Supplemental Data. 1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

26 

EXPERIMENTAL PROCEDURES

Inactivation of porS of P. gingivalis W50 - Chromosomal DNA was isolated from P. gingivalis W50

cultures grown for 24h using a PuregeneTM DNA isolation reagent (Flowgen) and all plasmids were

purified using ion-exchange chromatography (Qiagen). Restriction and modifying enzymes were

purchased from New England BioLabs. General manipulation of DNA, restriction and mapping of

plasmids and transformation of E. coli were as described elsewhere.

The organization of genes at porR locus is shown in Fig. S1A. TIGR’s nomenclature is used

throughout. A 2,550 bp region encompassing both porR-porS (PG1138-PG1137), was amplified using

oligonucleotide primers: 5’-atatatgcggccgcTCCCTACAACACGCCTATGC-3’ (PorRF1) and 5’-

atatatgcggccgcCTCTTGGGATGTGAGATGTCTG-3’ (PorRR1) [irrelevant sequences are italiced and

NotI recognition site is in bold] (1).The amplicon was cloned into the NotI site of pUC18not and a 2.1kp

erm cassette (2) was then inserted at the EcoRV site of the insert to inactivate porS in vitro (Fig. S1B).

The construct was restricted with NotI prior to electroporation into exponential cells of P. gingivalis W50

to generate clindamycin resistant mutants by allelic exchange. P. gingivalis mutant colonies were then

screened by PCR that showed that the erm cassette had been inserted in the right position.

Complementation of porS - Initially, a 1,843bp amplicon corresponding to the porS ORF and preceded

by an additional 459bp was amplified by PCR using primers RfbEF2

(atatatgcggccgcCAAACATGACATTA-TCGGATGC) and PorRR1; NotI sites are in bold and irrelevant

sequences are in lower case. Following restriction with NotI, the amplicon was cloned into the NotI site

of pUCET1 to inactivate the vector’s erm cassette. The flanking erm cassette was then used to target the

homologous region in porS via electroporation and allelic exchange with selection for the tagged tetQ on

blood agar plates (3). PCR was used to confirm chromosomal integration of the tetQ-tagged amplicon

(Fig. S1C). This method ensured complementation of porS in cis.

Cloning of porS in E. coli - The gene encoding the putative flippase porS was amplified from P.

gingivalis W50 genomic DNA using the primers PorSEcoRIfw

1  

(aagaattcATGACAAAAGAGGAGGGAGGCTCG) and PorSXbaI6xHisrv

(aatctagatta

1 

gtggtggtggtggtggtgGGAAGCACAAGAGTGAAGTCTGAGAC), irrelevant sequences are

italiced, EcoRI and XbaI sites are in bold and the sequence encoding the His6 tag is underlined. The PCR

product and the plasmid pEXT20 were digested with EcoRI and XbaI. Ligation of the restriction products

resulted in the plasmid pMFH3, encoding a PorS with a C-terminal His6 tag. The plasmid was verified by

DNA sequencing. The expression of the plasmid in E. coli DH5α was verified by Western-blot analysis

using αHis6 polyclonal antibody (Rockland Immunochemicals Inc.). 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

PorS flippase activity test in E. coli - First we decided to test if PorS was able to replace other Wzx

translocase involved in the LPS biosynthesis. The experiments were performed in E. coli K-12 O16. The

wild type strain W3110 and the Δwzx mutant strain (CLM17) were transformed with the plasmid pMF19,

which encodes a functional wbbL (gene that encodes a rhamnosyltransferase involved in the O antigen

subunit synthesis), in order to restore the O16 antigen production (4,5). CLM17 cells were also

transformed either with pMFH3 or pEXT20. The LPS from those strains were purified and analyzed by

SDS-PAGE and silver staining (Fig. S2A).

We further examined if PorS is able to complement translocase deficiency in the C. jejuni N-protein

glycosylation system reconstituted in E. coli (6). In our experiments we used the E. coli strain SCM6,

which lacks wzx. This strain was transformed with pIH18 (encoding the protein acceptor AcrA),

pACYCpglKmut (encoding the C. jejuni N-protein glycosylation machinery with a mutation in the

translocase gene pglK) (7) and pMFH3. In the positive control strain, the plasmid pACYCpgl (encoding

the wild type locus) was introduced instead of pACYCpglKmut and pMFH3. And in the negative control

strain, the plasmid pMFH3 was replaced by the empty vector pEXT20. AcrA glycosylation status was

detected by Western blot developed with αAcrA polyclonal antibody (SACRI Antibody Services,

University of Calgary, Canada) (Fig. S2B) and α-C. jejuni glycan-specific antiserum (R12), which

specifically recognizes glycosylated AcrA (Fig. S2C).

2  

LPS analysis - Cells corresponding to 3 or 6 OD600 units of an overnight culture of E. coli or P.

gingivalis, respectively, were resuspended in 150μl lysis buffer (2% SDS, 4% β-mercaptoethanol, 10%

glycerol, 1M Tris HCl pH 6.8), heated at 100°C for 10 min, and incubated with 2μl Proteinase K (20mg

ml-1, Roche Applied Science) at 60°C for 2h. Then, 150μl 95% phenol was added, incubated at 70°C for

15 min and cooled on ice for 10 min. After spinning at 18,000 g for 10min, the aqueous phase was

transferred to another tube, the LPS was precipitated by the addition of 2.5 vol. Ethanol (8). The

precipitated LPS was resuspended in 50μl H2O.

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

23 

24 

25 

Cell Immobilization for atomic force microscopy (AFM) Measurements - Immobilization of bacterial

cells onto a suitable support while avoiding denaturation is critical for successful (AFM) imaging in

liquid environments. The cells of P. gingivalis W50 were strongly bound to the surface of glass slides

coated with 3-aminopropyltrimethoxysilane (Genorama, Asper Biotech, Tartu, Estonia). A droplet of

concentrated bacterial suspension (5−10 μL) was placed onto a silanized glass slide. After 60 min of

settling at room temperature, the bacteria-coated glass was rinsed to remove loosely attached cells and

mounted on the AFM stage. Slides were kept hydrated prior to AFM analysis by soaking them in 0.1 M

phosphate buffer solution.

Imaging of Bacterial Cells with AFM - The AFM imaging was performed using a Molecular Force

Probe 3D (MFP 3D) from Asylum Research (Santa Barbara, CA) controlled with IGOR PRO software

(Wavemetrics, Portland, OR). All the AFM images were acquired in tapping mode to avoid the surface

damage that usually accompanies contact-mode imaging of soft samples. In the tapping mode of

operation, the AFM cantilever is vibrated near its resonance frequency and scanned over the surface of

interest using intermittent contact between the tip and the sample. This mode of operation reduces friction

forces and hence decreases the accumulation of cellular components on the tip when imaging soft

biological samples such as bacterial cells.

The scan rate ranged from 0.5 to 1Hz and the tip velocity was maintained between 25 and 50μm s-1.

Typically, we began by scanning a 20μm × 20μm area that would contain several bacterial cells.

3  

Gradually, the image size was reduced to isolate individual cells (2μm × 2μm). All the AFM images were

acquired using Olympus gold-coated cantilevers (Bio-levers, Asylum Research, Santa Barbara, CA) with

a nominal tip radius of 30−40nm and nominal spring constants of 27−50 pN/nm determined using the

Cleveland thermal noise method.

1 

2 

3 

4 

5 

6 

7 

8 

9 

To better resolve small surface features and borders, amplitude images were recorded in addition to

the more common topographic images. The amplitude images provide a better resolution of the finer

microstructural details on the bacterial cell surfaces, accentuating edges, and therefore tracking bacterial

roughness at the expense of the height information. Every set of AFM experiments was conducted with

new tips.

4  

REFERENCES 1 

2 3 4 5 6 7 8 9 

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 

1.  Aduse‐Opoku,  J.,  Slaney,  J.  M.,  Hashim,  A.,  Gallagher,  A.,  Gallagher,  R.  P.,  Rangarajan,  M., 

Boutaga, K., Laine, M. L., Van Winkelhoff, A. J., and Curtis, M. A. (2006)  Infect  Immun 74, 449‐460 

2.  Fletcher, H. M., Schenkein, H. A., Morgan, R. M., Bailey, K. A., Berry, C. R., and Macrina, F. L. (1995) Infect Immun 63, 1521‐1528 

3.  Rangarajan, M., Aduse‐Opoku, J., Paramonov, N., Hashim, A., Bostanci, N., Fraser, O. P., Tarelli, E., and Curtis, M. A. (2008) J Bacteriol 190, 2920‐2932 

4.  Feldman, M. F., Marolda, C. L., Monteiro, M. A., Perry, M. B., Parodi, A. J., and Valvano, M. A. (1999) J Biol Chem 274, 35129‐35138 

5.  Marolda, C. L., Vicarioli, J., and Valvano, M. A. (2004) Microbiology 150, 4095‐4105 6.  Wacker, M.,  Linton, D., Hitchen, P. G., Nita‐Lazar, M., Haslam,  S. M., North,  S.  J., Panico, M., 

Morris, H. R., Dell, A., Wren, B. W., and Aebi, M. (2002) Science 298, 1790‐1793 7.  Alaimo, C., Catrein,  I., Morf, L., Marolda, C. L., Callewaert, N., Valvano, M. A., Feldman, M. F., 

and Aebi, M. (2006) Embo J 25, 967‐976 8.  Marolda, C. L., Lahiry, P., Vines, E., Saldias, S., and Valvano, M. A. (2006) Methods Mol Biol 347, 

237‐252 9.  Dykxhoorn, D. M., St Pierre, R., and Linn, T. (1996) Gene 177, 133‐136 10.  Hug, I., Couturier, M. R., Rooker, M. M., Taylor, D. E., Stein, M., and Feldman, M. F. (2010) PLoS 

Pathog 6, e1000819 11.  Linton, D., Dorrell, N., Hitchen, P. G., Amber, S., Karlyshev, A. V., Morris, H. R., Dell, A., Valvano, 

M. A., Aebi, M., and Wren, B. W. (2005) Mol Microbiol 55, 1695‐1703  

5  

Supporting Figure Legends. 1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 

20 

21 

22 

Fig. S1. Construction of porS mutant and complemented strains. (A) Genetic organization of the

wild type porR locus in P. gingivalis chromosome. (B) A region encompassing porR-porS was

amplified using the oligonucleotide primers PorRF1 and PoRR.A erm cassette was then inserted

at the EcoRV site of the insert to inactivate porS in vitro. porS- was generated by insertion of this

construct by allelic exchange into wild type P. gingivalis. (C) The porS ORF was amplified

using primers RfbEF2 and PorRR1.The flanking erm cassette was then used to target the

homologous region in porS via electroporation and allelic exchange with selection for the tagged

tetQ on blood agar plates (porS+).

Fig. S2. Activity of the PorS was examined with complementation assays in flippase deficient E.

coli strains. (A) Purified LPS of E. coli strains containing a functional flippase wzx (wzx+), a wzx

mutation (Δwzx) or Δwzx complemented by porS (Δwzx + porS), was visualized by silver

staining. The typical smooth LPS ladder only was observed in the presence of a functional

translocase. (B, C) Glycosylation of the acceptor protein AcrA in the N-glycosylation pathway of

C. jejuni reconstituted in E. coli was analyzed by Western blot with α-AcrA antibody (B) and

αC. jejuni glycan (C). Glycosylated AcrA was detected in the presence of a functional C. jejuni

translocase (pglK+) and pglK mutation complemented by porS (pglK- + porS).On the other hand,

unglycosylated AcrA was detected in the absence of translocases (pglK-).

Fig. S3. Analysis of OM and OMV. (A, B) SDS-PAGE of purified OM (A) and OMV (B). The

numbers next to the protein bands represent the Sample ID. These protein bands were excised

6  

from the gels. The protein identities were determined by MS/MS fingerprinting and the results

are summarized in Table S2. Lane 1: Wild type, 2: porS-, 3: porS+ and 4: waaL-.

1 

2 

3 

4 

5 

6 

7 

8 

9 

10 

Fig. S4. . RagA/B are not gingipain substrates. (A) Wild type OM was purified in the presence

(+) or absence (-) of protease inhibitors and analyzed by silver staining. RagA/B were detected

even after incubating a 37°C for 2h the OM in the absence of protease inhibitors (t2h), whereas

other proteins were degraded during this procedure. These proteins were identified by MS/MS

fingerprinting as PG1414 (*) and PG0694and PG0695 (**). (B) Western blot analysis showing

presence at comparable levels of RagB in the OM fractions purified in the absence of protease

inhibitors. Prot. Inh.: protease inhibitor.

7  

SUPPORTING TABLES 1 

2  Table S1. Strains and Plasmids used in this study.

Strain or plasmid Description Reference Strains P. gingivalis W50 Wild type ATCC53978 porS- porS::erm This study porS+ erm::tet-porS (in porS-) This study waaL- PG1051:: erm (3) E. coli DH5α F- ϕ80lacZ M15 endA recA hsdR(r-K m-

K) supE thi gyrA relA? Δ(lacZYA-argF)U169

Laboratory stock

W3110 rph-1IN(rrnD-rrnE) Laboratory stock CLM17 W3110, Δwzx (5) SCM6 SØ874,Δwec, ΔwaaL (7) Plasmids pEXT20 Cloning vector, AmpR (9) pMFH3 P. gingivalis porS-his6 in pEXT20 This study pACYCpgl C. jejuni pgl-locus in pACYC184 (6) pIH18 C. jejuni acrA in pEXT21 (10) pMF19 E. coli rhaT in pEXT21 (4) pACYCpglKmut (pACYCwlaB::kan)

pACYCpgl with pglK::kan (11)

3 

8  

Table S2. Proteins identified in purified OM and OMV fractions. The Sample ID corresponds to the

protein bands excised from the SDS-PAGE in Fig. S3.

1 

2 

Sample ID Acc. number Protein Description 1 CAA10226 RagA TonB-dependent outer membrane receptor 2 CAA10226 RagA TonB-dependent outer membrane receptor 3 NP_905570 PG1414 TonB-linked OM receptor PG47 4 AAA99810

NP_905248 Kgp PG1028

Lysine gingipain Probable OM lipoprotein P61

5 NP_904522 NP_905755

RagB PG1626

TonB-dependent outer membrane binding protein Possible outer membrane-associated protein P58

6 AAA99810 Kgp Lysine gingipain 7 AAA69539 Rgp Arginine gingipain 8 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 9 NP_904968 PG0694 Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) 10 NP_904968

NP_904969 PG0694 PG0695

Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) Immunoreactive 43kDa antigen PG32 (OmpA/Omp41)

11 NP_906020 NP_906019

RplC RplD

50S ribosomal protein L3 50S ribosomal protein L4

12 CAA10226 RagA TonB-dependent outer membrane receptor 13 CAA10226 RagA TonB-dependent outer membrane receptor 14 NP_905570 PG1414 TonB-linked OM receptor PG47 15 AAA99810

NP_905248 Kgp PG1028

Lysine gingipain Probable OM lipoprotein P61

16 NP_904522 NP_905755

RagB PG1626

TonB-dependent outer membrane binding protein Possible outer membrane-associated protein P58

17 AAA99810 Kgp Lysine gingipain 18 AAA69539 Rgp Arginine gingipain 19 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 20 NP_904968 PG0694 Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) 21 NP_906223 PG2173 Outer membrane lipoprotein Omp28 22 NP_904968

NP_904969 PG0694 PG0695

Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) Immunoreactive 43kDa antigen PG32 (OmpA/Omp41)

23 NP_906020 NP_906019

RplC RplD

50S ribosomal protein L3 50S ribosomal protein L4

24 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 25 CAA10226 RagA TonB-dependent outer membrane receptor 26 CAA10226 RagA TonB-dependent outer membrane receptor 27 NP_905570 PG1414 TonB-linked OM receptor PG47 28 AAA99810

NP_905248 Kgp PG1028

Lysine gingipain Probable OM lipoprotein P61

29 NP_904522 NP_905755

RagB PG1626

TonB-dependent outer membrane binding protein Possible outer membrane-associated protein P58

30 AAA99810 Kgp Lysine gingipain 31 AAA69539 Rgp Arginine gingipain 32 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 33 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 34 NP_904968 PG0694 Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) 35 NP_906020

NP_906019 RplC RplD

50S ribosomal protein L3 50S ribosomal protein L4

36 CAA10226 RagA TonB-dependent outer membrane receptor 37 NP_905248 PG1028 Probable OM lipoprotein P61 38 NP_904522 RagB TonB-dependent outer membrane binding protein

9  

10  

NP_905755 PG1626 Possible outer membrane-associated protein P58 39 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 40 NP_904968 PG0694 Immunoreactive 42kDa antigen PG33 (OmpA/Omp40) 41 NP_905858

NP_906223 FbaB PG2173

Fructose-bisphosphate aldolase Outer membrane lipoprotein Omp28

42 NP_906020 NP_906019

RplC RplD

50S ribosomal protein L3 50S ribosomal protein L4

43 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 44 NP_904969 PG0695 Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) 45 NP_904969

NP_906118 PG0695 PG2054

Immunoreactive 43kDa antigen PG32 (OmpA/Omp41) Outer membrane lipoprotein PG3

46 AAA99810 Kgp Lysine gingipain 47 AAA69539 Rgp Arginine gingipain 48 AAA69539

AAA99810 Rgp Kgp

Arginine gingipain Lysine gingipain

49 NP_904383 PG0027 Hypothetical protein PG0027 50 CAA10226 RagA TonB-dependent outer membrane receptor 51 CAA10226 RagA TonB-dependent outer membrane receptor 52 CAA10226 RagA TonB-dependent outer membrane receptor 53 CAA10226 RagA TonB-dependent outer membrane receptor 54 NP_904522

NP_905755 RagB PG1626

TonB-dependent outer membrane binding protein Possible outer membrane-associated protein P58

55 AAA99810 Kgp Lysine gingipain 56 AAA69539 Rgp Arginine gingipain 57 NP_904383 PG0027 Hypothetical protein PG0027 58 AAA99810 Kgp Lysine gingipain 59 AAA69539 Rgp Arginine gingipain 60 NP_904383 PG0027 Hypothetical protein PG0027 61 CAA10226 RagA TonB-dependent outer membrane receptor 62 NP_904522

NP_905755 RagB PG1626

TonB-dependent outer membrane binding protein Possible outer membrane-associated protein P58

63 NP_904383 PG0027 Hypothetical protein PG0027 1 

2 

11

12

13

14