Supplementary

1. Over expression and purification of AFP:

AFP extracts were prepared as described by Hurst et al. (5); however 0.4x Luria Broth (LB) media was used for induction instead of 0.5xLB. Sedimentation of AFP particles was undertaken using a modified procedure of Nguyen et al. (10). Polyethylene glycol 6000 and NaCl were added to the supernatant to a final concentration of 8% (w/v) and 0.5 M, respectively. The mixture was gently stirred overnight on ice. AFP particles were sedimented by centrifugation at 8000 x g for 30 minutes using Eppendorf centrifuge 5810R, Rotor F34, and suspended at 4°C in 16 ml of 20 mM Tris, 20mM MgCl2 (TM) and incubated on a Labnet minilab roller at ambient temperature for 2hrs. The sample was centrifuged at 13000 x g for 10 min, at 4°C, and the clear supernatant was divided into 8 ml aliquots and then centrifuged further at 151139 x g for 2 hours at 4°C in a Beckman coulter Optima TM L-100K ultracentrifuge. After removing most of the supernatant, the pellet was resuspended in the residual 500 ml buffer in the tube, filtered through a 0.2-µm-pore-size filter and applied onto a Sephacryl S-400 HR (Pharmacia Biotech) gel filtration column (1.5 by 46 cm bed volume). The column was washed with eluant buffer (25 mM TBS with 130 mM NaCl, pH 7.5), at a flow rate of 1 ml/min at room temperature. Independent, combined column fractions I, II and III were concentrated in an Ultracel-50K device (Amicon). The retenate was resuspended in ~20 µl TM buffer with 0.02% NaN3 and assessed by SDS-PAGE (Figure S_1) and electron microscopy.

Figure S_1(A) SDS-PAGE of the column load amb2 induced (negative control) and amb2/pAF6 induced cultures, . Lane1: negative control. Lane 2: induced culture producing AFP. (B) SDS-PAGE analysis of the column eluent (lane 1). M, Bio-Rad Broad range marker, molecular weights (kDa). Arrows indicate the putative AFP bands; * denotes spectinomycin adenyltransferase.

2. Electron microscopy and processing of images of AFP

Minimal-dose images of AFP specimens were recorded using a Tecnai 12 electron microscope operated at an accelerating voltage of 120 kV at a nominal magnification of 52,000. Micrographs showing minimal drift and astigmatism were digitized using a NIKON Cool Scan 9000 corresponding to a pixel size of 2.44 Å on the specimen. Corrections for phase-flipping due to the contrast transfer function at the defocus used for cryo images (defocus range 1.5–2.7 mm) were applied using ‘bshow’ and ‘bctf’ in the BSOFT package (4). Images of an AFP particle and the regularly-arranged section of the sheath were “boxed” using the program “ helixboxer” of the EMAN suite (7). An averaged power spectrum for the latter was generated using ‘bctf’ of BSOFT software package (4) and the contributing image boxes were next sliced into 1658 overlapping segments of dimensions 150x150 pixels. The overlapping segments were binned 2x2 to reduce computational processing that resulted in the reduction in the box size of the overlapping segments to75 x75 pixels with the value of each pixel equals to 4.88 Å. 3D reconstruction was initiated based on the helical parameters (see below) using the iterative helical real-space reconstruction (IHRSR) algorithm (2).

From the power spectrum (in Figure 1D of the main text) we measured the following:

The axial Rise (Dx) = 78.5 Å

Helical Pitch = 118 Å and calculated Angular Rise (Dy) = 240°

Initially, reconstruction of the AFP sheath was started assuming that the whole assembly is a single continuous helical strand defined by the translation and twist (Dx, Dy) of an asymmetric unit. In IHRSR, in each iteration, a density map is initially generated (using back projection technique) with no helical symmetry but this symmetry is subsequently applied and the map refined (2). This map then serves as the reference for the next iteration. Examination of the density map, after a few iterations, revealed the in-plane 4-fold point group symmetry (C4) (Figure S_2a). This symmetry element was then imposed in subsequent reconstructions. A B

Figure S_2a. 15Å slabs excised from the density maps generated before (A) and after (B) imposing the in-plane C4 symmetry. Bar: 100Å.

In order to validate the deduced C4 symmetry, 8 individual density maps with point-group symmetries ranging from C2 to C9 but with the helical parameters (Dx, Dy) kept fixed, were calculated. A total of 720 projections, at a step size of 4°, were generated from each of these 8 density maps, which were then used as references to classify the raw image segments. This process was carried out using the algorithm APNQ of the SPIDER software package (3). The result from the classification is shown as a histogram plot in Figure S_2b, which clearly indicates that majority of the segments belong to the point group symmetry of C4. Segments belonging to other symmetry groups likely result from heterogeneity in the pitch or in other helical parameters (Dx, Dy). Such heterogeneity in biopolymers is common and has been well studied (8, 14). We also deduced the same C4 symmetry when images of negatively stained specimens were subject to analysis as above (data not shown).

Figure S_2b. Classification of the segments of cryo images of the AFP sheath into various point group symmetries. The bins corresponding to different point group symmetries are indicated on the X-axis while the percentage of the sheath segments in each of these bins is shown on the Y-axis.

The 3-fold helical symmetry of the AFP sheath: The angular rise of 240° corresponds to a 3-fold helix wherein 3 copies of the repeating motif in 2 turns of the helix constitute one helix repeat. This is represented as a so-called 32 helix. The 3-fold helical assembly is illustrated in Figure S_2c by following the spatial locations of an arbitrarily chosen, marked region of density along the helix. A semi-transparent blue sphere identifying one of the elevated regions of the bottom “stack” of the final density map of the AFP sheath marked ‘0’ is subjected to the helix operation using “bhelix” (4) with the helical parameters (Dx, Dy). The resulting positions indicated by the next 3 spheres ‘1, 2, 3’, marked by black arrows, are shown in a side view (left) and in a view down the helical axis (right) demonstrating the 3-fold helix symmetry. The total angular translation from the position “0” to “3” is 720° (or 2 turns).

Figure S_2c Orthogonal views of the density map with the semi-transparent blue spheres positioned at the elevated regions displayed at ~1.5s. Bar: 200Å

3. Heterogenity in the helical pitch of the AFP sheath

Heterogeneity created by the variation in the pitch was examined by assigning a range of pitch values, 90Å to 170Å in steps of 10Å to generate 9 reconstructions followed by generation of reference projections for each of the different reconstructions using APNQ of SPIDER (3). Figure S_3 is a plot of the result of such a classification.

Figure S_3. Classification of a total of 1658 overlapping cryo segments, of the sheath of AFP particles into various pitch values. The bins corresponding to different pitch value are indicated on the X-axis while the percentage of the AFP segments in each of these bins is shown on the Y-axis.

4. Reconstruction of helical sheath using negatively stained images of AFP particles

The reconstruction process employed was exactly the same as described above for the cryo images. Similar classification results used for determining the point group symmetry and the pitch heterogeneity were obtained (data not shown). Projections calculated using the final density map for the negatively stained (8968 overlapping segments) and frozen-hydrated specimens (694 overlapping segments) are shown together with the global average of the AFP particle (Figure S_4) for comparison. This figure demonstrates the agreement in the overall, reconstructed structural features of the sheath generated using the two, specimen preparation methods.

Figure S_4. Surface projections of density maps of the helical sheath obtained from the global averaged image (A) of AFP particle, from the datasets of negatively stained (B) and frozen hydrated (C) AFP sheath sections. (a), (b) and (c) are blow-ups of indicated small segments from (A), (B) and (C) respectively. The characteristic inverted “V” shaped feature, marked by black arrows, appears narrower in (C, c) in comparison to (A, a and B, b). This is likely due to flattening in heavy-metal stain embedded AFP specimens and air-drying of the sample after staining. In less-perturbed frozen-hydrated preparations, ridges of protein density inclined by about 40° to the helix (C, c) are prominent features. Vertical Bar: 200 Å.

5. Determination of the resolution of the density map of the AFP sheath

3D image reconstructions, using 1658 cryo and 26008 negatively stained segments of the AFP sheath, were carried out separately using a featureless solid cylinder as the starting reference map. The final density map computed using the negatively stained segments of the AFP sheath was then used as a starting model for another round of reconstruction using the 1658 cryo segments. The two final density maps computed from the cryo segments using the two separate starting models were used to estimate the resolution using the algorithm ‘bresolve’ (4). Figure S_5 is the plot of the Fourier shell correlation (FSC) values as a function of resolution. At a resolution of ~21.5Å where the FSC reaches 0.5 is deemed as the resolution of the final density map (12).

Figure S_5 Fourier shell correlation curve calculated for reconstruction carried out using the cryo-image data.

6. STEM mass analysis

Freshly prepared AFP samples were freeze-dried and were prepared for STEM analysis by following the standard protocol of the Brookhaven STEM facility (www.biology.bnl.gov/stem/stem.html).

Analysis of STEM Images— Digital dark-field micrographs (512 × 512 pixels) of freeze-dried specimens were recorded at raster steps of 1.0 or 2.0 nm per pixel. The images were analyzed using the PCMass (Brookhaven STEM resource (# 30)) program. Boxes of 150Å length and appropriate width (220Å for AFP and tobacco mosaic virus particles) were used for the measurements. The resulting data were normalized using the known mass-per-unit-length of tobacco mosaic virus (13.1 kDa/Å).

Histograms were calculated using a 0.3 kDa/Å bin. A Gaussian distribution was then fitted to the main peak using the graphical software package ORIGIN6. The result is shown in Figure 3.

The sheath helix comprises of 3 motifs in 2 turns i.e. in 119x2 = 238Å (where 119Å is the pitch calculated from the refined final density map). Total mass for 238Å is 9.8x238 = 2332 kDa. Mass/length of the inner tube is ~2.5kDa/Å. So excluding the contribution of inner tube, mass for the protein comprising the outer component of the sheath is 2332-2.5x238 = 1737 kDa for 3 motifs. For a 4-fold symmetric motif, the mass contribution per subunit = (1737/3)/4 = 145kDa.

7. Sequence similarity analysis of putative sheath proteins Afp2, Afp3 and Afp4 with other known sheath proteins of phage-like particles.

Analysis of the amino acid alignment of Afp2 (38kDa), Afp3 (48kDa) and Afp4 (45kDa) with the sheath proteins of the R-type pyocins and the Enterobacteria phage T4, identified good amino acid alignment at amino terminus and significant amino acid similarity at the carboxyl terminus, which encompasses the F1 tail sheath domain (Figure S_7a) (1, 9). This data in conjunction with the similar length of the aligned proteins may reflect the visual structural similarity of the Afp to the R-type pyocins and the contractile sheath of the Enterobacteria phage T4. It was also found that the Afp proteins Afp1 and Afp6 share 42 % amino acid protein similarity and align to the similarly sized gp19 protein involved in the inner tube formation of bacteriophage T4 tail (1, 6; Figure S_7b).


Afp2 : MTVTTTYPGVYLSEDAVSSFSV-NSAATAVPLFAYD------SENTNTIN------: 43
Afp3 : MATVTSVPGVYIEEDASPAMSV-SASATAVPLFVARFTPLKPELAGVITRIGSWLDYTIL------: 59
Afp4 : --MTMVLPGVSYNETLLTQASN-DDPVTMPLFIGYTPP------: 35
PA0662 : ---MSFFHGVTVTNVDIGARTI-ALPASSVIGLCDVFTPG------: 36
P2STM : -MAQDYHHGVRVVEINEGTRPI-TTVSTAIVGMVCTG------: 35
F1 : -MS-DFHHGTKVIEINDGTRVI-STVATAIVGMVWTA------: 34
gp18 : --MTLLSPGIELKETTVQSTVVNNSTGTAALAGKFQWGPAFQIKQVTNEVDLVNTFGQPTAETADYFMSAMNFLQYGNDLRVVRAVDRDTAKNSSPIAGNIDYTISTPGSNYAVG : 113
Afp2 : ------KPIQVF---RNWAEFTVEYPTPLEDAFYTS------: 70
Afp3 : ------FDSNVPSSARVTVSSTAVEPSPEFDALETASSKA------T : 94
Afp4 : ------DTAIPVTVMQPVSVGSLTQANSLF------G : 60
PA0662 : ------AQASAKPNVPVLLTSKKDAAAAFG------I : 61
P2STM : ------DDADASVFPLNKPVLLTDVLTASGKA------G : 62
F1 : ------SDADAETFPLNEPVLITNVQSAIAKA------G : 61
gp18 : DKITVKYVSDDIETEGKITEVDADGKIKKINIPTGKNYAKAKEVGEYPTLGSNWTAEISSSSSGLAAVITLGKIITDSGILLAEIENAEAAMTAVDFQANLKKYGIPGVVALYPG : 228
Afp2 : ------LSLWFMHGGGKCYLVNEANIADAVAQYDDITLIVAAGTDTTTY------TAFTTVVGQGYRIFGLFDGPKEKIAGTAKPDEVMEEYPTSPFG--- : 156
Afp3 : TTYTYQIDDTEVVDPTASVALRLYFQNGGGPCYLYPLEKADDNGPLAALPDLIDEVGEITLLASPDPDETYRTAVYGALAASLDQHKG------YFLLADSVNGDAPSAVGGS-- : 201
Afp4 : QRG------TLAYSLRHFFENGGLQCYVLPLGPGKG-EPAARLQELIAALQTPQMLET------LLADDKTGLVLVPELSELNEVSSTSLSAEGV : 142
PA0662 : GSS------IYLACEAIYNRAQAVIVAVGVETAETPEAQASAVIGGISAAGER------TGLQALLDGKSRFNAQPRLLVAPGH : 133
P2STM : ESG------TLARSLDAIADQAKPVTVVVRVAQG---ETEAETTSNIIGGVTSD------GKKTGMKALLSAQSQLGVKPRILGVPGH : 135
F1 : KKG------TLSASLQAIADQSKPVTVVVRVAEGTGDDAEAQTTSNIIGG-TDEN------GKYTGIKALLTAEAVTGVKPRILGVPGL : 137
gp18 : ELGDKIEIEIVSKADYAKGASALLPIYPGGGTRASTAKAVFGYGPQTDSQYAIIVRRNDAIVQSVVLSTKRGEKDIYDSNIYIDDFFAKGGSEYIFATAQNWPEGFSGILTLSGG : 343
Afp2 : ----AVFYP------: 161
Afp3 : -AQVAVYYPNVEVPHTRKLDD------AEVAIDGYLDDEGKAV------TTLA-ALRVVN------TEFAGEIAQSL : 258
Afp4 : DAAEVDADALWYQGWQVLLTL------CRQAPQRFALLELPEDPASAVTLTQQSFSADQCQRGAA------WWPRLETSYQD : 212
PA0662 : SAQQAVATAMDGLAEKLRAIA------ILDGPNS-TDEAAVAY------AKNFGSKR------LFMVDPGVQVW : 188
P2STM : -DTQAVATELLGVAQSLRGFA------YLAANGCKTVEEAIAY------RENF-SQREG------MLIWPDFINF : 190
F1 : -DTQEVATALASVCISLRAFG------YVSAWGCKTISEAMAY------RENF-SQREL------MVIWPDFLAW : 192
gp18 : LSSNAEVTAGDLMEAWDFFADRESVDVQLFIAGSCAGESLETASTVQKHVVSIGDARQDCLVLCSPPRETVVGIPVTRAVDNLVNWRTAAGSYTDNNFNISSTYAAIDGNHKYQY : 458
Afp2 : WGTLASGAAVPPSAIAAASITQTDRTRGVWKAPANQA---VNGVTPAFAVSDDFQGKYNQGKALN------MIRTFSGQGTVVWGARTLED--SDNWRYIPVRRLFNAVERD : 262
Afp3 : SGDLSAPLSLPPSALIAGVYGKTDGERGVWKAPANVV---LNGVSDVSVRVTNEQQAELNPKGIN------VIRHFSDRGLVVWGSRTQKD--DDDWRYIPVRRLFDAAERD : 359
Afp4 : ESSAPVVLSPLPAVAAAIQRSAHDN--GVWKAPAN------IALAKTRRPTQSILTSQALLDNQGV-SCNLIRSFVGKGVRLWGCRTLLNEENTAWRYIQIRLLVSSVEHY : 314
PA0662 : DSATNAARNAPASAYAAGLFAWTDAEYGFWSSPSNKE---IKGVTGTSRPVEFLDGDETCRANLLNNA----NIATIIRDD-GYRLWGNRT--L--SSDSKWAFVTRVRTMDLVM : 291
P2STM : DTVLKADATAYASARALGLRAKIDEQIGWHKTLSNVG---VNGVTGISADVFWDLQDPATDAGLLNKN----DVTTLIRKD-GFRFWGSR--CL--SDDPLFAFENYTRTAQVLA : 293
F1 : DTTANATATAYATARALGLRAYIDQTIGWHKTLSNVG---VQGVTGISASVFWDLQASGTDADLLNEA----GVTTLVRKD-GFRFWGNRT-C---SDDPLFLFENYTRTAQVLA : 295
gp18 : DKYNDVNRWVPLAADIAGLCARTDNVSQTWMSPAGYNRGQILNVIKLAIETRQAQRDRLYQEAINPVTGTG------GD---GYVLYGDKTATS-VPSPFDRINVRRLFNMLKTN : 563
Afp2 : IQKSLNKLVFEPNSQPTWQRVKAAVDSYLHSLWQQGALAGNTPADAWFVQVGKDLTMTQEEI-NQ-----G-KMIIKIGLAAVRPAEFIILQFSQDIAQ------: 354
Afp3 : IKKALQPMVFEPNSQLTWKRVQTAIDNYLYRLWQQGALAGNKAEEAYFVRVGKGITMTQDEI-NQ-----G-KMIIQVGMAAVRPAEFIILKFTQDMSQ------: 451
Afp4 : LSKLARAYLFEPNTAPTWMKLKGQVWTWLRQQWLAGAFFGTVEDEAFSLSIGLDETMTEDDIRHG------KMILQVRLALLAPAEFIAISLTLDLRDGTASAQTGGQS : 417
PA0662 : DAILAGHKWA------VDRGITKTYVKDVTEGLRAFMRDLKNQ--GAVINFEVYADPDLNSASQLAQGKVYWNIRFTDVPPAENPNFRVEVTDQWLTEVLDVA--- : 386
P2STM : DTMAEAHMWA------VDGVLNPSLARDIIEGLRAKMRSL-VNQGYLIGGDCWLD-ESVNDKDTLKAGKLTIDYDYTPVPPLENLMLRQRITDRYLVDFASRVAA- : 390
F1 : DTMAEAHMWA------VDKPITASLIRDIVDGINAKFRELKSN-GYIVDGECWFDEES-NDKETLKAGKLYIDYDYTPVPPLESLTLRQRITDKYLVNLAESVNS- : 392
gp18 : IGRSSKYRLFELN------N-AFTRSSFRTETAQYLQGNKALGGIYEYRVVCDTTNNTPSVIDRNEFVATFYIQPARSINYITLNFVATATGADFDELTGLAG----- : 659


Figure S_7a. Amino acid alignment of the predicted AFP sheath proteins Afp2 (YP_026138) Afp3 (YP_026139) and Afp4 (YP_026140) to the R-type pyocin sheath protein PA06620 (NP_249313; 11); the tail sheath gene protein FI of phage P2 STM27, (NP_461629) (9); tail sheath protein FI Enterobacteria phage 186 (AAC34166; 15) and the contractile sheath protein gp18 protein of the Enterobacteria phage T4 (NP_049780) (11). The amino terminus of Afp2 and Afp3 are similar and the carboxyl terminus of Afp2, Afp3 and Afp4. Similar amino acids are shaded.

Afp1 : MAI------TADDIAVQYPIPTYRFIVTLGDEQVPFTSASGLDINFDTIEYRDGTGNWFKMPGQRQAPNIT : 65
Afp6 : MSTP------AVSHRFLVNFLFNNIPNPFDIAFQRISGLSRTLEVSQHREGGENVRNLWLAEQVNHG : 61
gp19 : MFVDDVTRAFESGDFARPNLFQVEISYLGQNFTFQCKATALPAGIVEKIPVGFMNRKINVAGDRTFDDWTVTVMNDEAHDARQKFVDWQSIAAGQGNEIT : 100
Afp1 : LSKGVFPGKNAMYEWINAIQLNQVEKKDIMISLTNEAGTEVLVSWNVSNAFPTSLTSPSFDATSN-EIAVQQITLMADRVTIQTA--- : 149
Afp6 : SLVLERGVMNASPLTLQFDRVLRRESTQWANVVIMLLNELSLPVTTWTLSHALPVRWQMGDLDAGSNQVLINTLELRYQDMRMLGVKL : 149
gp19 : GGKPAEY------KKSAIVRQYARDAKTVTKEIEIKGLWPTNVGELQLDWDSNNEIQTFEVTLALDYWE------: 163