Studies And Applications Of Double-Stranded DNA Microarray Technologies

For the advantages of easy manipulation, cheap, high throughput and sensitivity, DNA microarray has been widely used in the fields of genomics, pharmacology and medical test, etc. The interactions of DNA-binding proteins and DNA-binding drugs with double-stranded DNA (dsDNA) in genome are involved in many important biological functions. Most of diseases are associated with the abnormal interactions between DNAs and DNA binding proteins. Most of the DNA microarray is applied in the nucleic acid detections by immobilized single-stranded oligonucleotides. However, DNA binding proteins mainly recognize sequence specific double-stranded DNAs. So, it is a very significant work to develop several kinds of double-stranded DNA (dsDNA) microarrays for the studying DNA/protein interaction. Furthermore, we have improved the design of immobilized dsDNA probes in the SNP detection. The main works during my Ph.D. study are as follows:1. The manufacture of the dsDNA microarraysIn this part, we have developed and compared several methods to fabricate dsDNA microarray. Method I: Immobilized a single-stranded oligonucleotide and annealed with a short one, then converted into a double-stranded probe by on-chip polymerase elongation; Method II: Immobilized a hairpin structure single-stranded oligonucleotide and converted into a double-stranded probe by on-chip polymerase elongation; Method III: Annealed a long single-stranded oligonucleotide with a hairpin probe to converted a"U"shaped oligonucleotide, and then converted into a double-stranded probe by on-chip polymerase elongation after immobilization; Method IV: Immobilized a whole double-stranded probe with a hairpin structure, directly self-hybridization to convert into a double-stranded DNA probe. We compared the four methods and found that Method II and Method IV are the most effective ways to fabricate dsDNA microarray. However, for the need of various research applications, every method has its own advantages.2. Optimization of dsDNA microarray fabricationIn order to make dsDNA microarray effectively and economically, some preliminary works were performed on the dsDNA microarray fabrication by using Method II. First of all, we have optimized the design principle of the hairpin structure, including the design of the span sequence, the base number of the loop area and the palindrome area in the hairpin. After the optimization, the design principle of the hairpin structure was summarized as follows: (1) Longer span sequence of immobilized probe is favorable for the DNA/protein interaction. For the NF-κB protein, at least six bases should be used. (2) The palindrome area in the hairpin should not shorter than four bases. All of the palindrome area used in our latter work was five bases. (3) In the five bases palindrome area, good elongation result also could be obtained without any loop base. Furthermore, we have displaced the Klenow DNA polymerase into Taq DNA polymerase in the elongation system and optimized the new on-chip reaction conditions. The results were shown that 50℃of reaction temperature and 2.5mM of Mg2+ concentration were optimal for the on-chip extension.3. Investigation of DNA/protein sequence specific interactions by using double-stranded microarrayA series of dsDNA probes with multi operation sites of restriction proteins in the middle sequence were immobilized to investigate DNA-protein sequence-specific interaction including methylation. We arranged EcoR I site and Rsa I site at the same duplex DNA probe to fabricate ds-DNA arrays and then used the dsDNA arrays to study DNA-restriction enzyme reactions before and after duplex DNA methylation under different probe concentrations and reaction time. Our results indicated that the dsDNA microarray can be further biochemically modified and accessible for interactions between DNAs and proteins in complex multi step gene-regulation processes.4. Evaluating the binding affinities of NF-κB P50 homodimer to the wild-type and single-nucleotide mutant binding sites by the dsDNA microarrayWe have investigated the binding affinities of NF-κB p50 homodimer to the wild-type and single-nucleotide mutant sites by the dsDNA microarray which was fabricated with Method III. The importance of each nucleotide consisting of NF-κB site for the sequence-specific p50p50/Ig-κB interaction was thus evaluated. The results demonstrate that the nucleotides at different positions contribute differently to the p50p50/Ig-κB binding interaction. The G1, G2 and C10 are most important for p50p50/Ig-κB binding interaction and determine the specificity of p50p50/Ig-κB interaction, which replacements with any other nucleotide could result in the similarly greatest binding affinity loses. Comparatively, the G3, A4, T8 and C9 are less important for p50p50/Ig-κB interaction and regulate the binding affinity, which substitutions with the variant nucleotide could change the binding affinity differently. The C5 is most unimportant for p50p50/Ig-κB interaction, the randomized nucleotide exchange of which effects little on p50p50/Ig-κB binding affinity. Among all possible single-nucleotide mutants, the T8 to C mutation could strength p50p50/Ig-κB interaction. The T7 acts differently from its symmetric C5 and the axial T6 is necessary for high-affinity p50p50/Ig-κB interaction. The dsDNA microarray provides a reliable method for exploring the binding affinities of DNA-binding proteins with a larger number of DNA targets.5. Evaluating the binding affinities of NF-κB P50 homodimer to the wild-type and single-nucleotide mismatch binding sites by the dsDNA microarrayWe have immobilized a series of double-stranded DNA probes by using Method IV to investigate the binding affinity of NF-κB p50 homodimer to the single-nucleotide mispairs (G?A or T?C) of the 10bp protein binding sites. The results demonstrate that the nucleotides at different positions contribute differently to the p50p50/DNA binding interaction. Within the 10 bp binding sites, the 5tG or 6cA mispair has less effect on the protein-DNA binding affinity. Even the 5tG mispair may have the ability to enhance the protein-DNA interaction (5t/w=1.07). On the other hand, the 7cA or 10tG mispair blocked the protein-DNA interaction more significantly than other six single-nucleotide mispairs. (7c/W=0.37, 10t/W=0.35). At the same time, we have compared the variation of DNA/ NF-κB binding affinities under single-nucleotide mispair and mutation sites. It was found that the binding affinities changes were different in some bases between mispair and mutation. It also has been suggested that this kind of unimolecular dsDNA microarray can be used to analyze the protein-DNA interactions not only in single-nucleotide mutations but also in mispairs.6. NF-κB protein detection methods based on dsDNA microarrayIn this part, we have developed four different methods to detect NF-κB protein by using dsDNA microarray: (1) Detecting NF-κB protein by fluorescent labeling the protein and hybridization to the dsDNA microarray. (2) Detecting NF-κB protein by fluorescent labeling the protein's antibody and hybridization to the dsDNA microarray. (2) Endonuclease-based free labeled method for detecting NF-κB protein. (4) Detecting NF-κB protein by using a novel designed half-site dsDNA microarray. The first two methods should label the proteins. Itwas not suitable for the protein detection, especially for the multi proteins detection. The other two methods did not need to label the protein. From the Fluorescence intensity-Protein concentration curve, it suggested that the two free labeled methods have the potential application in DNA binding protein detection. But for quantitative detection, it should be needed more improvements on dsDNA microarray fabrication and the binding reaction environment of DNA-protein.7. Studies on the nucleic acid hybridization abilities of the immobilized dsDNA probes We have compared the nucleic acid hybridization abilities of three kinds of immobilized probes with different structures. Under the conventional structure design, we found that the share-stem hairpin structure probe (SHP) was more easily hybridized with the target than the linear probe and conventional hairpin shaped probe (HP). The HP probe had the lowest hybridization ability. However, it was shown a contrary result in the single-nucleotide mismatch discrimination ratio. Subsequently, we increased the Tm of the two hairpin-shaped probes and found both of the two probes were improved on the hybridization specificities even though the hybridization abilities were decreased. The result of the hybridization temperature optimization experiment showed that under the 45℃, the Dr ratio (mismatch discrimination ratio) of HP-28T and SHP-32T were all as high as 8.5 while as the ratios of linear probe were～0.3-0.7. The results of this part suggested that the immobilized hairpin shaped probes also had the ability to discrimination single-nucleotide under proper design.8. Studies on the nucleic acid hybridization abilities of the immobilized dsDNA probesWe have developed a new kind of immobilized dsDNA probe used to discriminate single-nucleotide mismatch. We ulteriorly extended the stem area of the SHP probe until it was hardly to hybridize to its perfect matched target. And then placed all the stem sequence into the hybridization area and arranged the possible mutation nucleotide in the middle hybridization area. In the perfect match SHP probe, we introduced an inner mismatch at the complementary position of the middle base. After the introduction of the inner mismatch, the hybridization ability of the SHP probe with a long stem was enhanced significantly. On the other hand, the mismatch probe was not able to hybridize to the perfect matched target. After the hybridization experiment under different temperature and Mg2+ concentration, it was demonstrated that this kind of probe pair had an excellent discrimination ratio for single nucleotide mismatches. We hypothesized that such a dsDNA microarray would find practical applications in high-throughput mutation analysis and disease diagnosis.