Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: Adv Exp Med Biol. 2021;1323:81–90. doi: 10.1007/5584_2020_573

Synovial fluid mediated aggregation of clinical strains of four enterobacterial species.

Alicia Macias-Valcayo 1,, Amelia Staats 2,5,, John-Jairo Aguilera-Correa 1, Jack Brooks 2, Tripti Gupta 2, Devendra Dusane 2,, Paul Stoodley 2,3,4,, Jaime Esteban 1
PMCID: PMC7882635  NIHMSID: NIHMS1625274  PMID: 32797406

Abstract

Septic arthritis and prosthetic joint infection (PJI) are conditions commonly associated with Gram-positive cocci, however, a drastic increase in cases derived from enterobacterial species has recently been observed. Recently it has been reported by multiple groups that staphylococci rapidly form free-floating aggregates in the presence of synovial fluid. These aggregates are comparatively more resistant to antimicrobial challenge than their planktonic counterparts, and thus may play a role in the pathogenesis of joint infection. While staphylococcal aggregates have been the primary focus of interest in the field, it is unclear just how widespread synovial fluid mediated aggregation (SFMA) is in Gram negative enterobacteria (GNE). Through this work we have evaluated SFMA in clinical GNE isolated from PJIs. Two PJI clinical strains each of Enterobacter cloacae, Escherichia coli, Klebsiella pneumonia and Proteus mirabilis strains representing a range of antibiotic susceptibilities were exposed to 10% bovine synovial fluid supernatant (BSF) using a relatively simple, quick semi-quantitative method using an imaging plate reader. BSF stimulated aggregation within 0.5 hr both strains of E. cloacae and P. mirabilis and one strain of E.coli. In one strain of P. mirabilis and E.coli, the size of the aggregates significantly increased from 0.5 to 2 hr exposure. In contrast, neither K. pneumoniae strain aggregated in BSF. These preliminary findings show that aggregation can occur quickly in GNE, but the extent appears strain and species specific. Further work is required to assess the impact of SFMA on antibiotic tolerance, host innate immunity and the establishment of biofilms.

Keywords: Septic arthritis, Enterobacteria, aggregates, biofilm, prosthetic joint infection, rapid screen

1. Introduction

The frequency of both native and prosthetic joint infections is rapidly increasing (Nair et al. 2017). There are many factors contributing to this rise including an aging population, the growing prevalence of degenerative joint diseases, immunosuppressive treatments, and the use of invasive joint operations (Salar et al. 2014; Kaandorp et al. 1997). Septic arthritis (SA) is an invasion of a joint by an infectious agent resulting in localized inflammation. This pathology is infrequent, but it is considered a medical emergency associated with substantial morbidity and mortality (Margaretten et al. 2007). A recent retrospective study following the outcomes of 10,195 patients with septic arthritis stated that about 7% died within 90 days of receiving an arthroscopic knee washout as treatment, 1% required knee arthroplasty within 1 year and 9% within 15 years of the treatment (Abram et al. 2020). The development of SA can occur through a myriad of sources including hematogenous seeding, introduction of bacteria during joint surgery, or direct extension from a contiguous focus of the joint space by a pathogen microorganism (Shirtliff and Mader 2002; Goldenberg and Reed 1985). The most common microorganisms involved are Gram-positive cocci with Staphylococcus aureus making up 52% of all infections (Ross 2017) whilst Gram-negative bacilli represent about 10 to 20% of cases (Deesomchok and Tumrasvin 1990). SA events associated with Gram-negative bacilli often yield outcomes far more severe than those caused by Gram-positive bacteria (Chiu and Wang 2013). These infections may occur in both the native (non-prosthetic SA) and artificial joint, for example after a knee or hip replacement (Nair et al. 2017). Thus, patients who have artificial joints are at an increased risk of developing a joint infection (Del Pozo and Patel 2009). Recent studies suggest that although Gram-positive cocci represent the dominant source of prosthetic joint infections, the number of cases attributed to Gram-negative bacilli are continuously rising. (Benito et al. 2016). It has been suggested that GNE are the causative agents of 10–23% of all PJI (Rodriguez-Pardo et al. 2014).

PJIs are characteristically difficult to treat (Tande and Patel 2014) due to the ability of microorganisms to form biofilms on a variety of surfaces (Donlan and Costerton 2002). While the process of biofilm formation in the joint space is still being elucidated, multiple groups have shown that exposure to synovial fluid can stimulate bacteria to rapidly form macroscopic aggregates. It has been demonstrated that S. aureus aggregates formed in the presence of either human or bovine synovial fluid display an increased antimicrobial tolerance and resistance to immune system defenses (Gilbertie et al. 2019). Synovial fluid mediated aggregation (SFMA) has also been shown to occur in other bacterial species (Streptococcus zooepidemicus, Pseudomonas aeruginosa and Escherichia coli) stimulated by either equine or porcine synovial fluid (Gilbertie et al. 2019). Aggregates of P. aeruginosa biofilms have been observed in clinical specimens (Howlin et al. 2017) and have been shown to seed biofilm formation in vitro (Kragh et al. 2016). In view of these observations and evidence that biofilm-based infections related to Gram-negative enterobacteria are increasing, additional in vitro investigations are needed to better understand the interactions between these microorganisms and synovial fluid. The aim of the present study was to evaluate bovine SFMA in respect to time and aggregate size of four enterobacteria species isolated from PJI.

In order to semi-quantify the SFMA potential in the GNE clinical strains we first needed to develop a relatively quick, cheap, and simple rapid screen method of assessment. To accomplish this, we based our assay on a multiwall plate platform using an imaging plate reader in which aggregation patterns in the wells could be captured and semi-quantified. Previously we found that the viscosity and heterogeneous composition of 100% bovine synovial fluid posed difficulties in pipetting and also contained fibrous and particulate material which caused optical interference in the plate reader. Therefore, we first evaluated the use of 10% bovine synovial fluid in Ringer’s solution as a sufficient stimulus to demonstrate bacterial aggregation. Prior to testing our Gram-negative strains, conditions and usage of 10% bovine synovial fluid supernatant as an aggregating agent was demonstrated using a S.aureus 1276 strain known to aggregate in the presence of synovial fluid (Pestrak et al. 2020).

2. Materials and methods

2.1. Flow Cytometry

A single colony of GFP-expressing S. aureus strain, AH1726, a constitutive GFP expressing derivative of the USA300 LAC strain (Chiu et al. 2013) which is known to form synovial fluid mediated aggregates (Pestrak et.al. 2020) (Ibberson et al. 2016), was used to inoculate 5mL of Tryptic Soy Broth (BD, Germany). The culture was grown shaking in a 37°C incubator overnight (Innova 44, New Brunswick Scientific). The optical density at 600 nm was measured and 0.75 OD600 of cells from stationary phase cultures was pelleted at 21,000 xg, washed twice and resuspended in 500μL of Ringer’s solution buffer (BR0052G, Fisher Scientific). After the washes, cells were pelleted and re-suspended in 500 μl of Ringer’s solution or 1%, 5%, or 10% whole bovine synovial fluid or bovine synovial fluid supernatant (Lampire Biological Laboratories, Pipersville,PA, USA) in Ringer’s solution. Cells were incubated in the treatment for 1 hour prior to flow cytometry. After the incubation time, 100 μl of the cells was collected from the bottom of the microcentrifuge tube and transferred to a 5 ml round bottom polystyrene tube. As described elsewhere (Pestrak et al. 2020) bacterial aggregation can be quantified using a BD FACsCanto II flow cytometer (BD sciences). In order to exclude synovial fluid debris, only the forward and side scatter of the GFP+ population was quantified. Data collected from the flow cytometer was analyzed using FlowJo 9.0. The population of single cells was determined by gating a population of single bacterial cells in the negative control (no treatment). This population was confirmed previously by microscopy to be planktonic. Calculations to determine the percentage of the population existing as aggregates was accomplished by subtracting the single cell population from the total population. At least 10,000 events were measured during flow cytometry for 3 biological replicates per treatment. Additionally, the median fluorescence intensity (MFI-FSC) of the forward scatter was calculated to determine the relative aggregate size within the population. Statistical significance of variation in aggregation or MFI-FSC between whole bovine synovial fluid and bovine synovial fluid supernatant groups was determined by two-way ANOVA (GraphPad Prism version 5.0b software).

2.2. Antimicrobial Susceptibility Testing

Two bacterial strains of each of four enterobacterial species: Enterobacter cloacae (E.cl7 and E.cl8), Escherichia coli (E.c3 and E.c4), Klebsiella pneumonia (K.p1 and K.p2), and Proteus mirabilis (P.m5 and P.m6) were used in the study. All strains were isolated from samples of patients with PJI in the Clinical Microbiology Department of a metropolitan university hospital. They were frozen at −80°C and subsequently sent to the reference laboratory (Department of Microbial Infection and Immunity, Ohio State University). All PJI were diagnosed according to Infectious Diseases Society of America (IDSA) internationally accepted criteria (Osmon et al. 2012). Antimicrobial susceptibility testing was performed based on VITEK® 2 Systems (Biomerieux, France) and categorized according the EUCAST breakpoints (Table 1).

Table 1.

Details and antibiogram of GNE isolates. Infection type (EPI: Elbow prosthetic infection; KPI: knee prosthetic infection, HPI: hip prosthetic infection) and antibiogram of the clinical isolates. Susceptibility rates were interpreted according to EUCAST breakpoints. Abbreviations: AMK: Amikacin; GEN: Gentamicin; CXM: Cefuroxime; CTX: Cefotaxime; AMC: Amoxicillin/clavulanic acid; CIP: Ciprofloxacin; SXT: Trimethoprim/sulfamethoxazole; IMP: Imipenem; Bold values are considered resistant by EUCAST criteria.

Species E.cloacae E.coli K. pneumoniae P. mirabilis
Strain designation E.cl8 E.cl7 E.c3 E.c4 K.p1 K.p2 P.m5 P.m6
Origin KPI HPI HPI HPI KPI EPI HPI HPI
Antibiogram
AMK ≤2 ≤2 32 ≤2 4 ≤2 ≤2 ≤2
AMC ≥32 ≥32 ≥32 4 ≥32 ≤2 8 4
CTX ≥64 ≤1 ≥64 ≤1 ≥64 ≤1 ≥64 ≤1
CXM ≥64 16 ≥64 4 ≥64 ≤1 ≥64 ≤1
CIP ≥4 ≤0,25 ≥4 ≤0,25 ≥4 ≤0,25 ≥4 ≥4
GEN ≤1 ≤1 ≤1 ≤1 ≥16 ≥16 ≤1 ≤1
IMP 2 ≤0,25 ≤0,25 ≤0,25 2 ≤0,25 ≤0,25 ≤0,25
SXT ≤20 ≤20 ≥320 ≤20 40 ≤20 ≥320 ≤20
Open in a new tab

2.3. Plate Reader Imaging

First, the bovine synovial fluid was centrifuged, and the supernatant was taken. To assess bacterial aggregation formation, an overnight bacterial tryptic-soy broth culture (BD, Germany) was centrifuged and washed three times. 25 μL of the pellet were added to 100 μL of Ringer’s Solution (Sigma Aldrich, Missouri, United States) with or without 10% bovine synovial fluid (Lampire Biological Laboratories, Pennsylvania, United States) following a methodology previous described (Pestrak et al. 2020) in a 96-well plate (ThermoFisher Scientific, Massachusetts, United State). After 0.5 and 2hr incubation at 37°C and 5% CO2, images of the aggregate formation were taken using an imaging plate-reader (SpectraMax i3x, Molecular Devices). The experiment was performed in triplicate. The images were analyzed using ImageJ (O’Brien et al. 2016). Data were statistically evaluated using non-parametric Wilcoxon test. The values were represented as median and interquartile range. As a positive control, the GFP-expressing S. aureus strain, AH1726, which has previously been shown to aggregate in synovial fluid, was utilized. Methods for imaging after 1 hour exposure were followed as described above using 1%, 5%, or 10% whole bovine synovial fluid or bovine synovial fluid supernatant.

3. Results

3.1. Evaluation of Bovine Synovial Fluid Treatments

Regardless of whether whole synovial fluid or synovial fluid supernatant was utilized, a dose-dependent increase in aggregation was observed using plate reader imaging (Figure 1a-1f). Additionally, visually the aggregate size was comparable between the two treatments. These results were corroborated using flow cytometry to quantify the percentage of the population within aggregates after 1 hour of exposure to either whole bovine synovial fluid (Figure 2c) or bovine synovial fluid supernatant (Figure 2d). There was no statistically significant variability in aggregation (Figure 2a) or the median fluorescence intensity size of the forward scatter (MFI-FSC) (Figure 2b) between the two treatment groups.

Figure 1. Plate reader imaging of bovine synovial fluid induced aggregation.

Figure 1.

Open in a new tab

Plate reader imaging of AH1726 was used to observe aggregation 1-hour post exposure to bovine synovial fluid supernatant (1a-1c) or whole bovine synovial fluid (1d-1f) diluted in Ringer’s Solution. Control wells containing Ringer’s Solution (1g) or 10% bovine synovial fluid in Ringer’s Solution (1h).

Figure 2. Synovial fluid supernatant is sufficient to stimulate S. aureus aggregation.

Figure 2.

Open in a new tab

Flow cytometry was used to calculate percent of population within aggregates of AH1726 after 1-hour exposure to whole bovine synovial fluid or bovine synovial fluid supernatant (2a). Aggregation was calculated at 0%, 1%, 5%, and 10% of treatment in Ringer’s Solution. The median fluorescence intensity of the forward scatter (MFI-FSC) was calculated in order to assess the average aggregate size within the sample (2b). Flow cytometry outputs of whole bovine synovial fluid treatments (2c) and bovine synovial fluid supernatant (2d) 1-hour post exposure. Gated area represents single cells within population while anything outside gate indicates aggregation. Error bars indicate mean ± SEM. Statistical significance was determined by two-way ANOVA. Statistical analysis indicates treatment is not a statistically significant source of variation (ns p=0.9935) while treatment concentration is (*** p<0.0001).

3.2. Bacterial Aggregation of GNE

The use of 10 % of bovine synovial fluid favored the bacterial aggregation of enterobacteria in 5 of our 8 strains tested and these aggregates were visible after 2 hr as illustrated by data with E. coli. The presence of synovial fluid significantly increased the size of aggregates of both E. cloacae and P. mirabilis strains and one E. coli strain (E.c4), regardless of exposure time (p<0.05) (Figure 3). The size of aggregates significantly increased from 0.5 to 2 hr of exposure in P. mirabilis P.m5 and P.m6 and E. coli E.co4 (p<0.05). Neither of the K. pneumoniae strains showed evidence of aggregation patterns by our plate assay either with or without synovial fluid and there was no significant difference in aggregate size (p<0.05) (Figure 4).

Figure 3.

Figure 3.

Open in a new tab

GNE isolates visualized by an imaging plate reader (SpectraMax i3x, Molecular Devices, magnification x3) incubated 0.5hr and 2 hrs.at 37°C and 5% CO2 in Ringer solution and with 10% bovine synovial fluid.

Figure 4.

Figure 4.

Open in a new tab

Average size of enterobacterial aggregates in Ringer solution (black) and with 10% bovine synovial fluid of strains of E. cloacae E.cl8 and E.cl 7 (A), E. coli E.co3 and Eco.4 (B), K. pneumoniae K.p1 and K.p2 (C), and P. mirabilis P.m5 and P.m6 (D) at 0.5 and 2 hrs. *:p-value<0.05.

4. Discussion

4.1. Treatment Evaluation

In order to increase consistency in our experiments and minimize the presence of artifact during image analysis, we opted for the use of synovial fluid supernatant over whole synovial fluid. As it has been speculated that aggregation is predominantly mediated by fibrinogen and fibronectin components within the synovial fluid (Pestrak et al. 2020), we hypothesized that the supernatant alone would yield comparable degrees of induced bacterial aggregation. S. aureus strain AH1726 was able to aggregate just as well when stimulated with bovine synovial fluid supernatant as it was with whole bovine synovial fluid. Not only will utilizing the supernatant increase the accuracy of image analysis for our multi-well plate reader assay, but it also allows for greater consistency in future experiments. Whole bovine synovial fluid contains macroscopic bits of floating tissue and particulates which are eliminated after centrifugation, creating a more homogenous media. Utilization of this refined synovial fluid in future in vitro studies will allow for a more accurate analysis of other contributing factors, such as strain specific discrepancies which may influence aggregation.

4.2. GNE Synovial Fluid Induced Aggregation

Five of the eight enterobacterial strains evaluated formed aggregates in presence of synovial fluid within a 2 hr exposure. These results are consistent with previous work, where different grown patterns in synovial fluid were observed for different species (Dastgheyb et al. 2015; Gilbertie et al. 2019; Perez and Patel 2015) . Gilbertie et. al observed that after 24h incubation, non-staphylococcal aggregates were smaller than those formed by S. aureus. In our study, while 10% of synovial fluid was able to increase bacteria aggregation in at least one E. cloacae, E. coli and P. mirabilis strain, K. pneumoniae did not show this characteristic. SFMA reached steady state at 0.5 hr for both E. cloacae strains and the antibiotic-susceptible E. coli strain, suggesting that the rate of formation was also strain dependent.

Aggregation of both P. mirabilis strains increased over time in presence of synovial fluid. This finding implicates the formation of bacterial aggregates as a potential virulence factor of P. mirabilis which may contribute to the difficulty associated with diagnosis and treatment of these infections. Early aggregation may explain how in some cases septic arthritis evolves into a chronic infection despite administration of antibiotic therapy. It remains to be seen how aggregate formation may vary with exposure to other host fluids (such as serum). Additionally, future studies are required to assess the influence on susceptibility to natural immunity and antimicrobials as well as the initiation of biofilm formation.

While we acknowledge that our findings that synovial fluid can mediate aggregation in GNE is somewhat confirmatory of other studies, our work expands on the range of species and strains and demonstrates that synovial fluid mediated aggregation is strain dependent. Another important aspect of our work is the development of a relatively simple and cheap rapid screening technique for synovial fluid mediated aggregation which may be used to further characterize clinical strains allowing possible correlations to be found between aggregation and clinical outcomes.

5. Limitations

While this study displays the capacity of Gram-negative bacteria to aggregate when exposed to bovine synovial fluid, there are limitations to our methodologies which constrain the scope of these findings. First, this research was restricted to a limited number of clinical strains as well as a single type of inoculum growth media. In the future, additional strains need to be included to determine if there are any general correlations in aggregation within and between species and whether antibiotic resistance is related to this phenomenon. Second, we used only 10% bovine synovial fluid supernatant in order to avoid optical interference from fibrous and particulate matter in the synovial fluid. Along with our restriction to relatively short incubation times, it is possible that we may have underestimated the full degree of aggregation and potential size of the aggregates. Therefore, although our rapid screen appears capable of identifying aggregation differences in strains, this methodology may require further refinement. Finally, we used relatively low resolution imaging of the plate reader to semi-quantify SFMA potential through particle size analysis. While we were able to observe clear differences in aggregation patterns in the wells with and without synovial fluid, across 5 of the strains, a higher resolution technique such as confocal microscopy in conjunction with flow cytometry may be required to more precisely size the aggregates. In the future, these more in-depth techniques can be used to “calibrate” the imaging plate reader as a rapid screen method.

6. Conclusions

In conclusion, the ability to form biofilm-like aggregates by enterobacterial species which is mediated by synovial fluid appears to be dependent on strain and time of contact. Further work is required to study the aggregate phenotype and how aggregate formation might play a role in the establishment of GNE biofilm infection in the joint space. Development of a rapid screen assay to semi-quantify SFMA aggregation may be a useful tool to characterize this phenotype in clinical strains and to allow correlations be made between clinical outcomes and aggregation.

Acknowledgments

This work was funded by a grant from the Spanish Society for Clinical Microbiology and Infectious Diseases (SEIMC, COLBETA-19 Project) and NIH R01GM124436 (PS). AM-V was also funded by the SEIMC with a grant for stays in reference centers.

We would like to thank Dr. Alexander Horswill at the University of Colorado Anschutz Medical Campus for providing us with the GFP-expressing USA300 strain (AH1726) used in this study. We would also like to thank Dr. Matthew J. Pestrak for development of the flow cytometry based methodology for quantifying bacterial aggregation.

Contributor Information

Alicia Macias-Valcayo, Email: Aliciamacias_02hotmail.com.

Amelia Staats, Email: Amelia.Staats@osumc.edu.

John-Jairo Aguilera-Correa, Email: john.aguilera@fjd.es.

Jack Brooks, Email: brooks.992@osu.edu.

Tripti Gupta, Email: Tripti.Gupta@osumc.edu.

Devendra Dusane, Email: Devendra.Dusane@osumc.edu.

Paul Stoodley, Email: Paul.Stoodley@osumc.edu.

7 References

  1. Abram SGF, Alvand A, Judge A, Beard DJ, Price AJ (2020) Mortality and adverse joint outcomes following septic arthritis of the native knee: a longitudinal cohort study of patients receiving arthroscopic washout. Lancet Infect Dis 20 (3):341–349. doi:S1473–3099(19)30419–0 [pii] 10.1016/S1473-3099(19)30419-0 [DOI] [PubMed] [Google Scholar]
  2. Benito N, Franco M, Ribera A, Soriano A, Rodriguez-Pardo D, Sorli L, Fresco G, Fernandez-Sampedro M, Dolores Del Toro M, Guio L, Sanchez-Rivas E, Bahamonde A, Riera M, Esteban J, Baraia-Etxaburu JM, Martinez-Alvarez J, Jover-Saenz A, Duenas C, Ramos A, Sobrino B, Euba G, Morata L, Pigrau C, Coll P, Mur I, Ariza J, Infections RGftSoPJ (2016) Time trends in the aetiology of prosthetic joint infections: a multicentre cohort study. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 22 (8):732 e731–738. doi: 10.1016/j.cmi.2016.05.004 [DOI] [PubMed] [Google Scholar]
  3. Chiu IM, Heesters BA, Ghasemlou N, Von Hehn CA, Zhao F, Tran J, Wainger B, Strominger A, Muralidharan S, Horswill AR (2013) Bacteria activate sensory neurons that modulate pain and inflammation. Nature 501 (7465):52–57 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chiu LQ, Wang W (2013) A case of unusual Gram-negative bacilli septic arthritis in an immunocompetent patient. Singapore medical journal 54 (8):e164–168. doi: 10.11622/smedj.2013162 [DOI] [PubMed] [Google Scholar]
  5. Dastgheyb S, Parvizi J, Shapiro IM, Hickok NJ, Otto M (2015) Effect of biofilms on recalcitrance of staphylococcal joint infection to antibiotic treatment. J Infect Dis 211 (4):641–650. doi: 10.1093/infdis/jiu514 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Deesomchok U, Tumrasvin T (1990) Clinical study of culture-proven cases of non-gonococcal arthritis. Journal of the Medical Association of Thailand = Chotmaihet thangphaet 73 (11):615–623 [PubMed] [Google Scholar]
  7. Del Pozo JL, Patel R (2009) Clinical practice. Infection associated with prosthetic joints. The New England journal of medicine 361 (8):787–794. doi: 10.1056/NEJMcp0905029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Donlan RM, Costerton JW (2002) Biofilms: survival mechanisms of clinically relevant microorganisms. Clinical microbiology reviews 15 (2):167–193. doi: 10.1128/cmr.15.2.167-193.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Gilbertie JM, Schnabel LV, Hickok NJ, Jacob ME, Conlon BP, Shapiro IM, Parvizi J, Schaer TP (2019) Equine or porcine synovial fluid as a novel ex vivo model for the study of bacterial free-floating biofilms that form in human joint infections. PLOS ONE 14 (8):e0221012. doi: 10.1371/journal.pone.0221012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Goldenberg DL, Reed JI (1985) Bacterial arthritis. The New England journal of medicine 312 (12):764–771. doi: 10.1056/NEJM198503213121206 [DOI] [PubMed] [Google Scholar]
  11. Howlin RP, Cathie K, Hall-Stoodley L, Cornelius V, Duignan C, Allan RN, Fernandez BO, Barraud N, Bruce KD, Jefferies J (2017) Low-dose nitric oxide as targeted anti-biofilm adjunctive therapy to treat chronic Pseudomonas aeruginosa infection in cystic fibrosis. Molecular Therapy 25 (9):2104–2116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ibberson CB, Parlet CP, Kwiecinski J, Crosby HA, Meyerholz DK, Horswill AR (2016) Hyaluronan modulation impacts Staphylococcus aureus biofilm infection. Infection and immunity 84 (6):1917–1929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Kaandorp CJ, Krijnen P, Moens HJ, Habbema JD, van Schaardenburg D (1997) The outcome of bacterial arthritis: a prospective community-based study. Arthritis and rheumatism 40 (5):884–892. doi: 10.1002/art.1780400516 [DOI] [PubMed] [Google Scholar]
  14. Kragh KN, Hutchison JB, Melaugh G, Rodesney C, Roberts AE, Irie Y, Jensen PØ, Diggle SP, Allen RJ, Gordon V (2016) Role of multicellular aggregates in biofilm formation. MBio 7 (2) [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Margaretten ME, Kohlwes J, Moore D, Bent S (2007) Does this adult patient have septic arthritis? Jama 297 (13):1478–1488. doi: 10.1001/jama.297.13.1478 [DOI] [PubMed] [Google Scholar]
  16. Nair R, Schweizer ML, Singh N (2017) Septic Arthritis and Prosthetic Joint Infections in Older Adults. Infect Dis Clin North Am 31 (4):715–729. doi:S0891–5520(17)30067–3 [pii] 10.1016/j.idc.2017.07.013 [DOI] [PubMed] [Google Scholar]
  17. O’Brien J, Hayder H, Peng C (2016) Automated Quantification and Analysis of Cell Counting Procedures Using ImageJ Plugins. J Vis Exp (117):54719. doi: 10.3791/54719 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Osmon DR, Berbari EF, Berendt AR, Lew D, Zimmerli W, Steckelberg JM, Rao N, Hanssen A, Wilson WR (2012) Diagnosis and Management of Prosthetic Joint Infection: Clinical Practice Guidelines by the Infectious Diseases Society of Americaa. Clinical Infectious Diseases 56 (1):e1–e25. doi: 10.1093/cid/cis803 [DOI] [PubMed] [Google Scholar]
  19. Perez K, Patel R (2015) Biofilm-like aggregation of Staphylococcus epidermidis in synovial fluid. J Infect Dis 212 (2):335–336. doi: 10.1093/infdis/jiv096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pestrak MJ, Gupta TT, Dusane DH, Guzior DV, Staats A, Harro J, Horswill AR, Stoodley P (2020) Investigation of synovial fluid induced Staphylococcus aureus aggregate development and its impact on surface attachment and biofilm formation. PloS one 15 (4):e0231791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Rodriguez-Pardo D, Pigrau C, Lora-Tamayo J, Soriano A, del Toro MD, Cobo J, Palomino J, Euba G, Riera M, Sanchez-Somolinos M, Benito N, Fernandez-Sampedro M, Sorli L, Guio L, Iribarren JA, Baraia-Etxaburu JM, Ramos A, Bahamonde A, Flores-Sanchez X, Corona PS, Ariza J, Infection RGftSoP (2014) Gram-negative prosthetic joint infection: outcome of a debridement, antibiotics and implant retention approach. A large multicentre study. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 20 (11):O911–919. doi: 10.1111/1469-0691.12649 [DOI] [PubMed] [Google Scholar]
  22. Ross JJ (2017) Septic Arthritis of Native Joints. Infectious disease clinics of North America 31 (2):203–218. doi: 10.1016/j.idc.2017.01.001 [DOI] [PubMed] [Google Scholar]
  23. Salar O, Baker B, Kurien T, Taylor A, Moran C (2014) Septic arthritis in the era of immunosuppressive treatments. Annals of the Royal College of Surgeons of England 96 (2):e11–12. doi: 10.1308/003588414X13814021678196 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Shirtliff ME, Mader JT (2002) Acute septic arthritis. Clinical microbiology reviews 15 (4):527–544. doi: 10.1128/cmr.15.4.527-544.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Tande AJ, Patel R (2014) Prosthetic joint infection. Clinical microbiology reviews 27 (2):302–345. doi: 10.1128/CMR.00111-13 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES