- Email: [email protected]
Paper-based ELISA to rapidly detect Escherichia coli
Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Paper-based ELISA to rapidly detect Escherichia coli Cheng-Min Shih a,b,1, Chia-Ling Chang c,1, Min-Yen Hsu b,d,e, Jyun-Yu Lin e, Chen-Meng Kuan e, Hsi-Kai Wang e, Chun-Te Huang f, Mu-Chi Chung f, Kui-Chou Huang a, Cheng-En Hsu a, Chun-Yuan Wang d, Ying-Cheng Shen d, Chao-Min Cheng e,n a
Department of Orthopedic Surgery, Taichung Veterans General Hospital, Taichung 40705, Taiwan Rong Hsing Research Center for Translational Medicine, National Chung Hsing University, Taiwan c Department of Nephrology, Tungs' Taichung MetroHarbor Hospital, Taichung, Taiwan d Department of Ophthalmology, Taichung Veterans General Hospital, Taichung 40705, Taiwan e Institute of Nanoengineering and Microsystems, National Tsing Hua University, Hsinchu 30013, Taiwan f Department of Nephrology, Taichung Veterans General Hospital, Taichung 40705, Taiwan b
art ic l e i nf o
a b s t r a c t
Article history: Received 16 April 2015 Received in revised form 14 July 2015 Accepted 15 July 2015
Escherichia coli is a generic indicator of fecal contamination, and certain serotypes cause food- and waterborne illness such as O157:H7. In the clinic, detection of bacteriuria, which is often due to E. coli, is critical before certain surgical procedures or in cases of nosocomial infection to prevent further adverse events such as postoperative infection or sepsis. In low- and middle-income countries, where insufficient equipment and facilities preclude modern methods of detection, a simple, low-cost diagnostic device to detect E. coli in water and in the clinic will have significant impact. We have developed a simple paperbased colorimetric platform to detect E. coli contamination in 5 h. On this platform, the mean color intensity for samples with 105 cells/mL is 0.118 7 0.002 (n ¼4), and 0.014570.003 (P o0.01nn) for uncontaminated samples. This technique is less time-consuming, easier to perform, and less expensive than conventional methods. Thus, paper-based ELISA is an innovative point-of-care diagnostic tool to rapidly detect E. coli, and possibly other pathogens when customized as appropriate, especially in areas that lack advanced clinical equipment. & 2015 Elsevier B.V. All rights reserved.
Keywords: Paper-based ELISA Escherichia coli Point-of-care diagnostics Urinary tract infection
1. Introduction A frequent contaminant in food and water, Escherichia coli accounts for 70–95% of all urinary tract infections (UTI) [1,2], which are one of the most common infections, and especially prevalent in the elderly and in women [3–6]. In the USA, UTI results in 1 million visits to the emergency room yearly, as well as more than 100,000 hospitalizations [7,8]. Urinary tract infection is due to bacteria or other microorganisms in the urine or in genitourinary tissues, which are normally sterile, and may affect the bladder, kidney, or, in men, the prostate. Bacteriuria, the presence of bacteria in urine, may progress to bacteremia and sepsis without appropriate treatment. In surgical procedures such as elective joint replacement, detection of bacteriuria even in the asymptomatic patient is critical, because it increases the risk of postoperative infection of the prosthetic joint [9]. Quantitative cultures of urine are the standard method to n
Correspondence to: No. 101, Sec. 2, Kuang-Fu Rd., Hsinchu 30013, Taiwan. E-mail address: [email protected] (C.-M. Cheng). 1 These authors contributed equally.
definitively diagnose UTI, and are recommended soon after onset of infection and prior to empiric topical therapy. According to the 2005 guidelines of the Infectious Diseases Society of America, bacteriuria is diagnosed in women when the same bacterial species is isolated from two consecutive voided urine specimens at concentrationsZ105 cfu/mL. In men, bacteriuria is defined as the presence of a single bacterial species at concentrationsZ105 cfu/mL in a single, clean-catch voided urine specimen [10]. However, quantitative cultures may not always be possible due to time constraints, cost, and availability. Although urinary traction infection can be resolved with administration of antibiotics at a dose at which the pathogen is susceptible, increasing prevalence of antibiotic resistance is of great concern and presents a considerable challenge [3,11,12]. In the past 10 years, resistance to ciprofloxacin and trimethoprim/ sulfamethoxazole has increased substantially in urinary E. coli isolates from outpatients [13]. Thus, prudent use of antibiotics to treat UTI is also becoming increasingly important, as resistance results in long treatment courses, easy relapse, or re-infection [14]. On the other hand, treatment that targets the wrong pathogen may allow UTI to further progress. Conventional urine culture requires 3–5 days. Thus, a point-ofcare diagnostic tool that significantly reduces turn-around time
http://dx.doi.org/10.1016/j.talanta.2015.07.051 0039-9140/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
may provide the basis for timely clinical decisions. For example, in cases where the infection is symptomatic and is not due to E. coli, the clinician may choose to select different antibiotics to target another organism, including Enterococcus sp. In this paper, we describe a platform that can confirm E. coli UTI or asymptomatic bacteriuria in 5 h. In addition, the method is simple and costeffective.
2. Materials and methods 2.1. Paper selection To select the optimal paper to use, 200-μL samples containing 109 cells/mL E. coli DH-5α were spotted on all available paper materials. Some materials were immediately excluded because of poor absorbency or excessive or limited wicking. As a result, colorimetric ELISA reactions were tested only on the remaining materials. Based on these experiments, Whatman Fusion 5TM paper was selected because of good absorption, acceptable wicking area, optimal wicking time, and homogeneous color development (see Supplementary content). This paper is a hydrophilic, singlelayered material based on glass fibers with a maximum pore size of 11 mm, and a thickness of 370 mm. This material has been used in a wide range of tests, especially in rapid diagnostic tests such as HIV viral loading detection and ocular VEGF level [15,16].
Fig. 2. The colorimetric results of different cell concentrations in standard solution at 40, 80, and 120 s. Over time, all the colorimetric colors were more and more distinguishable. The left column indicates a concentration of 109 cfu/mL and the right column indicates the control (with no bacteria).
2.2. Paper-based ELISA Test samples were prepared by suspending E. coli DH-5α as follows: DH5α competent cells were inoculated into 5 mL LB
Fig. 1. Schematic illustration of paper-based ELISA for E. coli. After proper paper selection, sample was added onto the paper surface (a). Then, the 1st antibody was added to conjugate with the sample (b). After proper blocking, secondary antibody was added followed by washing (c). Finally, colorimetric reaction was carried out (d). With different concentrations of sample, representative intensity can be shown after the final protocol and can be read on a portable scanner or smartphone camera (e).
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Mean intensity of paper-based ELISA for E. coli on readout. The intensity showed a significant difference between the concentration of 105 cells/mL and the control (Po 0.01**). The standard deviation of our paper-based ELISA was smaller and appeared to provide a more reliable and reproducible method compared with conventional techniques. P-valuea
Number of E. coli cells/mL
Mean intensity on paper-based ELISA (M 7 SDb, n¼ 4) a b
109
108
107
106
105
Control (No cell)
105 vs. control
1.4067 0.005
1.312 7 0.037
1.1467 0.015
0.587 7 0.011
0.118 70.002
0.01457 0.003
Po 0.01**
Kruskal–Wallis test. M 7SD ¼ mean 7standard deviation.
with 1% BSA, and washed 4 for 5 min each with 200 μL 0.05% Tween 20 in phosphate-buffered saline (Fig. 1b). Papers were then incubated for 1 h with 200 μL HRP-conjugated streptavidin, washed as described (Fig. 1c), and developed for 120 s with a mixture of 3,3′,5,5′-tetramethylbenzidine and H2O2 (Fig. 1d). Papers were serially photographed at 40 s, 80 s, and 120 s using a commercially available smartphone camera (HTC Inc., Taiwan) (Fig. 1e). The entire protocol is completed within 5 h. To obtain color intensity and analyze assay sensitivity, photographs were converted to 8-bit grayscale at 600 dpi using off-the-shelf image processing software (Adobe Photoshops CS5).
3. Results
Fig. 3. The association between colorimetric result intensity from paper-based ELISA for E. coli and E. coli concentration (cells/mL). The linear regression is Y ¼ 0.33X 1.396, R2 ¼0.911. The intensity difference between lowest colony number, 105 cells/mL, and control (no cells) is significant (Po 0.01, double asterisks).
medium and incubated overnight at 37 °C with shaking at 250 rpm. The overnight culture was used to inoculate LB medium. We adjusted the volume of LB medium until the absorbance at 600 nm was around 1 (optical density at 600 nm, OD600 ¼1 E1 109 cells/mL, in LB medium). The bacterial sample (1 109 cells/mL) was then serially diluted 10-, 102-, 103-, 104-fold with LB medium. In total, six different samples were prepared: “1 106 cells/mL”, “control, 0 cells/mL”, “1 105 cells/mL”, “1 107 cells/mL”, “1 108 cells/mL”, and “1 109 cells/mL”. Samples without cells were used as control. Aliquots of 200 μL were spotted on Whatman Fusion 5 (3 3 cm2), air-dried for 1 h, and lysed with phenol/chloroform. Papers were blocked with 200 μL 1% BSA) for 1 h (Fig. 1a), probed for 1 h with 200 μL anti-E. coli biotin conjugate(ABM inc., Canada) antibody diluted 1:1000
To evaluate the performance of paper-based ELISA to detect E. coli, we spotted 105–109 cells/mL on Whatman Fusion 5 paper, and photographed the colorimetric ELISA reaction using a smartphone camera (Fig. 2). On these photographs, samples with 105 E. coli cells/mL are visually distinguishable from samples without cells. To quantify signals, photographs were converted to 8-bit grayscale using Adobe Photoshop CS5. The mean intensity of 105 cells/mL was 0.118 70.002 (n ¼4), which is 8.1 fold higher than the negative control, the intensity of which was 0.014570.003 (Po 0.01) (Table 1). Moreover, the standard deviations of color measurements were small, indicating that the assay was robust and reproducible. Further, the intraclass correlation coefficient of the E. coli dilution series was 0.999, suggesting that multiple replicates of the same concentration generated very similar color measurements. Finally, the color intensity was linearly correlated with the logarithm of E. coli concentration, with R2 ¼0.911 (Fig. 3).
4. Discussion Patients who develop fever in a clinical setting receive tests to
Fig. 4. Schematic illustrations of possible flow-chart assisted by P-ELISA for E. coli. In the lower column, the urine culture for routine antibiotics prescription before surgery demands 3 5 days and further shift antibiotics or more specific can be carried out later. Under the assistance of P-ELISA for E.coli, physicians can shift antibiotics or prescribe more specific antibiotics after 5 h protocol of P-ELISA for E. coli.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
Table 2 Comparison of paper-based ELISA and conventional urine culture.
Advantages
Paper-based ELISA
Conventional urine culture
Fast ( o 5 h )
Detection of other uropathogens or possible antibiotic resistance Gold standard for UTI diagnosis
Low cost ( o US $1, per sample) Short training time Less medical waste Exam in ER or home
Disadvantages Detection of E. coli only Readout required
Science Council (NSC 101-2628-E-007-011-MY3 to C.-M. Cheng) and Taichung Veterans General Hospital (TCVGH-1046901B to M.Y. Hsu; TCVGH-1045101A to C.-M. Shih). The authors declare no conflicts of interest.
Appendix A. Supplementary Information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2015.07.051. 3–5 days for bacteria culture Cost US $30–60, per sample Professional training More medical waste (agar plate, urine culture tube, etc.) Exam in biosafety level 2 laboratory
identify the infecting organism, ideally before antibiotics are prescribed [17]. Such tests include cultures of sputum, urine, and blood. However, routine cultures require 3–5 days, and empiric or broad-spectrum antibiotics are typically prescribed while results are pending. Treatment may then be adjusted or modified depending on results. Thus, point-of-care diagnostics may enable physicians to quickly identify pathogens, and prescribe the most suitable antibiotics in a timely manner (Fig. 4). Rapid diagnosis is especially critical in patients with sepsis, which may end in septic shock or death without appropriate treatment [18]. Moreover, specimens delivered to a laboratory require refrigeration at 4 °C until tested, a requirement that can be a challenge in developing countries. This issue is especially critical for urine samples, in which contaminating organisms grow rapidly at room temperature and ultimately obscure or compromise results. Finally, pointof-care diagnostics for E. coli can be a vital surveillance tool to monitor food and water quality in resource-poor communities [19]. Paper-based ELISA for E. coli promises to be one such point-ofcare diagnostic tool, because it can be performed without modern medical equipment (Table 2). The method is fast, reliable, reproducible, and requires neither advanced procedures like PCR nor complex preprocessing of specimens. Therefore, the technique offers a number of advantages in comparison to other methods [20]. However, its performance remains to be validated in the clinic. In theory, paper-based ELISA should detect 105 CFUs/mL in urine, the quantitative threshold for diagnosis of UTI. Therefore, we anticipate that this method can be combined with standard urinary leukocyte tests to facilitate rapid detection of UTI due to E. coli.
5. Conclusion We have developed a paper-based platform to detect E. coli within 5 h. The platform can detect 105 cells/mL without requiring expensive or advanced equipment. However, its performance using actual clinical urine samples remains to be evaluated.
Acknowledgments We acknowledge financial support from Taiwan's National
References [1] A.R. Manges, J.R. Johnson, B. Foxman, T.T. O'Bryan, K.E. Fullerton, L.W. Riley, Widespread distribution of urinary tract infections caused by a multidrugresistant Escherichia coli clonal group, N. Engl. J. Med. 345 (2001) 1007–1013. [2] F. Stéphan F, H. Sax, M. Wachsmuth, P. Hoffmeyer, F. Clergue, D. Pittet, Reduction of urinary tract infection and antibiotic use after surgery: a controlled, prospective, before-after intervention study, Clin. Infect. Dis. 42 (2006) 1544–1551. [3] S. Salvatore, S. Salvatore, E. Cattoni, G. Siesto, M. Serati, P. Sorice, M. Torella, Urinary tract infections in women, Eur. J. Obstet. Gynecol. Reprod. Biol. 156 (2011) 131–136. [4] B.A. Lipsky, Prostatitis and urinary tract infection in men: what's new; what's true? Am. J. Med. 106 (1999) 327–334. [5] G.G. Anderson, J.J. Palermo, J.D. Schilling, R. Roth, J. Heuser, S.J. Hultgren, Intracellular bacterial biofilm-like pods in urinary tract infections, Science 301 (2003) 105–107. [6] J.B. Kaper, J.P. Nataro, H.L. Mobley, Pathogenic Escherichia coli, Nat. Rev. Microbiol. 2 (2004) 123–140. [7] T.S. David, M.S. Vrahas, Perioperative lower urinary tract infections and deep sepsis in patients undergoing total joint arthroplasty, J. Am. Acad. Orthop. Surg. 8 (2000) 66–74. [8] R. Orenstein, E.S. Wong, Urinary tract infections in adults, Am. Fam. Physician 59 (1999) 1225–1234 1237. [9] B. Foxman, Epidemiology of urinary tract infections: incidence, morbidity, and economic costs, Am. J. Med 113 (Suppl 1A) (2002) 5S–13S. [10] L.E. Nicolle, S. Bradley, R. Colgan, J.C. Rice, A. Schaeffer, T.M. Hooton, Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults, Clin. Infect. Dis. 40 (2005) 643–654. [11] J.A. Karlowsky, L.J. Kelly, C. Thornsberry, M.E. Jones, D.F. Sahm, Trends in antimicrobial resistance among urinary tract infection isolates of Escherichia coli from female outpatients in the United States, Antimicrob. Agents Chemother. 46 (2002) 2540–2545. [12] S.B. Levy, B. Marshall, Antibacterial resistance worldwide: causes, challenges and responses, Nat. Med. 10 (2004) S122–S129. [13] G.V. Sanchez, R.N. Master, J.A. Karlowsky, J.M. Bordon, In vitro antimicrobial resistance of urinary Escherichia coli isolates among U.S. outpatients from 2000 to 2010, Antimicrob. Agents Chemother. 56 (2012) 2181–2183. [14] H.C. Chung, C.H. Lai, J.N. Lin, C.K. Huang, S.H. Liang, W.F. Chen, Y.C. Shih, H. H. Lin, J.L. Wang, Bacteremia caused by extended-spectrum-beta-lactamaseproducing Escherichia coli sequence type ST131 and non-ST131 clones: comparison of demographic data, clinical features, and mortality, Antimicrob. Agents Chemother. 56 (2012) 618–622. [15] C.-M. Cheng, A.W. Martinez, J. Gong, C.R. Mace, S.T. Phillips, E. Carrilho, K. A. Mirica, G.M. Whitesides, Paper-based ELISA, Angew. Chem. Int. Ed. 49 (2010) 4771–4774. [16] M.-Y. Hsu, C.Y. Yang, W.H. Hsu, K.H. Lin, C.Y. Wang, Y.C. Shen, Y.C. Chen, S. F. Chau, H.Y. Tsai, C.-M. Cheng, Monitoring the VEGF level in aqueous humor of patients with ophthalmologically relevant diseases via ultrahigh sensitive paper-based ELISA, Biomaterials 35 (2014) 3729–3735. [17] W.E. Stamm, T.M. Hooton, Management of urinary tract infections in adults, N. Engl. J. Med 329 (1993) 1328–1334. [18] P.I. Tarr, C.A. Gordon, W.L. Chandler, Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome, Lancet 365 (2005) 1073–1086. [19] N.P. Pai, C. Vadnais, C. Denkinger, N. Engel, M. Pai, Point-of-care testing for infectious diseases: diversity, complexity, and barriers in low- and middleincome countries, PLoS Med. 9 (2012) e1001306. [20] F.B. Myers, L.P. Lee, Innovations in optical microfluidic technologies for pointof-care diagnostics, Lab Chip 8 (2008) 2015–2031.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Paper-based ELISA to rapidly detect Escherichia coli Cheng-Min Shih a,b,1, Chia-Ling Chang c,1, Min-Yen Hsu b,d,e, Jyun-Yu Lin e, Chen-Meng Kuan e, Hsi-Kai Wang e, Chun-Te Huang f, Mu-Chi Chung f, Kui-Chou Huang a, Cheng-En Hsu a, Chun-Yuan Wang d, Ying-Cheng Shen d, Chao-Min Cheng e,n a
Department of Orthopedic Surgery, Taichung Veterans General Hospital, Taichung 40705, Taiwan Rong Hsing Research Center for Translational Medicine, National Chung Hsing University, Taiwan c Department of Nephrology, Tungs' Taichung MetroHarbor Hospital, Taichung, Taiwan d Department of Ophthalmology, Taichung Veterans General Hospital, Taichung 40705, Taiwan e Institute of Nanoengineering and Microsystems, National Tsing Hua University, Hsinchu 30013, Taiwan f Department of Nephrology, Taichung Veterans General Hospital, Taichung 40705, Taiwan b
art ic l e i nf o
a b s t r a c t
Article history: Received 16 April 2015 Received in revised form 14 July 2015 Accepted 15 July 2015
Escherichia coli is a generic indicator of fecal contamination, and certain serotypes cause food- and waterborne illness such as O157:H7. In the clinic, detection of bacteriuria, which is often due to E. coli, is critical before certain surgical procedures or in cases of nosocomial infection to prevent further adverse events such as postoperative infection or sepsis. In low- and middle-income countries, where insufficient equipment and facilities preclude modern methods of detection, a simple, low-cost diagnostic device to detect E. coli in water and in the clinic will have significant impact. We have developed a simple paperbased colorimetric platform to detect E. coli contamination in 5 h. On this platform, the mean color intensity for samples with 105 cells/mL is 0.118 7 0.002 (n ¼4), and 0.014570.003 (P o0.01nn) for uncontaminated samples. This technique is less time-consuming, easier to perform, and less expensive than conventional methods. Thus, paper-based ELISA is an innovative point-of-care diagnostic tool to rapidly detect E. coli, and possibly other pathogens when customized as appropriate, especially in areas that lack advanced clinical equipment. & 2015 Elsevier B.V. All rights reserved.
Keywords: Paper-based ELISA Escherichia coli Point-of-care diagnostics Urinary tract infection
1. Introduction A frequent contaminant in food and water, Escherichia coli accounts for 70–95% of all urinary tract infections (UTI) [1,2], which are one of the most common infections, and especially prevalent in the elderly and in women [3–6]. In the USA, UTI results in 1 million visits to the emergency room yearly, as well as more than 100,000 hospitalizations [7,8]. Urinary tract infection is due to bacteria or other microorganisms in the urine or in genitourinary tissues, which are normally sterile, and may affect the bladder, kidney, or, in men, the prostate. Bacteriuria, the presence of bacteria in urine, may progress to bacteremia and sepsis without appropriate treatment. In surgical procedures such as elective joint replacement, detection of bacteriuria even in the asymptomatic patient is critical, because it increases the risk of postoperative infection of the prosthetic joint [9]. Quantitative cultures of urine are the standard method to n
Correspondence to: No. 101, Sec. 2, Kuang-Fu Rd., Hsinchu 30013, Taiwan. E-mail address: [email protected] (C.-M. Cheng). 1 These authors contributed equally.
definitively diagnose UTI, and are recommended soon after onset of infection and prior to empiric topical therapy. According to the 2005 guidelines of the Infectious Diseases Society of America, bacteriuria is diagnosed in women when the same bacterial species is isolated from two consecutive voided urine specimens at concentrationsZ105 cfu/mL. In men, bacteriuria is defined as the presence of a single bacterial species at concentrationsZ105 cfu/mL in a single, clean-catch voided urine specimen [10]. However, quantitative cultures may not always be possible due to time constraints, cost, and availability. Although urinary traction infection can be resolved with administration of antibiotics at a dose at which the pathogen is susceptible, increasing prevalence of antibiotic resistance is of great concern and presents a considerable challenge [3,11,12]. In the past 10 years, resistance to ciprofloxacin and trimethoprim/ sulfamethoxazole has increased substantially in urinary E. coli isolates from outpatients [13]. Thus, prudent use of antibiotics to treat UTI is also becoming increasingly important, as resistance results in long treatment courses, easy relapse, or re-infection [14]. On the other hand, treatment that targets the wrong pathogen may allow UTI to further progress. Conventional urine culture requires 3–5 days. Thus, a point-ofcare diagnostic tool that significantly reduces turn-around time
http://dx.doi.org/10.1016/j.talanta.2015.07.051 0039-9140/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
may provide the basis for timely clinical decisions. For example, in cases where the infection is symptomatic and is not due to E. coli, the clinician may choose to select different antibiotics to target another organism, including Enterococcus sp. In this paper, we describe a platform that can confirm E. coli UTI or asymptomatic bacteriuria in 5 h. In addition, the method is simple and costeffective.
2. Materials and methods 2.1. Paper selection To select the optimal paper to use, 200-μL samples containing 109 cells/mL E. coli DH-5α were spotted on all available paper materials. Some materials were immediately excluded because of poor absorbency or excessive or limited wicking. As a result, colorimetric ELISA reactions were tested only on the remaining materials. Based on these experiments, Whatman Fusion 5TM paper was selected because of good absorption, acceptable wicking area, optimal wicking time, and homogeneous color development (see Supplementary content). This paper is a hydrophilic, singlelayered material based on glass fibers with a maximum pore size of 11 mm, and a thickness of 370 mm. This material has been used in a wide range of tests, especially in rapid diagnostic tests such as HIV viral loading detection and ocular VEGF level [15,16].
Fig. 2. The colorimetric results of different cell concentrations in standard solution at 40, 80, and 120 s. Over time, all the colorimetric colors were more and more distinguishable. The left column indicates a concentration of 109 cfu/mL and the right column indicates the control (with no bacteria).
2.2. Paper-based ELISA Test samples were prepared by suspending E. coli DH-5α as follows: DH5α competent cells were inoculated into 5 mL LB
Fig. 1. Schematic illustration of paper-based ELISA for E. coli. After proper paper selection, sample was added onto the paper surface (a). Then, the 1st antibody was added to conjugate with the sample (b). After proper blocking, secondary antibody was added followed by washing (c). Finally, colorimetric reaction was carried out (d). With different concentrations of sample, representative intensity can be shown after the final protocol and can be read on a portable scanner or smartphone camera (e).
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Table 1 Mean intensity of paper-based ELISA for E. coli on readout. The intensity showed a significant difference between the concentration of 105 cells/mL and the control (Po 0.01**). The standard deviation of our paper-based ELISA was smaller and appeared to provide a more reliable and reproducible method compared with conventional techniques. P-valuea
Number of E. coli cells/mL
Mean intensity on paper-based ELISA (M 7 SDb, n¼ 4) a b
109
108
107
106
105
Control (No cell)
105 vs. control
1.4067 0.005
1.312 7 0.037
1.1467 0.015
0.587 7 0.011
0.118 70.002
0.01457 0.003
Po 0.01**
Kruskal–Wallis test. M 7SD ¼ mean 7standard deviation.
with 1% BSA, and washed 4 for 5 min each with 200 μL 0.05% Tween 20 in phosphate-buffered saline (Fig. 1b). Papers were then incubated for 1 h with 200 μL HRP-conjugated streptavidin, washed as described (Fig. 1c), and developed for 120 s with a mixture of 3,3′,5,5′-tetramethylbenzidine and H2O2 (Fig. 1d). Papers were serially photographed at 40 s, 80 s, and 120 s using a commercially available smartphone camera (HTC Inc., Taiwan) (Fig. 1e). The entire protocol is completed within 5 h. To obtain color intensity and analyze assay sensitivity, photographs were converted to 8-bit grayscale at 600 dpi using off-the-shelf image processing software (Adobe Photoshops CS5).
3. Results
Fig. 3. The association between colorimetric result intensity from paper-based ELISA for E. coli and E. coli concentration (cells/mL). The linear regression is Y ¼ 0.33X 1.396, R2 ¼0.911. The intensity difference between lowest colony number, 105 cells/mL, and control (no cells) is significant (Po 0.01, double asterisks).
medium and incubated overnight at 37 °C with shaking at 250 rpm. The overnight culture was used to inoculate LB medium. We adjusted the volume of LB medium until the absorbance at 600 nm was around 1 (optical density at 600 nm, OD600 ¼1 E1 109 cells/mL, in LB medium). The bacterial sample (1 109 cells/mL) was then serially diluted 10-, 102-, 103-, 104-fold with LB medium. In total, six different samples were prepared: “1 106 cells/mL”, “control, 0 cells/mL”, “1 105 cells/mL”, “1 107 cells/mL”, “1 108 cells/mL”, and “1 109 cells/mL”. Samples without cells were used as control. Aliquots of 200 μL were spotted on Whatman Fusion 5 (3 3 cm2), air-dried for 1 h, and lysed with phenol/chloroform. Papers were blocked with 200 μL 1% BSA) for 1 h (Fig. 1a), probed for 1 h with 200 μL anti-E. coli biotin conjugate(ABM inc., Canada) antibody diluted 1:1000
To evaluate the performance of paper-based ELISA to detect E. coli, we spotted 105–109 cells/mL on Whatman Fusion 5 paper, and photographed the colorimetric ELISA reaction using a smartphone camera (Fig. 2). On these photographs, samples with 105 E. coli cells/mL are visually distinguishable from samples without cells. To quantify signals, photographs were converted to 8-bit grayscale using Adobe Photoshop CS5. The mean intensity of 105 cells/mL was 0.118 70.002 (n ¼4), which is 8.1 fold higher than the negative control, the intensity of which was 0.014570.003 (Po 0.01) (Table 1). Moreover, the standard deviations of color measurements were small, indicating that the assay was robust and reproducible. Further, the intraclass correlation coefficient of the E. coli dilution series was 0.999, suggesting that multiple replicates of the same concentration generated very similar color measurements. Finally, the color intensity was linearly correlated with the logarithm of E. coli concentration, with R2 ¼0.911 (Fig. 3).
4. Discussion Patients who develop fever in a clinical setting receive tests to
Fig. 4. Schematic illustrations of possible flow-chart assisted by P-ELISA for E. coli. In the lower column, the urine culture for routine antibiotics prescription before surgery demands 3 5 days and further shift antibiotics or more specific can be carried out later. Under the assistance of P-ELISA for E.coli, physicians can shift antibiotics or prescribe more specific antibiotics after 5 h protocol of P-ELISA for E. coli.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i
C.-M. Shih et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
Table 2 Comparison of paper-based ELISA and conventional urine culture.
Advantages
Paper-based ELISA
Conventional urine culture
Fast ( o 5 h )
Detection of other uropathogens or possible antibiotic resistance Gold standard for UTI diagnosis
Low cost ( o US $1, per sample) Short training time Less medical waste Exam in ER or home
Disadvantages Detection of E. coli only Readout required
Science Council (NSC 101-2628-E-007-011-MY3 to C.-M. Cheng) and Taichung Veterans General Hospital (TCVGH-1046901B to M.Y. Hsu; TCVGH-1045101A to C.-M. Shih). The authors declare no conflicts of interest.
Appendix A. Supplementary Information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2015.07.051. 3–5 days for bacteria culture Cost US $30–60, per sample Professional training More medical waste (agar plate, urine culture tube, etc.) Exam in biosafety level 2 laboratory
identify the infecting organism, ideally before antibiotics are prescribed [17]. Such tests include cultures of sputum, urine, and blood. However, routine cultures require 3–5 days, and empiric or broad-spectrum antibiotics are typically prescribed while results are pending. Treatment may then be adjusted or modified depending on results. Thus, point-of-care diagnostics may enable physicians to quickly identify pathogens, and prescribe the most suitable antibiotics in a timely manner (Fig. 4). Rapid diagnosis is especially critical in patients with sepsis, which may end in septic shock or death without appropriate treatment [18]. Moreover, specimens delivered to a laboratory require refrigeration at 4 °C until tested, a requirement that can be a challenge in developing countries. This issue is especially critical for urine samples, in which contaminating organisms grow rapidly at room temperature and ultimately obscure or compromise results. Finally, pointof-care diagnostics for E. coli can be a vital surveillance tool to monitor food and water quality in resource-poor communities [19]. Paper-based ELISA for E. coli promises to be one such point-ofcare diagnostic tool, because it can be performed without modern medical equipment (Table 2). The method is fast, reliable, reproducible, and requires neither advanced procedures like PCR nor complex preprocessing of specimens. Therefore, the technique offers a number of advantages in comparison to other methods [20]. However, its performance remains to be validated in the clinic. In theory, paper-based ELISA should detect 105 CFUs/mL in urine, the quantitative threshold for diagnosis of UTI. Therefore, we anticipate that this method can be combined with standard urinary leukocyte tests to facilitate rapid detection of UTI due to E. coli.
5. Conclusion We have developed a paper-based platform to detect E. coli within 5 h. The platform can detect 105 cells/mL without requiring expensive or advanced equipment. However, its performance using actual clinical urine samples remains to be evaluated.
Acknowledgments We acknowledge financial support from Taiwan's National
References [1] A.R. Manges, J.R. Johnson, B. Foxman, T.T. O'Bryan, K.E. Fullerton, L.W. Riley, Widespread distribution of urinary tract infections caused by a multidrugresistant Escherichia coli clonal group, N. Engl. J. Med. 345 (2001) 1007–1013. [2] F. Stéphan F, H. Sax, M. Wachsmuth, P. Hoffmeyer, F. Clergue, D. Pittet, Reduction of urinary tract infection and antibiotic use after surgery: a controlled, prospective, before-after intervention study, Clin. Infect. Dis. 42 (2006) 1544–1551. [3] S. Salvatore, S. Salvatore, E. Cattoni, G. Siesto, M. Serati, P. Sorice, M. Torella, Urinary tract infections in women, Eur. J. Obstet. Gynecol. Reprod. Biol. 156 (2011) 131–136. [4] B.A. Lipsky, Prostatitis and urinary tract infection in men: what's new; what's true? Am. J. Med. 106 (1999) 327–334. [5] G.G. Anderson, J.J. Palermo, J.D. Schilling, R. Roth, J. Heuser, S.J. Hultgren, Intracellular bacterial biofilm-like pods in urinary tract infections, Science 301 (2003) 105–107. [6] J.B. Kaper, J.P. Nataro, H.L. Mobley, Pathogenic Escherichia coli, Nat. Rev. Microbiol. 2 (2004) 123–140. [7] T.S. David, M.S. Vrahas, Perioperative lower urinary tract infections and deep sepsis in patients undergoing total joint arthroplasty, J. Am. Acad. Orthop. Surg. 8 (2000) 66–74. [8] R. Orenstein, E.S. Wong, Urinary tract infections in adults, Am. Fam. Physician 59 (1999) 1225–1234 1237. [9] B. Foxman, Epidemiology of urinary tract infections: incidence, morbidity, and economic costs, Am. J. Med 113 (Suppl 1A) (2002) 5S–13S. [10] L.E. Nicolle, S. Bradley, R. Colgan, J.C. Rice, A. Schaeffer, T.M. Hooton, Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults, Clin. Infect. Dis. 40 (2005) 643–654. [11] J.A. Karlowsky, L.J. Kelly, C. Thornsberry, M.E. Jones, D.F. Sahm, Trends in antimicrobial resistance among urinary tract infection isolates of Escherichia coli from female outpatients in the United States, Antimicrob. Agents Chemother. 46 (2002) 2540–2545. [12] S.B. Levy, B. Marshall, Antibacterial resistance worldwide: causes, challenges and responses, Nat. Med. 10 (2004) S122–S129. [13] G.V. Sanchez, R.N. Master, J.A. Karlowsky, J.M. Bordon, In vitro antimicrobial resistance of urinary Escherichia coli isolates among U.S. outpatients from 2000 to 2010, Antimicrob. Agents Chemother. 56 (2012) 2181–2183. [14] H.C. Chung, C.H. Lai, J.N. Lin, C.K. Huang, S.H. Liang, W.F. Chen, Y.C. Shih, H. H. Lin, J.L. Wang, Bacteremia caused by extended-spectrum-beta-lactamaseproducing Escherichia coli sequence type ST131 and non-ST131 clones: comparison of demographic data, clinical features, and mortality, Antimicrob. Agents Chemother. 56 (2012) 618–622. [15] C.-M. Cheng, A.W. Martinez, J. Gong, C.R. Mace, S.T. Phillips, E. Carrilho, K. A. Mirica, G.M. Whitesides, Paper-based ELISA, Angew. Chem. Int. Ed. 49 (2010) 4771–4774. [16] M.-Y. Hsu, C.Y. Yang, W.H. Hsu, K.H. Lin, C.Y. Wang, Y.C. Shen, Y.C. Chen, S. F. Chau, H.Y. Tsai, C.-M. Cheng, Monitoring the VEGF level in aqueous humor of patients with ophthalmologically relevant diseases via ultrahigh sensitive paper-based ELISA, Biomaterials 35 (2014) 3729–3735. [17] W.E. Stamm, T.M. Hooton, Management of urinary tract infections in adults, N. Engl. J. Med 329 (1993) 1328–1334. [18] P.I. Tarr, C.A. Gordon, W.L. Chandler, Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome, Lancet 365 (2005) 1073–1086. [19] N.P. Pai, C. Vadnais, C. Denkinger, N. Engel, M. Pai, Point-of-care testing for infectious diseases: diversity, complexity, and barriers in low- and middleincome countries, PLoS Med. 9 (2012) e1001306. [20] F.B. Myers, L.P. Lee, Innovations in optical microfluidic technologies for pointof-care diagnostics, Lab Chip 8 (2008) 2015–2031.
Please cite this article as: C.-M. Shih, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.07.051i