Asssment of heavy metal bioavailability in contaminated diments and soils using green fluorescent protein-bad bacterial bionsors
Vivian Hsiu-Chuan Liao *,Ming-Te Chien,Yuen-Yi Tng,Kun-Lin Ou
Department of Bioenvironmental Systems Engineering,National Taiwan University,No.1,Sec.4,Roovelt Road,Taipei 106,Taiwan,ROC
Received 28May 2005;received in revid form 21September 2005;accepted 29September 2005
Nonpathogenic GFP-bad bacterial bionsor is applicable in determining
the bioavailability of heavy metals in environmental samples.
Abstract
A green fluorescent protein (GFP)-bad bacterial bionsor Escherichia coli DH5a (pVLCD1)was developed bad on the expression of gfp under the control of the cad promoter and the cadC gene of Staphylococcus aureus plasmid pI258.DH5a (pVLCD1)mainly responded to Cd(II),Pb(II),and Sb(III),the lowest detectable concentrations being 0.1nmol L ÿ1,10nmol L ÿ1,and 0.1nmol L ÿ1,respe
ctively,with 2h exposure.The bionsor was field-tested to measure the relative bioavailability of the heavy metals in contaminated diments and soil samples.The results showed that the majority of heavy metals remained adsorbed to soil particles:Cd(II)/Pb(II)was only partially available to the bionsor in soil e water extracts.Our results demonstrate that the GFP-bad bacterial bionsor is uful and applicable in determining the bioavailability of heavy metals with high nsitivity in contaminated diment and soil samples and suggests a potential for its inexpensive application in environmentally relevant sample tests.Ó2005Elvier Ltd.All rights rerved.
Keywords:Green fluorescent protein;Bacterial bionsor;Bioavailable heavy metals;Contaminated diments and soils
1.Introduction
Environmental contamination by heavy metals is a world-wide problem.It is important to aware the possible effects of increasing levels of environmental heavy metals pollution on human health and the environment.Therefore,it is neces-sary to develop nsitive,effective,and inexpensive methods which can efficiently monitor and determine the prence and amount of hazardous metals in the environment.Tradi-tionally,the environmental risk caud by heavy metal pollu-tion is determined by
quantification of total metals after digestion with strong acids by using conventional analytical methods such as atomic absorption spectrometry and ion
chromatography.Additionally,before using the analytical methods,environmental samples require laborious treatment to solubilize the metal ions from the solid matrix (i.e.soils or diments).However,conventional analytical methods are not able to distinguish between available (potentially hazard-ous)and non-available (potentially non-hazardous)fractions of metals to biological systems.This is of particular interest with respect to solid soils,becau of the great adsorption capability of heavy metals to solid pha (Vanhala and Ahtiainen,1994).Moreover,the main drawback of chemical methods is the question of the transfer of the re-sults obtained on nonbiological systems to the biological ones.Becau the bioavailability of metals in environmental sam-ples is not a constant value,but varies with changing environ-mental conditions,it may be a crucial issue in environmental monitoring.To detect the bioavailable fraction of certain metals,veral approaches have been followed.Of which,
*Corresponding author.Tel.:þ886233665239;fax:þ886233663462.E-mail address:vivianliao@ntu.edu.tw (V .H.-C.Liao).
财政投资评审0269-7491/$-e front matter Ó2005Elvier Ltd.All rights rerved.
doi:10.vpol.2005.09.021
one approach is bad on the u of bacteria that are genetically engineered so that a measurable signal is produced when the bacteria are in contact with bioavailable metal ions.In nsor bacteria,expression of a reporter gene is controlled by a metal-responsive regulatory element,which usually originates from bacteria that are naturally resistant to a particular heavy metal. The regulatory element can be coupled to a reporter gene through a gene fusion that upon expression produces a readily measurable signal in respon to the particular metal.Thus,in the prence of the particular heavy metal,the amount of re-porter protein inside the cell increas.Hence,the amount of a given metal was detected by measuring the reporter protein produced by the nsor bacteria.
Several metal-specific bacterial nsors for the detection of bioavailable metals have been developed(Corbisier et al., 1993;Selifonova et al.,1993;Tauriainen et al.,1998;Biran et al.,2000;Ivask et al.,2001).Of which,bacterial bionsors for Cd(II)/Pb(II)have been previously described mostly utiliz-ing reporter genes such as lacZ,lux,and luc in the transcrip-tional fusion constructs(Tauriainen et al.,1998;Riether et al.,2001;Shetty et al.,2003).Although the colorimetric en-zyme assay and bioluminescence have been very successful as a reporter for Cd(II)/Pb(II)detection in their studies,the de-tection methods require addition of exogenous substrates or cofactors for signal pr
oduction.The gene for greenfluorescent protein(GFP)from the jellyfish Aequoria Victoria(Chalfie et al.,1994)is increasingly being ud as a reporter gene, although it has not been ud extensively as a reporter for measuring biologically relevant concentrations of pollutants. GFPfluorescence is stable and can be monitored non-invasively in living cells.GFP is also an attractive reporter system be-cau it is easy to u and does not require any exogenous sub-strates or cofactor.The u of GFP as a reporter protein in the bacterial bionsing system therefore can obviate the need for centrifugation,cell lysis,pH adjustment,and subquently kinetic enzyme activity measurements.
In this work,we describe the construction of a nonpathogenic Escherichia coli whole-cell bionsor for the detection of Cd(II),Pb(II),and Sb(III)by employing red-shifted GFP (rs-GFP)as a reporter protein.The nsor plasmid is bad on the expression of rs-GFP under the control of the cad pro-moter and the cadC gene of the cadA resistance determinant of Staphylococcus aureus plasmid pI258(Nucifora et al.,1989; Tynecka et al.,1981).In the abnce of an effector,the expres-sion of gfp gene is represd.Gene expression is induced and fluorescence can be measured in the prence of the effector. Moreover,despite of many different metal bionsors avail-able little attention has been drawn to the u of the bion-sors for the analysis of environmental samples.Therefore,the feasibility of the bacterial bionsor for measuring bioavail-able metals in environmental samples ha
s not been well tested. To demonstrate the usability of the GFP-bad bionsor,we describe the u of the GFP-bad bionsor to measure bio-available fractions of metals in metal-contaminated diment and agricultural soil samples.The feasibility of using such a strain to analyze the bioavailability of pollutants in the envi-ronment is also discusd.2.Materials and methods
2.1.Chemicals
Unless otherwi stated,all chemicals ud were analytical reagent grade or better and were purchad from Sigma e Aldrich(St.Louis,MO,USA).All media and buffer solutions were prepared using deionized distilled water (Barnstead,Dubuque,IA,USA).Restriction endonucleas and T4DNA liga were supplied from New England Biolabs(Beverly,MA,USA).The DNA polymera ud in polymera chain reaction(PCR)was from Qiagen (Qiagen,Hilden,Germany).
2.2.Construction of bionsor plasmid
Recombinant plasmid was constructed as a transcriptional fusion.Plasmid pI258isolated from S.aureus(NCTC50581;National Collection of Type Cul-tures,Colindale,London,UK)was ud as a template for PCR to generate the 572ba pairs DNA fragment consisting of the promoter/operator of the cad operon and cadC gene.PCR primers were designed with either Eco RI(for-ward primer)or
Bam HI(rever primer)recognition quence extensions. DNA amplification was carried out in an automated thermal cycler(Eppen-dorf,Hamburg,Germany).The amplified PCR product was purified using QIAquick PCR purification kit(Qiagen).The purified PCR-amplified DNA fragment was digested with Eco RI and Bam HI and was purified from an aga-ro gel by QIAEX II gel extraction kit(Qiagen).Subquently,the fragment was cloned into the Eco RI and Bam HI sites of pPROBE-NT#(Miller et al., 2000).The resulting recombinant plasmid,pVLCD1(Fig.1),was transformed li DH5a by the CaCl2competent cell method.
2.3.Cultivation of bacteria and induction of GFPfluorescence
by effectors
A single colony li harboring pVLCD1was grown overnight in Luria e Bertani(LB)media supplemented with50m g mLÿ1of kanamycin at 37 C.The overnight culture was diluted100-fold in fresh L
B medium supple-mented with50m g mLÿ1kanamycin and incubated at37
C in an orbital shaker at225rpm until the optical density at600nm(OD600)reached0.6.Var-ious conce
ntrations of Cd(II),Pb(II)or Sb(III)were added to2-mL aliquots of bacterial cultures.Optical density of cultures at600nm and thefluorescent in-tensity produced by the bacteria were measured.At least three independent ex-periments were performed for each effector.
2.4.Measurement of GFPfluorescence in culture
The transcriptional activity of the bionsor was estimated by the measure-ment of the GFPfluorescence of cells grown in LB medium containing a range of different metal ions or time periods.Cell growth was monitored by the mea-surement of optical density at600nm with a spectrophotometer(Eppendorf). Thefluorescence of GFP-producing cells that were grown in culture was mea-sured using a VersaFluor Fluorometer that wasfitted with a490Æ5nm exci-tationfilter and a510Æ5nm emissionfilter(Bio-Rad,Hercules,CA,USA).
(4)
O/P cadC GFP T1Km r
Fig.1.Schematic organization of the bionsor plasmid pVLCD1.Plasmid harbors genes required for replication(rep)and mobilization(mob).Abbrevi-ation:km r,gene encoding kanamycin resistance;T1,Escherichia coli rrnB rRNA T1terminator,GFP,greenfluorescent protein.The diagram is not drawn to scale.
18V.H.-C.Liao et al./Environmental Pollution142(2006)17e23
is defined as the rawfluorescence intensity expresd in RFU divided by the optical density at600nm measured at each time point.At least triplicate meas-urements were obtained for each sample.
2.5.Selectivity studies
The induction of the nsing system by a variety of metal ions,including As(III),Co(II),Cu(II),Fe(II),Hg(II),Mn(II),Ni(II),Sn(II),Cd(II),Pb(II), Sb(III),and Zn(II)was studied by measuring the greenfluorescence produced. Each metal ion(1m mol Lÿ1)was added to bacterial nsor culture at a cell density of0.6OD600.The cells were incubated for2h at37 C,and then the spec
ificfluorescence intensity was measured as described above.At least three independent experiments were performed for each kind of metal ion and mixtures of metal ion assays.
2.6.Testing of contaminated diment and soil samples with bacterial bionsor
Environmental samples were collected from veral canals and agricultural lands,known to have heavy metals contamination,from Chunghua County, Taiwan in August of2003.Before the analysis,the samples were air-dried and sieved to2mm.Soil e water extracts were prepared by mixing the air-dried soil with deionized water using the soil e water ratio of1/9(w/v).Subquently, the suspensions were shaken at room temperature for24h,followed by centri-fugation at13,000g for10min,and then the supernatants(soil e water extracts)were ud for chemical and bionsor analys.
For chemical analysis,concentrations of Cd(II),Pb(II)in water extracts of soil samples were determined with inductively coupled plasma atomic emis-sion spectroscopy analyzer(PerkinElmer3000SC,Norwalk,CT,USA).Certi-fied standards(PerkinElmer)were run with every determination.
For bionsor assay,the environmental sample was tested by adding 500m L of soil e water extracts
开学班会sample to250m L of6Âconcentrate of LB medium,10m L of LB medium,and740m L of DH5a cells harboring the pVLCD1plasmid in LB medium at a cell density of0.6OD600.The cells were incubated for2h at37 C,and then the specificfluorescence intensity was measured using the procedures described above.Samples containing known concentrations of Cd(II)in place of the10-m L portion of LB medium were tested in parallel with500m L of deionized,distilled laboratory water in place of environmental sample to generate a standard curve.Standard curve was derived from linear regression of the averagefluorescence value at each particular Cd(II)concentration,and then the concentrations of Cd(II)equiva-lent in the environmental samples were calculated from the standard curve.In order to examine possible inhibitory effects onfluorescence resulting from chemicals besides the effector compounds in the environmental sample, 10m mol Lÿ1of Cd(II)was added to the environmental sample,and the green fluorescence emission was compared to that for a positive control containing the same concentration of Cd(II)in deionized water.
2.7.Data analysis
The experiments were performed at least three times for error analys. The data were ud to calculate the standard deviations,reprented by error bars in thefigures.Student’s t test analysis at a¼0.05level was performed to check results for significance.Standard curvefits were done by linear r
egres-sion analysis.
3.Results
3.1.Description of the bacterial bionsor
The cad promoter and the cadC gene of the cadA resistance determinant of S.aureus plasmid pI258was cloned into the broad-host-range vector pPROBE-NT#(Miller et al.,2000) upstream from the gfp gene,creating a P cad-gfp transcriptional fusion that was designated pVLCD1as shown in Fig.1.When DH5a cells were transformed with pVLCD1recombinant plasmid,GFPfluorescence was obrved in respon to Cd(II),Pb(II),Zn(II),and Sb(III)resulting in a statistically significant increa in thefluorescence intensity relative to that of cells with no-effector control.A reprentative green fluorescence image of the bionsor strain expod to an effec-tor is shown in Fig.2.
The lectivity of the bacterial bionsor to metal ions was evaluated.The bacterial cells harboring pVLCD1plasmid were treated with1m mol Lÿ1of various metal ions for2h prior tofluorescence measurements as described in Section2.The levels offluorescence of the nsing system subjected to the metal ions are also plotted in Fig.3.In our experimental treat-ments,a positive respon was obrved for Cd(II),Pb(II), Sb(III)and Zn(II).No statistically significant change in green fluorescence w
as obrved for As(III),Co(II),Cu(II),Fe(II), Hg(II),Mn(II),Ni(II),Sn(II)compared to the control as shown in Fig.3.
Since pVLCD1responds to more than one effector,and since contaminated sites often contain multiple metal species, the respon of the bacterial bionsing system to the mixtures of the prent metals was investigated.To examine this poten-tially complex situation,we performed pairwi metal assays with the bionsor.Each metal ion(1m mol Lÿ1)
was Fig.2.Fluorescence of bionsor strain carrying pVLCD1expod to Sb(III). DH5a(pVLCD1)was treated with1m mol Lÿ1of Sb(III)for2h at37 C in Luria e Bertani(LB)medium,after which greenfluorescent protein expression in bacterial cells was visualized by using epifluorescence microscope.Images were captured by using a cooled charge coupled device(CCD)camera.(A)No Sb(III)treatment,(B)bacterial cells treated with Sb(III).Magnification,Â1000.
19
V.H.-C.Liao et al./Environmental Pollution142(2006)17e23
combined in the same treatment.Additionally,the respons from mixtures of effectors:Cd(II),Pb(II),Zn(II),and Sb(III)were also examined to test whether the individual effector acted in an additive manner.As shown in Fig.3,the effec-tors appear to act in an additive manner.The level of green fluorescence for cells treated with other metal combinations other than the aforementioned effector combinations was not significantly different from that of control or effector alone (data not shown).
3.2.Time-dependent induction of green fluorescence with effectors
Time-dependent induction of the bacterial nsor in respon to Cd(II),Pb(II),and Sb(III)ion was determined by incubating the cells with metal ions for various time intervals as described in Section 2.Although the bionsor showed re-spon to Zn(II)but to a lesr extent,thus we did not generate time-dependent curve for Zn(II).The induction of green fluo-rescence of the DH5a (pVLCD1)strain toward the exposure of the metal ions showed a time-dependence (Fig.4).As shown in Fig.4,as the time of induction incread,there was an increa in the green fluorescence emitted by the bac-terial DH5a (pVLCD1)strain.Interestingly,DH5a (pVLCD1)bionsor readily responds to Sb(III)with the highest induc-tion efficiency.The background fluorescence exhibited by the untreated bionsors did not have any statistically signifi-cant fluorescence change during the incubation period (data not shown).The kinetic profile of the bionsor respon also showed that during the first 5e 6h of incubation,the
specific fluorescence intensity continuously incread from the background value (Fig.4).
3.3.Do-dependent induction of green fluorescence with effectors
The do e respon relationship of DH5a cells harboring the pVLCD1plasmid was examined for the effectors:Cd(II),Pb(II),and Sb(III)as described in Section 2.For assay develop-ment,a 2-h induction p
eriod was chon since the green fluorescence signal obtained during this time period was sufficiently high enough.Moreover,a 2-h incubation also allows the complete formation of the GFP fluorophore.The fluorescence intensity incread with increasing concentrations of Cd(II),Pb(II),and Sb(III)ions in the sample.Plots of the do e respon relationships of the bionsor to the effectors as measured by fluorometer were shown in Fig.5.As shown in Fig.5,the intensity of fluorescence signal emitted
incread
50
100
150
200
250
A s
C o
C u
F e
H g
M n
N i
S n
C d
P b
S b
Z n
C d +P b
C d +S b
C d +Z n
空的成语P b +S b
P b +Z n娓娓动听什么意思
S b +Z n
冷暖锋C d +P b +S b +Z n
Metal ions
I n d u c t i o n i n t e n s i t y ( )
高考试卷全国都一样吗
Fig.3.Selectivity of the bacterial bionsor to metal ions.DH5a (pVLCD1)was treated with 1m mol L ÿ1of various individual metal ions or mixtures of metal ions for 2h.Induction intensity (in %)is defined as value of culture spe-cific fluorescence (in SFI)with metal treatment minus culture specific fluores-cence (in SFI)of control then divided by culture specific fluorescence (in SFI)of control.Specific fluorescence (in SFI)was measured as described in Section 2.Control refers to no-metal treatment bionsor bacteria.The data prented here are the mean values of at least three independent experiments with the standard deviations.
010002000300040005000
60007000
8000
S F I
我很伤心01000
20003000400050006000
70008000
9000
S F I
010000
20000
3000040000
5000060000
Induction time (h)
S F I
Fig.4.Time-dependent induction of green fluorescence with effectors.The DH5a cells harboring the p
VLCD1plasmid were expod to 4m mol L ÿ1Cd(II),Pb(II),or Sb(III),and the specific fluorescence intensity (in SFI)was determined after different exposure periods.Fluorescence (in SFI)measured with a fluorometer is defined as culture fluorescence divided by culture at a cell density of optical density at 600nm.The data prented here are the mean values of at least three independent experiments with the standard deviations.The SFI scales for the panels are different.
20
V.H.-C.Liao et al./Environmental Pollution 142(2006)17e 23
with the concentrations of Cd(II),Pb(II),and Sb(III)to a certain level.At concentrations lower than the detection limit of an effector,binding of cadC to the cad O/P quence represd transcription and translation of rs -GFP.Addition of Cd(II),Pb(II),and Sb(III)ions de-represd rs -GFP in the cells.
The detection limits for Cd(II),Pb(II),and Sb(III)obrved in our study are well below the metal ions concentrations typ-ically regulated water standard.For Cd(II),0.1nmol L ÿ1(0.01m g L ÿ1)Cd(II)was necessary to induce a statistically significant change (p <0.05)of gfp expression.The intensity of green fluorescence incread with increasing amount of Cd(II)to a concentration of 750m mol L ÿ1,after which the fluorescence started to decrea.This might be due to the toxicity of Cd(II)ions to t
he bacterial cells.For Sb(III),0.1nmol L ÿ1(0.01m g L ÿ1)Sb(III)was necessary to induce statistically significant change (p <0.05)of gfp expression and 10m mol L ÿ1Sb(III)caud a maximum gfp induction.Toxic effect was also noted for Sb(III)at concentration greater than 10m mol L ÿ1.For Pb(II),the lowest concentration re-quired to induce a statistically significant change (p <0.05)of gfp expression was 10nmol L ÿ1(2.07m g L ÿ1).At concen-tration of 10m mol L ÿ1Pb(II),gfp expression was induced to a maximal level.Toxic effect was not obrved for Pb(II)at concentrations between 10and 250m mol L ÿ1.3.4.Bioavailability of heavy metals
To demonstrate the utility of this bionsor in measuring actual environmental contamination,diment and agricultural soil samples with a known contaminant concentration were ex-amined and the results of the bionsor assays were compared to tho known concentrations.Sediment and agricultural soil samples were collected from contaminated sites in region of Chunghua County,Taiwan in August of 2003.The bioavail-able fraction of the metals was determined from water extract of diment and soil by using the DH5a (pVLCD1)bionsor.The standard curve was generated with known concentration of Cd(II),and the resulting equation (y ¼1.622x þ2441.9,r 2¼0.9961)was ud to calculate the Cd(II)equivalent con-centrations of the samples.By calculating the contaminant concentrations from the standard curve and taking the diluti
on factor of the assay into account (e Section 2),the final concentrations of contaminants in the samples were shown in Table 1.Becau we could not differentiate between possi-ble effectors,the data were expresd as Cd(II)equivalents.Table 1also compares the results of acid-soluble,water-soluble,and bioavailable fractions of heavy metals in diment and soil samples by analyzing same batches of environmental samples.Results prented in Table 1suggest that Cd/Pb contents of the soil samples,as measured by chemical method,are only partially available to E.coli DH5a (pVLCD1)bionsor.
Additionally,possible inhibitory effects that might be caud by chemicals besides the effector compounds in the en-vironmental sample were also assd by spiking the sample with a known concentration of Cd(II)(10m mol L ÿ1).Sub-quently the total SFI was measured and then compared to that for a positive control containing the same concentration in deionized water.No inhibitory effect was detected in this study.Therefore,it is unlikely that the constituents besides the effector compounds in the samples interfered with GFP fluorescence.
20004000600080001000012000
恒作文600字14000160001800020000
nmol L -1nmol L -1nmol L -1nmol L -1nmol L -1µmol L -1µmol L -1µmol L -1mmol L -1mmol L -1
Metal concentration
S F I
Fig.5.Do-dependent induction of green fluorescence by effectors.Fluorescence from DH5a cells harboring the pVLCD1plasmid was determined after 2-h incubation with various concentrations of m
etal ions,as described in Section 2.Fluorescence (in specific fluorescence intensity [SFI])measured with a fluorometer is defined as culture fluorescence divided by culture at a cell density of optical density at 600nm.The data prented here are the mean values of at least three independent experiments with the standard deviations.
21
V.H.-C.Liao et al./Environmental Pollution 142(2006)17e 23
