Fungal opportunist pathogens can cause life-threatening as well as minor infections, but non-lethal phenotypes are frequently ignored when studying virulence. Therefore, we developed a nematode model that monitors both the survival and reproduction aspects of host to investigate fungal virulence.
While pathogens can be deadly to humans, many of them cause a range of infection types with non-lethal phenotypes. Candida albicans, an opportunistic fungal pathogen of humans, is the fourth most common cause of nosocomial infections which results in ~40% mortality. However, other C. albicans infections are less severe and rarely lethal and include vulvovaginal candidiasis, impacting ~75% of women, as well as oropharyngeal candidiasis, predominantly impacting infants, AIDS patients and cancer patients. While murine models are most frequently used to study C. albicans pathogenesis, these models predominantly assess host survival and are costly, time consuming, and limited in replication. Therefore, several mini-model systems, including Drosophila melanogaster, Danio rerio, Galleria mellonella, and Caenorhabditis elegans, have been developed to study C. albicans. These mini-models are well-suited for screening mutant libraries or diverse genetic backgrounds of C. albicans. Here we describe two approaches to study C. albicans infection using C. elegans. The first is a fecundity assay which measures host reproduction and monitors survival of individual hosts. The second is a lineage expansion assay which measures how C. albicans infection affects host population growth over multiple generations. Together, these assays provide a simple, cost-effective way to quickly assess C. albicans virulence.
Candida albicans is an opportunistic fungal pathogen of humans residing in different niches, including the oral cavity, gastrointestinal, and urogenital tracts1. While typically commensal, C. albicans causes both mucosal and bloodstream infections, the latter of which can be deadly. The severity of C. albicans infection is dependent on host immune function, with immunocompromised individuals more susceptible to infection than healthy individuals1. In addition to host-related factors, C. albicans has several virulence traits which include, hyphae, biofilm formation, and production of secretory aspartyl proteinases (SAPs), which function to promote adhesion and invasion of C. albicans into host epithelial cells2, and candidalysin, a cytolytic peptide toxin3,4. Together, this suggests that C. albicans virulence is a complex phenotype resulting from an interaction between the pathogen and its host environment. Therefore, investigating virulence is best studied using model organisms that serve as host environments, in contrast to in vitro approaches.
Several host models, including both vertebrate and invertebrate organisms, have been developed to study C. albicans infection. The murine model, considered the gold standard, is often used for its adaptive and innate immune system, and ability to monitor disease progression both systemically and in specific organs5. However, there are significant limitations to this host model, including maintenance costs, small number of offspring, and decreased power and reproducibility associated with small sample sizes5. Therefore, other, more simple model organisms such as zebrafish (Danio rerio), fruit fly (Drosophila melanogaster), wax moth (Galleria mellonella), and nematode (Caenorhabditis elegans) have been developed. These non-mammalian model organisms are smaller, require less laboratory maintenance and larger sample sizes allow for greater power and reproducibility compared to murine models. Each of these models have specific advantages and disadvantages that need to be considered when choosing an infection model. G. mellonella offers the most physiologically similar environment to humans as it can be grown at 37 °C and has various phagocytic cells7. Furthermore, this model allows for the direct injection of a specific inoculum7. However, there is no fully sequenced genome, and no established method of creating mutant strains. Similar to G. mellonella, the D. rerio model allows for direct injection of a specific inoculum5,7. It also has both adaptive and innate immune system5, which is unique to this non-mammalian model, yet requires aquatic breeding tanks to maintain. D. melanogaster and C. elegans have similar advantages and disadvantages, which include fully sequenced genomes that are easy to manipulate and generate mutant strains7 but do not have adaptive immunity or cytokines7. Of all these non-mammalian models, C. elegans has the most rapid life cycle, self-fertilize to generate large numbers of genetically identical offspring, and are the most amenable to large-scale screens6,7,8. C. elegans has been extremely powerful for high-throughput screening of antifungal drugs9,10, characterizing virulence factors7, and identifying C. albicans-specific host defense networks11. The innate immune system in C. elegans has multiple components that are highly conserved with humans12. Host innate defenses include production of antimicrobial peptides13 (AMPs) and reactive oxygen species14,15,16.
The severity of C. albicans infection is predominantly measured by host survival but cannot capture non-lethal virulence phenotypes. An often-overlooked aspect of host fitness is reproduction, but several studies suggest that C. albicans impacts reproduction by reducing sperm viability17,18, suggesting that this may be an important aspect of host fitness to study. Therefore, the impact of C. albicans infection on host fecundity is a useful way to study non-lethal virulence phenotypes. We have developed two infection assays using C. elegans to investigate both survival and reproduction phenotypes in healthy hosts19,20. Here we describe both the fecundity and lineage expansion assays. Fecundity measures both progeny produced and survival of single hosts, and lineage expansion assesses the consequences of infection over three host generations. We demonstrate how these assays can be utilized to screen C. albicans deletion mutants to capture both dramatic and subtle differences in lethal and non-lethal virulence phenotypes.
1. Preparatory steps for the experiments
2. Fecundity assay
NOTE: Representative data in shown in Supplementary Table 1 and a schematic in shown in Figure 1A.
For 1 replicate | E. coli (OP50) control condition | C. albicans & E. coli (OP50) treatment condition | |||||
OP50 | H2O | Total | OP50 | C. albicans | H2O | Total | |
Day 0 | 6.25 ul | 43.75 ul | 50 ul | 6.25 ul | 1.25 ul | 42.5 ul | 50 ul |
Days 2-7 | 1.25 ul | 8.75 ul | 10 ul | 1.25 ul | .25 ul | 8.5 ul | 10 ul |
Table 2: Mastermix volumes of E. coli and C. albicans cultures needed to infect nematodes for the fecundity assay.
Day of Experiment (Day 0)
3. Lineage Expansion Assay
NOTE: Representative data in shown in Supplementary Table 2 and a schematic is shown in Figure 2A.
E. coli (OP50) control condition | C. albicans & E. coli (OP50). treatment condition | ||||||
For 1 replicate | OP50 | H2O | Total | OP50 | C. albicans | H2O | Total |
1.25 ul | 8.75 ul | 10 ul | 37.5 ul | 7.5 ul | 255 ul | 300 ul |
Table 3: Mastermix volumes of E. coli and C. albicans cultures needed to infect nematodes for the lineage expansion assay.
Nematode population growth: Day 0
Here we present two assays that measure C. albicans virulence as a non-lethal phenotype using C. elegans as an infection model. The first assay, fecundity, monitors how C. albicans infection impacts single hosts for progeny production and survival. The second assay, lineage expansion, measures how C. albicans infection impacts population growth over multiple generations.
The fecundity assay has multiple measures of host fitness during C. albicans infection. To assess how C. albicans infection impacts two distinct measures of host fitness, survival, and reproduction, we developed the fecundity assay19, which monitors individual hosts. Briefly, single hosts are isolated to control plates (uninfected) or C. albicans treatment plates and monitored every day for seven days for survival, and the number of daily progeny produced (Figure 1A). As this assay encompasses multiple measures of host fitness, it is an efficient way to assess C. albicans virulence. First, we evaluated total brood size by calculating the sum of daily progeny per host. Hosts infected with wildtype C. albicans produce a significantly smaller brood size on average compared to uninfected hosts (Figure 1B, black vs grey bars, p < 0.0001, Kuskal-Wallis test, and post-hoc Dunn's multiple comparison test). By monitoring daily progeny production, we can also detect differences in the timing of reproduction. We previously demonstrated that C. albicans infected hosts have delayed reproduction, with a large fraction of their progeny produced later in their adulthood compared to uninfected hosts19. The fraction of late reproduction is calculated by dividing the progeny produced on Days 4-7 by the total number of progeny produced. Uninfected hosts produce ~20% of their progeny during this late reproduction window whereas wildtype C. albicans infected hosts have 60% of their total offspring in this late reproduction window (Figure 1C).
The fecundity assay not only provides data to assess non-lethal virulence phenotypes, the data can also be used to assess host survival in a seven-day period. We plotted host survival (Figure 1D) and observed a decrease in survival for wildtype C. albicans infected hosts compared to uninfected hosts (grey vs. black lines, Figure 1D), but this difference was not statistically significant (p = 0.687, Log-rank test). This can be attributed to the short seven-day window of time this experiment covers. We previously showed it takes eight days for C. albicans infected hosts to reach 50% mortality19. Thus, this assay can only detect differences in host survival at very early time points.
Fecundity assays capture how C. albicans infection alters host fitness during early adulthood. We used this assay to screen for differences in virulence across reproductive and survival phenotypes using two C. albicans deletion strains (Table 1) previously identified to decrease virulence in murine, D. melanogaster, and C. elegans infection models11,19,23,24,25. CAS5 is a gene that encodes a transcription factor that regulates cell wall homeostasis, adherence, and stress response26. Here, hosts infected with C. albicans cas5Δ/Δ have significantly larger brood sizes and a smaller fraction of late reproduction compared to wildtype C. albicans infected hosts (Figure 1B,C, orange bars, p < 0.0001, Kruskal-Wallis test, and post-hoc Dunn's multiple comparison test). Additionally, hosts infected with C. albicanscas5Δ/Δ do not have significantly different brood sizes or fraction of late reproduction compared to uninfected hosts (p > 0.9999, Kruskal-Wallis test, and post-hoc Dunn's multiple comparison test), indicating that cas5Δ/Δ strains are avirulent. Furthermore, we found that cas5Δ/Δ reduces host mortality compared to wildtype C. albicans (Figure 1D, orange), although this difference is not statistically significant (p = 0.687, Log-rank test), likely due to the short timeframe in which host fitness is evaluated. RIM101 is a gene that encodes a transcription factor required for alkaline-induced hyphal growth27. Hosts infected with rim101Δ/Δ C. albicans had a significantly smaller fraction of late reproduction compared to wildtype C. albicans (Figure 1C, blue bar, p = 0.0184, Kruskal-Wallis test, and post-hoc Dunn's multiple comparison test) infected hosts, despite having similar total brood sizes (Figure 1B, blue bar, p = 0.6979, Kruskal-Wallis test, and post-hoc Dunn's multiple comparison test). Additionally, rim101Δ/Δ infected hosts had similar mortality to hosts infected with wildtype C. albicans (Figure 1D, blue, p = 0.687, Log-rank test). Taken together, we demonstrated the utility of this assay to distinguish subtle difference between C. albicans strains.
To quickly assess how C. albicans infection impacts host fecundity and survival over three generations, we developed the lineage expansion assay19, which monitors the progeny production of individual hosts. Briefly, single hosts are isolated to control plates (uninfected) or C. albicans treatment plates and after seven days the total number of viable progeny in the F1 and F2 generations are counted (Figure 2A). Uninfected hosts produced a progeny population ~35,000 in this timeframe (Figure 2B), compared to wildtype C. albicans infected hosts which produced a progeny population size of ~25,000, nearly a 30% reduction (Figure 2B). This simple assay can be used to rapidly screen through mutant strains of C. albicans. Both cas5Δ/Δ and rim101Δ/Δ C. albicans infected hosts produced progeny populations that were significantly larger than wildtype C. albicans infected hosts (Figure 2B, p < 0.0001 & p = 0.0185 respectively, Tukey's multiple comparisons test). Furthermore, cas5Δ/Δ C. albicans infected hosts produced progeny populations that were comparable to uninfected hosts, suggesting that this C. albicans strain is avirulent, and the rim101Δ/Δ C. albicans strain has reduced virulence for this virulence phenotype.
Figure 1: Fecundity and survival assessed in single hosts infected with different C. albicans strains. A) Experimental schematic of the fecundity assay. B) Total brood size for uninfected (OP50) hosts (n = 23), and hosts infected with wildtype (WT; SN250, n = 21), cas5Δ/Δ (n = 26), and rim101Δ/Δ (n = 26) C. albicans strains. The box represents the interquartile range, the midline indicates the median, and the whiskers represent the range. Error bars are the normalized range of the data. Treatments that share letters are not significantly differ, whereas treatments with differing letters are statistically significant, Kruskal-Wallis test and post-hoc Dunn's multiple comparison test. C) Fraction of late reproduction for uninfected hosts, and hosts infected with wildtype, cas5Δ/Δ, and rim101Δ/Δ C. albicans strains. Bars represent the mean, error bars represent +/- 1 SD, and symbols represent individuals hosts. Treatments that share letters are not statistically significantly different, whereas treatments with differing letters are significant, Kruskal-Wallis test and post-hoc Dunn's multiple comparison test. D) Survival curves of uninfected hosts, and hosts infected with wildtype, cas5Δ/Δ, and rim101Δ/Δ C. albicans strains for the first seven days of adulthood. Error bars represent ± 1 SD. Data from B, C, and D were collected from the same experiment and host sample sizes are the same in each panel. Please click here to view a larger version of this figure.
Figure 2: Virulence of C. albicans strains measured by host reproduction and death over multiple generations A) Experimental schematic of the lineage expansion assay. B) Box and whiskers plot of the population size (representing the number of F1 and F2 progeny) produced within 7 days from a single founder host exposed to OP50 food source alone (uninfected n=10, black), WT C. albicans (SN250, n=12, grey) or C. albicans cas5ΔΔ (n=12, orange), and rim101ΔΔ (n=12, teal) mutant strains (pink). Boxes indicate the 25-75th quartiles with median indicated. Error bars are the normalized range of the data. Treatments that share letters are not significantly different, whereas treatments with differing letters are statistically significant, one-way ANOVA and post-hoc Tukey multiple comparison test. Data was initially published in Feistel et al. 201916. Please click here to view a larger version of this figure.
Strain | Genotype | Source |
SN250 (WT) | his1Δ/his1Δ, leu2Δ::C.dubliniensis HIS1/leu2Δ::C.maltosa LEU2, arg4Δ /arg4Δ, URA3/ura3Δ::imm434, IRO1/iro1Δ::imm434 | Noble et al. (2010) |
rim101∆/∆ | his1Δ/his1Δ, leu2Δ::C.dubliniensis HIS1/leu2Δ::C.maltosa LEU2, arg4Δ /arg4Δ, URA3/ura3Δ::imm434, IRO1/iro1Δ::imm434 orf19.7247Δ::C.dubliniensisHIS1/orf19.7247Δ::C.maltosaLEU2 | Noble et al. (2010) |
cas5∆/∆ | his1Δ/his1Δ, leu2Δ::C.dubliniensis HIS1/leu2Δ::C.maltosa LEU2, arg4Δ /arg4Δ, URA3/ura3Δ::imm434, IRO1/iro1Δ::imm434 orf19.4670Δ::C.dubliniensisHIS1/orf19.4670Δ::C.maltosaLEU2 | Noble et al. (2010) |
Table 1: C. albicans strains used in this study
Supplemental Table 1: Sample Fecundity Data Please click here to download this table.
Supplemental Table 2: Sample Lineage Expansion Data Please click here to download this table.
Here, we present two simple assays that measure fungal virulence. Both assays leverage C. elegans as a host system that includes monitoring for both lethal and non-lethal host phenotypes. For example, fecundity assays investigate the reproductive success of individual infected hosts while also measuring individual survival. The daily monitoring provides not only total brood size, but also reproductive timing, and time of death. The lineage expansion assay was developed as a simplified version of the fecundity assay that is less cumbersome as it requires fewer host transfers and daily counting. The lineage expansion assay provides a multigenerational and quantitative measure that combines multiple aspects of host fitness. Together, these assays are a powerful way to quickly screen the virulence of C. albicans strains, including mutant strains and diverse clinical isolates. Furthermore, given the ease of the infection assay, where hosts are reared on the pathogen as the food source, this makes it easy to apply to other microbial pathogens to assess their virulence.
There are three main technical considerations to be mindful of. First, when synchronizing host populations to collect eggs, the timing of the bleach step needs to be closely monitored, by assessing host movement and/or host shape (dead hosts will no longer be sinusoidal) under a microscope. Once L1-adult hosts are dead, quickly centrifuge and remove the bleach to ensure the integrity of the eggs. Second, for both fecundity and lineage expansion it is critical that L4 hosts are isolated and transferred 48 hours following synchronization (Day 2), as this is the last developmental stage prior to reproductive maturity and non-L4 hosts can shift the reproductive timing. While 48 hours is typically when C. elegans reach L4, this can vary slightly. Third, when counting the final progeny population for lineage expansion, the concentration of the sample is important. If the sample is too dilute, the counts will be too low and cause significant technical variation. The sample concentration can be adjusted by centrifugation and removing excess buffer. If the sample is too concentrated, it will be difficult to count every host in that aliquot and cause significant technical variation. The sample concentration can be adjusted by adding more buffer. Keeping these considerations in mind will ensure the success of these relatively simple assays.
The assays we describe have some distinct differences from other C. elegans infection assays, including the use of solid NGM media and using OP50 as a food source for both uninfected and infected treatments. When hosts are reared only in the presence of C. albicans, their development is slower than when reared in conjunction with E. coli28,29. During our experiments, hosts are exposed to C. albicans throughout development and early larval stages cannot consume large yeast cells.
While this is a highly adoptable system that allows for high replication and quantification, there are a few limitations when using C. elegans to study host-pathogen interactions. First, interactions between the pathogen and host immune function are limited as C. elegans lack an adaptive immune system and pro-inflammatory cytokines and chemokines30. Second, since pathogens are introduced via the C. elegans diet, it is difficult to control the inoculum and ensure that all nematodes ingest the same amount7. However, with fluorescently tagged pathogens, and recent methods to extract yeast31 and bacteria32 from the C. elegans gut, we can enumerate the number of colonies ingested. Finally, C. elegans cannot survive at temperatures similar to the human body, instead are grown at temperatures ≥ 5°C physiologically relevant33. Thus, growth and proliferation of the pathogen will be different in C. elegans compared to humans.
Fungal virulence is predominantly assessed using murine models and is often restricted to monitoring survival. However, using C. elegans as a simple host model for fungal infection offers three unique advantages: First, the ease of infectivity and laboratory handling makes it amenable to scientists of all training levels. Second, the large number of hosts that can be infected and monitored more reliably and confidently captures the phenotypic variation in and between treatments. Third, in C. elegans, the innate immune pathways genes have highly conserved mammalian homologues, including the p38 MAP kinase PMK-113, the TIR-1 (SARM) protein which functions to activate the PMK-1 pathway in C. elegans immunity34, and the dual oxidase BLI-3, which generates reactive oxygen species in response to pathogen infection in C. elegans14,15,16. The conservation of immunity and the availability of different immune mutants at the Caenorhabditis Genome Center, makes it easy to investigate the impact of host immunity on the pathogenicity of C. albicans and other pathogens. We recently showed that these assays can be used with sek-1 hosts to demonstrate that immunocompromised are highly susceptible to fungal infection19,20. Together, these assays our results and the other applications described here offer many reasons to use C. elegans to investigate the virulence of C. albicans.
The authors have nothing to disclose.
We thank Dorian Feistel, Rema Elmostafa, and McKenna Penley for their assistance in developing our assays and data collection. This research is supported by NSF DEB-1943415 (MAH).
1.5 mL eppendorf microtubes 3810X | Millipore Sigma | Z606340 | |
100 mm x 15 mm petri plates | Sigma-Aldrich | P5856-500EA | |
15 mL Falcon Conicals | Fisher Scientific | 14-959-70C | |
50 mL Falcon Conicals | Fisher Scientific | 14-432-22 | |
Adenine | Millipore Sigma | A8626 | |
Agar (granulated, bacterilogical grade) | Apex BioResearch Produces | 20-248 | |
Aluminum Wire (95% Pt, 32 Gauge) | Genesee Scientific | 59-1M32P | |
Ammonium Chloride | Millipore Sigma | 254134 | |
Bacterial Cell Spreader | SP Scienceware | 21TP50 | |
BactoPeptone | Fisher BioReagants | BP1420-500 | |
Disposable Culture Tubes (20 x 150 mm) | FIsherBrand | 14-961-33 | |
Dissection Microscope (NI-150 High Intensity Illuminator) | Nikon Instrument Inc. | ||
E. coli | Caenorhabditis Genetics Center | OP50 | |
Glucose | Millipore Sigma | 50-99-7 | |
Medium Petri Dishes (35 X 10 mm) | Falcon | 353001 | |
Metal Spatula | SP Scienceware | 8TL24 | |
Nematode Growth Media (NGM) | Dot Scientific | DSN81800-500 | |
Potassium Phosphate monobasic | Sigma | P0662-500G | |
Sodium Chloride | Fisher Scientific | BP358-1 | |
Sodium Phosphate | Fisher Scientific | BP332-500 | |
Streptomycin Sulfate | Thermo-Fisher Scientific | 11860038 | |
Tryptone | Millipore Sigma | 91079-40-2 | |
Uridine | Millipore Sigma | U3750 | |
Wildtype C. elegans | Caenorhabditis Genetics Center | N2 | |
Yeast Extract | Millipore Sigma | 8013-01-2 |
.