LOADING LOADING

Introduction

Our wet lab efforts focus on simulating a seamless experimental workflow that starts with AI sequence generation and ends with application-ready spider silk based fiber, gel, and film; we hope that by first establishing a clear lab-scaled platform, we can begin to standardize the tedious process of synthetic biomaterial production and accelerate its eventual scaling for larger yields. To obtain accurate and verified information as “ground-truths” to train GPSS (Generative Personalized Spider Silk), our LLM-based sequence generation tool, we also designed and performed specified property testings on our post-processed spider silk materials.

Given the complex nature of the entire process, this page acts as an overview outlining each stage of SAC’s platform. To view the experimental results corresponding to each stage of this overview, visit our engineering success page.

Guideline

    • Our software tool, GPSS, is capable of generating custom sequences with 4 key mechanical properties:

      • Toughness (T): How much energy the silk can absorb before breaking.
      • Elastic Modulus (E): How much the silk stretches when force is applied.
      • Tensile Strength (TS):: How much force the silk can withstand before breaking.
      • Strain at Break (S): The extent to which the silk stretches before it breaks.

      We used GPSS to create four unique spider silk protein sequences, each optimized for different mechanical properties based on their potential applications. Property levels are normalized on a 1 - 10 scale (1 being the lowest level and 10 being the highest) and the nomenclature for each sequence follows: T[1-10]-E[1-10]-TS[1-10]-S[1-10]. The generated sequences are:

      1. T9-E9-TS9-S9: This sequence offers high toughness, elasticity, and tensile strength, making it ideal for industries like aerospace and construction. It’s perfect for applications where both strength and flexibility are required, such as in airplane wings or parachute ropes[1].
      2. T9-E9-TS9-S1: Designed for protective gear, this sequence offers high toughness and strength but low strain at break, making it excellent for bulletproof vests and helmets, where rigidity under stress is key[2].
      3. T7-E7-TS7-S3: With balanced properties, this sequence is well-suited for sportswear and medical sutures, offering both strength and flexibility for activewear or post-surgery recovery materials[3].
      4. T3-E3-TS3-S7: This sequence is optimized for high tensile strength but lower toughness and elasticity, making it ideal for structural reinforcement in cables or bridges, where stretching is minimal, but strength is critical[4].
      sample2

      T9-E9-TS9-S9

      High toughness, elasticity, and tensile strength. Ideal for aerospace and construction (e.g., airplane wings, parachute ropes).

      sample3

      T9-E9-TS9-S1

      High toughness and strength with low strain at break. Suitable for protective gear (e.g., bulletproof vests, helmets).

      sample4

      T7-E7-TS7-S3

      Balanced properties. Well-suited for sportswear and medical sutures (e.g., activewear, post-surgery materials).

      sample4

      T3-E3-TS3-S7

      High tensile strength with lower toughness and elasticity. Perfect for structural reinforcement (e.g., cables, bridges).

    • To produce these AI-generated spider silk sequences, they need to be introduced into E. coli bacteria. We used pET28a-T7-mEGFP as our plasmid backbone to control and optimize protein production; it has the following genetic elements:

      • T7 Promoter: Ensures high-level production of spider silk proteins.
      • lacO and IPTG: These elements allow us to control when the protein production starts. By adding IPTG, we can turn the production on or off as needed.
      • mEGFP Reporter: A green fluorescent protein that helps us visually track and measure the amount of spider silk being produced.

      Due to pET28a-T7-mEGFP’s favorable genetic elements, we designed our silk production plasmids by putting the 4 GPSS-generated sequences downstream of T7 and lacO and upstream of mEGFP to optimize production and quantification (shown in figure #1). All designs were performed and previewed via Benchling.

      Construct Designs

      1. PCR extension & Gibson Assembly

        Spider silk proteins contain many repeating Alanine and Glycine residues, which form beta-sheet structures that contribute to the silk’s extraordinary strength[5]. These repetitive sequences, however, prove to be challenging for current commercial synthesis techniques to produce; to overcome this limitation, we synthesized sequences in fragments and used a combination of Primer Extension and PCR-extension techniques to build the sequences manually. Once the fragments are joined, we use Gibson Assembly to combine the spider silk sequences with pET28a backbone, creating the final plasmids. Transformation into NEB® Stable Competent E. coli strain is performed for preservation and future use.

      2. Multi-Step DNA Sequence Validation

        The complexity of spider silk sequences calls for multiple validation methods to confirm sequence accuracy; after assembly, we performed double digestion (cut at existing restriction sites on the plasmid) and colony PCR (primer specified extension) to verify plasmid size and integrity. After both double digestion and colony PCR yield successful results, the plasmids are sent out for sequencing to confirm an exact sequence match.

      3. Induced Small-Scale Protein Expression

        We transformed the validated plasmids into BLR(DE3) E. coli for protein expression. FLEXS (Our real-time data collecting hardware)-equipped shake flasks were used here to monitor our small-scale protein expression not only to grow bacteria to peak state but also to identify optimal conditions for the next stage of SAC’s workflow: large-scale fermentation; the metrics monitored include pH, temperature, optical density (OD), and fluorescence.

      4. Multi-Step Protein Validation

        The induced bacterial cultures are subjected to SDS-PAGE and Western blotting to qualitatively confirm spider silk protein production. After initial confirmation of protein presence from SDS-PAGE, Western blotting is performed by using mEGFP antibodies for further verification. Finally, microscope observation is then used to observe fluorescence and micelle formation.

      1. Fermentation

        After we have confirmed that the protein expressed is our desired spider silk, we are prepared to transfer the FLEXS-optimized bacteria samples to a 5-liter bioreactor for large-scale fermentation and purification; this process will yield enough protein samples for post-processing and property testing in stage 4.

        Here, we utilized a mixture of molasses wastewater (a byproduct of sugar refining) and crude glycerol (a byproduct of biodiesel refining) as a sustainable carbon source for bacteria cultivation; pH, temperature, and nutrient levels are monitored and optimal conditions defined by FLEXS is maintained to maximize bacteria density and protein yield. Our results concluded that our industrial waste based carbon source is on-par with commercial mediums.

      2. Purification

        After maximum density is reached, sonication is performed to disrupt and lyse bacterial cells, releasing the spider silk proteins. Initial centrifugation is performed to separate cell debris from silk proteins and SDS (Sodium Dodecyl Sulfate) used to treat the proteins, dissolving impurities to improve protein solubility. A second and final centrifugation is performed to pellet the final silk proteins; the proteins are then washed with ddH2O to obtain a final yield of 1.7g/L, a promising result that brings us closer to large-scale production

    • After purification, we conducted three different post-processing steps to transform spider silk proteins into usable materials for various industrial applications. Each method allows us to tailor the silk for specific uses, ensuring application versatility.

      HFIP Dissolution

      • We use Hexafluoro-2-propanol (HFIP) to dissolve the spider silk proteins before further processing. HFIP is effective in breaking down the strong hydrogen bonds that stabilize the protein's beta-sheet structures, allowing the silk to be reconstituted into different forms. This solvent ensures that the silk maintains its structural integrity while being reshaped for industrial applications[6].

      Post-Processing Methods

      1. Fiber Spinning:
        • After dissolving in HFIP, the spider silk is spun into fibers. These fibers are used in high-performance materials such as textiles, aerospace components, and other industries where strength and flexibility are crucial.
      2. Gelation:
        • The silk can also be processed into hydrogels, which are particularly useful in medical applications, such as wound dressings and surgical sutures. The hydrogels provide a biodegradable and biocompatible alternative to synthetic medical materials[7].
      3. Film Casting:
        • Thin films of spider silk can be cast and used as biodegradable packaging materials, replacing plastics in industries that require eco-friendly solutions[8].

    • Once post-processing is complete, we conduct thorough testing to verify the mechanical properties of the spider silk, specifically focusing on fibers and films, to validate the predictions made by our AI models. The resulting data can be fed back into GPSS as “Ground Truths” to train our sequence generation’s accuracy, creating a feedback loop. Our testing methods are:

      1. Mechanical Testing:
        • Using a tensile machine, we can measure the key mechanical properties that match our sequence generation metrics: Strength, Elasticity, Tensile Strength, and Strain at Break.
      2. X-ray Diffraction (XRD) Analysis:
        • We also perform X-ray diffraction analysis (XRD) to study the crystalline structure of the spider silk proteins. By analyzing the diffraction patterns, we can determine crystalline arrangement, nanocrystal size, and degree of crystallinity. This allows us to pinpoint the specific regions of the silk protein that contribute to its superior mechanical properties.

      By combining mechanical testing and XRD, we gain valuable insights into how the AI-generated spider silk sequences behave at both macroscopic and microscopic levels, ensuring that our materials meet various industry standards while providing feedback to improve the design process.

    • This workflow acts as a proof of concept for a larger scaled production platform. In the future, we hope to achieve automation of the labor-intensive steps within this workflow and create a more seamless pipeline for large yield biomaterial production.

References

    • [1] Babu, K. M. (2017). Silk from silkworms and spiders as high-performance fibers. In Structure and properties of high-performance fibers (pp. 327-366). Woodhead Publishing.

      [2] Kluge, J. A., Rabotyagova, O., Leisk, G. G., & Kaplan, D. L. (2008). Spider silks and their applications. Trends in biotechnology, 26(5), 244-251.

      [3] Liu, Y., Huang, W., Meng, M., Chen, M., & Cao, C. (2021). Progress in the application of spider silk protein in medicine. Journal of Biomaterials Applications, 36(5), 859-871.

      [4] Li, J., Li, S., Huang, J., Khan, A. Q., An, B., Zhou, X., ... & Zhu, M. (2022). Spider silk‐inspired artificial fibers. Advanced Science, 9(5), 2103965.

      [5] Adrianos, S. L., Teulé, F., Hinman, M. B., Jones, J. A., Weber, W. S., Yarger, J. L., & Lewis, R. V. (2013). Nephila clavipes flagelliform silk-like GGX motifs contribute to extensibility and spacer motifs contribute to strength in synthetic spider silk fibers. Biomacromolecules, 14(6), 1751-1760.

      [6] Brooks, A. E., Stricker, S. M., Joshi, S. B., Kamerzell, T. J., Middaugh, C. R., & Lewis, R. V. (2008). Properties of synthetic spider silk fibers based on Argiope aurantia MaSp2. Biomacromolecules, 9(6), 1506-1510.

      [7] Khan, A. Q., Shafiq, M., Li, J., Yu, K., Liu, Z., Zhou, X., & Zhu, M. (2023). Recent developments in artificial spider silk and functional gel fibers. SmartMat, 4(6), e1189.

      [8] Spiess, K., Lammel, A., & Scheibel, T. (2010). Recombinant spider silk proteins for applications in biomaterials. Macromolecular bioscience, 10(9), 998-1007.