Introduction
Anoplophora glabripennis (Motschulsky) (Coleoptera: Cerambycidae: Lamiini), which has caused great economic losses in ecology and society, is a wood-boring beetle native to eastern China and Korea. The insect attacks many species of hardwood trees (
Xiao, 1992;
Lingafelter et al., 2002;
Zhang and Luo, 2003) and has become an invasive pest in other parts of the world. It has been found infesting urban shade trees in North America in the cities of New York, Chicago and Jersey (
Haack, 2003), and in Europe in cities such as Braunau of Austria. Adults undergo a 1-2-week period of maturation, feeding on the foliage and tender bark of twigs of host trees before oviposition (
Keena, 2002;
Smith et al., 2002). Females then chew slits or holes through the bark of host trees and lay their eggs under the bark. Only a single egg is laid in each oviposition site (
Lingafelter et al., 2002), and females often chew many potential oviposition sites that they do not ultimately use for oviposition (
Keena, 2002). Larvae feed in the cambium during their first few stadia and then bore into the wood, where they continue to feed, eventually forming a pupal chamber. Larval boring produces structural weaknesses and disrupts the flow of water and nutrients within the host trees, leading to the death of the branches and ultimately the whole trees.
Major programs have been undertaken by the related Chinese agencies to control this insect (
Zhang et al., 1999). However, there are not so many effective biological methods to control the insect pests. Therefore, it is necessary to find an effective biological way to control the pest. One way for the control that has been investigated is to apply the non-woven fiber bands containing cultures of entomopathogenic fungi. This delivery system is used for the control of adult lamiine cerambycid pests, particularly
Anoplophora chinensis (
A. malasiaca) and
Psacothea hilaris, in Japanese orchards with
Beauveria brongniartii (
Higuchi et al., 1997). Fungal bands are placed around tree trunks and branches, and the pest adults will be inoculated when creeping across the bands. The key elements that make this control method effective in Japan are as follows: (1) adult lamiine pests in Japanese orchards are susceptible to
B. brongniartii; (2) the non-woven bands serve as the environment for maintaining viable infective fungal spores for several months (
Higuchi et al., 1997;
Hajek et al., 2007); and (3) the normal daily activity of lamiine orchard pests in Japan is worming up and down the bark surface. The other way is to use the INA bacteria to kill the larvae of
A. glabripennis through freezing-induction (
Sun et al., 1997) and exterminate the
Nosema glabripennis Zhang parasitizing the longhorned beetle (
Zhang et al., 2003), etc.
Metarhizium (Metsch.) Sorokin, formerly known as
Entomophthora anisopliae, is a fungus that infects insects as a biological control agent, and it has been successfully used in the control of many agricultural and forest pests. But there is no systematic study on the application of
Metarhizium Sorokin to control
A. glabripennis. Adults of
A. glabripennis are susceptible to the fungal pathogen
Metarhizium anisopliae (
Dubois, 2003;
Dubois et al., 2007) and they commonly walk on tree branches and trunks (
Lance et al., 2000). Field trials with fungal bands made with
M. anisopliae against
A. glabripennis in China demonstrate a faster mortality of beetles in the treated plots compared with the control plots (
Hajek et al., 2006).
Previous studies on the pathogenic mechanism shows that one of the reasons why the infected larvae died wasthe histopathologic dysfunction in the tissues and organs (
Huang et al., 2005;
Wang et al., 2005;
Zhang et al., 2007), but there has been no symptomatic and histopathological study of
A. glabripennis larvae infected with
Metarhizium Sorokin until now. In this experiment, we observed the symptoms and studied the sections through light microscopy of the infected larvae to study the histopathologic process, the method of infection and the speed of lesion development, all of which can provide a scientific basis for the pathogenesis of
Metarhizium Sorokin to determine the best time of use.
Materials and methods
Materials
Metarhizium Sorokin MS01 was selected from nine strains with high virulence against A. glabripennis larvae in the collecting field in 2006. A. glabripennis larvae were collected from the breeding adults in the laboratory, who laid eggs that were hatched in 90 days.
Methods
Treatment of test insects
The larvae of the same size were chosen to be immersed directly in the spore suspension of 1×108 conidia•mL-1 for 30 s so that spores could attach themselves to the surfaces of the larvae. Thereafter, the infected larvae were put into carved willow branches and sampled at 1, 2, 3, 4 and 5 days in turn.
Preparation of section for light microscopy
The samples were put into a fixative named Bragil Fluid with 1 g picric acid dissolved in 80% ethanol 150 mL, then 60 mL formaldehyde and 15 mL acetic acid were added. After being fixed for 24 h, the larvae were rinsed several times by 50% and 70% ethanol and stored in 70% ethanol. After a series of ethanol dehydration treatment, we used the conventional methods of section for light microscopy to cut the larvae into 8-12 μm slices. They were then observed and photographed with an Olympus optical microscope.
Observation of external symptom of infected larvae
For inoculation, each treatment of 20 larvae was immersed in a spore suspension of 1×106 conidia•mL-1 and 1×108 conidia•mL-1 for 30 s, and then moved to the carved willow branches. The treated insects were observed and photographed daily, with mortality and external characteristics recorded in the case of the larvae. The dead bodies were moisturized and cultured, to determine whether the larvae died of infection with MS01 when the post-mortem white mycelial and green conidia appeared.
Results
Observation of external symptom of A. glabripennis larvae infected with MS01
Early in the infection process, there was no significant change in either activities or feedings. However, 24-48 h after inoculation, the high concentration inoculation larvae changed in color gradually, and the soaked dark spots appeared in some parts (Fig. 1 (a)). A similar case occurred at 48 hpi in the low concentration inoculation larvae. This was because of the host reactions, including the accumulation of the corpuscle within the dermis, as well as the hypertrophy and melanin of the cuticle. Subsequently, the infected larvae reduced their feedings and became unresponsive, numb and died gradually at 72 hpi. The dead larvae with the high concentration inoculation became swollen and hard or rigid throughout the body because the mycelium absorbed the water in the body rapidly. Meanwhile, the dead body with the low concentration inoculation was a little softer initially, and gradually turned into a corpse. The mycelium first grew on the intersegmental membrane when moisturizing and culturing the two types of the dead bodies. Then, a white hyphal layer formed on the surface of the body (Fig. 1 (b)), and finally the whole body was covered by green conidia (Fig. 1 (c)).
Histopathologic changes of A. glabripennis larvae infected with strain MS01
Paraffin sections show the spores of MS01 attached to the partial cuticle of A. glabripennis larvae. The conidia germinated and penetrated through the cuticle, and the germ tubes passed through the epithelial layers into the body cavity, multiplied and invaded the tissues nearby the penetration point, such as the fat body, muscle, tracheal tissue, malpighian tubule, alimentary canal, etc. Most tissues and organs had histopathologic changes with the multiplication of the mycelium.
The conidia attached to the cuticle of
A. glabripennis larvae at 24 hpi, germinated and penetrated through the cuticle at 36 hpi, passing through the epithelial layers into the body cavity at 48 hpi, and the color of the penetration point deepened, resulting from the function of mechanical stress and enzyme, leading to degradation in the epicuticle of
A. glabripennis larvae (
Hajek and St Leger, 1994;
Vestergaard and Butt, 1999;
Wang et al., 2004;
Wang et al., 2008). With the germ tubes passing through continually, the endocuticle gradually filled with hyphae, and the protein-chitin composite decomposed, leaving the exocuticle alone at last, with the cuticle separated from the epidermis (Fig. 2 (b)). Mycelium began to damage the dermis at 72 hpi, trichogen cells disintegrated, and the bristle of infected larvae fell, showing trichopore (Fig. 2 (c)). The surface had visible black spots in the process of cuticle infection, resulting from the cuticle defense of phenoloxidase.
Trachea
The structure of a healthy trachea was regular and tubular, and there were taenidiums on the surface (Fig. 3 (a)). The structure of spiral-wound was destroyed in the latter infection. The taenidiums were damaged and the cavity emerged partially at 48 hpi (Fig. 3 (b)), then disappeared gradually. Most trachea were disintegrated at 96 hpi (Fig. 3 (c)).
Fat body
Healthy fat body was irregular, structural affinity and well-defined smooth verge was also visible (Fig. 4 (a)). It became significantly deformed, loose in structure, and vacuoled, with the edge incised at 72 hpi (Fig. 4 (b)), and then it became blurred, releasing fat particles like grapes. These particles were dissolved into fat cells and mixed gradually with debris net, and the fat body was covered up with hyphae at 120 hpi (Fig. 4 (c)).
Muscle
The healthy muscle tissue had a tight structure, shaped like a bar or a lump (Fig. 5 (a), (c)), while the muscle tissue became loose and cracked at 72 hpi, and then separated into fragments gradually (Fig. 5 (b), (d)), and finally disintegrated.
Malpighian tube
Healthy malpighian tubes were composed of filamentous processes separated from each other, showing small diameters (Fig. 6 (a)), while the infected malpighian tubes shows a prominent extermal perimeter at 72 hpi (Fig. 6 (b)), and was covered with hyphae at 96 hpi, showing larger diameters (Fig. 6 (c)).
Alimentary canal
The healthy alimentary canal comprises peritrophic membrane, surrounded by regular columned cells, clinging to the muscle of wall cells (Fig. 7 (a) and Fig. 7 (d)). When the hyphae passed through the hemocoel, the alimentary canal became susceptible to being destroyed. The peritrophic membrane of the infected larvae disappeared at 72 hpi (Fig. 7 (b)). Subsequently, they invaded the mid-gut epithelium directly, and the columned cells were disorganized in the alimentary canal at 96 hpi (Fig. 7 (e)). The columned cells became slack and fell off, with parts of them having vacuoles, and were separated from the muscle of wall cells at 120 hpi (Fig. 7 (c)). All of the alimentary canal was decomposed, and cracks emerged. Part of the blood cells flew into hemocoel followed by blood, and then the alimentary canal was gradually disorganized (Fig 7 (f)). The intestinal cavity became turbid at last.
Discussion
In our study, we found that the intersegmental membrane to be the first invasion site when
A. glabripennis was infected with
Metarhizium Sorokin, as the layer of chitin was thin in the intersegmental membrane, making it possible for germ tubes to pass through. These data supported the idea proposed by
Wang et al. (2004) that the head of
Plutella xylostella susceptible to
Metarhizium Sorokin may result from large quantities of intersegmental membrane in the head. When moisturizing and culturing the two types of the dead bodies, the white mycelium first grew on the intersegmental membrane, also further proving that the site was vulnerable to the invasion of
Metarhizium Sorokin. Symptoms appeared earlier in the high concentration inoculation larvae than in the low one in the treatment of immersion in spore suspension, demonstrating that the quantity of conidia attached to the cuticle may be associated with the toxicity of
Metarhizium Sorokin and the sensitivity of the host.
Based on the section for light microscopy, we knew the tissues in the whole body were damaged to different degrees, and obvious pathological changes took place. After the inoculation of insects for three days, a variety of solid organs were filled with hyphal bodies, and the external distribution had a larger density than the inner. Besides, there were two types of hyphal bodies: One was small granular or short-rod with a diameter smaller than blood cells, generally located in the hemolymph, while the other was a long filament, far larger than the length of the blood cells and was usually located with a solid-state organization around it. All these show that the reproduction of the hyphal bodies were mainly in the hemolymph of the larvae. Following the blood circulation, the small type of hyphal bodies then passed into the peripheral tissues and organs. With the growth of mycelium, the hosts died and the mycelium gradually penetrated into the inner lumen of the solid-state organs. This mode of infection was similar to the process of
Galleria mellonella infection with
Metarhizium anisopliae (
Prasertphon and Tanada, 1968). Although the direct cause of death was not clear yet, we could infer that most tissues and organs were damaged when the hyphae passed through, so that the normal physiological activities were disrupted. The larvae hemocoel was full of hyphae with the multiplication of the mycelium, leading to the lack of nutrients and blocking of fluid circulation, which caused physical hunger and metabolic disorders.
In the near-death larvae or 48 h after inoculation, a few number of hypha were filling the bodies, while a large mount were concentrated in hemocoel in the vicinity of the invasion site. Only a very small amount passed into the adjacent body fat. This reveals that the death of the larvae was not due to direct undermining of tissues and organs of the host. We can infer that the invasion of mycelium weakened the function of all the tissues and organs, and the infected larvae died of without normal physical activities. In addition, the physiological and biochemical changes may also be the cause of death when infected with
Metarhizium Sorokin. Ferron suggested that what really caused the larvae infected with imperfect fungi to die was a fungal toxin. The toxin could not only inhibit the immune function of the host, but also affect the central nervous system and cause some pathogenic effects, thereby speeding up the death process of the host (
Sloman and Reynolds, 1993). However,
Wang and You, (1999) believed that the pathogenic mechanism of these toxins was not very clear, only to find that the pure toxin had effects on immune pressure, muscle paralysis and malpighian tube damage Therefore, whether or not the infected
A. glabripennis larvae died of toxins is a question that needs further studies.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(434KB)
