**4.1. Identification of S-RNase**

The physiology and mechanisms of GSI are actually most substantially studied in solanaceous vegetable species. The development of cDNA glycoproteins co-segregated together with S alleles was first cloned via *Nicotiana alata* [2-3]. The deduced amino p sequence clearly implicated stylar RNase involvement in the recognition and rejection reaction inside the style. Along with other studies, it has been shown that in Solanaceae, the S allele product inside pistil is usually a highly simple glycoprotein comprising sequence motif characteristics in the active site in the fungal RNase T2 [58] and RH [38], termed S-RNase [83, 84].

At one point when transgenic analyses with *Petunia inflata* and *Nicotiana* proved that the S-RNase alone is enough for determining the specificity in the GSI pollen rejection response inside Solanaceae [67, 88], Sassa et al. (1996) and Broothaerts et al. (1995) reported that SRNases are associated together with GSI of *Pyrus* and *Malus*in Rosaceae. The finding that these families recruited the same molecule as the GSI pistil determinant has been unexpected because Solanaceae (*Asteridae*) and Rosaceae (*Rosidae*) are phylogenetically remote [12, 45]. Although deduced amino p sequences from cDNAs of S-RNases of *Pyrus* and *Malus* may be similar and support the active site of T2/S-type RNases, differences were clear involving the rosaceous and solanaceous S-RNases [105-108, 126]. Later, Xue et al. (1996) cloned the real S-RNase through‐ out *Antirrhinum* inside Plantaginaceae, a tightly related family of the Solanaceae.

As *Prunus* is one of the Rosaceae, it had been readily predicted that *Prunus* even offers an S-RNase-based GSI system. However, S-RNases remained unidentified for countless years after the real cloning in the *Pyrus* and *Malus* S-RNases, because polymerase chain reaction (PCR) cloning solutions for *Prunus* S-RNase were hindered by its fairly low DNA string, similar to *Pyrus* and *Malus* SRNases as well as the presence of *Prunus* RNase genes which cannot involved with GSI. The first clue for the cloning of *Prunus* S-RNase was obtained when N-terminal sequences of almond (*Prunus dulcis*) SRNase were reported [120]. By the N-terminal amino p sequences, sweet cherry (*Prunus avium*) [120] and almond [125] S-RNases were cloned. Currently, sequences of over 100 *Prunus* S-RNase alleles are actually deposited in GenBank.

#### **4.2. Pollen S determinant**

#### *4.2.1. Identification of F-box gene*

The pollen S determinant in the S-RNase-based GSI in Rosaceae, Solanaceae, and Plantagina‐ ceae was discovered decades after the real stylar determinant, S-RNase. The subcentromeric location in the solanaceous and plantaginaceous S locus experienced had long prevented chromosome walking [23, 142]. The first clue for the identification in the pollen S was from the S locus in the Plantaginaceae. Sequencing analysis in the *Antirrhinum hispanicum* S locus exposed the presence of the pollen-expressed F-box gene (AhSLF for a. *hispanicum* S locus Fbox) located 9 kb downstream of S2-RNase [65].

Even though it was speculated that the F-box protein gene encoded the real pollen S, only one particular allele has been cloned. The S locus of *Prunus*, which is located right at the end of the linkage group 6 [20], was much more compact than those of Solanaceae and Plantaginaceae. Two separate groups in Japan successfully sequenced the real S locus of *Prunus* beginning from the S-RNase [22, 127]. Ushijima et al. (2003) did DNA sequencing and transcriptional analyses for the genomic areas that flank real almond (*P. dulcis*) S-RNase and identified polymorphic and non-polymorphic S locus F-box genes, called SFB of S haplotype-specific F-box gene and SLF for S locus F-box, respectively. At present, SLF continues to be referred to as SLFL1 [79] after the nomenclature of Entani et al. (2003). The options that come with SFB, including the high degree of allelic polymorphism, pollen-specific appearance, and the close physical distance to the S-RNase many supported that SFB may be the male determinant of GSI in almond. The same research group also found SFB in the cherry (*Prunus*) S locus in their attempt to compare the same S-RNase allele on SC and SI kinds of cherries, *R. cerasus* and *P. avium*, respectively [140]. Another study group in Japan reviewed the S locus place of a couple different S haplotypes in Japanese apricot (*Prunus mume*) and found some F-box genes [22]. Among them, SLF of S locus F-box, that includes a different name but can be orthologous to SFB throughout almond and cherries, shows a high level of allelic string diversity and was supposed to be a prospect of pollen S. The other F-box genes found, SLFL1, SLFL2, SLFL3 of SLF including gene 1, 2 and 3, respectively, showed far lower allelic string diversity. The pollen S candidate of Japanese apricot has been independently cloned in a study by Yamane et al. (2003d) and named differently as PmSFB. Since then, the *Prunus* pollen S was initially referred to by a couple of different words or terms such as ''SFB'' and also ''SLF''. We used ''SFB'' in this particular review to show the various features in *Prunus* SFB than the SLF in Solanacae and Plantaginaceae.

Direct evidence that S locus F-box gene adjustments of allele specificity in the pollen was from a transgenic research in *R. inflata* [109]. This kind of experiment employed a well-known phenomenon termed ''competitive interaction'', where heteroallelic pollen which has two unique pollen S alleles work in pistils together with one as well as both cognate S haplotypes [18, 26-29].

Although transgenic analyses in *Prunus* are hindered by a long period and lacking a useful transformation system, molecular characterization in the S haplotypes throughout SC mutants supplied indirect but very good supporting facts for SFB staying the pollen S [125]. Later, multiple F-box genes, called SFB (SFBBs), were found in candidate pollen S genes in *Malus* and *Pyrus* [105-107]. Even though it has also been shown that the SFBB-gamma gene is not likely involved with the specificity determination of GSI reaction [131], it is still possible that just as with the other families and genera, one or a number of the SFBBs could be the pollen S.
