Vertical transmission of Prunus necrotic ringspot virus: hitch-hiking from gametes to seedling

Transcription

1 Journal of General Virology (2009), 90, DOI /vir Vertical transmission of Prunus necrotic ringspot virus: hitch-hiking from gametes to seedling Khalid Amari, Lorenzo Burgos, Vicente Pallás3 and Maria Amelia Sánchez-Pina Correspondence Maria Amelia Sánchez-Pina Received 16 December 2008 Accepted 4 March 2009 Dpto. de Biología del Estrés y Patología Vegetal, CEBAS-CSIC, Campus Universitario de Espinardo, PO Box 164, Espinardo-Murcia, Spain The aim of this work was to follow Prunus necrotic ringspot virus (PNRSV) infection in apricot reproductive tissues and transmission of the virus to the next generation. For this, an analysis of viral distribution in apricot reproductive organs was carried out at different developmental stages. PNRSV was detected in reproductive tissues during gametogenesis. The virus was always present in the nucellus and, in some cases, in the embryo sac. Studies within infected seeds at the embryo globular stage revealed that PNRSV infects all parts of the seed, including embryo, endosperm and testa. In the torpedo and bent cotyledon developmental stages, high concentrations of the virus were detected in the testa and endosperm. At seed maturity, PNRSV accumulated slightly more in the embryo than in the cotyledons. In situ hybridization showed the presence of PNRSV RNA in embryos obtained following hand-pollination of virus-free pistils with infected pollen. Interestingly, tissue-printing from fruits obtained from these pistils showed viral RNA in the periphery of the fruits, whereas crosses between infected pistils and infected pollen resulted in a total invasion of the fruits. Taken together, these results shed light on the vertical transmission of PNRSV from gametes to seedlings. INTRODUCTION Seed transmission is a common mechanism of virus perpetuation, as about 20 % of plant viruses are seed transmitted (Hull, 2002; Mink, 1993). This process may play an important role in dissemination of a virus, especially as a result of breeding programmes, the international exchange of germplasm and the use of seedlings as rootstocks. The low seed transmission rate of some viruses is not a good indicator of the potential importance of a disease. The introduction of a poorly seedtransmitted virus in a new area can be catastrophic if an efficient vector is available. In general, viral seed transmission (Mink, 1993; Maule & Wang, 1996) depends on infection of the embryo, which can occur by three alternative pathways: the virus can (i) reach the embryo during fertilization after being transported by the male gametophyte, (ii) invade the ovule and then the embryo, also during fertilization (both ways are indirect) and/or (iii) directly invade the embryo during its development after fertilization. However, little is known about how viruses can move from sporophytic to gametophytic and back again into sporophytic tissues, in 3Present address: Instituto de Biología Celular y Molecular de Plantas, UPV-CSIC, Avda. de los Naranjos s/n, Valencia, Spain. A supplementary figure showing dwarfed plants grown from PNRSVinfected seeds is available with the online version of this paper. a process known as vertical transmission. The best example of the processes involved in vertical transmission is the work of Wang & Maule (1997) who demonstrated that pea early browning virus is transmitted by seed after infection of both gametes (indirect way) and that pea seed borne mosaic virus (PSbMV) is seed-transmitted, although it is not able to infect the gametes (direct way). The model of PSbMV transmission was elucidated by Roberts et al. (2003) who showed that the virus was able to first infect the testa and then invade the endosperm using the existing symplastic connection. The virus subsequently reached the embryo through the pore-like structures that are located in the sheath separating the embryo from the suspensor. To improve our knowledge of how viruses infect seeds, we have studied, in this work, the possible routes by which Prunus necrotic ringspot virus (PNRSV) is vertically transmitted in Prunus, more specifically, in apricot trees. Apricot is an important stone fruit tree crop in the Mediterranean region and there is a relatively high incidence of PNRSV in orchards of this species (Myrta et al., 2002). PNRSV is the causal agent of several diseases that affect most cultivated stone fruit trees. It causes serious economic damage (Uyemoto & Scott, 1992), decreasing tree growth by up to 30 % and their yield by %; PNRSV infection also makes infected trees very susceptible to chilling damage. PNRSV belongs to the genus Ilarvirus and has G 2009 SGM Printed in Great Britain 1767

2 K. Amari and others the same genomic organization as, and encodes translation products functionally similar to, those of the genera Bromo-, Cucumo-, Olea- and Alfamovirus, which all belong to the family Bromoviridae (Aparicio et al., 2001, 2003, 2006; Aparicio & Pallás, 2002; Sánchez-Navarro & Pallás, 1997; Sánchez-Navarro et al., 2006; Herranz & Pallás 2004; Herranz et al., 2005). The virus is pollen- and seedtransmitted (Cole et al., 1982; Hamilton et al., 1984; Kelley & Cameron, 1986; Aparicio et al., 1999; Amari et al., 2007), which contributes to its rapid spread in stone fruit trees (Mink, 1992). The seed transmission rate for PNRSV varies among the infected species. Previous studies have demonstrated that, in the case of cherry, PNRSV was transmitted in % (Vértesy, 1976; Kryczynski et al., 1992) of cases, while in almond it was only 9 % (Barba et al., 1986). The routes by which the seeds become infected are not known in detail, mainly due to the difficulties involved in working with stone fruit trees, which have long generation cycles and bloom only once a year. This paper studies the progress of infection of PNRSV in apricot reproductive organs during their development up until the seedling stage, in an attempt to explain its mode of transmission. METHODS Plant material. Apricot trees (Prunus armeniaca L.) cultivar Búlida from an open field at Calasparra (Murcia, Spain) were checked for the presence of PNRSV during routine virus indexing, using a previously described RT-PCR method (Sánchez-Navarro et al., 2005) that permits simultaneous detection and differentiation of eight important viruses that affect stone fruit trees, among them PNRSV. Flower buds at different developmental stages were collected from both virus-free and infected trees and examined for the presence of PNRSV. The ovaries were separated from pistils prior to fixation and paraffin embedding. Seeds were obtained from fruits at different ripening stages. Total RNA extraction. Several embryos, testa and cotyledons were excised from ten immature seeds, of both virus-free and infected trees, and homogenized in liquid nitrogen. Samples of homogenized material and carefully separated endosperm were extracted with 2 volumes of 1 : 1 ratio of extraction buffer (350 mm glycine, 48 mm NaOH, 34 mm NaCl, 40 mm EDTA and 4 % SDS) and phenol/ chloroform/isoamylalcohol (25 : 24 : 1). After shaking for 2 min, the homogenate was centrifuged for 10 min at g at 4 uc; the upper phase was extracted twice with phenol, followed by precipitation with 1/10 volume of 3 M sodium acetate, ph 5.2, and 2.5 volumes of ethanol. The pellet was resuspended in the appropriate volume of nuclease-free water and total RNA integrity was checked by agarose gel electrophoresis before being quantified using a Nanodrop ND Spectrophotometer (Thermo Fisher Scientific) and equalized to 0.2 mg ml 21. Dot-blot hybridization. Five microlitres (1 mg) total RNA extracted from infected and virus-free immature seeds were applied to nylon membranes (Roche), either directly or in a serial fivefold dilution. Membranes were air-dried and the nucleic acids were cross-linked by exposure to UV irradiation on a transilluminator. Pre-hybridization and hybridization were carried out as described previously (Pallás et al., 1998). Chemiluminescent detection with CSPD reagent as substrate was performed as recommended by the manufacturer (Roche). A PNRSV non-radioactive riboprobe was obtained by transcribing a ppn4-890 clone corresponding to an 890 nt full-length cdna of PNRSV subgenomic RNA 4, that codes for the coat protein (Sánchez-Navarro & Pallás, 1997). Tissue-printing hybridization. Tissue-printing of fruits was carried out as described previously (Más & Pallás, 1995). A total of 12 fruits divided into groups of three (making a total of four membranes) from different sides of the trees were freshly cut and imprinted directly onto nylon membranes. The membranes were treated as described for dot-blot hybridization. Seed germination. Seeds collected from infected and virus-free trees were cold-stratified in humid vermiculite for 3 months in a cold room (4 7 uc). When the seeds began to germinate, they were individually transplanted to soil and the seedlings were maintained in the greenhouse to observe any symptoms and detect PNRSV. In situ hybridization. Ten flower buds, ovaries and seeds were fixed by immersion in a freshly made mixture of 4 % paraformaldehyde and 2.5 % glutaraldehyde in 0.1 M phosphate buffer (ph 7.2), at 4 uc for 4 h and embedded in paraffin (Paraplast Plus) as described previously (García-Castillo et al., 2001; Amari et al., 2007). The same probe used for dot-blot hybridization was used for in situ hybridization. In situ hybridization was performed at 55 uc overnight. Slide pre-treatments, hybridization and colourigenic detection were carried out as described previously (García-Castillo et al., 2001; Amari et al., 2007). Sections were examined and photographed with a Leica DC 500 digital camera using a DMRB Leitz microscope. RESULTS PNRSV invades apricot reproductive organs from early stages of development PNRSV was detected in the early stages of ovule development (Fig. 1a) in both the ovule and the ovary. As expected, no viral RNA was detected in virus-free flower buds at the same developmental stage (Fig. 1b). PNRSV was detected in ovules in the megaspore stage (Fig. 1c) when it was present in the nucellus, while the megaspore itself remained free of the virus. PNRSV was localized, in later developmental stages, in the funiculus as well as in its corresponding ovule (Fig. 2a). However, at this developmental stage, distribution of virus was not homogeneous inside the ovary, since the other funiculus and its corresponding ovule were both virus-free (Fig. 2a). The study of later developmental stages showed that PNRSV invades all parts of the ovule, including inner and outer integuments surrounding the nucellus, which encloses the embryo sac (Fig. 2b, c). At this stage, two different infection patterns were observed in the ovule: in the first, the whole ovule was infected (Fig. 2b) and in the second, the nucellus was free of PNRSV except for the cells located close to and surrounding the embryo sac (Fig. 2c). However, PNRSV was always observed in the nucellar cells located around the embryo sac, even in the more advanced developmental stage studied (Fig. 2d), where viral RNAs were rarely detected inside the embryo sac and, when found, they were in the fused polar cells (Fig. 2e) Journal of General Virology 90

3 Vertical transmission of PNRSV in apricot Fig. 1. In situ hybridization of PNRSV in cross sections of young winter flowers. (a) Infected young flower bud showing viral RNA (blue) in the pistil (P) and ovule (Ov) inside the ovary (O), as well as in anthers (A). (b) No viral RNA was detected in flower buds from virus-free trees. (c) Ovule at the early developmental stage showing viral RNA in the nucellus (Nu) surrounding the megaspore (M), which is not infected. Infection of apricot embryos and seeds by PNRSV Dot-blot hybridization of PNRSV RNA obtained from different parts of the seeds (embryo, endosperm, testa and cotyledons) showed that in the torpedo (Fig. 3a) and bent cotyledon (Fig. 3b) stages, the lowest amount of PNRSV was detected in the embryos, while the greatest amount was detected in the endosperm and testa (Fig. 3a, b). In the mature seed stage, endosperm was totally absent and so embryos were separated from their cotyledons for the analysis; which showed that the PNRSV signal in the embryos was strongest at this stage, and even slightly higher than in the corresponding cotyledons (Fig. 3c). The PNRSV invasion of the apricot embryo in early developmental stages was confirmed by in situ hybridization studies of immature infected seeds, where PNRSV was detected even in embryos at the globular stage; here, it showed a heterogeneous distribution but was mostly present in the apical region (Fig. 4a). In addition, PNRSV RNA was detected in embryos in the globular developmental stage obtained from the hand-pollination of virus-free flowers with infected pollen grains (Fig. 4b). No in situ hybridization signal was observed in virus-free embryos at this stage (Fig. 4c). We also observed viral infection of the seed integuments, nucellus and testa in immature seeds (Fig. 4d g) using in Fig. 2. Distribution of PNRSV in female reproductive organs. (a) PNRSV RNA was localized in the funiculus (F) and its corresponding ovule (Ov) tissues (on the right). (b) Ovule showing infected outer and inner integument (OI and II, respectively), as well as the totally infected nucellus (Nu) surrounding the embryo sac (ES). (c) Infected ovule showing the uninfected nucellar tissue, except for the zone enclosing the embryo sac. (d, e) High magnification micrographs of the infected nucellus and their corresponding embryo sacs, including the egg cell (EC) and central (CC) cell, which is invaded by the virus in (e). situ hybridization. Infection of the testa (Fig. 4e) mainly affected areas in and around its vascular tissue (Fig. 4f, g). Seed transmission of PNRSV in apricot Tissue-printing hybridization analysis of apricot fruits resulting from the hand-pollination of virus-free plants with infected pollen revealed the presence of PNRSV only in the outer part of these fruits (Fig. 5a). However, in fruits

4 K. Amari and others Fig. 3. Dot-blot hybridization to detect PNRSV RNAs from different parts of the seed. (a, b) Serial dilutions of total RNA extracted from embryo, endosperm and testa at torpedo (a) and bent cotyledon (b) developmental stages were blotted and hybridized against viral RNA. (c) Total RNA from mature embryos and cotyledons was analysed separately showing a slightly higher accumulation of PNRSV in the embryos. Note that no positive hybridization signal was observed from virusfree controls. The positive control was RNA extracted from PNRSV-infected cucumber. resulting from the hand-pollination of infected plants with infected pollen, the virus was uniformly distributed (Fig. 5b). No signal was obtained from the tissue-printings of virus-free fruits (Fig. 5c). Furthermore, when seedlings obtained by germination of seeds from infected trees were analysed by dot-blot hybridization using total RNA extraction from leaves, 10 % of the seedlings were seen to be infected (data not shown) and the plants that were obtained showed a severe dwarfing (Supplementary Fig. S1, available in JGV Online). DISCUSSION It has been demonstrated previously that seed transmission is related to the ability of the virus to infect flower organs, especially the meristematic reproductive tissue, during early developmental stages (Mink, 1993; Bennett, 1969; Vazquez Rovere et al., 2002). We have already demonstrated that PNRSV invades apricot reproductive organs, including the pollen mother cell (PMC) stage, early in development (Amari et al., 2007). In that work, we showed that the virus reduces pollen fitness, causing poor germination and delayed growth inside the pistil. However, in the south-eastern Spanish climate (where this study was carried out), the delay in ovule maturity (Egea & Burgos, 1998; Alburquerque et al., 2000) gives the infected pollen the opportunity to fertilize the ovules (Amari et al., 2007); this was confirmed in the present study by obtaining infected embryos after hand-pollinating virus-free pistils with infected pollen. In the present study, we investigated whether female reproductive organs are infected as early in their development as the male gametophyte. Our results show that PNRSV invades the ovary in early developmental stages, since viral RNA was detected from the beginning of ovary formation (Fig. 1). Unlike our earlier work with pollen, we were unable to detect viral RNA in megaspores from young ovules by using in situ hybridization (Fig. 1c), which suggests that PNRSV cannot invade the megaspore, although the possibility that the amount of viral RNA present in this tissue was below the detection limit of the in situ hybridization method cannot be ruled out. The detection of PNRSV in the funiculus and its corresponding ovule (Fig. 2a) explains the mechanism of PNRSV invasion of the ovule, since the funiculus constitutes the last vascular connection of the ovule with the pericarp (Reiser & Fischer, 1993). Moreover, infection of the integuments (Fig. 2b) would be an early indication of testa infection, since the testa originates from the integuments (Swamy & Krishnamurthy, 1980) Journal of General Virology 90

5 Vertical transmission of PNRSV in apricot Fig. 4. In situ hybridization to determine the localization of PNRSV RNAs in embryos and seeds. (a) Infected embryo (E) at the globular stage from an infected tree, where viral RNAs are concentrated in the apical region. (b) Immature seed obtained from hand-pollination of a virus-free flower with infected pollen grains, showing infection of embryo and endosperm (En). (c) Virus-free embryo at the globular stage, showing no positive signal. (d) Cross-section of infected immature seed showing PNRSV RNAs in all parts of the seed (Mi, micropil; II, inner integument; Nu, nucellus; En, endosperm; E, embryo). (e) Infected immature seed at a later developmental stage than those shown before, showing that PNRSV RNA is infecting the testa (T) where it is located in and around its vascular tissue, as can be seen at higher magnification (f and g). Unlike in the megaspore stage, PNRSV was detected in the mature embryo sac (Fig. 2e), which may explain its vertical transmission in apricot. This infection takes place after invasion of the embryo by virus from the highly infected Fig. 5. Detection of viral RNAs by tissue-printing hybridization. (a) Cross-section of three different fruits obtained from hand pollination of virus-free flowers by PNRSV-infected pollen, where viral RNA is located only on the periphery of fruits. (b) Fruits obtained from infected trees showing uniform infection by PNRSV. (c) Tissue-printing from two different virus-free fruits. nucellar tissue (Fig. 2b). This hypothesis is supported by the fact that the nucellus is responsible for providing nutrients to the embryo sac during its development (Reiser & Fischer, 1993; Radchuk et al., 2006), as well as by previous studies which have demonstrated that the callose layer separating the megaspore from the maternal tissues is discontinuous, unlike in the case of PMCs (Johansen et al., 1994). Moreover, according to Kapil & Bhatnagar (1981), this callose layer of the female organ is absorbed and replaced by a pectocellulase layer later in development. This process could well be exploited by PNRSV to invade the embryo sac from the nucellus. The process of embryo development in dicots is often divided into four stages: globular, heart- and torpedoshaped and cotyledonary. The invasion of apricot seeds by PNRSV was studied from early (globular in immature seeds) to mature embryo stages. Our studies on viral accumulation in infected apricot seeds have shown that PNRSV is able to infect young embryo tissues during their development, at least during fertilization, since viral RNA was detected during the globular developmental stage (Fig. 4b) in young embryos obtained by the handpollination of virus-free pistils with infected pollen. Infection of the endosperm (Fig. 3a, b) may occur during double fertilization by one of the two sperm cells of

6 K. Amari and others infected pollen or by the central cell which, as we have shown, can be infected by PNRSV (Fig. 2e). Even if PNRSV cannot move from the gametes to the endosperm, infection of the latter could originate in the sporophytic maternal tissue, as has already been shown in the case of PSbMV, which is not able to infect the gametes but is able to reach the endosperm through the existing connections between the testa and the endosperm (Roberts et al., 2003). Moreover, PNRSV was localized in the testa of infected apricot seeds (Figs 3a, b and 4e g). This localization might explain the seed transmission of the virus but only if PNRSV can invade the embryo via the endosperm from the testa, because until now, no virus except tobacco mosaic virus (TMV) has been shown to be transmitted to seedlings directly from seeds. However, in the case of TMV, the stability of the virus allowed transmission to seedlings as a result of external contamination during seed manipulation (Broadbent, 1965; Taylor et al., 1961). The increase in PNRSV accumulation observed in the testa from the torpedo and bent cotyledon stages (Fig. 3a, b) could be explained by the existence of viral replication in this tissue or by continuous viral invasion from the maternal tissue following the route of plant assimilates, which are directed towards the fruit during development. In our particular case, we cannot rule out any of these hypotheses, although it is most probable that embryo infection takes place during fertilization, since PNRSV was detected in pollen grains, particularly in the generative cell (Amari et al., 2007), as well as in the embryo sac (Fig. 2e), both of which are responsible for the formation of the zygote and, later on, the embryo. The slightly higher level of accumulation of PNRSV in the embryo compared with the cotyledon (Fig. 3c) suggests that the virus could replicate in the embryo tissue, since RNA integrity was checked. This is not the case with other seed-transmitted viruses such as alfalfa mosaic virus, which only infect the embryo (Bailiss & Offei, 1990). In the case of PSbMV, the virus invades the embryo and partially invades the cotyledons (Wang & Maule, 1997). In our study, PNRSV invaded both the embryo and cotyledons (Figs 3c and 4a, b). It seems that infection of the embryo is a prerequisite for seed transmission, given that other viruses that are able to infect only the cotyledons are not seed-transmitted, as in the case of plum pox virus (Pasquini & Barba, 2006). The seed transmission rate obtained for PNRSV in apricot trees was similar (10 %) to that obtained for other genera of Prunus (Mink, 1992; Bassi & Martelli, 2003). It must be taken into account that the low rate of seed transmission of a given virus is not a good indicator of its ability to invade the embryo, since virus inactivation can occur during seed maturation or storage, which will, in some cases, reduce virus transmission (Bowers & Goodman 1979; Bailiss & Offei, 1990; Johansen et al., 1994). Thus, invasion of the embryo is not sufficient to classify the virus as seedtransmitted, since it must also be able to survive during the maturation and storage of the seed. In our study, we detected viral RNA in immature fruits obtained by hand-pollinating virus-free pistils with infected pollen (Fig. 5a), which demonstrates that PNRSV, like many other seed-transmitted viruses, can be transmitted into a pollinated virus-free plant or, at least, to the fruits in apricot. It has been described that some viruses, including PNRSV in cherry trees, can invade the pollinated plant (Cameron et al., 1973; Davidson, 1976; George & Davidson, 1964; Murant et al., 1974). However, the percentage of this type of transmission, called retroinfection by Mandahar (1985), is very low. This mode of transmission has been questioned, since even if a virus is able to escape from the ovule during or after fertilization, it would not be able to rapidly invade the adjacent maternal tissue to establish a systemic infection in the whole plant. Phloem transport is mostly directed to the developing fruits and the cell-to-cell movement would be too slow for the virus to leave the fruit and invade other parts of the plant (Bennett, 1969). Our observation of the external localization of PNRSV in the fruits (Fig. 5a) suggests an alternative route that this virus can use to escape from pollen and infect the fruit. This might take place during pollen tube germination, since the callose layer separating pollen and pistil is not present in the growing zone of the pollen tip (Keijzer & Willemse, 1989). This callose layer is formed by deposition of vesicular components of the pollen tube cytoplasm (Franklin-Tong, 1999) and we suggest that PNRSV could be transported within these vesicles and so leave the pollen tube. This suggestion is based on our previous work, where we observed a high concentration of PNRSV in the growing zone of germinating apricot pollen tubes (Amari et al., 2007). Taking all our data into account, we suggest that PNRSV is most probably transmitted by apricot seeds through gametes and not indirectly from the testa. ACKNOWLEDGEMENTS We thank Mr P. Thomas for revising the English grammar. This research was supported by grants BIO C02-01, BIO and BIO from the Spanish granting agency DGYCIT and FEDER. REFERENCES Alburquerque, N., Burgos, L. & Egea, J. (2000). Consequences to fertilization of the development stage of apricot ovules at anthesis. J Hortic Sci Biotechnol 75, Amari, K., Burgos, L., Pallás, V. & Sánchez-Pina, M. A. (2007). Prunus necrotic ringspot virus early invasion and its effects on apricot pollen grain performance. Phytopathology 97, Aparicio, F. & Pallás, V. (2002). The molecular variability analysis of the RNA 3 of fifteen isolates of Prunus necrotic ringspot virus sheds light on the minimal requirements for the synthesis of its subgenomic RNA. Virus Genes 25, Aparicio, F., Sánchez-Pina, M. A., Sánchez-Navarro, J. A. & Pallás, V. (1999). Location of Prunus necrotic ringspot ilarvirus within pollen 1772 Journal of General Virology 90

7 Vertical transmission of PNRSV in apricot grains of infected nectarine trees: evidence from RT-PCR, dot-blot and in situ hybridization. Eur J Plant Pathol 105, Aparicio, F., Sánchez-Navarro, J. A., Pallás, V. & Bol, J. (2001). Recognition of cis-acting sequences in RNA 3 Prunus necrotic ringspot virus by the replicase of Alfalfa mosaic virus. J Gen Virol 82, Aparicio, F., Vilar, M., Pérez-Payá, E. & Pallás, V. (2003). The coat protein of Prunus necrotic ringspot virus specifically binds and regulates the conformation of its genomic RNA. Virology 313, Aparicio, F., Sánchez-Navarro, J. A. & Pallás, V. (2006). In vitro and in vivo mapping of the Prunus necrotic ringspot virus coat protein C- terminal dimerization domain by bimolecular fluorescence complementation. J Gen Virol 87, Bailiss, K. W. & Offei, S. K. (1990). Alfalfa mosaic virus in lucerne seed during seed maturation and storage, and in seedlings. Plant Pathol 39, Barba, M., Pasquini, G. & Quacquarelli, A. (1986). Role of seeds in the epidemiology of two almond viruses. Acta Hortic 193, Bassi, D. & Martelli, G. P. (2003). Correlation between breeding techniques and gamic transmisson of viruses in fruit tree crops, viruslike diseases of stone fruits, with particular reference to the Mediterranean region. In Proceedings of the Mediterranean Network on Fruit Tree Virus (MNFTV): activity report. Options Méditerranéennes, Série B, no. 45. Bari: CIHEAM. Bennett, C. W. (1969). Seed transmission of plant viruses. Adv Virus Res 14, Bowers, G. R., Jr & Goodman, R. M. (1979). Soybean mosaic virus: infection of soybean seed parts and seed transmission. Phytopathology 69, Broadbent, L. (1965). The epidemiology of Tomato mosaic virus. XI. Seed-transmission of TMV. Ann Appl Biol 56, Cameron, H. R., Milbrath, J. A. & Tate, L. A. (1973). Pollen transmission of Prunus necrotic ringspot virus in prune and sour cherry orchards. Plant Dis Rep 57, Cole, A., Mink, G. I. & Regev, S. (1982). Location of Prunus necrotic ringspot virus on pollen grains from infected almond and cherry trees. Phytopathology 72, Davidson, T. R. (1976). Field spread of Prunus necrotic ringspot virus in sour cherries in Ontario. Plant Dis Rep 60, Egea, J. & Burgos, L. (1998). Fructification problems in continental apricot cultivars growing under Mediterranean climate. Ovule development at anthesis in two climatic areas. J Hortic Sci Biotechnol 73, Franklin-Tong, V. E. (1999). Signaling and the modulation of pollen tube growth. Plant Cell 11, García-Castillo, S., Marcos, J. F., Pallás, V. & Sánchez-Pina, M. A. (2001). Influence of the plant growing conditions on the translocation routes and systemic infection of carnation mottle virus in Chenopodium quinoa plants. Physiol Mol Plant Pathol 58, George, J. A. & Davidson, T. R. (1964). Further evidence of the pollen transmission of Necrotic ringspot and Sour cherry yellows viruses in sour cherry. Can J Plant Sci 44, Hamilton, R. I., Nichols, C. & Valentine, B. (1984). Survey of Prunus necrotic ringspot virus and other viruses contaminating the exine of pollen collected by bees. Can J Plant Pathol 6, Herranz, M. C. & Pallás, V. (2004). RNA-binding properties and mapping of the RNA-binding domain from the movement protein of Prunus necrotic ringspot virus. J Gen Virol 85, Herranz, M. C., Sánchez-Navarro, J.-A., Sauri, A., Mingarro, I. & Pallás, V. (2005). Mutational analysis of the RNA-binding domain of the Prunus necrotic ringspot virus (PNRSV) movement protein reveals its requirement for cell-to-cell movement. Virology 339, Hull, R. (2002). Matthews Plant Virology, 4th edn. New York: Academic Press. Johansen, E., Edwards, M. C. & Hampton, R. O. (1994). Seed transmission of viruses: current perspectives. Annu Rev Phytopathol 32, Kapil, R. N. & Bhatnagar, A. K. (1981). Ultrastructure and biology of female gametophyte in flowering plants. Int Rev Cytol 70, Keijzer, C. J. & Willemse, M. T. M. (1989). Cell isolation, recognition and fusion during sexual reproduction. In Cell Separation in Plants, pp Edited by D. J. Osborne & M. B. Jackson. Berlin: Springer. Kelley, R. D. & Cameron, H. R. (1986). Location of Prune dwarf and Prunus necrotic ringspot viruses associated with sweet cherry pollen and seed. Phytopathology 76, Kryczynski, S., Szyndel, M. S., Stawiszynska, A. & Piskorek, W. (1992). The rate and the way of Prunus necrotic ringspot virus spread in sour cherry orchard and in the rootstock production. Acta Hortic 309, Mandahar, C. L. (1985). Vertical and horizontal spread of plant viruses through seed and pollen: an epidemiological view. Perspect Plant Virol 1, Más, P. & Pallás, V. (1995). Non-isotopic tissue-printing hybridization: a new technique to study long-distance plant virus movement. J Virol Methods 52, Maule, A. J. & Wang, D. (1996). Seed transmission of plant viruses: a lesson in biological complexity. Trends Microbiol 4, Mink, G. I. (1992). Ilarvirus vectors. In Advances in Disease Vector Research, pp Edited by K. F. Harris. New York: Springer. Mink, G. I. (1993). Pollen- and seed-transmitted viruses and viroids. Annu Rev Phytopathol 31, Murant, A. F., Chambers, J. & Jones, A. T. (1974). Spread of raspberry bushy dwarf virus by pollination, its association with crumbly fruit, and problems of control. Ann Appl Biol 77, Myrta, A., Terlizzi, B. D., Savino, V. & Martelli, G. P. (2002). Sanitary status of the Mediterranean stone fruit industry. Acta Hortic 582, Pallás, V., Más, P. & Sánchez-Navarro, J. A. (1998). Detection of plant RNA viruses by non-isotopic dot-blot hybridization. In Plant Virus Protocols: from Virus Isolation to Transgenic Resistance, pp Edited by G. D. Foster & S. C. Taylor. Totowa, NJ: Humana Press. Pasquini, G. & Barba, M. (2006). The question of seed transmissibility of Plum pox virus. EPPO Bulletin 36, Radchuk, V., Borisjuk, L., Radchuk, R., Steinbiss, H., Rolletschek, H., Broeders, S. & Wobusa, U. (2006). Jekyll encodes a novel protein involved in the sexual reproduction of barley. Plant Cell 18, Reiser, L. & Fischer, R. L. (1993). The ovule and the embryo sac. Plant Cell 5, Roberts, I. M., Wang, D., Thomas, C. L. & Maule, A. J. (2003). Pea seed-borne mosaic virus seed transmission exploits novel symplastic pathways to infect the pea embryo and is, in part, dependent upon chance. Protoplasma 222, Sánchez-Navarro, J. A. & Pallás, V. (1997). Evolutionary relationships in the Ilarviruses: nucleotide sequence of Prunus necrotic ringspot virus RNA 3. Arch Virol 142, Sánchez-Navarro, J. A., Aparicio, F., Herranz, M. C., Minafra, A., Myrta, A. & Pallás, V. (2005). Simultaneous detection and identifica

8 K. Amari and others tion of eight stone fruit viruses by one-step RT-PCR. Eur J Plant Pathol 111, Sánchez-Navarro, J. A., Herranz, M. C. & Pallás, V. (2006). Cell-to-cell movement of Alfalfa mosaic virus can be mediated by the movement proteins of Ilar-, Bromo-, Cucumo-, Tobamo-, and Comoviruses and does not require virion formation. Virology 346, Swamy, B. G. L. & Krishnamurthy, K. V From Flower to Fruit, Embryology of Flowering Plants. New Delhi: McGraw-Hill. Taylor, R. H., Grogan, R. G. & Kimble, K. A. (1961). Transmission of Tobacco mosaic virus in tomato seed. Phytopathology 51, Uyemoto, J. K. & Scott, S. W. (1992). Important diseases of Prunus caused by viruses and other graft-transmissible pathogens in California and South Carolina. Plant Dis 76, Vazquez Rovere, C., Del Vas, M. & Hopp, H. E. (2002). RNAmediated virus resistance. Curr Opin Biotechnol 13, Vértesy, J. (1976). Embryological studies of Ilarvirus infected cherry seeds. Acta Hortic 67, Wang, D. & Maule, A. J. (1997). Contrasting patterns in the spread of two seed-borne viruses in pea embryos. Plant J 11, Journal of General Virology 90