Synthesis and structure investigation of stabilized aromatic oligoamides and their interaction with g quadruplex structures 3

The amide protons resonating at low fields 9.87-10.36 ppm are in line with the existence of strong intramolecular H-bonds in 5b.2c,2d Since the inter-atomic distances between amide and m

3.2 Result and Discussion

3.2.1 Helical Conformation in Pentamer 5 and Hexamer 6

1

5

6 10 11

15 16

20 21 25

N H O

O N N

O

H O

O N N

O

H O

oligomers 5 and 6 Although the corresponding helical conformation for longer

oligomers in solution has not been investigated, early molecular modeling5a on a

hexamer like 6a using the MM3 force field suggested it to adopt a crescent,

two-dimensional conformation that encloses an acyclic, open-ended cavity of ~ 4.3

in radius defined by six interior methoxy oxygen atoms (or 2.9 after deducting a covalent radius of 1.4 for oxygen atom); on this basis, a hexameric, head to tail binding mode was proposed to account for the helicity induction in porphyrin-modified hexamer by six chiral C60-incorporating histidines.3c The existence of such a large cavity of 2.9 , however, may seem uncertain as amide linkages are known to possess a great degree of plasticity in bond angles, instructing

the H-bond rigidified backbone to curve toward the H-bonded side.1f,2c Our

computational molecular modeling at the level of B3LYP/6-31G* on 5a showed that

oligomers of higher than tetramer should take up a helical backbone rather than the planar conformation as proposed previously.3c,5a Consequently, a helical cavity of

~1.4 in radius that is much smaller than 2.9 should result Consistent with these modeling results, we now provide here the solid state evidences involving helical

organizations in 5a and 6a by a continuous H-bonding network and the convincing 2D NOESY studies that support the helically folded conformations adopted by 5b and

6b and 6c, respectively, in solution These distinct helically folded conformations may

better explain the helicity induction observed previously.3c

Figure 3.1 Side and top views of oligoamides 5 and 6: (a) crystal structure of pentamer 5a, (b) ab initio

calculated structure of 5a and (c) crystal structure of hexamer 6a Interior methoxy methyl groups are

omitted for clarity

As mentioned in chapter 2, Oligomers 5 and 6 were synthesized from commercially available salicylic acid and 2,5-dihydroxybenzoic acid in 12-18 steps

Crystals of 5a and 6a suitable for X-ray structure determination were obtained by slow evaporation of 5a and 6a in mixed solvents containing hexane and chloroform

(1:1 v/v) for 5a and hexane and dichloromethane (1:1 v/v) for 6a at room temperature

Their crystal structures viewed along or perpendicular to helical axis are presented in

Fig 3.1a and 3.1c The common structural features shared between 5a and 6a are the

following: (1) both unit cells contain two enantiomeric helices of opposite helical senses (e.g., right/left handed) that tightly stack on each other with no inclusion of solvent molecules; (2) both structures possess a helical periodicity of ~five repeating units per turn; (3) the interior methoxy groups, pointing up and down alternatively, fill the helix hollow of ~1.4 in radius and completely prevent them from the encapsulation of guest molecules; and (4) as expected, all the inward-pointing amide protons and methoxy oxygen atoms participate in the formation of a continuous internally located H-bonding network that comprises up to ten intramolecular H-bonds (NH…OMe = 1.933-2.306 Å) One disparity between 5a and 6a involves the helical pitch While hexamer 6a has a pitch of ~ 3.4 Å typically observed in aromatic

foldamers,2a-f an appreciably larger helical pitch of 5 Å is observed in 5a Such a large

pitch may stem, in part, from the steric crowdedness involving the two end methoxy groups in the absence of favorable π-π stacking interactions and was also found in an aliphatic peptoid foldamer in solid state.2g The solid state structures of 5a and 6a (1)

provide the most conclusive evidence for the adoption of a helical conformation by

the foldable molecular strands 5a and 6a, (2) validate our conceptual reasoning and

others’ observations1f,2c detailing a significant effect the H-bonding forces may have

on the backbone curvature of aromatic folamers and (3) corroborate the power of ab

initio molecular modeling at the B3LYP/6-31G* level in the satisfactory prediction of 3D topologies of aromatic foldamers.2f,6

Figure 3.2 1H NMR spectra of pentamers (a) pentamer 5a (500 MHz, 5 mM, 298 K, CDCl3) and (b) pentamer 5b (800 MHz, 25 mM, 298 K, CDCl3 )

The highly repetitive nature of 5a (Figure 3.2a) and 6a6 led to the extensive 1H NMR signal overlaps among aromatic protons, hampering the elucidation of their folded structures in solution To overcome this difficulty, linear and branched alkoxyl

side chains as well as a methyl group para to the interior methoxy groups are

deliberately introduced into 5b At 25 mM, the 1H NMR spectrum of 5b recorded in

CDCl3 displays highly dispersed proton resonances for most of its protons (Figure 3.2b) The amide protons resonating at low fields (9.87-10.36 ppm) are in line with

the existence of strong intramolecular H-bonds in 5b.2c,2d Since the inter-atomic

distances between amide and methoxy protons from the crystal structure of 5a range

from 2.279 to 3.690 Å, two NOE contacts between every amide proton and its adjacent methoxy methyl groups should be seen in the 2D NOESY spectrum if a folded

conformation does prevail for 5b in solution The well-resolved amide protons and internal methoxy groups of 5b indeed permit us to observe the expected eight NOE

cross peaks, two for each amide protons Presumably owing to its unusually large helical pitch (5 Å), the NOEs between the end residues that are indicative of the helical

conformation of 5b was not detected (Figure 3.3)

Figure 3.3 NOE contacts (NOESY, 800 MHz, 25 mM, 298 K, 500 ms, CDCl3 ) seen between amide

protons and their adjacent interior methoxy protons of 5b

Figure 3.4 demonstrated full spectrum of NOESY and partial spectrum of TOCSY The TOCSY sequence is a homonuclear experiment which produces a COSY-like plot COSY gives correlations between protons that are coupled to one another, while TOCSY gives correlations between all protons in a given spin system In our experiment, TOCSY turns out to be the most efficient method to determine aromatic protons that are adjacent to each other The contacts between amide protons and aromatic protons could also be observed easily by TOCSY (Figure 3.4c)

a)

b) c)

Figure 3.4 2D NOESY (25 mM, 800 MHz, 298 K, 500 ms) & TOCSY (80ms) studies of pentamer 5b

in CDCl 3 (a) Full spectrum (2D NOESY), (b) Partial spectrum (2D NOESY) showing contacts

observed between external side chain protons and aromatic protons), (c) Partial spectrum (2D TOCSY) showing contacts observed between amide protons and aromatic protons

Such a helical conformation, however, can be confirmed by 2D NOESY study for

longer oligomers such as 6 where the favorable π-π stacking interactions between the

first and sixth aromatic rings bring the two end residues closer to each other, leading

to a typically observed helical pitch of ~ 3.4 Å for aromatic foldamers

Correspondingly, a strong NOE cross peak between the end ester methyl protons 1 and the aromatic proton 24, serving as an indicator for helical formation, was

observed for 6c at 10 mM in CDCl3 The existence of this helical conformation can be

nicely supported by the ab initio molecular modeling at the level of B3LYP/6-31G*

and the determined crystal structure of 6a (Figure 3.1c), both of which point to an energetically optimized configuration for 6c where the six interior methoxy methyl

groups point up and down alternatively along the aromatic backbone helically biased

by a continuous H-bonding network

Before pentamer 5b and hexamer 6b were designed, pentamer 5c and hexamer 6c

modified with various side chains were actually made first to probe the helical conformation that may be adopted by these longer oligomers However, the side

chains introduced into 5c do not give rise to a complete resolution of amide protons and internal methoxy methyl groups As to 6c, although good end-to-end NOE

contacts between either proton 1 or proton 5 and aromatic proton 24 can be detected

in CDCl3 at 283 K but not 298K, the signal overlap between protons 1 and 5 makes the accurate assignment of 1H NMR signals difficult This issue can be solved by addition of up to 25% DMSO-d6 into CDCl3, leading to a good separation between

protons 1 and 5 in both 6b and 6c (Figure 3.6) However, 2D NOESY collected for 4

hrs at 10 mM in 3:1 CDCl3/DMSO-d6 at either 283 K or 298 K failed to yield

detectable NOE contact between protons 1 and 24 in 6c Consequently, either crescent

or helical conformation concerning oligomers 5c and 6c can not be confidently

deduced

Figure 3.5 2D NOESY (500 MHz, 10 mM, 500 ms) & ROESY (500 MHz 10 mM, 200 ms) studies of 6c, showing end-to-end contact between protons 1 and 24 that is indicative of helical conformation (a)

2D NOESY (283 K, CDCl 3 , 4 hrs), (b) 2D NOESY (298 K, CDCl 3 , 4 hrs), (c) 2D NOESY (283 K, CDCl 3/DMSO-d 6 (3:1 v/v), 4 hrs), (d) 2D NOESY (283 K, CDCl 3/DMSO-d 6 (3:1 v/v), 8 hrs), (e) 2D NOESY (298 K, CDCl 3/DMSO-d 6 (3:1 v/v), 8 hrs) and (f) 2D ROESY (298 K, CDCl 3/DMSO-d 6 (3:1 v/v), 4 hrs) Compared to (d), NOE peak intensity in (e) is much weaker

Figure 3.6 Effect of DMSO-d 6 percentage in CDCl 3 on the 1 H NMR signal dispersion involving

protons 1 and 5 in hexamers: (a) hexamer 6b and (b) hexamer 6c

Before it was realized that lengthening the total acquisition time of 2D NOESY from 4 hrs to about 8 hrs in 3:1 CDCl3/DMSO-d6 allows us to detect an end-to-end

contact between proton 1 and proton 24 in 6c (Figures 3.5d and 3.5e), we decided to

replace the methyl side chain at the nitro end with an isopropyloxy side chain to

generate 5b and with an octyloxy side chain to generate 6b Such a minute difference

in structure pleasingly leads to completely resolved 1H NMR signals for critically important protons that allow us to unambiguously confirm a crescent conformation

for 5b in pure CDCl3 and a helical structure for 6b in 3:1 CDCl3/DMSO-d6 at room

temperature 6c was also confirmed to assume a helical conformation in solution by

2D NOESY experiment with a longer acquisition time (Figures 3.5d and 3.5e) or by 2D ROESY study (Figure 3.5f)

As mentioned above, the incorporation of two different side chains into 6b at only

one end differentiates the aromatic proton signals of the modified end units from all the other remaining units in the same molecule in CDCl3 Addition of 25% DMSO-d6into CDCl3 further separates ester methyl protons 1 from the interior methoxy methyl protons 5 As such, the ROE cross peak between the end units (methyl protons 1 and aromatic proton 24, Figure 3.7) is clearly identifiable This end-to-end ROE, along

with the observation of numerous ROE contacts among amide protons and their adjacent interior methoxy protons, evidently support the presence of the H-bond

enforced helical ordering in 6b in solution that leads to the stacking of one end over the other Observation of these end-to-end NOE or ROE contacts in 6b (Figure 3.7)

fully accords with the shortest inter-atomic distance (2.748 Å) found between protons

1 and 24 in the crystal structure of 6a

O O

O O

O

O O

O

O O

24 1

c) d )

e) f )

Figure 3.8 2D NOESY (500 MHz, 20 mM, 283 K, CDCl3 , 500 ms) and ROESY (500 MHz, 15 mM,

298 K, CDCl 3/DMSO-d 6 (3:1 v/v), 200 ms) studies of hexamer 6b (a) Full spectrum (2D NOESY)

showing end-to-end contacts, (b) Full spectrum (2D ROESY) showing end-to-end contacts, (c) Partial

spectrum (2D NOESY) showing contacts observed between aromatic protons and their adjacent

alkoxyl protons, (d) Partial spectrum (2D ROESY) showing contacts observed between aromatic

protons and their adjacent alkoxyl protons, (e) Partial spectrum (2D NOESY) showing contact

observed between amide protons and their adjacent interior methoxy protons, (f) Chemical structure of

6b

More NOESY, ROESY and TOXSY studies of 6b were demonstrated in Figure 3.8

and Figure 3.9 Figure 3.8c and 3.8d show the contacts between aromatic protons and

their adjacent alkoxyl protons, observed by NOESY and ROESY, respectively

N H O

O N N

O

H O

O N

O H

O

O

O2N 1 2 3 4

5 6 7

O

O

CH2(CH2)6CH3

8 9

10 11

12 13 14 15

16

17 18 19

20 21

22

23 24

25 26

27

28 29

6b

Contacts between amide protons and their adjacent interior methoxy protons in 6b

were shown in Figure 3.7e The well-resolved amide protons and internal methoxy

groups of 6b permit us to observe the expected ten NOE cross peaks, two for each

3.2.2 Helical Conformation of Heptamer 7

Scheme 3.1 Synthesis of heptamer 7a

O

OC8H17

N H O

O O N

O O

O N N

O

H O

+

a a) 3g, (COCl)2 , DMF, CH 2 Cl 2 ; b) H 2 , Pd/C, THF, 40 oC, 4a, then TEA/CH2 Cl 2

Heptamer 7 was obatained by a convergent route with relatively high yield 15%

For high-level oligomers, convergent coupling was used more frequently than

a) b)

stepwise fashion in order to saving time However, the reaction yield was usually low due to the steric hindrance imposed by helical conformation of oligoamides

Figure 3.10 1H NMR spectra of heptamer 7 (500 MHz, 5 mM, 298 K, CDCl3 )

One dimensional 1H NMR of heptamer 7 was presented in Figure 3.10 Similar to

pentamer and hexamer, the amide protons resonate at low fields (9.71-10.15 ppm),

suggesting the existence of strong intramolecular H-bonds in 7 Furthermore, the crescent and helical structure in 7 was probed by 2D NOESY study (Figure 3.11)

NOE contacts between amide proton and its adjacent methoxy methyl groups could be

observed in the 2D NOESY spectrum, suggesting a folded conformation of 7 in solution However, the aromatic protons in 7 were too overlapping, making it hard to

assign them Accordingly, amide protons and methoxy methyl group could not be

determined The only aromatic protons (27, 29, 32, 34) that could be assigned are the

ones at the last two benzyl units adjacent to nitro group end, due to their interaction with exterior side chains (Figure 3.11c) Although the position of each aromatic proton is unclear, a strong NOE cross between methyl proton and aromatic proton was

observed, which is similar to end-to-end contact in 6b and 6c Obviously, the aromatic

proton was not belongs to monomeric units at the nitro end (27, 29, 32, 34) Thus, it was assumed be the NOE between the end ester methyl protons 1 and the aromatic

proton 24

a)

b) c)

Figure 3.11 2D NOESY (500 MHz, 20 mM, 298 K, CDCl3, 500 ms) study of heptamer 7 (a) Full

spectrum, (b) Partial spectrum showing contact observed between amide protons and their adjacent

interior methoxy protons, (c) Partial spectrum showing contact observed between aromatic protons and

their adjacent interior methoxy protons

To validate this assumption, single crystal was tried to be grown by our traditional

slow evaporation method Unfortunately, solid state structure of heptamer 7 could not

be obtained despite of various solvent pairs To predict the structure of 7, ab initio

molecular modeling was performed As seen in Figure 3.12, heptamer 7 possess a

1 5

30 31 35

N H O

O N N

O H O

O

O N H O

23 24