Active Surface Hydrophobicity Switching and ... - ACS Publications

Subscriber access provided by University of Glasgow Library

Communication

Active Surface Hydrophobicity Switching and Dynamic Interfacial Trapping of Microbial Cells by Metal Nanoparticles for Preconcentration and In-plane Optical Detection Yuyeon Kim, Kwangyeong Jung, Jeehan Chang, Taejin Kwak, Youngwook Lim, Seonghak Kim, Jeong Geol Na, Jinwon Lee, Inhee Choi, Luke P. Lee, Dongchoul Kim, and Taewook Kang Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.9b03163 • Publication Date (Web): 03 Sep 2019 Downloaded from pubs.acs.org on September 3, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Active Surface Hydrophobicity Switching and Dynamic Interfacial Trapping of Microbial Cells by Metal Nanoparticles for Preconcentration and Inplane Optical Detection Yuyeon Kim1,†, Kwangyeong Jung1,†, Jeehan Chang1,†, Taejin Kwak3, Youngwook Lim3, Seonghak Kim1, Jeonggeol Na1, Jinwon Lee1, Inhee Choi4, Luke P. Lee5,6, Dongchoul Kim3,*, and Taewook Kang1,2,* 1Department

of Chemical and Biomolecular Engineering, Sogang University, Seoul, 04107,

Korea 2Institute

of Integrated Biotechnology, Sogang University, Seoul, 04107, Korea

3Department

of Mechanical Engineering, Sogang University, Seoul, 04107, Korea

4Department

of Life Science, University of Seoul, Seoul, 02504, Korea

5Berkeley

Sensor and Actuator Center, University of California Berkeley, CA 94720, USA

6Department

of Bioengineering, Electrical Engineering and Computer Science, University of

California Berkeley, CA 94720, USA

† These authors contributed equally to this work

ACS Paragon Plus Environment

1

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

*[email protected] *[email protected]


2

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

ABSTRACT

The surface hydrophobicity of a microbial cell is known to be one of the important factors in its adhesion to an interface. To date, such property has been altered by either genetic modification or external pH, temperature, and nutrient control. Here we report a new strategy to engineer a microbial cell surface and discover the unique dynamic trapping of hydrophilic cells at an air/water interface via hydrophobicity switching. We demonstrate the surface transformation and hydrophobicity switching of Escherichia coli (E. coli) by metal nanoparticles. By employing real-time dark-field imaging, we directly observe that hydrophobic gold nanoparticle-coated E. coli, unlike its naked counterpart, is irreversibly trapped at the air/water interface due to elevated hydrophobicity. We show that our surface transformation method and resulting dynamic interfacial trapping can be generally extended to gram-positive and gram-negative bacteria, and fungi. As the dynamic interfacial trapping allows the preconcentration of microbial cells, high intensity of scattering light, in-plane focusing, and near-field enhancement, we are able to directly quantify E. coli as low as 1.0 × 103 cells/ml by using a smartphone with an image analyzer. We also establish the identification of different microbial cells by the characteristic Raman transitions directly measured from the interfacially-trapped cells.

KEYWORDS Surface

hydrophobicity,

Microbial

cells,

Metal

nanoparticles,

Interfacial

trapping,

Preconcentration, In-plane optical detection


3

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

INTRODUCTION The hydrophobicity of a microbial cell plays an important role in its adhesion to various interfaces such as water/solid and air/water interfaces1,2. For example, microbial cells comprising a hydrophobic cell surface preferentially adhere to the interface between water and a hydrophobic solid, and microbial colonies are then developed within the extracellular matrix (i.e., biofilm)3. Such hydrophobicity has been explored in a wide variety of applications. Decreasing the hydrophobicity of the bacterial pathogen Pseudomonas aeruginosa has been regarded to be a key factor in the treatment of lethal infections that often occur in cystic fibrosis patients as it could prevent the formation of a biofilm and render bacteria susceptible to antimicrobial agents4. The surface hydrophobicity of materials used in implanted medical devices such as intravenous devices, urinary catheters, or artificial joints has been considered as a means to prevent infections resulting from the adhesion and growth of hydrophobic fungal pathogens (e.g., Candida albicans) in the form of a biofilm5,6. In addition, the hydrophobicity of bacteria is known to be critical in the bacterial degradation of pollutants such as heavy metal ions and hydrocarbons in soil as the efficiency of the adsorption of bacteria on soil depends on the hydrophobicity7,8. In order to control the surface hydrophobicity of microbial cells, specific genes whose functions are associated with the biosynthesis of cell walls are regulated, and their molecular composition (e.g., lipopeptides, fatty acids, lipopolysaccharides (LPS), or membrane proteins) is determined9-11. For example, expressing the lapF gene in Pseudomonas putida leads to the increase in the hydrophobicity of the cell walls via the production of hydrophobic proteins in the outer membrane9. Another method used to alter the cell surface hydrophobicity is to induce a change in the membrane composition in response to external stimuli such as temperature, pH, or nutrients12-15. For example, the Antarctic bacteria Pseudomonas syringae exhibits a less


4

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

hydrophobic surface at a low temperature (e.g., 4 °C) due to an increased ratio of hydroxylated fatty acids to LPS in the outer membrane12. In addition, some bacteria are known to produce capsular polysaccharides, which can mask the hydrophobic protein layer in the outer cell membrane under either low pH conditions or growth media with high carbon and nitrogen concentrations, thereby decreasing the hydrophobicity of their surfaces13,14. Recently, the combination of metal nanoparticles and microbial cells has been of interest as these particles can render their excellent electrical and optical properties to the cells, while the cells can serve as a biocompatible reactor for the synthesis of metal nanoparticles16-18. For example, various microbial cells such as bacteria, fungi, and actinomycetes can be examined for their function as bioreactors for the synthesis of metal (e.g., gold, silver, or bimetallic gold/silver) nanoparticles without the need for toxic reducing chemicals16. On the other hand, gold nanoparticles also can dramatically enhance the current output of a bacterial fuel cell by facilitating extracellular electron transfer17. Similarly, gold nanoparticles are known to enable the semi-artificial photosynthesis of non-photosynthetic bacteria via electron generation18. Herein, we propose an active switching of surface properties of various microbial cells via coating with metal nanoparticles. In addition, we achieve unique dynamic trapping of inplane assembly for hydrophilic microbial cells at an air/water interface through hydrophobicity switching, which can provide an effective solution for straightforward optical detection of the microbial cells. Figure 1 shows our strategy to engineer the surface property of a microbial cell using metal nanoparticles and resulting dynamic interfacial trapping phenomenon. First, colloidal metal nanoparticles are designed to be densely and uniformly attached to a microbial cell surface


5

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

by adjusting both the surface charge and relative number of the metal nanoparticles (Figure 1a). Then, depending on the hydrophobicity (or hydrophilicity) of the metal nanoparticles, the microbial cell surface can be transformed to either a hydrophilic or hydrophobic surface. If the hydrophilic microbial cell surface is transformed to be more hydrophobic by coating with hydrophobic nanoparticles, the cell is expected to be dynamically trapped at an air/water interface due to elevated hydrophobicity and the decrease in the interfacial energy (Figure 1b)1920.

This interfacial trapping phenomenon could be accelerated by mechanical agitation as an

additional flow force can lead to an increased number of cells approaching the interface at a higher rate. Finally, this dynamic interfacial trapping with optical metal nanoparticles could be expected to simultaneously achieve (1) the facile preconcentration of trace cells, (2) the in-plane focusing of incident light at the interface, (3) high intensity of scattering light from the cells, and (4) strong near-field enhancement between the metal nanoparticles. To realize these benefits, as proof-of-concept experiments, first, a smartphone is used to directly and sensitively quantify trace cells via the measurement of an area of interfacially trapped cells with metal nanoparticles by using a commercial image analyzer. This is performed as coating with the metal nanoparticles leads to microbial cells with a considerably high scattering intensity than bare microbial cells. Second, the selective identification of several microbial cells also can be achieved by employing surface-enhanced Raman spectroscopy (SERS). For the surface transformation of a microbial cell, Escherichia coli (E. coli) was selected due to its biological and clinical importance. Positively charged hydroxylamine hydrochloride (HAHC)-capped gold nanospheres (HAHC-GNS) were prepared by replacing the citrate ligands of negatively charged citrate-capped molecules with HAHC. Then, 0.5 ml of E. coli was mixed with 4.5 ml of HAHC-GNS with a diameter of 30 nm (Figure 2a) for 2 h. Scanning electron


6

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

microscopy (SEM) images in Figure 2a (top, rightmost) and Figure 2b directly reveal that the HAHC-GNS are uniformly and densely attached onto the surface of the E. coli. To further examine this attachment, UV–vis spectra of HAHC-GNS, E. coli, and a mixture of HAHC-GNS with E. coli are recorded (Figure 2c). HAHC-GNS exhibit a surface plasmon resonance (SPR) band at 523 nm. By mixing HAHC-GNS with the E. coli solution, another SPR peak in a nearinfrared (NIR) region is observed, while the SPR band of HAHC-GNS is red-shifted from 523 nm (Δλ = 15 nm). These results also support the binding between HAHC-GNS and E. coli, and the decrease in the interparticle distance between the neighboring particles due to this binding would account for the observed additional band in the NIR and the red-shift21. To examine the effects of the surface charge of the metal nanoparticles on their attachment onto the E. coli surface, negatively charged citrate-capped GNS (citrate-GNS) with the same diameter (30 nm) were also examined (Supporting Information Figure 1). Photograph and SEM images recorded after mixing E. coli with citrate-GNS reveal no significant binding of the citrate-GNS. Notably, few citrate-GNS can be attached onto the surface of these bacteria during water evaporation for SEM sample preparation. The UV–vis spectra of the citrate-GNS were recorded before and after mixing with E. coli. Unlike the previous result obtained for HAHC-GNS, neither a shift in the SPR band nor an additional SPR peak in the NIR region is observed. The zeta potentials of E. coli, HAHC-GNS, and citrate-GNS are estimated to be -80.1 ± 0.1, 28.0 ± 5.6, and 42.4 ± 15.7 mV, respectively. These results indicate that the electrostatic attraction between HAHC-GNS and E. coli can be mainly related to the observed attachment. Next, the viability of GNS-E. coli formation was investigated under growth conditions similar to those of bare E. coli by measuring the optical density at 600 nm (OD600), as well as by using a luminescent adenosine triphosphate (ATP) assay. The OD600 of GNS-E. coli increases


7

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

over time and reaches a stationary phase after 4 h, with a similar tendency observed for the OD600 of E. coli (Figure 2d). In addition, ATP concentrations of E. coli and GNS-E. coli increase with time and decrease as they enter a stationary phase. These results reveal that there is no significant difference in the cell viability of E. coli and GNS-E. coli and that E. coli maintains its viability even after surface coating with GNS. To examine the change in the surface hydrophobicity of microbial cells obtained by coating with metal nanoparticles, contact angle (θc) measurements were carried out (Figure 2e). First, a closely packed monolayer of E. coli is formed by the simple evaporation of an E. coli solution onto a Si wafer (Supporting Information Figure 2). Then, a closely packed monolayer of HAHC-GNS or gold–silica core–shell nanoparticles (Au@SiO2) obtained from ethanol-induced self-assembly at an air/water interface22,23 is directly deposited on the E. coli monolayer. Au@SiO2 was selected as it is more hydrophilic than the gold surface due to its silica shell24. The θc of each sample was measured by the analysis of an image of a 5 µl droplet of water on a sample substrate. With the formation of an E. coli monolayer onto the Si wafer, θc dramatically decreases from 86.7° to 22.6° due to the higher hydrophilicity of E. coli compared to the Si wafer. After the deposition of the nanoparticle monolayers, the θc of HAHC-GNS on the E. coli monolayer increases to 69.3°, while that of Au@SiO2 on the E. coli monolayer is similar to that of the E. coli monolayer (θc = 28.7°). Coating with hydrophobic nanoparticles renders the surface of E. coli less hydrophilic (i.e., more hydrophobic), while E. coli remains hydrophilic after coating with hydrophilic nanoparticles. Our surface transformation method was further exploited to induce the dynamic interfacial trapping of microbial cells at an air/water interface. To achieve trapping, 30 nm HAHC-GNS was mixed with E. coli, and this mixture was mildly shaken using a rotary shaker.


8

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Notably, dark-colored aggregates are observed in the GNS-E. coli solution at the interface, while the bare E. coli solution does not exhibit aggregate formation at the interface and remains clear (Figure 2f). The SEM image for the aggregates (Figure 2g and Supporting Information Figure 3) indicates that GNS-E. coli is considerably trapped at the interface. Aggregates are not observed without mechanical shaking due to the slow diffusion rate of GNS-E. coli25. To obtain direct evidence for this dynamic interfacial trapping phenomenon, real-time dark-field imaging of the air/water interface was performed by using a specially designed fluidic chamber (Figure 2h). In the dark-field images of the bare E. coli solution, several greenish light spots corresponding to E. coli with a weak scattering intensity are observed (Supporting Information Figure 4)26. Supporting Information Movie 1 and the time-resolved snapshots of Figure 2i reveal the random movement of the greenish spot near the interface by diffusion. In addition to the lateral movement of the spot in XY plane, the spot is out-focused at 2 and 12 s and then re-focused at 4 and 14 s, respectively. On the other hand, for the GNS-E. coli solution, bright yellowish spots corresponding to GNS-E. coli with a higher scattering intensity (owing to the attached gold nanoparticles) than bare E. coli are observed27. Similar to the analysis of E. coli, the yellowish spot exhibits random and lateral movement (Supporting Information Movies 2 and 3 and Figure 2j). However, neither out-focusing nor re-focusing of the spot is observed at the interface. This difference would mainly result from the non-trapping and stable trapping mode of E. coli and GNS-E. coli, respectively. The reversible adsorption and desorption of E. coli at the air/water interface (i.e., a focal plane) would be responsible for the observed and repetitive outand re-focusing of its scattering spot (Figure 2k). On the other hand, the elevated surface hydrophobicity would increase the stability of GNS-E. coli at the interface, thereby limiting its escape from the focal plane (Figure 2l). Note that GNS is found not to be trapped at the interface


9

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

due to its small size (Supporting Information Movie 4)19,20. Hence, the interfacially trapped GNS-E. coli assemble into large aggregates by the reduced negative charge of E. coli resulting from the surface-attached GNS and are further stabilized at the interface. To investigate whether our method and the observed interfacial trapping are generally applicable to various microbial cells and metal nanoparticles, gold nanoparticles with different sizes and shapes were first examined. HAHC-GNS of sizes greater than 50 nm and hexadecyltrimethylammonium bromide (CTAB)-capped gold nanorods (GNRs) were selected as both are positively charged (Supporting Information Figure 5). Photographs in Figure 3a and 3b clearly indicate that interfacial trapping is commonly observed for both cases. SEM images reveal that GNR and HAHC-GNS of sizes of 50 nm are uniformly attached onto E. coli, and E. coli with the nanoparticles are commonly trapped at the interface. Next, various microbial cells such as gram-negative bacteria (Pseudomonas putida), gram-positive bacteria (Corynebacterium glutamicum), and fungi (Saccharomyces cerevisiae) were also examined (Figure 3c–e). Photographs and SEM images were recorded after shaking P. putida (Figure 3c), C. glutamicum (Figure 3d), and S. cerevisiae (Figure 3e) with 30 nm HAHC-GNS. All of the images reveal that cells are trapped at the air/water interface. SEM images (Figure 3c–e, rightmost) reveal that HAHC-GNS are attached onto the cell surface. These microbial cells exhibit a negative surface charge due to lipid A for gram-negative bacteria, teichoic acid for gram-positive bacteria, and phosphate groups for fungi28. Notably, the transparent color of the GNS–S. cerevisiae solution after shaking is related to the low residual GNS concentration of the solution, resulting from the ten-fold higher surface area of S. cerevisiae compared to E. coli. To further elucidate the effect of mechanical shaking on the observed dynamic interfacial trapping of GNS-microbial cells, a three-dimensional fluid dynamics simulation was performed.


10

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

The diameter and height of cylindrical container were set to be 26 and 10 mm, respectively, and a microbial cell was assumed to be a sphere with a diameter of 1 μm. The flow force (Ff) generated by the bulk flow, the stochastic force (Fs) from the random cell diffusion, and the friction force (Ffric) were considered to determine the motion and location of the cell (Figure 4a). The cell was assumed to be irreversibly trapped at the air/water interface. The behavior of the cells according to their normalized velocity (vnorm), which is defined as the fluid velocity divided by the diffusion rate of the cell, was investigated. Figure 4b shows the time-sequential images of the 5.0 × 107 cells with (vnorm = 120) and without (vnorm = 0) shaking. In the absence of shaking, most of the cells are randomly dispersed in water after 3 h, and trapped cells are not observed at the interface. In contrast, with the application of shaking, most of the cells are shown to be trapped at the interface after 3 h. To further examine the mechanism of this dramatic difference, Figure 4c shows the trajectories of a single cell over 3 h. Due to the slow cell diffusion, without shaking, the cell is strongly limited to random movement in a relatively small, confined region that is still far from the interface. On the other hand, mechanical shaking dramatically increases the velocity of the cell, thereby moving in the entire container and providing a greater opportunity for the cells to reach the interface. To examine the effect of the shaking speed on the interfacial trapping, the number of interfacially trapped cells at different vnorm with respect to time was calculated (Figure 4d). At vnorm = 0, the number of trapped cells does not significantly change over 3 h. With the application of shaking (vnorm = 40, 80, and 120), the number of trapped cells rapidly increases in the first 1 h and then gradually increases. Notably, with the increase in vnorm from 0 to 120, the number of the trapped cells dramatically increases. For example, 99% of the cells are trapped after 3 h at vnorm = 120, while only 1% of the cells are trapped at vnorm = 0. For the comparison of this simulation


11

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

result with the experiment, GNS-E. coli solutions containing 5.0 × 107 cells were subjected to shaking at different speeds. In good agreement with the simulation, interfacially trapped GNS-E. coli are observed after 3 h only in the presence of the shaking, while the interface of the solution remains clear in the absence of shaking (Figure 4e). In addition, with the increase in the shaking speed from 50 to 250 rpm, the area of interfacially trapped GNS-E. coli dramatically increases by 15 times, indicating that increased amount of GNS-E. coli is trapped at higher shaking speeds. Our method and the resulting dynamic interfacial trapping can be applied to the sensitive and selective detection of microbial cells from an interface. First, the concentration of trace microbial cells was estimated via the measurement of an area of the interfacially trapped cells with a high scattering intensity by using a smartphone (Figure 5a). With the increase in the microbial cell concentration, the number of trapped GNS-microbial cells concomitantly increases, leading to the increased area of the trapped GNS-microbial cells. An image of the trapped cells was then recorded by using a smartphone, and the area (i.e., scattering intensity) was quantified by using a commercial image analyzer. If a wide dynamic range of scattering intensity is obtained, the smartphone-based quantification of microbial cells could be utilized in several areas of the public health industry. As a proof-of-concept, the quantification of E. coli was examined. Photographs of GNSE. coli with an E. coli content ranging from 1.0 × 103 to 5.0 × 105 cells/ml were recorded by using a smartphone (Supporting Information Figure 6), and the area of trapped GNS-E. coli was measured by using an image analyzer (Figure 5b and Supporting Information Figure 7). The measured results reveal that the area of trapped GNS-E. coli exponentially increases for E. coli concentrations ranging from 1.0 × 103 to 5.0 × 105 cells/ml. The limit of detection for E. coli is 1.0 × 103 cells/ml. Similar to E. coli, the limit of detection for S. cerevisiae is also found to be


12

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

1.0 × 103 cells/ml (Supporting Information Figure 8). This result implies that our method can provide straightforward and timesaving optical detection of pathogens (a diagnostic threshold is generally set to be higher than 1.0 × 103 CFU/ml)29. Second, microbial cells were identified by SERS. Four microbial cells, E. coli, P. putida, C. glutamicum, and S. cerevisiae, respectively, were selected as they are environmentally and industrially relevant. Raman spectra of the interfacially trapped GNS-microbial cells under a laser excitation of 785 nm were recorded (Supporting Information Figure 9 and Figure 5c). All bare microbial cells without GNS do not exhibit distinctive Raman transitions, while GNSmicrobial cells exhibit specific characteristic Raman transitions. Gram-negative bacteria (E. coli and P. putida) exhibit three Raman transitions at 957 and 1170 cm-1, corresponding to C–N stretching and glycosidic ring breathing, respectively30,31. These transitions correspond to lipid A that is the main component of a gram-negative bacterial cell wall. In addition, specific Raman transitions for each cell are observed at 731 cm-1 for E. coli, 824 and 1233 cm-1 for P. putida, 1154 cm-1 for C. glutamicum, and 1316 cm-1 for S. cerevisiae30-34. The different molecular compositions of outer cell walls are mainly responsible for the differences between Raman spectra. Supporting Table 1 summarizes more detailed assignments of the Raman transitions. In addition, various types of water (i.e., bottled and tap water) were investigated for potential applications toward water safety inspection. After 1 h of shaking, the interfacially trapped cells are observed in bottled and tap water (Figure 5d), while the mixture of GNS and these water samples do not reveal aggregates at the interfaces (Supporting Information Figure 10). In addition, characteristic transitions of E. coli (731, 957, and 1170 cm-1) are observed in the trapped cells (Figure 5e).


13

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

To examine the origin of these high signal enhancements, an electromagnetic (EM) simulation for a single GNS-microbial cell was conducted. Figure 5f shows the simulation schemes for a GNS-microbial cell on a Si wafer (left) and its EM field distributions (right). Two types of geometry according to the surface curvature (denoted as plane and curved surfaces, respectively) were considered. The maximum local electric field enhancement on a plane surface is estimated to be 2.7 times greater than that on curved surface as the electric dipole of each GNS on a curved surface is not parallel to the substrate (Figure 5g)35. This result indicates that GNS, which are assembled in parallel to the XY plane, are mainly responsible for the enhancement of the SERS signal. In conclusion, a method to transform the cell surface hydrophobicity by completely coating the surface with a dense layer of metal nanoparticles was described. The dynamic trapping of these cells at the air/water interface induced by the increase in the hydrophobicity of these cells was achieved to preconcentrate them for the in-plane optical detection. By employing one-step colloidal mixing, gold nanoparticles were spontaneously assembled into uniform, dense layers on the cell surface, leading to the change in the intrinsic hydrophobicity of the cell to the particle hydrophobicity. In addition, real-time dark-field imaging revealed that the increased hydrophobicity can induce the irreversible trapping of the cells at the air/water interface, while the bare cells escape from the interface by random diffusion. In addition, our method can be applied regardless of the size or shape of the particle, as well as the type of microbial cells including gram-positive bacteria, gram-negative bacteria, and fungi. Numerical simulations revealed that our interfacial trapping and preconcentration method can be further facilitated by mechanical shaking due to the increased cell velocity. Finally, our method was extended to the simultaneous quantification and identification of microbial cells using optical devices. From the


14

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

images of the interfacially trapped cells by using a smartphone and the subsequent analysis of the area of the cells via a commercial imaging program, trace amounts of the cells as low as 1.0 × 103 cells/ml can be quantified. Further identification of various microbial cells including grampositive bacteria, gram-negative bacteria, and fungi was achieved by surface-enhanced Raman spectroscopy. The microbial cells can be distinguished by each Raman transition from their intrinsic molecular compositions of their outer membranes via the strong electromagnetic field enhancement of surface-assembled gold nanoparticles. We strongly anticipate that our method may be used to enable the on-demand engineering of microbial cells for future applications ranging from microbial diagnostics to environmental monitoring, which requires the simple, portable, and standoff detection of trace amounts of pathogen.


15

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

REFERENCES (1)

Berne, C.; Ellison, C. K.; Ducret, A.; Brun, Y. V. Nat. Rev. Microbiol. 2018, 16, 616-627.

(2)

Van Loosdrecht, M. C.; Lyklema, J.; Norde, W.; Schraa, G.; Zehnder, A. J. Appl. Environ. Microbiol. 1987, 53, 1893-1897.

(3)

Koo, H.; Allan, R. N.; Howlin, R. P.; Stoodley, P.; Hall-Stoodley, L. Nat. Rev. Microbiol. 2017, 15, 740–755.

(4)

Drenkard, E.; Ausubel, F. M. Nature 2002, 416, 740.

(5)

Kojic, E. M.; Darouiche, R. O. Clin. Microbiol. Rev. 2004, 17, 255-267.

(6)

Finkel, J. S.; Mitchell, A. P. Nat. Rev. Microbiol. 2011, 9, 109.

(7)

Zhong, H.; Liu, G.; Jiang, Y.; Yang, J.; Liu, Y.; Yang, X.; Liu, Z.; Zeng, G. Biotechnol. Adv. 2017, 35, 490-504.

(8)

Ren, X.; Zeng, G.; Tang, L.; Wang, J.; Wan, J.; Liu, Y.; Yu, J.; Yi, H.; Ye, S.; Deng, R. Sci. Total Environ. 2018, 610–611, 1154–1163.

(9)

Lahesaare, A.; Ainelo, H.; Teppo, A.; Kivisaar, M.; Heipieper, H. J.; Teras, R. PloS one 2016, 11, e0166078.

(10)

El-Kirat-Chatel, S.; Beaussart, A.; Derclaye, S.; Alsteens, D.; Kucharikova, S.; Van Dijck, P.; Dufrêne, Y. F. ACS Nano 2015, 9, 1648–55.

(11)

Zara, S.; Bakalinsky, A. T.; Zara, G.; Pirino, G.; Demontis, M. A.; Budroni, M. Appl. Environ. Microbiol. 2005, 71, 2934-2939.


16

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(12)

Kumar, G. S.; Jagannadham, M. V.; Ray, M. K. J. Bacteriol. 2002, 184, 6746–6749.

(13)

Masson, L. U. K. E.; Holbein, B. E. Infect. Immun. 1985, 47, 465-471.

(14)

Bonet, R.; Simon-Pujol, M. D.; Congregado, F. Appl. Environ. Microbiol. 1993, 59, 2437–2441.

(15)

Koza, A.; Hallett, P. D.; Moon, C. D.; Spiers, A. J. Microbiol. 2009, 155, 1397-1406.

(16)

Singh, P.; Kim, Y. J.; Zhang, D.; Yang, D. C. Trends Biotechnol. 2016, 34, 588-599.

(17)

Fan, Y.; Xu, S.; Schaller, R.; Jiao, J.; Chaplen, F.; Liu, H. Biosens. Bioelectron. 2011, 26, 1908-1912.

(18)

Zhang, H.; Liu, H.; Tian, Z.; Lu, D.; Yu, Y.; Cestellos-Blanco, S.; Sakimoto, K. K.; Yang, P. Nat. Nanotechnol. 2018, 13, 900–905.

(19)

Pieranski, P. Phys. Rev. Lett. 1980, 45, 569.

(20)

Böker, A.; He, J.; Emrick, T.; Russell, T. P. Soft matter 2007, 3, 1231-1248.

(21)

Whang, K.; Lee, J. H.; Shin, Y.; Lee, W.; Kim, Y. W.; Kim, D.; Lee, L. P.; Kang, T. Light Sci. Appl. 2018, 7, 68.

(22)

Shin, Y.; Song, J.; Kim, D.; Kang, T. Adv. Mater. 2015, 27, 4344–4350.

(23)

Chang, J.; Lee, J.; Georgescu, A.; Huh, D.; Kang, T. Nano Lett. 2019, 19, 2074–2083.

(24)

Wang, F.; Peters, S.; Guzda, J.; Blunk, R. H.; Angelopoulos, A. P. Langmuir 2009, 25, 4384–4392.

(25)

Kumar, M. S.; Philominathan, P. Biophys. Rev. 2010, 2, 13-20.


17

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(26)

Page 18 of 30

Kinoshita, T.; Ishiki, K.; Nguyen, D. Q.; Shiigi, H.; Nagaoka, T. Anal. Chem. 2018, 90, 4098–4103.

(27)

Liu, M.; Chao, J.; Deng, S.; Wang, K.; Li, K.; Fan, C. Colloids Surf. B Biointerfaces 2014, 124, 111–117.

(28)

Malanovic, N.; Lohner, K. Biochim. Biophys. Acta - Biomembr. 2016, 1858, 936–946.

(29)

Foxman, B. Nat. Rev. Urol. 2010, 7, 653.

(30)

Liu, Y.; Zhou, H.; Hu, Z.; Yu, G.; Yang, D.; Zhao, J. Biosens. Bioelectron. 2017, 94, 131140.

(31)

Walter, A.; März, A.; Schumacher, W.; Rösch, P.; Popp, J. Lab. Chip. 2011, 11, 10131021.

(32)

Chisanga, M.; Muhamadali, H.; Ellis, D. I.; Goodacre, R. Appl. Spectrosc. 2018, 72, 9871000.

(33)

Paret, M. L.; Sharma, S. K.; Green, L. M.; Alvarez, A. M. Appl. Spectrosc. 2010, 64, 433441.

(34)

Sujith, A.; Itoh, T.; Abe, H.; Anas, A. A.; Yoshida, K.; Biju, V.; Ishikawa, M. Appl. Phys. Lett. 2008, 92, 103901.

(35)

Seo, S.; Chang, T. W.; Liu, G. L. Sci. Rep. 2018, 8, 3002.


18

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge on the ACS Publications website. Experimental section; A comparison between citrate-GNS and HAHC-GNS; A SEM image of uniform E. coli monolayer; A representative SEM image of preconcentrated microbial cells at an air/water interface; A representative dark-field scattering microscope image of E. coli; UV-vis spectra and Zeta potentials of HAHC-GNS and CTAB-GNR; Photographs of trapped GNS-E. coli with respect to cell concentration; Image analysis through open-source imaging program; Photographs and area measurement of trapped GNS-S. cerevisiae with respect to cell concentration; Raman measurement of bare microbial cells; Photographs of bare GNS mixed with various types of water; Dark-field scattering movie of bare E. coli at the interface; Dark-field scattering movie of GNS-E. coli at the interface; Dark-field scattering movie of colloidal HAHC-GNS at the interface; Raman transition assignments of microbial cells.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] *E-mail: [email protected] Author Contributions Y.K., K.J., and J.C. contributed equally to this work. T.K. and D.K. conceived the concept. Y.K., K.J., J.C., and T.K. designed and organized the experiments. Y.K., K.J., and J.C. performed the


19

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

experiments. T.J.K., Y.L, and D.K. contributed to simulations. S.K., J.N., and J.L. contributed to the cell growth and viability test. All authors discussed the results and wrote the manuscript. Funding Sources This research was supported by the C1 Gas Refinery Programs (No. 2018M3D3A1A01055759 and

No.

2015M3D3A1A01064929)

and

Basic

Science

Research

Program

(No.

2016R1A6A1A03012845) through the National Research Foundation of Korea funded by the Ministry of Science, ICT, and Future Planning. Notes The authors declare no competing financial interest.


20

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 1. Schematic illustrations of generalized surface transformation of microbial cells by metal nanoparticles and its application to dynamic interfacial trapping at an air/water interface. (a) Surface hydrophobicity switching of microbial cells by using metal nanoparticles via electrostatic interaction. (b) Dynamic interfacial trapping of microbial cells at an air/water interface owing to coating the cells with hydrophobic surface gold nanoparticles (GNP).


21

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30


22

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2. Detailed characterizations of the surface transformation, hydrophobicity switching and dynamic interfacial trapping of microbial cells. (a) Surface transformation of E. coli by hydroxylamine hydrochloride (HAHC)-capped gold nanosphere (HAHC-GNS) via electrostatic interaction. From left to right, cartoon representation of HAHC-GNS (the left-most), a representative transmission electron microscope (TEM) image of HAHC-GNS (top) and a photograph (bottom) of HAHC-GNS solution, a representative scanning electron microscope (SEM) image of bare E. coli and a photograph of E. coli solution, and a representative SEM image of HAHC-GNS attached E. coli (GNS-E. coli) and a photograph of GNS-E. coli solution. (b) The SEM image of the GNS-E. coli in a lower magnification. (c) UV-vis spectra of bare E. coli, HAHC-GNS, and GNS-E. coli solutions, respectively. (d) Optical density (OD600) test and


23

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

luminescent adenosine triphosphate (ATP) assay (inset) of the bare E. coli and GNS-E. coli solutions over time for viability test. (e) Contact angle measurements (top row), cartoon illustrations (inset), and SEM images (bottom row) of bare silicon wafer, E. coli monolayer, HAHC-GNS monolayer on E. coli monolayer, and gold-silica core-shell nanoparticle (Au@SiO2) monolayer on E. coli monolayer. (f) From left to right, photographs of E. coli (top) and HAHCGNS (bottom) solutions and photographs of the E. coli solution (top, no interfacial trapping observed) and GNS-E. coli solution (interfacial trapping observed) after shaking for 2 h. (g) A representative SEM image of interfacially trapped GNS-E. coli. (h) A dark-field scattering microscope with a special fluidic chamber for real-time tracking of bare E. coli and GNS-E. coli at an air/water interface. (i and j) Time-resolved snapshots showing non-trapping mode of E. coli and stable trapping mode of GNS-E. coli at the interface. All images are captured from the Supporting Information Movies 1 and 2 and are top view. Scale bars are 2 µm. (k and l) Cartoon illustrations for out- and re-focused dark-field scattering images of E. coli and focused GNS-E. coli due to interfacial trapping at the interface.


24

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 3. General applicability of our method and the dynamic interfacial trapping phenomenon. From left to right, photographs of microbial cell solutions (perspective and top view), representative SEM images of microbial cells, cartoons of gold nanoparticles with different size and shape, photographs of gold nanoparticle attached microbial cell solutions (top view), and SEM images of gold nanoparticle attached microbial cell with different magnifications for (a) E. coli and 50 nm HAHC-GNS, (b) E. coli and gold nanorod (GNR), (c) P. putida and 30 nm HAHC-GNS, (d) C. glutamicum and 30 nm HAHC-GNS, and (e) S. cerevisiae and 30 nm HAHC-GNS, respectively.


25

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

Figure 4. Numerical simulations for interfacial trapping of hydrophobic nanoparticle-attached microbial cells under dynamic fluidic condition. (a) A schematic illustration of force exerting on


26

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

bacteria resulting from rotational speed. (b) Three-dimensional simulations of the distribution of microbial cells over time with or without shaking. (c) Representative trajectories of a GNSmicrobial cell with or without shaking during the same time. (d) Calculated number of trapped cells after 3 h varying normalized velocity. (e) Experimentally measured trapped cell area (graph) and photographs of trapped GNS-E. coli after 3 h shaking under different shaking speed.


27

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30


28

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 5. Sensitive and selective optical detections of microbial cells. (a) Sensitive optical detection of microbial cells by measuring an area of interfacially trapped microbial cells by using a smartphone. (b) Measured area of interfacially trapped E. coli with respect to bare E. coli concentration by using commercial image analyzer. (c) Selective Raman detection from interfacially trapped various microbial cells with the aid of the surface gold nanoparticles. Each Raman spectrum is obtained from silicon wafer (black), bare GNS monolayer (pink), GNS-S. cerevisiae (blue), GNS-C. glutamicum (yellow), GNS-P. putida (green), and GNS-E. coli (red), respectively. (d) Photographs of GNS-E. coli dispersed in bottled water (top row), tap water (middle row), and DI water (bottom row) before (left column) and after 1 h of shaking (right column). Scale bars are 1 cm. (e) Raman spectra of silicon wafer (black), GNS monolayer (pink), GNS-E. coli in DI water (brown), GNS-E. coli in tap water (orange), and GNS-E. coli in bottled water (light green). (f) Model GNS-microbial cell for electromagnetic (EM) simulation and EM field distribution images for curved and plane surfaces of GNS-microbial cell at different magnifications. (g) Normalized maximum electric field amplitude (|E|/|E0|) for the curved and plane surfaces.


29

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 30

TOC graphic


30

Active Surface Hydrophobicity Switching and ... - ACS Publications

Recommend Documents