Quick identification and quantification of Proteus mirabilis by polymerase chain reaction (PCR) assays

Abstract

Proteus mirabilis is an opportunistic pathogen that can cause urinary tract infection in human beings. The accurate and rapid identification and quantification of P. mirabilis is necessary for early treatment. In this study, a pair of specific primers according to the conserved ureR sequence of P. mirabilis was designed and novel systems which consisted of a polymerase chain reaction (PCR) and a real-time PCR to identify and quantify P. mirabilis were developed. For the qualitative identification by ordinary PCR, a 225-bp DNA product was amplified from P. mirabilis and separated on an agarose gel. The corresponding DNA product is present in three P. mirabilis strains isolated from different geographical locations, but is absent in 20 strains representing 18 different species, including the ureR homolog contained Providencia stuartii and Escherichia coli strains, the other common pathogens Klebsiella sp., Edwarsiella sp., Vibrio sp., Enterobacter sp., and Escherichia sp., and other environmental bacteria Pseudomonas sp. and Acinetobacter sp. Proteus mirabilis at concentrations higher than 1.0 × 10³ CFU ml⁻¹ was detectable by ordinary PCR; P. mirabilis at concentrations higher than 10 CFU ml⁻¹ was quantified by real-time PCR. The specific, sensitive and time-efficient PCR methods were demonstrated to be applicable to rapid identification and quantification of P. mirabilis.

Introduction

Urinary tract infections (UTIs) are among the most frequently occurring human bacterial infections, accounting for about 20 % of all infections acquired outside hospitals. Almost 90 % of UTIs are ascending, with bacteria gaining access to the urinary tract via the urethra to the bladder and then to the upper part of the urinary tract (Hooton 2003; Hryniewicz et al. 2001; Stankowska et al. 2008). Proteus mirabilis, a Gram-negative and rod-shaped bacterium, is one of the most common opportunistic pathogens of UTIs in individuals with long-term indwelling catheters or complicated UTIs (Mobley 1996; Mobley and Belas 1995). The traditional methods to detect P. mirabilis were to culture this bacterium or amplify its 16S rRNA, followed by biochemical reaction or serological test, and these methods were inconvenient, lowly sensitive and time-consuming (Penner and Hennessy 1980; Suter et al. 1968). Moreover, certain methods may be strongly inaccurate for detection of viable but non-cultivable cells (Colwell 2009). Development of techniques that are specific, sensitive, and time-efficient for identification and quantification of P. mirabilis are urgently recommended.

Nowadays, PCR-based methods, in particular quantitative PCR, are used primarily to identify and quantify either pathogens or beneficial populations, including Bacillus sp., Campylobacter sp., Legionella sp., Pseudomonas sp., Salmonella sp., and Vibrio sp., etc., based on the 16S rRNA genes or their specific functional genes (Amri et al. 2007; Anbazhagan et al. 2010; Brolund et al. 2010; dos Santos et al. 2001; Fiume et al. 2005; Goarant and Merien 2006; Le Dréan et al. 2010; Maligoy et al. 2008; Masco et al. 2007; Mashsouf et al. 2008; Nakano et al. 2003; Postollec et al. 2011; Sauer et al. 2005; Smith and Osborn 2008; Sun et al. 2009; Vanniasinkam et al. 1999; Wehrle et al. 2010). Development of multiple PCR to simultaneously detect common pathogens has also been recommended, such as the PCR methods developed to detect Enterobacteriaceae and clinically important bacteria (Cheng et al. 2006; Lu et al. 2000). Species-specific detection of P. mirabilis by PCR methods based on 16S rRNA, ureC and a function unknown gene have been reported previously (Limanskiĭ et al. 2005; Mansy et al. 1999; Takeuchi et al. 1996). The ureC-based PCR method amplified a 533-bp DNA product with a detection limit of 7.0 × 10³ CFU ml^–1 of P. mirabilis, and the PCR method based on the function unknown gene amplified a 3,500-bp DNA product with a detection limit of 10 fg DNA of P. mirabilis.

ureR of P. mirabilis is the only one true regulatory gene that has been identified and is present only in those gene clusters that are inducible by urea. ureR of P. mirabilis and its close homolog (also designated ureR) in the plasmid-encoded urease gene clusters of Providencia stuartii and Escherichia coli also act as an AraC-like positive activator of gene expression in the presence of urea (Mobley et al. 1995). The ureR regulatory gene of P. mirabilis, P. stuartii and E. coli does not have a homolog in the urease gene clusters of Helicobacter pylori, Klebsiella aerogenes, Bacillus sp. or other bacterial species that have been examined thus far (D’Orazio and Collins 1993; Mulvaney and Bremner 1981). Here, by designing a novel pair of specific primers according to ureR nucleotide sequence, a ureR-based PCR method was introduced to detect P. mirabilis for the first time.

In this report, we have developed a P. mirabilis-specific ordinary PCR system amplifying part of the ureR gene uniquely existed in P. mirabilis to qualitatively detect P. mirabilis from lake water, seawater and urine. Another real-time PCR system was also developed to quantitatively detect P. mirabilis from aquatic environments.

Materials and methods

Bacterial strains and culture media

Three P. mirabilis strains and 20 other bacterial strains included in this study are listed in Table 1. All bacterial strains were grown in Luria-Bertani (LB) media (Sambrook et al. 1989) at 28 °C until OD₆₀₀ reached 0.8. The cultures were stored in 50 % glycerol at −80 °C till for use.

Table 1 Bacterial strains used in this study

DNA manipulation

Genomic DNA was extracted from bacterial strains using the DNA extraction kit (Tiangen, China) according to the manufacturer’s specifications. DNA concentration was determined using a nanophotometer (Nanodrop, USA). DNA was stored at −20 °C until used for PCR amplification. T4 DNA ligase was purchased from Takara (Dalian, China) and used in accordance with the manufacturer’s specifications. When the cloning strategy was employed, PCR products were directly ligated to the cloning vector pBS-T (Takara) and then transformed into the competent cell of E. coli. Sequencing was performed by Beijing Genomics Institute (Beijing, China)

Sample collection

Seawater was collected from costal zone (Yantai, Shandong province, China); Lake water was collected from Yantai University; urine of humans was collected when needed. To detect P. mirabilis from these samples, 1 μl each original sample or P. mirabilis-spiked sample was added into PCR system. To avoid the presence of DNase in urine, the urine was heated at 95 °C for 30 min, and then heated urine spiked with P. mirabilis was used as template of PCR system.

Primers and PCR amplification

The pair of primers was designed using the Primer 5 software according to the ureR sequence with accession number Z18752. The forward primer was ureRF1: 5′-GGTGAGATTTGTATTAATGG-3′, and the reverse primer was ureRR1: 5′- ATAATCTGGAAGATGACGAG-3′. Both primers were synthetized by Beijing Genomics Institute.

Each PCR amplification was carried out in a total volume of 15 μl consisting of 1 μl template, 0.5 μl 10 μmol l⁻¹ forward primer, 0.5 μl 10 μmol l⁻¹ reverse primer, 1.5 μl 10× PCR buffer [200 mmol l⁻¹ Tris–HCl (pH 8.4), 200 mmol l⁻¹ KCl, 15 mmol l⁻¹ MgCl₂], 1.5 μl dNTP mix (2.5 mmol l⁻¹ each), 0.25 μl Taq DNA polymerase and 9.75 μl distilled water. The amplification conditions were denaturation at 94 °C for 4 min, then 30 cycles of denaturation at 94 °C for 40 s, annealing at 58 °C for 1 min and extension at 72 °C for 20 s, followed by an extension at 72 °C for 10 min. PCR products were electrophoresed on 1 % agarose gel to determine the size of DNA products. Cells of bacterial strains listed in Table 1 were used as templates respectively.

SYBR-green based real-time PCR assay

Real-time PCR was carried out in an ABI 7300 real-time detection system (Applied Biosystems) by using the Sybr ExScript RT-PCR kit (Takara). Dissociation analysis of amplification products was performed at the end of each PCR to confirm that only one PCR product was amplified and detected. PCR mix without template DNA was used as negative control. The amplification conditions were run under the following conditions: each 20 μl reaction mixture consisted of 10 μl SYBR premix, 0.5 μl 10 μmol l⁻¹ forward primer, 0.5 μl 10 μmol l⁻¹ reverse primer, 0.4 μl ROX reference dye, 1 μl template DNA, and 7.6 μl sterile distilled water. After a denaturation at 94 °C for 30 s, the reaction mixture was run through 40 cycles of denaturation at 94 °C for 5 s, annealing and extension at 58 °C for 20 s, followed by dissociation stage, with 1 cycle of denaturation at 94 °C for 15 s, annealing and extension step at 58 °C for 1 min and then at 94 °C for 15 s.

Results

Viable count of P. mirabilis V7

In order to detect the exact cell numbers of P. mirabilis V7 contained in one OD₆₀₀, colony counting experiment was carried out. Proteus mirabilis V7 was cultured in LB media to OD₆₀₀ of 0.7, and then the culture was diluted 10⁵- and 10⁶-fold respectively. One hundred microliters of each dilution was spread on LB agar plates in triplicate. After incubation at 28 °C for 16 h, the colonies that emerged on plates were enumerated. Five hundred, 480, and 442 colonies emerged on the plates when 100 μl of 10⁵-fold dilution was spread on plates, and 55, 50, and 49 colonies emerged on plates when 100 μl of 10⁶-fold dilution was spread on the plates. Calculation from these data showed that OD₆₀₀ of P. mirabilis culture at 1.0 approximately corresponded to 7.1 × 10⁸ CFU ml⁻¹ viable cells.

Specificity of P. mirabilis-specific PCR

As shown in Fig. 1, using the pair of designed specific primers as forward and reverse primers in PCR amplification, a 225-bp DNA product was amplified from P. mirabilis. The DNA product from P. mirabilis V7 was cloned into pBS-T and sequenced. The nucleotide sequence of DNA product, amplified by PCR using ureRF1 and ureRR1 as primers and P. mirabilis V7 cell as template, is shown in Fig. 2. The sequence showed 100 % similarity to the ureR sequence of P. mirabilis strain HI4320 with NCBI accession numbers AM942759 and Z18752. The result of the sequence alignment that the ureR homologue was found in no other bacterial species but P. mirabilis led us to wonder whether ureR is a gene unique to P. mirabilis. To investigate this speculation, we examined the prevalence of ureR in P. mirabilis strains and other different bacterial species collected from different environments. As shown in Fig. 1, the 225-bp DNA product presented in all three P. mirabilis strains V7, SW11 and SY2, isolated from differently geographical environments; but no DNA product was observed in the other bacterial strains, including Acinetobacter sp., Klebsiella sp., Vibrio sp., Enterobacter sp., Edwarsiella sp., and Pseudomonas sp., especially the ureR homolog contained P. stuartii and E. coli strains. Together these results demonstrated that using the designed primers, the 225-bp DNA fragment only appeared in PCR product using P. mirabilis cell as template, and thus the ureR-based PCR method can be used for specific detection of P. mirabilis.

Fig. 1

PCR amplifications using P. mirabilis cells and non-P. mirabilis cells as templates and ureRF1 and ureRR1 as primers. Templates used were the cells of E. hormaechei Pd9 and E. cloacae V2 (lanes 1 and 2); Acinetobacter sp. HS51 and A. calcoaceticus T32 (lanes 3 and 4); Klebsiella sp. Pd2 and K. pneumoniae Pd8 (lanes 5 and 6); Vibrio sp. V134, V. harveyi T4, V. parahaemolyticus BL, V. ichthyoenteri BX (lanes 7–10), E. tarda T2 (lane 11); Pseudomonas sp. J61 and DX7, P. plecoglossicida DJ3, P. aeruginosa YK, P. putida SP1 (lanes 12–16); P. stuartii Ps1 (lane 17); Escherichia sp. HS21 and E. coli Top10 and HS11 (lanes 18–20); P. mirabilis V7, SW11 and SY2 (lanes 21–23); DNA marker (lane 24)

Fig. 2

Nucleotide sequence of the amplified DNA product from P. mirabilis V7. The sequence of the primers is underlined

Sensitivity of P. mirabilis-specific PCR

To determine the sensitivity of the developed ureR-based PCR method, P. mirabilis V7 was grown in LB media to logarithmic phase and then was 10-fold serially diluted from 1.0 × 10⁷ CFU ml⁻¹ to 1.0 × 10 CFU ml⁻¹. One microlitre of each diluted culture was applied to PCR amplification. As shown in Fig. 3 (lanes 23–29), clear positive DNA bands were observed with concentrations of 1.0 × 10³–1.0 × 10⁷ CFU ml⁻¹. A weak positive result (i.e. a faint band) was observed with concentration of 1.0 × 10² CFU ml⁻¹, and this result was not always reproducible.

Fig. 3

Sensitivity of P. mirabilis-specific PCR. Lane 1 DNA marker; lanes 2–8, P. mirabilis spiked in urine at concentrations of 1.0×10¹, 1.0×10², 1.0×10³, 1.0×10⁴, 1.0×10⁵, 1.0×10⁶ and 1.0×10⁷ CFU ml^-1; lane 9–15, P. mirabilis spiked in seawater at concentrations of 1.0×10¹, 1.0×10², 1.0×10³, 1.0×10⁴, 1.0×10⁵, 1.0×10⁶ and 1.0×10⁷ CFU ml^-1; lane 16–22, P. mirabilis spiked in lake water at concentrations of 1.0×10¹, 1.0×10², 1.0×10³, 1.0×10⁴, 1.0×10⁵, 1.0×10⁶ and 1.0×10⁷ CFU ml^-1; lane 23–29, P. mirabilis spiked in ddH2O at concentrations of 1.0×10¹, 1.0×10², 1.0×10³, 1.0×10⁴, 1.0×10⁵, 1.0×10⁶ and 1.0×10⁷ CFU ml^-1

Detection of P. mirabilis from environmental samples

With the above results, we wondered whether the ureR-based PCR method could be applicable for detection of P. mirabilis from environmental samples. To investigate this speculation, natural seawater, lake water and urine of humans were collected. One microliter of each original sample was used as template for PCR amplification. However, no DNA product was obtained after amplification using natural samples as templates. Natural samples spiked with different P. mirabilis concentrations were then used for P. mirabilis detection. Results showed that clearly positive DNA bands were observed with concentrations of 1.0 × 10³–1.0 × 10⁷ CFU ml⁻¹ P. mirabilis in lake water (Fig. 3, lanes 16–22); clear positive DNA bands were observed with concentrations of 1.0 × 10³–1.0 × 10⁷ CFU ml⁻¹ P. mirabilis in seawater, but with weaker DNA bands (Fig. 3, lanes 9–15); however, positive DNA bands were observed with concentrations of 1.0 × 10⁴–1.0 × 10⁷ CFU ml⁻¹ P. mirabilis in urine (Fig. 3, lanes 2–8), and both PCR products obtained using P. mirabilis spiked in urine and in heated urine at concentration of 1.0 × 10⁴ CFU ml⁻¹ as template showed the same faint DNA bands on agarose gel.

Quantification of P. mirabilis by real-time PCR

To detect sensitivity of the ureR-based PCR when using DNA as template, serial dilutions of genomic DNA of P. mirabilis V7 were applied to real-time PCR. The negative control showed the largest cycle threshold values (Ct), the number of cycles required for the fluorescent signal to cross the threshold (i.e. exceeds background level). A linear relationship between DNA input, with DNA concentrations ranging from 85 fg μl⁻¹ to 85 ng μl⁻¹, and Ct values was observed with R ² = 0.9967 (Fig. 4a). In order to quantify the exact cell number of P. mirabilis, 1 ml of 10-fold diluted P. mirabilis culture was used to extract genomic DNA, and 1 μl of each extracted DNA was used as template in real-time PCR. Fig. 4b shows that the Ct values had a good linear relationship within a certain cell concentrations ranging from 1.0 × 10 CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹ with R ² = 0.9397. The Ct values had a better linear relationship within cell concentrations ranging from 1.0 × 10⁴ CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹ with R ² = 0.9905 (Fig. 4c). This result showed that using the ureR-based real-time PCR method, P. mirabilis concentration at 10 CFU ml⁻¹ can be detected and quantified. Higher P. mirabilis concentrations, from 1.0 × 10⁴ CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹, was more accurately quantified through the developed ureR-based real-time PCR method.

Fig. 4

Relationship between 10-fold serially diluted DNA and Ct value (a). A linear range was observed for DNA concentrations from 85 fg μl⁻¹ to 85 ng μl⁻¹. Relationship between the serially diluted P. mirabilis cells and Ct values obtained (b, c). Lines indicate results of regression analysis (triplicate samples). A linear range was obtained from 1.0 × 10¹ CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹ P. mirabilis cells (b), and a more rigorous linear range was obtained from 1.0 × 10⁴ CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹ P. mirabilis cells (c)

Discussion

In the last two decades, culture-independent molecular approaches have undergone considerable development in microbial ecology. Compared with culture-based methods, the PCR method is faster, more sensitive and more specific to detect bacterial strains. Moreover, it allows detection of dead cells and viable but non-cultivable cells (Postollec et al. 2011). Several PCR-based methods have been developed to uniquely and rapidly detect P. mirabilis specimens. 16S rRNA, ureC and a function unknown gene, were employed in previous PCR methods for P. mirabilis detection. Compared with the genetic markers 16S rRNA, the housekeeping gene that is preserved in almost all bacterial species, one may speculate that ureR-based molecular method should be more specific and discriminating in the detection of P. mirabilis (Huang et al. 1999; Limanskiĭ et al. 2005). The result of sequence blast in NCBI that ureR uniquely presented in P. mirabilis further strengthened this speculation.

For practical reasons, a good diagnostic method should possess the qualities of being sensitive, specific and time-efficient. The ureR-based PCR method can be applied directly to crude environmental samples, without, as is required in the previously reported PCR detection methods, cell culturing and DNA preparation. The whole detection process of the ureR-based PCR method can be completed in less than 3 h, involving only a PCR amplification and subsequent resolution of the PCR products by electrophoresis in an agarose gel. The comparison between the methods developed previously and the one developed in this study was listed in Table 2. Compared with the PCR method developed by Mansy et al. (1999), the product of ureR-based PCR was much smaller and thus it was more time-saving. ureR-based PCR detected less bacterial cell numbers than the ureC-based PCR method developed by Huang et al. (1999) and Takeuchi et al. (1996) . By ordinary PCR, P. mirabilis concentrations higher than 1.0 × 10³ CFU ml⁻¹ could be detected from the distilled water, lake water and seawater; however, P. mirabilis concentrations higher than 1.0 × 10⁴ CFU ml⁻¹ could be detected from urine. These features should enable this method to be applicable to various situations often encountered in practical industries, especially to those emergency situations that demand instant diagnosis. The still less DNA product from heated urine excluded the possibility of degrading DNA product by DNase, and thus the less DNA product was probably due to the presence of a PCR inhibitor, such as urea, in urine (Abolmaaty et al. 2007).

Table 2 Comparison of the PCR-based method for the detection of P. mirabilis

The real-time PCR protocol described in this study, including the process of DNA extraction, amplification, dissociation and data collection, can be done within 3 h. Moreover, the ureR-based real-time PCR described here can be used for quantitatively detecting P. mirabilis DNA from 85 fg μl⁻¹ to 85 ng μl⁻¹, and P. mirabilis cell number from 1.0 × 10 CFU ml⁻¹ to 1.0 × 10⁸ CFU ml⁻¹, respectively. The detection of such a small amount of DNA and small cell numbers may enable the DNA product to be amplified directly from environmental samples and may avoid the extra step of propagating bacteria overnight in culture media. This publication has shown the possibility to follow the growth of P. mirabilis in complex environments and has highlighted the potential of molecular approaches in assisting in controlling industrial processes such as has previously been reported (Hagi et al. 2010; Nakayama et al. 2007).