Brennan et al., Appl. Environ. Microbiol. 2002, 68(2): 820-830. What organisms grow on the surface of a semi-soft, pasteurized washed rind cheese during the course of its development? How does inoculation with a commercial strain of Brevibacterium linens affect the development of the rind microbial community? This paper sheds some light on these very pertinent questions.
Cheeses from a single batch of Gubbeen cheese were split into two sub-lots; one had an initial treatment at 3 days with a solution of salt and commercial B linens culture (called ‘BL2’), while the other cheeses only received a wash with salt solution. In all other respects the cheeses were treated identically, and were washed on a regular basis with a saline solution to promote normal rind development. Sampling was done at four points during maturation (at 4, 16, 23, and 37 days of age). For each sample, a thin layer of rind was removed and the total number of bacteria and yeasts present were counted. Fifty colonies were also isolated and tests were run to identify them based on their appearance, metabolic activity, and their acid and salt sensitivity. A set of molecular tests were also run on each cultured bacterial colony, which allowed more minute differences between strains to be detected and their relationships mapped.[i] The interactions between several of the strains isolated from the cheese rinds and two commercial strains of B linens, BL1 and BL2, were examined.
Summary of Findings
The team found that there was no significant difference in the total number of yeasts or bacteria on the inoculated and non-inoculated cheeses at any point during their maturation.[ii],[iii] The bacteria on the surface of the cheeses were identified as primarily coryneforms (including the genera Arthrobacter, Brevibacterium, Corynebacterium, and Microbacterium) along with a few coagulase-negative (non-toxin-producing) staphylococci. Several newly-identified species made up the majority of the rind population, including two species of Corynebacterium and the aptly-named Microbacterium gubbeenense. None of the strains isolated from the cheese rinds was identical to the inoculated B linens culture strain. The team found that all of the strains of Staphylococcus and several of the strains of Corynebacterium inhibited the growth of the strain of B linens used in the smear culture for the inoculated cheese, which might explain why the inoculated culture was not able to grow successfully on the rind of the cheese.
This paper suggests something quite interesting: that the natural organisms that make up the ‘microbiome’ of cheese dairies may not only provide a rich and unique source of ripening bacteria for the cheeses matured there, but that these organisms may in some cases outcompete commercial strains that are applied directly to the cheeses. Anything that comes into contact with the cheese (including brine, racks, and shelves) could a source of these microbes. Coagulase-negative staphylococci and coryneforms are also common members of the human skin microbiome, so it is possible that the people doing the maturing may also be lending some of their own microbes to the surface community of the cheeses. The amount of each species found in the rind community is not necessarily a reflection of the amount applied to the cheese by hands or equipment. Rather, the rind community’s composition reflects the organisms’ interactions with each other and relative fitness to succeed in that specific environment (including factors such as availability of nutrients, temperature, salt, acidity, and moisture level). A small change in any of these factors can tip the balance toward a different set of microorganisms and substantially change the rind community makeup. The authors of the paper felt that a preparation of the species unique to Gubbeen cheese might be a more useful inoculant than BL2. (Since then, the scientists have developed a proprietary mix of these ripening strains for use on Gubbeen cheese.) One might question whether this inoculation is really necessary, given that the organisms were thriving there already without inoculation. Furthermore, the question of ‘evolution’ is also worth exploring: how have the environment and conditions at Gubbeen’s dairy changed over the last decade? Might we find that the M gubbeenense isolated in 2002, currently used for inoculation, is now completely inhibited or outcompeted by another native strain? How much money may be wasted on commercial ripening cultures when the important organisms are already present in the environment of the dairy? Have cheesemakers who have done trials comparing surface-ripened cheeses with and without inoculation with ripening cultures achieved similar results?
The full text of this paper is available here.
[i] The scientists limited their analysis to colonies that they were able to isolate through culture-based methods in the laboratory. Use of newer, so-called ‘high-throughput techniques’, which have become more common since 2002, might allow a more complete identification of the inhabitants of the cheese rind community and all of its members.
[ii] During the early stages of ripening (up through day 10), yeasts were the most numerous inhabitants of the rinds. At the end of the maturation, bacteria outnumbered yeasts by a factor of 100. This is a classic example of succession, where early colonisers (in this case yeasts) deacidify a surface, making it more hospitable for later inhabitants (bacteria).
[iii] The surface of the mature cheese was between pH 5.7-6 and 2-2.5% salt. The moisture level at the surface of the cheese was about 25% (compared to around 45% within the cheese).
Post written by Bronwen Percival