Multilayer analysis reveals extensive hidden population complexity in a dairy starter microbiome comprising few species

In the world of food science, some of the most important work happens in the invisible realm of bacteria. When making traditional cheeses like Gouda or Cheddar, producers rely on starter cultures: mixtures of microscopic organisms that turn milk into solid cheese and create its distinctive flavor. For over a century, these cultures have been kept alive and passed from batch to batch, a practice known as using an "undefined" starter because the exact mix of bacteria inside is not fully known or controlled. Scientists have long suspected that these mixtures are more complex than they appear, hiding a vast array of different bacterial strains that work together to keep the cheese safe and tasty. Understanding this hidden diversity is crucial because it determines how well the cheese resists attack from viruses that infect bacteria, known as phages, which can ruin a whole batch of production.

A team of researchers set out to map this hidden world inside a specific Danish starter culture called Tistrup, which has been in use for more than a hundred years. They knew that simply growing bacteria in a lab dish to count them was not enough, as many strains might be missed or overlooked. Instead, they combined two powerful methods: growing a large collection of individual bacterial strains in the lab and reading the genetic code of the entire mixture directly from the culture without growing anything first. By comparing what they found in the lab with what they found in the raw mixture, they discovered that the community was far more intricate than either method could reveal on its own. The study shows that even in a group of bacteria that looks simple on the surface, there is a deep layer of genetic variety that allows the community to survive and thrive.

The researchers began by looking at the Tistrup culture, which is dominated by just a few species of bacteria, primarily Lactococcus cremoris and Lactococcus lactis. Because there are so few species, they expected the genetic differences between individual bacteria to be small. However, when they sequenced the DNA of 111 distinct bacterial strains they grew in the lab and compared them to the genetic data from the whole mixture, they found a surprising amount of hidden complexity. The two methods did not just agree on the same bacteria; they found different parts of the population. The lab-grown strains represented only a fraction of the diversity present in the culture, while the direct genetic reading of the mixture revealed many lineages that the lab methods completely missed.

One of the most striking findings concerned the cell walls of these bacteria. The outer surface of a bacterium is covered in a sugary coating called cell wall polysaccharide, which acts like a specific keyhole that viruses use to enter and infect the cell. The researchers found that the most common type of this sugar coating in the culture, making up nearly thirty percent of the population, was completely absent from the bacteria they grew in the lab. This means that the most abundant bacteria in the culture were the ones that the scientists failed to grow, leaving a major part of the community invisible to traditional methods. The lab-grown bacteria mostly carried different, less common versions of this sugar coating, suggesting that the conditions used to grow them in the lab favored a specific, less dominant group.

Beyond the cell walls, the researchers looked at other genetic features that help bacteria defend themselves against viruses. They examined the presence of viruses that hide inside the bacterial DNA, known as prophages, and the various defense systems the bacteria use to fight off attacks. They found that these defense tools were mixed and matched in countless combinations among the different strains. Even bacteria that looked almost identical in their core genetic code carried different sets of defense systems and hidden viruses. This variety acts as a shield for the community; if a virus attacks one type of bacteria, others with different defenses can survive and keep the cheese-making process going. The study showed that this defensive diversity was much greater than what could be seen by looking at just the lab-grown strains.

The team also investigated the mobile genetic elements, which are small loops of DNA called plasmids that bacteria can swap with one another. These plasmids often carry genes that are essential for making cheese, such as those needed to digest milk sugar or break down proteins. The researchers found that these plasmids were also highly diverse and were carried by different strains in different combinations. By using a special technique that reads chemical marks on the DNA, they were able to link many of these plasmids to the specific bacteria that carried them, even when those bacteria were not grown in the lab. This allowed them to see that the same plasmid could be found in different species, moving between them and spreading important traits throughout the community.

The study concludes that the complexity of this dairy starter is not just about having many different species, but about the vast number of different genetic combinations existing within just a few species. The researchers found that no single method, whether growing bacteria in a lab or reading the DNA of the whole mixture, could capture the full picture. The lab methods missed the most abundant strains, while the direct DNA reading could not always tell which specific strain carried which gene. Only by combining both approaches could they begin to see the true extent of the diversity. This hidden variety, involving cell walls, defense systems, and mobile DNA, provides the community with a robust ability to resist viral attacks and maintain the quality of the cheese. The findings suggest that the resilience of these traditional starters comes from a deep, combinatorial diversity that remains largely hidden unless scientists look at the community through multiple lenses.