Capreomycin (CMN) is an important second-line antituberculosis antibiotic isolated from Saccharothrix mutabilis subspecies capreolus. The gene cluster for CMN biosynthesis has been identified and sequenced, wherein the cph gene was annotated as a phosphotransferase likely engaging in self-resistance. Previous studies reported that Cph inactivates two CMNs, CMN IA and IIA, by phosphorylation. We, herein, report that (1) Escherichia coli harboring the cph gene becomes resistant to both CMN IIA and IIB, (2) phylogenetic analysis regroups Cph to a new clade in the phosphotransferase protein family, (3) Cph shares a three-dimensional structure akin to the aminoglycoside phosphotransferases with a high binding affinity (K D) to both CMN IIA and IIB at micromolar levels, and (4) Cph utilizes either ATP or GTP as a phosphate group donor transferring its γ-phosphate to the hydroxyl group of CMN IIA. Until now, Cph and Vph (viomycin phosphotransferase) are the only two known enzymes inactivating peptide-based antibiotics through phosphorylation. Our biochemical characterization and structural determination conclude that Cph confers the gene-carrying species resistance to CMN by means of either chemical modification or physical sequestration, a naturally manifested belt and braces strategy. These findings add a new chapter into the self-resistance of bioactive natural products, which is often overlooked while designing new bioactive molecules.

Nanodiamond (ND) is a carbon-based nanomaterial with potential for a wide range of biological applications. One of such applications is to facilitate the nucleation of protein crystals in aqueous solution. Here, we show that NDs (nominal diameters of 30 and 100 nm) after surface oxidation in air and subsequent treatment in strong acids are useful as heterogeneous nucleating agents for protein crystallization. Tested with lysozyme, ribonuclease A, proteinase K, and catalase, the nanomaterials in either aggregate or film form are found to be able to increase the crystallization efficiency of all proteins. Particularly, for 30 nm NDs, the films with an area of ∼2 mm can effectively induce the crystallization of lysozyme at a concentration as low as 5 mg/mL. The efficiency can be further improved by adding preformed protein clusters (∼300 nm in diameter) as inherent nucleation precursors, as demonstrated for ribonuclease A. This combined approach is easy to implement, highly compatible with existing technologies, and can be applied to other protein samples as well.

Plant type III polyketide synthases produce diverse bioactive molecules with a great medicinal significance to human diseases. Here, we demonstrated versatility of a stilbene synthase (STS) from Pinus Sylvestris, which can accept various non-physiological substrates to form unnatural polyketide products. Three enzymes (4-coumarate CoA ligase, malonyl-CoA synthetase and engineered benzoate CoA ligase) along with synthetic chemistry was practiced to synthesize starter and extender substrates for STS. Of these, the crystal structures of benzoate CoA ligase (BadA) from Rhodopseudomonas palustris in an apo form or in complex with a 2-chloro-1,3-thiazole-5-carboxyl-AMP or 2-methylthiazole-5-carboxyl-AMP intermediate were determined at resolutions of 1.57 Å, 1.7 Å, and 2.13 Å, respectively, which reinforces its capacity in production of unusual CoA starters. STS exhibits broad substrate promiscuity effectively affording structurally diverse polyketide products. Seven novel products showed desired cytotoxicity against a panel of cancer cell lines (A549, HCT116, Cal27). With the treatment of two selected compounds, the cancer cells underwent cell apoptosis in a dose-dependent manner. The precursor-directed biosynthesis alongside structure-guided enzyme engineering greatly expands the pharmaceutical repertoire of lead compounds with promising/enhanced biological activities.

Human bleomycin hydrolase (hBH) catalyzes deamidation of the anticancer drug bleomycins (BLM). This enzyme is involved in BLM detoxification and drug resistance. Herein, we report the putative BLM-binding site and catalytic mechanism of hBH. The crystal structures and biochemical studies suggest that hBH cleaves its C-terminal residue without significant preference for the type of amino acid, and therefore can accordingly accommodate the β-aminoalanine amide moiety of BLM for deamidation. Interestingly, hBH is capable of switching from a cysteine protease to a serine protease that is unable to cleave the secondary amide of hBH C-terminus but reacts with the primary amide of BLMs.

In the development of new functionalities of transketolase for the industrial strain Pichia stipitis (TKps) the structural information of TKps would allow us to gain insight into the enzyme's reaction mechanisms, substrates selectivity and reaction directionality to help reach the goal. We here report seven TKps crystal structures of wild type (WT) and mutants in complex with various physiological ligands. These complexes were refined to resolutions at 1.6-1.03 Å. Both biochemical and mutagenic analyses concluded that residues His27, His66, His100, His261, His478, Asp473, Arg356 and Arg525 play important roles in coenzyme binding and substrates recognition. In general, His66 and His261 hold thiamine diphosphate in place; Arg356 and Arg525 serve as gatekeepers interacting with the terminal phosphate group of sugar-phosphates. His27, His66, His100, His478 and Asp473 are critical for sugars recognition/binding, in which His27 is relatively more important in interaction with sedoheptulose-7-phosphate (S7P) than xylulose-5-phosphate (X5P) in terms of molecular recognition/binding affinity. Kinetically, the reactions with X5P (forward) which were catalyzed by WT or H27A are indistinguishable, while in the reactions with S7P (backward) H27A exhibits weaker activity relative to WT. As a result, given TKps(H27A) as the biocatalyst the overall reactivity reverses from the backward reaction preference to forward, thus facilitating net xylose assimilation.