Primers for Castilleja and their utility across Orobanchaceae: II. Single-copy nuclear loci.

PREMISE OF THE STUDY: We developed primers targeting nuclear loci in Castilleja with the goal of reconstructing the evolutionary history of this challenging clade. These primers were tested across other major clades in Orobanchaceae to assess their broader utility. METHODS AND
RESULTS: We assembled low-coverage genomes for three taxa in Castilleja and developed primer combinations for the single-copy conserved ortholog set (COSII) and the pentatricopeptide repeat (PPR) gene family. These primer combinations were designed to take advantage of the Fluidigm microfluidic PCR platform and are well suited for high-throughput sequencing applications. Eighty-seven primers were designed for Castilleja, and 27 were found to have broader utility in Orobanchaceae.
CONCLUSIONS: These results demonstrate the utility of these primers, not only across Castilleja, but for other lineages within Orobanchaceae as well. This expanded molecular toolkit will be an asset to future phylogenetic studies in Castilleja and throughout Orobanchaceae.

Although the plastome has long been considered the workhorse of phylogenetic inference in plants, reliance on chloroplast data alone may limit the ability to identify inheritance patterns of polyploids, as well as introgression and hybridization events (Godden et al., 2012; Twyford and Ennos, 2012). The reliable estimation of an underlying species tree also depends upon the acquisition of multiple, unlinked loci—especially for recent and rapid radiations—shifting the focus toward the development of single- or low-copy nuclear gene regions for phylogenetic analyses. The process of identifying and developing these nuclear gene regions using traditional methods can be time consuming and costly, but the increasing availability of high-throughput sequencing data, as well as new bioinformatic approaches, allows for the efficient and cost-effective exploration of the nuclear genome.

Here we focus on developing a suite of putatively single-copy nuclear gene regions in Castilleja L. (Orobanchaceae; “the paintbrushes”), a clade rich with polyploid and hybrid taxa, and the product of an ongoing rapid radiation (Tank and Olmstead, 2008, 2009). Previous studies using the nuclear ribosomal ITS and ETS regions, the low-copy nuclear gene waxy, and the plastid trnL-F and rps16 intron regions hinted at cytonuclear discordance in some taxa, and most relationships among closely related taxa were unresolved (Tank and Olmstead, 2008, 2009). We recently developed primer combinations targeting the most variable regions of the plastome in Castilleja (Latvis et al., 2017a), and now present a companion set of nuclear primers with the goal of obtaining a resolved species tree for this challenging clade, as well as to aid in the detection of introgression and hybrid speciation. Primers for microsatellite markers have also been developed by Fant et al. (2013) for population-level investigations. In part, we follow the approach outlined by Blischak et al. (2014) to develop nuclear primer combinations from genome-skimming data, while following specifications for the Fluidigm Access Array microfluidic PCR system (Fluidigm, South San Francisco, California, USA) (see Latvis et al., 2017a). Thus, all primer combinations use the same annealing temperature of 60°C and may be amplified in parallel prior to high-throughput sequencing or traditional Sanger sequencing. We specifically target putatively single-copy genes from the conserved ortholog set (COSII) and pentatricopeptide repeat (PPR) domains, both of which have been highlighted for their phylogenetic utility in plants (COSII: Wu et al., 2006; PPR domain: Li et al., 2008; Yuan et al., 2009, 2010). We also test these primers for their broader applicability across Orobanchaceae following the approach outlined in Latvis et al. (2017a) with the goal of finding a subset of nuclear gene regions that would amplify across this more inclusive clade. Previous phylogenetic studies within Orobanchaceae employed the nuclear phytochrome genes PHYA and PHYB, the nuclear ribosomal ITS region, and the plastid matK and rps2 genes. Orobanchaceae is the largest clade of parasitic angiosperms, and plastome reduction and accelerated rates of molecular evolution in retained plastid genes have been documented (see discussion in Bennett and Mathews, 2006). Additionally, phytochrome genes regulate responses to light and can be significantly modified in parasites (Bennett and Mathews, 2006). Therefore, the development of additional single-copy nuclear regions would provide a much-needed source of phylogenetic information in Orobanchaceae, and may provide a more reliable estimate of branch lengths for further studies of diversification and character evolution.

METHODS AND RESULTS

We assembled contigs from raw reads of three low-coverage Castilleja genomes, C. cusickii Greenm., C. foliolosa Hook. & Arn., and C. tenuis (A. Heller) T. I. Chuang & Heckard (Latvis et al., 2017a; National Center for Biotechnology Information [NCBI] Sequence Read Archive [SRA] accession SRP100222) using CAP3 (Huang and Madan, 1999) with the default settings. The accessions were sequenced as 100-bp single-end reads on an Illumina HiSeq 2000 (Illumina, San Diego, California, USA), yielding ∼12.5 million reads per taxon (Uribe-Convers et al., 2014) and an average depth of coverage of ∼0.8×. These taxa include both annual and perennial lineages of Castilleja and span the phylogenetic breadth of the clade (Tank and Olmstead, 2008, 2009). These assemblies were then culled to include only contigs of 1 Kb or larger using a custom R script. The culled assemblies and script are available from the Dryad Digital Repository (http://doi.org/10.5061/dryad.52v62; Latvis et al., 2017b).

To search for hits among our contigs, available COS sequences were obtained from Sol genomics (https://solgenomics.net), and PPR loci were mined from the Mimulus L. genome on Phytozome (Hellsten et al., 2013; https://phytozome.jgi.doe.gov) using the 127 PPR orthologs identified in Arabidopsis Heynh. by Yuan et al. (2009) as references. Both gene sets may be found in Uribe-Convers et al. (2016) and were used to construct local BLAST databases for the search (-makeblastdb). We used TBLASTX to search each Castilleja CAP3 assembly (with contigs of 1 KB or greater) against both the COS and PPR databases, indicating tab-delimited output (-outfmt 6). Output files were filtered for alignment length >200 and a maximum E-value of 1e-10, and were culled to include only unique hits.

Hits shared between C. cusickii, C. foliolosa, and C. tenuis were placed together into individual FASTA files (data available from the Dryad repository: http://doi.org/10.5061/dryad.52v62; Latvis et al., 2017b), imported into Geneious R7 version 7.0.6 (Kearse et al., 2012), and aligned with MAFFT version 7.017b under the default settings (Katoh and Standley, 2013). We designed primer pairs using Primer3 (Untergasser et al., 2012) using the same specifications for the Fluidigm Access Array system as Latvis et al. (2017a), but with a size range between 400–525 bp and an optimal size of 500 bp. We designed 10–30 primer pairs for each identified locus and prioritized them based on desired size and the presence of multiple G or C bases at the 3′ end of the primers (GC clamp). This also allowed us to design overlapping sets of primers with the potential to combine them after sequencing to produce longer contigs for downstream phylogenetic analyses. Suitable primer pairs were validated for Castilleja with PCR following the same amplification protocol and using the same Castilleja accessions as in Latvis et al. (2017a) and visualized on an agarose gel. We present 87 nuclear primer combinations specifically designed and validated for Castilleja (Table 1).

Table 1.

Nuclear primer pairs designed for Castilleja (locus and region amplified), amplicon lengths, and validation results for Orobanchaceae and outgroup taxon Paulownia. All pairs were designed for an annealing temperature of 60°C (±1°C). Boldfaced rows correspond to core Orobanchaceae primers, defined by successful amplification in two or more major clades in Orobanchaceae (see Fig. 1).

Locus (Region)	Primer sequences (5′–3′)a	Amplicon length (bp)b	Clade I: Lindenbergia sp.c	Clade II: Schwalbea americanac	Clade III: Orobanche californicac	Clade IV: Castilleja lineariloba, C. pumila, C. lemmoniid	Clade IV: Lamourouxia virgatae	Clade IV: Pedicularis sp.c	Clade V: Neobartsia peruvianac	Clade V: Rhinanthus alectorolophusc	Clade VI: Harveya purpureac	Clade VI: Physocalyx majorc	Paulowniaceae: Paulownia fortuneic (outgroup)
At2g34560_688F At2g34560_1186R (COSII)	F: CTGTTTGTGGCAGCAAGTACC	498	X	X	X	X	X	X	X		X	X	X
	R: AAAGGCTGTTGCAACTGAAGC
At5g46630_851F At5g46630_1350R (COSII)	F: TGACAAACCAGTTCCAAATGCA	499	X	X	X	X	X	X		X	X		X
	R: TTGGCAGGACGGGATTTAAGT
At3g04260_165F At3g04260_665R (COSII)	F: GGAGTGGGAAGTTGAATTGGC	500		X		X	X		X	X	X		X
	R: AAGGCTGAAGGCATAGGAGC
At5g12370_3513F At5g12370_4024R (COSII)	F: ATGCACTGCTGAAGCCCTTT	511		X	X	X	X		X	X			X
	R: GAGATGGCCTGACTGAAGCT
At2g28390_1096F At2g28390_1595R (COSII)	F: TGCAATGCAGGTTATCAGTCTTG	499	X	X	X	X	X	X					X
	R: GACATGGGATTGTATCTTGGCAG
At2g26430_2136F At2g26430_2632R (COSII)	F: TTGCAAGATTCCAAGCTTCTTGTC	496	X	X	X		X	X	X	X
	R: GCCGCCAGTTGTGTTTGG
At4g24190_1278F At4g24190_1777R (COSII)	F: CTGAAGTCCCAGATTTAGCAATAGTAC	499		X	X	X	X	X	X		X
	R: TCAGGCATTGGACAAGATTAGGT
At5g27620_538F At5g27620_1337R (COSII)	F: CCGCATAATCTTATCAACTTCCGC	799	X	X	X	X	X	X	X
	R: CTTGGGAAAGGGATTGAACAGG
At3g62010_1226F At3g62010_1725R (COSII)	F: CAACATCTGCTTTGCTCTCAGG	499	X	X		X			X	X	X
	R: GCTTCAGATGTTGATTCCATCCA
At5g48300_1121F At5g48300_1621R (COSII)	F: GCTCCAGGAATAACTTCACTTCC	500		X	X	X	X	X					X
	R: GCAGTAAGGCAATATTTGTGGTTG
At3g62010_1425F At3g62010_1926R (COSII)	F: GATAGCCATAGCATAAGCAACCC	501		X	X	X	X		X
	R: AGGCAATCTGGTGGTAGGATG
At4g24830_948F At4g24830_1447R (COSII)	F: AAACAAACAACCTCGCAGCC	499	X	X	X	X	X
	R: AAGTGGTCCTGGCCTACAGT
At3g04260_147F At3g04260_646R (COSII)	F: TCCTAGAAGATGGTCACAGGAGT	499		X		X	X		X	X
	R: CTCGTATAGCAGCCCGAAGT
At5g27620_1180F At5g27620_1680R (COSII)	F: CGAATCCTAAACTCTGCAATGAAAGA	500	X	X	X	X
	R: CGATGCCTTCAAATTTCCACGT
At5g26360_1312F At5g26360_1811R (COSII)	F: GGCATTGGTTTGGTTCTCACC	499		X	X	X	X
	R: TGTTGGGACTTGTGAAGAGTTGT
At2g29210_514F At2g29210_1014R (COSII)	F: TGAGACCATGGATTGCAAAGC	500		X		X	X			X
	R: GCTGTGCACTCAGAAGCAG
At3g62010.2_1674F At3g62010.2_2174R (COSII)	F: AGCCAAGCACTTACTCATCCT	500				X
	R: CAGATCTAAGGGAAGAGGCTGTAG
At3g62010_1674F At3g62010_2290R (COSII)	F: AGCCAAGCACTTACTCATCCT	616	X		X		X
	R: ATATCCTGTGTATAGCATGGCAACT
At2g38020_1797F At2g38020_2297R (COSII)	F: GTCGTATTAATGCACTGGACTTGT	500		X	X	X	X
	R: CAGCATGAGCAGCAACATCTG
At2g38770_387F At2g38770_886R (COSII)	F: TTTCTAACACGCCCAAGCCA	499			X	X			X
	R: CCTAGCTCTTCCCTGGGCA
At4g11120_1226F At4g11120_1716R (COSII)	F: TCCAGCAATACGACCTATGCC	490		X		X			X
	R: AATTTATTGCCTCAATTGCTTGGC
At1g80150_1016F At1g80150_1528R (PPR)	F: GCATGTAGGCCAGCAGCTTA	512				X	X					X
	R: GCCGTCGGGAAGGGTTTATA
At1g04200_1908F At1g04200_2395R (COSII)	F: GAGACGTACTTGCGTGAAAGC	487			X	X	X	X
	R: GACTCCTCTTTCAATGCCAGTG
At2g07050_1067F At2g07050_1534R (COSII)	F: GCCAAACTGTAAGAAGAGCCCA	467	X			X	X
	R: TGTAGAGAAAGAATAACATTCACAGCA
At4g02720_1652F At4g02720_2152R (COSII)	F: CGAAATCTGTTTCCGAGAGCG	500				X	X				X
	R: CCACCTCACCTCTACGTGG
At3g09920_917F At3g09920_1416R (COSII)	F: CATCCTCCATCGACTCTGAGG	499				X				X
	R: TGCACGGAGTTAAGAATTCATCG
At3g09920_1316F At3g09920_1859R (COSII)	F: ACGTCCCAATGAAGAGCCTT	543			X	X	X
	R: GTCGATATGCGGAAATGATGCC
At5g26360_322F At5g26360_818R (COSII)	F: TTCACCGGTATTCCCATCTATGC	496				X					X
	R: ATGCTATGTCTGTGGCACGA
At5g49970_503F At5g49970_998R (COSII)	F: AGCAGCTACAAGACCATCGC	495				X	X
	R: AGCTGCGGAAATTGATGAAATCC
At5g52210_456F At5g52210_955R (COSII)	F: ATCCAGCACGAACCCTCAAC	499				X	X
	R: GGATTTGCAAGGAGCTCCAC
At1g09620_1187F At1g09620_1688R (COSII)	F: GGTCTACAGCGAGCCTGAAA	501				X	X
	R: ACCCGACACCCTCAGATCAA
At1g09620_1490F At1g09620_1989R (COSII)	F: AGCTCATATGTTGCAGGGAGG	499				X	X
	R: CCTCCTCCATCCATGCAATCT
At2g28190_483F At2g28190_983R (COSII)	F: GTTTCCCAGGTCACCCGC	500				X	X
	R: TGCAACCTGTTTCCTGAATGG
At2g38020_892F At2g38020_1391R (COSII)	F: TTTGACATTTCCTGTTGCTTACCA	499				X	X
	R: CGTACAGCATTTAAGACGCGC
At4g22670_93F At4g22670_589R (COSII)	F: GAATTTGGTTGTTGGCATACAGC	496				X	X
	R: AGTCTAGCTTTGACGCCACA
At5g26360_1322F At5g26360_1821R (COSII)	F: TGGTTCTCACCTTTCTTGTACTCA	499				X	X
	R: GGTGCGATTATGTTGGGACTTG
At2g29210_902F At2g29210_1401R (COSII)	F: TTGAATGCAGGCAGTTAATGGAA	499				X	X
	R: ACCGCCCTACAGCAGTTTG
At1g01940_828F At1g01940_1227R (COSII)	F: TCATAACTATGTAAGCCAACCATGATT	399				X	X
	R: TATAAAGGGTTTCATGATACAAGGTGG
At4g24620_1780F At4g24620_2223R (COSII)	F: CTCGTTCCCTGTATACCAAGATTTG	443				X	X
	R: CAGTTATATTCAACAGCTCAGAGATGG
At5g62530_439F At5g62530_938R (COSII)	F: TGGAACAAATGCTTCGATTACTCC	499				X	X
	R: TGCCGGTGAAGAGAGTCATG
At3g62010_1468F At3g62010_1969R (COSII)	F: ACAAATAGATATGACCCTGCCGA	501					X
	R: GCACATGGAGGCACGTTTG
At4g11120_1399F At4g11120_1859R (COSII)	F: CCCATGTGAAGGTGCGGATA	460				X
	R: GCTTGTGCAGCCATTCCAG
At2g41680_402F At2g41680_902R (COSII)	F: GTACACAAACAAGTCTCGGGC	500				X
	R: GGATCCGGATGTATTGCTGCT
At5g22830_933F At5g22830_1432R (COSII)	F: AAAGAGCCTAATGCATCTATTGTGAAC	499				X
	R: ACCAGTCTCTGTTTGCTAATACGA
At3g26730_1284F At3g26730_1770R (COSII)	F: AGATAACATCTTATCTGGATCCATGGA	486				X
	R: TGTTTCAGGACAAGTGGCTCAG
At2g17670_501F At2g17670_998R (PPR)	F: ATCCAATAGCCCGGCCTTAC	497				X
	R: AAGGAGTTCGGTATAAAGCCTGAT
At4g37170_962F At4g37170_1466R (PPR)	F: CAACTCATACCGGGTTTCTTCAC	504				X
	R: ATGGGTATGCTCAGAATGGTCAG
At4g30825_1603F At4g30825_2102R (PPR)	F: GGTGACATGGTGATGATCGGT	499				X
	R: TTCAAAGACCCGGACTTGACA
At3g14730_878F At3g14730_1378R (PPR)	F: ACGGGCTTGTACCTAACAGG	500				X
	R: AACGGGCTTCTCTCATGCTC
At5g18475_417F At5g18475_915R (PPR)	F: ACGGATGAAGCTATCGAGATGC	498				X
	R: AACCATGCACACAACATAATATGCT
At1g13040_1457F At1g13040_1958R (PPR)	F: TGTAGCTCACCATGTCAGGC	501				X
	R: AGTGGAGCTCAAATGCATAGGT
At1g16890_816F At1g16890_1315R (PPR)	F: AAACCCATCGATCGTCATTACTTTG	499				X
	R: TCCACCTCCATTTGATGAAACTTG
At1g20300_782F At1g20300_1286R (PPR)	F: GTGATCGACGCTTTGTGCC	504				X
	R: TCGGCTCCACCTCATTCTCA
At1g07230_362F At1g07230_870R (COSII)	F: GGACCTCATCAACGGGTTCC	508				X
	R: GTCAATCCACGGCGAAATCAA
At1g17760_1370F At1g17760_1867R (COSII)	F: CTTGCTACATTCCCAACTTCGG	497				X
	R: AACATCATTCACGTATTTGAAACCCA
At1g26640_1171F At1g26640_1670R (COSII)	F: TGCAAGAGCCCTGACAATTTC	499				X
	R: TGGGTCCTCAGAAATTCCATCTG
At1g30580_2218F At1g30580_2718R (COSII)	F: TGGAAGGCTGCTGAAGTTGA	500				X
	R: GGCATGGATCTTTGGGAGAAAC
At1g43860_2011F At1g43860_2515R (COSII)	F: AGCATACAATTTCAGGAGTGCAATC	504				X
	R: CTCGGCGACCATCACCAC
At1g59600_576F At1g59600_1075R (COSII)	F: GGAGACCTGAAACTAGGTCCC	499				X
	R: AAGGTGGTTTGCTGAGCCTC
At1g69220_607F At1g69220_1075R (COSII)	F: AGCCTCGTCCGTAACATTCA	467				X
	R: GGGCAGTGTATAAAGCTCGAGA
At2g03120_425F At2g03120_925R (COSII)	F: TGGGATCTGTTGTAGTTTATCCAGA	500				X
	R: ACGAAAGAGCAGCAATCCCT
At2g27170_2888F At2g27170_3366R (COSII)	F: CCCATCTTCATCCCTCCTGC	475				X
	R: TTGACAGTCAATGGCTCAGCC
At2g42490_4066F At2g42490_4564R (COSII)	F: ACTTAGCCTTTGGACCAGCC	475				X
	R: GCCTGGAGAAACGCACAATC
At3g03600_524F At3g03600_1023R (COSII)	F: AGGGCTCGAACTGTTTGTCA	499				X
	R: GCCCTTAAACCCTGCAACTC
At3g04870_957F At3g04870_1412R (COSII)	F: TCTTCCCATCTTCCCAAGGTC	455				X
	R: ACTCAAACACGGCTCAACTCA
At3g05000_1412F At3g05000_1912R (COSII)	F: ACCAACTAAACAGAGCCTCAATCA	500				X
	R: TTGCATCTTTACTTTCCCTGTGG
At3g09090_1427F At3g09090_1912R (COSII)	F: TTTCGTGTTTGCATAGTTTAGCAGT	485				X
	R: ATGGCCGCACAAATGATCCA
At3g13235_2063F At3g13235_2545R (COSII)	F: GGCAACATGTATCCGGCCTA	482				X
	R: TTGGGAAACTAAACCAACTCTGC
At3g16150_956F At3g16150_1360R (COSII)	F: ACGACAAGTTCAACGATATCAATGG	404				X
	R: CCTTATTGCTCGGAATGCTCG
At3g17205_926F At3g17205_1425R (COSII)	F: TCAAGTCGAAACCAGCATCTGA	499				X
	R: GGACTGCTGGTAATGCATCTG
At3g55360_121F At3g55360_620R (COSII)	F: CAGAACTCCGGAGACTTCTCAAC	499				X
	R: GCCCACCGTCCTTCATTACA
At3g62580_1318F At3g62580_1817R (COSII)	F: CATAATCGTGCTTGGGATTAACCG	499				X
	R: GCAATCTGGTTATCTTTACGCCC
At4g00560_2034F At4g00560_2528R (COSII)	F: AAGCTGCCATGTCTTTAAATGTCC	494				X
	R: GGAACAGTCTCATCCTCTTCCTTG
At4g12230_1167F At4g12230_1584R (COSII)	F: TGTAGATTGGATCAAAGCAATGGC	417				X
	R: TCAAATGCTTGAATCCACTCATGG
At4g23100_808F At4g23100_1307R (COSII)	F: ATCACCTCAAAGAGAAGGCGG	499				X
	R: CCTTCATATATCGTATTTCTACCACGG
At4g29490_2406F At4g29490_2905R (COSII)	F: CCAACCAAACTGCATAGTCAGC	499				X
	R: CCTAGCTCACCTCATGGCAG
At4g32280_1754F At4g32280_2254R (COSII)	F: CCGATGGCTAATCACCGCC	500				X
	R: ATACGTTGCACAGACTCGACA
At4g33625_852F At4g33625_1351R (COSII)	F: AGGTAGGCGTCATTGTAGAACC	499				X
	R: CATCTCCTGCTGCAAGAATTGG
At4g38630_507F At4g38630_1005R (COSII)	F: GGTACCTGTTTCTAGGACGTGC	498				X
	R: ATCCAAGTTGGGATCCACACC
At5g06830_1899F At5g06830_2398R (COSII)	F: GGCCCAACCCATTGCCATTA	499				X
	R: ACTAGACTCTGGACTTTGCAGG
At5g07270_1779F At5g07270_2278R (COSII)	F: CACATCCTTGCAACCTCAAGC	499				X
	R: ACTGCTCTTCATATGGCTGCT
At5g12200_438F At5g12200_937R (COSII)	F: TCGGTTCCTACTAGCTGCAAC	499				X
	R: ATCAGAGTTAGGGATGCTTGTGT
At5g13640_387F At5g13640_885R (COSII)	F: GGGAACTCTGATTACAGGGACTC	498				X
	R: CTAGCCTACGAGACCATGGC
At5g18580_933F At5g18580_1432R (COSII)	F: GCTGCTTTATTGCACGGTGT	499				X
	R: ACAGTTACCTGCGTGAAGAGAG
At5g20080_925F At5g20080_1424R (COSII)	F: ATACGAGAATAGCATTGGTACTCGA	499				X
	R: AGATGCCATCCTTAAGAACCCTG
At5g23540_1611F At5g23540_2102R (COSII)	F: AATCCTTCCATGAGTGAAACCCA	491				X
	R: ATTACTCTCAACATGGATGATAGTTGG
At5g58490_254F At5g58490_664R (COSII)	F: GCTATCATTCCCAGCCCTAATTG	410				X
	R: TTGCAAATCTGTAATTATAGGACCCAA

Primer sequence for the “Castilleja-specific primer.” To make the target-specific primer for subsequent microfluidic PCR, conserved sequence tags CS1 (5′-ACACTGACGACATGGTTCTACA) and CS2 (5′-TACGGTAGCAGAGACTTGGTCT) were added to each forward and reverse primer, respectively.

Amplicon length (bp) estimated from BLAST hit alignments.

PCR validations using DNAs from Bennett and Mathews (2006).

PCR validations were considered successful for Castilleja when amplification occurred for all three taxa, representing one annual lineage (C. lineariloba) and two perennial lineages (C. pumila and C. lemmonii).

Taxa that were both PCR validated and had primer combinations evaluated in silico against their respective assemblies (raw read files available in the NCBI Sequence Read Archive submission SRP100222).

Fig. 1.

Relationships among major clades within Orobanchaceae modified from McNeal et al. (2013) with taxa used for primer validation indicated (see text). Bootstrap support values for clades are indicated along the branches and follow McNeal et al. (2013). Reproduced from Latvis et al. (2017a).

To investigate whether any of these primer combinations would amplify successfully across Orobanchaceae, we searched for our selected Castilleja primers against an assembled low-coverage genome for Lamourouxia multifida Kunth using BLAST (Altschul et al., 1990). Lamourouxia multifida was sequenced on an Illumina HiSeq 2000 at the University of Oregon as 100-bp paired-end reads, and contigs were assembled using SPAdes (Bankevich et al., 2012) under the default settings. BLAST search parameters, assessment of suitable hits, and subsequent PCR validation with Lamourouxia virgata Kunth, Physocalyx major Mart., and Neobartsia filiformis (Wedd.) Uribe-Convers & Tank are described in Latvis et al. (2017a). Primer combinations with successful amplification in Lamourouxia Kunth and at least one other taxon were selected for further PCR testing with other major lineages in Orobanchaceae (sensu McNeal et al., 2013; Fig. 1). This second round of PCR validation follows Latvis et al. (2017a), except that two of the accessions used for testing were changed. As in Latvis et al. (2017a), we also included a negative control and conserved sequence–tagged “universal” primers for the trnL-F region as a positive control for all primer pairs. We used Neobartsia peruviana (Walp.) Uribe-Convers & Tank instead of N. filiformis, and Paulownia fortunei (Seem.) Hemsl. instead of P. elongata Siebold & Zucc. (Appendix 1). Of the 87 nuclear primer combinations specifically designed for Castilleja, we identified 27 with broader applicability in Orobanchaceae, chosen if they successfully amplified in Pedicularideae (Clade IV; including Castilleja, Lamourouxia, and Pedicularis L.) and at least one of the other major clades highlighted in Fig. 1. Validation results are presented in Table 1 with these “core Orobanchaceae” combinations boldfaced.

CONCLUSIONS

We present 87 nuclear primer pairs specifically designed for Castilleja that target COSII and PPR loci. Although we target the same putative single-copy nuclear domains as previous studies (Wu et al., 2006; Li et al., 2008; Yuan et al., 2010; Blischak et al., 2014; Uribe-Convers et al., 2016), we developed primers for different loci and present unique primer combinations in this study. As with our chloroplast primers (Latvis et al., 2017a), all combinations were designed with the Fluidigm microfluidic PCR system in mind, allowing for parallelization of amplification for downstream high-throughput sequencing platforms. Of these, we identify a set of 27 primer combinations with broader utility across Orobanchaceae. The development of primers for putatively single-copy nuclear loci will greatly enhance efforts to understand evolutionary history at multiple taxonomic scales, both for Castilleja and across Orobanchaceae.

Appendix 1.

Voucher information for species used in this study.

Species	Voucher accession no. (Herbarium)a	Collection locality	Geographic coordinates
Castilleja cusickii Greenm.	Tank 2009-01 (ID)	Idaho, USA	45.884241°N, 116.230195°W
Castilleja foliolosa Hook. & Arn.	A. Colwell 03-09 (YM)	California, USA	35.3926°N, 120.3522°W
Castilleja lemmonii A. Gray	Jacobs 2015-088 (ID)	California, USA	37.907982°N, 119.258583°W
Castilleja lineariloba (Benth.) T. I. Chuang & Heckard	Tank 2002-04 (WTU)	California, USA	37.41387°N, 120.10833°W
Castilleja pumila Wedd.	Uribe-Convers 2011-120 (ID)	La Libertad, Peru	7.99506°S, 78.44197°W
Castilleja tenuis (A. Heller) T. I. Chuang & Heckard	Tank 2001-13 (WTU)	Washington, USA	46.118133°N, 121.5158°W
Harveya purpurea Harv.	Randle 79 (OS)	NA	NA
Lamourouxia multifida Kunth	Mejia 695 (CAS)	Chiapas, Mexico	NA
Lamourouxia virgata Kunth	Mejia 581 (CAS)	Chiapas, Mexico	16.713611°N, 92.614722°W
Lindenbergia sp. Kunth	Armstrong 1163 (ISU)	NA	NA
Neobartsia peruviana (Walp.) Uribe-Convers & Tank	Uribe-Convers 13-011 (ID)	NA	NA
Orobanche californica Cham. & Schltdl.	Bennett 72 (A)	Cultivated	Cultivated
Paulownia fortunei (Seem.) Hemsl.	s.n. (A)	Cultivated	Cultivated (https://sheffields.com)
Pedicularis sp. L.	Krasek and Bennett s.n. (A)	Slovenia	NA
Physocalyx major Mart.	G. O. Romão 2528 (ESA)	Minas Gerais, Brazil	19.2635°S, 43.5508°W
Rhinanthus alectorolophus (Scop.) Pollich	Bennett 85 (A)	NA	NA
Schwalbea americana L.	Kirkman s.n. (PAC)	NA	NA

Note: NA = not available.

Herbarium acronyms are per Index Herbariorum (http://sweetgum.nybg.org/science/ih/).