W. Broothaerts

cDNA cloning and molecular analysis of two self-incompatibility alleles from apple

Complementary DNA clones representing two alleles of the self-incompatibility (S) locus of apple (Malus × domestica Borkh.) have been isolated and characterised. One of the alleles corresponds to a 29 kDa ribonuclease (S-RNase) that was purified from pistil tissue. On northern blots, both cDNAs hybridized to a transcript that was only present in pistils and not in the other plant tissues analysed. Corresponding genomic sequences, amplified by PCR, were found to contain a single intron of 138 bp and 1100 bp respectively. Comparison of both sequences shows that the cDNAs encode mature proteins containing 65% of identical residues. Eight invariable cysteine residues, conserved regions around two histidines thought to play a role in RNA catalysis, and a number of other distinct residues are conserved between the apple S-RNases and similar proteins in the family Solanaceae. As this is the first report of sequences of S-alleles from a species belonging to a family that is not related with the Solanaceae, the structural features of S-RNase deduced from a comparison of their sequences are discussed.

A molecular method for S-allele identification in apple based on allele-specific PCR

AbstractcDNA sequences corresponding to two self-incompatibility alleles (S-alleles) of the apple cv ‘Golden Delicious’ have previously been described, and now we report the identification of three additional S-allele cDNAs of apple, one of which was isolated from a pistil cDNA library of cv ‘Idared’ and two of which were obtained by reverse transcription-PCR (RT-PCR) on pistil RNA of cv ‘Queens Cox’. A comparison of the deduced amino acid sequences of these five S-allele cDNAs revealed an average homology of 69%. Based on the nucleotide sequences of these S-allele cDNAs, we developed a molecular technique for the diagnostic identification of the five different S-alleles in apple cultivars. The method used consists of allele-specific PCR amplification of genomic DNA followed by digestion of the amplification product with an allele-specific restriction endonuclease. Analysis of a number of apple cultivars with known S-phenotype consistently showed coincidence of phenotypic and direct molecular data of the S-allele constitution of the cultivars. It is concluded that the S-allele identification approach reported here provides a rapid and useful method to determine the S-genotype of apple cultivars.

New findings in apple S-genotype analysis resolve previous confusion and request the re-numbering of some S-alleles

Abstract.Apple trees display gametophytic self-incompatibility which is controlled by a series of polymorphic S-alleles. To resolve the discrepancies in S-allele assignment that appeared in the literature, we have re-examined the identity of S-alleles known from domestic apple cultivars. Upon an alignment of S-allele nucleotide sequences, we designed allele-specific primer pairs to selectively amplify a single S-allele per reaction. Alternatively, highly similar S-alleles that were co-amplified with the same primer pair were discriminated through their distinct restriction digestion pattern. This is an extension of our previously developed allele-specific PCR amplification approach to reveal the S-genotypes in apple cultivars. Amplification parameters were optimised for the unique detection of the 15 apple S-alleles of which the nucleotide sequences are known. Both the old cultivars with a known S-genotype and a number of more common cultivars were assayed with this method. In most cases, our data coincided with those obtained through phenotypic and S-RNase analysis. However, three S-alleles were shown to relate to RNases that were previously proposed as being encoded by distinct S-alleles. For another S-allele the corresponding gene product has not been discriminated. Consequently, we propose the re-numbering of these four S-alleles. Furthermore, two alleles that were previously identified as S27a and S27b now received a distinct number, despite their identical S-specificity. To ease widespread future analysis of S-genotypes, we identified common cultivars that may function as a witness for bearing a particular S-allele. We discuss the assignment of new S-alleles which should help to avoid further confusion.

Use of the multi-allelic self-incompatibility gene in apple to assess homozygocity in shoots obtained through haploid induction

Abstract To obtain homozygous genotypes of apple, we have induced haploid development of either the female or the male gametes by parthenogenesis in situ and anther culture, respectively. Of the shoots obtained, which were mainly of a non-haploid nature, some could be derived from fertilised egg cells or from sporophytic anther tissue. In order to select the shoots having a true haploid origin, and thus homozygotes, we decided to use the single multi-allelic self-incompatibility gene as a molecular marker to discriminate homozygous from heterozygous individuals. The rationale behind this approach was that diploid apple cultivars contain 2 different alleles of the S-gene and therefore the haploid induced shoots obtained from them should have only one of the alleles of the single parent. The parental cultivars used were ‘Idared’ (parthenogenesis in situ) and ‘Braeburn’ (androgenesis), and their S-genotypes were known, except for 1 of the ‘Braeburn’S-alleles. To stimulate parthenogenetic development ‘Idared’ styles were pollinated with irradiated ‘Baskatong’ pollen, the S-alleles of the latter (2n) cultivar were also unknown. The cloning and sequence analysis of these 3 unidentified S-alleles, 1 from ‘Braeburn’ and 2 from ‘Baskatong’ is described, and we show that they correspond to the S24-, S26- and S27-alleles. We have optimised a method for analysis of the S-alleles of ‘Idared/Baskatong’- or ‘Braeburn’-derived in vitro plant tissues and have shown that this approach can be applied for the screening of the in vitro shoots for their haploid origin.

Petunia hybrida S-proteins: ribonuclease activity and the role of their glycan side chains in self-incompatibility

SummarySelf-incompatibility in flowering plants is controlled by the S-gene, encoding stylar S (allele-specific) glycoproteins. In addition to three previously characterized Petunia hybrida S-proteins, we identified by N-terminal sequence analysis another stylar S-protein, co-segregating with the Sb-allele. Purified S-proteins reveal biological activity, as is demonstrated for two of them by the allele-specific inhibition of pollen tube growth in vitro. Moreover, the four isolated S-proteins are ribonucleases (S-RNases). Specific activities vary from 30 (S1) to 1000 (S2) units per min per mg protein. We attempted to investigate the functionality of the carbohydrate portion of the S-RNases. Deglycosylation studies with the enzyme peptide-N-glycosidase F (PNGase F) reveals differences in the number of N-linked glycan chains present on the four S-RNases. Variability in the extent of glycosylation accounts for most of the molecular weight differences observed among these proteins. By amino acid sequencing, the positions of two of the three N-glycosylation sites on the S2-RNase could be located near the N-terminus. Enzymic removal of the glycan side chains has no effect on the RNase activity of native S-RNases. This suggests another role of the glycan moiety in the self-incompatibility mechanism.

S-Allele characterization in self-incompatible pear (Pyrus communis L.)

Abstract. Gametophytic self-incompatibility, a natural mechanism occurring in pear and other fruit-tree species, is usually controlled by the S-locus with allelic variants (S1, S2, Sn). Recently, biochemical and molecular tools have determined the S-genotype of cultivars in various species. The present study determined the S-locus composition of ten European pear cultivars via S-PCR molecular assay, thereby obviating time-consuming fieldwork whose results are often ambiguous because of environmental effects. To verify the S-PCR assay, two putative S-allele DNA fragments of Japanese pear were isolated; their sequences proved to be identical to those reported in the databank. Six S-allele fragments of European pear were then sequenced. While field data confirmed the molecular results, fully and half-compatible field crosses were not distinguishable.

Self-fertile apple resulting from S-RNase gene silencing

Self-incompatibility (SI) restricts fertilisation and fruit setting in many tree fruit crops. In apple, we have produced transgenic trees harbouring extra copies of the endogenous S-gene controlling SI. Two independent transgenic genotypes were characterised in detail. Controlled self- and cross-pollination of the flowers of trees from both genotypes over a 3-year-period showed that the transgenic lines produced normal levels of fruit and seeds after selfing. In contrast, the controls produced much less fruit following self- compared to cross-pollination. Fruit set data correlated with the results of microscopic evaluation of pollen tube growth through the pistil, which revealed inhibition after selfing in the controls but not in the transgenic lines. The self-fertile phenotype was associated with the complete absence of pistil S-RNase proteins, which are the products of the targeted S-gene. These results confirm that self-fertility was due to inhibition of expression of the S-RNase gene in the pistil, resulting in un-arrested self-pollen tube growth, and fertilisation.

Re-examination of the self-incompatibility genotype of apple cultivars containing putative 'new' S-alleles

Abstract Recently, the self-incompatibility (S-) genotypes of 56 apple cultivars were examined by protein analysis, which led to the identification by Boskovic and Tobutt of 14 putative ’new’ S-alleles, S12 to S25. This paper reports a re-examination of the S-genotypes of some of these cultivars through S-allele ’specific’ PCR and sequence analysis. The results obtained by this analysis indicated that the number of S-alleles that are present in apple is probably smaller than the number proposed by Boskovic and Tobutt. The existence of three ’new’ S-alleles (S20, S22 and S24) was confirmed. The existence of two other putative ’new’ S-alleles (S23 and S25) was, however, contradicted. The coding sequences of the S-alleles that correspond to the S10 and the S25 ribonuclease bands as well as those corresponding to the S22 and the S23 ribonuclease bands were shown to be identical in sequence. Interestingly, the S-allele corresponding to the S22 and the S23 ribonuclease bands shared a high sequence identity (99% identity) with S27, which was previously cloned and sequenced from Baskatong, but which was not included in the analysis conducted by Boskovic and Tobutt. Both S-alleles only differ in point mutations, which are not translated into differences in amino-acid sequence. To our knowledge, this is the first report of two S-alleles that differ at the nucleotide level but still encode for identical S-RNases. The implications of these observations for determining the S-genotypes of plants by PCR analysis or protein analysis are discussed.

Purification and N-terminal sequencing of style glycoproteins associated with self-incompatibility in Petunia hybrida

We report isolation and N-terminal amino acid sequencing of three style glycoproteins, which segregate with three S (self-incompatibility) alleles of Petunia hybrida. The S-glycoproteins were expressed mainly in the upper part of the pistil and showed an increasing concentration during flower development. The glycoproteins were purified by a combination of ConA-Sepharose and cation exchange fast protein liquid chromatography. The amount of S-glycoproteins recovered from style extracts varied from 0.5 to 1.6 μg per style, which was 40–60% of the amount recovered by a simplified analytical method. N-terminal amino acid sequences of S1-, S2- and S3-glycoprotein showed homology within the fifteen amino terminal residues. These amino acid sequences were compared with the previously published sequences of S-glycoproteins from Nicotiana alata and Lycopersicon peruvianum.

S-RNases in apple are expressed in the pistil along the pollen tube growth path

Abstract The molecular bases of self-incompatibility have been intensively studied in a restricted number of model species, but for most families the expression and distribution of S-proteins is unknown. In this work, pistil cryosections from apple were used for in situ detection of S-proteins. Two specific antibodies, one against the S3-protein and another against all apple S-proteins were used. S-proteins were shown to be localised in the intercellular space of the transmitting tissue, both in the stigma and style, which agrees with the proposed mechanism of action for S-RNases in gametophytic self-incompatibility. Some intracellular labelling was also observed in all ovary sections, confined to one layer of the nucellus surrounding the embryo sac, but this labelling was found to be non-S-allele-specific. Nevertheless, the signal in the ovary was tissue-specific, which may indicate that some component not encoded by the S-locus but similar to S-proteins was detected. To the best of our knowledge this is the first report on the precise distribution of S-RNases in a rosaceous species.