- Email: [email protected]
Defining the peptide nucleic acids (PNA) length requirement for PNA binding-induced transcription and gene expression1
doi:10.1006/jmbi.2001.5109 available online at http://www.idealibrary.com on
J. Mol. Biol. (2001) 313, 933±940
Defining the Peptide Nucleic Acids (PNA) Length Requirement for PNA Binding-induced Transcription and Gene Expression Gan Wang1*, Kai Jing1, R. Balczon2 and Xiaoxin Xu1 1
Institute of Environmental Health Sciences, Wayne State University, 2727 Second Avenue, Room 4000, Detroit MI 48201, USA 2 Department of Cell Biology and Neuroscience, University of South Alabama College of Medicine, Mobile AL 36688, USA
Induction of gene expression has great potential in the treatment of many human diseases. Peptide nucleic acid (PNA) as a novel DNA-binding reagent provides an ideal system to induce gene-speci®c expression. In our recent studies, we have demonstrated that PNA bound to doublestranded DNA targets and, therefore, generated single-stranded D-loops and induced transcription of target genes both in vitro and in vivo. Most importantly, we have demonstrated that treatment of cultured human cells with PNAs led to expression of an endogenous target gene. Therefore, the study of the molecular mechanism of PNA binding-induced gene expression will have great implications for the gene therapy of many human diseases. In the current study, we have investigated the PNA length requirement for PNA binding-induced transcription initiation. Using a series of PNAs with different lengths, we have determined that PNAs with lengths of 16 18 nt induce very high levels of transcription in a HeLa nuclear extract in vitro transcription system. Transfection of the PNA-bound GFP reporter gene plasmid into human normal ®broblast (NF) cells led to a similar result. Gel-mobility shift assays revealed very strong binding af®nities of these PNAs. DNA footprinting analysis further demonstrated the speci®city of PNAs binding to the targets. These results lead to important understanding of the molecular mechanism of transcription initiation and highly valuable information in PNA design, especially for PNA binding-induced, gene-speci®c expression. # 2001 Academic Press
*Corresponding author
Keywords: peptide nucleic acids (PNA); PNA binding-generated D-loop; transcription; transcription initiation; gene-speci®c expression
Introduction Abnormal gene expression has been linked to many human diseases, such as cancer and agerelated diseases. Induction of speci®c gene expression has a tremendous impact on the treatment of these diseases. Several drugs have been developed to induce gene expression.1 ± 4 Unfortunately, none of these drugs could induce genespeci®c expression. Peptide nucleic acids (PNAs) are synthetic oligonucleotides with modi®ed polyamide backbones.5 PNAs bind sequence-speci®cally to DNA and RNA targets to form Watson-Crick double helix Abbreviations used: PNA, peptide nucluic acid; NF, normal ®broplast; GFP, green ¯uorescent protein. E-mail address of the corresponding author: [email protected] 0022-2836/01/050933±8 $35.00/0
structures.5 When bound to a homopurine/homopyrimidine target of double-stranded DNA targets, the PNA molecule replaces one DNA strand by strand displacement to form a [PNA]2/DNA triplex structure, and the displaced DNA strand forms a D-loop structure at the PNA-binding site.6 ± 10 The triplex structure can stabilize the D-loop formed at the PNA binding sites signi®cantly. In our recent studies, PNA binding-generated D-loops have been demonstrated to have the capability of inducing transcription both in vitro and in vivo.11 More importantly, treatment of cultured human K562 erythroleukemia cells with PNAs designed to bind to the g-globin gene 50 ¯anking region resulted in transcription of the endogenous g-globin gene from the PNA target sites.11 This result suggests that a PNA binding-induced target gene expression strategy may provide a novel approach to induce # 2001 Academic Press
934 gene-speci®c expression of endogenous genes, which will have great implication on the treatment of many human diseases. However, the molecular mechanism of PNA binding-induced transcription has not been established. More importantly, the PNA length requirement for ef®cient transcription has not been determined. In this study, we have investigated the PNA length requirement for PNA binding-induced transcription and gene expression. Using reporter gene plasmid constructs with special PNA-binding sequences, the lengths of PNAs required for suf®cient transcription induction have been determined in a HeLa nuclear extract in vitro transcription system. Our results suggest that PNAs with lengths of 16-18nt induce very strong transcription. When the PNA-bound GFP reporter gene plasmid DNA was transfected into normal ®broblast (NF) cells, a similar result was obtained for the PNA length requirement. The gel-mobility shift assay has demonstrated strong binding of PNAs of 14-20 nt in length. DNase I-based footprinting analysis also demonstrated speci®c binding of PNAs to the target sites. All these results suggest that PNAs of 16-18 nt in length can induce strong transcription and target gene expression. The results obtained from this study lead to important understanding of the molecular mechanism of PNA binding-induced gene expression and will be essential in PNA designing, especially for PNA binding-induced target gene expression as a novel approach for gene therapy of many human diseases.
Results Designing and testing of PNAs that binds to a novel PNA-binding site To study the PNA length requirement for PNA binding-mediated transcription, a potential PNA-binding sequence was designed and cloned into promoter reporter plasmid vectors. A 20 bp DNA fragment with a sequence of 50 AGAAGAGA AGGAAGAGAAGA30 was designed and synthesized. This 20 bp DNA sequence was inserted into two promoter reporter vectors: pGL3-Basic, which carries a promoterless luciferase gene (Promega Corp., Madison, WI), and pEGFP-1, which carries a promoterless green ¯uorescent protein (GFP) gene (Clontech, Palo Alto, CA), to obtain the pJK1 and pJK3 plasmids, respectively (Figure 1). This 20 bp DNA fragment contains a homopurine/ homopyrimidine sequence for forming a [PNA]2/ DNA triplex structure at the PNA-binding site. The invert repeat sequence design in the homopurine/ homopyrimidine sequence enables the PNA molecule to bind to the DNA target either as a duplex or as a triplex strand to form the [PNA]2/DNA triplex. A series of PNAs with different lengths (8mer 20mer) were designed and synthesized to bind to the 20 bp homopurine/homopyrimidine target (Table 1). The binding af®nity of these PNAs
PNA Binding-induced Transcription
Figure 1. Structure of plasmids pJK1 and pJK3. The plasmid pJK 1 was constructed by inserting a 20 bp PNA-binding sequence into the MluI-BglII sites of pGL-3 Basic plasmid (Promega, Madison, WI). pJK3 was constructed by inserting the 20 bp PNA-binding sequence into the MluI-BglII site of plasmid pEGFP-1 (Clontech, Palo Alto, CA).
was determined by a gel-mobility shift assay (Figure 2). No detectable binding was observed with the 8mer PNA, even at the highest concentration tested (1 10ÿ5 M) (Figure 2). A weak binding was detected with the 10mer PNA at the concentration of 1 10ÿ5 M. Strong bindings were observed in the PNAs of 12mer or longer. The strongest binding af®nity was detected with the 16mer PNA: clear PNA binding was detected with the PNA concentration as low as 1 10ÿ7 M (Figure 2). No detectable binding was observed for a 15mer control PNA (data not shown). These results suggest that PNAs with length of 12mer or greater may be required for suf®cient binding to double-stranded DNA targets. Determining the levels of transcription of pJK1 plasmid DNA induced by individual PNA in HeLa nuclear extracts PNA binding-mediated transcription was studied in a HeLa nuclear extract in vitro transcription system. The pJK1 plasmid DNA was incubated with individual PNAs to form the [PNA]2/DNA triplex and the single-stranded Dloop structures. Then the PNA-plasmid DNA complex was incubated in a HeLa nuclear extract in vitro transcription system that was supplemented with rNTPs for RNA synthesis. The nascent RNA was isolated from the in vitro transcription reactants. A reverse transcription assay was performed to determine the initiation sites of the transcripts and a 20mer oligonucleotide that was complementary to the luciferase gene 50 mRNA sequence was used as a primer (Figure 3). No speci®c RNA transcripts initiated from the PNA-binding site were detected from pJK1 plasmid DNA alone, pJK1 plasmid DNA treated with the 10mer PNA or pJK1 plasmid
935
PNA Binding-induced Transcription Table 1. Oligonucleotides and PNAs used in the study A. Oligonucleotides used as PNA binding target Oligo 1: 50 AGCTAGTCATGCGATCTCTTCTCTTCCTTCTCTTCTAATGCACGTAACGG30 Oligo 2: 50 CCGTTACGTGCATTAGAAGAGAAGGAAGAGAAGAGATCGCATGACTAGCT30 B. RT PCR primer 50 GCCTTATGCAGTTGCTCTCC30 C. PNAs used in the study 8 mer PNA: 10 mer PNA: 12 mer PNA: 14 mer PNA: 16 mer PNA: 18 mer PNA: 20 mer PNA:
NH2-CTTCCTTC-COOH NH2-TCTTCCTTCT-COOH NH2-CTCTTCCTTCTC-COOH NH2-TCTCTTCCTTCTCT-COOH NH2-TTCTCTTCCTTCTCTT-COOH NH2-CTTCTCTTCCTTCTCTTC-COOH NH2-TCTTCTCTTCCTTCTCTTCT-COOH
D. Control PNA
NH2-TACGGTTATCCACAG-COOH
DNA treated with 12mer PNA in the in vitro transcription reaction (Figure 3, lanes 1, 2, and 7). In contrast, speci®c RNA transcripts initiated from the PNA-binding site were detected when the pJK1 plasmid DNA treated with either 14mer, 16mer, 18mer, or 20mer PNAs were used as the templates (Figure 3, lanes 3 to 6). Phosphorimaging analysis indicated that the highest transcription signal was obtained when the 18mer PNA-treated pJK1 plasmid DNA was used as a substrate (Figure 3, lane 5 versus other lanes). Noticeably, the transcription initiation position shifted to the 30 side of the PNA-bind-
ing site in the DNA template as the length of the bound PNA increased. Determining the PNA length requirement for PNA binding-induced GFP expression in human NF cells To determine if the PNA length requirement obtained from the in vitro transcription assay is similar in the living cells, PNA binding-induced GFP expression was studied in human NF cells. The pJK3 plasmid DNA was incubated with individual PNAs in vitro to form the [PNA]2/DNA triplex and D-loop structures. Then the PNA-plasmid
Figure 2. Comparison of the binding of different PNAs. A ®xed concentration of the 32P-labeled 50 bp DNA fragment (5 10ÿ9 M) was incubated with increasing concentrations of PNA (from 1 10ÿ9 to 1 10ÿ5 M) in TE buffer (10 mM Tris (pH 6.0), 1 mM EDTA) at 37 C for 24 hours. The reactants were analyzed by polyacrylamide gel electrophoresis using a 20 % acrylamide gel followed by autoradiography. A GS-250 PhosphorImager (BioRad, Hercules, CA) was used for quanti®cation of the reaction products. The concentration at which PNA binding was half-maximal was taken as the equilibrium dissociation constant (Kd). The Kd was used to determine the binding af®nity of individual PNA.
936
PNA Binding-induced Transcription
Figure 3. Determination of transcription activity of individual PNA-generated single-stranded D-loop via reverse transcription assay. Nascent RNA was generated from individual PNA-bound pJK1 plasmid DNA via HeLa nuclear extract in vitro transcription. The RNA was used as a template in the reverse transcription assay. The amount of RNA transcripts initiated from the PNA-binding site region was quanti®ed using phosphorimaging analysis. The initiation sites of the transcripts were determined by comparison with the DNA sequence ladder. (a) Reverse transcription assay to detect speci®c RNA transcripts initiated from the PNA-binding site region. (b) Comparison of the transcription activity of the D-loops generated by individual PNAs. The results are mean data from ®ve individual experiments. The level of RNA transcription induced by the 18mer PNA at the PNA-binding site was counted as 100 %, and the amount of RNA transcription induced by other PNAs at the PNA-binding site was calculated as the percentage of that of 18mer PNA-induced transcription. Lane1, pJK1 plasmid DNA 10mer PNA; lane 2, pJK1 plasmid DNA 12mer PNA; lane 3, pJK1 plasmid DNA 14mer PNA; lane 4, pJK1 plasmid DNA 16mer PNA; lane 5, pJK1 plasmid DNA 18mer PNA; lane 6, pJK1 plasmid DNA 20mer PNA; lane 7, pJK1 plasmid DNA alone.
DNA complexes were transfected into the NF cells using cationic liposomes. As an internal control for the transfection, plasmid pDsRed1-C1, which expresses a red ¯uorescent protein (Clontech Laboratories, Inc., Palo Alto, CA), was co-transfected with the PNA-pJK3 complexes. Expression of the GFP and DsRed proteins were monitored in the transfected NF cells by ¯uorescent microscopic analysis after two days of incubation (Figure 4). No GFP was detected in either NF cells or NF cells transfected with pJK3 plasmid DNA (Figure 4(e) and (f)). In contrast, high levels of GFP were detected when the NF cells were trasnfected with pJK3 plasmid that was pretreated with 14mer, 16mer, 18mer, and 20mer PNAs (the 16mer and 18mer PNA-induced GFP gene expression is shown in Figure 4(g) and (h)). As a control, high levels of DsRed protein were detected in all the cotransfection experiments (Figure 4(a)-(d)). This result indicated that the PNAs of 14 20mer in length induced strong GFP expression in NF cells. This result indicates that the PNA length requirement for inducing expression of target genes in living cells is similar to that observed in HeLa nuclear extract in vitro transcription system. Determining the PNA-binding specificity To determine that the PNAs used in the study indeed bind to the designed target site of DNA template and generate single-stranded D-loops,
a DNase I protection-based DNA footprinting analysis was performed. A 300 bp DNA fragment containing the DNA sequence around the PNAbinding site was ampli®ed from the pJK1 plasmid DNA by PCR. The 50 end of the 300 bp DNA fragment was labeled with [g-32P]ATP. The labeled DNA fragment was incubated with individual PNAs to form the [PNA]2/DNA triplex and D-loop structure. The DNA-PNA complex then was digested with DNase I and analyzed by polyacrylamide gel electrophoresis using a denaturing 6 % (w/v) acrylamide gel (Figure 5). As expected, a speci®c DNA region located at the PNA-binding site was protected from the DNase I digestion with the binding of PNAs that are 14mer or longer. This result further demonstrated that the 14mer, 16mer, 18mer, and 20mer PNAs used in the study bound to the designed target sequence in a highly sequence-speci®c manner.
Discussion Induction of gene expression may have great implications for the treatment of many human diseases. However, there exists the challenge of identifying reagents that can induce expression of speci®c target genes. In our previous studies, we demonstrated that binding of PNAs to doublestranded DNA targets led to the synthesis of speci®c RNA transcripts from the PNA-binding sites. Therefore, the strategy of PNA binding-
PNA Binding-induced Transcription
937
Figure 4. PNA binding-mediated GFP expression from pJK3 plasmid in NF cells. Individual PNA was incubated with pJK 3 plasmid DNA for 48 hours for PNA binding to plasmid DNA. Then the PNA-plasmid DNA complexes were mixed with pDsRed1 C1 plasmid at a 1:1 ratio and transfected into the NF cells. Expression of GFP and DsRed was determined 48 hours after the transfection. The DsRed ¯uorescence is shown in the top panels, the GFP expression is shown in the middle panel and the visible light view of the same area is shown at the bottom panel. (a), (e), and (i), NF cells; (b), (f), and (j), NF cells transfected with pDsRed1 C1 and pJK3 plasmid DNA; (c), (g), and (k), NF cells transfected with pDsRed1 C1 plasmid DNA and pJK3 plasmid DNA that was pre-incubated with the 16mer PNA; (d), (h), and (l), NF cells transfected with pDsRed1 C1 plasmid DNA and pJK3 plasmid DNA that was pre-incubated with the 18mer PNA.
mediated target gene expression may provide a new approach for gene therapy of many diseases. In this study, we have investigated the PNA length requirement for PNA binding-induced transcription in model systems. Using a series of PNAs with different lengths, we have determined that PNAs with lengths of 14mer to 20mer induced strong transcription signals in a HeLa nuclear extract in vitro transcription system. The highest transcription activity was detected with the PNAs of 16mer and 18mer. A similar result was obtained when the PNA-bound plasmid DNAs were transfected into the human NF cells. All these results suggest that PNAs of 16 18mer in lengths are more effective at inducing transcription from the PNA-binding sites than other PNAs. This result provides important and valuable information for the design of PNAs for inducing speci®c target gene expression. The results obtained from this study indicate that the single-stranded D-loops generated by PNAs of 16 18mer are most effective in transcription initiation. This result is comparable with the normal transcription initiation documented in both prokaryotic and eukaryotic cells. The established transcription initiation model suggests that during
the normal transcription initiation process, a 12 20 bp DNA region is unwound and a singlestranded DNA region of approximately 17 bp is most effective in transcription.12 Although we did not measure the single stranded DNA region generated by PNA binding directly, the DNA footprinting analysis result suggests that the D-loops generated by PNA binding are comparable to the size of PNAs used in the study. Our results strongly suggest that a similar single-stranded DNA length requirement exists for both PNA binding-induced transcription initiation and normal protein-related transcription initiation of mammalian cells. Although the PNA length requirement for PNA binding-induced transcription has been determined in this study, the molecular basis of PNA bindingmediated transcription initiation process has not been established. For example, it has been demonstrated that during normal protein-related transcription initiation, the recognition and binding of TATA binding protein (TBP), a component of TFIID basal transcription factor, to the TATA sequence is the ®rst step in initiation of transcription. Then, RNA polymerase II, with the assistance of TFIIE and TFIIH, recognizes and binds to the
938
PNA Binding-induced Transcription
Figure 5. DNase I-based footprinting analysis to determine the PNA-binding speci®city. A 300 bp DNA fragment was ampli®ed from pJK1 plasmid DNA by PCR and end-labeled with [g-32P]ATP. The DNA was incubated with individual PNA in 10 mM Tris (pH 6.0), 1 mM EDTA at 37 C for 24 hours to form the [PNA]2/DNA triplex and single-stranded D-loop structure. Then the DNA-PNA complexes were incubated with DNase I at room temperature for two minutes and terminated by addition of stop solution. The reactants were denatured at 95 C for ten minutes and then analyzed by DNA sequencing gel electrophoresis using a 6 % acrylamide gel. Lane1, pJK1 plasmid DNA alone; lane 2, pJK1 plasmid DNA Control PNA; lane 3, pJK1 plasmid DNA 8mer PNA; lane 4, pJK1 plasmid DNA 10mer PNA; lane 5, pJK1 plasmid DNA 12mer PNA; lane 6, pJK1 plasmid DNA 14mer PNA; lane 7, pJK1 plamsidDNA 16mer PNA; lane 8, pJK1 plasmid DNA 18mer PNA; lane 9, pJK1 plamsid DNA 20mer PNA.
promoter region, resulting in opening of the promoter region for approximately 12 20 bp.12 ± 14 At the same time, RNA polymerase II undergoes a major conformational change, yielding a stable and highly processive elongation complex that completes synthesis of the nascent RNA transcripts. However, in the case of PNA binding-induced transcription, single-stranded loops have already been generated by binding of PNA to the target sites prior to the binding of the transcription machinery. Therefore, whether a similar transcription initiation process takes place for PNA bindinginduced transcription is unknown. The molecular mechanism of PNA binding-mediated transcription initiation is currently under investigation in this laboratory. In this study, PNA binding-induced transcription was investigated using in vitro bound PNAplasmid DNA complexes, and the PNA binding conditions have been optimized. However, many factors may affect PNA binding in living cells and result in reduced levels of target gene expression in vivo. For example, several studies have reported that high levels of salts prevent binding of PNA to DNA targets.7,10,15 High levels of salts are also known to be present in most living cells. Therefore, enhancing the binding of PNAs to target DNAs in the presence of high levels of salts will be important. Studies have demonstrated that increasing positive charges of PNAs can improve PNA binding ef®ciency in the presence of salt.16 Transcription of target genes can greatly enhance PNA binding to the DNA target in the presence of high levels of salt.17 Modi®ed PNAs with salt-independent binding may need to be developed to over-
come the limitation of the PNA binding in the presence of high levels of salt. PNAs were delivered into the living cells as PNA-DNA complexes using a cationic liposome transfection procedure in the current study and relatively high transfection ef®ciency was achieved. However, delivery of PNAs into living cells still encounters some dif®culty in general, due to the neutral charge property of PNAs. Improving PNA delivery ef®ciency to living cells will be important for using PNAs to regulate gene expression. A ligand-mediated delivery strategy may provide an ideal system to introduce PNAs into living cells.18 ± 23 This issue is pursued actively in our laboratory.
Materials and Methods Oligonucleotides, PNAs, plasmids, and cells Oligonucleotides used in this study are listed in Table 1 and were synthesized by either Oligos Etc. Inc. (Wilsonville, OR) or W. M. Keck Biotechnology Resource Center at the Yale University (New Haven, CT). The PNAs used in the study are listed in Table 1 and were synthesized from P.E. Biosystems (Framingham, MA). The human NF cells (GM00637F) were obtained from the NIGMS Human Genetic Mutant Cell Repository (Camden, NJ) and maintained in MEM supplemented with 10 % (v/v) fetal bovine serum (GIBCO BRL, Gaithersburg, MD). The plasmids used in this study were either purchased commercially or constructed in this laboratory. The pDsRed1-C1 plasmid was purchased from Clontech Laboratories, Inc. (Palo Alto, CA). The JK1 and pJK3 plasmids were constructed by a standard molecular cloning procedure.24 Brie¯y, two 20mer complementary oligonucleotides with a sequence of 50 AGAAGAGAAG
939
PNA Binding-induced Transcription GAAGAGAAGA30 were synthesized. These two oligonucleotides were annealed to form a 20 bp DNA fragment. The 20 bp DNA fragment was inserted into the BamHI-MluI sites of both pGL3-basic, a promoter reporter vector with a promoterless luciferase gene (Promega, Madison, WI), and pEGFP-1, a promoter reporter vector with a promoterless GFP gene (Clontech, Palo Alto, CA), to obtain the plasmids pJK1 and pJK3, respectively (Figure 1). The orientation of the inserted DNA fragment was con®rmed by DNA sequence analysis to be the same orientation as that of the reporter genes. PNA binding assay To measure PNA binding, two complementary 50mer oligonucleotides containing the PNA-binding sequence were synthesized (Table 1). One of the oligonucleotides was end-labeled with [g-32P]ATP. The labeled oligonucleotide was annealed with the complementary oligonucleotide at a 1:1 ratio to form a 50 bp DNA probe. A ®xed concentration of the labeled 50 bp DNA probe (4 10ÿ8 M) was incubated with increasing concentrations of PNA in 10 mM Tris (pH 6.0) at 37 C for 48 hours. The samples were analyzed by polyacrylamide gel electrophoresis in a 20 % acrylamide gel, followed by autoradiography. A GS-250 PhosphorImager (BioRad, Hercules, CA) was used for quanti®cation of the reaction products. The concentration at which PNA binding was half-maximal was taken as the equilibrium dissociation constant (Kd) In vitro transcription assay The in vitro transcription assays were performed in a HeLa nuclear extract in vitro transcription system under conditions recommended by the manufacturer (Promega; Madison, WI). Brie¯y, an 800 bp DNA fragment containing the PNA-binding site sequence was ampli®ed from the pJK1 plasmid DNA by PCR. The 800 bp DNA fragment (1 mg) was incubated with individual PNAs (10 mM) in 10 mM Tris (pH 6.0) at 37 C for 48 hours to induce the [PNA]2/DNA-loop formation. The PNADNA complexes were incubated in a 25 ml reaction containing 200 mM each ATP, CTP, GTP and UTP, and eight units of HeLa nuclear extract at 30 C for 60 minutes. Termination of the reaction was performed by addition of 175 ml of Stop Mix. The products were extracted with 200 ml of phenol/chloroform and were precipitated with 500 ml of ethanol. The pellets were resuspended in 10 ml of nuclease-free water and used for reverse transcription assay. Reverse transcription (RT) assay The RT reaction was set up in a volume of 25 ml containing M-MLV buffer (50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM DTT), 10 ml of nascent RNA, 200 mM each dATP, dCTP, dGTP, and dTTP, 1 fmol of end-labeled RT oligonucleotide, 200 units of MMLV reverse transcriptase and one unit of RNase inhibitor. The RNA and the primer were pre-incubated at 95 C for ten minutes and then cooled to 37 C. dNTPs, M-MLV reverse transcriptase, and RNase inhibitor were added, and the reactants were incubated at 37 C for 60 minutes. The reaction was terminated by addition of 12.5 ml of Stop solution. The reactants were denatured at 95 C for ten minutes and then analyzed by DNA
sequencing gel electrophoresis using a 7 % gel followed by autoradiography. DNA footprinting assay The DNA footprinting assay was performed using a Core Footprinting System (Promega, Madison, WI). Brie¯y, a 200 bp DNA fragment containing the PNAbinding site sequence was ampli®ed from pJK1 plasmid DNA by PCR. The purine-rich strand of the DNA fragment was end-labeled with [g-32P]ATP. The DNA fragment was incubated with individual PNAs to form the PNA-DNA complexes. DNase I digestion then was performed in a volume of 30 ml containing 1 mg of DNA, 10 mM Tris (pH 7.5), 6 mM MgCl2, 0.5 mM CaCl2, and 0.06 unit of DNase I. The reactants were incubated at room temperature for two minutes and then terminated by addition of 15 ml of Stop solution (200 mM NaCl, 30 mM EDTA, 1 % (w/v) SDS, 100 mg/ml of yeast RNA). The reactants were analyzed by DNA sequencing gel electrophoresis using 6 % gel and a DNA sequence ladder of the pJK1 plasmid DNA using the same primer was applied. Visualization of the results was achieved by autoradiography. Transfection of plasmid DNA into NF cells and detection of GFP expression Transfection of the PNA-bound plasmid DNA was performed using a Superfect transfection reagent (Qiagen, Santa Clarita, CA). The pJK3 plasmid DNA was pre-incubated with individual PNAs in vitro for two days to form the PNA-pJK3 plasmid DNA complexes. Then the pDsRed1-C1 plasmid DNA was mixed with PNA-pJK3 plasmid DNA complex at a 1:1 ratio. The plasmids were mixed with the Superfect reagent at a ratio of 10 mg DNA/50 ml Superfect reagent in a total volume of 300 ml. The mixture was incubated at room temperature for 40 minutes and then added into the cell culture dish. The cells were incubated in the liposomecontaining medium for eight hours, and then the medium was replaced with fresh cell growth medium. The cells were incubated in the tissue culture incubator at 37 C for an additional 48 hours, and the presence of GFP and DsRed was detected using an Olympus ¯uorescence microscope.
Acknowledgments We thank Dr Zhewen Chen, J. Harrison, N. Ramby, and M. George for their assistance in the experiments. We also thank Dr G. L. Wilson and Dr M. Bhatnagar for their helpful discussion and critical readings of the manuscript. Performance of this work was facilitated by the Cell Culture Facility Core and the Imaging and Cytometry Facility Core of the Environmental Health Sciences Center in Molecular and Cellular Toxicology with Human Applications at Wayne State University (P30 ES06639). This work is supported by a grant (R01 HL62551) from the National Heart, Lung, and Blood Institute of National Institute of Health (NIH) to G. W.
References 1. Dover, G. J., Brusilow, S. & Samid, D. (1992). Increased fetal hemoglobin in patients receiving
940
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
PNA Binding-induced Transcription
sodium 4-phenylbutyrate. N. Engl. J. Med. 327, 569570. Perrine, S. P., Ginder, G. D., Faller, D. V., Dover, G. H., Ikuta, T., Witkowaka, H. E. et al. (1993). A short-term trial of butyrate to stimulate fetal-globingene expression in the b-globin disorders. N. Engl. J. Med. 328, 81-86. Charache, S., Dover, G. J., Moyer, M. A. & Moore, J. W. (1987). Hydroxyurea-induced augmentation of fetal hemoglobin production in patients with sickle cell anemia. Blood, 69, 109-116. Fucharoen, S., Siritanaratkul, N., Winichagoon, P., Chowthaworn, J., Siriboon, W., Muangsup, W. et al. (1996). Hydroxyurea increase hemoglobin F levels and improve the effectiveness of erythropoiesis in bthalassemia/hemoglobin E disease. Blood, 87, 887892. Nielsen, P. E., Egholm, M., Berg, R. H. & Buchardt, O. (1991). Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science, 254, 1497-1500. Hanvey, J. C., Peffer, N. J., Bisi, J. E., Thomson, S. A., Cadilla, R., Josey, J. A. et al. (1992). Antisense and antigene properties of peptide nucleic acids. Science, 258, 1481-1485. Cherny, D. Y., Belotserkovskii, B. P., FrankKamenetskii, M. D., Egholm, M., Buchardt, O., Berg, R. H. & Nielsen, P. E. (1993). DNA unwinding upon strand-displacement binding of a thymine-substituted polyamide to double-stranded DNA. Proc. Natl Acad. Sci. USA, 90, 1667-1670. Demidov, V. V., Cherny, D. I., Kurakin, A. V., Yavnilovich, M. V., Malkov, V. A., FrankKamenetskii, M. D. & Sonnichsen, S. H. (1994). Electron microscopy mapping of oligopurine tracts in duplex DNA by peptide nucleic acid targeting. Nucl. Acids Res. 22, 5218-5222. Demidov, V. V., Yavnilovich, M. V., Belotserkovskii, B. P., Frank-Kamenetskii, M. D. & Nielsen, P. E. (1995). Kinetics and mechanism of polyamide (``peptide'') nucleic acid binding to duplex DNA. Proc. Natl Acad. Sci. USA, 92, 2637-2641. Peffer, N. J., Hanvey, J. C., Bisi, J. E., Thomson, S. A., Hassman, C. F., Noble, S. A. & Babiss, L. E. (1993). Strand-invasion of duplex DNA by peptide nucleic acid oligomers. Proc. Natl Acad. Sci. USA, 90, 1064810652. Wang, G., Xu, X., Pace, B., Dean, D. A., Glazer, P. M., Chan, P. et al. (1999). Peptide nucleic acid (PNA) binding-mediated induction of human g-globin gene expression. Nucl. Acids Res. 27, 2806-2813. Goodrich, J. A. & Tjian, R. (1994). Transcription factors IIE and IIH and ATP hydrolysis direct promoter clearance by RNA polymerase II. Cell, 77, 145-156.
13. Holstege, F. C. P., van der Vliet, P. C. & Timmers, H. T. M. (1996). Opening of an RNA polymerase II promoter occurs in two distinct steps and requires the basal transcription factors IIE and IIH. EMBO J. 15, 1666-1677. 14. Zawel, L. & Reinberg, D. (1995). Common themes in assembly and function of eukaryotic transcription complexes. Annu. Rev. Biochem. 64, 533-561. 15. Nielsen, P. E., Egholm, M. & Buchardt, O. (1994). Sequence-speci®c transcription arrest by peptide nucleic acid bound to the DNA template strand. Gene, 149, 139-145. 16. Mollegaard, N. E., Buchardt, O., Egholm, M. & Nielsen, P. E. (1994). Peptide nucleic acid. DNA strand displacement loops as arti®cial transcription promoters. Proc. Natl Acad. Sci. USA, 91, 3892-3895. 17. Larsen, H. J. & Nielsen, P. E. (1996). Transcriptionmediated binding of peptide nuclei acid (PNA) to double-stranded DNA: sequence-speci®c suicide transcription. Nucl. Acids Res. 24, 458-463. 18. Chinnery, P. F., Taylor, R. W., Diekert, K., Lill, R., Turnbull, D. M. & Lightowlers, R. N. (1999). Peptide nucleic acid delivery to human mitochondria. Gene Ther. 6, 1919-1928. 19. Branden, L. J., Mohamed, A. J. & Smith, C. I. E. (1999). A peptide nucleic acid-nuclear localization signal fusion that mediates nuclear transport of DNA. Nature Biotechnol. 17, 784-787. 20. Cutrona, G., Carpaneto, E. M., Ulivi, M., Roncella, S., Landt, O., Ferrarini, M. & Boffa, L. C. (2000). Effects in loive cells of a c-myc anti-gene PNA linked to a localization signal. Nature Biotechnol. 18, 300-303. 21. Pooga, M., Soomets, U., Hallbrink, M., Valkna, A., Saar, K., Rezaei, K. et al. (1998). Cell penetrating PNA constructs regulate galanin receptor levels and modify pain transmission in vivo. Nature Biotechnol. 16, 857-861. 22. Aldrian-Herrada, G., Desarmenien, M. G., Orcel, H., Boissin-Agasse, L., Mery, J., Brugidou, J. & Rabie, A. (1998). A peptide nucleic acid (PNA) is more rapidly internalized in cultured neurons when coupled to a retro-inverso delivery peptide. The antisense activity depresses the target mRNA and protein in magnocellular oxytocin neurons. Nucl. Acids Res. 26, 49104916. 23. Scar®, S., Giovine, M., Gasparini, A., Damonte, G., Millo, E., Pozzolini, M. & Benatti, U. (1999). Modi®ed peptide nucleic acids are internalized in mouse macrophages RAW 264.7 and inhibit inducible nitric oxide synthase. FEBS Letters, 451, 264-268. 24. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd edit., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Edited by J. Karn (Received 23 April 2001; received in revised form 19 September 2001; accepted 24 September 2001)
J. Mol. Biol. (2001) 313, 933±940
Defining the Peptide Nucleic Acids (PNA) Length Requirement for PNA Binding-induced Transcription and Gene Expression Gan Wang1*, Kai Jing1, R. Balczon2 and Xiaoxin Xu1 1
Institute of Environmental Health Sciences, Wayne State University, 2727 Second Avenue, Room 4000, Detroit MI 48201, USA 2 Department of Cell Biology and Neuroscience, University of South Alabama College of Medicine, Mobile AL 36688, USA
Induction of gene expression has great potential in the treatment of many human diseases. Peptide nucleic acid (PNA) as a novel DNA-binding reagent provides an ideal system to induce gene-speci®c expression. In our recent studies, we have demonstrated that PNA bound to doublestranded DNA targets and, therefore, generated single-stranded D-loops and induced transcription of target genes both in vitro and in vivo. Most importantly, we have demonstrated that treatment of cultured human cells with PNAs led to expression of an endogenous target gene. Therefore, the study of the molecular mechanism of PNA binding-induced gene expression will have great implications for the gene therapy of many human diseases. In the current study, we have investigated the PNA length requirement for PNA binding-induced transcription initiation. Using a series of PNAs with different lengths, we have determined that PNAs with lengths of 16 18 nt induce very high levels of transcription in a HeLa nuclear extract in vitro transcription system. Transfection of the PNA-bound GFP reporter gene plasmid into human normal ®broblast (NF) cells led to a similar result. Gel-mobility shift assays revealed very strong binding af®nities of these PNAs. DNA footprinting analysis further demonstrated the speci®city of PNAs binding to the targets. These results lead to important understanding of the molecular mechanism of transcription initiation and highly valuable information in PNA design, especially for PNA binding-induced, gene-speci®c expression. # 2001 Academic Press
*Corresponding author
Keywords: peptide nucleic acids (PNA); PNA binding-generated D-loop; transcription; transcription initiation; gene-speci®c expression
Introduction Abnormal gene expression has been linked to many human diseases, such as cancer and agerelated diseases. Induction of speci®c gene expression has a tremendous impact on the treatment of these diseases. Several drugs have been developed to induce gene expression.1 ± 4 Unfortunately, none of these drugs could induce genespeci®c expression. Peptide nucleic acids (PNAs) are synthetic oligonucleotides with modi®ed polyamide backbones.5 PNAs bind sequence-speci®cally to DNA and RNA targets to form Watson-Crick double helix Abbreviations used: PNA, peptide nucluic acid; NF, normal ®broplast; GFP, green ¯uorescent protein. E-mail address of the corresponding author: [email protected] 0022-2836/01/050933±8 $35.00/0
structures.5 When bound to a homopurine/homopyrimidine target of double-stranded DNA targets, the PNA molecule replaces one DNA strand by strand displacement to form a [PNA]2/DNA triplex structure, and the displaced DNA strand forms a D-loop structure at the PNA-binding site.6 ± 10 The triplex structure can stabilize the D-loop formed at the PNA binding sites signi®cantly. In our recent studies, PNA binding-generated D-loops have been demonstrated to have the capability of inducing transcription both in vitro and in vivo.11 More importantly, treatment of cultured human K562 erythroleukemia cells with PNAs designed to bind to the g-globin gene 50 ¯anking region resulted in transcription of the endogenous g-globin gene from the PNA target sites.11 This result suggests that a PNA binding-induced target gene expression strategy may provide a novel approach to induce # 2001 Academic Press
934 gene-speci®c expression of endogenous genes, which will have great implication on the treatment of many human diseases. However, the molecular mechanism of PNA binding-induced transcription has not been established. More importantly, the PNA length requirement for ef®cient transcription has not been determined. In this study, we have investigated the PNA length requirement for PNA binding-induced transcription and gene expression. Using reporter gene plasmid constructs with special PNA-binding sequences, the lengths of PNAs required for suf®cient transcription induction have been determined in a HeLa nuclear extract in vitro transcription system. Our results suggest that PNAs with lengths of 16-18nt induce very strong transcription. When the PNA-bound GFP reporter gene plasmid DNA was transfected into normal ®broblast (NF) cells, a similar result was obtained for the PNA length requirement. The gel-mobility shift assay has demonstrated strong binding of PNAs of 14-20 nt in length. DNase I-based footprinting analysis also demonstrated speci®c binding of PNAs to the target sites. All these results suggest that PNAs of 16-18 nt in length can induce strong transcription and target gene expression. The results obtained from this study lead to important understanding of the molecular mechanism of PNA binding-induced gene expression and will be essential in PNA designing, especially for PNA binding-induced target gene expression as a novel approach for gene therapy of many human diseases.
Results Designing and testing of PNAs that binds to a novel PNA-binding site To study the PNA length requirement for PNA binding-mediated transcription, a potential PNA-binding sequence was designed and cloned into promoter reporter plasmid vectors. A 20 bp DNA fragment with a sequence of 50 AGAAGAGA AGGAAGAGAAGA30 was designed and synthesized. This 20 bp DNA sequence was inserted into two promoter reporter vectors: pGL3-Basic, which carries a promoterless luciferase gene (Promega Corp., Madison, WI), and pEGFP-1, which carries a promoterless green ¯uorescent protein (GFP) gene (Clontech, Palo Alto, CA), to obtain the pJK1 and pJK3 plasmids, respectively (Figure 1). This 20 bp DNA fragment contains a homopurine/ homopyrimidine sequence for forming a [PNA]2/ DNA triplex structure at the PNA-binding site. The invert repeat sequence design in the homopurine/ homopyrimidine sequence enables the PNA molecule to bind to the DNA target either as a duplex or as a triplex strand to form the [PNA]2/DNA triplex. A series of PNAs with different lengths (8mer 20mer) were designed and synthesized to bind to the 20 bp homopurine/homopyrimidine target (Table 1). The binding af®nity of these PNAs
PNA Binding-induced Transcription
Figure 1. Structure of plasmids pJK1 and pJK3. The plasmid pJK 1 was constructed by inserting a 20 bp PNA-binding sequence into the MluI-BglII sites of pGL-3 Basic plasmid (Promega, Madison, WI). pJK3 was constructed by inserting the 20 bp PNA-binding sequence into the MluI-BglII site of plasmid pEGFP-1 (Clontech, Palo Alto, CA).
was determined by a gel-mobility shift assay (Figure 2). No detectable binding was observed with the 8mer PNA, even at the highest concentration tested (1 10ÿ5 M) (Figure 2). A weak binding was detected with the 10mer PNA at the concentration of 1 10ÿ5 M. Strong bindings were observed in the PNAs of 12mer or longer. The strongest binding af®nity was detected with the 16mer PNA: clear PNA binding was detected with the PNA concentration as low as 1 10ÿ7 M (Figure 2). No detectable binding was observed for a 15mer control PNA (data not shown). These results suggest that PNAs with length of 12mer or greater may be required for suf®cient binding to double-stranded DNA targets. Determining the levels of transcription of pJK1 plasmid DNA induced by individual PNA in HeLa nuclear extracts PNA binding-mediated transcription was studied in a HeLa nuclear extract in vitro transcription system. The pJK1 plasmid DNA was incubated with individual PNAs to form the [PNA]2/DNA triplex and the single-stranded Dloop structures. Then the PNA-plasmid DNA complex was incubated in a HeLa nuclear extract in vitro transcription system that was supplemented with rNTPs for RNA synthesis. The nascent RNA was isolated from the in vitro transcription reactants. A reverse transcription assay was performed to determine the initiation sites of the transcripts and a 20mer oligonucleotide that was complementary to the luciferase gene 50 mRNA sequence was used as a primer (Figure 3). No speci®c RNA transcripts initiated from the PNA-binding site were detected from pJK1 plasmid DNA alone, pJK1 plasmid DNA treated with the 10mer PNA or pJK1 plasmid
935
PNA Binding-induced Transcription Table 1. Oligonucleotides and PNAs used in the study A. Oligonucleotides used as PNA binding target Oligo 1: 50 AGCTAGTCATGCGATCTCTTCTCTTCCTTCTCTTCTAATGCACGTAACGG30 Oligo 2: 50 CCGTTACGTGCATTAGAAGAGAAGGAAGAGAAGAGATCGCATGACTAGCT30 B. RT PCR primer 50 GCCTTATGCAGTTGCTCTCC30 C. PNAs used in the study 8 mer PNA: 10 mer PNA: 12 mer PNA: 14 mer PNA: 16 mer PNA: 18 mer PNA: 20 mer PNA:
NH2-CTTCCTTC-COOH NH2-TCTTCCTTCT-COOH NH2-CTCTTCCTTCTC-COOH NH2-TCTCTTCCTTCTCT-COOH NH2-TTCTCTTCCTTCTCTT-COOH NH2-CTTCTCTTCCTTCTCTTC-COOH NH2-TCTTCTCTTCCTTCTCTTCT-COOH
D. Control PNA
NH2-TACGGTTATCCACAG-COOH
DNA treated with 12mer PNA in the in vitro transcription reaction (Figure 3, lanes 1, 2, and 7). In contrast, speci®c RNA transcripts initiated from the PNA-binding site were detected when the pJK1 plasmid DNA treated with either 14mer, 16mer, 18mer, or 20mer PNAs were used as the templates (Figure 3, lanes 3 to 6). Phosphorimaging analysis indicated that the highest transcription signal was obtained when the 18mer PNA-treated pJK1 plasmid DNA was used as a substrate (Figure 3, lane 5 versus other lanes). Noticeably, the transcription initiation position shifted to the 30 side of the PNA-bind-
ing site in the DNA template as the length of the bound PNA increased. Determining the PNA length requirement for PNA binding-induced GFP expression in human NF cells To determine if the PNA length requirement obtained from the in vitro transcription assay is similar in the living cells, PNA binding-induced GFP expression was studied in human NF cells. The pJK3 plasmid DNA was incubated with individual PNAs in vitro to form the [PNA]2/DNA triplex and D-loop structures. Then the PNA-plasmid
Figure 2. Comparison of the binding of different PNAs. A ®xed concentration of the 32P-labeled 50 bp DNA fragment (5 10ÿ9 M) was incubated with increasing concentrations of PNA (from 1 10ÿ9 to 1 10ÿ5 M) in TE buffer (10 mM Tris (pH 6.0), 1 mM EDTA) at 37 C for 24 hours. The reactants were analyzed by polyacrylamide gel electrophoresis using a 20 % acrylamide gel followed by autoradiography. A GS-250 PhosphorImager (BioRad, Hercules, CA) was used for quanti®cation of the reaction products. The concentration at which PNA binding was half-maximal was taken as the equilibrium dissociation constant (Kd). The Kd was used to determine the binding af®nity of individual PNA.
936
PNA Binding-induced Transcription
Figure 3. Determination of transcription activity of individual PNA-generated single-stranded D-loop via reverse transcription assay. Nascent RNA was generated from individual PNA-bound pJK1 plasmid DNA via HeLa nuclear extract in vitro transcription. The RNA was used as a template in the reverse transcription assay. The amount of RNA transcripts initiated from the PNA-binding site region was quanti®ed using phosphorimaging analysis. The initiation sites of the transcripts were determined by comparison with the DNA sequence ladder. (a) Reverse transcription assay to detect speci®c RNA transcripts initiated from the PNA-binding site region. (b) Comparison of the transcription activity of the D-loops generated by individual PNAs. The results are mean data from ®ve individual experiments. The level of RNA transcription induced by the 18mer PNA at the PNA-binding site was counted as 100 %, and the amount of RNA transcription induced by other PNAs at the PNA-binding site was calculated as the percentage of that of 18mer PNA-induced transcription. Lane1, pJK1 plasmid DNA 10mer PNA; lane 2, pJK1 plasmid DNA 12mer PNA; lane 3, pJK1 plasmid DNA 14mer PNA; lane 4, pJK1 plasmid DNA 16mer PNA; lane 5, pJK1 plasmid DNA 18mer PNA; lane 6, pJK1 plasmid DNA 20mer PNA; lane 7, pJK1 plasmid DNA alone.
DNA complexes were transfected into the NF cells using cationic liposomes. As an internal control for the transfection, plasmid pDsRed1-C1, which expresses a red ¯uorescent protein (Clontech Laboratories, Inc., Palo Alto, CA), was co-transfected with the PNA-pJK3 complexes. Expression of the GFP and DsRed proteins were monitored in the transfected NF cells by ¯uorescent microscopic analysis after two days of incubation (Figure 4). No GFP was detected in either NF cells or NF cells transfected with pJK3 plasmid DNA (Figure 4(e) and (f)). In contrast, high levels of GFP were detected when the NF cells were trasnfected with pJK3 plasmid that was pretreated with 14mer, 16mer, 18mer, and 20mer PNAs (the 16mer and 18mer PNA-induced GFP gene expression is shown in Figure 4(g) and (h)). As a control, high levels of DsRed protein were detected in all the cotransfection experiments (Figure 4(a)-(d)). This result indicated that the PNAs of 14 20mer in length induced strong GFP expression in NF cells. This result indicates that the PNA length requirement for inducing expression of target genes in living cells is similar to that observed in HeLa nuclear extract in vitro transcription system. Determining the PNA-binding specificity To determine that the PNAs used in the study indeed bind to the designed target site of DNA template and generate single-stranded D-loops,
a DNase I protection-based DNA footprinting analysis was performed. A 300 bp DNA fragment containing the DNA sequence around the PNAbinding site was ampli®ed from the pJK1 plasmid DNA by PCR. The 50 end of the 300 bp DNA fragment was labeled with [g-32P]ATP. The labeled DNA fragment was incubated with individual PNAs to form the [PNA]2/DNA triplex and D-loop structure. The DNA-PNA complex then was digested with DNase I and analyzed by polyacrylamide gel electrophoresis using a denaturing 6 % (w/v) acrylamide gel (Figure 5). As expected, a speci®c DNA region located at the PNA-binding site was protected from the DNase I digestion with the binding of PNAs that are 14mer or longer. This result further demonstrated that the 14mer, 16mer, 18mer, and 20mer PNAs used in the study bound to the designed target sequence in a highly sequence-speci®c manner.
Discussion Induction of gene expression may have great implications for the treatment of many human diseases. However, there exists the challenge of identifying reagents that can induce expression of speci®c target genes. In our previous studies, we demonstrated that binding of PNAs to doublestranded DNA targets led to the synthesis of speci®c RNA transcripts from the PNA-binding sites. Therefore, the strategy of PNA binding-
PNA Binding-induced Transcription
937
Figure 4. PNA binding-mediated GFP expression from pJK3 plasmid in NF cells. Individual PNA was incubated with pJK 3 plasmid DNA for 48 hours for PNA binding to plasmid DNA. Then the PNA-plasmid DNA complexes were mixed with pDsRed1 C1 plasmid at a 1:1 ratio and transfected into the NF cells. Expression of GFP and DsRed was determined 48 hours after the transfection. The DsRed ¯uorescence is shown in the top panels, the GFP expression is shown in the middle panel and the visible light view of the same area is shown at the bottom panel. (a), (e), and (i), NF cells; (b), (f), and (j), NF cells transfected with pDsRed1 C1 and pJK3 plasmid DNA; (c), (g), and (k), NF cells transfected with pDsRed1 C1 plasmid DNA and pJK3 plasmid DNA that was pre-incubated with the 16mer PNA; (d), (h), and (l), NF cells transfected with pDsRed1 C1 plasmid DNA and pJK3 plasmid DNA that was pre-incubated with the 18mer PNA.
mediated target gene expression may provide a new approach for gene therapy of many diseases. In this study, we have investigated the PNA length requirement for PNA binding-induced transcription in model systems. Using a series of PNAs with different lengths, we have determined that PNAs with lengths of 14mer to 20mer induced strong transcription signals in a HeLa nuclear extract in vitro transcription system. The highest transcription activity was detected with the PNAs of 16mer and 18mer. A similar result was obtained when the PNA-bound plasmid DNAs were transfected into the human NF cells. All these results suggest that PNAs of 16 18mer in lengths are more effective at inducing transcription from the PNA-binding sites than other PNAs. This result provides important and valuable information for the design of PNAs for inducing speci®c target gene expression. The results obtained from this study indicate that the single-stranded D-loops generated by PNAs of 16 18mer are most effective in transcription initiation. This result is comparable with the normal transcription initiation documented in both prokaryotic and eukaryotic cells. The established transcription initiation model suggests that during
the normal transcription initiation process, a 12 20 bp DNA region is unwound and a singlestranded DNA region of approximately 17 bp is most effective in transcription.12 Although we did not measure the single stranded DNA region generated by PNA binding directly, the DNA footprinting analysis result suggests that the D-loops generated by PNA binding are comparable to the size of PNAs used in the study. Our results strongly suggest that a similar single-stranded DNA length requirement exists for both PNA binding-induced transcription initiation and normal protein-related transcription initiation of mammalian cells. Although the PNA length requirement for PNA binding-induced transcription has been determined in this study, the molecular basis of PNA bindingmediated transcription initiation process has not been established. For example, it has been demonstrated that during normal protein-related transcription initiation, the recognition and binding of TATA binding protein (TBP), a component of TFIID basal transcription factor, to the TATA sequence is the ®rst step in initiation of transcription. Then, RNA polymerase II, with the assistance of TFIIE and TFIIH, recognizes and binds to the
938
PNA Binding-induced Transcription
Figure 5. DNase I-based footprinting analysis to determine the PNA-binding speci®city. A 300 bp DNA fragment was ampli®ed from pJK1 plasmid DNA by PCR and end-labeled with [g-32P]ATP. The DNA was incubated with individual PNA in 10 mM Tris (pH 6.0), 1 mM EDTA at 37 C for 24 hours to form the [PNA]2/DNA triplex and single-stranded D-loop structure. Then the DNA-PNA complexes were incubated with DNase I at room temperature for two minutes and terminated by addition of stop solution. The reactants were denatured at 95 C for ten minutes and then analyzed by DNA sequencing gel electrophoresis using a 6 % acrylamide gel. Lane1, pJK1 plasmid DNA alone; lane 2, pJK1 plasmid DNA Control PNA; lane 3, pJK1 plasmid DNA 8mer PNA; lane 4, pJK1 plasmid DNA 10mer PNA; lane 5, pJK1 plasmid DNA 12mer PNA; lane 6, pJK1 plasmid DNA 14mer PNA; lane 7, pJK1 plamsidDNA 16mer PNA; lane 8, pJK1 plasmid DNA 18mer PNA; lane 9, pJK1 plamsid DNA 20mer PNA.
promoter region, resulting in opening of the promoter region for approximately 12 20 bp.12 ± 14 At the same time, RNA polymerase II undergoes a major conformational change, yielding a stable and highly processive elongation complex that completes synthesis of the nascent RNA transcripts. However, in the case of PNA binding-induced transcription, single-stranded loops have already been generated by binding of PNA to the target sites prior to the binding of the transcription machinery. Therefore, whether a similar transcription initiation process takes place for PNA bindinginduced transcription is unknown. The molecular mechanism of PNA binding-mediated transcription initiation is currently under investigation in this laboratory. In this study, PNA binding-induced transcription was investigated using in vitro bound PNAplasmid DNA complexes, and the PNA binding conditions have been optimized. However, many factors may affect PNA binding in living cells and result in reduced levels of target gene expression in vivo. For example, several studies have reported that high levels of salts prevent binding of PNA to DNA targets.7,10,15 High levels of salts are also known to be present in most living cells. Therefore, enhancing the binding of PNAs to target DNAs in the presence of high levels of salts will be important. Studies have demonstrated that increasing positive charges of PNAs can improve PNA binding ef®ciency in the presence of salt.16 Transcription of target genes can greatly enhance PNA binding to the DNA target in the presence of high levels of salt.17 Modi®ed PNAs with salt-independent binding may need to be developed to over-
come the limitation of the PNA binding in the presence of high levels of salt. PNAs were delivered into the living cells as PNA-DNA complexes using a cationic liposome transfection procedure in the current study and relatively high transfection ef®ciency was achieved. However, delivery of PNAs into living cells still encounters some dif®culty in general, due to the neutral charge property of PNAs. Improving PNA delivery ef®ciency to living cells will be important for using PNAs to regulate gene expression. A ligand-mediated delivery strategy may provide an ideal system to introduce PNAs into living cells.18 ± 23 This issue is pursued actively in our laboratory.
Materials and Methods Oligonucleotides, PNAs, plasmids, and cells Oligonucleotides used in this study are listed in Table 1 and were synthesized by either Oligos Etc. Inc. (Wilsonville, OR) or W. M. Keck Biotechnology Resource Center at the Yale University (New Haven, CT). The PNAs used in the study are listed in Table 1 and were synthesized from P.E. Biosystems (Framingham, MA). The human NF cells (GM00637F) were obtained from the NIGMS Human Genetic Mutant Cell Repository (Camden, NJ) and maintained in MEM supplemented with 10 % (v/v) fetal bovine serum (GIBCO BRL, Gaithersburg, MD). The plasmids used in this study were either purchased commercially or constructed in this laboratory. The pDsRed1-C1 plasmid was purchased from Clontech Laboratories, Inc. (Palo Alto, CA). The JK1 and pJK3 plasmids were constructed by a standard molecular cloning procedure.24 Brie¯y, two 20mer complementary oligonucleotides with a sequence of 50 AGAAGAGAAG
939
PNA Binding-induced Transcription GAAGAGAAGA30 were synthesized. These two oligonucleotides were annealed to form a 20 bp DNA fragment. The 20 bp DNA fragment was inserted into the BamHI-MluI sites of both pGL3-basic, a promoter reporter vector with a promoterless luciferase gene (Promega, Madison, WI), and pEGFP-1, a promoter reporter vector with a promoterless GFP gene (Clontech, Palo Alto, CA), to obtain the plasmids pJK1 and pJK3, respectively (Figure 1). The orientation of the inserted DNA fragment was con®rmed by DNA sequence analysis to be the same orientation as that of the reporter genes. PNA binding assay To measure PNA binding, two complementary 50mer oligonucleotides containing the PNA-binding sequence were synthesized (Table 1). One of the oligonucleotides was end-labeled with [g-32P]ATP. The labeled oligonucleotide was annealed with the complementary oligonucleotide at a 1:1 ratio to form a 50 bp DNA probe. A ®xed concentration of the labeled 50 bp DNA probe (4 10ÿ8 M) was incubated with increasing concentrations of PNA in 10 mM Tris (pH 6.0) at 37 C for 48 hours. The samples were analyzed by polyacrylamide gel electrophoresis in a 20 % acrylamide gel, followed by autoradiography. A GS-250 PhosphorImager (BioRad, Hercules, CA) was used for quanti®cation of the reaction products. The concentration at which PNA binding was half-maximal was taken as the equilibrium dissociation constant (Kd) In vitro transcription assay The in vitro transcription assays were performed in a HeLa nuclear extract in vitro transcription system under conditions recommended by the manufacturer (Promega; Madison, WI). Brie¯y, an 800 bp DNA fragment containing the PNA-binding site sequence was ampli®ed from the pJK1 plasmid DNA by PCR. The 800 bp DNA fragment (1 mg) was incubated with individual PNAs (10 mM) in 10 mM Tris (pH 6.0) at 37 C for 48 hours to induce the [PNA]2/DNA-loop formation. The PNADNA complexes were incubated in a 25 ml reaction containing 200 mM each ATP, CTP, GTP and UTP, and eight units of HeLa nuclear extract at 30 C for 60 minutes. Termination of the reaction was performed by addition of 175 ml of Stop Mix. The products were extracted with 200 ml of phenol/chloroform and were precipitated with 500 ml of ethanol. The pellets were resuspended in 10 ml of nuclease-free water and used for reverse transcription assay. Reverse transcription (RT) assay The RT reaction was set up in a volume of 25 ml containing M-MLV buffer (50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM DTT), 10 ml of nascent RNA, 200 mM each dATP, dCTP, dGTP, and dTTP, 1 fmol of end-labeled RT oligonucleotide, 200 units of MMLV reverse transcriptase and one unit of RNase inhibitor. The RNA and the primer were pre-incubated at 95 C for ten minutes and then cooled to 37 C. dNTPs, M-MLV reverse transcriptase, and RNase inhibitor were added, and the reactants were incubated at 37 C for 60 minutes. The reaction was terminated by addition of 12.5 ml of Stop solution. The reactants were denatured at 95 C for ten minutes and then analyzed by DNA
sequencing gel electrophoresis using a 7 % gel followed by autoradiography. DNA footprinting assay The DNA footprinting assay was performed using a Core Footprinting System (Promega, Madison, WI). Brie¯y, a 200 bp DNA fragment containing the PNAbinding site sequence was ampli®ed from pJK1 plasmid DNA by PCR. The purine-rich strand of the DNA fragment was end-labeled with [g-32P]ATP. The DNA fragment was incubated with individual PNAs to form the PNA-DNA complexes. DNase I digestion then was performed in a volume of 30 ml containing 1 mg of DNA, 10 mM Tris (pH 7.5), 6 mM MgCl2, 0.5 mM CaCl2, and 0.06 unit of DNase I. The reactants were incubated at room temperature for two minutes and then terminated by addition of 15 ml of Stop solution (200 mM NaCl, 30 mM EDTA, 1 % (w/v) SDS, 100 mg/ml of yeast RNA). The reactants were analyzed by DNA sequencing gel electrophoresis using 6 % gel and a DNA sequence ladder of the pJK1 plasmid DNA using the same primer was applied. Visualization of the results was achieved by autoradiography. Transfection of plasmid DNA into NF cells and detection of GFP expression Transfection of the PNA-bound plasmid DNA was performed using a Superfect transfection reagent (Qiagen, Santa Clarita, CA). The pJK3 plasmid DNA was pre-incubated with individual PNAs in vitro for two days to form the PNA-pJK3 plasmid DNA complexes. Then the pDsRed1-C1 plasmid DNA was mixed with PNA-pJK3 plasmid DNA complex at a 1:1 ratio. The plasmids were mixed with the Superfect reagent at a ratio of 10 mg DNA/50 ml Superfect reagent in a total volume of 300 ml. The mixture was incubated at room temperature for 40 minutes and then added into the cell culture dish. The cells were incubated in the liposomecontaining medium for eight hours, and then the medium was replaced with fresh cell growth medium. The cells were incubated in the tissue culture incubator at 37 C for an additional 48 hours, and the presence of GFP and DsRed was detected using an Olympus ¯uorescence microscope.
Acknowledgments We thank Dr Zhewen Chen, J. Harrison, N. Ramby, and M. George for their assistance in the experiments. We also thank Dr G. L. Wilson and Dr M. Bhatnagar for their helpful discussion and critical readings of the manuscript. Performance of this work was facilitated by the Cell Culture Facility Core and the Imaging and Cytometry Facility Core of the Environmental Health Sciences Center in Molecular and Cellular Toxicology with Human Applications at Wayne State University (P30 ES06639). This work is supported by a grant (R01 HL62551) from the National Heart, Lung, and Blood Institute of National Institute of Health (NIH) to G. W.
References 1. Dover, G. J., Brusilow, S. & Samid, D. (1992). Increased fetal hemoglobin in patients receiving
940
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
PNA Binding-induced Transcription
sodium 4-phenylbutyrate. N. Engl. J. Med. 327, 569570. Perrine, S. P., Ginder, G. D., Faller, D. V., Dover, G. H., Ikuta, T., Witkowaka, H. E. et al. (1993). A short-term trial of butyrate to stimulate fetal-globingene expression in the b-globin disorders. N. Engl. J. Med. 328, 81-86. Charache, S., Dover, G. J., Moyer, M. A. & Moore, J. W. (1987). Hydroxyurea-induced augmentation of fetal hemoglobin production in patients with sickle cell anemia. Blood, 69, 109-116. Fucharoen, S., Siritanaratkul, N., Winichagoon, P., Chowthaworn, J., Siriboon, W., Muangsup, W. et al. (1996). Hydroxyurea increase hemoglobin F levels and improve the effectiveness of erythropoiesis in bthalassemia/hemoglobin E disease. Blood, 87, 887892. Nielsen, P. E., Egholm, M., Berg, R. H. & Buchardt, O. (1991). Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science, 254, 1497-1500. Hanvey, J. C., Peffer, N. J., Bisi, J. E., Thomson, S. A., Cadilla, R., Josey, J. A. et al. (1992). Antisense and antigene properties of peptide nucleic acids. Science, 258, 1481-1485. Cherny, D. Y., Belotserkovskii, B. P., FrankKamenetskii, M. D., Egholm, M., Buchardt, O., Berg, R. H. & Nielsen, P. E. (1993). DNA unwinding upon strand-displacement binding of a thymine-substituted polyamide to double-stranded DNA. Proc. Natl Acad. Sci. USA, 90, 1667-1670. Demidov, V. V., Cherny, D. I., Kurakin, A. V., Yavnilovich, M. V., Malkov, V. A., FrankKamenetskii, M. D. & Sonnichsen, S. H. (1994). Electron microscopy mapping of oligopurine tracts in duplex DNA by peptide nucleic acid targeting. Nucl. Acids Res. 22, 5218-5222. Demidov, V. V., Yavnilovich, M. V., Belotserkovskii, B. P., Frank-Kamenetskii, M. D. & Nielsen, P. E. (1995). Kinetics and mechanism of polyamide (``peptide'') nucleic acid binding to duplex DNA. Proc. Natl Acad. Sci. USA, 92, 2637-2641. Peffer, N. J., Hanvey, J. C., Bisi, J. E., Thomson, S. A., Hassman, C. F., Noble, S. A. & Babiss, L. E. (1993). Strand-invasion of duplex DNA by peptide nucleic acid oligomers. Proc. Natl Acad. Sci. USA, 90, 1064810652. Wang, G., Xu, X., Pace, B., Dean, D. A., Glazer, P. M., Chan, P. et al. (1999). Peptide nucleic acid (PNA) binding-mediated induction of human g-globin gene expression. Nucl. Acids Res. 27, 2806-2813. Goodrich, J. A. & Tjian, R. (1994). Transcription factors IIE and IIH and ATP hydrolysis direct promoter clearance by RNA polymerase II. Cell, 77, 145-156.
13. Holstege, F. C. P., van der Vliet, P. C. & Timmers, H. T. M. (1996). Opening of an RNA polymerase II promoter occurs in two distinct steps and requires the basal transcription factors IIE and IIH. EMBO J. 15, 1666-1677. 14. Zawel, L. & Reinberg, D. (1995). Common themes in assembly and function of eukaryotic transcription complexes. Annu. Rev. Biochem. 64, 533-561. 15. Nielsen, P. E., Egholm, M. & Buchardt, O. (1994). Sequence-speci®c transcription arrest by peptide nucleic acid bound to the DNA template strand. Gene, 149, 139-145. 16. Mollegaard, N. E., Buchardt, O., Egholm, M. & Nielsen, P. E. (1994). Peptide nucleic acid. DNA strand displacement loops as arti®cial transcription promoters. Proc. Natl Acad. Sci. USA, 91, 3892-3895. 17. Larsen, H. J. & Nielsen, P. E. (1996). Transcriptionmediated binding of peptide nuclei acid (PNA) to double-stranded DNA: sequence-speci®c suicide transcription. Nucl. Acids Res. 24, 458-463. 18. Chinnery, P. F., Taylor, R. W., Diekert, K., Lill, R., Turnbull, D. M. & Lightowlers, R. N. (1999). Peptide nucleic acid delivery to human mitochondria. Gene Ther. 6, 1919-1928. 19. Branden, L. J., Mohamed, A. J. & Smith, C. I. E. (1999). A peptide nucleic acid-nuclear localization signal fusion that mediates nuclear transport of DNA. Nature Biotechnol. 17, 784-787. 20. Cutrona, G., Carpaneto, E. M., Ulivi, M., Roncella, S., Landt, O., Ferrarini, M. & Boffa, L. C. (2000). Effects in loive cells of a c-myc anti-gene PNA linked to a localization signal. Nature Biotechnol. 18, 300-303. 21. Pooga, M., Soomets, U., Hallbrink, M., Valkna, A., Saar, K., Rezaei, K. et al. (1998). Cell penetrating PNA constructs regulate galanin receptor levels and modify pain transmission in vivo. Nature Biotechnol. 16, 857-861. 22. Aldrian-Herrada, G., Desarmenien, M. G., Orcel, H., Boissin-Agasse, L., Mery, J., Brugidou, J. & Rabie, A. (1998). A peptide nucleic acid (PNA) is more rapidly internalized in cultured neurons when coupled to a retro-inverso delivery peptide. The antisense activity depresses the target mRNA and protein in magnocellular oxytocin neurons. Nucl. Acids Res. 26, 49104916. 23. Scar®, S., Giovine, M., Gasparini, A., Damonte, G., Millo, E., Pozzolini, M. & Benatti, U. (1999). Modi®ed peptide nucleic acids are internalized in mouse macrophages RAW 264.7 and inhibit inducible nitric oxide synthase. FEBS Letters, 451, 264-268. 24. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd edit., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Edited by J. Karn (Received 23 April 2001; received in revised form 19 September 2001; accepted 24 September 2001)