Embed Size (px)
Citation preview
1
Supporting Information 1
Monitoring F1651 P-like fimbriæ expression at the single-cell level 2
reveals a highly heterogeneous phenotype 3
Richard Graveline1*†, Rémi Lavoie1†, Philippe Garneau1, France Daigle2, Serge 4
Sénéchal2, Christine Martin3 and Josée Harel1# 5
Université de Montréal, 1Groupe de Recherche sur les Maladies Infectieuses du Porc (GREMIP) and 6
Centre de Recherche en Infectiologie Porcine (CRIP), Faculté de Médecine Vétérinaire, Université de 7
Montréal, St-Hyacinthe, CANADA; 2Département de Microbiologie et Immunologie, Faculté de 8
Médecine, Université de Montréal, Montréal, CANADA ; 3Institut National de la Recherche 9
Agronomique, Saint-Genès-Champanelle, FRANCE 10
†These authors contributed equally to the work 11
# Corresponding author: 12
Josée Harel, 13
Université de Montréal, Faculté de Médecine Vétérinaire 14
Groupe de Recherche sur les Maladies Infectieuses du Porc (GREMIP) and Centre de 15
Recherche en Infectiologie Porcine (CRIP), 16
3200 rue Sicotte 17
St-Hyacinthe, QC, Canada 18
450-773-8521 19
21
* Current Address 22
Table of Contents 23
24
2
Text S1: Material and Methods 25
26
Movie S1: Phase variation in a growing microcolony of JH301 (papI-pap) 27
Movie S2: Phase variation in a growing microcolony of JH302 (fooI-pap) 28
Movie S3: Phase variation in a growing microcolony of JH303 (fooI-foo) 29
30
Table S1: Primers used in this study 31
32
Figure S1: Gate acquisition corresponding to Fig. 3. 33
34
Figure S2. Molecular model of the steps involved in P and F1651 fimbriae 35
switching event mechanism. 36
37
Figure S3: F1651 fimbrial production of E. coli clinical animal strains 38
39
Figure S4: Differential methylation of pap and foo intergenic region 40
41
42
43
44
45
46
47
3
Text S1. Material and Methods 48
49
Media and growth conditions. Luria-Bertani (LB) broth, M63, and M9 minimal 50
broths were prepared as described previously (1, 2). When necessary, media were 51
supplemented with antibiotics at the following concentrations (unless otherwise 52
noted): ampicillin (amp;, 100 g.ml-1); chloramphenicol (Cm; 10 g.ml-1); 53
gentamicin (10 g.ml-1); kanamycin (40 g.ml-1); tetracycline (tet; 12.5 g.ml-1); and 54
5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside (X-gal; at a final concentration 55
of 40 g.ml-1). For microscopy experiments, M9 minimal salts x 1 media 56
supplemented with 0.2% glycerol and gentamicin (10 µg/ml) (Sigma Aldrich) and 57
2% BD BactoTM-Agar. 58
59
Real-time Microscopy 60
Microscopy sample preparation. Bacteria were first cultivated on LB agar and then 61
isolated on minimal M9 media with glycerol and grown for 24 h at 37°C. A few colonies 62
were harvested and resuspended in sterile 1 ml PBS solution. The prewarmed cell 63
suspension was then diluted in PBS at 1/100 so that the density of cells allows the 64
observation of single cells. Five ul of the bacterial suspension was dropped on the 65
agar strip. An adhesive silicone isolator (Grace Bio-labs; 19 mm x 32 mm x 1 mm) 66
was affixed onto a microscope slide to create a chamber that contains the solid 67
media. A volume of 700 µl culture media consisting of M9 glycerol medium with 2% 68
agar was poured into the silicone isolator. A coverslip (22 mm x 60 mm; Fisher) 69
coated with Sigmacote (Sigma) was then placed over the isolator. Once the agar was 70
4
solidified, the coverslip was removed. The now flat-surfaced agar was trimmed into 71
a thin strip of about 19 mm x 15 mm. The rest of the cavity acted as an air reservoir 72
for cell growth. The cells spread evenly and became immobile. Over time, each single 73
cell grew into a microcolony. 74
75
Image acquisition. Microscope slides and bacterial suspensions were placed 76
under the 37°C incubation chamber of the Zeiss Axiovert inverted microscope to 77
limit the cold shock when the cells are transferred onto the microscope slides. Five 78
ul of the bacterial suspension was dropped on the agar strip. When the sample was 79
immediately closed using another coverslip, it was ready for observing the growth 80
of a single ON phase cell. The cell had to be identified and isolated (such as during 81
growth there is no overlap of the microcolonies, which would be detrimental to the 82
tracking analysis). Growth was followed for 10 h at 37°C with images taken every 10 83
min; both DIC (differential interference contrast) and fluorescence images were 84
captured. Both sets of images are needed to follow the growth and the switching 85
events occurring for each bacterium. Imaging was performed with a Zeiss Aksiovert-86
A1 inverted microscope. 87
88
Data analysis. Cell fluorescence quantification and lineage tree construction 89
were performed with Schnitzcell, software written in MATLAB and kindly provided 90
by M. Elowitz (Young et al., 2011). Schnitzcell uses the Image Processing and the 91
Statistics toolboxes, along with the basic MATLAB module. 92
93
5
Immunofluorescence microscopy 94
To evaluate the fimbrial production of clinical strains producing F1651 (3) 95
immunofluorescence microscopy was performed as described (4). Briefly, bacteria 96
were grown at 37oC for 12h in M9-glycerol 0.2% media above were fixed in sodium 97
phosphate buffer, pH 7.4, containing 2.5% (w/v) paraformaldehyde and 0.08% 98
(w/v) glutaraldehyde. Fixed bacteria were resuspended in 50 mM glucose, 10 mM 99
EDTA, 20 mM Tris HCl, pH 7.5 and transferred to microscope slides coated with poly 100
L-lysine (1 mg ml_1) and incubated for 30 min. After 2 washes, fixed bacteria were 101
blocked with PBS containing 2% bovine serum albumin (BSA) for 30 min and 102
incubated for 30 min with rabbit polyclonal anti-F165 antibodies diluted 1:100 in 103
blocking solution at 37oC (5). Immune complexes were revealed by incubation with 104
diluted 1:2000 goat anti-rabbit IgG Alexa Fluor 488 conjugate (Invitrogen) and the 105
microscopic slides were examined with a confocal microscope (Olympus FV1000 106
IX81) at 40X and a numerical zoom of 2.5X. Three washes in PBS were done 107
between each incubation. 108
109
Differential methylation assay. The methylation pattern of GATC-I and GATC-II 110
sites of the foo/pap intercistronic region of strains JH200 (fooI-foo-papBA–lacZ) 111
(6) and strain DL4388 (papI-pap-papBA–lacZ) (7) were studied using a qPCR 112
method of digested genomic DNA using restriction enzymes that recognize 113
methylated or non-methylated GATC sites (8). Primer sequences are presented in 114
Table S1. For DNA purification, a single blue or a single white colony was picked 115
from M9-glycerol-X-Gal agar, resuspended in M9 salts, and used to inoculate 116
6
three flasks containing M9 glycerol medium as described previously (6, 9). 117
Genomic DNA was isolated from overnight cultures as described previously (10). 118
DNA concentrations were measured using ND-1000 Spectrophotometer 119
(NanoDrop Technologies, Wilmington, DE). Digestions of 1 μg of DNA with 120
restriction endonucleases DpnI (cuts methylated GATC sites), Sau3A I (cuts all 121
GATC sites, whether they are methylated or not) were performed as specified by 122
the manufacturer. After incubation, the enzymes were inactivated and DNA 123
fragments were purified using the MinElute system from Qiagen (Hilden, 124
Germany). DNAs were quantified by measuring the A260 using ND-1000 125
Spectrophotometer. Using the Smart Cycler apparatus (Cepheid, Sunnyvale, CA), 126
reactions were conducted with qPCR using 1.25 μM of each primer, 5 mM MgCl2, 127
12.5 μl of QuantiTect SYBR Green PCR Kits (Qiagen) and 4 μl of digested DNA (20 128
ng) in microcapillary tubes in a final volume of 25 μl. Primers fooDI and fooR1 for 129
amplification of GATCdist of foo intergenic region (IR), foo_prox-F and fooR2 for 130
GATCprox. Primers for the pap IR were Pap4-6F and Pap4-6R for amplification of 131
GATCdist region and Pap_prox-F, Pap1-3 R for amplification of GATCprox region. 132
133
Standard curves for clp quantification were obtained by PCR amplification of 134
genomic DNA from DL4388 (pap)(7) and MC4100.λ1 (foo) (11). The PCR products 135
were purified with the MinElute PCR purification kit (Qiagen) and DNA amounts 136
were quantified by measuring the A260. This amount was converted to molecule 137
number as previously described (12). Then PCR products were 10-fold serially 138
diluted from 107 to 103 molecules; three qPCRs were carried out in a Smart Cycler 139
7
apparatus (Cepheid). The standard curve for rpoA quantification was obtained in the 140
same way using the genomic DNA of MC4100.λ1 and the primers rpoA_F and 141
rpoA_R. Primers were selected in a region devoid of GATC sites in order to not be 142
digested by the enzymes used. Data were normalized by making the ratio N (copies 143
of tested gene)/ N (copies of rpoA). Here rpoA is used as the reference gene since it 144
is present at only one copy per genome. Thus it gives an indication of the number of 145
genomes and therefore an indication of the number of copies of the intergenic 146
region that were to be digested. The results represent the mean of two independent 147
experiments.148
8
Table S1. Primers used in this study 149
150
Primer Sequence (5’ - 3’)
i-cat2 CCTGGTAATCCGTTACCGCCAGCGCCTCTGCAATTTCTGCAAAATGA
GACGTTGATCGGCACGT
fooUB-sacB TACTCTTCACGCAATAAGTTAAATTTAGTTTTTTATGTTGATCAAAG
GGAAAACTGTCCATATGCACAGATG
pI/fI-fwd CCTGGTAATCCGTTACCGCCAG
fooI->papI-F AATCCGTTACCGCCAGCGCCTCTGCAATTTCTGCCGTTTTCCCTCCAT
CATGCCTGTTCAGAAATTCC
papI->fooI-F AATCCGTTACCGCCAGCGCCTCTGCAATTTCTGCAGTTTTCCCTCCAT
TATGCCTGTTCAGAAATTCC
fooUB40 TACTCTTCACGCAATAAGTTAAATTTAGTTTTTTATGTTG
fooI-fwd AGTTGTGGAAGAACAGCTTTGCCC
foopB-rvs ACTTCATGATGCGCCATGTTTCCC
lacZ-T7pol_fwd GCATTTTAACTTTCTTTATCACACAGGAAACAGCTACACGATTAACA
TCGCTAAGAAC
lacZ-CAM_R CGCTCATCGCCGGTAGCCAGCGCGGATCATCGGTCAGACGATGGAGT
TCTGAGGTCATTACTG
SacI-genta-F TACCGAGCTCGAATTGGCCGCGGCG
AatII-genta-R ATTAGACGTCGGAATTGCCAGCTGGGGCGCCCTCTGGTAAGGTTGGG
AAGCCCATTGACATAAGCCTGTTCGGTTCGTAAACTGTAATGC
9
fooDI GTATTTGCTTTATTTGCG
fooR1 CAAACAACAAGAATACAC
Foo_prox-F CACCATGATGTTTTTATCTGAGTGTATTCTTGTTGTTTGTG
fooR2 GTGTATTCTTGTTGTTTG
Pap4-6 F TTCTCTATGTTTGCTTTATTTGTT
Pap-4-6 R TAATAGCAAGAGGGTACTCAGATA
Pap_prox-F TGCCATGATGTTTTTATCTGAGTACCCTCTTGCTATTAGTG
Pap1-3 R TACTCTTCACGCAATAAGTTAAAT
rpoA_F CGTATCAAAGTTCAGCGCGGTCGT
rpoA_R CAGCTTGTCCAGGTCGGTACGC
151
152
153
154
155
156
157
10
Movie S1. Phase variation in a growing microcolony of JH301 (papI-pap) 158
This film shows 720 min (72 frames of fluorescence images) of microcolony growth 159
from a single ON cell condensed in 8 seconds. Images were taken every 10 min. Cells 160
were switched to OFF at the start and then ON switching events can be seen at 2 161
seconds. Medium is M9-glycerol. 162
163
Movie S2. Phase variation in a growing microcolony of JH302 (fooI-pap) 164
This film shows 720 min (72 frames of fluorescence images) of microcolony growth 165
from a single ON cell condensed in 8 seconds. Images were taken every 10 min. 166
Multiple ON and OFF switching events can be seen. Medium is M9-glycerol. 167
168
Movie S3. Phase variation in a growing microcolony of JH303 (fooI-foo) 169
This film shows 720 min (72 frames of fluorescence images) of microcolony growth 170
from a single ON cell condensed in 8 seconds. Images were taken every 10 min. 171
Multiple ON and OFF switching events can be seen throughout. Medium is M9-172
glycerol. 173
174
175
11
Figure S1: Gate acquisition corresponding to Fig. 3. 176
177
178
179
Figure S1. 180
181
182
12
Figure S2. Molecular model of the steps involved in P and F1651 fimbriae 183
switching event mechanism. (adapted from Hernday A, Krabbe M, Braaten B, Low 184
D. 2002. Self-perpetuating epigenetic pili switches in bacteria. Proceedings of the 185
National Academy of Sciences of the United States of America 99 Suppl 4:16470-186
16476. © (2002) National Academy of Sciences, U.S.A.)(13). During phase variation 187
Dam and Lrp compete for binding to the unmethylated switch region. Binding of Lrp 188
to the unmethylated switch region blocks Dam, thereby preventing DNA 189
methylation. As Lrp can also bind to hemimethylated sites, during the replication of 190
an ON cell (Replication I), Lrp can either bind the unmethylated distal site or the 191
hemimethylated proximal site. Binding to the proximal site leads to the OFF (or 192
partial) state. At the second round of replication (Replication II) of the daughter cell 193
in OFF state, Lrp will either bind hemimethylated distal or proximal sites, or 194
unmethylated proximal site. Binding of Lrp to the proximal site will then generate 195
cells in the OFF state and binding to the distal site will lead to cells in the ON (or 196
partial state). The intermediate expression evidenced by foo could be mainly during 197
ON to OFF switching of daughter cells that inherit only GFP protein and messenger 198
RNA, which can be diluted by growth and thus be decreased in concentration if the 199
cell lineage remains in the OFF state. However the high rate of foo switching points 200
out a cell population in the intermediate state. 201
202
203
13
204
205
Figure S2 206
207
208
14
Figure S3. F1651 fimbrial production of E. coli clinical animal strains. E. coli strains 209
4787 and 5131 were grown on M9 glycerol. The bacteria were fixed, transferred to 210
microscope glass slides and labeled with polyclonal anti-F1651 antibodies. Goat anti-211
rabbit serum conjugated Alexa 488 fluorescent dye was used as secondary antibody 212
and the slides were mounted for examination using confocal microscope. The 213
fluorescence (A, D) and the corresponding phase contrast (B, E) and merged fields 214
(C, F) are shown from right to left images, respectively. The upper panel 215
corresponds to E. coli strain 4787and the lower panel to E. coli strain 5131. 216
217
218
219
Figure S3 220
221
15
Figure S4: Differential methylation of papWT and fooWT intergenic region. The 222
fraction of hemimethylated sites of both repressor and activator regions is more 223
important in foo than in pap, suggesting a frequent change of the methylation DNA 224
status during foo cell division. In contrast, GATCprox site of the repressor region of 225
pap remains methylated in greater proportion than GATCdist of the activator region. 226
This correlates with the prevalent OFF state of pap cells. The percentage of 227
amplification of each GATC site, proximal (GATCprox) and distal (GATCdist), is 228
evaluated after enzymatic digestion by DpnI. Values are normalized by the 229
amplification of a house-keeping gene (rpoA). Amplification is conducted on DNA 230
from two conditions, ON and OFF cells for a pap- and a foo-bearing strains. 231
232
233
234
Figure S4 235
236
16
References 237
238
1. Miller JH. 1972. Experiments in Molecular Genetics. Cold Spring Harbor, NY: 239
Cold Spring Harbor Laboratory. 240
2. Thomason L, Court DL, Bubunenko M, Costantino N, Wilson H, Datta S, 241
Oppenheim A. 2007. Recombineering: genetic engineering in bacteria using 242
homologous recombination. Curr Protoc Mol Biol Chapter 1:Unit 1.16. 243
3. Harel J, Daigle F, Maiti S, Désautels C, Labigne A, Fairbrother JM. 1991. 244
Occurrence of pap-, sfa-, and afa-related sequences among F165-positive 245
Escherichia coli from diseased animals. FEMS Microbiol Lett 66:177-182. 246
4. Berthiaume F, Rutherford N, Mourez M. 2007. Mutations affecting the 247
biogenesis of the AIDA-I autotransporter. Res Microbiol 158:348-354. 248
5. Harel J, Daigle F, Forget C, Tessier MC, Crost C, Martin C. 2000. Phase 249
variation of F165(1) (Prs-like) fimbriae from Escherichia coli causing 250
septicaemia in animals. Can J Microbiol 46:1101-1107. 251
6. Graveline R, Mourez M, Hancock MA, Martin C, Boisclair S, Harel J. 2011. 252
Lrp-DNA complex stability determines the level of ON cells in type P fimbriae 253
phase variation. Mol Microbiol 81:1286-1299. 254
7. Braaten BA, Nou X, Kaltenbach LS, Low DA. 1994. Methylation patterns in 255
pap regulatory DNA control pyelonephritis-associated pili phase variation in 256
E. coli. Cell 76:577-588. 257
8. Tessier MC, Graveline R, Crost C, Desabrais JA, Martin C, Drolet M, Harel 258
J. 2007. Effects of DNA supercoiling and topoisomerases on the expression of 259
17
genes coding for F165(1), a P-like fimbriae. FEMS microbiology letters 260
277:28-36. 261
9. Crost C, Garrivier A, Harel J, Martin C. 2003. Leucine-responsive regulatory 262
protein-mediated repression of clp (encoding CS31A) expression by L-263
leucine and L-alanine in Escherichia coli. Journal of bacteriology 185:1886-264
1894. 265
10. Harel J, Forget C, Ngeleka M, Jacques M, Fairbrother JM. 1992. Isolation 266
and characterization of adhesin-defective TnphoA mutants of septicaemic 267
porcine Escherichia coli of serotype O115:K-:F165. Journal of general 268
microbiology 138:2337-2345. 269
11. Crost C, Harel J, Berthiaume F, Garrivier A, Tessier MC, Rakotoarivonina 270
H, Martin C. 2004. Influence of environmental cues on transcriptional 271
regulation of foo and clp coding for F165(1) and CS31A adhesins in 272
Escherichia coli. Research in microbiology 155:475-482. 273
12. Fronhoffs S, Totzke G, Stier S, Wernert N, Rothe M, Bruning T, Koch B, 274
Sachinidis A, Vetter H, Ko Y. 2002. A method for the rapid construction of 275
cRNA standard curves in quantitative real-time reverse transcription 276
polymerase chain reaction. Molecular and cellular probes 16:99-110. 277
13. Hernday A, Krabbe M, Braaten B, Low D. 2002. Self-perpetuating 278
epigenetic pili switches in bacteria. Proceedings of the National Academy of 279
Sciences of the United States of America 99 Suppl 4:16470-16476. 280
281
