materials · manufacturing · family: making one is easy. making a million is the problem
spider silk without thespider or the scale
Bio-fabricated structural proteins cannot be produced at textile-industry scale or cost
Problem statement
Recombinant structural proteins — particularly spider silk (spidroins) — are strong and elastic, biodegradable yet waterproof, and made from renewable raw materials, but after 15 years of development and over $330M raised, Bolt Threads has not produced them at textile-industry scale or cost: as of early 2017 the company had told C&EN it was targeting $100 per kg, against an analyst's threshold of under $10 per kg for competing with conventional textiles in mass-market applications, and by 2024 it had turned its b-silk protein toward beauty and personal care. Making the protein is not settled even in a pioneer's lab: Randy Lewis of Utah State University, a spider silk pioneer, told C&EN that production in E. coli was not yet sorted out and that the major challenge is a purification protocol that is both efficient and as cheap as possible. Spinning is a second hurdle: Bolt's commercial spinning process missed its promised summer-2016 date, and a Karolinska Institute study showed that fiber strength rises markedly when the silk is spun in a way that mimics the spider's spinning glands.
Why this matters
Spider silk is made from renewable raw materials, is biodegradable yet waterproof, and is strong and elastic; in 2017 Adidas, The North Face and Patagonia were jostling to be first with a mass-market product made from it, and cosmetics and medical-device developers (silk-coated catheters and implants, sutures, wound patches) were interested too. Bolt Threads' parallel mycelium leather product (Mylo) stalled short of commercial scale: it was paused in 2023 after the company was unable to raise additional funds to scale production, with its CEO saying the material was "very close" to commercial scale, and by October 2024 the company's market capitalization had fallen 94% in the two months after it went public. In the CEO's words, "Scaling is the most expensive and challenging part, and there are no shortcuts." Japan's Spiber has a factory in Thailand in production since 2022 and another under construction in Iowa, and consumer products with Goldwin and The North Face, but in 2017 it had publicly stated a commercial-scale cost of $20–30 per kg, above the under-$10-per-kg an analyst gave for competing with conventional textiles.
What’s been tried and why it hasn’t worked
Bolt Threads engineered yeast to produce recombinant spider silk protein and forms fibers by wet spinning, a process similar to the one used for Lyocell. Its commercial spinning process, promised for summer 2016, was reported in place in early 2017 as the company began transferring its lab-scale process to commercial-scale operations for three customers; by 2024 it had stopped Mylo, gone public through a SPAC, and turned b-silk toward personal care, counting on third-party factories that already run at large scale to bring prices down. On the protein side, Randy Lewis's lab, with two 500 L fermentation reactors, expected to make about 1 kg of protein per week in E. coli and named a cheap, efficient purification protocol as the major challenge; his best transgenic goats give 8 L of milk a day with a protein yield of 2 g/L. Spiber ferments its silk with recombinant E. coli and, like Bolt, delayed its first product (the $1,000 Moon Parka) without saying why. On spinning, the C&EN piece points to a Karolinska Institute study (Nat. Chem. Biol. 2017) in which spinning that mimics the spider's spinning glands markedly enhanced fiber strength.
What would unlock progress
Alternative expression hosts — Lewis is developing genetically engineered E. coli, goats, silkworms and alfalfa, and Kraig Biocraft uses transgenic silkworms that make a hybrid of worm and spider silk — could change the cost of the protein if purification can be made cheap and efficient, which Lewis calls the major challenge — though Lewis himself says spider silk will never be made for $10 to $15 per kg even with a perfect fermentation and purification process. Biomimetic spinning processes that mimic the spider's spinning glands, as in the Karolinska study, could improve fiber strength, as that study reported. Computational protein design that identifies minimal sequence motifs required for silk-like assembly could enable shorter proteins that are easier to produce, and alternative structural proteins (mussel byssus, hagfish slime thread, insect resilin) may offer better production economics for specific applications; neither of these two directions is discussed in the sources and they are offered here as ideas, not findings. Adding functionality — antibiotic or fluorescent ligands conjugated to spider silk, as in Neil Thomas's lab at the University of Nottingham — is, in the C&EN piece's words, "a likely key to long-term success."
Entry points for student teams
A team could prototype a biomimetic spinning device that imposes controlled pH and shear gradients on a recombinant silk protein solution, characterizing fiber structure as a function of spinning parameters. Alternatively, a team could use computational protein design tools (Rosetta, AlphaFold) to identify minimal beta-sheet motifs that self-assemble into silk-like structures, then test expression in E. coli. Relevant disciplines: materials science, bioengineering, mechanical engineering, computational biology.
Genes
Structural cousins: stuck the same way, in other fields
Sources
LUXEPLACE, "Once Famous for “Spider Silk” and “Mushroom Leather,” Why Did Bolt Threads’ Valuation Crash 94% in Two Months?", 18 October 2024, Alex Scott, "Delivering on spider silk’s promise," C&EN, vol. 95, issue 8, 20 February 2017, Sarah Kent, "Bolt Threads Pauses Operations of Leather-Alternative Mylo," The Business of Fashion, 4 July 2023, accessed 2026-02-20 go to source 1 ↗ go to source 2 ↗ go to source 3 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
Bolt Threads (over $330M raised; market capitalization down 94% in the two months after its August 2024 listing) is the most prominent case, but the scale-up challenge is shared across the field. Spiber (Japan) has a factory in Thailand and another under construction in Iowa, and consumer products with Goldwin and Sacai; Kraig Biocraft (transgenic silkworms; a new factory with an annual capacity of 25 tons, eight times the size of its previous facility, began production in September 2024) and AMSilk (Germany; protein made by E. coli fermentation, sold for cosmetics since 2014, acquired by Givaudan in 2022) take different routes. Related: chemistry-synbio-commodity-molecule-cost-floor (same fermentation cost-floor challenge but for small molecules rather than structural proteins).
Reconciliation 2026-09-30: Three sources were read. The LUXEPLACE article (https://luxeplace.com/once-famous-for-spider-silk-and-mushroom-leather-why-did-bolt-threads-valuation-crash-94-in-two-months/, fetched 2026-09-30, full text, dated 18 October 2024) and the C&EN feature by Alex Scott (https://cen.acs.org/articles/95/i8/Delivering-spider-silks-promise.html, fetched 2026-09-30, full text, 20 February 2017, vol. 95 issue 8). The Business of Fashion article by Sarah Kent (4 July 2023) is paywalled and blocks scripted requests; a 2026 Internet Archive copy gave only the headline, byline, date and opening sentence, and an earlier capture (web.archive.org/web/20240116075222/www.businessoffashion.com/news/sustainability/bolt-threads-mylo-alternative-leather-mushroom-pause-operations/) carries the full text, which says Bolt Threads paused Mylo "after struggling to fund raise," that products containing the material "have been slow to reach the market and the company has been unable to raise additional funds to scale production," and quotes the company's statement that the macroeconomic climate made it difficult "to secure the necessary capital to support the scale up of emerging technologies." (1) The Problem Statement said spider silk's strength-to-weight ratio exceeds Kevlar and its toughness exceeds steel, that its properties are superior to any synthetic fiber, and that Bolt could not reach prices competitive with even premium textiles. None of the sources says any of this; C&EN says spider silk is "strong and elastic," "biodegradable yet waterproof," that Bolt was targeting $100 per kg, and that an analyst put the threshold for competing with conventional textiles at under $10 per kg. The paragraph now says that, with the $100 per kg figure dated to early 2017. (2) The same section explained low yields by repetitive, aggregation-prone sequences, batch failures from small changes in temperature, pH and viscosity, and spinning that must replicate a hierarchical nanostructure. No source says this. The text now carries what C&EN does report: Randy Lewis saying E. coli production is not yet sorted and that cheap, efficient purification is the major challenge; Bolt's spinning process missing its promised summer-2016 date; and a Karolinska Institute study in which spinning that mimics the spider's glands markedly enhanced fiber strength. (3) Why This Matters said the textile industry is the third-largest polluter globally and that synthetic fibers are about 60% of global fiber production, and named collagen and elastin applications. None of this is in the sources; removed. The applications now listed (footwear and apparel, cosmetics, catheter and implant coatings, sutures, wound patches) are C&EN's. (4) The same section said Spiber "has made the most progress," operates at "relatively small scale," and has "premium pricing far above commodity synthetics." The sources do not rank the companies or describe Spiber's scale; LUXEPLACE reports a factory in Thailand in production since 2022 and one under construction in Iowa, and C&EN reports Spiber's stated commercial-scale cost of $20–30 per kg. The sentence now says that, with the cost figure dated to 2017. The scare-quoted phrase "valley of death" was not in any source and is gone; the Mylo sentence now says what BoF reports (slow to reach the market; unable to raise additional funds to scale production) and carries the CEO's quoted words on scaling from LUXEPLACE. (5) What's Been Tried explained low yeast yields by DNA-repair clipping of repetitive genes, purification cost by chaotropic agents, and the spinning problem by a liquid crystalline phase transition, and said Spiber uses a fibroin-like protein with lower mechanical performance. None of this is in the sources; C&EN says Spiber ferments its silk with recombinant E. coli. The section now reports what the two readable sources say about Bolt's process and timeline, Lewis's fermenters, yields and goats, Spiber's delay, and the Karolinska study. (6) What Would Unlock Progress attributed yield gains to avoiding "microbial recombination of repetitive silk genes" and listed spinneret pH, ionic and shear gradients; these mechanisms are not in the sources. The section now names the alternative hosts C&EN reports (Lewis's E. coli, goats, silkworms and alfalfa; Kraig's transgenic silkworms), carries Lewis's own caveat that spider silk will never be made for $10 to $15 per kg even with a perfect fermentation and purification process, drops the unsourced clause that biomimetic spinning would work "without requiring the protein to be identical to natural spidroins," keeps the computational-design and alternative-protein ideas but marks them as not discussed in the sources, and adds C&EN's own closing point that adding functionality is "a likely key to long-term success." (7) Source Notes said Spiber has raised about $800M and pivoted to more processable proteins at the expense of performance, called AMSilk "fibroin-based," said Kraig and AMSilk take "biological rather than fermentation-based approaches," and said the protein case is worse than small molecules because proteins aggregate and misfold. No source gives a Spiber funding figure or a pivot; C&EN says AMSilk's protein is made by E. coli fermentation, which contradicts the "rather than fermentation-based" claim. The paragraph now carries what LUXEPLACE and C&EN report about Spiber (a factory in Thailand, one under construction in Iowa), Kraig Biocraft and AMSilk, and the cross-reference is kept without the aggregation claim. (8) The Source line now carries the LUXEPLACE title in full, the C&EN author, volume, issue and date, and the BoF author, date and address, all copied from the pages as fetched. Confirmed in the sources and left as written: Bolt founded 2009 and fifteen years in the field; over $330M raised; the 94% fall in market capitalization in two months; yeast fermentation and wet spinning; Mylo paused in 2023 for lack of funding; the entry points. All three sources were read in full. Titles unchanged. Genome Tags untouched. A second reader went over the repairs on the same day, read the BoF body through the January 2024 archive capture, and reworded six sentences so that each says no more than its source: the goat yield is now attributed to the best goats (8 L a day at 2 g/L), as C&EN says; two 2017 cost figures are dated; the Mylo sentence follows BoF's wording; the claim that spinning matters "as much as" making the protein, which no source makes, is now "a second hurdle"; and Lewis's price caveat is included.