Estrogenicity in the Sacramento and San Joaquin river system

Identifying the Causes of Feminization of Chinook Salmon in the Sacramento and San Joaquin River System

University of California, Berkeley

University of California, Riverside

Applied Marine Sciences

Funding Agency: California Bay-Delta Authority

[Project Overview] [Participants] [Objectives] [Background] [Results] [Current and future work] [Publications and presentations ] [Relevant Links] [Contact Information ]

PROJECT OVERVIEW

Some organic compounds such as steroid hormones from livestock and surfactant metabolites can induce feminization of aquatic species when present at relatively low concentrations in water. The purpose of this project is to assess the estrogenicity of surface waters in the Sacramento/San Joaquin river system using both bioassays and targeted chemical analysis for steroid hormones and alkylphenolethoxylates. Additionally, samples exhibiting significant biological activity unexplained by known estrogenic compounds will be subjected to chemical fractionation and exhaustive analysis to identify the causative agents. Results of the research will be used to identify cost-effective approaches for controlling or preventing feminization of salmon and other important fish species. This research project is relevant to CALFED’s efforts to protect and restore Chinook salmon and other critical species because it provides much needed information about a family of chemical stressors that have not received much attention from the CALFED program.

[back to top]

PARTICIPANTS

· David L. Sedlak, UC Berkeley, Project Leader/Principal Investigator. [contact information]

· Daniel Schlenk, UC Riverside, Principal Investigator. [contact information]

· Robert Spies, Applied Marine Sciences, Principal Investigator. [contact information]

· Ramón Lavado Parra, UC Riverside, Participating Investigator. [contact information]

· Jorge E. Loyo-Rosales, UC Berkeley, Project Manager/Participating Investigator. [contact information]

· Edward P. Kolodziej, University of Reno, External Advisor. [contact information]

[back to top]

OBJECTIVES

The primary objective of the project is to determine if endocrine-disrupting chemical contaminants are present in the Sacramento-San Joaquin river systems at concentrations capable of feminizing Chinook salmon or other fish species. A secondary objective is to develop and test a biomarker to assess the prevalence of feminized male fish in this river system.

Specifically, the project addresses the following questions:

· What (xeno)estrogens are present in the Sacramento-San Joaquin River system and do concentrations exceed levels at which biological effects have been observed?

· What is the total concentration of estrogen agonists in water samples and what is their identity?

· What is the potential for water and sediment samples from the Sacramento-San Joaquin River system to induce feminization in rainbow trout or Chinook salmon?

· What are the main sources of the most potent (xeno)estrogens and what is the temporal and spatial variability of these (xeno)estrogen discharges?

· Will the identified (xeno)estrogenic chemical contaminants feminize salmon?

· What species of fish other than salmonids demonstrate evidence of feminization in the Sacramento-San Joaquin River system?

[back to top]

RESULTS

Summary. In the first phase of the project, more than 110 surface water samples were collected from 16 different locations throughout the Sacramento/San Joaquin river system on 8 dates over a 14-month period. The samples were analyzed by GC/MS-MS for steroid hormones and alkylphenol ethoxylates (APEs), and their vitellogenin-induction capacity was evaluated with both in vivo and in vitro bioassays. Results of this survey showed that steroid hormones were either absent or present at very low concentrations, usually below quantitation limits, in most samples. APEs were present in all of the samples, but total concentrations were always at least an order of magnitude below levels believed to cause estrogenic responses. In contrast, high levels of estrogenic activity were observed in the bioassays in several sites, such as the Napa and Sacramento River sites. These findings raise the possibility that compounds other than the APEs and the steroid hormones analyzed are responsible for the estrogenic activity. Initial TIE results suggest that the compound(s) responsible for the estrogenicity in the Napa River are different from the compound(s) in the Sacramento River site. Additionally, Chinook salmon was found to be only slightly more sensitive than rainbow trout to E2, suggesting that the latter species is a good surrogate for salmon in estrogenicity tests.

Survey results. Survey results showed that most of the samples contained very low amounts of the hormones measured—17α-estradiol (aE2), 17β-estradiol (E2), estrone (E1), estriol (E3), progesterone, medroxyprogesterone, testosterone, and androstenedione. The estrogenic hormones were seldom detected above quantitation limits (0.1-0.5 ng/L): E1 was quantifiable in only 28% of the samples, E2 14%, aE2 6% and E3 3%. Moreover, the concentration of these hormones was below 1.5 ng/L in 95% of the samples. In contrast, the other family of estrogenic compounds measured, the alkylphenol ethoxylates (APEs), was detected in most samples; 79-84% of the samples contained these compounds in concentrations above quantitation limits. APEs analyzed included nonylphenol (NP), nonylphenol monoethoxylate (NP1EO), nonylphenol diethoxylate (NP2EO), octylphenol (OP), octylphenol monoethoxylate (OP1EO), and octylphenol diethoxylate (OP2EO). Reflecting commercial use patterns, NPEOs were present in higher concentrations than OPEOs; e.g. the 90th percentile for NP was 56 ng/L, whereas for OP it was 3.5 ng/L. In contrast to the chemical analysis, the bioassays showed relatively high estrogenic activity in 6 of the 16 sites investigated. Activity tended to be higher in the summer months, and the geographic pattern was consistent throughout the sampling period: 2 sites showed in vitro but not in vivo activity, 2 had in vivo but not in vitro activity and 2 produced both in vitro and in vivo responses.

Correlation between chemical analysis and bioassay response. Results of the chemical analysis did not correlate with the bioassays. Estrogen equivalents (chem EEQs) were calculated from the chemical concentrations of the estrogenic compounds using literature values (a high and a low estimate were calculated) and no correlation was found between chem EEQs and bioassay EEQs (bio EEQs). The analysis of individual sites did not show correlation either. Only in one of the sites there was some correlation, but even in this case the chemicals measured could only explain <1% of the activity observed in the bioassay. No correlation was found between the bio EEQs and any other of the parameters measured, i.e. dissolved organic carbon, suspended solids, nitrate, pH, temperature, or conductivity.

Storm events. Samples from 3 different storm events have been collected and analyzed to study the effect of precipitation on the estrogenicity of the sites. In the sites where water discharge information from USGS gauges was available and estrogenicity levels were above quantitation limits, in vitro bioassay response appeared to be related to discharge magnitude. In the case of the storm events, no correlation was observed between chem EEQs and bio EEQs.

TIE. Samples from the 2 sites with consistently high in vivo and in vitro estrogenicity—Sacramento River in Delta and Napa River—were chosen for TIE. Additionally, a sample from the Mokelumne River site, which had no detectable activity in the bioassays, was spiked with 100 ng/L E2 and used as a positive control. These samples were extracted on C18 discs as before, but the discs were eluted with a sequence of methanol/water mixtures with increasing concentrations of methanol: 20, 40, 60, 80 and 100%. All fractions were analyzed with the bioassays and GC/MS. The spiked control and the Sacramento River sample showed in vitro activity only in the 60% fraction, whereas the Napa sample showed in vitro activity in both the 80 and 100% fractions, with the 100% fraction showing almost an order of magnitude higher response than the 80%. In vivo activity was observed in a larger number of fractions: 40, 60 and 80% for the spiked control, fractions 60 and 80% for the Sacramento River, and fractions 60, 80, and 100% for Napa. In the spiked sample and the Sacramento River, in vivo activity was very similar in all fractions, whereas in Napa activity increased from the 60 to the 100% fraction. Therefore, the 60% fractions of the spiked control and the Sacramento River, and the 100% fraction for Napa were chosen for HPLC fractionation. GC/MS analysis of the different fractions confirmed the presence of E2 in the estrogenic fractions of the spiked sample (Mokelumne). As previously observed for the whole samples, E2 was not detected in any of the fractions from the Sacramento and Napa rivers.

The fractions exhibiting the highest estrogenic activity (i.e., the 60% methanol fractions from the Sacramento River Delta and the spiked Mokelumne sample, and the 100% methanol fraction from the Napa River) were subjected to a more precise separation using high-performance liquid chromatography (HPLC). HPLC fractionation was performed in a C18 reversed-phase column using the following gradient: 0-3 min, 40% methanol; 3-30 min, 40-100% methanol. 3-mL (or 3 min) samples were collected and analyzed using the in vitro and in vivo bioassays. The highest in vitro activity was observed in fraction 9 of the spiked control, where E2 was expected to elute (confirmed with a separate E2 injection); additionally, activity was also observed in fractions 8 and 10 of the spiked sample, but the source of the activity is unknown. From the unspiked samples, only fraction 8 of the Sacramento River and fraction 6 of the Napa showed in vitro activity and it was present at relatively low levels (under 5 ng/L); the source of the activity is also unknown.

Selected HPLC fractions were analyzed at the Southern Nevada Water Authority for a suite of pharmaceuticals, personal care products, phytoestrogens, pesticides (including pyrethroids) and hormones. Low concentrations (< 5 ng/L) of some steroid hormones and pharmaceuticals were detected in some fractions, but they cannot explain the activity detected in the bioassays.

CURRENT AND FUTURE WORK

Delta samples. Analysis (both GC/MS and bioassays) of a series of samples collected in April 2008 in the Sacramento/San Joaquin Delta by Prof. Inge Werner’s group from UC Davis is in progress. Prof. Werner studies pelagic organism decline in the Delta, and her group monitors a series of sites in the Delta for organic pollutants, including pyrethroids, and they also conduct toxicity tests using delta smelt and inland silversides. GC/MS analysis of the samples showed no steroid hormones or APEs in most Delta sites. The sites at the mouths of the Sacramento and Napa rivers had very low concentrations of some of the hormones and low concentrations of the APEs, consistent with our previous observations, but the levels detected would not be expected to cause feminization of sensitive fish species. VTG analysis, both in vivo and in vitro, is still in progress.

TIE. A new round of TIE analysis is currently being conducted. TIE analysis using both in vitro and in vivo bioassays will be conducted for the same sites discussed above (Sacramento River in Delta and Napa River) using larger sample volumes (10 L) to improve sensitivity, especially in the HPLC fractions. TIE analysis will also be conducted on water from the site in the Tuolumne River because it consistently shows high (often the highest) in vitro responses of all the sites. Additionally, aliquots of all TIE samples will be extracted and stored for full scan GC/MS and LC/QToF to identify unknowns in the samples that show activity.

Estrogenicity sources in Napa and Tuolumne. A more comprehensive investigation of the estrogenicity sources in Napa and Tuolumne will be conducted. It will include sampling additional sites in both the Napa and Tuolumne rivers for GC/MS-MS, bioassay analysis and full scan GC/MS and LC/QToF to identify unknowns in samples showing estrogenic activity.

[back to top]

LINKS

· University of California, Berkeley

Department of Civil and Environmental Engineering

· University of California, Riverside

Department of Environmental Sciences

· Applied Marine Sciences

· CALFED

[back to top]

PUBLICATIONS AND PRESENTATIONS

· Quantification of steroid hormones with pheromonal properties in municipal wastewater effluent (Kolodziej et al. 2003; DOI: 10.1897/03-42)

· Platform presentation. Kolodziej, E.; Sedlak, D.: Agricultural sources of steroids hormones in surface waters. 4th Biennial CALFED Science Conference. Sacramento, CA. October 2006.

· Platform presentation. Lavado, R.; Loyo-Rosales, J. E.; Kolodziej, E. P.; Sedlak, D. L.; Schlenk, D.: Evaluation of steroid estrogen and estrogenic activity in surface waters from Central California. Southern California SETAC Annual Meeting. Los Angeles, CA. April 2007.

· Poster presentation. Lavado, R.; Loyo-Rosales, J. E.; Kolodziej, E. P.; Sedlak, D. L.; Schlenk, D.: Evaluation of steroid estrogen and estrogenic activity in surface waters from Central California. PRIMO (Pollutant Responses in Marine Organisms) 14th International Symposium. Florianopolis, Brazil. May 2007.

· Platform presentation. Kolodziej, E. P.; Loyo-Rosales, J. E.; Sedlak, D. L.: Steroid hormone sources and fate in agricultural watersheds. Micropol and Ecohazard Conference. Frankfurt, Germany. June 2007.

· Poster presentation. Lavado, R.; Loyo-Rosales, J. E.; Kolodziej, E. P.; Sedlak, D. L.; Schlenk, D.: Evaluation of steroid estrogen and estrogenic activity in surface waters from Central California. SETAC 28th Annual Meeting. Milwaukee, WI. November 2007.

· Platform presentation. Loyo-Rosales, J. E.; Sedlak, D. L.; Lavado, R.; Schlenk, D.: Estrogenic activity in Central Valley rivers. 18th Annual Meeting of the Northern California Regional Chapter of SETAC. Berkeley, CA, May 2008.

· Platform presentation. Floyd, E.; Lavado, R.; Loyo-Rosales, J.; Kolodziej, E.; Sedlak, D.; Snyder, S. A.; Vanderford, B. J.; Schlenk, D.: Detection of estrogenic activity in surface waters from the Central Valley of California. Platform presentation. 2008 Annual Meeting of the Southern California Regional Chapter of SETAC. Dana Point, CA, May 2008.

CONTACT INFORMATION

David L. Sedlak, Ph.D.

Professor of Environmental Engineering

Department of Civil and Environmental Engineering

657 Davis Hall

University of California

Berkeley, CA 94720-1710

(510) 643-0256

sedlak@ce.berkeley.edu

http://www.ce.berkeley.edu/~sedlak/

Daniel Schlenk, Ph.D.

Professor of Aquatic Toxicology