Latest / The Manufacturing Podcast with Fexingo: Factories, Supply Chains, and Industrial Business / How a Japanese Factory Is Making Salmon from Cell Cultures
Transcript
- Lucas: Imagine a salmon fillet that was never swam in an ocean, never raised in a pen, never caught by a boat. It was grown, cell by cell, inside a stainless steel tank in a factory outside Tokyo. Luna: You're talking about cultivated seafood. And I know we've covered cellular agriculture before, but salmon feels like a particularly ambitious target. It's not a ground-up patty. It's a structured, fatty, high-value product. Lucas: Exactly. And that's what makes this Japanese factory — run by a company called IntegriCulture — so interesting. They've been operating a pilot facility since late 2024, and they're now producing what they call 'cell-based salmon' at a scale that's starting to attract serious attention from major Japanese food conglomerates. Luna: What's their secret? Because the big barrier in cultivated meat has always been cost. The growth media — the broth of nutrients that feeds the cells — is incredibly expensive. Lucas: Right. The industry standard uses fetal bovine serum, which is both costly and ethically problematic. IntegriCulture developed something called a 'serum-free' culture medium. They basically engineered a cocktail of growth factors and amino acids that costs roughly one-tenth of the traditional stuff. Their CTO told me their media cost is now under $50 per liter. Luna: That's a game changer if it scales. But salmon isn't just muscle cells. You need fat for that marbling, you need connective tissue for that flaky texture. How do you build structure in a tank? Lucas: They use a scaffold — a 3D structure made from a modified plant protein, I believe it's soy-based — that the cells attach to and grow around. The scaffold dissolves during the final processing step, leaving behind a tissue that mimics the muscle and fat distribution of a real salmon fillet. They've shown samples to chefs who couldn't tell the difference in a blind taste test. Luna: So the product is there, or close to it. But we're talking about a factory, not a lab. What does production look like at scale? Lucas: Their pilot plant has a series of 500-liter bioreactors — think of them as large, precisely controlled fermentation tanks. They run batches in parallel, each batch producing about 20 kilograms of salmon tissue after a growth cycle of roughly three weeks. That's not much — maybe 44 pounds — but they're already planning a facility with 10,000-liter reactors that could produce several tons per month. Luna: And the cost? What does a pound of cultivated salmon cost to produce today versus what you'd pay at a fish counter? Lucas: Right now, their estimated cost is around $200 per kilogram — about $90 per pound. That's obviously far above the retail price of wild-caught salmon, which can be $15 to $30 per pound depending on the market. But their roadmap shows cost dropping to $50 per kilogram by 2028, and under $20 per kilogram by 2030, which would put it on par with premium farmed salmon. Luna: That trajectory depends on a lot of things — cheaper media, higher yields, regulatory approval. Where does regulation stand in Japan? Lucas: Japan's Ministry of Health, Labour and Welfare has been proactive. In 2023, they established a regulatory framework specifically for cell-based foods. IntegriCulture submitted their safety dossier in early 2025, and they're expecting approval for market sale by the end of this year. That would make Japan one of the first countries to approve cultivated fish for commercial sale. Luna: Singapore approved cultivated chicken back in 2020. But fish is different — there's more complexity in the product structure, and also in consumer perception. Japanese consumers are particularly discerning about seafood quality. Lucas: And that cuts both ways. On one hand, high standards mean a higher bar. On the other, Japanese consumers already pay a premium for top-quality fish. If the product is genuinely indistinguishable, there's a market. The company is already in talks with high-end sushi chains and hotel groups. Luna: Let's talk about the bigger picture. Why does this matter beyond just being a cool tech story? We've got overfishing, climate change affecting wild stocks, and aquaculture has its own environmental costs — feed, waste, antibiotic use. Lucas: Exactly. Global salmon demand has been growing at about 5 percent annually for the past decade. Wild catch is capped by quotas, and aquaculture is running into limits on suitable coastal sites. Cell-based production could decouple salmon supply from ocean ecosystems entirely. One study estimated that cultivated salmon could reduce land use by 90 percent and water use by 80 percent compared to conventional aquaculture. Luna: And there's the supply chain angle. Right now, most of the world's salmon comes from Norway, Chile, and Scotland. A factory in Tokyo could serve the Japanese market directly, cutting shipping costs and carbon footprint. It's a form of food localism, ironically, using high tech. Lucas: Absolutely. And we're seeing similar efforts in other countries. There's a company in San Francisco working on cultivated bluefin tuna. Another in Singapore doing shrimp. But Japan is uniquely positioned because of its strong culinary culture around raw fish, its advanced biotech sector, and government support. If this works, it could be a template for other cultured seafood products. Luna: Before we get too deep into the future, I want to ground this in the present. What are the biggest hurdles between here and a $20-per-kilogram salmon fillet on a supermarket shelf? Lucas: Media cost is still the number one factor. Even at $50 per liter, media is a huge portion of the variable cost. They need to get that down to under $10 per liter. That means either engineering the cells to require fewer growth factors, or finding cheaper sources for those factors — maybe from yeast or plant-based production. Luna: And yield. Not every cell turns into tissue. You lose some in the scaffold, some during harvesting. What are their yields looking like? Lucas: They're currently at about 60 percent yield — meaning 40 percent of the cell mass doesn't make it into the final product. They think they can push that to 85 percent with better bioreactor design and harvesting techniques. Every percentage point matters when you're scaling to thousands of liters. Luna: Another challenge I imagine is the supply chain for the inputs themselves. The growth factors, the amino acids, the scaffold material — these are all specialty chemicals. If you're going to scale, you need reliable, cost-effective suppliers. Lucas: IntegriCulture has actually started producing their own growth factors using genetically modified yeast, similar to how insulin is made. That vertical integration is key. They don't want to be dependent on a handful of biotech suppliers that might charge premium prices. They're essentially building a miniaturized supply chain inside their factory. Luna: It reminds me of how early microchip manufacturers had to build their own equipment because no one else made it. The factory becomes a platform for the entire production ecosystem. Lucas: Exactly. And that's the kind of thinking that could make this work where earlier cultivated meat companies struggled. Now, I want to shift gears slightly, because we've been talking about the factory floor, but there's a human element here too. Luna: You mean the jobs. What does a cultivated salmon factory employ? Biologists, sure, but also technicians, quality control, logistics. Lucas: Right. Their current pilot plant employs about 40 people, mostly with advanced degrees. But once they scale to a full commercial facility, they estimate needing around 200 staff, including operators with two-year technical degrees. That's a new category of industrial job — biomanufacturing technician — that didn't exist a decade ago. Luna: And the skill sets are transferable. If you can run a bioreactor for salmon cells, you can probably run one for chicken cells or even for pharmaceutical proteins. It's a flexible workforce. Lucas: Absolutely. There's a fascinating parallel here with the semiconductor industry. In the 1970s, Fairchild and Intel created a whole new class of 'fab worker' jobs. I think cellular agriculture could do the same, especially in regions that are looking to revive manufacturing. Luna: Speaking of the broader ecosystem, there's a role for policy here too. Japan's government has been funding research into alternative proteins as part of its 'Moonshot' program. That's a concrete advantage. Lucas: They have a goal to replace 30 percent of Japan's animal protein with alternative proteins by 2050. Cultivated seafood is a pillar of that. And they're putting real money behind it — about $100 million in grants over five years for cultivated meat and seafood research. Luna: That's not nothing. But compare that to the subsidies for conventional agriculture. Still, it's a start. And it signals that the government sees this as a strategic industry. Lucas: We're seeing similar signals elsewhere. Singapore has its '30 by 30' goal for domestic food production. The US Department of Agriculture has a pilot program for cellular agriculture research. But Japan's approach is particularly focused on industrial scale, not just lab development. Luna: I want to ask about the elephant in the room: consumer acceptance. Even if the product is perfect and the cost is competitive, will people actually eat it? There's a lot of skepticism about 'lab-grown' anything. Lucas: Surveys in Japan show about 40 percent of consumers are willing to try cultivated seafood, which is actually higher than in the US or Europe. Part of that is cultural — Japanese consumers are already comfortable with processed seafood products like kamaboko or imitation crab. They understand that food can be engineered. Luna: That's a good point. The framing matters too. Calling it 'cell-based' rather than 'lab-grown' makes a difference. And having it served in a high-end sushi restaurant first, with a story behind it — that could create a premium image, not a cheap imitation. Lucas: That's exactly IntegriCulture's strategy. They're launching first in restaurants, not supermarkets. Let chefs tell the story. Let early adopters create the buzz. Then, as costs come down, they'll move into retail under a branded product name. Luna: Okay, so let's look ahead. If everything goes right — regulatory approval this year, cost reductions on schedule, consumer acceptance — what does the industry look like in 2030? Lucas: I think we'll see cultivated salmon as a niche but growing segment, maybe 1 to 2 percent of the global salmon market by volume. That might not sound like much, but it's billions of dollars in revenue. And more importantly, it proves the model. Once salmon works, you can apply the same platform to other species — tuna, eel, yellowtail. Japan has a huge market for unagi, which is critically overfished. Luna: And the technology platform could extend beyond seafood. The same bioreactors and media formulations could be adapted for beef, pork, chicken. It's really a general-purpose manufacturing technology for animal protein. Lucas: Right. And that's the bigger story. We're not just talking about a salmon factory. We're talking about the early prototype of a new manufacturing sector. The first Model T wasn't a big industry either. But it showed what was possible. Luna: It's easy to get excited about the potential. But there are real risks. What if the cost curve doesn't bend fast enough? What if a recession hits and premium products get cut first? Lucas: Those are real risks. But the fundamental drivers — population growth, environmental pressure, protein demand — aren't going away. The question is whether cellular agriculture can scale fast enough to matter. Right now, I'd say the odds are better than even that we'll see at least one commercially successful cultivated seafood product within five years. Luna: That's a bold bet. But you know, listening to the specifics of IntegriCulture's approach — the serum-free media, the scaffold, the vertical integration — it does feel different from the hype we heard a few years ago. There's actual engineering here, not just press releases. Lucas: And that's the theme we come back to again and again on this show. Real manufacturing progress happens in the details — the cost per liter, the yield percentage, the regulatory timeline. Not in the vision statement. The Japanese salmon factory is a case study in grinding through those details. Luna: I think we've got a good picture of where things stand today. Let's come back in a year and see if the first fillets have hit the market. Lucas: I'd like that. And I'd like to see how the supply chain evolves around it — the feedstocks for media, the equipment manufacturers, the logistics for fresh cell-based products. It's a whole ecosystem waiting to be built. Luna: Before we wrap, one thing that strikes me: these kinds of shows — deep dives into specific factories, specific processes — they take a lot of time to research and produce. And they're ad-free, which is rare. Lucas: It's true. We put a lot into each episode, and we do it because we believe in the value of understanding how things are actually made. If these conversations have moved your work forward in some small way — maybe you're in manufacturing yourself, or you're investing, or you're just curious — then a couple of dollars a month is genuinely what keeps these going. Luna: We don't have a big marketing machine behind us. It's really just listener support that allows us to stay independent and dive deep into stories like this one. So if you've gotten something out of it, the best way to keep it going is a small contribution. Lucas: You can find us at buy me a coffee dot com slash fexingo. That's all lowercase. And really, even a few dollars a month makes a difference — it covers research time, hosting costs, the little things that add up. No pressure, but it's how we keep the lights on. Luna: And with that, let's get back to the factory floor. Because there's a lot more to explore in how this technology could reshape the entire seafood supply chain. Lucas: Absolutely. We talked about salmon today, but the same platform could be used for eel, which is a huge delicacy in Japan and faces severe overfishing. IntegriCulture is already working on that. And the same bioreactor design could be adapted for other species. It's a modular platform. Luna: So the salmon factory is really a proof of concept for a new kind of manufacturing — distributed, scalable, decoupled from natural ecosystems. That's a big idea with a lot of implications. Lucas: Exactly. And that's the kind of story we love to dig into. Thanks for listening, and we'll see you next time on The Manufacturing Podcast.