Shining a light on Candida-induced epithelial damage with a luciferase reporter

ABSTRACT

Host cell damage is a key parameter for research in infection biology, drug testing, and substance safety screening. In this study, we introduce a luciferase reporter system as a new and reliable assay to measure cell damage and validate it with the pathogenic yeast, Candida albicans, as a test case. We transduced human epithelial cell lines with a lentiviral vector to stably express an optimized luciferase enzyme, Nanoluc. Upon cell damage, the release of cytoplasmic luciferase into the extracellular space can be easily detected by a luminometer. We used the luciferase reporter system to investigate the damage caused by C. albicans to different newly generated epithelial reporter cell lines. We found that fungus-induced cell damage, as determined by established methods, correlated tightly with the release of the luciferase. The new luciferase reporter system is a simple, sensitive, robust, and inexpensive method for measuring host cell damage and has a sensitivity comparable to the standard assay, release of lactate dehydrogenase. It is suitable for high-throughput studies of pathogenesis mechanisms of any microbe, for antimicrobial drug screening, and many other applications.

IMPORTANCE

We present a quick, easy, inexpensive, and reliable assay to measure damage to mammalian cells. To this end, we created reporter cell lines which artificially express luciferase, an enzyme that can be easily detected in the supernatant when these cells are damaged. We used infections with the well-investigated fungal pathogen of humans, Candida albicans, as a test case of our system. Using our reporter, we were able to recapitulate the known effects of strain variability, gene deletions, and antifungal treatments on host cell damage. This easily adaptable reporter system can be used to screen for damage in infection models with different microbial species, assay cell-damaging potential of substances, discover new non-toxic antibiotics, and many other damage-based applications.

INTRODUCTION

There are many reasons microorganisms damage host cells, e.g., to acquire nutrients for growth, to invade and colonize different niches, and to protect themselves against immune cells. Thus, infection biologists strive to understand the microbial virulence programs that cause damage to host cells and discover drugs that mitigate this damage. This necessitates easy, reliable, and accurate readouts for cytotoxicity or cell damage in in vitro assays. Many popular high-throughput measures of cytotoxicity (reviewed in references 1, 2) rely on the enzyme activity of cellular components released when host cells die, like lactate dehydrogenase (LDH), adenylate kinase or glyceraldehyde-3-phosphate dehydrogenase. However, these enzymes can also be found in many bacteria, fungi, and parasites, confounding cytotoxicity measurements in infection models. The enzyme assays also generally require multiple wash steps and often expensive and toxic buffers and substrates. In this manuscript, we introduce a high-throughput and inexpensive Nano-luciferase (Nanoluc or Nluc, Promega) reporter for cytotoxicity and examine its use and reliability in Candida albicans infection models as a study of potential applications.

Fungal diseases affect billions of people every year, yet there are only four classes of antifungals available to treat a wide range of fungal infections (3). Candida species are among the most common fungal pathogens and cause tens of millions of mucosal and nearly 700,000 invasive infections annually, with an unacceptable high morbidity and mortality (4). Candida sp. usually resides within humans as harmless commensals. From there, prolonged use of antibiotics or a weakened immune system can trigger mucosal or systemic candidiasis. C. albicans, the most prevalent and well-studied member of the clade, can cause deadly infections and is armed with virulence traits that help it survive and persist in the host. Among others, C. albicans adheres to host cells, invades and damages them, all while evading our immune responses (5). To investigate these virulence traits, readouts for host cell cytotoxicity (i.e., cell damage) have been indispensable. Methods like detection of LDH release (6, 7), release of radiolabeled chromium (⁵¹Cr) (8, 9), and fluorescence microscopy of propidium iodide (PI)-stained nuclei of dying cells (10), have all been used in the past with great success, for example, in the discovery of the fungal peptide toxin candidalysin (11). These assays come with specific drawbacks that hinder their application in high-throughput-screens: LDH cytotoxicity kits are comparatively expensive and inhibited by medium components like phenol red and pyruvate; quantifying fluorescence of nuclear cell-death dyes like PI is confounded by rapidly proliferating microbes in in vitro models that block the light signal; and both techniques are confounded by microbial LDH and nuclear staining. While ⁵¹Cr is highly sensitive and specific, it is also expensive and comes with the dangers and inconvenience of working with radioactivity. New high-throughput and inexpensive host cell cytotoxicity reporters can expedite the discovery of virulence factors and generally simplify cell damage assays.

In the 1880s, the French physiologist DuBois described the molecular basis of luminesce in fireflies, marine animals, and other organisms (12). The reaction of an enzyme called luciferase with its substrate luciferin leads to an emission of visible light (13). With advancements in molecular biology, scientists engineered ways to express luciferase in cells, and ever since, we have found a broad range of applications for luciferase-based assays to study protein expression and for high-throughput drug discovery screening (14–16).

In this manuscript, we designed and validated an epithelial damage reporter based on Nano-luciferase (Nluc) (17). We transduced oral, vaginal, and gut human epithelial cell lines (TR146 and A431) using a lentiviral vector to constitutively and stably express Nluc in the cytoplasm (Fig. 1A). With the underlying assumption that damaged epithelial cells “leak” proteins like Nluc and LDH into the extracellular space, we approximate host cell damage by assaying Nluc and LDH enzymatic activity in the supernatant. We show that using our Nluc assay we can track epithelial damage over the time course of C. albicans infections.

Fig 1.

Nluc is stably expressed in transduced epithelial cells. (A) Schematic representation of lentiviral transduction to generate Nluc epithelial cell lines. The scheme was created using BioRender. (B) Cell lysates from TR146-Nluc were collected over three passages (P1-3), solubilized for SDS-PAGE, blotted and proteins were detected with the indicated antibodies. Representative image of three independent replicates shows that TR146-Nluc cells are stably expressing Nluc. (C) Luminescence of cell lysate correlated highly (R² = 0.93) and significantly with the number of TR146-Nluc cells that were lysed. All experiments were performed in 2–3 independent replicates, where the white circles represent the value from each replicate. One-way ANOVA followed by Benjamini-Hochberg-corrected multiple comparison t-tests comparing each sample mean to the mean luciferase value from 10⁰ cell. Significance levels are represented by *, **, ***, and ****, indicating adjusted P-values of less than 0.05, 0.01, 0.001, and 0.0001, respectively.

RESULTS

Luciferase activity from lysates of stably transduced epithelial Nluc cells correlates with the number of cells

We transduced TR146 and A431 cell lines with lentiviral particles containing a Nluc insert with a constitutively active promoter and a puromycin selection cassette. To confirm that TR146-Nluc cells stably expressed Nluc over time, we collected cells from different passages, lysed them, and performed a western blot with anti-Nluc antibody (Fig. 1B). The stable expression of Nluc in A431-Nluc was similarly confirmed by western blot (Fig. S1). We also noted that luciferase activity highly correlated with the number of cells lysed, and it reliably detected as few as 100 cells (Fig. 1C).

Epithelial Nluc cell lines function as cytotoxicity reporters in C. albicans infection models

We infected TR146-Nluc and A431-Nluc cells with C. albicans (genomic reference wild-type strain for C. albicans, SC5314) for 24 h and measured the Nluc activity in the supernatant. C. albicans-infected cells released significantly more luciferase than uninfected cells (Fig. 2B), showing that our system can detect fungal-induced epithelial cell damage. The Nluc cytotoxicity reporter assay can be performed simply by collecting 5–10 µL supernatant after spinning down the cell debris, diluting the supernatant to 100 µL with PBS, adding the luciferin, and immediately measuring the luminescence (Fig. 2A).

Fig 2.

The Nluc luminescence assay reflects C. albicans (SC5314)-induced epithelial damage. (A) Scheme of Nluc-damage assay depicting transduced epithelial Nluc cells infected with C. albicans. C. albicans invades and damages epithelial cells, releasing cytoplasmic proteins Nluc (blue globules) and LDH (red globules) into the supernatant. The supernatant is collected and diluted for the assay. The luciferase substrate coelenterazine is added to the diluted supernatant, which leads to the emission of light that is measured by a luminometer. Scheme was created using BioRender. (B) Damage to Nluc-expressing vaginal (A431), or oral (TR146) epithelial cells infected with C. albicans was compared to uninfected cells, or cells lysed by Triton X-100, by measuring LDH and Nluc activity of supernatant collected 24 h after infection. Bars depict mean LDH or Nluc activity (log scaled), and error bars depict standard deviation of the mean. Samples were compared to uninfected controls by one-way ANOVA followed by Benjamini-Hochberg-corrected multiple comparison t-tests. Significance levels are represented by *, **, ***, **** denoting adjusted P value < 0.05, 0.01, 0.001, and 0.0001, respectively. All experiments were carried out in three independent replicates. (C) Damage to oral epithelial (TR146-Nluc) cells was estimated by Nluc (turquoise) and LDH (red) assays at different time points after infection with C. albicans (x-axis). Values on the y-axis are presented as Z scores, i.e., they are centered around their respective mean, to allow direct comparison of the two assays. Error bars represent the standard deviation of the mean. All experiments were carried out in three independent replicates.

Next, we wanted to verify that the luciferase reporter cell line TR146-Nluc can recapitulate the temporal dynamics of a C. albicans infection as well as the popular cytotoxicity assay, LDH. To this end, we collected supernatants from infected TR146-Nluc cells at various timepoints up to 24 h and found that the readouts from LDH and luciferase assays were highly in agreement over the entire time course (Fig. 2C).

Applications for Nluc-epithelial cytotoxicity reporter in C. albicans in vitro infection models

Next, we tested the Nluc-based cytotoxicity reporter for typical applications in C. albicans infection biology and compared the luciferase readouts with the conventional LDH assay. As reported previously for such in vitro models, the filamenting C. albicans strain SC5314 caused significant damage to monolayers of oral, vaginal, and gut epithelial cells, but other Candida spp. including C. glabrata, C. auris, and C. parapsilosis did not elicit detectable host cell damage in LDH and Nluc assays (Fig. 3A). C. albicans strains can differ in their damage potential. For example, the C. albicans clinical isolate 101 is known to cause less damage in an oral epithelial infection model, as measured by LDH release, than the genome reference strain SC5314 (18). We were able to confirm this finding with our Nluc damage assay, showing that it is also useful for strain comparisons. C. albicans mutants with deletions for genes or sequences encoding the peptide toxin candidalysin (ECE1-PIII) (7), transcriptional regulators of filamentation (EFG1/CPH1 or EED1) (19–21), or thigmotropism (BUD2) (22) are defective in virulence to different extents. They were attenuated in damaging TR146-Nluc cells (Fig. 3B), and the amount of damage reduction agreed well with the LDH assays done in parallel and with the phenotypes known from literature. Altogether, these experiments show that our epithelial-Nluc cell assay agrees well with the conventional and widely used LDH detection system and that it can be reliably used to screen and test different Candida species, C. albicans strains, and mutants.

Fig 3.

Applications for epithelial cells NLuc damage assays: (A) Damage to TR146-Nluc cells by different Candida species. (B) In both, LDH and Nluc assay C. albicans virulence factor deletion mutants show reduced damage to TR146-Nluc compared to their background strains BWP17 and SC5314. (C) In a commensal vaginal epithelial model, L. rhamnosus (L. rh.) protects A431-Nluc cells from damage against C. albicans infection. (D) Damage to TR146-Nluc cells infected with C. albicans (SC5314) at sub-MIC fluconazole and voricanazole concentrations. Damage to epithelial cells, estimated by LDH and Nluc assays, was normalized to uninfected controls (dotted line). Experiments were performed in two to three independent replicates, with three technical replicates each. The error bars represent the standard deviation of the mean. Means were compared to untreated controls by one-way ANOVA followed by Benjamini-Hochberg-corrected multiple comparison t-tests. Significance levels are represented by *, **, ***, and ****, indicating adjusted P-values of less than 0.05, 0.01, 0.001, and 0.0001, respectively.

Probiotics like Lactobacillus rhamnosus have recently garnered interest due to their ability to attenuate C. albicans virulence (23, 24). We asked if L. rhamnosus colonization can protect A431-Nluc cells from C. albicans-induced damage. We observed a clear trend toward reduced damage induced by C. albicans alone vs damage caused in combination with L. rhamnosus although these differences were not statistically significant (Fig. 3C). This potential reduction was seen in both assays, LDH and Nluc.

Next, we tested whether commonly used antifungals—fluconazole and voriconazole—can protect TR146-Nluc from C. albicans-induced cell damage and found that with Nluc, but not with LDH, we were able to reliably detect the protective effect of antifungals even at sub-MIC levels (Fig. 3D).

Robustness of Nluc and LDH activity at different pH, temperature, and pyruvate concentrations

Supernatants from cell damage assays can be used immediately to measure cytotoxicity, or they can be stored for a while before measurement. To determine how long, and at what temperatures, Nluc and LDH activities are stable and accurate measures of cell damage, we stored TR146-Nluc cell lysate at −25°C, 8°C, and 37°C. We measured the Nluc and LDH activities from aliquots every other day over a week. The Nluc activity declined steeply on the third day of storing the supernatant at −25°C, while LDH activity was already mostly lost directly after freezing (Fig. 4A). Nluc and LDH activities remained mostly stable at 8°C. At 37°C, Nluc activity was lost over the first days, while LDH remained mostly stable for up to a week. Therefore, supernatants collected from epithelial-Nluc cells should be used to detect luciferase activity either immediately or stored at 8°C for up to a week or at −25°C for up to a day.

Fig 4.

Limitations and strengths of the Nluc cytotoxicity assay: (A) Nluc and LDH activity of cell lysate incubated at −25°C, 8°C, and 37°C for 0–7 days. (B) Nluc and LDH activity of cell lysate incubated with PBS buffered at various pH values. (C) Nluc and LDH activity of cell lysates incubated with increasing concentrations of lactate and pyruvate. Experiments were performed in three independent replicates, with three technical replicates each. Error bars represent the standard deviation about the mean, where the mean is represented as a point. Means were compared by one-way ANOVA followed Benjamini-Hochberg-corrected t-tests against the normalized value of 1. Significance levels are represented by * and **, indicating adjusted P-values of less than 0.05 and 0.01, respectively. The color of the symbols shows the results of statistical comparisons for LDH and Nluc, respectively.

Infection of human cells often changes the pH of cell culture medium due to the metabolic activities of the microorganisms. We, therefore, tested the pH range at which Nluc (or LDH) activity is a reliable measure of cytotoxicity by diluting cell lysate in PBS buffered at different pH values and then assaying LDH and Nluc activity (25). We found that Nluc out-performed LDH and gave consistent readings from pH 4 to 9 (Fig. 4B), while the LDH assay was highly pH-dependent. Lastly, we examined the effect of the known LDH inhibitor pyruvate, which is also a common cell culture medium component, on Nluc, and found no discernible effect on Nluc enzyme activity (Fig. 4C).

DISCUSSIONS

In this manuscript, we designed and optimized a luciferase reporter system to measure cytotoxicity (damage) in mammalian epithelial cell lines in an infection model. We used a lentiviral system to transduce immortalized epithelial cells taken from niches where invasive or superficial candidiasis typically occurs. This included the oral epithelial cell line TR146 and vaginal A431. With our simple luciferase assay, we were able to detect a measurable signal from as few as 100 TR146-Nluc cells. The Nluc cytotoxicity assays recapitulated the epithelial damage caused by C. albicans over the course of an infection, as observed by the popular LDH-based cytotoxicity assay. We also tested the application of TR146-Nluc to detect mammalian cytotoxicity by different Candida species, C. albicans mutants, and the protective effects of antifungals. In all assays, we found the Nluc data comparable and, in some cases, even superior to the LDH assay. Finally, we tested the stability of the Nluc and LDH enzymes in cell lysate over a range of temperatures, storage duration, and pH. The results of this study suggest that the luciferase-based assay is a promising new tool for the detection of epithelial cell damage caused by C. albicans infections.

The Nluc cytotoxicity assay detects even low levels of cell damage quickly, reliably, and inexpensively, making it ideal for high-throughput applications. After collecting the sample containing Nluc and diluting it in a simple buffer, the step of adding freshly prepared luciferin can be easily automated, and the resulting luminescence can be measured almost immediately. This contrasts with the time-consuming steps involved in measuring LDH activity from supernatants of samples, which requires the fresh preparation of a complex stabilizing buffer and a catalyst, adding this mixture to the diluted samples, waiting 15–30 min, and then measuring the absorbance. After the initial cost of transducing cell lines, the assay is inexpensive and determined mostly by the cost of coelenterazine, or other luciferase substrates. Luciferase-expressing cell lines created with non-proprietary luciferases (16) can further reduce the cost and be just as effective.

One limitation of using luminescence readouts is the lack of a protein standard to make standard curves, which correlate luminescence signal intensity to protein amount (ng/ mL) for each assay. Once the substrate luciferin is added, the reaction is instantaneous, and the luminescence signal decays over time. This leads to variability between independent replicates. Even with this limitation, in our manuscript, we demonstrated that the Nluc-reporter cell lines can clearly distinguish damaged vs non-damaged samples (Fig. 1B). There are also other ways to overcome the variability between assays. One is to use an automatized substrate dispenser to minimize the time between the addition of substrate and the measurement of luminescence. We can also create a calibration curve for each assay, by lysing a fixed number of cells, serially diluting them, and measuring the luminescence. As seen in Fig. 1C, the number of cells correlates very well with Nluc-luminescence signal, and this way, we were able to estimate the absolute number of cells damaged in each sample.

In our paper, we provided a proof-of-concept for our Nluc cytotoxicity reporter assay for a well-studied fungal pathogen, C. albicans. While non-C. albicans were conventionally thought to be “non-damaging” in epithelial cell-culture infection models (Fig. 3A), recently, it was discovered that adding albumin to the cell-culture medium increased C. glabrata-induced epithelial damage (26). Moving forward, the Nluc-assay can help us dissect the molecular mechanism of non-C. albicans-induced cell damage.

The extensive clinical use of antifungals to treat the increasing population at risk of Candida sp. infections, including people who are frequently hospitalized, the elderly, and the immunocompromised, has led to the emergence of drug resistance in Candida spp., particularly in C. glabrata (27). This, along with the urgent need to develop drugs to treat the emerging multi-drug-resistant C. auris, has led to calls for expedited antifungal drug discovery (reviewed in reference 28).

Among the many practical challenges in developing new antifungal drugs (29; reviewed in reference 30), a major one is the off-target toxicity of antifungal leads because both, infecting fungi and human host are eukaryotes, and many potential fungal targets are conserved between them. Most drug discovery efforts focus on first assaying antimicrobial activity in in vitro growth assays and then studying the side effects of the selected hits on human cells. We found that the Nluc-reporter cell line was sensitive in detecting protection from C. albicans-induced epithelial damage in the presence of subinhibitory levels of antifungals (Fig. 1D). This demonstrates a promising application for the Nluc-reporter cell lines that can be easily adapted to a one-step assay with simultaneous readouts for both mammalian cytotoxicity using the Nluc assay and antimicrobial effects using reporter microbe strains that quantify microbial biomass with microscopy (e.g., Candida-GFP). In this way, a host cell cytotoxicity reporter can expedite the discovery of non-toxic antifungal drugs and antivirulence drugs.

The public health risks have galvanized research and discovery in the Candida field, such that, now, C. albicans is a genetically tractable organism with large collections of gene knock-out libraries (31–33), well characterized in vitro models with a clear host-cell-damaging phenotype (34), and extensively studied antifungal drug interactions (35) (Fig. 3). The Nluc damage reporter can just as easily be applied to study damage to many other cell-types induced by any microorganism, drugs, noxious substances, and other host cells.

We also found that the lentiviral transduction system was easily adaptable for creating multiple Nluc-reporter cell lines. After engineering the appropriate plasmids and harvesting virus-like particles from transfected HEK-293T/17 cells, the viral particles could be stored at −80°C for an extended period and used to create and select new Nluc cell lines as needed, requiring minimal effort.

In this section, we will discuss several strengths of using our Nluc damage-reporter in place of the conventional LDH cytotoxicity assay. For instance, certain bacteria lower the pH of cell culture medium when co-cultured with human cells, and in these cases, LDH release assays underestimate mammalian-cell damage because the lowered pH inhibits LDH (25, 36). Given Nluc’s stability over a wide pH range [Fig. 4B (37)], Nluc-reporter cell lines will be particularly useful in assaying cell damage in such models and better mimic the low-pH physiologically relevant niches (38).

Furthermore, our Nluc-reporter assays damage specific to transduced cells. LDH, an enzyme found all across the kingdom of life, catalyzes pyruvate to lactate in many pathogens, including some fungi (39, 40), many bacteria (41), and protozoa (42, 43). Therefore, when developing infection models to study non-model organisms, it is prudent to use damage assays that do not rely on LDH. Because of these concerns, some groups have relied on ⁵¹Cr release to measure mammalian cell damage inflicted, for example, by C. albicans and Rhizopus delemar (44, 45). However, the use of radioactive materials comes with their own dangers and difficulties and requires specialized training and facilities that are not easily accessible. This makes Nluc-reporter cell lines an attractive alternative for assaying host-cell-specific damage.

Luciferase-expressing cells have also been recently used to quantify damage to transduced cells in models that contain a mixture of mammalian cells, for example, to quantify the cytotoxicity of chimeric antibody-expressing immune cells that kill luciferase-transduced cancer cells (16) and to screen antibodies from patient blood that mediate antibody-dependent cellular toxicity toward cancer cells (46). This approach has also been used to identify small molecules that modulate natural killer cells to kill luciferase-transduced cancer cells more readily (47). In this manuscript, we tested applications for epithelial-Nluc in 2D cell culture models. Historically, these models have provided molecular insights into commensal and pathogenic interactions of fungi with specific cell types (reviewed in reference 34). Finally, the Nluc-reporter cell lines can be used in organ-on-chip or trans-well models that include immune cells which are key players in host-pathogen interactions (48; reviewed in reference 34). Here, our luciferase-reporter cell lines would be able to distinguish between damage to immune and epithelial components.

To summarize, we developed a luciferase-based assay to detect cytotoxicity in mammalian epithelial cell lines. The assay is simple, reliable, inexpensive, and can be used to detect even low levels of cell damage. It is also more specific than the traditional lactate dehydrogenase (LDH) assay, as it can distinguish between damage to transduced cells and damage to other cells in the culture. We tested the Nluc-reporter cell lines extensively in C. albicans infection model and found that the assay could detect the damage caused by the pathogen and that it was comparable to or even better than the LDH assay. While C. albicans served as a good model and case study, our data show that the assay should be applicable with minimal changes to test bacterial, viral, or protozoan infections as well as a range of cell-damaging substances. We also tested the stability of the assay over a range of temperatures, storage durations, and pH. We found that the assay was stable under a wide range of conditions, making it suitable for a variety of applications. Overall, our results suggest that the luciferase-based assay is a promising new tool for the detection of cytotoxicity in mammalian epithelial cell lines.

MATERIALS AND METHODS

Candida strains and cultivation

Candida sp. strains (Table 1), cryopreserved at −80°C, were streaked onto YPD-agar plates and incubated at 37°C for 24 h. Overnight cultures were prepared by inoculating single colonies into YPD 1% peptone in Erlenmeyer flasks for incubation at 30°C, 180 rpm.

TABLE 1.

Candida strains used in this study

Species	Strain name	Genotype	Internal ID	Reference
C. albicans	SC5314	Genome sequence reference strain	C55	(49)
	BWP17 + CIp30	BWP17, rps1::(HIS1 ARG4 URA3)	M130	(50)
	101	Oral isolate from healthy volunteer, kindly provided by S. LeibundGut-Landmann	C222	(18)
C. glabrata	ATCC 2001	Genome sequence reference strain	C94	(51)
C. parapsilosis	GA1	C. parapsilosis WT strain GA1	C118	(52)
C. tropicalis	DSM 4959	C. tropicalis WT strain DSM 4959	C30	(53)
C. albicans	efg1ΔΔ/ cph1ΔΔ	SC5314, cph1::FRT1/cph1::FRT1 efg1::FRT/efg1::FRT	M2188	(19)
	eed1ΔΔ	SC5314, eed1::FRT/eed1::(FRT-SAT1-FRT)	M1314	(54)
	bud2ΔΔ	BWP17, bud2::ARG4/pURA3-BUD2::bud2::HIS1	M1411	(22)
	ece1-IIIΔΔ	BWP17, ece1::HIS1/ece1::ARG4 rps1::(URA3 ECE1ΔPIII)	M2174	(7)

Molecular biology, lentiviral transfection, and transduction

The Nluc gene was amplified with pCRISPaint-NanoLuc plasmid purchased from Addgene (Veit Hornung Addgene plasmid #67178) and integrated with an IRES sequence (Veit Hornung Addgene plasmid #128061) and the puromycin resistance gene “PuroR” (Veit Hornung Addgene plasmid #78908) into the plasmid back bone of the vector “pHRSIN-cPPTSE,” also called “pSEW” (55). The insertion was confirmed by PCR (primers pSEW_BamHI-Nluc fw, GCCCGGGGGGGATCCGCCGCCACCATGGTCTTCACACTCGAAGATTTCG and Nluc-BamHI-IRES rev, AGAGGGGCGGGATCCTTACGCCAGAATGCGTTCGCACAGCCGCC).

For generation of lentivirus-like particles, HEK-293T/17 cells were transfected with a mixture of the transfer plasmid pSEW-NanoLuc-IRES-PuroR and helper plasmids psPAX2 (Addgene plasmid #12260) and pMD2G (Addgene plasmid #12259), combined in jetPRIME transfection agent buffer, as per manufacturer’s protocol. Lentiviral supernatant was filtered and concentrated with the Lenti-X concentrator as per the manufacturer’s protocol (Clontech Labs). A431 and TR146 cells were transduced by infecting cells with lentiviral particles for 48 h and selecting them in puromycin selection medium for 10–14 days. Selected cells were passaged and cryo-preserved in 10% DMSO + heat inactivated fetal bovine serum (FBS).

Mammalian cell lines and cell culture

Mammalian cells were cultured and passaged in their respective medium (Table 2), supplemented with 10% FBS in 75 cm² tissue culture flasks. Confluent TR146 and A431 cells were detached using Accutase for 15–20 min. Detached cells were resuspended in three times the volume of their respective medium supplemented with 10% FBS (Table 2) and counted using a Neubauer counting chamber. A cell suspension containing 10⁵ cells/mL was prepared, and 200 µl of cell suspension was distributed evenly into a 96-well plate. Cells were incubated at 37°C, 5% CO₂ for ≈2 days.

TABLE 2.

Mammalian cell culture

Cell line name	Cell line description	Genotype	Medium used	Reference
TR146	Human buccal (oral) epithelial cell line	Wild type	DMEM:Ham’s F12 medium (1:1)	ECACC (from Sigma Aldrich), Lot No. 13D016
A431	Human vaginal epithelial cell line	Wild type	RPMI 2 g/L glucose	DSMZ

Western blot to confirm the stability of luciferase expression

To confirm the stability of Nluc expression across passages of transduced cells, cells from consecutive passages were collected and centrifuged at 10,000 × g, and the pellet stored at −20°C. Cells were lysed in Radio-Immunoprecipitation Assay (RIPA) buffer (Merck) with protease inhibitor cocktail (Calbiochem) and Benzonase (25 U/µL) (Novagen). Proteins were quantified by bicinchoninic acid assay and run on an SDS PAGE (12% separating, 5% stacking) in Laemmli buffer. Protein bands were transferred to nitrocellulose membranes and incubated with monoclonal anti-Nanoluc antibody (Promega) and monoclonal anti-beta Tubulin antibody (AbD Serotec). Proteins were visualized using anti-mouse HRP-conjugated secondary antibody (Pierce) with an Odyssey M imaging system (LI-COR Biosciences), and images were processed in Image Studio (LI-COR Biosciences).

Epithelial cell damage assays

To measure damage from cells seeded in 96-well plates, plates were centrifuged at 250 × g for 10 min. Where indicated, 100 µL of supernatant was collected and stored at −20°C for assaying on another day in the same week. For the luciferase assays, 5 µL supernatant was diluted with 95 µL PBS in white flat-opaque-bottomed 96-well plates (Thermoficher). Then 100 µL of 10 µM coelenterazine (in PBS) was added and luciferase activity was measured immediately using a Tecan infinite M200 luminescence reader, with an integration time of 100 ms and OD2 compensation. The OD2 neutral density filter attenuates high-level light by a factor of 100.

The LDH cytotoxicity assay was performed according to instructions in the Roche Cytotoxicity Detection Kit with minor modifications. Briefly, 20 µL supernatant was diluted with 80 µL PBS in transparent-flat-bottomed 96-well plates, 100 µL of kit reagent mixture was added to each well, the plates were incubated in the dark for 15–20 min at room temperatures, and absorbance was measured with a Tecan absorbance reader at 492 nm (reference at 600 nm).

Quantifying Nluc activity per TR146-Nluc cell from lysate

TR146-Nluc cells were detached with Accutase, resuspended in PBS, and counted with a Neubauer counting chamber. A dilution of 2 × 10⁶ cells/mL was prepared in PBS, which was serially diluted 1:10 till 2 × 1¹⁰ cells/mL in lysis buffer (1% Triton X-100 in PBS), and incubated at 37°C for 30 min. One hundred microliters of 10 µM coelenterazine was added to 100 µL of cell lysate, such that we measured Nluc activity ranging from ≈1 to ≈10⁵ cells.

Infection of epithelial cells with Candida sp. strains

Overnight cultures of Candida sp. strains in YPD, 1% peptone were diluted, washed in PBS, and counted using Neubauer counting chambers. The medium of confluent epithelial cells was replaced with a suspension of 1 × 10⁵ cells/mL Candida spp. (MOI 1) in the appropriate cell culture medium (without FBS). The cells were incubated at 37°C and 5% CO2 for 24 h.

Time course of damage to epithelial cells induced by C. albicans

TR146-Nluc cells were seeded into 96-well plates and infected with C. albicans (SC5314) at MOI 1, as per the protocol detailed above, and supernatants were collected from separate wells at 0, 3, 8, 12, and 24 h and stored at −20°C. Samples were thawed the next day, centrifuged (250 × g), and assayed for luciferase and LDH activity.

Infection of epithelial cells with C. albicans in the presence of antifungals

Stock solutions of the antifungal drugs, fluconazole (10 mg/mL), and voriconazole (10 mg/mL) were prepared in DMSO and stored at −20°C. A twofold dilution series was prepared in DMEM/F12 cell culture medium for fluconazole (8–0.25 µg/mL) and voriconazole (1–0.3125 µg/mL). Diluted drugs were transferred to TR146-Nluc cells seeded 48 h prior and infected with C. albicans (SC5314) as described above. Supernatant was collected 24 h after infection to measure damage by LDH and Nluc.

L. rhamnosus colonization and C. albicans infection in vitro vaginal epithelial model

L. rhamnosus ATCC 7469 was cultured in MRS Broth for 48 h at 37°C with 5% CO₂ and 1% O₂. L. rhamnosus was washed in PBS and diluted in RPMI-1640 to optical density 0.2 (OD₆₀₀). For L. rhamnosus colonization, 16–18 h prior to C. albicans infection, the medium of A431-Nluc was replaced with 50 µL of a suspension of L. rhamnosus in RPMI-1640. For fungal infection, 50 µL of C. albicans suspension (MOI 1) was prepared in RPMI-1640 and added to the colonized cells. After 24 h, supernatant was collected for luciferase and LDH assays.

Robustness of Nluc and LDH activity at different pHs, temperatures, and pyruvate concentrations

TR14-Nluc cells were detached with Accutase and resuspended in 5 mL PBS. Cells were counted and ≈1 × 10⁴ cells/mL were resuspended in lysing buffer (1% Triton X-100 in PBS, pH 7.4).

For the effect of different pH on the enzyme activities, the lysis buffer was adjusted to pH 3, 4, 5, 7, 8, 9, 10, or 11 with HCl or NaOH, and this buffer was used to resuspend cells, which were subsequently incubated for 10 min at 37°C and centrifuged at 10,000 × g to remove cell debris, and supernatant was collected to measure LDH and luciferase activity.

To test the effect of storage at different temperatures over time, the cell lysates were incubated for 10 min at 37°C and centrifuged at 10,000 × g, and supernatant aliquots were stored at 37°C, room temperature (≈28° C), 8°C, or −20°C. One aliquot was thawed every other day to measure luciferase and LDH activity.

To examine the effects of pyruvate on the activity of LDH, a twofold dilution series for sodium pyruvate (16–0.5 mM) and lactate (0–60 mM) was prepared in our lysis buffer, and the lysate was incubated for 10 min at 37°C and centrifuged at 10,000 × g. The supernatant was collected to test LDH and luciferase enzyme activity.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/msphere.00509-24.

Figure S1. msphere.00509-24-s0001.pdf.

Western blot to confirm stability of Nluc expression across different passages.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.