Week 3 HW: Cell-Free Reaction Design

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

Copy and paste this section (section 3, A-F) into your website! Note, coopetition.htgaa.org is intended to support you as you answer the following questions.

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)

A1. Why does the concentration of potassium glutamate and magnesium glutamate differ between compositions? Which is more sensitive?

My response

My understanding is that different CFPS recipes need different amounts of potassium and magnesium because the other ingredients also affect how these salts work. Potassium helps create a suitable environment for making proteins. Magnesium helps ribosomes, the machinery that makes proteins, work properly. Some ingredients can bind magnesium, so not all the magnesium added is freely available. For example, PEP binds magnesium more strongly than pyruvate. This means that changing the energy source can also require changing the magnesium concentration.

Magnesium is generally more sensitive. In one experiment by Jewett and Swartz (2004), increasing added magnesium glutamate from 20 to 24 mM reduced protein production to 73% of its original level. Increasing potassium glutamate from 175 to 200 mM only reduced it to 94%. This example helped me understand why adding more salt does not always improve a reaction. The best amount depends on the recipe, and magnesium needs especially careful adjustment.

Reference: Jewett, M. C., & Swartz, J. R. (2004). Mimicking the Escherichia coli cytoplasmic environment activates long-lived and efficient cell-free protein synthesis. Biotechnology and Bioengineering, 86(1), 19–26. https://doi.org/10.1002/bit.20026. See Table II for the comparison above; these values describe the tested conditions, not a universal optimum.

A2. Explain how a PEP+NTP reaction formulation is different from a Ribose+NMP reaction formulation.

My response

A PEP+NTP formulation adds ATP, CTP, GTP, and UTP directly as the building blocks for making RNA. PEP helps regenerate ATP to supply energy for protein synthesis.

A Ribose+NMP formulation uses simpler starting materials. It adds AMP, CMP, GMP, and UMP, which enzymes in the cell lysate must convert into NTPs. Ribose supports the metabolism needed for this process, often together with other energy sources.

My understanding is that the first approach supplies more ready-to-use materials, while the second relies more on the reactions already possible in the lysate. Ribose+NMP formulations can reduce reagent costs, but their performance depends on the whole recipe and the activity of the lysate.

A3. Choose two cell-free reaction compositions from the literature, compare and contrast the choice of reagents and the concentration of reagents.

My response

I compared PANOx-SP and RFopt using the formulations listed in Table 1 of Olsen et al. (2026). Both contain potassium glutamate, magnesium glutamate, amino acids, and a buffer, but they use different ingredients to support energy production and RNA synthesis.

PANOx-SP contains 30 mM PEP and directly supplied NTPs: 1.2 mM ATP and 0.85 mM each of CTP, GTP, and UTP. RFopt instead contains 50 mM ribose, 10 mM glucose, and NMPs: 3 mM AMP and 2.15 mM each of CMP, GMP, and UMP. These NMPs need to be converted into NTPs by enzymes in the lysate.

The concentrations of other ingredients also differ. RFopt uses more potassium glutamate (362 versus 130 mM), amino acids (5 versus 2 mM), and HEPES (75 versus 57 mM). Both use 8 mM magnesium glutamate. RFopt also leaves out some added components, such as CoA, NAD, and tRNA.

What I learned from this comparison is that a lower-cost formulation does not simply use less of everything. RFopt removes some ingredients while increasing others. Its performance depends on how the ingredients work together.

Reference: Olsen, M. L., et al. (2026). Design-driven optimization of low-cost reagent formulations for reproducible and high-yielding cell-free gene expression. Nature Communications, 17, 3478. https://doi.org/10.1038/s41467-026-69605-8. Table 1.

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)

B1. Sign in to coopetition.htgaa.org and publish your first unique 20 µL CFPS reaction composition for each of the following (don’t worry, you can resubmit later):

• sfGFP
• PETase
• Reteplase

Note: if you’re unable to login with your HTGAA keycloak account, please post in Zulip so that you can participate in the coopetition!

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

My response

I submitted three 20 µL compositions. For each one, I started with the matching reference formulation and changed one reagent. These are exploratory designs, not confirmed improvements.

sfGFP: a small increase in magnesium

Sample ID: qingyun_liu_sfgfp_ribose_mg885_v1

I started with the sfGFP Ribose reference formulation and increased magnesium glutamate from 8.225 to 8.85 mM. This corresponds to increasing the magnesium stock addition from 225 to 250 nL. All other reagent additions stayed the same, and water was adjusted to 0.15 µL to keep the total volume at 20 µL.

I want to test whether a small increase in magnesium improves sfGFP production. Magnesium is important for protein synthesis, but too much can also reduce production. I chose one 25 nL addition step as an exploratory test, not a known optimum.

PETase: less DsbC

Sample ID: qingyun_liu_petase_pep_dsbc375_v1

I started with the PETase PEP reference formulation and reduced DsbC from 5 to 3.75 µM, a 25% reduction. The DsbC stock addition decreased from 1,000 to 750 nL. All other reagent additions stayed the same, and water was adjusted to keep the total volume at 20 µL.

I want to test whether less DsbC can maintain PETase activity while reducing reagent cost. I chose this concentration as an exploratory test point, not a known optimum. If activity drops too much, the cost per unit of activity may increase instead.

Reteplase: a small change in glutathione balance

Sample ID: qingyun_liu_reteplase_pep_gsh1125_v1

I started with the Reteplase PEP reference formulation and increased GSH from 1 to 1.125 mM by adding an extra 25 nL of GSH stock, increasing its addition from 200 to 225 nL. GSSG and DsbC were kept unchanged, along with the other reagent additions. Water was adjusted to 0.60 µL to maintain a total volume of 20 µL.

I want to test whether this small change in the glutathione balance improves Reteplase activity. I chose the smallest allowed volume increase as an exploratory test, not a known optimum. The change could improve or reduce activity, so its effect needs experimental testing.

All three designs retain the reference energy sources, amino acids, nucleotides, and cofactors. The changes focus on a salt for sfGFP, a folding helper for PETase, and a redox component for Reteplase. Each reaction retains 2,000 nL base buffer, 5,000 nL lysate, and 2,000 nL of the matching DNA template.

Reference formulations: Ginkgo sfGFP Ribose, PETase PEP, and Reteplase PEP, available in the Cell-Free Reaction Designer.

B2. 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.

My response

I upvoted the existing FkpA suggestion by Xiangtao Shi because I wanted to explore another way to help Reteplase fold correctly. Groff et al. (2014) found that FkpA and DsbC supported antibody folding in a cell-free system.

My Reteplase formulation already contains DsbC, so I think FkpA is worth testing as an additional folding helper. However, the study tested antibodies, not Reteplase, so it does not prove that FkpA will improve our reaction. I would compare Reteplase activity with and without added FkpA and check whether any improvement is worth the extra reagent cost.

Reference: Groff et al. (2014), “Engineering toward a bacterial ‘endoplasmic reticulum’ for the rapid expression of immunoglobulin proteins.” https://doi.org/10.4161/mabs.28172.

B3. 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.

CFPS coopetition forum

My response

I accessed the cfps-coopetition channel and confirmed that I could view the discussion.

C. ASIMOV KERNAL (OPTIONAL)

C1. Build T7 compatible plasmids for cell-free expression using the amino acid coding sequences below:

sfGFP

MSKGEELFTGVVPILVELDGDVNGHKFSVRGEGEGDATIGKLTLKFICTTGKLPVPWPTLVTTLTYGVQC
FSRYPDHMKRHDFFKSAMPEGYVQERTISFKDDGKYKTRAVVKFEGDTLVNRIELKGTDFKEDGNILGHK
LEYNFNSHNVYITADKQKNGIKANFTVRHNVEDGSVQLADHYQQNTPIGDGPVLLPDNHYLSTQTVLSKD
PNEKGTRDHMVLHEYVNAAGITWSHPQFEK

PETase

MEKKIWSHPQFEKGGSGENLYFQSGGASPAGGGANPYERGPNPTDALLEASSGPFSVSEENVSRLSASGF
GGGTIYYPRENNTYGAVAISPGYTGTEASIAWLGERIASHGFVVITIDTITTLDQPDSRAEQLNAALNHM
INRASSTVRSRIDSSRLAVMGHSMGGGGTLRLASQRPDLKAAIPLTPWHLNKNWSSVTVPTLIIGADLDT
IAPVATHAKPFYNSLPSSISKAYLELDGATHFAPNIPNKIIGKYSVAWLKRFVDNDTRYTQFLCPGPRDG
LFGEVEEYRSTCPFGGSGVSGWRLFKKIS

Reteplase

MEKKIWSHPQFEKGGSGENLYFQSGGASPAGGGNSDCYFGNGSAYRGTHSLTESGASCLPWNSMILIGKV
YTAQNPSAQALGLGKHNYCRNPDGDAKPWCHVLKNRRLTWEYCDVPSCSTCGLRQYSQPQFRIKGGLFAD
IASHPWQAAIFAKHRRSPGERFLCGGILISSCWILSAAHCFQERFPPHHLTVILGRTYRVVPGEEEQKFE
VEKYIVHKEFDDDTYDNDIALLQLKSDSSRCAQESSVVRTVCLPPADLQLPDWTECELSGYGKHEALSPF
YSERLKEAHVRLYPSSRCTSQHLLNRTVTDNMLCAGDTRSGGPQANLHDACQGDSGGPLVCLNDGRMTLV
GIISWGLGCGQKDVPGVYTKVTNYLDWIRDNMRPGGSGVSGWRLFKKIS

My progress: sfGFP draft assembled and sequence-checked

I started the optional section by building HW3_sfGFP_T7_v1 in Asimov Kernel. I used the sfGFP amino acid sequence provided in the assignment to make my own DNA coding part, with AI assistance. The 723 bp coding sequence includes start and stop codons, and its translation was checked to match the provided 240-amino-acid sequence.

I combined this part with public parts available in Kernel:

The construct map shows a total length of 2,887 bp. I added the four-base spacer TATT between the RBS and the coding sequence, giving a seven-base gap between the GAGGAG motif and the ATG start codon. This is an initial design choice, not an experimentally tested optimum.

I exported the draft as HW3_sfGFP_T7_v1.gb and checked the exported sequence with AI assistance. The file contains 2,887 bases and is marked as circular DNA. The sfGFP coding sequence translates exactly to the supplied 240-amino-acid sequence, with no internal stop codons. The part order and the seven-base RBS-to-start gap also match the design.

I also compared the backbone with the pUC19 reference sequence, L09137.2, using the NEB pUC19 map to locate its features. The 589 bp replication-origin region matches exactly. The full AmpR coding region has two synonymous DNA differences, but encodes the same 286-amino-acid protein.

These checks confirm the sequence matches the intended design; they do not show whether it will express well. I have not tested protein expression or completed the PETase and Reteplase plasmids.

C2. Bonus question: design a T7 helper plasmid that you could add to the cell-free reaction to improve the yield of functional reteplase (general chaperones, disulfide-bond machinery, etc).

I have not started the optional helper plasmid design.

AI assistance

I used OpenAI Codex as a discussion tool while working on this assignment. I asked for explanations of the readings and discussed how changes in magnesium, DsbC, and glutathione might affect cell-free protein expression and folding. These discussions helped me compare possible reaction changes. I made the final design choices.

For the optional section, I used AI for instruction for generating a DNA coding sequence from the provided sfGFP amino acid sequence, understanding the roles of the plasmid parts, and check the exported sequence. I assembled the construct in Asimov Kernel, using public parts and my own coding sequence.