Highly Effective Dual-Function Near-Infrared (NIR) Photosensitizer for Fluorescence Imaging and Photodynamic Therapy (PDT) of Cancer

Article abstract of DOI:10.1021/acs.jmedchem.6b00890

We report herein the synthesis and biological efficacy of near-infrared (NIR), bacteriochlorin analogues: 3-(1′-butyloxy)ethyl-3-deacetyl-bacteriopurpurin-18-N-butylimide methyl ester (3) and the corresponding carboxylic acid 10. In in vitro assays, compared to its methyl ester analogue 3, the corresponding carboxylic acid derivative 10 showed higher photosensitizing efficacy. However, due to drastically different pharmacokinetics in vivo, the PS 3 (HPLC purity >99%) showed higher tumor uptake and long-term tumor cure than 10 (HPLC purity >96.5%) in BALB/c mice bearing Colon 26 tumors. Isomerically pure R- and S- isomers of 3 (3a and 3b, purity by HPLC > 99%) under similar treatment parameters showed identical efficacy in vitro and in vivo. In addition, photosensitizer (PS) 3 showed limited skin phototoxicity and provides an additional advantage over the clinically approved chemically complex hematoporphyrin derivative as well as other porphyrin-based PDT agents, which makes 3 a promising dual-function agent for fluorescence-guided surgery with an option of phototherapy of cancer.

Full text of DOI:10.1021/acs.jmedchem.6b00890

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Highly Effective Dual-Function Near-Infrared (NIR) Photosensitizer for
Fluorescence Imaging and Photodynamic Therapy (PDT) of Cancer
Nayan Patel, Paula Pera, Penny Joshi, Mykhaylo Dukh, Walter A Tabaczynski, Kevin E.
Siters, Mark Kryman, Ravindra R. Cheruku, Farukh Durrani, Joseph R. Missert, Ramona
Watson, Tymish Y. Ohulchanskyy, Erin Tracy, Heinz Baumann, and Ravindra K. Pandey
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00890 • Publication Date (Web): 17 Oct 2016
Downloaded from http://pubs.acs.org on October 18, 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.
Journal of Medicinal Chemistry 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 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
Highly Effective Dual-Function Near-Infrared (NIR) Photosensitizer for
Fluorescence Imaging and Photodynamic Therapy (PDT) of Cancer
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
Nayan Patel,1,2 Paula Pera,1 Penny Joshi,1 Mykhaylo Dukh,1 Walter A. Tabaczynski,1
Kevin Siters,3 Mark Kryman,3 Ravindra R. Cheruku,1 Farukh Durrani,1 Joseph R. Missert,1
Ramona Watson,1 Tymish Y. Ohulchanskyy,4,5 Erin Tracy,6 Heinz Baumann6 and
Ravindra K Pandey1,2,3*
1Photodynamic Therapy Center, Department of Cell Stress Biology, 2Department of
Pharmacology, 6Department of Molecular Biology, Roswell Park Cancer Institute, Buffalo, NY
14263, 3Photolitec, LLC, 73 High Street, Buffalo, NY 14203 and College of Optoelectronic
Engineering, Shenzhen University, Shenzhen, Guangdong, China, 518060 and
5Institute of Lasers, Photonics and Biophotonics, University of Buffalo, Buffalo, NY 14260
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 42
1
2
3
4
ABSTRACT
5
6
7
8
We report herein, the synthesis and biological efficacy of nearꢀinfrared (NIR),
bacteriochlorin analogs: 3ꢀ(1’ꢀbutyloxy)ethylꢀ3ꢀdeacetylꢀbacteriopurpurinꢀ18ꢀNꢀbutylimide methyl
ester (3) and the corresponding carboxylic acid 10. In in vitro assays, compared to its methyl
ester analog 3, the corresponding carboxylic acid
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
derivative 10 showed higher photosensitizing efficacy.
However, due to drastically different pharmacokinetics
in vivo, the PS 3 (HPLC purity > 99%) showed higher
tumor uptake and longꢀterm tumor cure than 10 (HPLC
purity > 96.5%) in BALB/c mice bearing Colon26
tumors. Isomerically pure R- and S- isomers of 3 (3a
and 3b, purity by HPLC >99%) under similar treatment parameters showed identical efficacy in
vitro and in vivo. In addition, photosensitizer (PS) 3 showed limited skin phototoxicity, and
provides an additional advantage over the clinically approved chemically complex
hematoporphyrin derivative as well as other porphyrinꢀbased PDT agents, which makes 3 a
promising dualꢀfunction agent for fluorescenceꢀguided surgery with an option of phototherapy of
cancer.
2
ACS Paragon Plus Environment

Page 3 of 42
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
Porphyrinꢀbased compounds, especially the chlorins and bacteriochlorins in which one
or two pyrrole units (diagonal to each other) are reduced respectively have shown enormous
interest for NIR fluorescence imaging and photodynamic therapy (PDT) of cancer1ꢀ3. These
tetrapyrrolic compounds can be synthesized either via a multiꢀstep synthesis or by partial
modification of naturally occurring chlorophyllꢀa and bacteriochlorophyllꢀa, which can be isolated
from Spirulina algae and Rb. sphaeroides respectively4. We and others have previously
reported a large number of effective PDT agents derived from chlorophyllꢀa and
bacteriochlorophyllꢀa5. Even though some of these compounds are highly fluorescent and quite
effective both in vitro and in vivo, but due to limited Stokes shift in the NIR, these compounds
are not ideal candidates for fluorescence imaging or fluorescenceꢀguided PDT of large and
deeply seated tumors6,7. The optical properties of a tissue affect both diagnostic and therapeutic
applications of light. The ability of an appropriate wavelength of light to penetrate tissue with
less scattering, interact the tumor–avid photosensitizer (PS), then emit from the tissue for
detection is essential to diagnostic applications8. The irradiation of a PS in tissue, excite the PS
the first excited singlet state. After inter system crossing (ISC), the PS excited triplet state can
undergo two kinds of reactions. First, it can react with the substrate to form radical species,
which on further reaction with oxygen to yield oxygenated products (type I reaction). It can also
undergo a photophysical process known as Type II reaction, which converts the triplet molecular
oxygen present in tumor to highly reactive singlet oxygen, believed to be a key cytotoxic agent
for the destruction of tumor cells by PDT9. However, the depth of light penetration also depends
on the tumorꢀtype. Therefore, it is of utmost importance to define the optical properties of a
tissue in designing the devices for optimization of imaging and therapeutic potential of a
photosensitizer/imaging agent.9
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
Among the bacteriochlorins investigated so far, a water soluble palladiumꢀ
bacteriopheophorbide photosensitizer is currently in Phase III trials in Europe as a vascular
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 42
1
2
targeted prostate cancer treatment10. The Pdꢀbacteriopheophorbide’s NIR absorption (763 nm)
and water solubility make it an excellent PDT agent. However, it has a very short halfꢀlife in the
body. It is classified as a vascular targeted PDT agent and since it is ineffectively taken up by
tumor cells at extravascular sites, it has no target selectivity. The presence of palladium as a
central metal in Pdꢀbacteriopheophorbideꢀa increases its singlet oxygen yield (heavy atom
effect)11, which enhances PDT, but it has limited fluorescence quantum yield, which diminishes
its utility as dualꢀfunction (imaging and therapy) agent.
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
We have been investigating the use of certain tumorꢀavid PS as dualꢀimaging
(fluorescence/PET12ꢀ14, fluorescence/SPECT15 and fluorescence/MRI)16,17 agent(s). The main
objective has been to develop a single molecule or a nanoplatform for imaging with an option of
directing surgery and photodynamic therapy. Surgery is the prime and essential component in
the curative treatment for the majority of cancer patients. Therefore, NIR fluorescenceꢀguided
surgery in combination with PDT could be extremely useful in treating various cancer types. For
developing a NIR fluorescence agent it is of the utmost importance that the compound exhibits a
significant Stokes shift at nearꢀinfrared region with high fluorescence. The agent should be
retained by the tumor for a sufficient length of time for detection and photoreaction. This
fluorescence guided resection (FGR) has the compelling ability to allow the surgeon a precise
location of the tumor bed which can improve tumor resection and tumor control, without any
injury to normal tissue18.
Herein, we report the synthesis, characterization, photophysical properties, tumorꢀ
imaging, therapeutic potential of NIR bacteriochlorin 3, and the corresponding stereoisomers (R-
& S-) as potential candidates for tumorꢀimaging and PDT. The advantages of PS 3, over purified
form of , 3ꢀ(1’ꢀhexyloxy)ethylꢀ3ꢀdevinylpyropheophorbideꢀa (HPPH, 1) and purpurinimide 2 are
also presented. The structures of these photosensitizers are shown in Figure 1.
4
ACS Paragon Plus Environment

Page 5 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
RESULTS AND DISUSSION
Chemistry
7
8
9
For developing tumorꢀspecific PDT agents both active and passive approaches have
been investigated by us and others, which resulted in some interesting findings. However most
of the PDT agents which are currently at various stages of preclinical and clinical trials were
developed by following the structureꢀactivity relationship (SAR) and quantitative structure
activity relationship (QSAR) studies19ꢀ21. The Roswell Park Cancer Research (RPCI) group was
successful in identifying 1 with a long wavelength absorption at λmax 665 nm (in vivo) as the
most effective candidate from a homologous series of alkyl ether analogs of pyropheophorbideꢀ
a,21ꢀ23 and will be undergoing Phase II clinical trials of head & neck cancer24.
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
For developing photosensitizers with longer wavelength absorption, the long wavelength
absorption of 1 was redꢀshifted by replacing the five member isocyclic ring with a six member Nꢀ
alkyl imide ring system, and the resulting purpurinimide class of compound exhibited a NIR
absorption at 700 nm25. Among a series of alkylꢀether analogs evaluated for PDT efficacy, the
Nꢀbutyl and Oꢀbutyl purpurinimide 2 was selected due to its excellent biological efficacy and
ease of synthesis.
.
Figure 1: Photosensitizers derived from chlorophyllꢀa and bacteriochorophyllꢀa bearing a five member
isocyclic or a six member imide ring system. Arrows denote the reduced pyrrole ring(s).
To further understand the impact of structural correlation of the chlorin vs. bacteriochlorin
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 42
1
2
3
4
5
6
system in biological efficacy, we retained all the structural parameters of purpurinimide, but
replaced the ring ‘B’ with a reduced pyrrole unit, which produced a significant red shift (82 nm)
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
Scheme 1: Synthesis of bacteriopurpurinimides
in its long wavelength absorption (λmax 782 nm). The desired bacteriopurpurinimides as methyl
ester
3
(3ꢀ(1’ꢀbutyloxy)ethylꢀ3ꢀdeacetylꢀbacteriopurpurinꢀ18ꢀNꢀbutylimide methyl ester;
Photobac) and the corresponding carboxylic acid 10, were prepared by following the reaction
sequences shown in Scheme 1
6
ACS Paragon Plus Environment

Page 7 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In our approach, bacteriochlorophyllꢀa 4 was extracted from Rb. sphaeroides, which on
treating with trifluoroacetic acid (TFA) and then with CH2N2 (diazomethane) afforded methyl
bacteriopheophorbideꢀa 526. The five member isocyclic ring present in 5 was converted into a
six member anhydride ring system 6 by following the methodology reported for the preparation
of bacteriopurpurinꢀ18 methyl ester27. Methyl bacteriopurpurinimide 6 on reacting with nꢀbutyl
amine gave an isomeric mixture of the intermediate amides, which were not isolated and
immediately treated with methanolic KOH to afford the 3ꢀacetylꢀbacteriochlorinꢀnꢀbutylimide 7 in
75 % yield. Due to a significant difference in long wavelength absorptions between the
anhydride vs. imide ring systems, as well as the intermediate amides, the conversion of 6 to 7
can be monitored by UVꢀvis spectroscopy (Figure 2).
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
1.4
1.2
1
0.8
0.6
0.4
0.2
0
300
400
500
600
700
800
900
1000
Wavelength (nm)
Figure 2: Absorption spectra of bacteriochlorins containing an anhydride (6), intermediate amide or an
imide (7) ring system in methanol (not in equimolar concentrations).
The 3ꢀ(1’ꢀhydroxy)ethyl analog 8 obtained on reacting 7 with sodium borohydride was
reacted with HBr gas at room temperature. The intermediate bromoꢀ derivative was not isolated
but immediately reacted with nꢀbutanol in presence of anhydrous potassium carbonate. The
desired butyl ether analog 3 was obtained as a major product (83% yield) along with 3ꢀvinylꢀNꢀ
butylpurpurinimide 9 as a minor product in 5ꢀ10 % yield. In a separate approach, we followed a
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 42
1
2
3
4
5
6
7
8
similar methodology except the starting material 6 was directly isolated from Rb. sphaerodes in
almost the same yield. The second approach is advantageous as it eliminated the initial two
steps of the original synthetic procedure.27
9
The structures of the intermediates and the final products were confirmed by NMR (1H
and 13Cꢀ), high resolution mass (HRMS) and/or elemental analyses. The purity of the final
product as methyl ester and carboxylic acid analogs was also confirmed by HPLC analysis.
Due to the presence of a chiral center at positionꢀ3(1’) of bacteriochlorin 3, it was isolated as a
mixture of R- & S- stereoisomers. To compare the in vivo and in vitro biological efficacy of the
individual isomers, both isomers were separated by HPLC using a Waters Symmetry
preparative column and methanol as mobile phase (Figure 3). The purity of both the possible
isomers was confirmed by NMR and mass spectrometry analyses. However, no difference in
biological efficacy of individual isomers was observed. Therefore, the stereochemistry (R- or S-)
of individual isomers at position 3(1’) was not established. The isomer with retention time 15.3
min was coded as 3a and the other eluted at 16.9 min was coded as 3b.
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
Proton NMR spectral assignment was aided by information obtained from cross peaks
observed in 2ꢀD NMR spectra (Supplemental material, Figure S19) acquired for compound 8.
1
These spectra include homonuclear 2D H COSY, NOESY, and TOCSY, as well as 2D
heteronuclear 1Hꢀ13C HSQC. NOESY spectra were especially useful in making the meso proton
assignments. NOESY cross peaks for compound 8 indicate throughꢀspace interactions
between the meso proton at 8.81 ppm and the 31ꢀH (6.21 ppm) and 31ꢀCH3 (2.05 ppm) protons.
These cross peaks identified this meso proton as 5ꢀH. Additional cross peaks, between 5ꢀH
and the 7ꢀH (4.20 ppm) and 7ꢀCH3 (1.79 ppm) protons, further substantiated this assignment.
The assignment of the 10ꢀH meso proton was made in a similar way. Here, the NOESY
spectrum showed that the meso proton at 8.61 ppm interacts with the 12ꢀCH3 (3.62 ppm), 8ꢀH
(4.01 ppm), 8ꢀCH2CH3 (2.03 and 2.33 ppm), and 8ꢀCH2CH3 (1.11 ppm) protons. Likewise, 20ꢀH
was assigned using correlations observed between the meso proton at 8.33 ppm and the 2ꢀCH3
8
ACS Paragon Plus Environment

Page 9 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(3.27 ppm), 18ꢀH (4.19 ppm) and 18ꢀCH3 (1.68 ppm) protons. Further interpretation of the 2D
spectra allowed assignment of the remainder of the protons of 8. The 1H spectral assignment of
8 was useful in interpreting and assigning the 1H spectra of 3, 7, 9, and 10, which have similar
structures.
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
1
Figure 3:HPLC chromatograms and H NMR spectra (meso region) of photosensitizers 3, 3a, and 3b
(purity >99%). Signals demonstrating the presence of both stereoisomers are seen for 3, while signals of
the individual isomers are observed for 3a and 3b. Two distinct 5ꢀH NMR resonances (meso proton at
position 5)are observed at ~8.8 ppm for the mixture of the two isomers (A), while each of the isolated
stereoisomers shows only one 5ꢀH peak (B and C).
Compound 3 is a mixture of two unresolved stereoisomers. These two forms are
possible because of chirality at C31. The influence of this stereoisomerism is nicely illustrated
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 42
1
2
3
4
5
6
7
8
9
by the meso region of the proton NMR spectrum of 3, where two distinct signals (at 8.832 ppm
and 8.784 ppm; Figure 3A) were observed for the 5ꢀH proton. In contrast, each of the two
isolated stereoisomers (3a and 3b, HPLC purity >99%) showed only a single resonance for 5ꢀH
(Figure 3B and 3C), demonstrating a complete resolution of the R- and S- forms. Chemical
shifts of the 5ꢀH signals of the isolated forms closely match those of the pair of 5ꢀH signals
observed for the mixture. Unlike 5ꢀH, the 10ꢀH and 20ꢀH meso protons did not show a
significant difference in chemical shifts observed for the two stereoisomers. These protons are
further removed from the chiral center and were observed as singlets in 3 (Figure 3A), showing
no peak doubling due to stereoisomerism.
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
For evaluating the biological efficacy of methyl ester vs. the corresponding carboxylic
acid analog, 3 was reacted with aqueous lithium hydroxide and corresponding carboxylic
analog 10 was isolated in 40% yield. The structure was confirmed by NMR (1H, 13Cꢀ), HRMS
and purity was confirmed by HPLC [Column: Waters C18 Symmetry (4.6 x 150 mm, 5 ꢀ particle
size), mobile phase 0.5% (v/v) acetic acid in methanol; flow rate: 1.0 ml/min), retention times:
9.39 and 10.00 minutes], see Figure S21, Supplemental Information.
Photophysical properties
Photophysical properties for both bacteriopurpurinimide analogs 3 (ꢀmethyl ester) and 10
(ꢀcarboxylic acid) were examined in methanol and both analogs showed identical absorbance
and fluorescence characteristics with similar extinction coefficients values. Figure 4 displays
the absorbance and fluorescence spectra of bacteriopurpurinimide 3 in methanol. Structurally,
the only difference between 3 and 10 is that compound 3 contains a methyl ester functionality at
positionꢀ172 whereas compound 10 has a carboxylic acid functionality at the same position.
Since neither of these functional groups alters the electronic structures of the aromatic system
of the PS, these compounds are not expected to have a significant difference in their optical
properties. Both photosensitizers exhibit wavelength absorption at 782 nm (in methanol) and a
broad fluorescence emission is observed at 830nm. The fluorescence band is quite broad with
10
ACS Paragon Plus Environment

Page 11 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
peaks at 850 and 920 nm with no overlap with the absorption band beyond 830 nm. Such an
inherent characteristic provides an opportunity to excite the photosensitizer at its longest
absorption wavelength and measure the fluorescence > 830 nm. This is a unique characteristic
9
specific to this particular class of
compounds
and
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
was not observed so far with other PDT agents related to chlorins and purpurinimide systems
derived from chlorophyllꢀa.
Figure
4:
Relative absorption
(solid line) and fluorescence (dotted line) spectra of 3 in methanol at 5 ꢁM concentration. However,
compound 3 in the presence of BSA (bovine serum albumin) or HSA (human serum albumin) gave a red
shift of 5 nm with a broad NIR absorption at 782 nm.
1.4
1.2
1
0.8
0.6
0.4
0.2
0
300
400
500
600
700
800
900
1000
Wavelength (nm)
Figure 5: Comparative absorption spectra of HPPH 1 (a chlorin with 5 member fused isocyclic ring),
purpurinimide 2 (a chlorin with six member fused imide ring) and bacteriopurpurinimide (3), a
bacteriochlorin containing a six member fused imide ring, in methanol (not in equimolar concentrations).
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 42
1
2
3
4
5
6
7
8
We have previously reported a detailed photophysical and electrochemical properties of
related bacteriochlorin analogs. The relative singlet oxygen quantam yield was in the range of
0.45ꢀ0.48.28
9
In vitro cell-uptake, subcellular localization and photosensitizing efficacy of bacterio-
purpurinimides 3 and 10:
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
To investigate a correlation between cell uptake in in vitro photosensitizing activity and
PDT efficacy, the photosensitizing efficacy of 3 and its carboxylic acid analog 10 was
determined in Colon26 tumor cells by following the standard MTT assay.
B
Figure 6: In vitro cellꢀuptake (A), subcellular distribution (B), and photosensitizing efficacy (C) of 3
and the carboxylic acid 10. (A) Cellꢀuptake by flow cytometry, (B) subcellular localization in
mitochondria (red) and lysosome (green). Distribution was measured in terms of similarity score which
is a measure of coꢀlocalization with mitochondrial (CMXRos – Ex, 561 nm, Em, 595ꢀ660 nm) or
lysosome (green Fluoropheres ꢀEx. 488 nm, imaging window. > 500 nm). (C) In vitro PDT (A): Colon
26 cells were incubated with various concentrations of the PSs for 24h and then irradiated with light
0.25 J/cm2 (787 nm) and the cell viability was determined by MTT assay. All 3 experiments were done
in triplicate. Error bars represent standard deviation, p < 0.05. These compounds did not show any
dark toxicity (no light exposure).
The tumor cells were incubated with PS for 24h and uptake was measured by flow
cytometry (Figure 6A) and the subcellular site of retention was defined by image stream
analyses following the standard methodology.29 Compared to 3 the carboxylic acid 10 showed
almost 2ꢀfold higher uptake and enhanced localization in mitochondria (Figure 6B). Treatment
of Colon26 tumor cells with light at 787 nm at a fluence rate of 0.25J/cm2, and compound 10
was more effective than 3 (Figure 6C).
12
ACS Paragon Plus Environment

Page 13 of 42
Journal of Medicinal Chemistry
1
2
In vivo uptake (by fluorescence and 14C-) and comparative PDT efficacy of 3 and the
corresponding carboxylic acid 10:
3
4
5
6
7
8
9
Based on in vitro information, we expected a similar trend regarding in vivo tumor uptake
and PDT efficacy of 3 and 10. Both compounds were evaluated in BALB/c mice bearing
Colon26 tumors. The tumorꢀuptake of these compounds at variable time points was determined
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
by fluorescence imaging. In brief, three tumor bearing mice (BALB/c) per group were injected
with 3 and 10. The mice were imaged over the course of 48h (Figure 7A). Using the builtꢀin
Maestro image analysis software, the fluorescence signal in tumors was quantified at variable
time points for both the methyl ester and the carboxylic acid.
Unlike the in vitro results, 3 showed a higher level of uptake than the corresponding
4
12
8 h
24 h
48 h
12000
10000
8000
6000
4000
2000
Tumor
SD
Figure 7: (A) Whole body fluorescence imaging of 3 (top row) and 10 (bottom row) in a BALB/c
0
mouse bearing Colon26 tumors. The photosensitizers were injected (i. v.) into mice (3 mice/group) at
24
hour
48
hour
72
hour
a dose of 0.25 ꢀmol/kg and images were acquired using Maestro Deep Red filter set (2 second
exposure). Fluorescence signal was unmixed from background (red) and then overlaid with white light
scan (green) using ImageJ software. (B) Illustrating a difference in tumor uptake of 3 and 10 at
variable timeꢀpoints. Error bars represents standard deviation, p<.05. (C) Eight to twelve week old
female BALB/c mice implanted with Colon26 tumors, and 1.0 ꢁmole/kg [14]Cꢀlabeled 3 (0.1
ꢁCi/mouse) was administered i. v. At 24, 48 and 72 h postꢀinjection, 3 mice/group/time point were
euthanized. Tumors and the surrounding muscle were removed, weighed, solubilized and the
radioactivity was measured. Results are expressed as the mean cpm/gram tissue ±SD of three
mice/time point. Error bars represents standard deviation, p<.05 (14C- analog was synthesized by
following the reaction sequences shown in Scheme 2). (D) In vivo tumor response of 3 (red) and 10
13
(blue) in BALB/C mice implanted with Colon26 tumors. Twenty four hours post injection of PS 3, 0.25
ꢀmole/kg) mice were treated with laser light at 787 nm at a dose of 135 J/cm2 and 35 mW/cm2. Tumor
ACS Paragon Plus Environment
growth was monitored for a period of 60 days post PDT treatment. The PDT response was statistically
different from one another (p<0.05 by MantelꢀCox test). SD: Standard deviation. For control
experiment only PS 3 (no light) was used at a therapeutic dose.

Journal of Medicinal Chemistry
Page 14 of 42
1
2
3
4
5
6
7
8
9
carboxylic acid 10 at each time point (4, 8, 12, 24 and 48 h postꢀinjection), which starts declining
after 24 h postꢀinjection. During the course of imaging, the retention of PS 3, which was
statistically higher than 10 (p < 0.05), and the maximum uptake was observed at 24h postꢀ
injection (Figure 7B). Additionally, there was a significant difference in the tumorꢀretention
between the two PSs. Compound 3 retained in the tumor for an extended time, whereas the
maximal uptake of the corresponding carboxylic acid 10 was observed at 4h postꢀinjection. The
data suggest a distinct pharmacokinetic property of 3 bearing a methyl ester functionality and its
carboxylic acid derivative suggesting the latter having several fold faster clearance. The detailed
pharmacokinetic (PK), pharmacodynamics (PD) and toxicity studies of 3 in rats and dogs
following the US FDA guidelines are underway.
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
Fluorescence optical imaging in animals or organs is useful for determining the relative
distribution of a fluorophore. The technique is acceptable for comparison of two optically
identical fluorophores within the same organ, e. g., tumor vs. another tumor. However, it is not
reliable to compare two different organs, e. g. accumulation in tumor vs. liver. This is due to the
difference in the optical properties within different tissues which can interfere with signal
detection. For example, dark and dense tissues such as the liver and kidney allow less light to
penetrate and reflect compared to lighter tissues such as intestine and muscle. Therefore to
14
truly understand the tissue distribution of 3, the Cꢀ analog of 3 (14C-3) was synthesized (as a
mixture of R- and S- isomers, Figure 8) by following the methodology depicted in Scheme 2.
Similar to the in vivo fluorescence optical imaging, 14C-3 showed maximum tumor uptake at 24ꢀ
14
ACS Paragon Plus Environment

Page 15 of 42
Journal of Medicinal Chemistry
1
2
3
4
48 h postꢀinjection, with no statistical difference (Figure 7C).
5
6
7
8
The in vivo accumulation studies indicated that 3 (fluorescence and 14Cꢀ) and carboxylic
acid 10 (fluorescence only) show optimal accumulation at 24 h and 12 h postꢀinjection
respectively (Figure 7B). Therefore, to correlate the tumorꢀuptake data with PDT efficacy, both
compounds were evaluated in BALB/c mice bearing Colon26 tumors, using the drug doses
ranging from 0.25 to 0.5 ꢀmol/kg. For 3, the tumors were exposed to 787 nm light (135 J/cm2,
75 mW/cm2) at 24 h postꢀinjection and for 10 at 12h postꢀinjection using the same light fluence
and fluence rate . The regrowth of the treated tumors was monitored daily. In the initial study, 5
mice/group were used and 3 at a dose of 0.25 ꢀmol/kg, gave 80% complete response (4/5
mice) at day 60 postꢀtreatment. Under similar treatment parameters, the corresponding
carboxylic acid 10 showed limited PDT efficacy. Both compounds using identical conditions
were again evaluated in larger groups of mice. The combined results obtained from treating 66
tumor bearing mice with 3, and 22 mice with PS 10 are summarized in Figure 7D. Combined
data from in vivo PDT experiments demonstrated a correlation with distinctive pharmacokinetics
of the PS, and a higher PDT efficacy was observed for 3 than the corresponding carboxylic acid
derivative 10. In separate experiments, BALB/c mice bearing Colon26 tumors were also treated
with light at 12 and 48 h post injection of the PSs, resulting a lower PDT response (longꢀterm
cure).
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
Cellular uptake, subcellular localization and in vitro PDT efficacy of stereoisomers 3a and
3b:
Structurally, compound 3 exists as a mixture of two stereoisomers and, as a part of preꢀ
clinical testing required by the US FDA, the individual isomers were separated by HPLC, and
characterized by NMR (1H & 13Cꢀ) (see the experimental part and supplemental information)
before evaluating their in vitro and in vivo uptake and biological efficacy. The intracellular
localization of epimeric mixture 3 and the related isomers 3a (retention time: 15.39) and 3b
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 42
1
2
3
4
5
6
7
8
9
(retention time: 16.44) were subjected to similar in vitro experiments in Colon26 tumor cells, as
discussed previously, for compounds 3 and 10. Results summarized in Figure 9A showed no
difference in the cellular uptake of the isomeric mixture and the individual isomers. Imaging
analysis (Figure 9B) indicated some difference in subcellular localization of the individual
isomers, especially isomer 3b with an apparent higher accumulation in mitochondria than the
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
Figure 9: In vitro uptake and subcellular distribution of 3 and isomers (3a, eluted first in HPLC, and 3b,
eluted later). Colon 26 cells were incubated with the PSs (0.8 ꢁM) for 24 hours and analyzed using
ImageStreamx (Ex. 405 nm, Em. 740ꢀ800 nm). (A) Uptake of 3 and both the isomers was measured by
flow cytometry (B) Subcellular distribution of 3 as isomeric mixture in mitochondria (red) and lysosome
(green). Distribution was measured in terms of similarity score which is a measure of coꢀlocalization with
mitochondrial (CMXRos ꢀ Ex. 561 nm Em. 595ꢀ660 nm) or lysosome (Green Fluorospheres ꢀ Ex. 488 nm
Em. 480ꢀ560 nm). The more positive the similarity score the more closely the fluorescent probes are coꢀ
localized. Both A and B are the average of 3 experiments done in duplicate, error bars represent
standard deviation, p<0.05. (C) In vitro PDT activity of 3 (red), isomer 3a (green), and isomer 3b (blue)
was determined by using the cell viability MTT assay. Colon 26 cells were incubated with 0.025 ꢁM
compound for 24 hours and then irradiated with a laser light (787 nm) at various doses of light (0 – 1.0
J/cm2). Values are expressed as cell percent survival (set to 100%). Data represents the results of a
single experiment in triplicate, error bars represent standard deviation.
3a
3b
Figure 10: (A) Whole body fluorescence images of BALB/c mice bearing Colon26 tumors (3
mice/group) injected with isomer 3a (upper row) and 3b (lower row) at a dose of 0.25 ꢀmol/kg
(therapeutic dose) and imaged at 24h postꢀinjection. (B) Comparative uptake of 3a (green) and 3b
(blue) in Balb/c mice (3 mice/group) at variable time points. Both isomers produced maximal uptake at
24 h postꢀinjection. (C) Comparative in vivo PDT efficacy of bacteriopurpurinimide 3 and its
stereoisomers 3a and 3b in BALB/c mice bearing Colon 26 tumors (10 mice/group) at a dose of 0.25
mmol/kg treated with light (787 nm, 135 J/cm2, 75 mW/cm2) at 24 h postꢀinjection. Tumor regrowth was
monitored for day 60. PS 3 and its stereoisomers 3a and 3b showed similar tumor cure (8/10 mice in
16
each group were tumor free on day 60. Retention time (r.t): 3a = 15.39 min, 3b = 16.94 min.
ACS Paragon Plus Environment

Page 17 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
lysosomes. To correlate the cellular uptake of the isomers to PDT efficacy in vitro, a series of
MTT assay was performed using Colon26 tumor cells incubated with the photosensitizers for
24h under similar PS concentrations and light doses. The results summarized in Figure 9C
show the in vitro PDT efficacy from a single PS concentration (0.025 ꢀM) and the cells were
exposed to light at 787 nm in the range of 0ꢀ1.0 J/cm2, and no significant difference in
photosensitizing ability was observed.
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
For investigating in vivo efficacy, similar experiments with these isomers were then
extended in BALB/c mice bearing Colon26 tumors and the results are depicted in Figure 10. As
indicated, both isomers showed maximal tumorꢀuptake at 24 h postꢀinjection with a similar
clearence profile (Figure 10B). The isomeric mixture and the individual isomers were also
evaluated for in vivo PDT efficacy. The results illustrated in Figure 10C showed similar longꢀ
term tumor cure by the isomeric mixture and the individual isomers.
Compared to chlorins (HPPH 1 and purpurinimide 2), bacteriopurpurinimide (3)
shows enhanced PDT efficacy:
Among the chlorins with long wavelength absorption in the range of 660 to 700 nm
evaluated in our laboratories, (665 nm) 1 (665 nm) and 2 (702 nm) were noted to mediate
effective PDT in several tumor models.25 To rate these PS against 3, in vivo biological efficaces
of HPPH 1, purpurinimide 2 and 3 were assessed under identical conditions. The results
summarized in Figure 11 indicate that the PDTefficacy ranked as 3 > 2 > 1.
Figure 11: Comparative in vivo PDT efficacy of 1, 2
and 3 in BALB/c mice bearing Colon26 tumors.
Compound 1 at a dose of 0.25 and 0.47 ꢀmol/kg (665
nm)
purpurinimide
2
(702
nm)
and
bacteriopurpurinimide 3 at a dose of 0.25 ꢀmol/kg
was injected in tumored mice. The tumors were
exposed to light (787 nm, 135 J/cm2, 75 mW/cm2) at
24 h postꢀinjection (optimal drug uptake time point for
all the photosensitizers). As per approved protocol,
mice with tumor >400 mm3 should be sacrificed.
However, compounds 3, 2 and 1 gave 80%, 60% and
40% complete cure Mice with no tumors) respectively.
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
Injecting compounnd 1 at the higher dose (0. 47 mmol/Kg), yielded 40% tumor cure. To confirm
that the longer wavelength of light alone has no impact on tumor inhibition, the tumors (3
mice/group) were irradiated with light at therapeutic wavelength 665, 702 or 787 nm of PSs 1, 2
and 3 respectively, and no tumor inhibition was observed (see Figure S21, Supplemental
Information).
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
Fluorescence imaging of lung tumors with PS 3: To further expand the imaging capability of
3, we experimentally introduced tumors into the lungs of the mice (BALB/c, 3 mice/group) by
injecting 4T1ꢀluc cells intravenously. After 9 days of
tumor implantation, the bioluminescence images of
tumors were recorded.
Compound 3 (0.25
ꢀmole/kg) was then injected to mice intravenously
and the fluorescence images of tumors were
obtained after 24 h by IVIS SpectrumꢀCT. It also
showed high tumorꢀavidity and made it possible to
locate the tumors implanted in the lungs. The multiꢀ
modal image (fluorescence (red) merged with
(Merged Image)
Figure 12: Bioluminescence images of
tumors were taken at dayꢀ9 after injecting
(i. v.) 4T1ꢀluc tumor cells in BALB/c
mouse. Mice were injected with 3 (0.25
ꢀmol/kg) and the fluorescence images
(Ex. 785 nm, Em. > 830 nm) of tumors
were recorded at 24h postꢀinjection (A).
The merged image (pink) (fluorescence–
red, + bio luminescenceꢀblue) shows the
tumorꢀspecificity of 3 (B). The imaging
experiment was performed at the Pekinꢀ
Elmer Imaging Facility, California, USA.
bioluminescence (blue)
=
pink) shows the
localization of 3 within the tumor. The intensity of
the localization matched with the epiꢀfluorescence
of the PS, which further indicates excellent tumor specificity (Figure 12).
Photobleaching of compound 3 in vivo:
To determine the rate of photobleaching30ꢀ32 of compound 3 during the in vivo light
treatment, BALB/c mice bearing Colon26 tumors were injected with 0.25 mmol/kg of the
photosensitizer and the fluorescence intensity of 3 from the tumor (in situ) was recorded prior to
lightꢀirradiation (t = 0) and used as 100% value for each mouse. The tumors were then exposed
to 787 nm light and in regular interval the irradiation was stopped. The fluorescence intensity
18
ACS Paragon Plus Environment

Page 19 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was recorded, and the light treatment was continued. This procedure was repeated until the
total time of combined light treatment reached 30 min. The percentage of fluorescence
observed in the tumor was plotted against the time of exposure with light. The results obtained
are summarized in Figure 13 indicate an initial rapid rate of photobleaching of the PS, followed
by a decreased rate of photobleaching with approximately 40% of the photosensitizer
fluorescence still present (not photoꢀbleached) after the light exposure used for PDT. However
in this study, in contrast to a continued light exposure during the PDT treatment, the light
treatment was stopped at regular intervals for determining the photoꢀbleaching of the PS by
fluorescence, which may help in reꢀoxygenating the tumor tissue. To avoid tissue reꢀ
oxygenation, the fluorescence measurement of the PS in tumors before and after the PDT
treatment was performed in a separate set of animals, and a very similar end results were
observed (data not shown).
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
Figure 13: In vivo photobleaching of 3. BALB/c mice
(3) bearing colon 26 tumors were irradiated at 24 hours
after injection of 3 (dose, 0.25 ꢁmol/kg with 797 nm
light (135 J/cm2 at 75 mW/cm2). At regular intervals of
a 30 minute span, mice were imaged with Maestro
Deep Red filter set for PS signal which was plotted
against preꢀirradiated signal (the decay of the
absorption band at 787 nm at each time point was
measured and plotted with time). The data presented
are from a single experiment, error bars represent
standard deviation, p<0.05.
STAT3 photo-induced dimerization correlates to PDT response:
Previous reports from our laboratory have shown that murine and human cancer cell
lines react to PDT treatment by an immediate covalent crossꢀlinking of STAT3 to homodimeric
and other complexes33. A direct correlation between the photoꢀinduced STATꢀ3 dimerization
after irradiating the cells with light and concentration of the photosensitizer (PS) was observed.34
The light dose did produce a significant difference in crossꢀlinking of STAT3, but light alone, or
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 42
1
2
3
4
PS alone did not had any impact in cell death. To further correlate the in vitro and in vivo PDT
5
6
response of 3 and the corresponding carboxylic acid 10, Colon26 cells were incubated with
7
8
9
either 3 or 10 were exposed to light at 787 nm and quantified the lightꢀdependent STAT3 crossꢀ
linking by Western blot. The data summarized in Figure 14, clearly indicates that compound 10,
which was more effective in vitro than 3, showed higher STAT 3 dimerization. However, in vivo
PS 3, which produced higher uptake and longꢀterm tumor cure than 10,
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
also showed a direct correlation with higher STAT3 dimerization
Figure 14: A, Cellꢀassociated compounds 3 and 10 were visualized by fluorescent microscopy in
Colon26 cell cultures incubated for 2 h in medium containing 3200 nM photosensitizer. Phase and
fluorescent images of the cultures at 100X magnification are shown. B & C, Photoreaction mediated
by compounds 3 and 10 in Colon26 cells in vitro and in vivo. Crosslinking of STAT3 as measure for
photosensitizing activity. Left panel (B): Colon26 cell monolayers were incubated for 24 hours in
culture medium containing the indicated concentrations of compounds 3 and 10. After washing the
cultures with serumꢀfree medium, the cells were treated with 797ꢀnm light (3 J/cm2) and immediately
extracted. Right panel (C): Mice bearing a Colon26 tumor(~6ꢀ8 mm largest dimension) were
injected with compounds 3 or 10 (0.47 ꢀmol/Kg). After 24 hours, the tumors were treated in situ with
787ꢀnm light (ꢀꢀꢀ) and then immediately excised, Each tumor mass was cut into two parts (a and b)
and extracted. Aliquots for the extracts from tissue cell cultures and from the tumor pieces,
containing equal amount of proteins (40 ꢀg), were analyzed by Western blotting for STAT3. The
percent conversion of monomeric STAT3 into homodimeric crosslinked STAT3 was calculated and
indicated at the top each lane.
Compound 3 produces limited skin phototoxicity:
To investigate the skin phototoxicity of 3, nonꢀtumor bearing mice were injected with the
therapeutic dose (0.25 ꢀmol/kg) of 3. A various times of postꢀinjection (day 1 to 7) one of the
hind feet of each mice (3 mice/group/time point) was irradiated with solar light, the other foot
was used as a control35. The severity of the skin reaction was graded according to Table 1
(Experimental Section) and a foot response score was assigned to each mouse36. Mice were
20
ACS Paragon Plus Environment

Page 21 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scored daily and the average foot response was plotted against the day of light treatment. The
results summarized in Fig. 15 indicate that mice treated with laser light at 24 h after injecting the
PS had the strongest PDT response and were classified as moderate to strong edema. Reꢀ
examination 24 h later showed that the feet were healing and after an additional 48h the feet of
these mice returned to pretreatment level. Mice treated with light at 48ꢀ96 h postꢀinjection had a
less severe response than the 24 h postꢀinjection group, and in all cases the mice responded
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
Figure 15: Skin phototoxicity assessment of NIR
PS 3. BALB/c mice were injected at a dose of 0.25
ꢁmol/kg (therapeutic dose). At intervals of 24 h
(circle), 48 h (square), 72 h (triangle) and 96 h
(diamond) post injection mice were restrained and
one hind foot was illuminated with solar light for 30
min. The foot response for each mouse (3 mice/
group) was observed and scored ranging from 0
(no apparent difference from normal untreated
foot) to 3 (loss of foot). Data plotted are from a
single experiment, error bars represent standard
deviation.
well and their feet returned to preꢀtreatment level by 4ꢀ5 days of postꢀtreatment. These results
are similar to those observed for 1, which shows limited skin phototoxicity not only in Colon26
bearing mouse model, but also in patients undergoing PDT treatment.31
1.1. Determination of logP values of 3 and 10
The Pallas program37 which is currently wellꢀaccepted in drug development by most of
the pharmaceutical companies was used to determine the logP values. The calculated logP
values for photosensitizers 3 and 10 are 6. 32 and 6.11 respectively.
Stability of methyl ester functionality of 3 in biological media
To confirm the stability of ester group in PS 3 in biological media, the compound was
incubated in Colon26 cells for 24 h, the media was removed and the cells were washed.
Methanol was then added to cells, stirred and centrifuged. The pallet of cells was removed. The
solvent was concentrated by blowing a slow stream of nitrogen. It was then chromatographed in
analytical TLC with compounds 3 (methyl ester) and 10 (carboxylic acid), The hydrolysis of
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 42
1
2
3
4
5
6
7
8
methyl ester to carboxylic acid was not observed (Figure S22, Supplemental Information).
The metabolic studies of PS 3 at variable doses in rats and dogs (GLP facility) following the US
FDA guidelines are underway.
9
CONCLUSION
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
Our present study indicates that the presence and nature of the fused exocyclic ring
structure in the tetrapyrrolic system makes a significant difference in photophysical
characteristics of the compounds. For example 1, a pyropheophorbideꢀa analog containing a
five member isocyclic ring shows long wavelength absorption in methanol at 660 nm, whereas
the purpurinimide 2 in which the five member ring is replaced with a six member imide ring
system exhibits its long wavelength absorption near 700 nm. Reducing the ring ‘B’ of
purpurinimide 2 to produce bacteriopurpurinimide 3 further extends its long wavelength
absorption to 782 nm, a shift of 87 nm with a remarkable difference in the Stokes shift between
the absorption and fluorescence emission. The in vivo data (imaging and therapy) discussed in
this manuscript are based on a large number of tumor bearing mice in several groups and
indicate the potential of 3 in longꢀterm tumor cure than that achieved by 1 and 2. The inherent
more favorable optical and biological properties render compound 3 as a NIR fluorophore for
imageꢀguided surgery with an option of photodynamic treatment to a variety of tumor types, and
these studies are currently underway. The fundamental comparison of carboxylic acid and
methyl ester derivatives has also revealed a striking difference in pharmacokinetic properties
that remain to be defined. Similar to most of the porphyrinꢀbased compounds, as expected, 3
also showed photobleaching characteristics on irradiating with light. However, more than 50% of
the PS was still available at postꢀtreatment time, which again makes it a suitable candidate for
PDT.
On the basis of ease of synthesis, in vivo biological efficacy and photophysical
characteristics of the bacteriochlorinꢀbased photosensitizers developed so far in our laboratory,
3 is the most promising dualꢀfunction NIR agent for fluorescence guided PDT.38ꢀ40
22
ACS Paragon Plus Environment

Page 23 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
EXPERIMENTAL SECTION
1. Chemistry Methods.
7
8
9
1.1 Synthesis. All reactions were carried out in heat gunꢀdried glassware, under an
atmosphere of nitrogen and magnetic stirring. Thinꢀlayer chromatography (TLC) was done on
precoated silica gel sheets (layer thickness: 0.25 mm) or aluminum oxide sheets. Column
chromatography was performed either using silica gel 60 (70ꢀ230 mesh) or neutral alumina
grade III. In some cases preparative TLC was used for the purification of compounds. Solvents
were dried following the standard methodology. Purity of the compounds was ascertained by
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
1
TLC or by HPLC analyses. All compounds (including the intermediates were >95% pure. H
NMR spectra were recorded at room temperature in CDCl3 solution using either a Varian
VNMRSꢀ400 spectrometer or a Bruker Avance III 400 spectrometer. All chemicals shifts are
reported in parts per million (δ). 1H NMR (400 MHz) spectra were referenced to residual CHCl3
13
(7.26 ppm) or TMS (0.00 ppm). C NMR (100 MHz) spectra were referenced to CDCl3 (77.02
ppm) or TMS (0.00 ppm). Mass spectrometry analyses were performed at the Mass
Spectrometry Facility, Michigan State University, East Lansing. UVꢀvisible spectra were
recorded on FT UVꢀvisible spectrophotometer using dichloromethane/THF as solvent.
31-(1-Butoxy)ethyl-3-deacetylbacteriopurpurin-18-N-butylimide Methyl Ester (3): Bacteriochlorin
8 (40 mg) was dissolved in dry dichloromethane (5.0 mL). HBr gas was bubbled through the
solution for 2 min, and stirred at room temperature under an argon atmosphere for 5 min. The
solvent and the excess hydrogen bromide were then evaporated under anhydrous high vacuum
(at or below 30 °C) to give a residue, which was immediately redissolved in dry dichloromethane
(10 mL) and nꢀbutanol (0.6 mL to give 100 fold excess, Sigma Aldrich, St. Louis, MO and
American Radiolabeled Chemicals, St. Louis, MO for radioactive) in the presence of anhydrous
potassium carbonate (200mg). The reaction mixture was stirred for 30 min. After the standard
workup, the residue was purified on PTLC using Ethyl acetate/Hexanes (40% v/v), and the title
compound was obtained in as a solid. Yield 35 mg (83%). UVꢀvis (CH2Cl2, λmax, nm, (ε)): 355,
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 42
1
2
3
4
1
366, 416, 472, 537, 702, 787 (5.44 x104); H NMR (400 MHz, CDCl3, δ ppm): 8.83/8.78 (s, 1H,
5
6
7
8
5ꢀH), 8.60 (s, 1H, 10ꢀH), 8.31 (s, 1H, 20ꢀH), 5.66/5.64 (q, 1H, 31ꢀH, J = 6.7 Hz), 5.25 (m, 1H, 17ꢀ
H), 4.42 (m, 2H, ꢀNCH2(CH2)2CH3), 4.11ꢀ4.23 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.00 (m, 1H, 8ꢀ
H), 3.62 (s, 3H, 12ꢀCH3), 3.562/3.561 (s, 3H, ꢀCOOCH3), 3.48ꢀ3.63 (m, 2H, OCH2CH2CH2CH3),
3.23/3.22 (s, 3H, 2ꢀCH3), 2.63 (m, 1H, ꢀCH2CHHCOOMe), 2.22ꢀ2.42 (m, 3H, 1H for ꢀ
CHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for ꢀCH2CHHCOOMe), 1.99 (d, 3H, 31ꢀCH3, J = 6.6
Hz), 1.88ꢀ2.12 (m, 4H, 1H for 8ꢀCHHCH3, 1H for 17ꢀCHHCH2COOMe, 2H for ꢀ
NCH2CH2CH2CH3), 1.79/1.78 (d, 3H, 7ꢀCH3, J = 7.1 Hz), 1.66ꢀ1.73 (m, 2H, OCH2CH2CH2CH3),
1.69 (d, 3H, 18ꢀCH3, J = 7.3 Hz), 1.62 (m, 2H, ꢀNCH2CH2CH2CH3), 1.30ꢀ1.52 (m, 2H,
OCH2CH2CH2CH3), 1.13/1.10 (t, 3H, 8ꢀCH2CH3, J = 7.4 Hz), 1.08 (t, 3H, ꢀN(CH2)3CH3, J = 7.4
Hz), 0.86/0.85 (t, 3H, OCH2CH2CH2CH3, J = 7.4 Hz), 0.05 (s, 1H, NH), ꢀ0.32 (s, 1H, NH); 13C
NMR (100 MHz, CDCl3, δ ppm): 173.97, 173.81/173.79, 172.41, 170.44/170.37, 167.73, 163.92,
160.89/160.84, 141.12/141.03, 138.26/138.20, 137.63/137.62, 133.26, 132.48/132.46,
131.51/131.50, 129.42/129.41, 113.97/113.96, 101.37/101.35, 99.78/99.77, 99.31/99.13,
94.76/94.73, 72.56/72.37, 69.42/69.26, 55.36/55.26, 53.76/53.73, 51.47, 49.33, 48.25/48.23,
40.03, 32.28/32.22, 32.22/32.17, 31.24, 31.11, 30.29/30.20, 24.47/24.33, 23.57/23.55,
22.87/22.71, 20.77, 19.57/19.55, 14.07, 13.93, 11.92, 10.85/10.82, 10.81/10.68. HRMS (ESI)
for C42H56N5O5 [MH+] calculated: 710.4266; found: 710.4282. Elemental analysis calculated for
C42H55N5O5, C 71.06, H 7.81, N 9.87. Found: C 71.29, H 8.10, N 9.84.
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
Besides the desired major compound 3, a minor product 9 in 5ꢀ7% yield was also
isolated and was identified as 3ꢀvinylꢀbacteriopurpuinꢀ18ꢀNꢀbutylimide methyl ester, possibility
obtained by dehydration or dehyrobromination of the starting material or the intermediate
1
unstable bromoderivative. H NMR (400 MHz, CDCl3, δ ppm): 8.62 (s, 1H, mesoꢀH), 8.49 (s,
1H, mesoꢀH), 8.38 (s, 1H, mesoꢀH), 7.77 (dd, 1H, 31ꢀH, J = 11.5, 17.9 Hz), 6.18 (dd, 1H, 32ꢀH, J
= 1.3, 17.9 Hz), 6.07 (dd, 1H, 32ꢀH, J = 1.4, 11.6 Hz), 5.26 (m, 1H, 17ꢀH), 4.43 (m, 2H, ꢀ
24
ACS Paragon Plus Environment

Page 25 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NCH2(CH2)2CH3), 4.15ꢀ4.24 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.02 (m, 1H, 8ꢀH), 3.63 (s, 3H, 12ꢀ
CH3), 3.57 (s, 3H, ꢀCOOCH3), 3.28 (s, 3H, 2ꢀCH3), 2.65 (m, 1H, ꢀCH2CHHCOOMe), 2.25ꢀ2.43
(m, 3H, 1H for ꢀCHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for ꢀCH2CHHCOOMe), 1.88ꢀ2.11 (m,
4H, 1H for 8ꢀCHHCH3, 1H for 17ꢀCHHCH2COOMe, 2H for ꢀNCH2CH2CH2CH3), 1.79 (d, 3H, 7ꢀ
CH3, J = 7.2 Hz), 1.70 (d, 3H, 18ꢀCH3, J = 7.3 Hz), 1.62 (m, 2H, ꢀNCH2CH2CH2CH3), 1.11 (t, 3H,
8ꢀCH2CH3, J = 7.4 Hz), 1.08 (t, 3H, ꢀN(CH2)3CH3, J = 7.4 Hz), 0.01 (s, 1H, NH), ꢀ0.34 (s, 1H,
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
13
NH); C NMR (100 MHz, CDCl3, δ ppm): 174.0, 173.5, 172.5, 170.3, 167.7, 163.9, 161.0,
141.1, 138.2, 134.0, 133.4, 131.7, 131.6, 129.8, 128.7, 122.4, 114.1, 101.4, 99.7, 98.4, 95.3,
55.4, 53.8, 51.5, 49.3, 48.2, 40.0, 32.3, 31.3, 31.1, 30.2, 23.6, 22.8, 20.8, 14.1, 11.94, 11.92,
10.7. HRMS (ESI) for C38H46N5O4 [MH+] calculated: 636.3584; found: 636.3541. HPLC retention
time: 13.73 min (see Figure S20b, purity > 96.4% Supplemental Information).
1.2 HPLC Method for the separation of isomers present in bacteriopurpurinimde 3a and
Isomer 3b. The purity of isolated isomers 3a and 3b eluted by column chromatography were
injected on an analytical column (Waters C18 Symmetry, 4.6 x 150 mm, 5 m particle size),
mobile phase 100% methanol at 1.0 ml/min). The retention times of 3a and 3b were 15.39 min
and 16.44 min respectively.
Isomer 3a (retention time 15.39 min): UVꢀvis (MeOH, λmax, nm, (ε)): 344, 366, 416, 537, 702,
1
780 (4.73 x104); H NMR (400 MHz, CDCl3, δ ppm): 8.79 (s, 1H, 5ꢀH), 8.60 (s, 1H, 10ꢀH), 8.31
(s, 1H, 20ꢀH), 5.66 (q, 1H, 31ꢀH, J = 6.7 Hz), 5.25 (m, 1H, 17ꢀH), 4.42 (m, 2H, ꢀNCH2(CH2)2CH3),
4.11ꢀ4.23 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.01 (m, 1H, 8ꢀH), 3.62 (s, 3H, 12ꢀCH3), 3.47ꢀ3.64
(m, 2H, OCH2CH2CH2CH3), 3.563 (s, 3H, ꢀCOOCH3), 3.23 (s, 3H, 2ꢀCH3), 2.63 (m, 1H, 17ꢀ
CH2CHHCOOMe), 2.21ꢀ2.42 (m, 3H, 1H for ꢀCHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for 17ꢀ
CH2CHHCOOMe), 2.00 (d, 3H, 31ꢀCH3, J = 6.6 Hz), 1.88ꢀ2.12 (m, 4H, 1H for 8ꢀCHHCH3, 1H for
17ꢀCHHCH2COOMe, 2H for ꢀNCH2CH2CH2CH3), 1.79 (d, 3H, 7ꢀCH3, J = 7.2 Hz), 1.66ꢀ1.77 (m,
2H, OCH2CH2CH2CH3), 1.69 (d, 3H, 18ꢀCH3, J = 7.3 Hz), 1.62 (m, 2H, ꢀNCH2CH2CH2CH3),
1.31ꢀ1.52 (m, 2H, OCH2CH2CH2CH3), 1.10 (t, 3H, 8ꢀCH2CH3, J = 7.4 Hz), 1.08 (t, 3H, ꢀ
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
N(CH2)3CH3, J = 7.4 Hz), 0.86 (t, 3H, OCH2CH2CH2CH3, J = 7.4 Hz), 0.05 (s, 1H, NH), ꢀ0.32 (s,
1H, NH); 13C NMR (100 MHz, CDCl3, δ ppm): 173.99, 173.81, 172.41, 170.38, 167.72, 163.92,
160.83, 141.11, 138.18, 137.60, 133.24, 132.48, 131.49, 129.41, 113.93, 101.35, 99.76, 99.13,
94.75, 72.35, 69.25, 55.24, 53.74, 51.49, 49.30, 48.24, 40.03, 32.20, 32.16, 31.21, 31.10, 30.21,
24.34, 23.55, 22.88, 20.77, 19.57, 14.08, 13.93, 11.93, 10.85, 10.68; MS (ESI): m/z 710 (M+).
HPLC chromatograms: (Figure 3 of the main text, purity > 99%).
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
Isomer 3b (retention time: 16.94 min): UVꢀvis (MeOH, λmax, nm, (ε)): 344, 366, 416, 537, 702,
1
780 (4.73 x104); H NMR (400 MHz, CDCl3, δ ppm): 8.83 (s, 1H, 5ꢀH), 8.59 (s, 1H, 10ꢀH), 8.31
(s, 1H, 20ꢀH), 5.64 (q, 1H, 31ꢀH, J=6.7 Hz), 5.25 (dd, 1H, 17ꢀH, J=2.8, 8.8 Hz), 4.42 (m, 2H, ꢀ
NCH2(CH2)2CH3), 4.14ꢀ4.23 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.00 (m, 1H, 8ꢀH), 3.62 (s, 3H, 12ꢀ
CH3), 3.59 (m, 1H, OCHHCH2CH2CH3), 3.560 (s, 3H, ꢀCOOCH3), 3.52 (m, 1H,
OCHHCH2CH2CH3), 3.22 (s, 3H, 2ꢀCH3), 2.63 (m, 1H, ꢀCH2CHHCOOMe), 2.21ꢀ2.42 (m, 3H, 1H
for ꢀCHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for ꢀCH2CHHCOOMe), 1.99 (d, 3H, 31ꢀCH3, J=6.7
Hz), 1.87ꢀ2.12 (m, 4H, 1H for 8ꢀCHHCH3, 1H for 17ꢀCHHCH2COOMe, 2H for ꢀ
NCH2CH2CH2CH3), 1.77 (d, 3H, 7ꢀCH3, J = 7.2 Hz), 1.65ꢀ1.73 (m, 2H, OCH2CH2CH2CH3), 1.69
(d, 3H, 18ꢀCH3, J = 7.3 Hz), 1.61 (m, 2H, ꢀNCH2CH2CH2CH3), 1.38 (m, 2H, OCH2CH2CH2CH3),
1.12 (t, 3H, 8ꢀCH2CH3, J = 7.4 Hz), 1.08 (t, 3H, ꢀN(CH2)3CH3, J = 7.4 Hz), 0.85 (t, 3H,
OCH2CH2CH2CH3, J = 7.4 Hz), 0.05 (s, 1H, NH), ꢀ0.32 (s, 1H, NH); 13C NMR (100 MHz, CDCl3,
δ ppm): 173.99, 173.79, 172.41, 170.45, 167.72, 163.92, 160.88, 141.02, 138.24, 137.62,
133.25, 132.49, 131.48, 129.40, 113.93, 101.37, 99.74, 99.32, 94.73, 72.55, 69.42, 55.34,
53.71, 51.48, 49.31, 48.21, 40.03, 32.27, 32.20, 31.21, 31.11, 30.29, 24.48, 23.57, 22.72, 20.77,
19.55, 14.08, 13.94, 11.93, 10.82, 10.81; MS (ESI): m/z 710 (M+). HPLC Chromatograms:
(Figure 3 of the main text, purity > 99%).
3-Acetylbacteriopurpurin-18-N-butylimide Methyl Ester (7): 3ꢀAcetylbacteriopurpurinꢀ18 methyl
ester 6 was prepared from methyl bacteriopheophorbideꢀa 4 following a published procedure.27
3ꢀAcetylbacteriopurpurinꢀ18 methyl ester 6 (300 mg) dissolved in CH2Cl2 (25 mL) was treated
26
ACS Paragon Plus Environment

Page 27 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with a butylamine (4.0 mL) at room temperature for 18 h. The disappearance of the absorption
at 813 nm and the appearance of a new absorption peak at 753 nm indicated the formation of
the intermediate amide system. The reaction mixture was then treated with diazomethane for 10
min and then diazomethane was removed by passing argon through it for 30 min. This was
further treated with a catalytic amount of methanolic KOH was added and the mixture was
stirred for 1ꢀ2 min. The disappearance of the absorption at 753 nm and appearance of a new
peak 822 nm indicated the completion of the reaction. Following the standard workup, the
residue was chromatographed over a grade III alumina column eluting with 30%
hexanes/CH2Cl2. The appropriate fractions were combined. Yield 176 mg (53%). UVꢀvis
(CH2Cl2, λmax, nm): 363, 414, 546, 753, 822; 1H NMR (400 MHz, CDCl3, δ ppm): 9.23 (s, 1H, 5ꢀ
H), 8.81 (s, 1H, 10ꢀH), 8.63 (s, 1H, 20ꢀH), 5.31 (m, 1H, 17ꢀH), 4.44 (m, 2H, ꢀNCH2(CH2)2CH3),
4.23ꢀ4.34 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.11 (m, 1H, 8ꢀH), 3.70 (s, 3H, 12ꢀCH3), 3.57 (s, 3H,
ꢀCOOCH3), 3.55 (s, 3H, 2ꢀCH3), 3.17 (s, 3H, COCH3 ), 2.68 (m, 1H, ꢀCH2CHHCOOMe), 2.28ꢀ
2.43 (m, 3H, 1H for ꢀCHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for ꢀCH2CHHCOOMe), 2.07 (m,
1H, 8ꢀCHHCH3), 1.89ꢀ2.02 (m, 3H, 1H for 17ꢀCHHCH2COOMe, 2H for ꢀNCH2CH2CH2CH3), 1.82
(d, 3H, 7ꢀCH3, J=7.3 Hz), 1.73 (d, 3H, 18ꢀCH3, J=7.3 Hz), 1.63 (m, 2H, ꢀNCH2CH2CH2CH3), 1.11
(t, 3H, 8ꢀCH2CH3, J=7.4 Hz), 1.09 (t, 3H, ꢀN(CH2)3CH3, J=7.4 Hz), ꢀ0.52 (s, 1H, NH), ꢀ0.73 (s,
1H, NH); 13C NMR (100 MHz, CDCl3, δ ppm): 198.6, 173.9, 172.8, 169.74, 169.73, 167.5,
164.5, 163.5, 136.5, 135.6, 135.4, 134.7, 134.3, 133.7, 131.7, 116.3, 101.9, 101.7, 99.2, 97.1,
56.4, 54.4, 51.5, 48.5, 47.5, 40.1, 33.2, 32.3, 31.2, 31.1, 30.0, 24.0, 23.3, 20.7, 14.1, 13.6, 12.2,
10.8. HRMS (ESI) for C38H46N5O5 [MH+] calculated: 652.3457; found: 652.3492. Elemental
analysis calculated for C38H45N5O5, C 70.02, H 6.96, N 10.74. Found: C 69.88, H 6.86, N 10.57.
3-(1-Hydroxy)ethyl-3-deacetylbacteriopurpurin-18-N-butylimide Methyl Ester (8): Bacteriochlorin
7 (170 mg) was dissolved in 25 mL CH2Cl2. Sodium borohydride (360mg) in two lots was added
and 3 mL methanol were added to the reaction mixture. The reaction was monitored by TLC
and UVꢀvis spectrophotometry. The disappearance of the 822 nm band and the appearance of
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
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 42
1
2
3
4
5
6
7
8
a new band at 787 nm indicated the completion of the reaction (reaction time: 20 min). This was
then neutaralized using 2% acetic acid/water and then extracted with CH2Cl2.(3X100ml), dried
over sodium sulphate and concentrated at rotary evaporator to give 8. Yield quantitative. UVꢀvis
9
1
(CH2Cl2, λmax, nm): 366, 417, 534, 726, 787; H NMR (400 MHz, CDCl3, δ ppm): 8.82/8.80 (s,
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
1H, 5ꢀH), 8.61 (s, 1H, 10ꢀH), 8.33 (s, 1H, 20ꢀH), 6.21 (q, 1H, 31ꢀH, J=6.6 Hz), 5.24 (m, 1H, 17ꢀ
H), 4.42 (m, 2H, ꢀNCH2(CH2)2CH3), 4.14ꢀ4.24 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 4.01 (m, 1H, 8ꢀ
H), 3.62 (s, 3H, 12ꢀCH3), 3.57/3.56 (s, 3H, ꢀCOOCH3), 3.27 (s, 3H, 2ꢀCH3), 2.63 (m, 1H, ꢀ
CH2CHHCOOMe), 2.52/2.50 (d, 1H, 31ꢀOH, J =1.6 Hz), 2.24ꢀ2.42 (m, 3H, 1H for ꢀ
CHHCH2COOMe, 1H for 8ꢀCHHCH3, 1H for ꢀCH2CHHCOOMe), 2.05 (d, 3H, 31ꢀCH3, J = 6.6
Hz), 1.88ꢀ2.10 (m, 4H, 1H for 8ꢀCHHCH3, 1H for 17ꢀCHHCH2COOMe, 2H for ꢀ
NCH2CH2CH2CH3), 1.79/1.78 (d, 3H, 7ꢀCH3, J = 7.2 Hz), 1.69/1.68 (d, 3H, 18ꢀCH3, J = 7.3 Hz),
1.62 (m, 2H, ꢀNCH2CH2CH2CH3), 1.11 (t, 3H, 8ꢀCH2CH3, J = 7.3 Hz), 1.08 (t, 3H, ꢀN(CH2)3CH3,
J = 7.4 Hz), ꢀ0.03 (s, 1H, NH), ꢀ0.39 (s, 1H, NH); 13C NMR (100 MHz, CDCl3, δ ppm):
173.99/173.97, 173.61/173.60, 172.45/172.43, 170.2, 167.7, 163.9, 161.06/161.05,
140.83/140.82, 139.17/139.14, 137.20/137.11, 133.4, 131.6, 131.49/131.45, 129.74/129.73,
114.1, 101.4, 99.75/99.74, 99.32/99.29, 94.98/94.94, 65.54/65.50, 55.4, 53.81/53.76, 51.5, 49.3,
48.22/48.18, 40.0, 32.27/32.24, 31.3, 31.1, 30.2, 25.64/25.62, 23.60/23.57, 22.90/22.82, 20.8,
14.1, 11.9, 11.13/11.11, 10.77/10.76. HRMS (ESI) for C38H48N5O5 [MH+] calculated: 654.3627;
found: 654.3649.
31-(1-Butoxy)ethyl-3-deacetylbacteriopurpurin-18-N-butylimide carboxylic acid (10): In a dry RB
flask, 60 mg of 3 was dissolved in 40 ml of dry acetonitrile and stirred well. A 650 mg of LiOH
dissolved in 13 ml of water was added slowly. The entire reaction mixture was stirred overnight
at room temperature. TLC after 18 hours showed some starting material. Added 100 mg of LiOH
in 2ml of water and stirred for another 4 hours. The mixture was washing with 2% Acetic
acid/water solvent mixture. The combined organic layer was dried over sodium sulphate and the
filtrate was evaporated under reduced pressure. The crude obtained was purified by silica gel
28
ACS Paragon Plus Environment

Page 29 of 42
Journal of Medicinal Chemistry
1
2
3
4
PTLC using 55% hexanes/Ethyl acetate. Yield 24 mg (40%). UVꢀvis (CH2Cl2, λmax, nm, (ε)): 365,
5
6
1
413, 472, 537, 701, 784 (5.44 x104); H NMR (400 MHz, CDCl3, δ ppm): 8.83/8.78 (s, 1H, 5ꢀH),
7
8
9
8.58 (s, 1H, 10ꢀH), 8.30 (s, 1H, 20ꢀH), 5.64/5.62 (q, 1H, 31ꢀH, J = 6.7 Hz), 5.22 (m, 1H, 17ꢀH),
4.40 (m, 2H, ꢀNCH2(CH2)2CH3), 4.10ꢀ4.23 (m, 2H, 1H for 7ꢀH, 1H for 18ꢀH), 3.99 (m, 1H, 8ꢀH),
3.60 (s, 3H, 12ꢀCH3), 3.46ꢀ3.63 (m, 2H, OCH2CH2CH2CH3), 3.21/3.20 (s, 3H, 2ꢀCH3), 2.63ꢀ2.74
(m, 1H, ꢀCH2CHHCOOH), 2.25ꢀ2.41 (m, 3H, 1H for ꢀCHHCH2COOH, 1H for 8ꢀCHHCH3, 1H for ꢀ
CH2CHHCOOH), 1.99/1.98 (d, 3H, 31ꢀCH3, J = 6.7 Hz), 1.86ꢀ2.09 (m, 4H, 1H for 8ꢀCHHCH3, 1H
for 17ꢀCHHCH2COOH, 2H for ꢀNCH2CH2CH2CH3), 1.78/1.77 (d, 3H, 7ꢀCH3, J = 7.1 Hz), 1.64ꢀ
1.71 (m, 2H, OCH2CH2CH2CH3), 1.67 (d, 3H, 18ꢀCH3, J = 7.3 Hz), 1.59 (m, 2H, ꢀ
NCH2CH2CH2CH3), 1.30ꢀ1.50 (m, 2H, OCH2CH2CH2CH3), 1.11/1.08 (t, 3H, 8ꢀCH2CH3, J = 7.4
Hz), 1.05 (t, 3H, ꢀN(CH2)3CH3, J = 7.4 Hz), 0.85 (t, 3H, OCH2CH2CH2CH3, J = 7.4 Hz), 0.05 (s,
1H, NH), ꢀ0.31 (s, 1H, NH); 13C NMR (100 MHz, CDCl3, δ ppm): 177.44, 173.71/173.70, 172.21,
170.50/170.43, 167.86, 163.86, 160.92/160.87, 141.19/141.10, 138.32/138.26, 137.65/137.64,
133.22, 132.53/132.51, 131.51/131.50, 129.44/129.43, 113.91/113.90, 101.42/101.40,
99.67/99.66, 99.42/99.23, 94.76/94.74, 72.55/72.35, 69.42/69.25, 55.33/55.23, 53.76/53.74,
49.27, 48.24/48.22, 40.04, 32.26/32.15, 31.90, 31.08, 30.83, 30.26/30.18, 24.45/24.31,
23.55/23.53, 22.85/22.69, 20.72, 19.54, 14.03, 13.92/13.91, 11.90, 10.82/10.80, 10.78/10.66.;
MS (ESI): m/z 695.5 (M+). HPLC chromatograms: Retention times 9.49 and 10.00 minutes,
purity: 96.52%, Figure S21c, Supplemental Information.
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
1.3 Formulation of the photosensitizers. 1% Tween 80. The amount of compound and
volume of solution were calculated for a desired concentration. The final solution was made to
be a 1% Tween 80/D5W solution. The compound and necessary volume of Tween80 were
added to a mortar and mulled to a paste. The resulting paste was allowed to sit overnight and
the next day the calculated amount of D5W was added to the paste and mixed. The solution
was the filtered through a 0.2 ꢁm syringe filter and the concentration was measured
spectrophotometrically. The resulting drug solutions were
29
ACS Paragon Plus Environment