Thread by @elliotswartz

If you asked a room of cell biologists 5 years ago if cell culture media could productively grow cells while costing just $0.63 per liter, you would have been laughed out of it. No longer.

Let's dive into this important study that poses challenges for #cultivatedmeat skeptics:

The 1st innovation was developing a way to replace recombinant albumin in serum-free media. Albumin is critical in most serum-free media, but makes up >1/3 of the cost. It's been shown that albumins from plants can functionally replace recombinant animal albumin at low cost:

But this study takes it one step further. They show that low-cost, food-grade ingredients such as methylcellulose, cyclodextrin, & antioxidants can individually replace the functions of albumin (shear protection, nutrient carrier, oxidation) in a protein-free manner.

At $0.63/L, this makes it the lowest-cost media published to date. It's also fully animal-component-free.

Importantly, the media cost is ~80% lower than Humbird's analysis (~$3/L), which has been used by skeptics to question CM's economic viability:

thecounter.org/lab-grown-cult…

The media also performs well. They were able to generate max cell densities of 130M cells/mL using tangential flow filtration (TFF) perfusion at lab-scale, which outcompetes alternating tangential flow filtration (ATF) due to higher flux rates:

I will say with confidence that other cultivated meat companies are also in this media cost range.

And that I'm confident we will see even lower published media costs in the near future, as originally predicted by @LizSpecht:

Continuous experiments were run for 20 days at densities of ~20M cells/mL (~200g/L by back-calculating cell mass at ~9.8ng, which is >3x higher than what previous CM modeling studies used)

More empirical cell mass data are needed to clarify what to use in future models.

The product is ~80% protein and 13% fat on a dry mass basis, with consistent values generated across successive harvests:

Importantly, the lactate/glucose ratio was 0.52, which is close to Humbird's "enhanced metabolism" value of 0.50.

This implies that the chicken fibroblast lines have an efficient metabolism without needing additional modification:

Back-calculating data in the paper shows this to be the case, with 232g AAs, 409g sugar, & 436g salts/vitamins per kg of wet biomass

This aligns well w/ our previous LCA estimates, suggesting the feed conversion here is nearly 3x as efficient as conventional chicken production:

These empirical cell and media data provide support for the favorable climate impact of #cultivatedmeat compared to conventional animal agriculture, outlined here:

https://x.com/elliotswartz/status/1638508832929488897

They then use this empirical data for the cells and media to calculate total costs in a hypothetical 50,000L facility using 2kL x 25 ATF reactors, 5kL x 10 reactors for TFF, and 25kL x 2 for future large-scale perfusion

Costs come out to $21.49 to $38.54/kg wet mass:

Media remains the cost driver at around 60-70% of total cost.

TFF's reusable filters show significant cost advantages for manufacturing & consumables compared to ATF, resulting in a lower total capital investment for the facility:

They model the final product cost at a 50% inclusion rate, with a 50/50 product being split between CM & plant-based ingredients, which contribute $3/kg to the total cost.

Altogether, hybrid product costs are estimated at $6.22 to $10.08/lb, in the range of organic chicken:

Speaking to others, some costs in the supplemental tables (e.g., labor, buildings, equipment) may be optimistic. But this is outside my expertise to judge, so I defer to others for feedback

Regardless, w/ media making up the bulk of costs, inaccuracies here carry less weight

In summary, this paper challenges previous cost models & the prevailing perception that #cultivatedmeat is not economically viable.

It's the 1st study to model costs based on real empirical data, thus it should IMO be viewed as a new standard for comparison.

That said, the data are primarily lab-scale & it remains to be seen if these numbers will be borne out in a real, scale-up facility

@believermeats can prove this to the world in their commercial facility opening in North Carolina in the coming months.

It's also important to note that the data presented here are now well over a year old, and it's possible that further improvements have been made in the interim time to publication.

vegconomist.com/company-news/b…

The team @believermeats led by @KNahmias31077 deserves a lot of praise

Openly publishing data benefits everyone & will further trust in the technological readiness of #cultivatedmeat. There are many potential ways to manufacture CM, but the scientific ecosystem collectively needs detailed studies like this to resolve the most prudent paths forward

I truly hope this will spur other companies to follow in their footsteps

The full article can be found below.

Please get in touch if you notice any inaccuracies in my calculations, the paper, or have strong opinions on the costing details. I want to hear about them!

nature.com/articles/s4301…

@threadreaderapp unroll