Liquid-Crystal-Based Immunosensor for Diagnosis of Tuberculosis

However, when clinical serum samples from healthy people or latent TB ..... the background signal, and random samples were chosen as representatives...
0 downloads 0 Views 3MB Size
Download PDF
Research Article www.acsami.org

Liquid-Crystal-Based Immunosensor for Diagnosis of Tuberculosis in Clinical Specimens Hyeong Jin Kim, Jinseob Rim, and Chang-Hyun Jang* Department of Chemistry, Gachon University, Seongnam-Daero 1342, Bokjeong-Dong, Sujeong-Gu, Seongnam-Si, Gyeonggi-Do 13120, Korea S Supporting Information *

ABSTRACT: Tuberculosis (TB) is a serious global health problem, although it is easily preventable and curable. The prevalence of TB is caused by a lack of simple, rapid, cheap, and effective TB diagnostic tests. In this study, we have established a novel liquid crystal (LC)-based immunosensor to diagnose TB in clinical specimens. The clinical serum samples were incubated on a TB antigenimmobilized substrate, and their optical LC responses were observed using a polarized light microscope. Specific binding of anti-TB antibodies to the TB antigen-immobilized surface only occurred in clinical specimens from TB patients, inducing the disruption of the orientation of LCs. This was followed by the distinctive change in the optical appearance of LCs from uniform to random. However, when clinical serum samples from healthy people or latent TB patients were incubated, the orientation of LCs remained uniform. Through the change of optical LC images, in 76% of TB patients, this essay correctly identified the patients as having antibodies to TB in their serums. 91% of healthy people free of TB were correctly identified as not having antibodies to TB. Thus, this LC-based immunosensor is a promising platform, particularly in clinical TB diagnostics, which does not require complicated preparation of clinical specimens or complex instrumentation. KEYWORDS: liquid crystal, 4-cyano-4′-pentylbiphenyl (5CB), nanostructured surface, tuberculosis (TB), clinical serum specimen

1. INTRODUCTION

be tested completely and requires sophisticated techniques or complex instrumentation.4−6 In recent years, many methods have been researched to detect TB effectively, such as polymerase chain reaction (PCR) assays,7−9 interferon-γ release assays (IGRA),10−12 and enzymelinked immunosorbent assays (ELISAs).13−15 Although nucleic acid amplification with PCR has high specificity, it has low sensitivity and requires 1 to 2 days for diagnosis of TB. Both IGRA and ELISA are commonly performed for serological diagnostic tests; however, IGRA cannot distinguish active TB and latent TB, and ELISA requires an additional labeling process and has low sensitivity. Over the past decade, liquid crystals (LCs) have attracted great interest in the domain of chemical and biological sensing systems.16−18 Owing to LC characteristic properties like optical anisotropy, long-range order, fluidity, and simple arrangement on a substrate, the LC-based sensing systems may be simple, rapid, cost-efficient, optical, portable, and promising techniques in the field of detecting biomolecular interactions taking place at biological membranes. Compared with other conventional analytical methods, LC-based sensing techniques have several advantages. They do not require labeled molecules,19 laborious techniques,20 or a well-equipped laboratory.21 Many types of

Tuberculosis (TB) caused by bacteria (Mycobacterium tuberculosis) remains one of the major health problems in the world. It is estimated that more than 8 million new TB cases occur each year, and approximately 1.5 million patients die from the disease annually. Over 95% of TB cases and deaths occur in developing countries, which have poor hygiene, few health care resources, and large numbers of people infected with the human immunodeficiency virus (HIV).1,2 Until the TB has advanced, the symptoms (fever, cough, loss of weight, poor appetite, night sweats, etc.) might be attributed to other diseases. This leads to serious delays in diagnosing and treating TB, causing disease transmission to others. Even though most TB can be successfully cured with timely diagnosis and treatment, the number of casualties is extremely high.3 Therefore, the development of simple and rapid diagnostic methods for TB is required. A complete medical diagnosis for TB includes the following: a medical history, physical examination, test for TB infection (the Mantoux tuberculin skin test or the TB blood test), chest radiograph, and diagnostic microbiology.1,4 The skin test is simple and commonly used as a diagnostic tool, but it is not always accurate when performed on certain populations.5,6 For diagnostic microbiology, many countries have relied on the widely applied sputum smear microscopy method to diagnose TB. However, sputum smear microscopy takes 4 to 8 weeks to © 2017 American Chemical Society

Received: May 3, 2017 Accepted: June 6, 2017 Published: June 15, 2017 21209

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces

Korea). All clinical specimens from TB patients, latent TB patients, and healthy people were provided by the Tuberculosis Specimen Bank in Masan National Tuberculosis Hospital (South Korea) and the Korean Institute of Tuberculosis (South Korea). These clinical samples were in the form of serum separated from the blood cells by centrifuging, and these serum samples were used in this study without any additional preparation process. All aqueous solutions were prepared with high purity deionized (DI) water (18 MΩ/cm) generated using a Milli-Q water purification system (Millipore, Bedford, MA, USA). In this research, the clinical tests using humanderived materials were approved by the Institutional Review Board (IRB) at Gachon University. 2.2. Fabrication of the Nanostructured Substrates. Glass microscope slides were treated with piranha solution (70% H2SO4, 30% H2O2) following the procedure reported previously.23−26 After cleaning, thin-film layers of titanium (with thickness of 80 Å) and gold (with thickness of 200 Å) were obliquely deposited (45°) on the slides, using an electron beam evaporator.27−29 2.3. Immobilization of TB Antigen. The obliquely deposited gold substrates were immersed in 1 mM ethanolic solution (7:3 mixture of MUA and 1-decanethiol) for 1 h to functionalize the surfaces with a mixed self-assembled monolayer (SAM). To activate surface carboxyl groups on the mixed SAM, NHS/EDC (50 mM/200 mM) solutions were used.23−26 After activation of surface carboxylic groups on the mixed SAM, the substrates were incubated with an aqueous solution of TB antigen (ESAT-6) in PBS at room temperature for 0.5 h. 2.4. Detection of Anti-TB Antibodies in Clinical Serum Specimens. Before the incubation of clinical serum samples on the nanostructured surface, the TB antigen-immobilized surface was treated with 20 mM ethanolamine to block nonspecific binding of proteins to the surface. Then, clinical serum specimens from TB patients and healthy people were incubated on the TB antigenimmobilized surface in humid conditions at 37 °C. After 3 h of incubation, the substrates were sequentially rinsed with Tween-20 and DI water and dried under gaseous N2 streams. 2.5. Fabrication of LC Optical Cells. LC optical cells were fabricated by pairing the obliquely deposited gold surface with an OTS-treated glass with a film of spacers (thickness of 12 μm). OTStreated glass was prepared by a procedure reported previously.23−26 Then, the 5CB molecules were heated by a heat gun (∼40 °C, isotropic phase of 5CB) and applied to the space between these paired surfaces with a glass syringe. Next, the LC optical cells were cooled slowly to room temperature (nematic phase of 5CB). 2.6. Imaging of LC Optical Cells, Using a Polarized Light Microscope. The LC optical cells were imaged under a polarized light microscope (ECLIPSE LV100POL, Nikon, Tokyo, Japan). All LC images were acquired by a digital camera (DS-2Mv, Nikon, Tokyo, Japan) affixed to the polarized light microscope with a 10× objective lens under the two crossed polarizers at a resolution of 1600 × 1200 pixels, a gain of 1.00×, and a shutter speed of 1/30 s. The horizontal scale of LC images was 1.25 mm. For quantitative analysis of LC optical images, the luminosity was measured by Adobe Photoshop CS6 (San Jose, CA, USA). The LC texture was converted to gray scale, and the average brightness was determined. 2.7. Atomic Force Microscopy (AFM). AFM (NanoScope IIIa, Veeco Metrology, Santa Barbara, CA, USA) was performed to measure the surface topography for two purposes:29,30 (1) characterization of the obliquely deposited gold surface and (2) analysis of root-meansquare (RMS) roughness change at each stage of the reaction. For the characterization, the average values of the peak-to-trough and peak-topeak distances were calculated to determine the “amplitude” and “wavelength” of the topography, respectively. For more precise roughness analysis, we chose flat gold surfaces on silicon wafers instead of obliquely deposited gold surfaces to minimize background roughness of the surface. Every image was acquired with a resolution of 256 × 256 points at a scan rate of 1.0 line/s in tapping mode. 2.8. Ellipsometry. An ellipsometer (Elli-SE(UV)-FM8, Ellipsotechnology, Korea) was used to measure the optical thickness change at each stage of the reaction. The light sources were halogen and

sensing applications employing an LC-based sensor have been developed, including for detection of enzyme reactions,22 biotinylated proteins,23 DNA strands,24 viruses,25 and antigen− antibody reactions.26,27 In this study, we report a new immunosensor to diagnose TB sensitively in clinical serum specimens with a rapid optical response to the orientational change of LCs occurring on the nanostructured surface (Figure 1). Inspired by the optical

Figure 1. Schematic illustration of TB diagnosis in clinical specimens, using a LC-based immunosensor. With clinical serum samples from TB patients, binding of anti-TB antibodies onto a TB antigendecorated surface causes the orientational transition of LCs, resulting in a random texture of optical LC image. However, with clinical serum samples from healthy people, the orientation of LCs is not disrupted, resulting in a uniform texture of optical LC image.

anisotropic property of LCs and their high sensitivity for detecting pure anti-TB antibodies,18,26 we developed an immunosensor for clinical TB diagnosis, using LCs that exhibit specific orientational transitions with clinical serum samples from TB patients by TB antigen−antibody reactions on the TB antigen-immobilized substrate. This leads to random optical texture of LCs. However, clinical serum samples from healthy people cannot induce orientational transitions of LCs, resulting in uniform optical texture of LCs. Furthermore, through the observance of optical LC images under a polarized light microscope, this LC-based immunosensor can be employed for simple TB diagnosis in clinical serum specimens, with a sensitivity of 76% and specificity of 91%.

2. MATERIAL AND METHODS 2.1. Materials. Nematic liquid crystal, 4-cyano-4′-pentylbiphenyl (5CB), was purchased from Tokyo Chemical Industry Co., Ltd. (C1555, Japan). Glass microscope slides were acquired from Matunami (S-1215, Japan). Polished silicon (1 0 0) wafers were obtained from Silicon Sense (Nashua, NH, USA). Gold (99.999%) and titanium (99.999%) were acquired from Unlimited Enhanced Technology (South Korea). Octyltrichlorosilane (OTS) was purchased from Alfa Aesar (Ward Hill, MA, USA). N-Hydroxysuccinimide (NHS), N-(3-(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride (EDC), 1-decanethiol, 11-mercaptoundecanoic acid (MUA), phosphate-buffered saline (PBS), Tween-20, and ethanolamine were purchased from Sigma-Aldrich (St. Louis, MO, USA). TB antigen, early secretory antigenic target with (ESAT)-6, secreted as a potent T cell antigen, was obtained from Genes Laboratories Inc. (South 21210

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces deuterium lamps with output wavelengths of 380−1100 nm. Samples for ellipsometric analysis were prepared in the same manner as those for AFM. A three-phase model (ambient/organic layer/substrate) was involved to calculate the optical thickness of the organic layer with the Sellmeier equation. The resulting data were gathered from 5 points on the spot area.

nanostructured pattern size of the obliquely deposited gold surface, the concentration of TB antigens was controlled (5 μg/ mL in 10 mM PBS) and incubated not to disrupt anisotropic nanostructured patterns. After immobilization of TB antigens, the uniform texture of the optical LC image was observed (Figure 3(A)) and suggested that the ability of the surface anisotropy to trigger a uniform orientation of LCs was sufficiently retained despite the presence of TB antigens. The antigen−antibody reactions of TB were examined by monitoring the optical responses of LCs, using the TB antigenimmobilized surface after incubation with clinical serum samples. We speculated that the orientation of LCs would be changed by the binding of anti-TB antibodies to the TB antigen-decorated surfaces, masking the anisotropic surface topography. These specific antigen−antibody reactions of TB solely occurred during introduction of clinical serum samples from TB patients in which anti-TB antibodies existed. To verify our hypothesis, human serum samples from TB patients and healthy people were incubated on TB antigen-immobilized surfaces. Before the introduction of clinical specimens, the TB antigen-decorated surface was immersed in ethanolamine solution to deactivate the remaining carboxyl groups on the surface to block the nonspecific binding of serum proteins to the surface. Then, human serum samples from TB patients and healthy people were incubated on the TB antigen-decorated surfaces for 3 h, upon consideration of the diffusion rate of antiTB antibodies and regioselectivity between TB antigens and anti-TB antibodies. Figure 3(B) and (C), in sequence, shows the optical LC images after incubation of human serum samples from TB patients and healthy people on the TB antigen-decorated surface. As shown in Figure 3(B), the alternation from a uniform to random texture occurred by the introduction of TB patient serum specimens, while a uniform texture remained after introduction of normal serum specimens (Figure 3(C)). Since the optical transition of LC texture to a random texture originated from the random rearrangement of LCs, we confirmed that the specific binding of anti-TB antibodies (from the clinical serum sample of a TB patient) to the TB antigen-decorated surface masked the anisotropic surface topography and induced the random orientation of LCs. From these results, TB antigen−antibody reactions could be easily monitored on the nanostructured substrate by observing the optical LC images, resulting in a clinical diagnosis of TB using human serum samples. Additionally, luminosity was investigated for quantitative analysis of the uniform and random textures of the LC image. LC texture was converted to gray scale by Adobe Photoshop CS6, and then the average brightness was calculated. Figure 4 indicates the luminosity of the optical LC images of obliquely deposited gold surfaces after immobilization of TB antigens, followed by the incubation of clinical serum samples from TB patients and healthy people. The blue and orange bars represent the optical image captured with an angle of 0° and 45°, respectively, under polarized light microscopy. After reaction with clinical serum specimens from TB patients, the luminosity of two optical LC images (captured with an angle of 0° and 45°) had a similar value, indicating randomness. However, after immobilization of TB antigens followed by the incubation of clinical serum specimens from healthy people, there was a significant difference in the luminosity between two optical LC images (captured with an angle of 0° and 45°), representing the uniformity of the LC textures.

3. RESULTS AND DISCUSSION 3.1. Characterization of the Nanostructured Substrate. In other studies, nanostructured substrates produced by various methods were demonstrated not only to introduce the uniform orientation of LCs on these anisotropic surfaces but also to play an essential role in the detection of biomolecular interactions (e.g., antigen−antibody reaction and receptor−ligand interaction).18,23,27 For sensitive response of LCs to an interaction with target molecules, the anisotropic surface topography was designed based on two parameters: (i) amplitude size, which is smaller than target molecules, and (ii) optimum surface coverage of target molecules, which can disrupt the surface anisotropy. Here, considering these two conditions, we manufactured the obliquely deposited gold surfaces and then characterized their surface topography by tapping mode AFM.29 The cross-sectional AFM analysis confirmed the surface topographical features of the gold-coated substrate, which have anisotropic patterns with an amplitude of 1−4 nm and a wavelength of 20−50 nm (Figure S1). Corresponding to our aim for diagnosis of TB by monitoring the TB antigen−antibody reaction, an anisotropic surface topography of obliquely deposited gold surfaces could be disrupted by TB antigen−antibody complexes (estimated size: ∼9.4 nm), resulting in an orientational transition of LCs from uniform to random.26,27 To identify the orientational behavior of LCs on the obliquely deposited gold substrates, the mixed self-assembled monolayers (SAMs) were formed with 1 mM ethanolic solutions (7:3 mixture of MUA and 1-decanethiol). After the fabrication of LC optical cells by pairing with OTS-treated glass, the optical LC response was observed under a polarized light microscope. The uniform textures of the optical LC image indicated their uniform arrangement of LCs on the gold-coated surface (Figure 2). Thus, this result suggests that LC molecules

Figure 2. Optical LC images acquired by polarized light microscopy: obliquely deposited gold surface treated with carboxylate-terminated mixed SAM. The oval and two arrows on the left side of the images indicate the optical cell, polarizer, and analyzer, respectively. The orientation of the optical cell was defined as the angle between the polarizer and analyzer (0°, 45°).

are arranged in a line parallel to the anisotropic nanopatterns created by obliquely gold deposition, resulting in a uniform orientation of LCs.28 3.2. Monitoring Antigen−Antibody Reactions of TB on the Nanostructured Substrate. We immobilized the TB antigens (ESAT-6, molecular weight: 6 kDa, estimated size: ∼2.4 nm) on the mixed SAMs, where surface carboxyl groups were activated by NHS-EDC chemistry. Considering the 21211

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces

Figure 3. Optical LC images acquired by polarized light microscopy: obliquely deposited gold surface after (A) immobilization of TB antigens and incubation of clinical serum samples from (B) TB patients and (C) healthy people. The oval and two arrows on the left side of the images indicate the optical cell, polarizer, and analyzer, respectively. The orientation of the optical cell was defined as the angle between the polarizer and analyzer (0°, 45°).

Figure 4. Luminosity of the optical LC images with different angles (0°, 45°). After immobilization of TB antigens, the obliquely deposited gold surfaces were incubated with clinical serum samples from TB patients and healthy people.

3.3. Sensitivity and Specificity of LC-Based Sensor for Clinical Diagnosis of TB Using Human Serum Samples. The sensitivity of the LC-based sensor for clinical diagnosis of TB was evaluated using 50 clinical specimens from TB patients, depending on the optical LC images observed after incubation of those specimens. In other words, the random and uniform textures of LC images were, respectively, designated a positive and negative result of TB. In the first clinical trial, 39 clinical specimens resulting in random textures of LC image were diagnosed as TB-positive cases, and 11 clinical specimens resulting in uniform textures of LC image were determined as TB-negative cases. From these results, the diagnostic sensitivity of TB was determined as 78% (39/50) in the first clinical test. For more accurate investigation of sensitivity, we conducted a second clinical trial using 50 clinical serum samples from TB patients identical to those used in the first test. In the second clinical trial, 37 clinical samples were determined as TB-positive cases, and 13 clinical samples were determined as TB-negative cases, resulting in a sensitivity of 74% (37/50). From the results of two clinical trials, the average sensitivity was calculated as 76% (76/100) (Table 1). This indicated that the chance of receiving a TB-positive result using clinical serum samples from TB patients is 76%. However, the sensitivity of this LC-based sensor could be increased by several repetitive tests for diagnosis of TB. For example, when clinical trials for TB

Table 1. Sensitivity of TB Detection in Clinical Specimens, Using LC-Based Sensor LC-based sensor TB patients

positive

negative

total

sensitivity (%)

1st test 2nd test total

39 37 76

11 13 24

50 50 100

78% 74% 76%

were performed two times, the chance of misdiagnosis of TB (probability of receiving TB-negative results in both tests) was only theoretically 5.8% (0.24 × 0.24 × 100). Through experiments we conducted, it was confirmed that when the clinical assays were repeated twice, the chance of a TB patient’s serum sample being determined as a negative would be highly decreased. In practice, only three clinical serum samples from TB patients were determined as TB-negative in both clinical trials, and 47 clinical serum samples from TB patients were diagnosed as TB-positive more than once in two clinical trials, showing a sensitivity of 94% (47/50). Next, we examined the specificity of the LC-based sensor for clinical diagnosis of TB using the 53 clinical serum specimens from healthy people, depending on the optical LC images observed after incubation of clinical serum samples from healthy people using the same method as that in clinical 21212

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces

roughness (1.953 ± 1.219 nm), and the AFM image also changed notably (Figure 5(D)). This increment (about 0.80 nm), which was larger than 0.41 nm, is reasonable because the size of antibody molecules (∼7.0 nm) is bigger than the size of TB antigens (ESAT-6: ∼2.4 nm). However, the normal serum samples did not make any meaningful difference in both the roughness value and the image compared with those of the TB antigen-immobilized surface (1.174 ± 0.381 nm, Figure 5(E)), which meant an insignificant number of biomolecules settled down on the surface because the normal serum did not contain anti-TB antibodies. We also performed a 2-sample t test to determine whether the population means of the TB and normal cases were different in a 95% confidence interval (CI) and if the P-value was less than 0.001, which would indicate that they were meaningfully different from each other. These results supported that the orientational transition of LCs was triggered when anti-TB antibodies in the serum bound to the TB antigen-immobilized surface and changed the surface anisotropy. For ellipsometric analysis, the optical thickness of the organic layer was measured based on the flat gold surface (Figure 6).

sensitivity tests. Only 5 clinical samples resulted in random LC textures, and they were determined as TB-positive cases; meanwhile, 48 clinical samples resulted in uniform LC textures, and these samples were determined as TB-negative cases, indicating a specificity of 91% (48/53). These results suggested that the chance of receiving a TB-negative result using clinical serum specimens from healthy people is 91% (Table 2). Table 2. Specificity of TB Detection in Clinical Specimens, Using LC-Based Sensor LC-based sensor healthy people

positive

negative

total

specificity (%)

test

5

48

53

91%

The sensitivity and specificity of this LC-based immunosensor for diagnosis of TB was analogous to that of a polymerase chain reaction (PCR) assay, which had 91.4% sensitivity and 75.9% specificity for the diagnosis of tuberculosis meningitis (TBM).9 However, this label-free LC-based sensor for clinical diagnosis of TB is a simple, rapid, and promising technique that does not require complex sample preparation and complicated instrumentation. 3.4. Surface Topographical Analysis by AFM and Ellipsometry. We investigated the surface topography of the substrate by AFM and ellipsometry to confirm that the optical response came from morphologic change after the incubation of clinical serum specimens.30 In both AFM and ellipsometry analyses, a flat gold surface on a silicon wafer was used as a substrate to minimize the background signal, and random samples were chosen as representatives. Figure 5 shows AFM images of the height data in a 5 nm data scale. The root-mean-square (RMS) roughness of a bare gold surface was 0.689 ± 0.051 nm, and its AFM image appeared smooth (Figure 5(A)). Following the formation of a mixed SAM on the surface, the roughness slightly increased to 0.743 ± 0.105 nm, and the AFM image did not change considerably (Figure 5(B)). After the immobilization of TB antigens, there was an evident increase in the roughness (1.155 ± 0.377 nm, Figure 5(C)). The roughness increased about 0.41 nm compared with that of the mixed SAM surface. Meanwhile, the incubation of the TB patient serum specimens, which may contain a high level of anti-TB antibodies, further increased the

Figure 6. Ellipsometric data of the optical thickness on the flat gold surface after formation of mixed SAM, immobilization of TB antigens, and incubation of TB patient and normal serum on the TB antigen layer (*P < 0.001, 2-sample t test in 95% CI).

Figure 5. AFM images of the height data on the flat gold surface after (A) no treatment, (B) formation of mixed SAM, (C) immobilization of TB antigens, and incubation of (D) TB patient serum and (E) normal serum on the TB antigen layer (*P < 0.001, 2-sample t test in 95% CI). 21213

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces The thickness difference between the mixed SAM (0.531 ± 0.165 nm) and antigen-immobilized layer (0.948 ± 0.213 nm) was approximately 0.41 nm. This increment was not the same as the estimated size of TB antigens (ESAT-6) because the optical thickness was measured by an optical method and thus was different from the actual thickness. When the TB patient serum was incubated on the TB antigen-immobilized surface, the thickness jumped to 3.611 ± 0.269 nm, which might be derived from TB antigen−antibody interaction. On the other hand, when the normal serum was incubated, the thickness increased to 2.158 ± 0.413 nm, which was statistically different from the thickness after the incubation of TB patient serum (P < 0.001, 2-sample t test in 95% CI). These ellipsometric results also supported that the optical response was induced by the interaction of anti-TB antibodies in the clinical serum samples from TB patients. 3.5. Studying the Detection of Latent TB Using Human Serum Samples. After confirming that TB could be diagnosed by this LC-based sensor using the clinical specimens from TB patients and healthy people, our interest shifted to latent TB patients. Using the same method as that in clinical tests of TB patients and healthy people, we conducted the clinical test for detection of anti-TB antibodies using the latent TB patient serums by observing the optical LC response after incubation of the serum samples. As shown in Figure S2, the uniform textures of the LC image originated from the uniform orientation of LCs appeared after incubation with most of the latent TB patient serum specimens. Among 16 clinical serum samples from latent TB patients, only 4 clinical samples revealed random textures of the LC image, and they were determined as TB-positive. With this LC-based sensor for diagnosis of TB, anti-TB antibodies from the latent TB patient serum could not be detected since there was not enough antiTB antibodies in the latent TB patient serum to induce an orientational transition of LCs. This result of the latent TB patient cases was also underpinned by AFM and ellipsometry analysis of surface topography. After the incubation of the latent TB serum samples, the RMS roughness value and optical thickness were gauged to be 1.811 ± 0.159 nm and 2.197 ± 0.623 nm, respectively (Figures S3 and S4). These measured values of RMS roughness and optical thickness were much less than those of the TB patient serum samples and similar to those of normal serum samples. Although anti-TB antibodies in the latent TB patient serum could not be detected, further research would be required because only 15 clinical serum specimens from latent TB patients were used in this study. Additionally, we expect that the detection of anti-TB antibodies from the latent TB patient serum using an LC-based sensing technique would be feasible by increasing the sensitivity of antiTB antibodies through changing several experimental conditions, such as surface topography, serum incubation, or preparation of clinical specimens.

the orientation of LCs was retained, resulting in uniform texture of the LC image. The sensitivity and specificity of this LC-based sensor for TB diagnosis were 76% and 91%, respectively. On the basis of these results, this rapid, simple, and label-free LC-based immunosensor is a promising technique, particularly in clinical diagnosis of TB, that does not require intricate preparation of clinical specimens and complicated instrumentation. Moreover, this LC-based immunosensing technique can be utilized in the establishment of correct and sensitive clinical diagnostics for various diseases.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b06189. Details about the characterization of nanostructured substrate and the experimental results from latent TB patients (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +82-31-750-8555. Fax: +82-31-750-8774. E-mail: [email protected] (C.-H. Jang). ORCID

Chang-Hyun Jang: 0000-0003-3165-4747 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2016R1D1A1B03935782).

■

REFERENCES

(1) Society, A. T. Diagnostic Standards and Classification of Tuberculosis in Adults and Children. Am. J. Respir. Crit. Care Med. 2000, 161, 1376−1395. (2) WHO, Global Tuberculosis Report 2016; 2016. (3) Albert, H.; Heydenrych, A.; Brookes, R.; Mole, R.; Harley, B.; Subotsky, E.; Henry, R.; Azevedo, V. Performance of a Rapid Phagebased Test, FASTPlaqueTB, to Diagnose Pulmonary Tuberculosis from Sputum Specimens in South Africa. Int. J. Tuberc. Lung Dis. 2002, 6 (6), 529−537. (4) Society, A. T. Infectious Diseases Society of America: Treatment of Tuberculosis. Am. J. Respir. Crit. Care Med. 2003, 167, 603−662. (5) Bates, J. H. Diagnosis of Tuberculosis. Chest 1979, 76, 757−763. (6) Andersen, P.; Munk, M.; Pollock, J.; Doherty, T. Specific Immune-based Diagnosis of Tuberculosis. Lancet 2000, 356 (9235), 1099−1104. (7) Eisenach, K. D.; Sifford, M. D.; Cave, M. D.; Bates, J. H.; Crawford, J. T. Detection of Mycobacterium Tuberculosis in Sputum Samples Using a Polymerase Chain Reaction. Am. Rev. Respir. Dis. 1991, 144, 1160−1163. (8) Kim, J. W.; Choi, Y. J.; Kim, H. J.; Park, J. S.; Nam, H. S.; Hwangbo, Y.; Kim, D. U.; Park, K. S. Comparison of PNA Probebased Real-time PCR and Cobas TaqMan MTB for Detection of MTBC. BioChip J. 2013, 7 (2), 85−88. (9) Deshpande, P. S.; Kashyap, R. S.; Ramteke, S. S.; Nagdev, K. J.; Purohit, H. J.; Taori, G. M.; Daginawala, H. F. Evaluation of the IS6110 PCR Assay for the Rapid Diagnosis of Tuberculous Meningitis. Cerebrospinal Fluid Res. 2007, 4 (10), 10. (10) Brock, I.; Munk, M.; Kok-Jensen, A.; Andersen, P. Performance of Whole Blood IFN-γ Test for Tuberculosis Diagnosis Based on PPD

4. CONCLUSION In conclusion, we have developed an LC-based immunosensor for TB diagnosis, using clinical serum specimens. The specific binding of anti-TB antibodies from TB patient clinical samples to the TB antigen-immobilized surface disrupted the anisotropic surface topography, inducing the orientational transition of LCs. This resulted in an alteration of optical appearance of LCs from uniform to random. However, in the case of clinical samples from healthy people or latent TB patients, the TB antigen−antibody reaction did not occur, and 21214

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215

Research Article

ACS Applied Materials & Interfaces or the Specific Antigens ESAT-6 and CFP-10. Int. J. Tuberc. Lung Dis. 2001, 5 (5), 462−467. (11) Brock, I.; Weldingh, K.; Lillebaek, T.; Follmann, F.; Andersen, P. Comparison of Tuberculin Skin Test and New Specific Blood Test in Tuberculosis Contacts. Am. J. Respir. Crit. Care Med. 2004, 170 (1), 65−69. (12) Meier, T.; Eulenbruch, H.-P.; Wrighton-Smith, P.; Enders, G.; Regnath, T. Sensitivity of a New Commercial Enzyme-linked Immunospot Assay (T SPOT-TB) for Diagnosis of Tuberculosis in Clinical Practice. Eur. J. Clin. Microbiol. Infect. Dis. 2005, 24 (8), 529− 536. (13) Baelden, M.-C.; Vanderelst, B.; Dieng, M.; Prignot, J.; Cocito, C. Serological Analysis of Human Tuberculosis by an ELISA with Mycobacterial Antigen 60. Scand. J. Infect. Dis. 1990, 22 (1), 63−73. (14) Al Zahrani, K.; Al Jahdali, H.; Poirier, L.; Rene, P.; Gennaro, M.; Menzies, D. Accuracy and Utility of Commercially Available Amplification and Serologic Tests for the Diagnosis of Minimal Pulmonary Tuberculosis. Am. J. Respir. Crit. Care Med. 2000, 162 (4), 1323−1329. (15) Wilkins, E.; Bothamley, G.; Jackett, P. A rapid, Simple Enzyme Immunoassay for Detection of Antibody to Individual Epitopes in the Serodiagnosis of Tuberculosis. Eur. J. Clin. Microbiol. Infect. Dis. 1991, 10 (7), 559−563. (16) Brake, J. M.; Daschner, M. K.; Luk, Y.-Y.; Abbott, N. L. Biomolecular Interactions at Phospholipid-decorated Surfaces of Liquid Crystals. Science 2003, 302 (5653), 2094−2097. (17) Brake, J. M.; Abbott, N. L. Coupling of the Orientations of Thermotropic Liquid Crystals to Protein Binding Events at Lipiddecorated Interfaces. Langmuir 2007, 23 (16), 8497−8507. (18) Woltman, S. J.; Jay, G. D.; Crawford, G. P. Liquid-crystal Materials Find a New Order in Biomedical Applications. Nat. Mater. 2007, 6 (12), 929−938. (19) Mo, Z.-H.; Yang, X.-C.; Guo, K.-P.; Wen, Z.-Y. A Nanogoldquenched Fluorescence Duplex Probe for Homogeneous DNA Detection Based on Strand Displacement. Anal. Bioanal. Chem. 2007, 389 (2), 493−497. (20) Redl, G.; Husain, F. T.; Bretbacher, I. E.; Nemes, A.; CichnaMarkl, M. Development and Validation of a Sandwich ELISA for the Determination of Potentially Allergenic Sesame (Sesamum Indicum) in Food. Anal. Bioanal. Chem. 2010, 398 (4), 1735−1745. (21) Pandey, A.; Mann, M. Proteomics to Study Genes and Genomes. Nature 2000, 405 (6788), 837−846. (22) Hu, Q.-Z.; Jang, C.-H. Imaging Trypsin Activity through Changes in the Orientation of Liquid Crystals Coupled to the Interactions between a Polyelectrolyte and a Phospholipid Layer. ACS Appl. Mater. Interfaces 2012, 4 (3), 1791−1795. (23) Han, G.-R.; Jang, C.-H. Measuring Ligand−receptor Binding Events on Polymeric Surfaces with Periodic Wave Patterns Using Liquid Crystals. Colloids Surf., B 2012, 94, 89−94. (24) Park, S.-J.; Jang, C.-H. Using Liquid Crystals to Detect DNA Hybridization on Polymeric Surfaces with Continuous Wavy Features. Nanotechnology 2010, 21 (42), 425502. (25) Han, G.-R.; Song, Y.-J.; Jang, C.-H. Label-free Detection of Viruses on a Polymeric Surface Using Liquid Crystals. Colloids Surf., B 2014, 116, 147−152. (26) Kim, H. J.; Rim, J.; Jang, C.-H. Diagnosis of Tuberculosis Using a Liquid Crystal-based Optical Sensor. Macromol. Res. 2016, 24 (2), 123−130. (27) Gupta, V. K.; Skaife, J. J.; Dubrovsky, T. B.; Abbott, N. L. Optical Amplification of Ligand-receptor Binding Using Liquid Crystals. Science 1998, 279 (5359), 2077−2080. (28) Gupta, V. K.; Abbott, N. L. Uniform Anchoring of Nematic Liquid Crystals on Self-assembled Monolayers Formed from Alkanethiols on Obliquely Deposited Films of Gold. Langmuir 1996, 12 (10), 2587−2593. (29) Skaife, J. J.; Abbott, N. L. Quantitative Characterization of Obliquely Deposited Substrates of Gold by Atomic Force Microscopy: Influence of Substrate Topography on Anchoring of Liquid Crystals. Chem. Mater. 1999, 11 (3), 612−623.

(30) Skaife, J. J.; Brake, J. M.; Abbott, N. L. Influence of Nanometerscale Topography of Surfaces on the Orientational Response of Liquid Crystals to Proteins Specifically Bound to Surface-immobilized Receptors. Langmuir 2001, 17 (18), 5448−5457.

21215

DOI: 10.1021/acsami.7b06189 ACS Appl. Mater. Interfaces 2017, 9, 21209−21215