The present invention relates to a method for detecting and assaying microorganisms, to agents for use in such a method, and to test kits comprising essential reagents for carrying out the method-

All living organisms utilise adenosine triphosphate (ATP) as a source of chemical energy and it is known to assay this using the ATP driven luciferase/luciferin reaction. Light generated by this enzymic reaction can be measured using a luminometer and related to the amount of ATP present. The usefulness of ATP as an index of microbial numbers has been known since the mid 196θ's (see ATP Luminescence Rapid Methods in Microbiology (1989) editor Stanley et al.; Blackwell Scientific Publications, London, see pages 1-10); its main advantage being speed and sensitivity. Utilising this assay format simple samples can be analysed in a matter of minutes while complex ones routinely take only half an hour with a detection capability provided down to 10" ¹² mol/1 ATP. There is however a need for methods which provide still further sensitivity when detecting microorganisms or their contents while retaining speed and ease of performance.

The present inventor has now determined that the speed and sensitivity of ATP based method can be enhanced significantly by shifting the target of the assay from ATP to an enzyme which generates it, particularly to adenylate kinase. Adenylate kinase is an enzyme used by all organisms for the conversion of adenosine diphosphate (ADP) to adenosine triphosphate (ATP). The targeting of this enzyme in preference to ATP, by using the preferred method, reagents and kits of the invention, allows the detection of down to at least 10 ^"20 moles of intracellular marker adenylate kinase.

It is known to assay adenylate kinase using the luciferase/luciferin system (see Brolin et al Journal of Biochemical and Biophysical Methods 1 (1979) 163-169 and Shutenko et al. Biotekhnologiya, No 4,

(1988) 542-547) for the purpose of determining its activity and this has been applied to study of certain mammalian and plant tissues (eg. see Rodionova et al Fiziologiya Rastenii (1978) 25, 4, P73 34). The use of such assay system for the detection and assay of microorganisms however has not been suggested and the advantages of doing such, ie. enhanced sensitivity so provided, have not been relevant to those studying the enzyme itself.

Although adenylate kinase is present in smaller quantities than ADP or ATP, its use as a biological marker for microorganisms provides enhanced sensitivity with a typical amplification available of 400,000 by measuring its presence through the ATP it produces; that is for every mole of enzyme present 400,000 moles of ADP are converted to ATP in a 10 minute incubation. Thus estimation of the enzyme by measuring the substrate or product of the reaction it catalyses provides for detection down to as low as 10 ^"20 moles.

The applicant's copending PCT application PCT/GB94/00118 relates to the general method of estimation of microorganisms in a sample from the sample's ability to convert ADP to ATP, and relating that to the presence of microorganisms or their intracellular materials. That application exemplifies the method wherein magnesium ions, necessary for the reaction of two molecules of ADP with each adenylate kinase active site, are not added as a reagent but are provided by any bacteria cells present and as an impurity in the other reagents. The number of cells detected in the examples using such technique proved to be about 10 ² , with results of a more statistically valid nature being obtained at 10 ³ or more; a linear relationship between luminometry counts and cell numbers then being obtainable.

The present invention relates to an improved technique in which adenylate kinase activity is measured in an optimised fashion using reaction conditions in which magnesium ions are supplied to the ADP conversion reaction, and wherein the reagents used are treated to

remove adenylate kinase to higher degrees of purity whereby the number of microorganisms that can be detected is of the order of tens rather than hundreds per 200μl sample, and readings with linear relation between cells and ATP derived light are possible down to 10 cells.

In a first aspect of the present invention there is provided a method for determining the presence and/or amount of microorganisms and/or their intracellular material present in a sample characterised in that the amount of adenylate kinase in the sample is estimated by mixing it with adenosine diphosphate (ADP), determining the amount of adenosine triphosphate (ATP) produced by the sample from this ADP, and relating the amount of ATP so produced to the presence/or amount of adenylate kinase and to microorganisms and/or their intracellular material, wherein the conversion of ADP to ATP is carried out in the presence of magnesium ions at a molar concentration sufficient to allow maximal conversion of ADP to ATP. The amount of magnesium present is preferably such that there is sufficient to provide one mole of magnesium for one mole of ADP such that all of the ADP molecules may be associated with at least one magnesium ion.

In preferred embodiments of this aspect of the invention the sample is provided in the form of an aqueous suspension or solution and the estimation of adenylate kinase, and thus microorganisms and/or intracellular material in the sample, is carried out by adding ADP and magnesium ions to the sample under conditions whereby any adenylate kinase present will convert ADP to ATP, incubating the sample for a predetermined period to effect such conversion, adding luciferase and luciferin agents, determining the amount of light emitted from the sample and relating that to presence and amount of adenylate kinase.

The amount of ADP with which the sample is mixed is preferably sufficient to provide an ADP concentration in the mixture in excess of 0.005mM, more preferably in excess of O.OlmM and most preferably in excess of O.OδmM. A particularly preferred amount of ADP in the

conversion step mixture is about O.lmM.

Where reagents are to be used which contain magnesium ion depleting agents, eg. chelating/sequestering agents such as EDTA and phosphate buffers, it will be realised that in order to provide the ADP with sufficient magnesium ions for it to undergo optimal conversion, it will be preferred for an excess of magnesium ions to be present. For the preferred concentrations of ADP set out above, the preferred concentration of magnesium ions in the suspension or solution during conversion of ADP to ATP is ImM or more, more preferably 5mM or more and most preferably lOmM or more. The magnesium ions may be provided in the form of any magnesium salt, preferably as magnesium acetate.

A further preferred format of the present invention adds luciferin /luciferase luminometry reagents to the sample at the beginning of the incubation, preferably as a single reagent with the ADP and magnesium ion source. Lucifease is preferably stored separately from extractant.

In formats of the invention where all the reagents are included at the start of the conversion of ADP to ATP in this manner, and/or where luminometer counting is continued after luciferin/luciferase addition where that is a separate step, magnesium may be provided by the luciferin/luciferase reagent. However, due to binding of magnesium ions by EDTA and phosphate it is necessary that the amount of magnesium ions is positively ensured by prior experiment or calculation. It will be realised by those skilled in the art that the optimal amount of magnesium salt to be added to a given ADP, sample and luciferin /luciferase mixture will be readily deteπninable by routine experiment using a sample containing a known amount of bacteria, eg. E. coli. whereby maximal signals are obtained. Figure 3 below gives an indication of the optimal amount of magnesium acetate to be added to a mixture as used in the Examples below.

As Mg ^2* ions facilitate ADP depletion due to contaminant adenylate

kinase, it is preferred not to keep them in solution together prior to use; chelating agent such as EDTA may be included in the ADP to prevent this. Preferably the magnesium and ADP are brought together just prior to use or in the ADP conversion step. Where the reagents are to be kept together it is preferred that they are kept in freeze dried form to avoid any premature ADP conversion to ATP.

As stated above, although other assays may be used, ATP is preferably detected by use of the luciferin/luciferase system to provide a photometrically detectable signal indicative of the amount of ATP in the sample. Luciferin/luciferase preparations and methods for their use in assaying ATP will be well known to those skilled in the art and are commercially available (eg. see Brolin et al). A typical formulation contains eg. 0.1 to lOmg/litre luciferase, 15 to lOOOμmol/litre D-luciferin, and agents such as MgCl ₂ (2.5 ^_ 25mmole), EDTA, BSA. and pH7 buffer (see eg. EP 054676).

For single reagent use with adenylate kinase testing methods as described herein it is preferred that the pH is adjusted to that which is optimal for both enzymes, ie. a compromise, in order that counting might continue while converting ADP to ATP. This may be determined by routine experiment using known bacterial numbers in a sample.

The sample, ADP and magnesium ion source may be mixed in any buffer providing a pH suitable for the adenylate kinase reaction; no other reagents are necessary. Thus any buffer providing a pH of between 5»5 and 8.5 might be used, with optimal pH lying between pH6 and 7. preferably pH6.5- Examples of suitable buffers include Tris and phosphate buffers. Most suitably the sample is collected and/or diluted in such a buffer in preparation for carrying out the method of the invention.

As with any amplified assay, the sensitivity of the adenylate kinase assay of the present invention is limited by the purity of the

reagents. In this case the significant contaminants are ATP in the ADP substrate and adenylate kinase in the luciferase preparation. For use as a sensitive assay for microorganisms, particularly where these may be potentially harmful and need detecting in low numbers, it is necessary that the purity of each of the reagents be as high as possible with respect to the substance with which it is to react in the assay.

To address the first problem, high purity commercial ADP (>99-5% purity) is preferably used after further purification by column chromatography. This is desirable because even small amounts of contaminating ATP may be sufficient to cause a high background reading. For example, using a diethylaminoethylcellulose column and 0.02mM hydrochloric acid eluent, ATP is eluted more slowly from the column than ADP to a degree enabling substantial separation. Other chromatographic media and eluent combinations may also be used to similar effect, for example HPLC using Nucleosil column packings (available from Technicol, Stockport Cheshire UK), such as Nucleosil 3 and Nucleosil 5, using 0.06M KH ₂ P0 _ή :methanol in 77:23 ratio v/v at pH6 with 5mM tetrabutylammonium hydrogen sulphate. Fractions with high ADP to ATP ratios are retained for use and purity is assessed by luciferin/luciferase reagent mediated bioluminescence after adenylate kinase action to measure ADP levels and without adenylate kinase to measure ATP contaminant levels.

Using a preferred "Econopaq Q strong anion exchange gel cartridge (Biorad) equilibrated with 20mM potassium phosphate at pH4.6 and eluting with steps of KP. concentration up to 400mM, ADP was found to be strongly retained and eluted as a coherent peak, with ATP eluting after it. In this manner ADP with a molar % ATP upper limit of 2 x 10 ^"8 was obtainable. The most pure ADP the applicants are aware of from the literature is 0.001. (see Shutenko et al, as above) thus the present invention provides ADP for use in the method of the present invention that has less than 0.001 molar % ATP, more preferably 2 x 10 ^"8 molar % or less.

A further method for removing ATP from the ADP substrate uses enzymes that specifically degrade ATP, such as luciferase or apyrase. Such enzymes may also be used to further purify chromatographically purified ADP, or alternatively enzymically purified ADP may be treated by column chromatography. It will be noted that apyrase is also an ADPase, but as some are more active on ATP and the ADP is present at much higher levels this does not present a significant problem.

With regard to the second problem, adenylate kinase, as an essential "housekeeping" enzyme, is present in virtually all organisms and is generally present in luciferase preparations. It may only be a minor contaminant, but since the aim is to measure very low adenylate kinase levels in samples, its presence in the luciferase may be a limiting factor. In fact the applicant has determined that, defining one unit (U) of activity as the amount of enzyme which converts lμmol ADP to lμmol ATP per minute in the presence of 0.5mM ADP and 4.5mM Mg ^2* at pH7.8 at 20°C, commercial luciferases may contain 10 ^"7 U/ml or more adenylate kinase activity whereas luciferin, its substrate, comprises very little if any activity. Furthermore it is common to stabilise luciferase reagents with a stabiliser, commonly a protein such as bovine serum albumin (BSA) , and commercial preparations of this have been determined by the applicant to possess significant adenylate kinase activity.

The molecular weights of luciferase and adenylate kinase are significantly different, being 6lkD and 21kD respectively. Furthermore luciferase is a membrane bound protein and therefore relatively hydrophobic, whereas adenylate kinase occurs as a soluble enzyme. It is thus possible to remove adenylate kinase from luciferase preparations by, eg. size exclusion chromatography, reverse phase chromatography, or both. Alternatively or in addition to this, the problem of adenylate kinase contamination of luciferase can be avoided by adding the bioluminescent reagents (luciferase and luciferin) just before or as measurements are taken so that any contaminating

adenylate kinase does not have the time to produce a significant effect.

Suitable methods for purifying luciferase use column chromatography fractionation with a low porosity gel, eg Sephadex G-25 (see Nielsen and Rasmussen, Acta Chemica Scandinavica 22 (1968) pl757~1762; use of Sephadex and Sepharose columns (eg Blue Sepharose) in series and/or SDS electrophoresis (see Devine et al, Biochimica et Biophysica Acta 1172 (1993) 121-132) or ageing for a period at elevated ambient temperature.

In order to remove adenylate kinase activity from agents such as bovine serum albumin it is similarly possible to use column chromatography. A further treatment that has proved successful in this regard is chemical treatment of the BSA such that its ability to stabilise the luciferase is retained, but adenylate kinase activity is reduced or depleted altogether. Any conventional chemical treatment for the depletion of enzymic activity from proteins may be equally be applied for this purpose. Alternatively a non-protein luciferase stabiliser, eg. glycerol, may be used as a supplement or replacement for the BSA.

For example, the applicant has determined that commercially available BSA can have its adenylate kinase activity reduced to less than 2% of its original activity or less merely by heat treatment at acid or alkaline pH. One suitably effective treatment heats the BSA at pH5.6 or pHIO at 50°C for 24 hours. A further source of adenylate kinase free BSA is the chemically treated reagent acetylated-BSA, available from Sigma and BDH. It will be realised by those skilled in the art that other chemically treated BSAs will also be suitable.

In order to render all the adenylate kinase associated with a target microorganism available to the ADP, magnesium ions and luciferase /luciferin assay reagents of the invention it will be necessary to disrupt them such that intracellular material is released or otherwise exposed to the reagents. Such disruption might be carried out using

mechanical means such as an ultrasonic generator, by use of osmotic shock optionally in association with cold shock or such agents as lysozyme or, more conveniently, by use of detergents. Such detergents are commercially available and commonly referred to as 'extractants' . Typical extractants include generic cationic detergents such as CTAB (cetyl trimethyl ammonium bromide) , and proprietory agents such as Biotrace XM extractant (available from Biotrace, Bridgend UK), Celcis UK cationic extractants and Lumac NRM (nucleotide releasing agent available from Lumac BV, Holland). When using CTAB a convenient preparation will include 0.01 to 1% CTAB in water, eg 0.2% , but other concentrations may occur to those skilled in the art.

Thus before adding ADP and luciferase/luciferin reagent(s) to an assay sample suspected of containing microorganisms it is preferred to disrupt these to render their intracellular contents accessible to luminometry reagents by use of disrupting agent. If it is desired to distinguish between target cells and cells such as those of fungal spores it is possible to run two separate assays treating one with a nonionic detergent capable of disrupting only these spores and multi -cellular 'somatic' animal cells (eg.Triton TX-100) and the other with cationic detergent 'extractants' detailed above for disrupting all cells. It is possible to carry out these assays on the same sample if an ATPase such as apyrase is added between detergent/luciferase /measurement cycles; one cycle using nonionic and the other cationic detergent in a first cycle step with filtration steps between.

The effect of extractant upon the luciferase/luciferin system is known to be important (see eg. Simpson et al (1991) J« Biolumin Chemilumin 6(2) pp97-106) ; with cationic detergents being known to potentiate the reaction but to cause gradual inactivation of luciferase, anionic detergent inhibiting the reaction and nonionic and zwitterionic detergents being known to potentiate over a wide range. A mixture of 0.155- cationic detergent together with 0.25% tertiary dia ine surfactant (obtained from Celcis, Cambridge, UK) was found to be

satisfactory for present purposes, but those skilled in the art will have no problem screening for other 'extractants' that yield an optimal mix of adenylate kinase and luciferase activity when copresent in the same solution.

The light given off from the mixture after all the essential steps are complete, ie. ADP conversion to ATP and subsequent action of luciferase upon luciferin, may be measured by residence of the sample volume, eg. luminometer tube, within a light detector immediately after or simultaneously with addition of the luciferase and luciferin or other agents which enable the essential steps to proceed.

In a second aspect of the present invention there is provided a test kit comprising the essential reagents required for the method of the invention, ie. adenosine diphosphate, a source of magnesium ions and preferably luciferase and luciferin. Preferably the kit includes all these reagents, with the luciferase and luciferin being provided as a single reagent solution, with a detergent reagent in the kit suitable for disrupting the target cells for which the assay is intended. Usually for assaying microorganisms only cationic detergent is needed, whereas if fungal spores and somatic cells are likely to be significant then a further nonionic detergent reagent might be included to assess their numbers. The kit is in the form of a single package preferably including instructions as to how to perform the method of the invention; the reagents being provided in containers and being of strength suitable for direct use or after dilution.

It may be appropriate to provide the magnesium ions with the luciferase/luciferin reagent if this is to be added before the ADP to ATP coversion has begun, but then they should be in excess over that bound to the EDTA or phosphate in that reagent and should be optimised to accomodate both adenylate kinase and luciferase requirements. For microbial detection magnesium ions are preferably provided with a sample collection/dilution buffer but other formats may be preferred

for particular applications. Most conveniently the magnesium ions are provided with a sample collection or dilution buffer, the ADP is provided with detergent/surfactant extractants and optionally with a stabiliser such as EDTA, and the luciferase and luciferin are provided together, thus providing a three reagent test kit. Alternatively these agents may be provided in the form of a single reagent that is freeze dried such that they do not interact to cause degradation of eg. the ADP, prior to use.

A preferred test kit of the invention comprises ADP reagent which is of purity higher than 99-9992t and a luciferase/luciferin reagent, including BSA, that is substantially free of adenylate kinase activity. Alternatively the luciferase/luciferin ratio used, reflected in the kit instructions for use and/or in their relative concentrations, is such that the luciferase is capable of acting upon the luciferin substrate sufficiently quickly such that any luciferase associated adenylate kinase produces ATP after the initial emission is finished; thus microorganism derived adenylate kinase will be indicated by a flash kinetic reaction and contaminant ATP by a glow.

The preferred purified reagents may be provided by the methods described above. It is noted that adenylate kinase activity in luciferase may also be depleted by leaving the luciferase to stand for a period of months or years.

The methods, apparatus, reagents and kits of the present invention will now be illustrated by way of example only with reference to the following non-limiting Examples and Figures. Further embodiments of the invention will occur to those skilled in the art in the light of these.

FIGURES

Figure 1: plots log luminometer signal against log number of E. coli

in a 200 μl sample using the improved assay of the invention using 1 and 5 minute incubations prior to luciferin/luciferase addition.

Figure 2: plots log luminometer signal against log number of cells of E. coli in absence of magnesium.

Figure 3- shows the effect of magnesium ion concentration upon the luminometer signal derived from a set number of P. aeruginos at pH 7-5 and pH 8.0 showing the 10 fold increase over no added magnesium.

Example 1: Preparation of purified adenosine diphosphate reagen . Liquid chromatography was used to further purify commercial high purity (>99.95#) ADP (Sigma) using a 5ml Econopac Q cartridge (Biorad) equilibrated with 20mM potassium phosphate pϋ4.6 and loaded with ml of the ImM ADP (2.1 mg) . Elution was carried out by steps of KP _i concentration up to 400mM whereupon the ADP was strongly retained and eluted as a peak at approximately 340mM KP _j . Setting the system up with a pump (5ml/min) and gradient mixer, a gradient of 50 to 1M KP ₄ in 200ml total was provided and ml fractions collected. ADP eluted as a sharp peak between fractions 12 and 17 with ATP beginning to appear at the end of the gradient. A step in [KP _j ] to 1 M eluted the remaining ATP. The purest ADP fractions from this column were of less than 2x 10 ^"8 mole % ATP.

Example 2: Preparation of adenylate kinase free luciferase reagents. Adenylate kinase activity was deleted from commercially available luciferin/luciferase reagents (Biotrace HM) by ageing, including several months at high ambient temperature (circa 30°C) over a period of 12 months in dry form.

Example : Alternative preparation of kinase free luciferase reagents. Commercially available luciferase is purified using column chromatography by the method of Devine et al (1993) using Blue Sepharose as referred to above.

Example 4: Preparation of adenylate kinase free BSA. Sigma Fraction V (RIA Grade, Cat. No. A-7888) BSA was made up at 1% weight/volume in 200ml sterile water to give a starting pH of 5.6. Two 50ml samples of this were put into 100ml Duran bottles, the remainder made to pHIO using 5M NaOH and 50ml put into each of two Durans. Thimerosal was added to 0.02% final concentration as a preservative to prevent microbial growth and the bottles incubated at 37°C or 50°C for 24 hours before the pH of each was readjusted to 7.6 with M HC1 or 5M NaOH as appropriate. Adenylate kinase activity was measured by mixing lOOμl BSA sample as prepared above with lOOμl 30mM magnesium acetate solution, placing the resultant mixture in a 3-5ml luminometer tube in a luminometer and adding lOOμl ADP solution prepared in Example 1 and lOOμl luciferin /luciferase reagent (Celcis, Cambridge UK) that was prepared free of adenylate kinase activity by use of column chromatography and chemically treated BSA. After a 5 second delay light emission integrated over 10 seconds was measured and recorded on a computer and a total of 10 sequential 10 second readings were made to determine the rate of ATP production; analyses being performed in duplicate. Calibration was made using 4 replicate measurements of the light emitted from 5ul of lOng/ml (91 femtomoles) of ATP in water: mean signal 2950 per femtomole.

Results: The BSA sample incubated at 37°C remained clear whilst those at 50°C formed a precipitate which was slight at pHIO and very heavy at pH5.6. At pHIO and 50°C there was slight discolouration. Adenylate kinase activity remaining in these samples is shown below in Table 1 below as represented by luminometer counts per minute.

It is recommended that still milder forms of inactivation are used, with longer duration, or that the BSA is immediately freeze dried, if it is intended to store it for any length of time as after 2 weeks even the pHIO 50°C sample became unuseable due to increased discolouration. The 37°C samples did not go off in this manner and thus offer better scope for reducing stable adenylate kinase free BSA

by increasing the incubation time. The fact that the Biotrace HM agent lost its activity in dry form after storage at 4θ°C demonstrates the possibilities here.

TABLE 1: d[ATP]/dt

Treatment Counts t5-15 t95-105 Difference (fm.sec ^"1 )

37/5.6 9350 41727 23207 7.1

9845 26041 (means)

11896 32945

37/10 7192 30602 17943 5-5 5047 20557 4469 19377

50/5-6 606 1191 595 0.18 343 948

50/10 460 1014 500 0.15 342 847 314 754

Example : Preparation of luciferin/luciferase reagent with BSA. Commercial preparations of luciferin/luciferase commonly contain BSA as necessary. BSA chemically treated as set out in Example 4 above or as commercially available as acetylated BSA (eg BDH or Sigma) was admixed with adenylate kinase free luciferase in normal proportions together with other standard Celcis agents such as to provide a Celcis LDR luciferin/luciferase luminescence reagent of adenylate kinase activity less than 10 ^"9 U assay volume (ie. 300μl).

Example 6: Test kit of tha invention.

A test kit of the invention is provided consisting of the following: (i) a container holding 15mM magnesium acetate solution for collection /dilution of samples;

(ii) a container holding purified ADP solution (>99-99999998% pure with regard to ATP) prepared as described in Example 1 in a concentration of 0.3mM in potassium phosphate (7.5mM pH6.5) buffer solution further including 0.2mM EDTA and a mixed extractant of 0.15% cationic detergent and 0.252 tertiary diamine surfactant.

(iii) a container holding luciferin/luciferase LDR (Celcis, Cambridge, UK) bioluminescence reagent of adenylate kinase activity less than 10' ⁸ U/100μl.

Optionally included in the package is a container of nonionic detergent solution (Triton X-1000.2 or equivalent) and/or a container holding an ATPase such as apyrase for the destruction of ATP released by the action of the nonionic detergent on a sample rendering it suitable for reassay by addition of the cationic detergent.

Example 7: Assay of known amounts of E.coli using method of the invention.

A one week old E. coli broth culture containing approximately 2.2 x 10 ⁷ microorganisms per 200μl of phosphate buffered saline pH7.4 was used as stock and diluted in successive dilutions of 10 with the collection/dilution reagent containing magnesium ions ((i) in Example 6) to give a range of samples of from 10 ⁷ to 0.1 microorganisms per 200μl sample.

Each 200μl sample was added to a 3-5ml luminometer tube, lOOμl of ADP/extractant reagent ((ii) in Example 6) added and the mixture, total volume 300μl, was incubated at room temperature for 1 or 5 minutes. On completion of the incubation lOOμl of modified Celcis LDR bioluminescence reagent, ((iii in Example 6 above) was added and the light emitted determined over a first 10 second interval and then over 10 second intervals up to one minute to determine the increase in light in cumulative fashion using a Biotrace M3 luminometer. The initial signal value was subtracted from the final reading to gain a measure of the signal in counts per minute.

The efficacy of the present method can be seen by reference to Figure 1 where statistically valid linear response between number of E.coli and light emitted by the sample mix after minute incubation with ADP is obtained for 10 organisms per sample and upward, for 100 organisms and upward with a 1 minute incubation, and a detection limit of about 10 organisms is given in both cases. This compares very favourably with the method of PCT/GB94/00118 which gives a difference of only 26 cpm after a 1 minute incubation with 100 organisms and 67 cpm with 1000; linear responses only being obtained with 1000 cells and over. By comparison 1000 cells per sample in the present method gives a signal increase of several thousand cpm after 1 minute.

It will be realised that in order to perform an assay for an unknown number of microorganisms in a sample using the present method, a calibration curve may be provided plotting known numbers of microorganisms against luminometer counts as shown in the Figures 1 and 2 (eg. as log values), deriving a number of counts from a sample containing the unknown number of microorganisms (including zero organisms) using the same protocol, and estimating the number of microorganisms in the sample as being that corresponding to the same number of counts on the curve.

It will be realised by those skilled in the art that the amount of adenylate kinase present in a particular microorganisms, eg.bacteria, may vary from other microorganisms. For example, yeasts contain more adenylate kinase than bacteria by virtue of their size, and indeed single yeasts can be detected by this method. Thus for a given microorganism a particular calibration curve may be required, and it may be necessary to provide such curves for different states of the same microorganism, eg. for weakened, pH or oxygen stressed organisms. However, a further advantage of the present method over exitsing ATP based methodology is that adenylate kinase content will be more closely correlated to cell numbers than the highly variable ATP content which is depleted by cell metabolism.