Hydroxynitrile lyases: Functions and properties

Plant hydroxynitrile lyases (Hnl) have attracted the attention of bioorganic scientists for more than 90 years. However, the most important increase in knowledge of this class of enzymes has only arisen in the recent decade. The industrial application of these enzymes as biocatalysts for the synthesis of enantiomerically pure α-cyanohydrins may be responsible for the growing interest in this area.
The Hnls are involved in the catabolic degradation of cyanogenic glycosides, releasing HCN which serves as defense agent against herbivores and microbial attack, or as a nitrogen source. Hydroxynitrile lyases from various plant families appear to represent a new example of enzymes that originated from the convergent evolution of different precursor proteins. The enzymes have been classified into non-FAD- and FAD-containing proteins. FAD-containing enzymes have been isolated exclusively from the Rosaceae, whereas the FAD-independent Hnls, which are more heterogenous in structure, have been characterized from various plant families (Poaceae, Euphorbiaceae, Linaceae, Olacaceae. Filitaceae). The aim of this review is to present a general survey of the natural function and localization of this class of enzymes and a comprehensive summary of the biochemical and genetic data of the isolated proteins.     

Structural and Functional Comparison of Polysaccharide-Degrading Enzymes

Sugar molecules as well as enzymes degrading them are ubiquitously present in physiological systems, especially for vertebrates. Polysaccharides have at least two aspects to their function, one due to their mechanical properties and the second one involves multiple regulatory processes or interactions between molecules, cells, or extracellular space. Various bacteria exert exogenous pressures on their host organism to diversity glycans and their structures in order for the host organism to evade the destructive function of such microbes. Many bacterial organism produce glycan-degrading enzymes in order to facilitate their invasion of host tissues. Such polysaccharide degrading enzymes utilize mainly two modes of polysaccharide-degradation, a hydrolysis and a β-elimination process. The three-dimensional structures of several of these enzymes have been elucidated recently using X-ray crystallography. There are many common structural motifs among these enzymes, mainly the presence of an elongated cleft transversing these molecules which functions as a polysaccharide substrate binding site as well as the catalytic site for these enzymes. The detailed structural information obtained about these enzymes allowed formulation of proposed mechanisms of their action. The polysaccharide lyases utilize a proton acceptance and donation mechanism (PAD), whereas polysaccharide hydrolases use a direct double displacement (DD) mechanism to degrade their substrates.     

Cover Image

Synthetic microbial communities have become a focus of biotechnological research since they can overcome several of the limitations of single-specie cultures. A paradigmatic example is Clostridium cellulovorans DSM 743B, which can decompose lignocellulose but cannot produce butanol. Clostridium beijerinckii NCIMB 8052 however, is unable to use lignocellulose but can produce high amounts of butanol from simple sugars. In our previous studies, both organisms were cocultured to produce butanol by consolidated bioprocessing. However, such consolidated bioprocessing implementation strongly depends on pH regulation. Since low pH (pH 4.5–5.5) is required for butanol fermentation, C. cellulovorans cannot grow well and saccharify sufficient lignocellulose to feed both strains at a pH below 6.4. To overcome this bottleneck, this study engineered C. cellulovorans by adaptive laboratory evolution, inactivating cell wall lyases genes (Clocel_0798 and Clocel_2169), and overexpressing agmatine deiminase genes (augA, encoded by Cbei_1922) from C. beijerinckii NCIMB 8052. The generated strain WZQ36: 743B*6.0*3△lyt0798△lyt2169-(pXY1-P_thl-augA) can tolerate a pH of 5.5. Finally, the alcohol aldehyde dehydrogenase gene adhE1 from Clostridium acetobutylicum ATCC 824 was introduced into the strain to enable butanol production at low pH, in coordination with solvent fermentation of C. beijerinckii in consortium. The engineered consortium produced 3.94 g/L butanol without pH control within 83 hr, which is more than 5-fold of the level achieved by wild consortia under the same conditions. This exploration represents a proof of concept on how to combine metabolic and evolutionary engineering to coordinate coculture of a synthetic microbial community.     

Over-expression of hydroxynitrile lyase in transgenic cassava roots accelerates cyanogenesis and food detoxification

Cassava (Manihot esculenta, Crantz) roots are the primary source of calories for more than 500 million people, the majority of whom live in the developing countries of Africa. Cassava leaves and roots contain potentially toxic levels of cyanogenic glycosides. Consumption of residual cyanogens (linamarin or acetone cyanohydrin) in incompletely processed cassava roots can cause cyanide poisoning. Hydroxynitrile lyase (HNL), which catalyses the conversion of acetone cyanohydrin to cyanide, is expressed predominantly in the cell walls and laticifers of leaves. In contrast, roots have very low levels of HNL expression. We have over-expressed HNL in transgenic cassava plants under the control of a double 35S CaMV promoter. We show that HNL activity increased more than twofold in leaves and 13-fold in roots of transgenic plants relative to wild-type plants. Elevated HNL levels were correlated with substantially reduced acetone cyanohydrin levels and increased cyanide volatilization in processed or homogenized roots. Unlike acyanogenic cassava, transgenic plants over-expressing HNL in roots retain the herbivore deterrence of cyanogens while providing a safer food product.     

A kinetic model for enzyme interfacial activity and stability: pa-hydroxynitrile lyase at the diisopropyl ether/water interface

A kinetic framework is developed to describe enzyme activity and stability in two-phase liquid-liquid systems. In particular, the model is applied to the enzymatic production of benzaldehyde from mandelonitrile by Prunus amygdalus hydroxynitrile lyase (pa-Hnl) adsorbed at the diisopropyl ether (DIPE)/aqueous buffer interface (pH = 5.5). We quantitatively describe our previously obtained experimental kinetic results (Hickel et al., 1999; 2001), and we successfully account for the aqueous-phase enzyme concentration dependence of product formation rates and the observed reaction rates at early times. Multilayer growth explains the early time reversibility of enzyme adsorption at the DIPE/buffer interface observed by both enzyme-activity and dynamic-interfacial-tension washout experiments that replace the aqueous enzyme solution with a buffer solution. The postulated explanation for the unusual stability of pa-Hnl adsorbed at the DIPE/buffer interface is attributed to a two-layer adsorption mechanism. In the first layer, slow conformational change from the native state leads to irreversible attachment and partial loss of catalytic activity. In the second layer, pa-Hnl is reversibly adsorbed without loss in catalytic activity. The measured catalytic activity is the combined effect of the deactivation kinetics of the first layer and of the adsorption kinetics of each layer. For the specific case of pa-Hnl adsorbed at the DIPE/buffer interface, this combined effect is nearly constant for several hours resulting in no apparent loss of catalytic activity. Our proposed kinetic model can be extended to other interfacially active enzymes and other organic solvents. Finally, we indicate how interfacial-tension lag times provide a powerful tool for rational solvent selection and enzyme engineering.     

Biosynthesis of the leucine derived α‐, β‐ and γ‐hydroxynitrile glucosides in barley (Hordeum vulgare L.)

Barley (Hordeum vulgare L.) produces five leucine‐derived hydroxynitrile glucosides (HNGs), of which only epiheterodendrin is a cyanogenic glucoside. The four non‐cyanogenic HNGs are the β‐HNG epidermin and the γ‐HNGs osmaronin, dihydroosmaronin and sutherlandin. By analyzing 247 spring barley lines including landraces and old and modern cultivars, we demonstrated that the HNG level varies notably between lines whereas the overall ratio between the compounds is constant. Based on sequence similarity to the sorghum (Sorghum bicolor) genes involved in dhurrin biosynthesis, we identified a gene cluster on barley chromosome 1 putatively harboring genes that encode enzymes in HNG biosynthesis. Candidate genes were functionally characterized by transient expression in Nicotiana benthamiana. Five multifunctional P450s, including two CYP79 family enzymes and three CYP71 family enzymes, and a single UDP‐glucosyltransferase were found to catalyze the reactions required for biosynthesis of all five barley HNGs. Two of the CYP71 enzymes needed to be co‐expressed for the last hydroxylation step in sutherlandin synthesis to proceed. This observation, together with the constant ratio between the different HNGs, suggested that HNG synthesis in barley is organized within a single multi‐enzyme complex.     

Mass balance analysis of contaminated heparin product

A quantitative analysis of a recalled contaminated lot of heparin sodium injection U.S. Pharmacopeia (USP) was undertaken in response to the controversy regarding the exact nature of the contaminant involved in the heparin (HP) crisis. A mass balance analysis of the formulated drug product was performed. After freeze-drying, a 1-ml vial for injection afforded 54.8 ± 0.3 mg of dry solids. The excipients, sodium chloride and residual benzyl alcohol, accounted for 11.4 ± 0.5 and 0.9 ± 0.5 mg, respectively. Active pharmaceutical ingredient (API) represented 41.5 ± 1.0 mg, corresponding to 75.7 wt% of dry mass. Exhaustive treatment of API with specific enzymes, heparin lyases, and/or chondroitin lyases was used to close mass balance. HP represented 30.5 ± 0.5 mg, corresponding to 73.5 wt% of the API. Dermatan sulfate (DS) impurity represented 1.7 ± 0.3 mg, corresponding to 4.1 wt% of API. Contaminant, representing 9.3 ± 0.1 mg corresponding to 22.4 wt% of API, was found in the contaminated formulated drug product. The recovery of contaminant was close to quantitative (95.6–100 wt%). A single contaminant was unambiguously identified as oversulfated chondroitin sulfate (OSCS).     

Dickeya dadantii pectic enzymes necessary for virulence are also responsible for activation of the Arabidopsis thaliana innate immune system

Soft‐rot diseases of plants attributed to Dickeya dadantii result from lysis of the plant cell wall caused by pectic enzymes released by the bacterial cell by a type II secretion system (T2SS). Arabidopsis thaliana can express several lines of defence against this bacterium. We employed bacterial mutants with defective envelope structures or secreted proteins to examine early plant defence reactions. We focused on the production of AtrbohD‐dependent reactive oxygen species (ROS), callose deposition and cell death as indicators of these reactions. We observed a significant reduction in ROS and callose formation with a bacterial mutant in which genes encoding five pectate lyases (Pels) were disrupted. Treatment of plant leaves with bacterial culture filtrates containing Pels resulted in ROS and callose production, and both reactions were dependent on a functional AtrbohD gene. ROS and callose were produced in response to treatment with a cellular fraction of a T2SS‐negative mutant grown in a Pels‐inducing medium. Finally, ROS and callose were produced in leaves treated with purified Pels that had also been shown to induce the expression of jasmonic acid‐dependent defence genes. Pel catalytic activity is required for the induction of ROS accumulation. In contrast, cell death observed in leaves infected with the wild‐type strain appeared to be independent of a functional AtrbohD gene. It was also independent of the bacterial production of pectic enzymes and the type III secretion system (T3SS). In conclusion, the work presented here shows that D. dadantii is recognized by the A. thaliana innate immune system through the action of pectic enzymes secreted by bacteria at the site of infection. This recognition leads to AtrbohD‐dependent ROS and callose accumulation, but not cell death.