Latest News

Dairy Defined Podcast:

Milk Doesn’t Come From a Vat, Wisconsin’s Lucey Says

July 25, 2022

While “lab-based” dairy is touted by startups and investors, the very concept that dairy-protein replicas made in vats holds equal value for consumers is suspect, says Dr. John Lucey, professor of food science at the University of Wisconsin, in a Dairy Defined podcast released today.

“We have to separate out the fact there’s a lot of marketing hype,” Lucey said. “The reality is, to produce these original structures like milk fat or casein, micelles and stuff that are present naturally in milk, is really complicated.”

Lucey details the many differences between dairy, a natural animal product, and “animal-free” imitators in composition and sustainability, noting why labeling for such products need to be clearly differentiated. The full podcast is here. You can also find the podcast on Apple Podcasts, Spotify, Google Podcasts and Amazon Music. A transcript is also linked below. Broadcast outlets may use the MP3 file below. Please attribute information to NMPF.


Alan Bjerga, NMPF: Hello, and welcome to the Dairy Defined podcast. Lab-based dairy, it’s being touted as the future of food by startups and investors, but is it really that simple? How you regulate, how you ensure safety, even what you can say on a label, all of these questions need answers when new technologies enter a market. Joining us to at least better understand what we’re talking about this week is John Lucey, professor of food science at the University of Wisconsin. Lucey’s research interests include dairy chemistry and technology with work involving cheese, yogurt and dairy beverages. The focus of these interests is to try to understand the basic mechanisms that can then be used to explain applied problems or gain insights into how food functions. Dr. Lucey, welcome to Dairy Defined.

Dr. John Lucey, WI Center for Dairy Research: Well, thank you. I’m happy to be here.

Alan Bjerga, NMPF: And in some ways, our topic today really is defining dairy. What, from a scientific standpoint is a dairy product?

Dr. John Lucey, WI Center for Dairy Research: Well, I think dairy products come from milk. And so milk comes from cows. More generally, we technically would say from mammals. So cows are an example of that and it’s the milk from a mammal. And we use it then to produce a product. The product could be actually literally the milk itself, or it could be something as complicated as cheese or dried to make a powder. But any of these things, dairy products, the key part of it is it comes from a mammal. Now that could be cow’s milk, goat’s milk, deer milk. It could be from anything that produces milk, could be a dairy product.

Alan Bjerga, NMPF: What’s important about that animal connection?

Dr. John Lucey, WI Center for Dairy Research: Well, there’s very unique components in milk that’s produced by the mammal because it’s there for the nutrition of their young. So for example, when we were talking about humans, the mother is putting into this milk all of the critical components needed for the growth of the young. Because whether it’s a cow trying to feed a calf, or whether it’s a human trying to feed their infant, that first milk that they will be consuming has to have everything that is needed for the growth of the young. Everything. So there’s very unique components in there, they’re organized in various ways and nothing can be absent. So this is not just going out and having a bar or a snack or a cereal. This material is the only food that the young will have for a while. So there’s all kinds of interesting things in there, but it has to be a complete food.

Alan Bjerga, NMPF: So when we hear people talking about an animal-free dairy product, what are they really talking about?

Dr. John Lucey, WI Center for Dairy Research: Well, it’s not even clear that they are truly animal free, because, for example, one of the technologies takes a gene from the animal, tries to put it into a yeast or a bacteria and get it to produce it, a single protein. Even that single protein, is that really animal free? Because actually, the gene came from an animal. So I think there’s some questions about that kind of terminology. But is it really the same as milk if I take one protein and we know there are hundreds, perhaps thousands of components in milk, and I successfully make in a lab one component and put that into a solution, is that really milk? I mean, it doesn’t make a lot of sense to me. It is very limited in terms of what can be produced and put into a system for you in terms of the complexity of a natural system.

Alan Bjerga, NMPF: So part of the shorthand we’re hearing about these products is the term lab-based, gets bandied about. But when researching for this piece, came across another term, precision fermentation, that didn’t seem as common, but it seems important. Could you explain a little bit what precision fermentation is and how that relates to the term lab-based?

Dr. John Lucey, WI Center for Dairy Research: I think for a scientist like me, none of the technologies that are really out there at the moment are truly lab-based, I would put that comment out there first. So I think the vast majority of the technologies that are being employed by these startup companies fall under the bucket of what we call precision fermentation. That’s a new word, but it’s been around for a long time. In simple terms, a gene from a cow or a calf, to make a protein is taken out from the DNA of that animal and then they put that into another organism. They include that genetic material into another organism. That organism is usually a yeast or a bacteria. These are kind of industrial strains that are used to make lots of vitamins and all kinds of things. And then they get that organism, that bacteria or yeast to produce lots of that individual protein. They kind of trick it to think it’s part of their genetic material and they kind of force it to produce lots of it.

That’s usually done in big fermentors or reactors, big tanks. They grow these bugs to high numbers, and then they burst it up, release it, purify it. So it’s more like a fermentation type of process than anything else. And the word precision is really about, they put in a gene for a protein, for example, into one bug. So it’s targeted, but remember, this is one protein. So what you’re doing is, every time you want to make another component, you have to go and do the same whole process again.

So back to your question about lab-based, none of these really fit my kind of definition of lab based, they’re all fermentation-based, using genetically modified organisms to do this. There are some technologies out there that could be lab-based and they would be actually taking not the genetic material from an animal or from a cow or calf, these would actually be taking the actual lactating cells themselves and putting them in a lab and actually growing the cells and actually having those cells, which are little mini milk factories, produce everything in the lab for you. It’s kind of like you have the cow right there in front of you in the lab to produce your stuff. That’s been explored right now and that may be something that might happen in the future, but there is no commercial products out there on that lab-based, using the actual mammary cells themselves.

Alan Bjerga, NMPF: So seeing what’s out there now, do you see a technology that can make a 100% identical dairy protein without a cow?

Dr. John Lucey, WI Center for Dairy Research: It’s not 100% clear that even if you take the genetic material from a calf and put it into an organism that that organism, you can trick it to make identical to what the calf will make. There’s usually some small differences because it has to be put into their genetic material encoded and expressed, et cetera. There’s a lot of steps that go on there. So generally what that means is that yeast or bacteria will make some changes themselves. How important those changes are? We don’t know, but there’s usually some changes. This technology is not new, although there’s a lot of hype and stuff going on right now.

Back in the early 1990s, the dairy industry was the first to pioneer this technology of precision fermentation. And it did it for making calf rennet. We were running out of calf rennet globally, not just here in the US, and there wasn’t enough calves to provide enough of this calf derived rennet to make cheese. So the industry looked to put in the gene from the calf into these organisms, these genetically modified organisms, got it to produce lots of rennet in a vat or a tank or fermentor, and purified that. And those had small differences from the calf, but not enough to be a concern. They didn’t affect the functionality to make cheese, but back to your question, they weren’t 100% identical. How important are these small changes? We don’t know, because we don’t know enough about these changes from these new ingredients that the startups and others are coming up at.

I’ll give you one example. Some people have heard about A2 milk and A1 milk, and then there’s some products on the market at the moment, and I’m not going to comment about whether or not they are really different or better. However, A2 and A1 milk have only one amino acid difference between them. One, out of about 200 amino acid. So one, that’s less than 1%. That’s about half a percent change in the protein sequence. So that’s a small change and yet there’s a lot claim for the A2 milk. So I would say we don’t know enough about if there’s changes, how much changes and what they do, because it’s just too new. We haven’t studied it.

Alan Bjerga, NMPF: One of the unique properties of dairy compared to say, other commodities is that it can be turned into different things and it has different components. One of those components are milk fats. Is a cow-free milk fat even possible?

Dr. John Lucey, WI Center for Dairy Research: If you take milk fat as produced by, for example, the cow, there are 400 approximately fatty acids that are found in milk fat from the cow. 400 different types of fatty acids. Those are assembled onto what we call a triglyceride. Without getting too complicated, those fats are added like 1, 2, 3 on a backbone. The order of which they’re put on, whether the first position, second or third is not random. Some types of fatty acids go on the first, some on the second, some on the third. And that affects things like melting point, crystallization, digestion. We study this in class. So this is a very complicated type of system to create. In the lab, what they are talking about right now is making one or two fatty acids and just sending that into an emulsion, a simple [inaudible] mixture, and giving that in a beverage. That isn’t a fraction of what we’re talking about in terms of what we call biological components or bio assemble components.

The milk fat also has a membrane around it because it’s produced in the cow, in the mammary cells. And then when it comes out of that cell, it gets another membrane. So the actual membrane have a lot of interesting complexity as well with rich and phospholipids and proteins. The degree of complexity in something like milk fat is staggering and very difficult to reproduce. I always think everything is possible, but trying to reproduce that in the lab, to me, there would be a lot of Nobel prizes given up for all this kind of stuff, if somebody can figure that out. And nobody has yet.

Alan Bjerga, NMPF: Does safety come into play in any of these considerations, concerns that people may have?

Dr. John Lucey, WI Center for Dairy Research: We don’t know enough yet would be one answer. However, I’ll give you a couple of comments. One is that when the first food ingredient, which was this rennet, was developed and approved by FDA, the FDA took a very rigorous multi-year approach to reviewing it. In fact, they made it a case study because it was the first food ingredient approved by using this precision fermentation technology. And to get it approved, multiple, what I would call kind of kill steps, had to be put into the process to make sure none of the material from the yeast or bacteria ever made it to the final product. I haven’t seen all the process flows from these new startups and these new companies. The ones that they have shared publicly do not include all of these kill steps. That’s what I can tell you. So that makes me nervous.

That’s one thing. The second is that the rennet example that we talk about as kind of the first of these. Rennet is a tiny, tiny fraction of what we’ll find in cheese. It’s a tiny, tiny fraction of 1% of what you would eat. But if we replace all the protein in a cheese and using this precision fermentation type of material, that would be about 20% to 25% of the mass of the cheese. It’s a lot. You would be consuming not a trivial, trace amount. So I think we have to be a little bit careful because when we’re talking about proteins, many of these proteins, we’re talking about milk proteins, they can cause allergies. We don’t know enough about whether these new types of proteins would cause more allergies, different types of allergies. We just don’t know. We just haven’t studied them. And until there’s been a period of detail study, both for how they are and how identical they are and then also whether there’s any issues. I think it’d be very risky to start thinking about widespread use of them.

Alan Bjerga, NMPF: That said, these products would not be introduced to the market if there wasn’t some sort of consumer appeal to them.

Dr. John Lucey, WI Center for Dairy Research: Right now, what we have is a lot of marketing hype from these startups. We have zero published studies that indicate that there’s any sustainability improvement by them. They could make a case for this to see if it would be more cost effective or more sustainable. They could do that. And the way you do that is you do a study. Right now, there’s no study. But I would say I’m very dubious on some of these claims. Let’s put it this way. Why? Because our experience of using bioreactors and biofermentors are all over our industry, both for pharmaceutical use, for making amino acids and nutrients. It’s not that this technology is not out there. So we do know some of the cost and inputs in there. It is extremely expensive to build a modern bioreactor facility, extremely expensive. Because you have to grow them in steps from small batches, to medium batches, to large batches, and to control this process within very tight limits. That’s going to cost a lot of input costs and capital costs and operational costs in here.

And then you have to purify this system. And what I understand for some of these startups that are related to dairy proteins is the yields are very low. No more than probably what we’re getting out of cheese whey, for example, for some of their type of systems. So they’re going to be faced with the same concentration and spray drying and other technologies that are needed by the dairy industry. Right now, it’s not clear what the benefit from an economic point of view is for switching to these kind of processes.

Let’s put it another way. We feed the cow by having our walk around in the morning, eating grass in the field and by afternoon, she’s producing gallons of the stuff already produced, assembled, biofermented and we can milk her. I actually sat down and calculated how much it would cost to put in bioreactors to replace all of the milk produced here in Wisconsin. It would be greater than the national debt of some countries. I think we need to have a little reality check that the cost to produce this kind of modern biofermentor technologies is extremely high. So I think that issue is out there.

Right now, most of them are venture capital funded companies. They’re not like normal companies where they have to make a profit or they go out of business, they’re surviving on venture capital. So I think what you’ll see is those early wave of venture capital supported companies will divert away from trying to target milk. Because my bottled water is more expensive than milk. It’s not a high value added material. Instead they’ll divert their technologies to higher value ingredients because then they can make a return for it.

Back to the sustainability angle, as I said, there’s no studies right now. And the ones that I have seen, even from websites, the few that they have published on their websites or claimed on their websites, milk production, like skim milk, has a much lower carbon footprint than some of these ingredients that I’ve seen. Much lower. So I do not understand where all these claims are and it’s not backed up by any studies. But that doesn’t make them stop having all these marketing claims, for sure.

Alan Bjerga, NMPF: Let’s go back and talk about some of those higher value ingredients that you’re talking about because indeed when you’re talking about coming up with a facsimile of say, a single dairy protein, you’re really talking about something that would be more likely to be channeled into ingredients than a fully formed dairy product. We’ve talked about some very specific terms that get batted around, precision fermentation, et cetera. Here’s another one, bioactive compounds. A lot of your research is about looking into the unknown. What don’t we understand about how dairy works? And how does that play into our consideration of acceptance of a fermented product as a dairy product?

Dr. John Lucey, WI Center for Dairy Research: We’re having a kind of a paradigm shift at the moment in terms of, let’s say, nutrition and food ingredients. And it’s particularly very important for the dairy sector as well. What I mean by that is that historically, if you look at nutritionist, they talk about protein, fat, sugar, all that kind of stuff. They just talk about him in very bulk things. And if you had them in a food, you can consume X, Y, and Z, and you would get that amount of fat, carbohydrates and sugar, et cetera, or minerals or whatever. But that’s not what we’re consuming. We’re consuming real food products that are assembled into a matrix in a very complicated fashion.

So right now in the food science side of things, there’s huge interest in understanding how foods are assembled or separately, how to create different personalized foods that have different types of delivery. And then how are they released when they come into our body? All of those things are very important for affecting the question you are making about bioactivity. I’ll give you an example. If we were consuming a cheese, we are consuming a whole bunch of different components, protein, fats, minerals, et cetera, bacteria. But they’re not all individually packaged, they’re all part of a system. And then how slowly they’re released and how they’re released could affect how much we can absorb from them and what effects they have in our body.

For example, we’ve spent a lot of time worrying about and interested in bioactive peptides that come from proteins. So what that means is instead of just thinking about protein as something to build muscle or something like that, that as you break down the protein in our bodies, we are releasing small chunks of this protein that might reduce your blood pressure, that might help with hypertension, that might help your immune function, that might help you bind minerals, that might be antimicrobial and affect your gut metabolism. So we’re learning that, as we think about food and just the historical way of thinking about it, just as a collection of major components, macronutrients, isn’t really what we’re doing when we’re consuming a food product. And that the complexity and how it interacts with us is very unknown, to be honest. And I think that’s where I think about, like a simple giving one or two proteins in a solution and swallow it and thinking that’s the same as milk. It falls down very quickly in terms of how our body is thinking about it.

And another completely unknown area is then how this affects our gut microbiome, because we are learning more and more every day that our gut microbiome, which is kind of our internal defenders of our system against pathogens and other kind of conditions, they are very much affected by the type of food that we consume. Not just because it’s fat, proteins and sugars, et cetera. It’s because of how they respond to the kind of food we’re putting in there. So I think there’s a lot we don’t know, but it really pushes me to think about real foods. Real assemble foods and complicated foods are quite different than taking a couple of individual components in a tablet or in a simple drink.

Alan Bjerga, NMPF: Well, and bringing this back to where we began this conversation about the actual definition of dairy there’s the ever popular, but also incredibly important topic of labeling. Is there a scenario where such products, even if they are a, quote unquote, “carbon copy” of a single dairy protein, for example, does it make sense in the marketplace, in the name of consumer utility and transparency to use dairy terms when labeling these products?

Dr. John Lucey, WI Center for Dairy Research: I think it’s very confusing for consumers. So let’s say I have a cow’s milk protein allergy, and I say, I can’t have cow products. And somebody labels this and says, it’s cow-free. I might mistakenly consume that product, but the protein, as you say, could be almost identical to the one I’m supposed to avoid. And I could have an allergic reaction to them. So I think it’s very confusing for the consumers to understand what terms like cow-free means or dairy-free really means. I think if they’re expecting to now no longer have an allergic reaction to it, they would be mistaken. They would have the same reaction to it. But that might lead them to make mistakes.

So I think the consumers need to know what it is they’re consuming and is this different? I would say that beyond having the surprises of people consuming it and thinking they won’t have allergies to it now because it’s cow-free, the other thing is that are they really, truly getting their nutritional package from this? They’re thinking this is identical to milk, and this is identical to a dairy product. Where simplifying it down to a couple of components and putting it in some sugar and putting it a white solution and saying, that’s milk, the problem there is it’s not nutritionally equivalent, it’s inferior in all kinds of ways to the original product and the impact on people’s health and wellness is unknown, but it’s probably detrimental. If you say it’s kind of equivalent, but it actually nutritionally is not equivalent. That’s the problem I have.

Alan Bjerga, NMPF: We’re speaking with Dr. John Lucey, professor of food science at the University of Wisconsin. Dr. Lucey, anything else that we need to be covering while we have you with us today?

Dr. John Lucey, WI Center for Dairy Research: Well, I think these new technologies, we can use and we have used. And I think there are very important things that we can use, for example, we have already used it to make a minor ingredient like rennet to make cheese. We could use it to make components that are at trace levels or very low levels in milk. And then add it back into infant form or nutritional products and potentially have some benefit on those populations. So I think there are examples of where we should be using these technologies.

I honestly think there’s a lot of hype about replacing the milk from the cow and so on and making truly identical dairy products, that’s not easy to do to. To make cheese, for example, the way I would expect to make a cheese, which is take milk, add some rennet, coagulate it, and so on, to actually perform that function with an isolated, one individual casein is not possible. So there has to be some solution to that to make them into the micelles and the structures that are present in milk. And I’ve seen no evidence that they’re able to do that.

So I think we have to separate out the fact there’s a lot of marketing hype about doing it because they’re trying to attract venture capital money. But the reality is, to produce these original structures like milk fat or casein, micelles and stuff that are present naturally in milk, is really complicated. I spent my whole career trying to understand them, and it’s really complicated. But not coming soon to any place near us for something like that.

Alan Bjerga, NMPF: Dr. Lucey, thank you for your time.

Dr. John Lucey, WI Center for Dairy Research: Thank you.