Microsatellite Typing of Catheter-Associated Candida albicans Strains

Candida albicans is the most common pathogenic fungus and occurs frequently in the digestive tract (Bernhardt, 1998; Doskey, 2004). Vaginal candidiasis (Mohanty et al. 2007; Paulitsch et al., 2006; Sobel, 2007) is also a wide spread problem. This species can become invasive, causing infections on many different sites in patients with severe underlying diseases (Marol & Yukesoy, 2008; Odds et al., 2007).


Introduction
Candida albicans is the most common pathogenic fungus and occurs frequently in the digestive tract (Bernhardt, 1998;Doskey, 2004). Vaginal candidiasis (Mohanty et al. 2007;Paulitsch et al., 2006;Sobel, 2007) is also a wide spread problem. This species can become invasive, causing infections on many different sites in patients with severe underlying diseases (Marol & Yükesoy, 2008;Odds et al., 2007).
The classical picture of yeast cells as unicellular life forms is based on the pure-culture model of growth. In their natural habitat microorganisms including yeasts are mostly organized in biofilm ecosystems which are often ´multicultural´, made not only of yeasts but also of bacteria (El-Aziz et al., 2004;López-Ribot, 2005;Ramage et al., 2005;Nobile et al., 2006). The possibility to adhere to a surface is a very important factor for the development of fungal (Hogan, 2006;Verstrepen & Klis, 2006) and bacterial biofilms (Dolan, 2001).
Microsatellites, which are also known as short tandem repeats, are repeated nucleotide sequences with a length from 2 up to 7 base pairs. These polymorphic DNA loci are variable within a population and in this way multiple alleles are created for a single microsatellite locus. These different multilocus genotypes are used to distinguish strains within a single species (Applied Biosystems [AB], 2005). Microsatellite markers provide the possibility to discriminate strains of the same species and to trace their epidemiological pathways (Botterel et al., 2001;Sampaio et al., 2005).
For this study, pairs for three loci (CDC3, EF3, and HIS3) on three different chromosomes developed by Botterel et al. (2001) were used to compare the C. albicans strains which were found to produce a biofilm, with those strains which did not produce a biofilm on the investigated catheter material. The differentiation of biofilm and non-biofilm forming strains was based on scanning electron microscopical findings (Paulitsch et al., 2009). Different primer pairs and also different combinations of primer pairs for the subtyping of C. albicans were reported elsewhere, see e.g. the works of Sampaio et al. (2005) or Fan et al. (2007).
For each marker and for a given isolate one or two bands were observed, and each observed band was assigned to an allele. Because C. albicans is diploid each strain can be characterized by six alleles with the method used.
The discriminatory power (DP) is a numerical index to describe the probability that two unrelated samples of a test group are placed in two different typing groups. The DP of EF3 is 0.86, the DP of CDC3 is 0.77, and the DP of HIS3 is 0.91 (Botterel et al., 2001). The combined DP of all three markers was 0.97. In order to get reliable results, this index has to be greater than 0.90 (Botterel et al., 2001).

Material and methods
The 123 C. albicans (64 [52%] of them biofilm positive) strains for this study were collected during a study in biofilm forming abilities of yeast on indwelling devices (Paulitsch et al., 2009). The strains were stored at -70°C until examination.
Strains were subcultured on Sabouraud agar plates for 24 h at 35°C. For DNA extraction the PrepManTM Ultra Kit (Applied Biosystems [AB], Foster City, California) was used. For the microsatellite typing three different primer pairs were used (Botterel et al., 2001). The unmarked primers were HIS3R, CDC3R, and EF3 (Invitrogen, Lofer, Austria). The fluorescence labeling of the primers HIS3 (NEDTM, yellow), CDC3 (VICTM, green), and EF3R (6-FAMTM, blue) (all AB) was fitted to the DyeSet DS-33 (AB) which is recommended for 5-dye custom primer analyses. PCR was performed using the 96 well GeneAmp PCR System 9700 or the 96 well 2700 Thermal Cycler (both AB). PCR reactions were carried out as singleplex reactions for each primer pair. The samples were initially incubated for 2 minutes at 94°C to activate the Taq Polymerase (Eppendorf, Hamburg, Germany) and to denature the DNA. After thermal cycling (30 cycles; 94°C for 45 s, 48°C for 45 s, 68°C for 90 s) samples were kept at 68°C for another 5 minutes to complete partial polymerization.
Sample preparation for the injection in the 3100 Automatic Sequencer (AB) was done following the instructions. For analysis 1 µL of PCR product, 0.3 µL of size standard (GeneScanTM 500-LIZ®, AB) and 10 µL Hi-DiTM Formamide (AB) were mixed and transferred into a 96 well plate. The samples were denatured for 4 minutes at 94°C in a thermal cycler and immediately placed on ice. In every run three samples were used as internal control. The plate was transferred in the sequencer and processed using the Foundation Data Collection 3.0 software of the sequencer.
Data analysis was done with the GeneMapper® v3.7 software. Therefore it was necessary to set up the microsatellite analyses following the instructions of the manual (AB, 2005). The peaks were automatically detected (Auto Binning) with the created bin set, low quality data were checked manually and corrected. The results were exported in a Microsoft Excel sheet for documentation.

Results
Typing of 123 C. albicans strains was done with the above mentioned three primer pairs. Only from strain number 85 (sample W60) no data from the EF3 locus was producible. Although the DNA was isolated a second time and several PCR reactions were done for this locus, no peaks could be generated. In A comparison of the results did not reveal information of typical microsatellite models for C. albicans strains which produced biofilms in this study. Only 41 of the investigated strains showed a similarity with one or up to six other strains (  The most convergent data were generated with the CDC3 primer pair, only 12 different allele pairs were; found, with the EF3 primer pair 25 different pairs were located, and HIS3 primers provided 50 different pairs of alleles.
From six patients two strains were available, each of them originated from different samples and showed C. albicans infections in routine diagnostics. Both samples from one patient were biofilm positive, from another patient both samples were negative. The microsatellite data of these catheters are listed in table 3. When only HIS3 and CDC3 alleles were compared, five out of the six patients showed the same strain two times, when they were also compared with EF3 primer alleles, only one patient had the same strain two times.
The comparison of the genotyping of biofilm forming C. albicans strains (e.g. see figure 1) with non-biofilm forming C. albicans species shows also a consistent distribution of genotypes.   Table 3. Microsatellite models of 12 strains from six patients (two strains each).

Discussion
The catheters which were investigated in this study originated from many different stations of mainly two hospitals. The analyses of the genotypes of 123 C. albicans strains collected from these samples give many interesting points to think about. The comparison of the CDC3, EF3, and HIS3 genotyping results from the two hospitals (data not shown) did not provide suitable data for distinguishing the epidemiological distribution of C. albicans. The contribution of the genotypes was consistent within the University Hospital of Graz compared with the AKH Vienna hospital. This was also true for the aggregation of the data, no significantly dominant genotype was detected, only a group of 11 (8.9%) strains (Table 2) was found to be the most frequent genotype with the multilocus genotype characterised by CDC3: 117-129, EF3: 130-139, and HIS3 154-154. All other groups within this study consist of at most 5 strains. These results are comparable to those of Eloy et al. (2006) who studied the genotypes of C. albicans in two different hospitals using the CDC3, EF3, and HIS3 typing system. An overall number of 67 isolates were tested and 50 different genotypes were found. Eight patients shared the same genotype in one hospital; the same genotype was also present in 3 strains in the second hospital. Botterel et al. (2001) tested 100 isolates for their microsatellite profile. They detected 5, 12, and 18 alleles in the CDC3, EF3, and HIS3 system, respectively. The different associations of this alleles led to 10 CDC3, 22 EF3, and 25 HIS3 allele associations within this system. A group of 17 isolates was found to share the genotype.
This genotype was the same as reported by Eloy et al. (2006) in the group of 11 genotype identical strains. Both authors reported the multilocus genotype characterised by  for their most common strains.

www.intechopen.com
Totally different data were provided from Shi et al. (2007) who collected isolates by female and male patients with genital infection, rectal and oral samples. The authors reported 54.9% of the strains investigated to show the same multilocus genotype, these results were clearly different from all other studies.
The CDC3 locus showed 12 different allele pairs, the EF3 locus 25 allele pairs, and the HIS3 locus 50 allele pairs. This is convergent with the data within the three loci and leads to 94 multilocus genotypes. When compared with the results of Botterel et al. (2001) who reported 65 different multilocus genotypes with different allele associations of 10 for CDC3, 22 for EF3, and 25 for HIS3, it is obvious that the HIS3 locus was clearly more divergent within the current study. However, it remains unclear whether this variation is typical for C. albicans strains collected from BSI, or if the discriminatory power (DP) of the HIS3 locus (0.91) is not strong enough. The calculated overall DP for the CDC3, EF3, and HIS3 multilocus genotyping was 0.97. It is worth noting that the DP of HIS3 alone was the highest of the three loci (CDC3: 0.77, EF3: 0.86) (Botterel et al., 2001). Nevertheless, a comparison of the typing information without the HIS3 locus showed that the groups of strains sharing the same genotype do not increase significantly (data not shown).
The comparison of the genotyping of biofilm forming C. albicans strains with non-biofilm forming C. albicans species shows also a consistent distribution of genotypes. There is no literature to compare these specific results with, but as aforementioned, a consistent contribution of genotype data collected with the CDC3, EF3, and HIS3 multilocus genotyping system seems to be normal for C. albicans strains.
The collected information about strains from the same patients are worth a closer look: Only one patient out of six showed 2 strains sharing the multilocus genotype. Using the same typing system, Beretta et al. (2006) investigated 14 isolates of eight patients and reported 4 strains with the same genotype for one patient out of three. Another patient had 2 of 3 strains sharing the genotypes (Beretta et al., 2006). When only HIS3 and CDC3 alleles were compared, five out of the six patients in the current study show the same strain twice. Because of these findings, the typing was done without EF3 locus information, and as it is mentioned above for the typing without HIS3 allele information, no significant increase in the numbers of strains sharing the same multilocus genotype could be seen (data not shown).
Recapitulating the multilocus genotyping with the CDC3, EF3, and HIS3 system during this study, the data presented here is in good agreement with the authors mentioned above.

Conclusion
The multilocus genotyping with the CDC3, EF3, and HIS3 system during this study did work well and provided data comparable to former studies. Therefore it is strongly indicated that the genotyping of C. albicans strains should be continued in future studies.
Aditionally the results give possible evidence that genotypes do not matter in the connection to biofilm forming abilities, so that potentially all C. albicans strains are able to www.intechopen.com form such ecosystems. In that case, studies like the recent one can only give evidence of epidemiological behavior of the species investigated.
Another set of microsatellite markers is likely to give more information about those strains which are able to form biofilms on indwelling devices or about the epidemiological behavior of clinically important strains.

Acknowledgment
This work was partly funded by the Hygiene Fund of the Medical University of Graz. This work was performed in the TTIW-cooperation framework of Wetsus, centre of excellence for sustainable water technology (www.wetsus.nl). Wetsus is funded by the Dutch Ministry of Economic Affairs. The authors like to thank the participants of the research theme "DNA based detection technologies" for the fruitful discussions and their financial support.