ABSTRACT
Catheter-associated urinary tract infections (CAUTIs) are a serious public health problem and account for approximately 40% of hospital-acquired infections worldwide. Candida spp are a major causative agent of CAUTI (17.8%) – specifically Candida albicans– that has steadily increased to become the second most common CAUTI uropathogen1. Yet, there is poor understanding of the molecular details of how C. albicans attaches, grows in the bladder, forms biofilms, survives, and persists during CAUTI 2. Understanding of the mechanisms that contribute to CAUTI and invasive fungal infection will give insights into the development of more effective therapies, which are needed due to the spread of antifungal resistance and complex management of CAUTI in patients that require a urinary catheter 3. Here, we characterize the ability of five Candida albicans clinical and laboratory strains to colonize the urinary catheter, grow and form biofilm in urine, and their ability to cause CAUTIs using our mouse model. Analysis of C. albicans strains revealed that growth in urine promotes morphological transition from yeast to hyphae, which is important for invasive infection. Additionally, we found that biofilm formation was dependent on the presence of fibrinogen, a protein released on the bladder to promote bladder healing4,5. Furthermore, deletion of hyphae regulatory genes resulted in defective bladder and catheter colonization and abolished dissemination. These results indicate that novel antifungal therapies preventing the morphological transition of C. albicans from yeast to hyphae have considerable promise for the treatment of fungal CAUTIs.
INTRODUCTION
Fungal infections and diseases have become a serious public health concern. Candida, Aspergillus, and Cryptococcus spp. are among the most prominent fungal pathogens contributing to mycoses-related deaths6,7. Of these pathogens, Candida spp. is the prevailing opportunistic pathogen in addition to being the fourth most common hospital-acquired bloodstream infection (candidemia) in the United States6,8. More specifically, Candida albicans is the main contributor to invasive candidiasis, responsible for 50% to 70% of candidiasis cases6,8. As Candida spp. are becoming increasingly resistant to antifungal therapy9, the management of this fungal pathogen is a major challenge for the medical community8.
C. albicans inhabits the gastrointestinal and genitourinary tracts and mucosal membranes along with forming biofilms on dentures, pacemakers, prosthetic joints, and intravenous and urinary catheters10-12. The 2016 National Healthcare Safety Network (NHSN) review found that Candida spp., specifically Candida albicans, have become the second most prevalent uropathogen causing 17.8% of catheter associated urinary tract infections (CAUTIs)1. CAUTIs are the most common hospital-acquired infection (HAI) with more than 150 million individuals acquiring such an infection each year13,14. The use of catheters and subsequent infections are not limited to hospital-care settings. In long-term care facilities such as skilled nursing facilities (SNF), 12.6% of admitted patients use an indwelling catheter. Further, in nursing homes, 11.9% of residents use indwelling catheters15, and ∼50% of nursing home residents with indwelling urinary catheters will experience symptomatic CAUTIs16. Despite Candida spp. prevalence in CAUTIs, we lack a clear understanding of the factors required for Candida pathogenesis during CAUTIs 1,2.
C. albicans, among other fungal species, can grow in yeast, pseudohyphal, and hyphal morphologies17-19. The pseudohyphal and hyphal form have a filamentous structure with the pseudohyphae containing constrictions at sites of septation and hyphae having long tube-like filaments with no constrictions at the site of septation. C. albicans’ ability to undergo morphological changes is a virulence determinant. The hyphae morphology is critical in the spread of infection and results in Candida becoming virulent. When C. albicans is in its hyphae morphology, it can invade epithelial and endothelial cells, resulting in tissue damage and an invasive infection. C. albicans can be converted to its hyphae form via serum, neutral pH, and other environmental conditions such as CO2, and is controlled by the master regulator Enhanced filamentous growth protein 1 (Efg1), among other transcription factors20.
The switch between yeast and hyphal forms is essential to pathogenesis and biofilm formation10,19-22. Most C. albicans infections are associated with biofilm formation, making biofilm formation one of the main virulence traits of candidiasis. The biofilm initiates with the adherence of yeast cells to a surface, forming microcolonies followed by proliferation of hyphae and pseudohyphae21. Maturation of the biofilm occurs when the hyphal scaffold is encased in an extracellular polysaccharide matrix, other carbohydrates, proteins, nucleic acids and lipids23,24. The final step in biofilm formation involves dispersion of non-adherent yeast cells for establishment of colonization and biofilms into the surrounding environment21. The resulting biofilm poses a serious threat to the host as the biofilm cells grow increasingly resistant to antifungal therapy and can evade protection from the host defenses24. New antifungal treatments, such as the echinocandins, that target the β-1,3 glucan component of the cell wall have proven to successfully prevent C. albicans biofilms11,25; unfortunately, recently echinocandin-resistant Candida spp are emerging26,27.
Candiduria (Candida in urine) is primarily associated with patients with predisposing factors including diabetes mellitus, genitourinary structural abnormalities, diminished renal function, renal transplantations, metabolic abnormalities, and indwelling urinary tract catheters or devices28. C. albicans colonization and overgrowth in different infection models have been associated with dysbiosis in the host environment such as alterations in host immunity, stress, and resident microbiota10. Importantly, urinary catheterization disrupts the bladder homeostasis, mechanically compromising the integrity of the urothelium, inducing inflammation, interfering with normal micturition (voiding)29, and disrupting host defenses in the bladder30-32. Hence, we are interested in understanding how these host environment changes, specifically in the bladder, can promote fungal CAUTI.
The inflammation response caused by catheterization exposes cryptic epithelial receptors or recruits host factors that can be recognized by the pathogen, enabling microbial colonization, multiplication, and persistence within the urinary tract30-33. We have found that one of these factors is fibrinogen (Fg), which is released into the bladder to heal damage tissue and prevent bleeding due to catheter-induced inflammation in both mice and humans5,34,35. However, due to the constant mechanical damage caused by the urinary catheter, Fg accumulates in the bladder and deposits onto the catheter at concentrations that increase with prolonged catheterization5,34,36-38. This Fg is used by Enterococcus faecalis and Staphylococcus aureus for colonization of catheters and the bladder through the adhesins Ebp and ClfB, respectively; blocking this interaction resulted in defective to no colonization34,36,39. Similarly, we have found that in Acinetobacter baumannii and Proteus mirabilis CAUTI, interactions with Fg are important40,41; however, the bacterial factors responsible have not yet been described. Consistently, in an ex vivo study we showed that a C. albicans CAUTI isolate binds to urinary catheters via Fg5. This strongly suggests that urine is inducing changes in the expressed Fg-binding adhesins, prompting us to study this interaction more carefully in C. albicans.
In this study, we characterized five C. albicans clinical and laboratory strains’ growth in urine, biofilm formation, and capability to cause CAUTI. We found that for the C. albicans strains that grew poorly in urine, growth was promoted when supplemented with a nitrogen source. Urine itself promoted morphological changes, and biofilm formation was enhanced by the presence of protein and Fg in urine conditions. Furthermore, C. albicans was able to infect the catheterized bladder and form a biofilm on the urinary catheter; importantly, a hyphae-deficient mutant was unable to cause CAUTI, suggesting that filamentation was critical for this infection model. Based on these findings, we conclude that the morphological change of C. albicans, which is promoted by urine and the catheterized environment, is crucial to colonize, persist, and cause CAUTIs.
RESULTS
C. albicans survives and grows in urine
C. albicans is able to infect a wide range of human sites including oropharyngeal, gastrointestinal, intra-abdominal, skin, genital, and urinary tract, demonstrating its plasticity to survive and replicate in different host environments42,10,11,13,14,20. Since C. albicans has become a prominent CAUTI pathogen, we explored its ability to grow in urine using clinical and laboratory strains. We used three urinary clinical isolates Pt62, Pt65, and PCNL15,43 (obtained from Washington University School of Medicine) and two well-characterized laboratory strains, DAY230 and SC5314 (Table S3). Since hyphal formation is important for promoting disease, we used a hyphae defective mutant in the SC5314 background, efg-1Δcph-1Δ, to test whether hyphal formation is important in the catheterized bladder environment (Table S3). We tested their ability to grow in a variety of environments ranging from nutrient rich to restrictive conditions. For rich environment, we used yeast extract peptone dextrose (YPD) and brain and heart infusion (BHI) in shaking and static conditions; YPD is a standard C. albicans growth media and BHI was used since the clinical strains were isolated on this media. Static growth was used to mimic the bladder environment and shaking growth was used as a comparison with standard lab culture conditions. For restrictive environment, we used human urine and to further mimic the plasma protein extravasation in the catheterized bladder37, urine was supplemented with either 3% human serum or different nitrogen sources including amino acids (AA), bovine serum albumin (BSA), or Fg. Serum albumin and Fg are two of the most abundant host proteins on catheters retrieved from human and mice and have shown to be used by other uropathogens as a nutrient source34,36. Samples were taken at 0 hours, 24 hours, and 48 hours to assess growth by enumeration of colony forming units (CFUs). As expected, C. albicans clinical and laboratory strains grow in higher densities in rich media and aeration while growth in rich media in static conditions was similar to the urine conditions (Fig. 1). In restrictive environment, C. albicans strains were able to survive and replicate in urine, but it varied widely between the strains and supplementation. Human serum supplementation promoted growth of all strains by 24 hrs with a subsequent decline, possibly because all nutrients were consumed (Fig. 1). The growth of the lab strains and Pt62 was not enhanced, or was inhibited, with BSA or Aa supplementation when compared with urine alone (Fig. 1A, D-F). However, Fg enhanced growth of all strains, at different magnitudes when compared with urine alone (Fig. 1), except for PNCL1, which already exhibited good growth in urine alone (Fig. 1C).
Urine conditions promotes C. albicans hyphal formation
Environmental conditions induce C. albicans morphological changes that are associated with virulence10,19. C. albicans can exhibit different morphologies21, but the main ones are vegetative yeast, pseudohyphae, and hyphae19. Hyphae morphology leads to the spread of infection and increased virulence of the pathogen20,22. Therefore, we wanted to determine how the bladder environment affects C. albicans morphology. To assess the pathogen morphologies, C. albicans’ laboratory and clinical strains were grown in urine with 3% human serum to mimic plasma protein extravasation in the catheterized bladder37,38. YPD alone was used as a negative control and when supplemented with serum was used as a positive control for inducing hyphal morphology44-46. Strains were incubated at 37°C with 5% CO2 for 48 hrs and samples were collected at 0, 24, and 48 hours. C. albicans strains were stained with calcofluor white to assess morphology using fluorescence microscopy (Zeiss Axio Observed inverted scope). The cell morphology was analyzed automatically using CellProfiler software (available from the Broad Institute at www.cellprofiler.org)47 to quantify the percentages of yeast, pseudophyphal, or hyphal forms based on the circularity value of each outlined cell (Fig. 2G, Table S1). All strains showed predominantly yeast morphology in YPD media and YPD with serum induced pseudohyphal and hyphal formation in all strains, except Pt65 and SC5314 efg-1Δcph-1Δ. Notably, our analysis showed that urine conditions promote pseudohyphal and hyphal formation in all strains (Fig. 2, Table S1) except SC5314 efg-1Δcph-1Δ (Fig. 2F, Table S1). Pseudohyphal and hyphal morphologies were further induced when urine was supplemented with human serum in Pt62, PNCL1, DAY230, and SC45314 but not in Pt65. This suggests that the catheterized environment triggers C. albicans morphological change from the yeast cell form to hyphae.
Fibrinogen enhances C. albicans’ biofilm formation
Scanning electron microscopy analyses of C. albicans biofilms on urinary catheters from CAUTI patients2,43 and rats48,49 have revealed pseudohyphal and hyphal morphology. During candidiasis, C. albicans pseudohyphal and hyphal formation induces expression of virulence genes including host adhesion factors18,21,50,51. One of those factors is Mp58, a Fg-binding protein4,52. In an ex-vivo study, we showed that C. albicans PNCL1 clinical isolate colocalized with Fg-deposited on urinary catheters retrieved from a patient with preoperative negative urine culture5. Fg deposition has been shown to be a platform for biofilm formation by diverse uropathogens5,34,36,41,53, suggesting that Fg may be an important factor to promote C. albicans CAUTI pathogenesis. Based on these and our previous findings (Fig. 2), we hypothesized that urine conditions induce factors responsible for Fg-binding and biofilm formation. Thus, we assessed biofilm formation under rich (YPD and BHI) and restrictive conditions (human urine) and compared between BSA- and Fg-coated microplates as we have previously described54. At 48 hrs, immunostaining analyses were performed to assess fungal biofilm formation by using anti-Candida antibodies and biofilm biomass was quantified by fluorescence intensity54. We found that for the clinical strains, Fg promoted biofilm formation in all conditions but Fg-dependent biofilm formation was further enhanced in human urine condition (Fig. 3A-C). For the laboratory strains, we found a similar Fg-dependent biofilm formation in YPD, BHI, and urine, but Fg had no effect in DAY230 when grown in BHI (Fig. 3D and E). In contrast, SC5314 efg-1Δcph-1Δ was not able to form biofilms in human urine regardless of the coated surface (Fig. 3F), highlighting the importance of filamentation for biofilms under urine conditions. Notably, we observed that the hyphae-defective mutant was able to form Fg-dependent biofilms in YPD and BHI (Fig. 3F), suggesting that the mechanisms of biofilm formation are different. This difference could be related to adhesins that are expressed during yeast form that may contribute to biofilm but not in urine conditions. Therefore, using conditions that closely mimic the in vivo environment are important to identify physiologically-relevant determinants for biofilm formation.
Fibrinogen promotes C. albicans biofilm formation
Furthermore, we analyzed the C. albicans strains biofilm by immunofluorescence (IF) microscopy. C. albicans strains’ biofilms were grown for 48 hrs in human urine using glass-bottom petri dishes coated with BSA or Fg. Biofilms on BSA were barely monolayers or small aggregates composed of yeast, pseudohyphae, and hyphae (Fig. 3G-K), except for SC5314 efg-1Δcph-1Δ, where all cells were in yeast form (Fig. 3L). On the other hand, Fg promoted a robust biofilm in all strains when compared with BSA-dependent biofilms (Fig. 3G-K); except for the hyphae-deficient mutant where the colonization was composed of small yeast aggregates similar to the colonization on BSA-coated surfaces (Fig. 3L), suggesting filamentation is important for Fg-dependent biofilm formation.
Furthermore, in the damaged tissue environment, like the catheterized bladder, Fg is converted into fibrin fibers or nets to stop bleeding and allow healing. Therefore, we explored the fungal-fibrin interactions in vitro. Fibrin fibers and nets were formed in glass-bottom petri dishes by adding thrombin to soluble Fg and incubating at 37° C for an hour; then, C. albicans strains were added and incubated for 48 hrs at 37° C in human urine. Visualization of the interaction was done by confocal microscopy and 3D reconstruction at 10x and 40x. We found that most of the cells had a pseudohyphal and hyphal morphology surrounding and going through the fibrin fibers and nets (Fig. 4A-E), except for SC5314 efg-1Δcph-1Δ (Fig. 3F).
C. albicans hyphal formation is critical for establishment of CAUTI
Based on our previous results, we hypothesized that a hyphae-deficient mutant of C. albicans would have a defective colonization in the catheterized bladder. To test this, we assessed the ability of the clinical and laboratory strains to colonize and form biofilm on the urinary catheter using our established CAUTI mouse model34,36-41,55-57. Furthermore, to get a better understanding of the contribution of the catheterized bladder environment to C. albicans’ colonization, we carried out infections in catheterized and non-catheterized mice. Mice with or without catheters were challenged with 1×106 CFU of each strain grown in YPD media overnight at 37° C under static conditions. After 24 hours post infection (hpi), the mice were euthanized and their organs and catheters (when catheterized) were harvested to quantify colonization by CFU enumeration. The results showed that catheterization significantly increased colonization on the bladder of Pt62, Pt65, PNCL1, DAY230 and SC5314 and the catheter and bladder were colonized to the same extent (Fig. 5). Importantly, we found that the hyphae-deficient mutant, SC5314 efg-1Δcph-1Δ, had a significantly defective bladder and catheter colonization when compared with the SC5314 wild-type (WT) strain (p-value < 0.005; Fig. 5E). Interestingly, in the absence of a catheter, the hyphae-deficient mutant behaved and colonized to the same extent as the WT strain.
In human infection, the incidence of candidemia and systemic dissemination arising from Candida UTI or candiduria Candida in urine are relatively low (1-8% of all candidemia cases)58. However, the prevalence of candidemia due to candiduria increases in critically ill and immunocompromised patients59,60, especially if the patients are undergoing urinary catheterization61. Since our mouse model of CAUTI allows us to assess dissemination, we analyzed fungal burden of kidneys, spleen, and heart after 24 hpi (Fig. 5). We found that urinary catheterization significantly contributes to the fungal spread of DAY230 to the kidneys and spleens (p-value < 0.05; Fig. 5D) and in SC5314, spreads to the kidneys (p-value < 0.005; Fig. 5E). Additionally, colonization of the kidneys by Pt62 and Pt65 was 2-3 logs higher than non-catheterized mice, trending to significance. Furthermore, the hyphae-deficient mutant, SC5314 efg-1Δcph-1Δ, did not show differential dissemination between catheterized and non-catheterized mice (Fig. 5E). This data showed that the changes induced by the presence of a catheter are necessary for fungal colonization of the urinary tract and further demonstrate that hyphal morphology as well as pathways regulated by efg-1 and cph-1 are crucial for establishment of CAUTI.
Interactions with fibrinogen and hyphal morphology are important for C. albicans colonization the catheterized bladder
To further understand the C. albicans’ morphology, its interaction with Fg, and its spatial colonization in the bladder during CAUTI, we performed histological analyses using hematoxylin and eosin (H&E) staining, and also IF microscopy of 24 hpi implanted and infected bladders with 1 × 106 CFU of each C. albicans strain. For the IF analysis, we stained for C. albicans (red), Fg (green), and for neutrophils (white) (Fig. 6, merge images; Fig. S1-S6, single channels). We focused on neutrophils for two reasons: we have shown they are highly recruited into the catheterized bladder37,38,62 and the fact that neutropenic patients are more susceptible to C. albicans and bacterial dissemination from CAUTI59,63-65, suggesting a role in controlling candidemia from candiduria. We found that bladder colonization was so robust that it was visible in the H&E-stained whole bladders (blue arrow heads) (Fig. 6). Consistently, our IF analysis showed the presence of C. albicans’ hyphal and pseudohyphal morphologies in the lumen of the bladder in all clinical and laboratory strains (Fig 6A-D), except for the hyphal mutant, SC5314 efg-1Δcph-1Δ (Fig. 6F). As seen with other uropathogens, including E. faecalis, S. aureus, A. baumannii, and P. mirabilis5,34,36,40,41,53, C. albicans cells in the catheterized bladder are found in close association with Fg (Fig. 6 and Fig. S1-S6).
Importantly, neutrophils were highly recruited into the bladder, specifically in the areas with fungal colonization (Fig. 6 and Fig. S1-S6). We found that C. albicans breached the urothelium, encountering a strong neutrophil response at the site of entry (Fig. 6C-E and Fig. S3-S4). For example, we found that PNCL1 was able to reach the bladder lamina propria, inducing a massive neutrophil recruitment to contain the infection (Fig. 6C and Fig. S3). Pt62 and Pt65 cells were primarily found on the bladder lumen and the fungal cells were interacting with Fg and neutrophils (Fig. 6A-B and Fig. S3-S4). On the other hand, robust fungal colonization and neutrophil recruitment was not observed in the SC5314 efg-1Δcph-1Δ infected bladder (Fig. 6F and Fig. S6). These data demonstrate that in the catheterized bladder, C. albicans is mostly in hyphal and pseudohyphal morphology and it is able to interact with Fg. Furthermore, neutrophils are recruited to control fungal infection.
C. albicans interacts with deposited fibrinogen on the catheter during CAUTI
Based on our in vitro Fg-binding results, we assessed if C. albicans-Fg interaction occurs in vivo on the catheter during CAUTI. Catheters from mice infected with each strain were retrieved 24 hpi and immunostained for C. albicans (red) and Fg (green). Except for the hyphal mutant, we found that all strains form a robust biofilm on the implanted catheter (Fig. 7), colonizing 59% to 79% of the surface of the catheter (Table S2). SC5314 WT showed 78.9 ± 14% colonization of the catheters while SC5314 efg-1Δcph-1Δ catheters’ colonization was 10.4 ± 7.5%, exhibiting a significant defective colonization (Fig. 7F-G, L; Table S2). Our IF analysis showed that C. albicans strains were preferentially binding onto deposited Fg on the catheter (Fig. 7B-G). We then further quantified the percentage of the catheter-colonizing fungal population that was colocalizing with Fg. We found that 75% to 91% of the C. albicans strains’ staining was colocalized with deposited Fg (Fig. 7H-L). Moreover, in the SC5314 efg-1Δcph-1Δ strain, that was only able to colonize 10% of the catheter (Fig. 7G, Table S2), 79% of it was colocalizing with Fg (Fig. 7L). This result further corroborates that hyphal formation is important for a robust biofilm formation during CAUTI and that Fg serves as a platform for catheter colonization in vivo.
DISCUSSION
In this study, we have shown that hyphal morphology and fungal interactions with Fg and fibrin are critical for establishment of CAUTI. We showed that C. albicans is able to survive and grow in urine and supplementation with 3% serum and Fg promotes growth. Importantly, we found that hyphal formation is induced by urine conditions in vitro and C. albicans strains exhibit pseudohyphal and hyphal morphology in vivo during CAUTI. The presence of Fg and fibrin enhances biofilm formation in vitro and in vivo. Thus, the catheterized bladder (consisting of urine with serum protein and high Fg and fibrin) creates the ideal environment for C. albicans to colonize and persist in the host.
Our results showed that hyphal formation was critical for Fg-dependent biofilm formation in urine conditions but not in YPD and BHI media. This seemingly contradictory result is not surprising since we have observed that laboratory growth media do not fully recapitulate conditions found within the host. For example, several factors critical for bacterial biofilm formation in CAUTI are dispensable in in vitro biofilm assays when using conventional laboratory growth media34,38,54-56,66. Moreover, host proteins have been shown to contribute to fungal biofilm formation49,67. A study done by the Andes group found Fg, as well as other host proteins associated with C. albicans biofilms in urinary catheters. retrieved from rats24. These results, taken together with our previous data showing that Fg is accumulated in the bladder and deposited on the catheters 1,5,13, and that Fg promotes fungal biofilm formation in urine conditions and during urinary catheterization (Fig. 3, 5-6), suggest that expression of Fg binding proteins could mediate fungal biofilm formation in CAUTI.
It has been reported that C. albicans encodes a Fg-binding protein, Mp58, which is expressed during candidiasis4,52. Furthermore, other cell surface adhesins such as Agglutinin-like sequence (ALS) glycoproteins, hyphal regulated gene 1 (Hyr1), and hyphal wall protein 1 (Hwp1) have shown to be important in biofilm formation68-72. From these adhesins, ALS1, ALS3, and ALS9 have been shown to bind to Fg in vitro72-74 and structural analyses have shown binding to Fg γ-chain via protein-protein interaction, similar to Clf adhesins in S. aureus74,75. However, the role of Mp58, ALSs, Hyr1, and Hwp1 on binding to Fg and their contribution to catheter and bladder colonization during CAUTI have not been described. These adhesins will be explored in further studies.
Interestingly, Mp58, ALSs, Hyr1, and Hwp1 are specifically expressed during hyphal formation4,52,76-79. Transition between yeast and hyphae is central to virulence and this shift is responsive to the environment10,19,20,22,79. Our results have shown that urine conditions and the catheterized bladder environment induce hyphal formation, which suggest these adhesins and other virulence factors may be expressed during CAUTI. Importantly, we have shown that SC5314 efg-1Δcph-1Δ hyphal mutant exhibited defective biofilm formation and Fg binding in urine condition. Furthermore, it displayed deficient catheter and bladder colonization during urinary catheterization. Therefore, Efg1 and Cph1 downstream targets such as Hwp1, Hwp2, Hyr1, ALS8, and secreted aspartyl proteinases (SAP)-4, −5, −6, and −980,81 may play independent roles in C. albicans CAUTI pathogenesis. Since the Efg1 and Cph1 regulatory networks are known, dissection of their contributions to fungal infection in the catheterized bladder will be further explored.
Hyphal cells are important for the formation of biofilms and tissue invasion70,82-84. This correlates with our observation that strains were in primarily hyphal morphology on the biofilms and we observed hyphal invasion of urothelium and lamina propria during CAUTI. Importantly, massive neutrophil recruitment occurred into the areas of fungal tissue invasion; this was not observed in the SC5314 efg-1Δcph-1Δ hyphal mutant. Studies have shown that neutrophils respond to C. albicans site of entry, responding to epithelial-released cytokines and chemokines in addition to recognizing fungal factors such as SAPs 85-88. Neutrophils are able to phagocytize and kill yeast cells and short hyphae while large hyphae are killed by inducing neutrophil extracellular traps (NETs), which releases DNA, granule enzymes, and antimicrobial peptides 85,88-91. Furthermore, it has been shown that neutropenic patients developed candidemia from candiduria, suggesting that bladder recruited neutrophils are critical to control fungal systemic dissemination59. Our future studies will be focused on understanding the immune cell strategies against the fungal CAUTI and their role in containing the fungal infection in the bladder.
C. albicans occupies many niches in the human body, and morphological changes are associated with the establishment of diseased states. This is most important in the bladder, since it is an open and dynamic system, where urine is constantly passing through. Therefore, in order to establish a successful colonization, adhesion and biofilm formation on the urinary catheter is essential13,34,36,39,92. Our results are consistent with that, Candida biofilms not only ensures colonization but can protect the growing cells from the hostile environment and potentiate establishment of the infection93,94. Targeting these hyphal genes so filamentation cannot occur could be a possible therapeutic avenue for preventing C. albicans CAUTIs. Moreover, our results highlight the importance of Fg and fibrin in the process, hence blocking deposition of these proteins onto the catheter might prevent fungal biofilm formation as well. As these fungal pathogens are becoming more commonplace in the healthcare setting, it is essential that the pathogenesis of Candida spp. is better understood in order to decrease the spread of infection and mortality rates. Understanding key characteristics of C. albicans’ for CAUTI pathogenesis is the foundation to understanding and subsequently preventing Candida spp. infections.
MATERIALS AND METHODS
Ethics statement
All animal care was consistent with the Guide for the Care and Use of Laboratory Animals from the National Research Council. The University of Notre Dame Institutional Animal Care and Use Committee approved all mouse infections and procedures as part of protocol number 18-08-4792MD. For urine collections, all donors signed an informed consent form and protocols were approved by the Institutional Review Board of the University of Notre Dame under study #19-04-5273.
Urine Collection
Human urine from at least two healthy female donors between the ages of 20 - 35 were collected and pooled. Donors did not have a history of kidney disease, diabetes, or recent antibiotic treatment. Urine was sterilized with a 0.22 μm filter (VWR 29186-212) and pH was normalized to 6.0-6.5. BSA (VWR 97061-48) supplemented urine was sterilized again using a 0.22 μm filter. When urine was supplemented with Fg (Enzyme Research Laboratories FIB 3), it was added directly to the sterilized urine and the urine was not sterilized after the addition of Fg.
Fungal Culture Conditions
All strains of Candida albicans were cultured at 37 °C with aeration in 5 mL of YPD (10g/L Yeast Extract (VWR J850-500G), 20g/L Peptone (VWR J636-500G), 20 g/L Dextrose (VWR BDH9230-500G)) broth. For in vivo mouse experiments, C. albicans strains were grown static for ∼5 hrs in 5 mL of YPD followed by static overnight culture in human urine.
Growth Curve
Growth curves were performed in glass test tubes (Thermo Fisher Scientific 14-961-29). Overnight cultures (all in stationary phase; measured using a UV/Vis Spectrophotometer) were normalized to ∼1×107 CFU/ml in 1xPBS (Sigma–Aldrich 1002786391). The culture was then diluted (1:1000) into human urine (supplemented with 1 mg/mL BSA, 1 mg/mL Fg, 50X amino acids, or 1 mg/mL human serum), BHI (incubated statically or shaking), or YPD (incubated statically or shaking) and were incubated in the test tube at 37°C for 48 hours. At 0, 24, and 48 hours, samples of each condition were taken and analyzed by CFU counts. All serum donors signed an informed consent form and protocols were approved by the Institutional Review Board of the University of Notre Dame under study #18-08-4834.
Morphology Assay and CellProfiler Analysis
All strains of C. albicans were grown in YPD with or without serum and in human urine with or without serum. At 0, 24, and 48 hours, a sample of each condition was taken, fixed with 10% formalin, and stained with 100 µg/mL of calcofluor. Samples were viewed under a Zeiss inverted light microscope (Carl Zeiss, Thornwood, NY) with the DAPI fluorescent channel. Random images were taken at 100x magnification and processed with CellProfiler. A pipeline (CellProfiler) was created to identify fungal cells and measure the form factor (circularity) of each outlined cell. Based on the form factor value (form factor of a straight line is 0 and form factor of a perfect circle is 1), each cell was assigned to a particular morphology as follows: form factors <0.25, hypha; 0.25 – 0.5, pseudohypha; >0.5, yeast. Details on the pipeline are provided as supplementary materials. Images (consisting of a 3 x 3 tiled region, i.e. 9 fields of view) were randomly acquired and at least three images were analyzed per condition.
The total number of cells per phenotype were summed and divided by the total number of cells to give the overall percentage of each cell type on Microsoft Excel.
Antibodies and dyes used in this study
Primary antibodies
Goat anti-fibrinogen (Sigma-Aldrich F8512), rabbit anti-Candida (ThermoFisher Scientific PA1-27158), and rat anti-mouse Ly6G (BD Pharmingen 551459).
Secondary antibodies
Alexaflour 488-labeled donkey anti-goat (ThermoFisher Scientific SA5-10086); Alexaflour 594-labeled donkey anti-rabbit (ThermoFisher Scientific SA5-10039); Alexaflour 647-labeled donkey anti-rat (ThermoFisher Scientific SA5-10029); IRDye 800CW donkey anti-goat; and IRDye 680LT donkey anti-rabbit. Alexaflour secondary antibodies were purchased from Invitrogen Molecular Probes and IRDye conjugates secondary antibodies from LI-COR Biosciences. Dyes: Hoechst dye (Thermo Fisher Scientific 62249) staining; Hematoxylin and Eosin (H&E) (vector Laboratories #H-3502).
Biofilm Formation in 96-well plates
Biofilm formations were performed in 96 well flat-bottomed plates (VWR 10861-562) were coated with 100uL of BSA or Fg (150 µg/mL) incubated overnight at 4° C. The various strains were grown as described above and the inoculum normalized to ∼1×106 CFU/ml. Cultures were then diluted (1:1000) into YPD, BHI, or human urine. 100uL of the inoculum were incubated in the wells of the 96 well plate at 37°C for 48 hours while static.
Following the 48hr incubation, the supernatant was removed from the plate and washed three times with 200uL 1x PBS to remove unbound fungi. Plates were fixed with 10% neutralizing formalin (Leica 3800600) for 20 minutes and followed by three washes with PBS containing 0.05% Tween-20 (PBS-T). Blocking solution (PBS with 1.5% BSA and 0.1% sodium azide (Acros Organics 447811000)) was added to the plate for one hour at room temperature and then washed with PBS-T (3x). Biofilms were incubated with anti-Candida antibodies diluted into dilution buffer (PBS with 0.05% Tween (VWR M147-1L), 0.1% BSA) for two hours. Plates were washed three times with PBS-T and incubated for one hour with IRDye 680 LT donkey anti-rabbit secondary antibody solution at room temperature and washed with PBS-T (3x). As a final step, the biofilms were visualized by scanning the plates using the Odyssey Imaging System (LI-COR Biosciences) and the analyzed with Image Studio software to obtain the fluorescence intensities (LI-COR Version 5.2, Lincoln, NE).
BSA or Fg-coated dishes and formation of Fibrin fibers/nets
For these assays No. 0 coverglass glass-bottom 35 mm petri dish with a 14 mm microwell (MatTek P35G-0-14-C) were used. The dishes were coated with 150 µg/mL of BSA or Fg overnight at 4° C. For fibrin fiber/nets formation, Fg and thrombin (Sigma-Aldrich T6884-250UN) were thawed at 37° C. 100 µl of 0.5 mg/ml Fg in PBS was added into the microwell glass-bottom and then 10 µl of 2 U/ml thrombin was added to polymerize Fg into fibrin. Dishes were incubated at 37° C for 1 hour and kept overnight at 4° C
Visualization of biofilms and fungal-fibrin interaction
The various strains were grown as described above and the inoculum normalized to ∼1×107 CFU/ml in PBS. These cultures were then diluted (1:1000) into human urine, added to the BSA-, Fg-, or fibrin coated dishes and then were incubated at 37°C for 48 hours under static conditions. After incubation, dishes were then washed three times with 1x PBS to remove unbound fungi, then dishes were fixed with 10% neutralizing formalin solution for 20 minutes and washed with 1x PBS three times. Dishes were blocked with blocking solution was added for an hour at room temperature as described above. Then BSA- and Fg-coated dishes were incubated in primary antibody (rabbit anti-Candida) and for fibrin-coated dishes were incubated with rabbit anti-Candida and goat anti-Fg antibodies. Incubation with the primary antibodies was done for two hours followed by three washes with PBS-T. Then, dishes were incubated for 1 hour with Alexaflour 594-labeled donkey anti-rabbit secondary antibody for BSA- and Fg-coated dishes and Alexaflour 594-labeled donkey anti-rabbit and Alexaflour 488-labeled donkey anti-goat antibodies for fibrin-coated dishes, followed by three washes with PBS-T. BSA- and Fg-coated dishes were visualized with a Zeiss inverted light microscope, and images were taken at different magnifications (10x, 20x, 40x and 100x). Zen Pro and Fiji-ImageJ95 softwares were used to analyze the images. For the fungal-fibrin interaction, fibrin-coated dishes were visualized by Nikon A1-R/Multi-Photon Laser Scanning Confocal Microscope and images were analyzed by IMARIS Image Analysis software and ImageJ software95.
In Vivo Mouse Model
Mice used in this study were ∼6-week-old female wild-type C57BL/6 mice purchased from Jackson Laboratory. Mice were subjected to transurethral implantation of a silicone catheter and inoculated as previously described57. Briefly, mice were anesthetized by inhalation of isoflurane and implanted with a 6-mm-long silicone catheter (BrainTree Scientific SIL 025). Mice were infected immediately following catheter implantation with 50 μl of ∼ 1 × 107 CFU/mL in PBS, of one of the fungal strains introduced into the bladder lumen by transurethral inoculation. Mice were sacrificed at 24 hours post infection by cervical dislocation after anesthesia inhalation and catheter, bladder, kidneys, spleen and heart were aseptically harvested for fungal CFU enumeration. A subset of catheters were fixed for imaging as described below and a subset of bladders were fixed and processed for immunofluorescence and histology analysis as described below.
Catheter Imaging and Analysis
Harvested catheters were fixed for imaging via standard IF procedure as previously described54. Briefly, catheters were fixed with formalin, blocked, washed with 1x PBS, and incubated with the appropriate primary antibodies overnight. Catheters were then incubated with secondary antibodies for two hours at room temperature. Catheters were washed with PBS-T and then a final wash with PBS. Catheters were visualized with the Odyssey Imaging System and then analyzed using color pixel counter from Fiji-ImageJ software95. The number of pixels of each color was compared to the total number of pixels to identify percent coverage of the catheter.
Bladder IHC and H&E Staining of Mouse Bladders
Mouse bladders were fixed in 10% formalin overnight, before being processed for sectioning and staining as previously described53. Briefly, bladder sections were deparaffinized, rehydrated, and rinsed with water. Antigen retrieval was accomplished by boiling the samples in Na-citrate, washing in tap water, and then incubating in 1x PBS three times. Sections were then blocked (1% BSA, 0.3% TritonX100 (Acros Organics 21568-2500) in 1x PBS) washed in 1x PBS, and incubated with appropriate primary antibodies diluted in blocking buffer overnight at 4 °C. Next, sections were washed with 1x PBS, incubated with secondary antibodies for 2 h at room temperature, and washed once more in 1x PBS prior to Hoechst dye staining. Secondary antibodies for immunohistochemistry were Alexa 488 donkey anti-goat, Alexa 550 donkey anti-rabbit, and Alexa 650 donkey anti-rat. Hematoxylin and Eosin (H&E) stain for light microscopy was done by the CORE facilities at the University of Notre Dame (ND CORE). All imaging was done using a Zeiss inverted light microscope. Zen Pro and ImageJ software were used to analyze the images.
Statistical Analysis
Data from at least 3 experiments were pooled for each assay. Two-tailed Mann-Whitney U tests were performed with GraphPad Prism 5 software (GraphPad Software, San Diego, CA) for all comparisons described in biofilm, CAUTI, and catheter coverage experiments. Values represent means ± SEM derived from at least 3 independent experiments. *, P<0.05; **, P<0.005; ***, P<0.0005; ns, difference not significant.
Funding
This work was supported by institutional funds from the University of Notre Dame (to A.L.F.M. and F.H.S.T).
Author contributions
A.L.F.M., and F.H.S.T designed the experiments. A.L.L., M.J.A., A.M., P.S. performed the studies. A.L.L., A.L.F.M, and F.H.S.T. wrote the paper.
Competing interests
The authors declare no competing financial interests.
Acknowledgments
We thank members of the A.L.F.M. and F.H.S.T. laboratories for their helpful suggestions and insightful comments. Especial thank you to Dr. Michael G. Caparon for his comments.
Footnotes
Mailing address: 351 Galvin Life Science Center, Notre Dame, IN 46556, USA. Phone: (574) 631-4964 (ALFM) and (574) 631-4096 (FHST).