The Alu family of mobile retroposons provide an interesting alternative model system with which to study the mechanisms involved in the synthesis and posttranscriptional handling of pol III transcripts. B1-Alu elements have been evolving within their mammalian hosts and may have coopted cellular factors to facilitate their proliferation through pathways that are unique. Human Alu sequences are short interspersed elements (SINEs) that are found at about one milion sites in our genomes. In rodents, the homologous elements are called B1 or Alu-equivalent sequences -for convenience, we can refer to these SINEs collectively as B1-Alu. Population studies by Deininger and colleagues suggest that a new Alu insertion establishes itself in the genome of about one in one hundred live births. Transposition into new sites does not always cause genetic disruptions since these would be tolerated in recessive alleles. However, new insertions do cause problems (genetic disorders such as neurofibromatosis 1 and hemophelia) when they jump into dominant or X-linked genes. Alu elements have been classified as retroposons because they are mobilized to new sites through RNA intermediates, somewhat similar to retroviruses. Yet, the mechanisms used to mobilize to new sites in the genome are largely unknown. B1-Alu retroposons are distinguishable from retrotransposons, an example of which are the long interspersed elements (LINEs) that encode polypeptides such as reverse transcriptase, that assist their retrotransposition. It is thought that B1-Alu transcripts are reverse transcribed and transposed through the actions of the retrotransposon-encoded factors. Since Alu elements outnumber retrotransposons in the human genome, they must be excellent templates for retroposition and would appear to pirate LINE machinery.
The internal pol III promoter is found in high copy numbers in mammalian genomes because it has been mobilized by Alu transposition. This feature, the internal promoter, provides newly inserted B1-Alu elements with an inherrent potential for expression. However, the sequences flanking the newly inserted element can influence that potential and mechanisms exist to keep B1-Alu sequences silenced most of the time, although viral infection and other forms of cellular stress activate their transcription. While the internal promoter within a newly inserted element can direct accurate initiation, there is no termination signal in the B1-Alu sequence and the elongating polymerase must therefore rely on the first fortuitous tract of consecutive dTs downstream of the insertion site for termination. The sequence adjacent to a dT tract can have significant effects on the expression of a B1-Alu element (Maraia et al., 1992, Goodier and Maraia, 1998). Likewise, upstream sequences can influence promoter function. Thus, the site into which an element inserts influences its subsequent activity.
Nuclear versus cytoplasmic partitioning. In cultured cells and animal tissues, B1 and Alu sequences accumulate as stable transcripts known as small cytoplasmic (sc)B1 and scAlu RNAs. These RNAs are associated with a heterodimeric protein known as SRP9/14 which recognizes the structured 5' end region of these RNAs. SRP9/14 also associates with another RNA known as signal recognition particle (SRP) RNA (also known as 7SL RNA) which shares substantial sequence and structural homology with scB1/scAlu RNAs. SRP consists of one RNA of 300 nucleotides and six polypeptides. SRP directs polypeptides that are being synthesized on ribosomes that are destined for secretion, to the endoplasmic reticulum. The function of scAlu and scB1 RNAs are presently unknown.
La is predominantly a nuclear protein while SRP9/14 is predominantly cytoplasmic. By homology to SRP RNA, the structured region at the 5' end of the B1-Alu transcript would direct nuclear export (He et al., 1994). Thus, the 3' end of the nascent B1-Alu RNA would be bound by La and therefore tethered in the nucleus while the 5' structured region would want to direct cytoplasmic accumulation. 3' processing would liberate the Alu domain which would then be exported and stabilized by SRP9/14 in the cytoplasm.
3' processing diverts nascent B1-Alu transcripts away from a nuclear retroposition pathway.* B1-Alu SINEs that jump into new loci do so with a 3' poly(A) tract. Moreover, studies of another type of poly(A)-type retroelement suggest that the 3' poly(A) tract directs the retroinsertion process. In that case, the poly(A) tract at the end of the RNA is used as the template on which the new cDNA is initiated by reverse transcriptase which uses the 3'OH-end of a nick in the target DNA as the primer (Luan et al., 1993 ). Thus, for this type of element, reverse transcription is catalyzed at the site of insertion and the poly(A) tail is an important determinant of the process. However, scB1/Alu RNAs are produced as a result of processing that removes their poly(A) tracts and downstream sequences. Moreover, the absence of genomic B1-Alu sequences that lack poly(A) (or the A-rich tails that have accumulated mutations in older elements) is further evidence that these 3'-processed RNAs are not competent for retroinsertion. Thus, it would appear that 3'-processing of nascent B1-Alu transcripts diverts them away from the retroposition pathway, to accumulate instead in the cytoplasm. However, La does more than simply retain RNAs in the nucleus; it also stabilizes these 3'-processing. In the case of a nascent B1-Alu transcript, La maintains it in a state that would appear to be competent for retroposition.