The Society for Integrative and Comparative Biology
Projecting Population-Level Responses of Mysids Exposed to an Endocrine Disrupting Chemical1
1 U.S. EPA Gulf Ecology Division, 1 Sabine Island Drive, Gulf Breeze, Florida 32561
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
To fully understand the implications of a chemical's effect on the conservation of a species, effects observed at the physiological or individual level must be expressed in terms of the population. Since long-term field experiments are typically not feasible, vital rates such as survival and reproduction of individual organisms are measured in life table response experiments (LTRE) and employed to extrapolate the effects of a pollutant on the population. The population-level response of the mysid, Americamysis bahia, to varying concentrations of methoprene (0, 4, 8, 16, 31, 62 µg/L) was determined using age-structured population models. Models were parameterized from the results of an LTRE conducted throughout the entire mysid life cycle. A density-independent matrix model with time invariant demographic parameters was developed to measure the change in population growth rate,
, with change in methoprene concentration. The values of were greater than one for all methoprene concentrations, indicating that populations exposed to the concentrations reported here would not become extinct. However, a general decrease in occurred with increasing methoprene concentration and would result in reduced population sizes. Sensitivity and decomposition analyses were conducted to determined the relative roles of the vital rates on altered population growth rates and determined that impaired reproduction was the primary influence on the observed decrease in . The model constructed was a useful tool for linking the individual-level effects to the population-level consequences of methoprene exposure on mysids, as well as defining the mechanism (reduced reproduction) responsible for the observed effects on population. |
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Biological conservation is a multidisciplinary science that relies on the synthesis of all hierarchical levels, from cellular mechanisms to individual organism responses to population-level consequences. Toxic chemicals that act as endocrine disrupting compounds (EDCs) first affect organisms at the cellular level through hormone disruption and can lead to a diversity of physiological effects at the individual-level, such as abnormal growth and reproduction (Oberdörster and Cheek, 2000
). However, to understand the ecological impacts of such compounds, physiological effects must be interpreted in terms of the population. This task can be accomplished by using the individual level as a link; effects measured on individuals take their mechanisms from the level below and express their consequences at the level above (Caswell, 1996). This study projects the effects of the known endocrine disruptor, methoprene, on populations of the mysid shrimp, Americamysis bahia, based on physiological responses observed at the individual level.
Methoprene is a juvenile hormone (JH) mimic and is registered for use as an insect growth regulator in the control of mosquitoes, fleas, ants, and stored products pests (U.S. EPA, 2001). By disrupting the balance of hormones that facilitate molting, metamorphosis, and gametogenesis, methoprene has the potential to retard growth and reproduction of both target and non-target invertebrates where applied (Dhadialla et al., 1998). Non-target effects of methoprene are a concern primarily for invertebrates that utilize JH or other juvenoids as regulatory hormones (e.g., arthropods, nematodes). Several studies have measured the adverse physiological responses in non-target organisms exposed to methoprene (Templeton and Laufer, 1983; Yin et al., 1987; McKenney and Matthews, 1990; McKenney and Celestial, 1993). However, population-level effects and potential ecological consequences of methoprene exposure on non-target organisms are often insinuated, yet have never been convincingly connected to these lower orders of biological organization.
Life table response experiments (LTREs) were conducted to measure the effect of various methoprene concentrations on mysid survival and reproduction (McKenney and Celestial, 1996). Demographic matrix models were used to project the population-level consequences of methoprene-altered reproduction of mysids as measured in the LTRE (Caswell, 2001). A density-independent, stage-structured model was used to determine the change in population growth rate, , with increasing methoprene concentration. In a deterministic model such as this, a population growth rate greater than one will results in an increasing population size over time, whereas a population growth rate that has been reduced to less than one by a pollutant will decline. As such, population growth rate is commonly used to measure population-level pollution effects and provides a valuable endpoint for population risk projection.
Elasticity and decomposition analyses are less applied in toxicity studies and can further break down the changes in in terms of the responses of individual organisms. An elasticity analysis determines which stage-specific individual-level response (survival, reproduction) has the largest influence on population growth rate (Caswell, 2001). The precursor of the elasticities, vital rate sensitivities, are combined with the degree and magnitude of the non-linear changes in each vital rate in a decomposition analysis, which determines the altered vital rate that contributed to the observed changes in at each treatment concentration (Caswell, 1996). Together with the stage-structured model, these analyses were used to bridge the gap between individual-level responses and population-level effects of methoprene on mysids exposed to methoprene.
|
METHODS |
|---|
|
TOP
SYNOPSISINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
The effects of methoprene on A. bahia survival and reproduction were determined by exposing individual organisms to several methoprene concentrations through a complete life cycle as described in the methods of McKenney (1994)
. Throughout the course of two bioassays, mysids were exposed to several nominal methoprene concentrations (0, 4, 8, 16, 32, 62, and 125 µg methoprene/L) that were selected to test mysid responses over a range of sub-lethal and lethal concentrations. Newly released juvenile A. bahia were obtained from ovigenerous females maintained in static, recirculating cultures. Fifteen juveniles less than 24 hours old were placed into each of three replicates of each treatment concentration. Every three minutes, methoprene was added to the treatments using a proportional diluter as described by Schoor and McKenney (1983). Daily survival observations were made at each exposure concentration throughout the 29 day life cycle. Ovigenerous females were isolated and paired with mature males in Nitrix® brood cups and daily observations were made of the number of young released by each female. Stage-specific survival and reproduction data were used for model input; detailed results of the methoprene exposures on individual survival and reproduction are reported in McKenney and Celestial (1996). At 125 µg/L, total mortality occurred within 4 days of exposure, thus matrix models were parameterized using only the survival and reproduction data from concentrations 
62 µg methoprene/L.
The life cycle of A. bahia was divided into 7 stages that were representative of various mysid life stages: (1) early juveniles—less than 24 hours old, (2) juveniles—one to four days old, (3) advanced juveniles— five to 10 days old, (4) young adult—11 to 16 days old, (5) early breeders—17 to 20 days old, (6) intermediate breeders—21 to 24 days old, and (7) late breeders—25 to 29 days old. The stages are herein referred to by the oldest age in days of the individuals within each class. For example, stage 24 is composed of individuals between 21 and 24 days old. Reproductive adults were separated into 3 stages (early, intermediate, and late breeders) because methoprene delayed brood release (McKenney and Celestial, 1996) and the distinction of the three adult stages allowed us to identify potential effect of delayed brood release on population dynamics.
For each methoprene concentration, x, we developed a density-independent, deterministic population projection matrix, A(x) that was composed of probability of surviving and remaining in stage i, P(x)i; the probability of an individual growing from stage i to stage i+1, G(x)i; and the reproductive output of females in stage i, F(x)i such that:
|
|
The vital rates P(x)i, G(x)i, and F(x)i, denoted collectively as aij, were calculated from the data obtained in the bioassays using the equations:
|
|
|
|
where 
i(x) was the stage-specific survival probability, 
i(x) was the transition probability, and mi was the average number of offspring per female in stage i. The lower order vital rates, i(x) and i(x), were calculated according to Caswell (2001):
|
|
In Equation 6, T was the duration of the stage interval and (x) was the population growth rate calculated as the dominant eigenvalue of the matrix A(x). Because the estimate of i(x) required the eigenvector of the matrix that it was projecting, an iterative approach was used to calculate values of i(x) by setting an initial value of (x) to 1.1. The initial value of can be set as 1.0 for the first iteration (Caswell, 2001); however, 1.1 was selected as the initial value here to avoid mathematical complications where the denominator of Equation 6 was reduced to zero. The resulting values of i(x) were used to estimate the entries of a second matrix, which was then used to produce a second value of (x). Parameters were recalculated until the resulting values of (x) were stabilized (Caswell, 2001). Parameter estimates for each treatment matrix are shown in Table 1. The final values of were compared among treatments to determine the change in population growth rate with increased methoprene concentration. Population growth rate, , can be related to the continuous-time rate of increase as r = log .
|
To determine the contribution of each vital rate, aij, to
for each concentration, x, the sensitivity (sij) of to changes in aij was first determined for each concentration as:
|
|
where w, the stable age distribution, was the right eigenvector of the matrix and v was the corresponding left eigenvector representing the reproductive value of each stage, and 
vw
was the scalar product. A sensitivity matrix derived from Equation 7 determined how much would change if a change occurred to aij. An elasticity matrix for each treatment concentration determined the relative contribution of each vital rate to and was related to sensitivity as:
|
|
Elasticities, which sum to one within a matrix, are presented for each treatment concentration in Table 2.
|
We used a regression design of the decomposition method outlined by Caswell (1996)
to determine how each vital rate changed with methoprene concentration and the relative contribution of these impacts on changes in . The decomposition analysis was based on the sensitivity of to the vital rates, 
/aij, and the change of each vital rate with methoprene concentration, aij(x)/x. The latter expression was obtained as the slope of a nonparametric regression using a loess smoother. The smoothing parameter was selected using cross-validation to minimize the prediction error. The decomposition of the change in with methoprene concentration was modeled as:
|
|
and is a function of the sensitivity value of each vital rate and magnitude of its change.
|
RESULTS |
|---|
|
TOP
SYNOPSISINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
In the stage-structured model, population growth rate experienced an approximate 5% decline with an increase from 0 to 62 µg methoprene/L (R2 = 0.91, P = 0.003) (Fig. 1). Based on the elasticity analysis, Fi and Gi had relatively small contributions to
compared to Pi of intermediate stages (P10, P16, P20) at all concentrations (Table 2).
|
The nonparametric regression of the effects of treatment on Pi, Gi, and Fi showed that each vital rate was affected at some concentration (Fig. 2). At lower concentrations (
16 µg/L), Pi of early juveniles through early breeders were less than control populations, with the largest decrease occurring in young adults (ages 11–16 days old). Pi of breeding adult stages (ages 17 to 28 days) were higher than control populations at concentrations less than 16 µg/L but decreased between 16 and 31 µg/L and then stabilized between 31 and 62 µg methoprene/L (Fig. 2a). In general, Gi was not greatly impacted by methoprene and was greater than the control at lower concentrations (8 µg/L). At higher concentrations, very little impacts were observed to Gi (Fig. 2b). Overall, the most dramatic effects occurred in the reproduction of all adults stages. Compared with the effects on Pi and Gi, the effects to Fi were two orders of magnitude more severe, which can be observed in Figure 2 as the difference in scale of the z axes. A delay in reproduction was not observed in this analysis, which would have been noted by a decrease in reproduction of early breeders (ages 17–20 days old) that was not observed in intermediate (ages 21–24 days old) or late (ages 25–28 days old) breeders. Rather, decreased reproduction occurred in all adult stages for concentrations of 4 and 8 µg/L (Fig. 2c). Young production then increased at 16 µg/ L in all stages. At the highest concentration, the greatest impact occurred on the reproduction early breeders (ages 17–20 days old).
|
The decomposition analysis determined which changes measured for Pi, Gi, and Fi contributed most to the change in
. Despite the high values of Pi measured by the elasticity analysis, neither Pi nor Gi contributed much to the change in , (Fig. 3 a, b). Conversely, the small relative contribution of reproduction on provided the most influence on the change in , which is demonstrated by the identical patterns of change in Fi and contribution of Fi (Figs. 2c, 3c).
|
|
DISCUSSION |
|---|
|
TOP
SYNOPSISINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
In this study, population growth rate,
, of A. bahia generally declined with increasing methoprene concentrations throughout the range of concentrations used. Without further information on the dynamics of natural populations to compare with the deterministic model used here, these results can not directly infer population viability in natural environments as a result of methoprene exposure. However, reduced with increasing concentration does indicate a population-level effect, which may find its effect may find its consequences as increased probability of extinction. As population growth rate decreases, the probability of population extinction increases in stochastic environments (Tanaka, 2003). Additionally, mysids exposed to the highest dose in the bioassays, 125 µg methoprene/L, experienced 100% mortality within four days of exposure (McKenney and Celestial, 1996). Although the population-level response of this lethal effects was not able to be modeled, populations exposed to 125 µg methoprene/L or higher would decline toward extinction at an exponential rate.
Mysid populations with lower growth rates would also result in smaller population sizes, which may trickle up to a community-level impact. Mysids are a critical link in the food web of estuarine habitats, where they graze on primary producers and are an important food source for several species of fish (Mauchline, 1980; Mees and Jones, 1997). Chemically-reduced mysid populations would presumably alter the trophodynamics of their food web by reducing the production of dominant secondary producers (Munkittrick and McCarty, 1995). A more informative analysis based on field populations and community interactions would provide a better estimate of the ecological consequences of methoprene-induced reductions in mysid population growth rate.
The physiological endpoints measured on A. bahia in this study were survival and reproduction. We used a nonparametric regression to determine how these vital rates were affected by methoprene at different concentrations and in different life stages. In the decomposition analysis, the results of the nonparametric regression were combined with the sensitivities of each concentration projection matrix to determined how the effects of methoprene on each vital rate influenced the change in population growth rate. Pi of intermediate stages provided the largest relative contribution to population growth rate based on large elasticities. However, the impact of methoprene on Pi was small and provided only a minimal contribution to the changes in population growth rate. Conversely, the relative role (elasticity) of Fi on population growth rate was small (<0.05 for each stage), yet the impact of methoprene on reproduction was large enough to impact population growth rate. Similar patterns observed for the change in reproduction with methoprene concentration (Fig. 2c) and the contribution of those impacts to changes in population growth rate (Fig. 3c) identifies altered reproduction as the primary cause of decreased growth rate with increased methoprene concentration.
This finding is reflective of the endocrine disrupting nature of methoprene. The mode of action of methoprene is to directly interfere with normal hormone function by mimicking the insect juvenile hormone (JH). In juvenile insects, elevated JH levels disrupt the balance of molting hormones which may result in supernumerary larvae or prevent metamorphosis and molting (Dhadialla et al., 1998). In adults, JH mimics have resulted in a number of reproductive impairments such as sterility (King and Bennett, 1990; Hicks and Gordon, 1992), altered gamete production (Glancy and Banks, 1988), and inhibited oviposition (Hatakoshi, 1992). In crustaceans, juvenoids such as methyl farnesoate are physiologically analogous and structurally similar to JH. As such, methoprene and other JH mimics may have similar effects on the crustacean endocrine system as methyl farnesoate, thereby disrupting normal hormone function (McKenney, 2005). Due to the broad spectrum of potential effects, methoprene may impact a variety of life processes in several life stages of non-target crustaceans. For example, methoprene has retarded growth and inhibited metamorphosis of larval grass shrimp (McKenney and Matthews, 1990; McKenney and Celestial, 1993), inhibited both male and female gametogenesis in the mud-crab (Payen and Costlow, 1977), and inhibited growth and reproduction in Daphnia (Templeton and Laufer, 1983; Olmstead and LeBlanc, 2001).
Our analysis was based on a simple stage-structured model that did not incorporate density dependence or environmental stochasticity, both of which play critical roles in modeling population risk estimation (Grant, 1998; Tanaka, 2003). Although density-dependence is known to occur in both A. bahia juveniles and adults (Lussier et al., 1988), detailed information on the strength of density dependence acting on each stage is lacking. Applying an arbitrary strength of density dependence to all stages yield results proportional to those of a density-independent model (S. R., unpublished data). Additionally, Caswell (2001) noted that elasticities of density-dependent vital rates were highly correlated with those of corresponding density-independent vital rates. Without more detailed information characterizing the density dependence of A. bahia, incorporating it into stage-structured models such as the one used here are not likely to improve the predictability of the model.
We used a common modeling approach to estimate the population-level effects of methoprene on A. bahia based on the response of individuals in laboratory life-cycle experiments. Although models lack the strength of long-term field studies, they allowed us to identify the vital rate that was primarily responsible for decreased population growth rate. Because A. bahia is one of the more sensitive estuarine invertebrates and a valuable organism for biomonitoring ecosystem health (Nimmo and Hamaker, 1982; Roast et al., 1998, Verslycke et al., 2004), its population-level response to a pollutant provides a valuable contribution where estuarine conservation is a concern. Furthermore, as research expands the available information on physiological and individual-level effects of xenobiotics on non-target species, it becomes increasingly more important to translate these effects in terms of the population for the comprehensive analysis that is required for biological conservation.
|
ACKNOWLEDGMENTS |
|---|
The data used in this study were generated by the staff of the U. S. Environmental Protection Agency, Gulf Ecology Division. We wish to thank Doug Middaugh, Shea Tuberty and the anonymous reviewers for their suggestions that improved this manuscript. The information in this document does not necessarily reflect the views and policies of the U. S. Environmental Protection Agency.
|
FOOTNOTES |
|---|
1 From the Symposium EcoPhysiology and Conservation: The Contribution of Endocrinology and Immunology presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 5–9 January 2004, at New Orleans, Louisiana.
2 E-mail: raimondo.sandy{at}epa.gov
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Caswell, H. 1996. Demography meets ecotoxicology: Untangling the population level effects of toxic substances. In M. C. Newman and C. H. Jagoe (eds.), Ecotoxicology: A hierarchical treatment, pp. 255–292. Lewis Publishers, New York.
Caswell, H. 2001. Matrix population models. Sinauer Associates, Sunderland, Massachusetts.
Dhadialla, T. S., G. R. Carlson, and D. P. Le. 1998. New insecticides with ecdysteroidal and juvenile hormone activity. Annu. Rev. Entomol, 43:545-569.[CrossRef][ISI][Medline]
Glancy, M. B., and W. A. Banks. 1988. Effects of the insect growth regulator fenoxycarb on the ovaries of queens of the red imported fire ant (Hymenoptera: Formicidae). J. Econ. Entomol, 81:642-648.
Grant, A. 1998. Population consequences of chronic toxicity: Incorporating density dependence into the analysis of life table response experiments. Ecological Modelling, 105:325-335.[CrossRef]
Hatakoshi, M. 1992. An inhibitory mechanism over oviposition in the tobacco cutworm, Spodoptera litura, by juvenile hormone analogue pyriproxyfen. J. Insect Physiol, 38:793-801.[CrossRef]
Hicks, B. J., and R. Gordon. 1992. Effects of the juvenile hormone analog fenoxycarb on various developmental stages of the eastern spruce budworm, Choristoneura fumiferana (Clemens) (Lepidoptera: Tortricidae). Can. Entomol, 124:117-123.
King, J. E., and G. W. Bennett. 1990. Comparative sterilizing and ovicidal activity of fenoxycarb and hydroprene in adults and oothecae of the German cockroach (Dictyoptera: Blattellidae). J. Med. Entomol, 27:642-645.[Medline]
Lussier, S. M., A. Kuhn, M. J. Chammas, and J. Sewall. 1988. Techniques for the laboratory culture of Mysidopsis species (Crustacea: Mysidacea). Environ. Toxicol. Chem, 7:969-977.
Mauchline, J. 1980. The biology of mysids and euphausiids. Adv. Mar. Biol, 18:1-369.
McKenney, C. L., Jr. 1994. Alterations in growth, reproduction, and energy metabolism of estuarine crustaceans as indicators of pollutant stress. In J. Salanki, D. Jeffrey, and G. M. Hughes (eds.), Biological monitoring of the environment: A manual of methods, pp. 111–115. CAB International, Wallingford, England.
McKenney, C. L., Jr. 2005. The influence of insect juvenile hormone agonists on metamorphosis and reproduction in estuarine crustaceans. Integr. Comp. Biol. (In print).
McKenney, C. L., Jr., and D. M. Celestial. 1993. Variations in larval growth and metabolism of an estuarine shrimp Palaemonetes pugio during toxicosis by an insect growth regulator. Comp. Biochem. Physiol, 105:239-245.[CrossRef]
McKenney, C. L., Jr., and D. M. Celestial. 1996. Modified survival, growth and reproduction in an estuarine mysid (Mysidopsis bahia) exposed to a juvenile hormone analogue through a complete life cycle. Aquat. Toxicol, 35:11-20.[CrossRef]
McKenney, C. L., Jr., and E. Matthews. 1990. Influence of an insect growth regulator on the larval development of an estuarine shrimp. Environ. Poll, 64:169-178.[CrossRef][Medline]
Mees, J., and M. B. Jones. 1997. The hyperbenthos. Oceanogr. Mar. Biol. Annu. Rev, 35:221-255.
Munkittrick, K. R., and L. S. McCarty. 1995. An integrated approach to aquatic ecosystem health: Top-down, bottom-up or middle-out? J. Aquat. Ecosystems Health, 4:77-90.
Nimmo, D. R., and T. L. Hamaker. 1982. Mysids in toxicity testing— a review. Hydrobiologia, 93:171-178.[CrossRef]
Oberdörster, E., and O. Cheek. 2000. Gender benders at the beach: Endocrine disruption in marine and estuarine organisms. Environ. Toxicol. Chem, 20:23-36.[Medline]
Olmstead, A. W., and G. L. LeBlanc. 2001. Low exposure concentration effects of methoprene on endocrine-regulated processes n the crustacean Daphnia magna. Toxicol. Sci, 62:268-273.
Payen, G. G., and J. D. Costlow. 1977. Effects of a juvenile hormone mimic on male and female gametogenesis of the mud-crab, Rhithropanopeus harrish (Gould) (Brachyura: Xanthidae). Biol. Bull, 152:199-208.
Roast, S. D., R. S. Thompson, J. Widdows, and M. B. Jones. 1998. Mysids and environmental monitoring: A case for their use in estuaries. Mar. Freshwater Res, 49:827-832.[CrossRef]
Schoor, W. P., and C. L. McKenney Jr. 1983. Determination of fenvalerate in flowing-seawater exposure studies. Bull. Environ. Contam. Toxicol, 30:84-92.[Medline]
Tanaka, Y. 2003. Ecological risk assessment of pollutant chemicals: Extinction risk based on population-level effects. Chemosphere, 53:421-425.[Medline]
Templeton, N. S., and H. Laufer. 1983. The effects of a juvenile hormone analog (Altosid ZR-515) on the reproduction and development of Daphnia magna (Crustacea: Cladocera). Int. J. Invertebr. Repr. Dev, 6:99-110.
U.S. EPA., 2001. Update of the March 1991 methoprene reregistration decision (R.E.D.) fact sheet. http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factsheet_105401.pdf.
Verslycke, T. A., N. Fockedey, C. L. McKenney Jr., S. D. Roast, M. B. Jones, J. Mees, and C. R. Janssen. 2004. Mysids as potential test organisms for the evaluation of environmental endocrine disruption: A review. Environ. Toxicol. Chem. (In press).
Yin, C.-M., M. Takeda, and Z.-S. Wang. 1987. A juvenile hormone analogue, methoprene as a circadian and developmental modulator in Diatraea grandiosella (Pyralidae). J. Insect Physiol, 33:95-102.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||