Hemisynthesis of 132,173-cyclomesopheophorbide-a-enol

Article abstract of DOI:10.1016/j.tetlet.2010.12.079

Cyclopheophorbides are often major chlorophyll-a degradation products in biogeochemical situations. Authentic standards are unavailable and facile generation of pure compounds is required. 13²,17³- Cyclomesopheophorbide-a-enol (mesoCYCLO) was prepared in moderate yields via a known Dieckmann-like condensation of mesopyropheophorbide-a methyl ester (mpPBIDaME). MesoCYCLO was purified by flash chromatography over a polymeric reversed phase (PRP-1) material, negating the requirement for a crystallization step. Structural verification included ultraviolet-visible spectrometry, high resolution matrix (sulfur/CS₂) assisted laser desorption mass spectrometry, and nuclear magnetic resonance.

Full text of DOI:10.1016/j.tetlet.2010.12.079

Tetrahedron Letters 52 (2011) 1078–1081
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Hemisynthesis of 132,173-cyclomesopheophorbide-a-enol
⇑
Mitra (Mortezaei-Rad) Khalesi, J. William Louda
Department of Chemistry and Biochemistry, Florida Atlantic University, Boca Raton, FL 33431, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Cyclopheophorbides are often major chlorophyll-a degradation products in biogeochemical situations.
Authentic standards are unavailable and facile generation of pure compounds is required. 132,173-Cycl-
omesopheophorbide-a-enol (mesoCYCLO) was prepared in moderate yields via a known Dieckmann-like
condensation of mesopyropheophorbide-a methyl ester (mpPBIDaME). MesoCYCLO was purified by flash
chromatography over a polymeric reversed phase (PRP-1™) material, negating the requirement for a
crystallization step. Structural verification included ultraviolet-visible spectrometry, high resolution
matrix (sulfur/CS2) assisted laser desorption mass spectrometry, and nuclear magnetic resonance.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 21 August 2010
Revised 9 December 2010
Accepted 19 December 2010
Available online 23 December 2010
Keywords:
Mesocyclopheophorbide-a-enol
Cyclomesopheophorbide-a-enol
Chlorophyll
Pigments
Tetrapyrroles
Cyclopheophorbides-a-enol (CYCLO, 2a) is a well known bio-
geochemical degradation product of chlorophyll-a.1,2 Several re-
ports reveal the generation of CYCLO from chlorophyll-a (CHLa)
in filter feeding marine organisms, such as sponges3 and mol-
lusks.4–6 Geochemically, CYCLO is reported from a variety of sedi-
ments1,7 and has been observed to form directly from
pyropheophorbide-a (pPBIDa, 1a1) in a downhole sequence of sul-
fidic carbonates.8 The exocyclic seven-membered ring forms via a
dehydration–cyclization of pyropheophorbide-a free acid (1a1).
Interest in the cyclopheophorbides also arises from their apparent
direct relationship as diagenetic precursors to a variety of bicy-
cloalkylporphyrins (BiCAP) in oil shales and petroleum
crudes.2,7,9,10
CYCLO (2a) has been synthesized (Scheme 1) in vitro by Dieck-
mann-like (intramolecular Claisen)11–13 cyclizations of pyropheo-
phorbide-a methyl ester (1a2). Though a wide variety of 132,173-
cyclopheophorbide-enols were synthesized by Falk et al.,11 as far
as we could find, mesoCYCLO has not been formed in vitro until
the present Letter.
The cyclopheophorbides have been found to be extremely
unstable and often oxidatively rearrange to 132-(S/R)-hydroxychlo-
rophyllones (3), 132-oxopheo-phorbide-a (4), and/or chlorophyl-
lonic acid-a (5) and are therefore often referred to as anti-
oxidants.4,5,14–17 However, to date, there is no proof that these
compounds are physiologically active as such in nature. The struc-
tures of these compounds are given in Figure 1.
In this Letter, we report the hemisynthesis of CYCLO (2a) and
describe the analogous generation of mesoCYCLO (2b) from meso-
pyropheophorbide-a methyl ester (mpPBIDaME, 1b). Each proce-
dure followed the more recently modified methodology of
Ocampo et al.13;cf11,15 Additionally, we present proof (UV/Vis, 1H
NMR, HR-MALDI-TOF) of the structure of mesoCYCLO (2b) and
add to the NMR data on CYCLO (2a). Significant analytical develop-
ment was required and we describe the successful and facile puri-
fication of this and other cyclopheophorbides, as well as providing
UV/Vis, mass and NMR spectra of the highly purified pigments.
All procedures were performed either in the dark or subdued
yellow light and solutions were kept cold/frozen and under argon
whenever possible.mpPBIDaME (1b) was prepared from hydroge-
nation of the vinyl group of pPBIDaME18 (1a2) as follows; pPBIDa-
ME(100 mg, 0.18 mmol) was dissolved in 99.9% anhydrous
tetrahydrofuran (THF, 25 mL) and added to 4 mg (10%) palladium
on charcoal catalyst.19 This follows the procedure outlined by Jean-
don et al.20 Hydrogenation was performed under 1 atmosphere
pressure (relative) for 200 min in a Parr hydrogenator using an
ACE glass reaction vessel. The catalyst was removed by filtering
the mixture through Celite and the solvent was removed in vacuo.
CYCLO (2a) or mesoCYCLO (2b) was prepared from pPBIDaME
(1a2) or mpPBIDaME (1b), respectively, following a reported proce-
dure13 according to Scheme 1. For example, in the case of mesoCY-
The title compound of the present study, 132,173-cyclomes-
opheophorbide-a-enol (mesoCYCLO, 2b), has been identified by
LC–PDA–MS as a minor constituent of sedimentary organic matter
in Peru margin sediments.8 In that Letter, the authors identified
mesoCYCLO (2b) as cyclopheophorbide-518 (CPP518), indicating
the nominal mass of the pigment. Previously, our group2 has also
termed CYCLO (2a) as phorbide686 and mesoCYCLO (2b) as chlo-
rin678, indicating the wavelengths (nm) of their band I absorptions
in ethyl ether.
⇑
Corresponding author. Tel.: +1 561 297 3309; fax: +1 561 297 2759.
E-mail address: blouda@fau.edu (J.W. Louda).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.12.079

Mitra (Mortezaei-Rad) Khalesi, J. W. Louda / Tetrahedron Letters 52 (2011) 1078–1081
1079
R1
R1
R1
N
NH
N
N
N
NH
NH
(TMS)2NNa
HN
HN
HN
N
N
H
O
H
O
O
H
C
OH
O
OR2
O
'intermediate'
1a1 R1 = -CH=CH2 R2 = -H
1a2 R1 = -CH=CH2 R2 = -CH3
1b
2a
R1= -CH2-CH3 R2 = -CH3
R1 = -CH=CH2 2b R1= -CH2-CH3
Scheme 1. Structural comparisons and cyclization reaction forming the cyclopheophorbide-a-enols.
N
N
N
NH
N
NH
N
NH
N
HN
HN
HN
COOCH3
O
O
O
O
OH
OR
O
(3)
(5)
(4)
O
O
Figure 1. Structures14–16 of oxidatively generated degradation products of (2a) CYCLO: (3) 132(S/R)-hydroxychlorophyllones, (4) 132-oxopheophorbide-a, (5) chlorophyllonic
acid-a.
CLO (2b), mpPBIDaME (1b, 55.07 mg, 1.09 mmol) was added to
6 mL tetrahydrofuran (THF; 99.9% anhydrous inhibited with
250 ppm BHT) under a stream of ultra-high purity (UHP)-argon.
Ten milliliters (1.0 mmol) of sodium bis(trimethylsilyl)-amide21
in THF was added to that solution and stirred for about 8 min.
The reaction mixture was quenched in an Ar-sparged deoxygen-
ated mixture of CH2Cl2 (80 mL), saturated NaH2PO4 (20 mL) and
degassed deionized ice (20 g). Following liquid:liquid extraction,
the organic (CH2Cl2) layer was dried over anhydrous Na2SO4, fil-
tered and evaporated in vacuo.
The crude product was then flash chromatographed over poly-
meric reversed phase (PRP-1™) using an isocratic solvent (90% ace-
tonitrile) as we detailed previously.22 This rapid and mild
purification also negated the need for crystallization/recrystalliza-
tion procedures. As we were interested in only the pure product,
we did not trace yield, which would be difficult as we ‘heart cut’
the main fraction of the target compound (2b). A very gentle puri-
fication method was required since cyclopheophorbide-a-enols
have been found to be very unstable compounds that can be easily
oxidized to a variety of alternate cyclic pheophorbides. CYCLO (2a)
is routinely altered to different oxidative artifacts, mainly 132(S/R)-
hydroxychlorophyllones-a (3), during chromatographic and other
isolation/purification procedures.14–16,22 As with CYCLO (2a), when
mesoCYCLO (2b) is chromatographed over normal phase silica or
alumina in conventional column chromatography or analytical
HPLC, it is mainly oxidized to artifacts and becomes barely detect-
able. The main artifacts compared to CYCLO (2a) analog have been
found to be highly polar and are presumed to be the meso-chloro-
phyllonic acids (cf compd 22 in Ref. 15). To prevent undesirable
oxidations, and due to successful purification using PRP-1™ as
packing material,22 a PRP-1™ analytical or semi-prep column
was used for HPLC analysis. The HPLC (PRP-1™) chromatogram
for the purification of mesoCYCLO (2b) is presented in Figure 2.
The two peaks immediately following the solvent front are the arti-
facts generated over the column. To confirm this hypothesis, we
collected a ‘heart cut’ of the mesoCYCLO (2b) peak between
ꢀ23–27 min and reinjected it. This was repeated several times
and each time a new pair of these artifacts was generated from a
previous run’s heart cut. The reappearance of these two peaks on
the chromatogram supported our hypothesis of artifact formation
being directly related to injectate preparation and injection into
the HPLC system. This occurred regardless of continual He sparging
of solvents and/or the addition of anti-oxidants such as butylated-
hydroxytoluene (BHT) or ascorbic acid to the injectate and devel-
oping solvent.
Like cyclopheophorbide-a-enol (CYCLO, 2a), the Soret Band in
the UV/Vis spectrum of pure mesoCYCLO (2b) is complex and con-
sists of at least 4 overlapping individual bands. Additionally, the
spectrum contains an intense band
I appearing at 676 nm
(Fig. 3a), which exhibits a bathochromic shift compared to the par-
ent compound mesopyropheophorbide-a methyl ester (1b;
k1 = 666 nm). Both band I and the Soret band exhibit hypsochromic
shifts when compared to the main maxima of the CYCLO analog
(Fig. 3b). This is expected and occurs as a consequence of reduction
of the vinyl to ethyl at position 3 of the macrocycle.8,23 The UV/Vis
spectrum, an overall characteristic of the cyclopheophorbides, has
broad split Soret(S) bands. Absorption maxima recorded in the
HPLC eluant and relative intensities are given in brackets:
kS = 358 [1.000], (407) [0.796], 423 [0.907], 449 [0.593], kIV = 528
[0.065], kIII = 572 [0.093], kII = 619 [0.139], kI = 676 [0.426] nm.

1080
Mitra (Mortezaei-Rad) Khalesi, J. W. Louda / Tetrahedron Letters 52 (2011) 1078–1081
32
31
0.03
7
3
mesoCYCLO
8
0.02
0.01
2
N
ARTIFACTS
NH
HN
N
12
18
0.00
13
131
17
132
171
0
10
20
30
Time (min)
40
50
O
172
173
OH
Figure 2. HPLC chromatogram of pure mesoCYCLO (2b) on analytical PRP-1™.
Figure 5. Structure of 132,173-cyclomesopheophorbide-a (mesoCYCLO, 2b) with
proton containing carbons numbered. CYCLO (2a) has a double bond between 31
and 32 (vinyl in place of ethyl).
Electronic absorption spectra ( Fig. 3a and b) were recorded in
CH2Cl2 with a Perkin–Elmer Model Lambda-2 Spectrometer which
is routinely standardized versus holmium oxide.
We obtained high resolution MALDI-TOF mass spectra24 of mes-
oCYCLO (2b) in 1:1 ratio with a carbon disulfide (CS2)/sulfur ma-
trix, an example being given herein as Figure 4. The matrix
derives from the method of Brune.25 The base peak equates to
the mesoCYCLO (2b) molecular ion (M+ = 518.3068 m/z) and the
other peaks below ꢀm/z = 400 belong to the sulfur matrix
(S8 = 256; Matrix = 256 32 m/z). Constraining elemental composi-
tion to C, H, N and O, the mass spectral software calculated a
molecular formula of C33H34N4O2 with an expected exact mass of
518.6487 Da. Sporadically, the dimer (1,035.0051 m/z; expec-
ted = 1,035.2835 Da) would be present in low (65% of 518 m/z) rel-
ative amounts. MALDI induced dimers are reported with other
compounds as well.26
Nuclear magnetic resonance (NMR) spectra27 were obtained at
400 MHz for pyropheophorbide-a methyl ester (1a2), CYCLO (2a),
mesopyropheophorbide-a methyl ester (1b) and mesoCYCLO (2b)
in order to compare similarities and differences among this family
of dihydroporphyrins. In most cases, NOESY and gCOSY 2D spectra
were also obtained in order to substantiate assignments.
The 400 MHz 1H NMR spectrum peak assignments for both CY-
CLO (2a) and mesoCYCLO (2b) are summarized in Appendix 1. Peak
assignments were made in concert with previous literature.
cf6,10,12,15 Differences in proton shifts for positions 171, 172 between
CYCLO (2a) and mesoCYCLO (2b) may be related to keto–enol tau-
tomerism effects (compds. 8 and 9 in Ref. 7) as reported to occur by
Ocampo et al.7
For the consideration of the NMR data, the structure and carbon
numbering system for CYCLO (2a) and mesoCYCLO (2b) are given
herein as Figure 5.
In summary, the preparation, rapid purification, and spectro-
metric characterization of 132, 173-cyclomesopheophorbide-a-enol
0.50
0.50
0.40
0.40
CYCLO (2a)
mesoCYCLO (2b)
0.30
0.30
0.20
0.20
0.10
0.10
(b)
(a)
0.00
0.00
350
450
550
λ, nm
650
750
350
450
550
λ, nm
650
750
Figure 3. UV/Vis absorption spectra of cyclopheophorbides in CH2Cl2. (a) mesoCYCLO (2b), (b) overlay of mesoCYCLO (2b) and CYCLO (2a), as indicated.
100
Sulfur
518.3068
Matrix
80
255.6016 + 32
60
40
Dimer
1,035.0051
20
0
80.0
478.4
850.5
1240.2
2001.0
1620.6
m/z
Figure 4. MALDI-MS spectrum of mesoCYCLO (2b) using a sulfur (S8)/CS2 matrix.

Mitra (Mortezaei-Rad) Khalesi, J. W. Louda / Tetrahedron Letters 52 (2011) 1078–1081
1081
References and notes
(aka mesoCYCLO), a requisite known for comparison to biogeo-
chemical isolates, were described. This compound and its data
add to the growing list of known and fully characterized chloro-
phyll derivatives. In the future, alteration of the pyropheophor-
bide-a precursors (e.g., methyl-desvinyl, H-desvinyl, and others)
used for the cyclization reaction should certainly give additional
novel compounds.
1. Goericke, R.; Strom, S. L.; Bell, M. A. Limnol. Oceanogr. 2000, 45, 200–211.
2. Baker, E. W.; Louda, J. W. In Treibs Memorial Volume; Prashnowsky, A. A., Ed.;
University of Wurzburg: Wurzburg, 2002; pp 3–128.
3. Karuso, P.; Berquist, P. R.; Buckelton, J. S.; Cambie, R. C.; Clark, G. R.; Rickard, C.
E. Tetrahedron Lett. 1986, 27, 2177–2178.
4. Sakata, K.; Yamamoto, K.; Ihshikawa, H.; Yagi, A.; Etoh, H.; Ina, K. Tetrahedron
Lett. 1990, 31, 1165–1168.
5. Watanabe, N.; Yamamoto, K.; Ihshikawa, H.; Yagi, A.; Sakata, K.; Brinen, L. S.;
Clardy, J. J. Nat. Prod. 1993, 56, 305–317.
Acknowledgments
6. Louda, J. W.; Neto, R. R.; Magalhaes, A. R. M.; Schneider, V. F. Comp. Biochem.
Physiol. 2008, B-150, 385–394.
7. Ocampo, R.; Sachs, J. P.; Repetam, D. J. Geochim. Cosmochim. Acta 1999, 63,
3743–3749.
8. Louda, J. W.; Loitz, J. W.; Rudnick, D. T.; Baker, E. W. Org. Geochem. 2000, 31,
1561–1580.
9. Chiller, X. F. D.; Gülaçar, F. O.; Buchs, A. Chemosphere 1993, 27, 2103–2110.
10. Chicarelli, M. I.; Maxwell, J. R. Tetrahedron Lett. 1986, 27, 4653–4654.
11. Falk, H.; Hoornaert, G.; Isenring, H.-P.; Eschenmoser, A. Helv. Chim. Acta 1975,
58, 2347–2357.
12. Isenring, H. –P.; Zass, E.; Smith, K.; Falk, H.; Luisier, J. L.; Eschenmoser, A. Helv.
Chim. Acta 1975, 58, 2357–2367.
This work was supported in parts by the Department of Chem-
istry and Biochemistry of Florida Atlantic University and by a con-
tract (to J.W.L.) from the South Florida Water Management District.
We thank Drs. Verid Marks, Frank Mari, and Vincent Storhaugh for
assistance with different facets of NMR. The assistance of Dr. Victor
Asirvathim with mass spectrometry is appreciated. Email ‘conver-
sations’ with Dr. Rubin Ocampo regarding certain facets of the
hemisynthesis are also noted and appreciated. The authors also
thank the editorial staff of Tetrahedron Letters for their profes-
sional attention to detail and help throughout this process.
13. Ocampo, R.; Sachs, J. P.; Repeta, D. J. Geochim. Cosmochim. Acta 1999, 63, 3743–
3749.
14. Yamamoto, K.; Sakata, K.; Watanabe, N.; Yagi, A.; Brinen, L. S.; Calardy, J.
Tetrahedron Lett. 1992, 33, 2587–2588.
15. Ma, L.; Dolphin, D. Tetrahedron: Asymmetry 1995, 6, 313–316.
16. Ma, L.; Dolphin, D. J. Org. Chem. 1996, 61, 2501–2510.
17. Harris, P. G.; Pearce, G. E. S.; Peakman, T. M.; Maxwell, J. R. Org. Geochem. 1995,
23, 183–187.
18. Pyropheophorbide-a methyl ester (pPBIDa ME, 1a2) was purchased (Cat. I.D.
MPPa) from Frontier Scientific in Logan Utah, USA. Alternately, it was produced
Appendix 1. 1H NMR assignments of 132,173-mesocyclopheo-
phorbide-a (mesoCYCLO, 2b) as compared to 132,173-cyclopheo-
phorbide-a (CYCLO, 2a: [data in brackets])
through reflux of pheophorbide-a ME obtained either as
commodity (Pha-592: Cat. I.D. PPa) from the same source or prepared from
chlorophyll-a by standard techniques. (Ref. 22).
a purchased
Proton No. of
Multiplicity (COSY)
(coupling constant)
Chemical shift
(ppm)
No.
protons
19. Palladium (10%) on charcoal catalyst (#205699) was purchased from the
Sigma–Aldrich Chemical Company, St. Louis, MO, USA.
21
3 [3]
2 [1]
3 [1]
s [s]
3.12 [3.22]
3.97 [7.83]
1.64 [6.1]
31
Q [dd(18.01, 11.650]
t (6.81) [d(11.51)]
20. Jeandon, C.; Ocampo, R.; Callot, H. J. Tetrahedron 1997, 53, 16107–16114.
21. Bis-trimethylsilyl-sodium amide in tetrahydrofuran (#245585) was purchased
from the Sigma–Aldrich Chemical Company, St. Louis, MO, USA.
22. Mortezaei-Rad, M.; Louda, J. W. J. Liq. Chromatogr. Related Technol. 2007, 30,
1361–1369.
32 [32a]
[32b]
5
— [1]
[d(17.78)]
[6.2]
1 [1]
3 [3]
2 [2]
3 [3]
1 [1]
3 [3]
1 [1]
2 [2]
s [s]
s [s]
q [q(7.89, 7.80)]
t [t(7.68)]
s [s]
8.86 [8.91]
3.16 [3.1]
3.52 [3.5]
1.62 [1.59]
8.97 [8.95]
3.35 [3.35]
3.7 [3.66]
1.75 [2.37,
2.41]
23. Baker, E. W.; Louda, J. W. In Biological Markers in the Sedimentary Record; Johns,
R. B., Ed.; Elsevier: Oxford, 1986; pp 125–225.
71
24. Matrix Assisted Laser Desorption Ionization-Time of Flight (MALDI-TOF) mass
spectra were obtained using Applied Biosystems voyager DE-STR instrument
fitted with a 337 nm nitrogen laser. For sample preparation, a sulfur (S8) matrix
was used following Brune.25 A 0.6 M solution of sulfur in carbon disulfide was
prepared. After dissolving a tiny amount of sample in carbon disulfide, a 1:1–
1:5 ratio mixture of sample to matrix was spotted on a 100 cell MALDI plate.
Each spectrum was obtained as an accumulation of 4 Â 50 laser shots.
25. Brune, D. Rapid Commun. Mass Spectrom. 1999, 13, 384–389.
26. Strupat, K.; Sagi, D.; Bonisch, H.; Schafer, G.; Peter-Katalinic, J. Analyst 2000,
563–567.
81
82
10
121
17
171
s [s]
2nd order [2nd order]
2nd order [2nd order]
172
2 [2]
t (6.74) [2nd order]
4.27 [2.93,
3.07]
27. 2D and 1D 1H NMR spectra in CD2Cl2 were obtained using a 400 MHz Varian
Mercury spectrometer.
O–H
18
1 [1]
1 [1]
3 [3]
1 [1]
2 Â 1
[2 Â 1]
s [s]
13.44 [13.43]
3.97 [4.02]
1.98 [1.99]
8.06 [7.98]
À0.75, 0.093
[0, À0.929]
2nd order [2nd order]
d (7.13) [d(6.99)]
s [s]
s [s]
181
20
N–H