Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
letter
. 2003 Jan;69(1):719–721. doi: 10.1128/AEM.69.1.719-721.2003

Savinase Is a Bactericidal Enzyme

Edward M Smith 1,*, Laura E Green 1, David Mason 1
PMCID: PMC152419  PMID: 12514070

Mastitis is probably the most important disease affecting dairy cows worldwide and has been estimated to cost $1.5 to 2.0 billion/year in the United States alone (4). Although detection methodology continues to improve, microbiological culture is still the recommended method (2) for determining the bacterial content of a milk sample, despite this procedure being both time-consuming and expensive. Consequently, more rapid methods of identification and/or enumeration of bacteria in milk are being developed (3, 5). Many of these techniques, e.g., PCR-based methods, can be completed in a matter of hours, but they lack the specificity to be able to determine the bacterial species present or are unable to enumerate the bacteria present. There can also be a problem with techniques that require a period of enrichment prior to analysis.

The instrument of choice for overcoming problems of enumeration is a flow cytometer (FCM). In the paper by Gunasekera and colleagues published in Applied and Environmental Microbiology in March 2000, a method for the rapid detection and enumeration of bacteria in milk, without the need for enrichment, was outlined (1). Gunasekera et al. stated that an enzymatic treatment was required to remove and/or modify protein globules present in milk to enable the distinction of bacteria. Two alternative enzymes were suggested for this purpose, proteinase K (Sigma-Aldrich, Dorset, England) and savinase (Novozymes, Bagsvaerd, Denmark), a protease enzyme developed for the detergent industry.

However, when we used savinase to recreate and develop this method to identify Staphylococcus aureus in milk, we found that bacterial recovery from the original milk sample was below 50% (data not shown). Our investigations led us to determine the effect (if any) that savinase itself was having on S. aureus. The results indicated that two separate batches of savinase were immediately bactericidal towards 12 strains of S. aureus in pure culture. These strains included the one used by Gunasekera et al. (NCTC 4163), the Newbould 305 strain (NCIMB 702892), and 10 strains isolated from cattle with clinical or subclinical bovine mastitis.

The effects of savinase, proteinase K, and alcalase (Novozymes), the latter being an enzyme designed to break down proteins in milk to produce infant milk formula, on S. aureus Newbould 305 inoculated into pasteurized milk were also investigated. Savinase took between 3 and 5 h (at 37°C) to kill all the bacteria in the inoculated milk sample (Fig. 1) and resulted in a significantly lower (P < 0.001) bacterial count than the other two treatments after only 60 min. This is well within the time frame (60 to 90 min) of the protocol described by Gunasekera and colleagues.

FIG. 1.

FIG. 1.

Open in a new tab

How protease enzymes can affect the growth of S. aureus Newbould 305 in pasteurized milk. Shown are the effects of no enzyme (control) (⋄), proteinase K (□), alcalase (+), and savinase (×).

These results may be due to a difference in the formulations of savinase supplied in the United Kingdom and Australia, and we would be interested to know whether this is the case. However, we would like to point out that, in the United Kingdom, savinase appears to be unsuitable for use in the detection of S. aureus in milk.

REFERENCES

  • 1.Gunasekera, T. S., P. V. Attfield, and D. A. Veal. 2000. A flow cytometry method for rapid detection and enumeration of total bacteria in milk. Appl. Environ. Microbiol. 66:1228-1232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Hogan, J. S., R. N. González, R. J. Harmon, S. C. Nickerson, S. P. Oliver, J. W. Pankey, and K. L. Smith. 1999. Laboratory handbook on bovine mastitis, rev. ed. National Mastitis Council, Inc., Madison, Wis.
  • 3.Pheuktes, P., P. D. Mansell, and G. F. Browning. 2001. Multiplex polymerase chain reaction assay for simultaneous detection of Staphylococcus aureus and streptococcal causes of bovine mastitis. J. Dairy Sci. 84:1140-1148. [DOI] [PubMed] [Google Scholar]
  • 4.Wells, S. J., S. L. Ott, and A. Hillberg Seitzinger. 1998. Key health issues for dairy cattle—new and old. J. Dairy Sci. 81:3029-3035. [DOI] [PubMed] [Google Scholar]
  • 5.Yazdankhah, S. P., and E. Olsen. 1998. Simple and direct detection of Staphylococcus aureus in milk by a tube coagulase test. Lett. Appl. Microbiol. 27:111-115. [DOI] [PubMed] [Google Scholar]
Appl Environ Microbiol. 2003 Jan;69(1):719–721.

Author's Reply

Thusitha S Gunasekera 1,*, Duncan A Veal 1

This letter is written in response to the Letter to the Editor written by Smith et al., titled “Savinase Is a Bactericidal Enzyme.” These authors used our method (1) and standard plate count methods to assess the total count and the viability of savinase-treated S. aureus bacteria. Their experiments demonstrated that savinase is a bactericidal enzyme and may not be suitable for clearing milk proteins as a sample preparation procedure for FCM detection and enumeration of S. aureus bacteria in milk.

The original article published in Applied and Environmental Microbiology (1) described the use of FCM for rapid detection and enumeration of total bacteria in milk. In this method, savinase 16L (type EX; activity, 16 KNPU/g) and proteinase K were used as milk-clearing agents, and the nucleic acid binding fluorescent dye SYTO BC was used for staining bacteria in milk. Ultra-high-temperature-treated (UHT) milk was inoculated with Escherichia coli or S. aureus cells at concentrations between 103 and 108 ml−1 and analyzed by the FCM and plate count methods. To provide more-realistic data, raw milk samples were analyzed by the FCM method and the results were compared with those obtained by the standard plate count method. Samples used for analysis by the FCM method were subjected to milk clearing, but subsamples used for analysis by the plate count method were not treated with savinase or proteinase K. Results from the FCM method had good correlation with those from the plate count method.

To address the concerns of Smith et al., we have determined, (i) the effects of savinase (10 μl per 100 μl of UHT milk) on total counts and numbers of CFU over time (0 to 300 min) and (ii) the effects of a higher concentration of savinase (50 μl per 100 μl of UHT milk) on both total counts and numbers of CFU determined by FCM after incubation of the sample for 40 min. Using the concentration recommended for UHT milk (10 μl per 100 μl of UHT milk) (1), we are unable to reproduce the effects of savinase described by Smith et al., either on total counts or on numbers of CFU determined by FCM (Fig. 1). Unfortunately, Smith et al. did not provide the FCM results nor did they indicate the concentrations of savinase that they used for comparison. Using the higher concentration recommended for raw milk (50 μl per 100 μl), we found no reduction of the total count obtained by FCM but found a reduction in the number of CFU (Fig. 2). Savinase at even higher concentrations and longer incubation times than those recommended (1) did not decrease the total bacterial count (Fig. 1 and 2).

FIG. 2.

FIG. 2.

Open in a new tab

Effect of savinase on the viability of S. aureus. UHT milk was inoculated with S. aureus, and 100-μl milk samples were treated with 10 μl of savinase for different time intervals. Samples were processed as described by Gunasekera et al. (1), and the total numbers of bacteria were determined by FCM. Viable counts were determined by the plate count method. Ovals, plate counts; squares, FCM counts.

The effect of the milk-clearing procedure on bacterial viability was not an issue, since the method we described (1) was primarily developed to detect and enumerate the total number of bacteria in milk. SYTO BC stains both live and dead bacterial cells (R. Haugland, Handbook of Fluorescent Probes and Research Chemicals. Molecular Probes, Inc. Eugene, Oreg. 1996). Thus, any effect that savinase might have on bacterial colony-forming ability is not going to affect the total count (Fig. 2). Thus, the effect of savinase on colony-forming ability is not an important issue (as described by Smith et al.) for our method (1). The effect of higher concentrations of savinase on CFU is not surprising. In fact, for viability assays, we use alternative milk-clearing agents or low levels of savinase (2).

Dr. Smith and his colleagues have used the numbers of CFU to determine the viability of savinase-treated, stressed bacteria. I would contend that the numbers of CFU do not accurately reflect the viability of all stressed bacteria (3). This is especially the case where cells may be viable but not culturable. Other parameters, such as cell vitality, cell permeability, membrane potential, and enzyme activity, are also useful indicators in measuring cellular activities.

Another point of concern of Dr. Smith and his colleagues is the formulation of savinase supplied in the United Kingdom and Australia. Savinase is available in six different formulations, and for our studies in Australia we used savinase 16L, type EX (declared activity, 16 KNPU/g). Batch-to-batch variation is also worth considering. We would therefore be happy to supply a sample of the savinase we used for the above experiments to Dr. Smith, or we can test the savinase used by Smith et al. We certainly expect that different formulations have different effects on milk proteins and, more importantly, on the bacteria in milk. Therefore, care in choosing the correct enzyme type for milk clearing is important, particularly when viability testing is required.

FIG. 3.

FIG. 3.

Open in a new tab

Effect of 50 μl of savinase on total S. aureus populations. UHT milk was inoculated with S. aureus, and 100-μl milk samples were treated with 50 μl of savinase for 40 min. Samples were processed as described by Gunasekera et al. (1), and the total numbers of bacteria were determined by FCM. Viable counts were determined by the plate count method. Controls (bacteria in buffer) were treated with sterilized distilled water. Filled bars, FCM counts; open bars, plate counts. Data were analyzed by using Student's t test (an asterisk denotes significant difference at a P value of <0.05).

REFERENCES

  • 1.Gunasekera, T. S., P. V. Attfield, and D. A. Veal. 2000. A flow cytometry method for rapid detection and enumeration of total bacteria in milk. Appl. Environ. Microbiol. 66:1228-1232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gunasekera, T. S., A. Sørensen, P. V. Attfield, S. I. Sørensen, and D. A. Veal. 2002. Inducible gene expression by nonculturable bacteria in milk after pasteurization. Appl. Environ. Microbiol. 68:1988-1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Porter, J., C. Edwards, and R. W. Pickup. 1995. Rapid assessment of physiological status in Escherichia coli using fluorescent probes. J. Appl. Bacteriol. 79:399-408. [DOI] [PubMed] [Google Scholar]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES