Week 1 HW: Final Project Ideas

HTBAA Homework #1: Biomanufacturing Proposal Aya Ahmed Jalil Iraq-Baghdad 22Sep26

1. What I would like to biomanufacture and why?

I propose to biomanufacture recombinant spider silk nanoparticles encapsulating a topical corticosteroid (e.g., dexamethasone or hydrocortisone) for controlled, sustained transdermal drug delivery. The target product is a biocompatible, biodegradable nanoparticle formulation that can be incorporated into a cream or hydrogel for topical application, releasing the anti-inflammatory corticosteroid over an extended period to treat chronic skin conditions such as eczema, psoriasis, or dermatitis.

My motivation is both utilitarian and personal. Topical corticosteroids are a mainstay of dermatological therapy, but their efficacy is often limited by rapid clearance from the skin, requiring frequent reapplication and risking side effects like skin atrophy. A spider silk-based delivery system could provide sustained release, reduce application frequency, and improve patient compliance. Spider silk proteins (spidroins) are renowned for their exceptional biocompatibility, slow biodegradation, and tunable mechanical properties, making them ideal candidates for drug delivery. Recombinant spider silk nano- and microcapsules have already been demonstrated to encapsulate a wide range of pharmaceutical agents, including corticosteroids such as dexamethasone and other glucocorticosteroids. The ability to functionalize spider silk coatings with therapeutic agents that release on a predictable timescale has been patented, specifically listing dexamethasone as an example.

2. What living system would I use?

I would use the methylotrophic yeast Pichia pastoris (also known as Komagataella phaffii) as the production chassis for the recombinant spider silk protein. P. pastoris is a well-established expression host for large, repetitive spidroin genes, offering significant advantages over bacterial systems: it achieves stable gene integration, reduces protein truncation, and enables extracellular secretion, which simplifies downstream purification. It can be grown to very high cell densities in fed-batch fermentation, and synthetic spider dragline silk protein analogs have been readily expressed at high levels under the control of the methanol-inducible AOX1 promoter.

How it would work: A synthetic gene encoding a spider silk protein—such as a MaSp1 repeat analog from Trichonephila clavipes, which has been expressed in P. pastoris at yields of ~663 mg/L—would be codon-optimized and placed under the AOX1 promoter. The yeast would be cultured in a minimal medium with glycerol, then switched to methanol for induction. The silk protein would be secreted into the culture supernatant. After harvest and purification, the silk protein would be processed into nanoparticles via a phase-separation or emulsion-based method, during which the corticosteroid would be encapsulated. The drug-loaded nanoparticles would then be formulated into a topical cream or hydrogel.

Assumption: I assume the secreted silk protein can be recovered at >90% purity and that encapsulation efficiency of the corticosteroid is sufficient (e.g., >50%) for a viable therapeutic dose. I also assume the expression titer is in the range of 0.5–1.0 g/L, which is conservative but realistic for large repetitive silk proteins based on reported yields for similar constructs.

3. How would I use automation?

I would use a desktop liquid-handling robot (Opentrons OT-2 or Flex) for high-throughput screening of expression and encapsulation conditions, and a cloud lab / biofoundry for scale-up and process optimization.

Automated step: The Opentrons platform would automate high-throughput screening of spider silk gene variants and encapsulation formulations. The APEX (Automated Protein Expression) pipeline demonstrates that the Opentrons OT-2 can be configured to perform heat shock transformation, selective plating, colony sampling, microculturing, and protein expression induction with high precision and reproducibility. For this project, I would adapt APEX to P. pastoris to:

· Screen different silk gene constructs (e.g., varying repeat lengths, solubility tags) for expression titer and protein integrity. · Test different nanoparticle formulation conditions (e.g., silk concentration, solvent, corticosteroid ratio) in 96-well format. · Automate sampling and preparation of supernatants for protein quantification and drug release assays.

This automation would accelerate the design-build-test-learn cycle, reducing human error and enabling rapid optimization of both the silk production and nanoparticle encapsulation steps.

Why automate: Spider silk proteins are large, repetitive, and difficult to express; the design space of gene constructs, process conditions, and formulation parameters is enormous. Manual experimentation would be slow and inconsistent. Automation enables rapid iteration and reproducible data generation. For scale-up, a cloud lab with robotic bioreactors and automated downstream processing could run parallel 1–5 L fed-batch fermentations to identify the best scale-up parameters before transferring to a 1000 L single-use bioreactor.

4. How much would I need to make, and how long would it take?

(a) How much?

I define the scale as: the amount of drug-loaded spider silk nanoparticles required to treat 10,000 patients for one month of topical therapy.

· Assume a typical topical corticosteroid cream contains 0.1% (w/w) corticosteroid (a common concentration for hydrocortisone or dexamethasone formulations). · Assume a patient uses 30 g of cream per month (approximately 1 g/day). · Total cream required = 10,000 patients × 30 g/patient = 300,000 g = 300 kg. · Total corticosteroid required = 300 kg × 0.001 = 0.3 kg = 300 g. · Assume the spider silk nanoparticle carrier is formulated at 5% (w/w) in the cream. · Total silk nanoparticle mass required = 300 kg × 0.05 = 15 kg. · Assume 70% encapsulation efficiency of the corticosteroid in the silk nanoparticles. · Total corticosteroid needed for encapsulation = 300 g / 0.70 = ~429 g. · Total silk protein required (assuming nanoparticles are ~90% silk protein by mass) = 15 kg / 0.90 = ~16.7 kg.

Assumption: The 0.1% corticosteroid concentration and 30 g/month usage are estimates based on typical topical corticosteroid dosing. Actual values would depend on the specific drug and formulation. This calculation is intended as a back-of-the-envelope estimate.

(b) How long?

Production system: 1,000 L fed-batch bioreactor with P. pastoris.

· Assume a conservative secreted silk protein titer of 0.5 g/L at harvest. (This is lower than the 663 mg/L reported for a MaSp1 repeat in P. pastoris but appropriate for a large, complex silk protein intended for nanoparticle formation.) · Protein per 1,000 L batch = 1,000 L × 0.5 g/L = 500 g = 0.5 kg. · Number of batches required for 16.7 kg = 16.7 kg / 0.5 kg/batch = 33.4 batches (round to 34 batches).

Batch time: A typical P. pastoris fed-batch fermentation for recombinant protein production takes approximately 120–168 hours (5–7 days) from inoculation to harvest. Assume 6 days per batch for a 1,000 L run, including turnaround (cleaning, sterilization, setup).

Parallelization: If the facility operates two 1,000 L bioreactors in parallel, then:

· Effective batches per 6-day cycle = 2. · Total cycles needed = 34 batches / 2 = 17 cycles. · Total time = 17 cycles × 6 days/cycle = 102 days ≈ 3.4 months.

This does not include downstream processing (protein concentration, purification, nanoparticle formation, drug loading, cream formulation), which could add another 4–6 weeks. So the total timeline from start of fermentation to finished cream is roughly 4.5–5 months.

Prototyping vs. manufacturing: For prototyping, the Opentrons and small-scale (1–5 L) bioreactors would be used to optimize titer, protein integrity, and encapsulation efficiency. The 1,000 L scale would only be used once the process is robust. If titer can be improved to 2 g/L through strain and process engineering (as has been achieved for other spidroin constructs, with bioreactor yields reaching 25 g/L in some engineered systems), the number of batches would drop to ~8.4 batches, and total fermentation time could be under 6 weeks with parallel reactors.

5. What don’t I know yet?

6. Encapsulation efficiency and release kinetics of corticosteroids in spider silk nanoparticles. While spider silk particles have been used for drug delivery, the release profile of small, hydrophobic corticosteroids from silk nanoparticles is not well characterized. I would need to experimentally determine the loading capacity, encapsulation efficiency, and in vitro release kinetics for the specific drug and silk construct. Early studies suggest that release from spider silk particles can be very fast at physiological conditions, indicating that incorporation into the protein matrix may not always lead to sustained release. This is a critical bottleneck that would need to be addressed through formulation optimization.

7. Secretion efficiency and protein integrity of large silk proteins in P. pastoris at scale. While P. pastoris is superior to E. coli for spidroin expression, the secretory pathway may still limit yields for highly repetitive, high-molecular-weight proteins. I would need to measure the actual titer, the proportion of full-length protein, and the extent of proteolytic degradation in progressively larger bioreactors (1 L → 10 L → 100 L → 1000 L). Oxygen transfer and shear stress at large scale could also affect protein quality.

8. Stability and shelf-life of the final cream formulation. Spider silk proteins are biodegradable, which is advantageous for the drug delivery application but could also limit the shelf-life of the cream. I would need to determine the stability of the drug-loaded nanoparticles in the cream base over time, including the effects of temperature, pH, and preservatives on both the silk carrier and the corticosteroid. This would require accelerated stability studies and long-term real-time stability testing. References

9. Lammel A, Schwab M, Hofer M, Winter G, Scheibel T. Recombinant spider silk particles as drug delivery vehicles. Biomaterials. 2011;32(8):2233–2240.

10. Hofer M, Winter G, Myschik J. Recombinant spider silk particles for controlled delivery of protein drugs. Biomaterials. 2012;33(5):1554–1562.

11. Fahnestock SR, Bedzyk LA. Production of synthetic spider dragline silk protein in Pichia pastoris. Applied Microbiology and Biotechnology. 1997;47(1):33–39.

12. Tokareva O, Michalczechen-Lacerda VA, Rech EL, Kaplan DL. Recombinant DNA production of spider silk proteins. Microbial Biotechnology. 2013;6(6):651–663.

13. Heidebrecht A, Scheibel T. Recombinant production of spider silk proteins. Advances in Applied Microbiology. 2013;82:115–153.

14. Widhe M, Johansson J, Hedhammar M, Rising A. Current progress and limitations of spider silk for biomedical applications. Biopolymers. 2012;97(6):468–478.

15. APEX: Automated Protein Expression in Opentrons OT-2. Opentrons application note / bioRxiv preprint. Verify exact citation details before submission. I used “DeepSeek” AI application, the prompt I used: “My Idea is to biomanufacturer spider silk drug delivery system such as topical corticosteroids answer this homework assignment based on research studies and provide them for me with each answer”

Final Project