Detection of Mycobacterium tuberculosis complex and Mycobacterium gordonae on the same portable surface plasmon resonance sensor

Biosensors and Bioelectronics 26 (2010) 908–912

Contents lists available at ScienceDirect

Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios

Short communication

Detection of Mycobacterium tuberculosis complex and Mycobacterium gordonae on the same portable surface plasmon resonance sensor Memed Duman, Erhan Piskin ∗ Hacettepe University, Chemical Engineering Department and Bioengineering Division, and R&D Center for Bioengineering and Biyomedtek, 06800 Ankara, Turkey

a r t i c l e

i n f o

Article history: Received 4 April 2010 Received in revised form 29 June 2010 Accepted 30 June 2010 Available online 8 July 2010 Keywords: Tuberculosis Mycobacterium tuberculosis complex Mycobacterium gordonae Surface plasmon resonance sensors

a b s t r a c t In the present study, we have developed a specific detection system for Mycobacterium tuberculosis complex (MTB complex) and Mycobacterium gordonae by using a commercially available SPR based portable-multichannel sensor system. The probe single-strand oligodeoxynucleotides (probe-ssODNs), which also contain suitable spacer arms, against the target characteristic sequence (target-ssODNs) of both species were selected and synthesized. The SPR sensors were prepared by direct coupling of thiolated probes on gold-coated sensor surfaces. 6-Mercapto-1-hexanol was used as orientation helper and surface blocking agent. Immobilization protocol was optimized. The validation test results showed that the detection limit of sensor platform has been found to be 30 ng ␮l−1 . Developed sensor can be used to detect specific DNA hybridization at a concentration of 0.05 ␮M. The sensor chip surface can be regenerated with exposure to 2.5 mM HCl quite effectively and reused several times without losing the signal intensity. The SPR sensors carrying the probe-ssODNs were kept in vacuum at room temperature in the dark for about 12 weeks, and used effectively. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Mycobacterium tuberculosis has been present in the human population since antiquity—fragments of the spinal column from Egyptian mummies from 2400 B.C. showed definite pathological signs of tubercular decay. Tuberculosis (TB) is a common and deadly infectious disease caused by several mycobacteria especially M. tuberculosis complex (MTB complex) consisting of M. tuberculosis, M. bovis, M. africanum, M. microti. TB is considered a widely spread infectious disease with a significant social impact around the world. According to World Health Organization (WHO Global Tuberculosis Control Report, 2009), globally, there were about 10 million incident cases and 13.7 million prevalent cases of TB in 2007. There were also about 1.7 million deaths from TB in 2007 occurring world-wide. MTB complex is among the most common mycobacterial species causing TB in human, therefore it was selected the main target in this study. Mycobacterium gordonae, typically regarded as a colonizing organism, is a nontuberculous mycobacterium (NTM) which is found primarily in natural and tap water, but is also found in soil, dust, animals, and foods (Falkinham, 1996, 2002). M. gordonae is not defined as a pathogen since it rarely causes pulmonary disease in human and no evidence of person-to-person transmission. Beside the low pathogenic potential of M. gordonae causing pulmonary

∗ Corresponding author. Tel.: +90 532 707 9468; fax: +90 312 440 6214. E-mail address: [email protected] (E. Piskin). 0956-5663/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2010.06.071

disease in humans, a few cases of infection have been reported (Kumar and Varkey, 1980; Guarderas et al., 1986). However, it is hard to detect real mycobacterial infection from colonization due to M. gordonae, which is present throughout the environment and is a harmless organism. Moreover, the presence of M. gordonae in hospital water system and other environments causes contamination and detection as a false positive result in many conventional TB tests (Jackson et al., 1996; Eckburg et al., 2000). Therefore, distinguishing M. gordonae among MTB complex species is essential. Methods for diagnosis of tuberculosis vary in complexity from simple and inexpensive smear microscopy to more complex and expensive nucleic acid or antibody based techniques. Both conventional and new diagnostic tests for tuberculosis have limitations in terms of long assay time (days), lack of uniformity and reproducibility among different laboratories, high reagent cost, requirement of sophisticated automation and skilled personnel (Arora et al., 2006). It is clear that there is still a strong need for new diagnostic technologies with higher sensitivity, selectivity, reproducibility, reusability, and portability, which are cost effective, rapid and simple to use. Over the last decade, a great deal of research has focused on developing optical biosensors for the detection of microorganisms, cells and viruses while allowing rapid, real time and label-free identification with increased sensitivity. Surface plasmon resonance (SPR) is a rather new optical technique where interactions between biomolecules can be monitored without using any labels. SPR has been applied to the measurements of ligand–receptor interactions, drug screening, DNA hybridization, enzyme–substrate interactions,

M. Duman, E. Piskin / Biosensors and Bioelectronics 26 (2010) 908–912

polyclonal antibody characterization, epitope mapping and labelfree immunoassays (Shankaran et al., 2007; Fan et al., 2008; Hoa et al., 2007; Homola et al., 1999; McDonnell, 2001; Englebienne et al., 2003). In the present work, we have developed a three channel mini (portable) SPR system for label-free detection of MTB complex and M. gordonae. A thiol-derivatized single-strand oligodeoxynucleotides (ssODNs), which are complementary of the target characteristic sequence of MTB complex and M. gordonae were used as the “probes” and immobilized different channel of SPR chip. A probe-immobilization protocol was developed and the sensor carrying the probes as nano-sublayers was used for the detection of the both “target” ssODNs of MTB complex and M. gordonae. A non-complementary ssODNs was also used as control to show the selectivity of the sensor.

909

(MCH) molecules and used (without probe immobilization) as a control. The following procedure was applied to determine optimum probe-ssODNs immobilization conditions: (i) probe-ssODNs with different concentrations (0.05–2.0 ␮M) in the immobilization solution (1 M KH2 PO4 with a pH of 3.8) were incubated with the sensor surface for different periods of time (10–1200 min), (ii) for better orientation of the probe-ssODNs molecules and to eliminate the non-specific absorption on the sensor surfaces, MCH (“orientation helper and surface blocking agent”) was co-immobilized by injecting solutions having MCH with different concentrations (0–4.0 mM) for different periods of time (0–240 min), and (iii) to check the effects of these immobilization conditions on hybridization, pre-determined amounts of the target-ssODNs (1.0 ␮M) in the hybridization buffer (150 mM NaCl, 20 mM Na2 HPO4 , 0.1 mM EDTA, 0.005% Tween 20, pH 7.4) were introduced to the sensor channels and sensor response monitored for each case.

2. Materials and methods 2.4. Performance of SPR sensor

2.1. Instrumentations and reagents A SPREETATM , SPR3 Model 1146643 (Sensata, USA) was used as the SPR device and 3 channel gold-coated SpreetaTM , SPR3SM type TSPR1K23 sensors were used as the SPR chips. All other reagents that were used for blocking solution (6-mercapto-1hexanol, MCH), immobilization solution (1 M KH2 PO4 , pH 3.8) and hybridization buffer (150 mM NaCl, 20 mM Na2 HPO4 , 0.1 mM EDTA, 0.005% Tween 20, pH 7.4) were purchased from Sigma–Aldrich (USA). The gold-coated SpreetaTM sensors’ surfaces were cleaned by using UV/ozone chamber (Irvine, CA: Model 42, Jelight Company Inc., USA) before use. The buffer and chemical solutions were prepared by using MilliQ water (18 M). All SPR measurements were conducted at 25 ◦ C. ISMATEC (MCP-Process IP 65, Württemberg, Germany) peristaltic pump was used to maintain 5 ␮l min−1 flow rate. 2.2. Single-strand oligonucleotides Single-strand target oligodeoxynucleotides (target-ssODNs) for MTB complex and M. gordonae were selected according to their internal transcribed spacer (ITS) sequence (Park et al., 2005). The complementary sequences of both target-ssODNs were used as probe-ssODNs. For active immobilization of the probes onto the sensor surfaces, a “HS-(CH2 )6 -(T)15 ” spacer arm group was added to the 5 ends of the probes. A non-complementary ssODNs, which is the same size with the MTB complex’s target-ssODNs but having different a nucleotide sequence was used to exhibit the specificity of the sensors. The oligonucleotide sequences, purchased from TIB-MOLBIOL (Germany) and used from the 5 to the 3 end within this text were: MTB complex target-ssODNs M. gordonae target-ssODNs MTB complex probe-ssODNs M. gordonae probe-ssODNs Non-complementary ssODNs

(MTB-target-ssODNs)

CCAACTTTGTTGTCATGCACCC

(GOR-target-ssODNs)

GACAGCACCCGACGGTG

(MTB-probe-ssODNs)

(HS-(CH2 )6 -(T)15 GGGTGCATGACAACAAAGTTGG (HS-(CH2 )6 -(T)15 CACCCTCGGGTGCTGTC AGTCAATGCGGTAAACCCGACT

(GOR-probe-ssODNs) (NCS-ssODNs)

2.3. Immobilization of probe-ssODNs onto SPR sensor surfaces The gold sensor surfaces were cleaned and characterized before immobilization step as described previously (Duman et al., 2009). The MTB-probe-ODNs and GOR-probe-ODNs were immobilized onto the two different channels of the same SPR sensor chips while the last channel was just covered with mercaptohexanol

In order to determine sensor performance (sensitivity and selectivity) and to obtain the calibration curve of both targetssODNs, two different SPR chips were used. Two channels of each of the gold-coated SPR chips were prepared with immobilized probe-ssODNs. One of the two channels was treated with hybridization buffer containing different amounts (from 0.05 ␮M to 2.0 ␮M) of the target-ssODNs. The other channel was treated with hybridization buffer containing non-complementary strands (“NCS-ssODNs”). The 3rd channel was always used to determine the non-specific interaction of ssODNs with sensor surface. For this reason, this channel was only covered with MCH molecules. Both target oligonucleotides were incubated with each channel separately and hybridization was monitored online. Hybridizations on the sensor surfaces were monitored by using a SPR device for “real-time” detection of the hybridization reaction with a kinetic data acquisition mode for 20 min at room temperature and a constant flow rate of 5 ␮l min−1 . The analytical signal was reported as relative refractive index unit (RIU). Hybridizations were studied at four different temperatures (10 ◦ C, 25 ◦ C, 37 ◦ C, and 45 ◦ C). Regeneration of the SPR sensor and reuse was also tested by using 2.5 mM HCl as the regeneration medium, and 1 ␮M targetssODNs (Manneli et al., 2005), respectively. 3. Results and discussion 3.1. Optimization of SPR sensor surface The first step in the preparation of an oligonucleotide based sensor is the immobilization of the probe-ssODNs onto the sensor platform at a correctly oriented confirmation and concentration. The ssODNs-SH can be self-assembled onto gold surfaces via the thiol end groups rather easily. However, the interactions of the immobilized ssODNs with their complements at the surface are limited because of either densely packed self-assembled ssODNsSH molecules monolayers (SAMs) or their non-specific absorption on the gold surface from nitrogen-containing purine and pyrimidine bases (Herne and Tarlov, 1997). Therefore, the probe-surface concentration should be optimized and probe-ODNs should be oriented in a vertical conformation on the surfaces. Generally 6mercapto-1-hexanol (“SH-(CH2 )6 -OH” (MCH)) has been used as a helper molecule in several similar studies to prevent these two drawbacks (Herne and Tarlov, 1997; Steel et al., 2000; Ito et al., 2007). Use of the same length hydrocarbon chains of MCH molecules with both probe-ssODNs’ methylene group spacer, allows the MCH molecules to push probe molecules to reorganize (orient) on the

910

M. Duman, E. Piskin / Biosensors and Bioelectronics 26 (2010) 908–912

Fig. 1. Changes in the relative refractive indices: (A) operating curve obtained with MTB-target-ssODNs in a concentration range 0.05–2 ␮M; and (B) the linear range was up to 0.5 ␮M with a linear regression coefficient of 0.962.

surfaces in vertical positions. Moreover, the“–OH” end groups make the surface (after self-assembling) quite hydrophilic, which prevents non-specific interactions of the target-ssODNs molecules with the bare gold surfaces. On the basis of our previous findings (Duman et al., 2009), 1 mM MCH concentration and 1 h incubation time are the optimal conditions to achieve maximum coverage of MCH molecules onto sensor surfaces. In the second part of the optimization studies, different concentrations of probe-ssODNs and immobilization times were applied. Relative refractive indices increased initially with immobilization time but reached almost plateau values in about 120 min. Moreover, by increasing the probe-ssODNs concentration, the relative refractive indices of the oligonucleotide layers formed on the gold sensor surfaces first increased significantly and reached almost plateau values in a concentration of 1 ␮M (82 ± 8.2 × 10−6 RIU). For higher concentrations, this value decreased which may be due to the increase of steric hindrances between the competing oligonucleotide molecules (within the medium) on the surface (Duman et al., 2009). Considering the results given in our previous study, the following self-assembling protocol were applied to prepare the SPR sensor platforms for the further steps: first, 1 ␮M probe-ssODNs were self-assembled on the sensor platforms for 2 h and then

they were incubated with 1 mM MCH molecules for 1 h in order to both prevent the non-specific adsorption and orient probessODNs (Herne and Tarlov, 1997; Levicky et al., 1998; Steel et al., 1998). 3.2. Performance of SPR sensor In this part of the study, aqueous solutions of the targetssODNs with different concentrations in the range of 0.05–2.0 ␮M were flowed through the SPR sensor carrying the probe-oligos at room temperature and a constant flow rate of 5 ␮l min−1 . The hybridization reactions were monitored for about 20 min and the relative refractive indices of the sublayers were measured. Fig. 1A shows the changes in the relative refractive indices after the hybridization of MTB-probe-ssODNs with its complementary oligonucleotide molecules (MTB-target-ssODNs) with the increase of the target concentration from 0.05 ␮M to 2.0 ␮M. The maximum relative refractive index (82 ± 8.2 × 10−6 RIU) was reached when the target concentration was 1.0 ␮M. However, the SPR signal decreased to 57 ± 4.2 × 10−6 RIU when the MTB-target-ssODNs concentration was further increased to 2.0 ␮M. Most probably high concentration of target-ssODNs caused high mass transfer rates

Fig. 2. Changes in the relative refractive indices: (A) operating curve obtained with GOR-target-ssODNs in a concentration range 0.05–2 ␮M; and (B) the linear range was up to 1 ␮M with a linear regression coefficient of 0.989.

M. Duman, E. Piskin / Biosensors and Bioelectronics 26 (2010) 908–912

911

3.4. Reusability and stability Regeneration and storage stability of the SPR sensors prepared in this study were also investigated. Regeneration of the sensor was performed with the same sensor by treating surface with 2.5 mM HCl. After rinsing with buffer solution, hybridization was repeated using 1 ␮M target-ssODNs in hybridization buffer. Sensor still can be used even after 75th regeneration and sensor response decreased about 30%. Storage stability of SPR sensor was also evaluated by storing the probe-ssODNs carrying SPR sensors under vacuum at room temperature in the dark for about 12 weeks and then their performances were measured. The results showed that the changes in the measured values were not more than 10%. These observations clearly imply that the SPR sensors developed in this study are highly reusable and can be stable after storage. 4. Conclusion

Fig. 3. Changes in the refractive index responses in channel A (), B () and C (䊉) which were modified with MTB-probe-ssODNs, GOR-probe-ssODNs and only MHC molecules, respectively. All 3 channels of SPR sensor chips were incubated with mixed sample solution which contained MTB-target and non-complementary sequences. The responses are the representative results of three separate experiments.

within the medium directed to the sensor surfaces (due to high concentration difference-“the driving force”). This prevents complete hybridization and causes less mass accumulation (means low relative refractive index). The change of the relative refractive indices with the target concentration was linear only up to 0.5 ␮M with a linear regression coefficient of 0.962 (Fig. 1B). The prepared sensor performance and calibration curve were also examined for detection of M. gordonae. A steady increase in the SPR signal was observed when the target-ssODNs concentration increased up to 1 ␮M and the maximum relative refractive index was measured as 146 ± 13 × 10−6 RIU (Fig. 2A). Moreover, no significant increase in the SPR signal (at equilibrium) was observed when the sensing surface was treated 2 ␮M (148 ± 11 × 10−6 RIU). A calibration curve of GOR-target-ssODNs was also performed when the target ODNs concentration in 0–2 ␮M and linear relationship can be observed up to 1 ␮M with a regression coefficient of 0.989 (Fig. 2B). 3.3. Cross-selectivity and direct determination of MTB complex and M. gordonae Cross-selectivity of the SPR sensors was tested to determine the MTB complex in mixed sample solutions that contained both MTBtarget and non-complementary sequences. All experiments were carried out by 1 ␮M target oligonucleotides at room temperature and a flow rate 5 ␮l min−1 . The reaction was monitored for about 20 min. Fig. 3 presents changes in the refractive index responses in all 3 channels of SPR sensor chips when they were incubated with mixed sample solution. The samples were tested in triplicate (n = 3). As expected, we only observed an increase in relative refractive index in channel A which was carrying only the MTBprobe-ODNs. When the oligonucleotides in mixed sample solution reached sensor surfaces of the channel B and C, relative refractive index first increased up to 20 × 10−6 RIU and then decreased during the time in both channels. Since, the channel B was just modified with the GOR-probe-ODNs and channel C was as a control channel, the changes in relative refractive index (9.2 × 10−6 RIU), was caused of non-specific interactions between the MTB-target-ODNs and the surface of the channels.

MTB complex is among the most common mycobacterial species causing TB in humans, therefore it was selected the main target in this study. M. gordonae is found in hospital water systems and other environments causing false positives result in many conventional TB tests performed for the MTB complex. In the present study, we have successfully applied the use of a portable SPR system for specific detection of MTB complex and M. gordonae with the same sensor. The probe-ssODNs were selected according to the specific genes in the internal transcribed spacer (ITS) sequence of both MTB complex and M. gordonae and synthetically produced. The probessODNs have also spacer arms which increase the availability of the probe molecules immobilized onto the sensor platform for better interaction with their target-ssODNs. The SPR sensors were prepared by direct coupling of thiolated probes on gold-coated sensor surfaces. The results showed that developed sensor can be used to detect specific DNA hybridization at a concentration of 0.05 ␮M with a detection limit 30 ng ␮l−1 . In very similar study (Prabhakar et al., 2008), it has been showed that SPR based PNA biosensor showed better detection limit (1 ng ␮l−1 ) when they compared with SPR based DNA biosensor. These limits are quite good when they compare with most common TB detection methods, such as; microscopy and culturing methods. However, selectivity and long assay time are the major drawbacks of these techniques. Although, PCR based methods (Mishra et al., 2005) show highest sensitivity and selectivity, the high reagent cost, complexity of assay, labsize-equipments and requirement of skilled personnel are the main disadvantages of this technique. Furthermore, the developed platform operates with specificity in aqueous solutions at room temperature and completes hybridization within 20 min at the flow rates used in the present study. It was also demonstrated that the sensor platform can be regenerated with 2.5 mM HCl quite effectively and reused several times without losing the signal intensity. The shelf-life can be considered acceptable when the platforms carrying the probe-ssODNs are kept in vacuum at room temperature in the dark for about 12 weeks. This study showed that the SPREETATM SPR based portablemultichannel sensor system can be configured to allow detection of both specific MTB complex and M. gordonae via their specific DNA sequences in aqueous media by quite simple, rapid and relatively inexpensive approach for routine analysis. Acknowledgements This study, which was performed in the contest of Graduate Programme of Hacettepe University (a part of Ph.D. thesis of Memed Duman under the supervision of Prof. Erhan Pis¸kin), was also sup-

912

M. Duman, E. Piskin / Biosensors and Bioelectronics 26 (2010) 908–912

ported by Turkish Scientific and Technological Council (Turkish Scientific and Technological Research Council, TÜBI˙ TAK, Projects: 105T509 and 104T551. Erhan Pis¸kin was supported by Turkish Academy of Sciences (TÜBA) as a full member. Supports of Biyomedtek, and H.U.-R&D Center for Bioengineering is also kindly acknowledged. References Arora, K., Chand, S., Malhotra, B.D., 2006. Anal. Chim. Acta 568, 259–274. Duman, M., Caglayan, M.O., Demirel, G., Piskin, E., 2009. Sens. Lett. 7 (4), 535–542. Eckburg, P.B., Buadu, E.O., Stark, P., Sarinas, P.S.A., Chitkara, R.K., Kuschner, W.G., 2000. Chest 117, 96–102. Englebienne, P., Hoonacker, A.V., Verhas, M., 2003. Spectroscopy 17, 255–273. Falkinham III, J.O., 1996. Clin. Microbiol. Rev. 9, 177–215. Falkinham III, J.O., 2002. Clin. Chest Med. 23, 529–551. Fan, X., White, I.M., Shopova, S.I., Zhu, H., Suter, J.D., Sun, Y., 2008. Anal. Chim. Acta 620, 8–26. Guarderas, J., Alvarez, S., Berk, S.L., 1986. South Med. J. 79, 505–507. Herne, T.M., Tarlov, M.J., 1997. J. Am. Chem. Soc. 119, 8916–8920.

Hoa, X.D., Kirk, A.G., Tabrizian, M., 2007. Biosens. Bioelectron. 23, 151–160. Homola, J., Yee, S.S., Gauglitz, G., 1999. Sens. Actuators B 54, 3–15. Ito, M., Nakamura, F., Baba, A., Tamada, K., Ushijima, H., Lau, K.H.A., Manna, A., Knoll, W., 2007. J. Phys. Chem. C 111, 11653–11662. Jackson, J., Leggett, J.E., Wilson, D., Gilbert, D.N., 1996. Am. J. Infect. Control 24, 19–23. Kumar, U.N., Varkey, B., 1980. Br. J. Dis. Chest 74, 189–192. Levicky, R., Herne, T.M., Tarlov, M.J., Satija, S.K., 1998. J. Am. Chem. Soc. 120, 9787–9792. Manneli, I., Minunni, M., Tombelli, S., Wang, R., Spiriti, M.M., Mascini, M., 2005. Bioelectrochemistry 66, 129–138. McDonnell, M.J., 2001. Curr. Opin. Chem. Biol. 5, 572–577. Mishra, A., Singhal, A., Chauhan, D.S., Katoch, V.M., Srivastava, K., Thakral, S.S., Bharadwaj, S.S., Sreenivas, V., Prasad, H.K., 2005. J. Clin. Microbiol. 43, 5670–5678. Park, H., Jang, H., Song, E., Chang, C.L., Lee, M., Jeong, S., Park, J., Kang, B., Kim, C., 2005. J. Clin. Microbiol. 43, 1782–1788. Prabhakar, N., Arora, K., Arya, S.K., Solanki, P.R., Iwamoto, M., Singh, H., Malhotra, B.D., 2008. Analyst 133, 1587–1592. Shankaran, D.R., Gobi, K.V., Miura, N., 2007. Sens. Actuators B 121, 158–177. Steel, A.B., Herne, T.M., Tarlov, M.J., 1998. Anal. Chem. 70, 4670–4677. Steel, A.B., Levicky, R.L., Herne, T.M., Tarlov, M.J., 2000. Biophys. J. 79, 975–981. World Health Organization, 2009. Global Tuberculosis Control 2009 Report, http://www.who.int/tb/publications/global report/2009/en/index.html.