Nonautonomous Long Terminal Repeat Retrotransposons in Plants: Progress and Perspective

Retrotransposons are a class of mobile elements, which initiate their retrotransposition through a copy-and-paste mechanism via RNA intermediates. Retrotransposons can be divided into five orders on the basis of their structural features, long terminal repeat retrotransposons (LTR-RTs), DIRS-like elements, Penelope-like elements (PLEs), LINEs and SINEs [1]. Among them, LTR-RTs are the major genomic components in flowering plants, particularly in species with complex genomes. For example, ~20% of rice [2], ~42% of soybean [3], ~55% of sorghum [4], and >75% of the maize genomes [5] are composed of LTR-RTs. Recent studies indicate that in diploid species, genome size and TE content show a strong positive correlation [6].

 A typical LTR-RT element contains two identical LTRs, a primer-binding site (PBS), a polypurine tract (PPT), gag, and pol, two genes necessary for retrotranspositional process [7]. The LTR region can be further divided into three parts, including U3, R and U5 [7]. Because two LTRs of an element are identical at the time of insertion, the insertion time can be estimated based on the divergence time of the two LTRs and the evolutionary rate of LTR sequences [8]. For example, the majority of LTR-RTs in soybean were amplified in the last 1 million years (Mys) [9]. Usually LTR-RTs are subclassified into Copia and Gypsy superfamilies based on the order of IN and RT in pol [10]. Occasionally some elements were found to contain an additional ORF1 gene upstream of gag, and/or envelope (env)-like gene after the pol. Because both genes are not required for the retrotranspositional process, their origin and functional role remains mysterious. Recent genome-wide analysis and multispecific comparisons revealed that these elements were anciently evolved, and lineage-specific [1, 9]. Besides intact LTRRTs, a large number of Solo-LTRs and truncated elements have also been found in plant genomes [9, 11, 12]. These incomplete elements, together with numerous LTR remnants were presumed to be the products of unequal recombination and illegitimate recombination, two molecular mechanisms counterbalancing genome expansion [11, 12]. For instance, it was estimated that >190 Mb of DNA had been removed from the rice genome in the past 8 Mys, leaving the current rice genome ~400 Mb with ~97 Mb DNA of detectable LTR-RTs [12].

 Based on their structural completeness and retrotranspositional capability, LTR-RTs can also be classified into autonomous and nonautonomous types. An intact element is defined as autonomous if it encodes all the protein-coding domains necessary for catalyzing its retrotransposition [1]. By contrast, an element lacking one or more protein coding domains, but still keeping its retrotranspositional activity within a time frame, is generally defined as nonautonmous. Large retrotransposon derivatives (LARDs) and terminal-repeat retrotransposons in miniature (TRIM) are two groups of LTR-RTs belonging to nonautonomous types [13, 14]. Since both LARDs and TRIM have no open reading frames in the internal part, they were presumed to transpose by borrowing proteins from their autonomous partners. But for most cases, the relationships between autonomous and nonautonomous elements have not been established yet, and the exact mechanism(s) governing the activity of nonautonomous elements remains unclear. Although the transpositional mechanism of Ds1 in maize, and MITEs in rice have been discussed previously [15-19], these nonautonomous elements transpose via a "cut-and-paste" mechanism, and do not undergo reverse transcription process. Thus the transposition mechanism for these elements may be essentially different from those of LTR-RTs. 

We have previously identified 510 LTR-RT families in the sequenced soybean genome, and conducted further comprehensive analysis of the largest family SNARE [20, 21]. This family contains both autonomous and nonautonomous subfamilies. We found that nonautonomous elements frequently exchanged the LTR domains with their autonomous partners in different timeframes of soybean evolutionary history, thus providing the evidence that autonomous and nonautonomous LTR-RTs can interact and communicate with each other. Here we review the recent studies on plant nonautonomous elements with respect to the nature, timing, origin, and evolutionary process, thereby providing insights into the retrotranspositional process of LTR-RTs.