Dehydrins (DHNs) are part of a large group of highly hydrophilic proteins known as LEA (Late Embryogenesis Abundant). They were originally identified as group II of the LEA proteins. The distinctive feature of all DHNs is a conserved, lysine-rich 15-amino acid domain, EKKGIMDKIKEKLPG, named the K-segment. It is usually present near the C-terminus. Other typical dehydrin features are: a track of Ser residues (the S-segment); a consensus motif, T/VDEYGNP (the Y-segment), located near the N-terminus; and less conserved regions, usually rich in polar amino acids (the Φ-segments). They do not display a well-defined secondary structure. The number and order of the Y-, S-and K-segments define different DHN sub-classes: YnSKn, YnKn, SKn, Kn and KnS. Dehydrins are distributed in a wide range of organisms including the higher plants, algae, yeast and cyanobacteria. They accumulate late in embryogenesis, and in nearly all the vegetative tissues during normal growth conditions and in response to stress leading to cellular dehydration (e.g. drought, low temperature and salinity). DHNs are localized in different cell compartments, such as the cytosol, nucleus, mitochondria, vacuole, and the vicinity of the plasma membrane; however, they are primarily localized to the cytoplasm and nucleus. The precise function of dehydrins has not been established yet, but in vitro experiments revealed that some DHNs (YSKn-type) bind to lipid vesicles that contain acidic phospholipids, and others (KnS) were shown to bind metals and have the ability to scavenge hydroxyl radicals [Asghar, R. et al. Protoplasma 177 (1994) 87–94], protect lipid membranes against peroxidation or display cryoprotective activity towards freezing-sensitive enzymes. The SKn-and K-type seem to be directly involved in cold acclimation processes. The main question arising from the in vitro findings is whether each DHN structural type could possess a specific function and tissue distribution. Much recent in vitro data clearly indicates that dehydrins belonging to different subclasses exhibit distinct functions.

/pdf/plant-dehydrins-tissue-location-structure-and-function-2d07s83jsi.pdf

Plant dehydrins--tissue location, structure and function.

A 745 bp sequence (pSau3A9) located at the centromeres of several cereal species was isolated from a sorghum BAC library by Jiang et al. (1996, Proc. Natl Acad. Sci. USA, 93, 14210-14213). We have amplified a partially homologous 809 bp sequence from barely genomic DNA by PCR and localized it to the centromeres of barley, wheat and rye chromosomes by fluorescent in situ hybridization (FISH). Sequence analysis showed this barley homolog of pSau3A9 to have high similarity to the integrase region of the polyprotein gene of Ty3/gypsy group retrotransposons. Using this integrase sequence as a probe, several clones were isolated from a lambda library constructed of genomic barley DNA. One of the lambda clones contained coding regions for all five catalytic sites characteristic of the retrotransposon polyprotein. Two direct repeats flanking the polyprotein gene are homologous to the cereal centromeric sequence described by Aragon-Alcaide et al. (1996, Chromosoma, 105, 261-268) and may represent all or part of the long-terminal repeats (LTRs). Different plasmid subclones containing various regions of the lambda clone were used in FISH to show that the entire polyprotein gene and upstream flanking sequences, including the presumed LTR, are present at barley centromeres. The preferential (or exclusive) localization of an apparently complete retroelement within the centromeric regions of several cereal species raises interesting questions about its role in karyotype evolution and centromere function.

A Ty3/gypsy retrotransposon-like sequence localizes to the centromeric regions of cereal chromosomes.

Retroviruses and retrotransposons make up the broad class of retroelements replicating and transposing via reverse transcriptase. Retroelements have recently been found to be ubiquitous in the plants. We report here the isolation, sequence and analysis of a retroelement from barley (Hordeum vulgare L.) with all the features of a copia-like retrotransposon. This is named BARE-1 (for BArley RetroElement 1), the first such element described for barley. BARE-1 is 12 088 bp, with long terminal repeats (LTRs) of 1829 bp containing perfect 6 bp inverted repeats at their ends and flanked by 4 bp direct repeats in the host DNA. Between the long terminal repeats is an internal domain with a derived amino acid sequence of 1285 residues, bearing homology to the gag, pro, int and rt domains of retroviruses and both plant and non-plant copia-like retrotransposons. Cultivated barley contains about 5000 elements in the genome similar to the BARE-1 putative gag domain, but ten-fold more hybridizing to rt or LTR probes. The particular BARE-1 element reported here appears to be inactive, as the putative protein-coding domain is interrupted by four stop codons and a frameshift. In addition, the 3′ LTR is 4% divergent from the 5′ LTR and contains a 3135 bp insertion. Nevertheless, we have recently detected transcripts hybridizing to BARE-1 on northern blots, presumably from active copies. Analysis of BARE-1 expression and function in barley is currently underway.

BARE-1, a copia-like retroelement in barley (Hordeum vulgare L.)

Simple sequence repeat (SSR)-based genetic markers are being actively developed for the majority of crop plant species. In barley, characterization of 290 dinucleotide repeat-containing clones from SSR-enriched libraries has revealed that a high percentage are associated with cereal retrotransposon-like and other dispersed repetitive elements. Associations found were with BARE-1, WIS2-1A, PREM1 and the dispersed repetitive element R173. Additional similarities between different SSR clones, which have no matches in DNA sequence databases, indicate that this phenomenon is probably widespread in the barley genome. Sequence homologies to the non-coding regions of several cereal genes were also explained by homology to mobile genetic elements. The SSRs found can therefore be classified into two types: (1) those with unique sequences on either flank, and (2) those which are intimately associated with retro-transposons and other dispersed repetitive elements. As the cereal genome is thought to consist largely of this type of DNA, some random association would be expected. However, the conserved positions of the SSRs, relative to repetitive elements, indicate that they have arisen non-randomly. Furthermore, this class of SSRs can be classified into three subtypes: (1) those which are positioned 3' of a transposable element with unique sequence on the other flank, (2) those positioned 5' of a transposable element, and (3) those which have arisen from an internal sequence and so have transposable element sequence on both flanks. The first appear to be analogous to the class of SSRs in mammalian systems which are associated with Alu elements and SINEs (short interspersed elements) and which have been postulated to arise following integration of an extended and polyadenylated retro-transcript into the host genome, followed by mutation of the poly(A) tract and expansion into an SSR. For the second, we postulate that a proto-SSR (A-rich sequence) has acted as a 'landing pad' for transposable element insertion (rather than being the result of insertion), while the third includes those which have evolved as a component of an active transposable element which has spread throughout the genome during bursts of transposition activity. The implications of these associations for genome and SSR evolution in barley are discussed.

Intimate association of microsatellite repeats with retrotransposons and other dispersed repetitive elements in barley.

DNA from Triticum tauschii (the D-genome donor to hexaploid wheat, Triticum aestivum) has been cloned using the restriction endonuclease PstI to generate fragments for insertion into the plasmid pB...

The molecular–genetic analysis of Triticum tauschii, the D-genome donor to hexaploid wheat

The rye-specific R173 family of repeated DNA sequences consists of ca. 15 000 individual copies per diploid rye (Secale cereale) genome and is distributed over all 7 rye chromosomes in a dispersed manner. Individual R173 elements vary in size between 3 and 6 kb, are generally not arranged as tandem repeats and are flanked by both multi-copy and single-copy sequences. DNA sequence analysis of three R173 elements (R173-1, R173-2 and R173-3) demonstrated a high degree of homology in conserved domains. The structure of R173-1 was quite different from the other two elements: long direct repeats, which represent a rye-specific repetitive sequence, were found at the ends and a 600 bp long domain was replaced by an unrelated sequence of approximately equal size. R173-2 and R173-3 were extremely similar to each other with the exception of a terminal truncation of R173-2. No open reading frames for proteins >20 kDa were present and a database search failed to detect significant homologies to published protein sequences. Despite the transposon like genomic organisation of the R173 family, individual elements lacked sequence features frequently associated with transposons and retrotransposons. In contrast, two of the regions flanking R173 elements showed strong DNA homologies to a 850 bp long region of a proposed wheat retrotransposon and to a 300 bp long region downstream of the wheatGlu-D1 gene.

Structural heterogeneity in the R173 family of rye-specific repetitive DNA sequences.

In contrast to all other characterised families, repetitive DNA sequences of the rye-specific R173 family occur generally as a monomer and have an unusually long repeat unit. A genomic library was generated from a wheat line with three copies of the short arm of chromosome 1 of rye, 1RS. Seventy-seven λ clones, representing independent members of the family, were isolated by hybridization to pAW173. They share a common region of approximately 3.5 kbp, which is free of large internal repeats and therefore constitutes the basic unit of the dispersed R173 family. The analysis of flanking regions showed that individual members of the R173 family are generally not found in the vicinity of other characterised families of repetitive DNA. The flanking regions of four selected λ clones were different from each other and comprised both repetitive and low-copy sequences. A restriction fragment length polymorphism probe, mapping to the short arm of chromosome 1, was obtained by subcloning of flanking regions of the R...

Jin-Yuan Liu

Papers

Structural heterogeneity in the R173 family of rye-specific repetitive DNA sequences.

The R173 family of rye-specific repetitive DNA sequences: a structural analysis