Effects of Increasing Nitrogen and Phosphorus Concentrations on

Subscriber access provided by NEW YORK UNIV

Article

Effects of increasing nitrogen and phosphorus concentrations on phytoplankton community growth and toxicity during Planktothrix blooms in Sandusky Bay, Lake Erie Timothy W. Davis, George S. Bullerjahn, Taylor Tuttle, Robert Michael McKay, and Susan B Watson Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b00799 • Publication Date (Web): 20 May 2015 Downloaded from http://pubs.acs.org on May 27, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39

Environmental Science & Technology

1

Effects of increasing nitrogen and phosphorus concentrations on phytoplankton

2

community growth and toxicity during Planktothrix blooms in Sandusky Bay, Lake Erie

3

Timothy W. Davis1†*, George S. Bullerjahn2, Taylor Tuttle2, Robert Michael McKay2,

4

Susan B. Watson1

5

1: Canadian Centre for Inland Waters, Environment Canada, Burlington, ON, L7S 1A1,

6

Canada

7

2: Department of Biological Sciences, Bowling Green State University, Bowling Green, OH, 43403, USA

8 9

†: Current Address: NOAA Great Lakes Environmental Research Laboratory, Ann

10

Arbor, MI 48108, USA

11

*Corresponding author

12

4840 South State Road

13

Ann Arbor, MI 48108-9719

14

Phone: (734) 741-2286

15

Fax: (734) 741-2055

16

Email: [email protected]

17

Running title: N promotes Planktothrix growth and bloom toxicity

1

ACS Paragon Plus Environment


18

Abstract:

19

Sandusky Bay experiences annual toxic cyanobacterial blooms dominated by

20

Planktothrix aghardhii. To further understand the environmental drivers of these events,

21

we evaluated changes in the growth response and toxicity of the Planktothrix-dominated

22

blooms to nutrient amendments with phosphate (PO4) and inorganic and organic forms of

23

dissolved nitrogen (N; ammonium (NH4), nitrate (NO3) and urea) over the bloom season

24

(June – October). We complemented these with a metagenomic analysis of the planktonic

25

microbial community. Our results showed that bloom growth and microcystin (MC)

26

production responded more frequently to additions of dissolved N than PO4, and that the

27

dual addition of NH4 + PO4 and Urea + PO4 yielded the highest MC concentrations in

28

54% of experiments. Metagenomic analysis confirmed that P. aghardhii was the primary

29

MC producer. The phylogenetic distribution of nifH revealed that both heterocystous

30

cyanobacteria and heterotrophic proteobacteria had the genetic potential for N2 fixation in

31

Sandusky Bay. These results suggest that as best management practices are developed for

32

P reductions in Sandusky Bay, managers must be aware of the negative implications of

33

not managing N loading into this system as N may significantly impact cytanobacterial

34

bloom size and toxicity in Sandusky Bay.

2


Page 2 of 39

Page 3 of 39

35


Table of Contents Graphic:

36 37

Key words: Lake Erie, microcystins, nitrogen limitation, Planktothrix, harmful algal

38

bloom

3



39

Introduction:

40

Cyanobacterial harmful algal blooms (CHABs) threaten the Laurentian (North

41

American) Great Lakes, a major freshwater resource containing roughly 18% of Earth’s

42

available surface freshwater (1). Over the past several decades these ecosystems have

43

been subject to many anthropogenic pressures such as nutrient over-enrichment

44

(eutrophication), contaminant inputs, structural engineering (impoundments, shoreline

45

development and dredging), wetland loss, invasive species (e.g., dreissenid mussels and

46

round gobies). Lake Erie is the most socio-economically important of the Great Lakes

47

(1,2,3), serving the recreational, commercial, and drinking water needs of over 11.5

48

million people (3). Lake Erie is also is the smallest and shallowest Great Lake, making it

49

the most susceptible to increased nutrient loading. Lake Erie has been impacted by poor

50

water quality, including CHABs, as far back as the 1960s (4,5,6). However, with the

51

introduction of phosphorus reduction programs in 1972 as part of the Great Lakes Water

52

Quality Agreement (7), the water quality (e.g. reduction in cyanobacterial biomass) in

53

Lake Erie improved significantly throughout the late 1970s and 1980s (6,8). However,

54

since the mid-1990s, Lake Erie and its watershed have been increasingly affected by

55

toxic CHABs, largely fuelled by agricultural nonpoint nutrient sources (9,10,11). To date

56

there has been no consensus on the key factors driving HAB occurrence, extent and

57

timing, even in well-studied systems such as Lake Erie (12).

58

Traditionally, it has been thought that phosphorus (P) was the primary limiting

59

nutrient in freshwater systems (13,14), including Lake Erie (9,15). However, recently,

60

the importance of nitrogen (N) in the control of eutrophication and CHAB events has

61

been gaining attention (16,17,18).

Previous studies on Lake Erie blooms have

4


Page 4 of 39

Page 5 of 39


62

investigated bottom-up controls such as nutrient availability and light (19,20,21),

63

physical factors like wind strength (22) and top-down controls including pelagic (23) and

64

benthic grazing (25). Furthermore, differences and dynamics among the genetic strains

65

of cyanobacteria within Lake Erie blooms have been investigated through field and

66

laboratory experiments. (11,26,27).

67

Planktothrix, a globally distributed genus, (28), is less common than Microcystis

68

in the offshore Lake Erie blooms but can predominate in bays and tributaries, notably

69

Sandusky Bay (29) and the Maumee River (30). Anabaena (currently called

70

Dolichospermum; 31) spp. blooms also occur in the western and central basins of Lake

71

Erie (21,23,32) but to date, the ecology and toxicity of these events have been largely

72

overlooked.

73

Only a few studies have investigated the phytoplankton population in Sandusky

74

Bay, (29), bloom phylogenetics (33) and the growth response of the Planktothrix

75

community to N or P additions (34).

76

Planktothrix spp. as the primary MC-producers in Sandusky Bay.

77

Bridgeman (34) have investigated the short-term growth response of the total

78

phytoplankton community to N and P amendments, but did not examine toxins and their

79

study was limited to a single experiment (July) in Sandusky Bay. Furthermore, even

80

though it has been demonstrated that DIN in Sandusky Bay decreases rapidly during the

81

summer months, the non-diazotrophic cyanobacterium, Planktothrix, dominates

82

throughout the growing season. However little is known about the Sandusky Bay N-

83

fixing community, except that the diazotrophic cyanobacterium, Cylindrospermopsis

84

raciborskii, can be present at significant densities (23,29). A better understanding the

Rinta-Kanto and Wilhelm (33) identified

5


Chaffin and


85

composition of the N-fixing community may provide significant insight into the

86

ecological reasons why a non-diazotrophic cyanobacterium can dominate a chronically

87

N-limited system. Therefore, our study aimed to evaluate changes in the growth response

88

of the Planktothrix-dominated blooms to increases in inorganic phosphorus and inorganic

89

and organic forms of nitrogen over the growing season (June – October) while

90

complementing these results with measurements of microcystin production and

91

metagenomic data on the planktonic microbial community.

92

Methods:

93

Study Sites:

94

Sandusky Bay is a shallow (mean depth: 2.6 m; 29), well-mixed drowned river

95

mouth at the southeast corner of the western basin of Lake Erie (Fig. 1). Sandusky Bay

96

has an hourglass shape and is divided into the outer bay (eastern half) and the inner bay

97

(western half; Fig. 1).

98

amendment experiments were conducted on samples from the outer bay at two

99

Environment Canada sites (EC 1198; EC 1163, Fig. 1). Samples were collected from a

100

small boat (EC1198) or from the CCGS Limnos (both sites); EC 1163 was accessible

101

only from the CCGS Limnos, and due to ship scheduling no sampling or experiments

102

were conducted between early June and mid-July (Supplemental Table 1 & 2,

103

respectively).

104

Sample collection:

During May through October 2013, short-term nutrient

105

Physicochemical data were measured at each site at approximately the same time

106

each sampling trip using a calibrated water quality probe (YSI 6600, Yellow Springs,

107

OH, USA), measuring surface water temperature, dissolved oxygen concentration, pH,

6


Page 6 of 39

Page 7 of 39


108

and conductivity. Water samples (60L) were collected using a rosette from a depth of 1 m

109

and either processed immediately when onboard CCGS Limnos or kept at room

110

temperature until returned to Bowling Green State University (BGSU) for processing

111

within two hours. From each site, subsamples were processed for total chlorophyll a (chl

112

a; 47 mm GF/F), particulate microcystins (MCs; 1.2 µm polycarbonate membrane) and

113

cell-bound DNA analysis (0.22µm Sterivex filter cartridge; Millipore Corp., Billerica,

114

MA, USA). The filter cartridges were immediately frozen with chl a and MC samples

115

stored at -20 °C and the Sterivex filter cartridges stored at -80 °C until analysis. Samples

116

for phytoplankton community composition and biomass determination were preserved

117

with Lugol’s iodine solution (1% v/v. final conc.). Rainfall data for the area surrounding

118

Sandusky

119

(http://www.usclimatedata.com/).

120

through October 2013.

Bay

were

collected

from

the

U.S.

Climate

Data

website

Weekly average rainfall was calculated from May

121

Duplicate water samples for the analyses of dissolved nutrients (nitrate/nitrite

122

[NO3- + NO2-], ammonia [NH3] and soluble reactive phosphorus [SRP]) were collected by

123

filtering lake water through a 0.45µm polycarbonate filter into triple-rinsed 20 mL plastic

124

bottles and stored at -20 °C until analysis. Water samples for total phosphorus [TP]

125

analysis were collected by filling a triple-rinsed 20 mL plastic vial with whole lake water

126

followed by storage at -20 °C. All samples were analyzed at the National Center for

127

Water Quality Research, Heidelberg University, Tiffin, OH, USA using standard

128

techniques (35).

129

Nutrient amendment experiments: Short-term nutrient enrichment experiments were conducted to assess the impact

130

7



Page 8 of 39

131

of increased organic and inorganic N or PO4 concentrations on the Planktothrix

132

populations in Sandusky Bay. During the summer of 2013, a total of 13 experiments

133

were conducted on samples from sites EC 1163 and EC 1198.

134

experiments, sets of triplicate, 1L clear polycarbonate bottles (n = 24) were filled with

135

surface water from each experimental site and were either left unamended to serve as a

136

control, or amended with various forms of N: nitrate (combination of Ca(NO3)2 plus

137

KNO3), ammonium (NH4Cl), urea, orthophosphate (KH2PO4) or a combination of each

138

individual N form and KH2PO4 (referred to as PO4 from hereon). All N treatments were

139

amended to a final concentration of 178.9µM and all P treatments were spiked to a final

140

concentration of 1µM. In order to facilitate comparisons between these field experiments

141

and planned laboratory experiments, the N concentration of 178.9µM was chosen because

142

it represented 10% of the total N concentration in the growth media used in our

143

laboratory. Experiments were conducted in two locations (See Supplemental Table 2).

144

Those carried out onboard the CCGS Limnos were incubated in a flow-through deck

145

incubator under ambient water and light conditions. Neutral density screening was used

146

to achieve light levels similar to in situ conditions. The experiments conducted at BGSU

147

were incubated in a stand-alone incubator using temperatures, light levels (50 µmol m-2 s-

148

1

149

photoperiods (14:10 h: June, July and August; 12:12 h: September and October).

150

Incubator temperatures during all experiments were measured every five minutes with in

151

situ loggers (Onset Computer Corporation, Bourne, MA, USA) and indicated that

152

incubation temperatures and light levels remained within the same range (i.e. light: 50 -

153

100 µmol m-2 s-1; temperature ambient ± 1 o C) as those found at each site. All bottles

To commence

) similar to measured field conditions on each sampling date and seasonally appropriate

8


Page 9 of 39


154

were gently inverted every 4 to 6 h. After 48 h, each bottle was sampled as described

155

above to quantify chl a, phytoplankton community composition and biomass and

156

particulate concentrations of MCs. The effects of P and inorganic and organic N on the chl a concentration, MCs

157 158

concentration were analyzed with one-way ANOVAs.

Post-hoc comparisons of

159

significant impacts were elucidated with Tukey’s multiple comparison tests. For all

160

results, the standard variance presented is ± one standard error (SE). Since only one

161

replicate per treatment was enumerated for phytoplankton community composition and

162

biomass, statistical tests could not be preformed.

163

Extraction and analysis of microcystins:

164

Particulate MCs were extracted from samples using a combination of physical and

165

chemical lysis techniques. All samples were resuspended in 1mL molecular grade water

166

(pH = 7; Sigma Aldrich, St. Louis, MO, USA) and subjected to three freeze/thaw cycles

167

before the addition of the QuikLyse™ reagents (Abraxis LLC; Warminster, PA, USA) as

168

per the manufacturer’s instructions. The samples were then centrifuged for five minutes

169

at 2 ×103 × g to pellet cellular debris. The concentrations of microcystins (reported as

170

microcystin-LR equivalents) were measured using an enhanced sensitivity microcystin

171

enzyme-linked immunosorbent assay (Abraxis LLC) following the methodologies of

172

Fischer et al. (36). This assay is congener-independent as it detects the ADDA moiety,

173

which is found in almost all MCs. These analyses yielded a detection limit of 0.04 µg L-

174

1

175

Phytoplankton community composition and biomass:

.

9



After evaluating the chlorophyll a data, one replicate per treatment in each of the

176 177

seven representative seasonal experiments (Figures 2 & 3) was chosen for enumeration to

178

support the pigment-based results. Further, since the initial chlorophyll a concentrations

179

and control bottles after 48 hours were statistically similar in all but one experiment we

180

only enumerated the control from each experiment and not the initial. Aliquots of the

181

preserved samples from were allowed to settle overnight in sedimentation chambers

182

following the procedure of Lund et al. (37). Algal units were counted from randomly

183

selected transects on a Zeiss Axiovert 40 CFL inverted microscope. Counting units were

184

individual cells, filaments or colonies depending on the organization of the algae. A

185

minimum of 400 units were counted for each sample. The bulk of the counts are made at

186

600X magnification. The large and rare organisms (such as Ceratium) were scanned for

187

at 350X magnification. Fresh weight biomass was calculated from recorded abundance

188

and specific biovolume estimates, based on geometric solids (38), assuming unit specific

189

gravity (1.0 g = 1.0 cm3). The biovolume (mm3 m-3 fresh weight) of each species was

190

estimated from the average dimensions of 10 to 15 individuals. The biovolumes of

191

colonial taxa were based on the number of individuals in a colony. All calculations for

192

cell concentration and biomass were performed with Hamilton’s (39) computer program.

193

Sandusky Bay metagenome:

194

Microbial biomass from 200 mL surface water collected on 17 June 2013 at EC

195

1198 was concentrated on Sterivex cartridge filters (0.22 µm) and immediately frozen in

196

liquid nitrogen. DNA was extracted from the Sterivex cartridges using the PowerWater

197

Sterivex DNA Isolation Kit (MO BIO Laboratories, Inc, Carlsbad, CA, USA) following

198

manufacturer’s instructions. The metagenome was obtained by Illumina sequencing on a

10


Page 10 of 39

Page 11 of 39


199

HiSeq platform at the DOE-Joint Genome Institute (Walnut Creek, CA, USA). The

200

annotated metagenome dataset comprised 7.3 x 109 total bases and 3.8 x 107 total

201

sequences.

202

interest (ntcA, nifHDK, mcyA) were identified via BLAST (40), followed by alignment by

203

CLUSTAL W and phylogenetic analysis in MEGA (41). Trees were constructed by

204

neighbor-joining with 1000 bootstrap replicates.

205

Accession numbers:

206

Sequences were deposited into the JGI IMG database (accession numbers 10000275,

207

10348616, 10111557, and 100007422).

208

Results:

209

Water quality and rainfall data

Cyanobacterial and bacterial metagenome scaffolds containing genes of

210

The Sandusky Bay bloom persisted from May through October, 2013. At both

211

sites, surface water temperatures ranged from 16.6 – 27.0 °C, dissolved oxygen

212

concentrations were generally supersaturating (mean: 10.7 ± 1.1 mg L-1) and the pH and

213

conductivity were consistent throughout the sampling period (mean: 8.4 ± 0.1 and 401 ±

214

20 µS cm-1, respectively).

215

ranged from 18.4 ± 0.2 µg L-1 to 66.4 ± 7.7 µg L-1 (Supplemental Table 1). Cyanobacteria

216

dominated the phytoplankton community biomass (95-99%) at all times except during

217

mid-July when cyanobacteria only comprised 11% and 32% of the total phytoplankton

218

biomass at stations 1163 and 1198, respectively (Figures 2 & 3). Planktothrix agardii

219

dominated the cyanobacterial community comprising > 73% of the biomass on all but

220

one date (July 16) at site 1163 (Figures 2 & 3). In spring and early summer,

221

concentrations of MCs frequently exceeded the State of Ohio’s Public Health Advisory

Chlorophyll a concentrations of all samples at both stations

11



Page 12 of 39

222

guideline level of 6 µg L-1 with values falling below this limit in mid-summer

223

(Supplemental Table 1). May and June samples ranged from 5.2 ± 1 µg L-1 to 11.7 ± 2.6

224

µg L-1, peaking on June 19 at EC 1198 (Supplemental Table 1). By contrast, July –

225

October concentrations of MCs ranged from a mid-summer low of 0.4 ± 0.1 µg L-1 on

226

July 16 at EC 1198 and increased to 5.3 ± 0.5 µg L-1 on October 9 (Supplemental Table

227

1).

228

Profiles of nutrient concentrations (SRP and DIN) measured at the time of each

229

experiment indicated that whereas SRP remained well above DL and showed only

230

moderate changes with time (EC 1163: 0.08 ± 0.03 – 0.41 ± 0.06 µM; EC 1198: 0.06 ±

231

0.02 – 1.21 ± 0.03, Figure 4B), DIN was highly variable (Figure 4A). DIN declined at

232

both stations in late August – October, and at EC 1198 in late May and early June.

233

However, DIN concentrations peaked in late June and early July at values almost 100

234

fold higher, following a series of mid-summer rain events and increased runoff (Figure 4;

235

www.usclimatedata.com/climate/sandusky/ohio/united-states/usoh0855/2013/8).

236

Nutrient amendment experiments In total, seven experiments were conducted at EC 1163 and six at EC 1198.

237

A

238

representative experimental date was chosen from early spring, early summer mid-

239

summer and early fall for a visual comparison of seasonal responses both within and

240

between stations. 1163 had no early summer experiment (Figures 2 & 3). Both stations

241

showed similar overall responses in algal growth, (as inferred by increasing chl a

242

concentrations) and concentrations of MCs as both were primarily stimulated by N in

243

experiments conducted during the summer of 2013 (Figures 5 & 6; Supplemental Table

12


Page 13 of 39


244

2). However, each station yielded different responses to nutrient additions at different

245

points throughout the summer.

246

During early summer (late May - early June), concentrations of chl a and MCs did

247

not exhibit significantly positive responses from the control in any treatment (p > 0.05;

248

Figures 5 & 6; Supplemental Table 2). Phytoplankton biomass response was consistent

249

with the overall chl a response and no shift in community composition was observed

250

(Figures 2 & 3). No experiments were conducted at EC 1163 during June, but all

251

experiments conducted throughout this month at 1198 indicated that by mid-June, algal

252

growth was strongly N-limited, with additions of any form of N yielding chl a

253

concentrations that were nearly double the control treatments (Figure 6; Supplemental

254

Table 2). MC concentrations responded to the addition of any form of N in two of the

255

three experiments, suggesting that bloom toxigenicity was also strongly N-limited (p

Effects of Increasing Nitrogen and Phosphorus Concentrations on

Recommend Documents