Cultural Revolution

Words by Andrew Janjigian / Images by Adam Detour

In a sleek new laboratory off Oxford Street in Cambridge, a small group of Harvard researchers are leading a quiet—if sometimes odiferous—revolution in the world of cheesemaking. There, Rachel Dutton, along with postdocs Ben Wolfe and Julie Button, are using modern technology—along with some old-fashioned scientific fieldwork—to answer fundamental questions about one of our most traditional foods.

It’s true that a cheese is a living organism, one that develops over time and has a limited life span, but it’s more accurately thought of as a community, or—as Dutton likes to put it—an "environment." Cheese is made of milk, but the function of the milk is to provide food and housing to the microbes that give each cheese its true identity. In aged cheeses, it's the microbes that matter most.

Over the past few years, Dutton and her colleagues have traveled throughout the US and Europe collecting samples from the rinds of more than 160 different cheeses (including many from Jasper Hill Farm in Vermont, an early and ongoing supporter of their work). Once back in the lab, they transferred each to a sterile mixture of cheese curds and agar (a medium Dutton dubbed "in vitro cheese"), in order to isolate and identify all of the individual organisms present, and to compare the community of microbes on each cheese to one another.

The technology that accurately and rapidly characterizes microorganisms in this way has been available for some time, but until now no one had thought to apply it to cheese (or any other "live" foods, for that matter). Before Dutton's team came along, these kinds of questions were asked only in the context of food safety. Scientists were mainly interested in keeping harmful microorganisms out of our food, not looking at those that are meant to be there.

Dutton herself didn't start out with the goal of studying cheese either. A microbiologist by training, she was interested in looking at the interactions between microbes in the environment. But, because of the sheer numbers of organisms involved, studying real-world microbial relationships is a daunting task. Scientists instead tend to look for simplified systems that can serve as models for what happens in more complex communities.

One day, while browsing the pages of Harold McGee's On Food and Cooking, Dutton realized that cheese might make an ideal model system, because each cheese is a real-world microbial island unto itself, complex, yet circumscribed. (That it tended also to be delicious didn't hurt either.)


To get a better sense of what she and her colleagues are up to, it helps to understand how milk is transformed into cheese. There are two general classes of cheese microbes: starter bacteria and ripening microbes. Starter bacteria are added at the very beginning of the cheesemaking process; they convert sugars in the milk into lactic acid, which—along with the enzymes supplied by rennet—helps to create curds, which are then formed into cheese. For cheeses that are consumed "young", such as mozzarella and feta, that's more or less the end of the story. The flavor and texture of these cheeses derives mainly from the milk itself and the way it has been processed.

But for aged cheeses, the tale—thankfully—is far more complex. The powdery blue veins in a Roquefort, the pungent, unmistakeable bouquet of a Limburger, the characteristic holes in a Swiss, the snowy white rind and oozy consistency of a ripe Brie all come thanks to the action of ripening microbes on the cheese during the aging process. Nearly all of the diversity in the cheese world comes from the myriad ways that these sorts of bacteria, yeasts, and fungi behave.

While producers do deliberately introduce ripening microbes during the cheesemaking process, either during the raw curd stage or on the surface of the already formed cheese, they normally only add no more than a few key cultures to each, such as the Penicillium responsible for the blue of a Roquefort, or the Brevibacterium that gives Limburger its dirty-sock aroma (the very same Brevi that's hiding between your toes right now, by the way).

But—according to the data that Dutton and her colleagues have generated—the typical aged cheese is home to anywhere from 5 to 20 unique microbes. If they are not put there deliberately by the cheesemaker, where do they come from? And—more importantly—what are they up to? These are exactly the sort of questions that the Dutton lab is attempting to elucidate.

With a comprehensive bank of cheese microorganisms on hand, the work of piecing apart the interactions among each is now possible. It's time-consuming, given the numbers involved, but not complicated: you just mix and match the organisms in all possible combinations on an agar plate, and then observe what happens over time. (The bulk of the interactions work falls on the shoulders of Julie Button.) This fungus and that bacterium together produce a certain rosy-hued pigment, these two bacteria with the unmistakeable aroma of Kraft Macaroni & Cheese. Each data point can then be used to explain the appearance of these same phenomena out in the "real world" on the cheeses themselves.

One of first key discoveries the lab made was the fact that cheeses of the same style, no matter the origin, were remarkably similar to one another in terms of their microbiology. Not just similar, but nearly identical in many cases, with the same sets of species on each. "We thought maybe we would find completely different things in French cheeses than we did in the US cheeses," Dutton told me. "Instead, what we are finding is that the way you make a cheese creates a specific environment, and then you get the microbes that are associated with that type of environment."

These results upend the notion of terroir, the belief that the essential character of certain foods derives from their place of origin. The term was first used by the French to explain why grapes grown in a particular climate and soil type produce wines that taste a certain way, while the same vines transplanted to another region can yield a very different product.

The idea of terroir has long applied to cheesemaking as well. The notion that the character of a cheese is tied directly to the unique microbial makeup of the cave in which it is ripened is an old one. You might make a similar cheese elsewhere, but—lacking the precise mixture of microbes found only in that one cave—it'll never be quite the same. Dutton's results suggest otherwise.

The microbes found in cheese appear to be ubiquitous, rather than local. What's important—at least as regards cheeses that derive much of their flavor from rind microbes—is not so much where you make the cheese, but rather how you make it. If you look closely you'll find the same set of organisms on a blue cheese from England (Stichelton, for example) as on a blue from Vermont (like Jasper Hill's Bailey Hazen Blue). As postdoc Ben Wolfe is fond of saying, "If you build it, they will come."

Of course, the composition and quality of the milk does have a strong influence on the cheese that it's made into, especially on larger-sized cheeses, where rind microbes have much less of an impact on the final flavor of the curd. Cheese microbes have the greatest impact on smaller washed-rind or bloomy-rind cheeses.


Dutton and her colleagues admit that cheese terroir might still exist at the level of the individual strain rather than that of species or genus, as had been previously assumed.

But if so, that fact might present new commercial opportunities for United States cheesemakers. Almost to a one—largely for historical reasons—American cheeses are made using cultures isolated from and produced in Europe. If Dutton and her team identify strains of bacteria or fungi that are unique to North America, these could potentially serve as stock for locally produced cultures, freeing the American cheesemaking industry from its present reliance on European ones.

Their work has more immediate practical implications for cheesemakers as well: the results of their ongoing interactions studies will someday serve as a database of information that can be used by cheesemakers to troubleshoot cheese cultures gone awry.

"We're not anywhere near it yet because we are obviously still doing the very basic science that needs to get done," explains Wolfe. "But from all these little tidbits that we're seeing, we're starting to think of things that could help cheesemakers figure things out that they are encountering. That's pretty exciting."

"For example, we had a case of a rind that just didn't grow in the United Kingdom. There was a bloomy-rind cheese and the bloominess just stopped growing after a while. It turned out that the milk had had a lot of Pseudomonas in it, and the Pseudomonas was probably producing some anti-fungal compound that was killing off the Geotrichum that existed on the outside of the cheese. [Cheesemakers] know there's something wrong, but until you start growing [these organisms] in the lab and pulling them out and pairing things together—which is what we do all the time now—you don't know exactly what's wrong or how to really fix it out there."

While most of the cheesemakers they've approached welcome the light that such studies shed upon the cheesemaking process, there remain some who are resistant to the notion that the their work can or should be reduced to a science.

"Most of the people who come hear us talk are the kind of people that want to know more and don't really care about ruining the magic," explains Wolfe. "But there is a cohort of people out there that have been making these wonderful, delicious cheese products for however many years, and [the process] works fine. They don't need to know exactly what's there."

Others, he suggests, are less worried about having the magic drained from their work as they are having their personal recipes and methods made public.


While much of their work outside of the lab involves educating cheesemakers about the various organisms that go into making their cheeses, Dutton and her colleagues consider outreach to the general cheese-buying public as of equal importance. The cheesemakers they work with feel that as their own understanding of the biology of cheese grows more sophisticated, so should that of their customers.

"We all promise cheesemakers that we'll educate lay people about what's good and what's bad," says Wolfe. In the United States, the appearance of patches of yellow or black molds on our cheese would likely condemn it to the dustbin, while in Europe consumers know better: they might also be a sign of deliciousness within. "Cheesemongers and the people who buy cheese need to be better educated about the sorts of things that people in France have known forever."

It's obviously a long way off, but one result of Dutton's work could be that by entering a set of endpoints into a computer, you could engineer a cheese—one microbe at a time—to whatever specifications a cheesemaker might desire or dream up.

"The system is complex, but still simple enough that I think that we could get there," predicts Wolfe. "We have to figure out a lot of things, like what is the salt range that every microbe wants to grow in, and what happens if you add every combination of every microbe with each other. I feel like if we measured all of those things—which is not that hard to do—we might actually be able to come up with a predictive model that could say this is exactly what your cheese rind would be like if you do this. It would be kind of crazy. And really cool."

Perhaps the coolest thing—for me at least—is that these smart people get to work with cheese all day. While eating is forbidden in the lab, sniffing is encouraged, and there's lots of things to smell, some more appetizing than others. On the day I visited, the dominant odor was that of rotting lettuce. At other times, the smell of cauliflower, mushrooms, or barnyard animals might waft from their lab bay. Fortunately, most of their neighbors are more curious than offended by these aromas. "One of the labs next door to ours works on human feces, so how bad could our cheese smells be in comparison?" says postdoc Julie Button. "And we often have samples available in the lunch room for tasting, which is great for public relations and probably leads to greater tolerance of our smelly lab bay!”


Andrew Janjigian is an Associate Editor at Cook's Illustrated Magazine. When he's not dismantling recipes for hire, he's more often than not baking bread. He tweets regularly as @wordloaf and blogs much less regularly at