Livestock Research for Rural Development 18 (4) 2006 Guidelines to authors LRRD News

Citation of this paper

Effects of starter cultures and heat treatments of milk on Bafut cheese; a local cheese from the High Lands of Cameroon

A Kameni, J Mbanya and S Mendi

Institute of Agricultural Research for Development,
IRAD, Box 415 Garoua, Cameroon


The effects of starter cultures (CH1 or CH44) and pasteurisation of cheese milk on the chemical and microbiological composition of Bafut cheese, a local cheese from Cameroon, were investigated. Bafut Cheeses were manufactured from raw and pasteurised fresh morning milk adjusted to 2.5 % fat. A randomised block design was used in a 2*2 factorial experiment. FActor 1 was the types of starters (CH1 versus CH44) and factor 2 was the types of milk (raw versus pasteurised). A mesophilic culture (CH44), containing Lactococcus lactis, Lactococcus cremoris, Leuconostoc cremoris and Lactococcus lactis biovar. diacetylactis, and a thermophilic yogurt culture (CH1) containing Lactobacillus bulgaricus and Streptococcus thermophilus were included in cheese milk at a rate of 1.5%.

Starter cultures CH1 and CH44 affected cheese yields, moisture content, acidity, FFA and TVS. Heat treatment destroyed all the coliform organisms and probably other undesirable micro-organisms of raw milk and this improved the microbiological quality of the cheeses. Lactobacilli were found to be the most important organism by their presence in large numbers in all the cheeses. Cheese made with starter cultures were preferred by panelists with culture CH1 producing the best cheese.

This overall pattern confirms the view that starter culture (CH1) is likely to be a suitable culture for Bafut cheese and its regular use in cheese-making could be beneficial to consumers.

Key words: Bafut cheese, cow milk, heat treatment, starter cultures


Milk was originally converted into cheese primarily to preserve its excellent nutritional properties using natural acidifying action of undefined lactic bacteria. Development in the dairy industry of the West has brought about the use of various selected commercial starter cultures for cheese production. However, in sub-Sahara Africa where surplus milk is sometimes available for processing into cheese, highly efficient techniques for cheese making have been timidly applied (IDRC 1984). Attempts to develop or introduce cheese making in this part of the continent had been initiated with encouraging results (Ogundiwin and Oke 1983; Boor et al 1987; Faist et al 1987). The International Livestock Centre for Africa (ILCA) introduced the production of Halloumi, Quesco Blanco, Feta, Domiati and Gibbneh Beda cheeses in Ethiopia and Sudan (O'Mahony and Peters 1987). Acidification of cheese milk is achieved by inoculating fresh raw milk with a bit of old fermented milk as starter culture followed by rennet coagulation.

An example of cheese resulting from technology transfer in Cameroon (Herwig and Schwerdtfeger 1980) is Bafut Cheese found on the local market as a cylindrical block weighing about 1 to 2 kg (Kameni et al 1994). It is produced from raw milk and involves renneting, salting and pressing. Maturation is carried out at room temperature (17 to 22°C) for about 4 to 6 weeks. For the valorisation of such local cheeses emphasis be should be laid on yield and ripening, since both have economic impact on manufacturing cost and returns. The addition of some selected micro-organisms for their ability to serve as starter cultures for the production of Wara, a West African soft cheese (Sanni et al 1999) proved useful in improving cheese quality. The possibilities for improving and/or standardising the properties of Bafut cheese, particularly through modifications of the manufacturing process (eg. by the introduction of pasteurisation and the use of starter cultures), need to be examined. This work evaluates the effects of starter cultures (CH1 and CH44) and the heat treatment of cheese milk on chemical and microbiological composition of Bafut Cheese.

Materials and Methods

Fresh morning milk from a mixed herd of dairy cows of the Institute of Agricultural Research for Development (IRAD) was used, and the required fat content obtained by mixing six volumes of raw whole milk with four volumes of raw skimmed milk to give a final fat content of 2.5 to 2.7%. Two types of starter culture were considered, and both were supplied by Chr. Hansen's Laboratory, Reading, UK. The first was a standard, mesophilic culture for cheesemaking (Code : CH44), which contained a mixture of four bacteria species, namely: two acid producing species Lactobacillus lactis biovar. diacetylactis, Lactococcus lactis sub-sp. lactis and Lactococcus lactis sub-sp. cremoris in the ratio of 1:1 according to the manufacturer's instructions. In addition, two flavour-enhancing species: Leuconostoc mesenteroides sub-sp. cremoris and Lactobacillus lactis biovar. diacetylactis were included at a lower level. The second culture was a thermophilic yogurt culture (Code : CH1) which contained a mixture of Lactobacillus delbrueckii sub-sp. bulgaricus and Streptococcus thermophilus in the ratio 1:1 as specified by the manufacturer. This culture was chosen in case of need to speed up the ripening stage prior to the addition of rennet, or desire to increase the level of acidity attained. Cultures were purchased as lyophilized granules, and were reactivated by the method of Cox and Lewis (1972) through three subcultures before being used. This approach can give rise to some minor changes in the balance between the species, but it is the most practical system for areas remote from culture suppliers.

Experimental design

A 2X2 factorial arrangement in a randomised block design was used for the experiment, with a block being the batch of milk (40 l) used. The treatments were the two types of starters (CH1 and CH44) and the two types of milk (raw versus pasteurised). Cheese was manufactured according to the procedure of the Sister of Emmanuel (Kameni et al 1994) in batches of four vats each containing 10 kg of milk. Two vats contained raw milk at 350C, and the other two contained pasteurised milk (72C for 1 minute) cooled to 35C. Starter cultures (CH1 and CH44) were inoculated into one vat of raw milk and pasteurised milk respectively, at the rate of 1.5 %. The experiment was repeated six times giving a total of 24 vats of cheese manufactured..

Chemical and microbiological analyses

Matured cheeses were analysed after four weeks with respect to, titratable acidity (TA), dry matter, (DM), fat, pH., and salt (Marth 1978), total protein by the Kjeldahl method (AOAC 1980), tyrosine index, (Hull 1947), total volatile substances (TVS) (Hempenius and Liska 1968) and free fatty acids (FFA), (Dulley and Grieve 1974). The microbiological analyses included total colony counts (TCC), aerobic spore counts on Milk Agar, counts for lactobacilli on Rogosa Agar, presumptive coliform counts on Violet Red Bile Agar and yeasts and moulds on Rose Bengal Chloramphenicol Agar (Marth 1978) on poured plates.

Taste panel analysis

A panel of 6 women selected for their past habit of consuming cheese was used to assess the organoleptic quality of the cheeses. A vocabulary to evaluate de features of the product was developed using the continuous marking scheme of Powers (1984). A total of four training sessions were used to prepare panel members followed by four evaluation sessions. The panellists were presented with the experimental samples, identified by coded letters. They were asked to marked the position of a sample on a 10 cm line which corresponded to its score. The features under assessment were elasticity, acidity, peppery flavour, close texture, cheese-like flavour and off odours. The controls were commercial Babut cheese.

Statistical analyses

For statistical analysis, the GLM procedure of the SAS (1985) was used. Means separation was done by the Duncan multiple range method to separate means for significant differences

Results and discussions

Effects on cheese composition

The treatments given to cheese milk lead to significant (P<0.05) differences in cheese yields, moisture content, TA, and TVS (Table 1). Raw milk treated with starter cultures (CH1 or CH44) produced cheeses with relatively higher dry matters yields, fat content, than similar cheese from heat treated milk. TA values were significantly higher (P<0.05) in raw cheeses with starter culture CH44 leading to high levels of lactic acid in the curd compared to CH1. However, the combined effects of starter culture and the heat treatment given to cheese milk did not affected (p>0.05) the levels of protein and salt contents, the tyrosine indices and the pH of the resultant cheeses.

Table 1.  Effects of treatments on chemical composition of Bafut cheese made  from raw and pasteurised milk with starter cultures CH1 and CH44

Parameters evaluated

Treatments of cheese milk


Raw milk /CH1

Raw milk /CH44

Pasteurised Milk /CH1

Pasteurised Milk/CH44

 Dry matter, %

57.7 1.2 b

61.2 0.70 a

53.4 0.50 c

56.6 1.3 b


Cheese yields, Kg/100kg

8.40 0.4 b

8.16 0.30 c

8.75 0.6 a

8.76 0.9 a


Fat, %

27.7 0.70


25.1 1.4

26.1 0.5


Fat in DM*, %






TA as lactic acid, %

2.49 0.06 a

2.46 0.03 a

1.51 0.03 c

2.21 0.06b



4.53 0.02

4.57 0.03

4.73 0.07

4.60 0.05


Proteins, %

26.6 1.2


26.8 0.3

26.7 0.1


Tyrosine, mg/5 ml of cheese filtrate

0.04 0.01

0.05 0.01

0.04 0.01

0.05 0.01


TVS, m-moles NaOH/100 ml of distillate

3.05 0.22 b

2.13 0.2 c

3.85 0.04 a

3.20 0.09 b


FFA, m-moles of NaOH/100g

6.76 1.3 b

4.91 1.10 c

7.80 0.5 a

6.56 040 b


Salt, %

1.47 0.3

1.57 0.2

1.55 0.13

1.43 0.1


SED= Standard errors of differences of means

Values with different letters in the same line are different (P<0.5)

*Calculated values

Higher yields observed with pasteurised milk are in line with the results of Lau et al (1990) who reported that pasteurisation of cheese milk had little effect on fat recovery, but improved the recovery of nitrogen. In effect, heat treatment of milk proteins reduces their ability to clot under rennin action, and increases their hydration potential. This reduces the rate of whey expulsion, so resulting in high moisture cheese. However observed yields were lower than those reported by Bruhn and Franke (1991) for Edam cheese, and Cheddar cheese made with starter cultures (Coggins 1991). The low yield may be explained by the standardization of milk prior to cheese making (milk was adjusted to 2.5% fat), and excessive drying during maturation.

The high level of acidity in these cheese is a good indicator of starter activity and vitality. In raw milk, the natural micro-flora, composed in part of lactic acid bacteria, would also ferment lactose to lactic acid. Thus, the action of the natural milk micro-flora, together with that of the starter cultures, is responsible for the high level of acidity observed in cheese made from raw milk.

Culture CH1 exhibited a marked tendency to produce more volatile compounds and free fatty acids, particularly in pasteurised milk cheeses. Higher values of FFA obtained for pasteurised milk cheese are in contradiction with earlier results from McSweeney et al (1993) who reported higher levels of free fatty acids in cheese made from raw milk. High values of FFA is desirable in cheese and could be attributed to the specificities of the strains of culture micro organisms used. Free fatty acids in cheese are important in defining the specific flavour of each cheese. They derived from two majors sources: breakdown of fat by lipolysis and metabolism of carbohydrates and animo acids by bacteria. Lipolysis is the principal contributor of free fatty acids of chain length C4 or greater (Foda et al 1974; Dulley and Grieve 1974). These results suggest that the lipolytic activities of the two starter cultures were not similar.

Effects on microbiological contents

The mean total colony counts for the major micro-organisms of the cheeses are listed in Table 2. The levels of various micro-organisms (TCC, aerobic spore formers, and yeast and moulds) in cheese were not significantly (p>0.05) affected by the starter cultures used, although pasteurised milk cheeses had slightly higher TCC. All finished cheese had similar numbers of micro-organisms in each group except for lactobacilli and Coli aerogenes. Cheeses made with starter culture (CH1) had significantly (p<0.05) higher level of lactobacilli compared to cheese made with starter culture (CH44). There was no growth of Coli aerogenes in cheese made from pasteurised milk. The heat treatment was enough to destroy these organisms. However, raw milk cheese made with starter cultures had relatively lower coliforms.

Table 2.  Micro-biological composition of Bafut cheese made from raw and pasteurised milk with starter cultures CH1 and CH44 (mean values as cfu / g of cheese)

Micro-organisms evaluated

Treatments of cheese milk


Raw milk /CH1

Raw milk /CH44

Pasteurised Milk /CH1

Pasteurised Milk/CH44


Total colony counts

8.7 x 109

1.2 x 1010

4.6 x 1010

1.8 x 1010


Aerobic Spores counts

2.0 x 103

1.6 x 103

8.1 x 102

1.6 x 103


Lactobacilli counts

56 x 108b

33 x 108b

240 x 108a

1.4 x 108c


Presumptive coliform counts

4.9 x 102a

2.0 x 102a




Yeasts and moulds

7.6 x 103

7.2 x 103

7.8 x 103

4.1 x 103


SED = Standard errors of differences of means

Values with different letters in the same line are different (P<0.5)

The total colony counts of bacteria per gram of cheese were quite high compared to other cheeses like Edam, Cheddar and Bombel whose counts usually settle between 106 and 108 cfu/g (Price and Call 1969). The addition of Lactobacillus bulgaricus through the culture would have increased the initial number of lactobacilli in the cheese. Lactobacilli were found in appreciable numbers in cheese made with pasteurised milk and starter culture (CH44) suggesting that some of them might have survived heat treatment of cheese milk, and grew later in the cheese. Lactobacilli are a major organism encountered in traditional cheese (Poznaski et al 2004). The aerobic spore-forming bacteria found in cheese came from raw milk, and pasteurisation would likely not cause any reduction in their numbers (Saubois et al 1991). Various groups of micro-organisms identified are part of those generally encountered in raw milk cheese (Badis et al 2003).

Taste panel appreciation of cheeses

Taste panellists found significant differences (P<0.05) between commercial Bafut cheese and similar cheeses made with starter cultures (Table 3). Commercial Bafut cheese, the control received the lowest scores in all the features under investigation except for off odours. Similarly, cheese made with starter culture CH1 received significantly higher scores for elasticity, mouth feel and cheese flavour. They found that cheese made from starter cultures were better from the overall scores given. The preferred cheeses were in order: cheese from culture CH1 followed by CH44 and the commercial cheese. The panel preference in favour of PCH1 suggests that starter culture CH1 is likely to be a suitable culture for Bafut cheese.

Table 3.  Mean taste panel scores of Bafut cheese attributes and similar cheeses made from fresh milk and starter cultures

Cheese parameters

Bafut cheese

Pasteurised Milk/CH44

Pasteurised Milk /CH1


4.46 1.29a

6.34 1.09 b

8.65 1.05 c

Close texture

3.76 0.72 a

7.12 0.52 b

7.06 0.68 b

Mouth feel

5.44 0.59 a

7.06 0.68 b

8.70 0.87 c


5.86 0.53 a

7.82 0.74 b

7.06 0.68 b

Pepper flavour

5.12 0.88 a

8.40 0.58 b

7.91 0.45 b

Cheese flavour

6.97 1.29 a

8.53 0.38 b

9.56 0.27 c

Off odour

5.86 0.53 b

2.89 0.86 a

2.76 0.69 a

Values with different letters in the same line are different (P<0.5)



AOAC 1980 Official methods of analysis. 13th edition Association of Official Analytical Chemists. Washington DC

Badis A, Guetarni D, Moussa-Baoudjemâa B, Henni D E, Tornadijo M E. and M Kihal 2003 Identification of cultivable lactic acid bacteria isolated from Algerian raw goat's milk and evaluation of their technological properties. Food Microbiology, Volume 21 (3)343-349.

Boor K J, Brown DL and Fitzhugh A H 1987 The potential for goat milk production. World Animal Review, Volume 62:31-40.

Bruhn J C and A A Franke 1991 Raw milk composition and cheese yield in California:1987 and 1988. Journal of Dairy Science. Volume 74 (3) 1108-1114

Coggins J S 1991 Predicting Cheddar cheese yield in an individual plant: Van Slyke revisited Journal of Dairy Science Volume 74 359-368.

Cox W A and Lewis J E 1972  Safety in microbiology. Editor D A Shapton, Academic Press, Rig Board, London.

Dulley J R and Grieve P A 1974 Volatile fatty acids in Cheddar Cheese. The Australian Journal of Dairy Technology, September 120-123.

Foda E A, Hammond E G, Reinhold G W and Hotchkiss D K 1974 Role of fat in flavor of Cheddar cheese. Journal of Dairy Science, Volume 57:1137-1142.

Faist B Chardonnet P and Nibogora J 1987 Technologie fromagère artisanale en zone tropicale: Résultats techniques d'une fruitière en milieu rural au Burundi. Revue d'Elevage et Médecine Vétérinaire des Pays Tropicaux, Volume 40 (2) 191-197.

Hempenius W L and Liska B J 1968 Methods of determining volatile acids in cultured dairy products. Journal of Dairy Science, Volume 51:221-222.

Herwig M K and Schwerdtfeger K 1980 Final report on a short term assignment in Wum Cameroon. German Appropriate Technology Exchange. P. O. Box 5180 D-6236 Eschborn 1, Federal Republic of Germany

Hull M E 1947 Studies on milk proteins. II Colorimetric determination of the partial hydrolysis of the proteins in milk. Journal of Dairy Science, Volume 30:881-884.

InternationalDevelopment Research Center (IDRC) 1984 The potential for small scale milk production in Eastern and Southern Africa. Ottawa, Canada. IDRC MR 98E.

Kameni A, Mbanya N J and Robinson R K 1994 Bafut: A study of the manufacture and characteristics of Bafut, a local cheese produced at a convent in Cameroon. Dairy Industry International, Volume 59 (6), 27, 29.

Lau K Y, Barbano D M and Rasmussen R R 1990 Influence of pasteurization on fat and nitrogen recoveries on Cheddar cheese yield. Journal of Dairy Science, Volume 73:561-570.

Marth E H 1978 Standard methods for the examination of dairy products. 14th edition American Public Health Association, Washington DC.

McSweeney P H, Fox P F, Lucey J A, Jordan K N and. Cogan T N 1993 Contribution of the indigenous microflora to the maturation of Cheddar cheese, International Dairy Journal 3 (7) 613-634.

Ogundiwin J O and Oke L O 1983 Factors affecting the processing of Wara-A Nigerian white cheese Food Chemistry, Volume 11, 2 (1) 1-13.

O'Mahony F and Peters J 1987 Sub-Saharan Africa Options for smallholder milk processing. Word Animal Review, Volume 62:16-30.

Powers J J 1984 Current Practices and Application of Descriptive Methods: In Sensory Analysis of Food. Editor J R Piggot, Elsevier Applied Science Publisher, London

Poznanski E, Cavazza A, Cappa F and Cocconcelli P S 2004 Indigenous milk microbiota influences the bacterial development in traditional cheese from an alpine natural park. International Journal of Food Microbiology, Volume 92 (2) 141-151.

Price W V and Call A O 1969 Cheddar Cheese: comparison of effects of raw and heated milk on quality and ripening. Journal of Milk and Food Technology. Volume 32:304.

Sanni A I Onilude A A and Monoh M O 1999 Selection of starter and a starter-mediated novel procedure for production of Wara, a West African soft cheese. International. Journal of Food Technology, Volume 34:325-333.

Saubois A, Basilico J C and Simonetta A 1991 Calidad microbiológica de leche cruda. III and- Incidencia de bacterias esporuladas arobias y anaerobias, levaduras y hongos. Revista Argentina de Lactología, Año 3 (5) 79-90

SAS Institute Inc1985 SAS user's guide: Basis; version 5 edition . Cary, NC.: SAS Institute Inc.1290 pp

Received 9 January 2006; Accepted 23 February 2006; Published 10 April 2006

Go to top