Photocontrolled DNA minor groove interactions of imidazole/pyrrole polyamides

Article abstract of DOI:10.3762/bjoc.16.8

Azobenzenes are photoswitchable molecules capable of generating significant structural changes upon E-to-Z photoisomerization in peptides or small molecules, thereby controlling geometry and functionality. E-to-Z photoisomerization usually is achieved upo

Full text of DOI:10.3762/bjoc.16.8

Photocontrolled DNA minor groove interactions of
imidazole/pyrrole polyamides
Sabrina Müller1, Jannik Paulus1, Jochen Mattay2, Heiko Ihmels3, Veronica I. Dodero1
and Norbert Sewald*1
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2020, 16, 60–70.
1Organic and Bioorganic Chemistry, Department of Chemistry,
Bielefeld University, PO Box 100131, D-33501 Bielefeld, Germany,
2Organic Chemistry I, Department of Chemistry, Bielefeld University,
PO Box 100131, D-33501 Bielefeld, Germany and 3Organic
Chemistry II, Department Chemistry – Biology, Siegen University,
Adolf-Reichwein-Str. 2, D-57068 Siegen, Germany
doi:10.3762/bjoc.16.8
Received: 26 October 2019
Accepted: 20 December 2019
Published: 09 January 2020
Associate Editor: K. N. Allen
Email:
Norbert Sewald* - norbert.sewald@uni-bielefeld.de
© 2020 Müller et al.; licensee Beilstein-Institut.
License and terms: see end of document.
* Corresponding author
Keywords:
azobenzene; DNA minor groove; N-methylimidazole; N-methylpyrrole;
photoswitches; polyamide
Abstract
Azobenzenes are photoswitchable molecules capable of generating significant structural changes upon E-to-Z photoisomerization in
peptides or small molecules, thereby controlling geometry and functionality. E-to-Z photoisomerization usually is achieved upon ir-
radiation at 350 nm (π–π* transition), while the Z-to-E isomerization proceeds photochemically upon irradiation at >400 nm (n–π*
transition) or thermally. Photoswitchable compounds have frequently been employed as modules, e.g., to control protein–DNA
interactions. However, their use in conjunction with minor groove-binding imidazole/pyrrole (Im/Py) polyamides is yet unprece-
dented. Dervan-type Im/Py polyamides were equipped with an azobenzene unit, i.e., 3-(3-(aminomethyl)phenyl)azophenylacetic
acid, as the linker between two Im/Py polyamide strands. Only the (Z)-azobenzene-containing polyamides bound to the minor
groove of double-stranded DNA hairpins. Photoisomerization was exemplarily evaluated by 1H NMR experiments, while minor
groove binding of the (Z)-azobenzene derivatives was proven by CD titration experiments. The resulting induced circular dichro-
ism (ICD) bands of the bound ligands, together with the photometric determination of the dsDNA melting temperature, revealed a
significant stabilization of the DNA upon association with the ligand. The (Z)-azobenzene acted as a building block inducing a
reverse turn, which favored hydrogen bonds between the pyrrole/imidazole amide and the DNA bases. In contrast, the E-config-
ured polyamides did not induce any ICD characteristic for minor groove binding. The incorporation of the photoswitchable azoben-
zene unit is a promising strategy to obtain photoswitchable Im/Py hairpin polyamides capable of interacting with the dsDNA minor
groove only in the Z-configuration.
60

Beilstein J. Org. Chem. 2020, 16, 60–70.
Introduction
The development of small chemical agents to modulate gene [14-20]. Mascareñas et al. reported the first photoisomerizable
expression in an organism has attracted considerable interest in transcription factor (Tf) that recognized its target sequences by
molecular medicine [1]. However, the missing selectivity of major groove recognition [21,22]. Woolley et al. reported other
chemical agents, such as chemotherapeutics, often results in photoisomerizable Tf mimetics that interacted through the
undesirable cytotoxicity, harming healthy cells, and thereby in- major groove [23,24]. Hybridization of DNA [25-27] or
ducing a multitude of side effects. A major breakthrough con- PNA/DNA [28] can be light-controlled by azobenzene-modi-
cerning selective small DNA-targeting molecules was the use of fied DNA. While DNA covalently modified with azobenzene
pyrrole/imidazole hairpin polyamides. They are conceptually moieties was proven amenable for the light-induced triggering
derived from the natural products netropsin and distamycin A, of transcription [24], photoswitchable polyamides that can act
which selectively bind to particular sequences of the DNA as selective DNA minor groove binders displaying tunable
duplex [2-4]. Covalent tethering of two antiparallel polyamide affinity have not yet been reported [29,30].
segments increases the sequence specificity and the affinity of
the polyamides to their cognate dsDNA. Different linker strate- We envisaged that control of the geometry and functionality
gies were used, with γ-aminobutyric acid (γ) being the most could be achieved upon replacement of the γ-aminobutyric acid
successful representative. The resulting hairpin polyamides linker between the two hairpin strands by a photoswitchable
bound with 100-fold higher affinity than the corresponding azobenzene motif. As the azobenzene moiety undergoes a con-
untethered moieties, with the reverse turn module showing siderable steric reorganization upon isomerization from the E-
selectivity for AT base pairs [5]. The reason for this selectivity to the Z-configuration, the end-to-end distance in a 4,4'-disub-
was the steric repulsion with the exocyclic amino function of stituted azobenzene changes from 9 Å to 5.5 Å upon E-to-Z
guanine that pointed inside the DNA groove [2]. The group of isomerization [31,32]. If we consider the minor groove of
Dervan has systematically increased the number of binding double-stranded B-DNA to have a width of approximately
motifs, and thus achieved sequence-specific binding [6-9]. 5.7 Å [21], the Z-isomer of a 4,4’-azobenzene-tethered Im/Py
Selectively binding polyamides adopt an antiparallel hairpin polyamide would fit into the minor groove, while the two
structure where a base pair of the DNA is addressed by a pair of polyamide strands establish direct contacts to the bottom of the
the heterocyclic amino acids in the hairpin assembly. Overall, cognate minor groove. However, previous studies on
specific binders have been developed for all possible base pair- distamycin A-like derivatives containing only pyrrole moieties
ings (AT, TA, GC, CG; 'pairing rules'). A GC pair is specifi- and 4,4'-substituted azobenzene as a photoswitchable linker
cally being addressed in the minor groove by N-methylimida- showed that the binding properties of the Z- and E-isomers were
zole/N-methylpyrrole (Im/Py), while AT associates with the quite similar. Hence, there was no switch in binding activity
pair N-methylpyrrole/N-methylhydroxypyrrole (Py/Hp). The [29,30]. Importantly, it was demonstrated that the DNA binding
same applies to CG (Py/Im) and TA (Hp/Py). However, was highly dependent on the linker length between the azoben-
N-methylhydroxypyrrole (Hp) is neither easy to synthesize nor zene and the distamycin moiety.
sufficiently stable. In practice, therefore, Py/Py is used to
address both AT and TA.
Results and Discussion
Here, we report on the design, synthesis, and characterization of
Along this line, it is attractive to incorporate a molecular switch photoswitchable minor groove binders based on pyrrole/imida-
for the selective activation of an initially inactive substance, zole hairpin polyamides functionalized with an azobenzene unit
once located at the DNA target, with an external stimulus. in which the Z-isomer is sufficiently stable to allow for accu-
Azobenzenes are photoswitchable molecules capable of gener- rate measurements. Im/Py polyamides bind to the minor groove
ating significant structural changes in peptides or small mole- of DNA in a sequence-specific manner, encoded by antiparallel
cules, thereby controlling their geometry and functionality upon side-by-side pairs of pyrrole and imidazole carboxamides with
irradiation [10-13]. While isomerization of the E- to the Z-form higher binding affinities than the pyrrole analogs alone [33,34].
usually takes place at 350 nm irradiation (π–π* transition), the We selected 3-(3-(aminomethyl)phenyl)azophenylacetic acid as
reverse is photochemically induced at 450 nm (n–π* transition) the linker between both polyamide strands because it was
or thermally. The rate of thermal relaxation, inter alia, depends shown to induce hairpins in peptides upon E-to-Z photoisomer-
on the substituents, and only derivatives with very slow ther- ization [35,36]. According to molecular dynamics calculations,
mal relaxation are suitable as bi-stable switches.
the 3,3'-substituted azobenzenes are more suitable as photo-
switchable building blocks to induce a hairpin motif than the
Azobenzene and other small photoreactive molecules have been 4,4'-substituted correlates [35]. For 3,3'-substituted azoben-
employed as ligands to control DNA or RNA assembly by light zenes, the Z-form is expected to display higher thermal stability
61

Beilstein J. Org. Chem. 2020, 16, 60–70.
than for the 4,4'-substituted correlates, as the substituents are ucts was detected by MALDI–TOF MS. This irreversible
not conjugated to the N=N double bond [35,37-39]. Moreover, N-guanylation of the polyamide N-terminus resulted from slow
it was demonstrated that in the case of peptides, only the carboxy activation of the building block by HBTU and the pres-
Z-isomer adopts the hairpin conformation. In our setup, this ence of an excess of HBTU and could not be completely
could influence the hydrogen bonding contacts between the prevented [44,45]. Since this side reaction did not occur with
pyrrole/imidazole amide with DNA bases, helping to achieve phosphonium salts, P1 and P3 were successfully obtained by
proper shape complementarity only when the system is in the using PyBOP as an activating reagent [45,46]. After final Fmoc
Z-configuration. Therefore, this azobenzene represents a prom- cleavage, the polyamides were released from the resin with a
ising approach to replace the γ-aminobutyric acid linker in the solution of 2.5% TFA in dichloromethane. The free C-terminus
Im/Py polyamide systems.
of the polyamides was modified in solution with N,N-dimethyl-
aminopropylamine by using PyBOP as an activating reagent to
install the corresponding end group. The completeness of all
reaction stages was checked by MALDI–TOF MS. Purification
Design and synthesis of photoswitchable
polyamides
From a pool of new azobenzene-containing polyamides, we of the final products by preparative reversed-phase HPLC gave
present the most promising photoswitchable representatives the TFA salts of the polyamides P1, P2, and P3 in 10%, 2%,
containing six or eight Im/Py units (P1, P2, and P3, Scheme 1). and 6% yield, respectively (see Experimental section).
Their binding capability to dsDNA was analyzed by titration
and temperature-dependent circular dichroism experiments.
E-to-Z photoconversion
The functional properties and photoconversion of 3-(3-(amino-
The assembly of the photoswitchable pyrrole/imidazole methyl)phenyl)azophenylacetic acid were already established
polyamides (P1–P3) was performed stepwise on solid phase previously [35,36,39]. NMR spectroscopy was employed to
using the acid-labile 2-chlorotrityl resin and Fmoc as a tempo- quantify the photoconversion of the azobenzene-containing
rary protecting group, according to the synthetic route published polyamides P1–P3 as well as the stability of the Z-isomer. For
by the Dervan group [40]. The synthesis of the Fmoc-protected this purpose, two proton signals were selected, which were not
photoswitchable ω-amino acid 1 (Scheme 2A) was performed superimposed by other signals and where the chemical shift
based on a procedure by Aemissegger et al. [35]. The ω-amino changed as a result of the switching process. The amide proton
acid 1 was subsequently used in the solid-phase synthesis or for of the azobenzene moiety at 8.7 ppm and the adjacent methy-
the synthesis of the dimer 7 in solution (Scheme 2B).
lene group at 4.5 ppm met these requirements, and hence were
used for analysis (Figure 1). The proportions of the individual
The incorporation of β-alanine as a 'molecular spring' was re- species in the mixture were determined by integration of the
quired because this allowed for an alignment between hydro- corresponding signals. A representative series of spectra is
gen-bonding groups in long polyamides and in the minor presented in Figure 1 for polyamide P1. In the thermal equilib-
groove of DNA [41].
rium, approximately 6% of the Z-isomer was present
(Figure 1A).
The Fmoc-protected heterocyclic amino acids 2 were obtained
from N-methylpyrrole and N-methylimidazole, respectively Upon irradiation at 350 nm, this fraction steadily increased and
(Scheme 2A). The N-terminal N-methylpyrrole and N-methyl- reached a level of 65% after 50 min irradiation (Figure 1B and
imidazole units were introduced by employing diamino deriva- Figure 1C). However, the Z-isomer content could not be in-
tives 3 (Scheme 2A) [40]. Moreover, it turned out that coupling creased further by UV irradiation. The Z-isomer could subse-
the subsequent building block to an N-terminal Im building quently be switched back to the E-isomer (19% Z-configuration
block was problematic because the amino group of the imida- after 30 min) by irradiation at 420 nm (Figure 1D). The thermal
zole derivative 2b is a weak nucleophile, and therefore the conversion to the E-isomer in the dark was slow at room tem-
Fmoc–Py–Im–OH dimer 5 was obtained [42,43]. Owing to the perature, and even after 18 days, 13% of the Z-isomer were still
poor coupling results of Fmoc–Py–OH to the amino function of present. The thermal equilibrium (5% Z-configuration) was
a terminal Im moiety, the Fmoc–Azo–Im–OH dimer 7 was syn- reached only after 40 days in the dark (data not shown).
thesized in 73% yield by coupling to the amine 4, as shown in
Scheme 2B.
In a previous study, the same azobenzene species incorporated
in a hairpin peptide reached a photostationary state of 85%
HBTU as a coupling reagent and DIPEA as a base were em- Z-isomer [35]. The photostationary state depends on the nature
ployed in DMF for the synthesis of P2. However, in the case of of the azobenzene, but also on the concentration, as the azoben-
P1 and P3, the formation of tetramethylguanidinium side prod- zene is known to form supramolecular assemblies at high con-
62

Beilstein J. Org. Chem. 2020, 16, 60–70.
Scheme 1: Pyrrole–imidazole–azobenzene polyamides and the dsDNA target sequences employed in this study.
centration, compromising photoconversion [11,12]. When the tively) was reached after 15 min irradiation at 350 nm. It has
concentration of polyamide P1 was reduced to 0.8 mM, 80% of been reported that the E-to-Z isomerization of 3,3’-substituted
the Z-isomer were obtained after 15 min of irradiation azobenzene compounds proceeds substantially slower than for
(Figure 2). In the case of P2 and P3, similar results were ob- 4,4’-substituted azobenzene analogs [36]. Moreover, the solu-
tained, and the photostationary Z-state (83% and 80%, respec- tions were kept in the dark to evaluate the thermal stability of
63

Beilstein J. Org. Chem. 2020, 16, 60–70.
Scheme 2: Building blocks required for the synthesis of the photoswitchable Im/Py polyamides. A) Fmoc–Azo–OH 1, N-methylpyrrole and N-methyl-
imidazole monomers 2–4, and Fmoc–Py–Im–OH 5. B) Synthesis of the dimer Fmoc–Azo–Im–OH 7: a) DCC, THF, rt, 5 min, then 4, rt, 15 h, 41%;
b) BF3∙Et2O, dichloromethane, 0 °C→rt, 12 h, 73%.
Figure 1: Section of the 1H NMR (600 MHz) spectrum of polyamide P1. A) Initial thermal equilibrium. B) After irradiation at 350 nm for 10 min.
C) After irradiation at 350 nm for 50 min. D) After irradiation at 420 nm for 30 min (in DMSO-d6, c = 8 mM).
the Z-isomer, and it was noted that after one or two days, the dsDNA [47-49]. The B-DNA has a positive CD band at
percentage of Z-isomer was slightly reduced (after 2 d: 62% P1; 260–280 nm and a negative band at about 245 nm [47]. In
67% P2; after 1 d: 76% P3).
contrast, achiral ligands, such as the polyamides, do not give a
CD signal. However, by associating the achiral ligand with
chiral DNA, an induced circular dichroism (ICD) signal can be
detected at wavelengths longer than 300 nm where dsDNA does
Analysis of polyamide–DNA interaction by in-
duced circular dichroism
CD spectroscopic analysis is suitable for the characterization of not absorb. Therefore, the occurrence of ICD is a strong indica-
DNA minor groove binders, providing semiquantitative infor- tion of the interaction between ligand and DNA [50,51]. This
mation about binding affinity and thermal stability of the target ICD signal originates from the coupling of the transition dipole
64

Beilstein J. Org. Chem. 2020, 16, 60–70.
Figure 2: E/Z isomer ratio of the polyamides P1–P3. Values were obtained from the respective 1H NMR experiments (c = 0.80 mM) after irradiation
(λ = 350 nm, t = 15 min) followed by evaluation after one or two days at 5 °C in the dark. For details, see Supporting Information File 2.
moments of the nucleobases and the ligand and is usually posi- PA:hairpin DNA from 0 to 4.8, where saturation was reached
tive for polyamide groove binders on B-DNA at about 335 nm for Z-P1 and Z-P3 at the ratio 1:3.3. Considering that only the
[47,52]. Initially, it was difficult to predict, which sequences 1:1 PA:dsDNA binding mode was possible for our setup, the
would be optimal for the new polyamides containing the observed behavior was indicative of dynamic binding at the
azobenzene linker. Hence, we employed the target sequences concentration tested. Furthermore, for these compounds, hyper-
for the Im/Py hairpin polyamides based on Dervan’s pairing chromicity (increase in signal intensity) and hypsochromic
rules, such as 5´-TTGTC-3´ for P1, 5´-TTGTT-3´ for P2, and shifts of the B-DNA maximum at ≈275 nm and minimum at
5´-TTGTCA-3´ for P3 (Scheme 1). Furthermore, as proof of ≈250 nm were detected upon addition of polyamide.
concept, short hairpin oligonucleotide sequences were em-
Considering that PAs do absorb light at wavelengths < 300 nm,
the observed changes cannot be unambiguously attributed to the
DNA because the ICD of the PA is combined with the intrinsic
CD spectrum of the oligonucleotide. Nevertheless, the hyper-
chromicity of the signal on increasing PA concentration and the
spectral behavior observed were characteristic for a B-DNA
structure [51,53].
ployed to investigate the interaction of the polyamides with
dsDNA because six- and eight-membered polyamides can only
bind in a 1:1 mode, and thus simplify the analysis [41,42,47].
Figure 3 demonstrates that only (Z)-polyamides show the char-
acteristic positive ICD signal with the dsDNA sequences tested.
The position of the maximum of the positive ICD signal in-
creased with the length of the polyamide (Z-P1 and Z-P2:
λmax = 325 nm; Z-P3: λmax = 333 nm), and a new negative band
at 280–310 nm was observed, too. The intensity of these new
signals increased with increasing concentration of the
polyamides. Titrations were performed at molar ratios
In contrast, Z-P2 did not reach saturation in the tested concen-
tration range (Figure 3B), which is indicative of a lower binding
affinity in comparison with the other PAs. For Z-P2,
65

Beilstein J. Org. Chem. 2020, 16, 60–70.
Figure 3: Titration experiments of target DNA sequences with P1–P3 in the photostationary Z-state and the thermal E-state. See Table 1 for the com-
plete sequences.
hypsochromic shifts of the B-DNA maximum at ≈275 nm and that only the (Z)-azobenzene moiety is able to induce the
the minimum at ≈250 nm upon addition of polyamide were polyamide hairpin conformation required for binding. We also
accompanied by slightly reduced hyperchromicity. The appear- tested the interaction of the polyamides upon single base muta-
ance of clear-cut isodichroic points in the CD titration (red tion in the dsDNA sequences to explore the recognition speci-
arrows in Figure 3) suggests the formation of one dominant ficity. Specifically, we tested the sequence 5´-TTGTT-3´ for
type of DNA/PA complex, which, together with the positive P1, 5´-TTGTC-3´ for P2, and 5´-TTGTTA-3´ for P3 and ob-
ICD bands observed at wavelengths > 300 nm, indicate minor tained essentially similar results. Only the Z-isomers were able
groove binding [51,53]. Similar results have been obtained by to bind, while the E-isomers did not show any interaction with
Wang et al., who evaluated the interaction between hairpin the DNA minor groove (Figure 4).
Im/Py PA and short hairpin dsDNA sequences [51-53]. Strik-
ingly, no positive ICD effect on the CD spectrum was observed While the complexes of Z-P1 and Z-P2 with their target DNA
upon addition of the E-configured polyamides to the same (D1 and D2) and single-point mutation DNA (D1’ and D2’) had
dsDNA sequences, which indicates that they did not bind to the similar ICD spectra, the complex of Z-P3 with the single-point
minor groove of the short dsDNA. Importantly, no precipita- mutation DNA D3’ showed only a blue shift of the B-DNA
tion was observed during the measurement. We hypothesize maximum at ca. 275 nm, but no shift of the minimum at
66

Beilstein J. Org. Chem. 2020, 16, 60–70.
Figure 4: Titration of DNA containing single mutations (in bold) with P1–P3 in the photostationary Z-state and the thermal E-state. See Table 1 for
complete sequences.
≈250 nm, lacking this isodichroic point (cf. Figure 3C and quences D1–D3. However, considering their polyamide struc-
Figure 4C). Finally, Z-P3 did not reach saturation with the ture, they should at least exhibt a weak affinity toward double-
single-point mutation DNA D3’ at the tested concentration stranded DNA. The fluorescent indicator displacement (FID)
range, suggesting lower binding affinity compared to target assay was performed with E-P1 as representative ligand to ex-
dsDNA D3.
amine the propensity of these derivatives to bind to DNA. The
known intercalator thiazole orange (TO) was chosen as indi-
The maxima of the ICD signals at 325 and 333 nm were plotted cator because it exhibits an intense fluorescence band at 526 nm
against the concentration of polyamide to quantify the binding when bound to DNA, whereas it is only weakly fluorescent in
of the PA to the cognate dsDNA. The dissociation constants solution. The displacement of TO from its DNA binding site
KD, calculated from the binding isotherms, were in the micro- was monitored by emission spectroscopy. The addition of
molar range (11.8 6.0–52.9 29.6 μM). The binding affinity 2.3 equiv of E-P1 led to a displacement of 50% of TO (cf. Sup-
decreased in the order P1 > P3 > P2 for the target sequences porting Information File 3), which indicates only a weak
and P1 > P2 > P3 for the single-point mutation (Table 1).
binding of E-P1 because as a groove binder, the latter occupies
more binding sites than TO. These data indicate that E-P1 does
The CD spectroscopic experiments indicate that the ligands bind to DNA in general, but the affinity of this ligand was obvi-
E-P1–E-P3 did not bind significantly to the oligonucleotide se- ously too low to bind strongly enough to the duplexes D1–D3,
67

Beilstein J. Org. Chem. 2020, 16, 60–70.
Table 1: Binding constants and melting temperatures of the interaction of the polyamides with dsDNA obtained by circular dichroism.
recognition sequence (5´→3´)
KD/μM
Tm/°C
(DNA)a
Tm/°C
(PA/DNA)b
ΔTm/K
PA:DNA
Z-P1
CATTGTCAGCCTTTTTGGCTGACAATG D1
CATTGTTAGCCTTTTTGGCTAACAATG D1‘
11.8 6.0
13.2 7.8
Z-P2
72.2
66.5
75.7
68.6
3.5
2.1
3.3
3.3
CATTGTTAGACTTTTTGTCTAACAATG D2
CATTGTCAGACTTTTTGTCTGACAATG D2‘
52.3 21.0
42.2 6.3
Z-P3
65.4
68.7
67.5
70.1
2.0
1.4
3.9
3.6
CATTGTCATGCCTTTTTGGCATGACAATG D3
CATTGTTATGCCTTTTTGGCATAACAATG D3‘
21.3 11.2
52.9 29.6
72.9
69.6
76.4
70.0
3.5
0.4
3.3
3.7
ac (dsDNA) = 8.87 μM. bFor the Z-isomers, the employed ratio is given. For specific experimental conditions see Experimental section.
as indicated by the lack of any significant induced circular di- was obtained for Z-P2 (ΔTm = 2 K). On the other hand, for the
chroism signal.
single-point mutation, the Tm values were lower: ΔTm = 2.1 K
for Z-P1, ΔTm = 1.4 K for Z-P2, and ΔTm = 0.4 K for Z-P3.
We tested the stability of each dsDNA by determining their These results are in agreement with the data of the titration ex-
melting curve during thermal DNA denaturation in the absence periments and indicate a higher sequence specificity, e.g., of
or the presence of P1–P3 under otherwise identical experimen- Z-P3. The highest binding affinity (lowest KD) was obtained for
tal conditions for further information on the PA/DNA interac- Z-P1, with a slight preference for the target DNA sequence
tion (see Experimental section). If a polyamide is a minor compared to the single mismatch sequence. Binding affinity as
groove binder, this affects the melting temperature of the well as duplex stabilization was lowest for Z-P2 for both
dsDNA [34,47,53-55]. The DNA double strand is stabilized by dsDNA sequences tested (Table 1). The affinity of Z-P3 to the
additional hydrogen bonds between ligand and oligonucleotide, target D3 was twofold higher than to D3’ with the single muta-
and hence the melting temperature is expected to increase. This tion sequence. However, this value is still very much different
effect is of particular interest when considering the sequence from the 29-fold selectivity observed for the γ–Im/Py
selectivity of the polyamides, which should therefore also be re- polyamide analogs of P3 obtained by DNase footprinting exper-
flected in a change of the melting temperature. The melting iments [3]. The DNA sequence selectivity observed for Z-P3
curves after the addition of polyamide were recorded at the was underlined by an increase of ΔTm = 3.5 K observed for the
highest concentration of polyamide used in the titration experi- interaction with the target sequence D3 compared to
ments (Table 1). Since the hyperchromicity of DNA increases ΔTm = 0.4 K for D3’.
the absorption of ssDNA at 260 nm compared to that of
dsDNA, denaturation can be observed by CD, too. The melting Conclusion
temperature (Tm) corresponds to the point of inflection where We were able to show for the first time that designed hetero-
50% of the DNA molecules are single-stranded. Tm depends on cyclic polyamides P1–P3, equipped with a photoisomerizable
the number and strength of intramolecular interactions between azobenzene, bind to double-stranded DNA hairpins in the
the single strands, providing information on stabilization or Z-configuration. In this case, the (Z)-azobenzene acted as a
destabilization of the DNA duplex in the presence of the corre- building block inducing a reverse turn. Minor groove binding
sponding PA. In general, the more energy needed to break the was proven by CD titration experiments and melting tempera-
hydrogen bonds between the single strands of the duplex and ture determinations. The E-configured azobenzene polyamides
the stacking interactions of the bases, the higher the melting did not induce ICD assigned to the polyamide chromophores,
temperature. Hence, it also increases with higher GC content. possibly because of the formation of small soluble aggregates.
The melting temperature Tm of the dsDNA increased in the Z-P3, with a higher number of heterocyclic units, displayed
presence of the three polyamides Z-P1, Z-P2, and Z-P3, al- some sequence specificity, albeit not in a range that was re-
though it is also possible that during heating, thermally induced ported for other nonphotoswitchable polyamides. The (Z)-
isomerization to the E-isomer occurred (Table 1). The highest azobenzene linker may be sterically more demanding than the
stabilization value of the corresponding target sequences was γ-aminobutyric acid linker, which would cause a dilation of the
obtained for Z-P1 and Z-P3 (ΔTm = 3.5 K). The lowest value minor groove, and thus reducing the binding affinity and selec-
68

Beilstein J. Org. Chem. 2020, 16, 60–70.
References
tivity of the tested sequences. The previously reported analog of
P3 with γ-aminobutyric acid instead of azobenzene has been
used to exert genomic effects on mRNA expression [56]. There-
fore, the photoactivation of such interaction may be a future ap-
plication. The incorporation of the photoswitchable 3-((3-
(aminomethyl)phenyl)diazenyl)phenylacetic acid linker upon
replacement of the γ-aminobutyric acid linker is a powerful
strategy to obtain photoswitchable Im/Py hairpin polyamides
capable of interaction with the dsDNA minor groove only in the
Z-configuration.
1. Rahman, A.; O'Sullivan, P.; Rozas, I. Med. Chem. Commun. 2019, 10,
26–40. doi:10.1039/c8md00425k
2. Mrksich, M.; Parks, M. E.; Dervan, P. B. J. Am. Chem. Soc. 1994, 116,
7983–7988. doi:10.1021/ja00097a004
3. Greenberg, W. A.; Baird, E. E.; Dervan, P. B. Chem. – Eur. J. 1998, 4,
796–805.
doi:10.1002/(sici)1521-3765(19980515)4:5<796::aid-chem796>3.0.co;
2-g
4. Bhaduri, S.; Ranjan, N.; Arya, D. P. Beilstein J. Org. Chem. 2018, 14,
1051–1086. doi:10.3762/bjoc.14.93
5. Tse, W. C.; Boger, D. L. Chem. Biol. 2004, 11, 1607–1617.
doi:10.1016/j.chembiol.2003.08.012
6. Kielkopf, C. L.; White, S.; Szewczyk, J. W.; Turner, J. M.; Baird, E. E.;
Dervan, P. B.; Rees, D. C. Science 1998, 282, 111–115.
doi:10.1126/science.282.5386.111
7. Dervan, P. B. Bioorg. Med. Chem. 2001, 9, 2215–2235.
doi:10.1016/s0968-0896(01)00262-0
Supporting Information
Supporting Information File 1
Experimental section.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-8-S1.pdf]
8. Dervan, P. B.; Edelson, B. S. Curr. Opin. Struct. Biol. 2003, 13,
284–299. doi:10.1016/s0959-440x(03)00081-2
9. Kielkopf, C. L.; Baird, E. E.; Dervan, P. B.; Rees, D. C. Nat. Struct. Biol.
1998, 5, 104–109. doi:10.1038/nsb0298-104
Supporting Information File 2
NMR spectra of P1–P3 and dimers 6 and 7.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-8-S2.pdf]
10.Brieke, C.; Rohrbach, F.; Gottschalk, A.; Mayer, G.; Heckel, A.
Angew. Chem., Int. Ed. 2012, 51, 8446–8476.
doi:10.1002/anie.201202134
11.Sequeira, M. A.; Herrera, M. G.; Quirolo, Z. B.; Dodero, V. I. RSC Adv.
2016, 6, 108132–108135. doi:10.1039/c6ra20933e
12.Quirolo, Z. B.; Sequeira, M. A.; Martins, J. C.; Dodero, V. I. Molecules
2019, 24, No. 2508. doi:10.3390/molecules24132508
13.Heinrich, B.; Bouazoune, K.; Wojcik, M.; Bakowsky, U.; Vázquez, O.
Org. Biomol. Chem. 2019, 17, 1827–1833. doi:10.1039/c8ob02343c
14.Bergen, A.; Rudiuk, S.; Morel, M.; Le Saux, T.; Ihmels, H.; Baigl, D.
Nano Lett. 2016, 16, 773–780. doi:10.1021/acs.nanolett.5b04762
15.Dohno, C.; Uno, S.-n.; Nakatani, K. J. Am. Chem. Soc. 2007, 129,
11898–11899. doi:10.1021/ja074325s
Supporting Information File 3
Photoisomerization of P1–P3 and TO displacement
experiments.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-8-S3.pdf]
Supporting Information File 4
HRMS spectra of P1–P3.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-8-S4.pdf]
16.Dohno, C.; Yamamoto, T.; Nakatani, K. Eur. J. Org. Chem. 2009,
4051–4058. doi:10.1002/ejoc.200900323
17.Basak, A.; Mitra, D.; Kar, M.; Biradha, K. Chem. Commun. 2008,
3067–3069. doi:10.1039/b801644e
18.Wang, X.; Huang, J.; Zhou, Y.; Yan, S.; Weng, X.; Wu, X.; Deng, M.;
Zhou, X. Angew. Chem., Int. Ed. 2010, 49, 5305–5309.
doi:10.1002/anie.201002290
19.Sánchez, M. I.; Vázquez, O.; Vázquez, M. E.; Mascareñas, J. L.
Chem. Commun. 2011, 47, 11107–11109. doi:10.1039/c1cc13355a
20.Iqbal, E. S.; Richardson, S. L.; Abrigo, N. A.; Dods, K. K.;
Osorio Franco, H. E.; Gerrish, H. S.; Kotapati, H. K.; Morgan, I. M.;
Masterson, D. S.; Hartman, M. C. T. Chem. Commun. 2019, 55,
8959–8962. doi:10.1039/c8cc10192b
Acknowledgements
The authors gratefully acknowledge experimental assistance by
Anke Nieß and Marco Wißbrock and support by the NMR spec-
troscopy and the mass spectrometry facilities of Bielefeld
University.
21.Boga, S.; Bouzada, D.; García Peña, D.; Vázquez López, M.;
Vázquez, M. E. Eur. J. Org. Chem. 2018, 249–261.
doi:10.1002/ejoc.201700988
22.Caamaño, A. M.; Vázquez, M. E.; Martínez-Costas, J.; Castedo, L.;
Mascareñas, J. L. Angew. Chem., Int. Ed. 2000, 39, 3104–3107.
doi:10.1002/1521-3773(20000901)39:17<3104::aid-anie3104>3.0.co;2-
0
23.Zhang, F.; Timm, K. A.; Arndt, K. M.; Woolley, G. A.
Angew. Chem., Int. Ed. 2010, 49, 3943–3946.
doi:10.1002/anie.201000909
24.Morgan, S.-A.; Woolley, G. A. Photochem. Photobiol. Sci. 2010, 9,
1320–1326. doi:10.1039/c0pp00214c
Funding
The authors gratefully acknowledge financial support by the
Deutsche Forschungsgemeinschaft (SE609/13-1 and IH24/13-
1).
ORCID® iDs
Jochen Mattay - https://orcid.org/0000-0002-1511-7960
Heiko Ihmels - https://orcid.org/0000-0003-0969-0426
Veronica I. Dodero - https://orcid.org/0000-0001-7937-1880
Norbert Sewald - https://orcid.org/0000-0002-0309-2655
69

Beilstein J. Org. Chem. 2020, 16, 60–70.
25.Asanuma, H.; Ito, T.; Yoshida, T.; Liang, X.; Komiyama, M.
Angew. Chem., Int. Ed. 1999, 38, 2393–2395.
51.Šmidlehner, T.; Piantanida, I.; Pescitelli, G. Beilstein J. Org. Chem.
2018, 14, 84–105. doi:10.3762/bjoc.14.5
doi:10.1002/(sici)1521-3773(19990816)38:16<2393::aid-anie2393>3.0.
co;2-7
52.Qiao, H.; Ma, C.; Zhang, X.; Jing, X.; Li, C.; Zhao, Y.
Bioconjugate Chem. 2015, 26, 2054–2061.
26.Liang, X.; Mochizuki, T.; Asanuma, H. Small 2009, 5, 1761–1768.
doi:10.1002/smll.200900223
27.Asanuma, H.; Liang, X.; Nishioka, H.; Matsunaga, D.; Liu, M.;
Komiyama, M. Nat. Protoc. 2007, 2, 203–212.
doi:10.1021/acs.bioconjchem.5b00309
53.Wang, S.; Aston, K.; Koeller, K. J.; Harris, G. D., Jr..; Rath, N. P.;
Bashkin, J. K.; Wilson, W. D. Org. Biomol. Chem. 2014, 12,
7523–7536. doi:10.1039/c4ob01456a
doi:10.1038/nprot.2006.465
28.Stafforst, T.; Hilvert, D. Angew. Chem., Int. Ed. 2010, 49, 9998–10001.
doi:10.1002/anie.201004548
54.Lacy, E. R.; Cox, K. K.; Wilson, W. D.; Lee, M. Nucleic Acids Res.
2002, 30, 1834–1841. doi:10.1093/nar/30.8.1834
55.Mergny, J.-L.; Lacroix, L. Oligonucleotides 2003, 13, 515–537.
doi:10.1089/154545703322860825
56.Supekova, L.; Pezacki, J. P.; Su, A. I.; Loweth, C. J.; Riedl, R.;
Geierstanger, B.; Schultz, P. G.; Wemmer, D. E. Chem. Biol. 2002, 9,
821–827. doi:10.1016/s1074-5521(02)00174-6
29.Ghosh, S.; Usharani, D.; Paul, A.; De, S.; Jemmis, E. D.;
Bhattacharya, S. Bioconjugate Chem. 2008, 19, 2332–2345.
doi:10.1021/bc800130u
30.Ghosh, S.; Usharani, D.; De, S.; Jemmis, E. D.; Bhattacharya, S.
Chem. – Asian J. 2008, 3, 1949–1961. doi:10.1002/asia.200800151
31.Renner, C.; Kusebauch, U.; Löweneck, M.; Milbradt, A. G.; Moroder, L.
J. Pept. Res. 2005, 65, 4–14. doi:10.1111/j.1399-3011.2004.00203.x
32.Wachtveitl, J.; Zumbusch, A. ChemBioChem 2011, 12, 1169–1170.
doi:10.1002/cbic.201100185
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
33.Trauger, J. W.; Baird, E. E.; Dervan, P. B. Nature 1996, 382, 559–561.
doi:10.1038/382559a0
(https://creativecommons.org/licenses/by/4.0). Please note
that the reuse, redistribution and reproduction in particular
requires that the authors and source are credited.
34.Chenoweth, D. M.; Meier, J. L.; Dervan, P. B. Angew. Chem., Int. Ed.
2013, 52, 415–418. doi:10.1002/anie.201205775
35.Aemissegger, A.; Kräutler, V.; van Gunsteren, W. F.; Hilvert, D.
J. Am. Chem. Soc. 2005, 127, 2929–2936. doi:10.1021/ja0442567
36.Kräutler, V.; Aemissegger, A.; Hünenberger, P. H.; Hilvert, D.;
Hansson, T.; van Gunsteren, W. F. J. Am. Chem. Soc. 2005, 127,
4935–4942. doi:10.1021/ja044253u
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
37.Ulysse, L.; Cubillos, J.; Chmielewski, J. J. Am. Chem. Soc. 1995, 117,
8466–8467. doi:10.1021/ja00137a023
38.Nishimura, N.; Sueyoshi, T.; Yamanaka, H.; Imai, E.; Yamamoto, S.;
Hasegawa, S. Bull. Chem. Soc. Jpn. 1976, 49, 1381–1387.
doi:10.1246/bcsj.49.1381
39.Dong, S.-L.; Löweneck, M.; Schrader, T. E.; Schreier, W. J.; Zinth, W.;
Moroder, L.; Renner, C. Chem. – Eur. J. 2006, 12, 1114–1120.
doi:10.1002/chem.200500986
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.16.8
40.Wurtz, N. R.; Turner, J. M.; Baird, E. E.; Dervan, P. B. Org. Lett. 2001,
3, 1201–1203. doi:10.1021/ol0156796
41.Trauger, J. W.; Baird, E. E.; Dervan, P. B. Chem. Biol. 1996, 3,
369–377. doi:10.1016/s1074-5521(96)90120-9
42.Su, W.; Gray, S. J.; Dondi, R.; Burley, G. A. Org. Lett. 2009, 11,
3910–3913. doi:10.1021/ol9015139
43.Wetzler, M.; Wemmer, D. E. Org. Lett. 2010, 12, 3488–3490.
doi:10.1021/ol1013262
44.Arttamangkul, S.; Arbogast, B.; Barofsky, D.; Aldrich, J. V.
Lett. Pept. Sci. 1997, 3, 357–370. doi:10.1007/bf00127967
45.Albericio, F.; Bofill, J. M.; El-Faham, A.; Kates, S. A. J. Org. Chem.
1998, 63, 9678–9683. doi:10.1021/jo980807y
46.Coste, J.; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990, 31,
205–208. doi:10.1016/s0040-4039(00)94371-5
47.Dodero, V. I.; Quirolo, Z. B.; Sequeira, M. A.
Front. Biosci., Landmark Ed. 2011, 16, 61–73. doi:10.2741/3676
48.Dodero, V. I.; Mosquera, M.; Blanco, J. B.; Castedo, L.;
Mascareñas, J. L. Org. Lett. 2006, 8, 4433–4436.
doi:10.1021/ol061551i
49.Jiménez-Balsa, A.; Dodero, V. I.; Mascareñas, J. L. Tetrahedron 2013,
69, 7847–7853. doi:10.1016/j.tet.2013.05.126
50.Chang, Y.-M.; Chen, C. K.-M.; Hou, M.-H. Int. J. Mol. Sci. 2012, 13,
3394–3413. doi:10.3390/ijms13033394
70