Journal of General Virology

Comparative reactions of recombinant papaya ringspot viruses with chimeric coat protein (CP) genes and wild-type viruses on CP-transgenic papaya

Chu-Hui Chiang,¹ Ju-Jung Wang,¹ Fuh-Jyh Jan,¹ Shyi-Dong Yeh² and Dennis Gonsalves¹

¹ Department of Plant Pathology, Cornell University NYSAES, Geneva, NY 14456, USA
² Department of Plant Pathology, National Chung Hsing University, Taichung 402, Taiwan, ROC

Transgenic papaya cultivars SunUp and Rainbow express the coat protein (CP) gene of the mild mutant of papaya ringspot virus (PRSV) HA. Both cultivars are resistant to PRSV HA and other Hawaii isolates through homology-dependent resistance via post-transcriptional gene silencing. However, Rainbow, which is hemizygous for the CP gene, is susceptible to PRSV isolates from outside Hawaii, while the CP-homozygous SunUp is resistant to most isolates but susceptible to the YK isolate from Taiwan. To investigate the role of CP sequence similarity in overcoming the resistance of Rainbow, PRSV HA recombinants with various CP segments of the YK isolate were constructed and evaluated on Rainbow, SunUp and non-transgenic papaya. Non-transgenic papaya were severely infected by all recombinants, but Rainbow plants developed a variety of symptoms. On Rainbow, a recombinant with the entire CP gene of YK caused severe symptoms, while recombinants with only partial YK CP sequences produced a range of milder symptoms. Interestingly, a recombinant with a YK segment from the 5´ region of the CP gene caused very mild, transient symptoms, whereas recombinants with YK segments from the middle and 3´ parts of the CP gene caused prominent and lasting symptoms. SunUp was resistant to all but two recombinants, which contained the entire CP gene or the central and 3´-end regions of the CP gene and the 3´ non-coding region of YK, and the resulting symptoms were mild. It is concluded that the position of the heterologous sequences in the recombinants influences their pathogenicity on Rainbow.

Pathogen-derived resistance can be mediated by either protein or RNA (Baulcombe, 1996; Beachy, 1993, and accompanying articles; Dougherty & Parks, 1995; Lomonossoff, 1995; Sanford & Johnston, 1985). Lindbo et al. (1993) were the first to show that the resistance of transgenic plants expressing various forms of the coat protein (CP) of a potyvirus is RNA-mediated. They provided evidence that RNA-mediated protection was sequence-specific and thus effective only when the transgene has high similarity to the attacking virus. Numerous other laboratories have confirmed and extended these observations to viruses in other genera (English et al., 1996; Pang et al., 1996, 2000; Prins & Goldbach, 1996). The underlying mechanism of RNA-mediated virus resistance, also referred to as homology-dependent resistance, is post-transcriptional gene silencing (PTGS) (Baulcombe, 1999 a, b; English et al., 1997; Meins, 2000; Wassenegger & Pelissier, 1998).

Various models have been proposed to explain the mechanisms that trigger PTGS and produce virus resistance in transgenic plants that express a transgene that is homologous to the attacking virus. These include an RNA threshold model (Dougherty & Parks, 1995; Smith et al., 1994), an ectopic pairing and aberrant RNA model (Baulcombe, 1996; Baulcombe & English, 1996; English et al., 1996) and a dsRNA-induced PTGS model (Metzlaff et al., 1997; Montgomery & Fire, 1998; Waterhouse et al., 1998). However, all of these models propose a common sequence-specific RNA-degradation process. Briefly, RNA-dependent RNA polymerase synthesizes short antisense RNA from the transgene mRNA and the antisense RNA binds to the complementary regions of the mRNA in the cytoplasm to form RNA duplexes, which are then degraded by dsRNA-specific nucleases (Dalmay et al., 2000 a, b; Mourrain et al., 2000). Viral RNA in the cytoplasm is also a target for degradation. Indeed, several recent papers report the identification of small RNA molecules, 21–25 nt in length, that correspond to sense and antisense pieces of the dsRNA or transgene that is introduced into the cytoplasm (Bass, 2000; Dalmay et al., 2000 a, b; Hamilton & Baulcombe, 1999).

Papaya ringspot virus (PRSV), from the genus Potyvirus, is the major limiting factor for economic papaya production throughout the tropics and subtropics, including the state of Hawaii (Gonsalves, 1998). Two transgenic cultivars, Rainbow and SunUp, that are resistant to PRSV in Hawaii were recently commercialized (Gonsalves, 1998; Manshardt, 1999). SunUp was derived from transgenic papaya line 55-1 (Fitch et al., 1992) and is homozygous for a single insert of the CP gene of PRSV HA 5-1 (Tennant et al., 2001), a mild mutant of PRSV HA (Yeh & Gonsalves, 1984). Rainbow is a hybrid of SunUp and the non-transgenic cultivar 'Kapoho'. It is therefore hemizygous for the CP gene (Manshardt, 1999). Tennant et al. (1994, 2001) reported that Rainbow and hemizygous plants of line 55-1 are resistant to PRSV isolates from Hawaii that share at least 97 % nt identity to the CP transgene but are susceptible to isolates from outside Hawaii that have 89–94 % identity to the transgene. In contrast, SunUp is resistant to a number of isolates from outside Hawaii.

We recently developed infectious transcripts of PRSV HA (Chiang & Yeh, 1997), which provide us with a unique opportunity to produce PRSV HA chimeras that are different from PRSV HA in their CP sequences. Such chimeras can be used to determine the relative importance of CP sequence similarity in breaking the resistance of Rainbow. We constructed a series of such CP recombinants by using whole or partial CP sequences of PRSV YK, a PRSV isolate with 90 % nt identity to PRSV HA in the CP sequence (Wang & Yeh, 1997). Recombinant viruses were able to overcome the resistance of Rainbow but the symptoms varied from very mild to severe, depending on the region of the CP gene that was substituted.

Virus isolates. Two PRSV isolates were used in this study. PRSV HA is a severe strain originally from Hawaii (Gonsalves & Ishii, 1980) and PRSV YK is the most common strain in Taiwan (Wang & Yeh, 1997). The complete nucleotide sequences of HA and YK have been determined (Wang & Yeh, 1997; Yeh et al., 1992). Both genomes are 10326 nt in length, excluding the poly(A) tail.

Transgenic papaya lines. The commercial transgenic papaya SunUp and Rainbow used in this work were originally derived from transgenic line 55-1 (Manshardt, 1999). Line 55-1 was developed by transforming the Hawaiian papaya cultivar 'Sunset' with the CP gene of PRSV HA 5-1 (Fitch et al., 1992), which is a nitrous acid-induced mutant from the parent strain PRSV HA (Yeh & Gonsalves, 1984). Comparison of the 3´-terminal 2235 nt of HA with its mild mutant HA 5-1 showed 99.4 % identity (Wang & Yeh, 1992). Their CP gene sequences differ by 2 nt but their 3´ non-coding region (NCR) sequences are identical. The CP-homozygous SunUp was from the R3 generation and was obtained by crossing R0 transgenic line 55-1 with hermaphroditic 'Sunset' and then self-crossing of progenies (Manshardt, 1999). Rainbow is an F1 derived from a cross of SunUp and non-transgenic cultivar 'Kapoho'. Rainbow and SunUp express the transgene from sequence positions 9257 to 10168 of PRSV HA 5-1, which corresponds to the entire CP gene (Quemada et al., 1990) and 51 nt of the 3´ NCR (Fig. 1 B). Additionally, the 5´ end of PRSV CP contains an extra 48 nt that encode 16 amino acids of cucumber mosaic virus (CMV) CP, and an extra 22 nt of the CMV 3´ NCR is fused to the end of the PRSV 3´ NCR (Ling et al., 1991).

Generation of recombinant viruses between HA and YK. A full-length infectious cDNA clone of PRSV HA, designated pT3-HAG (Chiang & Yeh, 1997), was used to construct different recombinants between PRSV HA and PRSV YK at the 3´ region of the genome.

Clone p3´YKCP, which contains the 1.2 kb 3´ region of PRSV YK, was constructed by using RT–PCR with an upstream primer, 5´ GGCAGGGCCCCATATGTGTCTG 3´, that contains a created ApaI site (underlined) between positions 9053 and 9074 of YK and with an oligo(dT) oligonucleotide that has a 5´-terminal NotI site as a downstream primer. A full-length hybrid virus, designated pHA-3´YK, was obtained by replacing the ApaI–NotI fragment of pT3-HAG with the corresponding region of p3´YKCP (Fig. 1 A, B). Thus, clone HA-3´YK (Fig. 1 B) contains the entire sequence of PRSV HA except that the 3´-proximal 1.2 kb, consisting of 200 nt of the nuclear inclusion b (NIb) gene, the complete CP gene (861 nt) and 209 nt of the 3´ NCR, is from PRSV YK.

Fig. 1. Schematic strategy for the construction of various PRSV HA recombinants with CP gene segments from PRSV YK and a summary of their reactions on transgenic Rainbow papaya. (A) Genetic map of PRSV. ApaI–NotI indicates the region used for replacing the sequence between HA and YK. (B) HA and YK represent the 3´ regions of HA and YK. The restriction enzymes chosen for the replacements (ApaI, SwaI, EcoRI and NotI) are indicated. The numbers between the arrows indicate the distances in nucleotides. Numbers in parenthesis indicate the numbers of nucleotides that were mismatched with the corresponding segment of the transgene. Open rectangles indicate HA sequences and shaded rectangles represent YK sequences. All constructs are aligned to the same scale. Symptom types are those of Rainbow plants that were inoculated at a young (5–6 leaves) stage: Sev, severe; NS, no symptoms; Rec, recovery; I, prominent vein clearing; II, many vein flecks; III, very few vein flecks. See Fig. 3 for pictures of symptoms.

We constructed six other recombinant clones that contained YK sequences in the 5´ region, the central region or the 3´ region of the CP gene (Fig. 1 B). These six full-length chimeric CP constructs were obtained by replacing cDNA fragments with the common restriction enzyme sites (ApaI, SwaI, EcoRI and NotI) between pT3-HAG and pHA-3´YK (Fig. 1 B). Clones YK-AS, YK-SE, YK-EN, YK-AE and YK-SN were constructed by exchanging the ApaI–SwaI, SwaI–EcoRI, EcoRI–NotI, ApaI–EcoRI and SwaI–NotI restriction fragments of pT3-HAG with those from pHA-3´YK. YK-AS/EN was obtained by replacing the SwaI–EcoRI fragment of pHA-3´YK with the corresponding fragment from pT3-HAG. The sequences of the recombinants were verified by digestion with enzymes NdeI, SpeI and SacII to identify replacement regions between HA and YK and by sequencing to confirm the replacement.

Inoculation of papaya. RNA transcripts were synthesized in vitro by T3 RNA polymerase from NotI-linearized plasmids as described by Chiang & Yeh (1997). Capped RNA transcripts were then applied mechanically onto non-transgenic plants of papaya (Carica papaya) with three true leaves. Initially, inocula (1 g leaves in 15 ml buffer) were from papaya infected with the in vitro transcript. Subsequently, non-transgenic papaya and another systemic host, Cucumis metuliferus (Naud.), were also inoculated and used as the source of recombinant virus for subsequent tests. However, only tissues from up to three inoculation transfers were used as inocula. After that, inocula were again obtained from the original papaya that was infected by the in vitro transcripts. Papaya plants were inoculated at a young stage, with 5–6 true leaves, or at an older stage, with 10–12 true leaves. All inoculated plants were kept in a greenhouse at 21–24 °C and observed for symptoms for 90 days.

Virus detection. Total RNA was extracted from papaya leaves as described by Levy et al. (1994). The 3´ region of the viral genome was amplified by RT–PCR with upstream and downstream primers respectively corresponding to PRSV HA positions 8868–8897 and 10083–10117. The RT–PCR-generated DNA fragments were sequenced with an ABI 373 automated sequencer (DNA Sequencing Services, Cornell University, Ithaca, NY, USA).

Northern blot analysis was used to estimate viral RNA accumulation. Ten µg total RNA, extracted 45 days post-inoculation (p.i.) from transgenic Rainbow and non-transgenic papaya, was electrophoresed in a denaturing formaldehyde–1.2 % agarose gel and blotted onto a Gene Screen Plus nylon membrane as described by the manufacturer's manual (DuPont). A ³²P-labelled, random-primed, ApaI/NotI-digested, 1.2 kb DNA fragment from pT3-HAG, which contained 200 bp of NIb, the complete CP gene (861 nt), 209 bp of the NCR and a 36 residue poly(A) sequence from pT3-HAG, was used as a probe. Bark extracts from stems of plants that did not have symptoms on leaves but had water-soak lesions on the stems 4 months after inoculation were assayed by double-antibody sandwich ELISA (Clark & Adams, 1977) with antiserum to intact PRSV HA virus (Ling et al., 1991).

Construction and biological activity of recombinant viruses on non-transgenic plants

PRSV HA was used as the backbone for creating recombinant viruses with PRSV YK sequences because the former causes severe symptoms on non-transgenic papaya and is nearly homologous to PRSV HA 5-1, a mild, nitrous acid-induced mutant of HA. HA 5-1 has only two nucleotide differences from PRSV HA in the CP and none in the 3´ NCR (Wang & Yeh, 1992; Fig. 1). SunUp and Rainbow express the CP gene of PRSV HA 5-1, and initial work showed that these plants are susceptible to YK.

Seven full-length PRSV HA recombinant constructs were generated by replacing segments of the HA genome with corresponding segments from YK (Fig. 1). Since a suitable restriction enzyme site at the 5´ end of the CP gene in PRSV HA was not available, an ApaI site in the NIb gene 200 bp upstream from the CP gene was chosen to perform the DNA replacements between HA and YK. Consequently, recombinant viruses HA-3´YK, YK-AS, YK-AE and YK-AS/EN also contained 200 bp of NIb from YK. A NotI site was created downstream of the poly(A) tail to make constructs HA-3´YK, YK-EN, YK-SN and YK-AS/EN. The NotI restriction site was also used to linearize the plasmids prior to in vitro transcription (Chiang & Yeh, 1997). Thus, recombinants HA-3´YK, YK-EN, YK-SN and YK-AS/EN contained an extra 158 nt of the 3´ NCR sequence from YK compared with the transgene (Fig. 1).

Table 1. Infectivity of tissue extracts from non-transgenic papaya inoculated with PRSV HA, YK or HA-3´YK

Non-transgenic papaya plants at the 5–6 leaf stage were inoculated on the lowest two leaves. Leaf extracts (1:20 dilution) from inoculated papaya were taken 15 days after inoculation and applied to leaves of Chenopodium quinoa. The position of each leaf was designated at the time of inoculation. Numbers of local lesions are means from four inoculated Chenopodium quinoa leaves. Differences in numbers of local lesions were not significant (=0.05). Analyses were done with the SAS general linear models and Tukey's studentized range test.

Comparisons of the YK segments of the recombinant viruses to corresponding regions of the transgene are shown in Fig. 1. The replacement segments of recombinant clone HA-3´YK showed 76 nt differences out of 861 in the PRSV HA 5-1 CP sequence and 8 nt differences out of 51 in the 3´ NCR region. The YK replacement segment of the recombinant YK-AS had the lowest nucleotide sequence identity to the CP transgene (87.5 %; 33 of 263 nt different); this region corresponded to the variable N terminus and part of the core region of the CP (Shukla et al., 1988). The YK segment of recombinant YK-SE had an identity of 92.3 % (32 of 415 nt different); the YK segment originated from the core region of the CP (Shukla et al., 1988). The recombinant YK-EN contained a YK segment that corresponded to the conserved C-terminal and core regions of the CP (94.0 % identity; 11 of 183 nt different) and the first 51 nt of the 3´ NCR (84.3 % identity; 8 of 51 nt different). Recombinants YK-AE, YK-SN and YK-AS/EN contained combinations of two YK segment replacements, as shown in Fig. 1(B).

The biological activity of the recombinants was tested on non-transgenic papaya. Papaya mechanically inoculated with in vitro transcripts corresponding to the recombinants showed symptoms similar to those induced by PRSV HA. Symptoms developed 8–11 days p.i. and consisted of severe mosaic, leaf distortion and stunting of the plants. RT–PCR and sequencing from the inoculated non-transgenic plants verified that the infection was from the proper recombinant viruses (data not shown). Furthermore, these recombinants appeared stable in that they induced similar symptoms in non-transgenic papaya following serial passages for over a year.

Fig. 2. ELISA detection of PRSV HA ( ) and YK ( ) and recombinants YK-AS/EN (x) and YK-AE () in leaves of infected non-transgenic papaya. Samples were collected at periodic intervals from the same three leaves of test plants and tested by ELISA. Sampled leaves were positioned as the top expanded leaf at 8 (A), 12 (B), and 14 (C) days p.i. The results are the mean ELISA readings of four or five plants from two experiments. , Healthy non-transgenic papaya. The readings were taken after 1 h of substrate hydrolysis.

The relative titres of several recombinants (HA-3´YK, HA-AE and HA-AS/EN) in non-transgenic papaya were also compared with those of HA and YK. Two or three selected non-inoculated leaves (top expanded leaf at 8, 12 and 14 or 16 days p.i.) were monitored for virus by ELISA at about 2 day intervals up to 20 days p.i. Additionally, comparisons of local lesion production on Chenopodium quinoa were done with HA-3´YK, HA and YK. ELISA readings of HA-AE and HA-AS/EN were similar to HA and YK over time (Fig. 2 A–C). Virus was first detected by ELISA in the top expanded leaf (Fig. 2 A) at 12–14 days p.i. and detection by ELISA coincided with the appearance of symptoms on the sampled leaves. The virus titre was maximal in all leaves starting 16–18 days p.i. The recombinant HA-3´YK also showed virus titres similar to those of HA and YK (data not shown). Furthermore, leaf extracts of papaya sampled 15 days after inoculation with HA, YK or HA-3´YK induced similar numbers of local lesions (Table 1). In similar tests, ELISA analysis of Cucumis metuliferus inoculated with the isolates showed that these plants also developed similar titres (data not shown). Taken together, these results show that the recombinants HA-3´YK, HA-AE and HA-AS/EN replicate and move in a similar way to HA and YK in non-transgenic papaya and Cucumis metuliferus.

Comparative reactions of SunUp and Rainbow to PRSV HA and YK

Plants of homozygous SunUp and hemizygous Rainbow, which express the CP gene of PRSV HA 5-1, were resistant to PRSV HA when inoculated at the 5–6 (young) or 10–12 (older) true-leaf stages (Table 2; Fig. 3 A). In contrast, transgenic plants challenged with PRSV YK at either young or older developmental stages showed severe mosaic symptoms (Table 2; Fig. 3 B), although symptom expression was delayed compared with non-transgenic plants. Rainbow showed 2–3 day and 4–6 day symptom delays, respectively, at the young and older stages, while SunUp showed 3–5 day and 10–18 day delays, respectively, when challenged at the young and older stages.

Fig. 3. Symptoms of transgenic Rainbow plants inoculated with different viruses. (A) NS: No symptoms, inoculation with HA. (B) Sev: Severe symptoms caused by YK. (C) Sev: Severe symptoms, similar to (B), caused by HA-3´YK. (D) Type I: vein clearing, typically caused by YK-SE, -EN and -SN. (E) Type II: many vein flecks, caused by YK-AE and YK-AS/EN. (F) Type III: few vein flecks, caused by YK-AS. Plants were inoculated at a young (5–6 leaves) stage. Symptoms were recorded at 45 days p.i.

Recombinant viruses induce differential symptoms on Rainbow

Transgenic plants were challenged with PRSV HA-3´YK, which contained the whole viral genome from HA except that the CP gene, 200 nt of NIb and 209 nt of the NCR were from YK. All of the inoculated Rainbow plants became infected and developed severe symptoms when they were challenged at the young stage (Table 2; Fig. 3 C), whereas only 27 % (12/44) of the SunUp plants became infected, and these plants showed mild symptoms consisting of vein flecks (Table 2). Symptom development was delayed by 4–6 days in Rainbow and 14–20 days in SunUp compared with non-transgenic plants. When transgenic plants were challenged at an older stage, 72 % (13/18) of Rainbow plants became infected, showing milder symptoms with some yellow spots on the leaves and less leaf distortion, and symptoms were delayed by 11–22 days. None of the transgenic SunUp plants inoculated at the older stage developed symptoms.

All six recombinant viruses with various segments of YK CP (Fig. 1) induced severe symptoms on Cucumis metuliferus and non-transgenic papaya (Table 3), but variable reactions appeared on Rainbow and SunUp plants. None of the recombinants with only partial CP sequences of YK were able to infect SunUp with the exception of recombinant YK-SN, which infected only 16 % of the inoculated plants and caused very mild symptoms, consisting of a few small yellow spots (Table 3). However, Rainbow plants that were challenged with these recombinant viruses developed a range of symptoms, which were milder than those caused by HA-3´YK (Fig. 3 D–F; Table 3). The symptoms were grouped into three types. Type I symptoms (Fig. 3 D) were characterized by extensive vein clearing on leaves early in the test (45 days p.i.) and leaf distortion at a later stage (90 days p.i.). Type II symptoms (Fig. 3 E) were less severe than type I and consisted of many vein flecks in newly developed leaves (45 days p.i.), with variable symptom expression later on (90 days p.i.). Type III symptoms (Fig. 3 F) consisted of a few vein flecks at 45 days p.i., and the new leaves were symptomless at 90 days p.i.

Table 2. Response of papaya plants inoculated with PRSV isolates HA and YK and hybrid virus HA-3´YK

Symptoms at 45 days p.i. are scored as: NS, no symptoms; Sev, severe symptoms; M, mild symptoms. The frequency gives the number of plants with symptoms/number of plants tested.

* Papaya plants were inoculated at the 5–6 true leaf stage (young) or the 10–12 true leaf stage (old).
† Mild symptoms with vein flecks.
‡ Mild symptoms with few yellow spots.

Recombinant viruses YK-SE, YK-EN and YK-SN caused type I symptoms in 75–95 % of the inoculated Rainbow plants. These recombinants contained the YK fragments at the central and/or the 3´ end of CP and 3´ NCR. Transgenic Rainbow plants inoculated with recombinant virus YK-AS, on the other hand, showed type III symptoms. It will be recalled that recombinant YK-AS contained the full length of the HA sequence except that the 3´ end of NIb (200 nt) and the 5´ end of CP (263 nt) were from YK (Fig. 1). The recombinant viruses YK-AE and YK-AS/EN respectively caused type II symptoms on 70 and 43 % of inoculated Rainbow plants. These recombinants had YK sequences for the first two-thirds of the CP region and for the first and third parts of the CP region, respectively (Fig. 1). Furthermore, plants infected with these recombinants showed variable symptoms at 90 days p.i. For example, 14 of 20 transgenic Rainbow plants challenged with YK-AE showed type II symptoms at 45 days p.i. and 2 of 14 developed type I symptoms at 90 days p.i., while the other 12 plants showed recovery, with no symptoms or type III symptoms on young leaves. Similarly, when YK-AS/EN was inoculated to Rainbow, 9 of 21 inoculated plants showed type II symptoms at 45 days p.i. However, at the later development stage (90 days p.i.), three of these nine Rainbow plants developed type I symptoms, five showed recovery and the other plant remained with type II symptoms.

Northern blot analysis of total RNA from Rainbow plants infected with different recombinants revealed that symptom severity was correlated with viral RNA accumulation (Fig. 4). Rainbow plants infected with YK-SE, YK-EN or YK-SN, which caused type I symptoms, and YK-AE or YK-AS/EN, which caused type II symptoms, had relatively high levels of RNA accumulation (Fig. 4). In contrast, very little or no viral RNA was detected in Rainbow plants infected with YK-AS (type III symptoms). As expected, a large amount of viral RNA was detected in non-transgenic plants infected by PRSV HA, while no viral RNA was detected in Rainbow plants inoculated with HA. The weaker signals in the YK- and HA-3´YK-infected plants (Fig. 4) were probably due to the relatively low similarity of the probe to the YK segment (described in Methods). The probe was derived from PRSV HA, which has 88.8 % sequence identity to YK in the corresponding region (Wang & Yeh, 1997).

Fig. 4. Analysis of viral RNA accumulation in the fourth (from top) leaf from transgenic Rainbow (RB) and non-transgenic Sunrise (SR) plants. Papaya plants were inoculated with HA, a Hawaii PRSV strain; YK, a Taiwan PRSV strain, or the various recombinant viruses. The sample labelled YK-AS1 was taken from the fourth (from top) leaf with very few flecks. Sample YK-AS2 was taken from the second (from top) leaf without symptoms. Fifty ng of in vitro transcripts from pT3-HAG was used as a positive control. Mock, Inoculated with buffer. (A) Northern blot analysis of total plant RNA hybridized with a ³²P-labelled, 1.2 kb DNA specific for the 3´ end of PRSV HA, including 200 nt of NIb, the complete CP and 3´ NCR. The positions of markers (in kb) are shown on the left. (B) Ethidium bromide-stained gel.

Since recombinant YK-AS induced very mild type III symptoms in 70 % (14/20) of inoculated Rainbow plants, we wanted to determine whether higher doses of YK-AS would infect a higher percentage of Rainbow plants and cause more severe symptoms. Crude leaf saps from YK-AS-infected Cucumis metuliferus plants diluted 1:5 and 1:10 were applied onto papaya with 5–6 true leaves. Although 85 % (6/7) and 50 % (3/6) of the Rainbow plants inoculated with the 1:5 and 1:10 dilutions became infected, only type III symptoms were observed.

Some Rainbow plants that initially showed type II symptoms recovered at a later stage, with leaves being symptomless, although the stems still showed water-soak lesions. Virus was apparently still present in the stem tissue, since ELISA analysis of six recovered Rainbow plants gave 2- to 5-fold higher readings than mock-inoculated Rainbow plants (A₄₀₅ of 1.8–1.0 compared with 0.4).

We have shown that Rainbow, a transgenic papaya that expresses the CP gene of the mild mutant of PRSV HA, is resistant to PRSV HA but develops severe symptoms when infected with PRSV YK or with a PRSV HA recombinant containing the full-length CP gene of YK. Furthermore, PRSV HA recombinants with less than full-length CP gene segments of PRSV YK induce severe symptoms on non-transgenic papaya and variable, but on the whole milder, symptoms on Rainbow. Interestingly, an HA recombinant with a YK CP gene segment from the 5´ region, which has the lowest comparative nucleotide sequence identity to the transgene, produced much milder symptoms than recombinants with YK CP gene segments from the middle and 3´ end. Thus, we show for the first time that the virulence of recombinant PRSV on Rainbow is influenced more by the position than by the degree of sequence similarity between the recombinant CP gene and the transgene.

Our conclusion that the virulence of our recombinants on Rainbow is affected more by the position rather than the degree of sequence similarity is based on several observations. The YK-AS recombinant, which has a 263 nt YK segment with 87.5 % sequence identity (33 mismatched nt) to the corresponding region of the transgene, induced very mild type III symptoms on Rainbow, in contrast to the more prominent type I symptoms induced by YK-SE and YK-EN recombinants, which respectively have 92.3 and 91.9 % nt identity to the transgene. Furthermore, the length of the YK replacement segment does not account for the different symptoms induced, since the YK segments in the YK-EN and YK-AS recombinants are similar in length (234 and 263 nt; Fig. 1). Also, the different symptoms that the recombinants produced on Rainbow are apparently not due to their inherent capacity to replicate, as all recombinants produced severe symptoms on non-transgenic plants.

A plausible explanation for the differential virulence of the recombinants on Rainbow is that the PTGS mechanism is preferentially targeted to the middle and 3´ regions of the virus transgene. Thus, recombinant YK-AS, which has 99.9 % identity to the middle and 3´ end of the virus transgene, would be degraded more effectively (and thus produce very mild type III symptoms) than YK-SN, which has only 92.2 % identity to the middle and 3´ end of the transgene. This would also explain the type I symptoms produced by the YK-SE and YK-EN recombinants. Other investigations on homology-dependent virus resistance have shown that the PTGS mechanism is preferentially directed against the 3´ end region of the transgene (e.g. English et al., 1996; Metzlaff et al., 1997; Sijen et al., 1996; Sonoda et al., 1999). On the other hand, reports have shown (i) that the target sites of some transgenic lines are scattered throughout the transgene (Jacobs et al., 1999; Sonoda et al., 1999) and (ii) that the 5´- and 3´-terminal coding regions of the mRNA may be relatively inefficient targets for the silencing machinery (Jacobs et al., 1999).

Table 3. Inoculation of Rainbow and SunUp plants with PRSV HA recombinants containing segments of the CP gene of PRSV YK

Inocula were from papaya plants originally infected with in vitro-capped RNA transcripts or subsequently from serially infected plants. Plants were inoculated at the young (5–6 true leaf) stage. SE, Symptom expression. Symptoms were observed at 45 days p.i. and are scored as: NS, no symptoms; Sev, severe symptoms; I, prominent vein clearing; II, many vein flecks; III, very few vein flecks; M, mild symptoms with few yellow spots. Numbers of plants with symptoms/numbers of plants tested are also given.

The suggested preferential PTGS targeting to the middle and 3´ end of the transgene does not account fully for the intermediate type II symptoms induced on Rainbow by the recombinants YK-AE and YK-AS/EN, which contain YK segments of the 5´ end plus the middle or 3´ end of the CP gene (Table 3; Fig. 1). We would expect these recombinants to produce type I symptoms, since other recombinants with YK segments from the middle or 3´ regions of the CP gene produced type I symptoms. Furthermore, these type II symptom-producing recombinants infected an average of 56 % (23/41) of the inoculated Rainbow plants compared with 89 % (54/61) of the plants inoculated with type I symptom-producing recombinants (Table 2). Taken together, it seems that the presence of the YK 5´ CP segment reduced the virulence of recombinants that would otherwise produce type I symptoms on Rainbow. Furthermore, these recombinants were as virulent as YK or HA on non-transgenic papaya, which rules out the possibility that the differences in symptoms were due the recombinants being inherently less virulent. At present, we have no explanation for this observation. It should be noted, however, that plants with type II symptoms at 45 days p.i. showed variable symptoms (type I or II or recovery; see Fig. 1) at 90 days p.i. In contrast, plants with type I symptoms at 45 days p.i. still had the same symptoms at 90 days p.i. (Fig. 1).

Additionally, the differential virulence of the HA-3´YK recombinant and YK on Rainbow and SunUp is difficult to explain solely by the concept of homology-dependent resistance. Since the genomes of HA-3´YK and YK have identical CP and 3´ NCR sequences, they should have equal virulence on Rainbow and SunUp. Instead, HA-3´YK produced only mild symptoms on older Rainbow and on young SunUp plants, and did not infect older SunUp plants, whereas YK caused severe symptoms on Rainbow and SunUp plants that were inoculated at all stages (see Table 1). The observed differences are not due to HA-3´YK being inherently less virulent than YK, as our results show that HA-3´YK replicates and moves as well as YK and HA in non-transgenic papaya (Table 1). Thus, our results suggest that virus sequences or genes that do not correspond to the transgene may affect the phenotypic reaction of the transgenic plant. Several reports (Anandalakshmi et al., 1998; Brigneti et al., 1998; Kasschau & Carrington, 1998; Voinnet et al., 1999) have shown that HC-Pro of potyviruses can act as a suppressor of PTGS. Thus, if the HC-Pro of YK is more effective than the HC-Pro of HA in suppressing PTGS of infected plants, YK would be expected to replicate better than HA-3´YK in Rainbow and SunUp plants. The HC-Pro of HA and YK share 86.5 and 95.6 % nucleotide and amino acid sequence identity (Wang & Yeh, 1997). The availability of infectious clones of HA (Chiang & Yeh, 1997) and the stability of recombinants that contain segments of HA and YK (this work) will allow us to test experimentally whether HC-Pro contributes to the above observation on Rainbow and SunUp.

We also show here that zygosity and development stage affect the resistance of transgenic plants to recombinants. Our results confirm and extend those of Tennant et al. (2001), who tested Rainbow and SunUp with different PRSV isolates but did not use recombinants. Others have shown similar effects of zygosity and development stage on the resistance of other virus–transgenic plant systems (Goodwin et al., 1996; Jan et al., 2000; Pang et al., 1996), but the experiments were not done with recombinant viruses.

Resistance-breaking strains could conceivably emerge through recombination of PRSV strains from Hawaii with transgenic papaya that express the CP or other genes of PRSV strains that overcome the resistance of Rainbow or SunUp. Thus, the Rainbow–PRSV system is a good model for investigating critically the risk of viruses arising through recombination of PRSV with heterologous PRSV transgenes in papaya.

This work was supported in part by grants from the Biotechnology Risk Assessment Research Grants Program (no. 97-39210-5005) and the Hawaii Papaya Administrative Committee to D. Gonsalves. C.-H. Chiang was partly supported by a fellowship from the Ministry of Education, Taiwan.

Anandalakshmi, R., Pruss, G. J., Ge, X., Marathe, R., Mallory, A. C., Smith, T. H. & Vance, V. B. (1998). A viral suppressor of gene silencing in plants. Proceedings of the National Academy of Sciences, USA 95, 13079–13084.

Bass, B. L. (2000). Double-stranded RNA as a template for gene silencing. Cell 101, 235–238.

Baulcombe, D. C. (1996). Mechanisms of pathogen-derived resistance to viruses in transgenic plants. Plant Cell 8, 1833–1844.

Baulcombe, D. (1999 a). Viruses and gene silencing in plants. Archives of Virology Supplementum 15, 189–201.

Baulcombe, D. C. (1999 b). Fast forward genetics based on virus-induced gene silencing. Current Opinion in Plant Biology 2, 109–113.

Baulcombe, D. C. & English, J. J. (1996). Ectopic pairing of homologous DNA and post-transcriptional gene silencing in transgenic plants. Current Opinion in Biotechnology 7, 173–180.

Beachy, R. N. (1993). Introduction: transgenic resistance to plant viruses. Seminars in Virology 4, 327–328.

Brigneti, G., Voinnet, O., Li, W. X., Ji, L. H., Ding, S. W. & Baulcombe, D. C. (1998). Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana. EMBO Journal 17, 6739–6746.

Chiang, C.-H. & Yeh, S.-D. (1997). Infectivity assays of in vitro and in vivo transcripts of papaya ringspot potyvirus. Botanical Bulletin of Academia Sinica 38, 153–163.

Clark, M. F. & Adams, A. N. (1977). Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. Journal of General Virology 34, 475–483.

Dalmay, T., Hamilton, A., Mueller, E. & Baulcombe, D. C. (2000 a). Potato virus X amplicons in Arabidopsis mediate genetic and epigenetic gene silencing. Plant Cell 12, 369–379.

Dalmay, T., Hamilton, A., Rudd, S., Angell, S. & Baulcombe, D. C. (2000 b). An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not by a virus. Cell 101, 543–553.

Dougherty, W. G. & Parks, T. D. (1995). Transgenes and gene suppression: telling us something new? Current Opinion in Cell Biology 7, 399–405.

English, J. J., Mueller, E. & Baulcombe, D. C. (1996). Suppression of virus accumulation in transgenic plants exhibiting silencing of nuclear genes. Plant Cell 8, 179–188.

English, J. J., Davenport, G. F., Elmayan, T., Vaucheret, H. & Baulcombe, D. C. (1997). Requirement of sense transcription for homology-dependent virus resistance and trans-inactivation. Plant Journal 12, 597–603.

Fitch, M. M. M., Manshardt, R. M., Gonsalves, D., Slightom, J. L. & Sanford, J. C. (1992). Virus resistant papaya derived from tissues bombarded with the coat protein gene of papaya ringspot virus. Bio/Technology 10, 1466–1472.

Gonsalves, D. (1998). Control of papaya ringspot virus in papaya: a case study. Annual Review of Phytopathology 36, 415–437.

Gonsalves, D. & Ishii, M. (1980). Purification and serology of papaya ringspot virus. Phytopathology 70, 1028–1032.

Goodwin, J., Chapman, K., Swaney, S., Parks, T. D., Wernsman, E. A. & Dougherty, W. G. (1996). Genetic and biochemical dissection of transgenic RNA-mediated virus resistance. Plant Cell 8, 95–105.

Hamilton, A. J. & Baulcombe, D. C. (1999). A species of small antisense RNA in posttranscriptional gene silencing in plants. Science 286, 950–952.

Jacobs, J. J. M. R., Sanders, M., Bots, M., Andriessen, M., van Eldik, G. J., Litiere, K., Van Montagu, M. & Cornelissen, M. (1999). Sequences throughout the basic -1,3-glucanase mRNA coding region are targets for homology dependent post-transcriptional gene silencing. Plant Journal 20, 143–152.

Jan, F.-J., Pang, S.-Z., Tricoli, D. M. & Gonsalves, D. (2000). Evidence that resistance in squash mosaic comovirus coat protein-transgenic plants is affected by plant developmental stage and enhanced by combination of transgenes from different lines. Journal of General Virology 81, 2299–2306.

Kasschau, K. D. & Carrington, J. C. (1998). A counterdefensive strategy of plant viruses: suppression of posttranscriptional gene silencing. Cell 95, 461–470.

Levy, L., Lee, I. M. & Hadidi, A. (1994). Simple and rapid preparation of infected plant tissue extracts for PCR amplification of virus, viroid, and MLO nucleic acids. Journal of Virological Methods 49, 295–304.

Lindbo, J. A., Silva-Rosales, L., Proebsting, W. M. & Dougherty, W. G. (1993). Induction of a highly specific antiviral state in transgenic plants: implications for regulation of gene expression and virus resistance. Plant Cell 5, 1749–1759.

Ling, K., Namba, S., Gonsalves, C., Slightom, J. L. & Gonsalves, D. (1991). Protection against detrimental effects of potyvirus infection in transgenic tobacco plants expressing the papaya ringspot virus coat protein gene. Bio/Technology 9, 752–758.

Lomonossoff, G. P. (1995). Pathogen-derived resistance to plant viruses. Annual Review of Phytopathology 33, 323–343.

Manshardt, R. M. (1999). 'UH Rainbow' papaya. University of Hawaii College of Tropical Agriculture and Human Resources, New Plants for Hawaii-1. Honolulu: University of Hawaii College of Tropical Agriculture and Human Resources.

Meins, F., Jr (2000). RNA degradation and models for post-transcriptional gene silencing. Plant Molecular Biology 43, 261–273.

Metzlaff, M., O'Dell, M., Cluster, P. D. & Flavell, R. B. (1997). RNA-mediated RNA degradation and chalcone synthase A silencing in petunia. Cell 88, 845–854.

Montgomery, M. K. & Fire, A. (1998). Double-stranded RNA as a mediator in sequence-specific genetic silencing and co-suppression. Trends in Genetics 14, 255–258.

Mourrain, P., Beclin, C., Elmayan, T., Feuerbach, F., Godon, C., Morel, J. B., Jouette, D., Lacombe, A. M., Nikic, S., Picault, N., Remoue, K., Sanial, M., Vo, T. A. & Vaucheret, H. (2000). Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101, 533–542.

Pang, S. Z., Jan, F. J., Carney, K., Stout, J., Tricoli, D. M., Quemada, H. D. & Gonsalves, D. (1996). Post-transcriptional transgene silencing and consequent tospovirus resistance in transgenic lettuce are affected by transgene dosage and plant development. Plant Journal 9, 899–909.

Pang, S. Z., Jan, F. J., Tricoli, D. M., Russell, P. F., Carney, K. J., Hu, J. S., Fuchs, M., Quemada, H. D. & Gonsalves, D. (2000). Resistance to squash mosaic comovirus in transgenic squash plants expressing its coat protein genes. Molecular Breeding 6, 87–93.

Prins, M. & Goldbach, R. (1996). RNA-mediated virus resistance in transgenic plants. Archives of Virology 141, 2259–2276.

Quemada, H., L'Hostis, B., Gonsalves, D., Reardon, I. M., Heinrikson, R., Hiebert, E. L., Sieu, L. C. & Slightom, J. L. (1990). The nucleotide sequences of the 3´-terminal regions of papaya ringspot virus strains W and P. Journal of General Virology 71, 203–210.

Sanford, J. C. & Johnston, S. A. (1985). The concept of parasite-derived resistance – deriving resistance genes from the parasite's own genome. Journal of Theoretical Biology 113, 395–405.

Shukla, D. D., Strike, P. M., Tracy, S. L., Gough, K. H. & Ward, C. W. (1988). The N and C termini of the coat proteins of potyviruses are surface-located and the N terminus contains the major virus-specific epitopes. Journal of General Virology 69, 1497–1508.

Sijen, T., Wellink, J., Hiriart, J.-B. & van Kammen, A. (1996). RNA-mediated virus resistance: role of repeated transgenes and delineation to targeted regions. Plant Cell 8, 2277–2294.

Smith, H. A., Swaney, S. L., Parks, T. D., Wernsman, E. A. & Dougherty, W. G. (1994). Transgenic plant virus resistance mediated by untranslatable sense RNAs: expression, regulation, and fate of nonessential RNAs. Plant Cell 6, 1441–1453.

Sonoda, S., Mori, M. & Nishiguchi, M. (1999). Homology-dependent virus resistance in transgenic plants with the coat protein gene of sweet potato feathery mottle potyvirus: target specificity and transgene methylation. Phytopathology 89, 385–391.

Tennant, P. F., Gonsalves, C., Ling, K. S., Fitch, M., Manshardt, R., Slightom, J. L. & Gonsalves, D. (1994). Differential protection against papaya ringspot virus isolates in coat protein gene transgenic papaya and classically cross-protected papaya. Phytopathology 84, 1359–1366.

Tennant, P., Fermin, G., Fitch, M. M., Manshardt, R. M., Slightom, J. L. & Gonsalves, D. (2001). Papaya ringspot virus resistance of transgenic Rainbow and SunUp is affected by gene dosage, plant development, and coat protein homology. European Journal of Plant Pathology (in press).

Voinnet, O., Pinto, Y. M. & Baulcombe, D. C. (1999). Suppression of gene silencing: a general strategy used by diverse DNA and RNA viruses of plants. Proceedings of the National Academy of Sciences, USA 96, 14147–14152.

Wang, C. H. & Yeh, S. D. (1992). Nucleotide sequence comparison of the 3´-terminal regions of severe, mild, and non-papaya infecting strains of papaya ringspot virus. Archives of Virology 127, 345–354.

Wang, C.-H. & Yeh, S.-D. (1997). Divergence and conservation of the genomic RNAs of Taiwan and Hawaii strains of papaya ringspot potyvirus. Archives of Virology 142, 271–285.

Wassenegger, M. & Pelissier, T. (1998). A model for RNA-mediated gene silencing in higher plants. Plant Molecular Biology 37, 349–362.

Waterhouse, P. M., Graham, M. W. & Wang, M. B. (1998). Virus resistance and gene silencing in plants can be induced by simultaneous expression of sense and antisense RNA. Proceedings of the National Academy of Sciences, USA 95, 13959–13964.

Yeh, S. D. & Gonsalves, D. (1984). Evaluation of induced mutants of papaya ringspot virus for control by cross protection. Phytopathology 74, 1086–1091.

Yeh, S.-D., Jan, F.-J., Chiang, C.-H., Doong, T.-J., Chen, M.-C., Chung, P.-H. & Bau, H.-J. (1992). Complete nucleotide sequence and genetic organization of papaya ringspot virus RNA. Journal of General Virology 73, 2531–2541.

JGV Direct table of contents

© 2001 SGM

This article is now available in the November 2001 print issue of JGV (vol. 82, 2827–2836). The complete issue of the journal may be seen in electronic form on JGV Online.