‘ . u
I .. . . ,
éﬁﬁg ..
1.3.

. . :.
Yr!

:mﬁww

é a.

v
_. .1.
.33317

3.139
I I .31?
nxituz. 2;; ‘.

diagﬁdﬁn

 

 

 

5%

IQIO

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

IDENTIFICATION OF NOVEL GENES INVOLVED IN ZEBRA
MUSSEL (DREISSENA POLYMORPHA) UNDERWATER
ADHESION MECHANISM

presented by

WEI XU

has been accepted towards fulﬁllment
of the requirements for the

Doctoral degree in Pathology

 

 

MO‘AQW‘LQ g Eat S. 0L

Major Professor’s Signature

la} IA 1 2009

Date

MSU is an Afﬁnnative 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.

 

DATE DUE

DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KIProj/Acc&Pres/ClRC/DateDue.indd

 

 

 

IDENTIFICATION OF NOVEL GENES INVOLVED IN
ZEBRA MUSSEL (DREISSENA POL YMORPl-IA)
UNDERWATER ADHESION MECHANISM

By

Wei Xu

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Pathology

2009

ABSTRACT

IDENTIFICATION OF NOVEL GENES INVOLVED IN
ZEBRA MUSSEL (DREISSENA POLYMORPHA)
UNDERWATER ADHESION MECHANISM

By

Wei Xu

Zebra mussels (Dreissena polymorpha) have invaded North America causing economic
and ecologic devastation. The zebra mussel attaches ﬁrmly to underwater substrates
through a complex system of exocrine glands, byssal threads, and adhesive plaques. In
this study, tools were developed and experiments were designed in order to better
understand the zebra mussel underwater adhesion mechanisms. A normalized cDNA
library of the zebra mussel byssus was constructed with the subtractive suppression
hybridization (SSH) technique. 750 non-redundant expressed sequence tags (ESTs) were
obtained from the library. A cDNA microarray was developed with the PCR products of
716 ESTs selected from the cDNA library. The newly developed cDNA microarray was
successfully used to compare between two groups of mussels with different byssogenenic
activities. Between the two groups, 16 genes were differentially expressed with statistical
signiﬁcance. The results were validated by the quantitative PCR (qPCR). To follow up on
genes that are either up or downregulated along the course of byssogenesis, a microarray
time-course experiment was designed. Samples were collected at seven time intervals; 12
hours, 1 day, 2 days, 3 days, 4 days, 7 days, and 2| days after severing the byssal threads

in the treatment group. The numbers of differentially expressed genes identiﬁed at these

time points were 13, I3, 20, l7, I6, 20, and 29 respectively. An additional experiment
was designed to identify the differentially expressed genes in response to the changes of
byssogenesis status and environmental factors, including temperature, current velocity,
and dissolved oxygen levels, as well as the status of byssogenesis on the expression of
foot-unique genes. The expression profiles of 18 genes were found to be altered by two
experimental factors, while 117 genes had differentially expressed proﬁles in response to
only one experimental factor. The numbers of the genes modulated by byssogenesis
status, temperature, dissolved oxygen, and current velocity were 59, 27, 26, and 9,
respectively. Seven genes identiﬁed by the time-course experiment and four genes
identiﬁed by factorial analysis assay were validated by qPCR with the results consistent
to microarray results. Detected by RNA ﬂuorescent in situ hybridization (FISH), both
genes BG20_AOl and 8697/ l92_BO6 were found to be expressed within the thread-
forrning glandular cells. The genes BGIS_FO3 and BGI6_H05 were expressed by stem-
forming gland cells and plaque-forming gland cells, respectively. A full length gene in
the microarray, homologous to insect defensin A, was cloned from the zebra mussel foot.
The analysis of the D. polymorpha defensin (Dpd) suggested that the Dpd is homologous
to the insect defensin A. The expression of Dpd in hemocytes was found to be inducible
by stimulation with lipopolysaccharides (LPS), peptidoglycan (PGN), and zymosan
(ZYM) using qPCR. The mature recombinant Dpd showed inhibition activity against four
strains of Gram-negative bacteria and one Gram-positive bacterium. Dpd was present not
only in the foot, but also in a number of other tissues. Interestingly, expression levels of
Dpd during the early stage of byssogenesis were mostly higher than the non-byssogenesis

status, which suggested that the activities of Dpd were associated with byssogenesis.

Copyright by
Wei Xu

2009

 

DEDICATION

This dissertation is dedicated to my wife, my parents, and my dearest friends.

ACKNOWLEDGEMENTS

It is a pleasure to thank those who made this thesis possible.

Foremost, I would like to express my sincere gratitude to my major advisor Prof.
Mohamed Faisal for the motivation, enthusiasm and inspiration, which was always there
when I needed it. This thesis would not have been possible without his sincere guidance
and encouragement.

I would also like to thank the rest of my thesis committee: Prof. Paul Coussens, Prof.
Matti Kiupel, Prof. Orlando Samelle, and Prof. Robert Bowker, for their greatest support
during my Ph.D. study and invaluable advice.

Special appreciation goes to Prof. Robert Tempelman who gave great help in
experimental design and microarray data analyses.

I am indebted to my many of my colleagues for a stimulating and fun environment in
which to learn and grow, especially grateful to Thomas Loch, Michelle Gunn, Carolyn
Schultz, Robert Kim, Elena Millard, and Drew Winters.

I must thank the current and former members in the Center for Animal Functional
Genomics, Sue Sipkovsky, Christopher Colvin, Xiaoning Ren, and Robert Halgren for
their effort in developing the microarray.

I would like to extend my appreciation to Michigan Sea Grant College Program, from
National Sea Grant, NOAA, US. Department of Commerce, and Dissertation
Completion Fellowship from MSU Graduate School.

Last, but not least, I wish to thank my wife, Qianqian, for her understanding and love

during the past few years. Her support and encouragement was also the source of my

vi

strength and motivation. My parents, Jianyun and Ruiyi, receive my deepest gratitude and

love for their dedication and the many years of support through my candidature.

vii

Preface

Native to Russian waters, the zebra mussel (Dreissena polymorpha Pallas 1771)
was ﬁrst introduced to the Great Lakes in the late 19805. The non-native species has
colonized an alarming number of North American inland waterways and lakes (Benson &
Raikow 2009). Zebra mussels reproduce and spread rapidly, leaving behind a trail of
habitat destruction, harmful algal blooms, loss of native species, and economic
devastation. Developing an effective strategy to eradicate zebra mussels, or to at least
control their spread, remains a remote possibility despite concerted efforts by scientists
and managers.

Why are zebra mussels so abundant when most bivalve mollusks populations
worldwide are either declining or at the brink of extinction? One likely reason is the
zebra mussel’s extraordinary ability to attach tightly and expeditiously to any underwater
hard surface including adults of their own species (Johnson & Padilla 1996). Additionally,
optimal environmental conditions, lack of competition, and scarcity of natural predators
or pathogens may have also contributed to the explosive increase in zebra mussel
populations in the Great Lakes.

Zebra mussel invasions have severe impacts on the health of the ecosystem and
the local economy. Industrial facilities devote signiﬁcant efforts to the cleaning of water
intake pipes and heat exchangers clogged with zebra mussels. Water circuits in the
facilities must be cleaned manually, where possible, or with various chemicals in
processes that are both expensive and environmentally harmful (Cope et al. 1997). The
profound economic impact of zebra 'mussel expansion in the Great Lakes basin alone has

mounted billions of dollars in increased operating costs of affected industries and

viii

municipalities over decades (Benson & Raikow 2009). This situation has overwhelmed
North American scientists and water resource managers who unexpectedly became
confronted with an ecologic catastrophe of an unprecedented magnitude. The major
obstacle that impedes immediate intervention or development of an effective
management strategy is the dearth of knowledge on the basic biology of zebra mussels
and the survival strategies used by this successful intruder.

The zebra mussel is unique in many biological properties. For example, zebra
mussels and other dreissenids are the only freshwater bivalves to produce highly
dispersible planktonic veligers and to retain their attachment organ into adulthood
(Morton 1993). These two properties give zebra mussels the capability to spread quickly
and adhere tenaciously in the Great Lakes. The attachment apparatus of the zebra mussel
is an extraorganismic, shock absorbing structure, called the byssus. The molecular
mechanism of the zebra mussel attachment performed by the byssus remains largely
unexplored. To this end, this study attempts to unravel the nature and regulation of zebra
mussel byssus activities at the molecular level. The overarching hypothesis is that
byssogenessis in D. polymorpha is mediated through expression of multiple genes; each
should play a role in a cascade of events. Proteins of some of the expressed genes should
possess a regulatory function during a certain stage of zebra mussel byssal thread
production. The expression of genes involved in byssogenesis of dreissenids can be
altered by a number of environmental factors such as temperature, current rate, and
dissolved oxygen levels.

To test these hypotheses, a number of objectives have to be identiﬁed: 1) To

identify the genes unique to zebra mussel foot and determine if they can be potentially

ix

involved in the activities of zebra mussel byssus; 2) To develop a cDNA microarray
containing all the genes unique to zebra mussel byssus; 3) To ﬁnd out the genes up- or
downregulated in response to the changes of environmental factors, as well as the change
of byssogenesis status; 4) To identify the zebra mussel byssus unique genes with
differential expression proﬁles at the different stages of byssogenesis; and 5) To study the
molecular characteristics, protein activities and potential functions of the genes identiﬁed
by the microarray analyses.

Beginning with the construction of a normalized cDNA library containing the
genes unique to zebra mussel byssus, this study applied a high-throughput technique,
cDNA microarray to identify the gene expression proﬁles during different byssus
activities in large scale. The cDNA library described in chapter 11 provided more than
700 expressed sequence tags (ESTs) that are unique to zebra mussel byssus for the
development of the zebra mussel cDNA library. The putative functions of the proteins
encoded by these ESTs have a broad spectrum such as, foot structural proteins (Dpfp),
exocrine gland peptides (EGP), host defense related, other normal functional proteins,
and the proteins without known homologues in database.

The purpose of the development of cDNA microarray in chapter 111 is to study the
expression patterns of the zebra mussel byssus unique genes during the activities of the
byssus. For example, in this chapter, two different zebra mussel attachment statuses were
created, attachment and non-attachment. By comparing the expression proﬁles of the
genes between these two status, we found seventeen genes were either up or

downregulated in the attachment. It is the ﬁrst time to use microarray technique to study

the zebra mussel attachment in molecular level. Also, the results of this study suggest that
the newly developed zebra mussel microarray is efﬁcient and reliable.

The zebra mussel microarray was also applied in chapter IV and V in order to
identify the genes differentially displayed along the course of byssogenesis (CHAPTER
IV) and those modulated in response to changes in environmental factors (CHAPTER V).
From both of the assays, a number of genes encoding the proteins homolgous to D.
polymorpha foot protein (Dpfp), EGP, and neuopeptide-like proteins (nlp) were identiﬁed.
This suggested that the three families of proteins are very likely to play important roles in
the different activities of zebra mussel byssus and possibly as new foot proteins that have
never been reported before.

Interestingly, a host defense related molecule, D. polymorpha defensin (Dpd), was
found signiﬁcantly expressed by zebra mussel byssus gland cells and also had the
expression proﬁles associated with the production of zebra mussel byssal threads. It is
suggested that beside the basic antimicrobial activities reported in CHAPTER VI, Dpd
may be necessary in the process of attachment by protecting byssal threads against
microbial degradation.

The successful application of cDNA microarray to this study provided a great tool
to understand byssus activities of the zebra mussel. The differentially expressed genes in
response to the change of byssogenesis status, environmental factors and stages of
byssogenesis identiﬁed in this study are of interest in understanding the mechanism of

zebra mussel underwater adhesion.

xi

TABLE OF CONTENTS

LIST OF TABLES ...................................................................................................... xv
LIST OF FIGURES .................................................................................................. xvii
Introduction .................................................................................................................. 1
CHAPTER I .................................................................................................................... 5
Invasion of Zebra Mussels .................................................................................................. 5
Impacts to Economy and Ecology ...................................................................................... 6
Control Methods against Zebra Mussel .............................................................................. 7
Zebra Mussel Byssus .......................................................................................................... 8
Zebra Mussel Byssogenesis and Environmental Factors .................................................. 13
cDNA Library Construction and Suppression Subtractive Hybridization cDNA ............ 15
cDNA Microarray ............................................................................................................. 17
Zebra Mussel Host Defense Mechanism and Byssus Activities ....................................... 20
CHAPTER II ................................................................................................................. 28
Abstract ........................................................................................................................ 28
Introduction ................................................................................................................ 29
Materials and Methods ......................................................................................... 3O
Zebra mussel tissue source, total RNA isolation and mRNA puriﬁcation .................... 30
Construction of the suppression subtractive hybridization cDNA library .................... 31
ESTs sequencing ........................................................................................................... 31
Data Analysis ................................................................................................................ 32
Results .......................................................................................................................... 33
ESTs sequencing and assembling ................................................................................. 33
Homologous gene searching and putative function prediction ..................................... 33
Discussion ................................................................................................................... 36
CHAPTER III ............................................................................................................... 45
Abstract ........................................................................................................................ 45
Introduction ................................................................................................................ 47
Materials and Methods ......................................................................................... 49
Construction of zebra mussel foot cDNA microarray .................................................. 49
cDNA microarray validation ......................................................................................... 50
False discovery rate (FDR) determination .................................................................... 52
Effects of the presence of attachment substrate on gene expression in the foot ........... 52
Microarray data analysis ............................................................................................... 54

xii

Excretory gland peptide (EGP)—like zebra mussel protein amino acid analysis ........... 55

Amino acid analysis of differentially expressed genes without putative function ....... 55
Validation of selected differentially expressed genes ................................................... 55
Results .......................................................................................................................... 57
The tissue speciﬁcity of microarray templates ............................................................. 57
Byssogenesis in group AD and NAD ........................................................................... 57
FDR determination ........................................................................................................ 58
Effects of adhesion status on gene expression .............................................................. 58
EGP-like sequences analysis ......................................................................................... 59
Validation for microarray results .................................................................................. 60
Discussion ................................................................................................................... 61
CHAPTER IV ............................................................................................................... 76
Abstract ........................................................................................................................ 76
Introduction ............................................................. 78
Materials and Methods ......................................................................................... 81
Zebra mussel sample collection and experimental design ............................................ 81
RNA extraction and cDNA synthesis ........................................................................... 82
Microarray hybridization and data analyses ................................................................. 82
Quantitative reverse transcription PCR (qRT-PCR) validation of microarray data ..... 83
Visualization of two differentially expressed genes in foot tissues using RNA
ﬂuorescence in situ hybridization (FISH) ..................................................................... 84
Tissue distribution of representative expressed genes .................................................. 86
Results .......................................................................................................................... 87
Differentially expressed genes during the byssogenesis identiﬁed by microarray ....... 87
Validation of the results obtained from microarray ...................................................... 88
The in situ expression of selected genes within the zebra mussel foot ......................... 89
The relative expression levels of selected genes in zebra mussel tissues ..................... 90
Discussion ................................................................................................................... 91
CHAPTER V ............................................ 107
Abstract ...................................................................................................................... 107
Introduction .............................................................................................................. 109
Materials and Methods ....................................................................................... 112
Zebra mussel collection and maintenance .................................................................. 1 12
Treatments of zebra mussels and experimental design ............................................... 1 12
RNA extraction and cDNA synthesis ......................................................................... 113
Microarray hybridization and data analyses ............................................................... 114
Microarray data validation .......................................................................................... l 15
RNA ﬂuorescence in situ hybridization (FISH) ......................................................... 116
Results ........................................................................................................................ 118
Overview of the factorial analysis results ................................................................... 1 18
Validation .................................................................................................................... 120
RNA in situ hybridization ........................................................................................... 121

xiii

Discussion ................................................................................................................. 123

CHAPTER VI ............................................................................................................. 145
Abstract ...................................................................................................................... 145
Introduction .............................................................................................................. 147
Materials and Methods ....................................................................................... 151
The collection and maintenance of zebra mussels ...................................................... 151
Gene cloning of zebra mussel defensin ...................................................................... 151
Primary and tertiary structure analyses of zebra mussel defensin .............................. 152
The expression of zebra mussel defensin in E. coli in vitro expression system ......... 152
The antimicrobial activity of recombinant Dpd .......................................................... 154
Determination of Dpd expression in zebra mussel tissues .......................................... 156
The expression proﬁles of Dpd in the different statuses of byssogenesis .................. 158
Results ........................................................................................................................ 160
The full length gene sequence and primary protein structure of Dpd ......................... 160
The three dimensional structure of Dpd ...................................................................... 160
The minimal inhibition concentrations (MICs) of recombinant Dpd ......................... 162
The in situ expression of Dpd in zebra mussel ........................................................... 162

Determination of relative expression level of Dpd in zebra mussel tissues by qPCR 163
The expression of Dpd in the zebra mussel foot during the early stages of byssogenesis

..................................................................................................................................... 164
Discussion ................................................................................................................. 165
CHAPTER VII ............................................................................................................ 179
Conclusions .............................................................................................................. 179
Future Studies ......................................................................................................... 182
REFERENCE .............................................................................................................. I 89

xiv

LIST OF TABLES

TABLE 2-1 SUMMARY OF THE ESTS SEQUENCED FROM THE SSH cDNA LIBRARY. ............ 43

TABLE 2-2 THE ESTS WITH HOMOLOGY TO ZEBRA MUSSEL (D. POL YMORPHA) BYSSAL
PROTEIN DPFPI. ......................................................................................................... 43

TABLE 2-3 THE OTHER FOOT PROTEINS WITH PUTATIVE POTENTIAL ATTACHMENT
FUNCTIONS. ................................................................................................................ 44

TABLE 2-4 ESTS AND THEIR HOMOLOGOUS PROTEINS WITH PUTATIVE HOST DEFENSE

FUNCTIONS. ................................................................................................................ 44
TABLE 3-1 THE PRIMERS USED FOR QRT-PCR. ................................................................. 73
TABLE 3-2 THE ESTS WITH UPREGULATED EXPRESSION LEVEL IN AD GROUP ................... 74
TABLE 3-3 THE ESTS WITH DOWNREGULATED EXPRESSION LEVEL IN AD GROUP. ............ 75
TABLE 4-1 THE GENE SPECIFIC PRIMERS USED FOR QRT-PCR. .......................................... 99

TABLE 4-2 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 12-HOUR
REATTACHMENT ....................................................................................................... 100

TABLE 4-3 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER l-DAY
REATTACHMENT ....................................................................................................... 101

TABLE 44 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 2-DAv
REATTACHMENT ....................................................................................................... 102

TABLE 4-5 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 3-DAY
REATTACHMENT ....................................................................................................... I 03

TABLE 4-6 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 4-DAY
REATTACHMENT ....................................................................................................... 1 04

TABLE 4-7 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 7-DAY
REATTACHMENT ....................................................................................................... I 05

TABLE 4-8 THE GENES THAT WERE DIFFERENTIALLY EXPRESSED AFTER 21-DAY
REATTACHMENT ....................................................................................................... 106

TABLE 5-] THE NULL HYPOTHESIS AND THE INTERPRETATION OF TESTED PROBESETS. I37

XV

TABLE 5-2 THE GENE SPECIFIC PRIMERS USED FOR oRT-PCR. ........................................ 138

TABLE 5-3 THE GENES WHOSE EXPRESSION PROFILES HAVE BEEN SIGNIFICANTLY ALTERED
BY THE STATUS OF BYSSOGENESIS ............................................................................ 139

TABLE 5-4 THE GENES WHOSE EXPRESSION PROFILES ARE SIGNIFICANTLY ALTERED DUE TO
THE CHANGE OF TEMPERATURE ................................................................................ 142

TABLE 5-5 THE GENES WHOSE EXPRESSION PROFILES ARE SIGNIFICANTLY ALTERED DUE TO
THE CHANGE OF D.O. ............................................................................................... 143

TABLE 5-6 THE GENES WHOSE EXPRESSION PROFILES ARE SIGNIFICANTLY ALTERED DUE TO
THE CURRENT VELOCITY. ......................................................................................... 144

TABLE 6-1 THE MINIMAL INHIBITION CONCENTRATIONS (MICS) OF DPD AGAINST THE
BACTERIA AND YEAST. ............................................................................................. 178

xvi

LIST OF FIGURES

FIGURE l-I THE ANATOMIC STRUCTURE OF ZEBRA MUSSEL FOOT AND BYSSUS. ................ 23

FIGURE 1-2 THE STRUCTURE OF MYTILID BYSSUS AND DISTRIBUTIONS OF FOOT PROTEINS.

................................................................................................................................... 24
FIGURE 1-3 THE CONSTRUCTION OF SUPPRESSION SUBTRACTIVE HYBRIDIZATION CDNA

LIBRARY ..................................................................................................................... 25
FIGURE 1-4 THE ANATOMIC STRUCTURE OF ZEBRA MUSSEL FOOT AND RETRACTOR

MUSCLES. ................................................................................................................... 26

FIGURE 1-5 THE NON-REFERENCE INTERWOVEN LOOP DESIGN WITH DIFFERENT NUMBERS OF

CONDITIONS AND REPLICATES. ................................................................................... 27
FIGURE 2-1 THE STRUCTURE OF ZEBRA MUSSEL FOOT AND RETRACTOR MUSCLES. ............ 39
FIGURE 2-2 PERCENTAGE OF ESTS WITH DIFFERENT POSSIBLE FUNCTIONS ........................ 40

FIGURE 2-3 THE ANNOTATION OF PROTEINS ENCODED BY ESTS FROM CDNA LIBRARY. 41

FIGURE 2- 4 THE COMBINED GRAPHIC 0F ANNOTATION OF THE ZEBRA MUSSEL BYSSUS

UNIQUE GENES. .......................................................................................................... 42
FIGURE 3-1 ZEBRA MUSSEL BYSSUS CDNA MICROARRAY. ................................................ 66
FIGURE 3-2 QRT-PCR WITH SELECTED GENES USING DIFFERENT TISSUES. ........................ 67
FIGURE 3-3 NUMBER OF REGROWN BYSSAL THREADS OVER 16 DAYS. ............................... 68

FIGURE 3-4 HISTOGRAM OF DISTRIBUTION OF THE NUMBER OF DIFFERENTIALLY EXPRESSED
GENES AND FOLD-CHANGE. ........................................................................................ 69

FIGURE 3-5 NEIGHBOR-JOINING PHYLOGENETIC ANALYSIS WITH EXCRETORY GLAND

PEPTIDE-LIKE MOLECULES .......................................................................................... 70
FIGURE 3-6 MULTIPLE ALIGNMENTS WITH THE SEVEN EGP-LIKE SEQUENCES. .................. 71
FIGURE 3-7 VALIDATION OF MICROARRAY RESULTS WITH QRT-PCR ................................ 72

FIGURE 4-1 THE HIERARCHICAL CLUSTER WITH THE MICROARRAY IDENTIFIED GENES. 95

xvii

FIGURE 4-2 THE COMPARISONS OF GENE EXPRESSION PROFILES BETWEEN THE MICROARRAY
RESULTS AND QRT-PCR. ........................................................................................... 96

FIGURE 4-3 THE [N SITU EXPRESSION OF THE GENE BG 1 5_F03-DPFP AND BG 1 6__HOS-EGP
IN ZEBRA MUSSEL FOOT TISSUE. ................................................................................. 97

FIGURE 4-4 THE DISTRIBUTION OF THE MRNA PRODUCTS OF THE SELECTED GENES WITHIN
ZEBRA MUSSEL TISSUES .............................................................................................. 98

FIGURE 5-1 THE NON-REFERENCE INTERWOVEN LOOP DESIGN FOR MICROARRAY ANALYSIS.

................................................................................................................................. 128
FIGURE 5-2 THE HIERARCHIAL CLUSTER OF MICROARRAY IDENTIFIED GENES. ................ 129
FIGURE 5-3 THE NUMBERS OF GENES WHOSE EXPRESSION PROFILES ARE ALTERED BY

SINGLE FACTOR OR MULTIPLE FACTORS. .................................................................. 130

FIGURE 5-4 MULTIPLE ALIGNMENT WITH THE IDENTIFIED EGP ENCODING GENES WHOSE
EXPRESSION PROFILES CAN BE ALTERED BY BYSSOGENIC ACTIVITY. ........................ I31

FIGURE 5-5 COMPARISON OF PRIMERAY STRUCTURES OF THE EGP ENCODING GENES WHOSE
EXPRESSION PROFILES CAN BE ALTERED BY WATER TEMPERATURES ........................ 132

FIGURE 5-6 THE DISTRIBUTION OF THE MRNA PRODUCTS OF THE SELECTED GENES WITHIN
ZEBRA MUSSEL TISSUES ............................................................................................ 133

FIGURE 5-7 THE QRT-PCR RESULTS DEMONSTRATED THE RELATIVELY EXPRESSION LEVELS
OF THE GENE DURING THE BYSSIOGENESIS. .............................................................. 134

FIGURE 5-8 THE DISTRIBUTION OF THE ZEBRA MUSSEL BYSSUS GLAND CELLS IN MUSSEL
FOOT. ....................................................................................................................... 135

FIGURE 5-9 THE [N SITU EXPRESSION OF DPFP-BGzo_A01 AND EGP—BG97/192_BO6 IN
FOOT. ....................................................................................................................... 136

FIGURE 6-1 PHYLOGENETIC ANALYSES OF ZEBRA MUSSEL DEFENSIN (DPD) AND THE
HOMOLOGUES. ......................................................................................................... 168

FIGURE 6-2 THE MULTIPLE ALIGNMENT WITH THE HOMOLOGOUS INSECT AND MOLLUSK

DEFENSINS. .............................................................................................................. 169
FIGURE 6-3 THE TERTIARY STRUCTURES OF DPD AND DISULFIDE BONDS ......................... 170
FIGURE 6-4 THE CONSERVED AMINO ACID RESIDUES IN DPD. . ......................................... 171
FIGURE 6- 5 FISH RESULT OF DPD WITHIN ZEBRA MUSSEL FOOT. .................................... 172

xviii

FIGURE 6-6 NEGATIVE CONTROL FOR FISH RESULTS OF DPD IN ZEBRA MUSSEL FOOT
TISSUE. ..................................................................................................................... 173

FIGURE 6-7 FISH RESULTS OF DPD IN OTHER TISSUES OF THE ZEBRA MUSSEL. ................ 174

FIGURE 6-8 THE NEGATIVE CONTROL OF FISH DPD IN OTHER TISSUES OF THE ZEBRA
MUSSEL. ................................................................................................................... 175

FIGURE 6- 9 THE RELATIVE ABUNDANCE OF DPD IN DIFFERENT ZEBRA MUSSEL TISSUES
WITH QPCR .............................................................................................................. 176

FIGURE 6-10 THE RELATIVE EXPRESSION LEVELS OF DPD UNDER THE CONDITION OF NON-
BYSSOGENESIS AND BYSSOGENESIS .......................................................................... 177

Xix

Introduction

Native to Russian waters, the zebra mussel (Dreissena polymorpha Pallas 1771)
was ﬁrst introduced to the Great Lakes in the late 19803. The non-native species has
colonized an alarming number of North American inland waterways and lakes (Benson &
Raikow 2009). Zebra mussels reproduce and spread rapidly, leaving behind a trail of
habitat destruction, harmﬁil algal blooms, loss of native species, and economic
devastation. Developing an effective strategy to eradicate zebra mussels, or to at least
control their spread, remains a remote possibility despite concerted efforts by scientists
and managers.

Why are zebra mussels so abundant when most bivalve mollusks populations
worldwide are either declining or at the brink of extinction? One likely reason is the
zebra mussel’s extraordinary ability to attach tightly and expeditiously to any underwater
hard surface including adults of their own species (Johnson & Padilla 1996). Additionally,
optimal environmental conditions, lack of competition, and scarcity of natural predators
or pathogens may have also contributed to the explosive increase in zebra mussel
populations in the Great Lakes.

Zebra mussel invasions have severe impacts on the health of the ecosystem and
the local economy. Industrial facilities devote signiﬁcant efforts to the cleaning of water
intake pipes and heat exchangers clogged with zebra mussels. Water circuits in the
facilities must be cleaned manually, where possible, or with various chemicals in
processes that are both expensive and environmentally harmful (Cope et al. 1997). The

profound economic impact of zebra mussel expansion in the Great Lakes basin alone has

mounted billions of dollars in increased Operating costs of affected industries and
municipalities over decades (Benson & Raikow 2009). This situation has overwhelmed
North American scientists and water resource managers who unexpectedly became
confronted with an ecologic catastrophe of an unprecedented magnitude. The major
Obstacle that impedes immediate intervention or development of an effective
management strategy is the dearth of knowledge on the basic biology of zebra mussels
and the survival strategies used by this successful intruder.

The zebra mussel is unique in many biological properties. For example, zebra
mussels and other dreissenids are the only freshwater bivalves to produce highly
dispersible planktonic veligers and to retain their attachment organ into adulthood
(Morton 1993). These two properties give zebra mussels the capability to spread quickly
and adhere tenaciously in the Great Lakes. The attachment apparatus of the zebra mussel
is an extraorganismic, Shock absorbing structure, called the byssus. The molecular
mechanism of the zebra mussel attachment performed by the byssus remains largely
unexplored. To this end, this study attempts to unravel the nature and regulation of zebra
mussel byssus activities at the molecular level. The overarching hypothesis is that
byssogenessis in D. polymorpha is mediated through expression of multiple genes; each
should play a role in a cascade of events. Proteins of some of the expressed genes should
possess a regulatory function during a certain stage of zebra mussel byssal thread
production. The expression of genes involved in byssogenesis of dreissenids can be
altered by a number of environmental factors such as temperature, current rate, and

dissolved oxygen levels.

To test these hypotheses, a number of objectives have to be identiﬁed: 1) To
identify the genes unique to zebra mussel foot and determine if they can be potentially
involved in the activities of zebra mussel byssus; 2) TO develop a cDNA microarray
containing all the genes unique to zebra mussel byssus; 3) To ﬁnd out the genes up- or
downregulated in response to the changes of environmental factors, as well as the change
of byssogenesis status; 4) To identify the zebra mussel byssus unique genes with
differential expression proﬁles at the different stages of byssogenesis; and 5) To study the
molecular characteristics, protein activities and potential functions of the genes identiﬁed
by the microarray analyses.

Beginning with the construction of a normalized cDNA library containing the
genes unique to zebra mussel byssus, this study applied a high-throughput technique,
cDNA microarray to identify the gene expression proﬁles during different byssus
activities in large scale. The cDNA library described in chapter 11 provided more than
700 expressed sequence tags (ESTS) that are unique to zebra mussel byssus for the
development of the zebra mussel cDNA library. The putative functions of the proteins
encoded by these ESTS have a broad spectrum such as, foot structural proteins (Dpfp),
exocrine gland peptides (EGP), host defense related, other normal functional proteins,
and the proteins without known homologues in database.

The purpose of the development of cDNA microarray in chapter 111 is to study the
expression patterns of the zebra mussel byssus unique genes during the activities of the
byssus. For example, in this chapter, two different zebra mussel attachment statuses were
created, attachment and non-attachment. By comparing the expression proﬁles of the

genes between these two status, we found seventeen genes were either up or

downregulated in the attachment. It is the ﬁrst time to use microarray technique to study
the zebra mussel attachment in molecular level. Also, the results of this study suggest that
the newly developed zebra mussel microarray is efﬁcient and reliable.

The zebra mussel microarray was also applied in chapter IV and V in order to
identify the genes differentially displayed along the course of byssogenesis (CHAPTER
IV) and those modulated in response to changes in environmental factors (CHAPTER V).
From both of the assays, a number of genes encoding the proteins homolgous to D.
polymorpha foot protein (Dpfp), EGP, and neuopeptide-like proteins (nlp) were identiﬁed.
This suggested that the three families of proteins are very likely to play important roles in
the different activities of zebra mussel byssus and possibly as new foot proteins that have
never been reported before.‘

Interestingly, a host defense related molecule, D. polymorpha defensin (Dpd), was
found signiﬁcantly expressed by zebra mussel byssus gland cells and also had the
expression proﬁles associated with the production of zebra mussel byssal threads. It is
suggested that beside the basic antimicrobial activities reported in CHAPTER VI, Dpd
may be necessary in the process of attachment by protecting byssal threads against
microbial degradation.

The successful application of cDNA microarray to this study provided a great tool
to understand byssus activities of the zebra mussel. The differentially expressed genes in
response to the change of byssogenesis status, environmental factors and stages of
byssogenesis identiﬁed in this study are of interest in understanding the mechanism of

zebra mussel underwater adhesion.

CHAPTER I

Literature Review

Invasion of Zebra Mussels

In 1988, disturbing reports announced that zebra mussels (Dreissena polymorpha)
were found in Lake St. Clair; a water body connecting Lake Huron and Lake Erie. This
non-native species found a refuge in the Laurentian Great Lakes watershed and has
successfully invaded the majority of the water shed (Hebert et al. 1989). By 1990, zebra
mussels had been found in all the Great Lakes. Zebra mussels escaped the Great Lakes
basin and found their way into the Illinois and Hudson rivers in 1991. Invasion into the
Illinois River led to their introduction into the Mississippi River drainage system which
covers over 1.2 million square miles. By 1992, populations of zebra mussels were
established in the following rivers: Arkansas, Cumberland, Hudson, Illinois, Mississippi,
Ohio, and Tennessee. Zebra mussels continued their Spread across the continental U. S.
and by 1994 were reported in water bodies of the following states: Alabama, Arkansas,
Illinois, Indiana, Iowa, Kentucky, Louisiana, Michigan, Minnesota, Mississippi, Missouri,
New York, Ohio, Oklahoma, Pennsylvania, Tennessee, Utah, Vermont, West Virginia,
and Wisconsin (Benson & Raikow 2009). More recently, Connecticut (2002), Nebraska
(2003), and South Dakota (2003) have been listed as additional states where zebra
mussels were found (Benson & Raikow 2009). The zebra mussel has continued its spread
west, south, and east; in January 2008, zebra mussels reached the southwestern United

States, in the San Justo Reservoir in central California (Benson & Raikow 2009).

Currently in june 2009, zebra mussels were found in Lake Texoma on the border of
Texas and Oklahoma (Benson & Raikow 2009). It is believed that the zebra mussel was
ﬁrst introduced to North America through the ballast exchange of a commercial cargo
ship traveling from the north shore of the Black Sea to the Great Lakes of North America
(McMahon 1996). This rapid range expansion by the zebra mussel was accomplished not
only by the passive driﬁing of the larva, but also by its ability to attach to boats
navigating lakes and rivers (Benson & Raikow 2009). An adult zebra mussel under cool
and humid conditions stays alive for several days when taken out of water. That provides
the mussel enough time to be transmitted to inland lakes and rivers through trailers and

vehicles (Benson & Raikow 2009).

Impacts to Economy and Ecology

The underwater attachment of the zebra mussel has made this invasive species
notorious for its biofouling capabilities. Zebra mussels colonize in underwater
constructions, such as water supply pipes of hydroelectric and nuclear power plants,
public water supply plants, and industrial facilities. Moreover, zebra mussels also cause
deterioration and corrosion of underwater structures such as navigational buoys, ﬁshing
gears, boats, and docks (Benson & Raikow 2009). The impacts of the zebra mussel to the
ecological system are even more profound. The huge consumption of phytoplankton, as
well as other suspended materials including bacteria, protozoa, microzooplankton, and
silt, has seriously reduced the biomass of local aquatic systems. The increased water
clarity caused by the decrease of biomass allows light to penetrate further to the deep
level of the water body which potentially promotes macrophyte populations (Skubinna et

al. 1995). As most of the phytoplankton are consumed by the zebra mussel, the

6

concentration of dissolved organic carbon (DOC) can drop to an extremely low level;
which has been reported in some inland lakes colonized by zebra mussels (Benson &
Raikow 2009). The negative effects of zebra mussels on the ecosystem have induced
major disturbances in the foodweb. The reductions of phytoplankton and zooplankton
biomass cause huge impacts to ﬁsh populations, such as the increase of competition,
decrease of survivals and the extinction of planktivorous ﬁsh. Since the
microzooplanktons are more heavily affected by zebra mussels, ﬁsh larvae, which feed
on the microzooplanktons may be more greatly affected than other later life stages
(Raikow 2004). Moreover, native unionids were also affected by the zebra mussel
invasion (Schloesser et al. 1996; Baker & Hombach 1997). By simply attaching to the
shells of unionids, the zebra mussels restrict valve operation, cause shell deformity,
smother siphons, compete for food, impair movement, and deposit metabolic waste onto
unionid clams. According to a previous study, the survival rates of native unionid
mussels in the Mississippi River, Minnesota signiﬁcantly declined with the zebra mussel
colonization (Hart et al. 2001). The unionids in Lake St. Clair and western Lake Erie, to

date, have been extirpated (Benson & Raikow 2009).

Control Methods against Zebra Mussel

The profound economic and ecological impact of zebra mussel expansion in the
Great Lakes basin has mounted to over billions of dollars in increased operating costs of
affected industries and municipalities (Roberts 1990). A variety of control methods have
been developed against the attachment and distribution of zebra mussels. Physically,
zebra mussels can be detached by exoteric force. Many types of equipment have been

invented to remove the attached zebra mussels from underwater substrates, such as high

7

pressure water hoses and acoustical vibration. Alternatively, a number of chemical
molluscicides including chlorine, chlorine dioxide, and other oxidizing and non—oxidizing
chemicals have been applied in some areas to eliminate the zebra mussels. Some other
physical changes of the environmental conditions may also lead to the mortalities or
detachment of zebra mussels. Such physical elimination methods include
dewatering/desiccation, thermal controlling, electrical current, C02 injection, and
ultraviolet light. Instead of detaching the zebra mussels from the substrates, people also
apply Special coating materials on the surface of the structures, such as copper, zinc, and
silicone-based paint (Benson & Raikow 2009). However, all the strategies mentioned
above are either not sufﬁcient or not environment-friendly. Even the most recently
introduced biological control, such as the introduction of natural mussel predators (Bially
& Maclsaac 2000), selective toxic microbes and parasites (Molloy 1998), and disrupting
reproductive process (Snyder et al. 1997), have been proven to be ineffective. The
ultimate goal of these control methods is to disrupt or prevent attachment of zebra
mussels. However, the major obstacle that impedes immediate intervention or
development of an effective management strategy is the dearth of knowledge on

attachment mechanism employed by this successful intruder.

Zebra Mussel Byssus

The attachment organ of the zebra mussel is an extraorganismic, shock absorbing
structure, called the byssus, which was ﬁrst described by van der F een in 1949 (Waite
2002). Except for the morphological characteristics, the biochemical nature of the byssus
and processes involved in zebra mussel’s adhesion to substrates remain largely

unexplored. The byssus apparatus is considered to be a masterpiece of underwater

8

bioadhesion because of the rapid and robust production, as well as the relatively simple
components of adhesive proteins (Waite 2002). The silver hair-like structures that can be
seen out of shells in most natural conditions are called byssal threads. All byssal threads
are generated from the stem located at the root of the foot. The enlarged plaque end of the
byssal thread is the structure holding all the adhesive proteins at the interface of the
byssal thread and substrata surface. In the zebra mussel, the byssus also includes three
main glands located in the foot, namely, the stem-forming gland, thread-forming land,
and plaque-forming gland. The proteins needed for byssal thread formation and adhesion
are secreted by the glands and expelled to the groove of the ventral side in the foot and
transported out of the organ (Figure 1-1). The anatomic structure of the zebra mussel
byssus and the production of the adhesive proteins have been well described by Rzepecki
and Waite (Rzepecki & Waite 1993a, b).

Two distinct types of byssal threads are produced by zebra mussels. One of them
is the temporary byssal thread which possesses extraordinary plasticity. It is often seen in
the juvenile stage and early stage of attachment of adult zebra mussels. This type of
byssal thread helps to explore the precipitous surfaces of underwater substrates.
Morphologically, the temporary threads are anchored by an elastic, mucous ﬁlament
which is compositionally distinct from the threads and plaques (Rzepecki & Waite
1993a). The other type of byssal threads are permanent byssal threads, which appears
mostly at the adult stage of the zebra mussel and help mussels for longer term settlement.
There are two distinct sections observed in the permanent byssal threads, the proximal
portion that is close to the stem of byssal threads, and the distal portion which is close to

the attachment plaque of the thread. Observations of the exterior surface characteristics of

the byssal threads show that the surface of proximal region is relatively smooth, whereas
the surface is rough at the distal portion (Eckroat et al. 1993).

Once the zebra mussels are relocated, the majority of byssal threads are formed
within one week to ensure a rapid, strong attachment. Thereafter, the rate of byssal thread
formation remains constant (Eckroat et al. 1993). As a zebra mussel prepares to produce
byssal threads, the foot is moved along the surface of the substrate searching for suitable
attachment site. At the same time, the space between the zebra mussel foot and the
substrate is ﬁlled with mucus provided by the ventral groove. The mucus does not
resemble the byssal threads, but is there to create a hydrophobic and clean environment
for the formation of the byssal threads. The actual protein components are produced by
the three major byssus glands which have been described above. The ﬁrst formed section
of byssal threads is the plaque; thereaﬁer, the foot Slowly retracts back into the shell
while the clear zebra mussel byssal threads start becoming visible (Eckroat et al. 1993).
Three main protein components have been reported to be involved in the formation of
zebra mussel byssal threads, namely, D. polymorpha foot protein (Dpfp) -l, -2, and -3
(Rzepecki & Waite 1993b, a). However, only two of them, Dpfp-l and -2 have been
sufﬁciently puriﬁed for characterization. Preliminary studies on Dpfp-l and -2 suggested
that both of these two foot proteins may serve as the component maintaining byssal
structures, rather than adhesive proteins (Anderson & Waite 2000). Dpfp-l and Dpfp-2
contain a number of 3,4-dihydroxyphenylalanines (DOPAs) in their primary sequence.
The DOPA proteins are tandemly repetitive with unique oligopeptide motif sequences
(Rzepecki & Waite 1993b). Dpfp-l was found to have at least 10-15 variants which

constituted a polymorphic protein family speciﬁc to the zebra mussel (Rzepecki & Waite

10

1993b). Crosslinks formed by the Dpfp-l variants along with DOPA create a dovetail
frame, which is the primary component to maintain the byssal thread structures.
Meanwhile, DOPA also participates in protein crosslinking during maturation of byssal
threads, following oxidation to DOPA quinone. These crosslinks will structurally
reinforce the weakest components of the byssus.
Similar structures have been observed in many marine byssus-carrying species, such as
Mytz'lus sp. (Brown 1952; Price 1983; Waite 1983a, 1992). At least 12 proteins have been
identiﬁed from the byssus of Mytilus species and eight of them are characterized as
adhesive proteins located in the plaque. M edulis foot protein-l and -2 (pr-l and pr-2)
are the two most well studied foot proteins that can be easily puriﬁed from the M. edulis
foot. The pr-l, which constitutes about 5% of the plaque matrix proteins, is a molecule
to mediate bonding to the intended substratum during attachment (Benedict & Waite
1986; Waite et al. 1998; Sun & Waite 2005; Waite et al. 2005), while the pr-2
performs as a structural component comprising 25% of the plaque content (Papov er a1.
1995). Also pr-l covers all the exposed portion of the byssus to provide protection
from the environment (Lin et al. 2007). pr-3 is the real adhesion proteins acting as glue
in the plaque proteins (Lin et al. 2007). pr-4, with high levels of histidine, lysine and
arginine in the primary structure, is proposed to be the protein that serves as a coupling
agent in the thread plaque junction (Cha et al. 2008). pr-S is a plaque speciﬁc protein
with the highest 3,4—dihydroxyphenyl-L-alanine (DOPA) level (~30 mol.%) among all
the foot proteins (Waite & Qin 2001).

Previous studies demonstrate that mussel adhesion largely depends on DOPA

(Deming 1999; Waite 2002), the higher the DOPA content, the stronger the adhesion (Yu

11

& Deming 1998; Lee et al. 2006). pr-3 and pr-S are two adhesive molecules that
exist in the contact area between plaques and solid surfaces in the water (Waite 2002;
Zhao et a1. 2006; Zhao & Waite 2006). The main component of the byssal threads are
two collagens, preCol-D and —NG. They dominate from the ﬁbrous core of byssal threads
(Waite et al. 1998) to the plaque matrix with their frayed ends (Waite et al. 2005).
Recently, another plaque protein, pr-6, was identiﬁed from Mytilus californianus,
which is a close relative to M edulis. pr-6 is a thiol-rich protein that has potential
ﬁinction of mediating the cysteinyldopa cross-link between the adhesive proteins and the
solid surface (Zhao & Waite 2006). The distributions of these foot proteins of Mytulids
are shown in Figure l-2.

Details about byssal thread formation also mostly comes from M eulis, a marine
byssate species. The medium for habitat of water mussels, perhaps poses some challenges
to adhesion (Cayless 1991). Interactions between the adhesive proteins in the plaque and
solid surfaces in the water are mostly noncovalent including charge-charge, hydrogen
bond, and van der Waals forces which contain dipole-dipole, induced dipole-dipole, and
nonpolar coupling interactions. However, the resistance of adhesion formation with the
noncovalent bonds in water is about 80 times stronger than that in the air. If the mussel
attaches to a hard surface in water, the individual should be able to remove weak
boundary layers including water, dirt, microbes, etc, from the surface. Moreover, the
mussel has to keep noncovalent bonds of adhesion from being subverted by water all
around the plaques. In fact, adhesion of the mussels relies on different interfacial
interactions, and can happen on a wide range of underwater materials. That suggests that

the mussel is able to recognize and respond to different kinds of surfaces (Waite 2002).

12

Information about the mechanism of byssal thread formation is very limited according to
the previous studies. In M edulis, the byssal proteins are synthesized by byssus glands
and stocked in the foot. During thread formation, granular thread precursors are secreted
by the glands and collected in the ventral groove in the foot. With the contraction of the
foot, the secreted precursors are mixed and shaped, thereafter loaded onto the surface of
the substratum. The maturation of byssal threads is very similar to formation of
intermolecular cross-links derived from oxidized DOPA (McDowell et al. 1999).
Although the Dfp and pr have some similarities in their characteristics, the
classiﬁcation of dreissenids indicates that adhesion mechanisms employed by zebra and
quagga mussels will be different than that of blue mussels. The Dreissena is classiﬁed in
the superorder Veneroida with a non-byssate North American invader, Corbicula
ﬂuminea; while the mytilids belong to the superorder Pterioida (Allen 1985). This
suggests that the dreissenid byssus evolved independently of that in mytilids (Morton
1993). Therefore, there may be other adhesive proteins that haven’t been isolated or

regulation mechanisms of the two byssate mussels are completely different.

Zebra Mussel Byssogenesis and Environmental Factors

In adult stages of the zebra mussel, a number of environmental factors have been
reported to inﬂuence byssogenesis, or the rate of production of new byssal threads.
Clarke and McMahon designed a series of experiments to test the effects of physical
factors on byssogenesis (Clarke & McMahon 1996b, 0; Clarke & McMahon 1996a;
Clarke 1999). The byssogenesis rate of zebra mussels at the ﬁrst week was boosted with
an increase of temperature, within the range of 5 — 30 °C (Clarke & McMahon 1996c).

Maximum byssal thread generation rate appeared when the current velocity was 0.2 m/s.

13

Byssogenesis rate was decreased with current Speeds higher or lower than 0.2 m/s (Clarke
& McMahon 1996a). With dissolved oxygen level at 30.9 torr, the zebra mussels had
maximum byssogenesis rate while low dissolved oxygen level led to hypoxia, which
caused a signiﬁcant decrease of byssogenesis rate, as well as mussel mortalities. The
byssogenesis rate decreased when dissolved oxygen levels increased to a very high level
at 154.3 torr (Clarke & McMahon 1996b). Interestingly, the byssogenesis rate is also
higher when more food is available (Clarke 1999). These results indicate that, at the
molecular level, expression levels of genes involved in byssogenesis regulation can be
affected by changes in environmental factors. The response to the changes in
environment may be controlled by different byssogenesis genes. What are those genes
involved in the byssogenesis and how do these genes regulate the byssogenesis to
accommodate the environment are two very interesting questions about the byssogenesis
mechanism.

In fact, the zebra mussel underwater adhesion mechanism does not only include
the byssal thread structural proteins and plaque adhesive proteins, but also combines
many other regulation processes including protein release and modiﬁcation, byssus
protection, and even signal transduction. The study of individual molecules has helped us
learn more about the characteristics of the byssus components; however, it did not give
the answer to the questions about the molecular regulation mechanism of zebra mussel
attachment. At this point, a new strategy of the zebra mussel adhesion mechanism study

is necessary to solve this complex puzzle.

14

cDNA Library Construction and Suppression Subtractive

Hybridization cDNA

Given that a number of proteins participate in the regulation Of zebra mussel
attachment, conventional methods focusing on a few zebra mussel proteins may not be
efﬁcient enough to understand the complicated mechanism employed by zebra mussels
during attachment. Therefore, we opted to employ assays from which we can obtain
accurate information on a relatively large number of molecules of relevance to the
attachment process. Most high throughput approaches in the ﬁeld of functional genomics
start with the construction of a complimentary DNA (cDNA) library to messenger RNA
(mRN A) of expressed genes at the time of collection. The mRNA collected from target
cells is reverse-transcribed to stable complementary DNA. One potential beneﬁt to this
approach is that cDNA libraries lack information about enhancers, introns, and other
regulatory elements found in a genomic DNA library.

Although the cDNA library contains much less redundant sequence information
compared to a genomic DNA library, the abundance of each gene in the library is
tremendously different than the others. A typical eukaryote somatic cell contains three
different classes of mRNAs based on their frequencies (Bishop et al. 1974; Davidson &
Britten 1979). The frequencies of highest and lowest are 40-45% and 10% respectively.
On the average, in a single cell there are about 10 most prevalent mRNA species and
each of them is represented by 5,000 copies. On the other hand, the class of high
complexity category includes about 15,000 mRN A species, with each of them
represented by only 1-15 copies (Soares et al. 1994). Hence, the identiﬁcation of the rare

mRNA species in regular cDNA libraries is extremely difﬁcult compared to mRNAs with

15

high frequencies in cells. It is therefore highly recommended to normalize the cDNA
library so that each expressed gene is represented at similar frequency in the cDNA
library.

Suppression subtractive hybridization (SSH) is a simple and efﬁcient method for
generating cDNAs highly enriched for differentially expressed genes of both high and
low abundance (Diatchenko et al. 1996; Diatchenko et al. 1999; Wang & F euerstein 2000;
J i et al. 2002). A high level of enrichment of rare transcripts has been reported from a
cDNA library normalized by SSH; a 1,000- to 1,500-fold enrichment for rare cDNAs
(Diatchenko et a1. 1996). SSH has been successfully applied in studies on tissues (Li et al.
2000; Villalva et al. 2001) and cell lines (Eleveld—Trancikova et al. 2002; Langley et al.
2003)

In SSH, cDNAs synthesized from mRNAs of target tissues or cells are used as
tester cDNA while the cDNAs from the mRN A isolated from control tissues or cells are
drivers (Figure 1-3). The objective of SSH is to reduce cDNAs that are not unique to
tester tissues or cells by subtracting cDNAs abundant in both testers and drivers from the
tester cDNAs. Brieﬂy, the tester cDNAs are divided into two groups, tester 1 and tester 2,
and connected with two different DNA adapters, namely, adapter 1 and adapter 2
respectively. An excessive amount of driver cDNAs without any adapters is used to
hybridize with tester 1 and tester 2, respectively. The hybridized tester 1 and tester 2 are
then hybridized to each other. Thereafter, the cDNA pool contains a variety of
hybridization products and is used for PCR with a pair of primers designed based on the

sequence of adapter 1 and adapter 2. The only type of cDNA strand that can be ampliﬁed

l6

by PCR is the one with different adapters connected at the 5’ and 3’ ends; this cDNA
strand is also unique to tester tissues or cells (Figure 1-3).

Anatomic studies of the zebra mussel foot indicated that other than muscular
tissue and epithelial covering, there are a number of byssus glands responsible for the
development and structure of byssal threads and end plaques. If the cDNAs synthesized
from the foot mRNA are used as a tester, the cDNAs abundant in the muscle cells can be
greatly subtracted with the cDNAs from adductor or retractor mussels of the zebra mussel
(Figure 1-4). The enriched cDNAs in the tester are likely to reﬂect the composition of

mRNA in zebra mussel byssus glands.

cDNA Microarray

The normalized cDNA library will provide a number of genes unique to the zebra
mussel byssus; however, identiﬁcation of expression proﬁles and the potential functions
involved in zebra byssus activities requires another technique to analyze gene expression
on a large scale. Evolved from Northen blotting techniques, the DNA microarray
technique was ﬁrst applied and described as a high throughput method to identify genes
whose expression is modulated by interferon (Kulesh et al. 1987). With further
development and improvement during the past two decades, the DNA microarray
technique has been applied to many ﬁelds, such as discoveries of the exons and genes for
annotation of human genome draft sequence (Shoemaker et al. 2001), the analysis of
genomic DNA for detection of ampliﬁcations and deletions in tumors (Hodgson et a1.
2001; Fritz et al. 2002), and the differential gene expression analysis of the networks of

genes within common pathways of regulation (Zhu et al. 2000; Miki et al. 2001).

17

Two main types of DNA microarray have been developed, spotted DNA
microarrays and in situ synthesized Oligonucleotide arrays (Afﬁmetrix) (Dufva 2009).
Although the Affymetrix microarray is more advanced in many aspects, including the
lower background of hybridization, higher speciﬁcity, and more accessibility for
hybridization, the spotted DNA microarray is currently widely used because it can be
produced in academic laboratories with affordable prices (Bowtell & Sambrook 2003).

In the cDNA microarrays, singled strand DNAS obtained from PCR are Spotted and ﬁxed
in a glass Slide. Separately labeled with different ﬂuorescent dyes (usually green and red),
two cDNA samples from different cells populations or tissues are mixed and then
hybridized Simultaneously with one microarray. These two cDNA probes competitively
anneal to their complementary nucleotide sequences on the microarray slide and the rest
of the probes that cannot be anchored by the DNA strands on the Slides will be washed
away. The expression levels of each probe from each sample can be detected by
capturing the color and strength of the signals in each spot on the microarray. The
intensity ratio at a spot is thus a measure of the relative abundance of the gene in the two
samples (Nguyen et al. 2002; Stears et al. 2003).

In the present study, the use of the cDNA microarray has enabled us to understand
the genetic regulation of the byssus unique genes involved in the process of byssogenesis,
from the beginning of byssal threads being produced, until they mature within 21 days.
As discussed above, byssogenesis can be divided into a number of stages such as the
secretion of byssal proteins, the generation of temporary byssal threads, the formation of
permanent threads, and structural modiﬁcations of the permanent threads. These steps

involve various proteins, and therefore, a time-course microarray, as applied in this study,

18

proved helpful in the identiﬁcation of the genes differentially regulated over the course of
byssogenesis. The time-series analyses of microarray have been applied Since the
invention of the cDNA microarray technology. For example, DeRisi et al in 1997
successfully applied a time-course analysis on the study of the gene expression patterns

in yeast during metabolic shift from fermentation to respiration (DeRisi et al. 1997). A
similar study was performed by Chu et al in 1998 also with time-course microarray
design (Chu et al. 1998).

The effects of the surrounding environmental factors on byssogenesis are complex
to study, whether single or combined. Therefore, the non-reference factorial design of
microarray analysis has been used in this study, based on the recommendations of a
number of statisticians (Draghici et al. 2001; Churchill 2002; Kerr 2003). A particularly
popular non-reference factorial design, the loop design (Kerr & Churchill 2001), involves
at least two different arrangements of biological replicates (Dobbin et al. 2003). For an
experiment dealing with the effects of more than two factors, the interwoven loop design
(Figure 1—5) with more than one loop, is statistically more efﬁcient than a common
reference design, as it leads to less disparity in precision and power comparisons between
any two treatments (Tempelman 2005). Indeed this approach allows the researcher to
obtain more information from the microarray experimental design including multiple

environmental factors with least number of hybridizations.

19

Zebra Mussel Host Defense Mechanism and Byssus
Activities

Along with the generation of byssal threads, other proteins are probably being
secreted along with the byssal threads. The highly proteinaceous components of fresh
byssal threads are likely to be degraded by microbial communities, if there were no
efﬁcient mechanisms employed by the zebra mussels to protect their threads. On the
other hand, the bioﬁlms existing on the surface of underwater substances form a barrier
between the substrates and adhesive proteins at the ends of the byssal threads. Some
bioﬁlms have negative effects on adhesion of both larval (Kavouras & Maki 2003a) and
adult stages (Kavouras & Maki 2003b, 2004; Angarano et al. 2009) of zebra mussel
byssus.

Like other bivalve mollusks, the host defense system employed by the zebra
mussel consists of both cellular and humoral components. The cellular defense iS
primarily performed by hemocytes, which are found in a semiclosed circulation system,
including the heart, vessels, and sinus with different sizes located in major organs
(Auffret 2005). Phagocytosis by hemocytes represents the main process of cellular
defense and goes through a number of phases that encompass recognition, adhesion,
ingestion, destruction and elimination of foreign cells (Pipe 1990, 1992; Tiscar & Mosca
2004). There are different methods for hemocytes to destroy the pathogens, such as using
lysosomal enzymes and respiratory burst. Similar to mammalian phagocytes, the
respiratory burst of bivalve hemocytes can be induced by certain stimuli. A number of
reactive oxygen species (ROS) will be produced during the respiratory burst activity,

such as superoxide, the hydroxyl radical, singlet oxygen, hydrogen peroxide, hypohalides,

20

halidamines, nitric oxide and the peroxynitrite. The reactions of these oxidants involved
in the respiratory burst has been identiﬁed from different mollusc hemocytes such as
Patinopecten yessoensis, Pecten maximus, Crassostrea virginica, C. gigas, Ostrea edulis,
Mya arenaria, Mercenaria mercenaria, Mytilus edulis (Roch 1999), M galloprovincialis
(Arumugam et al. 2000), and Ruditapes decussatus (Tafalla et al. 2003). However, the
mechanism of ROS production and regulation against infections of pathogens remains
largely unknown (Roch 1999).

The humoral defense system also includes lysosomal enzymes (Tiscar & Mosca
2004), opsonins (Renwrantz & Stahmer 1983) and antimicrobial peptides (AMPS)
(Hubert et al. 1996; Mitta et al. 2000a). Produced in hemoytes, the lysosomal enzymes
(B-glucuronidase, acid and alkaline phosphatase, lipase, aminopeptidase and lysozyme)
are released into serum during phagocytosis (Pipe 1990). The lysosomal enzymes can
also be found in digestive glands due to the fact that ﬁltered bacteria represent
nourishment for marine bivalves and the enzymes produced by digestive glands have to
function in both digestion and protection simultaneously (Tiscar & Mosca 2004). Two
major types of opsonin, agglutinins and lectins have been found in many bivalve tissues.
The main targets of the agglutinins are erythrocytes, bacteria, protozoa and algae (Chu
1988), while the lectins have a speciﬁc opsonizing function in hemocyte aggregation and
foreign cell agglutination. The lectins also manage to promote the recognition of foreign
cells because they are speciﬁc for hemocyte and bacteria glycoconjugates (Tiscar &
Mosca 2004).

As a main component of the bivalve humoral defense system, the AMPS have

been identiﬁed not only from bivalve mollusks, but also from a variety of organisms in

21

other phyla (Boman 2003; Bulet et al. 2004). Over 1,000 identiﬁed AMPS have been
classiﬁed in three groups based on their secondary structures (Bulet et al. 2004),
including linear peptides with (it-helices, highly disulphide—bonded (cysteine-rich) [3-
sheets, and those with proline- or glycine-rich character (Douglas et al. 2003; Gueguen et
al. 2006). By inserting themselves into membranes, the AMPS cause the malfunctioning
of pathogen membranes, thereby leading to lysis of pathogens. Moreover, depending on
the tissue distribution, the AMPS are also involved in a number of other reactions, such as
mediating inﬂammation (Hancock & Lehrer 1998). To date, AMPS have been identiﬁed
from several bivalve species and deﬁned as several types, such as defensins, mytilins,
myticins, and mytimycin. Molluskan AMPS have been found to inhibit the growth of a
broad spectrum of pathogens, primarily bacteria and fungi (Mitta et al. 2000b).

Studies on zebra mussels, in past decades, have provided us with basic knowledge
about zebra mussel underwater adhesion, such as the biochemical characteristics of the
foot proteins, the structures of the byssus apparatus, the stage of byssogenesis, and the
activities of byssus under the impact of different environmental factors. This suggested
that the zebra mussel employs a sophisticated system to produce, maintain, and modify
the byssal threads. This process may involve several regulation systems that cannot be
simply reﬂected by the morphological change of byssal threads. To this end, this study
was undertaken with the following objectives: 1) to obtain the genes unique to zebra
mussel byssus and participate in zebra mussel byssus activities; 2) to identify the genes
differentially expressed during the formation of byssal threads; 3) to determine how the
environmental factors affect the zebra mussel byssogenesis on gene level; 4) to ﬁnd out

the molecules with potential functions related to byssal threads protection.

22

    
 
    
    

Stem-forming gland
5

Root \
Temporary byssal threads

Thread-forming gland

 
  

I ,

 
 

Ventral groove

  

\
Permanen
byssal
threads

I

 

Plaque-forming gland “ ﬂ ‘
" ‘ Adhesive
' p roteins

  

Figure l—l The anatomic structure of zebra mussel foot and byssus.

The stem-forming gland surrounds the root of byssal threads. The cells of thread-forming
gland are distributed along the ventral groove in the middle of foot. The plaque-forming
gland is located at the tip of zebra mussel foot. The foot proteins produced by thread-
forming gland and the thread-forming gland releases the protein into the ventral groove
transported to the root of the foot. Two types of byssal threas, temporary threads and
permanent threads, are produced with the protein materials secreted by the three byssus
glands.

23

4— Stem

 

Collagen gland

 

Rigid fibers pro - Col-D 8. N6

 

 

Accessory gland

 

 

Ventral Groove 4—Coatlng - Mefp-1
I d
”I“ ”“5““ MR adheslve - Mefp 2 a. 4
Phenol gland ' :-.;. .,.;.; -'1:'

mer-Mefp38-5

 

 

Figure 1-2 The structure of Mytilid'byssus and distributions of foot proteins.

Three byssus glands were located in the foot of Mytilid. Seven types of foot proteins, by
far, have been identiﬁed from a variety of Mytilids. The structural proteins of Mytilids
byssal threads are pre-Col-D and NG coated with Mefp-1. The proteins ﬁlled in the
plaques are bulk adhesive proteins Mefp 2 and 4, and the primer proteins, Mefp 3 and 5.

24

Muscle mRNA
Tester with adwtor 1F Driver Tester with adaptor 2R

% I

Foot mRNA ___i__1snrybudizam_i_ Foot mRNA

 

 

 

 

 

—|—2nd Hybﬂdtalion |

 

Figure 1-3 The construction of suppression subtractive hybridization cDNA library.
Tester cDNAs are connected to different adaptors, 1F and 2R. After two rounds of
hybridizations, there are many forms of cDNA templates that were generated; however,
only the templates connected to different adaptors at 5’ and 3’ ends can be ampliﬁed by
the adaptor speciﬁc primers. These ampliﬁed templates were originally unique in tester
cDNAs.

This ﬁgure is in color

25

 

Figure 1-4 The anatomic structure of zebra mussel foot and retractor muscles.

The foot is circled by dashed line. The foot is connected with the anterior part of the shell
through anterior retractor muscle (ARM) and anchored on the posterior side of the shell
through posterior retractor muscle (PRM).

26

 

Figure 1-5 The non-reference interwoven loop design with different numbers of
conditions and replicates.

A: the design for 2 conditions with 2 replicates in each condition; B: the design for 2
conditions with 4 replicates in each condition; C: the design for 3 conditions with 3
replicates in each condition; D: the design for 3 conditions with 4 replicates in each
condition.

27

CHAPTER II

Putative Identification of Expressed Genes
Associated with Attachment of the Zebra Mussel
(Dreissena polymomha)

Abstract

Through its ﬁrm attachment to substrata, the zebra mussel (Dreissena polymorpha) has
caused severe economic and ecologic problems since its invasion into North America.
The nature and details of attachment of this nuisant mollusk remain largely unexplored.
Byssus, a special glandular apparatus located at the root of the mussel’s foot produces
threads and plates through which ﬁrm attachment of the mollusk to underwater objects
takes place. In an attempt to better understand the zebra mussel’s adhesion mechanism,
we employed the suppression subtractive hybridization (SSH) assay to produce a cDNA
library with genes unique to the mussel’s foot. Analysis of the SSH cDNA library
revealed the presence of 750 new expressed sequence tags (ESTS) including 304 contigs
and 446 singlets. Using BLAST search, 365 zebra mussel ESTS showed homology to
other gene sequences with putative functions. The putative functions of the homologues
included proteins involved in zebra and blue mussels’ byssal thread formation, exocrine
gland secretion, host defense, and housekeeping. The generated data provide, for the ﬁrst
time, some useﬁil insights into the foot structure of the zebra mussel and its underwater

adhesion.

28

Introduction

Originally from Eurasia, the zebra mussel (Dreissena polymorpha) invaded the
Great Lakes in North America in the early 19808 causing economic and ecologic
devastation. Through its extraordinary ability to attach to hard surfaces underwater, the
zebra mussel has blocked water intake pipes for industrial facilities, eroded underwater
structures, and destroyed native mussel species (Johnson & Padilla 1996). The zebra
mussel attaches to hard surfaces by the byssus, a structure embedded in the mussel’s foot
and consists of excretory glands and byssal threads that end with adhesive plaques
(Morton 1993). There is a dearth of knowledge on the biochemical and physiological
processes leading to byssus formation, attachment and detachment in the zebra mussel
(Frisina & Eckroat 1992; Bonner & Rockhill 1994b). Moreover, there seems to be a large
number of proteins involved in byssus formation and maintenance. The nature of proteins
associated with the byssal apparatus structure and their functions have not been
thoroughly studied due to the technical difﬁculties associated the isolation and
puriﬁcation of proteins using standard biochemical procedure. This study is the ﬁrst step
needed to identify the genes encoding byssal proteins. To achieve this goal, the
suppression subtractive hybridization (SSH) assay was employed to enrich genes

encoding proteins unique to the zebra mussel foot.

29

Materials and Methods

Zebra mussel tissue source, total RNA isolation and mRNA purification

Zebra mussel samples were collected from the Huron River in Ann Arbor, MI,
USA (Latitude: 42.270N; Longitude -83.726W), transported alive in river water to the
Aquatic Animal Health Laboratory at Michigan State University, East Lansing, Michigan.
This sampling Site was selected due to the absence of the quagga mussel (US. Geological
Survey Investigation 2007). The mussels were then identiﬁed according to their
morphological criteria as described elsewhere (Pathy & Mackie 1993). The mussels were
then allowed to acclimate for eight weeks in glass aquaria kept at room temperature and
supplied with well water. While the majority of the mussels were allowed to attach by
placing them on the ventral Side, the rest of mussels were placed on their dorsal side and
stayed, therefore, unattached. Both attached and detached mussels were fed a pure culture
of the algus Ankistrodesmusfalcatus once a week. Five individual mussels from each
attachment status were randomly selected to develop cDNA libraries that emcompass
genes expressed in both attached and detached status. Shell length (measured from umbo
to the opposite shell margin) of selected mussels ranged from 2 to 3 cm. Two types of
tissues were excised from the zebra mussel: the entire foot, after removing byssal threads
(ZMF), which consists of byssal glands and muscular tissues (Rzepecki & Waite 1993b,
a); and the retractor muscles (anterior and posterior, ZMR, Figure 2-1). The Samples
were immediately cryo-preserved in liquid nitrogen until RNA extraction. Total RNA was
extracted with the RNeasy Protect Midi Kit (Qiagen, Valencia, CA) following the
manufacturer’s protocol. All the Poly A+ RNA (mRNA) templates were collected from

the total RNA solution.
30

Construction of the suppression subtractive hybridization cDNA library
BD ClontechTM PCR-SelectTM cDNA Subtraction Kit and PCR-SelectTM cDNA

Subtraction Kit (BD Biosciences, Palo Alto, CA) were applied in the construction of the
cDNA library following the manufacturer’s protocols. In this study, the ZMF samples
were used as the testers that were connected to the speciﬁc adaptor offered by the kit
before the ﬁrst hybridization, and ZMR was treated as the driver, which was not
connected to any adaptor. After two cycles of hybridization and nested PCR, only those
expressed sequence tags (ESTS) that appeared in ZMF (tester) but not in the ZMR (driver)
were maintained and enriched. The cDNA from the library was cloned using the

QIAGEN PCR CloningPlus Kit (Qiagen), which included the vector, T4 DNA ligase, and
the competent cell strain for cDNA cloning. All the steps of the ESTS ligation,
recombinant transformation and cells incubation were performed as described in the

manufacturer’s protocol.

ESTS sequencing

The Eschrechia coli cells were screened with isopropyl-B-D-
thiogalactopyranoside (IPTG) and X-gal in Luria Broth (LB) agar plates (Invitrogen,
Carlsbad, CA). The clones with white colors were picked for inoculation with LB under
200 rpm 37°C overnight. The inoculated bacteria were used for sequencing. All the ESTS
were sequencedusing the Applied Biosystems 3730xl DNA Analyzer (Applied
Biosystems, CA). The sequencing data are processed automatically and quality values are
assigned using the Phred algorithm (Ewing & Green 1998; Ewing et al. 1998). The vector

sequences were masked using the vectors database from GenBank.

31

Data Analysis

The assembling and clustering of ESTS were performed using CAP3 (Contig
Assembly Program) (Huang & Madan 1999). Homologous gene searching was conducted
using local BLAST (http://www.ncbi.nlm.nih.gov/BLAST) in conjunction with the

GenBank protein database. All homologous sequences were conﬁrmed using their

respective E-value. All results with E-Values less than 10.5 were considered highly

homologous. The putative functions of the ESTS that had homologous genes within the

GenBank database were predicted based upon the functions of their homologous genes.
The proteins encoded by identiﬁed ESTS were annotated to Gene Ontology (GO)

terms with software package Blast2GO (Conesa et al. 2005). The G0 terms for the

annotated proteins were based on their molecular functions.

32

Results

ESTs sequencing and assembling

Sequencing and primary data analysis yielded 2336 ESTS (Table 2-1). The
average read length for analyzed ESTs, using the Applied Biosystems 3730xl DNA
Analyzer, was 673 bp. The 2336 ESTS were assembled and the results from CAP3
analysis are displayed in Table 2-1; 304 contigs were assembled with 1890 single ESTS,

leaving 446 singlets that did not have overlaps with other ESTS.

Homologous gene searching and putative function prediction

Homologous genes for the 304 contigs and 447 singlets were searched using the
BLASTX program. Among the results, 365 well-assembled ESTS hit homologous genes
present within the GenBank database with putative functions. The remaining 385 ESTs
either did not have homologous proteins or had no reported putative functions within
GenBank. As displayed in Figure 2-2, 3.0% of the assembled ESTS had putative functions
related to byssal proteins and byssogenesis. 37.0% of the ESTS were homologous with
tick exocrine gland-secreted proteins. 1.0% of the ESTS were predicted to have some host
defense functions involved in different pathways based upon the function of their
homologues. The ﬁnal 6.0% of the ESTs with putative ﬁinctions were grouped into
proteins that have basic cell functionality or with housekeeping gene characteristics. All
the 725 assembled ESTS with length larger than 100 bp were submitted to GenBank and
were assigned accession numbers from AM229723 toAM230448.

Of the sequences with putative byssal proteins and byssogenesis functions, 21 had

homology with the zebra mussel byssal protein Dpfpl (GenBank accession no.

33

AAF75279) (Table 2-2). The other ESTS in this group had homology with a number of
proteins within GenBank that may be involved in zebra mussel attachment, such as the
polyphenolic adhesive protein (323760) from Mytilus edulis (Table 2-3). In the group of
putative exocrine gland proteins, 110 ESTS (accession numbers in GenBank from
AM229751 to AM229861 except for AM229807) had high homology with the western

black-legged tick (Ixodes paciﬁczrs) transcriptome of the salivary glands NPL-2

(AAT92111) and the E-values of the homology are from 3><10'6 to 5X10“; 169 ESTS

(The accession numbers of this group of ESTS are from AM229863 to AM230033 except
for AM229950 and AM229951) were homologous to the black-legged tick, Ixodes

scapularis, putative secreted salivary gland peptide (AAV80789) with the E-values from

7><10-6 to 6X10-12; two of the ESTS in this group (The accession numbers of these two

ESTS are AM230043 to AM230044) had homology with I. scapularis putative salivary
protein that contains GYG repeats (AAY66521) and their E-values are 2><10.9 to 6><10-7

respectively.

Six ESTS showed sequence homology homologous to Rhinoceros beetle (Oryctes
rhinoceros) defensin precursor (096049), as well as some other host defense related
proteins including different types of protease and protease inhibitors (Table 2-4). The
other 122 ESTs with known putative functions were either housekeeping genes or
involved in normal cell functions.

The annotation of the ESTS based on their molecular functions showed that eight
ESTS encoded proteins that were involved in electron carrier activity. Another eight ESTS

encoded binding proteins. Five proteins were predicted to have structural molecule

34

activity. Only one protein was classiﬁed into each of the following categories: receptor
activity, receptor binding, transportor activity, molecular transducer activity, protein

binding, and signal transducer activity. Another 21 proteins were found to have certain
molecular functions (Figure 2-3). The graphic combination of the annotation displayed

the relation between categories and the score of each node was Shown in Figure 2-4.

35

Discussion

Since the SSH was applied, it is assumed that the generated cDNA library was
dominated by the foot transcripts including those involved in the formation of byssal
threads. Numerous ESTS in our cDNA library exhibited very high homology with the
zebra mussel foot protein, Dpfpl, which is present in secretory granules within a gland
that surrounds the ventral groove of the mussel’s foot (Benedict & Waite 1986). The

degree of homology with Dpfpl varied among ESTs; while some showed extraordinarily

high homology (e.g., BG12_F05 and BG22_E08 whose E-values were 3X10'69) others

exhibited barely signiﬁcant homology (e.g., BG_CON_98 whose E-value was 2x10-8).

Interestingly, the 24 sequences that exhibited high homology with the Dpfp I gene did
not share the same sequence as was apparent from the cluster analysis. According to
previous studies, the Dpfpl protein is composed of tandemly repeated and segregated
motifs (Anderson & Waite 1998). However, the absence of the overlaps among the 24
Dpfpl-like ESTs of our study demonstrated that these ESTS are probably different
variants of byssal proteins rather than different tandem repeats of a single gene. This
phenomenon can also be explained by Mytilus edulis foot protein-3 precursor (Mefp3)
model, which encodes the major protein of the blue mussel adhesive plaque. Variants of
Mefp 3 have been reported to exist due to the alternative splicing and RNA editing of the
gene (Warner & Waite 1999). However, ﬁndings of this study cannot verify that if the
multiple copies of Dpfp 1 in our library represent a number of variants or just the tandem
repeats from a single gene. Also, from this study, it is apparent that zebra and the marine

blue mussels (Mytilus edulis) share a number'of similarities in their mechanisms of

36

underwater attachment. Two ESTS (BG13_D08 and BG02_A07) of the zebra mussel
exhibited homology with the polyphenolic adhesive protein 1 of M edulis, which is
considered a major player in the blue mussel underwater attachment in the marine
environment (Waite 1983b).

Another important ﬁnding of the current study is the relatively high number
(37.0%) of zebra mussel ESTS that exhibited high homology to salivary gland peptides of
two Ixodes spp. In the absence of knowledge about the functions of both families of
proteins, it is impossible to make any inference about the functions of the zebra mussel
molecules that resemble the tick salivary gland proteins. The zebra mussel foot harbors a
number of exocrine glands, which are probably similar to salivary glands in terms of
secretion and regulation. Whether these proteins form byssal threads or control their
production remains to be elucidated.

It seems that the zebra mussel guards its foot and the byssus apparatus in the
microbe-rich aquatic environments with a number of molecules whose homologues are
involved in host defense mechanisms (Table 2-4). Homologues of antimicrobial peptides
(e.g., defensin) may be vital in protecting byssal threads from bacterial degradation.
Defensin, in particular, has been found in other mollusks such as the Mediterranean
mussel (Mytilus galloprovincialis) (Yang et al. 2000), the American oyster (Crassostra
virginica) (Seo et al. 2005) and the bay scallop (Argopecten irradians) (Zhao et al. 2007).
These Molluscan defensins have high sequence homology with arthropod defensins (Seo
et al. 2005). Additional defensive molecules, such as tumor suppressor factors and
interleukin enhancer, may also be of importance to the zebra mussel’s innate immunity,

similar to those found in other bivalve Species (Wiens et al. 1999).

37

Protease and protease inhibitors were identiﬁed in the zebra mussel SSH cDNA
library (Table 2-4). Their exact role in the foot or in byssogenesis is currently unknown.
The zebra mussel ESTS included also genes involved in signal transduction, cell division
and development, metabolism, cell structure maintenance, and DNA/protein synthesis.
Since the zebra mussel byssus is an extremely complex system, composed of a number of
tissues including glands, it’s very likely that the byssus employs multiple genes with
various functions. Unlike the other molluscans, such as oysters (Ostreidae) and scallops
(Pectinidae), there is very limited data available on this important organism.

In general, data obtained from this study constitute the ﬁrst basis of knowledge
gained and current studies are directed toward identifying their potential functions.
Moreover, zebra mussel ESTS, with known and unknown potential functions, are
currently being assembled into a microarray to further determine their roles in

byssogenesis.

38

 

Figure 2-1 The structure of zebra mussel foot and retractor muscles.

The organ within the dotted circle is the foot (the source of ZMF-RNA). Black arrows
point to zebra mussel anterior retractor muscles while the non-ﬁlled arrows point to the
posterior retractor muscles. ZMR RNA was extracted fom both anterior and posterior
muscles.

39

Foot protlen genes
3.0%

  
  
  

Exocrine gland secreted
protein genes

0
Unknown byssal 37-0 A

protein genes
53.0%

Host defense related genes
1.0%

Normal cell function genes
6.0%

Figure 2-2 Percentage of ESTS with different possible functions.

40

Signal transducer activity
Protein binding

Molecular function
Molecular transducer activity
Transportor activity

Receptor binding

Receptor activity

Structural molecule activity
Binding

Electron carrier activity

GO-term

 

 

 

 

 

 

 

 

I I I I I I

0 2 4 6 810121416182022

 

Number of Sequence

Figure 2-3 The annotation of proteins encoded by ESTs from cDNA library.
The proteins translated from the ESTs in the library were annotated and classiﬁed into 10
groups based on their molecular functions.

41

 

 

 

 

 

 

 

 

 

 

Molecular Function

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

if“ . . . I “‘-__“
/.---~ is a Sade-.21. Score.0.00 is a ~‘\\._\
/ A: is a \‘S a 1\
Molecular. . ‘ ' Structural ‘ " Binding Electron Carrier T A t' . .
Transducer Act'v'lv Molecular Activity 1 Seq8:8; H Activity ll 3"" W1, '8'“ 87% 1
Seqs:1:Score:0.00 ,Seqs:5;Score:5.00/ ~~ Score:7.00 " ~._Seqs:8;Score:5.00 eqs. ‘ core. ' ,, '
is a - \ 5 I I
- #1 IS a . \ \a /_I.\
Signal Transducer / 600005198 Molecular I/TL‘\\ \ 1,,
_S£§IIScore:0.00 \\ ' »' Seqs:1:Score:0.00 .\ Seqs: 7 x Seqs:1 x
. Its a “" is a ‘f “‘ kp/ ’
I ‘\ 1 '1/ ‘.
r Receptor ACIIVIIy , ‘I. Receptor BIUUII'IQ , C 6080009355
I 333°? .7160 ' seq“; ‘\ m. x
' e. ' Score:7.00 "W”
//A\“\ A
5 6020094872 " co:ooos1o2
. Seqs.1 .- \. Seqs: 5 ,1
_.\\__M ’/_.. ___\\_ _,,..

Figure 2- 4 The combined graphic of annotation of the zebra mussel byssus unique genes.
The number of sequences and the annotation score are labled in each node.

42

Table 2-1 Summary of the ESTS sequenced from the SSH cDNA library.

 

 

ESTS Overall
Number of ESTs 2336
Average length from the sequencer 673bp
Number of contigsa 304
The number of ESTS contained in the contigs 1890
Number of singletsb 446
Total ESTS aﬁer assembling 750

 

a The sequences assembled by many single ESTS with overlaps.

The ESTs have no overlaps with any others in database.

Table 2-2 The ESTS with homology to zebra mussel (D. polymorpha) byssal protein
Dpfpl.

 

 

Zebra mussels Accession # Zebra mussels Accession #
Clone IDs of this given by E-Value Clone le of this given by E-value

study (contigs) EMBL study (singlets) EMBL
BG_CON_212 AM229723 2.0013-63 BG12_F05 AM229736 3 .00E-69
BG_CON_252 AM229724 3.00E—38 BG22_E08 AM229737 3.00E-69
BG_CON_279 AM229725 5.00E-26 BGO9__G08 AM22973 8 3 .00E-2 1
BG_CON_181 AM229726 8.00E-26 BG 17__H08 AM229739 1.00E-12
BG_CON_70 AM229727 3.00E-20 BG13_D02 AM229740 4.00E-12
BG_CON_203 AM229728 1.00E-18 BG07_GO7 AM229741 7.00E-12
BG_CON_149 AM229729 1.00E-12 BG08_F07 AM229742 2.00E-l 1
BG_CON_193 AM229730 5.00E-12 BG09_C04 AM229743 3005-1 1
BG_CON_135 AM22973] 8.00E-10 BG15_D06 AM229744 5.0013-10
BG_CON_156 AM229732 1.00E-9 BGO6_C02 AM229745 2.00E-9
BG_CON_98 AM229733 2.00E-8

 

43

Table 2-3 The other foot proteins with putative potential attachment functions.

 

 

Zebra mussels Accession # le of homologous
Clone IDS of given by Functions of homologous proteins genes from E-value
this study EMBL GenBank
BG13_D08 AM229747 Polyphenolic adhesive protein I 523760 1.00E-9
BGOZ_A07 AM229748 Blue mussel (Mytilus edulis) 2.00E-7

 

Table 2-4 ESTs and their homologous proteins with putative host defense functions.

 

 

Zebra mussels Accession IDs of homologous
Clone IDs of # given by Functions of homologous proteins genes from E-value
this study EMBL GenBank
BG11_E08 AM230099 Defensin precursor 096049 4.00E-6
Rhinoceros beetle (Or-yctes
rhinoceros)
BG10_D04 AM230107 Hemicentin AAK68690 6.00E-l 8

Human (Homo sapiens)

BG23_A03 AM230108 Putative tumor suppressor CAC80049 8.00E-48
Marine sponges (Suberites
domuncula)

BG_CON_268 AM230114 Kunitz-like protease inhibitor AAN 10061 4.00E-8

Dog hookworm

(A ncylostoma caninum)

BG_CON_124 AM2301 15 Tissue factor pathway inhibitor AAB31955 2.00E-8
Rhesus monkey (Macaca
mulatta)

BG02_E12 AM230116 Cathepsin L precursor BAA0397O 1.00E-9

Flesh ﬂy (Sarcophaga peregrina)

 

44

CHAPTER III

Development of a cDNA Microarray of Zebra
Mussel (Dreissena polymorpha) Foot and its Use
in Understanding the Early Stage of Underwater

Adhesion

Abstract

The underwater adhesion of the zebra mussel (Dreissena polymorpha) to substrates is a
complex process that is controlled by a delicate apparatus, the byssus. AS a critical
activity of the byssus glands embedded in zebra mussel feet, byssogenesis is highly active
to produce numerous byssal threads from the settled juvenile stage through the adult
stage in its life cycle. This lifelong activity helps the zebra mussel to ﬁrmly attach to
substrata underwater, thereby causing severe economic and ecologic impacts. In an
attempt to better understand the zebra mussel’s byssus activity, a cDNA microarray
(ZMB) including 716 genes, generated from a Suppression Subtractive Hybridization
(SSH) cDNA library, was printed and used for the comparison of gene expression during
zebra mussel adhesion and non-adhesion. To better understand the byssogenesis
mechanism, RNA samples from the zebra mussel feet with byssogenesis and without
byssogenesis were used in a two-color hybridization to reveal the gene differential
expression in the two states. Based on the P values (P < 0.05), F ifty-two ESTS were
found as differentially expressed genes and were divided into two groups, upregulated
and downregulated group according to there logF C values. With the false discovery rate

(FDR) adjustment, seven were identiﬁed from the upregulated group and nine from the

45

downregulated group. Phylogenetic analysis indicated that the four excretory gland
peptide-like proteins (EGP) encoding genes in the upregulated group are structurally
different than the two in the downregulated list. The amino acid composition analysis on
the proteins, which were encoded by the up- or downregulated ESTS without homologues
(NH) suggested that seven of the NH proteins are biochemically similar to the novel foot
proteins from other mussels. The quantitative reverse transcription PCR (QRT-PCR)
proved the uniqueness of the templates in the array, and also conﬁrmed the differentially
expressed genes identiﬁed by microarray experiment. Our ﬁndings demonstrated that the
zebra mussel byssus cDNA microarray is an efﬁcient tool for the studies of differential
gene expression in different byssogenesis states, thereby revealing important details of

the underwater adhesion.

46

Introduction

The underwater attachment of the zebra mussel (Dreissena polymorpha) to
underwater substrates causes catastrophic ramiﬁcations in the fragile ecosystem of the
Laurentian Great Lakes, USA. Attachment of the zebra mussel is primarily mediated by a
structure called the byssus which is comprised from exocrine byssal glands embedded in
the mussel’s foot, byssal threads located at the root of zebra mussel foot, and adhesive
proteins forming a plaque at the distal end of the threads through which adhesion takes
place (Waite 1992; Rzepecki & Waite 1993a, b). The unique characteristics of the byssus
have helped the zebra mussel survive a number of physical, chemical and biological
control methods. Indeed, in the absence of solid knowledge on the mechanism of zebra
mussel attachment, an effective control strategy cannot be developed.

Current knowledge on the biochemical nature and functions of underwater
adhesion by mussels is very scarce. Three families ofDreissena polymorpha foot
proteins (Dpfpl-3) within the plaques of the byssal threads have been identiﬁed
(Rzepecki & Waite 1993b, a). Dpfp-1 and -2 are believed to play a role in the
maintenance of the byssal structures and have no adhesive properties (Anderson & Waite
2000). Studies performed on marine mussels of the genus Mytilus demonstrated the
presence of 12 byssal proteins, eight of which are believed to play a major role in
adhesion (Benedict & Waite 1986; Waite et al. 1998; Waite & Qin 2001; Sun & Waite
2005; Waite et al. 2005; Lin et al. 2007; Cha et al. 2008). How each of these proteins
contributes to the underwater adhesion and potential differences in mussel attachment
mechanisms between the freshwater and marine mussel species remains largely

unexplored.

47

To better understand the molecular mechanism of zebra mussel attachment, a
suppression subtractive hybridization (SSH) cDNA library was constructed using zebra
mussel foot and muscle tissues (Xu & Faisal 2008) in which foot RNA was used to
generate tester cDNA (ZMF) while the retractor muscle RNA was used as the templates
for driver cDNA (ZMM). By subtracting ZMM from ZMF, all the cDNAs unique to the
zebra mussel foot was assembled. This SSH cDNA foot-Speciﬁc library entails 716 ESTS
encoding proteins with various predicted ﬁinctions such as adhesion, host defense and
structure maintenance. Hence, the ﬁrst aim of the study was to develop and validate a
cDNA microarray that can be used to unravel mechanisms involved in byssogenesis of
the zebra mussel.

Byssogenesis is inﬂuenced by a number of physical, chemical and biological
factors, the most important of which is the presence of a suitable surface (substrate) to
which the mussel can ﬁrmly attach (Clarke & McMahon 1996b, 0; Clarke & McMahon
1996a; Clarke 1999). Indeed, it has been reported that the formation of zebra mussel
byssal threads would not start unless the zebra mussel foot is tightly touching an
underwater substrate (Eckroat et al. 1993). Whether the accessibility of the foot to
attachment substrate triggers the synthesis of byssal thread proteins or modulates their
release is yet to be elucidated. To this end, we report the use of the newly developed
zebra mussel foot cDNA microarray to identify if access to substrates inﬂuences the

differential expression of genes involved in byssogenesis.

48

Materials and Methods

Construction of zebra mussel foot cDNA microarray

The templates for microarray spotting were from the SSH cDNA library
previously constructed in our laboratory(Xu & Faisal 2008). Accession numbers of the
ESTS obtained from the library are from AM229723 to AM230448 in the GenBank. All
ESTS were ampliﬁed using PCR before being spotted on the slides.

The preparation Of DNA templates for microarray followed the protocols detailed
in (Chandrasekharappa et al. 2003). Brieﬂy, seven hundred and sixteen Escherichia coli
clones were selected from thirty-four 96-well plates and rearranged into eight new 96-
well plates. The overnight cultures from the eight 96-well plates were used to prepare the
PCR templates. Ten microliters of each bacterial culture was taken to mix with 90 ul
sterile double distilled water. The mixture was then incubated at 100 °C for 10 minutes
followed by centrifugation for 15 min at 15200 g. The supernatant was used as PCR
templates. M13 forward (5’- GTT TTC CCA GTC ACG ACG TTG -3’) and reverse (5’-
TGA GCG GAT AAC AAT TTC ACA CAG -3’) primers were applied with GoTaq
Green Mastermix (Promega Co., Madison, WI) to yield 100 Lil PCR product for each
EST. PCR products puriﬁcation was done with the MinElute 96 UF PCR Puriﬁcation Kit
(Qiagen Inc., Valencia, CA). All puriﬁed PCR products were dissolved in 20 ul of 50%
dimethyl sulfoxide (DMSO). One pl of each puriﬁed amplicon was loaded onto 1.2%
agarose Tris—acetate-EDTA (TAE) gel to detect the presence of products. If a sample
produced a faint band on gels, the concentration of this sample was determined using

Qubit F luorometer and Quant-iTTM dsDNA BR Assay Kit (Invitrogen, Carlsbad, CA).

49

The wells with template concentrations <100 ng/ul were not used and their contents
replaced with freshly prepared templates whose concentration was 2100 ng/ul. All
puriﬁed amplicons were then transferred into two 384-well plates with the help of the
Tecan Genosys 150 liquid handling robot (Tecan U.S., Research Triangle Park, NC)
before microarray spotting.

The SuperAmine microarray substrate (Array1t®, TeleChem International Inc,
Sunnyvale, CA) was used throughout this study. The cDNA microarray spotting was
performed by the GeneTAC G3 arraying robot (Genomic Solutions, Ann Arbor, MI)
equipped with a 48-pin head. The diameter of each pin is 200 pm. The zebra mussel foot
cDNA microarray was designed to contain the 716 genes arranged in 48 sub-arrays (4
X 12). In each sub array, 64 spots were distributed as an 8 X 8 matrix. The 716 amplicons
were spotted in triplicates, which are randomly located in each sub-array. Zebra mussel
[i—actin was used as the positive control spotted in every sub-array while the spotting
solutions were used as negative controls in each sub-array. Meanwhile, 12 spots in each
sub-array were left blank and considered as both background and negative controls

(Figure 3-1).

cDNA microarray validation

This experiment was designed to ascertain that the ESTs used in the developed
microarray are, indeed, more enriched in the foot tissue as compared to other major tissue
types of the zebra mussel. Zebra mussels used for total RNA extraction were obtained
from the Huron River in Ann Arbor, MI, USA (Latitude: 42°16’12”N; Longitude -
83°43’34”W). Hemolymph was extracted from the anterior adductor muscle sinus by

using a 25G 5/8” needle inserted through the dorsal side of the mussel’s shell. The

50

mussels were then dissected under a dissecting scope, and samples of foot, muscle,
ctenidium, mantle, and gonad were collected separately. Two total RNA samples from
each of the tissue sample were extracted with 5-PR1ME PerfectPure RNA Tissue Kit (5
PRIME Inc, Gaithersburg, MD) and the concentrations were quantiﬁed by a Qubit

F luorometer (Invitrogen, Carlsbad, CA). Then samples were diluted to 10 ng/ul with
RNase and DNase Free Water (Fisher Scientiﬁc, Pittsburgh, PA).

Eight genes were randomly selected for quantitative reverse transcription PCR
(QRT-PCR). Table 3-1 displays the gene speciﬁc primers designed with software
PimerExpress 2.0 (Applied Biosystem, Foster City, CA). As an internal reference, the
188 ribosomal RNA (rRNA) was used to eliminate potential errors brought by
quantiﬁcation. The QRT-PCR was applied with Brilliant SYBR Green QRT-PCR Master
Mix (Stratagene, La Jolla, CA) in the real-time thermocycler, Mastercycler ep realplex S
(Eppendorf, Westbury, NY). For each EST biological replicate, two technical replicates
were applied. This gave four replicates for each reaction. Twenty nano gram total RNA
from each sample was added in each PCR reaction. The program used for QPCR is: 30
minutes at 50 °C, 10 minutes at 95 °C, followed by 40 repeats of a 3-step temperature
cycle (30 seconds at 95 °C, 1 minute at 55 °C and 30 seconds at 72 °C). The values
generated by QRT-PCR were cycle threshold (Ct), which can be used as a parameter to
indicate the relative abundance of the genes. The algorithm of gene differential
expression was 2'AAC' as described by Livak and Schmittgen (Livak & Schmittgen 2001).
The multiple comparisons were performed by Tukey’s test aﬁer analysis of variance

(ANOVA) test with SAS (SAS Institute Inc, Cary, NC).

51

False discovery rate (FDR) determination

Four RNA samples isolated from zebra mussel feet (designated T1, T2, T3, and
T4) were used as biological replicates. T1 and T2 were labeled with Alexa555 and
Alexa647 respectively and applied to the ﬁrst array slide (MTl). T3 and T4, also labeled
with AleanSS and Alexa647 respectively, were applied to a second Slide (MT2). MTl

and MT2 were used as replicates to determine FDR of the newly developed microarray.

Effects of the presence of attachment substrate on gene expression in the foot

Zebra mussels used in this experiment were also collected from the Huron River
in Ann Arbor, MI, USA. The byssal threads of mussels were cut off with a sharp scalpel
at the point of their emergence from the foot groove. Mussels were then randomly
assigned to one of two groups: AD and NAD. Mussels in Group AD were allowed to
attach to a matrix by placing them in glass petri dishes with their ventral Sides facing
down so that their feet have access to the glass. In the second group (NAD), mussels were
placed in petri dishes on their dorsal sides; thereby giving their feet no access to a
substrate to which they can attach. Petri dishes, containing both AD and NAD mussel
groups were placed in an aquarium, supplied with 0.22 pm ﬁltered Huron River water,
and kept at room temperature. The mussels in both groups were fed daily with a pure
culture of the alga Ankistrodesmusfalcatus.

The generation and growth of byssal threads after their severing was observed and
recorded. On days 1, 2, 4, 5, 6, 7, 8, 9, 12, 14, and 16, six mussels from each group were
randomly selected and their byssal threads counted under a dissecting microscope.

In order to identify genes differentially displayed due to the ability of the mussel to attach,

total RNA was extracted from mussels’ feet 48 hours post-treatment as described above.

52

In each group, total RNA extraction involved four biological replicates, with four
individual feet pooled in each replicate. The concentration and the integrity of the RNA
samples were detected by 2100 Bioanalyzer (Quantum Analytics, Foster City, CA).
Started with 10 pg total RNA, the cDNA synthesis and dye labeling were performed by
SuperScript Plus Indirect cDNA Labeling Kit (Invitrogen) containing Alexa Fluor 555 &
647 dyes. The Oligo(dT) primer was used in cDNA synthesis. In each group, two cDNA
replicates were labeled with Alexa 555 while the other two were labeled with Alexa 647.
One cDNA sample from each group labeled with different dyes were mixed, therefore,
two replicates for each combination were obtained: the combination of Alexa 555 labeled
AD-cDNA and Alexa 647 labeled NAD-cDNA, as well as the combination of Alexa 647
labeled NAD-cDNA and Alexa 555 labeled AD-cDNA. The design for this study can be
described as the following:

Slide 1: ADl —> NADl; Slide 2: AD2 <— NAD2;

Slide 3: AD3 —> NAD3; Slide 4: AD4 <— NAD4.

The start point of the arrow stands for being labeled with Alexa555 while the end point of
the arrow means being labeled with Alexa647. Each mixed aliquot was transferred to a
Microcon YM-30 Centrifugal Filter Unit (Millipore, Billerica, MA) to be concentrated.
Before the samples were loaded on the GeneTAC Hbetation (Genomic Solutions), 110
III SlideHyb buffer 3 (Ambion, Austin, TX) was added to each concentrated mixture and
subsequently incubated in 70 °C for 5 minutes. An 18-hour step-down protocol was
applied to all hybridizations; brieﬂy, 3 hours at 60 °C, 3 hours at 55 °C, 12 hours at 50 °C.
The washing procedure followed the hybridizations, and was performed with three

different washing buffers containing SSC buffer (Ambion) and sodium dodecyl sulfate

53

(SDS) under different conditions: 30 seconds in each cycle at 50 0C for three cycles with
medium stringency wash buffer (2 X SSC and 0.1% SDS), 30 seconds in each cycle at 42
°C for three cycles with high stringency wash buffer (0.2 X SSC and 0.1% SDS), and 30
seconds in each cycle at 42 0C for three cycles with post wash buffer (0.2 X SSC). All
slides were rinsed once in 0.2 X SSC and once in double distilled water after they were
downloaded from the Hbetation, and immediately dried by centrifugation at 1,000 g for
two minutes.

Hybridized cDNA microarrays were scanned by GenePix 4000B two-laser
Scanner (Molecular Devices, Downingtown, PA), and GenePix Pro 6.0 (Molecular
Devices) software was then used for image processing and spot intensity ﬁle creation.
The data obtained from the microarray and used in the following analysis have been
deposited in Gene Expression Omnibus (GEO) at the National Center for Biotechnology

Information (NCBI), with the series accession number GSE10851.

Microarray data analysis

The four spot density ﬁles output from GenePix Pro 6.0 were analyzed with
Limma (Wettenhall & Smyth 2004) software package. The raw spot intensity data were
normalized within and between the slides. Least square regression was also performed to
determine the differentially expressed ESTS and their fold change. The P values were
adjusted by Benjamini-Hochberg (BH) correction. The signiﬁcance of the differential

expression was also decided by P values and the FDR as described above.

54

Excretory gland peptide (EGP)-like zebra mussel protein amino acid analysis
This analysis was performed in order to determine the difference between the up-
and downregulated EGP-like genes. EGP-like ESTS, from the up- and downregulated
groups, were translated into amino acid sequences, a phylogenetic tree of the seven
sequences constructed using Mega4 software (Tamura et a1. 2007), and phylogenetic
analysis performed with the Neighbor-Joining algorithm. Multiple alignments of the

EGP-like peptides were done by ClustalW (Thompson et al. 1994).

Amino acid analysis of differentially expressed genes without putative
function

Twenty-one differentially expressed ESTS without putative hits in both the up-
and downregulated groups were translated with the aid of the proteomics and sequence
analysis tools of the Expert Protein Analysis System (Gasteiger et al. 2003). Six
translation frames were obtained from each EST, with the longest continuous amino acid
sequence without a stop codon chosen for further analysis. Corrections were made
manually and the percentage of each amino acid residue in each sequence calculated by

Protein Stat in Sequence Manipulation Suite (Stothard 2000).

Validation of selected differentially expressed genes

QRT-PCR was utilized to verify the differentially expressed genes identiﬁed from
the cDNA microarray. For this experiment, additional AD and NAD groups were created
as described above. Total RNA samples were extracted from both groups two days aﬁer
the mussels settled and were labeled as AD-2 and NAD—2. The QRT-PCR was also

performed as described above. In this QRT-PCR assay, the NAD samples were treated as

55

control and the relative abundance of AD-2 was calculated also with the 2 AACt algorithm

as explained above. Four replicates for each sample were used in this experiment (two
biological replicates with two technical replicates in each biological replicate). Statistical

analysis was also done by ANOVA followed by Tukey’s test (SAS).

56

Results

The tissue speciﬁcity of microarray templates

To test if the templates Spotted on this zebra mussel byssus microarray was
unique to zebra mussel foot. Eight ESTs that were randomly selected from the array were
detected within different tissues of zebra mussel by QRT-PCR. As displayed in Figure 3-
2, the QRT-PCR demonstrates that the eight selected differentially expressed genes were
much more expressed in the zebra mussel foot than in other tissues tested. The

differences in transcription levels of the selected genes among the tissues were highly

signiﬁcant (P< 0.01).

Byssogenesis in group AD and N AD

None of the mussels in the NAD group grew byssal threads during the 16-day
observation period. Mussels in the AD group grew byssal threads that averaged 14.0 j;
1.8 (standard error) per individual within one day to 104.2 1 3.9 threads/individual after
16 days of observation (Figure 3-3). Regression analysis indicated that the trend of the

increase of byssal threads in AD group ﬁts the linear model very well with the correlation

coefﬁcient R2 = 0.9705. The ANOVA analysis with Tukey’s test indicated that Day 2 is

the earliest stage with the number of byssal threads signiﬁcantly different than Day 0 (P
< 0.05). Except for Day 1, the numbers of byssal threads at other time points are all
signiﬁcantly more than Day 0 (P < 0.05). The numbers of byssal threads increased
signiﬁcantly within every two days (P < 0.05), while it is mentionworthy that within the

observation period the number of new byssal threads plateaued between the 7th and the

57

9th day (P>0.05), then continued to increase throughout the remainder of the observation

period (Figure 3-3).

FDR determination

Results obtained from the two microarrays hybridized by differentially labeled
AD-cDNA samples demonstrated that the FDR is much lower than expected at each P
value level. If the P value was set at 0.01 as criteria, only one false positive gene was
obtained (Figure 3-4), suggesting that our design of the ZMB microarray, combined with
LOESS normalization and background subtraction, yields a very low rate of false

positives.

Effects of adhesion status on gene expression

Data obtained from hybridized zebra mussel cDNA microarray using the SSH
library and analyzed with the limma revealed 52 differentially expressed genes in the AD
Group with P < 0.05. Twenty-four genes were upregulated (i.e., logFC > 0, Table 3-2),
and 19 genes were downregulated (i.e., logFC < 0, Table 3-3). Using the cutoff of P<0.01,
based on FDR determination experiment, seven upregulated and nine downregulated
ESTs were found differentially expressed genes (P < 0.01). In tables 3-2 and 2b, the
positive log values Show the upregulated genes highly expressed in AD samples, while
the negative ratios indicated the downregulated genes that are more abundant in NAD
samples. Among the upregulated genes, three of them were homologous to excretory
gland peptides (EGP) identiﬁed from western black-legged tick (Ixodes paciﬁcus,
GenBank accession no. AAT92111) or the black-legged tick (I. scapularis, GenBank

accession no. AAV80789). One of the upregulated ESTS were homologous to shematrin

58

4 of the pearl oyster, Pinctaa’aﬁrcata. The rest of the upregulated genes had no
homologues in GenBank database. The downregulated group had two sequences
homologous to EGP of I. scapularis and one EST homologous to polypeptide release
factor 3 from the yeast, Yarrowia lipolytica. The other downregulated ESTS had no
putative functions based on BLASTX search.

With the cutoff of P <0.05, twenty-four ESTS were found upregulated in AD
group while 28 ESTS were classiﬁed as downregulated. Among the 24 upreguated genes,
ﬁfteen of them had no homologues in GenBank. Nine of the upregulated sequences have
putative functions based on their homologues. AM229726 was found homologous to
zebra mussel foot protein 1 (Dpfp-1) (Table 3-2). Among the downregulated genes,

eighteen of were not homologous to any sequences in GenBank database (Table 3-3).

EGP-like sequences analysis

As displayed in Figure 3-5, the four upregulated EGP-like sequences clustered
together, while the two downregulated EGP-like sequences clustered together.
AM229866, AM229964, AM229813, and AM229911 showed the closest relatedness in
the upregulated EGP-like ESTS clade. The AM229892 and AM229917 formed the other
clade. Multiple alignments of the six EGP-like EST sequences demonstrated a similar
dichotomy between the four upregulated EGPS and the two downregulated sequences.
Within the sequences of the six EGP-like ESTS, The bases at both 3’ and 5’ ends

exhibited more variance than those within the sequences (Figure 3—6).

59

Validation for microarray results

As displayed in Figure 3-7, QRT-PCR conﬁrmed that AM229866, AM230104,
and AM229726 were signiﬁcantly upregulated (P <0.05) in the AD-2 group, while the
AM230072, AM230157, and AM230114 were dramatically downregulated (P < 0.05).

This was in accordance with the results from microarray analyses.

60

Discussion

The cDNA microarray developed along the course Of this study is unique for a
number of reasons. First, it is the ﬁrst of the zebra mussel, a nuisance species that is
causing severe economic and ecologic consequences. There are only a handful of DNA
microarrays from other bivalve mollusks, such as a cDNA microarray of the American
(Crassostrea virgim'ca) and Paciﬁc (C. gigas) oysters (Jenny et al. 2007), that consists of
27,496 ESTs obtained from sequences deposited in the GenBank. Second, it is the ﬁrst
microarray speciﬁcally designed to study underwater attachment mechanisms at the
molecular level. Recently, a relatively small low-density Oligonucleotide microarray has
been constructed from 24 ESTS of Mytilus spp. from sequences deposited in the GenBank
and used to determine gene expression levels in response to pollution stresses (Dondero
et al. 2006). The ESTs in this Mytilus microarray are not related to the foot function. Last,
the zebra mussel foot cDNA microarray was constructed based on an SSH cDNA library,
which allowed the enrichment of foot-speciﬁc expressed genes, therefore, the likelihood
of this microarray reveals novel attachment mechanisms is high. The validation
experiment performed in this study (Figure 3-2) is evidance that the ESTs of the zebra
mussel cDNA microarray are, indeed, highly expressed in the foot and not (or much less)
in other mussel tissues.

False positive results have always been a problem in analyzing microarray data.
The problems stem from a number of factors such as the proportion of truly differentially
expressed genes, distribution of the true differences, measurement variability, and sample
size (Pawitan et al. 2005). Among the tools used to control false positive rate, FDR

correction is the most common statistical method. However, the FDR correction usually

61

gives very high adjusted P values when the sample of the experiment is small (Pawitan et
a1. 2005). In our study, when the BH correction was implemented, most adjusted P values
increased substantially (table 3-2 and 3-3). It iS likely caused by the small sample size of
this study (n = 4). To avoid this problem, Nobis et al. (2003) suggested hybridization of’a
microarray slide with two identical samples, a step that helps determining the actual false
discovery rate. When Nobis et al. (2003) protocol was applied in this study, 13 genes
were found falsely identiﬁed as differentialy expressed at P<0.05; while at P<0.01, only
one falsely positive gene was found. Therefore, subsequent analyses of the AD-NAD
microarray experiment used P < 0.01 as the cutoff. This modiﬁcation in analysis

revealed that 16 genes (seven upregulated and nine downregulated) out of the 52

differentially expressed genes obtained using P<0.05 as the cutoff value (marked with *

in Table 3-2 and 3-3). This method certainly increased the speciﬁcity of the microarray in
ﬁnding the adhesion associated genes on one hand, however it decreased its sensitivity as
shown by qRT-PCR assay of the EST AM229726 and AM2301 14 which were rejected
by P < 0.01 but are deﬁnitely associated with adhesion (Figure 3-7).

Experiments performed in this study also demonstrated that the presence of a
suitable substrate for attachment is vital for byssogenesis. In the AD mussel group, byssal
threads grew in as early as two days and increased thereafter, while in the NAD mussel
group, no byssal threads were formed over the entire observation period. Statistical
analysis between the AD and NAD mussel groups indicated that the difference in the
number of newly generated byssal threads becomes statistically signiﬁcant as early as
'two days post-treatment, even though there have been no threads formed in the NAD

mussel group. Based on this ﬁnding, samples were collected and microarray analysis was

62

performed to determine the differences in gene expression between the AD and NAD
mussel groups at the early stage of byssal thread regeneration, an important stage of zebra
mussel attachment.

Interestingly, the genes identiﬁed from this microarray represent a wide range of
proteins with different putative functions. Some genes are known for their involvement in
adhesion, such as EGP-like sequences (AM229866, AM229964, AM229911, AM229892,
and AM229917 in Table 3-2 and 3-3) that are reported to function as the main adhesive
host defense molecules in black tick’s salivary gland (F rancischetti et al. 2005;
Narasimhan et al. 2007). The shematrin-like molecule is homologous to a shematrin
isolated from the mantle of a pearl oyster Pinctadafucata, providing a framework for
Shell classiﬁcation (Yano et a1. 2006). The actual function of this shematrin-like protein
of the zebra mussel remains unknown; however, the genes exclusively expressed in zebra
mussel foot indicated that its function is more likely to be related to foot activity rather
than to shell classiﬁcation.

Surprisingly, none of the zebra mussel foot proteins, originally identiﬁed in the
SSH cDNA library (Xu & Faisal 2008) were differentially expressed in this early phase
of byssogenesis. This can be attributed to a number of reasons. First, it is possible that
these genes become differentially expressed at a later stage of byssogenesis (i.e., later
than 48 hours post-treatment). Indeed, some of these proteins function as links and
dovetails between adjacent structural proteins, and therefore will only be needed at a later
stage. For example, two proteins (preCol-P) have been identiﬁed from the blue mussel M
edulis, with one being distributed in the foot and the other joining the proximal threads to

the byssal stem (Coyne & Waite 2000). Second, there is a possibility that these proteins

63

are constitutively expressed and their encoded proteins accumulated in the byssal
glandular cells, yet their excretion to form new threads is triggered, directly or indirectly,
by the presence of adhesion substrate. AS previously reported (Eckroat et al. 1993;
Anderson & Waite 2000), byssal proteins are produced prior to the thread formation,
stocked in the foot’s ventral groove, and then released upon attachment. Last, the
accessibility to the substratum may not be the only triggering factor to stimulate the
differential expression of foot proteins. Environmental factors, such as temperature,
dissolved oxygen, current velocity and food availability are known to play an important
role in byssal thread formation (Clarke & McMahon 1996b, c; Clarke & McMahon 1996a;
Clarke 1999). On the other hand, some genes without homologues (NH) in our
microarray exhibited the highest fold increase as demonstrated by microarray analysis
and RT-PCR. This suggests that the transcripts of these genes are very likely involved in
byssal thread regeneration 48 hours post-treatment. These NH protein fragments share
some characteristics to foot proteins identiﬁed from marine or freshwater mussels in the
amino acids composition. For example, in the novel Dpfpl, the proline, tyrosine, aspartic
acid/asparagine, lysine, threonine, and glycine residues together account for more than
50% of the amino acid composition (Anderson & Waite 1998). It is also noted in other
marine byssal precursors, such as M edilus foot protein 1 (Filpula et al. 1990), M
galloprovincialis foot protein 1 (Inoue & Odo 1994), M coruscus foot protein 1 (Inoue et
al. 1996), and, to a lesser extent, M galloprovincialis foot protein 2 (Inoue et al. 1995).
Some other foot proteins, such as M californianus foot protein 3 have amino acid
compositions dominated by glycine, asparagine and tyrosine (Zhao et al. 2006). The high

content of these amino acids in the sequences listed in Table 4 suggests that one of the

64

NH molecules upregulated in the early stage of byssogenesis may have Similar
biochemical characters with the novel foot proteins.

Homologous to a salivary gland peptide ofIxodes paciﬁcus (F rancischetti et al.
2005) and I. scapularis (N arasimhan et al. 2007), the EGP-like proteins were found in
both up- and downregulated molecule groups. According to Fracischetti et al, the salivary
gland peptides isolated from I. paciﬁcus can be divided by 16 groups that are
functionally related to anti-hemostatic, anti-coagulant, anti-microbial, oxidant
metabolism, and housekeeping (Francischetti et al. 2005). The EGP-like proteins
identiﬁed by microarray are all homologous to the salivary gland peptide with putative
anti-microbial activity. However, there is some difference between the upregulated and
downregulated EGPS. The phylogenetic analysis indicated that upregulated EGPS and I
downregulated EGPS belong to different clades (Figure 3-5). Structurally, the main
difference is caused by the sequence areas close to 3’ or 5’ ends, as well as by a longer
fragment consisting of six amino acids (Figure 3-6). It is not clear that if this divergence
will cause difference in tertiary structures of the proteins; however, the structural
difference has been reﬂected by differential expression in this study.

In general, experiments performed in this study underscore the importance of the
newly developed cDNA microarray to better understand mechanisms of adhesion in the
zebra mussel. The microarray analysis directed our attention to a number of proteins of
importance to the early byssogenessis, such as the EPG-like peptides, shamatrin-like
proteins, and some sequences with unknown functions. Future studies to determine their
exact function, location within the byssus, and factors that regulate their expression will

be necessary to unravel the mechanism of zebra mussel attachment.

65

 

Figure 3—1 Zebra mussel byssus cDNA microarray.

This ﬁgure depicts one of the 48 sub—arrays after hybridization. The 64 printed spots were
arranged as an 8 X 8 matrix. The spots (P) located at four comers are ﬁlled with B-actin
DNA. The white crosses demarcate the blank spots that were used as negative controls.

66

I Foot
I Muscle

t I Ctenidlum
I Mantle

I. IHemocyte
e

3 FIST?“ a: LEI-"RA

1 "r4;.~:4
‘ M -"-.‘.L_«'

El
Ii
iii
{.1

 

 

 

1 I I I

AMZMZ AM229808 AM229952 AM229947 AM230153 AM229942 AM230153 AM2301“

TheIDIorSeIectedESTs

Figure 3-2 qRT-PCR with selected genes using different tissues.
The Ct values in foot tissue were used as control with the value 1. The relative expression

levels of the genes in other tissues were calculated using Z'AACt model. * Signiﬁcant
diﬂ‘erence (P<0.05). Bars represent average of two biological replicates istandard errors.

This ﬁgure is in color.

67

120
100

a)

1: 80

ﬂ

0

I-

5 60

T:

g 40

JD

“5 20

L-

3

E o

3

z -20
-40

 

——e—-IH)
—..-NAD

-------- Unear
(AD)

++-—-I—l—a—e—I——u—————-I———n———l

 

 

T r I T I I I j r r

1 2 3 4 5 6 7 8 9 10111213141516
Days post treatment

Figure 3-3 Number of regrown byssal threads over 16 days.

Each point represents an average and standard error of thread numbers of six randomly
selected mussels. The linear regression analysis of the samples in AD group was
performed by Excel with the equation y = 5.9425x + 9.6942. The correlation coefﬁcient

was calculated as R2=0.9705.

68

14-

   
  

 

 

—-e—<0.0S
Q 12 ‘ --I-<0.025
0
g 10 - —.— <0.01
.2
‘: 8-
8
z 6‘
2
.8 4.
S
2 2-
0 I I TI ﬂ
0 1 2 3 4 5

Fold Change

Figure 3-4 Histogram of distribution of the number of differentially expressed genes and
fold-change.

The values of x-axis are fold changes. The y-axis stands for the number of the genes.
Each curve is drawn under a certain P value cutoff: 0.05, 0.025 and 0.01. In each P value
level, the numbers of differentially expressed genes (y-axis) with different fold change
levels (x-axis) were plotted.

69

0.025

AM 229866
AM 229964
AM 22981 3
AM 22991 1
AM 229892
AM 22991 7

 

I l l L I l

I I I I I I
0.025 0.020 0.015 0.010 0.005 0.000

Figure 3-5 Neighbor-Joining phylogenetic analysis with excretory gland peptide-like
molecules.

All the proteins selected in this phylogenetic tree are homologous to excretory gland
peptide of the western black-legged tick (Ixodes paciﬁcus, GenBank accession no.
AAT92111) or the black-legged tick (I. scapularis, GenBank accession no. AAV80789).
The upregulated AM229866, AM229964, AM229813, and AM229911 ESTS clustered
together while the downregulated AM229892 and AM229917 ESTS formed another
cluster. The number above each branch is the actual length of the branch.

70

AM229866
AM229964
AM22991 I
AM229813
AM229892
AM229917

 

AM229866 GTAGTACGCTTCTAGCGTGCTCCGGTAGCAAACGCTCAAT
AM229964 GTAGTTACCGCTTCTAGCGTGCTCCGTAGCAAACGCCAAT
AM229911GTAGTATGCGCTTCTAGCGTGCTCCGTAGCAAACGC(‘AAT
AM229813 GTAGTGCTTCTAGCGATGCTGCTCCGTAGCAAACGC(‘AAT
AM229892 GTAGCAGCTTCTAGCGTGTGCTCCGTAGCAAACGCCCAAT
AM229917 'CG C T T C T A G (‘G AT T G C T CC AG T A GCA A A C G C C C A A '1‘

   
   
  
     
  
 

  
  
  
  
  
 
  
  
  

 

AM229866
AM229964
AM22991 1
AM229813
AM229892
AM229917

 

AT—--GATTACGGATATGGCGGTAACAACTATGGTTACCC W4
ACTACGATTACGGATATGGCGGTAACAACTATGGCTACCC
\TTATGATTACGGATATGGCGGIAATAACTATGGCTACCC
ACTATGATTACGGATATGGCGGTAACAACTATGGCTACCC
ACTATGACTACGGATATGGCGGTAACAACTATGGCTACCC
ACTATGACTACGGATATGGCGGTAACAACTATGGCTACCC

AM229866 A '1‘ (1 (1G G T G G T '1‘ A T G G C G G A '1’ A T (i (i C A A A 'l‘ A A T A A A T G T C G 354

 

     
 
      
   

AM229964 A’I’GGGGTGACTTA(iGC - - - TA TGGCAAATAAT AATCI ' ‘ 334
AM229911 ATGGGG ’7' ' ‘ 353
AM229813 7‘ 311
AM229892 335
AM229917 293
AM229866 I.\.AACCATTGCTTAAAGG\.\CTAAA.1\CGATC’I'ACGACCI 394
AM229964 " 363
AM229911 l' \AAACCATTGCTTAAAGG\ \ACCICATC’1'T.\CAA(iACC.»\T 393
AM229813 1 \AA ------------ AA A ACGA I’('TACGACCA ’l‘ 339
AM229892 I \AAAACGA'I‘CT,\C(iACC. " 375
AM229917

AM229866 ,
AM229964
AM22991 l .
AM229813
AM229892
AM229917

Figure 3-6 Multiple alignments with the seven EGP-like sequences.

EST AM229866, AM229964, AM229813, and AM229911 were all found upregulated in
microarray experiment. Sequence AM229892 and AM229917 were proved to be
downregulated. ClustalW was used to align all the EGP-like ESTS. The dark stripe
indicates that more than 70% nucleotides in this site are identical. The grey shading
highlights the site with similar nucleotides (similar biochemical characters). The hyphens
Show the gaps existing in the sequences.

71

z-AACt
DJ
J

 

 

 

 

AM229866 AM230104 AM229726 AM230072 AM2301S7 AM230114

Selected ESTs

Figure 3-7 Validation of microarray results with qRT-PCR.

The Ct values of NAD were used as control with the value 1. A baseline was drawn to

Show the expression level of NAD group. The relative expression levels of the genes
AA t

under attachment in each day were calculated using 2- model. * Signiﬁcant
difference (P <0.05).

72

Table 3-1 The primers used for QRT-PCR.

 

Primer names

Sequence

 

AM230 188_Forward
AM230 18 8_Reverse
AM229952_Forward
AM229952_Reverse
AM229808_Forward
AM229808_Reverse
AM230072_Forward
AM230072_Reverse
AM229947_Forward
AM229947_Reverse
AM230 1 57_Forward
AM230 1 57_Reverse
AM229866_Forward
AM229866_Reverse
AM230104_Forward
AM230 104_Reverse
AM229726_Forward
AM229726_Reverse
AM2301 14_Forward
AM2301 14_Reverse
AM229942_Forward
AM229942_Reverse
AM230042_Forward
AM230042_Reverse
AM230153_Forward
AM2301 53_Reverse
* 1 8SrRNA_Forward
* 1 88rRNA_Reverse

5’- TGC GAG CCG ACG TTA TCC -3’

5’- GCT ATC GGC CGC AAC AAT AT -3’

5’- GCG GCC GAG GTT TGT TAG T -3’

5’ -ATC CGT AGT CAT AGT ATT GGG CG -3’
5’ -GAT GAA CAC CAA ACA GTT GAT GTG -3’
5’ -TAT TGG GCG TTT GCT ACC G -3’

5’- CTT CCA CGT TAT CCG CAT CA -3’

5’- TCC CCG TCT GGA GCC TAT C -3’

5’- GGC AGG TAA AAC ACC TGA TGT G -3’
5’- TCA TAG TAT CGG GCG TTT GCT -3’

5’- GAT CCT ATT GGC CTG GAC CAT -3’

5’- ATA CCC GCT TAC GGA ATT ACC TT -3’
5’- CCT CCG AGC AGC GAC AAA -3’

5’- CTA CCG GAG CAG TCG CTA GAA -3’
5’- TCC TTG GGT GTG GCA TCA G -3’

5’- GAC GAC TGC CCC AAG TTC TC -3’

5’- CCG ATG GGC CAT ATG ATA AGA A -'3’
5’- GAT GCC TTG CTT GTC TTG TTG A -3’
5’- GAA CTG TGC GCT GGA TAC GA -3’

5’- GGA TCA CCC AGG CTC CAT T -3’

5’- CCA GGA AAT TAT GGC GAC TAT GA -3’
5’- CCA CCC CCT CCG AGG AT -3’

5’- GAT GAA CAC CAA CCA GCT GAT G -3’
5’- TAC TGG GCG TTT GCT ACC G -3’

5’- TCC TAT TGG CCT GGA CCA TG -3’

5’- CCC GCT TAC GGA ATT ACC TTG -3’

5’- GAC ACG GCT ACC ACA TCC AA -3’

5’- CTC GAA AGA GTC CCG CAT TG -3’

 

* The primers used as internal reference for QRT-PCR.

73

Table 3-2 The ESTs with upregulated expression level in AD group.

 

 

Accession # Genes ID LogF C Fold-change P Putative Function

AM229866* BG21_BO8 2.402 5.285354 4.00E-05* salivary gland peptide
[lxodes scapularrs]

AM230108* BG34_H05 2.679 6.4041 18 0.00018 N/A

AM230253* BG28_F07 1.81 1 3.508854 0.00196 N/A

AM229964* BGl3_A07 1.535 2.897884 0.00244 salivary gland peptide
[lxodes scapularis]

AM229911* BG33_C07 1.402 2.642677 0.00563 salIvary gland peptide
[Ixodes paczﬁcus]

AM229799* BG30_F03 1.567 2.96288 0.00737 Shematrin-4
[Pmctadaﬁicata]

AM229776* BG22_A01 1.348 2.54559 0.00778 N/A

AM229813 MF030105_E04 2.585 6.000156 0.01072 salivary gland peptide
[Ixodes paciﬁcus]

AM230109 BG03_A09 1.281 2.430074 0.01 141 N/A

AM230094 BG33_H08 1.273 2.416636 0.01193 choriogenin Hminor
[Oryzias Iatipes]

AM230117 BG33_H09 1.272 2.414961 0.01206 N/A

AM230104 BG10_D04 1.271 2.413288 0.01206 hemicentin l
[Homo sapiens]

AM230205 BGZO_C08 1.21 1 2.31498 0.01679 N/A

AM230183 BG14_F10 1.181 2.267339 0.01975 N/A

AM230042 BGZS_H08 l. 14 2.20381 0.02435 N/A

AM229726 BGZO_A02 1.311 2.481 135 0.02499 Byssal protein Dpfpl
[Dreissena polymorpha]

AM230399 BG18_F 10 1.585 3.000078 0.02692 N/A

AM229876 BG14_C02 l . 1 13 2.16295 0.02803 N/A

AM230435 BG28_BO4 1.105 2.150989 0.02913 N/A

AM230179 BG08_BO3 1.261 2.396618 0.03104 N/A

AM230100 BG34_A07 1.063 2.089272 0.03586 N/ A

AM230249 BG28_I-109 1 .046 2.064797 0.03897 N/A

AM229789 BG06_C03 1.463 2.75681 0.041 1 1 N/A

AM229935 BG97- l 92_BO6 1.028 2.039195 0.04241 Shematrin-4
[Pinctadafucata]

 

* The differentially expressed ESTs with P <0.01.

74

Table 3-3 The ESTs with downregulated expression level in AD group.

 

 

Accession # Genes ID Coef Fold P Putative Function
C hang;

AM230401 1: BG 1 8_D08 -2.447 0.183392 1.35E-06 N/A

AM230072* BG18_H07 -3.022 0.123108 2.00E-05 N/A

AM229892* BG31_E06 -2.01 0.248273 0.00059 Salivary gland peptide
[lxodes scapzilaris]

AM229917* BG33_F04 -1.98 0.25349 0.00071 Salivary gland peptide
[lxodes scapularis]

AM229853* BG97-192_H1 1 -1.348 0.392836 0.00778 N/A

AM229749* MF030105_C07 -1.543 0.343171 0.00833 N/A

AM230237* BG34_G06 —1.326 0.398873 0.00884 N/A

AM230157* BG34_A05 -1.318 0.401091 0.00929 Polypeptiderelease factor
[Yarrowra lipolytica]

AM230394* BG 17_G 10 -l.313 0.402483 0.00955 N/A

AM229918 BG18_D06 - l .292 0.408384 0.01072 N/A

AM230372 BGlS_H07 - l .285 0.410371 0.01117 N/A

AM230377 BG16_D10 -l..792 0.288771 0.01234 N/A

AM229777 BG31_E04 -1.259 0.417833 0.01293 N/ A

AM229781 BG29_F09 - 1 .245 0.421908 0.014 N/A

AM230109 BG23_D06 -1.235 0.424842 0.01479 N/A

AM230268 MF030105_B 12 -1.422 0.373195 0.01502 N/A

AM229973 BG28_E10 -l .231 0.426022 0.01505 Salivary gland peptide
[Ixodes scapularis]

AM230114 B020_C01 -1.733 0.300826 0.01555 Kallikrein
[Sus scrofa]

AM229947 BG26_E11 -1.209 0.432568 0.01694 Salivary gland peptide
[Ixodes paciﬁcus]

AM230178 MF030105_C01 -1. l 17 0.461052 0.02738 N/A

AM230258 BG 16_G06 -1.287 0.409802 0.0278 N/A

AM229831 BG32_F09 -1.286 0.410086 0.0279 Salivary gland peptide
[Ixodes paciﬁcus]

AM230384 BG17_C09 -1.285 0.410371 0.02799 N/A

AM230406 BG18_G03 - l .249 0.42074 0.03277 N/A

AM229920 BG33_E08 -l.45 0.366021 0.04291 Salivary gland peptide
[Ixodes scapularis]

AM229864 BG16_H05 -1.181 0.441046 0.04338 Salivary gland peptide
[Ixodes scapularis]

AM229892 BG34_BO8 -1.013 0.495515 0.0455 Salivary gland peptide
[Ixodes scapularis]

AM230248 BG 17_D01 -l .419 0.373971 0.04754 N/A

 

* The differentially expressed ESTs with P <0.01.

75

CHAPTER IV

Gene Expression Profiling During the
Byssogenesis of Zebra Mussel (Dreissena
polymorpha)

Abstract

Since its invasion to the North American waters 20 years ago, the zebra mussel
(Dreissena polymorpha) has negatively impacted the ecosystems it has infested through
its ﬁrm underwater adhesion. The molecular mechanisms governing the ﬁinctions of the
zebra mussel byssus, the main structure responsible for maintaining the underwater
adhesion, have received little attention. Our previously developed zebra mussel foot
byssus cDNA microarray was applied in this study to identify the genes involved in
different stages of the byssal threads generation. Byssal threads of zebra mussels were
manually severed under laboratory conditions and the formation of new byssal threads
was followed over a three week course. By comparing the gene expression proﬁles in
different stages of byssal threads generation (byssogenesis) to their baseline values, we
found that the number of unique byssus genes differentially expressed at 12-hour, l-day,
2 days, 3 days, 4 days, 7 days, and 21 days post-treatment was 13, 13, 20, 17, 16, 20, and
29, respectively. Comparisons were also made between two subsequent samples (e.g.,
12h vs. 1d, 1d vs. 2d, 2d vs. 3d, and so on).

Seven differentially expressed genes were selected for validation by using quantitative

reverse transcription PCR (qRT-PCR) and the results were consistent with those from the

76

microarray analysis. By using ﬂuorescent in situ hybridization (FISH), we found that two
microarray identiﬁed genes, BG15_F 03-DPF P and BG16_H05-EGP, were expressed in
two major byssus glands located in the zebra mussel foot: the stem-forming gland and
plaque—forming gland, respectively. Moreover, the qRT-PCR of seven microarray
identiﬁed genes with different zebra mussel samples suggested that they were also
expressed in other mussel tissues beside the foot, albeit at much lower levels. This
suggests that the microarray identiﬁed genes were produced primarily by the foot, and are
likely associated with byssogenesis. The differentially expressed genes identiﬁed in this
study suggested that multiple molecules are involved in byssogenesis, most likely
performing multiple functions during the generation of byssal threads. These results
obtained herein represent the ﬁrst logical step toward understanding underwater

attachment mechanisms employed by this invasive species.

77

Introduction

Native to eastern Eurasia, the zebra mussel (Dreissena polymorpha) has invaded
North America since the late 19803. Zebra mussel invasion led to economical and
ecological devastation. The mussel attaches to substrates underwater with hair-like byssal
threads produced by glands within the zebra mussel foot (Morton 1969; Claudi & Mackie
1993; Benson & Raikow 2009). The strength of zebra mussel attachment is mostly
decided by the number of its byssal threads (Dormon et al. 1997), thus the generation of
byssal threads, also known as byssogenesis, is critical for zebra mussel underwater
attachment and survival. The mechanism of zebra mussel byssogenesis remains largely
unknown, especially at the molecular level. Based on previous morphological and
biochemical studies, it has been determined that byssogenesis in the zebra mussel foot
goes through multiple stages, ending up with the formation of byssal threads with
adhesive pads (Eckroat et al. 1993).

Three major byssus glands are involved in this process. The stem-forming gland,
located at the root area of the zebra mussel foot, was the starting point of the zebra
mussel byssal threads. The thread-forming gland cells evenly distributed along the ventral
groove in the middle of the foot produce two main zebra mussel foot proteins, D.
polymorpha foot protein -1 and -2 (Dpfp-1 and -2), which serve as the component
maintaining the byssal structures (Rzepecki & Waite 1993b, a). Another major byssus
gland, the plaque-forming gland, is located in the tip area of the foot and involved in the
formation of byssal thread plaques (Rzepecki & Waite 1993b). The byssogenesis in the
zebra mussel foot ceases when the mussels are fully attached to suitable substrates

(Eckroat et al. 1993). On the contrary, the byssogenesis can be initiated by the removal of

78

existing byssal threads (Clarke & McMahon l996c, b; Clarke & McMahon l996a; Clarke
1999)

Morphological studies Of the zebra mussel foot revealed the presence Of three
major exocrine glands that are involved with byssal thread formation: the stem-forming
gland, thread-forming gland, and plaque-forming gland (Rzepecki & Waite 19933).
Located at the root of the zebra mussel foot, the stem-forming gland forms and secretes
the proteins that build the trunk of the byssal threads. This process is considered the
initiation event of byssogenesis. The thread-forming gland cells are distributed along the
foot’s ventral groove and produce a large amount of Dpfp-l and -2, both important in
holding the byssal thread structure. The plaque-forming gland, involved in the formation
of the enlarged end of byssal threads, is located at the tip section of the zebra mussel foot.
Along with the thread-forming gland, the plaque-forming gland releases numerous foot
proteins into the ventral groove. The mixed proteins in the ventral groove are transpOrted
to the root of the foot, and thereafter, expelled into the hydrophobic environment created
between the foot and underwater substrates to form the complete byssal thread (Eckroat
et al. 1993). Clearly, the process of byssogenesis seems to be regulated by a number of
intrinsic and extrinsic factors.

Most recently, we have developed a cDNA microarray (716 genes) based on
normalized zebra mussel foot cDNA library (Xu & Faisal 2008; Xu & Faisal 2009a). In a
previous study, we identiﬁed molecules with potential involvement in zebra mussel
attachment which are differentially displayed when the mussels have an access to
attachment substrates for 48 hours. Since the byssal threads take up to 21 days to be fully

formed (Waite 2002; Farsad et al. 2009), this study was designed with a holistic approach

79

the would allow the identiﬁcation Of shifts in gene expression during the entire length of
byssogenesis. In speciﬁc, we employed cDNA microarray to compare gene expression
patterns between quiescent, fully attached mussels, and those in mussels at different time

points during the formation of new byssal threads.

80

Materials and Methods

Zebra mussel sample collection and experimental design

Zebra mussels used in this study were collected from the Vineyard Lake in
Brooklyn, MI, USA (Latitude: 42°4’59”N; Longitude: 84°12’34”W). 1n the laboratory,
the mussels were thoroughly cleaned, placed in submersed glass PYREX® 100 X 15mm
Petri-dish (Corning Inc, Coming, NY) in aerated, ﬁlter-sterilized (0.22 pm ﬁlter)
Vineyard Lake water in a glass aquarium (30 gal). The mussels were fed weekly with a
pure culture of the algus A nkistrodesmus falcatus and were allowed to be acclimated to
the laboratory conditions for eight weeks. The adapted mussels were then divided into .
two groups. Mussels of the ﬁrst group were detached, their byssal threads severed as
close as possible to their roots by a sharp scalpel, and then allowed to reattach in the
submersed Petri-dishes. In the second group, the mussels were allowed to maintain their
attachment status and were considered the control group.

Total RNA was extracted from zebra mussel feet of both groups at 12 hours, 1
day, 2 days, 3 days, 4 days, 7 days, and 21 days post mussel reattachment. For Group 1,
four biological replicates were obtained at each sampling time point and each replicate
contained total RNA pooled from ﬁve mussel feet. RNA Samples were collected from
mussels in Group 2 in the same manner at the same time intervals. Twenty-six microarray
slides were used for hybridizations. Each slide was hybridized with two different samples
labeled with Alexa 555 and Alexa 647, respectively. Thus, the gene expression proﬁles of
the two samples on each slide could be compared. Fourteen slides were used to compare

mussels in Group 1 and Group 2. The other 12 slides were applied to differentiate the

81

gene expression proﬁles between any two Group 1 samples at connected time points (i. e.,
12 hours vs. 1 day, 1 day VS. 2 days, and so on). The detailed sample labeling and
hybridization combination can be found in the description of the deposited microarray
data in ‘NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) with the

series accession number GSE16407.

RNA extraction and cDNA synthesis

The total RNAS used for microarray hybridizations were extracted with 5-PRIME
PerfectPure RNA Tissue Kit (5 PRIME Inc, Gaithersburg, MD) and the protocol
followed the manufacturer’s instructions for the kit. The reverse transcription of the total
RNAS was performed with SuperScript Plus Indirect cDNA Labeling Kit (Invitrogen,
Carlsbad, CA) containing Alexa Fluor 555 & 647 ﬂuorescent dyes. Fifteen micrograms
of total RNA from each sample was used for cDNA synthesis. The procedures of cDNA

syntheses and labeling were all described by the instructions for the kits.

Microarray hybridization and data analyses

Details of the zebra mussel foot cDNA microarray used in this study can be found
in our previous report (Xu & Faisal 2009a). Microarray hybridizations were performed
on the GeneTAC Hbetation (Genomic Solutions, Ann Arbor, MI), and the 18-hour step-
down protocol for hybridization was used as described previously (Xu & Faisal 2009a).
The hybridized dual-channel array slides were scanned by GenePix 4000B two-laser
Scanner (Molecular Devices, Downingtown, PA), and the images from the scanner were
processed by GenePix Pro 6.0 (Molecular Devices) software to generate the spot intensity

ﬁle of each array image. The spot density ﬁles were analyzed with Limma soﬁware

82

package (Smyth 2005). The preliminary data processing includes background correction,
within array normalization and between array normalization. The differentially expressed
genes were determined by least square regression. The relative expression level of each
gene at each time point was calculated compared to the common reference. Meanwhile,
the differential expression of each gene between any of the two time points was also
calculated and demonstrated by the logarithmic value of the ratio between the two time
points. The hierarchical clustering was done by the software Genesis (Sturn et al. 2002)
and the dataset of this microarray study was also deposited in NCBI GEO (series
accession number GSE16407). The putative ﬁmctions of the identiﬁed genes were
predicted by the homologues searching method, Basic Local Alignment Search Tool
(BLAST) (Altschul et al. 1990) with GenBank nucleotides and protein database. For the
genes without homologues in the database, the sequences were scanned with
InterProScan function (Zdobnov & preiler 2001) in software package Blast2G0

(Conesa et al. 2005) to detect the signature sequences of unknown molecules.

Quantitative reverse transcription PCR (qRT-PCR) validation of microarray data
The qRT-PCR was applied to validate the expression proﬁles of the genes that

were affected by attachment status identiﬁed by microarray. Additional zebra mussels

were collected and maintained as described above. The treatment and RNA sampling of

mussels were the same as for the samples used for microarray study. The one-step qRT-

PCR was performed with Power SYBR® Green RNA-to-CTTM l-Step Kit (Applied

Biosystems Inc, Foster City, CA) and Mastercycler ep realplex S thermocycler system
(Eppendorf, Westbury, NY). The thermocycler program used for the qRT-PCR was 30
minutes at 48 °C, 10 minutes at 95 °C and 40 repeats of a 2-step temperature cycle (15

83

seconds at 95 °C and 1 minute at 60 0C). The relative expression levels of the gene

expression were calculated with 2.AACt algorithm using cycle threshold (Ct) Of PCR

described by Livak and Schmittgen (Livak & Schmittgen 2001). The gene Speciﬁc
primers and the internal reference control used in this assay were designed by Primer

ExpressTM (Applied Biosystem Inc.), listed in Table 4-1.

Visualization of two differentially expressed genes in foot tissues using RNA
ﬂuorescence in situ hybridization (FISH)

Fluorescence in situ hybridization (FISH) was applied in order to locate the
transcripts of two genes in situ, of potential importance to byssogenesis. The product of
BG15_F03-DPFP is homologous to D. polymorpha foot protein -1 (Dpfp-1), which is
believed to play a major role in the maintenance of the byssal thread structure, while the
protein encoded by BG16_H05-EGP is homologous to one of the peptides secreted by I.
paciﬁcus salivary gland. The zebra mussel foot tissue was immediately ﬁxed in 2%

paraformaldehyde phosphate buffer saline after severing from the mussel. The ﬁxation

solution contained 0.02 M NaH2P04, 0.0077 M NazHP04, 1.4 M NaCl, and 2% w/v

paraformaldehyde (pH 8.0). Excised tissue was incubated in the ﬁxation buffer for ﬁve
hours at room temperature and then kept in 100% ethanol until use. The dehydrated tissue
sample was then covered totally with parafﬁn and longitudinally sectioned (5 pm). Ten
serial longitudinal sections from the middle of the foot were mounted on RNase- and
DNase-free glass slides (Corning Inc.).

The probes used for the FISH were RNA probes which were complementary

strands of the transcripts of the two selected genes. The PCR product of the gene

84

fragment was cloned into pGEM®-T Easy Vector Systems (Promega U.S., Madison, WI).
Then the positive colony was picked for sequencing with M13 forward primer to clarify
the direction of the insert of the recombinant plasmid. The PCR was applied on the
selected colonies with both M13 forward and reverse primers. The PCR products were
puriﬁed with Wizard SV Gel and PCR Clean—Up System (Promega) and used as the
templates of RNA transcription. The T7 or Sp6 RNA polymerase was selected according
to the insert direction of the recombinant plasmid to synthesize the antisense RNA.
Besides the genes BG15_F 03-DPF P and BG16_H05-EGP, RNA probe of zebra mussel
cytoplasmic actin (AF082863) gene was also produced in the same way described above.
The probe BG15_F03-DPF P (DPFP-G) was labeled with ﬂuorescent dye Alexa 488
(Invitrogen) while the probe BGl6_H05-EGP (EGP-R) was labeled with Alexa 594
(Invitrogen). The control probe cytoplasmic actin was divided into two groups and

labeled with Alexa 594 (ACT-R) and Alexa 488 (ACT-G), respectively. The steps of the

RNA synthesis and labeling can be found in the instructions of the two kits: FISH TagTM

RNA Green Kit (Invitrogen) and Fish TagTM RNA Red Kit (Invitrogen). The probes

DPFP-G and ACT-R were used to hybridize with same zebra mussel foot section slide
while the EGP-R and ACT-G were applied on the other slide. The RNA-FISH was
performed as recommended by manufacturer (Invitrogen). The hybridized Slides were
visualized by Olympus BX41 (Olympus America Inc., Center Valley, PA) microscope
under the excitation of mercury lamp and the images were captured by Olympus DP25
digital camera (Olympus America Inc.). The image processing software DP2-BSW
(Olympus America Inc.) was used to combine the different color channels into a Single

image.

85

Tissue distribution of representative expressed genes

In order to ﬁnd whether the the differentially expressed genes (DEG) are also
expressed in other tissues besides the zebra mussel foot, the one-step qRT-PCR was
performed to detect the amount of transcripts of selected DEG in zebra mussel hemocyte,
retractor mussel, ctenidium, mantle, as well as foot. The mussels were collected,
maintained, and handled Similarly to the mussels described above. Immediately
following the acclimation period, RNA samples were extracted from hemocytes, feet,
muscles, ctenidia, and mantles of zebra mussels. The protocol for hemocyte collection
was described by our previous studies (Xu & Faisal 2007, 2009b) and the other samples
were taken by sterilized scissors and scalpels. The primers used in this study were listed

in Table 4-1.

86

Results

Differentially expressed genes during the byssogenesis identiﬁed by
microarray

Compared to the non-byssogenesis group, several genes have been up- and
downregulated at the different time points during byssogenesis; their numbers and nature
varied, however, along the course of byssogenesis. Numberwise, 9, 8, 3, 2, 10, 12, and 15
genes were upregulated at 12 hours, 1 day, 2 days, 3 days, 4 days, 7 days, and 21 days
post-treatment, respectively. On the contrary, 4, 5, 17, 15, 6, 8, and 14 genes were
downregulated after 12 hours (Table 4-2), 1 day (Table 4-3), 2 days (Table 4-4), 3 days
(Table 4-5), 4 days (Table 4-6), 7 days (Table 4-7), and 21 days (Table 4-8), respectively.
Comparative gene expression studies were performed between any two different time
points for each differentially expressed gene.

Based on the expression patterns of each differentially expressed gene along the
time course of the byssogenesis, the genes were hierarchically arranged into 12 clusters
(A-L, Figure 4-1). As demonstrated in Figure 4-1, cluster A included 16 genes which
were mostly upregulated during the early stages of byssogenesis (12 hours to 1 day of
byssogenesis) and the upregulation was diminished after that. The eight genes included in
cluster B had opposite expression patterns than cluster A, which were not upregulated
until 21 days of byssogenesis. The expression patterns of the eight genes in cluster C
were similar to those in cluster B; however, they were suddenly downregulated at day 7
during byssogenesis before they were signiﬁcantly upregulated in byssogenesis at day 21.

Along the time course of the byssogenesis, the expression levels of the genes in cluster D,

87

E, F, and H were downregulated at the time points 4-, 2-, 3-, and 21-days, respectively.
The genes in cluster G appeared to have downregulated expression levels during the
byssogenesis of the whole experiment. The downregulation became more and more
signiﬁcant from the beginning of the byssogenesis until day 7, where the downregulation
of the genes in cluster G reached maximum level. The three genes in cluster G included
an exocrine gland peptide-like gene (BG16_H05), a gene with anaphylatoxin domain and
an epidermal growth factor region (BG16_GO6), and a gene homologous to neuropeptide-
like protein 31 (MF030105_E03). 0n the contrary, the cluster I contained three genes
whose expression patterns all appeared to be upregulated during the byssogenesis with
the maximum expression levels at 7 days post-treatment. Two of the genes in cluster 1,
BG l8_G06 and BG16_D10, had no homologues found in the database. The other one in
this cluster, BG29_D08, was a homologue to another exocrine gland peptide, which was
structurally different than that in cluster G. Eleven genes were included in cluster J and
were upregulated at days 4 and 7 during byssogenesis. Expression patterns of the Six
genes in cluster K were ﬁrst upregulated at day 7, and subsequently downregulated at day
21 byssogenesis. The last cluster (cluster L) had all the genes with the expression proﬁles
downregulated at the early stage of byssogenesis (12-hour) but the expression levels were

continuously increased until day 7 (Figure 4-1).

Validation of the results obtained from microarray

The expression patterns of seven genes identiﬁed by microarray were validated by
qRT-PCR. According to the qRT-PCR results, the expressions of BG20G04,
BG97/192_E09, and BG15_H03 were all upregulated at time points 12 hours, 1 day and

4 days post the initiation of byssogenesis, respectively (Figure 4-28), which is consistent

88

with the microarray results (Figure 4-2A). The other four genes were all differentially
expressed at two time points during the byssogenesis based on the microarray results.
BG16_H05 was downregulated after 4 days and 7 days reattachment (Figure 4-2A).
Similarly, the downregulation of the gene expression of BG97/192_E09 during the
byssogenesis appeared at 12 hours and 1 day time point, along the time course (Figure 4-
2A). 0n the contrary, BG30_H08 demonstrated upregulated expression proﬁles 2 days
and 3 days post reattachment (Figure 4-2A). Moreover, the downregulation of
BG03_H04 expression started after 7 days reattachment while it was signiﬁcantly
upregulated at day 21 post reattachment. The expression proﬁles of all four of these

genes were conﬁrmed by qRT-PCR (Figure 4-2B).

The in situ expression of selected genes within the zebra mussel foot

Within the foot tissues, the expression locations of two genes, BG15_F03-DPF P
(Figure 4-3A) and BG16_H05-EGP (Figure 4-3B), were observed by using ﬂuorescence
in situ hybridization (FISH). The yellow signal in Figure 4-3A represented the stem-
forrning gland area where both beta-actin (red) and BG15_F03-DPFP (green) were
expressed. The RNA of BG15_F 03-DPF P in the thread—forming gland and plaque-
forming gland was not detected. The expression of the gene BG16_H05-BOP (labeled
with red dye) can only be observed at the tip of the zebra mussel foot where the plaque-
forming gland is located (Figure 4-3B). The strong orange in Figure 4-3B suggests that
the expression level of BG16_H05-EGP gene in the plaque-forming gland was very high
compared to the background stained by B-actin (green). The other byssus gland areas, the
stem-forming gland and thread-forming gland, did not Show signiﬁcant RNA signal of

BGl6_H05-EGP.

89

The relative expression levels of selected genes in zebra mussel tissues

The distribution of the gene expression within the zebra mussel was detected also
by using RT-PCR. The results demonstrated that the expression levels of all the seven
genes selected in this assay were extremely high in the foot tissue. Although the
transcripts of Bg91/192_E09, BG04_H04, BG15_F03, BG16_H05, and BG20_GO4 were
detected in all the other four tissue and hemocyte samples, the expression levels of these
genes in the zebra mussel foot are signiﬁcantly higher than in any of the other tissues
(Figure 4-4). The expression area of the gene MF030105_H05 and BG30_H08 was
smaller. MF030105_H05 was also Slightly expressed in the hemocyte and muscle
samples other than the foot, while the BG30_H08 can be only detected within the foot
and hemocyte. Similarly, the expression levels of these two genes outside the zebra

mussel foot were extremely low compared to that within the foot (Figure 4-4).

90

Discussion

Byssogenesis is a vital physiological activity of healthy zebra mussels to ensure
ﬁrm underwater attachment, primarily by a number of strong byssal threads with
adhesive pads. Detached mussels try to resume attachment through the production of
fresh byssal threads (Rajagopal et al. 2002). Two distinct types of byssal threads are
produced during byssogenesis: temporary and permanent threads. Within a short time
after detachment, the temporary threads are produced, followed by the generation of
permanent threads without interrupting the production of additional temporary threads.
There is also signiﬁcant morphological difference between the fresh (< 1 week) and the
aged (3 week) permanent byssal threads, due to the crosslink reaction of the foot proteins
with 3,4—dihydroxyphenylalanine (DOPA) and the melanization of the exogenous L-
DOPA with catechol oxidase (F arsad et al. 2009). It suggests that at least three stages
exist in byssogenesis: temporary byssal thread generation, permanent byssal thread
generation, and permanent byssal thread modiﬁcation. By comparing the genes
differentially expressed at the different stages of zebra mussel byssogenesis, we found
some genes that were differentially expressed at one time period only. According to Clark
and MacMahon (Clarke & McMahon 19960, b; Clarke & McMahon 1996a; Clarke 1999),
after 12 hours reattachment, most of the byssal threads are of the temporary type.
Therefore, the genes with the signiﬁcantly differentiated expression patterns on or around
day 1 post-treatment are associated with the generation of temporary byssal threads. This
may explain the drop in the number of upregulated genes following the Day 1 sampling.
Similarly, since the three-week permanent byssal threads are morphologically different

than the fresh threads of less than one week, the genes identiﬁed by microarray after 21

91

days reattachment may play certain roles in the biochemical and morphological
modiﬁcation noticed in the permanent byssal threads. The other genes that were
differentially expressed between day l and day 7 are possibly involved in the generation
and maintenance of the permanent byssal threads. Unfortunately, a large number of
identiﬁed genes do not have predictable functions due to the limited availability of
homologues in gene bank database. The putative functions and their roles in the
byssogenesis require additional analysis.

Among the identiﬁed genes with predicted functions, a number of sequences were
homologous to the Dreissena polymorpha foot protein 1 (Dpfp-1) whose expression
seems to have ﬂuctuated over our observation period. This particular protein has been
identiﬁed as a major foot protein with at least 10—15 variants which constituted a
polymorphic protein family speciﬁc to the zebra mussel (Rzepecki & Waite 1993b).
Therefore, we were not surprised with the ﬂuctuation of Dpfp-1 homologues between up-
and downregulation in our study. FISH has demonstrated that the transcription products
of a Dpfp-1 homologue had been detected in the byssal thread stem-forming gland, which
probably contributes to the protein components forming the stem of the byssal thread.
Functionally, the Dpfp-l serves as a protection or structure supporting material instead of
adhesive matrix (Anderson & Waite 1998). Based on the observation of F arsad et al
(Farsad et al. 2009), it is likely that other DOPA-rich foot proteins with adhesive
characteristics are deposited between the plaques and the hard surface to thereby
accomplish the underwater adhesion.

A large number of EGP-like encoding genes identiﬁed at all the time points in this

study may also play a role in adhesion. Structurally, the amino acid compositions of these

92

EGP-like proteins are very similar to zebra mussel Dpfp whose amino acid sequences are
dOminanted by proline, tyrosine, aspartic acid/asparagine, lysine, threonine, and glycine
residues (Anderson & Waite 1998). It structurally allows these molecules to form
crosslinks with DOPA which largely exists in the zebra mussel byssal threads and seems
to be crucial in their attachment to underwater substrates (Waite 2002). As demonstrated
in this study, several EGP-like genes were differentially expressed at all time points in
this study, suggesting their heavy involvement in byssal thread production, with up- or
downregulation representing a cascade of feedback mechanisms. Our FISH results
indicated that one of the BOP homologues was indeed found abundantly expressed in the
plaque-forming gland of zebra mussel; hence this protein is involved in zebra mussel
thread formation.

Based on the results of this study, it seems that byssal thread formation is also
associated with the differential expression of a battery of genes whose transcribed
proteins seem to serve multiple purposes. For example, our results suggest that genes
encoding neuropeptide-like proteins of Caenorhabditis elegans, known to be involved in
neuronal signaling, development and antimicrobial activities (Nathoo et al. 2001;
Couillault et al. 2004; McVeigh et al. 2008). Whether these molecules have a similar
ﬁmction in the zebra mussels and are, therefore, regulated due to their importance in
development, sensation, antimicrobial activities, is currently unknown and deserves
additional studies. Likewise, the differentially displayed zebra mussel genes whose
transcripts found no homologues in databases are likely important in byssogenesis and
related processes. Based only on signiﬁcant differential expression of genes on day 2

post-treatment, we have previously demonstrated that the zebra mussel byssogenesis

93

environment is likely affected by surrounding environmental factors such as temperature,
dissolved oxygen, and water ﬂow (Rajagopal et al. 2005).

The cDNA microarray technique has been proven to be a powerful tool to
differentiate the gene expression patterns along the time-course of a biological process
(Evans & Somero 2008; Pritchard & Nelson 2008; Suh et al. 2008), and in this case, the
zebra mussel byssus cDNA microarray was successfully applied to identify the
differentially expressed genes during the different stages of zebra mussel byssogenesis.
Given that all the genes identiﬁed in this study are indeed involved in certain stages of
the byssogenesis, the mechanism of zebra mussel underwater adhesion is still an
intriguing puzzle. What do these proteins encoded by the up or downregulated genes
exactly do during the byssal thread generation? How are they regulated by the other
components involved in this process? Are the foot proteins constitutively produced by
byssus glands or can they be induced by external stimuli? These questions will not be
answered until detailed analyses are performed on the functional genes identiﬁed by this
study. Further analyses of these genes will be a great progress in understanding the zebra

mussel attachment mechanism.

94

 

B-BGenes

 

 

 

 

 

c-BGenes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

 

 

 

 

3°°832°BF°RS°UBFFSS

Figure 4-1 The hierarchical cluster with the microarray identiﬁed genes.
The hierarchical cluster of the genes differentially expressed during the byssogenesis

based on their expression patterns at the different time points along the time-course.

27‘ldd

Twelve clusters are generated by hierarchical cluster analysis. The pink curve in each
cluster ﬁgure represents the expression proﬁles of the genes in this cluster during the

byssogenesis.

This ﬁgure is in color.

95

Day 21 I MF030105_HOS

 

 

 

 

 

I 3503_H04
Day 7 I BGlS_F03
Day 4 I BG30_H08
D 3 - ”97/192109
3" - BGZO_GO4
Day 2 - s616_Hos
Day 1
Day 0.5
0
Fold change
A
: - moaorouios
Day 21 n 3603304
Dav 7 J_ - $15103
Day 4 , .H I 3634mm
0 3 u. - seer/192_toe
——i
av .. - sozo_coc
Day 2 _-o - scrubs
Dav 1 _——i
DaVO-S —,,, . ...___.
o
o 05 1 15 2 25 3 35 4 45 5
Fold chemo
3

Figure 4-2 The comparisons of gene expression proﬁles between the microarray results
and qRT-PCR.

A. The differentially expressed genes identiﬁed by cDNA microarray and their
expression proﬁles during byssogenesis. B. The qRT-PCR results validating the relative
expression levels of the microarray identiﬁed genes during the byssogenesis. The
byssogenesis samples and non-byssogenesis reference samples were treated as described
in Materials and Methods, and the non-byssogenesis sample was used as control with
value 1. The amount of the transcripts of the gene was calculated by using 2'AAC’ model.

This ﬁgure is in color.

96

g - - - -Stem-forming
' Gland

<— - - -Thread-forming
Gland

:‘l' ' W
,. .

 

" ;‘Ih"~Plaque-forming
' Gland

 

Figure 4—3 The in situ expression of the gene BG15_F03-DPFP and BG16_H05-EGP in
zebra mussel foot tissue.

A. The synthesized antisense RNA of BG15_F03-DPF P was labeled with Alexa 488 dye
(green) while the complementary RNA for beta-actin was labeled with Alexa 594 (red). B.
The RNA probe for BGl6_H05-EGP was labeled with Alexa 594 (red) and the reference
probe, beta-actin in this case was labeled with Alexa 488 (green). The foot was cut
longitudinally as shown by the two parallel dashed lines across the foot. The positions of
the byssus glands were marked on the foot. The dashed lines connected between slides
and foot indicated the position of the three slides on the foot.

This ﬁgure is in color.

MF030105_H05-EGP 3915;030pr

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

q, 200 ‘ 10000 -
0
5 150 - 8000 .
'2
3 100 - 6000 ‘
2:, 4000 -
> 50 - + _ _
E (+) ( 1 (1 i) 2000 ' (+) (+) (+) (+)
8:). 0 r 1’ ' I —-L— ﬂ I 0 y 1 1 1 I
BGOB_HO4-Shamatrin BG15_F03—DPFP
8000 — 10000 -
8 (+)
S 6000 - f
.0 4000 - J
1:, 4000 - , -.
> _ .
E 2000 (+) (+) (+) 2000 - (+) L“i (+) (+) (+)
E 0 . . 0 . . . . ﬂ
Eels-H05‘EGP BGZO_GO4-ANXN
0 2500 - (+) 8000 u
g 2000 - f" (+)
“o _ 6000 - -
g 1500 - 1 FL!
.0
i 1000 - . 4°00 ’ 1i i
.2 ' . '
g 500 (+) (+) (+) (+) 2000 (+) U (+) (+) (+)
m 0 I I. r — ’ r 1 0 I T I - I 1
Hemocyte Foot Muscle Ctenidium Mantle
BG30_H08-PRF
0 150 -
o
I:
8
c 100 -
3
Q
g 50 "
g (+) (-i H <-)
a, 0 I I I l I
m

 

Hemocyte Foot Muscle Ctenidium Mantle

Figure 4-4 The distribution of the mRNA products of the selected genes within zebra
mussel tissues.

(+) indicated the existence of the gene in the tissue; (-) suggested no detected transcripts
in the tissue. For each gene, the relative expression levels of the gene in other tissues

were calculated by using 2-AACt model.

98

Table 4-1 The gene speciﬁc primers used for qRT-PCR.

 

Target gene

Name of the primer

Sequence

 

MF030105_H05

BG97/l92_E09

BGO3_H04

BGlS_F03

BGl6_H05

BG20_GO4

BG30_H08

*rSSrRNA

MF030105_H05_EG P-RT-Fw
MF030105_H05_EGP-RT-Rv
BG97/ l 92_E09_DPFP-RT-Fw
BG97/ 192_EO9_DPFP-RT-Rv
BG03_H04_Smtrn—RT-Fw
BG03_H04_Smtrn-RT-RV
BG15_F03_DPFP-RT-Fw
BG15_F03_DPFP-RT-Rv

BG 1 6_H05_EGP-RT- FW

30 l 6_H05_EGP-RT—Rv
BGZO_GO4_1RCl—RT-Fw
BG20_GO4_IRCI-RT-Rv
BG30_H08_PRF-RT-Fw
BG30_HO8_PRF-RT-Rv

1 8SrRNA_Forward

18SrRN A_Reverse

5’ -TCC GCC ACC ATA GTC GTC AT- 3’

5’ -GGT AGC GAA CGC CCA AAA C- 3’

5’ -CCG ATG GGC CAT ATG ATA AGA A- 3’
5’ -ACG TCG ATG CCT TGC TTG TC- 3’

5’ -TGG CGG GTA TCC AGG AAA- 3’

5’ -CCG TAA CCA CCC CAT TTG C- 3’

5’ -TCG ATG CCT TGC TTG TCT TG- 3’

5’ -ACA CCG ATG GGC CAT ATG AT- 3’

5’ -CGA TTG CTC CGG TAG CAA AC- 3’

5’ -TAT AAC CAT CCG CCA CCA TAG TT- 3’
5’ -CGG CGA CCA AGA AAA ATA CG- 3’

5’ -GAC CCA CCA ACG GCA TTC- 3’

5’ -GCG GCG TTG CGT TGA G- 3’

5’ -CAT GGT CCA GGC CAA TAG GA- 3’

5’- GAC ACG GCT ACC ACA TCC AA -3’

5’- CTC GAA AGA GTC CCG CAT TG -3’

 

* The primers used as internal reference for QRT-PCR.

99

Table 4-2 The genes that were differentially expressed after 12-hour reattachment.
Selected by t-test P < 0.05, when Log (PC) > 0, the gene was upregulated during the

byssogenesis; otherwise was downregulated.

 

 

Accession # Gene [D P Log(FC) Homologues

AM230194* BG28_F04 0.00235 - l .03 N/A

AM230436* BG29_A08 0.00548 0.362 CAA55094| Cytochrome oxrdase Ill
[Kellra Iaperousu]

at BG20_GO4 0.00783 0.925 XP_001134471| Annexin VII

AM230139 [Dictyostelium discoideum AX4]

AM229777* BG31_H08 0.00875 0432 AAV80789| Exocrine gland peptide
[Ixodes scapularis]

AM229723 BG32_C03 0.01453 -0.334 AAC39039| Foot protein 1 precursor
[Dreissena polymorpha]

AM229947 BG26__E11 0.01684 0.717 N P__504107| Neuropeptide-like protein 31
[Caenorhabditis elegans]

AM230020 MF030105_H05 0.02056 -0.637 AAV80789| Exocrine gland peptide
[Ixodes scapularis]

AM229916 BG13_E10 0.02386 0.429 AAT921 1 1| Exocrine gland peptide NFL-2
[Ixodes paciﬁczrs]

AM230391 BG17_E04 0.02853 0.275 N/A

AM230219 B012_BIO 0.02952 0.647 N/A

AM230428 BG24_C06 0.0342 0.345 N/A

AM230236 BG15_C08 0.0445 0.33 N/A

AM230086 BG10_C01 0.04607 0.364 BAE93436| Shematrin-4 [Pinctadafucata]

 

* Differentially expressed genes with t-test P < 0.01

100

Table 4-3 The genes that were differentially expressed after l-day reattachment.
Selected by the t-test P < 0.05, when Log (F C) > 0, the gene was upregulated during the
byssogenesis; otherwise was downregulated.

 

 

Accession # Gene ID P Log(FC) Homologues

AM230212* BG10_A09 0.00135 -0.51 1 N/A

AM229999* BG12_D08 0.00312 0.335 AAV80789I Exocrine gland peptide
[Ixodes scapularrs]

AM230238 BG34_F07 0.01997 0.657 N/A

AM230370 BG 1 7_D01 0.02299 -0.503 N/A

AM229820 BG12_A09 0.02401 0.307 AAT92111| Exocrine gland peptide NFL-2
[Ixodes paciﬁc-us]

AM230030 BG33_E04 0.02495 0.296 N/A

AM229726 BG97/192_E09 0.02747 0.685 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

AM230280 BG97/ 192_H07 0.03323 0.428 N/A

AM230020 MF030105_H05 0.0338 -0.505 AAV80789| Exocrine gland peptide
[Ixodes scapularis]

AM230222 BG21_A07 0.03851 -0.196 N/A

AM230177 BG09_C10 0.0396 0.253 P27085| Ribosomal protein 826
[Octopus vulgaris]

AM229905 BGlS_E06 0.04041 0.432 AAV80789| Exocrine gland peptide
[lxoa'es scapularr's]

AM230398 BG18_C01 0.04853 -0.2 N/ A

 

* Differentially expressed genes with t-test P < 0.01

101

Table 4-4 The genes that were differentially expressed after 2-day reattachment.
Selected by the Host P < 0.05, when Log (F C) > 0, the gene was upregulated during the

byssogenesis; otherwise was downregulated.

 

 

Accession # Gene ID P Log(FC) Homologues

AM230170* BG14_A11 0.00099 -0.62 A3152762| 60$ ribosomal protein L27
[Argas monolakenszs]

AM230268* MF030105_B 12 0.0031 1 -0.862 N/A

AM230436* BG29_A08 0.00836 -0.291 CAA55094| Cytochrome oxidase 111
[Kc/Ira laperousu]

AM230156 BG30_H08 0.0103 0.477 BAB12683| Polypeptide release factor 3
[Yarrowia lipolytica]

AM230135 BG07_E12 0.01653 -0.33 BAE93433| Shematrin-l
[Pinctada ﬁrcata]

AM229907 BG33_D04 0.01699 -0.295 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM229915 BG10_B03 0.01749 0.403 N/A

AM230168 BG16_D03 0.01875 -0.29 AAN05585| Ribosomal protein L22
[A rgopecten irradians]

AM230262 BG29_D12 0.02504 -0.626 N/A

AM230201 BG97-192_C04 0.0251 -0.324 N/A

AM230279 BG33_BO6 0.02726 -0.392 N/A

AM230213 BG03_E10 0.02799 0.387 N/A

AM230299 BG05_A01 0.02951 -0.382 N/A

AM230428 BG24_C06 0.03067 -0.301 N/A

AM229867 BGl6_BOl 0.03153 -0.275 AAV80789| Exocrine gland peptide
[Ixodes scapularis]

AM230089 BG20_F04 0.0351 1 -0.376 N/A

AM229837 BG97/192_G04 0.03959 -0.339 AAT92111| Exocrine gland peptide NPL—2
[Ixodes paciﬁcus]

AM230302 BG05_D06 0.04032 -0.255 N/A

AM230270 MF030105_D06 0.04264 -0.264 N/A

AM229724 BG07_H12 0.04388 -0.269 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

 

* Differentially expressed genes with Host P < 0.01

102

Table 4-5 The genes that were differentially expressed after 3-day reattachment.
Selected by the t—test P < 0.05, when Log (FC) > 0, the gene was upregulated during the
byssogenesis; otherwise was downregulated.

 

 

Accession # Gene ID P Log(FC) Homologues

AM230436* BG29_A08 0.00248 -0.341 CAA55094| Cytochrome oxidase III
[Kellia laperousii]

AM230391* BG17_E04 0.00254 -0.339 N/A

AM230270* MF030105_D06 0.00562 -0.375 N/A

AM229957 BG26_BO9 0.01 178 -0.322 N/A

AM230080 BG23_C05 0.01255 -0.314 CAA10192| Glycine-rich protein 2
[C icer arietinum]

AM230258 BG16_G06 0.01553 -0.518 With anaphylatoxin domain (P801177) and
EGF-like region (P800022)

AM229881 BG97/l92_812 0.02036 -0.558 N/A

AM230428 BG24_C06 0.02568 -0.31 1 N/A

AM230440 BG3 1_F01 0.02893 -0.281 N/A

AM230145 BG28_F03 0.03174 -0.197 N/A

AM230214 BG16_G04 0.03261 0.242 N/A

AM230387 BG17_D06 0.03391 -0.212 N/A

AM230156 BG30_H08 0.03449 0.382 BABIZ683| Polypeptide release factor 3
[Yarrowia lipolyrica]

AM230306 BG28_E01 0.04006 -0.227 N/A

AM230220 BG08_BOl 0.04406 -0.285 N/A

AM229950 MF030105__E03 0.04621 -0.404 NP_504107| Neuropeptide-like protein 31
[C aenorhabditis elegans]

AM229946 BGl 1_E06 0.04767 0.339 AAV80789| Exocrine gland peptide

[lxodes scapularis]

* Differentially expressed genes with t-test P < 0.01

103

 

Table 4-6 The genes that were differentially expressed after 4-day reattachment.
Selected by the Host P < 0.05, when Log (F C) > 0, the gene was upregulated during the

byssogenesis; otherwise Was downregulated.

 

 

 

Accession # Gene [D P Log(FC) Homologues
AM229866* BGZ8_E10 0.00716 0.321 AAV80789| Exocrine gland peptide
[Ixodes scapularis]
AM229864* BG16_H05 0.00983 -0.775 AAV80789| Exocrine gland peptide
[Ixodes scapularrs]
AM230217 BG11_F12 0.01064 0.31 1 N/A
AM230230 BGlS_A03 0.01095 0.317 N/A F
AM230173 BG22_A12 0.0148 0.464 ABR23471| 4OS ribosomal protein S11
[Ornithodoros parkeri]
AM229737 BG32_C03 0.01774 0.276 AAC39039| Foot protein 1 precursor
[Dreissena polymorpha]
AM229913 BG3 1_D01 0.02009 —0.459 NP_505834| Neuropeptide-like protein 33
[Caenorhabditis elegans] L
AM230212 BG10_A09 0.02227 0.335 N/A "
AM230209 BG08_E 12 0.02575 0.256 N/A
AM230170 BG14_A11 0.02715 -0.3 88 ABISZ762| 6OS ribosomal protein L27
[A rgas monolakensis]
AM229738 BG15_F03 0.03366 0.254 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]
AM230148 BG09_D03 0.03713 0.24 N/A
AM230094 BG33_H08 0.03736 -0.433 BAA76901| Choriogenin Hminor
[Oryzias latipes]
AM230064 BG29_BO7 0.04237 -0.439 BAE93436| Shematrin-4
' [Pinctadafucata]
AM230258 BG16_GO6 0.04583 -0.42 With anaphylatoxin domain (PS01 177) and
EGF-like region (PS00022)
AM229789 BG06_C03 0.04902 0.332 N/A

 

* Differentially expressed genes with t-test P < 0.01

104

Table 4-7 The genes that were differentially expressed after 7-day reattachment.

Selected by the Host P < 0.05, when Log (F C) > 0, the gene was upregulated during the

byssogenesis; otherwise was downregulated.

 

 

Accession # Gene ID P Log(FC) Homologues
* BG16 G06 0.00027 -0.876 With ana h latoxin domain P801177 and

AM230258 — EGF-likepreilgion (P800022) ( )

AM229950* MF030105_E03 0.00129 -0.721 NP_504107| Neuropeptrde—lrke protein 31
[C aenorhabdms elegans]

AM229748* BG31_E04 0.0031 0.817 N/A

AM229864* BG16_H05 0.00944 —0.798 AAV80789| Exocrine gland peptide
[Ixodes scapularls]

AM229810 BGZ9_D08 0.01067 0.629 AAT92111| Exocrine gland peptide NFL-2
[Ixodes paciﬁcus]

AM230230 BG15_A03 0.01075 0.325 N/A

AM230148 BG09_D03 0.01202 0.302 N/A

AM229949 BG05_C08 0.0221 0.523 N/A

AM230173 BG22_A12 0.02354 0.437 ABR23471| 4OS ribosomal protein S1 1
[Ornithodoros parkeri]

AM229798 BG06_C03 0.0271 0.3 85 N/A

AM230036 BG03_H04 0.02725 —0.379 BAE93436.1| Shematrin—4
[Pinctadaﬁrcata]

AM230106 BG05__C07 0.03377 0.442 N/A

AM230377 BG16_D10 0.03458 0.973 N/A

AM230211 BG05_D02 0.03577 -0.449 N/A

AM230407 BGl8~_G06 0.04034 0.778 N/A

AM229725 BG12_C01 0.04359 0.175 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

AM229736 BG32_C03 0.04386 0.236 AAC39039| Foot protein 1 precursor
[Dreissena polymorpha]

AM230404 BG18_F03 0.04685 -0.548 N/A

AM230344 BG12_G06 0.04945 -0.44 N/A

AM229727 BGIS_H10 0.0499 -0.668 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

 

* Differentially expressed genes with t—test P < 0.01

105

 

Table 4-8 The genes that were differentially expressed after 21-day reattachment.
Selected by the t-test P < 0.05, when Log (F C) > 0, the gene was upregulated during the

byssogenesis; otherwise was downregulated.

 

 

 

Accession # Gene ID P Log(FC) Homologues

AM230363* BG 1 5_D08 0.00051 -0.697 N/A

AM229907* BG33_D04 0.00192 0.47 AAV80789| Exocrine gland peptide
[lxodes scapular/s]

AM230179* BG08_BO3 0.00247 0.792 N/A

AM229753* BGZ3_D01 0.00559 0.553 AAT92111|.Exocrine gland peptide NPL-2 g:-
[lxodes pacrﬁcus]

AM230251* BG29_BO9 0.00668 -0.823 N/A

‘ AM230111* BG30_G02 0.00835 -0.749 AAN10061| Kunitz-like protease inhibitor

[A ncylostoma canmum]

AM230206 BGZ9_BOS 0.01314 -0.735 N/A L

AM230302 BG05_D06 0.01414 0.3 64 N/A —

AM230408 BG97/192_E03 0.01813 -0.592 N/A

AM502285 BG09_F07 0.01987 -0.642 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

AM230248 BGZS_H1 1 0.01997 0.398 With Antifreezei domain and
Thiolase domain

AM230143 BG29_F09 0.02004 0.529 N/A

AM230253 BG34_F09 0.02196 -0.392 N/A

AM229730 BG05_D07 0.02324 0.517 N/A

AM229728 BG21_E12 0.02548 -0.586 AAF75279| Byssal protein Dpfpl precursor
[Dreissena polymorpha]

AM230344 BG12_GO6 0.02555 0.58 N/A

AM229901 BG15_E03 0.02856 -0.815 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM230390 BG33_C06 0.03041 0.425 N/A

AM230036 BG03_H04 0.03334 0.416 BAE93436| Shematrin-4
[Pinctadaﬁrcata]

AM229892 BG32_F04 0.03801 0.898 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM230172 BG26_BOI 0.03833 -0.394 AAK95191| 408 ribosomal protein S9
[Ictalurus punctatus]

AM230048 BG16_E06 0.04033 0.689 N/A

AM229879 BG32_F05 0.04249 -0.427 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM229876 BG23_F02 0.04295 0.363 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM229878 BG34_H02 0.04315 -0.573 AAV80789| Exocrine gland peptide
[lxodes scapularis]

AM229802 BG22_H06 0.04316 0.381 N/A

AM230095 BGl6_C02 0.04396 -0.408 BAE93436.1| Shematrin-4
[Pinctadaﬁicata]

AM230123 BG12_G10 0.04409 -0.545 N/A

AM229752 BG33_BOZ 0.04517 0.453 AAT92111| Exocrine gland peptide NPL-2
[lxodespaciﬁcuﬂ

 

* Differentially expressed genes with t-test P < 0.01

CHAPTER V

Factorial Microarray Analysis of Zebra Mussel
(Dreissena polymorpha) Adhesion

Abstract

The zebra mussel (Dreissena polymorpha) has been well known for its expertise in

attaching to substances under the water. Studies in past decades on this underwater

 

adhesion focused on the adhesive protein isolated from the byssogenesis apparatus of the
zebra mussel. However, the mechanism of the initiation, maintenance, and determination
of the attachment process remains largely unknown. In this study, we used a zebra mussel
cDNA microarray previously developed in our lab and a factorial analysis to identify the
genes that were involved in response to the changes of four factors: temperature (Factor
A), current velocity (Factor B), dissolved oxygen (Factor C), and byssogenesis status
(Factor D). One hundred and seventeen probes in the microarray were found to be altered
by at least one of the factors, while 18 genes were modulated by two of them. The
transcription products of four selected genes, DPFP-BG20_A01, EGP-BG97/192_BO6,
EGP-BG13_G05, and NH—BGI7_C09 were unique to the zebra mussel foot based on the
results of quantitative reverse transcription PCR (qRT-PCR). The expression proﬁles of
these four genes under the attachment and non-attachment were also conﬁrmed by qRT-
PCR and the result is accordant to that from microarray assay. The in situ hybridization
with the RNA probes of two identiﬁed genes DPFP-BGZO_A01 and EGP-BG97/192_BO6
indicated that both of them were expressed by a type of exocrine gland cell located in the

middle part of the zebra mussel foot. It turns out that the factorial design and analysis of

107

the microarray experiment is a reliable method to identify the inﬂuence of multiple
factors on the expression proﬁles of the probesets in the microarray; therein it provides a

powerful tool to reveal the mechanism of zebra mussel underwater attachment.

108

 

in1

It;

1111
an
of
Se

all

51‘

Us

Fa

Introduction

The Eurasian, non-native mollusk, the zebra mussel (Dreissena polymorpha),
invaded the Laurentian Great Lakes in the 19805 and has expanded since then to other
regions in North America (Hebert et al. 1989; Benson & Raikow 2009). The spread of
the zebra mussel has been followed by ecologic and economic devastation of
unprecedented magnitude (N alepa & Schloesser 1993). Through their unique ability to
attach to underwater surfaces, the invading mussels interfered with navigation and ﬂow
of industrial efﬂuents, a problem that has caused billions of dollars in losses (N ew-York-
Sea—Grant 1994b, a). Moreover, the selective ﬁltering capacity of the zebra mussel has
altered the microbial communities in the Great Lakes to the extent that some resident
Species have become endangered and others are facing extinction (N alepa & Schloesser
1993). The problem is compounded by the fact that all control methods used to minimize
the zebra mussel invasion impacts have been unsuccessful (Snyder et al. 1997; Molloy
1998; Bially & Maclsaac 2000). As a result, a pressing need has emerged to unravel the
strategies used by the invading mollusk to survive and mechanisms governing its ﬁrm
attachment to underwater surfaces.

The zebra mussel is unique among freshwater bivalves in that in the complete life
cycle it keeps its byssus, a group of exocrine glands embedded in the mussel’s foot that
secretes threads with attachment pads through byssogenesis (Rzepecki & Waite 1993a, b).
Our previous studies have identiﬁed 716 genes that are unique to zebra mussel byssus
using the suppression subtractive hybridization (SSH) cDNA library technique (Xu &

Faisal 2008). Besides the classic D. polymorpha foot protein -1 and -2 (Dpfp-1 and -2), a

109

number of molecules with different putative functions have also been produced by byssus
glands, such as the Exocrine gland peptides homologous to the salivary gland peptides of
Ixodes spp. (Francischetti et al. 2005), host defense related molecules, and the proteins
without predictable functions. The subsequently developed cDNA microarray with these
716 unique genes was used to compare the expression proﬁles of the gland unique genes
under different attachment status. The results indicated that the expressions of 52 genes

are either up or down regulated in the attachment status (Xu & Faisal 2009a). It is

 

suggested that the microarray technique can be used as a high-throughput tool to study L
the mechanism of zebra mussel attachment at the molecular level.

Zebra mussel attachment is inﬂuenced by several environmental factors including
water temperature (Clarke & McMahon 19960), dissolved oxygen level (Clarke &
McMahon 1996b), and current velocity (Clarke & McMahon 1996a). The mechanism(s)
by which these factors inﬂuence zebra mussel attachment is currently unknown and
involves a large number of molecules. Identifying the inﬂuence of environmental
conditions on gene encoding is rather complicated and may require a novel way of
computation and statistical analysis. In this context, the studies of retinal development in
zebra ﬁsh (Danio rerio) have adapted the multifactorial analysis to cDNA microarray
data and proved useful in deciphering the complicated developmental process (Leung et
al. 2008). The factorial microarray analysis allows us to analyze the effects of more than
one independent variable (the main factors involved in this study) on a dependent
variable (the differentiation of gene expression) (Wu & Hamada 2000; Montgomery
2005). In this study, a non-reference loop experimental design was performed to test the

effects of multiple environmental factors on the expression proﬁles of the byssus unique

110

genes. This experimental design has been reported as a reliable design that leads to less
disparity in precision and power comparisons between any two groups (Tempelman
2005)

The purpose of the factorial design that we are going to describe here is to
identify the inﬂuence of the temperature, water current velocity, dissolved oxygen, and
byssogenesis status on the expression proﬁles of the molecules unique to zebra mussel
byssus glands to thereby to better understand the mechanism of zebra mussel attachment

under the complex underwater environment.

111

Materials and Methods

Zebra mussel collection and maintenance

Zebra mussels used in this study were collected from Vineyard Lake in Brooklyn,
MI, USA (Latitude: 42°4’59”N; Longitude: 84°12’34”W). The mussels were thoroughly
cleaned and rinsed with deionized water several times before they were allowed to
acclimate in aerated, ﬁltered Vineyard Lake water for eight weeks. The mussels were ‘
kept in a glass tank and fed weekly with a pure culture of the algus Ankistrodesmus

falcatus.

Treatments of zebra mussels and experimental design
Three environmental factors were involved in this study including temperature
(Factor A), current velocity (Factor B), and dissolved oxygen level (Factor C). The status

of byssogenesis was considered as Factor D. The model used in this study is:

yg=pg+A+B+C+D+e

This one way model includes only the basal expression value of the probe (,ug), the main

effects coefﬁcients (A, B, C, D), and the error term (a). In each effect, two levels were
created. To create two different attachment statuses of the mussels, we severed the byssal
threads from randomly selected mussels and put them back into the tanks in different
orientations. The mussels lying on the ventral side of their shells for 48 hours were
considered to be attached individuals with byssogenic activity (A) while the ones with
ventral sides facing up for 48 hours were taken as detached mussels without byssogenesis

(D). Two temperature levels were used in this study: room temperature (R) and a low

112

temperature level at 4 °C. The two dissolved oxygen (D. 0.) levels used were the normal
D. 0. level (N), which was maintained by airstones at about 10 mg/L, and the lower D. 0.
level, which was about 5 mg/L without airstones. The current of the water was performed
by using VWR 371 Hotplate/Stirrer (VWR International LLC, West Chester, PA) and a
magnetic stirring bar with a stirring Speed of 60 rpm (F). The static water was used as
non-current condition of Factor B in the experiment (S).

Sixteen treatment combinations were obtained from this 2X2><2><2 factorial
experimental design. To maintain high statistical power for the analyses of each factor
effect, a loop experimental design was performed with four biological replicates for each
treatment (Figure 5-1) (Tempelman 2005). Every two samples connected with an arrow
were used to hybridize with one microarray slide. The dye labeling and hybridizations are
indicated by Figure 5-1. Four null hypotheses were tested and the interpretation of the

rejection of each null hypothesis is listed in Table 5-1.

RNA extraction and cDNA synthesis

The total RNAS used for microarray hybridization were extracted from pools of
zebra mussel feet with six individuals in each pool. The total RNA puriﬁcation kit, 5-
PRIME PerfectPure RNA Tissue Kit (5 PRIME Inc, Gaithersburg, MD) was used for
RNA puriﬁcation and the protocol followed the manufacturer’s instructions. The RN As
were then reverse transcribed to single strand cDNAs by using SuperScript Plus Indirect
cDNA Labeling Kit (Invitrogen, Carlsbad, CA) containing Alexa F luor 555 & 647
ﬂuorescent dyes. The cDNA syntheses and dye labeling procedures were all followed as

described by the kit’s instructions.

113

Microarray hybridization and data analyses

The zebra mussel byssus cDNA microarray was developed and printed in the
Center for Animal Functional Genomics, Michigan State University (Xu & Faisal 2009a).
The GeneTAC Hbetation (Genomic Solutions) was used as the hybridizer for all
hybridizations and an l8-hour step-down protocol was applied as described in our
previous study (Xu & Faisal 2009a). The hybridized dual—channel array slides were
scanned by GenePix 40003 two-laser Scanner (Molecular Devices, Downingtown, PA),
and GenePix Pro 6.0 (Molecular Devices) software was then used for image processing
and spot intensity ﬁle generation. The spot density ﬁles were analyzed with the Limma
software package (Smyth 2005). The preliminary data processing includes background
correction, within array normalization, and between array normalization. The
differentially expressed genes were determined by least square regression. The
coefﬁcients of the analysis were the logarithmic values of the ratios between two levels in
each factor. The homologues of the genes are listed with the accession numbers and
putative functions, as well as with the names of the species (Table 2-5). The up- or down-
regulation of the factor to probes was decided by the coefﬁcient. The hierarchical
clustering was done with the soﬁware Genesis (Sturn et al. 2002). The dataset of this
microarray study was deposited in Gene Expression Omnibus (GEO) with the series
accession number GSE16397. Some of the identiﬁed genes with similar primary
structures were differentiated by multiple sequence alignment with ClustalW (Thompson

et al. 1994).

114

Microarray data validation

The one-step quantitative reverse transcription PCR (qRT-PCR) was performed

with Power SYBR® Green RNA-to-CTTM I-Step Kit (Applied Biosystems Inc., Foster

City, CA) to identify the distribution of the candidate genes in zebra mussel tissues. The
mussels were also collected from Vineyard Lake in Brooklyn, MI. The shells of
randomly selected individuals were swabbed with 70% ethanol, followed by a soak in
150 m1 sterilized double-distilled water containing 5,000 U penicillin, 5 mg streptomycin,
and 10 mg gentamicin for 30 minutes (Davids & Yoshino 1998). Thereafter, the mussels
were transferred to sterilized double distilled water. The RNA samples were extracted
from hemocytes, feet, muscles, ctenidia, and mantles of zebra mussels. The protocol for
hemocyte collection was described by our previous studies (Xu & Faisal 2007, 2009b)

and the other samples were taken by sterilized scissors and scalpels. The primers were
. . TM . . . . .
desrgned by Primer Express (Applied Biosystem Inc.). The primers used In this study

are listed in Table 6.

The qRT-PCR was also applied to validate the expression proﬁles of the genes
that were affected by the attachment status identiﬁed by microarray. The zebra mussels
were allowed to fully attach themselves on glass Petri-dishes for six weeks. Then byssal
threads from all individuals were cut off with scalpels; thereafter, ﬁve mussels were
dissected and the RNA extracted from their feet was considered as group 0. The rest of
the mussels were put back in the Petri-dishes in water, and allowed to regenerate their
byssal threads. The RNA samples were extracted from the feet of the mussels at 12 hours,
one day, two days, and three days post re-attachment. Two biological replicates were
used for each group. Then for each candidate gene, the qRT-PCR was performed with

115

these ﬁve group samples. The information of the gene speciﬁc RT Primers are listed in

Table 6.

RNA ﬂuorescence in situ hybridization (FISH)

In order to locate the transcripts of two selected genes in situ, ﬂuorescence in situ
hybridization (FISH) was used. The zebra mussel foot tissue was ﬁxed in 2%
paraformaldehyde phosphate buffer saline (0.02M NaH2P04, 0.0077M NazHP04, 1.4M
NaCl, 2% w/v paraformaldehyde, pH 8.0) for ﬁve hours at room temperature. The sample
was then conserved in absolute ethanol until needed. The conserved sample was buried in
parafﬁn and longitudinally sectioned at the same level. The ﬁve pm thick sectioned
sample was obtained and put on an RN ase- and DNase-free glass slide. Ten slides were
obtained with one stained by H&E method.

The antisense RNA strand of a fragment of each gene was produced. Brieﬂy, the
PCR product of the gene fragment was cloned into pGEM®-T Easy Vector Systems
(Promega U.S., Madison, WI). Then the positive colony was picked for sequencing with
the M13 forward primer to clarify the direction of the insert of the recombinant plasmid.
Once the direction of the insert was known, the T7 or Sp6 RNA polymerase was selected
to synthesize the antisense RNA with the linearized recombinant plasmid. In this study,
T7 polymerase was used on both probes. The probe DPFP-BG20_A01 was labeled with
ﬂuorescent dye Alexa 488 (Invitrogen) while the probe EGP-BG97/192_B06 was labeled

with Alexa 594 (Invitrogen). The steps of the RNA synthesis and labeling can be found in

the instructions of the two kits: FISH TagTM RNA Green Kit (Invitrogen) and Fish

TagTM RNA Red Kit (Invitrogen). The RNA FISH was performed following the

116

recommended protocol in the appendix of the Fish TagTM RNA Kit instruction. The

hybridization buffer mentioned in the protocol was replaced by ULTRAhbeM
Ultrasensitive Hybridization Buffer (Applied Biosystems/Ambion, Austin, TX). The
hybridized slides were visualized by Olympus BX41 (Olympus America Inc., Center
Valley, PA) microscope under the excitation of a mercury lamp and the images were
captured by Olympus DP25 digital camera (Olympus America Inc.). The image
processing software DP2-BSW (Olympus America Inc.) was used to combine the

different color channels into a Single image.

117

Results

Overview of the factorial analysis results

Microarray data analyses suggest that each of the factors tested (temperature,
stirring, dissolved oxygen level and accessibility to attachment surfaces) had altered the
expression Of zebra mussel foot genes. The genes whose expressions had been altered by
each of the four factors are listed in Tables 2-5. While the gene expression proﬁles had
been altered, however, the type and numbers of differentially expressed genes differed
with each experimental factor used. At a 95% conﬁdence level (P <0.05), 73 genes (10%
of the total genes on the Slide) were differentially expressed by the accessibility of the
mussel to attachment surfaces (Table 2). At the same conﬁdence level, 27 genes (4% of
the total number of genes) were found either up- or down-regulated with the temperature
increased from 4 °C to the room temperature of 22 °C (Table 3). Twenty-six of the genes
(4% of the total genes) were modulated by the level of dissolved oxygen (D0), which
was close to that in the temperature group (Table 4). Surprisingly, only nine genes, 1% of
the whole probesets on the array, were identiﬁed with their expression levels inﬂuenced
by the current velocity (Table 5). With the P values set at 0.01, fewer probes were
identiﬁed. Eleven probes had their expression proﬁles altered by the status of attachment
(Table 2). Five genes demonstrated signiﬁcantly different expression levels with the
temperature change (Table 3). With the level change, each of the other two
environmental factors, DO. and current velocity, modulated the transcription proﬁles of

two genes only (Table 4 and 5).

118

Ten major clusters were generated using the hierarchical clustering method on the
probesets whose expression proﬁles were modulated by at least one of the factors (Figure
5-2). The genes in each cluster had similar expression patterns under the effects of the
four factors. For instance, three genes had been grouped into cluster A, suggesting that
the expression proﬁles of the three genes were signiﬁcantly upregulated when the
temperature was decreased or when the mussel was undergoing byssogenesis. Similarly,
cluster B contained seven genes whose expression patterns were upregulated by water
ﬂow, through were downregulated by byssogenesis (Figure 5-2).

Among the genes identiﬁed by microarray analysis, some genes demonstrated
dramatic expression level changes in response to the combination of more than one factor
(Figure 5-3). For example, from the 59 genes affected by the the attachment status
(Factor D), six genes responded to changes in temperature (Factor A); another set of six
genes were altered by dissolved oxygen levels (Factor C); and two other genes were
affected by water current velocity (Factor B). Similarly, two genes were regulated by
both Factor C and Factor D, while only one gene was affected by the combination of
Factor B and C or Factor B and D (Figure 5-3).

Among the genes that were altered by byssogenesis at the P < 0.01, ﬁve were
homologous to invertebrate exocrine gland peptides (GenBank accession number
AAV80789 and AAS92593), while three genes from those modulated by temperature
were homologous to two other invertebrate exocrine gland peptides (Accession numbers
AAV80789 and AAT92111). The multiple sequences alignment suggested that the ﬁve
genes in Table 2 whose expression was altered at P < 0.01 were not identical. BG31_H01

and BG10_C05 shared some Similarities while BG13_F10 and BG23_BO3 were

119

 

structurally close to each other. The sequence BG97/ 192_BO6 Shared the least similarity
to the other four genes in nucleotide sequences (Figure 5-4). Similarly, among the three
genes from the temperature modulated gene group, BG23_D02 and BG03_BOI were
closest in their primary structure and both of them were downregulated in response to the
lower temperature. On the contrary, EST, BG27_808 had its expression proﬁle

upregulated at low temperature (Figure 5-5).

Validation

Four genes that were signiﬁcantly altered by byssogenesis status were selected for
validation (Table 6). The qRT-PCR with RNA samples from ﬁve tissues of zebra mussels
showed that the abundance of all four selected genes in the zebra mussel foot were
extremely high as compared to other tissues. The RNA product of an excocrine gland
peptide-like gene (EGP-BG97/ 192_BO6) was detected in all selected samples; however,
the expression level in the foot was signiﬁcantly higher than in any of the others. The
lowest expression level of EGP-BG97/192_BO6 was in the ctenidium, which was 17399
i- 844 times less than that in the foot. Although the quantity of the gene EGP-
BG97/192_B06 transcripts in the muscle was also higher than that in the ctenidium, the
foot tissue produced 15 i 1 times more gene product than the ctenidium cells. Another
EGP gene, EGP-BG13_G05, was exclusively expressed in the foot and hemocyte while
the ratio between expression levels in foot and hemocyte was 954 i 56. The product of a
Dpfp-like protein encoded gene, DPFP-BG20_A01 was detected in foot, muscle, and
hemocyte samples. The abundance of the gene product in the foot was the highest, at 383
i 63 times that in the hemocyte and 49 d: 8 times more than that in the muscle. Another

gene that was expressed in all ﬁve tissues was a gene without known homologues in

120

GenBank database, NH-BG17_C09. Compared to the least amount of RNA in the
ctenidium, the foot provided 87 3: 3 times more NH-BG17_C09 product (Figure 5-6).
The comparisons of the expression levels Of four selected genes between attached and
detached status were also performed by qRT-PCR. As shown in Figure 5-7, compared to
the detached mussels (D), the gene DPFP-BG20_A01 had signiﬁcantly higher expression
levels in attached groups (A) at 12 hours, 1 day, 2 days, and 3 days post-treatment. The
most signiﬁcant difference between A and D was shown 2 days post-treatment. The two
EGP-like protein encoding genes, EGP-BG97/ 192_BO6 and EGP-BG13_G05,
demonstrated similar expression proﬁles during the byssogenesis; both were upregulated
in attachment status at 1 day and 2 days post-treatment, while down-regulated at 12 hours
and 3 days post-treatment. However, the most obvious difference between A and D
appeared at Day 1 in gene EGP-BG97/192_B06 and one day later in gene EGP-
BG13_G05. At most time points, gene NH-BG17_C09 acted as downregulated in group
A. Only at Day 2, the expression level of this gene demonstrated higher level in the
attached status than the detached. All four candidate genes demonstrated the same

expression proﬁles to those identiﬁed by microarray assay.

RNA in situ hybridization

Three major byssus gland cells were observed from longitudinally sectioned zebra
mussel foot with H&E stain; namely, stem-forming gland (SFG) cell, thread-forming
gland (TF G) cell, and plaque-forming gland (PFG) cell (Figure 5-8). The SFG was
located at the root of the foot containing the cells with the smallest size among the three
types of gland cells (Figure 5-8a). The TF G cells were distributed in the middle section of

the foot, along the epithelial cells on the surface. The color of the TFG cells was lighter

121

 

than the other two types of gland cells, and the size of TF G cells was larger than SF G
cells but smaller than PFG cells (Figure 5-8b). The cells of PFG had the largest size but
the amount of this type of cells was the least and all the PFG cells were embedded in a
small area of the foot tip (Figure 5-80).

The in situ hybridization results indicated that the transcription of both genes
DPFP-BG20_A01 (Alexa 488 labeled) and EGP-BG97/ 192_BO6 (Alexa 594 labeled)
were not detectable at the foot root or tip sections (Figure 5-9 a and 0); however, strong
signals of DPFP-BG20_A01 mRNA and EGP-BG97/192_B06 mRNA were detected in
the middle area of the foot containing the TFG cells. The cells that expressed DPFP-
BG20_A01 and those that contain EGP-BG97/192_BO6 mRNAs were largely overlapped,
demonstrating yellow signal. Some of the TFG cells that exclusively expressed either
DPFP-BG20_A01 (green) or EGP-BG97/ 192_BO6 (red) were also observed (Figure 5-

9b).

122

 

Discussion

The factorial analysis used in this study was designed as three enclosed loops
encompassing four biological replicates for each treatment. This design allowed the
estimation of the effects of the experimental factors with a minimal number of hybridized
slides and a high statistical power. In contrast to conventional microarray analysis, the
factorial analysis is unique as it can properly estimate the potential effects of an
experimental condition that was not directly included in the hybridization (Leung et al.
2008). Most importantly, the factorial analysis of microarrays employed in this study is
optimal for studies involving aquatic animals, as they are exposed Simultaneously to a
multitude of environmental factors that are in continuous ﬂuctuation. Similar loop
designs of microarray data were successfully applied by Wit et al (Wit et al. 2005) and
Zou et al (Zou et al. 2005) and proved efﬁcient in comparing gene expression alteration
due to multiple treatments. Despite the clear advantages of the factorial analysis of
microarray data, the false discovery rate (FDR) continues to be a problem that cannot be
ignored (Pawitan et al. 2005). To minimize FDR, several statistical methods have been
integrated into the microarray data analysis (28-32). In this study, we presented the data
using both P < 0.05 and P < 0.01 as the cutoff for the microarray results. Subsequent
validation demonstrated that many of the differential expression, of several genes with
either P values in our study were indeed positive.

Among the complete probesets identiﬁed by this array study, 62% were altered in
response to adhesion status (Factor D), therefore, subsequent study of individual probes
focused on this group of genes that were most probably associated with the zebra

mussel’s ability to attach. The four candidate genes selected for qRT-PCR were from this

123

category. It was conﬁrmed by qRT-PCR that after 48 hour treatment, the expression
levels of these four genes were all upregulated during byssogenesis; however, curves of
the relative expression levels of the four genes during the three-day byssognesis were not
consistent. It is likely that these genes are involved at different stages during
byssogenesis and are not always up- or downregulated. This periodical expression or
depression may reﬂect the sequence of events of byssogenesis.

One of the identiﬁed genes, DPFP-BG20_A01, that had been validated by qRT-
PCR is homologous to the Dreissena polymorpha foot protein-1 (Dpfpl), which was
previously identiﬁed as an important protein for byssal thread structure (Rzepecki &
Waite 1993b, a) and was located by immunohistochemistry along the ventral groove of
the zebra mussel foot (Anderson & Waite 2000). The in situ hybridization performed in
this study located DPFP-BG20_A01 in an area along the groove of the mussel’s foot that
is dominated by thread forming glands (Bonner & Rockhill 1994a). It is noticeable in this
study that the expression proﬁles of many Dpfp-1 homologues were found in response to
the change of adhesion status. This suggests that multiple foot proteins are involved in
the byssogenesis of the zebra mussel in addition to the Dpfp-1. A number of foot proteins
produced by byssus glands and playing different roles in the byssogenesis of Mytilids,
another genus of mussels with a lifelong byssus apparatus, have been found over past
decades (Benedict & Waite 1986; Papov et al. 1995; Waite et al. 1998; Waite & Qin
2001; Zhao & Waite 2006; Lin et al. 2007; Cha et al. 2008). It is very likely that, in the
zebra mussel, there is also a group of foot proteins serving in different stages of

byssogenesis.

124

Not only were foot proteins identiﬁed, but also the homologues of exocrine gland
peptides (EGP) Narasimhan et al. 2007) were commonly found to be differentially
expressed during byssogenesis. The transcription of one of the EGP-like molecules, EGP-
BG97/ 192_BO6, was also Observed in the TF G cells area. This observation underscores
the importance of these molecules in zebra mussel adhesion and that byssogenesis in the
zebra mussel involves a myriad of proteins. The primary sequence of the EGP-
BG97/192_B06 suggests its distinction from Dpfp-l, despite the domination of its amino
acids with proline, tyrosine, aspartic acid, lysine, threonine, and glycine residues, which
is similar to Dpfp-1 (Anderson & Waite 1998). However, according to Francischetti et al,
the homologous genes encode for salivary gland peptides with a variety of functions in
ticks, such as anti-coagulant, anti-microbial, and oxidant metabolism (Francischetti et al.
2005). In our analysis, the ClastalW with zebra mussel EGP-like genes in response to the
byssogenic activities suggested that these molecules were structurally similar but not
identical to each other. Therefore, their potential functions during byssogenesis are most
likely different. It suggested that during the adhesion a number of molecules with various
functions were involved in this process. Some of the molecules may be directly involved
in the generation of the byssal threads and adhesive proteins, while others may play
alternate roles in the complicated process such as protecting the byssal proteins from
degeneration by microbes, releasing the adhesive proteins, and modiﬁcation of byssal
thread components.

The results of the study demonstrated that that very few genes had their
expression proﬁles altered by more than one experimental factor. This suggests that each

of the experimental factors affected only a few zebra mussel genes involved in

125

byssogenesis independently; i.e., no uniformal response to all factors exists. A series of
experiments was done by Clark and McMahon in 1996 to analyze the effects of
temperature (Clarke & McMahon 1996c), hypoxia (Clarke & McMahon 1996b), and
current velocity (Clarke & McMahon 1996a) on byssogenesis rate under laboratory
conditions. The change of the water temperature seems to have the most signiﬁcant effect
on byssogenesis rate. From 5 °C to 30 °C, every 10 °C shift caused dramatic difference in
the rate of byssal thread regeneration. Even 5 °C increase led to a signiﬁcant increase of
the byssogenesis rate when the temperature was above 25 °C (Clarke & McMahon
1996c). Our results demonstrated that the number of differentially expressed genes in
response to temperature changes was the highest of the four factors tested (27 genes). 0n
the contrary, the byssogenesis rate of the zebra mussel showed the least sensitivity to the
change of current velocity. Only in a certain ﬂow speed (0.2 m/s) was the byssogenesis
rate of zebra mussel signiﬁcantly higher than that in the other speeds (0.1 m/s, 0.15 m/S,
0.27 m/s) (Clarke & McMahon 1996a). Consistently, the least number of genes (9 genes)
identiﬁed by microarray showed the signiﬁcantly differential expressions during the
change of the current velocity. Comparing the genes signiﬁcantly altered by the four
experimental factors, we can easily see that adhesion status, temperature, and dissolved
oxygen level tends to affect the byssogenesis by modulating Similar genes. For example,
the homologues of Dpfp-l, EGP, C. elegans neuropeptide-like proteins (N P_504109 and
NP_505834), and oyster shematrins (BAE93436) (Y ano et al. 2006) were widely
identiﬁed. It is possible that the temperature and oxygen level along with the adhesion
status, affect the byssogenesis through the similar pathway, while the current velocity

seems to control this process through a different pathway.

126

For the genes that were altered by multiple experimental factors, it is also very
interesting to observe how environmental factors can alter gene expression in different
directions. For example, BG23_BO3 is downregulated in attached status, while is
upregulated with the decrease of the temperature; while BG97/192_BO6 can be
upregulated in the attached status and downregulated by the dissolved oxygen level.
What these gene-environmental factor interactions mean to zebra mussel attachment and
associated mechanisms, cannot be easily answered, primarily due to the fact that the
putative functions of these genes remain largely unknown. In conclusion, the factorial
analysis of the microarray results provides important information on how their expression
is inﬂuenced by important environmental factors. This newly generated knowledge will
help to better understand the molecular mechanism of zebra mussel underwater adhesion.
The factorial analysis of zebra mussel adhesion mechanism has provided for us a great
tool to learn about the role of the four experimental factors during the zebra mussel
attachment under controlled laboratory conditions. One can easily speculate that in the
mussel’s complex aquatic habitat more genes are likely to be involved in the attachment
process. Additional studies, also using cDNA microarray factorial analysis, deem
necessary to decipher the mechanisms governing zebra mussel byssogenesis, a process
that has devastated the fragile ecosystem in infested waters of the Laurentian Great Lakes

basin.

127

 

 

 

 

 

Figure 5-1 The non-reference interwoven 100p design for microarray analysis.

Effects of temperature, dissolved oxygen, current velocity, and byssogenesis status on
gene expression in the zebra mussel foot using cDNA microarray analysis. The ﬁgure
displays the different treatment combination selected as per the loop design approach
(Churchill 2002). Each arrow represents one microarray hybridization. The start of the
arrow stands for the sample labeled with dye Alexa 647 while the end point of an arrow
represents the sample labeled with dye Alexa 555. L stands for low temperature (4 °C)
while R stands for a higher temperature (22 0C); S stands for static water while F means
ﬂow water stirred by magnetic stirring bars; H represents the low dissolved oxygen level
(5mg/L) while the N represents the normal dissolved oxygen level (10 mg/L).

128

 

 

 

 

 

 

 

1 \A-3Genes B-TGenes C-IIGenes D-BCenes
o / I L“ ._...,.__-.—.-.:f:j M: _g/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.-

Figure 5-2 The hierarchial cluster of microarray identiﬁed genes.

The identiﬁed probesets with P < 0.05 were hierarchically clustered based on their
expression proﬁles under the effects of the four factors. Ten clusters are generated
representing the genes with the expressions induced by the change of the four factors.

This ﬁgure is in color.

129

 

 

Temperature Current veloclty

 

t «~31 sizi . ..g ii 599 Byssogenesis

 

 

 

Figure 5-3 The numbers of genes whose expression proﬁles are altered by single factor or
multiple factors.

The values in the ellipse were the numbers of the probes which were altered by the factor
with the same color. For example, the sum of the numbers in red ellipse is l9+l+1+6=27
and that means 27 probes were found modulated by the temperature. The underline(s)
under each number suggested the amount and types of factors could regulate its
expression proﬁles. For instance, the number 19 with a read underline means the
expression proﬁles of the 19 genes can only be altered by factor A while the value 1 with
a red and blue lines suggests that one gene was regulated by both temperature and current
velocity.

This ﬁgure is in color.

130

BG97/192_B06
BG13_F10
BC23_B03 . ,; ;
BC3IH01 ACACTTCATCTCTTTCTTACTCC CGGCCGTF GC CCTTC TE
BCIQCOS ACACTTCATCTCTTTCTTACTCCCCCCCCTACCCCTTCr

BG97/192_B06

BG13_F10

BG23ﬁBO3 ' TTGCTCCGGTL‘ GCAAF‘LCG
Bc3iﬁor TTGCTCCIGTZ CCAAACC
Bcropos TTCCTCCCCTA GCAAACC

BG97/192_BO6
BGl3_F10 ICITACC
BG23_BO3 zurTAeroAT;ITGru:.i
BG34_H01 HITACGGATATGCr-imt
BG10_C05 ’TTFJ‘IJnTATGI'W great

BG97/l92_BO6 . .. . . ‘ V .~
BCLLFIO . TArrccooAAAcrArncinoiT rccnca;

BG23_BO3
BG34_H01 / \TCCi TAC‘GGGGG
BGIO_C05 fTATGGCTRIZ‘GGGGGAPACTF=.TGGIT“’1I§ZVITit"

BG97/l92_BO6

BG13_F10

BG23_BO3 :‘ r” ‘ H T ,i T ..TCCVTGIT;
BG34_H01 1‘“ T. “.117. _ 7. T‘L TlJleL‘l‘;l—‘ (,_,‘I
BGlO_C05 .a. r *r': -:. a .*T ..r C

BG97/192_BO6
BG13_F10
BG23_B03
BG34_H01
BG10_C05

BG97/192_BO6
8613 F10
BG23:BO3
BG34 _H01
BG10 _C05 ‘ ' .' ‘ " ' ‘ ‘ "

Figure 5- 4 Multiple alignment with the identiﬁed EGP encoding genes whose expression

proﬁles can be altered by byssogenic activity.

The comparison was performed by using ClustalW multiple sequences alignment. The

protein product encoded by the EST BG97/ l 92_BO6 is homologous to an exocrine

protein isolated from a ﬁlarial nematode L. sigmodonris while the rest of the EGPS

encoded by selected ESTS have a salivary gland peptide from blacklegged tick (1.

scapularis) as their homologue. The nucleotides were colored with black when more than

60% nucleotides in this locus were identical.

 

131

BG27_BO8 45
BG03_BOl 80
BG23_D02 76
BGZ7_BOB 78
BG03_BOl 120
BGZ3_D02 116
BG27_BOB 116
BGO3_BOl _-,. _;C_{.. I, . 160
BGZ3_D02 ~.;3*T1R~»'**“*J ;' - .T”TATCH3CGC; 156

BGZLﬁOB TRRTAACTATGGCTACgCCGGAAACTATGGCTACGGGGGH 156
BGmxﬁOI TRACAACTATGGCTACCCCGGAAACTATGGCTACGGGGGL 200
BGZipOZ TLACAACTATGGCTACCCCGGAAACTATGGCTACGGGC47 196

 

BG21ﬁ08 RRCTATGGCGGGTATCCAGGAAATTATGGW ---------- 186
BGWXﬁOl RACTATGGCGGGTATCCAGGAAATTATGGF ---------- 230
BGZXPOZ ~VPTATGGCGGGTATCCAGGAAATTATGGCAACTATGGCA 236

 
  
  
   
 
  

BG21ﬁO8 -------- TRWTATGACAACTATGGTGGCGGATGGTTRF7 218
BGWTﬁOl -------- JRCTATGACAACTATGGTGGCGGATGGTTAT} 262
BGZLPOZ ACTATGGCGACTATGACAACTATGGTGGCGGATGGTTATT 276

 

BGZLﬁOB wAAACTCCTCCGACCCGCTCC ------ v u— 251
8631601 “RRAATCCTCGGAGGGGGTGCAAAAGGC‘\AGGIutxfWG 302
BGZLPOZ ‘T‘\~TFFTFGGAGGGGGTQG ------ 3 AnpA- 309
BGZ7_808 -------- 'TTA'”“ 283
BGWLﬁOl GGTGGTTACGG‘ 342
BGZB_D02 ----------- “in C . . .C; A2;- 338
BG27_BO8 323
BG03_BOl 37o
BG23_D02 366
BG27_BOB . . _ _ ,. 359
BGWLﬂOl ;nATACATCTAAAwiAACCCTTTICTT “TITTTTA 3Tr- 410
BG23_D02 ' "- 7 .72 7 7 T 7‘ T? % 7,‘ 7 .7“. ’1‘ < ‘ I' I ‘T T T i T T. '73 T 7“. 7 T T T 7 C} . 7 ' 406

 

Figure 5-5 Comparison of primeray structures of the EGP encoding genes whose
expression proﬁles can be altered by water temperatures.

The three EGP genes are all differentially expressed under the change of water
temperature. The EST BG27_B08 is homologous to salivary gland peptide identiﬁed
from the western black-legged tick (I. paciﬁcus). The other two probesets are all
homologous to the blacklegged tick (I. scapularis). The nucleotides were colored with
black when more than 60% nucleotides in this locus were identical. The gap between the
nucleotides were labeled as “-”.

132

EGP—BG97/192_BOS EGP-BG13_BOS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20000 - (+) 1200 -

“’ (+)

8 HA 1000 .

(0 15000 - l

'8 800 -

B

4; 10000 - 600 «

g 400 ~ .

E 5000 - ﬂ .. .

c? (+) ‘ i (+) (+) (+) 200‘ (+) H H H

0 . ' . i7- . . 0 . . . .
DPFP-BG20_A01 NH-BGl7_C09

500 q 100 1 (+)

a: Flu,

g 400 « 80 .

g 300 . 60 -

2 200 . 4o . f

323 100‘ (+) . . (+) (-> <-) 20‘ (+) . -. (+) (+) (+)

% O r I 7 V —._ I I 0 I T7 7 1 L7; I 1

07 Hemocyte Foot Muscle Ctenidium Mantle Hemocyte Foot Muscle Ctenidium Mantle

Figure 5-6 The distribution of the mRNA products of the selected genes within zebra
mussel tissues.

(+) indicated the existence of the gene in the tissue; (-) suggested no detected transcripts
in the tissue. For each gene, the lowest (+) sample was used as control and the relative
expression levels of the gene in other tissues were calculated by using IMO model.

133

Relative Expression Level

Relative Expression Level Relative Expression Level

Relative Expression Level

DPFP—BG20_AO1

__l

     
 

4.5 - EGP—BGQ7/192_BOG

EGP-BG13_BOS

3 _ NH-BG17_COQ

 

i

O 12hr Day1 Day2 Day

 

Figure 5-7 The qRT-PCR results demonstrated the relatively expression levels of the
gene during the byssiogenesis.

The byssogenesis and non-byssogenesis samples were treated as described in materials
and methods, and the non-byssogenesis sample was used as control with value 1. The
amount of the transcripts of the gene was calculated by using Z'AAC‘ model. The
comparisons of the gene expression levels between byssogenesis and non-byssogenesis
and detached sample were made at O, 12 hours, 1 day, 2 days, and 3 days post treatment.

134

 

Figure 5-8 The distribution of the zebra mussel byssus gland cells in mussel foot.

Figure 5-8. The distribution of the zebra mussel byssus gland cells in mussel foot. The
zebra mussel foot sections are stained by H&E method. The sections are made along the
longitude axis as demonstrated by the paralleled dash lines across the foot. Arrows with
dashed line indicate the position of the three major byssal glands embeded in the mussel’s
foot.

3: A longitudinal section in the root of the mussel’s foot. Arrows point to the stem-
forming glandular cells (S).

b: The section of middle area of the foot. The light purple cells alsong the epethelial
surface of the foot were thread-forming cells which were demostrated by arrows and the
letter T.

c: The tip of the foot containing the deep purple stained plaque-forming gland cells that
were labled as P.

This ﬁgure is in color.

135

   
 
  

__ Stem-forming
Gland

I_ (.--

Thread-forming
Gland

‘(.-__

 

av 7Flaque-forming
“ Gland

Figure 5-9 The in situ expression of DPFP-BGZO_A01 and EGP-BGQ7/ 192_BO6 in foot.
Figure 5-9. The in situ expression of the gene DPFP-BGZO_A01 and EGP-
BG97/192_B06 in zebra mussel foot tissue. The synthesized antisense RNA of DPFP-
BGZO_A01 was labeled with Alexa 488 dye (green) while the complementary RNA for
EGP-BG97/192_BO6 was labeled with Alexa 594 (red). The foot was cut along longitude
as shown by the two paralleled dash lines across the foot. The positions of the byssus
glands were labeled on the foot. The dash lines connected between slides and foot
indicated the position of the three slides on the foot.

a: The root area of zebra mussel foot including stem-forming gland;

b: The middle region of the mussel foot containing thread-forming gland;

c: The tip section of the foot inchrding plaque-forming gland.

T: Thread-forming gland cells.

This ﬁgure is in color.

136

Table 5-1 The null hypothesis and the interpretation of tested probesets.

 

Null hypothesis Interpretation when null hypothesis is rejected

 

H 0, A = 0 The expression proﬁle of certain genes is altered in response to
the surrounding temperature.

H 0, B ___ 0 The expression proﬁle of the gene is altered by the change of
water movement.

H 0, C = 0 The expression proﬁle of the gene is altered by dissolved oxygen
level.

H0: D = 0 The expression proﬁle of the gene is altered by byssogenic
actIVIty.

 

137

 

Table 5-2 The gene speciﬁc primers used for qRT-PCR.

 

Target EST

Primer Name

Primer Sequence

 

AM229726

AM229885

AM230073

AM230384

*IssrRNA

BGZOAO 1_DPF P_RT_F w
BGZOAO 1_D.PFP_RT_Rv
BG 13 BOS_EGP_RT__FW

BG l3BOS_EGP_RT_Rv
8097/ l 9ZBO6_EGP_RT_Fw
8697/ l 9ZBO6_EG P_RT_Rv
BG 1 7CO9_NH_RT_Fw

BG 1 7C09_NH_RT__Rv

* 1 8SrRNA_Forward

* l SSrRNA_Reverse

5’ -ATG GGC CAT ATG ATA AGA AAC CA- 3’
5’ -TCC AGG AGG TTC CAA TGG AA- 3’

5’ -CGG TTG CTA TAC ATG TGT CCA AGT- 3’
5’ -GGG AGG TTA CGG CGG CTA T- 3’

5’ -CAT CCC CGT ATG GGA TCC A- 3’

5’ -GGT GCA ACG GCC AAG TTT AT- 3’

5’ -TCC GGA TAT TGG TTG TCC TCA T- 3’

5’ —TTC TCC GTA GCC ACA CCA TTT- 3’

5’- GAC ACG GCT ACC ACA TCC AA -3’

5’— CTC GAA AGA GTC CCG CAT TO -3’

 

31‘

The primers used as internal reference for QRT-PCR.

138

 

Table 5-3 The genes whose expression proﬁles have been signiﬁcantly altered by the
status of byssogenesis.
When Log(F C) > O, the gene is upregulated in attachment status.

 

 

Gene ID Accession # p.value Log(FC) Homologue
AM230231 0.00011 -0.3 N/A
BG17_G02
AM229883 0.00114 0.073 AAV80789.1| Exocrine gland peptide
B G10_C05 [lxodes scapularis]
BG33_H03* AM230185 0.0013 -0.084 N/A

* C AM230076

0.00317 0.107
BG97/192_Bo6

AAS92593. 1| Excretory/secretory protein

[Litomosoides sigmodontis]

* AM230089 0.00339 —0.06l N/A
BG20_F04
* AM230013 0.00374 -0.065 AAV80789.1| Exocrine gland peptide
BG13_F10 [lxodes scapularis]
* AM229897 0.00498 -0.092 AAV80789.1| Exocrine gland peptide
BG23_B03 [lxodes scapularis]
* AM230170 0.00523 0.136 ABISZ762.1| 60S ribosomal protein L27
B G14-All [A rgas monolakensis]
* C AM230168 0.00801 0.084 AAN05585.1| Ribosomal protein L22
B (716—003 [A rgopecten irradians]
9: AM230205 0.00946 -0.062 N/A
BGl3_C ll _
* AM229894 0.01029 0.087 AAV80789.1| Exocrine gland peptide
‘ B 034—1101 [lxodes scapularis]
BG20_A01 AM229724 0.01 149 0. 185 AF 265353_1 | Byssal protein Dpfpl
[Dreissena polymorpha]
BGZ6_C06 AM230431 0.01 19 -0.079 N/A
BG05_D06 AM230302 0.01 191 0.076 N/A
BGI7_C09 AM230384 0.01206 0.158 N/A
BGOS_BIO AM230189 0.01278 0.048 N/A
BG10_F10 AM230328 0.01321 -0.076 N/A
BG32_C03 AM229736 0.01371 -0. 141 AAC39039.1| Foot protein 1 precursor
[Dreissena polymorpha]
BG32_D04 AM230156 0.01374 0.058 BAB12683.1| Polypeptide release factor 3
[Yarrowia lipolytica]
BG15_D07 AM230362 0.01379 -0.055 N/A
BGZ3_E03 AM229779 0.01382 -0.057 BAE93436.1| Shematrin-4 [Pinctada
fucata]
BGl3 805C AM229816 0.0139 0.15 AAV80789.1| Exocrine gland peptide
— [lxodes scapularis]
BG23_C05 AM230080 0.01572 -0.058 N/A
BGl4_C07 C AM230353 0.01679 0.096 N/A
BGZZ_C12 AM229893 0.01706 -0.072 N/A
BGO7_F07 AM230102 0.01764 -0.057 NP_504109.1| Neuropeptide-Like protein
29 [Caenorhabditis elegans]
BG06__E09 AM230224 0.01777 -0.073 AAF75279.1| Byssal protein Dpfpl
[Dreissena polymorpha]
BG04_A09 AM230090 0.02014 -0.054 N/A
BGlZ_D10 AM230249 0.02035 0.146 N/A
30311311 C AM230254 0.02171 0.147 N/A
BG25_H08 AM230042 0.0221 1 0.168 N/A

 

139

 

 

Table 5-3 (cont’d-Q

 

 

Gene ID Accession # p.value Log(FC) Homologue

BGO4_GO4 AM230066 0.02267 -0.05 ABD62888.1| Serine proteinase-like
protein [Penaeus monodon]

BGO4_F03 AM230188 0.02308 -0. 161 N/A

MF030105_BO9 AM230050 0.02341 -0.076 N/A

BG28_H05 A AM229934 0.02359 0.1 14 1121126523436J| Shematrin-4 [Pinctac/a

BG34_F03 AM230174 0.02459 0.07 ABG81984.1| Ribosomal protein Sl4e
[Diaphorina citri]

BG33_E04 AM229997 0.02496 -0.067 N/A

BG18_BO7 AM230216 0.0271 0.061 N/A

BGlO_D04 AM230'104 0.02748 -0. 143 AAK68690.1| Hemicentin
[Homo sapiens]

BG20_BO9 AM230000 0.02829 -0.057 AAV80789.1| Exocrine gland peptide
[lxodes soapy/art's]

BG30_H12 AM229764 0.02853 -0.053 N/A

BG25_F07 AM229895 0.0296 0.058 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]

MF030105_G10 AM229731 0.02971 0.163 AAF75279.1| Byssal protein Dpfpl
[Dreissena polymorpha]

BGOB_BO8 AM230321 0.0341 0.109 N/A

MF030105_H09 AM229864 0.03422 -0.041 AAV80789.” Exocrine gland peptide
[lxodes scapularis]

BG06_C03 AM229866 0.03508 -0.142 N/A

BGZ3_BO6 AM229738 0.03609 0.046 AF265353_1 Byssal protein Dpfpl
precursor [Dreissena polymorpha]

BG05_E08 AM230198 0.03709 0.057 N/A

MF030105_F07 AM230273 0.0371 1 -0.047 N/A

BGZ6_BO9 A AM229855 0.03759 -0.048 N/A

MF030105_C07 B AM229749 0.03772 0145 N/A

BG34_B04 AM230256 0.03796 0.043 N/A

MF030105_F 02 AM230272 0.03899 -0.1 1 N/A

BGBl_All AM230107 0.04079 0.047 EAY59350.1| Serine protease pepD
[Mycobacterium tuberculosis C]

BG03 801A AM229901 0.04086 0.064 AAV80789.1| Exocrine gland peptide

— [lxodes scapularis]

30053304 A AM23007O 0.04134 0.055 N/A

BG10_809 AM230213 0.0415 0.105 N/A

BG97/ 192_A06 AM230225 0.04155 0.07 N/A

BG08_F 02 AM229725 0.04254 0.052 AAF75279.1| Byssal protein Dpfpl
[Dreissena polymorpha]

BG16_A07 AM230120 0.04305 -0.06 ABEO3 741 . 1| Prophenoloxidase
activating factor [Penaeus monodon]

BGZG_G12 AM229957 0.04376 -0.049 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]

BG04_A11 AM230155 0.04379 0.05 BAB12683.1| Polypeptide release factor 3
[Yarrowia lipolytica]

BGZ9_E08 AM229745 0.04467 -0.062 AAF75279. 1|Byssa1 protein Dpfpl
precursor [Dreissena polymorpha]

BGl4_D01 AM230121 0.04495 -0.043 N/A

BG13_A02 AM230345 0.04566 0.087 N/A

 

140

 

 

Table 5-3 (cont’d-Z)

 

 

Gene ID Accession # p.value Log(FC) Homologue

BG04_H02 AM229815 0.04732 -0.05 AAT92111.1| Exocrine gland peptide
NPL-Z [lxodes paciﬁcus]

BG14__D12 AM229862 0.04786 0.045 BAE93436.1| Shematrin-4 [Pinctada
fucata]

BG15_F03 AM229727 0.04786 0.048 AAF75279.1|Byssa1 protein Dpfpl
precursor [Dreissena polymorpha]

BG17_F03 AM230143 0.04871 -0.129 N/A

BG10_H04 AM2301 18 0.04908 0.069 N/A

BG31_D01 AM230039 0.04942 0.133 NP_505834. 1| Neuropeptide—Like protein
33 [Caenorhabditis elegans]

BG15_D03 AM229780 0.04982 0.097 N/A

BG30 D12C AM229805 0.04991 -0.045 BAE93436.1| Shematrin-4

— [Pmctadafucata]

 

* The differentially expressed ESTs with P <0.01.
A Also altered by Factor A (Temperature); B Also altered by Factor B (Agitation);

C Also altered by Factor C (D.O.)

141

 

Table 5—4 The genes whose expression proﬁles are signiﬁcantly altered due to the change

 

 

of temperature.
Log(FC) > 0 means the gene is upregulated with the decrease of temperature.
Genes [D Accession # p.value Log(F C) Homologues
D -
BG17_C09* AM230384 7.00E 05 0.314 N/A
BGZ7 808* AM230019 0.00043 0.1 16 AAT921 1 Lil Exocrine gland peptide NFL-2
— [lxodes pacrﬁcus]
* D AM229934 0.0052 -O.161 BAE93436.1| Shematrin-4
BGZS_H05 .
[Pmctadafucata]
BG03 801* AM229901 0.00653 -0.098 AAV80789.1| Exocrine gland peptrde
— [lxodes scapularrs]
BG23 D02 AM229752 0.0068 -0.095 AAV80789.” Exocnne gland peptide
- [lxodes scapularzs]
BG28_C09 AM229790 0.01034 -0.127 BAE93436.1| shematrin-4[Pinctadafucata]
BG17_F07 AM229867 0.01123 -0.057 BAE93436. 1| shematrin-4 [Pinctadafucata]
BG27_H05 AM230068 0.01216 0.091 AAC05725. 1| RNA helicase A
[Mus musculus]
BG27_E09 AM229903 0.01427 0.1 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]
BG26_D08 AM230432 0.01818 0.054 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]
BG13_806 AM229798 0.01919 0.152 AAT92111.1| Exocrine gland peptide NFL-2
[lxodes paciﬁcus]
BGZB_BO4 AM230435 0.02361 -0.19 N/A
BG29_E12 AM230154 0.02449 -0.065 N/A
BG05_BO4 AM230070 0.02598 -0.067 N/A
BGl6_F01 AM230081 0.02693 0.096 AAS92593. 1| Excretory/secretory protein
[Litomosoides sigmodontis]
AM229897 0.02749 0.077 NP_505834.1|Neuropeptide-Like protein 33
BG23 803 . .
— [Caenorhabdrtzs elegans]
BG26_C05 AM229917 0.02753 0.08 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]
BG18_D01 AM229730 0.02863 0.201 N/A
BG16_G06 AM230258 0.03025 -0.212 N/A
BG05_A11 AM229750 0.03037 0.069 Q25460| Adhesive plaque matrix protein
[Mytilus edulis]
BG27_BIO AM230153 0.03739 0.065 BAB12683.1| Polypeptide release factor 3
[Yarrowia lipolytica]
BG33_A08 AM229799 0.03749 0.093 AAT92111.1| Exocrine gland peptide NFL-2
[lxodes paciﬁcus]
BG97/ 192_D02 AM229740 0.03991 -0.096 N/A
B . .
BG07 H06 AM230138 0.04305 0.078 ABN1-3415.1|Chorrogen1n H
- C [Oryzras melastrgma]
B 628_E01 AM229772 0.04418 0.053 N/A
BG28_F03 AM230145 0.04425 0.069 N/A
D
B G2 6_B09 AM229879 0.0449 0.051 N/A

 

* The differentially expressed ESTs with P <0.01.

B Also altered by Factor B (Agitation); C Also altered by Factor C (D.O.); D Also altered
by Factor D (Adhesion).

142

Table 5-5 The genes whose expression proﬁles are Isigniﬁcantly altered due to the change

of DO.

Log (F C) > 0 indicates the gene is upregulated when the dissolved oxygen in water is at

lower level.

 

 

Gene ID Accession # p.value Log (F C) Homologue
BG23_D06* AM230109 0.0026 -0.135 N/A
BG14_F09* AM230247 0.00639 0.08 N/A
BGlZ_BlO AM230219 0.01343 0.098 N/A
D ' _
BG30 D12 AM229805 0.0138 0.048 BAE93436.1| Shematrm 4
— [Pmctaa'aﬁzcata]
MF030105_D04 AM230269 0.01606 -0.125 N/A
BGIS_BO3 AM230397 0.02033 0. l 17 N/A
D . .
BGI6 D03 AM230168 0.02521 0.056 AAN05585.1'| Ribosomal protein L22
- [A rgopecten trradzans]
BG05_D02 AM2302] 1 0.02647 0. 103 N/A
D
BG14_C07 AM230353 0.02743 0.071 N/A
BG97/ 192_C05 AM230277 0.02815 0.063 N/A
BG13_E01 AM229847 0.02927 0.044 BAE93436.1| Shematrin-4
[Pinctadafucata]
B -
B G2 5_H1 l AM230248 0.02979 0.047 N/A
BGl6_BOI AM229753 0.03044 0.052 AAV80789.1| Exocrine gland peptide
[lxodes scapularis]
MF030105_H01 AM229821 0.03044 0.134 AAT92111.1| Exocrine gland peptide NPL-
2 [lxodes paciﬁcus]
BGO7_F03 AM229726 0.03069 -0.083 AAF75279.1| Byssal protein Dpfpl
precursor [Dreissena polymorpha]
BG12_G10 AM230123 0.03203 0.054 N/A
D . .
BG97/l92 806 AM230076 0.03528 0.059 AAS92593.1| Exocrine/secretory protein
’ [Lttomosozdes szgmodontzs]
BG27_F10 AM230146 0.04161 0.047 N/A
BG24_Bll AM230213 0.04259 0.041 N/A
A
BG28_E01 AM229772 0.04306 0.039 N/A
BG26_BOl AM230172 0.04342 -0.04l AAK95191.1| 40$ ribosomal protein S9
[Ictalurus punclatus]
D _
BG31_E11 AM230254 0.04511 0.103 N/A
D AM229885 0.04647 0.096 AAV80789.1| Exocrine gland peptide
BG13 B05 .
- [lxodes scapularts]
B _
BG22_H0 6 AM229789 0.04694 0.039 N/A
BGZO_F01 AM229782 0.04702 -0.107 N/A
BGZS_301 AM229884 0.04732 0.089 AAV80789.1| Exocrine gland peptide

* The differentially expressed ESTs with P <0.01.

[lxodes scgularis]

A Also altered by Factor A (Temperature); B Also altered by Factor B (Agitation);

D Also altered by Factor D (Adhesion).

143

 

Table 5-6 The genes whose expression proﬁles are signiﬁcantly altered due to the current

velocity.

Log(FC) > 0 indicates the gene is upregulated when the mussel is in ﬂow water.

 

 

Gene ID Accession # p.value Log (FC) Homologue
BG29_BO9* AM230251 0.00415 0.1 16 N/A
C _
BG22_H06* AM229789 0.00952 0.078 N/A
BG10_F01 AM230379 0.01 195 0.074 N/A
BG29_A05 AM230250 0.02185 0.053 N/A
D -
MF030105_C07 AM229749 0.02546 0.188 N/A
A . .
BG07 H06 AM230138 0.0265 0.094 ABN1.3415.1| Choriogemn H
"' [OryZIas melasngma]
D
BG17_G02 AM230231 0.03856 0.166 N/A
C -
BG25_H11 AM230248 0.03927 0.065 N/A
BG12_H05 AM230222 0.0491 0.076 N/A

 

* The differentially expressed ESTs with P <0.01.
A Also altered by Factor A (Temperature); C Also altered by Factor C (D.O.);
D Also altered by Factor D (Adhesion)

144

CHAPTER VI

Defensin of the Zebra Mussel (Dreissena
polymorpha): Molecular Structure, in vitro
Expression, Antimicrobial Activity, and Potential
Functions

Abstract

A 409 bp full length defensin cDNA was cloned and sequenced based on an expressed
sequence tag (EST) obtained from a normalized zebra mussel (Dreissena polymorpha)
foot cDNA library developed in our laboratory. The D. polymorpha defensin (Dpd) gene
encoded a peptide with 76 amino acid residues. The mature Dpd contains 54 amino acids
with a fully functional insect defensin A domain. Homologue searching against GenBank
database suggested that this Dpd was phylogenetically close to defensins from a group of
insects with six conserved Cysteine residues. Predicted with homology modeling method,
the three dimensional structure of Dpd also demonstrated a signiﬁcant similarity with
insect defensin A. The recombinant Dpd was in vitro expressed through an Escherichia
coli expression system. The antimicrobial activities of the re-folded recombinant Dpd
were found against the growth of Morganella sp., Plesiomonas shigelloides,
Edwardsiella tarda, Escherichia coli, and Staphylococcus aureus aureus (ATCC12598)
with the minimal inhibition concentration (MIC) 0.35, 0.43, 1.16, 6.46, and 30.39 pM,
respectively. However, with less than 50 uM no detectable inhibition activities were
observed against the other four Gram-negative bacteria (Aeramanas salmom'cida

salmonicia’a, Motile A eromonas, F lavobacterium sp., Pseudomonasﬂuorescens,

145

 

Shewane/la putrifaciens), as well as the other Gram-positive bacteria (Bacillus
megaterium ATCC14581, Carnobacterium maltaromaiicum ATCC27865, Enterocaccus
faecalis ATCC19433, and Microcaccus leteus ATCC4698), and the yeast Saccharomyccs
cerevisiae. The RNA ﬂuorescent in situ hybridization (FISH) of Dpd in the zebra mussel
suggested that the Dpd was expressed in a variety of tissues, such as foot, retractor
muscle, ctenidium, mantle, hemocytes, gonad, digestive gland, and intestine. By using the F
quantitative PCR, the expression level of Dpd in the zebra mussel foot was the highest,

followed by the muscle. By comparing the amount of Dpd transcripts in the zebra mussel

 

foot under two different byssogenesis conditions (byssogenesis and non-byssogenesis)
with qPCR, we found that the higher expression levels of Dpd were always associated
with the byssogenesis status from two days to four days after the start of byssogenesis.
Findings of this study suggest that the expression of the Dpd gene was upregulated in

vivo by byssogeneis, and by hemocytes in vitro upon stimulation with microbial antigens.

146

Introduction

The invasion of the zebra mussel (Dreissena polymorpha) in North America has
raised several questions pertaining to the ability of a non-native mollusk to adapt and

reproduce in a new environment. While the lack of natural predators and ability to attach

 

ﬁrmly to underwater substrates are logical explanations, scientists believe that zebra '5‘
mussels may have extraordinary host defense mechanisms that allow them to survive
microbial attacks in the pathogen-rich aquatic environment (F rischer et al. 2000). While L

the zebra mussels could be experimentally infected with pathogens such as Aeromonas
spp. (Maki et al. 1998) and Pseudomonasﬂuorescens (Molloy 2002), no spontaneous
infections by either pathogens were observed since the zebra mussel was ﬁrst introduced
to North America in the 19803 (Hebert et al. 1989).

Like other invertebrates, bivalves, to which zebra mussels belong, have both
cellular and humoral innate host defense mechanisms (Cheng 1983; Hine 1999). The
former includes two main pathways, phagocytosis or encapsulation, followed by the
destruction of pathogens via enzymes or the release of oxygen metabolites (Pipe 1990,
1992), while the latter contains two families of molecules, opsonins (Renwrantz &
Stahmer 1983) and antimicrobial peptides (AMPS) (Hubert et al. 1996; Mitta et al.
2000a). Most of the knowledge about bivalve innate immunity comes from marine
mussels or oysters, such as Mytilus edulis (Pipe et al. 1997), M. gallaprovincialis
(Carballal et al. 1997), Crassostrea virginica, and C. gigas (Oliver & Fisher 1995). The
studies on the host defense mechanism of the zebra mussel are very limited (Giamberini

et al. 1996).

147

Previous study of zebra mussel hemocyte-mediated host defense mechanisms
demonstrated that exposure of hemocytes to microbial-derived proteins causes a sharp
rise in the expression of genes with potential defense and cell protection functions such as
matrilin and heat shock proteins (Xu & Faisal 2007, 2009b). Interestingly, when a cDNA
library was constructed with zebra mussel foot and normalized by suppression subtractive
hybridization technique, 45 immune-related genes were identiﬁed (Xu & Faisal 2008).
Considering that the foot cDNA library was developed through subtraction of the
mussels’ retractor muscle, which contains a sinus full of hemocytes, the immune-related
genes found in the subtracted foot cDNA were either overly expressed in foot cells as
compared to hemocytes or unique to the foot. Located in the zebra mussel foot, there are
three major glands that produce foot proteins, which build unique structures called byssal
threads to maintain underwater attachment. The generation of byssal threads, also known
as byssogenesis, is affected by many environmental factors such as temperature (Clarke
& McMahon 1996c), current velocity (Clarke & McMahon l996a), dissolved oxygen
(Clarke & McMahon 1996b), and microbial communities (Kavouras & Maki 2003b;
Kavouras & Maki 2003a; Kavouras & Maki 2004). In particular, the interference of
microbial communities during the zebra mussel byssogenesis has been observed in both
the larval and adult stages of zebra mussels. The host defense related genes identiﬁed
from our previous study are probably involved in the self protection employed by zebra
mussel byssus. Among these, an AMP molecule with high similarity to defensin was
found in the zebra mussel foot SSH cDNA library.

AMPS play important roles in the innate immune system of a variety of organisms

(Boman 2003; Bulet et al. 2004). Over 1,000 antimicrobial peptides have been reported

148

 

(Bulet et al. 2004), including three basic types based on their secondary structure of
sequences, linear peptides with a—helices, highly disulphide-bonded (cysteine-rich) [3-
sheets, and those with proline- or glycine-rich character (Douglas et al. 2003; Gueguen et
al. 2006). The amphipathic character of most members in the AMP family enables them
to be inserted into biological membranes. Although, in most cases, the primary action of
AMP is to lyse the membranes of pathogens, they are also involved in a number of other
reactions depending on their tissue distributions, such as mediating inﬂammation
(Hancock & Lehrer 1998).

Among AMPS, the defensin represents a family of peptides with the antimicrobial
activities against a wide range of bacteria and fungi (Boulanger et al. 2004). Structurally,
the animal defensin molecules contain three or four disulﬁde bonds and can be classified
into four major groups: a-, B-, 0-, and invertebrate defensin. In many vertebrates, the a-
defensin is constitutively expressed in neutrophils and acts intracellularly within the
phagolysozom (Cunliffe 2003). The B-denfensin is inducible by a variety of pathogens in
epithelial cells and neutrophils, and also functions to initiate the adaptive immune system
of vertebrates (Yang et al. 1999). The O-defensin, expressed in the neutrophils and
monocytes of non-human primates, was most recently was found to have lectin—like
activities that can mediate the protection of lymphocytes from infection by viruses (Cole
et al. 2002; Munk et al. 2003; Wang et al. 2003).

The ﬁrst insect defensin, ﬁrst puriﬁed from the larvae of the ﬂesh ﬂy Phormia
terranova (Hofﬁnann & Hetru 1992), has been isolated from a broad range of insect

species which includes six insect orders (Cociancich et al. 1993). As the most diverse

149

 

animal group, the insects have the most remarkable diversity not only reﬂected by their
morphology and lifecycles but also by various biomolecules including defensins.

A common insect defensin contains Six cysteine residues engaged in three
disulﬁde bonds. Three distinct domains can be found in a mature insect defensin
including an amino-terminal loop, an amphipathic a-helix, and a carboxyl-terminal anti-
parallel B-Sheet (Bonmatin et al. 1992). Like the vertebrate defensins, the insect defensins
also inhibit the growth of a wide range of microorganisms (Hoffmann & Hetru 1992).
Interestingly, the defensins identiﬁed from mollusks were similar to insect defensins.
Thus far, two defensins, Cg—def and MGD-l, have been isolated from Crassostrea gigas
(Gueguen et al. 2006) and Mytilus galloprovincialis (Li et al. 2000), respectively.
Antimicrobial assays demonstrated that insect and molluscan defensins share functional
Similarities. To this end, this study was designed to identify the structure of zebra mussel
defensin, and to determine its tissue distribution, inducibility, antimicrobial activities, and

potential involvement in byssogenesis.

150

Materials and Methods

The collection and maintenance of zebra mussels

The zebra mussels used in all the assays of this study were collected from
Vineyard Lake in Brooklyn, MI, USA (Latitude: 42°4’59”N; Longitude: 84°12’34”W).
The mussel shells were thoroughly cleaned with a brush and rinsed with deionized water.
Then the mussels were maintained in aerated, sterile Vineyard Lake water in a glass tank
and fed weekly with a pure culture of the algus Ankistrodesmusfalcatus. The zebra
mussels used in this study were allowed to acclimate to the laboratory environment for

six weeks before use.

Gene cloning of zebra mussel defensin

The expressed sequence taq (EST) of Dpd was obtained from a normalized cDNA
library constructed with zebra mussel foot RNA (Xu & Faisal 2008). A pair of gene
Speciﬁc primers (BGl 1E08-RT- Fw: 5’ - CGA AGG CGG ATA TTG CTA CTG - 3’ and
BG11E08-RT- Rv: 5’ - TGA CCA TAA CAC TCC ATC GGT G - 3’) was designed
based on the Dpd EST with PimerExpress 2.0 (Applied Biosystem, Foster City, CA). The
full length cDNA library was constructed with SMARTTM RACE cDNA Ampliﬁcation
Kit (Clontech Laboratories, Mountain View, CA). The 3’ and 5’ ends of the Dpd gene
were ampliﬁed following the instructions of the kit. Puriﬁed PCR products were
sequenced at the Research Technology Support Facility of Michigan State University.
The full sequence of Dpd was deposited in GenBank with the accession number

GU139954.

151

 

Primary and tertiary structure analyses of zebra mussel defensin

The homologue searching of the full length Dpd gene was performed with the
Basic Local Alignment Search Tool (BLAST) through NCBI server. The Dpd gene was
translated into amino acid sequence by the DNA-Protein translation tool provided by
Expert Protein Analysis System (ExPASy) (Gasteiger et al. 2003). The primary structure
of Dpd protein was compared to defensin molecules from other species by using the
multiple sequence alignment program CLUSTAL W (Thompson et al. 1994).

The tertiary structure of zebra mussel defensin was predicted with a homology-
modeling method via ESyPred3D using neural networks (Lambert et al. 2002). The
visualization and characterization of the three dimensional structure of the zebra mussel
defensin was performed with software PyMOL (DeLano 2009). The tertiary structure of
zebra mussel defensin was compared to its homologue, the insect defensin A (PDB entry:
IICA) isolated from Phormia terraenovae (Cornet et al. 1995), as well as two defensin
molecules, MGD-l (IFIN) of M. galloprovincialis (Li et al. 2000) and Cg—def(2B68) of

C. gigas (Gueguen et al. 2006).

The expression of zebra mussel defensin in E. coli in vitro expression system
The recombinant zebra mussel defensin was produced in vitro with prokaryote
expression system, QIAexpression® System (Qiagen, Valencia, CA). The total RNA was
extracted from the zebra mussel foot followed by cDNA synthesis with SuperScript® III
Reverse Transcriptase (Invitrogen). A pair of primers (Forward: 5’ - GCG CGC ATG
CGG CAC CCC AGA AGC GTA TTA C - 3’; Reverse: 5’ - ATA TAT GTC GAC TTA
ACC AAG GAT TTC CGA GAA GG - 3’) was designed to amplify the whole functional

domain of the zebra mussel defensin as well as to introduce the recognition site of

152

 

restricted endonuclease Sphl and SalI to the 5’ and 3’ ends of the sequence, respectively.
The PCR product was obtained through RT-PCR with zebra mussel cDNA template and
was puriﬁed with Wizard® SV Gel and PCR Clean-Up System (Promega, Madison, WI).
The puriﬁed PCR product was sequentially digested with endonuclease Sphl and SalI
(New England Biolabs, Ipswich, MA). Meanwhile, the expression vector PQE-l from the
QIAexpression® System (Qiagen) was also digested by Sphl and Ball in sequence. The
digested PCR product and PQE-l vector were puriﬁed, and then connected to each other
with T4 DNA ligase from pGEM®-T Easy Vector System I (Promega). Thereafter, the
recombinant plasmid was transformed to competent M15 E. coli cells. The preparation of
competent cells, transformation, and the bacteria culture were based on the protocol in
the manufacturer’s instruction of the QIAexpreSSionistTM (Qiagen). The Isopropyl B-D-l-
thiogalactopyranoside (IPTG) with a ﬁnal concentration lmM was used to induce the
expression of the defensin gene in E. coli cells. The matured zebra mussel defensin with
poly-histidines at its carboxyl-terminus (C-terminus) were abundantly produced after six-
hour of IPTG induction. The 6XHiS-tagged defensin was then puriﬁed with the Ni-NTA
Agarose ﬁlled polypropylene column (Qiagen) under the naive condition. The detailed
puriﬁcation steps were also available in the manual of QIAexpressionistTM (Qiagen). The
puriﬁed 6XHiS-tagged defensin was overnight desalted with Slide-A-Lyzer® Dialysis
Cassette (Therrno Scientiﬁc, Rockford, IL) immersed in 1X TAGZymeTM Buffer (20 mM
NaH2P04, 0.15 M NaCl, pH 7.0) at 4 °C. The 6XHis-tag at the C-terminus was then
cleaved by DAPaseTM from TAGZymeTM Kit (Qiagen). The cleavage and enzyme

removal were described in the TAGZymeTM Handbook (Qiagen). The cleaved product

153

 

was detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)

that was prepared according to the protocol written by Walker (Walker 2002).

The antimicrobial activity of recombinant Dpd

To have the recombinant defensin recovering the maximum activities, the mature
defensin was refolded under laboratory conditions (Gueguen et al. 2006). First of all, the
zebra mussel defensin dissolved in 1X TAGZymeTM buffer was overnight dialyzed at
4 °C with refolding buffer containing 100 mM Tris-HCl at pH 8.0. Then the cleaved
defensin protein was reduced in the refolding buffer in the presence of 100 mM
dithiothreitol (DTT). Finally, the defensin solution was dialyzed again at 4 °C with pure
refolding buffer overnight. The defensin in refolding buffer was allowed to refold at room
temperature for 48 hours. The concentration of the refolded protein was detected by
QubitTM ﬂuorometer with the Quant-iTTM Protein Assay Kit (Invitrogen). Then a series of
dilutions were made to create a set of Dpd solutions with different concentrations. The
minimal inhibition concentrations (MICS) of the Dpd were tested with radial diffusion
assay following the steps described by Lehrer et al (Lehrer et al. 1991). The
microorganisms we used included nine Gram-negative bacteria (Aeromonas salmonicida
salmonicida, Motile Aeromonas, Morganella sp., Edwardsiella tarda, E. coli DHSa,

F lavobacterium sp., Plesiomonas shigelloides, Pseudomonasﬂuorescens, and

Shewanella putrifaciens), four Gram-positive bacteria (Bacillus megaterium ATCC14581,
Carnobacterium maltaromaticum ATCC27865, Enterococcusfaecalis ATCC19433,
Micrococcus leteus ATCC4698, and Staphylococcus aureus aureus ATCC12598), and
one yeast (Saccharomyces cerevisiae). Except for E. coli DHSa and the Gram-positive

bacteria, all the other microorganisms were isolated from ﬁsh disease cases by the

154

 

 

Aquatic Animal Health Laboratory at Michigan State University and identiﬁed by
biochemical and/or molecular testing. The bacteria and yeast isolates were inoculated and
incubated in 3% trypticase soy broth (TSB) at room temperature overnight. One milliliter
aliquot of each culture was transferred into a 1.5 ml microcentrifuge tube. The
microorganisms were centrifuged at 900 X g for 10 minutes at 4 °C. After the supernatant
was discarded, the pellet of each microorganism was re-suspended with 1 ml ice cold

phosphate buffered saline (PBS, pH 7.4). The optical density of the suspension was

measured at 620 nm. The 1 ml suspended solution of each microorganism (4 X 106 CF U)

was mixed with 20 ml previously autoclaved warm (42°C) PBT (PBS with 0.05% Tween
20, pH7.4) containing 3 mg TSB powder and 1% low-electroendosmosiS-type (LE)
agarose (Applied Biosystems /Ambion, Austin, TX). The mixture was poured into a 150
X 15 mm Petri-dish after the microorganism was well dispersed. After the medium was
solidiﬁed, a ~10 mm deep layer was formed at the bottom of the Petri-dish. A sterile
disposable transfer pipet (VWR International, West Chester, PA) was then used to make
wells with ~3 mm diameter evenly spaced on the surface of the medium plate. Five
microliters of each Dpd dilution was added in each well and incubated at room
temperature for at least three hours. On the top of each plate, sterilized overlay medium
containing 6% TSB and 1% LE agarose was poured at 42 °C. The double layer medium
plates were kept in room temperature overnight to 24 hours.

After the inhibition rings were visualized on each plate, the diameter of the ring
was measured with 10 X Hastings Measuring Magniﬁer (Bausch & Lomb Inc., Rochester,
NY). The diameters of the inhibition rings (y—axiS) were plotted against the logic values

of peptide concentrations (x-axis). The linear regression of the diameter / concentration

155

was estimated and the x-axis intercept of the regression line corresponds to log (MICS)

(Ciomei et al. 2005).

Determination of Dpd expression in zebra mussel tissues

The RNA ﬂuorescent in situ hybridization (FISH) was performed to identify the
expression levels of defensin in different zebra mussel tissues. The zebra mussel was
ﬁxed in Davidson’s ﬁxative solution. One liter solution includes 115 ml glacial acetic
acid, 330 ml 95% ethanol, 220 ml 100% formalin and 335 m1 double distilled water with
pH 3-4 (Andrade 81 al. 2008). The mussel was conserved in the ﬁxative solution for 24 a
hours until the mussel Shell is dissolved. The ﬁxed mussel was buried in parafﬁn and
sectioned along the longitude axis of the mussel. The 5 pm thick sectioned sample was
put on an RN ase- and DNase- free glass slide. The probes used for the FISH were RNA
probes which were complementary strands of the Dpd mRNA. The PCR product
(ampliﬁed with primer BG11E08-RT-Fw and BG11E08-RT-Rv) of the gene fragment
was cloned into pGEM®-T Easy Vector Systems (Promega U.S., Madison, WI). Then
the positive colony was picked for sequencing with M13 forward primer to ﬁnd the
direction of the insert of the recombinant plasmid. The PCR was applied on the selected
colonies with both M13 forward and reverse primers. The PCR products were puriﬁed
with Wizard SV Gel and PCR Clean-Up Sysyem (Promega) and used as the templates of
RNA transcription. The T7 RNA polymerase was selected based on the insert direction of
the recombinant plasmid to synthesize the antisense RNA. An RNA probe of zebra
mussel cytoplasmic actin (AF 082863) gene was also produced in the same way as a

reference. Meanwhile, the sense-strand of Dpd probe was also synthesized as negative

control (NC).

156

 

The probe for Dpd (Dpd-R) was labeled with ﬂuorescent dye Alexa 594
(Invitrogen) while the probe for actin (Act-G) was labeled with Alexa 488 (Invitrogen).

The steps of the RNA synthesis and labeling can be found in the manual of either FISH

TagTM RNA Red Kit (Invitrogen) or Fish TagTM RNA Green Kit (Invitrogen). The

probe Dpd-R and ACT—G were used to hybridize with the same zebra mussel whole body

section Slide. The RNA FISH was performed following the recommended protocol in the V?

appendix of the Fish TagTM RNA Kit instruction. The hybridized slides were visualized

 

by Olympus BX41 (Olympus America Inc., Center Valley, PA) microscope under the “+-
excitation of mercury lamp and the images were captured by Olympus DP25 digital
camera (Olympus America Inc.). The image processing software DP2-BSW (Olympus
America Inc.) was used to combine the different color channels in a single image. To
enhance the contrast of Signals given by Dpd and actin, the blue color instead of green
was assigned to actin by the soﬁware. Similarly, the probe NC was also labled with
Alexa 594 and hybridized to a Slide with Alexa 488 labeled actin as background.

The quantitative RT-PCR (qRT-PCR) was used to detect the relative abundance
of defensin mRNA within the different tissues of zebra mussels. The Shells of the zebra
mussels used in this assay were swabbed with 70% ethanol, followed by a soak in 150 ml
sterilized double distilled water containing 5,000 U penicillin, 5 mg streptomycin, and 10
mg gentamicin for thirty minutes (Davids & Yoshino 1998). Thereafter, the mussels were
transferred to sterilized double distilled water. The RNA samples were extracted from
hemocytes, feet, muscles, ctenidia, and mantles of zebra mussels. The hemolymph was
extracted from the adductor muscle of 10 zebra mussels as and mixed with equal volume

of a buffer that consisted of 0.05 M TRIS/HCl, 2% glucose, 2% NaCl (pH 7.4), as

157

 

described by Pipe et al (Pipe et al. 1997). The mixture was centrifuged at 900 g for 10
minutes at 4 °C. The other tissues were collected under a dissecting microscope with
sterile scalpels and scissors. The total RNA of each sample was extracted with 5-PRIME
PerfectPure RNA Cell & Tissue Kit (5 PRIME Inc, Gaithersburg, MD). The relative

expression level of zebra mussel defensin in each sample was estimated by a one-step

qRT-PCR performed with Power SYBR® Green RN A-to-CTTM l-Step Kit (Applied

Biosystems). The paired primers designed for full length Dpd cloning (BG11E08-RT-Fw

and BG11E08-RT-Rv) were used in qRT-PCR. The primers for internal reference of the

qRT-PCR were also designed by PrimerExpressTM based on the sequence of zebra

mussel 18S ribosomal RNA (18S—Fw: 5’- GAC ACG GCT ACC ACA TCC AA -3’ and
l8S-Rv: 5’- CTC GAA AGA GTC CCG CAT TG -3’). The thermocycler used for qRT-
PCR was a Mastercycler ep realplex S (Eppendorf, Westbury, NY). The thermocycler
program used for this one-step RT-PCR was 30 minutes at 48 0C, 10 minutes at 95 °C and

40 repeats of a 2—Step temperature cycle (15 seconds at 95 °C and 1 minute at 60 °C). The

calculation of gene expression. level was accomplished with Z'AACt algorithm using cycle

threshold (Ct) of PCR described by Livak and Schmittgen (Livak & Schmittgen 2001).

This experiment was repeated once to have two biological replicates.

The expression proﬁles of Dpd in the different statuses of byssogenesis

In this assay, the zebra mussels were divided into two different groups. The ﬁrst
group contained mussels fully attached underwater to Petri-dishes for at least 30 days. In
the second group, the mussels were detached by removal of the byssal threads with

scalpels at their stems; thereafter, they were put back into water. The mussels were

158

 

randomly taken from both groups at day 1, day 2, day 3 and day 4 after the detachment of
the mussels in the second group. The total RNA samples were extracted from the feet of
the mussels collected every day. The qRT-PCR was performed with the RNA samples
from both groups. Comparisons were made between the two groups at the same day or
between two time points in the same group. The steps of qRT-PCR were previously
mentioned in the sections above and a two-way ANOVA test has been performed to

analyze the effects of the byssogenesis status and time to the expression level of Dpd.

159

Results

The full length gene sequence and primary protein structure of Dpd

The full sequence of the zebra mussel Dpd gene obtained from our study
consisted of 409 bp, including an open reading frame which encoded Dpd protein with 76
amino acid residues. The BLAST result showed that Dpd was homologous to a number
of defensins identiﬁed from a variety of Species in non-redundant protein sequence
database (nr), such as the Japanese disc abalone (Haliotis discus hannai, ABF 69125), the
parasitic wasp (Nasonia vitripennis, NP_001159944), the bumblebee (Bombus ignitus,
AAQ90412), the western honey bee (Apis mellifera, NP_001011616), and the scarab
beetle (Anomala cuprea, BAD77966), etc. Phylogenetically, the Dpd is closest to the
defensin of Japanese disc abalone. The defensins of these two molluscan species were
closely related to two types of defensins identiﬁed from Apocrita (wasps, ants and bees)
(Figure 6-1). Aligned with another 15 defensin molecules from 14 Species, the Dpd was
found to have a conserved motif of common defensins, six conserved cysteine residues.

Two continuous glycines in all the defensin molecules were also observed (Figure 6-2).

The three dimensional structure of Dpd

The tertiary structure of Dpd was predicted to be the closest to an insect defensin
A model (IICA) in protein database. Both of these two molecules Share a number of
structural similarities. First, both had one a-helix and two B-sheets. Second, both have Six

cysteine residues in their amino acid sequences with three disulﬁde bonds. Last, the

disulﬁde bonds in Dpd (Cys28-CySS7, Cys43-Cys62, and Cys47-CySS4) and in insect

160

 

 

defensin A (CySS7-CyS84, Cys70-Cy590, Cys74-Cy582) existed in Similar positions

(Figure 6-3). On the other hand, there was some minor difference between these two
defensin molecules. The u—helix and B-sheets in Dpd are shorter than the corresponding
secondary structures in the insect defensin A. The a-helix in Dpd consists of seven amino
acid residues, while in the insect defensin A, there are nine amino acid residues.
Similarly, both B-sheets in Dpd had four amino acid residues, which was one residue
shorter than those in the insect defensin A. Moreover, the 3D structure of the insect
defensin A was more compactable than that of Dpd (Figure 6-3). The other two

molluscan defensin molecules, MGD-l and Cg-def, were structurally close to each other.

Both of them have four disulﬁde bonds instead of three: Cys4-Cy325, Cyle-Cys33,
Cysl4-Cys35, and CysZI-Cys38 in MGD-l (Figure 6—3D); CyS4-Cys25, CyS1 l-Cys34,

CyslS-Cys36, and CySZO-Cys39 in Cg-def (Figure 6-3D).

Besides the conserved cysteine residues in both defensin molecules, there were

also some other amino acid residues with hydrophobic or positively-charged characters

conservatively distributed. In Dpd, His46 and Arg65 were positively charged when at pH

7.0 (Figure 6-4A) while the His73 and Arg93 were also observed in insect defensin A

(Figure 6—4B). They were all considered as conserved amino acid residues based on their

primary protein structures (Figure 6-4C). Similarly, there were ﬁve conserved

hydrophobic amino acid residues found in both Dpd (Leu30, Leu31, lle36, Ala“, and

Na“) and insect defensin A (Leusg, Leu60, lle65, Ala69, and Alan).

161

The minimal inhibition concentrations (MICS) of recombinant Dpd

The in vitro expressed recombinant Dpd contained the complete functional

domain including 58 amino acid residues (from Ala22 to Gly75). Among all the

microorganisms used in MICS test, the growth inhibition was observed in four Gram-
negative bacteria strains. The lowest MIC was found in the inhibition test against
Morganella sp. (0.35 uM) while the P. shigelloides showed the second highest sensitivity
to the Dpd with the MIC equal to 0.43 uM. E. tarda and E. coli Showed stronger
resistance to Dpd with the MICS 1.16 uM and 6.46 uM, respectively. Only one Gram-
postive bacterium strain Staphylococcus aureus aureus was inhibited by Dpd with MIC
30.39 11M. With the concentration of Dpd less than 50 pM, there was no obvious

inhibition observed in the growth of the other bacteria, nor in the yeast S. cerevisiae.

The in situ expression of Dpd in zebra mussel

The ﬂuorescent in situ hybridization with Dpd RNA probe demonstrated the
constitutive expression of Dpd in multiple zebra mussel tissues. The Dpd was mildly
expressed by the epithelial cells covering the surface of the mussel’s foot (Figure 6-5A)
while it was abundantly expressed in a number of granular cells in the middle section of
the foot (Figure 6-SB). A strong positive signal was noted at the position of thread-
forming glands in zebra mussel foot (Figure 6-5C). The negative control of the FISH in
zebra mussel foot was shown in Figure 6—6 where only the actin probe was observed
(blue). The expression level of Dpd in the retractor muscle was also present, albeit the
signal was weaker than in the foot (Figure 6-7A). Hemocytes that were distributed within
the muscle Sinus exhibited a relatively strong signal (Figure 6-7B). In the section of zebra

mussel ctenidium (gill), the signal of Dpd mRNA was detected in the epithelial lamellae
162

 

but not in the water channel or connected tissue between the two epithelial lamellae
(Figure 6—7C). The expression of the Dpd was also found in zebra mussel mantle,
exclusively in the mantle epithelium (Figure 6-7D). In general, all hemocytes, regardless
of their location, exhibited strong signals (Figure 6—7E). Germ cells in the gonads
exhibited strong Dpd Signals as well (Figure 6-7F). The signiﬁcant level of Dpd
expressions was also observed in the digestive gland cells (Figure 6-7G) and in the
intestinal mucosa] lining (Figure 6-7H). The negative control for the FISH in the zebra

mussel tissues other than foot was displayed in Figure 6-8.

Determination of relative expression level of Dpd in zebra’ mussel tissues by
qPCR

The expression levels of Dpd within zebra mussel hemocytes, foot, retractor
muscle, ctenidium, and mantle were different, and the order from highest to lowest was
foot, retractor muscle, ctenidium, mantle, and hemocytes (Figure 6-9), sequentially. Out
of the ﬁve tissues tested in this assay, the zebra mussel foot had the most abundant Dpd
expression compared to the other four tissues (P < 0.01). The muscle tissue had the
second highest expression level of Dpd. The statistical analysis indicated that the
expression level of Dpd in muscle is dramatically higher than in hemocyte, ctenidium,
and mantle, but Signiﬁcantly lower than in the foot (P < 0.01). The Dpd was also
expressed in the other tissues tested, hemocyte, ctenidium, and mantle; however, the
expression levels were fairly low compared to that in foot or muscle (Figure 6-9). The
comparisons were also made between any two of the rest of the tissues and it turned out
that the expression levels of Dpd in hemocyte, ctenidium, and mantle of zebra mussel

were not Si gniﬁcantly different (P > 0.05)

163

The expression of Dpd in the zebra mussel foot during the early stages of
byssogenesis

To understand how the expression of Dpd iS associated with the zebra mussel
byssogenesis, quantitative PCR was used to compare the expression levels in two
different byssogenesis status, non-byssogenesis and byssogenesis status. After one-day
adhesion, the zebra mussel did not show signiﬁcant difference of Dpd expression in the
feet of non-byssogenesis and byssogenesis individuals. Starting from Day 2 post
initiation of byssogenesis, the expression levels of Dpd in the individuals with
byssogenesis were signiﬁcantly higher than the non-byssogenesis (P < 0.05 at Day2, P <
0.01 at Day 3, 4, and 5). It suggested that the higher expression levels were intended to

associate with the mussels under the byssogenesis status (Figure 6—10).

164

 

Discussion

Compared to the studies on oyster defensin (Cg—def) (Gueguen et al. 2006), the
Dpd seems to have a broader range of tissue distribution in the zebra mussel. The
expression level of Cg-def was found signiﬁcantly higher in the mantle than any of the
other tissues of C. gagas including hemocytes, adductor muscle, digestive gland, gills,
heart, and labial palps. The FISH analysis in our study suggested that the Dpd were

constitutively expressed in all the tissues or cells examined, especially the epithelial

 

lining, such as ctenidium (gill), mantle, intestine, and foot. This is similar to what has
been reported on vertebrate B-defensin, which was overly expressed in epithelium (Yang
et al. 1999). That defensin was also expressed in the zebra mussel ovary is also consistent
with the previous report (Radhakrishnan et al. 2007), which demonstrated that vertebrate

B-defensin was expressed in the murine ovary. The authors suggested that the presence

 

of defensin in murine ovaries has the function of protection of fertilization. It was
noticeable that the relative abundance of Dpd is extremely high in the zebra mussel foot.
The FISH results demonstrated that the Dpd expressed by the epithelial lining of the foot
was much less than Dpd expression in byssal glands. Such high level of Dpd within the
byssus gland cells suggests that some of the zebra mussels attempt to protect the
relatively young byssal threads from degradation in the bacteria-rich aquatic
environment. In a previous study, it was demonstrated that these same exocrine glands
produce a number of foot proteins among other secretory proteins. The Dpd, which had
extremely high expression levels within the foot, and was continuously expressed in high
levels during the byssogenesis (Figure 6-10), may be the ideal antimicrobial molecule to

provide sufﬁcient protection during the byssogenesis. Whether Dpd is the only

165

antimicrobial peptide employed by the zebra mussel to protect its byssus or not is
currently unknown and deserves further investigation.

The Dpd showed certain growth inhibition activity to four Gram-negative bacteria
including Morganella sp., P. shigelloides, E. tarda, and E. coli, as well as a Gram-
positive bacterium Staphylococcus aureus aureus, but no obvious inhibition to the only
one Gram-positive bacterium or the yeast species used in this study. Both Cg-def and
MGD-l were demonstrated to mostly inhibit the growth of many Gram-positive bacteria
with only limited inhibition to some E. coli strains. The difference in the activities of these
three antimicrobial peptides can be possibly attributed to the difference in their three-
dimensional structures. Both Cg-def and MGD-l have eight cysteines in their amino acid
sequences, which help proteins to form four disulﬁde bonds, instead of the three that exist
in most insect defensin molecules. Moreover, the homologues searching analysis
indicated that the Dpd is closer to the defensins from insects, such as that of the bee (N.
vitripennis, B. ignitus, and A. mellifera) and beetle (A. cuprea), instead of their molluscan
relatives. In insects, the antimicrobial activities of defensins have broader spectrum
against the pathogens compared to Cg-def and MGD-l, including gram-positive/
negative bacteria and fungi (Hoffmann & Hetru 1992; Cociancich et al. 1993; Hetru et al.
1998).

The recombinant Dpd peptide used in the MIC study was produced in E. coli and
refolded in vitro. There are concerns about the correct in vitro refolding of eukaryote
proteins with multiple disulﬁde bonds when expressed in prokaryote cell system (Harder
et al. 2001). However, the same system has been successfully applied to study several

defensin species, such as human a— and B-defensins (Ouellette et al. 2000; Harder et al.

166

 

 

2001), as well as the oyster defensin (Gueguen et al. 2006). The same method we used in
this study produced considerable amount of proteins with the MICS analysis. It was
noticeable that the recombinant proteins puriﬁed under the denatured conditions did not
Show any inhibition activity against the growth of bacteria and ﬁingus. This suggests that
the refolded zebra mussel defensin retains some of its biological activities. Due to the
lack of the Similar study with freshwater bivalves, the molecules we can use to compare
to the Dpd do not exist to the best of our knowledge.

The known functions of defensin have always centered on its microbicidal effects.
However, in this study, it is the ﬁrst time that defensin was found associatd with a
physiological process such as byssal thread formation. Given that the Dpd is largely
produced by zebra mussel byssus gland cells, it has to be transported through the byssus
apparatus to the outside of the mussel where the microbial communities exist. Whether
the role of defensin in byssogenesis is limited to biological protection of its microbial
degradation or is an integral part of a cascade that leads to the production, secretion, or
activation of another proteins remains to be determined. AS displayed above (Figure 6-
11), Dpd level was elevated as. early as Day 2 of byssogenesis denoting to its importance
to the process. The data obtained in this study Shed more light on the process of
byssogenesis in general, and the host defense mechanisms of this nuisant Species in

particular.

167

 

IOU AA 090412 Bombus ignitus

 

50 IACI04166 Bombus hypocrite sapporoensis
._ NP 001011616 Apis mellifera
91 ,_ ACH96387 Apis cerana cerana

 

 

 

 

76

 

 

—q;—— NP 001159944 Nasonia vitn'pennis B
91 NP 001159943 Nasonia vitn'pennis A

 

 

 

 

 

 

GU139954 Dreissena polymorpha
88 ABF69125 Haliotis discus hannai
l— BAD77966 Anomala cuprea A
15L BAD77967 Anomala cuprea B
CAI83768 Pseudoplectania nigre/Ia

 

 

 

 

 

 

 

 

 

0 1

Figure 6—1 Phylogenetic analyses of zebra mussel defensin (Dpd) and the homologues.
Neighbor-Joining bootstrap test was used for this analysis. The value above each node
stands for the percentage of bootstrapping after 1000 replications. The Dpd was mostly
close to a defensin molecule identiﬁed from the Japanese disc abalone (Haliotis discus
hannai). The clade of Dpd and abalone defensin was phylogenetically related to a group
of defensins from Apocrita (wasps, ants and bees). The defensin A and B of the scarab
beetle (Anomala cuprea) had less identity to the defensin of Mollusca and Apocrita. The
fungus (Pseudoplectania nigrella) defensin was used as an outgroup.

168

TCDLLG---GVWlLGADTACAGHCYTLN-HPGG

  

HCEG-—-GY(YCRM

      
 

    

   

   

      

TCDLLSHMGGFSFG-DSACAAHCIVLH-HNGGHLSN---GVCVCRB
CP ------------ NDYPCHRHCKSIPGRXGUYCGGXHRLRCTCY%
P ------------ NNYQCHRHCKSIPGRCGGYCGGWHRLRCTCY%

P ------------ NNYACHQHCKSIRGYCUUYCASWFRLRCTCY%

P ----------- LNQGACHNHCRSlR-RRGUYCSGIIKQTCTCY%

P ----------- LNQGACHNHCRSlG-RRGUYCAGIlKQTCT(Y%

P ----------- FNQGACHRHCRSlR-RRGGYCAGLIKQTCTLYM

P ----------- FDQGACHRHCQS[G-RRGGY(AGFIKQTCT(Y%
-—-—GTGIN-HSACAAHOLLRG-NRUUYCNGK--GV(VCRW
----FGGVVGDSACAANCLSMG-KAGUSCNG---GICECR%

--——lKGVAEHSACAANLLSMG-KAUURCEN---GV(LCRW
P ----------- LDQMQCHRHCQTITGRSGUYtSGPLKLT(T(YN
—---——--FNQGACHRHORSlR-RRGUYIAGLFKQT(T(Y%
P ----------- LNQGACHRHCRSIR-RRUGYtAGFFKQTIT(Y%

Figure 6—2 The multiple alignment with the homologous insect and mollusk defensins.
The conserved amino acid residues and their positions in defensin molecules are
highlighted in dark. Typically, the defensin molecules have at least six conserved
cysteines which form three disulﬁde bonds. Also, there are two glycines that are
conserved in mollusks, insects, arthropods and fungus. Besides the Dpd (1), another 15
defensins identiﬁed from 14 different species were selected in the multiple sequences
alignment, including the Japanese disc abalone (2, Haliotis discus hannai, ABF69125),
the blue mussel (3, Mytilus edulis, P81610), the Mediterranean mussel (Mytilus
galloprovincialis, 4, P80571 and 5, AAD52660), the American dog tick (6, Dermacentor
variabilis, AA024323), the bush tick (7, Haemaphysalis longicorm's, BAD93183), the
cattle tick (8, Boophilus microplus, AAO48943), the deer tick (9, Ixodes scapularis,
AAV74387), the Northern blowﬂy (10, Protophormia terraenovae, P1089l), the
parasitic wasp (11, Nasonia vitripennis, NP_001159944), the bumblebee (12, Bombus
ignitus, AAQ90412), the dragonﬂy (l3, Aeschna cyanea, P80154), the fat-tailed scorpion
(14, Androctonus australis, P56686), the yellow scorpion (15, Leiurus quinquestriatus,
P41965), and a fungus (l6, Pseudoplectania nigrella, CA183768).

169

 

Figure 6-3 The tertiary structures of Dpd and disulﬁde bonds.

A. the 3D structure of Dpd; B. the 3D structure of insect defensin A(11CA). Both Dpd
and insect defensin A had similar secondary structures, one a—helix (red) and two [5-
sheets (yellow). The tertiary structures of Dpd and insect defensin A are also very similar.
The relative positions of six Cysteins (labeled with blue) and the three disulﬁde bonds
(blue lines between Cysteines) formed by Cysteines in the molecules were conserved.
The two defensin molecules from another two mollusks, Mytilus galloprovincialis and
were also compared to Dpd. C. defensin MGD-l of M galloprovincialis (lFJN); D.
defensin Cg-def of Crassostrea gigas (2B68). Both MGD-l and Cg-def have four
disulﬁde bonds instead of three in Dpd and insect defensin A.

This ﬁgure is in color.

170

 

C

D. polymorpha I

 
 
 
 

HicE - YCY PGT FS 71
P. terraenovae A ' ‘ YCNIKIVCV N— —— - 94
Figure 6-4 The conserved amino acid residues in Dpd.
The amino acid residues of both Dpd and insect defensin A are characterized with
different colors. Both hydrophobic (green) and positively charged (red) residues at pH7.0
are demonstrated. The two positively charged residues His and Arg in Dpd (A) were
also observed in insect defensin A (B) at conserved ositions (His73 and Arg”). Five
hydrophobic residues were found in both Dpd (Leu3 , Leu31, Ile’é, Ala“, and Ala“) and
insect defensin A (Leusg, Leuf’o, Ile65, Ala69, and Alan). The six conserved Cysteines are
also labeled (yellow) in both tertiary structures (A and B) and primary structures (C) of
both defensin proteins. The other conserved amino acid residues are highlighted with
dark and the residues with similar biochemical characters at same positions are colored
with grey.

This ﬁgure is in color.

171

1011111

2pm

 

Figure 6- 5 FISH result of Dpd within zebra mussel foot.

The Dpd was observed abundantly expressed in the zebra mussel foot. The Dpd was
slightly expressed in the epithelial cells on the surface of zebra mussel foot (A). The main
resource of expressed Dpd was observed in the middle of zebra mussel foot (B), ﬁlled
with byssus gland cells (C). The blue background was the signal given by the stained
cytoplasmic actin probe. The pink ﬂuorescent color was the combination of the Dpd
probe signal (red) and the cytoplasmic actin probe (blue).

This ﬁgure is in color.

172

- B
c D
2 pm
G
2 11
Figure 6—8 The negative control of FISH Dpd in other tissues of the zebra mussel.
The in situ hybridization with sense-Strand of Dpd RNA probe (red) and anti-sense-Strand
actin RNA probe (blue). The signal for actin probe can be clearly observed in all tissues
but there is no signiﬁcant ﬂuorescent signal of sense-strand of Dpd. (A) Retractor
muscle; (B) Ctenidium; (C) Mantle; (D) Hemocytes; (E) Gonad; (F) Digestive gland; (G)

Intestine.
This ﬁgure is in color.

 

177

 

 

Figure 6-7 FISH results of Dpd in other tissues of the zebra mussel.
The Dpd was also found also expressed in other tissues besides that the zebra mussel foot
(red). The expression level of Dpd in the retractor muscle was not Signiﬁcant (A). The
sporadically distributed Dpd mRNA was likely from the hemocytes scattered in the
muscle (B) Since the existence of hemocyte sinus in it. The transcription of Dpd was also
detected from ctenidia (C), out layers of mantles (D), hemocytes (E), gonads (F),
digestive gland cells (G), and epithelial cells of intestines (H). The cytoplasmic actin was
stained as background (blue).

This ﬁgure Is in color.

174

 

2 pm

Figure 6-8 The negative control of FISH Dpd in other tissues of the zebra mussel.

The in situ hybridization with sense—strand of Dpd RNA probe (red) and anti-sense-strand
actin RNA probe (blue). The signal for actin probe can be clearly observed in all tissues
but there is no signiﬁcant ﬂuorescent Signal of sense-strand of Dpd. (A) Retractor
muscle; (B) Ctenidium; (C) Mantle; (D) Hemocytes; (E) Gonad; (F) Digestive gland; (G)
Intestine.

This figure is in color.

175

1400 u:

 

Relative Expression Level

O‘Nwhalm‘lﬂ

Hemocyte Foot Muscle Mantle Ctenidium

Figure 6— 9 The relative abundance of Dpd in different zebra mussel tissues with qPCR.
The expression levels of Dpd in zebra mussel foot and muscle are signiﬁcantly higher
than other tissues (P < 0.01). The abundance of Dpd in the zebra foot is also signiﬁcantly
higher than that in muscle. The ANOVA with Tukey’s test was used for the multiple
comparisons. ** P < 0.01.

176

5 _ . I Non-byssogenesis
ii-

4.5 — a Byssogenesis

Relative Expression Level

 

 

Day 1 Day 2 Day 3 Day 4 Day 5 Day 6

Time Post Byssogenesis

Figure 6-10 The relative expression levels of Dpd under the condition of non—
byssogenesis and byssogenesis.

Compared to the non-byssogenesis status, the expression levels of Dpd in zebra mussel
with adhesion are signiﬁcantly higher at most time point post the start of byssogenesis
except for Day 1. The ANOVA with Tukey’s test was used for the multiple comparisons.
* P < 0.05; ** P < 0.01.

177

Table 6-1 The minimal inhibition concentrations (MICS) of Dpd against the bacteria and
yeast.

 

Microorganisms MIC (11M)

 

Bacteria

Gram-negative

Morganella sp. 0.35
Plcsiomonas shigclloidcs 0.43
E dwardsiella tarda 1.16
Escherichia coli DHSa 6.46
Aeromonas salmom'cida salmonicida >50
Motile Aeromonas >50
F lavobacterium sp. >50
Pseudomonasﬂuorescens >50
Shewanella putrifacicns >50
Gram-positive

Staphylococcus aureus aureus ATCC12598 30.39
Bacillus megaterium >50
Carnobacterium maltaromaticum ATCC27865 >50
Enterococusfaecalis ATCC19433 >50
Micrococcus luteus ATCC4698 >50
Fungus

Saccharomyces cerevisiae >50

 

178

 

CHAPTER VII

Conclusions and Future Studies

Conclusions

With the results obtained from this project, identiﬁcation of the genes associated
with the byssus activities, we were able to gradually understand the regulation
mechanism of zebra mussel attachment. The methodology of this study has been proven
sufﬁcient and reliable.

The application of suppression subtractive hybridization (SSH) cDNA technique
dramatically decreased the population of the genes of interest. In eukaryotes, the mRNAs
of a typical somatic cell are distributed in three frequency classes (Bishop et al. 1974;
Davidson & Britten 1979). The two classes with the most abundance (> 90% of total
mRNA population) consist of the genes involved in the common metabolism of the cells.
On average, the most prevalent class consists of about 10 mRNA species, each
represented by 5000 copies per cell, whereas the class of high complexity comprises
15,000 different species each represented by l-15 copies only (Soares et al. 1994). The
SSH cDNA library technique enriched the cDNAs unique to zebra mussel byssus by
subtracting the muscle RNA from the foot RNA. Compared to the 3000 genes obtained
from non-normalized hemocyte cDNA library of scallop Argopecten irradians (Song et
al. 2006) and Chlamysfarreri (Wang et al. 2009), respectively, the 750 ESTS generated
from normalized zebra mussel foot cDNA library with each of the rare genes represented
that is related to the function of the foot. By quantitative PCR, the genes selected from

179

the microarray were all conﬁrmed to be Signiﬁcantly expressed in zebra mussel foot
compared to the other tissues. The decrease of the interference caused by the gene
constitutively expressed in all the tissues dramatically improved the efﬁciency of the
cDNA library. However, to identify the genes related to byssus activities of zebra mussel
from more than 700 ESTS still seems unpromising until the application of the cDNA
microarray technology.

Our newly constructed Drcisscna polymomha microarray is the ﬁrst cDNA
microarray designed for the study of the freshwater bivalves and also the ﬁrst array to in
the study of bivalve underwater adhesion mechanism. The cDNA microarray developed
in this study contains 716 genes selected from the SSH cDNA library and represents the
genes unique to the byssus of zebra mussel. Based on the low false discovery rate (FDR)
of this microarray in identifying the differentially expressed genes (Chapter III), the zebra
mussel byssus cDNA microarray was proven to be efﬁcient and reliable.

With the applications of zebra mussel cDNA microarray, multiple differentially
expressed genes were found from different stages of byssogenesis. Furthermore, the
genes with the expression patterns varied in response to the change of environmental
factors and byssogenesis status were also identiﬁed. Among all these identiﬁed genes, not
only are the genes encoding the Dpfp-1 variants, but also some other genes with different
putative functions found to be involved. The protection to the byssus components
provided by zebra mussel has not been widely studied. However, the abundant expression
of D. polymorpha defensin (Dpd) in zebra mussel byssus gland makes us believe that the
byssus gland cells of zebra mussel, primarily as the byssal threads producer, also provide

essential protection to the proteins produced by themselves. Combined with the current

180

knowledge of zebra mussel attachment that was obtained from previous studies, the
results of our study partially explained basic questions of the molecular mechanism of
zebra mussel underwater adhesion, such as what kind of molecules are produced by zebra
mussel byssus glands, which genes were involved in byssogenesis At the same time, so
many questions have been raised that the follow-up investigation in this delicate

mechanism is necessary in particular.

181

 

 

Future Studies

For as long as history has been recorded, people have been fascinated with
underwater adhesion. Water can weaken many of the chemical bonds mediating adhesion.
There are many examples of both permanent and temporary underwater attachment that
are mediated by complex polymer glues (Waite 1987; F lammang 1996; Smith & Callow
2006). For some aquatic animals, adhesion is vital for survival and completion of their
life cycle. One such organism is the zebra mussel (Dreissena polymorpha), the bivalve
mollusk I have chosen to elucidate with regards to the molecular aspects of its underwater
adhesion. The ﬁrst stumbling block I faced in my studies was the severe lack of
background information on the details of underwater attachment processes employed by
the zebra mussel. Overwhelmed by the economic losses caused by the zebra mussel in the
Laurentian Great Lakes basin, managers and scientists focused their efforts on ﬁnding
creative ways to clean pipes from blocking by the invading mollusk, with relatively
minimal efforts directed towards studying its biology. It is therefore that I ﬁnd the study
of the physiology, defense mechanisms, and attachment mechanisms of the two nuisant
dreissenids (zebra and quagga mussels) a ﬁrst priority for future research and funding.

Studies addressing underwater adhesion by marine invertebrates, which live in
dynamic ocean environments by adhering tightly to underwater substratum using their
holdfasts, were the only sources available upon starting my research project.
Unfortunately, they were of little help as they focused on the biochemical nature of the
proteinaceous glue produced by the marine mussels. The approach used in my studies
ﬁlls a void in our knowledge of the zebra mussel physiology and underwater attachment.

Due to the successful application of cDNA microarray technique, many zebra mussel

182

 

novel genes were identiﬁed and their kinetics of expression followed along the
byssogenic cycle.

While the nature and functions of these genes have been identiﬁed, whether
through experiments performed in my studies or through the functions of their
homologues, there is a great need to deﬁne these genes, complete sequencing of them,
and deduce the nature of the proteins they encode and their potential functions. In this
respect, I would argue that the EGP genes constitute a ﬁrst priority in future research as

they may represent important components of a cascade leading to the formation of

 

adhesive particles, alone or combined with other secreted foot proteins. Good candidates
are the shematrin-like and neuropeptide (nlp) encoding genes; both were found
differentially displayed in most of the experiments performed in this study. The
shematrins were ﬁrst isolated from the mantle tissue of the pearl oyster (Pinctadafucata)
and deﬁned as a glycine-rich protein family believed to form a scaffold for various
structures (Yano et al. 2006). Whether zebra mussel shematrins are also essential for the
byssus structure warrants further studies. The functions of nlp family cover a broad range
including neuronal signaling, development, and antimicrobial activities (N athoo et al.
2001; Couillault et al. 2004; McVeigh et a1. 2008). Whether the nlp could be essential for
a kind of “sensation” by which the zebra mussel communicates with the newly formed
byssal thread, and thereby determine the surrounding environmental factors, including the
suitability of the substrates for adhesion, is deﬁnitely important to study in the future.
Shematrin and nlp are examples of those zebra mussel proteins that have hits in
the gene databases, however, almost 40% of the zebra mussel foot genes have had no hits,

and based on the experiments presented in Chapter III-V, these functionally-unknown

 

183

genes seem to play a vital role in zebra mussel adhesions. A study to elucidate their
function relative to byssogenesis is deemed necessary for future studies. Indeed, some of
the genes without putative functions in the library encode amino acid residues that share
great similarities with the zebra and blue mussel foot proteins.

The expressions of many genes in the zebra mussel byssus cDNA library seem to
be modulated by the ﬂuctuation in environmental factors and status of byssogenesis
(Chapter V). It was surprising to determine that most of the differentially expressed genes
are responsive to changes in one or two factors, and not to all four of the factors studied.
This observation sheds light on the fact that the expression of these genes may be very
Speciﬁc to one particular factor and not to another. For example, the expression of EGP
homologues was not affected by the current velocity. Since EGP seems to play a central
role in underwater adhesion, it is empirical that future studies address the role of the
surrounding environmental factors in the initiation and magnitude of EGP gene
expression. Some other examples werealso found Speciﬁc to certain factors, such as a
polypeptide release factor encoding gene whose expression proﬁles were only altered by
water temperature; and choriogenin H-like protein encoding gene with the expression
pattern only affected by temperature and current velocity. This suggests that zebra mussel
underwater adhesion is not a universal mechanism, rather a number of either independent
or sequential responses to various stimuli.

In our study, only two different levels were created for each factor, which is not
enough to determine a solid correlation between the expression of certain zebra mussel
genes identiﬁed in this study for their role in byssogenesis and surrounding

environmental factors. In previous morphological studies, multiple levels have been

184

 

 

created for each environmental factor to study the regression of the byssogenesis rate and
each experimental factor (Clarke & McMahon l996b, 0; Clarke & McMahon 1996a;
Clarke 1999). Applying multiple levels in each factor will allow further study of how the
expression levels of certain genes are associated with changes within the environment.
This will help us to unravel the regulation mechanism of the byssus unique genes during
the byssogenesis in natural conditions. Additionally, the genes found to be involved in
byssogenesis in the present study Should be further analyzed, their functions deciphered,
and factors affecting their expression determined.

The time course study involving the developed cDNA microarray successfully
identiﬁed genes whose differential expression patterns coincided with morphological
changes occurring in byssal threads (Chapter IV). For example, homologous to annexin
Vll encoding gene, the gene BG20_GO4 was only upregulated 12 hours after the start of
byssogenesis. Within the ﬁrst 12 hours, the temporary threads are dominant in freshly
produced byssal threads, and the protein product of BG20_GO4 gene probably provides
material for temporary threads. This particular gene is an excellent candidate for future
studies as it may provide details regarding the early stages of byssogenesis.

Besides the zebra mussel models used in our microarray analysis, adhesion vs.
non-adhesion (Chapters III and V) and byssogenesis vs. non-byssogenesis (Chapter IV),
there is another model that may also be studied with our cDNA microarray: attachment vs.
detachment. According to the observations in earlier studies, both adult and juvenile
zebra mussels can detach themselves and move using their feet. The crawl rate of
juvenile zebra mussel is about 7 cm/night while the adults can move at a rate up to 36

cm/h (Ackerman et al. 1994; Toomey et al. 2002). This suggests that zebra mussels can

185

 

 

detach themselves, probably by dissolving their previously secreted glue with other
molecules of lytic nature (Claudi & Mackie 1994). Finding the nature of the zebra mussel
lytic molecule may be a cornerstone toward developing control agents against the nuisant
organism. If a laboratory model can be developed to force attached mussels to detach,
then the cDNA microarray can be used to identify the gene(s) encoding the glue-
dissolving molecule(s) employed by the zebra mussel.

In order to settle down in a particular area, zebra mussels sense the substance
surfaces and surrounding environment using their feet during movement (Eckroat et al.
1993). This cannot be accomplished without the presence of certain sensory elements.
The conduction through the pedal ganglion of Mytilus has been reported (Richards 1929),
but to date, this phenomenon is very under—appreciated and little understood. It remains
unclear whether the zebra mussel foot sensing process is accomplished by the nerves and
pedal ganglion like the blue mussel, or if the zebra mussel employs other yet to be
discovered, mechanisms. AS stated previously, the genes homologous to neuropeptide
encoding genes can be potentially involved in the neuronal signaling (N athoo et al. 2001;
Couillault et al. 2004; McVeigh et al. 2008). The analysis of this family of genes may
contribute to the understanding of surface recognizing mechanisms employed by zebra
mussel, which is a primary step of underwater attachment.

The Dpfp-1 encoding gene has been identiﬁed in most microarray experiments
performed in this study. The Dpfp-1 contains tandemly repeated and segregated motifs: a
heptapeptide and a tridecapeptide consensus motif. At least ten Dpfp-1 variants exist in
the protein mass puriﬁed from zebra mussel byssus. The multiple protein variants

encoded by a Single Dpfp gene is probably caused by an alternative splicing mechanism

186

 

 

 

on gene transcription level which is Similar to the other collagen-like proteins
(Pihlajaniemi & Tamminen 1990; Rzepecki & Waite 1993). (The transcripts of the Dpfp-l
identiﬁed in our data can be directly used to study the Dpfp-l expression regulation in
transcription level, which will be essential to obtain more information on byssal thread
formation in the zebra mussel.

Although the microarray assays greatly narrowed down the genes Of our interests,
there are still a large number of genes that need to be studied. The difference of the gene
expression on mRN A level may not lead to the change of protein level due to the decay
of mRNA in post-transcriptional regulation (Jacobson & Peltz 1996; Ross 1996). Among
the three previously studied D. polymorpha foot proteins, only two of them were puriﬁed
directly from the zebra mussel foot (Rzepecki & Waite 1993). It is also likely that, among
the EGP genes and nlp-like genes identiﬁed by microarray, there are only a few whose
differential transcriptional patterns will be also reﬂected by the differential translation
proﬁles, which means most of the identiﬁed genes may not be as heavily involved in
zebra mussel byssus activities, as one may think. At this point, analyses of the proteins
encoded by the genes identiﬁed by microarray, such as Western blot, spectrophotometry,
and enzyme-linked immunosorbent assay (ELISA), will help us validate the results on
protein level, thereupon discover the genes that are indeed involved in zebra mussel
adhesion.

On the protein level, it is hard to locate the zebra mussel byssal thread adhesive
proteins due to the lack of commercially available antibodies. The difﬁculties in
obtaining the pure adhesive proteins from zebra mussel byssal thread plaques are the

main obstacle in producing the antibodies against adhesive proteins (Farsad et al. 2009).

187

 

With the genes identiﬁed from this study, we can easily produce a large amount of pure
foot proteins through in vitro expression systems; whereafter, the antibodies against many
types of foot proteins can be developed with the recombinant foot proteins. The
availability of the foot protein antibodies will contribute to the discovery of novel
proteins produced by zebra mussel byssus glands, as well as the structure of the bulk

proteins in the plaques of zebra mussel byssal threads.

188

 

REFERENCE

Ackerman J .D., Smith B., Nichols S.J. & Claudi R. (1994) A review of the early life
history of zebra mussels (Dreissena polymorpha): comparisons with marine
bivalves. Can J Zool 72, 1169-79.

Allen J .A. (1985) The recent Bivalvia: their form and function. Academic Press, New
York.

Altschul S.F., Gish W., Miller W., Myers E.W. & Lipman DJ. (1990) Basic local
alignment search tool. J Mol Biol 215, 403-10.

Anderson K.E. & Waite J .H. (1998) A major protein precursor of zebra mussel

(Dreissena polymorpha) byssus: deduced sequence and signiﬁcance. Biol Bull
194, 150—60.

Anderson K.E. & Waite J .H. (2000) Immunolocalization of Dpfpl, a byssal protein of the
zebra mussel Dreissena polymorpha. J Exp Biol 203, 3065-76.

Andrade T.P.D., Redmana R.M. & Lightner D.V. (2008) Evaluation of the preservation
of Shrimp samples with Davidson's AF A ﬁxative for infectious myonecrosis virus
(IMNV) in situ hybridization. Aquaculture 278, 179-83.

Angarano M.B., McMahon R.F. & Schetz J .A. (2009) Cannabinoids inhibit zebra mussel
(Dreissena polymorpha) byssal attachment: a potentially green antifouling
technology. Biofouling 25, 127-38.

Arumugam M., Romestand B., Torreilles J. & Roch P. (2000) In vitro production of
superoxide and nitric oxide (as nitrite and nitrate) by Mytilus galloprovincialis

haemocytes upon incubation with PMA or laminarin or during yeast phagocytosis.
Eur J Cell Biol 79, 513-9.

Auffret M. (2005) Bivalves as models for marine immunotoxicology. In: Investigative
Immunotoxicology (eds. by Tryphonas H, F oumier M & Blakley BR). CRC Press,
Boca Raton, FL.

189

Baker S.M. & Hombach DJ. (1997) Acute physiological effects of zebra mussel

(Dreissena polymorpha) infestation on two unionid mussels, Actinonaias
ligmentina and Amblema plicata. Can J Fish Aquat Sci 54, 512-9.

Benedict C.V. & Waite J .H. (1986) Location and analysis of byssal structural proteins of
Mytilus edulis. J Morphol 189, 171-81.

Benson A.J. & Raikow D. (2009) Dreissena polymorpha. USGS nonindigenous aquatic
species database. URL http://naS.e11usgs.gov/queries/FactShcet.asp?specieslD=5.

 

Bially A. & Maclsaac H.G. (2000) Fouling mussels (Dreissena spp.) colonize soft
sediments in Lake Erie and facilitate benthic invertebrates. Freshwater Biol 43,
85-97.

Bishop J .O., Morton J.G., Rosbash M. & Richardson M. (1974) Three abundance classes
in HeLa cell messenger RNA. Nature 250, 199—204.

Boman H.G. (2003) Antibacterial peptides: basic facts and emerging concepts. J Intern
Med 254, 197-215.

Bonmatin J .M., Bonnat J.L., Gallet X., Vovelle F., Ptak M., Reichhart J.M., Hoffmann
J .A., Keppi E., Legrain M. & Achstetter T. (1992) Two-dimensional 1H NMR
study of recombinant insect defensin A in water: resonance assignments,
secondary structure and global folding. J Biomol NMR 2, 235-56.

Bonner T.P. & Rockhill R. (1994a) Functional morphology of the zebra mussel byssus.
Dreissena polymorpha. Info Rev. 5, 4—7.

Bonner T.P. & Rockhill R.L. (1994b) Ultrastructure of the byssus of the zebra mussel
(Dreissena polymorpha, Mollusca: Bivalvia). Trans Am Microsc Soc 113, 302-15.

Boulanger N., Lowenberger C., Volf P., Ursic R., Sigutova L., Sabatier L., Svobodova
M., Beverley S.M., Spath G., Brun R., Pesson B. & Bulet P. (2004)
Characterization of a defensin from the sand ﬂy Phlebotomus duboscqi induced
by challenge with bacteria or the protozoan parasite Leishmania major. Infect
Immun 72, 7140-6.

Bowtell D. & Sambrook J. (2003) DNA microarrays. A molecular cloning manual. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

190

Brown CH. (1952) Some structural proteins of Mytilus edulis L. Q J Microsc Sci 93,
487-502.

Bulet P., Stocklin R. & Menin L. (2004) Anti-microbial peptides: from invertebrates to
vertebrates. Immunol Rev 198, 169-84.

Carballal M.J., Lopez M.C., Azevedo C. & Villalba A. (1997) Hemolymph cell types of
the mussel Mytilus galloprovincialis. Dis Aquat Org 29, 127-35.

Cayless RA. (1991) Interfacial chemistry and adhesion: the role of surface analysis in the
design of strong stable interfaces for improved adhesion and durability. Sur Inter
Anal 17, 430-8.

Cha H.J., Hwang D.S. & Lim S. (2008) Development of bioadhesives from marine
mussels. Biotechnol J.

Chandrasekharappa S., Holloway A., lyer V., Monte D., Murphy M. & Nowak NJ. (2003)
Generation of probes for spotted microarrays. In: DNA microarrays: a molecular
cloning manual (ed. by Bowtell DaS, J .), pp. 1-60. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.

Cheng TC. (1983) Bivalves. In: Invertebrate blood cells (eds. by Ratcliffe NA & Rowley
AF), pp. 234-300. Academic Press, London.

Chu F LE. (1988) Humoral defense factors in marine bivalves. In: Disease Processes in
Marine Bivalve Molluscs (ed. by Fisher WS), pp. 178-88. American Fisheries
Society Special publication, Bethesda, MD.

Chu S., DeRisi J., Eisen M., Mulholland J., Botstein D., Brown P.O. & Herskowitz I. _
(1998) The transcriptional program of sporulation in budding yeast. Science 282,
699-705.

Churchill GA. (2002) Fundamentals of experimental design for cDNA microarrays. Nat
Genet 32 Suppl, 490-5.

Ciomei C.D., Sigurdardottir T., Schmidtchen A. & Bodelsson M. (2005) Antimicrobial
and chemoattractant activity, lipopolysaccharide neutralization, cytotoxicity, and

inhibition by serum of analogs of human cathelicidin LL—37. Antimicrob Agents
Chemother 49, 2845-50.

191

Clarke M. (1999) The effect of food availability on byssogenesis by the zebra mussel
(Dreissena polymorpha Pallas). J Moll Stud 65, 327-33.

Clarke M. & McMahon R.F. (l996a) Effects of current velocity on byssal-thread
production in the zebra mussel (Dreissena polymorpha). Can J Zool 74, 63-71.

Clarke M. & McMahon R.F. (l996b) Effects of hypoxia and low-frequency agitation on
byssogenesis in the freshwater mussel Dreissena polymorpha (Pallas) Biol Bull
191, 413-9.

Clarke M. & McMahon R.F. (1996c) Effects of temperature on byssal thread production

by the freshwater mussel, Dreissena polymorpha (Pallas). Am Malacol Bull 13,
105-10.

Claudi R. & Mackie G. (1994) Practical Manual for Zebra Mussel Monitoring and
Control. Lewis Publishers, Boca Raton, FL.

Claudi R. & Mackie G.L. (1993) Practical manual for zebra mussel monitoring and
control. Lewis Publishers, Boca Raton, FL.

Cociancich S., Ghazi A., Hetru C., Hoffmann J .A. & Letellier L. (1993) Insect defensin,
an inducible antibacterial peptide, forms voltage-dependent channels in
Micrococcus luteus. J Biol Chem 268, 19239-45.

Cole A.M., Hong T., Boo L.M., Nguyen T., Zhao C., Bristol G., Zack J.A., Waring A.J.,
Yang 0.0. & Lehrer RI (2002) Retrocyclin: a primate peptide that protects cells
from infection by T— and M-tropic strains of HIV-1. Proc Natl Acad Sci U S A 99,
1813-8.

Conesa A., Gotz S., Garcia-Gomez J .M., Terol J ., Talon M. & Robles M. (2005)
Blast2GO: a universal tool for annotation, visualization and analysis in functional
genomics research. Bioinformatics 21, 3674-6.

Cope W.G., Bartsch M.R. & Marking LL. (1997) Efﬁcacy of candidate chemicals for
preventing attachment of zebra mussels (Dreissena Polymorpha). Environ T oxicol
Chem 16, 1930-4.

192

Cornet B., Bonmatin J .M., Hetru C., Hoffmann J.A., Ptak M. & Vovelle F. (1995)
Reﬁned three-dimensional solution structure of insect defensin A. Structure 3,
435-48.

Couillault C., Pujol N., Reboul J ., Sabatier L., Guichou J.F., Kohara Y. & Ewbank J].
(2004) TLR-independent control of innate immunity in Caenorhabditis elegans
by the TIR domain adaptor protein TIR-1, an ortholog of human SARM. Nat
Immunol 5, 488-94.

Coyne K.J. & Waite J .H. (2000) In search of molecular dovetails in mussel byssus: from
the threads to the stem. J Exp Biol 203, 1425-31.

Cunliffe RN. (2003) Alpha-defensins in the gastrointestinal tract. Mol Immunol 40, 463-
7.

Davids B.J. & Yoshino T.P. (1998) Integrin-like RGD-dependent binding mechanism
involved in the spreading response of circulating molluscan phagocytes. Dev
Comp Immunol 22, 39-53.

Davidson E.H. & Britten R.J. (1979) Regulation of gene expression: possible role of
repetitive sequences. Science 204, 1052-9.

DeLano W.L. (2009) The PyMOL molecular graphics system.
URL http://wwwpymolorg.

 

Deming T.J. (1999) Mussel byssus and biomolecular materials. Curr Opin Chem Biol 3,
100-5.

DeRisi J .L., Iyer V.R. & Brown PC. (1997) Exploring the metabolic and genetic control
of gene expression on a genomic scale. Science 278, 680-6.

Diatchenko L., Lau Y.F., Campbell A.P., Chenchik A., Moqadam F., Huang 3.,
Lukyanov S., Lukyanov K., Gurskaya N., Sverdlov E.D. & Siebert PD. (1996)
Suppression subtractive hybridization: a method for generating differentially
regulated or tissue-speciﬁc cDNA probes and libraries. Proc Natl Acad Sci U S A
93, 6025-30.

193

Diatchenko L., Lukyanov S., Lau Y.F. & Siebert PD. (1999) Suppression subtractive
hybridization: a versatile method for identifying differentially expressed genes.
Methods Enzymol 303, 349-80.

Dobbin K., Shih J .H. & Simon R. (2003) Questions and answers on design of dual-label
microarrays for identifying differentially expressed genes. J Natl Cancer Inst 95,
1362-9.

Dondero F ., Piacentini L., Marsano F ., Rebelo M., Vergani L., Venier P. & Viarengo A.
(2006) Gene transcription proﬁling in pollutant exposed mussels (Mytilus spp.)
using a new low-density Oligonucleotide microarray. Gene 376, 24-36.

Dormon J .M., Coish C., Cottrell C., Allen D.G. & Spelt J.K. (1997) Modes of byssal
failure in forced detachment of zebra mussels. J Environ Eng 123, 933-8.

Douglas S.E., Gallant J .W., Liebscher R.S., Dacanay A. & Tsoi SC. (2003) Identiﬁcation
and expression analysis of hepcidin—like antimicrobial peptides in bony ﬁsh. Dev
Comp Immunol 27, 589-601.

Draghici S., Kuklin A., Hoff B. & Shams S. (2001) Experimental design, analysis of
variance and slide quality assessment in gene expression arrays. Curr Opin Drug
Discov Devel 4, 332-7.

Dufva M. (2009) Introduction to microarray technology. Methods Mol Biol 529, 1-22.

Eckroat L.R., Masteller E.C., Shaffer J .C. & Steele L.M. (1993) The byssus of the zebra
mussel (Dreissena polymorpha): morphology, byssal thread formation and

detachment. In: Zebra mussels: biology, impacts, and control (eds. by Nalepa TF
& Schloesser DW), pp. 239-63. Lewis, B0ca Raton, FL, USA.

Eleveld—Trancikova D., Kudela P., Majerciak V., Regendova M., Zelnik V., Pastorek J .,
Pastorekova S. & Bizik J. (2002) Suppression subtractive hybridisation to isolate
differentially expressed genes involved in invasiveness of melanoma cell line
cultured under different conditions. Int J Oncol 20, 501-8.

Evans T.G. & Somero ON. (2008) A microarray-based transcriptomic time-course of
hyper- and hypo-osmotic stress signaling events in the euryhaline ﬁsh Gillichthys
mirabilis: osmosensors to effectors. J Exp Biol 211, 3636-49.

194

Ewing B. & Green P. (1998) Base-calling of automated sequencer traces using phred. 11.
Error probabilities. Genome Res 8, 186-94.

Ewing B., Hillier L., Wendl M.C. & Green P. (1998) Base-calling of automated
sequencer traces using phred. 1. Accuracy assessment. Genome Res 8, 175-85.

F arsad N., Gilbert T.W. & Sone ED. (2009) Adhesive structure of the freshwater zebra
mussel, Dreissena polymorpha. In: Mater. Res. Soc. Symp. Proc, pp. 1187-
KK02—03 Warrendale, PA.

Filpula D.R., Lee S.M., Link R.P., Strausberg S.L. & Strausberg R.L. (1990) Structural
and functional repetition in a marine mussel adhesive protein. Biotechnol Prog 6,
171-7.

Flammang P. (1996) Adhesion in echinoderms. In: Echinoderm studies (eds. by J angoux
M & Lawrence JM), pp. 1-60. A. A. Balkema, Rotterdam, Netherlands.

Francischetti I.M., My Pham V., Mans B.J., Andersen J.F., Mather T.N., Lane R.S. &
Ribeiro J .M. (2005) The transcriptome of the salivary glands of the female

western black-legged tick Ixodes pacificus (Acari: Ixodidae). Insect Biochem Mol‘
Biol 35, 1142-61. '

Frischer M.B., Nierzwicki-Bauer S.A., Parsons R.H., Vathanodom K. & Waitkus KR.
(2000) Interactions between zebra mussels (Dreissena polymorpha) and microbial
communities. Can. J. Fish. Aquat. Sci. 57, 591-9.

F risina A.C. & Eckroat LR. (1992) Histological and morphological attributes of the
byssus of the zebra mussel Dreissena polymorpha (Pallas). J PA Acad Sci 66, 63-
7.

Fritz B., Schubert F ., Wrobel G., Schwaenen C., Wessendorf S., Nessling M., Korz C.,
Rieker R.J., Montgomery K., Kucherlapati R., Mechtersheimer G., Eils R., Joos S.
& Lichter P. (2002) Microarray-based copy number and expression proﬁling in
dedifferentiated and pleomorphic liposarcoma. Cancer Res 62, 2993-8.

Gasteiger E., Gattiker A., Hoogland C., Ivanyi 1., Appel R.D. & Bairoch A. (2003)
ExPASy: The proteomics server for in-depth protein knowledge and analysis.
Nucleic Acids Res 31, 3784-8.

195

Giamberini L., Auffret M. & Pihan J .C. (1996) Haemocytes of the freshwater mussel,
Dreissena polymorpha (Pallas): cytology, cytochemistry, and E-ray microanalysis.
J Moll Stud 62, 367-79.

Gueguen Y., Herpin A., Aumelas A., Gamier J ., F ievet J ., Escoubas J.M., Bulet P.,
Gonzalez M., Lelong C., F avrel P. & Bachere E. (2006) Characterization of a
defensin from the oyster Crassostrea gigas. Recombinant production, folding,
solution structure, antimicrobial activities, and gene expression. J Biol Chem 281,
313-23.

Hancock R.E. & Lehrer R. (1998) Cationic peptides: a new source of antibiotics. Trends
Biotechnol 16, 82-8.

Harder J ., Bartels J ., Christophers E. & Schroder J .M. (2001) Isolation and
characterization of human beta -defenSin-3, a novel human inducible peptide
antibiotic. J Biol Chem 276, 5707-13.

Hart R.A., Miller A.C. & Davis M. (2001) Empirically derived survival rates of a native
mussel, Amblema plicata, in the Mississippi and Otter tail rivers, Minnesota. Am
Midland Naturalist 146, 254-63.

Hebert P.D.N., Muncaster B.W. & Mackie G.L. (1989) Ecological and genetic studies on
Dreissena polymorpha (Pallas): a new mollusc in the Great Lakes. Can J Fish
Aquat Sci 46, 1587-91.

Hetru C., Hoffmann D. & Bulet P. (1998) Antimicrobial peptides from insects. In:
Molecular Mechanisms of Immune Responses in Insects (eds. by Brey PT &
Hultmark D), pp. 40-66. Chapman & Hall, London.

Hine RM. (1999) The inter-relationships of bivalve haemocytes. Fish Shellﬁsh Immunol
9, 367-85.

Hodgson G., Hager J .H., Volik S., Hariono S., Wernick M., Moore D., Nowak N.,
Albertson D.G., Pinkel D., Collins C., Hanahan D. & Gray J .W. (2001) Genome
scanning with array CGH delineates regional alterations in mouse islet
carcinomas. Nat Genet 29, 459-64.

Hofﬁnann J .A. & Hetru C. (1992) Insect defensins: inducible antibacterial peptides.
Immunol Today 13, 411-5.

196

Huang X. & Madan A. (1999) CAP3: A DNA sequence assembly program. Genome Res
9, 868-77.

Hubert F., Noel T. & Roch P. (1996) A member of the arthropod defensin family from
edible Mediterranean mussels (Mytilus galloprovincialis). Eur J Biochem 240,
302-6.

Inoue K. & Odo S. (1994) The adhesive protein cDNA of Mytilus galloprovincialis
encodes decapeptide repeats but no hexapeptide motif. Biol Bull 186, 349-55.

Inoue K., Takeuchi Y., Miki D. & Odo S. (1995) Mussel adhesive plaque protein gene is
a novel member of epidermal growth factor-like gene family. J Biol Chem 270,
6698-701.

Inoue K., Takeuchi Y., Takeyama 8., Yamaha E., Yamazaki F., Odo S. & Harayama S.
(1996) Adhesive protein cDNA sequence of the mussel Mytilus coruscus and its
evolutionary implications. J Mol Evol 43, 348-56.

Jacobson A. & Peltz SW. (1996) Interrelationships of the pathways of mRNA decay and
translation in eukaryotic cells. Annu Rev Biochem 65, 693-739.

Jenny M.J., Chapman R.W., Mancia A., Chen Y.A., McKillen D.J., Trent H., Lang P.,
Escoubas J .M., Bachere E., Boulo V., Liu Z.J., Gross P.S., Cunningham C., Cupit
P.M., Tanguy A., Guo X., Moraga D., Boutet I., Huvet A., De Guise S., Almeida
J .S. & Warr G.W. (2007) A cDNA microarray for Crassostrea virginica and C.
gigas. Mar Biotechnol (NY) 9, 577-91.

Ji W., Wright M.B., Cai L., F lament A. & Lindpaintner K. (2002) Efﬁcacy of SSH PCR
in isolating differentially expressed genes. BMC Genomics 3, 12.

Johnson L.E. & Padilla D.K. (1996) Geographic Spread of exotic species: ecological
lessons and opportunities from the invasion of the zebra mussel (Dreissena
polymorpha). Biol Conserv 78, 23-33.

Kavouras J .H. & Maki J .S. (2003a) Effects of bioﬁlms on zebra mussel postveliger
attachment to artiﬁcial surfaces. Invert Biol 122, 138-51.

Kavouras J .H. & Maki J .S. (2003b) The effects of natural bioﬁlms on the reattachment of
young adult zebra mussels to artiﬁcial substrata. Biofouling 19, 247-56.

197

 

Kavouras J .H. & Maki J.S. (2004) Inhibition of the reattachment of young adult zebra

mussels by single-species bioﬁlms and associated exopolymers. J Appl Microbiol
97, 1236-46.

Kerr MK. (2003) Design considerations for efﬁcient and effective microarray studies.
Biometrics 59, 822-8.

Kerr M.K. & Churchill GA. (2001) Experimental design for gene expression microarrays.
Biostatistics 2, 183-201.

 

Kulesh D.A., Clive D.R., Zarlenga D.S. & Greene J .J . (1987) Identiﬁcation of interferon-
modulated proliferation-related cDNA sequences. Proc Natl Acad Sci U S A 84,
8453-7.

ii

Lambert C., Leonard N., De Bolle X. & Depiereux E. (2002) ESyPred3D: Prediction of
proteins 3D structures. Bioinformatics 18, 1250-6.

Langley R.R., Ramirez K.M., Tsan R.Z., Van Arsdall M., Nilsson M.B. & Fidler II.
(2003) Tissue-speciﬁc microvascular endothelial cell lines from H—2K(b)-tSA58
mice for studies of angiogenesis and metastasis. Cancer Res 63, 2971-6.

Lee B.P., Chao C., Nunalee F.N., Motan E., Shull K.R. & Messersmith PB. (2006) Rapid
gel formation and adhesion in photocurable and biodegradable block copolymers
with high DOPA content. Macromolecules 39, 1740-8.

 

Lehrer R.I., Rosenman M., Harwig S.S., Jackson R. & Eisenhauer P. (1991)
Ultrasensitive assays for endogenous antimicrobial polypeptides. J Immunol
Methods 137, 167-73.

Leung Y.F., Ma P., Link B.A. & Dowling J .E. (2008) Factorial microarray analysis of
zebraﬁsh retinal development. Proc Natl Acad Sci U S A 105, 12909-14.

Li Y.J., Tian F ., Chen Z.C., Guan Y.J., He C.M., Yang X.M. & Xie DH. (2000) Isolation
and Identiﬁcation of cDNA Sequences Differentially Expressed in Laryngeal
Carcinoma. Sheng Wu Hua Xue Yu Sheng Wu Wu Li Xue Bao (Shanghai) 32, 153-
7.

198

Lin Q., Gourdon D., Sun C., Holten-Andersen N., Anderson T.H., Waite J .H. &
Israelachvili J .N. (2007) Adhesion mechanisms of the mussel foot proteins mfp-l
and mfp-3. Proc Natl Acad Sci U S A 104, 3782-6.

Livak K.J. & Schmittgen TD. (2001) Analysis of relative gene expression data using
real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25,
402-8. ‘

Maki J .S., Patel G. & Mitchell R. (1998) Experimental pathogenicity of aeromonas spp.
for the zebra mussel, Dreissena polymorpha. Curr Microbiol 36, 19-23.

McDowell L.M., Burzio L.A., Waite J .H. & Schaefer J. (1999) Rotational echo double
resonance detection of cross-links formed in mussel byssus under high-ﬂow stress.
J Biol Chem 274, 20293-5.

McMahon RF. (1996) The physiological ecology of the zebra mussel, Dreissena
polymorpha, in North America and Europe. Am Zoologist 36, 339-63.

McVeigh P., Alexander-Bowman S., Veal E., Mousley A., Marks N.J. & Maule A.G.
(2008) Neuropeptide-like protein diversity in phylum Nematoda. Int J Parasitol.

Miki R., Kadota K., Bono H., Mizuno Y., Tomaru Y., Caminci P., Itoh M., Shibata K.,
Kawai J ., Konno H., Watanabe S., Sato K., Tokusumi Y., Kikuchi N., Ishii Y.,
Hamaguchi Y., Nishizuka 1., Goto H., Nitanda H., Satomi S., Yoshiki A.,
Kusakabe M., DeRisi J.L., Eisen M.B., Iyer V.R., Brown P.O., Muramatsu M.,
Shimada H., Okazaki Y. & Hayashizaki Y. (2001) Delineating developmental and
metabolic pathways in vivo by expression proﬁling using the RIKEN set of
18,816 full-length enriched mouse cDNA arrays. Proc Natl Acad Sci U S A 98,
2199-204.

Mitta G., Vandenbulcke F ., Hubert F ., Salzet M. & Roch P. (2000a) Involvement of
mytilins in mussel antimicrobial defense. J Biol Chem 275, 12954-62.

Mitta G., Vandenbulcke F., Noel T., Romestand B., Beauvillain J.C., Salzet M. & Roch P.
(2000b) Differential distribution and defence involvement of antimicrobial
peptides in mussel. J Cell Sci 113 (Pt 15), 2759-69.

Molloy DP. (1998) The potential of using biological control technologies in the
management of Dreissena. In: the Eighth International Zebra Mussel and Other
Nuisance Species Conference. California Sea Grant, Sacramento, California.

199

 

Molloy DP. (2002) Biological control of zebra mussels. In: Third California Conference
on Biological Control, pp. 86-94, University of California, Davis, CA.

Montgomery DC. (2005) Design and Analysis of Experiments. Wiley, .Hoboken, NJ.

Morton B. (1969) Studies on the biology of Dreissena polymorpha pall: I. General
anatomy and morphology. J. Moi/us. Stud. 38, 301-21.

Morton B. (1993) The anatomy of Dreissena polymorpha and the evolution and success
of the heteromyarian form in the Dreissenoidea. In: Zebra Mussels: Biology,
Impact, and Control (eds. by Nalepa TF & Schloesser DW), pp. 185-215. Lewis
Press, Boca Raton, FL.

Munk C., Wei G., Yang 0.0., Waring A.J., Wang W., Hong T., Lehrer R.I., Landau N.R.
& Cole A.M. (2003) The theta-defensin, retrocyclin, inhibits HIV-1 entry. AIDS
Res Hum Retroviruses 19, 875-81.

Nalepa T.P. & Schloesser D.W. (I993) Zebra mussels: biology, impact and control.
Lewis Publisher, Boca Raton, Florida.

Narasimhan S., Deponte K., Marcantonio N., Liang X., Royce T.E., Nelson K.F., Booth
C.J., Koski B., Anderson J.F., Kantor F. & F ikrig E. (2007) Immunity against
Ixodes scapularis salivary proteins expressed within 24 hours of attachment
thwarts tick feeding and impairs Borrelia transmission. PLoS ONE 2, e451.

Nathoo A.N., Moeller R.A., Westlund B.A. & Hart A.C. (2001) Identiﬁcation of
neuropeptide-like protein gene families in Caenorhabditiselegans and other
Species. Proc Natl Acad Sci U S A 98, 14000-5.

New-York-Sea—Grant (1994a) "Agency Activities." Dreissena! , pp. 5(3), 1—2. New York
Sea Grant, Zebra Mussel Clearinghouse, 250 Hartwell Hall, SUNY College at
Brockport, Brockport, NY 14420-2928.

New-York-Sea-Grant (1994b) Dreissena polymorpha information review. pp. 5(1), 14-5.
New York Sea Grant, Zebra Mussel Clearinghouse, 250 Hartwell Hall, SUNY
College at Brockport, Brockport, NY 14420-2928.

Nguyen D.V., Arpat A.B., Wang N. & Carroll R.J. (2002) DNA microarray experiments:
biological and technological aspects. Biometrics 58, 701-17.

200

Oliver L.M. & Fisher W.S. (1995) Comparative form and function of oyster Crassostrea
virginica hemocytes from Chesapeake bay (Virginia) and Apalachicola Bay
(Florida). Dis Aquat Org 22, 217-25.

Ouellette A.J., Satchel] D.P., Hsieh M.M., Hagen S.J. & Selsted ME. (2000)
Characterization of luminal paneth cell alpha-defensins in mouse small intestine.

Attenuated antimicrobial activities of peptides with truncated amino termini. J
Biol Chem 275, 33969-73.

Papov V.V., Diamond T.V., Biemann K. & Waite J .H. (1995) Hydroxyarginine-
containing polyphenolic proteins in the adhesive plaques of the marine mussel
Mytilus edulis. J Biol Chem 270, 20183-92.

Pathy D.A. & Mackie G.L. (1993) Comparative shell morphology of Dreissena
polymorpha, Mytilopsis leucophyta, and the “quagga” mussel (Bivalvia:
Dreissenidae) in North America. Can J Zool 71, 1012-23.

Pawitan Y., Michiels S., Koscielny S., Gusnanto A. & Ploner A. (2005) False discovery
rate, sensitivity and sample Size for microarray studies. Bioinformatics 21, 3017-
24.

Pihlajaniemi T. & Tamminen M. (1990) The alpha 1 chain of type XIII collagen consists
of three collagenous and four noncollagenous domains, and its primary transcript
undergoes complex alternative splicing. J Biol Chem 265, 16922-8.

Pipe R.K. (1990) Hydrolytic enzymes associated with the granular haemocytes of the
marine mussel Mytilus edulis. Histochem J 22, 595-603.

Pipe R.K. (1992) Generation of reactive oxygen metabolites by the haemocytes of the
mussel Mytilus edulis. Dev Comp Immunol 16, 111-22.

Pipe R.K., Farley S.R. & Coles J .A. (1997) The separation and characterisation of
haemocytes from the mussel Mytilus edulis. Cell Tissue Res 289, 537-45.

Price HA. (1983) Structure and formation of the byssus complex in Mytilus (Mollusca,
Bivalvia). J Moll Stud. 49, 9-17.

Pritchard C.C. & Nelson PS. (2008) Gene expression proﬁling in the developing prostate.
Differentiation 76, 624-40.

201

 

Radhakn'shnan Y., Fares M.A., French F.S. & Hall SH. (2007) Comparative genomic
analysis of a mammalian beta-defensin gene cluster. Physiol Genomics 30, 213-22.

Raikow DP. (2004) Food web interactions between larval bluegill (Lepomis macrochirus)

and exotic zebra mussels (Dreissena polymorpha). Can. J. Fish. Aquat. Sci. 61,
497-504.

Rajagopal S., Van Der Velde G. & Jenner HA. (2002) Does status of attachment
inﬂuence survival time of zebra mussel, Dreissena polymorpha, exposed to
chlorination? Environ Toxicol Chem 21, 342-6.

Rajagopal S., van der Velde G., van der Gaag M. & Jenner HA. (2005) Byssal
detachment underestimates tolerance of mussels to toxic compounds. Mar Pollut
Bull 50, 20-9.

Renwrantz L. & Stahmer A. (1983) Opsonizing properties of an isolated hemolymph
agglutinin and demonstration of lectin-like recognition molecules at the surface of
hemocytes from Mytilus edulis. J Comp Physiol 149, 535-46.

Richards O.W. (1929) The conduction of the nervous impulse through the pedal ganglion
of Mytilus Biol Bull 56, 32-40.

Roberts L. (1990) Zebra mussel invasion threatens U.S. waters: damage estimates soar
into the billions for the zebra mussel, just one of many invaders entering U.S.
waters via ballast water. Science 249, 1370-2.

Roch P. (1999) Defense mechanisms and disease prevention in farmed marine
invertebrates. Aquaculture 172, 125-45.

Ross J. (1996) Control of messenger RNA stability in higher eukaryotes. Trends Genet
12, 171-5.

Rzepecki L.M. & Waite J .H. (1993a) The byssus of the zebra mussel, Dreissena
polymorpha. I: Morphology and in situ protein processing during maturation. Mol
Mar Biol Biotechnol 2, 255-66.

Rzepecki L.M. & Waite J .H. (1993b) The byssus of the zebra mussel, Dreissena
polymorpha. 11: Structure and polymorphism of byssal polyphenolic protein
families. Mol Mar Biol Biotechnol 2, 267-79.

202

 

Schloesser D.W., Nalepa T.P. & Mackie G.L. (1996) Zebra mussel infestation of unionid
bivalves (Unionidae) in North America. Am Zoologist 36, 300-10.

Seo J.K., Crawford J.M., Stone K.L. & Noga E]. (2005) Puriﬁcation of a novel
arthropod defensin from the American oyster, Crassostrea virginica. Biochem
Biophys Res Commun 338, 1998-2004.

Shoemaker D.D., Schadt E.E., Armour C.D., He Y.D., Garrett-Engele P., McDonagh
P.D., Loerch P.M., Leonardson A., Lum P.Y., Cavet G., Wu L.F., Altschuler S.J.,
Edwards S., King J ., Tsang J .S., Schimmack G., Schelter J .M., Koch J ., Ziman M.,
Marton M.J., Li B., Cundiff P., Ward T., Castle J ., Krolewski M., Meyer M.R.,
Mao M., Burchard J., Kidd M.J., Dai H., Phillips J.W., Linsley P.S., Stoughton R.,
Scherer S. & Boguski MS. (2001) Experimental annotation of the human genome
using microarray technology. Nature 409, 922-7.

Skubinna J .P., Coon T.G. & Batterson TR. (1995) Increased abundance and depth of
submersed macrophytes in response to decreased turbidity in Saginaw Bay,
Michigan. J Great Lake Res 21, 476-88.

Smith A.M. & Callow J .A. (2006) Biological adhesives. Springer, Berlin, Germany.

Smyth GK. (2005) Limma: linear models for microarray data. In: Bioinformatics and
Computational Biology Solutions using R and Bioconductor (eds. by Gentleman
R, Carey V, Dudoit S, Irizarry R & Huber W), pp. 397 - 420. Springer, New York.

Snyder F.L., Hilgendorf M.B. & Garton D.W. (1997) Zebra Mussels in North America:
The invasion and its implications. Ohio Sea Grant, Ohio State University,
Columbus, OH.

Soares M.B., Bonaldo M.F., Jelene P., Su L., Lawton L. & Efstratiadis A. (1994)
Construction and characterization of a normalized cDNA library. Proc Natl Acad
Sci USA 91, 9228-32.

Song L., Xu W., Li C., Li H., Wu L., Xiang J. & Guo X. (2006) Development of
expressed sequence tags from the bay scallop, Argopecten irradians irradians.
Mar Biotechnol (N19 8, 161-9.

Stears R.L., Martinsky T. & Schena M. (2003) Trends in microarray analysis. Nat Med 9,
140-5.

203

 

Stothard P. (2000) The sequence manipulation suite: JavaScript programs for analyzing
and formatting protein and DNA sequences. Biotechniques 28, 1102, 4.

Sturn A., Quackenbush J. & Trajanoski Z. (2002) Genesis: cluster analysis of microarray
data. Bioinformatics 18, 207-8.

Suh Y.J., Cho S.A., Shim J.H., Yook Y.J., Yoo K.H., Kim J.H., Park E.Y., Noh J.Y., Lee
SH, Yang M.H., Jeong H.S. & Park J .H. (2008) Gene discovery analysis from

mouse embryonic stem cells based on time course microarray data. Mol Cells 26,
338-43.

Sun C. & Waite J .H. (2005) Mapping chemical gradients within and along a ﬁbrous
structural tissue, mussel byssal threads. J Biol Chem 280, 39332-6.

Tafalla C., Gomez-Leon J ., Novoa B. & F igueras A. (2003) Nitric oxide production by

carpet shell clam (Ruditapes decussatus) hemocytes. Dev Comp Immunol 27 , 197-
205.

Tamura K., Dudley J ., Nei M. & Kumar S. (2007) MEGA4: Molecular Evolutionary
Genetics Analysis (MEGA) software version 4.0. Mol Biol E vol 24, 1596-9.

Tempelman R.J. (2005) Assessing statistical precision, power, and robustness of
alternative experimental designs for two color microarray platforms based on
mixed effects models. Vet Immunol Immunopathol 105, 175-86.

Thompson J .D., Higgins D.G. & Gibson T.J. (1994) CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence

weighting, position-Speciﬁc gap penalties and weight matrix choice. Nucleic
Acids Res 22, 4673-80.

Tiscar P.G. & Mosca F. (2004) Defense mechanisms in farmed marine molluscs. Vet Res
Commun 28 Suppl 1, 57-62.

Toomey M.B., Mccabe D. & Marsden J .E. (2002) Factors affecting the movement of
adult zebra mussels (Dreissena polymorpha). J. N. Am. Benthol. Soc. 21, 468-75.

Villalva C., Trempat P., Zenou R.C., Delsol G. & Brousset P. (2001) Gene expression
proﬁling by suppression subtractive hybridization (SSH): an example for its
application to the study of lymphomas. Bull Cancer 88, 315—9.

204

Waite J .H. (1983a) Adhesion in byssally attached bivalves. Biol Rev 58, 209-31.

Waite J .H. (1983b) Evidence for a repeating 3,4-dihydroxyphenylalanine- and
hydroxyproline-containing decapeptide in the adhesive protein of the mussel,
Mytilus edulis L. J Biol Chem 258, 2911-5.

Waite J .H. (1987) Nature's underwater adhesive specialist. Int J Adhes Adhes 7, 9—14.

Waite J .H. (1992) The formation of mussel byssus: anatomy of a natural manufacturing
process. Results Probl Cell Differ 19, 27-54.

Waite J .H. (2002) Adhesion a la Moule. Integr Comp Biol 42, 1 172-80.

 

Waite J.H., Andersen N.H., Jewhurst S. & Sun C. (2005) Mussel adhesion: ﬁnding the
tricks worth mimicking. J Adhes 81, 297-317.

Waite J .H. & Qin X. (2001) Polyphosphoprotein from the adhesive pads of Mytilus edulis.
Biochemistry 40, 2887-93.

Waite J.H., Qin X.X. & Coyne K.J. (1998) The peculiar collagens of mussel byssus.
Matrix Biol 17, 93-106.

Walker J .M. (2002) SDS polyacrylamide gel electrophoresis of proteins. In: The protein
protocols handbook (ed. by Walker JM), pp. 61-7. Humana Press, Totowa, NJ.

Wang L., Song L., Zhao J., Qiu L., Zhang H., Xu W., Li H., Li C., Wu L. & Guo X.
(2009) Expressed sequence tags from the zhikong scallop (Chlamysfarrerz):

discovery and annotation of host-defense genes. Fish Shellfish Immunol 26, 744-
50.

Wang W., Cole A.M., Hong T., Waring AJ. & Lehrer R.I. (2003) Retrocyclin, an
antiretroviral theta-defensin, is a lectin. J Immunol 170, 4708-16.

Wang X. & Feuerstein G.Z. (2000) Suppression subtractive hybridisation: application in
the discovery of novel pharmacological targets. Pharmacogenomics 1, 101-8.

205

 

Warner S.C. & Waite J .H. (1999) Expression of multiple forms of an adhesive plaque
protein in an individual mussel, Mytilus edulis. . Mar Biol 134, 729-34.

Wettenhall J .M. & Smyth GK. (2004) limmaGUI: a graphical user interface for linear
modeling of microarray data. Bioinformatics 20, 3705-6.

Wiens M., Koziol C., Hassanein H.M., Muller I.M. & Muller W.E. (1999) A homolog of
the putative tumor suppressor QM in the sponge Suberites domuncula:

downregulation during the transition from immortal to mortal (apoptotic) cells.
Tissue Cell 31, 163-9. 1

Wit E., Nobile A. & Khanin R. (2005) Near-optimal designs for dual channel microarray
studies. J Royal Stat Soc, C 54, 817-30.

Wu C.-F. & Hamada M. (2000) Experiments: Planning, Analysis, and Parameter Design
Optimization. Wiley, New York.

Xu W. & Faisal M. (2007) Matrilin-like molecules produced by circulating hemocytes of
the zebra mussel (Dreissena polymorpha) upon stimulation. Dev Comp Immunol
31, 1205-10.

Xu W. & Faisal M. (2008) Putative identiﬁcation of expressed genes associated with
attachment of the zebra mussel (Dreissena polymorpha). Biofouling 24, 157-61.

Xu W. & Faisal M. (2009a) Development of a cDNA microarray of zebra mussel
(Dreissena polymorpha) foot and its use in understanding the early stage of
underwater adhesion. Gene 436, 71-80.

Xu W. & Faisal M. (2009b) Identiﬁcation of the molecules involved in zebra mussel

(Dreissena polymorpha) hemocytes host defense. Comp Biochem Physiol B
Biochem Mol Biol 154, 143-9.

Yang D., Chertov 0., Bykovskaia S.N., Chen Q., Buffo M.J., Shogan J ., Anderson M.,
Schroder J .M., Wang J .M., Howard O.M. & Oppenheim J .J . (1999) Beta-

defensins: linking innate and adaptive immunity through dendritic and T cell
CCR6. Science 286, 525-8.

206

Yang Y.S., Mitta G., Chavanieu A., Calas B., Sanchez J .F., Roch P. & Aumelas A. (2000)
Solution structure and activity of the synthetic four-disulﬁde bond Mediterranean
mussel defensin (MGD-l). Biochemistry 39, 14436-47.

Yano M ., Nagai K., Morimoto K. & Miyamoto H. (2006) Shematrin: a family of glycine-
rich structural proteins in the Shell of the pearl oyster Pinctadafucata. Comp
Biochem Physiol B Biochem Mol Biol 144, 254-62.

Yu M. & Deming T.J. (1998) Synthetic Polypeptide Mimics of Marine Adhesives.
Macromolecules 31, 4739-45.

Zdobnov E.M. & preiler R. (2001) InterProScan--an integration platform for the
signature-recognition methods in InterPro. Bioinformatics 17, 847-8.

Zhao H., Robertson N.B., Jewhurst S.A. & Waite J .H. (2006) Probing the adhesive
footprints of Mytilus califomianus byssus. J Biol Chem 281, 11090-6.

Zhao H. & Waite J .H. (2006) Linking adhesive and structural proteins in the attachment
plaque of Mytilus califomianus. J Biol Chem 281, 26150-8.

Zhao J., Song L., Li C., Ni D., Wu L., Zhu L., Wang H. & Xu W. (2007) Molecular
cloning, expression of a big defensin gene from bay scallop Argopecten irradians
and the antimicrobial activity of its recombinant protein. Mol Immunol 44, 360-8.

Zhu G., Spellman P.T., Volpe T., Brown P.O., Botstein D., Davis T.N. & Futcher B.
(2000) Two yeast forkhead genes regulate the cell cycle and pseudohyphal growth.
Nature 406, 90-4.

Zou J., Rodriguez-Zas S., Aldea M., Li M., Zhu J ., Gonzalez D.O., Vodkin L.O.,
DeLucia E. & Clough SJ. (2005) Expression proﬁling soybean response to
Pseudomonas syringae reveals new defense-related genes and rapid HR-Speciﬁc
downregulation of photosynthesis. Mol Plant Microbe Interact 18, 1161-74.

207

 

 

I
”1
Ii

A m

m Ii

mil