Research report

​Variability of Alkaline Report

1 documents for this subject


Hannah Research
Banks, J ; Muir, D

The microbiological safety of milk depends on efficient pasteurisation and prevention of recontamination of the finished product. Pasteurisation, a heat treatment equivalent to a minimum holding of milk at no less than 71.8°C for 15 seconds - inactivates almost all potential pathogens found in raw milk. Validation of the effectiveness of the pasteurisation process is based on destruction of a natural milk enzyme-alkaline phosphatase (ALP). The sensitivity of this test hinges on the initial concentration of ALP in raw milk. The higher the initial activity, the more sensitive is the test. In addition, the sensitivity depends on the method of analysis of residual enzyme activity.

The detection limit for the reference test for ALP is equivalent to contamination of properly pasteurised cows milk by 0.1% raw milk. The implications of applying the test to pasteurised goat milk were explored because goat milk was reported to have natural levels of ALP around 10% of the activity found in cows milk. In such circumstances the sensitivity of the reference method for ALP would be reduced ten fold i.e. contamination of pasteurised goat milk by raw milk could reach a level of 1% before it would be detected. Variations in the initial pool of indigenous alkaline phosphatase in milk lead to different amounts of raw milk being allowed in the pasteurised milk product at the statutory pass level.

The research undertaken in this project aimed to reduce the potential threat to public safety associated with consumption of inadequately processed goat milk.

The effectiveness of the ALP test is determined by three factors: (1) The level of ALP in milk and its variability, (2) The sensitivity of the method of measuring ALP (spectrophotmetric, fluoroescence, bioluminescence) and (3) The levels set in legislation as acceptable standards for residual ALP activity.

The work undertaken within this project encompassed: a lactational study of the variability of ALP in a herd of British Saanen goats; a comparison of the effectiveness of bioluminescence, fluorescence and spectrophotometric measurements of ALP; a study of the formation of heat stable ALP; and a limited survey of residual ALP and microbiological quality of pasteurised goat milk retailed in Scotland.

There is little information in the literature regarding the variability levels of ALP in individual goat milks or the influence of lactational effects on secretion of ALP into milk. Seventeen British Saanen goats from the Hannah Research Institute herd were therefore sampled on a weekly basis throughout a full lactation, i.e. April 2001 to early January 2002. Individual morning and evening milking samples were tested for ALP daily during the first eight weeks of lactation Thereafter the morning milk from individual goats was sampled once a week throughout the remaining lactation period. Alkaline phosphatase was determined by the Fluorophos method (IDF Standard 155A:1999). Statistical modelling of the lactational data explored relationships between ALP levels in milk and goat genotype, age, lactation history and milk composition.

ALP levels in milk were lowest in the early stages of lactation and increased as the lactation progressed and milk yields declined. ALP levels were higher in milk samples from evening milk as compared with morning milk. The lowest mean value recorded for ALP in an individual goat for morning milk in May was 5823 mU/L. 15 of the 16 animals producing milk in May had ALP levels under 32000mU/L. The lowest level of ALP in an individual goat milk sample was 3630 mU/L. Values for ALP in goat milk increased as lactation progressed. In November, the minimum mean value for ALP in an individual goat morning milk was 18658 mU/L. The minimum value for an individual goat was 10410 mU/L. Mean vales for ALP in November ranged from 18658 mU/L to 782000mU/L.

The mean value throughout lactation (11 samples) for ALP in bulk herd goat milk was 38880 mU/L. The equivalent value for cows milk (mean of 22 samples) was more than tenfold higher at 560049 mU/L.

Early lactation pasteurised bulk goat milk which was contaminated with raw milk at levels of 1.0% raw milk did not fail the Fluorophos, Bioluminescence or Sanders and Sager ALP test. In mid lactation failures for goat milk were obtained at levels ranging from 0.7 to 0.9% contamination. In cows milk failures for the ALP tests were evident at 0.08 to 0.1% contamination of pasteurised milk with raw milk.

Results indicate that all tests currently available are not suitable for ALP determination in goat milk.

Twenty nine samples of pasteurised goat milk were collected from retail outlets in Ayrshire. All samples had satisfactory residual phosphatase levels but two of the samples had unacceptably high counts for Enterobacteriaceae.

Project Code: S01003