Monitoring of Mycoplasma genitalium growth and evaluation of antibacterial activity of antibiotics tetracycline and levofloxacin using a wireless magnetoelastic sensor

Biosensors and Bioelectronics 24 (2009) 1990–1994

Contents lists available at ScienceDirect

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

Monitoring of Mycoplasma genitalium growth and evaluation of antibacterial activity of antibiotics tetracycline and levofloxacin using a wireless magnetoelastic sensor Bo He a,b , Lifu Liao a,b,∗ , Xilin Xiao a,c , Shuqin Gao a , Yimou Wu d a

College of Chemistry & Chemical Engineering, University of South China, Hengyang, Hunan 421001, China College of Public Health, University of South China, Hengyang, Hunan 421001, China State Key Laboratory of Chemo/Biosensing and Chemometrics, Hunan University, Changsha 410082, China d Institute of Pathogenic Biology, University of South China, Hengyang, Hunan 421001, China b c

a r t i c l e

i n f o

Article history: Received 11 July 2008 Received in revised form 20 September 2008 Accepted 6 October 2008 Available online 22 October 2008 Keywords: Magnetoelastic sensor Mycoplasma genitalium Wireless Tetracycline Levofloxacin

a b s t r a c t Mycoplasma genitalium (Mg) is the smallest and simplest self-replicating bacteria lacking of cell wall and is a human pathogen causing various diseases. This paper describes the real-time, long-term and in situ monitoring of the growth of Mg and evaluation of the effect of the antibiotics tetracycline and levofloxacin on the growth using a wireless magnetoelastic sensor. The sensor is fabricated by coating a magnetoelastic strip with a polyurethane protecting film. In response to a time-varying magnetic field, the sensor longitudinally vibrates at a resonance frequency, emitting magnetic flux that can be remotely detected by a pick-up coil. No physical connections between the sensor and the detection system are required. The wireless property facilitates aseptic operation. The adhesion of Mg on the sensor surface results in a decrease in the resonance frequency, which is proportional to the concentration of Mg. The shift of the resonance frequency–time curves shows that under routine culture condition the growth curve of Mg is composed of three phases those are lag, logarithmic and stationary phase, respectively. In the presence of the antibiotics, the lag phase in the growth inhibition curves is prolonged obviously and the stationary phase is substituted by a decline phase. The growth inhibition of Mg is related to the concentration of the antibiotics. The MIC50 (minimal inhibitory concentration) of Mg incubated in the presence of the antibiotics for 120 h is calculated to be 1.5 and 0.5 ␮g/mL for tetracycline and levofloxacin, respectively. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Mycoplasmas are the smallest and simplest self-replicating bacteria (Razin et al., 1998). These microorganisms lack a rigid cell wall and are bound by a single membrane, the plasma membrane. Mycoplasmas are resistant to some antibiotics such as penicillins and cephalosporins due to the lack of cell wall. Mycoplasma genitalium (Mg) is the smallest one among the mycoplasmas with a cell diameter of only 300 nm and a genome size of only 580 kbp (Fraser et al., 1995). Mg has a specialized tip structure by which it can attach to human urogenital epithelia. Mg can also adhere tightly on glass or plastic surface and glide on the surface. Mg is a human pathogen causing urogenital diseases, such as urethritis

∗ Corresponding author at: College of Chemistry & Chemical Engineering, University of South China, Xueyan Road, Hengyang, Hunan 421001, China. Tel.: +86 734 8280649; fax: +86 734 8282375. E-mail address: [email protected] (L. Liao). 0956-5663/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2008.10.005

(Yoshida et al., 2002; Deguchi and Maeda, 2002), cervicitis (Uno et al., 1997), endometritis (Cohen et al., 2002) and infertility (Clausen et al., 2001). For the investigation of the growth stage and the causing diseases mechanism of Mg, and for selecting proper antibiotics to inhibit the growth of Mg, the efficient methods those can be used to real-time, long-term and in situ monitor the growth and growth inhibition process of Mg are required. At present, several instrumental methods have been used to monitor microorganism viability and proliferation, such as flow cytometry (Assunc¸ão et al., 2005), quartz crystal microbalance (Nivens et al., 1993; Zhou et al., 2000; Reipa et al., 2006), acoustic wave impedance technique (Chang et al., 2007), and electrochemical-based sensor (Berrettoni et al., 2004). These methods are valuable and each method has its own advantages. However, due to the fact that these methods require either electrical connections or line-of-sight telemetry, there are some disadvantages in these methods. For example, it is not easy to execute aseptic operations and to remotely monitor the growth of microbe in sealed dark vessels with these meth-

B. He et al. / Biosensors and Bioelectronics 24 (2009) 1990–1994

ods. Furthermore, some of the methods require relatively expensive equipment. For detecting bacteria, although some counting methods, such as colony counting method (Putman et al., 2005) and color change unit (CCU) counting method (Stemke and Robertson, 1982), have also been used, these methods cannot be used to track bacterial growth process, and some of them such as colony counting method cannot be used to detect Mycoplasmas due to the lack of cell wall in Mycoplasmas. Magnetoelastic materials are made of amorphous ferromagnetic alloys composed of iron, nickel, molybdenum and boron. A magnetoelastic sensor oscillates at a fundamental resonant frequency in the presence of a time-varying magnetic field. These oscillations generate a magnetic flux around the sensor that can be detected by a non-contacting pickup coil. Therefore, magnetoelastic sensor techniques have the unique characteristics of being able to wirelessly detect the resonant frequency changes of the magnetoelastic strip. The resonant frequency is dependent on the physical dimensions of the magnetoelastic strip, the compositions attached on the strip and the environment around the strip, which facilitates magnetoelastic sensors to be used in various measurements. In recent years, this kind of sensors have been developed for the determination of physical parameters and chemical substances including pressure, temperature, humidity, flow rate, liquid viscosity, pH, carbon dioxide and ammonia (Grimes et al., 2002), and for the determination of biological substances such as glucose (Cai et al., 2004), bacteria (Ruan et al., 2003; Pang et al., 2007; Huang et al., 2008), trypsin (Wu et al., 2006), protein (Ruan et al., 2004), blood coagulation (Puckett et al., 2003), antigen immunoassay (Guntupalli et al., 2007) and breast cancer cell (Xiao et al., 2008). However, little attention has been directed to the use of the magnetoelastic sensors for the real-time, long-term and in situ monitoring of the growth of mycoplasmas. In the study described herein, a wireless magnetoelastic sensor was developed for the real-time, long-term and in situ monitoring of Mg growth and evaluation of the inhibiting activity of the antibiotics tetracycline and levofloxacin to Mg. The sensor is fabricated by coating the magnetoelastic ribbon-like strip with a polyurethane film that protects the iron-rich sensor from oxidation and provides a microbe-compatible surface. The adhesion of Mg on the sensor surface resulted in the decrease of the resonance frequency. The curves of the shift of the resonance frequency to the time clearly demonstrate Mg growth process and quantify the inhibiting activity of the antibiotics with satisfactory results. 2. Materials and methods 2.1. Materials Mg G37 strain was obtained from Institute of Pathogenic Biology, University of South China (Hengyang, China). Tetracycline, levofloxacin and the materials for the preparation of SP-4 liquid culture medium were purchased from BioDev-Tech. Co., Ltd. (Beijing, China). SP-4 culture medium was prepared according to the method described by Tully et al. (1979). All other chemicals were of analytical grade. A magnetoelastic ribbon of Metglas alloy 2826MB, with a composition of Fe40Ni38Mo4B18, was obtained from Honeywell Corporation (Morristown, NJ). The ribbon was then cut into strips as sensors with dimensions of 18 mm × 6 mm × 28 ␮m. The magnetoelastic sensors were ultrasonically cleaned in water and acetone, respectively, and then dried in a stream of nitrogen. To protect the iron-rich magnetoelastic sensors from corrosion and offered a microbe-compatible surface, on both sides of the sensor, a layer of Bayhydrol 110 (Bayer Corporation, Pittsburgh, Pennsylvania) was applied by dip-coating, and the polyurethane-coated sensors were

1991

Fig. 1. (a) Schematically showing the test setup. The magnetoelastic sensor is placed in a vial. The tube-like vial is then inserted into a small coil which was connected to a ‘sensor-reader’ box. (b) The measured resonant frequency at the beginning (1) and end (2) of an experiment lasted for 192 h.

dried in air and then annealed at 150 ◦ C for 2 h to form a robust membrane, and sterilized at 160 ◦ C for 120 min prior to use. 2.2. Methods A polyurethane-coated sensor was placed in a small tube-like vial containing 2 mL SP-4 culture medium. The vial was inserted into a small coil that was connected to a microprocessor-based frequency counting device, see Fig. 1(a), used for signal telemetry (Zeng et al., 2002). Mg G37 strain was inoculated in the culture medium and the vial as well as the coil was put in a CO2 incubator (37 ◦ C, 5% CO2 ). The Mg was incubated in the absence or in the presence of tetracycline or levofloxacin. During the process of incubation, the resonance frequency of the magnetoelastic sensor was detected continuously. 3. Results and discussion 3.1. Principle of operation Fig. 1 illustrates the operating principle of the magnetoelastic sensor. A time-varying magnetic field is used to excite the sensor, causing it to longitudinally vibrate at a resonant frequency. The mechanical vibrations of the magnetostrictive material, in turn, generate a magnetic flux that can be remotely detected using a simple pick-up coil. A microprocessor-based frequency count system

1992

B. He et al. / Biosensors and Bioelectronics 24 (2009) 1990–1994

Eq. (2) shows that the resonance frequency shifts linearly, decreasing with increasing mass on the sensor surface. Hence the binding of the target organism onto the sensor surface causes a mass increase with a corresponding decrease in fundamental resonant frequency. 3.2. Sensor responses and Mg growth curves

Fig. 2. The shift of the resonant frequency during the Mg G37 culture process in the absence of antibiotics.

is employed to detect the resonant frequency. No physical connections between the sensor and the detection system are required for signal telemetry. The fundamental resonant frequency f0 of the longitudinal vibrations is given by (Stoyanov and Grimes, 2000): f0 =

1 2L



E (1 −  2 )

(1)

where E is Young’s modulus of elasticity,  is the Poisson’s ratio,  is the density of the sensor material, and L is the longitudinal dimension of the sensor. When a small mass load m adhere to the surface of the sensor, the shift in the resonant frequency of the sensor is given by (Cai and Grimes, 2000): f = f − f0 = −f0

m 2m

(2)

where f is the shift in the resonant frequency of the sensor, f0 is the initial resonant frequency, m is the initial mass and m is the mass change.

The sensor was firstly incubated in a blank sterile SP-4 culture medium without adding Mg in a CO2 incubator (37 ◦ C, 5% CO2 ) for several hours to detect any possible nonspecific adsorptions, which were not observed. The resonant frequency of the sensor in the SP4 culture medium is 117.40 kHz. A certain amount of Mg G37 strain was then inoculated in the culture medium. During the process of incubation, the resonant frequency of the magnetoelastic sensor was detected continuously. Fig. 2 shows the change of the resonant frequency during the Mg G37 culture process in the absence of antibiotics. The initial concentration of Mg G37 strain was 2.0 × 102 ccu mL−1 . The resonant frequency decreased slightly in the first 48 h, then decreased rapidly in the 48–168 h, and finally hardly decreased after 168 h. The decrease in the resonant frequency was due to the adhesion of Mg G37 on the sensor surface. The total change value of resonance frequency during the process was about 850 Hz. The f–t curves indicated that the growth curve of Mg G37 in the absence of antibiotics was composed of three phases: lag phase in the 0–48 h, logarithmic phase in the 48–168 h and stationary phase after 168 h. The arrival of stationary phase meant that after 168 h the nutriment was almost exhausted and the amount of Mg G37 was hardly changed. To confirm the hypothesis of three phases, after being cultured for 24, 72, 120 and 168 h, the Mg G37 adhered to the sensor surface were imaged with a high-power microscope (Nikon ECLIPSE TE2000-S ×1000). The microscopy images are shown in Fig. 3. It can be observed that Mg G37 adhered and proliferated on the sensor surface. The increasing amount of Mg G37 with the incubation

Fig. 3. Microscopy images of Mg adhered on polyurethane film-coated magnetoelastic sensor during the culture course in the SP-4 culture medium.

B. He et al. / Biosensors and Bioelectronics 24 (2009) 1990–1994

1993

time is in accord with the results determined by the magnetoelastic sensor device. A CCU method was used to confirm the results further. Mg can decompose glucose to produce acid, causing the change of the color of the SP-4 culture medium from red to orange when phenol red is present. In the operating process of CCU method, a Mg sample obtained from a certain growth phase was diluted step by step using SP-4 culture medium containing phenol red as diluent to form serial 10-fold dilutions. The serial 10-fold dilutions were incubated together in the CO2 incubator (37 ◦ C, 5% CO2 ) for 120 h. If the color of 1–n dilutions in the serial 10-fold dilutions changed from red to orange, the initial density of Mg in the sample would be 10n ccu mL−1 . One ccu corresponds to 10–100 cells (Bernet et al., 1989). The results given by CCU method are also consistent with those obtained by the sensor. The results show that in the range of 1.9 × 103 to 2.7 × 104 ccu mL−1 the linear relationship of f and CCU is f (Hz) = −0.0284 CCU(ccu mL−1 ) – 7.95 with the correlation coefficient of R = 0.998, and the detection limit of the sensor is 3.2 × 102 ccu mL−1 , which indicates further that the proposed sensor can be applied for the real-time, long-term and in situ monitoring of Mg growth. 3.3. Evaluation of anti-microbial activity of antibiotics Two antibiotics, tetracycline and levofloxacin, were chosen as the model antibiotics to test the ability of the sensor system for evaluation of antibacterial activity. These two antibiotics have different inhibition mechanisms. Tetracycline was selected as the representative of the antibiotics those inhibit the synthesis of protein (Fraterrigo and Perlman, 1971). Levofloxacin was selected as the representative of the antibiotics those inhibit the copy of DNA (Ernst et al., 1997). Fig. 4(a) shows the typical f–t curves of the sensor incubated in the presence of tetracycline, and Fig. 4(b) shows that in the presence of levofloxacin. Compared with the control experiment showed in Fig. 2, the addition of the antibiotics obviously inhibited the proliferation of Mg G37. In the presence of the antibiotics, the lag phase in the growth inhibition curves is prolonged obviously, the extent of Mg G37 proliferation is narrow in the logarithmic phase, and the stationary phase is substituted by a decline phase. The inhibition effect of the antibiotics on the growth of Mg can be described quantitatively with the inhibition efficiency (IE) of the antibiotics. The inhibition efficiency of an antibiotic can be defined as IE (%) =

f0 − f × 100 f0

where f0 and f are the shifts in the resonance frequency of the magnetoelastic sensor without and with addition of the antibiotic, respectively. For a certain antibiotic, IE is related with the concen-

Fig. 4. The shift of the resonant frequency during the Mg G37 culture process in the presence of (a) tetracycline with the concentrations of (1) 1.0 ␮g/mL, (2) 2.0 ␮g/mL and (3)3.0 ␮g/mL, and (b) levofloxacin with the concentrations of (1) 0.5 ␮g/mL, (2) 1.0 ␮g/mL and (3) 1.5 ␮g/mL.

tration of the antibiotic. The experiments showed that 1.0 ␮g/mL of tetracycline inhibited the proliferation of Mg by 34.4% after 72 h, while 2.0 ␮g/mL of tetracycline could kill 56.5% of Mg during the same period. The inhibition efficiency of levofloxacin was different from that of tetracycline. The experiments showed that 0.5 ␮g/mL of levofloxacin inhibited the proliferation of Mg by 39.3% after 72 h, while 1.0 ␮g/mL of levofloxacin could kill 57.1% of Mg during the same period. Moreover, MIC50 (minimal inhibitory concentration, concentration of a drug that reaches 50% inhibition efficiency) for Mg incubated with the antibiotics for 120 h was calculated to be 1.5 ␮g/mL for tetracycline and 0.5 ␮g/mL for levofloxacin. The difference in the inhibition efficiency of these two antibiotics is due to the difference of their inhibition mechanism. The profiles of the growth inhibition curves in the presence of tetracycline and levofloxacin also revealed the difference in the inhibition mechanism of these two antibiotics. Tetracycline worked on Mg G37 rapidly, while levofloxacin worked on Mg G37 slowly. However, the long-

Table 1 Inhibition efficiency of antibiotics on Mg G37 (%)a . Conc. (␮M)

Time after drug addition (h) 48

72

96

120

144

168

192

216

240

Tetracycline 1.0 2.0 3.0

34.1 ± 1.8 55.0 ± 2.2 58.1 ± 2.7

34.4 ± 2.6 56.5 ± 2.3 78.6 ± 2.4

34.9 ± 1.4 57.7 ± 4.1 82.5 ± 4.4

35.3 ± 4.3 57.9 ± 4.8 83.6 ± 5.8

35.4 ± 3.1 56.8 ± 4.1 82.4 ± 4.5

35.6 ± 3.6 56.0 ± 3.3 79.7 ± 6.2

34.7 ± 2.7 56.6 ± 5.6 78.8 ± 5.2

34.5 ± 2.0 56.8 ± 2.5 81.5 ± 3.8

33.9 ± 1.8 57.8 ± 3.0 85.2 ± 4.6

Levofloxacin 0.5 1.0 1.5

14.3 ± 1.0 21.4 ± 1.1 35.7 ± 1.5

39.3 ± 1.0 46.4 ± 1.5 57.1 ± 2.7

45.0 ± 2.5 55.0 ± 2.8 60.1 ± 2.6

50.7 ± 2.0 60.0 ± 2.1 67.3 ± 3.6

58.1 ± 3.2 63.5 ± 3.1 70.3 ± 2.0

51.9 ± 2.8 58.3 ± 2.7 69.6 ± 2.7

42.6 ± 3.4 51.2 ± 2.3 67.5 ± 3.3

44.4 ± 1.1 53.1 ± 1.9 71.6 ± 4.0

50.0 ± 1.9 58.2 ± 3.0 76.6 ± 2.2

Where f0 and f are the shifts in the resonance frequency of the magnetoelastic sensor without and with addition of the antibiotic, respectively. a The inhibition efficiency is defined as IE (%) = ((f0 − f)/f0 ) × 100 (data presented are mean ± S.D., n = 3).

1994

B. He et al. / Biosensors and Bioelectronics 24 (2009) 1990–1994

term inhibition efficiency of levofloxacin was better than that of tetracycline (Table 1). 4. Conclusions In this work, a wireless magnetoelastic sensing device has been developed for the real-time, long-term and in situ monitoring of Mg G37 growth, and it has been successfully used for the evaluation of the inhibition effect of the antibiotics tetracycline and levofloxacin on Mg G37 growth. In the device, no direct physical connections between the sensor and the detection system are required for signal telemetry, which greatly facilitates aseptic biological operation. The proposed magnetoelastic sensing device costs effectively and can be readily used for the monitoring of the growth of other Mycoplasma and evaluation of the effects of antibiotics on the Mycoplasma based on the f–t curves. Acknowledgements The authors thank the National Natural Science Foundation of China (NSFC Nos. 20877038, 20775024, 30770115) and the Science Foundation of Department of Public Health of Hunan Province, China (No. B2006-103) for financial support. The authors also thank Prof. Qingyun Cai (Hunan University, China) for the help with the wireless magnetoelastic-sensing device developed by Craig A. Grimes et al. (The Pennsylvania State University, University Park, USA). References Assunc¸ão, P., Diaz, R., Comas, J., Ruiz de Galarreta, C.M., González-Llamazares, O.R., Poveda, J.B., 2005. J. Appl. Microbiol. 98, 1047–1054. Bernet, C., Garret, M., de Barbeyrac, B., BéBéar, C., Bonnet, J., 1989. J. Clin. Microbiol. 27, 2492–2496. Berrettoni, M., Carpani, I., Corradini, N., Conti, P., Fumarola, G., Legnani, G., Lanteri, S., Marassi, R., Tonelli, D., 2004. Anal. Chim. Acta 509, 95–101.

Cai, Q.Y., Grimes, C.A., 2000. Sens. Actuators B 71, 112–117. Cai, Q.Y., Zeng, K.F., Ruan, C.M., Desai, T.A., Grimes, C.A., 2004. Anal. Chem. 76 (14), 4038–4043. Chang, K.-S., Chang, C.-K., Chien-Yuan Chen, C.-Y., 2007. Sens. Actuators B 125, 207–213. Clausen, H.F., Fedder, J., Drasbek, M., Nielsen, P.K., Toft, B., Ingerslev, H.J., Birkelund, S., Christiansen, G., 2001. Hum. Reprod. 16, 1866–1874. Cohen, C.R., Manhart, L.E., Bukusi, E.A., Astete, S., Brunham, R.C., Holmes, K.K., Sinei, S.K., Bwayo, J.J., Totten, P.A., 2002. Lancet 359, 765–766. Deguchi, T., Maeda, S.-I., 2002. J. Urol. 167 (3), 1210–1217. Ernst, M.E., Ernst, E.J., Klepser, M.E., 1997. Am. J. Health-Syst. Pharm. 54 (22), 2569–2584. Fraser, C.M., Gocayne, J.D., White, O., et al., 1995. Science 270, 397–404. Fraterrigo, C.C., Perlman, D., 1971. J. Antibiot. 24 (3), 185–188. Grimes, C.A., Mungle, C.S., Zeng, Z.F., Jain, M.K., Dreschel, W.R., Paulose, M., Ong, K.G., 2002. Sensors 2, 294–313. Guntupalli, R., Hu, J., Lakshmanan, R.S., Huang, T.S., Barbaree, J.M., Chin, B.A., 2007. Biosens. Bioelectron. 22 (7), 1474–1479. Huang, S., Pang, P., Xiao, X., He, L., Cai, Q., Grimes, C.A., 2008. Sens. Actuators B 131, 489–495. Nivens, D.E., Chambers, J.Q., Anderson, T.R., White, D.C., 1993. Anal. Chem. 65, 05–09. Pang, P.F., Huang, S.J., Cai, Q.Y., Yao, S.Z., Zeng, K.F., Grimes, C.A., 2007. Biosens. Bioelectron. 18 (5–6), 675–681. Puckett, L.G., Barrett, G., Kouzoudis, D., Grimes, C.A., Bachas, L.G., 2003. Biosens. Bioelectron. 23, 295–299. Putman, M., Burton, R., Nahm, M.H., 2005. J. Immunol. Methods 302, 99–102. Razin, S., Yogev, D., Naot, Y., 1998. Microbiol. Rev. 63, 1094–1156. Reipa, V., Almeida, J., Cole, K.D., 2006. J. Microbiol. Methods 66, 449–459. Ruan, C., Zeng, K., Varghese, O.K., Grimes, C.A., 2003. Anal. Chem. 75 (23), 6494–6498. Ruan, C., Zeng, K., Varghese, O.K., Grimes, C.A., 2004. Biosens. Bioelectron. 19 (12), 1695–1701. Stemke, G.W., Robertson, J.A., 1982. J. Clin. Microbiol. 16, 959–961. Stoyanov, P.G., Grimes, C.A., 2000. Sens. Actuators A 80, 8–14. Tully, J.G., Rose, D.L., Whitcomb, R.F., Wenzel, R.P., 1979. J. Infect. Dis. 139, 478–482. Uno, M., Deguchi, T., Komeda, H., Hayasaki, M., Iida, M., Nagatani, M., Kawada, Y., 1997. Sex Transm. Dis. 24, 284–286. Wu, S.H., Cai, Q.Y., Grimes, C.A., 2006. Sens. Lett. 4 (2), 160–164. Xiao, X., Guo, M., Li, Q., Cai, Q., Yao, S., Grimes, C.A., 2008. Biosens. Bioelectron. 24, 247–252. Yoshida, T., Deguchi, T., Ito, M., Maeda, S., Tamaki, M., Ishiko, H., 2002. J. Clin. Microbiol. 40, 1451–1455. Zeng, K.F., Ong, K.G., Mungle, C., Grimes, C.A., 2002. Rev. Sci. Instrum. 73 (12), 4375–4380. Zhou, T., Marx, K.A., Warren, M., Schulze, H., Braunhut, S.J., 2000. Biotechnol. Prog. 16, 268–277.