Capillary Electrophoresis of Conidia from Cultivated Microscopic

Apr 22, 2009 - Marie Horka´ ,*,† Filip Ru˚ z icˇ ka,‡ Anna Kubesova´ ,† Veronika Hola´ ,‡ and Karel Sˇ lais†. Institute of Analytical ...
0 downloads 0 Views 315KB Size
Download PDF
Anal. Chem. 2009, 81, 3997–4004

Capillary Electrophoresis of Conidia from Cultivated Microscopic Filamentous Fungi ˇ lais† Marie Horka´,*,† Filip Ru˚zˇicˇka,‡ Anna Kubesova´,† Veronika Hola´,‡ and Karel S Institute of Analytical Chemistry Academy of Sciences of the Czech Republic, v. v. i., Veverˇ´ı 97, 602 00 Brno, Czech Republic, and Department of Microbiology, Faculty of Medicine, Masaryk University Brno, Czech Republic In immunocompromised people fungal agents are able to cause serious infections with high mortality rate. An early diagnosis can increase the chances of survival of the affected patients. Simultaneously, the fungi produce toxins and they are frequent cause of allergy. Currently, various methods are used for detection and identification of these pathogens. They use microscopic examination and growth characteristic of the fungi. New methods are based on the analysis of structural elements of the target microorganisms such as proteins, polysaccharides, glycoproteins, nucleic acids, etc. for the construction of antibodies, probes, and primers for detection. The above-mentioned methods are time-consuming and elaborate. Here hydrophobic conidia from the cultures of different strains of the filamentous fungi were focused and separated by capillary zone electrophoresis and capillary isoelectric focusing. The detection was optimized by dynamic modifying of conidia by the nonionogenic tenside on the basis of pyrenebutanoate. Down to 10 labeled conidia of the fungal strains were fluorometrically detected, and isoelectric points of conidia were determined. The observed isoelectric points were compared with those obtained from the separation of the cultured clinical samples, and they were found to be not host-specific. Filamentous fungi, including Penicillium sp., Aspergillus sp., and Fusarium sp., and their conidia are widespread in both outdoor and indoor environment. Exposure to their antigens is associated with a wide range of adverse health effects, especially genesis of allergic reactions, e.g., asthma.1 They have become an important cause of nosocomial infections with high mortality rate, especially in immunocompromised patients.2-4 Many fungal species have been recognized as allergens. Further, the microscopic filamentous fungi (FF) are equipped with the ability to produce mycotoxins (e.g., aflatoxin) which contaminates foods5 and represent serious risk for humans especially due to their carcinogenic effect. By reason of different pathogenesis and * To whom correspondence should be addressed. E-mail: [email protected]. Fax: (++420) 5-41212113. † Institute of Analytical Chemistry Academy of Sciences of the Czech Republic. ‡ Masaryk University Brno. (1) Douwes, J.; Pearce, N. Am. J. Epidemiol. 2003, 158, 203–206. (2) Lucas, G. M.; Tucker, P.; Merz, W. G. Clin. Infect. Dis. 1999, 29, 1594– 1596. (3) Latge´, J.-P. Clin. Microbiol. Rev. 1999, 12, 310–350. (4) Fridkin, S. K.; Narcis, W. R. Clin. Microbiol. Rev. 1996, 9, 499–511. (5) Wogan, G. N. Bacteriol. Rev. 1966, 30, 460–70. 10.1021/ac900374v CCC: $40.75  2009 American Chemical Society Published on Web 04/22/2009

variable sensitivity to antifungal agents of these pathogens, early detection and correct identification of the fungal pathogens is essential for choosing an optimal therapy and for assessing the prognosis of the infection.6 Most common methods7 used for routine detection and identification of filamentous fungi are based on morphological (conidia or hyphae arrangement, shape of conidia, etc.) and physiological characteristics (growth, colony character, etc.).8 These methods are laborious and time-consuming. Especially, growth of these microorganisms takes from several days to several weeks. Moreover, the reliability of their results depends on a number of conditions (the media quality and culture conditions, etc.) and they may be easily influenced by subjective error in results evaluation. Currently, some new methods9-12 appeared, for example, the rapid identification and characterization products and antigens of FF13-16 and methods based on detection of nucleic acids by polymerase chain reaction (PCR).8 The reliability of the methods is high. On the other hand, these methods are laborious and expensive. Fungi in culture are identified by the shape and formation of the conidia in routine microbiological laboratories. Species differentiation is based on the formation of conidia as well as their color, shape, texture, and size. They are usually produced at the tip or side of hyphae, the body of a typical fungus, or on the special spore-producing structures called conidiophores. Mature spores detach and spread to the environment.8 (6) Nucci, M.; Spector, N.; Bueno, A. P.; Slza, C.; Perecmanis, T.; Bacha, P. C.; Pulcheri, W. Clin. Infect. Dis. 1997, 24, 575–579. (7) Douwes, J.; Thorne, P.; Pearce, N.; Heederik, D. Ann. Occup. Hyg. 2003, 47, 187–200. (8) De Hoog, G. S.; Guarro, J.; Gene, J.; Figueras, M. J. General techniques. In Atlas of Clinical Fungi, 2nd ed.; De Hoog, G. S., Guarro, J., Gene´, J., Figueras, M. J., Eds.; Centraalbureau voor Schimmelcultures: Utrecht, The Netherlands, 2000; pp 39-54. (9) Floriano, P. N.; Christodoulides, N.; Romanovicz, D.; Bernard, B.; Simmons, G. W.; Cavell, M.; McDevitt, J. T. Biosens. Bioelectron. 2005, 20, 2079– 2088. (10) Madonna, A. J.; Voorhess, K.; Taranenko, N. I.; Laiko, V. V.; Doroshenko, V. M. Anal. Chem. 2003, 75, 1628–1637. (11) James, G. S.; Sintchenko, V. G.; Dickeson, D. J.; Gilbert, G. L. J. Clin. Microbiol. 1996, 34, 1572–1575. (12) De Vos, M. M.; Nelis, H. J. J. Microbiol. Methods 2003, 55, 557–564. (13) Farrell, S.; Halsall, H. B.; Heineman, W. R. Analyst 2005, 130, 489–497. (14) Maertens, J.; Verhaegen, J.; Demuynck, H.; Brock, P.; Verhoef, G.; Vandenberghe, P.; Van Eldere, J.; Verbist, L.; Boogaerts, M. J. Clin. Microbiol. 1999, 37, 3223–3228. (15) Aquino, V. R.; Gildami, L. Z.; Pasqualotto, A. C. Mycopathologia 2007, 163, 191–202. (16) Linder, M. B.; Szilvay, G. R.; Nakari-Setala, T.; Pentlila, M. E. FEMS Microbiol. Rev. 2005, 29, 877–896.

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

3997

The cell surface of conidia consists of various components including proteins, lipids, acids, and polysaccharides.17,18 The hydrophobins are generally found on the outer surface of conidia and of the hyphal wall.19 Hydrophobins are a class of small, cysteine-rich proteins that are expressed only by FF. They are known for their capability of forming a hydrophobic coating16 on a surface of an object and may be involved in mediating contact and communication between the fungus and its environment. The hydrophobic effect is also the tendency of nonpolar species to aggregate in water solution.20 To date, little is known about the molecular mechanisms of spore aggregation/adhesion.21 From the capillary electromigration techniques, CE, most often capillary zone electrophoresis, CZE, and capillary isoelectric focusing, CIEF, are used. At CZE separation the properties of microorganisms, MOs, can be characterized according to their electrophoretic mobilities, which are dependent on the background electrolytes, BGE, composition. There are no papers concerned with detection of the fungal conidia. In the majority of previous CE studies involving bacteria and/or bacterial spores they were detected using ultraviolet-visible (UV-vis) absorbance.17,22-27 One study28 involved the detection of Bacillus thuringiensis spores by measuring dipicolinic acid levels, an acid found in spore coats. Here, using CE with UV-vis absorbance 7.2 × 105 spores mL-1 was possible to detect. However, the low intrinsic fluorescence possessed by spores in their native state disables their direct detection by fluorescence methods.29,30 When utilizing fluorescent dye bound to spore surfaces sensitive fluorometric detection can be used. A rapid and sensitive capillary electrophoresis-laser-induced fluorescence (CE-LIF) laserinduced method for the separation and detection of low levels of Bacillus globigii (B. globigii) spore concentrations, which requires minimal sample preparation, was introduced in ref 28. Several dyes, including fluorescamine, C-10, NN-127, Red-1c, and indocyanine green, were utilized as noncovalent labels for proteins on the B. globigii spore surface. Resulting electropherograms showed unique fingerprints for each dye with B. globigii spores. Different fluorescent dye labels yield different sensitivities and fingerprints. For the collection of the Bacillus subtilis spores, native31 or fluorescently32 labeled dielectrophoresis was also used.

At CIEF separation the amphoteric bioparticles can be characterized by the isoelectric point,17,25,33-35 pI, which can be one of the potentially suitable issues for their identification.36 At the trace analysis of the hydrophobic conidia, some difficulties35 such as their adsorption onto the capillary wall, the aggregation between conidia and the additives in buffer solution, and the sensitivity of the detection can be expected. Generally, the adsorption of the cells can be prevented and the peak shape can be improved by use of permanent capillary coating37 or by use of the additives such as poly(ethylene glycol) (PEG).25,35,38-40 The sensitive fluorometric detection of the conidia depends mainly on the tagging the cells by fluorophores40,41 without significant change of the respective isoelectric points.41,42 In this contribution the possibility to separate of FF conidia by CZE and CIEF with UV and fluorometric detection was verified, and then the electromigration properties of the filamentous fungi were monitored. For these experiments the conidia of FF strains, Aspergillus niger (A. nigersconidia are brown to black, very rough, globose, and measure 4-5 µm in diameter8), Aspergillus fumigatus (A. fumigatussconidia are smooth to finely roughened, subglobose, 2-3.5 µm in diameter8), Aspergillus flavus (A. flavussconidia are smooth to very finely roughened, globose to subglobose, 3-6 µm in diameter8), Fusarium solani (F. solanisconidia are borne from long monophialides, are one- to three-celled, 2-5 × 8-16 µm long8), and Penicillium chrysogenum (P. chrysogenumsconidia are 2.5-5 µm in diameter, round, unicellular, and visualized as unbranching chains at the tips of the phialides8) were selected. The electromigration properties including isoelectric points of various native and labeled conidia of each FF species including cultured clinical samples were determined and compared with each other. Conidia were dynamically modified by the nonionogenic tenside on the basis of pyrenebutanoate, poly(ethylene glycol) 4-(1-pyrene) butanoate (PB-PEG).38

Kenndler, E.; Blaas, D. Trends Anal. Chem. 2001, 20, 543–551. Ho, J. Anal. Chim. Acta 2002, 457, 125–148. Whiteford, J. R.; Spanu, P. D. Fungal Genet. Biol. 2001, 32, 159–168. Doyle, R. J. Microbes Infect. 2000, 2, 391–400. van der Aa, B. C.; Asther, M.; Dufrene, Y. F. Colloids Surf., B 2002, 24, 277–284. Palenzuela, B.; Simonet, B. M.; Garcia, R. M.; Rios, A.; Valca´rcel, M. Anal. Chem. 2004, 76, 3012–3017. Ebersole, R. C.; McCormick, R. M. Biotechnology 1993, 11, 1278–1282. Klodzinska, E.; Dahm, H.; Rozycki, H.; Szeliga, J.; Jackowski, M.; Buszewski, B. J. Sep. Sci. 2006, 29, 1180–1187. Armstrong, D. W.; Schulte, G.; Schneiderheinze, J. M.; Westenberg, D. J. Anal. Chem. 1999, 71, 5465–5469. Armstrong, D. W.; Schneiderheinze, J. M. Anal. Chem. 2000, 72, 4474– 4476. Schneiderheinze, J. M.; Armstrong, D. W.; Schulte, G.; Westenberg, D. J. FEMS Microbiol. Lett. 2000, 189, 39–44. He, J.; Luo, X. F.; Chen, S. W.; Cao, L. L.; Sun, M.; Yu, Z. N. J. Chromatogr., A 2003, 994, 207–212. Chichester, K. D.; Silcott, D. B.; Colyer, Ch. L. Electrophoresis 2008, 29, 641–651. Hairston, P. P.; Ho, J.; Quant, F. R. J. Aerosol Sci. 1997, 28, 471–482.

(31) Hoettges, K. F.; Hughes, M. P.; Cotton, A.; Hopkins, N. A. E.; McDonell, M. B. IEEE Eng. Med. Biol. Mag. 2003, 22, 68–74. (32) Lapizco-Encinas, B. H.; Davalos, R.; Simmons, B. A.; Cumings, E. B.; Fitschenko, Y. J. Microbiol. Methods 2005, 62, 317–326. (33) Horka´, M.; Ru˚zˇicˇka, F.; Horky´, J.; Hola´, V.; Sˇlais, K. J. Chromatogr., B 2006, 841, 152–159. (34) Shen, Y.; Berger, S. J.; Smith, R. D. Anal. Chem. 2000, 72, 4603–4607. (35) Horka´, M.; Horky´, J.; MatouSˇkova´, H.; Sˇlais, K. Anal. Chem. 2007, 79, 9539–9546. (36) Rijnaarts, H. H. M.; Norde, W.; Lyklema, J.; Zehnder, A. J. B. Colloids Surf., B 1995, 4, 191–197. (37) Horka´, M.; Planeta, J.; Ru˚zˇicˇka, F.; Sˇlais, K. Electrophoresis 2003, 24, 1383– 1390. (38) Horka´, M.; Ru˚zˇicˇka, F.; Horky´, J.; Hola´, V.; Sˇlais, K. Anal. Chem. 2006, 78, 8438–8444. (39) Kaper, H. J.; Busscher, H. J.; Norde, W. J. Biomater. Sci., Polym. Ed. 2003, 14, 313–324. (40) Desai, M. J.; Armstrong, D. W. Microbiol. Mol. Biol. Rev. 2003, 67, 38– 51. (41) Li, Y.; Buch, J. S.; Rosenberger, F.; De Voe, D. L.; Lee, C. S. Anal. Chem. 2004, 76, 742–748. (42) Sze, N. S. K.; Huang, T. M.; Pawliszyn, J. J. Sep. Sci. 2002, 25, 1119–1122.

(17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30)

3998

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

EXPERIMENTAL SECTION Chemicals. The buffer component tris(hydroxymethyl)aminomethane, Tris, γ-glycidoxypropyltrimethoxysilane (GOPTMS), and Brij 35 were obtained from Sigma (St. Louis, MO). Octadecylcyclotetrasiloxane (D4 reagent) was from VCHZ Synthezia (Kolı´n, Czech Republic). The high-resolution ampholyte, pH 2-4, was from Fluka Chemie GmbH (Buchs, Switzerland). Poly(ethyl-

ene glycol) (Mr 400, 1000, 4000, and 10 000), taurine, and 4(1-pyrene)butyric acid were from Aldrich (Milwaukee, WI). The solution of synthetic carrier ampholytes, Biolyte, pH 3-10, was obtained from Bio-Rad laboratories (Hercules, CA), L-aspartic acid (Asp) was from LOBA Chemie, Wien, Austria, and 2[4-(2-hydroxyethyl)-1-piperazinyl]-ethanesulfonic acid (HEPES) was from Merck, Darmstadt, Germany. All chemicals were analytical grade. Poly(ethylene glycol) pyrenebutanoate, UV-detectable pI markers, pI 2.0, 4.0, 6.6, 7.0, and fluorescein-based pI markers, pI 1.8, 3.0, 4.0, 4.7, 5.5, 6.6, were synthesized in the Institute of Analytical Chemistry Academy of Sciences of the Czech Republic, v. v. i., Brno. PB-PEG was prepared by the reaction of 4-(1-pyrene) butyric acid and PEG 400.38 The specifications43,44 of the used spacers, simple ampholytes, are described in ref 45. FF strains. The strains included in this study, A. niger CCM 8222, CCM 8189, A. fumigatus CCM 3960, P. chrysogenum CCM F-362, F. solani CCM 8014, and A. flavus CCM F-449 were obtained from the Czech Collection of Microorganisms (CCM; Brno, Czech Republic). The other strains (A. niger A-33, A. fumigatus BAR07, A. fumigatus PAS07, A. fumigatus A-19, and P. chrysogenum A-246) were isolated from clinical material at Department of Microbiology (Faculty of Medicine Masaryk University and St. Anna Faculty Hospital in Brno, Czech Republic). Preparation of the Microbial Sample. The tested strains were cultivated on Sabouraud dextrose agar (HiMedia, Mumbai, India) at 30 °C for 3 days. The microbial cultures were poured over by 5 mL of physiological saline solution (PSS), and conidia were released from mycelium by vortexing. The concentration of the resuspended spores was estimated by the suspension turbidity detector DEN-1 McFarland densitometer. The numbers of microorganisms in the reference samples were controlled by serial dilution and plating of 100 µL of the suspension on Sabouraud dextrose agar. After the cultivation at 30 °C for 3 days, the colonies were counted. Safety. According to the international standards the used strains of filamentous fungi belong to maximum to the biohazard group 2smicroorganisms that may cause human and/or animal diseases and which might be a hazard to laboratory workers but are unlikely to spread in the community. Laboratory exposure rarely produces infection, and effective prophylaxis or effective treatment is available.46 CZE and CIEF: Equipment and Procedure. The capillary zone electrophoretic and capillary isoelectric focusing experiments were carried out using the laboratory-made apparatus44 at constant voltage (-) 20 kV supplied by a high-voltage unit Spellman CZE 1000 R (Plainview, NY). The length of the fused-silica capillaries (FS), 0.1 mm i.d. and 0.25 mm o.d. (Pliva-Lachema a. s., Brno, Czech Republic), was from 300 to 550 mm and from 150 to 300 mm to the detector; effective volume of the column was 1.2-2.4 µL. The ends of the fused-silica capillary were dipped in 3 mL glass vials with the electrodes and the BGE in CZE or with the anolyte or the catholyte solutions in CIEF. During the CIEF (43) Hirokawa, T.; Nishino, M.; Aoki, N.; Sawamoto, Y. K. T. Y.; Akiyama, J.-I. J. Chromatogr., A 1983, 271, D1–D106. (44) Acevedo, F. J. Chromatogr., A 1991, 545, 391–396. (45) Horka´, M.; Ru˚zˇicˇka, F.; Hola´, V.; Sˇlais, K. Anal. Bioanal. Chem. 2006, 385, 840–846. (46) http://www.sci.muni.cz/ccm/index.html.

experiments, the current decreased from 40 to 60 µA at the beginning of the experiment down to 3 or 6 µA at the time of detection, depending on the sampling time interval and the sample solution. The on-column UV-vis detector LCD 2082 (Ecom, Prague, Czech Republic) connected to the detection cell by optical fibers (Polymicro Technologies, Phoenix, AZ) was used at the wavelength of 280 nm. The PU4027 programmable fluorescence detector (Philips Scientific, Cambridge, Great Britain) was modified for the on-column fluorometric detection. The excitation wavelength, λEX, was 335 nm, the emission wavelength, λEM, was 480 nm, and the width of the detection window was 1 mm. The optical density of the suspension of spores was measured using the suspension turbidity detector DEN-1 McFarland densitometer (BioSan, Riga, Latvia). The sample injection was accomplished by siphoning action obtained by elevating of the inlet reservoir on the side of the anode relative to the outlet reservoir side of the cathode. The height difference of the reservoirs for the sample injection, ∆h, was in range from 100 to 200 mm; the time of injection, tinj, was from 8 to 45 s at CZE or CIEF experiments. At CIEF the segmental injection of the sample pulse was used. Before separation of the microbial sample it was necessary to homogenize it. The clusters of the conidia were disrupted by the sonication of the microbial suspension in the ultrasound bath Sonorex, Bandelin electronic (Berlin, Germany). The sonication was processed for 2 min at the temperature 25 °C and at frequency of 35 kHz. After the sonication the sample of conidia was vortexed for 10 min (Vortex-Genie 2, Scientific Industries, Bohemia, U.S.A.) and then immediately used. The detector signals were acquired and processed with the chromatography data station Clarity (DataApex s.r.o., Praha, Czech Republic). Preparation of the FS Capillaries. (1) In one of the experiments the capillary was dynamically coated with D4 reagent and then heated for 3 h at 380 °C after sealing both ends. Then the capillary was washed with a mixture of ethyl alcohol (EtOH) and acetone (1:1) and subsequently with the anolyte for 3 min and before using at CIEF.47 (2) In the other experiment the capillary was rinsed with 5% (v/v) of GOPTMS47 dissolved in methanol for 10 min and then heated for 5 min at 90 °C. Subsequently the capillary was washed by EtOH for 10 min, for 3 min with catholyte before using at CIEF. Electrolyte Systems. At CIEF 3 × 10-2 to 4 × 10-2 mol L-1 sodium hydroxide and 0.1 mol L-1 orthophosphoric acid were used as the catholyte or the anolyte solution, respectively, with or without addition of 3% (v/v) ethanol. The solution of 0.1% (w/v) Brij 35, 0.3% (w/v) PEG 1000 or 4000 or the solution of 7 × 10-5 mol L-1 PB-PEG, 0.3% (w/v) and 1% (w/v) PEG 10 000 were added into the catholyte and the anolyte, when we used the UV or the fluorometric detection, respectively. At CZE the BGEs were composed of 2.0 × 10-3 mol L-1 taurine-Tris buffer (pH 8.4), 3% (v/v) EtOH. In case of the usage of the UV detection or eventually in case of the fluorometric detection 0.3% (w/v) PEG 1000 and 0.1% (w/v) (47) Blomberg, L.; Markides, K.; Wannman, T. In Proceedings of the Fourth International Symposium on Capillary Chromatography; Kaiser, R. E., Ed.; Hindelang, Hu ¨ thig Verlag: Heidelberg, Germany, 1981; p 73.

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

3999

Brij 35 or 7 × 10-5 to 8 × 10-5 mol L-1 PB-PEG and 0-0.3% (w/v) PEG 10 000, respectively, were added into BGE. Before each injection the capillaries were rinsed with ethanol for 10 min and then back-flushed with BGE (CZE) or catholyte (CIEF) for 2 min. The rinsing procedures were carried out hydrodynamically. Preparation of the Samples, CZE. The model mixture of 108 conidia/mL from all tested strains was dissolved in PSS with 1% (v/v) PEG 1000 when we used the UV detection. The injected volumes of the analytes were 40 nL (∆h ) 100 mm, tinj ) 10 s), which introduces thousands of conidia of each from FF injected into the capillary. At CZE with the fluorometric detection the model mixture of 104-106 conidia/mL of FF was dissolved in 15 × 10-3 mol L-1 PSS, 2 × 10-4 to 5 × 10-4 mol L-1 PB-PEG, and 20% (v/v) EtOH. The sample of conidia was stored for 10 min at 20 °C before use. The injected volumes of the analytes were 10-90 nL (∆h ) 100-200 mm, tinj ) 8-20 s, which introduces maximum of hundred conidia of each from the fungi injected into the capillary. Composition of the Sample Pulse at CIEF. At the segmental injection33 the sample was injected into the capillary in three parts: segment of the spacers,45 solution of the selected simple ampholytic electrolytes HEPES and Asp (2:1) dissolved in the catholyte, the segment of the sample mixture of conidia, and the segment of the mixture of 5% (w/v) commercial carrier ampholytes and pI markers for UV detection, pI 2.0, 4.0, 6.6, 7.0 (25 µg mL-1 of each), or fluorescent pI markers, pI 1.8, 3.0, 4.0, 4.7, 5.5, and 6.6 (5 µg mL-1 of each), for the tracing of the used pH gradient of 2.0-8.0. The applied synthetic carrier ampholytes, Biolyte pH 3-10 and ampholyte pH 2-4, were used in the ratio 2:1. The height differences of the reservoirs at the injection of the segments were 100-200 mm, tinj of the segment of spacers was 25 s, sample segment was 8-20 s, and segment of carrier ampholytes and pI markers was 45 s. When we used the UV detection the second segment of the sample mixture was composed of the solution of 108-109 conidia/ mL (see the section Preparation of the Samples, CZE) dissolved in PSS with or without 0.1% (w/v) PEG 4000 or 1.0% (w/v) PEG 1000. The injected volumes of the analytes were 30-60 nL (∆h ) 100, tinj ) 8-15 s). The fresh model mixture of conidia of FF were dissolved in 15 × 10-3 mol L-1 PSS, 5 × 10-4 mol L-1 PB-PEG, and 20% (v/v) EtOH at CIEF with the fluorometric detection. The sample mixtures were stored for 10 min at 20 °C before use like as in the section Preparation of the Samples, CZE. The injected volumes of the analytes were approximately 45 nL (∆h ) 200 mm, tinj ) 10 s) which represents maximum tens of conidia of each from FF injected into the capillary. RESULTS AND DISCUSSION CIEF of the Conidia with UV Detection. The genus Aspergillus belongs to the significant members of the group of the filamentous fungi. One of them, A. fumigatus (belonging to the group of the most common Aspergillus species to cause disease in immunocompromised individuals), was chosen for the preliminary researching of the conditions for the separation of its conidia by CIEF with UV detection. Seeing that the separation of unknown 4000

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

Figure 1. Separation of the conidia of A. fumigatus by CIEF with UV detection, the optimization. (A) The catholyte, 3 × 10-2 mol L-1 sodium hydroxide, the anolyte, 0.1 mol L-1 H3PO4; 3% (v/v) EtOH and 0.1% (w/v) PEG 4000 were added into the catholyte and the anolyte; (B and C) see (A) FS capillary was prepared by D4 reagent or GOPTMS, respectively; other conditions and designations, FS 0.1 mm i.d., 0.25 mm o.d., length 300 mm, 150 mm to the detection cell; applied voltage (-) 20 kV; wavelength, 280 nm; sample composition (∆h, 100 mm), segment of spacers, HEPES, and Asp (2:1) dissolved in the catholyte, tinj, 25 s; segment of the 5% (w/v) water solution of the synthetic carrier ampholytes, Biolyte, pH 3-10 and ampholyte pH 2-4 (2: 1), tinj, 45 s; sample, 2 × 108 conidia/mL of A. fumigatus CCM 3960, dissolved in PSS, tinj, 8 s; rinsing procedure, EtOH for 10 min and then back-flushed with the catholyte for 2 min; migration time, t [min].

microbial analytes, conidia, can be made, we extended the commonly used pH gradient at the separation of the yeasts (pH range of 2.0-5.5)49 preferably on the pH range of 2.0-8.0, see Figures 1 and 2. The problem with reproducibility and the linearity of the pH gradient was solved once again by applying the segmental injection.33,45 In these experiments the first segment was composed of the suitable spacers HEPES and Asp in the ratio 2:1, dissolved in the catholyte and the third segment, the segment of the commercial carrier ampholytes, was composed of Biolyte pH 3-10 and ampholyte 2-4 in the ratio 2:1. The pH gradient was traced by the UV-detectable pI markers 2.0, 4.0, 6.6, and 7.0, see Figure 2. For the estimation of the adherence of the conidia to the inner surface of the capillary the catholyte, 3 × 10-2 mol L-1 NaOH, the anolyte, 0.1 mol L-1 H3PO4, without additives, and the sample segment composed of 2 × 108 conidia/mL (48) Horka´, M.; Willimann, T.; Blum, M.; Nording, P.; Friedl, Z.; Sˇlais, K. J. Chromatogr., A 2001, 916, 65–71. (49) Horka´, M.; Ru˚zˇicˇka, F.; Hola´, V.; Sˇlais, K. Electrophoresis 2007, 28, 2300– 2307.

Table 1. Isoelectric Points of the Conidia from Different Strains of the Filamentous Fungi Included in This Study

Figure 2. CIEF of the conidia of FF in the pH gradient pH range of 2.0-8.0 with UV detection. Conditions and designations, see Figure 1; the catholyte, 4 × 10-2 mol L-1 sodium hydroxide, and the anolyte, 0.1 mol L-1 H3PO4, were dissolved in 3% (v/v) EtOH, 0.1% (w/v) Brij 35, and 0.3% (w/v) PEG 1000; UV-detectable pI markers pI 2.0, 4.0, 6.6, and 7.0 were added to the segment of the carriers; sample composition, conidia of A. niger CCM 8222, A. fumigatus CCM 3960, P. chrysogenum CCM F-362, F. solani CCM 8014, and A. flavus CCM F-449 were dissolved in PSS and 1% (w/v) PEG 1000, tinj, 15 s.

from A. fumigatus CCM 3960 (approximately 6 × 103 particles), resuspended only in PSS, were used at first. According the procedure described in the section CZE and CIEF: Equipment and Procedure it was necessary to sonicate and vortex the suspensions of the conidia to prevent the aggregation20 as a result of the hydrophobic effect.36 Without the use of this procedure the aggregates of the conidia in the sample segment were injected into the capillary; the zone of the sample was not fully focused and therefore was poorly detectable. In addition, without additives in the catholyte and the anolyte solutions the conidia were partly adsorbed onto the capillary surface also, and their focusing was insufficient. After addition of 3% (v/v) of EtOH in the catholyte and the anolyte, the adherence of the conidia to the inner surface of the capillary was decreased and the peak of the focused conidia of A. fumigatus was detected. Anyway for the reproducible results it was necessary to rinse the capillary between the focusing runs as the prevention of the massive adsorption of the conidia onto the surface of the capillary. The sequential addition of 0.3% or 0.1% (w/v) PEG 4000 in the catholyte and the anolyte or in the suspension of the conidia, respectively, increased the adherence of the nonpolar conidia to the capillary dynamically coated by PEG 4000, see Figure 1A. In the next

strain

isoelectric point, pI

A. niger CCM 8222 A. niger CCM 8189 A. niger A-33 A. fumigatus CCM 3960 A. fumigatus BAR07 A. fumigatus PAS07 A. flavus CCM F-449 F. solani CCM 8014 P. chrysogenum CCM F-362 P. chrysogenum A-246

2.0 2.1 2.0 3.1 3.2 3.1 5.9 4.6 4.4 4.5

two experiments, Figure 1, parts B and C, we changed the surface properties of the separation capillaries by their silylating with low hydrophobic GOPTMS or with high hydrophobic D4 reagent, respectively. The other separation conditions were left unchanged. As results from these CIEF experiments the inner surface of the capillary would have been low hydrophobic, see Figure 1, part B versus part C, the addition of EtOH and PEG, see Figure 1, parts B and C, in the catholyte and the anolyte solution or in the sample segment, respectively, were suitable. At the optimized conditions the catholyte, 4 × 10-2 mol L-1 NaOH, and the anolyte, 0.1 mol L-1 H3PO4, were dissolved in 3% (v/v) EtOH, 0.3% (w/v) PEG 1000, and 0.1% (w/v) Brij 35. The samples, conidia of A. niger CCM 8222, A. fumigatus CCM 3960, P. chrysogenum CCM F-362, F. solani CCM 8014, and A. flavus CCM F-449 (each of them 2 × 108 conidia in 1 mL-1, approximately 12 000 particles), were resuspended in PSS and 1% (w/v) PEG 1000 in accordance with the experimental requirement. The UV-detectable pI markers for the tracing of the pH gradient were selected according the preliminary ascertained isoelectric points of the conidia from individual strains of FF. The narrow zones of the separated analytes from the sample pulse were detected at the optimized separation condition, see Figure 2A-E. The values of pIs of conidia were calculated from the migration times of the selected pI markers and their isoelectric points. The isoelectric points of the conidia from these collection strains and the clinical isolates are summarized in the Table 1. The average values of the isoelectric points of the examined conidia were calculated from a minimum of three measurements for each of the strains from this table. The values of pIs in the pH gradient pH range of 2.0-8.0 were found to be not host-specific for the conidia from the strains A. niger, A. fumigatus, and P. chrysogenum like as in ref 35. The pIs were determined as 2.0 for A. niger (three strains, RSD ) 2.5%), 3.1 for A. fumigatus (three strains, RSD ) 1.7%), 5.9 for A. flavus CCM-F449 (one strain, nine measurements, RSD ) 2.9%), 4.6 for F. solani CCM 8014 (one strain, nine measurements, RSD ) 2.0%), and for P. chrysogenum 4.4 (two strains, RSD ) 1.8%). Of course, in this broad pH gradient pH range of 2.0-8.0 the values of pIs are only approximate. Nevertheless, it may be supposed that at least we can separate the conidia from different strains of the genus Aspergillus. For the conidia of A. niger calibration curve yielded a correlation coefficient of 0.98. CZE of the Conidia with the UV Detection. As the basic buffer component 2 × 10-3 mol L-1 taurine-Tris buffer (pH 8.4) Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

4001

Figure 3. CZE of the FF conidia with UV detection. Conditions and designations, see Figures 1 and 2; BGE composition, 2 × 10-3 mol L-1 taurine-Tris buffer (pH 8.4), 3% (v/v) EtOH, 0.3% (w/v) PEG 1000, and 0.1% (w/v) Brij 35; sample composition, 108 conidia/mL of FF dissolved in PSS with 1% (w/v) PEG 1000; ∆h ) 100 mm, tinj ) 10 s.

was used here according to the experiences from the separation of yeast by CZE in ref 49. The selection of the buffer additives, 3% (v/v) EtOH, 0.3% (w/v) PEG 1000, and 0.1% (w/v) Brij 35, and the sample segment composition, 108 conidia in 1 mL-1 dissolved in PSS with 1% (w/v) PEG 1000, was the same as in the section CIEF of the Conidia with UV Detection. The number of the detected conidia was approximately 9000. The separation of the conidia from the strains of A. niger CCM 8222, A. fumigatus CCM 3960, P. chrysogenum CCM F-362, F. solani CCM 8014, and A. flavus CCM F-449 according their electrophoretic mobilities is depicted in the electropherogram in Figure 3. Under optimized conditions narrow zones of the separated conidia were detected. RSDs from the replicated measurements of the migration times, t, of the conidia from the individual examined strains listed in Table 1 were up to 2.2%. CZE of the Conidia Dynamically Modified by PB-PEG with the Fluorometric Detection. In the experiments, see Figure 4A-G, we would like to demonstrate the optimization procedure for the sensitive detection of the labeled conidia from the strains of FF mentioned above. Here, the additives diminish the adherence of the conidia to the inner surface of the capillary, the nonionogenic tenside and PEG 1000 (see Figure 3), were substituted by PB-PEG. The basic buffer composition, 2 × 10-3 mol L-1 taurine-Tris buffer (pH 8.4) with addition of 3% (v/v) EtOH, was the same as in previous CZE experiments with UV detection. The samples of the fungal conidia injected into the capillary were always resuspended in 15 × 10-3 mol L-1 PSS and 20% (v/v) EtOH. At the individual CZE runs the concentration of PB-PEG in the sample and/or in BGE were changed. At the experiments 4002

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

Figure 4. CZE of the fungal conidia dynamically modified by PBPEG, the optimization. (A) BGE, 2 × 10-3 mol L-1 taurine-Tris buffer (pH 8.4), and 3% (v/v) EtOH, 7 × 10-5 mol L-1 PB-PEG; sample composition, 106 conidia/mL of A. niger resuspended in 2 × 10-4 mol L-1 PB-PEG, 15 × 10-3 mol L-1 PSS, and 20% (v/v) EtOH, ∆h ) 200 mm, tinj ) 20 s; (B) see (A) BGE, 8 × 10-5 mol L-1 PB-PEG; sample composition, 105 conidia/mL of A. niger, tinj ) 10 s; (C) see (B) sample composition, 105 conidia/mL of F. solani, dissolved in 4 × 10-4 mol L-1 PB-PEG; (D) see (C) sample composition, 105 conidia/ mL of A. flavus; (E) see (D) BGE, 7 × 10-5 mol L-1 PB-PEG; sample composition, 105 conidia/mL of A. fumigatus, dissolved in 5 × 10-4 mol L-1 PB-PEG; (F) see (E) sample composition, 104 conidia/mL of each of A. niger, F. solani, A. flavus, A. fumigatus, 105 conidia/mL of P. chrysogenum, tinj ) 20 s; (G) see (F) BGE, 0.3% (w/v) PEG 1000; sample, 104 conidia/mL, tinj ) 12 s; conditions and designations, see Figures 1-3; FS 0.1 mm i.d., 0.25 mm o.d., length 550 mm, 250 mm to the detection cell; applied voltage (-) 20 kV, λEX ) 335 nm, λEM ) 480 nm; rinsing procedure, 10 min with acetone/ethanol mixture (10:1 v/v), 2 min with buffer.

depicted in Figure 4A, 2 × 10-4 and 7 × 10-5 mol L-1 of PBPEG were added in the sample (106 conidia/mL, approximately 100 conidia in the sample zone from the strain of A. niger) or dissolved in BGE, respectively. The incubation time was 10 min. The broad peak of the conidia was detected. The concentration of the nonionogenic tenside, PB-PEG, in the sample/BGE was too low on its sufficient adsorption on the hydrophobic surfaces of all injected conidia in the sample pulse. Therefore, the conidia were adhered/adsorbed on the inner surface of the capillary. At the enhancement of PB-PEG in the BGE on 8 × 10-5 mol L-1 and 20-fold reduction of the number of the conidia in the sample pulse, see Figure 4B, the migration velocity of the conidia was reduced and the peak of the conidia was slimmer. The higher response on the relatively large residue of PB-PEG from the electrolyte system sample pulse/BGE was detected; the migration velocity of the nonionogenic tenside matches to the velocity of the electroosmotic flow and the components from the sample pulse. In the next Figure 4C the same number of each of the conidia of A. niger and F. solani as in the previous Figure 4B were injected into the capillary in the sample pulse. With regard to the large concentration sensitivity of this technique on the change of the number of separated particles the concentration of the noninonogenic tenside in the sample was increased on 4 × 10-4 mol L-1. A relatively narrow peak of the conidia of F. solani was detected; the migration time of the conidia of A. niger was increased with respect to the residue of PBPEG. At the follow-up enrichment of the sample zone any of the additional conidia of A. flavus (the same number of the conidia as from the strains A. niger and F. solani, see Figure 4C) three peaks of the conidia zone were detected, see Figure 4D. The migration times of the conidia of F. solani and A. niger were decreased. The smallish residue of PB-PEG was detected. We presumed that the residue of PB-PEG in Figure 4B-D depends on the concentration of PB-PEG in the BGE. Hence, we decreased the concentration of the nonionogenic tenside in BGE on 7 × 10-5 mol L-1 and increased in the sample on 5 × 10-4 mol L-1, see Figure 4E. Into the basic mixture of the conidia from the previous experiments, Figure 4B-D, was added the same number of the conidia of A. fumigatus. Four peaks were detected; the residue of PB-PEG in the BGE was very diminished. At the same separation conditions the number of conidia in 1 mL from the strains examined in Figure 4E was decreased to 104, the number of the newly added conidia of P. chrysogenum was 105 in 1 mL, see Figure 4F. The broad peak of the conidia of P. chrysogenum was detected; the peak area of the rest of the conidia from other strains was decreased. The migration time of the conidia of A. niger was increased through low concentration of PB-PEG in the sample with respect to the total number of the conidia in the sample pulse. Very sensitive CZE separation and detection of the approximately down to 10 conidia of all five examined strains, 104 conidia/mL, is depicted in Figure 4G. The composition of the sample pulse and BGE electrolytes was the same as in Figure 4F, only 0.3% (w/v) PEG 1000 was added in BGE. Therefore, the migration times of the conidia zones were increased, and the time for the dynamic coating of the conidia

Figure 5. CIEF of fluorescent pI markers and the fungal conidia modified by PB-PEG. Conditions and designations, see Figures 1, 2, and 4; the catholyte, 4 × 10-2 mol L-1 sodium hydroxide, and the anolyte, 0.1 mol L-1 H3PO4, were dissolved in 7 × 10-5 mol L-1 PBPEG, 3% (v/v) EtOH, and 0.3% (w/v) PEG 10 000; sample composition, 104 conidia/mL of each of A. flavus, F. solani, P. chrysogenum, and A. niger, were dissolved in 15 × 10-3 mol L-1 PSS, 5 × 10-4 mol L-1 PB-PEG, 20% (v/v) EtOH, ∆h, 200 mm, tinj ) 10 s; fluorescent pI markers pI 3.0, 4.0, and 5.5 were added to the segment of the carriers.

was extended. Narrow peaks of all conidia of the examined strains were detected. CIEF of the Conidia Dynamically Modified by PB-PEG with the Fluorometric Detection. The optimized concentrations of PB-PEG at CZE, 7 × 10-5 mol L-1 in the BGEs and 5 × 10-4 mol L-1 in the sample, were used as the starting conditions at CIEF, see Figure 5. The nonionogenic tenside was dissolved together with 3% (v/v) EtOH and 0.3% (w/v) PEG 10 000 in the catholyte, 4 × 10-2 mol L-1 NaOH, and in the anolyte, 0.1 mol L-1 H3PO4. The sample pulse except PBPEG was composed of 15 × 10-3 mol L-1 PSS, 20% (v/v) EtOH, and A. niger CCM 8222, P. chrysogenum CCM F-362, F. solani CCM 8014, and A. flavus CCM F-449, 104 conidia/ mL. In the preliminary experiments the fluorescent pI markers 1.8, 3.0, 4.0, 4.7, 5.5, and 6.6 were added in the segment of the carrier ampholytes and used for the tracing of the pH gradient, see the section CIEF of the Conidia with UV Detection. This separation condition appeared to be suitable for the sensitive and reproducible separation and detection down to 10 labeled conidia of FF. The calibration curve for the conidia of A. niger yielded correlation coefficients of 0.99. The values of pIs of conidia were calculated again from the migration times of the pI markers and their isoelectric points. The isoelectric points of the examined conidia dynamically modified by PB-PEG are in good agreement with the values of the native conidia. CONCLUSIONS The possibility to separate of the conidia of filamentous fungi, A. niger, A. fumigatus, P. chrysogenum, F. solani, and A. flavus, by capillary electromigration techniques was verified in this study. The optimized protocols for their separation by CIEF and CZE with UV and fluorometric detection were developed. According to the preliminary experiments the isoelectric points of the conidia of different strains were not host-specific. The isoelectric points of the native conidia and the conidia dynamiAnalytical Chemistry, Vol. 81, No. 10, May 15, 2009

4003

cally modified by PB-PEG are comparable. At the electromigration techniques with fluorometric detection down to 10 conidia injected into the capillary and dynamically modified by the nonionogenic tenside were detected. These capillary techniques appear to be useful for fast detection and identification of FF after their cultivation and isolation from both clinical samples and samples from environment. For the purpose of the microbiological laboratories more measurements of the isoelectric points of the conidia for considerable number of FF strains would be necessary to make.

4004

Analytical Chemistry, Vol. 81, No. 10, May 15, 2009

ACKNOWLEDGMENT This work was supported by the Grant Agency of the Academy of Sciences of the Czech Republic No. IAAX00310701 and by the Institutional research plan AVO Z40310501.

Received for review February 18, 2009. Accepted April 2, 2009. AC900374V