- Email: [email protected]
Impact of peripheral substituents in regioselective synthesis of position-10 or position-20 bromo-bacteriochlorins
Tetrahedron Letters 58 (2017) 851–854
Contents lists available at ScienceDirect
Tetrahedron Letters journal homepage: www.elsevier.com/locate/tetlet
Impact of peripheral substituents in regioselective synthesis of position-10 or position-20 bromo-bacteriochlorins Mykhaylo Dukh a, Penny Joshi a, Kei Ohkubo b, Walter Tabaczynski c, Nayan J. Patel a,d, Joseph R. Missert a, Steve Zador a, Shunichi Fukuzumi b,e,f,⇑, Ravindra K. Pandey a,d,⇑ a
Chemistry Division, PDT Center, Cell Stress Biology, Roswell Park Cancer Institute, Buffalo, NY 14263, USA Department of Material and Life Science, Graduate School of Engineering, Osaka University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Suita, Osaka 565-0871, Japan c NMR Facility, Roswell Park Cancer Institute, Buffalo, NY 14263, USA d Pharmacology and Therapeutics, Roswell Park Cancer Institute, Buffalo, NY 14263, USA e Department of Bioinspired Science, Ewha Womans University, Seoul 120-750, Republic of Korea f Faculty of Science and Technology, Meijo University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Tempaku, Nagoya, Aichi 468-8502, Japan b
a r t i c l e
i n f o
Article history: Received 29 November 2016 Revised 9 January 2017 Accepted 13 January 2017 Available online 18 January 2017 Keywords: Chlorin Bacteriochlorin Bovine serum albumin Density functional calculations
a b s t r a c t Bacteriochlorins derived either from chlorophyll-a or bacteriochlorophyll-a on reacting with pyridinium bromide or N-bromosuccinimide (NBS) gave the corresponding 10- or 20-bromo analogues. In contrast to methyl bacteriopyropheophorbide-a, which afforded 10-bromo derivative, the 7-keto and 8-ketobacteriochlorins under similar reaction conditions gave the corresponding 20-bromo analogues exclusively. In both series, the nature of substituents present at position-3 did not make any difference in the reaction outcome. Density functional calculations were carried out to clarify the difference in reactions outcome. Density functional calculations were carried out to clarify the difference in reactivity of bromination at 10- and 20- meso positions. Ó 2017 Elsevier Ltd. All rights reserved.
Halogenated porphyrins and reduced porphyrins (chlorins and bacteriochlorins) have been of significant interest for the preparation of supramolecular structures with defined configuration as models for photosynthetic reaction centres.1,2 Such tetrapyrrolic systems with variable number of halogen substituents have also been investigated to understand the impact of the type and number of halogen functionalities in altering their electrochemical and photophysical properties.3–7 Reactions of the halogenated porphyrins, chlorins and bacteriochlorins with a variety of small molecules have shown potential in developing cell-specific photosensitizers for cancer-imaging and photodynamic therapy (PDT).8 The presence of halogen at the periphery of tetrapyrrolic system also provides an opportunity to introduce a wide variety of functionalities necessary for altering overall lipophilicity, increasing the number of targeting groups and also in developing multifunctional agents for cancer imaging and therapy.9 ⇑ Corresponding authors at: Department of Material and Life Science, Graduate School of Engineering, Osaka University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Suita, Osaka 565-0871, Japan (S. Fukuzumi). Chemistry Division, PDT Center, Cell Stress Biology, Roswell Park Cancer Institute, Buffalo, NY 14263, USA (R.K. Pandey). E-mail addresses: [email protected] (S. Fukuzumi), Ravindra. [email protected] (R.K. Pandey). http://dx.doi.org/10.1016/j.tetlet.2017.01.044 0040-4039/Ó 2017 Elsevier Ltd. All rights reserved.
In chlorin systems, e.g., methyl mesopyropheophorbide-a 1, we and others have shown9,10 that halogen can be selectively introduced at position-20 (i. e., chlorin 2) due to higher electron density at this particular position, which is next to the reduced ring [the other adjacent position is not available, and is occupied with a fused 5 member isocyclic ring system (ring E)]. However, in ringB reduced chlorin 3 in which the calculated electron density at position-5 (electron density: 0.316) was slightly higher than position-10 (electron density: 0.311), but gave only 10-bromo analogue 4, and it could be due to steric hindrance of the substituents present at position-3 of the chlorin moiety.9 Interestingly, the presence of electron-withdrawing substituents at position-3 in both chlorin and bacteriochlorin systems did not make any difference in selectivity for introducing a bromo- functionality. However, replacing chlorin 1 (ring D reduced) with chlorin 3 (ring B reduced) and bacteriochlorin 5 (rings B and D reduced) as substrates and subjecting them under similar reaction conditions afforded exclusively10-bromo- analogues 4 and 6 respectively (see Schemes 1–3). For the preparation of 20-bromo bacteriochlorins, we followed two different approaches. In our first approach, the methyl mesopyropheophorbide-a 7 was reacted with pyridinium bromide and the corresponding 20-bromo analogue 8 was isolated in 80%
852
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854 O-hexyl H Me Et NH N H
Me
Me
R 3
Me
Et
NH N B
H Me
N HN D
Me
N HN
Me
Me
O
1
Me
Et
Me
10
Br H
N HN
Me
Me
Me
Me
MeO2C
2
H Me Et H Br
NH N
H
N HN
Me
Me
NH N
Et
6
MeO2C
Me
R = ET H 2SO 4
Me
O
O
O
H
Me
N
Et
N
Et
19
O
H
N
Et
N
7-Keto
Me Me Et
20
7-Keto
Scheme 4. A possible ketobacteriochlorins.
mechanism
H
O
4
MeO2C
Me
Me OH OH
H 2SO 4 R = CHO
O
O
Me
R
Me
5
MeO2C
H O
N HN H
O
O-hexyl H Me Et NH N H Br N HN Me
Me
NH N
20
3
MeO2C
Me
R
H
NH N
Me
H MeO2C
Me Et
H
H
Me
Me
O
Me
+
O N
Me Et
H
O N
Et
21
8-Keto
O
R = Electron donating or electron withdrawing groups
for
the
formation
of
7-
and
8-
Scheme 1. Regioselective synthesis of bromo-chlorins and bacteriochlorins.
up
Me
Et Me
Et Et
NH N
b
Br H
Me
N HN
Me
NH N N HN H
O
NH N
H
Me
N HN
9
MeO2C
H
O
Me
a Me 7
5
NH N
20 H
Et
Et 8 10
N HN
Me
OH OH
Me
NH N
b H
Me
Me
Et
N HN
Me
NH N
c H
N HN
H
O
O
Me
Me
H
7
Et
Me
Me
MeO2C
O
10
MeO2C
a Et
Me
Me
H
8
O
Br
H Me
MeO2C
Et Me
Et c
Br
Me
Me
s wn ci do OH & Cis OH
O
11
MeO2C
H MeO2C
12 O
a. P yr idinium bromide; b. OsO4 ; c. H 2 SO4 Scheme 2. Synthetic approaches for the preparation of 20-bromo-7-keto-mesopyropheophorbide-a.
CHO Me OH OH Me
NH N
CHO Me
Et
H
N HN
Me
H MeO2C
H
H
14 O
N HN
MeO2C
H
15 O
MeO2C
Me
16 O
c
c
CHO
O
CHO
Et
NH N
Me
Me
NH N
Br
N HN
Me
H
N HN
H
Me
MeO2C
N HN
Me
Me H
O
O NH N
Br
Me
H
13
H
Me
Me
Me MeO2C
O NH N
Me
a
H
and
N HN
Me
CHO Me
CHO Me
NH N
b Me
O
17
O
H MeO2C
18
O
a. OsO4 , b. H 2 SO4 , c. py r idinium bromide Scheme 3. Synthetic approaches for the preparation of 20-bromo-3-formyl-7-keto and 8-ketobacteriochlorins.
yield. The intermediate on reacting with osmium tetroxide produced the corresponding bis-hydroxy bacteriochlorin 9, which on subjecting for Pinacol-Pinacolone reaction conditions produced the desired 20-bromo-7-ketobacteriochlorin 10 in excellent yield. In another approach, the methyl mesopyropheophorbide-a 7 was
converted to 7-ketobacteriochlorin 12 by following our well-established methodology,11 which on reacting with pyridinium bromide or NBS gave 20- bromo-7-ketobacteriochlorin 10 as a sole product. These results are in contrast to those obtained from bacteriochlorin 5 when subjected under similar reaction conditions, and suggests that the nature of substituents present in periphery makes a remarkable difference in bromo- substitution. To further explore the presence of an electron withdrawing group in the regioselectivity of bromination, 3-formyl-3-devinylpyropheo phorbide a 13 (the ethyl group at position-3 was replaced with a formyl group) was converted to keto-bacteriochlorins 15 and 16, does not make any difference in the formation of intermediate carbocation 19 (Scheme 4) resulting in the migration of methyl- group over the ethyl under Pinacol-Pinacolone reaction conditions. However, replacing the methyl group with a formyl group at position-3 resulted in the formation of intermediate carbocations 20 and 21 (Scheme 4) in which preferential migration of either methyl- or ethyl group produced an isomeric mixture of ketobacteriochlorins 17 and 18, which under bromination conditions also gave the corresponding 20-bromo analogues 17 and 18 respectively. These results further suggest that the nature of substituents at position-3 in both 7- and 8-ketobacteriochlorin did not make any difference in regioselective outcome of the brominated products. The formation of 7-keto-bacteriochlorin 10 from the corresponding vic-dihydroxybacteriochlorins 9 and 11 clearly indicates that the presence of bromo-functionality at position-20 does not make any difference in the preferential migration of the methyl group over the ethyl functionality under Pinacol-Pinacolone reaction conditions. The presence of a bromo- functionality at position-20 of ketobacteriochlorins 10, 17 and 18 was confirmed by detailed NMR study. The full assignment of the 1D proton NMR spectra was aided by correlations obtained from 2D 1H COSY and NOESY spectra. The COSY spectra were used to obtain through-bond correlations that sort the proton signals into separate spin systems. NOESY cross peaks were used to make specific assignments of individual proton signals based on their proximity to neighbouring protons. This approach was used to obtain the complete 1H NMR assignment of the three ketobacteriochlorins (see ‘‘Supporting Material” information). Substitution at position-20 is indicated because no meso proton signal could be assigned to that site. The two meso proton signals observed for each of the three ketobacteriochlorins were clearly and unambiguously assigned to H-5 and H-10. H-5 was identified by NOESY cross peaks correlated with the protons of the 3-CH2CH3, 3-formyl, 7-CH3 and 7-CH2CH3 groups. Similarly, H-10 was identified by NOESY cross peaks correlated with the protons of the 8-CH3, 8-CH2CH3 and 12-CH3 groups. None of these meso
853
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854
protons exhibited NOESY cross peaks correlated with either H-18 or 18-CH3. This is consistent with substitution at position-20. A comparison of the 1H NMR spectra of the 7- and 8-keto compounds revealed differences in chemical shifts. These can be seen in Fig. 1, where each sub-spectrum (A-F) shows 1H NMR data for the 7-keto (17) and 8-keto (18) forms. Substantial shift differences are illustrated for the (A) aldehyde, (B) H-5 and (C) H-10 protons. In each compound, the meso proton nearest the keto group is deshielded. Smaller shift differences are observed for the (D) 12-CH3 at 3.57 ppm, and (F) 7- or 8-CH2CH3 protons. In addition, the presence of two distinct stereoisomers (due to the chiral carbon at C-7 or C-8) is demonstrated by the ‘‘doubling” of several proton resonances. This can be seen in the spectra of 17 and 18 (Fig. 1), where stereoisomerism at C-8 of the 7-keto compound (17) results in doubling of peaks due to the (B) H-5, (E) 8-CH3 and 18-CH3, and (F) 8-CH2CH3 protons. Similarly, stereoisomerism at C-7 of the 8-keto compound (18) results in doubling of peaks due to the (C) H-10, (D) COOCH3 at 3.63 ppm, and (F) 7-CH2CH3 protons. The electronic absorption spectra of methyl 20-bromo-3ethyl-3-devinyl-8-keto-bacteriopyropheophorbide-a 10, methyl20-bromo-3-formyl-3-devinyl-8-keto-bacteriopyropheophorbide-a 17 and methyl-20-bromo-3-formyl-3-devinyl-7-keto-bacteriopyro pheophorbide-a 18 in methanol are illustrated in Fig. 2. As expected, replacing the ethyl group at position-3 of bacterioketochlorin 10 with a formyl- group exhibited a significant red shift in the electronic absorption spectra of bacteriochlorins 17 and 18. However, no significant shift in the long wavelength absorption
Fig. 1. 1H NMR spectral regions (A-F) showing chemical shift differences between the 7-keto (17) and 8-keto (18) forms. Doubling of some resonances due to chirality at the 7- or 8-position is also shown here (see text for details).
Fig. 3. Electronic absorption spectra of bacteriochlorins 10 (blue), 17 (red) and 18 (green) in bovine calf serum (BCS).
was observed. Interestingly, these compounds showed a significant difference in the pattern of ‘‘Soret” bands, especially for 3-formyl8-keto-20-bromobacteriochlorins, which was much shorter than observed for other analogues. To our surprise diluting methanolic solutions of these bacteriochlorins with 30-fold excess of aqueous bovine calf serum albumin (BCS) exhibited a dramatic red-shift in their electronic absorption spectra, which was in a range of 27–96 nm (Fig. 3). Interestingly, it was more prominent with compounds containing a formyl group (instead of ethyl) at position-3. Further addition of BCS did not make any significant changes. Such red shifts in the presence of albumin result from the solvatochromic effect of albumin which provides hydrophobic environments. Literature survey revealed that such a phenomenon mainly caused by the formation of non-symmetric aggregates has been reported previously by Smith et al. 12,13 in bacteriochlorophylls c, d and e, which are actually chlorins. Bacteriochlorins 10, 17 and 18 being a class of real bacteriochlorin system, the long wavelength band showed the shift in longer wavelength region, especially with the 3-formyl substituted bacteriochlorins 17 and 18 respectively. Among these two compounds, compared to 8-ketobacteriochlorin 18 (kmax ¼ 815 nm), the 7-keto analogue exhibited longer wavelength absorption (kmax ¼ 838 nm), possibly due to the formation of more non-symmetric aggregation.14 Density functional calculations were carried out to clarify the difference in reactivity of bromination at the meso-positions. Tables 1 and 2 show the electronic charges of the starting materials and the Hertree energies of the brominated products obtained by DFT with the B3LYP/6-31G(d) basis set. In contrast to bacteriopyro pheophorbide-a (derived from bacteriochlorophyll-a), which afforded 10-bromo analogues, the 7- or 8-ketobacterio chlorins obtained from methylpyropheophorbide-a produced the corresponding 20-bromo analogues, and the selectivity directly correlated to higher electron density at this position (Table 1). Table 2 shows the total energies of the brominated compounds in reference to the energies of 20-bromo compounds. The smallest energies were obtained in the case of brominated compounds 7, 12 and 16. However, the energy of 20-bromo compound 15 is
Table 1 Electronic densities on carbons on natural population analyses (NPA) calculated at the B3LYP/6-31G(d) level of theory.
Fig. 2. Electronic absorption spectra of bacteriochlorins 10 (blue), 17 (green) and 18 (red) in methanol.
Compound
3
5
Position 5 Position 10 Position 20
0.316 0.311 0.242
0.302 0.298 0.288
Compound
7
12
15
16
Position 5 Position 10 Position 20
0.248 0.226 0.316
0.294 0.232 0.306
0.252 0.279 0.295
0.286 0.240 0.295
854
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854
Table 2 Total energy differences (in kcal mol1) of the brominated compounds in reference to the 20-bromo compounds of certain chlorins and bacteriochlorins. Compound
7
12
15
16
Position 5 Position 10 Position 20
+6.8 +1.8 0.0
+4.2 +1.9 0.0
+13.1 0.9 0.0
+10.1 +1.8 0.0
0.9 kcal mol1 larger than the value of 10-bromo compound. Thus, bromination to certain chlorins and bacteriochlorins occur via nucleophilic attack of bromo cation to the 20-carbon nucleus rather than enthalpic control. In conclusion, this manuscript presents a facile approach for the synthesis of 10- and 20-bromoderivatives, and provides an opportunity to introduce a variety of substituents/targeting groups at an additional position of the near infrared (NIR) photosensitizing agents.15 These compounds could be useful in introducing variable number and types of tumour-targeting groups to target multiple receptors by a single agent. Besides, their applications in cancerimaging and therapy, the brominated analogues under proper synthetic design also provide useful substrates in constructing supramolecular structures as models for photosynthetic reaction centres. A remarkable shifts, especially at long wavelength absorptions of the ketobacteriochlorins 10, 17 and 18 in organic and aqueous BCS solutions was quite surprising and further studies to understand more about such characteristics with these and other compounds are underway. Acknowledgements RKP is highly thankful to Professor K. M. Smith, University of Louisiana, Baton Rouge, for valuable discussions on the aggregation characteristics of chlorin and bacteriochlorin systems. The help rendered by Aimee Marko in preparing the manuscript is highly appreciated. This work was supported by Photolitec, LLC, the
National Institutes of Health (NIH) (CA55791, CA127369) and the shared resources of the Roswell Park Cancer Support Grant (P30A15056) and Grants-in-Aid (nos. 26620154 and 26288037 to K.O.) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT); ALCA and SENTAN projects from JST, Japan (to S.F.). Mass spectrometry analyses of the compounds were performed at the Michigan State University, East Lansing, Michigan and the SUNY, Buffalo. NMR analyses were performed at RPCI, NMR facility. A. Supplementary material The experimental details and NMR spectra (1D, 2D, 13C, COSY/ NOESY) of compounds 10, 17 and 18 are discussed. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tetlet.2017.01.044. References 1. Kadish KM, Smith KM, Guilard R, editors. Handbook of porphyrin science, vol. 1, supramolecular chemistry. New Jersey: World Scientific; 2010 [and references therein]. 2. Kelley RF, Tauber MJ, Wilson TM, Wasielewski MR. Chem Commun. 2007;4407. 3. Kelley RF, Tauber MJ, Wasielewski MR. Angew Chem Int Ed. 2006;45:7979. 4. Tomizaki K-y, Yu L, Wei L, Bocian DF, Lindsey JSJ. Org Chem. 2003;68:8199. 5. Nakamura Y, Jang SY, Tanaka T, et al. Chem Eur J. 2008;14:8279. 6. Tokuji S, Yurino T, Aratani N, Shinokubo H, Osuka A. Chem Eur J. 2009;15:12208. 7. Kodis G, Liddell PA, de la Garza L, Moore AL, Moore TA, Gust D. Helv Chem Acta. 2001;84:2765. 8. Kadish KM, Smith KM, Guilard R, editors. Handbook of porphyrin science, vol. 4, phototherapy, radioimmunotherapy and imaging. New Jersey: World Scientific; 2010 [and references therein]. 9. Ethirajan M, Joshi P, William WH, Ohkubo K, Fukuzumi S, Pandey RK. Org Lett. 2011;1956:13 [and references therein]. 10. Smith KM, Goff DA, Simpson DF. J Am Chem Soc. 1985;107:4946. 11. Pandey RK, Isaac M, MacDonald I, et al. J Org Chem. 1987;62:1463. 12. Smith KM, Kehres LA, Fajer J. J Am Chem Soc. 1983;105:1385. 13. Abraham RJ, Smith KM, Goff DA, Bob FW. J Am Chem Soc. 1985;1085:105. 14. Kuriedo M, Tamiaki H. J Org Chem. 2005;70:820. 15. Ethirajan M, Chen Y, Joshi P, Pandey RK. Chem Soc Rev. 2011;40:340.
Contents lists available at ScienceDirect
Tetrahedron Letters journal homepage: www.elsevier.com/locate/tetlet
Impact of peripheral substituents in regioselective synthesis of position-10 or position-20 bromo-bacteriochlorins Mykhaylo Dukh a, Penny Joshi a, Kei Ohkubo b, Walter Tabaczynski c, Nayan J. Patel a,d, Joseph R. Missert a, Steve Zador a, Shunichi Fukuzumi b,e,f,⇑, Ravindra K. Pandey a,d,⇑ a
Chemistry Division, PDT Center, Cell Stress Biology, Roswell Park Cancer Institute, Buffalo, NY 14263, USA Department of Material and Life Science, Graduate School of Engineering, Osaka University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Suita, Osaka 565-0871, Japan c NMR Facility, Roswell Park Cancer Institute, Buffalo, NY 14263, USA d Pharmacology and Therapeutics, Roswell Park Cancer Institute, Buffalo, NY 14263, USA e Department of Bioinspired Science, Ewha Womans University, Seoul 120-750, Republic of Korea f Faculty of Science and Technology, Meijo University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Tempaku, Nagoya, Aichi 468-8502, Japan b
a r t i c l e
i n f o
Article history: Received 29 November 2016 Revised 9 January 2017 Accepted 13 January 2017 Available online 18 January 2017 Keywords: Chlorin Bacteriochlorin Bovine serum albumin Density functional calculations
a b s t r a c t Bacteriochlorins derived either from chlorophyll-a or bacteriochlorophyll-a on reacting with pyridinium bromide or N-bromosuccinimide (NBS) gave the corresponding 10- or 20-bromo analogues. In contrast to methyl bacteriopyropheophorbide-a, which afforded 10-bromo derivative, the 7-keto and 8-ketobacteriochlorins under similar reaction conditions gave the corresponding 20-bromo analogues exclusively. In both series, the nature of substituents present at position-3 did not make any difference in the reaction outcome. Density functional calculations were carried out to clarify the difference in reactions outcome. Density functional calculations were carried out to clarify the difference in reactivity of bromination at 10- and 20- meso positions. Ó 2017 Elsevier Ltd. All rights reserved.
Halogenated porphyrins and reduced porphyrins (chlorins and bacteriochlorins) have been of significant interest for the preparation of supramolecular structures with defined configuration as models for photosynthetic reaction centres.1,2 Such tetrapyrrolic systems with variable number of halogen substituents have also been investigated to understand the impact of the type and number of halogen functionalities in altering their electrochemical and photophysical properties.3–7 Reactions of the halogenated porphyrins, chlorins and bacteriochlorins with a variety of small molecules have shown potential in developing cell-specific photosensitizers for cancer-imaging and photodynamic therapy (PDT).8 The presence of halogen at the periphery of tetrapyrrolic system also provides an opportunity to introduce a wide variety of functionalities necessary for altering overall lipophilicity, increasing the number of targeting groups and also in developing multifunctional agents for cancer imaging and therapy.9 ⇑ Corresponding authors at: Department of Material and Life Science, Graduate School of Engineering, Osaka University, ALCA and SENTAN, Japan Science and Technology Agency (JST), Suita, Osaka 565-0871, Japan (S. Fukuzumi). Chemistry Division, PDT Center, Cell Stress Biology, Roswell Park Cancer Institute, Buffalo, NY 14263, USA (R.K. Pandey). E-mail addresses: [email protected] (S. Fukuzumi), Ravindra. [email protected] (R.K. Pandey). http://dx.doi.org/10.1016/j.tetlet.2017.01.044 0040-4039/Ó 2017 Elsevier Ltd. All rights reserved.
In chlorin systems, e.g., methyl mesopyropheophorbide-a 1, we and others have shown9,10 that halogen can be selectively introduced at position-20 (i. e., chlorin 2) due to higher electron density at this particular position, which is next to the reduced ring [the other adjacent position is not available, and is occupied with a fused 5 member isocyclic ring system (ring E)]. However, in ringB reduced chlorin 3 in which the calculated electron density at position-5 (electron density: 0.316) was slightly higher than position-10 (electron density: 0.311), but gave only 10-bromo analogue 4, and it could be due to steric hindrance of the substituents present at position-3 of the chlorin moiety.9 Interestingly, the presence of electron-withdrawing substituents at position-3 in both chlorin and bacteriochlorin systems did not make any difference in selectivity for introducing a bromo- functionality. However, replacing chlorin 1 (ring D reduced) with chlorin 3 (ring B reduced) and bacteriochlorin 5 (rings B and D reduced) as substrates and subjecting them under similar reaction conditions afforded exclusively10-bromo- analogues 4 and 6 respectively (see Schemes 1–3). For the preparation of 20-bromo bacteriochlorins, we followed two different approaches. In our first approach, the methyl mesopyropheophorbide-a 7 was reacted with pyridinium bromide and the corresponding 20-bromo analogue 8 was isolated in 80%
852
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854 O-hexyl H Me Et NH N H
Me
Me
R 3
Me
Et
NH N B
H Me
N HN D
Me
N HN
Me
Me
O
1
Me
Et
Me
10
Br H
N HN
Me
Me
Me
Me
MeO2C
2
H Me Et H Br
NH N
H
N HN
Me
Me
NH N
Et
6
MeO2C
Me
R = ET H 2SO 4
Me
O
O
O
H
Me
N
Et
N
Et
19
O
H
N
Et
N
7-Keto
Me Me Et
20
7-Keto
Scheme 4. A possible ketobacteriochlorins.
mechanism
H
O
4
MeO2C
Me
Me OH OH
H 2SO 4 R = CHO
O
O
Me
R
Me
5
MeO2C
H O
N HN H
O
O-hexyl H Me Et NH N H Br N HN Me
Me
NH N
20
3
MeO2C
Me
R
H
NH N
Me
H MeO2C
Me Et
H
H
Me
Me
O
Me
+
O N
Me Et
H
O N
Et
21
8-Keto
O
R = Electron donating or electron withdrawing groups
for
the
formation
of
7-
and
8-
Scheme 1. Regioselective synthesis of bromo-chlorins and bacteriochlorins.
up
Me
Et Me
Et Et
NH N
b
Br H
Me
N HN
Me
NH N N HN H
O
NH N
H
Me
N HN
9
MeO2C
H
O
Me
a Me 7
5
NH N
20 H
Et
Et 8 10
N HN
Me
OH OH
Me
NH N
b H
Me
Me
Et
N HN
Me
NH N
c H
N HN
H
O
O
Me
Me
H
7
Et
Me
Me
MeO2C
O
10
MeO2C
a Et
Me
Me
H
8
O
Br
H Me
MeO2C
Et Me
Et c
Br
Me
Me
s wn ci do OH & Cis OH
O
11
MeO2C
H MeO2C
12 O
a. P yr idinium bromide; b. OsO4 ; c. H 2 SO4 Scheme 2. Synthetic approaches for the preparation of 20-bromo-7-keto-mesopyropheophorbide-a.
CHO Me OH OH Me
NH N
CHO Me
Et
H
N HN
Me
H MeO2C
H
H
14 O
N HN
MeO2C
H
15 O
MeO2C
Me
16 O
c
c
CHO
O
CHO
Et
NH N
Me
Me
NH N
Br
N HN
Me
H
N HN
H
Me
MeO2C
N HN
Me
Me H
O
O NH N
Br
Me
H
13
H
Me
Me
Me MeO2C
O NH N
Me
a
H
and
N HN
Me
CHO Me
CHO Me
NH N
b Me
O
17
O
H MeO2C
18
O
a. OsO4 , b. H 2 SO4 , c. py r idinium bromide Scheme 3. Synthetic approaches for the preparation of 20-bromo-3-formyl-7-keto and 8-ketobacteriochlorins.
yield. The intermediate on reacting with osmium tetroxide produced the corresponding bis-hydroxy bacteriochlorin 9, which on subjecting for Pinacol-Pinacolone reaction conditions produced the desired 20-bromo-7-ketobacteriochlorin 10 in excellent yield. In another approach, the methyl mesopyropheophorbide-a 7 was
converted to 7-ketobacteriochlorin 12 by following our well-established methodology,11 which on reacting with pyridinium bromide or NBS gave 20- bromo-7-ketobacteriochlorin 10 as a sole product. These results are in contrast to those obtained from bacteriochlorin 5 when subjected under similar reaction conditions, and suggests that the nature of substituents present in periphery makes a remarkable difference in bromo- substitution. To further explore the presence of an electron withdrawing group in the regioselectivity of bromination, 3-formyl-3-devinylpyropheo phorbide a 13 (the ethyl group at position-3 was replaced with a formyl group) was converted to keto-bacteriochlorins 15 and 16, does not make any difference in the formation of intermediate carbocation 19 (Scheme 4) resulting in the migration of methyl- group over the ethyl under Pinacol-Pinacolone reaction conditions. However, replacing the methyl group with a formyl group at position-3 resulted in the formation of intermediate carbocations 20 and 21 (Scheme 4) in which preferential migration of either methyl- or ethyl group produced an isomeric mixture of ketobacteriochlorins 17 and 18, which under bromination conditions also gave the corresponding 20-bromo analogues 17 and 18 respectively. These results further suggest that the nature of substituents at position-3 in both 7- and 8-ketobacteriochlorin did not make any difference in regioselective outcome of the brominated products. The formation of 7-keto-bacteriochlorin 10 from the corresponding vic-dihydroxybacteriochlorins 9 and 11 clearly indicates that the presence of bromo-functionality at position-20 does not make any difference in the preferential migration of the methyl group over the ethyl functionality under Pinacol-Pinacolone reaction conditions. The presence of a bromo- functionality at position-20 of ketobacteriochlorins 10, 17 and 18 was confirmed by detailed NMR study. The full assignment of the 1D proton NMR spectra was aided by correlations obtained from 2D 1H COSY and NOESY spectra. The COSY spectra were used to obtain through-bond correlations that sort the proton signals into separate spin systems. NOESY cross peaks were used to make specific assignments of individual proton signals based on their proximity to neighbouring protons. This approach was used to obtain the complete 1H NMR assignment of the three ketobacteriochlorins (see ‘‘Supporting Material” information). Substitution at position-20 is indicated because no meso proton signal could be assigned to that site. The two meso proton signals observed for each of the three ketobacteriochlorins were clearly and unambiguously assigned to H-5 and H-10. H-5 was identified by NOESY cross peaks correlated with the protons of the 3-CH2CH3, 3-formyl, 7-CH3 and 7-CH2CH3 groups. Similarly, H-10 was identified by NOESY cross peaks correlated with the protons of the 8-CH3, 8-CH2CH3 and 12-CH3 groups. None of these meso
853
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854
protons exhibited NOESY cross peaks correlated with either H-18 or 18-CH3. This is consistent with substitution at position-20. A comparison of the 1H NMR spectra of the 7- and 8-keto compounds revealed differences in chemical shifts. These can be seen in Fig. 1, where each sub-spectrum (A-F) shows 1H NMR data for the 7-keto (17) and 8-keto (18) forms. Substantial shift differences are illustrated for the (A) aldehyde, (B) H-5 and (C) H-10 protons. In each compound, the meso proton nearest the keto group is deshielded. Smaller shift differences are observed for the (D) 12-CH3 at 3.57 ppm, and (F) 7- or 8-CH2CH3 protons. In addition, the presence of two distinct stereoisomers (due to the chiral carbon at C-7 or C-8) is demonstrated by the ‘‘doubling” of several proton resonances. This can be seen in the spectra of 17 and 18 (Fig. 1), where stereoisomerism at C-8 of the 7-keto compound (17) results in doubling of peaks due to the (B) H-5, (E) 8-CH3 and 18-CH3, and (F) 8-CH2CH3 protons. Similarly, stereoisomerism at C-7 of the 8-keto compound (18) results in doubling of peaks due to the (C) H-10, (D) COOCH3 at 3.63 ppm, and (F) 7-CH2CH3 protons. The electronic absorption spectra of methyl 20-bromo-3ethyl-3-devinyl-8-keto-bacteriopyropheophorbide-a 10, methyl20-bromo-3-formyl-3-devinyl-8-keto-bacteriopyropheophorbide-a 17 and methyl-20-bromo-3-formyl-3-devinyl-7-keto-bacteriopyro pheophorbide-a 18 in methanol are illustrated in Fig. 2. As expected, replacing the ethyl group at position-3 of bacterioketochlorin 10 with a formyl- group exhibited a significant red shift in the electronic absorption spectra of bacteriochlorins 17 and 18. However, no significant shift in the long wavelength absorption
Fig. 1. 1H NMR spectral regions (A-F) showing chemical shift differences between the 7-keto (17) and 8-keto (18) forms. Doubling of some resonances due to chirality at the 7- or 8-position is also shown here (see text for details).
Fig. 3. Electronic absorption spectra of bacteriochlorins 10 (blue), 17 (red) and 18 (green) in bovine calf serum (BCS).
was observed. Interestingly, these compounds showed a significant difference in the pattern of ‘‘Soret” bands, especially for 3-formyl8-keto-20-bromobacteriochlorins, which was much shorter than observed for other analogues. To our surprise diluting methanolic solutions of these bacteriochlorins with 30-fold excess of aqueous bovine calf serum albumin (BCS) exhibited a dramatic red-shift in their electronic absorption spectra, which was in a range of 27–96 nm (Fig. 3). Interestingly, it was more prominent with compounds containing a formyl group (instead of ethyl) at position-3. Further addition of BCS did not make any significant changes. Such red shifts in the presence of albumin result from the solvatochromic effect of albumin which provides hydrophobic environments. Literature survey revealed that such a phenomenon mainly caused by the formation of non-symmetric aggregates has been reported previously by Smith et al. 12,13 in bacteriochlorophylls c, d and e, which are actually chlorins. Bacteriochlorins 10, 17 and 18 being a class of real bacteriochlorin system, the long wavelength band showed the shift in longer wavelength region, especially with the 3-formyl substituted bacteriochlorins 17 and 18 respectively. Among these two compounds, compared to 8-ketobacteriochlorin 18 (kmax ¼ 815 nm), the 7-keto analogue exhibited longer wavelength absorption (kmax ¼ 838 nm), possibly due to the formation of more non-symmetric aggregation.14 Density functional calculations were carried out to clarify the difference in reactivity of bromination at the meso-positions. Tables 1 and 2 show the electronic charges of the starting materials and the Hertree energies of the brominated products obtained by DFT with the B3LYP/6-31G(d) basis set. In contrast to bacteriopyro pheophorbide-a (derived from bacteriochlorophyll-a), which afforded 10-bromo analogues, the 7- or 8-ketobacterio chlorins obtained from methylpyropheophorbide-a produced the corresponding 20-bromo analogues, and the selectivity directly correlated to higher electron density at this position (Table 1). Table 2 shows the total energies of the brominated compounds in reference to the energies of 20-bromo compounds. The smallest energies were obtained in the case of brominated compounds 7, 12 and 16. However, the energy of 20-bromo compound 15 is
Table 1 Electronic densities on carbons on natural population analyses (NPA) calculated at the B3LYP/6-31G(d) level of theory.
Fig. 2. Electronic absorption spectra of bacteriochlorins 10 (blue), 17 (green) and 18 (red) in methanol.
Compound
3
5
Position 5 Position 10 Position 20
0.316 0.311 0.242
0.302 0.298 0.288
Compound
7
12
15
16
Position 5 Position 10 Position 20
0.248 0.226 0.316
0.294 0.232 0.306
0.252 0.279 0.295
0.286 0.240 0.295
854
M. Dukh et al. / Tetrahedron Letters 58 (2017) 851–854
Table 2 Total energy differences (in kcal mol1) of the brominated compounds in reference to the 20-bromo compounds of certain chlorins and bacteriochlorins. Compound
7
12
15
16
Position 5 Position 10 Position 20
+6.8 +1.8 0.0
+4.2 +1.9 0.0
+13.1 0.9 0.0
+10.1 +1.8 0.0
0.9 kcal mol1 larger than the value of 10-bromo compound. Thus, bromination to certain chlorins and bacteriochlorins occur via nucleophilic attack of bromo cation to the 20-carbon nucleus rather than enthalpic control. In conclusion, this manuscript presents a facile approach for the synthesis of 10- and 20-bromoderivatives, and provides an opportunity to introduce a variety of substituents/targeting groups at an additional position of the near infrared (NIR) photosensitizing agents.15 These compounds could be useful in introducing variable number and types of tumour-targeting groups to target multiple receptors by a single agent. Besides, their applications in cancerimaging and therapy, the brominated analogues under proper synthetic design also provide useful substrates in constructing supramolecular structures as models for photosynthetic reaction centres. A remarkable shifts, especially at long wavelength absorptions of the ketobacteriochlorins 10, 17 and 18 in organic and aqueous BCS solutions was quite surprising and further studies to understand more about such characteristics with these and other compounds are underway. Acknowledgements RKP is highly thankful to Professor K. M. Smith, University of Louisiana, Baton Rouge, for valuable discussions on the aggregation characteristics of chlorin and bacteriochlorin systems. The help rendered by Aimee Marko in preparing the manuscript is highly appreciated. This work was supported by Photolitec, LLC, the
National Institutes of Health (NIH) (CA55791, CA127369) and the shared resources of the Roswell Park Cancer Support Grant (P30A15056) and Grants-in-Aid (nos. 26620154 and 26288037 to K.O.) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT); ALCA and SENTAN projects from JST, Japan (to S.F.). Mass spectrometry analyses of the compounds were performed at the Michigan State University, East Lansing, Michigan and the SUNY, Buffalo. NMR analyses were performed at RPCI, NMR facility. A. Supplementary material The experimental details and NMR spectra (1D, 2D, 13C, COSY/ NOESY) of compounds 10, 17 and 18 are discussed. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tetlet.2017.01.044. References 1. Kadish KM, Smith KM, Guilard R, editors. Handbook of porphyrin science, vol. 1, supramolecular chemistry. New Jersey: World Scientific; 2010 [and references therein]. 2. Kelley RF, Tauber MJ, Wilson TM, Wasielewski MR. Chem Commun. 2007;4407. 3. Kelley RF, Tauber MJ, Wasielewski MR. Angew Chem Int Ed. 2006;45:7979. 4. Tomizaki K-y, Yu L, Wei L, Bocian DF, Lindsey JSJ. Org Chem. 2003;68:8199. 5. Nakamura Y, Jang SY, Tanaka T, et al. Chem Eur J. 2008;14:8279. 6. Tokuji S, Yurino T, Aratani N, Shinokubo H, Osuka A. Chem Eur J. 2009;15:12208. 7. Kodis G, Liddell PA, de la Garza L, Moore AL, Moore TA, Gust D. Helv Chem Acta. 2001;84:2765. 8. Kadish KM, Smith KM, Guilard R, editors. Handbook of porphyrin science, vol. 4, phototherapy, radioimmunotherapy and imaging. New Jersey: World Scientific; 2010 [and references therein]. 9. Ethirajan M, Joshi P, William WH, Ohkubo K, Fukuzumi S, Pandey RK. Org Lett. 2011;1956:13 [and references therein]. 10. Smith KM, Goff DA, Simpson DF. J Am Chem Soc. 1985;107:4946. 11. Pandey RK, Isaac M, MacDonald I, et al. J Org Chem. 1987;62:1463. 12. Smith KM, Kehres LA, Fajer J. J Am Chem Soc. 1983;105:1385. 13. Abraham RJ, Smith KM, Goff DA, Bob FW. J Am Chem Soc. 1985;1085:105. 14. Kuriedo M, Tamiaki H. J Org Chem. 2005;70:820. 15. Ethirajan M, Chen Y, Joshi P, Pandey RK. Chem Soc Rev. 2011;40:340.