Miniaturized Spotter-Compatible Multicapillary Stamping Tool for Microarray Printing
2007; Future Science Ltd; Volume: 42; Issue: 5 Linguagem: Inglês
10.2144/000112535
ISSN1940-9818
AutoresAlexei L. Drobyshev, Nikolai N Verkhodanov, А. С. Заседателев,
Tópico(s)Silk-based biomaterials and applications
ResumoBioTechniquesVol. 42, No. 5 BenchmarksOpen AccessMiniaturized spotter-compatible multicapillary stamping tool for microarray printingAlexei L. Drobyshev, Nikolai N. Verkhodanov & Alexander S. ZasedatelevAlexei L. Drobyshev*Address correspondence to Alexei L. Drobyshev, Engelhardt Institute of Molecular Biology, Vavilov str. 32, Moscow, 119991, Russian Federation. e-mail: E-mail Address: dro@newmail.ruEngelhardt Institute of Molecular Biology, Moscow, Nikolai N. VerkhodanovEngelhardt Institute of Molecular Biology, Moscow & Alexander S. ZasedatelevEngelhardt Institute of Molecular Biology, MoscowMoscow Institute of Physics and Technology, Dolgoprudny, Russian FederationPublished Online:28 Jun 2018https://doi.org/10.2144/000112535AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinkedInRedditEmail Microarray technology is the most successful example of miniaturization in modern life science. Taking advantage of high parallelism resulting from miniaturization microarrays has become a tool of choice in numerous applications (1). However in contrast to microarrays themselves, the methods of their manufacturing are generally completely macroscopic.Currently, most microarrays are produced by pin or piezo-jet technology. Both use 384-well (seldom 96-well) microtiter plates filled with the solutions to be spotted. Pins or piezo dispensers are placed in the holder so that they match the pattern of the plate (i.e., in the case of a 384-well plate, they have a 4.5-mm pitch). This imposes some restrictions on the shape, size, and density of arrays produced in multipin mode. These arrays consist of clusters of spots matching the pattern of the microtiter plate they were printed from. For example, no microarray smaller than 4.5 × 4.5 mm can be printed in multipin mode. On the other hand, in a highly parallel mode with 48 (typically 4 × 12) pins, the resultant array will cover the entire 25 × 75 mm microscopic slide, which is unacceptable for 100-element arrays (i.e., these arrays are currently produced in low-throughput single pin mode).A few exceptions from this general trend are based on parallel depositing of presynthesized species by contact (2–5) or noncontact (6–9) method. Both exploit planar microcapillary chips with an array of spotting or shooting nozzles. Although these methods provide a high-throughput for low-complexity arrays, they require special machinery and are not easily compatible with existing microarray spotters available in many biological laboratories.Here we propose a spotter-compatible microstamping tool capable of printing up to 127 spots within a 3 × 3 mm are in single touch. We use it for production of a 70-element hydrogel-based microarray for identification of drug-resistant strains of Mycobacterium tuberculosis (10).The key element of our printing tool is a glass multicapillary funnel with 127 microchannels forming a hexagonal array (Figure 1A). The funnel was manufactured at an industrial environment [Technology and Equipment for Glass Structures (TEGS), Saratov, Russian Federation] by forming a hexagonal array of glass tubes and baking them together. Then this multichannel structure was drawn in a special oven to obtain the funnel of desired shape. This technology was used earlier, for example, for the manufacturing of polycapillary lenses (11).Figure 1. Multicapillary printing tool and spotted arrays.(A) General view of assembly: a glass funnel with 127 microchannels, 70 of them have inserted polypropylene capillaries filled with solution to be spotted, the remaining 57 channels are not used for spotting. The upper clamp is normally tightly fixed at the broad end of funnel blocking the capillaries from moving out in an upward direction. (B) Front view of narrow (printing) end of funnel with 70 channels having inserted capillaries and empty channels. (C) Side view of narrow end of funnel with 70 polypropylene capillaries protruding out of channels. (B and C) The arrows point to the same channels. (D) Microphotograph of 70-element microarrays of 4 M betaine droplets printed with a 70-capillary printing tool in a single touch in quadruplicate.At the broad end of the funnel, the diameter of the microchannels is 1 mm with a 1.3-mm pitch. At the narrow end of the funnel, the diameter is 0.21 mm with a 0.27-mm pitch, and the funnel's length is 64.5 mm. These microchannels serve as guides for polypropylene capillaries used for printing, (Figure 1, B and C). They were drawn manually from 1-mL insulin syringes (Becton Dickinson, Franklin Lakes, NJ, USA). Every capillary was inserted into the funnel and adjusted so that its outer diameter at the narrow end of funnel (monitored by means of optical microscope) was from 0.15 to 0.18 mm and then trimmed at this position by a sharp blade (Stanley Works, Sheffield, England). The blade sharpness is essential, as it is a major factor affecting the quality of the printing end of the capillary, so the blades should be replaced as soon as necessary. At the opposite end, every capillary was trimmed to achieve the necessary length (from 65 to 66 mm, depending on its position in the funnel). Then the capillaries were filled with spotting solutions and inserted into the microchannels again. At the broad end of funnel, the capillaries were blocked from moving out by an upper clamp (Figure 1A), so that they protrude from the narrow end of funnel by approximately 0.5 mm. The upper clamp is normally tightly attached to the funnel and removed only to access the capillaries. This makes the capillary working like a spring: when it touches the surface of substrate with its protruding end, it shrinks, exerting some pressure and forming the tight contact necessary to transfer liquid to the surface.The entire assembly was placed into a QArray spotter (Genetix, Hampshire, England) instead of a standard pin tool by means of a specially designed holder (see Supplementary Figure S1 available online at www.BioTechniques.com). In the course of printing, this stamp was brought in contact with slides, so that it stamped the pattern formed by the array of microcapillaries (Figure 1D).To test the efficiency of the stamped arrays in mutation screening by hybridization analysis, the funnel was loaded with 70 capillaries filled with hydrogelbased spotting solution with oligonucleotide probes for M. tuberculosis multidrug resistance (TB-Biochip-MDR set; Biochip-IMB, Moscow, Russian Federation) (10) and 10 μM Texas Red™ fluorescent dye. Then the funnel was loaded into the spotter, and a 70-element array was stamped to 10 slides in triplicate with 900 ms contact time. After spotting, the tool was stored for 10 days at 4°C and successfully used for another spotting again (as demonstrated below).Spotted slides were polymerized (12), and Texas Red fluorescence images were acquired and quantified with ImageWare software (Biochip-IMB). Relative standard deviation of integral signals from 70 elements was from 14% to 23% for different slides.Then the microarrays were hybridized with wild-type (H37Rv strain) and mutant DNA target (prepared as described earlier, see Reference 10, and Supplementary Figure S2, A and B, respectively). Every microarray contained two replicates hybridized in the same volume. The fluorescent images were acquired and automatically analyzed for mutations with ImageWare. Wild-type and mutant strains were identified correctly for both replicates in agreement with earlier direct sequencing and hybridization analysis (10). Separately, the same pair of hybridizations was performed with the microarrays stamped after 10 days storage of the tool at 4°C. Both wild-type and mutant DNA targets were successfully identified again (not shown). This proves that our stamping tool provides sufficient printing accuracy for correct detection of single nucleotide substitutions in M. tuberculosis DNA even without internal control like dual-color hybridization.To assess the effect of 10 days storage of the tool quantitatively, the wild-type hybridization signals of one array were plotted against that of another array stamped in the same batch and against an array stamped after 10 days storage of the tool (Figure 2A, solid triangles and open squares, respectively). The correlation coefficients for the data before and after storage were 0.94 and 0.95, respectively. To assess the source of this variability, a correlation plot of two replicates hybridized in the same volume was made (Figure 2B). The correlation coefficient was 0.993, indicating that the error coming from stamping is much smaller than the error from individual hybridizations or 10 days storage of the loaded tool at 4°C.Figure 2. Correlations of hybridization signals of wild-type DNA target in different experiments.(A) Correlation of hybridization signals for microarrays hybridized separately. Solid triangles, microarrays from the same batch; open squares, microarray from the first batch versus microarrays from the batch stamped after 10 days storage of the tool at 4°C. (B) Correlation of hybridization signals of two microarray replicates stamped on the same slide and hybridized in the same volume. For both plots, fluorescent background was subtracted from the mean fluorescent intensities of the microarray features.As the data from Figure 2B demonstrates a good reproducibility of droplets spotted by the same capillary, it was assessed in a special experiment. The funnel was loaded with only one capillary containing hydrogel-based spotting solution with a Cy™5-labeled oligonucleotide marker from the TBBiochip-MDR set and placed into the spotter. An array of 1600 droplets was spotted on a single slide with this capillary and polymerized (12); then a Cy5 fluorescence image was acquired and quantified. Relative standard deviation of integral signals for randomly selected massive of 70 (7 × 10) spots was 6%.Next, the maximal number of replicates that could be stamped without refilling the capillaries was assessed. Obviously it depends on the initial loading of the capillaries and the average volume of spots. The latter depends on the capillary diameter and the wetting of substrate by spotting solution. To study this second factor, only one capillary was left in the funnel, and three different solutions were used for spotting: (i) the solution for hydrogelbased microarrays (12) (50% glycerol, 4.75% methacrylamide, 0.25% methylenebisacrylamide, and Cy5-labeled marker oligonucleotide with an amino group from the TB-Biochip-MDR set in 0.01 M sodium-borate buffer, pH 10.5); (ii) 3 M betaine with 25% glycerol; and (iii) 4 M betaine. Initial loading of the capillary was adjusted to be maximal with the resulting spots' diameter <0.27 mm (the pitch of the printing end of the funnel). Then the funnel was placed into the spotter, and a microarray with a 0.27-mm pitch was printed at 900 ms contact time on different slides from different batches until the exhaustion of the solution. For hydrogel-based spotting solution, four maximal achieved numbers of spots were 4150, 4050, 2850, and 2650 (the average is 3425); for 25% glycerol and 3 M betaine, these numbers were 2500, 2400, 1650, and 1250 (the average is 1950); and for 4 M betaine, these numbers were 3400, 1850, 1200, and 1100 (the average is 1888). The average maximal number of spots for hydrogel-based solution is significantly higher than this number for 3 M betaine with 25% glycerol and for 4 M betaine (P values are 0.025 and 0.059, respectively), in agreement with the suggestion that the transferred volume decreases with the decrease of wetting of the substrate with spotting solution. Nevertheless, several hundreds of replicates are confidently achievable for any of the tested solutions (e.g., for N = 500, P values are 0.005, 0.017, and 0.079, respectively).The proposed method of microarray printing combines simplicity and flexibility because of the ability to add, remove, or replace capillaries in the funnel if necessary. This ability could be used to achieve the desired feature size distribution across the array. If desired, better spot size uniformity could be also achieved by imposing more stringent restrictions on the diameters of capillaries and adjusting the proper speed of tool lifting after the stamping, which is impossible in the current QArray interface. The stamping array could be easily expanded beyond the current 127 elements. Since the funnel is a three-dimentional structure, this expansion will not require any change of interconnection layout necessary for planar chip-based microstamping tools.After spotting, the tool could be stored and used again without recharging the capillaries. In this case, it is important that the solution does not dry or crystallize [e.g., because of the presence of betaine (13) or glycerol (12)]. With the QArray spotter, it takes no more than 3 s/stamping of up to 127 features; no washing, drying, or refilling are required, making our multicapillary funnel stamp a powerful tool for the manufacturing of small-or middle-size microarrays demanded for examples in genotyping (14–16) and the microRNA profiling (17,18) application.Larger arrays could be produced by combining two approaches: (i) increasing the number of channels in a funnel and (ii) replacing the used funnels in a spotter. For example, with current stamping speed and 384-channel funnel, a batch of one-hundred 4000-element arrays could be produced by manual replacing of 11 different funnels in approximately 1 h. Automatic funnel replacement will also improve the potential of this tool for high complexity microarrays production.AcknowledgmentsThis work was partially supported by International Science and Technology Center (ISTC) grant no. 2906. We are grateful to V.M. Mikhailovich, N.V. Zakharova, O.A. Zasedateleva, E.N. Timofeyev, A.V. Chudinov, and S.V. Pan'kov, for helpful discussions, to O.G. Somova and O.V. Moiseyeva for post-spotting microarray processing, and to O.V. Markova, A.Y. Kozlova, E.E. Fesenko, and O.V. Antonova for the microarray hybridizations.Competing Interests StatementThe authors declare no competing interests.References1. Schena, M. 2000. Microarray Biochip Technology. Eaton Publishing, Natick, MA.Google Scholar2. Lin, S.C., F.G. Tseng, H.M. Huang, C.Y. Huang, and C.C. Chieng. 2001. Microsized 2D protein arrays immobilized by microstamps and micro-wells for disease diagnosis and drug screening. Fresenius J. Anal. Chem. 371:202–208.Crossref, Medline, CAS, Google Scholar3. Tseng, F.G., S.C. Lin, H.M. Huang, C.Y. Huang, and C.C. Chieng. 2002. Protein micro arrays immobilized by mu-stamps and -protein wells on PhastGel(R) pad. Sens. Actuators B Chem. 83:22–29.Crossref, CAS, Google Scholar4. Lin, S.C., C.E. Ho, F.G. Tseng, Y.C. Tsai, H.M. Huang, and C.C. Chieng. 2003. A novel protein micro stamper with back-filling reservoir for simultaneous immobilization of large protein arrays, p. 299–302. In Micro Electro Mechanical Systems. IEEE The Sixteenth Annual International, Kyoto.Google Scholar5. Lin, S.C., F.G. Tseng, H.M. Huang, Y.F. Chen, Y.C. Tsai, C.E. Ho, and C.C. Chieng. 2004. Simultaneous immobilization of protein microarrays by a micro stamper with backfilling reservoir. Sens. Actuators B Chem. 99:174–185.Crossref, CAS, Google Scholar6. de Heij, B., M. Daub, O. Gutmann, R. Niekrawietz, H. Sandmaier, and R. Zengerle. 2004. Highly parallel dispensing of chemical and biological reagents. Anal. Bioanal. Chem. 378:119–122.Crossref, Medline, CAS, Google Scholar7. Gutmann, O., R. Niekrawietz, R. Kuehlewein, C.P. Steinert, S. Reinbold, B. De Heij, M. Daub, and R. Zengerle. 2004. Non-contact production of oligonucleotide microarrays using the highly integrated TopSpot nanoliter dispenser. Analyst 129:835–840.Crossref, Medline, CAS, Google Scholar8. Gutmann, O., R. Kuehlewein, S. Reinbold, R. Niekrawietz, C.P. Steinert, B. de Heij, R. Zengerle, and M. Daub. 2005. Fast and reliable protein microarray production by a new drop-in-drop technique. Lab Chip 5:675–681.Crossref, Medline, CAS, Google Scholar9. Gutmann, O., R. Kuehlewein, S. Reinbold, R. Niekrawietz, C.P. Steinert, B. de Heij, R. Zengerle, and M. Daub. 2004. A highly parallel nanoliter dispenser for microarray fabrication. Biomed. Microdevices 6:131–137.Crossref, Medline, CAS, Google Scholar10. Gryadunov, D., V. Mikhailovich, S. Lapa, N. Roudinskii, M. Donnikov, S. Pan'kov, O. Markova, A. Kuz'min, et al.. 2005. Evaluation of hybridisation on oligonucleotide microarrays for analysis of drug-resistant Mycobacterium tuberculosis. Clin. Microbiol. Infect. 11:531–539.Crossref, Medline, CAS, Google Scholar11. Bjeoumikhov, A., N. Langhoff, R. Wedell, V. Beloglazov, N. Lebed'ev, and N. Skibina. 2003. New generation of polycapillary lenses: manufacture and applications. X-Ray Spectrom. 32:172–178.Crossref, CAS, Google Scholar12. Rubina, A.Y., S.V. Pan'kov, E.I. Dementieva, D.N. Pen'kov, A.V. Butygin, V.A. Vasiliskov, A.V. Chudinov, A.L. Mikheikin, et al.. 2004. Hydrogel drop microchips with immobilized DNA: properties and methods for large-scale production. Anal. Biochem. 325:92–106.Crossref, Medline, CAS, Google Scholar13. Diehl, F., S. Grahlmann, M. Beier, and J.D. Hoheisel. 2001. Manufacturing DNA microarrays of high spot homogeneity and reduced background signal. Nucleic Acids Res. 29:E38.Crossref, Medline, CAS, Google Scholar14. Denkin, S., D. Volokhov, V. Chizhikov, and Y. Zhang. 2005. Microarray-based pncA genotyping of pyrazinamide-resistant strains of Mycobacterium tuberculosis. J. Med. Microbiol. 54:1127–1131.Crossref, Medline, CAS, Google Scholar15. Park, H., H. Jang, E. Song, C.L. Chang, M. Lee, S. Jeong, J. Park, B. Kang, and C. Kim. 2005. Detection and genotyping of Mycobacterium species from clinical isolates and specimens by oligonucleotide array. J. Clin. Microbiol. 43:1782–1788.Crossref, Medline, CAS, Google Scholar16. Sergeev, N., D. Volokhov, V. Chizhikov, and A. Rasooly. 2004. Simultaneous analysis of multiple staphylococcal enterotoxin genes by an oligonucleotide microarray assay. J. Clin. Microbiol. 42:2134–2143.Crossref, Medline, CAS, Google Scholar17. Wang, Y., T. Weng, D. Gou, Z. Chen, N.R. Chintagari, and L. Liu. 2007. Identification of rat lung-specific microRNAs by micoRNA microarray: valuable discoveries for the facilitation of lung research. BMC Genomics 8:29.Crossref, Medline, Google Scholar18. Beuvink, I., F.A. Kolb, W. Budach, A. Garnier, J. Lange, F. Natt, U. Dengler, J. Hall, et al.. 2007. A novel microarray approach reveals new tissue-specific signatures of known and predicted mammalian microRNAs. Nucleic Acids Res. Mar 13; [Epub ahead of print].Crossref, Medline, Google ScholarFiguresReferencesRelatedDetailsCited ByPolymerizing immobilization of acrylamide-modified nucleic acids and its applicationBiosensors and Bioelectronics, Vol. 24, No. 7 Vol. 42, No. 5 Follow us on social media for the latest updates Metrics Downloaded 321 times History Received 4 December 2006 Accepted 26 March 2007 Published online 28 June 2018 Published in print May 2007 Information© 2007 Author(s)AcknowledgmentsThis work was partially supported by International Science and Technology Center (ISTC) grant no. 2906. We are grateful to V.M. Mikhailovich, N.V. Zakharova, O.A. Zasedateleva, E.N. Timofeyev, A.V. Chudinov, and S.V. Pan'kov, for helpful discussions, to O.G. Somova and O.V. Moiseyeva for post-spotting microarray processing, and to O.V. Markova, A.Y. Kozlova, E.E. Fesenko, and O.V. Antonova for the microarray hybridizations.Competing Interests StatementThe authors declare no competing interests.PDF download
Referência(s)