Exploring the loblolly pine (Pinus taeda L.) genome by BAC sequencing and Cot analysis

Abstract

Loblolly pine (LP; Pinus taeda L.) is an economically and ecologically important tree in the southeastern U.S. To advance understanding of the loblolly pine (LP; Pinus taeda L.) genome, we sequenced and analyzed 100 BAC clones and performed a Cot analysis. The Cot analysis indicates that the genome is composed of 57, 24, and 10% highly-repetitive, moderately-repetitive, and single/low-copy sequences, respectively (the remaining 9% of the genome is a combination of fold back and damaged DNA). Although single/low-copy DNA only accounts for 10% of the LP genome, the amount of single/low-copy DNA in LP is still 14 times the size of the Arabidopsis genome. Since gene numbers in LP are similar to those in Arabidopsis, much of the single/low-copy DNA of LP would appear to be composed of DNA that is both gene- and repeat-poor. Macroarrays prepared from a LP bacterial artificial chromosome (BAC) library were hybridized with probes designed from cell wall synthesis/wood development cDNAs, and 50 of the "targeted" clones were selected for further analysis. An additional 25 clones were selected because they contained few repeats, while 25 more clones were selected at random. The 100 BAC clones were Sanger sequenced and assembled. Of the targeted BACs, 80% contained all or part of the cDNA used to target them. One targeted BAC was found to contain fungal DNA and was eliminated from further analysis. Combinations of similarity-based and ab initio gene prediction approaches were utilized to identify and characterize potential coding regions in the 99 BACs containing LP DNA. From this analysis, we identified 154 gene models (GMs) representing both putative protein-coding genes and likely pseudogenes. Ten of the GMs (all of which were specifically targeted) had enough support to be classified as intact genes. Interestingly, the 154 GMs had statistically indistinguishable (α = 0.05) distributions in the targeted and random BAC clones (15.18 and 12.61 GM/Mb, respectively), whereas the low-repeat BACs contained significantly fewer GMs (7.08 GM/Mb). However, when GM length was considered, the targeted BACs had a significantly greater percentage of their length in GMs (3.26%) when compared to random (1.63%) and low-repeat (0.62%) BACs. The results of our study provide insight into LP evolution and inform ongoing efforts to produce a reference genome sequence for LP, while characterization of genes involved in cell wall production highlights carbon metabolism pathways that can be leveraged for increasing wood production.

Comparative study of pine reference genomes reveals transposable element interconnected gene networks

“…This could be explained by the lower GC content of LTRs of some RLXs distributed in genes; in addition, the database of conifer-interspersed repeats used for searches contained more sequences speci c to the P. taeda genome. The mean GC content of the gene transcripts was 44% for P. lambertiana and for P. taeda, which was higher than any average estimate for introns and published estimates for whole BAC clones and whole genome sequences of gymnosperms (38%, [90,91]).…”

Comparative Study of Pine Reference Genomes Reveals Transposable Element Interconnected Gene Networks

“…This could be explained by the lower GC content of LTRs of some RLXs distributed in the gene introns. The mean GC content of the gene transcripts (involved in the Daugava RLX network) was 44% for P. lambertiana and P. taeda , which was higher than any average estimate for introns and published estimates for whole BAC clones and whole genome sequences of gymnosperms (38%, [ 91 , 92 ]).…”

“…Additionally, the TE insertion patterns in the investigated pine introns were found to have a lower average GC content (39%) than nearby transcripts. The GC content of gene transcripts in the studied genes in P. taeda and P. lambertiana was comparable (44%) and higher than the reported genome average of 38% [ 91 , 92 ]. Similarly, a lower GC content in plant gene introns has been reported for other plant species [ 130 , 131 ], indicating that intron sequences may have a more relaxed DNA conformation and are more accessible to transcription and other regulatory factors [ 132 , 133 ].…”

Comparative study of pine reference genomes reveals transposable element interconnected gene networks

“…Additionally, the TE insertion patterns in investigated pine introns were found to have lower average GC content (39%) than nearby transcripts. The GC content of gene transcripts in the studied genes in P. taeda and P. lambertiana were comparable (44%) and higher than the reported genome average of 38% (Gonzalez-Ibeas et al, 2016;Perera et al, 2018). A lower GC content in plant gene introns has been reported for other plant species (Mizuno and Kanehisa, 1994;Singh et al, 2016), indicating that intron sequences may have a more relaxed DNA conformation and are more accessible to transcription and other regulatory factors (Schwartz et al, 2009;Gelfman and Ast, 2013).…”

“…This could be explained by the lower GC content of LTRs of some RLXs distributed in genes; in addition, the database of conifer-interspersed repeats used for searches contained more sequences speci c to the P. taeda genome. The mean GC content of the gene transcripts was 44% for P. lambertiana and for P. taeda, which was higher than any average estimate for introns and published estimates for whole BAC clones and whole genome sequences of gymnosperms (38%, (Gonzalez-Ibeas et al, 2016;Perera et al, 2018)).…”