2004 Progress Report: Individual Level Indicators: Reproductive Function in Estuarine Fishes

EPA Grant Number: R829458C005
Subproject: this is subproject number 005 , established and managed by the Center Director under grant R829458
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: EAGLES - Consortium for Estuarine Ecoindicator Research for the Gulf of Mexico
Center Director: Brouwer, Marius
Title: Individual Level Indicators: Reproductive Function in Estuarine Fishes
Investigators: Thomas, Peter , Rose, Kenneth A. , Cheek, Ann , Nunez, Scott
Institution: The University of Texas at Austin , Louisiana State University - Baton Rouge
Current Institution: The University of Texas at Austin , University of Southern Mississippi
EPA Project Officer: Packard, Benjamin H
Project Period: December 1, 2001 through November 30, 2005 (Extended to May 20, 2007)
Project Period Covered by this Report: December 1, 2003 through November 30, 2004
RFA: Environmental Indicators in the Estuarine Environment Research Program (2000) RFA Text |  Recipients Lists
Research Category: Water , Aquatic Ecosystems , Ecological Indicators/Assessment/Restoration


The overall long-term objective of this component of the Consortium for Estuarine Ecoindicator Research for the Gulf of Mexico (CEER-GOM) is to evaluate biomarkers of reproductive function in Atlantic croaker and gulf killifish Fundulus grandis as early warning indicators of fish population hazards resulting from the degradation of estuarine environments. For Atlantic croaker, the specific objectives for Year 3 were to: (1) determine the effects of hypoxia on biomarkers of vitellogenin, estrogen receptor mRNA levels, and hypoxia inducible factor expression in croaker collected from Pensacola Bay, Florida, and in the controlled laboratory study in 2003; and (2) collect croaker from hypoxic and reference sites in the Mobile Bay, Alabama, and compare site differences in selected reproductive indicators. For F. grandis, the specific objectives of Year 3 were to: (1) evaluate the effects of size on reproductive parameters in F. grandis; and (2) compare reproductive biomarker responses in F. grandis collected from hypoxic and normoxic sites in Terrebonne Bay, Louisiana; Pensacola Bay, Florida; and Weeks Bay, Alabama.

Progress Summary:

Croaker (Thomas, Nunez)

Objective 1: Complete Analyses of Reproductive Indicators in Atlantic Croaker Exposed to Hypoxic and Normoxic Conditions in Pensacola/East Bay, Florida, and in the Controlled Laboratory Experiments

Previously, it was shown that circulating levels of estradiol-17β were lower in croaker collected from the hypoxic sites in East Bay, suggesting an impairment of estrogen signaling in the exposed fish. Hepatic estrogen receptor mRNA levels were significantly lower in these croaker collected from the hypoxic sites (P12, P13, P14, PB5; Figure 1) compared to those sampled from the two reference sites (Bridge and Pensacola Bay). Circulating plasma vitellogenin concentrations were dramatically decreased in croaker exposed to hypoxic conditions compared to those under normoxic conditions, further indicating an impairment of estrogen signaling (Figure 2). The decline in vitellogenin levels correlated with the decreased gonadal growth and oocyte development observed previously in these fish from the hypoxic sites. Fish collected in November showed a similar hypoxic site-specific pattern of impaired ovarian development and vitellogenin levels. It is concluded from the field data that the croaker collected from all the low dissolved oxygen (DO) sites in East Bay showed a consistent pattern of impairment of reproductive indicators associated with vitellogenesis, as well as oocyte and gonadal growth and vitellogenin. In contrast, fish collected near the Bridge site with normal DO at the edge of Pensacola Bay, less than 2 miles from some of the East Bay sites, showed normal gonadal development and vitellogenin, as well as ER mRNA profiles. Hypoxia inducible factor 1α (HIF1α), HIF2α, aryl hydrocarbon nuclear translocator 1 (ARNT1), and ARNT2 were cloned from Atlantic croaker and sequenced. Croaker from one of the hypoxic sites (P13) showed approximately a two-fold induction of ovarian HIF 2α mRNA levels compared to fish collected from a reference site (Bridge), suggesting the gene is upregulated in response to environmental hypoxia.

Similar analyses of hepatic endoplasmic reticulum (ER) mRNA levels and plasma vitellogenin concentrations were also conducted with croaker samples from the laboratory hypoxia studies. ER mRNA levels in the livers of female chronically exposed to an intermediate DO level (2.7ppm) were approximately 50 percent those of the control group and showed no further decrease in the low DO group (1.7ppm). Plasma vitellogenin levels showed a similar dramatic decrease in the low DO experimental groups compared to the normoxic controls as observed in the field samples. Thus, the patterns of disruption of estrogen signaling in croaker exposed to low DO in the laboratory and in the field are remarkably similar. The gonadotropin (luteinizing hormone, LH) response to the neuropeptide, LHRH, was examined in croaker exposed to low DO to determine whether the decreased estrogen signaling resulted from impaired neuroendocrine function. Plasma gonadotropin levels 1 hour after LHRH challenge were significantly reduced in croaker exposed to the lowest DO concentration compared to controls (Figure 3), suggesting this environmental stressors is acting, at least partially, at the neuroendocrine level to disrupt reproduction.

Finally, HIF2α mRNA levels showed a concentration-dependent increase in croaker exposed to low DO in the laboratory, thereby confirming the preliminary results from the field study and further suggesting the potential utility of this measurement as a molecular indicator of exposure to hypoxia.

Objective 2: Collect Croaker From Hypoxic and Reference Sites in the Mobile Bay, Alabama, and Compare Site Differences in Selected Indicators of Ovarian Growth and Vitellogenesis

Hurricane Ivan made landfall at Mobile Bay, Alabama, prior to the sampling in the fall of 2004, dramatically altering the profiles of DO. Analysis of all the DO data collected during the summer and fall in the vicinity of the collection sites is ongoing. Preliminary analyses, however, suggest that two of the sample sites were under hypoxic conditions (sites A and B) for extended periods and two sites (sites C and D) were normoxic for a length of time prior to sampling. Atlantic croaker (30 croaker/station) samples were collected between dawn and noon with a trawl from the four field stations in Mobile Bay, Alabama, twice during the period of gonadal recrudescence (10/5/04 and 10/26/04). The fish were processed on board and the blood, gonadal tissues, liver tissues, and brain tissues were collected for analysis. Male and female fish collected during both sampling periods showed significant inhibition of gonadal growth (GSI) at the high DO sites (Table 1). Histological examination of the ovaries from the females showed that the decreased ovarian growth was associated with a significant decrease in the percentage of oocytes that were fully grown and contained large amounts of yolk reserves (Table 1). Plasma concentrations of vitellogenin were significantly reduced in females from the low DO sites compared to those from the reference sites, and this was associated by minor decreases in ER mRNA levels, as measured by RT-PCR (Table 1). In contrast, plasma estradiol levels varied considerably among the field-caught croaker and did not show site differences, suggesting this measurement may not be suitable as a reproductive indicator for field assessments.


Hypoxia stressor-induced changes in eight out of nine endocrine indicators, and all the morphometric six and functional six reproductive indicators were observed in the controlled laboratory studies. Because reproductive impairment was severe in field-caught fish, some stage specific analyses could not be conducted and others were impractical because the method of collection by trawling (e.g., effects of capture stress on gonadotropin levels). All five endocrine indicators and five morphometric indicators measured in the field samples were significantly altered in fish collected from the hypoxic sites in East Bay/Pensacola Bay in 2003. Four out of the five indicators used to assess ovarian growth and vitellogenesis were significantly altered in females collected from Mobile Bay sites in 2004. It is concluded from these studies that up to 11 reproductive indicators in Atlantic croaker are potentially useful for environmental monitoring. The probability of successfully applying these indicators is considered high, based on the field testing conducted to date.

Table 1. Reproductive Indicators in Female Atlantic Croaker Collected From Hypoxic (hyp) and Normoxic (norm) Sites in Mobile Bay in October 2004



A (hyp)

B (hyp)

C (norm)

D (norm)

Estradiol 10/5/04 (ng/mL)


1.52 ± 0.16

1.47 ± 0.37

1.82 ± 0.34

1.28 ± 0.34

Vitellogenin 10/5/04 (µg/mL)


227 ± 43***

180 ± 43 ***

643 ± 75

469 ± 91*

ER mRNA 10/5/04 (%)


69.8 ± 5.1 ++

78.7 ± 4.7 +

78.5 ± 4.8 +

88.4 ± 3.2

Mature Oocytes 10/5/04 (%)


5.6 ± 4.8 **

7.6 ± 4.1**

31.8 ± 7.8

31.1 ± 10.6

GSI 10/5-10/26/04 (%)


2.8 ± 0.4**

2.9 ± 0.5**

4.4 ± 0.6

4.3 ± 0.6

*:p<0.10, **:p<0.05, ***:p<0.0001 compared to C; +:p<0.10, ++:p<0.05, +++:p<0.0001 compared to D

Fundulus grandis (Cheek)

Objective 1: Evaluating the Effects of Body Size on Reproductive Indicators

Across all three estuaries, indicators varied with body size during the reproductive season (April and July), but not in the fall (November). In the early spawning season (April) condition factor in males and females increased with body size in Pensacola Bay, but not Terrebonne Bay. In April, steroid hormones in males (T and 11KT) and females (T and E2) increased with body size in both Pensacola Bay and Terrebonne Bay, but GSI did not, nor did vitellogenin. Later in the spawning season (July), male steroid hormones and GSI were positively related to size in Pensacola Bay, but not in Terrebonne Bay or Weeks Bay. In July, female steroid hormones and vitellogenin were generally unrelated to size, but female GSI increased with body size in PB and TB. Perhaps only larger females continue to produce eggs late in the spawning season, but all sizes of males seem to be capable of producing sperm. Late in the spawning season, female, but not male, condition factor was positively related to body size in Terrebonne Bay and Weeks Bay, but not Pensacola Bay, suggesting that large females have greater energy reserves at the end of the spawning season.

Objective 2: Comparison of Reproductive Biomarkers in F. grandis Captured From Hypoxic and Normoxic Sites in Terrebonne Bay, Pensacola Bay, and Weeks Bay

Marshes in the three estuaries differed in estimated duration of diel hypoxia at the hypoxic sites. In Weeks Bay, diel minima of less than or equal to 2 mg/L persisted for approximately 5 weeks prior to fish sampling. In Pensacola Bay, minima of less than or equal to 3.5 mg/L persisted for at least 3 months before sampling. Duration could not be estimated from DO profiles in Terrebonne Bay as profiles were taken only during sampling.

Focusing on indicator responses in July when persistent diel hypoxia is likely to occur revealed a clear difference in the pattern of indicator responses. After medium term diel hypoxia (5 weeks– Weeks Bay), male GSI and steroid hormones (T and 11KT) are significantly reduced, but condition factor is not. Female GSI is reduced and steroid hormones probably are as well (samples to be analyzed), but condition factor and vitellogenin are unaffected. After long-term (> 3 months – Pensacola Bay) diel hypoxia, male condition factor is reduced, but GSI and steroid hormones do not differ. Female condition factor also is reduced, as is vitellogenin. This pattern of responses confirms the predicted time scale of stressor responsiveness for each indicator: GSI -days to weeks; steroid hormones -hours to days and weeks; condition factor and vitellogenin -months. Using the pattern of indicator responses observed at Pensacola and Weeks Bay sites, we hindcast the likely duration of diel hypoxia in Terrebonne Bay as being longer than 5 weeks but less than 3 months. Like Weeks Bay, male GSI and steroid hormones were significantly reduced but condition factor was not. No parameters were reduced in females, but GSI was significantly related to body size as it was at Pensacola Bay.


Reproductive function in F. grandis was shown to be sensitive to moderate hypoxia in the field. Comparing field sites with differing durations (weeks -months) of diel hypoxia demonstrated variation in temporal sensitivity of indicators. GSI and steroid hormones are reduced in response to short- to medium-term hypoxia while condition factor and vitellogenin are reduced by long term (several months) hypoxia.

Future Activities:

F. grandis

All the major objectives of the first 3 years of the project have been completed and the remaining analyses will be completed by June 20. The results of the field studies will be confirmed with follow-up sampling at the same sites in the Pensacola and Weeks Bay systems this year.

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other subproject views: All 24 publications 5 publications in selected types All 4 journal articles
Other center views: All 171 publications 54 publications in selected types All 48 journal articles
Type Citation Sub Project Document Sources
Journal Article Popular article featuring Cheek & Thomas projects: Janet Raloff. Choked Up: How dead zones affect fish reproduction. Science News 2004;166(20):309. R829458C005 (2004)
not available

Supplemental Keywords:

population, community, ecosystem, watersheds, estuary, estuaries, Gulf of Mexico, nutrients, hypoxia, innovative technology, biomarkers, water quality, remote sensing, geographic information system, GIS, integrated assessment, risk assessment, fisheries, conservation, restoration, monitoring/modeling, benthic indicators, ecological exposure, ecosystem monitoring, environmental indicators, environmental stress, estuarine ecoindicator, estuarine integrity,, RFA, Scientific Discipline, Health, ECOSYSTEMS, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Ecosystem/Assessment/Indicators, Endocrine Disruptors - Environmental Exposure & Risk, Aquatic Ecosystem, Aquatic Ecosystems, endocrine disruptors, Ecological Monitoring, Environmental Monitoring, Ecology and Ecosystems, Biology, Endocrine Disruptors - Human Health, Ecological Indicators, Gulf of Mexico, monitoring, ecoindicator, ecological exposure, molecular ecology, estuaries, estuarine integrity, ecosystem assessment, biomarkers, fish, endocrine disrupting chemicals, ecological assessment, estuarine ecoindicator, fish reproduction, animal models, environmental indicators, environmental stress, reproductive processes, water quality, nutrient fluxes

Relevant Websites:

http://www.usm.edu/gcrl/ceer_gom/ Exit
http://www.usm.edu/gcrl/ceer_gom/fish_reproduction.php Exit

Progress and Final Reports:

Original Abstract
  • 2002 Progress Report
  • 2003 Progress Report
  • 2005 Progress Report
  • 2006
  • Final

  • Main Center Abstract and Reports:

    R829458    EAGLES - Consortium for Estuarine Ecoindicator Research for the Gulf of Mexico

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R829458C001 Remote Sensing of Water Quality
    R829458C002 Microbial Biofilms as Indicators of Estuarine Ecosystem Condition
    R829458C003 Individual Level Indicators: Molecular Indicators of Dissolved Oxygen Stress in Crustaceans
    R829458C004 Data Management and Analysis
    R829458C005 Individual Level Indicators: Reproductive Function in Estuarine Fishes
    R829458C006 Collaborative Efforts Between CEER-GOM and U.S. Environmental Protection Agency (EPA)-Gulf Ecology Division (GED)
    R829458C007 GIS and Terrestrial Remote Sensing
    R829458C008 Macrobenthic Process Indicators of Estuarine Condition for the Northern Gulf of Mexico
    R829458C009 Modeling and Integration