Identification of charged amino acids required for nuclear localization of human L1 ORF1 protein.

BACKGROUND: Long Interspersed Element 1 (LINE-1) is a retrotransposon that is present in 500,000 copies in the human genome. Along with Alu and SVA elements, these three retrotransposons account for more than a third of the human genome sequence. These mobile elements are able to copy themselves within the genome via an RNA intermediate, a process that can promote genome instability. LINE-1 encodes two proteins, ORF1p and ORF2p. Association of ORF1p, ORF2p and a full-length L1 mRNA in a ribonucleoprotein (RNP) particle, L1 RNP, is required for L1 retrotransposition. Previous studies have suggested that fusion of a tag to L1 proteins can interfere with L1 retrotransposition. RESULTS: Using antibodies detecting untagged human ORF1p, western blot analysis and manipulation of ORF1 sequence and length, we have identified a set of charged amino acids in the C-terminal region of ORF1p that are important in determining its subcellular localization. Mutation of 7 non-identical lysine residues is sufficient to make the resulting ORF1p to be predominantly cytoplasmic, demonstrating intrinsic redundancy of this requirement. These residues are also necessary for ORF1p to retain its association with KPNA2 nuclear pore protein. We demonstrate that this interaction is significantly reduced by RNase treatment. Using co-IP, we have also determined that human ORF1p associates with all members of the KPNA subfamily. CONCLUSIONS: The prediction of NLS sequences suggested that specific sequences within ORF1p could be responsible for its subcellular localization by interacting with nuclear binding proteins. We have found that multiple charged amino acids in the C-terminus of ORF1p are involved in ORF1 subcellular localization and interaction with KPNA2 nuclear pore protein. Our data demonstrate that different amino acids can be mutated to have the same phenotypic effect on ORF1p subcellular localization, demonstrating that the net number of charged residues or protein structure, rather than their specific location, is important for the ORF1p nuclear localization. We also identified that human ORF1p interacts with all members of the KPNA family of proteins and that multiple KPNA family genes are expressed in human cell lines.

The impact of multiple splice sites in human L1 elements.

LINE-1 elements represent a significant proportion of mammalian genomes. The impact of their activity on the structure and function of the host genomes has been recognized from the time of their discovery as an endogenous source of insertional mutagenesis. L1 elements contain numerous functional internal polyadenylation signals and splice sites that generate a variety of processed L1 transcripts. These sites are also reported to contribute to the generation of hybrid transcripts between L1 elements and host genes. Using northern blot analysis we demonstrate that L1 splicing, but not L1 polyadenylation, is delayed during the course of L1 expression. L1 splicing can also be negatively regulated by EBV SM protein known to alter this process. These results suggest a potential for L1 mRNA processing to be regulated in a tissue- and/or development-specific manner. The delay in L1 splicing may also serve to protect host genes from the excessive burden of L1 interference with their normal expression via aberrant splicing.

Human retroelements may introduce intragenic polyadenylation signals.

In the human genome, the insertion of LINE-1 and Alu elements can affect genes by sequence disruption, and by the introduction of elements that modulate the gene's expression. One of the modulating sequences retroelements may contribute is the canonical polyadenylation signal (pA), AATAAA. L1 elements include these within their own sequence and AATAAA sequences are commonly created in the A-rich tails of both SINEs and LINEs. Computational analysis of 34 genes randomly retrieved from the human genome draft sequence reveals an orientation bias, reflected as a lower number of L1s and Alus containing the pA in the same orientation as the gene. Experimental studies of Alu-based pA sequences when placed in pol II or pol III transcripts suggest that the signal is very weak, or often not used at all. Because the pA signal is highly affected by the surrounding sequence, it is likely that the Alu constructs evaluated did not provide the required recognition signals to the polyadenylation machinery. Although the effect of pA signals contributed by Alus is individually weak, the observed reduction of "sense" oriented pA-containing L1 and Alu elements within genes reflects that even a modest influence causes a change in evolutionary pressure, sufficient to create the biased distribution.

Alu insertion polymorphisms for the study of human genomic diversity.

Genomic database mining has been a very useful aid in the identification and retrieval of recently integrated Alu elements from the human genome. We analyzed Alu elements retrieved from the GenBank database and identified two new Alu subfamilies, Alu Yb9 and Alu Yc2, and further characterized Yc1 subfamily members. Some members of each of the three subfamilies have inserted in the human genome so recently that about a one-third of the analyzed elements are polymorphic for the presence/absence of the Alu repeat in diverse human populations. These newly identified Alu insertion polymorphisms will serve as identical-by-descent genetic markers for the study of human evolution and forensics. Three previously classified Alu Y elements linked with disease belong to the Yc1 subfamily, supporting the retroposition potential of this subfamily and demonstrating that the Alu Y subfamily currently has a very low amplification rate in the human genome.

Large-scale analysis of the Alu Ya5 and Yb8 subfamilies and their contribution to human genomic diversity.

We have utilized computational biology to screen GenBank for the presence of recently integrated Ya5 and Yb8 Alu family members. Our analysis identified 2640 Ya5 Alu family members and 1852 Yb8 Alu family members from the draft sequence of the human genome. We selected a set of 475 of these elements for detailed analyses. Analysis of the DNA sequences from the individual Alu elements revealed a low level of random mutations within both subfamilies consistent with the recent origin of these elements within the human genome. Polymerase chain reaction assays were used to determine the phylogenetic distribution and human genomic variation associated with each Alu repeat. Over 99 % of the Ya5 and Yb8 Alu family members were restricted to the human genome and absent from orthologous positions within the genomes of several non-human primates, confirming the recent origin of these Alu subfamilies in the human genome. Approximately 1 % of the analyzed Ya5 and Yb8 Alu family members had integrated into previously undefined repeated regions of the human genome. Analysis of mosaic Yb8 elements suggests gene conversion played an important role in generating sequence diversity among these elements. Of the 475 evaluated elements, a total of 106 of the Ya5 and Yb8 Alu family members were polymorphic for insertion presence/absence within the genomes of a diverse array of human populations. The newly identified Alu insertion polymorphisms will be useful tools for the study of human genomic diversity.

Cis-acting influences on Alu RNA levels.

The human short interspersed repeated element (SINE), Alu, amplifies through a poorly understood RNA-mediated mechanism, termed retroposition. There are over one million copies of Alu per haploid human genome. The copies show some internal variations in sequence and are very heterogeneous in chromosomal environment. However, very few Alu elements actively amplify. The amplification rate has decreased greatly in the last 40 million years. Factors influencing Alu transcription would directly affect an element's retroposition capability. Therefore, we evaluated several features that might influence expression from individual Alu elements. The influence of various internal sequence variations and 3' unique flanks on full-length Alu RNA steady-state levels was determined. Alu subfamily diagnostic mutations do not significantly alter the amount of Alu RNA observed. However, sequences containing random mutations throughout the right half of selected genomic Alu elements altered Alu RNA steady-state levels in cultured cells. In addition, sequence variations at the 3' unique end of the transcript also significantly altered the Alu RNA levels. In general, sequence mutations and 3' end sequences contribute to Alu RNA levels, suggesting that the master Alu element(s) have a multitude of individual differences that collectively gives them a selective advantage over other Alu elements.