-
Title
-
Eng
Investigating the Effectiveness of Scanning Electron Microscopy to Identify Densovirus in Pisaster Ochraceus Infected with Sea Star Wasting Disease
-
Date
-
2018
-
Creator
-
Eng
Azar Averi
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
INVESTIGATING THE EFFECTIVENESS OF SCANNING ELECTRON
MICROSCOPY TO IDENTIFY DENSOVIRUS IN PISASTER OCHRACEUS
INFECTED WITH SEA STAR WASTING DISEASE
by
Averi A. Azar
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2018
2018 by Averi A. Azar. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Averi A. Azar
has been approved for
The Evergreen State College
by
______________________________________________
Erin Martin, Ph.D.
Member of the Faculty
________________________
Date
ABSTRACT
Investigating the effectiveness of Scanning Electron Microscopy to
identify densovirus in Pisaster ochraceus infected with sea star wasting
disease
Averi A. Azar
Sea stars on the Pacific Northwest coast of North America experienced massive sea star
wasting disease induced die-offs from 2013 to 2015. Some of the sea stars most affected
by wasting disease were the keystone predator, the purple-orange ochre sea star, Pisaster
ochraceus. In 2014 evidence demonstrated sea star wasting is strongly correlated with
presence of the sea star-associated densovirus (SSaDV). Traditional methods for viral
identification have involved metagenomics, a highly expensive and technical method. This
research aimed to develop a financially accessible method to identify densovirus in P.
ochraceus. This thesis implemented the use of Scanning Electron Microscopy (SEM), to
test its ability to resolve densovirus in P. ochraceus and potentially develop a SEM method
that could help others with limited expertise and funding contribute to sea star wasting
disease literature. This project found viral isolation from P. ochraceus tissue to be
challenging and difficult to successfully execute. Thus, this project performed whole P.
ochraceus tissue analysis using traditional SEM sample preparation methods. Through
scanning, P. ochraceus tissue was resolved under multiple magnifications, covering a
range of X to Y. Bacteria were not present in any of the samples. Even at the highest
magnification of 300,000X, the SEM was not strong enough to produce a quality image at
high resolution and was therefore unsuccessful at identifying densovirus at 20-25 nm in
necrotic P. ochraceus tissue. The SEM implemented for this research (Model JEOL JSM6480LV), is not sufficient to identify densovirus in P. ochraceus
Table of Contents
List of Figures __________________________________________________________ vi
Introduction ____________________________________________________________ 1
Literature Review _______________________________________________________ 5
Pisaster ochraceus ____________________________________________________ 5
Sea Star Wasting Disease _______________________________________________ 8
The Densovirus ______________________________________________________ 14
Testing for the Densovirus ___________________________________________ 19
SEM and Densovirus _____________________________________________ 20
Scanning Electron Microscopy __________________________________________ 23
Components of the SEM _____________________________________________ 23
SEM Operation ____________________________________________________ 25
Standard SEM Methods _____________________________________________ 26
Methods______________________________________________________________ 29
Methodological Introduction ___________________________________________ 29
Tissue Sample Collection ______________________________________________ 30
Environmental Data Collection__________________________________________ 30
Method 1 ___________________________________________________________ 32
Filtration ___________________________________________________________ 32
Filtration Testing ___________________________________________________ 32
Result & Discussion ________________________________________________ 33
Method 2 ___________________________________________________________ 35
Vacuum Filtration __________________________________________________ 35
Result and Discussion _______________________________________________ 36
Method 3 ___________________________________________________________ 38
Swinlock Filter Holders & Polycarbonate Membranes _____________________ 38
Tissue Sample Preparation for SEM ____________________________________ 39
Result and Discussion _______________________________________________ 40
Whole Tissue Sample Preparation _______________________________________ 42
Experimental Drying ________________________________________________ 42
Sputter Coating ____________________________________________________ 43
SEM Calibrations __________________________________________________ 44
Sample Analysis _________________________________________________ 44
iv
Densovirus Characteristics _____________________________________________ 45
Results _______________________________________________________________ 47
Introduction of Results ________________________________________________ 47
Sample 1 ___________________________________________________________ 47
Sample 2 ___________________________________________________________ 59
Sample 3 ___________________________________________________________ 65
Sample 3.5__________________________________________________________ 69
Sample 4 ___________________________________________________________ 76
Sample 4.5__________________________________________________________ 80
Sample 5 – Healthy Tissue _____________________________________________ 84
Conclusive Results ___________________________________________________ 88
Discussion ____________________________________________________________ 91
Conclusion ___________________________________________________________ 95
Bibliography __________________________________________________________ 97
Appendices __________________________________________________________ 105
Appendix A ________________________________________________________ 105
Sample 1 ________________________________________________________ 105
Sample 3 __________________________________________________________ 108
Sample 3.5_________________________________________________________ 110
Sample 4 __________________________________________________________ 112
Sample 4.5_________________________________________________________ 116
Sample 5—Healthy Tissue ____________________________________________ 118
v
List of Figures
Figure 1. ______________________________________________________________ 6
Figure 2. _____________________________________________________________ 10
Figure 3. _____________________________________________________________ 11
Figure 4. _____________________________________________________________ 16
Figure 5. _____________________________________________________________ 21
Figure 6. _____________________________________________________________ 24
Figure 7. _____________________________________________________________ 26
Figure 8. _____________________________________________________________ 48
Figure 9. _____________________________________________________________ 49
Figure 10. ____________________________________________________________ 50
Figure 11. ____________________________________________________________ 51
Figure 12. ____________________________________________________________ 52
Figure 13. ____________________________________________________________ 53
Figure 14. ____________________________________________________________ 54
Figure 15. ____________________________________________________________ 55
Figure 16. ____________________________________________________________ 56
Figure 17. ____________________________________________________________ 57
Figure 18. ____________________________________________________________ 58
Figure 19. ____________________________________________________________ 59
Figure 20. ____________________________________________________________ 60
Figure 21. ____________________________________________________________ 61
Figure 22. ____________________________________________________________ 62
Figure 23. ____________________________________________________________ 63
Figure 24. ____________________________________________________________ 64
Figure 25. ____________________________________________________________ 65
Figure 26. ____________________________________________________________ 66
Figure 27. ____________________________________________________________ 67
Figure 28. ____________________________________________________________ 68
Figure 29. ____________________________________________________________ 69
Figure 30. ____________________________________________________________ 70
Figure 31. ____________________________________________________________ 71
Figure 32. ____________________________________________________________ 72
Figure 33. ____________________________________________________________ 73
Figure 34. ____________________________________________________________ 74
Figure 35. ____________________________________________________________ 75
Figure 36. ____________________________________________________________ 76
Figure 37. ____________________________________________________________ 77
Figure 38. ____________________________________________________________ 78
Figure 39. ____________________________________________________________ 79
Figure 40. ____________________________________________________________ 80
Figure 41. ____________________________________________________________ 81
Figure 42. ____________________________________________________________ 82
Figure 43. ____________________________________________________________ 83
Figure 44. ____________________________________________________________ 84
vi
Figure 45.
Figure 46.
Figure 47.
Figure 48.
Figure 49.
Figure 50.
Figure 51.
Figure 52.
Figure 53.
Figure 54.
Figure 55.
Figure 56.
Figure 57.
Figure 58.
Figure 59.
Figure 60.
Figure 61.
Figure 62.
Figure 63.
Figure 64.
Figure 65.
vii
____________________________________________________________ 85
____________________________________________________________ 86
____________________________________________________________ 87
____________________________________________________________ 88
___________________________________________________________ 105
___________________________________________________________ 106
___________________________________________________________ 107
___________________________________________________________ 107
___________________________________________________________ 108
___________________________________________________________ 109
___________________________________________________________ 110
___________________________________________________________ 111
___________________________________________________________ 112
___________________________________________________________ 113
___________________________________________________________ 113
___________________________________________________________ 114
___________________________________________________________ 115
___________________________________________________________ 116
___________________________________________________________ 117
___________________________________________________________ 118
___________________________________________________________ 119
Acknowledgements
This project was made possible through the generosity and interest of the Clean
Energy Committee and the Master of Environmental Studies Association. In
combination, these groups were responsible for fully funding my research and I am so
grateful for their contribution. I would like to thank my thesis reader, Erin Martin, for
supporting me throughout this journey. I began my work on this project in Spring of 2017
and completed it during the Spring of 2018. Erin was my advisor and reader throughout
this process and I’m so grateful for her patience, encouragement and mentorship despite
the series of unfortunate events which seemed to plague my project.
I would also like to thank Trisha Towanda, my SIT, for teaching me to use the
SEM and supporting my learning through Advanced Microscopy training. Trisha’s
unyielding confidence in me and my project was essential for my success and I’m so
thankful for her tutelage and support. Likewise, Erik Thuesen and Gerardo Chin-Leo
often provided critical guidance and advice throughout my SEM analysis which helped
informed decisions and results in this thesis.
My success during the thesis process was highly influenced by the professional
support of my boss Andrea Martin, Assistant Director of the Master of Environmental
Studies program. Andrea has been the best supervisor, mentor and support system during
the last year and a half. Her excellent organization and planning abilities have made
working in the MES office a true joy and frequently the easiest part of each day. I
couldn’t have done this without her. Additionally, my coworkers in the MPA program,
Jan Hays (whom I’ve had the distinct privilege of sharing an office with) and Puanani
Nihoa, Assistant Director to the MPA Tribal Governance program, have been a constant
source of light and happiness. I was always buoyed by their humor and molded by their
experience and mentorship, an honor I am so thankful to have experienced during this
time in my life.
I would like to thank my family and friends for supporting me throughout my
time in graduate school. Thank you for always listening to me when I was stressed or
upset about my thesis project and for offering words of wisdom when my project would
encounter another challenge. Thank you for also putting up with my physical absence and
letting me focus so I can accomplish this monster of a task: write a thesis.
Most of all though, I would like to thank my cohort, the best cohort ever! It has
been a true joy to work with and get to know all of you. Your experiences and guidance
have informed my work and I truly value the influence you all have had on me both
personally and professionally. I would like to specifically thank those that have
continuously offered me emotional and academic support throughout my thesis work.
Nicole George & Leslie Carman—my primary thesis support system—for being my
writing buddies, helping me understand the SEM and being totally awesome. My lab
group, Amanda Mintz, Morgan Maupin, Stephanie Blair, Max Calloway, Alex CaseCohen and John Messina for being so helpful and filling the lab with laughter.
viii
ix
Introduction
The Pacific Northwest coast of North America from Juneau, Alaska to Baja,
California was plagued with massive sea star die-offs from 2013 to 2015. This mass
mortality event was caused by sea star wasting disease (Hewson et al. 2014, Eisenlord et
al. 2016, Fuess et al. 2015). A combination of disease outbreak in 2013, followed by
warming ocean temperatures in 2014, exacerbated the disease and further enhanced sea
star mortality to wasting (Miner et al. 2018). The disease looks as its name suggests, as a
form of sea star tissue disintegration and melting in a mass of white lesions; an
appearance which has left many on the Northeast Pacific coast alarmed and baffled
(Fuess et al. 2015). The mass mortality event is often considered a two-year incident,
ending in 2015, due to the collective mass dying which occurred in this period. Post2015, sea star wasting still plagues the Northeast Pacific coast (Eisenlord et al. 2016).
However, populations of sea stars were thrown into such a sharp decrease that presence
of wasting disease is not felt as critically with fewer sea stars to be affected.
The strongest correlations to the source of wasting disease have been temperature
increases in the Pacific (Miner et al. 2018) and the presence of the sea star-associated
densovirus (SSaDV) in diseased individuals (Hewson et al. 2014). Much of our
knowledge of sea star wasting causation has only recently been gained in the last couple
years with the largest breakthrough occurring in 2014 by Hewson et al. The challenge
with studying wasting disease is the deductive reasoning involved in learning about the
phenomena. The disease is not only present in infected individuals but appears to persist
in the environment and seems to be transmissible from sick to healthy individuals. These
observations lead Hewson et al. (2014) to examine the likelihood that sea star wasting
1
could be caused by a pathogen much smaller than bacteria, parasites or fungi: a virus.
This observation proved to be likely with the implementation of viral metagenomics,
Hewson et al. (2014) successfully discovered the presence of a densovirus—a member of
the Parvoviridae family—in sick sea star tissue. This virus was therefore named, sea starassociated densovirus (SSaDV) for its apparent relationship to the presence of wasting
disease and its genetic difference to other known densoviruses (Hewson et al. 2014).
Continued research into the source of sea star wasting disease is imperative as the
impact of the disease is still felt on the Pacific coast. In the Pacific Northwest of North
America, a species of great importance to the coastal ecosystems is the purple orange
ochre sea star, Pisaster ochraceus (Paine 1969). P. ochraceus are one of the organisms
responsible for maintaining species biodiversity in rocky intertidal zones. Their role as a
keystone species is felt most intensely in their voracious appetites for mussels, especially
the highly prolific California mussels (Mytilus californianus) (Paine 1995, Lawrence
2013). Without P. ochraceus, many intertidal ecosystems have experienced biodiversity
loss from mussels that frequently outcompete other organisms for prime intertidal habitat
(Miner et al. 2018). Although P. ochraceus populations are attempting to rebound from
the 2013-2015 mass die-off, sea star wasting is still present and remains a threat to the
future of P. ochraceus and subsequently, the health of rocky intertidal habitats (Moritsch
& Raimondi 2018).
Questions about the source of wasting disease continue to swirl amongst the
confusion and despair about the state of P. ochraceus. Sea star wasting disease is not
definitively caused by a single source or pathogen (Bucci et al. 2017). Additionally,
viruses in marine invertebrates as a whole are highly understudied so our knowledge of
2
viral-host interactions is murky and undefined (Menge et al. 2016, Suttle 2005). The
relationship between densovirus and sea star wasting disease is the strongest pathogendisease relationship to date. Therefore, a greater understanding of this relationship is
necessary to further our knowledge of this viral-host interaction in P. ochraceus. While
viral metagenomics is the most accurate form of viral identification in sea stars, it is also
a highly technical and expensive method, requiring a great deal of genetic expertise and
money to execute successfully. As a result, this project aimed to develop a less technical,
cost-effective method for identifying densovirus in P. ochraceus infected with sea star
wasting disease. This would aid in expanding the accessibility of sea star wasting
literature to those with limited genetic expertise and restricted funds.
The method implemented to achieve this project’s research goals was Scanning
Electron Microscopy (SEM). SEM is a fantastic method historically used to visualize
anatomical structures in biologic organisms and identify pathogens such as bacteria,
fungi, and viruses in organisms and people inundated with disease. The traditional
methods for SEM sample preparation are achievable for those with limited research
funding. Furthermore, Scanning Electron Microscopes do not require extensive
knowledge and lengthy training. Instead, advanced microscopy training is obtainable in a
matter of weeks and comprehension of the theory behind SEM functionality is much
easier to attain and interpret than metagenomics. Therefore, this thesis focused on
investigating if the SEM could be a viable method for identifying densovirus in P.
ochraceus to assist in developing an accessible and cost-effective method to study sea
star wasting disease.
3
This thesis will begin with an overview of the keystone species central to this
project, Pisaster ochraceus. The literature review will focus only on those aspects of P.
ochraceus integral to understanding the nature of sea star wasting disease. These include
P. ochraceus’ relationship to the wasting phenomenon such as physiological responses
and will address the impact of reduced P. ochraceus populations on the Pacific rocky
intertidal zones. Next this thesis will review the literature on sea star wasting disease,
placing the greatest focus on research performed since 2014. This was the year Hewson
et al. (2014) discovered densovirus in P. ochraceus and is responsible for launching sea
star wasting disease literature into a new chapter of understanding. Topics discussed
within sea star wasting literature are disease characteristics and potential causes and
correlations to disease outbreak and spread.
Following sea star wasting disease literature, this literature review will pay
particular focus to current knowledge of sea star-associated densovirus and other related
densoviruses which similarly affect invertebrates in terrestrial and marine hosts. Next the
feasibility of implementing Scanning Electron Microscopy will be examined as a method
for identifying SSaDV in P. ochraceus tissues. A review of traditional SEM methods and
theory will proceed, with care taken to explain its distinction from Transmission Electron
Microscopy. The literature review will conclude with an explanation of SEM
implementation for this research.
4
Literature Review
Pisaster ochraceus
One of the organisms’ integral to maintaining biodiversity on the rocky intertidal
zone is the purple-orange ochre sea star, Pisaster ochraceus (Figure 1). P. ochraceus are
a keystone species, meaning their presence in the rocky intertidal ecosystem is important
because of their influence on other species in these environments, particularly mussels
(Paine 1969). These are the organisms which inspired Paine in 1969 to propose the
concept of a keystone species. A keystone species exists as a highly important member of
an ecosystem because of the influence it has on other organisms and their interactions. It
is a species that regulates the relationships in an ecosystem, and without them, the
composition of the ecosystem would become drastically different or could potentially fail
(Paine 1969, Eisenlord et al. 2016, Menge et al. 2016).
5
Figure 1. A healthy orange P. ochraceus in the rocky intertidal zone.
This keystone species concept, has been adopted most especially in conservation
biology to explain complex ecosystem relationships and champion species conservation.
Contrarily, this concept has also been refuted because it has the potential to oversimplify
complex species interactions within ecosystems (Paine 1995). Nevertheless, Paine’s
observations about the influence P. ochraceus have over molluscs and crustaceans such
as bay mussels (Mytilus tossulus), acorn barnacles (Balanus glandula) and the California
6
mussel (Mytilus californianus) are relevant to the Pacific Northwest intertidal ecosystems
today (Lawrence 2013). It exists as the basis for continued research exploring P.
ochraceus and their importance on North American Pacific coastlines.
The sea stars which inhabit the rocky intertidal ecosystem are fascinating
organisms in that they have remained relatively static in their physiology and anatomy
since the Triassic period, 208-245 million years ago (Lawrence 2013). Through their
unchanging characteristics they have been named a “living fossil.” Although it is
understood that the class of asteroids (Asteroidea) have remained fairly static through
time and evolution, the phylogeny or rather, the genetic origins of the current Class of
Asteroidea we see today, come from a single ancestor. This ancestor was most likely able
to survive the Permian-Triassic boundary or so called the Great Permian Extinction,
responsible for the dinosaur extinction 252 million years ago (Lawrence 2013). From that
single surviving ancestor, we see the rise of a clade of asteroids known as the
Neoasteroidea. These Neoasteroidea are the sea stars which all currently living
Asteroidea derive from (Gale 1987), including Pisaster ochraceus.
The North American Pacific coast is where P. ochraceus inhabit the dynamic and
often harsh environments of the rocky intertidal zones. They range from coasts in Alaska
to sunny Baja California. Their bright purple and orange pigmentation are a treasure to
the Pacific Northwest. Their presence in the rocky intertidal zone is largely responsible
for maintaining species biodiversity in this extreme habitat (Paine 1969). P. ochraceus
have a highly voracious appetite and they use it to eat the most abundant shellfish on the
intertidal zone: mussels (Lawrence 2013, Eisenlord et al. 2016). Mussels are a concern
for this ecosystem along the North American Pacific coast because their presence in the
7
rocky zones can be prohibitive to other organisms seeking prime intertidal habitat (Paine
1969). Often, if left unchecked, mussels rapidly multiply and outcompete other species
for space such as sponges, chitons, limpets, crabs, anemones and algaes, reducing
biodiversity in the intertidal zone (Paine 1969, Eisenlord et al. 2016). This decreases the
number of species existing in prime rocky intertidal habitat, altering the species
composition of the coastline.
Sea Star Wasting Disease
In the Pacific Northwest P. ochraceus have been a perpetual staple to beaches and
coastal systems (Paine 1967). However, in recent years this steadiness has been
interrupted by a disease which has swept the North American Pacific coastline,
decimating populations of sea stars. This disease is called sea star wasting and has been
most devastating to P. ochraceus since 2013 when the first signs of disease outbreak took
hold (Fuess et al. 2015). Since the mass wasting event which lasted into late 2015,
populations of P. ochraceus have rebounded in many of their usual habitats and recently
even began approaching pre-outbreak numbers along the North American Pacific
coastline. Although the number of sea star individuals is reaching pre-outbreak levels, sea
star population densities have not rebounded in the same way, with populations more
spread out along the coast (Moritsch & Raimondi 2018, Eisenlord et al 2016). As a result,
mussels are still not experiencing the same level of predation pressure as populations
prior to 2013. Sea star recovery is not uniform and the rocky intertidal zone will continue
to be altered by the impact of sea star wasting disease in years to come (Moritsch &
Raimondi 2018).
8
Sea star wasting is a disease that most negatively affects sea stars of the Class
Asteroidea. When the animals contract the disease, the first sign of infection are white
lesions that appear on the top of the body either on the arms or the central disk (Staehli et
al. 2009) (Figure 1). As the disease progresses, the lesions grow in size and multiply
across the body of the sea star. The animal loses its turgidity so they are no longer plump
and full but instead deflated and appear to collapse on themselves (Eisenlord et al. 2016,
Fuess et al. 2015). Typically, a sea star will lose the ability to move their tube feet and
slowly become immobile. They are unable to hunt and feed, so as the disease spreads,
their bodies lose their coloring, turning their tissues into a white mass of lesions (Menge
et al. 2016, Hewson et al. 2014, Fuess et al. 2015). Sea stars usually collapse and die
within one to two weeks of contracting wasting disease (Eisenlord et al. 2016, Hewson et
al. 2014).
9
White
Lesions
Figure 2. Image of a P. ochraceus infected with early stages of sea star wasting disease.
The sea star was a deflated appearance with white lesions on the top of the central disk
and sides of the arms. The left most arm is beginning to curl, a typical response to
wasting disease in this organism.
10
Loss of Turgidity
Figure 3. A P. ochraceus infected with sea star wasting disease. This organism exhibits
signs of turgidity loss with increased sponge-like, hollow qualities in the arms and
decreased mobility.
Sea star wasting disease is highly fatal in the majority of sea stars infected such as
the sunflower sea star (Pynapodia helianthoides), the morning sun star (Solaster
dawsoni) (Eisenlord et al. 2016), the forbes sea star (Asterias forbesi), and the
purple/orange ochre sea star (Pisaster ochraceus) (Bucci et al. 2017). There is however, a
silver lining in that it is not entirely fatal in adults of the species, P. ochraceus. The
purple/orange ochre sea star implements the unique asteroid anatomy that allows the
animal some protection from wasting syndrome. The tissues of P. ochraceus are held
together by what look like a woven mesh of tissue. P. ochraceus has complete control
over the fluidity of the tissues and are therefore able to relax their tissues and tear
themselves apart (Mothersill & Austin 2000). This act is called autotomy and is
11
particularly useful in staving off death from sea star wasting disease. When P. ochraceus
contract a lesion on an arm, they are able to autotomize the limb to save the central
nervous system and prevent disease spread throughout the rest of the body. When the
infected limb is removed, P. ochraceus can heal and regenerate the autotomized limb by
growing another in its place (Mothersill & Austin 2000).
Although P. ochraceus have the ability to save their bodies from sea star wasting
disease through autotomy, this did not save them from experiencing mass die-offs from
wasting syndrome from 2013 to 2015. Populations of P. ochraceus from Baja California
to Juneau Alaska were decimated by sea star wasting disease (Hewson et al 2014, Menge
et al. 2016). In Washington State, P. ochraceus mortality from sea star wasting was most
pronounced in the San Juan region with significant losses incurred on the coast as well as
the Puget Sound (Eisenlord et al. 2016) The rapid decline in P. ochraceus populations
coupled with the length of the population rebound period, has made it difficult for the
rocky intertidal zone to maintain species composition on the Pacific coast of North
America (Menge et al. 2016, Eisenlord et al 2016, Miner et al. 2018).
Temperature appears to have a significant effect on sea stars and their
susceptibility to wasting disease (Kohl et al. 2016, Bonaviri 2017). Often, increases in
temperature are implicated with the presence and mortality from wasting in P. ochraceus
which can be especially rapid for juveniles (Bates et al. 2009, Eisenlord et al 2016,
Staehli et al 2009, Bates et al. 2009, Miner et al. 2018). Adults in warmer waters contract
sea star wasting disease faster than juveniles. However, juvenile mortality to wasting
disease is more dependent on increases in temperature than it is for adults (Eisenlord et
al. 2016). The greatest correlation between temperature and wasting events exists in the
12
relationship between the wasting disease outbreaks in 2009, and then from 2014-2015. In
2009, sea star wasting appeared as a minor blight in the Puget Sound. Bates et al. (2009)
found that at the time, minor increases in ocean temperatures had a strong influence on
the presence and mortality of wasting disease in sub-tidal zones in Vancouver, Canada.
In 2013 the mass mortality P. ochraceus experienced on the coastline began but
was then exacerbated in conjunction with warming ocean temperatures in the Pacific
(Miner et al. 2018). Wasting continued through 2014 with the increases in water
temperatures and then hit a peak in 2015 with ongoing warming in the Puget Sound and
San Juan Islands. In the Puget Sound region of Washington state, wasting events spiked
especially from Winter to Spring in 2014 (Eisenlord et al. 2016, Miner et al. 2018). P.
ochraceus populations were immersed in higher temperatures, experienced advanced
proportions of infected sea stars. Since adults’ contract wasting disease faster than
juveniles particularly in warmer waters, the wasting events that occurred during these
years severely reduced adult populations by a quarter of their original population size,
thus decreasing the number of reproductive individuals on the rocky intertidal zone
(Eisenlord et al. 2016).
Although the relationship between temperature and sea star mortality to wasting
disease is strong, the prevalence of wasting may be driven by factors which are not
exclusively environmental. In 2009, Pankey & Wares identified a genetic mutation
present in P. ochraceus called EF1A. This gene is overdominant in P. ochraceus which
means that the gene is phenotypically outside the range of the homozygous parent genes.
Those organisms which carry this gene are thus heterozygous (Chandler & Wares 2017).
This means that the sea stars with this new gene carry two different sets of alleles for the
13
same gene, while those sea stars without the mutation carry two of the same sets of
alleles. Therefore, unmutated sea stars are homozygous. Overdominance of the EF1A
gene is quite possibly a positive contribution to P. ochraceus genetic makeup because it
is linked to greater disease tolerance (Pankey & Wares 2009). Chandler & Pankey (2017)
recently found that individuals with the EF1A mutation carried an increased ability to
respond to stresses from increased temperatures. Therefore, heterozygous P. ochraceus
carrying EF1A may have greater fitness against sea star wasting disease and increased
environmental temperatures (Chandler & Pankey 2017).
During the 2014 wasting event, temperature seemed to increase the prevalence of
sea star wasting disease on the Pacific Northwest coast of North America. However, later
in 2014, Hewson et al. presented evidence that sea star wasting is also strongly correlated
with presence of a densovirus, from the Parvoviridae family called the sea star associated
densovirus (SSaDV) (DelSesto et al. 2015, Eisenlord et al. 2016, Menge et al. 2016). The
relationship between SSaDV and sea star wasting disease has been explored in multiple
sea star species on both the east and west coasts of North America (Hewson et al. 2014,
Bucci et al. 2017). A great deal of evidence exists which implicates the SSaDV in the sea
star wasting disease epidemic and sea star mortality as a result. However, the topic of
correlation versus causation is challenging because of the nature of pathogen detection
and the variability surrounding sea star wasting disease and SSaDV presence.
The Densovirus
A well-known viral family which infects both terrestrial and marine organisms is
the Parvoviridae or better known as parvo for short (Cotmore 2014). Parvoviridae viruses
14
are nonenveloped meaning they do not have the lipid membrane around them that carries
viral proteins and often acts as a method for binding to the cells of their host (Yang et al.
2016, Bruemmer et al. 2005). Instead, these viruses must replicate through cell division
and have been known to be more lethal in young hosts as a result (Bruemmer et al. 2005).
A genus of the Parvoviridae family is the Densovirinae or rather densoviruses.
Densoviruses are single-stranded DNA viruses that typically infect invertebrates in
terrestrial and marine ecosystems such as crustaceans and insects (Jackson et al. 2016,
Bruemmer et al. 2005, Bucci et al. 2017, Tijssen et al. 2016). The sea star-associated
densovirus (SSaDV) is a circular single-stranded DNA virus suspected of contributing to
sea star wasting disease in echinoderms (Fahsbender et al. 2015, Hewson et al. 2014,
Bucci et al. 2017).
15
Figure 4. The above figure is a graph from research performed by Hewson et al. (2014).
It depicts the relationship between presence of SSaDV and sea star wasting disease in
symptomatic individuals. For this analysis 107 sea stars of three species Pisaster
ochraceus, Pycnopodia helianthoides, and Evasterias troschelii were tested for presence
of SSaDV. Physiological characteristics such as sea star size, and population abundance
with SSaDV presence in the organism were statistically analyzed against each other to
test for independent variables which may predict sea star wasting disease in these three
species of asteroids. Hewson et al. (2014) found that presence of SSaDV was the variable
which most effectively predicted the presence of wasting disease in the sea stars tested.
In 2014, the densovirus became the first virus associated with echinoderms in
Hawaii and Asteroids in the Northeast Pacific Coast. Previous research on densoviruses
demonstrates an expanding host range in marine plants and invertebrates such as the
oyster, Crassostrea ariakensis (Kang et al. 2017), sea barley (Hordeum marinum)
(Francois et al. 2014), and three urchin species off the coast of Hawaii; the helmet urchin
(Colobocentrotus atratus), collector urchin (Tripneustes gratilla) and burrowing urchin
(Echinometra mathaei) (Gudenkauf et al. 2014). Since then, sea star-associated
densovirus (SSaDV) has been found in many of the sea stars infected with sea star
wasting disease such as the sunflower sea star (Pynapodia helianthoides) and over 20
16
other asteroids in Northeast Pacific coast (Hewson et al. 2014, Smith 2017). Presence of
densovirus in sea stars has also been found on the Northeast coast of North America in
the forbes sea star (Asterias forbesi). Phylogenetically, the densovirus nearest to SSaDV
is an Ambidensovirus identified in a freshwater crayfish (Cherax quadricarinatus) called
Decapod ambidensovirus variant Cherax quadricarinatus densovirus (CqDV) (Bochow
et al. 2015). At the time of discovery, the SSaDV had not been genetically identified.
Thus, this new densovirus was named for its association to the wasting disease and the
organisms most affected, sea stars (Hewson et al. 2014).
Although SSaDV holds the strongest pathogen correlation to sea star wasting
disease to date, questions about SSaDV causation of wasting disease are still present. The
enigma of this relationship, lies in the fact that SSaDV cannot fulfill the decisive test
which determines the relationship between a pathogen and a disease. This test is
demonstrated by Henle-Koch Postulates or Koch’s postulates as they’re more commonly
known. These guidelines offer the most secure avenue for demonstrating a pathogen is
responsible for causing a disease. The first of these postulates states that “The parasite
occurs in every case of the disease in question and under circumstances which can
account for the pathological changes and clinical course of the disease” (Evans 1976).
This is not true for the relationship between SSaDV and instances of sea star wasting
disease as not all animals who have shown symptoms of the disease tested positive for
SSaDV presence using viral metagenomics (Hewson et al. 2014, Bucci et al. 2017).
The second postulate states that “It occurs in no other disease as a fortuitous and
nonpathogenic parasite” (Evans 1976). At present the sea star-associated densovirus has
not been found in any other disease though it is present in sea stars that do not have
17
symptoms of wasting and has been found in specimens from natural history collections,
stored for 70 years, without any symptoms of wasting (Hewson et al. 2014). The third,
and most challenging to fulfill, is the postulate which demands that “After being fully
isolated from the body and repeatedly grown in pure culture, it can induce the disease
anew” (Evans 1976). This third postulate is nearly impossible with our current
technology and understanding because no cell lines exist for asteroids (Mothersill &
Austin 2000, Bucci et al. 2017). Furthermore, isolating pure densovirus from the host has
not yet been achieved and has the potential to be highly improbable to nearly impossible
(Bucci et al. 2017). Subsequently, the inability for SSaDV to fulfill Koch’s Postulates to
sea star wasting disease is why testing for SSaDV transmissibility in sea stars has largely
involved challenge experiments.
Although the relationship between SSaDV and the disease, sea star wasting
remains inconclusive based on our current standards for pathogen detection, it is
important to continue to study this relationship especially in invertebrates. Viruses in
invertebrates are highly understudied since most viral work tends to focus on vertebrates,
as the risk of disease spillover to humans is higher in these organisms (Gudenkauf et al.
2014, Jackson et al. 2016, Francois et al. 2016). Therefore, our understanding of viral
interactions in echinoderms such as sea stars and urchins is limited (Gudenkauf &
Hewson 2016, Jackson et al. 2016). The discovery of the SSaDV in tissues of sick sea
stars has sparked further research into viruses that may be associated with echinoderms to
better understand the host ranges of viral pathogens in invertebrates.
Prior to 2014 (Gudenkauf et al.) no echinoderm-associated virus had been named.
Francois et al. (2016) demonstrated a much larger host range for viruses in the
18
Parvoviridae family, such as densoviruses, than had been previously thought, including
two asteroid orders, Paxillosida and Spinulosida. Furthermore, viral presence and
influence in the aquatic environment is vastly underestimated (Suttle 2005). Gudenkauf et
al. (2014) posits viruses may be causing many echinoderm diseases, however, this cannot
be confirmed given the current gaps in the echinoderm-viral literature. It is therefore
necessary to continue to increase our knowledge of viral interactions in these organisms.
This is important not only to understand how viruses may be linked to sea star wasting
disease but to increase our knowledge of the influence they may have on trophic cascades
in the marine ecosystems (Buck 2017).
Testing for the Densovirus
Viral Metagenomics is the primary method Hewson et al. (2014) used to identify
correlations between densovirus presence and sea star wasting disease. Metagenomics
has been a breakthrough in sea star wasting research because it does not require sea star
cell cultures, a feat which has not been possible since no cell lines exist in marine
invertebrates as a whole (Jackson et al 2016, Bucci et al. 2017). Most of P. ochraceus’s
ability to regenerate derive from their unique tissue and cell structure. Many organisms in
the Class Asteroidea are comprised predominantly of tissues with multiple cell types and
functions within each tissue (Mothersill & Austin 2000). This has made it challenging to
create cell cultures of sea star tissues which, until the use of metagenomics, has greatly
hindered research on viruses and diseases in these organisms.
Although metagenomics has allowed for wonderful scientific breakthroughs
regarding the study of viruses in echinoderms because of its use of molecular diagnostic
assays (Hewson et al. 2014, Jackson et al. 2016), this method is extremely technical and
19
highly expensive, costing on average $1000.00 to process one sample for accurate viral
identification. Another method used to identify densovirus in sea stars infected with
wasting disease is Transmission Electron Microscopy (Hewson et al. 2014). This method
uses electrons to penetrate the sample and obtain a microscopic image of the virus’s
form. It is a proven method for effective identification and although it does not provide
the same positive assurances as genetic testing, it can be accomplished without species
specific knowledge of protein structures or organism cell cultures.
SEM and Densovirus
In 2014, Hewson et al. successfully captured an image of the SSaDV using
Transmission Electron Microscopy. Their image resolved a virus that was 0.020-0.025
microns in diameter. The shape of the virus was icosahedral so that it appears as many
equilateral triangles arranged together to create a geometric hexagon shape. Similarly, a
Transmission Electron Microscope was also successful at resolving an image of an insect
parvovirus, Junonia coenia densovirus (JcDNV) captured by Bruemmer et al. (2005),
which belongs to the same densovirus genus as the SSaDV. Bruemmer et al. describes the
shape of JcDNV as icosahedral which illustrates the shape as a collection of equilateral
triangles symmetrically arranged. This description of JcDNV coincides with the SSaDV
description by Hewson et al. (2014) using Transmission Electron Microscopy after
having performed the metagenomic analysis to understand the physical characteristics of
SSaDV.
20
Figure 5. Image of an icosahedral single stranded circular DNA virus captured by
Hewson et al. (2014) using Transmission Electron Microscopy. The viral particle is
0.020-0.025 microns in diameter. Hewson et al. (2014) believes this image to be a
representation of the structure of the sea star-associated densovirus (SSaDV).
Transmission Electron Microscopy is an important method used most frequently
for viral identification because electrons emitted by the electron beam are able to
penetrate the sample. For this to occur, the samples themselves must be of a very thin
size. The resulting image captured by TEM is two dimensional, therefore the outline of a
potential virus is all that is resolvable by the TEM (Kessel & Shih 2012).
Another member of the microscopy family is the Scanning Electron Microscope
(SEM). For an SEM to capture images at the microscopic scale, electrons do not need to
pass through the sample. Instead the electrons bounce off of the sample and emit smaller,
weaker secondary electrons (Scanning Electron Microscope A To Z 2009, Kessel & Shih
2012). These electrons are collected and produce a light image which appears three
dimensional. This allows for better topographical visibility at the microscopic level that
would not otherwise be seen with the TEM (Kessel & Shih 2012). When it comes to viral
identification though, the methods for sample preparation are part of the more
21
challenging issues associated with successful viral identification with a SEM (Golding et
al. 2016).
While Hewson et al. (2014) resolved an image of densovirus, their methods for
sample preparation are different than those that are typically implemented for SEM. The
nature of TEM which allows electrons to pass through a sample, are beneficial to sample
preparation involving aqueous solutions (Golding et al. 2016, Kessel & Shih 2012). Often
TEM samples undergo staining so it is easier to discern between background noise and
the object itself. This allows for fewer obstructions and more contrast in the sample
(Kessel & Shih 2012). Sample preparations for SEM require that the sample be coated in
a metal such as gold or aluminum. This ensures that the sample will not experience
charging while inside the vacuum chamber. This is also important because the electrons
must be able to bounce off of the sample inside the chamber, meaning the sample itself
must be conductive. The challenge of resolving an image of a sample as small as a virus,
lies in the fact that the metal coating may also coat the virus, preventing successful
identification of densovirus (Golding et al. 2016). Nonetheless, this method is important
to pursue because if imaging is successful, it will not only provide another method for
viral identification, but can also help inform other SEM users about the ideal amount of
time sputter coating should occur for viral resolution.
SEM is a method used for quick and cost-effective identification of viruses and
bacteria. A SEM can successfully identify viruses in biologic organisms (Kessel & Shih
2012) but has not been attempted for viral identification of densovirus in P. ochraceus.
Since TEM offers the greatest contrast for viral resolution, work with SEM and viral
particles has been less popular. When SEM is implemented for research and is
22
unsuccessful at resolving images or obtaining the desired result, the trials go unpublished.
Consequently, lack of failed research contributes to gaps in SEM knowledge for methods
which have worked and those that haven’t (Golding et al. 2016). If the SEM is
unsuccessful at resolving densovirus in P. ochraceus, the method can be eliminated as a
source for sea star wasting disease research.
Scanning Electron Microscopy
A Scanning Electron Microscope (SEM) is one type of electron microscope able
to resolve images of inorganic and organic matter at microscopic levels (Kessel & Shih
2012). To put it in perspective, a light microscope is able to visualize samples at a
maximum of 1000X magnification, while a SEM can scan a sample at a maximum of
300,000X magnification (Wells 1974). This great difference in magnification ability is
due to the incredibly small size of the electron (Hayat 1974, Scanning Electron
Microscope A To Z 2009). Since the electron is 2.82x10-15 meters in diameter, the
theoretical size a sample can be visualized is slightly greater than the electron’s size
(Kessel & Shih 2012). Therefore, the electron’s incredibly small size is how the SEM is
able to provide a window to the microscopic world.
Components of the SEM
A SEM is comprised of several important parts which allow the machine to scan a
sample using electrons, they are the electron gun, microscope column and the sample
chamber (Hayat 1974). The electron gun rests at the top of the microscope column, it is
where the electrons are separated from the atoms and the electron beam, used to scan
23
samples, is created (Wells 1974, Scanning Electron Microscope A To Z 2009). The beam
is then sent down the second component of the SEM, the microscope column, which
focuses the beam of electron energy so that it can be manipulated to scan a sample (Hayat
1974). The beam is sent from the column into the sample chamber, a small box at the end
of the column (Kessel & Shih 2012). Inside the chamber are the microscope stage where
the beam of electrons is sent out, the sample placed at the base of the chamber, and
sensors to detect electrons (Scanning Electron Microscope A To Z 2009, Wells 1974).
Figure 6. Image of The Evergreen State College’s Scanning Electron Microscope. Model
JEOL JSM-6480LV. To the left is the microscope and to the right is the computer
monitor with SEM control panel between them.
24
SEM Operation
As the beam of electrons comes down the microscope column and into the sample
chamber, the electrons are focused onto the sample. When electrons hit a part of the
sample, they release weaker negative electrons (Wells 1974). These weaker electrons are
collected by the sensor which converts the electron reading into light that is projected
onto the computer monitor as an image (Heywood 1971). When electrons hit a part of the
sample that is higher up from the sample holder, weaker electrons are released. When the
electron beam touches an area lower to the sample holder, few or no electrons are
released. The more electrons released when it hits the sample, the greater the light
produced by the microscope will be (Scanning Electron Microscope A To Z 2009). Thus,
the images produced are in a type of grayscale and the differences in light and dark are
actually the differences in topography of the sample.
25
Figure 7. Image of The Evergreen State College Scanning Electron Microscope sample
chamber.
Standard SEM Methods
The priority for this project was to test the ability of the Scanning Electron
Microscope to detect and successfully visualize the densovirus in P. ochraceus infected
with sea star wasting disease. The standard Scanning Electron Microscopy method coats
the sample in a metal such as gold or aluminum (Kessel & Shih 2012, Hajibagheri 1999).
This is important because samples in the SEM need to be conductive so the electrons can
react to the surface of the specimen and produce a light image (Heywood 1971). If the
sample is not conductive, the electrons cannot react and bounce off the surface of the
specimen and the electrons will refract, making it impossible to obtain an image. Prior to
26
coating the specimen with a metal, it is important that the tissue is dried in a Critical
Point Dryer so that there is no water present in the sample (Hajibagheri 1999). This is due
in large part to the fact that the SEM holds the specimens in a vacuum chamber so any
organic matter will dry out and shrink. Wet samples are therefore difficult to process in
the SEM (and also highly dangerous because a wet sample can combust inside the
vacuum chamber). It is necessary to perform critical point drying to remove water from
the sample prior to sputter coating (Kessel & Shih 2012, Heywood 1971).
There are several potential issues with these methods in relation to studying
viruses and sea star tissue samples. The first is that the drying process will have the
negative effect of altering the topography of the tissue sample and thereby reducing the
ability of the SEM to detect fine details. The second issue is that sputter coating the
sample with metal to create conductivity may obscure fine details making it challenging
to see any viruses that may be present in the sample (Golding et al. 2016). However, it is
important to try these methods with P. ochraceus tissues as they are the standard and
have been successful at producing quality images of viruses in other organisms and in
people for rapid disease diagnosis (Heywood 1971, Kessel & Shih 2012, Golding et al.
2016)). Thus, this project seeks to understand if standard SEM methods are a viable
option for studying densovirus in P. ochraceus.
Successful development of a method which implements SEM would provide
additional resources to study the phenomena of sea star wasting disease. Furthermore,
this method would offer researchers with limited funding, an opportunity to contribute to
the sea star wasting literature without the financial burden genetic testing often imposes.
With a developed method for successful viral identification in echinoderms using SEM,
27
there is potential for other students at Universities housing SEMs to participate in larger
research discussions and experimentation on sea star wasting disease.
28
Methods
Methodological Introduction
The primary focus of this project was to explore and develop a method to identify
densovirus in P. ochraceus infected with sea star wasting disease. This thesis is a
methods development thesis and highly experimental in nature. Thus, the process for
resolving the viral material was complex and involved a great deal of trial and error with
few fruitful results. Consequently, the layout of this methods is arranged in order of
experimental trials and results.
These methods begin with an overview of the field collection of P. ochraceus
tissue samples with a brief explanation of the data collection process using the GIS
Collector application. Then the first filtration method, which utilized the pop-top filter
head unit, is described. Following Method 1 is a Results & Discussion, elucidating why it
was not appropriate for this project. Next the second attempted method, vacuum
filtration, is introduced with a proceeding Results & Discussion section. The third method
follows the same model as the prior two methods sections but chronicles the final
filtration method which implemented the Swinlock filter head unit. A Results &
Discussion is then presented for this method to explain why filtration as a whole was
unsuccessful, with discussion of what could have been done differently and possibilities
for the future in regards to micro-filtration recommendations.
Finally, the fourth method utilized in this project is described. This method
involves whole P. ochraceus tissue sample preparation for SEM analysis. Following this
method, an explanation of how this project intends to identify densovirus is discussed
based on current literature with Transmission Electron Microscopy (TEM) images of this
29
viral genus as reference sources. Finally, the methods reinforce the comparison between
SEM and TEM to clarify the difference between the method used for this project and the
method used in the literature to resolve densovirus. Finally, these methods discuss the
implementation of the GIS Collector application for data collection and explains the
usefulness of displaying sea star data through this platform. The Results section,
following the full Methods section, will expand on findings from whole P. ochraceus
tissue sample preparation and analysis.
Tissue Sample Collection
P. ochraceus were collected from the Puget Sound and Pacific Coast in
Washington state. Individuals were chosen as samples based on their degree of wasting
disease. Sea star tissues decimated with wasting were collected because this is the tissue
that carries the largest quantity of densovirus if it is present in the organism (Hewson et
al. 2014). Diseased sea star tissues were removed using an 8 mm sterile punch biopsy
commonly used in veterinary medicine. The animal tissue was then placed in tin foil and
preserved in a cooler with dry ice. The cooler was taken back to The Evergreen State
College lab and tissue samples were stored in a deep freezer at -80 degrees Celsius.
Environmental Data Collection
Tissue samples were obtained from Tacoma Narrows, Narrows Marina and
Boston Harbor Marina in South Puget Sound, Washington. Each collection site was
chosen based on historical presence of P. ochraceus with wasting disease as documented
in Eisenlord et al. (2016), the Rocky Intertidal Monitoring Group citizen science data
30
collected during sea star wasting outbreaks from 2013-present. Sites were also selected
based on personal observations of historically dense sea star populations in the Puget
Sound prior and during the 2013-2015 sea star wasting disease outbreaks. A GIS
collector app was created using ArcGIS 10.5 prior to sample collection with all pertinent
environmental and physiological attributes that would be recorded in the field. The
following information was recorded: GPS coordinates, Total P. ochraceus at site, Color
(purple/orange), Size (radius in centimeters), Age (juvenile/adult), SSWD presence
(yes/no), average water Temperature (celsius), Salinity (ppt), Dissolved oxygen (mg/L),
and Conductivity (spc). P. ochraceus age was determined according to length of the sea
star radius. Juveniles are any sea star unable to reproduce; P. ochraceus can reproduce
after 75 mm radius (Eisenlord et al. 2015). Therefore, juveniles are any P. ochraceus
under 75 mm and adults are sea stars with a radius greater than 75 mm.
A picture was captured of each sea star when a tissue sample was removed at
every sample location. Measurements were taken of sea star radius size in inches, the
color and approximate life stage as well as the degree of wasting disease. Other
environmental variables such as salinity, dissolved oxygen, and conductivity were
measured on site during the day of tissue collection using a Yellow Springs Institute
(YSI) meter. Water quality measurements were taken from 5 separate points in the
intertidal zone where wasting was present to capture ancillary environmental data while
recognizing the inherent variability of ecological measurements in the intertidal zone.
31
Method 1
Filtration
To best visualize densovirus under SEM, P. ochraceus tissue sample filtration
was attempted for this research. This is important to eliminate background noise that
would be present in a sample prepared with tissue and other organic matter around the
viral material. Viral isolation from tissue was successfully performed by Hewson et al.
(2014). This is possible through homogenization and the fact that densovirus can persist
in the environment and subsequently, sea water bases.
The filtration method for this project was developed through a combination of
methods by Hewson et al. (2014) and Golding et al. (2016). Both sources implemented
viral isolation, one for TEM analysis and the other for SEM respectively. Although
Hewson et al. (2014) isolated densovirus successfully from sea star tissue, their method,
which created a viral pellet suspended in fluid, was not appropriate for SEM sample
preparation methods as it would not provide a platform to dry out the viral material to
mount on a sample stud. The resulting viral material needed to be mounted on a filter so
that ethanol could be passed through the sample to fully dry it out and mount on a 10 mm
SEM sample stud. Therefore, sample homogenization and centrifuge were performed
following Hewson et al. (2014), while viral filtration using polycarbonate track-etched
membrane filters was adapted from Golding et al. (2016).
Filtration Testing
Proper sample filtration and viral isolation was the primary focus of this stage of
method development. The first filtration method attempted, utilized procedures published
32
by Golding et al. (2016). The filters were 13 mm polycarbonate track etched membrane
filters with a pore size of 0.015 µm. This filter diameter was chosen in part because it is
easily seated on SEM sample studs with diameters of 10 mm and would not require
trimming to mount the filters to the studs, post-filtration. The pore size is necessary to
capture densovirus measured by Hewson et al. (2014) at approximately 0.020-0.025 µm.
The corresponding filter holders utilized would be compatible with 3 mL Luer-Lok
syringes and 13 mm polycarbonate track etched membrane filters. The filter holders
implemented were the Whatman Pop-Top Syringe-Type Reusable Membrane Filter
Holders, with a 13 mm diameter at dimensions 2.7 cm x 2.7 cm.
Result & Discussion
Initial testing of the Pop-Top filter holder and filter membrane was performed
using deionized water to assess the filtration rate without tissue samples. This was
important to understand how the filter should behave under normal conditions to
recognize how it acts when it clogs. This was a concern because clogging was frequently
an issue Golding et al. (2016) experienced with 13 mm polycarbonate track etched filters
having a 0.05 µm pore size. While Golding et al. (2016) implemented a Swinnex filter
head unit for the 25 mm filter diameter, this filter head was unavailable for this research.
Therefore, this project relied on the most compatible filter holder for the 13 mm diameter
membrane, the Whatman Pop-Top Syringe Filter Holder.
This Pop-Top Filter Holder is reusable and autoclavable. It seats the filter
membrane between the filter head components and the third piece locks them together by
snapping the front and back of the unit to seal the washer around the filter membrane. To
33
test this apparatus, 3 mL of deionized water was filtered through the unit and 13 mm
membrane filter. The filter head unit immediately leaked water out the sides of the filter
head and no water was able to make it through the tip of the syringe head. This meant
that the water wasn’t actually able to penetrate the filter because the unit could not create
a solid seal between the washer and the membrane. The challenges with this filter unit
exist in the fact that the filter head snap was highly variable and wasn’t easily seated onto
the syringe head. Furthermore, when a solid fit could be achieved, the filter unit lost most
of the water out the sides due to inefficiencies in the filter head unit design.
Correspondence with the manufacturing company, Whatman, elucidated a couple
issues with this filter set-up. One issue is the diameter of the filter membrane in
conjunction with the extremely fine pore size used for this project provides inherent
challenges with filtration. The pore size itself is so small, filtration should take a very
long time, however, a decent amount of pressure should be placed on the filter to force
the water through the membrane. The recommendation was to increase the surface area of
the filter membrane in hopes that a larger surface area would evenly disperse the pressure
placed on the filter during filtration. Therefore, the filter membrane was increased from
13 mm to 25 mm diameter to attempt this method with Whatman’s recommendations.
The other challenge this project experienced with the 13 mm diameter filter
membrane is that it was only compatible with one type of filter head unit, the Pop-Top
Syringe Type Filter Holder. This unit was unable to withstand the pressure necessary to
force water through the membrane. Whatman admitted this filter unit was not their most
reliable apparatus for micro filtration. Increasing the filter head diameter to a 25 mm
polycarbonate track etched membrane filter would also allow this project to experiment
34
with another, more reliable filter head unit, the Whatman Swinlock filter holder. Since
the Pop-Top Filter Head Unit and the 13 mm diameter were unable to produce an
effective filtration result with deionized water, this set up was not pursued as a viable
filtration option for this project.
Method 2
Vacuum Filtration
One of the challenges syringe filtration presented is that it only allowed a single
sample to be processed at a time. An alternative filtration method which could support
filtration of multiple samples and circumvent potential faulty filter head units was
vacuum filtration. This method utilizes a vacuum pump with side-arm Erlenmeyer flasks,
and a funnel to force a sample through the filter unit. Filtration is achievable because the
vacuum decreases the pressure in the chamber of the side-arm Erlenmeyer flask below
the funnel. The funnel seats the filter sample and as the pressure decreases, the sample is
filtered through the membrane. Multiple side-arm flasks can be connected to a manifold
to process up to 10 samples at once.
This method was first tested using deionized water and the Whatman 25 mm
polycarbonate track-etched membrane filters with the 0.015 µm pore size. Side-arm 250
mL Erlenmeyer flasks were connected to The Evergreen State College lab’s house
vacuum. This trial implemented rubber stoppers, to hold the glass towers, with 20 mm
diameter glass funnels. Polycarbonate track etched membrane filters were placed in the
bottom of the glass filters against the filter base. Deionized water was added to the funnel
with the filter seated at the base and the vacuum initiated. Filtration with this method was
35
unsuccessful as the polycarbonate filter immediately lifted from the base of the glass
filter. Thus, this setup was not effective filtration with this filter type.
To prevent filters lifting from the funnel, a filter tower with screw cap and
washer, was used as a replacement for the glass funnel because it effectively seats the
filter and seals it between the apparatus. The filter was secured inside the filter tower and
the side-arm flask was attached to the vacuum, then 20 mL of deionized water was added
to the top of the tower. After 2 minutes, the deionized water did not filter through the
membrane and the vacuum was shut off. Since the house vacuum was not successful at
pulling water through the filter, the Erlenmeyer flask was attached to a standing vacuum
because it offered controlled pressure settings and could increase the vacuum power
greater than the house vacuum. The standing vacuum was adjusted to pull at maximum
pressure and was allowed to sit for 1 minute. The standing vacuum was unsuccessful at
pulling deionized water through the filter membrane.
Result and Discussion
None of the vacuum filtration options attempted were able to successfully pull
deionized water through the 0.015 µm Whatman polycarbonate track-etched filter
membrane. The purpose of this filtration method was to increase the number of samples
that could be filtered at one time. Multiple sample filtration was to be accomplished
through the use of a filtration manifold. Unfortunately, the pressure needed to pull
deionized water through the filter membrane was insufficient with a manifold and not
implemented for this project. This is because a manifold distributes the same level of
pressure to ten side-arm Erlenmeyer flasks. The same pressure was inadequate at drawing
36
deionized water down through a single filter unit, thus diluting the pressure through a
manifold would be ineffective.
Vacuum filtration with the funnel was unsuccessful for known and unknown
reasons. The known issue with this method deals again with the incredible smallness of
the pore size. Communication with Whatman indicated that vacuum filtration may not
have provided enough differential pressure to filter deionized water through the 0.015 µm
pores. The maximum pressure that could have pulled the water down through the
membrane was achieved with the standing vacuum. Prolonged vacuum connection
beyond the 1 and 2 minute pressure intervals tested would not have increased the pressure
in the Erlenmeyer flask or the draw down from the funnel. This is because vacuum
filtration pulls the maximum pressure capacity within the first couple minutes and this
pressure would not increase with extended exposure to the vacuum. Therefore, the
maximum pulling capacity for the house vacuum was achieved and did not result in
deionized water filtration through the 0.015 µm polycarbonate track-etched filter
membrane. The polycarbonate membranes are hydrophilic so that they are able to
maximize their contact with water because of the filter surface’s natural attraction to
water molecules (Xie et al. 2005). Hence, polycarbonate filters should have no issue
filtering water-based solutions as they are the most well-equipped micro filter material
for this process. As such, it is curious that the pressure is insufficient at drawing water
through the filter membrane, even one as small as 0.015 µm.
37
Method 3
Swinlock Filter Holders & Polycarbonate Membranes
The final filtration method attempted referred again to the method implemented
by Golding et al. (2016) with some modifications based on obstacles encountered in
Method 1. Instead of 13 mm polycarbonate track-etched membrane filters, this method
used 25 mm diameter polycarbonate track-etched filters and the Whatman Swinlock
Filter Holders. Filters in the sizes 5 µm, 2 µm, 0.05 µm and 0.015 µm were initially
tested using deionized water similar to the previous methods, to test the flow rate and the
firmness of the filter holder seal. These pore sizes were initially used because they were
available as free samples through Whatman. Later, filters were chosen by powers of 5
beginning with the filter size goal of 0.015 µm. Filters purchased and implemented by
this research were as follows: 0.015 µm, 0.08 µm, 0.4 µm, and 2 µm.
Deionized water filtered easily through 5 µm and 2 µm filters. At 0.05 µm the
Luer-Lok syringe required more pressure but 3 mL could be pressed through the filter at
a rate of 1 mL per 5 minutes. The most challenging filter pores size was again, the 0.015
µm filter. Deionized water could pass through the filter head but with great care because
if the pressure placed on the filter head exceeded 52 psi the Swinlock filter head washer
seal would leak water through the sides of filter unit. If a pressure balance was achieved
that could both filter sample through the membrane while preventing leaks, then 3 mL of
deionized water could filter through the membrane in approximately 25 minutes. Thus,
the Swinlock Filter Holder and polycarbonate track-etched membrane filters were utilized
for sea star tissue filtration methods development.
38
Tissue Sample Preparation for SEM
Frozen tissue samples were placed in 10 mL of artificial buffered sea water to
prevent contamination from an environmental water sample. The sea water was used to
homogenize the samples using a mortar and pestle. After homogenization, samples were
centrifuged at 1000 X g for 5 minutes to separate larger tissue material following
methods published by Hewson et al. (2014). The liquid material was filtered using a
Whatman Swinlock filter holder and Whatman 25 mm polycarbonate nuclepore tracketched membrane filters. Tissue homogenate filtration was attempted using a 3 mL LuerLok syringe through descending pore sizes of 2 µm, 0.4 µm, 0.08 µm to the final pore
size of 0.015 µm. Unfortunately, tissues homogenized in buffered sea water were unable
to pass through even the largest filter pore size of 2 µm. The few remaining 5 µm
polycarbonate track-etched membrane filters from the first trials were then used for tissue
filtration. The 5 µm was also unsuccessful at passing tissue homogenate through the filter
membrane. As a result, tissue homogenates in sea water were too dense to pass through
the polycarbonate filter membranes and thus this was not a successful filtration method.
Sea star tissue was therefore homogenized in 10mL of deionized water and
centrifuged at 1000 X g for 5 minutes. Sample filtration was tested first on the largest
polycarbonate pore size available, the 5 µm filter membrane. Filtration with this
homogenate presented its own set of problems because, while some of the sample was
able to filter through the membrane, the pressure required to perform this filtration
exceeded the 52psi the Swinlock Filter Holder could withstand without leaking. Although
some of the sample could be filtered through the membrane, at least half the sample was
lost through the sides of the filter holder. Filtration with deionized water was
39
inconsistent. The sample was repeatedly diluted with increases of 20mL of deionized
water to create better filtration without too much pressure but the Swinlock Filter Holder
always leaked. Filtration of sea star tissue to separate potential viral particles, even at the
larger pore sizes, was unsuccessful at producing the desired filtration result and further
filtration decreases were not attempted after these failed trials.
Result and Discussion
This filtration method, though successful with deionized water, proved
challenging with samples of sea star tissue. Tissue samples that were homogenized and
centrifuged in artificial sea water were first filtered through the 2 µm polycarbonate filter.
Continuous, firm pressure was placed on the syringe for 30 minutes but the sample would
not filter through the membrane and Swinlock holder. The sample did not carry visible
particles but the solution did possess a thick gelatinous quality that potentially clogged
the filter. The sample was then removed from the syringe and diluted in more sea water.
After dilution, the sample was still resistant to filtration through a 2 µm filter. Therefore,
a 5 µm filter was then used to perform sample filtration but a similar result occurred with
the 5 µm filter and the solution was unable to pass through the membrane.
It is curious why the samples, with such high dilution, refused to pass through the
filters even those with larger pore sizes. Questions about the polycarbonate filter’s
compatibility with sea water arose but Whatman would not disclose detailed information
about its’ filter membranes, citing that the information was proprietary. Correspondence
with Evergreen faculty, Gerardo Chin-Leo, confirmed the suspicion that the homogenate
was in fact too dense with the use of sea water plus tissue. It was recommended that this
40
project attempt to homogenize sea star tissue in deionized water instead to reduce the
density of the sample. This recommendation proved to be true, as deionized sea water
created a fluid, less dense homogenate than samples homogenized with buffered sea
water which were pretty gelatinous and thick. However, filtration with deionized water
presented its own set of challenges, in that the homogenate was more prone to break the
Swinlock filter head unit washer tight seal and leaked through the filter head unit with
less pressure placed on it than the sea water homogenate. This caused at least half the
sample homogenate to be lost and was not an effective filtration solution for this project.
A component this project could not financially support is the implementation of a
syringe pump similar to the one used by Golding et al. (2016) to filter their sample
homogenates through 0.05 µm filter membranes. Syringe pumps are machines that can
accurately control the pressure placed on the filter units and hold continuous pressure for
long periods in ways a user cannot. Golding et al. (2016) was able to filter their samples
at a rate of 1000 µL per minute through a 0.05 µm filter unit using a syringe pump. Their
filters did experience clogging at this pore size and multiple dilutions were necessary for
full sample filtration but the syringe pump allowed them to have more control over the
filter unit than this project could achieve. Implementing a syringe pump for micro
filtration would be a recommendation this project would make for others pursuing a
similar methodology. Those with sufficient resources or syringe pump availability in their
laboratories should budget for at least one of these machines as filtration variability and
lack of pressure control was a primary issue with this methods development.
41
Whole Tissue Sample Preparation
Since sea star tissue filtration could not be achieved with any of the filtration
methods implemented by this project, whole tissue samples were processed and analyzed
with the SEM to scan for viral presence. This method was also implemented to examine
the difference between wasting and non-wasting sea star tissues at the microscopic level.
Tissues were frozen at -80 degrees Celsius then placed in 3% glutaraldehyde solution for
72 hours following methods published by Becker et al. (2004) (Hajibagheri 1999). Six
samples removed from wasted P. ochraceus, were submerged in increasing ethanol
concentrations of 20 mL of 50%, 70%, 90% and three 100% treatments each for 15
minutes per SEM tissue preparation methods from Knox University Faculty, Linda Dybas
which are standard methods implemented for soft tissue SEM sample preparation.
Although both methods from Dybas and Becker et al. (2004) did not indicate the
necessity to also place the samples in 100% acetone, this method was chosen, as acetone
acts as a better transition fluid to CO2 for optimum soft tissue desiccation during CO2
Critical Point Drying (Thuesen & Towanda, Personal Communication). Therefore,
samples were submerged in acetone and then critical point dried in acetone in the CO2
Critical Point Dryer sample chamber.
Experimental Drying
To compare the difference between tissue dried with ethanol and acetone, and
tissue dried in exclusively acetone, a seventh wasted sea star tissue sample was removed
from the top of an adult purple P. ochraceus. This sample was held frozen at -80 degrees
Celsius then placed in 3% glutaraldehyde solution for 72 hours following methods
42
published by Becker et al. (2004). This sample was then submerged in increasing ethanol
concentrations of 20 mL of 50%, 70%, 90% and three 100% treatments for 15 minutes
each (Linda Dybas). The sample was then submerged in ethanol in the CO2 Critical Point
Dryer during the final drying process. Upon Critical Point Drying completion, the sample
was removed from the sample chamber and found expanded and soaked in ethanol. This
indicates that the ethanol was insufficient at completely drying the soft tissue and was
unable to fully bleed out as CO2 was added to the sample chamber. This permanently
damaged the sample and so was thrown out.
Sputter Coating
The standard SEM method of sputter coating the sample in gold-palladium was
applied to this methodology for P. ochraceus tissue sample preparation. Samples were
dried in a CO2 Critical Point Dryer, then mounted on SEM studs, purged with argon and
sputter coated in gold-palladium. Standard SEM methods typically sputter coat a sample
for 120 seconds, however, the samples were coarse with variable topography so more
sputter coating was required to completely coat the sample in gold-palladium. Sample
studs, with sample mounts, were laid on their sides and sputter coated on both left and
right angles of the samples to ensure full gold-palladium coverage of each sample.
Although Golding et al. (2016) found that aluminum was also a viable conductive metal
which produced quality viral images in the SEM, The Evergreen State College lab does
not have the ability to attempt this method as they exclusively use gold-palladium.
43
SEM Calibrations
Each sample was mounted on a 10 mm carbon disk and affixed to a 10 mm SEM
stud. Samples were then placed in a 32 mm stud holder and inserted in to the SEM. The
SEM was initially calibrated with the aperture set to stage 2 and the voltage to 5kV with a
spot size of 45 and a working distance of 8 mm for all samples. The capacity of the stud
holder was for 3-cylinder studs or for 5-point studs depending on the style of stud the
samples were mounted on. The samples were first observed under View and then
gradually adjusted to Stage 1 then 2 for finer detail. Auto brightness and contrast were
selected first, and then auto focus second, to obtain the best image at each magnification
level. Magnifications of 38,000X and 40,000X were used to capture fine detail images of
densovirus per methods used by Bruhmmer et al (2005).
Sample Analysis
A total of 7 wasted sea star tissue samples were collected and prepared for whole
tissue SEM analysis. The first sample, removed from the top of an arm of an adult, purple
sea star was examined. The second sample was taken from the top of an adult orange sea
star. Sample three was removed from the top of an adult purple sea star and a secondary
sample was removed from this organisms’ underside to examine the tube feet, they are
called Sample 3 and Sample 3.5 respectively. This was also performed for the fourth
organism collected, so that we have both a Sample 4 representing the top of the organism
and Sample 4.5 representing the underside of the same specimen. A fifth sample was
collected from an infected sea star; however, this sample was botched due to improper
sample preparation in exclusively ethanol. The sample was not completely dried during
44
CO2 Critical Point drying which cause the tissues to expand. It was no longer usable and
was discarded.
To visually compare characteristics in wasted sea star tissue to healthy sea star
tissue, an 8th sample was collected from a healthy P. ochraceus for whole tissue analysis
under SEM. This sample was dried in both ethanol and acetone, since this method was
the most successful SEM sample preparation approach. It was analyzed under SEM and
used as a “control” sample to compare features and characteristics between wasted and
non-wasted sea star tissue. This sample was imaged after Sample 4.5, and will therefore
be referred to as Sample 5 to maintain continuity and avoid confusion.
Densovirus Characteristics
Successful identification of viruses in the Paroviridae family inside of P.
ochraceus tissues was performed with reference to images of an insect parvovirus,
Junonia coenia densovirus (JcDNV) captured by Bruemmer et al. (2005) using
Transmission Electron Microscopy (TEM). JcDNV belongs to the same densovirus genus
as the sea star-associated densovirus (SSaDV). Reference for the characteristics and a
single image of the SSaDV were taken by Hewson et al. (2014) using TEM (see Figure 5
in Literature Review) after having performed a metagenomic analysis to understand the
physical characteristics of SSaDV. Both Hewson et al. (2014) and Bruemmer et al.
(2005) were able to describe densoviruses as icosahedral viruses which looks like many
equilateral triangles symmetrically placed together. Hewson et al. (2014) successfully
captured a TEM image of the densovirus and recorded its size as being between 20 to 25
nm. These size and magnification specifications allowed this thesis to have a reference
45
for viral image resolvability using the SEM. The images taken by the TEM from both
papers were able to provide visual examples of how the densovirus would appear under
microscopy if present in the samples.
It is important to note, however, that both reference papers (Hewson et al. 2014 &
Bruemmer et al. 2005) implemented the Transmission Electron Microscope to produce
their sample images and this research used the Scanning Electron Microscope. Although
both technologies are able to see at the microscopic level, the TEM is different in that the
electrons emitted by the machine pass through the sample to resolve an image of the
object. Since TEM requires that the electrons be able to pass through the object, the
sample itself must be very thin so that this process can occur (Kessel & Shih 2012).
Subsequently, the image obtained is two dimensional which limits the depth of field for
the microscope.
Conversely, The SEM is able to detect material on the surface of a sample, but on
a microscopic scale. The SEM scans the topography of an object and is able to resolve
differences in topography through light given off by electrons when they touch the
surface of an image. The greater the light, the higher the sample is from the sample stage.
Therefore, the image obtained is often 3-dimensional, and is able to demonstrate
differences in depth of field better than the TEM. The benefit of using the SEM is that the
samples do not have to be particularly thin, in fact, whole organisms such as insects and
small animals and plants can be desiccated and inspected under the SEM. This provides a
better understanding of the cellular structures of organisms while they are intact and can
offer a better picture of the assemblages which comprise some of the smallest organisms
on earth.
46
Results
Introduction of Results
The following Results section chronicles images of whole P. ochraceus tissue
analyzed under the SEM. Samples were individually scanned to observe surface features
present within each sample to better understand the nature of P. ochraceus tissue infected
with wasting disease. The primary goal for this project was to ascertain whether or not
the SEM would be a good method for resolving densovirus in P. ochraceus. Thus,
samples were scanned with the focus of obtaining a quality, high resolution image of
densovirus in high magnification. The images shown below are the result of careful
sample scanning at multiple scales to increase image resolvability as magnification was
increased.
The Results are arranged in order of samples chronologically removed from the
field. Each sample is introduced as its own section with a description of the images at the
beginning of the sample section. Each sample section begins with a wide-view
micrograph of the sample, followed by images taken at multiple magnifications. The final
image at the end of each sample section was captured at the highest magnification
achievable by The Evergreen State College SEM.
Sample 1
The first image of sample 1 (Figure 8) was captured to observe the characteristics
of its full surface area. Sample 1 was removed from the top of an adult, P. ochraceus arm
infected with sea star wasting disease. The orientation of the tissue on the sample stud is
the top of the organism’s tissue. On the top right and bottom left of Figure 5 are spines
47
from the top of the sea star body. The middle of the sample appears to have many cracks
and crevices, with rounded characteristics, henceforth referred to as nodes in this Results
section.
Figure 8. SEM image of P. ochraceus tissue, sample 1. Image captured at 35
magnification, with the voltage calibrated to 5 and spot size of 44. The Z distance and
working distance were set to 20 mm. Image represents the full aspect of sample 1, the full
diameter is approximately 8 mm. The textured region around the sample is the carbon
disk the sample is seated atop.
The SEM was able to successfully resolve fine detail represented in the P.
ochraceus spine. Figure 9 showcases the ridges around the circumference with great
detail. The spine appears to have a fine texture similar to pockmark-like circular divots
around the surface. A closer look at the spine can be found in Figure 15, where the ridges
are easily resolved and showcases the ridge characteristics around the edges of the
sample.
48
Figure 9. The above image is a spine of sample 1. Image captured at 60 magnification,
with the voltage calibrated to 5 and spot size of 43. The Z distance and working distance
were set to 20, however, the autofocus function automatically reset the working distance
to 11 mm. The resolution for this image was so good with this working distance, the
modification by the autofocus function was not changed to match the Z distance. Final
image captured using auto contrast and brightness with autofocus.
Figures 10 and 11 illustrate the fine features present in tissue structure. The most
apparent characteristics are the cracks and fissures clearly present on the surface of the
sample. Sample 1 was treated in increasing concentrations of ethanol to three treatments
of 100% acetone. Acetone was also used to dry the sample in the Critical Point Dryer,
and it is likely these cracks are a result of acetone treatments and CO2 critical point
drying. These cracks are unlikely to be present as a characteristic of sea star wasting
disease (Erik Thuesen, personal communication).
49
Figure 10. The above image is a mid-section of sample 1 between the two spines captured
in this sample. Image captured at 100X magnification, with the voltage calibrated to 5
and spot size of 46. The Z distance and working distance were set to 20, however, the
autofocus function automatically reset the working distance to 13 mm. Working distance
was not changed to match the Z distance after autofocus modification. Final image
captured using auto contrast and brightness with autofocus.
50
Figure 11. The above image is a larger view of the center of sample 1 between the two
spines. Image captured at 30 magnification, with the voltage calibrated to 5 and spot size
of 44. The Z distance and working distance were set to 20, however, the autofocus
function automatically reset the working distance to 11 mm. The working distance was
not changed to match the Z distance.
Figure 12 demonstrates more of the fissures present in the tissue sample and
better resolves some of the finer cracks. This image also reveals a texture that appears
sponge-like on the surface of the organism. Tissue imaged in Figure 12 reinforces the
idea that the cracking is a result of SEM sample preparation methods and this is most
clearly illustrated in the bottom left region of this micrograph. Smaller holes in the tissue
appear very fine while others are as large as 10 µm across. However, it seems that as the
holes increase in size, they begin to form larger cracks, creating the fissures presented in
Figure 9.
51
Figure 12. The above image was captured of a more flat surface in sample 1. Image
captured at 270 magnification, with the voltage calibrated to 5 and spot size of 43. The Z
distance and working distance were set to 20. Autofocus function automatically reset the
working distance to 12 mm and this modification was not changed. Final image captured
using auto contrast and brightness with autofocus.
Figure 13 is a close-up of one of the rounded nodes seen on the surface of the
tissue sample. These details are most curious because the clumps of spongy matter either
appear to be coming out of the center from this rounded feature or were once a part of the
structure. The texture of this spongy matter is also attached to the side of the top left node
so it is challenging to tell from this sample if the sponge-like matter is decimated sea star
tissue or other material on the surface of the organism. It is challenging to determine if
this sponge-like matter is the same between the two areas or if it is completely different.
Figure 14 is a close-up view of Figure 13’s node on the left side. It exhibits the porosity
of the sea star tissue and presents a finer scale look into the sponge-like material. On the
52
bottom left of Figure 13 is a side-view of a fissure that opened up in the tissue which
occurred during the drying process.
Figure 13. The above image is a region of the spine of sample 1. Image captured at 220
magnification, with the voltage calibrated to 5 and spot size of 41. The Z distance and
working distance were set to 20, however, the autofocus function automatically reset the
working distance to 11 mm and this was accepted as the working distance for this
micrograph. Image was captured by manually blacking out the contrast and brightness
then the brightness was increased until a gray screen was just visible. Contrast was
increased until fine features could be visualized, then autofocus was initiated to capture
this final image.
53
Figure 14. The above image is a close-up image of Figure 10. Image captured at 3500X
magnification, with the voltage calibrated to 5 and spot size of 40. The Z distance and
working distance were set to 20, and working distance autofocused to 12 mm. Image was
calibrated with auto contrast and brightness and autofocus to capture this final image.
54
Figure 15. The above image is a closer look at the ridges in the spine of sample 1. Image
captured at 430 magnification, with the voltage calibrated to 5 and spot size of 43. The Z
distance and working distance were set to 20, however, the autofocus function
automatically reset the working distance to 12 mm and this was accepted as the working
distance for this micrograph. Image was captured by manually blacking out the contrast
and brightness then the brightness was increased until a gray screen was barely visible.
Contrast was increased until fine features could be visualized, then focus was adjusted to
capture this final image.
The features shown in Figure 16 are replicated in different sponge-like regions
throughout Sample 1. They consist of web-like tendrils connected to a center and spread
out across the surface. These were mostly found on the nodes and the sponge-like matter
within Sample 1. They appear to be features either attached to the surface or protruding
from it. It is unclear what they are or if they are even P. ochraceus tissue.
55
Figure 16. Image captured at 7500X magnification, with the voltage calibrated to 5 and
spot size of 39. The Z distance was calibrated to 20 and the working distance autofocused
to 11 mm. Image was captured by manually blacking out the contrast and brightness then
the brightness was increased until a gray screen was just observable. Contrast was
increased until fine features could be visualized.
The final three images, Figure 17, Figure 18, and Figure 19 are close ups of a
sponge-like region captured with increasing magnifications of 13,000X, 70,000X and
300,000X respectively. Increasing the magnification, decreases the crispness of the
images so that less and less fine detail is resolvable. The final image Figure 19, is the
most important for this research project. It was taken at the highest magnification the
SEM could achieve. A couple items are important to note, firstly the visibility of the
image’s contents is not at all sharp enough to resolve fine viral material let alone with
proper accuracy. Next, the maximum measurement the SEM scale bar is set to at
300,000X magnifications is 50 nanometers. The densovirus measures at 20-25
nanometers (Hewson et al. 2014), half the size of the scale bar shown in Figure 16. The
56
SEM is both challenged in its ability to produce crisp and clear images at very high
magnification necessary to identify densovirus. This particular model of SEM at
Evergreen cannot also zoom in far enough to see the incredibly small size of densovirus.
Therefore, this project concludes that the SEM model at Evergreen appears to be
insufficient for detecting densovirus in P. ochraceus due to the smallness of virus’s size.
Figure 17. Image captured at 13,000X magnification, with the voltage calibrated to 5 and
spot size of 38. The Z distance was adjusted to 20 and the working distance autofocused
to 12 mm. Image was captured by manually blacking out the contrast and brightness then
the brightness was increased until a gray screen was just visible Contrast was increased
until fine features could be pictured.
57
Figure 18. The above image is a close up of Figure 14. Image captured at 13,000X
magnification, with the voltage calibrated to 5 and spot size of 37. The Z distance was
calibrated to 20 and the working distance autofocused to 20 mm. Image was captured by
manually blacking out the contrast and brightness then the brightness was increased until
a gray screen was just observable. Contrast was increased until fine features could be
visualized.
58
Figure 19. The above image is a close up of Figure 15. Image captured at 300,000X
magnification, with the voltage calibrated to 5 and spot size of 34. The Z distance was
adjusted to 20 and the working distance to 20 mm. Image was captured by manually
blacking out the contrast and brightness then the brightness was increased until a gray
screen was just observable. Contrast was increased until fine features could be visualized.
This procedure was performed 3 times for this image until the best contrast and
brightness was present. Focus was calibrated manually using fine adjustment tool until
the above image was resolved.
Sample 2
The first image of Sample 2 is a full view which showcases the variety of features
present on the sample (Figure 20). Sample 2 was removed from the top of an adult P.
ochraceus arm using an 8 mm punch biopsy. The sample is mounted on a carbon disk
atop a 10 mm SEM sample stud whose pock-like marks can be seen around the outside of
the sample. Figures 21 and 22 are magnified views of the P. ochraceus spine, which is
located in the middle, at the top of Figure 20. Figure 21 captures clear views of the nodes
present in this sample, to the left of the spine; a characteristic that was also seen in
59
Sample 1.
Sample 2 was also particularly challenging to capture images of, even at lower
magnifications. This sample was most challenging when trying to adjust for brightness
and contrast. The contrast would be too bright and the brightness would cause the sample
to wash out so that no fine features could be visualized. These images of Sample 2 were
therefore captured at very low brightness and moderate contrast. These challenges also
made it difficult to obtain a sharp focus. It was better overall to adjust the spot size to a
higher range to smooth out the surfaces, which made it easier for the SEM to achieve a
quality focus on the sample.
Figure 20. The above image is from diseased P. ochraceus tissue. This is a large view
image of Sample 2. Image captured at 14 magnification, with the voltage calibrated to 5
and spot size of 45. The Z distance was adjusted to 20 and the working distance to 36
mm.
60
Figure 21. The above image is a close up of a spine in Sample 2, pictured at the top of
Figure 17. Image captured at 50 magnification, with the voltage calibrated to 5 and spot
size of 43. The Z distance was adjusted to 20 and the working distance to 18 mm.
61
Figure 22. The above image is an even closer look at the spine of Sample 2. Image was
captured at 300X magnification, with the voltage calibrated to 5 and spot size of 43. The
Z distance was adjusted to 20 and the working distance to 18 mm.
The features present in Figures 23 and 24 illustrate more familiar sample
characteristics. On the surface are cracks and fissures from the SEM sample preparation
drying process. In Figure 24 the same porosity observed in Sample 1 can also be
visualized in this image. The final image captured of Sample 2 (Figure 25), was the
highest magnification The Evergreen State College SEM was capable of achieving. This
image is a highly magnified region of the features on Figure 24. The white region in the
middle of the sample is tissue and the dark regions around it are the cracks which can be
seen Figure 24. This image at 300,000X magnification was not captured at as high a
quality resolution as Figure 19, but illustrates the limitations of the imaging at such a high
magnification.
62
Figure 23. This is an image taken from Sample 2. Image captured at 130 magnification,
with the voltage calibrated to 5 and spot size of 45. The Z distance was adjusted to 20 and
the working distance to 36 mm.
63
Figure 24. Image captured at 300X magnification, with the voltage calibrated to 5 and
spot size of 38. The Z distance was adjusted to 20 and the working distance to 18 mm.
64
Figure 25. The above image demonstrates the SEM at maximum magnification for
Sample 2. Image captured at 300,000X magnification, with the voltage calibrated to 5
and spot size of 28. The Z distance was adjusted to 20 and the working distance to 18
mm.
Sample 3
Sample 3 was taken from the top of an adult P. ochraceus arm. Figure 26 is the
widest view captured of this sample, in large part because the machine was already
calibrated to a Z distance of 20. This was the widest view that could be achieved by the
machine at this distance and captures the overall topography of the sample very well. In
Figures 26 and 27, the nodes in the P. ochraceus tissue are easily observed, however, the
nodes appear less well-defined than those in Samples 1 and 2. This may be due to
damage to the sample either during the drying process or in transit from the field to the
lab. This may also simply be differences in sample features between organisms.
65
Figure 26. The above image is a wide-view picture of Sample 3. Image captured at 25
magnification, with the voltage calibrated to 5 and spot size of 50. The Z distance was
adjusted to 20 and the working distance to 17 mm.
Figure 27 is a magnified image of a region in the left of Figure 26. It illustrates
some of the unique webbing present on many tissue sample images, where the web-like
structures appear to lie on top of the nodes and stretch out between them. It is unclear if
this webbing is a part of the P. ochraceus tissue or if it is an anomalous characteristic
present on the samples. Sample 3 definitely carries the largest quantity of the web-like
matter of all samples scanned under SEM.
66
Figure 27. Image captured at 140 magnification, with the voltage calibrated to 5 and spot
size of 34. The Z distance was adjusted to 20 and the working distance to 18 mm.
The third image, Figure 28, is a close up of a central region pictured in Figure 27.
It is scaled at 160,000X magnification and was calibrated to have very high contrast. This
image demonstrates the challenges with achieving a quality image at this scale range.
This image was useful in visualizing the tissue with the highest contrast before
approaching the final image, Figure 29, at 300,000X magnification. In Figure 29, the
white space on the right is P. ochraceus tissue and the dark space on the left is a gap
around the tissue from the central region of Figure 27. Of the images taken at 300,000X
magnification, Figure 29 is the poorest quality and resolution, however, it is the best high
magnification image captured for Sample 3.
67
Figure 28. Image captured at 160,000X magnification, with the voltage calibrated to 5
and spot size of 40. The Z distance was adjusted to 20 and the working distance to 18
mm.
68
Figure 29. The above image is sample 3 under the SEM at maximum magnification.
Image captured at 300,000X magnification, with the voltage calibrated to 5 and spot size
of 35. The Z distance was adjusted to 20 and the working distance to 18 mm.
Sample 3.5
A second tissue sample was taken from the third P. ochraceus sampled for this
project and is thus called Sample 3.5. The tissue was removed not from the wasted region
at the top like Sample 3, but from the underside of the organism. An 8 mm punch biopsy
was removed from the tube feet of the third P. ochraceus; these tube feet were not yet
necrotic and still appeared to possess mobility. Figure 30 is a wide-view of Sample 3.5.
The tube feet protrude from the base of the sample and curve at an angle, down and to the
left. Half of the tube feet are mostly in-tact from the top to the tube feet pictured in the
middle in the left. On the right, the tube feet have been slightly damaged either from the
punch biopsy, transportation to the lab or during sample preparation.
69
Figure 30. This figure is representative of a second sample taken from the underside of
the third P. ochraceus sampled for this research. Pictured are the desiccated tube feet
from a diseased individual. Image captured at 17 magnification, with the voltage
calibrated to 5 and spot size of 45. The Z distance was adjusted to 20 and the working
distance to 36 mm.
The nodes which have been visible in Samples 1, 2, and 3, are also present around
the base of the tube feet but with longer shape variations. Figure 31 is a closer view of the
tube foot at the bottom of Figure 30. This tube foot is interesting because it was sliced in
half during biopsy and illustrates a dissected P. ochraceus tube foot. The center of this
feature appears to have a sponge-like quality a mass of connective tissues riddled with
small holes.
70
Figure 31. This micrograph demonstrates a tube foot of Sample 3.5 which has been sliced
in half. Image captured at 80 magnification, with the voltage calibrated to 5 and spot size
of 32. The Z distance was adjusted to 20 and the working distance to 16 mm.
The external texture of the P. ochraceus tube feet are evident in Figures 31 and
32. The outside, resembles desert cracks with pieces of the tissue almost lifting off of the
surface in a flaking or peeling manner. Figure 33 is a close-up view of tissue between
two tube feet imaged in the upper right corner of Figure 30. These features resemble the
nodes present in Samples 1-3. However, they appear elongated with a symmetrical
arrangement around a center which is not seen in this image but looks to begin at the base
of the tube feet.
71
Figure 32. The above image is an intact but desiccated tube foot. Image captured at 100X
magnification, with the voltage calibrated to 5 and spot size of 32. The Z distance was
adjusted to 20 and the working distance to 18 mm.
A close-up view of Figure 32 can be seen in Figures 34 and 35, where the region
in the bottom center of Figure 32 is magnified to reveal the nature of the white matter
protruding between the tube feet. Upon closer inspection, the white material in Figure 34,
appears to be tattered or shredded looking matter lying on top of the tissue. Within this
tattered material, are small rounded ovals pictured in Figure 35. These rounded features
do not appear anywhere but the tattered material and give the impression like they’re
nested within it.
72
Figure 33. This image is a close up of a region between two tube feet. Image captured at
100X magnification, with the voltage calibrated to 5 and spot size of 18. The Z distance
was adjusted to 20 and the working distance to 16 mm.
73
Figure 34. Pictured above is a region of the tube feet. Image captured at 1,400X
magnification, with the voltage calibrated to 5 and spot size of 28. The Z distance was
adjusted to 20 and the working distance to 15 mm.
74
Figure 35. This picture is a close up of the central region of Figure 31. The rounded, oval
shapes are characteristically similar in size and feature to bacteria. This image depicts the
only example of bacteria-like structures present on any of the samples imaged for this
project. However, their legitimacy as bacteria cannot be confirmed without genetic tests.
Image captured at 5,500X magnification, with the voltage calibrated to 5 and spot size of
28. The Z distance was adjusted to 20 and the working distance to 16 mm.
The final image of Sample 3.5 (Figure 36) is a high magnification view of Figure
35. The light, white regions are the P. ochraceus tissue and the dark regions are the
spaces and gaps between the sample. This image is one of the better images captured at
300,000X magnification because it is at least able to resolve some of the variability and
shapes in the tissue. This image was challenging to resolve because Sample 3.5
experienced points of high light which is best illustrated in Figure 35 behind the tattered
materials and the right-most oval where the light is particularly bright. This is due in
large part to SEM electron scatter and the microscope was challenged by its ability to
hone in on a region and remain still. The scatter shifted the background of Figures 35 and
75
36 quite a lot and so capturing these photos involved a great deal of image tinkering and
fine adjustment to the wobbler to reduce some of this image shifting.
Figure 36. This micrograph is an image of Sample 3.5 at the SEM’s highest
magnification. Image captured at 300,000X magnification, with the voltage calibrated to
5 and spot size of 10. The Z distance was adjusted to 20 and the working distance to 16
mm.
Sample 4
The first image of Sample 4 (Figure 37) is a wide-view of the fourth P. ochraceus
sampled for this project. The tissue removed for this sample was necrotic and taken from
the top of an adult sea star arm. In the top left region of Figure 37, is another large,
rounded spine. Around the spine are more of the nodes which have been seen in the
previous samples. The nodes in Sample 4 do not appear to be as pronounced or abundant
as those seen in Samples 1, 2, and 3, but are instead spread out across the sample, being
flat in topography.
76
Figure 37. This image is a wide-view of Sample 4. Image captured at 12 magnification,
with the voltage calibrated to 5 and spot size of 45. The Z distance was adjusted to 40 and
the working distance to 49 mm.
Figure 38 is a closer view of a bottom right protrusion in Figure 37. It seems too
be either another rounded protruding feature different than the spines, or may be the early
stage of a spine growing from the tissue. To understand this structure, another image was
taken of a magnified region of Figure 38 to observe the smaller spines on top of the
sample. Figure 39 shows a smaller protrusion rising up from the tissue, ending in a fine
point. It may be that this is a microscopic spine on top of an early forming spine. The
texture of the spines appear to be porous and cracking, similar to the previous samples
observed.
77
Figure 38. Image captured at 50X magnification, with the voltage calibrated to 5 and spot
size of 45. The Z distance was adjusted to 20 and the working distance to 17 mm.
78
Figure 39. The above image is a close-up of a small spine on Figure 35. Image captured
at 2,700X magnification, with the voltage calibrated to 5 and spot size of 45. The Z
distance was adjusted to 20 and the working distance to 18 mm.
The final image (Figure 40) is a micrograph of Sample 4 at the highest
magnification the SEM is capable of achieving, 300,000X magnification. This image has
a smoother texture, especially when compared to Figure 36 because the spot size to
capture this image was calibrated to 39. This smooths out the sample features and makes
the image appear less grainy. This image further reinforces the challenges with resolving
P. ochraceus tissue and viral material that may be present, at high magnifications with
the SEM.
79
Figure 40. This micrograph demonstrates Sample 4 at the highest magnification
achievable by the SEM. Image captured at 300,000X magnification, with the voltage
calibrated to 5 and spot size of 39. The Z distance was adjusted to 20 and the working
distance to 19 mm.
Sample 4.5
Sample 4.5 was taken from the same organism as Sample 4. The tissue was
removed from the underside of an adult P. ochraceus, capturing tube feet of the sea star.
The tissue on the underside of the P. ochraceus did not exhibit signs of wasting disease at
the time of removal. The tube feet appear arranged in a similar fashion to those in Sample
3.5. The tissue in Figure 41 appears to have sustained slightly more damage either during
biopsy, in transit or through sample preparation than Sample 3.5. The light spaces visible
in Figure 41 are potentially a product of the SEM’s poor temperament during imaging. A
closer view of the tube feet is visible in Figure 42. The same porosity and cracking
present in Sample 3.5 is also present in this image, though these tube feet to appear to
have fewer smooth surfaces.
80
Figure 41. The above image is a wide-view of Sample 4.5. Image captured at 14X
magnification, with the voltage calibrated to 5 and spot size of 45. The Z distance was
adjusted to 20 and the working distance to 45 mm.
81
Figure 42. The above image is a close-up of tube feet imaged in Figure 41. Image
captured at 37X magnification, with the voltage calibrated to 5 and spot size of 45. The Z
distance was adjusted to 20 and the working distance to 16 mm.
Figure 43 demonstrates more of the odd nodes present at the base of the tube feet
where the nodes appear to have a center in which they radiate outwards from ending in
rounded tips with harsh lines down the center. It is possible these features are present at
the base of P. ochraceus tube feet as a species. However, too few images were captured
of these features to be certain they are a physiological characteristic on the underside of
all P. ochraceus.
82
Figure 43. The above image illustrates a region at the base of the tube feet in Sample 4.5.
Image captured at 45X magnification, with the voltage calibrated to 5 and spot size of 45.
The Z distance was adjusted to 20 and the working distance to 16 mm.
Sample 4.5 was definitely the most challenging to image. Many times, the
scanning area appeared much too bright with light hot spots that prevented much of the
surface details from being visible. This seemed to come from issues with the SEM which
was acting up when the microscope tried to zoom into magnifications greater than
5,000X. Consequently, the highest magnification this Sample 4.5 was imaged at was
190,000X. On the right and left sides of Figure 44, are white spaces with a large dark
streak down the center. The lighter spaces are P. ochraceus tissue and the dark region in
the center represents the gaps in between tissues. The image has a blurry texture
characteristic of imaging with electron drift, where the electron beam experiences
challenges focusing in on the samples surface and drags the image across the computer
83
monitor. The image was not stationary as it was being captured and therefore, appears out
of focus and mobile.
Figure 44. The above image is the highest magnification that could be achieved with
Sample 4.5 due to electron drift. Image captured at 190,000X magnification, with the
voltage calibrated to 5 and spot size of 45. The Z distance was adjusted to 20 and the
working distance to 16 mm.
Sample 5 – Healthy Tissue
A single tissue sample was captured from a healthy, adult P. ochraceus to observe
the surface characteristics of tissue uninfected with sea star wasting disease. Figure 45
captures the entire surface of the 8mm diameter tissue sample. The tissue was removed
from the top of a P. ochraceus arm. Figure 45 illustrates the arrangement of sea star
spines with nodes dispersed in between them. Sample 5 appears to have less of the
sponge-like matter which was present on the top of the sea star spines infected with
84
wasting disease. The nodes appear fairly intact and snugly fit around the base of each
spine.
Figure 45. The above image is a wide-view of Sample 5 which is tissue taken from a
healthy adult P. ochraceus. Image captured at 14X magnification, with the voltage
calibrated to 5 and spot size of 45. The Z distance was adjusted to 20 and the working
distance to 35 mm.
A closer view of the spines is represented in Figure 46. These spines were
arranged on the left side of the tissue imaged in Figure 45 and protrude away from the
center. This image demonstrates P. ochraceus spines at different levels of development
on the top of the body. The left most spine is the largest, with fully developed ridges. The
center spine is the smallest and carries visible bumps but no ridges around the outside of
the spine. The right most spine is in what appears to be a mid-level developmental stage
where the smaller bumps are present on the top and the spine is now beginning to form
85
ridges around the edges as it expands. Nodes are inset in between the spines in this
micrograph.
Figure 46. The above image is a closer view of spines and nodes on Sample 5. Image
captured at 50X magnification, with the voltage calibrated to 5 and spot size of 42. The Z
distance was adjusted to 20 and the working distance to 14 mm.
The nodes in Sample 5 are the most curious feature because they appear very
smooth, unlike the nodes present in all necrotic tissue which is usually covered by the
sponge-like matter. The nodes in Figure 47 are plump with a soft rolling texture that
exhibits the same porosity as all other samples, from the drying process. Nodes appear to
wrap around spines and lay against each other in a woven pattern. This structure was not
seen in the tissues of diseased individuals due to the large amounts of background noise
present in each sample.
86
Figure 47. The above image demonstrates the texture of the nodes in Sample 5. Image
captured at 270X magnification, with the voltage calibrated to 5 and spot size of 36. The
Z distance was adjusted to 20 and the working distance to 14 mm.
The final image of Sample 5 was captured at the highest magnification capable by
The Evergreen State College SEM. This image demonstrates a higher quality resolution
than most of the images captured at 300,000X magnification. The lighter gray regions are
P. ochraceus tissue and the dark gray to black areas are space in between the tissue.
Figure 48 demonstrates the closest image of P. ochraceus tissue porosity as a result of the
drying process necessary for SEM sample preparation.
87
Figure 48. The above image is Sample 5 captured at the highest magnification the SEM
was capable of achieving. Image captured at 300,000X magnification, with the voltage
calibrated to 5 and spot size of 42. The Z distance was adjusted to 20 and the working
distance to 14 mm.
Conclusive Results
Images captured of P. ochraceus tissue at the highest magnification achievable by
The Evergreen State College’s SEM were not successful at producing high resolution
images. Fine features in Figures 19, 25, 29, 36, 40, and 48 were not resolvable at the
300,000X magnification. The scale bar at this highest magnification was 50 nm, while the
diameter of the densovirus this project attempted to detect were 20-25 nm; half the
distance of this scale bar range. It is clear from the images captured at 300,000X
magnification that SEM is not strong enough to resolve potential features of densovirus
in P. ochraceus.
88
This project also found a lack of abundant or accurately discernable bacterial
features. Four of the 5 tissue samples analyzed under the SEM, were removed from
necrotic P. ochraceus tissue infected with sea star wasting disease. Of these samples,
only Sample 4 in a small region pictured in Figure 34, had any discernable rounded
bacteria-like structures. However, the ovals may be features contaminating the sample
and are not conclusively bacterial in origin (Erik Thuesen, personal communication).
Thus, bacteria were not definitively present on necrotic P. ochraceus tissue samples
analyzed for this project.
In conjunction with the viral and bacterial findings from the SEM analysis, this
project also identified successful methods for P. ochraceus soft tissue preparation for
SEM. One of the methods implemented for this research, involved comparing the
difference between P. ochraceus tissue sample drying using ethanol versus ethanol and
acetone. This project found that samples were fully dried after having been submerged in
increasing ethanol concentrations and 3 treatments of 100% acetone. The single
experimental drying trial used ethanol exclusively in increasing concentrations and
included ethanol in the CO2 Critical Point Dryer for the final stage of the sample drying
process. Upon critical point drying completion, the sample was removed, still wet with
ethanol. The ethanol, by itself, was not capable of completely drying the P. ochraceus
tissue. The Critical Point Dryer was not able to fully bleed off the ethanol and the tissue
was warped from this final stage of drying. Therefore, this project found that acetone is a
necessary component for effective P. ochraceus soft tissue SEM sample preparation.
89
90
Discussion
This project attempted to detect the densovirus in P. ochraceus tissue infected
with sea star wasting disease. The final method implemented, which prepared whole
necrotic P. ochraceus tissue found that the SEM at The Evergreen State College (JEOL
JSM-6480LV) was not strong enough to detect the virus with a diameter of 20-25 nm.
Although, magnification has been a critical reference point for this project’s analysis, it is
only one component of SEM. The most important aspect of quality SEM micrograph
imaging, lies in the challenge of attaining high resolution. Resolution deals with the
ability to visualize features within an image. This project found it difficult to achieve a
high-resolution image above 25,000X magnification.
Resolution is a component of SEM imaging that is essential for achieving quality
micrographs. Scanning at high magnification, places stress on the user’s ability to also
obtain high quality resolution. For successful densovirus identification, it was necessary
to achieve both high magnification and quality resolution so that viral characteristics
could be visualized and used as a tool for identification. At the magnification necessary
to visualize densovirus (300,000X), excellence resolution is incredibly challenging to
achieve, especially for a new user. A more advanced user with an abundance of
experience scanning at high magnification, may have more skills in their ability to
generate high quality micrographs at high magnification.
The other challenge this research encountered was that the SEM at Evergreen was
temperamental; the machine was broken for two months and repaired half-way through
the data collection process for this project. This presented a new set of challenges, since
Evergreen’s SEM had already experienced technical difficulties prior to this episode.
91
Issues such as electron drift were frequent, especially at high magnification, where
electrons were unstable and the electron beam was unable to focus on a specified region
causing the image to shift and blur. It is possible that an advanced SEM user with a more
reliable microscope could achieve higher resolution images at very high magnification.
However, there is little evidence in the literature to suggest that high quality resolution at
300,000X magnification is regularly accomplished (Wells 1974). Thus, this thesis does
not recommend this project’s methods, as an avenue to resolve densovirus in P.
ochraceus infected with sea star wasting disease.
One method this project had originally been inspired to attempt was to use ionic
suspension fluid in the SEM sample chamber for analysis. This was a method utilized by
Golding et al. (2016) which was effective at resolving quality images of Ebola virus
compared to traditional methods of drying and sputter coating. An issue Golding et al.
(2016) reconciles in their research is the sample coverage sputter coating imposes to
create conductivity in the sample. Sputter coating can coat fine details such as viruses
that may be present in the sample and reduce the resolvability of microscopic structures.
Ionic suspension fluid was a method Golding et al. (2016) applied to avoid these
challenges with virus resolution under SEM. However, this project was not able to
include ionic suspension fluid in this method development because the solution is too
dangerous to use. Ionic suspension fluid has the unfortunate quality of degrading into
hydrofluoric acid, an extremely toxic substance harmful to humans. Handling this
solution would require full personal protective equipment.
The primary concern with using ionic suspension fluid, was the already
temperamental nature of The Evergreen State College’s SEM. The microscope had
92
experienced mechanical challenges in recent years; it was decided that this project would
not attempt ionic suspension fluid out of fear it could damage the machine. In the future,
it would be interesting to see if a virus as small as densovirus, could be successfully
resolved using ionic suspension fluid in a SEM.
A brief and limited comparison of necrotic P. ochraceus tissue to a single healthy
tissue sample was executed by this project. Major deductions cannot be made from these
few images, but the distinct lack of sponge-like matter in Sample 5 is highly intriguing.
Most of the wasted P. ochraceus tissue samples which exhibited definitive signs of sea
star wasting disease in the field, appeared to have large masses on top of the nodes which
lie in between the sea star spines. In the healthy P. ochraceus tissue, only nodes were
present and the tissue appeared smoother in Sample 5 than in Samples 1, 2, 3, or 4. It is
challenging to know if these differences in P. ochraceus tissue characteristics are a
product of sea star wasting disease. With only visual observations, there does appear to
be a perceptible difference between tissues collected from diseased individuals and the
single healthy P. ochraceus tissue sample. More research and images of the physical
changes P. ochraceus tissue experiences at the microscopic level from wasting disease is
needed. Visualizing tissue decay and comparing it further to healthy organisms, may
elucidate the nature of diseased tissues and help researchers understand what wasting
looks like and how it behaves microscopically.
Although this project was unsuccessful at developing a new method for viral
identification in P. ochraceus, more research into viral host interactions in sea stars and
echinoderms as a whole, is crucial. At present, our knowledge of existing viral-host
relationships in the rocky intertidal zone is severely limited. The sea star-associated
93
densovirus (SSaDV) was the first pathogen to exhibit a strong correlation with sea star
wasting disease (Hewson et al. 2014). However, the inadequate baseline knowledge of
viruses in echinoderms, undermines SSaDV as a conclusive disease-causing pathogen,
raising more questions than answers as to the source behind sea star wasting disease.
The ambiguity of sea star wasting, and the nature of viral detection in asteroids,
makes studying this phenomena complicated and frustrating. There still exists a need to
understand the mass wasting event from 2013-2015, how it impacted Northeast Pacific
asteroid populations, and how it will continue to affect the coastlines. The future of sea
star wasting disease is uncertain with its continued presence in sea stars and an
incomplete knowledge of its cause. For a time, the long-term influences of this disease on
P. ochraceus was also unclear as their populations experienced dramatic decreases
especially during the sea star wasting event. Today, populations of P. ochraceus do
appear to be rebounding on the coastline, reclaiming their place in intertidal ecosystems
as keystone predators (Moritsch & Raimondi 2018, Miner et al 2018). With time, there is
hope P. ochraceus can create the same predation pressure on mussels, and maintain
intertidal species biodiversity, as they did prior to the 2013 sea star wasting outbreak.
94
Conclusion
This project set out to develop an accessible method which would allow others the
opportunity to contribute to sea star wasting disease research. Originally, it was thought
that the key to answering this question would lie in developing a successful filtration
method that would simultaneously isolate densovirus from the host tissue and prepare the
sample for SEM analysis. The challenges faced in developing a successful filtration
method were unanticipated. However, failure did not deter this project which turned to
preserving and preparing whole P. ochraceus tissue for SEM analysis. The images
captured of necrotic and healthy P. ochraceus tissue were examined for this research.
None of the images demonstrated incontrovertible presence of bacterial assemblages,
further reinforcing that sea star wasting disease is not caused by a bacterial pathogen. At
the highest magnification of 300,000X, the SEM was not capable of producing a high
magnification image at a high-quality resolution. This project concludes that the methods
implemented for this research were insufficient to identify densovirus in P. ochraceus
infected with sea star wasting disease.
95
96
Bibliography
Bates, A.E., B.J. Hilton, C.D.G. Harley, 2009, Effects of Temperature, season and
locality on wasting disease in the keystone predatory sea star Pisaster ochraceus.
Diseases of Aquatic Organisms, Vol. 86, pp245-251.
Becker, P., Gillan, D., Lanterbecq, D., Jangoux, M., Rasolofonirina, R., Rakotovao, J.,
and Eeckhaut, I, 2004. The skin ulceration disease in cultivated juveniles of
Holothuria scabra (Holothuroidea, Echinodermata). Aquaculture 242: 13-30
Bochow, S., Condon, K., Elliman, J., & Owens, L. (2015). First complete genome of an
Ambidensovirus; Cherax quadricarinatus densovirus, from freshwater crayfish
Cherax quadricarinatus. Marine genomics, 24, 305-312.
Bonaviri, C., Graham, M., Gianguzza, P., & Shears, N. T. (2017). Warmer temperatures
reduce the influence of an important keystone predator. Journal of Animal
Ecology.
Bruemmer, A., Scholari, F., Lopez-Ferber, M., Conway, J. F., & Hewat, E. A. (2005).
Structure of an insect parvovirus (Junonia coenia Densovirus) determined by
cryo-electron microscopy. Journal of molecular biology, 347(4), 791-801.
97
Bucci, C., Francoeur, M., McGreal, J., Smolowitz, R., Zazueta-Novoa, V., Wessel, G.
M., & Gomez-Chiarri, M. (2017). Sea Star Wasting Disease in Asterias forbesi
along the Atlantic Coast of North America. PloS one, 12(12), e0188523.
Buck, J. C., & Ripple, W. J. (2017). Infectious Agents Trigger Trophic Cascades. Trends
in Ecology & Evolution, 32(9), 681-694.
Chandler, V. K., & Wares, J. P. (2017). RNA expression and disease tolerance are
associated with a “keystone mutation” in the ochre sea star Pisaster ochraceus.
PeerJ, 5, e3696.
Cotmore, S. F., Agbandje-McKenna, M., Chiorini, J. A., Mukha, D. V., Pintel, D. J., Qiu,
J., ... & Davison, A. J. (2014). The family parvoviridae. Archives of virology,
159(5), 1239-1247.
Eisenlord, M. E., Groner, M. L., Yoshioka, R. M., Elliott, J., Maynard, J., Fradkin, S., ...
& Harvell, C. D. (2016). Ochre star mortality during the 2014 wasting disease
epizootic: role of population size structure and temperature. Phil. Trans. R. Soc.
B, 371(1689), 20150212.
Evans, A. S. (1976). Causation and disease: the Henle-Koch postulates revisited. The
Yale journal of biology and medicine, 49(2), 175.
98
Fahsbender, E., Hewson, I., Rosario, K., Tuttle, A. D., Varsani, A., & Breitbart, M.
(2015). Discovery of a novel circular DNA virus in the Forbes sea star, Asterias
forbesi. Archives of virology, 160(9), 2349-2351.
François, S., Filloux, D., Roumagnac, P., Bigot, D., Gayral, P., Martin, D. P., ... &
Ogliastro, M. (2016). Discovery of parvovirus-related sequences in an
unexpected broad range of animals. Scientific Reports, 6.
François, S., Bernardo, P., Filloux, D., Roumagnac, P., Yaverkovski, N., Froissart, R., &
Ogliastro, M. (2014). A novel itera-like densovirus isolated by viral
metagenomics from the sea barley Hordeum marinum. Genome announcements,
2(6), e01196-14.
Fuess, L. E., Eisenlord, M. E., Closek, C. J., Tracy, A. M., Mauntz, R., GignouxWolfsohn, S., ... & Friedman, C. S. (2015). Up in arms: immune and nervous
system response to sea star wasting disease. PLoS One, 10(7), e0133053.
Gale, A. S. (1987). Phylogeny and classification of the Asteroidea (Echinodermata).
Zoological Journal of the Linnean Society, 89(2), 107-132.
Golding, C. G., Lamboo, L. L., Beniac, D. R., & Booth, T. F. (2016). The scanning
electron microscope in microbiology and diagnosis of infectious disease.
Scientific reports, 6, 26516.
99
Gudenkauf, B. M., & Hewson, I. (2016). Comparative metagenomics of viral
assemblages inhabiting four phyla of marine invertebrates. Frontiers in Marine
Science, 3, 23.
Gudenkauf, B. M., Eaglesham, J. B., Aragundi, W. M., & Hewson, I. (2014). Discovery
of urchin associated densoviruses (family Parvoviridae) in coastal waters of the
Big Island, Hawaii. Journal of General Virology, 95(3), 652-658.
Gudenkauf, B. M., & Hewson, I. (2015). Metatranscriptomic analysis of Pycnopodia
helianthoides (Asteroidea) affected by sea star wasting disease. PloS one, 10(5),
e0128150.
Hajibagheri, M. N. (Ed.). (1999). Electron Microscopy: Methods and Protocols. Springer
Science & Business Media.
Hayat, M.A. (1974). Principles and Techniques of Scanning Electron Microscopy. Van
Nostrand Reinhold Company.
Heywood, V.H. (1971). Scanning Electron Microscopy. London: Academic Press.
100
Hewson, I., Button, J. B., Gudenkauf, B. M., Miner, B., Newton, A. L., Gaydos, J. K., ...
& Fradkin, S. (2014). Densovirus associated with sea-star wasting disease and
mass mortality. Proceedings of the National Academy of Sciences, 111(48),
17278-17283.
Jackson, E. W. (2016). The diversity and community composition of microorganisms
associated with echinoderms (Doctoral dissertation, Cornell University).
Jackson, E. W., Bistolas, K. S., Button, J. B., & Hewson, I. (2016). Novel Circular Single
Stranded DNA Viruses among an Asteroid, Echinoid and Holothurian (Phylum:
Echinodermata). PloS one, 11(11), e0166093.
Kang, Y. J., Huang, W., Zhao, A. L., Lai, D. D., Shao, L., Shen, Y. Q., ... & Zhang, W.
(2017). Densoviruses in oyster Crassostrea ariakensis. Archives of Virology, 1-5.
Kessel, R. G., & Shih, C. Y. (2012). Scanning Electron Microscopy in BIOLOGY: A
Students’ Atlas on Biological Organization. Springer Science & Business Media.
Kohl, W. T., McClure, T. I., & Miner, B. G. (2016). Decreased temperature facilitates
short-term sea star wasting disease survival in the keystone intertidal sea star
Pisaster ochraceus. PloS one, 11(4), e0153670.
Lawrence, J. M. (2013). Starfish: biology and ecology of the Asteroidea. JHU Press.
101
Menge, B. A., Cerny-Chipman, E. B., Johnson, A., Sullivan, J., Gravem, S., & Chan, F.
(2016) Sea star wasting disease in the keystone predator Pisaster ochraceus in
Oregon: insights into differential population impacts, recovery, predation rate, and
temperature effects
from long-term research. PloS one, 11(5), e0153994.
Miner, C. M., Burnaford, J. L., Ambrose, R. F., Antrim, L., Bohlmann, H., Blanchette, C.
A., ... & Miner, B. G. (2018). Large-scale impacts of sea star wasting disease
(SSWD) on intertidal sea stars and implications for recovery. PloS one, 13(3),
e0192870.
Montecino-Latorre, D., Eisenlord, M. E., Turner, M., Yoshioka, R., Harvell, C. D.,
Pattengill Semmens, C. V., ... & Gaydos, J. K. (2016). Devastating
Transboundary Impacts of Sea Star Wasting Disease on Subtidal Asteroids. PloS
one, 11(10), e0163190.
Moritsch, M. M., & Raimondi, P. T. (2018). Reduction and recovery of keystone
predation pressure after disease‐related mass mortality. Ecology and Evolution.
Mothersill, C., & Austin, B. (2000). Aquatic invertebrate cell culture. Berlin: Springer.
Paine, R. T. (1969). The Pisaster‐Tegula Interaction: Prey Patches, Predator Food
Preference, and Intertidal Community Structure. Ecology, 50(6), 950-961.
102
Paine, R.T. (1995). "A Conversation on Refining the Concept of Keystone Species".
Conservation Biology. 9 (4): 962–964.
Pankey, M. S., & Wares, J. P. (2009). Overdominant maintenance of diversity in the sea
star Pisaster ochraceus. Journal of evolutionary biology, 22(1), 80-87.
Scanning Electron Microscope A To Z. (2009). JEOL Ltd. http://www.jeol.com/.
Smith, A. (2017). The Impact of the Mass Mortality of Ochre Stars (Pisaster ochraceus)
and Sunflower Stars (Pycnopodia helianthoides) on Intertidal and Subtidal
Communities in Puget Sound (Doctoral dissertation).
Staehli, Annette, Schaerer, Rolf, Hoelzle, Katharina, Ribi, Georrg, 2009. Temperature
induced disease in the starfish Astropecten jonstoni. Marine Diversity Records 2:
1-5.
Suttle, C. A. (2005). Viruses in the sea. Nature, 437(7057), 356-361. van Oers, M. M.,
Bateman, K. S., & Stentiford, G. D. (2017). Viruses of invertebrates related to the
food chain.
Tijssen, P., Pénzes, J. J., Yu, Q., Pham, H. T., & Bergoin, M. (2016). Diversity of small,
single stranded DNA viruses of invertebrates and their chaotic evolutionary past.
Journal of invertebrate pathology, 140, 83-96.
103
Wells, O.C. (1974). Scanning Electron Microscopy. New York: McGraw-Hill Book
Company.
Xie, R., Chu, L. Y., Chen, W. M., Xiao, W., Wang, H. D., & Qu, J. B. (2005).
Characterization of microstructure of poly (N-isopropylacrylamide)-grafted
polycarbonate track-etched membranes prepared by plasma-graft pore-filling
polymerization. Journal of Membrane Science, 258(1-2), 157-166.
Yang, W. T., Shi, S. H., Jiang, Y. L., Zhao, L., Chen, H. L., Huang, K. Y., ... & Wang, C.
F. (2016). Genetic characterization of a densovirus isolated from great tit (Parus
major) in China. Infection, Genetics and Evolution, 41, 107-112.
104
Appendices
Appendix A
Sample 1
Figure 49. Necrotic P. ochraceus tissue captured at 950X magnification. Voltage is 5kv,
spot size of 42, Z distance is 20 and working distance is 12 mm.
105
Figure 50. Necrotic P. ochraceus tissue captured at 1,400X magnification. Voltage is
5kv, spot size of 41, Z distance is 20 and working distance is 15 mm.
106
Figure 51. Necrotic P. ochraceus tissue captured at 2,300X magnification. Voltage is
5kv, spot size of 42, Z distance is 20 and working distance is 12 mm.
Figure 52. Necrotic P. ochraceus tissue captured at 2,500X magnification. Voltage is
5kv, spot size of 42, Z distance is 20 and working distance is 20 mm.
107
Sample 3
Figure 53. Tissue from the tube feet of Sample 3, captured at 150X magnification.
Voltage is 5kv, spot size of 50, Z distance is 20 and working distance is 18 mm.
108
Figure 54. P. ochraceus tissue from the tube feet of Sample 3 captured at 1,200X
magnification. Voltage is 5kv, spot size of 40, Z distance is 20 and working distance is 18
mm.
109
Sample 3.5
Figure 55. Tissue from the tube feet of Sample 3 captured at 550X magnification.
Voltage is 5kv, spot size of 45, Z distance is 20 and working distance is 15 mm.
110
Figure 56. P. ochraceus tissue from the underside of the organism captured at 4,000X
magnification. Voltage is 5kv, spot size of 41, Z distance is 20 and working distance is 15
mm.
111
Sample 4
Figure 57. Necrotic P. ochraceus tissue captured at 40X magnification. Voltage is 5kv,
spot size of 45, Z distance is 20 and working distance is 18 mm.
112
Figure 58. Necrotic P. ochraceus tissue captured at 100X magnification. Voltage is 5kv,
spot size of 45, Z distance is 20 and working distance is 17 mm.
Figure 59. Necrotic P. ochraceus tissue captured at 200X magnification. Voltage is 5kv,
spot size of 45, Z distance is 20 and working distance is 19 mm.
113
Figure 60. Necrotic P. ochraceus tissue captured at 1,000X magnification. Voltage is
5kv, spot size of 45, Z distance is 20 and working distance is 19 mm.
114
Figure 61. Necrotic P. ochraceus tissue captured at 95,000X magnification. Voltage is
5kv, spot size of 45, Z distance is 20 and working distance is 19 mm.
115
Sample 4.5
Figure 62. Necrotic P. ochraceus tissue captured at 200X magnification. Voltage is 5kv,
spot size of 45, Z distance is 20 and working distance is 16 mm.
116
Figure 63. Necrotic P. ochraceus tissue captured at 1,400X magnification. Voltage is
5kv, spot size of 45, Z distance is 20 and working distance is 16 mm.
117
Sample 5—Healthy Tissue
Figure 64. Healthy P. ochraceus tissue captured at 5,000X magnification. Voltage is 5kv,
spot size of 39, Z distance is 20 and working distance is 14 mm.
118
Figure 65. Healthy P. ochraceus tissue captured at 60X magnification. Voltage is 5kv,
spot size of 39, Z distance is 20 and working distance is 14 mm.
119
Azar_Thesis_2018.pdf