Error bars represent standard deviation from the mean values. The equation describing the head capsule width increase across instars, along with the coefficient of determination (R²), is shown.

Occlusion bodies (OBs) of the local AcMNPV isolate were produced and purified from its original host (R. nu larvae). They were checked by scanning electron microscopy (FEI Quanta 250, FEI Co., Eindhoven, Netherlands) to discard possible contaminations (Fig. 2a) and counted on a Neubauer chamber. In a first experiment, groups of 37-44 neonate H. gelotopoeon larvae were supplied with different doses of AcMNPV OBs (0, 2, 5, 10 and 50), one group per dose, by droplet-feeding (Hughes & Wood, 1981), assuming a mean ingestion of 0.014 µl per specimen (Ferrelli et al., 2016). Larvae were then maintained individually on artificial diet as referred earlier and observed daily until death or pupation. AcMNPV caused 100 % mortality to H. gelotopoeon larvae only at the highest dose of 50 OBs. In this case, half of the tested insects were dead by day 8 post inoculation (p. i.). Irrespective of the dose, speed of kill was consistently slow (Fig. 3a). Accordingly, though a different methodology was used, more than 9 days were required to attain 50 % mortality in H. zea neonates (Vail et al., 1978). For biocontrol purposes, these results were rather poor: a similar study on nenonate H. gelotopoeon infected with HegeSNPV revealed that 18 OBs/larva were sufficient to induce 100 % mortality, while the median lethal dose was below 2 OBs and the average time to death slightly above 4 days (Ferrelli et al., 2016). AcMNPV infection was verified in all dead larvae by direct observation of cubic OBs by phase-contrast microscopy. A transmission electron microscopy confirmation was also carried out. Two larvae were infected in parallel to the main experiment; at 7 days p. i. they were processed as previously reported (Ferrelli et al., 2016) and the ultrathin sections examined in a TEM JEOL 1200 EX II (JEOL, Tokyo, Japan). AcMNPV was relatively sparse across the host tissues. A few heavily infected cells displayed intranuclear OBs. Multiple nucleocapsids per envelope were observed (Fig. 2b). The shape, size and general ultrastructure of OBs and virions were typical of AcMNPV. To further corroborate disease etiology, total DNA was extracted (CTAB method) from two dead larvae, and the polyhedrin and gp64 genes amplified by PCR using primer pairs 5’-NRCNGARGAYCCNTT-3’ / 5’-DGGNGCRAAYTCYTT-3’ and 5’-ACTGCAGTTAATATTGTCTATTACGG-3’ / 5’-ACTGCAGCGGCCGCCCACTTGGCGCC-3’, respectively. Amplicons were sequenced in an ABI PRISM 3500 XL genetic analyzer (Applied Biosystems, USA) at Instituto de Biotecnología-CICVyA-INTA (Hurlingham, Argentina). Both sequences, available at GenBank under accession numbers MF066628 (polyhedrin) and MF066627 (gp64), were identical to those of AcMNPV-R. nu (DQ345451, polyhedrin gene) and AcMNPV isolates WP10 (KM609482, complete sequence) from Trichoplusia ni (France) and E2 (KM667940, complete sequence) from A. californica (USA).

Figure 2.

a. Purified AcMNPV cubic OBs under the scanning electron microscope. b. Ultrathin section of a nucleus of an AcMNPV-infected Helicoverpa gelotopoeon larva. Virions containing multiple nucleocapsids per envelope in longitudinal (arrows) and transversal (arrowheads) view. c. Epidermis of an AcMNPV-infected H. gelotopoeon larva after Azan staining, showing bright-red OBs. d. Same as before, but in a HegeSNPV-infected larva.

A second experiment was performed to simultaneously evaluate the pathogenicity of AcMNPV and HegeSNPV to 3^rd instar H. gelotopoeon larvae. The AcMNPV isolate used was the same as above, while HegeSNPV OBs (also checked by microscopic and molecular analyses) were propagated and purified from H. gelotopoeon larvae. Because older H. gelotopoeon larvae were not attracted by the sugar solution, the diet surface contamination method (Eberle et al., 2012) was used to deliver the virus isolates to two groups of 40 larvae each. Based on prior results showing low susceptibility of H. gelotopoeon to the generalist baculovirus, a considerably high dose of 400 OBs/mm² was selected in order to induce detectable effects. Starved, newly moulted larvae were exposed to one or the other virus for 48 h and transferred to individual cups containing artificial diet for further rearing. Negative (healthy) controls (n = 32) were also considered. The effect of AcMNPV and HegeSNPV on larval development and survival, and the associated symptomatology, was strikingly different. In HegeSNPV-inoculated individuals, 100 % mortality was achieved by day 7 p. i. (Fig. 3b). As in most baculoviral infections, post-mortem liquefaction of the integument was consistently observed. In contrast, specimens supplied with AcMNPV developed in an apparently normal manner until the prepupal stage. However, 90 % of tested larvae failed to pupate and died between 12 and 31 days p. i. (Fig. 3b). The typical disintegration of insect cadavers was not recorded. Such retarded larval development and unusual symptomatology had also been noticed by Vail & Jay (1973) when assessing the susceptibility of H. zea to AcMNPV.

Figure 3.

a. Neonate larvae inoculated with different doses of AcMNPV through the droplet-feeding method. b. Third-instar larvae exposed to 400 OBs/mm² of HegeSNPV or AcMNPV by diet surface contamination. HC: healthy (uninfected) controls.

Larval liquefaction favours the release of OBs from the cadavers into the environment and, therefore, the dissemination and horizontal transmission of the virus. The disruption of the cuticle is promoted by viral enzymes, which have been identified in several baculovirus species, including AcMNPV (Rohrmann, 2019). Such fragility could be boosted by the viral replication in the epithelial cells following systemic infection. Hence, a histopathological analysis was conducted on AcMNPV- and HegeSNPV-infected H. gelotopoeon larvae, aiming to detect possible differences in tissue tropism. Third-instar larvae were inoculated by diet contamination as explained above. At 3 (HegeSNPV) or 5 (AcMNPV) days p. i., larvae were fixed, embedded, sectioned and stained with Azan (Hamm, 1966). A preliminary sampling of AcMNPV-infected larvae at 10 days p. i. appeared to be inappropriate for histology because of tissue reorganisation during metamorphosis (not shown). Light microscopic examinations of semithin sections disclosed bright-red stained OBs in AcMNPV- (Fig. 2c) and HegeSNPV-infected (Fig. 2d) larval tissues. No obvious differences were found with respect to the main sites of virus colonisation which were, in both cases, the fat bodies, epidermis and tracheal matrix. Thus, AcMNPV was polyorganotropic in H. gelotopoeon, as in full permissive lepidopteran hosts (Barret et al., 1998). The same was true for HegeSNPV. Larvae treated with HegeSNPV showed, however, higher OB proliferation in a comparatively larger proportion of infected cells than those exposed to AcMNPV. In addition, the nuclear hypertrophy associated to baculoviral infections was more clearly appreciated in the former case. No OBs were seen in the healthy controls (not shown). Again, regarding tissue specificity and OB production, the gross pathology produced by AcMNPV in H. gelotopoeon resembled that described for H. zea (Vail & Jay, 1973).

The present work demonstrated that H. gelotopoeon is a semi-permissive host for AcMNPV. As for H. zea, an extremely slow speed of kill, together with the relatively high doses required to produce mortality, makes this virus an ineffective biocontrol agent for H. gelotopoeon. Such response is particularly noteworthy since H. gelotopoeon often co-occurs in space and time with other polyphagous noctuid pests for which AcMNPV-based formulations are available or in development, e. g. Spodoptera spp and Rachiplusia nu (Haase et al., 2015; Decker-Franco et al., 2021). For the latter species, AcMNPV was shown to act even faster than its specific baculovirus, RanuNPV (Decker-Franco et al., 2021). On the other hand, the experiments provided additional evidence on the biocontrol potential of HegeSNPV, which deserves further research for its effective use in IPM strategies. Although HearNPV-based bioinsecticides are already available in the market (Rowley et al., 2011; Knox et al., 2015), the incorporation of new, virulent isolates may help improve current management of Heliothinae and delay the emergence of resistance.