D. radiodurans strain R1 (ATCC 13939) was grown in tryptone/glucose/yeast extract broth (TGY) (Murray, 1992) for 24 h at 30°C, with shaking at 250 rpm. Cells were harvested by centrifugation of 1 l cultures at 5000 × g for 10 min at 4°C and resuspended in 50 mM Na Phosphate pH 7.8 (Buffer A).

Whole cell membrane fractions were purified at 4°C according to Farci et al. (2014). After centrifugation the cells were resuspended in Buffer A, treated with DNase and disrupted using a French Pressure Cell. Unlysed cells were removed by low speed centrifugation (4°C, 2 × 2000 × g for 10 min). The final supernatant was centrifuged again (4°C, 48,000 × g for 10 min) and the pink pellet resuspended in 10 ml of Buffer A. A second step of lysis was performed by using a French Pressure Cell followed by centrifugation and resuspension (4°C, 48,000 × g for 10 min). To remove surface polysaccharides the membrane suspension was incubated under agitation (800 rpm) with 100 μg/ml lysozyme for 8 h at 30°C. The membrane suspension was then centrifuged (4°C, 48,000 × g for 10 min) in order to obtain the protein DR_2577 in solution.

The protein sample obtained after the solubilization step was loaded on to a 20 ml size exclusion chromatography column (Superose 6 10/300GL, GE Healthcare) previously equilibrated in Buffer B [50 mM Na Phosphate pH 7.4, 0.06% (w/v) β-DDM] at a flow rate of 0.3 mL/min. The molecular weight of the DR_2577 complex resolved by the size exclusion chromatography was estimated by plotting the elution volume vs. the logarithm of the molecular weight of the standard proteins (Gel Filtration Standard, Biorad) using a polynomial curve fit (second-order polynomial best-fit).

For denaturing Sodium Dodecyl Sulphate-Polyacrylamide Gel Electrophoresis (SDS-PAGE), 5 or 10% (w/v) separating polyacrylamide/urea gels with 4% (w/v) stacking gels were used (Schägger and Von Jagow, 1987). Monomeric samples were resolved by denaturing with Rotiload (Roth) and boiling for 10' or alternatively by treating them with 6 mM tris-2-carboxy-ethylphosphine at room temperature before loading. Dimeric samples were resolved by denaturing with Rotiload (Roth) but avoiding to boil or to treat them with tris-2-carboxy-ethylphosphine. After the electrophoretic separation the gels were stained with Coomassie Brilliant Blue G250. Blue Native-Polyacrylamide Gel Electrophoresis (BN-PAGE) was carried out using 3–12% (w/v) continuous gradient gels, according to Schägger and Von Jagow (1991). Pink envelopes were mixed with 0.25 volumes of Coomassie Blue Solution 5%, (v/v) Serva Blue G, 750 mM aminocaproic acid and 35% (w/v) sucrose. Electrophoresis was carried out at 205 V for 5 h at 4°C. The molecular weight of the native DR_2577 complex resolved by the BN-PAGE or of the denatured DR_2577 samples resolved by SDS-PAGE was estimated by plotting the retardation factor values (Rf, length of the band migration/length of the dye front) vs. the logarithm of the molecular weight of the molecular marker (NativeMARK, Invitrogen for the BN-PAGE and the Prestained Standard high range, Biorad for the SDS-PAGES) using a polynomial curve fit (second-order polynomial best-fit), according to manufacturer's instructions.

In order to characterize the protein DR_2577, the whole cell membrane fraction was purified and its quality checked by electron microscopy as described in Farci et al. (2014). The intact cell membrane fragments obtained were subsequently subjected to several cycles of disruption and centrifugation. By this procedure, starting from a cell membrane fraction already enriched in DR_2577 (Farci et al., 2014) (Supplementary Figure 1), it was possible to obtain a selective delivery of DR_2577 from the bacterial cell wall fragments to the soluble fraction resulting in an almost pure sample (Supplementary Figure 1). A subsequent step of Size Exclusion Chromatography (SEC, Figure 1) was able to increase further the level of purity as confirmed by SDS-PAGE (Figure 2). After purification, the identity of the isolated DR_2577 was confirmed by Mass spectrometry (data not shown).

SDS-PAGEs on the protein samples, treated according to the standard procedures of denaturation, showed two very near bands that migrated with a double apparent molecular weight with respect to the expected DR_2577 mass of 124 kDa, indicating that the protein may occur as stable dimers (Figure 2, Table 1). Complete dissociation to monomers was reached only under harsh conditions of denaturation such as long boiling or strong treatment with tris-2-carboxy-ethylphosphine. In these conditions the protein samples when resolved by SDS-PAGE presented the expected mass of 124 kDa (Figure 2).

SDS-PAGEs of samples not treated by a strong denaturation shows an exclusive presence of dimers, suggesting that DR_2577 might be prone to form specific intermolecular disulfide bonds. The DR_2577 sequence contains only two cysteine residues localized in the amino acid residues 896 and 929, respectively. In order to clarify the possible bases of the presence of dimers we have performed a basic analysis on the protein sequence, using the software DiANNA (Ferre and Clote, 2005a,b, 2006) for disulfide bonds prediction (http://clavius.bc.edu/~clotelab/DiANNA/), and found that only the residue 929 is likely to be a semi-cystine (and hence be prone to form a disulfide bond) so that only this residue may be potentially part of a disulfide bond (Supplementary Table 1). In the light of these facts it is most likely that two monomers could make a bridge by the interaction of the residues 929 from two different monomers entertaining an intermolecular disulfide bond.

DR_2577 showed small retention volumes when separated by SEC, strongly suggesting that under native conditions the protein could occur as a high molecular weight complex (Figure 1). With the aim to have a more detailed description of DR_2577 oligomeric state, we used the retention volumes of the SEC profiles and defined the size of the DR_2577 peak with respect to molecular standards. By this analysis a calculated mass out of the linearity range for this system was obtained (Figure 1; Table 1) not allowing to identify a precise size of the complex but showing that the complex has a mass greater than 700 kDa. Next, we performed a similar analysis by means of BN-PAGEs. Due to the typical auto-assembling properties of S-layer proteins (Pum et al., 2013), we performed these experiments loading about 0.1 μg of protein in order to reduce the tendency to smear that appears when the protein sample is loaded at concentrations greater than 0.5 μg. Using this method the sample was resolved into two isoforms (Figure 3) in which the heaviest band appeared as the most representative. The size of the complexes was estimated for both bands confirming that the protein occurs as a homo-oligomeric complex which can carry masses of ~740 and 770 kDa, respectively (Figure 3; Table 1). According to this finding and considering that a DR_2577 monomer accounts for 124 kDa it is reasonable to conclude that in vivo the DR_2577 complexes occur in form of hexamers (Table 1).

Growing evidence suggests a primary role of the protein DR_2577 in the structural organization of the S-layer of D. radiodurans (Rothfuss et al., 2006; Farci et al., 2014). From early works it was observed how the knockout mutants for this protein typically lack of their S-layer integrity and their ability to resist to extreme conditions (Rothfuss et al., 2006). More recently was reported how the incidence of DR_2577 in the cell wall is far from being secondary (Farci et al., 2014).

In this work was found that DR_2577 can be isolated in purity by several steps starting from extracted S-layers in which DR_2577 is shown to be a dominant component (Supplementary Figure 1).

The protein DR_2577 can be separated into two different bands, either of dimers, when analyzed by SDS-PAGE (Figure 2), or of hexamers, when analyzed by BN-PAGE (Figure 3; Table 1). In both cases the presence of a double band can be explained by the characteristic tendency of S-layer proteins to assemble into ordered structures so that patterns similar to the observed in the PAGE experiments may be ascribed to rate-limited self-association phenomena in which the presence of two bands may be attributed to the strong imbalance between the dominant assembling reaction (expected for these kind of proteins) and its inverse disassembling (Shunong et al., 1991). Such unbalance would also explain the absence of an expected third band related to monomers in SDS-PAGE and to dimers in BN-PAGE. However, it cannot be excluded, even if less likely, that the two bands observed in both PAGE experiments could represent two different isoforms of the same protein. These observations suggest that in vivo the dimeric form should be the structural unit from which homo-oligomeric complexes of higher order are built, providing a structural cohesion to the S-layer. This hypothesis is further supported by the observation that the dimeric structural unit was found to be exceptionally resistant, so that only harsh denaturing conditions such as long boiling or strong treatment with tris-2-carboxy-ethylphosphine were able to induce formation of monomers. These results are also consistent with our preliminary bioinformatic analysis indicating that out of the two residues of cysteine present in the DR_2577 sequence, only one is a strong candidate to entertain a disulfide bond, allowing for the formation of DR_2577 homo-dimers between the same cysteine of two different monomers. Such stability and resistance to denaturation were previously observed also on the SlpA protein from Thermus thermophilus, a homolog of DR_2577, providing a further indication of similarity (Berenguer et al., 1988; Castón et al., 1988). Integration of these results with the observations emerged from SEC and BN-PAGE analyses provided also evidence that DR_2577 is organized into hexamers which accordingly with the description provided above must be constituted by triads of stable dimers. Previously, DR_2577 was shown to be exclusively associated to three complexes having masses of about 860, 916, and 1117 kDa, respectively (Farci et al., 2014). In the present work we have found that the purified DR_2577 assembles into slightly smaller complexes of about 750 kDa (Table 1), suggesting that DR_2577 hexamers in vivo may occur in association with other proteins. According to this hypothesis it is most likely that, as well as in the purified form, DR_2577 present “in vivo” the same oligomeric behavior, basing its assembling properties on a precise hierarchical organization for which stable dimers originate from disulfide bonds between two monomers and hexamers from non-covalent interaction between three stable dimers.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.