A New Design Strategy and Diagnostic to Tailor ... - ACS Publications

Sep 26, 2016 - ABSTRACT: The classical model for DNA groove binding states that groove binding molecules should adopt a crescent shape that closely ...
0 downloads 0 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

A New Design Strategy and Diagnostic to Tailor the DNABinding Mechanism of Small Organic Molecules and Drugs Phi H. Doan, Demar R. G. Pitter, Andrea Kocher, James N Wilson, and Theodore Goodson III ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00448 • Publication Date (Web): 26 Sep 2016 Downloaded from http://pubs.acs.org on October 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

A New Design Strategy and Diagnostic to Tailor the DNA-Binding Mechanism of Small Organic Molecules and Drugs Phi Doan,† Demar R. G. Pitter,‡ Andrea Kocher,† James N. Wilson,‡ Theodore Goodson III†* † ‡

Department of Chemistry, University of Michigan, Ann Arbor, MI 48109 Department of Chemistry, University of Miami, Coral Gables, FL 33146

Corresponding Authors * E-mail: [email protected]

ABSTRACT The classical model for DNA groove binding states that groove binding molecules should adopt a crescent shape that closely matches the helical groove of DNA. Here, we present a new design strategy that does not obey this classical model. The DNA-binding mechanism of small organic molecules was investigated by synthesizing and examining a series of novel compounds that bind with DNA. This study has led to the emergence of structure-property relationships for DNA-binding molecules and/or drugs, which reveals that the structure can be designed to either intercalate or groove bind with calf thymus dsDNA by modifying the electron acceptor properties of the central heterocyclic core. This suggests that the electron accepting abilities of the central core plays a key factor in the DNA-binding mechanism. These small molecules were characterized by steady-state and ultrafast nonlinear spectroscopies. Bio-imaging experiments were performed in live cells to evaluate cellular uptake and localization of the novel small molecules. This report paves a new route for the design and development of small organic molecules, such as therapeutics, targeted at DNA as their performance and specificity is dependent on the DNA-binding mechanism.

INTRODUCTION Cancer is the leading disease that results in the most deaths in the United States.1 Despite extensive research on chemotherapy, there remains considerable interest in establishing design strategies for therapeutics aimed at DNA. Anti-cancer drugs that target DNA are some of the most effective drugs used in cancer chemotherapy.2 Consequently, significant effort has been placed in the development of new drugs that are more selective and less toxic. Active compounds 1 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

that are aimed at biological targets, such as DNA, are referred to as targeted therapeutics because they can bind and influence specific molecular targets.3 DNA has become one of the major biomolecular targets for cancer because it is required for the proliferation of cancer cells. Increased efforts are now focused on rational approaches, including design strategies, to control the DNA-binding mode of small molecules. This provides a new means to structurally modify drugs to enhance performance and selectivity based on the DNA-binding mechanism. Small DNA-binding molecules can interact with DNA predominately through an intercalating or groove binding mechanism. In general, intercalating molecules have planar aromatic rings incorporated into the structure that bind with DNA by inserting between the DNA nucleobases to form a binding pocket. Lengthening, unwinding, and distortion of the DNA helical axis are most prominent upon intercalation,4 which requires important driving factors, such as π-stacking, dipole-dipole interaction, electrostatic factors, and dispersive interaction with the aromatic nucleobases in DNA.5 On the other hand, groove binding is characterized by little to no perturbation in the DNA structure. Groove binding molecules generally contain unfusedaromatic structures with terminal basic functions.6 Additionally, groove binding molecules require conformational flexibility that allow the molecule to fit into the DNA groove and functional groups that interact with the nucleobases through H-bonding and/or van der Waals interactions with minimal steric hindrance.7 Groove depth, groove width, electrostatic potential, and floor functionality are structural features found to be critical for groove binding recognition.8 Understanding the mechanisms of the DNA-binding interactions of small molecules can provide valuable information towards the development of therapeutics targeted at DNA. For example, common effects of DNA intercalative drugs include inhibition of cell growth, cell transformation, and cell death, which have applications as antitumor, antibacterial, and antiparasitic agents9 while common effects of groove binding drugs involve interferences with cellular processes by targeting enzyme and protein access to DNA.10 Hence, structure-property relationships are of considerable interest to guide in the design of small organic molecules in which the DNA-binding mechanism can be modified and tailored. These relationships will contribute to the design of therapeutics aimed at DNA. Qualitative methods have been employed to elucidate the DNA-binding modes of small organic molecules. However, a combination of select conventional methods must be implemented to determine the DNA-binding mode while avoiding ambiguity.11 Therefore, new 2 ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

techniques that provide detailed information about the DNA-binding mechanism are of interest. Two-photon absorption (TPA) can yield valuable information inaccessible with steady-state spectroscopy. The TPA cross-section is directly proportional to the square of the transition dipole moment and change of the static dipole moment after excitation.12 Thus, TPA can detect changes in environmental conditions, charge transfer character, and excited-dipoles with high sensitivity.13–22 Recently, we developed a highly sensitive methodology utilizing TPA spectroscopy to diagnose the DNA-binding mode of small organic molecules unambiguously by analyzing the change in the TPA cross-section upon complexing with DNA.23 The change in the TPA cross-section is based on the DNA electric field influence on the transition dipole of the binding molecule and/or drug upon binding with DNA. Groove binding and intercalating molecules will exhibit different orientations relative to the DNA helical axis. As a consequence, groove binding molecules will have a dipole aligned more parallel to the DNA electric field resulting in an increase in the TPA cross-section upon binding with DNA. On the other hand, an intercalating molecule will have a dipole oriented more perpendicular to the DNA electric field leading to a decrease in the TPA cross-section upon complexing with DNA. A series of small organic molecules that bind with DNA were investigated. These molecules will serve as a model system to rationalize structure-property relationships for future drug development. We report a design strategy that allows the DNA-binding mode to be tailored in the case of small organic molecules that adopt a crescent or V-shaped scaffold. According to the classical model for DNA groove binding, molecules that possess a crescent or curved scaffold should groove bind with DNA because they can closely fit or match the groove of the DNA helix. This is significant as our design strategy does not obey the classical model. The combined approach utilizing ultrafast nonlinear and steady-state spectroscopies were employed to establish structure-property relationships. Herein, the results reveal that a change in the DNA-binding mode can be achieved by modifying the electron acceptor properties of the central heterocyclic core of the binding molecule. Our recently developed methodology based on TPA was implemented to examine the DNA-binding mode of these small molecules. The fluorophores were imaged in live HeLa cells by confocal microscopy to correlate cellular uptake and localization to the established structure-property relationships. The impact of this work is significant as the design strategy can be applied towards the development of drugs targeted at DNA. 3 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

RESULTS AND DISCUSSION Design. The fluorophores (Figure 1) were designed with three key elements: (i) Recognition units were incorporated into the pendant arms of the fluorophores. Since DNA is a polyanionic molecule, cationic recognition arms are necessary for the fluorophore to electrostatically interact with the DNA backbone, which plays an important role in the formation of DNA-drug complexes.24 N-methylpiperazine was selected because protonation of the pendant arms is achieved at physiological conditions. (ii) The pedant aryl arms were designed to control the optical switching of fluorescence. Quenching is noted when the pendant arms are allowed free rotation in the absence of DNA, resulting in an ultrafast non-radiative deactivation. The pendant arms of the fluorophores are restricted of rotation when bound with DNA, which leads to a planar conformation. As a consequence, a large fluorescence enhancement originates from the loss of rotation of the pendant arms due to the constrictive DNA environment as well as a reduction in the non-radiative deactivation pathways upon binding with DNA. (iii) The donoracceptor-donor π-system motif is responsible for the optical properties. The incorporation of heterocyclic cores with varying electron acceptor properties into the conjugated π-system leads to an enhancement in the photophysical properties, which allows for a variety of molecular designs. OH N

F F B O

N

O

1

5

N

N

N N

N

N

OH N

N

N

F F B O

O

N

6

2 N

N

N

N N

N

N

N OH N

F F B O

N

O

7

3 N

N

N

N N

N

N

N

NC

N

N O

4 N N

CN

N

N

8

N

N

4 ACS Paragon Plus Environment

N N

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 1. Chemical structures of the fluorophores. The fluorophores adopt a donor-acceptordonor motif with varying conjugation length between the electron donor arms and electron acceptor core. UV-Vis Absorption. The steady-state absorption spectra of the fluorophores in the absence of DNA are presented in Figure 2a. As expected, the fluorophores (1, 4, 5) with shorter conjugation length displayed the shortest wavelength absorption maxima while the fluorophores (3, 7, 8) with increased conjugation length on both pendant arms exhibited the longest wavelength absorption maxima. The fluorophores displayed a single absorption band, which can be attributed to the π – π* transition. In the case of the fluorophores with a difluoroboron βdiketonate or dicyanomethylene pyran heterocyclic core (5 – 8), a second less intense absorption band at a lower wavelength was noted, which can be ascribed to the σ – σ* transition. Interestingly, 7 had the broadest absorption band with a bathochromic shift of 63 nm compared to 5. A direct comparison of the fluorophores with (3, 7, 8) and without (1, 4, 5) increased conjugation length on the pendant arms will allow us to assess the effect of the electron accepting abilities of the heterocyclic central core on the absorption spectra in the absence of DNA. The fluorophores (1, 4, 5) without increased conjugation length on both pendant arms will be discussed first. As expected, the addition of an electron donating hydroxyl substituent to the pyrimidine core (1) resulted in a bathochromic shift of 39 nm compared to 4. The shift can be ascribed to both the electron acceptor properties of the core as well as the increase in conjugation length. Comparing with 4, a bathochromic shift of 124 nm was achieved by replacing the pyrimidine core with a difluoroboron β-diketonate core as demonstrated with 5. The fluorophores (3, 7, 8) with increased conjugation length on both pendant arms were also investigated. 3 exhibited a maximum absorption band near 435 nm. 8 displayed a bathochromic shift of 20 nm compared to 3 while 7 exhibited a bathochromic shift of 80 nm relative to 8. The absorption spectra of the fluorophores were also examined in the presence of DNA (Figure 2b). Bathochromic shifts of 2 – 4 nm were observed in the absorption maxima upon DNA binding for the fluorophores (1, 4, 5) without increased conjugation length on both pendant arms, which was the smallest shift noted. Interestingly, the fluorophores (3, 7, 8) with increased conjugation length on both pendant arms exhibited the largest bathochromic shifts of 12 – 20 nm upon complexing with calf thymus dsDNA. It is reasonable to suggest that the difference in the 5 ACS Paragon Plus Environment

ACS Chemical Biology

absorption shift upon binding with DNA can be directly related to the conjugation length. The fluorophores with increased conjugation length on the pendant arms exhibit more conformational isomers compared to the fluorophores without increased conjugation length. A larger bathochromic shift is observed since the fluorophores with increased conjugation length on both pendant arms result in a larger reduction of conformational isomers upon complexing with DNA. The bathochromic shift can also be ascribed to the change in the environment of the fluorophore when bound with DNA.25 Upon complexing with DNA, the solvation of the fluorophore displaces water molecules, leading to a more hydrophobic environment that prevents fluorescence quenching and provides steric protection.26 b

1 2 3 4 5 6 7 8

1.00

0.75

0.50

Normalized Absorption

Normalized Absorption

a

0.25

0.00 300

400

500

600

700

0.75

0.50

0.25

0.00 300

800

1 2 3 4 5 6 7 8

1.00

400

Wavelength (nm)

c 0.8 0.6 0.4 0.2 0.0 400

500

600

700

600

700

d

1 2 3 4 5 6 7 8

1.0

500

800

Wavelength (nm)

Normalized Intesnity

Normalized Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

0.8 0.6 0.4 0.2 0.0 400

800

1 2 3 4 5 6 7 8

1.0

500

600

700

800

Wavelength (nm)

Wavelength (nm)

Figure 2. Normalized absorption spectra of the fluorophores in the (a) absence and (b) presence of DNA. Normalized emission spectra of the fluorophores in the (c) absence and (d) presence of DNA. Steady-State Fluorescence. The emission spectra of the fluorophores were investigated in the absence of DNA (Figure 2c). The fluorophores displayed weak fluorescence in buffered 6 ACS Paragon Plus Environment

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

aqueous solution with the fluorescence quantum yield ranging on the order of 10-4 to 10-2. In particular, the fluorophores with increased conjugation length on both pendants arms exhibited the lowest quantum yields. Similar to the absorption spectrum, shifts in the emission band were also noted by varying the conjugation length and electron acceptor properties of the heterocyclic core. By comparing 1, 4, and 5, the difluoroboron β-diketonate core possesses the strongest electron accepting abilities, leading to the most bathochromically shifted maximum emission band near 549 nm as noted for 5. It has been reported that the vacant p-orbital from the boron atom overlaps with the conjugated moiety, which leads to a decrease in the lowest unoccupied molecular orbital (LUMO), resulting in a bathochromic shift in both the absorption and emission spectra.27 A similar trend was also observed by comparing 3, 7, and 8. The results indicate that the optical properties of the fluorophores can be finely tuned by varying the electron acceptor properties of the central heterocyclic core as well as the conjugation length. The modifications in the optical properties can be directly attributed to the change in the highest occupied molecular orbital (HOMO) and/or LUMO. Presented in Figure 2d, the emission spectra of the fluorophores were examined in the presence of DNA. With the exception of 8, the fluorophores displayed a fluorescence fold increase of up to 74-fold as well as narrowing of the emission band upon binding with DNA. The broader emission band can be attributed to a large distribution of conformational isomers of the free fluorophore in solution.25 In particular, the emission band of 3 is broad compared to the remaining complexes. It is important to note that the small peak near 515 nm arises from the Raman scattering of the solvent. The broadening of the emission band can be ascribed to several equilibrated excited state rotational conformational isomers of the unbound fluorophore.25 In the presence of DNA, the broad emission band was not observed. Narrowing of the emission band implies that the fluorophore adopts a more planar conformation upon binding with DNA, which restricts free rotation of the pendant arms. As a consequence, there is a reduction of conformational isomers leading to fluorescence enhancement.

7 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

Table 1. Summary of steady-state properties of the fluorophores in the absence (top) and presence of DNA (bottom). ε (M-1 cm-1)

523

Stokes Shift (cm-1) 6720

36,500

0.005

598

7370

37,400

0.012

435

619

6680

36,600

7 x 10-4

4

348

514

9320

39,900

0.006

5

472

549

2970

29,000

0.009

6

508

647

4230

43,000

0.005

7

535

696

4320

27,200

2 x 10-4

8

455

661

6850

42,500

0.002

Compound

λmax, abs (nm)

λmax, em (nm)

1

387

2

415

3

Φ

ε (M-1 cm-1)

Φ

501

Stokes Shift (cm-1) 5790

32,700

0.14

Fluorescence Fold Increase 73.8

423

586

6490

30,100

0.15

31.2

3

448

620

6140

26,300

0.02

61.1

4

352

476

7400

29,600

0.10

29.0

5

474

555

3180

23,800

0.06

28.5

6

521

642

3590

44,600

0.18

39.3

7

555

682

3310

32,900

0.007

33.7

8

467

659

5260

29,200

0.004

1.3

Compound

λmax, abs (nm)

λmax, em (nm)

1

389

2

Circular Dichroism (CD) Spectroscopy. CD has been widely used to study the function and interactions of DNA complexes in buffered aqueous solution.28 CD was employed to gain an insight on the DNA-binding mechanism. The CD spectra of the fluorophores in the presence of DNA were examined (see Supplementary Information). With the exception of 8, an ICD signal was detected, which confirms DNA-dye binding interactions. A positive ICD signal was observed for 1 – 4, which suggests that the fluorophores are groove binding with DNA. Additionally, this indicates that the transition dipole of the fluorophore is oriented along the groove.29 Interestingly, a bisignate ICD signal was observed for 1 – 7, which is due to the interactions between the fluorophore that reflect excitonic coupling in dimers or aggregates formed at the groove or surface of DNA.30 1 and 4 were previously found to groove bind with 8 ACS Paragon Plus Environment

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

DNA, which corresponds with the results.23 In a previous report, 1 was found to groove bind at AT-rich sequences. However, the fluorescence is quenched from GC sequences, which is likely due to the photoinduced electron transfer (PET) from guanine to the excited chromophore.31 The CD spectra were investigated in case of the fluorophores with a difluoroboron βdiketonate core (5 – 7). 5 was previously reported to intercalate with DNA utilizing linear dichroism (LD).32 Unexpectedly, 5 exhibited a strong positive ICD band near 475 nm, which implies that the fluorophore is groove binding with DNA. However, when a positive ICD signal arises from an intercalator, this indicates that the transition dipole is oriented along the long axis of DNA whereas a negative ICD signal is observed when the transition dipole is aligned perpendicular with the long axis of the DNA structure.33 Based on the CD results, it is reasonable to suggest that the transition dipole of 5 – 7 is oriented parallel with the long axis of DNA. However, it is difficult to determine the DNA-binding mode based solely on CD. Two-Photon Absorption (TPA). The TPA cross-sections (δ) of the unbound fluorophores are summarized in Table 2. The TPA cross-section increased as conjugation length increased in the case of the fluorophores with the same central core. For example, the TPA cross-section of 1 was 3.4 GM. Increasing conjugation length to a single pendant arm resulted in a TPA crosssection of 6.1 GM as detailed with 2 while increasing conjugation length to both pendant arms led to a TPA cross-section of 66.8 GM as shown with 3. A similar trend was also noted for 5 – 7. Interestingly, there was a significant enhancement in the TPA cross-section when a stronger electron acceptor core was incorporated into the structure in the case of the fluorophores with increased conjugation length on a single pendant arm. For instance, 2 had a TPA cross-section of 6.1 GM. A dramatic increase in the TPA cross-section was found by replacing the hydroxypyrimidine core with a difluoroboron β-diketonate core as demonstrated with 6, which displayed a TPA cross-section of 37.5 GM. The findings suggest that the electron acceptor properties of the core and conjugation length play an important role in the TPA cross-section of the unbound fluorophore. Table 2. TPA cross-section (δ) of the fluorophores in the absence of DNA. The TPA crosssections were measured with [fluorophore] = 5 μM in PBS. δ (GM) 3.4 6.1 66.8 2.1

Compound 1 2 3 4

9 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

1.4 37.5 50.6 87.6

5 6 7 8

TPA was employed to assess the DNA-binding mode of the fluorophores as described in a previous report where an overall increasing or decreasing trend in the TPA cross-section indicates a groove binding or intercalating binding mode, respectively.23 The TPA cross-sections were plotted as a function of DNA concentration as given in Figure 3. 1 and 4 were previously found to display an increasing TPA cross-section trend, suggesting a groove binding mode 23 The increasing TPA cross-section trend implies that the dipole of the fluorophore is oriented more parallel with the DNA electric field, which is indicative of a groove binder. Interestingly, the TPA cross-section enhancement of 4 is significantly greater than 1. 1 exhibited a TPA crosssection enhancement of 4.1-fold whereas 4 displayed a TPA cross-section enhancement of 13.6fold upon binding with DNA. Since the structures are nearly identical, this can be attributed to the hydroxyl substituent on the heterocyclic core of 1. It has been shown that intermolecular interactions can influence the TPA cross-section.34 Additionally, it has been reported that Hbonding interactions and aggregation can reduce the TPA cross-section of V-shaped hydroxypyrimidine complexes.35 Hence, the difference in the TPA cross-section enhancement upon binding with DNA is directly related to the hydroxyl substituent on 1 interacting with the DNA nucleobases and/or surrounding water molecules.

10 ACS Paragon Plus Environment

Page 11 of 30

a

b

15

30 25

δ (GM)

δ (GM)

12

9

20 15 10

6

5 3

0 0

100

200

300

400

500

0

600

100

[DNA] (µM)

c

200

300

400

500

600

[DNA] (µM)

d

4

40

δ (GM)

3

δ (GM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

2

30 20

1

10 0

0 0

100

200

300

400

500

0

600

100 200

300 400

500

600

[DNA] (µM)

[DNA] (µM)

Figure 3. TPA cross-section (δ) plotted as a function of DNA concentration. The TPA crosssection was measured for (a) 1,23 (b) 4,23 (c) 5, (d) 6 with [fluorophore] = 5 μM at increasing DNA concentrations to determine the DNA-binding mode. The red line is to guide the eye. 5 was previously shown to intercalate with DNA utilizing LD.32 It has been reported that CD cannot determine the binding mode unambiguously without the orientation of the transition dipole moment being known.7 In order to determine the DNA-binding mode unambiguously, TPA was employed. Our TPA results suggest that both 5 and 6 intercalate with DNA as observed with the decreasing TPA cross-section trend. For example, 6 had a TPA cross-section of 37.5 GM in the absence of DNA. The TPA cross-section decreased to 8.6 GM at 624 μM of DNA. The overall decreasing TPA cross-section trend indicate that the transition dipole of both 5 and 6 is oriented more perpendicular to the DNA electric field, which is a characteristic of an intercalating binding mode. Interestingly, the TPA cross-section of 5 increased from 1.4 to 3.5 GM when the DNA concentration was increased from 0 to 2.1 μM. This can be directly

11 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

attributed to the fluorophore stacking or aggregating at the surface of DNA at high dye-to-DNA ratios.36 Another increase in the TPA cross-section was noted for 5 when increasing the DNA concentration from 125 to 624 μM. This is likely due to the dye forming dimers at the surface or groove of the DNA structure. Based on our findings, it is suggested that 5 and 6 intercalate with DNA, which does not necessarily agree with the CD interpretation. Time-Resolved Fluorescence. The fluorescence lifetime of the fluorophores in the absence of DNA was investigated to better understand the excited-state dynamics (see Supplementary Information). The fluorophores displayed a two-component decay, suggesting that multiple excited states exist. The fluorescence lifetimes are summarized in Table 3. The multi-exponential decay indicates that there are two deactivation pathways, which includes the emission from the locally excited (LE) state and the intramolecular charge transfer between the fluorescent LE state to the non-fluorescent twisted intramolecular charge transfer (TICT) state (LE → TICT transition) by twisting of the molecular moieties.37 The short component is likely due to an intramolecular charge transfer from the fluorescent LE to the non-fluorescent TICT state. When the pendant arms are allowed free rotation, the rate of the LE → TICT transition process may compete with other radiative and non-radiative processes.38 By hindering this process or restricting the pendant arms, the LE → TICT transition is suppressed, leading to enhanced emission and a longer decay rate. Interestingly, the short component of the fluorophores is near 2 ps with the exception of 8. Because of this, it is reasonable to suggest that the fast component may also be ascribed to the photoisomerization of the fluorophore. It has been reported that photoisomerization of cis-stilbene is 1 ps in n-hexane.39 This indicates that the photoisomerization activation barrier is minimal. The longer component is not likely due to solvent or vibrational relaxation, but to the excited-state population dynamics of the free fluorophore. In other words, this is due to the population decay of the charge transferred state coupled with the triplet state.40 The predominant long component can be attributed to the LE state or the deactivation of the LE state to the ground state. The fluorescence decay dynamics were investigated to analyze the effect of conjugation length. Increasing π-conjugation length to one pendant arm led to a longer-lived fluorescence lifetime as observed with 2 and 6. For example, 2 exhibited a long component of 45.1 ps, which was the longest lifetime noted. The longer-lived fluorescence lifetime can be directly ascribed to the asymmetrical configuration of the fluorophore. The fluorophores have two charge transfer 12 ACS Paragon Plus Environment

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

excited states that are different in energy because of the asymmetric pathway. Thus, there is an increase in the ICT character due to the asymmetric structural configuration. It has been previously reported that the fluorescence lifetime increases as the ICT character is increased, which corresponds with our results.37 The shortest lifetimes were observed for the fluorophores with increased π-conjugation length on both pendant arms. These observations are consistent in the case of the fluorophores that possess a pyrimidinol or difluoroboron β-diketonate core. 3 exhibited a long component of 16.4 ps while 7 had a long component of 7.4 ps. The low quantum yield and fast decay of 3 and 7 can be directly related to the increase in conjugation length on both pendant arms. Increasing conjugation length results in the delocalization of the excited state. Photoinduced alterations in the dipole occur when the π-conjugated system connecting the donor and acceptor moieties is lengthened.41 The insertion of double bonds between the donor and acceptor moieties leads to a decrease in the overall aromaticity of the system resulting in a reduced band gap. Additionally, this leads to an increase in the rigidification of the fluorophore. As a consequence, a faster fluorescence lifetime is observed.

Table 3. Summary of the fluorescence lifetime data given in the absence (top) and presence of DNA (middle). The fluorescence lifetime values were fitted to the equation y(t) = A1 e−t/τ1 + A2 e−t/τ2 . The rotational correlation decay parameters of the fluorophores are presented (bottom). Compound 1 2 3 4 5 6 7 8

A1 0.55 0.38 0.71 0.62 0.50 0.32 0.58 0.56

τ1 (ps) 2.0 1.7 1.4 2.3 1.1 2.4 0.8 4.0

A2 0.45 0.62 0.29 0.38 0.50 0.68 0.42 0.44

τ2 (ps) 22.3 45.1 16.4 30.6 10.6 43.3 7.4 27.3

Compound 1 2 3 4 5 6 7 8

A1 0.05 0.11 0.13 0.05 0.14 0.08 0.05 0.16

τ1 (ns) 0.57 0.40 0.52 0.55 0.33 0.35 0.32 0.34

A2 0.95 0.89 0.87 0.95 0.86 0.92 0.95 0.84

τ2 (ns) 2.76 2.65 1.33 3.66 2.64 2.30 1.08 1.90

Compound 1

Absence of DNA τ (ps) ro 31.8 0.39

13 ACS Paragon Plus Environment

Presence of DNA τ (ps) ro 5194 0.19

ACS Chemical Biology

104.0 85.4 57.3 11.3 24.7 85.4 N/A

2 3 4 5 6 7 8

0.31 0.19 0.35 0.34 0.25 0.12 0.12

2020 4880 5806 452 1079 2200 2276

Normalized Intensity

a

0.16 0.17 0.27 0.05 0.03 0.11 0.07

1 2 3 4 5 6 7 8

1.0 0.8 0.6 0.4 0.2 0.0 0

2

4

6

8

Time (ns) 0.4

c

0.3

Anisotropy

b

Anisotropy

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

0.2 0.1

0.4 0.3 0.2 0.1 0.0

0.0 0

2

4

6

0

8

2

4

6

8

Time (ns) Time (ns) Figure 4. (a) Fluorescence lifetime of the fluorophores in the presence of DNA. Fluorescence anisotropy (b) 1 and (c) 5 in the presence of DNA.

TCSPC was employed because of the increase in the fluorescence lifetime of the DNA bound fluorophore (Figure 4a). The fluorescence lifetime in the presence of DNA is summarized in Table 3. A multi-component decay in the nanosecond time range was recorded. A short compound was observed ranging between 0.32 ns and 0.57 ns. A long component was noted ranging between 1.08 and 3.66 ns. The short component is likely attributed to the dye 14 ACS Paragon Plus Environment

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

aggregating or stacking at the surface of DNA or the unbound dye.42,43 The long component is directly due to the DNA bound dye in the planar conformation. To further understand the fluorescence dynamics of the unbound fluorophores, ultrafast fluorescence anisotropy decay measurements were conducted (see Supplementary Information). The rotational anisotropy decay values are summarized in Table 3. Anisotropy measurements can provide detailed information about the directionality of the material. A higher anisotropy implies that the material is more symmetrical and less freely moving.44 Interestingly, the anisotropy value decreases as conjugation length increased. This indicates that the highest direction dependence is observed for the fluorophores without increased conjugation length regardless of the electron acceptor properties of the central heterocyclic core. This observation can also be attributed to the increase in the rotational diffusion and/or energy transfer processes. Specifically, the fluorophores with increased conjugation length reorient faster after excitation. Anisotropy values can also provide information about the transition dipole of the fluorophore. A high anisotropy (near 0.40) indicates that the emission is not delocalized and is subsequent from localized transition dipoles,45 which was noted for 1 (0.39). With the exception of 8, the fluorophores exhibit anisotropy decay in the absence of DNA. This suggests that the angle between the absorption and emitting dipoles does not change during intersystem crossing for 8.45 In other words, the emission from 8 has equal intensities along the different axes of polarization. The DNA-binding interactions of the bound fluorophores were investigated by polarized excitation measurements. The bound fluorophore exhibits different excited states that lead to altering depolarization times. The rotational correlation times of the bound fluorophore were longer than the local motion of the free fluorophore under the same conditions, indicating a restricted tumbling motion (see Supplementary Information). For example, 1 had an anisotropic decay of 31.8 ps in the absence of DNA while the anisotropic decay was 5.2 ns when bound with DNA. The values are summarized in Table 3. Both groove binding and intercalation inhibit the rapid tumbling motion of the fluorophore upon binding with DNA. High residual anisotropy was not observed for 5 – 7 in the presence of DNA. On the other hand, high residual anisotropy was noted in the case of 1 – 4, which implies that the fluorophores exhibit highly restricted rotation upon binding with DNA. The anisotropy for 1 – 7 can be ascribed to the locking of the fluorophore due to the protonated N-methylpiperazine recognition units electrostatically interacting with the negatively charged DNA helix. In addition, the 15 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

heterocyclic core of the fluorophore undergoes intermolecular interactions with the DNA nucleobases to further stabilize the binding molecule. It has been reported that intercalating molecules can form aggregates, which are disrupted upon the addition of DNA to the system.46 This disruption of the aggregates for 5 – 7 likely results in the non-high residual anisotropy at high DNA concentrations as the rotation of the fluorophore is not as restricted. The high residual anisotropy for 1 – 4 is attributed to the local reorientation and geometric restrictions of the bound fluorophore at the microenvironment due to torsional and bending of the DNA structure.47 In other words, this high residual anisotropy may reflect the overall motion of DNA as the fluorophore binds at the groove of DNA to form a highly geometric restricted complex.43 Live Cell Imaging. To explore the utility of the fluorophores, a series of imaging experiments were carried out in live HeLa cells. The fluorophores demonstrated a dramatic increase in fluorescence upon binding with DNA, which makes them suitable for bio-imaging applications. Cellular uptake of the fluorophore was verified by confocal microscopy. The fluorophores displayed excellent cellular uptake. Cell morphology was not found during the bioimaging experiments. Single-photon fluorescence confocal microscopy was utilized to image the fluorophores (see Supplementary Information). 1, 2, and 4 demonstrated distinguishable nuclei staining with high selectivity for DNA with less pronounce staining of the cytoplasm. The images displayed specific staining of DNA localized in the cell nucleus with high signal-to-noise ratio. 3 also exhibited high selectivity for DNA localized in the nucleus; however, staining of the cytoplasm was more pronounced. To confirm these observations, the cells were stained with 1 μM of 3. Staining of the cytoplasm was noted at the lower fluorophore concentration. 5 – 8 showed vastly different staining trends as the fluorophore was mainly localized in the cytoplasm. This is likely due to the increase in the electron acceptor properties of the heterocyclic core causing the fluorophore to penetrate the nucleus membrane with lower efficiency. From the steady-state studies, 8 was not expected to label the nuclei with a high fluorescent enhancement. It should be emphasized that localized bright spots were observed in the cytoplasm, which is likely attributed to direct staining of organelles. It was previously reported that cationic molecules can be used as fluorescent probes for lysosomes, endoplasmic reticulum, cell membrane, and mitochondria.48 Therefore, these bright spots localized in the cytoplasm may constitute as labeled organelles, such as lysosomes or mitochondria.

16 ACS Paragon Plus Environment

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Presented in Figure 5, two-photon excited microscopy (TPEM) was utilized to image the fluorophores in live cells. TPEM offers several advantages over traditional one-photon fluorescence microscopy (OPEM) as it demonstrates superior spatial resolution and sensitivity. For example, the features of a single cell are remarkably pronounced by TPEM compared to OPEM. A significant portion of the fluorescence collected with single-photon microscopy arises from autofluorescence due to the higher energy excitation wavelength. Autofluorescence is less pronounced with TPEF since the excitation wavelength is near the IR range, which also minimizes photodamage. The excitation for TPEF is confined to a small volume in a focal plane, leading to reduced photobleaching as well as deeper tissue penetration capabilities for imaging applications. Thus, fluorophores that exhibit TPA properties are of interest. 1, 2, and 4 displayed optical properties that are promising as biological markers utilizing TPEM. The labeled nuclei are less pronounced with OPEM whereas the contrast and resolution is more dramatic by TPEM. Cell permeable small organic molecules that can target specific DNA sequences and interfere with gene expression are important for drug research and development.49 Hence, imaging of such processes with high spatial resolution is of considerable interest.

a

b

c

d

e

f

g

h

Figure 5. Two-photon excited fluorescence confocal microscopy images. Live HeLa cells stained with (a) 1, (b) 2, (c) 3, (d) 4, (e) 5, (f) 6, (g) 7, (h) 8. The cells were incubated at 2 μM of fluorophore for 1 h. λex = 800 nm. Bar: 20 μm.

17 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

Discussion. The investigated small molecules provide a unique sample to elucidate the DNA-binding mechanism. Although design criteria have been reported for groove binding and intercalating drugs,4,50,51 it has yet to be demonstrated that a change in the DNA-binding mechanism can be achieved in the case of small molecules that possess similar structural scaffolds. A common feature shared by these small molecules is that they adopt a crescent shaped donor-acceptor-donor motif. Based on the structure-property relationships, we propose a design strategy to tailor the DNA-binding mechanism. By controlling the DNA-binding mode, the binding preferences of drugs targeted at DNA can be tailored. This provides a more effective route to design selective drugs that exhibit high performance. One of the most challenging tasks in drug discovery is developing drugs that demonstrate high specificity. Our results indicate that a change in the DNA-binding mechanism can be achieved by simply modifying the electron acceptor properties of the core, which will allow the specificity and localization of the drug to be controlled. CD spectroscopy was employed to gain an insight on the DNA-binding mechanism of the fluorophores. Overall, the positive ICD bands correspond with the absorption bands. Interestingly, the fluorophores displayed a bisignate shape in the presence of DNA with the exception of 8. Because of this, it is difficult to determine the DNA-binding mode with solely the use of CD. A major issue with CD is that the DNA-binding mode cannot be determined unambiguously without the inclusion of other techniques.7 In previous reports, it has been noted that common methodologies, such as CD and LD, to evaluate the DNA-binding must be made with caution because a combination of selected methods only provides sufficient information to determine the DNA-binding mode.4,11 In order to better understand the DNA-binding mechanism, our recently developed TPA methodology was implemented. To assess the DNA-binding mode, TPA measurements were carried out. In a previous report, it was shown that the local electric field can influence the TPA cross-section of green fluorescent proteins (GFP).13 We have demonstrated that the local DNA electric field can influence the transition dipole of the binding molecule, which affects the TPA cross-section of the binding molecule upon complexing with DNA.23 5 and 6 displayed an overall decreasing TPA crosssection trend as the concentration of DNA was increased, indicating that the fluorophores are intercalating with DNA. This is critical as it illustrates a major drawback of CD whereas our TPA approach can provide sufficient information to determine the DNA-binding mode. A major 18 ACS Paragon Plus Environment

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

limiting issue with CD is that the signal indicative of an intercalating binding mode may not be observed because a stronger, more intense signal due to other interactions, such as aggregation, can overwhelm the intercalating signal.52 Based on the results, it is suggested that a change in the DNA-binding mode can be achieved by modifying the electron acceptor properties of the central heterocyclic core. This is significant as crescent shaped DNA-binding molecules have been reported to groove bind with DNA because they can match the curvature of the DNA groove for binding.53 Herein, we demonstrated that crescent shaped molecules can bind with DNA through intercalation, which does not obey the classical model for groove binding. This is in strong contrast to other intercalative binding molecules as they generally adopt planar structural scaffolds. The change in the DNA-binding mode can be rationalized by the charge-transfer interactions between the intercalator and DNA nucleobases. Previous calculation studies have shown that intercalators, such as ethidium bromide, are good electron acceptors that exhibit large charge delocalization while DNA base pairs are good electron donors.54 As the electron acceptor properties of the central heterocyclic core are increased, we observe a change in the DNA-binding mode from groove binding to intercalation by altering the pyrimidine-based to a difluoroboron β-diketonate central core, respectively. This implies that intercalating molecules are not only stabilized by interactions, such as electrostatic and dispersion interactions, but also charge-transfer contributions.54 Chargetransfer reactions can occur over short distances (on the order of Angstroms),55 which provides evidence of charge-transfer interactions between the DNA nucleobases and intercalating molecule. This is significant as it suggests that the DNA-binding mode can be tailored by simply modifying the electron acceptor properties of the central heterocyclic core. Based on these conclusions, one would elucidate that 8 would bind with DNA through intercalation. Although the electron acceptor properties of the central core are enhanced for 8, the results indicate that it does not bind with DNA. The findings reveal that 8 is neither intercalating nor groove binding with DNA. Narrowing of the emission band was not noted along with a low fluorescence enhancement (1.3-fold) in the presence of DNA provide evidence that 8 is not binding with DNA. Moreover, an ICD signal was not detected for 8 in the presence of DNA. This implies that the fluorophore is not binding with DNA, but rather electrostatically interacting with DNA. This was further supported by the anisotropy data as 8 exhibited a rotational correlational time of 2.3 ns in the presence of DNA, 19 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

suggesting some sort of interaction between the small molecule and DNA structure. The Nmethylpiperazine recognition units can electrostatically interact with the negatively charged DNA helical axis. The results confirm that the dicyanomethylene pyran core is responsible for 8 not binding with DNA since 3 and 7 were found to bind with DNA. The required interactions for DNA-binding with the DNA nucleobases, such as van der Waals and/or H-bonding interactions, is directly due to the dicyanomethylene pyran preventing these interactions. It has been reported that the orientation of a molecule upon binding with DNA can be directly related to the electrostatic and steric effects.54 Thus, the dicyanomethylene substituent may undergo steric hindrance with the DNA nucleobases, which does not allow the fluorophore to fit in the groove or the pocket of DNA. Bio-imaging experiments were performed to provide insight into the uptake and distribution of the fluorophores in live cells. Confocal microscopy is a convenient technique to visualize cellular localization and investigate trafficking and uptake of a drug.56 These experiments allowed us to correlate the performance of the small molecules based on our established structure-property relationships. Cellular uptake and accumulation as well as distribution and localization to the cell nucleus are critical parameters in the mechanism of DNA-targeted drugs.57 Three levels of localization are required for efficient drug delivery to the nucleus of cancer cells, which include targeting the cancerous tissue, accumulation in the cancer cell, and intracellular localization in the cell nucleus.58 The strength of the electron acceptor core with respect to the ability of the small molecule to localize in the cell nucleus was assessed. It should be emphasized that the fluorophore was more localized in the cytoplasm as the electron acceptor properties of the core were increased. In particular, 7 and 8 did not exhibit effective nuclear localization evident by the empty nucleus. This is important because it suggests that increasing the electron acceptor properties of the core reduces the ability of the drug to target DNA in the cell nucleus, thus making it an ineffective drug. On the other hand, the incorporation of a weak electron acceptor core resulted in effective uptake and localization of the small molecule in the cell nucleus, which is valuable because DNA-targeted therapies are expected to terminate cancer cells more efficiently and directly.59 This demonstrates that there is a balance between modifying the electron acceptor properties relative to the performance and the localization of the drug. This property can also be important for designing drugs targeted at organelles located in the

20 ACS Paragon Plus Environment

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

cytoplasm, such as the mitochondria,60 especially if the toxicity of the drug is enhanced when bound with DNA.

CONCLUSION In summary, we developed a design strategy that allows the DNA-binding mode to be tailored by investigating a series of small organic molecules. Steady-state and nonlinear ultrafast spectroscopies were employed to analyze the DNA-binding interactions of these small molecules. Structure-property relationships were established for small organic molecules and/or drugs that adopt a crescent shaped scaffold. The results reveal that the DNA-binding mode can be tailored by implementing structural modifications to the central heterocyclic core. This was achieved by replacing the central heterocyclic core from a pyrimidine-based moiety with a difluoroboron β-diketonate moiety, which modified the DNA-binding mode from groove binding to intercalation, respectively, suggesting that the central heterocyclic core plays a critical role in the DNA-binding mechanism based on our limited results. Cellular uptake and localization of the small molecules were evaluated by confocal microscopy. The findings indicate that increasing the electron acceptor properties of the core leads to a reduction in the ability of the small molecule to localize in the cell nucleus, thus making it an ineffective drug. The impact of this work opens new strategies for the design and development of therapeutics aimed at DNA as their performance and selectivity is dependent on the DNA-binding mechanism.

METHODS Materials. Chemicals and solvents were obtained from commercial suppliers and used without further purification, unless otherwise indicated. HeLa cells (CCL-2) were purchased from ATCC and used as received. Fetal bovine serum (FBS), penicillin/streptomycin (10,000 U/ml), and Live Cell Imaging Solution were provided by Life Technologies. Dulbecco's Modified Eagle Medium (DMEM) was supplied by Corning. Calf-thymus dsDNA was purchased from Sigma Aldrich and used as received. Further purification steps were not taken. A stock solution of fluorophore (500 μM) was constructed in DMSO. Samples were prepared in PBS (1X) by diluting the stock solution. Measurements for the DNA system were performed with [fluorophore] = 5 μM and [DNA] = 100 μM, unless otherwise indicated. PBS with a pH of 7.2 containing no calcium and magnesium was utilized in the experiments. 21 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Synthesis.

1

and

5

were

synthesized

previously.32

Page 22 of 30

Reaction

of

4,6-bis(4-

fluorophenyl)pyrimidine with N-methylpiperazine by nucleophilic aromatic substitution afforded 4.23 2, 3, 6, and 7 were synthesized elsewhere.61 Synthesis of 8 can be found in the Supplementary Information. Absorption and Fluorescence. Absorption spectra were measured on an Agilent 8341 spectrophotometer. Emission spectra were collected on a Fluoromax-2 fluorimeter with slits set at 4 nm and an integration time of 0.100 s. Fluorescence quantum yields were obtained for 1 and 3 using Coumarin 153 dissolved in methanol; 2 using Rhodamine B dissolved in water; 4 and 6 using cresyl violet dissolved in methanol; 5 and 8 using Rhodamine 6G dissolved in water; 7 using Coumarin 30 dissolved in methanol as the fluorescence standards. Circular Dichroism (CD). CD spectra were recorded on an Aviv model 202 circular dichroism spectrometer using a wavelength step of 1.0 nm and a bandwidth of 1.0 nm. CD scans were averaged with n = 3. Quartz cells with 10 mm path lengths were used for all measurements. All optical measurements were carried out at 25 oC under N2. Two-Photon Absorption (TPA). Two-photon spectroscopy was performed using a Kapteyn Murnane Laboratories diode-pumped mode-locked Ti:sapphire laser with pulses of ~30 fs. All emission scans were recorded at 800 nm excitation. TPA cross-sections were measured utilizing the two-photon excited fluorescence (TPEF) method.62 The fluorescence was collected perpendicular to the incident beam. A focal-length plano-convex lens was utilized to direct the fluorescence into a monochromator whose output was coupled to a photomultiplier tube. A counting unit was employed to convert the photons into counts. Coumarin 307 dissolved in methanol was used as a standard ((φδ)800 nm = 15 GM).62 Fluorescence Lifetime Measurements. The time-correlated single-photon counting (TCSPC) system has been described in detail elsewhere.16 A Kapteyn Murnane Laboratories (KML) mode-locked Ti:sapphire laser system (90 MHz) delivering ∼30 fs output pulses at 800

nm was used as the excitation source. Second harmonic generation in a β-barium borate (BBO) crystal was used to convert the 800 nm pulsed light to 400 nm excitation pulses. The fluorescence was collected at the maximum emission to a right angle excitation and detected by a photomultiplier tube (PMT) module coupled to a monochromator. A time to amplitude converter (TAC) was used to create time resolution. TimeHarp 200 (PicoQuant) software was implemented for the TCSPC measurements. Fluorescence anisotropy measurements were carried out by 22 ACS Paragon Plus Environment

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

recording the fluorescence with a polarizer oriented parallel or perpendicular with respect to the excitation source. The fluorescence upconversion setup used for time-resolved measurements has been reported previously.20 A Tsunami Ti:sapphire laser operated at 800 nm with 120 fs pulses and a repetition rate of 82 MHz was utilized to produce an excitation wavelength of 400 nm using a BBO crystal. The gate pulse was directed to a delay line while the excitation light was upconverted to excite the samples by a FOG-100 system. Fluorescence anisotropy measurements were collected by changing the polarization of the excitation source utilizing a Berek compensator. All samples were placed in a 1 mm thick rotating sample cuvette. Cell Culture and Fluorescence Microscopy. HeLa cells were cultured in DMEM (with phenol red) supplemented with 10% dialyzed FBS serum, 100 units of penicillin, and 100 μg/ml streptomycin in a humid incubator at 37 °C and 5% CO2. Prior to imaging, the cells were seeded in a 35 mm glass bottom dish supplied by MatTek. The cells were incubated for 24 h or until a monolayer was visible at 60 – 70% confluency. The working medium was removed and 2 μM of fluorophore in DMEM (serum-free) was added to the imaging dish and incubated for 1 h. The medium was removed and the cells were washed with PBS to remove excess fluorophore. Live Cell Imaging Solution was used as the physiological medium during the imaging experiments. The imaging experiments were carried out on a Leica Inverted SP5X Confocal Microscope System equipped with a 10X – 100X objective. Avalanche diode detectors were used when capturing and collecting images. A 405 nm diode (UV excitation), multi-line argon laser (470 – 670 nm), and Spectra Physics Mai-Tai two-photon tunable laser (640 – 1040 nm) were used as the excitation wavelength source. The excitation power was adjusted to 5 – 10 % to reduce cell apoptosis.

ASSOCIATED CONTENT Supplementary Information Detailed experimental procedures, synthesis of chromophores, CD spectra, fluorescence lifetime decay, and fluorescence microscopy images. This material is available free of charge via the Internet at http://pubs.acs.org.

23 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

AUTHOR INFORMATION Corresponding Author [email protected]

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT T.G.III acknowledges the National Science Foundation for support (1306815). J.N.W. acknowledges the support from the Florida Department of Health Bankhead-Coley New Investigator Research Program (3BN08).

References (1) Palchaudhuri, R., and Hergenrother, P. J. (2007) DNA as a target for anticancer compounds: methods to determine the mode of binding and the mechanism of action. Curr. Opin. Biotechnol. 18, 497–503. (2) Hurley, L. H. (2002) DNA and its associated processes as targets for cancer therapy. Nat. Rev. Cancer 2, 188–200. (3) Kim, J. A. (2003) Targeted therapies for the treatment of cancer. Am. J. Surg. 186, 264–268. (4) Suh, D., and Chaires, J. B. (1995) Criteria for the mode of binding of DNA binding agents. Bioorg. Med. Chem. 3, 723–728. (5) Chattoraj, M., King, B. A., Bublitz, G. U., and Boxer, S. G. (1996) Ultra-fast excited state dynamics in green fluorescent protein: multiple states and proton transfer. Proc. Natl. Acad. Sci. U. S. A. 93, 8362–8367. (6) Welch, J. J., Rauscher, F. J. 3rd, and Beerman, T. A. (1994) Targeting DNA-binding drugs to sequence-specific transcription factor.DNA complexes. Differential effects of intercalating and minor groove binding drugs. J. Biol. Chem. 269, 31051–31058. (7) Ihmels, H., Faulhaber, K., Vedaldi, D., Dall’Acqua, F., and Viola, G. Intercalation of organic dye molecules into double-stranded DNA. Part 2: the annelated quinolizinium ion as a structural motif in DNA intercalators. Photochem. Photobiol. 81, 1107–1115. 24 ACS Paragon Plus Environment

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

(8) Tse, W. C., and Boger, D. L. (2004) Sequence-selective DNA recognition: natural products and nature’s lessons. Chem. Biol. 11, 1607–1617. (9) Berman, H. M., and Young, P. R. (1981) The interaction of intercalating drugs with nucleic acids. Annu. Rev. Biophys. Bioeng. 10, 87–114. (10) Mukherjee, A., Lavery, R., Bagchi, B., and Hynes, J. T. (2008) On the molecular mechanism of drug intercalation into DNA: a simulation study of the intercalation pathway, free energy, and DNA structural changes. J. Am. Chem. Soc. 130, 9747–9755. (11) Long, E. C., and Barton, J. K. (1990) On demonstrating DNA intercalation. Acc. Chem. Res. 23, 271–273. (12) Pawlicki, M., Collins, H. A., Denning, R. G., and Anderson, H. L. (2009) Two-photon absorption and the design of two-photon dyes. Angew. Chem. Int. Ed. Engl. 48, 3244–3266. (13) Drobizhev, M., Makarov, N. S., Tillo, S. E., Hughes, T. E., and Rebane, A. (2011) Twophoton absorption properties of fluorescent proteins. Nat. Methods 8, 393–399. (14) Wang, Y., Clark, T. B., and Goodson, T. (2010) Two-photon and time-resolved fluorescence conformational studies of aggregation in amyloid peptides. J. Phys. Chem. B 114, 7112–7120. (15) Clark, T. B., Orr, M. E., Flynn, D. C., and Goodson, T. (2011) Synthesis and Optical Properties of Two-Photon Absorbing GFP-type Probes. J. Phys. Chem. C 115, 7331–7338. (16) Lahankar, S. A., West, R., Varnavski, O., Xie, X., Goodson, T., Sukhomlinova, L., and Twieg, R. (2004) Electronic interactions in a branched chromophore investigated by nonlinear optical and time-resolved spectroscopy. J. Chem. Phys. 120, 337–344. (17) Varnavski, O. P., Goodson, T., Mohamed, M. B., and El-Sayed, M. A. (2005) Femtosecond excitation dynamics in gold nanospheres and nanorods. Phys. Rev. B 72, 235405. (18) Clark, T. B., Ziółkowski, M., Schatz, G. C., and Goodson, T. (2014) Two-photon and timeresolved fluorescence spectroscopy as probes for structural determination in amyloid-β peptides and aggregates. J. Phys. Chem. B 118, 2351–2359. (19) Varnavski, O. P., Ostrowski, J. C., Sukhomlinova, L., Twieg, R. J., Bazan, G. C., and Goodson, T. (2002) Coherent Effects in Energy Transport in Model Dendritic Structures Investigated by Ultrafast Fluorescence Anisotropy Spectroscopy. J. Am. Chem. Soc. 124, 1736– 1743. (20) Varnavski, O., and Goodson, T. (2000) Femtosecond fluorescence dynamics and molecular 25 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

interactions in a water-soluble nonlinear optical polymeric dye. Chem. Phys. Lett. 320, 688–696. (21) Narayanan, A., Varnavski, O. P., Swager, T. M., and Goodson, T. (2008) Multiphoton Fluorescence Quenching of Conjugated Polymers for TNT Detection. J. Phys. Chem. C 112, 881–884. (22) Bhaskar, A., Ramakrishna, G., Lu, Z., Twieg, R., Hales, J. M., Hagan, D. J., Van Stryland, E., and Goodson, T. (2006) Investigation of two-photon absorption properties in branched alkene and alkyne chromophores. J. Am. Chem. Soc. 128, 11840–11849. (23) Doan, P. H., Pitter, D. R. G., Kocher, A., Wilson, J. N., and Goodson, T. (2015) TwoPhoton Spectroscopy as a New Sensitive Method for Determining the DNA Binding Mode of Fluorescent Nuclear Dyes. J. Am. Chem. Soc. 137, 9198–9201. (24) Strekowski, L., and Wilson, B. (2007) Noncovalent interactions with DNA: an overview. Mutat. Res. 623, 3–13. (25) Cosa, G., Focsaneanu, K.-S., McLean, J. R. N., McNamee, J. P., and Scaiano, J. C. (2007) Photophysical Properties of Fluorescent DNA-dyes Bound to Single- and Double-stranded DNA in Aqueous Buffered Solution¶. Photochem. Photobiol. 73, 585–599. (26) Reynisson, J., Schuster, G. B., Howerton, S. B., Williams, L. D., Barnett, R. N., Cleveland, C. L., Landman, U., Harrit, N., and Chaires, J. B. (2003) Intercalation of trioxatriangulenium ion in DNA: binding, electron transfer, x-ray crystallography, and electronic structure. J. Am. Chem. Soc. 125, 2072–2083. (27) Matsumi, N., and Chujo, Y. (2007) π-Conjugated Organoboron Polymers via the Vacant pOrbital of the Boron Atom. Polym. J. 40, 77–89. (28) Nordén, B., and Kurucsev, T. (1994) Analysing DNA complexes by circular and linear dichroism. J. Mol. Recognit. 7, 141–155. (29) Wilson, W. D., Tanious, F. A., Ding, D., Kumar, A., Boykin, D. W., Colson, P., Houssier, C., and Bailly, C. (1998) Nucleic Acid Interactions of Unfused Aromatic Cations: Evaluation of Proposed Minor-Groove, Major-Groove, and Intercalation Binding Modes. J. Am. Chem. Soc. 120, 10310–10321. (30) Seifert, J. L., Connor, R. E., Kushon, S. A., Wang, M., and Armitage, B. A. (1999) Spontaneous Assembly of Helical Cyanine Dye Aggregates on DNA Nanotemplates. J. Am. Chem. Soc. 121, 2987–2995. (31) Wilson, J. N., Wigenius, J., Pitter, D. R. G., Qiu, Y., Abrahamsson, M., and Westerlund, F. 26 ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

(2013) Base pair sensitivity and enhanced ON/OFF ratios of DNA-binding: donor-acceptordonor fluorophores. J. Phys. Chem. B 117, 12000–12006. (32) Pitter, D. R. G., Wigenius, J., Brown, A. S., Baker, J. D., Westerlund, F., and Wilson, J. N. (2013) Turn-on, fluorescent nuclear stains with live cell compatibility. Org. Lett. 15, 1330–1333. (33) Garbett, N. C., Ragazzon, P. A., and Chaires, J. B. (2007) Circular dichroism to determine binding mode and affinity of ligand-DNA interactions. Nat. Protoc. 2, 3166–3172. (34) Woo, H. Y., Liu, B., Kohler, B., Korystov, D., Mikhailovsky, A., and Bazan, G. C. (2005) Solvent effects on the two-photon absorption of distyrylbenzene chromophores. J. Am. Chem. Soc. 127, 14721–14729. (35) Liu, Z., Shao, P., Huang, Z., Liu, B., Chen, T., and Qin, J. (2008) Two-photon absorption enhancement induced by aggregation due to intermolecular hydrogen bonding in V-shaped 2hydroxypyrimidine derivatives. Chem. Commun. 2260–2262. (36) Zama, M., and Ichimura, S. (1970) Circular dichroism of acridine orange bound to DNA. Biopolymers 9, 53–63. (37) Grabowski, Z. R., Rotkiewicz, K., and Rettig, W. (2003) Structural changes accompanying intramolecular electron transfer: focus on twisted intramolecular charge-transfer states and structures. Chem. Rev. 103, 3899–4032. (38) Stsiapura, V. I., Maskevich, A. A., Kuzmitsky, V. A., Uversky, V. N., Kuznetsova, I. M., and Turoverov, K. K. (2008) Thioflavin T as a molecular rotor: fluorescent properties of thioflavin T in solvents with different viscosity. J. Phys. Chem. B 112, 15893–15902. (39) Todd, D. C., Jean, J. M., Rosenthal, S. J., Ruggiero, A. J., Yang, D., and Fleming, G. R. (1990) Fluorescence upconversion study of cis-stilbene isomerization. J. Chem. Phys. 93, 8658. (40) Banerjee, D., and Pal, S. K. (2006) Ultrafast charge transfer and solvation of DNA minor groove binder: Hoechst 33258 in restricted environments. Chem. Phys. Lett. 432, 257–262. (41) Blanchard-Desce, M., Wortmann, R., Lebus, S., Lehn, J.-M., and Krämer, P. (1995) Intramolecular charge transfer in elongated donor-acceptor conjugated polyenes. Chem. Phys. Lett. 243, 526–532. (42) Furstenberg, A., Julliard, M. D., Deligeorgiev, T. G., Gadjev, N. I., Vasilev, A. a, and Vauthey, E. (2006) Ultrafast Excited-State Dynamics of DNA Fluorescent\rIntercalators: New Insight into the Fluorescence\rEnhancement Mechanism. J. Am. Chem. Soc. 128, 7661–7668. (43) Sarkar, R., and Pal, S. K. (2007) Interaction of hoechst 33258 and ethidium with histone1 27 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

DNA condensates. Biomacromolecules 8, 3332–3339. (44) Kuśba, J., and Lakowicz, J. R. (1999) Definition and properties of the emission anisotropy in the absence of cylindrical symmetry of the emission field: Application to the light quenching experiments. J. Chem. Phys. 111, 89. (45) Flynn, D. C., Ramakrishna, G., Yang, H.-B., Northrop, B. H., Stang, P. J., and Goodson, T. (2010) Ultrafast optical excitations in supramolecular metallacycles with charge transfer properties. J. Am. Chem. Soc. 132, 1348–1358. (46) Fürstenberg, A., and Vauthey, E. (2007) Ultrafast excited-state dynamics of oxazole yellow DNA intercalators. J. Phys. Chem. B 111, 12610–12620. (47) Millar, D. P., Robbins, R. J., and Zewail, A. H. (1982) Torsion and bending of nucleic acids studied by subnanosecond time‐resolved fluorescence depolarization of intercalated dyes. J. Chem. Phys. 76, 2080–2094. (48) Abbotto, A., Baldini, G., Beverina, L., Chirico, G., Collini, M., D’Alfonso, L., Diaspro, A., Magrassi, R., Nardo, L., and Pagani, G. A. (2005) Dimethyl-pepep: a DNA probe in two-photon excitation cellular imaging. Biophys. Chem. 114, 35–41. (49) Belitsky, J. M., Leslie, S. J., Arora, P. S., Beerman, T. A., and Dervan, P. B. (2002) Cellular uptake of N-methylpyrrole/N-methylimidazole polyamide-dye conjugates. Bioorg. Med. Chem. 10, 3313–3318. (50) Werner, M. H., Gronenborn, A. M., and Clore, G. M. (1996) Intercalation, DNA Kinking, and the Control of Transcription. Science 271, 778–784. (51) Schumacher, M., Choi, K., Zalkin, H., and Brennan, R. (1994) Crystal structure of LacI member, PurR, bound to DNA: minor groove binding by alpha helices. Science 266, 763–770. (52) Wilson, W. D., Barton, H. J., Tanious, F. A., Kong, S. B., and Strekowski, L. (1990) The interaction with DNA of unfused aromatic systems containing terminal piperazino substituents. Intercalation and groove-binding. Biophys. Chem. 35, 227–243. (53) Munde, M., Ismail, M. A., Arafa, R., Peixoto, P., Collar, C. J., Liu, Y., Hu, L., DavidCordonnier, M.-H., Lansiaux, A., Bailly, C., Boykin, D. W., and Wilson, W. D. (2007) Design of DNA minor groove binding diamidines that recognize GC base pair sequences: a dimeric-hinge interaction motif. J. Am. Chem. Soc. 129, 13732–13743. (54) Řeha, D., Kabeláč, M., Ryjáček, F., Šponer, J., Šponer, J. E., Elstner, M., Suhai, S., and Hobza, P. (2002) Intercalators. 1. Nature of Stacking Interactions between Intercalators 28 ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

(Ethidium, Daunomycin, Ellipticine, and 4‘,6-Diaminide-2-phenylindole) and DNA Base Pairs. Ab Initio Quantum Chemical, Density Functional Theory, and Empirical Potential Study. J. Am. Chem. Soc. 124, 3366–3376. (55) Ly, D., Sanii, L., and Schuster, G. B. (1999) Mechanism of Charge Transport in DNA: Internally-Linked Anthraquinone Conjugates Support Phonon-Assisted Polaron Hopping. J. Am. Chem. Soc. 121, 9400–9410. (56) Lang, P., Yeow, K., Nichols, A., and Scheer, A. (2006) Cellular imaging in drug discovery. Nat. Rev. Drug Discov. 5, 343–356. (57) Ding, S., Qiao, X., Suryadi, J., Marrs, G. S., Kucera, G. L., and Bierbach, U. (2013) Using fluorescent post-labeling to probe the subcellular localization of DNA-targeted platinum anticancer agents. Angew. Chem. Int. Ed. Engl. 52, 3350–3354. (58) Langer, R. (1998) Drug delivery and targeting. Nature 392, 5–10. (59) Pan, L., He, Q., Liu, J., Chen, Y., Ma, M., Zhang, L., and Shi, J. (2012) Nuclear-targeted drug delivery of TAT peptide-conjugated monodisperse mesoporous silica nanoparticles. J. Am. Chem. Soc. 134, 5722–5725. (60) Torchilin, V. P. (2006) Recent approaches to intracellular delivery of drugs and DNA and organelle targeting. Annu. Rev. Biomed. Eng. 8, 343–375. (61) Pitter, D. R. G., Brown, A. S., Baker, J. D., and Wilson, J. N. (2015) One probe, twochannel imaging of nuclear and cytosolic compartments with orange and red emissive dyes. Org. Biomol. Chem. 13, 9477–9484. (62) Xu, C., and Webb, W. W. (1996) Measurement of two-photon excitation cross sections of molecular fluorophores with data from 690 to 1050 nm. J. Opt. Soc. Am. B 13, 481.

29 ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

Page 30 of 30