<%BANNER%>

Molecular Investigations of Cylindrospermopsin Production in Florida Lakes

Permanent Link: http://ufdc.ufl.edu/UFE0021749/00001

Material Information

Title: Molecular Investigations of Cylindrospermopsin Production in Florida Lakes
Physical Description: 1 online resource (79 p.)
Language: english
Creator: Yilmaz, Mete
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: 16srdna, aphanizomenon, cylindrospermopsin, cylindrospermopsis, peptide, polyketide, synthase, synthetase
Fisheries and Aquatic Sciences -- Dissertations, Academic -- UF
Genre: Fisheries and Aquatic Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Large numbers of toxic cyanobacterial blooms have been reported for lakes, reservoirs, and rivers around the world. The presence of toxic species in surface waters that are used for drinking water and recreation have received special attention due to health concerns and effects on ecosystems. Cylindrospermopsin (CYN), a cyanobacterial toxin, shows hepatotoxic, cytotoxic and genotoxic activities. In an attempt to identify CYN-producers in Florida, USA, different species and strains of filamentous cyanobacteria were isolated from lakes in the state. Cylindrospermopsis raciborskii is generally assumed to be the source of CYN in lakes and rivers in Florida. However, none of the eight Florida isolates of this species that were tested contained the genetic determinants involved in toxin production, nor did they produce CYN. I show for the first time that Aphanizomenon ovalisporum isolated from a pond in this state has the genes associated with CYN production. Analysis by liquid chromatography with mass spectrometric detection (LC/MS) revealed that it produced CYN in the range of 7.39 to 9.33 ?g mg-1 freeze-dried cells. The 16S rDNA sequences of this strain showed 99.6% and 99.9% identity to published A. ovalisporum and Anabaena bergii 16S sequences, respectively. To determine if CYN genes were present in the lakes examined and to reveal the variety of CYN producers, water samples were collected from different freshwater lakes in Florida. However, due to inhibitors co-purifed with genomic DNA, it was not possible to amplify CYN genes by using these templates. Humic substances are believed to be the most common inhibitors of enzymes used in molecular biology, such as Taq polymerase in PCR. A new approach to genomic DNA isolation from environmental cyanobacteria samples was developed. The genomic DNA obtained by the new method was of high quality, had a high molecular weight, was amplified in the Polymerase Chain Reaction (PCR), and was amenable to restriction nucleases. These methods, to my knowledge, are the first to describe the removal of PCR inhibitors from lake samples during cyanobacterial DNA isolation. CYN genes, polyketide synthase (PKS) and peptide synthetase (PS) were detected in three samples from Crescent Lake, Buffalo Bluff, and Lake George collected in May 2007. All sites had high numbers of C. raciborskii and low numbers of A. ovalisporum. Restriction Fragment Length Polymorphism (RFLP) analysis of PS genes suggested A. ovalisporum as the likely source of CYN in these systems.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mete Yilmaz.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Phlips, Edward J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021749:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021749/00001

Material Information

Title: Molecular Investigations of Cylindrospermopsin Production in Florida Lakes
Physical Description: 1 online resource (79 p.)
Language: english
Creator: Yilmaz, Mete
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: 16srdna, aphanizomenon, cylindrospermopsin, cylindrospermopsis, peptide, polyketide, synthase, synthetase
Fisheries and Aquatic Sciences -- Dissertations, Academic -- UF
Genre: Fisheries and Aquatic Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Large numbers of toxic cyanobacterial blooms have been reported for lakes, reservoirs, and rivers around the world. The presence of toxic species in surface waters that are used for drinking water and recreation have received special attention due to health concerns and effects on ecosystems. Cylindrospermopsin (CYN), a cyanobacterial toxin, shows hepatotoxic, cytotoxic and genotoxic activities. In an attempt to identify CYN-producers in Florida, USA, different species and strains of filamentous cyanobacteria were isolated from lakes in the state. Cylindrospermopsis raciborskii is generally assumed to be the source of CYN in lakes and rivers in Florida. However, none of the eight Florida isolates of this species that were tested contained the genetic determinants involved in toxin production, nor did they produce CYN. I show for the first time that Aphanizomenon ovalisporum isolated from a pond in this state has the genes associated with CYN production. Analysis by liquid chromatography with mass spectrometric detection (LC/MS) revealed that it produced CYN in the range of 7.39 to 9.33 ?g mg-1 freeze-dried cells. The 16S rDNA sequences of this strain showed 99.6% and 99.9% identity to published A. ovalisporum and Anabaena bergii 16S sequences, respectively. To determine if CYN genes were present in the lakes examined and to reveal the variety of CYN producers, water samples were collected from different freshwater lakes in Florida. However, due to inhibitors co-purifed with genomic DNA, it was not possible to amplify CYN genes by using these templates. Humic substances are believed to be the most common inhibitors of enzymes used in molecular biology, such as Taq polymerase in PCR. A new approach to genomic DNA isolation from environmental cyanobacteria samples was developed. The genomic DNA obtained by the new method was of high quality, had a high molecular weight, was amplified in the Polymerase Chain Reaction (PCR), and was amenable to restriction nucleases. These methods, to my knowledge, are the first to describe the removal of PCR inhibitors from lake samples during cyanobacterial DNA isolation. CYN genes, polyketide synthase (PKS) and peptide synthetase (PS) were detected in three samples from Crescent Lake, Buffalo Bluff, and Lake George collected in May 2007. All sites had high numbers of C. raciborskii and low numbers of A. ovalisporum. Restriction Fragment Length Polymorphism (RFLP) analysis of PS genes suggested A. ovalisporum as the likely source of CYN in these systems.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mete Yilmaz.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Phlips, Edward J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021749:00001


This item has the following downloads:


Full Text

PAGE 1

1 MOLECULAR INVESTIGATIONS OF CY LINDROSPERMOPSIN PRODUCTION IN FLORIDA LAKES By METE YILMAZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Mete Y lmaz

PAGE 3

3 To my parents, who supported me in every stage of my PhD.

PAGE 4

4 ACKNOWLEDGMENTS I would lik e to express my sincere appreciation to my supervisor Dr. Edward J. Phlips for his support and encouragement. I have always ad mired his relationship with his students and his philosophy in science will lead me in my career as a scientist. I thank all my committee members for useful di rections and suggestions They helped me to think the results of my experiments in a wide r perspective. I thank Dr. Nancy Szabo from the Center for Human and Environmental Toxicology at UF and Dr. Daniel Tillett from the School of Pharmacy and Applied Sciences, La Trobe Univer sity in Australia. Their help and suggestions in experiments improved the overall quality of my dissertation. Special thanks to my good friend and collea gue Dana Bigham for collecting samples from lakes Dora and Griffin. Special appreciation also goes to Dr. Jim Austin for lending me a thermal cycler, electrophoresis equipment and UV-transilluminator. I am grateful to Dr. Hans Paerl from the Univ ersity of North Carolina, Chapel Hill and Dr. Pia Moisander from the University of Californi a, Santa Cruz for providing me with five C. raciborskii cultures. I am also thankful to Dr. Ge offrey Eaglesham from the Queensland Health Scientific Services, Australia and Dr. Geoffrey Codd from the University of Dundee, Scotland for providing the CYN standards and a C. raciborskii culture.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4LIST OF TABLES................................................................................................................. ..........7LIST OF FIGURES.........................................................................................................................8ABSTRACT.....................................................................................................................................9 CHAP TER 1 INTRODUCTION..................................................................................................................112 A COMPARATIVE STUDY OF FLORIDA STRAINS OF Cylindrospermopsis AND Aphanizomenon FOR CYLINDROSPERMOPSI N PRODUCTION..................................... 16Introduction................................................................................................................... ..........16Materials and Methods...........................................................................................................18Isolation, Culture and Morphometric M easurements of the Cyanobacteria....................18Genomic DNA isolation, Amplification and Sequencing............................................... 19Cylindrospermopsin Analysis......................................................................................... 20Results.....................................................................................................................................21Morphology of A. ovalisporum and C. raciborskii Strains.............................................21Comparison of 16S rRNA Gene Sequences.................................................................... 22Amplification and Sequences of Partial PKS and PS Genes........................................... 23Toxin Content of Florida and Australian Isolates...........................................................23Discussion...............................................................................................................................243 AN IMPROVED METHOD FOR THE ISOL ATION OF INHIBITOR-FREE CYANOBACTERIAL DNA FROM ENVIRONMENTAL SAMPLES............................... 35Introduction................................................................................................................... ..........35Materials and Methods...........................................................................................................37Cyanobacterial Strains Used and Collection of Lake Samples.......................................37Genomic DNA Isolations................................................................................................ 38DNA Quantification........................................................................................................ 39Polymerase Chain Reaction............................................................................................. 40Restriction Endonuclease Di gestion of Genomic DNA.................................................. 41Results.....................................................................................................................................41DNA Yield with Different Measurement Methods.........................................................41Isopropanol versus PEG Precipitation............................................................................. 42Restriction Digesti on of Genomic DNA......................................................................... 43Discussion...............................................................................................................................44

PAGE 6

6 4 RESTRICTION FRAGMENT LENGTH POLYMORPHIS M (RFLP) ANALYSES OF CYLINDROSPERMOPSIN (CYN ) BIOSYNTHESIS GENES IN FLORIDA LAKES...... 53Introduction................................................................................................................... ..........53Materials and Methods...........................................................................................................56Control Cyanobacterial Strains........................................................................................ 56Lake Samples...................................................................................................................56Genomic DNA Isolation.................................................................................................. 56Polymerase Chain Reaction (PCR) and Inte rnal Control Fragment (ICF) Design......... 57Restriction Digesti on of PS Fragments........................................................................... 58Phylogenetic Analysis.....................................................................................................58Results.....................................................................................................................................58Determination of Optimum ICF Concentration.............................................................. 58Screening of Lake Samples for CYN Biosynthesis Genes.............................................. 59Purification and RFLP Analyses of PS Frag ments from the St. Johns River System Samples........................................................................................................................59Phylogenetic Relationship of A. ovalisporum to Other Cyanobacteria........................... 60Discussion...............................................................................................................................615 CONCLUSIONS.................................................................................................................... 69LIST OF REFERENCES...............................................................................................................71BIOGRAPHICAL SKETCH.........................................................................................................79

PAGE 7

7 LIST OF TABLES Table page 1-1 Classification of cyanotoxins............................................................................................. 152-1 List and origins of the strains used in the study................................................................. 292-2 Morphometric measurements (m) of Aphanizomenon ovalisporum ...............................292-3 Concentrations of CYN in C. raciborskii strains and A. ovalisporum ...............................303-1 Lake samples used, collection dates and locations............................................................ 483-2 Oligonucleotide primers used in the study......................................................................... 483-3 Total genomic DNA yield from two lake samples and A. ovalisporum ............................494-1 Lake samples used in PCR amplifications for CYN genes................................................ 63

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Structure of cylindrospermopsin........................................................................................ 152-1 Trichomes of isolated A.ovalisporum strain FAS-AP1......................................................312-2 PCR products for partial PKS and PS genes......................................................................322-3 Selected ion chromatograms for QHSS/NR/CYL/03, FAS-AP1, and FAS-C1................ 333-1 PCR results for the 16S rDNA, PC-IGS a nd mcyA for the Lake George sample............. 503-2 mcyA PCR results for lake samples and two cultures........................................................513-3 Restriction digestion of genomic DNA with BmrI............................................................ 524-1 PCR results with different concentr ations of ICF for PKS and PS genes.........................644-2 PCR results of PKS genes from Hillsborough County lake samples................................. 654-3 PCR results for PKS and PS genes fr om the St. Johns River Samples.............................. 664-4 RFLP analyses on the PS fragment s from the St. Johns River Samples............................ 674-5 Neighbor-joining tree of selected cya nobacteria based on 16S rDNA sequences............. 68

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MOLECULAR INVESTIGATIONS OF CY LINDROSPERMOPSIN PRODUCTION IN FLORIDA LAKES By Mete Y lmaz December 2007 Chair: Edward J. Phlips Major: Fisheries a nd Aquatic Sciences Large numbers of toxic cyanobacterial blooms have been reported for lakes, reservoirs, and rivers around the world. The presence of toxi c species in surface waters that are used for drinking water and recreation have received special attention due to health concerns and effects on ecosystems. Cylindrospermopsin (CYN), a cyanobacterial toxin, shows hepatotoxic, cytotoxic and genotoxic activities. In an attempt to identify CYN-producers in Florida, USA, different species and strains of filamentous cyanob acteria were isolated from lakes in the state. Cylindrospermopsis raciborskii is generally assumed to be the source of CYN in lakes and rivers in Florida. However, none of the eight Florida isolates of this species that were tested contained the genetic determinants involved in toxin produc tion, nor did they produce CYN. I show for the first time that Aphanizomenon ovalisporum isolated from a pond in this state has the genes associated with CYN production. Analysis by liquid chromatography with mass spectrometric detection (LC/MS) revealed that it pr oduced CYN in the range of 7.399.33 g mg-1 freeze-dried cells. 16S rDNA sequences of this strain showed 99.6 % and 99.9 % identity to published A. ovalisporum and Anabaena bergii 16S sequences, respectively. To determine if CYN genes were present in the lakes examined and to reve al the variety of CYN producers, water samples were collected from different freshwater lakes in Florida. However, due to inhibitors co-purifed

PAGE 10

10 with genomic DNA, it was not possible to amplif y CYN genes by using these templates. Humic substances are believed to be the most common inhibitors of enzymes used in molecular biology, such as Taq polymerase in PCR. A new approach to genomic DNA isolation from environmental cyanobacteria samples was deve loped. The genomic DNA obtained by the new method was of high quality, had a high molecula r weight, was amplified in the Polymerase Chain Reaction (PCR), and was amenable to re striction nucleases. These methods, to my knowledge, are the first to describe the removal of PCR inhibitors from lake samples during cyanobacterial DNA isolation. CYN genes, polyke tide synthase (PKS) an d peptide synthetase (PS) were detected in three samples from Cr escent Lake, Buffalo Bluff, and Lake George collected in May 2007. All sites had high numbers of C. raciborskii and low numbers of A. ovalisporum Restriction Fragment Length Polymorphism (RFLP) analysis of PS genes suggested A. ovalisporum as the likely source of CYN in these systems.

PAGE 11

11 CHAPTER 1 INTRODUCTION Bacteria, cyanobacteria and fungi produce chem ically and functionally different types of peptides and polyketides, which are collectively called secondary m etabolites. Some of these secondary metabolites act as proteinase inhib itors, anti-algal and an ti-fungal agents, immunosuppressors, and promoters of cell differentiation Cyanobacterial toxins, such as microcystins and cylindrospermopsin, are examples of such secondary metabolites (Dittmann et al. 2001) A large number of toxic cyanobacterial blooms ha ve been reported for lakes, reservoirs, and rivers around the world (Paerl et al. 2001; Phlips 2002). The presence of toxic species in surface waters represents a potential threat to the integrity and sust ainability of aquatic ecosystems. Due to use of some affected systems as sources of drinking water and recreation, they are receiving special attention as a threat to human and animal health (Hitzfeld et al. 2000). In 1996 more than 50 dialysis patients died as a result of cyanobacterial toxins in a water supply in Brazil, and in 1989, a Microcystis bloom in a water storage reservoir in England killed 20 lambs and 15 dogs (Haider et al. 2003). Classification of cyanobacterial toxins is generally based on two features: Chemical structure and mode of action (Tab le 1-1). Cylindrospermopsin is an unusual alkaloid (Figure 11) having both a cyclic guanidine group and a su lphate group that make it zwitterionic and highly water soluble, with a molecular weight of 415 (No rris et al. 2002; Griffi ths and Saker 2003). It is stable in aqueous solutions of different pH and li ght conditions (Chiswe ll et al. 1999). The pyrimidine ring is essential for to xicity (Griffiths and Saker 2003) There are three different chemical forms of cylindrospermopsin: Cylindrospermopsin, 7epicylindrospermopsin, and deoxyc ylindrospermopsin. 24 hour LD50 of cylindrospermopsin in a mouse bioassay is 2 mg kg whereas 5 day LD50 of this toxin is 0.2 mg kg-1. In the mouse

PAGE 12

12 bioassay, cylindrospermopsin and 7-epicylindrospermopsin have the same toxicity. Deoxycylindrospermopsin shows greatly re duced toxicity (Metcalf et al. 2002). In Florida, USA, two species of cyanobacteria have been singl ed out as the most important sources of toxins in freshwater ecosystems, Microcystis aeruginosa and Cylindrospermopsis raciborskii (Chapman and Schelske, 1997; Phlips, 2002). The focus of this dissertation was C. raciborskii which is one of the most widespread and prolific bloom-forming species of algae in Florida. Cylindrospermopsin has been reported to be produced by Cylindrospermopsis raciiborskii isolated from various regions globall y (Griffiths and Sake r 2003). In 1979, C. raciiborskii caused 148 residents of Palm Island, Australia to be hospitalized with he patitis-like symptoms. In addition, deaths of cattle have been repor ted from different regions of Australia where C. raciborskii blooms occurred (Griffiths and Saker 2003) Symptoms of cylindrospermopsin damage in liver are distinguishable from da mages caused by other hepatotoxins such as microcystins (Saker et al. 2003). There are no pathological differences of cylindrospermopsin toxicity in mice when the toxin is administered either as a suspension of freeze dried culture or purified compound, and there are no differences if it is administered orally or intraperitoneally (Norris et al. 2002). The initial objectives of this dissertation were to define 1) The relationship between Cylindrospermopsi s blooms and the presence of the toxi n cylindrospermopsin (CYN), and 2) the environmental and genetic factors that ma y regulate the level of toxin production. Cylindrospermopsis raciborskii is generally assumed to be the source of CYN in lakes and rivers in Florida (Chapman and Schelske 1997). Early studies of CYN concentrations in Florida lakes have shown a poor relationship between the amount of toxin present and the size of

PAGE 13

13 Cylindrospermopsis blooms (Aubel et al., 2006). Orig inally it was hypothesized that this observation was based on environmental regulation of toxin pr oduction. In the early phases of the current research effort, strains of C. raciborskii from a number of lake s and rivers in Florida were isolated and tested for both presence of the toxin and the genetic determinants for CYN production. Surprisingly, none of the Cylindrospermopsis isolates contained the genes for the production of the toxin. This led to a more intensive search for the organisms that may be responsible for the toxin production, which led to the discovery of a species of Aphanizomenon with the genetic determinants. This search fo rms the basis of Chapter 2 of this dissertation. Building upon the unexpected results of the fi rst phase of researc h, the search for CYN genes was extended to lake and river water samples in an effort to establish the scope of influence of A. ovalisporum as a producer of CYN. Cyanobacteria were harvested by filtration and used in whole community genomic DNA extraction. The presence of inhibitors in natural water samples co-purified with genomic DNA, however, made it impossible to amplify CYN genes. Humic substances are believed to be th e most common inhibitors of enzymes used in molecular biology, such as Taq polym erase in PCR (Pan et al. 2002). The third chapter of the dissertation describes the developm ent of a new approach to cyanobacterial DNA isolation from lake samples. This method resolves problems associated with inhibition of PCR reactions. The genom ic DNA obtained using the improved method was of high quality, had a high molecular weight, and was digested with restriction nucleases. This method, to my knowledge, is the first work to desc ribe the removal of PCR inhibitors from lake samples during cyanobacterial DNA isolation. The fourth chapter describes screening of lake samples from Hillsborough County, the Harris Chain of Lakes, and the St. Johns River system for CYN biosynthesis genes using the

PAGE 14

14 improved DNA isolation method from environmen tal cyanobacteria. Amplification products from lakes were subjected to Restriction Frag ment Length Polymorphism (RFLP) analyses and compared with fragments from C. raciborskii and A. ovalisporum Results indicated that A. ovalisporum is the likely CYN producer in sa mples containing the CYN genes.

PAGE 15

15 Table 1-1. Classifica tion of cyanotoxins. Type of toxin Toxin Chemical structure Mode of action Cyclic peptides Alkaloids Hepatotoxins Neurotoxins Cytotoxins microcystin, nodularin anatoxin-a, anatoxin-a( s), cylindrospermopsin microcystin, cylindrospermopsin, nodularin anatoxin-a, anatoxin-a(s), saxitoxin cylindrospermopsin Figure 1-1. Structure of cylindrospermopsin.

PAGE 16

16 CHAPTER 2 A COMPARATIVE STUDY OF FLORIDA STRAINS OF Cylindrospermopsis AND Aphanizomenon FOR C YLINDROSPERMOPSIN PRODUCTION Introduction Cyanobacteria are prom inent f eatures of many freshwater and brackish systems around the world. Under suitable conditions they can reach high biomass levels and form blooms (Paerl et al., 2001; Phlips, 2002). One of the major concerns with blooms is that certain cyanobacteria have the ability to produce cyanotoxins as a s econdary metabolite. Cyanotoxins not only pose a risk to the integrity of aquatic ecosystems, but also to human health (Chorus and Bartram, 1999). One of the cyanotoxins of concern is cylindrospermopsin (CYN). First identified from Cylindrospermopsis raciborskii CYN is a tricyclic guanidine alkaloid associated with production of harmful metabolites an d the inhibition of protein synthesis (Froscio et al., 2003). It has been linked to gastrointestinal distress, liver damage, and, to a lesser degree, damage to other organs in a variety of animals (Falconer et al., 1999; Seawright et al., 1999; Shaw et al., 2000). CYN is genotoxic, causing DNA strand break age (Humpage et al., 2000; Shen et al., 2002). The human health risk associated with C. raciborskii was highlighted by a poisoning incident on Palm Island, Australia in 1979, in whic h 148 people were hospitalized with hepatitislike symptoms after exposure to CYN-contaminat ed water (Bourke et al., 1983; Hawkins et al., 1985). Initially described as a tropi cal and sub-tropical species, Cylindrospermopsis has also been found in temperate lakes across Asia, Europe, Nort h and South America, which demonstrates its wide physiological tolerance (P adisk, 1997; Briand et al., 2004) Although found worldwide, only the Australian and some Asian strains of Cylindrospermopsis have been shown to produce CYN (Hawkins et al., 1985; Li et al., 2001a). Some Brazilian strains have been shown to produce saxitoxin (Lagos et al., 1999), and some European strains of C. raciborskii have shown

PAGE 17

17 varying degrees of toxicity to mice, rat he patocytes and human cell lines, but without the demonstrated ability to produce CYN or saxitoxin (Fastner et al., 2003; Sake r et al., 2003). Over the past decade, species of cyanobacteria other than C. raciborski have been shown to produce CYN, including Umezakia natans from Japan (Harada et al., 1994), Raphidiopsis curvata from China (Li et al., 2001b), Anabaena bergii from Australia (Schembri et al., 2001), Aphanizomenon ovalisporum from Israel (Banker et al., 1997), Aphanizomenon flos-aquae from Germany (Preussel et al., 2006), Anabaena lapponica from Finland (Spoof et al., 2006) and Lyngbya wollei from Australia (Sei fert et al., 2007). The recent observation that CYN is generate d by a range of cyanobacteria species has initiated efforts to define the spatial distributi on and sources of CYN. The goal of this study was to explore the capacity for CYN production among strains of Cylindrospermopsis encountered in Florida, USA, as well as other potential toxin-producing genera. Cylindrospermopsis was first reported in Florida in 1986 (Hodgson et al., 19 86). Subsequent research has shown that Cylindrospermopsis is a major bloom-forming species in th is state (Cichra et al., 1995; Chapman and Schelske, 1997; Phlips et al., 2003). Alt hough CYN has been detected in Florida lakes (Aubel et al., 2006), none of the isolated stra ins of this genus produce CYN or contain the genetic determinants involved in toxin produc tion (Neilan et al., 2003; Kellmann et al., 2006). Schembri et al. (2001), by utilizing degenerate primers, determined for the first time that genes encoding polyketide synthase (PKS) and pep tide synthetase (PS) enzymes are found in cylindrospermopsin-producing C. raciborskii and Anabaena bergii ( var. limnetica ), but are absent in strains that do not produce the toxin. Later work by Shalev-Alon et al. (2002) identified three genes putati vely involved in cylindros permopsin biosynthesis in Aphanizomenon ovalisporum These genes, aoaA encoding an amidinotransferase, aoaB encoding a hybrid non-

PAGE 18

18 ribosomal peptide synthetase/polyketide synthase, and aoaC encoding a polyketide synthase, were located in the same genomic region, s uggesting their putative involvement in CYN biosynthesis. In the current study, individual strain s of cyanobacteria were isolated from environmental samples collected from various site s in Florida and were screened for the presence of the two genes putatively associated with the production of CYN, PKS a nd PS. Representative isolates were tested for CYN production. Genetic relationships between isolates were also examined based on the gene sequence anal yses of 16s rRNA. While none of the Cylindrospermopsis isolates produced CYN, I report for the first time the isolation of a CYNproducing Aphanizomenon ovalisporum from Florida. Materials and Methods Isolation, Culture and Morphometric Measurements of the Cyanobacteria Water sam ples collected from lakes and rivers in Florida (USA) were enriched in screw cap tubes with BG-11 medium (Stanier et al., 19 71) excluding a nitrogen source. After growth of cultures, individual filaments were transferred into 5 ml of fresh medium by micro-pipette. Isolated species were grown at 25 C havi ng a 12:12-h light-dark photoperiod under 30 mol photons m-2 s-1 supplied with cool white fluorescent lamp s. They were screened for PKS and PS determinants after reaching suffi cient density for DNA extraction. Two C. raciborskii strains (QHSS/NR/Cyl/03 and FAS-C1) and an A. ovalisporum strain (FAS-AP1) were grown in 4-L flasks containing 2-L of medium in a manner similar to that described above. The remaining strains were kept in culture tubes and used in the Polymerase Chain Reaction (PCR) and Enzyme-Linked Immuno-Sorbent Assay (ELISA). A C. raciborskii strain QHSS/NR/Cyl/03 obtained from Queensla nd Health Scientific Services (Queensland, Australia) was used as a positive control for CYN production and genetic determinants. The list of strains used in this study and th eir sources are provided in Table 2-1.

PAGE 19

19 Key morphological charact eristics of cultured A. ovalisporum and C. raciborskii, preserved with Lugols were determined microscopically using a Leica DM IL pha se contrast inverted microscope (Leitz, Wetzler Germany). Meas urements of cells were at 400X and 1000X. Genomic DNA isolation, Amplification and Sequencing Cyanobacterial cells were harvested by centr ifugation at 13000 rpm for 5 min. Genomic DNA isolation was performed using the Wizard Genomic DNA Purification Kit (Promega, WI, USA) following the method for gram-positive bact eria in accordance with the manufacturers instructions. Amplification of PKS and PS ge ne fragments were performed on a Biometra thermal cycler (Biometra GmbH i. L., Goetti ngen, Germany) according to Schembri et al. (2001). Each PCR reaction contained 10-20 ng of genomic DNA, 20 pmol of each primer (MWG-Biotech Inc., NC, USA), 200 M of each deoxynucleoside triphosphates (Fisher Scientific Company L.L.C, PA., USA), 1.5 mM MgCl2, 10 l of 5X green buffer, and 1.5 units of GoTaq DNA polymerase (Promega, WI, USA) in a total volume of 50 l. For amplification of partial PKS gene (624 bp), the PKS-specific primer pair M4 and M5 was used. PS-specific primers M13 and M14 were used to amplify a region of the PS gene (597 bp). Approximately the 1470 bp region of the 16s rDNA was amplified with the primer pair 27F1 (Neilan et al. 1997) and 1516R (5-ATCCAGCCACACCTTCCGG), the latter designed in this study. Each PCR reaction contained appr oximately 10-50 ng of genomic DNA, 20 pmol of each primer (MWG-Biotech Inc., NC, USA), 200 M of each deoxynucleoside triphosphates (Fisher Scientific Company L.L.C, PA., USA), 2.5 mM MgCl2, 10 l of 5X green buffer, and 1.25 units of GoTaq DNA polymerase (Promega, WI, USA) in a total volume of 50 l. Amplification conditions for all we re 95 C for 3 min, 30 cycles of 95 C for 30 s, 60 C for 30 s, 72 C for 1 min, and a final extension step at 72 C for 5 min. To determine sizes, 100 bp DNA

PAGE 20

20 ladder (Invitrogen, CA, USA) was run with PCR pr oducts on a 1.5% agarose gel. Gels were visualized under UV transillumination after staining with ethidium-bromide. PCR generated fragments of PKS, PS a nd 16S rDNA were used for sequencing. Fragments were purified from 1.5% agarose gels using a Qiaquick MinElute gel extraction kit (Qiagen, CA, USA). For PKS and PS, the same primers used in PCR amplification were used for sequencing of both strands. 16S rDNA fr agments were sequenced by using the newly designed sequencing primers 379F (5-GAATTTTCCGCAATGGGC), 1118F (5AACGAGCGCAACCCTCG), 522R (5-CCG TATTACCGCGGCTG), and 1180R (5CTTGACGTCATCCCCACCTT) along with PCR prim ers to obtain sequences on both strands. This was accomplished with a Perkin Elmer 377 automated DNA sequencer (Perkin Elmer, MA, USA), using the flourescent dideoxy terminator method of cycle se quencing (Smith et al., 1986). Sequences were corrected manually using Vect or NTI software (Invitrogen, CA, USA). Published sequences were obtained from the National Center for Bi otechnology Information (NCBI) databases (http:/ /www.ncbi.nlm.nih.gov/). DNA sequence alignments were performed by using Vector NTI software. GenBank accession numbers of the sequences obtained in this work are:. EU076457 to EU076459 (16S rRNA gene); EU076460 and EU076462 ( aoaC ); EU076461 and EU076463 ( aoaB ). Cylindrospermopsin Analysis Two C. raciborskii strains (QHSS/NR/Cyl/03 and FAS-C1) and one A. ovalisporum strain (FAS-AP1) were analy zed for CYN by LC/MS. Cultures in the late exponential phase were harvested by filtration and freeze-dried. Two sets of cultures, 30-day old and 40-day old, were analyzed for each strain. FAS-C1 was the only Florida strain of C. raciborskii that reached sufficient density to harvest and freeze-dry at the time of analysis. The remaining Florida C. raciborskii isolates were tested with a CYN ELISA kit (Abraxis LLC, PA, USA).

PAGE 21

21 For LC/MS analysis, CYN was isolated from freeze-dried tissues by extraction into a 1.5% solution of formic acid using a method based on the work by Trkne et al. (2004). Resulting extracts were analyzed using an HP1100 hi gh-performance liquid chromatography system (Hewlett-Packard Company, Wilmington, DE) with UV detection in series with an LCQ MS (Thermo Electron, Cincinnati, OH). The analytes were separated across an Adsorbosphere HS C18 column (150 x 4.6 mm, 5m; Alltech Chro motography, Deerfield, IL) under isocratic conditions at a flow rate of 0.6 ml/min in a mob ile phase consisting of 5:95 methanol: water with 0.1% formic acid. All samples were monitored fi rst at 262 nm before being introduced into the mass spectrometer via ESI probe in positive mode, where an m/z range of 300-2000 was evaluated for characteristic CYN fragme ntation pattern. Based on the M + H+ ion ( m/z 416), quantitation of CYN was determined ag ainst standard calibration curves (R2 > 0.995) having at least five points. The limit of detection was < 0.01 g mg-1 freeze-dried cells. Both freeze-dried samples of 40-day old cultures and the remainder of the strains were also analyzed by an ELISA kit for CYN. This is a direct competitive ELISA kit utilizing polyclona l antibodies. Exponential phase cultures of each strain were harveste d by centrifugation and were washed once with distilled water. Approximately 12 mg wet weight from each stra in was extracted for CYN with glass beads in distilled water. After centrifugation, supernatan ts were used in the ELISA. Freeze-dried cultures were extrac ted in distilled water by glass-beads. After centrifugation, supernatants were used in the ELISA according to manufacturers instructions. Results Morphology of A. ovalisporum and C. raciborskii Strains Strain FASAP1 was identified as Aphanizomenon ovalisporum in accordance with morphological features desi gnated by Komrek and Kov ik (1989) and Pollingher et al. (1998) (Figure 2-1). Morphological measurements (n= 30) of this strain are given in Table 2-2.

PAGE 22

22 Terminal cells were elongated and blunt. Am ong the 100 trichomes observed, 48 had terminal cells that were completely hyaline, 31 were partially hyaline and 21 were not hyaline. Heterocysts were roughly spherical. Terminal he terocysts were identifie d in 13% of trichomes (n=30) observed in the morphologi cal analysis. Heterocysts we re also located at various locations throughout the trichomes. Akinetes were either spherica l or broadly oval. Cylindrospermopsis raciborskii strains were identified ac cording to the morphological features outlined by Komrek and Kling ( 1991). Morphological characteristics of C. raciborskii (n=30) revealed solitary, strai ght or slightly curved trichomes that had a length range of 41m182m, with a mean of 89m. The width of trichomes varied from 1.9m to 2.7m, with a mean of 2.4m. Heterocysts were ovoid in shap e and always terminal, with lengths from 6.0m to 13m and a mean of 8.2m. The width at th e base of the heterocyst s ranged from 1.9m to 2.7m with a mean of 2.4m. There were eith er one or two terminal heterocysts on each trichome. Comparison of 16S rRNA Gene Sequences 16S rDNA s equence for Aphanizomenon ovalisporum strain FAS-AP1 (1431bp) showed 99.6 % identity to a previously published A. ovalisporum 16S rDNA sequence in the NCBI databases (1131/1136 identity with accession number AY335547). Similarly, the same sequence showed 99.9% identity to a published 16S rDNA sequence for Anabaena bergii (1398/1399 identity with accession number AF160256). C. raciborskii Florida strain FAS-C1 and the Australian strain of C. raciborskii QHSS/NR/Cyl/03 were also very similar in terms of their 16 S rDNA sequences with 99.7% (1428/1432) identity. Identity of 16S rDNA sequence for strain A. ovalisporum FAS-AP1 to sequences for the Florida strain of C. raciborskii, FAS-C1 and Australian strain QHSS/NR/Cyl/03 were lower, 93.1 % ( 1336/1435) and 93.2 % ( 1338/1435), respectively.

PAGE 23

23 Amplification and Sequences of Partial PKS and PS Genes Nine strains of C. raciborskii were included in this study, eigh t from Florida waters. Three of these strains (FL-D, FL-F, FL-I) were nega tive for PKS and PS determinants in previous studies (Neilan et al., 2003; Kellm ann et al., 2006). The five rema ining Florida isolates (FL-E, FL-L, FAS-C1, FAS-C2, FAS-C3) also tested ne gative for PKS and PS genes (Fig. 2-2). The known CYN-producing strain of C. raciborskii from Australia (QHSS/NR/Cyl/03) and the newly isolated and identified cyanobacterium Aphanizomenon ovalisporum (FAS-AP1) were the only strains giving positive amplification produc ts for both of the genes (Fig.2-2). Partial PKS sequence (583 bp) from A. ovalisporum strain FAS-AP1 shared 99.8% (539/540) and 100% (583/583) identity with published PKS sequences for A. bergii (accession number AF170844) and A. ovalisporum (accession number AF395828), respectively, in the NCBI databases. Partial PKS sequence (583 bp) from C. raciborskii strain QHSS/NR/Cyl/03 showed 99% to 100% identity to other published PKS sequences from C. raciborskii Identity between strains FAS-AP1 and QHSS/NR/Cyl/03 was 99.3% (579/583) for the PKS sequences. Partial PS sequence (552 bp) of A.ovalisporum strain FAS-AP1 shared 99.4% (531/534) and 100% (552/552) identity with sequences from A. bergii (accession number AF170843) and A. ovalisporum (accession number AF395828). Similar ity of PS sequences between the C. raciborskii strain QHSS/NR/Cyl/03 (552bp) and other published C. raciborskii PS sequences ranged from 99% to 100%. PS sequence of QHS S/NR/Cyl/03 showed 96.4% (532/552) identity to PS sequence for FAS-AP1. Toxin Content of Florida and Australian Isolates CYN was not detected in the 30or 40-day old cultures of C. raciborskii strain FAS-C1 by either the L C/MS (Figure 2-3) or ELISA methods. CYN was also not detected in the remaining seven Florida strains of C. raciborskii using ELISA. Concentrations of CYN obtained by

PAGE 24

24 LC/MS and ELISA for strains QHSS/NR/CYL/03 and FAS-AP1 are shown in Table 2-3. In general, 40-day old cultures contained higher leve ls of CYN than the 30-day old cultures for these strains. As evaluated by LC/MS, CYN leve ls for FAS-AP1 were consistently higher than for QHSS/NR/CYL/03. Although the same freeze-dried materials used for LC/MS evaluation were extracted and tested independently with di fferent lots of the CYN ELISA kit, the ELISA results were never in agreement with the LC/M S results for FAS-AP1; in each case the ELISA results were about half the le vel determined by LC-MS. In contrast, CYN concentrations, determined by LC/MS and ELISA for strain QHSS/ NR/CYL/03, were in excellent agreement. The difference in performance between the LC/MS and ELISA methods for FAS-AP1, as compared to QHSS/NR/CYL/03, may be due to the presence of an unidentified CYN isomer or congener produced by one of the strains. While a full comparative method evaluation was beyond the scope of this study, I intend to investigat e the cause(s) of these differences in future work. Discussion The results of this study dem onstrate that th e ability to produce cer tain toxins can vary considerably among strains of cyanobacterial spec ies that are largely i ndistinguishable from a morphological perspective. This is a significant observation considering the fact that cultural europhication of surface waters has increased th e world-wide occurrence s of putatively toxic algal blooms (Chorus and Bartram, 1999; Phlips, 2002). In many freshwater ecosystems, intense algal blooms are dominated by cyanobacteria (Can field et al., 1989). Fl orida contains over 7,000 lakes, many of which are eutrophic, and the to xinproducing ability of ubiquitous species like C. raciborski is a critical water management issue. C. raciborskii blooms have been observed in numerous lakes and rivers in this state (Cichra et al., 1995; Chapman and Schelske, 1997; Phlips et al., 2003). Owing to the fact that Australian and Brazilian strains of C. raciborskii produce the

PAGE 25

25 toxic compounds, CYN and saxitoxins widespread occurrences of C. raciborskii blooms have raised serious concerns about ri sks to the integr ity of affected ecosystems, as well as human health. To my knowledge, however, none of the isolated strains of this genus from Florida produce CYN or carry the genetic determinants i nvolved in CYN production (Neilan et al., 2003; Kellmann et al., 2006). The eight Florida strains of C. raciborskii examined in this study also did not produce PCR bands corresponding to PKS and PS gene fragments. These conclusions are further supported by the results of CYN analysis of unialgal cultures of Florida isolates of C. raciborski using LC/MS and ELISA, which did not yield measurable quantities of CYN, in contrast to significant quantities present in cultures of C. raciborskii from Australia (QHSS/NR/Cyl/03). While these data do not rule out the possibility of a CYN-producing C. raciborskii in Florida, it adds weight to the hypothesis that most Florida strains do not produce CYN. These results also help explain the ge neral lack of a defined relationship between abundance of C. raciborskii in freshwater ecosystems of Flor ida and observed concentrations of CYN (Aubel et al., 2006). The latte r observation raises the possibili ty that previous reports of CYN may have been caused by coincidental but unrecorded presence of another CYNproducing species. In an effort to identify species of cya nobacteria that may be responsible for CYN production, a number of lake and river water samples were screened in this study for genes putatively associated with CYN production. A sample from Duva l County, Florida gave positive bands for PKS and PS determinants (data no t shown). This sample contained both C. raciborskii and an unidentified filamentous cy anobacterium. Isolation and te sting of both species revealed that the genetic determinants for PKS and PS were present only in the unidentified cyanobacterium. Morphological observations of th e latter isolate (FAS-AP1 ) indicate that it is

PAGE 26

26 most closely allied with Aphanizomenon ovalisporum Intracellular concentrations of CYN associated with A. ovalisporum strain FAS-AP1 (i.e. 7.39 g mg-1 freeze-dried cells) fall within the higher range of reported CYN concentrations in all species tested (Fastner et al., 2003; Trkne et al., 2004; Preussel et al., 2006). To my knowledge, this is the first report of a CYNproducing cyanobacterium in the USA. Morphological observations and measurements of A. ovalisporum FAS-AP1 overlap with shapes and sizes of vegetative cells, heterocysts, and akinetes of strains previously identified as A. ovalisporum Anabaena bergii and A. minderi ( A. bergii var. limnetica ) (Pollingher et al., 1998; Shaw et al., 1999; Hind k, 2000). Komrek and Kov ik (1989) suggest that all three species should be pl aced in the genus Aphanizomenon. However, the debate as to whether these species belong to Anabaena or Aphanizomenon, or whether they should be in a different genus, continues (Komrek and Kov ik, 1989; Hindk, 2000; Gkelis et al., 2005). According to our observations and measurements, strain FAS-AP1 is most similar to A. ovalisporum and A. bergii. The akinetes of strain FAS-AP1 are either spherical or broadly oval, contrary to long elliptical akinetes in A. minderi ( A. bergii var limnetica ) (Hindk, 2000). It should also be noted that different culture conditions may affect shapes and sizes of cells, heterocysts and akinetes (Pollingher et al., 1998 ), complicating morphological comparisons. The results of sequence analyses of 16S rDNA of the Florida strain of A. ovalisporum show 99.6 % and 99.9% identity to published 16S rDNA sequences for A. ovalisporum and Anabaena bergii ( var. limnetica ), respectively. Although FAS-AP1 is more similar to A. bergii ( var. limnetica ) in terms of their 16S sequences, base d on the morphological criteria discussed above, I identify FAS-AP1 as Aphanizomenon ovalisporum In addition, there are no available morphological descriptions for A. bergii ( var. limnetica ) for which the 16S rDNA sequence is

PAGE 27

27 available. It has been suggest ed that these two species are mo rphological variants of the same species (Schembri et al., 2001). In a phyloge netic study based on 16S rRNA gene sequences, A. bergii ( var. limnetica ) clustered closely with some Nodularia Cyanospira and Anabaenopsis strains (Rajaniemi et al., 2005). A nother phylogenetic study also clustered A.ovalisporum A. bergii ( var. limnetica ), a Umazekia, an Anabaenopsis and a Cyanospira strain together. Based on similarity of 16S rDNA sequen ces, Kellmann et al. (2006) suggest ed that these species can be included in a revised genus. My results further support the need for a taxonomic revision of Aphanizomenon ovalisporum, Anabaena bergii and Anabaena minderi In the case of Cylindrospermopsis comparisons of 16S rDNA sequences between the Australian strain of C. raciborskii QHSS/NR/Cyl/03 and the Florid a strain FAS-C1 reveal high sequence similarity between these two strains (99.7 % identity). This is in good agreement with previous reports of Cylindrospermopsis strains from different ge ographic regions having high sequence similarities and forming a monophyletic group. However, slightly lower similarities are found between FAS-C1 and some of the pr eviously published 16S rDNA sequences for Florida strains,~ 99.1% as already reported by Neila n et al. (2003) for other Florida strains. PKS sequences for A. ovalisporum A. bergii and C. raciborskii share high sequence identity (99 100%). Although PS sequence identities within A. ovalisporum A. bergii and C. raciborskii strains are very high (99-100%), they are lower when compared with each other (9596%). Kellmann et al. (2006) suggest that a horizontal gene transfer event may have occurred between these species. Considering the variety of CYN producers observed around the world, A. ovalisporum may be one of several species of cyanobacteria that produce CYN in freshwater ecosystems of

PAGE 28

28 Florida. More species should be isolated and test ed from these systems. Furthermore, this study demonstrates that C. raciborskii is probably not a major CYN producer in Florida.

PAGE 29

29 Table 2-1. List and origins of the strains used in the study. Species Strain Origin C. raciborskii A. ovalisporum FL-D FL-E FL-F FL-I FL-L FAS-C1 FAS-C2 FAS-C3 QHSS/NR/Cyl/03 FAS-AP1 St. Johns River, Florida St. Johns River, Florida St. Johns River, Florida St. Johns River, Florida St. Johns River, Florida Lake Griffin, Florida Fishpond, Jacksonville, Florida Lake Brant, Florida Australia Fishpond, Jacksonville, Florida Table 2-2. Morphometric measurements (m) of Aphanizomenon ovalisporum strain FAS-AP1. Mean SD (n=30) Minimum Maximum Trichome Vegetative cell Terminal vegetative cell Heterocyst Akinete Length Length Width Length Width Length Width Length (n=11) Width (n=11) 243.9 4.0 4.5 4.9 3.8 6.4 7.1 10.7 7.6 127.1 0.9 0.9 1.0 0.9 1.1 1.1 1.8 1.8 34.0 2.5 2.0 3.0 3.0 4.8 5.8 9.0 6.0 628.0 5.5 5.0 7.9 6.0 8.2 8.0 15.0 12.0

PAGE 30

30 Table 2-3. Concentrations of CYN in C. raciborskii strains and A. ovalisporum as determined by LC/MS and CYN ELISA. Strain CYN (LC/MS) (g mg-1 freeze-dried cells) CYN (ELISA) (g mg-1 freeze-dried cells ) C. raciborskii QHSS/NR/Cyl/03 C. raciborskii FL-D C. raciborskii FL-E C. raciborskii FL-F C. raciborskii FL-I C. raciborskii FL-L C. raciborskii FAS-C1 C. raciborskii FAS-C2 C. raciborskii FAS-C3 A. ovalisporum FAS-AP1 3.44a e ND a 7.39a 6.73b ND b 9.33b 6.61 0.20b (n=3)c NDd ND ND ND ND ND b ND ND 5.03 0.36b (n=3)c a30-day old culture; b40-day old culture; cSamples for ELISA were run in triplicate, dNot detected, eNot analyzed.

PAGE 31

31 Figure 2-1. Trichomes (a, b) of isolated A.ovalisporum strain FAS-AP1 showing heterocysts (H), akinete (A), and terminal hyaline cell (arrow).Bar 10 m.

PAGE 32

32 Figure 2-2. Lanes 1 and 11, size marker; Lanes 2-10, PCR products for partial PKS genes: Lanes 2) QHSS/NR/CYL/03; 3) FAS-C1; 4) FAS-C2 ; 5) FAS-C3; 6) FL-D; 7) FL-E; 8) FLF; 9) FL-L; 10) FAS-AP1. Lanes 12-20, P CR products for partial PS genes: 12) QHSS/NR/CYL/03; 13) FAS-C1; 14) FASC2; 15) FAS-C3; 16) FL-D; 17) FL-E; 18) FL-F; 19) FL-L; 20) FAS-AP1. 600 denot es the 600 bp band in the size marker.

PAGE 33

33 Figure 2-3. Selected ion chromotograms ( m/z 416) for QHSS/NR/CYL/03 (A), FAS-AP1 (B), and FAS-C1 (C). The characteristic peak for CYN (*) appears at a retention time of 16.8 min for the two strains shown to produce the toxin.

PAGE 34

34 Figure 2-3. Continued.

PAGE 35

35 CHAPTER 3 AN IMPROVED METHOD FOR THE ISOL ATION OF INHIBITOR-FREE CYANOBACTERIAL DNA FROM ENVIRONMENTAL SAMPLES Introduction Molecular approaches have greatly expa nded understanding of the diversity of m icroorganisms in freshwater and marine envi ronments (Beja 2004; Tringe and Rubin 2005). Given that the majority of these microorga nisms cannot be cultured in the laboratory, environmental DNA isolation and Polymerase Chai n Reaction (PCR) of samples from the natural environment, along with cloning and gene seque ncing, provide a powerful tool for identifying novel microorganisms and their metabolites (Amann et al. 1995). Cyanobacteria are a large, diverse and important group of organisms in freshwater environments. They commonly reach very high cell densities and form blooms in many freshwater ecosystems (Phlips 2002). Species within the group have various competitive advantages over other algae, such as the ability to fix atmospheric nitrogen, the ability to regulate buoyancy in the water column, and production of toxi ns. The ability to produce toxins also make certain species of cyanobacteria a concern for human and animal welfare (Haider et al. 2003). Isolation of individual species of cyanobacteria has been an effective method for the identification of toxins produced by certain key sp ecies (Sivonen et al. 1990; Harada et al. 1994). Cultures of isolated cyanobacteria have also been used to identify genes responsible for the biosynthesis of toxins (Tille tt et al. 2000; Schembri et al 2001). Degenerate or specific primers designed from these gene sequences have been used to search for toxin genes in environmental samples (Schembri et al. 2001; Hisbergues et al. 2003; Y lmaz et al. 2007). This approach has been used to dete rmine the species responsible for toxin production in natural water samples, either by cloning, sequenc ing strategies or RFLP methods (Hisbergues et al. 2003). All these methods require so-called inhibitor-free DNA, because these molecular methods are

PAGE 36

36 easily inhibited by environmental contaminants (Pan et al. 2002). The presence of contaminants also poses serious barriers to the quantification of environm ental DNA (Zipper et al. 2003). Cyanobacterial DNA exists free in the cel lular cytoplasm surr ounded by a multilayered envelope, demonstrating both Gram-positive and Gram-negative features. Some filamentous cyanobacteria possess S-layers, fibrillar structures and sheaths outside of the outer membrane, making it difficult to lyse cells for extrac tion of DNA (Hoiczyk and Baumeister 1995). Methods exist for DNA isolation from culture d cyanobacteria. Not every method for DNA extraction performs equally well for different cyanobacterial strains due to variations in cell wall characteristics, cell size, and the presence or absence of sheat hs. Modifications of basic methods, however, can be appl ied to yield good quality geno mic DNA from cyanobacterial isolates cultured under laboratory conditions (Mak and Ho 1992; Fiore et al. 2000; Tillett and Neilan 2000). Problems, however, can arise when cyanobacteria are harves ted directly from the natural environment. Contaminants such as hum ic (HA) and fulvic acids are often co-purified along with DNA from environmental samples. Th ese substances inhibit PCR and restriction digestion of the DNA (Zhou et al. 1996; Jackson et al. 1997; Pan et al. 2002). Current methods of cyanobacterial DNA isolation do not effectively el iminate these inhibitory substances. Bovine serum albumin or skim milk has been used to ci rcumvent the effects of inhibitors (Pan et al. 2002). Chelex-based kits have also been applied successfully in some DNA isolations (Wilson et al. 2000). Both approaches, however have yielded only partial success. Similar inhibitors of PCR and restriction di gestion of DNA are commonly encountered in DNA isolation from bacteria in so il samples. Soils rich in or ganic content often contain high levels of humic and fulvic acids. Many methods exist for eliminating inhibitors from soil DNA, however, no single method is used universally (S chneegurt et al. 2003). In most cases, after

PAGE 37

37 isolation, DNA is purified using chromatography columns, agarose gel el ectrophoresis, or glass milk to remove humics and othe r inhibitors, th ereby increasing the cost and time needed to obtain good quality DNA and reducing ultimate DNA yi elds (Miller et al. 1999; England et al. 2001). Humic substances are also very abundant in the freshwater and marine environments, constituting a significant portion of the colo red dissolved organic matter (CDOM) in these systems. In freshwater systems, humic substanc es originate from decay of algae, bacteria and plants, including soil leaching and surface runoff fr om terrestrial environments (Anesio et al. 2005). During a study on the distribution of toxic cyanobacteria in lakes in Florida (USA), I encountered problems in isola ting inhibitor-free DNA from thes e freshwater ecosystems. Several approaches including commercial kits and published methods did not yield DNA suitable for PCR. Building upon published methods on soil DNA extraction, and modifying the xanthogenate nucleic acid isolation method of Tillett and Neilan (2000), I developed an improved method for removing inhibitory substan ces from DNA of environmental cyanobacteria samples collected from natural environments characterized by hi gh amounts of colored dissolved organic matter (CDOM). Because a large number of lakes, rivers and estuaries throughout the world contain high levels of CDOM, this method will provide a valuable new tool for expanded application of molecular genetics to limnological and ocea nographic research. Materials and Methods Cyanobacterial Strains Used and Collection of Lake Samples Microcystis aeruginosa PCC 7806 a nd Aphanizomenon ovalisporum FAS-AP1 were grown in BG-11 medium (Stani er et al. 1971) at 25 C. Light was provided on a 12:12hr light:dark photoperiod at 30 mol photons m-2 s-1 from cool white fluorescent lamps. Cells in

PAGE 38

38 exponential growth phase were harvested by centrifugation at 13,000 rpm for 10 minutes. Wetweight of cells was recorded, and tubes were kept frozen at -20 C until genomic DNA isolation. Natural water samples were collected from a suite of Florida lakes (Table 3-1). Water samples were collected using a vertical sampli ng pole and filtered onto 5 or 10 m nylon filters depending on the phytoplankton species present and algal cell density of the sample. Microorganisms were scraped off filters and susp ended in distilled water by pipetting. Equal amounts of cells were distribu ted in pre-weighed microcentr ifuge tubes, and tubes were centrifuged for 5-7 minutes at 13,000 rpm. Supernatants were removed, tubes were weighed and samples were kept frozen at -20oC until DNA isolation. In the case of cells with gas vacuoles, these were disrupted by pressure in a syringe before centrifugation. Genomic DNA Isolations Genom ic DNA was isolated from laborator y cultures and lake samples using the xanthogenate nucleic acid isol ation method (XS-isop method) (T illett and Neilan 2000) with modifications introduced in this study. In addition, Instagene matrix (Bio-Rad, CA, USA) was tested for its ability to remove PCR inhibitors. Instagene matrix isolations were performed according to Neilan (2002). Approximately 10-20 mg wet-weight of cells was used for each DNA isolation. In the standard XS-isop method, cells were suspended in 50 l TE buffer at pH 7.4 (10 mM Tris-HCl, pH 7.4; 1 mM ED TA, pH 8). Then, 750 l of XS buffer (1% potassium ethyl xanthogenate; 100 mM Tris-HCl, pH 7.4; 20 mM EDTA, pH 8; 800 mM ammonium acetate; and 1% SDS) were added to the suspension. Tubes we re incubated for 2 h at 70 C. After incubation, the tubes were vortexed for 30 s and kept on ic e for 30 min. Following centrifugation for 10 min at 13,000 rpm, supernatants were transferred to cl ean micro-centrifuge tubes, and 0.7 volumes of isopropanol were added. Tubes were then incu bated at room temperature for 10 min. After

PAGE 39

39 centrifugation at 13,000 rpm for 10 min, pellets were washed in 70% ethanol, air dried and suspended in double distilled and autoclaved water. New modifications to the aforementioned meth od were: (1) DNA was precipitated with 7% final concentration of polyethylen e glycol 8000 (PEG 8000) with 10mM MgCl2 (XS-PEG) instead of 0.7 volumes of isopropanol. (2) 3% polyvinylpolypyrrolidone (PVPP) was included in the XS buffer, and DNA was precipitated with either 0.7 volumes of isopropanol (XS-3%PVPPisop) or 7% PEG 8000 with 10 mM MgCl2 (XS-3%PVPP-PEG). (3) Cells were suspended in 50 l of TE buffer at pH 7.4. Before addition of XS buffer with 3% PVPP, cells were prewashed with 1 ml of TE buffer pH 7.4 containing 3% PV PP by vortexing at high speed for 30 s. Two l of 3 M CaCl2 was added to precipitate PVPP, followed by another 30 s vortexing. After three minutes on the bench, supernatants were transfe rred to clean micro-centrifuge tubes and cells were pelleted by centrifugation at 13,000 rpm for 5 minutes (Wechter et al. 2003). Cells were resuspended in 50 l of TE buffer at pH 7.4, a nd 750 l of XS buffer with 3% PVPP was added. DNA was precipitated with either 0.7 volumes of isopropanol (prewashed -isop) or 7% PEG 8000 with 10 mM MgCl2 (prewashed-PEG). For all methods, DNA pellets were washed with 70% ethanol (twice if PEG was used), air-dried, and suspended in double distilled and autoclaved water. DNA Quantification DNA was quantified by three m ethods: (1) Spectrophotometric measurements were performed on 2 l aliquots of processed sa mples using a Nanodrop ND-1000. When calculating DNA concentrations, 340 nm absorbance values we re used to normalize 260 nm readings with this instrument. In trials, is opropanol and PEG methods resulted in significant differences in 340-nm values. Absorbance values around 340 nm can be a consequence of humic substances in the samples. For this reason I used 260-nm r eadings without 340-nm normalization to calculate

PAGE 40

40 DNA concentrations. (2) Flouromet ric measurements were performed using 5-10 l aliquots of processed genomic DNA solution with a Quant-iT dsDNA HS assay kit and a Qubit flourometer (Invitrogen, CA, USA), in accordance with the manufacturers instructions. (3) DNA concentrations were also determined using elect rophoretic methods. Ten l aliquots of genomic DNA samples were run on 0.8% agarose gels along with 0.5 g -HindIII DNA fragments (Invitrogen, CA, USA). After ethidium bromid e staining, gels were visualized under UVtransillumination. Band intens ities of genomic DNAs were compared to those of the -HindIII fragments using the ImageJ software (http://r sb.info.nih.gov/ij/). Samples from Lakes Dora (06/26/2007) and Griffin (05/30/2007) and the A. ovalisporum FAS-AP1 culture were used for DNA yield comparisons. Duncans multiple range test (p=0.05) was used to compare mean values obtained using different methods. Polymerase Chain Reaction To test effectiveness of the new DNA isol ation m ethods in removing PCR inhibitors, genomic DNAs isolated with different met hods were amplified for a 1470 bp region of 16S rDNA, 685 bp region of the phycocyanin intergenic spacer region (PC-IGS), and approximately 300 bp of one of the condensation domains of the mcyA gene within the mi crocystin synthetase operon. Primers used in the study are listed in Table 3-2. 16S and PC-IGS reactions contained 2l of genomic DNA, 20 pmol of each primer (MWG-Biotech Inc., NC, USA), 200 M of each deoxynucleoside triphosphates (Fisher Scientif ic Company L.L.C, PA, USA), 2.5 mM MgCl2, 10 l of 5X green buffer, and 2 units of GoTaq DNA polymerase (Promega, WI, USA) in a total volume of 50 l. Reaction components for mcyA PCR were similar except that 300 M of each deoxynucleoside triphosphate was added, and the MgCl2 concentration was increased to 4.5 mM (Hisbergues et al. 2003). Amp lification for PC-IGS and mcyA started with denaturation of the genomic DNA at 95 C for 3 min, followed by 30 cycles of 95 C for 30 s, annealing at 55 C for

PAGE 41

41 30 s, extension at 72 C for 1 min, and a final ex tension step at 72 C for 5 min. Amplification conditions for 16S rDNA were similar except that annealing temperature was increased to 60 C. PCR products were pooled, normalized and run on a 2% agarose gel in TBE buffer (89 mM Tris, 2 mM EDTA, 89 mM Boric Acid), stained with ethidium brom ide and visualized with UV transillumination. Restriction Endonuclease Digestion of Genomic DNA Geno mic DNAs from Microcystis aeruginosa PCC 7806 laboratory cultures and lake samples were isolated using XS-isop, XS-PEG, or XS-3%PVPP-PEG methods Each restriction reaction contained 1 U of the restriction endonucl ease BmrI (New England Biolabs, Inc), 5 l of buffer, and 40 l of genomic DNA solution (80400 ng DNA), in a total volume of 50 l. Samples were incubated overnight at 37 C. Negativ e controls used the same amount of template without the enzyme BmrI. Digests were run on a 0.8% agarose gel in TBE buffer, stained with ethidium bromide and visualized with UV transillumination. Results DNA Yield with Different Measurement Methods For alm ost all cases, the spectrophotometric (Nanodrop) quantification method yielded more genomic DNA than flourometric (Qubit) or ge l densitometry (Image J) methods. In ten of twelve comparisons, flourometry and gel densito metry values were not significantly different from each other (Table 3-3). For lake samples, in general, there was no significant difference between XS-isop and XSPEG methods in terms of DNA yield obtained using the flourometric and gel densitometry quantification methods. Only the Lake Griffin sample showed a statistically sign ificant decrease in flourometry between the mentioned isolati on methods. There were significant differences between the two isolation methods when the spectrophotometric method was used.

PAGE 42

42 DNA yield was similar between the XS-isop and XS-3%PVPP-isop methods for the lake samples, except for the Lake Griffin sample with fluorometry and Lake Dora sample with spectrophotometry. Only the Lake Griffin sample with the sp ectrophotometric method, and the Lake Dora sample with the fluorometric method yielded signif icantly different DNA concentrations when XS-PEG and XS-3%PVPP-isop methods were compared. In general, the XS-3%PVPP-PEG method yi elded the lowest DNA concentrations. Concentrations with this method were significantly lower when compared to those of the XSisop method. The XS-3%PVPP-PEG method yielde d a similar concentration only in the gel densitometry for the Lake Dora sample. For this sample, there were no significant differences in gel densitometry concentrations regardless of the isolation methods used. XS-3%PVPP-PEG values were significantly diffe rent from XS-PEG and XS-3%P VPP-isop methods in seven of twelve comparisons. The situation was different with the A. ovalisporum FAS-AP1 culture. All isolation methods yielded significantly different concentrations when spectrophotometry was used. With flourometry and gel densitometry, XS-isop and XS-P EG yields were significantly different most of the time, while there was no significant di fference between XS-isop and XS-3%PVPP-isop. XS-PEG and XS-3%PVPP-isop methods yielded si gnificantly different concentrations for all measurement methods. For flourometry, the yiel d was significantly different between XS-PEG and XS-3%PVPP-PEG, while no difference was obs erved when yield was analyzed with gel densitometry. XS-3%PVPP-PEG concentrations were significantly different from XS-PEG and XS-3%PVPP-isop in five of six comparisons. Isopropanol versus PEG Precipitation Cells collected fro m Lake George (07/10/2007) were used in DNA isolations to determine effects of isopropanol versus PEG 8000 preci pitation of genomic DNA in obtaining PCR

PAGE 43

43 amplifiable templates,. The samples from La ke George were char acterized by high CDOM levels (>50 platinum cobalt units ). Amplification of 16S rD NA, PC-IGS and mcyA genes was inhibited when genomic DNA was precipitated with isopropanol whether in regular XS buffer or XS buffer with 3% PVPP (Figure 3-1). Isopropanol precipitation was only successful when cells were prewashed with PVPP and DNA isolated in XS buffer with 3% PVPP (data not shown). PEG 8000 precipitated genomic DNA was amplified for all three genes. Instagene matrix was not effective in removing PCR inhibitors from this sample. Aphanizomenon ovalisporum FASAP1 and Microcystis aeruginosa PCC 7806 were used as positive controls for each DNA isolation method. Expected amplifications from each strain were obtained from all isolation methods. A. ovalisporum did not have any mcyA bands, in agreement with microcystin ELISA results that were negative for this strain (d ata not shown). Instea d, this strain produced cylindrospermopsin. The effect of the XS-PEG method in removing P CR inhibitors was also tested for six other lake samples for the mcyA gene (Figure 3-2). Lake samples used in these analyses exhibited high CDOM levels (i.e. 50 platinum cobalt units ). While all XS-isop preparations were inhibited, the XS-PEG method provided DNA that was amplified for all lakes. Restriction Digestion of Genomic DNA Sa mples from Lake Griffin (06/26/2007), Crescent Lake (07/10/2007), Buffalo Bluff (07/10/2007), and Lake George (08/09/2007) were used in genomic DNA isolations with XSisop, XS-PEG and XS-3%PVPP-PEG methods. Microcystis aeruginosa PCC 7806 genomic DNA was used as the positive cont rol, and its DNA was cut with Bm rI for all isolation methods. The restriction enzyme was inhibi ted with XS-isop preparations fo r all lake samples, whereas all samples from XS-PEG and XS-3%PVPP-PEG prepara tions were digested with the enzyme used.

PAGE 44

44 A gel sample is shown in Figure 3-3 for Crescent Lake, Buffalo Bluff and M. aeruginosa PCC 7806 comparing activity of the restriction enzyme on XS-isop and XS-PEG isolations. Discussion The proposed new m ethods provided DNA that is easily amplified in PCR, and cut with restriction enzymes. Additional PCR additives such as BSA were not necessary. Obtained DNA was of high molecular weight, and yield was similar to that of the original protocol (XS-isop). XS-isop protocol was previously shown to yield high quality genom ic DNA not only from cyanobacteria, but from Gram-positive, Gram-n egative bacteria and the Archea. The new improved methods should be applicable for a wide range of microorganisms in aquatic environments, facilitating removal of inhibitors. This has important implications in an era when non-culture methods of microbial community analyses have become very important. My initial intention was to incorporate PVPP in the extraction solution. PVPP has been used to remove humics from soil DNA isolations eith er as part of the extraction buffer or as part of a purification step after DNA isolation (Zhou et al. 1996; Ar beli and Fuentes 2007). PVPP did not cause any negative effects on the XS buffe r, and a range of PVPP concentrations in the buffer was tested in terms of recovery of DNA and prevention of inhibition of PCR (data not shown). Inclusion of PVPP alone (1-5%) did not remove PCR inhibitors from lake samples. Wechter et al. (2003) reported pre-washing of soil samples in potassium phosphate buffer containing PVPP, followed by another PVPP additi on before the protein precipitation step. When I incorporated pre-washing of cells in TE buffer containing 3% PVPP, followed by isolation of DNA in XS buffer with 3% PVPP and isopropanol precipitation, I was able to obtain PCR amplifiable templates from all tested lake samples. DNA yield with pre-washing was comparable to that from XS-3%PVPP-isop for cultu res. There was, however, a significant loss

PAGE 45

45 in yield when lake samples were used. This was probably due to fr eezing of cells before isolation, and some cells and DNA were lost during pre-washing (data not shown). Polyethylene glycol (PEG) precip itation of DNA has been reported in the literature mainly for size fractionation of DNA. Si ze of DNA precipitated by PEG is dependent on concentration of PEG used, and higher molecular weight DNA is recovered by lower PEG concentrations (< 10% PEG ) (Lis and Schleif 1975; Paithankar and Prasad 1991). Lis and Sc hleif (1975) reported that 15% PEG 6000 precipitated DNA fragments down to 100 bp, while fragments smaller than 700 bp were not precipitated when 7% final concen tration of PEG was used. More recently, use of PEG has appeared in several papers as an approach to remove humic contaminants from soil DNA crude extracts. A four fold reduction in humic substances is observed when compost DNA is precipitated with 10% PEG 8000 and 1.1 M Na Cl, compared with isopropanol precipitation (LaMontagne et al. 2002). Arbeli and Fuen tes (2007) obtained 48.5-136.8-fold lower humic contaminants compared to isopropanol precipitation when genomic DNA was precipitated with 5% PEG 8000 in the presence of 0.6 M NaCl with no difference in DNA yield. In my case, replacement of isopropanol prec ipitation of DNA with PEG 8000 in the XS buffer (XS-PEG), resulted in PCR amplifiable template, without ne ed for addition of PVPP or prewashing. These DNA preparations were also free of DNases and were cut with the restriction enzyme BmrI. Initially I used 5% PEG in the XS buffer, however this resulted in signifi cant loss in DNA yield. With 7% PEG, yield of DNA was similar to isop ropanol precipitation (T able 3-3). If the inhibitors in my samples are humic substances, I may have prevented th e precipitation of lower molecular weight humics by using 7% PEG in the extraction buffer. In general, aquatic humic substances are smaller than terrestrial humic substances in terms of molecular weight (MW). An MW range of 0.8-1.5 kDA has been reported for Su wannee stream humics (Beckett et al. 1987).

PAGE 46

46 For two rivers in Florida, dominant molecular masses of humics were determined to be around 2 Kda, < 4.5% being in the higher molecular weight fraction (13 Kda) (Zanardi-Lamardo et al. 2002). The MW distribution of humic substances if any, in my lake samples is not known. However the color of the DNA pellets after XS-isop isolation from all lake samples was brownish suggesting presence of humics. This color decreased gradually in the order XS-PEG, XS-3%PVPP-isop, XS-3%PVPP-PEG, prewashed-isop and pre-washed-PEG. The effect of PEG compared to isopropanol precipitation can also be seen in Figure 3-3, where fragments smaller than 564 bp are absent in XS-PEG DNA templates. Among the three quantification methods tested on lake samples and A. ovalisporum culture, spectrophotometric measurements yielded greater DNA values than flourometric and gel analyses measurements. Spectrophotometry overestimates DNA concentrations, especially from environmental samples. Zipper et al.,(2003) re ported 10 to 30 times higher DNA concentrations compared to flourometric methods, when HA is pr esent in samples. HA also absorbs in the UV range, thereby complicating DNA quantification. Different HA preparations differ in their absorbances in the UV region, and it is not possi ble to correct for HA absorbance to quantify DNA with UV-VIS spectrophotometry (Zipper et al. 2003). For th is reason, flourometry and gel densitometry are commonly preferred over spectr ophotometry in soil DNA isolations (Zhou et al. 1996; LaMontagne et al. 2002; Howele r et al. 2003). Although flourometry or gel densitometry is more reliable than spectr ophotometry, binding of humic substances to fluorescent dyes has also been repo rted (Zipper et al. 2003). I have obtained values 10 times higher with spectrophotometry than flourometry and gel analyses. In general, the latter two methods agreed well for DNA values, and I suggest using either of them for both environmental samples and cyanobacteria cultures. I did not a ttempt to quantify humic acids in my samples

PAGE 47

47 simply by using specific wavelengths for the reasons mentioned above. Instead, inhibition was tested by PCR and restriction assays. For environmental samples, DNA yields using XS-isop and XS-PEG were similar and the improved method can be used safely for these sa mples without loss in DNA yield. It is likely that others may encounter more difficult samples in terms of preventing the inhibition of PCR. If the XS-PEG method is not sufficient, I sugge st pre-washing the cells with 3%PVPP and precipitation with PEG if DNA yield is not the major concern. It is advisa ble to wash cells just after harvest and before freezing so th at loss of cells and DNA is minimized.

PAGE 48

48 Table 3-1. Lake samples used, collection dates and locations. Source Collection date Location Harris Chain of Lakes St. Johns River System Lake Dora Lake Griffin Buffalo Bluff Crescent Lake Lake George 05/30/2007 06/26/2007 05/30/2007 06/26/2007 07/10/2007 07/10/2007 08/09/2007 07/10/2007 08/09/2007 Lake County, Florida Lake County, Florida Palatka, Florida Putnam County, Florida Volusia County, Florida Table 3-2. Oligonucleotide primers used in the study. Name Sequence (5-3) Reference 16S rRNA 27F1 1516R PC-IGS PC F PC R mcyA mcyA-Cd1F mcyA-Cd1R AGAGTTTGATCCTGGCTCAG ATCCAGCCACACCTTCCGG GGCTGCTTGTTTACGCGACA CCAGTACCACCAGCAACTAA AAAATTAAAAGCCGTATCAAA AAAAGTGTTTTATTAGCGGCTCAT (Neilan et al. 1997) (Y lmaz et al. 2007) (Neilan et al. 1995) (Neilan et al. 1995) (Hisbergues et al. 2003) (Hisbergues et al. 2003)

PAGE 49

49 Table 3-3. Total genomic DNA yield (g) from two lake samples and A. ovalisporum FAS-AP1 culture with different methods of isolation and DNA quantification*. Sample Source Quantification Method XS-isop DNA XS-PEG Isolation XS-3%PVPPisop Methods XS-3%PVPP -PEG Lake Griffin Lake Dora A.ovalisporum Nanodrop Qubit Image J Nanodrop Qubit Image J Nanodrop Qubit Image J 5.65.65 (A) (a) 0.63.02 (A) (b) 0.82.10 (A) (b) 2.55.54 (A) (a) 0.26.07 (AB) (b) 0.23.08 (A) (b) 5.00.82 (A) (a) 1.17.07 (A) (b) 0.84.07 (A) (b) 1.72.20 (B) (a) 0.50.06 (B) (b) 0.86.14 (A) (c) 1.27.65 (BC) (a) 0.20.02 (B) (b) 0.26.03 (A) (b) 0.93.12 (B) (a) 0.36.01 (B) (b) 0.56.03 (B) (c) 4.30.02 (A) (a) 0.53.03 (B) (b) 0.86.07 (A) (b) 1.82.32 (B) (a) 0.28.04 (A) (b) 0.25.11 (A) (b) 3.13.36 (C) (a) 1.06.06 (A) (b) 0.87.02 (A) (b) 2.31.18 (B) (a) 0.33.01 (C) (b) 0.36.11 (B) (b) 0.89.04 (C) (a) 0.19.02 (B) (b) 0.22.02 (A) (b) 2.22.02 (D) (a) 0.53.12 (C) (b) 0.56.02 (B) (b) *Each extraction was performed in triplicate. Ea ch measurement was repeated twice, except for Qubit, which was conducted once. Sample means for each lake and culture were compared to other values using Duncans multiple range test. Means with the same letter are not significantly different from each other. Upper case letters are for comparisons among the four DNA isolation methods for each quantification method (rows). Lower case letters are for comparisons among the three different quantification methods for each DNA isolation method (columns).

PAGE 50

50 Figure 3-1. PCR results for the 16S rDNA (uppe r bands), PC-IGS (middle bands) and mcyA (lower bands) with genomic DNA from the Lake George sa mple (lanes 1-5) collected on 07/10/2007. A. ovalisporum FAS-AP1 (lanes 6-7) and M. aeruginosa PCC 7806 (lanes 8-9) were used as positive cont rols. Size marker (SM) is the 100 bp DNA ladder.

PAGE 51

51 Figure 3-2. mcyA PCR results for lake samples and two cultures isolated either with XS-isop (odd lanes) or XS-P EG (even lanes). M. aeruginosa PCC 7806 was the positive control, and A. ovalisporum FAS-AP1 was the negative control for the mcyA gene. Size marker (SM) is the 100 bp DNA ladder.

PAGE 52

52 Figure 3-3. Restriction digestion of genomic DNA with BmrI, isol ated either with XS-isop or XSPEG. Odd lanes are samples incubated with the enzyme. Even lanes are samples incubated without the enzyme. Lanes (1-4), Crescent Lake (0 7/10/2007); lanes (5-8) Buffalo Bluff (07/10/2007); lanes (9-12), Microcystis aeruginosa PCC 7806 as the positive control. Size marker (SM) is 0.5 g DNA Hind-III fragment.

PAGE 53

53 CHAPTER 4 RESTRICTION FRAGMENT LENGTH POLYMORPHIS M (RFLP) ANALYSES OF CYLINDROSPERMOPSIN (CYN) BIOSYNT HESIS GENES IN FLORIDA LAKES Introduction Cyanobacteria are a m orphologically diverse group of microorganisms capable of plantlike photosynthesis. They contain chlorophyll-a in addition to other accessory pigments such as phycocyanin and phycoerythrin (Bryant and Frigaar d 2006). They are considered one of the oldest life forms on earth, and have adaptations to survive in a variety of environments from lakes to oceans, and from deserts to polar regions (Eldridge and Gr eene 1994; Paerl et al. 2000). Under appropriate environmental conditions cyanobacteria can reach very high cell densities in lakes and oceans (Phlips 2002). They are a concern in aquatic ecosystems around the world due to the ability of some bloom -forming species to produce toxic secondary metabolites (Haider et al. 2003). Although the function of these t oxins is a matter of debate, thesy can have adverse affects on the function of the ecosystem, including human resources (Kaebernick and Neilan 2001). Concerns regard ing the effect on humans are related to the drinking water supply, recreati onal activities, a nd food supply (Haider et al. 2003). Many cyanobacterial toxins are either hepato toxic (e.g., microcystins, cylindrospermopsin) or neurotoxic (e.g., saxitoxins, anatoxin-a) (Kaebernick and Ne ilan 2001). Some toxins are produced by a variety of cyanobacteria species, and some species can pr oduce several types of toxins. The ability to produce toxins depends on th e presence of toxin biosynthesis genes, most commonly polyketide synthases, peptide synthetases or a combination of both (Tillett et al. 2000; Schembri et al. 2001). For some species such as Microcystis aeruginosa the gene clusters responsible for the biosynthesis of microcys tins (MC) have been id entified and sequenced (Dittmann et al. 2001). In the case of cylindr ospermopsin (CYN), biosynthesis genes from Cylindrospermopsis raciborskii and Aphanizomenon ovalisporum have been identified and

PAGE 54

54 partially sequenced (Schembri et al. 2001; Shalev -Alon et al. 2002). The av ailability of genetic information has opened new areas of research, in cluding: (1) regulation of toxin biosynthesis (Kaebernick et al. 2000; Dittmann et al. 2001; Kaebernick et al. 2002), (2) determination of toxic species from isolated cultures (Li et al. 2001) (3) rapid screen ing of lakes for toxic cyanobacteria (Pan et al. 2002), and (4) determin ation of species respons ible for toxin production in mixed phytoplankton populations (Hisbergues et al. 2003). As described in Chapter 2, I have identified a CYN-producing Aphanizomenon ovalisporum from a freshwater pond in Florida usi ng Polymerase Chain Reaction (PCR) for the putative CYN biosynthesis genes, namely polyketide synthases ( pks ) and peptide sythetases ( ps ) (Schembri et al. 2001; Kellmann et al. 2006). A. ovalisporum FAS-AP1 PKS sequences were 99.8% and 100% similar to prev iously published sequences from A. bergii and A. ovalisporum Similarity to PKS sequences from C. raciborskii was also high (i.e., over 99%). Identities of PS sequences from A. ovalisporum FAS-AP1 to PS sequences of A. bergii and A. ovalisporum were 99.4% and 100%, respectively. Contrary to PK S sequences, PS sequence identity between A. ovalisporum FAS-AP1 and C. raciborskii strains was lower (~ 96.4%). Amplification of toxin genes from environmen tal samples has previously been performed for MC genes. These studies formed a model to screen lakes for CYN biosynthesis genes, followed by RFLP analyses to determine the producer organisms. MC can be produced by several different cya nobacteria including Microcystis Anabaena and Planktothrix (Haider et al. 2003). Measuring MC concentratio ns with a biochemical method alone, however, does not give any information regarding the pr oducer(s) of the toxin. One wa y to find the producers is to amplify the toxin genes by PCR from an envir onmental sample, purify the fragments and clone them, followed by sequencing of the fragments (Tringe and Rubin 2005). Obtained sequences

PAGE 55

55 will reveal the variety of toxin producers, or, if there is sufficient sequence information from isolated cultures, will identify the toxin-producer organisms in that sample. The other method to differentiate sequences of toxin-producer organisms again uses purified toxin gene fragments. These are then restricted with a variety of restriction enzymes. Different sequences are cut with different enzymes or from differe nt sites. The resulting so-cal led restricted fragments are visualized on an agarose gel and compared with fragments from known toxin-producer isolates (Hisbergues et al. 2003). This method is called Restriction Fragment Length Polymorphism (RFLP). Gene sequence information of toxin genes was used previously to detect MC genes from lakes in Germany and China (Pan et al. 20 02; Hisbergues et al. 2003). By taking advantage of differences in sequences of MC genes from different cyanobacteria, Hisbergues et al. (2003) purified amplified MC fragments from lake sample s and digested them with different restriction enzymes. In this way, they were able to iden tify the source of MC in the investigated lakes. In this chapter I use sequence information from A. ovalisporum FAS-AP1, along with previously published sequences from other CYN producing cyanobacteria, to identify CYN producers in Florida lakes. C. raciborskii is a major bloom-forming species in Florida and has long been assumed to be the source of CYN in lakes (Chapman and Sc helske 1997). In Chapter 2, I demonstrated that none of the eight Florida isolates of this specie s obtained in this study pr oduced this toxin. Using the newly established DNA isolation method from environmental samples described in chapter 3, I undertook screening of lake samples in Flor ida for CYN biosynthesis genes to determine producer species by RFLP methods.

PAGE 56

56 Materials and Methods Control Cyanobacterial Strains Cyanobacterial s trains A. ovalisporum FAS-AP1, C. raciborskii strains QHSS/NR/CYL03 and FAS-C1 were grown in BG11 medium (Stanier et al. 1971) at 25 C. Light was provided on a 12:12hr light:dark photope riod, at 30 mol photons m-2 s-1 from cool white fluorescent lamps. Cells in exponential phase of growth we re collected by centrifugation at 13,000 rpm for 5 minutes and frozen at -20 C until DNA isolation. Lake Samples Lake sam ples were collected using a vertic al sampling pole. After transport to the laboratory, they were filtered onto 5-10 m nylon filters, scraped off the filters and suspended in double distilled water in 1.5 ml centrifuge tubes. Tubes were centrifuged for 5-7 minutes at 13,000 rpm. Supernatants were removed, and cells were frozen at -20 C until DNA isolation. A list of lake samples used in this study is provided in Table 4.1. C YN concentrations were determined using a CYN ELISA kit (Abraxis, LLC, PA, USA) according to manufacturers instructions. Genomic DNA Isolation Genom ic DNA from lake water samples was isolated according to the XS-PEG method described in chapter 3. Approximately 10-20 mg we t weight of cells was suspended in 50 l of TE buffer, to which 750 l of freshly prepared XS buffer was adde d. Tubes were incubated at 70 C for 2 h. Following incubation on ice for 30 min to precipitate proteins, tubes were centrifuged for 10 min at 13,000 rpm. 750 l of the supernatan ts were transferred to clean microcentrifuge tubes and 187.5 l of PEG stock so lution (35% PEG with 50 mM MgCl2) was added to each tube. Tubes were mixed by inverting several ti mes and incubated at room temperature for 10 min, followed by 10 min centrifugation at 13,000 rpm. Supernatants were removed and the DNA

PAGE 57

57 pellet washed twice with 70% Ethanol. Wash ed DNA was pelletted by ce ntrifugation for 5 min at 13,000 rpm and air dried. DNA was dissolved eith er in double distilled autoclaved water or TE buffer. Polymerase Chain Reaction (PCR) and In ternal Control Fragment (ICF) Design A 624 bp region of the PKS gene and 597 bp re gion of the PS gene were am plified from lake samples and isolated cultures on a Biometra thermal cycler (Biometra GmbH i. L., Goettingen, Germany) according to Schembri et al. (2001). Each PCR r eaction contained 10-20 ng of genomic DNA, 20 pmol of each primer (M4-5 and M13-14) (MWG-Biotech Inc., NC, USA), 200 M of each deoxynucleoside triphosphate s (Fisher Scientific Company L.L.C, PA, USA), 1.5 mM MgCl2, 10 l of 5X green buffer and 1.5 units of GoTaq DNA polymerase (Promega, WI, USA) in a total volume of 50 l Amplification conditions for both included denaturation of the template DNA at 95 C for 3 min, 30 cycles of 95 C for 30 s, annealing of the primers at 60 C for 30 s, extension at 72 C for 1 min and a final extension step at 72 C for 5 min. PCR products were run on a 1.5% agar ose gel along with a 100 bp DNA ladder, stained with ethidium bromide and visual ized under UV transillumination. Internal Control Fragments for PKS and PS genes were designed as described in Wilson et al. (2000) for the DNA-dependent RNA polymerase ge ne. They were designed so that the same primers used in toxin gene amplification were used to amplify ICFs. For the construction of PKS-ICF, a 33 bp primer PKS-INT (5 -TGGAATCCGGTAATTGGTCTCAGTTCATCAAGC ) was designed. A 20 bp fragment at the 3 end of PKS-INT matched exactly to the sequence 220 bp downstream of the primer M4. The 13 bp overhang at the 5 end of PKS-INT matched exactly with the fragment at the 3 end of M4. PKS-INT and M5 were used in the first PCR and the resulting product was used in the second PCR w ith the primers M4 and M5 to give an ICF of 404 bp. ICF for PS was constructed in a similar fashion. A 33 bp primer PS-INT (5-

PAGE 58

58 TGATAGCCACGAGCTTTTCTGAAAGAGGTTGGC) was designed to match a contiguous 19 bp fragment 206 bp downstream of primer M13. The 14 bp overhang at the 5 end of PS-INT matched exactly with a fragment at the 3end of M13. PS-INT and M14 were used in the first PCR and their product was used as a template in the second PCR with M13 and M14 to give an ICF of 391 bp. PCR conditions were the same as described for the PKS and PS amplifications described above. ICF fragments were added in PCR reactions during amplification of CYN genes to visualize the success of the PCR. PKS, PS and ICF fragments were purified from agarose gels with the Qiaquick MinElute ge l extraction kit (Qiagen, CA, USA), according to manufacturers instructions. Restriction Digestion of PS Fragments Approxim ately 150 ng of purified PS PCR product s were used in each restriction assay with 1.5 U of the enzymes BmgBI or NciI (New England Biolabs) in a total volume of 50 l with the buffers supplied with the enzymes. They were incubated overnight at 37 C. Phylogenetic Analysis Partial 16S rDNA seque nces were aligned us ing Clustal W within the Mega 4 package (Tamura et al. 2007). Sequence alignments were checked manually. Phylogenetic trees were constructed from pairwise distance matrix us ing neighbor-joining. Alignm ents were bootstrapped with 1000 resamplings. Results Determination of Optimum ICF Concentration Three different ICF concentrations (10, 20, 30 fg) were tested in PCR reactions for PKSICF and PS-ICF, with the sam e amount of template DNA from A. ovalisporum FAS-AP1. Both the gene fragment, and the ICF were amplified in all cases, with the expected length fragments (Figure 4.1). Ten fg was used as the am ount of ICF in subsequent PCR reactions.

PAGE 59

59 Screening of Lake Samples for CYN Biosynthesis Genes Fourteen samples obtained from five la kes in Hillsborough County were chosen for genetic screening based on the high abundances of C. raciborskii. Genomic DNAs obtained from these sites were subjected to amplifi cation for the PKS and PS genes along with the corresponding ICFs. ICFs were used to confirm the success of PCR reacti ons. As described in chapter 3, environmental DNA samples usually c ontain PCR inhibitors, but the inhibition was removed using the newly developed method of is olation. None of the 14 samples contained CYN biosynthesis genes. Results for the PKS gene are shown in Figure 4-2. Seven other samples from Lakes Dora, Griffin, and the St. Johns River system collected in April and May of 2007 were also screened for CYN biosynthesis genes, i.e. PKS and PS. All sites had high densities of C. raciborskii (~ 100,000 cells/ml) as obs erved by microscopy (Mary Cichra, personal communication). Detection of C YN genes was initially tested with ICFs prior to final PCR for purifaction. CYN genes were absent in samples from Lakes Dora and Griffin. However both PKS and PS genes were amplified in samples from the St. Johns River System (Figure 4-3). Despite presence of CYN genes, ELISA analysis did not detect measurable quantities of CYN. Samples from th e St. Johns River System also had A. ovalisporum in low numbers. Low concentrations of CYN have b een reported for lakes in Florida (Aubel et al. 2006). CYN concentrations in St. Johns Rive r samples were below detection limits, in agreement with the low numbers of A. ovalisporum present in the samples. Purification and RFLP Analyses of PS Fr agments from th e St. Johns River System Samples High sequence similarity be tween PKS sequences of A. ovalisporum and C. raciborskii Australia strain QHSS/NR/CYL/03 (> 99%) did not permit the use of restriction enzymes to differentiate between the two species. However, PS sequence similarity between the two species

PAGE 60

60 was lower (96.4%). Two restriction enzymes, na mely BmgBI and NciI, yielded different length fragments for the PS fragments from C. raciborskii and A. ovalisporum The C. raciborskii PS fragment was not cut with BmgBI, whereas th is enzyme yielded two fragments 310 bp and 287 bp, in A. ovalisporum FAS-AP1. PS fragments from both species were cut with the enzyme NciI. C. raciborskii yielded three fragments, 329 bp, 234 bp, and 51 bp, whereas A. ovalisporum yielded two fragments, 319 bp and 278 bp. These fragment sizes obtained from the sequence information for the two species were in agreem ent with the fragments visualized after gel electrophoresis (Figure 4-4). Amplified PS frag ments from Crescent Lake, Buffalo Bluff, and Lake George, in the St. Johns River ecosystem collected in May of 2007, were purified from agarose gels and restricted with the mentioned enzymes. The frag ment patterns of all three sites for both enzymes were the same as patterns obtained from A. ovalisporum but differed from C. raciborskii There are faint bands at 600 bp (Figur e 4-4), which was not cut by the enzyme BmgBI. This is probably due to incomplete di gestion by this enzyme. Cyanobacterial DNA is not cut by all restriction enzymes, and can sometime s be incompletely digested. It may, however, represent amplifications from a C. raciborskii PS fragment. This is unlikely since the same fragments were cut with the enzyme NciI, giving the same length fragments as A. ovalisporum Phylogenetic Relationship of A. ovalisporum to Other Cyanobacteria Sequence infor mation and genetic relatedness of A. ovalisporum to other strains of A. ovalisporum and A. bergii were described in Chapter 2. To illustrate its relationship in other cyanobacteria, partial 16S rDNA sequences obtaine d in this work and from published sequences have been used to construct the tree presented in Figure 4-5. The A. ovalisporum FAS-AP1 strain clustered together with other A. ovalisporum and A. bergii strains from around the world. Discrepancies in identification and naming of th ese two species were discussed in Chapter 2. Interestingly both species were separated from other Anabaena and Aphanizomenon strains. C.

PAGE 61

61 raciborskii Florida strain FAS-C1 and Australian strain QHSS/NR/C YL/03 clustered with other C. raciborskii strains from Florida and around the world forming a monophyletic group. Discussion C. raciborskii is one of the m ost common bloom-fo rming species of cyanobacteria in Florida lakes. Year-round blooms have been ob served in some lakes (Chapman and Schelske, 1997; Phlips et al, 2003). It has been considered a threat to th e health of the ecosystems in Florida due to its putative ability to produce C YN, as reported in other countries (Chorus and Bartram, 1999). There has, however, been no confirmation of CYN production from isolated species of this genus in Florida (Kellmann et al 2006). I showed in Chapter 2 that none of the eight Florida isolates of C. raciborskii produce this toxin. Instead, A. ovalisporum isolated from a pond in Duval County, Florida contained CYN bi osynthesis genes and pr oduced the toxin. To further investigate CYN producers in different lakes, I used PCR for the putative CYN biosynthesis genes on environmental samples. The results provided the first example of screening of lake water samples in Florida by molecular methods for the production of CYN. All of the Hillsborough County lakes, as well as Lakes Dora and Gri ffin, were negative for the presence of CYN genes despite high C. raciborskii biomass. However, CYN genes were detected in samples from the St. Johns River System, the first reported observation of these genes in the US. It was possible to discriminate between PS fragments in C. raciborskii and A. ovalisporum using two different restriction enzymes. Following purification of these fragments from lake samples and cultures, they were subj ected to RFLP analysis. Patterns obtained from lakes were the same as those from A. ovalisporum but different from C. raciborskii These results suggest that PS genes involved in CYN biosynthesis in Crescent Lake, Buffalo Bluff and Lake George are from A. ovalisporum There is a possibility th at a CYN-producing strain of C. raciborskii exists in lakes not included in this study. It is advisable to test with more restriction

PAGE 62

62 enzymes, preferably after a nested PCR. Cl oning and sequencing of these PS fragments would also reveal the variety of PS se quences in these samples, and such a test should be performed before arriving at definitive conclusions. The phylogenetic tree illustrated in Figure 4-5 shows the clustering of A. ovalisporum and A. bergii strains. As they are separated from other Anabaena and Aphanizomenon strains, a taxonomic revision of these two species is needed as discussed in Chapter 2. Along with the newly developed DNA isolation methods from lake water samples, a variety of sites can be screened, and the or ganisms responsible for toxin production can be identified, which is not possible by chemical methods or microscopy alone. This study provides the first example of such re search in Florida lakes.

PAGE 63

63 Table 4-1. Lake samples used in PCR amplifications for CYN genes. Source Collection Date Location Lake Brant Lake Carroll Egypt Lake Little Lake Wilson (LLW) Lake Wilson Lake Dora Lake Griffin Buffalo Bluff Crescent Lake Lake George 07/27/2005 10/13/2005 07/18/2006 08/15/2006 09/28/2005 07/18/2006 07/27/2005 07/26/2005 07/17/2006 08/14/2006 07/26/2005 09/27/2005 07/17/2006 08/14/2006 04/15/2007 05/30/2007 04/15/2007 05/30/2007 05/20/2007 05/20/2007 05/20/2007 Hillsborough County, Florida Hillsborough County, Florida Hillsborough County, Florida Hillsborough County, Florida Hillsborough County, Florida Lake County, Florida Lake County, Florida Palatka, Florida Putnam County, Florida Volusia County, Florida

PAGE 64

64 Figure 4-1. PCR results with di fferent concentrations of ICF (10, 20, 30 fg) for PKS (A) and PS (B) genes with template from A. ovalisporum Size marker (SM) is the 100 bp DNA ladder.

PAGE 65

65 Figure 4-2. PCR results for the amplificati on of PKS genes from Hillsborough County lake samples. None of the sites had PKS genes. 400 bp fragment is the ICF added into the PCR reactions for conformation of proper P CR function. Size marker (SM) is the 100 bp DNA ladder. See table 4-1 for lake locations and abbreviations.

PAGE 66

66 Figure 4-3. PCR results for the PKS (A) and PS (B) genes. Amplifications were obtained only from the St. Johns River Samples. A. ovalisporum and C. raciborskii were the positive controls. SM is the 100 bp DNA ladder.

PAGE 67

67 Figure 4-4. RFLP analyses on the PS fragment s obtained from the St. Johns River Samples, A. ovalisporum and C. raciborskii. SM is the 100 bp DNA ladder. 51 bp fragment from the NciI treated C. raciborskii PS fragment was not visible in the gel due to its low concentration.

PAGE 68

68 Figure 4-5. Neighbor-joining tr ee of selected cyanobacteria based on partial 16S rDNA sequences. Bootstrap values above 50% ar e shown. PSP, paralytic shellfish poisons, and CYN, cylindrospermopsin.

PAGE 69

69 CHAPTER 5 CONCLUSIONS A significant percentage of la kes in F lorida, USA are eutr ophic and subject to blooms of planktonic algae, including potenti ally toxic species. The threat that toxic algae pose for the integrity of aquatic ecosystems and human health in Florida has become a major issue for water managers. Although there has b een considerable research on phytoplankton in many eutrophic lakes by universities and government agencies, da ta on toxic algae have only recently been forthcoming (Phlips, 2003; Phlips et al. unpublished, Y lmaz et al. unpublished). Two species of cyanobacteria have been identified as th e major toxic algae threats in Florida, M. aeruginosa and C. raciborskii (Phlips, 2002; Phlips et al. 2003). The link between M. aeruginosa and the hepatotoxin microcystin has been cl early established. For example, a M. aeruginosa bloom that covered many square kilometers in the St. Lu cie Estuary was tested for the presence of microcystins, revealing up to 800 g/L microc ystin-LR equivalents (Phlips et al. unpublished), 800 times higher than the World Health Organizati ons provisional guideline value of 1 g/L in drinking water (Chorus and Bartram, 1999). A tw o year investigation of different systems resulted in a wide range of MC concentrations, generally lower than 20 g/L (Bigham et al. unpublished; Y lmaz et al. unpublished). Establishing the link between C. raciborskii and the toxin CYN has been less clear cut despite the widespread distributi on and intensity of blooms around Fl orida. This study helped to explain the basis for this confusion. The fact that all isolates of C. raciborskii from lakes and rivers in Florida lacked genes associated w ith CYN production suggests that the CYN observed in Florida lakes is produced by other species of cyanobacteria. Here, I reported on the isolation of a CYN-producing A. ovalisporum for the first time in the USA. The fact that A. ovalisporum is not a common bloom-formi ng species in Florida helps

PAGE 70

70 explain why very high concentrations of CYN ha ve not been observed in Florida (Aubel et al., 2006) despite the common occurrence of intense C. raciborskii blooms. In order to identify the variety of CYN producers in the freshwater sy stems of Florida by molecular methods, lake samples were collected during a period of two years from lakes in Hillsborough County, the Harris Chain of Lakes, and the St. Johns River System. Whole community genomic DNA isolated from these samples proved to be uns uitable for molecular analyses. The improved methods described in chapter 3 yielded DNA suita ble for many molecular biology applications. To my knowledge, this is the first report on removal of inhibitors from cyanobacterial DNA isolations in freshwater system s. These new methods should have broad application in aquatic molecular ecology because the modified DNA isol ation method was previously shown to yield high quality DNA from cyanobacteria, other Gram -negative and Grampositiv e bacteria, as well as the Archea (Tillet and Neilan, 2000). Chapter 4 of this dissertation illustrates th e usefulness of the established DNA isolation methods. As explained in that chapter, lakes with high C. raciborskii biomass were screened for the putative CYN biosynthesis genes, and only three contained orga nisms harboring these genes. RFLP analyses on the amplified gene fr agments suggested the newly isolated A. ovalisporum was the source of CYN genes in these system s. Cloning and sequencing of these gene fragments, however, are needed to ascertain the variety of CYN genes in the positive samples. In addition, isolation and screening of more cyanobacteria sp ecies are needed to understand the variety of toxin producers fully. This dissertation supports the usefulness of molecular methods in ecological research and describes novel methods that can be used to answ er a variety of questions in aquatic molecular ecology.

PAGE 71

71 LIST OF REFERENCES Aubel, M., D'Aiuto, P., Chapm an, A. D., Casama tta, D., Reich, A., Ketchen, S. & Williams, C. 2006. Blue-green algae in St. Johns River, FL. LakeLine 26:40-5. Amann, R. I., Ludwig, W. & Schleifer, K. H. 1995. Phylogenetic iden tification and in-situ detection of individual microbial-cells without Cultivation. Microbiol. Rev. 59:143-69. Anesio, A. M., Graneli, W., Aiken, G. R., Kieb er, D. J. & Mopper, K. 2005. Effect of humic substance photodegradation on bacterial growth and respiration in lake water. Appl. Environ. Microbiol. 71:6267-75. Arbeli, Z. & Fuentes, C. L. 2007. Improved pur ification and PCR amplification of DNA from environmental samples. FEMS Microbiol. Lett. 272:269-75. Banker, R., Carmeli, S., Hadas, O., Teltsch, B., Porat, R. & Sukenik, A. 1997. Identification of cylindrospermopsin in Aphanizomenon ovalisporum (Cyanophyceae) isolated from Lake Kinneret, Israel. J. Phycol. 33:613-6. Beckett, R., Jue, Z. & Giddings, J. C. 1987. Determ ination of molecular weight distributions of fulvic and humic acids using flow field-flow fractionation. Environ. Sci. Tech. 21:289-95. Beja, O. 2004. To BAC or not to BAC: marine ecogenomics. Curr. Opin. Biotech. 15:187-90. Bourke, A. T. C., Hawes, R. B., Neilson, A. & Stallman, N. D. 1983. An outbreak of hepatoenteritis (the Palm Island mystery diseas e) possibly caused by algal intoxication. Toxicon 21 (Suppl 3):45-8. Briand, J. F., Leboulanger, C., Humbert, J. F., Bernard, C. & Dufour, P. 2004. Cylindrospermopsis raciborskii (Cyanobacteria) invasion at mid-latitudes: Selection, wide physiological tolerance, or global warming? J. Phycol. 40:231-8. Bryant, D. A. & Frigaard, N. U. 2006. Prokaryo tic photosynthesis and phototrophy illuminated. Trends Microbiol. 14:488-96. Burns, B. P., Saker, M. L., Moffitt, M. C. & Neilan, B. A. 2004. Molecular detection of genes responsible for cyanobacterial toxin production in the genera Microcystis, Nodularia, and Cylindrospermopsis. Methods in molecular biology (Clifton, N.J 268:213-22. Canfield, D. E., Phlips, E. & Duarte, C. M. 1989. Factors influencing the abundance of bluegreen-algae in Florida Lakes. Can. J. Fish. Aquat. Sci. 46:1232-7. Chapman, A. D. & Schelske, C. L. 1997. Recent appearance of Cylindrospermopsis (Cyanobacteria) in five hype reutrophic Florida lakes. J. Phycol. 33:191-5.

PAGE 72

72 Chiswell, R. K., Shaw, G. R., Ea glesham, G., Smith, M. J., Norris, R. L., Seawright, A. A. & Moore, M. R. 1999. Stability of cylindrosperm opsin, the toxin from the cyanobacterium, Cylindrospermopsis raciborskii : Effect of pH, temper ature, and sunlight on decomposition. Environ. Toxicol. 14:155-61. Chorus, I. & Bartram, J. 1999. Toxic cyanobacteria in water: A guide to their public health consequences, monitoring, management. E & FN Spon London, 416 pp. Cichra, M. F., Badylak, S., Henderson, N., Rueter, B. H. & Phlips, E. J. 1995. Phytoplankton community structure in the open water zone of a shallow subtropical lake (Lake Okeechobee, FL, USA). Arch. Hydrobiol., Special Issues, Advances in Limnology 45:157-75. Dittmann, E., Erhard, M., Kaebernick, M., Scheler, C., Neilan, B. A., von Dohren, H. & Borner, T. 2001. Altered expression of two light-dep endent genes in a microcystin-lacking mutant of Microcystis aeruginosa PCC 7806. Microbiol. 147:3113-9. Dittmann, E., Neilan, B. A. & Borner, T. 2001. Molecular biology of peptide and polyketide biosynthesis in cyanobacteria. Appl. Microbiol. Biotech. 57:467-73. Eldridge, D. J. & Greene, R. S. B. 1994. Microbiotic soil crusts a review of their roles in soil and ecological processes in th e rangelands of Australia. Aust. J. Soil Res. 32:389-415. England, L. S., Trevors, J. T. & Holmes, S. B. 2001. Extraction and detection of baculoviral DNA from lake water, detritus and forest litter. J. Appl. Microbiol. 90:630-6. Falconer, I. R., Hardy, S. J., Hu mpage, A. R., Froscio, S. M., Tozer, G. J. & Hawkins, P. R. 1999. Hepatic and renal toxicity of the blue-green alga (cyanobacterium) Cylindrospermopsis raciborskii in male Swiss albino mice. Environ. Toxicol. 14:143-50. Fastner, J., Heinze, R., Humpage, A. R., Misc hke, U., Eaglesham, G. K. & Chorus, I. 2003. Cylindrospermopsin occurrence in two Germ an lakes and preliminary assessment of toxicity and toxi n production of Cylindrospermopsis raciborskii (Cyanobacteria) isolates. Toxicon 42:313-21. Fiore, M. F., Moon, D. H., Tsai, S. M., Lee, H. & Trevors, J. T. 2000. Miniprep DNA isolation from unicellular and filamentous cyanobacteria. J. Microbiol. Methods. 39:159-69. Fischer, W. J., Hitzfeld, B. C., Tencalla, F., Eriksson, J. E., Mikhailov, A. & Dietrich, D. R. 2000. Microcystin-LR toxicodynamics, induced pathology, and immunohistochemical localization in livers of blue-gr een algae exposed rainbow trout ( Oncorhynchus mykiss ). Toxicol. Sci. 54:365-73. Froscio, S. M., Humpage, A. R ., Burcham, P. C. & Falconer, I. R. 2003. Cylindrospermopsininduced protein synthesis inhibition and its dissociation from acute toxicity in mouse hepatocytes. Environ. Toxicol. 18:243-51.

PAGE 73

73 Gkelis, S., Moustaka-Gouni, M., Sivonen, K. & Lanaras, T. 2005. First report of the cyanobacterium Aphanizomenon ovalisporum Forti in two Greek lakes and cyanotoxin occurrence. J. Plankton. Res. 27:1295-300. Griffiths, D. J. & Saker, M. L. 2003. The Palm Island mystery disease 20 years on: a review of research on the cyanotoxin cylindrospermopsin. Environ. Toxicol. 18:78-93. Haider, S., Naithani, V., Viswanathan, P. N. & Kakkar, P. 2003. Cyanobacterial toxins: a growing environmental concern. Chemosphere 52:1-21. Harada, K., Ohtani, I., Iwamoto, K., Suzuki, M. Watanabe, M. F., Watanabe, M. & Terao, K. 1994. Isolation of cylindrospermopsin from a cyanobacterium Umezakia Natans and its screening method. Toxicon 32:73-84. Hawkins, P. R., Runnegar, M. T. C., Jackson, A. R. B. & Falconer, I. R. 1985. Severe hepatotoxicity caused by the tropica l cyanobacterium (blue-green-alga) Cylindrospermopsis-raciborskii (Woloszynska) Seenaya and S ubba Raju isolated from a domestic water-supply reservoir. Appl. Environ. Microbiol. 50:1292-5. Hindk, F. 2000. Morphological variation of four planktic nostocalean cyanophytes members of the genus Aphanizomenon or Anabaena ? Hydrobiologia. 438:107-16. Hisbergues, M., Christiansen, G., Rouhiaine n, L., Sivonen, K. & Borner, T. 2003. PCR-based identification of microcystin -producing genotypes of differe nt cyanobacterial genera. Arch. Microbiol. 180:402-10. Hitzfeld, B. C., Hoger, S. J. & Dietrich, D. R. 2000. Cyanobacterial toxins: Removal during drinking water treatment, and human risk assessment. Environ. Health Persp. 108:11322. Hodgson, L. M., Linda, S. B. & Canfield, D. E. J. 1986. Periphytic algal growth in a hypereutrophic Florida lake following a winter decline in phytoplankton. Florida Sci. 49:234-41. Hoiczyk, E. & Baumeister, W. 1995. Envelope stru cture of 4 gliding filamentous cyanobacteria. J. Bacteriol. 177:2387-95. Howeler, M., Ghiorse, W. C. & Walker, L. P. 2003. A quantitative analysis of DNA extraction and purification from compost. J. Microbiol. Methods. 54:37-45. Humpage, A. R., Fenech, M., Thomas, P. & Fa lconer, I. R. 2000. Micronucleus induction and chromosome loss in transformed human white cells indicate clast ogenic and aneugenic action of the cyanobacterial toxin, cylindrospermopsin. Mut. Res. 472:155-61.

PAGE 74

74 Jackson, C. R., Harper, J. P., Willoughby, D., Rode n, E. E. & Churchill, P. F. 1997. A simple, efficient method for the separation of humic substances and DNA from environmental samples. Appl. Environ. Microbiol. 63:4993-5. Kaebernick, M., Dittmann, E., Borner, T. & Neil an, B. A. 2002. Multiple alternate transcripts direct the biosynthesis of microcystin, a cyanobacterial nonribosomal peptide. Appl Environ Microbiol 68:449-55. Kaebernick, M. & Neilan, B. A. 2001. Ecologica l and molecular investig ations of cyanotoxin production. FEMS Microbiol. Ecol. 35:1-9. Kaebernick, M., Neilan, B. A., Borner, T. & Dittmann, E. 2000. Light and the transcriptional response of the microcystin biosynthesis gene cluster. Appl Environ Microbiol 66:338792. Kellmann, R., Mills, T. & Neilan, B. A. 2006. Fu nctional modeling and p hylogenetic distribution of putative cylindrospermops in biosynthesis enzymes. J. Mol. Evol. 62:267-80. Komrek, J. & Kling, H. 1991. Variation in 6 pl anktonic cyanophyte genera in Lake Victoria (East-Africa). Arch. Hydrobiol. Suppl. 88:21-45. Komrek, J. & Kov ik, L. 1989. Trichome structure of 4 Aphanizomenon taxa (cyanophyceae) from Czechoslovakia, with notes on th e taxonomy and delimitation of the genus. Plant Syst. Evol. 164:47-64. Krienitz, L., Ballot, A., Kotut, K., Wiegand, C., Putz, S., Metcalf, J. S., Codd, G. A. & Pflugmacher, S. 2003. Contribution of hot spring cyanobacteria to the mysterious deaths of Lesser Flamingos at Lake Bogoria, Kenya. FEMS Microbiol. Ecol. 43:141-8. Lagos, N., Onodera, H., Zagatto, P. A., Andrinol o, D., Azevedo, S. & Oshima, Y. 1999. The first evidence of paralytic shellfish toxins in the freshwater cyanobacterium Cylindrospermopsis raciborskii isolated from Brazil. Toxicon 37:1359-73. LaMontagne, M. G., Michel, F. C., Jr., Hold en, P. A. & Reddy, C. A. 2002. Evaluation of extraction and purification methods for obt aining PCR-amplifiable DNA from compost for microbial community analysis. J. Microbiol. Methods. 49:255-64. Li, R., Carmichael, W. W., Brittain, S., Eagl esham, G. K., Shaw, G. R., Mahakhant, A., Noparatnaraporn, N., Yongmanitchai, W., Ka ya, K. & Watanabe, M. M. 2001a. Isolation and identification of the cyanotoxin cylindr ospermopsin and deoxy-cylindrospermopsin from a Thailand strain of Cylindrospermopsis raciborskii (Cyanobacteria). Toxicon 39:973-80.

PAGE 75

75 Li, R. H., Carmichael, W. W., Brittain, S., Ea glesham, G. K., Shaw, G. R., Liu, Y. D. & Watanabe, M. M. 2001b. First report of the cyanotoxins cylindrospermopsin and deoxycylindrospermopsin from Raphidiopsis curvata (Cyanobacteria). J. Phycol. 37:1121-6. Lis, J. T. & Schleif, R. 1975. Size fractiona tion of double-stranded DNA by precipitation with polyethylene-glycol. Nucl. Acids Res. 2:383-9. Mak, Y. M. & Ho, K. K. 1992. An improved me thod for the isolation of chromosomal DNA from various bacteria and cyanobacteria. Nucl. Acids Res. 20:4101-2. Metcalf, J. S., Lindsay, J., Beattie, K. A., Birm ingham, S., Saker, M. L., Torokne, A. K. & Codd, G. A. 2002. Toxicity of cylindros permopsin to the brine shrimp Artemia salina : comparisons with protein synthesis inhibitors and microcystins. Toxicon 40:1115-120. Miller, D. N., Bryant, J. E., Madsen, E. L. & Ghiorse, W. C. 1999. Evaluation and optimization of DNA extraction and puri fication procedures for soil and sediment samples. Appl. Environ. Microbiol. 65:4715-24. Neilan, B. A., Jacobs, D., Del Dot, T., Blackall, L. L., Hawkins, P. R., Cox, P. T. & Goodman, A. E. 1997. rRNA sequences and evolutiona ry relationships among toxic and nontoxic cyanobacteria of the genus Microcystis Int. J. Syst. Bacteriol. 47:693-7. Neilan, B. A., Jacobs, D. & Goodman, A. E. 1995. Genetic diversity and phylogeny of toxic cyanobacteria determined by DNA polymor phisms within the phycocyanin locus. Appl Environ Microbiol 61:3875-83. Neilan, B. A., Saker, M. L., Fastner, J., Tor okne, A. & Burns, B. P. 2003. Phylogeography of the invasive cyanobacterium Cylindrospermopsis raciborskii Mol. Ecol. 12:133-40. Norris, R. L., Seawright, A. A., Shaw, G. R., Senogles, P., Eaglesham, G. K., Smith, M. J., Chiswell, R. K. & Moore, M. R. 2002. Hepatic xenobiotic metabolism of cylindrospermopsin in vivo in the mouse. Toxicon 40:471-6. Paerl, H. W., Pinckney, J. L. & Steppe, T. F. 2000. Cyanobacterial-bacterial mat consortia: examining the functional unit of microbial surv ival and growth in extreme environments. Environ. Microbiol. 2:11-26. Paerl, H. W., Fulton, R. S., 3rd, Moisander, P. H. & Dyble, J. 2001. Harmful freshwater algal blooms, with an emphasis on cyanobacteria. Sci. World J. 1:76-113. Paithankar, K. R. & Prasad, K. S. N. 1991. Precipitation of DNA by polyethylene-glycol and ethanol. Nucl. Acids Res. 19:1346-.

PAGE 76

76 Pan, H., Song, L. R., Liu, Y. D. & Born er, T. 2002. Detection of hepatotoxic Microcystis strains by PCR with intact cells from both culture and environmental samples. Arch. Microbiol. 178:421-7. Phlips, E. J. 2002. Algae and eutrophication. In Bitton, G. [Ed.] Encyclopedia of Environmental Microbiology. Wiley, p. 3527. Phlips, E. J., Bledsoe, E., Cichra, M. F. & Ba dylak, S. 2003. The distribu tion of potentially toxic cyanobacteria in Florida. In Johnson, D. & Harbison, R. D. [Eds.] Proceedings of health effects of exposure to cyanobacterial toxins: State of the science. Florida Department of Health and the Center for Disease Control, St. Peterburg, Florida. Pollingher, U., Hadas, O., Yacobi, Y. Z., Zohary, T. & Berman, T. 1998. Aphanizomenon ovalisporum (Forti) in Lake Kinneret, Israel. J. Plankton. Res. 20:1321-39. Preussel, K., Stuken, A., Wiedner, C., Chor us, I. & Fastner, J. 2006. First report on cylindrospermopsin producing Aphanizomenon flos-aquae (Cyanobacteria) isolated from two German lakes. Toxicon 47:156-62. Rajaniemi, P., Hrouzek, P., Kastovska, K., Willame, R., Rantala, A., Hoffmann, L., Komarek, J. & Sivonen, K. 2005. Phylogenetic and mor phological evaluation of the genera Anabaena, Aphanizomenon, Trichormus and Nostoc (Nostocales, Cyanobacteria). Int. J. Syst. Evol. Microbiol. 55:11-26. Saker, M. L., Nogueira, I. C., Vasconcelos, V. M. Neilan, B. A., Eaglesham, G. K. & Pereira, P. 2003. First report and toxicological assessment of the cyanobacterium Cylindrospermopsis raciborskii from Portuguese freshwaters. Ecotoxicol. Environ. Safety 55:243-50. Schembri, M. A., Neilan, B. A. & Saint, C. P. 2001. Identification of genes implicated in toxin production in the cyanobacterium Cylindrospermopsis raciborskii Environ. Toxicol. 16:413-21. Schneegurt, M. A., Dore, S. Y. & Kulpa, C. F ., Jr. 2003. Direct extraction of DNA from soils for studies in microbial ecology. Curr. Iss. Mol. Biol. 5:1-8. Seawright, A. A., Nolan, C. C., Shaw, G. R., Chis well, R. K., Norris, R. L., Moore, M. R. & Smith, M. J. 1999. The oral toxicity fo r mice of the tropical cyanobacterium Cylindrospermopsis raciborskii (Woloszynska). Environ. Toxicol. 14:135-42. Seifert, M., McGregor, G., Eaglesham, G ., Wickramasinghe, W. & Shaw, G. 2007. First evidence for the production of cylindrosperm opsin and deoxy-cylindrospermop sin by the freshwater benthic cyanobacterium, Lyngbya wollei (Farlow ex Gornont) Speziale and Dyck. Harmful Algae 6:73-80.

PAGE 77

77 Shalev-Alon, G., Sukenik, A., Livnah, O., Sc hwarz, R. & Kaplan, A. 2002. A novel gene encoding amidinotransferase in the cy lindrospermopsin producing cyanobacterium Aphanizomenon ovalisporum FEMS Microbiol. Lett. 209:87-91. Shaw, G. R., Seawright, A. A., Moore, M. R. & Lam, P. K. S. 2000. Cylindrospermopsin, a cyanobacterial alkaloid: Evaluation of its toxicologic activity. Ther. Drug Monit. 22:8992. Shaw, G. R., Sukenik, A., Livne, A., Chiswell, R. K., Smith, M. J., Seawright, A. A., Norris, R. L., Eaglesham, G. K. & Moore, M. R. 1999. Blooms of the cylindrospermopsin containing cyanobacterium, Aphanizomenon ovalisporum (Forti), in newly constructed lakes, Queensland, Australia. Environ. Toxicol. 14:167-77. Shen, X. Y., Lam, P. K. S., Shaw, G. R. & Wickramasinghe, W. 2002. Genotoxicity investigation of a cyanobacter ial toxin, cylindrospermopsin. Toxicon 40:1499-501. Sivonen, K., Carmichael, W. W., Namikoshi, M., Rine hart, K. L., Dahlem, A. M. & Niemela, S. I. 1990. Isolation and characte rization of hepatotoxic microcystin homologs from the filamentous fresh-water cyanobacterium Nostoc sp strain-152. Appl. Environ. Microbiol. 56:2650-7. Smith, L. M., Sanders, J. Z., Kaiser, R. J., H ughes, P., Dodd, C., Connell, C. R., Heiner, C., Kent, S. B. H. & Hood, L. E. 1986. Fluores cence detection in automated DNA-sequence analysis. Nature 321:674-9. Spoof, L., Berg, K. A., Rapala, J., Lahti, K., Lepi sto, L., Metcalf, J. S., Codd, G. A. & Meriluoto, J. 2006. First observation of cylindrospermopsin in Anabaena lapponica isolated from the boreal environment (Finland). Environ. Toxicol. 21:552-60. Stanier, R. Y., Kunisawa, R., Mandel, M. & Cohenbaz.G. 1971. Purification and properties of unicellular blue-green alg ae (order chroococcales). Bacteriol. Rev. 35:171-&. Tillett, D., Dittmann, E., Erhard, M., von Dohr en, H., Borner, T. & Neilan, B. A. 2000. Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide-pol yketide synthetase system. Chem. Biol. 7:753-64. Tillett, D. & Neilan, B. A. 2000. Xanthogenate nucleic acid isolation from cultured and environmental cyanobacteria. J. Phycol. 36:251-8. Trkne, A., Asztalos, M., Bankine, M., Bickel H., Borbely, G., Carmeli, S., Codd, G. A., Fastner, J., Huang, Q., Humpage, A., Metcalf, J. S., Rabai, E., Sukenik, A., Suranyi, G., Vasas, G. & Weiszfeiler, V. 2004. Interlaboratory comparison trial on cylindrospermopsin measurement. Anal. Biochem. 332:280-4. Tringe, S. G. & Rubin, E. M. 2005. Metagenomics: DNA sequencing of environmental samples. Nature Rev. Genet. 6:805-14.

PAGE 78

78 Wechter, P., Williamson, J., Robertson, A. & Kluepfel, D. 2003. A rapid, cost-effective procedure for the extraction of microbial DNA from soil. World J. Microbiol. Biotech. 19:85-91. Wilson, K. M., Schembri, M. A., Baker, P. D. & Saint, C. P. 2000. Molecular characterization of the toxic cyanobacterium Cylindrospermopsis raciborskii and design of a species-specific PCR. Appl. Environ. Microbiol. 66:332-8. Y lmaz, M., Phlips, E. J., Szabo, N. J. & Ba dylak, S. 2007. A compara tive study of Florida strains of Cylindrospermopsis and Aphanizomenon for cylindrospermopsin production. Toxicon doi:10.1016/j.toxicon.2007.08.013. Zanardi-Lamardo, E., Clark, C. D., Moore, C. A. & Zika, R. G. 2002. Comparison of the molecular mass and optical properties of colore d dissolved organic material in two rivers and coastal waters by flow field-flow fractionation. Environ. Sci. Tech. 36:2806-14. Zhou, J., Bruns, M. A. & Tiedje, J. M. 1996. DNA recovery from soils of diverse composition. Appl. Environ. Microbiol. 62:316-22. Zipper, H., Buta, C., Lammle, K., Brunner, H., Bernhagen, J. & Vitzthum, F. 2003. Mechanisms underlying the impact of humic acids on DNA quantification by SYBR Green I and consequences for the analysis of soils and aquatic sediments. Nucl. Acids Res. 31.

PAGE 79

79 BIOGRAPHICAL SKETCH Mete Y lm az was born in Ankara, Turkey in 1976. He started his undergraduate degree at Ege University in 1995, majoring in aquaculture. Upon graduation in 1999, he started his M.Sc. in aquaculture there and completed the program in 2001. He worked on mass cultivation of a cyanobacterium, Spirulina platensis, to be used in human consumption, and optimized conditions for the production of a pigment (phycocyanin) produced by the algae. He became interested in pigment biosynthesis and went to Brown Univers ity in 2002 to work on this topic with Sam Beale. He came to the University of Florid a in 2003 and started his PhD under the supervision of Edward J. Phlips. He has been working on the molecular ecology of to xic cyanobacteria since then.