Great Lakes Restoration Initiative: Reassessment of Wildlife Reproduction and Health

Impairments in the Saginaw Bay and River Raisin Areas of Concern and Grand Traverse BayK. A. Grasman, D. Bouma, A. Mahn, J. Van Bruggen, Department of Biology, Calvin College, Grand Rapids, MI

J. Moore, L. Williams, U.S. Fish and Wildlife Service, Region 3, East Lansing, MI

Introduction

Results and Discussion

Persistent organochlorines such as polychlorinated biphenyls (PCBs) and chlorinated dioxins have

been suspected to be causing reproductive problems in the Great Lakes since the late 1950s, which

was later supported in the early 1970’s when scientists noted increased mortality and deformities

such as crossed bills in fish-eating birds living in the lakes (Grasman et al., 1998). PCBs have been

shown to cause chick edema disease in chickens, and there is also strong evidence to suggest that

they cause an equivalent illness known as GLEMEDS (Great Lakes Embryo Mortality, Edema, and

Deformities Syndrome) in fish-eating birds (Gilbertson et al., 1991). Another organochlorine,

dichlorodiphenyltrichlorethane (DDT) and its breakdown product dichlorodiphenyldichlorethylene

(DDE) are known to induce eggshell thinning in Great Lakes waterbirds (Grasman et al., 1998). In

addition, organochlorines have been shown to cause decreased immune function in Great Lakes

fish-eating birds. In previous studies, scientists have shown that birds nesting in areas with higher

concentrations of organochlorines than those nesting at reference sites had suppressed t-cell

mediated immunity (Grasman et al., 1996).

In 1972, the Great Lakes Water Quality Agreement was created between the United States and

Canada; since then, it has expanded to include 14 Beneficial Use Impairments to measure the

health of the Great Lakes (Environment Canada, 2013). These guidelines are used in designating

and maintaining Areas of Concern (AOCs) which are sites with severe degradation or high

contamination in the Great Lakes. The current study monitors the River Raisin and Saginaw Bay

AOCs as part of the US Fish and Wildlife Service’s efforts under the broader Great Lakes

Restoration Initiative and reassesses two Beneficial Use Impairments: 1) Degradation of fish and

wildlife populations and 2) Bird or animal deformities or reproduction problems. In 2014, Bellow

Island in Grand Traverse Bay was added to this study due to an unusual mixture of contaminants.

This site has a high level of dioxin toxic equivalents (TEQs), a relatively low concentration of

PCBs, and a high level of residual DDE. The study compared the data obtained from these polluted

sites to reference sites at the Pipe Island Twins, Chantry Island, and Two Tree Island in Lake Huron

(Figure 1).

This study reexamined the reproduction, growth, and immunological responses of colonial

waterbirds (herring gulls, Caspian terns, and black-crowned night herons) in the Great Lakes in an

attempt to evaluate the effects of these pollutants. To this end, the study employed the endpoints of

embryonic mortality, deformities, reproductive success, chick growth, T cell-mediated immune

function (phytohemagglutanin (PHA) skin test), and antibody response (sheep red blood cell

(SRBC) test). The PHA skin test and the SRBC test are well known measures of immune function

in avian test subjects (Grasman, 2002).These tests are also known to be ecologically relevant as

studies have shown that wild songbirds demonstrating a high immune response correlates with

higher survival and establishment of new populations (Moeller and Saino 2004; Moeller and

Cassey 2004).

Embryonic Nonviability

• Three egg herring gull nests were marked during laying. During mid incubation (21 days),

embryonic viability was assessed using an Avitronics Digital Egg Monitor (Cornwall England),

which measures the embryo's heart rate. Nonviable eggs were opened to determine infertility,

stage of development, and deformities.

Growth and Reproductive Success

• Body mass and size measurements were taken at 3 and 4 weeks of age in order to measure the

rates of growth (or degradation) of chicks in the colonies.

• Chick productivity in enclosed nests was measured by counting the number of chicks alive at 3

weeks divided by the number of nests.

Immune Function

• T-cell mediated immune test - To test the T cell-mediated immunity, 3 week gulls and terns or 2

week herons were injected with PHA (a lymphocyte activator) in phosphate buffered saline

(PBS) in one wing web and a placebo of PBS in the other wing web. Wing web thickness

measurements were taken with pressure sensitive calipers before and one day after injections.

The PHA stimulation index was calculated as the difference between the change in the PHA

wing web thickness and the change in the control wing web thickness.

• Antibody response test- A sheep red blood cell (SRBC) suspension was injected into the wing

of 3 week old chicks to stimulate antibody production. The hemagglutination assay was

performed with plasma from 4 week old chicks by serial dilutions with addition of SRBCs.

Total antibody response titer value was determined as the highest dilution where agglutination

formed between the antigen and antibody. IgG antibody response was assessed by denaturing

IgM antibodies with 2-mercaptoethanol prior to the addition of SRBCs.

Embryonic Nonviability and Deformities

Embryonic nonviabiliy rates in late incubation herring gull eggs were elevated at

Monroe in the River Raisin AOC (multi-year mean of 7.7%), SB CDF (4.5%) and L.

Charity Is. (7.4%) in the Saginaw Bay AOC, and Bellow Is. in Grand Traverse Bay

(5.6%) compared to reference sites (2.6% at Pipe Is. Twins, <2% at other reference sites

in studies by Craig Herbert, Environment Canada, personal communication; Fig. 2).

Infertility was the primary cause of nonviability at the Pipe Is. Twins reference site and

contributed substantially to nonviability at all AOC colonies and Grand Traverse Bay.

Embryonic death (failed development) also accounted for a sizable portion of total

nonviability at Monroe in the River Raisin AOC and L. Charity Is. in the Saginaw Bay

AOC. Crossed-bill deformities were found in two herring gull chicks at the River Raisin

AOC during 2012 (7mm offset of the upper and lower bill) and 2013 (4mm offset; Fig

3). Embryonic nonviability and deformities such as crossed bills are characteristic of

GLEMEDS and are associated with exposure to PCBs and dioxins (Gilbertson et al.

1991, Grasman et al. 1998).

Reproductive Success

Reproductive success at AOCs was variable across years, but consistently strong at

reference colonies. Gull populations that produce at least 0.75 chicks/nest are usually

stable (Kadlec & Drury 1968). In 2010-14, gulls in the Saginaw Bay AOC reproduced at

sustainable rates (0.75-2.18 chicks/nest) (Fig. 4A). Productivity at the Monroe colony

fluctuated between the years, with good productivity in 3 out of 5 years, but complete

failure in 2010. Gulls in Grand Traverse Bay had a good productivity, comparable to

that of the reference site. Caspian terns on Charity R. had low productivity in 2010 (0.3

chicks/nest), followed by good productivity in 2014 (0.95 chicks/nest). Higher

productivities were observed on the SB CDF (Fig. 4B) until 2014, when productivity

fell to 0.2 chicks/nest. Previous studies of this state threatened species have shown low

recruitment of young terns raised in Saginaw Bay and other highly contaminated

ecosystems (Mora et al. 1993). In 2014, the reference colony nests were lost due to

flooding from and higher water levels and weather.

Environment Canada. 2013. History of the Great Lakes Water Quality Agreement.

Gilbertson, M., T. Kubiak, J. Ludwig, and G. Fox. 1991. Journal of Toxicology and Environmental Health 33:455-520.

Grasman, K.A., G. Fox, P. Scanlon, and J. Ludwig. 1996. Journal of Toxicology and Environmental Health 104:29-842.

Grasman, K.A., P. Scanlon, and G. Fox. 1998. Environmental Monitoring and Assessment 53:117-145.

Grasman, K.A. 2002. Integrative and Comparative Biology 42: 34-42

Moeller, A. and P. Cassey. 2004. Journal of Animal Ecology 73:1035-1042.

Moeller, A. and N. Saino. 2004. Oikos 104:299-304.

Herring gulls, Caspian terns, and black-crowned night herons in the Saginaw Bay

AOC exhibited adverse health effects, including a decreased immune responses and

(or) decreased productivity, which were consistent with past studies:

• Embryonic non viability was higher at the AOCs

• Chick growth was significantly less at Little Charity

• AOC has less reproductive success

• Suppressed T- cell mediated immunity in herring gulls, Caspian terns, and black-

crowned night herons

• Suppressed total antibody and IgG response in herring gulls

Herring gulls at the River Raisin AOC showed impairments in immune responses

and reproductive success, consistent with past studies

• Suppressed T cell-mediated immune response

• Suppressed total antibody and IgG response

• Decreased growth rates for chicks, excluding 2013 and 2015

• Higher than normal embryonic nonviability

• Complete reproductive failure in 2010 and decreased productivity in 2012

Herring gulls at Grand Traverse Bay, a site with high concentrations of TEQs and

DDE, showed impairments in immune responses and growth

• Suppressed T cell-mediated immune response

• Suppressed total antibody response

• Low growth rate between three and four weeks of age

This project was funded by the Great Lakes Restoration Initiative Funding for Area of Concern and Remediation and

Restoration of Contaminated Sediments through the U. S. Fish and Wildlife Service, Region 3. Ted Ludwig, Jim

Ludwig, Dave Best, and Therese Best provided logistical support in the field while Calvin VanOpstall provided

contributed greatly to lab work. Research stipends from Calvin College were provided to Jenna Van Bruggen by the Star

Fellowship and to Alaina Mahn by the Dornbush De Vries Family Student Research Fellowship.

Conclusions

Materials and Methods

Acknowledgements

References

Figure 2. Embryonic nonviability in herring gulls at the

Saginaw Bay and River Raisin AOCs and Grand Traverse

Bay during 2010, 2012-14. Numbers on bars are egg

totals. “Undetermined” categorizes infertile or failed eggs

that were difficult to distinguish due to addling or rot.

Figure 6. Mean phytohemagglutinin (PHA) stimulation index, a measure of the T cell-mediated immunity, in pre-fledging

fish-eating birds at AOCs, Grand Traverse Bay, and reference sites in 2010-15. Bars sharing the same letter were not

statistically different by Tukey’s test (p<0.05). Numbers on bars indicate sample sizes. Error bars indicate the standard error.

Figure 4. Reproductive productivity (number of 3 week old chicks surviving per nest) at AOCs, Grand Traverse

Bay, and reference sites in 2010-15. For herring gulls, productivity of 0.75 chicks/nest or greater is necessary to

maintain a stable population (Kadlec and Drury 1968).

Figure 1. Colonial waterbird colonies studied in Great Lakes

AOCs, other contaminates sites, and reference areas for this

assessment of health and reproduction in fish-eating birds during

2010-15.

Figure 3. Crossed-billed herring gull chicks (3 weeks old) found at

the Monroe colony in the River Raisin AOC during 2012 (A) and

2013 (B).

(A) (B)

Figure 5. Mean growth in body mass between 3 and 4 weeks of age in pre-fledgling fish-eating birds at Great Lakes AOCs, Grand

Traverse Bay, and reference sites in 2010-15. Normal average growth rates are 4-18 g/day for Caspian terns and 14-20 g/day for

herring gulls (Grasman et al. 1996). Numbers on bars are sample sizes. Error bars indicate standard error.

A) Herring GullTwo-way ANOVA

site: p < 0.0001

year: p < 0.0015

B) Caspian Tern Two-way ANOVA

site: p <0.0001

year: p < 0.094

Two-way ANOVA

site: p < 0.0001

year: p < 0.0001

A) Herring Gull B) Caspian Tern Two-way ANOVA

site: p = 0.0262

year: p =0.1342

Growth

Growth in herring gull chicks varied by year and site (Two-way ANOVA for site and year:

p<0.0001) (figure ?). Gulls at Monroe and Little Charity Island had lower growth than the

reference sites during 2011-2012 and 2014 and the overall mean growth for Little Charity was

significantly lower than Pipe Island Twins (Tukey p<0.05). In 2013 and 2015 the growth

trends differed in that the mean growth for the reference site was lower than the other sites

with the exception of Bellow Island in 2015. Bellow Island had lower growth than the

reference site in both 2014 and 2015 (Figure 5A).

The normal mean growth rate for herring gulls is 14-20 g/day (Grasman et al., 1996). In our

study, normal growth rates were achieved in 2010 at the CDF, in 2011 and 2014 at the Pipe

Island Twins, and in 2013 and 2015 at Monroe, CDF, and Little Charity; weight loss was seen

at the CDF in 2014 and at Little Charity Island in 2012 and 2014 (Figure 5A).

The normal mean growth rate for Caspian terns is 4.0-18.0 g/day (Grasman et al., 1996) In

2011 through 2013 at the reference site, terns ranged from 3.6-10.8 g/day in mean growth rate.

However, at the CDF between 2010-2012 and in 2014, terns only averaged between 2.0-5.0

g/day for mean growth rate (Figure 5B).

T Cell-Mediated Immunity

Herring gulls and Caspian terns had a lower T-cell mediated immune response at all AOC sites

as well as at Grand Traverse Bay compared to the reference sites at Pipe Island Twins and Two

Tree Island (Two-way ANOVA p<0.0001) (Figure ?). The mean PHA skin response of gulls

was 53-57% lower at AOCs compared to the Pipe Island Twins. Gulls in Grand Traverse Bay

had a 50% lower mean PHA skin response than the reference site at Pipe Island Twins (Tukey

p<0.05). Caspian terns at AOCs had a 46-50% lower response as compared to Two Tree

Island. Black-crowned night herons on the CDF showed a 39% lower T-cell response

compared to herons at the Chantry Island reference site (Figure 6)

Antibody Titer of Plasma Samples

Herring gulls at the AOC sites and Grand Traverse Bay had significantly lower total antibody

response and lower IgG response than the herring gulls at the Pipe Island Twins reference site

(Two-way ANOVA p<0.0001, Tukey p<0.05). At AOCs mean titer values for total antibodies

(4.3-4.9) and IgG responses (2.1-3.0) were significantly lower than values for the reference

site (Total: 5.6, IgG: 4.1). For Caspian Terns, total and IgG responses differed by year and no

significant differences between site were found (Two-way ANOVA Total: p =0.09, IgG:

p=0.066) (Figure 7).

Other Polluted Sites

Areas of Concern

C) Black-crowned

Night Heron

T-test

p < 0.05

(N/A)(N/A)(N/A)(N/A)(N/A) ( N/A ) (N/A) (N/A) (N/A) (N/A) (N/A)

Study Sites (See Fig. 1):

Reference

Pipe Island Twins (gulls 2011-15)

Two Tree Island (terns 2011-13)

Chantry Island (herons 2010)

River Raisin AOC

Monroe (gulls 2010-2015)

Grand Traverse Bay

Bellow Island (2014-15)

Saginaw Bay AOC

Confined Disposal Facility (CDF)

(gulls 2010-15, terns 2010-14, herons 2010)

Little Charity Island (gulls 2010,

2012-15 and terns 2015)

Charity Reef (terns 2010, 2014)

B) Caspian TurnA) Herring Gull

SITE

OLD OLD

Figure 7. Mean total and IgG antibody titer in pre-fledging

herring gulls at AOCs, Grand Traverse Bay, and reference site

in 2010-2015. Bars sharing the same case and letter were not

statistically different by Tukey’s test (p<0.05). Numbers on bars

indicate sample sizes. Error bars indicate the standard error.

Two-way

ANOVA

Site: p<0.001

Year: p=

0.0015

OLD

OLD