Week 1 HW: Final Project Ideas

(1) What you would like to biomanufacture and why?

I want to engineer a microbial chassis for low molecular weight hyaluronic acid production. Hyaluronic acid (HA) is a glycosaminoglycan distinguished by its high water loading capacity - it can hold up to 1000 times its own weight in water. It is found naturally across vertebrate species where it plays a vital role in hydration, healing, fluid balance, and structural cushioning. Its behaviour is largely a function of its molecular weight with high MWs having more structural and lubricating roles while low MWs have greater permeability and participate in cellular signalling, wound healing, and molecular transport. Its widespread presence in the body gives it diverse biomedical applications. Low MW-HA (<300 kDa) is used often in dermatology as well as cosmetics for its hydration and anti-aging effects. It is also being explored for engineering nanoparticles, liposomes, and micelles for drug delivery with 200 kDa HA having high affinity to the CD44 receptor on cancer cells (Yasin et al., 2022) (Fakhari & Berkland, 2013). As of 2025, hyaluronic acid has a global market value of USD 11.5 billion with dermal fillers and cosmetic applications holding a bulk of the market share (Grand View Research, 2026). This marks the importance of developing a sustainable and safe mechanism for efficient hyaluronic acid production.

My interest in producing low MW-HA stems from my research group’s larger goal of producing defined molecular weight hyaluronic acid. I am also very eager to engineer a HA producing strain outside of the natural HA producers by taking into account all the relevant genes and fluxes worth optimising. Finally, I get a particular sense of satisfaction from manufacturing a compound that I see so often in all my skincare products.

1. Yasin, A., Ren, Y., Li, J., Sheng, Y., Cao, C., & Zhang, K. (2022). Advances in hyaluronic acid for biomedical applications. Frontiers in Bioengineering and Biotechnology, 10, 910290. https://doi.org/10.3389/fbioe.2022.910290
2. Fakhari, A., & Berkland, C. (2013). Applications and emerging trends of hyaluronic acid in tissue engineering, as a dermal filler, and in osteoarthritis treatment. Acta Biomaterialia, 9(7), 7081–7092. https://doi.org/10.1016/j.actbio.2013.03.005
3. Grand View Research. (2026). Hyaluronic acid market size, share & trends analysis report, 2026–2033. https://www.grandviewresearch.com/industry-analysis/hyaluronic-acid-market

(2) What living system would you use?

Currently, large-scale production of hyaluronic acid is done using streptococcus species which naturally produce high MW-HA. After production, the high MW-HA is enzymatically cleaved to produce lower MWs. This means of production is unfavourable due to the pathogenic nature of streptococcus as well as the co-production of endotoxins (Soundiriraj et al., 2026). For this reason, I want to work on using a generally recognised as safe (GRAS) organism for industrial scale synthesis to alleviate some of the downstream purification concerns. I also want to reduce the need for post-production processing by consciously limiting the MW-HA to a narrow range as opposed to purifying out the range I want after.
The microbial chassis I have chosen for this project is Corynebacterium glutamicum as it has previously been reported to produce hyaluronic acid at a lower MW than the other tested GRAS organisms. Additionally, C. glutamicum is already a major industrial workhorse for the production of amino acids, carotenoids, and biofuels among other things. C. glutamicum is a non-sporulating, gram-positive, facultative anaerobe that is preferred for its stress tolerance, usage of diverse carbon substrates, ability to reach high cell densities, and well characterised genetic and metabolomic pathways (Becker et al., 2018) (Wolf et al., 2021). The precursors for hyaluronic acid production, UDP-glucuronic acid and UDP-N-acetyl-glucosamine are already produced in C. glutamicum as components of the cell wall. The only missing enzyme for HA synthesis is hyaluronan synthase (HasA) that links the precursors together to make an alternating chain of glucuronic acid and N-acetyl-glucosamine units. While hyaluronic acid synthesis is theoretically possible with just the addition of HasA, the output has been low as the insufficient amount precursors compensate either on cell-growth or HA production. For this reason, many studies have co-expressed genes from the Streptococcus has operon; most commonly UDP-glucose dehydrogenase (HasB) and/or UDP-glucose pyrophosphorylase (HasC) are coexpressed to increase the existing levels of UDP-glucuronic acid which is largely the limiting precursor. Additionally, production of UDP-glucuronic acid requires NAD+, making it favourable for hosts to have a Δldh that increases the NAD+/NADH ratio as necessary. Finally, different Streptococcus species have minor differences in their HasA enzyme that ultimately determines the maximum length that the HA chain can reach. This makes it important to test different HasA enzymes to determine which is capable of producing the desired range of MWs under the tested conditions (Soundiriraj et al., 2026). Taking all the above into account, I intend to focus on adding hasA-hasB into C.glutamicum with the hasA gene sequences taken from differing sources. These sources would include S. pyogenes, S. uberis, S. parauberis, and S. zooepidemicus. Previous literature has found that HasA from S. pyogenes produced the lowest MW HA in *L.lactis), approximately 0.3 MDa (Soundiriraj et al., 2026). By using C. glutamicum I hope to bring this number lower as well as improve the generated yield. Additionally I plan to test a Δldh strain for differences in production titre as is frequently reported in microbial HA production studies. While first testing these conditions using a pXMJ19 vector with the hasA-hasB genes under a strong IPTG-inducible tac promoter, I ultimately want to integrate the genes into the chromosome using a pK18mobsacB suicide vector to create a stable HA producing host.

1. Soundiraraj, S., Ravishankar, N., Jeeva, P., et al. (2026). Tailored molecular weight hyaluronic acid production by engineered Lactococcus lactis. Microbial Cell Factories, 25, 64. https://doi.org/10.1186/s12934-026-02945-8
2. Becker, J., Rohles, C. M., & Wittmann, C. (2018). Metabolically engineered Corynebacterium glutamicum for bio-based production of chemicals, fuels, materials, and healthcare products. Metabolic Engineering, 50, 142–163. https://doi.org/10.1016/j.ymben.2018.07.008
3. Wolf, S., Becker, J., Tsuge, Y., Kawaguchi, H., Kondo, A., Marienhagen, J., Bott, M., Wendisch, V. F., & Wittmann, C. (2021). Advances in metabolic engineering of Corynebacterium glutamicum to produce high-value active ingredients for food, feed, human health, and well-being. Essays in Biochemistry, 65(2), 197–212. https://doi.org/10.1042/EBC20200134

(3) How would you use automation?

Since I plan on testing at least 8 different strain variants (HasA from 4 different sources x Δldh /wild-type C. glutamicum) and different fermentation substrates, I would use an Opentrons automated liquid handling robot through the screening and sampling stages to reduce manual burden and experimental variability. Here I would automate the dispensing of substrate media, antibiotics, IPTG, and the inocula into 96-deep-well plates such that the various experimental conditions can be tested parallelly. This would also benefit the periodic testing of OD600 through plate-readers. Additionally HA titre could be quantified easily from the 96 well-plates directly using the CTAB turbidimetric method while quantifying the MW would have to be done separately by HPLC (Wen et al., 2010).

1. Wen, C., Xu, H. M., Sun, Y. F., Wang, Y., & Chen, Z. Y. (2010). Efficient quantification of hyaluronic acid in fermentation broth by modified CTAB method. China Biotechnology, 30(2), 89–93.

(4) How much would you need to make, and how long would it take?

(a) How much? Previously HA production titers from strains similar to the ones I will be testing were 21.6 g/L for a MW range from 1.28 MDa to 260 kDa. This was achieved using a 5L fermenter with 2L working volume and 48 h of fed-batch fermentation (Cheng et al., 2017). Since I plan on producing HA around the smaller MW range, I expect our titre to be increased from this amount to roughly 25 g/L as a result of increased precursor availability. Ideally, I want to produce 1 kg of purified low-MW HA as a proof of concept study. Ambitiously assuming that I can produce 25 g/L of HA using the published data (Cheng et al., 2017) as a benchmark, I would require roughly 40L of fermentation broth to produce 1 kg of HA. If I assume I have a 20% downstream loss (as downstream recovery is not often reported), I can produce 1 kg of HA using 50L of fermentation broth. If aeration can be optimized to maintain this 25g/L production, I can scale up production from a 5L fermenter to a 70L fermenter with ~50L working volume.

(b) How long? 1 kg of HA could be produced in 48 h using a 70L fermenter with two additional days for harvesting, purification, and characterisation (Cheng et al., 2017). The prototyping stage could take weeks to identify a productive HA producing strain. With automation testing various conditions in parallel, I expect the screening time to be reduced: 3 weeks for developing the strains (constructing plasmids and primers, assembling the plasmids and inserts, transforming and validating the resultant strains), 2 weeks for 96 well plate screening and identifying promising strains, 4 days to characterise the HA produced by each of the strains, and 4 weeks to optimise fermenter conditions. Overall, I optimistically expect this process to take roughly 9-10 weeks

1. Cheng, F., Luozhong, S., Guo, Z., Yu, H., & Stephanopoulos, G. (2017). Enhanced biosynthesis of hyaluronic acid using engineered Corynebacterium glutamicum via metabolic pathway regulation. Biotechnology Journal, 12(10), 1700191. https://doi.org/10.1002/biot.201700191

(5) What don’t you know yet?

My three biggest technical unknowns are

(a) Can different HasA variants actually produce different MW-HA as previously reported in L. lactis? Since this is the first study testing HasA from different sources, it would be interesting to know whether the previously reported patterns with S. pyogenes producing the lowest MW repeats itself or whether HA-MW below 300 kDa can be achieved at all in our experimental conditions with C. glutamicum. Whether the HA-MW can be produced in a narrow range without outliers is also something to watch out for (Soundiriraj et al., 2026).

1. Soundiraraj, S., Ravishankar, N., Jeeva, P., et al. (2026). Tailored molecular weight hyaluronic acid production by engineered Lactococcus lactis. Microbial Cell Factories, 25, 64. https://doi.org/10.1186/s12934-026-02945-8

(b) Is the gene combination selected from the has operon appropriate for the desired production values? While this is uncertain, previous studies testing 8 different has operon combinations found that hasAB and hasABC were the best performing in C. glutamicum (Chen et al., 2017).

1. Cheng, F., Luozhong, S., Guo, Z., Yu, H., & Stephanopoulos, G. (2017). Enhanced biosynthesis of hyaluronic acid using engineered Corynebacterium glutamicum via metabolic pathway regulation. Biotechnology Journal, 12(10), 1700191. https://doi.org/10.1002/biot.201700191

(c) Can production successfully be scaled up from pilot-scale 2 L working volume studies to 50 L without compromising on titre As previous studies have mostly been pilot scale studies with smaller volume fermenters, the problem of aeration, heating, mixing, and nutrient gradient optimisation arise again

AI agent: ChatGPT

Week 2 HW

Part 1: Define your expression target

What molecule are you expressing? I want to express the enzyme urate oxidase to aid in the breakdown of uric acid, the end metabolite of purine metabolism.

What is the sequence? The protein sequence for uricase will be taken from Vibrio vulnificus and can be found from NCBI accession number: TDL85923.1

MTIQNENRTMFYGKGDVFVYRTFVQPLTGLRKIPESSFTERDNTIFGFNCQISLKGEAFFSSFTEGDNSL VVATDSMKNIIHRQAANYQGNTAEGFLKFICEVFLDKYSHIDAVELTATEVPFDHVLVPKGEGHENSDVV YRCSRNERATTTIEVKRTPTGSKIVKHSSGIVDVHLIKVKGSSFYGFIQDEYTTLPEAQDRPLFIYLDFD WEYSRWEDGTGANPEKYVAAEQICDIANTVFHELNNRSIQQLIYHIGIRILERFPQLANVQFKTNNRTWE TVVETIPNSEGSVYQEPRPPFGFQGFVVTQEDVKQKKAADSTVGASR

What host or chassis? I want to use a probiotic organism to act as a transient gut microbe to help relieve hyperuricemia. For this, I would use Saccharomyces boulardii that already has applications as a biotherapeutic probiotic and is typically cleared from the gut within 2-5 days post-supplementation.

Where does the product need to go? The product will be expressed intracellularly in the cytoplasm of Saccharomyces boulardii. In order for it to come in contact with uric acid, I will have to additionally express a urate transporter (Such as UapA)

References:

1. Wenzhuo Wang, Lei Pan, Huansha He, Huiyuan Xue, He Huang, Audrey Mihewi Samosir, Xian Fu, Yue Shen; Systematic Engineering for Efficient Uric Acid-Degrading Activity in Probiotic Yeast Saccharomyces boulardii. ACS Synth. Biol. 20 June 2025; 14 (6) 2030 2043. https://doi.org/10.1021/acssynbio.4c00831
2. National Center for Biotechnology Information. (n.d.). Urate oxidase [Vibrio vulnificus] (Protein accession No. TDL85923.1). NCBI Protein. NCBI Protein record

Part 6. What dont you know yet?

My biggest unknown at the moment is whether the host system and transporter protein will be functional at a 37°C body temperature as opposed to their preferred 30°C Additionally, since this probiotic is mainly meant to reduce the side effects of high nucleic acid single cell proteins, I wonder if administering both at the same time would be optimal for uric acid removal.

Week 3 HW

3. ASSIGNMENT: CFPS REACTION DESIGN FOR SFGFP, PETASE, RETEPLASE

Goal: using the collective intelligence of the HTBAA community and accessible AI systems, design novel CFPS reaction compositions for each of the three target proteins: sfGFP, PETase, and Reteplase to reduce the cost per gram of functional protein. Each design should be compatible with the experimental framework described above and should reflect the distinct expression and folding requirements of that protein. You can learn more about the details of the coopetition here.

A. COMPARE CELL-FREE REACTION COMPOSITIONS (20 minutes)

1. Why does the concentration of potassium glutamate and magnesium glutamate differ between compositions? Which is more sensitive?

The concentrations of K-glutamate and Mg-glutamate differ between CFPS formulations because K⁺ and Mg²⁺ have distinct biochemical functions and their optimal concentrations depend on the composition of the entire reaction. K⁺ primarily establishes the ionic environment required for efficient translation, whereas Mg²⁺ directly interacts with ribosomes, nucleic acids, ATP and other metabolites. Changes in NTP, phosphate and other charged components alter Mg²⁺ availability, so the Mg²⁺ concentration must be adjusted accordingly. While the concentration of both matter, research has shown that Mg2+ has a relatively narrower optimum, compared to that of K+, with deviations strongly affecting protein synthesis (Banks et al., 2022).

References:

1. Banks, A. M., Whitfield, C. J., Brown, S. R., Fulton, D. A., Goodchild, S. A., Grant, C., Love, J., Lendrem, D. W., Fieldsend, J. E., & Howard, T. P. (2022). Key reaction components affect the kinetics and performance robustness of cell-free protein synthesis reactions. Computational and Structural Biotechnology Journal, 20, 218 229. https://doi.org/10.1016/j.csbj.2021.12.013

2. Explain how a PEP+NTP reaction formulation is different from a Ribose+NMP reaction formulation.

PEP + NTP and ribose + NMP formulations differ mainly in how they provide energy and nucleotides for protein synthesis. In a PEP + NTP system, PEP is used as a high-energy phosphate donor to regenerate ATP, while ATP, GTP, CTP and UTP are directly supplied for transcription. In a ribose + NMP system, AMP, GMP, CMP and UMP are supplied instead, which can then be converted into NTPs by enzymes present in the cell extract. Ribose can also enter the cell extracts metabolic pathways and contribute to maintaining the energy and nucleotide pools. Therefore, the ribose + NMP system relies more on the metabolic activity of the cell extract, whereas the PEP + NTP system supplies more of the required components directly. The NMP-based system can also reduce the cost of the reaction because NMPs are cheaper than NTPs (Jewett et al., 2008)

3. Choose two cell-free reaction compositions from the literature, compare and contrast the choice of reagents and the concentration of reagents.

The two formulations are both based on E. coli cell extracts, but they use different strategies for energy regeneration and protein folding. Michel et al. (2012) was designed for disulfide-containing proteins, while Colant et al. (2021) focused on high protein production.

Overall, Michel et al. uses higher concentrations of the basic salts and NTPs and includes a creatine phosphate-based energy system and additional redox components (GSH, GSSG and DsbC) for disulfide-bond formation. Colant et al. uses lower NTP and salt concentrations but includes PEP, CoA, NAD and polyamines to support energy regeneration and metabolism. Thus, the reagent composition is adapted to the requirements of each CFPS system.

References:

1. Michel, E., & Wüthrich, K. (2012). Cell-free expression of disulfide-containing eukaryotic proteins for structural biology. The FEBS Journal, 279(17), 3176 3184. https://doi.org/10.1111/j.1742-4658.2012.08697.x
2. Colant, N., Melinek, B., Teneb, J., Goldrick, S., Rosenberg, W., Frank, S., & Bracewell, D. G. (2021). A rational approach to improving titer in Escherichia coli-based cell-free protein synthesis reactions. Biotechnology Progress, 37(1), e3062

Note: A recommended starting point would be to review a least one published formulation, such as PANOx-SP (Jewett & Swartz 2004), a cost-optimized formulation from Olsen et al. (2026), and/or an AI-designed formulation from Smith et al. (2026).

B. DESIGN ONE REACTION COMPOSITION FOR EACH OF THE THREE PROTEINS (40 minutes)

1. Sign in to coopetition.htgaa.org and publish your first unique 20 µL CFPS reaction composition for each of the following (dont worry, you can resubmit later):

• sfGFP
• PETase
• Reteplase

For each composition, specify the reagents you would adjust from the relevant categories, including salts, energy sources, building blocks, cofactors, and redox components.

2. As a HTBAA research team, we will have the opportunity to suggest new reagents to add to the cell-free reagent inventory. Contribute (or upvote) at least one reagent to the “Community Reagents” section of the website and provide a short rationale. Include in your homework documentation how you decided on that specific reagent and provide links to any relevant research.

3) Visit the cfps-coopetition forum page to ensure you will be able to take part in the discussion. Bonus: send a note in the channel about a CFPS optimization idea that you have! If you would like clarification on aspects of the project, please message in this dedicated channel.

C. ASIMOV KERNAL (OPTIONAL)

1. Build T7 compatible plasmids for cell-free expression using the amino acid coding sequences below:

sfGFP

MSKGEELFTGVVPILVELDGDVNGHKFSVRGEGEGDATIGKLTLKFICTTGKLPVPWPTLVTTLTYGVQCFSRYPDHMKRHDFFKSAMPEGYVQERTISFKDDGKYKTRAVVKFEGDTLVNRIELKGTDFKEDGNILGHKLEYNFNSHNVYITADKQKNGIKANFTVRHNVEDGSVQLADHYQQNTPIGDGPVLLPDNHYLSTQTVLSKDPNEKGTRDHMVLHEYVNAAGITWSHPQFEK

PETase

MEKKIWSHPQFEKGGSGENLYFQSGGASPAGGGANPYERGPNPTDALLEASSGPFSVSEENVSRLSASGFGGGTIYYPRENNTYGAVAISPGYTGTEASIAWLGERIASHGFVVITIDTITTLDQPDSRAEQLNAALNHMINRASSTVRSRIDSSRLAVMGHSMGGGGTLRLASQRPDLKAAIPLTPWHLNKNWSSVTVPTLIIGADLDTIAPVATHAKPFYNSLPSSISKAYLELDGATHFAPNIPNKIIGKYSVAWLKRFVDNDTRYTQFLCPGPRDGLFGEVEEYRSTCPFGGSGVSGWRLFKKIS

Reteplase

MEKKIWSHPQFEKGGSGENLYFQSGGASPAGGGNSDCYFGNGSAYRGTHSLTESGASCLPWNSMILIGKVYTAQNPSAQALGLGKHNYCRNPDGDAKPWCHVLKNRRLTWEYCDVPSCSTCGLRQYSQPQFRIKGGLFADIASHPWQAAIFAKHRRSPGERFLCGGILISSCWILSAAHCFQERFPPHHLTVILGRTYRVVPGEEEQKFEVEKYIVHKEFDDDTYDNDIALLQLKSDSSRCAQESSVVRTVCLPPADLQLPDWTECELSGYGKHEALSPFYSERLKEAHVRLYPSSRCTSQHLLNRTVTDNMLCAGDTRSGGPQANLHDACQGDSGGPLVCLNDGRMTLVGIISWGLGCGQKDVPGVYTKVTNYLDWIRDNMRPGGSGVSGWRLFKKIS

Final Project