There are many ways in which traditional cheeses can be classified. Criteria such as country of origin, type of milk used, species of animal used to produce the milk, fat content, moisture content, texture, whether mould ripened or not, cheese making process used, moisture in the non-fat solids have been and continue to be used. These criteria have been used either singly or in combination.
These descriptive approaches are limited in that they provide no theoretical insight into why one cheese is different from another. In other words, why is Gouda cheese different than Cheshire or what makes Emmental different than Cheddar cheese?
For many years researchers in New Zealand, the UK, Ireland, the Netherlands and elsewhere were aware that there were significant differences in the pH and mineral concentrations of the major cheese varieties.
Lawrence et. al. (1984) first suggested that cheeses could be classified on the basis of two criteria, pH and calcium content. This approach is illustrated in fig. 1 for Swiss, Gouda, Cheddar and Cheshire cheeses.
Swiss, Gouda and Cheshire cheeses (fig. 1) exhibit a narrow range for pH and calcium concentration compared with Cheddar. Cheddar has a much wider range for both pH and calcium content. Lawrence et. al. (1984) suggested that Cheddar is a popular variety for this reason, this cheese type can have a wide range for pH and calcium content and still meet customer expectations; this suggests that Cheddar is easy to make!
The model proposed by Lawrence and his colleagues suggests that the differences between the various traditional cheese types are based to a large extent on the basic structure of the cheese. This basic structure is ultimately determined by the properties of the protein matrix in the cheese. This is dependent on the pH of the whey at the point when the curds and whey are separated. It is at this point where the mineral content of the cheese and its residual lactose content are largely determined. The residual lactose will influence the lowest pH, the highest level of acidity, that can be attained. Note that the buffering capacity of the curd will also influence the final pH. Buffering capacity will also be affected by the pH at which the whey is drained.
Why does pH at whey drainage determine cheese type? The explanation involves the effect of pH on the casein micelle and ultimately the properties of the protein aggregates in cheese.
The casein micelle is composed of numerous sub-micelles, which are held together by colloidal calcium phosphate. As pH decreases, acidity increases, and the colloidal calcium phosphate becomes soluble. Put simply, the concentration of the binding agent between the sub-micelles decreases. As the calcium phosphate becomes soluble, the casein micelle starts to disassemble. This process can be monitored by measuring the size of protein aggregates using electron microscopy.
Note this is a somewhat simplistic explanation since many micelle properties can only be explained by a “dual binding” model such as the one proposed by Horne (1998)*. This suggests that casein micelles are formed because of two binding mechanisms namely hydrophobic attraction and colloidal calcium phosphate bridging.
The mineral content of cheese is largely determined by the quantity of calcium phosphate lost from the curd, which is mainly dependant upon the pH of the whey at drainage; the pH of the whey at this point is dependant on starter activity. The latter depends on several factors including:
- concentration of starter
- starter type
- time which has elapsed since starter addition
It is this loss of calcium phosphate, and its effect on the properties of protein aggregates in cheese, that is critical to the development of cheese type.
Low acid cheeses (high pH) such as Swiss have a high mineral content and have protein aggregates largely composed of intact casein micelles. Electron microscopy reveals an extensive protein matrix composed of strings of protein aggregates. Such cheeses have relatively elastic properties.
At the other extreme Cheshire cheese has a low mineral content and has very small protein aggregates. This cheese has virtually no elasticity; it is ‘crumbly' and will break easily. Electron microscopy reveals a very weak internal protein matrix.
Cheddar is intermediate between Gouda and Cheddar with regard to the size of protein aggregates and the density and strength of the internal protein matrix.
The basic structure of a cheese affects proteolysis and influences cheese flavour
In general proteolysis in cheeses made with calf rennet is dependent on the activity of two major proteases, chymosin and plasmin. Chymosin is one of several enzymes present in rennet. Plasmin is a native milk protease.
Chymosin is active at low pH whereas plasmin is active at higher pH, pH optimum ca. 7.5. For this reason chymosin activity during ripening is highest in Cheshire followed by Cheddar followed by Gouda and is least active in Swiss cheese. Plasmin has the opposite activity profile. Plasmin is mainly responsibly for the hydrolysis of ß-casein which only weakly hydrolysed by chymosin.
Application of acidity and calcium model of Lawrence and colleagues
The Lawrence model provides an explanation of why Cheshire and similar cheeses are crumbly or easily broken; they have a weak internal protein matrix. This is associated with low pH and calcium content. On the other hand, cheeses that have a more elastic texture and flex/bend to some extent e.g. Cheddar have a higher pH and calcium content and a stronger internal protein matrix.
The previous discussion has clearly identified the importance of pH at whey drainage on the development of cheese type and the requirement to regulate acidity in cheese manufacture. pH regulation in the manufacture of Cheddar, Cheshire, Swiss and Gouda cheeses will be explored elsewhere.
Modeling the grade value of Cheddar Cheese
The above information on pH and calcium can be used to model the grade value of Cheddar cheese. Using this model readers can use the calculator to predict the grade of Cheddar cheese and if you have access to factory data see how well the calculator performs compared to a commercial grader.
Horne, D.S. (1998). Casein interactions: casting light on the Black Boxes, the structure in dairy products. International Dairy Journal. 8, 171-177.
Lawrence, R.C., Heap, H.A. and Gilles, J. (1984). A controlled approach to cheese technology. Journal Dairy Science 67, 1632-1645.
Search for literature cited
How to cite this article
Mullan, W.M.A. (2005).
[On-line]. Available from: https://www.dairyscience.info/index.php/cheese-manufacture/114-classification-of-cheese-types-using-calcium-and-ph.html . Accessed: 24 May, 2016.