A novel set of polymorphic chloroplast microsatellite markers for northern red oak (Q. rubra L.)

Abstract

Oaks are model species due to their importance in various ecosystems, worldwide distribution, high economic value and emerging genomic resources. As such, knowledge on population differentiation across their distribution range is of high importance for sustainable forest management. As in most angiosperms, chloroplasts are maternally (via seeds) inherited in oak species and acorns are dispersed over comparably short distances. Consequently, chloroplast markers reveal comparatively high differentiation between populations, making them highly viable for the analysis of historic migration patterns and the certification of reproductive material. Despite the existence of various white oak (section: Quercus) chloroplast markers, red oak (section: Lobatae) chloroplast markers remain limited. Northern red oak is one of the most important North American oak species and a widespread non-native plantation tree species in European forests. We took advantage of chloroplast genomes of Q. rubra L. and related oak chloroplast genomes to develop a set of robust and easy-to-use chloroplast microsatellite primers for northern red oak. Furthermore, we tested transferability of those novel red oak primer pairs to Q. robur L. and Q. petraea Matt. Liebl. The new set of fifteen polymorphic chloroplast microsatellite markers revealed three additional northern red oak haplotypes after screening 80 northern red oak individuals representing seven haplotypes, identified based on formerly available markers. Therefore, they provide a higher resolution of haplotypes as compared to currently available markers.

Introduction

Northern red oak is one of the most important North American lumber species. It ranges from the peninsula of Nova Scotia to Minnesota and south to eastern Oklahoma, Alabama, and North Carolina (Sander, 1990). By now, northern red oak has become widespread in European forests with notable populations in Germany (Schmitz, 2014), and various other countries (Nicolescu et al., 2020).

Previous studies identified northern red oak haplotypes using PCR-RFLP (Aldrich et al., 2003; Magni et al., 2005; Feng et al., 2008; Birchenko et al., 2009; Pettenkofer et al., 2020) and cpSSR markers (Zhang et al., 2015; Pettenkofer et al., 2019; Götz et al., 2020), which were originally developed for angiosperms (Weising and Gardner, 1999) or white oak species (Deguilloux et al., 2002; Zhang et al., 2015; Pettenkofer et al., 2020). Chloroplast markers are commonly used for the certification of reproductive material due to their maternal inheritance (via seeds) and strong phylogeographic signature (Petit et al., 2002; Chmielewski et al., 2015). Here, we introduce fifteen novel and easy-to-score polymorphic cpSSRs for Q. rubra chloroplast haplotype identification. We also tested the transferability of those primers to the closely related species Q. ellipsoidalis E. J. Hill and the white oak species Q. robur L. and Q. petraea (Matt.) Liebl. (section Quercus).

Experimental

Chloroplast genome

The Northern red oak chloroplast genome was first completely sequenced in 2014 (Alexander and Woeste, 2014) and again in 2019 (Pang et al., 2019). Chloroplast genome sequences of additional oak species from several subsections were also included in the sequence alignment (online Supplementary Table S1). All sequences were downloaded from the NCBI genomic database (Sayers et al., 2020) in November 2021.

Plant material and haplotypes

Northern red oak DNA was obtained from previously conducted red oak studies with known haplotypes based on cpSSR and cpCAPS (PCR-RFLP) analyses (Pettenkofer et al., 2020; online Supplementary Table S2), Q. ellipsoidalis DNA from a previous study (Lind and Gailing, 2013), Q. robur samples with known chloroplast haplotypes (Burger et al., 2021 and unpublished data), and sessile oak samples from three geographic regions were used for transferability tests.

Motif detection and primer design

A red oak chloroplast genome (Alexander and Woeste, 2014) was browsed for repetitive motives (minimum six repetitions for mononucleotide, and three for longer repeat motives) using the tandem repeats finder online application (Benson, 1999). An alignment of Quercus chloroplast genome sequences was performed using the Qiagen CLC Genomics Workbench (Hilden, Germany). All detected candidate loci were visually verified for the variability of the potential SSR, revealing 44 promising candidates. Primer design was done using the NCBI Primer blast online application (Ye et al., 2012) with all available chloroplast reference genomes of Quercus rubra (taxid:3512). Target fragment size range was set from 70 to 500 bp, melting temperature (T _m) between 57 and 63°C with differences ⩽3°C between both primers and at least two mismatches to unintended targets within the last five base pairs of the 3′ end. All primer pairs were modified with an appendix (Forward primer: 5′-CACGACGTTGTAAACGAC-3′; reverse primer: 5′-GTTTCTT-3′). This enabled the use of HEX (Sigma Aldrich; St. Louis, MO, USA) or 6-FAM™ (Sigma Aldrich; St. Louis, MO, USA) labelled universal M13 primers for economic reasons (Schuelke, 2000).

Primer testing

For primer testing, PCRs were conducted following the touchdown PCR protocol and reagent mix of Götz et al. (2020) in volumes of 14 μl, each using 1 μl of genomic DNA (5–10 ng), 1 μl labelled M13 forward primer (5 μM), 0.2 μl of the designed forward primer (5 μM), and 0.5 μl reverse primer (5 μM). Primer pairs producing distinguishable fragment sizes were multiplexed (Table 1). Electrophoretic separation for allele scoring of 1:150 diluted PCR products was done using an ABI Genetic Analyzer 3130xl (Applied Biosystems, Foster City, USA) with the GeneScan™ 500 ROX™ internal size standard (Applied Biosystems, Foster City, USA) using GeneMapper V 3.7 (Applied Biosystems, Foster City, USA).

Table 1. Primer sequences and fragment size range for allele binning of novel polymorphic primer pairs

Primer sequences of monomorphic markers are reported in online Supplementary Table S3.

^a Including M13 primer appendix. Fragment sizes will be shorter if labelled primers without M13 tail (online Supplementary Table S3) are used.

^b Additional white oak allele: 188.

^c Comparatively poor amplification during PCR.

Initial tests were conducted for eight samples of four different chloroplast haplotypes (Pettenkofer et al., 2020). Polymorphic markers were subsequently multiplexed in 80 northern red oak samples of seven different haplotypes (Table 2, online Supplementary Table S2), and eight white oak samples with two and three previously defined haplotypes, respectively (online Supplementary Tables S2 and S4). Three of the fifteen polymorphic markers (QRcp01, QRcp07, QRcp17) exhibited various alleles within previously defined widely distributed red oak chloroplast haplotypes A.1 and A.3 (Götz et al., 2020), revealing sub-haplotypes A.1_1, A.1_2, A.3_1, A.3_2, and A.3_3 (Table 2, online Supplementary Table S4).

Table 2. Origin and novel haplotypes (H_N) of northern red oak populations of previously defined haplotypes (H_P, following Götz et al., 2020)

Alleles of each haplotype are reported in online Supplementary Table S4.

¹ Götz et al. (2020), ² Pettenkofer et al. (2019), ³ Liesebach and Schneck (2011), ⁴ Lind-Riehl et al. (2014), ⁵ Lind and Gailing (2013).

^a Taken from a German provenance trial.

Discussion

Since our novel primer pairs enabled the distinction of additional haplotypes within two common northern red oak haplotypes of the native range, they provide a powerful tool for further differentiation between northern red oak chloroplast haplotypes. Reproductive material for the establishment of forest plantations, such as for northern red oak, is increasingly demanded due to climate change. However, certified plantations within Europe are rather limited (Steiner, 2012). German red oak probably originated from a limited geographic range within the species' natural distribution (Pettenkofer et al., 2019; Götz et al., 2020). Additional chloroplast markers will be valuable tools to identify novel and rare haplotypes with a restricted native range as a prerequisite to narrow down the geographic origin of reproductive material and potential admixture. German red oak population haplotypes, for example, consist predominantly of haplotype A.1 (Pettenkofer et al., 2019; Götz et al., 2020). Our sample test set included samples with haplotype A.1 from different states of Germany (Brandenburg), Canada (Ontario), and the USA (Wisconsin, Tennessee). Interestingly, we observed an association between populations´ geographic origin and novel sub-haplotypes. Sub-haplotype A.1_1 was observed in all individuals from Germany and Ontario. Wisconsin and Tennessee populations, on the other hand, consisted of samples of sub-haplotype A.1_2. This result provides further evidence for the capability of our marker set to narrow down the origin of European red oak and supports the assumption of a geographically restricted origin of German Q. rubra stands (Pettenkofer et al., 2019; Götz et al., 2020).

All fifteen polymorphic red oak markers amplified in the closely related species Q. ellipsoidalis, but were monomorphic in the tested samples. Additional alleles might be revealed in future Q. ellipsoidalis investigations. All primer pairs amplified in both white oak species and ten primer pairs were polymorphic (Table 1, online Supplementary Table S3). We observed comparable variation within white oak and red oak samples (32 versus 31 total alleles for all polymorphic markers), even though white oak samples were underrepresented (16 versus 80 samples). This result supports the observation of higher variation in white oak haplotypes as compared to red oaks (Birchenko et al., 2009). However, we observed no differentiation within previously defined white oak haplotypes (online Supplementary Fig. S3). Our new cpSSR markers could be used as a fast and effective method to define chloroplast haplotypes in North American red oaks as well as in European white oaks and might potentially differentiate white oak haplotypes in larger datasets.