Rapid preparation of Candida genomic DNA: combined use of enzymatic digestion and thermal disruption

Abstract

Nucleic acid based molecular technologies are the most promising tools for the early diagnosis of Candida infection. A simple and effective DNA preparation method is of critical for standardizing and applying molecular diagnostics in clinic laboratories. The goal of this study was to develop a Candida DNA preparation method that was quick to do, easy to perform, and bio-safe. Snailase and lyticase were screened and combined in this work to enhance the lysis of Candida cells. The lysis solution composition and metal bath were optimized to boost amplification efficiency and biosafety. A duplex real-time PCR was established to evaluate the sensitivity and specificity of the preparation method. Using the supernatant from the rapid preparation method as templates, the duplex PCR sensitivities for five common Candida species were determined to be as low as 10⁰ CFUs. When compared to conventional preparation methods, the samples prepared by our method showed higher PCR detection sensitivity. PCR identification and ITS sequencing were 100% consistent, which was better than biochemical identification. This study demonstrates a rapid method for Candida DNA preparation that has the potential to be used in clinical laboratories. Meanwhile, the practical application of the method for clinical samples needs to be proven in future investigations.

Introduction

Invasive candidiasis (IC), including candidemia, is a significant cause of morbidity and death, particularly in individuals with severe medical conditions (Cleveland et al. 2012; Lamoth et al. 2018; Magill et al. 2014). Even when IC patients were given antifungal medication, the death rate in severely sick patients was as high as 58.6% (Al-Dorzi et al. 2020). Previous epidemiological data showed that Candida species are among the top four pathogens responsible for nosocomial bloodstream infection (Pfaller and Diekema 2007; Wisplinghoff et al. 2004). The major 5 leading Candida pathogens that cause IC were: Candida albicans, Candida glabrata, Candida tropicalis, Candida parapsilosis and Candida krusei. Although C. albicans is the most common cause of Candida infection, the prevalence of IC caused by non-albicans spp. has steadily grown (Borjian Boroujeni et al. 2021). Notably, the human race has always been vulnerable to emerging yeast infection. Candida auris, a novel pathogen known as a "superbug fungus," is a multidrug-resistant yeast that has caused invasive infection and mortality in recent years (Chakrabarti and Singh 2020; Plachouras et al. 2020; Satoh et al. 2009). In 2019, many countries, including the United States, experienced nosocomial C. auris outbreaks. According to Adams E et al., approximately half of C. auris infected individuals died within 90 days of being identified with the fungus (Adams et al. 2018).

Traditional laboratory diagnosis and empirical therapy strategies revealed significant limits in dealing with IC, suggesting that prompt identification and susceptibility testing of Candida species may not always be possible. Culture-based identification and characterization approaches are sometimes hampered by long turnaround times and poor sensitivities (Clancy and Nguyen 2013; Ness et al. 1989; Pfeiffer et al. 2011). Different guidelines have recently suggested the G test, which is designed to detect (1,3)-D-glucan, for IC screening(Hage et al. 2019; Martin-Loeches et al. 2019). However, this approach does not give enough information for IC management, such as Candida species and yeast susceptibility test results. Because glucans are a non-specific polysaccharide component of fungal cell walls that may be detected in a wide range of species, false positive results are a methodological problem for the G test(Egger et al. 2018; Ito et al. 2018).

Technological advances in nucleic acid-based molecular diagnostics bring new insights into IC diagnosis. Several nucleic acid detection tests, such as polymerase chain reaction (PCR), hybridization, and the loop-mediated isothermal amplification (LAMP), have been shown to identify the infection even in its early stages (Inácio et al. 2008). Among them, PCR was regarded as the most representative and promising diagnostic tool because to its high sensitivity, specificity, reproducibility, and short detection time (Derveaux et al. 2010; Huggett et al. 2005). The majority of nucleic acid-based molecular diagnostic methods involve two steps: DNA extraction and amplification. The fundamental assurance for amplification is the quality and efficiency of DNA preparation. Tough cell walls of Candida spp., which are made up of Chitin, glucan, mannan, and glycoprotein, make DNA preparations challenging.

To address this issue, the researchers followed different strategies (Liu et al. 2000; Romanelli et al. 2014). However, most of these methods are incapable of meeting the criteria of clinical laboratories for easy standardization and biosafety. Thus improve the DNA preparation method would boost the application of IC molecular diagnostics. Here we aimed to demonstrate a Candida DNA preparation method with shorter sample turnaround time, easy to perform, and bio-safe. Furthermore, the prepared DNA samples can be directly applied to the subsequent PCR reaction, removing the need for traditional procedures of DNA extraction and purification.

Materials and methods

Candida strains and culture

C. albicans strain SC5314 was used in this study (He et al. 2015). The standard Candida species involved, including C. tropicalis ATCC1369, C. glabrata ATCC15126, C. parapsilosis ATCC22019 and C. krusei ATCC6258, as well as a standard Saccharomyces cerevisiae strain ATCC9763 were obtained from the American Type Culture Collection. The fungi were cultured to exponential phase at 35 °C, 5% CO₂ in yeast extract peptone dextrose (YPD) medium (1% yeast extract, 2% peptone, 2%D-glucose) and harvested by centrifugation at 400 × g for 5 min. After two washes with PBS, the fungal density was determined using a cell counting plate. A Candida count of 5 × 10⁷ CFUs was used to choose the suitable digestive enzymes; a Candida count of 5 × 10⁵ CFUs was utilized to validate the ideal lysis conditions, which may be closer to the fungal load in real applications.

Candida lysis ability of digestive enzymes

Digestive enzymes, including snailase (Takara Biotechnology, China), lyitcase (Sigma, USA), zymolyase (Zymo Research, USA) and glucanase (Sigma, USA), were tested for their capacity to lyse Candida cells. Aliquots of enzyme at different concentrations were prepared according to the manufacturer’s instructions.

The precipitated Candida cells were resuspended in 1 mL of suspension buffer (0.5 M sorbitol, 25 mM EDTA, pH = 8.0) containing 2 μL of β-mercaptoethanol. The different digestive enzymes then were added to each Eppendorf tube at the following concentrations: (1) snailase: 0, 130, 650, 1040 and 2080 μg/mL; (2) lyitcase: 0, 100, 200, 500 and 800 U/mL; (3) zymolyase: 0, 40, 80, 160 and 320 U/mL; and (4) glucanase: 0, 40, 80, 160 and 320 U/mL. All of the digestive reaction systems described above were incubated at the temperatures recommended by the enzymes, which were 35, 30, 30 and 50 °C for snailase, lyitcase, zymolyase and glucanase, respectively. The lysing effects were assessed after 5 h by counting the remaining intact Candida blastospores with a cell counting plate.

Preparation of genomic DNA control

Overnight cultured C. albicans SC5314 cells were collected by centrifugation. Following a PBS wash, the cells were treated with a yeast lytic enzyme kit (Solarbio Life Sciences, China). SC5314 cells were resuspended in 470 µL of sorbitol buffer (0.6 M sorbitol, 5 mM EDTA, 50 mM Tris pH 7.4) and incubated for 10 min at room temperature. Then, 25 µL of enzyme solution and 5 µL of β-mercaptoethanol were added into the SC5314 suspension. The mixture was incubated in a water bath at 30 °C for 4 h before being centrifuged at 14,000 g for 30 min. The genomic DNA was extracted from the collected precipitates using the Yeast DNA Extraction Kit (Omega BioTec, USA) according to manufacturer’s instructions.

Analyze PCR inhibitors in the lysis solution

WE designed an interference experiment to evaluate the impact of the components on the PCR because the lysis solution comprises a variety of components. The purified SC5314 genomic DNA was used as the PCR template. Snailase, lyticase, β-mercaptoethanol, sorbitol and EDTA were separately added into the 50 µL PCR reaction systems to observe their effects on amplification. Primer and probe sequences were as follows: forward primer, 5′- GGG TTT GCT TGA AAG ACG GTA -3′; reverse primer, 5′- TTG AAG ATA TAC GTG GTG GAC GTT A -3′; TaqMan probe, 5′-FAM- ACC TAA GCC ATT GTC AAA GCG ATC CCG -TAMRA-3′. PCR reactions were performed in 50 µl volumes, containing 2 µl of each primer (forward and reverse) at a concentration of 10 µmol/L, 25 µl of 2 × Taq PCR Mix (Tiangen Biotech, China), 2 µl Taqman probe, 4 µl of DNA template (∼50 ng) and a specified volume of lysate components; The reaction volume was adjusted to 50 μL with double-distilled water. The qPCR was performed using a SLAN 96P real time PCR System (HONGSHI, China), and cycling conditions were as follows: initial denaturation at 95 °C for 2 min, followed by 40 cycles of 94 °C for 15 s and 60 °C for 1 min.

Impact of high-temperature metal bath

Cultured C. albicans SC5314 cells were washed with PBS and divided into two equal aliquots. Following centrifugation, the aliquots were incubated with optimized lysis solution at 30 °C for 1 h. Then, one aliquot was heated at 100 °C in a DB100C metal bath (JOANLAB, China) for 10 min, while the second aliquot was left untreated. After centrifuging the lysates for 5 min, the supernatants were utilized. The parameters of the amplification curve and cycle threshold (Ct) values were used to assess the effect of high temperature on PCR performance.

Duplex real-time PCR

Prior to developing the multiplex real-time PCR test, singleplex real-time PCR techniques were optimized to identify the five Candida species. The primers and probes were designed as previously reported (Brinkman et al. 2003; Guiver et al. 2001; Guo et al. 2016). The matrix dilution method was used to establish the optimal concentration of primers and probes.

As shown in Table 1, five sets of specialized primers/probes and one set of universal primers/probes were matched into three duplex PCR reactions. The 50-μl PCR reaction system, equipment, and amplification conditions were configured exactly as described in “Analyze PCR inhibitors in the lysis solution.”

Table 1 Candida TaqMan primer and probe sequences

Internal transcribed spacer (ITS) sequencing

ITS sequencing was used as a gold standard to identify Candida spp.. Total DNA was isolated from Candida isolates and amplified using ITS1/4 primers. Shanghai Biological Engineering Technology Co., Ltd. sequenced the amplicons, edited them with the BioEdit Sequence Alignment program, and compared them to reference ITS regions deposited in the GenBank Database.

Statistical analysis

Statistical analysis was performed using software GraphPad Prism 7.0. Descriptive statistics, including means and standard deviations, were used to summarize continuous measures. For continuous variables, independent t-tests were applied. Real-time PCR efficiency, statistical significance between slops and intercepts were calculated through linear regression analysis. P < 0.05 was considered significant.

Results

Determine the optimal combination use of digestive enzymes

Table 2 shows the lysis efficiency of various digestive enzymes and their combinations (for details, see Additional file 1: Table S1–S6 and Additional file 2: Figure S1–S5 in supplemental materials). The snailase and lyticase could effectively lyse the blastospores of C. albicans. The blastospore lysis rates were steadily raised in line with the increasing doses of snailase and lyticase. According to the findings, 1040 μg snailase or 200 U lyticase could disrupt 5 × 10⁷ CFUs of the blastospores of C. albicans in 1 mL lysis solution. The enzymes glucanase and zymolyase showed no effect on blastospore lysis. Following that, a combination of snailase and lyticase was used to achieve higher lysis efficient. Dose dependent studies were performed to determine the optimal concentrations for the combination use of lyitcase and snailase. It was confirmed that 650 μg snailase combined with 100 U lyticase could completely disrupt 5 × 10⁷ CFUs of C. albicans blastospores in 1 ml lysis solution. Correspondingly, 130 μg snailase combined with 20 U lyticase could totally disrupt 5 × 10⁵ CFUs of C. albicans blastospores in 1 h.

Table 2 Lysis efficiency of digestive enzymes and their combinations against Candida albicans SC5314

PCR inhibitors in the lysis solution

The interference experiment displays the Ct values under different conditions using purified SC5314 DNA as the template. As shown in Table 3, the principal chemicals that interfere with PCR are sorbitol and EDTA. When 12.5 mM sorbitol or 0.625 mM EDTA were added to the PCR reaction, no amplification curve was observed, which was identical to the negative control. While in the presence of 16 µg snailase, 2.5 U lyticase or 0.05 µL β-mercaptoethanol, the Ct values were identical to those obtained from the positive control.

Table 3 Impact of the lysis solution components on PCR amplification

Optimization of the lysis solution

Removing sorbitol and EDTA from the lysis solution had no effect on lysis ability when co-incubated with SC5314 blastospores. A 100 μL lysis solution containing 130 μg snailase, 20 U lyticase and 1 μL β-mercaptoethanol could totally disrupt 5 × 10⁵ CFUs of SC5314 blastospores under 30 °C for 1 h. Further investigation indicated that among the five Candida spp. investigated, the lysis solution could successfully lyse blastospores of C. albicans, C. tropicalis and C. parapsilosis. The blastospores of C. glabrata and C. krusei, on the other hand, appear to stay mostly intact even when the enzyme concentration was raised significantly (shown in Additional file 2: Figure S7).

High-temperature metal bath improved the PCR amplification effect

As shown in Fig. 1, PCR on high-temperature metal bath treated C. albicans lysates produced a typical S-shaped amplification curve with significantly higher ΔRn (1.35 ± 0.03 VS 1.06 ± 0.02) values and lower Ct values (19.08 ± 0.06 VS 20.26 ± 0.11) when compared to untreated samples.

Fig. 1

Better PCR amplification effect was obtained after enzyme-lysed C.albicans sample treated with high temperature metal bath. The amplification curve produced by samples treated with high temperature metal bath (a), which displays a typical S-shape with significantly higher ΔRn and lower Ct values when comparing with the curve produced by the sample without metal bath treatment (b)

Sensitivity and specificity of the duplex PCR

Five Candida spp.’s cultured blastospores were collected and adjusted to indicate densities. The samples were treated with a high temperature metal bath after being incubated with the optimized lysis solution. When lysate supernatants were used as templates for PCR amplification, the sets of specific primers and probes were able to effectively amplify the corresponding Candida samples, with detection limits as low as 10⁰ CFUs (shown in Additional file 2: Figure S8). The detection limits of Candida universal primers (ITS1 and ITS4) were varied when detecting different Candida species, which was 10¹ CFUs for C. glabrata and C. parapsilosis and 10⁰ CFUs for C. albicans, C. tropicalis, and C. krusei (shown in Additional file 2: Figure S9).

The cross-amplification of primers and probes with non-Candida pathogens was also assessed. The non-Candida samples were included a variety of clinically prevalent bacterial isolates and sera from various virus-infected individuals. As shown in Table 4 and Additional file 2: Figure S10, the duplex PCR assay correctly identified the corresponding Candida spp., with no cross-reaction with DNA prepared from other pathogens.

Table 4 Species tested by the duplex PCR assay

Comparing the rapid DNA preparation method with conventional method

We evaluate the PCR amplification impact of our DNA preparation method with another method that used lyticase and a DNA purification kit (Ma 2009). There were clear differences between the PCR Ct values of the five different Candida spp. DNA templates obtained by the two methods at a gradient blastospore load ranging from 10⁰ to 10⁵ CFUs. The DNA template obtained from our method yielded lower Ct values in the five Candida species investigated (for details, see Additional file 1: Table S7–S11). Statistical analysis revealed that the y-intercept of linear regression for PCR Ct values obtained by different DNA preparation methods differed significantly. The rapid DNA preparation method provided a lower intercept of linear regression than the conventional method (Table 5).

Table 5 Comparison of linear regression parameters for PCR Ct values obtained from different DNA preparation methods

Comparing PCR with microbial culture for clinical isolate identification

A total of 30 biochemically identified Candida clinical isolates were collected and subjected to duplex PCR and ITS sequencing in this study (He et al. 2021). Table 6 and Additional file 1: Table S12 illustrate the results. Except for one C. parapsilosis isolate that was misidentified as C. Krusei by duplex PCR and ITS sequencing, the results of microbial culture, duplex PCR and ITS sequencing were completely concordant for all of the remaining Candida isolates included. The overall consistency for the three identification methods was 96.67%, with duplex PCR and ITS sequencing being 100% consistent.

Table 6 The identification results of PCR, microbial culture and ITS sequencing for the 30 Candida clinical isolates

Discussion

Rapid species identification methods including PCR are becoming increasingly important for clinical control of Candida infection due to the fact that the incidence rate of IC is sustained high in recent years. Since the time-consuming steps of DNA extraction and purification were necessary, conventional PCR methods for Candida spp. detecting usually take 6 h or more (Zhang et al. 2016). Furthermore, the likehood of DNA loss and tedious operation are notable limitations for typical DNA extraction methods when used in clinical labs with high sample flow (Dalla-Costa et al. 2017). In this study, we provide a rapid and bio-safe Candida DNA preparation method for PCR detection that takes only about 3 h from sample collection to obtain the results.

DNA sample preparation method is one of the key factors that affecting PCR identification for Candida species (Dalla-Costa et al. 2017). As previously described (Kim et al. 2016; Mazoteras et al. 2015; Rickerts et al. 2012), enzyme digestion and high temperatures are efficient ways for releasing Candida DNA. To achieve higher lysis efficiency, we combined two enzymes. Contrary to our expectation, the combined enzyme solution was able to lyse C. albicans, C. tropicalis and C. parapsilosis, although it had no impact in C. glabrata and C. krusei. However, the PCR results showed that the incomplete lysis had no effect on the sensitivity of the following amplification after the high temperature metal bath. We hypothesize that the high temperature metal bath that impacted on the rather fragile cell wall that had been treated by digestive enzymes completed the DNA release. Previously, we demonstrated that the sensitivity of PCR could be increased to 10¹ CFUs utilizing a nucleic acid sample generated solely by the "heat-shock" approach (He et al. 2020). High temperature treatment also inactivates various bioactive components of the fungal lysate, which aids in reducing the influence of crude DNA template on the PCR process (Chen et al. 2019).

The lysed Candida supernatant was directly used as a PCR template to reduce the difficulty and labor of DNA sample preparation. To this end, we designed interference experiments to identify and remove the inhibitors, as well as a metal bath to inactivate the protein PCR inhibitors in the samples. The PCR amplification curve of a template prepared by enzyme lysis coupled with a metal bath was of the standard "S" type, which was superior to samples not treated with high temperature. Another key issue with sample processing in clinical labs is bio-safety. The Candida DNA preparation method presented in this paper could be completed in a closed test tube and then finally treated in a high-temperature metal bath, considerably reducing the biohazard risk.

Our study so far suggests a general experimental workflow for preparing Candida nucleic acid samples as indicated by Fig. 2. After simple pretreatment for different types of clinical samples (eg, sputum, urine or whole blood), the precipitated Candida cells were digested by the lysis solution for 1 h and treated with a high temperature metal bath for 10 min to fully release the Candida DNA. After centrifugation, the supernatant can be directly used as a template for further nucleic acid analysis.

Fig. 2

Schematic of the rapid Candida DNA preparation and detection workflow. Candida blastospores were applied to enzyme digestion and high temperature metal bath after separated from pretreated clinical samples. After centrifugation, the supernatants were directly used as templates for nucleic acid analysis. The metal bath instrument was drawn by JP, and the other figure elements were taken from the free medical image site at https://smart.servier.com/

Five sets of specific and one set of universal primers/probes were used to identify common clinical Candida spp.. The PCR sensitivity of the specific primers/probes to their corresponding Candida species was 10⁰ CFUs, which was higher than that assessed by Foongladda et al. (Foongladda et al. 2014). The sensitivity of the universal primers/probes ranged from 10⁰ to 10¹ CFUs, yielding superior results than previous studies using the identical primers/probes (Guo et al. 2016). Although the improved sensitivity may be due to various PCR reagents and the settings of the amplification, the higher DNA yield of the preparation method described in this work may be a more relevant reason.

We selected 30 Candida clinical isolates to evaluate the diagnosis accuracy of PCR using ITS sequencing as the gold standard. The PCR and ITS sequencing were completely consistent. It is worth noting that a strain of C. krusei validated by ITS sequencing was misidentified as C. parapsilosis by microbial culture. Among the possible causes are: (i) the incidence rate of C. albicans and C. tropicalis infection is relatively high, and the detection system for them is mature (Xiao et al. 2018); (ii) the CHROMagar coloration and biochemical reaction characteristics are clear for C. albicans and C. tropicalis, but not for some C. parapsilosis, C. glabrata and C. krusei isolates (Jafari et al. 2017).

It should be mentioned that, in a variety of nucleic acid detection methods, this study only employed PCR as a representative to validate the rapid Candida DNA preparation method. Other detection methods, such as the LAMP technique, are capable of detecting the Candida species with high sensitivity and specificity, without the need for expensive instruments (Fallahi et al. 2020). Kasahara et al. devised a multiplex LAMP test with a detection limit of 10⁰ CFU/mL for medically important yeasts (Kasahara et al. 2014). Hence, combining the rapid DNA preparation method and other nucleic acid detection methods might provide a feasible and easy protocol for Candida detection in the future.

Here we present a novel method of rapid Candida DNA preparation that has the benefits of high DNA yield, simple operation, easy standardization, and bio-safety. When combined with the duplex PCR test, the rapid preparation method might have practical utility in accurately diagnosing Candida infection in the clinical laboratories. Meanwhile, the practical application of the method for clinical samples needs to be proven in future investigations.