Abstract

This article will discuss the estimation of genotyping errors for Mini-DogFiler panel A and B in group of wolfs (n = 39) where the source material for DNA isolation were faeces and control group of dogs (n = 170), represented by Czechoslovakian wolfdog breed (CSW), where DNA was obtained from buccal swabs.

DNA samples isolated from buccal swabs had a very low frequency of genotyping error estimated as mean error rate per allele (ea = 0.09 %), mean error per locus (el = 0.18 %) and error rate per multilocus genotype (eobs = 2.17 %). Completely different situation was observed for DNA samples isolated from faeces and affirms how complicated is to get reliable results from such material. Total performance was very poor where only 7 out of 39 samples replicated (∼18 %) at least 2 times out of 4 replicates. In such case would be eobs = 86.5 %. For this type of samples is typical very fragmented DNA with low concentration. This fact leads to the occurrence of many of negative effects like an allelic dropout, null allele or presence of PCR artefacts. Results confirm necessity of multiple DNA isolates from one sample and a multiple amplification of these samples for creating a consensus for each genotype.

Moreover, both types of samples suffered by relatively high frequency of allelic dropout occurring in locus VGL2136, but it did not increase error rate because no signal was detected in all cases. The cause of such behaviour remained unknown but it raised the question whether is this locus appropriate for analysis of wolf and CSW samples.

Keywords1. IntroductionPopulation studies made on wolves, dogs and their hybrids often used problematic biological materials, which might affect quality of analyses and subsequent results [1Kusak J. Fabbri E. Galov A. et al.Wolf-dog hybridization in Croatia., 2Beyond wild and domestic: human complex relationships with dogs, wolves, and wolf-dog hybrids., 3Salvatori V. Godinho R. Braschi C. et al.High levels of recent wolf× dog introgressive hybridization in agricultural landscapes of central Italy.]. Examples of such material are excrements, urine or saliva [4Canu A. Mattioli L. Santini A. et al.‘Video-scats’: combining camera trapping and non-invasive genotyping to assess individual identity and hybrid status in gray wolf., 5Lorenzini R. Fanelli R. Grifoni G. et al.Wolf–dog crossbreeding:“Smelling” a hybrid may not be easy., 6Sundqvist A.K. Ellegren H. Vilà C. Wolf or dog? Genetic identification of predators from saliva collected around bite wounds on prey.]. Our study of genotyping errors of STR markers was connected with samples collected from excrements.Identification panel Mini-DogFiller was selected for this study. It is first forensic identification system designed for identification of individuals from degraded DNA. Parameters of this system are from “DogFiler” adapted to standards of human forensic identification panels [[7]Kun T. Lyons L.A. Sacks B.N. et al.Developmental validation of Mini-DogFiler for degraded canine DNA.].2. Material and methods2.1 Biological material

From 170 samples of Czechoslovakian wolfdogs (CSW), which were source of allelic ladder too, 23 samples were randomly selected. These specimens were newly isolated from buccal swabs and independently tested in four repeats.

Wolf (W) DNA was isolated from excrements. From 41 samples obtained from universities and ZOOs from the Czech Republic and Poland, 39 samples were used for the analysis. These samples were isolated and tested in four independent runs. Isolation of DNA was performed by commercial kits and was proceed according to manufacturer recommendation. For DNA isolation from buccal swabs was used NucleoSpin Tissue kit (Macherey-Nagel) and for excrement DNA QIAmp DNA Mini Kit (Qiagen).

2.2 Optimalization and PCR reaction

For this study, A and B panels of Mini-DogFiler were selected. Primers and subsequently whole panels were tested on four specimens firstly on buccal swab and secondly on excrements. Primers were tested on 8 genotypes, firstly in singleplex and after that in multiplex. For validation of parallel runs, allelic ladder, developed on the department of genetics and breeding (CULS) by mixture of genotypes which represented majority of founded alleles, was used.

Amplification was done in multiplex PCR manner. The final compound of reaction mixture (10 μl) was as followed: 1x Multiplex PCR Plus Kit (Qiagen), 1 μl primer mix made of 7 (panel A) or 5 (panel B) pairs of primers, 10 ng DNA and 3 μl ddH2O. The difference in the PCR mixture for the excrements was in the amount of DNA 4 μl.

Before separation by capillary electrophoresis, PCR products were diluted 1:9 or 1:19 by ddH2O. After that 1 μl diluted PCR product was mixed with 12 μl Hi-Di formamide and 0.2 μl size standard GeneScan LIZ600 (Life Technologies). PCR was performed on Termocycler C1000 (Bio-Rad). The PCR parameters for buccal samples consisted of an initial 1 min denaturation at 95 °C followed by 35 cycles of: 30 s denaturation at 94 °C, annealing for 90 s at 60 °C, extension for 1 min at 72 °C. A final extension was performed for 80 min at 72 °C with 4 °C hold until the plate was removed from the thermocycler.

2.3 StatisticsBasic population descriptors like number of alleles per locus (k), heterozygosity observed (HO) and expected (HE) were calculated by program Cervus v.3.0.3 [[8]Kalinowski S.T. Taper M.L. Marshall T.C. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment.]. Potential departure from H-W equilibrium and null allele frequency were estimated both Cervus v.3.0.3 and MicroChecker v2.2.3 softwares [[9]van Oosterhout C. Hutchinson W.F. Wills D.P. et al.MICRO‐CHECKER: software for identifying and correcting genotyping errors in microsatellite data.]. Capability to distinguish individual genotypes was tested by pairwise comparison of each individual in Cervus v.3.0.3. Three metrics suggested by Pompanon et al. [[10]Pompanon F. Bonin A. Bellemain E. et al.Genotyping errors: causes, consequences and solutions.] were applied for evaluation of genotyping errors, it included mean error rate per allele (ea), mean error rate per locus (el) and error rate per multilocus genotype (eobs).3. Results

During genotyping process possibly all situations, that might lead to errors, were noticed – environmental contamination of DNA sample, allelic dropout, PCR artefacts and technical null allele. Although these phenomena prevails in low quality and problematic samples, technical null allele was identified also for DNA samples from buccal swabs. Overall, 23 samples of CSW in four runs, 92 replicates, were successfully amplified.

The allele and genotype error values for each panel A and B locus were with an average allele error of 0.09 % and an average genotype error of 0.18 %. In terms of the strictest criterion, the error rate on the genetic profile was 2.17 %. Furthermore, the occurrence of a technical null allele at the VGL2136 locus was found in 9.8 % of the genotypes. This phenomenon did not lead to error bias because no allele of the respective locus was amplified and therefore heterozygotes were not mistyped as homozygotes.

In wolf samples, only 8 (panel A) and 7 (panel B) genotypes were repeatedly successfully amplified in at least two replicates out of 4. The occurrence of allelic dropout at several loci was also detected, namely VGL2009 (6.25 % genotypes), VGL3112 (68.75 %) and VGL1606 (31.25 %).

4. Conclusions

DNA isolated from buccal swabs had very low error rate of the amplified products. Nevertheless, allelic dropout was detected in the VGL2136 STR locus, the cause of which is unclear. But it seems this phenomenon was not caused by mutation because the occurrence of the dropout was noticed only in one of the four repeats. Only 2 alleles were detected at the VGL1228 locus, indicating its low variability. This fact raised doubt on the suitability of using the Mini-DogFiler microsatellite system for the VGL1228 locus in the Czechoslovakian Wolfdog.

Results of the analysis with DNA isolated from wolf excrements are strongly biased because the final number of suitable samples was too small. The biggest problem has been the amplification of loci that amplify large alleles. VGL1063, VGL1541, VGL3008 loci showed very low variability (2 alleles). The largest number of alleles (4 alleles) was detected in the VGL2009 and VGL3438 loci.

In wolf samples only 8 (panel A) and 7 (panel B) genotypes were repeatedly amplified in at least two replicates of four. The occurrence of allelic dropout at several loci was also detected VGL2009 (6.25 % genotypes), VGL3112 (68.75 %) and VGL1606 (31.25 %), which is probably due to strong DNA fragmentation, since they are the largest loci amplicon sizes.

5. DiscussionBefore the start of this study, the "ASCH" identification panel [[11]van Asch B. Alves C. Gusmão L. et al.A new autosomal STR nineplex for canine identification and parentage testing.] was also tested and validated, and the Mini-DogFiler panel was then tested on the same protocol.

First, it was important to equalize the primer concentrations themselves, and then the amount of DNA (3−10 ng per reaction) and the number of PCR cycles (30–40) were tested. PCR artefacts were identified in 40 cycles. On the basis of testing, a methodology was proposed which had constant and relevant results also in archival DNA isolates case. Consequently, 35 PCR cycles with 10 ng of DNA should correspond to 4065 copies of a given locus.

Further testing and validation, consistent with procedures in human forensic genetics, would be needed to enhance the quality and relevance of this study [[12]Hansson O. Gill P. Egeland T. STR-Validator: an open source platform for validation and process control.].Funding source

This project was supported by Internal Grant Agency Czech University of Life Science in Prague, SGS project SV19-07-21360.

Declaration of Competing Interest

The authors of this manuscript declare no conflict of interest.

ReferencesKusak J. Fabbri E. Galov A. et al.

Wolf-dog hybridization in Croatia.

Vet. Arh. 88: 375-395

Beyond wild and domestic: human complex relationships with dogs, wolves, and wolf-dog hybrids.

Hybrid Communities. Routledge, : 65-79Salvatori V. Godinho R. Braschi C. et al.

High levels of recent wolf× dog introgressive hybridization in agricultural landscapes of central Italy.

Eur. J. Wildl. Res. 65: 73Canu A. Mattioli L. Santini A. et al.

‘Video-scats’: combining camera trapping and non-invasive genotyping to assess individual identity and hybrid status in gray wolf.

Wildlife Biol. : 4Lorenzini R. Fanelli R. Grifoni G. et al.

Wolf–dog crossbreeding:“Smelling” a hybrid may not be easy.

Mamm. Biol.-Zeitschrift für Säugetierkunde. 79: 149-156Sundqvist A.K. Ellegren H. Vilà C.

Wolf or dog? Genetic identification of predators from saliva collected around bite wounds on prey.

Conserv. Genet. 9: 1275-1279Kun T. Lyons L.A. Sacks B.N. et al.

Developmental validation of Mini-DogFiler for degraded canine DNA.

Forensic Sci. Int. Genet. 7: 151-158Kalinowski S.T. Taper M.L. Marshall T.C.

Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment.

Mol. Ecol. 16: 1099-1106van Oosterhout C. Hutchinson W.F. Wills D.P. et al.

MICRO‐CHECKER: software for identifying and correcting genotyping errors in microsatellite data.

Mol. Ecol. Notes. 4: 535-538Pompanon F. Bonin A. Bellemain E. et al.

Genotyping errors: causes, consequences and solutions.

Nat. Rev. Genet. 6: 847-859van Asch B. Alves C. Gusmão L. et al.

A new autosomal STR nineplex for canine identification and parentage testing.

Electrophoresis. 30: 417-423Hansson O. Gill P. Egeland T.

STR-Validator: an open source platform for validation and process control.

Forensic Sci. Int. Genet. 13: 154-166Article InfoPublication History

Published online: November 26, 2019

Accepted: November 19, 2019

Received in revised form: November 19, 2019

Received: September 17, 2019

Identification

DOI: https://doi.org/10.1016/j.fsigss.2019.11.014

Copyright

© 2019 Elsevier B.V. All rights reserved.

ScienceDirectAccess this article on ScienceDirect Related Articles