Mixed alkoxy/hydroxy 1,8-naphthalimides: expanded fluorescence colour palette and in vitro bioactivity

Elley E. Rudebecka, Rosalind P. Coxb, Toby D. M. Bellb, Rameshwor Acharyac, Zikai Fengc, Nuri Guevenc, Trent D. Ashton*ade and Frederick M. Pfeffer*a
aSchool of Life and Environmental Sciences, Deakin University, Waurn Ponds, 3216, Australia. E-mail: fred.pfeffer@deakin.edu.au
bSchool of Chemistry, Monash University, Clayton, 3800, Australia
cSchool of Pharmacy and Pharmacology, College of Health and Medicine, University of Tasmania, Hobart, 7001, Australia
dWalter and Eliza Hall Institute of Medical Research, Parkville, Victoria 3052, Australia. E-mail: ashton.t@wehi.edu.au
eDepartment of Medical Biology, The University of Melbourne, Parkville, 3010, Australia

Received 17th February 2020 , Accepted 28th April 2020

First published on 29th April 2020

An efficient and functional group tolerant route to access hydroxy 1,8-naphthalimides has been used to synthesise a range of mono- and disubstituted hydroxy-1,8-naphthalimides with fluorescence emissions covering the visible spectrum. The dialkoxy substituted compounds prepared possess high quantum yields (up to 0.95) and long fluorescent lifetimes (up to 14 ns). The method has been used to generate scriptaid analogues that successfully inhibit HDAC6 in vitro with tubulin acetylation assays confirming that these compounds are more effective than tubastatin.

The 4-hydroxy-1,8-naphthalimide scaffold is, like its 4-amino relative, a photostable fluorophore that possesses a large Stokes shift and two-photon absorption cross-section, as well as a high quantum yield suitable for imaging applications.1 With a pKa ∼ 5.52 the 4-hydroxy-1,8-naphthalimides are, essentially, fully deprotonated to the corresponding “phenoxide” (1, Fig. 1) at physiological pH. Under the same conditions the 4-alkoxy derivatives are blue emissive species (λem ∼ 440 nm) and the controllable interconversion of these spectrally distinct species has driven their increasing popularity as ratiometric fluorescent sensors in physiologically relevant conditions (Fig. 1).3–7
image file: d0cc01251c-f1.tif
Fig. 1 The physiologically relevant form of 4-hydroxy-1,8-naphthalimide (1) and examples of substituted derivatives.5,6,9

Nevertheless, the majority of examples reported to date possess relatively simple, single, substituents.8 Multi-substituted 1,8-naphthalimides offer interesting photophysical properties9,10 as recently illustrated by Jolliffe and New for 4-amino-1,8-naphthalimides.10 Nevertheless, examples in which disubstituted hydroxy/alkoxy 1,8-naphthalimides have been studied are rare.9,11

The most common approach to access 4-alkoxy (and 4-acyloxy) derivatives involves reacting the 4-hydroxy-1,8-naphthalimide with an alkylating (or acylating) agent. Critical to this approach, is the hydroxylation of the 1,8-naphthalimide core and this has conventionally been accomplished using a two-step sequence involving (i) conversion to the methoxy then (ii) cleaving the ether with strong acid (generally HI12 or Al/I213). These conditions are clearly incompatible with acid sensitive functional groups. Other methods include the multi-step Staudinger/Sandmeyer approach of Tsukamoto7 and the one-step Pd-catalysed hydroxylation developed by Fleming.14 Acid-labile groups were tolerated by the latter method, although extended reaction times (20–48 h) and high loading (8 mol%) of the t-BuXPhos or bippyphos ligand were required. A boronate ester/H2O2 approach has also been successfully used.15

Hydroxylation of halobenzenes by a tandem SNAr/Lossen-rearrangement was reported by Maloney16 using AcNHOH as the hydroxyl source. Similarly, Du used N-hydroxyphthalimide and K2CO3 to effect conversion of N-butyl-4-chloro-1,8-naphthalimide to the corresponding 4-hydroxy derivative.17 While encouraging, this methodology, with regards to 1,8-naphthalimides with multiple hydroxy/alkoxy substituents has not been explored.

Herein, using N-hydroxysuccinimide (NHS) as the key reagent, both 4-hydroxy and 3,4-disubstituted-1,8-naphthalimides have been accessed and the photophysical properties of the resultant fluorophores measured. The method has been exemplified in the development of a small family of highly potent inhibitors of tubulin deacetylation.

Beginning with the conditions reported by Du N-propyl-4-chloro-1,8-naphthalimide was heated (80 °C) in the presence of N-hydroxyphthalimide (1.1 equiv.) and K2CO3 (3.3 equiv.) in DMSO for 6 hours.17 However, following the published work-up (dilution in H2O and acidification to pH 3), the product collected from the reaction mixture, contained an aromatic impurity (apparent in the 1H NMR spectrum). In order to remove this impurity (likely derived from N-hydroxyphthalimide) column chromatography was required.

We reasoned that the related N-hydroxysuccinimide (NHS) should partake in the same mechanistic sequence proposed by Maloney16 to form the more soluble β-alanine which would be easily removed during aqueous workup. When NHS (1.1 equiv.) was used in place of N-hydroxyphthalimide (Scheme 1) consumption of starting material and formation of the fluorescent product (λex = 254 nm) was observed after 85 min using TLC analysis. Dilution of the reaction mixture with H2O, then adjustment of the pH to 1 resulted in the precipitation of the desired product 2 which was isolated by vacuum filtration in 97% yield. There was no evidence by 1H NMR that the β-alanine by-product persisted through the work-up.

image file: d0cc01251c-s1.tif
Scheme 1 Substrate scope for N-hydroxysuccinimide mediated hydroxylation. Unless otherwise stated X = Cl: a[thin space (1/6-em)]X = Br; b[thin space (1/6-em)]DMF was used in place of DMSO; c[thin space (1/6-em)]100 °C.

A range of N-substituted 4-halo-1,8-naphthalimides were then evaluated as substrates using the modified conditions (Scheme 1). The desired hydroxy product 3 was prepared from both the 4-chloro and 4-bromo-1,8-naphthalimide in high yields (87 and 88%, respectively). Various functional groups on the N-imide were well-tolerated, including carboxylic acids 4 and 5 (98 and 81% yield, respectively), methyl ester 6 (94%), a propargyl handle 8 (88%) and an aryl bromide 9 (98%).

The endoplasmic reticulum targeting fluorescent probe 10, that was reported by Xu,13 was prepared in 92% yield using these conditions. Similarly, Spring and Cui's probe 1118 that has demonstrated lysosomal accumulation was also prepared in 69% yield, although isolation was achieved by extraction with 10% MeOH in CH2Cl2 (the product failed to precipitate after acidification). For the synthesis of the acid-labile tetrahydropyranyl (THP) protected hydroxamic acid 7, DMF was used as the reaction solvent in place of DMSO to facilitate solvent removal prior to extractive workup. Following trituration with Et2O, compound 7 was isolated in 69% yield.

The use of NHS permits access to 3- and 4-substituted 1,8-napthalimides with different oxygen containing functional groups, a feat not possible using traditional preparations. First, a series of 3-alkoxy-4-chloro-1,8-naphthalimides was prepared by treating the corresponding 3-alkoxy-1,8-naphthalimide with N-chlorosuccinimide (NCS) (see ESI for details). Using the hydroxylation methodology, 3-methoxy-4-chloro-1,8-naphthalimde was reacted with NHS in the presence of K2CO3. In comparison with the single substituted 1,8-naphthalimides a higher temperature (100 °C vs. 80 °C) was required, nevertheless, the desired hydroxylated product 12 was isolated as a precipitate in 92% yield. Similarly, 3-benzyloxy-4-hydroxy-1,8-naphthalimide 13 was obtained in 77% yield, while a 3-(4-bromobenzyloxy) analogue 14 was isolated in 79% yield. The aryl bromide of 14 provides a synthetic handle which may permit access to numerous self-immolative linkers.19

Further functionalisation to afford differentially substituted 3,4-dialkoxy 1,8-naphthalimides was readily accomplished (Scheme 2). For example, treatment of 3-benzyloxy-4-hydroxy-1,8-naphthalimide 13 with benzyl bromide and K2CO3 gave the corresponding bis-benzyloxy compound 16 in 62% yield. Similarly, the use of iodomethane gave 15 in 63% yield. The 4-chloro-3-alkoxy-1,8-naphthalimides could also be directly converted to the corresponding 4-methoxy analogues using K2CO3 in MeOH at elevated temperatures (≥100 °C) using microwave irradiation. This procedure gives 15, 18 and 19 in yields of 80–93% while avoiding the use of iodomethane. Finally, the 3-hydroxy-4-methoxy derivative 22 was accessed in 67% yield by hydrogenolysis (Pd/C, H2, MeOH) of 20.

image file: d0cc01251c-s2.tif
Scheme 2 Preparation of 3,4-dialkoxy and 3-hydroxy-4-methoxy-1,8-napthalimides.

The optical (absorption and emission) properties of all compounds were determined in DMSO, with hydroxy substituted examples evaluated both in the presence of acid (TFA) and base (Et3N) (see ESI for full details, selected examples highlighted in Table 1 and Fig. 2 and 3). The disubstituted 3,4-dialkoxy series (e.g. 15, 16 and 20, Table 1) were efficient cyan fluorophores (λem ∼ 490 nm, λex ∼ 390 nm, ΦF up to 0.95) with absorption and emission maxima that were red-shifted approximately 20 and 40 nm, respectively, compared to the blue emissive 4-methoxy derivative 21. Time-resolved fluorescence studies of the 3,4-dialkoxy examples identified long fluorescence lifetimes (>12 ns) comprising a single exponential component (Table 1 and Fig. 2). The substitution pattern, or the effect of a methyl versus benzyl substituent, had little influence on emission wavelength.

Table 1 Photophysical data for selected alkoxy substituted compoundsa
Compd λabs (nm) λem (nm) ΦF τF (ns) χ2
a All experiments performed in DMSO. For full details see ESI.
15 391 493 0.76 14.4 1.05
16 387 487 0.86 12.3 1.13
20 386 490 0.95 14.1 1.08
21 367 449 0.47 5.6 1.07

image file: d0cc01251c-f2.tif
Fig. 2 For compounds 21 (blue), 16 (red), 20 (purple) and 15 (cyan) in DMSO. Top: Absorption (solid) and emission (dashed) spectra. Bottom: TCSPC decay histograms and fitted mono-exponential decay functions (λex = 375 nm). Instrument response function (IRF) shown in light-green.

image file: d0cc01251c-f3.tif
Fig. 3 Emission spectra in DMSO of S13 (purple), 21 (blue), 15 (cyan), 22 (green), 3 (Et3N, yellow), 13 (Et3N, orange), S11 (red).

Depending on the position and combination of 3- and 4-alkoxy/hydroxy substituents, emission maxima that span a significant portion of the visible spectrum can be accessed (Fig. 3). A single 3-benzyloxy substituent (e.g. S13) results in a hypsochromic shift of emission maxima (λem = 421 nm) relative to 4-alkoxy substitution (e.g. 21). As shown in Table 1, 3,4-dialkoxy substitution results in a cyan emission (λem ∼ 490 nm) while for the 3-hydroxy-4-methoxy substituent pattern (22) a greener emission is seen (λem = 508 nm). For the simple 4-hydroxy 3, in the presence of Et3N, a single green-yellow emission band (λem = 558 nm) is recorded; a red-shifted alternative to 4-amino-1,8-napthalimides.20 Inclusion of a 3-alkoxy group (e.g. 13 with Et3N) results in a bathochromic shift in emission maxima (λem = 590 nm) versus 3. Base induced red-shifts have also been recorded for amide containing 1,8-naphthalimides.21 The 3-hydroxy-1,8-naphthalimide S11 has a red emission (λem = 621 nm). Despite the quantum yields of hydroxy derivatives in DMSO being heavily reduced (<0.10), these compounds have considerable potential to be developed as more sophisticated cellular probes and sensing systems covering a wide range of spectral and intensity readouts.

To further illustrate the utility of the method we sought an additional application. The 1,8-naphthalimide scaffold is present in the histone deacetylase (HDAC) inhibitor scriptaid, and analogues of this compound have recently been developed as isoform selective HDAC inhibitors,22,23 including a potent and HDAC6 selective series of 4-amino-scriptaid derivatives.22 However the development of alkoxy variants has, only recently, received attention when Ho et al. reported JW-124 (Fig. 4) as a selective inhibitor with equivalent activity to the benchmark HDAC6 inhibitor, tubastatin25 (1000-fold selective for HDAC6 over HDAC1), in the in vitro inhibition of tubulin acetylation.

image file: d0cc01251c-f4.tif
Fig. 4 Top: Structures of JW-1 and the compounds prepared in this study. Bottom: Effect of scriptaid analogues on tubulin acetylation vs. non-treated cells (tubastatin +ve control, nullscript −ve control). Cells were treated with 1 μM of the compound (24 hr) and acetylated tubulin levels quantified by automated high content imaging using ≥ 1000 cells per condition. Data represent the average of 4 different experiments with 4 replicate wells for each experiment. Statistical significance determined using Welsh t-test (Graph pad Prism). Significance was set as p ≥ 0.05 = non-significant, 0.05 > p ≥ 0.01 = *, 0.01 > p ≥ 0.002 = **, and p < 0.002 = ***. Error bars represent SD.

Using the 4-amino-scriptaid analogues as templates22 a small library, consisting of seven alkoxy scriptaid analogues, was synthesised (23–29, Fig. 4, see ESI for full details) and evaluated in the human lung cancer cell line A549 using immunostaining against acetylated tubulin combined with high content imaging. While in this assay, increased tubulin acetylation by tubastatin did not reach statistical significance (p = 0.06) compared to non-treated and negative (nullscript treated) controls (Fig. 4), alkoxy analogues 23–29 displayed significantly higher levels (up to ∼1.5-fold higher) of tubulin acetylation compared with the non-treated and negative controls. This outcome suggests that the new analogues retain comparable HDAC6 inhibitory activity to either the 4-amino scriptaid analogues or JW-1 and further investigation into the influence of the oxy substituents on HDAC selectivity is warranted.

In conclusion a facile, functional group tolerant, method for the hydroxylation of 4-halo-1,8-naphthalimides has been developed. Products are isolated in excellent yield (typically >80%) with no requirement for chromatography. Based on the readily accessed 4-hydroxynaphthalimide scaffold, a highly active series of tubulin acetylation inhibitors were synthesised. The synthesis is equally effective using 4-bromo- or 4-chloro-1,8-naphthalimides and can be used to produce previously inaccessible 3-alkoxy-4-hydroxy-1,8-naphthalimides. Ultimately, the hydroxylation conditions allow for the functional-group tolerant synthesis of mixed 3,4-alkoxy/hydroxy 1,8-naphthalimides, a feat impossible to achieve using other literature conditions. The new fluorophores possess emissions covering the visible spectrum with the 3,4-dialkoxy analogues displaying high quantum yields and long fluorescent lifetimes.

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. (a) R. M. Duke, E. B. Veale, F. M. Pfeffer, P. E. Kruger and T. Gunnlaugsson, Chem. Soc. Rev., 2010, 39, 3936 RSC; (b) Z. Luo, B. Yang, C. Zhong, F. Tang, M. Yuan, Y. Xue, G. Yao, J. Zhang and Y. Zhang, Dyes Pigm., 2013, 97, 52 CrossRef CAS; (c) X. Wu, L. Li, W. Shi, Q. Gong, X. Li and H. Ma, Anal. Chem., 2016, 88, 1440 CrossRef CAS PubMed.
  2. (a) X.-X. Zhang, H. Wu, P. Li, Z.-J. Qu, M.-Q. Tan and K.-L. Han, Chem. Commun., 2016, 53, 8283 RSC; (b) Z. Qu, P. Li, X. Zhang, E. Wang, Y. Wang and P. Zhou, J. Lumin., 2016, 177, 197 CrossRef CAS.
  3. (a) T. D. Ashton, K. A. Jolliffe and F. M. Pfeffer, Chem. Soc. Rev., 2015, 44, 4547 RSC; (b) J. Tu, M. Xu, S. Parvez, R. T. Peterson and R. M. Franzini, J. Am. Chem. Soc., 2018, 140, 8410 CrossRef CAS PubMed.
  4. Z.-R. Dai, G.-B. Ge, L. Feng, J. Ning, L.-H. Hu, Q. Jin, D.-D. Wang, X. Lv, T.-Y. Dou, J.-N. Cui and L. Yang, J. Am. Chem. Soc., 2015, 137, 14488 CrossRef CAS PubMed.
  5. T. D. Nalder, T. D. Ashton, F. M. Pfeffer, S. N. Marshall and C. J. Barrow, Biochimie, 2016, 128–129, 127 CrossRef CAS PubMed.
  6. L. Feng, Q. Yan, B. Zhang, X. Tian, C. Wang, Z. Yu, J. Cui, D. Guo, X. Ma and T. D. James, Chem. Commun., 2019, 55, 3548 RSC.
  7. K. Tsukamoto, S. Shimabukuro, M. Mabuchi and H. Maeda, Chem. – Eur. J., 2016, 22, 8579 CrossRef CAS PubMed.
  8. (a) N. I. Georgiev and V. B. Bojinov, J. Lumin., 2012, 132, 2235 CrossRef CAS; (b) N. V. Marinova, N. I. Georgiev and V. B. Bojinov, J. Photochem. Photobiol., A, 2013, 254, 54 CrossRef CAS.
  9. R. W. Middleton, J. Parrick, E. D. Clarke and P. Wardman, J. Heterocycl. Chem., 1986, 23, 849 CrossRef CAS.
  10. K. G. Leslie, D. Jacquemin, E. J. New and K. A. Jolliffe, Chem. – Eur. J., 2018, 24, 5569 CrossRef CAS PubMed.
  11. T. Kasai, Kogyo Kagaku Zasshi, 1969, 72, 1128 CrossRef CAS.
  12. W. Adam, X. Qian and C. R. Saha-Möller, Tetrahedron, 1993, 49, 417 CrossRef CAS.
  13. S. Xu, H.-W. Liu, X.-X. Hu, S.-Y. Huan, J. Zhang, Y.-C. Liu, L. Yuan, F.-L. Qu, X.-B. Zhang and W. Tan, Anal. Chem., 2017, 89, 7641 CrossRef CAS PubMed.
  14. C. L. Fleming, T. D. Nalder, E. H. Doeven, C. J. Barrow, F. M. Pfeffer and T. D. Ashton, Dyes Pigm., 2016, 126, 118 CrossRef CAS.
  15. (a) W. Sun, W. Li, J. Li, J. Zhang, L. Du and M. Li, Tetrahedron, 2012, 68, 5363 CrossRef CAS; (b) B. Zhu, H. Jiang, B. Guo, C. Shao, H. Wu, B. Du and Q. Wei, Sens. Actuators, B, 2013, 186, 681 CrossRef CAS.
  16. P. S. Fier and K. M. Maloney, Org. Lett., 2016, 18, 2244 CrossRef CAS PubMed.
  17. W. Shu, L. Yan, J. Liu, Z. Wang, S. Zhang, C. Tang, C. Liu, B. Zhu and B. Du, Ind. Eng. Chem. Res., 2015, 54, 8056 CrossRef CAS.
  18. T. Liu, Z. Xu, D. R. Spring and J. Cui, Org. Lett., 2013, 15, 2310 CrossRef CAS PubMed.
  19. A. Alouane, R. Labruére, T. Le Saux, F. Schmidt and L. Jullien, Angew. Chem., Int. Ed., 2015, 54, 7492 CrossRef CAS PubMed.
  20. C. L. Fleming, T. D. Ashton and F. M. Pfeffer, Dyes Pigm., 2014, 109, 135 CrossRef CAS.
  21. K. N. Hearn, T. D. Nalder, R. P. Cox, H. D. Maynard, T. D. M. Bell, F. M. Pfeffer and T. D. Ashton, Chem. Commun., 2017, 53, 12298 RSC.
  22. C. L. Fleming, A. Natoli, J. Schreuders, M. Devlin, P. Yoganantharajah, Y. Gibert, K. G. Leslie, E. J. New, T. D. Ashton and F. M. Pfeffer, Eur. J. Med. Chem., 2019, 162, 321 CrossRef CAS PubMed.
  23. (a) C. L. Fleming, T. D. Ashton, C. Nowell, M. Devlin, A. Natoli, J. Schreuders and F. M. Pfeffer, Chem. Commun., 2015, 51, 7827 RSC; (b) Y. Zhang, J. Yan and T.-P. Yao, Eur. J. Med. Chem., 2017, 141, 596 CrossRef CAS PubMed.
  24. Y.-H. Ho, K.-J. Wang, P.-Y. Hung, Y.-S. Cheng, J.-R. Liu, S.-T. Fung, P.-H. Liang, J.-W. Chern and C.-W. Yu, Org. Biomol. Chem., 2018, 16, 7820 RSC.
  25. K. V. Butler, J. Kalin, C. Brochier, G. Vistoli, B. Langley and A. P. Kozikowski, J. Am. Chem. Soc., 2010, 132, 10842 CrossRef CAS PubMed.


Electronic supplementary information (ESI) available: Full details of synthesis, optical measurements and in vitro studies. See DOI: 10.1039/d0cc01251c

This journal is © The Royal Society of Chemistry 2020