Supporting Information

Brown et al. 10.1073/pnas.1114476109 SI Materials and Methods and silver enhancement was not required. Electron microscopy Strains and Conditions. A. tumefaciens C58 was grown in Luria– was performed on a JEOL JEM1010 at 60 kV, equipped with a Bertani (LB) broth medium or in AT minimal medium (1) TemCam-F416 (TVIPS) digital camera. supplemented with 0.5% (wt/vol) glucose and 15 mM ammonium sulfate and 0.1 mM acetosyringone at 26 °C. S. meliloti Electron Microscopy. Droplets of exponentially growing cultures fi 1021 was grown in LB medium supplemented with 2.5 mM were deposited onto a piece of para lm and E.M. carbon-for- fl CaCl and 2.5 mM MgSO at 26 °C. Brucella abortus 544 was mvar-coated copper grids (200 mesh) were oated onto the 2 4 droplets for 2 min. Excess liquid was removed with filter paper, grown in 2YT liquid medium at 37 °C and O. anthropi LMG 3331 fl was grown in LB medium at 30 °C. H. denitrificans was grown in and the grids were quickly washed four times oating onto Hyphomicrobium medium 337 containing 0.4% methylamine droplets of double distilled water, transferred to droplets of 1% hydrochloride (2) at 26 °C without shaking. P. hirschii was grown uranyl acetate in water for 2 min, washed once in double distilled on MMB medium (3) at 26 °C. When necessary, kanamycin was water, dried, and subjected to electron microscopy as described above for D-cys labeling. used at 100 μg/mL and 1 mM isopropyl-α-D-thio-galactoside (IPTG) was used as an inducer. TRSE Staining. The amine reactive dye, TRSE, binds to outer membrane proteins and the mobility of these proteins in E. coli is Construction and Imaging of FtsZ-eGFP Expression Strain. To enable severely restricted in the regions comprised of inert peptidoglycan localization studies of FtsZ (Atu2086), a translational fusion of fi FtsZ to eGFP was placed under the control of the lac promoter. (7). The TRSE staining protocol was modi ed to allow time-lapse A. tumefaciens The Atu2086 gene lacking the stop codon was amplified using the microscopy of individual growing cells. were grown to exponential phase and washed three times in 0.1 M NaHCO , FtsZ2-F-GFP (CATATGACGATACAGCTGCAAAAGCCT) and 3 pH 8.3 buffer by centrifugation for 5 min at 5000 × g.Washedcell FtsZ2-R-GFP (CTCGAGGTTGGACTGGCGGCGCAGGAA- pellets were resuspended in 200 μl of buffer and TRSE was added GGC) primers and cloned into the IPTG-inducible expression to a final concentration of 0.1 μg/mL. Cells were incubated in the vector pSRKKm (4), generating pJW164. eGFP was amplified dark for 5–10 min. B. abortus and O. anthropi cell were collected from pJZ383 (5) using primers XhoIgfp (CTCGAGATGAGT- during stationary phase, washed three times with PBS, and then AAAGGAGAAGAACTT) and gfpNheI (GCTAGCTCATTTG- incubated for 15 min with TRSE at a final concentration of 1 μg/mL TATAGTTCATCCATGCC). A fragment digested with XhoI and in the dark at room temperature. The bacteria were then washed NheI and containing eGFP was cloned into pJW164, generating the fi one time with buffer and two times with the appropriate medium. nal construct, pJW164G. pJW164G was introduced into wildtype A. tumefaciens cells were spotted on an agarose pad immediately A. tumefaciens. Before time-lapse microscopy, cells were diluted to after staining and observed using time-lapse microscopy. B. abortus an OD of 0.1 and grown in the presence of kanamycin (100 μg/mL) μ cellsweregrownfor3hinliquidmediumandO. anthropi cells were and 1 mM IPTG for 3 h. Dilute cell suspension (0.8 L) was then grown for 1 h in liquid medium before observation with time-lapse spotted on LB agarose pads containing kanamycin and IPTG and microscopy. imaged every 10 min using the methods for time-lapse microscopy described below. Time-Lapse Microscopy. For A. tumefaciens, LB medium (unless otherwise stated) containing 1% agarose was applied to a 25-mm D-cys Labeling. D-cys labeling was performed as previously de- × × fi μ by 75-mm glass slide to form an agarose pad (22 mm 22 mm scribed (6) with the following modi cations. D-cys (100 g/mL) 0.5 mm) capable of supporting the growth of the bacteria. Ex- was added to exponentially growing cultures (150 mL; OD600 = ponential phase cell culture was diluted to an OD600 of 0.05–0.2 0.1) of A. tumefaciens and the cultures were incubated further for and 0.8 μL was spotted on the agarose pad. The agarose pad was 150 min. At this stage cells (OD600∼0.6) were collected by cen- × covered with a coverslip and sealed using a 1:1:1 mixture of trifugation (5000 g; 5 min; room temperature), resuspended in Vaseline, lanolin, and paraffin. A Nikon Eclipse 90i light mi- an equal volume of prewarmed LB, centrifuged again as before, croscope equipped with a ×100 DIC Plan Apo VC oil objective and resuspended into 9 mL of prewarmed LB. A 3-mL sample was used for DIC microscopy and a Chroma 83700 triple filter (nonchased control) of the cell suspension was immediately cube was used with corresponding excitation and emission filters mixed with 6 mL of 6% (wt/vol) SDS in a boiling water bath under for epifluorescence microscopy. Images were captured every strong magnetic stirring. The remaining cell suspension was dis- 5 min using a Photometrics Cascade 1K cooled charge-coupled- tributed among an appropriate number of cultures with pre- device camera and Metamorph imaging software (Molecular warmed medium, and further incubated for the selected chase Devices). times, normally 45 and 90 min, roughly corresponding to one and For B. abortus and O. anthropi, time-lapse microscopy was two mass doublings. At the end of the chase time, cells were performed by placing cells on a microscope slide that was layered collected by centrifugation, resuspended into 3 mL of prewarmed with a 1% agarose pad containing 2YT medium and LB medium, medium, and immediately mixed with 6 mL of boiling 6% SDS as respectively. Fluorescence was observed at 583 nm. Samples described above. Samples were kept in a boiling water bath with were observed every hour at 37 °C for B. abortus and every 40 stirring for 6 h and then were left overnight at room temperature min at 32 °C for O. anthropi using a Nikon i80 fluorescence with moderate stirring. Further washing, biotinylation, immu- microscope and the NIS software from Nikon with a Orca ER nolabeling and silver enhancement were performed as described Hamamatsu camera. (6) except that a mouse monoclonal anti-biotin antibody (Mo- lecular Probes) and nanogold-conjugated anti-mouse antibody Agrobacterium Attachment to an Arabidopsis Root. Sterilized Ara- (Nanoprobes) were used as primary and secondary antibodies bidopsis thaliana seeds were placed on 1/2 MS salts plates with respectively. D-cys labeling for S. meliloti was completed as de- 1% agar and 1% sucrose and allowed to germinate and grow scribed above except that a 10 nm of Gold-proteinA conjugate until the roots were ∼3 cm long (8). Wild type A. tumefaciens was used in place of nanogold-conjugated anti-mouse antibody cells were grown in LB medium at 26 °C until reaching expo-

Brown et al. www.pnas.org/cgi/content/short/1114476109 1of16 nential phase. TRSE stained cells (200 μL) were washed and (pH 4.95), 15% (vol/vol) methanol. Elution was monitored by resuspended in a solution containing 1 mM calcium chloride and measuring the absorbance at 204 nm. 0.4% sucrose. The bacterial cell culture was spotted into a Petri For the identification of muropeptides, each peak of the HPLC dish and a 10 mm root segment was floated in the bacterial cell profile was collected, vacuum dried, desalted, and subjected to culture in the dark at room temperature for 4 h. The root seg- MALDI-TOF (Autoflex; Bruker Daltonics) to determine the ments were rinsed in 1 mM calcium chloride and 0.4% sucrose molecular mass of the components. The specific sequence of and placed on an agarose pad containing 0.5 μg/mL Alexa Fluor amino acids and amino sugars in each muropeptide was defined 488-conjugated WGA (Invitrogen Molecular Probes), 1 mM by means of electrospray-ion trap MS/MS (LCQ Classic; Thermo- calcium chloride, and 0.4% sucrose. The attachment and growth Finnigan) of the HPLC-purified muropeptides. of A. tumefaciens cells to the root segments was observed using time-lapse microscopy. Analysis of Mother Cell Growth. A. tumefaciens was grown in LB to exponential phase and then introduced into a 0.1 × 1 × 50 mm Determination of Peptidoglycan Composition of A. tumefaciens and rectangular glass capillary (VitroCom). After sufficient time to S. meliloti by HPLC and MS. Peptidoglycan purification was per- allow attachment by motile cells (5–10 min), constant flow of formed by a modification of the boiling SDS extraction method sterile LB at ∼100 μm/s permitted observation of dividing cells (9). Cells were collected from exponentially growing cultures by through multiple divisions while washing away nascent motile centrifugation (6,000 × g; 15 min; 20 °C), suspended into a small daughters. A Nikon 90i microscope with 60× phase-contrast, oil volume of LB, and mixed 1:2 with 6% SDS in a boiling water immersion objective and motorized stage collected images from bath under strong magnetic stirring. The suspensions were kept multiple stage positions at 4-min intervals. A custom ImageJ under those conditions for 4 h, and then were left overnight at plugin tracked the size of individual cells through time using room temperature with stirring. The SDS insoluble material was automatic thresholding and particle detection. recovered by centrifugation (Beckman TL100 desktop ultracen- trifuge, 300,000 × g; 12 min; 18 °C) and cleaned until free of SDS Phylogenetic Tree Construction. Phylogenetic trees were con- by repeated cycles of suspension in distilled water and centrifu- structed using either gyrA or 16S rRNA sequences from repre- gation. Sacculi were then subjected to protease digestion to re- sentative species. gyrA sequences were retrieved from genome move contaminating protein and cleave potential peptidoglycan sequence datasets using the Integrated Microbial Genomes (10) bound lipoproteins. The last pellet of the washing procedure was for the following GenBank accession numbers: NC_002696, suspended into 10 mM Tris·HCl (pH 7.5) and digested for 60 NC_003047, NC_003911, NC_007406, NC_007958, NC_008347, min with 100 μg/mL preactivated pronase-E at 60 °C. Digestion NC_008358, NZ_AALV01000002, NZ_AALY01000002, NZ_ was terminated by adding SDS to 1% (wt/vol) final concentration AAMQ01000001, NZ_AAOT01000029, NC_003304, NC_ and incubating the samples for 45 min in a boiling water bath. 007493, NZ_AAYA01000004, NC_009720, NZ_ABIG01000001, Samples were washed again until free of SDS by three cycles of NC_012982, NZ_ACQR01000016, NC_000913, NZ_ centrifugation and suspension in water as above. The final pellets ADFF01000011, NZ_ADVE01000034, NZ_DS990628, NZ_ were suspended in a small volume (100–400 μl) of 50 mM GG703765, NC_014313, and NC_014375. Sequences were phosphate buffer pH 4.5 and digested overnight with 40 μg/mL aligned using MUSCLE (11). muramidase (Cellosyl) at 37 °C. Digested samples were in- Aligned 16S rRNA sequences were obtained from the Ribo- cubated for 4 min in a boiling water bath, centrifuged in a bench somal Database Project (12) (http://rdp.cme.msu.edu) for the top Eppendorf centrifuge (10,000 × g; 15 min), and the soluble following GenBank accession numbers: X73041, AF338176, fraction was recovered and transferred to glass test tubes and CP000115, L11664, AM158980, U70978, EF117253, Y14308, mixed 1:1 with 0.5 M sodium borate buffer pH 9.0. A few grains X97693, AB095950, FJ560750, GQ221761, Y18946, GQ221766, of granulated NaBH4 were added to each sample to cause re- D14509, GQ221767, AE009348, AF399970, CP002292, Y11552, duction of sugars and avoid anomerization, and after 30 min the X78315, AY424896, U73725, X53853, DQ342322, X78312, pH of the samples was adjusted to pH 3.8, with phosphoric acid. Y11551, U58356, AF098491, Y13155, AJ535710, U20772, Samples were then filtered through Millex-GV, 4-mm-diameter GQ221763, and AP009048. filters (Millipore) and either injected into the HPLC system or RAxML (13) reconstructed the maximum likelihood phylog- kept at −20 °C. eny using the JTT amino acid substitution matrix (gyrA) or GTR For HPLC analysis, a binary pump Waters system was used. nucleotide substitution matrix (16S rRNA) and a four-category The column was a Thermo Scientific3μm ODS-Hypersil 250 × discrete gamma distribution and single invariant category of se- 4.6 mm. Elution conditions were: flow rate, 0.5 mL/min; tem- quence rate variation among sites. Analysis of 100 bootstrapped perature, 35 °C; 7 min isocratic elution in 50 mM pH 4.35 so- datasets determined support for clades occurring in the maxi- dium phosphate followed by a 113 min linear gradient to 75 mM mum likelihood phylogeny.

1. Tempé J, Petit A, Holsters M, Montagu M, Schell J (1977) Thermosensitive step 7. de Pedro MA, Grünfelder CG, Schwarz H (2004) Restricted Mobility of Cell Surface associated with transfer of the Ti plasmid during conjugation: Possible relation to Proteins in the Polar Regions of Escherichia coli. J Bacteriol 186:2594–2602. transformation in crown gall. Proc Natl Acad Sci USA 74:2848–2849. 8. Ramey BE, Matthysse AG, Fuqua C (2004) The FNR-type transcriptional regulator SinR 2. Hirsch P, Conti SF (1964) Biology of Budding Bacteria. I. Enrichment, Isolation and controls maturation of Agrobacterium tumefaciens biofilms. Mol Microbiol 52: Morphology of Hyphomicrobium Spp. Arch Mikrobiol 48:339–357. 1495–1511. 3. Staley JT (1984) Prosthecomicrobium hirschii, a New Species in a Redefined Genus. Int 9. Glauner B, Höltje JV, Schwarz U (1988) The composition of the murein of Escherichia J Syst Bacteriol 34:304–308. coli. J Biol Chem 263:10088–10095. 4. Khan SR, Gaines J, Roop, RM, 2nd, Farrand SK (2008) Broad-host-range expression 10. Markowitz VM, et al. (2006) The integrated microbial genomes (IMG) system. Nucleic vectors with tightly regulated promoters and their use to examine the influence of Acids Res 34(Database issue):D344–D348. TraR and TraM expression on Ti plasmid quorum sensing. Appl Environ Microbiol 74: 11. Edgar RC (2004) MUSCLE: Multiple sequence alignment with high accuracy and high 5053–5062. throughput. Nucleic Acids Res 32:1792–1797. 5. Cormack BP, Valdivia RH, Falkow S (1996) FACS-optimized mutants of the green 12. Cole JR, et al. (2009) The Ribosomal Database Project: Improved alignments and new fluorescent protein (GFP). Gene 173(1 Spec No):33–38. tools for rRNA analysis. Nucleic Acids Res 37(Database issue):D141–D145. 6. de Pedro MA, Young KD, Höltje JV, Schwarz H (2003) Branching of Escherichia coli 13. Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based phylogenetic cells arises from multiple sites of inert peptidoglycan. J Bacteriol 185:1147–1152. analyses with thousands of taxa and mixed models. Bioinformatics 22:2688–2690.

Brown et al. www.pnas.org/cgi/content/short/1114476109 2of16 Fig. S1. Position of the constriction and growth of the mother cell compartment in A. tumefaciens.(A) Position of the constriction relative to the old pole (gray diamonds) and new pole (open triangles) is plotted against the total cell length. (B) Representative images used in the quantiﬁcation of mother cell growth. An attached mother cell is tracked through several rounds of growth and division. Images shown are the frames just before and just after cell separation. Daughter cells are ﬂushed away after cell separation. (C) Each trace illustrates the growth of a single cell over time. Dark horizontal lines indicate the 0 μm position for the cell trace above. Each horizontal line is separated by 1 μm. Sudden decreases in cell length are indicative of cell division.

Brown et al. www.pnas.org/cgi/content/short/1114476109 3of16 A Growth at pole B

0 min 20 min 40 min 60 min 80 min 100 min 120 min

Growth at collar

Fig. S2. Cell growth of A. tumefaciens.(A) Predicted patterns of TRSE or D-cys labeling if growth occurs at the pole (Upper) or at the junction between the mother and daughter cell compartments (Lower). (B) Growth pattern of a TRSE-stained cell and D-cys labeling patterns of individual sacculi indicate that new growth is polar. (C) D-cys-labeled sacculi after a long chase (3.25 doublings) are either unlabeled or heavily labeled with D-cys, indicating that peptidoglycan is inert with little or no turnover.

A 0 min 100 min 200 min 300 min 400 min

B 0 min 40 min 80 min

120 min 160 min 200 min

C 0 min 20 min 40 min 60 min 80 min 100 min

Fig. S3. Unipolar growth is the primary mode of A. tumefaciens growth. (A) Time-lapse microscopy of an individual TRSE-strained A. tumefaciens cell grown under virulence-inducing conditions [AT minimal medium (1) supplemented with 0.5% (wt/vol) glucose and 15 mM ammonium sulfate and 0.1 mM acetosyringone]. (B) Time-lapse microscopy of TRSE-stained A. tumefaciens cells growing in association with an Arabidopsis thaliana root. White arrows track the position of three TRSE-stained cells throughout the panels. (C) Time-lapse microscopy of TRSE-stained A. tumefaciens cells attached to an A. thaliana root. Attachment is shown by the presence of the unipolar polysaccharide stained with ﬂuorescent wheat germ agglutinin (WGA)–Alexa Fluor 488 (green).

Brown et al. www.pnas.org/cgi/content/short/1114476109 4of16 A M4 D44

T444 204 A

Retention time (min)

B M4 D44 204

A T444

Retention time (min)

C D-cys treatment D-cys treatment Untreated No chase 120 min chase

Fig. S4. D-cys is incorporated into muropeptides M4, D44, and T444 (see Fig. S7 for detailed structure) and does not affect cell morphology. (A) HPLC profile of A. tumefaciens peptidoglycan following D-cys labeling. Red arrows indicate the peaks corresponding to the modified M4, D44, and T444 muropeptides. (B) HPLC profile of A. tumefaciens peptidoglycan following D-cys labeling and subsequent hydrogen peroxide treatment. The hydrogen peroxide oxidizes the cysteines to sulfonic acid derivatives, causing the peaks containing D-cys-modified muropeptides to disappear. (C) Morphology of cells after no treatment, D-cys treatment and no chase, and D-cys treatment and 120 min chase.

Brown et al. www.pnas.org/cgi/content/short/1114476109 5of16 Vasilyevaea enhydra Rhodomicrobium vannielii Filomicrobium insigne Hyphomicrobium denitrificans Pedomicrobium australicum Prosthecomicrobium hirschii Ancalomicrobium adetum Xanthobacter viscosus Rhodopseudomonas palustris Bradyrhizobium denitrificans Nitrobacter winogradskyi Methylosinus sporium Prosthecomicrobium sp. Agrobacterium tumefaciens Rhizobium aggregatus Sinorhizobium meliloti Ochrobactrum anthropi Brucella abortus Oceanicola granulosus Octadecabacter arcticus Antarctobacter heliothermus Sagittula stellata Sulfitobacter pontiacus Roseobacter litoralis Roseovarius tolerans Silicibacter pomeroyi Marinovum algicola Rhodobacter sphaeroides Rhodobacter blasticus Prosthecomicrobium sp. Bauldia consociata Stella humosa Blastobacter sp. Escherichia coli 0.1

Fig. S5. Maximum likelihood phylogeny inferred from 16S rRNA sequences. Taxon label text colors indicate bacteria described as reproducing by budding (red), binary fission (blue), budding or binary fission (green), and bacteria for which the mode of reproduction is not clearly described (black) (1). Highlighting indicates bacteria currently or formerly classified as members of the paraphyletic genera of budding bacteria Prosthecomicrobium (gray) and Blastobacter (yellow) and bacteria shown to exhibit polar growth in this study (blue). Node values >75 (filled squares) and between 50 and 74 (open squares) are indicated. Node values indicate clade frequency among 100 nonparametric bootstrapped datasets. The scale bar indicates the estimated number of substitutions per site. See SI Materials and Methods for details of phylogenetic reconstruction.

1. Garrity G, et al., eds (2005) Bergey’s Manual of Systematic Bacteriology (Williams & Wilkins, Baltimore) Vol 2, Part C.

Brown et al. www.pnas.org/cgi/content/short/1114476109 6of16 A 3 11 14 20 8 15 12 2 9 19 13 16 10 18 1 17 21 6 4 7 5

204 14

A 3 B 12

2 11 8

16 4 9 10 19 15 20 18 5 7 17 21

0 10 20 30 40 50 60 70 80 90 100 Retention time (min)

Fig. S6. HPLC elution proﬁle of A. tumefaciens (A) and S. meliloti (B) muropeptides. A sample (600 μg) of muramidase digested and NaBH4-reduced peptidoglycan was subjected to HPLC and the A204 of the eluate was continuously monitored. Material in the individual peaks was collected, freeze-dried, and further analyzed by MS. Arrows indicate the peaks corresponding to muropeptides shown in Fig. S7.

Brown et al. www.pnas.org/cgi/content/short/1114476109 7of16 Fig. S7. Structures of the individual muropeptides identified in A. tumefaciens peptidoglycan. Numbers above each structure correspond to the labeled peaks for the HPLC profile in Fig. S6. Numbers with an asterisk indicate minor muropeptides whose assigned structures were indistinguishable from those assigned to major muropeptides with different elution times; two asterisks indicate that the position of the LD Dap-Dap peptide bridge is ambiguous. The following conventions for muropeptide naming were used: first letter indicates whether the muropeptide is a monomer (M) or cross-linked dimer (D) or trimer (T); the numbers indicate the amino acids of the peptide side chains from donor to acceptor; G or GG indicate one or two Gly as part of the interpeptide bridge; and (Q) indicates substitution of Glu for Gln on the left peptide moiety; the right-hand side letters indicate one (D) or two (DD) LD-Dap-Dap cross-links in the muropeptide, and the presence of one (G) or two (GG) Gly residues at the C terminus of the terminal acceptor muropeptide.

Brown et al. www.pnas.org/cgi/content/short/1114476109 8of16 Table S1. Identiﬁcation of HPLC-puriﬁed muropeptides from A. tumefaciens by mass spectrometry Molecular mass‡ [M+H]+ HPLC Muropeptide † peak* structure Theoretical§ Experimental Difference¶ Comments

1 M3 871.37 871.15 −0.22 2 M3G 928.39 928.32 −0.07 3 M4 942.40 942.12 −0.28 4 M4b 942.40 942.25 −0.15 Isomer of M4. Identical MS/MS spectra 5 M44 1,385.60 1,385.43 −0.17 Product of N-acetylmuramyl-L-ala amidase digestion of muropeptide D44 6 M5 1,013.44 1,014.02 0.58 M43 1,314.57 1,315.01 0.44 Product of N-acetylmuramyl-L-ala amidase digestion of muropeptide D34 7 D3GG3 1,837.77 1,837.28 −0.49 Diglycine interpeptide bridge 8 D3G3 1,780.75 1,780.56 −0.19 Gly interpeptide bridge 9 D33D 1,723.73 1,723.10 −0.63 10 D43GG 1,908.81 1,908.88 0.07 Gly-Gly C-terminal dipeptide instead of D-Ala-D-Ala 11 D34D 1,794.76 1,794.13 −0.63 12 D34GD 1,851.78 1,851.90 0.12 13 D43 1,794.76 1,794.40 −0.36 14 D44 1,865.80 1,865.74 −0.06 15 T333GDD 2,633.11 2,632.81 −0.30 Trimer cross-linked exclusively by LD-Dap-Dap bridges 16 T333DD 2,576.11 2,576.03 −0.08 Trimer cross-linked exclusively by LD-Dap-Dap bridges D44b 1,865.80 1,865.50 −0.30 Isomer of D44. Identical MS/MS spectra. Minor component 17 T43(Q)3D 2,646.12 2,645.54 −0.58 Glutamine instead of glutamic acid in one of the tripeptide moieties 18 T434D 2,718.16 2,718.23 0.07 19 T4(Q)3(Q)4D 2,716.16 2,716.23 0.07 Isomer of T434D, with two glutamine residues 20 T444 2,789.20 2,789.24 0.04 21 T343GD 2,704.14 2,703.76 −0.38

*Numbers correspond with the indicated peaks in the HPLC shown in Figs. S4 and S6. Peaks 6 and 16 contain two distinct muropeptides. † The abbreviated notation corresponds with the structures shown in Fig. S7. ‡ Molecular mass was determined by MALDI MS and the proposed amino acid and amino sugar sequences of muropeptides were conﬁrmed by electro spray MS/MS. §Molecular masses were calculated using ChemSketch software and correspond to the reduced (alcoholic) forms of muropeptides with muramicitol instead of muramic acid. ¶Experimental-theoretical mass.

Brown et al. www.pnas.org/cgi/content/short/1114476109 9of16 Table S2. Muropeptide composition of peptidoglycan puriﬁed from actively growing A. tumefaciens and S. meliloti cells A. tumefaciens S. meliloti

HPLC peak* Muropeptide† Area (%)‡ M.F. (%)§ Area (%)‡ M.F. (%)§

1 M3 3.42 5.62 6.23 8.90 2 M3G 5.76 9.46 12.32 17.60 3 M4 16.50 27.09 16.89 24.13 M4G ——1.96 2.80 4 M4b 1.18 1.94 3.94 5.62 5 M8 0.56 0.92 1.36 1.94 6 M5 1.60 2.63 0.18 0.26 7 D3GG3 1.47 1.21 0.83 0.59 8 D3G3 6.64 5.45 8.97 6.41 9 D33D 4.54 3.73 3.78 2.70 10 D43GG 0.58 0.48 0.79 0.56 11 D34D 12.33 10.12 8.53 6.09 12 D34GD 4.11 3.38 9.75 6.96 13 D43 4.11 3.38 2.36 1.69 14 D44 15.53 12.75 13.53 9.66 15 T333GDD 2.60 1.42 2.45 1.17 16 T333DD 2.13 1.17 0.54 0.26 17 T43(Q)3D 2.13 1.17 0.97 0.46 18 T434D 4.21 2.30 0.70 0.33 19 T4(Q)3(Q)3D 2.10 1.15 0.87 0.42 20 T444 5.59 3.06 2.12 1.01 21 T433GD 0.74 0.41 0.91 0.43

*Corresponds to peaks in the HPLC shown in Figs. S4 and S6. † Muropeptide structure is shown in Fig. S7. ‡ Percentage of the total integrated area in the corresponding HPLC peak (Figs. S4 and S6). §Molar Fraction (M.F.) as calculated from the area values as described by Glauner et al. (9).

Table S3. Muropeptide groups in peptidoglycan from A. tumefaciens and S. meliloti A. tumefaciens S. meliloti

Muropeptide group* Area (%)† M.F. (%)‡ Area (%)† M.F. (%)‡

Monomers 29.03 47.66 42.89 61.25 Dimers 49.33 40.49 48.55 34.67 Trimers 21.65 11.85 8.56 4.08 Dap-Dap 30.74 22.57 26.93 18.07 Gly-muropeptides 24.06 22.98 37.98 36.52

*Muropeptides were clustered according to the speciﬁc property indicated. † Percentage of the total integrated area in the corresponding HPLC peak (Figs. S6 and S8). ‡ Molar fractions (M.F.) as calculated from the area values as described by Glau- ner et al. (9).

Table S4. Peptide crosslinks in peptidoglycan from A. tumefaciens and S. meliloti A. tumefaciens S. meliloti

Cross-linking (%)* Rel. con (%)† (%)* Rel. con (%)†

Dap-D-Ala-Dap 28.52 44.71 15.58 36.39 Dap-Gly-(Gly)-Dap 6.66 10.44 7.00 16.35 Dap-Dap 28.61 44.85 20.24 47.27 TOTAL 63.78 42.82

*Calculated as by Glauner et al. (9), the contribution of each kind of peptide bridge was considered separately. † Relative concentration (Rel. con) is the percentage contribution of each kind of peptide bridge to total cross-linkage.

Brown et al. www.pnas.org/cgi/content/short/1114476109 10 of 16 Movie S1. DIC image series of an A. tumefaciens cell growing on an LB agarose pad. The movie corresponds to 158 min and is shown at 30 frames per second. Images were acquired every minute.

Movie S1

Movie S2. Combined DIC and epiﬂuorescence images series of A. tumefaciens cells expressing FtsZ-eGFP cells growing on an LB agarose pad containing 1 mM IPTG. The movie corresponds to 100 min and is shown at ﬁve frames per second. Images were acquired every 10 min.

Movie S2

Brown et al. www.pnas.org/cgi/content/short/1114476109 11 of 16 Movie S3. Combined DIC and epiﬂuorescence images series of TRSE-stained E. coli cells growing on an LB agarose pad. The movie corresponds to 180 min and is shown at seven frames per second. Images were acquired every 5 min.

Movie S3

Brown et al. www.pnas.org/cgi/content/short/1114476109 12 of 16 Movie S4. Combined DIC and epiﬂuorescence images series of a TRSE-stained young A. tumefaciens cell growing on an LB agarose pad. The movie corresponds to 440 min and is shown at 10 frames per second. Images were acquired every 10 min.

Movie S4

Movie S5. Combined DIC and epiﬂuorescence images series of a TRSE-stained predivisional A. tumefaciens cell growing on an LB agarose pad. The movie corresponds to 440 min and is shown at 10 frames per second. Images were acquired every 10 min.

Movie S5

Brown et al. www.pnas.org/cgi/content/short/1114476109 13 of 16 Movie S6. Combined DIC and epiﬂuorescence images series of a TRSE-stained predivisional A. tumefaciens cell growing on an LB agarose pad. The movie corresponds to 440 min and is shown at 10 frames per second. Images were acquired every 10 min.

Movie S6

Movie S7. Combined DIC and epiﬂuorescence images series of a TRSE-stained A. tumefaciens cell growing on an AT minimal media agarose pad containing 0.5% (wt/vol) glucose, 15 mM ammonium sulfate, and 0.1 mM acetosyringone. The movie corresponds to 1090 min and is shown at ﬁve frames per second. Images were acquired every 10 min.

Movie S7

Brown et al. www.pnas.org/cgi/content/short/1114476109 14 of 16 Movie S8. Combined DIC and epiﬂuorescence images series of TRSE-stained A. tumefaciens cells growing in close proximity to an A. thaliana root on an agarose pad containing 1 mM calcium chloride and 0.4% sucrose. The movie corresponds to 280 min and is shown at ﬁve frames per second. Images were acquired every 10 min.

Movie S8

Movie S9. Combined DIC and epiﬂuorescence images series of TRSE-stained A. tumefaciens cells growing in close proximity to an A. thaliana root on an agarose pad containing 1 mM calcium chloride, 0.4% sucrose, and 0.5 μg/mL Alexa Fluor 488-conjugated WGA. The movie corresponds to 230 min and is shown at 10 frames per second. Images were acquired every 10 min.

Movie S9

Brown et al. www.pnas.org/cgi/content/short/1114476109 15 of 16 Movie S10. Combined DIC and epiﬂuorescence images series of TRSE-stained B. abortus cells growing on a 2YT agarose pad. The movie corresponds to 420 min and is shown at ﬁve frames per second. Images were acquired every 60 min.

Movie S10

Movie S11. Combined DIC and epiﬂuorescence images series of TRSE-stained O. anthropi cells growing on a LB agarose pad. The movie corresponds to 200 min and is shown at ﬁve frames per second. Images were acquired every 40 min.

Movie S11

Movie S12. Combined DIC and epiﬂuorescence images series of TRSE-stained P. hirschii cells growing on an MMB agarose pad. The movie corresponds to 460 min and is shown at 10 frames per second. Images were acquired every 10 min.

Movie S12

Brown et al. www.pnas.org/cgi/content/short/1114476109 16 of 16