Never the Two Shall Mix: Robust Indel Markers to Ensure the Fidelity of Two Pivotal and Closely-Related Accessions of Brachypodium distachyon

Brachypodium distachyon is an established model for monocotyledonous plants. Numerous markers intended for gene discovery and population genetics have been designed. However to date, very few indel markers with larger and easily scored length polymorphism differences, that distinguish between the two morphologically similar and highly utilized B. distachyon accessions, Bd21, the reference genome accession, and Bd21-3, the transformation-optimal accession, are publically available. In this study, 22 indel markers were designed and utilized to produce length polymorphism differences of 150 bp or more, for easy discrimination between Bd21 and Bd21-3. When tested on four other B. distachyon accessions, one case of multiallelism was observed. It was also shown that the markers could be used to determine homozygosity and heterozygosity at specific loci in a Bd21 x Bd3-1 F2 population. The work done in this study allows researchers to maintain the fidelity of Bd21 and Bd21-3 stocks for both transgenic and nontransgenic studies. It also provides markers that can be utilized in conjunction with others already available for further research on population genetics, gene discovery and gene characterization, all of which are necessary for the relevance of B. distachyon as a model species.


Introduction
Monocots are considered to be a highly valuable clade of plants. These organisms, such as orchids, lilies, bananas and oil palm, from the families Orchidaceae, Liliaceae, Musaceae and Arecaceae, respectively, have aesthetic, nutritional and medicinal uses, and therefore command a niche market with substantial economic value [1][2][3][4][5]. Another monocotyledonous family with a wide geographic range, used for food, feed, forage and fuel is the Poaceae (grass family) [6][7][8][9]. Within the grass family, the Triticeae tribe is well known for its significant contribution to global food security [10,11]. In order for Triticeae production to keep pace with or exceed demand, the continued exploration of critical developmental pathways and mechanisms for stress tolerance and yield is imperative [12][13][14][15][16][17][18].
This research can be accelerated with the use of model systems. As defined by Leonelli and Ankeny [19], a model organism is used to represent a larger group of species in order to investigate several pathways (genetic, developmental, physiological, evolutionary and ecological), and reach conclusions that can be extrapolated to the species in the group. To this end, the model organism must be small in size, be very fertile, have a rapid life cycle, be easy and cheap to maintain, and be amenable to genetic modification. In addition, infrastructure such as stock centers, easily accessible databases, and rules and regulations governing the utilization of standardized materials, must be in previous researchers can be useful in differentiating between Bd21 and Bd21-3, the process can be expedited with larger amplicon size variations which are easier to score. This has been shown in Arabidopsis, where Indel Group in Genomes (IGG) markers have been utilized for selectivity between Arabidopsis ecotypes Columbia (Col-0) and Landsberg erecta (Ler-0) [52]. IGG markers differ from indels because the former, predominantly located upstream and downstream of the translation start and stop sites, can contain several indels (called Indel Groups) with cumulative lengths of 1500 bp or more. This increases the resolution of genotyping due to larger amplicon size differences easily scored with lower percentage agarose gels [52].
In the current study, we leveraged the availability of the genomes of Bd21 and Bd21-3, the knowledge that indels were available to discriminate between these closely related accessions, as well as the demonstration of larger indels in Arabidopsis and B. distachyon [37,51,52], to develop 22 large indel markers upstream and downstream of designated genes, for an efficient discrimination between Bd21 and Bd21-3. There was also one incidence of multiallelism among the B. distachyon accessions tested. The markers developed will ensure that researchers can easily and efficiently maintain stocks of Bd21 and Bd21-3 for transgenic studies, as well as aid in genotyping the progeny of biparental crosses.

Identification of Indels or Indel Groups and Marker Development From Brachypodium distachyon Putative Orthologs of Arabidopsis thaliana Genes
Putative B. distachyon orthologs of A. thaliana genes were used to generate 22 indel markers from all five chromosomes for easy discrimination between Bd21 and Bd21-3 (Table 1 and Table S1). The only exception was BdindelWSU_8 obtained from one of the Bd21 candidate genes (Bradi3g00757) listed by Cui et al. [46], and its Bd21-3 ortholog from the Phytozome database (Version 12.1.6, default settings: https://phytozome.jgi.doe.gov/pz/portal.html). Alignments of upstream and downstream regions of the 22 curated genes yielded single or cumulative gaps of 150 nucleotides or more ( Figure S1). Primers were then successfully designed, flanking these regions, and then in silico PCR was used to predict the amplicon sizes of the two accessions (Table 1, Figure S1). Out of the 22 markers, 13 were designed from upstream regions, and 9 were designed from downstream regions of the designated genes (Table S1). In addition, for half of these markers, the predicted amplicon size was greater in Bd21 than in Bd21-3 (Table 1 and Table S1).

Marker Assessment on Six Brachypodium distachyon Accessions and F2 Population
All 22 markers tested produced amplicons that allowed for the discrimination between Bd21 and Bd21-3 ( Figure 1, Table 2 and Table S2). When tested against four other B. distachyon accessions, in most cases, the amplicon sizes followed those displayed by Bd21 or Bd21-3 (Table 2, Figure S2). There was one exception, (BdindelWSU_12), where multiallelism was observed within the six accessions tested ( Figure 2 and Figure S2, Table 2 and Table S2). Two markers (BdindelWSU_15 and BdindelWSU_22) were tested on a Bd21 x Bd3-1 F2 population to show homozygosity and heterozygosity, highlighting the efficiency of the markers designed ( Figure 3).

Discussion
Since the inception of B. distachyon as a model system, many tools including markers have been developed to aid in research and discovery. Markers are available for the discrimination of different species in the genus Brachypodium, and within multiple populations of the same species [44,[48][49][50]54]. In many cases, the length polymorphisms require the use of high percentage agarose gels. Following the work that has been done with indel markers in Arabidopsis as well as the evidence provided for larger indels in the Brachypodium genus [51,52], the current study aimed to design indel markers with sufficient length polymorphisms easily visualized on low percentage agarose gels, for the two highly utilized accessions needed for functional gene characterization, Bd21 and Bd21-3.
The parameters for these indel markers were as follows: (1) The markers had to be upstream or downstream of a designated gene. This was done to avoid designing primers from gene families which could lead to multiple amplification products. Although there is literature showing that an intron can be used to discriminate within the Brachypodium genus [50], the primers were designed

Discussion
Since the inception of B. distachyon as a model system, many tools including markers have been developed to aid in research and discovery. Markers are available for the discrimination of different species in the genus Brachypodium, and within multiple populations of the same species [44,[48][49][50]54]. In many cases, the length polymorphisms require the use of high percentage agarose gels. Following the work that has been done with indel markers in Arabidopsis as well as the evidence provided for larger indels in the Brachypodium genus [51,52], the current study aimed to design indel markers with sufficient length polymorphisms easily visualized on low percentage agarose gels, for the two highly utilized accessions needed for functional gene characterization, Bd21 and Bd21-3.
The parameters for these indel markers were as follows: (1) The markers had to be upstream or downstream of a designated gene. This was done to avoid designing primers from gene families which could lead to multiple amplification products. Although there is literature showing that an intron can be used to discriminate within the Brachypodium genus [50], the primers were designed upstream and downstream of the start and stop codons to increase the likelihood of robust and unequivocal results. This approach also increased our understanding of promoter region differences between Bd21 and Bd21-3 that could be exploited in future gene expression variation studies. (2) The markers needed to amplify in both accessions in order to avoid diagnostic ambiguity created by reaction failure. (3) The predicted amplicon sizes were between 150 bp and 2000 bp, a range easily covered by a cost-effective Taq polymerase routinely used in PCR. This would potentially increase the utility of the markers by multiple research groups. (4) The amplicon size difference generated by each marker had to be 150 bp or more for easy discrimination on a 1.5% agarose gel.
All 22 markers tested showed discrimination between Bd21 and Bd21-3 (Figure 1). With the exception of one case (BdindelWSU_3), the predicted amplicon sizes were obtained (Tables 1 and 2). With regard to the deviation from the expected amplicon size, this was a nonissue, as distinct and sizeable length polymorphism differences were still observed between Bd21 and Bd21-3, thus fulfilling the primary objective of the study. In some cases, other amplicons besides the ones predicted were noted (Figure 1, Figure S2, Table S2). This observation did not affect the primary aim of the study, as very distinct patterns could be observed between Bd21 and Bd21-3. It must be noted that these markers are yet to be tested on different populations of the same accession from a wide geographic range, or on other species in the Brachypodium genus. The expectation is that such analyses would yield multiple alleles, as has been shown by other studies [48,50,54]. The in silico sequence analyses showed that some of the regions selected for marker development were distant from areas that could reasonably be considered to be the promoter region for the designated gene (within 2 Kb upstream of the translation start site), and may in fact have been part of the structure of genes in a different orientation ( Figure S1). Out of the 22 pairs of sequences aligned, 13 were upstream of the start codons and 12 of these showed indels within 2 Kb of the designated gene, and may help to explain expression differences between these two accessions (Table S1, Figure S1). Conducting the marker development for discrimination between Bd21 and Bd21-3 on a larger scale with higher throughput resources, could show more promoter regions with indels that can explain transcriptomic differences between these two accessions, and contribute clues regarding observed phenotypic differences (e.g., transformation efficiency) [37,43]. This approach has merit as genome-wide scans have shown that single nucleotide polymorphisms can give clues about the environmental adaptation of genes [61,62]. Our approach could also help with comparative transcriptomics between Bd21 and Bd21-3 by ensuring that the starting plant materials used for the analyses are accurate, and that the data reported in the scientific literature are precise.
When the markers were screened on four other B. distachyon accessions, one case of multiallelism (BdindelWSU_12) was observed ( Figure 2, Table 2). Locus homozygosity and heterozygosity was also noted when selected markers (BdindelWSU_15 and BdindelWSU_22) were used on a Bd21 x Bd3-1 F2 population ( Figure 3). The observed multiallelism in different accessions indicates that our approach in conjunction with other markers developed can be utilized on a larger scale to verify accessions slated for biparental crosses. Broadening the scope of this study would add to the body of marker types available for population genetics and gene discovery. The markers developed ensure that researchers are able to optimize their B. distachyon transformation pipelines by actively selecting for the more efficient accession, Bd21-3. Researchers will therefore protect the accuracy of the transgenic data generated. For example, phenotypes (e.g., early or late flowering) obtained from overexpression and knockdown of a gene can be accurately reported as being a result of the transgenic event rather than cross-contamination of the accessions used for the independent transformation events. Researchers can also ensure the generation of transgenic or gene-edited lines in a genetic background consistent with those of publicly available stocks of T-DNA lines. This prevents false positives and negatives as well as the loss of valuable time and resources.
There is also the potential for cross-contamination during material transfer from one researcher to another. Rather than a reliance on unpredictable and environment-dependent morphological differences, our markers give a quick way to ensure that the right accession is being given or received. There is a very valid argument of simply obtaining Bd21 and Bd21-3 from seed banks. However some seed banks have a policy of denying repeated requests for the same material. It is the responsibility of the researcher to propagate and maintain the seed stocks, and the markers developed in the current study aid with fidelity during cultivation and stock maintenance. In addition, seed viability can decrease over time, and the markers developed in the current study help researchers maintain the desired accession fidelity long after the original seed material has been completely exhausted via repeated propagation. case of multiallelism) and visualized using the Bio-Rad ChemiDoc ™ Touch Imaging System under the ethidium bromide, faint band setting. The GoldBio ® 50 bp DNA ladder and 1 Kb PLUS ™ DNA ladder were used to estimate the amplicon sizes. For the four accessions (Bd3-1, Bd2-3, Bd30-1, Bd1-1) where no predictions were made before the markers were tested, the size determination was dependent upon comparison to the results of Bd21 or Bd21-3, as well as on the DNA ladders used.

Conclusions
There are many markers that have been developed for both population genetics and gene discovery that can distinguish between species in the Brachypodium genus. Generally the indel markers that have been developed tend to have small amplicon size differences, and very few of them have been reported to specifically distinguish between the transformation-optimal accession Bd21-3 and the reference genome accession Bd21. The current study sought to increase the knowledge in this area, and to this end, 22 indel markers were developed to distinguish between Bd21 and Bd21-3 with an amplicon size difference minimum of 150 bp. In addition, an incidence of multiallelism was observed, indicating that the approach utilized can be translated to other brachypodium accessions in the same species. The markers developed provide a quick, cost-efficient and easy way to score samples and accurately maintain the fidelity of Bd21 and Bd21-3, which are extremely critical to population genetics, evolutionary studies and functional gene characterization, all of which ensure that B. distachyon maintains its relevance as a model monocot.