![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
|
Vol. 8, Issue 5, 416-418, May 1998
INSIGHT/OUTLOOK
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ARTICLE |
|---|
Top
Article
References
|
|---|
The completion of the Saccharomyces genome sequence in 1996 signaled the beginning of a new chapter in repetitive element sequence
analysis (Cherry et al. 1997
). The genomic analysis
in the article by Kim et al. (1998) in this issue provides new insights into the nature and distribution of repetitive DNA sequences in the
Saccharomyces genome. It not only assesses the repetitive sequences present in the genome but also clarifies what is not found.
Saccharomyces has five long terminal repeat
(LTR)-retrotransposon families (Ty elements) but no nuclear LINE-like
or SINE retroelements and no identifiable DNA-based transposable
elements. Remarkably, each Ty family displays insertion specificity.
The parallels and differences between the organization of transposable
elements in the Saccharomyces genome, and what is thus far
known of the organization of elements in the genomes of other
eukaryotes, suggest that this chapter will have many interesting sequelae.
The genomic analysis by Kim et al. (1998) further defines the
relationships of the Ty families of yeast comprised of four copia-like
(Ty1, Ty2, Ty4, and Ty5) and one gypsy-like (Ty3) families. The
copia-like families have been divided further based on similarity of
encoded proteins and LTRs. Ty1 and Ty2 share significant similarity, particularly in the capsid domain, and share a common LTR, called 
. In addition, Kim et al. showed for the first time, that Ty1 and
Ty2 elements are distinct
differing consistently at a single position. Unlike retrovirus LTRs, yeast element LTRs flanking the
internal domain undergo recombination, thereby deleting one copy of the
LTR and the internal domain sequence. Interestingly, among the
families, there are widely differing copy numbers of complete elements
and LTRs: 32 Ty1 elements and 217 Ty1-type elements; 13 Ty2
elements and 34 Ty2-type elements; two Ty3 elements and 41 
elements; 3 Ty4 elements and 32 
elements; and one Ty5 element and
seven 
elements. Thus, ratios of LTR sequences to complete
elements range from 7 for Ty1 to 21 for Ty3. Because there are many
more Ty1 insertions than Ty3 insertions and because the Ty1-type elements are more degenerate than the elements, Ty1 probably
represents an older class of elements within yeast. The fact that the
ratio of LTR to complete copies is so much higher for Ty3 suggests that
generation of isolated LTRs occurs more frequently for some sequences
than for others.
With stunning completeness, the genome sequence reveals the nonrandom
nature of the retroelement insertions for each family (see Fig. 1)
(Chalker and Sandmeyer 1992; Devine and Boeke 1996; Zou et al. 1996;
Bryk et al. 1997; Smith and Boeke 1997). In addition to tRNA genes, 5S, U6, and PI RNA genes were also found associated with
insertions. The Tyl, Ty2, Ty3, and Ty4 elements target the upstream
region of RNA polymerase 111-transcribed genes: 196 of 217 Ty1; 28 of
34 Ty2; 40 of 41 Ty3; and 30 of 32 Ty4 insertions are within 750 bp of
a gene transcribed by RNA polymerase III. This listing is conservative
because it does not exclude the possibility that some of the seemingly
exceptional insertions are associated with unidentified targets, such
as unknown or degenerate polymerase III-transcribed genes. Ty5 does not
target tRNA genes. Instead, Ty5 inserts into silenced regions of the
genome, the telomeres and mating type loci.
|
In addition to these important insights, the analysis of Kim et al.
(1998) raises provocative questions concerning the relationship between
repetitive elements and their habitats: Why is Saccharomyces apparently lacking major classes of elements, including quite ancient
ones found in bacteria; what is the impact of these elements on genome
maintenance; and why do these elements target particular genomic
regions? These questions are addressed in the remainder of this article.
Why Does the Nuclear Yeast Genome Appear to Lack DNA Elements and SINE and LINE Classes of RNA elements?
DNA transposons (and DNA viruses), which are found in prokaryotic and eukaryotic organisms, including plants, have not been identified in the yeast genome sequence. One possible explanation is that DNA transposons do not amplify and integrate themselves as efficiently as retrovirus-like RNA elements. Cut-and-paste DNA transposons amplify by hopping into sister chromosomes that have not lost the donor copy. Replicative elements, other than viruses, transpose one new copy without excising the donor element. RNA elements are more similar to viruses in that they generate multiple genomic copies, although the first step is transcription, rather than replication. Thus, some organisms might escape or ultimately erase DNA elements while being colonized by their more prolific cousins, the RNA elements. Integration may provide a second point of discrimination against DNA elements. A fundamental distinction between prokaryotes, where DNA transposons abound, and eukaryotes is, of course, the nuclear membrane. Although the nuclear membrane disintegrates during mitosis in many species, it does not disintigrate in budding yeast. Therefore, DNA transposase would require mechanisms both for nuclear entry and for finding the DNA target. These requirements are also potentially more restrictive for DNA elements than for RNA elements that potentially associate integrase and substrate cotranslationally.
Transposition of DNA elements is initiated by nicking at the ends of
the genomic donor DNA, whereas the first genomic incision in
retrovirus-like transposition is a concerted strand transfer of donor
into target DNA. Nicked genomic DNA is likely unstable and therefore
subject to the reverse reaction or DNA repair. Properties of modern DNA
transposons reflect the labile nature of the transposition intermediate. For example, the bacterial element, Tn7, does
not initiate transposition with donor nicking until the target site is
identified (Bainton et al. 1993). Furthermore, many transposons in
prokaryotes and eukaryotes transpose more efficiently into sites
proximal to the donor site. Because the initiation steps for DNA
transposition and retroelement transposition differ, they represent
another point where yeast might discriminate against the proliferation
of DNA elements.
Although elements with reverse transcriptases related to LINE reverse
transcriptase exist in other simple eukaryotic organisms and sequences
encoding reverse transcriptases related to LINE reverse transcriptase
are found in yeast mitochondria, SINE- and LINE-type retrotransposons
appear to be absent from the yeast nuclear genome. Perhaps, similar to
DNA elements, primitive versions of these elements did not compete
effectively with DNA repair. According to the Eickbush model for
transposition of R2Bm, a site-specific LINE-like element in Bombyx
mori (Luan et al. 1993), transposition is initiated by nicking of
the target DNA by the element-encoded endonuclease followed by cDNA
synthesis in situ. In contrast, retrovirus-like elements copy RNA into
DNA in discrete particles and integrate that DNA in a concerted strand
transfer. It is generally thought that LINE element reverse
transcriptases provide the enzymatic means through which SINE elements
are mobilized. Thus, the general absence from the yeast genome of SINEs
and pseudogenes is consistent with the notion that their derivation is
linked to LINEs.
How Does the Presence of Multiple Elements Impact Maintenance of the Genome and Expansion of Element Populations?
Related sequences in nonhomologous contexts undergo ectopic
recombination in yeast, but meiotic recombination of Ty1 elements is
less than predicted based on studies of other sequences (Kupiec and
Petes 1988). Consistent with this, Kim et al. (1998) found no evidence
of chimeric Ty1 and Ty2 elements in the yeast genome. Nevertheless,
particular Ty1 elements were determined to be recombinant by inference
from sequences at their termini. Thus, although it may not be an efficient
process, retroelements do medate gross rearrangements of the yeast genome.
This study also confirmed that most Ty1 and Ty2 elements are likely
competent for transposition. This is important and interesting because of a long-standing paradox in the Ty1 field (Curcio et al.
1988). A significant percentage of cellular poly(A)+ RNA is Ty1 RNA,
yet transposition is normally infrequent. Furthermore, induction of
a single active element increases the frequency of transposition by
an amount disproportionate to the increase in RNA. These observations
could be accounted for if there were a cis bias by
retroelement proteins for genomic RNA or if most elements were
inactive, leading to synthesis of interfering proteins. Previous investigations have shown that transposition of a marked element can
be readily complemented in trans. This study, although not conclusive, makes the dominant-negative model less appealing, as only 7 of 45 Ty1 and Ty2 elements were found to contain inactivating mutations. Moreover, Kim et al. (1998) identified a discrete subclass of three Ty1 elements, the members of which differ from the main class
by a majority of silent mutations, indicating that they are under
selection. How did this come about in the presence of abundant Ty1 RNA
and in the absence of a cis-acting bias? It could be explained
if these elements have undergone mutations resulting in proteins and
RNA with specificity for one another (i.e., trans-acting bias). This could be tested. If it is not observed, perhaps it will be
necessary to finally resolve the fundamental mechanism by which
retrovirus-like elements and retroviruses partition genomic from
translated RNA copies.
Is the Propagation of Retroelements in the Yeast Genome a Case of Effective Parasitism or Mutualism?
The case for the latter is attractive, although not proven. Tests
of the potential effects of increasing Ty1 element copy number in yeast
have shown disadvantages of particular insertions but have not shown
that increases in copy numbereven large onesare particularly
burdensome (Boeke et al. 1991; Wilke and Adams 1992). In eukaryotic
cells, transposable element clustering is widespread (Moyzis et al.
1989; Pimpinelli et al. 1995; Andersson et al. 1998) and helps to
account for host tolerance of transposable elements. However, the yeast
elements put a new twist on this phenomenon by clustering, relative to
specific genomic features, RNA polymerase III-transcribed genes and
silenced regions. This pattern of facilitated clustering could be
particularly important given the compact nature of the yeast genome.
Targeting of Ty3 and Ty5, in particular, is likely to be mediated by
transcription factors and silencing proteins, respectively. Insertion
just upstream of tRNA genes does not disrupt gene expression, and
telomeric and mating-type regions targeted by Ty5 are transcriptionally silent. Thus, a single mechanism may guarantee that insertion is both
efficient and transparent to host function.
Eukaryotic genomes may have extracted benefits from retroelement
targeting as well. Regions dense in transposable element insertions are
distributed throughout the chromosomes of eukaryotic organisms. By
providing a common sequence at semiregular intervals, these elements
could be coopted for a structural role in chromosome organization or
movement. In chromosomes of human and Drosophila origin,
clusters of transposable elements are within heterochromatic regions.
Whether these regions contribute to chromosome function is not yet
known. In the case of Drosophila, HeT and TART elements have
been demonstrated to provide telomeric structure (Biessman and Mason
1997). Although clustering and targeting is observed for endogenous
transposable elements, external agents may not be constrained. Thus
far, although retrovirus insertion hot spots are observed in the
genome, strong regional preferences, such as those described for the
yeast elements, are not (Withers-Ward et al. 1994; Carteau et al. 1998).
In yeast, each Ty family displays de facto clustering as a consequence
of targeting. The telomeres and genes transcribed by polymerase III are
distributed throughout the chromosomes, as are the clusters of repeated
elements in eukaryotic cells, in a pattern that would be compatible
with structural functions. Furthermore, tRNA genes, rDNA (containing 5S
genes), and telomeres, the targets of Ty insertions, are all known to
recombine ectopically and also to repress RNA polymerase II
transcription. A zone of Ty elements could provide a barrier protecting
the integrity and expression of ORFs adjoining these regions. In yeast
there is a telomerase. Nevertheless, in addition to providing a buffer for recombination events, yeast transposable elements could provide redundancy in telomerase function or in repair of double-stranded breaks in the DNA (Kupiec and Petes 1988; Moore and Haber 1996).
Conclusions
Analyses such as the one reported here by Kim et al. (1998) are
invaluable. Without careful assessment of the numbers, types, and
patterns of repetitive elements, we would surely miss much of the
richness of the information made available from the genome project. It
has been clear for some time that repetitive elements are found in
heterochromatic clusters in many species and that this might be the
case in all organisms. However, neither the basis of the clustering nor
the basis of the heterochromatic configuration of this DNA nor the
implication for genomic structure is understood. These questions are
not merely academic. Expression of newly inserted sequences in
eukaryotic cells, whether introduced by transformation into plants or
by retrovirus vectors into eukaryotic, cells is frequently associated
with poor expression. As more complete genome sequence information
becomes available, revelation of transposable element patterns should
prompt renewed consideration of their interplay with their host genome.
This will surely lead to further insights into the workings of
chromosomes. In addition, if modification of genomes is sought, we must
look carefully at the mechanisms and consequences of nature's ongoing
modifications. Perhaps it should not be such a rude surprise that many
of our would-be modifications, similar to those in nature, fall silent
with time.
| |
FOOTNOTES |
|---|
1 E-MAIL sbandme{at}uci.edu; FAX (949) 824-2688.
| |
REFERENCES |
|---|
|
Top
Article
References
|
|---|
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Chung, O. Siol, T. Dingermann, and T. Winckler Protein Interactions Involved in tRNA Gene-Specific Integration of Dictyostelium discoideum Non-Long Terminal Repeat Retrotransposon TRE5-A Mol. Cell. Biol., December 15, 2007; 27(24): 8492 - 8501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Siol, M. Boutliliss, T. Chung, G. Glockner, T. Dingermann, and T. Winckler Role of RNA Polymerase III Transcription Factors in the Selection of Integration Sites by the Dictyostelium Non-Long Terminal Repeat Retrotransposon TRE5-A Mol. Cell. Biol., November 15, 2006; 26(22): 8242 - 8251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Bachman, M. E. Gelbart, T. Tsukiyama, and J. D. Boeke TFIIIB subunit Bdp1p is required for periodic integration of the Ty1 retrotransposon and targeting of Isw2p to S. cerevisiae tDNAs Genes & Dev., April 15, 2005; 19(8): 955 - 964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. L. Singleton and H. L. Levin A Long Terminal Repeat Retrotransposon of Fission Yeast Has Strong Preferences for Specific Sites of Insertion Eukaryot. Cell, February 1, 2002; 1(1): 44 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Yieh, G. Kassavetis, E. P. Geiduschek, and S. B. Sandmeyer The Brf and TATA-binding Protein Subunits of the RNA Polymerase III Transcription Factor IIIB Mediate Position-specific Integration of the Gypsy-like Element, Ty3 J. Biol. Chem., September 15, 2000; 275(38): 29800 - 29807. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
