Objectives:

Molecular modelling of SPM-1 showed that it has a very similar structure to IMP-1, differing primarily in the flexible 24 amino acid loop unique to SPM-1. SPM-1 and IMP-1 show very different kinetics, indicating a hydrolytic role for the loop. Accordingly, the loop was deleted and substrate binding and hydrolysis and zinc content were compared with wild-type SPM-1 (wtSPM-1).

Methods:

The 24 amino acid loop was deleted from SPM-1 (SPM-Ldel) by PCR overlapping primer extension techniques and confirmed by sequencing. The PCR product was over-expressed in Escherichia coli and confirmed by PCR using SPM-1-specific primers. RT-PCR was used to verify gene expression. SPM-Ldel was purified using an ammonium sulphate precipitation and standard protein chromatography, and the hydrolytic profile was determined for, cefuroxime, penicillin G, nitrocefin, imipenem and meropenem. Zinc assays were undertaken using Unicam 919 atomic absorption spectroscopy.

Results:

Sequence analysis verified the deletion and RT-PCR confirmed gene expression. The SPM-Ldel gave kcats of 0.07 s^-1 and 0.14 s^-1 for imipenem and meropenem, respectively, compared with 33 and 63 s^-1, for the wtSPM-1. K_m values decreased from 281 to 61 mm for meropenem and from 37 to 9.5 mm for imipenem. For cefuroxime, k_cat and K_m values changed from 37 to 0.6 s^-1 and from 4 to 5 mm, respectively. SPM-Ldel weakly hydrolysed penicillin with a kcat of 3.17 s^-1 and a K_m of 226 compared with the wtSPM-1 (k_cat 108 s^-1 and K_m of 38 mm). Interestingly, the nitrocefin k_cat value did not markedly change (0.53 s^-1 for wtSPM-1 compared with 0.51 s^-1 for SPM-Ldel) and the K_m increased from 4 to 23 mm.

Conclusions:

Unexpectedly, removal of the loop did not consistently result in weaker substrate binding.

The lower kcat values for SPM-Ldel demonstrate that the loop is integral to the hydrolysis of beta-lactams by SPM-1.