Limits...
Modular organization and reticulate evolution of the ORF1 of Jockey superfamily transposable elements.

Metcalfe CJ, Casane D - Mob DNA (2014)

Bottom Line: ORF1 type variations involving the PHD domain were found in many subgroups of the L2 and CR1 lineages.A Jockey lineage-like ORF1 with a PHD domain was found in both lineages.In conclusion, while the structure of the ORF2 appears to be highly constrained and its evolution tree-like, the structure of the ORF1 within the CR1 and L2 lineages is much more variable and its evolution reticulate.

View Article: PubMed Central - HTML - PubMed

Affiliation: Universidade de São Paulo, Instituto de Biociências, Rua do Matão 277, Cidade Universitária, São Paulo 05508-090 SP, Brazil.

ABSTRACT

Background: Long interspersed nuclear elements (LINES) are the most common transposable element (TE) in almost all metazoan genomes examined. In most LINE superfamilies there are two open reading frames (ORFs), and both are required for transposition. The ORF2 is well characterized, while the structure and function of the ORF1 is less well understood. ORF1s have been classified into five types based on structural organization and the domains identified. Here we perform a large scale analysis of ORF1 domains of 448 elements from the Jockey superfamily using multiple alignments and Hidden Markov Model (HMM)-HMM comparisons.

Results: Three major lineages, Chicken repeat 1 (CR1), LINE2 (L2) and Jockey, were identified. All Jockey lineage elements have the same type of ORF1. In contrast, in the L2 and CR1 lineage elements, all five ORF1 types are found, with no one type of ORF1 predominating. A plant homeodomain (PHD) is much more prevalent than previously suspected. ORF1 type variations involving the PHD domain were found in many subgroups of the L2 and CR1 lineages. A Jockey lineage-like ORF1 with a PHD domain was found in both lineages. A phylogenetic analysis of this ORF1 suggests that it has been horizontally transferred. Likewise, an esterase containing ORF1 type was only found in two exclusively vertebrate L2 and CR1 groups, indicating that it may have been acquired in a vertebrate common ancestor and then transferred between the lineages.

Conclusions: The ORF1 of the CR1 and L2 lineages is very structurally diverse. The presence of a PHD domain in many ORF1s of the L2 and CR1 lineages is suggestive of domain shuffling. There is also evidence of possible horizontal transfer of entire ORF1s between lineages. In conclusion, while the structure of the ORF2 appears to be highly constrained and its evolution tree-like, the structure of the ORF1 within the CR1 and L2 lineages is much more variable and its evolution reticulate.

No MeSH data available.


Related in: MedlinePlus

Neighbor-joining phylogeny based on ORF1 Type I domains. The ORF1 of CR1 group 3 sequences cluster with those of L2 group 2, suggesting that this type of ORF1 may have been horizontally acquired across lineages. The phylogeny was estimated using MEGA 6 [20] and inferred using the JTT substitution matrix. The robustness of the nodes was estimated by 1,000 bootstrap replicates. Only bootstrap values for major groups are shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4120745&req=5

Figure 8: Neighbor-joining phylogeny based on ORF1 Type I domains. The ORF1 of CR1 group 3 sequences cluster with those of L2 group 2, suggesting that this type of ORF1 may have been horizontally acquired across lineages. The phylogeny was estimated using MEGA 6 [20] and inferred using the JTT substitution matrix. The robustness of the nodes was estimated by 1,000 bootstrap replicates. Only bootstrap values for major groups are shown.

Mentions: The PHD and CCHC domains are quite small, 50 or so amino acids long [see Additional file 2]. An attempt was made to determine the relationship between PHD and CCHC domains from different subgroups by multiple alignments using muscle [27] and a phylogenetic analysis using MEGA 6 [20]. This resulted in trees for both the PHD and the CCHC domains with unresolved branches (data not shown). A Clans analysis [28], which is multiple alignment independent, using all PHD and CCHC domains, resulted in two large clusters, one for the CCHC domains and one for the PHD domains. No clear structuring was found within these two clusters.Six clusters were inferred in a Clans clustering of the individual RRM domains (Figure 7). Cluster 1 consisted of ORF1 type II RRM domains from CR1 subgroup 5 and L2 subgroups 3, 5 and 6. All other ORF1 type II RRM domain sequences did not cluster strongly with other subgroups. The ORF1 type III RRM domain was found in a single subgroup, CR1 subgroup 4. These sequences all fell into a single cluster, Cluster 5. The ORF1 type I RRM domains fell into 4 clusters. The upstream (‘U’) and downstream (‘D’) RRM domains from the Jockey subgroup 1 formed two separate clusters (Clusters 2 and 6). The RRM domains from L2 subgroup 2 and CR1 subgroup 3 fell into two clusters, the upstream RRM domains together in a single cluster (Cluster 4), and the downstream RRM domains together in a separate cluster (Cluster 3).The relationship between the type I ORF1 domains is shown in Figure 8. The ORF1 domains from the Jockey subgroup 1 and from L2 subgroup 8 fell into two well-supported groups. The ORF1 domains from CR1 subgroup 3 and L2 subgroup 2, however, clustered together, with the CR1 subgroup 3 sequences being embedded at two positions within the L2 subgroup 2-sequence phylogeny.


Modular organization and reticulate evolution of the ORF1 of Jockey superfamily transposable elements.

Metcalfe CJ, Casane D - Mob DNA (2014)

Neighbor-joining phylogeny based on ORF1 Type I domains. The ORF1 of CR1 group 3 sequences cluster with those of L2 group 2, suggesting that this type of ORF1 may have been horizontally acquired across lineages. The phylogeny was estimated using MEGA 6 [20] and inferred using the JTT substitution matrix. The robustness of the nodes was estimated by 1,000 bootstrap replicates. Only bootstrap values for major groups are shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4120745&req=5

Figure 8: Neighbor-joining phylogeny based on ORF1 Type I domains. The ORF1 of CR1 group 3 sequences cluster with those of L2 group 2, suggesting that this type of ORF1 may have been horizontally acquired across lineages. The phylogeny was estimated using MEGA 6 [20] and inferred using the JTT substitution matrix. The robustness of the nodes was estimated by 1,000 bootstrap replicates. Only bootstrap values for major groups are shown.
Mentions: The PHD and CCHC domains are quite small, 50 or so amino acids long [see Additional file 2]. An attempt was made to determine the relationship between PHD and CCHC domains from different subgroups by multiple alignments using muscle [27] and a phylogenetic analysis using MEGA 6 [20]. This resulted in trees for both the PHD and the CCHC domains with unresolved branches (data not shown). A Clans analysis [28], which is multiple alignment independent, using all PHD and CCHC domains, resulted in two large clusters, one for the CCHC domains and one for the PHD domains. No clear structuring was found within these two clusters.Six clusters were inferred in a Clans clustering of the individual RRM domains (Figure 7). Cluster 1 consisted of ORF1 type II RRM domains from CR1 subgroup 5 and L2 subgroups 3, 5 and 6. All other ORF1 type II RRM domain sequences did not cluster strongly with other subgroups. The ORF1 type III RRM domain was found in a single subgroup, CR1 subgroup 4. These sequences all fell into a single cluster, Cluster 5. The ORF1 type I RRM domains fell into 4 clusters. The upstream (‘U’) and downstream (‘D’) RRM domains from the Jockey subgroup 1 formed two separate clusters (Clusters 2 and 6). The RRM domains from L2 subgroup 2 and CR1 subgroup 3 fell into two clusters, the upstream RRM domains together in a single cluster (Cluster 4), and the downstream RRM domains together in a separate cluster (Cluster 3).The relationship between the type I ORF1 domains is shown in Figure 8. The ORF1 domains from the Jockey subgroup 1 and from L2 subgroup 8 fell into two well-supported groups. The ORF1 domains from CR1 subgroup 3 and L2 subgroup 2, however, clustered together, with the CR1 subgroup 3 sequences being embedded at two positions within the L2 subgroup 2-sequence phylogeny.

Bottom Line: ORF1 type variations involving the PHD domain were found in many subgroups of the L2 and CR1 lineages.A Jockey lineage-like ORF1 with a PHD domain was found in both lineages.In conclusion, while the structure of the ORF2 appears to be highly constrained and its evolution tree-like, the structure of the ORF1 within the CR1 and L2 lineages is much more variable and its evolution reticulate.

View Article: PubMed Central - HTML - PubMed

Affiliation: Universidade de São Paulo, Instituto de Biociências, Rua do Matão 277, Cidade Universitária, São Paulo 05508-090 SP, Brazil.

ABSTRACT

Background: Long interspersed nuclear elements (LINES) are the most common transposable element (TE) in almost all metazoan genomes examined. In most LINE superfamilies there are two open reading frames (ORFs), and both are required for transposition. The ORF2 is well characterized, while the structure and function of the ORF1 is less well understood. ORF1s have been classified into five types based on structural organization and the domains identified. Here we perform a large scale analysis of ORF1 domains of 448 elements from the Jockey superfamily using multiple alignments and Hidden Markov Model (HMM)-HMM comparisons.

Results: Three major lineages, Chicken repeat 1 (CR1), LINE2 (L2) and Jockey, were identified. All Jockey lineage elements have the same type of ORF1. In contrast, in the L2 and CR1 lineage elements, all five ORF1 types are found, with no one type of ORF1 predominating. A plant homeodomain (PHD) is much more prevalent than previously suspected. ORF1 type variations involving the PHD domain were found in many subgroups of the L2 and CR1 lineages. A Jockey lineage-like ORF1 with a PHD domain was found in both lineages. A phylogenetic analysis of this ORF1 suggests that it has been horizontally transferred. Likewise, an esterase containing ORF1 type was only found in two exclusively vertebrate L2 and CR1 groups, indicating that it may have been acquired in a vertebrate common ancestor and then transferred between the lineages.

Conclusions: The ORF1 of the CR1 and L2 lineages is very structurally diverse. The presence of a PHD domain in many ORF1s of the L2 and CR1 lineages is suggestive of domain shuffling. There is also evidence of possible horizontal transfer of entire ORF1s between lineages. In conclusion, while the structure of the ORF2 appears to be highly constrained and its evolution tree-like, the structure of the ORF1 within the CR1 and L2 lineages is much more variable and its evolution reticulate.

No MeSH data available.


Related in: MedlinePlus