-SpyTag fusions (red arrows). All scale bars indicate 5 μm. 66 Intercellular variation in the level of expression, localization, and accessibility of surface- displayed domains was difficult to ascribe to any specific cellular feature or pattern. Relative to previously reported results (Fedeson and Ducat, 2017), the SpyTag epitope was accessible at a comparable penetrance to when probed with a prolonged α-SpyTag antibody incubation. By contrast, the FLAG epitope showed lower accessibility as assessed by an overnight incubation with α-FLAG antibodies. Overall, there appeared to be poor correlation between cell lines with higher penetrance of modified SomA (as observed by fixed cell immunofluorescence, Figure 3.2) and the proportion of living cells that reacted strongly with soluble SpyCatcher (Figure 3.4). The source of the intra-population variation in staining cells could not be explained as a merodiploid genetic insertion and was not resolved in separate control experiments using other engineered SomA variants, increased incubation times, or chemical protocols for ‘stripping’ the outer surface components with chelating agents (Sumper et al., 1990; Messner and Sleytr, 1992; Fedeson and Ducat, 2017). Furthermore, SpyCatcher-reactive cells of the population exhibited chlorophyll a fluorescence (Figure 3.4), and we were unable to find an association between cells exhibiting SpyCatcher-mNG staining and uptake of vital dyes; therefore, we cannot directly attribute the spurious staining with a subpopulation of dead cells. The nature of the variability remains largely unknown, although the most consistent explanation of our results is that other components (as yet, unknown) of the cell surface or extracellular matrix might contribute to epitope inaccessibility and the “all-or-nothing” staining patterns. Use of heterologous intimin for surface display To evaluate if a separate surface display system could resolve the inconsistency, I expressed intimin proteins that have been previously characterized to display a protein binding domain on an extended projection from the cell surface (Figure 3.1; constructs 3 and 4). When expressed, intimin-fusion proteins appeared to express poorly and/or primarily localize to the cytosol, as assessed by inclusion body-like immunofluorescence staining of fixed cell populations (Figure 3.5). Consistent with these results, we did not observe any reactivity of living cells with α- FLAG or α-p53 antibodies above the level of paired negative controls. Our results appear to suggest that intimins are not properly localized to the outer-membrane when expressed heterologously in S. elongatus PCC 7942. 67 Figure 3.5. Immunofluorescence of fixed S. elongatus PCC 7942 cells with intimin. Background strains with constructs 3 (intimin-FLAG) and 4 (intimin-p53) labeled with α-FLAG or α-p53 antibodies, respectively. Cells exhibit limited expression (intimin-FLAG) or mis- localization of expressed protein (intimin-p53). All scale bars indicate 5 μm. Discussion In conclusion, while surface display technologies offer considerable biotechnological promise, our study highlights the complexities and challenges associated with achieving surface display of functional and accessible protein domains for direct adhesion in S. elongatus PCC 7942. There have been a limited number of successful reports of functional cyanobacterial surface display, such as the fusion of affibodies onto endogenous pili or S-layer proteins in Synechocystis sp. PCC 6803 and antibody-mediated adhesion in S. elongatus PCC 7942 (Fedeson and Ducat, 2017; Cengic et al., 2018). Strikingly, while some living cyanobacteria expressing SomA-SpyTag fusions reacted strongly with soluble SpyCatcher and displayed localization patterns that would be expected from a functional surface display system, the vast majority of cells in the clonal population were non-reactive even under prolonged incubation and/or chemical treatment to strip surface-associated proteins and polysaccharides (Figure 3.4). Despite genetic and chemical attempts to clear occlusions from the cell surface, expressing various linker types to extend further from the outer-membrane, and numerous variations to the binding procedures, we consistently observed low penetrance of binding to epitopes and SpyCatcher. One possible explanation is that the surface loops of SomA remain occluded by other endogenous cellular factors that project from the surface of S. elongatus PCC 7942, limiting accessibility even in surface-stripped cells. 68 Attempts to remedy this possible obstacle by mounting epitopes on intimins that project from the cell surface were unsuccessful, presumably due to a failure to correctly target these proteins to the outer-membrane (Figure 3.5). It is possible that rare mutations in a minority of cells give rise to phenotypes with surface features amenable to display: selective recovery and sequencing of these rare cells could be revealing of the mechanism(s) responsible in this case. Our results suggest that future surface display efforts may be more effective if they are focused on engineering targets that project further from the surface of the cyanobacterial outer membrane. For example, type IV pili could represent a promising avenue for further engineering, as they have been successfully employed for surface display purposes in Synechocystis sp. PCC 6803, and cell-encapsulating S-layer proteins can be utilized as a scaffold for high-density surface display rather than be removed as occlusions (Cengic et al., 2018; Charrier et al., 2019). Ultimately, these results emphasize the need for further study into the outer-membrane of S. elongatus PCC 7942 for better design and application of surface display technologies. Materials and Methods Cell culture and strain generation Cyanobacterial cultures were grown in baffled flasks in a Multitron Pro (Infors HT) incubator under constant illumination from fluorescent bulbs (150 μmol m−2 s−1; 15W Gro-Lux; Sylvania) in 32°C, 2% CO2, and with 125 rpm shaking. Cultures were routinely grown in liquid BG-11 medium supplemented with 1 g/L HEPES (H4034; Sigma-Aldrich) set to pH 8.3 with NaOH, with the appropriate antibiotic when needed (chloramphenicol, 25 μg/mL; kanamycin, 20 μg/mL; spectinomycin, 100 μg/mL). To improve consistency and maintain cultures in exponential rowth, cultures for immunostaining or protein co-localization were back-diluted to an OD750 of ~0.3 each day for three days before inducing with 1 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG; I2481C25; Goldbio). Cultures were induced for 24 hours prior to any subsequent cell staining or analysis. The knockout background strain (ΔwztΔslpA) was constructed by sequentially transforming with integration plasmids containing an antibiotic resistance selectable marker flanked by ~1 kilobase fragments upstream and downstream of the gene of interest on the 5’ and 3’ ends, respectively, as reported previously (Fedeson and Ducat, 2017). Surface display strains were constructed by combining synthesized gene fragments of adhesins (Glass & Riedel-Kruse, 2018; Integrated DNA Technologies) and PCR-linearized NS3 integration vector with the S. 69 elongatus PCC 7942 SomA with Gibson Assembly (Gibson et al., 2009; Fedeson and Ducat, 2017). Transformations were performed as outlined in Golden et al., 1987. Transformants were selected on BG-11 agar plates with appropriate antibiotics (chloramphenicol, 12.5 μg/mL; kanamycin, 16.7 μg/mL; and spectinomycin, 100 μg/mL), and verified by colony PCR and Sanger sequencing (RTSF Genomics Core, Michigan State University). Fluorescent protein expression, purification, and in vitro binding SpyCatcher-mNG and SpyTag-GFP plasmids were transformed into E. coli BL21(DE3) as is routine for protein purification, and transformants recovered on selective agar plates. Single colonies were cultured overnight at 37°C in 5 mL Lysogeny Broth (LB) with the appropriate antibiotic (ampicillin (50 μg/mL) for SpyCatcher-mNG, and kanamycin (50 μg/mL) for SpyTag- GFP). The following day, overnight cultures were back diluted 1:100 into fresh Terrific Broth (TB) with the appropriate antibiotic, and cultured at 37°C until a density of 0.4-0.6 OD600 was reached. At mid-exponential, cultures were chilled to approximately 23°C and induced with a final concentration of 400 μM IPTG for SpyCatcher-mNG and 500 μM anhydrous tetracycline (aTc) for SpyTag-GFP prior to culturing overnight at 23°C. Cell suspensions were chilled and transferred to a pre-weighed container to be pelleted, and the wet cell mass of the pellet was weighed before being resuspended with a minimal volume of lysis buffer and stored in -80°C until further processing. To prepare cells for lysing, frozen samples were thawed on ice and cold lysis buffer was added (final of 1:1 volume of lysis buffer to wet cell mass, w/w). During lysis, samples were kept on ice and a sonicator (Fisherbrand Model 120 Sonic Dismembrator; Fisher Scientific) equipped with a 1/8-inch probe microtip was used to disrupt cell integrity. Lysate was centrifuged 7,000 rpm for 25 minutes at 10°C on a Sorvall centrifuge equipped with the SS-34 fixed angle rotor (Sorvall) to pellet the cellular debris. Expressed proteins (SpyTag-GFP and SpyCatcher-mNG) were recovered by affinity purification using encoded Hisx6 tags. To prepare the Ni-NTA agarose resin (Qiagen) for purification, 3 mL of the slurry was aliquoted into an empty gravity flow column (Poly-Prep Chromatography Columns; Bio-Rad) and the resin was washed 3 times with 3 mL lysis buffer. The cleared cell lysate was then loaded onto the column and the column was washed 4 times with 3 mL of the lysis buffer. Bound protein was eluted with the elution buffer. To remove the high concentration of imidazole from the elution buffer, the collected protein was transferred to an Amicon Ultra-15 Centrifugal Filter (Millipore) with the appropriate molecular weight cut off (20 70 kDa for SpyCatcher-mNG and 10 kDa for SpyTag-GFP). The filters were centrifuged at 10°C to remove the bulk of the liquid before adding the protein storage buffer to dilute the remaining amount of imidazole before being centrifuged again; there were two total additions of protein storage buffer before the protein was collected, quantified, aliquoted, and flash frozen with liquid nitrogen. To ascertain functionality, in vitro binding assays were performed. Approximately 230 pM of each SpyCatcher-mNG and SpyTag-GFP were diluted in phosphate citrate buffer (pH 7.5) before combining at room temperature. To stop the reactions for time points (10 and 120 minutes), samples were mixed 1:1 with 2X Laemlli buffer (161-0737; Bio-Rad) and boiled at 95°C for 10 minutes. Samples were loaded onto 10% Mini-PROTEAN TGX Precast Protein Gels (Bio-Rad), and run with 75 V/10 mA for the first 10 minutes, and 120 V/30 mA for the remaining run time. Immunostaining and protein co-localization Cells were harvested (250 μL culture at ~1.5 OD750), pelleted, and washed in PBS. For immunofluorescence, cells were fixed, permeabilized, blocked, labeled with α-FLAG, α-SpyTag, or α-p53 antibodies, and labeled against the primary antibody with a fluorescent secondary antibody. To fix cells, washed pellets were resuspended with 200 μL pre-chilled methanol and incubated at -20˚C for 10 minutes. Methanol was removed, and cells were washed with 500 μL PBS three times. To permeabilize, fixed cells were resuspended with 300 μL of lysozyme solution, and incubated in 37˚C for 30 minutes. Lysozyme solution was removed, and cells were washed with 500 μL PBS three times. Permeabilized cells were then incubated with blocking buffer for 30 minutes at room temperature. Blocking buffer was removed, and cells were resuspended with 300 μL of a diluted primary antibody in blocking buffer (1:1000 dilution for α-FLAG (DYKDDDDK) (LT0420; Lifetein) and α-p53 (DO-1) Alexa Fluor 488 (sc-126 AF488; Santa Cruz Biotechnology), 1:5000 dilution for α-SpyTag (HCA406; Bio-Rad)). Samples with α-FLAG and α-p53 were incubated in 4˚C overnight, and α-SpyTag samples were incubated in 4˚C for ~36 hours. The primary antibody was removed, and cells were washed with 500 μL PBS three times before being resuspended in 500 μL blocking buffer and incubated at room temperature for 15 minutes. The blocking buffer was removed, and the samples were protected from light in the following steps until imaging. Cells were resuspended with 300 μL of a diluted secondary antibody in blocking buffer (α-mouse goat IgG DyLight™ 488 (35502; Thermo Fisher Scientific), or α- rabbit goat IgG DyLight™ 488 (35552; Thermo Fisher Scientific)), and incubated in 30˚C for 1 71 hour. After labeling, cells were pelleted, the secondary antibody removed, and cells were washed with 500 μL PBS three times. For protein co-localization, washed cells had the cell surface stripped by resuspending pellets with 200 μL 150 mM EDTA and incubated for 30 minutes at room temperature with mixing. The EDTA solution was removed, and cells were washed with 500 μL phosphate citrate buffer three times. To label living cells with the compatible soluble SpyCatcher-based reporter, cell pellets were collected as above, then resuspended with 300 μL phosphate citrate buffer with SpyCatcher- mNG (final molar ratio of ~1:20 SomA protein to SpyCatcher-mNG), and incubated for 3 hours with mixing at room temperature while protected from the light. After labeling, the protein solution was removed, and cells were washed with 500 μL phosphate citrate buffer three times prior to imaging. Fluorescence microscopy and image analysis To image, labeled cells were resuspended with 15-30 μL phosphate citrate buffer or PBS, spotted onto agarose pads (1.5% w/v agarose in distilled water), and mounted onto glass coverslips. Fluorescence images were collected using an Axio Observer.D1 microscope (ZEISS), Axiocam 503 camera (ZEISS), 63x1.3 NA objective, X-Cite 120Q (Lumen Dynamics), and the Zen Blue (ZEISS) software. The following filter sets were used: filter set 43 (000000-1114-101; ZEISS) for chlorophyll autofluorescence, filter set 38 (000000-1031-346; ZEISS) for DyLight 488 fluorescence, and filter set 46 (000000-1031-346; ZEISS) for mNG fluorescence. Exposure times were held constant for fluorescent antibodies and SpyCatcher reporter proteins, respectively, so that outcomes could be compared across experiments. Image analysis was performed with Python 3. Cell segmentation was conducted using custom Python scripts that used the Unet segmentation architecture (Ronneberger et al., 2015) for deep learning and implementation by Pytorch (Paszke et al., 2019). Mean fluorescence intensity plots were generated by segmenting the cells using the chlorophyll autofluorescence channel, rotating the cells such that the medial axis was horizontal, and rescaling the dimensions to 500 × 200 pixels to ensure consistent boundaries. The pixel intensity of the DyLight 488 and SpyCatcher- mNG was averaged from the segmented cells in the collection of images from each respective strain. ACKNOWLEDGEMENTS We would like to thank our laboratory postdoctoral fellows Dr. María Santos-Merino and 72 Dr. Amit Singh, and our predoctoral fellows Emmanuel Kokarakis and Rees Rillema for helpful comments on this manuscript. We would also like to thank Dr. Bryan Ferlez for guidance in protein purification. The plasmid for SpyTag-GFP was kindly gifted to us by Dr. Cheryl Kerfeld. This work was primarily supported by the National Science Foundation and the Division of Molecular and Cellular Bioscience (Grant: 1845463), and partially supported by a fellowship from the Plant Biotechnology for Health and Sustainability Training Program at Michigan State University (Grant: NIH T32-GM110523). 73 REFERENCES Cengic, I., Uhlén, M., and Hudson, E. P. (2018). Surface display of small affinity proteins on Synechocystis sp. strain PCC 6803 mediated by fusion to the major type IV pilin PilA1. J. Bacteriol. 200, 10.1128/jb.00270-18. doi: 10.1128/jb.00270-18 Charrier, M., Li, D., Mann, V. R., Yun, L., Jani, S., Rad, B., et al. (2019). Engineering the S-layer of Caulobacter crescentus as a foundation for stable, high-density, 2D living materials. ACS Synth. Biol. 8, 181–190. doi: 10.1021/acssynbio.8b00448 Chungjatupornchai, W., and Fa-Aroonsawat, S. (2008). Biodegradation of organophosphate pesticide using recombinant cyanobacteria with surface- and intracellular-expressed organophosphorus hydrolase. J. Microbiol. Biotechnol. 18, 946–951. Chungjatupornchai, W., and Fa-aroonsawat, S. (2009). Translocation of green fluorescent protein to cyanobacterial periplasm using ice nucleation protein. J. Microbiol. 47, 187–192. doi: 10.1007/s12275-008-0188-x Chungjatupornchai, W., Kamlangdee, A., and Fa-aroonsawat, S. (2011). Display of organophosphorus hydrolase on the cyanobacterial cell surface using Synechococcus outer membrane protein A as an anchoring motif. Appl. Biochem. Biotechnol. 164, 1048–1057. doi: 10.1007/s12010-011-9193-3 Fedeson, D. T., and Ducat, D. C. (2017). Cyanobacterial surface display system mediates engineered interspecies and abiotic binding. ACS Synth. Biol. 6, 367–374. doi: 10.1021/acssynbio.6b00254 Ferri, S., Nakamura, M., Ito, A., Nakajima, M., Abe, K., Kojima, K., et al. (2015). Efficient surface-display of autotransporter proteins in cyanobacteria. Algal Res. 12, 337–340. doi: 10.1016/j.algal.2015.09.013 Gibson, D. G., Young, L., Chuang, R.-Y., Venter, J. C., Hutchison, C. A., and Smith, H. O. (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6, 343–345. doi: 10.1038/nmeth.1318 Glass, D. S., and Riedel-Kruse, I. H. (2018). A synthetic bacterial cell-cell adhesion toolbox for programming multicellular morphologies and patterns. Cell 174, 649-658.e16. doi: 10.1016/j.cell.2018.06.041 Golden, S. S., Brusslan, J., and Haselkorn, R. (1987). "Genetic engineering of the cyanobacterial chromosome," in Methods in Enzymolology, eds. R. Wu, and L. Grossman (Academic Press), 153, 215–231. Han, L., Zhao, Y., Cui, S., and Liang, B. (2018). Redesigning of microbial cell surface and its application to whole-cell biocatalysis and biosensors. Appl. Biochem. Biotechnol. 185, 396–418. doi: 10.1007/s12010-017-2662-6 Kondo, A., and Ueda, M. (2004). Yeast cell-surface display—applications of molecular display. Appl. Microbiol. Biotechnol. 64, 28–40. doi: 10.1007/s00253-003-1492-3 Mathimani, T., and Mallick, N. (2018). A comprehensive review on harvesting of microalgae for biodiesel – Key challenges and future directions. Renew. Sustain. Energy Rev. 91, 1103– 1120. doi: 10.1016/j.rser.2018.04.083 74 Messner, P., and Sleytr, U. B. (1992). “Crystalline bacterial cell-surface layers,” in Advances in Microbial Physiology, ed. A. H. Rose (Academic Press), 213–275. doi: 10.1016/S0065- 2911(08)60218-0 Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, et al., Chintala S. 2019. PyTorch: An imperative style, high-performance deep learning library. In 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada. Robledo, M., Álvarez, B., Cuevas, A., González, S., Ruano-Gallego, D., Fernández, L. Á., et al. (2022). Targeted bacterial conjugation mediated by synthetic cell-to-cell adhesions. Nucleic Acids Res. 50, 12938–12950. doi: 10.1093/nar/gkac1164 Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation, in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, eds. N. Navab, J. Hornegger, W. M. Wells, and A. F. Frangi (Cham: Springer International Publishing), 234–241. doi: 10.1007/978-3-319-24574-4_28 Samuelson, P., Gunneriusson, E., Nygren, P.-Å., and Ståhl, S. (2002). Display of proteins on bacteria. J. Biotechnol. 96, 129–154. doi: 10.1016/S0168-1656(02)00043-3 Santos-Merino, M., Singh, A. K., and Ducat, D. C. (2019). New applications of synthetic biology tools for cyanobacterial metabolic engineering. Front. Bioeng. Biotechnol. 7, 33. doi: 10.3389/fbioe.2019.00033 Santos-Merino, M., Yun, L., and Ducat, D. C. (2023). Cyanobacteria as cell factories for the sucrose. Front. Microbiol. 14, 1126032. doi: photosynthetic production of 10.3389/fmicb.2023.1126032 Sumper, M., Berg, E., Mengele, R., and Strobel, I. (1990). Primary structure and glycosylation of the S-layer protein of Haloferax volcanii. J. Bacteriol. 172, 7111–7118. doi: 10.1128/jb.172.12.7111-7118.1990 Vandamme, D., Foubert, I., and Muylaert, K. (2013). Flocculation as a low-cost method for harvesting microalgae for bulk biomass production. Trends Biotechnol. 31, 233–239. doi: 10.1016/j.tibtech.2012.12.005 Zakeri, B., Fierer, J. O., Celik, E., Chittock, E. C., Schwarz-Linek, U., Moy, V. T., et al. (2012). Peptide tag forming a rapid covalent bond to a protein, through engineering a bacterial adhesin. Proc. Natl. Acad. Sci. 109, E690–E697. doi: 10.1073/pnas.1115485109 75 APPENDIX PRELIMINARY DATA FIGURES The following datasets elaborate upon the information published in µPublication Biology. Due to the short format of the journal style, I was unable to include a number of additional pieces of data that detail other efforts to troubleshoot and use of additional controls. This information is provided here in a concise format for the reader who may have additional interest and/or wish to follow up on a related project. 76 Figure S3.1. Immunofluorescence of fixed cells displaying nanobodies. (A) Table of constructs of SomA and intimin fusion proteins with p53 or AKT bound by various flexible linker types. (B) Immunofluorescence of fixed cells labeled with their respective antibodies (α-p53 or α-AKT). All images were taken with the same exposure as the figures in the main body text, but the brightness was adjusted for this figure due to the low fluorescence. The brightness adjustments remain the same for all images in this figure. All scale bars indicate 5 μm. (C) The relative fluorescence intensity of each strain and condition. Cells exhibit limited expression and/or mis-localization of expressed proteins. Nanobodies are markedly larger than FLAG (102 residues, p53; 118 residues, AKT; 24 residues, FLAG), suggesting their larger size prevents the protein from properly transporting to the outer-membrane. The best performing construct (construct 5) also had the shortest linkers, which may have aided their transport, but still formed inclusion bodies. Cells expressing intimin (construct 7) showed the least labeling, following the trend supporting poor transport of the heterologous outer-membrane protein. 77 Figure S3.2. Immunofluorescence of fixed cells displaying rigid linkers. (A) Table of constructs of SomA-SpyTag-FLAG fusion proteins bound by all-rigid or rigid-flexible combination linkers. (B) Immunofluorescence and brightfield of fixed cells labeled with α-FLAG antibodies. All images were taken with the same exposure as the figures in the main body text, but the brightness was adjusted for this figure due to the low fluorescence. The brightness adjustments remain the same for all images in this figure. All scale bars indicate 5 μm. (C) The relative fluorescence intensity of each strain and condition. Cells exhibit limited expression (construct 8) and/or mis-localization of expressed proteins (construct 9). In anticipation of the distance the surface display may need to traverse to be accessible to binding targets, versions of rigid linkers were designed with the reasoning that their stiffness may help extend beyond extracellular polysaccharides that may occlude the cell surface. However, the strains expressing all-rigid linkers (construct 8) exhibit nearly no labeling and strains with combination linkers (construct 9) show partial labeling or inclusion bodies. Rigid linker-containing constructs were also the most difficult to obtain and culture, with many strains failing to be generated. Cells expressing surface display of all types outlined in this work undergo some degree of enlargement in size. Construct 5 in the ΔwztΔslpA background exhibited the most extreme of this phenotype with elongation that, at times, doubled their length (C, brightfield). These data suggest that rigid linkers are not compatible with surface display in S. elongatus PCC 7942 in the approach undertaken in this dissertation. 78 CHAPTER 4: CONCLUDING REMARKS AND FUTURE DIRECTIONS 79 Overview With climate change and pollution worsening with current practices, the global economy will need to transition towards sustainable measures. Due to their metabolic flexibility, fast growth, and ability to fix CO2 using light, non-potable water, and non-agricultural land, cyanobacteria are ideal candidates to contribute to a circular economy. The central theme of my dissertation was related to the capacity of cyanobacteria to produce sucrose as an alternative to terrestrial plant carbon feedstocks, which is economically and chemically expensive to harvest, and competes with resources required to grow food (Graham-Rowe, 2011; Lang et al., 2017). There has been much progress in the decade since the development of sucrose-secreting Synechococcus elongatus PCC 7942: the system has been expressed in various strains, has been made inducible without salt stress, and multiple enzymes up- and down-stream of the sucrose biosynthesis pathway have been engineered in an effort to increase productivity. However, there are still more gaps to fill in before cyanobacterial sucrose biosynthesis can reach the theoretical possibility of surpassing plant output. The work in this dissertation includes examples of directions that have not yet been fully tapped into by the broader research community. Genetic manipulation is often the go-to in biotechnology: while such strategies are the cornerstone of the field and will always be important, there are other parameters beyond interrogating metabolic pathways. While the genes (or lack thereof) are required for the production of novel compounds in a more efficient manner, cells are still dependent on the resources they are provided. Without adequate light to power photosynthetic machinery or nutrition to support biomass production, biosynthesis will suffer as cells will prioritize surviving before thriving. In this dissertation, I explored the impact of light and CO2 availability and have outlined several findings to improve the production of sucrose-secreting cyanobacteria which can be applied more broadly to bioproductive aquatic photoautotrophs. Photosynthesis, as the name implies, requires light, but I found that currently utilized irradiance (~150 µmol photons m-2 s-1) is insufficient in the MC-1000, a bubble-stream style multicultivator. Growth and sucrose production of cultures with a starting density of 0.3 OD750 are light-limited until 500 µmol photons m-2 s-1, but light intensities beyond this become detrimental (Figure 2.1A), as by Takatani et al., 2015. The difference in sucrose yield between 150 and 500 µmol photons m-2 s-1 cultures become evident within the first 24 h, suggesting that the higher irradiance is not damaging at a lower OD750. After identifying 250 µmol photons m-2 s-1 as the optimal irradiance for low-starting density cultures 80 (0.3 OD750), I transitioned to the HDC, a system consisting of cultivators with a gas permeable bottom and a bicarbonate buffer providing CO2. The HDC can supply up to 8% CO2 and showed the MC-1000 was CO2-limited. Using the same irradiance and BG-11 medium, the HDC yielded 245% higher OD750 and 188% more sucrose (Figure 2.2A, B). Interestingly, the data showed that the commonly utilized 2% CO2 in incubators (e.g., Infors Multitron) is also carbon limiting. Normally, the growth of sucrose secreting strains is lower than those without the metabolic burden of exporting fixed carbon, but this pattern was not observed with the HDC. Furthermore, this indicates that sucrose production does not necessarily have to have a trade-off with growth if there is sufficient carbon to supply both the fixation of sucrose and cellular biomass. With the increase in light and carbon availability, I was concerned that the standard BG- 11 medium was not properly sustaining the stark increases I observed. Indeed, my suspicions were confirmed by replacing BG-11 with CD medium, a more nutrient-dense media recipe designed for high density photoautotrophic growth. The new medium increased sucrose yield by 139%, but also lead to a decrease in growth of induced sucrose-secreting strains, returning to the aforementioned pattern. I postulate this is due to the increased osmotic pressure the medium exerted on the endogenous and heterologous SPS, which is supported by the significant amounts of sucrose produced by strains not induced to export sucrose. Analysis of the proportions of total fixed carbon revealed that the stress response draws away a higher proportion of fixed carbon from growth to sucrose biosynthesis (Figure 2.2A, E). As cultures grow more dense, the light available to the cultures diminish due to self-shading; to address this, I further increased irradiances to 500 µmol photons m-2 s-1. I observed another increase in growth and sucrose yield of 173% and 182%, respectively. Interestingly, the differences observed in the MC-1000 were not significant (Figure 2.1A, B), but were so in this instance with the difference being CO2 availability – perhaps the carbon limitation inhibited the true potential of the tested light intensities. I revisited the idea of utilizing irradiances much higher than optimal for S. elongatus PCC 7942. Previously, I interrogated if high irradiances that may be damaging at low-starting densities may be beneficial for high-starting density cultures. However, after cultivating cultures of 1.5 and 2.0 OD750 with 2000 and 2500 µmol photons m-2 s-1, I found that the growth of these cultures was not sustainable and sucrose titer was not improved relative to low-starting cultures (Figure 2.1C, D). This may have been because overnight cultures were concentrated to the desired OD750. While experiments began with fresh media, the cells were “older”, and I have consistently observed sucrose 81 productivity decline with time. So, for my next foray into high light and high density, I began with a low-starting density of 0.3 OD750 with 500 µmol photons m-2 s-1 and increased the light intensity to 1000 µmol photons m-2 s-1 at 48 h once the cultures were on average 2.3 OD750. Although 1000 µmol photons m-2 s-1 is considered to be detrimental to S. elongatus PCC 7942, it increased growth by 17% and sucrose titer by 15% in sucrose exporting strains (Figure 2.3 C, D). Additionally, uninduced strains produced 126% more sucrose in this condition than sucrose-exporting strains grown with 250 µmol photons m-2 s-1 – albeit with 618% higher OD750. Unlike many genetic engineering attempts to increase sucrose production, enhancing light, CO2, and nutrition increases sucrose titers without sacrificing cell health. In the end, our highest productivity was 47.8 mg L-1 h-1; for reference, the highest productivity of S. elongatus PCC 7942 was achieved by a strain that forwent growth entirely to devote all fixed carbon had a productivity of 48 mg L-1 h-1 (Abramson et al., 2018). Previous attempts to improve sucrose yield relied heavily on genetic modification to increase the carbon flux towards sucrose, but this strategy came with the price of cell health. In the work outlined above, the cells remained resilient throughout: growth and sucrose continued to climb, despite prolonged incubation periods, high density, and high light. Interrogating long- standing common growth conditions is a promising avenue to improve the bioproductivity of existing strains while maintaining fitness, which will be required for cyanobacteria biotechnology to be commercially viable. Of the many steps that are being taken to bridge the distance between lab research and “real-world” application are attempts to reduce the cost of harvesting biomass by eliminating extraneous costs and to improve productivity by utilizing co-cultures – both of which can be aided by the development of direct cyanobacterial surface display. For example, the cost of biomass harvesting can account for 20-30% of total costs when employing traditional methods (Christenson and Sims, 2011). Inducible intercellular adhesion could allow cyanobacteria to create flocs without the added costs or risks associated with added chemicals or centrifugation, while being more reliable than natural methods such as gravity sedimentation (Christenson and Sims, 2011). On another note, the use of engineered co-cultures are gaining traction as photoautotrophs can directly provide for the carbon requirements of bioproductive heterotrophs. Moreover, these engineered co-cultures show improved yields relative to their axenic counterparts (Cooper and Smith, 2015; Fedeson and Ducat, 2021). However, these novel systems do not have the advantage of shared evolutionary time (i.e., most co-cultures characterized to date are from species that have not co- 82 evolved in a natural context), and are relatively unstable and difficult to maintain. Naturally occurring microbial communities often rely on spatial structure (e.g., biofilms) (Flemming et al., 2016; Nadell et al., 2016), and experiments have revealed that this can also help stabilize engineered co-cultures. Microfluidic devices (Kim et al., 2008; Gupta et al., 2020), agar (Pande et al., 2016), and hydrogels (Smith and Francis, 2017) have been employed thus far with success, but these are limited to small-scale applications. Flocs created by intercellular adhesion can introduce spatial structure while still being in liquid culture, and thus be scalable. Previous works in cyanobacteria have engineered pili in Synechocystis sp. PCC 6803 (Cengic et al., 2018) and a porin in S. elongatus PCC 7942 (Fedeson and Ducat, 2017) with some success. The work discussed in this dissertation continues to develop the porin, SomA, as well as utilizing a heterologous outer-membrane protein, intimin. The previous iteration of SomA for surface display required a mediating protein to assist in binding – likely due to occluding factors on the cell surface (Fedeson and Ducat, 2017). Anticipating the distance that may be required to bridge surface displayed components (FLAG, SpyTag, p-53) and targets of interest, I engineered linkers of various types and lengths in between SomA and adhesin domains (Figure 3.1). Likewise, intimin was employed due to its natural function of intercellular binding in Escherichia coli and a nanobody library developed for surface display (Glass and Riedel-Kruse, 2018). Our findings with immunofluorescence with α-FLAG initially agreed those outlined in Fedeson and Ducat (2017), that strains with genetically removed outer-membrane proteins (e.g., slpA, wzt) had greater binding (Figure 3.2A), though my strains exhibited greater heterogeneity within the population. Additionally, I found that a shorter, flexible linker sequence (GGGSx3) performed better than strains with long flexible linkers (GSAGSAAGSGEF), which may be excessive rather than useful in traversing the outer-membrane. Due to the possibility of the epitopes cleaving away from the cell due to the introduced linkers, I performed an extended incubation with α-SpyTag to confirm both domains were available, which further validated my previous observations (Figure 3.2B). After confirming the presence and proper transport of the adhesins to the cell surface, I evaluated the accessibility of these functionalizing tags through the use of a protein co-localization study of surface displayed SpyTag to soluble SpyCatcher-mNG. The overall relative fluorescence intensities of all strains were markedly reduced compared to when strains were tagged with an antibody, and the penetrance remained low throughout the population (Figure 3.4). Intimin proteins did not fare better; rather, we observed inclusion body-like labeling suggesting that the 83 proteins did not transport to the outer-membrane properly (Figure 3.5). Clearly, more research is required to understand the environment of the S. elongatus PCC 7942 outer-membrane, and how to transport heterologous proteins to the outer-membrane if interest regarding intimin continues. Future outlook and perspectives Genetic engineering first drew me into the field of biotechnology – it is remarkable that we can introduce heterologous, modified, or even entirely novel genes to gain new functions. I was truly inspired by the microbial synthesis of bioplastics and biofuels as an impressionable undergraduate student. They still are incredible feats that open doors to what the future could be, but I have since learned to read more carefully and think more skeptically. Now, I seek less novelty and more applicability. How can we bring these advancements to the broader public? We must make them scalable, affordable, and accessible to replace the products that contribute to climate change and pollution. For this reason, I devoted my graduate career in search of how to make that possible, rather than the metabolic engineering I initially thought I would pursue. In the past six years of this journey, I have learned from personal experience that nothing in science is easy without luck, and that much of lab research has not transitioned into industry, not for a lack of effort. Still, I am not without hope. There is still so much to try and to learn from, which is an iterative process that forever betters itself, and that is what has kept me in science. Regarding cyanobacterial biotechnology, I would like to see variations of my research topics to continue, though I admit I am biased. The idea of creating surface display for direct adhesion in cyanobacteria really stuck to me. My initial hopes were to create flocs in culture to use as a platform to investigate the impact of spatial structuring in safeguarding engineered mutualistic interactions, namely from “cheaters” which could easily arise in commercial applications by way of mutation or contamination. In the example outlined in Figure 4.1, S. elongatus PCC 7942 provides sucrose to the mutualistic E. coli in exchange for β-lactamase, which detoxifies the antibiotic in the medium. However, there is a low concentration of cheaters that benefit from the exchanged goods without bearing the metabolic burden of mutual contribution. This affords the cheater an advantage to out-grow the mutualistic bacteria (Figure 4.1B), rendering bioproductive cultures (axenic and co-cultures) useless. If mutualistic partners aggregate, the secreted public goods could localize to the floc, increasing their concentrations, thus improving the growth of the partners while effectively starving out the excluded cheater (Figure 4.1C). This approach could be a method to combat contaminants, but cheaters would also arise from E. coli losing the capacity 84 to provide β-lactamase while in the floc. In that case, the cheater would benefit from the localization of public goods and outgrow the mutualistic partners, but at its own demise because the cheater is dependent on both contributors to survive (Figure 4.1D). This could potentially lead to an interesting phenomenon called the Simpson’s paradox in which the local population (i.e., the aggregate) will be dominated by the cheater, but the global population (i.e., the culture) will contain greater populations of the mutualistic partners. Individual flocs may be quickly overrun by cheaters, but they cannot continue to do so without the support of the mutualists it out-grew, keeping them relatively static, while flocs containing a majority of partners will continue to thrive. However, this is hypothetical. An ideal antibiotic concentration would be to only attenuate the growth of S. elongatus PCC 7942, but initial antibiotic screening with 15 different concentrations of chloramphenicol and hygromycin showed nearly binary responses: death or normal growth. For cellular aggregates to be used as a platform to study the stabilizing effects of spatial structuring, at Figure 4.1. Co-culture of planktonic and attached S. elongatus PCC 7942 and E. coli. Cyanobacteria providing sucrose (green) in exchange for β-lactamase (yellow) from E. coli in the presence of a cheater (red) over time. (A) Surface display of an antigen-nanobody pair and localized metabolites. (B) A planktonic co-culture of collaborators and low concentration of cheaters that overrun the population. (C) Aggregated cooperators that exclude cheaters. (D) Co- aggregation of cooperators and cheaters. 85 least two obstacles must be overcome: 1) surface display in cyanobacteria and their heterotrophic partners need to be developed for specific intercellular adhesion; 2) a public good that is similarly detrimental to both the cyanobacteria and the cheater heterotroph needs to be identified. While the approach I outlined here for surface display was not successful, there have been at least two successful examples (Cengic et al., 2018; Lin et al., 2020), but have not been utilized for inter- species bindings. Some cyanobacteria, like S. elongatus PCC 7942, express S-layer proteins that could be engineered for surface display, but I am concerned about the extracellular polysaccharides and other occluding factors on the outer-membrane that may inhibit accessibility. There is limited literature characterizing the outer-membrane as most interest in cyanobacterial membranes often falls to those implicated in photosynthesis. Identifying a factor that is equally (or similarly) detrimental and recoverable to cyanobacteria and heterotrophs could prove difficult, but not impossible. From personal experience, S. elongatus PCC 7942 is more sensitive to antibiotics and environmental conditions than E. coli; a less robust heterotroph may need to be identified instead. One of the reasons engineered microbial communities first began to gain traction was due to their simplicity relative to their natural counterparts, but our brief venture through their construction has revealed their own complexities. Even between only two species, there is much to understand and optimize as we take on the role evolution once played (though directed evolution is still an option). Still, I believe this is well worth the effort. The seemingly simple aggregation of bioproductive partners opens so many opportunities. Utilizing it as a platform to understand the rise of emergent properties (e.g., robustness, division of labor) can not only guide engineering efforts to create more resilient co-cultures, but to also de-establish undesirable microbial communities for medical, commercial, and ecological applications. Cyanobacteria have transformed the Earth to the lush and bountiful planet we know today, and with their help, we may be able to maintain it. The field of cyanobacteria is growing in every direction: ecology, genomics, biomedicine, biofuel, bioenergy, and endlessly more. In 1980, there were only 237 papers regarding cyanobacteria, but the numbers have jumped to 1634, 1584, and 1600 in 2016, 2017, and 2018, respectively (Konur, 2019). It is truly exciting to see the progress cyanobacterial research has made in the past decade. 86 REFERENCES Abramson, B. W., Lensmire, J., Lin, Y. T., Jennings, E., and Ducat, D. C. (2018). Redirecting carbon to bioproduction via a growth arrest switch in a sucrose-secreting cyanobacterium. Algal Res. 33, 248–255. doi: 10.1016/j.algal.2018.05.013 Cengic, I., Uhlén, M., and Hudson, E. P. (2018). Surface display of small affinity proteins on Synechocystis sp. strain PCC 6803 mediated by fusion to the major type IV pilin PilA1. J. Bacteriol. 200, 10.1128/jb.00270-18. doi: 10.1128/jb.00270-18 Christenson, L., and Sims, R. (2011). Production and harvesting of microalgae for wastewater and bioproducts. Biotechnol. Adv. 29, 686–702. doi: treatment, biofuels, 10.1016/j.biotechadv.2011.05.015 Cooper, M. B., and Smith, A. G. (2015). Exploring mutualistic interactions between microalgae the omics age. Curr. Opin. Plant Biol. 26, 147–153. doi: in and bacteria 10.1016/j.pbi.2015.07.003 Fedeson, D. T., and Ducat, D. C. (2017). Cyanobacterial Surface Display System Mediates Engineered Interspecies and Abiotic Binding. ACS Synth. Biol. 6, 367–374. doi: 10.1021/acssynbio.6b00254 Fedeson, D. T., and Ducat, D. C. (2021). “Symbiotic Interactions of Phototrophic Microbes: Engineering Synthetic Consortia for Biotechnology,” in Role of Microbial Communities for Sustainability, eds. G. Seneviratne and J. S. Zavahir (Singapore: Springer), 37–62. doi: 10.1007/978-981-15-9912-5_2 Flemming, H. C., Wingender, J., Szewzyk, U., Steinberg, P., Rice, S. A., and Kjelleberg, S. (2016). Biofilms: An emergent form of bacterial life. Nat. Rev. Microbiol. 14, 563–575. doi: 10.1038/nrmicro.2016.94 Glass, D. S., and Riedel-Kruse, I. H. (2018). A Synthetic bacterial cell-cell adhesion toolbox for programming multicellular morphologies and patterns. Cell 174, 649-658.e16. doi: 10.1016/j.cell.2018.06.041 Graham-Rowe, D. (2011). Agriculture: Beyond food versus fuel. Nature 474, S6–S8. doi: 10.1038/474S06a Gupta, S., Ross, T. D., Gomez, M. M., Grant, J. L., Romero, P. A., and Venturelli, O. S. (2020). Investigating the dynamics of microbial consortia in spatially structured environments. Nat. Commun. 11. doi: 10.1038/s41467-020-16200-0 Kim, H. J., Boedicker, J. Q., Choi, J. W., and Ismagilov, R. F. (2008). Defined spatial structure stabilizes a synthetic multispecies bacterial community. Konur, O. (2019). “Cyanobacterial bioenergy and biofuels science and technology: a scientometric overview,” in Cyanobacteria, eds. A. K. Mishra, D. N. Tiwari, and A. N. Rai (Academic Press), 419–442. doi: 10.1016/B978-0-12-814667-5.00021-0 Lang, T., Schoen, V., Hashem, K., McDonald, L., Parker, J., and Savelyeva, A. (2017). “The Environmental, Social, and Market Sustainability of sugar,” in Advances in Food Security and Sustainability, ed. D. Barling (Elsevier), 115–136. doi: 10.1016/bs.af2s.2017.09.002 87 Lin, T. Y., Wen, R. C., Shen, C. R., and Tsai, S. L. (2020). Biotransformation of 5- hydroxymethylfurfural to 2,5-furandicarboxylic acid by a syntrophic consortium of engineered Synechococcus elongatus and Pseudomonas putida. Biotechnol J 15, e1900357. doi: 10.1002/biot.201900357 Nadell, C. D., Drescher, K., and Foster, K. R. (2016). Spatial structure, cooperation and competition in biofilms. Nat. Publ. Group 14. doi: 10.1038/nrmicro2016.84 Pande, S., Kaftan, F., Lang, S., Svato, A., Germerodt, S., and Kost, C. (2016). Privatization of cooperative benefits stabilizes mutualistic cross-feeding interactions in spatially structured environments. ISME J. 10, 1413–1423. doi: 10.1038/ismej.2015.212 Smith, M. J., and Francis, M. B. (2017). Improving metabolite production in microbial co-cultures using a spatially constrained hydrogel. Biotechnol Bioeng 114, 1195–1200. doi: 10.1002/bit.26235 Takatani, N., Use, K., Kato, A., Ikeda, K., Kojima, K., Aichi, M., et al. (2015). Essential role of Acyl-ACP synthetase in acclimation of the cyanobacterium Synechococcus elongatus strain PCC 7942 to high-light conditions. Plant Cell Physiol. 56, 1608–1615. doi: 10.1093/pcp/pcv086 88