- Email: [email protected]
Molecular quantification of genes encoding for green-fluorescent proteins
Journal of Microbiological Methods 52 (2003) 297 – 304 www.elsevier.com/locate/jmicmeth
Molecular quantification of genes encoding for green-fluorescent proteins A. Felske a,*, V. Vandieken a,b, B.V. Pauling a, H.F. von Canstein a, I. Wagner-Do¨bler a a
GBF (German Research Center for Biotechnology), Division of Microbiology, Mascheroder Weg 1, D-38124 Braunschweig, Germany b Max-Planck-Institute fu¨r Marine Microbiology, Celsiusstraße 1, D-28359 Bremen, Germany Received 21 May 2002; received in revised form 14 June 2002; accepted 3 September 2002
Abstract A quantitative PCR approach is presented to analyze the amount of recombinant green fluorescent protein (gfp) genes in environmental DNA samples. The quantification assay is a combination of specific PCR amplification and temperature gradient gel electrophoresis (TGGE). Gene quantification is provided by a competitively coamplified DNA standard constructed by point mutation PCR. A single base difference was introduced to achieve a suitable migration difference in TGGE between the original target DNA and the modified standard without altering the PCR amplification efficiency. This competitive PCR strategy is a highly specific and sensitive way to monitor recombinant DNA in environments like the efflux of a biotechnological plant. D 2002 Elsevier Science B.V. All rights reserved. Keywords: gfp; Competitive PCR; TGGE
1. Introduction Emerged from the deep sea, the green fluorescent protein (gfp) from the jellyfish Aqueorea victoria (Prasher et al., 1992) is now also spreading worldwide on the surface. Its gene, transferred by numerous human ‘vectors’, is introduced in more and more different organisms like various bacterial strains, yeast, Drosophila, C. elegans, zebrafish, Xenopus, mice, many plant species and even human cell lines. At the moment, the thriving of gfp is restricted to the
* Corresponding author. Tel.: +49-531-618-1408; fax: +49-531618-1411. E-mail address: [email protected] (A. Felske).
small area of more or less isolated laboratories, but it is only a question of time when gfp-carrying organisms are released deliberately or accidentally into uncontrolled environments. The application of gfp mainly served the investigation of the structure and dynamics of macromolecules in living cells and had quickly become a primary tool for analysis of DNA and protein localization (van Roessel and Brand, 2002). Strong fluorescence by external illumination and activity independently from external substrates made it an attractive and a versatile reporter. The selfcontained domain structure of gfp reduces the possibility of inhibition by cellular factors of the host (Yang et al., 1996). Also, the design of fused proteins is feasible, neither disturbing gfp fluorescence nor the activity of the protein to which it is fused (Rizzuto et
0167-7012/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 7 0 1 2 ( 0 2 ) 0 0 1 8 4 - 7
298
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
al., 1996). The stability of gfp also allows detection of its fluorescence after the producing host has died, like for instance, during protein purification or cell fixation for other staining protocols. Localization of gfp fusion proteins can be analyzed in live cells, enabling real-time monitoring of dynamics in situ (Romberg et al., 1998). Such analyses have been further enhanced by the selection of brighter, more soluble gfp variants (Heim et al., 1995; Crameri et al., 1996; Haseloff, 1999). Hence, there is little doubt that gfp will continue to gain ground and segregate into a high diversity of fluorogenic enzymes, and all this directed by the hands of humans. The gfp had become a domestic enzyme, transported and replicated in environments far beyond the reach of the original host A. victoria. Here, we report on the design of a molecular method to detect gfp on the DNA level. This is useful to monitor genetically engineered microbes (GEM) not only in their active state, but also in distorted biosamples where activity and cellular integrity had largely ceased. The most sensitive and also highly specific molecular test is the PCR. We used quantitative PCR to count down to very low target numbers by competitive coamplification of a standard molecule of known concentration (Freeman et al., 1999). It was the aim to design the standard with only one base difference to the original sequence for a temperature gradient gel electrophoresis (TGGE)-based detection system (Felske et al., 2001). In contrast, the original competitive PCR approach used standards of different length (Wang et al., 1989), which might, however, lack a consistent amplification efficiency (Stolovitzky and Cecchi, 1996). Here, we apply a new competitive PCR/TGGE protocol to quantify the release of gfp genes from mercury reduction biocatalysers inoculated with gfp-labeled mercury reducing bacteria.
2. Materials and methods 2.1. Strain cultivation and DNA extraction The genome of the strain Pseudomonas putida KT2442Dmer73Dgfp contained independently introduced DNA fragments carrying the merTPAB operon and the gfp gene as described before (Felske et al.,
2001). Both were constitutively expressed. The strain was grown in Luria –Bertani medium (trypton 10 g l 1, yeast extract 5 g l 1, NaCl 10 g l 1) at 30 jC for 16 h. Cells from a 5-ml culture were harvested by centrifugation at 2000 g for 10 min. The DNA was extracted from the bacteria pellet as previously described (Pitcher et al., 1989) and the DNA concentration measured via the saran wrap method (Sambrook et al., 1989). 2.2. Biocatalyser set-up and sampling Neutralized and aerated chloralkali wastewater (Hg2 + concentration between 5 and 9 mg l 1) was pumped with 160 ml h 1 into the biocatalyser (at approximately 25 jC) in upflow mode and supplemented with 0.1 g l 1 sucrose and 0.02 g l 1 yeast extract. The catalyser was a glass column (200 ml volume, 4.8 cm internal diameter), wherein, above a grid, approximately 80 cm3 pumice was placed (core diameter 2.5 – 3.5 mm, Raab, Neuwied, Germany). Columns and tubing were sterilized by autoclaving (121 jC, 20 min). Each sampling day, 51 ml efflux was collected. One milliliter was used for cfu counting, where 50 Al efflux and serial 10-fold dilutions were plated on rich agar medium (4 g l 1 sucrose, 2 g l 1 yeast extract and 15 g l 1 NaCl). Three replicates were prepared and checked for growth after a week. The suspended matter of the remaining 50 ml was harvested by centrifugation at 2000 g for 30 min. The pellet was forwarded to DNA extraction and quantification as described above. 2.3. Primer selection and standard construction The amplified DNA stretch used for TGGE analysis covered the positions 573 – 721 of the gfp gene (EMBL accession number X83959). The primers for standard construction and production covered a little larger stretch. The simulation of DNA denaturation was computed with the PC-software Poland V1.3 (Steger, 1994), the graphs were made with Excel 97 (Microsoft). The default parameters were used for the computation without adjusting to electrophoresis-specific conditions. The TGGE primers were GC-GFPf (5V-[CGCCCCCGCCGCCCCGCCGCCCGCCGCCCC]-GCCCTGTCCTTTTACCAGAC-3V) and GFPr (5V-CCATGTGTAATCCCAGCAGC-3V).
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
The GC-clamp, which is recommended for TGGE analysis (Abrams et al., 1990), is given in brackets. The GFP primer sequences are printed bold, and the PRO primer sequences are underlined to clarify the nested arrangement of the primer sets. The primers for introducing the point mutation (-C-) were MUTf (5V-CTCCAATTGGCGATGGCCCTGTC-3V) and MUTXr (5V-GTATAGTTCATCCATGCCATGTGTAATCCCAGCAGCTGTTACAAACTCAAGA-cGACCA-3V). Another primer pair was used to produce standard template DNA with or without mutation, the primers PROf (5V-GCGATGGCCCTGTCCTTTTACCAGAC-3V) and PROr (5VCATCCATGCCATGTGTAATC-3V). Amplification was performed with a Mastercycler Gradient (Eppendorf, Hamburg, Germany) using 30 cycles of 94 jC for 10 s, 56 jC for 20 s and 68 jC for 20 s. The PCR (20 Al) contained 10 mM Tris –HCl (pH 8.3), 50 mM KCl, 3 mM MgCl2, 100 AM each of dATP, dCTP, dGTP, and dTTP, 0.4 AM each of one forward and one reverse primer, 0.5 units of recombinant Taq DNA polymerase (LifeTechnologies, Paisley, UK) and 1 Al template. This PCR with MUT primers on genomic DNA of P. putida KT2442Dmer73Dgfp (1 ng Al 1) has been used to introduce the mutation. It was performed in serial 10-fold template DNA dilution to minimize template carry-over. Strong MUT-PCR products were obtained until dilution step 10 3 (1 pg Al 1). This product has been purified with the NucleoSpinR kit (Macherey-Nagel, Du¨ren, Germany) according to the instructions of the manufacturer. The same PCR protocol, but with PRO primers produced standard molecules by reamplification of the MUT product. Also, this PRO-PCR was performed in serial 10-fold template DNA dilution to minimize the contribution of original genomic DNA. Strong PRO-PCR product was obtained until dilution step 10 5. This PCR product was purified with the NucleoSpinR kit and used as standard (DNA concentration was measured as described above). The corresponding diluted template of purified MUT-PCR products were stored for future standard production. 2.4. Competitive PCR and TGGE Each competitive PCR experiment consisted of seven reactions of decreasing gradients of standard
299
DNA (400 fg and serial 3-fold dilutions, i.e. 1 106, 3.3 105, 1.1 105, . . . molecules; samples with very low DNA concentration were coamplified with 40 or 4 fg input and corresponding dilutions). The second competitor was a DNA sample representing 1 ml of the catalyser efflux, i.e. 1 Al of 50 Al DNA solution per reaction. The seven-reactions-mastermix for a competitive PCR assay was composed for PCR as described above, but with 5 Al template volume per reaction (1 Al sample + 4 Al standard): The catalyser efflux DNA was added in 7 Al to the seven-reactionsmastermix, and after distribution of the mastermix to the seven reaction tubes (7 16 Al), the DNA standards of 4 Al/tube were added. The PCR conditions were the same as described above. The Biometra TGGE maxi system (Biometra, Go¨ttingen, Germany) was used for separation of amplicons after PCR (2 Al PCR product input for TGGE). Electrophoresis took place in a 0.8-mm polyacrylamide gel (6% w/v acrylamide, 0.1% w/v bis-acrylamide, 8 M urea, 20% v/v formamide, 2% v/v glycerol) with 1 MN buffer (20 mM MOPS, 10 mM NaOH, pH f 7.0) at a fixed voltage of 5 V for 10 min (prerun to establish the temperature gradient) and then at 400 V for 2 h. A temperature gradient was built-up in electrophoresis direction from 30 to 45 jC. Silver-stained gels (Sanguinetti et al., 1994) were scanned with a GS700 Densitometer (Bio-Rad, Hercules, USA) and analyzed with Scion Image 3b software (Scion, Frederick, USA). In every competitive PCR assay, the lanes were analyzed, where the band intensities of the standard and the DNA template were most similar. Pixel (Pv) volumes of the band-images were quantified, and the original chromosome amount (C) in the genomic DNA extraction was calculated with these values and the standard molecule amount (St): C = PvC PvSt 1 St. Since the DNA input represented 1 ml catalyser efflux, the GEM chromosome number ml 1 catalyser efflux could be calculated. Efflux DNA preparation concentrations have been translated into chromosome equivalents with regards to the P. putida chromosome size of 6 Mbp (RamosDı´az and Ramos, 1998): 1 ng genomic DNA approximates 1.5 104 chromosomes. One standard PROPCR product molecule of 142 bp with 78.8 kg mol 1 was equivalent to one GEM chromosome (10 ng standard DNA = 7.6 1010 GEM chromosome equivalents).
300
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
2.5. Perpendicular TGGE The sample for perpendicular TGGE was a mixture of 100 Al gfp PCR product from the standard (with mutation) and 100 Al gfp PCR product from the P. putida KT2442Dmer73Dgfp genomic DNA applied along a slot covering almost the whole gel width. TGGE was done for 15 min without temperature gradient (28 jC), then interrupted to set-up a 28 –44 jC gradient in perpendicular direction to electrophoresis and continued for further 2 h (other TGGE parameters as above). In contrast to normal parallel TGGE, this TGGE variant did not require careful gradient optimization. Therefore, the 30 –45 jC mentioned before was the optimal gradient for parallel TGGE.
with the PRO primers and the MUT – PCR product as DNA template (10 5 dilution) was based on maximum 2000 MUT template molecules with virtually no genomic DNA remains (4 10 6 chromosomes Al 1). 3.2. Competitive PCR and TGGE The simulation of the GFP-PCR product migration speed at different temperatures and the experimental confirmation via perpendicular TGGE demonstrated a changed denaturation behavior (Fig. 1), resulting in
3. Results and discussion 3.1. Competitive PCR standard production Competitive PCR requires a concentration standard which is easily distinguishable from the target DNA. TGGE separates amplicons by their denaturation progress along a growing temperature gradient. The disproportionate DNA double strand stability, caused by the higher stability of C u G pairs compared to T = A pairs, could be used to manipulate the melting behavior. This effect could be simulated with the computer software Poland V1.3 in the search for the most suitable site to introduce a point mutation (data not shown). The point mutation was introduced 17 bp ahead of the GFPr 3V-end and exchanges an A by a C. The mutation-introducing MUTXr primer was covering the GFPr sequence for reamplification of the MUT product with the GFPr or PRO primers. The amplification of serial dilutions of genomic DNA from the P. putida KT2442Dmer73Dgfp11 with the MUT primer pair succeeded with a DNA input of as low as 1 pg (approximately 160 chromosomes, data not shown). This MUT –PCR product again could be reamplified with the PRO primers down to a template dilution of 10 6 (1 Al template in 20 Al reaction volume). Given the 200-fmol Al 1 nucleotide input (see above), the PCR could theoretically reach a concentration of approximately 2 108 amplicons Al 1. Thus, the routine standard DNA production
Fig. 1. Denaturation of the gfp PCR fragment with (st) and without (or) mutation. (a) The computer simulation shows a clear discrimination of the two fragments in the temperature range from 69 to 73 jC. (b) Perpendicular TGGE shows the predicted discrimination of the fragments, but temperatures were decreased by using urea and formamide in the gel to protect it from drying. st: standard product. or: genomic DNA product. ss: single-stranded DNA.
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
separated TGGE bands of the competitive PCR (Fig. 2). The high temperatures required to force the melting (Fig. 1a) could be reduced by approximately 36 jC by using the denaturing gel ingredient urea and formamide (Fig. 1b) to protect the gel from drying. There are two types of heteroduplexes visible in Fig. 2: st+/or and or+/st . Among them, the difference changing the melting behavior is a T p C mismatch compared to a weak GA match at the mutation site. TGGE may resolve such minor differences under favorable conditions, which were intentionally provided by selecting a mutation site critical for denaturation of the amplicons. Thus, we detected four bands: Firstly stops heteroduplex or+/st with one T p C mismatch, second is heteroduplex st+/or with weak G u A match, third comes homoduplex or+/ or with T = A match and the last molecule stopping on TGGE is the homoduplex st+/st with strong G u C match. Heteroduplex abundance is always limited by the yield of the less abundant amplicons because a heteroduplex requires one strand of each amplicons. Thus, heteroduplexes were most abundant in the reactions where both amplicons are equal. Further, only these reactions with the most similar amplicons amounts for standard and target are used for quantification (Fig. 2, marked with *), because reactions with growing discrepancy of standard and target are suffering increasing bias by the C0t effect (Mathieu-Daude´ et al., 1996). We also applied DNA samples without gfp for PCR control, yielding only TGGE bands of the standard and no heteroduplexes. This method could easily be transferred to an application in real-time PCR machines (Heid et al., 1996), which may detect melting temperature varia-
301
tions directly during PCR (Ririe et al., 1997). Where such machines are available, this method could be applied even more straightforward with high sample throughput and probably also more accurate. Nevertheless, the cumbersome good old TGGE also had advantages by providing a clearer separation of PCR products and eventually accompanying side products and heteroduplexes. There was also no need for concerns about primer specificity, since only amplicons of correct length and sequence would end on the standard and target positions of the TGGE gel. 3.3. Demonstration of gfp efflux monitoring from a biotechnical device A biocatalyser for mercury remediation was inoculated with a mercury-reducing gfp-labeled P. putida KT2442Dmer73Dgfp11. According to our experience, heavily loaded industrial wastewater like our sample (from chlor alkali electrolysis) typically contained 1 –10 mg l 1 Hg2 + (von Canstein et al., 1999). Our biocatalysers are designed for them as end-of-pipe solutions. While contaminated environments may contain approximately 1000-fold lower mercury concentrations, the mercury concentrations in the industrial wastewater were absolutely lethal to organisms without mercury resistance ( = mercury reduction). Therefore, the P. putida KT2442Dmer73 is constitutively overexpressing the merTPAB operon resulting in elevated levels of mercuric reductase activity (Horn et al., 1994). The biocatalyser outflow was analyzed over 94 days to compare the competitive PCR method with cfu counts and another competitive PCR approach targeting the merTPAB operon insertion region (Felske et
Fig. 2. Silver-stained TGGE gel with competitive PCR amplicons on it. The seven reactions on the left hand were from day 94 and the ones on the right hand from day 88. The lanes used for quantification are marked with *.
302
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
Fig. 3. Microscopic picture of a pumice surface after 3 days of growth, taken with a CLS microscope. The young growing cell layer of the gfp-labeled strain is visible as bright rod-shaped cells. Grey background signals are reflections from the pumice surface.
al., 2001). Direct microscopic observation of the gfplabeled GEM succeeded only in the early biofilmformation stages on the pumice filling in the reactor
(Fig. 3). When the biofilm became more dense by growth and the subsequent accumulation of dead biomass, single cells could hardly be distinguished. Torn fragments of this biofilm flushing out of the biocatalyser showed weak fluorescence, which could not be appointed, to single cells, hence, the origin of this fluorescence remained unproved (data not shown). The visualization of gfp-containing bacteria was much simpler with the cultivation approach. Putting Petri dishes under UV light revealed gfp presence as greenglowing colonies. The cfu quantification yielded a broad variation over three orders of magnitude across the whole period (Fig. 4). Irregular biomass release from the biocatalyser caused great variation of total cfu counts and DNA extraction yields over several magnitudes. However, the cultivation approach led to a massive underestimation of biomass and gfp release. The DNA extraction yields were translated into chromosome equivalents, following the simplification that the chromosome size of all bacteria in the system is the same like the GEM (6 Mbp). This estimation was rough, but good enough to serve our interpretations. The comparison to the total DNA yield was chosen to add a relative estimation to the absolute values and indicated up to two magnitudes more bacteria than detectable by cfu count (Fig. 4). Hence, we could
Fig. 4. Comparison of the GEM and invading bacteria in the catalyser efflux by cfu counts and DNA quantification. The dotted columns are the cfu counts of all bacteria and the only fluorescent cfu counts of the gfp-labeled GEMs. Some cfu counts were below 1000 ml 1, the detection limit of the used cfu counting, and therefore neglected. The white columns represented the DNA extraction yield translated into chromosomal equivalents (1 equivalent = 6.7 fg DNA). The stripped columns are the amount of merTPAB operons of the GEM as quantified by a previously published competitive PCR/TGGE protocol. The black columns are the numbers of gfp genes as quantified by the gfp competitive PCR/TGGE protocol.
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
demonstrate sometimes a much higher biomass efflux than indicated by cfu counts. This is more supported, but contraindicated by the fact that probably not all DNA could be extracted from the biofilm fragments glued by polysaccharides. Cell quantification via microscopic counts or biomass quantification would have been valuable, but was hampered by the distorted state of the (ex-) biofilm aggregates leaving the reactor. The comparison of gfp gene numbers to the total DNA extraction yield correlated well to the ratio of total cfu to the gfp-labeled cfu. Both results showed that the gfp-labeled strain has been challenged by contaminants from the nonsterile wastewater input. Further, the gfp quantification gave almost identical results to the strain-specific merTPAB operon insert region quantification. Mercury resistance and gfp of P. putida KT2442Dmer73Dgfp11 were both introduced into the host genome by genetic engineering, but independently. Thus, both genes are located in different areas of the genome and detected completely independently. While the gfp method described here was specific for gfp and its variants, the merTPAB operon method was exclusively designed for P. putida KT2442Dmer73: by composing one primer for the introduced merA gene (mercuric reductase) and the other for the adjacent genome (Felske et al., 2001). The little deviation of both values can best be explained by the limited resolution of the competitive PCR set-up. The three-fold dilution of standards and the subsequent densitometric measurement of electrophoresis bands are likely to be the most limiting steps for a higher resolution. 3.4. The labeled strain and the intruders The biofilm within the biocatalyser 3 days after the inoculation (Fig. 4) was clearly made of the gfplabeled strain as shown by cultivation and DNA analysis. In the beginning, all isolated colonies were fluorescent and the quantitative PCRs gave values almost identical to the estimation by DNA yield. However, from day 7 on the quantitative PCR results showed one (day 11) to three (day 59) magnitudes lower amounts of gfp and merTPAB operons than bacterial genomes (indicated by DNA yield). Also the cultivation approach yielded from this day on nonfluorescent colonies. Analysis of these colonies showed that they were contaminants (data not
303
shown). These intruders probably contaminated the nonsterile wastewater reservoir through the air and entered via this route the biocatalysers (WagnerDo¨bler et al., 2000). Since they were also highly efficient mercury reducers, they could successfully challenge the P. putida KT2442Dmer73Dgfp11. Also interesting is the period from days 45 to 59, when cfu counts dropped below reliable quantification. Appearing almost sterile, the biocatalyser output from these days may suggest the death of the bacterial community. However, the DNA yield was not reduced, indicating the continued presence of the biofilm. Consequently, the performance of the system was not ceased, because the biocatalyser depended more on the presence of the bacteria than their growth activity.
4. Conclusions The gfp is a useful marker to follow the ways of labeled microbes in living ones, but with this approach, also in dead ones. The study of dissemination of such common markers will serve to improve security measures for GEM use. The high sensitivity of the PCR quantification is applicable to monitor contamination events. Microbes released deliberately or by leakage accidents would probably be rapidly inactivated in a low-nutrient water, sediment or soil body. Thus, gfp fluorescence would fade while the GEMs, or their DNA, remain intact. Hidden in a huge diversity of environmental bacteria, the GEM may still be tracked back by its DNA. Even transformation events may be followed, since any other organism picking up the gfp gene should give a positive signal as well. The ease and efficiency of natural bacterial dissemination will surely lead to increasing appearance of gfp in the environment. Our method will help to assess this process in a straightforward way.
Acknowledgements This work was supported by a grant from a European Communities EC project (QLK3-CT-199901213). A.F. was supported by a grant from the European Communities EC project BACREX (QLK32000-01678).
304
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
References Abrams, E.S., Murdaugh, S.E., Lerman, L.S., 1990. Comprehensive detection of single base changes in human genomic DNA using denaturing gradient gel electrophoresis and a GC clamp. Genomics 7, 463 – 475. Crameri, A., Whitehorn, E.A., Tate, E., Stemmer, W.P., 1996. Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat. Biotechnol. 14, 315 – 319. Felske, A., Pauling, B.V., von Canstein, H.F., Li, Y., Lauber, J., Buer, J., Wagner-Do¨bler, I., 2001. Molecular monitoring of genetically improved mercury reducing bacteria by competitive PCR. BioTechniques 30, 142 – 148. Freeman, W.M., Walker, S.J., Vrana, K.E., 1999. Quantitative RTPCR: pitfalls and potential. BioTechniques 26, 112 – 125. Haseloff, J., 1999. Gfp variants for multispectral imaging of living cells. Methods Cell Biol. 58, 139 – 151. Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Real time quantitative PCR. Genome Res. 6, 986 – 994. Heim, R., Cubitt, A.B., Tsien, R.Y., 1995. Improved green fluorescence. Nature 373, 663 – 664. Horn, J.M., Brunke, M., Deckwer, W.-D., Timmis, K.N., 1994. Pseudomonas putida strains which constitutively overexpress mercury resistance for biodetoxification of organomercurial pollutants. Appl. Environ. Microbiol. 60, 357 – 362. Mathieu-Daude´, F., Welsh, J., Vogt, T., McClelland, M., 1996. DNA rehybridization during PCR: the ‘C0t effect’ and its consequences. Nucleic Acids Res. 24, 2080 – 2086. Pitcher, D.G., Saunders, N.A., Owen, R.J., 1989. Rapid extraction of bacterial genomic DNA with guanidium thiocyanate. Lett. Appl. Microbiol. 8, 151 – 156. Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G., Cornier, M.J., 1992. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229 – 233. Ramos-Dı´az, M.A., Ramos, J.L., 1998. Combined physical and genetic map of the Pseudomonas putida KT2440 chromosome. J. Bacteriol. 180, 6352 – 6363.
Ririe, K.M., Rasmussen, R.P., Wittwer, C.T., 1997. Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Anal. Biochem. 245, 154 – 160. Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R.Y., Pozzan, T., 1996. Double labelling of subcellular structures with organelle-targeted gfp mutants in vivo. Curr. Biol. 6, 183 – 188. Romberg, L., Pierce, D.W., Vale, R.D., 1998. Role of the kinesin neck region in processive microtubule-based motility. J. Cell Biol. 140, 1407 – 1416. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Saran wrap method. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, New York. appendix E6. Sanguinetti, C.J., Dias Neto, E., Simpson, A.J.G., 1994. Rapid silver staining and recovery of PCR products separated on polyacrylamide gels. BioTechniques 17, 915 – 919. Steger, G., 1994. Thermal denaturation of double-stranded nucleic acids: prediction of temperatures critical for gradient gel electrophoresis and polymerase chain reaction. Nucleic Acids Res. 22, 2760 – 2768. Stolovitzky, G., Cecchi, G., 1996. Efficiency of DNA replication in the polymerase chain reaction. Proc. Natl. Acad. Sci. U. S. A. 93, 12947 – 12952. van Roessel, P., Brand, A.H., 2002. Imaging into the future: visualizing gene expression and protein interactions with fluorescent proteins. Nat. Cell Biol. 4, E15 – E20. von Canstein, H., Li, Y., Timmis, K.N., Deckwer, W.-D., WagnerDo¨bler, I., 1999. Removal of mercury from chloralkali electrolysis wastewater by a mercury-resistant Pseudomonas putida strain. Appl. Environ. Microbiol. 65, 5279 – 5284. Wagner-Do¨bler, I., Lu¨nsdorf, H., Lu¨bbehu¨sen, T., von Canstein, H.F., Li, Y., 2000. Structure and species composition of mercuryreducing biofilms. Appl. Environ. Microbiol. 66, 4559 – 4563. Wang, A.M., Doyle, M.V., Mark, D.F., 1989. Quantification of mRNA by the polymerase chain reaction. Proc. Natl. Acad. Sci. U. S. A. 86, 9717 – 9721. Yang, F., Moss, L.G., Phillips, G.N., 1996. The molecular structure of green fluorescent protein. Nat. Biotechnol. 14, 315 – 319.
Molecular quantification of genes encoding for green-fluorescent proteins A. Felske a,*, V. Vandieken a,b, B.V. Pauling a, H.F. von Canstein a, I. Wagner-Do¨bler a a
GBF (German Research Center for Biotechnology), Division of Microbiology, Mascheroder Weg 1, D-38124 Braunschweig, Germany b Max-Planck-Institute fu¨r Marine Microbiology, Celsiusstraße 1, D-28359 Bremen, Germany Received 21 May 2002; received in revised form 14 June 2002; accepted 3 September 2002
Abstract A quantitative PCR approach is presented to analyze the amount of recombinant green fluorescent protein (gfp) genes in environmental DNA samples. The quantification assay is a combination of specific PCR amplification and temperature gradient gel electrophoresis (TGGE). Gene quantification is provided by a competitively coamplified DNA standard constructed by point mutation PCR. A single base difference was introduced to achieve a suitable migration difference in TGGE between the original target DNA and the modified standard without altering the PCR amplification efficiency. This competitive PCR strategy is a highly specific and sensitive way to monitor recombinant DNA in environments like the efflux of a biotechnological plant. D 2002 Elsevier Science B.V. All rights reserved. Keywords: gfp; Competitive PCR; TGGE
1. Introduction Emerged from the deep sea, the green fluorescent protein (gfp) from the jellyfish Aqueorea victoria (Prasher et al., 1992) is now also spreading worldwide on the surface. Its gene, transferred by numerous human ‘vectors’, is introduced in more and more different organisms like various bacterial strains, yeast, Drosophila, C. elegans, zebrafish, Xenopus, mice, many plant species and even human cell lines. At the moment, the thriving of gfp is restricted to the
* Corresponding author. Tel.: +49-531-618-1408; fax: +49-531618-1411. E-mail address: [email protected] (A. Felske).
small area of more or less isolated laboratories, but it is only a question of time when gfp-carrying organisms are released deliberately or accidentally into uncontrolled environments. The application of gfp mainly served the investigation of the structure and dynamics of macromolecules in living cells and had quickly become a primary tool for analysis of DNA and protein localization (van Roessel and Brand, 2002). Strong fluorescence by external illumination and activity independently from external substrates made it an attractive and a versatile reporter. The selfcontained domain structure of gfp reduces the possibility of inhibition by cellular factors of the host (Yang et al., 1996). Also, the design of fused proteins is feasible, neither disturbing gfp fluorescence nor the activity of the protein to which it is fused (Rizzuto et
0167-7012/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 7 0 1 2 ( 0 2 ) 0 0 1 8 4 - 7
298
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
al., 1996). The stability of gfp also allows detection of its fluorescence after the producing host has died, like for instance, during protein purification or cell fixation for other staining protocols. Localization of gfp fusion proteins can be analyzed in live cells, enabling real-time monitoring of dynamics in situ (Romberg et al., 1998). Such analyses have been further enhanced by the selection of brighter, more soluble gfp variants (Heim et al., 1995; Crameri et al., 1996; Haseloff, 1999). Hence, there is little doubt that gfp will continue to gain ground and segregate into a high diversity of fluorogenic enzymes, and all this directed by the hands of humans. The gfp had become a domestic enzyme, transported and replicated in environments far beyond the reach of the original host A. victoria. Here, we report on the design of a molecular method to detect gfp on the DNA level. This is useful to monitor genetically engineered microbes (GEM) not only in their active state, but also in distorted biosamples where activity and cellular integrity had largely ceased. The most sensitive and also highly specific molecular test is the PCR. We used quantitative PCR to count down to very low target numbers by competitive coamplification of a standard molecule of known concentration (Freeman et al., 1999). It was the aim to design the standard with only one base difference to the original sequence for a temperature gradient gel electrophoresis (TGGE)-based detection system (Felske et al., 2001). In contrast, the original competitive PCR approach used standards of different length (Wang et al., 1989), which might, however, lack a consistent amplification efficiency (Stolovitzky and Cecchi, 1996). Here, we apply a new competitive PCR/TGGE protocol to quantify the release of gfp genes from mercury reduction biocatalysers inoculated with gfp-labeled mercury reducing bacteria.
2. Materials and methods 2.1. Strain cultivation and DNA extraction The genome of the strain Pseudomonas putida KT2442Dmer73Dgfp contained independently introduced DNA fragments carrying the merTPAB operon and the gfp gene as described before (Felske et al.,
2001). Both were constitutively expressed. The strain was grown in Luria –Bertani medium (trypton 10 g l 1, yeast extract 5 g l 1, NaCl 10 g l 1) at 30 jC for 16 h. Cells from a 5-ml culture were harvested by centrifugation at 2000 g for 10 min. The DNA was extracted from the bacteria pellet as previously described (Pitcher et al., 1989) and the DNA concentration measured via the saran wrap method (Sambrook et al., 1989). 2.2. Biocatalyser set-up and sampling Neutralized and aerated chloralkali wastewater (Hg2 + concentration between 5 and 9 mg l 1) was pumped with 160 ml h 1 into the biocatalyser (at approximately 25 jC) in upflow mode and supplemented with 0.1 g l 1 sucrose and 0.02 g l 1 yeast extract. The catalyser was a glass column (200 ml volume, 4.8 cm internal diameter), wherein, above a grid, approximately 80 cm3 pumice was placed (core diameter 2.5 – 3.5 mm, Raab, Neuwied, Germany). Columns and tubing were sterilized by autoclaving (121 jC, 20 min). Each sampling day, 51 ml efflux was collected. One milliliter was used for cfu counting, where 50 Al efflux and serial 10-fold dilutions were plated on rich agar medium (4 g l 1 sucrose, 2 g l 1 yeast extract and 15 g l 1 NaCl). Three replicates were prepared and checked for growth after a week. The suspended matter of the remaining 50 ml was harvested by centrifugation at 2000 g for 30 min. The pellet was forwarded to DNA extraction and quantification as described above. 2.3. Primer selection and standard construction The amplified DNA stretch used for TGGE analysis covered the positions 573 – 721 of the gfp gene (EMBL accession number X83959). The primers for standard construction and production covered a little larger stretch. The simulation of DNA denaturation was computed with the PC-software Poland V1.3 (Steger, 1994), the graphs were made with Excel 97 (Microsoft). The default parameters were used for the computation without adjusting to electrophoresis-specific conditions. The TGGE primers were GC-GFPf (5V-[CGCCCCCGCCGCCCCGCCGCCCGCCGCCCC]-GCCCTGTCCTTTTACCAGAC-3V) and GFPr (5V-CCATGTGTAATCCCAGCAGC-3V).
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
The GC-clamp, which is recommended for TGGE analysis (Abrams et al., 1990), is given in brackets. The GFP primer sequences are printed bold, and the PRO primer sequences are underlined to clarify the nested arrangement of the primer sets. The primers for introducing the point mutation (-C-) were MUTf (5V-CTCCAATTGGCGATGGCCCTGTC-3V) and MUTXr (5V-GTATAGTTCATCCATGCCATGTGTAATCCCAGCAGCTGTTACAAACTCAAGA-cGACCA-3V). Another primer pair was used to produce standard template DNA with or without mutation, the primers PROf (5V-GCGATGGCCCTGTCCTTTTACCAGAC-3V) and PROr (5VCATCCATGCCATGTGTAATC-3V). Amplification was performed with a Mastercycler Gradient (Eppendorf, Hamburg, Germany) using 30 cycles of 94 jC for 10 s, 56 jC for 20 s and 68 jC for 20 s. The PCR (20 Al) contained 10 mM Tris –HCl (pH 8.3), 50 mM KCl, 3 mM MgCl2, 100 AM each of dATP, dCTP, dGTP, and dTTP, 0.4 AM each of one forward and one reverse primer, 0.5 units of recombinant Taq DNA polymerase (LifeTechnologies, Paisley, UK) and 1 Al template. This PCR with MUT primers on genomic DNA of P. putida KT2442Dmer73Dgfp (1 ng Al 1) has been used to introduce the mutation. It was performed in serial 10-fold template DNA dilution to minimize template carry-over. Strong MUT-PCR products were obtained until dilution step 10 3 (1 pg Al 1). This product has been purified with the NucleoSpinR kit (Macherey-Nagel, Du¨ren, Germany) according to the instructions of the manufacturer. The same PCR protocol, but with PRO primers produced standard molecules by reamplification of the MUT product. Also, this PRO-PCR was performed in serial 10-fold template DNA dilution to minimize the contribution of original genomic DNA. Strong PRO-PCR product was obtained until dilution step 10 5. This PCR product was purified with the NucleoSpinR kit and used as standard (DNA concentration was measured as described above). The corresponding diluted template of purified MUT-PCR products were stored for future standard production. 2.4. Competitive PCR and TGGE Each competitive PCR experiment consisted of seven reactions of decreasing gradients of standard
299
DNA (400 fg and serial 3-fold dilutions, i.e. 1 106, 3.3 105, 1.1 105, . . . molecules; samples with very low DNA concentration were coamplified with 40 or 4 fg input and corresponding dilutions). The second competitor was a DNA sample representing 1 ml of the catalyser efflux, i.e. 1 Al of 50 Al DNA solution per reaction. The seven-reactions-mastermix for a competitive PCR assay was composed for PCR as described above, but with 5 Al template volume per reaction (1 Al sample + 4 Al standard): The catalyser efflux DNA was added in 7 Al to the seven-reactionsmastermix, and after distribution of the mastermix to the seven reaction tubes (7 16 Al), the DNA standards of 4 Al/tube were added. The PCR conditions were the same as described above. The Biometra TGGE maxi system (Biometra, Go¨ttingen, Germany) was used for separation of amplicons after PCR (2 Al PCR product input for TGGE). Electrophoresis took place in a 0.8-mm polyacrylamide gel (6% w/v acrylamide, 0.1% w/v bis-acrylamide, 8 M urea, 20% v/v formamide, 2% v/v glycerol) with 1 MN buffer (20 mM MOPS, 10 mM NaOH, pH f 7.0) at a fixed voltage of 5 V for 10 min (prerun to establish the temperature gradient) and then at 400 V for 2 h. A temperature gradient was built-up in electrophoresis direction from 30 to 45 jC. Silver-stained gels (Sanguinetti et al., 1994) were scanned with a GS700 Densitometer (Bio-Rad, Hercules, USA) and analyzed with Scion Image 3b software (Scion, Frederick, USA). In every competitive PCR assay, the lanes were analyzed, where the band intensities of the standard and the DNA template were most similar. Pixel (Pv) volumes of the band-images were quantified, and the original chromosome amount (C) in the genomic DNA extraction was calculated with these values and the standard molecule amount (St): C = PvC PvSt 1 St. Since the DNA input represented 1 ml catalyser efflux, the GEM chromosome number ml 1 catalyser efflux could be calculated. Efflux DNA preparation concentrations have been translated into chromosome equivalents with regards to the P. putida chromosome size of 6 Mbp (RamosDı´az and Ramos, 1998): 1 ng genomic DNA approximates 1.5 104 chromosomes. One standard PROPCR product molecule of 142 bp with 78.8 kg mol 1 was equivalent to one GEM chromosome (10 ng standard DNA = 7.6 1010 GEM chromosome equivalents).
300
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
2.5. Perpendicular TGGE The sample for perpendicular TGGE was a mixture of 100 Al gfp PCR product from the standard (with mutation) and 100 Al gfp PCR product from the P. putida KT2442Dmer73Dgfp genomic DNA applied along a slot covering almost the whole gel width. TGGE was done for 15 min without temperature gradient (28 jC), then interrupted to set-up a 28 –44 jC gradient in perpendicular direction to electrophoresis and continued for further 2 h (other TGGE parameters as above). In contrast to normal parallel TGGE, this TGGE variant did not require careful gradient optimization. Therefore, the 30 –45 jC mentioned before was the optimal gradient for parallel TGGE.
with the PRO primers and the MUT – PCR product as DNA template (10 5 dilution) was based on maximum 2000 MUT template molecules with virtually no genomic DNA remains (4 10 6 chromosomes Al 1). 3.2. Competitive PCR and TGGE The simulation of the GFP-PCR product migration speed at different temperatures and the experimental confirmation via perpendicular TGGE demonstrated a changed denaturation behavior (Fig. 1), resulting in
3. Results and discussion 3.1. Competitive PCR standard production Competitive PCR requires a concentration standard which is easily distinguishable from the target DNA. TGGE separates amplicons by their denaturation progress along a growing temperature gradient. The disproportionate DNA double strand stability, caused by the higher stability of C u G pairs compared to T = A pairs, could be used to manipulate the melting behavior. This effect could be simulated with the computer software Poland V1.3 in the search for the most suitable site to introduce a point mutation (data not shown). The point mutation was introduced 17 bp ahead of the GFPr 3V-end and exchanges an A by a C. The mutation-introducing MUTXr primer was covering the GFPr sequence for reamplification of the MUT product with the GFPr or PRO primers. The amplification of serial dilutions of genomic DNA from the P. putida KT2442Dmer73Dgfp11 with the MUT primer pair succeeded with a DNA input of as low as 1 pg (approximately 160 chromosomes, data not shown). This MUT –PCR product again could be reamplified with the PRO primers down to a template dilution of 10 6 (1 Al template in 20 Al reaction volume). Given the 200-fmol Al 1 nucleotide input (see above), the PCR could theoretically reach a concentration of approximately 2 108 amplicons Al 1. Thus, the routine standard DNA production
Fig. 1. Denaturation of the gfp PCR fragment with (st) and without (or) mutation. (a) The computer simulation shows a clear discrimination of the two fragments in the temperature range from 69 to 73 jC. (b) Perpendicular TGGE shows the predicted discrimination of the fragments, but temperatures were decreased by using urea and formamide in the gel to protect it from drying. st: standard product. or: genomic DNA product. ss: single-stranded DNA.
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
separated TGGE bands of the competitive PCR (Fig. 2). The high temperatures required to force the melting (Fig. 1a) could be reduced by approximately 36 jC by using the denaturing gel ingredient urea and formamide (Fig. 1b) to protect the gel from drying. There are two types of heteroduplexes visible in Fig. 2: st+/or and or+/st . Among them, the difference changing the melting behavior is a T p C mismatch compared to a weak GA match at the mutation site. TGGE may resolve such minor differences under favorable conditions, which were intentionally provided by selecting a mutation site critical for denaturation of the amplicons. Thus, we detected four bands: Firstly stops heteroduplex or+/st with one T p C mismatch, second is heteroduplex st+/or with weak G u A match, third comes homoduplex or+/ or with T = A match and the last molecule stopping on TGGE is the homoduplex st+/st with strong G u C match. Heteroduplex abundance is always limited by the yield of the less abundant amplicons because a heteroduplex requires one strand of each amplicons. Thus, heteroduplexes were most abundant in the reactions where both amplicons are equal. Further, only these reactions with the most similar amplicons amounts for standard and target are used for quantification (Fig. 2, marked with *), because reactions with growing discrepancy of standard and target are suffering increasing bias by the C0t effect (Mathieu-Daude´ et al., 1996). We also applied DNA samples without gfp for PCR control, yielding only TGGE bands of the standard and no heteroduplexes. This method could easily be transferred to an application in real-time PCR machines (Heid et al., 1996), which may detect melting temperature varia-
301
tions directly during PCR (Ririe et al., 1997). Where such machines are available, this method could be applied even more straightforward with high sample throughput and probably also more accurate. Nevertheless, the cumbersome good old TGGE also had advantages by providing a clearer separation of PCR products and eventually accompanying side products and heteroduplexes. There was also no need for concerns about primer specificity, since only amplicons of correct length and sequence would end on the standard and target positions of the TGGE gel. 3.3. Demonstration of gfp efflux monitoring from a biotechnical device A biocatalyser for mercury remediation was inoculated with a mercury-reducing gfp-labeled P. putida KT2442Dmer73Dgfp11. According to our experience, heavily loaded industrial wastewater like our sample (from chlor alkali electrolysis) typically contained 1 –10 mg l 1 Hg2 + (von Canstein et al., 1999). Our biocatalysers are designed for them as end-of-pipe solutions. While contaminated environments may contain approximately 1000-fold lower mercury concentrations, the mercury concentrations in the industrial wastewater were absolutely lethal to organisms without mercury resistance ( = mercury reduction). Therefore, the P. putida KT2442Dmer73 is constitutively overexpressing the merTPAB operon resulting in elevated levels of mercuric reductase activity (Horn et al., 1994). The biocatalyser outflow was analyzed over 94 days to compare the competitive PCR method with cfu counts and another competitive PCR approach targeting the merTPAB operon insertion region (Felske et
Fig. 2. Silver-stained TGGE gel with competitive PCR amplicons on it. The seven reactions on the left hand were from day 94 and the ones on the right hand from day 88. The lanes used for quantification are marked with *.
302
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
Fig. 3. Microscopic picture of a pumice surface after 3 days of growth, taken with a CLS microscope. The young growing cell layer of the gfp-labeled strain is visible as bright rod-shaped cells. Grey background signals are reflections from the pumice surface.
al., 2001). Direct microscopic observation of the gfplabeled GEM succeeded only in the early biofilmformation stages on the pumice filling in the reactor
(Fig. 3). When the biofilm became more dense by growth and the subsequent accumulation of dead biomass, single cells could hardly be distinguished. Torn fragments of this biofilm flushing out of the biocatalyser showed weak fluorescence, which could not be appointed, to single cells, hence, the origin of this fluorescence remained unproved (data not shown). The visualization of gfp-containing bacteria was much simpler with the cultivation approach. Putting Petri dishes under UV light revealed gfp presence as greenglowing colonies. The cfu quantification yielded a broad variation over three orders of magnitude across the whole period (Fig. 4). Irregular biomass release from the biocatalyser caused great variation of total cfu counts and DNA extraction yields over several magnitudes. However, the cultivation approach led to a massive underestimation of biomass and gfp release. The DNA extraction yields were translated into chromosome equivalents, following the simplification that the chromosome size of all bacteria in the system is the same like the GEM (6 Mbp). This estimation was rough, but good enough to serve our interpretations. The comparison to the total DNA yield was chosen to add a relative estimation to the absolute values and indicated up to two magnitudes more bacteria than detectable by cfu count (Fig. 4). Hence, we could
Fig. 4. Comparison of the GEM and invading bacteria in the catalyser efflux by cfu counts and DNA quantification. The dotted columns are the cfu counts of all bacteria and the only fluorescent cfu counts of the gfp-labeled GEMs. Some cfu counts were below 1000 ml 1, the detection limit of the used cfu counting, and therefore neglected. The white columns represented the DNA extraction yield translated into chromosomal equivalents (1 equivalent = 6.7 fg DNA). The stripped columns are the amount of merTPAB operons of the GEM as quantified by a previously published competitive PCR/TGGE protocol. The black columns are the numbers of gfp genes as quantified by the gfp competitive PCR/TGGE protocol.
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
demonstrate sometimes a much higher biomass efflux than indicated by cfu counts. This is more supported, but contraindicated by the fact that probably not all DNA could be extracted from the biofilm fragments glued by polysaccharides. Cell quantification via microscopic counts or biomass quantification would have been valuable, but was hampered by the distorted state of the (ex-) biofilm aggregates leaving the reactor. The comparison of gfp gene numbers to the total DNA extraction yield correlated well to the ratio of total cfu to the gfp-labeled cfu. Both results showed that the gfp-labeled strain has been challenged by contaminants from the nonsterile wastewater input. Further, the gfp quantification gave almost identical results to the strain-specific merTPAB operon insert region quantification. Mercury resistance and gfp of P. putida KT2442Dmer73Dgfp11 were both introduced into the host genome by genetic engineering, but independently. Thus, both genes are located in different areas of the genome and detected completely independently. While the gfp method described here was specific for gfp and its variants, the merTPAB operon method was exclusively designed for P. putida KT2442Dmer73: by composing one primer for the introduced merA gene (mercuric reductase) and the other for the adjacent genome (Felske et al., 2001). The little deviation of both values can best be explained by the limited resolution of the competitive PCR set-up. The three-fold dilution of standards and the subsequent densitometric measurement of electrophoresis bands are likely to be the most limiting steps for a higher resolution. 3.4. The labeled strain and the intruders The biofilm within the biocatalyser 3 days after the inoculation (Fig. 4) was clearly made of the gfplabeled strain as shown by cultivation and DNA analysis. In the beginning, all isolated colonies were fluorescent and the quantitative PCRs gave values almost identical to the estimation by DNA yield. However, from day 7 on the quantitative PCR results showed one (day 11) to three (day 59) magnitudes lower amounts of gfp and merTPAB operons than bacterial genomes (indicated by DNA yield). Also the cultivation approach yielded from this day on nonfluorescent colonies. Analysis of these colonies showed that they were contaminants (data not
303
shown). These intruders probably contaminated the nonsterile wastewater reservoir through the air and entered via this route the biocatalysers (WagnerDo¨bler et al., 2000). Since they were also highly efficient mercury reducers, they could successfully challenge the P. putida KT2442Dmer73Dgfp11. Also interesting is the period from days 45 to 59, when cfu counts dropped below reliable quantification. Appearing almost sterile, the biocatalyser output from these days may suggest the death of the bacterial community. However, the DNA yield was not reduced, indicating the continued presence of the biofilm. Consequently, the performance of the system was not ceased, because the biocatalyser depended more on the presence of the bacteria than their growth activity.
4. Conclusions The gfp is a useful marker to follow the ways of labeled microbes in living ones, but with this approach, also in dead ones. The study of dissemination of such common markers will serve to improve security measures for GEM use. The high sensitivity of the PCR quantification is applicable to monitor contamination events. Microbes released deliberately or by leakage accidents would probably be rapidly inactivated in a low-nutrient water, sediment or soil body. Thus, gfp fluorescence would fade while the GEMs, or their DNA, remain intact. Hidden in a huge diversity of environmental bacteria, the GEM may still be tracked back by its DNA. Even transformation events may be followed, since any other organism picking up the gfp gene should give a positive signal as well. The ease and efficiency of natural bacterial dissemination will surely lead to increasing appearance of gfp in the environment. Our method will help to assess this process in a straightforward way.
Acknowledgements This work was supported by a grant from a European Communities EC project (QLK3-CT-199901213). A.F. was supported by a grant from the European Communities EC project BACREX (QLK32000-01678).
304
A. Felske et al. / Journal of Microbiological Methods 52 (2003) 297–304
References Abrams, E.S., Murdaugh, S.E., Lerman, L.S., 1990. Comprehensive detection of single base changes in human genomic DNA using denaturing gradient gel electrophoresis and a GC clamp. Genomics 7, 463 – 475. Crameri, A., Whitehorn, E.A., Tate, E., Stemmer, W.P., 1996. Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat. Biotechnol. 14, 315 – 319. Felske, A., Pauling, B.V., von Canstein, H.F., Li, Y., Lauber, J., Buer, J., Wagner-Do¨bler, I., 2001. Molecular monitoring of genetically improved mercury reducing bacteria by competitive PCR. BioTechniques 30, 142 – 148. Freeman, W.M., Walker, S.J., Vrana, K.E., 1999. Quantitative RTPCR: pitfalls and potential. BioTechniques 26, 112 – 125. Haseloff, J., 1999. Gfp variants for multispectral imaging of living cells. Methods Cell Biol. 58, 139 – 151. Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Real time quantitative PCR. Genome Res. 6, 986 – 994. Heim, R., Cubitt, A.B., Tsien, R.Y., 1995. Improved green fluorescence. Nature 373, 663 – 664. Horn, J.M., Brunke, M., Deckwer, W.-D., Timmis, K.N., 1994. Pseudomonas putida strains which constitutively overexpress mercury resistance for biodetoxification of organomercurial pollutants. Appl. Environ. Microbiol. 60, 357 – 362. Mathieu-Daude´, F., Welsh, J., Vogt, T., McClelland, M., 1996. DNA rehybridization during PCR: the ‘C0t effect’ and its consequences. Nucleic Acids Res. 24, 2080 – 2086. Pitcher, D.G., Saunders, N.A., Owen, R.J., 1989. Rapid extraction of bacterial genomic DNA with guanidium thiocyanate. Lett. Appl. Microbiol. 8, 151 – 156. Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G., Cornier, M.J., 1992. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229 – 233. Ramos-Dı´az, M.A., Ramos, J.L., 1998. Combined physical and genetic map of the Pseudomonas putida KT2440 chromosome. J. Bacteriol. 180, 6352 – 6363.
Ririe, K.M., Rasmussen, R.P., Wittwer, C.T., 1997. Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Anal. Biochem. 245, 154 – 160. Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R.Y., Pozzan, T., 1996. Double labelling of subcellular structures with organelle-targeted gfp mutants in vivo. Curr. Biol. 6, 183 – 188. Romberg, L., Pierce, D.W., Vale, R.D., 1998. Role of the kinesin neck region in processive microtubule-based motility. J. Cell Biol. 140, 1407 – 1416. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Saran wrap method. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, New York. appendix E6. Sanguinetti, C.J., Dias Neto, E., Simpson, A.J.G., 1994. Rapid silver staining and recovery of PCR products separated on polyacrylamide gels. BioTechniques 17, 915 – 919. Steger, G., 1994. Thermal denaturation of double-stranded nucleic acids: prediction of temperatures critical for gradient gel electrophoresis and polymerase chain reaction. Nucleic Acids Res. 22, 2760 – 2768. Stolovitzky, G., Cecchi, G., 1996. Efficiency of DNA replication in the polymerase chain reaction. Proc. Natl. Acad. Sci. U. S. A. 93, 12947 – 12952. van Roessel, P., Brand, A.H., 2002. Imaging into the future: visualizing gene expression and protein interactions with fluorescent proteins. Nat. Cell Biol. 4, E15 – E20. von Canstein, H., Li, Y., Timmis, K.N., Deckwer, W.-D., WagnerDo¨bler, I., 1999. Removal of mercury from chloralkali electrolysis wastewater by a mercury-resistant Pseudomonas putida strain. Appl. Environ. Microbiol. 65, 5279 – 5284. Wagner-Do¨bler, I., Lu¨nsdorf, H., Lu¨bbehu¨sen, T., von Canstein, H.F., Li, Y., 2000. Structure and species composition of mercuryreducing biofilms. Appl. Environ. Microbiol. 66, 4559 – 4563. Wang, A.M., Doyle, M.V., Mark, D.F., 1989. Quantification of mRNA by the polymerase chain reaction. Proc. Natl. Acad. Sci. U. S. A. 86, 9717 – 9721. Yang, F., Moss, L.G., Phillips, G.N., 1996. The molecular structure of green fluorescent protein. Nat. Biotechnol. 14, 315 – 319.