Current -

Abstract

Escherichia coli DegS is an integral inner membrane protein that breaks down incorrectly assembled proteins and helps regulate the σ ᴱ stress response pathway. When degS is deleted, approximately 200-fold more outer membrane vesicles (OMVs) are produced. As OMVs contain sections of the outer membrane and proteins from the periplasm, they may play a role in the virulence of E. coli. To test whether degS or OMVs play a role in uropathogenic E. coli (UPEC) virulence, we characterized a degS mutant. We found that the degS deletion did not affect the morphology of the bacterium, but did result in decreased virulence in vivo and in vitro. When we tested our mutant in a murine model of cystitis, mice infected with degS::cat bacteria had a 100-fold decrease in bacterial titers in the bladder at 24 hours post-infection. In vitro binding and invasion assays showed no difference in the ability to bind to host cells, but a significant decrease in the ability of mutant bacteria to invade cells. Together, these data suggest that degS deletion decreases the virulence of E. coli and that an increase in OMV production may be a detriment to virulence. Future studies can separate the role of degS and OMV production in UPEC virulence.

Introduction

Escherichia coli is a common species of bacteria that inhabits the intestines of humans and animals. It is a rod-shaped, gram-negative, and facultative anaerobic bacterium. While most strains are harmless, some are capable of causing disease in humans. One such strain is uropathogenic E. coli (UPEC), the causative agent of over 85% of urinary tract infections (UTIs) (1). UTIs can affect the bladder, termed cystitis, or spread to cause a kidney infection, termed pyelonephritis. Some infections will affect both the bladder and the kidneys. Over 50% of women will develop at least one UTI in their lifetime, and many of these women will have recurrent infections that affect their quality of life (1). While there are likely many host factors that contribute to the occurrence of UTI, UPEC are characterized by the expression of specific virulence factors such as type 1 pili, flagella, hemolysin, and siderophores (2, 3). These virulence factors aid the bacteria in binding the host bladder epithelial cells. Bound bacteria can be internalized, where they are protected from antibiotics and the immune system (3). Virulence factors help the bacteria acquire necessary nutrients and grow to high titers within the cells. Some of these intracellular bacteria will remain hidden for up to 6 months before emerging from the cells and causing a recurrent infection (3). In addition to the intracellular bacteria, bacteria are also able to persist in the lumen of the bladder and can ascend up the ureters to cause pyelonephritis. The discovery of additional UPEC virulence factors could lead to strategies for the prevention of UTIs in humans and the development of more effective drugs for the treatment of E. coli infections.

degS of E. coli encodes a 355-residue protein that is a homolog of the DegP protease (4). The DegS protein is an integral inner membrane protein with an active site in the periplasm (5). This protein is able to detect and break down misfolded proteins in the periplasm so that they can be cleared from the cell (5). Additionally, DegS provides proteolytic cleavage of the RseA anti-sigma factor when misfolded proteins are detected to regulate sigma E (σE) activity in E. coli cells (5). Activity of the σE pathway is essential under conditions of envelope stress, as well as in non-stress conditions, because it maintains homeostasis within cells (6). When degS is mutated in bacteria, the production of outer membrane vesicles (OMVs) is increased by approximately 200-fold, most likely in response to the accumulation of misfolded proteins (7). While much of this original work was done in the non-pathogenic E. coli strain DH5α (7), follow-up work confirmed that deletion of degS from a UPEC strain similarly increases the production of OMVs (Danka and D.A. Hunstad, unpublished).

Figure 1

Figure 1: Growth of UTI89, degS::cat and pdegS in LB broth. Optical density at 600 nm was measured every 40 minutes during growth at 37ºC. UTI89 is shown in blue, degS::cat is shown in red and pdegS is shown in green. LB broth and LB broth with ampicillin (used as blanks) are shown in purple and orange, respectively. The average optical density was graphed, error bars represent standard deviation.

Figure 2

Figure 2: TEM images of UTI89 and degS::cat. Representative images of UTI89 and degS::cat cells are shown under shaken and static growth conditions. The scale bar represents 500 nm.

OMVs are nano-sized (~10-300 nm in diameter), spherical, bilayered proteolipids that are produced when small portions of the bacterial outer membrane bulge away from the cell, break off, and release (8). The production of OMVs has been described as a bacterial stress response, but experimental evidence suggests that OMVs play a role in pathogenesis and are associated with virulence factors and toxins (9, 10). OMVs bear components that are thought to impact the course of infection and are likely to play a significant role in the virulence of gram-negative bacterial pathogens (9, 10). Although much work has been done on the role of OMVs in various settings, it is unknown whether OMVs are liberated by membrane instability or through a more directed process.

To investigate the role that degS and OMVs have on the virulence of UPEC, we characterized a degS::cat UPEC strain. We aimed to determine if degS deletion affects the virulence of UPEC. We hypothesized that degS::cat would be more virulent due to the increased production of OMVs, which can play an important role during infection. If degS::cat was more virulent, it would suggest that OMVs do increase virulence, but that degS is not essential for UPEC virulence. Conversely, if the wildtype strain was more virulent, this could mean that an increase in OMV production is detrimental to virulence, or that degS is essential for virulence in UPEC. We used a variety of in vitro and in vivo assays to compare our strains and to test the virulence of our mutant. Our results indicated that the deletion of degS slightly impairs growth of the bacteria, but does not change the morphology or protein expression in the membranes. Furthermore, degS::cat was less virulent than wild-type UPEC both in vitro and in vivo, demonstrating that the deletion of degS, accompanied by an increase in OMVs, decreases the virulence of UPEC.

Results

Wild-Type, degS::cat, and pdegS Strains Do Not Have Significantly Different Growth in LB

Growth curves of UTI89, UTI89 degS::cat, and UTI89 pdegS were developed to compare the effects that degS mutation has on bacterial growth. Cells were cultured in LB, and the optical density was measured every 40 minutes for 12 hours to create growth curves (Figure 1). Strains had similar growth at all time points. The mutant was less dense than the wild-type or complemented strains at most time points, although this was only significantly different at two points in early log phase.

degS::cat Does Not Have Altered Morphology Under Aerated or Nonaerated Conditions

Transmission electron microscopy was used to examine the effects that shaking and static growth conditions have on the size, shape, and pili of bacterial cells from the wild-type strain or degS::cat. Shaken cultures of either strain produced bacterial cells that were coccoid with few pili (Figure 2B, D). Conversely, cells from static cultures were rod-shaped with lots of pili (Figure 2A, C). The wild-type and mutant strains had similar morphological characteristics in both the static and shaken cultures, indicating that degS::cat does not alter morphology under either condition.

Protein Composition of Membranes and Cytosol is Similar Between Strains

SDS-PAGE gels were used to compare the protein content of whole cell lysates of wild-type and mutant cells to see if degS mutation affects the expression of other proteins. There were no observable differences in the protein composition of whole cell lysates from wildtype or mutant cells when stained with Coomassie blue (Figure 3A). Cellular fractionation and SDS-PAGE gels were used to more closely examine the composition of the inner membrane, outer membrane, and cytosol of the two strains. We did not see any differences in the proteins in the outer membranes of our two strains (Figure 3B). Surprisingly, we found that degS::cat was missing a high molecular weight protein band, approximately 200 kDa, which appeared in the cytosol from wild-type bacteria; however, we were not able to identify this protein (arrow, Figure 3B). Finally, a silverstained SDS-PAGE gel was used to analyze the protein expression in the inner membrane (Figure 3C). We did not note any major differences in the protein content of the inner membrane between the two strains.

Figure 3

Figure 3:Protein expression of whole cells and cellular fractions. (A) Coomassie stain was used to show the protein expression in whole cell lysates from both strains. (B) Coomassie stain was used to show protein content in the outer membrane and cytosol of both strains. Arrow indicates a missing high molecular weight protein in the cytosol of degS::cat. (C) Silver stain was used to show the protein content of the inner membrane of the two strains. Marker indicates the approximate protein size in kDa.

Figure 4

Figure 4: Protein expression in isolated OMVs. (A) Pellets of OMVs after ultra-centrifugation of filter-sterilized supernatants from overnight cultures. Pellets are circled in red. (B) A silver-stained protein gel revealed differences in the protein composition of OMVs. Examples are denoted with asterisks (*). Marker indicates the estimated protein size in kDa.

degS::cat Results in Increased Production of Outer Membrane Vesicles

In order to confirm that degS::cat produces more outer membrane vesicles (OMVs) than the wild-type strain, OMVs were isolated from overnight cultures of the wildtype, degS::cat, and pdegS strains. Visual inspection of the pellets of the OMVs recovered after ultracentrifugation revealed a much larger pellet in degS::cat, compared to either the wild-type or complemented strain (Figure 4A). This was confirmed by resuspending the pellets in equal volumes and determining the protein content in each (UTI89 = 31 μg/mL, UTI89 degS::cat = 910 μg/mL, pdegS = 51 μg/mL). Transmission electron microscopy confirmed that these pellets contained mostly intact OMVs, with minimal amounts of cellular debris (data not shown). A silver-stained SDS-PAGE gel was used to analyze variations in OMV protein content in each strain. We found that many of the proteins in OMVs isolated from degS::cat were different from those in OMVs from the wild-type or complementated strains, and that some of the proteins that were common between strains varied in amount (asterisks, Figure 4B).

degS::cat Does Not Alter Antibiotic Susceptibility

An antibiotic disk diffusion assay was used to compare the antibiotic susceptibility of the wild-type strain and degS::cat. This method was used as a simple way to assess changes in essential pathways (DNA replication, protein synthesis, cell wall synthesis). The assay showed no significant differences in susceptibility to ampicillin, kanamycin, novobiocin, spectinomycin, or tetracycline (p = 0.10, 0.09, 1.0, 0.07 and 0.34, respectively; Figure 5). We therefore conclude that degS deletion and the resulting increase in OMV production do not alter antibiotic resistance and that most essential pathways are intact. There was a significant difference in susceptibility to chloramphenicol (p < 0.000001, Figure 5). As a resistance cassette for this antibiotic was used to replace degS in the mutant, chloramphenicol functioned as our positive control for the assay.

degS::cat Bacteria Cause Less Severe Infections in a Murine Cystitis Model

Female C57BL/6J mice were infected using a wellestablished murine model of cystitis in order to compare the virulence of wild-type and mutant strains. At 24 hpi, mice inoculated with degS::cat had approximately 100- fold fewer bacteria in the bladder than mice inoculated with UTI89 (p < 0.0001, Figure 6).

Figure 5

Figure 5: Antibiotic susceptibility. The antibiotic susceptibility of the wild-type and mutant strains was determined by disc diffusion assay. The average zone of inhibition was graphed, error bars represent standard deviation.

Figure 6

Figure 6: Bladder bacterial titers from infected mice. Colony-forming units (CFU) recovered from mouse bladders at 24 hours post-infection (hpi). Horizontal lines indicate the geometric mean.

degS Mutation Does Not Alter Intracellular Growth In vivo

A murine model was originally used to characterize bacterial growth during UTI. Growth within bladder epithelial cells and the formation of biofilm-like intracellular bacterial communities (IBCs) are critical if an infection is to progress to high titers. The important role of IBCs was confirmed by the discovery of these structures in the urine of human patients with active cystitis (15). Knowing this, fluorescent confocal microscopy was used to examine the growth of wild-type and mutant strains within murine bladder epithelial cells to investigate a potential cause for the differences in bladder titers. The bladders of C57BL/6J mice were inoculated with UTI89/pcomGFP or degS::cat/pcomGFP. Bladders were harvested at 16 hpi and stained for microscopy. We found both early stage and mature IBCs in the bladders regardless of infecting strain, indicating that both strains were capable of growing within bladder epithelial cells and forming mature IBCs (Figure 7). It should be noted that IBCs were more difficult to identify in mice infected with the mutant strain, as all bacteria seemed to have lost the GFP-expressing plasmid for an unknown reason.

Figure 7

Figure 7: Confocal microscopy images of IBCs. Bladders of mice infected with UTI89/pcomGFP or UTI89 degS::cat/pcomGFP were harvested at 16 hpi. IBCs in the bladders were visualized at 63x. Nucleic acids are stained red. IBCs are indicated with an arrow.

Figure 8

Figure 8: Binding and invasion into 5637 human bladder epithelial cells. (A) The average percent of UTI89, degS::cat, pdegS or UTI89 fimH::kan (negative control for binding) cells that were able to bind to host cells. (B) The average percent of each strain that was able to invade into the host cells. (C) The average percent of bound bacteria that invaded into host cells. An asterisk (*) denotes significance as determined by Student t-test compared to wild-type UTI89 (p < 0.05). Error bars represent standard deviation.

degS::cat Has Decreased Ability to Invade Host Cells

Binding and invasion assays were conducted in order to compare how well the wild-type, degS::cat, and pdegS strains bound and invaded 5637 human bladder epithelial cells in vitro. We found that degS::cat bound to 5637 cells as well as the wild-type strain (p = 0.47, Figure 8A). However, the degS::cat strain invaded cells significantly less than the wild-type strain (p = 0.01, Figure 8B). We found that only 0.1% of the degS::cat bacteria that bound to cells actually invaded, as compared to 0.59% of wild-type bacteria (Figure 8CFigure 8A-B). degS::cat bound similarly to the complemented strain (p = 0.28, Figure 8A), but invaded significantly less (p = 0.001, Figure 8B). A strain that is unable to produce pili (UTI89 fimH::kan) functioned as a binding mutant and our negative control. As expected, UTI89 fimH::kan was unable to bind to host cells, and as a result, invaded significantly less than the wild-type strain (p = 0.04, 0.02, respectively; Figure 8AB).

Discussion

To investigate the role that degS and OMVs have on the virulence of UPEC, we characterized a degS mutant strain. This mutant has degS replaced with a chloramphenicol antibiotic resistance cassette. We predicted that degS::cat would be more virulent because of the increased production of OMVs. OMVs have previously been shown to contain outer membrane proteins, toxins, and other factors that can be delivered to host cells, which may in turn enhance infection (9,10). However, our results demonstrated that degS may instead be necessary for virulence and that increased numbers of OMVs do not promote infection.

Our experiments with degS::cat demonstrated that while degS mutation does not alter the morphology of the bacteria under static or shaking growth conditions, the mutant has slightly impaired growth in rich media in early log phase. This difference is no longer significant by mid-log phase, although the mutant does grow at a slightly decreased density. Complementation of the mutation restores the wild-type growth phenotype. These data may indicate that increased OMV production early in growth modifies nutrient allocation and limits the speed at which bacteria can replicate. Similarly, DegS function may be most important in early log phase, when proteases can easily be overwhelmed by misfolded proteins that accumulate as the bacteria rapidly synthesize proteins necessary for growth. In contrast to this idea, the comparison of isolated membranes did not reveal major changes in protein content. However, in the cytosol, a high molecular weight protein was missing in degS::cat, which could be related to a change in virulence.

We were able to confirm that degS deletion results in increased OMV production, and furthermore, were able to restore wild-type production of OMVs by complementing the mutant with the wild-type gene. Interestingly, in addition to increased numbers of OMVs, degS::cat also had changes in the proteins contained within those OMVs. In line with the previous suggestion, these likely represent misfolded proteins that have accumulated in the periplasm due to the lack of functional DegS. Future experiments could confirm this hypothesis by isolating and comparing the periplasm of our three strains. The variability in the protein content of the OMVs and the whole cells likely demonstrates the efficiency for packing misfolded proteins specifically into OMVs to clear them from the cells. We acknowledge that one or more of those proteins could be negatively affecting the virulence of the mutant in vitro and/or in vivo. The degS::cat and wild-type strains also had similar antibiotic susceptibility, indicating that the basic processes through which these antibiotics function are not affected by increased OMV production or by the presence of extra misfolded proteins present in the periplasm due to the decreased DegS activity.

In vitro binding and invasion assays and mouse infections showed that degS::cat is less virulent than wildtype. The binding and invasion assays demonstrated that degS::cat cells were able to bind to host cells just as well as the wild-type strain, indicating that degS::cat does not alter pili production or function. This confirmed our TEM result showing no obvious change in the number of pili produced by the mutant under static conditions. Surprisingly, degS::cat was not able to invade the cells. A complemented mutant strain was able to bind and invade as well as the wild-type strain. However, the degS::cat strain bound and invaded significantly less than its complement. This confirms that the deletion of degS specifically leads to decreased ability to invade. These results may mean that degS is important for the invasion mechanism in this model, or that increased production of OMVs is detrimental to mutant cells. Past research has shown that UPEC invade host cells via an HDAC6-modulated microtubule-dependent pathway (16); because degS::cat was not able to invade host cells, this may mean that excess numbers of OMVs, or a factor packaged in or on OMVs, can disrupt this pathway. It is also possible that degS mutants have limited ability to produce important virulence factors due to energy wasted on misfolded proteins and OMV production.

In a mouse model of cystitis, we found that mice infected with degS::cat had 100-fold fewer bacteria in their bladders at 24 hpi. Consistent with our results, past research has also shown that degS is necessary for virulence in a murine model of peritonitis and that σE regulation is essential for UTI (17, 18). We did not find any differences in intracellular growth of the bacteria when we assessed the bladders for IBC formation. However, IBCs seemed to occur less frequently in mice infected with degS::cat (observation by the authors).

Future experiments could quantify IBC formation in mice infected with the wild-type and mutant strains to determine if there is an invasion defect in vivo as well. Additional experiments could also confirm that the complemented strain is able to rescue the infection defects.

The function of the DegS protein is relatively well understood. Previous studies have indicated that DegS functions as a protease in vivo, contributes to the regulation of σE activity, and clears incorrectly assembled proteins out of the periplasm of the cell (4-6). When degS is not present, OMV production increases, likely as a result of stress on the cell envelope (7). In our strain, OMV production may be using more of the available energy, which could explain why degS::cat grew a little slower than wild-type. Although OMVs have previously been shown to positively impact infection (9, 10), we demonstrate decreased virulence in our UPEC strain when degS is knocked out. This suggests that increased OMV production in UPEC could be stressing the cell instead of aiding in infection. Alternatively, the extra vesicles present in mice infected with degS::cat could increase the activation of the host immune system, resulting in faster clearance of the mutant bacteria. Finally, our experiments do not conclusively determine whether the altered virulence is a due to greater numbers of OMVs or to a different factor specific to the degS mutant. Further research can be done to focus solely on the effects of OMVs in UPEC virulence. Specifically, wild-type UPEC can be tested in vitro and in vivo with exogenous OMVs to determine whether OMVs alter virulence.

By investigating bacterial virulence factors and important bacterial genes, we can better understand how bacteria function. This can allow chemists to create drugs that are more effective against specific bacteria and help doctors determine how best to treat infections. E. coli persists throughout our bodies and our environment, so an in-depth understanding of UPEC virulence could aid in reducing UTI occurrence and recurrence, as well as associated healthcare costs.

Materials and Methods

Bacterial Strains

The model uropathogenic E. coli strain UTI89 was used as the wild-type. This strain was originally isolated from the urine of a patient with cystitis (11). A degS deletion (UTI89 degS::cat) was previously made by replacing degS with a chloramphenicol (cat) resistance cassette using lambda red mutagenesis (12). A UTI89 fimH::kan strain was previously made by replacing the fimH gene with a kanamycin (kan) resistance cassette. This strain does not express type 1 pili. All strains were grown in fresh LB broth without antibiotics, unless otherwise indicated.

Molecular Biology

A degS complementation plasmid was created by amplifying degS from UTI89 genomic DNA (F primer: 5’-CCATCATGTTTGTGAAGCTCTTACGTTCCGTTGCGATTGG-3’, R primer: 5’-TTAGTTGGTCGCCGGATAT TCCTGAATGGTGACCTGC-3’). The product and the plasmid vector pTRC99A were digested with XbaI and HindIII and then ligated together. The resulting plasmid was transformed into TOP10 cells. Colony PCR identified colonies that contained the plasmid with the insert, and these colonies were sent for sequencing. Once the sequence was confirmed, the plasmid was electroporated into UTI89 degS::cat cells. The full strain is termed UTI89/pTRC99A-degS degS::cat and is referred to as pdegS.

Growth in Rich Media

UTI89, UTI89 degS::cat and UTI89 degS::cat/pdegS were cultured in a shaker overnight at 37°C in 3 mL LB broth (with 100 μg/mL ampicillin for pdegS). The next morning, cells were sub-cultured 1:100 into fresh LB broth with or without antibiotics as appropriate and grown with shaking until the OD600=1 (approximately 2x109 CFU/mL). The cells were diluted to 2x105 cells/ mL. Approximately 2000 cells were added to 200 μL of LB per well. Strains were plated in triplicate in LB or LB-Amp. Cells were grown at 37°C, and the optical density at 600nm was recorded every 40 minutes for 12 hr. Triplicate wells of LB and LB-Amp served as blanks and controlled for contamination. The experiment was repeated three times.

Transmission Electron Microscopy (TEM)

UTI89 and UTI89 degS::cat cells were grown in LB broth overnight; replicates for each strain were grown under shaking or static conditions. The cultures were centrifuged at 5000 x g for 2 min, and the pellets were resuspended in fresh phosphate-buffered saline (PBS). Samples were then fixed with 1% glutaraldehyde and negative-stained with uranyl acetate before analysis on a JEOL 1200EX transmission electron microscope (JEOL USA, Peabody, MA) (13).

Membrane Isolation

Overnight cultures of mutant and wild-type strains were sub-cultured and grown in LB broth until they reached an OD600 =0.8. Cells were pelleted and the pellets were stored at -20°C. The pellets were resuspended in 10 mM Tris pH 8.0 with 100 mM RNase and DNase. Cells were lysed in the Stansted Pressure Cell Homogenizer (model SPCH-10, United Kingdom). Whole cells were removed by centrifugation at 3000 x g for 15 minutes at 4˚C. The cytosolic and membrane proteins were resuspended in 50 mM Tris. Membranes and cytosol were separated by centrifugation at 33,000 rpm (82,500 x g) for 80 minutes at 4˚C. The pellet containing the membranes was resuspended in 50 mM Tris and centrifuged as before. The supernatant (inner membrane) was removed, and the pellet (outer membrane) was resuspended in 10 mM Tris. All samples were stored at -20°C.

Protein Expression Analysis

SDS-PAGE analysis was used to compare the proteins in whole bacteria, inner membrane, outer membrane, and cytosol from the wild-type and mutant strains. Samples from each fraction were separated on a 12% SDS-PAGE gel. Gels containing whole cell lysates, outer membranes, and cytosol were Coomassie stained for 30 minutes and then destained with 6:3:1 water:methanol:acetic acid until bands could clearly be seen. To visualize the inner membrane proteins of the wild-type and mutant strain, a Pierce Silver Stain Kit was used according to the manufacturer’s instructions (Thermo Scientific).

Bacterial Outer Membrane Vesicle Isolation and Protein
Expression Analysis

Wild-type, degS::cat, and pdegS strains were grown in 100 mL LB overnight, before centrifugation for 10 minutes at 5000 x g at 4°C. The supernatant was filter sterilized with 0.2 μm filter and ultracentrifuged for 2 hours at 40,000 rpm (100,000 x g). The pellet was resuspended in 200 μL of Dulbecco’s phosphate buffered saline (Gibco) with 0.2 M NaCl. A BCA protein assay was performed according to the manufacturer’s instructions (Pierce). The samples were separated by SDS-PAGE and silver stained to compare the proteins in the OMVs.

Antibiotic Susceptibility

UTI89 and UTI89 degS::cat were cultured in a shaker overnight at 37°C in 5 mL LB broth. Strains were sub-cultured the next morning and grown to an OD600=1. The cells were diluted 1:10 in sterile PBS. The cultures were then spread evenly across the entire plate with a sterile swab. Using sterile tweezers, three antibiotic-impregnated disks were placed onto each agar surface. A single strain and antibiotic combination was used per plate. The antibiotics for each disk were as follows: ampicillin (10 μg/disk), chloramphenicol (30 μg/ disk), kanamycin (30 μg/disk), novobiocin (30 μg/disk), spectinomycin (100 μg/disk), and tetracycline (30 μg/ disk). The plates were incubated overnight at 37°C, and the zone of inhibition around the disk was measured. The experiment was repeated three times.

Mouse Infections

C57BL/6J (Jackson Laboratories, Bar Harbor, Maine) mice were used for infections; all mice were female and aged-matched at 8-10 weeks. UTI89 and UTI89 degS::cat were grown in 20 mL LB broth overnight under static conditions. Catheters were made with 30 gauge needles and 0.28 mm polyethylene tubing and sterilized with UV light in a tissue culture hood overnight. Overnight cultures were spun down at 7000 rpm for 10 min, resuspended in sterile PBS to OD600=1, and diluted 1:1 with PBS for the inoculum. Under isoflurane anesthesia, three mice per strain were infected transurethrally with 50 μL of bacteria (~1 x 107 bacteria). Bladders were harvested 24 hours post-infection (hpi) using sterile technique, homogenized, and plated in order to count colony forming units. The experiment was performed three times.

Confocal Microscopy

Mice were infected with 50 μL of UTI89/pcomGFP or UTI89 degS::cat/pcomGFP (2 mice/bacterial strain) as described above. Mice were sacrificed at 16 hpi and bladders were harvested, bisected, and stretched. Bladders were fixed in 2.5% paraformaldehyde in PBS, permeabilized with 0.1% Triton X-100 in PBS, and stained with Syto 61 nucleic acid stain (1:1000 in PBS, Life Technologies). Stained bladders were washed 3 times with PBS, mounted on slides with ProLong Gold antifade reagent (Invitrogen), and imaged on a Zeiss LSM510 inverted confocal microscope.

Eukaryotic Cell Culture

Human bladder epithelial cells (strain 5637, ATCC HTB-9) were maintained in RPMI 1640 medium (Gibco) supplemented with 10% fetal bovine serum (Sigma). Cells were grown at 37°C in a humidified atmosphere with 5% CO2.

In vitro Binding and Invasion Assays

Binding and invasion assays were performed as described previously (14), with modifications. Four strains were compared: UTI89, degS::cat, pdegS, and UTI89 fimH::kan (negative control). Overnight static cultures of each strain were centrifuged and resuspended in sterile PBS to OD600=1. Binding, input, and invasion wells with confluent 5637 cells were prepared in triplicate for each strain in 24-well tissue culture plates (TPP). Old medium was aspirated and 1 mL of fresh medium was added to each well. Cells were infected by adding 10 μL of the desired bacterial suspension to each well. Plates were spun at 1500 rpm for 3-5 minutes at room temperature and incubated in a tissue culture incubator for desired length of time. Binding and input wells were removed from the incubator after 30 min. Supernatants were removed from the binding wells, and cells were washed five times with 1X PBS before lysis in 0.1% Triton X-100 (in PBS) at room temperature for 15 min. Input wells were not washed before cells were lysed with 10 μL of 10% Triton X-100. Each well was scraped and examined under the microscope to ensure that all cells were released. Lysed cells and bacteria were serially diluted and plated on LB agar plates to determine CFU. The invasion plates were removed from the incubator after 60 min, and wells were washed once with 1X PBS. Gentamicin (10 μg/mL in media) was added to each well, and the plate was returned to the incubator for 1.5 hours. After gentamicin treatment, wells were washed three times with 1X PBS before lysis in PBS with 0.1% Triton X-100. Lysed cells and bacteria were scraped, serially diluted, and plated. The LB agar plates were incubated overnight at 37˚C and CFU were counted the following day.

Statistical Analysis

Significance was determined by a Student t-test, with values less than 0.05 considered significant. For the mouse infections, significance was determined by the Mann-Whitney U statistical analysis as performed by GraphPad Prism 5 software.

Acknowledgements

Cardinal Ritter College Prep High School (ICB), Young Scientist Program Summer Focus (ICB), Division of Biology and Biomedical Sciences, Washington University in St. Louis (ESD)

We would like to thank Dr. David Hunstad for the lab space and supplies that were used for this project (NIH grants DK090752 and DK064540). We thank David Hunstad, Lisa Drewry, and Natalie Omattage for critical reading of the manuscript. We would also like to thank Wandy Beatty for sample preparation and assistance with TEM. I.C.B. was supported by the Summer Focus program of the Young Scientist Program at Washington University School of Medicine. E.S.D. is supported by a National Science Foundation Graduate Research Fellowship (DGE-1143954).

References

1. Foxman, B. The epidemiology of urinary tract infection. Nature Reviews Urology 7(12), 653-60 (2010).

2. Wiles, T.J, Kulesus, R. R, Mulvey, M.A. Origins and virulence mechanisms of uropathogenic Escherichia coli. Experimental and Molecular Pathology 85(1), 11-9 (2008).

3. Hannah, T.J, Totsika, M, Mansfield, K.J, Moore, K.H, Schembri, M.A, Hultgren, S.J. Host-pathogen checkpoints and population bottlenecks in persistent and intracellular uropathogenic Escherichia coli bladder infection. FEMS Microbiology Reviews 36(3), 616-48 (2012).

4. Waller, P.R, Sauer, R.T. Characterization of degQ and degS, Escherichia coli genes encoding homologs of the DegP protease. Journal of Bacteriology 178(4), 1146-53 (1996).

5. Meltzer, M, Hasenbein, S, Mamant, N, Merdanovic, M, Poepsel, S, Hauske, P, Kaiser, M, Huber, R, Krojer, T, Calusen, T, Ehrmann, M. Structure, function and regulation of the conserved serine proteases DegP and DegS of Escherichia coli. Research in Microbiology 160(9), 660-6 (2009).

6. Alba, B.M, Zhong, H.J, Pelayo, J.C, Gross, C.A. degS (hhoB) is an essential Escherichia coli gene whose indispensable function is to provide σE activity. Molecular Microbiology 40(6), 1323-33 (2001).

7. McBroom, A.J, Kuehn, M.J. Release of outer membrane vesicles by Gram-negative bacteria is a novel envelope stress response. Molecular Microbiology 63(2), 545-558 (2007).

8. Deatherage, B.L, Cookson, B.T. Membrane vesicle release in bacteria, eukaryotes, and archaea: a conserved yet underappreciated aspect of microbial life. Infection and Immunity 80(6), 1948-57 (2012).

9. Ellis, T.N, Kuehn, M.J. Virulence and immunomodulatory roles of bacterial outer membrane vesicles. Microbiology and Molecular Biology Reviews 74(1), 81-94 (2010).

10. Keuhn, M.J, Kesty, N.C. Bacterial outer membrane vesicles and the host-pathogen interaction. Genes and Development 19(22), 2645-55 (2005).

11. Mulvey, M.A, Schilling J.D, Hultgren S.J. Establishment of a persistent Escherichia coli reservoir during the acute phase of a bladder infection. Infection and Immunity 69(7), 4572-9 (2001).

12. Murphy, K.C, Campellone, K.G. Lambda Redmediated recombinogenic engineering of enterohemorrhagic and enteropathogenic E. coli. BMC Molecular Biology 4(11), doi: 10.1186/1471- 2199-4-11 (2003).

13. Danka, E.S, Hunstad, D.A. Cathelicidin Augments Epithelial Receptivity and Pathogenesis in Experimental Escherichia coli Cystitis. Journal of Infectious Diseases doi: 10.1093/infdes/jui577 (2014).

14. Nicholson, T.F, Watts, K.A, Hunstad D.A. OmpA of uropathogenic Escherichia coli promotes postinvasion pathogenesis of cystitis. Infection and Immunity 77(12), 5245-51 (2009).

15. Rosen, D.A, Hooten, T.M, Stamm, W.E, Humphrey, P.A, Hultgren, S.J. Detection of intracellular bacterial communities in human urinary tract infection. PLoS Medicine 4(12), e329 (2007).

16. Dhakal, B.K, Mulvey, M.A. Uropathogenic Escherichia coli invades host cells via an HDAC6- modulated microtubule-dependent pathway. Journal of Biological Chemistry 284(1), 446-54 (2009).

17. Redford, P, Roesch, P.L, Welch, R.A. degS Is Necessary for Virulence and Is among Extraintestinal Escherichia coli Genes Induced in Murine Peritonitis. Infection and Immunity 71(6), 3088-96 (2003).

18. Redford, P, Welch, R.A. Role of Sigma E-Regulated Genes in Escherichia coli Uropathogenesis. Infection and Immunity 74(7), 4030-4038 (2006).

File Type	File Name	Attachment	Description
	UPEC1.jpg
	UPEC2.jpg
	UPEC3.jpg
	UPEC4.jpg
	UPEC5.jpg
	UPEC6.jpg
	UPEC7.jpg
	UPEC8.jpg

Step Files:

File Name	Submitted By	Date Created
UPEC	JEI Admin	07/10/2015
UPEC1.jpg	JEI Admin	07/10/2015
UPEC1.jpg	JEI Admin	07/11/2015
UPEC2.jpg	JEI Admin	07/11/2015
UPEC3.jpg	JEI Admin	07/11/2015
UPEC4.jpg	JEI Admin	07/11/2015
UPEC5.jpg	JEI Admin	07/11/2015
UPEC6.jpg	JEI Admin	07/11/2015
UPEC7.jpg	JEI Admin	07/11/2015
UPEC8.jpg	JEI Admin	07/11/2015

Characterization of a UPEC degS mutant in vitro and in vivo by India C. Bradley and Elizabeth S. Danka