pIdsBB-DVR-HI- EVR-HI
Dids::Tn-Cm(R) producing the cognate DVR-HI-EVR-HI pair, in which the sequences encoding amino acids 761-865 of IdsDBB and amino acids 147 - 169 of IdsEBB have been
replaced with the equivalent sequences from IdsDHI and IdsEHI, respectively. The plasmid is also known as pCS29. This study KAG1599 CS190 CCS04 BB2000::∆ids c. pIdsHI ∆ids::Tn-Cm(R) producing the cognate DHIEHI pair. The plasmid is also known as pCS18. This study KAG1494 CS138 BB2000c. pKG101 Wild type BB2000 carrying plasmid pKG101. (10) KAG066 BB2000::∆ids c.
pIdsBB-CD-mt Dproducing EVR-BB and no ids::Tn-Cm(R) IdsC, or D. The plasmid is also known as
pidsBB∆1034.
(4) KAG023
BB2000::∆ids c.
pIdsBB-E-mt Dproducing DVR-BB and not ids::Tn-Cm(R) E. The plasmid is also known as pidsBB-312mt.
(4) KAG078
S17lpir E. coli mating strain for moving plasmids into P. mirabilis
Table 2.2 Primers used in this study
Plasmid Amplified product Primer pairs (5’ -> 3’) pIdsBB-DVR-
HI-EVR-BB
idsDHI variable region from pLC-048 FD: CCCGACTTACCACAAGATGC RV: TTAGATATTCTCACTGTTAATAAAGCCTA AAAGC pIdsBB-DVR-
HI-EVR-BB Quikchange to introduce A761V and A765T amino acid changes into IdsD protein sequence FD: GCAGGTGCCGTTGGTgttGCTCTTGCCactCG TGACGCATTAGAAGC RV: GCTTCTAATGCGTCACGagtGGCAAGAGCaa cACCAACGGCACCTGC pIdsBB-DVR- BB-EVR-HI
idsEBB upstream of the variable region from pLC- 013 FD: TTCTGCTTTTGGTGCACATC RV: TTGATGTAATATGACCCCCAAAATAAACA TGAAT pIdsBB-DVR- BB-EVR-HI
idsEHI variable region from pIdsHI FD: ATTCATGTTTATTTTGGGGGTCATATTACA TCAA RV: GTATTGGATTTTTTTTGAAAAAAGATTGA TAGCCA pIdsBB-DVR-
BB-EVR-HI idsEvariable region from pLC-BB downstream of the 013 FD: TGGCTATCAATCTTTTTTCAAAAAAAATC CAATAC RV: GAACAGCCACAAGCGACTTT pIdsBB-DVR- HI-EVR-HI
idsDHI variable region from pLC-048 FD: CCCGACTTACCACAAGATGC RV: TTAGATATTCTCACTGTTAATAAAGCCTA AAAGC pIdsBB-DVR-
HI-EVR-HI Quikchange to introduce A761V and A765T amino acid changes into IdsD protein sequence FD: GCAGGTGCCGTTGGTgttGCTCTTGCCactCG TGACGCATTAGAAGC RV: GCTTCTAATGCGTCACGagtGGCAAGAGCaa cACCAACGGCACCTGC
Results
To determine whether the in vitro binding interactions between IdsD (D) and IdsE (E) correlate with in vivo behaviors, I used an in vivo Ids expression system in which all BB2000- derived ids genes are expressed from a plasmid under the native control of the ids promoter (pIdsBB) in a BB2000-derived mutant strain lacking the ids genes (∆ids) (4). I used this simplified system in which all genes are identical except for the expressed ids genes to remove contributions to self recognition-dependent boundary formation due to differences at other loci. To test the hypothesis that in vitro binding interactions between D and E correlate with in vivo
self identity, I replaced the variable regions (VRs) in the idsD and idsE genes (amino acids 761 to 865 of D and 147 to 169 of E), individually or together, in plasmid pIdsBB with those from strain HI4320 and introduced each construct into the ∆ids strain.
These strains were subjected to boundary formation assays, which are currently the standard assay for studying self identity in P. mirabilis (1-4, 11). When two migrating
populations merge to form a single swarm upon meeting they are described as self and when a boundary forms between the two populations they are described as nonself (2-4). Expression of the ids genes from BB2000 (pIdsBB) in a ∆ids background results in a strain that merges with BB2000, indicating that BB2000 is recognized as self (4). By contrast, expression of the ids
genes from strain HI4320 (pIdsHI) in a ∆ids background led to a boundary with ∆ids carrying pIdsBB as well as the wild type strain HI4320 (Fig. 2.1, Fig. A.1). The boundary formation with strain HI4320 likely results from multiple factors independent from the ids genes, such as the putative cytotoxic idr (9) and/or pef (11) genes.
Surprisingly, ∆ids strains carrying pIdsBB-DVR-HI-EVR-BB or pIdsBB-DVR-BB-EVR-HI, which encode individual VR exchanges in D and E, respectively, did not clearly merge or form a
Figure 2.1 In vitro binding interactions between D and E correlate with self identity in vivo
Swarm-permissive agar surfaces were inoculated with strains expressing different D-E pairs. Strains are ∆ids complemented with pIdsBB, a low copy number plasmid encoding the entire ids
operon under control of its native promoter (4). Identities of expressed D and E variants are indicated. Variable region exchanges from BB2000 (BB) to HI4320 (HI) are indicated with the subscript “VR-HI”. DHI and EHI are D and E variants derived completely from HI4320 and encoded on pIdsHI. Both ∆ids and HI4320 carry the empty parent vector pKG101 (10). Close- ups of contact regions between approaching swarms are shown; the arrow below indicates where two swarms meet. In some cases, more than one swarm ring per swarm is visible. The dashed box defines the region of contact between indicated swarms when more than two swarms are visible within the frame. Green rectangles outline combinations of swarms that merged. Full images of swarm plates are shown in Fig. A.1 in Appendix A.
Figure 2.1 In vitro binding interactions between D and E correlate with self identity in vivo
boundary with strains expressing the ids genes from either BB2000 or HI4320 (Fig. 2.1, Fig. A.1). Strikingly, ∆ids carrying pIdsBB-DVR-HI-EVR-HI, which encodes VR exchanges in both D and E, simultaneously merged with the strain expressing HI4320-derived ids genes (pIdsHI) and formed a boundary against the strain expressing BB2000-derived ids genes (pIdsBB) (Fig. 2.1, Fig. A.1). This observation is consistent with the observed in vitro binding interactions between cognate D and E variants.
As a control, I tested whether the D and E VR exchange variants are actually produced in swarming P. mirabilis cells. For this purpose, I used custom antibodies that were raised against amino acids 4-18 (EVNEKYLTPQERKAR) of D and amino acids 298-312
(EQILAKLDQEKEHHA) of E. These antibodies do not produce any signals at the sizes expected for D and E in strains lacking D and/or E, respectively (Fig. 2.2). Thus, these antibodies specifically recognize D and E. What is more, the antibodies were raised against regions of D and E that are conserved between different species and as such they recognize D and E variants from different strains making them useful tools to detect D and E expression (Fig. 2.2). Indeed, exchanging the D and/or E VRs does not interfere with expression of these self- identity determinants (Fig. 2.2). Hence, any changes to self-recognition behaviors in these strains are due to the presence of altered VRs rather than a lack of D and/or E production.
Together, these experiments show that exchanging the VRs in both D and E, in otherwise isogenic strains, leads to the conversion of strain-specific identity from that of one isolate to another in vivo, indicating that the D/E binding interactions are one factor that contributes to the definition of strain identity.
Figure 2.2 Variable region exchange variants of D and E are produced Whole cell extracts collected from actively migrating swarm colonies were probed for the presence of D and E using custom anti-D (top) and anti-E (middle) antibodies. A Coomassie Blue-stained gel (bottom) was used as a loading control. For each sample, the strain background as well as D and E variants encoded on pIdsBB derivates are indicated. Variable region exchanges from BB2000 (BB) to HI4320 (HI) are indicated with the subscript “VR-HI”. DHI and EHI are D and E variants derived completely from HI4320 and encoded on pIdsHI. Both ∆ids and BB2000 (BB) carry the empty parent vector pKG101 (10).
Discussion
In this chapter, I have shown that in vitro binding interactions between the presumptive self-identity determinants D and E correlate with P. mirabilis self-recognition behaviors in vivo. If strains with cognate pairs of D and E meet on a surface, the two swarm colonies merge to form one larger colony. In contrast, when strains with noncognate pairs of D and E meet on a surface, the two swarm colonies form a macroscopic boundary.
Several open questions remain, however, regarding the mechanism of self-identity communication in P. mirabilis. For example, how do D and E establish and signal self identity? It is known that D is exported from cells by a type six secretion system, while E is not (9). This makes it likely that D is transferred between cells to communicate identity; if it binds a cognate E in the recipient cell, self recognition occurs, while lack of binding signals nonself recognition. In the following chapter, I probe this model.
References
1. Budding AE, Ingham CJ, Bitter W, Vandenbroucke-Grauls CM, Schneeberger PM. 2009. The Dienes phenomenon: competition and territoriality in swarming Proteus mirabilis. J Bacteriol 191:3892-3900.
2. Dienes L. 1947. Further observations on the reproduction of bacilli from large bodies in
Proteus cultures. Proc Soc Exp Biol Med 66:97-98.
3. Senior BW. 1977. The Dienes phenomenon: identification of the determinants of compatibility. J Gen Microbiol 102:235-244.
4. Gibbs KA, Urbanowski ML, Greenberg EP. 2008. Genetic determinants of self identity and social recognition in bacteria. Science 321:256-259.
5. Cardarelli L, Saak C, Gibbs KA. 2015. Two proteins form a heteromeric bacterial self- recognition complex in which variable subdomains determine allele-restricted binding. mBio 6:e00251.
6. Belas R, Erskine D, Flaherty D. 1991. Transposon mutagenesis in Proteus mirabilis. J Bacteriol 173:6289-6293.
7. Vallejo AN, Pogulis RJ, Pease LR. 2008. PCR Mutagenesis by Overlap Extension and Gene SOE. CSH Protoc 2008:pdb prot4861.
8. Simon R, Priefer U, Pühler A. 1983. A broad host range mobilization system for in vivo
genetic engineering: transposon mutagenesis in gram negative bacteria. Nat Biotechnology 1:784-791.
9. Wenren LM, Sullivan NL, Cardarelli L, Septer AN, Gibbs KA. 2013. Two independent pathways for self-recognition in Proteus mirabilis are linked by type VI- dependent export. mBio 4: e00374.
10. Gibbs KA, Wenren LM, Greenberg EP. 2011. Identity gene expression in Proteus mirabilis. J Bacteriol 193:3286-3292.
11. Alteri CJ, Himpsl SD, Pickens SR, Lindner JR, Zora JS, Miller JE, Arno PD, Straight SW, Mobley HL. 2013. Multicellular bacteria deploy the type VI secretion system to preemptively strike neighboring cells. PLoS Pathog 9:e1003608.
Chapter 3
The self-identity protein IdsD is communicated between cells in swarming Proteus mirabilis colonies
Most of the work presented in this chapter was published as Saak CC, Gibbs KA. 2016. The Self-Identity Protein IdsD Is Communicated between Cells in Swarming Proteus mirabilis
Colonies. J Bacteriol 198:3278-3286. Full publication can be found in Appendix B.
A patent application has been filed based on Saak & Gibbs (2016). The U.S. provisional application no. is 62/375,248.
Abstract
IdsD (D) and IdsE (E) encode identity information in the bacterium Proteus mirabilis. D and E bind in vitro in an allele-restrictive manner, and D-E binding is correlated with populations merging whereas a lack of binding correlates with populations separating. Key questions about in vivo D and E interactions remained, specifically whether D and E bind within a single cell or whether D-E interactions occur across neighboring cells, and if so, which of the two proteins are exchanged. Here, I demonstrate that D must originate from another cell to communicate identity via a type six secretion system and that this non-resident D interacts with E resident in a recipient cell. Further, I show that unbound D in the recipient cell does not cause cell death and instead appears to contribute to a restriction of the expansion radius of the swarming colony. I conclude that P. mirabilis communicates D between neighboring cells for non-lethal kin recognition, suggesting that the Ids proteins constitute a type of cell-cell communication system.
Introduction
IdsD (D) and IdsE (E) are the self-identity determinants in the bacterium P. mirabilis (1).
In vitro interactions between cognate D and E variants from the same strain correlate with self- recognition outcomes in vivo, while lack thereof (as is the case for noncognate D and E variants from different strains) predicts nonself recognition (1). It remained unclear, however, how D and E signal identity between cells. Both D and E contain transmembrane domains (1). Further, D has been found outside of cells, and its export has been shown to be dependent on a functional type six secretion system (T6SS) (2). Consistent with this data, D contains the recently described MIX motif, which has been found among multiple secreted T6SS effector proteins; this MIX motif is predicted to identify previously unknown substrates of the T6SS (3). In contrast, E has
not been found outside of cells and is predicted to be an integral inner membrane protein (1, 2). Given these data, the prevailing hypothesis was that D is delivered to a neighboring cell in a T6SS-dependent manner. Binding to a cognate E in the recipient cell would then signal self recognition, while lack thereof would signal nonself recognition. However, there was no experimental evidence for Ids transfer between cells. It was also unclear, whether Ids-mediated nonself recognition causes lethality of nonkin cells. Last, whether boundary formation results from interactions among the Ids proteins within one cell, or between cells, has not been previously addressed.
In this chapter, I demonstrate that D-E interactions, or lack thereof, do not cause lethality in P. mirabilis. Further, I provide evidence that D is communicated from one cell to another in a T6SS-dependent manner and that interactions with E in the recipient cell determine behavior. I also present first evidence that these D-E interactions impact the expansion of a swarming colony. These data demonstrate that kin recognition in P. mirabilis entails the cell-cell communication of an identity-encoding protein.
Materials and Methods Bacterial strains and media
Strains and plasmids used in this study are described in Table 3.1. P. mirabilis strains were maintained on LSW- agar (4). CM55 blood agar base agar (Oxoid, Basingstoke, England) was used for swarm-permissive nutrient plates. Overnight cultures of all strains were grown in LB broth under aerobic conditions at 37°C. Kanamycin was used at a concentration of 35 micrograms/milliliter (µg/ml) for plasmid maintenance and was added to all swarm and growth
I employed a previously published ids expression system in which the entire ids locus from P. mirabilis strain BB2000 is expressed from a low-copy number plasmid under control of its native promoter (pIdsBB) in a BB2000-derived strain lacking the chromosomal copy of the
ids operon (∆ids) (5). I engineered alterations to the ids locus on the vector; hence, all strains are isogenic except for the encoded ids genes.
Plasmid and strain construction
Construction of pIdsBB-ΔE
The pIdsBB-ΔE plasmid was constructed using the 390-basepair (bp) gBlock oCS80 (Integrated DNA Technologies, Inc., Coralville, IA) containing the last 266 bp of idsD, the last 18 bp of idsE and the first 106 bp of idsF (gBlock sequence: 5’-
GCGAACAATTAAAAATGGCAAGTGAAAAAGGTGATTGGAACCCTGAAACAGGTATA TTTAAATTTAGTTTGGAAGTACAGTCTCAATTAGTAAATACATATTCTGCTTTTGGTG CACATCCTAATAGCCGTATAGGTATTGAAGATTTATATTGGTATTATCAAGTCAATC CCGAGGTAACAACACCGATGCGTTATATCAATTGGGGGGGAGATACCCAAGAAAAC AATCAGCTTTTAGGCTTTATTAACAGTGAGAATATCTAAATCAGGAGAAAGAACACC ATGCGTAGTTTGGTAAACGGCAGAAAGATTATTTTAGAAAATGATACAACAAATAC CGGCGGTACCGTACTTACCGGCTCTTCTATTGCTAAACAAACACAAGGGG-3’). EcoNI and KpnI restriction sites within this fragment were used to replace the sequence between EcoNI and KpnI in pIdsBB (5). Ligations were transformed into One Shot® OmniMaxTM 2 T1R
chemically competent Escherichia coli (Thermo Fisher Scientific, Waltham, MA). Oligos were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA), and DNA sequencing was performed by Genewiz (South Plainfield, NJ).
Construction of the vipA mutation
A swarm-capable, spontaneous mutant strain of BB2000::∆ids carrying pIdsBB-E-mt (5) was isolated. This isolate was subjected to phenol-chloroform extractions to isolate genomic DNA (gDNA). gDNA was sheared using a Covaris S 220 (Covaris, Woburn, MA), and a library for whole genome sequencing was prepared using the PrepX ILM DNA Library Kit (WaferGen Biosystems, Fremont, CA) for the Apollo 324 NGS Library Prep System (WaferGen
Biosystems, Fremont, CA). The library was sequenced as 100-bp, paired-end reads using an Illumina HiSeq 2500 system (Illumina, San Diego, CA). Reads were aligned to the P. mirabilis
BB2000 genome (GenBank accession no. CP004022) using Geneious (Biomatters, Auckland, New Zealand). Suppressor-specific polymorphisms were identified by aligning the assembled genome to that of the ancestral strain, BB2000::∆ids. The identified mutation mapped to gene
bb2000_0821 (GenBank accession no. AGS59317.1) encoding a vipA homolog [T6SS_VipA (PFAM family PF05591)]. BB2000::∆ids, vipAT95G was then constructed by amplifying the vipA
containing fragment from gDNA of the isolated spontaneous mutant strain (forward primer oCS138: 5’-CGCGGGCCCGGTATTACCCCATAAATAGTGC-3’, reverse primer oCS139: 5’- GCGCGCTCTAGACCTTAAGTTAAACCAAATATAGCTG -3’). Restriction digestion with ApaI and XbaI was used to introduce this sequence into the suicide vector pKNG101 (6). The resulting vector pCS34 was introduced into mating strain E. coli SM10λpir (7) and then mated into BB2000::Δids (5). Matings were subjected to antibiotic selection on LSW- agar (15 µg/ml tetracycline and 25 µg/ml streptomycin). Candidate strains were subjected to sucrose counter- selection as previously described (8). Double-recombinants were confirmed using whole genome sequencing as described above. The Bauer Core Facility at Harvard University performed all
Colony expansion and co-swarm assays and viable cell counts
Overnight cultures were normalized to an optical density at 600 nm (OD600) of 0.1 and swarm-permissive nutrient plates were inoculated with one microliter (µl) of normalized culture. Plates were incubated at 37 °C for 16 hours, and radii of actively migrating swarms were
measured. Additionally, widths of individual swarm rings within the swarm colonies were recorded. For co-swarm assays, strains were processed as described and mixed at a ratio of 1:1 where indicated.
For viable cell counts after 16 hours, actively migrating swarms were resuspended in six ml of LB medium and 20 µl of the cell suspension were used for a 10-fold dilution series. A total of eight dilutions were prepared for each sample and 10 µl of each dilution were spotted onto LSW- agar.
For measuring viable cell counts over time, swarm plates were set up as above. Viable cell counts at time point zero were determined by preparing a 10-fold dilution series of the normalized overnight cultures and spotting 10 µl of each dilution on LSW- agar plates. Viable cell counts at time points two, four, six and eight hours post-inoculation were determined by resuspending swarm colonies in one ml LB medium and preparing 10-fold dilution series as described above. 10 µl of each dilution were spotted onto LSW- agar. Dilutions with countable numbers of colonies were used to determine viable cell counts of swarm colonies.
Swimming assay
Overnight cultures were normalized to OD600 0.1. An inoculation needle was used to inoculate 0.3% LB nutrient plates. Plates were incubated at 37 °C for nine hours and diameters of swim colonies were measured.
Measuring generation times
Overnight cultures were normalized to OD600 0.1 in LB medium. Normalized cultures were grown overnight at 37 °C shaking periodically in a Tecan Infinite® 200 PRO microplate reader (Tecan, Männedorf, Switzerland). Generation times were calculated from logarithmic phase growth measurements.
Trichloroacetic acid precipitations, SDS-PAGE, western blots and LC-MS/MS
Trichloroacetic acid precipitations were performed as previously described (2). Samples were normalized according to OD600 of the liquid cultures at collection, separated by gel
electrophoresis using 12% Tris-tricine polyacrylamide gels, transferred onto 0.45 micrometer (µm) nitrocellulose membranes, and probed with monoclonal rabbit-anti-FLAG (Sigma-Aldrich, St. Louis, MO) and mouse-anti-s70 primary antibodies (Thermo Fisher Scientific, Waltham, MA), followed by polyclonal goat-anti-rabbit (KPL, Inc., Gaithersburg, MD) and goat-anti- mouse secondary antibodies (KPL, Inc., Gaithersburg, MD), respectively. Membranes were developed with the Immun-Star HRP Substrate Kit (Bio-Rad Laboratories, Hercules, CA) and visualized using a Chemidoc (Bio-Rad Laboratories, Hercules, CA). TIFF images were exported and figures were made in Adobe Illustrator (Adobe Systems, San Jose, CA).
To detect secreted proteins in liquid supernatants by mass spectrometry, trichloroacetic