WO1994003632A1

WO1994003632A1 - Detection of microorganisms

Info

Publication number: WO1994003632A1
Application number: PCT/GB1993/001646
Authority: WO
Inventors: David Charles Cowell; Antony Alan Dowman; Stephen David Watkins
Original assignee: University Of The West Of England, Bristol
Priority date: 1992-08-05
Filing date: 1993-08-04
Publication date: 1994-02-17
Also published as: AU4721093A; GB9216582D0

Abstract

Catalase activity, particularly in microorganisms, is measured by (1) incubating a sample with a known amount of hydrogen peroxide for a known time; and then (2) determining residual hydrogen peroxide. The determination step may involve adding a fluorophenol (which may also serve to inhibit catalase activity) and peroxidase, so that fluoride is generated. This may then be measured using a fluoride ion selective electrode.

Description

Detection of Microorganisms

The present invention relates to a method and means for the detection of microorganisms. It particularly relates to microorganisms containing the enzyme catalase, and their quantitative determination.

A common problem in applied bacteriology is estimating the total numbers of living organisms or detecting the presence of specific organisms, rapidly. Traditional methods relying solely on cultural techniques are both time consuming and expensive and somewhat insensitive, requiring considerable amplification of the bacteria before they can be detected. There is, therefore, a considerable demand for an inexpensive, high sensitivity detection system able to detect changes from the normal bacterial load or to detect growth very early in selective amplification media. These criteria apply within many fields, e.g. the water/sewerage industries, medicine, the food industry and the defence industry with regard to biological warfare.

Catalase (EC 1.11.1.6) is an enzyme which catalyses the breakdown of hydrogen peroxide to water and oxygen. It occurs widely in plants, animals, and nearly all aerobic microorganisms. Wang et al. , Journal of Food Science 5J, (6) (1986) 1442-1444 have used a "Catalasemeter" (Bio-Engineering Group Ltd., New Haven, CT, USA) to determine microorganisms. The organism is applied to a membrane which is placed in hydrogen peroxide. Catalase activity produces oxygen bubbles which eventually cause the membrane to float, after a period dependant on the level of catalase activity. However, this technique is rather insensitive. Broadly, the present invention provides a method of detecting microorganisms by determining the catalase-type activity of a sample by allowing the sample to interact with a predetermined quantity of hydrogen peroxide for a predetermined time, and subsequently determining the quantity of hydrogen peroxide remaining. Preferably this determination is carried out by using the residual hydrogen peroxide to cause the production of a readily detectable species, which is then detected, preferably by means of a selectively sensitive electrode. Suitably, potentiometric measurements are effected by the kinetic mode or the fixed time mode. Thus the residual hydrogen peroxide may be used to oxidise a signal substrate to release a detectable signal ion. The oxidation may be catalysed by a peroxidase enzyme. The signal ion may be fluoride, which is detected using a fluoride ion selective electrode. Thus the signal substrate may be a compound containing one or more carbon-fluorine bonds susceptible to cleavage by hydrogen peroxide and peroxidase, e.g. a fluorophenol such as 4-fluorophenol or pentafluorophenol. We have discovered that a suitable fluorophenol signal substrate may also serve to inhibit the catalase.

Less preferred ways of determining hydrogen peroxide include amperometric and spectrophotometric analysis.

The sample for analysis may be (or contain) catalase-positive whole microorganisms, or catalase- active components thereof, e.g. lysed microorganisms. The method may be applied to determining the quantity of catalase-positive microorganisms in a sample. A sample may be pretreated to remove possible interfering species.

Typically the microorganism is allowed to convert the substrate, hydrogen peroxide, to oxygen and water over a controlled period of time under buffered conditions. At the end of this controlled period of time the remaining substrate is determined electrochemically by the peroxidase catalyzed rupture of carbon-fluorine bonds from an organo-fluoro compounds using a fluoride ion selective electrode. The microorganism may be presented in a variety of media e.g. water, buffer, growth media, or swab. In a second aspect the invention provides a kit of parts for carrying out a method of the invention, including means for providing a known quantity of hydrogen peroxide; and means for causing residual hydrogen peroxide to generate a signal (preferably quantitatively). The signal generating means may comprise an oxidisable signal substrate (e.g. an organofluorine compound ) and a signal ion detector system

(e.g. a fluoride ion selective electrode). Generally there will be a detector cell, comprising a vessel for holding the sample, the signal substrate, and peroxidase; a fluoride ion selective electrode and a reference electrode; and circuitry connecting the electrodes for providing an output dependent on oxidation of the signal substrate.

Some embodiments of the invention will now be described in more detail with reference to the accompanying drawings in which:

Fig. 1 is a representation of the reactions involved;

Fig. 2 is a graph of fluoride ion release rate versus catalase activity, showing variation in the detection limit for catalase activity with increasing incubation period of the primary reaction; Fig. 3 is a graph of fluoride ion release rate versus catalase activity, showing the effect of reducing concentrations of hydrogen peroxide (HP) on the sensitivity of the catalase determination;

Fig. 4 is a graph of fluoride ion release rate versus catalase activity, showing the effect of increasing horseradish peroxidase ( HRP ) activity on the sensitivity of the catalase determination;

Fig. 5 is a graph of fluoride ion release rate versus the concentration of colony forming units ( "cfu" ) illustrating the application of the catalase assay to the detection of Escher±ch±a col± ( intact cells ) ;

Fig. 6 is a graph similar to Fig. 5 but relating to the detection of Escher±ch±a col± as lysed cells;

Fig. 7 is a graph of fluoride ion release rate versus cfu concentration, showing the results of 10 sets of determinations of E.coli to demonstrate reproducibility;

Figs. 8 and 9 are graphs similar to Fig. 7 but relating to 5 sets of determinations of S.aureus and Proteus respectively;

Fig. 10 is a graph comparing results for three microorganisms; and

Fig. 11 is a schematic view of the apparatus employed for the determinations.

The microorganisms (whole cells or lysed cells) are suitably incubated in a citric acid (suitably 0.1M)- disodium orthophosphate (suitably 0.2M) buffer, pH 6.0. Other buffers and molarities may be used e.g. acetate, borate, or Tris. Generally the buffer is 0.05-0.5M. The pH may be varied from 5.0 to 8.5.

The substrate, hydrogen peroxide, is added to this incubation buffer to give a final concentration of 0.1 to 20 M in the reaction mixture. Following a fixed period of time, e.g. between 1 and 60 minutes, during which time the microorganism's intracellular catalase utilises the substrate, the catalase is inhibited with 4-fluorophenol (whose final concentration may be between 1 and 200 M in the reaction mixture). Other organo-fluoro compounds may be utilized e.g. pentafluorophenol.

The remaining hydrogen peroxide is then determined electrochemically by a fluoride ion selective electrode measuring the fluoride ion released from the catalytic cleavage of the carbon-fluorine bonds of the 4- fluorophenol ("4-FP"). The reactions are shown in Fig. 1.

In the examples, apparatus as shown in Fig. 11 was used. A combination fluoride ion-selective electrode 10 (Orion Model 9608, Orion UK, Forest Row, Sussex, UK) was used in conjunction with a BBC Master series microcomputer 12 (Acorn, Cambridge, UK) and a microprocessor (Cambridge Microprocessor Systems, Cambridge, UK) containing a 13-bit analogue to digital converter and high-impedance amplifier with a gain of 10. Fluoride ion measurements were carried out in a glass cell 14 having a water jacket 16, for maintaining the temperature of 30 +/- 0.1°C. The reagents are stirred by a stirrer 18. A saturated calomel reference electrode 20 is electrically coupled to the cell 14 via a tube 22 filled with saturated KC1. Fig 11 also shows schematically means for providing known amounts of hydrogen peroxide, fluorophenol and horseradish peroxodase, shown as respective vials 24,26,28.

A pH 6.0 citric acid - disodium hydrogen orthophoshate buffer was prepared by mixing appropriate volumes of 0.1 ol 1 citric acid and 0.2 mol 1 disodium orthophosphate. The basic isoenzyme of horseradish peroxidase (EC 1.11.1.7. ) was purchased from Sigma Chemical Co. Ltd. (Type IX; Poole, Dorset, UK) and Biozyme Laboratories Ltd. ( HRP 4B; Blaenavon, Gwent, UK); isoelectric focusing electrophoresis confirmed that these preparations were identical in their isoenzyme pattern. A working solution of the enzyme was prepared by dilution of a 200 U ml^-1 solution in the buffer and diluted to the required activity in the same buffer. Bovine liver catalase (EC 1.11.1.6), was also purchased from Sigma; a

1665 U ml^-1 solution in the buffer was prepared and diluted to the required activity in the same buffer. 4-

Fluorophenol was obtained from Aldrich (Gillingham,

Dorset, UK) as crystals of 99% purity which were further purified by sublimation and then dissolved in the buffer containing 4 //mol 1 sodium fluoride, ensuring a constant and stable electrode signal prior to enzymatically catalysed fluoride ion release from 4- fluorophenol by horseradish peroxidase. Hydrogen peroxide, standardised by iodometric titration, was prepared in distilled water. Working sodium fluoride standards of 10, 100 and 1000 μmol 1 were prepared by dilution of a 10 mmol 1 stock solution with the buffer.

Assay Procedures

Sequential Catalase Assay .

0.03 ml of catalase solution or microorganism suspension was incubated with 0.72ml citric-phosphate buffer pH 6.0 containing hydrogen peroxide (1.0 to 4.0 mmol 1 ) in the thermostatically controlled cell. After a fixed time incubation at 30°C, 0.72 ml of 4- fluorophenol (160 mmol l^-1) in the same buffer was added. 0.03 ml horseradish peroxidase solution ( 2 to 8 U ml^-1) was added to initiate the reaction. The reaction was followed for 200 s.

Prior to the analytical run, the fluoride ion- selective electrode was calibrated using 10, 100 and 1000 μmol 1 standard solutions of sodium fluoride. To ensure that there was no drift in the E_Q value (i.e., the sum of the constant potentials for the experimental system), it was corrected by the use of a single sodium fluoride standard solution (100 //mol 1 ) between assays. During the 200 s reaction time, ten voltage readings (mV) were taken every second and the averages were used to calculate the concentration of fluoride ions using the pre-determined Nernstian slope and the E_Q value. The data generated were displayed in real time as a plot of the fluoride ion concentration versus time and stored. The stored data were analyzed by linear regression to calculate the rate of the reaction where zero order kinetics are obeyed with regard to substrate concentration. The fluoride ion-selective electrode was re-calibrated every six assays.

M±crob±al suspens±ons and viable counts

The microorganisms, were grown in Nutrient Broth E (lab M, Preston, UK) overnight at 30°C in a shaking incubator at 200 rpm. One ml aliquots of the suspensions were centrifuged at 10,000g for 3 min. The pellet formed was resuspended in the same volume of sterile 2% w/v sodium chloride. This washing procedure was repeated. The washed cells were resuspended in sterile citrate- phosphate buffer pH 6.0, containing 4 .mol 1 sodium fluoride. All solutions had been previously sterilised by ultra-filtration through a 0.45 urn disposable filter (Sartorius, Epsom, UK).

The viable count of the prepared suspensions was determined using two techniques. A series of ten-fold dilutions of each suspension was made in the same sterile citrate-phosphate buffer containing sodium fluoride. From these, viable counts were made by a surface drop technique, derived from that described by Miles and Misra J. of Hygiene, 38_, 732-749, (1937), onto Nutrient Agar (Lab M, Preston, UK). Counts were also performed on the series of dilutions by a surface spread technique (Collins and Lyne' s Microbial Methods, London 1987, 131- 132). All plates were incubated at 37°C aerobically overnight. Cell disruption was achieved by sonication of 10 ml volumes of the washed buffered suspensions for three periods of 30 seconds with cooling on ice (MSE ultrasonic disintegrator, Fisons Scientific Equipment,

Loughborough, UK).

Sequential Catalase Assay

Catalase was allowed to utilise the common substrate, hydrogen peroxide, for a fixed time period followed by inhibition of its catalytic action by 4- fluorophenol. The remaining hydrogen peroxide was then assayed via the horseradish peroxidase cleavage of the carbon-fluorine bonds of the inhibitor, 4-fluorophenol. The hydrogen peroxide concentration adopted in the reaction cell was 2 mmol 1 . This concentration was chosen because we have found that at optimum conditions the release of fluoride ions diminished rapidly below a hydrogen peroxide concentration of 2 mmol 1 . Consequently, substrate utilisation during the primary reaction by catalase activity may be more sensitively assessed by the indicator reaction, the horseradish peroxidase cleavage of fluoride ion from 4-fluorophenol.

The limit of detection of this assay was determined by defining the point at which the fluoride ion released for an assay was statistically different from that of a blank assay, i.e. no catalase present. As a decrease in fluoride ion release was indicative of catalase activity, the blank assay gave the highest rate of fluoride ion release. It was necessary to reappraise the limit of detection whenever one of the assay parameters was varied during the development process. As an example the data for the limit of detection in Figure 2 are presented where the assay parameters were 4-fluorophenol 80 mmol 1 , hydrogen peroxide 2 mmol 1 , horseradish peroxidase activity 40 mU ml^"1 and temperature 30°C. A within batch reproducibility study of the blank assay (n=20) gave a mean of 5.35 //mol 1 s with a standard deviation ( SD) of 0.19 //mol l^-1 s . A 99% confidence limit was assigned to the detection limit i.e. mean -2.6 x SD. All assays with rates less than 4.86 //mol 1 s were thus deemed to be significantly different from the blank assay. It was apparent that the assay had three variables, incubation time, concentration of the common substrate hydrogen peroxide and the activity of the indicator enzyme, horseradish peroxidase, that would affect both the limit of detection and the sensitivity. These variables have been individually studied.

Catalase activity was studied between 3.33 x 10^~8 U ml^" and 3.33 U ml with varying incubation times for the primary reaction with the horseradish peroxidase activity at 40 mU ml^-1 (Fig. 2). The limit of detection for catalase activity decreased with increasing incubation time. After a 10 minutes incubation the detection limit was 1.0 x 10 U ml^-1; after 15 minutes

1.5 x 10~⁶ U ml^-1; after 20 minutes 3.0 x 10 ^~7 U ml^"1.

The sensitivity of the assay, defined as the rate of fluoride ion release per decade over the most linear part of the curve in the region of the limit of detection, was studied using a 15 minute incubation period, with the horseradish peroxidase activity at 80 U ml and varying the hydrogen peroxide concentration between 0.5 and 2.0 mmol 1 (Fig. 3). The sensitivities observed between 3.3 x 10^~7 and 3.3 x 10^"4 U ml^"1 of catalase were 0.17, 0.76 and 1.03 //mol 1 s for hydrogen peroxide concentration of 0.5, 1.0 and 2.0 mmol 1 respectively. Not only was the sensitivity of the assay greatest at 2.0 mmol 1 hydrogen peroxide but the calibration curve was no longer bimodal; the useful analytical range covering all activities of catalase studied.

Investigation of the activity of the horseradish peroxidase in the indicator reaction (Fig. 4) showed that it had a significant effect on the sensitivity of the assay. The rates of fluoride ion release per decade over the most linear part of the curve in the region of the detection limit (3.3 x 10^~6 to 3.3 x 10^~2 U ml^"1 of catalase) were 1.26, 1.90 and 2.42 //mol l^"1 s^"1 for horseradish peroxidase activity of 80, 120 and 160 mU ml^" respectively. The highest activity of peroxidase produced the curve with the largest analytical range; 3.3 x 10~⁷ to 3.3 x 10^"2 U ml^"1.

It is thus apparent that by the judicious variation of three parameters, the length of incubation, the concentration of the primary substrate and the indicator enzyme activity, both the detection limit and the sensitivity can be altered to suit the requirements of the assay.

Application of the catalase assay to the detection of Escherichia coli.

A fifteen minute incubation time, coupled with the optimum assay conditions of hydrogen peroxide 2.0 mmol 1 and horseradish peroxidase activity of 160 mU ml^"1, were used to demonstrate the use of the assay for the detection of the aerobic microorganism Escherichia coli (Fig. 5, curve C). The number of microorganisms present were expressed as colony forming units (cfu) present in the reaction mixture. The resulting curve had a useful analytical range covering 2.8 x 10 to 2.8 x 10 cfu ml^" , a sensitivity of 3.26 //mol 1 s per decade and a limit of detection of 1 x 10 cfu ml ^**** . The combination of sub-optimal assay conditions produced little effect on the limit of detection but considerable reductions in sensitivity (Fig. 5); curve A 0.27 /mol l^"1 s^"1 per decade and curve B 0.73 //mol l^"1 s^s_1 per decade.

The limit of detection achieved represents a considerable advance on previously published methods.

The limit of detection appeared to be dictated by the length of the incubation of the primary reaction.

However, the rate of diffusion of the substrate, hydrogen peroxide through the microbial cell wall must be a significant limiting factor to the further reduction in the detection limit. A repetition of the previous experiment with lysed microorganisms (Fig. 6)

_ ^*1 demonstrated the detection of less than 10 cfu ml . This result was achieved at non-optimal assay conditions of substrate concentration and indicator enzyme activity indicating that it may be possible to reduce the incubation period without having a detrimental effect on the limit of detection of the assay.

Similar studies with the catalase negative organism, Streptococcus sanguis (ATCC10556), showed no variation from the blank assay up to a concentration of lθ" cfu π_] ^~ This indicated that catalase was the intracellular enzyme being assayed.

To determine the reproducibility of the method, the E.coli was assayed repeatedly under the same conditions, substantially as used to produced curve C of Fig. 5. The results are shown in Fig. 7. It can be seen that the reproducibility is remarkably good. Similar experiments were carried out using S.aureus and Proteus ( 5 repetitions each) and the results are shown in Figs. 8 and 9 respectively. Fig. 10 shows the curves obtained for the three types of microorganism. The curves differ, but not greatly. This indicates that, for many practical purposes, the method of the invention can be used to measure the concentrations of microorganisms of unknown and/or mixed types. Detection limits and sensitivities for the organisms are presented in Table 1.

The work presented in this paper has demonstrated the possibilities of this versatile generic detection system for aerobic microorganism . The use of a solid state sensor, the fluoride ion selective electrode, opens up possibilities for the development of both automated instrumentation and disposable biosensors for the rapid potentiometric detection of microorganisms. TABLE 1

Claims

1. A method of determining the catalase activity in a sample comprising incubating the sample with a known amount of hydrogen peroxide for a predetermined time and then determining a value related to the amount of hydrogen peroxide remaining, said value being indicative of the catalase activity in the sample.

2. A method of determining the quantity of catalase-containing microorganisms in a sample comprising incubating the sample with a known amount of hydrogen peroxide for a predetermined time and then determining a value related to the amount of hydrogen peroxide remaining, said value being indicative of the quantity of catalase-containing microorganisms in the sample.

3. A method according to claim 2 wherein the microorganisms are lysed prior to incubation.

4. A method according to any preceding claim wherein, after said predetermined time, catalase activity in the sample is inhibited by addition of a fluorophenol.

5. A method according to any preceding claim wherein said value determination is effected by adding a signal substrate which is oxidisable by hydrogen peroxide and thereafter monitoring the oxidation of the signal substrate.

6. A method according to claim 5 wherein said signal substrate is a compound containing at least one C- F bond which is oxidatively rupturable with release of fluoride, and this is determined electrochemically.

7. A method according to claim 6 wherein a peroxidase is also added, and the signal substrate is one which undergoes peroxidase-catalysed C-F rupture.

8. A method according to claim 6 or 7 wherein the signal substrate is a fluorophenol which also acts to inhibit catalase activity.

9. A method according to claim 6, 7 or 8 wherein the fluoride is detected by means of an ion-selective electrode.

10. A method according to any of claims 6 to 9 wherein rate of release of fluoride is determined and serves as said value.

11. A method according to any preceding claim wherein a calibration curve relating said value to catalase activity or microorganism quantity is recorded, and then an unknown sample is assayed by determining the value for it and comparing this with the calibration curve.

12. A kit of parts for carrying out a method according to any preceding claim comprising: a vessel for holding a sample; means for providing a known quantity of hydrogen peroxide in the vessel for interaction with the substrate; and means for causing residual hydrogen peroxide to generate a signal whose value is related to the amount of hydrogen peroxide remaining.

13. A kit according to claim 12 wherein the signal generating means comprises a system which generates fluoride on reaction with hydrogen peroxide, a fluoride- sensitive electrode, a reference electrode, and circuitry for connecting said electrodes and providing an output dependent on said generation of fluoride.