Localization of Zucchini yellow mosaic virus to the veinal regions and role of viral coat protein in veinal chlorosis conditioned by the zym potyvirus resistance locus in cucumber

Physiological and Molecular Plant Pathology (2002) 60, 79±89 doi:10.1006/pmpp.2002.0379, available online at http://www.idealibrary.com on

Localization of Zucchini yellow mosaic virus to the veinal regions and role of viral coat protein in veinal chlorosis conditioned by the zym potyvirus resistance locus in cucumber Z A K I R UL L A H * and R E B E C CA G R U M E T { Department of Horticulture and Program in Plant Breeding and Genetics, Michigan State University, East Lansing, MI 48824, U.S.A. (Accepted for publication 17 January 2002) The zym locus in cucumber (Cucumis sativus L.) is marked by multiple alleles conferring resistance to the potyvirus, Zucchini yellow mosaic virus (ZYMV). Cotyledon inoculation of greenhouse grown plants expressing the zymDina allele results in a distinct pattern of veinal chlorosis limited to a single systemic leaf; leaf-inoculated plants remain free of symptoms. Inoculation of cotyledons of plants of di€erent growth stages indicated that the virus moved to the leaf that was newly emerging at the time of cotyledon infection where it then remained localized to the veinal regions for up to 30 days post-inoculation (d.p.i). Cotyledon removal experiments indicated that the inability of ZYMV to spread within the uninoculated leaf, or to infect inoculated leaves, was not due to the inability to replicate in leaves. Chimeric viruses generated from a strain of ZYMV that did not induce the veinal chlorosis response, indicated that the coat protein amino terminus, which has been shown to be involved in long distance movement, a€ected occurrence of veinal chlorosis. In the growth chamber, inoculation of the ®rst leaf, but not subsequent leaves, gave a high percentage of plants expressing veinal chlorosis. Together, these observations suggest that resistance conferred by zymDina is developmentally regulated and occurs at the level of systemic movement. c 2002 Elsevier Science Ltd. * Keywords: Cucumis sativus L., potyvirus; virus resistance; systemic movement.

INTRODUCTION Potyviridae form the largest and most economically important family of plant viruses [44]. Although almost all major crops are susceptible to one or more members of this family, naturally occurring sources of potyvirus resistance showing a range of resistance mechanisms have been identi®ed in numerous plant species [36, 38]. Examples of dominant alleles conferring hypersensitive responses include the Nv and Rysto alleles conferring resistance to Potato virus V [5], Tobacco etch virus (TEV) and Potato virus Y (PVY) [22] in potato; the I allele, which confers resistance to at least nine potyviruses in bean [26, 43]; and the RTM1 and RTM2 loci, which cooperate to restrict long distance movement of TEV in Arabidopsis [28, 48]. RTM1 encodes a protein with * Present address: Department of Biochemistry, Michigan State University, U.S.A. { To whom all correspondence should be addressed. E-mail: [email protected] Abbreviations used in text: MWMV, Moroccan watermelon mosaic virus; PVA, potato virus A; PVY, potato virus Y; TEV, tobacco etch virus; WMV, watermelon mosaic virus; ZYMV, Zucchini yellow mosaic virus.

0885-5765/02/020079+11 $35.00/00

similarity to a group of plant proteins that are implicated in defence against viruses, fungi and insects, while RTM 2 encodes a heat shock related protein [7, 47]. Numerous recessive alleles also have been described that interfere with various stages in the viral life cycle, possibly implicating an altered interaction between a host and viral factor. sbm-1 of pea [25], et a of Capsicum annum [8], and pvr1 of C. chinense [30], interfere with viral replication or RNA accumulation. Cell-to-cell movement is restricted for PVY by the ya gene in pepper plants [4], and for PVY and Tobacco vein mottling virus by the va gene in tobacco [15, 29]. The raadg allele provides strong resistance against vascular movement of Potato virus A (PVA) [21], and long distance movement of TEV is restricted in the resistant tobacco line V20, at the level of entry into or exit from the sieve elements [41]. In a few cases, speci®c potyviral factors have been shown to play a role in stimulating or overcoming host resistance, and can give insight into resistance mechanisms and/or viral gene function. For example, the VPg coding region of potyviruses is involved in overcoming host resistances operating at the level of replication [25] and movement [29, 31, 37, 42]. A single amino acid (aa) c 2002 Elsevier Science Ltd. *

80

Z. Ullah and R. Grumet

substitution (Ser47 to Pro) in the amino terminus (NT) of the coat protein (CP) of the NY isolate of Pea seed borne mosaic virus, enabled the virus to move systemically in Chenopodium quinoa rather than be restricted to necrotic local lesions in the inoculated leaf [2]. Studies showed that substitution of the CP-NT of Zucchini yellow mosaic virus (ZYMV) with the CP-NT from a potyvirus, that does not normally infect cucurbits (TEV), resulted in initial systemic infection followed by a recovery response in the upper leaves [20]. In cucumbers, two sources of naturally occurring resistance to ZYMV have been identi®ed [1, 24, 34, 35]. Although each is due to a single recessive allele occurring at the same locus, they di€er in phenotypic response to cotyledon inoculation with ZYMV [24]. Plants possessing the resistance allele from ``TMG-1'' (zymTMG), an inbred line derived from the Chinese hybrid ``Taichung Mou Gua'', remain symptom-free. Plants with the resistance allele from ``Dina-1'' (zymDina), an inbred line derived from the Dutch hybrid ``Dina'', show a distinct veinal chlorosis that is limited to one leaf; the remainder of the plant remains symptom-free. Unlike the response to cotyledon inoculation, veinal chlorosis is not generally observed in response to leaf inoculation. Segregation analyses indicated that in the absence of the dominant susceptible allele, Zym, the zymDina resistance allele, which allows for limited systemic infection, is dominant to the resistance allele from ``TMG-1'' zymTMG [24]. Both ``TMG-1'' and ``Dina-1'' plants also are resistant to the potyviruses, Watermelon mosaic virus (WMV), the watermelon strain of Papaya ringspot virus (PRSV-W) and Moroccan watermelon mosaic virus (MWMV) [23, 34, 35]. In both lines, ZYMV resistance completely cosegregated (within 1 cM) with the other potyvirus resistances, suggesting either a gene cluster or a single gene conferring multiple potyvirus resistance [19, 23, 24, 45, 46]. A recent mapping study by Park et al. [33] also showed a tight association (2.2 cM) between the resistances to ZYMV and PRSV-W. Although the veinal chlorosis phenotype has been observed with a number of ZYMV isolates of diverse origin, it is unique to infection with ZYMV; it did not occur in response to inoculation with WMV, MWMV, or PRSV [24]. In this study the veinal chlorosis phenotype conferred by zymDina was utilized to gain insight into the mechanism of resistance conferred by the zym allele. Evidence is provided that was utilized the veinal chlorosis phenotype of ``Dina-1'' is associated with the accumulation and distribution of is provided virus in the symptomatic leaves and that resistance operates at the level of virus movement. It is also shown that the ZYMV CP-NT is associated with the veinal chlorosis phenotype of ``Dina-1''.

MATERIALS AND METHODS

Plant material, virus stock and inoculation The resistant cucumber inbred lines used in this study, ``TMG-1'' and ``Dina-1'', were initially obtained from Dr J. Staub (USDA, University of Wisconsin, Madison) and Dr K. Owens (Seminis Peto Seed Company, Woodland, Calfornia), respectively, and subsequently maintained by self pollination in the greenhouse. The susceptible cultivar was ``Straight-8'' (W. Atlee Burpee and Company, Warminster, PA, U.S.A.). All seeds were germinated on moist ®lter paper for 24 hr at 308C before sowing in Baccto soil mix (Michigan Peat Company, Houston, TX, U.S.A.) in 15 cm clay pots. During winter months (October±March), greenhouse grown plants received supplemental lighting (16 h day). Growth chambers were set for a 16 hr day (248C day/208C night) with an average light intensity of 150 mE. Randomized complete block designs were used for experiments when applicable. Each experiment was performed at least three times. The number of plants used for each treatment is included in the ®gure and table legends. Unless otherwise indicated, experiments were performed using the Connecticut (Ct) isolate of ZYMV (ZYMV-Ct) [18, 35]. Maintenance of virus stock and rub-inoculation procedures were as described in Wai and Grumet [46]. Unless di€erent ages of inoculation were chosen for developmental studies (as indicated in the text and legends), the standard age for inoculation was 7±10 day old plants (cotyledons expanded and ®rst leaf just emerging). Plasmid DNA of the ZYMV NAA/Ct CP hybrid infectious constructs (described below) was prepared using the Wizard mini prep kit (Promega, Madison, WI, U.S.A.) and directly inoculated onto cotyledons of 7±10 day old ZYMV-susceptible squash (Cucurbita pepo L., cv. Midas; Willhite Seeds Inc., Poolville, TX, U.S.A.) plants using the particle bombardment procedure of Gal-On et al. [14]. Infected leaf material ( passage 0, P0) was stored at 808C and used to inoculate squash plants (P1) which were then used as a source of inoculum for all subsequent experiments.

ELISA and immunoblot analyses Polyclonal antibodies prepared to ZYMV-Ct coat protein were used for both ELISA and immunoblot analyses. ELISA was performed with leaf discs using the procedure of Wai and Grumet [46]. Leaves harvested for immunoblot analyses were frozen at 808C, thawed at room temperature and used to form a sandwich consisting of ®lter paper, nitrocellulose membrane (Protran, PH79 pore size 0.1 mm) pre-wet with distilled water, the leaf, and a single layer of paper towel. The sandwich was placed in a plastic sample bag (Nasco WHIRL-PAK

Localization of ZYMV to veinal regions in cucumber 18oz) and rolled through a pasta press (Atlas, Italy) using a pressure setting 4 which resulted in high reproducibility and minimal spread from the tissues [as evidenced by the clear delineation where samples were removed for ELISA analysis Fig. 1(g)]. Immediately after the run, the membrane was removed from the sandwich, placed on a piece of paper towel and allowed to dry at room temperature. The blotted membrane was developed using alkaline phosphatase conjugated secondary antibodies and 5-bromo-4-chloro-3 indolyl phosphate (BCIP, Sigma) and nitro blue tetrazolium (NBT, Sigma) reagents according to the Western blot protocol of Blake et al. [6].

Chimeric ZYMV infectious constructs The full length infectious clone of ZYMV-NAA was kindly provided by Drs A. Gal-On and B. Raccah (Volcani Center, Bet Dagan, Israel) [12±14]. To facilitate substi-

81

tutions in the CP coding region, a PstI site was introduced at the NIb-CP junction (Q/S; nucleotide position 8540 in the full length ZYMV) by PCR ampli®cation (using Vent DNA polymerase with proof reading activity; NE Biolabs Inc. Beverly, MA, U.S.A.) of the CP coding region and the 30 NTR using the NIb-CP junction primer RG81 (50 ATGCTGCAGTCAGGCACCAGCCA-3; restriction site is underlined) and 30 NTR primer RG40 (50 AGTGAATTCTCGAGCTTATTCGTGA-30 ). The RG40 primer included an engineered EcoRI and a natural XhoI site with a single base overlap. The ampli®ed fragment was digested with PstI and XhoI and cloned into PstI± XhoI digested pBlueScript KS (Stratagene, La Jolla, CA, U.S.A.) to produce pZPX. To clone the 1 kb 30 fragment of the NIb coding region, the forward primer (RG69: 50 GGTCATACGTATGATGTG-30 ) priming at nucleotide position 7280 in the ZYMV genome and the reverse primer (RG80 50 -TGACTGCAGCATTACAGTGCTCC30 ) introducing a PstI site at the NIb-CP junction were

F I G . 1. (a)±(c). Response of ``TMG-1'', ``Dina-1'', and ``Straight 8'' to inoculation with ZYMV. Cotyledons were inoculated when the plants were 7 days old, pictures were taken 12 d.p.i. ``TMG-1'' remained symptom-free, ``Dina-1'' showed a distinct pattern of veinal chlorosis limited to the ®rst leaf, while ``Straight 8'' showed systemic mosaic throughout the plant. (d)±( f). Immunoblots of the ®rst leaves of ``TMG-1'', ``Dina-1'', and ``Straight 8'' at 10 d.p.i. Note that where samples had been punched from the leaves for ELISA assay prior to harvesting, sharp edges of the holes were observed after blotting [( f) arrow].

82

Z. Ullah and R. Grumet

used. The PCR product was cloned into pZPX digested with SacI (a unique restriction site in the NIb; nucleotide position 7510) and PstI resulting in the construct pZCPSPX. pZCPSPX contained the 2 kb, 30 fragment of the infectious ZYMV-NAA construct with an engineered PstI site at the NIb-CP junction. The engineered PstI site resulted in substitution of two bases (CTCCAA to CTGCAG), but did not change the amino acid sequence. This fragment was then substituted into the full length ZYMV-NAA clone to produce pZNAA-P (Fig. 6), and veri®ed to be infectious on susceptible squash and cucumber plants. The full length CP coding region of ZYMV-Ct was introduced into pZCPSPX by using primer pair RG81 and RG56 (50 -TAACCTAGGTAGGCGACC-30 priming 15 bp downstream from the CP stop codon at the naturally occurring AvrII site; nucleotide position 9400) and resulted in the construct pCtCPSPX. The CP-NT of the infectious ZYMV-NAA construct was substituted with the CP-NT of ZYMV-Ct [20] by cloning the SacI± MluI fragment from pCtCPSPX into the infectious full length pZNNA-P construct digested with the same enzymes. MluI is a unique restriction site near the beginning of the core portion of the CP ( position 8746). The core and carboxy terminal portions of the CP were exchanged between the two constructs using the restriction sites of MluI and AvrII.

Cotyledon removal experiments To test for the ability of the virus to replicate in the leaves, cotyledons of 7 day old cucumber plants were inoculated; on successive days after inoculation (1±5 d.p.i.), leaf disc samples were removed from the cotyledons and ®rst leaf (when present) of each of ®ve plants to sample for ELISA. The discs from a single plant were pooled, each plant was regarded as a replicate. The cotyledons were then removed from the stem with a razor blade. Two weeks post-inoculation (w.p.i.), the ®rst and second leaves of all the plants were again sampled for ELISA. For immunoblot experiments, 7 day old ``Dina-1'' plants were inoculated on the cotyledons with ZYMV-Ct. Both the cotyledons and one half of the ®rst leaf were removed from ®ve plants at 3, 4, and 5 d.p.i. and stored at 808C. Ten d.p.i (after the appearance of the veinal chlorosis symptoms) the intact halves of the ®rst leaves were harvested for immunoblotting as described earlier. Both halves of each leaf were immunoblotted and probed at the same time. Healthy ``Dina-1'' plants were used as negative controls for each experiment, while positive controls included plants from which cotyledons were not removed after inoculation.

RESULTS

Symptom expression and virus accumulation in the resistant genotypes ``Dina-1'' and ``TMG-1'' When cotyledons of 7±9 day old plants ( ®rst leaf not yet, or barely, visible) were inoculated with ZYMV-Ct, the susceptible genotype, ``Straight-8'', responded with systemic mosaic symptoms and measurable virus accumulation (Figs 1 and 2). ``TMG-1'' plants remained symptom-free and at 2 w.p.i. did not contain detectable amounts of virus accumulation in cotyledons or leaves. ``Dina-1'' plants, however, showed a distinct veinal chlorosis limited to a single leaf, usually leaf 1, or occasionally leaf 2 (Fig. 1(b)). High virus titre was observed in the cotyledons and the chlorotic leaf of the ``Dina-1'' plants, but not in the upper, non-symptomatic leaves (Fig. 2). Similarly, only 1 of 36 ``Dina-1'' plants (less than 3 %) contained ZYMV-Ct in leaf 4 or 5 when tested by inoculation onto susceptible squash plants (Table 1). Four to six w.p.i., ``Straight-8'' plants had systemic mosaic symptoms throughout the plant. At that time, an additional one or two ``Dina-1'' leaves often showed detectable levels of virus accumulation and occasionally mild veinal chlorosis symptoms. ``TMG-1'' plants also showed virus accumulation in cotyledons and sometimes in the ®rst uninoculated leaf, indicating that the resistance conferred by both alleles greatly reduces, but does not completely prevent, systemic virus infection. On rare occasions, detectable levels of virus were observed in ``TMG-1'' cotyledons at two w.p.i. (data not shown). When cotyledons of slightly older ``Dina-1'' plants were inoculated (10±12 days old; leaf 1 unfolded, leaf 2 not yet, or barely visible), veinal chlorosis occurred on leaf 2 or leaf 3. The mean leaf position showing veinal chlorosis was 2.43 (n ˆ 16) for plants inoculated when 10±12 days old vs. mean leaf position of 1.15 (n ˆ 20) for plants inoculated when 7±9 days old. Thus veinal chlorosis developed on the leaf that was newly emerging at the time of cotyledon infection. In analogous experiments with ``Straight-8'', when cotyledons of di€erent ages (and so plants at di€erent stages of development) were inoculated (Fig. 3), symptoms and virus accumulation were ®rst detected at progressively higher leaf positions (i.e. in leaves that were newly expanding at the time of inoculation). When cotyledons of 7 day old plants were inoculated, the virus ®rst peaked in leaf 1; when 10 or 16 day old cotyledons were inoculated, the peak moved to leaf 2 and leaf 3, respectively. Together these observations suggest that in both ``Dina-1'' and ``Straight-8'' plants, the virus moved from inoculated cotyledons to the predominant sink leaf at the time. The distribution of virus within the ®rst leaf of ``TMG1'', ``Dina-1'' and ``Straight-8'' plants was examined by

Localization of ZYMV to veinal regions in cucumber

83

appeared to be primarily present along the veins. Nonsymptomatic leaves of ``Dina-1'' plants did not contain detectable amounts of virus (data not shown). A more uniform distribution of ZYMV-Ct occurred throughout the ®rst uninoculated leaf of ``Straight-8'' plants. These results suggest that the veinal chlorosis pattern in ``Dina1'' plants corresponds to virus distribution within the leaf, and that the symptomless phenotype of ``TMG-1'' is associated with the absence of detectable virus accumulation. If sampled at an earlier time (e.g. 4 d.p.i.) (Fig. 4), ``Straight-8'' leaves also showed a predominantly veinal pattern of virus distribution in the ®rst uninoculated leaf, but at 10 d.p.i. there was a more uniform distribution throughout the leaf, indicating that the virus was able to exit from the vascular tissue and move cell-to-cell. ``Dina-1'' plants, on the other hand, showed very little or no virus in the ®rst uninoculated leaf at 4 d.p.i. and a predominantly veinal distribution at 10 d.p.i. that continued up to 30 d.p.i. Unlike the response to cotyledon inoculation, when the ®rst or second leaves of greenhouse grown plants were inoculated, ``Dina-1'' plants generally remained symptom-free and did not contain detectable amounts of ZYMV-Ct (Fig. 2). Occasionally very mild veinal chlorosis symptoms were observed on leaf 3 of ``Dina-1'' plants when leaf 1 was inoculated (12 %; 8/64), whereas no symptoms were observed when leaf 2 was inoculated (0/30). Similarly, ``Dina-1'' plants inoculated with ZYMV-Ct on leaf 4 or 5 did not show veinal chlorosis, and only 2/20 plants (10 %) tested had the virus in leaf 7 or 8 based on inoculation to susceptible squash plants (Table 1). Leaf inoculation of ``Straight-8'' plants resulted in typical systemic mosaic and virus accumulation; ``TMG-1'' plants remained symptom- and virusfree (Fig. 2).

Nature of the block to successful virus infection in ``Dina-1''

F I G . 2. Virus titre in cotyledons (Cot) and leaves (L) in ``Straight 8'' (a), ``Dina-1'' (b), and ``TMG-1'' (c) in response to inoculation with ZYMV-Ct. Young seedlings were inoculated on the cotyledons (Cot inoc, Q), ®rst (L1, R) or second (L2, W) leaf. Virus titre was measured by ELISA 14 d.p.i. The x-axis shows the number of leaves from the base to the top of the plant and the y-axis shows absorbance at 405 nm. Each point represents the mean of four replications. The experiment was repeated three times with similar results. Control (W).

immunoblotting with antiserum to the ZYMV-Ct CP [Fig. 1(d)±( f)]. At 10 days post-inoculation (d.p.i.), no virus accumulation was detected in the ®rst leaf of ``TMG-1'' plants, whereas in ``Dina-1'' plants, the virus

Limitation of veinal chlorosis to a single leaf following cotyledon inoculation in ``Dina-1'' plants, localization of virus along the veins in the uninoculated leaf, and failure to see symptoms or virus accumulation with leaf inoculation, suggest either a block in replication or movement in the leaves. To test for the ability of the virus to replicate in the leaves, the inoculated cotyledons were removed at various time intervals after inoculation, and the plants scored for veinal chlorosis and virus accumulation with ELISA and immunoblotting. The virus was not detectable in the leaf discs removed from the cotyledons or leaves at 2 or 3 d.p.i. [Fig. 5(a) and (c)]. However, at 14 d.p.i., veinal chlorosis and high virus titre was observed in leaf 1 (or some cases on leaf 2) [Fig. 5(b)], even in those plants from which the cotyledons were removed at 2 or 3 d.p.i. This indicated

84

Z. Ullah and R. Grumet

T A B L E 1. Comparison of cotyledon and leaf inoculation of greenhouse grown ``Dina-1'' plants with ZYMV-Ct and ZYMV-NAA Treatment

Cotyledon inoculation L4/L5 inoculation

Presence of virus as determined by inoculation of squasha

Veinal chlorosis Z-Ct

Z-NAA

Leaves testedb

Z-Ct

Z-NAA

35/36c 0/20

0/12 0/40

L4 or L5 L7 or L8

1/36 2/20

Ð 32/40

Data are pooled from three experiments. a Half of the leaf from each ``Dina-1'' plant was used to inoculate two squash plants; symptoms were monitored for 30 days. b Leaves were sampled 2 w.p.i., all were non-symptomatic. c Number of plants showing symptoms/total number of plants inoculated.

F I G . 3. Virus accumulation (as determined by ELISA) in di€erent leaves of ``Straight 8'' plants as a€ected by plant age at the time of cotyledon inoculation. Each point represents the mean absorbance (405 nm) of ®ve replications. The experiment was repeated twice with similar results.

that virus replication had occurred in the absence of cotyledons. Similarly, when the cotyledons and half of leaf 1 were removed at 3 d.p.i., no virus was visible in leaf 1 at that time [Fig. 5(c)]. When the other half of leaf 1 was removed 7 days later (at the time of veinal chlorosis symptoms), detectable levels of the virus were observed [Fig. 5(d)]. Together these results suggest that virus replication in ``Dina-1'' plants occurred in the absence of cotyledons, and that the block to ZYMV infection of ``Dina-1'' plants occurred at the level of movement in the leaves. While the possibility that the block in movement is at the cell-to-cell level (e.g. initial unloading from the vascular tissue in the one systemically infected leaf, followed by an inability to move cell-to-cell) cannot be ruled out, immunoblot analysis showed accumulation of virus along the veins in the inoculated leaves of ``Dina-1''

F I G . 4. (a) Time course experiments showing the accumulation of ZYMV in the ®rst systemically infected leaf of ``Straight 8'' (S) and ``Dina-1'' (D) leaves at 4, 10 and 30 d.p.i. Plants were inoculated on the cotyledons when 7 days old. (b). Inoculated ®rst leaf immunoblotted at two w.p.i.

Localization of ZYMV to veinal regions in cucumber

85

F I G . 5. (a, b). Virus titre in the cotyledons and leaves of cotyledon-inoculated ``Dina-1'' plants at the time of removal of cotyledons (a) and in leaves of the same plants at 14 d.p.i. (b). The x-axis shows the d.p.i. at which cotyledons were removed; the y-axis shows absorbance at 405 nm. Data were normalized by subtraction of background ELISA readings. Cotyledon (L), leaf 1 (E) and leaf 2 (K) from the base of the plant. Each column represents the mean of ®ve replicate plants. (c, d). Immunoblots showing virus replication in the leaves of ``Dina-1'' plants. Plants were inoculated on the cotyledons with ZYMV-Ct when 7 days old; the cotyledons and one half of the ®rst leaf were removed 3 days later (c). The other half was removed at 10 d.p.i. when veinal chlorosis symptoms appeared (d). Both halves of the leaf were immunoblotted at the same time with ZYMV-CP antiserum. Both experiments were repeated three times with similar results.

plants [Fig. 4(b)]. This suggests cell-to-cell movement from primary infection foci to the vascular tissue. A possible e€ect on movement was further examined using infectious ZYMV clones encoding coat proteins chimeric for ZYMV strains that do, and do not, cause veinal chlorosis. Although most isolates of ZYMV tested on ``Dina-1'' plants caused veinal chlorosis [24], the NAA isolate did not cause veinal chlorosis or systemic symptoms. It did, however, show limited systemic movement as detected by ELISA (data not shown), or by inoculation of upper, non-inoculated leaves onto susceptible squash plants; 32/40 of the ``Dina-1'' inoculated plants (72 %) showed virus accumulation in non-inoculated leaves as detected by inoculation onto squash plants

(vs. less than 3 % of the plants that were inoculated with ZYMV-Ct) (Table 1). Chimeric ZYMV-NAA infectious constructs were made containing either the CP-NT (ZCtNT), or the core and carboxy terminal portions of the CP (ZCtCore), of ZYMV-Ct (Fig. 6). The conserved core region potyviral CPs are involved in cell-to-cell movement, while the surface exposed amino and carboxy termini are essential for systemic movement [10, 11, 40, 44]. Both constructs were fully infectious and produced systemic symptoms on susceptible squash and cucumber plants. When tested on the resistant genotype ``Dina-1'', the chimeric construct containing the CP-NT of ZYMV-Ct showed veinal chlorosis on either leaf 1 or 2, of approx.

86

Z. Ullah and R. Grumet

F I G . 6. Infectious ZYMV-NAA constructs derived from ZYMV-NAA. The complete ZYMV genome is shown at the top. pZNAAP: ZYMV-NAA with an engineered Pst1 site at the NIb-CP junction; pZCtNT and pZCtCore are chimeric viruses containing the NT or core and carboxy terminal portion of the CP of the ZYMV-Ct isolate, respectively. The veinal chlorosis response of ``Dina-1'' to these constructs is indicated at the right. Hatched areas denote ZYMV-Ct substitution into the ZYMV-NAA infectious clone. Data are compiled from three experiments.

60 % (25/43) of the inoculated plants, while ZCtCoreand ZYMV-NAA-inoculated plants remained largely symptomless. These results suggest that the CP-NT of ZYMV plays a role in the veinal chlorosis response of ``Dina-1'' plants. ZYMV-Ct and ZYMV-NAA di€er at two amino acid positions in the (Ser7 to Ala and Gly32 to Ser) in the N-terminal region.

Nature of leaf speci®c expression of the resistance response in Dina-1 Collectively the results described above indicated replication in, and movement from the cotyledons, followed by restricted movement of the virus after reaching the leaf. Successful infection of the cotyledons, but inhibition in the leaf, might be due to inherent structural di€erences between leaves and cotyledons or developmental changes in the host. Environmental conditions appeared to in¯uence the appearance of the veinal chlorosis on ``Dina-1'' plants following leaf inoculation. When experiments were performed in the growth chamber instead of the greenhouse, 81 % of the plants (13/16) showed veinal chlorosis following inoculation of leaf 1. Inoculation of leaf 2, however, resulted in a greatly reduced number of plants showing veinal chlorosis (19 %; 3/16 plants). These observations argue against inherent structural di€erences in the leaves preventing virus spread.

Together the di€erences between cotyledon and leaf inoculation in the greenhouse, and leaf 1 and leaf 2 inoculation in the growth chamber, suggest developmentally controlled expression of resistance.

DISCUSSION Resistance to ZYMV infection in ``Dina-1'' plants is controlled by a single recessive allele (zymDina) that confers an unusual phenotype. When cotyledons of 7±9 day old seedlings, whose ®rst leaf was not yet or just barely visible, were inoculated with ZYMV-Ct, a distinct veinal chlorosis pattern developed on the ®rst or occasionally second leaf. Subsequent leaves remained symptom-free, and direct inoculation of leaves generally did not result in veinal chlorosis or systemic infection. Immunoblot analyses revealed that the pattern of veinal chlorosis re¯ected the pattern of virus distribution within the leaf. In leaves expressing veinal chlorosis, virus accumulation was localized primarily along the veins even at 30 d.p.i.; non-symptomatic leaves did not have detectable virus accumulation. These observations may be contrasted with the resistance observed in ``TMG-1'' whose resistance is conferred by another allele at the same locus [24]. ``TMG-1'' plants remain symptom-free and there is very little or no detectable virus accumulation. In susceptible

Localization of ZYMV to veinal regions in cucumber ``Straight 8'' plants, virus was initially associated with the veins (at 4 d.p.i.), but by 10 d.p.i. the virus was more uniformly distributed throughout the leaf. The age of the plant at the time of cotyledon inoculation in¯uenced which leaf showed veinal chlorosis; for 10±12 day old plants, veinal chlorosis appeared predominantly in the second, instead of the ®rst leaf. Analogous patterns, showing a relationship between plant age at the time of cotyledon inoculation and ®rst leaf to accumulate virus, were observed with the ``Straight 8'' plants, suggesting that in both ``Straight 8'' and ``Dina-1'', the virus replicated in the inoculated cotyledons, and as is typical for virus infection [3, 27], was then transported to a leaf that was rapidly expanding at the time of infection. Interestingly, the observed pattern of veinal chlorosis and virus accumulation closely resembled the pattern observed for phloem unloading of carboxy¯orescein molecules in sink leaves [39]. Chlorosis was predominantly observed along class II and III veins, which are the main sites of phloem unloading [32, 39]. The pattern of virus distribution of ZYMV in susceptible ``Straight 8'', as determined by tissue prints, showed initial localization to veinal regions followed by spread throughout the leaf, analogous to that observed for Potato virus X-GFP in susceptible Nicotiana benthamiana hosts [39]. In several cases, evidence of the source±sink transition leading to a failure to unload in the more rapidly maturing apical portion of leaf was observed, at the level of both symptom development and virus distribution [e.g., Fig. 4(a)]. Failure to establish systemic infection beyond a single leaf of cotyledon-inoculated plants, and localization of virus to the veinal regions, suggested that the virus was either unable to replicate or spread in the systemically infected leaf. Cotyledon removal experiments showed that even when cotyledons were removed prior to the presence of detectable virus in leaves or cotyledons, measurable amounts of virus were observed in the leaves 1 week later, indicating that the block was not due to an inability to replicate in leaves. Although the possibility of a block in the cell-to-cell movement in the leaves following initial unloading from the vascular tissue cannot be eliminated, the pattern of veinal chlorosis and localization of virus to the veinal regions in uninoculated leaves of cotyledoninoculated plants suggested inhibition of long distance movement, possibly a block in exit from the vascular tissue. A block in virus unloading from vascular bundles has been observed with Brome mosaic virus in barley [9]. Similarly, the failure, in most cases, of leaf-inoculation to result in veinal chlorosis or movement to non-inoculated leaves, but accumulation of virus along the veins in the inoculated leaf, implies inhibition of long distance movement. Consistent with this observation, Schaad and Carrington [41] observed that in the resistant tobacco line V20, which interferes with long distance movement

87

of TEV, virus accumulation in the inoculated leaves appeared as an apparent tracking along the primary veins. Further evidence for a block in long distance movement in ``Dina-1'' leaves comes from the use of chimeric ZYMV viruses which substitute portions of the CP from the ZYMV-Ct isolate into the infectious clone of the ZYMV-NAA isolate. The conserved core portion of the potyvirus CP is involved in the encapsidation of viral RNA and cell-to-cell movement of the virus, whereas the variable surface exposed N- and C-terminal portions are required for systemic movement [10, 11, 40, 44]. The ZYMV-NAA isolate does not produce veinal chlorosis on ``Dina-1'' plants and is able to escape partially the block to systemic movement as detected by inoculation of upper leaves onto susceptible squash plants. When the ZYMVCt CP-NT was substituted with the respective region of ZYMV-NAA, the chimeric virus induced veinal chlorosis on ``Dina-1'' plants. This suggests that the CP-NT is involved in the di€erential response of ``Dina-1'' to these two isolates of ZYMV, and further suggests that the block to successful infection in ``Dina-1'' is occurring at the level of systemic movement. The observation of resistance in leaves but not cotyledons of ``Dina-1'' plants was intriguing. Since on some occasions, inoculation of leaf 1 resulted in mild chlorosis on leaf 3 in the greenhouse, and in the growth chamber, a majority of the plants showed veinal chlorosis following inoculation of the ®rst leaf; structural di€erences in the leaves vs. cotyledons are unlikely to account for the di€erential response to inoculation. On the other hand, since a much lower percentage (19 %) of the growth chamber grown plants showed symptoms when the second leaf was inoculated, developmental regulation of the zymDina resistance is implicated. This may be contrasted with the resistance conferred by the alternate zymTMG allele at the same locus, where resistance is already expressed in the cotyledon. Developmental di€erences in the expression of resistance have been observed in other cucurbit±potyvirus interactions [17, 18, 46]. For example when cotyledons of muskmelon (Cucumis melo) cultivar ``Cinbo'' were inoculated with PRSV-W, mild systemic symptoms were observed, but plants remained symptomless when leaves were inoculated [17]. Limited systemic movement of WMV and ZYMV from the cotyledon to the ®rst one or two leaves also has been observed in Cucurbita moschata line Menina 15 [18]. In ``TMG-1'', two independently assorting resistance factors to WMV were identi®ed, one expressed in both cotyledons and leaves and the other expressed only in the leaves [46]. It will be of interest to determine whether similar mechanisms or resistance gene products are involved in each of these examples. In summary, the results presented above indicate that the zymDina allele encodes a developmentally regulated

88

Z. Ullah and R. Grumet

resistance factor that restricts systemic ZYMV infection. The di€erences in developmental expression of resistance in ``Dina-1'', and possible involvement of the ZYMV CPNT, may provide approaches to allow for cloning and further characterization of the zym locus. The authors thank Drs Amit Gal-On and Benny Raccah for generously providing the infectious clone of ZYMVNAA, Dr John Everard for assistance in developing the tissue print procedure, and Drs Wayne Loescher and Richard Allison for critical reviews of the manuscript. This research was in part supported by BARD grant US2666-95 and the Michigan Agricultural Experiment Station. Z. Ullah also was a Quaid-e-Azam Scholar sponsored by the Ministry of Education, Government of Pakistan.

11.

12.

13.

14.

15. 16.

REFERENCES 1. Abul-Hayja Z, Al-Shahwan IM. 1991. Inheritance of resistance to zucchini yellow mosaic virus in cucumber. Journal of Plant Disease and Protection 98: 301±304. 2. Andersen K, Johansen IE. 1998. A single conserved amino acid in the coat protein of pea seed-borne mosaic potyvirus modulates the ability of the virus to move systemically in Chenopodium quinoa. Virology 241: 304±311. 3. Andrianifahanana M, Lovins K, Dute R, Sikora E, Murphy JF. 1997. Pathway for phloem dependent movement of pepper mottle potyvirus in the stem of Capsicum annuum. Phytopathology 87: 892±899. 4. Arryo R, Soto MJ, Martinez-Zapater JM, Ponz F. 1996. Impaired cell-to-cell movement of potato virus Y in pepper plants carrying the ya ( pr21) resistance gene. Molecular Plant±Microbe Interactions 9: 314±318. 5. Barker H. 1997. Extreme resistance to potato virus V in clones of Solanum tuberosum that are also resistant to potato viruses Y and A: evidence for a locus conferring broadspectrum potyvirus resistance. Theoretical and Applied Genetics 95: 1258±1262. 6. Blake MS, Johnson KH, Russel-Jones GT, Gotschlich EC. 1984. A rapid, sensitive method for detection of alkaline phosphatase-conjugated antibody on western blots. Analytical Biochemistry 136: 175±179. 7. Chisholm ST, Mahajan SK, Witham SA, Yamamoto ML, Carrington JC. 2000. Cloning of the Arabidopsis RTM1 gene, which controls restriction of long distance movement of tobacco etch virus. Proceedings of the National Academy of Sciences, USA 97: 489±494. 8. Deom CM, Murphy JF, Paguio OR. 1997. Resistance to tobacco etch virus in Capsicum annum: inhibition of virus RNA accumulation. Molecular Plant±Microbe Interactions 10: 917±921. 9. Ding XS, Flasinski S, Nelson RS. 1999. Infection of barley by brome mosaic virus is restricted predominantly to cells in and associated with veins through a temperature dependent mechanism. Molecular Plant±Microbe Interactions 12: 615±623. 10. Dolja VV, Haldeman-Cahill R, Montgomery AE, Vandenbosch KA, Carrington JC. 1995. Capsid protein determinants involved in the cell-to-cell and long distance

17.

18.

19.

20.

21.

22.

23.

24. 25.

26.

movement of tobacco etch potyvirus. Virology 206: 1007±1016. Dolja VV, Haldeman-Cahill R, Robertson NL, Dougherty WG, Carrington JC. 1994. Distinct functions of capsid protein in assembly and movement of tobacco etch potyvirus in plants. EMBO Journal 13: 1482±1491. Gal-On A, Antigus Y, Rosner A, Raccah B. 1991. Infectious in vitro RNA transcripts derived from cloned cDNA of the cucurbit potyvirus, zucchini yellow mosaic virus. Journal of General Virology 72: 2639±2642. Gal-On A, Antigus Y, Rosner A, Raccah B. 1992. A zucchini yellow mosaic virus coat protein gene mutation restores aphid transmissibility but has no e€ect on multiplication. Journal of General Virology 73: 2183±2187. Gal-On A, Meiri E, Huet H, Wu JA, Raccah B, Gaba V. 1995. Particle bombardment drastically increases the infectivity of cloned DNA of zucchini yellow mosaic potyvirus. Journal of General Virology 76: 3223±3227. Gibb KS, Hellmann GM, Pirone TP. 1989. Nature of resistance of a tobacco cultivar to tobacco vein mottling virus. Molecular Plant±Microbe Interactions 2: 332±339. Gibb KS, Padovan AC, Herrington ME. 1994. Nature of resistance in Cucumis melo cultivars to papaya ring spot virus type W. Australian Journal of Agricultural Research 45: 633±645. Gilbert-Albertini F, Lecoq H, Pitrat M, Nicolet JL. 1993. Resistance of Cucurbita moschata to watermelon mosaic virus type 2 and its genetic relation to zucchini yellow mosaic virus. Euphytica 69: 231±237. Grumet R, Fang G. 1990. cDNA cloning and sequence analysis of the 30 -terminal region of zucchini yellow mosaic virus RNA. Journal of General Virology 71: 1619±1622. Grumet R, Kabelka E, McQueen S, Wai T, Humphrey R. 2000. Characterization of sources of resistance to the watermelon strain of papaya ringspot virus in cucumber: allelism and co-segregation with other potyvirus resistances. Theoretical and Applied Genetics 101: 463±472. Grumet R, Ullah Z, Gal-On A, Raccah B. 2000. Cucurbit±potyvirus interactions in¯uencing virus movement and accumulation. In: Katzir N, Parris H, eds. Proceedings of Eucarpia 2000, Acta Horticulturae 510: 327±334. Hamalainen JH, Kekarainen T, Gebhardt C, Watanabe KN, Valkonen JPT. 2000. Recessive and dominant genes interfere with the vascular transport of potato virus A in diploid potatoes. Molecular Plant±Microbe Interactions 13: 402±412. Hinrichs J, Berger S, Shaw JG. 1998. A hypersensitive response like mechanism is involved in resistance of potato plants bearing the RYsto gene to the potyviruses potato virus Y and tobacco etch virus. Journal of General Virology 79: 167±176. Kabelka E, Grumet R. 1997. Inheritance of resistance to Moroccan watermelon mosaic virus in the cucumber line TMG-1 and its relationship to zucchini yellow mosaic virus resistance. Euphytica 95: 237±242. Kabelka E, Ullah Z, Grumet R. 1997. Multiple alleles for zucchini yellow mosaic virus resistance at the zym locus in cucumber. Theoretical and Applied Genetics 95: 997±1004. Keller KE, Johansen IE, Martin RR, Hampton RO. 1998. Potyvirus genome-linked protein (VPg) determines pea seed-borne mosaic virus pathotype-speci®c virulence in Pisum sativum. Molecular Plant±Microbe Interactions 11: 124±130. Kyle M, Provvidentii R. 1993. Genetics of broad spectrum viral resistance in bean and pea. In: Kyle MM, ed.

Localization of ZYMV to veinal regions in cucumber 27. 28.

29.

30.

31.

32. 33.

34. 35. 36.

Resistance to Viral Diseases of Vegetables: Genetics and Breeding. Portland, OR, U.S.A.: Timber Press, 153±166. Leisner SM, Turgeon R. 1993. Movement of virus and photoasimilate in the phloem: a comparative analysis. BioEssays 15: 741±748. Mahajan SK, Chisholm ST, Witham SA, Carrington JC. 1998. Identi®cation and characterization of a locus (RTM1) that restricts long distance movement of tobacco etch virus in Arabidopsis thaliana. Plant Journal 14: 177±186. Masuta C, Nishimura M, Morishita H, Hataya T. 1999. A single amino acid change in viral genome-associated protein of potato virus Y correlates with resistance breaking in Virgin A Mutant tobacco. Phytopathology 89: 118±123. Murphy JF, Blauth JR, Livingstone KD, Lackney VK, Jahn MK. 1998. Genetic mapping of the pvr locus in Capsicum spp, and evidence that distinct potyvirus resistance loci control responses that di€er at the whole plant and cellular level. Molecular Plant±Microbe Interactions 11: 943±951. Nicolas O, Dunnington SW, Gotow LF, Pirone TP, Hellmann GM. 1997. Variation in the VPg protein allows a potyvirus to overcome va gene resistance in tobacco. Virology 237: 452±459. Oparka KJ, Turgeon R. 1999. Sieve elements and companion cells-trac control centers of the phloem. Plant Cell 11: 739±750. Park YH, Sensoy S, Wye C, Antonise R, Peleman J, Havey MJ. 2000. A genetic map of cucumber composed of RAPDs, RFLPs, AFLPs, and loci conditioning resistance to papaya ringspot and zucchini yellow mosaic viruses. Genome 43: 1003±1010. Provvidenti R. 1985. Sources of resistance to viruses in two accessions of Cucumis sativus. Cucurbit Genetics Cooperative Reports 8: 12. Provvidenti R. 1987. Inheritance of resistance to a strain of zucchini yellow mosaic virus in cucumber. HortScience 22: 102±103. Provvidenti R, Hampton RO. 1992. Sources of resistance to viruses in Potyviridae. Archives of Virology 5: 189±211.

89

37. Rajamaki MI, Valkonen JPT. 1999. The 6K2 protein and the VPg of potato virus A are determinants of systemic infection in Nicandra physaloides. Molecular Plant±Microbe Interactions 12: 1074±1081. 38. Revers F, Gall OL, Candresse T, Maule AJ. 1999. New advances in understanding the molecular biology of plant potyvirus interactions. Molecular Plant±Microbe Interactions 12: 367±376. 39. Roberts AG, Santa Cruz S, Roberts IM, Prior DAM, Turgeon R, Oparka KJ. 1997. Phloem unloading in sink leaves of Nicotiana benthamiana: comparison of a ¯uorescent solute with a ¯uorescent virus. Plant Cell 9: 1381±1396. 40. Rojas MR, Zerbini FM, Allison RF, Gilbertson RL, Lucas WJ. 1997. Capsid protein and helper componentproteinases function as potyvirus cell to cell movement proteins. Virology 237: 283±295. 41. Schaad MC, Carrington JC. 1996. Suppression of longdistance movement of tobacco etch virus in a nonsusceptible host. Journal of Virology 70: 2556±2561. 42. Schaad MC, Lellis AD, Carrington JC. 1997. VPg of tobacco etch potyvirus is a host genotype-speci®c determinant for long distance movement. Journal of Virology 71: 8624±8631. 43. Scully B, Provvidenti R, Benscher D, Halseth DE, Miller JC, Wallace DH. 1995. Five multiple-virusresistant common bean breeding lines. HortScience 30: 1320±1323. 44. Shukla DD, Ward CW, Brunt AA. 1994. The Potyviridae. Wallingford, U.K.: CAB International. 45. Wai T, Grumet R. 1995a. Inheritance of resistance to watermelon strain of papaya ringspot virus in the cucumber line TMG-1. HortScience 30: 338±340. 46. Wai T, Grumet R. 1995b. Inheritance of resistance to watermelon mosaic virus in the cucumber line TMG-1: tissue-speci®c expression and relationship to zucchini yellow mosaic virus resistance. Theoretical and Applied Genetics 91: 699±706. 47. Whitham SA, Anderberg RJ, Chisholm ST, Carrington JC. 2000. Arabidopsis RTM2 gene is necessary for speci®c restriction of tobacco etch virus and encodes an unusual small heat shock like protein. Plant Cell 12: 569±582.