Investigating cross‐contamination by yeast strains from dental solid waste to waste‐handling workers by DNA sequencing

Abstract Trying to widen the discussion on the risks associated with dental waste, this study proposed to investigate and genetically compare yeast isolates recovered from dental solid waste and waste workers. Three samples were collected from workers' hands, nasal mucosa, and professional clothing (days 0, 30, and 180), and two from dental waste (days 0 and 180). Slide culture, microscopy, antifungal drug susceptibility, intersimple sequence repeat analysis, and amplification and sequencing of internal transcribed spacer regions were performed. Yeast strains were recovered from all waste workers' sites, including professional clothes, and from waste. Antifungal susceptibility testing demonstrated that some yeast recovered from employees and waste exhibited nonsusceptible profiles. The dendrogram demonstrated the presence of three major clusters based on similarity matrix and UPGMA grouping method. Two branches displayed 100% similarity: three strains of Candida guilliermondii isolated from different employees, working in opposite work shifts, and from diverse sites grouped in one part of branch 1 and cluster 3 that included two samples of Candida albicans recovered from waste and the hand of one waste worker. The results suggested the possibility of cross‐contamination from dental waste to waste workers and reinforce the need of training programs focused on better waste management routines.


Agency
. Caniato, Tudor, and Vaccari (2015) emphasized that a serious lack of reliable data regarding waste generation and its characteristics is a global issue, making the appropriate management solutions very difficult. This information is in agreement with Muhamedagic et al. (2009) who pointed out that the correct management and safe disposal of dental waste is one of the key ecological challenges of the modern world. There are some studies focusing on hospital and dental waste, although very few discuss specific aspects such as their potential biological risks, specially their fungi content. Previous results of our group (Vieira et al., 2010(Vieira et al., , 2011 and other studies (Madsen, Alwan, Ørberg, Uhrbrand, & Jørgensen, 2016) proved the presence of different species of viable bacteria and also of some yeast inside biomedical waste. To widen the discussion on the possible risks to waste workers, this study proposed to investigate fungal strains recovered from dental waste, comparing with those samples isolated from waste workers' tissues (hands and nasal mucosa) and their work wear. The susceptible profile of fungi samples to recommended antimicrobials was also surveyed.

| Institutional characteristics and sampling
The study was performed in a Public Dental Healthcare Service in Belo Horizonte, state of Minas Gerais, Brazil. The team responsible for processing environmental surfaces and waste collection was composed of 12 members, 6 in the first work shift (from 06:00 a.m. to 03:00 p.m.) and 6 in the second shift (01:00 p.m. to 10:00 p.m.).
To investigate the microbial load of waste workers, three complete samplings were performed on days 0, 30, and 180. Hands and coats were sampled using the methodology proposed by Snyder et al. (2008). Hand testing was carried out by moving three times a salinemoistened sterile swab across the dorsum of each finger and finished by drawing two circles on the palm, covering as much of the palm as possible. Sample collection begun with the nondominant hand and finished with the dominant one. Swabs were taken from professional coats by drawing a "W" with the saline-moistened swab on the belt line area. Two samples were collected from nasal mucosa according to Neves (2007). The swab was inserted into the anterior nare (<1 cm) and rotated gently for 10 sec. All swabs were inserted into a transport tube, properly labeled, and immediately transported to the laboratory. Sampling was performed by a trained professional. Waste sampling took place on days 0 and 180 adopting previously published methodology (Vieira et al., 2011). In this study, we considered the three main categories proposed by Kizlary, Iosifidis, Voudrias, and Panagiotakopoulos (2005) to separate dental solid waste: infectious waste, noninfectious waste, and domestic-type waste. To properly investigate the infectious category, dental solid waste produced in 24 hr was transferred to the waste storage room and were visually inspected and manually separated by specially trained personnel, using personal protective equipments (PPE). PPE included gloves, masks, goggles, gowns, and protective clothing. This waste category included those materials that were suspected to contain pathogens and that poses a risk of disease transmission (WHO, 2014). These materials represented 19.6 kg (mean values for the two samples) and represent 67.3% of the total amount of waste generated in 24 hr. From this portion, randomly 500 mg sample was obtained and placed in a sterilized container; to enhance microbial recovery, 500 ml of sterile saline solution supplemented with l-cysteine hydrochloride monohydrate (InLab ® ) and Tween 80 with pH adjusted to 7.0-7.2 was added (Vieira et al., 2011). Waste sample was aseptically mixed every 15 min and after 1 hr contact, 150 ml of the leached liquid was collected inside a sterile bottle. The flask was immediately transported to the laboratory and one aliquot of 0.1 ml was streaked onto Sabouraud dextrose agar (SDA, Difco ™ ). Morphologically different colonies were isolated as pure cultures for later investigation (Vieira et al., 2010). All yeast isolates except two were recovered from SDA. Yeast identification was obtained using conventional, automated, and genetic techniques.

| Routines of waste management
Dental waste management was implemented in 2001 and approved by the local government in the same year, in 2008 and in 2015. Wide wheeled containers (capacity 500 and 1,000 L), small washable and colored containers with pedals (100 L) and gram precision digital scale (capacity: up to 60 kg) were acquired. All infectious waste coming from dental offices was discarded inside small orange containers lined with white plastic bags with the international symbol for biological hazard.
At the end of the work shift (11:30 a.m., 16:30 p.m., and 08:30 p.m.) the bags were collected and transported to the waste storage room.
These plastic bags were daily weighed and sent for treatment by incineration. Common waste was discarded inside blue plastic bags disposed of inside small white containers and was collect by the government. All small containers were externally labeled as "infectious" or "common" waste depending on its content. All employees received training on segregation, collection, storage, transportation, and waste disposal, and a permanent educational program has been established with discussion, lectures, and seminars. All 12 waste workers were involved in waste collection from dental offices and also transport three times a day the containers to the waste storage room. All daily routines had been performed using protective clothing, gloves, masks, goggles, and gowns.

| Conventional and automated identification
Slide culture and microscopy were performed according to Sousa et al. (2016). One fragment of corn meal agar added with Tween 80 was placed over a sterile glass slide. After inoculating yeast isolates by making three central grooves, a sterile cover slip was placed on the agar surface. The culture was incubated inside a sterile Petri dish with a piece of cotton saturated with sterile water to maintain humidity at 28°C for nearly 7 days. From the fifth day of incubation, yeast growth was observed to perform the morphological identification. Yeast were speciated by visualizing the formation of pseudohyphae, hyphae, chlamydospores, and also the precise blastoconidium distribution over this structures. All samples isolated were identified by the automated system Vitek ® (bioMérieux) using Vitek ® 2 YST ID cards (Graf, Adam, Zill, & Go, 2000). Scores above 80% were considered for acceptable identification. Candida parapsilosis ATCC 22019 was included as control.

| Antifungal drug susceptibility
Antifungal susceptibility testing was performed according to the Clinical and Laboratory Standards Institute (CLSI, 2008) updated to CLSI 2012 parameters reported by some published papers (Alastruey-Izquierdo et al., 2015;Almeida et al., 2012;Nunes et al., 2013;Orasch et al., 2014;Ramos et al., 2015;Santos et al., 2014). The last edition of CLSI revised the azole and echinocandin clinical breakpoints against Candida species. The new breakpoints are now both drug and species specific, whereas the previous breakpoints were not (Fothergill, Sutton, McCarthy, & Wiederhold, 2014). Five antifungal drugs were selected considering the literature and clinical recommendations.
Final drug concentrations were of 0.125-64 μg/ml for fluconazole and 5-fluorocytosine, 0.015-8.0 μg/ml for caspofungin, and 0.0313-16 μg/ml for voriconazole and amphotericin B. When the MIC was lower than the breakpoint defined by CLSI, the isolate was considered susceptible, and if the MIC was higher than the breakpoint, the isolate was considered nonsusceptible. Samples considered dose-dependent susceptible, intermediate, and resistant were included in the last group (Orasch et al., 2014;Santos et al., 2014). Candida parapsilosis ATCC 22019 was included in all experiments as the quality control strain.

| DNA isolation
Initially, yeast cells were disrupted by heating and freezing steps, under denaturing conditions. Briefly, after growing on SDA at 28°C for 48 hr, 50-100 mg of biomass of each sample were suspended in 500 μl of extraction buffer (250 mmol/L Tris-HCl; 25 mmol/L EDTA; 0.5% (w/v) SDS; 250 mmol/L NaCl; pH 8.0), then vortexed for 1 min, and heated at 80°C for 20 min followed by freezing at −20°C for the same period. Then, the samples were centrifuged at 1,077g for harvesting the supernatant. DNA isolation from the upper aqueous phase was carried out by using a chloroform-isoamyl alcohol mixture (24:1) (14,475 g; 25°C for 10 min), followed by the DNA precipitation with 0.7 vol of isopropanol (14,475 g; 4°C for 30 min), and a washing step with 70% (v/v) ethanol (14,475 g; 4°C for 10 min). At last, DNA was suspended in 40 μl of sterile ultrapure water, and after measuring its concentration by using Nanodrop 1,000 instrument (Thermo Fisher Scientific), it was stored at −20°C.

| Ethical aspects
Ethics approval for the study was granted by the Ethics Committees of the Military Hospital (protocol number 006/2013) and Federal

University of Minas Gerais (Certificate of Presentation for Ethics
Appreciation-CAAE Number 24911213.5.0000.5149). All volunteers received verbal information and signed a written informed consent prior to participating in this study.

| Yeast identification
The mean values of microbial load detected on SDA, for the three samples collected from workers (0, 30, and 180 days) and from waste (0 and 180 days), varied from 2.23 log·CFU/ml (day 0: workers' hands of the first shift work) to 0 (0 day: dental waste; 30th day: workers' nasal mucosa of the second shift work; day 180th: workers' hands and coats of the first shift work, worker's nasal mucosa, and coats of the second shift work). According to Hou et al. (2016), C. haemulonii complex is an emerging group, which has been described as a cause of human infection T A B L E 1 Identification of 18 yeast isolates recovered from dental solid waste and waste-handling workers' nasal mucosa, hands, and professional clothing, using automated, conventional microbiological, and genetic methods

| Antifungal drug susceptibility
The minimal inhibitory concentrations (MICs) of tested antifungal drugs are depicted in

| Genetic investigation
The sequencing of ITS1-5.8S-ITS2 rDNA region was suitable to precisely identify 14 isolates to species level. High-quality chromatogram regions that were chosen for sequencing analysis displayed values among 20-30 as judged by Phred software (http://www.phrap.com/ phred/). These scores represent an accuracy of the base calling by sequencer reading of 99% and 99.9% (error probability of 1 in 100 and 1 in 1,000), respectively. DNA sequence chromatogram files judged by Phred <20 were not considered for molecular identification. The sequences submitted to online Blast tool (https://blast.ncbi. nlm.nih.gov/Blast.cgi) against GenBank database returned matched sequences with lengths between 268 and 534 bp, bit score greater than or equal to 472, E-value less than or equal to 3e-129, and identity greater than or equal to 98%, as shown in Table 2.

| Genetic variability
PCR with (GTG) 5 oligo (Data S1) performed to investigate ISSR polymorphisms generated bright and defined DNA bands patterns in their majority, as well as allowed to design a dendrogram (Figure 1), displaying genetic distances. All bands were less than 3,000 bp. The T A B L E 3 Susceptibility profile of 16 yeast isolates recovered from dental solid waste and waste-handling workers' nasal mucosa, hands, and professional clothing F I G U R E 1 Dendrogram of the 14 yeast recovered from dental solid waste and waste handling workers' nasal mucosa, hands, and professional clothing. Yeast isolates 1-5, 7, 9, 10, 12-14, 15, 16, and 18 (according to  Rhodotorula mucilaginosa and C. victoriae isolates were recovered from the same afternoon employee on the first and second sampling days, respectively. According to Wirth and Goldani (2012), Rhodotorula is a common environmental yeast found in air, soil, lakes, ocean water, milk, and fruit juice. Despite previously considered nonpathogenic, Rhodotorula species have emerged as opportunistic pathogens with the ability to colonize and infect susceptible hosts. The isolation of one Rhodotorula strain from the nasal mucosa of one waste-handling worker could be a matter of concern considering the possibility of cross-contamination. In cluster 3, C. albicans (#14 and 18) isolates were grouped with 100% similarity; they have shown complete dissimilarity with yeast grouped in clusters 1 and 2. It is important to highlight this result once C. albicans #14 and 18 were both recovered on the last sampling (day 180), one from dental waste and the other from the hand of one waste worker. This finding suggests that the employee had been contaminated during his work activities and that waste should be considered as an important reservoir of microorganisms inside a health care setting. In turn, O. polymorpha (#9) was also completely dissimilar compared to any other isolated yeast result of ISSR ( Figure 1). It was possible to observe that some yeast belong to the same clone and are circulating among waste workers. It was also confirmed that some strains were recovered from hands, nasal mucosa, professional clothing, and also inside dental waste. These findings prove that yeast isolates could be carried by waste workers that would be a source of infection especially for immunocompromised patients.

ACKNOWLEDGMENTS
We thank those workers who voluntarily participated in this study.