Plasmids and bacterial strains

Bacteria and plasmids used in this study are listed in Table S1 [22,23,24,25,26,27,28,29,30,31,32,33,34,35,36].

Plasmid constructions

Plasmids pHJS5 and pUZN10 contained the htrA wild-type (wt) gene from H. pylori strains 26695 and N6, respectively, fused with His6-tag at the C-terminus and E. coli pelB signal sequence at the N-terminus. These plasmids were used as templates for constructing plasmids pUZ20 and pUZN22, in which we exchanged the leucine at position 171 for serine, or serine at position 171 to leucine by site-directed mutagenesis using PrimeSTAR GXL DNA Polymerase (Clontech, Takara). The nucleotide substitutions were verified by Sanger sequencing (Eurofins Genomics, Ebersberg, Germany). The primers for the construction of plasmids pUZ20 (primers P1, P2), and pUZN22 (primers P3, P4) are listed in Table S2.

In vivo H. pylori mutagenesis

HtrA from H. pylori was mutated by gene replacement via double crossing-over recombination. This method was described earlier [13, 37] and resulted in the generation of functional htrA genes. The ΔhtrA deletion mutation containing an aphA-3 cassette (conferring kanamycin resistance) was replaced by mutated htrA from plasmids pUZ20 or pUZN22 along with a chloramphenicol resistance cassette using gene splicing by overlap extension PCR (SOE-PCR). First, we prepared three gel-purified DNA fragments amplified by PCR using primers and templates shown in Fig. S1. Next, a PCR with the two terminal primers P9 and P10 and three DNA fragments produced the spliced PCR (S-PCR) fragment. Finally, the purified S-PCR fragment was used (approx. 700 ng) to transform H. pylori N6∆htrA strain by natural transformation, followed by the selection of strains containing the corresponding htrA mutation at the correct chromosomal location. The introduced nucleotide substitutions were verified by Sanger sequencing (Eurofins Genomics, Ebersberg, Germany). Primers, templates, and PCR fragments are shown in Fig. S1 and summarized in Supplementary Tables S2 and S3. All wt and htrA mutated H. pylori strains used in this work are depicted in Fig. S2.

H. pylori growth conditions

After revival from frozen stocks, the H. pylori wt strains and the N6 derivatives were cultured for two days at 37 °C under microaerobic conditions produced by sachets (CampyGen) in 2.5 L anaerobic jars (Oxoid, Wesel, Germany) on the GC agar (Oxoid, Wesel, Germany) supplemented with 10% donor horse serum (Biowest, Nuaillé, France), protease peptone (Oxoid, Wesel, Germany), vitamin mix (1%), and antibiotics: vancomycin (10 µg/mL), trimethoprim (5 µg/mL), amphotericin (4 µg/mL) and colistin (10 µg/mL), as described [12, 26]. To allow for selection, the media were supplemented additionally with chloramphenicol (8 µg/mL). The cells were suspended in the BHI medium (Oxoid, Germany), and the number of bacteria was evaluated by measurement of the optical density at 600 nm (OD600). To investigate the survival of H. pylori the number of colony-forming units (CFU) was determined by plating serial dilutions of bacterial suspensions as described [12]. The various temperatures (39 °C and 41 °C) and pH (pH5.2) stress conditions were generated as described [12, 31]. All experiments were performed in triplicate.

Bacterial growth curve analysis

H. pylori culturing was basically carried out under conditions that were previously described in the literature [38]. H. pylori grown on plates were resuspended in BHI liquid broth medium with 10% FCS to an OD600 of 0.1, and then sub-cultured in 96-well microtiter plates (clear, flat-bottom, NEST Biotechnology Co.). Plates were incubated at 37 °C, 10% CO2, 5% O2, 500 rpm in a plate reader (Clariostar, BMG Labtech) with an atmospheric control unit (BMG Labtech). OD600 was automatically measured every 30 min until the stationary phase was reached. Growth curves were analyzed and processed using the MARS Data Analysis software 4.20 (BMG Labtech).

Expression and purification of the HtrA proteins

The E. coli strain BL21(DE3) transformed with pHJS5, pUZN10, pUZ20, or pUZN22 (plasmids containing htrA genes) were used to overproduce recombinant HtrA variants with the C-terminal His tag (6x) in the pET vector System (Novagen, San Diego, CA, USA). Bacteria were grown at 37 °C in Luria–Bertani (LB) broth supplemented with kanamycin (50 µg/mL) to OD600 of 0.7; then the expression of htrA was induced by addition of 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG) for 2–3 h (37 °C). After lysis and clearing of the lysates by centrifugation, all HtrA proteins were purified through nickel-affinity chromatography under native conditions as described previously [12].

SDS-PAGE, immunoblotting, and casein zymography

The SDS buffer was mixed with bacterial pellet or recombinant proteins and boiled for 10 min. Samples were separated by SDS-PAGE. The separated proteins were stained by Coomassie Brilliant Blue (Merck, Darmstadt, Germany) or blotted onto polyvinylidene difluoride (PVDF) membrane (Carl Roth, Karlsruhe, Germany), followed by blocking and incubating the blotting membrane with the appropriate antibodies in 5% non-fat dry milk in TBS-T buffer (200 mM Tris pH 7.4, 1.4 M sodium chloride and 1% Tween-20) [39]. For protein detection, the following primary antibodies were incubated with membranes for 1.5 h at room temperature: mouse monoclonal α-His-Tag (Proteintech, Manchester, UK, cat. #66005), rabbit polyclonal α-HtrA [13], rabbit α-UreB antibody (using the conserved peptide HDYTIYGEELK as antigen). The horseradish peroxidase-conjugated secondary antibodies goat α-rabbit and goat α-mouse IgGs (Thermo Fisher Scientific, cat. #31462 and cat. #31446) were used in combination with the ECL ™ Prime Western Blotting kit (GE Healthcare Chicago, IL, USA) to detect bound antibodies. Casein zymography was performed as described [12, 15] and shown in Fig S3. The appropriate amounts of recombinant protein or bacterial lysates of H. pylori were added to SDS- casein buffer without β-mercaptoethanol (final concentration 46 mM Tris pH 6.8, 7.4% glycerol, 2.2% SDS, 0.007% bromophenol blue) and were incubated 20 min at room temperature. The samples were separated into 10% SDS-PAGE gels containing 0.1% casein (Carl Roth, Germany). For protein renaturation after gel electrophoresis, the gels were incubated in renaturation buffer (2.5% Triton-X100) for 60 min at room temperature, with the buffer changed every 30 min, to remove the SDS from the gel and from the protein-SDS complexes. Given that HtrA has chaperone functions, the removal of SDS from the proteins initiated the recovery of the native HtrA structure and restored HtrA activity, as shown before [11, 12, 15, 17, 31, 43, 44]. After protein renaturation, casein cleavage was performed for 16 h at 37 °C. To investigate the effect of SDS concentration on HtrA trimer stability, SDS-casein buffer with different SDS concentrations (5; 2.2; 1; 0.1%) was prepared and conventional SDS-PAGE running buffer was used.

Analysis of HtrA proteolytic activity

The analysis of proteolytic activity of H. pylori HtrA against β-casein has been described [12], and the cleavage profile was recently determined [40]. In brief, 0.24 μM recombinant HtrA was mixed with 10 μM β-casein in 50 mM HEPES pH 6.2, 100 mM NaCl and incubated at 37 °C for 90 min in a final volume of 200 μL. Samples without HtrA were used as controls. The reaction was terminated by the addition of Laemmli lysis buffer (30 mM Tris-HCl, pH 6.8, 5% glycerol, 1.5% sodium dodecyl sulfate (SDS), 0.005% bromophenol blue) and freezing at − 20 °C immediately after boiling. The samples were then resolved in 15% SDS-PAGE and the gels were stained with Coomassie Brilliant Blue.

Analysis of HtrA sequence variations from various bacterial species

HtrA proteins from different bacterial species were examined using sequences obtained from the UniProt database and available full genome sequences. The hydrophobicity of amino acid residues was analysed using the Praline tool of the IBIVU server [41, 42].

Statistics

All data were generated from at least three independent experiments. The error bars in the corresponding figures represent the standard deviation values or the standard error of the mean (SEM). Significant differences were analysed using the Bonferroni test (ns- no significant differences; ***p < 0.001; **p < 0.01). The areas of the casein gels corresponding to the position of the HtrA trimer was examined densitometrically using a 1DScan EX program (Scanalytics Inc., United States).