Ecotoxicology (2016) 25:608–621 DOI 10.1007/s10646-016-1620-3

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL D. Iwanowicz1 • M. C. Black2 • V. S. Blazer1 • H. Zappia3 • W. Bryant4

Accepted: 28 January 2016 / Published online: 18 February 2016 Ó Springer Science+Business Media New York (outside the USA) 2016

Abstract During the spring and fall of 2001 and the spring of 2002 a study was conducted to evaluate the health of the largescale stoneroller (Campostoma oligolepis) populations in streams along an urban land-use gradient. Sites were selected from a pool of naturally similar sub-basins (ecoregion, basin size, and geology) of the Mobile River basin (MRB), using an index of urban intensity derived from infrastructure, socioeconomic, and land-use data. This urban land-use gradient (ULUG) is a multimetric indicator of urban intensity, ranging from 0 (background) to 100 (intense urbanization). Campostoma sp. have been used previously as indicators of stream health and are common species found in all sites within the MRB. Endpoints used to determine the effects of urban land-use on the largescale stoneroller included total glutathione, histology, hepatic apoptosis, condition factor and external lesions. Liver glutathione levels were positively associated with increasing urban landuse (r2 = 0.94). Histopathological examination determined that some abnormalities and lesions were correlated with the ULUG and generally increased in prevalence or severity with increasing urbanization. Liver macrophage aggregates were positively correlated to the ULUG. The occurrence of

& D. Iwanowicz [email protected] 1

U.S. Geological Survey, National Fish Health Research Laboratory, 11649 Leetown Road, Kearneysville, WV 25430, USA

2

Department of Environmental Health Science, University of Georgia, 150 E. Green Street, Athens, GA 30602, USA

3

Center for Threat Preparedness, 2007 Riffee Ridge, Given, WV 25425, USA

4

CK Associates Environmental Consultants, 17170 Perkins Road, Baton Rouge, LA 70810, USA

123

nucleosomal ladders (indicating apoptotic cell death) did not correspond with urban intensity in a linear fashion. Apoptosis, as well as prevalence and severity of a myxozoan parasite, appeared to have a hormetic dose–response relationship. The majority of the biomarkers suggested fish health was compromised in areas where the ULUG C 36. Keywords Biomarkers  Largescale stoneroller  Campostoma  Liver glutathione  Urban land-use gradient (ULUG)

Introduction Urbanization impacts many freshwater systems, particularly small streams. These impacts include effects on stream geomorphology (Paul and Meyer 2001; Tate et al. 2005; Ramirez et al. 2012), hydrology (Elkin et al. 2013), stream temperature (Hester and Bauman 2013) and water quality (Kim et al. 2013). Concentrations of nutrients, pesticides, organic chemicals and metals are often elevated in urban runoff and treated wastewater (Iannuzzi et al. 2011). Alterations in physical habitat and water quality due to urbanization are also associated with changes in aquatic biota (Weaver and Garman 1994). Understanding the effects of numerous stressors on aquatic assemblages is critical to preserve, rehabilitate, and manage ecosystems as urbanization continues (Ramirez et al. 2012; Bellucci et al. 2013). Many urban studies have used single measures such as population density, percent urban land, or percent imperviousness, to gauge effects caused by urbanization. Interpretation of these results varies depending on the selected endpoint (Tate et al. 2005). In an attempt to address the complexity of urbanized areas (i.e., differing infrastructure, human populations and socioeconomic characteristics) and

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

assign a single impact value, the urban land-use gradient (ULUG) was developed (McMahon and Cuffney 2000; McMahon et al. 2001). The ULUG is a multimetric indicator of urban intensity that can range from 0 (background) to 100 (intense urbanization). It combines individual condition measures, and together provides distinct information on all of the dimensions (infrastructure, human populations, and socioeconomic characteristics) of these complex systems (McMahon and Cuffney 2000; Tate et al. 2005). From 1995 to 2000, the U.S. Geological Survey’s National Water-Quality Assessment Program (NAWQA) conducted a five-year investigation of the Mobile River Basin (MRB). The MRB is the sixth largest basin in the Nation and encompasses 44,000 square miles in the states of Alabama, Georgia, Mississippi and Tennessee. Of this, 71 % of the MRB lies within Alabama (Johnson et al. 2002). Chemical and physical data were collected from selected areas throughout this study unit to describe water quality (McPherson et al. 2003; Atkins et al. 2004). In these preliminary studies it was determined that most urban streams and rivers in the MRB had high levels of insecticides (i.e.: chlorpyrifos, diazinon, malathion, DDT, dieldrin and carbaryl), as well as polychlorinated biphenyls and volatile organic compounds (VOCs). The most frequently detected VOCs included trichloromethane (chloroform—a byproduct of water chlorination), tetra- and trichloroethylene (commercial and industrial solvents and degreasers), cis-1,2dichlroethene (a metabolite of trichloroethylene), methyl tertbutyl ether (gasoline oxygenate), and the hydrocarbons benzene and toluene (Atkins et al. 2004). Organochlorines and trace elements were also measured in fish tissue and streambed sediments. Although sites with a ULUG C 36 were highest in most organochlorines and trace elements, sediment from selected sites with a ULUG \ 11 were highest in certain contaminants (o,p’-DDE, p,p’-DDE, o,p’-DDT, octochlorbiphenyl, 3-5-dichlorobiphenyl, and cis Nonachlor) from agricultural application. Fish tissues from these same sites were highest in the trace elements mercury, arsenic, cadmium, lead, aluminum, barium, copper, and manganese and the organochlorine p,p’-DDE. Concentrations of chlordane and hepatochlor epoxide in fish tissue increased with increasing ULUG (Zappia 2002). Although the effects of urbanization on fish community structure have been studied in a number of areas (Weaver and Garman 1994; Giddings et al. 2006; Fischer et al. 2012), there are few studies that evaluate the cumulative effects of urbanization on fish health. These studies tend to measure environmental health in terms of fish populations, population composition and tissue contaminant concentrations (Zandbergen 1998). An effective alternative is to assess the health of a tolerant, short-lived sentinel fish species. Largescale stonerollers (Campostoma oligolepis) are widespread and abundant throughout the Central and Mideastern United

609

States, and are often a major component of stream fish communities (Mettee et al. 1996). Stonerollers are herbivorous, consuming primarily periphyton and detritus (Hill and Napolitano 1997), that bioaccumulate PCBs and anthropogenic contaminants such as sewage derived organic matter, much higher than primary producers (Hill and Napolitano 1997; Colombo et al. 2007). Stonerollers also have a relatively small home range and are both abundant and geographically widespread. Since they are a food resource for many species, stonerollers may provide early warning of problems and potential adverse effects for a range of consumers (Burger et al. 2005; Azevedo et al. 2012). The goal of this study was to determine a method to examine the impacts of fish health in regards to different levels of urbanization. The primary objectives of this study was to (1) determine if there was an association between largescale stoneroller health metrics and ULUG values previously determined within the MRB, and (2) determine if stoneroller health is impaired along the ULUG gradient.

Materials and methods Site selection As part of the urban gradient studies, 30 sites were sampled within the Ridge and Valley ecoregion of Alabama to assess responses of fishes, invertebrates, and algae to urbanization (Cuffney et al. 2005; Meador et al. 2005; Zappia et al. 2005). In addition, this area was part of a study examining changes in aquatic biological communities, physical habitat, hydrology, temperature and water quality along gradients of urban intensity, in multiple environmental settings throughout the United States (Tate et al. 2005). Sites were selected based on ecoregion, basin size and geology. The ULUG was derived for 30 sites by McMahon and Cuffney (2000). Variables used to obtain the ULUG (Table 1) and the calculations have been reported (McMahon and Cuffney 2000; Tate et al. 2005). When appropriate, variables were normalized based on basin area (e.g., population density, percentage of basin in forest, number of toxic releases per 259 km2). Physical characteristics of each stream were an assessment of instream habitat and continuous measurements of stream stage and temperature. In-stream habitat characteristics (velocity, channel depth and width, aspect of flow, bed substrate, habitat cover, canopy closure and vegetation, and bank morphology) were assessed via standard NAWQA Program protocols (Fitzpatrick et al. 1998). Sites similar in water depth, stream width, water flow and temperature within the Birmingham, AL metropolitan (Fig. 1) area were chosen with very low (3); low (11–14); moderate (36–39) and high (70–84) ULUG values (Table 2).

123

610 Table 1 Generalized basin variables used to characterize the environmental setting and urban intensity index for Birmingham, AL

D. Iwanowicz et al. Characteristica

Basin variable

Environmental frame work characteristics

Basin area Topography Elevation (m) Basin slope (% of basin area) Slope 1 %, [midpoint Slope 1 %, Bmidpoint

Landuse characteristics

Soils Hydrologic soils Minimum infiltration rate Soil drainage (% of basin area) Well drained soils Poorly drained soils Average soil volume proportion of sand U.S. Environmental Protection Agency Ecoregions Ridge and valley Southern limestone/dolomite valleys Southern shale valleys Southern sandstone ridges Plateau escarpment Southern table platoes Land use/land cover (National Land-Cover Data) Basin level Developed Forest Shrubland

Infrastructure characteristics

Road density in basin (km/km2)

Socioeconomic characteristics

Socioeconomic indices (SEI) variables

Population characteristics

Population and housing variables 1999 population density (people/km2) Percent of housing units on public sewers, 1990

a

Complete table can be found at Tate et al. (2005)

Field collection and sample preparation Largescale stonerollers were collected using a Coeffelt Mark 10 backpack electrofisherTM (Smith-Root, Vancouver, WA) in March 2001, November 2001, and March 2002 (Table 2). The two March collections coincided with the stoneroller spawning dates which occur from February to July for the southern region. Water quality variables including water temperature, pH, dissolved oxygen and specific conductance were measured with a portable minisonde 4a memory water monitoring system during each sampling, and calibrated in accordance to manufacturer protocol. Global positioning system readings were recorded for each site. The target fish number for March 2001 was 20 fish per site. To increase the level of confidence, the target fish number for November 2001

123

and March 2002 was increased to 30 fish per site. Fish were euthanized with a lethal aqueous dose of tricaine methanesulfonate (MS-222, Redmond, WA), weighed, measured (total length, TL), and examined for external and internal lesions to provide organism-level indicators. Condition factor was calculated as K = [body weight in g/(total length in mm)3] 9 105. During March 2002, small sections (50 mg) of livers were snap frozen in liquid nitrogen for glutathione and apoptosis analyses. During November 2001 and March 2002, otoliths were removed for aging measurements following protocols previously described in Lai et al. (1996). A ventral incision was made in all fish and they were preserved whole in 10 % phosphate buffered formalin and returned to the National Fish Health Research Laboratory (Leetown, WV) for histopathological examination.

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

611

were only found in one tissue, a stage of 2 was assigned if they were found in two tissues, and a stage of 3 was assigned if they were found in all three tissues. Spore enumeration Fifteen additional fish from each site were evaluated for myxozoan spore enumeration. Skin and internal organs were removed and the muscle was individually weighed. Individual fish muscles were macerated, blended with water, and then sequentially filtered through a series of sieves (500, 250 and 45 lm). The filtrate was then pelleted (12000 rpm 9 12 min) and weighed. Pellets were suspended in 5 mL of water and the spores were quantified via hemacytometer. Spores per gram of fish tissue were determined. Glutathione analysis

Fig. 1 Map showing the location of the Mobile River Basin in Alabama, Georgia, Mississippi, and Tennessee. Location of sample sites (subbasins) selected for fish health evaluation surrounding Birmingham, AL are marked with numbers. The site numbers with the corresponding ULUG value in parenthesis are: Site 1 (3.0), site 2 (3.1), site 3 (11.4), site 4 (12.0), site 5 (14.0), site 6 (36.0), site 7 (39.0), site 8 (69.7), and site 9 (83.5)

Histopathology For the cellular level indicator, histological examinations of the sagittal midline sections of the entire body were examined. All internal organs and muscle sections were examined for abnormalities. Paraffin-embedded tissue sections (6 lm) mounted on glass slides were stained with hematoxlyin and eosin, Perl’s prussian blue, or Giemsa stains (Luna 1992). The percent of individuals demonstrating liver lesions (necrosis, fibrosis, vacuolated foci, macrophage aggregates, regenerative foci, inflammation and digenetic trematodes) was calculated for each site. A recently described myxozoan parasite, Myxobolus stanlii n.sp. (Iwanowicz et al. 2013), was identified and the severity of infection was staged with an intensity ranking. Parasite infection of each fish was assigned an intensity ranking or stage from 0 to 3. This value corresponded to the number of tissues occupied by the myxozoan parasite (areas included: muscle/connective tissue, kidney, and nerve tissue). A stage of 0 was assigned if the spores were not found in any of the above tissues, a stage of 1 was assigned if they

Liver reduced glutathione (GSH) concentrations (total glutathione including GSH and glutathione disulfide) of individuals were determined by the 5,5’-dithiobis(2-nitrobenzoic acid)-oxidized GSH (GSSG) reductase recycling assay following the modified methods of Conners and Ringwood (2000) for one of the sub-cellular indicators. Briefly, frozen liver samples were thawed and weighed. Liver samples were homogenized in 10 volumes of 5 % sulfosalicyclic acid (SSA), and centrifuged (14000 rpm, 5 min, 4 °C). The supernatant was removed and diluted 1:1 with 5 % SSA and mixed with the sodium phosphate buffer containing nicotinamide adenine dinucleotide phosphate (NADPH) and dithio-bis-nitrobenzoicacid (DTNB). GSSG reductase was added and the rate of thio-nitrobenzic acid (TNB) formation was monitored at 412 nm over a 90 s interval. GSH concentrations were estimated from a standard curve and reported as nM GSH/g wet weight. Apoptosis—ligation-mediated PCR In March 2001, genomic DNA was extracted from frozen liver tissue for apoptosis screening using the QIAmp Tissue Kit (Qiagen, Germantown, MD) for a second sub-cellular indicator. Images were analyzed using an ApoAlertTM Ligand mediated PCR ladder assay kit designed by ClontechTM. The ApoAlertTM LM-PCR ladder assay kit is a semiquantitative assay that detects nucleosomal ladders generated during apoptosis. The ApoAlertTM assay made quantifying the relative extent of apoptosis in different samples from scanned gels possible. Genomic DNA (1 lg) was mixed with 1 nmol each of 24-mer and 12-mer unphosphorylated oligonucleotides in 60 lL of T4 DNA ligase buffer (Fast-link DNA ligation kit, Epicentre Technologies, Madison, WI). Oligonucleotides were annealed by heating to 55 °C for 10 min and allowing the mixture to cool to 10 °C

123

612

D. Iwanowicz et al.

Table 2 Site name, urban land use gradient (ULUG), locations, average water temperature, specific conductance (SpC), pH, and dissolved oxygen (DO) from each of the nine sites sampled surrounding Birmingham, AL Site

Site name

ULUG value

Latitude

Longitude

Avg temperature (°C)

Avg SpC (mS/cm)

Avg pH (units)

Avg DO (ppm)

March 2001 1

Little Cahaba above Trussville

3

33.38

-86.36

8.51

88

5.83

5.44

2 3

Unnamed Tributary to Big Canoe Creek Cahaba Valley Creek at Pellham

3 11

33.79 33.18

-86.49 -86.48

11.5 13.25

208 189

7.98 8.93

9.37 9.47

4

Five Mile Creek at Nevel Road

12

33.36

-87.02

5

Little Cahaba near Braggsville

14

33.57

-86.12

9.92 13.4

249

8.35

9.32

298

5.9

3.66

6

Shades Creek at Homewood

36

33.27

-86.47

12.2

233

8.51

9.45

7

Shades Creek at Mountainbrook

39

33.44

-86.84

11.6

218

9.03

8.57

8

Village Creek at East Lake

70

33.57

-86.73

15.47

340

7.13

9.6

9

Valley Creek at Cleburn Avenue

84

33.47

-86.89

15.9

453

8.46

12.05

November 2001 1

Little Cahaba above Trussville

3

33.38

-86.36

8.2

ND

8.24

12.26

6

Shades Creek at Homewood

36

33.27

-86.47

9.8

ND

8.96

11.21

8

Village Creek at East Lake

70

33.57

-86.73

13.4

ND

7.24

12.24

9

Valley Creek at Cleburn Avenue

84

33.47

-86.89

11.6

ND

8.29

15.82

March 2002 1

Little Cahaba above Trussville

3

33.38

-86.36

15.12

112

6.6

9.45

2 3

Unnamed Tributary to Big Canoe Creek Cahaba Valley Creek at Pellham

3 11

33.79 33.18

-86.49 -86.48

17.96 19.15

206 216

7.32 7.6

9.71 9.47

4

Five Mile Creek at Nevel Road

12

33.36

-87.02

18.94

268

7.51

10

5

Little Cahaba near Braggsville

14

33.57

-86.12

17.83

320

7.43

9.16

6

Shades Creek at Homewood

36

33.27

-86.47

18.65

222

7.51

9.54

7

Shades Creek at Mountainbrook

39

33.44

-86.84

17.15

208

7.47

9.56

8

Village Creek at East Lake

70

33.57

-86.73

18.92

358

7.03

9.92

9

Valley Creek at Cleburn Avenue

84

33.47

-86.89

21.03

431

8.23

11.09

ND = Specific conductivity was not collected in the November 2001 sampling

over 55 min. The mixture was then incubated at 10 °C for 10 min. 1 lL of ATP and 1.5 lL of ligase buffer were added to the mixture and incubated for 16 h at 16 °C. The reactions were diluted to a final concentration of 5 ng/lL with TE buffer (10 mM Tris–HCL, 1 mM EDTA, pH 7.5). Samples were stored at -20 °C prior to PCR. Blunt-end linkers (24-mer: 50 -AGCACTCTCGAGCCT CTCACCGCA-30 and 12-mer: 50 -TGCGGTGAGAGG-30 ) were used to amplify the nucleosomal ladders as described by Staley et al. (1997). The PCR mixture was made by adding 5 lL of 24-mer (20 pM), 7.5 lL DNA and 12.5 lL of water to puRe TaqTMReady-To-GoTMPCR Beads (Amersham Biosciences, Piscataway, NJ). Samples were overlaid with 40 lL of sterile mineral oil. Tubes were heated to 72 °C for 3 min and 1.25 U of Taq polymerase (Gibco, Bethesda, MD) were added. The PCR amplification was conducted in a Hybaid Omnigene thermo-cycler using the following parameters (ApoAlertÒ LM-PCR Ladder Assay Kit User Manual PT3172-1): holding at 72 °C for 8 min, followed by 35 cycles with 1 min at 94 °C, 3 min at 72 °C, and 5 min at 72 °C. The final extension was

123

performed at 72 °C for 10 min. Calf-thymus DNA was used as an internal control. PCR products were resolved in a 0.75 % agarose and 1.5 % synergel matrix equilibrated in Tris–Acetate-EDTA buffer (TAE; 40 mM Tris base, 20 mM glacial acetic acid, and 10 mM EDTA, pH 8.5) for 4 h at 90 V. Gels were stained with ethidium bromide and photographed on a UV transilluminator. Images of scanned gels were analyzed with TotalLab analytical software (Biosystematica, Wales, United Kingdom). For apoptosis, nucleosomal ladders were measured at 200 bp intervals (data shown at 200, 400, and 600 base pairs lengths). Statistical analysis One-way analysis of variance (ANOVA; Statistica v11.00.01, Statsoft Inc. Software) was used to determine significant differences individually for condition factor, age and ligand-mediated apoptosis (P B 0.05) among sample sites within sampling dates. The unequal N HSD post hoc test was used to identify statistical differences among sites, sex, or sampling date (Winer et al. 1991). All variables

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

were examined for potential outliers (using the interquartile range) and tested for normality (Shapiro–Wilk W-test) and all assumptions of ANOVA were met. Non-normal variables were transformed when necessary (log10 or square root). All other data were assessed by forward stepwise multiple regression models, based on ordinary least squares for continuous dependent variables and ordered logit for categorical variables. Inspection of the bivariate plot of the dependent variable and lower order independent variable indicated that this model was appropriate (Cohen and Cohen 1983). To establish possible correlations between biomarkers and sites, the Spearman’s rank correlation coefficient (rho) and Bonferroni probability (P) were calculated. For statistical significance, preliminary and final models were tested at stations where n [ 8 for all endpoints. Data were significant if P \ 0.05.

613

8 (df = 59, F = 2.75, P = 0.01). Condition factors were different among sampling dates, with the condition factor lower in the November sampling than either March sampling (df = 447, F = 5.05, P = 0.01). Condition factor did not differ between male and female largescale stonerollers from any site at any sampling date (df = 448, F = 1.36, and P [ 0.05). Fish total length (mm) and weight (g) were not significantly different among the sample sites at any date (df = 448, F = 6.23, P = 0.75). Fish weight was different among sampling dates, with fish weighing significantly less in November than either of the March sampling (df = 447, F = 4.84, P = 0.03). Fish age (years) was not significantly different among the sampling sites (df = 287, F = 2.82, P = 0.26). Cellular-level indicators Histopathology

Results Environmental measurements Water quality metrics are reported in Table 2. Specific conductance was positively correlated with ULUG in both March 2001 (df = 17, rho = 0.83, P \ 0.0001) and March 2002 (df = 17, R = 0.78, P = 0.01). Temperature was positively correlated with ULUG only in March 2001 (df = 17, rho = 0.78, P = 0.02) sampling. Temperature, dissolved oxygen, and pH were not correlated during any sampling (P [ 0.05). Organism-level indicators External abnormalities External abnormalities primarily included black spot disease on the body surface and fins due to trematode metacercariae and the presence of leeches (Table 3). A few fish exhibited reddened or hemorrhagic areas on the body surface. In March 2001 there was a positive correlation between ULUG and black spot disease on the body surface (df = 182, rho = 0.60, P = 0.00). In November 2001 fish collected at all sites had a high prevalence of black spots but they were not significantly correlated between sites (df = 119, rho = -0.02, P = 0.79). No other abnormalities were noted. In March 2002, a similar spatial trend to that of the previous year was observed in terms of black spots and leeches but they were not significant (df = 192, rho = 0.10, P [ 0.36). Physiology, condition factor, and age Condition factor was significantly different among sites only in the March 2002 collection and only between site 5 and site

Lesions were most frequently observed in the liver and muscle tissue. In the liver, necrosis, fibrosis, vacuolated foci, macrophage aggregates, regenerative foci, and inflammation were the most common histopathological observations. In the muscle, digenetic trematode metacercariae were the second most common histological observation, after Myxobolus stanlii n.sp. cysts. In March 2001 histopathological samples from all nine sites were examined (Table 4). Liver necrosis (df = 192, rho = 0.78, P = 0.01), macrophage aggregates (df = 192, rho = 0.78, P = 0.03) and fibrosis (df = 192, rho = 0.83, P = 0.008) were positively correlated with ULUG, as were digenetic trematode metacercariae in the muscle (df = 192, rho = 0.57, P = 0.04). In November 2001, microscopic pathology of largescale stonerollers from four stations was examined (Table 4). In the liver, prevalence of macrophage aggregates (df = 118, F = 10.42, P = 0.04) and fibrosis (df = 118, F = 72.44, P = 0.01) were significantly different among sampling sites. The presence of macrophage aggregates in the liver was highest in fish from Valley Creek at Powderly (ULUG 84) and lowest in fish caught in Shades Creek at Sanford (ULUG 36). Liver fibrosis was highest in fish caught from Village Creek at East Lake (ULUG 70) and lowest in fish caught from Little Cahaba at Trussville (ULUG 3) and Valley Creek at Powderly (ULUG 84). In the muscle, metacercariae were not significantly different between sites (df = 118, F = 6.69, P = 0.08) in this season. In March 2002, histopathological samples from all nine stations were examined (Table 4). In the liver, macrophage aggregates were positively correlated with ULUG (df = 236, rho = 0.71, P = 0.04). The proportion of digenetic trematode metacercariae were significantly different between sites (df = 236, F = 24.65, P = 0.002),

123

614

D. Iwanowicz et al.

Table 3 Morphometric characteristics and external lesion incidence for largescale stonerollers (Campostoma oligolepis) collected along an urban land use gradient (ULUG) in different years and seasons Site

ULUG

Sample size

Mean age ± SEM (years)

Mean weight ± SEM (g)

Mean total length ± SEM (mm)

Condition factor mean ± SEM

Individual lesions (%) Black spots

Red areas

Leeches copepods

March 2001 1

3

20

NA

13.0 ± 3.6

92.7 ± 7.2

1.1 ± 0.04

20

0

5

2

3.1

17

NA

20.3 ± 3.6

114.9 ± 4.8

1.16 ± 0.07

5.8

0

76.4

4

12.3

22

NA

32.6 ± 7.4

125.9 ± 9.0

1.19 ± 0.03

32

0

0

5

14

20

NA

24.5 ± 2.3

125.0 ± 3.6

1.20 ± 0.05

25

0

5

6

36

20

NA

21.2 ± 1.2

125.0 ± 2.1

1.25 ± 0.04

20

0

0

7

39

20

NA

22.0 ± 2.4

116.7 ± 6.6

1.13 ± 0.07

20

0

0

8

69.7

20

NA

22.1 ± 3.5

120.8 ± 5.4

1.17 ± 0.05

70

5

0

9

83.5

20

NA

15.3 ± 2.2

108.1 ± 4.1

1.08 ± 0.02

95

0

0 0

November 2001 1

3

30

1.0 ± 0.5

11.9 ± 1.0

106.2 ± 2.6

0.94 ± 0.02

86.7

0

6

36

30

1.0 ± 0.0

6.2 ± 0.4

86.63 ± 1.9

0.92 ± 0.01

100

0

0

8

69.7

30

1.0 ± 0.3

8.7 ± 0.4

100.4 ± 1.5

0.85 ± 0.01

83.3

0

0

9 83.5 March 2002

30

1.0 ± 0.0

10.63 ± 0.8

102.7 ± 2.6

0.95 ± 0.01

90

0

0

1

3

25

1.0 ± 0.3

16.5 ± 1.4

111.9 ± 3.1

1.12 ± 0.03

60

0

12

2

3.1

17

1.0 ± 0.4

33.5 ± 4.3

135.1 ± 4.6

1.25 ± 0.08

47.1

0

47.1

3

11.4

10

1.0 ± 0.3

19.4 ± 4.2

115.3 ± 7.7

1.10 ± 0.02

20

0

0

5

14

30

1.0 ± 0.4

16.6 ± 2.6

104.4 ± 4.1

1.40 ± 0.03

3.3

3.3

0

6

36

21

1.0 ± 0.2

6.5 ± .1

82.9 ± 3.7

1.01 ± 0.03

100

0

0

7

39

30

1.0 ± 0.3

12.8 ± 0.8

105.2 ± 2.2

1.07 ± 0.02

100

5

0

8

69.7

30

1.0 ± 0.5

11.0 ± 0.8

101.8 ± 2.6

0.97 ± 0.02

66.7

0

0

9

83.5

30

1.0 ± 0.2

9.0 ± 0.7

93.2 ± 2.6

1.07 ± 0.03

90

0

0

NA = Otoliths were not collected in the March 2001 sampling

with them found primarily in fish from Shades Creek at Mountainbrook (ULUG 39), Village Creek at East Lake (ULUG 70) and Valley Creek at Powderly (ULUG 84). A recently described myxozoan parasite, Myxobolus stanlii n.sp., was observed in 100 % of the largescale stonerollers collected (Iwanowicz et al. 2013). Urban landuse gradient was a poor predictor of parasite spores per gram fish tissue. Prevalence was lower at sites with a ULUG C 39. Parasite load (spores per gram fish tissue) was highest in sites 3 and 5 and lowest at site 9 (r2 = 0.65). There was not a linear response to ULUG and parasite load (Table 5). Tissue distribution of this parasite included connective tissue of various organs, renal tubules, glomeruli, macrophage aggregates within the kidney, and in nerve tissue posterior to the eye (Iwanowicz et al. 2013). Myxozoan parasites were only found in the latter location during heavy infections. Of all fish collected, 95 % were either stage 1 or 2. Fewer than 5 % of all fish collected were at stage 3 parasitism (Table 5). There was no significant difference among sites as to the parasite stages and the ULUG (df = 457, F = 0.86,

123

P = 0.38). Since season (df = 457, F = 0.72, P = 0.47) and sampling year (df = 457, F = 4.24, P = 0.10) did not show any significant differences for parasite staging data shown in Table 5 are pooled. Sub-cellular level indicators Glutathione Mean glutathione levels were highest at Sites 8 (1304.9 ± 593.3) and 9 (1702.6 ± 842.8) and lowest at Sites 2 (474.9 ± 79.9) and 5 (377.8 ± 99.6). Urban land-use gradient was positively correlated with hepatic glutathione levels (df = 191, r2 = 0.94, P B 0.001; Fig. 2). Ligand-mediated PCR In March 2001, apoptotic nucleosomal ladders (measured as percent frequency) at approximately 200, 400 and 600 base pairs were measured to detect whether apoptosis was

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

615

Table 4 Percentage of liver lesion incidences for largescale stoneroller (Campostoma oligolepis) collected along an urban land use gradient (ULUG) Site

ULUG

Sample size

Necrosis

Fibrosis

Vacuolated foci

Macrophage aggregates

Regenerative foci

Inflammation

Digenetic trematodes

March 2001 1

3

20

5

25

0

5

55

0

0

2 4

3 12

17 22

0 5

29 14

65 27

24 5

0 27

6 9

0 0

5

14

20

0

25

20

5

20

0

0

6

36

20

0

35

20

20

0

5

0

7

39

20

5

30

0

0

0

0

5

8

70

20

15

50

30

45

10

10

0

9

84

20

15

50

40

75

0

25

15

November 2001 1

3

30

0

10

7

20

50

0

7

6

36

30

0

23

27

0

0

0

0

8

70

30

3

43

17

13

0

0

37

9

84

30

0

10

0

27

0

13

7

March 2002 1

3

25

28

16

0

4

12

0

0

2

3

17

6

41

35

0

6

0

0

3 5

11 14

10 30

20 7

10 20

0 10

10 3

10 20

0 0

0 0

6

36

21

0

24

24

5

0

0

0

7

39

30

0

27

0

3

0

0

3

8

70

30

7

43

27

7

0

0

20

9

84

30

3

27

0

33

0

7

3

Table 5 Mean parasite spores per gram of fish tissue (±std dev) versus urban land-use gradient (ULUG), along with pooledvalues for parasite staging on a scale of 1–3, where the value is related to the quantity of tissues infected by the myxozoan parasite, Myxobolus stanlii n.sp. (areas include: muscle/connective tissue, kidney, and nerve tissue)

Site

ULUG

n

Spores/g fish tissue

Stage 1

Percentage 2

3

1

2

3

1

3

75

354,569 ± 192,618

60

13

2

80

17.3

2.7

2

3

34

1,220,000 ± 865,965

27

4

3

79.4

11.8

8.8

3

11

10

2,086,500 ± 386,685

6

4

0

60

40

0

4

12

22

ND

8

14

0

36.4

63.6

0

5

14

50

2,209,090 ± 865,833

43

7

0

86

14

0

6

36

71

489,223 ± 866,645

59

10

2

83.1

14.1

2.8

7

39

50

318,968 ± 846,148

46

4

0

92

8

70

80

319,012 ± 829,855

23

51

6

28.8

63.8

8

7.5

0

9

84

80

225,160 ± 829,855

35

35

10

43.8

43.8

12.4

Also shown is the percent number of fish in each stage ND not detected

occurring. When compared to the control, evidence of hepatic apoptosis was significantly different to the ULUG at all three lengths measured, 200 bp (df = 231, F = 8.85, P = 0.02), 400 bp (df = 231, F = 6.03, P = 0.04) and 600 bp (df = 231, F = 5.98, P = 0.04). Data for the 200 base pair length is illustrated (Fig. 3).

Discussion Application of the ULUG made it possible to identify a population of potential study basins with similar natural characteristics (McMahon and Cuffney 2000) based on a gradient of urban intensity and evaluate biomarker responses

123

616 2500

Mean GSH Value (nmol)

Fig. 2 Mean glutathione (GSH) values (bars represent standard deviation) measured in nanomoles (nmol) versus urban land-use gradient, showing the positive linear regression (r2 = 0.94). Points refer to experimental data and the line represents a fitted model

D. Iwanowicz et al.

2000

y = 15.028x + 302.64 r² = 0.9376

1500

1000

500

0

0

10

20

30

40

50

60

70

80

90

ULUG

in resident fish. These intensity rankings were determined by a hierarchical environmental framework that limited the variability of natural landscape factors that might confound the water-quality response. The top level of this framework was to assess the physical, chemical, evolutionary, and biogeographic processes that influence watershed landscape and in-stream conditions. The variables that comprise this top level consist of location, geological substrate, climate, elevation, stream size, stream gradient, taxa richness, species composition, and adaptive strategies. The second level of this strategy measured landscape and stream conditions influenced by human activities such as land-use (agriculture, city), effluent discharge, water withdrawal, reservoir discharge, sport and commercial fisheries, and aesthetics. The third level evaluated social values important to an area such as education levels, labor, and income. Social values can alter biogeochemical processes known to influence one or more in-stream characteristics (flow regime, physical habitat structure, chemical water quality, energy sources, biological interactions). The fourth and final level measured the instream characteristics that altered stream health by measuring indicators of both geophysical condition (erosion rates, evapotranspiration, surface permeability, runoff amount and timing, groundwater recharge, water chemistry) and biological conditions (taxa richness, taxonomic composition, individual health ecological processes, and evolutionary processes). This ranking gave a composite index that we then used to compare to the health of a sentinel fish species within the same environment. This was completed by correlating the ULUG value to a series of biomarkers (cellular to organismal level effects) of the sentinel fish species, the largescale stoneroller. Basins had to (1) have similar natural (climate, elevation, soils) characteristics that would cause variability in the physical, chemical or biological water quality response,

123

(2) represent a range of urban intensity in the population of all potential study basins, and (3) be adequate to access for all required sampling (substrate, flow, landscape features, fish health). Knowledge of the degree of urbanization in the study basin allowed the choice of subbasins that together reflected the gradient of urbanization represented across a population of all sites and the degree of urbanization could not be adequately described by land-cover characteristics alone. Due to budgetary constraints we could only assess fish health from 9 of the original 30 study drainage basins within the Mobile River Basin. The sites chosen were used in conjunction with watershed urbanization data, which identified the final choice of watersheds that had limited variation in natural characteristics and a high variation in urbanization with an intensity scale ranging from 0 to 100. An urbanization intensity of 0 indicates a pristine site. In sites selected within the MRB, the lowest ULUG value was 3. Basin characterization suggests that a project of this scale can represent a gradient of urbanization while controlling for the effects of natural characteristics at the scale of a USEPA Level IV sub-region (McMahon and Cuffney 2000) The largescale stoneroller (family Cyprinidae) was utilized as a bioindicator of stream health within varying ULUG intensities in this study. Choosing an indicator species is often a daunting task as species can differ greatly in sensitivity to various contaminants. For example, using a tolerant species rather than a sensitive species potentially underestimates the ecological effects in an area. The use of smaller fish may also not address biomagnification issues that are more evident in larger fish. Often large predator fish are used as bioindicators of fish health. This is not possible in streams such as the ones studied in the MRB, which are smaller and often lack the presence of larger predatory fish. Moreover, lower trophic level organisms are

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

617

Fig. 3 Apoptosis intensity for nucleosomal ladders with values given in percent, where the points represent the experimental data and the line represents the fitted model. a Percent values for nucleosomal ladders measuring 200 bp in length versus ULUG for all sites (panel A) and for the four sites with the highest ULUG positive linear regression (r2 = 0.72) were observed between percent ladder intensity and ULUG. b Percent values for nucleosomal ladders measuring 200 bp in length versus the four highest ULUG sites (range 36.00–83.54). Data fit a positive linear regression model (r2 = 0.99). Points refer to the experimental data and the line represents a fitted model

often more abundant and less mobile than higher level species (Burger et al. 2005). The largescale stoneroller was the only fish that could be found at all sites selected within the ULUG study conducted at the MRB. The central stoneroller (C. anomalum), a closely related species, has also been used as an indicator species (Birge et al., 2000). It is possible that the use of relatively small fish, rather than the larger fish often used for assessments, may provide better estimates of regional extent of contamination in streams and small rivers because of their wide distribution (Lazorchak et al. 2003). The characteristics of rapid growth, a short lifespan and widespread distribution make largescale stonerollers good candidates for sentinel species. They occupy flowing water habitats ranging from small streams to large rivers, and are tolerant species that dominate the urban fish assemblage in the Birmingham area (Meador et al. 2005).

In the present study, we found that liver glutathione concentrations were positively correlated to the ULUG gradient (r2 = 0.97, P B 0.001). Our results suggest that largescale stonerollers were exposed to increasing levels of contaminants that initiate oxidative stress across the ULUG gradient (Srikanth et al. 2013). Glutathione is a non-enzymatic antioxidant that scavenges free radicals, facilitates intracellular transport, detoxifies electrophiles and maintains thiol-disulfides (DeLeve and Kaplowitz 1991). An elevated level of glutathione may indicate the presence of toxic substances, due to the need to detoxify harmful chemicals. Glutathione has been used as a biomarker of exposure in a variety of fish species (Otto and Moon 1995, Sayeed et al. 2003). Previous research in the MRB has demonstrated the presence of insecticides, PCBs, and volatile organic compounds (VOCs) in sediments. Zappia

123

618

(2002) measured organochlorines and trace elements in fish tissue and streambed sediments in the Mobile River Basin. Four of the sites in the current study were included in that survey (Table 2: sites 1, 3, 6 and 9) and the following chemicals were found at these sites: cis- and trans- chlordane, p,p’-DDE, p,p’-DDT, endrin, hexachlorobenzene, PCB’s, dieldrin, heptachlor epoxide, oxychlordane,and cisand trans-nonachlor. Their presence was attributed to urbanization and positively correlated with ULUG. Although organochlorines were found throughout the study sites, of the concentrations reported in fish-tissue, a majority of the sites sampled within the Mobile River Basin showed a potential for adverse effects to fish within these streams (Zappia 2002). This was believed to be primarily caused from the residues or breakdown products related to PCB’s, chlordane, and DDT. With respect to ULUG, these chemicals are found at increasing concentrations as ULUG increases (Zappia et al. 2005). Studies of other fish species have found that hepatic glutathione levels increase as these chemicals increase (Schmidt et al. 2005; Rocher et al. 2006). Unfortunately, polycyclic aromatic hydrocarbons (PAHs) were never measured at any of the sites. As PAHs are significant contaminants in urban areas it is recommended that these be tested in the future (Nowell et al. 2013). The addition of analyzing the presence of PAHs may help with determining stronger correlations between fish health and ULUG, as PAHs cause endocrine disruption, lesions, carcinogenic effects, and deformities. Although samples were not taken to examine reproductive health during this study, it should be noted that very few external lesions or deformities were observed on any of the fish examined in this study (Bryer et al. 2006; Rafferty et al. 2009). Liver apoptosis appeared to have a biphasic dose–response relative to the ULUG values studied. Apoptosis is an important homeostatic process that removes infected, transformed or damaged cells. At the cellular level it involves a series of morphological and nuclear changes that subsequently allow the cells to be ingested by phagocytic cells. This type of cell death protects tissues from the release of degradative enzymes that accompanies necrosis (Lockshin and Zakeri 2001). The apoptotic process includes an endonuclease-mediated fragmentation of the chromatin. Cellular endonucleases cleave genomic DNA between nucleosomes, producing fragments whose lengths vary by multiples of 180–200 base pairs (Arends et al. 1990; Staley et al. 1997). The resulting oligonucleosome-sized DNA fragments when visualized by gel electrophoresis, are described as a ‘nucleosomal ladder’, and have become a widely accepted biochemical indicator of apoptotic cell death (Staley et al. 1997). Credible evidence has shown that apoptosis displays a biphasic dose–response relationship, termed hormesis, in a wide range of animal models and cell types caused by a variety of stressors (Calabrese 2001; Osachoff et al. 2013).

123

D. Iwanowicz et al.

In general there is an adaptive response to low levels of stress or damage resulting in improved fitness for some physiological systems for finite periods (Jonas 2001; Calabrese and Baldwin 2001a, b). Our data suggest the biphasic dose–response relationship is occurring with regard to hepatic apoptosis and degree of urbanization as well, particularly in sites with ULUG values between 36 and 89. At these sites a J-shaped dose response (Fig. 3b) was noted, very similar to that produced by X-ray irradiation of mice thymocytes with inhibited low-dose response and stimulated high-dose response (Liu et al. 1996; Calabrese 2001). The key conceptual features of hormesis are disruption of homeostasis, modest overcompensation, reestablishment of homeostasis, and the adaptive nature of the process (Calabrese and Baldwin 2001b). As the ULUG is a composite index, this biphasic dose response may also reflect the concentration of one or more of the variables within the ULUG or variables that were not measured as part of the ULUG. At the present time, offering firm conclusions about such findings except to indicate that biphasic apoptotic responses reliably occur is premature, and more research is needed to understand these responses. We also found increasing densities of parasite load (spores per gram fish tissue) in fish throughout the lower and middle land-use gradients and a decreased density in fish from sites with the highest urban land-use values. At the sites with ULUG \ 36, water quality parameters such as eutrophication levels and the discharge by a thermal effluent, can raise rates of parasitism because the associated increased productivity can support a higher abundance of intermediate hosts and vectors (Lafferty and Kuris 1999), while these same stressors can decrease the host’s natural resistance to the parasites (Marcogliese and Pietrock 2011). At sites with ULUG C 36 a decrease in parasitism could occur if infected hosts suffer differentially higher mortality or the parasites were more susceptible to pollution than their hosts (Lafferty and Kuris 1999). Also habitat alterations such as urbanization can affect the intermediate host populations such that the abundance of their parasite densities is reduced (Lafferty and Kuris 1999). Evaluation of intermediate hosts was not included and would be a good follow up study. Histopathological changes in a variety of tissues have been used as biomarkers of contaminant exposure (Myers and Fournie 2002; Au 2004). The presence of macrophage aggregates was dramatically increased at site 9 (ULUG = 84). Macrophage aggregates are commonly used as a biomarker of exposure to toxicants (Blazer et al. 1987; Chang et al. 1998; van Dyk et al. 2012). In the present study, positive relationships were also seen between urbanization and prevalence of other lesions and abnormalities including liver necrosis and liver fibrosis. Interestingly, the occurrence of regenerative foci was highest in sites with the lowest ULUG value. Many factors could account for such data, including:

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL

levels of dietary protein, genetic variation within species, route of exposure, types of chemicals present, temperature, or dose (Bailey et al. 1989; Chang et al. 1998). There is also the possibility that the cause of increased regenerative foci in sites with the lowest ULUG values was not captured with the variables used to create the ULUG value. Urban land-use gradient information may be an important factor when determining impacts on freshwater ecosystems. McMahon and Cuffney (2000) determined that a rate of adverse effects on biological communities would be seen at index values [28 at their study site in New England. Furthermore, sites with an index value \28 were determined to be pre-effect and index values ranging from 28 to 66 were effect zones. Sites in the effect zone indicate the existence and shape of any relation between urbanization and water quality (McMahon and Cuffney 2000). Sites with high levels of urban intensity (ULUG [ 66) play less of a role in determining threshold responses and rates of impairment. Similarly, in this study we found positive correlations between ULUG index values and hepatic glutathione levels, in which hepatic glutathione levels were highest in ULUG sites C 36. There were also positive correlations between ULUG index values and macrophage aggregates at both spring samplings. We can also predict for stonerollers at ULUG sites C 36 that parasite spores per gram of fish tissue decrease significantly. The deleterious effects of urbanization on water quality are evident across the United States. Urbanization is reported to adversely affect the physical, chemical, and biological characteristics of the aquatic environment. Concern about the effects of urbanization has galvanized efforts to understand and manage urban development on the part of governmental organizations at all levels (McMahon and Cuffney 2000). Environmental science is continually confronted with issues related to assessing and evaluating the effects of stressors on the health of aquatic ecosystems. Some of the more challenging issues are validating ecologically significant factors, determining the importance of temporal and spatial variability (of physicochemical and biological factors in modifying responses to stress), and establishing cause and effect relationships between specific stressors as they relate to environmental damage (Adams 2001). Although at times there was a broad expanse of responses between ULUG values, the health of the stoneroller was consistently affected at ULUG values [36. It would be advantageous to expand the sites sampled between ULUG values of 14 and 36 to determine at what point fish health begins to be compromised for this species. Acknowledgments The authors are grateful to Beth Frankenberry, Brian Caskey, Sue Hartley, Kristin Justice, Wyman Turner and Sandy Page for help in collecting the largescale stonerollers for our research. We would also like to thank Dr. Mark Myers, Dr. Jack Fournie, Bane Schill, and Dr. Christine Densmore for critical review of the draft

619

manuscript. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

References Adams SM (2001) Biomarker/bioindicator response profiles of organisms can help differentiate between sources of anthropogenic stressors in aquatic ecosystems. Biomarkers 6:33–44 Arends MJ, Morris RG, Wyllie AH (1990) Apoptosis: the role of the endonuclease. Am J Pathol 136:593–608 Atkins JB, Zappia H, Robinson JL, McPherson AK, Moreland RS, Johnston BF, Harvill JS (2004) Water quality in the Mobile River Basin, Alabama, Georgia, Mississippi, and Tennessee, 1999–2001: U.S. Geological Survey. Circular 1231:40 Au DWT (2004) The application of histo-cytopathological biomarkers in marine pollution monitoring: a review. Mar Pollut Bull 48:817–834 Azevedo JS, Sarkis JES, Hortellani MA, Ladle RJ (2012) Are catfish (Arridae) effective bioindicators for Pb, Cd, Hg, Cu and Zn? Water Air Soil Pollut 223:3911–3922 Bailey GS, Goeger DE, Hendricks JD (1989) Factors influencing experimental carcinogenesis in laboratory fish models. In: Varanasi U (ed) Metabolism of polynuclear aromatic hydrocarbons in the aquatic environment. CRC Press, Boca Raton, pp 253–268 Bellucci CJ, Becker ME, Beauchene M, Dunbar L (2013) Classifying the health of Connecticut streams using benthic macroinvertebrates with implications for water management. Environ Manag 51:1274–1283 Birge WJ, Proice DJ, Shaw JR, Wigginton C, Hogstrand C (2000) Metal body burden and biological sensors as ecological indicators. Environ Tox Chem 19:1199–1212 Blazer VS, Wolke RE, Brown J, Powell CA (1987) Piscine macrophage aggregate parameters as health monitors: effect of age, sex, relative weight, season and site quality in largemouth bass (Micropterus salmoides). Aquat Toxicol 10:199–215 Bryer PJ, Elliot JN, Willingham EJ (2006) The effects of coat tarbased paemet sealer on amphibian development and metamorphosis. Ecotoxicology 15:241–247 Burger J, Campbell KR, Campbell TS, Shukla T, Dixon C, Gochfeld M (2005) Use of central stonerollers (Cyprinidae: Campostoma anomalum) from Tennessee as a bioindicator of metal contamination. Environ Monit Assess 110:171–184 Calabrese EJ (2001) Apoptosis: biphasic dose responses. Crit Rev Toxicol 31:607–613 Calabrese EJ, Baldwin LA (2001a) U-shaped dose-response in biology, toxicology, and public health. Annu Rev Publ Health 22:15–33 Calabrese EJ, Baldwin LA (2001b) Hormesis: a generalizable and unifying hypothesis. Crit Rev Toxicol 31:353–424 Chang S, Zdanowicz VS, Murchelano RA (1998) Associations between liver lesions in winter flounder (Pleuronectes americanus) and sediment chemical contaminants from north-east United States estuaries. ICES J Mar Sci 55:954–969 Cohen J, Cohen P (1983) Applied multiple regression/correlation analysis for the behavioral science, 2nd edn. Lawrence Erlbaum Associates, Publishers, Hillsdale Colombo JC, Cappelletti N, Migoya MC, Speranza E (2007) Bioaccumulation of anthropoenic contaminants by detritivorous fish in the Rı´o de la Plata estuary: 1-Aliphatic hydrocarbons. Chemosphere 68(11):2128–2135 Conners DE, Ringwood AH (2000) Effects of glutathione depletion on copper cytotoxicity in oysters (Crassostrea virginica). Aquat Toxicol 50:341–349

123

620 Cuffney TF, Zappia H, Giddings EMP, Coles JF (2005) Effects of urbanization on benthic macroinvertebrate assemblages in contrasting environmental settings: Boston, MA; Birmingham, AL; and Salt Lake City, UT. In: Brown LR, Gray RH, Hughes H, Meador MR (eds) Effects of urbanization on aquatic ecosystems. American Fisheries Society, Bethesda, pp 361–408 DeLeve LD, Kaplowitz N (1991) Glutathione metabolism and its role in hepatotoxicity. Pharmacol Ther 52:287–305 Elkin K, Lanier S, Rebecca M (2013) The interrelationships of hydrology and biology in a Tennessee stream, USA. Ecohydrology 6:355–362 Fischer JD, Cleeton SH, Lyons TP, Miller JR (2012) Urbanization and the predation paradox: the role of trophic dynamics in structuring vertebrate communities. Bioscience 62:809–818 Fitzpatrick FA, WaiteIR, D’Arconte PJ, Meador MR, Maupin MA, Gurtz ME (1998). Revised methods for characterizing stream habitat in the National Water Quality Assessment Program: U.S. Geological Survey Water Resources Investigations Report 98-4052, p 67 Giddings EA, Brown LR, Short TM, Meador MR (2006) Relation of fish communities to environmental conditions in urban streams of the Wasatch Front, Utah. West N Am Nat 66:155–168 Hester ET, Bauman KS (2013) Stream and retention pond thermal response to heated summer runoff from urban impervious surfaces. J Am Water Resour Assoc 49:328–342 Hill WR, Napolitano GE (1997) PCB Congener accumulation by periphyton, herbivores, and omnivores. Environ Contam Toxicol 32:449–455 Iannuzzi J, Butcher M, Iannuzzi T (2011) Evaluation of potential relationships between chemical contaminants in sediments and aquatic organisms from the lower Passaic River, New Jersey. Environ Toxicol Chem 30:1721–1728 Iwanowicz DD, Iwanowicz LR, Howerth EW, Schill WB, Blazer VS, Johnson RL (2013) Characterization of a new myxozoan species (Myxozoa: Myxobolidae: Myxosporea) in largescale stonerollers (Campostoma oligolepis) from the Mobile river basin (Alabama). J Parasitol 99:102–111 Johnson GC, Kidd RE, Journey CA, Zappia H, Atkins JB (2002) Environmental setting and water quality issues of the Mobile River basin, Alabama, Georgia, Mississippi, and Tennessee. Water Resource Inv Report 02-4162, Us Geological Survey, p 62 Jonas WB (2001) The future of hormesis: what is the clinical relevance of hormesis? Crit Rev Toxicol 31:655–658 Kim JH, Oh HM, Kim IS, Lim BJ, An KG (2013) Ecological health assessments of an urban lotic ecosystem using a multimetric model along with physical habitat and chemical water quality assessments. Int J Environ Res 7:759–768 Lafferty KD, Kuris AM (1999) How environmental stress affects the impacts of parasites. Limnol Oceanogr 44:925–931 Lai H-L, Gailucci VF, Gunderson DR, Donnelly RF (1996) Age determination in fisheries: methods and applications to stock assessment. In: Gallucci VF, Saila SB, Gustafson DJ, Rothschild BJ (eds) Stock assessment: quantitative methods and applications for small-scale fisheries. Lewis Publications, New York Lazorchak JM, McCormick FH, Henry TR, Herlithy AT (2003) Contamination in fish in streams of the mid-Atlantic region: An approach to regional indicator selection and wildlife assessment. Environ Toxicol Chem 22:545–553 Liu SZ, Zhang Y-C, Mu Y, Su X, Liu J-X (1996) Thymocyte apoptosis in response to low-dose radiation. Mutat Res 358:185–191 Lockshin RA, Zakeri Z (2001) Programmed cell death and apoptosis: origins of the theory. Nat Rev Mol Cell Bio 2:545–550 Luna LG (1992) Histopathological methods and color atlas of special stains and tissue artifacts. Am Histol, Gaithersburg

123

D. Iwanowicz et al. Marcogliese DJ, Pietrock M (2011) Combined effects of parasites and contaminants on animal health: parasites do matter. Trends Parasitol 27:123–130 McMahon G, Cuffney T (2000) Quantifying urban intensity in drainage basins for assessing stream ecological conditions. J Am Wat Res Assoc 36:1247–1261 McMahon G, Gregonis SM, Waltman SW, Omernik JM, Thorson TD, Freeouf JA, Rodrick AH, Keys JE (2001) Developing a spatial framework of common ecological regions for the conterminous United States. Environ Manag 28:293–316 McPherson AK, Moreland RS, Atkins JB (2003) Occurrence and distribution of nutrients, suspended sediment, and pesticides in the Mobile River Basin, Alabama, Georgia, Mississippi, and Tennessee, 1999–2001. U.S. Geological Survey Water-Resources Investigations Report 03-4203, p 109 Meador MR, Coles JF, Zappia H (2005) Comparison of fish assemblage responses to gradients of urban intensity in contrasting metropolitan areas: Birmingham, Alabama, and Boston, Massachusetts. In: Brown LR, Gray RH, Hughes H, Meador MR (eds) Effects of urbanization on aquatic ecosystems. American Fisheries Society, Bethesda Mettee MF, O’Neil PE, Pierson JM (1996) Fishes of Alabama and the mobile basin. Oxmoor House Inc, Birmingham Myers MS, Fournie JW (2002) Histopathological biomarkers as integrators of anthropogenic and environmental stressors. In: Adams SM (ed) Biological Indicators of Aquatic Ecosystem Stress. American Fisheries Society, Bethesda Nowell LH, Moran PW, Gilliom RJ, Calhoun DL, Ingersoll CG, Kemble NE, Kuivila KM, Phillips PJ (2013) Contaminants in stream sediments from seven United States Metropolitan Areas: Part I: distribution in relation to urbanization. Arch Environ Contam Toxicol 64:32–51 Osachoff HL, van Aggelan GC, Mommsen TP, Kennedy CJ (2013) Concentration-response relationships and temporal patterns in hepatic gene expression of Chinook salmon (Oncorhynchus tshawytscha) exposed to sewage. Comp Biochem Physiol D 8:32–44 Otto DME, Moon TW (1995) 3,3’,4,4’-Tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacol Toxicol 77:281–287 Paul MJ, Meyer JL (2001) Streams in the urban landscape. Ann Rev Ecol Syst 32:333–365 Rafferty SD, Blazer VS, Pinkney AE, Grazio JL, Obert EC, Boughton L (2009) A historical perspective on the ‘‘fish tumors or other deformities’’ beneficial use impairment at Great Lakes Areas of Concern. Great Lakes Res 35:496–506 Ramirez A, Engman A, Rosas KG, Perez-Reyes O, Martino-Cardona DM (2012) Urban impacts on tropical island streams: some key aspects influencing ecosystem response. Urban Ecosyst 15:315–325 Rocher B, Le Goff J, Peluhet L, Briand M, Manduzio H, Gallois J, Devier MH, Geffard O, Gricourt L, Augagneur S, Budzinski H, Pottier D, Andre´ V, Lebailly P, Cachot J (2006) Genotoxicant accumulation and cellular defense activation in bivalves chronically exposed to waterborne contaminants from the Seine River. Aquat Toxicol 79:65–77 Sayeed I, Parvez S, Pandey S, Bin-Hafeez B, Haque R, Raisuddin S (2003) Oxidative stress biomarkers of exposure to deltamethrin in freshwater fish, Channa punctatus Bloch. Ecotox Environ Safe 56:295–301 Schmidt K, Staaks GBO, Pflugmacher S, Steinberg CEW (2005) Impact of PCB mixture (Aroclor 1254) and TBT and a mixture of both on swimming behavior, body growth and enzymatic biotransformation activities (GST) of young carp (Cyprinus carpio). Aquat Toxicol 71:49–59

Effects of urban land-use on largescale stonerollers in the Mobile River Basin, Birmingham, AL Srikanth K, Pereira E, Duarte AC, Ahmad I (2013) Glutathione and its dependent enzymes’ modulatory responses to toxic and metalloids in fish-a review. Environ Sci Pollut Res 20:2133–2149 Staley K, Blaschke AJ, Chun J (1997) Apoptotic DNA fragmentation is detected by a semi-quantitative ligation-mediated PCR of blunt DNA ends. Cell Death Differ 4:66–75 Tate CM, Cuffney TF, McMahon G, Giddings EMP, Coles JC, Zappia H (2005) Use of an urban intensity index to assess urban effects on streams in three contrasting environmental settings. In: Hughes H, Meador MR (eds) LR, Gray RH. Effects of urbanization on aquatic ecosystems. American Fisheries Society, Bethesda, pp 291–316 van Dyk JC, Cochrane MJ, Wagenaar GM (2012) Liver histopathology of the sharptooth catfish Clarias gariepinus as a biomarker of aquatic pollution. Chemosphere 87:301–331 Weaver LA, Garman GC (1994) Urbanization of a watershed and historical changes in a stream fish assemblage. Trans Am Fish Soc 123:162–172

621

Winer BJ, Brown DR, Michels KM (1991) Statistical principles in experimental design, 3rd edn. McGraw-Hill, New York Zandbergen PA (1998) Urban watershed ecological risk assessment using GIS: a case study of the Brunette River watershed in British Columbia, Canada. J Hazard Mater 61:163–173 Zappia H (2002) Organochlorine compounds and trace elements in fish tissue and streambed sediment in the Mobile River Basin in Alabama, Mississippi, and Georgia, 1998. U.S. Geological Survey Water-Resources Investigations Report 02-4160, p 34 Zappia H, Cuffney T, McMahon G, Walsh S, Caskey B, Bryant W, Atkins JB (2005) Responses of fish, invertebrate, and algal communities to a gradient of urbanization in the Ridge and Valley Ecoregion of Alabama and Georgia. In: Brown LR, Gray RH, Hughes H, Meador MR (eds) Effects of urbanization on aquatic ecosystems. American Fisheries Society, Bethesda

123