MONITORING INTERSTRAIN DIFFICILE INTERACTIONS BY DIELECTROPHORETIC FINGERPRINTING Yi-Hsuan Su1, Cirle Warren2, Richard L. Guerrant2 and Nathan Swami1,* 1 Electrical and Computer Engineering, University of Virginia, USA and 2 Infectious Diseases, School of Medicine, University of Virginia, USA

ABSTRACT Clostridium difficile (C.difficile) (CDI) is quantified by enzyme immunoassays (EIAs) for the toxins produced by toxigenic C.difficile (TCD) strains. However, poor sensitivity due to toxin degradation and the lack of means for simultaneous monitoring of non-toxigenic C.difficile (NTCD) strains limit application of EIAs towards reducing CDI through antagonistic inter-strain interactions. Herein, we demonstrate that S (surface)-layer induced morphological differences within the region of C.difficile strains cause systematic variations in their dielectrophoretic (DEP) frequency spectra, due to alterations in their net wall capacitance. This enables the independent monitoring and separation of each strain from mixed samples.

KEYWORDS: Microbial, Microbiome, Clostridium difficile, Dielectrophoresis, S-layer proteins

INTRODUCTION Clostridium difficile (C.difficile) infection (CDI) is a global toxin-mediated intestinal disease that is commonly attributed to exposure to pathogenic C.difficile strains following the elimination of healthy microflora in the gut by broad-spectra antibiotics. A number of studies on animal model and a recent human phase one study suggest that asymptomatic colonization with non-toxigenic C.difficile (NTCD) strains can reduce the incidence of CDI from toxigenic C.difficile (TCD) strains [1]. State of the art enzyme immunoassays (EIAs) that detect the glutamate dehydrogenase (GDH) levels, as well as that of toxin A (TcdA) and/or toxin B (TcdB) levels are hampered by rapid degradation of the toxins. Furthermore, they do not offer means to simultaneously monitor NTCD and TCD strains, due to the lack of toxin production in NTCD strains. Herein we seek to address the need for simultaneously monitoring the levels and physiological alterations of each C.difficile strain within a mixed microbial sample, preferably in a label-free, non-destructive and real-time manner. S-layers proteins are part of the cell wall envelope in both gram positive and gram negative . S-layers are integral to surface recognition, gut colonization, host-pathogen adhesion and virulence. Studies have shown that antigenic variations of S-layers between C.difficile strains can be a potential alternative to serotyping by PCR-restriction fragment length polymorphism analysis and nucleotide sequencing [2], but these methods do not enable the recovery of intact microbials of each strain. Hence, we investigate the correlation of S-layer induced morphological variations to the cell electrophysiology for enabling inter-strain distinctions of intact C.difficile and other microbials exhibiting S-layer variations. Dielectrophoresis (DEP) causes the frequency-selective translation of polarized bio-particles in a spatially non-uniform electric field, either towards (by positive DEP or pDEP) or away (by negative DEP or nDEP) from the high field regions of a microfluidic device, depending on the polarizability of the bio- particle versus that of the medium. While prior work has focused on applying DEP towards the separation of microbial serotypes or in discerning persistent versus sensitive microbial subpopulations [3], the correlation of inter-strain DEP spectra to microbial morphology has not been reported. Herein, we demonstrate that differences in the S-layer constituting the cell wall of each C. difficile strain can enable their independent distinction due to morphological alterations to the cell wall.

EXPERIMENTAL DEP microfluidic chip fabrication: Standard PDMS (Poly-di-methyl-siloxane) micro-molding methods were used to microfabricate channels with sharp lateral constrictions (1000 m to 15 m). This so-called “constriction chip” was bonded using oxygen plasma treatment to a standard coverslip for easy

978-0-9798064-7-6/µTAS 2014/$20©14CBMS-0001 285 18th International Conference on Miniaturized Systems for Chemistry and Life Sciences October 26-30, 2014, San Antonio, Texas, USA microscopic viewing of DEP behavior. Using electrodes at the inlet and outlet, AC fields were applied over a wide-frequency range (0.05-5 MHz) by utilizing a power amplifier, for particle trapping towards or away from high field points at the constriction tips. The trajectory of the cells under this field was observed as high frames per second movies to quantify DEP velocity [3]. Bacteria strains preparation: C.difficile strains from ATCC were cultured in brain heart infusion (BHI) broth at 37°C under anaerobic condition prior to antibiotic treatment or dielectrophoretic analysis.

RESULTS AND DISCUSSION The morphological differences between two particular C.difficile strains: the high-toxigenic VPI10463 strain (HTCD) and the non-toxigenic strain VPI11186 (NTCD) are apparent from Transmission Electron Microscopy (TEM) images at 50k (Fig. 1a and c) and 100k (Fig. 1b and d) magnification. These show significantly higher cell wall roughness for the HTCD versus NTCD strains. One of the chief differences between each the respective strains is the S-layer proteins within the cell wall, which exhibit sequence variations and are higher within HTCD versus NTCD strains. This correlation is strengthened by prior results on the S-layer deficient mutant Tannerella forsythia, which shows much a smoother cell surface than the wild type [4].

Figure 1: Transmission electron microscope images of high-toxigenic C.difficile (HTCD) and non- toxigenic C.difficile (NTCD) strains at 50k (a&c) and 100k (b&d) magnification. Scale Bar: 0.2 m Dielectrophoresis (DEP) of biological particles, such as C.difficile, can be characterized using a shell model. Herein, particles exhibit a characteristic crossover frequency from nDEP to pDEP, as determined by the inverse RC time constant due to the net resistance (R) and capacitance (C) of the system. Using a parallel-plate model for the cell wall with spacing: d, and material permittivity, the capacitance is:

(1)

Hence, changes in surface roughness and area of the cell wall cause systematic differences in the net capacitance of each C.difficile strain, with the higher net capacitance HTCD versus the NTCD strain. The DEP crossover frequency (fxo) for each C.difficile strain can be related to the net wall capacitance (Cnet), at a given media conductivity (100 mS/m in our case), as follows:

(2) √

Hence, based on the higher net capacitance for the HTCD versus the NTCD strain, we anticipate a lower fxo for the HTCD versus the NTCD strain. In this current work, electrode-less rather than electrode- based DEP devices are utilized, since cell trapping under DEP does not occur at the electrode, where disruptions due to adsorption, electrolysis, electrothermal flow and electro-permeabilization of cells are considerable, but instead trapping occurs at or away from the tips of insulator constrictions where these disruptions are minimal [5, 6]. Upon optimization of conductivity at 100 mS/m, well separated DEP spectra for each strain are apparent, as per Fig. 2a, based on the translational velocity under the DEP trapping force for ~20 individual microbial cells to quantify the DEP spectra.

286 The lower crossover frequency for HTCD (300±75 kHz) versus NTCD (900±75 kHz) strains is consistent with its higher net wall capacitance due to higher surface area, as per TEM images in Fig. 1. The representative DEP behavior of each strains is showed in Fig. 2a-d. The well-separated DEP spectra suggest that the possibility of separation one strain from others with judicious choice of frequency. Based on the DEP response, we choose 400kHz at which HTCD strain exhibits strong pDEP as per Fig 2b while NTCD strain continues to show nDEP as per Fig. 2c. It is apparent from Fig. 2e, that such a separation can be accomplished in a facile manner, as confirmed by the DEP response and toxin production levels measured from pDEP trapped C.difficile. We envision that these characteristic spectral features in the 0.05-5 MHz range can offer the means for inter-strain microbial separations from heterogeneous samples.

Figure 2: DEP behavior of each C.difficile strains in the constriction region. HTCD strain shows (a) Negative DEP at 100kHz and (b) Positive DEP at 400kHz; NTCD strain shows (c) Negative DEP at 400kHz and (d) Positive DEP at 2MHz; and (e) Mixed sample where HTCD strain shows pDEP (red ar- row) vs NTCD strain shows nDEP (blue arrow).

CONCLUSION In this work, we demonstrate the capability to distinguish and separate C.difficile of varying strain- types based on differences in the S-layer that cause systematic alterations in their dielectrophoretic cross- over frequency. We envision the application of this microfluidic device towards point-of-care diagnostic applications for the development of therapies to arrest CDI by enabling the isolation of individual strains, the quantification of antibiotic resistant subpopulations and optimization of antibiotic treatments.

ACKNOWLEDGEMENTS This work was supported by University of Virginia’s Nanostar Seed Fund. The authors thank Dr. Glynis L. Kolling for helpful discussions.

REFERENCES [1] M. Natarajan, S. T. Walk, V. B. Young, D. M. Aronoff, "A Clinical and Epidemiological Review of Non-Toxigenic Clostridium Difficile," Anaerobe., 22, 1-5, 2013 [2] U. B. Sleytr, B. Schuster, E. M. Egelseer, D. Pum, "S-Layers: Principles and Applications," FEMS Microbiol. Rev., DOI: 10.1111/1574-6976.12063, 2014 [3] Y. H. Su, M. Tsegaye, W. Varhue, K. T. Liao, L. Adebe, J. A. Smith, R. L. Guerrant, N. S. Swami, "Quantitative Dielectrophoretic Tracking for Characterization and Separation of Persistent Subpopulations of Cryptosporidium Parvum," Analyst., 139, 66-73, 2014 [4] G. Sekot, G. Posch, P. Messne, et al. "Potential of the Tannerella Forsythia S-Layer to Delay the Immune Response," J. Dent. Res., 90, 109-114, 2011 [5] B. J. Sanghavi, W. Varhue, J. L. Chavez, C. F. Chou, N. S. Swami, "Electrokinetic Preconcentration and Detection of Neuropeptides at Patterned Graphene-Modified Electrodes in a Nanochannel," Anal. Chem., 86, 4120-4125, 2014 [6] V. Chaurey, A. Rohani, Y. H. Su, K. T. Liao, C. F. Chou, N. S. Swami, "Scaling Down Constriction- Based (Electrodeless) Dielectrophoresis Devices for Trapping Nanoscale Bioparticles in Physiological Media of High-Conductivity," Electrophoresis., 34, 1097-1104, 2013

CONTACT * Nathan S. Swami; phone: +1-434-924 -1390; [email protected]

287