Using Molecular Beacons To Probe Molecular Interactions between

Anal. Chem. 2000, 72, 3280-3285

Using Molecular Beacons To Probe Molecular Interactions between Lactate Dehydrogenase and Single-Stranded DNA Xiaohong Fang, Jianwei Jeff Li, and Weihong Tan*

Department of Chemistry and UF Brain Institute, University of Florida, Gainesville, Florida 32611

The interactions between two key macromolecular species, nucleic acids and proteins, control many important biological processes. There have been limited effective methodologies to study these interactions in real time. In this work, we have applied a newly developed molecular beacon (MB) DNA probe for the analysis of an enzyme, lactate dehydrogenase (LDH), and for the investigation of its properties of binding with single-stranded DNA. Molecular beacons are single-stranded oligonucleotide probes designed to report the presence of specific complementary nucleic acids by fluorescence detection. The interaction between LDH and MB has resulted in a significant fluorescence signal enhancement, which is used for the elucidation of MB/LDH binding properties. The processes of binding between MB and different isoenzymes of LDH have been studied. The results show that the stoichiometry of LDH-5/MB binding is 1:1, and the binding constant is 1.9 × 10-7 M-1. We have also studied salt effects, binding sites, temperature effects, pH effects, and the binding specificities for different isoenzymes. Our results demonstrate that MB can be effectively used for sensitive protein quantitation and for efficient proteinDNA interaction studies. MB has a signal transduction mechanism built within the molecule and can thus be used for the development of rapid protein assays and for real-time measurements. The investigation of the interactions between two key macromolecular species, nucleic acids and proteins, has been one of the most important areas and currently one of the most rapidly growing fields in contemporary molecular biology and biotechnology.1,2 Nucleic acids and proteins are both informationcontaining molecules due to their specific sequences. In the interaction of these two species, many different functions are carried out. The protein-nucleic acid interactions span a variety of physical conformations as well as varying degrees of specificity. Unfortunately, up to now, our knowledge of protein-nucleic acid interactions has been limited because of the lack of adequate methodologies for real-time detection. The detailed and funda* To whom correspondence should be addressed at the Department of Chemistry. Phone: 352-846-2410. Fax: 352-392-4651. E-mail: [email protected]. (1) Kneale, G. C., Ed. DNA-Protein interactions: principles and protocols; Humana Press: Totowa, NJ, 1994. (2) Vogel, H. J., Ed. Nucleic acid-protein recognition; Academic Press: New York, 1977.

3280 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000

mental nature of these interactions as well as their contribution to the total organization of biological systems is yet to be discovered. Better understanding of these interactions not only will benefit the understanding of many biological mechanisms but also is expected to be of great medical use in designing new drugs and in developing new disease treatment strategies. There has been considerable work in studying the molecular mechanisms of protein-nucleic acid interactions. Many methods have been developed to probe protein-nucleic acid interactions such as DNA footprinting, cross-linking techniques, filter binding assays, gel shift techniques, affinity chromatography, circular dichroism spectroscopy, X-ray crystallography, and fluorescence spectroscopy.1 These techniques and methods can provide information on the molecular basis of the interaction, such as the location of the DNA binding sites, the strength and specificity of binding, and the effects of protein binding on the gross conformation and local structure of DNA. Among these methods, spectroscopic techniques are especially useful in quantitative studies of protein-DNA interactions. Fluorescence spectroscopy is one of the most sensitive spectroscopic techniques, requiring low concentrations (typically in the micromolar or lower concentration range) for the study of many protein-DNA interactions. The measurements are quick and easy to perform. The fluorescence method used in protein-DNA studies usually takes advantage of the quenching of the protein’s intrinsic fluorescence arising from its tryptophan and/or tyrosine residues.3-5 When the residues are bound to the nucleic acid, the fluorescence of these aromatic amino acids is perturbed by direct interaction with the DNA or by subtle binding-induced conformational changes in protein. The quenching efficiency ranges from 0% to 80%, depending on the structure of the protein. It is known that a measurement of a reduced signal is often more difficult in determining trace amounts of analytes. In some cases, the intrinsic fluorescence of the protein cannot be used to monitor DNA binding because the changes may not be detectable. For example, if the tryptrophan and/or tyrosine residues are not located in the proximity of the DNA binding site, the spectrum will only exhibit minimum change upon binding. Moreover, if the protein contains a large amount of tryptophan and tyrosine residues unrelated to binding, the fluorescence intensity will not change significantly. To overcome (3) Kelly, R. C.; Jensen, D. E.; von Hippel, P. H. J. Biol. Chem. 1976, 251, 7240-7250. (4) McGhee, J. D.; von Hippel, P. H. J. Mol. Biol. 1974, 86, 469-489. (5) Alma, N. C. M.; Harmsen, B. J. M.; de Jong, E. A. M.; Ven, J. V. D.; Hilbers, C. W. J. Mol. Biol. 1983, 163, 47-62. 10.1021/ac991434j CCC: $19.00

© 2000 American Chemical Society Published on Web 06/10/2000

these problems, efforts have been made to explore other fluorescent probes which can offer an alternative approach to measure the protein-DNA binding event.6,7 The most promising one should be an approach where the binding between DNA and protein will result in a fluorescence signal increase. This will greatly improve the detection capability and thus the elucidation of even minute molecular alterations during the binding processes. Molecular beacons (MBs) are newly developed DNA probes with significant advantages over conventional DNA probes.8-11 They have been successfully used as highly sensitive fluorescent DNA probes in detecting specific DNA/RNA molecules.8-10 They emit intense fluorescence upon binding to their target molecules, with fluorescence enhancement as high as 200-fold under optimized conditions.8 Recently, we applied molecular beacons to study DNA-protein interactions and to quantify protein molecules.11 We found that an interaction between an Escherichia coli single-stranded DNA-binding protein (SSB) and a molecular beacon DNA molecule results in a significant fluorescence enhancement.11 The fluorescence enhancement caused by the SSB is comparable to that to the MB’s complementary DNA. This clearly indicates that the protein binding can disturb the conformation of the MB and therefore change the efficiency of fluorescence resonance energy transfer between the donor and the quencher of the MB. The study extended the applications of molecular beacons from pure nucleic acid hybridizations to nucleic acid-protein interactions. It also showed the potential of using an MB as a convenient and highly sensitive probe to quantify proteins in real time. It is generally known that proteins having an affinity for DNA are related to nucleic acid metabolism and control of gene expression. Many proteins that bind stoichiometrically to ssDNA have been isolated from a variety of prokaryotic and eukaryotic sources and have been well studied.12-13 Recently, a rat liver ssDNA-binding protein that had previously been proposed to play a role in transcription was identified as M chain lactate dehydrogenase.14 Lactate dehydrogenase (LDH) is a common intracellular enzyme in the glycolytic cycle that catalyzes the reversible interconversion of lactate and pyruvic acid.15 It has five isoenzymes, and all the isoenzymes’ activities in serum are critically important in supporting a host of diagnoses. Their binding with ssDNA suggests that LDH isoenzymes play many important biological roles. However, to date, few detailed characteristics of (6) Carpenter, M. L.; Kneale, G. C. In DNA-protein interactions: principles and protocols; Kneale, G. C., Ed.; Humana Press: Totowa, NJ, 1994; pp 313325. (7) Lefebvre, S. D.; Morrical, S. W. J. Mol. Biol. 1997, 272, 312-326. (8) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303-308. Vet, J. A. M.; Majithia, A. R.; Marras, S. A. E.; Tyagi, S.; Dube, S.; Poiesz, B. J.; Kramer, F. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6394-6399. (9) Fang, X.; Liu, X.; Schuster, S.; Tan, W. J. Am. Chem. Soc. 1999, 121, 29212922. Tan, W.; Fang, X.; Li, J.; Liu, X. Chem.sEur. J. 2000, 6, 1107-1111. (10) Sokol, D. L.; Zhang, X. L.; Lu, P. Z.; Gewitz, A. M. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11538. (11) Li, J.; Fang, X.; Schuster, S.; Tan, W. Angew. Chem., Int. Ed. Engl. 2000, 39, 1049-1052. (12) Kowalczykowski, S. C.; Bear, D. G.; von Hippel, P. H. In The Enzymes; Boyer, P., Ed.; Academic Press: New York, 1981; Vol. 14a, pp 373-444. (13) Williams, K. R.; Konigsberg, W. H. In Gene Amplification and Analysis Chirikjian, J. G., Papas, T. S., Eds.; Elsevier/North-Holland: New York, 1981; Vol. 2, pp 475-508. (14) Williams, K. R.; Reddigari, S.; Patel, G. L. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 5260-5264. (15) Searcy, R. L. Diagnostic Biochemistry; McGraw-Hill: New York, 1969; p 336.

LDH/ssDNA binding are known. Thus, it is of great interest to investigate their ssDNA-binding properties and to understand their physiologic importance. In this work, we present the first attempt at using a novel fluorescent probe to quantify a functional enzyme and to study its DNA-binding properties. The study opens a way to exploit molecular beacons for the quantitative characterization of proteinDNA interaction processes and for the real-time determination of lactate dehydrogenase. Using LDH isoenzymes enables us to elucidate how minor structural changes in proteins play critical roles in protein-DNA binding. EXPERIMENTAL SECTION Different LDH isoenzymes were purchased from Sigma (St. Louis, MO) and then dialyzed in 20 mM Tris/HCl buffer (pH 7.5) at 4 °C for about 24 h before use. Dialysis was carried out using a cellulose membrane (Fisher, MWCO 12 000-14 000). Two molecular beacons were used in this study: molecular beacon 1, 5′-TMR-CCT AGC TCT AAA TCG CTA TGG TCG CGC TAG G-DABCYL-3′, and molecular beacon 2, 5′-TMR-GCA CGT CCA TGC CCA GGA AGG AAC GTG C-DABCYL-3′. They were custom designed and synthesized by the Midland Certified Reagent Co. (Midland, TX). Unless otherwise specified, all experiments were carried out with molecular beacon 1. Glutamic acid monopotassium salt (KGlu) was obtained from Fluka, and all other chemical reagents were from Sigma. Superpurified water was used for the preparation of all solutions. Fluorescence measurements were performed on a SPEX Industries F-112A spectrofluorimeter (PE) with an external circulating water bath (Fisher) for temperature control. All experiments were carried out at room temperature (25 °C) unless otherwise specified. The sample cell was a 100 µL cuvette. The fluorescence of the MB was monitored by exciting the sample at 530 nm and measuring the emission at 580 nm. Both slits for excitation and emission were set at 10 nm. Control experiments were done to eliminate background signals and other systematic factors that might affect the absolute values of fluorescence measurements. Corrections were also made for potential dilution factors in the titration experiments. RESULTS Molecular Beacons as Fluorescent Probes for LDH/ ssDNA Binding Studies. Molecular beacons, as shown in Figure 1A, are single-stranded oligonucleotide probes that possess a stemand-loop structure.8 A fluorophore and a quencher are linked to the two ends of the stem. The stem keeps these two moieties in close proximity to each other, causing the fluorescence of the fluorophore to be quenched by energy transfer. When the MB encounters a target DNA/RNA molecule that is complementary to its loop sequence, the double-stranded hybrid formed between the loop and the target forces the stem apart, leading to the restoration of the fluorescence of the fluorophore. The LDH-5 isoenzyme has been reported to bind to ssDNA but not to dsDNA.14 When interacting with the molecular beacon DNA molecules, LDH-5 binds to the loop sequence of MB. This interaction is strong enough to dissociate the stem duplex, resulting in a conformational rearrangement of the MB and a spatial separation of the fluorophore and quencher. Our results have shown that there is a significant enhancement in the MB’s Analytical Chemistry, Vol. 72, No. 14, July 15, 2000

3281

A

Figure 2. Fluorescence intensity of molecular beacon 1 at different LDH concentrations. The concentration of MB was 0.4 µM. The buffer used was 20 mM Tris/HCl, 0.05 mM MgCl2, pH ) 7.8.

B

Figure 1. (A) The molecular beacon DNA molecule (MB1). (B) Fluorescence intensity time scan of the MB-binding reaction with LDH5. Measurements were taken to ensure that diffusion was not a factor in the reaction rate (15 s was required to fully mix the MB and LDH-5 solutions before the measurements). The concentrations of MB and LDH were 0.5 and 1 µM. The buffer used was 20 mM Tris/HCl, 0.05 mM MgCl2, pH ) 7.8.

fluorescence intensity upon binding with LDH-5. We have done control experiments to confirm that the fluorescence enhancement in MB/LDH solution is due to the separation between the dye and the quencher, not the changes in the dye itself. As shown in Figure 1B, the reaction is so rapid that the fluorescence intensity reaches a plateau within 20 s (with 15 s of the 20 s for mixing). The binding reaction between LDH and the MB is much faster than the hybridization between the cDNA and MB.8 This should provide the basis for the development of rapid protein assays and for real-time protein-DNA interaction studies. 3282 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000

The fluorescence restoration of the MB due to LDH-5 is about 40-50% of that due to cDNA. The lower fluorescence enhancement due to LDH may be related to a smaller conformational change in the MB upon protein binding. The binding between LDH-5 and the MB does not follow the Watson-Crick base-pairing rule. Therefore, the separation between the fluorophore and the quencher is relatively smaller than that for the fully straightened double-stranded DNA formed between the MB and its cDNA. Quantitative Analysis of LDH-5. A molecular beacon was used for the sensitive detection of LDH-5. Figure 2 shows that the MB fluorescence intensity increased as the LDH-5 concentration increased. When 0.4 µM MB was used, there was a linear relationship between the MB fluorescence intensity and the LDH-5 concentration in the low concentration range of LDH-5 ( LDH-5 (from rabbit muscle) > LDH-5 (from bovine muscle). This order is in agreement with that of their enzymatic activity (930 units/mg for LDH-5 from porcine, 912 units/mg for that from rabbit, and 720 units/mg for (19) Holbrook, J. J.; Liljas, A.; Steindel, S. J.; Rossmann, M. G. In The Enzymes; Boyer, P., Ed.; Academic Press: New York, 1970; Vol. 11a, p 229.

3284 Analytical Chemistry, Vol. 72, No. 14, July 15, 2000

that from bovine). This further confirms the similarities in binding capacities and binding sites between coenzymes and ssDNA when they interact with LDH. e. Effects of pH and Temperature. We also investigated the effects of pH and temperature on the binding between LDH-5 and the MB. There is no obvious change in the fluorescence of the LDH/MB complex when the pH of the buffer increases from 5.5 to 8. This indicates that pH is not a deciding factor in this specific binding reaction in this pH range. Temperature has different effects on the binding of the MB to LDH-1 and to LDH-5. The fluorescence intensity of the complex formed between the MB and LDH-5 gradually decreases as the temperature is increased from 10 to 40 °C, while that for LDH-1 increases in the same temperature range. This is due to the different heat stabilities of LDH-5 and LDH-1. It is generally agreed that LDH-5 is much more sensitive to heat and less stable than LDH-1 when the temperature is raised.17 The stability of the MB/ LDH-5 complex thus decreases at higher temperatures. On the other hand, the temperature effect on LDH-1 is minimal. Within a range, a higher temperature favors the formation and the stability of the MB/LDH-1 complex. DISCUSSION It is interesting to note that different isoenzymes have different binding capacities when an MB is used for LDH/DNA binding studies. We speculate that the different binding capabilities of LDH isoenzymes are due to the following two factors. First, it is known that there is a gradual increase in the content of basic amino acids (lysine and arginine) and a regular decrease in the content of acidic amino acids (aspartic and glutamic acid moieties) from LDH-1 to LDH-5.17 Therefore, the more positively charged basic amino acid residues in LDH-5 may favor the binding of negatively charged ssDNA. Second, if LDH binds to ssDNA at about the same site as that for NAD, the different amino acid compositions of LDH isoenzymes in their NAD binding locations will cause their different affinities for ssDNA. NAD has a nucleoside phosphate group similar to those in ssDNA. Previous studies have shown

that the adenosine end of NAD binds to a hydrophobic pocket in LDH.20 In this hydrophobic pocket, the three binding residues (residues Ile (96), Ile (119), and Met (55)) in H4 have shorter amino acid side chains than the corresponding ones in M4 (Val (94), Val (116), and Leu (54)). Therefore, NAD binds with reduced affinity to H4 compared to M4 in its initial binding. This may also be true for the ssDNA, which results in a lower affinity of H4 for an MB. It should be mentioned that LDH-1 actually binds better to its coenzyme NAD than LDH-5. This is because. after the initial binding of the adenosine end of NAD to LDH, one extra hydrogen bond will form between Gln (31) in H4 and the nicotinamide phosphate of NAD, while this hydrogen bond does not exist in M4. The net result is that LDH-1 binds more tightly with NAD. As there is no nicotinamide phosphate group in MB DNA molecules, the binding of an MB with LDH-1 will not follow this path. Therefore, LDH-1 has lower affinity for an MB. We also found that, even though LDH-2 has one M chain and three H chains, it did not show any obvious difference from LDH-1 in binding capacity. We thus believe that LDH binds to ssDNA in the form of a tetramer and only one M chain is not sufficient for strong binding with ssDNA like molecular beacons. When two molecular beacons (MB1 and MB2) with different loop sequences were used to interact with LDH-5, the degrees of MB fluorescence restoration were very similar. This indicates that, like most of the single-stranded DNA-binding proteins, LDH binds to an MB without DNA base specificity. However, our results demonstrate that there is a direct relationship between the restoration of MB fluorescence intensity and the DNA-binding capability of the isoenzymes. Even though nonspecific DNA-binding proteins are (20) Eventoff, W.; Rossmann, M. G.; Taylor, S. S.; Torff, H.-J.; Meyer, H.; Keil, W.; Kiltz, H.-H. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 2677-2681. (21) Benner, S. A.; Trabesinger, N.; Schreiber, D. Adv. Enzyme Regul. 1998, 38, 155-180. Osborne, S. E.; Matsumura, I.; Ellington, A. D. Curr. Opin Chem. Biol. 1997, 1, 5-9.

used here, the fact that different isoenzymes do have different MB-binding efficiencies is very encouraging. CONCLUSION We have applied novel molecular beacon DNA molecules to the study of protein-DNA interactions using fluorescence measurements. The interactions between lactate dehydrogenase isoenzymes and single-stranded DNA molecules have been investigated. The conformational change of an MB upon binding to LDH results in a quick and significant fluorescence enhancement. Using a simple fluorescence titration method, we were able to detect 1 × 10-8 M LDH-5. The binding properties of different LDH isoenzymes with different molecular beacons under various binding conditions have been studied. Useful information has been obtained for understanding the interaction between ssDNA and LDH proteins. Our results demonstrate that MBs are useful probe for protein quantitation and protein-DNA interaction studies. With further developments in MBs and related technologies, it is expected that MBs will be useful intracellular protein recognition probes for studying proteins in different environments, with ultrasensitivity, in real time and in living cells. Currently, we are pursuing a few approaches in this direction, including designing aptamer-based21 molecular beacons for better specificity and using a large array of molecular beacons for binding-pattern recognition. ACKNOWLEDGMENT This work was partially supported by Office of Naval Research Young Investigator Award N00014-98-1-0621, by NSF Career Award CHE-9733650, and by Whitaker Foundation Biomedical Engineering Award. Received for review December 14, 1999. Accepted April 20, 2000. AC991434J

Analytical Chemistry, Vol. 72, No. 14, July 15, 2000

3285