Trypanosomatid prevalence in Nezara viridula (L.), Euschistus heros (Fabricius) and Piezodorus guildinii (Westwood) (Heteroptera: Pentatomidae) populations in Northern Paraná, Brazil

Prevalência de Tripanossomatídeos em populações de Nezara viridula (L.), Euschistus heros (Fabricius) e Piezodorus guildinii (Westwood) (Heteroptera: Pentatomidae) no norte do Paraná

SCIENTIFIC NOTE

Trypanosomatid prevalence in Nezara viridula (L.), Euschistus heros (Fabricius) and Piezodorus guildinii (Westwood) (Heteroptera: Pentatomidae) populations in Northern Paraná, Brazil

Prevalência de Tripanossomatídeos em populações de Nezara viridula (L.), Euschistus heros (Fabricius) e Piezodorus guildinii (Westwood) (Heteroptera: Pentatomidae) no norte do Paraná

Daniel R. Sosa-Gómez^I; Ednéia Borges^II; Ivan H.T.L. Viera^III; Fernando Costa^III; Caroline N. Oliveira^III

^ICentro Nacional de Pesquisa de Soja, Embrapa Soja, C. postal 231, 86001-970, Londrina, PR [e-mail]

IIUndergraduate

^IIIGraduate students Embrapa Soja

ABSTRACT

The stink bug complex in soybeans has a great potential to cause damage, their control has been difficult due to high population densities at the end of the season. Studies related to entomopathogenic organisms in stink bugs are interesting for their possibilities to be used as biological control agents and because they could be an etiological agent in mass rearing. We determined the prevalence of trypanosomatids in the most important species of the stink bug complex in soybean, from Brazil, Nezara viridula (L.), Euschistus heros (Fabricius) and Piezodorus guildinii (Westwood). Insects were collected in soybean and on alternate hosts during the intercrop season. Hemolymph drops from each adult were mounted on slides and observed under microscope; stink bug numbers with and without flagellates were recorded. The flagellates found in P. guildinii or E. heros populations were smaller than that found in N. viridula. The higher prevalence of the trypanosomatids was observed on soybean reaching a peak of 30% of the sampled N. viridula specimens in January, 2001, and 44% prevalence in February, 2003. The maximum prevalences (16%) in P. guildinii were observed in January and February, 2003 and on July 27, 2003 with 20% of the specimens infected. Flagellates in the hemolymph of E. heros were less frequent (maximum prevalence of 8.7%), suggesting that this species is less susceptible than N. viridula and P. guildinii. Trypanosomatid prevalence was higher than that observed for entomopathogenic fungi, however they seem to be inefficient control agents in short term condition.

Key words: Insect, stink bug, protozoa, flagellate

RESUMO

O complexo de percevejos da soja possui um grande potencial de dano, seu controle é difícil ao final da safra por suas elevadas densidades populacionais. Trabalhos relacionados com organismos entomopatogênicos em percevejos são interessantes devido a suas possibilidades como agentes de controle e porque eles podem ser agentes causais de doenças em criação massal. A prevalência de tripanossomatídeos foi determinada nas espécies mais importantes da cultura da soja, no Brasil, Nezara viridula (L.), Euschistus heros (Fabricius) e Piezodorus guildinii (Westwood). Os insetos foram coletados na soja e hospedeiros alternativos durante a entressafra. Gotas de hemolinfa de cada adulto foram montadas em lâminas e observadas ao microscópio; o número de percevejos com e sem flagelados foi registrado. Os flagelados encontrados nas populações de P. guildinii e E. heros foram menores que os observados em N. viridula. A maior prevalência dos tripanossomatídeos foi observada nos espécimes de N. viridula coletados na soja alcançando 30% de infecção em janeiro, 2001, e 44% prevalência em fevereiro de 2003. Em P. guildinii as máximas prevalências foram observadas em janeiro e fevereiro de 2003 (16%) e em 27 de julho de 2003 com 20% dos espécimes infectados. Os flagelados na hemolinfa de E. heros foram menos freqüentes (máxima prevalência de 8,7%), sugerindo que esta espécie é menos suscetível que N. viridula e P. guildinii. Embora os níveis de prevalência sejam mais elevados que os observados para fungos entomopatogênicos, os tripanossomatídeos não parecem ser eficientes reguladores das populações de percevejos de percevejos em curto prazo.

Palavras-chave: Inseto, percevejo, protozoa, flagelado

The stink bug complex in soybeans has great potential to cause damage (Corrêa-Ferreira & Panizzi 1999), and their populations can reach unusual levels, in some cases 20 to 25 individuals per soybean row meter. At the end of the growing season, stink bug control is difficult due to high population densities and some cases of insecticide resistance were recently reported (Sosa-Gómez et al. 2001). Although an egg parasitoid can be used (Corrêa-Ferreira 1993) most of the time stink bugs are controlled by a large spectrum of insecticides.

Studies related to entomopathogenic organisms in stink bugs are interesting because of their possibilities to be used as biological control agents and also because they could be a source of diseases during mass rearing. The most common naturally occurring pathogens of stink bugs are fungi [Beauveria bassiana (Bals.) Vuill., Metarhizium anisopliae (Metsch.) Sorok., Paecilomyces fumosoroseus (Wize) Brown and Smith], some RNA viruses (ssRNA small picorna-like virus and a dsRNA Toti-like virus) (Williamson and Wechmar 1992, 1995), and protozoan flagellates (Gibbs 1957, Batistoti 1998, Fuxa et al. 2000). However, a search for key natural control agents in stink bugs field populations has been neglected. Among the pathogens associated with soybean stink bugs, none of them cause perceivable epizootics. Sosa-Gómez & Moscardi (1998) reported high prevalence of B. bassiana on Nezara viridula (L.), in the greenhouse, an artificially protected environment, but cases like this have not been observed in the field.

Studies on 17 heteropteran species from Brazil revealed that 13 of them had trypanosomatid infections (Godoi 2000). Among the sampled specimens, two pentatomid species, Euschistus and Edessa rufomarginata (DeGeer) were not infected. Thus, although the presence o trypanosomatid in Brazilian stink bugs is known (Batistoti 1998, Godoi 2000) no information is available about protozoan prevalence in the more important stink bug pests of soybean [Euschistus heros (Fabricius), N. viridula e Piezodorus guildinii (Westwood)].

Therefore our purpose was to monitor these pathogens in the hemolymph of soybean stink bugs for three years. We determined the prevalence [see Fuxa and Tanada (1987)] of flagellates in the three more important species of the stink bug complex from Brazil, N. viridula, E. heros and P. guildinii.

Adult insects were collected in soybean fields during the summer and on alternate hosts during the intercrop season. N. viridula samples were obtained on chinese motherwort (Leonorus sibiricus L.) and wild radish (Raphanus raphanistrum L.), P. guildinii on indigofera (Indigofera truxillensis H.B.K. and I. suffruticosa Mill. ) and pigeon pea (Cajanus cajan (L.) Millsp.), and E. heros in mulch of mango leaves (Manguifera indica L.) and on sunflower (Helianthus annuus L.). Collection sites were located in the North of Parana State: Warta, Guaravera (both from Londrina County), Jataizinho and Sertanópolis. Samples of stink bugs, ranging from 15 to 40 specimens of each species, were taken to the laboratory and bled through a shallow puncture on the lateral angle of the pronotum made with a stylet. The puncture was done inserting no more than 1 mm or 1.5 mm of the stylet tip inside the pronotal spine, far away enough of the digestive tract, to ensure that flagellates were not coming from the gut by the puncture. After gentle pressure, drops of hemolymph were collected on slides and observed with a phase-contrast microscope at 400 X; stink bugs with and without flagellates were recorded. The blood droplets were examined microscopically under the cover slip for the presence of flagellates, even though active forms were present in both hemolymph and feces. The droplet size was determined by weighting, assuming that the hemolymph have the same density of water. The flagellate bodies and flagella were measured with a OSM filar micrometer eyepiece (Ramsden type) (Olympus Optical Co. Ltd., Japan).

Because we did not initially detect flagellate infections in E. heros blood samples (n = 124), we suspected that this species was not susceptible to flagellatosis of the hemolymph. For this reason, N. viridula and P. guildinii samples were collected during 2000/2001, 2002/2003, and E. heros samples were collected starting from October 11, 2002 until August of 2003.

Comparisons of flagellates prevalence among stink bug species were done with the non-parametric test of Mann-Whitney Rank sum Test from Sigmastat Statistical Software (Jandel Scientific 1994).

Trypanosomatids were very common in N. viridula and P. guildinii populations (Fig. 1). Promastigotes and long forms were predominant. Flagellates found in P. guildinii and E. heros were smaller than those found in N. viridula (Table 1). Trypanosomatids from N. viridula were 50.1 ± 2.7 µm and the flagella were 31.5 ± 1.1 µm. One specimen found in E. heros had a body length of 25.1 µm and flagellum of 14.6 µm. In P. guildinii, the trypanosomatids were 22.9 ± 3.1 µm and 18.6 ± 3.1 µm, for body length and flagellum, respectively.

During the observational periods, flagellate prevalence increased in the populations of N. viridula and P. guildinii in the summer, during the wet and hot weather (January and/or February) reaching a peak of 30% in the sampled N. viridula specimens in January, 2001, and 44% prevalence in February, 2003 (Figs. 2 and 3). Therefore, we can consider that the rain distribution during the high prevalence period, could be positively related with the infection rates (Fig. 4). N. viridula had flagellates infection ranging from 2.5% to 44%, and P. guildinii had prevalence ranging from 2.9% to 20%. The maximum prevalences in P. guildinii were observed between January 29 and February 25, 2003 (16%) and on July 27, 2003 with 20% of the specimens infected. In E. heros, flagellates were observed on only three sampling dates in October, 2002, and only at prevalence rates of 3.3%, 8.7% and 4%. The prevalence in E. heros populations was significantly lower than in N. viridula populations (P < 0.0001, T = 424.0, n = 17, N = 17, Mann-Whitney Rank Sum Test) and P. guildinii populations (P = 0.03, T = 382.5, n = 17, N = 19, Mann-Whitney Rank Sum Test).

The average volume of hemolymph droplets obtained from N. viridula was higher (2.1 µl ± 0.7 (± SEM) than those from P. guildinii and E. heros (0.8 µl ± 0.4 and 0.5 µl ± 0.1, respectively). The number of flagellates per microliter of hemolymph in N. viridula (2.48 flagellates/µl) and P. guildinii (2.17 flagellates/µl) was not significantly different (P = 0.69, T = 781.0, n = 19, N = 67, Mann-Whitney Rank Sum Test) but was higher in N. viridula than in E. heros (0.50 flagellates/µl) (P < 0.009, T = 715.0, n = 22, N = 67, Mann-Whitney Rank Sum Test). No significant differences between P. guildinii and E. heros were observed. The number of trypanosomatids in the hemolymph of the three hemipterans ranged from 0 to more than 50 per blood sample (Fig. 5).

Although size differences were observed, morphological aspects are not conclusive to differentiate any taxon. Godoi (2000) stated that morphometric analysis was useless to resolve differencies among Genera. The taxonomy of this protozoan remains unclear at present; taxonomic studies of stink bug protozoans are necessary (Catarino et al. 2001). D.R. Sosa-Gómez & D.G. Boucias (unpubl.) observed infection by flagellates in N. viridula samples collected from corn during the summer of 1994, in Quincy, Florida, USA. Morphologically, the trypanosomatids from N. viridula, in the current study, are similar to those found in Louisiana (Fuxa et al. 2000) and to those that we found previously, from N. viridula collected from Quincy, FL. After isolation, we verified that the Florida flagellates reproduced very profusely in Trichoplusia ni cell cultures (TN368) immersed in TC100 medium (Life Technologies, Inc.), but Spodoptera frugiperda cells (Sf9) cultures were not permissive for the same trypanosomatids (D.R. Sosa-Gómez & D.G. Boucias, pers. observ.).

In the N. viridula samples from Quincy, FL, and in P. guildinii from Londrina, we found cysts attached to the flagellum (Fig. 1c). Cysts are thought to be forms of survival and propagation of trypanosomatids, and they occur in the Blastocrithidia and Leptomonas genera (Takata et al. 1996).

Although the mean size of body and flagella in the current study does not match those of Fuxa et al. (2000), there was an overlapping of the body and flagellum length. We observed a prevalence lower than that reported by Fuxa et al. (2000) in N. viridula, which reached almost 95% of the stink bugs infected. The higher number observed by Fuxa et al. (2000) could be due to the sampling method, because we sampled flagellates in the hemolymph, whereas these authors diagnosed infection by dissecting the alimentary tract of each individual. The infection peaks from November until March, were observed on soybean. At present little is known about the interactions between host plant species and levels of infection.

Unclean colonies of N. viridula in (food contaminated with feces) mass rearing had a high prevalence of infection (D.R. Sosa-Gómez & D.G. Boucias, unpub.), possibly due to the transmission of this pathogen through the feces left on the feeding substrate. The condensation of water on soybean plants early in morning may help the locomotion of the active phase and favor transmission. This behavior could explain the high prevalence during wet weather.

The presence of flagellates in N. viridula hemolymph is an indicative that flagellates can transpose the gut barrier more easily that in E. heros. This could be due to differences in the gut wall or in the flagellates virulence. Differences in virulence among trypanosomatid isolates have been observed by Baccan et al. (2001); for example, Leptomonas isolated from E. heros caused mortality very rapidly (24h after infection) in Veneza zonata (Pallas) but other isolates did not. Despite the high densities of flagellates in N. viridula and P. guildinii blood droplets, the insects were alive, revealing the nature of chronic diseases against these species. Until present the effect on longevity, fertility and fecundity is unknown.

Flagellates apparently do not invade and freely circulate in the hemolymph of E. heros as they do in N. viridula and P. guildinii. Their detection in E. heros required strong pressure on the body during the bleeding procedure. Thus, it is possible that they were released into the hemocoel after rupture of the gut wall. A Leptomonas sp. was detected in the digestive tract of E. heros by Batistoti (1998).

Low levels of trypanosomatid parasites in E. heros compared to N. viridula implies a lower susceptibility of this species. The mechanisms involved in these differences of susceptibility are unknown. Baccan et al. (2001) suggest that there are several mechanisms involved in the penetration of the gut wall towards the hemocoel, some of them rendering the wall permeable to the bacteria inhabiting the digestive tract.

The importance of pathogens as regulatory agents of insect populations is becoming more evident in several host-pathogen systems. Previously, life table studies of stink bugs usually did not consider pathogens as death cause. In biology studies stink bugs present high mortality rates but nothing is known about their etiological causes (Panizzi 1987). However pathogens such as viruses have been mentioned as the cause of mortality and diminished reproductive capability in the southern green stink bug (Williamson & Wechmar 1995). Flagellates could be a causal agent partly responsible for high mortality rates and reduced life span. Additional studies are necessary, because the impact of infection by these protozoans on pentatomid biology and reproduction at population level is still unknown. Although, trypanosomatid prevalence was higher than that observed for entomopathogenic fungi, they seem to be inefficient control agents in short term condition.

Acknowledgments

We thank Dr. James R. Fuxa for the critical review of the manuscript, and we acknowledge Jovenil José da Silva and José Jairo da Silva for their technical assistance. This work received financial support from the Embrapa Soja and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). This paper was approved for publication by the editorial board of Embrapa Soja as manuscript 02/2004.

Literature Cited

Received 20/VI/04. Accepted 25/X/04.

• Baccan, G., A.M.F. Falleiros, M. Cavazzana Jr., J.V. Jankevicius, M. Attias, W. de Souza & S.I. Jankevicius. 2001.Veneza zonata (Hemiptera: Coreidae)/ Trypanosomatid relationship: Action of hemolymph in vitro and experimental infection. J. Invertebr. Pathol. 77: 158-164.
• Batistoti, M. 1998. Tripanosomatídeos de insetos: Caracterização morfológica e fisiológica. Dissertação de mestrado, Universidade Estadual de Londrina, Londrina, 111p.
• Catarino, L.M., M.G. Serrano, M. Cavazzana Jr., M.L. Almeida, E.K. Kaneshina, M. Campaner, J.V. Jankevicius, M. M.G. Teixeira & S.I. Jankevicius. 2001. Classification of trypanosomatid from fruits and seeds using morphological, biochemical and molecular markers revealed several genera among fruit isolates. FEMS Microbiol. Lett. 201: 65-72.
• Corrêa-Ferreira, B.S. 1993. Utilização do parasitóide de ovos Trissolcus basalis (Wollaston) no controle de percevejos da soja. Londrina, Embrapa Soja. Circular técnica 1, 40p.
• Corrêa Ferreira, B.S. & A.R. Panizzi. 1999. Percevejos da soja e seu manejo. Londrina, Embrapa Soja, Circular técnica 24, 45p.
• Fuxa, J.E., J.R. Fuxa, A.R. Richter & E.H. Weidner. 2000. Prevalence of a trypanosomatid in the southern green stink bug, Nezara viridula J. Eukaryot. Microbiol. 47: 388-394.
• Fuxa, J.R. & Y. Tanada. 1987. Epidemiological concepts applied to insect epizootiology, p.3-21. In J.R. Fuxa & Y. Tanada (eds.), Epizootiology of insect diseases, New York, John Wiley Sons, 555p.
• Gibbs, A.J. 1957.Leptomonas serpens n. sp., parasitic in the digestive tract and salivary glands of Nezara viridula (Pentatomidae) and in the sap of Solanum lycopersicum (Tomato) and other plants. Parasitology 47: 297-303
• Godoi, M.M.I. de M. 2000. Phytomonas e outros tripanossomatideos em insetos no estado de Rondônia. Dissertação de mestrado, Universidade de São Paulo, São Paulo. 59p.
• Jandel Scientific. 1994. Sigmastat statistical software, User's manual. San Rafael, California, Jandel Scientific.
• Panizzi, A.R. 1987. Mortalidade e preferência alimentar de Nezara viridula e Piezodorus guildinii (Hemiptera: Pentatomidae) alimentados com vagens e sementes verdes de soja. Pesq. Agropec. Bras. 22: 345-347.
• Sosa-Gómez, D.R. & F. Moscardi. 1998. Laboratory and field studies on the infection of stink bugs, Nezara viridula, Piezodorus guildinii, and Euschistus heros (Hemiptera: Pentatomidae) with Metarhizium anisopliae and Beauveria bassiana in Brazil. J. Invertebr. Pathol. 71: 115-120.
• Sosa-Gómez, D.R., I.C. Corso & L. Morales. 2001. Insecticide resistance to endosulfan, monocrotophos and metamidophos in the neotropical brown stink bug, Euschistus heros (Fabr.). Neotrop. Entomol. 30: 317-320.
• Takata, C.S.A., E.P. Camargo & R.V. Milder. 1996. Encystment and excysment of a trypanosomatid of the genus Leptomonas Europ. J. Protistol. 32: 90-95.
• Williamson C. & B.V. Wechmar. 1995. The effects of two viruses on the metamorphosis, fecundity and longevity of the green stinkbug, Nezara viridula J. Invertebr. Pathol. 65: 174-178.
• Williamson C. & M.B. von Wechmar. 1992. Two novel viruses associated with severe disease symptoms of the green stinkbug Nezara viridula J. Gen. Virol. 73: 2467-2472.

Publication Dates

• Publication in this collection
  14 July 2005
• Date of issue
  Apr 2005

History

• Received
  20 June 2004
• Accepted
  25 Oct 2004