Biosensors & Bioelectronics Vol. 12. No. 7, pp. 613~518, 1997
© 1996 Elsevier Science Limited All rights reserved. Printed in Great Britain PII: S0956-5663(96)00080-2
ELSEVIER
0956-5663/97/$17.00
Quartz balance D N A sensor Claudio Nicolini, ",b Victor Erokhin, b Paolo Facci," S. Guerzoni, Andrea Ross" & Pavel Paschkevitsch b qnstitute of Biophysics, University of Genova, Via Giotto 2, 16135 Genova, Italy bELBA Foundation, Via Giotto 2, 16135 Genova, Italy (Received 29 May 1996; accepted 26 November 1996)
Abstract: Single-strand DNA-containing thin films were deposited onto quartz
oscillators by the Langmuir-Blodgett technique towards the realization of a device capable of sensing the presence of the complementary DNA sequences which hybridize with the immobilized ones. DNA, once complexed with aliphatic amines, appears as a monolayer in a single-stranded form by X-ray small angle scattering. A quartz nanobalance is then utilized to monitor mass increment related to specific hybridization with a complementary DNA probe. The crystal quartz nanobalance, capable of high sensitivity, indeed appears capable of obtaining a prototype of a device capable of sensing the occurrence of particular genes or sequences in the sample under investigation. The validity of the nanogravimetric assay was confirmed by independent fluorescence measurements utilizing DAPI and a CCD camera. ©1996 Elsevier Science Limited Keywords: Langmuir-Blodgett films, DNA-hybridization, DNA sensor
INTRODUCTION The commonly used hybridization technique (Kafatos et al., 1979; Martin, 1985) for sequencespecific detection of DNA is sensitive to - 1 0 p g of DNA. However, several factors render hybridization impractical for routine testing for DNA contaminants, since it is labor-insensitive, time consuming, strongly semiquantitative, and it usually requires a radioisotope. In addition, the specificity of the method means that some contaminating DNA may be missed. Several other approaches for detecting nucleic acids have been reported in the literature, based, for example, on a light-addressable potentiometric sensor (LAPS) (Kung et al., 1990), or on acoustic wave devices (Su et al., 1996).
E-mail:
[email protected]
It was indeed shown recently that it is possible to deposit a DNA-aliphatic amine complex onto the solid substrate by the Langmuir-Blodgett (LB) technique (Erokhin et al., 1992). In this method DNA was attached to a preliminary formed monolayer of octadecylamine (or hexadecylamine). The technique is based on the fact that at neutral pH DNA is charged negatively, while amine groups are positive (Frommer et al., 1970). Analysis of the film allowed the suggestion that DNA in such film is in a single-stranded form (Sukhorukov et al., 1993). A schematic model of the film deposited by such a technique is shown in Fig. 1. The single-stranded DNA layer is sandwiched between two aliphatic amine monolayers. Thus, the technique can be useful for our objectives, as it allows one to deposit single-stranded DNA to practically any substrate and does not demand a large quantity of DNA, as only one monolayer will be deposited. Never613
Claudio Nicolini et al.
Biosensors & Bioelectronics
was used as a probe (less than 10% of the random homology) (Pharmacia Biotech, Molecular and Cell Biology Catalogue, (1995, 1996)).
Film deposition
Fig. 1. Model of the elementary unit of the LB film. Single-stranded DNA is sandwiched between two octadecylamine monolayers. Octadecylamine bilayer thickness is about 4.9 nm and DNA layer thickness is about 0.9 nm.
theless, there is the question whether DNA in such a structure will hybridize. In fact, the film contains a single-stranded DNA monolayer between two amine monolayers and it is questionable whether the upper amine monolayer will not prevent hybridization with complementary DNA strands. The aim of the study was to check this possibility. A quartz nanobalance was chosen as a sensitive tool (Sauerbrey, 1964). The device allows one to monitor the mass attached to the surface of quartz oscillators (Facci et al., 1993). The method is simple, cheap and sensitive (as it should be for practical applications) and allows one to make parallel measurements in different media, also allowing a differential scheme of measurements.
MATERIALS AND METHODS The deposited DNA was the plasmid pUC19b5 (3650 bp) extracted and purified from E. coli by Birnboim and Doli (Maniatis et al., 1982). As a probe for the samples the same plasmid, transformed in the single-stranded form by boiling during 2 min, was used (100% of the continuous homology with the target). As a negative control the same plasmid was deposited and DNA of phage 614
The films were formed and transferred onto solid substrate by MDT trough (MDT, Russia) (Erokhin et al., 1994). A small Teflon trough (30ml volume) was used for the film formation. DNA solution (25 mg/ml) was used as a subphase. An octadecylamine monolayer was spread at the subphase surface and compressed to the surface pressure value of 15 mN/m. The adsorption of DNA at the amine head groups was done for 1 h. Then, the layer was compressed to 30 mN/m and the film was transferred onto solid substrates. The films were deposited onto solid substrates by a horizontal lift technique. One layer was deposited for the gravimetric and fluorescence measurements. Twenty layers were deposited for X-ray study.
X-Ray measurements X-ray measurements were carded out on automatic small angle X-ray diffractometer with a linear position-sensitive detector (Mogilevski et al., 1984; Erokhin et al., 1995). Cu K ( = 1.54/~) radiation was used. The sample was rotated with respect to the initial beam, while the intensity was registered in all channels of the detector.
Hybridization For the hybridization samples with deposited DNA were placed in plastic envelopes, containing 2 ml of the hybridization buffer (10mM TrisHC1, pH 7.6, 1 mM EDTA, 0.5% SDS). Twenty milliliters of the boiled probe was added to the same envelope. The envelopes were then sealed and placed into the water thermostat at 60°C with stirring overnight. After the hybridization, the samples was strongly washed with distilled water for 10 min, dried and measured. 'Cold' hybridization was performed at room temperature.
Fluorescence measurements After the hybridization and washing, the samples were stained with DAPI (4',6-diamidino-2phenylindole), which apparently associates in the
Biosensors & Bioelectronics
Quartz balance DNA sensor
minor groove of double-stranded DNA (Kapuscinski, 1990). DAPI from Sigma was used. Binding of DAPI to double-stranded DNA occurs with an about 20-fold fluorescence enhancement, which usually does not occur with single-stranded DNA (Haugland, 1992). Fluorescence microscopy measurements were performed with a Zeiss Axioplan microscope (Zeiss Co., Germany) equipped with a mercury lamp and a 40 x objective. Images were acquired using a CCD camera CH250 (Photometrix Co., Germany) cooled at -40°C by a liquid cooling unit CH260 (Photometrix Co., Germany). Images were acquired with a 5 s exposure time after having fixed the focus. The total intensity of fluorescence was estimated by the calculation of the integral of the gray level histogram. Gravimetric measurements
Calibration of the quartz oscillators was performed by consequent deposition of the cadmium arachidate layers (Facci et al., 1993). The dependence of the frequency shift upon the number of deposited bilayers is shown in Fig. 2. The analysis of the curve allows one to attribute a sensitivity of 1.482 pg/mm2 Hz to our transducer. A special setup has been developed for future applications which allows the measurement of up to eight oscillators in parallel or up to four in differential configuration. Two eight-channel digital multiplexers have been used to address the two ends of each oscil4000
lator. The selection operation consists of connecting the two ends of each oscillators to the driving circuit via the two multiplexers which are driven by three digital lines allowing the selection of the eight channels.
RESULTS AND DISCUSSION In order to check the geometry of the elementary cell containing DNA in the LB film, X-ray diffraction measurements were performed. The diffraction pattern is presented in Fig. 3. The angular position of Bragg reflections corresponds to the spacing of 5.8 nm. Taking into account that the amine used in the experiment was octadecylamine (bilayer thickness is about 4.9 nm), it appears that DNA layers incorporated into the film have a thickness of about 0.9 nm. These X-ray data along with independent IR spectroscopy (Sukhorukov et al., 1994) suggest that DNA is very likely in a single-stranded form in the film. The X-ray data confirm once more the film structure model shown in Fig. 1, where single-stranded DNA layer is sandwiched between two octadecylamine layers. Such a structure of the layer does not allow one to say a priori that hybridization of DNA will be possible, as it is protected by an octadecylamine layer. In order to control this possibility fluorescence measurements were performed. The first indication that hybridization was successful was that after the process the sample surface
T
3500
3000 2500
z 2000 ~. 1500 .=
u.
1000 500 0 0
2
4
6
8
10
number of bilayers
Fig. 2. Calibration curve f o r the quartz transducer, obtained by successive depositions of cadmium arachidate bilayers. 615
Claudio Nicolini et al.
Biosensors & Bioelectronics DNA - octodecylamine complex D = 5.7 nm (pedod of pure octodecylamine 4.9 nm)
700 600 500 >,, •~m C @
400 300 200 ;
iE
100 +
Ol . . . .
I
0
0.5
1
1.5
2
2.5
3
3.5
4
2®[°1 Fig. 3. X-ray diffraction pattern of the LB film, containing 20 periods." octadecylamine-DNA-octadecylamine. became wettable, while before it and after 'cold' hybridization it was not wettable at all. The results of the fluorescence measurements are summarized in Table 1. The results of the specific hybridization are three times greater with respect to unspecific hybridization and one order of magnitude greater with respect to 'cold' hybridization. Thus, it seems that during normal hybridization (100% homology) some structural changes and redistribution of the layer takes place. As a result, DNA becomes available for the specific reaction. Such a model also explains why the fluorescence level after unspecific hybridization (10% homology) is higher with respect to 'cold' hybridization. As the molecules have some mobility when the film is 'warmed', some DNA from the film can be hybridized on itself, while during 'cold' hybridization it is impossible. The results of the gravimetric measurements of TABLE 1
the hybridization are presented in Table 1 (see also Fig. 4). As in the case of fluorescence, normal hybridization results in a frequency shift much higher with respect to the unspecific and 'cold' hybridization. Non-specific hybridization (10% homology) gives a frequency shift higher then 'cold' hybridization, which is also consistent with the data of fluorescence. It is interesting to note that the error (standard deviation of the measurements of 10 different samples) in a case of unspecific hybridization (difference in frequency shift of different samples) is much higher with respect to specific hybridization (both in 'cold' and 'warm' conditions). This fact indicates that in a case of unspecific hybridization some random processes take place, as suggested earlier.
Results of the fluorescent and gravimetric measurements of the DNA hybridization against a reference containing only the LB immobilized single stranded DNA/amine sandwich
Type of experiment 100% homology hybridization at 60°C Cold 100% homology hybridization at 22°C Random 10% homology hybridization at 60°C
Integral intensity of the fluorescence respect to control (a.u.) 31.2 + 9.3 (100%) 3.00 + 0.18 (9.6%) 10.l + 4.8 (32.4%)
Frequency shift with respect to control (Hz) 688.5 + 10.8 (100%) (1.02 + 0.016 ng/mm 2 Hz) 26.3 + 17.5 (3.8%) (0.038 + 0.026 ng/mm 2 Hz) 297.5 + 70.0 (43-2%) (0-44 + 0.104 ng/mm 2 Hz)
The fluorescence data were obtained summarizing the intensity through total image area. The results are the average of two different samples prepared in the same way. Gravimetric data are the averaging of measurements on five samples. 616
Biosensors & Bioelectronics
Quartz balance DNA sensor NANOGRAVIMETRIC MEASUREMENTS HYBRIDIZATION
700 T N
,
600
"lq)-,
OF T H E D N A
sample 50o 400
>. o e-
300
...1 200 o"
I "
control cold hybri~zation
100 0
"
I
1
2
3
Fig. 4. Results of the gravimetric study of the film hybridization. Sample." normal specific hybridization; control." normal unspecific hybridization; cold hybridization." specific hybridization at room temperature.
CONCLUSIONS DNA-based biosensors are considered as one of the most promising elements in analytical biotechnology (Downs et al., 1987; Fawcett et al., 1988; Yevdokimov et al., 1992). Construction of the DNA chip (Nicolini, 1996) demands a single helix DNA immobilized at the surface of the device, sensitive to the binding of the complementary strand. This study has shown the possibility of using single-stranded DNA incorporated into a LB film as a sensitive layer of DNA-based biosensor. The nanogravimetric balance turned out to be a suitable candidate for a transducer of such a sensor as it provides 0.3 ng resolution. The present study represents the first positive step towards multiple DNA probe sensors with high sensitivity, based on LB films and nanogravimetric balance, as it establishes its feasiblity. Future development of this research will be in different directions, namely, the simultaneus detection in samples of several genes utilizing the corresponding DNA probes, the automation of genome sequencing and of course the optimization of the overall mechanics and hardware of the apparatus, until it becomes competitive with the existing hazardous radioactive labelling techniques also in terms of time, cost and efficiency.
ACKNOWLEDGEMENTS The work was supported by the Elba Foundation and the Polo Nazionale Bioelettronica.
REFERENCES Downs, M. E. A., Kobayashi, S. and Karube, J. (1987) New DNA technology and DNA biosensor. Analyr Letr 20, 1897-1927. Erokhin, V., Facci, P. and Nicolini, C. (1995) Twodimensional order and protein thermal stability: high temperature preservation of structure and function. Biosensors & Bioelectronics 10, 24-34. Erokhin, V., Popov, B., Samori, B. and Yakovlev, A. (1992) Immobilization of DNA fragments by Langmuir-Blodgett technique. Mol. Crysr Liq. Crysr 215, 213-220. Erokhin, V., Vakula, S. and Nicolini, C. (1994) Reversed micellar approach as a new tool for the formation and structural studies of protein Langmuir-Blodgett films. Thin Solid Films 238, 88-94. Facci, P., Erokhin, V. and Nicolini, C. (1993) Nanogravimetric gauge for surface density measurements and deposition analysis of Langmuir-Blodgett films. Thin Solid Films 230, 86-89. Fawcett, N. C., Evans, J., Chien, L. and Flowers, N. (1988) Nucleic acids hybridization detected by piezoelectric resonance. Analyt. Lett. 21, 10991114. 617
Claudio Nicolini et al. Frommer, M. A., Miller, I. R. and Kha'/at, A. (1970) Interaction of DNA with positively charged monolayers. Adv. Exp. Med. Biol. 7, 119-133. Haugland, R. P. (1992). Handbook of Fluorescent Probes and Research Chemicals. Molecular Probes, Inc., USA, 222 pp. Kafatos, F. C., Jones, C. W. and Efstratiadis, A. (1979) Nucleic Acid Res. 7, 1541-1552. Kapuscinski, J. (1990) Interactions of nucleic acids with fluorescent dyes: spectral properties of condensed complexes. J. Histochem. Cytochem. 38, 13231329. Kung, V. T., Panfili, P. R., Sheldon, E. L., King, R. S., Nagainis, P. A., Gomez, B., Ross, D. A., Briggs, J. and Zuk, R. (1990) Picogram quantification of total DNA using DNA-binding proteins in a silicon sensor-base system. Analyt. Biochem. 187, 220-227. Maniatis, T., Fritsch, E. E. and Sambrook, J. (1982). Molecular Cloning. Cold Spring Harbor Laboratory. Martin, F. H. (1985). Abnormal Cells: New Products & Risk, ed. H. E. Hopps & J. C. Petricciani. Tissue Culture Association, Gaithersburg, MD, pp. 90-93. Mogilevski, L. Yu., Dembo, A. T., Svergun, D. I. and Feigin, L. A. (1984) Small angle X-ray diffractometer with linear position sensitive detector. Crystallography 29, 587-591.
618
Biosensors & Bioelectronics Nicolini, C. (1996). Molecular Manufacturing. World Scientific, Singapore, 258 pp. Pharmacia Biotech, Molecular and Cell Biolgy Catalogue 1995. Pharmacia Biotech, Molecular and Cell Biolgy Catalogue 1996. Sauerbrey, G. Z. (1964) Z Phys. 178, 457-462. Su, H., Chong, S. and Thompson, M. (1996) Interfacial hybridization of RNA homopolymers studied by liquid phase acoustic network analysis. Langmuir 12, 2247-2255. Sukhorukov, B. I., Montrel, M. M., Sukhorukov, G. B. and Shabarchina, L. I. (1994) Optical properties and structure of Langmuir films of complexes on nucleic acids with lipids and synthetic amphiphilic molecules--I. Infrared spectra, structure and hydration of a multilayer Langmuir film of a complex of polyurididylic acid with octadecylamine. Biophysics 39, 273-282. Sukhorukov, G., Erokhin, V. and Tronin, A. (1993) Preparation and Investigation of Langmuir films of nucleic acids and oktadecylamine complexes. Biophysics 38, 243-248. Yevdokimov, Y. M., Skuridin, S. G. and Chernukha, B. A. (1992) Biosensors based on liquid crystal dispersion of double-stranded nucleic acids. Biotechnologiya (Russian) 5, 103-109.
