There are two classes of mobile elements whose transposition involves the conversion of an mRNA copy of the element into DNA via reverse transcription; their mechanisms of translocation are, however, quite distinct. One class contains the LTR retrotransposons - the retroviruses and retrovirus-like elements - described in the preceding section. These are characterized by long terminal repeats (LTRs) which play a central role in their replication, and DNA copies of these elements insert into the target chromosomal DNA after synthesis. The other class is composed of the non-LTR retrotransposons. Non-LTR retrotransposons make a double-stranded DNA copy of themselves in situ at the target site, using a 3'OH in the target DNA generated by the element-encoded endonuclease to prime reverse transcription, and the mRNA copy of the element as a template. This insertion reaction is called target-primed reverse transcription. As a consequence of this mechanism, the polyA sequences at the end of the transposon mRNA are incorporated into the new insertion site, and thus these non-LTR elements are also sometimes known as polyA+ retrotransposons.

Because of their distribution between and among genes, non-LTR retrotransposons are commonly called LINEs (long interspersed nuclear elements) and SINEs (short interspersed nuclear elements): these are illustrated schematically in Figure 13-13.1. Intact LINEs are autonomous elements that encode their own transposition functions whereas SINEs are non-autonomous elements that are believed to be mobilized by LINE recombination proteins. The origins of SINEs are discussed further below. Together, LINEs and SINEs make up a remarkably large proportion of the DNA of many eukaryotic genomes.

The number of non-LTR elements in the human genome is particularly striking. There are about 850,000 LINE and 1,500,000 SINE elements in the human genome, that is about 21% and 13% of the genome respectively, or 34% altogether. While LINE and SINE elements are particularly prominent in mammalian genomes, they are also present in other organisms including fungi, C. elegans, insects, reptiles, birds and plants. Indeed, the sequences encoding the reverse transcriptases of non-LTR and LTR transposons are the predominant eukaryotic DNA sequences on Earth.

Despite their prevalence, it is thought that only about 50 of the existing human genomic complement of LINE elements are likely to be capable of transposition. Most LINE elements are inactivated because of the inefficiency of reverse transcription, which is error-prone, so that the ORFs encoding the transposition machinery are likely to be disabled by mutations, and is not highly processive, so that 5' truncation of the elements often occurs during transposition. Like many other transposable elements, LINEs are expressed most highly in the germline, and gene disruption by LINE and SINE insertion is a known cause of some cases of several human diseases. Duchenne muscular dystrophy and hemophilia are common human genetic diseases in which several cases have been caused by insertional mutagenesis by LINE or SINE elements, probably because the genes for the affected proteins (dystrophin and factor VIII) are extensive and present large targets. Although most mutations occur in the germline, a somatic LINE insertion into a gene that plays a key role in tumorigenesis has been observed in a sporadic colon cancer.

LINEs and SINEs also contribute to genome instability and human disease because of their repeated character and interspersed distribution. This can lead to homologous recombination between them, resulting in unequal crossing over, and gene deletions and duplications.

An intact LINE element (Figure 13-13.1) is about 6 kb in length and has two open reading frames, ORF1 and ORF2, encoding two proteins required for its translocation. ORF1 encodes an RNA binding protein and ORF2 encodes a multifunctional protein having endonuclease and reverse transcriptase activity. Translocation of a LINE element begins with transcription of the element from an internal pol II promoter and export of the resulting mRNA to the cytoplasm. LINE proteins are then synthesized from ORF1 and ORF2, and preferentially associate with the LINE mRNA, and the resulting ribonucleoprotein complex reenters the nucleus. The sequence of events in translocation is outlined in Figure 13-13.2. Recombination begins with nicking of the target DNA by the element-encoded endonuclease which preferentially cleaves T-rich sequences: in humans the enzyme resembles an apurinic endonuclease and can act at many sites. The polyA+ end of the mRNA then associates with the target site by base pairing with the T-rich strand exposed by the endonuclease cleavage. The target DNA 3'OH exposed by endonuclease cleavage then acts as a primer for the synthesis of a new line DNA strand by the reverse transcriptase using the line mRNA as a template. Thus a new line DNA strand is produced at the insertion site. Finally, a second DNA strand is synthesized on the template of the DNA copy, and the target DNA is filled in to generate flanking direct repeats. The reactions by which the 5' end of the newly synthesized LINE DNA strand is attached to the target DNA, the second LINE DNA is synthesized and the target site duplication is completed remain to be established.

SINEs are DNA segments that do not encode recombination proteins but are thought to be mobilized by LINE element recombination proteins. The sequences of these elements are distinct from those of LINEs. Most SINEs are short (100—300 bp in length; see Figure 13-13.1) and are derived from tRNA genes. However, a prominent class of human SINEs derives from the 7SL RNA, the RNA component of the signal recognition protein involved in the targeting of some proteins to membranes. These 7SL elements are called Alu elements because they include an Alu restriction site. SINEs of both origins are transcribed from internal pol II promoters. Processed pseudogenes are thought to arise by the LINE-protein mediated insertion of a processed mRNA. They are closely related to other genes but lack 5' promoter sequences and introns, and have 3' polyA tracts, features consistent with reverse transcription of an mRNA. Most pseudogenes are inactive. Target-primed reverse transcription can also mobilize sequences flanking the 3' ends of LINE elements. Such transduction occurs when the LINE element transcript extends into flanking DNA. Insertion of the DNA corresponding to the extended transcript at a new genomic location can result in new combinations of regulatory and protein coding segments. It has been estimated that such LINE-transduced sequences represent about 1% of the human genome, an amount similar to the fraction of the genome present in exon sequences.

Alu elements: repeated genomic sequences derived from 7SL RNA, the RNA component of the signal recognition particle that targets protein to membranes.

LINEs (long interspersed nuclear elements): genomic sequences derived from the duplication and transposition of a class of retrotransposons that move by target-primed reverse transcription.

non-LTR retrotransposons: transposons that move by target-primed reverse transcription and lack the LTRs characteristic of retroviruses and retroviral-like transposons.

polyA+ retrotransposons: another name for non-LTR retrotransposons.

SINEs (short interspersed nuclear elements): genomic sequences derived from tRNA genes or 7SL RNA and that spread non-autonomously in the genome by target-primed reverse transcription mediated by LINE-encoded recombination proteins.

target-primed reverse transcription: mode of duplication and transposition of non-LTR retrotransposons, which spread through reverse transcription of retrotransposon RNA primed by DNA at the target site.

target-site duplications (TSDs): duplicated sequences at the ends of transposons, generated by repair of single-stranded DNA ends created by cleavage of the DNA at the target site.