Enterobacter cloacae GC1, a clinical strain isolated in 1992 in Japan, was found to produce a chromosomal class C beta-lactamase with extended substrate specificity to oxyimino beta-lactam antibiotics, significantly differing from the known E. cloacae beta-lactamases such as the P99 beta-lactamase. The 1560 nucleotides including the GC1 beta-lactamase gene were sequenced, and the amino acid sequence of the mature enzyme comprising 364 amino acids was deduced. A comparison of the amino acid sequence with those of known E. cloacae beta-lactamases revealed the duplication of three amino acids at positions 208-213, i.e. Ala-Val-Arg-Ala-Val-Arg. This duplication was attributed to a tandem duplication of a 9-nucleotide sequence. The chimeric beta-lactamases produced by the chimeric genes from the GC1 and P99 beta-lactamase genes indicated that the extended substrate specificity is entirely attributed to the 3-amino acid insertion. Two mutant beta-lactamases were prepared from P99 beta-lactamase by site-directed mutagenesis, i.e. an Ala-Ala-Ala sequence was inserted before or after the native Ala-Val-Arg at positions 208-210. These mutant enzymes revealed that the Ala-Val-Arg located from positions 211 to 213 in the GC1 beta-lactamase are the newly inserted residues, and this phenomenon is independent of the characteristics of the amino acids inserted.

The class C p-lactamase of Citrobacter fracndii GN346 is a typical cephalosporinase comprising 361 ammo acids. The aspartic acid at position 217 and glutamic acid at position 219 in this B-lactamase were, respectively, previously shown not to be the counterpart of G~u'~ (ABL166) in class AB-lactamases, even though sequence alignment of class A and C enzymes strongly suggested this possibility [(1990) FEBS Lett. 264,21 l-214; (1990) J. Bacterial. 172,4348-4351]. We tried again to assign candidates for the counterpart of 0111'~ through sequence alignment based on other criteria, the glutamic acids at positions 195 and 205 in the class C p-lactamase being selected. To investigate this possibility, these two glutamic acids were changed to glutamine, lysine or alanine, respectively. All the mutant enzymes showed more than SO% of the activity of the wild-type enzyme, indicating that the possibility was ruled out. These results strongly suggested the possibility that the class C b-lactamase lacks a functional acidic residue corresponding to Glu'= in class A enzymes.