Studies Towards the Synthesis of Thienamycin

Wardrop, Duncan J (1994) Studies Towards the Synthesis of Thienamycin. PhD thesis, University of Glasgow.

Abstract

Almost all successful syntheses of bicyclic beta-lactams involve the early synthesis of the monocyclic azetidin-2-one ring. Of the numerous methods developed to achieve this, the [2+2] cycloaddition of chlorosulphonyl isocyanate (CSI) to functionalised alkenes has proven to be particularly useful. In general, addition is performed on alkenyl acetates providing C-4 acetoxyazetidinones, which have found widespread use since the 4-acetoxy substituent can be replaced with various nucleophiles in an elimination/addition sequence. However, there are few functionalised alkenes which directly introduce the C-4 carbon substitution required for carbapenem synthesis upon cycloaddition with CSI. Earlier investigations within this group have shown that allylsilanes undergo regioselective cycloaddition with CSI to yield C-4 silylmethyl substituted beta-lactams.133,137 The regiochemistry of cycloaddition is controlled by the beta-effect of silicon, i.e., silicon's ability to stabilise the development of partial positive charge P to itself. Use of phenyldimethylsilyl substitution allows access, albeit in low yield, to the corresponding hydroxymethyl azetidinones, via oxidative cleavage of the C-Si bond. We were intrigued by reports from Fleming142 and Taddei141 that allylsilanes bearing a chiral centre directly adjacent to the double bond undergo reaction with various electrophiles, including CSI, with remarkably high stereoselectivity. The initial aim of this project was to prepare an ?-oxa-allylsilane, with such a chiral centre, which would undergo regio- and stereoselective cycloaddition with CSI to yield a C-3 hydroxyethyl substituted ?-lactam. Subsequent oxidative cleavage of the silylmethyl group would led to a useful precursor for the powerful carbapenem antibiotic Thienamycin. Unfortunately, all attempts to carry out cycloaddition with these substrates resulted in a rapid 1,4 silyl elimination to penta-l,3-diene. In an attempt to overcome this problem, an alternative allylsilane was designed and prepared. Disappointingly, however, this compound did not react with CSI. The second aim of this project was to improve the efficiency of the key oxidative cleavage step. Our interest was awakened by a report from Ito191 which briefly mentioned the use of iodine monochloride (ICl) for the iododesilylation of phenyldimethylsilyl moieties prior to oxidative cleavage. In order to study the applicability of this method to C-4 silylmethyl beta-lactams simple N-protected precursors were prepared. Gratifyingly, treatment with ICl resulted in cleavage of the Si-Ph bond and formation of the chlorodimethylsilyl beta-lactams which were hydrolysed to the corresponding silanols in high yield. Oxidation of these compounds using a modification of Tamao's191 procedure gave the potentially useful C-4 hydroxymethyl beta-lactams in good to moderate yield. Alternatively, one of the intermediate chlorosilanes could be directly oxidised using AcOOH/KF.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Adviser: E W Colvin
• Keywords: Organic chemistry, Pharmacology
• Date of Award: 1994
• Depositing User: Enlighten Team
• Unique ID: glathesis:1994-75650
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 19 Nov 2019 19:01
• Last Modified: 19 Nov 2019 19:01
• URI: https://theses.gla.ac.uk/id/eprint/75650