- Email: [email protected]
A firefly luciferase assay for subnanomolar concentrations of amphipathic substances
ANALYTICAL
BIOCHEMISTRY
190,
304-308
(1990)
A Firefly Luciferase Assay for Subnanomolar Concentrations of Amphipathic Substances’ Shokofeh Naderi and Donald L. Melchior’ Department of Biochemistry, University of MassachusettsMedical School, 55 Lake Avenue North, Worcester,Massachusetts01605
Received
May
4,199O
A sensitive assay is described for accurately quantitating subnanomolar aqueous concentrations of a wide variety of amphipathic and hydrophobic biological materials. This paper extends a luciferase-luciferin method previously used to measure aqueous concentrations of anesthetics to a variety of hormones, metabolites, and membrane active agents. The assay can cover analyte ranges from picomolar to micromolar. The sensitivity of the assay is shown to correlate with the hydrophobic nature of the analyte. The mechanism of the assay appears to result from competition of analyte with luciferin for a hydrophobic binding site on the luciferase molecule. This assay allows measurement of the partitioning of analytes into lipid bilayers from aqueous solution. 0 1990 Academic Press. Inc.
Biochemical investigations on the utilization of hydrophobic and amphipathic materials are becoming increasingly important. Hydrophobic and amphipathic substances participate in and modulate various biomembrane processes, frequently as metabolic substrates or as hormones, drugs, etc. In uiuo, these molecules can enter the membrane directly from an extracellular or cytosolic aqueous phase. A difficulty inherent in investigations of these materials is their low solubility in aqueous solution (1). Since biological processesusually require these substances as free molecular species, for in vitro investigations it is most relevant to work below critical micellar concentrations (2).
i This work DMB-8416219. 2 To whom 304
was funded by the National Science Foundation, We gratefully acknowledge this support. correspondence should be addressed.
Grant
CMCs3 for these substances are frequently in the subnanomolar range. Quantifying them in solution, therefore, can be difficult and imprecise. In addition to low solubility, hydrophobic, and amphipathic materials partition spontaneously into the lipid bilayer phase of biomembranes (3) where they can be recognized and act on or be acted on by membrane enzymes, receptors, etc. The development of methods that can measure low amounts of these substances in solution and also permit determinations of their partitioning into lipid bilayers are needed. In this paper, we extend a firefly luciferase assay previously employed to determine the partitioning of anesthetics into lipid bilayers (4) to the investigation of a wide variety of biologically relevant hydrophobic and amphipathic substances. Light production by firefly luciferase results from activation of the substrate luciferin in the presence of ATP and Mg2+ to form enzyme bound luciferyl adenylate. This complex reacts with oxygen which after several intermediate steps yields a photon of light and the ground-state enzyme-product complex (5). Deluca (6) demonstrated that certain dyes can act as competitive inhibitors of the luciferase enzyme by competing with luciferin for its normal hydrophobic site on the luciferase molecule. Middleton and Smith demonstrated (7) that the anesthetic diethyl ether inhibited in vitro light emission by the luciferase of the bacterium Vibrio fisheri by inhibiting binding of the aldehyde factor necessary s Abbreviations used: ACTH, adrenocorticotropin; ALP, alkyl lysolecithin (2-O-methyl-I-O-octadecylglycero-3-phosphocholine); CMC, critical micellar concentration; DML, dimyristoyl lecithin (1,2distearoyl-sn-glycero-3 phosphatidylcholinel; HEAT, 5-hydroxyeicosatetraenoic acid, MPL, monopalmitoyl lecithin (l-palmitoylsn - glycero 3 - phosphatidylcholine ) ; MSL , monostearoyl lecithin (I-stearoyl-sn-glycero-$phosphatidylcholine); LLS, luciferase-luciferin solution. 0003-x97/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
for production of luminescence by the bacterial luciferase. Franks and Lieb (8) using firefly luciferase demonstrated that a variety of alcohols and other small organic molecules having anesthetic properties inhibited the luciferase reaction. They used this to measure the partitioning of long-chain alcohols into lipid bilayers (4). MATERIALS
AND
METHODS
Reagents Sigma luciferase-luciferin reagent (LO633), luciferin, ATP, EDTA, cholesterol, myristic acid, ACTH, Substance P, insulin, and glycylglycine were obtained from Sigma (St. Louis, MO). Hexanoic acid and hexanol were obtained from Aldrich (Milwaukee, WI), hexane was from Fisher (Waukesha, WI), palmitoyl CoA was from Pharmacia (Piscataway, NJ), and neurotensin was from Peninsular (Belmont, CA). MPL, MSL, and DML were purchased from Avanti (Pelham, AL). PGEl, arachidonic acid, and 5-(U-HEAT was a gift from Professor S. Burstein (University of Massachusetts, Worcester, MA) and ALP was a gift from Professor A. Makryiannis (University of Connecticut, Storrs, CT).
ture (24’C) (17) and swirled. The large multilamellar liposomes formed were dispersed by 2-5 min sonication in a bath-type sonicator at a temperature above that of the lipid’s phase transition. Measurement
of light emission
Light emission was measured by means of a Hi-Tech S-4 Series stopped-flow spectrophotometer system (Salisbury, England). Solutions A and B were introduced into two reservoirs feeding two syringes. The reaction was initiated by triggering two pneumatically driven pistons to introduce solutions A and B at a ratio of l/l into a 40 ~1 quartz integral mixer/observation cell set for a lo-mm optical path length. Both the cell and reagent supply coils were immersed in a thermostatic bath adjustable from -100 to 100°C to within +-O.l”C. Light emission was transferred by a silica rod to a photomultiplier and the output amplified by the SF-40C control unit. Data were analyzed and processed by the Hightech ADS-2 Datapro Software Suite on an Apple IIe. Peak heights and areas were determined from hard copies of ADS-2 output using an Apple Graphics Pad. Assay procedure
Solutions Buffers. Buffer 1, 0.25 mM glycylglycine, 0.5 mM EDTA, pH 7.4 (adjusted with NaOH); Buffer 2, Buffer 1 made 1 mM in MgSO,. LLS consisted Luciferase-Luciferin Solution (LLS). of 1 mg luciferase-luciferin reagent per milliliter Buffer 2. This is prepared fresh daily. ATP solution consisted 100 mM ATP ATP Solution. in Buffer 2. Stock solutions. Stock solutions of MPL (160 PM), myristic acid (700 nM), hexane (1 mM), hexanol(1 mM), hexanoic acid (1 mM), ethanol (1 mM), cholesterol (300 nM), and insulin (200 nM) were made by adding appropriate substances to Buffer 1 and stirring at 22°C for 1 h. The cholesterol stock solution was sonicated (in a bath-type sonicator) additionally for $ h. Insulin stock solution was allowed to sit overnight at room temperature (this was required for the insulin to go into solution) . PGEl (140 PM), arachidonic acid (250 PM), and 5-aHEAT (300 nM) stock solutions were made by drying under nitrogen appropriate volumes of ethanolic solutions of the respective substances, pumping under vacuum (1 h, 100 pm Hg) and then adding Buffer 1. Liposomes Lipid bilayers The appropriate at a temperature
were formed from dimyristoyl lecithin. lipid was added (1 mg/ml) to Buffer 1 above the lipid’s transition tempera-
Assay solutions of substances of interest were made by adding appropriate volumes of stock solution to Buffer 2. Desired amounts of luciferin-luciferase reagent (as a powder) were added. Corresponding ATP assay solutions were made by adding equivalent amounts of the substances of interest to ATP solution. Both solutions were incubated at room temperature for various times. The assay solution and ATP solution were then put in their appropriate reservoirs in the stopped-flow spectrophotometer. Each point reported was taken a minimum of six times and a minimum of two samples were prepared for each substance. RESULTS
AND
DISCUSSION
When luciferin-luciferase solution is mixed with ATP solution in the stopped-flow spectrophotometer, a rapid initial burst of light is observed (maximum light emission occurs at about 6 ms) followed by a slow decay (Fig. 1A). This is in accord with findings by DeLuca and McElroy (5). In order to maximize the sensitivity of the assay, light output for the luciferase-luciferin reaction was determined as a function of [M%+], pH, temperature, and [ATP]. The curves obtained were similar to those reported in the literature (9). Maximum light emission was found at 2 mM [M$‘], pH 7.4, and 23°C. [ATP] above 2 mM resulted in no enhancement of light production. Saturating the solutions with 0, resulted in no enhancement of light emission. In agreement with the findings of Karl and Holm-Hansen (lo), additional luciferin added to the Sigma luciferase-luciferin re-
306
NADER1
AND
MELCHIOR TABLE
1
Sensitivity of Luciferase-Luciferin Assay to Various Biological Materials Concentration (nM) inhbition Material Myristic acid Prostaglandin El Hexane Arachidonic acid Hexanol 5-Hydroxyeicosatetraenoic Palmitoyl CoA
Seconds FIG. 1. Photomultiplier output of light emission upon mixing luciferase-luciferin solution in (B) the presence and (A) the absence of 50 nM monopalmitoyl lecithin. Ordinate, full scale 1 s; abcissa, full scale 0.2 v.
acid (HEAT)
Adrenocorticotropin (ACTH) Ethanol 2-0-methyl-1-0-octadecylglycero-3phosphocholine (ALP) Monopalmitoyl lecithin (MPL) Monostearoyl lecithin (MSL) Hexanoic acid Neurotensin”
20%
80%
0.25 0.25 0.5 0.5 0.5 1 1
20 100 10 10 50 30 -
5 5
40 40
5 5 5 10 (50%
50 50 100 100 inhibition at
agent resulted in increased light emission. Over the range investigated (0.25 to 2 mg/ml, luciferase-luciferin reagent, approx 4.5-36 nM luciferase), light emission as measured by photomultiplier output was linear with respect to the concentration of enzymes and reactants. The normal working concentration of luciferase-luciferin reagent used was 1 mg/ml. The luciferase-luciferin reaction was found to be sensitive to the presence of a variety of amphipathic and hydrophobic substances (Table 1). The presence of these materials in the luciferase-luciferin solution results in reduced light emission. The quantitative nature of the reduction in light emission allows the luciferaseluciferin reaction to be used as an analytical means to quantitate these materials (analytes). Figure 1B shows the affect of 50 nM monopalmitoyl lecithin on the spectra of light emission relative to its absence (Fig. lA, control). Suppression of light emission was initially quantified in two different ways: (1) the ratio of total light output (area under the curve) relative to control (absence of analyte from the reaction) or (2) the ratio of maximum peak height relative to that of control. Peak height comparison was found to be the more precise as well as simpler method and was used throughout this study. A time course was carried out to determine the extent of light diminution as a function of incubation time of analyte in the luciferase-luciferin solution. For monopalmitoyl lecithin (50 nM), maximum light reduction was obtained in 20 min. This incubation time was used in the following studies. Figure 2 shows the affect on light output of varying submicellar concentrations of MPL (CMC, 2 X 10e6 M (18)). Reduced light emission as a function of analyte concentration follows a smooth curve. This type of dose-response curve is typical for
Substnace
P”
5 nM)
(50% inhibition at 100 nM) 0 0
Cholesterol Insulin
0 0
a Only one concentration measured. Note. The concentrations at which these materials gave 20 and 80% reduction of light emission relative to control (no analyte) are presented to indicate the sensitivity of luciferase-luciferin assay to individual analytes. All these materials gave smooth dose-response curves.
materials that inhibit light emission. Table 1 lists various substances tested. The concentrations at which these materials gave 20 and 80% reduction of light emission are presented to indicate the sensitivity of lucifer-
loon
204 0
8 10
" 20
8 30
Concentration
r 40
" 50
60
(nM)
FIG. 2. Sensitivity of luciferase-luciferin light emission molar concentrations of monopalmitoyl lecithin. Control monopalmitoyl lecithin, is taken as 100% light emission.
to nanovalue, no
307
01 0
IO
20
30
[Luciferin] 0
100
200
300
400
Concentration FIG. 3. Sensitivity of luciferase-luciferin lar concentrations of prostaglandin El. din El, is taken as 100% light emission.
500
600
(PM)
01 20
40
60
Concentration
50
60
70
kg/mL
the luciferase assay as a function of in the absence of (0) and in the preslecithin.
light emission to picomoControl value, no prostaglan-
ase-luciferin assay for these substances. From Table 1, it is apparent that the luciferase-luciferin assay can be used to detect subnanomolar concentrations of a variety of types of materials. As an example, Fig. 3 shows light reduction for prostaglandin El over the picomolar concentration range. In order to confirm an apparent correlation between the effectiveness of a substance in suppressing light production and its hydrophobic versus hydrophilic nature, we tested the series hexane, hexanol, and hexanoic acid. These compounds-an alkane, an alcohol, and a carboxylic acid-all have the same six carbon chain as a hydrophobic moiety, but are progressively more hydrophilic. As seen in Fig. 4, the sensitivity of the luciferase-luciferin assay for these materials is hexane > hexanol > hexanoic acid.
0
FIG. 5. Light emission for increasing amounts of luciferin ence of (0) 50 nM monopalmitoyl
40
60
100
120
(nM)
FIG. 4. Sensitivity of luciferase-luciferin light emission to hexane (o), hexanol (0), and hexanoic acid (w). Control value, no analyte, is taken as 100% light emission.
Our findings support the basis of this assay to be an interaction of analyte with the hydrophobic luciferin binding site of the luciferase molecule (6). Competition experiments were carried out to determine the V,, and K,,, of luciferase for luciferin in the presence and absence of an analyte, in this case 50 nM MPL. Figure 5 shows the results of this study. The data were analyzed by the method of nonlinear curve fitting (11). The correlation coefficient for the fit of control values was 0.982 and for data in the presence of MPL, 0.987. In the absence of MPL, V,, was 78.5 and K,,, 1.14 pg/ml, while in the presence of MPL, V,., was 79.7 and K,,, 4.62 pg/ ml. This is indicative of competitive inhibition between analyte and luciferin, most likely for a hydrophobic site on the luciferase molecule. Many amphipathic and hydrophobic materials partition from aqueous suspension into lipid bilayers. Lipid bilayers form the underlying structure of biological membranes. Proteins embedded in membranes are responsible for mediating many cellular processes (12). The partitioning of substances such as drugs, metabolites, and hormones from an aqueous milieu into the hydrophobic environment provided by bilayers can play an important regulatory role (13). It was of interest, therefore, to determine whether the luciferase-luciferin assay could be used to determine the transfer of substances other than anesthetics from aqueous suspension into lipid bilayers. Figure 5 shows the results of an investigation of this question. Light emission by the luciferase assay is unaffected by the presence of lipid bilayers. When luciferase-luciferin solution was incubated for 20 min with model bilayers formed from dimyristoyl lecithin (0.025 mg/ml), light output (+DML) was the same as in the absence of bilayers (CONTROL) (Fig. 6). The presence of 50 nM MPL in luciferase-luciferin solution devoid of bilayers results in a 45% reduction in light output (+MPL). When MPL was preincubated with dimyristoyl lecithin bilayers (0.025 mg/ml, 25’C for 20 min) prior to addi-
308
NADER1
3 E
AND
60
s -5 8
40
20
0
Control
+DML
+MPL
DML,MPL
Conditions FIG. 6. Ability of the luciferase-luciferin assay to detect the absorption of the amphipath monopalmitoyl lecithin by lipid bilayers. Light emission for CONTROL, absence of analyte and lipid bilayers; +DML, in the presence of dimyristoyl lecithin bilayers; +MPL, in the presence of 50 nM monopalmitoyl lecithin; DML, MPL in the presence of 50 nM monopalmitoyl lecithin incubated with dimyristoyl lecithin bilayers.
tion of luciferase-luciferin solution and the analysis carried out in the regular manner, light emission, (DML, MPL) was the same as the control value (CONTROL). This reflects the absence of MPL in the aqueous phase due to its having partitioned into the bilayers (19) where it is inaccessible to the luciferase. In summary, the luciferase-luciferin assay can provide a simple means for quantitating in vitro picomolar to micromolar concentrations of a variety of important biological substances. It provides a means for not only measuring aqueous concentrations in such materials but also for measuring their partitioning into lipid bilayers from the aqueous phase. The sensitivity of this assay varies with the analyte to be investigated (Table 1). Assay sensitivity is in to a large extent dependent on the specifics of the substance under investigation (Fig. 4). The sensitivity of the assay at a given luciferinluciferase reagent concentration decreases with increasing analyte concentration (Fig. 2). Maximum assay sensitivity is obtained at lower analyte concentrations and is limited by instrumental noise. Our instrument at its most sensitive setting can reproducibly distinguish +2% quenching. For prostaglandin El (Fig. 3) we can detect differences in concentrations of 25 PM. The reaction cell used in our instrument is 40 ~1. Theoretically, for a material such as prostaglandin El, we can optimally detect differences of 1 X 10-l’ mol. Realistically, this is not possible since about 0.5 ml of solution is necessary to load the spectrophotometer and make
MELCHIOR
several measurements. About 1.3 x lo-l4 mol of material is therefore a best possible minimum. As in working with low amounts of any materials, a possible source of error is loss of material to surfaces. In our apparatus, the material is exposed only to glass Teflon and quartz and we found no indication of loss of material. A variety of hydrophobic and amphipathic peptides, lipids, and small organic molecules are susceptible to this assay. While we, as others (5), have employed a commercial stopped-flow spectrophotometer to make quantitative measurements of the luciferin-luciferase reaction, they can be made by a variety of other means. Amongst those approaches used are custom made devices employing a photomultiplier tube whose amplified signal is fed to an oscilliscope or computer (8,14), an LKB-1250 luminometer (15), a custom designed luminometer (16), and an SAI Technology Model 3000 ATP photometer coupled to a Omniscribe recorder (9). REFERENCES 1. Tanford, C. (1980) The Hydrophobic 2. Small, D. M. (1986) The Physical New York.
Effect, Chemistry
Wiley, New York. of Lipids, Plenum,
3. Gennis, R. B. (1989) Biomembranes, Chapter 7, Springer-Verlag, New York. 4. Franks, N. P., and Lieb, W. R. (1986) Proc. N&l. Acad. Sci. USA 83,5116-5120. 5. DeLuca, M., and McElroy, W. D. (1974) Biochemistry 13, 921925. 6. DeLuca, M. (1969) Biochemistry 8,160-166. 7. Middleton,
A. J., and Smith,
E. B. (1976)
Proc. R. Sot. London
B
193,173-190. 8. Franks, N. P., and Lieb, W. R. (1984) Nature (London) 310,599601. 9. Webster, J. J., and Leach, F. R. (1980) J. Appl. Biochem. 2,469479. 10. Karl, D. M., and Holm-Hansen, 0. (1976) Anal. Btichem. 75, 100-112. 11. Press, W. H., Flannery, B. P., Teukolsky, S. A., and Vetterling, W. T. (1988) Numerical Recipes: The Art of Scientific Computing Cambridge Univ. Press, London. 12. Aloia, R. C., Curtain, C. C., and Gordon, L. M. (1988) Lipid DOmains and the Relationship to Membrane Function, A. R. Liss, New York. 13. Adam, A., and Delbruck, M. (1968) in Structural Biology and Molecular Biology (Davidson, N. and Rich, A., Eds.), pp- 198-215, Freeman, San Francisco, CA. 14. Thompson, A., Nigro, J., and Seliger, H. H. (1988) Biochem. Biophys. Res. Commun. 140,888-894. 15. Ugarova, N. N., Dukhovich, A. F., Shrets, S. V., Philippova, N. Y., and Berezin, I. V. (1987) Biochim. Biophys. Acta 921,465-472. 16. Lundin, A., and Thore, A. (1975) Anal. Biochem. 66,47-63. 17. Melchior, D. L., and Steim, J. M. (1976) Annu. Rev. Biochem. Bioeng. 6,205-238. 18. Welzien, H. U. (1979) Biochim. Biophys. Acta 772, 239-243. 19. Naderi, S., Carruthers, A., and Mel&or, D. L. (1989) Biochim. Biophys. Acta 985,173-181.
BIOCHEMISTRY
190,
304-308
(1990)
A Firefly Luciferase Assay for Subnanomolar Concentrations of Amphipathic Substances’ Shokofeh Naderi and Donald L. Melchior’ Department of Biochemistry, University of MassachusettsMedical School, 55 Lake Avenue North, Worcester,Massachusetts01605
Received
May
4,199O
A sensitive assay is described for accurately quantitating subnanomolar aqueous concentrations of a wide variety of amphipathic and hydrophobic biological materials. This paper extends a luciferase-luciferin method previously used to measure aqueous concentrations of anesthetics to a variety of hormones, metabolites, and membrane active agents. The assay can cover analyte ranges from picomolar to micromolar. The sensitivity of the assay is shown to correlate with the hydrophobic nature of the analyte. The mechanism of the assay appears to result from competition of analyte with luciferin for a hydrophobic binding site on the luciferase molecule. This assay allows measurement of the partitioning of analytes into lipid bilayers from aqueous solution. 0 1990 Academic Press. Inc.
Biochemical investigations on the utilization of hydrophobic and amphipathic materials are becoming increasingly important. Hydrophobic and amphipathic substances participate in and modulate various biomembrane processes, frequently as metabolic substrates or as hormones, drugs, etc. In uiuo, these molecules can enter the membrane directly from an extracellular or cytosolic aqueous phase. A difficulty inherent in investigations of these materials is their low solubility in aqueous solution (1). Since biological processesusually require these substances as free molecular species, for in vitro investigations it is most relevant to work below critical micellar concentrations (2).
i This work DMB-8416219. 2 To whom 304
was funded by the National Science Foundation, We gratefully acknowledge this support. correspondence should be addressed.
Grant
CMCs3 for these substances are frequently in the subnanomolar range. Quantifying them in solution, therefore, can be difficult and imprecise. In addition to low solubility, hydrophobic, and amphipathic materials partition spontaneously into the lipid bilayer phase of biomembranes (3) where they can be recognized and act on or be acted on by membrane enzymes, receptors, etc. The development of methods that can measure low amounts of these substances in solution and also permit determinations of their partitioning into lipid bilayers are needed. In this paper, we extend a firefly luciferase assay previously employed to determine the partitioning of anesthetics into lipid bilayers (4) to the investigation of a wide variety of biologically relevant hydrophobic and amphipathic substances. Light production by firefly luciferase results from activation of the substrate luciferin in the presence of ATP and Mg2+ to form enzyme bound luciferyl adenylate. This complex reacts with oxygen which after several intermediate steps yields a photon of light and the ground-state enzyme-product complex (5). Deluca (6) demonstrated that certain dyes can act as competitive inhibitors of the luciferase enzyme by competing with luciferin for its normal hydrophobic site on the luciferase molecule. Middleton and Smith demonstrated (7) that the anesthetic diethyl ether inhibited in vitro light emission by the luciferase of the bacterium Vibrio fisheri by inhibiting binding of the aldehyde factor necessary s Abbreviations used: ACTH, adrenocorticotropin; ALP, alkyl lysolecithin (2-O-methyl-I-O-octadecylglycero-3-phosphocholine); CMC, critical micellar concentration; DML, dimyristoyl lecithin (1,2distearoyl-sn-glycero-3 phosphatidylcholinel; HEAT, 5-hydroxyeicosatetraenoic acid, MPL, monopalmitoyl lecithin (l-palmitoylsn - glycero 3 - phosphatidylcholine ) ; MSL , monostearoyl lecithin (I-stearoyl-sn-glycero-$phosphatidylcholine); LLS, luciferase-luciferin solution. 0003-x97/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
for production of luminescence by the bacterial luciferase. Franks and Lieb (8) using firefly luciferase demonstrated that a variety of alcohols and other small organic molecules having anesthetic properties inhibited the luciferase reaction. They used this to measure the partitioning of long-chain alcohols into lipid bilayers (4). MATERIALS
AND
METHODS
Reagents Sigma luciferase-luciferin reagent (LO633), luciferin, ATP, EDTA, cholesterol, myristic acid, ACTH, Substance P, insulin, and glycylglycine were obtained from Sigma (St. Louis, MO). Hexanoic acid and hexanol were obtained from Aldrich (Milwaukee, WI), hexane was from Fisher (Waukesha, WI), palmitoyl CoA was from Pharmacia (Piscataway, NJ), and neurotensin was from Peninsular (Belmont, CA). MPL, MSL, and DML were purchased from Avanti (Pelham, AL). PGEl, arachidonic acid, and 5-(U-HEAT was a gift from Professor S. Burstein (University of Massachusetts, Worcester, MA) and ALP was a gift from Professor A. Makryiannis (University of Connecticut, Storrs, CT).
ture (24’C) (17) and swirled. The large multilamellar liposomes formed were dispersed by 2-5 min sonication in a bath-type sonicator at a temperature above that of the lipid’s phase transition. Measurement
of light emission
Light emission was measured by means of a Hi-Tech S-4 Series stopped-flow spectrophotometer system (Salisbury, England). Solutions A and B were introduced into two reservoirs feeding two syringes. The reaction was initiated by triggering two pneumatically driven pistons to introduce solutions A and B at a ratio of l/l into a 40 ~1 quartz integral mixer/observation cell set for a lo-mm optical path length. Both the cell and reagent supply coils were immersed in a thermostatic bath adjustable from -100 to 100°C to within +-O.l”C. Light emission was transferred by a silica rod to a photomultiplier and the output amplified by the SF-40C control unit. Data were analyzed and processed by the Hightech ADS-2 Datapro Software Suite on an Apple IIe. Peak heights and areas were determined from hard copies of ADS-2 output using an Apple Graphics Pad. Assay procedure
Solutions Buffers. Buffer 1, 0.25 mM glycylglycine, 0.5 mM EDTA, pH 7.4 (adjusted with NaOH); Buffer 2, Buffer 1 made 1 mM in MgSO,. LLS consisted Luciferase-Luciferin Solution (LLS). of 1 mg luciferase-luciferin reagent per milliliter Buffer 2. This is prepared fresh daily. ATP solution consisted 100 mM ATP ATP Solution. in Buffer 2. Stock solutions. Stock solutions of MPL (160 PM), myristic acid (700 nM), hexane (1 mM), hexanol(1 mM), hexanoic acid (1 mM), ethanol (1 mM), cholesterol (300 nM), and insulin (200 nM) were made by adding appropriate substances to Buffer 1 and stirring at 22°C for 1 h. The cholesterol stock solution was sonicated (in a bath-type sonicator) additionally for $ h. Insulin stock solution was allowed to sit overnight at room temperature (this was required for the insulin to go into solution) . PGEl (140 PM), arachidonic acid (250 PM), and 5-aHEAT (300 nM) stock solutions were made by drying under nitrogen appropriate volumes of ethanolic solutions of the respective substances, pumping under vacuum (1 h, 100 pm Hg) and then adding Buffer 1. Liposomes Lipid bilayers The appropriate at a temperature
were formed from dimyristoyl lecithin. lipid was added (1 mg/ml) to Buffer 1 above the lipid’s transition tempera-
Assay solutions of substances of interest were made by adding appropriate volumes of stock solution to Buffer 2. Desired amounts of luciferin-luciferase reagent (as a powder) were added. Corresponding ATP assay solutions were made by adding equivalent amounts of the substances of interest to ATP solution. Both solutions were incubated at room temperature for various times. The assay solution and ATP solution were then put in their appropriate reservoirs in the stopped-flow spectrophotometer. Each point reported was taken a minimum of six times and a minimum of two samples were prepared for each substance. RESULTS
AND
DISCUSSION
When luciferin-luciferase solution is mixed with ATP solution in the stopped-flow spectrophotometer, a rapid initial burst of light is observed (maximum light emission occurs at about 6 ms) followed by a slow decay (Fig. 1A). This is in accord with findings by DeLuca and McElroy (5). In order to maximize the sensitivity of the assay, light output for the luciferase-luciferin reaction was determined as a function of [M%+], pH, temperature, and [ATP]. The curves obtained were similar to those reported in the literature (9). Maximum light emission was found at 2 mM [M$‘], pH 7.4, and 23°C. [ATP] above 2 mM resulted in no enhancement of light production. Saturating the solutions with 0, resulted in no enhancement of light emission. In agreement with the findings of Karl and Holm-Hansen (lo), additional luciferin added to the Sigma luciferase-luciferin re-
306
NADER1
AND
MELCHIOR TABLE
1
Sensitivity of Luciferase-Luciferin Assay to Various Biological Materials Concentration (nM) inhbition Material Myristic acid Prostaglandin El Hexane Arachidonic acid Hexanol 5-Hydroxyeicosatetraenoic Palmitoyl CoA
Seconds FIG. 1. Photomultiplier output of light emission upon mixing luciferase-luciferin solution in (B) the presence and (A) the absence of 50 nM monopalmitoyl lecithin. Ordinate, full scale 1 s; abcissa, full scale 0.2 v.
acid (HEAT)
Adrenocorticotropin (ACTH) Ethanol 2-0-methyl-1-0-octadecylglycero-3phosphocholine (ALP) Monopalmitoyl lecithin (MPL) Monostearoyl lecithin (MSL) Hexanoic acid Neurotensin”
20%
80%
0.25 0.25 0.5 0.5 0.5 1 1
20 100 10 10 50 30 -
5 5
40 40
5 5 5 10 (50%
50 50 100 100 inhibition at
agent resulted in increased light emission. Over the range investigated (0.25 to 2 mg/ml, luciferase-luciferin reagent, approx 4.5-36 nM luciferase), light emission as measured by photomultiplier output was linear with respect to the concentration of enzymes and reactants. The normal working concentration of luciferase-luciferin reagent used was 1 mg/ml. The luciferase-luciferin reaction was found to be sensitive to the presence of a variety of amphipathic and hydrophobic substances (Table 1). The presence of these materials in the luciferase-luciferin solution results in reduced light emission. The quantitative nature of the reduction in light emission allows the luciferaseluciferin reaction to be used as an analytical means to quantitate these materials (analytes). Figure 1B shows the affect of 50 nM monopalmitoyl lecithin on the spectra of light emission relative to its absence (Fig. lA, control). Suppression of light emission was initially quantified in two different ways: (1) the ratio of total light output (area under the curve) relative to control (absence of analyte from the reaction) or (2) the ratio of maximum peak height relative to that of control. Peak height comparison was found to be the more precise as well as simpler method and was used throughout this study. A time course was carried out to determine the extent of light diminution as a function of incubation time of analyte in the luciferase-luciferin solution. For monopalmitoyl lecithin (50 nM), maximum light reduction was obtained in 20 min. This incubation time was used in the following studies. Figure 2 shows the affect on light output of varying submicellar concentrations of MPL (CMC, 2 X 10e6 M (18)). Reduced light emission as a function of analyte concentration follows a smooth curve. This type of dose-response curve is typical for
Substnace
P”
5 nM)
(50% inhibition at 100 nM) 0 0
Cholesterol Insulin
0 0
a Only one concentration measured. Note. The concentrations at which these materials gave 20 and 80% reduction of light emission relative to control (no analyte) are presented to indicate the sensitivity of luciferase-luciferin assay to individual analytes. All these materials gave smooth dose-response curves.
materials that inhibit light emission. Table 1 lists various substances tested. The concentrations at which these materials gave 20 and 80% reduction of light emission are presented to indicate the sensitivity of lucifer-
loon
204 0
8 10
" 20
8 30
Concentration
r 40
" 50
60
(nM)
FIG. 2. Sensitivity of luciferase-luciferin light emission molar concentrations of monopalmitoyl lecithin. Control monopalmitoyl lecithin, is taken as 100% light emission.
to nanovalue, no
307
01 0
IO
20
30
[Luciferin] 0
100
200
300
400
Concentration FIG. 3. Sensitivity of luciferase-luciferin lar concentrations of prostaglandin El. din El, is taken as 100% light emission.
500
600
(PM)
01 20
40
60
Concentration
50
60
70
kg/mL
the luciferase assay as a function of in the absence of (0) and in the preslecithin.
light emission to picomoControl value, no prostaglan-
ase-luciferin assay for these substances. From Table 1, it is apparent that the luciferase-luciferin assay can be used to detect subnanomolar concentrations of a variety of types of materials. As an example, Fig. 3 shows light reduction for prostaglandin El over the picomolar concentration range. In order to confirm an apparent correlation between the effectiveness of a substance in suppressing light production and its hydrophobic versus hydrophilic nature, we tested the series hexane, hexanol, and hexanoic acid. These compounds-an alkane, an alcohol, and a carboxylic acid-all have the same six carbon chain as a hydrophobic moiety, but are progressively more hydrophilic. As seen in Fig. 4, the sensitivity of the luciferase-luciferin assay for these materials is hexane > hexanol > hexanoic acid.
0
FIG. 5. Light emission for increasing amounts of luciferin ence of (0) 50 nM monopalmitoyl
40
60
100
120
(nM)
FIG. 4. Sensitivity of luciferase-luciferin light emission to hexane (o), hexanol (0), and hexanoic acid (w). Control value, no analyte, is taken as 100% light emission.
Our findings support the basis of this assay to be an interaction of analyte with the hydrophobic luciferin binding site of the luciferase molecule (6). Competition experiments were carried out to determine the V,, and K,,, of luciferase for luciferin in the presence and absence of an analyte, in this case 50 nM MPL. Figure 5 shows the results of this study. The data were analyzed by the method of nonlinear curve fitting (11). The correlation coefficient for the fit of control values was 0.982 and for data in the presence of MPL, 0.987. In the absence of MPL, V,, was 78.5 and K,,, 1.14 pg/ml, while in the presence of MPL, V,., was 79.7 and K,,, 4.62 pg/ ml. This is indicative of competitive inhibition between analyte and luciferin, most likely for a hydrophobic site on the luciferase molecule. Many amphipathic and hydrophobic materials partition from aqueous suspension into lipid bilayers. Lipid bilayers form the underlying structure of biological membranes. Proteins embedded in membranes are responsible for mediating many cellular processes (12). The partitioning of substances such as drugs, metabolites, and hormones from an aqueous milieu into the hydrophobic environment provided by bilayers can play an important regulatory role (13). It was of interest, therefore, to determine whether the luciferase-luciferin assay could be used to determine the transfer of substances other than anesthetics from aqueous suspension into lipid bilayers. Figure 5 shows the results of an investigation of this question. Light emission by the luciferase assay is unaffected by the presence of lipid bilayers. When luciferase-luciferin solution was incubated for 20 min with model bilayers formed from dimyristoyl lecithin (0.025 mg/ml), light output (+DML) was the same as in the absence of bilayers (CONTROL) (Fig. 6). The presence of 50 nM MPL in luciferase-luciferin solution devoid of bilayers results in a 45% reduction in light output (+MPL). When MPL was preincubated with dimyristoyl lecithin bilayers (0.025 mg/ml, 25’C for 20 min) prior to addi-
308
NADER1
3 E
AND
60
s -5 8
40
20
0
Control
+DML
+MPL
DML,MPL
Conditions FIG. 6. Ability of the luciferase-luciferin assay to detect the absorption of the amphipath monopalmitoyl lecithin by lipid bilayers. Light emission for CONTROL, absence of analyte and lipid bilayers; +DML, in the presence of dimyristoyl lecithin bilayers; +MPL, in the presence of 50 nM monopalmitoyl lecithin; DML, MPL in the presence of 50 nM monopalmitoyl lecithin incubated with dimyristoyl lecithin bilayers.
tion of luciferase-luciferin solution and the analysis carried out in the regular manner, light emission, (DML, MPL) was the same as the control value (CONTROL). This reflects the absence of MPL in the aqueous phase due to its having partitioned into the bilayers (19) where it is inaccessible to the luciferase. In summary, the luciferase-luciferin assay can provide a simple means for quantitating in vitro picomolar to micromolar concentrations of a variety of important biological substances. It provides a means for not only measuring aqueous concentrations in such materials but also for measuring their partitioning into lipid bilayers from the aqueous phase. The sensitivity of this assay varies with the analyte to be investigated (Table 1). Assay sensitivity is in to a large extent dependent on the specifics of the substance under investigation (Fig. 4). The sensitivity of the assay at a given luciferinluciferase reagent concentration decreases with increasing analyte concentration (Fig. 2). Maximum assay sensitivity is obtained at lower analyte concentrations and is limited by instrumental noise. Our instrument at its most sensitive setting can reproducibly distinguish +2% quenching. For prostaglandin El (Fig. 3) we can detect differences in concentrations of 25 PM. The reaction cell used in our instrument is 40 ~1. Theoretically, for a material such as prostaglandin El, we can optimally detect differences of 1 X 10-l’ mol. Realistically, this is not possible since about 0.5 ml of solution is necessary to load the spectrophotometer and make
MELCHIOR
several measurements. About 1.3 x lo-l4 mol of material is therefore a best possible minimum. As in working with low amounts of any materials, a possible source of error is loss of material to surfaces. In our apparatus, the material is exposed only to glass Teflon and quartz and we found no indication of loss of material. A variety of hydrophobic and amphipathic peptides, lipids, and small organic molecules are susceptible to this assay. While we, as others (5), have employed a commercial stopped-flow spectrophotometer to make quantitative measurements of the luciferin-luciferase reaction, they can be made by a variety of other means. Amongst those approaches used are custom made devices employing a photomultiplier tube whose amplified signal is fed to an oscilliscope or computer (8,14), an LKB-1250 luminometer (15), a custom designed luminometer (16), and an SAI Technology Model 3000 ATP photometer coupled to a Omniscribe recorder (9). REFERENCES 1. Tanford, C. (1980) The Hydrophobic 2. Small, D. M. (1986) The Physical New York.
Effect, Chemistry
Wiley, New York. of Lipids, Plenum,
3. Gennis, R. B. (1989) Biomembranes, Chapter 7, Springer-Verlag, New York. 4. Franks, N. P., and Lieb, W. R. (1986) Proc. N&l. Acad. Sci. USA 83,5116-5120. 5. DeLuca, M., and McElroy, W. D. (1974) Biochemistry 13, 921925. 6. DeLuca, M. (1969) Biochemistry 8,160-166. 7. Middleton,
A. J., and Smith,
E. B. (1976)
Proc. R. Sot. London
B
193,173-190. 8. Franks, N. P., and Lieb, W. R. (1984) Nature (London) 310,599601. 9. Webster, J. J., and Leach, F. R. (1980) J. Appl. Biochem. 2,469479. 10. Karl, D. M., and Holm-Hansen, 0. (1976) Anal. Btichem. 75, 100-112. 11. Press, W. H., Flannery, B. P., Teukolsky, S. A., and Vetterling, W. T. (1988) Numerical Recipes: The Art of Scientific Computing Cambridge Univ. Press, London. 12. Aloia, R. C., Curtain, C. C., and Gordon, L. M. (1988) Lipid DOmains and the Relationship to Membrane Function, A. R. Liss, New York. 13. Adam, A., and Delbruck, M. (1968) in Structural Biology and Molecular Biology (Davidson, N. and Rich, A., Eds.), pp- 198-215, Freeman, San Francisco, CA. 14. Thompson, A., Nigro, J., and Seliger, H. H. (1988) Biochem. Biophys. Res. Commun. 140,888-894. 15. Ugarova, N. N., Dukhovich, A. F., Shrets, S. V., Philippova, N. Y., and Berezin, I. V. (1987) Biochim. Biophys. Acta 921,465-472. 16. Lundin, A., and Thore, A. (1975) Anal. Biochem. 66,47-63. 17. Melchior, D. L., and Steim, J. M. (1976) Annu. Rev. Biochem. Bioeng. 6,205-238. 18. Welzien, H. U. (1979) Biochim. Biophys. Acta 772, 239-243. 19. Naderi, S., Carruthers, A., and Mel&or, D. L. (1989) Biochim. Biophys. Acta 985,173-181.