Autodisplay of Human Hyaluronidase Hyal-1 on Escherichia coli and Identification of Plant-Derived Enzyme Inhibitors

Abstract

Hyaluronan (HA) is the main component of the extracellular matrix (ECM). Depending on its chain size, it is generally accepted to exert diverse effects. High molecular weight HA is anti-angiogenic, immunosuppressive and anti-inflammatory, while lower fragments are angiogenic and inflammatory. Human hyaluronidase Hyal-1 (Hyal-1) is one of the main enzymes in the metabolism of HA. This makes Hyal-1 an interesting target. Not only for functional and mechanistic studies, but also for drug development. In this work, Hyal-1 was expressed on the surface of E. coli, by applying Autodisplay, to overcome formation of inactive "inclusion bodies". With the cells displaying Hyal-1 an activity assay was performed using "stains-all" dye. Subsequently, the inhibitory effects of four saponins and 14 plant extracts on the activity of surface displayed Hyal-1 were evaluated. The determined IC50 values were 177 µM for glycyrrhizic acid, 108 µM for gypsophila saponin 2, 371 µM for SA1657 and 296 µM for SA1641. Malvae sylvestris flos, Equiseti herba and Ononidis radix extracts showed IC50 values between 1.4 and 1.7 mg/mL. In summary, Autodisplay enabled the expression of functional human target protein Hyal-1 in E. coli and facilitated an accelerated testing of potential inhibitors.

Keywords: Autodisplay; Hyal-1; hyaluronan; natural inhibitors.

Figure 1

Schematic structure of the Hyal-1 autotransporter fusion protein. The DNA- and amino acid sequences are plotted. Restrictions endonucleases recognition sites for Xhol and KpnI, used for the construction of the artificial gene encoding the fusion protein, are underlined.

Figure 2

Whole cell enzyme-linked immunosorbent assay (ELISA). White: E. coli F470 cells containing pAK009 for surface displaying of Hyal-1. Black: E. coli F470 cells without plasmid. Wells of a Maxisorp^® 96-plate were coated with cell suspensions of various optical densities at 578 nm (OD₅₇₈) of 0.05; 0.1 and 1. After labelling with the primary anti-Hyal-1 antibody and incubation with a secondary antibody conjugated with horse radish peroxidase the reaction was started by adding of 3,3′,5,5′-tetramethylbenzidine (TMB). A light‑protected incubation was followed for 10 min at RT. Subsequently, the reaction was stopped by adding sulphuric acid. The absorbance was recorded at 450 nm (n = 3, error bars ± SD).

Figure 3

Surface exposure of Hyal-1 on E. coli F470. M: PageRuler Prestained Protein Ladder (Fermentas); Lane 1: E. coli F470 without plasmid (control sample); Lane 2: E. coli F470 + pAK009; Lane 3: E. coli F470 + pAK009 with whole cell digestion using proteinase K before membrane protein preparation. (A) Coomassie stained SDS‑PAGE of enriched outer membrane protein from E. coli F470. Samples were dissolved in denaturation buffer containing 2% SDS and 100 mM dithiothreitol (DTT) (final concentration) and heated for 10 min at 95 °C; (B) Antigenic evaluation of enriched outer membrane proteins from E. coli F470 after SDS-PAGE with 10% PA and western blot. For detection, polyclonal murine anti-Hyal-1 antibody and the secondary anti‑mouse IgG antibody conjugated with horse radish peroxidase (HRP) were used. Visualisation was performed using ECL reagent. Chemoluminescence was detected after an exposure time of 30 min and 30 s, documented in inverse mode.

Figure 4

Activity determination of Hyal-1 displayed on the cell surface of E. coli F470. E. coli F470 displaying Hyal-1 and the control sample E. coli F470 not displaying Hyal-1 (host strain) were suspended in formate buffer (pH 3.5), and mixed with the substrate solution resulting in a final OD₅₇₈ of 10 and HA final concentration of 0.11 mg/mL. After an incubation time of 5 min the cells were removed by a centrifugation step at 4 °C. 25 µL of the supernatant was applied to the wells of a 96-microplate and mixed with “stains-all” solution following deionized water. A decrease in absorbance indicates activity. The absorbance was measured directly at 650 nm. Data are shown as mean values. (n = 3, error bars ± SD, *** p < 0.05, unpaired t test).

Figure 5

Activity measurement of Hyal-1 displayed on E. coli F470 at different pH values. E. coli F470 displaying Hyal-1 was suspended in formate buffer, pH ranges 3–5 and mixed with the substrate solution resulting in a final OD₅₇₈ of 10 and HA final concentration of 0.11 mg/mL. After an incubation time of 5 min the cells were removed by a centrifugation step at 4 °C. 25 µL of the supernatant was applied to the wells of a 96-microplate and mixed with “stains-all” solution following deionized water. The absorbance was immediately measured at 650 nm. The highest decrease in absorbance due to the cleavage of HA was set 100% activity. Mean values are given. There were no significant differences in activity at pH 3 and 3.5 as well as in activity at pH 4.5 and 5 (n = 3, error bars ± SD, *** p < 0.05, unpaired t test).

Figure 6

Optimal time (A) and cell density (B) for measuring the Hyal-1 activity displayed on E. coli F470; (A) Triangles: E. coli F470 pAK009 displaying Hyal-1; black quadrats: control cells without plasmid. The cell suspensions (final OD₅₇₈ of 10) were incubated with HA (0.11 mg/mL) over a period of 10 min. After separating the cells by centrifugation, the supernatant was mixed with stabilised “stains-all” solution and deionized water. The absorbance was measured at 650 nm; (B) E. coli F470 with pAK009 were resuspended in buffer pH 3.5 and mixed with HA (0.11 mg/mL) to give final optical densities (OD₅₇₈) of 10, 20, and 40. After an incubation time of 5 min at 37 °C, cells were removed and the supernatant was measured as described in (A). The highest decrease in absorbance was set 100% activity. The mean values are shown (n = 3, error bars ± SD, unpaired t test).

Figure 7

IC₅₀ value determination of glycyrrhizic acid. Inhibition of surface displayed Hyal-1 was measured after incubation with glycyrrhizic acid at concentrations ranging from 0 to 1 mM. Calculation of IC₅₀ values was done using GraphPadPrism5 Software. The IC₅₀ value was calculated from the inflectionpoint (50% inhibition of the Hyal-1 activity) of the plot (n = 3, error bars ± SD).

Figure 8

Compounds tested on inhibition of Hyal-1 of displayed on the surface of E. coli F470. The abbreviations Fuc (fucose), Gal (galactose), GlcA (glucuronic acid), Glc (glucose), Qui (chinovose), Rha (Rhamnose), Xyl (xylose) stand for different sugar residues.

Figure 9

IC₅₀ values of different plant extracts on surface displayed Hyal-1. (A) Equiseti herba; (B) Malvae sylvestris flos; (C) Ononidis radix. Activity of surface displayed Hyal-1 was measured after incubation with extract of each plant at concentrations ranging from 0 to 10 mg/mL. The IC₅₀ value was calculated from the inflection point (50% inhibition of the Hyal-1 activity) of the plot. The different shape of the curves probably results from the heterogeneous composition of the extracts and from the different solubility of the constituents in the extracts (n = 3, error bars ± SD).