90
H. Zhu et al.
In conclusion, the above investigations revealed that the early stages of cytochalasan biosynthesis depend on the co-expression of both PKS–NRPS and its trans-ER.
In the heterologous host A. oryzae, both of the co-expression products, amides A1
and A5, possess a hydroxy group moiety rather than the expected formyl group, and
suggested that an unwanted reduction occurred during the backbone formation. It
remains to be seen whether this excessive reduction was caused by native enzymes
or by the alternative splicing of the PKS–NRPS functional domain in the heterologous host. It has been observed that the co-expression of ccsA and ccsC in different
hosts led to two distinct intermediates, A1 and A2, respectively. A proposal made
by Skellam [2] is that native accessory enzymes in A. oryzae reduce the proposed
aldehyde but in A. nidulans and A. niger drive the formation of the pyrrolinone
unit. These modifications by the heterologous host may prevent the formation of the
required pyrrolinone and thus prevent also the [4 + 2] Diels–Alder cycloaddition
reaction occurring with the terminal diene [2].
4.2.2 The Middle Stage of Cytochalasan Biosynthesis
The middle stage of cytochalasan biosynthesis is defined as the conversion from
the linear PKS–NRPS intermediate to the iso-indolone fused macrocyclic structure.
Although there is only one intramolecular [4 + 2] Diels–Alder (DA) cycloaddition
involved, it is a critical step in the formation of the cytochalasan skeleton and a step
that has intrigued researchers for many years. It is believed that the DA reaction
is catalyzed by specific enzymes and several pieces of indirect evidence have been
provided through well-designed experiments.
Earlier evidence came from the total synthesis of cytochalasans, as reviewed by
Thomas in 1991 [263]. As a key step in the synthesis of proxiphomin (102), a naturally
occurring cytochalasan, the intramolecular DA reaction of B1 required relatively
rigorous conditions (100 °C for 5 h) to give a mixture of B2 and B3, suggesting that
spontaneous cycloaddition in fungal cells was unlikely (Scheme 3A).
In 1992, Oikawa et al. demonstrated that the stereospecificity of the cytochalasan
DA reaction might require enzymatic stereo-control [264]. As shown in Scheme 3B,
after the pyrolysis of prochaetoglobosin I (221), the authors found that both diastereomers 221 and diastereo-221 existed in the product mixture. From these results, it
could be speculated that 221 first undergoes a reverse DA reaction to form a linear
intermediate B4, namely, the proposed substrate for the DA reaction. As it cyclized
spontaneously under the reaction conditions used, an equal amount of diastereomers
were produced. The lack of stereoselectivity in this reaction indicated that the enzyme
responsible for [4 + 2] cycloaddition stabilizes the endo transition state and results
exclusively in the production of prochaetoglobosin I (221).
In addition to the above evidence from chemical reactions, progress made in
genetic engineering on fungi in recent years has accelerated the pace of searching
for DA enzymes. The formation of decalin, which involves a DA reaction similar
to that involved in producing the cytochalasans, was proved to be catalyzed by the
enzyme MycB by the group of Tang [265] (Scheme 3C). Highlights of this study
Précédent

- 96/341

Suivant