DESCRIPTION
MATERIALS AND METHODS FOR QUANTITATIVE PCR
Cross-Reference to Related Application
This application claims priority from provisional application U.S. Serial No. 60/064,823, filed November 7, 1997.
Background of the Invention Application of modern biotechnology to microbiological testing has resulted in the development of so-called 'rapid methods' that no longer rely solely on the use of culturing. These assays include miniaturized biochemical testing (hemolytic reactions), physiobiochemical measurements of metabolites (bioluminescence and fluroescence), antibody-based methods (ELIS A) and DNA-identification (PCR-amplification and DNA- probe hybridization) techniques. PCR-based assays have already been developed for Listeria monocytogenes (Fluit, A.C. et al, 1993; Furrer, B. et al, 1991; Wang, R-F. et al, 1992), enteroinvasive Escherichia coli (Anderson, M.R. et al, 1992), enterotoxic Escherichia coli (Stacy-Phipps, S. et al, 1995; Wernars, K. et al, 1991), and enterohemorrhagic Escherichia coli (Gannon, V.P.J. et al, 1992; Lang, A.L. et al, 1994). The polymerase chain reaction (PCR) is based on repeated cycles of denaturation of double-stranded DNA, followed by oligonucleotide primer annealing to the DNA template, and primer extension by a DNA polymerase (Mullis et al. Patent Nos. 4,683,195, 4,683,202 and 4,800,159; Saiki et al. 1985). The oligonucleotide primers used in PCR are designed to anneal to opposite strands of the DNA, and are positioned so that the DNA polymerase-catalyzed extension product of one primer can serve as the template strand for the other primer. The PCR amplification process results in the exponential increase of discrete DNA fragments whose length is defined by the 5' ends of the oligonucleotide primers.
Although it has been possible to detect and amplify large amounts of rare DNA or mRNA transcripts, it has been more difficult to quantitate the amount of the nucleic acid species in the starting material. While PCR can detect the presence of a targeted
nucleic acid species in the starting material, the results of conventional PCR cannot be used to calculate the pre-amplification levels of that targeted nucleic acid. This has precluded the use of PCR in many situations, for example, in an analysis of the fold induction of a specific mRNA in response to exogenous stimuli. Quantitation of mRNA by reverse transcriptase PCR (RT-PCR) has been attempted using endogenous internal standards or "housekeeping" genes. Unfortunately, inter- and intra-assay variation, due to either efficiency of amplification of individual primer pairs or actual variations in the levels of the standard within the compared samples, is often too great for this method to be reliable (Hoof et al, 1991; Murphy et al, 1990). The development of competition-based quantitation strategies should provide for more accurate and reproducible quantitation of specific mRNAs. This procedure is based on the use a synthetic exogenous standard, which shares identical sequences with the amplification primers, that is added to the reverse transcription reaction.
Currently, there are three general strategies for quantification of amplified target DNA. One approach involves coamplification of an internal standard together with the target of interest followed by size separation of the specific amplicons by electrophoresis (Tarauzzer, R.W. et al, 1996). Electrophoretic size separation schemes have several potential problems, including the dependency of amplification ratios for amplicons on DNA quality and bp length. A second approach uses hybridization with capture probes based on hybridization of the entire amplified sequence (Denis, M. et al, 1997). Capture probe methods are inappropriate for quantification of amplicons that contain homologous regions {i.e., competitive templates), but this technique can be suitably modified for automation (Denis, M. et al, supra; Rudi, K. et al, 1998). The latest approach is realtime detection using quantitiative PCR thermocyclers, like the ABI PRISM 7700 (Batt, C.A., 1997). Real time detection, while rapid and accurate, cannot be easily adapted to large multiplex assays due to the limited availability of fluorochromes with non- overlapping spectra.
In addition, PCR-based systems with coamplification of an internal standard with the targeted bacterial sequence have been established for Salmonella spp. (Jones, D.D. et al, 1993; Rahn, K. et al, 1992), Campylobacter spp. (Giesendorf, B.A.J. et al, 1992;
Wegmuller, B.J. et al, 1993), Yersinia enterocolitica and Vibrio vulnificus (Wang, R.F.
supra), as well as others (Hill, W.E., 1996). Many of these techniques remain deficient in sensitivity and require pre-enrichment steps. Selective enrichment has been used prior to PCR amplification as a way to concentrate the bacteria and dilute out potential inhibitory factors. Unfortunately, these steps tend to negate the sensitivity and speed attributed to PCR-based tests. Obviously, any procedure needed to either concentrate or enrich the bacteria sacrifices both time-savings and the ability to accurately enumerate the contaminating bacteria (Pillai, S.D. et al, 1995; Swaminathan, B. et al, 1994).
Although O. formigenes is attracting greater attention for its role in regulating oxalic acid absorption in both humans and animals, the difficulty in culturing this fastidious anaerobe has greatly limited research into its association with oxalate-related disorders. Until recently, special conditions have been required to culture Oxalobacter sp. and its detection has been generally based on the appearance of zones of clearance of calcium oxalate crystals surrounding colonies. Since O. formigenes expresses two unique genes, oxc (encoding oxalyl-CoA decarboxylase) and frc (encoding formyl-CoA transferase), required for its catabolism of oxalate, this bacterium appeared suitable for
DNA-based identification.
Brief Summary of the Invention The subject invention pertains to novel materials and methods for quantitating target RNA or DNA template using polymerase chain reaction methods. The subject invention is useful for quantitating the number of microorganisms, such as bacteria, in a clinical sample. In performing the methods of the subject invention, a target polynucleotide sequence can be quantitated using internal competitive template sequences during PCR amplification. The internal competitive template sequences used in conjunction with the subject methods comprise a polynucleotide sequence having the same primer binding sites as the target polynucleotide template sequences. The internal competitive template sequences can be designed such that a nucleotide sequence present in the target polynucleotide sequence is missing from the sequence of the internal competitive template. Accordingly, during PCR amplification, the internal competitive template will produce amplification products that are smaller in size than the amplification product of the target template sequences. This permits the amplification
products of the target template sequences and internal competitive sequences to be distinguished due to the differences in size. Relative band intensities of the amplification products can be determined and normalized for differences in molecular weight and then plotted against the log of the copy number of the internal competitive template added per reaction. The log of the ratio band intensities within each lane can be plotted against the log of copy number of the template added per reaction. Quantity of target template can then be readily determined where the ratio of internal competitive template and target template band intensity is about equal to 1.
Brief Description of the Drawings
Figure 1 shows synthesis of a competitive DNA template for quantifying the PCR. A truncated form of the 416 bp fragment of the oxc gene containing both the 5' and 3' primer sites was synthesized by PCR using a modified 3' primer. This truncated and modified sequence was ligated into the pCR2.1 vector system in order to produce high copy numbers.
Figures 2A-2E show quantification of the number of O. formigenes genomes using the QC-PCR. Dilutions, ranging from 50 to 250,000 molecules of purified plasmid containing the competitive template were used either as DNA template in PCR to establish standard curves (Figure 2B) or as competitive DNA by mixing with 2 x 104 (Figure 2A) or 2 x 102 (Figure 2C) copies of purified O. formigenes strain OxB genomes.
PCR band intensities were scanned, normalized for molecular mass, and the log ratios of O. formigenes to template band intensities graphed to determine log equivalence (Figures 2D and 2E).
Figure 3 shows a comparison of O. formigenes cfu in a bacterial culture as detected by culture and QC-PCR. An overnight culture of O. formigenes strain OxB containing approximately 7 x 108 cfu/ml, as determined by O.D.600, was serially diluted 10-fold. DNA, isolated from each dilution, was used as experimental DNA in QC-PCR. Log equivalence of O. formigenes to template band intensities were determined and compared to the estimated cfu for each dilution.
Detailed Disclosure of the Invention The subject invention concerns novel materials and methods for quantifying target RNA or DNA sequences in a sample using polymerase chain reaction (PCR) to amplify both target and exogenous competitive template sequences. In one embodiment, the subject invention can be used to quantitate the number of microorganisms, such as bacteria, in a clinical sample. For example, materials and methods described herein can be used to quantitate the number of enteric bacteria in a fecal sample.
In one embodiment of the invention, a target polynucleotide sequence is quantitated through the use of novel internal competitive templates using polymerase chain reaction. The internal competitive template used in the subject invention comprises a polynucleotide sequence that has the same or substantially similar primer binding sites at the 5' and 3' ends as the target polynucleotide sequence. However, in one embodiment, the internal competitive template lacks a portion of the sequence found in the target polynucleotide sequence. Thus, the internal competitive template when amplified produces an amplification product that is smaller in size and, therefore, can be distinguished from the amplification product of the target template. The differences in size between the target template sequence and the internal competitive template sequence should be selected so as to allow one to distinguish between the two template sequences.
Competition-based quantitative PCR is based on the principle that the upstream and downstream polynucleotide primers will equally compete for target template and the internal competitive template. This is particularly true if the exogenous internal competitive template contains identical complementary sequences as found in authentic target template. However, designing the internal competitive template to generate an amplification product of different size than that produced during PCR amplification of the authentic target template allows for the respective amplification products to be easily separated by standard sizing procedures, such as electrophoresis on agarose gels.
After PCR amplification and size separation of the amplification products by, for example, gel electrophoresis, the bands representing target template and internal competitive template amplification products are visualized. Relative band intensities are determined and normalized for differences in molecular mass. Band intensities are then
plotted against the log of the copy number of the internal competitive template added per reaction.
Using an embodiment of the invention exemplified herein, the subject methods were capable of quantifying as low as 10 cfu Oxalobacter formigenes per ml and this quantification was linear up to 108 cfu/ml.
The subject invention also concerns the internal competitive templates that can be prepared and used according to the methods of the subject invention. Internal competitive template can be prepared from any target template sequence by engineering a deletion of a portion of the internal sequence of the target template. The subject invention also concerns kits for quantifying a target polynucleotide in a sample. The kit can optionally comprise in one or more containers 3'- and 5'- oligonucleotide primers that can bind to primer binding sites on target and competitive template polynucleotides and are capable of functioning in PCR reaction; a competitive template polynucleotide and reagents for performing PCR. All publications and patents referred to herein are hereby incorporated by reference.
Materials and Methods Bacterial strains Oxalobacter formigenes strain OxB was used throughout this study as the standard. This strain was grown in medium B containing 30 mM oxalate, as described elsewhere (Allison, M.J. et al, 1985).
Preparation of genomic DNA from bacterial strains Cultures (10-15 ml) of O. formigenes were centrifuged at 10,000 x g, the supernate discarded, and the bacterial pellet stored at -80°C. Bacteria from O. formigenes were resuspended in 567 μl TE buffer (10 mM Tris-
Cl, pH 7.5 plus 1 mM EDTA, pH 8.0), 30 μl 10% sodium dodecyl sulfate (SDS) and 3 μl of proteinase K (20 mg/ml), and each mixture incubated 5 hr at 37°C to ensure bacterial cell lysis. Nucleic acids were extracted from the lysates using phenol/chloroform/isoamylalcohol (25:24:1). Chromosomal DNA was precipitated from the aqueous phase by adding l volume of 7.5 M ammonium acetate and 2 volumes of
100% ethanol. DNA was recovered by centrifugation (12,000 x g) and washed once with 70% ethanol. The pellet was resuspended in 15-20 μl H2O.
Preparation of genomic DNA from human fecal specimens Bacterial DNA was isolated directly from fresh stool samples obtained from individuals known to be positive or negative for O. formigenes using the procedures of Stacy-Phipps et al. (1995). Approximately 25 mg of feces was suspended in 1.5ml phosphate-buffered saline, centrifuged at low speed to remove debris, and the supernate centrifuged at 16,000 x g for five minutes to obtain a bacterial pellet. The pellet was resuspended in 0.6 ml of binding/lysis buffer (5.3 M guanidine thiocyanate, 1 OmM dithiothreitol, 1 % TWEEN 20,
0.3M sodium acetate, 50 mM sodium citrate) and incubated at 65 °C for 10 minutes. Glass matrix (50 l) (GlassPlac, National Scientific Supply Company, San Rafael, California) was added to absorb DNA. The DNA was eluted from the glass beads using the lOmM TE buffer.
PCR and QC-PCR For PCR, performed as described in Lung, H. et al. (1994), 50 μl reactions contained 1.5 mM MgCl2, 200 μM dNTP, 1.25 U Taq-polymerase (GIBCO- BRL, Bethesda, MD), 1 μg genomic DNA and 1 μM of a 5' and 3' primer. The optimal reaction profile proved to be 94°C for 5 min, then 45 cycles of 94°C for 1 min of denaturation, 55°C for 2 min of annealing and 72°C for 3 min of primer extension. For
QC-PCR, 1 μl of an appropriate dilution of competitive templates was added to the standard PCR reaction. PCR and QC-PCR products were size separated by gel electrophoresis in 1.2% agarose containing EtBr, illuminated with UV light and photographed for documentation. Gels were then scanned with an Alpha Inotech Imager ISO 1000 (Alpha Inotech Corp., San Leandro, CA) to determine the relative band intensities of the PCR products for quantification.
Following are examples which illustrate procedures for practicing the invention. These examples should not be construed as limiting.
Example 1 - Quantitative-Competitive DNA template-PCR (QC-PCR). O. formigenes expressed a unique gene required for the catabolism of oxalate: oxc (encoding oxalyl- CoA decarboxylase). This gene has been cloned and sequenced (Lung, H. et al, supra). Sequencing of the 5'-end of the oxc gene from numerous isolates of O. formigenes identified several unique, highly-conserved regions which permitted the synthesis of a genus-specific PCR-primer pair (forward primer: 5'-AATGTAGAGTTGACTGA-3' and reverse primer: 5'-TTGATGCTGTTGATACG-3') (Sidhu, H. et al, 1997a; Sidhu, H. et al, 1997b ). This PCR-primer pair, referred to as OXFfp and OXFrp, respectively, amplifies a 416 bp product of the 5 '-end of the oxc gene, fully unique to O. formigenes. QC-PCR is based on the assumptions that a genomic DNA template and a competitive DNA template containing homologous primer sites will compete equally for the PCR primers and that both experimental and competitive template PCR products will subsequently be amplified co-linearly. To construct a suitable competitive DNA template for use as the internal control, a 227 bp fragment of the oxc gene flanked by sequences homologous for the OXFfp/OXFrp primer pair and containing a genus-specific probe was generated (Figure 1). To accomplish this, a PCR reaction was performed using the OXFfp 5'-primer plus a modified OXFrp 3'-primer. The modified 3'-primer, (5'- TTGATGCTGTTGATACGGTCAAGCAAACGCC-3'). consisted of two portions: a 5'- end which contained the 3 '-primer sequence (underlined) within the oxc gene plus a 3'- end which annealed at a site located approximately 200 bp downstream of the 5'-primer site. The PCR using the primer pair OXFfp/modified-OXFrp amplified the 210 bp segment and synthesized the 17 bp OXFrp primer site at the 3'-end. This PCR fragment was purified and ligated into pCR-2.1 (Invitrogen, Inc., San Diego, CA). A recombinant pCR-2.1 plasmid with the proper insert (confirmed by sequencing) was selected for use as the internal competitive template.
Example 2 - Quantification of cfu using QC-PCR. To determine the accuracy of this QC- PCR to quantify cfu of O. formigenes, QC-PCR were established with two dilutions of an O. formigenes DNA preparation with a starting spectrophotometric reading of 1.126 μg DNA/μl. Assuming the genome of O. formigenes to be similar in size to that of E. coli (4.7 x 103 Kb), then 1 μg of genomic DNA would contain 1.8 x 108 molecules (or
gene copies). Thus, the original genomic DNA preparation of O. formigenes OxB contained approximately 2 x 108 molecules/μl. Two dilutions, 10~4 (20,000 genomes) and 10"6 (200 genomes) of this DNA were used as template in the QC-PCR with dilutions of competitive template ranging from 50 to 250,000 molecules. The PCR products were size separated by electrophoresis through 1.5% agarose gels and visualized with UN light
(Figure 2, top panels). PCR bands were scanned for intensities, normalized for differences in molecular mass, and the log ratios of O. formigenes to template band intensities plotted against the log of the copy number of synthetic template added per reaction to determine log equivalence. Quantitation of the number of oxc genes, thereby the total number of bacteria, revealed the accuracy of this QC-PCR detection system: the
Log equivalence revealed that the number of molecules of O. formigenes OxB in the reaction were estimated between 19,900-25,100 and 126-158, respectively (Figure 2, bottom panels).
Example 3 - Correlation between cfu detected by culture versus QC-PCR. To correlate the number of cfu of O. formigenes detected via culture/plating versus QC-PCR, and overnight culture of O. formigenes was prepared and determined to have a titer of approximately 7 x 108 cfu ml. This culture was diluted serially 10-fold. DΝA, isolated from each dilution, was used as experimental DΝA and mixed with competitive template diluted from 1 x 1010 to 1 x 102 copies/reaction. The PCR products were size separated, visualized with UN light and photographed. Photographs were scanned for relative band intensities, normalized for differences in molecular weight, and plotted against the log of the copy number of synthetic template added per reaction. As shown in Figure 3, the number of cfu for each dilution, as estimated by OD and determined by QC-PCR, were nearly identical.
Similar results were obtained when the number of cfu in clinical fecal specimens were detected by culture versus QC-PCR (Table 1). Fecal samples were collected from individuals known to be colonized or non-colonized with O. formigenes. The cfu were determined for each sample either by standard culture methods or by the QC-PCR detection system. Again, whether the sample was analyzed by culture or by QC-PCR, the number of cfu proved quite similar for all specimens.
Table 1. Correlation between cfu by standard culturing and QC-PCR
Clinical sample3 cfu / gm feces wet weight
Culture Method QC-PCR 1 0 0
2 0 0
3 0 0
4 2.3 x 106 2.5 x 106
5 5.5 x lO6 3.0 x l06 6 2.1 x lO7 3.5 x lO7
7 3.1 x lO7 3.1 x lO7
8 1.0 x lO7 1.6 x lO7
Specimens collected in North Carolina where cfu determined by culture; samples mailed to Florida where era were determined by QC-PCR
It should be understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and the scope of the appended claims.
References
U.S. Patent No. 5,213,961 (Bunn et al.)
U.S. Patent No. 4,683,195 (Mullis et al.)
U.S. Patent No. 4,683,202 (Mullis et al.)
U.S. Patent No. 4,800,159 (Mullis et al.)
U.S. Patent No. 5,219,727 (Wang et al.)
Allison, M.J., K.A. Dawson, W.R Mayberry, J.G. Foss. (1985) Oxalobacter formigenes gen. nov., sp. nov.: oxalate degrading bacteria that inhabit the gastrointestinal tract. Arch. Microbiol. 141: 1-7.
Anderson, M.R., C.J. Omiecinski. (1992) Direct extraction of bacterial plasmids from food for polymerase chain reaction amplification. Appl. Environ. Microbiol. 58: 4080-4082.
Batt, C.A. (1997) Molecular diagnostics for dairy-borne pathogens. J. Dairy Sci. 80: 220-229.
Denis, M. et al. (1997) Development of a semiquantitative PCR assay using internal standard and colorimetric detection on microwell plates for pseudorabies virus.
Mol. Cell. Probes 11: 439-448.
Fluit, A.C, R. Torensma, M.J.C. Visser, C.J.M. Aarsman, M.J.J.G. Poppeleir, B.H.I. Keller, P.Klawijk, J.Nerhoef. (1993) Detection of Listeria monocytogenes in cheese with the magnetic immuno-polymerase chain reaction assay. Appl. Environ.
Microbiol. 59: 1289-1293.
Furrer, B., U.Candrian, C.Hofelein, J.Leuthy. (1991) Detection and identification of Listeria monocytogenes in cooked sausage products and in milk by in vitro amplification of haemolysin gene fragments. J. Appl. Bacteriol. 70: 372-379.
Gannon, N.P.J., R.K.King, J.Y.Kim, E.J.G.Thomas. (1992) Rapid and sensitive method for detection of Shiga-like toxin-producing Escherichia coli in ground beef using the polymerase chain reaction. Appl. Environ. Microbiol. 58: 3804-3808.
Giesendorf, B.A.J., W.G.N.Quint, M.H.C. Henkens, H.Stegeman, F.A.Huf, H.G.M. Niesters. (1992) Rapid and sensitive detection of Campylobacter spp. in chicken products by using the polymerase chain reaction. Appl. Environ. Microbiol. 58: 3804- 3808.
Hill, W.E. (1996) The polymerase chain reaction: Applications for the detection of food-borne pathogens. Crit. Rev. Food Sci. & Nutrition 36: 123-173.
Hoof, T., J.R. Riordan, B. Tammler. (1991) Quantitation of mRNA by the kinetic polymerase chain reaction assay: a tool for monitoring P-glycoprotein gene expression. Anal. Biochem., 196:161-169.
Jones, D.D., R.Law, A.K.Bej. (1993) Detection oi Salmonella spp. in oysters using polymerase chain reaction (PCR) and gene probes. J. Food Sci. 58: 1191-1197.
Lang, A.L., Y-L.Tsai, C.L. Mayer, K.C.Patton, C.J. Palmer. (1994) Multiplex PCR for detection of the heat-labile toxin gene and Shig-like toxin I and II genes in
Escherichia coli isolated from natural waters. Appl. Environ. Microbiol. 60: 3145-3149.
Lung, H., A.L.Baetz, A.B.Peck. (1994) Molecular cloning, DNA sequence and gene expression of the oxalyl-CoA decarboxylase oxc gene, from the bacterium Oxalobacter formigenes. J.Bacteriol. 176: 2468-2472.
Murphy, L.D., C.E. Herzog, J.B. Rudick, A.T. Fojo, S.E. Bates. (1990) Use of polymerase chain reaction in the quantitation of mdr-I gene expression. Biochemistry 29:10351-10356.
Pillai, S.D., S.C.Ricke. (1995) Strategies to accelerate the applicability of gene amplification protocols for pathogen detection in meat and meat products. Crit. Rev. Microbiol. 21: 239-261.
Rahn, K. et al. (1992) Amplification of an invA gene sequence of Salmonella typhimurium by polymerase chain reaction as a specific method of detection of
Salmonella. Mol. & Cell Probes 6: 271-279.
Rudi, K., O.M. Skulberg, F.Larsen, K.S. Jakobsen. (1998) Quantification of toxic cyanobacteria in water by use of competitive PCR followed by sequence-specific labelling of oligonucleotide probes. Appl. Environ. Microbiol. 64: 2639-2643.
Saiki, R.K., S. Scharf, F. Faloona, K.B. Mullis, G.T. Horn, H.A. Erlich, N. Arnheim. (1985) Enzymatic Amplification of β-Globin Genomic Sequences and Restriction Site Analysis for Diagnosis of Sickle Cell Anemia. Science 230:1350-1354.
Sidhu, H, M.J. Allison, A.B. Peck. (1997a) Identification and classification of O. formigenes strains by using oligonucleotide probes and primers. J. Clin. Microbiol. 35:350-353.
Sidhu, H., L. Yenatska, S.D. Ogden, M.J. Allison, A.B. Peck. (1997b) Natural colonization of children in the Ukraine with the intestinal bacterium, O. Formigenes, using a PCR-based detection system. Mol. Diagnosis. 2:89-97.
Stacy-Phipps, S., J.J.Mecca, J.B.Weiss. (1995) Multiplex PCR assay and simple preparation method for stool specimens detect enterotoxigenic E.coli DNA during course of infection. J.Clin. Microbiol. 48: 401-426.
Swaminathan, B., P.Feng. (1994) Rapid detection of food-borne- pathogenic bacteria. Ann. Rev. Microbiol. 48: 401-426.
Tarnuzzer, R.W., S.P. Macauley, W.Farmerie, S.Cabellaro, M.R. Ghassemifar, C.Robinson, M.B.Grant, M.G. Humphrey s-Beher, L.Franzen, A.B.Peck, G.S. Schultz.
(1996) Competitive RNA templates for detection and quantitation of growth factors, cytokines, extracellular matrix components and matrix metallo-proteinases by RT-PCR.
Biotechnology 20: 670-674.
Wang, R.F., W.W.Cao, C.E. Cerniglia. (1992) 16S rRNA probes and polymerase chain reaction method to detect Listeria monocytogenes cells added to foods. Appl. Environ. Microbiol. 58: 2827-2831.
Wegmuller, B., J.Luthy, U.Candrian. (1993) Direct polymerase chain reaction detection of Campylobacter jejuni and Campylobacter coli in raw milk and dairy products. Appl. Environ. Microbiol. 59: 2161-2165.
Wernars, K., E.Delgou, P.S. Soentoro, S. Notermans. (1991) Successful approach for detection of low numbers of enterotoxigenic Escherichia coli in minced meat by using polymerase chain reaction. Appl. Environ. Microbiol. 57: 1914-1919.