Biomedical Microdevices

, 11:1289

Disposable plastic microreactors for genomic analyses


    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
    • National Nanotechnology Laboratory of INFM-CNRItalian Institute of Technology (IIT)
  • G. Vecchio
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • P. P. Pompa
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • G. Maruccio
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • L. Sanarica
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • A. Della Torre
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • G. De Bellis
    • Institute of Biomedical TechnologiesNational Research Council (ITB-CNR)
  • G. Caramenti
    • Institute of Biomedical TechnologiesNational Research Council (ITB-CNR)
  • C. Consolandi
    • Institute of Biomedical TechnologiesNational Research Council (ITB-CNR)
  • M. Severgnini
    • Institute of Biomedical TechnologiesNational Research Council (ITB-CNR)
  • R. Cingolani
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit
  • R. Rinaldi
    • National Nanotechnology Laboratory of CNR-INFMIIT Research Unit

DOI: 10.1007/s10544-009-9348-y

Cite this article as:
Sabella, S., Vecchio, G., Pompa, P.P. et al. Biomed Microdevices (2009) 11: 1289. doi:10.1007/s10544-009-9348-y


We show the design, development and assessment of disposable, biocompatible, fully plastic microreactors, which are demonstrated to be highly efficient for genomic analyses, such as amplification of DNA, quantitative analyses in real time, multiplex PCR (both in terms of efficiency and selectivity), as compared to conventional laboratory equipment for PCR. The plastic microreactors can easily be coupled to reusable hardware, enabling heating/cooling processes and, in the case of qPCR applications, the real-time detection of the signal from a suitable fluorescent reporter present in the reaction mixture during the analysis. The low cost production of these polymeric microreactors, along with their applicability to a wide range of biochemical targets, may open new perspectives towards practical applications of biochips for point of care diagnostics.


Plastic microreactorsGenomic analysesReal time PCRBiochip



Polymerase Chain Reaction


quantitative PCR






Bovine serum albumin



1 Introduction

In recent decades, semiconductor industry has accomplished considerable progress, due to several technological advancements in the miniaturization of complex integrated circuits, leading to the realization of novel and powerful portable electronic devices, such as mobile phones, computers, and cameras. Similar technological efforts are now being carried out worldwide to develop innovative diagnostic devices for healthcare (the so-called Lab-on-Chip technology). The opportunity to integrate conventional instrumentation for molecular biology in a biochip, performing one or more biochemical assays in short time, may revolutionize the way biological analyses are performed. Small-sized biochips may enable measurements from small volumes of biological samples and, in principle, without the need of expert operators, providing attractive possibilities for the development of new portable point-of-care (POC) medical diagnostic devices, to be exploited out of the well developed world of analytical laboratories (Yager et al. 2006).

In this frame, since the first PCR was carried out on a chip (Kopp et al. 1998), a large proliferation of biodevices with different geometries (single- or multi-well based) (Northrup et al. 1998; Belgrader et al. 2001; Neuzil et al. 2006) and/or equipped with microfluidic systems (Shaikh et al 2005) have been reported. Several materials (e.g., silicon (Weiping et al. 2005), quartz (Jacobson et al. 1995), glass (Giordano et al. 2001), ITO (Suna et al. 2002), etc.) have been exploited for the implementation of such biochips. In this respect, the use of polymers (e.g., PDMS, PMMA) (Duffy et al. 1998; Lee et al. 2004) provides an additional advantage in terms of fabrication of the biodevices, as such materials can be treated by means of low cost and easy processing methods, such as soft-lithography-based approaches (Xia and Whitesides 1998). Several hybrid biochips have also been developed, exploiting microstructured PDMS multi-layers assembled onto a substrate (usually glass) in order to localize holes acting as microchambers for PCR reactions (Xiang et al. 2005, 2007). Biochips may be integrated with heating/cooling elements and also with optical systems (e.g., optical fibers, fluorescence microscopes, etc.) in order to enable real-time measurements of the PCR products. However, several physical and biochemical problems concerning biocompatibility, adhesion, different thermal and/or mechanical behavior of materials in hybrid chips often arise, leading to serious drawbacks such as leaking, air bubble formation, evaporation and inhibition of the biological reactions. In addition, extensive analytical optimizations of the biochemical assays performed onto the disposable plastic chips (namely, in terms the selectivity, efficiency and repeatability) have not yet completely investigated, thus somewhat preventing the realization of reliable and high performance PCR chips.

In this work, we show the design, the physical characterizations and the biochemical assessment of a prototype PCR-Lab-on-Chip for genomic applications. The device is based on a disposable, biocompatible, plastic microreactor, which is demonstrated to be highly suitable for PCR analyses (amplification of DNA, quantitative analyses in real time, multiplex PCR) as probed by comparison with conventional laboratory equipment (i.e., PCR thermal cyclers). The plastic microreactors can be easily coupled to reusable hardware, enabling the PCR heating/cooling processes and, in the case of real-time qPCR applications, the detection of the signal from an appropriate fluorescent reporter present in the reaction mixture during analysis.

2 Experimental

2.1 PDMS microreactors fabrication

The PDMS microreactors were produced by means of a simple soft lithography technique, based on replica molding. PDMS (Liquid Sylgard 184, Dow Corning, Michigan, USA) was thoroughly mixed with curing agent in a 10:1 (w/w) ratio and poured into an appropriate system composed of moulds and punchers conferring different shapes and dimensions to the microreactor (possible volumes ranging from 50 to 1 μl). Depending on the shape of the moulds, it is possible to obtain single- or multiple-chambers systems, suitable for parallel PCR analyses. After curing in an oven at 70°C for 2 h, the PDMS microreactors were peeled off. Very thin PDMS layers, acting as the bottom of the microreactors, were then produced by spin coating. The thickness of the bottom layer was extensively investigated and optimized (it was in the range of few hundreds of micrometers) in order to guarantee a proper thermal cycling of the PCR solution. The efficiency of thermal transfer was experimentally evaluated by using different thermocouples inside the PCR solution in the plastic microreactors and, more importantly, by evaluating the efficiency of different PCR reactions performed in our PDMS microchambers. Afterwards, PDMS microreactors were made hydrophilic by means of oxygen plasma (PICO, Diener electronic GmbH + Co KG, Germany; experimental conditions: 25 W, 100 mTorr for 2 min). Because of the rapid hydrophobic recovery of PDMS, surfaces were immediately blocked with BSA or Si-PEG. In the first treatment, microreactors were immersed in a BSA (0.1%) + SDS (0.1%) solution and incubated for 15 min in oven at 60°C, followed by extensive washing with milliQ water. The Si-PEG passivation was performed by immersing PDMS substrates in polymer solutions (1 mM Si-PEG in water containing 0.8 mL concentrated HCl per liter) for 2 h at room temperature. The BSA-based treatment was found to be extremely effective and stable (the microreactors typically preserved their stability over several months) and was generally selected as the standard passivation method to make PDMS surfaces biocompatible. Interestingly, after passivation, PDMS microreactors can be sterilized by conventional autoclaves without losing any surface feature and their biocompatibility.

2.2 Validation tests of PDMS microreactors and PCR Lab-on-Chip

PDMS microreactors and PCR-Lab-on-Chip were validated by carrying out several PCR in few µL on different DNA targets. In particular, we used the following DNA templates: a) 2,958 bp plasmidic DNA (pBluescript II KS, Stratagene) in which is inserted a 449 bp sequence derived from RUNX1 gene containing a 272 bp target sequence (forward Primer: 5′- CCT ACG CAC TGG CGC TGC AAC AAG - 3′; reverse Primer: 5′- ATG GCT CTG TGG TAG GTG GCG ACT - 3′; PCR mixture: 1X PCR buffer, 0.2 μM of each primer, 200 μM of dNTPs, 2.5 mM of MgCl2, 1 unit of Taq DNA polymerase, 0.02 ng/μl of DNA template; b) bacterial genomic DNA which was extracted from Cyanobacterium Mycrocystis (length 105–106 bp) containing a 1,445 bp target sequence; PCR mixture: 1X Optimized DyNAzyme™ Buffer, 0.5 μM of each primer (forward: 5′-AGAGTTTGATCMTGGCTCAG-3′; reverse: 5′-CCTCGCCTCTGTGTGCCTAGGT-3′), 200 μM of dNTPs mix, 1.5 mM of MgCl2,1 unit of DyNazyme™ II DNA polymerase (Finzymes), 0.25 ng/μl of DNA template; c) human genomic DNA: CDK11 (length ~109 bp) containing the 206 bp target sequence which was amplified by the following PCR master mix: 47 μl or less of MegaMix (Microzone Ltd.), 0.4 μM of primers mix (forward: 5′-AGTGCTTCCCTTGCTTTCAA-3′; reverse: 5′-TGTGCTGTACATTTCCTGTGTG-3′), 1.5 ng/μl of DNA template. All the PCR experiments (250 independent analyses) were contemporaneously performed into vials (used as references) and microreactors inside a thermal cycle (T3000 Biometra, Goettingen Germany), and using PCR Lab-on-Chip (single and multi chamber format) using aliquots of the same master mix. Solutions were analyzed by agarose gel-electrophoresis by using O’GENERULER 1 kb DNA (Fermentas Life Sciences) as marker. Chip validation was carried out by analyzing two analytical parameters, namely specificity and efficiency (band position and PCR yield, respectively). Quantification of DNA, migrated as a band in the gel, was calculated by means of 1D Gel analysis software (UVIProplatinum 1.1).

Real-time PCR measurements were performed by the PCR Lab-on-Chip using different starting amounts of plasmidic DNA template (0.2 ng/μl and 0.002 ng/μl). We prepared the same master mix described above, with the addition of the fluorescent reporter SYBR Green I (λexc = 488 nm, λem = 522 nm). For such experiments a classical right-angle geometry was employed for the optical scheme, in combination with a square-shaped PDMS microreactor. We typically collected the fluorescence signal of SYBR-Green during the annealing phase of each PCR cycle, by using 0.5 s excitation from a low power light sources (e.g., monochromatized Xenon lamps or LEDs) and simultaneous detection by a PMT. In order to assess the yield and specificity of PCR amplification, we carried out an additional end-point detection (after 30 cycles) by means of agarose gel electrophoresis.

2.3 Micro-heaters fabrication

Micro-heaters and temperature sensors were defined in AZ5214E resist on Si/SiO2 substrates using standard photolithographic techniques. The photoresist was spin coated on the substrates at 4000 rpm for 40 s (final thickness of 1.5 μm). After exposure, samples were developed in AZ726MIF for 30 s. Finally, metallization was realized by E-beam evaporation (Ti/Pt, thicknesses 5/100 nm) followed by lift-off in an acetone bath for more than 15 min at moderate temperature (35–40°C). All the temperature sensors were calibrated on a hot plate using a thermocouple as reference, in order to evaluate the linearity of their behavior and determine the sensor temperature coefficient. Typical values for heating/cooling rates were 20 and 6°C/s, respectively.

3 Results and discussion

A schematic picture illustrating the fabrication steps and the optimization of the PDMS microreactors, along with their integration with a reusable micro-heater system to realize a PCR Lab-on-Chip, is reported in Fig. 1. Plastic transparent microreactors were easily and cheaply produced by means of replica molding, which was exploited combining metallic punchers and moulds to realize suitable substrates for replicas onto plastics (Fig. 1, steps I and II). Importantly, in order to facilitate heat diffusion into the biological solution, preventing the PDMS thermal inertia (which is a well known disadvantage of these plastic substrates) (Shin et al. 2003), the thickness of the PDMS layers, acting as the bottom of microreactors, was extensively investigated, finding out optimal values in the range of few hundreds of micrometers. Furthermore, as disposable PDMS microreactors have to be compatible with biological reactions, another aspect we deeply explored was the development and the optimization of procedures aimed at improving the wettability and the biocompatibility of the microchambers. To this purpose, we used physical and chemical treatments of the microreactors which elicited a very good and stable hydrophilicity of the material along with an excellent compatibility towards biomolecules over several months. Hence, mild conditions of O2 plasma (enabling a soft oxidation of the substrates) (Bhattacharya et al. 2005), immediately followed by polymer or protein blocking, were found to be highly efficient and reliable procedures for these substrates (Fig. 1, step III; see also Experimental for details). By changing some fabrication parameters, it is possible to obtain different kinds of microreactors, exhibiting several advantageous features: PDMS microchambers can be, in fact, produced with a wide variety of shapes and dimensions (master mix volumes ranging from conventional 25 µL down to just 1 µL, or even less), and in single- or multi-well formats, allowing for simultaneous PCR experiments onto the same chip (Fig. 1, step III); also, they are low-cost, disposable and suitable for different types of PCR analyses (i.e., amplification reactions of DNA and subsequent end-point gel electrophoresis detection, quantitative analyses in real time, multiplex PCR). Moreover, since PDMS is typically compatible with other plastic materials, adhesive films, such as “Microseal B”, were implemented as cover of the device (Fig. 1, step IV). This fully plastic configuration ensures a tight sealing of the biochip during PCR thermal cycling, strongly preventing liquid evaporation as well as air bubble formation. Analytical validations were carried out by performing PCR reactions into the plastic microreactors and into the conventional vials for thermal cyclers. Results, analyzed by agarose gel-electrophoresis, revealed a very good performance of the microreactors, comparable to standard PCR vials, both in terms of efficiency and selectivity of the reaction (see below). Interestingly, we observed that the PDMS microchambers can be sterilized, even after the passivation procedures, by conventional methods (e.g., by autoclave), as usually required for consumables dedicated to DNA or RNA analyses, without losing any positive surface feature, such as biocompatibility and long-term stability.
Fig. 1

Scheme of the fabrication steps of the PDMS microreactors and their integration with microheating elements to realize a prototype PCR Lab-on-Chip

Our disposable plastic microreactors can be easily combined with reusable micro-heating surfaces to form a fully integrated PCR Lab-on-Chip (Fig. 1, step V). Specifically, we integrated micro-heaters and temperature sensors onto the same silicon chip (see Experimental for details) and developed a home-made software, which allowed us to control all the elements (micro-heaters, sensors, cooling fan) and set-up the thermal parameters of the reactions. The micro-heaters and the temperature sensors were both obtained in a single lithographical step by thermal evaporation of Ti and Pt (5 nm and 100 nm, respectively), providing an efficient thermal cycling of the biological solution (typically, 20°C/sec heating ramp, 6°C/sec cooling). A PID feedback was implemented in the software for accurate control of temperature while cycling, including also an integrated fan control to speed the cooling. In any case, the PDMS microreactors are compatible with a wide range of heating/cooling systems (such as commercially available systems, e.g., Peltier) to perform efficient and very fast PCR reactions.

The plastic PCR Lab-on-Chip was extensively tested to assess its capability to selectively and efficiently amplify some specific DNA targets of increasing biomolecular complexity, i.e., plasmidic, bacterial and human genomic DNA. PCR experiments were performed in vials (control), in the PDMS microreactors inside a standard thermal cycler and into the PCR Lab-on-Chip, by using different aliquots of the same master mix. Results were analyzed by agarose gel-electrophoresis, in terms of two analytical parameters, namely specificity and efficiency of the reaction (250 independent experiments were carried out per each DNA type). Experimental data clearly indicate that disposable PDMS microreactors and PCR Lab-on-Chip show a high reliability, a good PCR yield (similar to conventional vials), as well as a very good specificity in amplifying the expected DNA targets (Fig. 2), even in the case of human genomic DNA (~109 bp sequence), where the risk of aspecific amplifications is considerable. Interestingly, the performances of the plastic microreactors and of the PCR Lab-on-Chip systems were not significantly affected by the molecular complexity of the template DNA (i.e., from ~103 to ~106 to ~109 bp sequence, from plasmidic to bacterial to human genomic DNA, respectively) and/or of the target DNA (i.e. target sequences ranging from ~200 to ~1,600 bp). Similar experiments were performed into the plastic multichamber systems (up to 36 parallel PCR reactions) and analogous results were obtained. In consideration of the high reliability of the plastic devices, these results clearly open up attractive perspectives towards the development of practical diagnostic applications. Our PDMS microreactors were found to be very stable and can be used after several months obtaining the same PCR efficiency, or used repeatedly several times after careful washing with pure water. However, thanks to the very low cost of the microreactors, they are intended as disposable and portable devices and their reuse is not convenient.
Fig. 2

Validation tests of PCR Lab-on-Chip for different PCR analyses of increasing biomolecular complexity: (a) plasmidic DNA, (b) bacterial genomic DNA and (c) human genomic DNA. Experiments were performed in standard vials (control), in PDMS microreactors in thermal cyclers, and by using the PCR Lab-on-Chip. Agarose gel results (lane 1: marker; lane 2: vials; lane 3: microreactors; lane 4: PCR Lab-on-Chip) and histograms reveal that the PDMS microreators and the PCR Lab-on-Chip systems show an efficiency and specificity comparable to the reference commercial instrumentation. 250 independent PCR experiments were carried out (the error bars represent the standard deviation)

Importantly, our PCR Lab-on-Chip offers the additional possibility to perform quantitative PCR analyses in real-time. In principle, thanks to the peculiar features of the optimized microreactors, which are nearly transparent in the whole visible spectral range, it is possible to use a wide variety of fluorescent reporters (such as molecular beacons, FRET probes, etc.). Also, there are virtually no specific constrains in the implementation of the optical set-up, with several possible geometries of the excitation/detection scheme. In this work, as a proof of principle, we used a widely exploited fluorescent reporter, i.e. SYBR-Green I, and a classical right-angle geometry for the optical scheme, combined to a square-shaped PDMS microreactor (a representative photograph of the PCR-Lab-on-Chip, used in “real-time mode”, is reported in Fig. 3, top). In this experiment, we typically collected the fluorescence signal of SYBR-Green during the annealing phase of each PCR cycle, by using 0.5 s excitation/detection times. As shown in Fig. 3(bottom), the fluorescence signals of the reporter increased over time (with a typical sigmoid-like shape), as a function of the increasing concentration of DNA in the solution, due to PCR amplification. The real-time monitoring of such signals, along with a proper set of calibration curves, allows for quantitative measurements of the DNA target starting amount. In particular, the two real-time PCR curves reported in Fig. 3 correspond to two different initial concentrations of DNA (the left curve was obtained with a concentration ca. 100-fold higher than the right curve); it is evident that the two curves reach a similar plateau, but the second curve undergoes a delay (in terms of PCR cycle) of the fluorescence increase. Importantly, consistent with “end-point” PCR data (see above), we observed that, also in the case of real-time measurements, the PCR-Lab-on-Chip ensures both a very good specificity and a high PCR yield, as demonstrated by gel-electrophoresis analysis performed at the end of the reaction (typically after 30–35 PCR cycles, see Fig. 3, bottom-right). Moreover, it is worth noting that the PCR Lab-on-Chip, in “real-time mode”, allows for melting curve analyses, providing the interesting opportunity of genotyping studies.
Fig. 3

(Top) Typical right-angle set-up for real-time quantitative PCR experiments by PCR Lab-on-Chip. (Bottom, left) Real-time PCR curves obtained by using SYBR-Green I as the fluorescent reporter. The left curve was obtained with a starting DNA concentration (~0.2 ng/μl) ca. 100-fold higher than the right curve (~0.002 ng/μl). Fluorescence signals were collected during the annealing phase of each PCR cycle. (Bottom, right) End-point agarose gel detection of DNA target (typically after 30–35 PCR cycles)

4 Conclusions

The PCR Lab-on-Chip allows, therefore, to perform reliable PCR analyses (single or multiplex PCR reactions, real-time quantitative measurements) on a wide range of biochemical targets, regardless of their biomolecular complexity. The plastic microreactors can be further integrated with microarray technology, directly developed onto the PDMS substrates, likely leading to a deeper insight in the characterization of the target DNA sequences (e.g., for detailed genotyping analyses of viruses of significant pathological relevance). Moreover, freeze-dried biological samples and reagents can be prepared into the disposable microreactors in minute amounts, thus providing “ready-to-use” formats that may be easily stored and shipped at ambient temperature, particularly suitable for low-cost point-of-care diagnostic applications.


The authors gratefully acknowledge E. Perrone, E. D’Amone, V. Fiorelli, D. Mangiullo and G. Caredda for the expert technical assistance. This work was supported by the Italian Ministry of Research through MIUR “FIRB” project (RBLA03ER38_001).

Copyright information

© Springer Science+Business Media, LLC 2009