Engineering Papers⌕ Search

Engineering topics

Sankaran, B.

Publications and source records attributed to Sankaran, B..

Robust deep learning–based protein sequence design using ProteinMPNN

Although deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here, we describe a deep learning–based protein sequence design method, ProteinMPNN, that has outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4% compared with 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using x-ray crystallography, cryo–electron microscopy, and functional studies by rescuing previously failed designs, which were made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target-binding proteins.

59 BASIC BIOLOGICAL SCIENCES↗

Purification, Crystallization, and Preliminary Crystallographic Studies of Human As(III) S-Adenosylmethionine Methyltransferase (hAS3MT)

Exposure to environmental arsenic is associated with serious of health issues such as cancer, diabetes and developmental delays in infants and children. In human liver, As(III) S-adenosylmethionine methyl transferase (hAS3MT) (EC 2.1.1.137) was proposed to be an detoxification process by methylation of inorganic arsenite into pentavalent methyl MAs(V) and dimethyl arsenite DMAs(V). More recently the first product was shown to be highly toxic and potentially carcinogenic trivalent methylarsenite (MAs(III)). Our studies are designed to elucidate the mechanism of AS3MT and its contribution to arsenic-related diseases. Here, in this paper, we report the first crystallization and preliminary X-ray diffraction analysis of the human AS3MT enzyme. The crystals belong to the monoclinic P12 1 1 space group with unit cell parameters of a = 135.03 Å, b = 260.44 Å, c = 279.03 Å, α = 90.00°, β = 93.36°, γ = 90.00°.

59 BASIC BIOLOGICAL SCIENCES↗

Structural Insights into the Functional Role of the Hcn Sub-domain of the Receptor-Binding Domain of the Botulinum Neurotoxin Mosaic Serotype C/D

Botulinum neurotoxin (BoNT), the causative agent of the deadly neuroparalytic disease botulism, is the most poisonous protein known for humans. Produced by different strains of the anaerobic bacterium Clostridium botulinum, BoNT effects cellular intoxication via a multistep mechanism executed by the three modules of the activated protein. Endocytosis, the first step of cellular intoxication, is triggered by the ~50 kDa, heavy-chain receptor-binding module (HCR) that is specific for a ganglioside and a protein receptor on neuronal cell surfaces. This dual receptor recognition mechanism between BoNT and the host cell’s membrane is well documented and occurs via specific intermolecular interactions with the C-terminal sub-domain, Hcc, of BoNT-HCR. The N-terminal sub-domain of BoNT-HCR, Hcn, comprises ~50% of BoNT-HCR and adopts a B-sheet jelly roll fold. While suspected in assisting cell surface recognition, no unambiguous function for the Hcn sub-domain in BoNT has been indentified. To obtain insights into the potential function of the Hcn sub-domain in BoNT, the first crystal structure of a BoNT with an organic ligand bound to the Hcn sub-domain has been obtained. Here, we describe the crystal structure of BoNT/CD-HCR determined at 1.70 Å resolution with a tetraethylene glycol (PG4) molecule bound in an hydrophobic cleft between B-strands in the B-sheet jelly fold roll of the Hcn sub-domain. The molecule is completely engulfed in the cleft, making numerous hydrophobic (Y932, S959, W966, and D1042) and hydrophilic (S935, W977, L979, N1013, and I1066) contacts with the protein’s side chain and backbone that may mimic in vivo interactions with the phospholipid membranes on neuronal cell surfaces. A sulfate ion was also observed bound to residues T1176, D1177, K1196, and R1243 in the Hcc sub-domain of BoNT/CD-HCR. In the crystal structure of a similar protein, BoNT/D-HCR, a sialic acid

botulinum neurotoxin↗

Structure of nitrilotriacetate monooxygenase component B from Mycobacterium thermoresistibile

Mycobacterium tuberculosis belongs to a large family of soil bacteria which can degrade a remarkably broad range of organic compounds and utilize them as carbon, nitrogen and energy sources. It has been proposed that a variety of mycobacteria can subsist on alternative carbon sources during latency within an infected human host, with the help of enzymes such as nitrilotriacetate monooxygenase (NTA-Mo). NTA-Mo is a member of a class of enzymes which consist of two components: A and B. While component A has monooxygenase activity and is responsible for the oxidation of the substrate, component B consumes cofactor to generate reduced flavin mononucleotide, which is required for component A activity. NTA-MoB from M. thermoresistibile, a rare but infectious close relative of M. tuberculosis which can thrive at elevated temperatures, has been expressed, purified and crystallized. The 1.6 Å resolution crystal structure of component B of NTA-Mo presented here is one of the first crystal structures determined from the organism M. thermoresistibile. The NTAMoB crystal structure reveals a homodimer with the characteristic split-barrel motif typical of flavin reductases. Surprisingly, NTA-MoB from M. thermoresistibile contains a C-terminal tail that is highly conserved among mycobacterial orthologs and resides in the active site of the other protomer. Based on the structure, the C-terminal tail may modulate NTA-MoB activity in mycobacteria by blocking the binding of flavins and NADH.

59 BASIC BIOLOGICAL SCIENCES↗

Structure of aldose reductase from Giardia lamblia

Giardia lamblia is an anaerobic aerotolerant eukaryotic parasite of the intestines. It is believed to have diverged early from eukarya during evolution and is thus lacking in many of the typical eukaryotic organelles and biochemical pathways. Most conspicuously, mitochondria and the associated machinery of oxidative phosphorylation are absent; instead, energy is derived from substratelevel phosphorylation. Here, the 1.75 Å resolution crystal structure of G. lamblia aldose reductase heterologously expressed in Escherichia coli is reported. As in other oxidoreductases, G. lamblia aldose reductase adopts a TIM-barrel conformation with the NADP+ -binding site located within the eight -strands of the interior.

59 BASIC BIOLOGICAL SCIENCES↗