Journal
of General Virology (1996), 77, 2847-2855. Printedin Great Britain
The satellite RNAs associated with the groundnut rosette
disease complex and pea enation mosaic virus: sequence
similarities and ability of each other's helper virus to support
their replication
S. A. Demler, 1 D. G. Rucker, 1 G. A. de Zoeten, 1 A. Ziegler, 2 D. J. Robinson z and A. F. M u r a n t 2
1Department of Botany and Plant Pathology, Michigan State University, East Lansing, HI 48824-1 31 2, USA
2Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, UK
Pea enation mosaic virus (PEMV) and the causal
agents of groundnut rosette disease are diverse
examples of disease complexes involving two RNA
species, one of which is related to the genomes of
luteoviruses and the other to those of umbraviruses.
In both complexes, these viral RNA components may
be supplemented with satellite RNAs that are dependent on the umbravirus component for replication and systemic movement, and on the luteovirus component for encapsidation and vector transmission. Sequence analysis identified regions of
similarity between the satellites of groundnut rosette virus (GRV) and PEMV, particularly at the 5'
and 3' termini and around duplicate sequence
repeats present in each satellite RNA. The umbravirus GRV and the umbravirus-like PEMV RNA-2
were each able to support the replication and
systemic spread of homologous and heterologous
Introduction
The groundnut rosette disease complex and pea enation
mosaic virus (PEMV) are each examples of associations
involving two viral RNAs, one of which is related to the
genomes of luteoviruses and the other to those of members of
the newly established genus Umbravirus (Falk & Duff-us, I981;
Murant, I993; de Zoeten & Demler, I995 ; Murant et al., I995;
Demler et al., i996). The groundnut rosette disease agents
comprise groundnut rosette assistor luteovirus (GRAV) and
groundnut rosette umbravirus (GRV) (Hull & Adams, 1968;
Reddy et al., I985a, b). GRV is a coat protein-deficient
Authorfor correspondence:D. J. Robinson.
Fax +44 1382 562426. e-mail [email protected]
0001-4110 © 1996 SGM
satellites. The presence of the PEMV satellite in
infections with GRV had no effect on symptom
expression in Nicotiono spp. or in Arochis hypogoeo.
Likewise, in Pisum sotivum, the GRV satellite had no
effect on the symptoms induced by PEMV. However,
the intense yellow blotch symptoms induced in
Nicotiono benthomiono by the YB3 GRV satellite in
conjunction with GRV were also manifested when
PEMV was the helper. Although PEMV RNA-1 was
capable of supporting the encapsidation and aphid
transmission of the GRV satellite, no evidence was
obtained that the essential role of the GRV satellite
in the aphid transmission of GRV could be supplied
by the PEMV satellite. These data further strengthen
the hypothesis of an evolutionary relationship between PEMV and the luteovirus-umbravirus complexes.
(Taliansky et al., 1996) yet independently infective RNA virus,
which depends on GRAV for encapsidation (A. F. Murant,
unpublished data) and aphid transmission (Hull & Adams,
1968). The genome of PEMV, the sole member of the
Enamovirus genus, is composed of a Iuteovirus-like RNA
(designated RNA-I) and an RNA (designated RNA-2) that is
umbravirus-like (Demler & de Zoeten, I991; Demler et aI.,
1993, 1994a; Taliansky et al., I996). PEMV RNA-2 resembles
GRV in being coat protein-deficient but independently infective; it depends on PEMV RNA-I for encapsidation and aphid
transmission. However, in contrast to GRAV, PEMV RNA-I
has not been shown to infect plants on its own (though it
infects intact protoplasts) and depends on the presence of
PEMV RNA-2 to establish a fully systemic infection.
Another similarity between the groundnut rosette complex
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
!843
and PEMV is the occurrence of satellite-like RNAs (Murant el
al., 1988; Demler & de Zoeten, 1989; Demler et aI., 1994 b). In
both systems, the satellite R N A is dependent on the umbravirus or umbravirus-like component for support of replication,
and on the luteovirus or luteovirus-like component for
encapsidation and vector transmission. Despite this similarity,
the biological activities of the GRV and PEMV satellites are
different. The approx. 900 nucleotide (nt) satellite R N A of
GRV, although not essential for the multiplication of GRV in
plants, is invariably present in natural infections and is
responsible for the symptoms of rosette disease in groundnut,
variants of the satellite being responsible for the differences in
symptoms between GRV isolates (Murant et al., 1988; Murant
& Kumar, 1990; Kumar et at., 1991). Moreover, the presence of
the GRV satellite in source plants, together with that of
GRAV, is essential for the aphid transmission of GRV (Murant,
1990). Thus, although the GRV-associated satellite fulfils many
of the criteria of satellite RNAs (Murant & M a y o , 1982), it
appears to be evolving towards becoming an essential genomic
RNA. In contrast, the 717 nt satellite R N A of PEMV has no
detectable effect on symptoms induced b y PEMV in its natural
host, Pisum sativum, although it does attenuate the symptoms
induced in Nicotiana benthamiana (Demler & de Zoeten, 1989;
Demler et al., 1994 b). Moreover, the presence or absence of the
PEMV satellite has no effect on the aphid transmission of either
PEMV genome component; thus the PEMV satellite is fully
dispensable, like a classical satellite.
In an effort to understand the role in viral pathogenesis and
vector transmission of these satellites, as well as their potential
evolutionary significance, we have compared their nucleotide
sequences and studied the ability of the heterologous helper
systems to support their replication and aphid transmission.
Methods
• Virus and satellite cultures. Experiments with GRV, GRAV and
the PEMV satellite RNA were done at the Scottish Crop Research
Institute. The GRAV isolate was the Nigerian isolate as in Rajeshwari et
al. (1987). The GRV isolate was a satellite-free culture, MC1, derived
experimentally by eliminating the satellite from MC, an isolate from
Malawian groundnut plants with the chlorotic form of rosette (Murant &
Kumar, 1990). Both virus isolates were maintained as previously
described. A satellite RNA (MC3a) derived from GRV isolate MC was
produced from plasmid pMC3a (see below) and the satellite RNA of
PEMV was produced by in vitro transcription from plasmid pPER3
(Demler et al., 1994 b). All virus and satellite cultures were imported and
maintained under licences issued by the Scottish Office Agriculture and
Fisheries Department.
Experiments with PEMV and the GRV satellite RNA were done at
Michigan State University. A satellite (YB3b) derived from the YB isolate
of GRV (Kumar et al., 1991) was produced by in vitro transcription from
a cDNA done (pYB3b, see below) imported and maintained under permit
no. 932946 issued by the USDA Animal and Plant Health Inspection
Service. The genomic RNAs of PEMV were produced in vitro from the
transcription vectors pPER1 (RNA-1) and pPER2d (RNA-2), originally
derived from the aphid non-transmissible WSG isolate of PFMV (Demler
!84~
et al., 1993, 1994a). RNA-1 transcripts of a cDNA clone derived from an
aphid-transmissible isolate of PEMV (designated pPERIAT + ) were also
used to study the effects of the GRV satellite RNA on the aphid
transmission of PEMV.
• Sequence analysis. The sequences of ten individual satellite RNAs
of GRV and the PEMV satellite are accessible through the GenEMBL
database (Blok et al., 1994; Demler et al., 1994b; accession numbers
Z29702-Z29711 and U03564, respectively). Nucleotide sequence alignments and comparisons of potential open reading frame (ORF) products
were done by using the GAP program of the UWGCG software package
(version 7.2; Devereux et al., 1984).
• Construction of pYB3b and pMC3a. ClonespYB3b and pMC3a
were derived from clones yb3b and mc3a, respectively, which contain
cDNA copies of GRV satellites from isolates YB and MC in plasmid
pGEM-T (Blok et al., 1994). The oligonucleotide primer 5' TACGACTCACTATAGGGTTTCAATAGGA 3' and a U-DNA mutagenesis kit
(Boehringer) were used to remove vector-derived sequences between the
T7 transcription start site and the 5' end of the satellite sequences, as
described by Kunkel (1985). At the 3' end of the satellite sequences, nine
vector-derived nucleotides remained before the SpeI site, which was used
to linearize the plasmid prior to transcription, but these did not prevent
biological activity of the transcripts.
• Production of RNA transcripts. Capped in vitro transcripts of the
genomic and satellite RNAs of PEMV were produced as described
previously (Demler et at., 1994 b). Capped, biologicallyactive RNA of the
GRV satellites YB3b and MC3a was produced similarly by T7 RNA
polymerase transcription from plasmids pYB3b and pMC3a, respectively.
All transcripts were evaluated for concentration and integrity by agarose
gel electrophoresis prior to use.
• Infectivity tests. Support of the replication of the GRV satellite
YB3b by the genomic RNA species of PEMV was evaluated in protoplasts
of P. sativum cv. 8221. Protoplasts were prepared and inoculated as
described by Demler et al. (1993). Tests in whole plants (P. sativum, N.
benthamiana and Arachis hypogaea cv. TMV-2) were done at 22 °C with
a 16 h photoperiod in isolated growth chambers. Seedlings were
inoculated as described by Demler et al. (1994a).
To evaluate support of the replication of the PEMV satellite by GRV,
RNA transcribed from plasmid pPER3 was mixed with total leaf nucleic
acid from GRV-MCl-infected N. benthamiana in a buffer containing
5 mM-Tris--HCl pH 8"0, 25 mM-NaCl and 0"6 mg/mi bentonite, and
inoculated into N. benthamiana. Control treatments contained either no
satellite RNA or GRV satellite MC3a RNA transcribed from plasmid
pMC3a. Subsequent propagation of isolates produced in this way was
done by inoculation of sap to N. benthamiana, N. clevelandii or N.
occidentalis. Isolates were transmitted to A. hypogaea by inoculation with
total leaf nucleic acid from infected IV. benthamiana, and subsequently
maintained by graft inoculation.
• Aphid transmission studies. RNA-1 transcripts, derived from
aphid-transmissible or aphid non4ransmissible PEMV isolates, and RNA2 transcripts were co-inoculated onto pea seedlings as described
previously (Demler eta]., 1994a), with or without GRV satellite YB3b
transcripts. After 7 days, these plants were used as sources for aphid
transmission: virus-free nymphs of Myzus persicae were allowed to feed
on the seedlings for 2 days, then transferred to recently emerged healthy
pea seedlings, approximately five aphids per seedling for a 2 day
inoculation access period. The aphid-inoculated plants were subsequently
maintained in a growth chamber and the success of aphid transmission
was monitored by symptom development and by Northem blot analysis
(see below). Each combination was evaluated in a minimum of three
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
IiiiiijIiiN N iii! !ii N!iNi!
independent trials, and each experiment included plants inoculated with
the parental aphid-transmissibleand non-transmissible PEMV isolates, as
well as mock-inoculated controls.
Groundnut plants were doubly inoculated by grafting with scions
from plants infected with GRAV and from plants infected with GRVMC1 plus the PEMV satellite. These plants were then used as sources in
transmission tests with Aphis craccivora, as described by Murant (1990).
The presence of the viruses and satellite in the source and test plants was
determined by ELISAfor GRAV (Rajeshwari et al., 1987; Rajeshwari &
Murant, 1988), by infectivity testing on Chenopodium amaranticolor and
N. benthamiana for GRV (Reddyet al., I985 a), and by dot-blotting for the
satellite (see below).
• Northern blot and dot-blot analyses. Protoplast RNA was
prepared and analysed as described by Demler et al. (1993). Total RNA
from pea leaves and RNA from purifiedPEMV particles were prepared as
described by Demler & de Zoeten (1989). For Northern blot analyses,
1 ,g aliquots of RNA were subjected to electrophoresis in 1% nondenaturing agarose gels and the gels were blotted onto nylon membranes
(MSI) as described by Demler et al. (1994b). PEMV nucleic acids were
detected with digoxigenin-ll-UTP-labelled single-stranded RNA probes
prepared from transcription vectors as described by Demler et al. (1994 b)
and GRV satellite RNA was detected with probes prepared in a similar
way from clone yb3b, Dot-blots were prepared as describedby Blok e~aI.
(1995) and probed with random-primed 3~P-labelled probes prepared
from clones mc3a or yb3b (Blok et al., 1994) for detection of the GRV
satellite, or from clone pPER3 for detection of the PEMV satellite.
Results
Sequence comparison of the satellite RNAs of PEMV
and GRV
Although the lengths of the satellite RNAs of GRV and
PEMV are dissimilar (895-902 nt and 717 nt, respectively),
sequence alignment revealed several striking similarities suggestive of a possible relationship between these RNAs. There
was 59% sequence identity over the entire sequence (Fig. I)
and four regions were identified that displayed strong
similarity. One region lies at the extreme 5' terminus, where I1
of the first I3 nucleotides are identical. A more substantial
region of similarity lies in the 3'-terminal one-third of the
molecules (nt 593-903 of the GRV satellite and nt 412-717 of
the PEMV satellite), including a direct match of the final 3'terminal 1I nucleotides. It should be noted that the sequence
identities at the extreme 5' and 3' termini of the satellite RNAs
are also present in PEMV RNA-2 (Demler et al., 1993), and that
the sequences at the termini of GRV RNA are also similar
(Taliansky et al., 1996). Internally, there are two additional
regions of similarity between the satellites, both of which
involve repeated sequences occurring in each molecule. The
PEMV satellite contains two 27 nt exact repeats (underlined in
Fig. 1), whereas the GRV satellite contains two longer
imperfect duplications of 93 and 94 nt (underlined in Fig. 1). As
shown in Fig. 1, not only does each molecule contain a
duplication, but the sequence duplicated in the PEMV satellite
is almost exactly repeated in the GRV satellite. The significance
of the duplications is at present unknown.
Although the sequences of ten GRV satellite variants
obtained to date encode a number of potential ORF products,
only one (designated ORF III) is conserved among all ten (Blok
ef aI., 1994). However, examination of the PEMV satellite
sequence failed to identify any analogue of ORF III or indeed
of any of the other potential ORFs. Thus, any functional
equivalence between the satellites must be manifested at the
nucleic acid level and not through translation products.
Support of the replication of the GRV satellite by
PEMV
To assess whether PEMV can support replication of the
GRV satellite RNA, P. sativum protopIasts were inoculated
with transcripts of GRV satellite YB3b and PEMV genomic
RNA in all combinations. Replication of the GRV satellite was
assessed in Northern blots of protoplast RNA isolated 24 h
after inoculation. As shown in Fig. 2, inoculation with
combinations that included both PEMV RNA-2 and GRV
satellite RNA (lanes 2 + G , I + 2 + G
and V + G ) led to
replication of the satellite RNA. The absence of detectable
satellite RNA 24 h after inoculation with satellite RNA alone
(lane G) supports the conclusion that the positive signals are
indeed due to de novo replication. Inoculation with PEMV
RNA-I and GRV satellite RNA did not allow satellite RNA
replication. Therefore, as previously demonstrated in similar
analyses for the satellite RNA of PEMV, RNA-2 of PEMV is
able to support replication of the GRV satellite RNA, whereas
PEMV RNA-1 is not.
To examine support of the GRV satellite in whole plants,
pea seedlings were inoculated with various combinations of
PEMV genomic RNA and GRV satellite YB3b RNA transcripts. After 7 days, total RNA was isolated and analysed by
Northern blots (Fig. 3). Fig. 3 (a) demonstrates that replication
and systemic movement of the GRV satellite are supported by
PEMV RNA-2, both alone (lane 2 + G) and in the presence of
RNA-1 (lane 1 + 2 + G). The homologous PEMV satellite
RNA is supported in a comparable way by RNA-2 or by RNA1 + RNA-2 (Fig. 3 b; see also Demler eta]., 1994 b).
It may be noted in Fig. 3 (a) that the GRV satellite-specific
probe recognized a smaller secondary band in RNA derived
from seedlings 7 days after infection, which was not evident
after the much shorter incubations used in protoplasts (Fig. 2).
Under the conditions used in this experiment, this band was
not recognized by the PEMV satellite RNA probe (Fig. 3 b),
and indeed experiments in denaturing gels (data not shown)
demonstrated that it was larger than the PEMV satellite RNA.
Observations on peas infected with the GRV satellite YB3b
RNA, together with PEMV RNA-1 and RNA-2, for up to 30
clays after inoculation showed that the presence of the satellite
had no detectable effect on the development of PEMV
symptoms or on virus yield. Similarly, the weak symptoms
associated with infection of peas by PEMV RNA-2 alone were
unaffected by the presence of this satellite RNA. These results
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
-~84£
!i!iiiiiiiiiiiiiliiiii
::::~::::::::::~:::~:
iiii!ililiiiii~
iiii
-~
~!
o~
U --
g
~1
n,
r,.)
~
t5
o
~
~
~
~
g
tO--
o,,_
~ ~_-
C~
CJ
r.,j
C;
--
--rJ
- - 0
U
=~
--
~=~
u -d
=:
=~
---i
-~.
~=~
g
,~
5~
--
~-~
-~
LI--O
I-I
~
~ -g
~
~
!-!
00
J.J
~
EO
~
~
-~
u ~_
rO
-°
_
~
cr~
..~
cQ
..~
-
Eo
~.--q
0"- "-~ - d
I:~ ~
r,.,O
o~_~
,,I,J
~ ~o~
r,,O
~ ~
E~
123
JJ
~_~
o-~
~-~
~
~
~
o
•
•
~
~
©
,~
~ - ~
~
o
00
.,~
~ ~
.10
~ ~
00
~ ~
:i
~
--U
..
0
~
~-~
~-~
~
~
~--~
~;~
~
~-~
g -g
g-~
~
~ - g
~5-
0
0
~o
o-~
~o
r'
0
If)
O0
r,J
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
0
0
~J
Ca
~,
~,
ca
ca
y.
ca
~
y.
ca
ev
¢v
y.
ev
(a)
GRV
satellite
Fig. 2
({
GRV
satellite
({
PEMV
satellite
(b)
Fig. 3
Fig. 2. Northern blot of total RNA,isolated from protoplasts of P. sativum, hybridized with a probe specific for GRVsatellite
sequences. Protoplasts were inoculated with various combinations of GRVsatellite YB3b RNAand the two PEMVgenomic RNA
species. The contents of the inocula are shown above each lane; 1, PEMVRNA-1; 2, PEMVRNA-2;G, GRVsatellite YB3b RNA;
V, PEMV virion RNA; IV], mock-inoculated control. The lane designated t = 0 represents protoplast RNA isolated immediately
after inoculation; all other RNA preparations were made from protoplasts 24 h p.i. The position at which GRV satellite RNA
migrated is indicated.
Fi 9. 3. Northern blots of total RNA, extracted from seedlings of P. sativum, hybridized with probes specific for the GRV satellite
(o) or the PEMV satellite (b). Plants were inoculated with various combinations of the PEMV genomic and satellite RNA species
and GRV satellite YB3b RNA. The contents of the inocula are shown above each lane; 1, PEMV RNA-t ; 2, PEMV RNA-2; G,
YB3b satellite of GRV; S, PEMV satellite RNA; M, mock-inoculated control. Samples were taken at 7 days p.i. from leaves above
those that had been inoculated. The positions at which satellite RNAs migrated are indicated.
infection by PEMV R N A - I + -2, although infected plants were
consistently stunted relative to mock-inoculated control plants.
However, the mosaic symptom typical of PEMV RNA-1 + -2
infections was significantly intensified, developing into a bright
yellow net or blotch pattern (Fig. 4, bottom row, centre) or, in
severe instances, into a general foliar chlorosis (Fig. 4, bottom
row, right). The weak leaf edge mottling associated with
infections with PEMV RNA-2 alone was also exacerbated, the
plants displaying the same bright yellow net or blotch patterns
described above. These symptoms were very similar to those
induced in N. benthamiana by the GRV-YB satellite in
conjunction with GRV (Kumar et al., 1991), and are characteristic of this particular satellite isolate. Indeed, the satellite
from strain YB was chosen for these experiments because it is
the only known variant that appreciably modifies the symptoms of GRV in hosts other than groundnut.
In groundnut, the natural host of GRV and its satellite,
replication of PEMV genomic RNA or of GRV satellite RNA
was not detected, suggesting that groundnut (at least the one
cultivar tested) is either a non-host or a subliminal host for
PEMV.
Support of the replication of the PEMV satellite by
GRV
Plants of N. benthamiana, N. clevelandii or N. occidentalis
were inoculated with GRV, alone or together with either the
GRV satellite MC3a or the PEMV satellite, and replication of
the satellites was assessed in dot-blots made from systemically
infected, symptom-bearing leaves 3 weeks after inoculation
(Fig. 5). In all three host species, GRV supported replication
and systemic movement of both satellites. Concentrations of
both satellites were lowest in N. occidentalis. Note that under
the conditions used in some experiments, there was a low level
of hybridization of the probes with the heterologous satellite;
for example, in Fig. 5 (a), the GRV satellite probe gave a weak
reaction with extracts from N. benthamiana inoculated with
GRV plus the PEMV satellite. The PEMV satellite was
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
!85
RNA-2
Mock inoculated
RNA-2 + GRV-YB
RNA-I+ RNA-2+GRV-YB
RNA-I+ RNA-2
RNA-1 + RNA-2 + GRV-YB
Fig. 4. Symptoms induced by the two PEMV genomic RNA species with or without the GRV YB3b satellite in N. benthomiana
14 days after inoculation.
C~
c~
(a)
Nc
Nb
No
fd
co
(b)
Nc
Nb
No
Fig. 5. Dot-blots of N. clevelandii (Nc), N. benthamiana (Nb) and N.
occidentalis (No) inoculated with GRV-MCl only (GRV), GRVMC1 +satellite MC3a (GRV+G), GRV-MC1 +PEMV satellite (GRV+S) or
uninoculated (U) and probed with GRV satellite-specific (a) or PEMV
satellite-specific probe (b).
'.85~
maintained in the culture with GRV in N. benfhamiana through
at least four passages by mechanical inoculation of sap. In none
of the three Nicotiana spp. did either of the satellites alter the
symptoms normally associated with infection by GRV, viz.
chlorotic mottle and leaf curling in N. benthamiana, mild mottle
and vein-clearing in N. develandii and mild mottle in N.
occidentalis. In this respect, these two satellites behave like
other GRV satellites, except that from strain YB, which is
exceptional in affecting symptoms in some Nicofiana spp.
Because it is difficult to infect groundnuts with GRV by
mechanical inoculation, tests were made by using, as inocula,
concentrated preparations of total leaf RNA from N. benthamiana plants in which infection with the GRV/satellite
combination was already established. About 2 months after
inoculation, the groundnut plants were tested for infection
with GRV by inoculation of sap extracts to C. amaranfico]or
and for presence of the satellites by dot-blotting. Three of
seven plants inoculated with GRV plus the PEMV satellite
became infected with both agents, as did one of seven plants
inoculated with GRV plus the GRV satellite MC3a. The
remaining plants were not infected with GRV and did not
contain detectable amounts of either satellite. Subsequently,
both satellite-containing isolates could be transmitted to, and
maintained in, groundnut by graft transmission. As has
regularly been found with other satellite-containing GRV
cultures, no instances of either satellite being eliminated from
the culture have been observed. Thus, replication of the PEMV
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
©
©
©
4F
(b)
(a)
:!:iiiiiiiiiiiii::::i:ii}ii:i:!iil
©
/
:::::'::::: : : : ; :
GRV
satellite
PEMV
RNA-2
GRV
satellite
Fig. 7
if:::: :
::
Fig. 6
Fig. 6. Northern blot analysis, with a probe specific for the GRV satellite, of RNA extracted from virus particles purified from pea
seedlings 10 days p.i. with the PEMV genomic RNA species only (-G) or supplemented with the GRV YB3b satellite RNA (+G).
The position at which GRV satellite RNA migrated is indicated.
Fig. 7. Northern blot analysis, with probes for the GRV satellite (a) or for PEMV RNA-2 (b), of total RNA preparations made
from the source plants used for the aphid acquisition feed (Sp) or from the recipient plants 10 days p.i. by aphids (R). The
source plants were infected with the PEMV genomic RNA species only ( - G ) or supplemented with the GRV YB3b satellite RNA
(+G) or were mock-inoculated (iv1).
satellite is supported by GRV in groundnut, a species that is
apparently not a host for its natural helper virus, PEMV.
Groundnut plants infected with GRV together with the PEMV
satellite were virtually symptomless, as were plants infected
with GRV alone. Plants infected with GRV together with the
plasmid-derived GRV satellite MC3a showed typical symptoms of chlorotic rosette.
Encapsidation of the GRV satellite by PEHV
To examine whether the particle protein of PEMV is
capable of encapsidating the GRV satellite RNA, RNA
extracted from virus particles purified from plants infected with
PEMV RNAs i + 2 and the GRV satellite YB3b was analysed
by Northern blotting (Fig. 6). Both full-length satellite RNA
and the smaller variant band noted above were detected. There
was no evidence of comparable bands from particles purified
from plants infected only with PEMV RNAs 1 + 2. Thus, the
particle protein specified by the luteovirus-like PEMV RNA-I
can encapsidate both the GRV satellite-specific RNAs identified in infected tissues.
GRV satellite YB3b with PEMV genomic RNAs i and 2,
PEMV was fully capable of mediating the aphid transmission
of the foreign satellite RNA and the smaller satellite-specific
variant RNA (Fig. 7a, lanes S p + G , R+G). Moreover, the
pattern of PEMV RNA-2-related species in the recipient plants
was the same whether the satellite was present or not (Fig. 7 b)
and included species smaller than RNA-2 itself, which have
consistently been observed in PEMV-infected tissue (Demler et
a]., 1993, 1994a, b). Thus, the GRV satellite had no discernible
effect on the aphid transmission of RNA-2 and, although it
plays an essential role in one aspect of its relationship with
GRV, it is non-essential in its heterologous association with
PEMV.
In one experiment, the PEMV satellite did not facilitate
aphid transmission of GRV. Of nine groundnut plants that
received A. craccivora which had previously fed on plants
infected with GRAV + GRV + the PEMV satellite, six became
infected with GRAV, but neither GRV nor the satellite was
detected in any of them.
Discussion
Role of heterologous satellite RNAs in aphid
transmission
A major difference between the satellite RNAs of GRV and
PEMV is the essential role played by the GRV satellite RNA in
the aphid transmission of the umbravirus GRV in association
with the luteovirus GRAV (Murant, 1990). There is no such
requirement for the PEMV satellite in the vector transmission
of PEMV RNA-2 in association with the luteovirus-like PEMV
RNA-1 (Demler eta/., 1994 b). In heterologous combinations of
Elucidation of the nucleotide sequences and biological roles
of the genomic RNAs of PEMV (Demler & de Zoeten, 1991;
Demler eta[., 1993, 1994a) invited comparison with luteovirus-umbravirus complexes, such as those involving GRV
and GRAV, or carrot mottle and carrot red leaf viruses. Subsequently, determination of the genome sequences of GRV
(Taliansky et aL, 1996) and of an Australian isolate of carrot
mottle virus (Gibbs et a[., i996) has shown that in genome
organization and in the amino acid sequences of its potential
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
!857.
....
iii
iii
iiiijiii
products, PEMV RNA-2 does indeed have affinities with
umbravirus RNA. The present finding that the satellite RNAs
that are associated with some isolates of PEMV, and with
GRV, have obvious sequence similarities adds further plausibility to the hypothesis that PEMV has evolved from a
luteovirus-umbravirus complex with increased mutual dependence of the two parts.
As helpers for replication of the two satellites, PEMV
RNA-2 and GRV RNA are essentially interchangeable. Each
will support the replication and systemic spread of either
satellite and, although no quantitative measurements were
attempted, the data in Figs 3 and 5 suggest that the two
satellites accumulated to similar levels when supported by the
same helper. There was no indication that the ability to support
either satellite was confined to particular hosts of the helper.
The molecular basis for this ability is probably the similar
sequences found at the extreme 5' and 3' termini of both
satellites, PEMV RNA-2, and GRV RNA, which are likely to
represent initiation sites for + and - strand synthesis,
respectively, by the replicases encoded by PEMV RNA-2 and
GRV RNA. The other striking feature of the sequence similarity
between the two satellites, the conservation of sequences that
are duplicated in each molecule, also seems likely to have
biological significance, but there are no indications of what that
might be.
Effects of the satellites on symptom production seem to be
of two kinds. One kind is specific to a particular satellite variant
and host plant species. Thus, GRV satellite YB3b produces
yellow blotch symptoms in N. benthamiana, whether the helper
virus is GRV or PEMV. Other satellite variants, such as MC3a,
have little or no effect on symptoms in this host, at least when
GRV is the helper. The effects of different satellites on the
symptoms of GRV in groundnut, where they are responsible
for the different forms of rosette disease (Murant & Kumar,
i990), are also of this kind, which seems to be an intrinsic
property of the satellite itself. The second kind of effect on
symptoms also seems to be specific to particular host species,
and to a particular helper virus, but not to an individual satellite
variant. Thus, in N. benthamiana, the GRV and PEMV satellites
have similar effects in ameliorating the symptoms of stunting
and malformation caused by PEMV but not the mottling and
leaf curling caused by GRV. This kind of effect probably
reflects an interaction of the satellites with processes induced
by the helper virus, and may be insensitive to variation in the
satellite.
Not only can replication of the GRV satellite be supported
heterologously, in association with PEMV RNA-2, but so also
can its aphid transmission, in association with PEMV RNA-1.
The satellite can be encapsidated by the coat protein of PEMV
and becomes transmissible by a novel vector, MI. persicae. In
contrast, no evidence was obtained that the PEMV satellite can
substitute for the GRV satellite in its essential role in aphid
transmission of GRV, and it is not clear whether the PEMV
satellite is encapsidated by GRAV coat protein. Although
'.854
these experiments were not exhaustive, they suggest that the
PEMV satellite may be unable to provide whatever function is
supplied by the GRV satellite.
The similarities in sequence between the GRV and PEMV
satellites, and the finding that they are in many ways
functionally equivalent, suggests that they have a common
evolutionary origin. However, they are now isolated from one
another, both geographically and biologically. In nature, GRV
occurs only in Africa, south of the Sahara, whereas there are no
confirmed reports of the occurrence of PEMV on the African
continent. Moreover, the viruses seem to be adapted to
different environmental conditions; PEMV is intolerant of heat
and grows best in temperate conditions, whereas GRV
flourishes in a sub-tropical or tropical environment. The only
known natural host of the groundnut rosette complex is
groundnut, which is not a host for PEMV. However, groundnut
is not native to Africa, and therefore GRV and GRAV must
have spread to it from some other species, as yet unidentified.
The experimental host range of GRAV is very limited,
although it has been transmitted to a few other species,
including pea. The experimental host range of GRV is wider,
though it does not include pea. However, species which are
hosts for GRV but not for GRAV are of little significance in
nature, because they cannot serve as sources for further
transmission by the vector. It seems unlikely, therefore, that
the GRV and PEMV satellites can have diverged after a recent
transfer from one helper complex to the other. Rather, it seems
that they have adapted to their respective helper systems in
isolation over a long period of time. Indeed, the additional
RNA species related to the GRV satellite that was detected in
infected peas (Fig. 3) and N. benthamiana (data not shown)
when PEMV was the helper may reflect the emergence of a
deletion variant or otherwise aberrant RNA derived from the
heterologous satellite RNA. Alternatively, this RNA may be
the product of a maturation process instigated by PEMV,
perhaps a step towards adaptation of the 'foreign' satellite to
greater compatibility with a novel helper virus. It will be
interesting to investigate the extent to which each satellite has
adapted to its natural helper by examining the outcome of
infections by mixtures of both satellites with each helper.
Incidental to the main purpose of this work, the production
of biologically active transcripts of GRV satellite RNAs has
clarified several points that were left unresolved by Blok et a].
(1994). Firstly, the sequence reported for clone yb3b included
an extra C residue at the 5' end, which had no counterpart in
most of the other clones. Transcripts from clone pYB3b lack
this residue but are biologically active, suggesting that it is
inessential and possibly artefactual. Secondly, it was speculated
that clone yb3b, which differed in about 5 % of residues from
other clones from Malawian chlorotic rosette isolates, might
represent the satellite variant responsible for the yellow blotch
symptoms in N. benthamiana. This has been confirmed. Thirdly,
when it is in combination with PEMV, the GRV satellite has all
the properties required of a true satellite. Previously, some
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
critics h a v e argued that its essential role in the natural survival
of G R V m e a n t that it was better r e g a r d e d as an ancillary
g e n o m e part. This v i e w n o w seems unprofitable.
SCRI is grant-aided by the Scottish Office Agriculture and Fisheries
Department, and work on this project was funded by the Plant Sciences
Research Programme of the Overseas Development Administration of
the United Kingdom. However, the Overseas Development Administration can accept no responsibility for any information provided or
views expressed. We thank Bridget Jones for assistance with the aphid
transmission experiments at SCRI.
References
Blok, V.C., Ziegler, A., Robinson, D.J. & Hurant, A.F. (1994).
Sequences of 10 variants of the satellite-like RNA-3 of groundnut rosette
virus. Virology 202,, 25-32.
Blok, V. C., Ziegler, A., Scott, K., Dangora, D. B., Robinson, D. J. &
Murant, A.F. (1995). Detection of groundnut rosette umbravirus
infections with radioactive and non-radioactive probes to its satellite
RNA. Annals of Applied Biology 127, 321-328.
Demler, S. A. & de Zoeten, G. A. (1989). Characterization of a satellite
RNA associated with pea enation mosaic virus. Journal of General Virology
70, 1075-I084.
Demler, S. A. & de Zoeten, G. A. (1991 ). The nucleotide sequence and
luteovirus-like nature of RNA 1 of an aphid non-transmissible strain of
pea enation mosaic virus. Journal of General Virology 72, 1819-1834.
Demler, S. A., Rucker, D. G. & de Zoeten, G. A. (1993). The chimeric
nature of the genome of pea enation mosaic virus: the independent
replication of RNA 2. Journal of General Virology 74, 1-14.
Demler, S. A., Borkhsenious~ O. N., Rucker, D. G. & de Zoeten, G. A.
(1994o). Assessment of the autonomy of replicative and structural
functions encoded by the luteo-phase of pea enation mosaic virus. Journal
of General Virology 75, 997-1007.
Demler, S.A., Rucker, D.G., Nooruddin, L. & de Zoeten, G.A.
(1994b). Replication of the satellite RNA of pea enation mosaic virus is
controlled by RNA 2-encoded functions. Journal of General Virology 75,
1399-1406.
Demler, S. A., De Zoeten, G. A., Adam, G. & Harris, K. F. (1996). Pea
enation mosaic enamovirus: properties and aphid transmission. In The
Plant Viruses, vol. 5, Polyhedral Virions and Bipartite RNA Genomes, chapter
12, pp. 303-334. Edited by B. D. Harrison & A. F. Murant. New York:
Plenum Press.
Devereux, l., Haeberli, P. & Smithies, O. (1984). A comprehensive set
of sequence analysis programs for the VAX. Nucleic Acids Research I2,
387-395.
de Zoeten, G. A. & Demier, S. A. (1995). Genus Enamovirus. In Virus
Taxonomy. Sixth Report of the International Committee on Taxonomy of
Viruses, pp. 384-387. Edited by F. A. Murphy, C. M. Fauquet, D. H. L.
Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo & M. D.
Summers. Vienna & New York: Springer-Verlag.
Folk, B. W. & Duffus, J. E. (1981). Epidemiology of helper-dependent
persistent aphid transmitted virus complexes. In Plant Diseases and
Vectors: Ecology and Epidemiology, pp. 161-179. Edited by K. Maramorosch & K. F. Harris. New York: Academic Press.
Gibbs, M. J., Cooper, J. I. & Waterhouse, P. H. (1996). The genome
organisation and affinities of an Australian isolate of carrot mottle
umbravirus. Virology (in press).
Hu||, R. & Adams, A. N. {1968). Groundnut rosette and its assistor virus.
Annals of Applied Biology 62, 139-145.
Kumar, I. K., Murant, A. F. & Robinson, D. J. (1991). A variant of the
satellite RNA of groundnut rosette virus that induces brilliant yellow
blotch mosaic symptoms in Nicotiana benthamiana. Annals of Applied
Biology 118, 555-564.
Kunkel, T.A. (1985). Rapid and efficient site-specific mutagenesis
without phenotypic selection. Proceedings of the National Academy of
Sciences, USA 82, 488-492.
Murant, A. F. (1990). Dependence of groundnut rosette virus on its
satellite RNA as well as on groundnut rosette assistor luteovirus for
transmission by Aphis craccivora. Journal of General Virology 71,
2163-2166.
Mutant, A. F. (1993). Complexes of transmission-dependent and helper
viruses. In Diagnosis of Plant Virus Diseases, pp. 334-357. Edited by R. E.
F. Mathews. Boca Raton: CRC Press.
Hurant, A. F. & Kumar, I. K. (1990). Different variants of the satellite
RNA of groundnut rosette virus are responsible for the chlorotic and
green forms of groundnut rosette disease. Annals of Applied Biology 117,
85-92.
Mutant, A. F. & Mayo, M. A. (1982). Satellites of plant viruses. Annual
Review of Phytopathology 20, 49-70.
Hurant, A. F., Rajeshwari, R., Robinson, D. J. & Raschk4, J. H. (1988).
A satellite RNA of groundnut rosette virus that is largely responsible for
symptoms of groundnut rosette disease. Journal of General Virology 69,
1479-1486.
Murant, A. F., Robinson, D. J. & Gibbs, M. J. (1995). Genus Umbravirus.
In Virus Taxonomy. Sixth Report of the International Committee on Taxonomy
of Viruses, pp. 388-391. Edited by F. A. Murphy, C. M. Fauquet, D. H. L.
Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo & M. D.
Summers. Vienna & New York: Springer-Verlag.
Rajeshwari, R. & Mutant, A.F. {1988). Purification and particle
properties of groundnut rosette assistor virus, and production of a
specific antiserum. Anneals of Applied Biology 112,, 403-414.
Rajeshwari, R., Murant, A.F. & Massalski, P.R. (1987). Use of
monoclonal antibody to potato leafroll virus for detecting groundnut
rosette assistor virus by ELISA. Annals of Applied Biology 111, 353-358.
Reddy, D. V. R., Murant, A. F., Duncan, G. H., Ansa, O. A., Demski, J.
W. & Kuhn, C. W. (1985 o). Viruses associated with chlorotic rosette and
green rosette diseases of groundnut in Nigeria. Annals of Applied Biology
107, 57--64.
Reddy, D. V. R., Murant, A. F., Raschk4, J. H. & Mayo, M. A. (1985b).
Properties and partial purification of infective material from plants
containing groundnut rosette virus. Annals of Applied Biology 107, 65-78.
Taliansky, M. E., Robinson7 D.J. & Hurant, A. F. (1996). Complete
nucleotide sequence and organization of the RNA genome of groundnut
rosette umbravirus. Journal of General Virology 77, 2335-2345.
Received 7 May 1996; Accepted 8 July 1996
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 02:55:07
!85!