a... ,.
nmﬁhuté

 

xt.....xx 1..
:1... ‘

 

.1...
I It... I .
. l... v1v3Iy
\wx...v3...:.q

. .. \,

 

 

 

57.1,:

 

Em $5.91..

11.:

 

 

This is to certify that the
dissertation entitled

VERTICAL AND HORIZONTAL DISTRIBUTION OF
DENITRIFIER COMMUNITIES IN PACIFIC NORTHWEST
AND ARCTIC MARINE SEDIMENTS

 

CD
‘5
4—:
E w a
< C 5 presented by
.m (u 51>)
O) .99 E
if) D V ' ' G tz
" eronlca riJn i
E 9

 

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Microbiology and Molecular
1 Genetics

 

\

 

I ijor Protessor'ﬁ Signature
Anzac/#275, QQ?‘

_ Date

 

MSU is an afﬁnnstive-action, equal-opportunity employer

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATEDUE

DATEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p./ClRC/DaIeDue.indd-p.1

 

VERTICAL AND HORIZONTAL DISTRIBUTION OF DENITRIFIER
COMMUNITIES IN PACIFIC NORTHWEST AND ARCTIC MARINE
SEDIMENTS

By

Veronica Grilntzig

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Microbiology and Molecular Genetics

2007

ABSTRACT

VERTICAL AND HORIZONTAL DISTRIBUTION OF DENITRIFIER
COMMUNITIES IN PACIFIC NORTHWEST AND ARCTIC MARINE
SEDIMENTS

By

Veronica Grtintzi g

Denitriﬁcation in marine sediments is a major sink for nitrogen in the global
nitrogen budget and the main sink of ﬁxed nitrogen in the oceans. Therefore,
understanding the distribution of denitriﬁer communities, their activity, and types of
denitriﬁers in sediments may help reﬁne denitriﬁcation estimates and better model the
marine N cycle. The vertical and horizontal distribution of denitriﬁers was studied in
sediments from various locations in the Paciﬁc Northwest (Puget Sound, continental
slope and abyssal sea ﬂoor) and the Arctic with different biogeochemical characteristics,
using the heme cd, nitrite reductase gene (nirS) as a molecular marker. Since a method
was lacking to rapidly and accurately quantify the organisms involved in this key
biogeochemical process, real-time PCR was tested for its applicability for the
quantiﬁcation of a functional gene, the nirS gene of Pseudomonas stutzeri, in the
environment. The assay was speciﬁc, sensitive down to one copy of the gene, precise and
accurate, and indicated that P. stutzeri, though fairly cosmopolitan, might not be as
dominant as previously thought. On the other hand, site-speciﬁc abundant nirS gene
sequences, unrelated to known denitriﬁers, were detected at each site. The vertical
distribution of denitriﬁers was correlated to the bioturbation of the sediments and not to

the redox gradients. A conserved denitriﬁer community structure, as well as

denitriﬁcation capacity, was detected throughout the mixed sediment layers, reaching as
deep as 37 cm in some locations, regardless of the presence of oxygen or nitrate. In the
horizontal dimension the different locations seem to harbor distinct denitrifying and
bacterial communities, as indicated by statistically signiﬁcant differences between nirS
gene clone libraries as well as 16S rRNA gene clone libraries at each site, in addition to
site-speciﬁc clustering of highly similar sequences for both genes. Differentiation of
denitrifying communities was mostly affected by geographic location, followed by
differing environmental factors, often related to water depth, within one area, and least
affected by sediment depth. The bacterial community, as studied by the 16S rRNA gene,
was less inﬂuenced by geographic location, probably due to the lower mutation rate of
this gene. Although all sites harbored rich and diverse denitriﬁer and bacterial
communities, a general trend towards higher diversity at intermediate productivity levels

was detected, suggesting a possible correlation between these two factors.

ACKNOWLEDGMENTS

Many, many people have contributed in one or the other way in the completion of
this work and although I’m not going to be able to name all of them, be assured that I’m
grateful towards, remember and cherish each one of them. First of all I want to thank Dr.
James Tiedje for giving me the opportunity to join his lab, for his guidance and teachings,
for always being available for me regardless of his busy agenda, taking the time to even
discuss my thesis in Germany, and for his constant support and encouragement. It has
been a great honor and pleasure to form part of the Tiedje lab. I want to thank my
committee members Dr. Allan Devol, Dr. Terry Marsh, and Dr. Mike Klug for their
guidance and suggestions, for always being available to answer my questions, for turning
every meeting into an interesting scientiﬁc discussion rather than an examination, and for
their patience, constant support and encouragement. Thanks also to Dr. Allan Devol for
allowing me to work in his lab and participate in the sampling cruises, which have been
wonderful experiences, for teaching me the oceanographic point of view of my work, for
his guidance during the sampling and microcosms experiment, and for even getting
muddy while helping out with them. Thanks to Dr. Terry Marsh for his guidance and
teachings throughout my Ph.D., but especially during my rotation through his lab, a great
experience, which has introduced me to many aspects of my later work. Thanks to Dr.
Mike Klug, for being absolutely encouraging every time I contacted him, making me
smile with each e-mail, and giving me the strength to think that I was able to ﬁnish this
task. So, again, thank you very much to Dr. Tiedje and my committee for having guided

me all the way through my Ph.D. to ﬁnally be able to write these acknowledgments!

iv

Thank you to all the past and present members of the Tiedje lab, a great and
stimulating working environment in which people are always open to help and share their
knowledge to make everybody’s work better. Special thanks go to my ﬁ'iends He'ctor,
Clary and Xiao, who have always been and still are there for me. Thanks for all the
scientiﬁc discussions and great suggestions, and also for the non-scientiﬁc suggestions
and baby-tips, for all the encouragement and most of all for your invaluable ﬁiendship.
Thanks to Steve and Gesche for introducing me to the world of denitriﬁcation in marine
sediments and for great guidance and suggestions. Thanks also go to many other friends
in and outside the Tiedje lab who have helped me in one or the other aspect of my work
and most of all, who have made my stay in Michigan a wonderful experience. Thanks to
Lizy, Blaz, Hanna, Kristian, Nid, Carmen, Claudia, Erick (thanks for the printing, picking
me up, etc, etc), Carlos, Vincent, Alban, Ryan, Debora, Jorge, Olga, Lycely, Aviaja,
Stephan, Tamara, and many, many others. A very special thank you goes also to Lisa,
who has been not only a friend, but like a “big sister” away from home, helping me and
my family in many, many ways that I can not be thankful enough for. In addition to Lisa,
thank you also to Pat, Nicky, Darlene, Angie, and Suzanne for all their assistance. Thank
you to Joyce, for all her help and her inﬁnite kindness. Thanks to Lindsey and Chia Yu
who have helped me in different parts of this research. Thanks to the RDP, especially to
Benli, for always being available to answer questions and helping me with the sequences.
Thanks to Christine at KBS for the help with the GC measurements. And a special thank
you also to Dr. Helmut Bertrand and Dr. Robert Hausinger who as Directors of Graduate
Studies have always been extremely kind and helpful. The list is long, but thanks to all

people at MSU that have contributed in my education and in the research presented here.

I am also very grateful towards the people at the University of Washington,
especially in Dr. Devol’s laboratory and the Department of Chemical Oceanography, in
addition to people from other institutions participating in the sampling cruises, that have
helped me with the sample retrieval, microcosms experiments, and made my stay there a
great experience, especially Gerard, Wendy, Cindy, Brooke, Ben, Heather, Linda, and
Erin.

Besides the people that have helped me directly in my research, there is a huge
number of people that have helped me to reach this point by giving me their constant
support and encouragement. I have to say thank you to my friends in Argentina,
Switzerland and Germany, who have always been at my side, cheering me up all along
the way. Thank you very much to all my family, for all their love and support. And a very
special thank you goes to my mom and my dad, for always being there for me and
believing in me, for encouraging me in all my decisions, even if that meant letting me go
to do a Ph.D. far away from home, and most of all, for giving me all their love now and
always. Danke ﬁir Alles, Mama und Papa, ich habe Euch sehr lieb!

And now, last but deﬁnitely not least, I want to thank the two most important
persons in my life, without whom none of this would have been possible, my husband
Gustavo and my daughter Marina. Thank you for always being there for me, for helping
me with everything you could, for encouraging me and supporting me all along the way,
and most of all, thank you for your unconditional love. Gracias por todo mis amores, los
quiero mucho como el cielo!

Thanks to everybody for helping me make this dream a reality.

vi

PREFACE

The work presented in this dissertation was part of a collaborative effort between
Michigan State University, the University of Washington, Oak Ridge National
Laboratory and the University of Puerto Rico, and funded by the Department of Energy.
The Arctic samples analyzed were retrieved by Dr. Allan H. Devol, of the University of
Washington. The Paciﬁc Northwest samples were retrieved and processed also under Dr.
Devol’s guidance. Dr. Devol’s laboratory was responsible for the biogeochemical
analysis of all samples from the Arctic and the Paciﬁc Northwest.

The preliminary alignment of nirS gene sequences with the Hidden Markov Model
aligner was performed by Dr. Benli Chai of the Ribosomal Database Project.

The in silico digestion of nirS sequences with a custom made Perl script was
performed by Dr. Héctor L. Ayala-del-Rio of MSU and latter the University of Puerto

Rico — Humacao.

vii

TABLE OF CONTENTS

 

LIST OF TABLES ............................................................................................................ xi
LIST OF FIGURES ............................................................................... xiii
CHAPTER 1
INTRODUCTION - -- - -- - - - - ..... - ----l
OBJECTIVES .................................................................................................................... 11
Objective 1 ................................................................................................................. 11
Objective 2 ................................................................................................................. 12
Objective 3 ................................................................................................................. 13
REFERENCES ................................................................................................................... 14
CHAPTER 2
PSEUDOMONAS S T U TZERI NITRITE REDUCTASE GENE ABUNDANCE IN
ENVIRONMENTAL SAMPLES MEASURED BY REAL-TIME PCR ................... 19
ABSTRACT ...................................................................................................................... 19
INTRODUCTION ............................................................................................................... 19
MATERIALS AND METHODS ............................................................................................ 22
Samples ...................................................................................................................... 22
Cultures ..................................................................................................................... 23
DNA extraction and quantitation .............................................................................. 24
Primers and probe ..................................................................................................... 25
Real-time PCR ........................................................................................................... 25
Specificity ................................................................................................................... 26
Sensitivity and detection limit .................................................................................... 27
Quantitation of nirS in environmental samples ......................................................... 28
Accuracy .................................................................................................................... 28
RESULTS ......................................................................................................................... 29
Specificity ................................................................................................................... 29
Sensitivity and detection limit .................................................................................... 30
Accuracy .................................................................................................................... 3]
Analyses of environmental samples ........................................................................... 32
DISCUSSION .................................................................................................................... 44
Acknowledgements ..................................................................................................... 46
REFERENCES ................................................................................................................... 48
CHAPTER 3

DENITRIFIER COMMUNITY STRUCTURE AND ACTIVITY WITH RESPECT
TO REDOX GRADIENTS AND BIOTURBATION-- - - -52

ABSTRACT ...................................................................................................................... 52

 

viii

INTRODUCTION ............................................................................................................... 53

MATERIALS AND METHODS ............................................................................................ 55
Stuay area and sediment sampling ............................................................................ 55
Chemical analyses of sediment samples .................................................................... 56
DNA extraction and purification ............................................................................... 56
T —RF LP analysis of nirS gene .................................................................................... 5 7
Analysis of nirS T -RFLP proﬁles ............................................................................... 59
Cloning of nirS sequences ......................................................................................... 60
Phylogenetic analysis of cloned nirS sequences ........................................................ 61
Quantification of nirS from specific clusters and strains by real-time PCR ............. 62
Denitrification capacity in sediments ........................................................................ 64

RESULTS ......................................................................................................................... 65
Biogeochemical characteristics of the sediments ...................................................... 65
Denitrifier community structure by T -RFLP of nirS genes ....................................... 66
Phylogenetic analysis of nirS clone sequences .......................................................... 69
Quantification of clone clusters by real-time PCR .................................................... 71
Denitrification capacity of sediments studied in microcosms ................................... 73

DISCUSSION .................................................................................................................. 102

REFERENCES ................................................................................................................. 11 1

CHAPTER 4

DENITRIFIER AND BACTERIAL DIVERSITY AND DISTRIBUTION
PATTERNS IN SEDIMENTS FROM THE ARCTIC AND THE PACIFIC

 

NORTHWEST _ - -- - 116
ABSTRACT .................................................................................................................... 116
INTRODUCTION ............................................................................................................. 1 17
MATERIALS AND METHODS .......................................................................................... 119

Study area and sediment sampling .......................................................................... 119
Chemical analyses of sediment samples .................................................................. 120
DNA extraction and purification ............................................................................. 120
T-RFLP analysis of nirS gene .................................................................................. 120
Cloning of nirS and 16S rRNA gene sequences ....................................................... 122
Phylogenetic analysis of cloned nirS and 16S rRNA gene sequences ..................... 123
Community description and comparison based on 16S rRNA and nirS clone libraries
................................................................................................................................. 1 24
Quantification of nirS from specific clusters and strains by real-time PCR ........... 125
RESULTS ....................................................................................................................... 126
Biogeochemical characteristics of the sediments .................................................... 126
Denitrifier community structure by T -RF LP of nirS genes ..................................... 127
Phylogenetic analysis of nirS gene clone sequences ............................................... I30
Quantification of clone clusters by real-time PCR .................................................. 132
Phylogenetic analysis of 16S rRNA clone sequences .............................................. 134
Community description and comparison based on nirS and 16S rRNA clone libraries
................................................................................................................................. 13 7
DISCUSSION .................................................................................................................. 192
REFERENCES ................................................................................................................. 200

ix

CHAPTER 5

 

 

CONCLUSIONS AND FUTURE PERSPECTIVES--. -- - _ ‘ 206
FUTURE PERSPECTIVES ................................................................................................. 210
REFERENCES ................................................................................................................. 213

APPENDIX

RECOVERY OF NOVEL nirS GENES FROM NATURE AND TESTING OF

THEIR FUNCTIONALITY - ...... - - - 215
REFERENCES ................................................................................................................. 217

LIST OF TABLES

TABLE 2.1. PRIMER AND PROBE SEQUENCES COMPARED To EXAMPLE POSITIVE AND
NEGATIVE CONTROL MRS GENE SEQUENCES ............................................................... 33

TABLE 2.2. SPECIFICITY OF REAL-TIME PCR REACTIONS FOR PSEUDOMONAS STUTZERI DNA

................................................................................................................................... 34
TABLE 2.3. P. STUTZERI MRS GENE ABUNDANCE IN VARIOUS HABITATS. ............................ 42
TABLE 3.1. LOCATIONS AND CHARACTERISTICS OF SAMPLING STATIONS. .......................... 78
TABLE 3.2. SEDIMENT SAMPLES ANALYZED FROM DIFFERENT SITES. ................................. 79
TABLE 3.3. DESCRIPTION OF CLUSTERS IDENTIFIED IN MRS PHYLOGENETIC TREE. ............. 91

TABLE 3.4. PRIMERS AND PROBES USED To QUANTIFY THE MRS GENE OF SPECIFIC CLUSTERS
AND STRAINS BY REAL-TIME PCR ............................................................................... 92

TABLE 3.5. NITROUS OXIDE PRODUCTION RATES IN MICROCOSMS AMENDED WITH 400 MM

NANO3. ...................................................................................................................... 99
TABLE 4.1. LOCATIONS AND CHARACTERISTICS OF SAMPLING STATIONS. ........................ 142
TABLE 4.2. SEDIMENT SAMPLES ANALYZED FROM DIFFERENT SITES. ............................... 143

TABLE 4.3. CLUSTER SPECIFIC AND P. STUYZERI MRS GENE ABUNDANCE IN PACIFIC
NORTHWEST AND ARCTIC SEDIMENTS As MEASURED BY REAL-TIME PCR. .............. 168

TABLE 4.4. CLUSTER SPECIFIC AND P. STUTZER! MRS GENE ABUNDANCE IN PACIFIC
NORTHWEST SEDIMENTS AS MEASURED BY REAL-TIME PCR AND DETERMINATION OF
SIGNIFICANT DIFFERENCES BY ANOVA ................................................................... 169

TABLE 4.5. ASSIGNMENT OF 168 RRNA CLONE SEQUENCES FROM PACIFIC NORTHWEST
AND ARCTIC SEDIMENTS INTO TAXONOMIC GROUPS ................................................. 184

TABLE 4.6.SIGNIFICANT DIFFERENCES IN THE TAXONOMIC COMPOSITION BETWEEN 16S
RRNA LIBRARIES FROM PACIFIC NORTHWEST AND ARCTIC SEDIMENTS .................. 185

TABLE 4.7. DIVERSITY AND PREDICTED RICHNESS OF 168 RRNA AND MRS GENE

FRAGMENTS IN SEDIMENTS FROM DIFFERENT SITES BASED ON OTU ASSIGNMENT OF
CLONES BY DOTUR. ................................................................................................ 189

xi

TABLE 4.8. DETERMINATION OF SIGNIFICANT DIFFERENCES BETWEEN LIBRARIES BY THE
APPLICATION OF THE INTEGRAL FORM OF THE CRAMER-VON MISES STATISTIC WITH I-
LIBSHUFF. ............................................................................................................. 190

TABLE 4.9. SIMILARITY BETWEEN LIBRARIES DESCRIBED BY THE ABUNDANCE-BASED
SORENSON SIMILARITY INDEX (LAM/ND) AND WITH A NONPARAMETRIC RICHNESS
ESTIMATOR OF SHARED OTUS ANALOGOUS TO CHAol (S AB CHAO) As DETERMINED
WITH SONS. ............................................................................................................. 191

xii

LIST OF FIGURES

Some of the ﬁgures in this dissertation are presented in color.

FIGURE 1.1. DIAGRAM OF THE NITROGEN CYCLE INDICATING PROCESSES AND NITROGEN
COMPOUNDS INVOLVED. ............................................................................................... 2

FIGURE 1.2. DENITRIFICATION PATHWAY SHOWING THE NITROGEN SPECIES AND ENZYMES
INVOLVED. .................................................................................................................... 6

FIGURE 2.1. GENERATION OF STANDARD CURVE. ............................................................... 35
FIGURE 2.2. LIMIT OF DETECTION OF THE P. STUTZERI MRS GENE USING REAL-TIME PCR. .37
FIGURE 2.3. RELATIONSHIP BETWEEN THRESHOLD CYCLE (CT) AND ERROR. ...................... 39

FIGURE 2.4. QUANTIFICATION OF MRS IN ARTIFICIAL MIXTURES OF P. STUTZERI KC, P.
AERUGINOSA, AND E. cou CELLS. ............................................................................... 40

FIGURE 2.5. QUANTIFICATION OF MRS IN KBS SOIL (0) AND WINTERGREEN LAKE
SEDIMENT (A) SAMPLES SPIKED WITH P. STUTZERI CELLS. ......................................... 41

FIGURE 2.6. P. STUTZERI MRS COPY NUMBER MEASURED BY REAL-TIME PCR WITHOUT
(BLACK BARs) AND WITH (GRAY BARS) THE ADDITION OF 106 P. STUTZERI MRS COPY
NUMBERS T0 DNA EXTRACTED FROM DIFFERENT ENVIRONMENTAL SAMPLES. .......... 42

FIGURE 3.1. LOCATIONS OF SAMPLING STATIONS AT PUGET SOUND AND THE WASHINGTON
MARGIN SLOPE. ........................................................................................................... 77

FIGURE 3.2. PORE WATER PROFILES FOR 02 AND NO3' FOR PUGET SOUND (SHALLOW BUDD
INLET, TURNING BASIN, AND CARR INLET) AND WASHINGTON MARGIN SEDIMENTS. 80

FIGURE 3.3. PORE WATER PROFILES FOR FE(II), SO42} MN(II), AND NH4+ FOR PUGET
SOUND (SHALLOW BUDD INLET, TURNING BASIN, AND CARR INLET) AND
WASHINGTON MARGIN SEDIMENTS. ........................................................................... 82

FIGURE 3.4. T-RFLP PROFILES OBTAINED BY AMPLIFICATION OF MRS GENES FROM PUGET
SOUND AND WASHINGTON MARGIN MARINE SEDIMENTS FROM VARIOUS DEPTHS ...... 84

FIGURE 3.5. DENDROGRAM OBTAINED BY CLUSTER ANALYSIS OF T-RFLP PROFILES ......... 85

FIGURE 3.6. PCA ORDINATION OF T-RFLPS OF MRS GENES FROM PUGET SOUND AND
WASHINGTON MARGIN SEDIMENTS. ........................................................................... 86

xiii

FIGURE 3.7. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF MRS CLONE SEQUENCES
RETRIEVED FROM SHALLOW BUDD INLET, PUGET SOUND, SEDIMENTS TO SELECTED

REFERENCE SEQUENCES. ............................................................................................. 87
FIGURE 3.8. DETAIL OF SUBTREE A AS DESCRIBED IN FIGURE 3.7. ..................................... 88
FIGURE 3.9. DETAIL OF SUBTREE B As DESCRIBED IN FIGURE 3.7. ..................................... 89

FIGURE 3.10. COMPARISON OF T-RFLP PROFILES OBTAINED BY AMPLIFICATION OF MRS
GENES FROM SHALLOW BUDD INLET, PUGET SOUND, SEDIMENTS FROM VARIOUS
DEPTHS To DISTRIBUTION OF T-RF SIZES OBTAINED BY IN sruco DIGESTION OF MRS
CLONES RETRIEVED FROM THAT SAME SITE (SEDIMENT DEPTH 2-2.5 CM). .................. 90

FIGURE 3.1 1. CLUSTER SPECIFIC AND P. STUTZERI MRS GENE ABUNDANCE IN SHALLOW
BUDD INLET AND TURNING BASIN SEDIMENTS FROM DIFFERENT DEPTHS As MEASURED
BY REAL-TIME PCR. ................................................................................................... 93

FIGURE 3.12. CLUSTER SPECIFIC AND P. STUTZERI MRS GENE ABUNDANCE IN CARR INLET
AND WASHINGTON MARGIN SEDIMENTS FROM DIFFERENT DEPTHS AS MEASURED BY
REAL-TIME PCR .......................................................................................................... 94

FIGURE 3.13. NITROUS OXIDE PRODUCTION IN MICROCOSMS OF SHALLOW BUDD INLET
(PUGET SOUND) SEDIMENTS FROM VARIOUS DEPTHS AFTER ADDITION OF DIFFERENT
ELECTRON ACCEPTORS ................................................................................................ 95

FIGURE 3.14. NITROUS OXIDE PRODUCTION IN MICROCOSMS OF CARR INLET (PUGET
SOUND) SEDIMENTS FROM VARIOUS DEPTHS AFTER ADDITION OF DIFFERENT ELECTRON
ACCEPTORS. ................................................................................................................ 96

FIGURE 3.15. NITROUS OXIDE PRODUCTION IN MICROCOSMS OF TURNING BASIN (PUGET
SOUND) SEDIMENTS FROM VARIOUS DEPTHS AFTER ADDITION OF DIFFERENT ELECTRON
ACCEPTORS. ................................................................................................................ 97

FIGURE 3.16. NITROUS OXIDE PRODUCTION IN MICROCOSMS OF WASHINGTON MARGIN
SEDIMENTS FROM VARIOUS DEPTHS AFTER ADDITION OF DIFFERENT ELECTRON
ACCEPTORS. ................................................................................................................ 98

FIGURE 3.17. T-RFLP PROFILES OBTAINED BY AMPLIFICATION OF MRS GENES FROM
SHALLOW BUDD INLET, PUGET SOUND, SEDIMENTS COMPARED To PROFILES
OBTAINED AFTER INCUBATION OF SEDIMENTS FROM THE SAME SITE WITH 400 MM
NANO3 IN MICROCOSMS. .......................................................................................... 100

FIGURE 3.18. PCA ORDINATION OF T-RFLPS OF MRS GENES FROM SHALLOW BUDD INLET,

PUGET SOUND, SEDIMENTS NOT SUBJECTED AND SUBJECTED To INCUBATION WITH 400
MM NANO3 IN MICROCOSMS. ................................................................................... 101

xiv

FIGURE 4.1. LOCATIONS OF SAMPLING STATIONS. ............................................................ 141

FIGURE 4.2. PORE WATER PROFILES FOR 02 AND NO;' IN ARCTIC (BARROW CANYON
SHALLOW, BARROW CANYON DEEP, EAST HANNA SHOAL SHALLOW, AND EAST
HANNA SHOAL DEEP) SEDIMENTS. ........................................................................... 144

FIGURE 4.3. PORE WATER PROFILES FOR FE(II), MN(II), AND NH4+ IN ARCTIC (BARROW
CANYON SHALLOW, BARROW CANYON DEEP, EAST HANNA SHOAL SHALLOW, AND
EAST HANNA SHOAL DEEP) SEDIMENTS. .................................................................. 146

FIGURE 4.4. PORE WATER PROFILES FOR 02, N03] AND NH4+ IN SEDIMENTS FROM THE
ABYSSAL SEA FLOOR WEST OF THE JUAN DE FUCA RIDGE. ........................................ 148

FIGURE 4.5. T-RFLP PROFILES OBTAINED BY AMPLIFICATION OF MRS GENES FROM ARCTIC
(EHS, EAST HANNA SHOAL SHALLOW; EHSD, EAST HANNA SHOAL DEEP; BC,
BARROW CANYON SHALLOW; BCD BARROW CANYON DEEP) AND PACIFIC
NORTHWEST (WJF, WEST OF JUAN DE FUCA; WM, WASHINGTON MARGIN; CI, CARR
INLET; SB, SHALLOW BUDD INLET; TB, TURNING BASIN) MARINE SEDIMENTS FROM
VARIOUS DEPTHS. ..................................................................................................... 149

FIGURE 4.6A. RICHNESS OF THE MRS-CONTAINING COMMUNITY IN PACIFIC NORTHWEST
(TB, TURNING BASIN; SB, SHALLOW BUDD INLET; CI, CARR INLET, WM,
WASHINGTON MARGIN; WJ F, WEST OF JUAN DE FUCA) AND ARCTIC (BC, BARROW
CANYON SHALLOW; BCD, BARROW CANYON DEEP; EHS, EAST HANNA SHOAL
SHALLOW; EHSD, EAST HANNA SHOAL DEEP) SEDIMENTS FROM DIFFERENT DEPTHS,
BASED ON NUMBER OF INDIVIDUAL T-RFS DETECTED AFTER RESTRICTION WITH HHAI.
................................................................................................................................. 150

FIGURE 4.6B. DIVERSITY OF THE MRS-CONTAINING COMMUNITY IN PACIFIC NORTHWEST
(TB, TURNING BASIN; SB, SHALLOW BUDD INLET; CI, CARR INLET, WM,
WASHINGTON MARGIN; WJ F, WEST OF JUAN DE FUCA) AND ARCTIC (BC, BARROW
CANYON SHALLOW; BCD, BARROW CANYON DEEP; EHS, EAST HANNA SHOAL
SHALLOW; EHSD, EAST HANNA SHOAL DEEP) SEDIMENTS FROM DIFFERENT DEPTHS,
AS DETERMINED BY THE SHANNON DIVERSITY INDEX BASED ON THE T-RFLP RESULTS.
THE T-RFLP ANALYSIS WAS PERFORMED WITH RESTRICTION ENZYME HHAI .............. 151

FIGURE 4.7. DENDROGRAM OBTAINED BY CLUSTER ANALYSIS OF T-RFLP PROFILES ....... 152

FIGURE 4.8. PCA ORDINATION OF T-RFLPS OF MRS GENES FROM PACIFIC NORTHWEST
(WM, WASHINGTON MARGIN; TB, TURNING BASIN; SB, SHALLOW BUDD INLET; CI,
CARR INLET; WJF, WEST OF JUAN DE FUCA) AND ARCTIC (BCD, BARROW CANYON
DEEP; BC, BARROW CANYON SHALLOW; EHSD, EAST HANNA SHOAL DEEP; EHS,
EAST HANNA SHOAL SHALLow) SEDIMENTS ............................................................ 153

FIGURE 4.9. CANONICAL CORRESPONDENCE ANALYSIS (CCA) BASED ON T-RFLPS OF MRS
GENES AND ENVIRONMENTAL VARIABLES FROM PACIFIC NORTHWEST (TB, TURNING

XV

BASIN; SB, SHALLOW BUDD INLET; CI, CARR INLET, WM, WASHINGTON MARGIN;

WJF, WEST OF JUAN DE FUCA) AND ARCTIC (BC, BARROW CANYON SHALLOW; BCD,
BARROW CANYON DEEP; EHS, EAST HANNA SHOAL SHALLOW; EHSD, EAST HANNA
SHOAL DEEP) SEDIMENTS. ........................................................................................ 154

FIGURE 4.10. SCHEMATIC REPRESENTATION OF THE MRS PHYLOGENETIC TREE SHOWN AND
DESCRIBED IN FURTHER DETAIL IN FIGURE 4.11. ...................................................... 155

FIGURE 4.1 1. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF MRS CLONE SEQUENCES
RETRIEVED FROM SHALLOW BUDD INLET (SB), WASHINGTON MARGIN (WM), WEST
OF JUAN DE FUCA (WJF), BARROW CANYON SHALLOW (BC), AND EAST HANNA

SHOAL SHALLOW (EHS) SEDIMENTS To SELECTED REFERENCE SEQUENCES. ........... 156
FIGURE 4.12. DETAIL OF SUBTREE A AS DESCRIBED IN FIGURES 4.10. AND 4.11. ............. 158
FIGURE 4.13. DETAIL OF SUBTREE B As DESCRIBED IN FIGURES 4.10. AND 4.1 1. ............. 160
FIGURE 4.14. DETAIL OF SUBTREE C AS DESCRIBED IN FIGURES 4.10. AND 4.11. ............. 162
FIGURE 4.15. DETAIL OF SUBTREE D As DESCRIBED IN FIGURES 4.10. AND 4.11. ............. 163
FIGURE 4.16. DETAIL OF SUBTREE E AS DESCRIBED IN FIGURES 4.10. AND 4.11. ............. 165

FIGURE 4.17. COMPARISON OF T-RFLP PROFILES OBTAINED BY AMPLIFICATION OF MRS
GENES FROM ARCTIC (EHS, EAST HANNA SHOAL SHALLOW; BC, BARROW CANYON
SHALLOW) AND PACIFIC NORTHWEST (WJF, WEST OF JUAN DE FUCA; WM,
WASHINGTON MARGIN; SB, SHALLOW BUDD INLET) SEDIMENTS FROM VARIOUS
DEPTHS To DISTRIBUTION OF T-RF SIZES OBTAINED BY IN SILICO DIGESTION OF MRS
CLONES RETRIEVED FROM THOSE SAME SITES. .......................................................... 167

FIGURE 4.18. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE
SEQUENCES FROM THE GAMMA PROTEOBACTERIA RETRIEVED FROM SHALLOW BUDD
INLET (SB), WASHINGTON MARGIN (WM), WEST OF JUAN DE FUCA (WJF), BARROW
CANYON SHALLOW (BC), AND EAST HANNA SHOAL SHALLOW (EHS) SEDIMENTS To
SELECTED REFERENCE SEQUENCES. .......................................................................... 170

FIGURE 4.19. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE
SEQUENCES FROM THE DELTA PROTEOBACTERIA RETRIEVED FROM SHALLOW BUDD
INLET (SB), WASHINGTON MARGIN (WM), WEST OF JUAN DE FUCA (WJF), BARROW
CANYON SHALLOW (BC), AND EAST HANNA SHOAL SHALLOW (EHS) SEDIMENTS TO
SELECTED REFERENCE SEQUENCES. .......................................................................... 172

FIGURE 4.20. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE

SEQUENCES FROM THE ALPHA, BETA, AND EPSILON PROTEOBACTERIA RETRIEVED FROM
SHALLOW BUDD INLET (SB), WASHINGTON MARGIN (WM), WEST OF JUAN DE FUCA

xvi

(WJF), BARROW CANYON SHALLOW (BC), AND EAST HANNA SHOAL SHALLOW
(EHS) SEDIMENTS To SELECTED REFERENCE SEQUENCES. ........................................ 174

FIGURE 4.21. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 168 RRNA CLONE
SEQUENCES FROM THE BA CTEROIDETES, CHLOROBI, DEFERRIBACTERES, SPIROCHAETES,
AND ACIDOBACTERIA RETRIEVED FROM SHALLOW BUDD INLET (SB), WASHINGTON
MARGIN (WM), WEST OF JUAN DE FUCA (WJF), BARROW CANYON SHALLOW (BC),
AND EAST HANNA SHOAL SHALLOW (EHS) SEDIMENTS To SELECTED REFERENCE
SEQUENCES. .............................................................................................................. 176

FIGURE 4.22. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE
SEQUENCES FROM THE CHLOROFLEXI, ACTINOBACTERJA, NITROSPIRA, ODl GROUP AND
OPl 1 GROUP RETRIEVED FROM SHALLOW BUDD INLET (SB), WASHINGTON MARGIN
(WM), WEST OF JUAN DE FUCA (WJF), BARROW CANYON SHALLOW (BC), AND EAST
HANNA SHOAL SHALLOW (EHS) SEDIMENTS To SELECTED REFERENCE SEQUENCES.
................................................................................................................................. 178

FIGURE 4.23. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE
SEQUENCES FROM THE FIRM/CUTES, FUSOBA CTERIA, AND FIBROBACTERES RETRIEVED
FROM SHALLOW BUDD INLET (SB), WASHINGTON MARGIN (WM), WEST OF JUAN DE
FUCA (WJF), BARROW CANYON SHALLOW (BC), AND EAST HANNA SHOAL SHALLOW
(EHS) SEDIMENTS To SELECTED REFERENCE SEQUENCES. ........................................ 180

FIGURE 4.24. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 16S RRNA CLONE
SEQUENCES FROM THE PLANCTOMYCETES, VERRUCOMICROBIA AND WS3 GROUP
RETRIEVED FROM SHALLOW BUDD INLET (SB), WASHINGTON MARGIN (WM), WEST
OF JUAN DE FUCA (WJF), BARROW CANYON SHALLOW (BC), AND EAST HANNA
SHOAL SHALLOW (EHS) SEDIMENTS To SELECTED REFERENCE SEQUENCES. ........... 181

FIGURE 4.25. PHYLOGENETIC TREE SHOWING THE AFFILIATION OF 168 RRNA CLONE
SEQUENCES FROM THE CYANOBACTERIA RETRIEVED FROM SHALLOW BUDD INLET (SB),
WASHINGTON MARGIN (WM), WEST OF JUAN DE FUCA (WJF), BARROW CANYON
SHALLOW (BC), AND EAST HANNA SHOAL SHALLOW (EHS) SEDIMENTS To SELECTED
REFERENCE SEQUENCES. ........................................................................................... 183

FIGURE 4.26. RELATIONSHIP BETWEEN THE NUMBER OF OTUS AND THEIR ABUNDANCE, AS
MEASURED BY THE NUMBER OF CLONES CORRESPONDING To EACH OTU, FOR 16S
RRNA AND MRS CLONE LIBRARIES FROM BARROW CANYON SHALLOW (BC), EAST
HANNA SHOAL SHALLOW (EHS), SHALLOW BUDD INLET (SB), WASHINGTON
MARGIN (WM), AND WEST OF JUAN DE FUCA (WJF) SEDIMENTS ............................ 187

FIGURE 4.27. RAREFACTION CURVES FOR 16S RRNA AND MRS CLONE LIBRARIES FROM
BARROW CANYON SHALLOW (BC), EAST HANNA SHOAL SHALLOW (EHS), SHALLOW
BUDD INLET (SB), WASHINGTON MARGIN (WM), AND WEST OF JUAN DE FUCA (WJF)
SEDIMENTS. .............................................................................................................. 188

xvii

CHAPTER 1

INTRODUCTION

Nitrogen is a component of fundamental building blocks of living organisms, such
as nucleic and amino acids. Nitrogenous compounds experience multiple transformations,
catalyzed primarily by microbes, which constitute the N cycle (Figure 1.1). One of the
principal biotic transformations of nitrogen compounds within the N cycle is
denitriﬁcation, the dissimilatory reduction of nitrate or nitrite to gaseous products,
principally N2 and N20, with the concomitant conservation of energy (Tiedje, 1994).

Denitriﬁcation is basically a bacterial respiratory process, although it is also found
in Archaea and in mitochondria of certain fungi (Zumft, 1997). Within the bacteria,
denitriﬁers are found in diverse phylogenetic groups, although they are notably absent
from the enterobacteria (Zumft, 1992). They are also Spread among diverse physiological
groups, including organotrophs (organic energy source), lithotrophs (inorganic energy
source), and phototrophs (light as energy source), although the most common energy
sources are organic substrates. Most denitriﬁers are aerobic bacteria, which have the

alternative capacity to use N oxides as electron acceptors, when oxygen becomes limiting

(Tiedje, 1988).

Nitriﬁcation

   
    
 

NO2
Assimilation - - - N
K, NH, groups Assnmllatlon 2
_ °f prawns Nitrogen ﬁxation
/ Deamination Aerobic

 

 

Assimilatma—
NH2 groups

of proteins Deaminatio

 
 

I Anaerobic

 

 

Denitriﬁcation

Figure 1.1. Diagram of the nitrogen cycle indicating processes and nitrogen
compounds involved.

Chemodenitriﬁcation (abiotic conversion of N03' to N2 by reaction with inorganic
species), chemoxidation (abiotic conversion of NH; to N2 by reaction with inorganic
species), and OLAND (oxygen-limited autotrophic nitriﬁcation-denitriﬁcation) (Brandes
et al., 2007) have not been included for clarity.

Denitriﬁcation is the main sink for ﬁxed nitrogen and, therefore, a fundamental
biogeochemical process (Knowles, 1982). Although its main signiﬁcance is the
completion of the nitrogen cycle, it also is of interest, from an anthropogenic point of
View. First, it accounts generally for 20 to 70% of fertilizer losses, which is of concern
since nitrogen is the most limiting nutrient to crop production. On the other hand, it is
helpful in removing excess combined nitrogen in waste treatment and N-enriched coastal
areas. Two of its intermediate products, N20 and NO, have been recognized to play a role
in the destruction of the ozone layer and contributing to global warming, while the latter
and NO2', are toxic and could produce local hazards (see Tiedje, 1988 and references
therein).

Denitriﬁcation in marine sediments is a major sink for nitrogen in the global
nitrogen budget and is the main sink of oceanic ﬁxed nitrogen (Codispoti, 1995). The
continental margins account for half of the nitrogen cycling taking place in the oceans,
although their area represents less than 20% of the world oceans (Walsh, 1991). One of
the major Sites of oceanic nitrogen cycling are marine sediments, of which the
Washington margin, in the Paciﬁc Ocean, is a good example. It is a high productivity
area, due to wind-driven coastal upwelling (Hickey, 1989), with a large ﬂux of organic
matter to the sediments, which results in very high denitriﬁcation rates of 3.7 pmol N cm'
2 s'1 in average (Devol, 1991). Growing attention has been given to the Arctic Ocean,
with 50% of its area consisting of shelf seas, which account for about 25% of the world
continental Shelves. High productivity has been detected in the Chukchi Sea in that area,
related especially with the northward ﬂow of water through the Bering Strait, driven by

sea level difference (Woodgate et al., 2005). This area also presents high denitriﬁcation,

with rates of the highly productive Barrow Canyon area in the Chukchi Sea comparable
to those of the Washington margin (Devol et al., 2005). Although denitriﬁcation rates in
deep sea sediments are signiﬁcantly lower as compared to those in continental margin
sediments, these rates have been revised upwards (Middelburg et al., 1996) and
denitriﬁers have been detected and isolated from this environment (Tamegai et al., 2007;
Tarnegai et al., 1997), suggesting its importance for further study.

The present day marine nitrogen budget seems to be unbalanced (Codispoti, 1995;
Middelburg et al., 1996). The input of nitrogen to the ocean by atmospheric precipitation,
riverine input and in situ nitrogen ﬁxation seems tO be smaller than its removal by
sedimentary denitriﬁcation, denitriﬁcation in the water column, permanent burial in
sediments and organic nitrogen exports (Codispoti, 1995; Middelburg et al., 1996). We
are currently in an interglacial period when higher denitriﬁcation rates are thought to
occur than in glacial periods (Ganeshram et al., 1995), therefore supporting the
possibility of an unbalanced nitrogen budget. One consequence would be a loss of ﬁxed
nitrogen by the ocean that could lead to lower photosynthesis and consequently to a lower
oceanic uptake of carbon dioxide. This, in turn, would result in a higher carbon dioxide
concentration in the atmosphere, which would enhance global warming (Codispoti, 1995;
Middelburg et al., 1996). Despite the above argument, the nitrogen budget could still be
in balance, as the estimates of sedimentary denitriﬁcation vary by more than an order of
magnitude (Devol, 1991; Middelburg et al., 1996). This variability is due to extrapolation
from different Site-Speciﬁc estimates and to varying results when direct or indirect
measurements are used for estimation of global sedimentary denitriﬁcation (Devol,

1991). In order to accurately model the oceanic nitrogen budget, the range in sedimentary

denitriﬁcation estimates needs to be narrowed. Better information on the distribution and
diversity of denitriﬁers and on the response of their activities to changes in environmental
factors may provide more fundamental insight into interpreting and predicting
denitriﬁcation activities and therefore help reﬁne these estimates. On the other hand,
other processes leading to loss of ﬁxed nitrogen from sediments and water column have
been recently described (Brandes et al., 2007), including chemodenitriﬁcation (Luther III
et al., 1997) and anaerobic ammonium oxidation (anammox) (van de Graaf et al., 1995;
Dalsgaard et al., 2005). It has been suggested that chemodenitriﬁcation, the reduction of
nitrate to N2 by Mn”, could account for up to 90% of the N2 formed in continental
margin sediments (Luther III et al., 1997), while anammox, the anaerobic ammonium
oxidation with nitrite leading to the formation of N2, may account for 30 to 50% of the N2
produced in the oceans (Devol, 2003). Furthermore, 16S rRNA sequences associated to
known anammox bacteria have been detected in a variety of marine sediments, in
addition to other environments, suggesting that this process might be more widespread
than previously thought (Penton et al., 2006). Dissimilatory nitrate reduction to
ammonium (DNRA) is also widespread and more important than denitriﬁcation in
environments with high carbon to electron acceptor ratios (Tiedje, 1994), however, it
does not constitute a loss term for the marine nitrogen budget, as nitrate is reduced to
ammonium, which stays in the system. The discovery of these additional processes
leading to nitrogen loss could suggest an even further unbalanced marine nitrogen budget
than suggested by the increase in marine denitriﬁcation estimations alone. However,

nitrogen ﬁxation estimates have also increased and this process seems to be more

widespread than previously thought, therefore leaving open the possibility for a balanced
marine nitrogen budget (Brandes et al., 2007).
Denitriﬁcation consists of four reaction steps, each catalyzed by one or two

different enzymes (Zumft, 1997; Philippot, 2002) (Figure 1.2).

N03; —> N02- —> NO —> N20 ——> N2
Nar Nir Nor Nos

Figure 1.2. Denitrification pathway Showing the nitrogen species and enzymes
involved.

Nar, nitrate reductase; Nir, nitrite reductase; Nor, nitric oxide reductase; Nos, nitrous
oxide reductase.

The key enzyme in denitriﬁcation is nitrite reductase, which catalyzes the reduction
of nitrite to nitric oxide, because it is the ﬁrst step where ﬁxed nitrogen is lost to the
biosphere (Zumft, 1997). For this reason, this enzymatic step has been the focus of
physiological and biochemical studies on denitriﬁcation and several probes and
antibodies have been designed to target the nitrite reductases (Ward, 1996; Bothe et al.,
2000). There are two variants of the dissimilatory nitrite reductase. One contains copper
and is coded by the gene nirK while the other type contains hemes c and d, and is coded
by gene nirS (Zumft, 1997). Although normally these two nitrite reductases are mutually
exclusive, the existence of both genes, nirS and nirK, has been detected in the genome of

Paracoccus denitrificans Pd1222 (Hallin and Lindgren, 1999). Heme cd, nitrite reductase

is a homodimer with a subunit mass of around 60 kDa. It is a tetraheme protein as each
subtmit has a heme c and a heme d1. Copper—containing nitrite reductase is a trimer with
three identical subunits of close to 40 kDa each and Cu as prosthetic metal (Zumft, 1997).
Although these enzymes are functionally equivalent, they are structurally different and
can therefore not be targeted with a general probe against both of them. Still, the transfer
of the nirK gene from Pseudomonas aureofaciens to a Pseudomonas stutzeri strain with a
mutated nirS gene led to the expression of a ﬁmctional copper-containing nitrite
reductase (Glockner et al., 1993). The heme cd, type seems to be numerically dominant
in nature, while the copper type is present in a wider range of physiological groups,
including, among others Gram-positive Spore formers, nitrogen ﬁxing bacteria, and
Archaea (Coyne et al., 1989; Coyne and Tiedje, 1990; Goregues etal., 2005).

The taxonomic diversity of denitriﬁers makes rRNA-based approaches to study the
diversity and distribution of this functional guild inappropriate. Instead, the genes
encoding the key enzymes involved in this pathway have been targeted in many studies
as molecular markers. Although primers have been designed for genes encoding the
nitrate (Flanagan et al., 1999; Cheneby et al., 2003), nitric oxide (Braker and Tiedje,
2003), and nitrous oxide (Scala and Kerkhof, 1998, Throback et al., 2004) reductases, the
ﬁrst and most targeted genes from the pathway have been the ones encoding the nitrite
reductases (Braker et al., 1998; Hallin and Lindgren, 1999, Throback et al., 2004).
Although nirK appears in bacteria of deeper systematic afﬁliation than nirS, it is more
conserved and design of broad-range primers is somewhat easier (Bothe et al., 2000).
However, environmental studies have often not been able to detect nirK or, as expected,

detected a lower diversity as for nirS genes, especially in marine and other aquatic

samples (Braker et al., 2000; Tarnegai et al., 2007, Santoro et a1, 2006; Nogales et al.,
2002). Soil studies, on the other hand, Showed similar results for marshland soils,
presenting a higher nirS than nirK diversity, however, nirS was not detected in upland
soils, opposite to the general observation in aquatic environments (Priemé et al., 2002).
This could be related to differential ampliﬁcation efﬁciencies of the primer sets used or
indicate a real difference in the distribution of these two genes in the environment
(Priemé et al., 2002). Therefore, most denitriﬁer diversity studies in aquatic
environments have focused on nirS, rather than nirK genes (Braker et al., 2001; Castro-
Gonzalez et al., 2005; Hannig et al., 2006; Jayakumar et al., 2004). Regardless of the
denitriﬁcation gene targeted, the denitriﬁer diversity seems to be extremely high in
estuarine and continental Shelf marine sediments (Braker et al., 2000; Braker et al., 2001;
Liu et al., 2003; Scala and Kerkhof, 1999; Tiquia et al., 2006), and lower in the water
column (Hannig et al., 2006; Castro-Gonzalez et al., 2005; Oakley et al., 2007).
However, all these denitriﬁer diversity studies reveal the presence of a large number of
denitriﬁcation gene sequences unrelated to cultured denitriﬁers. Environmental
denitriﬁcation gene clone sequences form major clusters, generally including clone
sequences from other environments, but with no overlap with isolated strains (Braker et
al., 2000; Hannig et al., 2006; Jayakumar et al., 2004; Liu et al., 2003; Scala and
Kerkhof, 1999; Oakley et al., 2007). On the other hand, isolation of denitriﬁers from
sediments and water column generally leads to the cultivation of known denitriﬁers, such
as Marinomonas Sp., Pseudomonas Sp. (often Pseudomonas stutzeri), and Halomonas Sp.
(Braker et al., 2000; Oakley et al., 2007; Goregues et al., 2005), not closely related to the

novel sequences found in the environment. Although only isolation of a denitriﬁer with

one of the novel sequences would be a deﬁnite proof of its functionality, expression of
some of these novel denitriﬁcation gene sequences has been detected in estuarine
sediments from the river Colne for narG, napA, nirS, and nosZ (Nogales et al., 2002;
Smith et al., 2007), suggesting that they might belong to active denitriﬁers. Therefore,
molecular methods are more appropriate for studying the overall diversity of
denitriﬁcation genes.

Several molecular approaches have been taken to detect denitriﬁer diversity in the
environment (Philippot and Hallin, 2005). Most studies aiming to describe diversity have
included the construction of clone libraries of the gene of interest. This method exhibits
the highest resolution, allowing the detection of individual nucleotide differences
between gene sequences. However, clone libraries generally only contain a limited
ntunber of gene sequences, often representing only a fraction of the diversity present, as
pristine marine sediments have been estimated to contain ~ 11,000 different genome
equivalents (Torsvik et al., 2002). Therefore, to explore the general diversity of a
community, based on the diversity of a speciﬁc gene, a ﬁngerprinting technique, such as
terminal restriction fragment length polymorphism (T-RFLP) (Liu et al., 1997) is often
applied (Braker et al., 2001; Castro-Gonzalez et al., 2005; Hannig et al.; Scala and
Kerkhof, 2000). This allows the ﬁngerprinting of a community and the rapid comparison
between different sites and samples, however, at a lower resolution than cloning of the
genes.

The full understanding of the distribution of denitriﬁers in the environment
requires, besides the description of their diversity, the determination of their abundance.

Several PCR-based methods, such as most-probable-number and competitive PCR, have

been used to quantify nirS (Michotey et al., 2000) and nirK (Qiu et al., 2004) from the
environment. However, in recent years, real-time PCR (Heid et al., 1996) has been more
widely used than these other techniques, due to its Simplicity and higher accuracy as it
focuses on the logarithmic phase of product accumulation rather than the end product.
Therefore, several denitriﬁcation genes, including narG, napA, nirS, nirK, and nosZ have
been quantiﬁed in the environment using this technique (Smith et al., 2007; Henry et al.,
2004; Henry et al., 2006; LOpez—Gutiérrez et al., 2004).

Denitriﬁer distribution in marine sediments follows speciﬁc horizontal and vertical
biogeographic patterns. A conserved denitriﬁer community structure has been Observed
along the vertical redox potential gradient in marine sediments from the Paciﬁc
Northwest (Braker et al., 2001). Although nitrate and oxygen were present only in the
sediment top 1 cm, the denitriﬁer community structure was conserved throughout the
analyzed depth of 6.5 and 10 cm in Puget Sound and Washington continental shelf
sediments, respectively. Furthermore, nirS genes have been cloned from these same 6.5
cm deep Puget Sound sediments (Braker et al., 2000). Bioturbation of the sediments by
marine invertebrates has been suggested as an explanation for the presence of
denitriﬁcation genes, which might correspond to functional denitriﬁers, in sediments
below the penetration depths of oxygen and nitrate, however, no sediments below the
bioturbated zone had been analyzed in order to evaluate this hypothesis. On the other
hand, large variations in denitriﬁer community structure have been detected on a
horizontal scale, which increases in correlation with increasing geographic distance
(Scala and Kerkhof, 2000). Puget Sound and Washington continental shelf sediments also

exhibited differential nirS-based community structures at different Sites (Braker et al,

10

2000; 2001). These differences were more signiﬁcant between these two geographic
locations, than between sediments from different water depths at the continental Shelf.
Different sediment depths, on the other hand, exhibited the lowest differentiation, as
mentioned earlier.

Increased knowledge on the distribution, activity and types of denitriﬁers may help

reﬁne denitriﬁcation estimates and aid the better modeling of the marine N cycle.

Objectives

Based on the above rationale, the following objectives and hypothesis are

proposed:

Objective 1

To develop a real-time PCR assay for the quantification of Pseudomonas

stutzeri nirS and evaluate its applicability in the environment.

In order to better model denitriﬁcation in the environment, the quantiﬁcation of the
catalyst is helpful. Real-time PCR is a simple and accurate quantiﬁcation method, which
has been widely applied in the medical ﬁeld, but only recently started to be used on
environmental samples. Therefore, it needed to be tested for its applicability in the
environment, in particular for the detection of a functional gene. P. stutzeri, a commonly
isolated denitriﬁer from soil (Gamble et al., 1977) and marine environments (Ward and
Cockcroft, 1993; Braker et al., 2000), may play a Signiﬁcant role in global denitriﬁcation.

The determination of its abundance in the environment will help elucidate its importance.

11

Objective 2

To elucidate the distribution and activity of denitrifiers in marine sediments,

with respect to the bioturbation zones and redox gradients.

As described earlier, the vertical distribution of denitriﬁcation genes in marine
sediments exhibited a conserved community structure within the bioturbated layer, and
the mixing of the sediments by marine invertebrates had been suggested as an
explanation for this observation (Braker et al., 2001). However, sediments reaching
below the mixed layer need to be analyzed in order to further support this hypothesis. In
addition, denitriﬁcation capacity has been previously detected in sediments depleted of
nitrate and oxygen for a long period of time (Jorgensen and Tiedje, 1993; Tiedje et al.,
1982), suggesting that some denitriﬁers can survive for extended periods even in the
absence of their normal electron acceptors. Based on these observations, the following

hypotheses will be tested:

Hypothesis]
Denitriﬁer populations are distributed homogeneously within bioturbated

sediments, regardless of the redox potential.

Hypothesis 2
The pattern of expression of denitriﬁcation genes is equal for denitriﬁers at any
depth within the bioturbated sediments, when subjected to equal environmental

conditions.

12

Objective 3

To study the denitrifier diversity and distribution in a broad range of

geographic locations with varying geochemical characteristics.

As stated above, geographic location has been previously determined to exert a
strong effect in the differentiation of denitriﬁer communities in the Paciﬁc Northwest
(Braker et al., 2000; 2001). Water depth within one geographic area, which is correlated
with gradually changing biogeochemical conditions, seemed to exert a less Signiﬁcant
effect, while sediment depth was the least inﬂuential of the scales studied. However, a
broader range of geographic locations with varying geochemical characteristics needed to

be analyzed in order to further test the following hypothesis:

Hypothesis

Geographic location, water depth within one geographic area, and sediment depth
within the bioturbated zone exert decreasing inﬂuence, in that order, on the

differentiation of denitriﬁer communities.

13

References

Bethe, H., G. Jost, M. Schloter, B. B. Ward, and K.-P. Witzel. 2000. Molecular
analysis of ammonia oxidation and denitriﬁcation in natural environments. FEMS
Microbiol. Rev. 24:673-690.

Braker, G., A. Fesefeldt, and K.-P. Witzel. 1998. Development of PCR primer systems
for ampliﬁcation of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria
in environmental samples. Appl. Environ. Microbiol. 64:3769-3775.

Braker, G. and J. M. Tiedje. 2003. Nitric oxide reductase (norB) genes from pure
cultures and environmental samples. Appl. Environ. Microbiol. 69:3476-3483.

Braker, G., H. A. Ayala-deI-Rio, A. H. Devol, A. Fesefeldt, and J. M. Tiedje. 2001.
Community structure of denitriﬁers, Bacteria, and Archaea along redox gradients in
Paciﬁc Northwest marine sediments by terminal restriction fragment length
polymorphism analysis of ampliﬁed nitrite reductase (nirS) and 16S rRNA genes. Appl.
Environ. Microbiol. 67: 1 893—1 901.

Braker, G., J. Zhou, L. Wu, A. H. Devol, J. M. Tiedje. 2000. Nitrite reductase genes
(nirK and nirS) as functional markers to investigate diversity of denitrifying bacteria in

Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol. 66: 2096—
2104.

Brandes, J. A., A. H. Devol, and C. Deutsch. 2007. New developments in the marine
nitrogen cycle. Chem. Rev. 107:577-589.

Castro-Gonzalez, M., G. Braker, L. Farias, and O. Ulloa. 2005. Communities of nirS-

type denitriﬁers in the water column of the oxygen minimum zone in the eastern South
Paciﬁc. Environ. Microbiol. 7:1298-1306.

Cheneby, D., S. Hallet, M. Mondon, F. Martin-Laurent, J. C. German, and L.
Philippot. 2003. Microb. Ecol. 46:113-121.

Codispoti, L. A. 1995. Is the ocean losing nitrate? Nature 376:724.

Coyne, M. S., A. Arunakumari, B. A. Averill, and J. M. Tiedje. 1989. Immunological
identiﬁcation and distribution of dissimilatory heme ed; and nonheme copper nitrite
reductase in denitrifying bacteria. Appl. Environ. Microbiol. 55:2924-2931.

Coyne, M. S., and J. M. Tiedje. 1990. Distribution and diversity of dissimilatory NOz'
reductases in denitrifying bacteria. In N. P. a. S. Revsbech, J. (ed.), Denitriﬁcation in soil
and sediment. Plenum Press, New York, NY.

Dalsgaard, T., B. Thamdrup, and D. E. Canfield. 2005. Anaerobic ammonium
oxidation (anammox) in the marine environment. Res. Microbiol. 156:457—464.

14

Devol, A. H. 1991. Direct measurement of nitrogen gas ﬂuxes from continental shelf
sediments. Nature 349:3 19-321 .

Devol, A. H. 2003. Solution to a marine mistery. Nature 422:575-576.

Devol, A., B. Chang, and J. Christensen. 2005. Denitriﬁcation and sulfate reduction in
Arctic continental margin sediments. Goldschmidt Conference Abstracts. A581.

Flanagan, D. A., L. G. Gregory, J. P. Carter, A. Karakas-Sen, D. J. Richardson, and
S. Spiro. 1999. Detection of genes for periplasmic nitrate reductase in nitrate-respiring
bacteria and in community DNA. FEMS Microbiol. Lett. 177 :263-270.

Gamble, T. N., M. R. Betlach, and J. M. Tiedje. 1977. Numerically dominant
denitrifying bacteria from world soils. Appl. Environ. Microbiol. 33:926-939.

Ganeshram, R. S., T. F. Pedersen, S. E. Calvert, and J. W. Murray. 1995. Large
changes in oceanic nutrient inventories from glacial to interglacial periods. Nature
376:755-758.

Glockner, A. B., A. Jungst, and W. G. Zumft. 1993. Copper-containing nitrite
reductase from Pseudomonas aureofaciens is functional in a mutationally cytochrome
cdl-free background (N irS’) of Pseudomonas stutzeri. Arch. Microbiol. 160:18-26.

Goregues, C. M., V. D. Michotey, and P. C. Bonin. Molecular, biochemical, and
physiological approaches for understanding the ecology of denitriﬁcation. Microb. Ecol.
49:198-208.

Hallin, S. and P.-E. Lindgren. 1999. PCR detection of genes encoding nitrite reductase
in denitrifying bacteria. Appl. Environ. Microbiol. 65:1652-1657.

Hannig, M., G. Braker, J. Dippner, and K. Jiirgens. 2006. Linking denitriﬁer
community structure and prevalent biogeochemical parameters in the pelagial of the
central Baltic Proper (Baltic Sea). FEMS Microbiol. Ecol. 57:260-271.

Heid, C. A., J. Stevens, K. J. Livak, and P. M. Williams. 1996. Real time quantitative
PCR. Genome Res. 6:986-994.

Henry, S., D. Bru, B. Stres, S. Hallet, and L. Philippot. 2006. Quantitative detection of
the nosZ gene, encoding nitrous oxide reductase, and comparison of the abundances of
16S rRNA, narG, nirK, and nosZ genes in soils. Appl. Environ. Microbiol. 72:5181-
5189.

Henry, S., E. Baudoin, J. C. Lépez-Gutiérrez, F. Martin-Laurent, A. Brauman, and
L. Philippot. 2004. Quantiﬁcation of denitrifying bacteria in soils by nirK gene targeted
real-time PCR. J. Microbiol. Methods 59:327-335.

15

Hickey, B. M. 1989. Patterns and processes of circulation over the Washington
continental shelf and Slope. In M. R. Landry and B. M. Hickey (eds), Coastal
Oceanography of Washington and Oregon. Elsevier, New York, NY.

Jayakumar, D. A., C. A. Francis, S. W. A. Naqvi, and B. B. Ward. 2004. Diversity of
nitrite reductase genes (nirS) in the denitrifying water column of the coastal Arabian Sea.
Aquat. Microb. Ecol. 34:69-78.

Jargensen, K. S., and J. M. Tiedje. 1993. Survival of denitriﬁers in nitrate-free,
anaerobic environments. Appl. Environ. Microbiol. 59:3297-3305.

Knowles, R. 1982. Denitriﬁcation. Microbiol. Rev. 46:43-70.

Liu, W.-T., T. L. Marsh, H. Cheng, and L. J. Forney. 1997. Characterization of
microbial diversity by determining terminal restriction fragment length polymorphisms of
genes encoding 16S rRNA. Appl. Environ. Microbiol. 63:4516-4522.

Liu, X., S. M. Tiquia, G. Holguin, L. Wu, S. C. Nold, A. H. Devol, K. Luo, A. V.
Palumbo, J. M. Tiedje, and J. Zhou. 2003. Molecular diversity of denitrifying genes in
continental margin sediments within the oxygen-deﬁcient zone off the Paciﬁc coast of
Mexico. Appl. Environ. Microbiol. 69:3549-3560.

Lépez-Gutiérrez, J. C., S. Henry, S. Hallet, F. Martin-Laurent, G. Catroux, and L.
Philippot. 2004. Quantiﬁcation of a novel group of nitrate-reducing bacteria in the
environment by real-time PCR. J. Microbiol. Methods 57:399-407.

Luther 111, G. W., B. Sundby, B. L. Lewis, P. J. Brendel, and N. Silverberg. 1997.
Interactions of manganese with the nitrogen cycle: Alternative pathways to dinitrogen.
Geochim. Cosmochim. Acta 61:4043-4052.

Michotey, V., V. Méjean, and P. Bonin. 2000. Comparison of methods for
quantiﬁcation of cytochrome ch-denitrifying bacteria in environmental marine samples.
Appl. Environ. Microbiol. 66:1564-1571 .

Middelburg, J. J., K. Soetaert, P. M. J. Herman, and C. H. R. Heip. 1996.
Denitriﬁcation in marine sediments: A model study. Global Biogeochem.Cycles. 10:661-
671.

Nogales, B., K. N. Timmis, D. B. Nedwell and A. M. Osborn. 2002. Detection and
diversity of expressed denitriﬁcation genes in estuarine sediments after reverse
transcription-PCR ampliﬁcation from mRNA. Appl. Environ. Microbiol. 68:5017-5025.

Oakley, B. B., C. A. Francis, K. J. Roberts, C. A. Fuchsman, S. Srinivasan, and J. T.
Staley. 2007. Analysis of nitrite reductase (nirK and nirS) genes and cultivation reveal
depauperate community of denitrifying bacteria in the Black Sea suboxic zone. Environ.

Microbiol. 9:118-130.

16

Penton, C. R., A. H. Devol, and J. M. Tiedje. 2006. Molecular evidence for the broad
distribution of anaerobic ammonium—oxidizing bacteria in freshwater and marine
sediments. Appl. Environ. Microbiol. 72:6829-6832.

Philippot, L. 2002. Denitrifying genes in bacterial and Archaeal genomes. Biochim.
Biophys. Acta 1577 :355-376.

Philippot, L. and S. Hallin. 2005. Finding the missing link between diversity and

activity using denitrifying bacteria as a model functional community. Curr. Opin.
Microbiol. 8:234-239.

Priemé, A., G. Braker and J. M. Tiedje. 2002. Diversity of nitrite reductase (nirK and

nirS) gene fragments in forested upland and wetland soils. Appl. Environ. Microbiol.
68:1893-1900.

Qiu, X.-Y., R. A. Hurt, L.-Y. Wu, C.-H. Chen, J. M. Tiedje, and J.-Z. Zhou. 2004.
Detection and quantiﬁcation of copper-denitrifying bacteria by quantitative competitive
PCR. J. Microbiol. Methods 59: 199-210.

Santoro, A. E., A. B. Boehm and C. A. Francis. 2006. Denitriﬁer community

composition along a nitrate and salinity gradient in a coastal aquifer. Appl. Environ.
Microbiol. 72: 2102-2109.

Scala, D. J. and L. J. Kerkhof. 1998. Nitrous oxide reductase (nosZ) gene-speciﬁc PCR
primers for detection of denitriﬁers and three nosZ genes from marine sediments. FEMS
Microbiol. Lett. 162:61-68.

Scala, D. J. and L. J. Kerkhof. 1999. Diversity of nitrous oxide reductase (nosZ) genes
in continental Shelf sediments. Appl. Environ. Microbiol. 65:1681-1687.

Scala, D. J. and L. J. Kerkhof. 2000. Horizontal heterogeneity of denitrifying bacterial

communities in marine sediments by terminal restriction fragment length polymorphism
analysis. Appl. Environ. Microbiol. 66:1980-1986.

Smith, C. J., D. B. Nedwell, L. F. Dong, and A. M. Osborn. 2007. Diversity and
abundance of nitrate reductase genes (narG and napA), nitrite reductase genes (nirS and

nrfA), and their transcripts in estuarine sediments. Appl. Environ. Microbiol. 73:3612-
3622.

Tamegai, H., L. Li, N. Masui, and C. Kato. 1997. A denitrifying bacterium from the
deep sea at 11,000 m depth. Extremophiles 1:207-211.

Tamegai, H., R. Aoki, S. Arakawa, and C. Kato. 2007. Molecular analysis of the
nitrogen cycle in deep-sea microorganisms from the Nankai Trough: genes for

nitriﬁcation and denitriﬁcation from deep-sea environmental DNA. Extremophiles
11:269-275.

17

Throbﬁck, I. N., K. Enwall, A. Jarvis, and S. Hallin. 2004. Reassessing PCR primers
targeting nirS, nirK and nosZ genes for community surveys of denitrifying bacteria with
DGGE. FEMS Microbiol. Ecol. 49:401-417.

Tiedje, J. M. 1988. Ecology of denitriﬁcation and dissimilatory nitrate reduction to
ammonium. In A. J. B. Zehnder (ed.), Biology of anaerobic microorganisms. John Wiley
& Sons,Inc, New York, NY.

Tiedje, J. M. 1994. Denitriﬁers. In R. W. Weaver, J. S. Angle, and P. S. Bottomley
(eds), Methods of Soil Analysis, Part 2. Microbiological and Biochemical Properties.
Soil Sci. Soc. Amer., Madison, WI.

Tiedje, J. M., A. J. Sexstone, D. D. Myrold, and J. A. Robinson. 1982. Denitriﬁcation:
ecological niches, competition and survival. Antonie van Leeuwenhoek. 48:569-583.

Tiquia, S. M., S. Gurczynski, A. Zholi, and A. Devol. 2006. Diversity of
biogeochemical cycling genes from Puget Sound sediments using DNA microarrays.
Environmental Technology. 27: 1377-1389.

Torsvik, V., L. Ovreas, and T. F. Thingstad. 2002. Prokaryotic diversity-magnitude,
dynamics, and controlling factors. Science 296:1064-1066.

van de Graaf, A., A. Mulder, P. de Bruijn, M. S. M. Jetten, L. A. Robertson, and J.
Gijs Kuenen. 1995. Anaerobic oxidation of ammonium is a biologically mediated
process. Appl. Environ. Microbiol. 61: 1246-1251 .

Walsh, J. J. 1991. Importance of continental margins in the marine biogeochemical
cycling of carbon and nitrogen. Nature 350:53-55.

Ward, B. B. 1996. Nitriﬁcation and denitriﬁcation: probing the nitrogen cycle in aquatic
environments. Microb. Ecol. 32:247-261.

Ward, B. B. and A. R. Cockcroft. 1993. Immunoﬂuorescence detection of the
denitrifying strain Pseudomonas stutzeri (ATCC 14405) in seawater and intertidal
sediment environments. Microb. Ecol. 25:233-246.

Woodgate, R. A., K. Aagaard, and T. J. Weingartner. 2005. A year in the physical

oceanography of the Chukchi Sea: Moored measurements from autumn 1990-1991.
Deep-Sea Res. 52:31 16-3149.

Zumft, W. G. 1992. The denitrifying prokaryotes. In A. Balows, H. G. Triiper, M.
Dworkin, W. Harder, and K.-H. Schleifer (eds), The prokaryotes, 2nd ed. Springer-
Verlag, New York, NY.

Zumft, W. G. 1997. Cell biology and molecular basis of denitriﬁcation. Microbiol. Mol.
Biol. Rev. 61:533-616.

18

CHAPTER 2

PSEUDOMONAS S T U T ZERI NITRITE REDUCTASE GENE
ABUNDANCE IN ENVIRONMENTAL SAMPLES MEASURED BY
REAL-TIME PCR

Abstract

We used real-time PCR to quantify the denitrifying nitrite reductase gene (nirS), a
functional gene of biogeochemical Signiﬁcance. The assay was tested in vitro and applied
to environmental samples. The primer-probe set selected was Speciﬁc for nirS sequences
that corresponded approximately to the Pseudomonas stutzeri Species. The assay was
linear from 1 to 106 gene copies (r2=0.999). Variability at low gene concentrations did
not allow detection of twofold differences in gene copy number at less than 100 copies.
DNA spiking and cell-addition experiments gave predicted results, suggesting that this
assay provides an accurate measure of P. stutzeri nirS abundance in environmental
samples. Although P. stutzeri abundance was high in lake sediment and groundwater
samples, we detected low or no abundance of this species in marine sediment samples
from Puget Sound (Wash) and from the Washington ocean margin. These results suggest

that P. stutzeri may not be a dominant marine denitriﬁer.

Introduction

Denitriﬁcation is one of the important biogeochemical processes in that it is the
main Sink for ﬁxed nitrogen (Knowles, 1982). In agriculture, it accounts for 20 to 30% of

fertilizer losses (Firestone, 1982) and in marine environments it is thought to account for

19

up to 80% of the loss of the nitrogen load to coastal areas (Seitzinger, 1990). Two of its
products, NO and N20, are involved in global warming and the destruction of the ozone
layer. Denitriﬁcation is also used in waste treatment facilities to remove excess combined
nitrogen (Tiedje, 1988).

More accurate understanding and modeling of denitriﬁcation should be possible if
the catalyst could be quantiﬁed. We have evaluated the application of real-time PCR to
the quantiﬁcation of the nitrite reductase gene in Pseudomonas stutzeri. P. stutzeri is a
denitriﬁer commonly isolated from both soil (Gamble et al., 1977) and marine
environments (Ward and Cockroft, 1993), and may be of general importance in global
denitriﬁcation. Nitrite reductase catalyzes a key step in the nitrogen cycle, in that the
reduction of nitrite (NO2_) to nitric oxide (NO) converts N to a form no longer available
to most of the biota. This enzyme is found as two different variants. One contains copper
and is encoded by nirK, while the other contains the hemes c and d; and is coded by nirS
(Zumft, 1997). nirK is found in a wider range of physiological groups, while nirS appears
to be more abundant in nature (Coyne et al., 1989). P. stutzeri contains nirS (KOrner et
aL,1987)

Several methods have been used to attempt to identify or quantify denitriﬁers,
including the design of speciﬁc PCR primers (Braker et' al., 1998; Hallin et al., 1999) or
probes (Smith and Tiedje, 1992) for genes involved in denitriﬁcation or rDNA genes
(Kerkhof, 1994), and immunoﬂuorescent assays using polyclonal antibodies against
speciﬁc denitrifying bacteria (Ward and Cockroft, 1993) or denitriﬁcation enzymes

(Coyne et al., 1989). Recently, competitive PCR and most-probable-number PCR were

20

applied to the quantiﬁcation of nirS—containing denitrifying bacteria (Michotey et al.,
2000)

The real-time PCR technique is based on the use of the 5' nuclease assay, ﬁrst
described by Holland et al. (Holland et al., 1991) and further improved by the use of
ﬂuorescent TaqMan methodology and the ABI Prism 7700 Sequence Detection System
(PE Applied Biosystems, Foster City, CA) (Gibson et al., 1977; Heid et al., 1996). The
system requires the design of a forward and a reverse primer, in addition to a probe that
hybridizes between them. The probe is ﬂuorescently labeled at both ends (Lee et al.,
1993). The ﬂuorescent dye at the 5' end serves as reporter and its emission spectra is
quenched by the dye at the 3' end of the probe. During the elongation step of each PCR
cycle, the DNA polymerase cleaves the annealed probe, using its 5' nuclease activity.
Once separated from the quencher, the reporter ﬂuorescence is detected, resulting in an
increase in ﬂuorescence emission. The ﬂuorescence increases logarithmically as the PCR
reaction proceeds, until a reagent becomes limiting. A threshold ﬂuorescence intensity is
deﬁned within the logarithmic phase. The higher the amount of initial template DNA, the
earlier the ﬂuorescence will cross the deﬁned threshold. Copy number of the initial target
DNA is thereby determined by comparison to a standard curve.

The advantage of the real-time PCR method over other PCR-based quantiﬁcation
methods is that it focuses on the logarithmic phase of product accumulation rather than
on the end product abundance. This technique is therefore more accurate, Since it is less
affected by ampliﬁcation efﬁciency or depletion of a reagent. In addition, real-time PCR
measures template abundance over a large dynamic range of around Six orders of

magnitude (Heid et al., 1996). Finally, this method allows the Simultaneous analysis of

21

96 samples in a short time and reduces the risk of contamination, as no post-PCR
manipulation is required. The main disadvantage of real-time PCR is the need for a
special thermocycler and reagents that are expensive compared to the equipment utilized
by other PCR-based quantiﬁcation methods.

Real-time PCR has been successfully applied in the medical ﬁeld, for example in
the quantiﬁcation of various DNA and RNA viruses in patients (Gut et al., 1999,
Josefsson et al., 1999; Kimura et al., 1999; Martell et al., 1999), in the detection of gene
ampliﬁcation (Bieche et al., 1998), mutations or chromosomal rearrangements (Dblken et
al., 1998; Luthra et al., 1998) and in the quantiﬁcation of gene expression (Bieche et al.,
1999; Wang and Brown, 1999) or detection of various splice variants (Kafert et al.,
1999). Recently, real-time PCR was applied to environmental samples in studies that
quantiﬁed conidia of a human pathogenic mold in airborne samples (Haugland et al.,
1999) and for determining the abundance of bacterioplankton in marine samples (Suzuki
etaL,2000)

In this paper we test real-time PCR for its use for quantifying an important

functional gene in environmental samples.

Materials and Methods

Samples

Soil samples were obtained from an agricultural plot at the Kellogg Biological
Station (MI) (Asuming-Brempong, 1999). Groundwater samples were taken from the

Schoolcraﬁ (MI) and Shiprock Uranium Mill Tailings Remedial Action (UMTRA) (NM)

22

sites. The uranium recovery process at the latter site used high concentrations of
ammonia, some of which entered the groundwater and led to the accumulation of nitrate.
The Site is located on an elevated terrace, along the south side of the San Juan River.
Samples were taken on the terrace (samples 813 and 826), on the ﬂoodplain (samples
602, 603, and 619), and from a seep ﬂowing from the base of the escarpment into the
ﬂoodplain (sample 425). They present increasing concentrations of nitrate in the
following order: 602, 826, 619, 425, 603, and 813 (Ivanova et al., 2000). P. stutzeri KC, a
strain that hydrolyzes carbon tetrachloride (Criddle et al., 1990), was injected into the
Schoolcraft aquifer for a bioremediation ﬁeld test 15 days before the samples were taken
from wells 2 m upstream, and 1 In and 2.5 m downstream from the injection site (MS,
M11, and M19, respectively) (Hyndman et al., 2000). The freshwater sediment sample
was collected from the surface 2 cm sediment from Wintergreen Lake (MI), a small
hypereutrophic lake. The marine sediment samples were obtained from the Washington
margin of the Paciﬁc Ocean and from Puget Sound (WA). Sediment cores were Sliced

into sections described later.

Cultures

Marine P. stutzeri isolates (strains A3-5, D7-6, D9-1, E4-2, and F9-2) were
obtained by L. Wu from the Washington marine sediments (Braker et al., 2000). nirS
clones were obtained by G. Braker by ampliﬁcation of DNA extracted from Washington
margin and Puget Sound sediments using Speciﬁc primers (Braker et al., 2000). All
culture collection strains and marine isolates used in this study were grown in nutrient

broth (DIFCO, Detroit, MI). E. coli transforrnants were grown in Luria- Bertani (LB)

23

broth (Sambrook et al., 1989) amended with kanamycin (50 ug/ml). Shewanella
oneidensis MR-l, P. stutzeri KC and P. aeruginosa were grown at 30°C, while all other

cultures were grown at 37°C.

DNA extraction and quantitation

Genomic DNA was extracted from late exponential phase cultures. Cells were
harvested by centrifugation and resuspended in lysis buffer (50 mM Tris pH 8, 50 mM
EDTA, 100 mM NaCl). The cells were incubated for 15 min at 37°C with lysozyme (3.5
mg/ml), achromopeptidase (70 ug/ml), and RNAse A (30 ug/ml). After two freeze-thaw
treatments, the cell suspension was incubated for 5 min at 37°C with SDS (1%), followed
by incubation with proteinase K (400 ug/ml) (l h at 60°C). The lysate was extracted
twice with phenol/chloroform/isoamyl alcohol. After isopropanol precipitation, the DNA
was resuspended in TE buffer (pH 8). Plasmid DNA was extracted from the E. coli
transforrnants with the Wizard® Plus SV Minipreps DNA Puriﬁcation System (Promega,
Madison, WI), according to the manufacturer's instructions. The genomic DNA
extractions from the Kellogg Biological Station soil, the Shiprock aquifer, and the
Wintergreen Lake sediment were performed using the Ultra CleanTM Soil DNA Kit (MO
BIO, Solana Beach, CA), following the manufacturer's instructions. Genomic DNA was
extracted from the Schoolcraft groundwater samples by the method of van Elsas and
Smalla (van Elsas and Smalla, 1995). This method was also used to extract genomic
DNA from the Washington marine sediment samples, with an additional proteinase K

treatment (50 pl of a 20 mg/ml) after the incubation with SDS. The protocol of Gray and

24

Herwig (Gray et al., 1996) was used to extract genomic DNA from the Puget Sound
samples.

The quality of the extracted DNA was analyzed by electrophoresis through a 0.8%
agarose gel. DNA concentrations were measured by absorbance at 260 nm. P. stutzeri
nirS gene copy number was estimated based on the P. stutzeri Zobell genome Size (4.29
Mbp) (Ginard et al., 1997), and on the assumption that only one copy of nirS is present

per genome (Jﬁngst et al., 1991); i.e. 4.4 fg P. stutzeri DNA = 1 genome mm = 1 nirS

copy.

Primers and probe

Nine P. stutzeri nirS sequences from isolates (Braker et al., 2000) and 52 non-P.
stutzeri nirS sequences from marine isolates, clones, and unrelated Species (Braker et al.,
2000) were compared to select conserved regions within the P. stutzeri nirS gene. The
primers and probe were designed within conserved regions using the program
PrimerExpress (PE Applied Biosystems, Foster City, CA) (Table 2.1). The probe was
dually labelled with the ﬂuorescent dyes FAM (6-carboxyﬂuorescein) and TAMRA (6-
carboxy-tetramethyl-rhodamine) at the 5' and 3' ends, respectively, as recommended by
the manufacturers. Primers and probe were synthesized by Integrated DNA Technologies

(Coralville, IA).

Real-time PCR

The increase in ﬂuorescence emission, due to the degradation of the probe by the

DNA polymerase in each elongation step, was monitored during PCR ampliﬁcation using

25

the 7700 Sequence Detector (PE Applied Biosystems, Foster City, CA). The ﬂuorescence
signal was normalized by dividing the emission of the reporter dye (FAM) by the
emission of the passive reference dye ROX (6-carboxy-X-rhodamine). The parameter CT
(threshold cycle) is the fractional cycle number at which the ﬂuorescence emission
crosses an arbitrarily deﬁned threshold within the logarithmic increase phase (0.1 in our
reactions). The higher the amount of initial template DNA, the earlier the ﬂuorescence
will cross the threshold, and the smaller will be the CT. The CT values obtained for each
sample are compared with a standard curve to determine the initial copy number of target
gene.

The reaction mixture for real-time PCR consisted of IX TaqMan® Universal PCR
Master Mix [containing AmpliTaq GoldTM DNA Polymerase, AmpErase® UNG (uracil-
N-glycosylase), which degrades PCR carry-over products from previous reactions,
dNTPS with dUTP, a passive reference (ROX), and Optimized buffer components] (PE
Applied Biosystems), 300 nM forward primer (FP), 900 nM reverse primer (RP), and 525
nM ﬂuorogenic probe. MicroAmp® optical caps and tubes were used (PE Applied
Biosystems, Foster City, CA). A total volume of 30 ul was used for the optimization
steps and 50 pl was used for the ﬁnal reactions. PCR reaction conditions were as follows:
2 min at 50°C, 10 min at 95°C, then 40 cycles of 15 S at 95°C and 1 min at 60°C.

Negative controls with no template DNA or no probe were run in each reaction.

Specificity

The DNA extracted from several strains, marine isolates (Braker et al., 2000) and

E. coli transforrnants (Braker et al., 2000) was used as positive and negative controls to

26

test the speciﬁcity of the primer-probe set. Template DNA (18 ng) was added to each

reaction tube.

Sensitivity and detection limit

Marine isolate E4-2 DNA was chosen as a standard for measuring the sensitivity of
the primer-probe set and for generating standard curves in subsequent determinations.
This isolate was previously identiﬁed as P. stutzeri based on 16S rDNA sequence identity
and physiological characters (Braker et al., 2000). Serial dilutions (IO-fold) of P. stutzeri
E4-2 DNA were prepared in herring Sperm DNA [1 pg-ml'l herring sperm DNA
(Boehringer Mannheim, Indianapolis, IN) in water] as a carrier. All determinations were
performed in triplicate and error bars are reported as 95% conﬁdence intervals.

Various calibration curves were constructed to determine the lower detection limit
of this assay and our ability to discriminate 2-fold differences in template concentration.
A dilution series of marine isolate E4-2 DNA was prepared in a solution of 1 pg-ml'l
herring sperm DNA. Different volumes (2 pl, 4 pl, 10 pl, and 20 pl) of template DNA
from this dilution series were added to individual reaction tubes and the difference was
made up with water.

The maximum allowable error (MAE) in order to distinguish a twofold difference
in copy number was calculated using the following equation:

ACT = m - log(2)

where ACT is the difference between CT obtained from samples with a twofold

difference in target copy number, and m is the slope of the standard curve. The maximum

allowable error (MAE) is:

27

MAE = ACT / 2
MAE was calculated for each standard curve generated and the mean MAE and the

95% conﬁdence interval were determined.

Quantitation of nirS in environmental samples

Reactions were performed using 100 ng template DNA. The template copy number
was determined from CT values using a standard curve. Samples that exhibited CT values
equal to or higher than the negative controls were considered as below the detection limit.

All results were normalized to the quantity of community DNA.

Accuracy

In order to test for the presence of PCR inhibitors, 106 strain E4-2 genome copies
(4.4 ng DNA) were added to 100 ng DNA extracted from one environmental sample from
each Site. Samples with and without the addition of this positive control DNA were
compared and the percent recovery of the added genome copy number was calculated. P.
stutzeri abundance was also evaluated in microbial communities constructed from P.
stutzeri KC, P. aeruginosa and E. coli JM109 in different proportions and added to l g of
sterile quartz sand (Sigma, St. Louis, MO). Direct cell counts were obtained before
mixing using a Petroff-Hauser counting chamber. P. stutzeri KC cells were added to the
mixtures in various ratios (1, 1/10, 1/102, 1/103, 1/104, 1/105, 1/107, 1/103, and 0). Equal
cell numbers of P. aeruginosa and E. coli JM109 were added to achieve 2 x 108 cells in

each 300 pl mixture. After cell addition, the mixture was vortexed for 3 s, and DNA was

extracted using the Ultra CleanTM Soil DNA Kit (MO BIO), following the manufacturer's

28

instructions. Triplicate samples were analyzed using 100 ng of template DNA. To avoid
any problem of different extraction efﬁciencies, the P. stutzeri KC nirS gene copy
number was expressed relative to total DNA extracted. P. stutzeri, P. aeruginosa and E.
coli genome Sizes were recently measured and reported as 4.29 Mbp (Ginard et al.,
1977), 5.9 Mbp (Ratnaningsih et al., 1990) and 4.6 Mbp (Blattner et al., 1997)
respectively; their percent G+C was considered to be 63%, 67% and 50%, respectively
(Krieg et al., 1984).

To measure the accuracy of real-time estimations of P. stutzeri abundance in
environmental samples, different numbers of P. stutzeri cells were added to Wintergreen
Lake sediment and KBS soil samples. The number of total cells in the soil or sediment
samples was determined by direct counts after staining with 5-(4,6-dichlorotriazine-2-yl)
aminoﬂuorescein (DTAF) (Bloem, 1995). Serially diluted cells (108, 107, and 10'5 cells)
were added to 0.5 g sediment or soil samples. After vortexing for 5 S, total DNA was
extracted as above. In order to normalize these values to the total community DNA, the
average genome size of soil or sediment bacteria was considered equal to the E. coli

genome size, or 4.6 Mbp (Blattner et al., 1997).

Results

Speciﬁcity

DNA extracted from a range of denitrifying isolates was used to test the speciﬁcity
of the primer—probe combination (Table 2.2). Logarithmic increase in ﬂuorescence

intensity was readily detected by real-time PCR in 17 of 21 P. stutzeri strains. However,

29

we were not able to detect by real-time PCR the P. stutzeri strains corresponding to
genomovars 4, 5, 7, and one strain in genomovar 1. Sequence analysis of nirS of the
strains from genomovars 4, 5, and 7 indicated a higher Similarity (80%, 81%, and 82%,
respectively) with the nitrite reductase gene of P. aeruginosa than with P. stutzeri. We
identiﬁed several missmatches between these strains and the primer-probe combination
used in this study (5 to 9 for the primers and 4 to 5 for the probe). No non-P. stutzeri

strain or clone gave a real-time PCR signal.

Sensitivity and detection limit

We tested the sensitivity of the real-time detection system using a dilution series of
P. stutzeri DNA (Fig. 2.1A). The threshold value for this and all subsequent analyses was
chosen to be 0.1. This value falls within the range of logarithmic ﬂuorescence increase,
yet avoids the signal from the no template control (open circle). Nearly all of our no
template controls Showed some increase in ﬂuorescence intensity similar to that in Fig.
2.1A. Since this increase is not logarithmic and similar signals were detected in control
reactions without DNA polymerase (data not shown), this ﬂuorescence increase is likely
due to probe degradation. The data obtained were used to draw a standard curve relating
CT values to the added mass of P. stutzeri DNA and number of gene copies (Fig. 2.1B). A
linear response was observed over more than six orders of magnitude, ranging from 14 to
4.05 x 106 nirS gene copies (r2 = 0.999, Fig. 2.13).

We constructed a series of calibration curves to study the detection limit of the
system and our ability to differentiate similar P. stutzeri DNA concentrations. Different

volumes (ranging from 2 pl to 20 pl) of a dilution series of template DNA (ranging from

30

0.1 to 1000 nirS gene copies - pl") were added to the PCR reaction. Given the 96 well
capacity, the analyses were done as a low and high concentration set (Fig. 2.2A and 2.2B,
respectively). Although the curves remained linear down to 1 nirS copy (20 pl of 0.05
nirS gene copies - pl") the variability associated with CT values from low copy numbers
preclude our ability to distinguish concentrations in samples with Similar amounts. Only
when the copy number was 100 or greater were we able to reliably differentiate a 2- fold
difference in P. stutzeri nirS concentration.

The increase in variability with cycle number is apparent when the upper 95%
conﬁdence interval from all determinations is plotted against CT (Fig. 2.3). Based on the
Slope of each individual standard curve, we calculated the maximal error in the CT value
that would Still allow detection of a two-fold difference in gene copy number (MAE).
The average of the different MAE is equal to 0.53 (Fig. 2.3). The corresponding 95%

conﬁdence interval is too small to be observed (Fig. 2.3).

Accuracy

P. stutzeri KC, E. coli JM109, and P. aeruginosa cells were added to sterile quartz
sand in various known amounts. P. stutzeri KC contains the targeted nirS gene while P.
aeruginosa has a nirS 67% similar in nucleotide sequence. The correlation between the
calculated and measured values was extremely high (slope = 0.98, r2 = 0.992) (Fig. 2.48).
The lowest proportions of nirS (1/107 and 1/108) overlapped with the background in the
real—time PCR measurements.

To further test the accuracy of the system, a Similar experiment was performed

adding known amounts of P. stutzeri cells to different soil and sediment samples.

31

Measured values of nirS were correlated well (slope = 0.97, r2 = 0.971) with the values

calculated from direct cell counts (Fig. 2.5).

Analyses of environmental samples

DNA extracted from environmental samples exhibited a wide range of P.stutzeri
nirS gene abundance measured by real-time PCR (Table 2.3). The groundwater and
freshwater sediment samples had the highest P. stutzeri nirS copies. Marine sediments
consistently displayed low P. stutzeri nirS abundance.

In order to test for the presence of any PCR inhibitors in the environmental
samples, 106 genome copies of strain E4-2 were added to each sample. Five of the Six
samples yielded the expected amount of nirS (Fig. 2.6). The KBS soil sample Showed

slightly less nirS copies than expected.

32

Table 2.1. Primer and probe sequences compared to example positive and negative
control nirS gene sequences

 

 

Example nirS Forward primerb Probeb Reverse primerb
sequence
5'ACAAGGAGCACAAC FAM- 5'CGCGTCGGCCCAGA3'
TGGAAGGT3' S'GGCAACCTGTTCGTCAAG

ACCCA3'-TAMRA

5'ACAAGGAGCACAAC 5'GGCAACCTGTTCGTCAAG 5'CGCGTCGGCCCAGA3'

Posrtrve control, T G G A AG GT3' ACCCA3'

P. stutzeri Zobell

la 5'ATCCGQAGIACG_CCT 5'GGCTCGCTGTTCATCAAGA S'QCACAGCIGQATCTT

Negative contro GGAAcgr CCCA3'

P. aeruginosa
”Underlined bases represent missmatches with the primer-probe set.

(Corresponding to P. stutzeri Zobell nirS positions 1260 to 1281, 1310 to 1332, and 1350 to 1363
(forward primer, probe, and reverse primer, respectively).

33

Table 2.2. Speciﬁcity of real-time PCR reactions for Pseudomonas stutzeri DNA

 

 

Species, strain, or Reference Denitriﬁ- nirS‘ Real-time
clonel (genonovar) or sourceb cation‘ PCR"
Positive controls

P- stutzeri (1) ATCC 17539 + + +
P- stutzeri (l) ATCC 27951 + + +
P. stutzeri (l) ATCC 17593 + + +
P. stutzeri (l) ATCC 17594 + + +
P. stutzeri (1) CCUGl 1256 + -
P. stutzeri (2) ATCC 17591 + + +
P. stutzeri (2) ATCC 17587 + + +
P. stutzeri Zobell (2) ATCC 14405 + + +
P. stutzeri (2) ATCC 17592 + + +
P. stutzeri (2) ATCC 17595 + + +
P. stutzeri (3) ATCC 50227 + + +
P. stutzeri l9SMN4 (4) DSM 6084 + + -
P. stutzeri DNSP21 (5) DSM 6082 + + -
P. stutzeri (7) DSM 50238 + + _
P. stutzeri JM300 (8) DSM 10701 + + +
P. stutzeri KC ATCC 55595 + + +
P. stutzeri A3-5 L. Wu + + +
P. stutzeri D7-6 L. Wu + + +
P. stutzeri D9-1 L. Wu + + +
P. stutzeri E4-2 L. Wu + + +
P. stutzeri F 9-2 L. Wu + + +
Negative controls

E. coli K12 DSM 498 - - -
Paracoccus denitrificans ATCC 17741 -
Azospirillum brazilense DSM 1690 -
Shewanella oneidensis MR-l ATCC 700550 - - -
Pseudomonas aeruginosa ATCC 15692 + + -
Pseudomonas balearica DSM 6083 + + -
Marine isolate CIO-ls L. Wu + + -
Marine isolate D4-14 L. Wu + + -
nirS clone A4 G. Braker + + -
nirS clone B6 G. Braker + + -
nirS clone A12 G. Braker + + -
nirS clone B76 G. Braker + + -

 

”Marine isolates and nirS clones were obtained from Puget Sound marine sediment (7).
”ATCC, American Type Culture Collection; DSM, Deutsche Sammlung von Mikroorganismen; CC UG, Culture Collection,

University of Goteborg; L. Wu, Oak Ridge National Laboratory, Oak Ridge, Tenn; G. Braker, MSU, East Lansing, Mich.

cPresence of denitriﬁcation and nirS from literature sources, except for the characterized genomovars, which were determined in

this study by PCR.

”4r, logarithmic ampliﬁcation readily detected; -, no ampliﬁcation detected.

34

Figure 2.1. Generation of standard curve.

(A) Increase of ﬂuorescence intensity with cycle number for serially diluted P. stutzeri
DNA. Symbols, from leﬁ to right: V , 17.8 ng; e , 5.9 ng; A , 0.59 ng; O , 59 pg; I , 5.9
pg; V , 0.59 pg; 0 , 59 fg; o , no template control. CT, Cycle at which the ﬂuorescence
intensity crosses an arbitrary threshold value. (B) Standard curve. Values represent means
i 95% conﬁdence interval (n=3).

35

Fluorescence intensity (ARn)

Threshold cycle (CT)

 

101 -

10° -

 

104

 

 

 

40

 

Cycle number

 

35 -

30 -

25 -

20'-

15 -

 

10

8:0.999
y= -3.46x + 38.84

 

 

 

100

l f I I

101 1O2 1O3 104 105 106 107
Number of P. stutzeri nirS gene copies

 

I I I I I

104 10'3 10'2 104 10° 10‘
ng P. stutzeri DNA

36

Figure 2.2. Limit of detection of the P. stutzeri nirS gene using real-time PCR.

The denoted volumes of serially diluted P. stutzeri DNA were added to different
reactions. (A), low concentrations, (B) high concentrations. Values represent means i
95% conﬁdence interval (n=3).

37

Threshold cycle (CT)

Threshold cycle (CT)

 

 

 

 

 

 

   
 

 

45 - A.
40 ‘ Volume added
.~~ to reaction
35 - ‘§£‘ i
. ~ 2 pl
‘ 4 pl
30 - ‘ 10 pl
20 pl
25 -
o -1 1 10 100
P. stutzeri nirS gene copies-pl DNA'1
45 - B.
40 -
Volume added
35 I to reaction
4 pl
10 pl
25 - 20 111
1 10 100 1000

P. stutzeri nirS gene copies-pl DNA'1

38

 

 

 

 

 

 

 

 

 

I- 10

O

a...

0

Tu 8‘ o

E

.15 6~ ‘

o

‘c’

o 4~

'0

I0:

8

o 2"

3?: 0.53

o: o-

a“)

S.

3 I I I
10 20 30 40

Threshold cycle (CT)

Figure 2.3. Relationship between threshold cycle (CT) and error.

The horizontal line represents the maximum allowable error to discriminate between
samples containing a twofold difference in P. stutzeri nirS cOpy number. Value Shown
represents the mean from 14 independent standard curves. The 95% conﬁdence interval
is Shown as a dashed line.

39

 

 

 
 
 

109 —.
<1
5 B.
108
107 i '8‘
S 51
2 cl
.‘Q
106 .g'
E‘ '
Ti . i
105 u.‘ 10! 10 1o- 10-

 

 

 

   
 

105 10' 1
Calculated

P. stutzeri nirS gene coples x pg DNA"

104
103
102
10‘
10°

 

 

 

 

 

 

P. stutzeri nirS gene copies x pg DNA'1

1 1/10 1/102 1/103 1/104 1/105 1/107 1/108

Ratio of P. stutzeri cells in mixture

Figure 2.4. Quantiﬁcation of nirS in artiﬁcial mixtures of P. stutzeri KC, P.
aeruginosa, and E. coli cells.

(A) Comparison between P. stutzeri nirS gene copies - pg DNA" measured by real-time
PCR (gray bars) and calculated values based on cell counts (black bars). (B) Correlation
between calculated and measured values (slope = 0.98, r2 = 0.992). Error bars represent
95% conﬁdence intervals (n=3) for both ﬁgures.

40

 

r2=0.971
1o8 4 y=0.97x + 0.60

 

 

 

 

'<
z
a
U)
:3.
X
(D
.02
'U Q.
93 8
E E
g 8, 107-
5g
2
't
0)
Is
5
"f 106 . .
Q 106 107 108

Calculated
P. stutzeri nirS gene copies x pg DNA'1

Figure 2.5. Quantiﬁcation of nirS in KBS soil (0) and Wintergreen Lake sediment
(A) samples spiked with P. stutzeri cells.

The line Shows the correlation between P. stutzeri nirS gene copies - pg DNA'l measured
by real-time PCR and calculated values. Error bars represent 95% conﬁdence intervals

(n=3).

41

Table 2.3. P. stutzeri nirS gene abundance in various habitats.

 

 

 

Habitat Sample No. of copies of P. stutzeri nirS P. stutzeri nirS DNA (p g
no. or DNA/pg of community DNAf of community DNA)
type
Mean Upper 95% Mean Upper 95% C1
C1
Soil“
2 .4
KBS (Mich) ‘1): 1&1} 4.331?
10x BDL BDL
100x BDL BDL
Groundwater”
Schoolcraft $4181 1 113 D116” 4131):}?
Bioremediation Site M19 2' 0x 107 ' 8x9
(Mich.)° - X
Shipmc" UMTRAC Site 425 1.921103 3.721103 8.421103 1621102
(N- Mex-1 602 3.121105 8.821104 1.4 3.92110'I
603 2.221105 2.121104 9.72110'1 9321102
619 2.01107 9.4x105 90 4.1
813 5.0x106 4.521105 22 2
826 5.021105 1.621105 2.2 6.92110"
Freshwater sediment”
Wintergreen Lake (MICII.) 2.2x106 3.1x105 9.7 1.3
Marine sediment‘
Paciﬁc Ocean (Wash) 0-0.5 cm BDL BDL
0.5-1 cm BDL BDL
1-2 cm 1.021102 2.011102 4.42110'4 8.8x10“'
2-3 cm BDL BDL
3-5 cm BDL BDL
5—10 cm BDL BDL
Puget Sound (Wash) 1-1.5 cm BDL BDL
1.52 cm 2.121104 2.421103 9.12110'2 1.02110'2
6-6.5 cm 7.0x102 2.021102 3.12110'3 8.82110“

 

“Soils were treated with 0, l, 10, and 100x the normal ﬁeld rate (1.1 kg ha") of the herbicide 2,4-dichlorophenoxyacetate (1).
”Samples were collected 2 m upstream (M8), 1 m downstream (M11) and 2.5 m downstream (M19) from the well where P. stutzeri
KC was injected (21).

° UMTRA, Uranium Mill Tailings Remedial Action

° The surface 2 cm containing the active denitriﬁers were analyzed

' Depths within the sediment core are reported.

I Results are from triplicate samples when 95% conﬁdence interval (CI) is indicated and from single samples, otherwise.

‘ BDL, below detection limit.

42

107

 

106 ..

P. stutzeri nirS copy number

 

  

103 —

 

 

 

KBS SCH SHI WIN WAS PUG
Samples

Figure 2.6. P. stutzeri nirS copy number measured by real-time PCR without (black
bars) and with (gray bars) the addition of 10'5 P. stutzeri nirS copy numbers to DNA
extracted from different environmental samples.

KBS, soil from Kellogg Biological Station (Ml); SCH, groundwater from Schoolcraft
Bioremediation Site (MI); SHI, groundwater from Shiprock (NM); WIN, freshwater
sediment from Wintergreen Lake (MI); WAS, marine sediment from Washington margin
(WA); PUG, marine sediment from Puget Sound (WA).

43

Discussion

Good detection methods share four features: speciﬁcity, sensitivity, precision, and
accuracy. Real-time PCR appears to satisfy these requirements. The primer-probe set we
designed for the P. stutzeri nirS gene ampliﬁed a group of strains that generally
corresponded to the P. stutzeri species (Table 2.2). Except for one strain, all the
representatives of the two more commonly isolated genomovars, l and 2, were readily
ampliﬁed. The sequence Similarity of the nirS gene of the P. stutzeri strains in
genomovars 4, 5, and 7 to the P. aeruginosa nitrite reductase gene explains the inability
to detect them by real—time PCR. In addition to cultured strains, we selected a range of
isolates and cloned nirS sequences from the Paciﬁc Northwest marine environment for
the primer-probe design. This makes our system especially appropriate for application in
this environment. Although there may be cross-reactivity with related DNA in the natural
microbial communities, none of our habitat-speciﬁc negative controls gave a positive
reaction. Furthermore, the phylogenetically closest relative, P. balearica, was not
detected. The need for the probe in real-time PCR requires the identiﬁcation of three
speciﬁc regions in the DNA sequence, rather than two, which provides for the high
speciﬁcity. This however, can make the design of an appropriate primer-probe set more
difﬁcult. The 16S rRNA gene would be an alternative target that is more conserved but
would sample a larger organismal group that may not correlate with function.

The method was linear over more than Six orders of magnitude and sensitive down
to 1 gene copy, Similar to the results obtained in other studies (Haugland et al., 1999;
Kimura et al., 1999). However, the high variability associated with low target copies

limits precision near the detection limit (Fig. 2.2). We are able to detect the presence of

44

only 1 copy, but not precisely, e.g. we cannot discriminate a 2-fold difference at 100
copies or less. The theoretical upper 95% conﬁdence interval to allow the discrimination
of a two-fold difference in copy number is 0.53 (Fig. 2.3). Conﬁdence intervals below
this value are achieved more frequently at lower CT values, equivalent to higher number
of target copies. The error values associated with the measurements of P. stutzeri nirS
abundance in various environmental samples is consistent with this characterization of
precision, i.e. high concentrations have an error value at least one order of magnitude
smaller than the measured value, while the error was in the same order of magnitude
when nirS copy number was small (Table 2.3).

The method proved to be extremely accurate. When P. stutzeri was mixed in
various proportions with E. coli and P. aeruginosa cells, it could be detected when
present at 100% to 0.001% in the mixture, presenting a high correlation between
expected and measured results over this whole range. Furthermore, spiked samples gave
the expected results.

Environmental samples analyzed with real-time PCR displayed a wide range of P.
stutzeri nirS abundances. P. stutzeri was highly abundant in freshwater sediment from
Wintergreen Lake (MI), which agrees with a study by Gamble et al. (1977), who
identiﬁed P. stutzeri as the dominant denitriﬁer in these sediments. P. stutzeri was either
absent or present at extremely low population densities in marine sediments from the
Washington margin and Puget Sound (WA). Ward and Cockcroft (Ward and Cockroft,
1993) measured the abundance of P. stutzeri in the water column of Monterrey Bay, CA,
and found that it represented only 0.02 to 0.08% of the total bacterial community. Our

ﬁndings indicate that P. stutzeri is also not abundant in marine sediments. This is in

45

agreement with the studies by Braker et a1. (2000), who observed nirS heterogeneity and
the lack of a dominant group in nirS clones isolated from these Sites. Furthermore, these
results are also consistent with studies done on other denitriﬁcation genes. Scala and
Kerkhof (1999) Observed a high diversity among nitrous oxide reductase (nosZ) genes in
marine sediments, with no overlap between environmental nosZ sequences and cultured
denitriﬁers. Groundwater samples from the Schoolcraft Bioremediation site 041)
followed expected trends, corresponding with injection of substrates and P. stutzeri strain
KC into a contamination plume. We detected no P. stutzeri upgradient from the
inoculation site, and found the highest P. stutzeri abundance just downgradient from the
inoculation Site. P. stutzeri was also prevalent at Shiprock perhaps resulting from the high
nitrate contamination at that Site.

Real-time PCR is a speciﬁc, sensitive, precise and accurate method for quantifying
a gene or organism group in a broad range of environmental sample types. It remains to
be seen whether functional gene sequences are conserved enough in a habitat so that a
reasonable number of probe-primer sets can provide useful quantitative information for
components of a group or process. This same method and primer-probe regions should be
effective in quantifying mRNA and hence in determining which group of organisms or

gene families are active in denitriﬁcation.

Acknowledgements

This work was supported by Department of Energy grants DE-FG02-98ER62535

and DE-FG02-97ER62469. We kindly thank Dr. Jorge Lalucat for providing the P.
stutzeri strains with genomovar classiﬁcations, Dr. Allan Devol for collecting the Puget

Sound and Paciﬁc Ocean sediment samples, Dr. Phillip Long and the UMTRA personnel

46

for the Shiprock groundwater samples, Liyou Wu for the marine isolates, Dr. Gesche
Braker for the nirS clones and for providing us DNA extracted from the Paciﬁc Ocean
sediments, and Katrina Linning for DNA extracted from the Schoolcraft samples. Oak
Ridge National Laboratory is managed by University of Tennessee-Battelle LLC for the

Department of Energy under contract DE-AC05-000R22725.

47

References

Asuming-Brempong, S. 1999. The effect of 2,4-dichlorophenoxyacetate selection on
microbial communities in microcosm and ﬁeld studies and the impact on ecosystem
function. Ph.D. thesis. Michigan State University, East Lansing.

Bieche, I., I. Laurendeau, S. Tozlu, M. Olivi, D. Vidaud, R. Lidereau, and M.
Vidaud. 1999. Quantitation of MYC gene expression in sporadic breast tumors with a
real-time reverse transcription-PCR assay. Cancer Res. 59:2759-2765.

Bieche, 1., M. Olivi, M.-H. Champeme, D. Vidaud, R. Lidereau, and M. Vidaud.
1998. Novel approach to quantitative polymerase chain reaction using real-time

detection: application to the detection of gene ampliﬁcation in breast cancer. Int. J.
Cancer 78:661-666.

Blattner, F. R., G.3. Plunkett, C. A. Bloch, N. T. Perna, V. Burland, M. Riley, J.
Collado-Vides, J. D. Glasner, C. K. Rode, G. F. Mayhew, J. Gregor, N. W. Davis, H.
A. Kirkpatrick, M. A. Goeden, D. J. Rose, B. Man, and Y. Shao. 1997. The complete
genome sequence of Escherichia coli K-12. Science 277: 1453-1474.

Bloem, J. 1995. Fluorescent staining of microbes for total direct counts, p.1-12. In
A.D.L. Akkermans, J .D. van Elsas, and F.J. de Brujin (ed.), Molecular microbial ecology
manual, Kluwer Academic Publishers, Dordrecht, The Netherlands.

Braker, G., A. Fesefeldt, and K.-P. Witzel. 1998. Development of PCR primer systems
for ampliﬁcation of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria
in environmental samples. Appl. Environ. Microbiol. 64:3769-3775.

Braker, G., J. Zhou, L. Wu, A. H. Devol, and J. M. Tiedje. 2000. Nitrite reductase
genes (nirK and nirS) as functional markers to investigate diversity of denitrifying

bacteria in Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol.
66:2096-2104.

Coyne, M. S., A. Arunakumari, B. A. Averill, and J. M. Tiedje. 1989. Immunological
identiﬁcation and distribution of dissimilatory heme cdl and nonheme copper nitrite
reductase in denitrifying bacteria. Appl. Environ. Microbiol. 55:2924-2931.

Criddle, C. S., J. T. DeWitt, D. Grbic-Galic, and P. McCarty. 1990. Transformation
of carbon tetrachloride by Pseudomonas Sp. strain KC under denitriﬁcation conditions.
Appl. Environ. Microbiol. 56:3240-3246.

Diilken, L., F. Schiller, and G. Diilken. 1998. Quantitative detection oft(14;18)-positive
cells by real-time quantitative PCR using ﬂuorogenic probes. BioTechniques 25:1058-
1064.

48

Firestone, M. K. 1982. Biological denitriﬁcation, p.289-326. In F.J. Stevenson (ed.),
Nitrogen in agricultural soils. Agronomy monograph 22., American Society for
Agronomy, Madison, Wis.

Gamble, T. N., M. R. Betlach, and J. M. Tiedje. 1977. Numerically dominant
denitrifying bacteria from world soils. Appl. Environ. Microbiol. 33:926-939.

Gibson, U. E. M., C. A. Heid, and P. M. Williams. 1996. A novel method for real time
quantitative RT-PCR. Genome Research 6:995-1001.

Ginard, M., J. Lalucat, B. Tiimmler, and U. Riimling. 1997. Genome organization of
Pseudomonas stutzeri and resulting taxonomic and evolutionary considerations. Int. J.
Syst. Bacteriol. 47: 1 32-143.

Gray, J. P. and R. P. Herwig. 1996. Phylogenetic analysis of the bacterial communities
in marine sediments. Appl. Environ. Microbiol. 62:4049-4059.

Gut, M., C. M. Leutenegger, J. B. Huder, N. C. Pedersen, and H. Lutz. 1999. One-
tube ﬂuorogenic reverse transcription-polymerase chain reaction for the quantitation of
feline coronaviruses. J. Virol. Methods 77:37-46.

Hallin, S., Lindgren, and P.-E. 1999. PCR detection of genes encoding nitrite reductase
in denitrifying bacteria. Appl. Environ. Microbiol. 65:1652-1657.

Haugland, R. A., S. J. Vesper, and L. J. Wymer. 1999. Quantitative measurement of
Stachybotrys chartarum conidia using real time detection of PCR products with the
TaqManTM ﬂuorogenic probe system. Mol. Cell. Probes 13:329-340.

Heid, C. A., J. Stevens, K. J. Livak, and P. M. Williams. 1996. Real time quantitative.
PCR. Genome Res. 6:986-994.

Holland, P. M., R. D. Abramson, R. Watson, and D. H. Gelfand. 1991. Detection of
Speciﬁc polymerase chain reaction product by utilizing the 5'-3'exonuclease activity of
T hermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. 88:7276-7280.

Hyndman, D. W., M. J. Dybas, D. Wiggert, X. Zhao, R. Wallace, T. Voice, A. Chan,
M. S. Phanikumar, and C. S. Criddle. 2000. Hydraulic characterization and design of a
full-scale biocurtain. Ground Water 38:462-474.

Ivanova, I. A., J. R. Stephen, Y.-J. Chang, J. Briiggemann, P. E. Long, J. P.
McKinley, G. A. Kowalchuk, D. C. White, and S. J. Macnaughton. 2000. A survey of
168 rRNA and amoA genes related to autotrophic ammonia-oxidizing bacteria of the [3-

subdivision of the class proteobacteria in contaminated groundwater. Can. J. Microbiol.
46:1012-1020.

Josefsson, A., K. .Livak, and U. Gyllenstein. 1999. Detection and quantitation of

human papillomavirus by using the ﬂuorescent 5' exonuclease assay. J. Clin. Microbiol.
37:490-496.

49

Jiingst, A., S. Wakabayashi, H. Matsubara, and W. G. Zumft. 1991. The nirSTBM
region coding for cytochrome cdl-dependent nitrite respiration of Pseudomonas stutzeri

consists of a cluster of monO-, di-, and tetraheme proteins. FEBS Lett. 279 :205-209.

Kafert, S., J. Krauter, A. Ganser, and M. Eder. 1999. Differential quantitation of
alternatively spliced messenger RNAS using isoform-Speciﬁc real-time RT-PCR. Anal.
Biochem. 269:210-213.

Kerkhof, L. 1994. A species-speciﬁc probe and a PCR assay for the marine bacterium,
Pseudomonas stutzeri strain Zobell. Microb. Ecol. 27 :201-212.

Kimura, H., M. Morita, Y. Yabuta, K. Kuzushima, K. Kato, S. Kojima, T.
Matsuyama, and T. Morishima. 1999. Quantitative analysis of Epstein-Barr virus load
by using a real-time PCR assay. J. Clin. Microbiol. 37 :132-136.

Knowles, R. 1982. Denitriﬁcation. Microbiol. Rev. 46:43-70.

Ktirner, H., K. Frunzke, K. Dohler, and W. G. Zumft. 1987. Immunochemical
patterns of distribution of nitrous oxide reductase and nitrite reductase (cytochrome cdl)
among denitrifying pseudomonads. Arch. Microbiol. 148:20-24.

Krieg, N. R. and J. G. Holt (Eds.). 1984. Bergey's Manual of Systematic Bacteriology.
Williams & Wilkins, Baltimore, MD.

Lee, L. G., C. R. Connell, and W. Bloch. 1993. Allelic discrimination by nick-
translation PCR with ﬂuorogenic probes. Nucleic Acids Res. 21:3761-3766.

Luthra, R., J. A. McBride, F. Cabanillas, and A. Sarris. 1998. Novel 5' exonuclease-
based real-time PCR assay for the detection of t(14;l8)(q32;q21) in patients with
follicular lymphoma. Am. J. Pathol. 153263-68.

Martel], M., J. Gémez, J. I. Esteban, S. Sauleda, J. Quer, B. Cabot, R. Esteban, and
J. Guardia. 1999. High-throughput real-time reverse transcription-PCR quantitation of
hepatitis C virus RNA. J. Clin. Microbiol. 37:327-332.

Michotey, V., V. Méjean, and P. Bonin. 2000. Comparison of methods for
quantiﬁcation of cytochrome cdl-denitrifying bacteria In environmental marine samples.
Appl. Environ. Microbiol. 66: 1564- 1571.

Ratnaningsih, E., S. Dharmsthiti, V. Krishnapillai, A. Morgan, M. Sinclair, and B.
W. Holloway. 1990. A combined physical and genetic map of Pseudomonas aeruginosa
PAO. J. Gen. Microbiol. 136:2351-2357.

Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory
manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Scala, D. J. and L. J. Kerkhof. 1999. Diversity of nitrous oxide reductase (nosZ) genes
in continental Shelf sediments. Appl. Environ. Microbiol. 65 :1681-1687.

50

Seitzinger, S. P. 1990. Denitriﬁcation in aquatic sediments, p.301-322. In N.P.
Revsbech and J. Sorensen (ed.), Denitriﬁcation in soil and sediment, FEMS Symposium,
Plenum Press, New York.

Smith, G. B. and J. M. Tiedje. 1992. Isolation and characterization of a nitrite reductase

gene and its use as a probe for denitrifying bacteria. Appl. Environ. Microbiol. 58:376-
384.

Suzuki, M. T., L. T. Taylor, and E. F. Delong. 2000. Quantitative analysis of small-
subunit rRNA genes in mixed microbial populations Via 5'-nuclease assays. Appl.
Environ. Microbiol. 66:4605-4614.

Tiedje, J. M. 1988. Ecology of denitriﬁcation and dissimilatory nitrate reduction to
ammonium, p.179-244. In A.J.B. Zehnder (ed.), Biology of anaerobic microorganisms,
John Wiley & Sons,Inc, New York, NY.

van Elsas, J. D. and K. Smalla. 1995. Extraction of microbial community DNA from

soils, p.1-11. In A.D.L. Akkerrnans, J.D. van Elsas, and F .J. de Brujin (ed.), Molecular
microbial ecology manual, Kluwer Academic Publishers, Dordrecht, The Netherlands.

Wang, T. and M. J. Brown. 1999. mRNA quantiﬁcation by real time TaqMan

polymerase chain reaction: validation and comparison with RNase protection. Anal.
Biochem. 269: 1 98-201.

Ward, B. B. and A. R. Cockcroft. 1993. Immunoﬂuorescence detection of the
denitrifying strain Pseudomonas stutzeri (ATCC 14405) in seawater and intertidal
sediment environments. Microb. Ecol. 25:233-246.

Zumft, W. G. 1997. Cell biology and molecular basis of denitriﬁcation. Microbiol. Mol.
Biol. Rev. 61:533-616.

51

CHAPTER 3

DENITRIFIER COMMUNITY STRUCTURE AND ACTIVITY WITH
RESPECT TO REDOX GRADIENTS AND BIOTURBATION

Abstract

Denitriﬁers have previously been detected at sediment depths below the penetration
depths of oxygen and nitrate, their normal electron acceptors. We were therefore
interested in studying the vertical distribution of denitriﬁers, with respect to redox
gradients and presence of macrobenthic fauna, which could redistribute bacteria within
the sediments and allow the penetration of overlying water deep into the sediments. Three
highly bioturbated Sites in Puget Sound, Washington, were compared to a Washington
Margin area with lower bioturbation. Despite the shallow oxygen and nitrate penetration
at the studied sites (<2cm), the heme cd; nitrite reductase gene (nirS) containing
community presented a conserved structure, as determined by T-RFLP, within the
bioturbated sediments (to at least 37 cm and ~15 cm below the sediment surface depth in
Puget Sound and Washington Margin samples, respectively), with a Sharp change in
community structure in deep unbioturbated sediments. PCA ordination and cluster
analysis grouped the sediments based on sampling location and mixing, with unmixed
sediments from all areas forming a separate group. Quantiﬁcation by real-time PCR of
clusters of nirS clones retrieved from these same sediments followed this same pattern,
with signiﬁcant reductions of abundance in deep unmixed sediments. On the other hand,
P. stutzeri was generally present at highly constant abundance throughout the sediment

depths within as well as below the bioturbated zone. Denitriﬁcation capacity, determined

52

as nitrous oxide production in microcosms after nitrate amendment, was detected
throughout the mixed layers at all Sites, conﬁrming the presence of denitriﬁers capable of

denitriﬁcation in sediments below the depth where nitrate is present in situ.

Introduction

Many near shore and continental shelf and slope areas are characterized by high
carbon ﬂux to the sediments, due to their high surface productivity and Shallow water
depth, leading to a high sedimentary respiration rate (Brandes and Devol, 1995). Puget
Sound and Washington continental margin sediments are an example of this
phenomenon. The high respiration rate is responsible for a rather shallow penetration
depth of nitrate, generally in the order of 1 cm (Braker et al., 2000; Brandes and Devol,
1995). Therefore, active denitriﬁcation is expected to occur in situ only in the top 1 cm,
as the presence of an N-oxide is generally required for the expression of the
denitriﬁcation genes (Zumft, 1997). Braker et al., however, were able to clone nirS gene
sequences from 6.5 cm deep sediments from Puget Sound (Braker et al., 2000) and
detected a highly Similar denitriﬁer community structure, as determined by T-RFLP, in
the top 6.5 and 10 cm of Puget Sound and Washington continental shelf sediments,
respectively (Braker et al., 2001). They suggested that sediment mixing by marine
invertebrates could have been responsible for the conserved community structure at
different depths despite the unfavorable electron acceptor gradients. Wadden Sea
sediments also presented a similar microbial community structure down to 4.75 cm when
major phylogenetic groups were studied by ﬂuorescent in situ hybridization (FISH),

regardless of the redox gradients (Llobet-Brosa et al., 1998). Furthermore, in Tokyo Bay

53

sediments, only minor bacterial groups changed in the top 10 cm, while the major
population structure remained stable, as revealed by T-RFLP and quinone proﬁling
(Urakawa et al., 2000). None of these studies, however, included deep sediments lying
below the bioturbated zone.

Besides detection of denitriﬁcation genes in nitrate depleted sediments,
denitriﬁcation capacity has also been detected in environments devoid of nitrate for a
long period of time, c. g. nitrate-free sludge, eutrophic lake sediments and river sediments
(Jorgensen and Tiedje, 1993; Tiedje et al., 1982). Oxygen was also absent from these
environments, hence, no aerobic respiration could take place. A low level of fermentation
was suggested as an explanation for the long-term survival of these denitriﬁers in the
absence of nitrate and oxygen (Jorgensen and Tiedje, 1993).

Puget Sound sediments support a dense benthic population, responsible for a high
sediment mixing rate of 43 cm2 y". Surface mixed layers of up to 22 cm have been
observed in these sediments by the analysis of excess me activity proﬁles (Carpenter et
al., 1985). On the other hand, in Washington slope sediments, though variable, surface
sediment mixing coefﬁcients are markedly lower than in Puget Sound (0.25-9.8 cm2 y'l),
with average depths of surface mixed layers of 6.2 :E 3.5 cm (Carpenter et al., 1982).
Furthermore, the sediment mixing coefﬁcients in this area suffer a particularly sharp
decrease below a water depth of 800 to 1400 m (<1 cm2 y'l). The low dissolved 02
concentration in waters over the Slope from about 600 to 1400 m probably accounts for
this decrease due to the inhibition of macrobenthic activity (Carpenter and Peterson,
1989). This clear difference in sediment mixing makes Puget Sound and the low 02 zone

of the Washington Slope ideal sites for comparison of sediments under different

54

bioturbation intensities. These Sites were therefore selected to test the following
hypotheses based on the above rationale:

° Denitriﬁer populations are distributed homogeneously within bioturbated
sediments, regardless of the redox potential.

0 The pattern of expression of denitriﬁcation genes is equal for denitriﬁers at any
depth within the bioturbated sediments, when subjected to equal environmental
conditions.

In order to test the hypotheses, sediments from mixed and unmixed layers were
analyzed in regard to denitriﬁer community structure and diversity by means of nirS
clone library construction and T-RFLP, abundance of certain nirS phylotypes determined

by real-time PCR, and denitriﬁcation capacity after nitrate addition to microcosms.

Materials and Methods

Study area and sediment sampling

Marine sediment samples were collected from three stations in Puget Sound,
Washington (Shallow Budd Inlet, Turning Basin, and Carr Inlet) and from the continental
slope at the Washington margin of the Paciﬁc Ocean (Figure 3.1, Table 3.1). The Puget
Sound and Washington margin samples were retrieved in July 2000 aboard the R/V
Clifford A. Barnes and in July 2001 aboard the R/V T.G. Thompson, respectively. One
core of at least 38 cm length was obtained from each site in Puget Sound using a Soutar
box corer and subsequent collection of subcores with 7.5- and lO-cm cast acrylic tubes
(Kristensen et al., 1999). A multicorer was utilized to retrieve the sample at the

Washington margin. Deep sediment samples were collected at Turning Basin (69 cm

55

sediment depth), Carr Inlet (135 cm sediment depth) and Washington margin (62 cm)
using a gravity corer.

Soutar box corer subcores and multicorer cores were sectioned into 0.5 cm Slices
and the sediments from the desired depth intervals (Table 3.2) were kept at -20°C for
DNA extraction. Intact cores from each station were kept at 4°C for microcosm

preparation.

Chemical analyses of sediment samples

The overlying waters and porewaters were sampled with a whole core squeezing
apparatus and analyzed for oxygen and nitrate plus nitrite concentrations as described by
Brandes and Devol (Brandes and Devol, 1995).

Sediment cores were sectioned and centrifuged at 1000 X g for 20 min. to separate
the pore water for ammonium (Strickland and Parsons, 1972), sulfate (Tabatabai, 1974),
iron (Stookey, 1970) and manganese (Brewer and Spencer, 1974) determination.

The sulfate reduction rate in the sediments was measured by the method of Fossing
and Jorgensen (Fossing and Jorgensen, 1989). The percentage of organic carbon was
obtained from other studies in the sampled area (Devol, pers. comm.) and was determined

by the method of Hedges and Stern (Hedges and Stern, 1984).

DNA extraction and puriﬁcation

Genomic DNA was extracted from 0.5 g of sediment from each sample using the

Ultra Clean Soil DNA kit (MO BIO, Solana Beach, CA), following the manufacturer’s
instructions. For Real Time PCR, the DNA was further puriﬁed by gel extraction from a

0.8% Sea Plaque agarose gel with B-agarase (Cambrex Bio Science, Rockland, ME), as

56

indicated by the manufacturer, followed by concentration with Microcon YM-IOO
columns (Millipore, Bedford, MA).

Genomic DNA was extracted as described in Chapter 2 (also, Griintzig et al., 2001)
from marine P. stutzeri isolate E4-2 (Braker et al., 2000) grown in nutrient broth (Difco,
Detroit, M1) at 37°C. Plasmid DNA was extracted from Escherichia coli transforrnants
grown in Luria-Bertani broth (Sambrook et al., 1989) at 37°C using the QIAprep Spin
Miniprep Kit (QIAGEN, Valencia, CA).

The quality of the extracted DNA was analyzed by electrophoresis on a 0.8% or 1%
agarose gel for genomic or plasmid DNA, respectively, and the concentration was

determined by absorbance at 260 nm.

T-RFLP analysis of nirS gene

Terminal restriction fragment length polymorphism (T-RFLP) analysis of the
sediment samples was performed as described by Braker et a1. (2000; 2001) with
modiﬁcations. The nirSlF primer labeled at the 5’ end with 6-carboxyﬂuorescein
(Integrated DNA Technologies, Coralville, IA) and nirS6R primer (Braker et al., 1998)
were used to amplify ~890 bp of the nirS gene in ﬁve replicate 50 pl PCR reactions for
each environmental DNA sample. The PCR reactions contained 150 ng environmental
DNA, 1 pM of each primer, 200 pM of each deoxyribonucleoside triphosphate, 2.5 mM
MgCl2, 0.4 pg pl'l bovine serum albumin (Roche Molecular Biochemicals, Indianapolis,
IN), 1.5 U T aq polymerase (Promega, Madison, WI), and 1X buffer provided with the
enzyme. After an initial denaturation step of 3 min at 95°C, a touchdown PCR was
performed as described by Braker et al. (2000). The replicate PCR reactions were pooled,

concentrated to approximately 60 pl with a Speedvac (Savant, Holbrook, NY) and the

57

ampliﬁed 890 bp nirS fragments were puriﬁed from a 2% preparatory agarose gel with
the QIAquick Gel Extraction kit (QIAGEN, Valencia, CA) eluting in 30 pl sterile-ﬁltered
cell culture tested water (Sigma, St. Louis, MO). Aliquots (10 pl) were digested with 5U
of restriction enzymes MspI and HhaI (New England Biolabs, Beverly, MA) in Single
enzyme reactions for 14 h at 37°C in the manufacturer’s recommended reaction buffers.
The restriction endonucleases were inactivated by incubating the reaction mixture at
65°C for 25 min. MspI and HhaI were used because Braker et al. (2001) have shown that
they yielded the highest number and most even distribution of T-RFS at these Sites.

Two different methods were applied for the separation and identiﬁcation of the
generated terminal restriction fragments (T-RFS). The digested DNA samples from Puget
Sound sediments were concentrated to 2 pl and analyzed on a 373 ABI Stretch automated
DNA sequencer (Applied Biosystems Instruments, Foster City, CA) by gel
electrophoresis, as described by Braker et al. (2001). The T-RFS generated from the
Washington margin sediment, on the other hand, were separated by capillary
electrophoresis with an Applied Biosystems PRISM 3100 Genetic Analyzer (Applied
Biosystems, Foster City, CA). This system loads the sample by electrokinetic injection,
which is highly affected by the presence of small ions in the sample, as they will be
injected preferentially, due to their high charge-to-mass ratio, interfering with the DNA
uptake. Therefore, an additional desalinization of the inactivated restriction digest was
performed, by adding sterile-ﬁltered cell culture tested water (Sigma, St. Louis, MO) up
to a ﬁnal volume of 500 pl, before concentration and desalinization on a Microcon YM-
10 column ((Millipore, Bedford, MA). The desalinized sample was ﬁrrther concentrated

to approximately 12 pl with a Speedvac (Savant, Holbrook, NY) and 4 pl was mixed with

58

forrnamide and internal standard and loaded onto the capillary electrophoresis system
with an injection time and voltage of 30 s and 3 kV, respectively. These conditions
revealed Similar results to those previously Obtained from Carr Inlet by gel
electrophoresis. Capillary electrophoresis reported longer T-RF Sizes as the fragment size
increased; this shift was normalized during data analysis to equate with the data from gel
electrophoresis.

The lengths of the T-RFS were determined by comparison with the internal

standard using GeneScan 3.7 (Applied Biosystems, Foster City, CA) software.

Analysis of nirS T-RFLP proﬁles

Electropherograms from each of the analyzed samples were compared with
Genotyper 3.7 software (Applied Biosystems, Foster City, CA) and peaks, over a
threshold of 50 ﬂuorescence units in at least one sample, representing the same T-RF
were manually aligned. The alignment was based primarily on fragment length, but the
pattern of peaks was also taken into consideration (Blackwood et al., 2003). An average
ﬁ'agment length was determined for each T-RF (Dunbar et al., 2001). Only T-RFS from
55 to 600 bp length were included in the analysis, to avoid the detection of the primers
and Size determination uncertainties. The presence and height of peaks was used to
characterize the communities. In order to normalize the results, the relative abundance of
peaks was determined for each sample by dividing the height of each peak by the total
ﬂuorescence intensity of that sample. The results were converted to percentages and
represented as histograms. The proﬁles resulting from the single-enzyme reactions were

analyzed independently, as suggested by Dunbar et al. (Dunbar et al., 2000).

59

The T-RFLP proﬁles from sections within and between cores were compared by
Principal Component Analysis (PCA) after calculating the Hellinger distance between
proﬁles a31ackwood et al., 2003). The Hellinger distance is determined by calculating the
Euclidean distance between proﬁles after square root transformation of relative peak
heights. This transformation allows the use of PCA on community composition data
containing many zeros as a result of representing a long gradient, with loss and
appearance of several Species (Legendre and Gallagher, 2001). A dendrogram was
constructed to visualize Similarities between proﬁles by the application of hierarchical
agglomerative cluster analysis to the Hellinger distances between proﬁles. Ward’s
method was chosen for group linking (Ward, 1963). PCA and cluster analysis were
performed using the Community Analysis Package software version 3.01 (Pisces
Conservation Ltd., Hampshire, United Kingdom).

An in silico digestion of the nirS clone sequences retrieved in this study and of 68
additional nirS sequences from strains and environmental clones from GenBank was
performed for HhaI and MspI with a custom made Perl script. The theoretical T-RFLP
proﬁle produced by the environmental clones retrieved in this study was compared to the

actual T-RFLP proﬁles.

Cloning of nirS sequences

A clone library of nirS sequences was constructed from Shallow Budd Inlet
sediments corresponding to a depth interval from 2 to 2.5 cm. This Speciﬁc depth was
chosen due to its high richness in nirS sequences based on T-RFLP. Environmental DNA
extracted from these sediments was PCR ampliﬁed as described above for T-RFLP

analysis with some modiﬁcations. Unlabelled primers nirSlF and nirS6R were used in 10

60

replicate PCR reactions. In addition, only 20 cycles, instead of 25 cycles, were performed
in the stable phase of the touchdown PCR to reduce PCR-induced artifacts and biases
(Qiu et al., 2001; Acinas et al., 2005). The replicate PCR reactions were pooled,
concentrated to approximately 140 pl with a Speedvac (Savant, Holbrook, NY) and the
ampliﬁed 890 bp nirS fragments were puriﬁed from a 1.2% preparatory agarose gel with
the QIAquick Gel Extraction kit (QIAGEN, Valencia, CA) following the manufacturer’s
instructions. The puriﬁed fragments were cloned using the TA cloning kit (Invitrogen,
Carlsbad, CA) and 188 white insert-bearing clones were randomly selected for
sequencing. These clones were grown on LB freezing buffer (Sambrook and Russell,
2001) with 50 pg ml'l kanamycin and inserts were sequenced with vector primer Ml3F
using an ABI 1730 Genetic Analyzer or an ABI Prism 3700 DNA Analyzer (Applied

Biosystems, Foster City, CA).

Phylogenetic analysis of cloned nirS sequences

The nirS sequences obtained in this study were compared to the GenBank database
by using BLAST to remove sequences with no homology to nirS. In addition, GenBank
sequences closely similar to the cloned sequences were identiﬁed. The sequences were
aligned with a Specialized Hidden Markov Model aligner utilized by the Functional Gene
Pipeline/Repository of the RDP. Closely Similar sequences chosen from GenBank were
added to the alignment by proﬁle alignment in Clustal X (Thompson et al., 1997). The
aligned sequences were translated and the alignment manually corrected in BioEdit (Hall,
1999). A phylogenetic tree was inferred by the neighbour-joining method (Saitou and
Nei, 1987) with Mega 2.1 (Kumar et al., 2001) using Poisson correction (Nei and Kumar,

2000) and 100 bootstraps.

61

Quantiﬁcation of nirS from Speciﬁc clusters and strains by real-time PCR

Five clusters of highly related nirS sequences as well as the nirS gene from P.
stutzeri were quantiﬁed in the studied sediments by real-time PCR. Syerreen
methodology was used to amplify four of the nirS clusters, while TaqMan methodology
was used to amplify cluster 2 and P. stutzeri nirS. ARB (Ludwig et al., 2004) and
PrimerExpress (Applied Biosystems, Foster City, CA) software were used to design
primers and probes speciﬁc for the studied clusters (Table 3.4). A previously described
primer-probe set was applied in the quantiﬁcation of P. stutzeri nirS (Griintzig et al.,
2001; Chapter 2). The probes used for TaqMan methodology were dually labeled with the
ﬂuorescent dyes 6-carboxyﬂuorescein (FAM) and 6-carboxytetramethyl-rhodamine
(TAMRA) at the 5’ and 3’ ends, respectively. Primers and probes were synthesized by
Integrated DNA Technologies (Coralville, IA).

Syerreen real-time PCR assays were carried out on an ABI Prism 7900HT
Sequence Detection System (Applied Biosystems, Foster City, CA) in 10 pl reactions
using the Syerreen PCR Core Reagents kit (Applied Biosystems, Foster City, CA)
following the manufacturer’s instructions with some modiﬁcations. The reaction mixture
consisted of 0.3 pM of each primer, lXSyerreen PCR buffer, 2.5 mM MgCl2, 0.2 pM of
each deoxyribonucleoside triphosphate, 0.25 U AmpliTaq Gold DNA Polymerase, 0.1 U
AmpErase UNG, and 10 ng environmental DNA.

TaqMan real-time PCR assays were performed on a 7700 Sequence Detector
System (Applied Biosystems, Foster City, CA) in 30 pl reactions containing 1X TaqMan
Universal PCR Master Mix (Applied Biosystems, Foster City, CA), 0.3 pM forward

primer, 0.3 pM or 0.9 pM reverse primer for cluster 2 or P. stutzeri, respectively, 0.6 pM

62

or 0.525 pM of the ﬂuorogenic probe for cluster 2 or P. stutzeri, respectively, and 10 ng
environmental DNA.

The PCR cycling used for all reactions was the same as described in Chapter 2
(Griintzig et al., 2001). A melting curve analysis was performed from 60°C to 95°C on
the Syerreen assay products to determine if only one product was formed during the
real-time PCR ampliﬁcation. Negative controls with no template DNA were run in each
reaction. Additionally, controls with no probe were prepared for TaqMan assays. Data
were analyzed with the SDS 2.1 software (Applied Biosystems, Foster City, CA).

The speciﬁcity of the primer sets was tested using several clones retrieved in this
study as positive and negative controls. Plasmid DNA (1 pg) extracted from the
corresponding E. coli transfonnants was added as template DNA in this case.

The nirS copy number of Speciﬁc clusters and of P. stutzeri was determined by
comparison to a standard curve. Plasmid DNA from the following clones was used to
generate standard curves for nirS clusters: NIRS4A C07 (cluster 1), NIRS4A F08 (cluster
2), NIRS4A A02 (cluster 3), NIRS4A E10 (cluster 5), NIRS4A G11 (cluster 6). Genomic
DNA from marine isolate E4-2, previously identiﬁed as P. stutzeri (Braker et al., 2000),
was used to construct standard curves for P. stuzeri nirS quantiﬁcation. Standard curves
were run in triplicate or quadruplicate, while duplicate reactions were performed for
environmental DNA. Samples that exhibited CT values (cycle number at which the
ﬂuorescence emission crosses a deﬁned threshold) equal to or higher than the negative
controls were considered as below the detection limit. All results were normalized to the

quantity of community DNA.

63

Denitriﬁcation capacity in sediments

In order to determine denitriﬁcation capacity, microcosms were prepared from
sediment samples from ﬁve different depths from each location (Carr Inlet: 0-2, 2-4, 10-
12, 34-38, and 135 cm. Shallow Budd Inlet: 0-2, 2-4, 10-12, 24-28, and 34-38 cm.
Turning Basin: 0-2, 2-4, 10-12, 34-38, and 69 cm. Washington Margin: 0-2, 2-4, 10-12,
34-38, and 62 cm.). A homogeneous slurry was prepared for each Site and depth and
further divided to set up individual microcosms, while a sample was conserved as
unamended control. Each microcosm consisted of 50 ml Slurry and 110 ml headspace.
The slurry contained 10 g sediment and artiﬁcial seawater (Kemp et al., 1993) to make
up a ﬁnal volume of 50 ml. Salinity was adjusted in Puget Sound microcosms to 28 ppt,
the salinity of the area’s water. Three sets of microcosms were prepared with either 40
pM NaNO3, 400 pM NaNO3, or 160 pM O2. Acetylene (10%) was added to each
microcosm to inhibit the reduction of nitrous oxide (Sorensen, 1978). All media and
bottles used were ﬂushed with He to displace O2. Microcosms were incubated at 20°C in
the dark shaking at 220 RPM. A control not subjected to shaking was also prepared, as
well as a control without added acetylene. Three microcosms of each type were incubated
for 1, 5, and 26 h, respectively, after which a gas sample was taken and the sediment
stored at -80°C for further analysis. Nitrous oxide concentrations in the headspace
samples were determined on a Hewlett Packard 5890 Series 11 gas chromatograph
(Hewlett Packard, Palo Alto, CA) with an electron capture detector by comparison to a
standard curve. The partition coefﬁcient of nitrous oxide (2.1 gas/water) was included in

the determination of total micromoles of nitrous oxide produced.

64

Nitrous oxide production rates were determined for the incubations with 400 pM
NaNO3, which presented a nearly linear accumulation of N20.
T-RFLP analysis of selected microcosm samples was performed as described

above.

Results

Biogeochemical characteristics of the sediments

As expected for near Shore and continental shelf and Slope sediments with high
sedimentary respiration rate (Brandes and Devol, 1995) oxygen and nitrate exhibited a
narrow penetration depth into the sediments (Figure 3.2), being depleted at Similar depths
ranging from 0.5 to 2 cm, consistent with previous measurements in the same areas
(Brandes and Devol, 1995; Braker et al., 2000). Shallow Budd Inlet showed the highest
oxygen concentration, reaching 435 pM, and extremely low nitrate concentrations, never
higher than 1.7 pM, due to the extensive growth of algae in the overlying water at that
Site. Turning Basing was characterized by a high sulfate reduction rate in the sediments
(Table 3.1), twice as high as the next highest measured value in Shallow Budd Inlet and
consistent with the consumption of sulfate observed in the pore water proﬁles (Figure
3.3). Nitrate was also extremely low at Turning Basin, never reaching concentrations
higher than 0.7 pM. Carr Inlet sediments, on the other hand, had much higher nitrate
concentrations than the other Sites in Puget Sound, reaching 14.8 pM. Besides individual
differences, the three studied Sites in Puget Sound presented a dense benthic macrofauna,
with invertebrates being Observed down to at least 38 cm depth, consistent with the high

mixing rate of 43 cm2 y'1 and surface mixed layers of up to 22 cm observed in sediments

65

from other Puget Sound areas by analysis of excess 21OPb activity proﬁles (Carpenter and
Peterson, 1989). Macrofauna was absent, on the other hand, in deep samples retrieved
with the gravity corer (Turning Basin, 69 cm; Carr Inlet, 135 cm). The Washington
Margin sample, in contrast, lies within a zone with a sharp decrease in sediment mixing
coefﬁcients, probably accounted for by the low dissolved oxygen concentration in waters
over the slope from about 600 to 1400 m, which leads to the inhibition of macrobenthic
activity (Carpenter and Peterson, 1989). This was corroborated by our observations of the
sediment cores of this Site, showing noticeably less macrobenthic fauna present only
down to about 15 cm in the sediments. Measured pore water oxygen concentrations, as
expected, never exceeded 12 pM, more than an order of magnitude smaller than the
measured concentration in any of the Puget Sound Sites (Figure 3.2). Nitrate, on the other
hand, reached 44.6 pM, 3-fold higher than the highest concentration detected in Carr Inlet

sediments.

Denitriﬁer community structure by T-RFLP of nirS genes

The denitriﬁer community structure changes along redox gradients were studied by
analyzing heme cd, nitrite reductase gene (nirS) T-RF S from different depth sediments at
the four sites. Restriction with Mspl yielded a higher number of T-RFS in the studied
samples than HhaI and was therefore chosen for the subsequent data presentation.
However, analysis performed with HhaI led to comparable results and further supported
the conclusions.

In Puget Sound samples, the pattern detected by T-RFLP was highly Similar in the
sediment samples down to 37 cm depth within each Site (Figure 3.4), with T-RF richness

from 22 to 41, 39 to 47, and 21 to 37 for Turning Basin, Shallow Budd Inlet and Carr

66

Inlet, respectively. A sharp decrease in T-RF richness was detected in deep samples
retrieved below the bioturbated sediments in Turning Basin (5 T-RFS at 69 cm depth) and
Carr Inlet (7 T-RF S at 135 cm depth). No samples below the bioturbated zone were
retrieved from Shallow Budd Inlet. In contrast, Washington Margin sediments, presented
a conserved nirS community structure only down to 10 cm in the sediments, with T-RF
richness between 40 and 46. A Sharp decrease in T-RF richness was already detected at
this site for the 36-37 cm deep sample (12 T-RF S), which lies below the bioturbated zone,
as well as for the even deeper sample from 62 cm depth (15 T-RFS). Two T-RFs, 97 bp
and 140 bp, were common to all sites and depths, and in Signiﬁcant relative abundance in
all cases (13.8 to 69.6% for T-RF of 97 bp and 1.0% to 22.9% for T-RF of 140 bp). Only
15 other T-RFS were detected in all areas, but not at all studied depths. The remaining T-
RFS were absent in at least one site, with 37 T-RFS being characteristic of one sampling
location only. Although some T-RFS were detected in sediments within and below the
bioturbated zone, most T-RF S were detected only within the mixed sediments being
absent in the unmixed zone. However, certain T-RFS were particular only to the deep
unmixed sediments at one site (75 bp and 208 bp at 62 cm deep Washington Margin
sediments, and 91 bp and 218 hp at 135 cm deep Carr Inlet sediments), although some
were detected also in Shallow sediments in other areas. T-RF 116 bp was detected in
shallow and deep sediments of Washington Margin, Turning Basin and Shallow Budd
Inlet, however its relative abundance was Signiﬁcantly higher in sediments below the
bioturbated zone.

Cluster analysis of the T-RFLP proﬁles grouped the samples according to Site and

depth (Figure 3.5), with samples from unmixed sediments from all sites forming a

67

separate cluster. Sediments from the bioturbated depths at each location formed distinct
clusters. This pattern was also observed when an ordination technique (principal
component analysis, PCA) was applied to the T-RFLP proﬁles (Figure 3.6), in order to
reduce the dimensionality of the data and summarize common correlations among
variables (T-RFS) into fewer variables (components) (Dollhopf et al., 2004). The ﬁrst two
dimensions explained 47% of the total variability (26% and 21% in dimensions 1 and 2,
respectively). T-RFLP proﬁles were grouped according to sampling location, however
samples from unmixed sediments were grouped together, regardless of the site they were
retrieved from, in agreement with the cluster analysis. PCA allows the identiﬁcation of
the main T-RFS responsible for the grouping of the samples, based on the size and
direction of the eigenvectors representing each T-RF. Eigenvectors in the Vicinity of a
sample, are most likely affecting it, and the size of the vector is a measure of the
magnitude of the effect. T-RFS present only in one Site or presenting a signiﬁcantly
higher relative abundance at that site, were mainly responsible for the dispersion of the
samples from different locations (data not Shown). So, T-RFS 236 bp and 62 bp, present
only or at Signiﬁcantly higher abundance, respectively, in Washington Margin samples,
were mainly responsible for the separation of the samples from this site. A set of
additional T-RF S detected only in Washington Margin, but with low relative abundance,
also had a signiﬁcant effect in the separation of these samples, e.g. 268, 350, and 428 bp.
Carr Inlet samples were mainly affected by T-RFS 87 and 406 bp, present only at this Site,
as well as by a number of additional T-RFS of low relative abundance, e.g. 410, 393, and

386 bp. Shallow Budd Inlet and Turning Basin were ordinated closely together, in

68

concordance with cluster analysis. The main T-RFS responsible for their ordination were

T-RFS present only at both of these Sites, e.g. 80, 110, 111, 178, and 221 bp.

Phylogenetic analysis of nirS clone sequences

Out of 188 sequenced clones, 137 were identiﬁed as nitrite reductase heme cdl
(nirS) sequences longer than 600 bp and used, after translation, to build a phylogenetic
tree by neighbor-joining analysis including also closely related sequences from public
databases (Figures 3.7, 3.8, and 3.9). A wide range of sequence divergence between nirS
clones was observed, with clones being from 49.6 to 100% Similar. Six individual
clusters of highly Similar nirS sequences, generating only one or two hypothetical T-RFS
by in silica digestion with the same restriction enzymes used in the T-RFLP analysis,
were identiﬁed (Table 3.3). These clusters were later targeted for quantiﬁcation by real-
time PCR (Table 3.4). Most nirS clones were only distantly related to any cultured
denitriﬁer, with the exception of clones grouped in cluster 4 (Table 3.3) being 80.4 to
83.3% Similar to Raseabacter denitrificans and clone NIRS4B A12 being 77.7 and 78.8%
similar to Paracoccus pantatrophus and Paracoccus denitrificans PD1222, respectively.
Two large clusters of sequences (subtrees A and B) contained almost all nirS clones
retrieved in this study (96% of sequences), as well as other environmental clones from
related and unrelated areas, but no cultured denitriﬁer with the exception of Azaarcus
talulyticus 2F B6. All clones were at least 75% similar to environmental nirS clones from
other studies with the relatedness reaching values of up to 98.5% Similarity for NIRS4B
F09 and pA33, a sediment clone retrieved from a different location in Puget Sound
(Braker et al., 2000). Clones in subtrees A and B are most closely related to sequences

found in these previously studied Puget Sound sediments as well as Washington Margin

69

sediments (Braker et al., 2000), in a Baltic Sea cyanobacterial aggregate (Tuomainen et
al., 2003), in the Baltic Sea water column (Hannig et al., 2006), and in sediments from
Huntington Beach, Califomia (Santoro et al., 2006). Furthermore, they are also related
(up to 97.1% Similarity) to an mRNA sequence (clone ANIS-54) detected in sediments of
the River Colne estuary in the United Kingdom (Nogales et al., 2002). Arabian Sea water
column clones (Jayakumar et al., 2004) and an Eastern South Paciﬁc water column clone
from the oxygen minimum zone (Castro-Gonzalez et al., 2005) clustered with clones in
subtree B, however more distantly.

1n silica digestion of these clones with the same restriction enzymes was compared
to the measured T-RFLP pattern at Shallow Budd Inlet (Figure 3.10) (see Figure 4.17 for
HhaI results). A shift between in silica and real T-RF Sizes of up to 6 bp was consistently
detected, coincident with the Shift observed between samples analyzed by gel and
capillary electrophoresis (see Materials and Methods). Most of the clones (89%)
generated a hypothetical T-RF of 102 bp, corresponding to the main T-RF at all depths
(97 bp), thus indicating that this abundant T-RF is not representing one dominant nirS
sequence, but rather a group of different sequences with a coincident restriction site. Only
three other hypothetical T-RFS were generated, 68, 140, and 172 bp, corresponding to T-
RF S of 62, 136, and 169 bp, respectively. The latter T-RF (hypothetical T-RF of 172 bp)
was generated by 12 clones in comparison to 1 and 2 clones generating hypothetical T-
RFS of 140 and 68 bp, respectively, in accordance with the higher relative abundance of
that T-RF in the observed proﬁle. Hypothetical T-RF 140 bp was generated by only one
clone (NIRS4B A12); hence its low abundance might have interfered with its detection

by T-RF LP, although it was present in other studied areas (data not Shown). On the other

70

hand, although partial restriction by the enzyme might be responsible for the detection of
certain additional T-RFS in the T-RFLP proﬁle compared to the clone library, the T-RF
sizes between 62 and 97 bp could not be explained by partial restriction of the clones
generating a hypothetical 68 bp fragment. Therefore, it is assumed that the lack of
identiﬁcation of clones corresponding to the additional T-RF sizes is principally due to an

incomplete coverage by the clone library of the nirS diversity at the studied site.

Quantiﬁcation of clone clusters by real-time PCR

Speciﬁc primer sets (Syerreen methodology) or primer-probe sets (TaqMan
methodology) (Table 3.4) were designed to quantify the clusters identiﬁed on the
phylogenetic tree of nirS sequences from Shallow Budd Inlet sediments (Table 3.3) by
real-time PCR. The goal was to better describe the abundance in the environment of some
of the groups of novel sequences unrelated to cultured denitriﬁers, as T-RFLP is only
considered semiquantitative (Liu et al., 1997), due to the possible PCR bias towards
different targets. These particular clusters were chosen as they generated only one or two
individual hypothetical T-RFS by in silica digestion and were related closely enough to
allow the design of primers against the whole group, but dissimilar from the other clones
to avoid cross ampliﬁcation of different sequences. All clusters were ampliﬁed using
Syerreen methodology, except for cluster 2, which was quantiﬁed by TaqMan real-time
PCR, as the primer set used for Syerreen real-time PCR led to non-speciﬁc
ampliﬁcation of cluster 3 clones and was therefore replaced. Also, sets of primers
designed against cluster 4, did not amplify any clone in these clusters and were therefore
discarded. In addition, the previously designed primer-probe set against P. stutzeri nirS

(Grttntzig et al., 2001) was also used. The detection limit was extremely low for each

71

primer or primer-probe set in all real-time PCR ampliﬁcations, always below seven nirS
copies per 10 ng community DNA, with the exception of the quantiﬁcation of cluster 2
nirS, which presented a detection limit of 65 copies per 10 ng community DNA.

In Shallow Budd Inlet, as well as in the geographically close Turning Basin Site,
cluster 2 was the most abundant nirS cluster (4 X 106 nirS copies per pg community DNA
at Shallow Budd Inlet, 24-25 cm) (Figure 3.11) as expected by the high number of clones
corresponding to this cluster (28 clones). Clusters l and 3 presented about an order of
magnitude lower abundances than cluster 2, consistent with the lower number of clones
identiﬁed in these clusters. The lowest abundances corresponded to clusters 5 and 6,
represented by only 4 and 3 clones, respectively. In correlation with the T-RF LP proﬁles,
the abundance of each cluster did not vary Signiﬁcantly at different depths within the
mixed sediments (down to 37 cm), however the nirS copy number diminished 100- to
1000-fold in the deep Turning Basin sediments (69 cm). Signiﬁcantly lower abundances
of clusters 1, 3 and 6 were detected in Carr Inlet sediments (Figure 3.12 and Table 4.4),
compared to Shallow Budd Inlet and Turning Basin. Furthermore, cluster 2 was below
detection limit at all depths, except at 10-10.5 cm, at this site. However, consistent with
the results at the other Puget Sound Sites, the abundance of clusters 1 and 3 did not vary
signiﬁcantly within the mixed sediments and was below detection limit at 135 cm depth.
Cluster 6, on the other hand, presented a low relative abundance and was below detection
limit already at 10-10.5 cm depth. Much lower abundances of the quantiﬁed clusters were
detected in Washington Margin sediments (Figure 3.12), often being below the detection
limit. This is in agreement with the T-RFLP results, which identiﬁed a more distantly

related community at this Site. Interestingly, P. stutzeri nirS was detected at a relatively

72

high abundance of up to 1.5 X 106 nirS copies per pg community DNA in Washington
Margin sediments. This is equivalent to 6.6 ng P. stutzeri DNA per pg community DNA,
i.e., 0.66% P. stutzeri DNA in total community DNA. In addition, its abundance in Carr
Inlet (up to 1.4 X 105 nirS copies per pg community DNA) was Signiﬁcantly higher than
the abundance of the quantiﬁed clusters (Table 4.4). On the other hand, P. stutzeri nirS in
Shallow Budd Inlet (2.7 x 104 nirS copies per pg community DNA) and Turning Basin
(3.2 X 105 nirS copies per pg community DNA) was generally less abundant than the nirS
clusters identiﬁed in the Shallow Budd Inlet clone library. In silica digestion of P.
stutzeri ZoBell nirS with Mspl generates a hypothetical T-RF of 141bp. A T-RF of 137
bp, which could correspond to P. stutzeri was observed at signiﬁcant relative abundance
in the T-RFLP proﬁles of Washington Margin (up to 14.1%) and Carr Inlet (up to 6.6%)
(Figure 3.4), but at lower abundance in Shallow Budd Inlet and Turning Basin, in
accordance with the real-time PCR results. Besides being more abundant in marine
sediments than previously measured (Griintzig et al., 2001), a high abundance of P.
stutzeri nirS was also detected in deep unmixed sediments from Turning Basin and
Washington Margin, similar to the abundance detected in shallow sediments. This again

is consistent with the presence of T-RF 137 bp in deep Washington Margin sediments.

Denitriﬁcation capacity of sediments studied in microcosms

Microcosms of sediments from each location and ﬁve different depths were
prepared to study their initial denitriﬁcation capacity, measured as nitrous oxide
production after nitrate addition. Puget Sound sediments from all sites and down to 38 cm
depth were able to denitrify immediately, as observed by signiﬁcant nitrous oxide

production after NaNO3 amendment (Figures 3.13, 3.14, and 3.15). On the other hand, in

73

Washington Margin sediments, Signiﬁcant denitriﬁcation capacity was detected down to
12 cm depth only, with sediments from 34 to 38 cm depth showing a minor accumulation
of nitrous oxide (Figure 3.16). No nitrous oxide accumulation was detected in
microcosms amended with oxygen, as expected, neither in controls lacking acetylene.
Addition of 400 pM NaNO3 led to an accumulation of nitrous oxide close to linearity in
all cases. Therefore, it was used to calculate nitrous oxide production rates in the
microcosms (Table 3.5). The highest rate was measured in Shallow Budd Inlet sediments
from 2 to 4 cm depth (0.133 N20 h'l microcosm'l). NO lag time was detected before
initiation of nitrous oxide production, though a slight increase in production rate was in
some cases observed for the deepest mixed sediments devoid of nitrate, after addition of
400 pM NaNO3 (Shallow Budd Inlet, 24-28 cm; Carr Inlet, 2-4 cm, 10-12 cm, 34-38 cm).
The accumulation rate of nitrous oxide reached a plateau after 5 h in microcosms
amended with 40 pM NaNOg, suggesting a complete consumption of the added nitrate
after this time. Carr Inlet top sediment samples (0-2 and 2-4 cm) amended with 40 pM
NaNO3 accumulated about 1 pmol nitrous oxide after 26 h incubation. This represents a
complete conversion of the added N-NO3' to N-N2O. However, deeper samples within the
mixed zone of this Site and samples from the other Sites generally reached a plateau at
values below 0.45 pmol, indicating only partial conversion to nitrous oxide, maybe due to
concomitant occurrence of dissimilatory nitrate reduction to ammonium (DNRA).

In order to study the changes in community structure that occurred during
incubation with nitrate, T-RFLP analysis was done on the sediment from selected
microcosms from Shallow Budd Inlet, as these sediments had the highest accumulation of

nitrous oxide. The analyzed microcosms had been amended with 400 pM NaNO3, as

74

nitrate seemed not have been depleted even after 26 h incubation with this initial
concentration. Microcosms from the ﬁve studied depths (0-2, 2-4, 10-12, 24-28, and 34-
38 cm) incubated for 26 h were analyzed, as well as microcosms from 2-4 cm depth
incubated for different times (0, 5 and 26 h). The results were compared to the T-RFLP
proﬁles generated from the original sediments retrieved from Shallow Budd Inlet (Figure
3.17). A noteable change in community structure was detected in all microcosm samples
compared to the original sediments, with some T-RFS disappearing, appearing, or
changing in relative abundance after incubation. Besides the disappearance of some T-
RFS there was a general increase in richness after incubation (52 to 59 T-RFs after 26 h
incubation, compared to 39 to 47 T-RFS in the original sediments). Though different from
the original sediments, the community structure of incubated sediments from different
depths seemed highly Similar. Furthermore, no signiﬁcant variations were detected with
different incubation times. This is further supported by PCA ordination of these T-RFLP
proﬁles (Figure 3.18), which Shows a clear separation of proﬁles from original sediments
and from microcosm sediments along the ﬁrst dimension, which explained 59% of the
total variability. The second dimension explained an additional 11% of the variability.
Samples incubated for 0, 5 and 26 h were grouped closely together, indicating that the
main changes in community structure had not occurred during the incubation time, but
rather due to the original set up of the microcosms. Subcores used for DNA extraction
were immediately sectioned and samples stored at -20°C, while subcores used for
microcosm set up were kept at 4°C for 3 days after retrieval, maybe leading to changes in
the community structure. In addition, the treatment of sediments for microcosm set up

(e.g. dilution with artiﬁcial seawater, mixing, etc.) might have further affected the

75

denitriﬁer community structure. T-RFS present only in the original sediments, in the
incubated sediments, or whose relative abundance changed noteably, were mainly
responsible for the separation of the samples. Several T-RFS present only in the original
samples (e.g. 72, 79, 80, 89, 108, 110, 111 bp) had a strong inﬂuence in the separation of
the original samples from the rest. In addition, T-RF S 84 and 97 bp, the latter representing
most of the novel nirS clones retrieved from this environment, also strongly inﬂuenced
the separation, as their relative abundance diminished after incubation. On the other hand,
T—RF 103 bp, present only in the microcosms and absent in the original sediments, and T-
RFS 137, 167 and 169 bp, present at Signiﬁcantly higher abundance in the microcosms,
had a strong effect on the separation of the microcosm samples from the rest.
Interestingly, T-RF 137 bp coincides with P. stutzeri, a commonly isolated bacteria, and
T-RF 169 bp coincides with twelve of the retrieved clones from this Site, with
Raseabacter denitrificans, Azaarcus talulyticus, and the marine denitrifying isolate C10-

1, previously retrieved from sediments at the Washington margin (Braker et al., 2000).

76

 

vv

   

IX!
who'd

 

 

     
   

  
  
  

 

 

 

 

 

 

 

m H‘- n g/ Gm! Pnimll -
C I I t \
arr n e «I.»
Margin .- 0 \ j
. w, I ..
“IQ “ Turning Basin 3‘ a Q
1»
,,_ it» ..
420‘ "3' “’2' :3 m Shallow BUdd Inlet

 

 

 

Figure 3.1. Locations of sampling stations at Puget Sound and the Washington
margin slope.

77

Table 3.1. Locations and characteristics of sampling stations.

 

 

Sulfate
Water Organic . . reduction
Station Location depth carbon 1:3? $318133), rate lrrigation‘
(m) (%) (mmoles
m'2 day")

Shallow 47°04.99'N high
Budd Inlet 122°54.16'W 3 2'5 '9'15 26'2 '2 (38 cm)
Turning 47°03.13'N high

Basin 122°54.45'W '2 3'2 ”'80 2"2 25 (38 cm)
47°17.21'N high

Carr Inlet 12204295,“, 84 1.8 16.80 29.6 7.5 (38 cm)

Washington 46°25.57'N medium

Margin 12404150.“, 1138 3.1 3.65 34.386 0.04 (15 cm)

 

°Numbers in parenthesis indicate lowest depth at which macrobenthic fauna was still detected.

78

Table 3.2. Sediment samples analyzed from different sites.

 

Sediment depth analyzed (cm)

 

 

Analysis
Shallow Budd Washington
method Turning Basin Carr Inlet
Inleta Margin
T-RF LP 005 005 0-0.5 0-0.5
1-1.5 1-1.5 1-1.5 1-1.5
2-2.5 2-2.5 2-2.5 2-2.5
10-10.5 10-10.5 10-10.5 10-10.5
24-25 36-37 36-37 36-37
36-37 69 135 62
Cloning 2-2.5
Real-time PCR 0-0.5 0-0.5 0-0.5 0-0.5
0.5-1 0.5-1 0.5-1 0.5-1
1-1.5 1-1.5 1-1.5 l-l.5
1.5-2 1.5-2 1.5-2 1.5-2
2-2.5 2-2.5 2-2.5 225
10-10.5 10-10.5 10-10.5 10-10.5
24-25 36-37 36-37 36-37
36-37 69 135 62
Microcosms 0-2 0-2 0-2 0-2
2-4 2-4 2-4 2-4
10-12 10-12 10-12 10-12
24-28 34-38 34-38 34-38
34-38 69 135 62

 

"No deep sample (retrieved with gravity corer) was obtained at this Site.

79

Figure 3.2. Pore water proﬁles for O2 and N03' for Puget Sound (Shallow Budd
Inlet, Turning Basin, and Carr Inlet) and Washington Margin sediments.

80

Moron)? ? 1' L 1 1° 1.2 '.‘ 1° 1.3

20
500

 

02 (”Mi 0 100 200 390. 400

.2-
0

2l
4-
64
81

1° ‘ Shallow Budd Inlet

Depth (mm)

12‘
141

 

 

 

N03'(pM)0 2 4 6 6 10 12 14 16 18

20

 

02(pM) q 20 40 60 80 100 120 140169

Depth (mm)

12 Turning Basin
14 , , , , . . , ,

1103'me 2 4 6 8 1o 12 14 16 18

 

180

 

20

 

021“"10 50 100 150 __200

250

 

.4 1 r 1
-2

Depth (mm)

14 ‘ Carr Inlet

 

 

 

1103111141) 0 1o 20 30 40
02011111) 0 - so 100 150

 

Depth (mm)

12 Washington Margin
14 , . .

81

50
200

 

 

 

+0,

+ N03-

 

 

Figure 3.3. Pore water proﬁles for Fe(II), 8042', Mn(II), and NH4+ for Puget Sound
(Shallow Budd Inlet, Turning Basin, and Carr Inlet) and Washington Margin
sediments.

82

renown), 30421111141), Mn(ll)(pM)0 1o 20 30 4o
NH"(pM) o 50 100 150 200 250

 

 

I)

Shallow Budd ’5‘
8

15

 

 

20 . , . .
FO(II)(1.IM), 3042111101), Mn(uxnmo s 10 15 211 215 so

 

 

NH4+IPMI 0 so 100 150 200 250 300 350

 

L

Turning

Basin 10

Depth (cm)

15

L A IL L ‘1 A A AA A A

V V v r v v y 77

 

 

20- . . .
rotunda),NH4ttnM),so42'(mM) 0 1o 20 30 40 so so 70 so so 100
Mn (II) (11M) 0 so 100 150 200 250 300

oi.

 

 

 

 

 

 

 

 

 

 

 

A 5 "
E
8.
Carr Inlet 5 1o
1%
D
15-
20 ~ . .
Fe(l|)(pM), SO42'(mM), Mn(II)(pM) L 1.0 210 :10 40 50
NH 4'01") 0 so 100 150 200 250
'8
Washington 3' + Fe(ll)
Margin 3 + MM“)
0 '0'
+ NH4
—0— 30‘2'

 

 

 

 

 

 

 

83

 

0%

20%

40%

60%

80%

100%

 

 

 

g, 00.5 1:162 1:365
g 1-1-5 1370 1:175
= 2'25 I79 1:80
210-105 "
3 36-37 E1184 E1187
:5: 62 12189 12191
A 3 MS I94 G97 .
E § 1-1.5 5102 121105
E F 2-2.5 C1108 821110
10-10.5
E, 5 36-37 C1111 111116
E 135 13134 5137
2 3. I140 I154
C -
g 3 ‘13:: I167 @169
'5 E 2.25 C1175 I178
:21 E10-10.5 I199 113208
0 = 24-25 '
o 2 36-37 @218 [3221
In E1228 I236
g 005 I265 I275
a :12: 111344 111376
E10403 [1403 13406
g 36-37 8416 I474
" 59 C1484 C1598

 

 

Relative abundance of T-RFs (%)

Figure 3.4. T-RFLP proﬁles obtained by ampliﬁcation of nirS genes from Puget
Sound and Washington Margin marine sediments from various depths.

Presented T-RFS were generated after restriction with Mspl. Numbers in legend indicate
T-RF length in base pairs for fragments representing more than 1.5% relative abundance.

84

 

2.58 1.82 1.28

0.538

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TB 58

CI 135
WM 62
WM 315-3?
Cl 383?
CI 1010.5
CI 22.5

CI 1-1.5

CI 00.5
WM 101 0.5
WM 225
WM 1-1.5
WM 005
TB 383?
TB 0-0.5
TB 1010.5
TB 225
TB 1-1.5
SB 36-37
SB 24-25
SB 1-1.5
SB 10-1 0.5
SB 22.5
SB 00.5

Figure 3.5. Dendrogram obtained by cluster analysis of T-RFLP proﬁles.

Hierarchical clustering analysis was performed applying Ward’s method on the Hellinger
distances between T-RFLP proﬁles from Puget Sound and Washington Margin sediments
from various depths. TB, Turning Basin; SB, Shallow Budd Inlet; CI, Carr Inlet; WM,
Washington Margin. Numbers next to sample names indicate sediment depth in cm.

85

 

 

 

 

 

 

 

 

 

 

 

8 4
‘ A

6 ~ A ‘

4 d ‘
C; 1 ’A x .
1:\_ , 2 ‘ x deep sediments 0 TB
9‘. A n I SB
2 -B I: 5 -2 0 2 4 6 8 1b 4 CI
"" -2 - x WM
1: In

.4 4

*5 >51

—8 — x

Dim 1 (26%)

 

 

 

Figure 3.6. PCA ordination of T-RFLPS of nirS genes from Puget Sound and
Washington Margin sediments.

TB, Turning Basin; SB, Shallow Budd Inlet; CI, Carr Inlet; WM, Washington Margin.
Percentage of variance explained by each dimension is presented in brackets. Oval
indicates the deep samples from below the bioturbated zone. Dim, dimension.

86

 

‘ Subtree A
J L——- Subtree B
i ‘— Huntington Beach sed. clone th 46 (00159571)
7 L— Thiabaciilus denitrificans (NC 007404)

J~]—— Arabian Sea water col. clone GB40—4F (AY336818)
l I 99[ “" Pseudomonas aeruginosa PAO1 (AEOO4488)
. Pseudomonas fluorescens (AF 1 97466)
' rl—— Marine denitrifying isolate 89-12 (AJ2483Q3)
gél (— - Marine denitrifying isolate 04 14 (AJ248395)

73;, 1* Marinabacter aquaeoleiVTB (NZ AALGOtOOOOOZ)
. 100'—— Marine denitrifying isolate C10-1 (AJ248394)
j I 4122- * Cupriavidus necatar (X91394)
- T?“ ' Ralstonia metallidurans CH34 (NZ AAAI03000011)

|

l

5L

'— Huntington Beach sed. clone we so (00159648)
(1091:..— NIRS4A E11

_ Huntington Beach sed clone th 4C (DQ159496)

3 ”7 " Huntington Beach sed. clone hbD 5A (00159611)

. 7— Huntington Beach sed. clone hbD 2F (00159599)
.2 100.3 59 ll - Huntington Beach sed. clone hbD 80 (00159624)

l ' ~ - - —~ Silicibacterpameroyi ass-3 (CP000032)

100 f Paracoccus denitriﬁcans PD1222 (U05002)

.__.__1
_;ﬁ

1 903 ‘ Paracoccus pantatraphus (AJ4014S2)
1 i; 7 NIRS48 A12
‘ 53 J1 "'77 Raseabacter denitrificans (AJ224911)
a}, - Huntington Beach sed done has 213 (DQ159527)
8100 Cluster 4 (4 clones)
ﬁ—r “Tm Hydrogenobacter thermophilus (A8210046)

7 . 7 J _ Arabian Sea water col. clone GB40-5A (AY336820)
tQQjTT—T— Puget Sound sed. clone p816 (AJ2484ZO)
100?* -— * * -- -—— Wash. margin sed. clone wA20 (AJ248428)
100; Puget Sound sed. clone pB6 (N248419)
_ _, j _ __ j ' Puget Sound sed. clone pA17 (AJ248407)
I 1oo' l Baltic Sea water col. clone M150—85 (00072183)
100 Baltic Sea water col. clone MOO-54 (00072203)
67 i— Thauera selenatis AX (AY078264)
J ‘1 Thauera aramatica K172 (AY078256)
I 1 100:‘ j ’ Pseudomonas stutzeriZoBell (X56813)
95 U Marine denitrifying isolate E4-2 (AJ248398)
i 100‘ Alca/igenes faecalis A15 (AJ224913)
Azaspirillum braSi/ense Sp7 (AJ224912)

spa—4
01

 

 

531

Figure 3.7. Phylogenetic tree showing the afﬁliation of nirS clone sequences

retrieved from Shallow Budd Inlet, Puget Sound, sediments to selected reference

sequences.

The tree was generated by neighbor-joining analysis of approximately 215 aminoacids
with Azaspirillum brasilense Sp7 as the outgroup. Values at branch points indicate the
percentage of 100 replicate trees supporting that branch. Bootstrap values below 50%
were omitted. Names of clones retrieved in this study begin with NIRS4A and NIRS4B.
Clusters of highly Similar nirS clone sequences are identiﬁed by numbers, with quantity
of clones in the cluster indicated in brackets. For graphical purposes, detail of subtree A
and subtree B is presented in Figure 3.8 and 3.9, respectively. The scale bar represents
0.1 substitutions per sequence position. sed, sediment; Wash, Washington; col., column.

87

r" Cluster 2 (28 clones)
l‘: NIRS4A H01
NIRS4A 003
NIRS4A 802
. Li— NIRS4A 003
H){__ gNIRS4A H07
NIRS4A H09
NIRS4B (311
NIRS4B 011
62L)— NIRS4A 012
50 _‘ NIRS4A H02
lH+ ~ NIRS4A 010
;— NIRS4AG01
)J Cluster 3 $11 clones)
NIRS4B E 0
) 98)— NIRS4A 608
9 9 )3") NIRS4B 008
NIRS4A 005
I NIRS4B H10
NIRS4B c302
NIRS4B 501
L— NIRS4B E02
70 River Colne estuary sed. clone ANlS-54 (AJ440470)
JE Puget Sound sed. clone pA12 (AJ248405)

 

)_ NIRS4A Goz
: NIRS4A F11
NIRS4AA11
: NIRS4A F05
1 50: NIRS4A G10
H— NIRS4B 004
NIRS4B 512
NIRS4B c304
NIRS4A H03
NIRS4A 505
NIRS4B F06
NIRS4B co3
'uNlRS4AA05
60 L- NIRS4B 007
83 r NIRS4B B08
[90“ NIRS4B F04
, NIRS4A F10
. ._W_ NIRS4A E05
! FL;- NIRS4B 011
L Baltic Sea cyanob. aggregate clone BS1270 (AJ457196)
i NIRS4B 010
)F’NIRS4B 002
934 NIRS4AC10

 

 

F l

‘ NIRS4A 006
i NIRS4A 001
J, m- NIRS4A H10
1 63 71)) “NIRS4A C08
) 832r NIRS4A E01
l 52 NIRS4A F12
; NIRS4A 809
i “—96 L I— NIRS4A F02
_ ___ . : 109,5 ti” NIRS4A BO4
' NIRS4B C09
L47" — NIRS4B C10
'. I“ Puget Sound sed. clone p346 (AJ248422)
r, 65 “ “7) r“ NIRS4AG12
100 79g; Puget Sou nd sed. clone p849 (AJ248423)
Puget Sound sed clone p866 (AJ248424)
%_______ Baltic Sea water col. clone M60- 76 (00072204)
‘ NIRS4A 604
50 _-___ "ﬂ 7 Huntington Beach sed. clone th 86 (00159557)

F—a—m— ~ 4
010

Figure 3.8. Detail of Subtree A as described in Figure 3.7.

 

 

 

 

sed, sediment; col., column; cyanob., cyanobacterial.

88

56I—' — Huntington Beach sed. clone th 66 (00159579)
2, ‘——— Huntington Beach sed. clone th 7G (00159582)
' “ "‘“‘ Huntington Beach sed. clone hbD 4C (00159607)
50 0““ Arabian Sea water col. clone 6857-78 (AY336858)
:57 3 P— ES. Pacif. OMZ water col. clone AC100-135 (AJ811517)
* "‘7 Arabian Sea water col. clone V483—3EE (AY336925)
Huntington Beach sed. clone hbD 2G (00159600)
- ‘- Arabian Sea water col. clone V483-8H (AY336912)
‘ Azoarcus talulyticus 2FB6 (AY078272)
_ 85 l—‘ Puget Sound sed. clone p820 (AJ248421)
) , gl 1 “'7'“ ' Baltic Sea water col. clone M60-145 (00072215)
95 l T___ Arabian Sea water col. clone V483-4E (AY336904)

f 51909 L“ Arabian Sea water col. clone (3840-6A (AY336822)
l 52 " “7 Cluster 6 (3 clones)

 

 

 

 

 

 

’ 551 ' Puget Sound sed. clone pA5 (AJ248403)
-- Baltic Sea water col. clone M60-165 (00072223)
I Huntington Beach sed. clone th 8F (00159664)
’ F _._ NIRS4A F07
) 96l NIRS4A 808
L. 84 ﬂ NIRS4A 011

) [-991 *- NIRS4A 806
a; L— Huntington Beach sed. clone th 9F (00159587)
)— r— NIRS4A 810
100 Huntington Beach sed. clone th 6F (00159546)
64 52) Wash. margin sed. clone wA15 (AJ248427)
” 60 l' NIRS48 F09
100 L Puget Sound sed. clone pA33 (AJ248412)
— ‘~ Baltic Sea water col. clone M60-156 (00072220)
” 7 Wash. margin sed. clone wF16 (AJ248437)

10 Huntington Beach sed. clone th 4C (00159534)
50; 7 Cluster 5 (4 clones)

)Ew) Huntington Beach sed. clone th 70 (00159549)
A 100 Huntington Beach sed. clone th 1E (00159522)
as! 57 - NIRS4A H05

| NIRS4A C01
82 F NIRS4A E12

Cluster1 (21 clones)

 

 

 

 

 

Figure 3.9. Detail of Subtree B as described in Figure 3.7.

sed., sediment; Wash., Washington; 001., column; E.S. Pacif. OMZ, Eastern South Paciﬁc
Oxygen Minimum Zone.

89

) 062(68)
i I72

' 074

D 79

El 80

m 84

In 89

El 91
94
097(102)
D102
0105
8108
E 110
II] 111
[0116
0137
B 140
I167
0169(172)
i 376
0% 20% 40% 60% 80% 100% 9416

I474
Relative abundance of T-RFs (%) 9598

1-1.5

2-2.5

10-10.5

24-25

Depth in sediments (cm)

36-37

 

 

 

 

Figure 3.10. Comparison of T-RF LP profiles obtained by amplification of nirS genes
from Shallow Budd Inlet, Puget Sound, sediments from various depths to
distribution of T-RF sizes obtained by in silico digestion of nirS clones retrieved
from that same site (sediment depth 2-2.5 cm).

Presented T-RFs (real and hypothetical) were generated by restriction with Mspl.
Numbers in legend indicate real T-RF lengths in base pairs for fragments representing
more than 1% relative abundance. Values in brackets indicate corresponding T-RF size
generated by in silico digestion of clones. One clone generated a hypothetical T-RF of
140 bp (plain grey), however the corresponding real T-RF (136bp) was not detected at
this site and was not included in the legend.

90

Table 3.3. Description of clusters identified in nirS phylogenetic tree.

 

 

 

. . In silica In silica
Minimum
Cluster Clones in cluster similarity generated generated
name (0/ )a T-RFs with T-RFs with
0 Mspl HhaIb
NIRS4A NIRS4B
A03 A09 Al 1
B03, C07, C09, ’ ’ ’
Cluster 1 004, 1303, 509, C01’ C05’ E04’ 98 102 7° (20” 23"
G07 G09 H08 E06, E08, F05, (1)
’ ’ F07, G05, G08
A02, A05, A06,
A06, A09, A10, A07, A10, BOZ,
BOS, C02, C04, B1 1, C04, C06, 36 (21), 234
Chm“ 2 C06, 012, 007, 009, 805, 1307, 90 102 (7)
D08, F08, H06 F02, F12, G09,
H1 1
A02, A04, E02, A08, D02, F01 , 36 (7), 234
Chm” 3 F03, F06, F09 F08, 606 98 '02 (4)
Cluster 4 307, D09, H1 1 E09 97 172 590
Cluster 5 E10 A04, C07, El 1 98 102 70
Cluster 6 011 003, 010 98 102 47

 

aSimilarity is inferred from Poisson-corrected distances between amino acid sequences.
bNumber of clones generating a speciﬁc T-RF is indicated in parenthesis, when more than one T-RF size is

observed.

91

Table 3.4. Primers and probes used to quantify the nirS gene of specific clusters and
strains by real-time PCR.

 

 

Target Primer or probea Sequence (5’-3’) Positionc Method
Cluster 1 nirSRTl F ATGGATCCCAAGTTTGGTCC 1159-1178 Syerreen
nirS RTl R TCGAGAACGCGTACGACT’I‘T 1295-1276 Syerreen

Cluster 2 nirSTaq2F GGATAGCACGCACCGCTATT 1023-1042 TaqMan

nirSTanpr CATGACGGCCGCGAACAAGTCC 1044-1065 TaqMan

nirSTaq2R TCGATCACCGCGATC'ITGT 1085-1067 TaqMan
Cluster 3 nirSRT3F CGATGTGACTGAGATI‘CCTCATC 1115-1138 Syerreen
nirSRT3R AGGCATGCTCGGGATTTTC 1273-1255 Syerreen
Cluster 5 nirSRTSF AATCTCGAAGCACTGGTGGAT 1099-1119 Syerreen
nirSRTS R GGATGACCGTCGGGATCA 1247-123 0 Syerreen
Cluster 6 nirSRT6F GACCGTAAACTGGCTGCTCTT 1093-11 13 Syerreen
nirSRT6R ATGGCCCTCAGGGTCTGTT 1245-1227 Syerreen

P. stutzeri nirSTaqutFb ACAAGGAGCACAACTGGAAGGT 1259-1280 T aqMan

nirSTaqutprb GGCAACCTGTTCGTCAAGACCCA 1 309-133 1 TaqMan

nirSTaqutRb CGCGTCGGCCCAGA 1362-1349 TaqMan

 

a F, forward primer; R, reverse primer; pr, probe.
b As in Chapter 2 and Grilntzig et al., 200].

° Positions in the m'rS gene of Pseudomonas stutzeri ZoBell EMBL X56813.

92

Shallow Budd Inlet

nirS abundance
(copies per microgram community DNA)

1.E+O1 1.E+02 1.E+03 1.E+O4 1.E+05 1.E+06 1.E+07

5)---. __ﬂ,,

 

 

 

 

 

 

 

 

 

 

 

 

 

E 10 _, ___.-.-——- —-~—
9.
5 15~~vi~ ’
s \ I \
1: 20-#’1
E
g 25 -———~ - — “ ‘” “
.99
3 3° ‘ / / 7

35 -———-— ‘ -—~

teem-r it 1131
40

 

Turning Basin

nirS abundance
(copies per microgram community DNA)

1.E+01 1.E+02 1.E+03 1.E+O4 1.E+05 1.E+06 1.E+07

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 J ‘1
1O _H____.._~..
E 20 -
2.
5 3
a 0
,3 40 _ -I—Cluster1
‘E -Ei—Cluster2
° 50 — —
.§ +C|uster3
E 50 — +Cluster5
70 _ —A——Cluster6
+P.stutzeri
80

Figure 3.11. Cluster specific and P. stutzeri nirS gene abundance in Shallow Budd

Inlet and Turning Basin sediments from different depths as measured by real-time
PCR.

93

Carr Inlet

nirS abundance
(copies per microgram community DNA)

1.E+O1 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
0 1991.999, ..

l

 

  

 

10'

 

 

 

 

 

 

 

E
3
I: 20 / //
3
¢
3 30
5 9/ J/ 0
.§ 40
'0 \
o \\
tn \
130
140

 

Washington Margin

nirS abundance
(copies per microgram community DNA)

1.E+01 1.E+02 1.E+03 1.E+O4 1.E+05 1.E+06 1.E+07

 

 

 

 

 

 

 

 

 

0
10 —
g 20 -
.:
§ 30 -
1: +Cluster1
g 40 - ~8—Cluster2
% 50 +Cluster3
3 —x—Cluster5
so . - —A—Cluster6
+P.stutzeri
70

 

Figure 3.12. Cluster specific and P. stutzeri nirS gene abundance in Carr Inlet and
Washington Margin sediments from different depths as measured by real-time
PCR.

Error bars represent standard deviation of duplicate assays. Missing values, represented
as discontinuities in plots, were below detection limit. Note that all values were below
detection limit for 135 cm deep Carr Inlet sediments.

94

 

3.5

Shallow Budd Inlet

N
01
l

N
4

 

+ SAhighN
--O-- SAlowN
"O" SA02
+ SBhighN
--><-- SBlowN
---><-- 8802
—O— SChighN
—-O-- SClowN
~O-- SCOZ
——)K— SDhighN
--)K-- SDIowN
--)K-- 8002
—I— SEhighN
—-l-- SElowN
-----l SE02

.3
U1
1

Nitrous oxide (micromoles I microcosm)

0.5 -

 

 

 

 

 

 

0 10 20 30
Time (h)

Figure 3.13. Nitrous oxide production in microcosms of Shallow Budd Inlet (Puget
Sound) sediments from various depths after addition of different electron acceptors.

Sediment depth is indicated by letters A through E (A, 0-2 cm; B, 2-4 cm; C, 10-12 cm;
D, 24-28 cm; B, 34-38 cm). Different microcosm sets were amended with 400 M
NaNO3 (highN), 4O uM NaNO3 (lowN) or 160 uM 02 (02).

95

2.5

 

 

Carr Inlet

A 2 —

E

at

O

0

2

.2

E

9, 1.5 —

2

O .

E + CAhighN

2 --o- - CAlowN

‘é’ -----o 0A02

§ ,f - -><- - CBlowN

g ,5.” ---><-- 0B02

m ,9?" + CChighN

g 'l'i’. "C'- CCIOWN

E 1’ ----e 0002

z + CDhighN
- -)K- - CDlowN
--->i<-- 0002
+ CEhighN
- ---- 0151ch
---n- 0502

 

 

 

 

 

 

 

Figure 3.14. Nitrous oxide production in microcosms of Carr Inlet (Puget Sound)
sediments from various depths after addition of different electron acceptors.
Sediment depth is indicated by letters A through E (A, 0-2 cm; B, 2-4 cm; C, 10-12 cm;
D, 34-38 cm; E, 135 cm). Different microcosm sets were amended with 400 M NaNO;
(highN), 40 M NaN03 (lowN) or 160 uM 02 (02).

96

 

3.5
Turning Basin

2.5 -

 

+TAhighN
--O-- TAlowN
We TA02
+TBhighN
-->(-- TBlowN
--><-- T802
—O—TChighN
--O-- TClowN
-----0 TC02
+TDhighN
--)K-- TDIowN
---)l<- T002
+TEhighN
—-l-— TElowN
---l-- TE02

1.5 ~

Nitrous oxide (micromoles I microcosm)

 

 

 

 

 

 

 

Figure 3.15. Nitrous oxide production in microcosms of Turning Basin (Puget
Sound) sediments from various depths after addition of different electron acceptors.

Sediment depth is indicated by letters A through E (A, 0-2 cm; B, 2-4 cm; C, 10-12 cm;
D, 34-38 cm; E, 69 cm). Different microcosm sets were amended with 400 M NaNO3
(highN), 4O uM NaNO3 (lowN) or 160 M 02 (O2).

97

2.5

 

 

 

 

 

 

Washington Margin
A 2 -
E
(D
o
O
2
.2
E
m 1.5 1
2
o
E
B
.2
E.
0 17
:2
x
o
0)
3
.5
z 0.5 —
>6 —'-‘-'—“‘" '-'—'.-"_-,-.—v‘-‘x
/ 9999999 ’i— 99999999999
1" .......
07.
o- - , —-—-:---—; ------
0 10 20
Time (h)

+WAhighN
--O-- WAlowN
MO WA02
——X——WBhighN
--><-— WBlowN
--X-- W802
+WChighN
—-O-- WClowN
-----0 W002
+WDhighN
--X-- WDlowN
--X-- W002
—I—WEhighN
—-I-- WElowN
---I- WE02

 

Figure 3.16. Nitrous oxide production in microcosms of Washington Margin
sediments from various depths after addition of different electron acceptors.

Sediment depth is indicated by letters A through E (A, 0-2 cm; B, 2-4 cm; C, 10-12 cm;
D, 34-38 cm; E, 62 cm). Different microcosm sets were amended with 400 pM NaNO3

(highN), 40 uM NaNO3 (lowN) or 160 uM 02 (02).

98

 

Table 3.5. Nitrous oxide production rates in microcosms amended with 400 pM
NaN03.

 

Rates of N20 production

 

 

Depth (cm) (pmoles N20 h'l microcosm'l)

SB CI TB WC
0-2 0.129 0.0612 0.102 0.0509
2-4 0.133 0.0432 0.0471 0.0386
10-12 0.119 0.0528 0.077 0.0233

24-28 0.0915 ND ND ND
34-38 0.0126 0.0304 0.0768 0.0026
62 ND ND ND 0.0003

69 ND ND 0.0004 ND

135 ND 0.0001 ND ND

 

ND: not determined.

99

Original sediment

 

 

A 041.5 E62 1:172
E 1.1_5 E74 5179
f; 2-2.5 580 8384
E10-10.5 '89 391
Q
g 24.25 094 097
,3 999., 13102 I103
Amended 0105 I108
0110 0111
.: 0-2
a D116 0137
s 24 i
.. A 0140 3:167
S 5 10-12 l
g - )0169 0189
1: 4.
g 2 28 1.221 I228
3“” 1,0275 I329
l0344 0376
:33 0" )0406 D416
§ g 5" I461 I474
_‘.=’ "'5 26 h 0598
0% 20% 40% 60% 80% 100%

Relative abundance of T-RFs (%)

Figure 3.17. T-RF LP profiles obtained by amplification of nirS genes from Shallow
Budd Inlet, Puget Sound, sediments compared to profiles obtained after incubation
of sediments from the same site with 400 uM NaNO; in microcosms.

Relative abundances of T-RFs in microcosms from different sediment depths incubated
for 26 h, as well as from sediments of 2-4 cm depth incubated for different times, are
represented. T-RFs were generated after restriction with Mspl. Numbers in legend
indicate T-RF length in base pairs for fragments representing more than 1% relative
abundance.

100

 

 

 

 

 

 

 

 

 

 

 

10
8 9 36-37. cm
6 - 0 Original sediment
§ 4 — 34-38 cm
5 A El Amended with NaN03
N 2 ~ o n 2499 cm (time curve)
E 5 h DD A
a 0 ‘ 0_2 cm 2-4 cm /26h 24-25 cm 0'0-5 cm A Amended with NaN03
_2 _ 91042 cm 1.1 .5 0m (different sediment
’2-25 cm depths)
_4 _ 10-105 cm
'6 l l r
-10 -5 0 5 10
Dim 1 (59%)

 

Figure 3.18. PCA ordination of T-RFLPs of nirS genes from Shallow Budd Inlet,
Puget Sound, sediments not subjected and subjected to incubation with 400 [1M
NaNO3 in microcosms.

In addition to the untreated sediments from several depths, sediments from 2 to 4 cm
depth incubated for different times, as well as sediments from different depths incubated
for 26 h, were included in the analysis. Labels next to the points in the graph indicate
sediment depth or incubation time.

101

Discussion

Bioturbation of the sediments by the activity of benthic macrofauna has three major
implications: modiﬁcation of sediment texture, bio-irrigation, and dispersal of solid
particles (Meysman et al., 2006), notably ﬁrst recognized by Charles Darwin who
observed the reworking activity of earthworms in soil (Darwin, C., 1881). The dispersal
of solid particles, which include not only non-living particles, but also some of biological
nature, e.g. microbes, was suggested as an explanation for the conserved denitriﬁer
community structure observed in sediments regardless of the redox gradients (Braker er
al., 2001). I further tested this hypothesis, by analyzing sediments lying below the
bioturbated zone and compared sites with signiﬁcantly different bioturbation intensities.
As expected, a conservation of the denitriﬁer community structure, revealed by T-RFLP
of the nirS gene, was detected within the mixed sediments from all areas, with a sharp
decrease in richness of 3- to 8-fold in deep unmixed sediments (Figure 3.4). This
decrease in richness is not related, however, only to the disappearance of T-RFs in deep
samples, but also to a replacement by others, as some particular T-RFs are detected only
in deep samples, thus indicating the presence of microbes better adapted to the conditions
of these subsurface sediments. Furthermore, the predation by macrobenthic fauna on
bacteria in the bioturbated sediments, an effect absent in the deep unbioturbated
sediments, might be an additional factor leading to the differentiation of the communities
in these two depth zones, as well as the mucus secreted by these animals when gliding,
which constitutes a source of organic carbon and has been demonstrated to stimulate
microbial growth (Reise, 2002). The detected shiﬁ in community structure is consistent

with the observation of clear differences in the bacterial types isolated from‘surface

102

Wadden Sea sediments when compared to those from 50 and 100 cm deep layers, and
almost no overlap with those from > 200 cm deep sediments (Kepke et al., 2005), a trend
also observed by DGGE (Wilms et al., 2006). Furthermore, microarray analysis of
various functional genes, including nirS, in Puget Sound sediments also showed a
conserved community structure down to 25 cm, signiﬁcantly different from the one
observed below 50 cm depth (Tiquia et al., 2006). The highly conserved community
structure down to 38 and 10 cm in Puget Sound and Washington Margin sediments,
respectively, was directly correlated to the presence of macrobenthic fauna down to at
least those depths, but absent in the deeper analyzed samples. Burrows of benthic infauna
have also been observed in previous studies extending down to at least 35 cm in natural
sediments and microcosms (Satoh et al., 2007) and can even reach depths of around 1 m
for some polychaetes (Reise, 2002). These burrows not only lead to the physical mixing
of the sediments, but also to the irrigation of the sediments with overlying water.
Microelectrode measurements in a sediment microcosm allowed the detection of oxygen
down to a depth of 35 cm within a burrow (Satoh et al., 2007). This can therefore. explain
the presence of denitriﬁers as deep as 38 cm in Puget Sound sediments, and furthermore,
justify their ability to start denitrifying without lag time when incubated with nitrate
amendment (Figures 3.13, 3.14, 3.15, and 3.16), as nitrate might have penetrated deep
into the sediments through burrows, leading to concentrations in the surrounding
sediments similar to the surface sediments, differentiating them from the nitrate-depleted
bulk sediments.

PCA ordination as well as cluster analysis grouped the mixed sediments according

to site and separate from the deep sediments (Figures 3.5 and 3.6). Braker et al. had also

103

observed site-speciﬁc grouping of nirS and 16S genes from sediments retrieved from the
same area (Braker et al., 2001). Furthermore, the most geographically distant and
biogeochemically divergent site (Washington Margin) was also the most distantly related
in PCA and cluster analysis, while the two closest locations in Puget Sound with similar
nitrate proﬁles (Turning Basin and Shallow Budd Inlet) were grouped closer together.
This pattern was also observed by T-RF LP analysis of another denitriﬁcation gene (nosZ)
in continental shelf sediments from the Atlantic Ocean, with increasing variability when
going from a centimeter to a meter and further to a kilometer scale (Scala and Kerkhof,
2000)

Although nirS clones retrieved from Shallow Budd Inlet presented a sequence
divergence of up to 49.6% being also only distantly related to cultured denitriﬁers, their
similarity to other environmental clones never fell below 75% with a signiﬁcant fraction
(65%) being more than 90% similar to these previously retrieved nirS sequences (Figures
3.7, 3.8, and 3.9). This indicates an ever increasing database of environmental nirS
sequences of different environments, leading to a higher coverage of the nirS diversity in
environmental samples. However, these environmental sequences are still, with a few
exceptions, only distantly related to cultured denitriﬁers, which leads one to question
their functionality in the environment. The close similarity of several Puget Sound clones
to an mRNA nirS sequence (clone ANIS-54) from the River Colne estuary (up to 97.1%
similarity) (Nogales er al., 2002) might be an indication of the expression of these
sequences in nature, giving some support to the idea that they code for functional nitrite

reductases.

104

A nirS clone (pA33) previously detected in sediments at a different location in
Puget Sound (Braker er al., 2000) was, not surprisingly, the closest environmental nirS
sequence to a clone retrieved in this study (NIRS4B F09). Although quite closely related
(98.5% similarity), pA33 as well as four additional previously retrieved Puget Sound
clones (pA16, pA63, p816, and pB20) (Braker et al., 2000) were not detected by real-
time PCR quantiﬁcation with speciﬁc primers designed against them at the four sites
studied here (data not shown), indicating the absence of more closely related sequences
that could be detected by the highly speciﬁc real-time PCR method. Therefore, a more
site-speciﬁc occurrence of extremely similar sequences is inferred.

Although most clones were only distantly related to cultured strains, four clones
(cluster 4) were up to 83.3% similar to Raseabacter denitriﬁcans. The Raseabacter clade
is oﬁen found associated with dinoﬂagellates, as it has the ability to degrade
dimethylsulfonopropionate (DMSP) produced by the latter (Miller and Belas, 2004).
Interestingly, red tide was occurring at the time of sampling of Shallow Budd Inlet
sediments, leading to the speculation that cluster 4 nirS sequences might be associated
with Raseabacter clade members able to catabolize DMSP. Furthermore, nirS and 16S
phylogenetic trees, including also clones from other areas exhibited the same association
of Raseabacter denitrificans with clones retrieved from Shallow Budd Inlet (Chapter 4).

The commonly isolated marine denitriﬁer P. stutzeri was detected in all studied
areas by real-time PCR, being most abundant in Washington Margin sediments (0.66% P.
stutzeri DNA in total community DNA), compared to 1 and 2 orders of magnitude lower
abundance in Puget Sound sediments (Figures 3.1] and 3.12). Its abundance in

Washington Margin sediments was 10-fold and several orders of magnitude higher than

105

previously determined abundances in Monterey Bay water column (Ward and Cockcroﬁ,
1993) and sediments from other sites in Puget Sound and the coast of Washington
(Griintzig et al., 2001), respectively. This suggests that, though apparently cosmopolitan,
this bacterium presents a rather wide range of abundances even in geographically close
locations.

On the other hand, quantiﬁcation of clusters of novel nirS sequences by real-time
PCR showed signiﬁcantly higher abundances in sediments from which these sequences
were originally retrieved (Shallow Budd Inlet) and the closely located Turning Basin than
in Carr Inlet and Washington Margin sediments, giving further support to the above
stated idea of site-speciﬁc nirS sequences being present at these locations.

Each cluster of novel nirS sequences presented generally a similar abundance as
measured by real-time PCR throughout the mixed sediments, with a sharp decrease in
unmixed sediments, indicating that their relative abundance in the bacterial community,
and not only in the denitriﬁer community as observed by T-RF LP, does not change with
depth within the mixed zone. Interestingly, P. stutzeri nirS abundance determined by
real-time PCR did not diminish in deep samples from Washington Margin and Turning
Basin lying below the bioturbation zone, therefore, its relative abundance in the bacterial
community seems stable throughout the studied sediment depths, regardless of mixing.
As several other denitriﬁers, represented for example by the novel nirS clusters, seem to
diminish in deep unmixed sediments, the fraction of nirS DNA corresponding to P.
stutzeri in the total nirS DNA assemblage will increase. This is consistent with the
increase of T-RF 137 bp, which correlates with the expected size for P. stutzeri-nirS, in

deep Washington Margin samples (Figure 3.4). Therefore, P. stutzeri seems to be able to

106

survive and possibly also to thrive in sediments depleted of nitrate and oxygen for a long
period of time. Though not known to be fermenters, a very low level of fermentation has
been detected in Pseudomonas isolates depleted of nitrate and oxygen for several months
(Jorgensen and Tiedje, 1993) and could account for the maintenance of these denitriﬁers
in the absence of their normally used electron acceptors. Furthermore, iron reduction has
been recently detected in certain Pseudomonas strains (Naganuma er al., 2006),
constituting another alternative metabolicpathway for these denitriﬁers.

In addition to the conserved denitriﬁer community structure in mixed sediments
based on semi-quantitative (T-RF LP) and quantitative (real-time PCR) distribution of
nirS sequences, the denitriﬁcation capacity of the sediments throughout the mixed layer
was conﬁrmed by nitrous oxide accumulation after nitrate addition (Figures 3.13, 3.14,
3.15, and 3.16). No lag time was observed in all cases, suggesting that the denitriﬁcation
enzymes might have already been expressed in situ, even in sediments below the depth in
which nitrate is detected in bulk sediments. As suggested above, nitrate might have
penetrated deep into the sediments through macrobenthic fauna] burrows, leading to
denitriﬁcation activity in the sediments surrounding them. Satoh et al., observed a higher
number of 16S rRNA gene copy numbers of ammonia-oxidizing bacteria and Nitraspira-
like nitrite-oxidizing bacteria in sediments surrounding infaunal burrows thaniin bulk
sediments (Satoh et al., 2007). Furthermore, ammonium consumption was also higher at
the burrow wall, indicating that these sites were hot-spots for nitriﬁcation, due to the
penetration of oxygen through the burrows, although no oxygen was detected in the
surrounding bulk sediments. Therefore, besides the effect of burrows allowing the direct

penetration of nitrate into deeper sediments, the indirect effect of stimulation of

107

nitriﬁcation through the penetration of oxygen, has also been shown to increase the
denitriﬁcation rates, through coupled nitriﬁcation-denitriﬁcation (Dollhopf er al., 2005).

Although the presence of an N-oxide, as well as anaerobiosis, is normally required
for denitriﬁcation gene expression (Zumft, 1997), nirS gene expression has also been
observed in the absence of N-oxides in P. aeruginosa PAOl and P. stutzeri JM300 (Ka er
al., 1997). Therefore, some denitriﬁcation gene expression could also occur in nitrate-
depleted bulk sediments, maybe also related to the periodic exposure to nitrate, and
explain the fast response of denitriﬁers to nitrate amendment. No denitriﬁcation capacity
was, however, detected in sediments below the bioturbated zone, indicating that either no
denitriﬁers with all necessary enzymes to convert N03' to N20 are present in these
sediments, or that longer incubation times are required before detecting N20
accumulation.

Stoichiometric conversion to N20 of 40 M NaNO; was only observed for Carr
Inlet top sediments, with less than 45% recovery of the added N-NO3’ as N-N20 in the
other samples. This might be due to the conversion of nitrate to ammonium by DNRA,
which could be important in these sediments as this process is favored under high
available carbon to electron acceptor ratios (Tiedje, 1994). On the other hand, technical
problems during the determination of denitriﬁcation capacity, such as the imperfect
inhibition of the nitrous oxide reductase by acetylene in reduced sediment layers
(Binnerup er al. , 1992), could have led to underestimation of denitriﬁcation.

Surprisingly, the setup of microcosms for the determination of denitriﬁcation
capacity led to a noteable change in the denitriﬁer community structure, detectable

already before addition of nitrate to the sediments in the microcosm incubated for 0 h

108

(Figure 3.17). Natural and microcosm sediments were clearly separated by PCA
ordination, while sediment depth and incubation time did not exert such a signiﬁcant
effect on the community structure (Figure 3.18). Although spatial variations between
subcores used for original determination of denitriﬁer community structure by T-RF LP
and microcosm studies can not be ruled out, it seems improbable that the signiﬁcant
changes detected are a reﬂection of the cm scale separation of the subcores, as previous
studies have observed only minor variations in community structure at this scale (Scala
and Kerkhof, 2000; Berardesco er al., 1998). Therefore, it seems that the manipulation of
the sediments used for microcosm setup in addition to the time frame between retrieval
and processing of the cores was responsible for the observed changes. Particularly,
keeping the cores at 4°C for 3 days, 15°C below their in situ temperature, might have had
a signiﬁcant effect, as seasonal variability of the denitriﬁer community has been
determined to be signiﬁcant in sediments (Scala and Kerkhof, 2000), which might be
linked to changes in temperature, besides other factors. Furthermore, the disturbances
exerted on the sediments during microcosm setup, such as the dilution with artiﬁcial
seawater and mixing, might have also affected the bacterial community. Although the
surface of the bulk sediments was brieﬂy exposed to oxygen during the transfer to the
microcosms, the large bulk of reduced sediments was never exposed to it. Therefore, the
effect of oxygen on the community can be considered minimal. Interestingly, the
abundance of the T-RF representing most of the novel nirS genes retrieved from this
environment (97 bp) was lower in the microcosm sediments than in the original
sediments. On the other hand, the abundance of the T-RF of 137 bp, coincident with P.

stutzeri, and the T-RF of 169 bp, coincident with twelve of the clones, Raseabacter

109

denitriﬁcans, Azaarcus talulyticus, and a marine denitriﬁer (C10-1) isolated from
sediments at the Washington margin (Braker er al., 2000) showed the opposite trend,
being higher in the microcosm sediments. Therefore, manipulation of sediments early on
after retrieval might have already affected community structure favoring commonly
isolated bacteria, such as P. stutzeri (Ward and Cockcroﬁ, 1993).

In conclusion, denitriﬁer populations, using the nirS gene as marker, were
distributed homogeneously, regardless of the redox potential, within the bioturbated
sediments, with a sharp change in their community structure in deep unmixed sediments.
Denitriﬁcation capacity, indicative of potentially active denitriﬁers, was, as well, detected
throughout the mixed layer, highlighting the importance of considering them when

studying denitriﬁcation and its main players in the environment.

110

References

Acinas S. G., R. Sarma-Rupavtarm, V. Klepac-Ceraj, and M. F. P012. 2005. PCR-
induced sequence artifacts and bias: insights from comparison of two 16S rRNA clone
libraries constructed from the same sample. Appl. Environ. Microbiol. 71:8966-8969.

Berardesco, G., S. Dyhrman, E. Gallagher, and M. P. Shiaris. 1998. Spatial and
temporal variation of phenanthrene-degrading bacteria in intertidal sediments. Appl.
Environ. Microbiol. 64:2560-2565.

Binnerup, S. J., K. Jensen, N. P. Revsbech, M. H. Jensen, and J. Sarensen. 1992.
Denitriﬁcation, dissimilatory reduction of nitrate to ammonium, and nitriﬁcation in a
bioturbated estuarine sediment as measured with ”N and microsensor techniques. Appl.
Environ. Microbiol. 582303-313.

Blackwood, C. B, T. Marsh, S.-H. Kim, and E. A. Paul. 2003. Terminal restriction
fragment length polymorphism data analysis for quantitative comparison of microbial
communities. Appl. Environ. Microbiol. 69:926-932.

Braker, G., H. A. Ayala-del-Rio, A. H. Devol, A. Fesefeldt, and J. M. Tiedje. 2001.
Community structure of denitriﬁers, Bacteria, and Archaea along redox gradients in
Paciﬁc Northwest marine sediments by terminal restriction fragment length
polymorphism analysis of ampliﬁed nitrite reductase (nirS) and 16S rRNA genes. Appl.
Environ. Microbiol. 67 : 1893-1901 .

Braker, G., J. Zhou, L. Wu, A. H. Devol, J. M. Tiedje. 2000. Nitrite reductase genes
(nirK and nirS) as functional markers to investigate diversity of denitrifying bacteria in

Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol. 66: 2096—
2104.

Braker, G., A. Fesefeldt, and K.-P. Witzel. 1998. Development of PCR primer systems
for ampliﬁcation of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria
in environmental samples. Appl. Environ. Microbiol. 64:3769-3775.

Braker, G., J. Zhou, L. Wu, A. H. Devol, and J. M. Tiedje. 2000. Nitrite reductase
genes (nirK and nirS) as functional markers to investigate diversity of denitrifying

bacteria in Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol.
66:2096-2104.

Brandes, J. A., and A. H. Devol. 1995. Simultaneous nitrate and oxygen respiration in
coastal sediments: Evidence for discrete diagenesis. Journal of Marine Research. 53:771-
797.

Brewer, P. G. and D. W. Spencer. 1974. Colorimetric determination of manganese in
anoxic waters. Limnology and Oceanography. 16:107-110.

111

Carpenter, R., and M. L. Peterson. 1989. Chemical cycling in Washington's coastal
zone, p. 367-509. In M. R. Landry and B. M. Hickey (ed.), Coastal oceanography of
Washington and Oregon. Elsevier, Amsterdam.

Carpenter, R., M. L. Peterson, and J. T. Bennett. 1982. 2lon-derived sediment

accumulation and mixing rates for the Washington continental slope. Mar. Geol. 48:135-
164.

Carpenter, R., M. L. Peterson, and J. T. Bennett. 1985. 2”’Pb-derived sediment

accumulation and mixing rates for the greater Puget Sound region. Mar. Geol. 64:291-
312.

Castro-Gonzalez, M., G. Braker, L. Farias, and O. Ulloa. 2005. Communities .of nirS-
type denitriﬁers in the water column of the oxygen minimum zone in the eastern South
Paciﬁc. Environ. Microbiol. 7: 1298-1306.

Darwin, C. 1881. The Formation of Vegetable Mould Through the Action of Worms
With Observation of Their Habits. John Murray.

Dollhopf, S. L., H. L. Ayala-del-Rio, and J. M. Tiedje. 2004. Artiﬁcial neural network
and multivariate statistical analysis of microbial community ﬁngerprints. In Kowalchuk,
G. G., de Brujin, F. J., Head, 1. M., Akkermans, A. D., and van Elsas, J. D. (eds.).
Molecular methods in microbial ecology manual. Second edition. Springer Verlag.

Dollhopf, S. L., J.-H. Hyun, A. C. Smith, H. J. Adams, S. O’Brien, and J. E. Kostka.
2005. Quantiﬁcation of ammonia-oxidizing bacteria and factors controlling nitriﬁcation
in salt marsh sediments. Appl. Environ. Microbiol. 71:240-246.

Dunbar, J., L. O. Ticknor, and C. R. Kuske. 2000. Assessment of microbial diversity
in four southwestern United States soils by 16S rRNA gene terminal restriction fragment
analysis. Appl. Environ. Microbiol. 66:2943-2950.

Dunbar, J., L. O. Ticknor, and C. R. Kuske. 2001. Phylogenetic speciﬁcity and
reproducibility and new method for analysis of terminal restriction fragment proﬁles of
16S rRNA genes from bacterial communities. Appl. Environ. Microbiol. 67 :190-197.

Fossing, H. and B. B. Jorgensen. 1989. Measurement of bacterial sulfate-reduction in
sediments: evaluation of single-step chromium reduction method. Biogeochemistry.
8:205-222.

Griintzig, V., S. C. Nold, J. Zhou, and J. M. Tiedje. 2001. Pseudomonas stutzeri nitrite
reductase gene abundance in environmental samples measured by real-time PCR. Appl.
Environ. Microbiol. 67:760-768.

Hall, T. A. 1999. A user-friendly biological sequence alignment editor and analysis
program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41:95-98. ' .

112

Hannig, M., G. Braker, J. Dippner, and K. Jiirgens. 2006. Linking denitriﬁer
community structure and prevalent biogeochemical parameters in the pelagial of the
central Baltic Proper (Baltic Sea). FEMS Microbiol. Ecol. 57:260-271.

Hedges, J. and J. Stern. 1984. Carbon and nitrogen determinations of carbonate
containing solids. Limnology and Oceanography. 29:657-663.

Jayakumar, D. A., C. A. Francis, S. W. A. Naqvi, and B. B. Ward. 2004. Diversity of
nitrite reductase genes (nirS) in the denitrifying water coltunn of the coastal Arabian Sea.
Aquat. Microb. Ecol. 34:69-78.

Jargensen, K. S., and J. M. Tiedje. 1993. Survival of denitriﬁers in nitrate-free,
anaerobic environments. Appl. Environ. Microbiol. 59:3297-3305.

Ka, J.-O, J. Urbance, R. W. Ye, T.-T. Ahn, and J. M. Tiedje. 1997. Diversity of
oxygen and N-oxide regulation of nitrite reductases in denitrifying bacteria. FEMS
Microbiol. Lett. 156:55-60.

Kemp, P. F., B. F .Sherr, E. B. Sherr, J. J. Cole. 1993. Handbook of methods in aquatic
microbial ecology. Lewis Publishers.

Kiipke, B., R. Wilms, B. Engelen, H. Cypionka, and H. Sass. 2005. Microbial diversity
in coastal subsurface sediments: a cultivation approach using various electron acceptors
and substrate gradients. Appl. Environ. Microbiol. 71:7819-7830.

Kristensen, E., A. H. Devol, and H. E. Hartnett. 1999. Organic matter diagenesis in
sediments on the continental shelf and slope of the eastern tropical and temperate North
Paciﬁc. Cont. ShelfRes. 19:1331-1351.

Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular
evolutionary genetics analysis software. Bioinformatics. 17:1244-1245.

Legendre, P. and E. D. Gallagher. 2001. Ecologically meaningful transformations for
ordination of species data. Oecologia. 129:271-280.

Liu, W.-T., T. L. Marsh, H. Cheng, and L. J. Forney. 1997. Characterization of
microbial diversity by determining terminal restriction fragment length polymorphisms of
genes encoding 16S rRNA. Appl. Environ. Microbiol. 63:4516-4522.

Llobet-Brossa, E., R. Rosella-Mora, and R. Amann. 1998. Microbial community

composition of Wadden Sea sediments as revealed by ﬂuorescence in situ hybridization.
Appl. Environ. Microbiol. 64:2691-2696.

Ludwig, L., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A.
Buchner, T. Lai, S. Steppi, G. Jobb, W. Fﬁrster, I. Brettske, S. Gerber, A. W.
Ginhart, 0. Gross, S. Grumann, S. Hermann, R. Jost, A. Kiinig, T. Liss, R.
LiiBmann, M. May, B. Nonhoff, B. Reichel, R. Strehlow, A. Stamatakis, N.

113

Stuckmann, A. Vilbig, M. Lenke, T. Ludwig, A. Bode and K.-H. Schleifer. 2004.
ARB: a software environment for sequence data. Nucleic Acids Research. 32:1363-1371.

Meysman, F. J. R., J. J. Middelburg, and C. H. R. Heip. 2006. Bioturbation: a fresh
look at Darwin’s last idea. Trends in Ecology and Evolution. 21:688-695.

Miller, T. R. and R. Belas. 2004. Dimethylsulfoniopropionate metabolism by Pfiesteria-
associated Raseabacter spp. Appl. Environ. Microbiol. 70:3383-3391.

Naganuma, T., M. Sato, D., Hoshii, Y. Amano-Murakami, T. Iwatsuki, and K. W.
Mandernack. 2006. Isolation and characterization of Pseudomonas strains capable of

Fe(III) reduction with reference to redox response regulator genes. Geomicrobiology
Journal. 23:145-155.

Nei, M. and S. Kumar. 2000. Molecular evolution and phylogenetics. Oxford University
Press, New York.

Nogales, B., K. N. Timmis, D. B. Nedwell and A. M. Osborn. 2002. Detection and
diversity of expressed denitriﬁcation genes in estuarine sediments aﬁer reverse
transcription-PCR ampliﬁcation from mRNA. Appl. Environ. Microbiol. 68:5017-5025.

Qiu, X. Y., L.-Y. Wu, H. Huang, P. E. McDonel, A. V. Palumbo, J. M. Tiedje, and
J.-Z. Zhou. 2001. Evaluation of PCR-generated chimeras, mutations, and heteroduplexes
with 16S rRNA gene-based cloning. Appl. Environ. Microbiol. 67 : 880-887.

Reise, K. 2002. Sediment mediated species interaction in coastal waters. J. Sea Res.
48:127-141.

Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory
manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sambrook, J. and D. W. Russell. 2001. Molecular cloning: a laboratory manual. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y.

Saitou, N. and M. Nei. 1987. The neighbour joining method: a new tool for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.

Santoro, A. E., A. B. Boehm and C. A. Francis. 2006. Denitriﬁer community

composition along a nitrate and salinity gradient in a coastal aquifer. Appl. Environ.
Microbiol. 72: 2102-2109.

Satoh, H., Y. Nakamura, and S. Okabe. 2007. Inﬂuences of infaunal burrows on the
community structure and activity of ammonia-oxidizing bacteria in intertidal sediments.
Appl. Environ. Microbiol. 73: 1 341-1348.

Scala, D. J. and L. J. Kerkhof. 2000. Horizontal heterogeneity of denitrifying bacterial

communities in marine sediments by terminal restriction fragment length polymorphism
analysis. Appl. Environ. Microbiol. 66:1980-1986.

114

Serensen, J. 1978. Capacity for denitriﬁcation and reduction of nitrate to ammonia in a
coastal marine sediment. Appl. Environ. Microbiol. 35:301-305.

Stookey, L. L. 1970. Ferrozine —a new spectrophotometric reagent for iron. Analytical
Chemistry. 42:779-781.

Strickland, J. D. H. and T. R. Parsons. 1972. A practical handbook of seawater
analysis. Fisheries Research Board of Canada, Ottawa, Ontario.

Tabatabai, M. A. 1974. A rapid method for determination of sulfate in water samples.
Environ. Letters 7 :237-243. '

Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins.
1997. The ClustalX windows interface: ﬂexible strategies for multiple sequence
alignment aided by quality analysis tools. Nucleic Acids Research. 24:4876-4882.

Tiedje, J. M. 1994. Denitriﬁers. pp. 245-267. Methods of Soil Analysis, Part 2.
Microbiological and Biochemical Properties. Soil Sci. Soc. Amer., Madison, Wisconsin.

Tiedje, J. M., A. J. Sexstone, D. D. Myrold, and J. A. Robinson. 1982. Denitriﬁcation:
ecological niches, competition and survival. Antonie van Leeuwenhoek. 48:569-583.

Tiquia, S. M., S. Gurczynski, A. Zholi, and A. Devol. 2006. Diversity of
biogeochemical cycling genes from Puget Sound sediments using DNA microarrays.
Enviromnental Technology. 27:1377-1389.

Tuomainen, J. M., S. Hietanen, J. Kuparinen, P. J. Martikainen, and K. Servomaa.
2003. Baltic Sea cyanobacterial bloom contains denitriﬁcation and nitriﬁcation genes, but
has negligible denitriﬁcation activity. FEMS Microbiol. Ecol. 45:83-96.

Urakawa, H., T. Yoshida, M. Nishimura, and K. Ohwada. 2000. Characterization of
depth-related population variation in microbial communities of a coastal marine sediment
using 168 rDNA-based approaches and quinone proﬁling. Environ. Microbiol. 2:542-
554.

Ward, J. H. 1963. Hierarchical grouping to optimize an objective function. Journal of
the American Statistical Association. 58:236-244.

Ward, B. B. and A. R. Cockcroft. 1993. Immunoﬂuorescence detection of the
denitrifying strain Pseudomonas stutzeri (ATCC 14405) in seawater and intertidal
sediment environments. Microb. Ecol. 25:233-246.

Wilms, R., H. Sass, B. Kiipke, J. Kiister, H. Cypionka, and B. Engelen. 2006. Speciﬁc
bacterial, archaeal, and eukaryotic communities in tidal-ﬂat sediments along avertical
proﬁle of several meters. Appl. Environ. Microbiol. 72:2756-2764.

Zumft, W. G. 1997. Cell biology and molecular basis of denitriﬁcation. Microbiol. Mol.
Biol. Rev. 61:533-616.

115

CHAPTER 4

DENITRIFIER AND BACTERIAL DIVERSITY AND
DISTRIBUTION PATTERNS IN SEDIMENTS FROM THE ARCTIC
AND THE PACIFIC NORTHWEST

Abstract

Denitriﬁer and bacterial communities were studied in sediments from nine
geographically distant sites with different geochemical characteristics, including the
Arctic (high and low productivity areas) and the Paciﬁc Northwest (Puget Sound,
continental slope and abyssal sea ﬂoor). High diversity and richness were observed in all
sites for the denitrifying as well as the bacterial community, based on heme cal) nitrite
reductase (nirS) and 16S rRNA gene clone libraries, respectively, with the highest values
measured at the Paciﬁc Northwest continental slope. The different sites seem to harbor
distinct microbial communities, as indicated by the statistically signiﬁcant differences
observed in the clone libraries from different sites, with geographic distance as the main
determinant in differentiation of denitriﬁer, but not bacterial communities. This was
further conﬁrmed by station-speciﬁc clustering on phylogenetic trees and by low
similarity and estimated richness of shared OTUS, often close to zero, indicating little or
no overlap in membership between communities at different sites. Station-speciﬁc
clustering of different depth denitriﬁer communities was also observed based on nirS T-
RFLP proﬁles, with clusters from close geographic locations combining, in turn, into
larger clusters. These observations support the hypotheses that differences in denitriﬁer

community structure are inﬂuenced more strongly by geographic location than by

116

differing environmental factors within one area, often related to water depth, which in

turn have a stronger inﬂuence than sediment depth.

Introduction

In recent years, interest has grown regarding the study of the distribution of
bacteria in the environment. Marine sediments have been shown to contain a particularly
high abundance and diversity of microorganisms (Whitman er al., 1998; Torsvik et al.,
2002), which is only starting to be discovered. F urthermore, the distribution of
microorganisms seems to follow speciﬁc laws, similar to the ones governing the
distribution of macroorganisms, leading to the frequent observation of biogeographic
patterns (Green and Bohannan, 2006; Hughes Martiny er al., 2006; Ramette and Tiedje,
2007). Although not necessarily applicable to all microorganisms, some endemism has
been observed when the distribution of certain taxonomic groups or of bacteria in general
has been studied, often depending on the level of taxonomic resolution chosen (Cho and
Tiedje, 2000; Ramette and Tiedje, 2007; Homer-Devine er al., 2004).

The distribution of denitrifying bacteria, a functional group that can not be deﬁned
by a corresponding taxonomic group, also showed speciﬁc patterns in the environment,
when the diversity of denitriﬁcation genes was studied. So, horizontal heterogeneity at
different distance scales in marine sediments and along environmental gradients in a
beach aquifer was observed for nitrous oxide reductase (nosZ) (Scala and Kerkhof, 2000;
Scala and Kerkhof, 1999) and nitrite reductase (nirS and nirK) (Santoro et al., 2006)
genes, respectively. Furthermore, Paciﬁc Northwest marine sediment samples exhibited

differential nirS-based community structures, with geographic location, water depth and

117

sediment depth exerting decreasing inﬂuence in the differentiation of the communities, in
that order (Braker et al, 2000; Braker er al., 2001). Our previous work (Chapter 3) with
sediments from the same area also suggested this pattern, with the exception that
sediment depth played a more signiﬁcant role when communities above and below the
bioturbation depth were compared. In order to further describe the distribution of
denitriﬁers and the inﬂuence the above stated factors have on their differentiation, a
broad range of sediment samples from various locations and biogeochemical
characteristics was analyzed.

The Arctic Ocean comprises 25% of the total continental shelf areas, the main sites
for nitrogen cycling taking part in the oceans (Walsh, 1991), and climate models suggest
that the effects of global warming will be magniﬁed in this area (Houghton et al., 2001).
Sediment denitriﬁcation rates in some highly productive regions on the Arctic margins
reach values similar to the highly productive Paciﬁc Northwest area, while they are lower
in less productive regions (Devol er al., 2005). Therefore, this region constitutes an
interesting extreme environment to study the denitriﬁer community structure and
compare it to the geographically distant and biogeochemically different, yet similar in
certain aspects, Paciﬁc Northwest area.

Deep-sea sediments are still poorly explored, however, a high bacterial diversity,
including denitriﬁers, is being unraveled in this environment (Tamegai et al., 2007) and
functional denitriﬁers have been isolated even from 11,000 m deep sediments at the
Mariana Trench (Tamegai er al, 1997). This makes them interesting targets for denitriﬁer
community structure studies and comparison to continental shelf and slope sediments.

Therefore, in addition to the highly productive previously studied Paciﬁc Northwest

118

sediments from Puget Sound and Washington Margin (Chapter 3), abyssal sea ﬂoor
sediments from the same area, however located west of the Juan de Fuca ridge, a
signiﬁcant physical barrier for dispersion of sediment organisms, as well as Arctic
sediment samples from different depths within high and low productivity zones were
analyzed and their denitriﬁer and bacterial community structures determined and

compared between different sites, based on nirS and 16S rRNA genes, respectively.

Materials and Methods

Study area and sediment sampling

Marine sediment samples were collected from ﬁve stations in the Paciﬁc Northwest
and four in the Arctic (Figure 4.1, Table 4.1). The Paciﬁc Northwest stations included
three in Puget Sound (Shallow Budd Inlet, Turning Basin, and Carr Inlet), one on the
continental slope (Washington Margin) and one on the abyssal sea ﬂoor west of the Juan
de Fuca ridge (West of Juan de Fuca) in the Paciﬁc Ocean. The Arctic stations included a
shallow and a deep sample from transects along Barrow Canyon and east of Hanna Shoal
in the Chukchi Sea (Barrow Canyon Shallow, Barrow Canyon Deep, East Hanna Shoal
Shallow, East Hanna Shoal Deep). Puget Sound and Washington Margin samples were
retrieved and stored as described previously (Chapter 3). The West of Juan de Fuca
sample was collected aboard the R/V T.G. Thompson in July 2001. Barrow Canyon and
East Hanna Shoal samples were retrieved during a cruise on the US Coast Guard Ice
Breaker USCGC Healy in July 2002 and May-June 2004, respectively. Sediment cores
were obtained with a multicorer and sectioned into 0.5 to 2 cm slices. Sediments from the

desired depths (Table 4.2) were kept at -20°C or centrifuged at 12,500 X g to separate the

119

bulk sediments from pore water and stored at -80°C, for West of Juan de Fuca and Arctic

samples, respectively.

Chemical analyses of sediment samples

Chemical determinations for Puget Sound and Washington Margin samples were
performed as described previously (Chapter 3). In the West of Juan de Fuca and Arctic
sediments, a microelectrode was used for oxygen concentration determinations. The
porewater separated after centrifugation of the sediments at 12,500 X g was used for
nitrate (Armstrong et al., 1967), ammonitun (Strickland and Parsons, 1972), sulfate
(Tabatabai, 1974), iron (Stookey, 1970) and manganese (Brewer and Spencer, 1974)
determinations. The sulfate reduction rate in the sediments was measured by the method
of Fossing and Jargensen (Fossing and Jorgensen, 1989). The percentage of organic
carbon was obtained from other studies in the sampled area (Devol, pers. comm.) and

was determined by the method of Hedges and Stern (Hedges and Stern, 1984).

DNA extraction and puriﬁcation

Genomic and plasmid DNA extractions, puriﬁcations, and quality determinations

were performed as described previously (Chapter 3).

T-RFLP analysis of nirS gene

Terminal restriction fragment length polymorphism (T-RFLP) analysis of the nirS
gene was performed on sediment samples from selected depths from all sites (Tables 3.2
and 4.2), following the same procedure described in Chapter 3. Terminal reStriction
fragments (T-RFs) generated from West of Juan de Fuca and Arctic samples were

analyzed by capillary electrophoresis.

120

PCA ordination and cluster analysis of the T-RF LP proﬁles, as well as comparison
to the in silica digestion of nirS clones retrieved in this study was performed as described
previously (Chapter 3), with the exception that the Sarensen distance, which considers
only presence and absence of T—RFs, was applied in the cluster analysis. Canonical
correspondence analysis (CCA) (ter Braak, 1986) was applied on the T-RFLP results to
obtain an ordination of the samples (sites) based on their biological components (T-RFs)
by correspondence analysis, but optimized in terms of the inﬂuence of environmental
variables. One T-RFLP proﬁle from each site was used for the analysis, corresponding to
a sediment depth of 1-1.5, 1-2, or 0-2 cm, depending on the sample depths available from
each site (Tables 3.2 and 4.2). Nine environmental variables measured in all areas were
included in the analysis (oxygen, nitrate, ammonium, water depth, latitude, longitude,
temperature, salinity and organic carbon) (Table 4.1; Figures 4.2, 4.3, and 4.4). Overlying
water values were used for oxygen, nitrate and ammonium. CCA was performed using
the Environmental Community Analysis software version 1.37 (Pisces Conservation Ltd.,
Hampshire, United Kingdom).

In addition, to describe the nirS-containing community at each site and depth,
richness and diversity were determined based on the T-RFLP proﬁles. The total number
of T-RFs for each sample was presented as its richness, while the diversity was calculated
with the Shannon index H ’(Hill et al., 2003):

H ’=-Zp,- ln 1),,

where p,- is the proportion of the ith T-RF in the community.

An in silica digestion of the nirS clone sequences retrieved in this study was

performed for Hhal and Mspl with a custom made Perl script. The theoretical T-RFLP

121

proﬁle produced by the clones was compared to the actual T-RFLP proﬁles. The
Sarensen similarity index was calculated for hypothetical and real T-RFLP proﬁles from

the same sediment depths.

Cloning of nirS and 16S rRNA gene sequences

Clone libraries of nirS and eubacterial 16S rRNA gene sequences were constructed
from Shallow Budd Inlet, Washington Margin, West of Juan de Fuca, Barrow Canyon
Shallow, and East Hanna Shoal Shallow sediments from one depth at each site.(Tables
3.2 and 4.2). The depth chosen in each case was related to a high richness in nirS
sequences based on T-RF LP. The nirS clone libraries were constructed as described in
Chapter 3. The eubacterial 16S rRNA clone libraries were constructed after PCR
ampliﬁcation of environmental DNA extracted from the sediments. Primers 8-27F (5’-
AGAGTTTGATCMTGGCTCAG-3’ with M for A or C) and 1392-1407R (5’-
ACGGGCGGTGTGTACA-3’) (Braker er al., 2001) were used in three replicate PCR
reactions, containing 100 ng environmental DNA, 0.2 M of each primer, 200 M of
each deoxyribonucleoside triphosphate, 3 mM MgC12, 0.2 pg pl"l bovine serum albumin
(Roche Molecular Biochemicals, Indianapolis, IN), 2.5 U T aq polymerase (Promega,
Madison, WI), and 1X buffer provided with the enzyme. After an initial denaturation step
of 3 min at 95°C, 30 cycles of 45 sec at 94°C, 1 min at 57°C, and 2 min at 72°C were
carried out, followed by a ﬁnal extension of 7 min at 72°C. The replicate PCR reactions
were pooled, concentrated to approximately 70 pl with a Speedvac (Savant, Holbrook,
NY) and the ampliﬁed 1400 bp 16S rRNA fragments were puriﬁed from a 1.2%
preparatory agarose gel with the QIAquick Gel Extraction kit (QIAGEN, Valencia, CA)

following the manufacturer’s instructions. The puriﬁed fragments were cloned using the

122

TA cloning kit (Invitrogen, Carlsbad, CA) and 188 white insert-bearing clones were
randomly selected for sequencing. These clones were grown on LB freezing buffer
(Sambrook and Russell, 2001) with 50 pg ml'l kanamycin and inserts were sequenced
with vector primer M13F using an ABI 1730 Genetic Analyzer or an ABI Prism 3700

DNA Analyzer (Applied Biosystems, Foster City, CA).

Phylogenetic analysis of cloned nirS and 168 rRNA gene sequences

The nirS and 168 rRNA sequences included in the clone libraries constructed from
each site were compared to the GenBank database by using BLAST to remove sequences
with no homology and select sequences from cultured strains and closely related
environmental clones from other studies. Alignment and translation of the nirS
sequences, followed by the construction of a neighbour-joining tree (Saitou and Nei,
1987) with Mega 2.1 (Kumar et al., 2001) using Poisson correction (Nei and Kumar,
2000) and 100 bootstraps was performed as described in Chapter 3.

The 16S rRNA sequences retrieved in this study were checked for chimeras with
the Chimera_Check program from the Ribosomal Database Project II
(http://rdp.cme.msu.edu). The cloned sequences, in addition to the sequences selected
from the GenBank database, were then aligned against the RDP public alignment using
RNACAD in myRDP (Cole er al., 2006). The aligmnent was manually corrected in
BioEdit (Hall, 1999) and phylogenetic trees were inferred by the neighbour-joining
method (Saitou and Nei, 1987) with Mega 2.1 (Kumar et al., 2001) using Jukes-Cantor
correction (Jukes and Cantor, 1969) and 100 bootstraps.

The 16S rRNA sequences from individual libraries were assigned to taxonomic

groups by placing them into the RDP Hierarchy with the RDP classiﬁer (Cole er al.,

123

2005). Signiﬁcant differences in the taxonomic composition between the different
libraries was further determined with the RDP Library Compare tool (Cole et al., 2006).

Community description and comparison based on 16S rRNA and nirS clone
libraries

Sequences included in the 16S rRNA and nirS clone libraries from each site were
assigned to individual operational taxonomic units (OTUS) with the purpose of
community analysis. A Jukes-Cantor corrected distance matrix (Jukes and Cantor, 1969)
generated with the DNADIST program from PHYLIP
(http://evolution.genetics.washington.edu/phylip.html) based on the DNA alignments was
used for the OTU assignment by DOTUR (Schloss er al., 2005). The furthest neighbor
sequence assignment method was applied, with a 3% and 5% distance threshold for
assignment of 16S rRNA and nirS sequences to the same OTU, respectively. The
grouping of the sequences into OTUS allowed the evaluation of the sampling effort by the
construction of rarefaction curves, the determination of diversity with the Shannon Index,
and the estimation of richness with the bias-corrected Chaol nonparametric richness
estimator, also using DOTUR.

The statistical signiﬁcance of the differences between libraries constructed from
sediments retrieved at different sites was determined with l-LIBSHUFF (Schloss et al.,
2004), which uses the integral form of the Cramér-von Mises statistic. This statistic
compares the homologous and heterologous coverage curves between two libraries. A
Jukes-Cantor corrected distance matrix constructed as above, but including the sequences
retrieved from all sites was used for the calculations. To determine the probability that

the differences observed between the original libraries were due to chance, a Monte Carlo

124

procedure was further applied, by the construction of randomized libraries (10,000) and
comparing the value of the Cramér-von Mises statistic obtained in these random‘izations
to the original value. Signiﬁcant P values (P < 0.0026; 0 = 0.05) were determined after
applying the correction for multiple comparisons, considering that ﬁve libraries were
compared in 20 tests.

To better describe the similarity between the communities at different sites, the
richness of shared OTUS between each pair of communities as well as the community
overlap between different sites was determined with a nonparametric richness estimator
of shared OTUS analogous to Chaol and with the abundance-based Sorenson similarity
index (Labmd), respectively, using SONS (Schloss and Handelsman, 2006). The
preliminary assignment of sequences to OTUs was performed applying DOTUR on a
Jukes-Cantor corrected distance matrix constructed as above, including the sequences
retrieved from all sites. Also as above, sequences were assigned to OTUs with the
furthest neighbor sequence assignment method, with a 3% and 5% distance threshold for

16S rRNA and nirS sequences, respectively.

Quantiﬁcation of nirS from speciﬁc clusters and strains by real-time PCR

Five clusters of highly related nirS sequences identiﬁed previously in the Shallow
Budd inlet clone library (Chapter 3), as well as the nirS gene from P. stutzeri were
quantiﬁed in the studied sediments by real-time PCR. One depth from each Arctic site
and from West of Juan de Fuca (Table 4.2) was studied as described previously (Chapter
3). In addition, the results obtained at four sediment depths (1-1.5 cm, 1.5-2 cm, 2-2.5
cm, and 10-10.5 cm) in Puget Sound and Washington Margin (Chapter 3) were

considered biological replicates for the determination of statistically signiﬁcant

125

differences of cluster and P. stutzeri abundances between locations or vice versa by one-
factor analysis of variance (ANOVA) with Tukey’s HSD procedure as past hac test. No
two-way ANOVA was performed, as a different number of groups were included in each

comparison, as several samples had values below detection limit.

Results

Biogeochemical characteristics of the sediments

Puget Sound and Washington Margin sediment characteristics have been described
previously (Chapter 3). All Arctic sites showed high oxygen concentrations, reaching 320
M at Barrow Canyon Shallow. This highly productive site in the Arctic (Bates et al.,
2005; Hill and Cota, 2005) presented the same phenomenon observed in Puget Sound and
Washington Margin, namely oxygen and nitrate exhibiting narrow penetration and being
depleted at similar depths (Figure 4.2), as generally observed in highly respiring near
shore and continental shelf and slope sediments (Brandes and Devol, 1995). Barrow
Canyon Deep, as expected for deeper sediments, exhibited a slightly deeper oxygen
penetration of 8.5 mm. Nitrate was detected throughout the sediment core, although its
concentration was < 1 nM below 7.5 cm depth. The ammonium proﬁle at this site (Figure
4.3) indicates a sink at about 30 cm depth, maybe indicative of anammox occurring
within these sediments. East Hanna Shoal sites present a higher oxygen penetration depth
compared to Barrow Canyon of 1.5 and 3.8 cm for the Shallow and Deep sites,
respectively. This is consistent with the lower productivity determined in this area in
comparison to Barrow Canyon (Bates et al., 2005; Hill and Cota, 2005). Nitrate presented

a similar trend to the observed in Barrow Canyon Deep, namely being present throughout

126

the sediment core, but its concentration dropping signiﬁcantly below 1.25 and 3.5 cm
depth for East Hanna Shoal Shallow and Deep, respectively. The West of Juan de Fuca
site presented typical proﬁles for abyssal sea ﬂoor samples with very low organic carbon
input (Table 4.1). At this site oxygen penetrated down to 5.1 cm into the sediments, while
nitrate reached its maximum concentration at 1.2 cm depth (60.17 uM) and slowly

declined to a concentration of 15.8 M at the deepest analyzed sediment sample (30 cm).

Denitriﬁer community structure by T-RFLP of nirS genes

The denitriﬁer community structure at various sediment depths at each site was
studied by analyzing T-RFs generated after restriction of heme cd) nitrite reductase genes
(nirS). An extremely low number of T-RFs were detected after restriction of West of Juan
de Fuca samples with Mspl, therefore, Hhal was chosen for the subsequent data
presentation. However, analysis performed with Mspl led to similar results and further
supported the conclusions.

As previously observed in Puget Sound and Washington Margin (Chapter 3), T-
RFLP patterns in West of Juan de Fuca and Arctic sediments were highly' similar
throughout the studied sediment depths, with a notable shift in community structure in the
deep West of Juan de Fuca sample (20-21 cm) and a minor change detected in sediments
below 5 cm depthiat the Barrow Canyon Shallow and East Hanna Shoal Deep sites
(Figure 4.5). No correlation was observed between community structure and redox
gradients. From 130 identiﬁed T-RFs, only seven (64, 108, 232, 236, 348, 388, and 436
bp), were common to all sites, although none was detected at all depths. These were
generally present in signiﬁcant relative abundance, reaching values as high as 64.4% for

T-RF 64 bp in Carr Inlet sediments. The majority of T-RFs were absent in at least one

127

site, with 31% of the total number of T-RFs being characteristic for one speciﬁc sampling
location.

The highest richness, as determined by total number of individual T-RFs, was
detected in Barrow Canyon Shallow sediments (59 T-RFs) (Figure 4.6A). The other
Arctic sites, as well as Washington Margin sediments, also contain a rich denitriﬁer
community, while the lowest richness was detected in Puget Sound sediments, with only
one T-RF identiﬁed in the 69 cm deep sample at Turning Basin. The West of Juan de
Fuca sample was the most diverse, as determined by the Shannon diversity index (H ’ =
3.12 at 4-5 cm depth) (Figure 4.6B), reﬂecting the high evenness in T-RF distribution
detected at this site. On the other hand, the dominance of T-RF 64 bp in Carr Inlet
sediments combined with a low richness, led to a low Shannon diversity index throughout
the sediments at this site (1.44 < H’ < 2.07). Arctic and Washington Margin sediments
presented a relatively high diversity, coincident with the high richness detected in these
areas.

Cluster analysis of the T-RFLP proﬁles (Figure 4.7) grouped the samples according
to geographic location, with the exception of deep Turning Basin, Carr Inlet, and West of
Juan de F uca samples, which formed a separate cluster, coincident with previous
observations of clustering of Puget Sound samples (Chapter 3). In addition, Carr Inlet top
sample (0-0.5 cm) clustered with Turning Basin samples. Besides these exceptions,
clusters speciﬁc to each site were observed. Furthermore, clusters from close geographic
locations were combined into larger clusters. So, Barrow Canyon Shallow and Deep, as
well as East Hanna Shoal Shallow and Deep clusters, were combined into larger clusters

from each transect, which, in turn, were combined into an Arctic cluster, clearly

128

separated from the Paciﬁc Northwest cluster. The latter was, in turn, constituted of a
Puget Sound cluster, combination of the three site-speciﬁc clusters from this area, and
another one combining West of Juan de Fuca, Washington Margin and deep Puget Sound
samples.

The same pattern detected with cluster analysis, was also observed by principal
component analysis (PCA) (Figure 4.8), which explained 27% of the total variability in
the ﬁrst two dimensions. T-RFLP patterns were grouped according to their sampling
location, with samples from close geographic sites, being in turn placed adjacent to each
other. So, Barrow Canyon Shallow and Deep samples are placed closely together, as well
as East Hanna Shoal Shallow and Deep samples. F urtherrnore, Puget Sound samples form
a compact group adjacent to the Washington Margin samples. T-RFs present only or at a
higher abundance at one site or a group of sites were mainly responsible for the
separation of these samples from the rest (data not shown). So, T-RFs 275 and 391 bp,
present only in Arctic and West of Juan de Fuca samples, were mainly responsible for the
separation of these samples from Puget Sound and Washington Margin samples. T-RFs
299 and 183 bp detected only in Arctic samples, although at low relative abundance, were
mainly responsible for the separation of these samples along the ﬁrst dimension. Barrow
Canyon samples were mainly affected by T-RF 129 bp, in addition to several T-RFs of
low relative abundance, all present exclusively in this area. T-RF 230 bp, present in East
Hanna Shoal and West of Juan de Fuca samples had a main effect in their separation from
the rest along the second dimension. The separation of West of Juan de Fuca samples was
mainly affected by T-RFs detected only at this site, e.g. 198 and 328 bp. Samples from

Puget Sound and Washington Margin were ordinated closely together, coincident with

129

cluster analysis. Although no exclusive T-RF for all these sites was identiﬁed, two T-RFs
of higher relative abundance at these locations (104 and 272 bp) were mainly responsible
for their ordination. .
Canonical correspondence analysis was applied on one T-RFLP proﬁle ﬁom each
site, to study the inﬂuence of biogeochemical parameters in the differentiation of the
denitriﬁer communities (Figure 4.9). The ﬁrst two dimensions explained 67% of the total
variability (43% and 24% in dimensions 1 and 2, respectively). Similar to the ordination
achieved by PCA, Arctic samples were grouped together and the West of Juan de Fuca
sample was separated from the other sites. In addition, Puget Sound and Washington
Margin samples were ordinated closely together, with the exception of Turning Basin.
Higher latitude and oxygen concentrations were correlated to the separation of the Arctic
samples from the rest. The separation of the West of Juan de Fuca sample was mainly
inﬂuenced by the higher water depth and lower organic carbon at this site. In addition,
the higher nitrate concentration also inﬂuenced the separation of this sample, as well as
Washington Margin along the ﬁrst dimension. Higher, i.e. less negative, longitude,
ammonium concentration and temperature were correlated to the separation of Puget

Sound and Washington Margin samples from the remaining sites.

Phylogenetic analysis of nirS gene clone sequences

A total of 649 cloned sequences from ﬁve locations (153, 124, 137, 161, and 74
clones from Barrow Canyon Shallow, East Hanna Shoal Shallow, Shallow Budd Inlet,
Washington Margin, and West of Juan dc Fuca sites, respectively) (Table 4.7) were
identiﬁed as heme cd) nitrite reductase (nirS) sequences longer than 600 bp. These were

translated and used to construct a neighbor-joining tree including closely related and

130

reference database sequences (Figure 4.11). For graphical purposes, subtrees were
identiﬁed as depicted in Figure 4.10 and represented in detail in Figures 4.12 to 4.16. The
divergence range in the cloned sequences was very large, with clones presenting
similarity values between 5.3 to 100% (i.e. 38.8 to 100% amino acid identity). Six
previously identiﬁed clusters (Clusters 1 to 6) of highly similar Shallow Budd Inlet clone
sequences (Chapter 3) were again detected, in spite of the addition of nirS sequences
from additional sites. These clusters included the same nirS sequences as described in
Chapter 3, except for Cluster 2’, which differs from Cluster 2 due to the inclusion of
clones NIRS4A H01 and NIRS4A D03 and the exclusion of clone NIRS4B A10, all from
Shallow Budd Inlet, therefore no inclusion of sequences from other areas was observed.
In addition, 26 other clusters of highly similar sequences were identiﬁed (Clusters A
through Z). These generally included nirS sequences retrieved from one location only,
although several clusters were constituted of sequences from two (Barrow Canyon
Shallow and East Hanna Shoal Shallow) or three (Barrow Canyon Shallow, East Hanna
Shoal Shallow, and Washington Margin) locations. Only 9.9% of the clones were at least
80% similar to cultured denitriﬁers, with the highest similarity (86.9%) detected between
clone NIRSlB C12, retrieved from Washington Margin, and Silicibacter pamerayi DSS-
3, a member of the marine Raseabacter clade. On the other hand, all clones were at least
49.9% and up to 100% similar (60.6% and 100% identical, respectively) to
environmental clones retrieved from marine sediments (Braker er al., 2000; Santoro er
al., 2006; Nogales et al., 2002) or water column (Hannig et al., 2006; Jayakumar er al.,
2004; Castro-Gonzalez et al., 2005; Tuomainen et al., 2003) in other studies, with most

clones (95.1%) presenting a similarity higher than 80% to some previously described

131

environmental clones. Interestingly, four Washington Margin clones and one from East
Hanna Shoal Shallow (NIRSlA GlO, NIRSlA H11, NIRSlB A12, NIRSlA C11, and
EHNIRS2H07) were 100% similar to nirS clones (pB49, pB6, pB46, pA5, and pB66,
respectively) retrieved previously from Puget Sound sediments (Braker er al., 2000).

The in silica digestion of the nirS clones from East Hanna Shoal Shallow, Barrow
Canyon Shallow, West of Juan de Fuca, Washington Margin, and Shallow Budd Inlet
generated 17, 17, 11, 17, and 9 hypothetical T-RFs, respectively (Figure 4.17), almost all
correlated to T-RFs detected in those same sediments, after adjusting for a shift between
real and hypothetical T-RF sizes described earlier (Chapter 3). Similar patterns between
hypothetical and real T-RFs from the same sediment depths were therefore identiﬁed,
with Sorensen similarity indeces of 0.61, 0.59, 0.50, 0.43, and 0.36 for Shallow Budd
Inlet, East Hanna Shoal Shallow, Barrow Canyon Shallow, West of Juan de Fuca, and
Washington Margin, respectively. Furthermore, dominant measured T-RFs generally
corresponded to hypothetical T-RFs generated by a large number of clones, indicating
that some abundant T-RFs are the result of coincidence in restriction site between

different clones rather than by representation of a dominant nirS sequence.

Quantiﬁcation of clone clusters by real-time PCR

The nirS clone clusters previously identiﬁed in the Shallow Budd Inlet clone
library and quantiﬁed in Puget Sound and Washington Margin sediments by real-time
PCR (Chapter 3) were additionally quantiﬁed in Arctic and West of Juan de Fuca
sediments with the same primer and primer-probe sets used previously (Chapter 3;
Griintzig er al., 2001). Clusters 2 and 3, the most abundant clusters in Shallow Budd

Inlet, were below detection limit in all Arctic and West of Juan de F uca sediments (Table

132

4.3). Cluster 1 also was below detection limit at these areas, except for East Hanna Shoal
Shallow, which presented a similar abundance as Shallow Budd Inlet sediments (1.6 x
105 nirS copies per pg community DNA). Furthermore, cluster 6 was detected only in
Barrow Canyon Deep and East Hanna Shoal Deep, but at a low abundance three orders of
magnitude smaller than in Shallow Budd Inlet sediments. Interestingly, cluster 5 was
detected in West of Juan de Fuca and all Arctic sediments, although its abundance was an
order of magnitude lower as in Shallow Budd Inlet. This cluster had been also detected in
all Puget Sound samples, being absent or below detection limit in the Washington Margin
samples (Chapter 3). This suggests that this cluster, though not abundant, might represent
a fairly cosmopolitan group of denitriﬁers. P. stutzeri nirS was detected in West of Juan
de F uca and all Arctic samples, presenting an abundance similar to Puget Sound
sediments and between 1 and 2 orders of magnitude lower than in the Washington
Margin site studied earlier (Chapter 3), reaching a value of 9.0 X 104 nirS copies per pg
community DNA in East Hanna Shoal Shallow sediments, equivalent to 0.04% P. stutzeri
DNA in total community DNA.

In order to identify signiﬁcant differences in cluster-speciﬁc and P.stutzeri nirS
gene abundance in Puget Sound and Washington Margin sediments, the previously
determined abundances from different depth intervals between 1 and 10.5 cm sediment
depth (Chapter 3) were considered biological replicates for ANOVA (Table 4.4). Shallow
Budd Inlet had a signiﬁcantly higher abundance of cluster 2 compared to the other
clusters. This was also true for Turning Basin sediments, with the exception that it was
not signiﬁcantly higher as the abundance of cluster 3. In Carr Inlet, on the other hand, the

abundance of P. stutzeri nirS was signiﬁcantly higher than the abundance of the

133

quantiﬁed clusters. At this site cluster 1 and 3 abundances were signiﬁcantly lower than
in Turning Basin and Shallow Budd Inlet, respectively. Cluster 6 abundance was
signiﬁcantly different in all Puget Sound sites, while P. stutzeri nirS abundance was not
signiﬁcantly different between these locations, but signiﬁcantly higher in Washington

Margin sediments.

Phylogenetic analysis of 16S rRNA clone sequences

In order to study the general bacterial community, 16S rRNA clone libraries were
constructed from the same sediments used for nirS clone library construction (Table 4.2).
A total of 399 non-chimeric 16S rRNA clone sequences longer than 300 bp were
obtained from the studied areas (81, 83, 80, 80, and 75 from Barrow Canyon Shallow,
East Hanna Shoal Shallow, Shallow Budd Inlet, Washington Margin, and Westiof Juan
de Fuca, respectively) (Table 4.7). Assignment of the sequences into taxonomic groups
with the RDP Classiﬁer allowed the placement of 61% of the 16S rRNA clones into a
speciﬁc taxon with a bootstrap conﬁdence estimate higher than 80% (Table 4.5).
Proteabacteria were dominant at all stations, constituting between 33 and 51% of the
bacterial community (i.e. 56 to 72% of the clones assigned to a taxon). The most
abundant classes within this group were Gamma and Delta Proteabacteria, with similar
number of sequences assigned to each of them in all cases, with the exception of West of
Juan de F uca, which presented a lower abundance of Delta than Gamma Pratealiacteria,
signiﬁcantly lower (P S 0.05) than the number detected in East Hanna Shoal Shallow and
Washington Margin sediments (Table 4.6). Although at lower abundances, Alpha
Proteabacteria were identiﬁed at all sites, while Epsilon Proteabacteria were retrieved

from Barrow Canyon Shallow and Shallow Budd Inlet only, being its abundance

134

signiﬁcantly higher at this latter site than at East Hanna Shoal Shallow, Washington
Margin and West of Juan de Fuca (Table 4.6). A single clone, retrieved from West of
Juan de Fuca sediments, was assigned to the Beta Proteabacteria. The second most
abundant phylum identiﬁed in Barrow Canyon Shallow and Shallow Budd Inlet
sediments was Bacteraidetes, presenting a signiﬁcantly higher abundance (P _<_ 0.05) than
in the other areas, where they were detected at a similar abundance as Planctamycetes.
This latter group was detected at all sites except at Shallow Budd Inlet, leading to a
signiﬁcant difference in abundance of this phylum between this site and West of "Juan de
Fuca. F irmicutes, Cyanabacteria, Verrucamicrabia, Nitraspira and Actinabacteria, all
had representatives in at least two of the studied areas, although each constituted less than
5% of the bacterial community at a site. In addition, a few clones were assigned to
candidate divisions WS3 and CPU established in previous studies (Hugenholtz et al.,
1998; Dojka et al., 1998). Between 30 and 51% of the clones could not be assigned to a
speciﬁc taxonomic group with the chosen conﬁdence estimate threshold, these could
therefore, if assigned, increase the abundances of the detected taxons or belong to
different phyla not included in Table 4.5.

Individual phylogenetic trees for different taxonomic groups were built by
neighbor-joining analysis including the 16S rRNA clones, closely related sequences from
public databases and reference sequences (Figures 4.18 to 4.25). The assignment of
clones to individual trees was based on the clustering observed in a general 16S rRNA
tree including all sequences. The similarity between the retrieved clones reached from
41.4 to 100% (59.3 to 100% nucleotide identity), indicating a wide range in similarities.

Clusters of highly similar clone sequences (similarity > 98%) were generally from the

135

same site, however no speciﬁc clustering pattern was observed for less related clones,
with broader clusters including sequences from various different environments. Clusters
of Gamma, Delta, and Alpha Proteabacteria, Bacteraidetes, Chlarabi, Acidabacteria,
Chlaraﬂexi, F irmicutes, Planctomycetes, and Verrucamicrabia included clones from all
studied sites. In addition, clones from at least one, but not all sites clustered with
members of the Beta and Epsilon Proteabacteria, Deferribacteres, Spirachaetes,
Actinabacteria, Nitraspira, F usabacteria, F ibrabacteres, candidate divisions ODl,
OP11, and WS3, and Cyanabacteria. The latter were only detected in the shallower
studied sites, namely Shallow Budd Inlet, East Hanna Shoal Shallow, and Barrow
Canyon Shallow, being detected in the sediments probably due to deposition from the
overlying water column. Several clones were closely related to environmental 16S rRNA
sequences retrieved in other studies mostly from marine environments and also to
cultured strains, though generally more distantly. The closest similarity with a cultured
strain was detected in some Shallow Budd Inlet clones, which were highly similar (up to
99.6%) to Laktanella rasea, an Alpha Proteabacteria recently isolated from sediments of
the north-west Paciﬁc Ocean (Ivanova et al., 2005). The psychrophilic bacteria
F lavabacterium frigaris and Psychramanas antarctica were closely related to clones
from the Arctic sites Barrow Canyon Shallow (97.8% similarity) and East Hanna Shoal
Shallow (94.1% similarity), respectively. Although Beta Proteabacteria and Nitraspira
clusters included only one clone from Washington Margin and one from West of Juan de
F uca sediments, the similarity between cultured strains in these phyla and West of Juan
de Fuca clones was high. The ammonia oxidizer Nitrasaspira multifarmis from the Beta

Proteabacteria and the nitrite oxidizer Nitraspira marina from the Nitraspira phylum

136

were 96.5 and 94.5% similar, respectively, to West of Juan de Fuca clones. Ammonia
monooxygenase (amoA) gene sequences related to Nitraspira had also been detected
previously in Paciﬁc Northwest sediments, although at a different site (Nold et al., 2000).

Although some individual or small clusters of clone sequences retrieved in this
study could not be related to any cultured strain or environmental clone from other
studies, most clone clusters included at least one 16S rRNA sequence from other
environmental clone libraries. These related clones from other studies were often from
close environments to our studied sites, as sediment clones from the Arctic Ocean off the
coast of Spitsbergen island (Norway) (Ravenschlag et al., 1999), which clustered closely
to several Barrow Canyon Shallow and East Hanna Shoal Shallow clones, or seaﬂoor
basalt clones from the Juan de Fuca Ridge clustering closely to clones retrieved from the
same site in this study. However, most of the retrieved clones clustered with
environmental clones from distant geographic locations and often dissimilar
biogeochemical characteristics, including, for example, Mid-Atlantic Ridge (Lopez-
Garcia et al., 2003; Nercessian et al., 2005) and Guaymas Basin hydrothermal vent
sediments (Teske et al., 2002; Dhillon et al., 2003), Japan Trench cold-seep sediments
(Li et al., 1999), Wadden Sea sediments (Mussmann et al., 2005) and water column
(Stevens et al., 2005), and Mediterranean Sea sediments (Polymenakou et al., 2005; Heijs
et al., 2005).

Community description and comparison based on nirS and 16S rRNA clone
libraries

The nirS and 16S rRNA clone sequences were grouped into operational taxonomic

units (OTUS) (Table 4.7) using a 5% and 3% similarity threshold, respectively, to allow

137

the description of the communities at the studied areas in terms of richness and diversity
of phylotypes and to estimate the similarity between communities based on their shared
gene sequences. Most OTUs were represented by only one gene sequence (Figure 4.26)
in 16S rRNA as well as nirS clone libraries. However, these singletons constituted a
larger fraction of the 16S rRNA OTUS (81 to 91%) than of the nirS OTUS (43 to 79%)
from the different sites. On the other hand, the most abundant nirS OTU included three
times more sequences (27 sequences) than the most abundant 16S rRNA OTU (9
sequences), both from Barrow Canyon Shallow sediments. The large ﬁaction of
singletons in 16S rRNA clone libraries led to steep rarefaction curves far from reaching a
plateau at all sites (Figure 4.27), indicating that the sampling effort was insufﬁcient to
cover the large diversity present. Rarefaction curves for nirS clone libraries, on the other
hand, indicated that a larger fraction of the nirS community diversity has been detected,
as a tendency to reach a plateau could be observed. Rarefaction curves also suggest a
signiﬁcantly higher diversity of nirS sequences in Washington Margin sediments than at
the Arctic sites, which in turn present a higher diversity than Shallow Budd Inlet and
West of Juan de F uca sediments. This is consistent with the Shannon diversity index and
the Chaol nonparametric richness estimator results (Table 4.7), which indicated-that the
Washington Margin sediments contained the most diverse (H ’=3.94) and richest
(Chaol=348) nirS-containing community, followed by the Arctic sediment samples.
West of Juan de Fuca sediments presented the lowest nirS richness and diversity
(Chaol=41; H ’=3.07). Although Shallow Budd Inlet’s richness estimate was similar to
East Hanna Shoal Shallow and more than twice as high as for West of Juan de Fuca, its

diversity was as low as at the latter site. This is probably an indication of a less even

138

distribution of nirS sequences at Shallow Budd Inlet, which lowers its diversity. At all
sites, richness and diversity was higher for 16S rRNA than for nirS gene sequences, with
Washington Margin sediments presenting again the highest values (Chaol=574;
H’=4.18), indicating that these sediments contained the richest and most diverse bacterial
community from all studied areas. Although all the remaining sites presented similar
diversity values, Barrow Canyon Shallow sediments contained a richer bacterial
community, though less evenly distributed.

The 16S rRNA as well as the nirS gene libraries from all sites were signiﬁcantly
different from each other (P<0.0026; a=0.05) as determined by the comparison of the
difference between the homologous and heterologous coverage curves for each pair of
libraries with the difference obtained after multiple randomizations of the sequences in
the libraries (Table 4.8). This suggests that samples from sediments used to construct the
different libraries may harbor distinct general bacterial as well as nirS-containing
denitriﬁer populations.

The signiﬁcant differences between communities from different areas are further
supported by the generally low similarity between libraries, as determined by the
abundance-based Sarenson similarity index (Labund), as well as the low richness
estimation of shared OTUS between sites, determined with a nonparametric richness
estimator analogous to Chaol (SM; Chao) (Table 4.9). West of Juan de Fuca as well as
Shallow Budd Inlet libraries (16S rRNA and nirS) presented extremely low similarities to
all other libraries, with similarity values often equal to zero. This was correlated to the
absence or the presence of only few estimated shared OTUs between each of these sites

and the others. On the other hand, East Hanna Shoal Shallow, Barrow Canyon Shallow

139

and Washington Margin were more similar to each other and presented a higher number
of estimated shared OTUS. The highest similarity (Labunf0.676) and estimated richness
of shared OTUS (S A.B Chao=47) was observed between the two Arctic sites for the nirS
gene, consistent with the observation of several clusters of high similarity including
sequences from these two areas on the phylogenetic tree (Figures 4.10 to 4.16). The
estimated number of shared OTUs represent, however, only a 42 and 48% of the
estimated richness of nirS genes at Barrow Canyon Shallow and East Hanna Shoal
Shallow, respectively, consistent with them being signiﬁcantly different as determined
before (Table 4.8). For the 16S rRNA libraries, the highest similarity (LabumFO-239) and
estimated richness of shared OTUS (8A3 Chao=45) was observed between East Hanna
Shoal Shallow and Washington Margin sediments. However, the number of shared OTUS
represents only 21 and 8% of the estimated richness in the former and latter sediments,
respectively. The two Arctic sites presented the next highest similarity and richness of

shared OTUs estimation.

140

470' 465' 460' 45' siﬂ' 445' 440' 43'

I East Hanna Shoal Deep

     
  

    

Barrow Canyon Deep

     
     
   
 

72' I 72'
East Hanna Shoal Shallow .
| . 't-....esv--iy

 

Puget Sound (Shallow
Budd Inlet, Turning
Basin and Carr Inlet)

 

 

.

West ofJnan de Fuca
. 9 la. .27 -.
Juan de Fuca ridge Vin: ‘_ .

440'

 

     

‘ . S Washington Margin

   

 

_1‘

km

 

Figure 4.1. Locations of sampling stations.

For detailed view of Puget Sound stations refer to Chapter 3 (Figure 3.1).

Table 4.1. Locations and characteristics of sampling stations.

 

 

Water Or anic Sulfate
. . g Temp. Salinity reduction rate
Station Location depth carbon 0 .2
(m) (°/) ( C) (PSU) (mmoles m
0 day")'
Shallow 47°04.99'N
Budd Inlet 122054.1o'w 3 2'5 19'” 26‘2 12
Turning 47°03.13'N
Basin 1220549995,“, 12 3.2 14.80 21.2 25-
47°17.21'N
Carr Inlet 12294295,“, 84 1.8 16.80 29.6 7.5
Washington 46°25.57’N
Margin 12404150,“, 1138 3.1 3.65 34.3 86 0.04
West of Juan 46°47'N
de Fuca 133040,“, 3869 <0.5 1.56 34.661 ND
Barrow
71°36.24’N
Canyon 156°12.49W’ 186 1.6 -0.87 33.854 1.19
Shallow
Barrow
72°12.33’N
Canyon 154°5.56’W 2000 1.7 -0.402 34.942 0.3
Deep
East Hanna
72°39.9’N '
Shoal 158°44.7’W 160 1.9 -1.32 33.407 1.5
Shallow
East Hanna 72°53.64’N
Shoal Deg) 158°15.78’W 1450 1.2 -l .66 32.601 0.05

 

3ND, not determined.

142

Table 4.2. Sediment samples analyzed from different sites.

 

Sediment depth analyzed (cm)

 

Analysis

 

Barrow Barrow East Hanna West of
East Hanna
method Canyon Canyon Shoal Juan de
Shoal Deep
Shallow Deep Shallow Fuca
T-RFLP 0-2 0-2 0-2 005 0-1
2-4 2-4 2-4 1-1 .5 _ 1-2
8-10 8-10 4-6 2-3 2-3
20 20 6-8 5-6 3-4
8-10 9-10 4-5
10-12 7.5-8.5
12-14 20-21
14-16
16-18
18-20
Cloninga 2-4 4-6 2-3
Real-time 2-4 2-4 4-6 2-3 2-3
PCR

 

aAlso sediments ﬁ'om Washington Margin (2-2.5 cm) were used for cloning in addition to analyses
described in Table 3.2.

143

Figure 4.2. Pore water proﬁles for O; and N03” in Arctic (Barrow Canyon Shallow,
Barrow Canyon Deep, East Hanna Shoal Shallow, and East Hanna Shoal Deep)
sediments.

144

warm? 29 1° 9° .0
0219M) 0 50 100 150 200 250 300 350

 

 

 

 

 

o ..
l
A I
.5. 1 ~
5
§ .,
2 4
Barrow Canyon Shallow
3 l A l a 4
N0,‘(uM) 9 20 40 so so

 

02(pM) 0 50 100 150 200 250 300 350

0; =3 #‘ =
5.;

10 <1

 

 

.1
I

20‘
l

Depth (cm)

25 «,
Barrow Canyon Deep

 

 

30 .l

 

NO,'(pM) 0 20 40 60 80

OzluM) o 50 100 150 200 250 300 350

 

Depth (cm)

 

 

East Hanna Shoal Shallow

I; 1 4L %

N0,'(uivi) o 20 40 60 80

4L

0201”) 0 50 100 150 200 250 300 350
0‘

 

5!

101
I

15 1'

Depth (cm)

20‘

 

25 1
30 , East Hanna Shoal Deep

 

 

 

 

L

 

 

145

Figure 4.3. Pore water proﬁles for F e(II), Mn(II), and NH4+ in Arctic (Barrow
Canyon Shallow, Barrow Canyon Deep, East Hanna Shoal Shallow, and East Hanna
Shoal Deep) sediments.

146

F0000"). M00001") 0 29° 4° 9° 9° 19°
NH;(uM) 0 100 200 300 400 500 600 700

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

° 1
A 5:
E
2
i 10 .
8
15 1
Barrow Canyon Shallow
20
Fe("KIJ-Mli Maintain)? 29° 49° 59° 89° 109°
NHfhtM) o 20 40 so so
0 L l . . c
10 -
’e‘
2
i 20
8
30 .
Barrow Canyon Deep
40 . .
Fe("XuMI. M00001") 0 59 19° 19° 29° 2"” 39°
NHJpM) 0 20 40 60 80
E
2
8
East Hanna Shoal Shallow
25 r 1 n
Fe(ll)(pM), Mn(l|)(pM) o 50 100 150 200 250 300
NHﬂpM) 0 20 40 60 80
E
2.
5
D.
8
+ NH:
20 « ' —-— Fe(Il)
East Hanna Shoal Deep —o-—- Mn(Il)
25 ‘

 

 

 

147

N0,'(plvl), NH,*(uM) o 20 4o 60 so

 

0201'“) o 20 40 60 80 100 120 140 160

 

  

 

Depth (cm)
a

 

 

 

 

 

20 i + 02
25 %, —-— No,‘
30 9 West of Juan de Fuca ° ””4

 

Figure 4.4. Pore water proﬁles for 02, N03" and NHI' in sediments from the abyssal
sea floor west of the Juan de Fuca ridge.

148

 

 

EVI/I/I/I/II-‘ ‘ » \\\\\\\\\\\‘s§ 444'; .\
_E'l/l/l/I/l/A I4 -.1 “\\\\\\\\L+f ""WTW sg . I ~—
=7/l/IIIII- u. Iii! p» " a 58 as
g e-a:-_IIIIIIII;1II1 :1m\\\ M . aw _W A 1
m o = ”ll/Il/IIIIA- L» \\ _ . s» 1 I 72 a 76
:l/l/l/l/l/I/l/I/Ifl' . :\ t- \\ :6, 1 p
. _\ ‘ a 83 n 86
_ :I/I/II/II/III/II/III ' s\\\\‘v+7__=_ '93» .1 m 91 a 97
=II/IIIIIIIIIIIIIII-1-J . , l
a \\\\\\\\\\\\\\‘ am we: : .102 E104
a, :-\\\\\\\\\\\\\\\\ aw» ' ‘D 108 Um
I \\\\\
"4 g,“\“\{*?_"_’"' [3113 I116
12129 I137
33 I 141 I 158i
war—I; , z
,_ 1 1m161l173
A Q III/I In; :Illlll ' _ 1
E o )m M01 1: 176 I 194
2. m IIIII.
3198 I 2041
5 1 [:1 208 a 21 o i
E LL. ‘
b g a 230 a 232
3 1:1 234 236
E I 238 I 257
.. rIIIIIIIIII;-: am _ ' , 1
g E III/III/I/IIZ" -:--III 1 Kiss: 0265 C1270
a 3 'l/I/I/I/I uIII III ‘ ' v '
IIIIIIIIIIIIIII.. .II-I ‘ a 272 a 275
'l/I/I/I/I/I/I/I/I ‘ : 1
III/IIIIIII :1 -:-::n l- 1 [:1 281 n 295
- III/IIIIIIIIIIIIIIIIIIIIIIIIIII IIIII ' ' : _- 1:3 297 I 317
o =II/I/III/IIII/IIIII/IIIIIIII .- . V 1
-_IIIIII/IIIIIIIIIIIIIIIIIIIII l a 328 I 332
Vl/I/I/I/I/ll/I/I/I/I/I/I/I/II- .
V/I/I/I/I/I/I/I/Il/I/I/l/l/l/l/I/I/I/IIlEl i I 335 343
III/IIIII —mm:
VI/Illllli-l—Ilﬂllllll mm 7 7 j i a 353 " 356
m III/IIIIIIIIIIrel—mt _— w .- g 15380 E11383
w x ‘ 1.,,”"1
taxman» 1:1 433 E 436
an» ‘
,9 ”EH m ‘I 462 540
:23 546 I 589

 

 

 

Relative abundance of T-RFs (%)

Figure 4.5. T-RFLP profiles obtained by ampliﬁcation of nirS genes from Arctic
(EHS, East Hanna Shoal Shallow; EHSD, East Hanna Shoal Deep; BC, Barrow
Canyon Shallow; BCD Barrow Canyon Deep) and Paciﬁc Northwest (WJF, West of
Juan de Fuca; WM, Washington Margin; CI, Carr Inlet; SB, Shallow Budd Inlet;
TB, Turning Basin) marine sediments from various depths.

Presented T-RFs were generated after restriction with HhaI. Numbers in legend indicate
T-RF length in base pairs for fragments representing more than 2% relative abundance.

149

Richness (number of T-RFs)

 

 

 

 

 

 

 

 

 

 

 

 

0 10 20 30 40 50 60
0 ‘ , -, a L , g 1
‘r - ?"
20 q. . _ 1 ,_ _ ,
4o _l,~._~, _ _____.__ - , _ _‘—_s_., , ﬂ_
1 +TB
+83
E 60 - , ~ 7 k an +CI
3 / —x—WM
.: +WJF
H
0. +30
8 80 4 *ﬁ * “‘ ‘ +300
+EHs
+EHSD
100 + - . _.___.___ sewn
120 -- ~ —vw #77 , ifﬁmgW
140

 

 

 

Figure 4.6A. Richness of the nirS-containing community in Paciﬁc Northwest (TB,
Turning Basin; SB, Shallow Budd Inlet; CI, Carr Inlet, WM, Washington Margin;
WJF, West of Juan de Fuca) and Arctic (BC, Barrow Canyon Shallow; BCD,
Barrow Canyon Deep; EHS, East Hanna Shoal Shallow; EHSD, East Hanna Shoal
Deep) sediments from different depths, based on number of individual T-RFs
detected after restriction with Hhal.

150

Shannon diversity index
0 0.5 1 1.5 2 2.5 3 3.5

 

 

 

 

+TB
+88
+CI
-x-WM
+WJF
—o—BC
80 *- ___,,,, * " *—--' r _ —e—BCD
+EHS
—e—EHSD

 

 

Depth (cm)

 

 

 

 

100; ~ ~~~ -~~ ~ —-

 

 

120~——-— ——~**~ , —*A~~~~~———~——~*4

 

 

 

 

 

 

 

 

 

 

140

Figure 4.6B. Diversity of the nirS-containing community in Paciﬁc Northwest (TB,
Turning Basin; SB, Shallow Budd Inlet; CI, Carr Inlet, WM, Washington Margin;
WJF, West of Juan de Fuca) and Arctic (BC, Barrow Canyon Shallow; BCD,
Barrow Canyon Deep; EHS, East Hanna Shoal Shallow; EHSD, East Hanna Shoal
Deep) sediments from different depths, as determined by the Shannon diversity
index based on the T-RFLP results. The T-RFLP analysis was performed with
restriction enzyme HhaI.

151

 

54

405

2] 135 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MN
MN

?@OAN
‘wdmwmeQ

oawAummm
QM
m

wwmwmg
95.4%:
010101

?

o
. b
0001
u

-‘¢?&#?o

Migomm

m

AN-‘NDOJ
N
(.71

saw
'01

———_CDCDCDCDCDCD
_DN
01

dN-‘(AJ
:w9m

wwﬁﬁnnmmmmmmnqqqqq
ommm.

D

”#40
. o
pewmd
mm D

EHS

0"
#00000"

m
I

.Qodmmm
N I O I O
o

m
I
U)
_L
U)
_s
m

EHS1448
EH31244
EHSlOJZ
EHSBJD
EHSSG
EHS4£

EH824
EH502

Figure 4.7 . Dendrogram obtained by cluster analysis of T-RFLP profiles.

Hierarchical clustering analysis was performed applying Ward’s method on the
Serensen’s distances between T-RFLP proﬁles from Paciﬁc Northwest (TB, Turning
Basin; SB, Shallow Budd Inlet; CI, Carr Inlet, WM, Washington Margin; WJF, West of
Juan de F uca) and Arctic (BC, Barrow Canyon Shallow; BCD, Barrow Canyon Deep;
EHS, East Hanna Shoal Shallow; EHSD, East Hanna Shoal Deep) sediments from
various depths. Numbers next to sample names indicate sediment depth in cm.

152

 

 

 

 

 

 

 

 

 

 

 

x 6 ,
%X 0‘ XWM
xXO 4 4 mo . OTB
3 ° 0 e 088
E: ‘2‘ 0‘ ACI
;; I A 0 TO , AWJF
E -10 -5 2 o 5 10 1F 0 BCD
5 Eng? . 0 BC
A AA .4 - i D EHSD
am 'I I EHS
5; I
I
#3
Dim 1 (15%)

 

 

 

Figure 4.8. PCA ordination of T-RFLPs of nirS genes from Pacific Northwest (WM,
Washington Margin; TB, Turning Basin; SB, Shallow Budd Inlet; CI, Carr Inlet;
WJF, West of Juan de Fuca) and Arctic (BCD, Barrow Canyon Deep; BC, Barrow
Canyon Shallow; EHSD, East Hanna Shoal Deep; EHS, East Hanna Shoal Shallow)
sediments.

Percentage of variance explained by each dimension is presented in brackets. Dim,
dimension.

153

 

WM

1 ‘ Organic Carbon f3 °
Temperature Ammonium
Longnude

     
     

N'
Salinity itrate

  
  

_05 1 Latitude BC.D '3‘:

Dim 2 (24%)
O
O
a

' Water Depth
4 ~ EHSD

-2 - WJF

 

 

 

Dim 1 (43%)

Figure 4.9. Canonical correspondence analysis (CCA) based on T-RFLPs of nirS
genes and environmental variables from Pacific Northwest (TB, Turning Basin; SB,
Shallow Budd Inlet; CI, Carr Inlet, WM, Washington Margin; WJF, West of Juan
de Fuca) and Arctic (BC, Barrow Canyon Shallow; BCD, Barrow Canyon Deep;
EHS, East Hanna Shoal Shallow; EHSD, East Hanna Shoal Deep) sediments.

Dots and vectors represent the sampling stations and environmental variables,
respectively. Percentage of variance explained by each dimension is presented in
brackets. Dim, dimension.

154

 

 

 

Subtree E
Subtree D

f

 

 

 

—:
H:

 

 

 

 

Subtree C

 

 

E
P
E.—

(_— Subtrge B

 

 

 

 

 

 

 

 

 

 

Li Subtree A >

 

 

 

 

 

 

 

 

 

L— Azospirillum brasilense Sp7 (AJ224912)

l——|
0.1

Figure 4.10. Schematic representation of the nirS phylogenetic tree shown and
described in further detail in Figure 4.11.

For graphical purposes, detail of subtrees A, B, C, D, and E is presented in Figures 4.12,
4.13, 4.14, 4.15, and 4.16, respectively.

155

Figure 4.11. Phylogenetic tree showing the affiliation of nirS clone sequences
retrieved from Shallow Budd Inlet (SB), Washington Margin (WM), West of Juan
de Fuca (WJF), Barrow Canyon Shallow (BC), and East Hanna Shoal Shallow
(EHS) sediments to selected reference sequences.

The tree was generated by neighbor-joining analysis of approximately 215 aminoacids
with Azaspirillum brasilense Sp7 as the outgroup. Values at branch points indicate the
percentage of 100 replicate trees supporting that branch. Bootstrap values below 50%
were omitted. Names of clones retrieved in this study begin with NIRSl (Washington
Margin), NIRSZ (Barrow Canyon Shallow), NIRS3 (West of Juan de Fuca), NIRS4
(Shallow Budd Inlet), and EHNIRS (East Hanna Shoal Shallow). Clusters of highly
similar m'rS clone sequences are identiﬁed by numbers or letters, with quantity of clones
in the cluster and retrieval site indicated in brackets. Clusters identiﬁed by numbers were
previously identiﬁed (Chapter 3). Cluster 2’ is equivalent to Cluster 2 (Chapter 3), except
for the inclusion of clones NIRS4A H01 and NIRS4A D03 and the exclusion of clone
NIRS4B A10. The scale bar represents 0.1 substitutions per sequence position. sed.,
sediment; cl., clone; Wash, Washington; col., column. .

156

I so 1;“: Cuprgvli’dli’s nemlAr r)(X91394
l Ralstonia metallidurans CH34 (NZ AAAI03000011)

 

 

 

            

1 9 L Clusterz 2(3 WM clones
ﬂ NIRS1B 001
L ” ’ Huntington Beach sed. cl. th SD (DQ159648)
l -— A” NIRS1A 003
‘51 r EHNIRSZA03
1 Huntington Beach sed cl hbD 5A (DQ159611)
l 1;: NlRS1A
11 100 NIRSlBC H06
1L_,vJ NIRSSA cos
,‘ 11m ‘NIRSSB (307
1_ 93 1LNIR51ACOZ
1 .1 7‘7—1 NIRS1B D08
1 H L * NIRS1BG03
; _ NIRS1A B12
LH ‘ NIRS1A 303
1011 5* l} 2% EHNIRS1A02
7 a 99 NlRSZB H02
1 1“: EHNIRSZEO7
1, 1 ‘7 NIRSZAH10
9. 11“ Huntington Beach sed. c1 mm 40 (DQ159496)
7‘ j (NIRS1B
;: EHNIRS1H01
r NIRS1AA11
, 'u NlRS1A DOB
90 r NIRS4A E11
’ *i A Hydrogenobacterthermophilus(A8210046)
E RS1GO1
53 ‘L NIRS1A DOZ
NIRS1BCOQ
97 NIRSZA G11
,3 Lr NIRS1A E06
v L—“NIRSZA F05
,, _ a, 1 1 F EHNIRS1E09
so - a ,, l___J NiRs1eBoa
100 ‘NiRs1B F05
’ Wash. margin sed. d. wA20 (AJ248428)
_r NIRS1B E08
‘ 11. * Puget Sound sed. cl. p816 (AJ248420)
' ~ ** , ¥ *NIRS1B 1305
l __ ——* #’ "# Arabian Sea water col. cl. 6840-5A (AY336820)
L»— , ,- Too < 31 _791 :NIRS1ADOS
F 7—1 ‘~ NIRS1B B10
1. l7 NIRS1BH10
1* LB, ¥ NIRS1A H03

1 u Puget Sound sed. d. p36 (AJ248419)
1:H Puget Sound sed. cl pA17 (AJ248407)
1 NIRS1B
‘ ‘ NIRS1A H11
1 L NIRS1A £02

NIRSZB A05
1 ILNIRS1B F06

NIRS1B A08

ll NIRSZA GOQ

1 1 NIRSZB G09

1 NIRSZAA11

ELM NIRSZB CO1

5 1,1 Baltic Sea water col. d. M150-85 (DQO72183)
2 ‘ Baltic Sea water ool. cl. MSG-54 (00072203)

54—

06‘5"— Thauera selenatis AX (AY078264)
l ’ _Thauera aromatica K172 (AY078256)
10°—Pseudomonas stutzeri ZoBell (X56813)
,7 L.-. Marine denitrifying isolate E4- 2 (AJ248398)
100 Alcaligenes faeaaI/sA15 (AJ224913)

—Az “ll bra '/ nse 7 N224912
ospmfum 816 a) ( )

0.1

157

 

 

86

 

0.05

 

55 ,1 ’ Subtree B

l 1EHN|RSZE01

as “' 1%L Cluster U (2 WM, 11 BC and 6 EHS clones)
' Cluster V (1 BC and 2 EHS clones)

 

 

 

f ‘ " ' “ ’ * Huntington Beach sed cl. th 4G (DQ159571)

Thiobacillus denitriﬁcans (NC 007404)
100 1 H Pseudomonas aeruginosa PAO1 (AEOO4488)
L‘ ‘“ ’ “w Pseudomonas ﬂuorescens (AF 197466)
1—* —— “m ‘— Cluster W (4 BC clones)
" Marine denitrifying isolate B9-12 (AJ248393)
"‘——“‘" ‘ Marine de nitrifying isolate 04-14 (N248395)
,0 ____#' -___5. Marinobacter aquaeoleiVTB (NZ AALGO1000002)
99 L——-~~ Marine denitrifying isolate 010-1 (AJ2483Q4)
l I ‘ ‘ —_—Arabian Sea water col. cl. GB40-4F (AY336818)
l
1
1

 

 

 

 

 

EHNIR82F04
5555 ‘~°° L NIRSZA 004
1 —- EHNIRS1E11

 

,, L —-- , NIRS1B 003
1 92 1 ~_ NIRSZAA10
68 ; ,, , --55__ NIRSZA F04
91 1, ,, __ EHNIRSZBOS
100 H NIRSZA H01
LNIRSZB H01
” ' Cluster 4 (4 SB clonesI‘
I 82 ”ﬂ Huntington Beach sed cl DB 26(00159527)
H ’" ﬂ” *‘ ,,___ Roseobacter denitriﬁcans (AJ224911)
‘1—41 EHNIRS1FO9
1 100 EHNIR82H08
11 '— Silicibacter pomeroyi USS-3 (CP000032)
’1 : --- NIRS1B H05
.1 100 _ ‘ Huntington Beach sed. cl. hbD 2F (DQ159599)
1 J5 A ~ - NIRS1A G12
‘____ ‘ L 'H" _,“_ NIRS48 A12
34 1 Paracoccus denitriﬁcans PD1222 (005002)
100 fParacoccus pantotrophus (AJ401462)
I EHNIRS1GOI3
f EHNIRS1AO4
1*, EHNIRS1D11
5, 15 EHNIRS1AO7
" Cluster X (1 WM, 2 BC and 3 EHS clones)
__512 ClusterY (5 EHS clones)
I -I 5- NIRSZB E08
1 EHNIRSZFO1
1 I ~* NIRssA E01
r" ‘ NIRS1B C12
r ‘ ’4 Huntington Beach sed. cl. hbD BD (DQ159624)
151" ~~— NIRS1B H01
1 1 5— NIRSBA E02
151 I H NIRSSB GO8
NIRSBAA12
I' — NIRS1B F04
NIRS3AA1O
118—IL NIRSSA F06
“I 1— _ NIRS1AE10
1 __1 NIRSSA F10
11 L -5 NIRSSA H06
11* NIRSBB A06
‘i1’ NIRS3A 604
r NIRS1B A04
11 .555 NIRSSB G11
L1; * NIRS3A 804
' NIRS3B A11

531

159

Figure 4.12. Detail of Subtree A as described in Figures 4.10. and 4.11.

The scale bar represents 0.05 substitutions per sequence position. sed, sediment; 0].,
clone; col., column.

158

Figure 4.13. Detail of Subtree B as described in Figures 4.10. and 4.11.

The scale bar represents 0.05 substitutions per sequence position. sed, sediment; cl.,
clone; col., column; E.S. Pacif. OMZ, Eastern South Paciﬁc Oxygen Minimum Zone.

160

 

LA} ‘ ‘ “-9;— NIRS4A F0§Ubtree C
1 I‘— Cluster P (4 WJF clones)

“ “#3; I EHNIRS1BOG

. IEHNIRSZAOZ
93

I EHNIRS1E06

 

11 N1RS18 010
—Huntington Beach sed cl th 66 (00159579)
1 NIRS1B 007
99.5 NIRS1A E03
A L NIRS1B 006
1 — 7*1 . NIRS1A 808
1 ‘--—- N1R838 E12
5. _AI 1—-—- -—- 7- ,_ Huntington Beach sed. cl. hbD 4C (00159507)
113 i - 90 NIRS3A 805
L115 7 5&0 IL NIRS3B H05
1 I NIRS3B 809
.9951w NIRS1A H07
,5 1 9‘ II '- NIRS1B H03
1— NIRS3A 808

‘2 DI 'NIRSSB H01
76 Cluster Q (1 5 WJF clones)

, 7 94,1 "‘ ' " NIRS1A F04
I L——— Huntington Beach sed. cl. th TG (DQ159582)
14‘ -- —*~ -- - 'NIRS1A 012

 

 

_ _ 75 _r:—"“——NIRS4A (304

1 [~71 1 Huntington Beachsed. cl th BG(DQ159557)
‘ 1
1 ;_ NIRS1A 806
I
I
l

 

64 .1" "~ EHNIR82F12
. rI L NIRSZB 002
5__ 5. - * NIRS1A E04
“7 L— NIRSiB A03
. . . -- - --‘— EHNIRSiD10
' "7* Azoarws talulyticus 2FBG (AY078272)
"" Cluster R _(1538 WM clones)
SPacif OMZwaterooI d AC100—135 (AJ811517)

 

 

 

I if_‘ EHNIRS1GO7E
. ___ _JI NIRSZA 806
I1 100 NIRSZA F03
. ' " " Arabian Sea water col. cl. V483 -3EE (AY336925)
”ﬂ III—1”? NIRS1A DO1

 

1
I
I r4
I
1

Arabian Sea water col cl (5857-781AY336858)
1 HI LI; 1" EHNiRSS16804
EHNIR 1 10

I 10° LL“ NIRS1AA06
"‘ “m ‘ ‘ —‘TAtabian Sea water col. cl. V483-8H (AY336912)
I 611,71 EHNlRS1807

1 EHNIRS1GOG

M ‘ ‘ NIRS2B GOG
1’" I EHNIRszH11
1 NIRS1B E02
I 5’ As 1- NIRS1A BO4

 

 

 

1 I L—“‘NIRS1A 805
z 1 #555 Huntington Beach sed. cl. hbD 26 (00159600)
_I EHNIRS2F06
1 III rrrrr -- NIRS1BCOZ
'1 EHNIRS1BO3
I EHN1R32812
1. 5, N1RS1BA06
: I “‘7'”— NIRS1B 005
. I EHNIRS1E03
EHNIRS1007
* , ’NIRS1A H08
EHNIRSZGO1
.__.ea_; - —~N1RS1A cos
— NIRSiA (302
_I EHNIRS1E04
EHNIRS2F08
I EHNiRsone
IEHNIRS1A06
1;“ IEHNIR81C08
I” 8 11 'EHNIRS1A05
EHNIRS1B11
III EHNiRsona
EHNIRS2D10
:— EHNIRS1012
I1 EHNiRsons
I IIEHNIRS1DO4

I

 

 

 

. ., EHNiRszoos
I I ‘1 NIRS1B A10
'1 ‘NIRS1B H07
6‘ 1. EHNIRSZG10
7‘ -_ NIRS1B F09
MIII NIRSZB A03
NIRSZB c02
7' Cluster T (17 WM clones)

161

ﬂ“— ree E
77 Su btreells
NIRS1AA04

NIRSZA 807

109 'NIRSZA F01
62 I II NIRSZB E07
1 I 1 83 LNIRSZB H11
. 71 —7 EHNIRS1F12
77 I LIV—I— NIRSZB F09

7 - ___I 7—7NIRSZAA03

I 55 17 7 _ Baltic Sea water col. cl. M60-145 (DQO72215)
-— EHNIRS1CO4
1 - - - - - - -—— NIRS1A CO7
I ~ ~-—~- --r- -~ NIRSSB H10
’ '53' EHNIRS1G08

JP NIRSZB 607
1 96 L EHNIRSZEOS
”'7 7 777* 7' 7 Cluster M (7 BC and 2 EHS clones)

NrRszAA07
Puget Sound sed. cl. p820 (AJ248421)
I' NIRSZB D10
99 NIRSZB F11
71 1" I NIRSZB F08
.- EHNIRS1E07
L - -— NIRS2B 007
We I— EHNIRSZDO4
i‘II mmam1
‘ NrRssA (308

59 L__

1 as LI NIRSSA 806
1
1

 

 

 

 

 

 

I

91 't——- NIRSSB F04
I NIRS3A B10
51 , NIRSZA (:05
' . ' NIRS3B F02
' 53 r— NrRsaAAoe
"—TLI NIRS3AA05
12'--— NIRS3B B10
117-77 Cluster N (9 WM clones)
' .‘EHNrRszcrz
f__1 EHNIRSZA07
2 L~- NIRSZA E08
—— NIRS1A H10
'NIRS1B B11
77 r—— EHNIRszoos
1 L—— NIRS3A C11
Arabian Sea water col. cl. GB40-6A(AY336822)
I as I EHNIRS1B10
' 55 JL NIRSZB 005
1. I ’ EHNIRSZB10
8° 6‘iIII_N18328 803
NIRS1B E04
, EHNIRSZDOZ
. 01 ~-w NIRS1AA05
- as {~— NIRS1A E12
ﬁg ___“ NIRS1AA08
I L 61 L—~ NIRSZA (304
I ‘ .7 Arabian Sea water col cl V483-4E (AY336904)

7 '93 7 Cluster 0 (5 BC and 2 EHS clones)

 

 

 

 

 

 

 

 

I52__

 

rt -~1

0.05
Figure 4.14. Detail of Subtree C as described in Figures 4.10. and 4.11.

The scale bar represents 0.05 substitutions per sequence position. sed, sediment; 0].,
clone; col., column.

162

Figure 4.15. Detail of Subtree D as described in Figures 4.10. and 4.11.

The scale bar represents 0.05 substitutions per sequence position. sed, sediment; 0].,
clone; col., column; Wash., Washington.

163

so 1'{ -~ EHNIRS2E05
I" NIRSZA H02
. ._ Wash. margin sed. cl. wA15 (AJ248427)
Lu NIRS1AA1O
56 :1 95 L NIRS1A 006
f1. ; Cluster l (1 BC and 3 EHS clones)
98 ; {EHNIRS2801
' {*NleﬁincgaiAos
L -—1 .r-- NIRSZA £11
6' = 66 EHNIRSZCOS
! Ln NIRSZA 001
, I :— NIRSZB 805
1 99 ‘ NIRSZB H03
1 . ,100 1—*"'N|RS4A 810
1 t ‘ ~ Huntin ton Beach sed. cl. th 6F (DQ159546)
| a too 7 IRS4A 806
, 9' 99 8.2 T: NIRS4AC11
{ 62 l 4. NIRS4A 806
E t’ —— Huntington Beach sed. cl. th 9F (DQ159587)
; 72 l 1 -ﬁ~ NIRS1A C10
' 1 “snares
' 9° “Ct—EHNIRSZGO6
,0 or NIRS4B F09
I 93 L Puget Sound sed. cl. pA33 (AJ248412)
_29_J NIRSZA H05
.g t t— Baltic Sea water col. cl. M60-156(DQO72220)
9, {___ _1 EHNIRStDOQ
96 LJ— EHNIRS2006
93 77 1— EHNIRSZDOB
ﬁJ [ , 92 ; ~- Cluster J (4 BC clones)
c, 1 ‘ NIRSZA H03
1 La”, f_) NIRSSAD12
, mo 1_r~w~— NIRSBB 510
,. 89 L NIRS3B H09

 

 

( ,, M NIRSZA 001

1 , J , 190 tNlRS3B 801

L L_ 1 'NIRSSB 007

t 9917—— NIRSSB B12

; 1 5|— NIRSSB C06

‘ 55100 — NIRS3B F03

) 84 ; Cluster K (4 WJF clones)

3 viii L—a Wash. margin sed. cl. wF16 (AJ248437)
L ' L_; NIRSSB A05

 

 

100 ‘ NIRSBB E09

i
__ t_‘ _

t . ~ _. NIRSBA H02
i “411 » ~ NIRS1ADO3
; 122‘ — NIRStB D12

7‘ 89 r NIRS1A GOG
1 99 9‘ 1' 'NIRS1B c302
‘ ‘ 1 NIRS1B 804
t NIRS1B G11
1 L NIRS1A F12
y—“h - Huntington Beach sed. cl. th 4C (00159534)
1 L824 NIRSZB 009
W _J ‘NIRSZB H09

 

u.__1 NIRS3A 1503
100 L4 NIRSSB 007
76 ) NIRSBA 004
[a 19’1" NIRS3BE11
‘2 76,1 NIRS3B 004
1 Lar— EHNIRS1H05
1 1 75 L130“— NIRSZB G10
" 54 ‘r— Cluster 5 (4 SB clones;
‘ 7 _ _ 1 Huntington Beach sed. cl. hb 7C (DQ159549)
;__; 100 L Huntington Beach sed. cl. th 1E (DQ159522)
86 1 ;' ‘—_ ’ NIRS4A H05
~ : ea NIRS4A 001
82 .., 1NIRS4A E12
72 ‘85“ ‘ Cluster1 (21 SB clones)
1————~—+
0.05

 

164

Figure 4.16. Detail of Subtree E as described in Figures 4.10. and 4.11.

The scale bar represents 0.05 substitutions per sequence position. sed, sediment; 0].,
clone; 001., column; cyanob., cyanobacterial.

165

- Cluster 2' (29 SB clones)
NIRS4A H09
NIRS4B 611
——— NIRS4A Ho7
NIRS4A 802
NIRS4A 603
NIRS4B 011
9r __NIRS4A 012
L‘_NIRS4A Hoz
w NIRS4A 01o
Loo NIRS4A
11'_N Cluster BE(11013 SB clones)

1‘1‘BRiter Cohe estuary sed. cl. ANIS—54 (N440470)
Puget Sound 2sed. cl. pA12 (AJ248405)

F ' NIRS4A
F1: NIRS4A F11
NIRS48 602
_ 69:13:12“ Cluster A (14 88 clones)
1 —1 NIRSS4A 605

 

1
1 98 NIRS4A 608
1 ! 97 LNIRS4B 008
"—1N_1F}S1A F09
1 NIRSS4A 809
L NIR 4AF02
__63 82 L1 NIRS4A 804
551—" NIRS4B (:09

_I_" NIRS4B C11
L‘ﬁ Baltic Sea cyanob. aggregate cl. BS1270 (AJ457196)

1 ' NIRS4A H10
1 1 1 61_1--N1RS48 808
1 __ 77

J

_ﬂ

NIRS4B F04
NIRS4A F10
1 1 l 1 J ——- -' NIRS48A10
1 *~N1RS48 010
11 1—— NIRS4B (:02
---- NIR
'- Cluster BE (6 SB clones)
EHNIRS2807
: V 2_,, NIRS48 (:10
4-- J EHNtRS1605
571 93- N1RS18 A12
1 ,3 '— Puget Sound sed. ct. p846 (AJ248422)
11 ‘ -2 EHNIRS1GOZ
~— NIRS4A 612
EHNrRszHo7
1 88 N1RS1A 610
1 Puget Sound sed. cl. p849 (AJ248423)
_59- 1 - NIRSZA Hoe
11 Puget Sound sed. cl. p866 (AJ248424)
t 11—- N1R618 EDS
1 L N1RStB E12
Cluster C (5 BC and 5 EHS clones)
BalticSS1eam1 water col (:1. NBO- 76 (00072204)
--~~~—— NlR
11f N IR Cluster D (1 BC and 5 EHS clones)
99
$111 : C'UStgbe E (8 BC clones)

anglers/1F (80 BC and 2 EH8 clones)
—“-*—- NIRS1B (:05
52 N1RS18 E03
96 NIRS18 E10
'7 Cluster 6 (3 SB clones)
NIRS1A C11
001J1Puget Sound sed. cl. pA5 (AJ248403)
, N1RS18 608
NIRS1A Hoz
‘NIRS1A HOG
NtRS1A F01
55 NIRS18 511
6 F NIRStA 004
1 N1RS18 F06
£911N1RS18 604
! L N1RS18 CO1
LN1R518A02
_g' 58 NtRS18 612

 

 

 

M

 

51;; EHNIR81F10

. 511 '- "L’NIRSZADOZ

1 1 ._.r—N1RS18609
1

1‘ .1‘— —'““‘ NIRS1A (307
1

‘ - N1R838 602
7" ~~ Battic Sea water col. (:1 MSG-165 (00072223)
r NIRSZA C11
92 100 NIRSZB CO71
1 EHNIRszcto
11'th, NIRS1AA12
11605
L Cluster 61(36 BC and 10 EHS clones)

 

 

1—: EHN1R32811
- v» NIRS3AA06
96 ‘ "~— NIRS3AE12
51871— EHNIRS1DOB
5‘1 EHNIRSZEOZ
~ "‘NIRS1AAO3
— EHN1R51601

L1____1 :‘ ~Hunungton Beach sed ct th 8F (00159664)

95 ——--- -—~-- ~ Cluster H (5 WM clones)

1_—._4
0.05

166

8-10

I 10-1 2
ul 12-14
14-16
16-18
18-20

EHS clones
0-2

0 2-4

0 8-10

20

BC CIOHOS

WJF clones

Depth in sediments (cm)

04).!)

s :1:
g -
1040.5
3647

62

SB clones

 

(1 ( \ I < . I-l'l-l-III
7 .7'.\f.' 3.3 -'.'. .'.'.'.

“(41111“

L‘Iﬂl‘l‘tl

 

 

 

 

 

 

 

 

ttttttttttt
1).), ........

20% 40% 60% 80%
Relative abundance of T-RFs

111. ‘0 511(505) 540
|
P

'87 58 0 64(70)
(:1 83 a 86

m 91 n 97
‘0102(108)a104(109)
D108(112)C1111
:3 129(133) I 141
I 158 121 161
I 173 198
121208 n210(212)
(a 230 a 232(234)
a 234(237) n 236(239)
a: 238(241) I 257
:1 270 1:1 272(274)
n 275(277)13 281
1a 295 :a 297(299)
1I 299(302) 303(304)
I 317 (a 328
I 332 335(337),
I 348(350) a 353(356)
a 356 n 359(362)
a 380(382) m 383(385)
131 388(390) a 391(393)‘
n 433(435) 1:2 436(437)

 

 

 

E] 546 I 589(587)
1:1 595(590)

100%

Figure 4.17. Comparison of T-RFLP profiles obtained by ampliﬁcation of nirS genes
from Arctic (EHS, East Hanna Shoal Shallow; BC, Barrow Canyon Shallow) and
Pacific Northwest (WJF, West of Juan de Fuca; WM, Washington Margin; SB,
Shallow Budd Inlet) sediments from various depths to distribution of T-RF sizes
obtained by in silico digestion of nirS clones retrieved from those same sites.

Presented T-RFs (real and hypothetical) were generated by restriction with HhaI.
Numbers in legend indicate real T-RF lengths in base pairs for fragments representing
more than 2% relative abundance or correlated to a nirS clone. Values in brackets
indicate corresponding T-RF size generated by in silico digestion of clones.

167

Table 4.3. Cluster specific and P. stutzeri nirS gene abundance in Pacific Northwest
and Arctic sediments as measured by real-time PCR.

 

 

 

Sitea Average nirS gene copy number per 11g community DNAB
Cluster 1 Cluster 2 Cluster 3 Cluster 5 Cluster 6 P.
stutzeri
WJF BDL BDL BDL 1.3x 103 BDL 6.9x 104
(6.2><103)
BC BDL BDL BDL 4.5><103 BDL 2.7x 104
(2.7x103) (2.1x103)
BCD BDL BDL BDL 8.7x103 4.1><102 7.3><104
(5.1x102) (1.7) (1.7x10“)
EHS 1.6><105 BDL BDL 8.0x103 80L 9.0x10“
(2.0x10“) (3.3x10‘) (3.1x104)
EHS BDL BDL BDL 2.1><103 7.0><102 2.3x10“
D (2.0x103) (7.9x103)

 

aWJ F, West of Juan de Fuca; BC, Barrow Canyon Shallow; BCD, Barrow Canyon Deep; EHS, East Hanna
Shoal Shallow; EHSD, East Hanna Shoal Deep.
bValues in parenthesis are standard deviations of duplicate assays. Single assays were performed
when no standard deviation is indicated. BDL, below detection limit.

168

Table 4.4. Cluster specific and P. stutzeri nirS gene abundance in Pacific Northwest
sediments as measured by real-time PCR and determination of significant
differences by ANOVA

 

Sitea Average nirS gene copy number per 1.1g community DNAb
Cluster 1 Cluster 2 Cluster 3 Cluster 5 Cluster 6 . P.
stutzeri
SB 1.4x105 2.0x10" ”"3 5.3><10SM 5.2x104M 1.1><10SM 8.4x10T
“1"," (1.2x10") (3.3x105) (1.1x10‘) (3.1x10“) 9"
(3.9x10“) (4.4x103)
TB 1.6x105 81" 1.1><10""LB 4.3><105 “4",” 3.5><10‘”‘*A 5.3><104M 5.2><104
(9.8x 104) (7.7x105) (3.1x105) (4.5x104) (2.8x10“) “A

(1.4x104)

CI 2.7><10""’A NI 1.4><10‘””A 3.4x103M 2.1><103°’A 7.8x104
(1.4x104) (4.5x103) (1.6x103) (4.5x102) “’3

(4.7x10‘1

WM N1 N1 N1 N1 N1 1.1><106

(5.3x105)

 

aSB, Shallow Budd Inlet; TB, Turning Basin; CI, Carr Inlet; WM, Washington Margin.
bQuadruplicate samples were analyzed. Different lower case superscript letters (a,b,c) indicate
signiﬁcant differences between sampling locations as evaluated by one-way ANOVA (Tukey's
test) at the 0.05 level. Different capital superscript letters (A,B) indicate signiﬁcant differences
between cluster abundances as evaluated by one-way ANOVA (Tukey's test) at the 0.05 level;
Values in parenthesis are standard deviations. NI, not included in analysis.

169

Figure 4.18. Phylogenetic tree showing the afﬁliation of 168 rRN A clone sequences
from the Gamma Proteabacteria retrieved from Shallow Budd Inlet (SB), _
Washington Margin (WM), West of Juan de Fuca (WJF), Barrow Canyon Shallow
(BC), and East Hanna Shoal Shallow (EHS) sediments to selected reference
sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
Thermus thermophilus and Thermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 16$ 2 (Barrow Canyon Shallow), 16$ 3 (West of Juan de Fuca),
168 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone; Pac.,
Paciﬁc; Atl., Atlantic; hydroth., hydrothermal.

170

L EH1681 D08

5411 East Pac. Rise basalt glass cl. 9NBGBact 23 (00070798)
_ .. 168 38 E02

1—1 166 3AH09

1 66 3A E09
154. EH1682 A02
1 ~ Arctic sed cl SvaO115(AJ240974)
1 . 1661BF04
53— 166 418E 1
1, 1 EH16S1A05
‘oafLLL EH 1681 A07

EH16

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

160 S1 COS
11 __..________ 16S1ACOS
. ,:LLLLLL— 168 4C A10
LLJa taGnSTian%h cold- -seep sed cl. JT8255 (ABO15254)
1 EH16S1 E04
.4 Japan Trench deep~sea sed. cl NB1-h (ABO13829)
1! 16$ 3AE11
- _r-L L EH1681 H06
_1 1 ‘LL-- . LLLLLLLLL 16848811
1“ 1 LLLL16S4B H08
, _4- L L ___ EH1681 807
L—L— 16S 18 H08
i 168 38 805
‘ L __ 168 38 003
,. _.2..-V 991L168 38H10
V ‘9: Mid-115689., Ridge2 hydroth vent sed. cl. AT-62- 59 (AY225636)
._1 1 __ i ’168 3A007
1 78I —1 168 3A E
1 254.... ‘ "”128 noiSS 3””
i 1 L-LLLL- L-L . 16838001
L1—1 16$ 2AC11
1 L—LLL 168 38 E04
1 971*“ 99F EH1GS1 E09
L1 ,. Pac. whale carcass microbialmatcl131760(AY922232)
_ 1 __ EH1681 003
811 1"“LL‘ EH1682 F12
89L - -. 1EH16$1 (306
- 65 LLLLL1682A 009
! 1 EH16SZ 606
T L LP ; LLLLL1681AG12
1 "" ‘ "___1' L 168 40 C07
. 95 6,12, 166 48 002
1 ‘. __ Wadden Sea bulk water cl GWS- K5-2 (AY515458)
1 1 ~— -— —- .._ Beggiahtoa alba (L409 94)
1 1 L hioploca ingrica (L40998
. _ ._ ,, -. “.99 LLL— Pseudomonas stutzeri (U26 62g
1 L LPseudomonas aeruginosa(XO 684)
| .__ -2..,W_EH1631302
LL—, 1 Pseudoalteromonas denitriﬁcans X82138)
’ ”“L ““ ‘EL_____1LLL LL Psychromonas profunda (AJ4167 6)
981._____r—LL _E1H16S F10
L _ Psychromonas antarctica (Y14697)
__- i Leucothrix muoor (X87277)
A 1 “L L LL __ __ __ LOceanospi'n'I/um mans (A8006771)
LL LLL _ _ LL _ LL“ Alcanivorax borkumensrs(Y1257 9)
F1 .1—~~ _ ~77 - — EH1662 011
' Macrobu/bifer maritimus (AY 377986)
1 63.1 168 488
157611 LL168483F02
991 1 168 4CF01
, L Gua mas;3 Basin hydroth. vent sed. cl. C1 8038 (AF420367)
.~ 571 168 48360
i. ___- _2 V0 LL— 16S1AH10
r P» 168 4 F0 08
1 1___.__, - ”tangy/068mm? fibrata (AF177296)
1 égh 1 .1 L LL16$1B G02
1 777 ~64 163 28 008
1 2.. L VjééVﬁ- L 168 28 E02
1 68 _--L L 16828010
1 166 48 F09
1 168 38 A12
: 99L LLLL LLL1683BG1O
m: - 781L—LLLL—LL 16S 18 806
1 __1 'LLLL'LL LLLLL 16 G06
1‘_—1-_.__._-_ LLL LL LLL LL “L “168 2AE07

1 EH16S1 602
_ 9V ILArctic sed d. Sva0091 (AJ240987)
66 28 A08
1— EH11631 611
86 EH16S1 CO1
9’26 591 marine h Srocarbon seep sed cl SBseep4(AY456982)

1A 608
3° 791———1816618A02

_J|

 

 

1 LL LThermus thermophilus (1X07998)
99L L LThermus aquaticus( 09663

 

171

Figure 4.19. Phylogenetic tree showing the afﬁliation of 168 rRNA clone sequences
from the Delta Proteabacteria retrieved from Shallow Budd Inlet (SB), Washington
Margin (WM), West of Juan de F uca (WJF), Barrow Canyon Shallow (BC), and
East Hanna Shoal Shallow (EHS) sediments to selected reference sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
Thermus thermophilus and T hermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 168 2 (Barrow Canyon Shallow), 168 3 (West of Juan de Fuca),
168 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone.

172

 

L 99 ”—641 1663A 805

 

9994f 16S 48 F10
‘Wadden Sea bulk water Cl GWS- Kdna24 (AY515480)
1 168 4861 10
| 1_ _ 1168 4C E05
1 98" 1L 168 4C 803
l 54 163 4C E11
1 , "v168 48 608
1 LLL ' " L 168 28 C03
581 89 _1———16328G06

"" i 1—L— EH1682 E12
1 95 7., Arctic sedcl dSva1036 6110240990)
67 L LLDesquorhopalus vacuolatus L42
1 581________Aérc1166§eé<188156va10 13 (AJ 40982)
: 99 L-
67: 96'th EH1662 F09 ,
1 L—L—LL Desulfota/eagsychroghlla (AF099062)
F" “LLLL S 2AE 2

99

,1 ' -~L EH1BS1F01
74 1 9:3,1EH1681 C04

99‘ , L EH1682 C10
4 Arcticsed d Sva1041 (AJ240984
1Wadden Sea sed d SK4 (AY771 54)
9794 EH1682 E02

EH1682 608
991—7 168 1B 807
7 ~591 Caswdia Margin hydrate ridge sed. cl. H d89—23 (AJ535245)
__ 1 Desulfacinum hydrotherma/e (AF170 17)
" 1LL L LL LLLLLLL 16S 18 E08
1% ¥ 1 1m _ 64 168L18086ulfonema magnum (U45989)
99 IL—
_ : Wadden Sea sed cl SK6(AY771941)
9°! 3111682 603

 

 

 

797

 

 

 

 

 

 

93

1 i
51
1 1 168 48 E06
1 6891011 EH1662 801
EH1682 C08
LL L LLLL LLL LLLL L Desulfonatronovibno hydrogenovorans (X99234)
‘ '1 1‘ " 'LLLLLLL Malonomonas mbra (Y1 17 2
'LL—LLLLL Desu/furomongsgoa/mltatls (U28172)

1
991 --.-—r l 816328
[ 991 - WgSsedB d ESvaO556 (AJ241010)
1
1
1
l
1

 

L 168 3A C05
991 L. LL16S 3A H05
ﬁvc—W—u _. v~1__ L LL Juan de Fuca Rid e basalt lass cl JdFBGBact 10 (00070818)

 

LL Nitrosprna 6gracrH rs O(2L355)

 

 

_._ﬁ _,.

 

.1 a, __ LLLL 1633A 809
1 1. 1 99 LJamgikergggcoid- seep sed cl JT836(A8015242)

,,,_ 95 rL—L 16
‘1 99: EH1JESPan05 Trench deep-sea sed d N81 -1 (A8013831)

 

 

 

 

L L L .- ‘ L _ LL L‘ __ “ LLL ”L 168 AOE3
1 E2601 Pelyangium cellulosum (00256395)
1 1 ‘LL L L L LL L' ﬁannxyais exegearbs (AJ233946)

 

 

-- - 51:1— 16S1AG11
1 7 1662Ao11

1 ,_ , , , L__EH1166§2AH 00%

,, c. -, 1 M 1—A1rct1csec1 c1 Sva1009 (AJ297456)
1 1681 BG05

1 _EH1661 E151

 

 

 

‘16 9rEH 16 682 A07
LArctic sed. cl. Sva0537 (AJ297473)
_._ L LL LL 16S 3A6F05

 

 

- H g LArctic sed1mentc|oneSva0506 (AJ241006)
9" 1— ___ L - EH1662 010

 

Thermus thermophilus (X07998
99 LL— Thermus aquaticus (L0966 )

 

005

173

Figure 4.20. Phylogenetic tree showing the afﬁliation of 168 rRNA clone sequences -
from the Alpha, Beta, and Epsilon Proteabacteria retrieved from Shallow Budd Inlet
(SB), Washington Margin (WM), West of Juan de F uca (WJF), Barrow Canyon
Shallow (BC), and East Hanna Shoal Shallow (EHS) sediments to selected reference
sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
T hermus thermophilus and Thermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 16$ 1
(Washington Margin), 16$ 2 (Barrow Canyon Shallow), 16$ 3 (West of Juan de Fuca),
168 4 (Shallow Budd Inlet), and EH 163 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; 0]., clone; isol.,
isolate; hydroth., hydrothermal.

174

 

i J
1

r

1

519H,:LCalifomia marine sed. d. (AY193227)
LoktaneI/a msea (AY682199)
ﬂLArctic sea ice isol. ARK10226 (AF468375)
163 4C H02
641.1L Sulﬁtobacter mediterraneus (Y17387)
“sf—LL Raseabacter denrtnf cans (L01784)
L16S 40 E09
:168 48 D04
1___163 18 B03
L—LParaooccus denitriﬁcans (Y1 692 8)
1______1633A 3A802

Magnetospiril/um magnetotacticum (Y101 10)
F 1 85L— 168 1A 805

761 1 — 168 4C E03
__ Methylosinus sporium (Y18946)
LLL Methylocystis parvus (M29026)
.521 LE ,__J—165 1A C09
3LLL— 163 3A E10
‘ _rL LL- 1681B E02
1_1 __ EH1681 F09
—95—. 1,. - L BgréogglEﬁ/izabethae (L01260)
1 _ _ _ L L L 1
l 90;_____ .. L163 38 F05
moL Juan de Fuca Ridge basalt glass cl. JdFBGBact 41 (00070828)

168 2A F12
-EL 168 18 COB
__ E Spin'llum volutans (M34131)
‘, L—LLL LLLLL Alca/igenes faecalis (M2 2508)
LLLT ILL—LL Cupn'avr'dus necator(M32021)
L {31 Thauera aromatica (X77118)
1 _ J L w 168 38 D10
‘1001 Nitrosospira multifarmis (L35 509)
Nitrosospira bn'ensis (AY123800)
168 28 804
F109 L Mi.d-Atl Ridge hydroth. vent sed cl le3BBOS(AY354174)
11 East Pac Rise basalt glass 0|. 9NBGBact 3 (00070790)
( , 191 ___ LLLLL— Wolinella succinogenes (M88159)
1

L1; LLLLL 16$ 2ACO4
,- 64 ___ Thiomicrospira denitrificans (L40808)
LL LLLLL Aroobacter cryaerophilus (L14624)
1 _1L LLLLLLLL Sulfurospirillum arcaohonense (Y11561)
1. _ Sulfurospirillum bamesii (AF038843)
861- 16S 48 808
55;“ '166 4C 605
168 4C 811
60' . South Korea tidal sed. d. (AY304364)
, 9811 166 4c 801
1 L 166 4C 005

' ' _ 168 4C (309
99 70 1682AH10
L“ 168 28 G02
[—1168 28 H10

Guaymas Basin hydroth. vent sed cl a1b001 (AF420344)

 

 

 

 

 

 

 

1V6 16S 28 G05
1 51_m 166 48 H03
L _99;— 166 2A H06

1 64131.1" '168 213 H06
I

L 168 48 001
1 Ldeep-sea sed. d 807- 9 (A8015584)
168 1A 001
168 2A E03

95 L168 28A05
16S 48 010

 

__2, ILL Thermus thennophilus (X07998)
100 1.___ Thermus aquaticus (L09663)

._J

175

Figure 4.21. Phylogenetic tree showing the afﬁliation of 168 rRNA clone sequences
from the Bacteroidetes, Chlarabi, Deferribacteres, Spirochaetes, and Acidobacteria
retrieved from Shallow Budd Inlet (SB), Washington Margin (WM), West of Juan
de Fuca (WJF), Barrow Canyon Shallow (BC), and East Hanna Shoal Shallow .
(EHS) sediments to selected reference sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
Thermus thermophilus and T hermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 168 2 (Barrow Canyon Shallow), 16$ 3 (West of Juan de Fuca),
16$ 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone; Atl.,
Atlantic; hydroth., hydrothermal; Massach., Massachussetts; benzene-degr. nitrate-red.
cons., benzene-degrading nitrate-reducing consortium.

176

 

 

 

 

 

_. 100

100.; saltrriarshssedB.E509 cl. LCP- 26 (AF286031)

 

_ _ 1_Caldrthri'x abyssi BéAJ430587)
86751—1665 328 8A02
. 16$ 16 604

 

99861 ———---L16326A04

1L- ~ -— L163 26 001

 

 

168 40 D11
76 f— ‘99 Japan Trench cold- -seep sed. cl JT 8250 (A8015264)
-— —1 1618 H09

HEH1681 D02
___1_09r CaEsEﬁdeia Mar%in hydrate ridge sed. d. Hyd89—65 (N535255)

511 'L——— 16828 509
-—L~--~- 16318812
L100 LLL - — L1632AH03

L_____Cﬂ1;topha a fermentans (M 5876 62
7 _J __ - __ 168 riniBiaF i/ia saimonioo or (M62 22)

,, V1 ”h, ___1 " “ “ 163 4C F11
100 "'— 163 2A F11
" _163 2AA08
, 163 3A H07
1 __ 1002163 2A 306

‘11 ‘ n _‘LL1682AA07
1 V___

MId-Atl. 2ARidge hydroth. vent sed. cl. pIR38H02 (AY354148)

1001 _ JapanS Trench deep—sea sed. cl. NB1-m (A8013834)
100; 5C 168 38 804
168 28 DOS
95:,A 168 18 G10

 

 

 

_ [ -‘ "' ‘-‘—"

 

 

Arctic sea ice isolate 118 (AF542202)
168 2A F 03
Flavobacterium tripods AJ557887)
Flavpbacfgrsium agua rle ( 6279)

L V1 168 4C D09
Massach. estuarine sed. cl. C319a- RBC- E2 (AY678530)

10011 ,. 163 ZAP
L L L L L1608 28 H11

168 4B 810
LLLL168 4C DO3
__ LL 16S 48 C01

Tenacibaculum maritimum (M64629)

P Sf:hiLBoscE)rpens bartonensis (U62913)

  
 

 

 

 

 

 

 

 

7956
Air L EH16S1 F02

006616
1_' __ 168 4C H05

 

L' __ 2
..-._ __.. "_‘—— ‘L‘ ‘ Chlorobiumlimioola(Y10640)

,2 , L168 28 E07

100 L North Sea sed. cl. Belgim2005/10-120-13 (DQ351751)
L .. LLL L 168 3A 606

L i f“.

100 .mw,

16S 18 001 1 . .
"L- -__ LLLLL Sgirochaeta ba/am/rfornrensr‘s (N698859)
L L H1681 GO4

 

Spiroch aeta halophi/a (M88722)
48 809
East Mediterranean mud volcano cl MilanoWF1 8- 06 (AY592849)
168 2A 609
L EH1681 H03
1681 18 E07

..______. 68 3AC03
'61EH186?:1) 51%8

 

 

Acrdobacterr’um capsu/atum (026171)
168 3A E0
L L 1681 AF05

 

‘ ‘ 1 16318 H01
1 L -L L LL- 16838005

005

99?

,1001 LL 168 1A H05
‘L __hlan_ka_i T1rgtégh8d§egsea sed cl NK818 (A8013270)

“L ‘“‘ 16$ 38 G12

,- r163 3A (308
East Mediterranean Sea sed cl. S lonian- F04 (AY533909)

Thermus theWnophrlus (x0799 98

~—- Thermus aquaticus (L09 63)

177

Figure 4.22. Phylogenetic tree showing the afﬁliation of 16S rRNA clone sequences
from the Chloroﬂaxi, Actinobacteria, Nitraspira, ODl group and OH] group
retrieved from Shallow Budd Inlet (SB), Washington Margin (WM), West of Juan
de Fuca (WJF), Barrow Canyon Shallow (BC), and East Hanna Shoal Shallow
(EHS) sediments to selected reference sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
T hermus thermophilus and T hermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 168 2 (Barrow Canyon Shallow), 168 3 (West of Juan de Fuca),
16$ 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone; Pac.,
Paciﬁc; Atl., Atlantic; hydroth., hydrothermal; Mich, Michigan; aq., aquatic; microb.,
microbial.

178

 

 

 

 

_, ' LL EH16S1 DO4
L - LL marinesed cl BMS43 (AY193161)
1681 18005
_ LLLLL EH1681 806
L 168 38 806

168 18 F07
_ LLLL—L Mich aquifersed. d. WCHB1- 26 AF050599)
___ ___- Mich. aquifer sed cl. WCH81- 6(AF050605)
168 18A06
168 1B F10
Yellowstone hot spring cl. 0P892 (AF027030)

- ~ --—- - LL-L1683AG12
__ .L 1682A008
--L16818 D12
86 ,L ,W- 1 - Mid- Pac. deep-sea sed cl. M8AE49(AJ567599)
.L—— 168 38 H09
70 L 168 3A F09
LEH1682 803

1 1 lL1681AEO7

6, EH1681 808

Mid- Atl Ridge hydroth ventsed d lR3BDOS AY354150)
99‘ _ . Guaymas Basin hydroth vent sed a2b024 F419678)

1 r hydrocarbonse 8PC063(AF1540 3)

1g dee -sea sed cl 8 2 e1% A(8015539)

97 71—LJE 1681F04

55‘ 1681AC02
168 38 C10

 

 

 

 

 

 

 

 

 

L L 168 3A H01
' _ LLL-16S4CHO4
7 iL—LLLL—LLL 168 3AA097
“‘ wig—.4 L LL168188
100.. L LL LL Mid- Atl9 Ridge hydroth. ventsed d AT-sZ- 33 (AY225655)
EH1681 1605
Cand Microrhnxéiarwoella (X82546)

__,_ Ac/djmicrobium 1ferrooxidans (U75647)
, _LLL1681ADO3
- 199: ﬂtralia N1u6llgrggrgiaes aq microb formation cl. wb1 _P06 (AF 317769)

___ 1L Mid- -81? Rid%%2 hydroth vent sed. d le3BCOB (AY354165)
1L

 

 

 

 

100' L 556—16816F09

168 4C F 12
1 065,12— 1168 A10 2
1 10° 90... ”Neogthsw Sea04 sed cl. Belgica2005/10- ZG 15 (DQ351808)
Nitraspira04 man'na (X82559)
LLLL L L LLLLLLL Nitrospira mOSOOVIGNSlS (X82558)

l
j FL- L- L16S1AF02
._ , . 100L 165 3AH03

T , LL East Mediterranean Sea sedol as Ionian A11 (AY534050)
rL _ LL LLL 16 $8 agga/ocoocoides eihenogenes (AF 004928)
100 LLLPCB dtachlorinating enrichment culture d OTU-1O (AY559073)
3__1L 168 48 D1 12
~—-—; - ~* “’0' 168 46 (309
168 4C A12
_JﬂL 168 48 DO7
1_—*-- 168 48 H06
j 168 aCEugymas Basin hydroth. vent sed d. C1 8046(AF419699)
94f
1 98—1 L Southggrgga tidal sed d 881 -()-115 (AY304372)

 

1 4
LA LL16S4BAO4

. , LLL LLLL LLL L 168 4B COS

93: L--LLL---16S 4CA06
: 95731- 16818 003

10096 L 168 18 E04

1 . LGuaymas Basinh droth vent sed d. 804RO15 (AY197416)

1_4 571—— L—-1__'hy1droc 4seFep0 sedd PC110(AF154084)

16$2ADO3

 

- --—~-L: 165 1A F12

 

 

~ - 16S 48 COB
60; a ,- ~~— . ‘09. EH1681 F06
T {LLLLLLL16 60184CC
L East Mediterranean mud volcano cl Milano-WF1B-48 (AY592889)
1 ___- _100 168 4C A02
1 _f L East Mediterranean mud volcano cl. Amsterdam -28- 06 (AY592366)
;f_ _ L168 48 804
116$ 18 E03
00 V1 168 18 D06
168 18 E12
22,2 LThermus thermophilus (X07 998
100L Thermus aquaticus (L0966 )

179

1991-2 168 3A F12
9,9; 1. LL L- ----1--——- EH1681A04

 

1 ;______.___ 168 18 805
1 1 , 1 ' LLL L‘ _- LLL Propionigenium man's (X84049)
1 601-"LL —-- L 16848607
;ﬂ__74 _._ LL—LLLL L 168 3A 303
. 1 EH1682 004
1 881 L L LL Bacillus Iitora/is (AY608605)

i I __ __ 'LLL Clostn'dium bowmanii(AJ506120)
1 A/lisonel/a histaminifonnans (A F548373)

 

 

1 ! 1681A 301
i 1 - EH1681HO4
991 deep-sea cold seep cl. C8521 (A8069798)
L 168 2A (301

951-- 16838 E09

1 . -- 100 1166 3A C10

1 1 Mid-Atl. Ridge hydroth. vent sed. cl. AT-sso (AY225657)
168 48 A06

 

 

 

1 71LLLLLL L __ ‘L Fibrobacter suocinogenes (M62696)
621 L- L- 168 2A 005
1 L Thermus thermophi/us (X07998)

 

 

Thermus aqua ticus (L09663)

Figure 4.23. Phylogenetic tree showing the afﬁliation of 168 rRNA clone sequences
from the F irmicutes, F usobacteria, and F ibrobacteres retrieved from Shallow Budd
Inlet (SB), Washington Margin (WM), West of Juan de Fuca (WJF), Barrow
Canyon Shallow (BC), and East Hanna Shoal Shallow (EHS) sediments to selected
reference sequences.

The tree was generated by neighbor-joining analysis of approximately 550 hp with
T hermus thermophilus and T hermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 16$ 1
(Washington Margin), 16$ 2 (Barrow Canyon Shallow), 16$ 3 (West of Juan de Fuca),
16$ 4 (Shallow Budd Inlet), and EH 16S (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone; Atl.,
Atlantic; hydroth., hydrothermal.

180

Figure 4.24. Phylogenetic tree showing the afﬁliation of 168 rRNA clone sequences
from the Planctomycetes, Verrucamicrabia and WS3 group retrieved from Shallow
Budd Inlet (SB), Washington Margin (WM), West of Juan de Fuca (WJF), Barrow
Canyon Shallow (BC), and East Hanna Shoal Shallow (EHS) sediments to selected
reference sequences. ‘

The tree was generated by neighbor-joining analysis of approximately 550 hp with
Thermus thermophilus and Thermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 16$ 2 (Barrow Canyon Shallow), 168 3 (West of Juan de Fuca),
168 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The scale bar
represents 0.05 substitutions per sequence position. sed., sediment; cl., clone; hydroth.,
hydrothermal; G. of Mex., Gulf of Mexico.

181

00413.1 EH1681 C07
[I ———ﬁ , ool ___ EH1682 E06
"I Greenland lkka Fjord ikaite column cl. ikaite un-ch (N431347)
_- ‘6‘){163 BB 012
:——_— “’5 ”533353
H1 1
LJ WWW168 3B H04
62L? W 16S1AA07
‘ W Arctic sed. ct. Sva0800 (AJ297459)
J""_——‘ --1681A(302
WW 168 28 DO7
I :WrW EH16S1 A02
L _ . W16S1AG07
‘W Black Sea water column d. JK471 (00368274)
" , 168 38 801
99 ___ Blastopinellula man'na (X62912)
l WW' ' “ "WWW Pirellula staleyi (AJ231183)
‘ 168 1B 801

I 100I __7_ r— EH1682A03
100 L G. of Mex. sed. cl. GoM GB425 028—7 (AY542549)
l I Gemmata obscuriglobus (AJ231191)
73 _f' , ’ 168 3AD10

«I 66 .. ‘95 Le? ——-—-—— Planctomyces brasiliensis (AJ231190)

 

 

   
  

 

 

 

 

85 ‘ “ ' Planctomyces man's(AJ231184)
_100; ' W EH1681 GlO
"WW ‘ —WW 168 38 C08

‘ g," 168 3B F02
J 'WWQB ,_ I _II 16828 F11

100 168 4BE12
"“"M ‘ W "““1683ABO4

[ 75‘W ,, __ --WW16S4C F06
W "W" ‘W " ‘ ‘ 168 BABOG
76' -- ~——A~ EH16S1E06
mi . ——~' - 168 1A 808

J i-___ ____ 16S 28 006
88?: 100 Scotland anoxic marine sed. cl. LD1-PB2 (AY114333)
L—«ﬁ‘ 168 1A 803
I V I :“WWW— 168 2A 002
{g 100 ---~w~—H ~--——~ 16$ 28 (310
I
|

 

 

 

 

 

W “W —- -— 168 2A E12
" _ _Ioor- ~-— 168 38 I303
' [___ #91 ~ --— North Sea sed cl. BeIgica2005/10-130 26 (DQ351768)

 

 

 

r Opitutus terrae (AJ229235)
, 62 .W 163 2A E09
; * r ‘ L ----- W" W Verrucamicrabium spinosum (X90515)
f 95.’ ,.__1QQ.I” 1631B A12
I l: ‘- Barents Sea mud volcano cl. HMMVPog-57 (AJ704715)

 

 

*W- 168 28 608

~ W 1 -- - "WW EH16$2 H05

I 168 1A C01

50: ,-
{A _ II— 168 2A 804
100 t _ EH1682 B10
, f ’ inactive deep- sea hydroth ventdwimney cl lndB1- 5 (A8099995)
as EH1682 H01
99[ _; EH1681 DO5
94W EH16S1 F08
,5; EH1682 CO7
L 168 40 B10

1 [_.ﬂi’ , r EH1681 E01
I I ' L ‘ "W WW WScotland anoxic marine sed. cl. LD1-PA13 (AY114311)
: ! I “'W ‘WWWW‘W" W soil clone PBS-Ill-9 (AJ390450)
‘ 771 r— ' soil cione PBS-ll-5 (N390441)
J #4 I 1QrW EH1682 A09
I 80L? L— 16828 CO?
+ 90! __-’—1BS4CG12
‘ 100 W W East Mediterranean Sea sed. d. Therm01—68 (AY534052)
_ " WW' .,_ “ Scotland anoxic marine sed. cl. LD1-PB19 (AY114332)
ml ___ Thermus thermophilus (X07998)
100 " * Thermus aquaticus (L09663)
I-— ~-————i

005

 

 

 

 

182

70—J 1682A F10
57? W 16828 F08
66 ‘163 2A C06
51' [_[r— 1682AGl1
1386 “W 16828 802
l’163 28 (307
168 28 G12
168 2AA02
W 168 28 (311
168 2A H02
'16SZADO4
55 i EH1682 008
168 2A F04

[
1-9—3W 16S 28 012

{LBJ ’ EH1681 A11
.W 168 4C (304
W- Salt marsh cl. SIMO—823 (AY712360)
97‘ I", T“ 168 2A GOB
l 73: Shepherds Creek sub-watershed d. 182up (AY212634)
-- WW 168 28 H07
W~ - --W- W EH16S1 F03

I
i ' I ' —_ Dermocarpe/Ia incrassata (AJ344559)
l

l ‘ sat-W W- Anabaena cylindrica(AF091150)

 

 

__87

 

Pro chlorococcus marin us (AF180967)
__ _ _} 7hermusthennophilus (X07998)
100‘ ___ Thermus aquaﬁcus(L09663)

 

Figure 4.25. Phylogenetic tree showing the affiliation of 168 rRNA clone sequences
from the Cyanobacteria retrieved from Shallow Budd Inlet (SB), Washington-
Margin (WM), West of Juan de Fuca (WJF), Barrow Canyon Shallow (BC), and
East Hanna Shoal Shallow (EHS) sediments to selected reference sequences.

The tree was generated by neighbor-joining analysis of approximately 550 bp with
Thermus thermophilus and T hermus aquaticus as outgroup. Values at branch points
indicate the percentage of 100 replicate trees supporting that branch. Bootstrap values
below 50% were omitted. Names of clones retrieved in this study begin with 168 1
(Washington Margin), 16$ 2 (Barrow Canyon Shallow), 16$ 3 (West of Juan de Fuca),
168 4 (Shallow Budd Inlet), and EH 168 (East Hanna Shoal Shallow). The seale bar
represents 0.05 substitutions per sequence position. cl., clone.

183

Table 4.5. Assignment of 16S rRNA clone sequences from Pacific Northwest and
Arctic sediments into taxonomic groups.

 

 

 

Taxonomic groupa Stationb
BC EHS SB WM WJF
Proteabacteria 30 42 35 30 25
Alpha Proteobacteria 1 1 4 4 3
Beta Proteabacteria 0 O O O l
Gamma Proteabacteria 9 2O 13 l 1 14
Delta Proteobacteria 10 19 10 11 ‘ 3
Epsilon Proteabacteria 2 O 5 0 0
unclassiﬁed
Proteabacteria 8 2 3 4 4
Bacteroidetes l4 4 12 4 3
F lavobacteria 4 1 7 0 O
Sphingobacteria 2 0 O 2 2
Bacteroidetes l O O O O
unclassiﬁed '
Bacteroidetes 7 3 5 2 1
Planctomycetes 1 4 0 4 5
Planctomycetacia 1 4 O 4 5
F irmicutes 1 1 0 l 0
C Iostridia 1 l O O 0
Caldithrix O 0 0 1 0
C yanobacteria 4 3 1 0 0
Cyanobacteria 4 3 1 O ' O
Verrucamicrabia 3 3 2 l 0
Verrucomicrobiae 3 3 2 1 0
Nitraspira 0 0 0 l l
Nitraspira O 0 0 1 1
Actinobacteria 0 0 1 l 2
Actinobacteria O O l 1 2
Genera_incertae_sedis_WS3 l l 1 0 0
Genera_incertae_sedis_OPll 0 0 0 0 ' l

Unclassiﬁed Bacteria 27 25 28 38 38

 

aAssignment to taxonomic groups as determined by the placement of the sequences into the RDP Hierarchy
with the RDP Classiﬁer. Bootstrapping was performed to estimate the classiﬁcation reliability. Sequences
that could not be assigned to a taxon with a bootstrap conﬁdence estimate above the 80% selected threshold
were considered “unclassiﬁed”.

bBC, Barrow Canyon Shallow; EHS, East Hanna Shoal Shallow; SB, Shallow Budd Inlet; WM,
Washington Margin; WJF, West of Juan de Fuca.

184

Table 4.6. Significant differences in the taxonomic composition between 16S rRNA
libraries from Pacific Northwest and Arctic sediments.

 

 

 

 

Libraries comparedal Cloneiaisslllgn ed to
Taxonomic group . . Signiﬁcancec
Library 1 Library 2 L‘blra’y L‘bgary
BC EHS
Bacteroidetes 14 4 0.017
Gamma Proteabacteria 9 20 0.029
BC SB
Sulfurospirillum 0 5 0.030
BC WM
Bacteroidetes 14 4 0.02 1
BC WJ F
Bacteroidetes 14 3 0.012
EHS SB
Epsilon Proteabacteria 0 5 0.028
Campylobacterales 0 5 0.028
C ampylobacteraceae 0 5 0.028
Sulfurospirillum 0 5 0.028
Bacteroidetes 4 1 2 0.04 l
EHS WM
Desulfobulbaceae 8 0 0.005
Proteabacteria 42 28 0.044
EHS W] F -
Delta Proteabacteria 20 3 0.001
Desulfobulbaceae 8 0 0.006
Desulfobacterales 1 2 2 0.01 3
Proteabacteria 42 24 0.018
SB WM
F lavobacteria 7 0 0.008
F lavobacteriales 7 0 0.008
F Iavobacteriaceae 7 0 0.008
Desulfobulbaceae 6 0 0.01 6
Desulfobacterales 1 0 2 0.023
Epsilon Proteabacteria 5 0 0.031
C ampylobacterales 5 0 0.03 l
C ampylobacteraceae 5 0 0.03 1
Sulfurospirillum 5 0 0.03 1
Bacteroidetes 1 2 4 0.049
SB WJ F
F lavobacteria 7 0 0.010

185

Table 4.6 (cont’d) Significant differences in the taxonomic composition between
16S rRNA libraries from Paciﬁc Northwest and Arctic sediments.
Clones assigned to

 

Libraries compareda

 

 

 

Taxonomic group taxon Signiﬁcancec
Library 1 Library 2 Library 1 Library 2
F lavobacteriales 7 0 0.010
F Iavobacteriaceae 7 0 0.01 0
Desulfobulbaceae 6 0 0.020
Planctomycetes 0 5 0.026
Planctomycetacia 0 5 0.026
Planctomycetales 0 5 0.026
Planctomycetaceae 0 5 0.026
Bacteroidetes 1 2 3 0.030
Desulfobacterales 1 0 2 0.030
Epsilon Proteobacteria 5 O 0.038
C ampylobacterales 5 0 0.038
Campylobacteraceae 5 0 0.038
Sulfurospirillum 5 0 0.038
WM WJF
Delta Proteobacteria 12 3 0.030

 

aBC, Barrow Canyon Shallow; EHS, East Hanna Shoal; SB, Shallow Budd Inlet; WM, Washington
Margin; WJF, West of Juan de Fuca.

bAssignment to taxonomic groups as determined by the placement of the sequences into the RDP Hierarchy
with the RDP Classiﬁer, using a threshold value of 80% for the bootstrap conﬁdence estimate.

cOnly taxonomic groups presenting signiﬁcant differences (P S 0.05) between the compared pair of clone
libraries were included.

186

 

70

168 rRNA

 

 

 

 

(11
O
|
I
1
1
1
1
|

 

.h
0
1

 

(a)
O
l
I

 

No. of OTUs

 

 

 

 

"J ‘1' “I ".' '5" '1' “I "I "I "f "f "f "f ".’ "f "I "I"! ".' 't' f‘f'f'f‘f'

 

 

 

 

 

 

7O ,. - ___ __ ____ __ . ___. . -_ ..._ ”In _.__HLMLJ

 

 

 

 

50 n ., __ ._._ __ __ .__..___.__ .___ L____-_.._2__‘_.

 

 

 

No. of OTUs

 

 

 

 

 

 

13 5 7 91113 15 1719 2123 25 27
Abundance

EBCIEHS EISB IWM IWJF

 

 

Figure 4.26. Relationship between the number of OTUs and their abundance, as
measured by the number of clones corresponding to each OTU, for 168 rRNA and
nirS clone libraries from Barrow Canyon Shallow (BC), East Hanna Shoal Shallow
(EHS), Shallow Budd Inlet (SB), Washington Margin (WM), and West of Juan de
Fuca (WJF) sediments.

187

so LL#,W,, “77 rem—em , , «— 1

70 183 rRNA W

60 . Egg;-
‘ @1111 i i
30 - ggggﬁg ‘

20

Number of OTUs

 

o g 1 ‘ 1 I 1 1
0 10 20 30 40 5O 60 70 80 90

Number of clones analyzed

 

 

WnirS

70*
60*
50‘
40-

Number of OTUs

20 WW
10 W

 

 

 

0 25 50 75 100 125 150 175
Number of clones analyzed

EBC_'-giE_I?. SB WW

 

Figure 4.27. Rarefaction curves for 16S rRNA and nirS clone libraries from Barrow
Canyon Shallow (BC), East Hanna Shoal Shallow (EHS), Shallow Budd Inlet (SB),
Washington Margin (WM), and West of Juan de Fuca (WJF) sediments.

Error bars represent 95% conﬁdence intervals.

188

Table 4.7 . Diversity and predicted richness of 16S rRNA and nirS gene fragments in
sediments from different sites based on OTU assignment of clones by DOTUR.

 

 

Gene and No. of No. of Shannon Index of Chaol richness
stationa clones OTUSb diversityc estimatec
16S rRNA
BC 81 65 3.99 (3.78; 4.20) 493 (232; 1160)
EHS 83 62 3.99 (3.81; 4.17) 215 (129; 414)
SB 80 62 4.01 (3.83; 4.19) 221 (132; 427)
WM 80 70 4.18 (4.01; 4.35) 574 (269; 1348)
WJF 75 62 4.02 (3.83; 4.21) 266 (150; 537)
nirS
BC 153 60 3.55 (3.36; 3.74) 112 (82; 183)
EHS 124 53 3.52 (3.33; 3.72) 98 (71; 164)
SB 137 42 3.08 (2.87; 3.29) 93 (59; 192)
WM 161 87 3.94 (3.74; 4.14) 348 (209; 643)
WJF 74 30 3.07 (2.85; 3.30) 41 (33; 71)

 

aBC, Barrow Canyon Shallow; EHS, East Hanna Shoal Shallow; SB, Shallow Budd Inlet; WM,

Washington Margin; WJF, West of Juan de Fuca.

bDistance threshold applied for OTU assignment was 3% and 5% for 168 rRNA and nirS, respectively.
cLower und upper 95% conﬁdence intervals are indicated in parenthesis.

Table 4.8. Determination of signiﬁcant differences between libraries by the
application of the integral form of the Cramér—von Mises statistic with 1-

LIBSHUFF.

 

Sampling site

P value for comparison of heterologous with

 

 

Gene homgldgous homologous libraryb
“bran?" BC EHS SB WM WJF
1 6S rRNA
BC 0.0036 0.0104 0.0000 0.0000
EHS 0.0000 0.0000 0.0003 0.0000
SB 0.0000 0.0000 0.0000 0.0000
WM 0.0000 0.3476 0.0012 0.0000
W] F 0.0000 0.0003 0.0000 0.0002
nirS
BC 0.0009 0.0000 0.0000 0.0000
EHS 0.01 75 0.0000 0.0000 0.0000
SB 0.0000 0.0000 0.0000 0.0000
WM 0.0000 0.0000 0.0000 0.0000
WJ F 0.0000 0.0000 0.0000 0.0000

 

aBC, Barrow Canyon Shallow; EHS, East Hanna Shoal Shallow; SB, Shallow Budd lnlet; WM,
Washington Margin; W] F, West of Juan de F uca.

bSigniﬁcant P values (P < 0.0026; (1 = 0.05) are indicated in bold. Each pair of libraries is compared twice,
by switching the library used as homologous and heterologous for the second comparison. Libraries are
signiﬁcantly different from each other if at least one of the two comparisons between them leads to a
signiﬁcant P value.

190

Table 4.9. Similarity between libraries described by the abundance-based Sorenson
similarity index (Labund) and with a nonparametric richness estimator of shared

OTUs analogous to Chaol (SM; cm) as determined with SONS.

 

 

 

 

Libraries compareda 16S rRNA
Labund SA,B Chao Labund SA,B Chao

EHS WM 0.239 45 0.494 28
EHS BC 0.194 14 0.676 47
EHS WJF 0.020 2 0 0
EHS SB 0.049 3 0.011 _ 2
WM BC 0.083 13 0.158 10
WM WJF 0 0 0.011 2
WM SB 0.044 5 0.012 2

BC WJF 0.019 2 0 0

BC SB 0 0 0 0
WJF SB 0 0 0 i 0

 

aEHS, East Hanna Shoal Shallow; BC, Barrow Canyon Shallow, WM, Washington Margin; WJF, West of
Juan de Fuca; SB, Shallow Budd Inlet.

191

Discussion

In previous studies (Chapter 3) a strong conservation of denitriﬁer community
structure was observed at different sediment depths along redox gradients within the
bioturbated layer. Clear community differences were detected between sites within one
geographic location (Puget Sound) and increased when sites from two different
geographic locations were compared (Puget Sound and Washington Margin). This pattern
was also observed in the same area by Braker et al. (2000; 2001) when various sites with
differing water depths at the Washington Margin were compared to a Puget Sound site,
suggesting that geographic location, water depth, and sediment depth were involved in
the appearance of community differences with decreasing importance in that order. To
further describe this phenomenon, the denitriﬁer, as well as the bacterial community
diversity and distribution were studied in two distant geographic areas, the Paciﬁc
Northwest and the Arctic Ocean. The former included the previously studied locations in
Puget Sound and the continental slope at the Washington Margin (Chapter 3), in addition
to an abyssal sea ﬂoor site separated from the rest by a geographic barrier. The Arctic
sites included high and low productivity locations with various water depths.

All studied sites showed a high bacterial and denitriﬁer diversity. The former was
especially high at the continental slope of the Washington Margin, based on 16S rRNA
clone libraries (Table 4.7). This site also presented, as well as the Arctic, a high
denitriﬁer diversity, based on nirS gene clone libraries and T-RFLP (Table 4.7, Figure
4.6B). The abyssal sea ﬂoor (West of Juan de Fuca) presented the highest and lowest nirS
diversity, based on T-RFLP and clone library results, respectively. This incongruence

might be due to the different resolution and coverage levels of both techniques.

192

Furthermore, although the T-RF richness at this site was lower than in other areas (Figure
4.6A), the even distribution of T-RFs probably led to an increase of the measured
diversity. Surprisingly, Puget Sound samples (Shallow Budd Inlet, Turning Basin, and
Carr Inlet) consistently showed a lower denitriﬁer diversity (Table 4.7, Figure 4.68) than
the other sites, which is opposite the results obtained by Braker et al. (2000; 2001) who
detected a higher diversity in Puget Sound compared to the Washington margin area.
Samples analyzed in the two studies were, however, not retrieved from the same sites, so
local differences might be responsible for the observed pattern. The low diversity
determined in Puget Sound samples, similar to the one observed at West of Juan de Fuca,
might be related to the fact that these sites had the highest and lowest productivity of the
sampled areas, respectively. It has been observed that the richness of certain bacterial
taxonomic groups was affected by primary productivity, leading in some cases to higher
richness at intermediate productivity levels (Homer-Devine et al., 2003), as often
observed for plants and animals (Mittelbach et al., 2001). Denitriﬁers, although not a
taxonomic group, might, as a functional group behave in a similar manner and present
higher richness and diversity at intermediate productivity levels. In addition, the same
pattern was observed also on a vertical scale, as maximum richness and diversity of T-
RF 5 were generally measured in subsurface sediments, rather than in the top layer (0-0.5
cm depth), which receives the highest amount of degradable organic matter from the
overlying water and lies within the depth at which oxygen and nitrate are detected in the
pore water. This same phenomenon was also observed in the Mediterranean Sea, where
the highest bacterial richness, as studied by T-RFLP, was detected at 1-2 and 2-3 cm deep

sediments and not at the top layer (Luna et al., 2004).

193

As presented previously in Chapter 3, site-speciﬁc grouping of the nirS T-RFLP
proﬁles was observed by cluster analysis (Figure 4.7) and PCA ordination (Figure 4.8),
except for deep sediments, indicating that sediment depth within the bioturbated zone did
not play a major role in differentiating denitriﬁer communities. Water depth along a
transect or slightly differing biogeochemical characteristics within one geographic
location had a stronger effect, while the most differentiated communities were from
different geographic areas. This is consistent with observations by Braker et al. (2001) on
nirS and 168 rRNA distributions in sediments from Washington Margin and Puget
Sound. The distribution of another denitriﬁcation gene, nosZ, showed the same pattern,
with sediment depth, meter scale and kilometer scale horizontal distances having
increasing inﬂuence on the differentiation of the community (Scala and Kerkhof, 2000).
Differences in sulfate-reducing bacterial communities were also detected. at the
Washington Margin related to water depth, which has been associated with decreasing
carbon quantity and bioavailability at increasing depth (Liu et al., 2003).

Water depth and organic carbon were also identiﬁed by CCA as main
environmental factors inﬂuencing the denitriﬁer community structure with opposite
effects, in addition to geographic position (latitude and longitude), oxygen, nitrate and
ammonium (Figure 4.9). Nitrate and oxygen have both been suggested as key factors
affecting the sediment (Liu et al., 2003) and water column (Castro-Gonzalez et al., 2005)
denitriﬁer community structure. In the Baltic Sea water column ammonium was also
identiﬁed in addition to oxygen and nitrate for its role in shaping the denitriﬁer
community (Hannig et al., 2005). Interestingly, although higher salinity has been related

with lower denitriﬁer diversity in other studies (Santoro et al., 2006), it didn’t seem to

194

have a major role in this study. Furthermore, the lowest salinity area (Puget. Sound)
presented the lowest richness and diversity.

Clone libraries from all sites, except the abyssal sea ﬂoor (West of Juan de Fuca),
suggested a dominance of Gamma and Delta Proteabacteria (Table 4.5), as has been
observed in Arctic (Ravenschlag et al., 1999) and Antarctic (Bowman and McCuaig,
2003) continental shelf sediments. The abyssal sea ﬂoor, as expected for deep sea
sediments which generally show a lower sulfate reduction rate than shallow sediments
(Hartnett and Devol, 2003), presented a high abundance of Gamma Proteabacteria, but
signiﬁcantly lower abundance of Delta Proteabacteria than other sites. The high
abundance of Delta Proteabacteria in Puget Sound and Washington Margin sediments is
consistent as well with the detection of a broad range of dissimilatory (bi)sulﬁte
reductase gene sequences (dsrAB) in the former (Tiquia et al., 2006) and latter site (Liu et
al., 2003), respectively. Furthermore, sulfate reduction was reported to be responsible for
an average of 50% of the total carbon oxidation rate in Washington shelf and upper slope
sediments (Hartnett and Devol, 2003). The widely distributed Bacteroidetes, were also
detected at all sites and were particularly abundant in Barrow Canyon Shallow and Puget
Sound sediments. This phylum had been identiﬁed as a major contributor. to the
community incorporation of tritium-labelled thymidine in Puget Sound water column
(Van Mooy et al., 2004), suggesting an important role in this coastal environment.
Bacteroidetes seem to be highly adapted to metabolize high molecular weight compounds
(Kirchman, 2002), which are probably highly abundant in the sediments of the highly
productive and close to shore Puget Sound and Barrow Canyon Shallow sites.

Planctomycetes were detected at all sites, except at Shallow Budd Inlet in Puget Sound,

195

though none of the retrieved sequences were similar to anammox-related bacteria. These
have been, however, detected with speciﬁc primers targeting this group in Shallow Budd
Inlet, Turning Basin, Washington Margin, West of Juan de Fuca, East Hanna Shoal
Shallow and Deep sediments (Penton et al., 2006). This difference is expected when
general bacterial vs. group-speciﬁc primers are used and might also be related to the low
coverage of the 16S rRNA clone libraries in this study of the total bacterial diversity
present at these sites.

The clone libraries of nirS as well as 16S rRNA genes were signiﬁcantly different
between the studied locations (Table 4.8), suggesting that the nirS-containing denitriﬁer
populations as well as the general bacterial community were signiﬁcantly different at
each site. This is in agreement with the detection of signiﬁcantly different denitriﬁer
communities even along a 40 m nitrate and salinity gradient in a beach aquifer (Santoro,
2006), suggesting that denitriﬁer communities are highly site-speciﬁc and vary at
relatively small spatial scales. The Arctic (Barrow Canyon Shallow and East Hanna
Shoal Shallow) nirS clone libraries, though signiﬁcantly different, presented the highest
similarity and estimated number of shared OTUS between these two sites (Table 4.9).
Although on different transects with differing productivity levels, these two sites share
several biogeochemical characteristis and are geographically closer together than the
other sites, which could favor the development of similar denitriﬁer communities.
Interestingly, the Washington Margin nirS library was more similar to the Arctic libraries
than to the other Paciﬁc Northwest libraries. Despite the larger geographic distance,
biogeochemical similarities detected between these two continental margin environments

(Devol et al., 2005) might inﬂuence the development of more similar communities.

196

Castro-Gonzalez et al. (2005) also observed a closer, though still distant, clustering of
water column nirS sequences from the oxygen minimum zone in the eastern South
Paciﬁc with Arabian Sea oxygen minimum zone water column clones, as with sediment
clones from other Paciﬁc Ocean areas.

Clustering on the nirS tree further supported the signiﬁcant differences observed
between libraries. Clusters of highly similar sequences generally included clones
retrieved at only one location, however, several clusters including sequences from both
Arctic locations, as well as clusters including Arctic and Washington Margin clones,
were also observed, consistent with the similarities detected between those libraries.
Shallow Budd Inlet and West of Juan de Fuca nirS clones generally formed individual
clusters, consistent with their low similarity to other libraries. Despite individual site-
speciﬁc clustering of highly similar clones, no general separation of clones from different
locations was observed. Furthermore, most clones (95.1%) were more than 80% similar
to previously retrieved environmental clones from various geographic locations,
indicating an increasing coverage of nirS diversity by growing sequence databases.

Interestingly, the Shallow Budd Inlet nirS clusters previously identiﬁed and
quantiﬁed by real-time PCR (Chapter 3) did not include any sequences from other areas,
consistent with the lower or lack of detection of most of them at the other locations
(Table 4.3). This might suggest a speciﬁc adaptation to this environment. Decrease in
number of speciﬁc nirS gene and transcript phylotypes was also observed along the
Colne estuary from the head to the mouth (Smith et al., 2007). P. stutzeri nirS, on the
other hand, was detected at all studied locations, although it represented at most 0.66% of

the community in Washington Margin sediments and 1 to 2 orders of magnitude less at

197

the other sites. This is consistent with the frequent detection of this bacterium in marine
samples, although at varying abundances (Ward and Cockcroft, 1993; Griintzig et al.,
2001)

Shallow Budd Inlet and West of Juan de Fuca were also the most divergent sites
based on their bacterial community, while the Arctic and Washington Margin, presented
the most similar bacterial communities, coincident with the pattern observed for nirS.
However, the highest similarity was detected between East Hanna Shoal Shallow and
Washington Margin, rather than between the two Arctic sites, as observed for nirS. This
further supports a probable similarity in environmental conditions favoring the
development of similar communities at these two sites. The lower evolutionary rate of the
16S rRNA gene compared to nirS could be responsible for the establishment of more
similar communities regardless of the larger geographic distances. This pattern was also
observed in the 16S rRNA phylogenetic tree, with clusters of highly similar sequences
generally including clones from only one location, but no speciﬁc clustering of clones
from close geographic locations beyond that. Regretfully, it is not possible to draw a clear
correlation between 168 rRNA and nirS phylogenies, as strains with the same 16S rRNA
phylotype have been often observed to have notable differences in nirS phylotype
(Goregues et al., 2005).

In conclusion, the differentiation of denitriﬁer communities, as studied by the nirS
gene, seems to be mainly inﬂuenced by geographic location, followed by varying
environmental factors within one area, and least affected by sediment depth. This can be
correlated to the hypothesis proposed by Hughes Martiny et al. (2006) of multiple

microbial provinces and habitats, representing, respectively, historical and environmental

198

factors affecting the microbial community structure. Based on this hypothesis, if biotic
similarities between samples are correlated with environmental similarity and geographic
distance, multiple habitats and provinces can be deﬁned. On the other hand, for 16S
rRNA, due to its lower mutation rate, biotic similarities were less correlated to
geographic distance, therefore, provinces seem less differentiated and tend to form one

province with multiple habitats.

199

References

Armstrong, F. A. J., C. R. Stearns, and J. D. H. Strickland. 1967. The measurement
of upwelling and subsequent biological processes by means of the Technicon
Autoanalyser and associated equipment. Deep-Sea Res. 14:381-389.

Bates, N. R., M. H. P. Best, and D. A. Hansel]. 2005. Spatio-temporal distribution of
dissolved inorganic carbon and net community production in the Chukchi and Beaufort
Seas. Deep-Sea Res. 11 52:3303-3323.

Bowman, J. P. and R. D. McCuaig. 2003. Biodiversity, community structural shifts,
and biogeography of prokaryotes within Antarctic continental shelf sediment. Appl.
Environ. Microbiol. 69:2463-2483.

Braker, G., H. A. Ayala-deI-Rio, A. H. Devol, A. Fesefeldt, and J. M. Tiedje. 2001.
Community structure of denitriﬁers, Bacteria, and Archaea along redox gradients in
Paciﬁc Northwest marine sediments by terminal restriction fragment length
polymorphism analysis of ampliﬁed nitrite reductase (nirS) and 16S rRNA genes. Appl.
Environ. Microbiol. 67: 1893—1901 .

Braker, G., J. Zhou, L. Wu, A. H. Devol, J. M. Tiedje. 2000. Nitrite reductase genes
(nirK and nirS) as functional markers to investigate diversity of denitrifying bacteria in

Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol. 66: 2096—
2104.

Brandes, J. A., and A. H. Devol. 1995. Simultaneous nitrate and oxygen respiration in
coastal sediments: Evidence for discrete diagenesis. Journal of Marine Research. 53:771-
797.

Brewer, P. G. and D. W. Spencer. 1974. Colorimetric determination of manganese in
anoxic waters. Limnology and Oceanography 16:107-110.

Castro-Gonzalez, M., G. Braker, L. Farias, and O. Ulloa. 2005. Communities ‘of nirS-
type denitriﬁers in the water column of the oxygen minimum zone in the eastern South
Paciﬁc. Environ. Microbiol. 7: 1298-1306.

Cho, J.—C. and J. M. Tiedje. 2000. Biogeography and degree of endemicity of
ﬂuorescent Pseudomonas strains in soil. Appl. Environ. Microbiol. 66:5448-5456.

Cole, J. R., B. Chai, R. J. Farris, Q. Wang, A. S. Kulam-Syed-Mohideen, D. M.
McGarrell, A. M. Bandela, E. Cardenas, G. M. Garrity, and J. M. Tiedje. 2006. The
ribosomal database project (RDP-II): introducing myRDP space and quality controlled
public data. Nucleic Acids Res. 35:Dl69-D172.

Cole, J. R., B. Chai, R. J. Farris, Q. Wang, S. A. Kulam, D. M. McGarrell, G. M.

Garrity, and J. M. Tiedje. 2005. The Ribosomal Database Project (RDP 11): sequences
and tools for high-throughput rRNA analysis. Nucleic Acids Res. 33:D294-D296.

200

Devol, A., B. Chang, and J. Christensen. 2005. Denitriﬁcation and sulfate reduction in
Arctic continental margin sediments. Goldschmidt Conference Abstracts. A581.

Dhillon, A., A. Teske, J. Dillon, D. A. Stahl, and M. L. Sogin. 2003. Molecular
characterization of sulfate-reducing bacteria in the Guaymas Basin. Appl. Environ.
Microbiol. 69:2765-2772.

Dojka, M. A., P. Hugenholtz, S. K. Haack, and N. R. Pace. 1998. Microbial diversity,
in a hydrocarbon- and chlorinated-so[vent-contaminated aquifer undergoing intrinsic
bioremediation. Appl. Environ. Microbiol. 64: 3869-3877.

Fossing, H. and B. B. Jorgensen. 1989. Measurement of bacterial sulfate-reduction in

sediments: evaluation of single-step chromium reduction method. Biogeochemistry
8:205-222.

Goregues, C. M., V. D. Michotey, and P. C. Bonin. Molecular, biochemical, and
physiological approaches for understanding the ecology of denitriﬁcation. Microb. Ecol.
49:198-208.

Green, J. and B. J. M. Bohannan. 2006. Spatial scaling of microbial biodiversity.
Trends Ecol. Evol. 21:501-507.

Griintzig, V., S. C. Nold, J. Zhou, and J. M. Tiedje. 2001. Pseudomonas stutzeri nitrite
reductase gene abundance in environmental samples measured by real-time PCR. Appl.
Environ. Microbiol. 67:760-768.

Hall, T. A. 1999. A user-friendly biological sequence alignment editor and analysis
program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41:95-98.

Hannig, M., G. Braker, J. Dippner, and K. Jiirgens. 2006. Linking denitriﬁer
community structure and prevalent biogeochemical parameters in the pelagial of the
central Baltic Proper (Baltic Sea). FEMS Microbiol. Ecol. 57:260-271.

Hartnett, H. E. and A. H. Devol. 2003. Role of a strong oxygen-deﬁcient zone in the
preservation and degradation of organic matter: A carbon budget for the continental

margins of northwest Mexico and Washington State. Geochim. Cosmochim. Acta
67 :247-264.

Hedges, J. and J. Stern. 1984. Carbon and nitrogen determinations of carbonate
containing solids. Limnology and Oceanography 29:657-663.

Heijs, S. K., J. S. Sinninghe-Damste, and L. J. Forney. 2005. Characterization of a
deep-sea microbial mat from an active cold seep at the Milano mud volcano in the
Eastern Mediterranean Sea. FEMS Microbiol. Ecol. 54:47-56.

Hill, T. C. J., K. A. Walsh, J. A Harris, B. F. Moffet. 2003. Using ecological diversity
measures with bacterial communities. FEMS Microbiol. Ecol. 43:1-11.

201

Hill, V. and G. Cota. 2005. Spatial patterns of primary production on the shelf, slope
and basin of the Western Arctic in 2002. Deep-Sea Res. 52:3344-3354.

Horner—Devine, M. C., M. Lage, J. B. Hughes, and B. J. M. Bohannan. 2004. A taxa-
area relationship for bacteria. Nature 432:750-753.

Horner-Devine, M. C., M. A. Leibold, V. H. Smith, and B. J. M. Bohannan. 2003.
Bacterial diversity patterns along a gradient of primary productivity. Ecol. Lett. 6:613-
622.

Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K.
Maskell, and C. A. Johnson. 2001. Climate change 2001: the scientiﬁc basis.
Cambridge University Press, Cambridge.

Hugenholtz, P., C. Pitulle, K. L. Hershberger, and N. R. Pace. 1998. Novel division
level bacterial diversity in a Yellowstone hot spring. J. Bacteriol. 180:366—376.

Hughes Martiny, J. B., B. J. M. Bohannan, J. H. Brown, R. K. Colwell, J. A.
Fuhrman, J. L. Green, M. C. Horner-Devine, M. Kane, J. Adams Krumins, C. R.
Kuske, P. J. Morin, S. Naeem, L. Ovreas, A.-L. Reysenbach, V. H. Smith, and J. T.

Staley. 2006. Microbial biogeography: putting microorganisms on the map. Nature
42102-112. .

Ivanova, E. P., N. V. Zhukova, A. M. Lysenko, N. M. Gorshkova, A. F. Sergeev, V.
V. Mikhailov, and J. P. Bowman. 2005. Loktanella agm'ta sp. Nov. and Loktanella
rosea sp. Nov., from the north-west Paciﬁc Ocean. Int. J. Syst. Evol. Microbiol. 55:2203-
2207.

Jayakumar, D. A., C. A. Francis, S. W. A. Naqvi, and B. B. Ward. 2004. Diversity of
nitrite reductase genes (nirS) in the denitrifying water column of the coastal Arabian Sea.
Aquat. Microb. Ecol. 34:69-78.

Jukes, T. H. and C. R. Cantor. 1969. Evolution of protein molecules. In Munro, H. N.
(ed.). Mammalian Protein Metabolism. Academic Press, New York.

Kirchman, D. L. 2002. The ecology of Cytophaga—Flavobacteria in aquatic
environments. FEMS Microbiol. Ecol. 39:91-100.

Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular
evolutionary genetics analysis Software. Bioinformatics. 17 : 1244-1245.

Li, L., C. Kato, and K. Horikoshi. 1999. Microbial diversity in sediments collected
from the deepest cold-seep area, the Japan Trench. Mar. Biotechnol. 1:391-400.

Liu, X., C. E. Bagwell, L. Wu,A. H. Devol, andJ. Zhou. 2003. Molecular diversity of
sulfate-reducing bacteria from two different continental margin habitats. Appl. Environ.
Microbiol. 69: 6073- 6081.

202

Liu, X., S. M. Tiquia, G. Holguin, L. Wu, S. C. Nold, A. H. Devol, K. Luo, A. V.
Palumbo, J. M. Tiedje, and J. Zhou. 2003. Molecular diversity of denitrifying genes in
continental margin sediments within the oxygen-deﬁcient zone off the Paciﬁc coast of
Mexico. Appl. Environ. Microbiol. 69:3549-3560.

Lopez-Garcia, P., S. Duperron, P. Philippot, J. F oriel, J. Susini, and D. Moreira.
2003. Bacterial diversity in hydrothermal sediment and epsilonproteobacterial dominance
in experimental microcolonizers at the Mid-Atlantic Ridge. Environ. Microbioll 5:961-
976.

Luna, G. M., A. Dell’Anno, L. Giuliano, and R. Danovaro. 2004. Bacterial diversity in
deep Mediterranean sediments: relationship with the active bacterial ﬁaction and
substrate availability. Environ. Microbiol. 6:745-753.

Mittelbach, G. G., C. F. Steiner, S. M. Scheiner, K. L. Gross, H. L. Reynolds, R. B.
Waide, M. R. Willig, S. I. Dodson, and L. Gough. 2001. What is the observed
relationship between richness and productivity? Ecology 82:2381-2396.

Mussmann, M., K. Ishii, R. Rabus, and R. Amann. 2005. Diversity and vertical
distribution of cultured and uncultured Deltaproteobacteria in an intertidal mud ﬂat of the
Wadden Sea. Environ. Microbiol. 7 2405-418.

Nei, M. and S. Kumar. 2000. Molecular evolution and phylogenetics. Oxford University
Press, New York.

Nercessian, O., Y. Fouquet, C. Pierre, D. Prieur, and C. Jeanthon. 2005. Diversity of
Bacteria and Archaea associated with a carbonate-rich metalliferous sediment sample
from the Rainbow vent ﬁeld on the Mid-Atlantic Ridge. Environ. Microbiol. 7 :698-714.

Nogales, B., K. N. Timmis, D. B. Nedwell and A. M. Osborn. 2002. Detection and
diversity of expressed denitriﬁcation genes in estuarine sediments after reverse
transcription-PCR ampliﬁcation from mRNA. Appl. Environ. Microbiol. 68:5017-5025.

Nold, S. C., J. Zhou, A. H. Devol, and J. M. Tiedje. 2000. Paciﬁc Northwest marine
sediments contain ammonia-oxidizing bacteria in the [3 subdivision of the Proteabacteria.
Appl. Environ. Microbiol. 66:4532-4535.

Penton, C. R., A. H. Devol, and J. M. Tiedje. 2006. Molecular evidence for the broad
distribution of anaerobic ammonium-oxidizing bacteria in freshwater and marine
sediments. Appl. Environ. Microbiol. 72:6829-6832.

Polymenakou, P. N., S. Bertilsson, A. Tselepides, and E. G. Stephanou. 2005.
Bacterial community composition in different sediments from the Eastern Mediterranean
Sea: a comparison of four 16S ribosomal DNA clone libraries. Microb. Ecol. 50:447-462.

Ramette, A. and J. M. Tiedje. 2007. Biogeography: An emerging cornerstone for
understanding prokaryotic diversity, ecology, and evolution. Microbiol. Ecol. 53:197-
207.

203

Ramette, A. and J. M. Tiedje. 2007. Multiscale responses of microbial life to spatial

distance and environmental heterogeneity in a patchy ecosystem. Proc. Nat. Acad. Sci.
10422761-2766.

Ravenschlag, K., K. Sahm, J. Pernthaler, and R. Amann. 1999. High bacterial
diversity in permanently cold marine sediments. Appl. Environ. Microbiol. 65:3982-
3989.

Saitou, N. and M. Nei. 1987. The neighbour joining method: a new tool for
reconstructing phylogenetic trees. Mol. Biol. Evol. 41406-425.

Sambrook, J. and D. W. Russell. 2001. Molecular cloning: a laboratory manual. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Santoro, A. E., A. B. Boehm and C. A. Francis. 2006. Denitriﬁer community

composition along a nitrate and salinity gradient in a coastal aquifer. Appl. Environ.
Microbiol. 72: 2102-2109.

Scala, D. J. and L. J. Kerkhof. 1999. Diversity of nitrous oxide reductase (nosZ) genes
in continental shelf sediments. Appl. Environ. Microbiol. 65: 1681-1687.

Scala, D. J. and L. J. Kerkhof. 2000. Horizontal heterogeneity of denitrifying bacterial
communities in marine sediments by terminal restriction fragment length polymorphism
analysis. Appl. Environ. Microbiol. 66:1980-1986.

Schloss, P. D., B. R. Larget, and J. Handelsman. 2004. Integration of microbial

ecology and statistics: a test to compare gene libraries. Appl. Environ. Microbiol.
70:5485-5492.

Schloss, P. D. and J. Handelsman. 2005. Introducing DOTUR, a computer program for
deﬁning operational taxonomic units and estimating species richness. Appl. Environ.
Microbiol. 71:1501-1506.

Schloss, P. D. and J. Handelsman. 2006. Introducing SONS, a tool for operational

taxonomic unit-based comparisons of microbial community memberships and structures.
Appl. Environ. Microbiol. 72:6773-6779.

Smith, C. J., D. B. Nedwell, L. F. Dong, and A. M. Osborn. 2007. Diversity and
abundance of nitrate reductase genes (narG and napA), nitrite reductase genes (nirS and
nrfA), and their transcripts in estuarine sediments. Appl. Environ. Microbiol. 73:3612-
3622. '

Stevens, H., M. Stubner, M. Simon, and T. Brinkhoff. 2005. Phylogeny of
Proteobacteria and Bacteroidetes from oxic habitats of a tidal ﬂat ecosystem. FEMS
Microbiol. Ecol. 54:351-365.

Stookey, L. L. 1970. Ferrozine —a new spectrophotometric reagent for iron. Analytical
Chemistry 42:779-781.

204

Strickland, J. D. H. and T. R. Parsons. 1972. A practical handbook of seawater
analysis. Fisheries Research Board of Canada, Ottawa, Ontario.

Tabatabai, M. A. 1974. A rapid method for determination of sulfate in water samples.
Environ. Letters 7 2237-243.

Tarnegai, H., R. Aoki, S. Arakawa, and C. Kato. 2007. Molecular analysis of the
nitrogen cycle in deep- s-ea microorganisms from the Nankai Trough: genes for

nitriﬁcation and denitriﬁcation from deep-sea environmental DNA. Extremophiles
11:269-275.

Tamegai, H., L. Li, N. Masui, and C. Kato. 1997. A denitrifying bacterium from the
deep sea at 11,000 m depth. Extremophiles 1:207-211.

ter Braak, C. J. F. 1986. Canonical correspondence analysis: a new eigenvector
technique from multivariate direct gradient analysis. Ecology 67 :1 167-1 179.

Teske, A., K. U. Hinrichs,K, V. Edgcomb, A. de Vera Gomez, D. Kysela, S. P. Sylva,
M. L. Sogin, and H. W. Jannasch. 2002. Microbial diversity of hydrothermal sediments

in the Guaymas Basin: evidence for anaerobic methanotrophic communities. Appl.
Environ. Microbiol. 68: 1994-2007.

Tiquia, S. M., S. Gurczynski, A. Zholi, and A. Devol. 2006. Diversity of
biogeochemical cycling genes from Puget Sound sediments using DNA microarrays.
Environmental Technology. 27: 1377-1 389.

Torsvik, V., L. Ovreas, and T. F. Thingstad. 2002. Prokaryotic diversity-magnitude,
dynamics, and controlling factors. Science 296: 1 064-1066.

Tuomainen, J. M., S. Hietanen, J. Kuparinen, P. J. Martikainen, and K. Servomaa.
2003. Baltic Sea cyanobacterial bloom contains denitriﬁcation and nitriﬁcation genes, but
has negligible denitriﬁcation activity. FEMS Microbiol. Ecol. 45:83-96.

Van Mooy, B. A. s., A. H. Devol, and R. G. Keil. 2003. Quantifying 3H-thymidine
incorporation rates by a phylogenetically deﬁned group of marine planktonic bacteria
(Bacteroidetes phylum). Environ. Microbiol. 6: 1061-1069.

Walsh, J. J. 1991. Importance of continental margins in the marine biogeochemical
cycling of carbon and nitrogen. Nature 350:53-55.

Ward, B. B. and A. R. Cockcroft. 1993. Immunoﬂuorescence detection of the
denitrifying strain Pseudomonas stutzeri (ATCC 14405) in seawater and intertidal
sediment environments. Microb. Ecol. 25: 233- 246.

Whitman, W. B., D. C. Coleman, and W. J. Wiebe. 1998. Prokaryotes: The unseen
majority. Proc. Natl. Acad. Sci. 95:6578-6583.

205

CHAPTER 5

CONCLUSIONS AND FUTURE PERSPECTIVES

At the time this study was started, microbial ecology was lacking an accurate way
to rapidly quantify genes and the corresponding organisms in the environment, especially
when speciﬁc groups that may not be abundant or active in the studied environment
needed to be targeted. Real-time PCR effectively ﬁlled that niche and this study provided
the background testing necessary before implementing it on a wide scale for the
quantiﬁcation of functional genes in the environment. It was proved that real-time PCR
was speciﬁc, sensitive, precise and accurate on a wide array of environmental samples,
which paved the way for its broad use. The detection of P. stutzeri nirS by this technique
in a broad range of habitats, although at highly varying abundances, suggests that this
commonly isolated denitriﬁer, though fairly cosmopolitan, might not be as dominant a
member of the denitrifying community as most people had thought based on its frequent
isolation. However, its persistence even in deep unmixed sediments, not exposed to
oxygen or nitrate for years, in addition to its relative increase within the bacterial
community due to sample manipulation, indicates a high level of resistance to various
and changing environmental conditions and its responsiveness to favorable. growth
conditions, probably explaining its ease of isolation.

On the other hand, most of the detected nirS genes, as usually observed in diversity
studies, were unrelated to any cultured denitriﬁer. Their relative decrease in the bacterial
community after sample manipulation suggests that they might be sensitive to factors not

considered when their isolation is attempted, explaining their absence from the cultured

206

strain collections. Some of these novel nirS genes, however, might also be part of
alternative pathways in non-denitrifying organisms, as the nirS gene detected in the
genome of the uncultured anammox bacterium Kuenenia stuttgartiensis, which has been
suggested to play a role in an alternative anammox pathway (Strous et al., 2006). The
question remains if these novel genes are all functional in the environment, or if at least
some of them are remnants of nirS genes which have lost their functionality due to
mutation in the gene itself or its regulatory system. Only isolation of denitriﬁers with one
of these novel genes will answer this question.

The vertical distribution of denitriﬁers was correlated to the bioturbation of the
sediments and not to the redox gradients as had been often suggested. A highly conserved
community structure associated also to denitriﬁcation capacity was detected throughout
the mixed layer, but with its sharp shift in the deep unmixed sediments. Therefore, the
eventual exposure of these denitriﬁers to oxygen and nitrate by the particle mixing
activity of burrowers as well as the increase of oxygen and nitrate exchange surface by
the burrows themselves apparently led to the maintenance of an active denitrifying
community far below the nitrate penetration depth into bulk sediments. The rapid
initiation of denitriﬁcation after exposure to nitrate suggests that the denitrifying enzymes
must already have been present in the organisms, although most of the bulk sediment
bacteria were not exposed to nitrate in situ at the time of retrieval. Expression of these
enzymes regardless of the presence of nitrate may give them a competitive advantage in
an electron acceptor-limited environment. On the other hand, some of these bacteria may
also survive based on alternative metabolic pathways, such as fermentation, which has

been detected in the denitrifying bacterium Pseudovibrio denitrificans recently isolated

207

from shallow sea water (Shieh et al., 2004). In any case, this indicates that denitriﬁcation
or at least denitriﬁcation capacity, extends well beyond the top few centimeters to which
nitrate is often restricted, therefore suggesting the necessity of considering the whole
mixed layer when modeling or predictions are made about denitriﬁcation and the
responsible organisms.

Denitriﬁer communities presented a much higher differentiation on the horizontal
than on the vertical dimension, however different scales are involved for each. Sediments
from different sites within one area with gradually changing environmental conditions
revealed more distantly related communities than sediments from different depths within
one core, but lower differentiation than sediments from distant geographic locations.
Clone libraries suggested the existence of signiﬁcantly different bacterial and denitriﬁer
communities at Shallow Budd Inlet, Washington Margin, West of Juan de Fuca,Barrow
Canyon Shallow, and East Hanna Shoal Shallow sediments, represented as well by the
site-speciﬁc clustering of highly similar sequences observed for both genes. However,
while nirS exhibited a higher similarity between the closely located Arctic transects, no
clear grouping of 16S rRNA gene sequences from different sites was observed, and a
higher similarity between the Arctic East Hanna Shoal Shallow and Washington Margin
bacterial communities was suggested. This discrepancy could result from the lower
coverage of the 16S rRNA gene libraries compared to the nirS gene libraries, but it might
also result from the lower mutation rate of the 16S rRNA gene, allowing the further
dispersal of highly similar 16S rRNA gene sequences. Hughes Martiny et al. (2006)
suggested four alternative hypotheses about the effect of environmental, represented as

habitats, and historical, represented as provinces, effects on communities. Depending on

208

the presence of single or multiple habitats and provinces, different distributions of
communities would be observed. Correlation of biotic similarity with environmental
similarity as well as with geographic distance would be indicative of the existence of
multiple habitats as well as provinces for the distribution of microorganisms. The
distribution of the nirS genes seems to correlate well with this view, while the
distribution of the 16S rRNA genes suggests a tendency of certain provinces to fuse into
bigger provinces, moving towards, although not reaching, the Baas-Becking hypothesis,
“everything is everywhere, the environment selects”. Hence, the way in which we deﬁne
“everything” e.g. a species or strain becomes an important issue, as the results Will vary
according to that deﬁnition and the characters measured. If we concentrate on all genes in
a genome as used by Konstantinidis and Tiedje (2005), more differences, and therefore
more provinces will be established. This tendency will increase even further if intergenic
regions are also considered. However, the question remains as to how important these
differences are in the functioning of the organism and if they actually represent
adaptations to the environment or are the result of neutral ecological drift and other
factors as has been recently suggested (Ramette and Tiedje, 2007).

In this study, the analyzed sites were not located along a transect or some clear
gradient of environmental conditions, but rather formed a patchy array. This makes the
correlation of the genotypic patterns seen with environmental variables more difﬁcult, as
several variables vary in unrelated ways from one site to the other. However, a general
trend towards higher diversity in intermediate productivity areas was detected, a pattern
also observed on the vertical axis. If this results from a balance between competitive

ability and resistance to predators remains to be established. Understanding the effect that

209

 

 

productivity has on the development of the denitriﬁer and general bacterial community is
of great importance to better predict changes that could result from increased productivity
in areas such as the Arctic, which has been suggested will be most highly affected by

global warming.

Future perspectives

As stated above, most of the nirS gene sequences retrieved in this study, as well as
in other diversity studies, are unrelated to known denitriﬁers. However, their high
abundance suggests that the bacteria that harbor them might play an important role in the
community. This remains questionable until the functionality of these novel genes can be
proven. The detection of mRNA sequences (Nogales et al., 2002) closely related to some
of the novel nirS genes detected in this and other studies, and, furthermore, their high
abundance in environmental samples (Smith et al., 2007) indicates that at least some of
these novel genes are being expressed in nature, constituting a further evidence of their
functionality. However, the translation into a functional protein of one of these abundant
sequences has not yet been demonstrated. Several approaches have been tried in this
study (Appendix) to retrieve an integral novel nirS gene in order to test its functionality
in a complementation assay. The complexity of the environmental DNA sample was,
however, a limiting factor and impeded the successful retrieval of a complete nirS gene
from marine sediments. Enrichment of the target novel nirS gene sequence with a
magnetic capture-hybridization assay (Jacobsen, 1995) might help overcome this
problem. However, the regulation of the expression of this gene and the ultimate function

in the original organism can only be elucidated by isolation of a denitrifying organism

210

that carries the gene. This has proven to be a difﬁcult task and the apparent sensitivity of
some of these novel denitriﬁers to manipulation of the samples might be a cause as well
as our inability to ﬁnd the right selective conditions. Therefore, screening for the
presence of these novel genes along with careful dissection of the various steps involved
in the attempt to isolate a novel denitriﬁer might help decipher critical factors leading to
the loss of these organisms during the isolation process. Recently, several new SARll
and other uncultured strains have been isolated from seawater with an improved high-
throughput dilution-to-extinction protocol (Stingl et al., 2007). Although not necessarily
effective for the more complex sediment community, some of the improvements could be
considered to aid in the isolation of novel denitriﬁers from this environment.

A common denominator for all the clone libraries presented in this study was the
low coverage of the diversity present, especially for the 16S rRNA gene. Traditional
cloning and sequencing techniques limited the number of sequences that could be
obtained to several tens or hundreds. However, with the development of pyrosequencing
(Ronaghi et al., 1996) and its improvement to 454 pyrosequencing (Margulies et al.,
2005) for high-throughput generation of sequencing data, tens of thousands of sequences
can be generated in only a few hours, signiﬁcantly improving our ability to cover the
diversity present in the environment of interest. Roesch et al. (2007) recently studied the
microbial diversity in soil from four different areas with this technique, which allowed
them to obtain between 26,140 and 53,533 sequences from each site, more than two
orders of magnitude more than the number generated in this study. Most of the taxonomic
groups identiﬁed in these soil samples, were, however, present at less than 0.5% relative

abundance in the community, suggesting that they most probably would not have been

211

detected in a clone library of a similar size to the ones used in this study. Sediments have
been shown to be as diverse as soil, therefore, applying 454 pyrosequencing to the
sediments used here would as well give us further information about the broad diversity
present to better understand the function of the community and the players involved.
Furthermore, larger coverage would also help to better establish if the differences we are
seeing between samples are real or artifacts due to undersampling of the diversity present.
In any case, a deeper knowledge on the ﬁne tuning of microbial communities is

beginning to be possible.

212

References

Baas-Becking, L. G. M. 1934. Geobiologie of Inleiding Tot de Milieukunde. W. P. van
Stockum & Zoon N. V., The Hague, The Netherlands.

Hughes Martiny, J. B., B. J. M. Bohannan, J. H. Brown, R. K. Colwell, J. A.
Fuhrman, J. L. Green, M. C. Horner-Devine, M. Kane, J. Adams Krumins, C. R.
Kuske, P. J. Morin, S. Naeem, L. ﬂvrer‘is, A.-L. Reysenbach, V. H. Smith, and J. T.

Staley. 2006. Microbial biogeography: putting microorganisms on the map. Nature
4:102-112.

Jacobsen, C. S. 1995. Microscale detection of speciﬁc bacterial DNA in soil with a
magnetic capture-hybridization and PCR ampliﬁcation assay. Appl. Environ. Microbiol.
61:3347-3352.

Konstantinidis, K. T. and J. M. Tiedje. 2005. Genomic insights that advance the
species deﬁnition for prokaryotes. Proc. Natl. Acad. Sci. 102:2567-2572.

Margulies, M., M. Egholm, W. E. Altman, S. Attiya, J. S. Bader, L. A. Bemben, J.
Berka, M. S. Braverman, Y.-J. Chen, Z. Chen, S. B. Dewell, L. Du, J. M. Fierro, X.
V. Gomes, B. C. Godwin, W. He, S. Helgesen, C. H. Ho, G. P. Irzyk, S. C. Jando, M.
L. I. Alenquer, T. P. Jarvie, K. B. Jirage, J.-B. Kim, J. R. Knight, J. R. Lanza, J. H.
Leamon, S. M. Lefkowitz, M. lei, J. Li, K. L. Lohman, H. Lu, V. B. Makhijani, K. E.
McDade, M. P. McKenna, E. W. Myers, E. Nickerson, J. R. Nobile, R. Plant, B. P.
Puc, M. T. Ronan, G. T. Roth, G. J. Sarkis, J. F. Simons, J. W. Simpson, M.
Srinivasan, K. R. Tartaro, A. Tomasz, K. A. Vogt, G. A. Volkmer, S. H. Wang, Y.
Wang, M. P. Weiner, P. Yu, R. F. Begley, and J. M. Rothberg. 2005. Nature 437 :376-
380.

Nogales, B., K. N. Timmis, D. B. Nedwell and A. M. Osborn. 2002. Detection and
diversity of expressed denitriﬁcation genes in estuarine sediments after .reverse
transcription-PCR ampliﬁcation from mRNA. Appl. Environ. Microbiol. 68:5017-5025.

Ramette, A. and J. M. Tiedje. 2007. Multiscale responses of microbial life to spatial
distance and environmental heterogeneity in a patchy ecosystem. Proc. Natl. Acad. Sci.
104:2761-2766.

Roesch, L. F. W., R. R. Fulthorpe, A. Riva, G. Casella, A. K. M. Hadwin, A. D.
Kent, S. H. Daroub, F. A. O. Camargo, W. G. Farmerie, and E. W. Triplett. 2007.
Pyrosequencing enumerates and contrasts soil microbial diversity. The ISME Journal
1:283-290.

Ronaghi, M., S. Karamohamed, B. Petterson, M. Uhlén, and P. Nyrén. 1996. Real-
time DNA sequencing using detection of pyrophosphate release. Anal. Biochem. 242:84-
89.

213

Shieh, W. Y., Y.-T. Lin, and W. D. Jean. 2004. Pseudovibrio denitrificans gen. nov.,
sp. nov., a marine, facultatively anaerobic, ferrnentative bacterium capable of
denitriﬁcation. Int. J. Syst. Evol. Microbiol. 54:2307-2312.

Smith, C. J., D. B. Nedwell, L. F. Dong, and A. M. Osborn. 2007. Diversity and
abundance of nitrate reductase genes (narG and napA), nitrite reductase genes (nirS and

nrfA), and their transcripts in estuarine sediments. Appl. Environ. Microbiol. 73:3612-
3622.

Stingl, U., H. J. Tripp, and S. J. Giovannoni. 2007. Improvements of high-throughput
culturing yielded novel SARll strains and other abundant marine bacteria from the
Oregon coast and the Bermuda Atlantic Time Series study site. The ISME Journal 1:361-
371.

Strous, M., E. Pelletier, S. Mangenot, T. Rattei, A. Lehner, M. W. Taylor, M. Horn,
H. Daims, D. Bartol-Mavel, P. Wincker, V. Barbe, N. Fonknechten, D. Vallente, B.
Segurens, C. Schenowitz-Truong, C. Médigue, A. Collingro, B. Snel, B. E. Dutilh, H.
J. M. Op den Camp, C van der Drift, I. Cirpus, K. T. van de Pas-Schoonen, H. R.
Harhangi, L. van Niftrik, M. Schmid, J. Keltjens, J. van de Vossenberg, B. Kartal,
H. Meier, D. Frishman, M. A. Huynen, H.-W. Mewes, J. Weissenbach, M. S. M.
Jetten, M. Wagner, and D. Le Paslier. 2006. Deciphering the evolution and metabolism
of an anammox bacterium from a community genome. Nature 440:790-794.

214

APPENDIX

RECOVERY OF NOVEL nirS GENES FROM NATURE AND
TESTING OF THEIR FUNCTIONALITY

Molecular techniques have led to the retrieval from nature of an enormous number
of novel gene sequences which might play an important role in the chemistry of the
biosphere. However, the functionality of these sequences has generally not been tested
and therefore their in situ function remains questionable. This is also true for
denitriﬁcation genes, like the nirS gene coding for the heme ed, nitrite reductase. In
previous studies, three clusters of nirS clones which did not include any nirS from a
cultured denitriﬁer were detected in sediment samples from the Paciﬁc Northwest
(Braker et al., 2000). Furthermore, in Chapter 3, some clusters of novel nirS sequences
were identiﬁed in Shallow Budd Inlet sediments, which were more abundant than
commonly isolated denitriﬁers such as Pseudomonas stutzeri at this site. These novel
sequences could correspond to novel denitriﬁers, however their functionality is unknown.
A complementation assay was therefore applied in order to test the ability of these
sequences to restore the denitriﬁcation pathway in a NirS' mutant.

The complementation assay involved the cloning of a novel nirS gene into the
pSUP104 vector, a broad host range vector which can be maintained in Pseudomonas
(Priefer et al., 1985). Zumft et al. (1988) developed a Tn5-induced mutant strain of P.
stutzeri (mutant MK202), which lacks the heme cdl nitrite reductase and is therefore
defective in nitrite reduction. This mutant strain was used to test the ability of the cloned
nirS gene to complement the defect in nitrite reduction after the introduction of the vector
pSUPlO4 with the potential nirS insert into the mutant by conjugation following the
protocol by Zumﬁ et al. (1985). The restoration of the denitriﬁcation pathway was tested
by growth in presence of nitrite as the only electron acceptor.

The previously retrieved nirS clones represent only part of the nirS gene sequence.
Therefore, a complete gene sequence had to be retrieved ﬁrst in order to test its
functionality by complementation of the nirS mutant. Several different approaches were
used 1n my attempts to achieve this.

- Direct cloning of partially restricted environmental DNA (~2000-4500 bp long)
retrieved from Shallow Budd Inlet sediments into pSUPlO4.

- Inverse PCR to retrieve the ends and ﬂanking regions of a novel nirS gene,
which would allow the design of speciﬁc primers for the ampliﬁcation of a
complete novel nirS gene. Inverse PCR involved partial restriction of
environmental genomic DNA followed by self-ligation and ampliﬁcation by
primers representing reverse complements of general nirS primers nirSlF and
nirS6R (Braker et al., 1998).

215

- Inverse PCR, as above, but using primers representing reverse complements of
cluster—speciﬁc primers designed for the quantiﬁcation by real-time PCR of
abundant novel clusters retrieved from the same site (Chapter 3).

- Ampliﬁcation of ends and ﬂanking regions of novel nirS genes, by combination
of general or cluster-speciﬁc nirS primers with ERIC primers for repetitive
elements (Rademaker et al., 1998).

Regretfully, none of the approaches led to the retrieval of a novel nirS gene ﬁom
environmental samples. Direct cloning and the use of ERIC primers seem to be the least
appropriate, due to the large number of steps with loss of genetic material involved in the
former, and the signiﬁcantly higher ampliﬁcation of fragments by ERIC primers alone
rather than combined with nirS primers in the latter. Inverse PCR, especially with cluster-
speciﬁc primers, seems to be the more reasonable approach, however, some
improvements could be done, such as the use of restriction enzymes leaving protruding
ends, rather than blunt ends (AluI and RsaI were used) for the partial restriction of the
DNA, which would increase the ligation efficiency. The complexity of the sample was
probably the most detrimental factor. Therefore, previous enrichment of the target gene,
for example by the use of a magnetic capture-hybridization assay (Jacobsen, 1995), might
help overcome this problem.

216

References

Braker, G., A. F esefeldt, and K.-P. Witzel. 1998. Development of PCR primer systems
for ampliﬁcation of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria
in environmental samples. Appl. Environ. Microbiol. 64:3769-3775.

Braker, G., J. Zhou, L. Wu, A. H. Devol, J. M. Tiedje. 2000. Nitrite reductase genes
(nirK and nirS) as functional markers to investigate diversity of denitrifying bacteria in

Paciﬁc Northwest marine sediment communities. Appl. Environ. Microbiol. 66: 2096—
2104.

Jacobsen, C. S. 1995. Microscale detection of speciﬁc bacterial DNA in soil with a
magnetic capture-hybridization and PCR ampliﬁcation assay. Appl. Environ. Microbiol.
61:3347-3352.

Priefer, U. B., R. Simon, and A. Piihler 1985. Extension of the host range of
Escherichia coli vectors by incorporation of RSF 1010 replication and mobilization
functions. J. Bacteriol. 163:324-330.

Rademaker, J. L. W., F. J. Louws, and F. J. de Bruijn. 1998. Characterization of the
diversity of ecologically important microbes by rep-PCR genomic ﬁngerprinting. In A.
D. L. Akkermans, J. D. van Elsas, and F. J. de Bruijn (eds), Molecular Microbial
Ecology Manual. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Zumft, W. G., K. Dohler, and H. Kiirner 1985. Isolation and characterization of
transposon Tn5-induced mutants of Pseudomonas perfectomarina defective in nitrous
oxide respiration. J. Bacteriol. 163:918-924.

Zumft, W. G., K. Dohler, H. Kiirner, S. Lochelt, A. Viebrock, and K. Frunzke 1988.
Defects in cytochrome cdl-dependent nitrite respiration of transposon TNS-induced
mutants from Pseudomonas stutzeri. Arch.Microbiol. 149:492-4

217

IIIjlljljjjijilﬂljjljjl[WilliI