Environmental Geochemistry and Health 1994 16(3/4) page 229
Development of biosensors for the detection of mercury and copper ions David S. Holmes,* Santosh K. Dubey and Seema Gangolli Department of Biology, Faculty of Sciences, University of Chile, Casilla 653, Santiago, Chile Abstract
The development of genetically engineered biosensors for copper and mercury ions is described. The biosensors have been constructed by fusing the lux or light emitting genes from Vibrio fischeri with genetic regulating elements that respond to copper ions or mercury ions, derived respectively from Escherichia coil and Serratia marcescens. The fusions were placed into E. coli cells which then emitted light in response to copper or mercury ions. Data is presented describing the sensitivity, specificity, and dynamic range of the biosensors to their respective target metal ions. A preliminary description of experiments is provided indicating how these biosensors might be used to investigate the bioavailability of mercury and copper ions in environmental samples.
Introduction A biosensor is a detection or measuring device containing a biological molecule or whole cell, that acts as a sensor for a target analyte. Examples of biological sensors and their respective analytes include antibodies/antigens, enzymes/substrates, DNA p r o b e s / D N A s e q u e n c e s and w h o l e cells/metabolites. Advantages of using biological components as sensors include their sensitivity and specificity towards target analytes, especially if the targets are biological compounds. Thus, biosensors h a v e s p e c i a l a t t r a c t i o n s for use in the pharmaceutical and other biotechnology industries and in the h e a l t h , f o o d p r o c e s s i n g and environmental areas. However, despite their inbuilt sensitivity and specificity a general problem with the development of biosensors has been the difficulty in developing systems to transduce the recognition event of the sensor to a signal that can be readily collected and quantified. Signal transduction systems used routinely include detection of radioactivity as in ELISA, antibody/antigen recognition, autoradiography of DNA hybridisation reactions and in c h e m o l u m i n e s c e n t and colourometric determinations. Our approach to solving the problem of signal t r a n s d u c t i o n has been to use w h o l e cell bioluminescence to detect and quantify target analytes. The advantages of this approach include the ease, sensitivity and accuracy with which light can be recorded and measured. Also, measurements of biologically produced light can be made
*To whom correspondence should be addressed.
non-destructively, and approximately in real-time, permitting multiple or continuous readings to be taken in the same sample over time. The present paper describes recent experiments to develop microbial biosensors that report the concentration of mercury (Hg) and copper (Cu) ions by emitting light in response to these ions. Since the ions have to be taken up by the microbial cells before light is emitted these biosensors report bioavailability of the ions and not necessarily their absolute chemical concentration. The microbial biosensors were constructed as follows. A group of five genes (termed lux genes), encoding enzymes for bioluminescence were removed by genetic engineering from their natural host the marine bacterium Vibrio fischeri. (Figure 1). The removal also eliminated the natural genetic regulator that activates the expression of the five lux genes. Next, genetic regulators were removed from two microorganisms, Serratia marcescens and E. coli. These regulators normally control the e x p r e s s i o n of g e n e s r e s p o n s i b l e for Hg detoxification and Cu metabolism respectively. In separate experiments the regulators were placed next to the five lux genes in such a way that Hg or Cu ions activated the regulators resulting in the expression of the adjacent lux genes. The two constructions were introduced separately into a strain of E. coli resulting in microorganisms that emit light in the presence of Hg or Cu ions respectively. Simplified diagrammatic representations of these constructions are shown in Figure 1. More detailed descriptions have appeared elsewhere (Dubey et aL, 1993, and Holmes, et al.,1993). Experiments are described in this paper to test the sensitivity, specificity and dynamic ranges of these biosensors for their target analytes in
230
Development of biosensors
Genetic Element
Genetic
Lux Genes
Element
Lux Genes
I
T
Activated +2 by Hg
Light
Activated by Cu +or Cu+2
Light
lux B
lux E
B. Genetic Element
lux C
lux D
lux A
,,./
,,,,/
Aide hyde
Luciferase
RCHO + FMNH + O 2
Luciferase
2
Aide hyde
RCHO+FMN+H O+hv 2
Figure 1 A, Diagrammatic representations of the fusion of genetic elements next to the lux genes. The genetic elements activate the lux genes resulting in bioluminescence in response to Cu or Hg ions. B. Representation of the five structural lux genes encoding the luciferase enzymes and the enzymes responsible for the aldehyde substrate recycling. controlled laboratory conditions. Preliminary and exploratory experiments are also described that are designed to evaluate the feasibility of using biosensors to detect and quantify metal ions in environmental samples. Method
Plasmids pDU1003 (gift of Simon Silver. See Foster and Silver, 1987 for general review) and pUCD615 (gift of C.I. Kado) were used to construct the mercury biosensor plasmid termed pMerlux. Plasmids pA223 (gift of B. Lee. (See Rouch et al., 1989 for general review)) and pUCD615 were used to construct the copper biosensor plasmid termed pCulux. Plasmids were isolated by a mini-prep technique [Holmes and Quigley, 1981] and all manipulations of DNA were carried ou.t by standard techniques of molecular biology (Sambrook et al., 1989). Recombinant plasmids pMerlux and pCulux were transformed into E. coli JM 101. Transformed
cells were grown at 30~ in LB broth liquid medium or on LB solid medium (solidified with 1.5% agar) in the presence of 100 I.tg mL-1 ampicillin. Light emission from bacteria grown in liquid culture was recorded with a Bioorbit luminometer (Model 1250). Light emission was recorded in millivolts. Light emission from bacteria grown on solid media was recorded using an intensified CCD camera and an Argusl0 image processor, and analysed with an LG3 video capture board (Scion, Inc.) and the NIH image software (public domain) in a Macintosh IICX computer. Metal ions were added as HgC12 and CuSO4 salts and were dissolved in distilled water. Results
Copper (Cu) and mercury (Hg) biosensors were constructed by fusion of Hg or Cu inducible genetic elements (promoters) next to a group of five Iux genes that are responsible for producing biological
D. S. Holmes, S. K. Dubey and S. Gangolli
A
231
B E4
z=
O~
~- 2 (3 O= 0
4
0
.,J
3
l
~[-- Control i
0
I
1
;
I
,
Log of Conc of Cu
.21
I
2
Control ,
0
3
I
i
I
l
I
2 4 +2 6 Log of Conc of Hg (pM)
+2 (uM)
Figure 2 Log of light output from the Cu ion biosensor (A) or the Hg ion biosensor (B) in response to increasing concentrations of their respective ions. The control refers to the amount of light emitted by the biosensor in the absence of exogenous metal ion.
IB
A
HgC~:
5 O (/) 0
E
E
v
%..*"
"-
4
.~ 2
,iI
--4 le .
o
3
o
2
-J O
~
13' (.3
Z
~
-I 0
-2 Metal Salts ( 0.Im M )
Me~al Salts ( I uM )
Figure 3 Log of light output from the Cu ion biosensor (A) or the Hg ion biosensor (B) in response to different metal ions.
light. As a result of the fusion light production comes under the control of Hg or Cu ions (Figure 1). The biosensor microorganisms were grown in batch liquid cultures and light was induced with the appropriate metal ion at various times during their growth (data not shown). The timing of maximal light output was determined over a wide range of concentrations of the metal ions. This occurred at about mid-log in the growth cycle. Light output as a function of Hg or Cu ion concentration at mid-log is shown in Figure 2. In each case a control was included in which the biosensor cells were grown in the absence of added Hg or Cu ions. Maximum light output is approximately the
same for both biosensors and is in the range of 105 t o 106 millivolts. (106 millivolts corresponds to approximately 5 x 103 photons per second per cell). However, this maximal light output is reached after induction with 1 ktM Hg2+ compared to 103 lt]V[ Cu 2+. In addition, the Hg biosensor is capable of detecting down to about 0.1 nM Hg § 2 compared to 1~t M Cu 2+ for the Cu biosensor ( Figure 2). In the control, in the absence of added metal ions the Cu biosensor emits about 104 times more light than the Hg biosensor. There are probably several genetic and physiological reasons for this difference, which may result from the fact that copper is an essential element for metabolism whereas mercury is not. The reduced sensitivity of
232
Development of biosensors
Figure 4 A filter disc (A) or 0.1 of soil (B) was doped with 50 ~tL of O.1 mM Hg e+ and placed on to a lawn of biosensor microorganisms growing on nutrient agar. Light can be seen as a halo around the doped material. The light can be quantified using the NIH image analysis program. Photographs were taken using only the light emitted from the biosensors. the Cu biosensor may also result from the binding of Cu 2+ to ingredients in the medium used to grow the b i o s e n s o r m i c r o o r g a n i s m s , r e d u c i n g its bioavailability. This issue has not been investigated. The specificity of the biosensors was tested by challenging them with related metal ions by addition of the following salts to liquid cultures: COC12, CdSO4, NiSO4 or ZnCI2. The maximum light resulting from the addition of each metal ion is shown in Figure 3. The Hg biosensor emitted virtually no light above background for any of the metal ions tested. The Cu biosensor gave a very low response above background to Co z+ (0.5% of the light output compared with its response to Cu 2+) but virtually no response to the other metal ions tested. The Cu biosensor responds with equal sensitivity a n d specificity towards Cu + (CuzS) or Cu ~+ (CuSO4) (data not shown). These experiments were carried out on cultures of the biosensors grown in a liquid growth medium containing various potentially competing cations such as Na+ (0.08M), K+ (0.02M), Mg 2+ (2 x 10-3M) and Ca 2+ (10-3M). Clearly these ions do not interfere with the recognition of the biosensors towards their target metals. Experiments have recently been initiated to evaluate whether the biosensors can be used to detect target metals in soils, sludges and other environmental samples. In one such experiment the biosensor microorganisms were grown on solid nutrient media, and then exposed either to a filter disc or test soils doped with a known volume of metal ion solution. In both c a s e s a zone of bioluminescence is produced around the filter disc
or soil sample. An example of the Hg ion biosensor challenged with a test soil sample (100 mg) containing 0.1 mM Hg 2§ (27ppb) is shown in Figure 4. The position, size and intensity of the light can be quantified using an intensified CCD camera with supporting image analysis hardware and software,
Discussion The biosensors described in this paper are based on the concept of using microorganisms to detect Hg and Cu ions and to transduce the recognition into a quantifiable light signal. For a biosensor to be useful for practical applications it must be sufficiently sensitive and specific for its target analyte and it should be able to measure concentrations of the target analyte in a useful range. The Hg and Cu ion biosensors appear to have these minimum attributes. The Hg biosensor can detect as low as l nM Hg 2+ ion and the Cu biosensor can detect down to 1 gm Cu § or Cu 2+ ions. The maximum amount of the ions that can be detected before they become toxic to the microorganisms is about 1 gM Hg 2+ or lmM Cu § or Cu 2§ Thus the dynamic range of detection covers about four logs of concentration from 0.1 nM to 1 ~tM for the Hg biosensor and from 1 ~tM to lmM for the Cu biosensor. Both biosensors are very specific for their target analytes. The potentially competing metal ions tested induced no detectable light above background for the Hg biosensor and only Co induced a modest amount of light with the Cu biosensor (1/20 of the light output compared to Cu2+).
D. S. Holmes, S. K. Dubey and S. Gangolli
The mercury and copper biosensors reported here detect Hg and Cu ions in concentration ranges that are useful for environmental monitoring. Moreover, they appear to be sensitive and specific for their target metals and do not suffer interference from relatively high concentrations of potentially competing cations. A striking advantage of this approach is that the biosensors monitor metal ions in a nondestructive fashion and almost in real time. This could lead to the development of biosensors that could monitor compounds directly in the environment in real time, although clearly much additional research would be required to achieve this goal. A key concept is that the biosensors monitor the bioavailability of the metal ions to the biosensor microorganisms, since only metal ions taken up by those cells will result in light induction. A major thrust in our laboratory in the future will be to investigate whether these biosensors can be used to report the general bioavailability of Cu and Hg ions in environmental samples. Such samples could contain a wide variety of other microorganisms and, at times, potentially toxic compounds that could interfere with the ability of the biosensor cells to make measurements. Other laboratories are actively engaged in constructing microbial biosensors based upon the same principle as described in this paper. For e x a m p l e , a n a p h t a l e n e b i o s e n s o r has b e e n c o n s t r u c t e d w h i c h is c a p a b l e o f d e t e c t i n g naphtalene in contaminated soils (Burlage et al., 1990 and King et al., 1990). Other examples of biosensors include work of Guzzo et al., (1991), Selifonova et al., (1993) and Tescione and Belfort (1993). This work, together with our own results illustrate the broad potential for using this type of microbial biosensor for the evaluation of bioavailability in environmental samples.
Acknowledgements Work supported by grants from the National Science Foundation, Office of Naval Research, New York State Science and Technology Foundation, General Electric and the New York State Energy, Research and Development Authority.
References Burlage, R.S., Sayler, G.S. and Larimer, F. 1990. Monitoring of Naphtalene Catabolism by Bioluminescence with NahLux Transcriptional Fusions. J. Bacteriol, 172, 4749-4757
233
Dubey, S. K. Gangolli S.S. and Holmes, D.S. 1993. Development of Biosensors to Measure Metal Ion Bioavailability in Environmental Samples. N. E. GeoL Soc. Proc., 15, 188-194. Geiselhart, L., Osgood, M. and Holmes, D.S.,1991. Construction and Evaluation of a SelfLuminescent Biosensor. Ann. New York Acad. Sci , 646, 53-60. Foster, T.J and Silver, S. 1987. The Genetics and Biochemistry of Mercury resistance. CRC Crit. Rev. Microbiol, 15, 117-140. Guzzo, A., Dioro C. and Dubow, M.S. 1991. Transcription of the E. coli Gene is Regulated by Metal Ions. Appl. Environ. Microbiol., 57, 2255-2259. Holmes, D.S. and Quigley, M. 1981. A Rapid Method for the Preparation of Plasmids. Anal. Biochem., 114, 193-197. Holmes, D.S., Dubey, S.K. and Gangolli, S. 1983. Development of Biosensors to Measure Metal Ion Bioavailability in Mining and Metal Wastes, pp 659 - 666. In: Biohydrometallurgical Technologies. Vol. 2. Torma, A.E., Apel, M.L. and Brierley, C.L. (eds), TMS Warrendale, PA, USA. King, J.M.H., Digrazia, P.M., Applegate, B., Burlage, R., Sanseverino, J., Dunbar, P., Larimer, F. and Sayler G.S., 1990. Rapid Sensitive Bioluminescent Reporter Technology of Naphthalene Exposure and Biodegradation. Science, 249, 778-781. Nichols, A.B. 1991. Bioremediation: Potential and Pitfalls. Water Environ. Tech., 4, 52-56 (1991) Rouch, D., Lee, BTO and Camakaris, J. 1989. Genetic and Molecular Basis of Copper Resistance in E. Coli. In: Hamer, D.H. and Winge, D.R. (eds.), Metal Ion Homeostasis: Molecular Biology and Chemistry, pp. 439-446. Alan R. Liss, New York. Sambrook, J., Fritch, E. and Maniatis, T. 1989. Molecular Cloning, 2nd edn. Cold Spring Harbor
Labs. Press, New York. Selifonova, O., Burlage, R. and Barkay, T. 1993. Bioluminescent Sensors for Detection of Bioavailable Hg (II) in the Environment. Appl. Environ. Microbiol., 49, 3083-3090. Tescione, L. and Belfort, G. 1993. Construction and Evaluation of a Metal Ion Biosensor. Bio. Eng., 42, 945-952. (Manuscript No.317: accepted after revision October 10, 1993).
Development of biosensors for the detection of mercury and copper ions David S. Holmes,* Santosh K. Dubey and Seema Gangolli Department of Biology, Faculty of Sciences, University of Chile, Casilla 653, Santiago, Chile Abstract
The development of genetically engineered biosensors for copper and mercury ions is described. The biosensors have been constructed by fusing the lux or light emitting genes from Vibrio fischeri with genetic regulating elements that respond to copper ions or mercury ions, derived respectively from Escherichia coil and Serratia marcescens. The fusions were placed into E. coli cells which then emitted light in response to copper or mercury ions. Data is presented describing the sensitivity, specificity, and dynamic range of the biosensors to their respective target metal ions. A preliminary description of experiments is provided indicating how these biosensors might be used to investigate the bioavailability of mercury and copper ions in environmental samples.
Introduction A biosensor is a detection or measuring device containing a biological molecule or whole cell, that acts as a sensor for a target analyte. Examples of biological sensors and their respective analytes include antibodies/antigens, enzymes/substrates, DNA p r o b e s / D N A s e q u e n c e s and w h o l e cells/metabolites. Advantages of using biological components as sensors include their sensitivity and specificity towards target analytes, especially if the targets are biological compounds. Thus, biosensors h a v e s p e c i a l a t t r a c t i o n s for use in the pharmaceutical and other biotechnology industries and in the h e a l t h , f o o d p r o c e s s i n g and environmental areas. However, despite their inbuilt sensitivity and specificity a general problem with the development of biosensors has been the difficulty in developing systems to transduce the recognition event of the sensor to a signal that can be readily collected and quantified. Signal transduction systems used routinely include detection of radioactivity as in ELISA, antibody/antigen recognition, autoradiography of DNA hybridisation reactions and in c h e m o l u m i n e s c e n t and colourometric determinations. Our approach to solving the problem of signal t r a n s d u c t i o n has been to use w h o l e cell bioluminescence to detect and quantify target analytes. The advantages of this approach include the ease, sensitivity and accuracy with which light can be recorded and measured. Also, measurements of biologically produced light can be made
*To whom correspondence should be addressed.
non-destructively, and approximately in real-time, permitting multiple or continuous readings to be taken in the same sample over time. The present paper describes recent experiments to develop microbial biosensors that report the concentration of mercury (Hg) and copper (Cu) ions by emitting light in response to these ions. Since the ions have to be taken up by the microbial cells before light is emitted these biosensors report bioavailability of the ions and not necessarily their absolute chemical concentration. The microbial biosensors were constructed as follows. A group of five genes (termed lux genes), encoding enzymes for bioluminescence were removed by genetic engineering from their natural host the marine bacterium Vibrio fischeri. (Figure 1). The removal also eliminated the natural genetic regulator that activates the expression of the five lux genes. Next, genetic regulators were removed from two microorganisms, Serratia marcescens and E. coli. These regulators normally control the e x p r e s s i o n of g e n e s r e s p o n s i b l e for Hg detoxification and Cu metabolism respectively. In separate experiments the regulators were placed next to the five lux genes in such a way that Hg or Cu ions activated the regulators resulting in the expression of the adjacent lux genes. The two constructions were introduced separately into a strain of E. coli resulting in microorganisms that emit light in the presence of Hg or Cu ions respectively. Simplified diagrammatic representations of these constructions are shown in Figure 1. More detailed descriptions have appeared elsewhere (Dubey et aL, 1993, and Holmes, et al.,1993). Experiments are described in this paper to test the sensitivity, specificity and dynamic ranges of these biosensors for their target analytes in
230
Development of biosensors
Genetic Element
Genetic
Lux Genes
Element
Lux Genes
I
T
Activated +2 by Hg
Light
Activated by Cu +or Cu+2
Light
lux B
lux E
B. Genetic Element
lux C
lux D
lux A
,,./
,,,,/
Aide hyde
Luciferase
RCHO + FMNH + O 2
Luciferase
2
Aide hyde
RCHO+FMN+H O+hv 2
Figure 1 A, Diagrammatic representations of the fusion of genetic elements next to the lux genes. The genetic elements activate the lux genes resulting in bioluminescence in response to Cu or Hg ions. B. Representation of the five structural lux genes encoding the luciferase enzymes and the enzymes responsible for the aldehyde substrate recycling. controlled laboratory conditions. Preliminary and exploratory experiments are also described that are designed to evaluate the feasibility of using biosensors to detect and quantify metal ions in environmental samples. Method
Plasmids pDU1003 (gift of Simon Silver. See Foster and Silver, 1987 for general review) and pUCD615 (gift of C.I. Kado) were used to construct the mercury biosensor plasmid termed pMerlux. Plasmids pA223 (gift of B. Lee. (See Rouch et al., 1989 for general review)) and pUCD615 were used to construct the copper biosensor plasmid termed pCulux. Plasmids were isolated by a mini-prep technique [Holmes and Quigley, 1981] and all manipulations of DNA were carried ou.t by standard techniques of molecular biology (Sambrook et al., 1989). Recombinant plasmids pMerlux and pCulux were transformed into E. coli JM 101. Transformed
cells were grown at 30~ in LB broth liquid medium or on LB solid medium (solidified with 1.5% agar) in the presence of 100 I.tg mL-1 ampicillin. Light emission from bacteria grown in liquid culture was recorded with a Bioorbit luminometer (Model 1250). Light emission was recorded in millivolts. Light emission from bacteria grown on solid media was recorded using an intensified CCD camera and an Argusl0 image processor, and analysed with an LG3 video capture board (Scion, Inc.) and the NIH image software (public domain) in a Macintosh IICX computer. Metal ions were added as HgC12 and CuSO4 salts and were dissolved in distilled water. Results
Copper (Cu) and mercury (Hg) biosensors were constructed by fusion of Hg or Cu inducible genetic elements (promoters) next to a group of five Iux genes that are responsible for producing biological
D. S. Holmes, S. K. Dubey and S. Gangolli
A
231
B E4
z=
O~
~- 2 (3 O= 0
4
0
.,J
3
l
~[-- Control i
0
I
1
;
I
,
Log of Conc of Cu
.21
I
2
Control ,
0
3
I
i
I
l
I
2 4 +2 6 Log of Conc of Hg (pM)
+2 (uM)
Figure 2 Log of light output from the Cu ion biosensor (A) or the Hg ion biosensor (B) in response to increasing concentrations of their respective ions. The control refers to the amount of light emitted by the biosensor in the absence of exogenous metal ion.
IB
A
HgC~:
5 O (/) 0
E
E
v
%..*"
"-
4
.~ 2
,iI
--4 le .
o
3
o
2
-J O
~
13' (.3
Z
~
-I 0
-2 Metal Salts ( 0.Im M )
Me~al Salts ( I uM )
Figure 3 Log of light output from the Cu ion biosensor (A) or the Hg ion biosensor (B) in response to different metal ions.
light. As a result of the fusion light production comes under the control of Hg or Cu ions (Figure 1). The biosensor microorganisms were grown in batch liquid cultures and light was induced with the appropriate metal ion at various times during their growth (data not shown). The timing of maximal light output was determined over a wide range of concentrations of the metal ions. This occurred at about mid-log in the growth cycle. Light output as a function of Hg or Cu ion concentration at mid-log is shown in Figure 2. In each case a control was included in which the biosensor cells were grown in the absence of added Hg or Cu ions. Maximum light output is approximately the
same for both biosensors and is in the range of 105 t o 106 millivolts. (106 millivolts corresponds to approximately 5 x 103 photons per second per cell). However, this maximal light output is reached after induction with 1 ktM Hg2+ compared to 103 lt]V[ Cu 2+. In addition, the Hg biosensor is capable of detecting down to about 0.1 nM Hg § 2 compared to 1~t M Cu 2+ for the Cu biosensor ( Figure 2). In the control, in the absence of added metal ions the Cu biosensor emits about 104 times more light than the Hg biosensor. There are probably several genetic and physiological reasons for this difference, which may result from the fact that copper is an essential element for metabolism whereas mercury is not. The reduced sensitivity of
232
Development of biosensors
Figure 4 A filter disc (A) or 0.1 of soil (B) was doped with 50 ~tL of O.1 mM Hg e+ and placed on to a lawn of biosensor microorganisms growing on nutrient agar. Light can be seen as a halo around the doped material. The light can be quantified using the NIH image analysis program. Photographs were taken using only the light emitted from the biosensors. the Cu biosensor may also result from the binding of Cu 2+ to ingredients in the medium used to grow the b i o s e n s o r m i c r o o r g a n i s m s , r e d u c i n g its bioavailability. This issue has not been investigated. The specificity of the biosensors was tested by challenging them with related metal ions by addition of the following salts to liquid cultures: COC12, CdSO4, NiSO4 or ZnCI2. The maximum light resulting from the addition of each metal ion is shown in Figure 3. The Hg biosensor emitted virtually no light above background for any of the metal ions tested. The Cu biosensor gave a very low response above background to Co z+ (0.5% of the light output compared with its response to Cu 2+) but virtually no response to the other metal ions tested. The Cu biosensor responds with equal sensitivity a n d specificity towards Cu + (CuzS) or Cu ~+ (CuSO4) (data not shown). These experiments were carried out on cultures of the biosensors grown in a liquid growth medium containing various potentially competing cations such as Na+ (0.08M), K+ (0.02M), Mg 2+ (2 x 10-3M) and Ca 2+ (10-3M). Clearly these ions do not interfere with the recognition of the biosensors towards their target metals. Experiments have recently been initiated to evaluate whether the biosensors can be used to detect target metals in soils, sludges and other environmental samples. In one such experiment the biosensor microorganisms were grown on solid nutrient media, and then exposed either to a filter disc or test soils doped with a known volume of metal ion solution. In both c a s e s a zone of bioluminescence is produced around the filter disc
or soil sample. An example of the Hg ion biosensor challenged with a test soil sample (100 mg) containing 0.1 mM Hg 2§ (27ppb) is shown in Figure 4. The position, size and intensity of the light can be quantified using an intensified CCD camera with supporting image analysis hardware and software,
Discussion The biosensors described in this paper are based on the concept of using microorganisms to detect Hg and Cu ions and to transduce the recognition into a quantifiable light signal. For a biosensor to be useful for practical applications it must be sufficiently sensitive and specific for its target analyte and it should be able to measure concentrations of the target analyte in a useful range. The Hg and Cu ion biosensors appear to have these minimum attributes. The Hg biosensor can detect as low as l nM Hg 2+ ion and the Cu biosensor can detect down to 1 gm Cu § or Cu 2+ ions. The maximum amount of the ions that can be detected before they become toxic to the microorganisms is about 1 gM Hg 2+ or lmM Cu § or Cu 2§ Thus the dynamic range of detection covers about four logs of concentration from 0.1 nM to 1 ~tM for the Hg biosensor and from 1 ~tM to lmM for the Cu biosensor. Both biosensors are very specific for their target analytes. The potentially competing metal ions tested induced no detectable light above background for the Hg biosensor and only Co induced a modest amount of light with the Cu biosensor (1/20 of the light output compared to Cu2+).
D. S. Holmes, S. K. Dubey and S. Gangolli
The mercury and copper biosensors reported here detect Hg and Cu ions in concentration ranges that are useful for environmental monitoring. Moreover, they appear to be sensitive and specific for their target metals and do not suffer interference from relatively high concentrations of potentially competing cations. A striking advantage of this approach is that the biosensors monitor metal ions in a nondestructive fashion and almost in real time. This could lead to the development of biosensors that could monitor compounds directly in the environment in real time, although clearly much additional research would be required to achieve this goal. A key concept is that the biosensors monitor the bioavailability of the metal ions to the biosensor microorganisms, since only metal ions taken up by those cells will result in light induction. A major thrust in our laboratory in the future will be to investigate whether these biosensors can be used to report the general bioavailability of Cu and Hg ions in environmental samples. Such samples could contain a wide variety of other microorganisms and, at times, potentially toxic compounds that could interfere with the ability of the biosensor cells to make measurements. Other laboratories are actively engaged in constructing microbial biosensors based upon the same principle as described in this paper. For e x a m p l e , a n a p h t a l e n e b i o s e n s o r has b e e n c o n s t r u c t e d w h i c h is c a p a b l e o f d e t e c t i n g naphtalene in contaminated soils (Burlage et al., 1990 and King et al., 1990). Other examples of biosensors include work of Guzzo et al., (1991), Selifonova et al., (1993) and Tescione and Belfort (1993). This work, together with our own results illustrate the broad potential for using this type of microbial biosensor for the evaluation of bioavailability in environmental samples.
Acknowledgements Work supported by grants from the National Science Foundation, Office of Naval Research, New York State Science and Technology Foundation, General Electric and the New York State Energy, Research and Development Authority.
References Burlage, R.S., Sayler, G.S. and Larimer, F. 1990. Monitoring of Naphtalene Catabolism by Bioluminescence with NahLux Transcriptional Fusions. J. Bacteriol, 172, 4749-4757
233
Dubey, S. K. Gangolli S.S. and Holmes, D.S. 1993. Development of Biosensors to Measure Metal Ion Bioavailability in Environmental Samples. N. E. GeoL Soc. Proc., 15, 188-194. Geiselhart, L., Osgood, M. and Holmes, D.S.,1991. Construction and Evaluation of a SelfLuminescent Biosensor. Ann. New York Acad. Sci , 646, 53-60. Foster, T.J and Silver, S. 1987. The Genetics and Biochemistry of Mercury resistance. CRC Crit. Rev. Microbiol, 15, 117-140. Guzzo, A., Dioro C. and Dubow, M.S. 1991. Transcription of the E. coli Gene is Regulated by Metal Ions. Appl. Environ. Microbiol., 57, 2255-2259. Holmes, D.S. and Quigley, M. 1981. A Rapid Method for the Preparation of Plasmids. Anal. Biochem., 114, 193-197. Holmes, D.S., Dubey, S.K. and Gangolli, S. 1983. Development of Biosensors to Measure Metal Ion Bioavailability in Mining and Metal Wastes, pp 659 - 666. In: Biohydrometallurgical Technologies. Vol. 2. Torma, A.E., Apel, M.L. and Brierley, C.L. (eds), TMS Warrendale, PA, USA. King, J.M.H., Digrazia, P.M., Applegate, B., Burlage, R., Sanseverino, J., Dunbar, P., Larimer, F. and Sayler G.S., 1990. Rapid Sensitive Bioluminescent Reporter Technology of Naphthalene Exposure and Biodegradation. Science, 249, 778-781. Nichols, A.B. 1991. Bioremediation: Potential and Pitfalls. Water Environ. Tech., 4, 52-56 (1991) Rouch, D., Lee, BTO and Camakaris, J. 1989. Genetic and Molecular Basis of Copper Resistance in E. Coli. In: Hamer, D.H. and Winge, D.R. (eds.), Metal Ion Homeostasis: Molecular Biology and Chemistry, pp. 439-446. Alan R. Liss, New York. Sambrook, J., Fritch, E. and Maniatis, T. 1989. Molecular Cloning, 2nd edn. Cold Spring Harbor
Labs. Press, New York. Selifonova, O., Burlage, R. and Barkay, T. 1993. Bioluminescent Sensors for Detection of Bioavailable Hg (II) in the Environment. Appl. Environ. Microbiol., 49, 3083-3090. Tescione, L. and Belfort, G. 1993. Construction and Evaluation of a Metal Ion Biosensor. Bio. Eng., 42, 945-952. (Manuscript No.317: accepted after revision October 10, 1993).
