Previous work with tobacco vein mottling virus (TVMV) has established that a highly conserved three amino acid motif, asp-ala-gly (DAG), located near the N terminus of the coat protein (CP), is important for aphid transmission. However, several other potyviruses which have motifs other than DAG are aphid-transmissible. Creation of these motifs in TVMV through site-directed mutagenesis failed to render TVMV aphid-transmissible from infected plants, and the creation of a putative complementary motif in the helper component did not restore transmissibility. In an isolate of tobacco etch virus (TEV) that contains two consecutive DAG motifs separated by a single ala, transmissibility was abolished or reduced by mutations affecting the first motif, whereas mutations in the second motif had little or no effect. In a TEV mutant made non-transmissible due to an altered first motif, substitution of val for ala in the position immediately before the second DAG restored transmissibility, whereas changing val to ala in the location prior to the first DAG resulted in reduced TEV transmissibility. In contrast, a val to ala change in the position preceding the single DAG motif of TVMV did not affect transmission. Creation of another DAG motif at the beginning of the TVMV CP core, in a position where certain other potyviruses have a second DAG motif, did not restore transmissibility. Our results suggest that the mere presence of a DAG motif does not guarantee transmissibility and that the context in which the DAG or equivalent motif is found plays a role in the process.

The hypothesis that loss of aphid transmissibility of potyvirus mutants is due to non-retention of virions in the mouthparts was tested by feeding aphids through membranes on purified virions of aphid transmissible (AT or HAT) and non-aphid- (HC). TVMV virions were detected, by electron microscopic examination of immunogoldlabelled thin sections, in the food canal or cibarium of 57 % of 28 aphids fed on the transmissible combination of TVMV-AT and functional HC, while no virions were found in these structures in 25 aphids fed on the nontransmissible combinations: TVMV-NAT and PVY HC, or TVMV-AT and PVC HC. Autoradiography of intact stylets allowed the examination of much larger numbers of aphids, fed on 125I-labelled TEV; 48 % of 523 aphids fed on the TEV-HAT and PVY HC combination retained label in the stylets; this correlated well with the percentage transmission in bioassays. In contrast, in non-transmissible combinations, label was found in the stylets of 0.77% of 389 aphids fed on TEV-NAT and PVY HC, and 1-35 % of 223 aphids fed on TEV-HAT and PVC HC. No differences were found in the overall amount of label in the bodies of aphids fed on the transmissible and non-transmissible combinations. There was a strong tendency for virions to be retained in the distal third of the stylets; 56 % of aphids positive for TVMV, and 82 % of those positive for TEV, had label in this area. These data support the concept that virions retained within the stylets are those that are primarily involved in potyvirus transmission.

The recently introduced soybean aphid (Aphis glycines), which is widespread in the soybean-growing regions in the United States, is the only aphid able to develop large colonies on soybean. Although its potential as a vector of plant viruses is recognized, reports on virus transmission efficiency by this aphid species are limited. In the present study, we examined the ability of A. glycines to transmit several economically important viruses. The results showed that A. glycines transmitted the potyviruses Bean yellow mosaic virus (BYMV) and Soybean mosaic virus from soybean to soybean more efficiently than Myzus persicae. However, M. persicae transmitted the alfamovirus Alfalfa mosaic virus and the potyviruses Tobacco etch virus (TEV) and Tobacco vein mottling virus (TVMV) from tobacco to tobacco more efficiently than A. glycines. This is the first report to demonstrate that the soybean aphid can vector TEV and TVMV, two economically important tobacco viruses. This also is the first report to document successful transmission of BYMV by A. glycines. All attempts to transmit the nepovirus Tobacco ringspot virus by A. glycines were unsuccessful, regardless of the length of the acquisition and inoculation feeding periods. Although the luteovirus Soybean dwarf virus (SbDV) was widely distributed in red and white clover in Kentucky, it was not detected in soybean. All transmission experiments of SbDV by A. glycines were unsuccessful. A reverse-transcription polymerase chain reaction (RT-PCR) assay was developed to detect SbDV in single aphids using a pair of primers designed to amplify a 372-bp PCR fragment in the coding region of SbDV coat protein. Although A. glycines was not a vector of SbDV, the virus was detected in 100% of tested aphids by RT-PCR after a 24- to 48-h virus acquisition access feeding. The practical applications of RT-PCR in detecting persistently transmitted viruses are discussed.

Aphids (Myzus persicae), fasted after removal from healthy rearing plants, transmitted tobacco etch potyvirus (TEV) more efficiently than unfasted aphids whether virus acquisition was from infected leaves or through membranes. There was no difference in uptake of lZSl-labelled TEV by fasted or unfasted aphids as measured by liquid scintillation counting. When aphids acquired lZSl-labelled TEV, label was retained in the stylets (as determined by autoradiographic light microscopy) by 51% of 272 fasted aphids, as against 7"8% of 258 unfasted aphids. There was a close correlation between virus transmission by aphids and virion retention in stylets. The effect of pre-acquisition fasting disappeared when aphids reared on an artificial diet were used in virus transmission tests. The transmission rates obtained with such aphids were similar to the rates with fasted aphids reared on healthy plants. Our results support the hypothesis that fasting eliminates plant component(s) which interfere with the retention of virions in the food canal of aphid stylets.