Zentralbl. Mikrobiol. 142 (1987), 379-386 VEB Gustav Fischer Verlag Jena

[Dairy Research Laboratory and Water Pollution Control Laboratory, National Research Center, Dokki, Cairo, Egypt]

Status of Membrane-Filter Procedure in the Examination of 'Vater and Milk for Staphylococcu8 aureU8

Anwendung der Membranfilter-Technik bei der Untersuchung von Wasser und Milch auf die Anwesenheit von Staphylococcu8 aureU8

M. M. SABBOUR and R. T. EL-ZANFALY

With 4 Figures

Summary A comparative study was carried out to assess the performance of four selectiye media, namely

Baird·Parker (B-P), Mannitol-Salt Agar (M.S.A), Staphylococcus·110 (S-110), and Chapman-Stone (C.S) in solid and liquid forms for recovery of Staphylococcu8 aureU8 from water and milk, using membrane filter technique.

The descending order of efficiency for demonstrating the presence of inoculated S. aureU8 from unsterilized Nile water was (B-P) -(M.S.A) + (S-110) and (C.S) medium in solid form and changecl to (S-llO) - (l\I.S.A) - (C.S) when used in the liquid form. With sterile Xile water, the descending order of efficiency of the tested media was (B-P) + (M.S.A) - (S-110) and (C.S) medium in the solid form and (M.S.A) + (S·llO) and (C.S) in the liquid form.

In testing the inoculated milk using solid media, (S-110) and (M.S.A) tend to show similar results awl took intermediate position between (B-P) and (C.S). \Vhen liquid media were used, the descending order of productivity was (S-llO) - (l\I.S.A) and (C.S).

Zusammenfassung Untet Verwendung der Membranfilter·Technik wurde die Brauchbarkeit von 4 selektiven

l\Iedien (Baird-Parker-Medium, :>Iannitol-Salz-Agar, Staphylococcus-llO-Medium und Chapman­Stone-Agar) in fester wie in fltissiger Form zum Nachweis von Staphylococcu8 aU1'eus in \Vasser und Milch gepriift. Bei gleicher Zusammensetzung boten die Medien in beiden Formen gleiche Vorau'"etzungen fiir die Entwicklung von S. aUTeU8.

Since its introduction as a tentative method for coliform enumeration in the 10th edition of Standard Methods (A.P.R.A. 1955), the membrane filter has gained wide application not only for total coliform but also for faecal coliform (GELDREICH et al. 1965), enterococ~i (SLANETZ et al. 1965), staphylococci (ALlco and PALENCHAR 1975), r~'brio comma (FELslmFELD and ROKKAKU 1956), Salmonella and Shigella (ToMPKIN et al. 1963), iron bacteria (LUESCHOW and MACKENTHUM (1962), and a variety of other bacterial tests. The membrane filter method was adopted for micro­biological analysis of drinking water (CLARK and KABLER 1952), natural surface water (GcmTz et al. 1952), industrial water and sewage (BUCH 1953), swimming pools (FAvlmo and DRAKE 1964), food and food processing equipments (WINTER et al. 1971), milk (GRAVES and SCHIPPER 1966), beverages (HAYS and O'BRIEN 1971), and beer (NOBILE 1967). Most of the so-called limitations of the membrane filter method can be removed by review and correction of deficiencies in the labora-

380 ~I. 1\1. SABBOUR and H. T. EL-ZANFALY

tory procedures being used by the workers. It is appropriate to call attention to the culture media designed for selective recovery of bacterial marker groups or pathogens, using agar pour plates, streak plates, or broth cultures have to be designed to compen­sate for physico-chemical properties of membrane filter materials (WOLOCHOM 1958; GELDREICH 1976).

This study was designed to evaluate four selective media in solid and liquid form for Staphylococcu8 aureU8 recovery from water and milk via membrane filter technique.

Materials and Methods

Staphylococcus aureus strain (enterotoxin A producer) was obtained from stock culture of Food and Drug Administration (U.S.A.). The strain was cultivated in nutrient broth (Difco Cor­poration, Baltimore, Maryland, U.S.A.) at 37°C for 24 h. 1 ml of the culture was inoculated into 500 ml of the following media and kept at room temperature (20°C ± 2°C).

a) Fresh water from River Nile. b) Nile water after filtration through 0.8 flm pore size Sartorius membrane filter for large organ­

isms removal and autoclaved at 121°C, 15 psi for 15 min. c) Reconstituted skim milk powder (100 gil of sterile tap water). d) Sterile reconstituted powdered skim milk by autoclaving at 121°C, 15 psi for 15 min.

Comparisons were made between the use of laboratory-prepared Staphylococcus-lIO medium (S-lIO), ~Iannitol-Salt Agar (M.S.A), Chapman's Stone Agar (C.S), which was prepared according to Difco Manual, and commercial Baird-Parker medium (B-P), supplied by bio-Merieux-France (Lot. No. 01203) as nutrient sources for the survivors of bacterial cells on membranes. The same four media were also prepared in liquid form.

The technique for the membrane filter procedure was as follows: After quick mixing of the test organism through the vehicle under investigation and at various intervals, samples were withdrawn and sterile dilutions were prepared in sterile peptone water. Membranes (cellulose nitrate, MS 11460, 0,45/lm pore size, Sartorius-Membrane Filter GmbH G6ttingen, FRG) were inoculated with 1 ml of the appropriate dilution of the inoculated vehicle. The inoculum was suspended in 10 ml of potassium phosphate buffer (0.05 M) at pH 7.0 for regular distribution of bacterial cells on the membrane surface. The fluid was drawn-through by vacuum and the mem­brane was immediately transferred to the surface of the tested solid media or on absorbent pads (supplied by Millipore Corporation, Bedford, Massachusetts, U.S.A) saturated by liquid form of the same media. Each sample was, therefore, plated in duplicate repeatedly on both solid and liquid media, using the four media under investigation.

In the case of using raw Nile water and unsterilized milk, duplicate plates from diluted non­inoculated samples were prepared as control.

Incubation was carried out at 37°C for Ml h. Colony counts were made with binocular scope, employing 10 to 15 magnification.

Results and Discussion

The wide-spread use of the membrane filtration technique has confirmed its value, especially its high degree of producibility and its ability to yield definite results. The membrane filter method excludes possible errors, common with the plate method, such as incomplete mixing of the tested samples with agar, or putting the sample onto hot agar. To avoid wrong results, originating from unsuitable medium, four selective media designed for recovery of Staphylococcu8 aureU8 have been evaluated for their suitability in solid or liquid form with the membrane filter technique.

Fig. 1 shows that Staphylococcu8 aureU8 enumeration from inoculated raw Nile water on Baird-Parker (B-P) medium gave significantly higher recoveries than the other three media. The descending order of efficiency in tracing the presence of the coagulase-positive staphylococci was: Baird-Parker (B-P), Mannitol-Salt Agar (:W.S.A),

Examination of 'Vater and Milk for Staphylococcu8 aureU8 381

Staphylococcus-110 (S-110), and Chapman-Stone medium (C.S). The efficiency of (C.S) was only 12.7 % compared to (B-P). The drastic decrease in count is progressive with time. On the 8th day the effectiveness of the (C.S) mediulll was reduced to 10 %, compared to the (B-P) medium.

The evaluation of the same four media in liquid forlll, using saturated absorbent pads showed another trend. 8. aureus when inoculated in raw Nile water, was not able to grow in colonies on the (B-P) medium. At the beginning both (S-110) and (M.S.A) medium were similarly effective (Fig. 1). The efficiency of the (C.S) medium was only 16.4 % compared to (S-llO). Considerable differences in the counts on (S-llO) and (M.S.A) were attained. Tests comparing the solid and liquid forms of the same nutrient medium (M.S.A or C.S) showed that counts for the former were higher than for the latter. The descending order of effectiveness for liquid media was (S-110) -(M.S.A) - (C.S). Previous studies by DUTKA et al. (1974) demonstrated significant reduction in the coliform counts and colony size, associated with the use of absorbent in comparison to the same medium prepared with 1.5 % agar base. GELDREICH (1976) added that vapour blockage and uneven flow might cause changing in nutrient concentrations in the pad substrate, ultimately affecting bacterial growth. Only few colonies of staphylococci were detected on membranes, testing the raw Nile water without inoculation. The characteristics of S. aureus colonies on the three liquid media, namely M.S.A, S-110, and C.S, seemed less definite, smaller, and more difficult to count than on solid media.

In both the sterile and raw Nile water, solid media tended to show similar patterns of descending order for S. aureus recovery (Fig. 2). Generally, it was observed that the drop in counts was more drastic in non-sterilized Nile water. On the 8th day, the percentages of recovery from sterile Nile water were 5.8, 4.5, 5.5, and 3.4 for (B-P), (M.8A), (S-110), and (C.S), respectively.

Fig. 2 shows S. aureus counts, determined by membrane filter technique on selective media from inoculated sterile Nile water. It appears that the medium is particularly suitable in determining the bacterial growth. Data confirmed that the (B-P) medium in liquid form was inhibitory to the tested strain and failed to initiate significant multiplication. Progressive decrease in S. aureus numbers were noted, but the reduction was generally less sharp than in the non-sterilized Nile water. High water quality is generally considered unsuitable for the survival of many bac­teria because of the small concentration of critical nutrients (GELDREICH 1979).

The inoculation of S. aureu.s in the non-sterile reconstituted dry milk and its enumeration results, using four different solid selective media, are presented in Fig. 3. At the beginning of these experiments, the higher counts were attained by using B-P, while the lowest counts were attained on (C.S) medium, showing 3 log cycle reductions in count when compared by (E-P). It is noteworthy that both (S-110) and (M.S.A) are tending to show somewhat similar results and took an intermediate position between (B-P) and (C.S) media. Prolonged incubation for the inoculated milk at room temperature showed decreases in counts and significant differences in growth response between (M.S.A) and (S-110) (Fig. 3). The decline in nUlllhers may be due to the repressive effect of natural microflora of unsterilized milk which COlll­

pete with the inoculated strain. By the time, staphylococci could multiply rapidly in the mixed population, due to the comparative shortness of the generation time of this species and because of the lengthened lag phase of the other species. On the 8th day, it was possible to observe re-increase in 8. aureus numbers. The recovered numbers on (M.S.A) and (S-110) returned to the level of the original numbers and became similar again. The numbers of 8. aureU8 on the control plates ranged between 1O-102/ml.

382 M. M. SABBOl;R and H. T. EL,ZANFALY

107

106

E ~

Qo a.

'" ~ Qo -u

'" .J:l

'0 ~ Qo

.J:l E

" e .. = '0

103 (JI 0 ..J

102

101

Solid media ----

\ \.

. \~ . \ ". ,.

". I~ " I

I'

x 1\ ". I , I

" ... '.,1\ " \

)\'~ '. ~ ". \.

f> I \ ... \ \

.\. \\ '. I

I' ·· .. 1\ , \ '. \.

". \ \ . \

' . .I.' '. x \ . \

'. \,

;;'\\ . \

'. \

".1\ . I '. \ : I'

". \ \ '. \

\ . ". \ \

'. \

'. \ \ . \

". I I.

x

Liquid media

(Saturated filter pads)

o Baird-Parker medium (B. P.)

4 Mannitol·Salt Agar (M SA)

x Chapman Stone medium .1. (C S)

'. \ • Staphylococcus 110 '., medium (S 110) '. ,\

". \1 '. \1

..... ~\ ;'. \

" I . .\

.... \' ' .. \ x. \ '. \

'. 1

'. '\1 "." I . ·1

\\ '. '\1 , I

". I ' .. \ '. \ i '. 1-1

...... "\ \ . \ I ' .. \ .~ \1 " I ' .. I

'. \\ '. \ '. 'I '. \\ '. \

". '\\ '. I ' .. \ .... \ ~

..... \.

Time, day 5

Fig. 1. Comparison of four different media in solid and liquid forms for enumerating Staphylo. coccus aureU8 from raw Nile water.

Fig. 2. Comparison of four different llledia in solid ancl liquid forms for enumerating Staphylococ­cus ,wreus incculaterl In

sterile Nile water.

E ~

'" Co

." ~

~ u ." .0

'0 ~ .. .0 E ::> c ..

£ -'0

'" 0 ...J

Examination of Wnter and Milk for Staphylococcu8 aureU8 383

107

106

Solid media

\ \

\

\~ \ x, \ A

" \

\ , \ \

X

".\ \ ' \ . .. \ ..

\

\\ \

\

'\ x \

\'l:. \

-

o Baird-Parker medium(B,P,) A Mannitol Salt Agar ( M SA) x Chapman Stone medium (C S • Staphylococcus 110 medium

(S 110)

Liquid media

(Saturated filter pads)

I,

o

'\ \'

'\ \'

\~

x',

2

\'

\\ \\

\ ,

\\ 'x \ •

\\ " \ \

" \ '. \

x'" \'\ .', \

.... \.\

4 6

" \ " \

", \ ',\ ',~

x

8

Time, days

The performance of liquid media for S. aUl'eU8 re<.:overies from inoculated unsteril­ized milk are shown in Fig. 3. These results illustrated that (B-P) failed to deteet S. aUl'eU8 and showed wrong negative results. It is dear that both (M.S.A) and (S-llO) in solid form are better than the liquid form. On the other hand, (C.S) medium tends to show higher counts in the liquid form than the solid one. On the third day, less variabilities in numbers of S. aureU8 produced on the tested Illedia were ohserved. Continuous decrease in numbers was noticed till the 5th day. On the 8th day, an increase in num hers was ohserved and rea<.:hed the initial counts by (M.S.A). On (S-llO) 5-fold in<.:rease in counts was attained. The sequen<.:e of order changed to (S-llO)-(M.S.A), and finally (C.S). Less contaminants were ohserved with liquid media and also less 8. aureU8 counts appeared in the control samples, compared with t hose attained with solid media.

Based on the data presented in Fig. 4, whi<.:h represent the use of sterile reconsti­tuted dry milk as a vehicle, minor differences in connts are observed using the four

384 i\I. i\t SABBOUR and H. T. EL,ZAXFALY

Solid media

E 1010

4; I a. I'

'" '\ 4; , u \\ '" £)

'0 ~

" £)

E :l C

" :5 '0 CI

108

',6.

0 ...J

',x \ I \ \ I \ \ \ / i

107 I ' 'I x' I ~ I

\ I \ I

I I " " ~:

x

106

0 2 4 5 6

\

,x

B 0

Liquid media

(Saturated filter pads)

\

o Baird-Parker medium (B.P.) a Mannitol-Salt Agar (MSA) x Chapman Stone medium (C.S.) • Staphylococcus 110 medium

(5 110)

,

l I'

/ x'

t/ '\ I,

~ .\ t/ \ ' x, I, \6\ II

\ ' \ '"

\'~ I' . '

t! \ ' I', ,x

~" I'

\\ II 'i.

iI' '\ ,'I I' "

'\' 1/ : " 'J, " ":~ ... \' x

3 4 5 6 8

Time, days

Fig. 3. Comparison of four different media in solid and liquid forms for enumerating Staphylo­COCCU8 aureU8 from raw milk.

solid media. Multiplication proceeded at a high rate by the 1st day or during the first few hours. The use of (B-P) provides the higher counts, followed by (M.S.A) - (S-110), and finally (C.S) medium; the population increased by 5.4, 4, and 3 log cycles on the four media, respectively. Absence of the decrease in S. aureU8 counts, which was observed previously with unsterilized milk, may be explained on the basis that na­tural competitors were inhibited by sterilization process.

Of the four selective media examined in the liquid form for the enumeration of S. aureU8 from inoculated sterile milk, (S-110) is the best, followed by (M.S.A), and

v

'" .0

c­o

...J

Examination of Vf ater Hnd Milk for Staphylococcu8 aureU8 385

Solid media

I 1/

/ x / / : . /

I I: / :

i' /./

/ : f •

-/ " / .-

I /.-/ '

f .x

i /.-f :

f :

/

f,'

f: , ' A ,0:

/ l: , f,

/ /: . f:

/ f,:

f.­. 'I.:

Liquid media

(Saturated filter pads)

o Baird-Parker medium (B,P) .:. Mannitol·Salt Agar (M SA) x Chapman Stone m,edium (C S ) o Staphylococcus 110 medium

(S 110)

/'

I

// " //

I .

// x.

;

1' .,'

f .. I.. ,.'

3 " 5 6 7 8

Time, days

Fig. 4. Comparison of four (lifferent media in solid and liquid forms for enumerating Staphylo­coccus aureus inoculated in sterile milk.

finally (C.S) medium. On the other hand, as all cases studied, (B-P) in the liquid form failed to detect the presence of the inoculated strain. It may be pointed out that, although the results illustrate clearly that the liquid media are able to show higher productivity at the beginning of our experiments compared with the solid form, the counts at the 8th day showed the opposite picture (Fig. 4). According to the medium used, S. aureU8 showed an increase in numbers on the 8th day, reaching 4 log cycles with (S-llO), 3 cycles with (M.S.A), and only 2 cycles with (C.S) medium.

386 ::VI.lVI. SABBOUR and H. T. EL-ZANFALY, Examination of Water and Milk for S. aureU8

References

ALICO, R. K., and P ALENCHAR, C. A.: Staphylococcus aureus recoveries on varIOUS brands of membrane filters. Health Lab. Sci. 12 (1975), 341.

American Public Health Association: Standard Methods for the Examination of -Water and Wastewater. 10th ed. American Public Health Association Inc., New York 1955.

BucH, J. H.: New bacteriological technique for testing water and sewage. Water and Sewage Works 100 (1953), 151.

CLARK, H. F., and KABLER, P. 'V.: The membrane filter in water quality tests. Am. J. Pub. Health 42 (1952), 385.

- - and GELDREICH, E. E.: Advantages and limitations of the membrane filter procedure. Water and Sewage Works 104 (1957), 385.

DUTKA, B. J., JACKSOX, ~I. J., and BELL, J. B.: Comparison of autoclave and ethylene oxide sterilized membrane filters used in water quality studies. App!. Microbio!. 28 (1974), 474.

FAVERO, .M. S., and DRAKE, C. H.: The use of staphylococci as indicators of swimming pool pollution by the use of the membrane filter technique. Pub. Hlth. Reror's 79 (1964), 6l.

FELSENPEL'). {I. an,.! PUKKAKU, 'V. K.: Adaptation of the membrane filter technique to the re­covery of Vibrio comma from water samples. J. Bacteriol. 72 (1956), 869.

GELD REICH, E. E., CLARK, H. F., HUFF, C. B., and BEST, L. C.; Fecal coliform organism for the membrane filter technique. J.A.W.W.A. 57 (1965), 208. Performance variability of membrane filter procedures. Public Health Lab. 34 (1976), 100. Stressed organism recovery in membrane filter procedures. Lecture outline for use at the membrane filter technology workshop, May 6, 1979. Los Angeles, California (1979).

GOETZ, A., GILMAX, R. H., and RAWN, A.lVI.: Application of molecular filter membranes to specific problems in water analysis. J.A.W.W.A. 44 (1952), 471.

GRAVES, D. C., and SCHIPPER, 1. A.: Membrane filtration technique for isolation organisms from raw milk of normal udders. Appl. Microbiol. 14 (196(j), 535.

HAYS, G. L., and O'BRIEN, D. L.: Methods for microbiological survey of carbonated beverage canning lines. Proc. Soft Drink Technologists 1-G (1971).

LUESCHOW, L. A., and MACKENTHUM, K. M.: Detection and enumeration of iron bacteria in municipal water supplies. J.A.W.vV.A. 54 (1962),751.

NOBILE, J.: Use of membrane filter technique in microbiological control for the brewing industry. Appl. Microbiol. 15 (19G7), 736.

SLANETZ, L. VV., BENT, D. F., and BARTLEY, C. H.: Use of the membrane filter technique to enumerate enterococci in water. Pub. Health Rep. 70 (1955), 67.

TmrPKJN, R. B., vVEISER, H. H., and MALANEY, G. W.: Salmonella and Shigella organisms in untreated farm pond water. J.A.W.W.A. 55 (1963), 592.

vVJNTER, F. H., YORK, G. K., and EL-HAKHAL, H.: Quick counting method for estimating the number of viable microbes in food and food processing equipment. App!. Microbiology 22 (1971), 89.

\VOLOCHOW, H.: The membrane filter technique for estimating numbers of viable bacteria - some observed limitations with certain species. Appl. Microbiol. 6 (1958), 201.

Authors' addresses:

Dr. }I. M. SABBOFR, Assoc. Prof., Dairy Research Lab., and Dr. H. T. EL-ZANFALY, Prof. and Corresponding author, \Vater Pollution Control Lab., National Research Center, Dokki, Cairo. Egypt.