the previously postulated role for P450BioI in biotin biosyn-
thesis.4
The fragmentation patterns of monohydroxy fatty acid
esters have been investigated previously,6-8 and several
characteristic fragmentations have been reported. The frag-
ments that are indicative of the position of hydroxylation
correspond to ones arising from cleavage R to the hydroxyl
group (Figure 1, A) and from R-cleavage and loss of
From our data, however, we were not able to exclude the
possibility that ω-functionalization of the fatty acid was
occurring. This would provide an additional path to pimelic
acid for biotin biosynthesis and a possibly different biological
role for P450BioI (ω-hydroxylation versus C-C bond cleav-
age). Thus, it was necessary to unambiguously identify all
of the hydroxy fatty acids formed by P450BioI oxidation, and
synthesis of a suite of hydroxy fatty acid standards was
undertaken (Scheme 1).
Figure 1. Characteristic mass spectral fragments of monohydroxy
fatty acid methyl esters.
The methyl esters of the required compounds (1a-d and
2a-e) were accessed conveniently through standard trans-
formations of the commercially available, terminally difunc-
tionalized compounds, undec-10-en-1-ol and 1,12-dodecan-
diol. Use of these precursors allowed rapid assembly of all
the required compounds via similar methodology (methyl
esters of 2a-e (Scheme 1), methyl esters of 1a-d (see
Supporting Information)). With the standards in hand, we
were able to confirm the predicted trends in GC/MS behavior
with respect to both retention time and MS fragmentation
relative to the position of the hydroxyl on the fatty acid alkyl
chain. GC/MS comparison of the synthetic standards and
their TFA esters with the products of P450BioI-catalyzed
oxidation confirmed the presence of 1a-c and 2a-e within
the enzymatic turnover mixtures of the C14 and C16 fatty
acids, respectively.
Importantly, no product was observed within the detection
limits of the GC/MS (<1% of total product) corresponding
to the methyl ester of authentic 14-hydroxytetradecanoic acid
1d in the tetradecanoic acid oxidations. There was also no
unidentified peak in the hexadecanoic acid oxidations that
may have been the methyl ester of 16-hydroxyhexadecanoic
acid 2f. These would be the products of ω-hydroxylation,
and therefore P450BioI does not act as an ω-hydroxylase.
Rather, it is only capable of in-chain hydroxylation at
positions prior to the terminal carbon, as well as C-C bond
cleavage to yield pimelic acid equivalents.
methanol (Figure 1, B). A plausible mechanism for the
formation of this latter peak is lactonization with the resultant
loss of methanol, followed by radical scission of the carbon
chain R to the lactone ring oxygen. As previously reported,
the ω- and ω-1-hydroxy fatty acid esters follow different
major fragmentation paths resulting from specific hydrogen
transfer (Figure 1, C).6-8
Utilizing these characteristic fragmentation patterns of hy-
droxy fatty acid esters, we undertook GC/MS analysis of
the P450BioI-catalyzed oxidation of tetradecanoic and hexa-
decanoic acids.
11-Hydroxytetradecanoic acid 1a and 12-hydroxytetrade-
canoic acid 1b were tentatively identified as the major
products from tetradecanoic acid oxidation (major fragments
of the methyl esters at m/z 215 (A)/183 (B) for 1a and 229
(A)/197 (B) for 1b). 11- and 12-hydroxyhexadecanoic acids
(2a and 2b, respectively) were also identified from hexa-
decanoic acid oxidation. However, analysis of the (better
resolving) trifluoroacetate esters of the esterified products
of P450BioI oxidation indicated that there were three hydroxy
fatty acids produced from tetradecanoic acid oxidation and
five from hexadecanoic acid. Unfortunately, the mass spectra
of the TFA esters lacked the characteristic fragmentation
required to identify the position of hydroxylation. However,
we predicted that the position of oxidation in the unidentified
products was likely to be closer to the methyl terminus of
the fatty acid chain than in the 11- and 12-hydroxy isomers.
This was based upon experience of the change in retention
times relative to the position of the alcohol moiety in the
fatty acid chain.
(6) Kawamura, K.; Gagosian, R. B. J. Chrom. 1988, 438, 309.
(7) Kerwin, J. L.; Torvik, J. J. Anal. Biochem. 1996, 237, 56.
(8) Wilson, R.; Smith, R.; Wilson, P.; Shepherd, M. J.; Riemersma, R.
A. Anal. Biochem. 1997, 248, 76.
3342
Org. Lett., Vol. 5, No. 18, 2003