8670
C. Pichon-Santander, A. I. Scott / Tetrahedron Letters 46 (2005) 8669–8672
B12, as well as the production of novel porphyrinoid
P
A
compounds.
A
P
A
P
NH
NH
HN
HN
It has been demonstrated earlier that 2-hydroxymethyl
PBG (1a) was a good substrate for the enzyme PBG
deaminase.8 We already had available from previous
work a family of 2-formylpyrroles bearing one or two
methoxycarbonate b-substituents (Scheme 1). For our
experiments, the methyl esters were hydrolyzed and
the formyl group reduced with sodium borohydride.
The reaction completion and purity of the compounds
PBG
deaminase
HO
H
H
N
H
A
P
NH2
PBG
P
A
HMB
1
were checked by H and 13C NMR in D2O.
Uro'gen III
synthase
H+ chemically
A
Overexpression and purification of PBG deaminase were
conducted as described previously.9 The enzyme was
used pure or as a lysate after heat treatment (10 min,
50 °C)10 to eliminate any trace of the next enzyme in
the porphyrinoid pathway, Uroꢀgen III synthase.
P
A
P
A
P
A
P
HN
HN
NH
NH
HN
NH
The free-acid 2-hydroxymethylpyrroles were incubated
with PBG deaminase under standard conditions.11
Incubation conditions and work up were optimized for
the 2-hydroxymethyl PBG (1a): it was found essential
to oxidize the porphyrin products before esterification
to avoid acid-catalyzed rearrangement of the porphyrino-
gens. It was also found important to remove all unre-
acted pyrroles before esterification to avoid additional
chemical formation of porphyrinogens during the acid-
catalyzed esterification process.
NH
HN
A
A
A
P
P
P
P
A
Uro'gen III
Uro'gen I
(+ other 3 isomers)
Scheme 2. Formation of Uroꢀgen I and III.
two carboxylate substituents (3a, Scheme 1) for subse-
quent use in enzymatic studies on the biosynthetic path-
way to vitamin B12, the most direct approach by
polymerization of four pyrrole units followed by cycliza-
tion and oxidation to porphyrins seems the method of
choice for us. Although Smith et al. reported formation
of 100% type I coproporphyrin (3f),2b their reaction con-
ditions do not work well with pyrroles bearing two elec-
tron-donating groups. Other conditions developed by
Eschenmoser and co-workers3 and in our group4 give
high ratios of type I (89%,3 72–80%4), but contamina-
tion with other isomers remains a concern. In nature
however, porphobilinogen (PBG) deaminase, one of
the early enzymes in the porphyrinoid pathway, per-
forms the tetramerization of PBG to the tetrapyrrole
1-hydroxymethylbilane (HMB), which upon subsequent
chemical cyclization and oxidation generates pure Uro I
(Scheme 2).5 Earlier studies in our laboratory on the
enzyme have shown activity toward some substrate
analogs6 and we thought it would be of interest to inves-
tigate this approach further.
Results of the PBG deaminase incubation with 2-
hydroxymethylpyrroles (Scheme 1) are shown in Table
1. Determination of type I ratios were based on 1H
NMR or HPLC analyses, depending on the products.12
Comparison of the results (yields and type I ratios) of
incubations with and without deaminase reveals a signi-
ficant difference only in the case of pyrrole 1a; with all
other pyrroles, both yields and ratios do not appear sig-
nificantly affected by the presence of enzyme. However,
a comparison with other methods shows that porphyrin
formation in buffer often gives better type I ratios. Thus,
the conditions described in this paper constitute the best
tetramerization method to obtain porphyrins with car-
boxylated b-substituents.
Although PBG deaminase seems to be too substrate-
specific to be used in the synthesis of HMB analogs,
in situ tetramerization of 2-hydroxymethylpyrroles bear-
ing carboxylate as b-substituent in buffer (pH 7.8–8.0)
proceeds reasonably well (Table 1) and allows the study
of the effect of Uroꢀgen III synthase on the cyclization
process (Scheme 2).
The next enzyme in the pathway, Uroporphyrinogen
(Uroꢀgen) III synthase, uses the tetrapyrrole HMB as
substrate and cyclizes it to Uroꢀgen III after rearrange-
ment of the last pyrrole ring (Scheme 2).5,7 The synthesis
of type III porphyrins being rather lengthy,1 it would
certainly be interesting to explore the possibility of using
this enzyme in combination with PBG deaminase to
obtain some Uro III analogs.
Overexpression of Uroꢀgen III synthase was conducted
as described previously.13 The enzyme was used as
lysate, incubations were run in the same standard condi-
tions as for PBG deaminase and worked up following
the same protocol.11 The results are shown in Table 2.
With pyrrole 1a, good yields were obtained with very
high ratios of type III porphyrin. This was expected
since the product formed in situ by the tetramerization
of pyrrole 1a is HMB, the natural substrate of Uroꢀgen
III synthase (Scheme 2). With other pyrroles, yields of
The availability of such Uro I or III analogs would
allow further probing and understanding of the enzy-
matic steps that follow in the biosynthesis of vitamin