- Email: [email protected]
Epigenetic transmission of phase in the desert locust, Schistocerca gregaria: Determining the stage sensitive to crowding for the maternal determination of progeny characteristics
Journal of Insect Physiology 56 (2010) 1883–1888
Contents lists available at ScienceDirect
Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys
Epigenetic transmission of phase in the desert locust, Schistocerca gregaria: Determining the stage sensitive to crowding for the maternal determination of progeny characteristics Koutaro Maeno, Seiji Tanaka * Locust Research Laboratory, National Institute of Agrobiological Sciences at Ohwashi (NIASO), Tsukuba, Ibaraki 305-8634, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 July 2010 Received in revised form 6 August 2010 Accepted 6 August 2010
Desert locust female adults respond to crowded conditions by changing progeny characteristics such as egg size, clutch size (no. of eggs per pod), hatchling body size and coloration. This study was conducted to determine the stage sensitive to crowding in this locust. Reproductively active females reared in isolation increased egg size and decreased clutch size and the proportion of green hatchlings after exposure to crowded conditions (in which each female was kept with four male adults). These changes depended not only on the timing of exposure to crowded conditions during the reproductive cycle but also on the length of the exposure. By varying the time and length of the exposure, it was found that crowding had no influence on progeny characteristics during the last two days of egg development at 31 8C and that there was a four-day sensitive stage before this period. The sensitive stage coincided with the time when the affected oocytes were 1.5–4 mm long, while the sensitivity to crowding appeared to be constant over the sensitive stage. The larger the magnitude of the increase in egg size after exposure to crowding, the smaller the proportion of green hatchlings (and the larger the proportion of gregarized dark hatchlings); there was a sigmoidal relationship between the two variables. Based on these results, we propose a model for determining the stage sensitive to crowding in both the female parent and the oocytes. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Crowding Density-dependent phase polyphenism Maternal effect Locusts Schistocerca gregaria
1. Introduction Although the desert locust, Schistocerca gregaria, usually occurs at a low density and seldom causes any economic damage, outbreaks of this locust are extremely destructive. As a result, this locust is one of the most destructive pests in the Old World (Uvarov, 1966, 1977). Numerous studies reported on various aspects of the biology of this locust, particularly since Uvarov (1928, 1966, 1977) proposed the phase theory in locusts (Dale and Tobe, 1990; Pener, 1991; Applebaum and Heifetz, 1999; Hassanali et al., 2005; Tanaka, 2006; Pener and Simpson, 2009). Locusts show remarkable phenotypic plasticity termed ‘‘phase polyphenism’’ in which morphological, physiological and behavioral traits change in response to local population density. At low population densities, juvenile hoppers are green or brown and adults are sedentary (solitarious forms). As the population density increases, the locusts become more gregarious, their body color becomes darker, and adults often migrate over a long distance in large swarms (gregarious forms). A complete change from the solitarious to the
* Corresponding author. Tel.: +81 29 838 6110; fax: +81 29 838 6110. E-mail address: [email protected] (S. Tanaka). 0022-1910/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jinsphys.2010.08.010
gregarious phase does not occur in one generation, but requires several generations. Locusts in the intermediate phase are called transient forms (Uvarov, 1966, 1977; Pener, 1991; Pener and Simpson, 2009). The trans-generational transmission of phase is an epigenetic phenomenon and has attracted much attention (Uvarov, 1966; Pener and Simpson, 2009). However, the details of the mechanism controlling this phenomenon remain largely unknown. It has recently been demonstrated in the laboratory that S. gregaria requires two generations to undergo a complete shift in morphological characteristics from the solitarious to the gregarious phase and vice versa (Maeno and Tanaka, 2009a). Adults change the quality and quantity of the progeny quickly in response to local population density (Maeno and Tanaka, 2008b, 2009b). Hatchlings from crowd-reared females are larger and darker than those from isolated-reared females, and they develop into adults with morphological characteristics typical of gregarious forms if the hatchlings are also grown under crowded conditions. However, such individuals show intermediate values between the two extreme phases if they are maintained in isolation during nymphal development. Similar adult characteristics of the intermediate (transient) phase are also observed if hatchlings from isolatedreared females are reared under crowded conditions, whereas such hatchlings become adults with characteristics typical of the
1884
[(Fig._1)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
solitarious form if the hatchlings are also reared in isolation. Therefore, the maternal determination of hatchling characteristics plays an important role in the trans-generational changes in phase characteristics in this locust. The present study aimed to determine which stage was sensitive to a crowding stimulus in S. gregaria female adults. This information is essential for understanding the physiological, endocrine and molecular mechanism of the transgenerational transmission of phase characteristics in the future. In S. gregaria, large, black hatchlings are produced if the female parent is exposed to crowding, whereas small, green hatchlings are obtained if the female parent experiences isolation or low-density conditions (Faure, 1932; Hunter-Jones, 1958). The crowding conditions experienced as adults are important in the control of progeny characteristics (egg size, hatchling body size and coloration; Hunter-Jones, 1958; Maeno and Tanaka, 2008a; Tanaka and Maeno, 2006, 2008). In their review article, Pener and Simpson (2009, p. 212) discussed the effects of a gregarizing agent derived from the reproductive accessory glands after ovulation, first in the oviduct and then after oviposition based on the studies of McCaffery et al. (1998) and others. The authors concluded that exposure to this gregarizing agent causes hatchlings to develop gregarious coloration and behavior. However, more recent studies have suggested that the ovariole is the site where egg size, hatchling body size and coloration are pre-determined and that no pheromonal factor is involved (for a review, see Tanaka and Maeno, 2010). The present study examined the effects of maternal crowding on the progeny characteristics in solitarious (isolatedreared) female adults. The results indicated that for this locust there is a narrow window of time in which a specific stage is sensitive to a crowding stimulus in each reproductive cycle. This paper describes the results of these observations in relation to oocyte growth. We hypothesize that the female parent might produce a hormonal factor that controls egg size in response to crowding conditions and that oocytes at a particular stage respond to this factor. 2. Materials and methods 2.1. Insects The S. gregaria laboratory colony used in this study was originally from Ethiopia, and the rearing and handling methods used here have been described previously (Tanaka and Yagi, 1997; Maeno and Tanaka, 2008a). Solitarious (isolated-reared) females, also visually isolated from one another, were used in all experiments and were reared individually in small cages (28 cm 15 cm 28 cm), except for a short period to allow mating during the adult stage, for two or more generations. The rearing room was well ventilated, but whether or not the locusts were also completely isolated pheromonally was unknown. Both nymphs and adults were fed orchard grass, cabbage leaves and bran. Each female adult was allowed to deposit into sand in a transparent plastic cup (340 ml volume) so that deposited egg pods could be seen through the bottom of the cup without having to dig into the sand. All locusts, including test insects, were maintained at 31 1 8C with a light–dark (LD) cycle of 16:8 h. 2.2. Determination of the stage sensitive to crowding Each female reared in isolation was paired with a sexually mature male 14–18 days after adult emergence. All males used in the experiment had been reared in a group because their phase status has no influence on the progeny characteristics examined (Maeno and Tanaka, unpublished observations). When a female is paired with a male for a long period of time in the laboratory, the female can be stimulated to produce offspring with characteristics
Fig. 1. Experimental design used to determine the stage sensitive to crowding in isolated-reared Schistocerca gregaria female adults at 31 8C. Isolated-reared female adults were exposed to crowding conditions (four male adults) at different lengths of time after the deposition of an egg pod (pod no. 1) for various periods and were then returned to isolated conditions. All females were allowed to deposit at least two egg pods before being used for the experiment.
of the gregarious forms (Hunter-Jones, 1958). Therefore, the pairing period was limited to less than 24 h. Mated females were allowed to deposit at least two egg pods before being used for experiments because substantial numbers of dark hatchlings are typically produced from the first egg pod even under continuously isolated conditions (Maeno and Tanaka, 2008b). Most female adults deposited egg pods every four days under the conditions used in this study (mean SD = 4.3 1.1 days, n = 129; Maeno and Tanaka, 2009b), with the time period ranging from three to six days. In the present study, oviposition was checked once a day and only those females that showed ovipositional intervals of four days during crowding treatments were used. To determine the stage sensitive to crowding, each mated female was held in a small cage with four sexually mature males one, two or four days after the deposition of an egg pod; the females were later re-isolated (Fig. 1). In this locust, the pairing of a female with a single male induces crowding effects on the progeny that are as strong as those observed when the female is reared with many males, and crowding effects can be elicited even when an isolated female is reared with females (Hunter-Jones, 1958). Males introduced were often found mating with the test females. Our unpublished data (Maeno and Tanaka) indicate that the crowding effects on solitarious females were equally elicited by sexually active males and females in terms of progeny characteristics such as body size and coloration. HunterJones (1958) demonstrated that solitarious females reproducing parthenogenetically switched from green-hatchling producers to dark-hatchling producers after being kept with other female adults. In the present study, only males were used to create crowding, because females would deposit egg pods, making it difficult to distinguish these egg pods from those produced by test females. Some females were kept in isolation or with four males continuously as controls. After the first egg pod was deposited at the beginning of the experiment, a maximum of four more egg pods were collected consecutively from each female and incubated at 31 1 8C. Two days after deposition the number of eggs in each pod was recorded and ten eggs were randomly chosen from each pod to measure their lengths to the nearest 0.1 mm with an ocular micrometer in a microscope. These and the other eggs were again incubated until hatching. The hatchlings of some egg pods were categorized into five hatchling color groups (HCGs 1–5) based on the darkness of their coloration as follows (Maeno and Tanaka, 2007): hatchlings in HCG 1 are green, as typically observed in solitarious forms; those in HCG 5 are almost completely black, as observed in gregarious forms; and those in HCGs 2–4 are intermediate in color.
[(Fig._2)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
1885
Fig. 2. Egg lengths (mean SD), proportions of hatchlings with different body colorations and clutch sizes (mean no. + or SD) of eggs per pod) of egg pods produced by isolatedreared Schistocerca gregaria female adults after different treatments at 31 8C. For explanation of the different treatments, see Fig. 1. Hatchlings were divided into five hatchling color groups (HCGs 1–5) based on the darkness of their coloration according to Maeno and Tanaka (2007), as follows: hatchlings in HCG 1 are green, as typically observed in solitarious forms, those in HCG 5 are almost completely black, as observed in gregarious forms, and those in HCGs 2–4 are intermediate in color. The numbers in parentheses indicate sample sizes. Asterisks above the bars indicate that the value was significantly different from that of the first egg pod in each panel using a t-test; **P < 0.01; ***P < 0.001. The data (proportions of green hatchlings) for hatchling body coloration were arcsine-transformed before being evaluated using a t-test.
2.3. Oocyte growth during reproductive cycles
3. Results
At the first, second, third, or fourth day after an egg pod was deposited, five to eleven females were sacrificed to measure the lengths of oocyte in the ovarioles. The ovary consists of functional and unfunctional ovarioles (Uvarov, 1966). Because the functional ovarioles of the same ovary looked similar to one another in sizes of the growing oocytes, one functional ovariole was removed from each dissected female and the lengths of the largest three oocytes were measured with an ocular micrometer in a microscope.
3.1. Effects of long periods of exposure to crowding
2.4. Statistical analysis The egg lengths and clutch sizes of egg pods were compared with the values for the first egg pod (the egg pod deposited at the beginning of the experiments) using a t-test after differences among pods were analyzed using an ANOVA. Likewise, the proportions of green hatchlings were compared using a t-test or an ANOVA after arcsine transformation of the data.
Females constantly kept in isolation (treatment I, Fig. 1) produced similar numbers of eggs and eggs of similar size in all five egg pods collected during the experimental period (Fig. 2A and K). Most hatchlings from those eggs were green (Fig. 2F). Isolatedreared females that were kept with males after the deposition of an egg pod (treatment II) produced significantly larger eggs in the second egg pod compared to the first egg pod, and the proportion of green hatchlings (HCG 1) was greatly decreased in the second egg pod (Fig. 2B and G). The mean egg length further increased and reached beyond 7 mm, and the proportion of green hatchlings further decreased in the third egg pod. However, no significant differences were observed in either parameter among the last three egg pods (P > 0.05; ANOVA for egg length; P > 0.05; x2 test for the proportion of green hatchlings), indicating that crowding effects were fully expressed by the third egg pod. Upon exposure to crowding, the clutch size (the number of eggs per pod) decreased,
1886
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
and there was a significant difference between the first and third or subsequent egg pods (Fig. 2L). Mean clutch size, which decreased significantly to 79.1 (SD = 15.0) eggs in the third egg pod, was further decreased to 71.4 (SD = 18.7) eggs in the fourth egg pod (P < 0.05; t = 3.61; df = 26), but no further reduction was observed in the fifth pod (73.4 18.1 eggs; P > 0.05; t-test; Fig. 2L). This result indicated that changes in clutch size occurred later than egg length. Females exposed to crowding for six days, which covered oneand-a-half inter-ovipositional periods (treatment III), exhibited results similar to those in treatment II for the first three egg pods (Fig. 2C, H and M). However, crowding effects seemed to disappear thereafter, causing all parameters to shift towards the values observed before the females were exposed to crowding. It took two reproductive cycles or egg pods to shift from one extreme to the other. Similar results were observed for egg length and hatchling body coloration when the females were exposed to crowding for four days, covering one ovipositional interval (treatment IV; Fig. 2D and I). In this case, however, egg lengths were similar in the second and third egg pods but remained relatively small. This result suggests that the first two days of crowding influenced the second egg pod, while the following two days influenced the third egg pod. No significant change was detected in clutch size (Fig. 2N). A fourday exposure to crowded conditions resulted in different results when this period began two days before the deposition of a second egg pod (treatment V). In those females, crowding did not influence the second egg pod, indicating that all parameters, including egg length, clutch size and hatchling body coloration, had been determined two days before oviposition. Crowding effects were detected in the third egg pod in all parameters (Fig. 2E, J and O). In the third egg pod, egg length increased to approximately 7 mm (Fig. 2E), the majority of hatchlings were in HCG 5 (Fig. 2J) and the clutch size decreased significantly (Fig. 2O). These results indicate that the period during which locusts were sensitive to crowding (i.e. the period in which progeny characteristics are determined in each egg pod) is approximately four days long, and begins six days before the egg pod is deposited under the present conditions.
[(Fig._3)TD$IG]
In treatments IV and V, the fourth egg pod produced fairly dark hatchlings, although the hypothesized four-day period of sensitivity to crowding would have ended two days before the deposition of the third egg pod. This relatively high incidence of dark hatchlings is likely related to difficulties experienced in staging female adults in these treatments: the females were kept with males two days after the deposition of the second egg pod, but the accuracy of this staging was 24 h. Therefore, the crowding period likely extended to the sensitive stage of the fourth egg pod in some female adults. Another possible factor that might explain this trend was the increased mortality of adults towards the end of the experiment. 3.2. Effects of short periods of exposure to crowding To determine if the sensitivity to crowding varies during the four-day sensitive period, isolated-reared female adults were exposed to crowding during the first half (two days prior to deposition of the second egg pod; treatment VI) or the second half (two days after deposition of the second egg pod; treatment VII) of the sensitive stage. The two treatments showed similar patterns in terms of the changes in egg length, hatchling body coloration and clutch size after a two-day crowding period (Fig. 3A, B, D, E, G and H). No significant difference between the two treatments was found in any of these parameters for the third egg pod (P > 0.05), indicating that the sensitivity to crowding was similar between the first and second halves of the sensitive stage. To determine whether there were cumulative effects of crowding over successive reproductive cycles, isolated-reared females were exposed to crowing during the first half of the sensitive stage for two consecutive clutches (treatment VIII; Fig. 1). The results showed that eggs of both the second and the third egg pods became significantly larger than those of the first egg pod, but remained relatively small (Fig. 3C). The mean lengths of the second and third egg pods were almost identical (P > 0.05; t-test). More than 50% of the hatchlings from the fourth egg pod were dark in
Fig. 3. Egg lengths (mean SD), proportions of hatchlings with different body colorations and clutch sizes (mean no. of eggs per pod + SD) of egg pods produced by isolated-reared Schistocerca gregaria female adults after a two-day exposure to crowded conditions during the sensitive stage at 31 8C. For an explanation of treatments, see Figs. 1 and 2. The data (proportions of green hatchlings) for hatchling body coloration were arcsine-transformed before being evaluated with a t-test.
[(Fig._4)TD$IG]
[(Fig._5)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
Fig. 4. The mean proportions (+ or SD) of green hatchlings of Schistocerca gregaria plotted against the magnitudes of changes in egg length between the first egg pod and the subsequent egg pods produced during the experiments shown in Fig. 1. Asterisks indicate significant differences from the proportion of green hatchlings from eggs with the difference in egg length being 0 mm by a t-test; ***P < 0.001. The data (proportions of green hatchlings) for hatchling body coloration were arcsinetransformed before being evaluated with a t-test. Bars indicate one SD. Numbers in parentheses indicate sample sizes (no. of egg pods).
color and the proportion of green hatchlings was significantly smaller than that of the first egg pod (Fig. 3F). As in treatments IV and V, this result is probably related to the method of staging the females. No significant change was observed in the clutch size (Fig. 3I). 3.3. Relationships between the magnitude of increase in egg length and the proportion of green hatchlings The above results showed that egg length increased relative to the length of the first egg pod after the female parent was exposed to crowding, and that the magnitude of the increase depended on when and how long the female was exposed to crowding. Fig. 4 illustrates that there was a sigmoidal relationship between the magnitude of the increase in egg length and the proportion of green hatchlings in each pod in the above experiments (R2 = 0.982; n = 435). The proportion of green hatchlings gradually decreased as egg size increased, and a significant decrease was observed when the magnitude of the increase in egg length reached 0.3 mm. The proportion of green hatchlings decreased further, and most hatchlings became black when the magnitude of the increase in egg length is over 0.7 mm. Eggs from some pods deposited by females before they were exposed to crowding conditions decreased slightly in length. These egg pods predominantly produced green hatchlings, but there was no significant difference in the proportion of green hatchlings between these egg pods (P > 0.05; ANOVA). 4. Discussion The present study confirmed that the phase-related body-color polyphenism in hatchlings of S. gregaria is closely correlated with the egg size, which is determined in the ovarioles of the female parent. Egg size was found to be the first variable affected after the females experienced a shift in population density, followed by changes in clutch size (the number of eggs per pod; Tanaka and Maeno, 2010). This study also demonstrated that there is a short period in each reproductive cycle during which females are sensitive to population density, which determines the egg size and hatchling body coloration. Fig. 5 summarizes the responses of S. gregaria female adults reared in isolation to crowded conditions. The rearing density
1887
Fig. 5. A schematic diagram of the stage sensitive to crowding for the maternal determination of egg length and hatchling body coloration in the largest oocytes of a Schistocerca gregaria ovariole. Crowding was found to have no effect on the progeny characteristics during the last two days of oocyte growth (oocyte A) and the sensitive period occurs when oocytes are 1.5–4 mm long in the ovary (oocytes B and C). Crowding induces a maximum response when it occurs during the entire period of the sensitive stage, while it elicits a significant but small response when it occurs only during the first or second half of the sensitive stage. Crowding occurring during the first half of the sensitive stage over two consecutive reproductive cycles also induces a significant but small response, indicating that there is no cumulative effect of crowding over successive reproductive cycles. Responses are based on increases in egg length in Figs. 2 and 3; , no change; +, 0.30–0.69 mm; ++, 0.7 mm. The yellow, green and blue areas indicate the sensitive stage for oocytes A, B and C, respectively, and the pink area indicates the range of oocyte length in which egg size is modified by crowding.
during the last two days before oviposition had no influence on the progeny characteristics observed. This finding was consistent throughout the present study. Maternal density apparently has no effect on egg length once oocytes in the ovarioles attain a length of more than 4 mm. Significant and marked effects of maternal density were observed when crowding occurred during a four-day period prior to the last two days of oocyte development under the present conditions. This is the stage at which females are sensitive to crowding, and that determines the egg length and hatchling body coloration in each egg pod. This period coincides with the stage when oocytes are 1.5–4 mm in length. The mother’s sensitivity to crowding appears to be constant throughout this period because females responded to crowding in a similar way regardless of whether they experienced crowded conditions during the first or second half of the sensitive stage. However, females exposed to crowding during the entire period of the sensitive stage responded more strongly by producing larger eggs that gave rise to predominantly black hatchlings. The eggs produced by females exposed to crowding over two consecutive reproductive cycles but only during the first half of each sensitive period (treatment VIII) remained in an intermediate size. This result indicates that there is no cumulative effect of maternal density across consecutive reproductive cycles. Hunter-Jones (1958) was the first to demonstrate that adult rearing density plays an important role in determining the hatchling body coloration in S. gregaria, although at that time it was unclear which stage of the adult life cycle was sensitive to crowding. His observation was confirmed by Islam et al. (1994a,b) who exposed isolated-reared and crowd-reared adults to crowded or isolated conditions at mating and oviposition. The latter studies reported a highly significant effect of density at ‘oviposition’. Crowd-reared females isolated at oviposition produced offspring
1888
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
that were more likely to be green in color than black, but isolatedreared females crowded at oviposition did not produce black offspring (Islam et al., 1994a; Fig. 2e and f). In another study by Islam et al. (1994b) (Fig. 5), adults kept under crowded conditions at mating and oviposition produced offspring that were similar in hatchling body coloration to those produced by their counterparts that were kept in isolation during mating and oviposition. These results are difficult to reconcile with the present results which suggest the presence of a sensitive stage in each reproductive cycle. However, the differences between these studies may be explained as follows: in the studies of Islam et al. (1994a), the reproductive cycle was not considered. The period of time for which females were exposed to crowded or isolated conditions at oviposition was not described and might have varied among the females tested because the reproductive cycle was ignored. In another study (Islam et al., 1994b), female adults were kept in a group for mating and ovipositing for a week. However, sexually mature females often deposit an egg pod as soon as mating takes place (Maeno and Tanaka, unpublished observations), and as a result, some egg pods might have been deposited earlier than others during that week. Islam et al. (1994a,b) used extremely small sample sizes and also appeared to use the first egg pods alone. First egg pods are not suitable for observations of this kind because the proportion of green or black hatchlings from such egg pods varies from 0 to 100% in this locust (Maeno and Tanaka, 2008b; Tanaka and Maeno, 2008, 2010). In a model constructed to explain the trans-generational transmission of phase in S. gregaria (Tanaka and Maeno, 2010), hatchling body coloration and body weight are thought to be predetermined in the ovarioles of the mother. The observed crowding stimulus is transmitted to the brain which causes the factor controlling egg size to be released. If this factor is assumed to increase egg size and the effective organ of this factor is the ovary, an increased release of this factor would be expected when the female adult is crowded longer during the sensitive stage, and all oocytes growing in each ovariole would be exposed to it. However, the present study indicated that only oocytes of a certain size show the effects of crowding experienced by the mother, while those that are too small or too big do not (Fig. 5). One possible explanation for this phenomenon is that the mother releases the egg-size controlling factor in response to crowding at any time and that oocytes can respond to this factor only during a limited period when they are 1.5–4 mm long. If egg size depends on the time at which the chorion is formed, the follicle cells might be the cells that respond to the egg-size controlling factor and control the timing of chorionation or egg size. Although nothing is known about the nature of the factor involved, the present study suggests that the physiological process determining egg size and hatchling body coloration is considerably complex. The present study showed a high correlation between the magnitudes of the increase in egg length after exposure to crowding and the proportions of green hatchlings. Eggs that increased by 0.3 mm or more showed a significantly reduced proportion of green hatchlings. This observation may be useful for establishing a convenient assay system to determine the stimuli inducing crowding effects and the site where such stimuli are perceived. A study investigating these effects is currently underway.
Acknowledgements The authors thank Ms. Hiroko Ikeda for her assistance in the laboratory and Dr. Ken-ichi Harano for his cooperation and for stimulating discussions. This study was funded in part by the Japan Society for the Promotion for Science for a research fellowship for PD (No. 20-2008) to Maeno and a Kakenhi grant of Japan (No. 19380039) to Tanaka. The grass used in this experiment was raised by the Field Management Section of NIASO. References Applebaum, S.W., Heifetz, Y., 1999. Density-dependent physiological phase in insects. Annual Review of Entomology 44, 317–341. Dale, J.F., Tobe, S.S., 1990. The endocrine basis of locust phase polymorphism. In: Chapman, R.F., Joern, A. (Eds.), Biology of Grasshoppers. John Wiley and Sons, New York, pp. 393–414. Faure, J.C., 1932. The phases of locusts in South Africa. Bulletin of Entomological Research 23, 293–405. Hassanali, A., Njagi, P.G.N., Bashir, M.O., 2005. Chemical ecology of locusts and related acridids. Annual Review of Entomology 50, 223–245. Hunter-Jones, P., 1958. Laboratory studies on the inheritance of phase characters in locusts. Anti-Locust Bulletin 29, 1–32. Islam, M.S., Roessingh, P., Simpson, S.J., McCaffery, A.R., 1994a. Effects of population density experienced by parents during mating and oviposition on the phase of hatchling desert locusts, Schistocerca gregaria. Proceedings of Royal Society of London B 257, 93–98. Islam, M.S., Roessingh, P., Simpson, S.J., McCaffery, A.R., 1994b. Parental effects on the behaviour and colouration of nymphs of the desert locust Schistocerca gregaria. Journal of Insect Physiology 40, 173–181. Maeno, K., Tanaka, S., 2007. Effects of hatchling body colour and rearing density on body colouration in last stadium nymphs of the desert locust, Schistocerca gregaria. Physiological Entomology 32, 87–94. Maeno, K., Tanaka, S., 2008a. Phase-specific developmental and reproductive strategies in the desert locust. Bulletin of Entomological Research 98, 527–534. Maeno, K., Tanaka, S., 2008b. Maternal effects on progeny size, number and body color in the desert locust, Schistocerca gregaria: density- and reproductive cycledependent variation. Journal of Insect Physiology 54, 1072–1080. Maeno, K., Tanaka, S., 2009a. The trans-generational phase accumulation in the desert locust: morphometric changes and extra molting. Journal of Insect Physiology 55, 1013–1020. Maeno, K., Tanaka, S., 2009b. Is juvenile hormone involved in the maternal regulation of egg size and progeny characteristics in the desert locust? Journal of Insect Physiology 55, 1021–1028. McCaffery, A.R., Simpson, S.J., Islam, M.S., Roessingh, P., 1998. A gregarizing factor present in the egg pod foam of the desert locust Schistocerca gregaria. Journal of Experimental Biology 201, 347–363. Pener, M.P., 1991. Locust phase polymorphism and its endocrine relations. Advances in Insect Physiology 23, 1–79. Pener, M.P., Simpson, S.J., 2009. Locust phase polyphenism: an update. Advances in Insect Physiology 36, 1–272. Tanaka, S., 2006. Corazonin and locust phase polyphenism. Applied Entomology and Zoology 41, 179–193. Tanaka, S., Maeno, K., 2006. Phase-related body-color polyphenism in hatchlings of the desert locust, Schistocerca gregaria: re-examination of the maternal and crowding effects. Journal of Insect Physiology 52, 1054–1061. Tanaka, S., Maeno, K., 2008. Maternal effects on progeny body size and color in the desert locust, Schistocerca gregaria: examination of a current view. Journal of Insect Physiology 54, 612–618. Tanaka, S., Maeno, K., 2010. A review of maternal and embryonic control of phasedependent progeny characteristics in the desert locust. Journal of Insect Physiology. 56, 911–918. Tanaka, S., Yagi, S., 1997. Evidence for the involvement of a neuropeptide in the control of body color in the desert locust, Schistocerca gregaria. Japanese Journal of Entomology 65, 447–457. Uvarov, B.P., 1928. Locusts and Grasshoppers. Imperial Bureau of Entomology, London. Uvarov, B., 1966. Grasshoppers and Locusts, vol. 1. Cambridge University Press, Cambridge, UK. Uvarov, B., 1977. Grasshoppers and Locusts, vol. 2. Centre for Overseas Pest Research, London, UK.
Contents lists available at ScienceDirect
Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys
Epigenetic transmission of phase in the desert locust, Schistocerca gregaria: Determining the stage sensitive to crowding for the maternal determination of progeny characteristics Koutaro Maeno, Seiji Tanaka * Locust Research Laboratory, National Institute of Agrobiological Sciences at Ohwashi (NIASO), Tsukuba, Ibaraki 305-8634, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 July 2010 Received in revised form 6 August 2010 Accepted 6 August 2010
Desert locust female adults respond to crowded conditions by changing progeny characteristics such as egg size, clutch size (no. of eggs per pod), hatchling body size and coloration. This study was conducted to determine the stage sensitive to crowding in this locust. Reproductively active females reared in isolation increased egg size and decreased clutch size and the proportion of green hatchlings after exposure to crowded conditions (in which each female was kept with four male adults). These changes depended not only on the timing of exposure to crowded conditions during the reproductive cycle but also on the length of the exposure. By varying the time and length of the exposure, it was found that crowding had no influence on progeny characteristics during the last two days of egg development at 31 8C and that there was a four-day sensitive stage before this period. The sensitive stage coincided with the time when the affected oocytes were 1.5–4 mm long, while the sensitivity to crowding appeared to be constant over the sensitive stage. The larger the magnitude of the increase in egg size after exposure to crowding, the smaller the proportion of green hatchlings (and the larger the proportion of gregarized dark hatchlings); there was a sigmoidal relationship between the two variables. Based on these results, we propose a model for determining the stage sensitive to crowding in both the female parent and the oocytes. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Crowding Density-dependent phase polyphenism Maternal effect Locusts Schistocerca gregaria
1. Introduction Although the desert locust, Schistocerca gregaria, usually occurs at a low density and seldom causes any economic damage, outbreaks of this locust are extremely destructive. As a result, this locust is one of the most destructive pests in the Old World (Uvarov, 1966, 1977). Numerous studies reported on various aspects of the biology of this locust, particularly since Uvarov (1928, 1966, 1977) proposed the phase theory in locusts (Dale and Tobe, 1990; Pener, 1991; Applebaum and Heifetz, 1999; Hassanali et al., 2005; Tanaka, 2006; Pener and Simpson, 2009). Locusts show remarkable phenotypic plasticity termed ‘‘phase polyphenism’’ in which morphological, physiological and behavioral traits change in response to local population density. At low population densities, juvenile hoppers are green or brown and adults are sedentary (solitarious forms). As the population density increases, the locusts become more gregarious, their body color becomes darker, and adults often migrate over a long distance in large swarms (gregarious forms). A complete change from the solitarious to the
* Corresponding author. Tel.: +81 29 838 6110; fax: +81 29 838 6110. E-mail address: [email protected] (S. Tanaka). 0022-1910/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jinsphys.2010.08.010
gregarious phase does not occur in one generation, but requires several generations. Locusts in the intermediate phase are called transient forms (Uvarov, 1966, 1977; Pener, 1991; Pener and Simpson, 2009). The trans-generational transmission of phase is an epigenetic phenomenon and has attracted much attention (Uvarov, 1966; Pener and Simpson, 2009). However, the details of the mechanism controlling this phenomenon remain largely unknown. It has recently been demonstrated in the laboratory that S. gregaria requires two generations to undergo a complete shift in morphological characteristics from the solitarious to the gregarious phase and vice versa (Maeno and Tanaka, 2009a). Adults change the quality and quantity of the progeny quickly in response to local population density (Maeno and Tanaka, 2008b, 2009b). Hatchlings from crowd-reared females are larger and darker than those from isolated-reared females, and they develop into adults with morphological characteristics typical of gregarious forms if the hatchlings are also grown under crowded conditions. However, such individuals show intermediate values between the two extreme phases if they are maintained in isolation during nymphal development. Similar adult characteristics of the intermediate (transient) phase are also observed if hatchlings from isolatedreared females are reared under crowded conditions, whereas such hatchlings become adults with characteristics typical of the
1884
[(Fig._1)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
solitarious form if the hatchlings are also reared in isolation. Therefore, the maternal determination of hatchling characteristics plays an important role in the trans-generational changes in phase characteristics in this locust. The present study aimed to determine which stage was sensitive to a crowding stimulus in S. gregaria female adults. This information is essential for understanding the physiological, endocrine and molecular mechanism of the transgenerational transmission of phase characteristics in the future. In S. gregaria, large, black hatchlings are produced if the female parent is exposed to crowding, whereas small, green hatchlings are obtained if the female parent experiences isolation or low-density conditions (Faure, 1932; Hunter-Jones, 1958). The crowding conditions experienced as adults are important in the control of progeny characteristics (egg size, hatchling body size and coloration; Hunter-Jones, 1958; Maeno and Tanaka, 2008a; Tanaka and Maeno, 2006, 2008). In their review article, Pener and Simpson (2009, p. 212) discussed the effects of a gregarizing agent derived from the reproductive accessory glands after ovulation, first in the oviduct and then after oviposition based on the studies of McCaffery et al. (1998) and others. The authors concluded that exposure to this gregarizing agent causes hatchlings to develop gregarious coloration and behavior. However, more recent studies have suggested that the ovariole is the site where egg size, hatchling body size and coloration are pre-determined and that no pheromonal factor is involved (for a review, see Tanaka and Maeno, 2010). The present study examined the effects of maternal crowding on the progeny characteristics in solitarious (isolatedreared) female adults. The results indicated that for this locust there is a narrow window of time in which a specific stage is sensitive to a crowding stimulus in each reproductive cycle. This paper describes the results of these observations in relation to oocyte growth. We hypothesize that the female parent might produce a hormonal factor that controls egg size in response to crowding conditions and that oocytes at a particular stage respond to this factor. 2. Materials and methods 2.1. Insects The S. gregaria laboratory colony used in this study was originally from Ethiopia, and the rearing and handling methods used here have been described previously (Tanaka and Yagi, 1997; Maeno and Tanaka, 2008a). Solitarious (isolated-reared) females, also visually isolated from one another, were used in all experiments and were reared individually in small cages (28 cm 15 cm 28 cm), except for a short period to allow mating during the adult stage, for two or more generations. The rearing room was well ventilated, but whether or not the locusts were also completely isolated pheromonally was unknown. Both nymphs and adults were fed orchard grass, cabbage leaves and bran. Each female adult was allowed to deposit into sand in a transparent plastic cup (340 ml volume) so that deposited egg pods could be seen through the bottom of the cup without having to dig into the sand. All locusts, including test insects, were maintained at 31 1 8C with a light–dark (LD) cycle of 16:8 h. 2.2. Determination of the stage sensitive to crowding Each female reared in isolation was paired with a sexually mature male 14–18 days after adult emergence. All males used in the experiment had been reared in a group because their phase status has no influence on the progeny characteristics examined (Maeno and Tanaka, unpublished observations). When a female is paired with a male for a long period of time in the laboratory, the female can be stimulated to produce offspring with characteristics
Fig. 1. Experimental design used to determine the stage sensitive to crowding in isolated-reared Schistocerca gregaria female adults at 31 8C. Isolated-reared female adults were exposed to crowding conditions (four male adults) at different lengths of time after the deposition of an egg pod (pod no. 1) for various periods and were then returned to isolated conditions. All females were allowed to deposit at least two egg pods before being used for the experiment.
of the gregarious forms (Hunter-Jones, 1958). Therefore, the pairing period was limited to less than 24 h. Mated females were allowed to deposit at least two egg pods before being used for experiments because substantial numbers of dark hatchlings are typically produced from the first egg pod even under continuously isolated conditions (Maeno and Tanaka, 2008b). Most female adults deposited egg pods every four days under the conditions used in this study (mean SD = 4.3 1.1 days, n = 129; Maeno and Tanaka, 2009b), with the time period ranging from three to six days. In the present study, oviposition was checked once a day and only those females that showed ovipositional intervals of four days during crowding treatments were used. To determine the stage sensitive to crowding, each mated female was held in a small cage with four sexually mature males one, two or four days after the deposition of an egg pod; the females were later re-isolated (Fig. 1). In this locust, the pairing of a female with a single male induces crowding effects on the progeny that are as strong as those observed when the female is reared with many males, and crowding effects can be elicited even when an isolated female is reared with females (Hunter-Jones, 1958). Males introduced were often found mating with the test females. Our unpublished data (Maeno and Tanaka) indicate that the crowding effects on solitarious females were equally elicited by sexually active males and females in terms of progeny characteristics such as body size and coloration. HunterJones (1958) demonstrated that solitarious females reproducing parthenogenetically switched from green-hatchling producers to dark-hatchling producers after being kept with other female adults. In the present study, only males were used to create crowding, because females would deposit egg pods, making it difficult to distinguish these egg pods from those produced by test females. Some females were kept in isolation or with four males continuously as controls. After the first egg pod was deposited at the beginning of the experiment, a maximum of four more egg pods were collected consecutively from each female and incubated at 31 1 8C. Two days after deposition the number of eggs in each pod was recorded and ten eggs were randomly chosen from each pod to measure their lengths to the nearest 0.1 mm with an ocular micrometer in a microscope. These and the other eggs were again incubated until hatching. The hatchlings of some egg pods were categorized into five hatchling color groups (HCGs 1–5) based on the darkness of their coloration as follows (Maeno and Tanaka, 2007): hatchlings in HCG 1 are green, as typically observed in solitarious forms; those in HCG 5 are almost completely black, as observed in gregarious forms; and those in HCGs 2–4 are intermediate in color.
[(Fig._2)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
1885
Fig. 2. Egg lengths (mean SD), proportions of hatchlings with different body colorations and clutch sizes (mean no. + or SD) of eggs per pod) of egg pods produced by isolatedreared Schistocerca gregaria female adults after different treatments at 31 8C. For explanation of the different treatments, see Fig. 1. Hatchlings were divided into five hatchling color groups (HCGs 1–5) based on the darkness of their coloration according to Maeno and Tanaka (2007), as follows: hatchlings in HCG 1 are green, as typically observed in solitarious forms, those in HCG 5 are almost completely black, as observed in gregarious forms, and those in HCGs 2–4 are intermediate in color. The numbers in parentheses indicate sample sizes. Asterisks above the bars indicate that the value was significantly different from that of the first egg pod in each panel using a t-test; **P < 0.01; ***P < 0.001. The data (proportions of green hatchlings) for hatchling body coloration were arcsine-transformed before being evaluated using a t-test.
2.3. Oocyte growth during reproductive cycles
3. Results
At the first, second, third, or fourth day after an egg pod was deposited, five to eleven females were sacrificed to measure the lengths of oocyte in the ovarioles. The ovary consists of functional and unfunctional ovarioles (Uvarov, 1966). Because the functional ovarioles of the same ovary looked similar to one another in sizes of the growing oocytes, one functional ovariole was removed from each dissected female and the lengths of the largest three oocytes were measured with an ocular micrometer in a microscope.
3.1. Effects of long periods of exposure to crowding
2.4. Statistical analysis The egg lengths and clutch sizes of egg pods were compared with the values for the first egg pod (the egg pod deposited at the beginning of the experiments) using a t-test after differences among pods were analyzed using an ANOVA. Likewise, the proportions of green hatchlings were compared using a t-test or an ANOVA after arcsine transformation of the data.
Females constantly kept in isolation (treatment I, Fig. 1) produced similar numbers of eggs and eggs of similar size in all five egg pods collected during the experimental period (Fig. 2A and K). Most hatchlings from those eggs were green (Fig. 2F). Isolatedreared females that were kept with males after the deposition of an egg pod (treatment II) produced significantly larger eggs in the second egg pod compared to the first egg pod, and the proportion of green hatchlings (HCG 1) was greatly decreased in the second egg pod (Fig. 2B and G). The mean egg length further increased and reached beyond 7 mm, and the proportion of green hatchlings further decreased in the third egg pod. However, no significant differences were observed in either parameter among the last three egg pods (P > 0.05; ANOVA for egg length; P > 0.05; x2 test for the proportion of green hatchlings), indicating that crowding effects were fully expressed by the third egg pod. Upon exposure to crowding, the clutch size (the number of eggs per pod) decreased,
1886
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
and there was a significant difference between the first and third or subsequent egg pods (Fig. 2L). Mean clutch size, which decreased significantly to 79.1 (SD = 15.0) eggs in the third egg pod, was further decreased to 71.4 (SD = 18.7) eggs in the fourth egg pod (P < 0.05; t = 3.61; df = 26), but no further reduction was observed in the fifth pod (73.4 18.1 eggs; P > 0.05; t-test; Fig. 2L). This result indicated that changes in clutch size occurred later than egg length. Females exposed to crowding for six days, which covered oneand-a-half inter-ovipositional periods (treatment III), exhibited results similar to those in treatment II for the first three egg pods (Fig. 2C, H and M). However, crowding effects seemed to disappear thereafter, causing all parameters to shift towards the values observed before the females were exposed to crowding. It took two reproductive cycles or egg pods to shift from one extreme to the other. Similar results were observed for egg length and hatchling body coloration when the females were exposed to crowding for four days, covering one ovipositional interval (treatment IV; Fig. 2D and I). In this case, however, egg lengths were similar in the second and third egg pods but remained relatively small. This result suggests that the first two days of crowding influenced the second egg pod, while the following two days influenced the third egg pod. No significant change was detected in clutch size (Fig. 2N). A fourday exposure to crowded conditions resulted in different results when this period began two days before the deposition of a second egg pod (treatment V). In those females, crowding did not influence the second egg pod, indicating that all parameters, including egg length, clutch size and hatchling body coloration, had been determined two days before oviposition. Crowding effects were detected in the third egg pod in all parameters (Fig. 2E, J and O). In the third egg pod, egg length increased to approximately 7 mm (Fig. 2E), the majority of hatchlings were in HCG 5 (Fig. 2J) and the clutch size decreased significantly (Fig. 2O). These results indicate that the period during which locusts were sensitive to crowding (i.e. the period in which progeny characteristics are determined in each egg pod) is approximately four days long, and begins six days before the egg pod is deposited under the present conditions.
[(Fig._3)TD$IG]
In treatments IV and V, the fourth egg pod produced fairly dark hatchlings, although the hypothesized four-day period of sensitivity to crowding would have ended two days before the deposition of the third egg pod. This relatively high incidence of dark hatchlings is likely related to difficulties experienced in staging female adults in these treatments: the females were kept with males two days after the deposition of the second egg pod, but the accuracy of this staging was 24 h. Therefore, the crowding period likely extended to the sensitive stage of the fourth egg pod in some female adults. Another possible factor that might explain this trend was the increased mortality of adults towards the end of the experiment. 3.2. Effects of short periods of exposure to crowding To determine if the sensitivity to crowding varies during the four-day sensitive period, isolated-reared female adults were exposed to crowding during the first half (two days prior to deposition of the second egg pod; treatment VI) or the second half (two days after deposition of the second egg pod; treatment VII) of the sensitive stage. The two treatments showed similar patterns in terms of the changes in egg length, hatchling body coloration and clutch size after a two-day crowding period (Fig. 3A, B, D, E, G and H). No significant difference between the two treatments was found in any of these parameters for the third egg pod (P > 0.05), indicating that the sensitivity to crowding was similar between the first and second halves of the sensitive stage. To determine whether there were cumulative effects of crowding over successive reproductive cycles, isolated-reared females were exposed to crowing during the first half of the sensitive stage for two consecutive clutches (treatment VIII; Fig. 1). The results showed that eggs of both the second and the third egg pods became significantly larger than those of the first egg pod, but remained relatively small (Fig. 3C). The mean lengths of the second and third egg pods were almost identical (P > 0.05; t-test). More than 50% of the hatchlings from the fourth egg pod were dark in
Fig. 3. Egg lengths (mean SD), proportions of hatchlings with different body colorations and clutch sizes (mean no. of eggs per pod + SD) of egg pods produced by isolated-reared Schistocerca gregaria female adults after a two-day exposure to crowded conditions during the sensitive stage at 31 8C. For an explanation of treatments, see Figs. 1 and 2. The data (proportions of green hatchlings) for hatchling body coloration were arcsine-transformed before being evaluated with a t-test.
[(Fig._4)TD$IG]
[(Fig._5)TD$IG]
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
Fig. 4. The mean proportions (+ or SD) of green hatchlings of Schistocerca gregaria plotted against the magnitudes of changes in egg length between the first egg pod and the subsequent egg pods produced during the experiments shown in Fig. 1. Asterisks indicate significant differences from the proportion of green hatchlings from eggs with the difference in egg length being 0 mm by a t-test; ***P < 0.001. The data (proportions of green hatchlings) for hatchling body coloration were arcsinetransformed before being evaluated with a t-test. Bars indicate one SD. Numbers in parentheses indicate sample sizes (no. of egg pods).
color and the proportion of green hatchlings was significantly smaller than that of the first egg pod (Fig. 3F). As in treatments IV and V, this result is probably related to the method of staging the females. No significant change was observed in the clutch size (Fig. 3I). 3.3. Relationships between the magnitude of increase in egg length and the proportion of green hatchlings The above results showed that egg length increased relative to the length of the first egg pod after the female parent was exposed to crowding, and that the magnitude of the increase depended on when and how long the female was exposed to crowding. Fig. 4 illustrates that there was a sigmoidal relationship between the magnitude of the increase in egg length and the proportion of green hatchlings in each pod in the above experiments (R2 = 0.982; n = 435). The proportion of green hatchlings gradually decreased as egg size increased, and a significant decrease was observed when the magnitude of the increase in egg length reached 0.3 mm. The proportion of green hatchlings decreased further, and most hatchlings became black when the magnitude of the increase in egg length is over 0.7 mm. Eggs from some pods deposited by females before they were exposed to crowding conditions decreased slightly in length. These egg pods predominantly produced green hatchlings, but there was no significant difference in the proportion of green hatchlings between these egg pods (P > 0.05; ANOVA). 4. Discussion The present study confirmed that the phase-related body-color polyphenism in hatchlings of S. gregaria is closely correlated with the egg size, which is determined in the ovarioles of the female parent. Egg size was found to be the first variable affected after the females experienced a shift in population density, followed by changes in clutch size (the number of eggs per pod; Tanaka and Maeno, 2010). This study also demonstrated that there is a short period in each reproductive cycle during which females are sensitive to population density, which determines the egg size and hatchling body coloration. Fig. 5 summarizes the responses of S. gregaria female adults reared in isolation to crowded conditions. The rearing density
1887
Fig. 5. A schematic diagram of the stage sensitive to crowding for the maternal determination of egg length and hatchling body coloration in the largest oocytes of a Schistocerca gregaria ovariole. Crowding was found to have no effect on the progeny characteristics during the last two days of oocyte growth (oocyte A) and the sensitive period occurs when oocytes are 1.5–4 mm long in the ovary (oocytes B and C). Crowding induces a maximum response when it occurs during the entire period of the sensitive stage, while it elicits a significant but small response when it occurs only during the first or second half of the sensitive stage. Crowding occurring during the first half of the sensitive stage over two consecutive reproductive cycles also induces a significant but small response, indicating that there is no cumulative effect of crowding over successive reproductive cycles. Responses are based on increases in egg length in Figs. 2 and 3; , no change; +, 0.30–0.69 mm; ++, 0.7 mm. The yellow, green and blue areas indicate the sensitive stage for oocytes A, B and C, respectively, and the pink area indicates the range of oocyte length in which egg size is modified by crowding.
during the last two days before oviposition had no influence on the progeny characteristics observed. This finding was consistent throughout the present study. Maternal density apparently has no effect on egg length once oocytes in the ovarioles attain a length of more than 4 mm. Significant and marked effects of maternal density were observed when crowding occurred during a four-day period prior to the last two days of oocyte development under the present conditions. This is the stage at which females are sensitive to crowding, and that determines the egg length and hatchling body coloration in each egg pod. This period coincides with the stage when oocytes are 1.5–4 mm in length. The mother’s sensitivity to crowding appears to be constant throughout this period because females responded to crowding in a similar way regardless of whether they experienced crowded conditions during the first or second half of the sensitive stage. However, females exposed to crowding during the entire period of the sensitive stage responded more strongly by producing larger eggs that gave rise to predominantly black hatchlings. The eggs produced by females exposed to crowding over two consecutive reproductive cycles but only during the first half of each sensitive period (treatment VIII) remained in an intermediate size. This result indicates that there is no cumulative effect of maternal density across consecutive reproductive cycles. Hunter-Jones (1958) was the first to demonstrate that adult rearing density plays an important role in determining the hatchling body coloration in S. gregaria, although at that time it was unclear which stage of the adult life cycle was sensitive to crowding. His observation was confirmed by Islam et al. (1994a,b) who exposed isolated-reared and crowd-reared adults to crowded or isolated conditions at mating and oviposition. The latter studies reported a highly significant effect of density at ‘oviposition’. Crowd-reared females isolated at oviposition produced offspring
1888
K. Maeno, S. Tanaka / Journal of Insect Physiology 56 (2010) 1883–1888
that were more likely to be green in color than black, but isolatedreared females crowded at oviposition did not produce black offspring (Islam et al., 1994a; Fig. 2e and f). In another study by Islam et al. (1994b) (Fig. 5), adults kept under crowded conditions at mating and oviposition produced offspring that were similar in hatchling body coloration to those produced by their counterparts that were kept in isolation during mating and oviposition. These results are difficult to reconcile with the present results which suggest the presence of a sensitive stage in each reproductive cycle. However, the differences between these studies may be explained as follows: in the studies of Islam et al. (1994a), the reproductive cycle was not considered. The period of time for which females were exposed to crowded or isolated conditions at oviposition was not described and might have varied among the females tested because the reproductive cycle was ignored. In another study (Islam et al., 1994b), female adults were kept in a group for mating and ovipositing for a week. However, sexually mature females often deposit an egg pod as soon as mating takes place (Maeno and Tanaka, unpublished observations), and as a result, some egg pods might have been deposited earlier than others during that week. Islam et al. (1994a,b) used extremely small sample sizes and also appeared to use the first egg pods alone. First egg pods are not suitable for observations of this kind because the proportion of green or black hatchlings from such egg pods varies from 0 to 100% in this locust (Maeno and Tanaka, 2008b; Tanaka and Maeno, 2008, 2010). In a model constructed to explain the trans-generational transmission of phase in S. gregaria (Tanaka and Maeno, 2010), hatchling body coloration and body weight are thought to be predetermined in the ovarioles of the mother. The observed crowding stimulus is transmitted to the brain which causes the factor controlling egg size to be released. If this factor is assumed to increase egg size and the effective organ of this factor is the ovary, an increased release of this factor would be expected when the female adult is crowded longer during the sensitive stage, and all oocytes growing in each ovariole would be exposed to it. However, the present study indicated that only oocytes of a certain size show the effects of crowding experienced by the mother, while those that are too small or too big do not (Fig. 5). One possible explanation for this phenomenon is that the mother releases the egg-size controlling factor in response to crowding at any time and that oocytes can respond to this factor only during a limited period when they are 1.5–4 mm long. If egg size depends on the time at which the chorion is formed, the follicle cells might be the cells that respond to the egg-size controlling factor and control the timing of chorionation or egg size. Although nothing is known about the nature of the factor involved, the present study suggests that the physiological process determining egg size and hatchling body coloration is considerably complex. The present study showed a high correlation between the magnitudes of the increase in egg length after exposure to crowding and the proportions of green hatchlings. Eggs that increased by 0.3 mm or more showed a significantly reduced proportion of green hatchlings. This observation may be useful for establishing a convenient assay system to determine the stimuli inducing crowding effects and the site where such stimuli are perceived. A study investigating these effects is currently underway.
Acknowledgements The authors thank Ms. Hiroko Ikeda for her assistance in the laboratory and Dr. Ken-ichi Harano for his cooperation and for stimulating discussions. This study was funded in part by the Japan Society for the Promotion for Science for a research fellowship for PD (No. 20-2008) to Maeno and a Kakenhi grant of Japan (No. 19380039) to Tanaka. The grass used in this experiment was raised by the Field Management Section of NIASO. References Applebaum, S.W., Heifetz, Y., 1999. Density-dependent physiological phase in insects. Annual Review of Entomology 44, 317–341. Dale, J.F., Tobe, S.S., 1990. The endocrine basis of locust phase polymorphism. In: Chapman, R.F., Joern, A. (Eds.), Biology of Grasshoppers. John Wiley and Sons, New York, pp. 393–414. Faure, J.C., 1932. The phases of locusts in South Africa. Bulletin of Entomological Research 23, 293–405. Hassanali, A., Njagi, P.G.N., Bashir, M.O., 2005. Chemical ecology of locusts and related acridids. Annual Review of Entomology 50, 223–245. Hunter-Jones, P., 1958. Laboratory studies on the inheritance of phase characters in locusts. Anti-Locust Bulletin 29, 1–32. Islam, M.S., Roessingh, P., Simpson, S.J., McCaffery, A.R., 1994a. Effects of population density experienced by parents during mating and oviposition on the phase of hatchling desert locusts, Schistocerca gregaria. Proceedings of Royal Society of London B 257, 93–98. Islam, M.S., Roessingh, P., Simpson, S.J., McCaffery, A.R., 1994b. Parental effects on the behaviour and colouration of nymphs of the desert locust Schistocerca gregaria. Journal of Insect Physiology 40, 173–181. Maeno, K., Tanaka, S., 2007. Effects of hatchling body colour and rearing density on body colouration in last stadium nymphs of the desert locust, Schistocerca gregaria. Physiological Entomology 32, 87–94. Maeno, K., Tanaka, S., 2008a. Phase-specific developmental and reproductive strategies in the desert locust. Bulletin of Entomological Research 98, 527–534. Maeno, K., Tanaka, S., 2008b. Maternal effects on progeny size, number and body color in the desert locust, Schistocerca gregaria: density- and reproductive cycledependent variation. Journal of Insect Physiology 54, 1072–1080. Maeno, K., Tanaka, S., 2009a. The trans-generational phase accumulation in the desert locust: morphometric changes and extra molting. Journal of Insect Physiology 55, 1013–1020. Maeno, K., Tanaka, S., 2009b. Is juvenile hormone involved in the maternal regulation of egg size and progeny characteristics in the desert locust? Journal of Insect Physiology 55, 1021–1028. McCaffery, A.R., Simpson, S.J., Islam, M.S., Roessingh, P., 1998. A gregarizing factor present in the egg pod foam of the desert locust Schistocerca gregaria. Journal of Experimental Biology 201, 347–363. Pener, M.P., 1991. Locust phase polymorphism and its endocrine relations. Advances in Insect Physiology 23, 1–79. Pener, M.P., Simpson, S.J., 2009. Locust phase polyphenism: an update. Advances in Insect Physiology 36, 1–272. Tanaka, S., 2006. Corazonin and locust phase polyphenism. Applied Entomology and Zoology 41, 179–193. Tanaka, S., Maeno, K., 2006. Phase-related body-color polyphenism in hatchlings of the desert locust, Schistocerca gregaria: re-examination of the maternal and crowding effects. Journal of Insect Physiology 52, 1054–1061. Tanaka, S., Maeno, K., 2008. Maternal effects on progeny body size and color in the desert locust, Schistocerca gregaria: examination of a current view. Journal of Insect Physiology 54, 612–618. Tanaka, S., Maeno, K., 2010. A review of maternal and embryonic control of phasedependent progeny characteristics in the desert locust. Journal of Insect Physiology. 56, 911–918. Tanaka, S., Yagi, S., 1997. Evidence for the involvement of a neuropeptide in the control of body color in the desert locust, Schistocerca gregaria. Japanese Journal of Entomology 65, 447–457. Uvarov, B.P., 1928. Locusts and Grasshoppers. Imperial Bureau of Entomology, London. Uvarov, B., 1966. Grasshoppers and Locusts, vol. 1. Cambridge University Press, Cambridge, UK. Uvarov, B., 1977. Grasshoppers and Locusts, vol. 2. Centre for Overseas Pest Research, London, UK.