WO1999001768A1

WO1999001768A1 - Peroxidase-catalysed fluorescence

Info

Publication number: WO1999001768A1
Application number: PCT/GB1998/001888
Authority: WO
Inventors: Patricia Easton; Margaret Cobb
Original assignee: Nycomed Amersham Plc
Priority date: 1997-07-04
Filing date: 1998-06-29
Publication date: 1999-01-14
Also published as: AU8224998A

Abstract

An assay method comprises bringing together a leuco-dye and an oxidant and a peroxidase which catalyses oxidation of the leuco-dye to a fluorescent dye. The assay is performed in the presence of an enhancer which is a substituted aromatic hydroxy or amine or borate compound which enhances fluorescent light output. The dye may be for example a fluorescein or rhodamine or oxazine dye. Preferred enhancers include 6-hydroxybenzothiazole and 6-hydroxy-2-naphthoic acid.

Description

PEROXIDASE-CATALYSED FLUORESCENCE

This invention concerns an assay system involving the well known reaction between a peroxidase enzyme, e.g. horseradish peroxidase (HRP), an oxidant e.g. hydrogen peroxide, and a substrate or leuco-dye which on catalysed oxidation is converted to a product observable by its colour or because it has chemiluminescent or fluorescent properties. In an assay format, one of the three reagents is present in unknown concentration and is or represents an analyte. The reagents may be present in solution or any one of the three reagents may be immobilised e.g. on a membrane.

In one such system, the substrate is luminol or a related compound which on enzyme catalysed oxidation generates light by chemiluminescence. A p-substituted phenol may be present to enhance the chemiluminescence. The use of such phenols in this assay system has been described in the following patents: EP 87959; EP 1 16454; EP 296752; USP 4,729,950; USP 5,106,732; USP 5,279,940. In the peroxidase enzyme assay system, other known substrates are leuco-dyes which are oxidised to fluorescent dyes. This invention results from the discovery that, with certain families of such leuco-dyes, phenols and certain other aromatic compounds can be used to enhance the fluorescent light output. This enhancement may be by increasing the signal, or by reducing background light, so as to increase a signal-noise ratio, or by means of both effects together. As shown in the examples below, different families of dyes behave differently in this respect.

Thus the invention provides an assay method which comprises bringing together a leuco-dye and an oxidant and a peroxidase which catalyses oxidation of the leuco-dye to a fluorescent dye, wherein the assay is performed in the presence of an enhancer which is a substituted aromatic hydroxy or amine or borate compound which enhances fluorescent light output.

Assay systems in which a fluorescent signal is detected are quicker and cheaper to perform than those in which the signal is radioactive; and more accurate and sensitive than those in which the signal is colorimetric and provide greater signal intensity than chemiluminescence; and offer the possibility of multiplexing by using several different dyes simultaneously. Enzyme-catalysed fluorescence has the advantage over conventional fluorescent detection with single fluors in that it allows amplification of a fluorescent signal due to enzyme turnover. The assay systems of this invention thus have the potential to compete with known assay systems using other signals.

The leuco-dye may be a xanthane-based compound, including the families of dihydrofluoresceins and dihydrorhodamines, having the following structures:

Where L and M are the same or different and each is independently selected from -OR¹ or -NR²R³, where R¹ is H or -COR⁴,

R⁴ is a C1 - 12 aryl or alkyl group added to increase the stability of the leuco dye and prevent non-specific oxidation to the fluorescent dye, preferably -CH₃, R² and R³ are the same or different and each is independently selected from H, or any lower alkyl group of C1-10 which may carry a carboxyl or a sulphonic acid group, preferably CH₃, or C₂H₅,

A is H or any lower alkyl group of C1-10, preferably CH₃,

B is any electron withdrawing or donating group, preferably H, C1-C12 alkyl or alkoxy, CN, (CH₂)₂CO₂H, or a halogen (I, Br, Cl, F),

C and D are the same or different and each is independently selected from H, COOH, COOCH₃ or SO₃H, provided that when C is COOH this may form a lactone with position 9, or when L and/or M = -NR²R³, a fused ring structure may be formed between A and/or B as shown below:

The following dyes are preferred:

• 2',7'-dichlorodihydrofluorescein diacetate (2',7'- dichlorofluorescin diacetate).

• 5-(and-6)-carboxy-2',7'-dichlorodihydrofluorescein diacetate (5-(and-6)-carboxy-2',7'-dichlorofluorescin diacetate). • dihydrofluorescein diacetate.

• dihydrorhodamine 123.

• dihydrorhodamine 6G.

Alternatively the leuco-dye may be an oxazine-based compound including the following structures:

Where L and M are the same or different and each is independently selected from -OR¹ or -NR²R³, where R¹ = H or -COR⁴, R⁴ is a C1 - 12 aryl or alkyl group added to increase the stability of the leuco dye and prevent non-specific oxidation to the fluorescent dye, preferably -CH₃,

R² and R³ are the same or different and each is independently selected from H, or any lower alkyl group of C1-10 which may carry a carboxyl or a sulphonic acid group, preferably CH₃, or C₂H₅, F, G, J, K, F' and G' are the same or different and each is independently selected from H, or any lower alkyl or alkoxy group of C1-10 or aryl group which may carry a carbonyl or sulphonic acid group, preferably CH₃, C₂H _5l (CH₂)CO₂H, or a halogen (I, Br, Cl, F), or CN,

Or F-G and/or F'-G' form a ring structure which may or may not be conjugated to the chromophore,

E is a leaving group which may include a function that modifies the physical properties of the dye such as solubility, preferably -COR⁵Z¹ where R⁵ is C1-C12 alkyl and Z¹ is H or carboxyl or sulphonic acid.

Phenoxazine based compounds may include the following structures (where F-G forms a ring structure):

Where J, K, L, M, E, F, F' and G are as defined above.

Also envisaged are thiazine and phenothiazine dyes, having structures as above except that the ring O atom is replaced by S, e.g. methylene blue. Other dyes envisaged include coumarin and extended coumarin dyes including disperse red, squarate dyes and cyanine dyes. In their fluorescent or oxidised state, all these dyes have a conjugated unsaturated structure that is not present in their reduced or leuco-dye state.

Preferred is the dye N-benzoyl-leuco-methylene blue.

Also envisaged are leuco-derivatives of extended coumarins based upon the following structure:

Where Q and Q¹ are the same or different and each is independently selected from H, or any lower alkyl group of C1-10 which may carry a carboxyl or sulphonic acid group, preferably C₂H₅ or (CH₂)₂COOH.

M¹ is selected from O or NH.

L¹ is a benzimidazole, benzoxazole or benzothiazole ring system, with or without additional substituents. For example:

Where Z=N, V=NH, N-alkyl, O or S and W is any substituent. An example of an extended leuco-coumarin structure is given below:

Where Q and Q¹ are as given above.

M is selected from O, NH or NCOAr.

T¹ substituents in the benzimidazole ring may be H or any common substituent, but preferably -SO₃H to enhance water solubility.

T is preferably CN or CONH₂.

The preferred dyes in this family are Disperse Red 277 leucodye I (N-ethyl derivative) and II (N-butyl derivative). Butyl

The enhancer may be p-substituted phenol or amine or boronic acid of the formula

where R = -OH, or -NH₂or -B(OH)₂

X = Halogen (I, Cl, Br); amine; sulphonic acid; or a salt or ester thereof; any group that extends the resonance, preferably imidazole, -CH=CH-COOH, carboxy, nitrile, formyl, or -(CH=CH)_n-Ar (where n is 0-4), or N=N-Ar, where Ar has the formula:

where Y is hydrogen, hydroxy, nitrile, amine, sulphonic acid, formyl or a salt or ester thereof. Phenol compounds are described in

EP 116 454. Organic amine compounds are described in USP 4,729,950. Boronic acid phenol compounds are described in US Patent 5, 512,451. Or the enhancer may be substituted naphthol of the formula

where R¹ and R² are the same or different and each is independently selected from hydrogen, carboxylic acid, nitrile, amine, sulphonic acid, formyl, Cl, Br or I; or a salt or ester thereof; or any other group that extends the resonance of the structure. Or R¹ and R² together form a ring which extends conjugation of the naphthyl system and is preferably an imidazole or oxazole or thiazole.

The nature of the peroxidase enzyme is not material to the invention. Conveniently HRP maybe used. Typical concentrations are in the range 0.001 pM to 1 nM.

The nature of the oxidant is not material to the invention. Preferred are per-compounds such as sodium perborate and hydrogen peroxide. Typical concentrations are in the range 10 μM to 300 mM particularly 50 μM to 10 mM.

The leuco-dye is preferably used at a concentration in the range 0.5 μM to 200 mM particularly 1 μM to 1 mM. The enhancer is preferably used at a concentration in the range 1 μM to 100 mM particularly 10 μM to 10 mM. The assay may suitably be performed at a temperature in the range 10 - 50°C in a buffered aqueous medium at pH in the range 6 - 10 particularly 7 - 9. It will often be convenient to perform multiple assays in parallel in individual wells of a Multiwell plate. In one assay format, all the agents are present in solution.

Any one reagent may be present in unknown concentration and may be or represent an analyte. A preferred formulation in solution is 0.01 mM N- acetyl-3,7-dihydroxyphenoxazine, 0.5mM sodium perborate, 0.1M 6-hydroxybenzothiazole, in 0.1 M Tris pH 8.0.

In another assay format, an analyte may be immobilised, either on the surface of an assay vessel or as a blot on a porous membrane. It is preferred that a blot be blocked by application of an aqueous fluid containing casein or milk powder; and that casein or milk powder be also included in the liquid phase components of the assay; this has the effect of improving the signal-to-noise ratio. It is often convenient to immobilise the peroxidase enzyme on a membrane blot, and to measure the concentration of the peroxidase enzyme as an assay for the immobilised analyte. A preferred formulation for membrane blots is 0.01 mM dihydrorhodamine-123, 0.1 mM 6-hydroxy-2-naphthoic acid, 0.5mM sodium perborate, 0.01 % casein in 0.1 M borate buffer pH 9.5. Dihydrorhodamine 123 has the advantage that it does not migrate readily on the membrane.

Equipment for observing fluorescent signals is readily available, and the wavelength is readily adjusted to detect output from the particular fluorescent dye being used. The assay may be performed on a qualitative or quantitative basis. When the assay is quantitative, the fluorescent light output may be measured at a particular time interval (after excitation) or may be the sum of all fluorescent light output over a defined period of time.

The following examples illustrate the invention. Examples 1 to 3 show the use of rhodamine dyes. Examples 4 to 6 show the use of fluorescein dyes. Examples 7 to 9 show the use of phenoxazine dyes. These families of dyes respond differently to enhancers:

• The rhodamine family shows a large enhancement of signal together with (generally) a small increase in background.

• The fluorescein family shows a moderate enhancement of signal - the effect on background is not clear. • The phenoxazine family shows a small enhancement of signal but with (generally) a slightly decreased background giving rise to an improved signa noise ratio.

EXAMPLES

Example 1. Enhancement of fluorescence from Dihydrorhodamine in solution.

20μl of a 1 nM solution of horseradish peroxidase (HRP) in 0.1 M borate buffer pH 8.5 was added to 80μl of detection reagent containing dihydrorhodamine 123 (final concentration 0.1 mM), sodium perborate (final concentration 1 mM) and enhancer compounds tested at final concentrations of 3.0, 1.0, 0.3, 0., 0.03 and OmM in borate buffer pH 8.5. The enhancers were prepared as 5mM and 15mM stocks in the same buffer. The borate buffer was prepared from orthoboric acid and NaOH. The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using the Vistra Fluorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 570nm emission filter and with the PMT set to 500V. Unlike enhancement of chemiluminescence with luminol, the enhancers do not generally appear to act by reducing the background signal

For results see Table 1. Table 1. Relative enhancement activity of phenols with dihydrorhodamine in solution.

Enhancer Max. Max. Signal: Effect on enhancement¹ Noise ratio² background³ (mM Enhancer) (mM Enhancer)

PIMP 270 (1.0) 67.5 (0.3) x5⁴

BiPCN 149 (0.3) 123.6 (0.3) x5

PIP 148 (10) 88 (3.0) <0.5⁵

BiPCA 159 (10) 79 (0.3) <0.5

PHCA 121 (0.3) 129 (0.3) Decreases⁶

HBT 286 (10) 124 (10) <0.5

NA 262 (0.3) 1 1 1 (0.3) <0.5

BrN 114 (0. ) 82 (0.3) <0.5

AP 50 (0.3) 71 (10) Decreases

MA 31 (0.3) 54 (0.3) Decreases

UNENHANCED 2.2

1. The maximum enhancement values were obtained by comparing the maximum fluorescent signal for each enhancer relative to the signal from the system without enhancer at the same time point. Values in parentheses show the concentration of enhancer.

2. The signal:noise ratios (S/N) were obtained by comparing the fluorescent signals from the enhanced reactions in the presence and absence of HRP at the optimum time point for each enhancer (t=20mins for MA, t=10mins for all the rest) .

3. Effect on background as the enhancer concentration increases. 4. Background increases as enhancer concentration increases to a maximum of approximately 5 x unenhanced value.

5. Background increases as enhancer concentration increases, but to less than half of the unenhanced value.

6. Enhancers cause a decrease in fluorescent background (but by no greater than half of the unenhanced value).

PIMP 4-lmidazol-yl-phenol

BiPCN 4'-Hydroxy-4'-biphenyl carbonit le

PIP 4-lodophenol

BiPCA 4-(4'-hydroxyphenyl)benzoic acid

PHCA : 4-Hydroxycinnamic acid

HBT 6-Hydroxybenzothiazole

NA 6-Hydroxy-2-naphthoic acid

BrN 6-Bromo-2-naphthol

AP 4-Acetamidophenol

MA Methoxyaniline

Example 2. Relative activity of enhancer phenols with dihydrorhodamine on membrane blots.

A series of dot blots of mouse immunoglobulin G (IgG) were prepared on PVDF membrane at the following concentrations: 50ng, 5ng, 500pg, 50pg, 10pg, 5pg, 1 pg, 0.5pg and Opg.

Before each experiment the blots were wetted in methanol and rinsed in water before blocking in 5% milk powder in phosphate- buffered saline (pH 7.5) containing 0.1 % (v/v) Tween-20 (PBS-T). The blocking step took place overnight at 2-4°C.

The blots were next washed in PBS-T for 1x15 min and 2x5min before incubating with biotinylated anti-mouse IgG diluted 1 :5000 in PBS-T for 1 hour at room temperature. After washing as before, the blots were incubated with streptavidin-HRP conjugate at a dilution of 1 :5000 in PBS-T for 1 hour at room temperature. The blots were washed again in PBS-T for 1x15 min and 4x5 min before the addition of detection reagent.

The detection reagent contained 0.1 mM dihydrorhodamine- 123, 0.5mM sodium perborate and enhancer compounds at 1mM, 0.5mm and 0.1 mM (except PIP which was tested at 10mM, 2mM and 1mM) in 0.1 M borate buffer pH 9.5.

The blots were imaged using the Vistra Fluorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 570nm emission filter and with the PMT set to 500V. For results see Table 2.

Table 2.

The maximum signal: noise ratio for each enhancer was obtained from the 5ng dot in comparison to the negative control on each blot. βN : β-naphthol PPP : p-phenylphenol

Example 3. Detection of Southern blots with the preferred formulation of dihydrorhodamine-123.

Hind Ill-treated human genomic DNA was separated by agarose gel electrophoresis at loadings of 2μg-0.5μg and Southern blotted onto Hybond N+ membrane. Fluorescein-labelled N-Ras probe was prepared by the random prime reaction described in the Amersham random prime labelling and detection kit booklet (RPN 3040) using a 1-3 hour incubation. This was hybridised to the genomic blots at a probe concentration of 10-12.6 ng/ml for up to 16 hours at 60°C. The blots were washed at 60°C in an excess of 1xSSC, 0.1 %SDS(w/v) for 15 minutes, then in O.δxSSC, 0.1 % SDS for a further 15 minutes. The blots were rinsed in TBS (10mM Tris-HCI, 300mM NaCl, pH 7.5), then blocked for 1 hour at room temperature in an excess of liquid block diluted 1 :10 in TBS.

Sheep anti-fluorescein HRP conjugate was added at a 1 :1000 dilution in 0.5% BSA (w/v) in TBS for 1 hour at room temperature. The blots were next washed in an excess of 0.1% Tween-20 in TBS for 3x 10 minutes before the addition of detection reagent. The detection reagent contained O.OI mM dihydrorhodamine-123, 0.1mM 6-hydroxy-2-naphthoic acid, 0.5mM sodium perborate, 0.01% casein in 0.1 M borate buffer pH 9.5. The blots were imaged using the Vistra Fluorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 570nm emission filter and with the PMT set to 500V. The limit of detection was 0.25 pg DNA. Example 4. Enhancement of fluorescence from dichlorofiuorescin diacetate in solution.

20μl of a 2.5nM solution of horseradish peroxidase (HRP) in 0.1 M borate buffer pH 8.5 was added to 80μl of detection reagent containing 2',7'-dichlorofluorescin diacetate (DCF diacetate) (final concentration 0.1 mM), sodium perborate (final concentration 0.25mM) and enhancer compounds tested at final concentrations of 10.0, 4.0, 1.2, and OmM in borate buffer pH 8.5. The enhancers were prepared as 10mM stocks in the same buffer. The borate buffer was prepared from orthoboric acid and NaOH.

The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using the Vistra Fluorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 530nm emission filter and with the PMT set to 450V.

The results shown in Table 3 below are taken at t=16 minutes after the start of the reaction.

Table 3.

Example 5. Relative activity of enhancer phenols with dichlorofluorescin diacetate on membrane blots. λ-Hind 111 DNA was labelled with fluorescein in a random prime reaction as described in Example 3. This labelled DNA was then dotted onto Hybond N+ membrane at loadings of 3pg, 1 pg, 0.3pg, 0.1 pg, 0.03pg and Opg. The blots were blocked in a mixture of a 1 ;20 dilution of liquid block and 0.5% BSA in TBS pH 7.5 (see Example 3) for 1 hour at room temperature. The blots were now treated as described in Example 3, except that the final washes were 2x10 min and 2x 5 min in TBS-Tween. The detection reagent contained 0.02mM 2',7'-dichlorofluorescin diacetate, 0.5mM sodium perborate and a range of enhancers at 2.0, 1.0, 0.5 and 0.1 mM in 0.1 M borate buffer pH 9.5. The blots were imaged using the Vistra Fluorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 530nm emission filter and with the PMT set to 450V. For results see Table 4.

Table 4.

HHBP : 4-Hydroxy-3-[3-(p-hydroxyphenyl)-1-oxo-2-propenyl]-2H-1- benzopyran-2-one.

Example 6. Relative activity of enhancer phenols with fluorescein derivatives on membrane blots.

Hybond N+ blots were prepared and treated as described in Example 5. The detection reagents contained 0.5mM sodium perborate; 0.01 mM substrate consisting of either 2', 7'-dichlorofluorescin diacetate, dihydrofluorescein diacetate or 5-(&-6-)-carboxy-2',7'-dichlorofluorescin diacetate. The enhancers tested were 1mM NA, 0.5mM BiPCA and 0.5mM PHCBA in 0.1 M borate buffer pH 9.5. The blots were imaged using the Vistra Fiuorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 530nm emission filter and with the PMT set to 500V. For results see Table 5.

Table 5.

1. Limit of detection on dot blot.

2. Maximum signal:noise ratio for the 3pg DNA dot. 3. PHCBA: trans-4-(3-propenoic acid) phenyl boronic acid.

Example 7: Effect of enhancers on the fluorescent signal from N- acetyl-3,7-dihydroxyphenoxazine.

20μl of a 0.5nM solution of horseradish peroxidase (HRP) in 0.1 M Tris buffer pH 7.5 was added to 80μl of detection reagent containing N-acetyl-3,7-dihydroxyphenoxazine (final concentration 0.01 mM in 100μl), sodium perborate (final concentration 0.5mM in 100μl) and enhancer compounds tested at final concentrations of 3.0, 1.0, 0.3, 0.1 , 0.03 and OmM in 0.1 M Tris buffer pH 7.5. The enhancers were prepared as 10mM stocks in the same buffer.

The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using the Vistra Fiuorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 590nm emission filter and with the PMT set to 500V. For results see Table 6.

Table 6

1. The maximum enhancement values were obtained at the initial time point (t = 3 minutes) by comparing the maximum fluorescent signal for each enhancer relative to the signal from the system without enhancer. Values in parentheses show the concentration of enhancer.

2. The signal:noise ratios (S/N) were obtained by comparing the fluorescent signals from the enhanced reactions in the presence and absence of HRP. 3. Enhanced background as % of unenhanced background at t=3 mins. Although some enhancers increase the background above that of the unenhanced reaction, the majority of enhancers have the effect of lowering the background. The increase in background over time is also lower in the presence of many enhancers so that the optimal signai:noise ratio is between 30-120 minutes.

Example 8. HRP dilution curve with N-acetyl-3,7- dihydroxyphenoxazine in solution with three different enhancers. 20μl of a range of HRP dilutions was added to 80μl detection reagent. The final concentrations of the reactants in 100μl was as follows: 100pM to 0.01 pM HRP; 0.01 mM N-acetyl-3,7-dihydroxyphenoxazine; 0.5mM sodium perborate; 0.1 mM 6-hydroxybenzothiazole or 0.1 mM 4- iodophenol or 0.03mM 6-hydroxy-2-naphthoic acid in 0.1 M Tris buffer pH 7.5. The unenhanced reaction was carried out in 0.1 M sodium phosphate buffer, pH 7.0, since it had previously been established that this was the optimum pH for this reaction.

The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using the Vistra Fiuorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 590nm emission filter and with the PMT set to 500V.

The readings were taken 2 hours after the start of the reaction. The limits of detection (set at 2x background) were 0.1 I pmolL^"1 for HBT, 0.12pmolL^'1 for NA, 0.16pmol ¹ for PIP and 0.18pmolL¹ for the unenhanced reaction. Example 9. Comparison between N-acetyl-3,7-dihydroxyphenoxazine and TMB in the detection of human tumour necrosis factor-α.

The Biotrak™ human tumour necrosis factor-α [(h)TNFα] ELISA kit from Amersham International was used as a test system in both its normal (RPN 2758) and high sensitivity (RPN 2788) forms. The assays were carried out according to the protocols in the kit booklets. The chromogenic substrate, 3,3',5,5'-tetramethylbenzidine (TMB) was incubated with a (h)TNFα standard curve for 30 minutes as detailed in the protocol booklet before stopping with sulphuric acid and immediately reading the absorbance at 405nm.

The fluorogenic reaction mixture contained 0.01 mM N-acetyl- 3,7-dihydroxyphenoxazine (Phx) ; O.δmM sodium perborate; and 0.1 mM 6- hydroxybenzothiazole in 0.1M Tris buffer pH 7.5.

The reactions were carried out in the antibody coated-clear, flat-bottomed 96-well microtitre plates that were part of the kit.

The TMB absorbance readings were carried out in a Labsystems Multiskan MCC/340 plate reader at room temperature and a Molecular Dynamic's BioLumin 960 instrument at 37°C.

The fluorescent readings were taken using the Vistra Fiuorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a 590nm emission filter and with the PMT set to 500V at room temperature, and a Molecular Dynamic's BioLumin 960 instrument with an excitation filter of 560nm and an emission filter of 595nm at 37°C.

N-Acetyl-3,7-dihydroxyphenoxazine was approximately four- fold more sensitive than TMB irrespective of assay system or instrument used.

See Table 7 for results. Table 7

1. Limit of detection of (h)TNFα from a standard curve, with a limit of detection set at 2x background.

Example 10. Comparison of enhanced fluorescence from dihydrorhodamine and N-acetyl-3,7-dihydroxyphenoxazine against enhanced luminol chemiluminescence in solution.

DHR-123: 20μl of a 1 nM solution of horseradish peroxidase (HRP) in 0.1 M borate buffer pH 8.5 was added to 80μl of detection reagent containing dihydrorhodamine 123 (final concentration 0.1 mM), sodium perborate (final concentration 1mM) and enhancer compounds tested at final concentrations of 3.0, 1.0, 0.3, 0., 0.03 and OmM in borate buffer pH 8.5. The enhancers were prepared as 5mM and 15mM stocks in the same buffer. The borate buffer was prepared from orthoboric acid and NaOH. Each assay was performed in duplicate. N-AcPhx: 20μl of a 0.5nM solution of horseradish peroxidase (HRP) in 0.1 M Tris buffer pH 7.5 was added to 80μl of detection reagent containing N-acetyl-3,7-dihydroxyphenoxazine (final concentration 0.01 mM in 100μl), sodium perborate (final concentration O.δmM in 100μl) and enhancer compounds tested at final concentrations of 3.0, 1.0, 0.3, 0.1 , 0.03 and OmM in 0.1 M Tris buffer pH 7.δ. The enhancers were prepared as 10mM stocks in the same buffer. Each assay was performed in duplicate. The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using the Vistra Fiuorlmager SI with an excitation wavelength of 488nm (from an Argon laser), a δ70nm emission filter (for DHR-123) or a δ90nm emission filter (for N-AcPhx)and with the PMT set to 500V.

Luminol: 10μl of a 5nM solution of HRP was added to 90μl detection reagent containing 3mM sodium perborate, O.δmM luminol and enhancer compounds at final concentrations of 1.0, O.δ, 0.1 , O.OImM and OmM in 0.1 M borate buffer pH 8.δ. The borate buffer was prepared as above. The reactions were carried out in Microfluor "B" black, flat-bottom, 96-well microtitre plates from Dynatech Laboratories Inc. (US). The plates were read in an Amerlite plate reader from Amersham International pic (UK) set at a dwell time of 0.2 seconds/well. Each assay was performed in duplicate.

It can be seen that the enhancer compounds have different effects upon the three substrates so that the best enhancer for one substrate is not necessarily the best enhancer for the other two. Unlike enhancement of chemiluminescence with luminol and fluorescence with N- acetyl-3,7-dihydroxyphenoxazine, the enhancers do not generally appear to act by reducing the background signal with dihydrorhodamine 123. For results see Table 8. Table 8. Relative enhancement activity of phenols with dihydrorhodamine, N-acetyl-3,7-dihydroxyphenoxazine and luminol in solution.

1. The maximum enhancement values were obtained by comparing the maximum fluorescent signal for each enhancer relative to the signal from the system without enhancer at the same time point. 2. The maximum enhancement values were obtained by comparing the maximum fluorescent signal for each enhancer relative to the mean signal from the system without enhancer at a single time point.

Example 11. Comparison between N-acety 1-3,7- dihydroxyphenoxazine, dihydrorhodamine 123, 2',7'- dichlorodihydrofluorescein diacetate and 5'-(and- 6)-carboxy-2', 7'- dichlorodihydrofluorescein diacetate in the detection of human tumour necrosis factor-α. The Biotrak™ human tumour necrosis factor-α [(h)TNFα]

ELISA kit from Amersham International was used as a test system in its normal sensitivity form (RPN 2758) . The assays were carried out according to the protocols in the kit booklets.

The chromogenic substrate, 3,3',5,5'-tetramethylbenzidine (TMB) was incubated with a (h)TNFα standard curve for 30 minutes as detailed in the protocol booklet before stopping with sulphuric acid and immediately reading the absorbance at 405nm.

The fluorogenic reaction mixture contained eitherl OμM N- acetyl-3,7-dihydroxyphenoxazine (Phx) or 0.25μM dihydrorhodamine (DHR)123, 2',7'-dichlorodihydrofluorescein (DCF) diacetate or δ'-(and- 6)- carboxy-2', 7'-dichlorodihydrofluorescein diacetate plus O.δmM sodium perborate; and either no enhancer or 0.1 mM 6-hydroxybenzothiazole

(HBT) or 0.03mM 6-hydroxy-2-naphthoic acid (NA) in 0.1 M Tris buffer pH

7.5. The reactions were carried out in the antibody coated-clear, flat-bottomed 96-well microtitre plates that were part of the kit.

The TMB absorbance readings were carried out in a

Molecular Dynamic's BioLumin 960 instrument at 37°C set to absorbance mode. The fluorescent readings were taken using the BioLumin 960 instrument from Molecular Dynamics with the appropriate excitation and emission filters for each substrate at 37°C (see Table 9 for details).

The signal:noise levels were determined for each of the four substrates over a range of (h)TNFα concentrations over time. Table 9 below gives the results for 1 ng/ml (h)TNFα at t=20 minutes.

Table 9

Example 12. Effect of enhancer on Disperse Red leucodyes

The absorbance spectra of Disperse Red 277 (imine form) leucodye I (N-ethyl derivative) and II (N-butyl derivative) were investigated in the presence and absence of horseradish peroxidase (HRP) and enhancer phenol 4-(4'-hydroxyphenyl)benzoic acid (BiPCA). The reaction mixture was comprised of the following reagents at the final concentrations given below: 0.1 mM Disperse Red 277 leucodye I or II, 3 mM sodium perborate, O.δ mM BiPCA, δOO pM HRP in 0.1 M borate buffer (prepared from orthoboric acid and sodium hydroxide) pH 8.δ or 0.1 M phosphate buffer pH 6.0.

The reactions were carried out in a quartz cuvette and the changes in absorbance spectra monitored in a Phillips PU 8720 UVΛ/isible spectrophotometer over the range 2δ0-700 nm. The fluorescent spectra were monitored in a Hitachi F-4δ00 fluorimeter at PMT=700V, and slit widths of Ex/Em 2.δ nm/δ nm.

For results see Table 10.

Table 10

nd: Not determined.

1. Absorbance in optical density units 2. Emission wavelength. In phosphate buffer pH 6 there is a red shift of approximately 10 nm in both the excitation and emission wavelengths.

3. Values determined from emission scan with excitation wavelength of 5δδ nm.

4. Excitation at δ23 nm (DRI) and δ26 nm (DRIl) will not give rise to fluorescence and are probably due to dye aggregation. δ. Values determined from emission scan with excitation wavelength of 559 nm.

Example 13. Enhancement of HRP-catalysed N-benzoyl-leuco- methylene blue

The fluorescent signal from N-benzoyl-leuco-methylene blue was investigated in the presence and absence of the enhancer phenol 4- (4'-hydroxyphenyl)benzoic acid (BiPCA).

A substrate mix was prepared as follows: 8.0 ml 0.1 M borate buffer pH 8.δ, 4.0 ml 1-methyl-2-pyrrolidinone, 0.466 ml 60 mM sodium perborate, 0.07δ ml 100 mM N-benzoyl-leuco-methylene blue in dimethylformamide (DMF). To 100 μl of this mixture was added δO μl of BiPCA in DMF over the concentration range 2-20 mM (final concentration). Finally, 1.2δ units of HRP in δ μl was added to the reaction mixture which was left for ten minutes before imaging.

The reactions were carried out in clear, flat-bottomed 96-well microtitre plates. They were imaged using a prototype Vistra Fiuorlmager δ7δ with a helium/neon laser exciting at 633 nm and a 670 nm emission filter.

For results see table 11. Table 11

1. Signal from enhanced reaction relative to signal from unenhanced reaction.

2. The signahbackground ratios were obtained by comparing the signals in the presence and absence of HRP.

3. The rise in background with enhancer concentration was determined relative to the background signal from the unenhanced reaction.

Claims

1. An assay method which comprises bringing together a leucodye and an oxidant and a peroxidase which catalyses oxidation of the leuco-dye to a fluorescent dye, wherein the assay is performed in the presence of an enhancer which is a substituted aromatic hydroxy or amine or borate compound which enhances fluorescent light output.

2. An assay method as claimed in claim 1 , wherein the leucodye is a xanthene-based compound having the following structure

Where L and M are the same or different and each is independently selected from -OR¹ or -NR²R³, where R¹ is H or -COR⁴, R⁴ is a C1 - 12 aryl or alkyl group added to increase the stability of the leuco dye and prevent non-specific oxidation to the fluorescent dye, preferably -CH₃, R² and R³ are the same or different and each is independently selected from H, or any lower alkyl group of C1 -10 which may carry a carboxyl or a sulphonic acid group, preferably CH₃, or C₂H₅, A is H or any lower alkyl group of C1-10, preferably CH₃, B is any electron withdrawing or donating group, preferably H, C1-C12 alkyl or alkoxy, CN, (CH₂)₂CO₂H, or a halogen (I, Br, Cl, F), C and D are the same or different and each is independently selected from H, COOH, COOCH₃ or SO₃H, provided that when C is COOH this may form a lactone with position 9, or when L and/or M = -NR²R³, a fused ring structure may be formed between A and/or B as shown below:

3. An assay method as claimed in claim 1 , wherein the leuco- dye is an oxazine-based compound having the following structure:

R² and R³ are the same or different and each is independently selected from H, or any lower alkyl group of C1-10 which may carry a carboxyl or a sulphonic acid group, preferably CH₃, or C₂H₅, F, G, J, K, F' and G' are the same or different and each is independently selected from H, or any lower alkyl group of C1-10 or aryl group, preferably CH₃, C₂H ₅, (CH₂)CO₂H, or a halogen (I, Br, Cl, F),

E is a leaving group which may include a function that modifies the physical properties of the dye such as solubility, preferably -COR⁵Z¹ where R⁵ is C1-C12 alkyl and Z¹ is H or carboxyl or sulphonic acid. 4. An assay method as claimed in claim 1 , wherein the leucodye is a phenoxazine-based compound, or a thiazine, a phenothiazine, methylene blue, a coumarin, an extended coumarin, e.g. disperse red, a squarate or a cyanine dye. δ. An assay method as claimed in any one of claims 1 to 4, wherein the enhancer is a p-substituted phenol or amine or boronic acid of the formula

where R = -OH, or -NH₂ or -B(OR)₂, X = Halogen (I, Cl, Br); amine; sulphonic acid; or a salt or ester thereof; any group that extends the resonance, preferably imidazole, -CH=CH-COOH, carboxy, nitrile, formyl, or -(CH=CH)_n-Ar (where n is 0-4), or N=N-Ar, where Ar has the formula:

where Y is hydrogen, hydroxy, nitrile, amine, sulphonic acid, formyl or a salt or ester thereof.

6. An assay method as claimed in any one of claims 1 to 4, wherein the enhancer is a substituted naphthol of the formula

where R¹ and R² are the same or different and each is independently selected from hydrogen, carboxylic acid, nitrile, amine, sulphonic acid, formyl, Cl, Br or I or a salt or ester thereof; or any other group that extends the resonance of the structure. Or R¹ and R² together form a ring which extends conjugation of the naphthyl system and is preferably an imidazole or oxazole or thiazole.

7. An assay method as claimed in any one of claims 1 to 6, wherein all the reagents are in solution.

8. An assay method as claimed in claim 7, wherein the leucodye is N-acetyl-3,7-dihydroxyphenoxazine and the enhancer is 6- hydroxybenzothiazole.

9. An assay method as claimed in any one of claims 1 to 6, wherein the peroxidase is immobilised on a membrane.

10. An assay method as claimed in claim 9, wherein the leucodye is dihydrorhodamine 123 and the enhancer is 6-hydroxy-2-naphthoic acid.

11. An assay kit comprising:

• a supply of a leuco-dye selected from dihydrofluorosceins, dihydrorhodamines and oxazines,

• a supply of an oxidant, • a supply of a peroxidase enzyme,

• and a supply of an aromatic hydroxy or amine or borate compound as an enhancer.