SEARCH

SEARCH BY CITATION

Keywords:

  • Allelic ladder;
  • Aspergillus fumigatus;
  • finger printing;
  • microsatellite;
  • standardization

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

An interlaboratory study was performed with the aim of investigating the reproducibility of a multiplex microbial microsatellite-based typing assay for Aspergillus fumigatus in different settings using a variety of experimental and analytical conditions and with teams having variable prior microsatellite typing experience. In order to circumvent problems with exchange of sizing data, allelic ladders are introduced as a straightforward and universally applicable concept for standardization of such typing assays. Allelic ladders consist of mixtures of well-characterized reference fragments to act as reference points for the position in an electrophoretic trace of fragments with established repeat numbers. Five laboratories independently analysed six microsatellite markers in 18 samples that were provided either as DNA or as A. fumigatus conidia. Allelic data were reported as repeat numbers and as sizes in nucleotides. Without the use of allelic ladders, size differences of up to 6.7 nucleotides were observed, resulting in interpretation errors of up to two repeat units. Difficulties in interpretation were related to non-specific amplification products (which were resolved with explanation) and bleed-through of the different fluorescent labels. In contrast, after resolution of technical or interpretive problems, standardization of sizing data by using allelic ladders enabled all participants to produce identical typing data. The use of allelic ladders as a routine part of molecular typing using microsatellite markers provides robust results suitable for interlaboratory comparisons and for deposition in a global typing database.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Aspergillus fumigatus is the fungal species most frequently involved in human infections in some immunocompromised patient populations, such as allogeneic stem cell transplant recipients. To gain more insight into the genetic and epidemiological relationships between isolates obtained from various sources, several molecular typing techniques have been developed [1,2]. Ideally, typing results should be accurate, reproducible and easy to interpret. Importantly, methods should be transferable to other settings, so that individual results can be compared to each other irrespective of laboratory variables such as the use of different equipment, reagents and software. Unfortunately, many typing methods, e.g. those based on the analysis of random amplified polymorphic DNA [3], restriction fragment length polymorphism [4] and amplified fragment length polymorphism [5], yield fingerprint profiles that consist of complex banding patterns that are difficult to reproduce in different settings. Only a few fingerprinting methods have been developed for A. fumigatus that yield exact typing data that can be unambiguously and easily interpreted. One such format is multilocus sequence typing, which offers, however, only limited discriminatory power to distinguish among different A. fumigatus isolates [6]. Another exact and high-resolution typing format is based on microsatellite markers. Two such microsatellite-based fingerprinting assays for A. fumigatus have been reported by Bart-Delabesse et al. and de Valk et al. [7,8]. Microsatellites, or short tandem repeats (STRs), provide high-resolution analysis that is consistent with restriction fragment length polymorphism [9] and amplified fragment length polymorphism analyses [5]. The two methods differ primarily in the nature and number of markers that are analysed. Amplification of microsatellite loci under high-stringency conditions leads to reproducible amplifications that are hardly affected by minor experimental variables. Sizing of PCR fragments is done automatically using high-resolution electrophoresis platforms, and the data generated can be easily converted into the corresponding number of repeats by comparison to reference fragments with established repeat numbers. Because of the compact and numerical output, STR typing is a very attractive system for exchanging results among laboratories in a digital format and for establishing global typing databases. However, interlaboratory reproducibility and transferability of results obtained with microbial STR typing systems have not yet been demonstrated. The complicating factor, which has been well documented, is that the electrophoretic mobility (and thus the calculated size) of a DNA fragment in capillary electrophoresis platforms is influenced by multiple factors. These include the exact base composition and sequence of the DNA, the separation matrix, presence of denaturing compounds, temperature, and fluorescent labels [10,11]. Additionally, even the size standard and the DNA polymerase that are used for amplification may affect the calculated size of an allele [12]. Thus, sizing values alone are not suitable for exchange unless a careful calibration of the different platforms has been established [12,13]. Such a calibration could consist of the generation of calibration curves, as recently demonstrated by Pasqualotto et al. [13]. Alternatively, a more straightforward and universally applicable method for achieving such a calibration is through the use of allelic ladders [14,15]. An allelic ladder consists of a well-defined mixture of pre-amplified alleles with predetermined repeat numbers (by DNA sequencing), which can be used to create reference positions for the interpretation of typing results. Here, we report the results of an international multicentre study that provides the proof of concept for use of locus-specific allelic ladders in microsatellite-based microbial typing schemes.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

STR typing protocol

The STR markers used in this study are a subset of the panel described by de Valk et al. [8], and involve the M3 multiplex (containing the STRAf-3A, STRAf-3B and STRAf-3C trinucleotide repeat markers) and M4 multiplex (containing the STRAf-4A, STRAf-4B and STRAf-4C tetranucleotide repeat markers). In each multiplex reaction, different fluorescent labels were used to discriminate among the individual markers. For reasons of economy, the combination of carboxyfluorescein (FAM), hexachlorocarboxyfluorescein (HEX) and tetrachlorocarboxyfluorescein (TET) was chosen.

Construction of allelic ladders

Allelic ladders were constructed for each of the three trinucleotide markers in the STRAf3 panel (STRAf-3A, STRAf-3B and STRAf-3C, respectively). Briefly, selected alleles were amplified in monoplex PCR reactions. The resulting fragments were cloned into the PGEM-T Easy Vector System (Promega, Leiden, The Netherlands). DNA from recombinant clones was isolated using High Pure chemistry (Roche Diagnostics, Almere, The Netherlands), and quantified by UV measurements. Verification of the repeat number in each construct was performed by direct sequence analysis of the plasmid DNA preparations as described previously [8]. A mixture of plasmid DNA preparations was made containing inserts with established repeat numbers. The mixture of plasmids was then amplified using monoplex PCR reactions and analysed by capillary electrophoresis. Finally, the concentration of certain alleles was adjusted to allow unambiguous identification of each reference allele. As PCR amplification of short tandem repeats leads to the formation of so-called stutter peaks (e.g. additional fragments mostly containing fewer repeat numbers [8,12]), it was not necessary to include all alleles in the amplification mixture. The missing alleles were generated automatically during the amplification process. As a result, although the sample used for amplication contained only 22 selected alleles, after amplification of this plasmid mixture the STRAf3A allelic ladder contains 68 alleles with repeat numbers ranging from nine to 76. Likewise, the STRAf3B allelic ladder contains alleles with repeat numbers from 7 to 39, and the STRAf3C allelic ladder contains alleles with repeat numbers from 5 to 53, as shown in Fig. 1a–c. An example of the use of these allelic ladders is illustrated with marker STRAf3A in Fig. 2.

image

Figure 1.  Illustration of the allelic ladders for the trinucleotide repeat markers. The numbers above the peaks correspond to the repeat numbers of each allele in these ladders. Boldface alleles in a larger font were actually included in the samples; the other alleles are the result of the formation of stutter peaks upon PCR amplification of the included alleles. (a–c) The STRAf-3A, STRAf-3B and STRAf-3C allelic ladders, respectively.

Download figure to PowerPoint

image

Figure 2.  Application of allelic ladders for calibrating short tandem repeat (STR)-based typing data among different laboratories. The example shown is for the STRAf3A marker. Sample 1 yields a principal peak of size X. Size X corresponds to an allele with 27 repeats, as can be deduced from the pre-amplified allelic ladder, which is added to another capillary with the same size standard as used with regular samples. Sample 2 yields a principal peak of size Y, which corresponds to 31 repeats. Although sizes X and Y can be very precisely reproduced within a laboratory (usually within 0.2 nt), they can be very different in another laboratory on another platform and/or using different electrophoretic conditions.

Download figure to PowerPoint

Study design and protocol

Five laboratories (denoted A–E throughout this article) participated in the study. Each laboratory received 12 purified DNA samples (A1–A12), multiplex primer mixes M3 and M4 (amplifying the three trinucleotide markers in the STRAf3 panel and the three tetranucleotide markers in the STRAf4 panel, respectively) [8], a size standard (ET400-R from GE Healthcare, Diegem, Belgium), allelic ladders for the STRAf3 loci (including a graphical display of their appearance), and amplification instructions for the preparation of reaction mixes. Laboratories also received six A. fumigatus isolates (B1–B6, as conidia), which were included in order to examine the potential influence of the DNA quality on the outcome of the assay. Four of these six isolates were also provided as DNA samples and should have yielded identical typing results. All participating laboratories had access (either in-house or through an external party) to a capillary electrophoresis platform with multi-colour detection ability. Each participating laboratory was free to choose its DNA isolation procedure for samples B1–B6, enzyme to be used for amplification (as long as it did not involve a proofreading formulation), and thermocycler, size standard and capillary electrophoresis platform (including running conditions). Reference values for all samples were obtained independently in a blinded fashion by one of the participating laboratories.

Data collection and analysis

Several datasets were reported by the participating laboratories. First, the sizes of the peaks in the allelic ladders were determined with a free choice of size standard. These values were to be used as reference values for the interpretation of the results of the STRAf 3 markers. In order to examine the influence of various size standards, the sizes of peaks in the allelic ladders were also determined using the ET400-R size standard. The results for the markers in the STRAf3 panel for all 18 samples were reported as repeat numbers. Second, the repeat numbers of the markers in the STRAf4 panel were determined for all 18 samples using the ET400-R marker in combination with reference size values taken from the original publication [8]. For those alleles that did not exactly fit the reference positions in the allelic ladders or reference values from the literature, but that differed by one, two or three nucleotides, the notation n.1, n.2 or n.3 was used, indicating that the allele corresponded to a size of n repeats plus one, two or three nucleotides, respectively [8].

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

The wide variety of experimental variables and equipment used by the participating laboratories is shown in Table 1. All the participating laboratories used a different capillary electrophoresis platform. In addition, three distinct thermocyclers, at least three distinct DNA isolation procedures and four different commercially available size standards were used.

Table 1.   Overview of experimental variables in the participating laboratories
 ABCDE
  1. STR, short tandem repeat.

Prior STR experienceAmpleAmpleAmpleSomeNone
Pretreatment or cell lysisMagNa Lyser (Roche Diagnostics)MagNA Lyser (Roche Diagnostics)Phenol/chloroform extractionFastDNA Kit (Bio 101 systems)Phenol/chloroform extraction
DNA isolation or clean-upMagNA pure LC DNA Isolation Kit III (Roche Diagnostics)Qiagen AMP DNA Blood Mini Kit (Qiagen)Qiagen G-100 columns (Qiagen)DNeasy Plant Mini Kit (Qiagen)Chromaspin columns (Clontech)
EnzymeFastStart Taq (Roche Diagnostics)FastStart Taq (Roche Diagnostics)FastStart Taq (Roche Diagnostics)FastStart Taq (Roche Diagnostics)FastStart Taq (Roche Diagnostics)
ThermocyclerT1 (BioMetra)GeneAmp  9700 (Applied Biosystems)GeneAmp  9700 (Applied Biosystems)iCycler (BioRAD)GeneAmp  9700 (Applied Biosystems)
Capillary platformMegaBACE  500 (GE Healthcare)ABI Prism 3100 (Applied Biosystems)ABI Prism 310 (Applied Biosystems)ABI Prism 3130 (Applied Biosystems)ABI Prism 3700 (Applied Biosystems)
LocationIn-houseIn-houseIn-houseExternalExternal
Size standardET400-R (GE Healthcare)400HD Rox (Applied Biosystems)GeneScan  500-TAMRA (Applied Biosystems)GeneScan  500 LIZ (Applied Biosystems)ET400-R (GE Healthcare)

First, the effect of the choice of the internal size standard on the calculated sizes of the alleles in the STRAf3 ladders was determined. Differences in sizing data for alleles in the allelic ladders are shown in Fig. 3. The sizes of the alleles in the allelic ladders were calculated using two size standards: the size standard that was routinely used in each participating laboratory (see Table 1), and the ET400-R size standard as supplied to all laboratories. The results show that identical alleles, analysed on different electrophoretic platforms with different size standards, may differ by up to 6.7 nucleotides. Switching to the use of identical size standards still yielded size differences of up to 5.4 nucleotides on the various platforms (results not shown).

image

Figure 3.  Graphical display of the size inaccuracy for each allele in the allelic ladders obtained on different analytical platforms and with different size standards. The x-axes display the repeat number of the alleles, and the y-axes the difference between the calculated size and the actual size as determined by sequencing.

Download figure to PowerPoint

The results of the determination of the repeat numbers in the trinucleotide markers in all 18 samples by the five participating laboratories are shown in Table 2. Assignment of alleles in the STRAf3 panel for all samples was done by comparing the sizes of the fragments in the samples to the sizes of the fragments in the allelic ladders using the laboratory’s routine size standard (as exemplified in Fig. 2). Two laboratories (A and B) yielded 100% identical results for all markers in all samples. Laboratory C made one error in interpretation: sample B2 yielded a ‘9.2’ allele where the expected result was a ‘10’ allele. Examination of the corresponding electropherogram showed that this was the result of reduced electrophoretic resolution. Under the experimental conditions as used, the platform (ABI Prism 310; Applied Biosystems, Foster City, CA, USA) was unable to adequately resolve −A peaks from +A peaks. This was apparent from the broad and round appearance of the peaks. Laboratory D made three errors in interpretation: a ‘31’ allele in STRAf3A was reported as a ‘10.1’ allele, a ‘15’ allele in STRAf3B was reported as a ‘10’ allele, and a ‘25’ allele in STRAf3C was reported as a ‘7’ allele. As all errors occurred in a single sample (A10), we believe this to be a problem of sample handling. When the original electropherograms were re-examined in light of the expected results, low-intensity peaks of the correct size were detected (31, 15 and 25 repeats, respectively). Laboratory D also reported low fluorescent signals with samples A1–A12 that were markedly improved with use of a higher concentration of DNA in the PCR reactions. Repeating the A10 PCR with more DNA resulted in peaks of the expected sizes. Laboratory E made eight errors in interpretation. In all these cases, the reported alleles were reduced by one or two repeat units, probably because of the assignment of stutter peaks instead of the actual principal peaks. Interpretation of the results by laboratory E was also hampered by substantial bleed-through (also called cross-talk) of the different fluorescent labels on the ABI Prism 3700, which the team was unable to resolve. With the exception of the above incidental interpretation errors, identical typing results were obtained for all samples that were provided both as isolates and DNA samples.

Table 2.   Results of the trinucleotide repeat markers based on calibration with allelic ladders
 STRAf3ASTRAf3BSTRAf3C
ABCDEABCDEABCDE
  1. aIsolates B1, B3, B4 and B5 correspond to DNA samples A4, A5, A1 and A9, respectively.

A1a4949494949111111111177777
A24848484846111111111177777
A32727272727999982626262626
A4a515151515116161616162222222222
A5a262626262623232323232222222222
A62525252525888881010101010
A7353535353511111111113333333333
A83636363636111111111177777
A9a18181818189999966666
A1031313110.1301515151015252525 725
A11343434343424242424244040404039
A122727272727121212121277777
B1a515151515116161616162222222222
B2101010101010109.210101010101010
B3a262626262623232323212222222222
B4a4949494949111111111177777
B5a18181818179999966665
B6141414141434343434341818181817

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

A multicentre study was carried out to investigate the interlaboratory reproducibility and compatibility of a microsatellite-based typing assay in different settings. From a theoretical point of view, microsatellites are ideal targets for high-resolution and exact fingerprinting of microbial isolates. The exchange of microsatellite typing data among laboratories is greatly hampered by the inability to accurately determine absolute fragment sizes by capillary electrophoresis (Fig. 3). Accurate sizing of identical DNA fragments using different high-resolution platforms is dependent on many experimental and environmental factors, which often lead to differing results [10–12]. Intralaboratory sizing of alleles has been shown to be highly reproducible, yet, almost paradoxically, it is the sizing result itself that needs to be calibrated among laboratories. This is the first report of the use of allelic ladders to precisely and reproducibly determine repeat numbers of amplified alleles in an interlaboratory setting for typing A. fumigatus isolates. The use of allelic ladders for microsatellite analysis allows standardization of sizing results, which, in turn, allows the direct comparison of microsatellite-based typing data among different laboratories and from different experimental settings. Allelic ladders consist of a mixture of common alleles that act as reference points for the position in an electrophoretic trace of fragments with specific numbers of repeats. Amplification of microsatellites by PCR leads to the formation of stutter peaks (e.g. additional fragments, mostly with fewer repeat numbers [8]). In the construction of the ladders reported here, the formation of these stutter peaks was taken advantage of in order to fill in the gaps for those alleles that were not actually present in the mixture of cloned alleles; for example, the −1 stutter peak that is formed upon amplification of the allele with 12 repeats will reveal the reference position of the allele with 11 repeats (Fig. 1). A circumstantial advantage of this approach is that it should be quite easy to unambiguously assign all alleles in the ladders, as the included alleles will yield peaks that are higher than their associated stutter peak(s).

Some participants initially misassigned the repeat numbers in their chromatograms of the allelic ladders. As a consequence, the repeat numbers for all markers in the corresponding samples were effectively altered by one repeat unit. In retrospect, such misassignments could have been prevented by including the expected results for an additional reference DNA sample with known repeat numbers. In some cases, a non-specific peak was assigned. Non-specific amplification products usually lack the stutter peaks that are associated with true alleles containing an STR. Previously, we reported that the intensity of such non-specific amplification products is greatly influenced by the annealing temperature during amplification [12]. Hence, based on thermocycler specifications, some laboratories may experience more non-specific amplification products than others. After this had been pointed out, participants were given the opportunity to reinterpret their results (Table 2).

The STRAf assay was developed as a multiplex assay in order to increase sample and marker throughput. This was easily established by using amplification primers with different fluorescent labels for each of the markers in the multiplex PCR reactions. For reasons of economy, the combination of fluorescent labels involved FAM (carboxyfluorescein), HEX (hexachlorocarboxyfluorescein) and TET (tetrachlorocarboxyfluorescein).

In this study, five different capillary electrophoresis platforms were used. All platforms have multi-colour detection capabilities, either through the presence of optical (hardware) filters (MegaBACE; GE Healthcare) or through virtual (software) filters (ABI platforms). Theoretically, it should have been possible to calibrate each of these platforms to the combination of labels used in this study, but this proved to be problematic for some of the ABI platforms. As a result, some participants were unable to properly calibrate their instrument, and consequently suffered from substantial bleed-through of the different labels. This greatly complicated the interpretation of the results for laboratory E, which showed most of the data analysis errors. Hence, the ability to properly separate the different fluorescent labels seems to be crucial for successful interpretation of the data and implementation of this assay.

Previously, we showed that sizing identical alleles using different size standards on the same electrophoretic platform will result in sizing differences of up to 5.0 nucleotides [12]. As shown in this study, these size differences can be as large as 6.7 nucleotides when different size standards are used on different electrophoretic platforms, but can still amount to 5.4 nucleotides when the same size standard is used on different analytical platforms. Furthermore, it is obvious from Fig. 3 that sizing differences are not constant across the range of alleles for a marker, and that they change with repeat numbers. In addition, individual markers behave differently on different analytical platforms. The examples clearly demonstrate that calculated allele sizes cannot be used for the comparison of typing results among different laboratories (even when the same size standard is used) and that there is a clear need for size data to be calibrated in some manner. As demonstrated here, allelic ladders are ideally suited for such calibration purposes. All that needs to be done is to use the allelic ladders under the preferred running conditions to determine the relative sizes of all alleles using these conditions. Despite the substantial variations in sizing results and after resolution of technical or interpretive problems, all laboratories were able to report identical typing data when the allelic ladders were used to calibrate the analysis. Allelic ladders have already been in use in human forensics for more than a decade [14,15] but, up to now, their use with microbial microsatellite-based typing schemes has not been reported.

No allelic ladders were provided for the markers in the STRAf4 panel. Assignment of alleles in this panel for all samples was done by comparing the sizes of the fragments in the samples to the sizes of fragments in the original publication. In this case, all laboratories used the ET400-R marker. Identical alleles yielded sizing data differing in up to 3.8 nucleotides. As a consequence, the reported repeat numbers for individual alleles in the STRAf4 panel differed by up to one repeat unit from the expected values. Without the option to calibrate the results with respect to an allelic ladder, and with the results interpreted on the basis only of the calculated (and relative) size of an allele, 118 of 264 reported alleles were incorrect (results not shown).

As might be expected, most interpretation errors were made by the participants with the least prior experience of microsatellite marker analysis. They were also the participants who left the actual electrophoretic running of the samples to an external party. Participants who produced identical typing results (laboratories A, B and C) had prior experience with this type of analysis and an in-house capillary electrophoresis platform (and software) that ensured correct detection and separation of the fluorescent labels.

In a previous study, we demonstrated the robustness of the STRAf assay by deliberately and systematically changing various reaction components to suboptimal conditions and by varying thermocycling parameters [12]. In the current study, all participants were allowed to use their regular DNA isolation procedure, DNA polymerase for performing PCR reactions, thermocycler, capillary electrophoresis platform, and size standard. In practice, five distinct combinations of pretreatment and DNA isolation procedures, three different thermocyclers, five different capillary electrophoresis platforms and four size standards were used (Table 2). The fact that all participating laboratories were able to produce identical typing data for all the STRAf3 markers in nearly all samples with this wide variety of experimental conditions again confirms the high robustness of a microsatellite-based typing assay and demonstrates the reproducibility and compatibility of results obtained with this assay in different settings. Thus, microsatellite-based typing assays are as suitable for interlaboratory comparisons as are sequence-based typing assays such as multilocus sequence typing. However, microsatellites offer a number of additional advantages in terms of speed, throughput, cost and discriminatory power. The present study also shows that, although allelic ladders were proven to provide a straightforward means of standardizing microbial microsatellite-based typing assays, a certain level of technical experience is required to ensure correct data interpretation. However, this is as true for microsatellite-based typing as it is for any other molecular typing assay.

The use of allelic ladders provides a convenient means of standardizing microsatellite-based typing assays among laboratories. This requires that such ladders should be made available to other interested laboratories. However, construction of an allelic ladder requires considerable effort and therefore, at present, and with the aim of demonstrating their usefulness, allelic ladders were constructed only for the trinucleotide repeat markers. If there is sufficient interest, allelic ladders for the dinucleotide and tetranucleotide markers can be constructed as well. In addition, this approach will only be successful if other laboratories are willing to commit themselves to using the same primers and fluorescent labels as reported here.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

We thank D. Denning and M. Cuenca-Estrella for providing support for this work and for helpful comments and suggestions.

Transparency Declaration

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

A. C. Pasqualotto received a grant from CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nivel Superior) (Brazilian government). M. J. Anderson was funded by a Wellcome Trust (UK) programme grant. All other authors: no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References
  • 1
    Varga J. Molecular typing of aspergilli: recent developments and outcomes. Med Mycol 2006; 44 (suppl): S149S161.
  • 2
    Soll DR. The ins and outs of DNA fingerprinting the infectious fungi. Clin Microbiol Rev 2000; 13: 332370.
  • 3
    Aufauvre-Brown A, Cohen J, Holden DW. Use of randomly amplified polymorphic DNA markers to distinguish isolates of Aspergillus fumigatus. J Clin Microbiol 1992; 30: 29912993.
  • 4
    Girardin H, Latge JP, Srikantha T, Morrow B, Soll DR. Development of DNA probes for fingerprinting Aspergillus fumigatus. J Clin Microbiol 1993; 31: 15471554.
  • 5
    De Valk HA, Meis JFGM, De Pauw BE, Donnelly PJ, Klaassen CHW. Comparison of two highly discriminatory molecular fingerprinting assays for analysis of multiple Aspergillus fumigatus isolates from patients with invasive aspergillosis. J Clin Microbiol 2007; 45: 14151419.
  • 6
    Bain JM, Tavanti A, Davidson AD et al. Multilocus sequence typing of the pathogenic fungus Aspergillus fumigatus. J Clin Microbiol 2007; 45: 14691477.
  • 7
    Bart-Delabesse E, Humbert JF, Delabesse E, Bretagne S. Microsatellite markers for typing Aspergillus fumigatus isolates. J Clin Microbiol 1998; 36: 24132418.
  • 8
    De Valk HA, Meis JFGM, Curfs IM, Muehlethaler K, Mouton JW, Klaassen CHW. Use of a novel panel of nine short tandem repeats for exact and high-resolution fingerprinting of Aspergillus fumigatus isolates. J Clin Microbiol 2005; 43: 41124120.
  • 9
    Bart-Delabesse E, Sarfati J, Debeaupuis JP et al. Comparison of restriction fragment length polymorphism, microsatellite length polymorphism, and random amplification of polymorphic DNA analyses for fingerprinting Aspergillus fumigatus isolates. J Clin Microbiol 2001; 39: 26832686.
  • 10
    Tu O, Knott T, Marsh M et al. The influence of fluorescent dye structure on the electrophoretic mobility of end-labeled DNA. Nucleic Acids Res 1998; 26: 27972802.
  • 11
    Vainer M, Enad S, Dolnik V et al. Short tandem repeat typing by capillary array electrophoresis: comparison of sizing accuracy and precision using different buffer systems. Genomics 1997; 41: 19.
  • 12
    De Valk HA, Meis JFGM, Klaassen CHW. Microsatellite based typing of Aspergillus fumigatus: strengths, pitfalls and solutions. J Microbiol Methods 2007; 69: 268272.
  • 13
    Pasqualotto AC, Denning DW, Anderson MJ. A cautionary tale: lack of consistency in allele sizes between two laboratories for a published multilocus microsatellite typing system. J Clin Microbiol 2007; 45: 522528.
  • 14
    Skowasch K, Wiegand P, Brinkmann B. pMCT  118 (D1S80): a new allelic ladder and an improved electrophoretic separation lead to the demonstration of 28 alleles. Int J Legal Med 1992; 105: 165168.
  • 15
    Puers C, Hammond HA, Jin L, Caskey CT, Schumm JW. Identification of repeat sequence heterogeneity at the polymorphic short tandem repeat locus HUMTH01[AATG]n and reassignment of alleles in population analysis by using a locus-specific allelic ladder. Am J Hum Genet 1993; 53: 953958.