Cultures for surface ripening of smeared soft cheese

The surface microflora of smear soft cheeses showed distinct differences to semi-soft (Tilsit-type) cheeses. Highest cell counts were found for the yellow pigmented coryneform Microbacterium gubbeenense. A low proportion of yellow pigmented coryneform Arthrobacter nicotianae was also detected. Cream coloured Corynebacterium casei comprised 10-30 % of the total bacterial surface flora. Brevibacterium linens was found in variable but low cell counts in the surface microflora. For one cheese factory orange pigmented motile rods (Halomonas sp.) were typical. On all cheeses analysed, Debaryomyces hansenii and Geotrichum candidum were the predominant yeast species. Staphylococcus equorum typical for semi-soft cheeses was not detected. All cheeses showed bacterial contamination with enterococci and enterobacteria in variable degree. Mould contamination was not detected on any of the soft cheeses analysed. The screening for surface microorganisms showed that D. hansenii was essential for deacidification of the cheese surface and G. candidum was important for appearance and aroma development, which was similar to that of B. linens. Isolated strains of the abundant Microbacterium and Arthrobacter species showed similar aromatic profiles (urine-like) in liquid milk model systems where these bacteria showed yellow culture supernatants. Depending on degree of proteolysis and pH, these yellow strains could also liberate red pigments. In cultures mixed with brevibacteria, corynebacteria and Geotrichum candidum, microbacteria and Arthrobacter developed a typical smear cheese aroma dominated by aromatic sulfur components. Of all species tested the important B. linens und M. gubbeenense species showed the highest degrees of salt tolerance whereas A. nicotianae strains showed the lowest. For typical ripening D. hansenii (in cheese brines or smear) and G. candidum (in the cheese milk) were essential. The use of different Microbacterium and Arthrobacter species and B. linens strains had only a minor effect on appearance and aroma of cheeses. The use of S. equorum in cheese brines together with D. hansenii led to a faster deacidification of the cheese surface during the first 7 days as compared to typically old young smeared soft cheeses. A contribution to aroma development was not observed for S. equorum, however, due to the orange pigments a contribution to the development of the orange appearance of the soft cheese can be assumed. The presence of Corynebacterium species was not essential for typical ripening. Instead, the red discolouration observed at the edges of the cheeses could be related to a high percentage of C. casei. Orange areas showed higher cell counts of yellow microbacteria or Arthrobacter. A selected defined surface culture was used on pilot scale (4.000 1 cheese milk). Using lyophilised G. candidum, S. equorum, Halomonas sp., A. nicotianae and M. gubbeenense preparations, Limburg cheeses were smeared 4 times during ripening. Development of yeast cell counts was normal, growth of smear bacteria as well as colour- and aroma development was typical but slower than on lab scale.