Deliverables
1. The surface microflora of 3 commercial smear cheeses was analysed over the whole ripening time of 15 weeks. The flora mainly consisted of the yeast D. hansenii and the bacteria S. equorum, B. linens, C. casei and M. gubbeenense. This demonstrated the the basic concept of a 5 species surface starter developed in a previous EU project using Tilsit cheese was also applicable to other semi-hard or semi-soft smear cheeses, like Klovborg, Cave cheese or Leerdammer Caractere.
2. 36 experimental cheese batches were ripened on laboratory scale with various combinations of smear strains and species tested. The selection of suitable B. linens strains for typical aroma development was most important. The selection of C. casei and M. gubbeenense strains had a lower impact on aroma development. No influence could be attributed to S. equorum and D. hansenii.
3.The requirements for starters for Klovborg, Cave cheese and Caractere were shown to consist of the 5 species mentioned above. The addition of further species (e.g. C. flavescens, C. variabile) did not change the aroma in any detectable way.
4. Mechanisms of colour development are still not fully understood. It is clear that the orange pigments of B. linens and S. equorum can contribute to the surface colour but only when high surface cell counts of the particular species occurred. This was rarely observed. A more likely mechanism was the interaction of yellow M. gubbeenense pigments with proteolysis products of B. linens along with a high surface pH which transform the yellow colour to red.
However further mechansms must exist. In cheeses with low B. linens and S. equorum numbers, AND no M. gubbeenense applied, red colours were still developed to a lesser extent.
5. The key factors of smear development were determined. First, the smear cheese surface must be ready to absorb the smear liquid. All microorganisms that are supposed to grow on cheese should be present at the beginning, either inoculated via the cheese brines or via the smear liquid. To use smear bacteria, staphylococci and yeasts in the brines and the smear liquid means that the probability of other bacteria (other smear strains or contaminants) growing is minimized, because all ecological niches are used by the starter microorganisms.
6. The functionality of the defined starter was proved by the identification of the strains that had grown on the mature cheeses by using pulsed field gel electrophoresis. More than 500 strains were analysed. It was found that the starter strains formed the predominant part of the surface microflora. When a natural brine microflora was present in trials, the amount of "house" strains was higher in the population.
7. The original concept of the 5 species surface starter, developed for Tilsit-like cheeses could also be applied for the cheeses studied in the project. All had a higher dry mass than the Tilsit-type cheeses.
8. During screening of suitable strains their functioning should be proven by identifying the strain on cheeses. It should be noted though, that the surface flora in general is quite variable, and that by many methods strains can not be detected if present at 1% of total cell counts or lower. Thus, successfully ripened cheese batches should be analysed for growing strains repeatedly and at different stages of ripening to determine which strains are growing and are probably responsible for aroma development.
9. Help yourself: smear cheeses might be a too small market for starter suppliers to provide a full set of adapted smear bacteria. Right now just B. linens, S. equorum and D. hansenii are available. Cheese companies themselves could isolate strains from cheese brines, identify species and strains, and store a sufficient collection for use in the cheese brines. Based on 2 different studies between 10 and 50 smear strains mimic a natural cheese brine microflora. The necessary methods are routine today and the total cost should be between 5,000 and 10,000 Euro. With this set of strains the flora of brines can be kept at a suitable levels if necessary (10-100 cfu/ml). A single 100ml culture growing up to 1e9 cfu/ml over night is sufficient to inoculate 1e6 litres of cheese brines.
10. Addition of the surface starter bacteria not only to the smear but also to cheese brines means full control of the surface flora. It has to be determined by experimental cheese trials (see another part of the etip) and sensory analysis whether this approach provides the typical properties of the cheeses. It is possible that a low level of the natural house microflora is essential for typical ripening. If this influence is clear the brine microflora could be controlled not by adding starters but by propagating strains from the brines to maintain a desirable microbiological status of the cheese brines (see 9.).
For producing high quality smear cheeses, it is essential that bacteria show good smear and colour development, lead to good cheeses with sensoric properties and inhibit mould growth. The yeast should have good deacidification activity. An iterative routine for developing improved starters has been build. This routine shows the essential steps for delivering (1) a better-controlled process for the preparation of red-smear cheeses and (2) an approach useful for development of both new starters and new products. The steps in the process comprise: a. Create a target cheese sensory and analytical profile. b. Isolate microorganisms from target cheese and supply if necessary with strains with additional interesting properties. c. Test technological and microbiological properties of these strains. These properties of the strains are: Enzyme activities relevant for (volatile) flavour compounds, proteolytic activities, mould and undesired bacteria inhibititing activities, actual flavour formation by strains, growth properties. d. Choice of starter composition: dependent on product and technological requirements. The smear should consist of strains with flavour formation activities, proteolytic activities, colour forming activities and mould inhibition activities. The manifestation of these properties is determined by the cheese (production) environment. The best way to investigate the ability to develop a desired smear is to test in model cheese experiments and/or in pilot scale experiments. Start with the yeast; good deacidification of the cheese surface is a must for further development of the other microorganisms. Then bacteria of genera Staphylococcus, Arthrobacter, Brevibacterium and Corynebacterium preferably should be used. e. Actual testing on cheeses (model and experimental/pilot scale). It is important to stay close to production conditions for surface ripening. f. Analysis of experimental cheeses and compare with target cheeses profile (sensory and analytical, as determined under a.). g. If experimental cheeses are not acceptable, then change the microbial composition with strains with more or less of the desired activity, then repeat steps d, e, and f until the cheeses are acceptable. Using the approach explained above, a 5-species surface starter can be applied. In addition to their use in smear, starters should be used in the cheese brines. In spite of the large volumes of industrial brines, this application might still be practical because of the low concentrations needed.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available