Cyanobacterial Blooms: Toxins, Tastes, and Odors

USGS Kansas Water Science Center Algal Toxin Team

Jennifer L. Graham, Keith A. Loftin,

Michael T. Meyer, and Andrew C. Ziegler

USDA-CSREES National Water Conference

February 4, 2008

Overview

• Cyanobacterial (Blue-Green Algal) Toxins and Taste-and-Odor Compounds

• Microcystin in the Midwest

• Research Needs

• USGS Studies

Cyanobacterial toxins are “…anthropogenically amplified, but basically a natural phenomenon…”

I. Chorus, 1993

Binder Lake, IA Aug 2006

Thomas Lake, NE May 2006

• Ecologic Concerns– Zooplankton avoidance or death– Accumulation by mussels– Fish kills– Losses to bird and mammal populations

• Economic Concerns– Added drinking water treatment costs– Loss of recreational revenue– Death of livestock and domestic animals– Medical expenses

• Health Concerns– Tastes-and-Odors

• Olfactory sensitivity at low concentrations (< 0.01 µg/L) • Chronic effects?

– Toxins• Human and animal illness and death• EPA contaminant candidate list• Drinking water - microcystin

– WHO guideline – 1.0 µg/L– Drinking-water treatment processes effectively remove most toxins

• Recreational water – WHO guidelines for microcystin– Low Risk - < 10 µg/L– Moderate Risk - 10-20 µg/L– High Risk - > 20 µg/L

• Known chronic effects

Toxins and Taste-and-Odor Compounds Produced by Cyanobacteria

Dermatoxins Hepatotoxins Neurotoxins Taste/OdorCYL MC ANA BMAA GEOS MIB

Colonial/Filamentous

Aphanizomenon X X X X X X

Anabaena X X X X X X ?

Cylindrospermopsis X X X

Microcystis X X X

Oscillatoria/Planktothrix X X X X X X

Unicellular

Synechococcus X X X X X

Synechocystis X X X

Cyanobacterial Toxins and Taste-and-Odor Compounds Are Not Produced By The Same Biochemical Pathway But Patterns in Distribution Are Similar

• Extreme spatiotemporal variability

• Lack of relation with cyanobacterial community composition or chlorophyll concentration

• Coupling with lake/river processes as influenced by physiochemical, biological, hydrological, and meteorological factors

Upper Gar, IA Aug 2006

Upper Pine, IA August 2006

Algal Toxins

SAX

ATX-A(s

)

MC-L

R, LA, -

YR

ATX-A

MC-R

R

CYL (2

4h)

CYL (5

d)

LD

50 ( g

/kg

)

0100200300400500600

20002100

Cyanotoxins Exhibit a Wide Range of Toxicities and Toxic Effects and Are

Currently Listed on the U.S. EPA Contaminant Candidate List

• Acute Toxicity– Neurotoxic– Hepatotoxic– Dermatoxic

• Chronic Toxicity– Carcinogen– Tumor Promotion– Mutagen– Teratogen– Embryolethality

To

xic

ity

After Chorus and Bartram, 1999

Cyanobacteria Made the News in at Least 21 U.S. States During 2006

= news report

After Graham, 2006 USGS FS-2006-3147

At Least 35 U.S. States have Anecdotal Reports of Human or Animal Poisonings Associated with Cyanotoxins

= reported incident

During 1999-2006 Microcystin was Detected in INTEGRATED PHOTIC ZONE Samples from 78% of Lakes (n=359) and TOTAL Concentrations Ranged from <0.1 to 52 µg/L

After Graham and others 2004 and 2006

Trophic Gradient

Mean and Maximum TOTAL Microcystin Concentrations Significantly Increased Along the Natural Trophic Gradient in the Study Region

OH OP DT WL

Mic

rocy

stin

( g

/L)

0

1

2

3

4

5

30405060

a

a, b

b

cn=2546p<0.01

After Graham and others 2004 and 2006

maxima

a, b, and c indicate significant differencesin mean concentration

80% of All Lakes Sampled During 1999-2006 Had Maximum TOTAL Microcystin Concentrations ≤ 1 µg/L in Open Water Samples

After Graham and others 2004 and 2006

Microcystin Concentration (g/L)

ND 0-1 1-5 5-10 10-20 20-60

% O

cc

urr

en

ce

0

20

40

60

80

100Total n=355

61% of Lakes Sampled During 3-6 Years Always Had Detectable Microcystin During Summer, and Microcystin Maxima Were Greatest in These Lakes

Ma

xm

imu

m M

icro

cys

tin

C

on

cen

tra

tio

n ( g

/L)

Always Detected

OccassionallyDetected

n=101

n=65

100

1

0.1

0.01

0.001

10

After Graham and others 2004 and 2006

Seasonal Patterns in Microcystin Concentration are Unique to Individual Lakes and Peaks May Occur Anytime Throughout the Year

Marceline 1, MO

Mic

rocy

stin

( g

/L)

510152025

Bilby Ranch, MO

1

2

Harrison, MO

1

2

Mozingo, MO

1

2

Forest, MO

1

2

Sterling Price, MO

2004J F M A M J J A S O N D

1

2

Forest Lake, MO

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Mic

rocy

stin

( g

/L)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Peak Microcystin Values May Occur in the Winter

Oscillatoria

2004

Seasonal Patterns Were Relatively Consistent Between Years in Some Lakes

Mozingo Lake, MO

Jan

Feb

Mar

Ap

r

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Jan

Mic

rocy

stin

( g

/L)

0.0

0.5

1.0

1.5

2.0

2.5

3.020042001

After Graham and Jones, 2006

Regionally, Microcystin Was Significantly Correlated With Factors That Affect Cyanobacterial Growth

Variable rs p-value n

Latitude 0.66 <0.01 800

Total Nitrogen (TN) 0.58 <0.01 795

Total Phosphorus (TP) 0.46 <0.01 795

Secchi -0.27 <0.01 796

pH 0.17 <0.01 507

Alkalinity 0.15 <0.01 432

TN:TP -0.15 <0.01 791

After Graham and others 2004

Regional Associations Between Microcystin and Environmental Variables Were Complex

TN:TP

010

020

030

040

050

0

TP (g/ L)

020

040

060

080

010

00

TN (m/ L)

020

0040

0060

0080

00

1000

0

1200

0

1400

0

1600

0

1800

0

Mic

rocy

stin

( g

/L)

pH

2 4 6 8 10 12

Secchi (m)

0 2 4 6 8 10

Alkalinity (mg/ L)

0 50 100

150

200

250

300

350

r2=0.84

r2=0.70

r2=0.38 r2=0.99

r2=0.72 r2=0.94

5

0

1

2

3

4

5

0

1

2

3

4

5

0

1

2

3

4

5

0

1

2

3

4

5

0

1

2

3

4

5

0

1

2

3

4

After Graham and others, 2004

Microcystin Was Not Strongly Correlated With Measures of the Cyanobacterial Community

After Graham and others, 2004

Biovolume of Potential Microcystin Producers (m3/ L)

1e+5

1e+6

1e+7

1e+8

1e+9

1e+10

1e+11

1e+12

1e+13

Mic

rocy

stin

( g

/L)

0

1

2

3

4

5

6 r=0.31p<0.01

r2=0.96 r2=0.90 r2=0.83

Mozingo Lake, MO - Summer 2001

Log10 Dissolved Nitrogen2.70

2.75

2.80

2.85

2.90

2.95

3.00

Log

10 M

icro

cyst

in

1.0

1.5

2.0

2.5

3.0

3.5

Log10 Total Cations1.60

1.65

1.70

1.75

Log

10 M

icro

cyst

in

1.0

1.5

2.0

2.5

3.0

3.5

Log10 Chlorophyll>35 m

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Log

10 M

icro

cyst

in

1.0

1.5

2.0

2.5

3.0

3.5

Individual Lake Correlations Between Microcystin and Environmental Variables Were Linear

After Graham and others, 2006

Seasonal Patterns in Individual Lakes are Coupled with Seasonal Lake Processes, Including Stratification and Nutrient Loss from the Epilimnion

Epilimnion of Mozingo Lake, MO - Summer 2001

May June July Aug Sept

Dis

solv

ed N

itro

gen (

g/L)

500

550

600

650

700

750

800

850

900

Mic

rocy

stin

(ng/L)

0

100

200

300

400

500

600

700

800

900

1000

Net

Chlo

rophyll (

g/L)

0

5

10

15

20

25

30

35

Tota

l Cati

ons

(mg/L)

38

40

42

44

46

48

50

52

54

56

58

NetChlorophyll

Microcystin

Nitrogen

Cations

After Graham and others, 2006

Factors Most Strongly Correlated With Microcystin Vary Among Lakes and Years

Mozingo Lake, MO

Chlorophyll > 35 m (g/ L)

0 10 20 30 40 50

Mic

rocy

stin

( g

/L)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8 r2=0.30p<0.01

Forest Lake, MO

Chlorophyll > 35 m (g/ L)

0 2 4 6 8 10 12

Mic

rocy

stin

( g

/L)

0.0

0.1

0.2

0.3

0.4

0.5

0.6 r2=0.90p<0.01

Marceline 1, MO

Chlorophyll > 35 m (g/ L)

0 10 20 30 40 50 60

Mic

rocy

stin

( g

/L)

0

5

10

15

20

25 r2=0.17p<0.01

Microcystin in Midwestern Lakes - Conclusions

• Microcystin is common in the Midwest and may reach levels that can cause health concerns

• Seasonal patterns in microcystin are unique to individual lakes and maxima may occur in any season

• Regional relations between microcystin and environmental variables are complex

• Microcystin and environmental variables may be tightly coupled in individual lakes, but relations vary among lakes and years

Binder Lake, IA Aug 2006

Elysian Lake, MN Aug 2006

Research Needs

• Certified Standards

• Consistent Sampling Protocols

• Robust and Quantitative Analytical Methods for a Variety of Toxins

• Distribution of Microcystin Variants and Other Cyanobacterial Toxins

• Long Term Studies to Identify the Key Environmental Factors Leading to Toxic/Taste-and-Odor Producing Blooms

• Methods for Early Detection

• Predictive Models

Cheney Reservoir, KS June 2003Photo Courtesy of KDHE

Thomas Lake, NE May 2006Photo from Omaha NBC News

Consistent Sampling Protocols – Sample Location is Important

After Chorus and Bartram, 1999

Concentrations of Toxins and Taste-and-Odor Compounds May Vary by Orders of Magnitude at Different Sample Locations Within a Lake

Cheney Reservoir, KS September 2006

Microcystin: 13 µg/LGeosmin: 0.25 µg/L

Microcystin: 4 µg/LGeosmin: Not Detected

Actinomycetes Bacteria Also Produce Geosmin and MIB and May Contribute to Taste-and-Odor Problems in Drinking Water Supplies

Consistent Sampling Protocols – Collection Technique is Important

ToxinToxin

IntracellularToxin

DissolvedToxin

SorbedToxin

IntracellularToxin

DissolvedToxin

SorbedToxin

Total Toxin Dissolved Phase Toxin

= +

Particulate Toxin

Plankton Net Sampling Whole Water Sampling Filter/Filtrate Sampling

Microcystin Concentrations Decreased with Decreases in Cyanobacterial Size Class

Size Class (m)

> 100 53-100 35-53 10-35 1-10

Mic

rocy

stin

( g

/L)

0.0

0.2

0.4

0.6

0.8

1.0

a

a, b

bbb

n=24

Graham and Jones, 2007

Letters indicate significant differencesin mean concentration

Net Size (m)

% U

ndere

stim

ate

of

Tota

l M

icro

cyst

in

0

20

40

60

80

100

100 53 35 10

n=24

Use of Plankton Nets Consistently Underestimated Microcystin Concentrations Relative to Whole Water Samples

Graham and Jones, 2007

There are Currently Over 80 Known Microcystin Variants

From McKinnon, 2003

Analytical Methods for Cyanotoxins - Bioassays

BioassaysEnzyme-linkedimmunosorbent assays(ELISA) - Microcystins/Nodularin - Cylindrospermopsins - Saxitoxins

Inhibition Assays - Protein Phosphatase Inhibition (Microcystins/Nodularin)

Radioassays - Neurotoxicity (Anatoxins/

Saxitoxins)

AdvantagesEasy to Use

Rapid

Inexpensive

Useful screening tools

May indicate toxicity

DisadvantagesCross-reactivity

Matrix effects

Semi-quantitative

Radioassays use radio-labeled isotopes

Analytical Methods for Cyanotoxins – Gas Chromatography

AdvantagesSpecificity

Intermediate cost

Quantitative

DisadvantagesAvailability of analytical

standards

Derivitization likely required

Not all compounds are

amenable to derivitization

GC-FID requires further

confirmation

Sample concentrating may

be necessary

Gas Chromotography (GC)Flame ionization detector (FID)

Mass spectrometry (MS)

Analytical Methods for Cyanotoxins – Liquid Chromatography

AdvantagesSpecificity

Derivitization not typically

necessary

Many toxins amenable to

LC techniques

Multi-analyte methods

are cost-effective

TOFMS good for

determining unknowns

(not quantitative)

DisadvantagesAvailability of analytical

standards

Matrix effects

Expensive

Sample concentrating may

be necessary

Spectroscopic techniques

may require further

confirmation

Liquid Chromotography (LC)UV-Visible (UV-Vis)

Fluorescence

Mass spectrometry (MS)

Tandem MS (MS/MS)

Ion trap MS (ITMS)

Time of flight MS(TOFMS)

Robust and Quantitative Analytical Methods - Capabilities of the USGS Organic Geochemistry Research Laboratory

http://ks.water.usgs.gov/Kansas/researchlab.html

Geosmin and MIB MRL: 5 ppt

Algal Toxin AnalysisLC/MS/MS Chromatogram

Pea

k In

tens

ity

Elution Time - Minutes

CYL

ATX

MC-R

R

MC-L

RMC-Y

RM

C-LW

MC-L

F

MC-LA

Toxin MRL’s: ~25 ppt

Robust and Quantitative Analytical Methods - Capabilities of the USGS Organic Geochemistry Research Laboratory

http://ks.water.usgs.gov/Kansas/researchlab.html

Geosmin and MIB MRL: 5 ppt

Algal Toxin AnalysisLC/MS/MS Chromatogram

Pea

k In

tens

ity

Elution Time - Minutes

CYL

ATX

MC-R

R

MC-L

RMC-Y

RM

C-LW

MC-L

F

MC-LA

Toxin MRL’s: ~25 ppt

MC-LYDeoxycylindrospermopsinLyngbyatoxin-aNodularin-R

Total Microcystin Comparison – ADDA Specific ELISA vs LC/MS/MS for –LR, -RR, -LY, -YR, -LA, -LW, and –LF variants

Total Microcystin by ELISA (g/L)

0 10 20 30 40 50

To

tal M

icro

cyst

in

by

LC

/MS

/MS

( g

/L)

0

50

100

150

200

250r2=0.36n=22

ELISA (ADDA) can be a useful tool in conjunction with LC/MS/MS

Total Microcystin by ELISA (g/L)

0 10 20 30 40 50

To

tal M

icro

cyst

in

by

LC

/MS

/MS

( g

/L)

0

50

100

150

200

250r2=0.94n=18

Implies microcystin or microcystin-like congeners were measured by ELISA, but not LC/MS/MS

Distribution of Microcystin Variants and Other Cyanobacterial Toxins – August 2006 Midwestern Cyanotoxin Lake and Reservoir Reconnaissance

• Objectives:– Characterize occurrence and co-occurrence of taste and odor compounds and

cyanotoxins– Determine the specific toxins by LC/MS/MS

• Design:– States: IA, KS, MN, MO (23 Lakes and Reservoirs)– Targeted Sampling: Blooms and Scums– Analyses:

• Taste and Odor – SPME GC/MS• Toxins – TOTAL and Dissolved Concentrations

– ELISA – Microcystins (ADDA), Microcystin LR, Cylindrospermopsins, Saxitoxins– LC/MS/MS – 7 microcystins (LR, RR, YR, LW, LA, LF, LY), Nodularin, Anatoxin-

a, Cylindrospermopsin, Deoxycylindrospermopsin, Lyngbyatoxin a • Water Chemistry• Chlorophyll• Phytoplankton

During August 2006 100% of BLOOMS Sampled (n=23) Had Detectable Microcystin, 83% Had Detectable Geosmin, and 26% Had Detectable Anatoxin

TOTAL Microcystin Maxima (12,500 – 18,030 µg/L) in BLOOM Samples Were Orders of Magnitude Greater Than Maxima for Other Compounds

(Anatoxin Maxima = 13 µg/L, All Other Maxmima < 1 µg/L)

Mic

rocy

stin

(ELIS

A)

Mic

rocy

stin

(LC/M

S/MS)

Geosm

inM

IB

Anatoxi

n

Saxito

xin

Cylin

drosp

erm

opsin

Nodularin

Co

nce

ntr

atio

n ( g

/L)

0255075

100125150175200

12000

15000

18000

During August 2006 Toxins and Taste-and-Odor Compounds Co-Occurred in 87% of BLOOMS Sampled (n=23) and Anatoxin-a Always Co-Occurred with Geosmin

During August 2006 Toxins and Taste-and-Odor Compounds Co-Occurred in 87% of BLOOMS Sampled (n=23) and Anatoxin-a Always Co-Occurred with Geosmin

“Algae may make for stinky water, but it poses no health risks”

-Concord Monitor, Concord, NH July 7, 2006

Although Toxins and Taste-and-Odor Compounds Frequently Co-Occurred Concentrations Were Not Linearly Related

Microcystin by ELISA (g/L)

0 10 20 30 40 50

Microcystin by LC/MS/MS (g/L)

0 50 100 150 200

Geo

smin

( g

/L)

0.0

0.2

0.4

0.6

0.8

1.0r2=0.04p=0.34n=22

r2<0.01p=0.67n=22

Cyanobacterial BLOOM with TOTAL Microcystin = 0.6 μg/L, Anatoxin = 0.1 μg/L , and Geosmin = 0.02 μg/L

Cyanobacterial BLOOM with TOTAL Microcystin = 12.3 μg/L, Nodularin = 0.1 µg/L, Geosmin = 0.02 μg/L, and MIB = 0.06 μg/L

Cyanobacterial BLOOM with TOTAL Microcystin = 18,000 μg/L, Cylindrospermopsin = 0.12 μg/L Saxitoxin = 0.04 μg/L, and Geosmin = 0.69 μg/L

Microcystin-LR and –RR Were the Most Common Microcystin Variants, and 41% of Lakes Had All 7 Measured Variants Present

Microcystin-LR and –RR Comprised the Majority of TOTAL Microcystin Concentrations

IA1

IA2

IA3

IA4

IA5

IA6

IA7

IA8

IA9KS1

KS2KS3

KS4KS5

MN1

Mn2

MN3

MN4

MN5

MN6

MO1

MO2

To

tal

Mic

rocy

stin

(

g/L

)

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

LR RR LY YR LA LW LF

IA2

IA3

IA4

IA5

IA6

IA7

IA8

IA9KS1

KS2KS3

KS4KS5

MN1

Mn2

MN3

MN4

MN5

MN6

MO1

MO2

0

50

100

150

200

250

IA3

IA4

IA5

IA6

IA7

IA8

IA9KS1

KS2KS3

KS4KS5

MN1

Mn2

MN3

MN4

MN5

MN6

MO1

MO2

0

10

20

30

40

50

2006 Texas Reservoir Survey for DISSOLVED Microcystin in Surface Samples at OPEN WATER Locations

Results:

28% of reservoirs (n=36) had

detectable microcystin by ELISA

Maximum DISSOLVED

Microcystin concentrations: < 1 µg/L

69% of reservoirs had detectable

MIB

30% of reservoirs had detectable

Geosmin

After Kiesling and others, in prep

Microcystins and Taste-and-Odor Compounds Frequently Co-Occurred in Texas Reservoirs

After Kiesling and others, in prep

2007 US EPA National Lake Assessment: ~1200 Lakes and ReservoirsTOTAL Microcystin in Integrated Photic Zone Samples

Preliminary Results:

33% of samples (n=711) haddetectable microcystin by ELISA

Mean TOTAL microcystin concentration: 0.97 µg/L

Maximum TOTAL microcystinconcentration: 74 µg/L

Sample Location and Type are Important

Study Sample Location

Sample Type n

% Samples with MC

Maximum MC

(µg/L)

Graham and others 1999-2006

Open Water, Integrated Photic

Total 2546 39 52

Midwest Recon

2006

Targeted Blooms, Bloom Grab

Total 23 96 12,500

Texas Recon

2006

Open Water, Surface Grab

Dissolved 67 22 0.2

EPA NLA

2007

Open Water, Integrated Photic

Total 711 33 74

Microcystin was measured by ELISA in all studies

Long Term Studies – Assessment of Water Quality in the North Fork Ninnescah River and Cheney Reservoir, 1997-Present

• Concerns– Taste-and-odor occurrences related to algal blooms

– Relation between watershed inputs and taste-and-odor causing algae

• Approach– Describe current and historical loading inflow

• Sediment Cores• Continuous Water-Quality Monitoring

– Describe physical, chemical, and biological processes associated with cyanobacteria and cyanobacterial by-products

• Discrete Samples• Real-Time Monitors

Cheney Reservoir, KS June 2003Photo Courtesy of KDHE

North Fork Ninnescah RiverMarch 2006

http://ks.water.usgs.gov/Kansas/studies/qw/cheney/

Early Detection and Predictive Models – Continuous Real-Time Water-Quality Monitors

• Real-Time Variables– Specific conductance, pH,

temperature, turbidity, dissolved oxygen

– Chlorophyll

– Light

– Blue-green algae

– Nitrate

The J. W. Powell USGS Monitoring Station on Lake Houston, Texas

Station Developed by Michael J. Turco, Timothy D. Oden, William H. Asquith, Jeffery W. East, and Michael R. Burnich

http://waterdata.usgs.gov/tx/nwis/

Continuous Monitoring Allows the Identification and Description of Events that Occur Within Relatively Short Periods of Time

http://waterdata.usgs.gov/tx/nwis/

http://ks.water.usgs.gov/Kansas/rtqw/index.shtmlEstimated Geosmin Concentration 2003

Geosmin

ElevationE

stim

ate

d G

eo

smin

Co

nce

ntr

ati

on

(μ

g/L

)

Early Detection - Geosmin Concentrations in Cheney Reservoir Frequently Exceed the Human Detection Limit of 10 ng/L

log10(Geo) = 7.2310 - 1.0664 log10(Turb) - 0.0097 SCr2=0.71

After Christensen and others, 2006 USGS SIR 2006-5095

http://ks.water.usgs.gov/Kansas/rtqw/index.shtml

Jennifer Graham jlgraham@usgs.gov (785) 832-3511

Additional Information Available on the Web:

Cyanobacteria - http://ks.water.usgs.gov/Kansas/studies/qw/cyanobacteriaCheney - http://ks.water.usgs.gov/Kansas/studies/qw/cheneyOlathe - http://ks.water.usgs.gov/Kansas/studies/qw/olatheRTQW - http://ks.water.usgs.gov/Kansas/rtqw/index.shtml

Keith Loftinkloftin@usgs.gov(785) 832-3543