Normal and abnormal heme biosynthesis. 6. Synthesis and metabolism of a series of monovinylporphyrinogens related to harderoporphyrinogen. further insights into the oxidative decarboxylation of porphyrinogen substrates by coproporphyrinogen oxidase

Article abstract of DOI:10.1021/jo100083t

A series of vinylporphyrinogens were prepared to probe the enzyme coproporphyrinogen oxidase (CPO). Six (2-chloroethyl)porphyrins were synthesized from a common dipyrrylmethane via a,c-biladiene intermediates in excellent yields. Subsequent dehydrohalogen

Full text of DOI:10.1021/jo100083t

pubs.acs.org/joc
Normal and Abnormal Heme Biosynthesis. 6. Synthesis and Metabolism
of a Series of Monovinylporphyrinogens Related
to Harderoporphyrinogen. Further Insights into the
Oxidative Decarboxylation of Porphyrinogen Substrates by
Coproporphyrinogen Oxidase†
Timothy D. Lash,* Ukti N. Mani, Anna-Sigrid I. M. Keck, and Marjorie A. Jones
Department of Chemistry, Illinois State University, Normal, Illinois 61790-4160
tdlash@ilstu.edu
Received January 18, 2010
A series of vinylporphyrinogens were prepared to probe the enzyme coproporphyrinogen oxidase
(CPO). Six (2-chloroethyl)porphyrins were synthesized from a common dipyrrylmethane via a,c-
biladiene intermediates in excellent yields. Subsequent dehydrohalogenation with DBU in refluxing
DMF then gave the required vinylporphyrin methyl esters, including harderoporphyrin-I, hard-
eroporphyrin-III, and isoharderoporphyrin. The corresponding porphyrinogen carboxylic acids
were incubated with chicken red cell hemolysates, which contain the enzyme CPO, and the products
analyzed. The 17-ethyl analogue of harderoporphyrinogen-III, but not its 13-ethyl isomer, was
shown to be an excellent substrate for CPO in accord with a proposed model for the active site of this
enzyme. In addition, harderoporphyrinogen-VII, the monovinyl intermediate in the metabolism of
coproporphyrinogen-IV, was shown to be an equally good substrate for this enzyme. However,
isoharderoporphyrinogen, which lacks the correct ordering of peripheral substituents, was also a
substrate for CPO. Furthermore, a nonnatural type I isomer of harderoporphyrinogen was shown to
be acted on by CPO, but in this case further metabolism was noted and this afforded an
unprecedented trivinyl porphyrinogen product. The corresponding porphyrin methyl ester was
isolated and characterized by FAB MS and proton NMR spectroscopy. The results from these
studies allow the binding requirements of CPO to be further assessed and provide a series of
substrates to investigate this poorly understood enzyme.
Introduction
chlorophylls, and corrins (e.g., vitamin B12),1-4 although the
pathway to the corrins and related tetrapyrroles such as siroheme
diverges at this point.5 In vertebrates, uroporphyrinogen-III
Uroporphyrinogen-III (uro’gen-III, Scheme 1) is the first
macrocyclic intermediate in the biosynthesis of the hemes,
(4) (a) Akhtar, M. The Modification of Acetate and Propionate Side Chains
during the Biosynthesis of Haem and Chlorophylls: Mechanistic and Stereochem-
ical Studies. In The Biosynthesis of Tetrapyrrolic Pigments; Ciba Foundation
Symposium 180; Wiley: Chichester, 1994; pp 130-155. (b) Akhtar, M. Copro-
porphyrinogen III and Protoporphyrinogen IX Oxidase. In The Porphyrin Hand-
book; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Elsevier: San Diego, 2003;
Vol. 12, pp 75-92. (c) Lewis, C. A., Jr.; Wolfenden, R. Proc. Natl. Acad. Sci. U.
S.A. 2008, 105, 17328–17333.
† Part 5: Cooper, C. L.; Stob, C. M.; Jones, M. A.; Lash, T. D. Bioorg. Med.
Chem. 2005, 13, 6244-6251.
(1) Biosynthesis of Tetrapyrroles; Jordan, P. M., Ed.; Elsevier: London,
1991.
(2) (a) Dailey, H. A. Biochem. Soc. Trans. 2002, 30, 590–599. (b) Ajioka,
R. S.; Phillips, J. D.; Kushner, J. P. Biochim. Biophys. Acta 2006, 1763, 723–
736.
(5) (a) Scott, A. I. Acc. Chem. Res. 1990, 23, 308–317. (b) Battersby, A. R.
(3) Jones, M. A.; Lash, T. D. Curr. Top. Biotechnol. 2004, 1, 15–27.
Acc. Chem. Res. 1993, 26, 15–21.
DOI: 10.1021/jo100083t
r
Published on Web 04/13/2010
J. Org. Chem. 2010, 75, 3183–3192 3183
2010 American Chemical Society

Lash et al.
JOCFeatured Article
SCHEME 1
SCHEME 2
undergoes a sequential decarboxylation process mediated by
the cytoplasmic enzyme uroporphyrinogen decarboxylase
(URO-D, E.C. 4.1.1.37) to afford the tetracarboxylate inter-
mediate coproporphyrinogen-III (copro’gen-III).6-9 Follow-
ing transfer to the mitochondria, oxidative decarboxylation of
two of the propionate side chains by coproporphyrinogen
oxidase (CPO, E.C. 1.3.3.3) gives protoporphyrinogen-IX
(proto’gen-IX).4b,10 Dehydrogenation mediated by protopor-
phyrinogen oxidase affords the aromatic tetrapyrrole proto-
porphyrin-IX, and insertion of iron(II) by ferrochelatase then
yields heme b.1-4 Protoporphyrin-IX is also the precursor to
many other hemes and chlorophylls.1,11 Most of the key steps in
the heme biosynthesis have been known for over 50 years,1 but
a detailed understanding of the individual enzyme-mediated
processes remains somewhat incomplete. URO-D mediates
the decarboxylation of four acetate side chains by a well-
understood mechanistic process,4a,c but the specificity of the
decarboxylation pathway7-9 and the potential interplay of the
two protein subunits12 still requires further study. CPO is a
more enigmatic enzyme that converts two propionic acid side
chains into vinyl groups by a poorly understood oxidative de-
carboxylation process.3,4,10 In aerobic organisms, CPO requires
molecular oxygen for activity, but it is not a metalloprotein and
does not utilize any cofactors.13 It had been demonstrated that
the propionate side chains are either initially hydroxylated so
that dehydration with concomitant decarboxylation can give
rise to the vinyl moieties4,14 or that a hydride acceptor was
involved.4,15 However, these speculations are not consistent
with the available data for aerobic CPO.13 A structurally
unrelated version of CPO is found in anaerobic organisms that
incorporates a 4Fe-4S cluster and appears to generate proto-
porphyrinogen-IX by a radical pathway,16 but this type of
mechanism is not plausible for the oxygen-dependent form of
CPO.17 One of us recently proposed a new mechanism that
involves the base-catalyzed generation of a peroxide anion
intermediate from molecular oxygen, followed by intramole-
cular deprotonation of the propionate unit and loss of H2O2
and CO2 (Scheme 2).17 The formation of hydrogen peroxide
and carbon dioxide as byproduct from the metabolism of
copro’gen-III has been demonstrated,18,19 and the new me-
chanism has received further support from independent DFT
calculations.20 In addition, this mechanism resembles the one
proposed for the conversion of urate to 5-hydroxyisourate by
urate oxidase using molecular oxygen, again without the
(6) (a) Jackson, A. H.; Sancovich, H. A.; Ferramola, A. M.; Evans, E.;
Games, D. E.; Matlin, S. A.; Elder, G. E.; Smith, S. G. Philos. Trans. R. Soc.
London B 1976, 273, 191–206. (b) Shoolingin-Jordan, P. M. The Biosynthesis
of Coproporphyrinogen III. In The Porphyrin Handbook; Kadish,
K. M., Smith, K. M., Guilard, R., Eds.; Elsevier: San Diego, 2003; Vol. 12 pp
33-74.
(13) Medlock, A. E.; Dailey, H. A. J. Biol. Chem. 1996, 271, 32507–32510.
(14) (a) Sano, S. J. Biol. Chem. 1966, 241, 5276–5283. (b) Yoshinaga, T.;
Sano, S. J. Biol. Chem. 1980, 255, 4727–4731.
(15) Seehra, J. S.; Jordan, P. M.; Akhtar, M. Biochem. J. 1983, 209, 709–
718.
(7) Lash, T. D. Biochem. J. 1991, 278, 901–903.
(8) Luo, J.; Lim, C. K. Biochem. J. 1993, 289, 529–532.
(9) Lash, T. D.; Mani, U. N.; Lyons, E. A.; Thientanavich, P.; Jones,
M. A. J. Org. Chem. 1999, 64, 478–487.
(10) Lash, T. D.; Mani, U. N.; Drinan, M. A.; Zhen, C.; Hall, T.; Jones,
M. A. J. Org. Chem. 1999, 64, 464–477.
(16) (a) Layer, G.; Grage, K.; Teschner, T.; Schunemann; Breckau, D.;
Masoumi, A.; Jahn, M.; Heathcote, P.; Trautwein, A. X.; Jahn, D. J. Biol.
Chem. 2005, 280, 29038–29046. (b) Layer, G.; Pierik, A. J.; Trost, M.; Rigby,
S. E.; Leech, H. K.; Grage, K.; Breckau, D.; Astner, I.; Jansch, L.; Heathcote,
P.; Warren, M. J.; Heinz, D. W.; Jahn, D. J. Biol. Chem. 2006, 281, 15727–
15734.
(17) Lash, T. D. Bioorg. Med. Chem. Lett. 2005, 15, 4506–4509.
(18) Breckau, D.; Mahlitz, E.; Sauerwald, A.; Layer, G.; Jahn, D. J. Biol.
Chem. 2003, 278, 46625–46631.
(11) Interestingly, an alternative pathway for the biosynthesis of C-type
hemes has been discovered in the primative anaerobic bacterium Desulfovi-
brio vulgaris. See: (a) Akutsu, H.; Park, J.-S.; Sano, S. J. Am. Chem. Soc.
1993, 115, 12185–12186. (b) Ishida, T.; Yu, L.; Akutsu, H.; Ozawa, K.;
Kawanishi, S.; Seto, A.; Inubushi, T.; Sano, S. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 4853–4858. (c) Lobo, S. A. L.; Brindley, A.; Warren, M. J.; Saralva,
L. M. Biochem. J. 2009, 420, 317–325.
(12) Phillips, J. D.; Warby, C. A.; Whitby, F. G.; Kushner, J. P.; Hill, C. P.
J. Mol. Biol. 2009, 389, 306–314.
(19) Elder, G. H.; Evans, J. O. Biochem. J. 1978, 169, 205–214.
(20) Silva, P. J.; Ramos, M. J. Bioorg. Med. Chem. 2008, 16, 2726–2733.
3184 J. Org. Chem. Vol. 75, No. 10, 2010

Lash et al.
JOCFeatured Article
SCHEME 3
FIGURE 1. Type isomers of coproporphyrinogen.
by HPLC24,28 and MS studies,29 but the regioisomer iso-
hardero’gen (Scheme 3) has not been detected. Never-
theless, isohardero’gen has been reported to be a substrate
for CPO.14,30 In order to assess the selectivity and binding
requirements for CPO, isomeric copro’gens were inves-
tigated. There are four “type isomers” of copro’gen
(Figure 1), including the natural “type III” isomer.31 The type
I and II isomers are not metabolized by CPO,32 but copro’gen-
IV is a good substrate for this enzyme.32-37 Copro’gen-IV
initially forms a type IV isomer of hardero’gen, and this
is further converted into proto’gen-XIII (Scheme 4).33-37
URO-D metabolizes all four uro’gen isomers, as well as all of
the possible hepta-, hexa-, and pentacarboxylate porphyrino-
gens for the type I and III series6-9,38 and for this reason is
considered to be a relatively promiscuous enzyme compared to
CPO.4 However, CPO has been shown to metabolize a series of
substrate analogues 1a-d (Scheme 5) where the 13 and 17-
propionate residues have been replaced by methyl, ethyl,
propyl, or butyl groupings.10,39-41 An early report suggested
that the diethylporphyrinogen 1a (meso’gen-VI) was converted
to a divinylic product,39 but extensive studies using 1a-d with
crude avian CPO preparations or purified human recombinant
CPO have afforded only the monovinyl products 2a-d.10,42
involvement of transition metal ions or cofactors.21 A better
understanding of CPO is needed, not only due to its central role
in heme and chlorophyll biosynthesis but also because defects
in this enzyme can lead to a disease state known as hereditary
coproporphyria.22 This disease may result in the overproduc-
tion and accumulation of porphyrins and can lead to skin
photosensitivity and neurological disorders.22
It is known that CPO carries out the two oxidative
decarboxylations sequentially to give a monovinylporphyr-
inogen intermediate (Scheme 3).1-4 This process is regiospe-
cific, generating harderoporphyrinogen (hardero’gen) as an
intermediate by first converting the A ring side chain to a
vinyl unit.1-4 The corresponding porphyrin was first isolated
and characterized from the harderian glands of rats23 and the
structure was confirmed by total synthesis.24 Harderopor-
phyrin and metabolites have also been observed in extracts
from urine and feces,25-27 and although isomeric porphyrins
have been detected, these appear to arise from isomerization
of the porphyrinogens prior to oxidation.27 In enzyme
incubation studies using copro’gen-III, the formation of
hardero’gen (or the related porphyrin) has be demonstrated
(30) Games, D. E.; Jackson, A. H.; Jackson, J. R.; Belcher, R. V.; Smith,
S. G. J. Chem. Soc., Chem. Commun. 1976, 187–188.
(31) Smith, K. M. In Porphyrins and Metalloporphyrins; Elsevier: Am-
sterdam, 1975; pp 3-28.
(32) Porra, P. J.; Falk, J. E. Biochem. J. 1964, 90, 69–75.
(33) (a) Buldain, G.; Hurst, J.; Frydman, R. B.; Frydman, B. J. Org.
Chem. 1977, 42, 2953–2956. (b) Buldain, G.; Diaz, L.; Frydman, B. J. Org.
Chem. 1977, 42, 2957–2960.
(34) Al-Hazimi, H. M. G.; Jackson, A. H.; Ryder, D. J.; Elder, G. H.;
Smith, S. G. J. Chem. Soc., Chem. Commun. 1976, 188–189.
(35) Al-Hazimi, H. M. G.; Jackson, A. H.; Knight, D. W.; Lash, T. D.
J. Chem. Soc., Perkin Trans. 1 1987, 265–276.
(36) Battersby, A. R.; Hamilton, A. D.; McDonald, E.; Mombelli, L.;
Wong, O.-H. J. Chem. Soc., Perkin Trans. 1 1980, 1283–1289.
(37) Cooper, C. L.; Lash, T. D.; Jones, M. A. Med. Sci. Monit. 2005, 11,
BR420–425.
(38) Jackson, A. H.; Supphayen, D. M. J. Chem. Soc., Perkin Trans. 1
1987, 277–286.
(39) Jackson, A. H.; Jones, D. M.; Smith, S. G.; Elder, G. H. J. Chem.
Soc., Chem. Commun. 1976, 322–323.
(40) Lash, T. D.; Drinan, M. A.; Zhen, C.; Mani, U. N.; Jones, M. A.
Bioorg. Med. Chem. Lett. 1994, 4, 1607–1612.
(41) Lash, T. D.; Hall, T.; Mani, U. N.; Jones, M. A. J. Org. Chem. 2001,
66, 3753–3759.
(42) Jones, M. A.; He, J.; Lash, T. D. J. Biochem. 2002, 131, 201–205.
(21) (a) Kahn, K.; Serfozo, P.; Tipton, P. A. J. Am. Chem. Soc. 1997, 119,
5435–5442. (b) Kahn, K.; Tipton, P. A. Biochemistry 1998, 37, 11651–11659.
(22) (a) Moore, M. R. Int. J. Biochem. 1993, 25, 1353–1368. (b)
McDonagh, A. F.; Bissell, D. M. Semin. Liver Dis. 1998, 18, 3. (c) Martesek,
P. Semin. Liver Dis. 1998, 18, 25–32. (d) McCarroll, N. A. Anal. Chem. 1995,
67, 425–428.
(23) Kennedy, G. Y.; Jackson, A. H.; Kenner, G. W.; Suckling, C. J.
FEBS Lett. 1970, 6, 9–12.
(24) (a) Cavaleiro, J. A. S.; Kenner, G. W.; Smith, K. M. J. Chem. Soc.,
Perkin Trans. 1 1974, 1188–1194. (b) Kenner, G. W.; Quirke, J. M. E.; Smith,
K. M. Tetrahedron 1976, 32, 2753–2756.
(25) (a) Smith, S. G.; Belcher, R. V.; Mahler, R. F.; Yudkin, J. Clin. Chim.
Acta 1969, 23, 241–243. (b) Belcher, R. V.; Smith, S. G.; Mahler, R. F.
Enzyme 1974, 17, 53–60.
(26) Jackson, A. H.; Rao, K. R. N.; Smith, S. G.; Lash, T. D. Biochem. J.
1985, 227, 327–331.
(27) Danton, M.; Lim, C. K. Rapid Commun. Mass Spectrom. 2004, 18,
2309–2316.
(28) Evans, N.; Jackson, A. H.; Matlin, S. A.; Towill, R. J. Chromatogr.
1976, 125, 345–355.
(29) Jackson, A. H.; Jones, D. M.; Philip, G.; Lash, T. D.; Batlle, A. M.;
del, C.; Smith, S. G. Int. J. Biochem. 1980, 12, 681–688.
J. Org. Chem. Vol. 75, No. 10, 2010 3185

Lash et al.
JOCFeatured Article
SCHEME 4
SCHEME 5
FIGURE 2. Active-site models for substrate binding by copropor-
phyrinogen oxidase. The top version shows a 2D representation
where Y is the catalytic site, X recognizes and binds a second
propionate residue, and Z can accommodate only small nonpolar
groups like Me, Et, V, or H. The lower version envisages that the
porphyrinogen binds as a bowl-shaped conformation and illustrates
the proposed reaction with molecular oxygen.
as H, Me, vinyl (V), or Et but not propionate or acetate
groupings. In order for a porphyrinogen to fit these binding
requirements, it must possess the sequence of peripheral sub-
stituents R Me-P Me-P, where R is one of the small nonpolar
groups listed above and P = CH2CH2CO2H.10
In 2004, a crystal structure for the oxygen-dependent form
of CPO from Saccharomyces cerevisiae was published,47 and
a similar structure was subsequentially reported for human
recombinant CPO.48 These were obtained as homodimers49
that generate a nonpolar cleft that has been tentatively
assigned as the active site.47 Site-directed mutagenesis studies
were used to demonstrate that the invariant amino acids
aspartate 400, arginine 262, and arginine 401 were essential
for significant catalytic activity,50 and it was speculated that
In addition, when only the 13-propionate group is replaced by
an ethyl (13-Et porphyrinogen 1e), avian CPO still only
processes the substrate once to give 2e.41,42 However, while
the isomeric 17-Et porphyrinogen 1f is an equally good sub-
strate for CPO, it is further converted into the divinylporphyr-
inogen product 3.41,42 These and related results32-37,43-46 were
used to develop a model for the substrate binding sites in CPO
(Figure 2).10,17 In this model, three regions were designated.
Region Y corresponds to the catalytic site where the propionate
group is converted into the vinyl moiety, region X designates a
site that requires the presence of a second propionate side chain
for substrate binding, and position Z represents a region that
can accommodate the presence of small nonpolar groups such
(47) Phillips, J. D.; Whitby, F. G.; Warby, C. A.; Labbe, P.; Yang, P.;
Pflugrath, J. W.; Ferrara, J. D.; Robinson, H. A.; Kushner, J. P.; Hill, C. P.
J. Biol. Chem. 2004, 279, 38960–38968.
(48) Lee, D. S.; Flacksova, E.; Bodnarova, M.; Demeler, B.; Martasek, P.;
Raman, C. S. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 14232–14237.
(49) Martasek, P.; Camadro, J.-M.; Raman, C. S.; Lecompte, M. C.;
Le Caer, J. P.; Demeler, B.; Grandchamp, B.; Labbe, P. Cell Mol. Biol. 1997,
43, 47–58.
(43) Jackson, A. H.; Lash, T. D.; Ryder, D. J.; Smith, S. G. Int. J.
Biochem. 1980, 12, 775–780.
(44) Jackson, A. H.; Lash, T. D.; Ryder, D. J. J. Chem. Soc., Perkin Trans.
1 1987, 287–298.
(45) Cooper, C. L.; Stob, C. M.; Jones, M. A.; Lash, T. D. Bioorg. Med.
Chem. 2005, 13, 6244–6251.
(46) Lash, T. D.; Kaprak, T. A.; Shen, L.; Jones, M. A. Bioorg. Med.
Chem. Lett. 2002, 12, 181–186.
(50) Stephenson, J. R.; Stacey, J. A.; Morgenthaler, J. B.; Friesen, J. A.;
Lash, T. D.; Jones, M. A. Protein Sci. 2007, 16, 401–410.
3186 J. Org. Chem. Vol. 75, No. 10, 2010

Lash et al.
JOCFeatured Article
these residues could be involved in substrate binding and
deprotonation of a pyrrolic NH to initiate reaction
with molecular oxygen.50,51 Porphyrinogens appear to bind
with URO-D in a bowl-shaped conformation,52 and it has
been speculated that this is also the case for CPO.17,47,50,53
This hypothesis has been used to modify the proposed model
for substrate binding (Figure 2)17 and parallels observations
for anion binding by synthetic calix[4]pyrroles.54 Despite all
of these recent advances, our understanding of the precise
mode for substrate binding and oxidative decarboxylation
by CPO remains incomplete. In order to further develop
these studies, a series of harderoporphyrinogens were re-
quired in our investigations. Relatively few studies have been
carried out using vinylporphyrinogens, due in part to their
decreased stability and the more stringent requirements for
synthesis.24,33-36 In this paper, we report new syntheses of
the related harderoporphyrins and their activity as sub-
strates for avian CPO.55
Results and Discussion
Hardero’gen has one pyrrole unit substituted with Me-V
and three with Me-P (Figure 3). These units can be arranged
to give a total of 8 type isomers which are numbered in
accord with the principles originally developed by Hans
Fischer.31 Hardero’gens I and II are structurally related to
copro’gens I and II, respectively, although neither one of
these porphyrinogens can be formed by the action of CPO.32
There are four type III hardero’gens (III-VI) and two type
IV hardero’gens (VII and VIII). By this nomenclature, the
natural isomer is hardero’gen-III, isohardero’gen is hard-
ero’gen-IV, and the intermediate derived from copro’gen-IV
is hardero’gen-VII. These designations will be used through-
out the discussion, although the name isohardero’gen will be
used instead of hardero’gen-IV to avoid confusion with
copro’gen-IV. Isomers III and VII are known to be sub-
strates for CPO,32-36 and isohardero’gen is also reported to
be a significant, albeit poorer, substrate.30 Although the
V- and VI-type isomers have not been synthesized, the related
dihydrohardero’gens 1e and 1f (Scheme 5) are known sub-
strates, and this strongly implies that hardero’gens V and VI
would also be metabolized by CPO. Hardero’gens I and VIII
also have the correct sequence of substituents to be substrates
for CPO, and only isomer II is likely to be a nonsubstrate for
this enzyme.4 Therefore, it is a little surprising to find that 7
of the 8 type isomers of hardero’gen are likely to be sub-
strates for this supposedly selective enzyme. To further our
investigations, samples of hardero’gen-III and isohardero’gen
were required. In addition, hardero’gen-I was selected as
an intriguing potential substrate for CPO. Copro’gen-I is
not a substrate for CPO, but if it were converted hardero’-
gen-I would be formed. Ironically, our model suggests that
hardero’gen-I should be a good substrate for CPO and may
FIGURE 3. Eight type isomers of harderoporphyrinogen.
SCHEME 6
be converted into a trivinyl product (see below). The syn-
thesis of hardero’gen-VII was also conducted to further
investigate metabolism in the type IV series, and the mono-
vinylic porphyrinogens 2e and 2f that are formed from the
13- or 17-ethyl porphyrinogens 1e and 1f were also needed
for study.
The substrates for CPO are hexahydroporphyrins or
porphyrinogens (Scheme 6), but these types of reduced
tetrapyrroles are highly unstable and prone to oxidations
and acid-catalyzed rearrangements.56 For this reason, the
corresponding porphyrins were targeted for synthesis. Por-
phyrins can easily be reduced to porphyrinogens (e.g., with
sodium amalgam10,57), and this conversion could be carried
out immediately prior to the biochemical studies. The por-
phyrins were also conveniently prepared as methyl esters and
only converted into the corresponding carboxylic acids im-
mediately prior to reduction.10 Harderoporphyrin-III, iso-
harderoporphyrin, and harderoporphyrin-VII have been
synthesized previously,24,33a,35,36 but harderoporphyrin-I
(51) For related studies, see: Gitter, S. J.; Cooper, C. L.; Friesen, J. A.;
Lash, T. D.; Jones, M. A. Med. Sci. Monit. 2007, 13, BR1–10.
(52) Phillips, J. D.; Whitby, F. G.; Kushner, J. P.; Hill, C. P. EMBO J.
2003, 22, 6225–6233.
(53) He, J.; Kaprak, T. A.; Jones, M. A.; Lash, T. D. J. Porphyrins
Phthalocyanines 2005, 9, 170–185.
(54) (a) Gale, P. A.; Sessler, J. L.; Kral, V. Chem. Commun. 1998, 1–8. (b)
Blas, J. R.; Lopez-Bez, J. M.; Marquez, M.; Sessler, J. L.; Luque, F. J.;
Orozco, M. Chem.;Eur. J. 2007, 13, 1108–1116.
(56) (a) Mauzerall, D. J. Am. Chem. Soc. 1960, 82, 2601–2605. (b)
Battersby, A. R.; Leeper, F. J. Chem. Rev. 1990, 90, 1261–1274.
(57) Hydrogenation over palladium catalysts can also be used to generate
porphyrinogens, thereby avoiding the use of mercury, but this method is not
compatible with the presence of vinyl side chains. See: Bergonia, H. A.;
Phillips, J. D.; Kushner, J. P. Anal. Biochem. 2009, 384, 74–78.
(55) Preliminary communication: Lash, T. D.; Keck, A.-S.; I., M.; Mani,
U. N.; Jones, M. A. Bioorg. Med. Chem. Lett. 2002, 12, 1079–1082.
J. Org. Chem. Vol. 75, No. 10, 2010 3187

Lash et al.
JOCFeatured Article
SCHEME 7
SCHEME 8
SCHEME 9
give the corresponding carboxylic acid 5 (Scheme 7). Treat-
ment with TFA cleaves the tert-butyl ester and decarbox-
ylates both of the terminal units, and subsequent reaction
with 2 equiv of pyrrole aldehydes 6a or 6b gave the corre-
sponding a,c-biladienes 7a and 7b, respectively. These open-
chain tetrapyrrolic salts were precipitated with ether to give
red powders in 62-77% yield. Cyclization with copper(II)
chloride in DMF at room temperature,58 followed by deme-
talation with 15% sulfuric acid-TFA and reesterification
with 5% H2SO4-methanol, gave the chloroethyl precursors
and the 13- and 17-ethyl analogues of harderoporphyrin-III
were not previously known. In order to simplify these
investigations, all six targeted porphyrins were prepared
from a single dipyrrylmethane 4 via a,c-biladiene intermedi-
ates (Schemes 7-9). The vinyl unit must be installed after the
porphyrin macrocycle has been generated due to the highly
reactive nature of vinylic pyrroles, and this was accom-
plished by using a chloroethyl side chain as a protected vinyl
moiety.10 Dipyrrylmethane 4, which was prepared by a
literature procedure,10 has three types of ester units. The
benzyl ester is easily hydrogenolyzed in quantitative yield to
(58) Smith, K. M.; Minnetian, O. M. J. Chem. Soc., Perkin Trans. 1 1986,
277–280.
3188 J. Org. Chem. Vol. 75, No. 10, 2010

Lash et al.
JOCFeatured Article
SCHEME 10
SCHEME 11
FIGURE 4. Time-course experiments for incubations of hardero-
porphyrinogen-III (upper trace) and isoharderoporphyrinogen
(lower trace) with chicken red cell hemolysates at 37 °C showing
the percent conversion of the substrate to protoporphyrin-IX.
These values were corrected to take into account the presence of
endogenous protoporphyrin-IX that is present in these crude en-
zyme preparations by the method described in ref 42.
cleave the tert-butyl ester and reacted with pyrrole aldehyde
6a and HBr to give the tripyrrene benzyl ester 10 in 75%
yield. Further treatment with HBr in TFA for 6 h cleaved the
benzyl ester, and subsequent condensation with pyrrole
aldehyde 6c gave a,c-biladiene 7e in 66% yield. Similarly,
reaction with pyrrole aldehyde 6b gave the a,c-biladiene
precursor 7f to harderoporphyrin-I in 65% yield. Copper(II)
chloride mediated cyclization, demetalation, and reesterifi-
cation gave porphyrins 8e and 8f in 66% and 70% yields,
respectively. The vinyl groups were introduced by base
catalyzed dehydrohalogenation of the chloroethyl side
chains (Scheme 10). In earlier work,10 we had carried out
this transformation with KOH-pyridine based on a literature
procedure,61 but these conditions gave incomplete conver-
sions for the current series. This difficulty was easily over-
come by heating the chloroethylporphyrins 8 with DBU
in DMF for 1 h. Following chromatography and recrystalli-
zation, the vinyl porphyrins 11 were isolated in 75-80%
yield.
to harderoporphyrin-III and isoharderoporphyrin, 8a and
8b, respectively, in 54% and 38% yields.
The remaining porphyrins had to be prepared via stepwise
routes.59,60 Dipyrrylmethane carboxylic acid 5 was reacted
with 1 equiv of pyrrole aldehyde 6a in the presence of
p-toluenesulfonic acid, followed by brief treatment with
anhydrous HBr, to give the tripyrrene tert-butyl ester 9 in
52% yield (Scheme 8). Subsequent treatment with TFA and
reaction of formylpyrroles 6b and 6c gave the related a,c-
biladienes 8c and 8d in 74-81% yield. Cyclization with
CuCl2 in DMF, followed by demetalation and reesterifica-
tion, gave the chloroethylporphyrins 8c and 8d in 59% and
58% yields, respectively. The remaining two porphyrins were
prepared by an alternative tripyrrene strategy (Scheme 9).60
The mixed ester dipyrrylmethane 4 was treated with TFA to
In order to completely assess these substrates, samples of
coproporphyrin-IV (16) were also required. This was pre-
pared in two steps from dipyrrylmethane dicarboxylic acid
1762 (Scheme 11). Reaction of 17 with pyrrole aldehyde 6b in
(59) Baptista de Almeida, J. A. P.; Kenner, G. W.; Rimmer, R.; Smith, K.
M. Tetrahedron 1976, 32, 1793–1799.
(60) Smith, K. M.; Craig, G. W. J. Org. Chem. 1983, 48, 4302–4306.
(61) Clezy, P. S.; Fookes, C. J. R. Aust. J. Chem. 1977, 30, 217–220.
(62) Lash, T. D.; Armiger, Y. L. S.-T. J. Heterocycl. Chem. 1991, 28, 965–
970.
J. Org. Chem. Vol. 75, No. 10, 2010 3189

Lash et al.
JOCFeatured Article
SCHEME 12
FIGURE 5. Time-course experiments for incubations of hardero-
porphyrinogen-I with chicken red cell hemolysates showing the
formation of protoporphyrinogen-I (blue line) and trivinylporphyr-
inogen 14a (red line) over a 2 h incubation period at 37 °C.
SCHEME 13
ducted in triplicate and show reasonably reproducible re-
sults. As expected, hardero’gen-III was an excellent substrate
(Figure 4) and was rapidly converted to protoporphyrin-IX
(Scheme 12). Isoharderporphyrin is a much poorer substrate,
but our model for enzyme activity suggests that this por-
phyrinogen should not be a substrate at all as it does not
possess the usual sequence of substituents R Me-P Me-P.
However, careful analysis of the data showed that isohar-
dero’gen is indeed a substrate for CPO as has been noted
previously in the literature.30 The presence of endogenous
protoporphyrin-IX does make the results a little ambiguous,
but preliminary results using human recombinant CPO,37
which does not have this contaminant, also show the forma-
tion of protoporphyrin-IX, and there can be no doubt that
isohardero’gen is a substrate for this enzyme. Hardero’gen-
VII is a good substrate for CPO and is metabolized at a
comparable rate to hardero’gen-III, but the 13-ethyl analo-
gue of hardero’gen-III 2e showed no conversion to the
divinyl product (Scheme 12), confirming earlier results for
tripropionate porphyrinogens.41,42 However, the 17-ethyl
analogue was an excellent substrate, again as had been
expected. In earlier work, tripropionate porphyrinogen 1e
had been incubated with CRH to give the 13-ethylporphyr-
inogen 2e (Scheme 5), and this was further characterized by
NMR spectroscopy for the related porphyrin dimethyl ester.
Synthetic 13-ethylharderoporphyrin 11d gave a virtually
identical 300 MHz proton NMR spectrum to the product
isolated from those incubation studies41 and was easily
distinguishable from the proton NMR spectrum for the
isomeric 17-ethylharderoporphyrin 11e (see the Supporting
Information). Hence, the synthetic samples further confirm
the structure of this incubation product.
the presence of TFA and HBr gave the a,c-biladiene 18, and
this was cyclized with CuCl2 in DMF, demetalated, and
reesterified to give coproporphyrin-IV tetramethyl ester in
51% yield.63
The vinylporphyrins were hydrolyzed with 25% hydro-
chloric acid and reduced to the corresponding porphyrino-
gen carboxylic acids with 3% sodium amalgam as described
previously.10 These were then incubated at 37 °C with
chicken red cell hemolysates (CRH), which act as an active
source of CPO.10 Following the incubation, the products
were extracted as the corresponding porphyrins and ester-
ified prior to analysis. Product formation was assessed by
TLC and normal-phase HPLC. Although CRH is an ex-
cellent source of CPO, it is contaminated by small amounts
of protoporphyrin-IX and the data must be corrected for this
contaminant. In addition, the porphyrinogens are air sensi-
tive, but oxygen must be present for enzyme activity. These
factors lead to some variations that can produce large
error bars. However, the experiments were usually con-
As expected, hardero’gen-I (12a) also proved to be a good
substrate for CPO (Scheme 13, Figure 5). Following incuba-
tions with CRH, extraction, and esterification, the products
were analyzed by HPLC (Figure 6). These traces showed the
formation of two products with retention times that were
consistent with di- (peak B) and monoesters (peak C),
respectively. Although 12a is a non-natural type I porphyr-
inogen, it has the correct sequence of peripheral substituents
for binding to CPO. Oxidative decarboxylation would then
afford protoporphyrinogen-I (13a), but this species also has
(63) The biochemical studies conducted on copro’gen-IV are reported in
ref 37.
3190 J. Org. Chem. Vol. 75, No. 10, 2010

Lash et al.
JOCFeatured Article
12b. This dihydroharderoporphyrin was also a good substrate
and gave a monopropionate product 14b (Scheme 13). The
metabolite was characterized as the porphyrin methyl ester 15b
by HR FAB MS and proton NMR spectroscopy (see the
Supporting Information).
Conclusions
A series of monovinylic porphyrins have been synthesized,
specifically harderoporphyrin-III, three isomeric com-
pounds, and two ethyl-substituted analogues. These por-
phyrins were converted into the corresponding porphy-
rinogen carboxylic acids and used to probe the activity of
coproporphyrinogen oxidase (CPO). The results from these
studies confirm that a specific sequence of peripheral sub-
stituents is generally required for substrate binding and
metabolism, although isoharderoporphyrinogen lacks this
sequence and can still act as a poor substrate for this enzyme.
Nevertheless, the 17-ethyl analogue of harderoporphyrino-
gen-III has the required sequence and is a good substrate,
while its 13-ethyl isomer, which lacks the correct sequence, is
not metabolized. Harderoporphyrinogen-I is also a good
substrate for CPO, but in this case the first formed product is
also a substrate, and this allows further conversion to an
unusual trivinylporphyrinogen product. These results help
to clarify the binding requirements for CPO and will allow
further investigations into the activity of this important
enzyme in the heme biosynthetic pathway.
FIGURE 6. HPLC traces showing the porphyrin methyl ester
products for a 20 min incubation of harderoporphyrinogen-I with
CRH and a standard chromatogram for protoporphyrin-IX. The
HPLC analysis was performed on a normal-phase 5 μm Partisil
colum (250 ꢀ 4.6 mm), using a 20 μL injection loop, with a mobile
phase of 35/65 v/v ethyl acetate-cyclohexane. Peak A = hard-
eroporphyrin-I; peak B = protoporphyrin-I; peak C = trivinylpor-
phyrin 15a.
Experimental Section
Benzyl 3-(2-Chloroethyl)-8,13-bis(2-methoxycarbonylethyl)-
2,7,12,14-tetramethyl-5,16-dihydrotripyrrin-1-carboxylate Hy-
drobromide (10). Dipyrrylmethane 410 (1.00 g, 1.80 mmol) was
treated with trifluoroacetic acid (6.3 mL) with stirring under a
nitrogen atmosphere at ambient temperature for 5 min. A
solution of pyrrole aldehyde 6a9 (0.377 g, 1.80 mmol) in
methanol (40 mL) was added all at once, and the brownish
orange solution was stirred further for 1.5 h. Commercially
available 30% HBr-acetic acid (8 drops) was added, followed
by dropwise addition of ether (80 mL), and the mixture was
stirred for a further 15 min. The orange precipitate was filtered,
washed thoroughly with ether, and dried in vacuo overnight to
give the title tripyrrene hydrobromide (0.962 g, 1.32 mmol,
75%) as orange crystals: mp 177.5-178.5 °C; UV-vis
(CHCl3) λmax (log10 ε) 493 nm (4.90); 1H NMR (CDCl3) δ
2.07 (3H, s), 2.26 (3H, s), 2.35 (3H, s), 2.46-2.54 (4H, 2 over-
lapping triplets), 2.66 (3H, br s), 2.76 (2H, t, J = 7.6 Hz), 2.91
(2H, t, J = 7.8 Hz), 2.96 (2H, t, J = 7.6 Hz), 3.37 (2H, t, J = 7.8
Hz), 3.61 (3H, s), 3.67 (3H, s), 4.33 (2H, br s), 5.31 (2H, s),
7.25-7.3 (3H, m), 7.31 (1H, s), 7.49 (2H, d, J = 7.2 Hz), 10.71
(1H, br s), 13.11 (1H, br s), 13.15 (1H, br s); 13C NMR (CDCl3) δ
9.3, 10.4, 10.7, 13.3, 19.5, 20.3, 23.3, 28.5, 33.9, 34.7, 44.1, 52.0,
52.1, 65.6, 118.4, 119.1, 121.1, 123.1, 125.7, 127.3, 128.0, 128.4,
128.6, 137.0, 144.4, 144.8, 150.6, 157.4, 161.1, 172.7. Anal. Calcd
for C36H43N3O6ClBr.H2O: C, 57.87; H, 6.07; N, 5.62. Found: C,
57.98; H, 5.76; N, 5.47.
FIGURE 7. 300 MHz proton NMR spectrum of the trivinylpor-
phyrin methyl ester derived from incubations of harderoporphyr-
inogen-I with chicken red cell hemolysates.
the correct sequence of substituents (V Me-P Me-P) and is
apparently further processed to give the trivinylporphyrino-
gen 14a. This observation provides striking evidence in sup-
port of our substrate binding model. Protoporphyrinogen-I
does not accumulate in these incubation studies and probably
remains associated with CPO so that it is converted directly into
14a. A preparative study was conducted where 12a was in-
cubated with CRH and the trivinylporphyrin product 15a
derived from porphyrinogen 13a was isolated and character-
ized by FAB MS and proton NMR spectroscopy. High-
resolution FAB MS gave an [M þ H] peak at m/z 531.2758
which corresponds to the expected formula C34H34N4O2 þ H.
The proton NMR spectrum (Figure 7) was also in full agree-
ment with the proposed trivinylporphyrin structure 15a, show-
ing three singlets for the meso-protons at 10.28 (2H), 10.21
(1H), and 10.12 ppm (1H), and three 3H multiplets for the
vinyl units at 8.24-8.38, 6.35-6.45, and 6.17-6.25 ppm.
The chloroethyl precursor 8f was also hydrolyzed and reduced
with 3% sodium amalgam to give the 3-ethylporphyrinogen
8-(2-Chloroethyl)-3,13,18-tris(2-methoxycarbonylethyl)-1,2,7,-
12,17,19-hexamethyl-10,23-dihydrobilin Dihydrobromide (7f).
Tripyrrene 10 (124 mg, 0.17 mmol) was stirred with a mixture
of 30% HBr-acetic acid (0.5 mL) and trifluoroacetic acid (2.5
mL) at room temperature for 6 h. A solution of pyrrole aldehyde
6b9 (36 mg, 0.17 mmol) was added all at once to the brownish
orange solution, which immediately turned reddish orange, and
stirred for 30 min. Ether (30 mL) was added rapidly but
dropwise, and stirring was continued for 15 min. The resulting
J. Org. Chem. Vol. 75, No. 10, 2010 3191

Lash et al.
JOCFeatured Article
precipitate was suction filtered, washed thoroughly with ether,
and dried in vacuo overnight to give the a,c-biladiene (94 mg,
0.11 mmol, 65%) as a red powder: mp 176.5-177 °C; UV-vis
(CHCl3) λmax (log10 ε) 455 (4.42), 524 nm (5.32); 1H NMR
(CDCl3) δ 1.99 (3H, s), 2.03 (3H, s), 2.33 (3H, s), 2.37 (3H, s),
2.49 (4H, t, J = 7.3 Hz), 2.56 (2H, t, J = 7.2 Hz), 2.69 (3H, s),
2.72 (3H, s), 2.77 (2H, t, J = 7.5 Hz), 2.92-3.06 (8H, m), 3.60
(3H, s), 3.64 (3H, s), 3.68 (3H, s), 5.21 (2H, s), 7.30 (1H, s), 7.34
(1H, s), 13.31 (1H, br s), 13.35 (1H, br s), 13.46 (1H, br s),
13.48 (1H, br s); 1H NMR (CDCl3) δ 9.1, 9.5, 10.4, 10.7, 13.3,
13.4, 19.6, 20.1, 20.3, 25.9, 27.5, 33.9, 34.7, 34.8, 43.7, 52.0, 52.1,
52.2, 121.1, 121.3, 125.1, 121.56, 121.63, 126.1, 126.7, 127.9, 128.4,
142.1, 144.2, 144.9, 145.6, 148.5, 149.4, 157.3, 157.7, 172.6, 172.7,
pressure, and the purple residue was purified by chromatogra-
phy on a grade 3 alumina column eluting with dichloromethane.
Recrystallization from chloroform-methanol gave hardero-
porphyrin-I trimethyl ester (29 mg, 0.045 mmol, 77%) as small
purple needles: mp 178.5-180 °C; UV-vis (1% Et3N-CHCl3)
λ
max (log10 ε) 403 (5.20), 502 (4.17), 538 (4.10), 572 (3.94), 626
nm (3.80); UV-vis (1% TFA-CHCl3) λmax (log10 ε) 408 (5.55),
551 (4.29), 593 nm (4.02); 1H NMR (CDCl3) δ -3.68 (2H, br s),
3.24-3.32 (6H, m), 3.64 (3H, s), 3.66 (3H, s), 3.68 (6H, s), 3.69
(3H, s), 3.70 (3H, s), 3.74 (3H, s), 4.37-4.48 (6H, m), 6.18 (1H,
dd, J = 1.8, 11.1 Hz), 6.37 (1H, dd, J = 1.8, 17.4 Hz), 8.30 (1H,
dd, J = 11.1, 17.4 Hz), 10.07 (2H, s), 10.14 (1H, s), 10.23 (1H, s);
1H NMR (TFA-CDCl3) δ -3.47 (3H, br s), -3.35 (1H, br s),
3.12-3.18 (6H, m), 3.647 (3H, s), 3.653 (3H, s), 3.66 (3H, s), 3.68
(3H, s), 3.69 (6H, s), 3.75 (3H, s), 4.42-4.49 (6H, m), 6.30 (1H, d,
J = 17.6 Hz), 6.49 (1H, d, J = 11.6 Hz), 8.15 (1H, dd, J = 11.6,
17.6 Hz), 10.69 (1H, s), 10.78 (2H, s), 10.84 (1H, s); 13C NMR
(TFA-CDCl3) δ 12.0, 12.1, 12.7, 21.9, 35.6, 52.6, 99.1, 99.2,
99.8, 99.9, 127.9, 138.4, 138.5, 139.1, 139.2, 139.3, 140.1, 140.3,
141.0, 141.7, 141.8, 142.1, 142.5, 142.6, 174.4, 174.5; EIMS
(70 eV) m/z (relative intensity) 652 (1.6), 651 (4.7), 650
(10.2, Mþ), 578 (1.6), 577 (21, [M - CH2CO2CH3]þ); HRMS
(EI) m/z calcd for C38H421N4O6 650.3104, found 650.3102. Anal.
172.8. Anal. Calcd for C39H51Br2ClN4O6 1/2H2O: C, 53.47;
3
H, 5.98; N, 6.39. Found: C, 53.31; H, 5.76; N, 6.26.
3-(2-Chloroethyl)-8,13,18-tris(2-methoxycarbonylethyl)-2,7,12,-
17-tetramethylporphyrin (8f). a,c-Biladiene dihydrobromide 7f
(460 mg, 0.53 mmol) was added to a stirred solution of copper(II)
chloride (1.31 g) in DMF (195 mL), and the resulting mixture was
stirred in dark for 2 h. The mixture was diluted with dichloro-
methane (250 mL) and washed with water (3 ꢀ 250 mL). The
aqueous layers were back-extracted with dichloromethane, and
the combined organic layers were dried over sodium sulfate and
filtered. The solvent was evaporated on a rotary evaporator under
aspirator pressure, and then a vacuum pump was used to remove
any remaining DMF. The solid residue was taken up in 15% v/v
sulfuric acid-TFA (50 mL) and stirred in the dark at room
temperature for 45 min. The reaction mixture was diluted with
dichloromethane (250 mL) and washed with water (2 ꢀ 250 mL)
and 5% aqueous sodium bicarbonate solution (250 mL). The
aqueous layers were back-extracted with dichloromethane, the
combined organic layers were dried over sodium sulfate, and
the solvent was evaporated under reduced pressure. The residue
was dissolved in 5% sulfuric acid-methanol (50 mL) and stirred at
room temperature in the dark overnight. The mixture was diluted
with dichloromethane and washed with water and then with 5%
aqueous sodium bicarbonate solution. The aqueous layers were
back-extracted with dichloromethane at each stage, the combined
organic layers were dried over sodium sulfate, and the solvent was
evaporated under reduced pressure. The residue was chromato-
graphed on a grade 3 alumina column, eluting with dichloro-
methane. A dark violet product fraction was collected, the solvent
was evaporated under reduced pressure, and the residue recrys-
tallized from chloroform-methanol to give the chloroethylpor-
phyrin (254 mg, 0.37 mmol, 70%) as fluffy maroon crystals: mp
243-244 °C; UV-vis (1% Et3N-CHCl3) λmax (log10 ε) 401
(5.26), 499 (4.25), 533 (4.13), 568 (4.01), 622 nm (3.92); UV-vis
(1% TFA-CHCl3): λmax (log10ε) 408 (5.55), 551 (4.29), 593 nm
(4.02); 1H NMR (CDCl3) δ -3.76 (2H, br s), 3.28 (6H, t, J = 7.8
Hz), 3.65 (3H, s), 3.670 (3H, s), 3.672 (3H, s), 3.686 (6H, s), 3.688
(3H, s), 3.693 (3H, s), 4.33 (2H, t, J = 7.6 Hz), 4.39-4.52 (6H, m),
4.54 (2H, t, J = 8.0 Hz), 10.03 (1H, s), 10.09 (2H, s), 10.11 (1H, s);
1H NMR (TFA-CDCl3) δ -3.55 (4H, br s), 3.16 (6H, t, J = 7.4
Hz), 3.68 (9H, s), 3.69 (3H, s), 3.697 (3H, s), 3.701 (3H, s), 3.73
(3H, s), 4.20 (2H, t, J = 7.2 Hz), 4.45-4.50 (6H, m), 4.57 (2H, t,
J = 7.2 Hz), 10.69 (1H, s), 10.81 (1H, s), 10.82 (1H, s), 10.84 (1H, s);
13C NMR (TFA-CDCl3) δ 12.1, 12.2, 12.3, 22.0, 30.1, 35.7, 44.1,
52.8, 99.3, 99.4, 99.5, 99.6, 138.1, 139.3, 139.5, 140.2, 140.4, 141.8,
141.9, 142.1, 142.4, 142.5, 142.6, 174.8. Anal. Calcd for C38H43ClN4O6:
C, 66.41; H, 6.31; N, 8.15. Found: C, 66.04; H, 6.25; N, 8.00.
Calcd for C38H42N4O6 /10CHCl3: C, 69.05; H, 6.40; N, 8.45.
3
Found: C, 68.76; H, 6.36; N, 8.31.
Enzyme Incubation Studies. Enzyme incubations and analyses
of metabolic products were carried out as described pre-
viously.10 HPLC analyses were performed using normal-phase
columns (5 m Partisil silica, Alltech) eluting with appropriate
ratios of ethyl acetate and cyclohexane. Kinetic data are re-
ported as mean ( standard deviation for three replicate experi-
ments, except in the case of isohardero’gen where only two
were performed, and compared statistically using analysis of
variance (ANOVA) following Fisher’s LSD post test. Values are
considered different at p < 0.05.
Porphyrins Isolated from Preparative Enzymic Studies. These
metabolites were obtained by reducing the tricarboxylic acids derived
from 11f and 8f, respectively, with 3% sodium amalgam and incubat-
ing the resulting porphyrinogens with chicken red cell hemolysates.
The products were purified as their methyl esters by flash chromato-
graphy, eluting with 10% ethyl acetate-toluene, and characterized by
FAB MS and 300 MHz proton NMR spectroscopy.
18-(2-Methoxycarbonylethyl)-2,7,12,17-tetramethyl-3,8,13-
1
trivinylporphyrin (15a): H NMR (CDCl3) δ -3.49 (2H, br s),
3.29 (2H, t, J = 7.6 Hz), 3.66 (3H, s), 3.68 (3H, s), 3.72 (3H, s),
3.74 (3H, s), 3.76 (3H, s), 4.44 (2H, t, J = 7.6 Hz), 6.17-6.25
(3H, m), 6.35-6.45 (3H, m), 8.24-8.38 (3H, m), 10.12 (1H, s),
10.21 (1H, s), 10.28 (2H, s); HRMS (FAB) m/z calcd for
C34H34N4O2 þ H 531.2760, found 531.2758.
3-Ethyl-18-(2-methoxycarbonylethyl)-2,7,12,17-tetramethyl-
8,13-divinylporphyrin (15b): 1H NMR (CDCl3) δ -3.5 (2H, br s),
1.88 (3H, t, J = 7.5 Hz), 3.27 (2H, t, J = 7.6 Hz), 3.63 (3H, s),
3.65 (3H, s), 3.67 (3H, s), 3.74 (3H, s), 3.75 (3H, s), 4.11 (2H, q,
J = 7.5 Hz), 4.40 (2H, t, J = 7.6 Hz), 6.18-6.23 (2H, m), 6.38
(2H, d, J = 17.7 Hz), 8.31 (2H, m), 10.05 (1H, s), 10.14 (1H, s),
10.22 (1H, s), 10.29 (1H, s); HRMS (FAB) m/z calcd for
C34H36N4O2 þ H 533.2916, found 533.2917.
Acknowledgment. This work was supported by National
Institutes of Health under Grant No. 1 R15 GM/OD52687-
01A1 and by the National Science Foundation under Grant
No. CHE-0616555.
8,13,18-Tris(2-methoxycarbonylethyl)-2,7,12,17-tetramethyl-
3-vinylporphyrin (11f; Harderoporphyrin-I Trimethyl Ester).
DBU (5 drops) was added to a solution of chloroethylpor-
phyrin 8f (40 mg, 0.058 mmol) in DMF (40 mL), and the mixture
was refluxed with stirring under nitrogen for 1 h. The solu-
tion was cooled, diluted with dichloromethane, and washed with
5% hydrochloric acid, 5% aqueous sodium bicarbonate solu-
tion, and water. The solvent was evaporated under reduced
Supporting Information Available: Full experimental de-
tails, 1H NMR and 13C NMR spectra for selected compounds,
and selected MS and HPLC data. This material is available free
of charge via the Internet at http://pubs.acs.org.
3192 J. Org. Chem. Vol. 75, No. 10, 2010