DNA-binding oligoamides have been designed using the pairing rules discovered through studies on Distamycin A [97, 98]: A-T base pairs are recognized by Py/ Hp; T-A base pairs by Hp/Py; C-G base pairs are recognized by Py/Im; and G-C base pairs by Im/Py [97, 98, 100-103]. Single stranded oligoamides bind in an extended fashion in the DNA minor groove [96, 188]. Covalently linking the oli-
goamide strands into a hairpin increases the affinity and specificity of these oligomers [103, 189-195]. Hairpin compounds typically retain the orientation preferences of extended polyamides, aligning N to C with the 5' to 3' direction of the closest DNA strand. The classic example of DNA-binding oligoamides is the eight-ring hairpin, in which a g-aminobutryic acid serves as the linker (Fig. 8.16) [189, 191, 192]. Covalent linkage of complementary oligoamides connecting the carboxyl terminus of one oligoamide to the amino terminus of another to create a hairpin can result in an increase of DNA-binding affinities by 100-fold . Eight-ring hairpins have been demonstrated to bind specific DNA sequences (6bp) with affinities similar to that of DNA-binding proteins (KD < 1 nM) . Undesirable binding events of hairpin oligoamides, such as binding in the opposite intended orientation to target DNA or binding in a 1:1 rather than the intended 2:1 binding mode have been remedied by the introduction of an amino substituent at the a-position of the g-aminobutryic acid turn  or b-Ala .
H-pin [111, 194] oligoamides have also been designed, whereby an alkyl chain projecting from the minor groove is used to link oligoamide strands at the central position (Fig. 8.16). The introduction of b-Ala into covalent linkers introduces flexibility to the structure and can relax the curvature of oligoamides, yielding molecules designed after homo-dimeric complexes to bind 11 base pair sequences of DNA with subnanomolar affinities . Tandem hairpin oligoamides [106, 108], in which two hairpins are covalently linked by a 5-aminovaleric acid linker, bound specifically to 10 base pair sites.
Cyclic oligoamides [196, 197], in which the C and N termini of a hairpin have been covalently linked, eliminated the possibility of extended binding (Fig. 8.16). Such oligoamides have demonstrated higher affinities but lower specificities for target DNA sequences compared with analogous hairpin molecules with the same number of cationic groups.
Oligoamides can bind a multitude of specific DNA sites with affinities comparable to DNA-binding proteins . The binding of oligoamides to promoter sites on DNA have inhibited gene transcription by disrupting RNA polymerase activity. In particular, the suppression of 5S RNA transcription RNA polymerase III has been achieved both in vitro and in cultured Xenopus kidney cells by minor-groove binding oligoamides . Transcription of HIV-1 has also been inhibited through the binding of designed hairpin oligoamides to multiple transcription factor binding sites within the HIV-1 enhancer/promoter site as revealed by cell free assays . The same oligoamides were able to halt viral replication by more than 99% in human blood lymphocytes by directly effecting viral transcription without greatly affecting cell viability. Oligoamides have been designed to target GC rich regions flanking CRE sites to inhibit Tax protein binding and Tax transactivation in vitro . Other oligoamides targeting GC rich sequences bound to a 4 base pair site typically cleaved by bacterial gyrase, thus preventing strand cleavage at nanomolar concentrations . The NF-kB hetero-dimer, a transcription factor which binds in the DNA major groove, has been inhibited by oligoamides that bind the minor groove opposite one of the monomers . Oligoamides conjugated with the tripeptide sequence Arg-Pro-Arg
Fig. 8.16 Various types of DNA-binding oligoamides.
can interfere with major groove-binding DNA proteins by distorting DNA through charge neutralization, occupying the major groove to cause steric interference, or binding the backbone phosphate. Such oligoamides were able to inhibit the major groove binding transcription factor GCN4 . Gene expression has also been affected by oligoamides in Drosophila; oligoamides introduced into food sources resulted in the noticeable gain and loss of certain phenotypes, demonstrating the effectiveness of oligoamides in complex organisms .
Oligoamides have also been demonstrated to upregulate transcription, either through the inhibition of repressor proteins or recruitment of transcriptional machinery. A hairpin oligoamide was able to block binding of the repressor IE86, upregulating the transcription of human cytomegalovirus MIEP . Upregula-tion of the HIV-1 promoter has also been achieved by oligoamide binding to the repressor complex sequence in the HIV-1 long terminal repeat, effectively inhibiting the human protein LSF, a protein involved in HIV-1 latency . Oligoamides have also been created which act as artificial transcription factors; a hairpin oligoamide was linked to a 20-residue peptide activation domain by a 36 atom straight-chain linker, resulting in activation of transcription in cell free assays . Appendage of an even smaller yet more potent peptide activation domain was also achieved, resulting in even higher levels of transcription .
Hairpin oligoamides have been used to target specific DNA sequences in nucle-osomal DNA complexes . Accessibility to certain DNA sequences of the nu-cleosome is dependent on the positioning and structural implications of the DNA strand wound about 8 histone proteins. Results show that DNA base pairs fully or sometimes partially facing away from the histone octamer are fully accessible . Oligoamides were shown to bind the nucleosome positioning sequence of the sea urchin 5S gene with KD @ 1 nM, blocking heat-inducible nucleosomal translocation and transcription by the T7 RNA polymerase .
Peptide nucleic acids have been primarily used to affect gene transcription, either through enhancement or suppression. In general, transcription enhancement mechanisms are more complex since activity not only relies on DNA binding but also the recruitment of various agents needed by the transcriptional machinery (i.e. transcription factors). The H-bonding pattern between PNA and DNA base pairs is directly consequential to the binding mode. For instance, PNA can pair with single stranded DNA to form hetero-duplexes which are bound together by Watson-Crick base-pairing, while two PNAs can bind a single stranded DNA to form a triple helix via a combination of Watson-Crick and Hoogsteen base pairing. One PNA strand binds DNA via Watson-Crick pairing while the other PNA binds the PNA-DNA heteroduplex via Hoogsteen pairing.
PNAs have been shown to inhibit transcription. A 15-mer homopyrimidine PNA targeted to the IL2-Ra NF-kB binding site inhibited transcription factor binding through strand invasion . Transactivation was inhibited in vitro when the PNA was pre-incubated with the target DNA under low salt concentration prior to addition to nuclear extracts. Although direct addition of the PNA to HeLa cells did not inhibit NF-kB mediated transactivation, inhibitory effects were evident up to 24 hours after introduction of preincubated PNA-reporter plasmid complexes. Another strategy to block transcription has been achieved through prevention of transcribed strand elongation. Binding of a homo-thymidine 10-mer PNA or a mixed sequence 15-mer to the template strand of a G-free transcription cassette have been demonstrated to block site-specific pol II transcription elongation in vitro . Binding of the 10-mer to RNA also terminated reverse transcription and in vitro translation at the exact site of PNA binding independent of RNase H activity. Furthermore, microinjection of the 15-mer into the nucleus of cells expressing SV 40 T antigen inhibited T antigen expression. A homo-thymidine 10-mer has also been used to target the terminator elements of three yeast class III genes, which are vital to RNA polymerase III (pol III) recycling and thus transcription efficiency . In vitro transcription assays revealed nanomolar inhibition of the genes in supercoiled (but not linear) plasmid constructs . Furthermore, PNA concentrations that inhibited multiple rounds of transcription had no effect on the absolute amount of RNA output for a single transcription cycle, suggesting the inhibition of pol III recycling caused by a strand invasion induced ''roadblock'' to the terminator. Unmodified PNAs are capable of crossing the blood-brain barrier in rats upon intraperitoneal injection . Following intraperitoneal injection, PNAs that bind to the mRNA and DNA of rat neurotensin receptor (NTR1) were demonstrated to inhibit gene transcription and exert behavioral effects in rat specimens .
Peptide-linked PNAs have also been designed with antitranscriptional activities. A 17-mer PNA with an appended nuclear localization signal (NLS) peptide successfully localized to the nuclei of Burkitt's lymphomas cells and caused rapid down regulation (35% decrease in 7 h) of c-myc oncoprotein expression . Peptide-linked PNAs have also been used to inhibit the expression of human caveolin 1 in both cultured HeLa and primary endothelial cells . A 9-mer PNA conjugated to a nitrogen mustard suppressed HER-2/neu oncogene expression by 80% in intact HeLa cells . Appendage of an alkylating agent presumably facilitates strand invasion and ultimately stabilizes triple helix formation. A trifunctional PNA-peptide-diethylenetriamine conjugate showed sequence-specific RNA cleavage in vitro and has potential as an artificial cell-penetrating ri-bonuclease . A PNA prodrug complementary to the hepatic human microsomal triglyceride transfer protein (huMTP) is rapidly internalized by HepG2 cells due to the appendage of GalNAc sugars . Also, the PNA conjugates accumulate in the parenchymal liver cells of mice to a far greater extent than nonconju-gated PNAs following intravenous injection. The MTP mRNA levels in HepG2 cells was consequently down regulated by 35-40% at 100 nM.
PNAs have been demonstrated to act as artificial transcription promoters in vitro [214, 215] and in vivo [215, 217]. RNA polymerases and transcription factors initiate transcription through sequence-specific recognition of 12 base pair loop structures at the promoter sites of partially melted double stranded DNA (dsDNA). Similar loop structures evolve upon binding of PNAs to dsDNA, as the resulting (PNA)2-DNA complexes display stable D-loop structures on the template strand resembling those of transcription initiation sites [78, 87]. Loops formed by strand invasion of homopyrimidine PNAs are recognized by RNA polymerase which can initiate transcription at PNA-binding sites with efficiencies comparable to that of the robust E. coli lacUV5 promoter . However, the use of PNAs as gene promoters in vivo has been limited due to impaired cell delivery. Induction of the g-globin gene, a therapeutic target for sickle cell anemia, has been achieved both in vivo and in vitro according to studies with reporter gene constructs . Induction of the endogenous gene was achieved in K562 human erythroleukemia cells after introduction into cells by electroporation. Also, PNA length-dependence studies demonstrated that PNAs 14-20 nucleotides long can induce high levels of transcription in a HeLa nuclear extract in vitro transcription system . Furthermore, transfection of these same PNAs bound to GFP reporter gene plasmids into human normal fibroblast (NF) cells could induce GFP translation in vivo.
The RNA component of the ribonucleoprotein enzyme complex human telo-merase is accessible to incoming nucleic acids, and has thus been recognize as a suitable anticancer target. The RNA component of telomerase has previously been targeted by complementary DNA , phosphorothioate DNA , and 2'-O-methyl RNAs , however the inhibitory efficiency of these agents was limited by poor sequence selectivity and bioavailability. Therefore, peptide nucleic acids designed to complement the RNA component of telomerase may provide an alternative with more desirable pharmacokinetics.
Dose-dependent reduction of telomerase activity in cell extracts, tumors, and permeabilized cells has been demonstrated by PNAs . Assays with HME50-5 cell extracts revealed several PNAs of 11 with nanomolar IC50 values (0.9-10 nM). The most potent inhibitor also showed an IC50 of 50 nM towards permeabilized cells. In comparison to control phosphorothioate DNA oligomers, PNAs demonstrated 10-50-fold increase in binding affinity and increased sequence specificity. The appendage of cationic peptides to 11-mer and 13-mer PNAs were shown to increase cell permeabilization in pretreated melanoma cells, with IC50 values in the submicromolar range (360 nM) .
Was this article helpful?