Antibacterial activity of lysozyme in the desert locust , Schistocerca gregaria ( Orthoptera : Acrididae )

The ability of biocontrol agents to overcome the immune defense of pests is a crucial issue. This is the first study of lysozyme activity as an inducible humoral component of the defense of Schistocerca gregaria, which depends on the recognition of the elicitor molecules of pathogens and not on epidermal wounding or a spiking effect. The level of lysozyme activity in fat body, haemocytes and haemolymph plasma of naïve and immunologically challenged 5th instar S. gregaria was evaluated using the zone of inhibition test against Micrococcus lysodeikticus. Various Gram-positive and Gram-negative bacteria as well as peptidoglycans (PGN) and lipopolysacchrides (LPS) of bacterial cell walls induce and increase in the level of lysozyme activity. Escherichia coli induced an increase in the level of activity of lysozyme in the fat body, haemocytes and plasma, but not in mid gut epithelium, 6–12 h after an immunological challenge and then it decreased to the constitutive level after 72 h. This study revealed that in S. gregaria there is a constitutive and a bacteria-inducible level of lysozyme activity, which protects it against infection by both Gram-negative and Gram-positive bacteria. 559 * Corresponding author; e-mail: tahabasal@gmail.com the tissues where lysozyme is produced and clarify the time course in the change in activity of lysozyme. MATERIAL AND METHODS


INTRODUCTION
Insects lack an adaptive (acquired) immune system.They have an innate immune system that relies on germline encoded factors recognizing and killing foreign invaders, which effectively and rapidly protects them from infection caused by microorganisms and parasites (Tsakas & Marmaras, 2010).This system is highly developed and comprises cellular and humoral components that can be activated by the invasion of foreign bodies, including pathogens (Gillespie et al., 1997).Recognition by the host of pathogen-associated molecular patterns triggers signaling cascades, which activate immune cells and the transcription of antimicrobial peptides that trap or kill pathogens (Marmaras & Lampropoulou, 2009).
Both haemocytes and fat body, but primarily the latter, of various insects are reported to synthesize and release lysozymes into haemolymph (Lemaitre & Hoffmann, 2007).However, other tissues, such as epidermis, muscles and mid gut cells may also participate in lysozymes production (Lee & Brey, 1995;Hultmark, 1996).In Manduca sexta mid gut cells lysozymes are developmentally regulated independently of a bacterial infection (Russell & Dunn, 1991).
The pest in this study, S. gregaria (Forskål) (Orthoptera: Acrididae), like other locusts and grasshoppers, is abundant in dry grassland and desert (Lomer et al., 2001), and is the most damaging of the locusts (Lecoq, 2004).Microbial control measures coupled with immunosuppressive agents are needed as part of a control program for this pest.
To our knowledge, unlike for the well-studied holometabolous insects, there is little information on lysozyme as an inducible humoral component of the defense mechanisms in hemimetabolous insects, particularly the orthopteroids (Hoffmann & Brehelin, 1976;Hoffmann, 1980;Zachary & Hoffmann, 1984;Schneider, 1985;Adamo, 2004;Adamo et al., 2008;Srygley & Lorch, 2011).In locusts, the only known antimicrobial peptide (AMP) other than lysozyme is locustin (Swiss-Prot: P83428.1).Therefore, the aim of the present study is to estimate the antimicrobial activity of lysozyme in the plasma of naïve and bacterially challenged 5 th instar S. gregaria, identify the tissues where lysozyme is produced and clarify the time course in the change in activity of lysozyme.

Insect rearing
The 5 th instar nymphs of S. gregaria were obtained from a well-established laboratory colony at the Entomology Department, Faculty of Science, Cairo University, Egypt.There is a detailed description of this S. gregaria colony and the rearing technique in Maeno & Tanaka (2008).We used 5 th instar individuals because the immature stages of insects frequently become infected during the moulting process and in the insect studied the 5 th instar is the last nymphal instar prior to radical changes in the body, involving histogenesis and organogeny of the adult body.

Bacterial strains, culture and immune challenge
The bacteria, Escherichia coli JM109 (Fermentas, Canada), Micrococcus luteus (ATCC 49732) (Biomerieux, Canada), Neisseria gonorrhoeae (ATCC 19424)  Before the immunological challenge, the bacterial cells were pelleted by centrifugation at 2000 g for 10 min at 4°C and well washed with sterile Ringer's solution (113.7 mM NaCl, 1.9 mM KCl, 1.1 mM CaCl2, 0.12 mM NaHCO3, 0.07 mM Na2HPO4; pH 7.2) using centrifugation.The cells were pelleted and resuspended in sterile Ringer's solution.The cell concentration was adjusted to give a final concentration of 6 ×10 4 cells/ml and then immediately used in the experiments.
Fifth-instar S. gregaria, at mid-stadium age (day 5-7), were each injected with 20 µl of log-phase bacteria between the 2 nd and 3 rd abdominal sclerites.Gram-positive B. subtilis, M. luteus and S. aureus, and the Gram-negative E. coli, N. gonorrhoeae and P. aeruginosa were used in this study.Also, nymphs were immunologically challenged by injection of 20 µg/nymph of bacterial or fungal cell wall components (purchased from Sigma-Aldrich, unless otherwise indicated) like lipopolysaccharide (LPS) from Salmonella enterica serotype abortus equi [prepared by the PCP method of Galanos et al. (1979)], Laminarin from Laminaria digitata, soluble peptidoglycan (PGN) from S. aureus, chitin (from shrimp shells), muramic acid and glucosamine.

Collection and preparation of haemolymph and fat body and mid gut tissue
Haemolymph was obtained by puncturing the arthrodial membrane in the hind legs of naïve or immunologically challenged chilled nymphs with a sterile needle and the haemolymph collected in ice-cold tubes containing a few crystals of phenylthiourea (Fluka).To determine the time course activity of lysozyme, haemolymph was collected immediately after (control) or at different intervals (2, 4, 6, 12, 18, 24, 48, and 72 h) post-injection (immunologically challenged individuals).For each interval, haemolymph from 12 nymphs was pooled and used for measuring lysozyme activity (see below); each measurement was repeated 3 times.Plasma was obtained by centrifugation of haemolymph at 8000 g for 5 min at 4°C and stored at -20°C until used.
For preparation of haemocyte lysate supernatant, haemolymph from 25 nymphs per treatment (naïve; saline-or E. coli injected) was collected immediately after (control) or 6 h post injection (treated) and gently mixed with 300 µl ice-cold anticoagulant (62 mM NaCl, 100 mM glucose, 10 mM EDTA, 30 mM trisodium citrate and 26 mM citric acid, pH 4.6).The haemolymph was centrifuged at 8000 g for 5 min at 4°C.Pellets of haemocytes were washed twice with 500 µl of Ringer's solution and then re-suspended in 50 µl of lysis buffer (20 mM Tris-HCl, 137 mM NaCl, 2 mM EDTA, 0.1% Triton-X 100, 10% glycerol, pH 7.2).Haemocyte lysates were ultrasonicated for 1 min continuous pulse on ice and the cell debris removed by centrifugation at 10,000 g for 2 min at 4°C.The haemocyte lysate supernatant (HLS) was stored in aliquots at -20°C until used.
The level of induced lysozyme activity in different tissues was also compared quantitatively at 6, 12, 24, 48 and 72 h.Fat body and mid gut epithelium were collected and washed with 0.01 M sodium phosphate buffer (pH 7.0) and separately homogenized in 0.01 M sodium acetate buffer (pH 6.0) containing protease inhibitors (complete-protease inhibitor cocktail tablets; Roche).The supernatants, after centrifugation at 8000 g for 15 min at 4°C, were used for assaying lysozyme activity.The activity was corrected for tissue protein content, i.e. values were obtained by dividing activity by total protein.The concentration of protein was estimated (Bradford, 1976) using bovine serum albumin as a standard.Controls were: (i) not injected [naïve control], (ii) injected with sterile saline or Ringer's solution, and (iii) wounded by piercing the cuticle of the abdomen.

Determination of lysozyme activity using zone of inhibition test
Lysozyme activity was determined using a slightly modified form of the zone of inhibition test (Zachary & Hoffmann, 1984).Briefly, 5 µl of crude plasma or tissue aliquots were placed into 3 mm wells punched in 1% agarose plates.The plates were inoculated with 1 mg lyophilised M. lysodeikticus (ATCC 4698) cells (Sigma-Aldrich) per ml agarose.The activity was recorded by measuring the diameter (mm) of the clear zone after incubating the plates for 24-48 h at 37°C.The linear relationship between the diameter of the clear zone and the concentrations of a reference lysozyme, hen egg white lysozyme (HEWL) (Sigma-Aldrich, 40 000 units/mg) was established and used as a reference for expressing the activity of the different samples in terms of µg HEWL.

Statistical analyses
The time course of the activity of lysozyme in different tissues was analyzed using a one way analysis of variance (ANOVA).Duncan test was used to determine the homogeneity of the means.Statistical analysis was performed using SPSS software (version 15; SPSS, Chicago, IL).Data are expressed as means ± SD.

Constitutive lysozyme activity
From the standard curve (Fig. 1), the diameter of the clear zone, which ranges from 5 to 14 mm, is proportional to the logarithm of enzyme concentration, as previously described (Boman, 1994).The coefficient of determination, R 2 , of 0.99 of this relationship indicates this assay gives accurately reproducible results.
The activity of lysozyme in the pooled sample of haemolymph plasma of the insect used in this study, as in other studies, is considered to be an accurate measure of its titer in the whole insect body.This is because haemo-lymph circulates throughout an insect's body and therefore where the lysozyme is most likely to collect after release from the tissues in which it is synthesized.Table 1 gives the constitutive level of lysozyme activity in the haemolymph plasma of naïve 5 th instar S. gregaria as about 80 µg of HEWL/ml.This is similar to that recorded in the hemolymph of several hemimetabolous [Locusta migratoria (Zachary & Hoffmann, 1984) and Gryllus bimaculatus (Schneider, 1985)] and holometabolous [Apis mellifera (Mohrig & Messner, 1968), Galleria mellonella (Powning & Davidson, 1973;Fiolka, 2012), Spodoptera eridania (Anderson & Cook, 1979), Bombyx mori (Morishima et al., 1995), and Samia cynthia ricini (Fujimoto et al., 2001)] insects.This value is considered to be the background level of activity and the initial immune response to invading microorganisms.For instance, it is believed to degrade the bacterial debris released during the initial cellular immune response (Park et al., 1997).
Sequential immune responses, such as the synthesis and secretion of lysozymes and other antimicrobial peptides, are thought to be elicited by PGN fragments resulting from the lysis of microbes (Kaneko et al., 2004;Park et al., 2007).This is thought to be a key-step in pattern recognition, which is followed by transduction of signals and initiation of antimicrobial responses (Tsakas & Marmaras, 2010).

Induced lysozyme Inducers and induction
The activity of lysozyme in the body of a 5 th instar S. gregaria was elevated by injecting certain elicitors into its haemocoel.These elicitors (Table 1) included whole cells (of log phase) of different Gram-positive and Gramnegative bacteria, or certain components of bacterial or fungal cell-walls.The level of the induced lysozyme activity in haemolymph plasma of immunized nymphs is significantly dependent on the elicitor used (Table 1).
Firstly, the level of lysozyme activity induced by whole bacterial cells depends on the species of bacteria.For E. coli, in particular, the proportionality test indicates that there is a positive linear correlation between the induction of lysozyme activity and doses of up to 800 bacterial cells/nymph and then it remains fairly constant (Fig. 2).

561
a Haemolymph was sampled 6 h post-injection (the time of maximum activity); b lysozyme activity was measured in terms of µg/ml of HEWL; c injected 1200 cells in 20 µl saline/nymph; d injected 10 µg in 20 µl saline/nymph; e each value is the mean ± SD of the results of three different experiments (n = 3).Lytic activity e (µg/ml) Inducer TABLE 1. Levels of haemolymph a lysozyme lytic activity b recorded in 5 th instar S. gregaria induced by different species of bacteria, their cell wall components and non-microbial sugars.Activity was assessed using zone of inhibition test against M. lysodeikticus.This indicates that E. coli can elicit increased lysozyme activity in this insect.Therefore, in the present study, E. coli was used as a general elicitor of lysozyme activity.Similar dose-dependent inductions of lysozyme activity by E. coli are also reported in other insects, e.g.Spodoptera eridania (Anderson & Cook, 1979), B. mori (Abraham et al., 1995) and Gryllus texensis (Adamo, 2004), and E. coli is also considered to be an effective inducer of insect humoral immune responses in general (Erler et al., 2011) and of hemolymph lysozyme activity, in particular, e.g.B. mori (Sumida et al., 1992;Abraham et al., 1995), Melanoplus sanguinipes (Gillespie et al., 1993), Artogeia rapae (Bang & Yoe, 2005), S. frugiperda (Chapelle et al., 2009) and Anopheles gambiae (Kajla et al., 2010).Also, there is no clear-cut specificity in terms of pathogen recognition, signaling pathways, or effector molecules in the course of induction of humoral responses in insects (Marmaras & Lamproulou, 2009).
Secondly, the macromolecules PGN and LPS in bacterial cell walls are shown to be strong inducers of lyso-zyme activity in this hemimetabolous insect (Table 1).This has similarly been reported in holometabolous insects, e.g., PGN in M. sexta (Kanost et al., 1988), B. mori (Morishima et al., 1995) and S. cynthia ricini (Fujimoto et al., 2001), and LPS in S. eridania (Anderson & Cook, 1979), Heliothis virescens (Shelby et al., 1998) and G. mellonella (Fiolka, 2012).However, it is uncertain which of the molecular components of these conjugated carbohydrates (PGN and LPS) act as elicitors (Marmaras & Lamproulou, 2009).The macromolecular fragments and even the muropeptides of PGN are reported (Kaneko et al., 2004;Park et al., 2007) to be elicitors of antimicrobial peptide synthesis and secretion.However, in the present study, molecules below this size, i.e. the mono sugar mother-molecules of PGN glucosamine and muramic acid did not elicit a response (Table 1).On the other hand, the inductive  Lysozyme activity was assayed using the zone of inhibition test against M. lysodeikticus and expressed in terms of µg HEWL/mg protein.Means ± SD; n = 3.For a given treatment, means in rows with the same superscript letter are similar (insignificantly different; P > 0.05) whereas, those with different letters are significantly different; P < 0.05).ND -not recorded.2. Lysozyme activity recorded in haemocytes, plasma, fat body and mid gut of naïve and 5 th instar S. gregaria that were injected with saline or E. coli, respectively.Lysozyme activity in plasma and haemocytes was measured after different intervals of time of 6 to 72 h.effect of LPS is uncertain because relatively crude preparations of these molecules were used in the present study (Table 1) and previous reports have indicated it is an elicitor (Kaneko et al., 2004); whereas highly purified LPS has little or no effect (Leulier et al., 2003;Kaneko et al., 2004).
In contrast, both the polymers -1,3-glucan laminarin and -1,4-glucan chitin have no obvious eliciting effect.Also, the injection of saline did not affect the level of lysozyme activity, which excludes a spiking effect and regulatory mechanisms associated with the wounding and disintegration of the epidermis, as previously reported for the pro-phenoloxidase cascade (Cerenius & Söderhäll, 2004).

Time course of the increase in the activity of lysozyme
There was a change over time in the level of lysozyme activity in the haemolymph of 5 th instar S. gregaria following injection with E. coli (Fig. 3).The maximum level of activity was recorded 6 h post injection.The induced level was maintained for between 6 and 12 h and then declined to the constitutive level 72 h post injection.This may indicate that the duration of the elicitor effect only lasts for this interval of time.In B. mori (Abraham et al., 1995) and Spodoptera litura (Kim & Yoe, 2003) it is reported that lysozyme activity reaches a maximum at 48 h and is slightly lower 72 h post injection.In G. mellonella immunized with LPS, however, the maximum lysozyme activity is attained 24 h post injection (Fiolka, 2008).Therefore, it is important to conduct further research on lysozyme activity in S. gregaria.

Site of production
As potential sites for the biosynthesis of lysozyme, its activity was determined separately in haemocytes, fat body and mid gut tissue of 5 th instar S. gregaria (and the activity level was considered to correspond to its rate of synthesis, neglecting the rate of release into the haemolymph) before and after an immunological challenge with E. coli (Table 2).The data show that before the challenge, the lysozyme is constitutively synthesized at comparatively low levels in both haemocytes and fat body, but not in mid gut tissue.After the challenge, the time-course in the level of activity of lysozyme in these two tissues is similar to that recorded in haemolymph plasma.
Haemocytes and fat body are the two main immunityconferring tissues, which are collectively responsible for the production and secretion of antimicrobial agents (Tsakas & Marmaras, 2010).Also, they are the main sites of lysozyme synthesis and secretion (Zhang et al., 2009).The results indicate that the level of lysozyme activity in haemocytes is higher than in the fat body of 5 th instar S. gregaria (Table 2).This may also be the case for lysozyme activity in hemimetabolous insects such as L. migratoria (Zachary & Hoffmann, 1984), Rhodnius prolixus (Azambuja et al., 1999), Teleogryllus commodus (Drayton & Jennions, 2011) and other AMPs (Bulet & Stöcklin, 2005).
A comparison of the lysozyme activity in plasma and in haemocytes at two successive intervals following an immunological challenge (Table 2) revealed that it is lower in the plasma than haemocytes between 6-24 h.This is also recorded for other hemimetabolous insects such as L. migratoria (Zachary & Hoffmann, 1984) and Blaberus discoidalis (Wilson & Ratcliffe, 2000).Later, i.e. 48-72 h post-challenge, the level of lysozyme activity in both haemocytes and plasma was similar, possibly due to pooling in the plasma of lysozyme from different tissues and probably also to a decline in the level of synthesis in haemocytes; and the level recorded in un-challenged insects was similar to that recorded 72 h post-challenge (Table 2).It is possible that the level of lysozyme activity recorded in plasma is an over-estimate of the actual level in vivo, because a small degree of lysis certainly occurred due to the fragility of the haemocytes.
In summary, S. gregaria has a constitutive level of lysozyme activity and the de novo induction of its biosynthesis in haemocytes, fat body and probably other body tissues follows a signal transduction mechanism when the associated molecular patterns of the PGN and LPS of an invading pathogen are recognized.The induced increase in the level of activity of lysozyme reached a maximum within the first few hours of a pathogen being recognized and then decreased to the constitutive level after 72 h.There was no evidence to indicate that these responses were a result of regulatory mechanisms associated with wounding of the epidermis and disintegration of cells or even a spiking effect.

Fig. 1 .
Fig. 1.The relationship between the activities of lysozyme recorded using the zone of inhibition test and the logarithm of the concentration of lysozyme.

Fig. 2 .
Fig. 2. The relationship between the activities of lysozyme in plasma of 5 th instar S. gregaria induced by the injection of different doses of E. coli contained in 20 µl of Ringer's solution.

Fig. 3 .
Fig. 3.The concentration of lysozyme recorded in injected and control 5 th instar S. gregaria at different times after injection with log-phase E. coli (6 × 10 4 cells in 20 µl saline).Bars marked with the same letters are similar (insignificantly different; P > 0.05) whereas, those with different letters aren't (significantly different; P < 0.05).