ABSTRACT
OTEN, KELLY LYNN FELDERHOFF. Host-Plant Selection by the Hemlock Woolly Adelgid, Adelges tsugae Annand: Sensory Systems and Feeding Behavior in Relation to Physical and Chemical Host-Plant Characteristics. (Under the direction of Dr. Fred P. Hain).
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an invasive insect causing extensive mortality to hemlocks in the eastern
United States. It has become increasingly important to understand the insect-plant interactions of this system for management and eventual hemlock restoration. Insect-plant interactions were observed using light, scanning and transmission electron microscopy.
HWA may exploit alternative stylet insertion sites during high infestation (typical insertion site is between the pulvinus and the stem, below the abscission layer). Five dentitions occur at the tip of the mandibular stylets. Stylet bundle cross-section reveal separate salivary (0.24
- 0.54 µm in diameter, n=11) and food canals (0.48 - 1.0 µm in diameter, n=11), typical of
Hemiptera, and single stylet innervation. Sensilla of the labium appear mechanosensory and an antennal sensorium is present, indicating that morphology and chemical host characteristics may play roles in host-plant acceptance. Tarsal setae are likely used in adhesion to surfaces. Feeding biology was further investigated in an enzymatic survey. We detected the presence of trypsin-like protease, amylase, peroxidase, and polyphenol oxidase.
HWA had four times lower protease activity (0.259 BAEE U µg protein-1 min-1) and eight times lower amylase-like activity (0.1088 mU protein-1 min-1) than L. lineolaris, the positive control. The presence of protease and amylase suggests that, if injected into the plant, it could be used for digestion of insoluble plant proteins and starches. HWA had twice as much peroxidase (0.2323 mU µg protein-1 min-1) and three times as much polyphenol oxidase activity (0.0569 abs µg protein-1 min-1) compared to L. lineolaris, signifying the possibility of plant defense detoxification.
Hemlocks were scrutinized for differences in morphology and chemical composition of the epicuticle. Comparisons were made between species using low temperature-scanning electron microscopy and related to known HWA resistance. Trichomes likely do not relate to host-plant acceptance by HWA, as species that lack and are covered with trichomes are both colonized. Cuticle thickness is significantly thinner at the stylet insertion point than at other locations of the pulvinus. A thinner cuticle may facilitate tissue penetration, potentially the reason for specificity in insertion site. In addition, cuticle thickness at this point is significantly different between species, with the hybrid, Carolina, southern Japanese, and northern Japanese hemlocks possessing the thickest cuticles; therefore, cuticle thickness may confer some resistance to HWA. Wax structures were generally smoother in eastern and
Carolina hemlocks, hypothesized to facilitate sheath adhesion to the plant surface. Because there are many factors that contribute to wax morphology, this research should be pursued further. Extractable lipids of the hemlock surface were analyzed with gas chromatography- mass spectrometry. Results indicate inter- and intraspecific variation. In addition, an unidentified compound appears in the resistant species. This compound should be further scrutinized, both in its identification and its association to HWA behavior.
Interspecific variation in resistance among eastern, Carolina, and western hemlock was assessed in the field. Two of five blocks were artificially infested, two blocks were untreated, and one block was chemically treated. A binary assessment of HWA infestation revealed that eastern hemlock was 36 times more likely and Carolina hemlock was half as likely to become infested as western hemlock. Infestation rates declined significantly in subsequent years as a result of winter temperatures.
Cutting success was also studied at the species level and propagation methods are suggested for hardwood rooted cuttings for all species except western hemlock. Rooted cuttings offer a viable option for clonal propagation, which will be useful in breeding programs for resistance screening. In addition, the establishment of a common-garden is discussed and will be useful for ongoing and future research in the field. Host-Plant Selection by the Hemlock Woolly Adelgid, Adelges tsugae Annand: Sensory Systems and Feeding Behavior in Relation to Physical and Chemical Host-Plant Characteristics
by Kelly Lynn Felderhoff Oten
A dissertation submitted to the Graduate Faculty of North Carolina State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy
Entomology
Raleigh, North Carolina
2011
APPROVED BY:
______Fred P. Hain John Frampton Committee Chair Minor Representative
______Allen C. Cohen John S. King
______John F. Monahan Robert M. Jetton
DEDICATION
I would like to dedicate this dissertation to my niece and nephew, Evelyn Grace and
Corbin Ray. May you grow up with love and respect for the natural world around you and appreciate the majestic forests of the world in which we live.
ii
BIOGRAPHY
Kelly Lynn Felderhoff was born in Texas on April 16, 1983. Second of four children, she was raised in College Station, TX by her parents, Chris and Sandy Felderhoff.
Throughout her childhood, it was apparent that she had a fascination with insects. She was involved in 4-H, with entomology as her main project and may have been the only fourth- grader to bring an insect collection for show-and-tell. While in high school, Kelly worked in the Texas A&M University Insect Collection. She graduated from A&M Consolidated High
School in 2001, knowing she wanted to pursue Entomology as a career.
In August 2001, Kelly enrolled at Texas A&M University to study entomology.
While pursuing her degree, she worked on undergraduate research projects in the field of forensic entomology with Dr. Jimmy Olson. Her project results revealed the importance of using proper preservation techniques in forensic collections in order to calculate an accurate post-mortem interval; her data were used as preliminary results for a research proposal. She also participated in the Texas Integrated Pest Management Internship Program, and experienced the agricultural side of entomology. In addition, she evaluated the effectiveness of the program since its inception, resulting in her first publication. Kelly graduated with her
B.S. in Entomology in August 2005.
The day following graduation, Kelly moved to Knoxville, TN to pursue her M.S. degree in entomology at the University of Tennessee. Under the direction of Dr. Ernest
Bernard, her thesis included a revision of the family Tomoceridae (Collembola) in Great
Smoky Mountains National Park. Using morphological and molecular techniques, she
iii
described two new species, resurrected one species, and identified two species complexes.
She graduated in December 2007.
In January 2008, Kelly began her doctoral program in entomology with a minor in forestry at North Carolina State University. Under the direction of Dr. Fred P. Hain, she studied the feeding biology of the hemlock woolly adelgid and investigated mechanisms of host-plant resistance between hemlock species; this dissertation presents her findings. While a student, she participated in the Preparing the Professoriate Program, which gave her the opportunity to teach her own course. This experience convinced her that teaching in some context must be a part of her career. In the spring semester following graduation, she will teach the undergraduate forest entomology course at NCSU.
iv
ACKNOWLEDGMENTS
Throughout this journey towards my doctoral degree, I am conscious that my accomplishments could not have been possible without the support of many people. First, I would like to thank Fred Hain for having faith in my ability to successfully complete this program. He played an integral role in my professional development and encouraged me to take on things that I may not have otherwise. Under his guidance, I grew as an individual personally and professionally, and I view him as a friend and a significant influence as a young scientist. I hope to implement many of his qualities when I become a mentor myself someday. I am grateful to John Frampton as well, who challenged me and supported me throughout my project. Thanks go to my committee members for their guidance and encouragement: John King, Robert Jetton, John Monahan, and Allen Cohen. I especially want to thank Allen Cohen, for his friendship, his unwavering interest in my research, and for always making the time to meet with me. His insight was invaluable and his patience endless. Allen’s wife, Jackie, was also extremely supportive and always a friendly face. I also want to thank the Cohens for my wonderful lab mate, Spotty, who kept me company during long hours of dissertation writing.
I express gratitude to sources of funding for my project, the USDA Forest Service,
Northern Research Station and Forest Health Protection in Region 8. In addition, support from the NCSU Department of Entomology made this project possible. I thank the department for taking me on.
I am very grateful for the time, enthusiasm, and ideas from John Strider. He was always willing to help me with my research, whether it include travel to my field sites or
v
trouble-shooting difficult analytical equipment. My life as a graduate student was made so much easier with his assistance and his willingness to help. I am also very grateful to the faculty, staff, and students in the Department of Entomology and Department of Forestry and
Environmental Resources. In particular, I acknowledge and thank Micah Gardner, Gene
Dupree, Jean Carter, JC Domec, Coby Schal, Zac Arcaro, Leslie Newton, David Bednar, Erin
Mester, Kata Boroczky, Andy Whittier, and Navdip Kaur. I could not have completed this work without the help from numerous collaborators: Gary Bauchan of USDA Electron
Microscopy Unit, Glenn Kohler of Washington State Department of Natural Resources,
Kunso Kim and David Marin of the Morton Arboretum, Susan Bentz and Richard Olsen of the National Arboretum, and Rebecca Norris and Kathy Kidd of the NCDA&CS Plant
Quarantine Facility. In addition, I would like to thank Wes Watson and the Veterinary
Entomology Lab at NCSU for allowing me to use equipment and lending me freezer space.
Also, thanks to the many fellow forest entomologists I have met at conferences and with which I had many invaluable conversations. In particular, I wish to thank John Riggins for his friendship and for encouraging me to take on the role of manuscript reviewer.
I am so thankful for the many friends I have met in Raleigh who offered their companionship and a never-too-often, sanity-restoring distraction. I came to the area knowing no one and now feel surrounded by love and support. I especially thank Alaina and
Joe Pantoja, Rhiannon Potter and Tim Michalski, Alana Jacobson, Diane Silcox, Shana
Keating, Andriena and Chad Smith, Renee and Chris White, Kristen and Gabe Fritz, Tara and Josh Clowes, Sumar and Bill Wightman, and Brian and Andrea Pacetti.
vi
I owe so much to my family, who has been understanding and supportive of my pursuit for a higher education. Despite distance, I am conscious of their love and support and
I hope I can extend the same to my sister, Rachel, as she completes law school and my brother, Terry, as he begins his own doctoral program. I am so grateful for my niece and nephew and thank my sister, Lydia, and my brother-in-law, Collin, for sending pictures of them anytime I needed a smile.
I will eternally be grateful to my parents. Through our numerous family camping trips in our youth, I fell in love with nature and owe my dedication to the field of entomology to them. My dad instilled a love of science in me, with his passion to figure things out and his fascination with the natural world. My mom was encouraging to all of us, no matter what we chose to pursue in life. She gave me the confidence to follow through with my interest in insects, initiating my development as a young entomologist by creating an entomology project group within our local 4-H club. Without her guidance, I would not be the person I am today, both professionally and personally. I thank both of them for their continued friendship and support and I love them both so very much.
Finally, my deepest and most heartfelt thanks go to my husband, Brian Oten. His love was unwavering throughout this journey. He continually conveyed his faith in me to not only complete my degree, but to do it well. I depended on his patience, his guidance, and his sense of humor. Our mini celebrations following the smallest of feats kept me going. I know he will remain a steadfast supporter and an amazing partner in our many years together to come and I only hope I can be an anchor for him, as he has been for me.
vii
TABLE OF CONTENTS
List of Tables ...... xii
List of Figures ...... xiii
Chapter I: A Review of the Relevant Literature Relating to Hemlock and the Hemlock Woolly Adelgid ...... 1
Dissertation Introduction ...... 1
Biology of the Hemlock Woolly Adelgid ...... 4
Management of the Hemlock Woolly Adelgid ...... 8
Hemlocks of the World ...... 15
Literature Cited ...... 25
Figures...... 37
Chapter II: Morphological Observations of Hemlock Woolly Adelgid Host-plant Interactions using Light and Electron Microscopy ...... 40
Abstract ...... 40
Introduction ...... 41
Materials and Methods ...... 46
Results ...... 48
Discussion ...... 50
Literature Cited ...... 60
Figures...... 67
Chapter III: Fine Structure of Labial, Tarsal, and Antennal Sensilla and Setae of Sistens Crawlers of the Hemlock Woolly Adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae) ...... 73
Abstract ...... 73
viii
Introduction ...... 73
Materials and Methods ...... 76
Results ...... 77
Discussion ...... 79
Literature Cited ...... 87
Figures...... 93
Chapter IV: Trophically-related Enzymes of the Hemlock Woolly Adelgid, Adelges tsugae (Hemiptera: Adelgidae) ...... 99
Abstract ...... 99
Introduction ...... 100
Materials and Methods ...... 104
Results ...... 108
Discussion ...... 109
Literature Cited ...... 121
Figures...... 130
Chapter V: Biophysical Characteristics of Six Hemlock Species' (Tsuga spp.) Surfaces: Implications for Resistance to the Hemlock Woolly Adelgid, Adelges tsugae (Hemiptera: Adelgidae) ...... 139
Abstract ...... 139
Introduction ...... 140
Materials and Methods ...... 144
Results ...... 146
Discussion ...... 149
ix
Literature Cited ...... 158
Tables and Figures ...... 169
Chapter VI: Chemical Composition and Variability of Epicuticular Lipids of Eight Hemlock (Tsuga spp.) Taxa and their Relationship to Hemlock Woolly Adelgid (Adelges tsugae) Host-Plant Selection ...... 178
Abstract ...... 178
Introduction ...... 178
Materials and Methods ...... 183
Results ...... 185
Discussion ...... 186
Literature Cited ...... 190
Tables and Figures ...... 194
Chapter VII: Infestation of Hemlock Woolly Adelgid among Three North American Hemlock (Tsuga) Species and their Growth Rates ...... 211
Abstract ...... 211
Introduction ...... 212
Materials and Methods ...... 216
Results ...... 218
Discussion ...... 219
Literature Cited ...... 224
Figures...... 229
Chapter VIII: Comparative Vegetative Propagation of Seven Hemlock Species (Tsuga spp.) by Rooted Hardwood Cuttings ...... 234
x
Abstract ...... 234
Introduction ...... 235
Materials and Methods ...... 237
Results ...... 238
Discussion ...... 239
Literature Cited ...... 242
Tables and Figures ...... 246
APPENDICES ...... 250
Appendix A: Establishment of a Hemlock Common-Garden in Western North Carolina ...... 251
Abstract ...... 251
Introduction ...... 251
Materials and Methods ...... 254
Discussion ...... 255
Literature Cited ...... 258
Figures...... 260
Dissertation Conclusions ...... 262
xi
LIST OF TABLES
Chapter V
Table 1. Hemlock (Tsuga) species used in studies: resistance documented in literature, native range, and source of samples ...... 169
Chapter VI
Table 1. Hemlock species and sources analyzed with GC/MS ...... 194
Table 2. Peak identification numbers, associated retention times, and the summary of the species containing the peak ...... 197
Chapter VIII
Table 1. Hemlock (Tsuga) species used in comparative vegetative propagation ...... 245
xii
LIST OF FIGURES
Chapter I
Figure 1. Distribution of the hemlock woolly adelgid (Adelges tsugae) in the eastern United States as of 2010 ...... 37
Figure 2. Stylet penetration by members of Aphidoidea through a transverse section of plant tissue ...... 38
Figure 3. World-wide distribution of hemlocks (Tsuga spp.) ...... 39
Chapter II
Figure 1. Feeding sites of the hemlock woolly adelgid ...... 67
Figure 2. Hemlock woolly adelgid external morphology ...... 68
Figure 3. Labium and stylet morphology ...... 69
Figure 4. Stylet cross-sections ...... 70
Figure 5. Adaxial surface of needle petiole and pulvini with settled HWA ...... 71
Figure 6. Western North American HWA population compared to eastern North American population HWA ...... 72
Chapter III
Figure 1. Scanning electron micrograph of the hemlock woolly adelgid crawler ...... 93
Figure 2. Scanning electron micrographs of the hemlock woolly adelgid labium, segment I ...... 94
Figure 3. Scanning electron micrographs of the hemlock woolly adelgid labium segments II and III ...... 95
Figure 4. Scanning electron micrographs of the hemlock woolly adelgid labium tip, segment IV...... 96
Figure 5. Scanning electron micrograph of the hemlock woolly adelgid tarsus ...... 97
Figure 6. Scanning electron micrographs of the hemlock woolly adelgid antenna ...... 98
xiii
Chapter IV
Figure 1. Enzyme kinetics of trypsin-like assay ...... 130
Figure 2. Trypsin-like activity per µg protein per minute of A. tsugae and L. lineolaris ...... 131
Figure 3. Enzyme kinetics of amylase-like enzyme assay ...... 132
Figure 4. Amylase-like activity per µg protein per minute of A. tsugae and L. lineolaris ...... 133
Figure 5. Enzyme kinetics of peroxidase assay ...... 134
Figure 6. Peroxidase activity per µg protein per minute of A. tsugae and L. lineolaris .135
Figure 7. Enzyme kinetics of polyphenol oxidase assay ...... 136
Figure 8. Polyphenol oxidase activity per µg protein per minute for A. tsugae, A. tsugae incubated with PTU for 20 min, and L. lineolaris ...... 137
Figure 9. Tangential view of xylem ray parenchyma cells of eastern hemlock, Tsuga canadensis ...... 138
Chapter V
Figure 1. Hemlock woolly adelgid insertion site in relation to other sites measured on pulvini ...... 170
Figure 2. Cross-sections of hemlock stem ...... 171
Figure 3. Hemlock cuticle at hemlock woolly adelgid stylet insertion point ...... 172
Figure 4. Effect of hemlock species on cuticle thickness ...... 173
Figure 5. Effect of location on cuticle thickness ...... 174
Figure 6. Effect of species on cuticle thickness at insertion point ...... 175
Figure 7. Cuticle thickness across all measurement points for six species ...... 176
Figure 8. Wax structures ...... 177
xiv
Chapter VI
Figure 1. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-4 (Sample shown: A-14) ...... 198
Figure 2. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-4 and a small peak 6 (Sample shown: A-16) ...... 199
Figure 3. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-6 (Sample shown: A-18) ...... 200
Figure 4. Chemical profile of Chinese hemlock, Tsuga chinensis, indicating peaks 1-5 (Sample shown: A-27) ...... 201
Figure 5. Chemical profile of Carolina hemlock, Tsuga caroliniana, indicating peaks 1-5 (Sample shown: A-24) ...... 202
Figure 6. Chemical profile of Carolina hemlock, Tsuga caroliniana, indicating peak 6 (Sample shown: A-25) ...... 203
Figure 7. Chemical profile of Carolina hemlock, Tsuga caroliniana, with multiple additional peaks (Sample shown: A-10) ...... 204
Figure 8. Chemical profile of southern Japanese hemlock, Tsuga sieboldii, with numerous small peaks between 35 and 38 min (Sample shown: A-2) ...... 205
Figure 9. Chemical profile of northern Japanese hemlock, Tsuga diversifolia, with peaks 1-5, 6, 8, and 11 (Sample shown: A-33) ...... 206
Figure 10. Chemical profile of western hemlock, Tsuga heterophylla (Sample shown: A-13) ...... 207
Figure 11. Chemical profile of mountain hemlock, Tsuga metensiana (Sample shown: A-12) ...... 208
Figure 12. Chemical profile of the hemlock hybrid, Tsuga chinensis x T. caroliniana (Sample shown: A-45) ...... 209
Figure 13. Effect of the period of time since cutting material on epicuticular wax composition (T. canadensis). A) Wax extracted directly from the tree; B) Wax extracted one week after cutting; C) Wax extracted two weeks after cutting...... 210
xv
Chapter VII
Figure 1. Percent infestation level of Adelges tsugae for each species per block (2008) ...... 229
Figure 2. Percent infestation across species and block by year...... 230
Figure 3. Mean monthly growth by species and block for the growing period in 2008 ..231
Figure 4. Mean monthly growth by species and block for the growing period in 2009 ..232
Figure 5. Mean monthly growth by species and block for the growing period in 2010 ..233
Chapter VIII
Figure 1. Images of hemlock cuttings, indicating health rating ...... 246
Figure 2. Mean health of cuttings by species at approximately four months post-rooting ...... 247
Figure 3. Mean health of cuttings by sample source at approximately four months post- rooting...... 248
Figure 4. Number of cuttings per species/source for each health rating at approximately four months post-rooting...... 249
Appendix A
Figure 1. Hemlock common-garden at the Upper Mountain Research Station (Laurel Springs, NC), 25 June 2009 ...... 260
Figure 2. Diagram of the hemlock common-garden in Laurel Springs, NC ...... 261
xvi
Chapter I
A Review of the Relevant Literature Relating to Hemlock and the Hemlock Woolly Adelgid
DISSERTATION INTRODUCTION
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is a small (~1.0 mm) piercing-sucking insect pest and the largest threat to ornamental and forest hemlock in eastern North America. Native to the Pacific Northwest and parts of Asia (Annand 1924, Havill et al. 2006), it first appeared in the eastern United
States in 1951 near Richmond, Virginia (Gouger 1971, Souto et al. 1996). Initially thought to be a pest of the ornamental industry, populations became a problem to forest hemlock in the 1980s. Since then, HWA has spread to 18 eastern states, ranging from northern Georgia to southern Maine (USDA Forest Service 2010) (Figure 1).
The destruction HWA has caused to eastern (Tsuga canadensis (L.) Carriére) and
Carolina hemlocks (T. caroliniana Engelm.) to date is tremendous. Nearly 80% of hemlocks native to the eastern United States have already vanished as a result of this exotic insect
(Hale 2004). Certain natural areas have experienced even more severe density reduction – more than 90% of the hemlocks in Shenandoah National Park in Virginia are dead
(Townsend and Rieske-Kinney 2006). Moreover, the production of hemlocks in the ornamental industry has been virtually eliminated. Following initial infestation the tree declines in health, marked by needle drop, bud abortion, and inhibition of new growth
(McClure 1991). A healthy hemlock can be killed in as few as four years, but some survive beyond ten (McClure 1987, 1991).
1
Populations of HWA were first observed and originally described from individuals infesting western hemlock (T. heterophylla Sargent) in northwest North America in the early
1920’s (Annand 1924). The population responsible for the widespread hemlock mortality in eastern North America originated from populations infesting southern Japanese hemlock (T. sieboldii Carriére) in low elevations of Central Japan (Havill et al. 2006). Until its most recent invasion into the United States the biology of HWA remained largely unstudied
(McClure 1996).
Management of HWA is largely focused on biological control and chemical control, with minimal work on resistance (McClure 2001, Bentz et al. 2002). Of the current control tactics, chemical control is the most effective, but is not practical in the forest setting due to environmental impacts and prohibitive costs (McClure 1991, 1992). Biological control has been a major research focus for HWA management. Biological control programs have been in effect since the mid-1990s (Cheah et al. 2004), yet field efficacy remains uncertain.
Recovery at field sites is minimal (Salom et al. 2008, Grant et al. 2010), lab-reared beetles behave sub-par (Wallin et al. 2011), and introduced beetles are hybridizing with native species (Mausel et al. 2008, Havill et al. 2010). Moreover, the impact of released predators on hemlock health and/or HWA populations has yet to be seen (Grant et al. 2010).
With biological control seeming less and less promising, and chemical control being difficult, costly, and ecologically harmful, there is a need for understanding host-plant resistance to HWA. The hemlock woolly adelgid infests hemlocks in its native range of Asia and the Pacific Northwest, but there, the adelgid is described as a minor pest (Keen 1938,
Furniss and Carolin 1977, Bentz et al. 2002). For years it was believed that eastern and
2
Carolina hemlocks were entirely and exclusively susceptible to the hemlock woolly adelgid.
However, in the wake of large-scale destruction, anecdotal evidence suggested surviving individuals or stands of eastern and Carolina hemlocks may be less susceptible. This has since been corroborated by research and distinct populations have been selected and continue to be pursued as putatively resistant to the adelgid (Jetton et al. 2008b, Caswell et al. 2008,
Kaur 2009, Ingwell and Preisser 2010, Oten et al. 2011).
The research I present in this dissertation investigates host-plant resistance mechanisms from two perspectives, that of the insect pest and that of the host tree. Chapters
II, III, and IV investigate feeding behaviors of HWA. Detailed knowledge of insect-host plant interactions is a basic part of understanding resistance in any given case (Painter 1951) and a major part of this is understanding feeding behavior (Davis 1985). I assessed the host- plant interactions between HWA and hemlock using scanning electron microscopy (SEM) and was able to pinpoint insertion point and external sheath material production. I also assigned function to sensilla and setae of HWA based on morphological studies.
Understanding the type of sensory system utilized will assist in our understanding of the cues used by HWA in host-plant selection, a significant aspect of resistance. In Chapter III, trophically-related enzymes used by HWA are identified and discussed. Not only does this establish nutrients being ingested, but this could have implications for host plant resistance mechanisms, such as protease inhibition by plants. Chapters V, VI, and VII examine characteristics of the host plant, comparing the physical and chemical characteristics of the different hemlock species. These examinations give us insight into characteristics that may confer resistance to the more resistant hemlock species.
3
BIOLOGY OF THE HEMLOCK WOOLLY ADELGID
Life Cycle
HWA has two parthenogenetic generations per year that feed on hemlocks as their secondary host. The spring generation, present from March to June, is called progrediens, followed by the sistens generation, present from June until the following March (McClure
1989, Gray and Salom 1996). When progrediens mature in early summer, they produce up to
300 eggs within cottony white ovisacs. A large portion of these progeny become wingless adult progrediens, which are morphologically similar to their mothers and remain on hemlock to aestivate through the summer. In October, they break aestivation and commence feeding (McClure 1989). The remaining portion, a third generation, becomes winged adult sexuparae that leave hemlock, migrating to a primary spruce host (Picea spp.). In the eastern
U.S., a suitable host spruce is not available and this generation presumably departs from its hemlock host and does not survive (McClure 1987, McClure 1989).
This life cycle description is based off studies on Connecticut populations of HWA; however, developmental rates can fluctuate by a month or more in either direction depending on environmental conditions or geographic location (McClure 1996, Gray and Salom 1996,
Salom et al. 2001). The most important abiotic factors influencing HWA development are probably temperature and photoperiod (Zaslavski 1988). Diapause is maternally regulated using preconditioning temperature cues. It can be artificially prevented by maintaining colonies at 14.5°C or colder at an optimal 12:12 photoperiod (Salom et al. 2001).
Temperature not only governs HWA development, but low winter temperatures impede its northerly spread and contributes to HWA mortality. The rate of HWA spread is
4
approximately 12.5 km/yr, but can vary significantly with environmental factors (Evans and
Gregoire 2006). Low lethal mortality is evident with 11% survival at -20°C in artificial conditions. At -35°C, there is no HWA survival. Warmer climate in southern regions has enabled more rapid spread, while extreme lows in the northern range slow movement (Parker et al. 1999). In further studies, cold hardiness depending on geographical location and time of year is evident (Skinner et al. 2003).
Feeding Biology
The feeding behavior of HWA is poorly understood. Studies by Young et al. (1995) give the most thorough examination of HWA feeding behaviors to date, yet a large part of our understanding of HWA feeding is based on the aphid system as a model because they are the closest relative to adelgids (von Dohlen 2008).
HWA feeds on xylem ray parenchyma cells. These cells are comprised mostly of starches, act as stored energy for the hemlock, and are used in the production of new growth.
Correspondingly, one of the first signs of HWA infestation is failure to produce new growth
(McClure 1991). Parenchyma feeders, such as adelgids (Young et al. 1995), are known to secrete saliva continuously (Auclair 1963). Miles (1986) indicates that adelgids penetrate tissue intercellularly (Figure 2), but Young et al. (1995) report a mixed intercellular and intracellular pathway.
Stylets are the functional mouthparts of Hemiptera capable of reaching tissue far from the insect body (Kloft and Ehrhardt 1962). The stylet bundle, composed of four individual stylets, is the long flexible mouthpart inserted into the plant, able to reach tissues relatively far from the body of the insect (Miles 1968). The stylet bundle of HWA appears to
5
correspond with this aphid model (Young et al. 1995). The four stylets dovetail to form two canals: a food canal for transport of nutrients from the plant to the insect, and a salivary canal, from which products of the salivary glands are injected into the plant (Miles 1999).
After inserting its stylets into plant tissue, the insect may withdraw and then move forward again on a new path, branching to one side or the other of the old path but the resultant branches seem always to be more or less in the same dimension throughout feeding
(Miles1968). This behavior and evidence has been casually dubbed as "wandering sheaths" by Dr. Allen Cohen. Morphological evidence indicates the presence of nerves within the central duct of mandibular stylets within Hemiptera (Forbes 1966). These nerves may assist in the positioning of the stylets as they move about.
Stylets of liquid-feeding insects are different from those that use extra-oral digestive enzymes. The food canal of true fluid feeders is approximately 1 µm in diameter, compared to 10 µm in insects that exploit extra-oral digestion (Cohen 1996 in Cohen 2004). Previous research indicates the presence of amylase in HWA saliva, but it was surmised to originate from microorganisms of the gut (Kaur 2009).
The saliva of Hemiptera plays important roles as a physiochemical agent during tissue penetration and food digestion. All Hemiptera utilize two kinds of saliva: a sheath material, which solidifies rapidly upon ejection and watery saliva (Miles 1972). The hemipteran stylet morphology accommodates the saliva with a canal distinct from the food canal. Saliva is pumped down one side, and nutrients obtained from the food source are sucked up the other
(Miles 1968).
6
Stylet sheaths are mostly made of proteins (Srivastava 1987). Saliva plays many roles in Hemipteran feeding, including lubrication of the mouthparts, dilution of food, help in stylet penetration, excretion of some metabolites, initiation of food digestion, prevention of deleterious changes in the food source, and keeping the sieve tubes open (Dixon1975, Miles
1999).
The watery saliva may also contain digestive enzymes, pumped down the salivary canal for extra-oral digesiton. Digestive enzymes or mechanical disruption of tissues are used to partially break down food extra-orally (Cohen 1998). This feeding habit is used in
80% of the predaceous families of arthropods and may be the reason for much of their success (Cohen 1995).
Not all enzymes found in the watery saliva is digestive in nature; the function of some enzymes can be rather ambiguous. Known functions include gelling of the stylet sheath, facilitating intercellular stylet penetration, or detoxifying agents (Peng and Miles 1988,
1991). The injection of phytotoxins into a plant may cause visible damage (Burd et al.
1998). In most of the piercing-sucking herbivorous insects, enzymes dissolve or soften the pectin layers of mid-lamellae and allow stylet penetration through the primary cell wall
(McAllan and Adams 1961). Polyphenol oxidase, an enzyme which neutralizes phenolic compounds produced by plants, is present in the saliva of many Aphididae, helping herbivores overcome the plant defenses and converting them into phagostimulants (Peng and
Miles 1988, 1991, Jiang 1996). In the rose aphid, Macrosiphum rosae (L.), catechol-oxidase converts plant-produced deterrents into non-toxic phagostimulants (Peng and Miles 1988).
7
MANAGEMENT OF THE HEMLOCK WOOLLY ADELGID
Silvicultural Practices
Historically, cultural control has been widely used in programs to control forest pests.
For example, a common approach to reducing the severity of southern pine beetle,
Dendroctonus frontalis Zimmermann, is to thin dense pine stands (Brown et al. 1987).
While a major approach has not been identified for hemlock woolly adelgid control, several silvicultural strategies can be used to mitigate spread and infestation severity, particularly in ornamental settings. Preventative measures can be taken proactively to avoid adelgid infestations altogether. Many animals, especially bird and deer, are known to spread HWA from one tree to another (McClure 1990). Any action taken to prevent them from visiting uninfested hemlocks will reduce their risk for becoming infested (McClure 1995b). Human mediated dispersal should also be avoided. This can be done by cleaning vehicles and clothing following a visit to forest settings, parks, or anywhere adelgid infestations are known in addition to inspecting plant materials (plants, logs, bark, wood chips, etc.) that are being transported to uninfested areas (McClure 1995b, Ward et al. 2004). For this reason, quarantines should be observed. Wind is also a factor in the passive dispersal of HWA
(McClure 1990). Larger trees can act as a launching pad for crawlers; therefore, selective removal of larger trees or trees that may be particularly attractive to wildlife may deter the establishment of new infestations (McClure 1995b). Along with human activity, one of the primary factors for risk of infestation is the proximity of a hemlock stand to existing HWA infestations (Faulkenberry et al. 2009).
8
In addition, healthy trees succumb less quickly to HWA infestations than those with low vigor (McClure 1995b). Therefore, actions taken to maintain good growing conditions are recommended. Soil moisture is crucial for hemlocks and can be maintained by mulching or irrigating one inch per week during periods of drought (Ward et al. 2004). It is not recommended to fertilize infested trees since it has been shown to enhance HWA survival and fecundity (McClure 1991, Joseph et al. 2010). However, fertilization may be useful for tree recovery following pesticides use. Removal of dead and unhealthy limbs and branches may also encourage hemlock health through the promotion of new growth (McClure 1995b).
It has also been suggested that mechanical removal of HWA infestations can reduce populations. This can be done by physically dislodging settled HWA, spraying high-pressure water stream at infested branches, or clipping heavily infested branches (McClure 1995b).
Chemical Control
Applications of insecticides can be useful and effective in HWA management of ornamental and forest hemlock with high recreational value (Steward and Horner 1994,
Cowles and Cheah 2002, Webb et al. 2003, Dilling et al. 2010, Webster 2010). Imidacloprid, a systemic neonicotinoid, is the most widely used and available in several formulations: water soluble packets, soluble tablets, and liquid. In addition, the chemical is flexible in its application method; it can be applied by a foliar spray, soil drench or injection, stem injections, or a combination of application methods to maximize protection (Doccola et al.
2010). Soil injections are considered the most effective and their residual effects can last several years (Cowles et al. 2006).
9
Chemical control of HWA can be useful in ornamental and landscape settings; however, it is not practical in a natural forest setting due to environmental impacts and prohibitive costs (McClure 1991, 1992). Pesticide applications are expensive, provide ephemeral protection, and may be harmful to non-target organisms (Dilling et al. 2009).
Following chemical treatment of hemlock, non-target soil organisms show significant declines in abundance and richness, especially in soil-dwelling springtails (Reynolds 2008), and the danger of contamination to proximal water sources pose a threat as well.
Applications are also limited due to geographical and logistical constraints; there are operational difficulties in bringing equipment into a forest (Cowles et al. 2006, Lamb et al.
2006). In some situations applications are ineffective altogether, as in Joyce Kilmer
Memorial Forest near Asheville, NC (Bompey 2010). Moreover, hemlocks are major components of riparian ecosystems where the danger of environmental contamination is high and in some cases pesticide usage is not allowed. Despite these negative aspects, chemicals remain the best option in ornamental settings, but are entirely impractical in forests.
Sustainable, long-term management of HWA in the forest setting will likely require biological control and/or host-plant resistance (Bentz et al. 2002, Cheah et al. 2004, Del
Tradici and Kitajima 2004).
Biological Control
In its native range, the HWA is thought to be controlled by resistance, scattered hemlock populations, and a complex of natural enemies (McClure 1992, Montgomery and
Lyon 1996). Beginning in the mid-1990s, classical biological control has been a major focus for the control of HWA (Cheah et al. 2004). Native, established predators have not been
10
found associated with HWA in the eastern United States (Wallace and Hain 2000), causing researchers to explore Asia and Northwest North America, the native range of HWA, for predators to use in a classical biological control program.
To date, several beetle species native to Asia have been released in hopes of reaching control. The first was a coccinelid native to Japan, Sasajiscymnus tsugae (Sasaji and
McClure), first released in 1995 (Cheah and McClure 2000, Cheah et al. 2005). There have been over two million S. tsugae released throughout the eastern U.S. since 2002 (Grant 2008,
Salom et al. 2008). Scymnus sinuanodulus, a Chinese coccinelid, has been released in at least eleven eastern states since 2004, numbering 25,000, but data suggests their optimal range is likely restricted to the southern region of the hemlock range in the eastern U.S.
(Salom et al. 2008). A predator from the Pacific Northwest, Laricobius nigrinus Fender, has also been released: over 50,000 beetles in over 80 locations (Salom et al. 2008).
The efficacy of these beetles in the field is unclear. Recovery of beetles at release sites has been minimal and establishment is taking longer than expected (Grant et al. 2010).
Particularly, the failure to recover numerous S. tsugae has led to questions in its efficacy and ability to establish populations (Salom et al. 2008). It is also unclear if lab-reared predators are fit enough to succeed in a natural environment. In olfactory assays, field-captured L. nigrinus exhibited orientation towards volatiles from hemlock and other conifers. However, lab-reared L. nigrinus did not respond to host cues and appeared lethargic. Beetles of both sources were unable to detect odors from HWA (Wallin et al. 2011). In addition to these issues, L. nigrinus released in the eastern United States has been observed mating with and successfully hybridized with the native L. rubidus (Mausel et al. 2008, Havill et al. 2010).
11
The ecological impacts of this are unknown. Finally, the impact of these predators on hemlock health, growth, survival or on HWA populations in the field is entirely unknown
(Grant et al. 2010).
Several other predacious beetles are being explored as potential agents and may be released soon. Laricobius osakensis Montgomery and Shuyake (Coleoptera: Derodontidae),
Scymnus camptodromus, S. ningshanensis, S. coniferam (Coleoptera: Coccinelidae) have been discussed as potential agents for the biological control of HWA (Lamb et al. 2006,
Lamb et al. 2010, Keena and Montgomery 2010, Hoover et al. 2010, Wallin and Ott 2010).
Following release, recovery and efficacy should be monitored closely.
Other organisms that have been recognized as potential biological control agents.
Tetraphleps galchanoides (Hemiptera: Anthocoridae), Leucopis argenticollis (Diptera:
Chamaemyiidae), Diapterobates humeralis (Oribatida: Ceratozetidae) have all been minimally investigated and are candidates for a more rigorous appraisal as biological control agents (McClure 1995a, McAvoy et al. 2007, Salom et al. 2008). For Leucopis spp., challenges, such as lack of rearing methods or difficulty in identification, have slowed the development of release programs (Ross et al. 2010).
A fungal agent, Lecanicillium muscarium is also currently under investigation (Salom et al. 2008). When administered via Mycotal, a commercial biopesticide from The
Netherlands, the fungus caused over 50% reduction in HWA population growth, but only when augmented with whey-based fungal enhancer. Challenges to this approach lie in biotic and abiotic, practical, economic, and regulatory factors (Costa 2010).
12
Although biological control agents may help control adelgid levels in some situations, the relatively high susceptibility of T. canadensis to HWA presents a unique problem; biological control agents must cause extremely high mortality to HWA to maintain populations at innocuous levels (McClure 1996). Ultimately, their incorporation into a well- rounded integrated pest management program may be where biological control agents are the most successful.
Host-Plant Resistance
Although much of the literature cites T. canadensis and T. caroliniana as being the only two susceptible hosts for HWA (Del Tredici and Kitajima 2004; Lagalante et al. 2006;
McClure 1992, 1995b, 1996), recent studies cast doubt on the true susceptibility of T. caroliniana, inferring the possibility of resistance. A recent study was performed using three species of hemlocks (T. canadensis, T. caroliniana, and T. heterophylla). They were artificially inoculated with HWA and it was observed that both T. caroliniana and T. heterophylla had significantly lower number of feeding adelgids than T. canadensis (Jetton et al. 2008b). This suggests that settlement rates on Carolina hemlock may be lower than that of Eastern hemlock; its complete susceptibility remains dubius.
The remaining hemlock species in the western U.S. and Asia are putatively adelgid resistant (McClure 1995a, Pontius et al. 2006, Jetton et al. 2008b). HWA is found on T. heterophylla in the Pacific Northwest, but it seldom injures forest hemlocks there (McClure
1989). In British Columbia, hemlock trees are rarely injured by HWA in the natural forest setting (Furniss and Carolin 1977). Laricobius nigrinus is found in close association with the
HWA there (Zilahi-Balogh et al. 2003), most likely contributing to the natural enemy
13
complex in the region. In its native range in Asia, HWA inhabits forests and ornamental plantings of T. diversifolia, T. sieboldii, and T. chinensis but its populations there are regulated by a combination of host resistance and natural enemies (McClure 1992, 1995a,
Montgomery et al. 2000, Del Tredici and Kitajima 2004). It also occurs at low and innocuous densities in Formosa and in Japan (Takahashi 1937 in McClure 1989, McClure
1987). Resistance of hemlocks in the native range of HWA still occurs when these species are grown in the U.S. in close association with highly infested, susceptible trees. Bentz et al.
(2002) observed a cultivated plot in Washington, D.C. After eight years of exposure to an infested T. canadensis, specimens of T. diversifolia and T. chinensis showed strong levels of resistance, while specimens of T. sieboldii showed variable levels. In Philadelphia, T. chinensis and T. sieboldii in the Morris Arboretum show none and little infestation of HWA, respectively (Bentz et al. 2007). A cultivated T. chinensis at the Arnold Arboretum has been there since 1911 and as of 2004, it was uninfested (Del Tredici and Kitajima 2004). In the same location, specimens of T. canadensis and T. chinensis, with six and four years exposure to HWA, respectively, were grown concurrently. They were measured for infestation levels and growth. The trial group consisting of T. chinensis showed no presence of HWA ovisacs and 100% of the branches had new growth on the tips. The trial group consisting of T. canadensis had high levels of ovisacs and only 45% of the branches had new growth (Del
Tredici and Kitajima 2004).
The mechanisms for the host resistance demonstrated by the Asian and Western U.S. hemlocks are unknown. It has been suggested that foliar chemistry plays a major role in palatability of the needle to the insect, with high concentrations of N and K causing higher
14
HWA population levels. Likewise, high concentrations of Ca and P may deter HWA infestations (Pontius et al. 2006). Painter (1951) distinguished three forms of plant resistance to insects: non-preference, antibiosis, and tolerance. Non-preference is the form of resistance that inhibits the insect from selecting the plant altogether, either as a food, shelter, or oviposition site. Antibiosis is the plant’s ability to disrupt the normal functions of the insect.
Tolerance refers to the ability of a plant to recover after undergoing attack (Baliddawa 1985).
The type and level of resistance the hemlocks are experiencing is presently unknown.
Attempts to hybridize eastern North American hemlock species with Asian species to create an adelgid-tolerant species are being made. Bentz et al. (2002) attempted to cross five species of Asian and eastern North American species. Attempts to cross T. canadensis with three Asian species were unsuccessful, but more than 50 authentic hybrids were created by crossing T. caroliniana and T. chinensis.
HEMLOCKS OF THE WORLD
Description of the Genus
Hemlocks are conifers in the genus Tsuga, as established by Carrière in 1855, and are the most shade tolerant and drought-susceptible of the Pinaceae. Typical forest types for hemlock growth are cool and humid, receiving high amounts of precipitation. They may regenerate under their own canopy, and often germinate on nurse logs, litter, moss beds, and humus (Farjon 1990).
Hemlocks are identified by their short, linear, petiole-based needles positioned on pulvini, or needle cushions (Dallimore and Jackson 1948, Farjon 1990, Iosue 2008).
15
Hemlocks are monoecious and possess almost spherical cones (Eckenwalder 2009). As many as 20 species have been described by various authors (van Gelderen and van Hoey
Smith 1986, Montgomery et al. 2000), but only nine are generally accepted as true species.
The discontinuous distribution of hemlock occurs almost exclusively in the northern hemisphere, with species concentrated in parts of eastern Asia and two areas in North
America widely separated from each other (Figure 3).
In Asia, five species of hemlocks occur. Two species occur in Japan, northern
Japanese hemlock (T. diversifolia Masters) and southern Japanese hemlock (T. sieboldii
Carrière). Further west, three species occur in China, Chinese hemlock (T. chinensis
(Franchet) E. Pritzel), Himalayan hemlock (T. dumosa), and Forrest hemlock (T. forrestii)
(Farjon 1990, Montgomery et al. 2000).
Northern Japanese hemlock may reach heights of 25 m and is considered resistant to infestations of HWA (McClure 1992, Montgomery et al. 2009). Its distribution is entirely within Japan, in the mountainous region (700-2000 m elevation) of the islands Honshu and
Kyushu. At higher elevations, it may be shrubby in appearance (Eckenwelder 2009). The associated climate is cool and moist with high amounts of precipitation (Farjon 1990).
Southern Japanese hemlock also occurs in Japan, at elevations lower than northern
Japanese hemlock occurs (Eckenwelder 2009). This hemlock is the ancestral host species from which the invasive North American HWA population originated (Havill et al. 2006). It can reach heights of 30 m. It is found in the hills and lower mountains of southern Japan at
500-1500 m elevation (Farjon 1990). Although its range overlaps that of northern Japanese hemlock, the two do not hybridize naturally (Eckenwelder 2009). The habitat is moist-
16
temperate and it is typically associated with soils derived from granitic or volcanic rock.
Southern Japanese hemlock is typically associated with both conifers and broad-leaved trees: momi fir (Abies firma Siebold et Zuccarni), Japanese Douglas-fir (Pseudotsuga japonica
Kino), Japanese cypress (Chamaecyparis obtusa Siebold et Zuccarni), Japanese cryptomeria
(Cryptomeria japinica D. Don), Japanese red pine (Pinus densiflora Siebold et Zuccarni),
Japanese white pine (P. parviflora Siebold et Zuccarni), Japanese umbrella pine
(Sciadopitys verticillata Siebold et Zuccarni), tall stewartia (Stewartia monadelpha (Siebold et Zuccarni)), evergreen witch hazel (Distylium racemosum Siebold et Zuccarni), wheel tree
(Trochodendron aralioides Siebold et Zuccarni), and others (Farjon 1990).
Chinese hemlock is considered to be the most highly resistant of all hemlocks to infestations of HWA (Del Tradici and Kitajima 2004). It may reach heights of 50 m, but is typically shorter. It is widely distributed and occurs in the mountains of northwestern
Yunnan and Gansu to Zhejiang and Guangdong in China, with elevations ranging between
1200-3200 m. A typical habitat for Chinese hemlock is montane, humid, and temperate, found in mixed forests amongst other conifers and hardwoods (Farjon 1990, Eckenwelder
2009).
Himalayan hemlock may be 35-50 m in height occurring in a broad belt from northern Uttar Pradesh in India to southwestern Sichuan in China and northern Myanmar.
Typically, it is found on moist to wet slopes in the monsoon region at 2,200-3,000 m
(Eckenwelder 2009). In some areas of its range, up to 10,000 mm annual precipitation occurs. It is associated with conifers: fir (Abies spp.), spruce (Picea spp.), deodar cedar
(Cedrus deodara (Roxburgh)), and larch (Larix griffithiana Carrière) (Farjon 1990).
17
Forrest hemlock typically reaches heights of 25-30 m and occurs in the mountains of the southwestern plateau of China where the range of Chinese and Himalayan hemlocks overlap. The range is limited and not entirely delineated, with an elevation range between
2000-3500 m. Typical climate is cold-temperate with 1000-2000 mm annual precipitation
(Farjon 1990). It is similar to Himalayan hemlock, and some consider it a subspecies of
Himalayan hemlock (Farjon 1990), or a hybrid between Himalayan and Chinese hemlocks
(Eckenwalder 2009).
Within North America, two hemlock species are found in the Pacific Northwest.
Western hemlock (T. heterophylla Sargent) extends from northern California to central
Alaska east to the northern Rocky Mountain region from central Idaho to northern British
Columbia from 600-1800 m in elevation. It is considered the tallest of all hemlocks, usually reaching heights of 60-70 m. It may be found occasionally in pure stands near the coast, but is mostly a component of mesothermal coniferous forests. Additionally, western hemlock is believed to be sympatric with Sitka spruce (Picea sitchensis (Bong.) Carrière) (Farjon 1990).
The first description of HWA came from populations infesting western hemlock in the
Pacific Northwest, where it is considered to be a minor pest of western and mountain hemlocks (Annand 1924).
The other species occurring in the Pacific Northwest, mountain hemlock (T. mertensiana (Bong.), occurs from Alaska to the southern Sierra Nevadas, occupying coastal regions in its northern range and montane regions to the south. It is primarily found on moist, north-facing slopes. Mountain hemlock may grow in pure or mixed stands, often associated with subalpine fir (Abies lasiocarpa (Hook.) Nutt.)), Pacific silver fir (A. amabilis
18
(Dougl.) Forbes), white spruce (Picea glauca (Moench) Voss), or other species (Farjon
1990). Some taxonomists separate mountain hemlock from the other eight hemlock species into its own genus or botanical section (Hesperospeuce) based on morphological characteristics, but mounting DNA evidence seats it firmly within Tsuga, as a sister group to western hemlock (Eckenwalder 1999, Havill et al. 2008). It is also unique in that it can tolerate sites with increased radiation and water deficiencies (Farjon 1990).
Both hemlock species in the western U.S. are less susceptible to HWA infestations than their eastern counterparts (Annand 1924, Montgomery et al. 2009), but they are not well adapted to the climate of eastern North America (Dirr 1998, Del Tradici and Kitajima 2004).
Therefore, they are not an effective substitute for hemlock replacement in areas of hemlock loss or in the ornamental industry.
Hemlocks of the Eastern United States
The two species in the eastern U.S., eastern and Carolina hemlocks, are unique in that they presumably did not co-evolve with HWA, making them the only two species of hemlock susceptible to infestations (Havill et al. 2006). For this reason, research presented in this dissertation largely focuses on these species, especially in comparison to the other, less susceptible species.
Eastern hemlock is also called Canadian hemlock or hemlock spruce. They are shade-tolerant and can live over 900 years, making them major components of old growth forest communities. Countless flora and fauna species have adapted to the distinctive shade that hemlocks provide. The wood is important to the pulp and paper industry and it is a popular ornamental tree (Godman and Lancaster 1990, Montgomery et al. 2009).
19
Eastern hemlock is often 30-50 m in height, with a straight, rarely forked trunk covered in brown, scaly, furrowed bark. Needles are long and tapering, 0.5-1.8 cm in length, with toothed edges and a petiole at its base, a characteristic unique to Tsuga spp. Needle tips may be pointed or rounded and there are two stomatal bands on the underside of the needle with five or six lines of stomates each (Eckenwalder 2009). Numerous branches spread horizontally and droop at the ends (Farjon 1990).
The range of eastern hemlock extends throughout Northeastern North America. Its range extends west to southern Ontario, northern Michigan, Wisconsin, and eastern
Minnesota, and extends south to the southern Appalachians, reaching the northern parts of
Georgia and Alabama (Farjon 1990). The typical site for eastern hemlock is moist ridges, slopes, and valleys, from 600 - 1,800 m in elevation (Eckenwalder 2009). In Nova Scotia, its range nears sea level. Optimal soil type for eastern hemlock growth is moist to very moist, well-drained, and acidic (pH 3-4) (Farjon 1990, Godman and Lancaster 1990). The associated climate is cool and humid, receiving 700-1500 mm precipitation annually. While eastern hemlocks may grow in pure stands, they are typically components of mixed ecosystems with other conifers and broad-leaved trees. Associated species include: white pine (Pinus strobus L.), red pine (P. resinosa Ait.), balsam fir (Abies balsamea (L.) Mill.), red spruce (Picea rubens Sarg.), white spruce (P. glauca (Moench) Voss), tamarack larch
(Larix laricina (Du Roi) K. Koch), birch (Betula spp.), sugar maple (Acer saccharum
Marsh), northern red oak (Quercus rubra L.), American or white ash (Fraxinus americana
L.), black ash (F. nigra Marsh), American beech (Fagus grandifolia Ehrh.), poplars (Populus spp.), and others (Farjon 1990).
20
Carolina hemlock reaches 30 m in height, but is typically shorter. Its bark is similar to eastern hemlock: furrowed, scaly, and brown. Needles are 1.0-2.0 mm long with a petiole at its base. Needles edges are untoothed, and tips are bluntly rounded and often notched. Its stomatal bands consist of 8-10 lines of stomates (Eckenwalder 2009). Numerous branches spread horizontally, drooping at the ends (Farjon 1990). Its conical crown appears more compact than eastern hemlock (Eckenwalder 2009).
Carolina hemlock is much more restricted in its range, elevation, and site type. It is uncommon, with scattered distribution lying mostly within western North Carolina
(Eckenwalder 2009). It can also be found in southwestern Virginia, northeastern Tennessee, northwestern South Carolina, and northeastern Georgia. Although its limited range overlaps the ubiquitous range of eastern hemlock, the two are not known to hybridize in nature.
Typically, sites for Carolina hemlocks are rocky, moist, and either north or east facing slopes and ridges, from 600-1500 m in elevation. Soils are usually sandy-clay loam soils, but recent investigations have suggested that it may occur on a wider range of soil textural groups than previously thought (Jetton et al. 2008a). In cool ravines, it can grow alongside streams. The associated climate is cool and humid, with over 1000 mm annual precipitation. It may grow singly or in mixed stands (Farjon 1990).
Phylogenetically, Carolina hemlock is more closely related to Chinese hemlock than its neighbor, eastern hemlock (Havill et al. 2008). The two have been successfully hybridized in breeding studies underway at the U.S. National Arboretum (Bentz et al. 2002), and the resulting hybrids exhibit resistance to HWA infestations (Montgomery et al. 2009).
Greenhouse infestation studies between pure species of eastern, Carolina, and western
21
hemlocks indicate that Carolina hemlock is likely less susceptible to HWA than eastern hemlock (Jetton et al. 2008b).
Hemlock Response to Hemlock Woolly Adelgid
The two hemlock species native to the eastern United States, eastern hemlock and
Carolina hemlock, are the only species worldwide which readily die as a result of HWA infestations. In the Pacific Northwest, HWA occasionally kill ornamental western hemlock, but it has never had a severe impact on the natural stands (McClure 1987). In Asia, HWA is considered a minor pest of native hemlock species (Bentz et al. 2002). The cause of hemlock death in the eastern United States is unknown, although it is suspected that a lack of natural enemies and host-plant resistance are responsible (McClure 1987). Havill et al. (2006) suggested that the differences in hemlock susceptibility may be the result of co-evolution.
Hemlocks in eastern North America were geographically isolated from the hemlock woolly adelgid, and conceivably evolved defenses against defoliators instead.
The hemlock woolly adelgid feeds on the xylem ray parenchyma cells of hemlock, depleting its nutrient stores (Young et al. 1995). The xylem ray is rich in carbohydrates and other essential nutrients; as adelgids feed, less is available to the tree for life functions such as growth, metabolism, reproduction, storage, and defense (Shigo 1989). Following infestation, susceptible hemlocks exhibit needle loss, bud abortion, and eventual branch and tree death (McClure 1991).
Specific knowledge regarding host response to HWA feeding is growing, but there is still much we need to understand. Several physiological changes are understood to occur.
Post-infestation, increased amounts and compositional differences in volatile monoterpenes
22
emitted from branch tips are evident (Broeckling and Salom 2003). Hemlock volatiles are important as attractants for a predator native to northwestern North America, Laricobius nigrinus Fender (Wallin et al. 2011). Young et al. (1995) suggested a systemic response to
HWA feeding was occurring, causing tree mortality. Recent work reveals increased levels of
H2O2 in eastern hemlock foliage in response to adelgid feeding, an indication that an induced hypersensitive response may be occurring (Radville et al. 2011). In addition, infested trees exhibit drought-like stress signs and abnormal wood production in the xylem is evident
(Walker-Lane 2009, Rivera et al. 2010). A comparable host response occurs in Fraser fir
(Abies fraseri (Pursh) Poir) trees as a response to feeding of the balsam woolly adelgid, A. piceae (Ratzeburg). Susceptible trees exhibit an increased production of heartwood and altered xylem, termed rotholtz (Balch 1964). Both of these play roles in obstructing water movement within the tree, ultimately causing tree death (Puritch 1971, 1977). The extent of the physiological stress, if any, in more resistant hemlock species is unknown.
Another invasive pest, the elongate hemlock scale (Fiorinia externa Ferris), also infests hemlocks in eastern North America. Also native to Japan, populations of this scale compete with HWA, although hemlock mortality is more strongly related to HWA population densities (Preisser et al. 2008).
Importance of Forest and Ornamental Hemlock in Eastern North America
The impressive hemlock is considered by many a beautiful and charismatic tree in both the landscape and forest setting and is a keystone species in its natural habitat where its longevity, shade tolerance, and crown density put it in a unique ecological niche (Quimby
1996). Hemlocks are shade-tolerant and long-lived, known to live over 900 years as major
23
components of old growth forest communities. As a foundation species, many ecosystems are affected by the distinctive, acidic microclimates created by hemlocks which affect faunal species composition, succession dynamics, primary productivity, decomposition, and nutrient cycling (Evans et al. 1996, Stadler et al. 2005). The hemlock occurs in riparian zones alongside streams and on inclines through the eastern U.S. Its shading and cooling effects on aquatic systems provide tolerable water temperatures for many animals, especially trout
(Evans et al. 1996). Brook trout (Salvelinus fontinalis Mitchill) and brown trout (Salmo trutta L.) are three times as prevalent in streams bordered by hemlocks than those bordered by hardwood (Ross et al. 2003). The canopy microclimate is also strongly associated with birds (Evans et al. 1996). They serve as an irreplaceable habitat and foraging site for many bird species, especially neotropical migratory birds. Many mammals, such as deer, browse hemlock foliage and use them as bedding shelter in the winter. They are also recognized cavity trees, providing habitat to native bears (Ward et al. 2004). The loss of the hemlock from the ecosystem would be devastating (Fidgen et al. 2006, Quimby 1996).
The aesthetic value of hemlock surpasses that of many trees in the east; it is planted around homes, cities, parks, and cemeteries, to name a few. There are 274 cultivars of eastern hemlock, “making it one of the most cultured and cultivated landscape tree species”
(Swartley 1984, Quimby et al. 1996). It is tall and beautiful and is used as a hedge because of its good reaction to shearing (USDA 1970). It is desired for its foliage color, graceful habit, and, until recently, its freedom from disease and insects (Eckenwalder 2009).
According to 1995 nursery inventories in Tennessee and North Carolina, the estimated value of Tsuga was approximately $34 million (Bentz et al. 2002). The wood of hemlock is
24
principally used as lumber (framing, roofing, subflooring, boxes, crates, sauna exteriors) and low-quality paper pulp (McClure 1995b, USDA 1970). Land value is also deteriorating as a result of HWA infestations. A case study in residential New Jersey found significant property value decline of over $7,000 as a result of 25-50% hemlock defoliation by HWA
(Holmes et al. 2005).
Exact consequences of total hemlock loss are unknown. Tree decline has already begun to affect infested areas. In just nine years, hemlocks in a study in the Delaware Water
Gap National Recreational Area declined in health producing the following effects: increase in average transmitted radiation from 5.0% to 11.7%, increase in vascular plant cover from
3.1% to 11.3%, and the presence of invasive plants in 35% of the permanent vegetation plots
(Eschtruth et al. 2006).
LITERATURE CITED
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac. Entomol. 1: 79-82.
Auclair, J.L. 1963. Aphid feeding and nutrition. Ann. Rev. Entomol. 8: 439-490.
Balch R.E, J. Clark, and J.M. Bonca. 1964. Hormonal action in production of tumours and compression wood by an aphid. Nature 202: 721-722.
Baliddawa, C.W. 1985. Insect behavior and host plant resistance. Insect Sci. Applic. 6: 337- 340.
Bentz, S.E., L.G.H. Riedel, M.R. Pooler, and A.M. Townsend. 2002. Hybridization and self-compatibility in controlled pollinations of eastern North American and Asian hemlock (Tsuga) species. J. Arboric. 28: 200-205.
25
Bentz, S.E., R.J. Griesbach, M.R. Pooler, and A.M. Townsend. 2007. Tsuga chinensis as a source of host resistance to the hemlock woolly adelgid, pp. 24-25. In K.W. Gottschalk (ed.), Proceedings: 17th U.S. Department of Agriculture Interagency Research Forum on Gypsy Moth and other Invasive Species, 10-13 January 2006, Annapolis, MD. U.S. Department of Agriculture, Forest Service, Northern Research Station, Gen. Tech. Rep. NRS-P-10, Newtown Square, PA.
Bompey, N. 2010. US Forest Service increases efforts to save hemlocks in Asheville area. Asheville Citizen Times. http://www.citizen- times.com/article/20101108/NEWS/311080020/1007/COLUMNISTS
Broeckling, C.D., and S.M. Salom. 2003. Volatile emission of eastern hemlock, Tsuga canadensis, and the influence of hemlock woolly adelgid. Phytochem. 62: 175-180.
Brown, M.W., T.E. Nebeker, and C.R. Honea. 1987. Thinning increases loblolly pine vigor and resistance to bark beetles. South. J. Appl. For. 11: 28-31.
Burd, J.D., R.A. Butts, N.C. Elliott, and K.A. Shufran. 1998. Seasonal development, overwintering biology, and host plant interactions of Russian wheat aphid (Homoptera: Aphdidiae) in North America, pp. 65-99. In S.S. Quisenberry, and F.B. Peairs ( eds), Response model for an introduced pest – the russian wheat aphid. Thomas Say Publications in Entomology, Lanham, MD.
Caswell, T., R. Casagrande, B. Maynard, and E. Preisser. 2008. Production and evaluation of eastern hemlocks potentially resistant to the hemlock woolly adelgid. In B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008, Hartford, CT. USDA Forest Service, FHTET- 2008-01, Morgantown, WV.
Cheah, C.A.S.-J., and M.S. McClure. 2000. Seasonal synchrony of life cycles between the exotic predator, Pseudoscymnus tsugae (Coleoptera: Coccinellidae) and its prey, the hemlock woolly adelgid Adelges tsugae (Homoptera: Adelgidae). Agric. For. Entomol. 2: 241-251.
Cheah, C.A.S.-J., M.E. Montgomery, S. Salom, B.L. Parker, S. Costa, and M. Skinner. 2004. Biological control of hemlock woolly adelgid. USDA Forest Service, FHTET-2004-04. Morgantown, WV.
Cheah, C.A.S.-J., M.A. Mayer, D. Palmer, T. Scudder, and R. Chianese. 2005. Assessments of biological control of hemlock woolly adelgid with Sasajiscymnus tsugae in Connecticut and New Jersey, pp. 116-130. In B. Onken, and R. Rheardon (eds.), Third Symposium on Hemlock Woolly Adelgid in the Eastern United States, 1-3 February 2005, Asheville, NC. USDA Forest Service, FHTET 2005-01, Morgantown, WV.
26
Cohen, A.C. 1995. Extra-oral digestion in predaceous terrestrial Arthropoda. Annu. Rev. Entomol. 40: 85-103.
Cohen, A.C. 1996. Plant feeding by predatory Hemiptera: evolutionary and adaptational aspects of trophic switching. pp 1-17. In R. Weidenmann, O. Alomar, and D. Livingstone (eds), Biological control by predatory Hemiptera. Thomas Say Publications in Entomology, Entomological Society of America, Lanham, MD.
Cohen, A.C. 1998. Solid-to-liquid feeding: the inside(s) story of extra-oral digestion in predaceous Arthropoda. Am. Entomol. 44: 103-117.
Cohen, A.C. 2004. Insect diets: Science and technology. CRC Press, Boca Raton, FL.
Costa, S. 2010. Hemlock woolly adelgid suppression with aerial application of mycotal fungus in a microfactory formulation: can it work?, pp. 36-42. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Cowles, R.S., and C.A.S.-J. Cheah. 2002. Systemic control of the hemlock woolly adelgid, 1999. Arthropod Manage. Tests 27: G48.
Cowles, R.S., M.E. Montgomery, and C.A.S.-J. Cheah. 2006. Activity and residues of imidacloprid applied to soil and tree trunks to control hemlock woolly adelgid (Hemiptera: Adelgidae) in forests. J. Econ. Entomol. 99: 1258-1267.
Dallimore, W., and A.B. Jackson. 1948. A Handbook of Coniferae, including Ginkoaceae, 3rd ed. Edward Arnold and Co., London, UK.
Davis, F.M. 1985. Entomological techniques and methodologies used in research programmes on plant resistance to insects. Insect Sci. Applic. 6: 391-400.
Del Tredici, P., and A. Kitajima. 2004. Introduction and cultivation of Chinese hemlock (Tsuga chinensis) and its resistance to hemlock woolly adelgid (Adelges tsugae). J. Arboriculture 30: 282-287.
Dilling, C., P. Lambdin, J. Grant, and R. Rhea. 2009. Community response of insects associated with eastern hemlock to imidacloprid and horticultural oil treatments. Environ. Entomol. 38: 53–66.
27
Dilling, C., P. Lambdin, J. Grant, and R. Rhea. 2010. Through space and time: an evaluation of the translocation of imidacloprid in eastern hemlocks, p. 84-89. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States; 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Dirr, M.A. 1998. Manual of woody landscape plants: their identification, ornamental characteristics, culture, propagation and uses, 5th ed. Stipes Publishing, Champaign, IL.
Dixon, A.G. 1975. Aphids and translocation, pp. 154-170. In M.H. Zimmermann, and J.A. Milburn (eds), Transport in plants. Springer, Berlin, Germany.
Doccola, J.J., W. Hascher, J.J. Aiken, and P.M. Wild. 2010. A combined treatment strategy using imidacloprid to facilitate hemlock recovery in hemlock woolly adelgid infested trees, p. 75-83. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States; 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Eckenwalder, J.E. 2009. Conifers of the World. Timber Press, Portland, OR.
Eschtruth, A.K., N.L. Cleavitt, J.J. Battles, R.A. Evans, and T.J. Fahey. 2006. Vegetation dynamics in declining eastern hemlock stands: 9 years of forest response to hemlock woolly adelgid infestation. Can. J. For. Res. 36: 1435-1450.
Evans, A.M., and T.G. Gregoire. 2006. A geographically variable model of hemlock woolly adelgid spread. Biol. Invasions 9: 369-382.
Evans, R.A., E. Johnson, J. Shreiner, A. Ambler, J. Battles, N. Cleavitt, T. Fahey, J. Sciascia, and E. Pehek. 1996. Potential impacts of hemlock woolly adelgid (Adelges tsugae) on eastern hemlock (Tsuga canadensis) ecosystems. In S. M. Salom, T. C. Tigner, and R. C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review, 12 October 1995, Charlottesville, VA. USDA Forest Servie, FHTET 96-10, Morgantown, WV.
Farjon, A. 1990. Pinaceae. Drawings and descriptions of the genera Abies, Cedrus, Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea. Koeltz Scientific Books, Konigstein, Germany.
Faulkenberry, M., R. Hedden, and J. Culin. 2009. Hemlock susceptibility to hemlock woolly adelgid attack in the Chattooga River watershed. Southeastern Nat. 8: 129-140.
Fidgen, J.G., D.E. Legg, and S.M. Salom. 2006. Binomial sequential sampling plan for hemlock woolly adelgid (Hemiptera: Adelgidae) sistens’ infesting individual eastern hemlock trees. J. Econ. Entomol. 99: 1500-1508.
28
Forbes, A.R. 1966. Electron microscopic evidence for nerves in the mandibular stylets of the green peach aphid. Nature 212: 726.
Furniss, R.L., and V.M. Carolin. 1977. Western Forest Insects. USDA For. Serv. Misc. Publ. 1339, Lanham, MD.
Godman, R.M., and K. Lancaster. 1990. Eastern hemlock. In R.M. Burns, and B.H. Honkala (eds.), Silvics of North America. USDA Forest Service, Agriculture Handbook 654, Washington, D.C.
Gouger, R.J. 1971. Control of Adelges tsugae on hemlock in Pennsylvania. Sci. Tree Topics 3: 1-9.
Grant, J.F. 2008. Establishing Sasajiscymnus tsugae in the South, pp. 39-44. In: B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Grant, J.F., A. Hakeem, R. Rhea, G. Wiggins, P.L. Lambdin, and G. Taylor. 2010. Recovery and establishment of introduced predators of hemlock woolly adelgid in the Southern Appalachians, pp. 21-23. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Gray, D.R., and S.M. Salom. 1996. Biology of the hemlock woolly adelgid in the southern Appalachians, pp. 16-25. In S.M. Salom, T.C. Tigner, and R.C. Rheardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review, 12 October 1995. USDA Forest Service, FHTET96-10, Morgantown, WV.
Hale, F.A. 2004. The hemlock woolly adelgid: a threat to hemlock in Tennessee. The University of Tennessee: Agricultural Extension Service. SP503-G.
Havill, N.P., M.E. Montgomery, G. Yu, S. Shiyake, and A. Caccone. 2006. Mitochondrial DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic speciation and pinpoints the source of the introduction to eastern North America. Ann. Entomol. Soc. Am. 99: 195-203.
Havill, N.P., C.S. Campbell, T.F. Vining, B. LePage, R.J. Bayer, and M.J. Donoghue. 2008. Phylogeny and biogeography of Tsuga (Pinaceae) inferred from nuclear ribosomal ITS and chloroplast DNA sequence data. Syst. Bot. 33: 478-489.
29
Havill, N., G. Davis, S. Salom, J. Klein, and A. Caccone. 2010. Population genetics of Laricobius nigrinus and L. rubidus, predators of adelgids in North America, pp. 5-6. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET- 2010-07, Morgantown, WV.
Hoover, K., K. Cassidy, and M. Keena. 2010. Rearing and host-range studies of Scymnus camptodromus, pp. 29-30. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Ingwell, L.L., and E.L. Preisser. 2010. Using citizen science programs to identify host resistance in pest-invaded forests. Conserv. Biol. 25: 182-188.
Iosue, F.V. 2008. The role of phytochemicals and leaf cushion morphology in host resistance to the hemlock woolly adelgid (Adelges tsugae). M.S. thesis, Villanova University, Villanova, PA.
Jetton, R.M., W.S. Dvorak, and W.A. Whittier. 2008a. Ecological and genetic factors that define the natural distribution of Carolina hemlock in the southeastern United States and their role in ex situ conservation. Forest Ecol. Manag. 255: 3212–3221.
Jetton, R.M., F.P. Hain, W.S. Dvorak, and J. Frampton. 2008b. Infestation rate of hemlock woolly adelgid (Hemiptera: Adelgidae) among three North American hemlock (Tsuga) species following artificial inoculation. J. Entomol. Sci. 43: 438-442.
Jiang, Y. 1996. Oxidative interactions between the spotted alfalfa aphid (Therioaphis trifolii maculata) (Homoptera: Aphididae) and the host plant Medicago sativa. Bull. Entomol. Res. 86: 533–540.
Joseph, S.V., J.L. Hanula, S.K. Braman, and F. Byrne. 2010. Impact of fertilizer and imidacloprid on Adelges tsugae (Hemiptera: Adelgidae), p. 74. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
Keen, F.P. 1938. Insect enemies of western forests. USDA Miscellaneous Publication 273.
30
Keena, M., and M. Montgomery. 2010. Effects of temperature on the eggs and larvae of Scymnus camptodromus, pp. 25-28. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Kloft, W., and P. Ehrhardt. 1962. Studies on the assimilation and excretion of labeled phosphate in aphids, pp. 181-190. In Radioisotopes and radiation in rntomology, International Atomic Energy Agency, Vienna, Austria.
Lagalante, A.F., N. Lewis, M.E. Montgomery, and K.S. Shields. 2006. Temporal and spatial variation of terpenoids in eastern hemlock (Tsuga canadensis) in relation to feeding by Adelges tsugae. J. Chem. Ecol. 32: 2389-2403.
Lamb, A.B., S.M. Salom, L.T. Kok, and D.L. Mausel. 2006. Confined field release of Laricobius nigrinus (Coleoptera: Derodontidae), a predator of the hemlock woolly adelgid, Adelges tsugae (Hemiptera: Adelgidae), in Virginia. Can. J. For. Res. 36: 369-375.
Lamb, A., L. Cota Viera, S. Shiyake, H. Story, T. McAvoy, and S. Salom. 2010. Suitability of Laricobius osakensis as a potential biological control agent of hemlock woolly adelgid, pp. 17-19. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
LePage, B.A. 2003. A new species of Tsuga (Pinaceae) from the middle Eocene of Axel Heiberg Island, Canada, and an assessment of the evolution and biogeographical history of the genus. Bot. J. Linn. Soc. 141: 257-296.
Mausel, D.L., S.M. Salom, L.T. Kok, J.G. Fidgen. 2008. Propagation, synchrony, and impact of introduced and native Laricobius spp. (Coleoptera: Derodontidae) on hemlock woolly adelgid in Virginia. Environ. Entomol. 37: 1498-1507.
McAllan, J.W., and J.B. Adams. 1961. The significance of pectinase in plant penetration by aphids. Can. J. Zoo. 39: 305-310.
McAvoy, T.J., G.M.G. Zilahi-Balogh, S.M. Salom, L.T. Kok, and Z. Guoliang. 2007. Development and feeding of Tetraphleps galchanoides, a predator of the hemlock woolly adelgid. Biocontrol 52: 339-350.
McClure, M.S. 1987. Biology and control of hemlock woolly adelgid. Bull. Conn. Agric. Exp. Stn. 851.
McClure, M.S. 1989. Evidence of a polymorphic life cycle in the hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae). Ann. Entomol. Soc. Am. 82: 50-54.
31
McClure, M.S. 1990. Role of wind, birds, deer, and humans in dispersal of hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 19: 36-43.
McClure, M.S. 1991. Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis. Environ. Entomol. 20: 258-264.
McClure, M.S. 1992. Hemlock woolly adelgid. Am. Nurseryman 175: 82-89.
McClure, M.S. 1995a. Diapterobates humeralis (Oribatida: Ceratozetidae): An effective control agent of hemlock woolly adelgid (Homoptera: Adelgidae) in Japan. Environ. Entomol. 24: 1207-1215.
McClure, M.S. 1995b. Managing hemlock woolly adelgid in ornamental landscapes. Conn. AES Bull. 925.
McClure, M.S. 1996. Biology of Adelges tsugae and its potential for spread in the northeastern United States. pp. 16-25 In S.M. Salom, T.C. Tigner, and R.C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review, 12 October 1995. USDA Forest Service, FHTET 96-10, Morgantown, WV.
McClure, M.S. 2001. Biological control of hemlock woolly adelgid in the eastern United States. FHTET-2000-08. USDA Forest Service, Forest Health Technology Enterprise Team, Morgantown, WV.
Miles, P.W. 1968. Insect secretions in plants. Ann. Rev. Phytopath. 6: 137-164.
Miles, P.W. 1972. The saliva of Hemiptera. Adv. Insect Physiol. 9: 183-255.
Miles, P.W. 1986. Feeding process of Aphidoidea in relation to effects on their food plants, pp. 321-339. In A.K. Minks, and P. Harrewijn (eds.), Aphids: their biology, natural enemies and control. Elsevier, Amsterdam, The Netherlands.
Miles, P.W. 1999. Aphid saliva. Biol. Rev. Camb. Philos. Soc. 74: 41-85.
Montgomery, M.E., and S.M. Lyon. 1996. Natural enemies of adelgids in North America: their prospect for biological control of Adelges tsugae (Homoptera: Adelgidae). In S.M. Salom, T.C. Tigner, and R.C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review, 12 October 1995. USDA Forest Service, FHTET 96-10, Morgantown, WV.
32
Montgomery, M.E., D. Yao, and H. Wang. 2000. Chinese Coccinelidae for biological control of the hemlock woolly adelgid: description of native habitat. In K. A. McManus, K. S. Shields, and D. R. Souto (eds.), Proceedings: Symposium on Sustainable Management of Hemlock Ecosystems in Eastern North America, 22-24 June 1999, Durham, NH. USDA Forest Service, GTR-NE-267, Newtown Square, PA.
Montgomery, M.E., S.E. Bentz, and R.T. Olson. 2009. Evaluation of hemlock (Tsuga) species and hybrids for resistance to Adelges tsugae (Hemiptera: Adelgidae) using artificial infestation. J. Econ. Entomol. 102: 1247-1254.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid. http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
Painter, R.H. 1951. Insect resistance in crop plants. Macmillan, New York, NY.
Parker, B.L., M. Skinner, S. Gouli, T. Ashikaga, and H.B. Teillon. 1999. Low lethal temperature for hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 28: 1085-1091.
Peng, Z., and P.W. Miles. 1988. Studies on the salivary physiology of plant bugs: function of the catechol oxidases of the rose aphid. J. Insect. Physiol. 10: 1027-1033.
Peng, Z., and P.W. Miles. 1991. Oxidases in the gut of an aphid, Macrosiphum rosae (L.) and their relation to dietary phenolics. J. Insect Physiol. 37: 779-787.
Puritch, G.S. 1971. Water permeability of the wood of Grand fir (Abies grandis (Doug.) Lindl.) in relation to infestation by the balsam woolly aphid, Adelges piceae (Ratz.) J. Exp. Bot. 22: 936-945.
Puritch, G.S. 1977. Distribution and phenolic composition of sapwood and heartwood in Abies grandis and the effects of the balsam woolly aphid. Can. J. For. Res. 7: 54-62.
Pontius, J.A., R.A. Hallett, and J.C. Jenkins. 2006. Foliar chemistry linked to infestation and susceptibility to hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 35: 112-120.
Preisser, E., A. Lodge, D. Orwig, and J. Elkington. 2008. Range expansion and population dynamics of co-occurring invasive herbivores. Biological Invasions 10: 201-213.
33
Quimby, J.W. 1996. Value and importance of hemlock ecosystems in the eastern United States. In S. M. Salom, T. C. Tigner, and R. C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review; 12 October 1995, Charlottesville, VA. USDA Forest Service, FHTET 96-10, Morgantown, WV.
Radville, L., A. Chaves, and E.L. Preisser. 2011. Variation in plant defense against invasive herbivores: evidence for a hypersensitive response in eastern hemlocks (Tsuga canadensis). J. Chem. Ecol. 37: 592-597.
Reynolds, W.N. 2008. Imidacloprid insecticide treatments for hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae), affect a non-target soil arthropod community surrounding eastern hemlock, Tsuga canadensis Carriere. M.S. thesis. University of Tennessee, Knoxville.
Rivera, L.N., J.C. Domec, J. Frampton, F. Hain, I. Peszlen, and J.S. King. 2010. The effect of hemlock woolly adelgid (HWA) infestation on water relations of Carolina and eastern hemlock: Can ecophysiological investigation of tree water relations improve silvicultural management of HWA?, p. 129-131. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Ross, D.W., S.M. Grubin, G.R. Kohler, and K.F. Wallin. 2010. Evaluation of Leucopis species (Diptera: Chamaemyiidae) from the PNW as potential biological control for the hemlock woolly adelgid, p. 32. In B. Onken and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Ross, R.M., R.M. Bennett, C.D. Snyder, J.A. Young, D.R. Smith, and D.P. Lemarie. 2003. Influence of eastern hemlock (Tsuga canadensis L.) on fish community structure and function in headwater streams of the Delaware River basin. Ecology of Freshwater Fish 12: 60-65.
Salom, S.M., A.A. Sharov, W.T. Mays, and J.W. Neal, Jr. 2001. Evaluation of aestival diapauses in hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 30: 877- 882.
Salom, S., L. Kok, A. Lamb, C. Jubb, and B. Onken. 2008. Biological control of hemlock woolly adelgid: what is it going to take to make it work?, pp. 11-17. In B. Onken, and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008, Hartford, CT.. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
34
Shigo, A.L. 1989. A new tree biology: facts, photos, and philosophies on trees and their problems and proper care, 2nd ed. Shigo and Trees, Associates, Durham, NH.
Skinner, M., B.L. Parker, S. Gouli, and T. Ashikaga. 2003. Regional responses of hemlock woolly adelgid (Homoptera: Adelgidae) to low temperatures. Environ. Entomol. 32: 523-528.
Souto, D., T. Luther, and B. Chianese. 1996. Past and current status of HWA in eastern and Carolina hemlock stands, pp. 9-15. In S. M. Salom, T. C. Tigner, and R. C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review; 12 October 1995, Charlottesville, VA. USDA Forest Service, FHTET 96-10, Morgantown, WV.
Srivastava, P.N. 1987. Nutritional physiology, pp. 99-121. In A.K. Minks, and P. Harrewijn (eds), Aphids their biology, natural enemies and control, vol. 2A. Elsevier, Amsterdam, The Netherlands.
Stadler, B., T. Müller, D. Orwig, and R. Cobb. 2005. Hemlock woolly adelgid in New England forests: canopy impacts transforming ecosystem processes and landscapes. Ecosystems 8: 233-247.
Steward, V.B., and T.A. Horner. 1994. Control of hemlock woolly adelgid using soil injections of systemic insecticides. J. Arboric. 24: 54-56.
Swartley, J.C. 1984. The Cultivated Hemlocks. Timber Press, Portland, OR.
Townsend, L. and L. Rieske-Kinney. 2006. Meeting the threat of the hemlock woolly adelgid. University of Kentucky: Cooperative Extension Service. ENTFACT-452.
USDA. 1970. Eastern hemlock. USDA, Forest Service, American Woods FS-239.
USDA Forest Service. 2011. Counties with established HWA populations 2010. http://na.fs.fed.us/fhp/hwa/maps/2010.pdf. van Gelderen, D.M., and J.R.P. van Hoey Smith. 1986. Conifers: The Illustrated Encyclopedia. Timber Press, London, England. von Dohlen, C.D. 2008. Endosymbionts of hemlock woolly adelgid, p. 140. In B. Onken. and R. Reardon (eds.), Proceedings of the Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008, Hartford, CT. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
35
Walker-Lane, L.N. 2009. The effect of hemlock woolly adelgid infestation on water relations of Carolina and eastern Hemlock. M.S. thesis. North Carolina State University, Raleigh.
Wallace, M.S., and F.P. Hain. 2000. Field surveys and evaluation of native and established predators of the hemlock woolly adelgid (Homoptera: Adelgidae) in the southeastern United States. Environ. Entomol. 29: 638-644.
Wallin, K., and D. Ott. 2010. Using behavioral assays to develop fitness-based quality control for the HWA predator, Scymnis conifarum, p. 31. In B. Onken, and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 Augus 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Wallin, K.F., T.M. Latty, and D.W. Ross. 2011. Orientation behavior of the predator Laricobius nigrinus (Coleoptera: Derodontidae) to hemlock woolly adelgid and host tree odors in a multi-chambered olfactometer. Environ. Entomol. 40: 864-872.
Ward, J.S., M.E. Montgomery, C.A.S-J. Cheah, B.P. Onken, and R.S. Cowles. 2004. Eastern hemlock forests: guidelines to minimize the impacts of hemlock woolly adelgid. USDA Forest Service. NA-TP-03-04. Morgantown, WV.
Webb, R.E., J.R. Frank, and M.J. Raupp. 2003. Eastern hemlock recovery from hemlock woolly adelgid damage following imidacloprid therapy. J. Arboric. 29: 298-302.
Webster, J. 2010. Management of hemlock woolly adelgid in Great Smoky Mountains National Park, p. 57-60. In B. Onken, and R. Reardon (eds.), Fifth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 17-19 Augus 2010, Asheville, NC. USDA Forest Service, FHTET-2010-07, Morgantown, WV.
Young, R.F., Shields, K.S. and G.P. Berlyn. 1995. Hemlock woolly adelgid (Homoptera: Adelgidae): stylet bundle insertion and feeding sites. Ann. Entomol. Soc. Am. 88: 827-835.
Zaslavski, V.A. 1988. Insect Development: photoperiodic and temperature control. Springer, Berlin, Germany.
Zilahi-Balogh, G.M.G., L.M. Humble, A.B. Lamb, S.M. Salom, and L.T. Kok. 2003. Seasonal abundance and synchrony between Laricobius nigrinus Fender (Coleoptera: Derodontidae) and its prey, the hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae) in British Columbia. Can. Entomol. 135: 103-115.
36
Figure 1. Distribution of the hemlock woolly adelgid (Adelges tsugae) in the eastern United States as of 2010. Map produced by the USDA Forest Service.
37
Figure 2. Stylet penetration by members of Aphidoidea through a transverse section of plant tissue. Key to groups: (a) Salivary track resulting from intercellular stylet penetration of non-vascular tissue, e.g. Adelges spp., Aphis fabae Scopoli, Macrosiphum euphorbiae (Thomas); (b) track left by brief intercellular probe; (c) intracellular beaded epidermal track and intercellular compressed track into parenchyma, e.g. Aphis gossypii Glover, Myzus persicae (Sulzer); (d) track resulting from stomatal entry; (e-g) intracellular tracks as produced in cortical tissues of stems and roots. Figure taken from Miles (1986).
38
Figure 3. World-wide distribution of hemlocks (Tsuga spp.). Key to scientific and common names: (1) Tsuga canadensis (L.) Carrière (eastern hemlock; (2) T. caroliniana Engelm. (Carolina hemlock); (3) T. heterophylla (Raf.) Sarg. (western hemlock); (4) T. mertensiana (Bong.) Carrière (mountain hemlock); (5) T. sieboldii Carrière (southern Japanese hemlock); (6) T. diversifolia (Maxim.) Mast. (northern Japanese hemlock); (7) T. chinensis (Franch.) E. Pritz. (Chinese hemlock); (8) T. forrestii Downie (Forrest hemlock); (9) T. dumosa (D. Don) Eichler. Map taken from LePage (2003).
39
Chapter II
Morphological Observations of Hemlock Woolly Adelgid Host-plant Interactions using Light and Electron Microscopy
ABSTRACT
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an invasive pest of eastern and Carolina hemlocks in the eastern U.S. This research focuses on the basic biology of HWA, with a concentration on feeding behaviors, using light, scanning electron, and transmission electron microcopy. We corroborate insertion sites identified by Young et al. (1995) and also identify alternative insertion sites, on the woody stem and through stomata of the needle, used when infestations are high.
Morphological comparisons are made between the crawler, adult, and winged stage, and between the western and eastern populations of North American HWA. The crawler possesses appendages equipped with unidentified sensilla and setae that likely assist its host- plant selection behavior. Stylet bundle morphology is observed, revealing five dentitions on the mandibular stylets. Cross-sections of the stylet indicate that HWA stylet bundle morphology is similar to that of aphids (Hemiptera: Aphididae) and the balsam woolly adelgid, Adelges piceae (Ratz.), with two separate canals (a salivary and a food canal) and neurons within tubules of the mandibular stylets. The salivary canal was 0.24 - 0.54 µm in diameter and the food canal measured 0.48 - 1.0 µm, representing the limiting diameter width for fodd ingestion.
40
INTRODUCTION
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is a major pest of eastern forests, causing extensive mortality to eastern and
Carolina hemlocks (Tsuga canadensis (L.) Carriére and T. caroliniana Engelmann). Native to eastern Asia and the Pacific Northwest, HWA was first recorded in the eastern United
States in 1951, presumably through the introduction of infested nursery stock (McClure and
Cheah 1999, Havill et al. 2006, Lambdin et al. 2006). Its range now includes 18 eastern states, where hemlocks are susceptible and 80-90% mortality of infested stands occurs (Hale
2004, Townsend & Rieske-Kinney 2006, USDA-FS 2011). Signs of HWA infestation of hemlock include a loss of vigor, inhibition of new growth, premature bud abortion, needle drop, and in many cases, tree death in four to ten years (McClure 1987, 1991, Fidgen et al.
2006). In the native range of HWA, hemlocks do not succumb to infestations. Although infrequent, some individual trees and populations of eastern and Carolina hemlock exhibit less susceptibility to HWA; sources of such resistance are unknown (Jetton et al. 2008, Kaur
2009, Oten et al. 2011).
Its status as a significant pest of trees important in forest ecosystems and the ornamental industry has prompted numerous research projects focusing on population management. Despite being a focus of research for over 20 years, the biology of HWA remains understudied. Adelgids belong to the superfamily Aphidoidea, are regularly labeled
“aphid-like,” and HWA is referred to as the hemlock woolly aphid in some texts (e.g., Pirone
1978, MacKenzie 2002, Speight 2007). Because of these associations, much of our
41
knowledge of the basic biology of HWA is based on a more comprehensive understanding of aphid biology, despite several major biological differences.
The feeding biology of HWA was investigated by Young et al. (1995), and their research produced conclusions that constitute most of what we know about HWA feeding behaviors. Young et al. (1995) ascertained that the stylet bundle insertion point occurs on the adaxial side of eastern hemlock needles, proximal to the twig with respect to the abscission layer. In addition, stylet length was measured and determined it to be almost three times the length of a crawler itself, possibly having to do with its feeding site in xylem ray parenchyma cells. This extensive stylet length is not unusual within the family Adelgidae and contrasts with the much shorter stylets of aphids. In the balsam woolly adelgid (A. piceae (Ratz.)), stylet penetration reaches a depth of 1.6 mm to reach parenchyma cells in a fir tree; penetration by the cotton aphid (Aphis gossypii Glover) is shallower, penetrating depths of
106-118 µm in its herbaceous host plant (Kloft 1955, Pollard 1958). Although lacking confirmation with a stylet cross-section image, Young et al. (1995) deduced that the stylet bundle morphology of HWA is morphologically similar to aphids based on lateral views of the stylet bundle fitted together. The aphid stylet bundle consists of two pairs of sclerotized stylets: a mandibular pair (external) and a maxillary pair (internal), which fit together via interlocking grooves to form two canals that transport saliva down to the plant and nutrient from the plant to the aphid (Fig. 4A) (Parrish 1967). This description also describes the stylets of the balsam woolly adelgid, A. piceae (Ratz.) (Fig. 4B), a close relative to HWA and another invasive adelgid, responsible for mortality of different fir species (Forbes and
Mullick 1970). Similarly, the maxillary stylets of HWA lie within grooves on the
42
mandibular stylets and the entire bundle is retracted from the plant during molting in order to also shed the stylet cuticle (Young et al. 1995, McClure 2001). Aphids mechanically pierce the plant cuticle with an alternating sawing motion of these four individual stylets to feed on the phloem of the plant (Pollard 1973). Conversely, HWA feeds on the xylem ray parenchyma cells (Young et al. 1995), storage tissues of the plant. Mode of entry to internal plant tissues for HWA is unstudied.
Knowledge of the external morphology of HWA is restricted to the original description of the insect (Annand 1924). However, this description is based on the population infesting western hemlock (T. heterophylla (Raf.) Sarg.) in northwestern North
America where HWA is endemic and does not kill its host trees, western and mountain hemlock (T. mertensiana (Bong.). Also, this older description is limited by lack of high magnification, high resolution instrumentation such as the scanning electron microscope.
The population infesting and killing hemlocks in the eastern U.S. originated from southern
Japan and, although not fatal in its native range, successfully infests and kills hemlocks in the eastern U.S. (Havill et al. 2006). It has since been suggested that these two populations are distinct species (Havill et al. 2008), but morphological comparisons between the two populations have not been done.
Crawlers, the host-seeking stage, lack wings and compound eyes. Consequently, transportation between host trees is entirely passive: dispersal occurs either by crawling to nearby branches, wind, phoresy on macro-vertebrates (e.g., birds, deer, humans) or vehicles, or by human movement of infested stock (Butin et al. 2007, McClure 1990). Therefore, habitat and host location, the initial two steps in host-plant selection by phytophagous insects
43
(Bernays and Chapman 1994), does not occur as part of the host selection behaviors of
HWA.
As in aphids, two types of saliva are produced from the salivary glands of HWA: the watery saliva and the salivary sheath material (Miles 1959a, 1968, 1972, Pollard 1973,
Young et al. 1995). The salivary sheath material is produced as a viscous bead at the tip of the stylets and begins to solidify, enclosing the stylets in a sleeve of the material (Miles 1968,
1972). As the bead solidifies, the stylets push through, creating another bead. This continues and, unless constricted by substrate, will take on a beaded appearance (McLean and Kinsey
1965). The sheath material and it associated flange may be produced multiple times during host exploration, prior to ingestion (Tjallingii 1994). Following stylet insertion, the sheath material remains intact even after stylet removal. The remaining stylet sheaths represent a history of the movement of the stylet bundle within the plant and can be used to determine probing behavior (Cohen et al. 1998).
An intact sheath is necessary for successful feeding of many homopteran insects
(Cohen et al. 1998). Therefore, it has received attention from those exploring feeding behaviors of pestiferous phytophagous insects. As a result, many hypotheses have been formed to explain the function of the salivary sheath. Suggested functions include an increase in sucking force during feeding, as lubrication or a mechanical advantage for stylet maneuvering and insertion, enabling stylet bundle reinsertion following a molt, to prevent leakage during the ingestion process, and to protect the insect against host-plant defenses
(Arnaud 1918, Pollard 1973, Smith 1985, Cohen 1990, Tjallingii and Hogen Esch 1993,
Miles 1999). Cohen (1990) discusses the use of a sheath flange on the surface of prey
44
exclusively used to consume solid food. He asserted that this assists labial tip stabilization and acts as a fulcrum for stylet maneuvering, an advantage when considering the musculature used to manipulate the stylet bundle. Miles (1999) rejected some of these hypotheses, namely theories by Arnaud (1918), Pollard (1973), Smith (1985), and Mittler (1988). The theory by Arnaud (1918) may be the earliest theory described for sheath function. He suggested the primary function of the sheath was to prevent leakage of sap from the stylet canals and/or to prevent unwanted materials from entering the canals. However, Miles
(1999) suggested that this was unlikely because the four stylets that make up the stylet bundle fit very snugly together and no evidence supports the need to seal the stylets for the passive uptake of nutrients by aphids. Theories by Pollard (1973) and Smith (1985) prescribe a physical function to the sheath. They suggest it acts to keep the stylet bundle together and as a lubricant for the sliding action of the stylet bundle, respectively. Miles
(1999) rebukes these theories based on the successful feeding of other phytophagous hemipterans without the use of a salivary sheath (e.g., Miles 1959b). Mittler (1988) suggests the salivary sheath serves to prevent loss of sieve tube sap, which Miles (1999) discards based on observations that sap does not exude from the opening when aphids are brushed off, with no time to seal the tip of the sheath. Rather, Miles (1999) supports the theory that the stylet sheath is likely involved in a biochemical function, insulating the host plant from the deleterious effects of feeding. He also suggests that sheath material plugs damage to cell walls, providing a surface biochemically sufficient for membrane repair by the host
(Tjallingii and Hogen Esch 1993), preventing the release and/or production of autotoxic
45
compounds involved in defensive reactions. Additionally, evidence indicates that sheath material can absorb and immobilize phenolic compounds and oxidases (Miles 1987, 1999).
Understanding the feeding biology of HWA is important to increase our knowledge of its basic biology, to aid in the development of an artificial diet for applications to research and biological control, and to investigate mechanisms of host-plant resistance. Much of this fundamental knowledge remains unknown and what we do know has been based off a single publication, Young et al. (1995). The objectives of our study are to examine stylet bundle morphology, determine alternative stylet bundle insertion points, examine the sheath material, and compare western North American populations of HWA with eastern North
American populations of HWA.
MATERIALS AND METHODS
Branches of infested T. canadensis were field collected from locations in Crossnore,
NC and Laurel Springs, NC. Material was transported to Raleigh, NC, trimmed, and kept in water at 22°C and approximately 65% humidity. Materials from western populations of
HWA and hemlocks were collected in Olympia, WA, bagged in quart sized Ziploc® bags, and shipped to the NCDA&CS Plant Industry Quarantine Facility (Cary, NC). Samples were shipped overnight with a bag of ice and appropriate permit for interstate movement of live plant pests and noxious weeds (APHIS-PPQ, permit no. P526P-08-03385).
Light microscopy
Samples were observed with a Meiji Stereomicroscope (Meiji Microscope, Osaka,
Japan) and imaged with a Canon EOS T2i Digital SLR (Canon, Toronto, Ontario, Canada).
46
Images of scale bars were taken at equal magnifications of each image to ensure scale bar accuracy. Images were modified for brightness and contrast by adjusting levels, sized, and placed together to produce figures using Adobe Photoshop 8.0.
Scanning electron microscopy (SEM)
All plant and insect material was fixed in 3.0% glutaraldehyde in 0.1 M Sorenson’s buffer, pH 7.0, for 48 hours. Samples were postfixed in 1% osmium tetroxide in the same buffer for 1 hour at 4°C, then dehydrated in ethanol baths of 30%, 50%, 70%, 95% and 100%
(3X) for 1 hour each at 4°C. Samples were critical-point dried, mounted, and coated with gold-palladium at a deposition thickness of 20-25Å on four sides and top. Samples were examined with a JEOL JSM-5900LV scanning electron microscope (JEOL, Tokyo, Japan) at an accelerating voltage of 15 or 20 kV. Measurements were taken with a planimeter on printed images. Images were modified for brightness and contrast by adjusting levels, sized, and placed together to produce figures using Adobe Photoshop 8.0.
Transmission electron microscopy (TEM)
Infested eastern hemlock material was sent via overnight mail to Charles A. Murphy
(USDA-ARS, Beltsville Agricultural Research Center, Beltsville, MD), who performed and imaged the TEM portion of this study. Tissue was fixed overnight at room temperature by immersion in fixative comprised of 3% glutaraldehyde, 2% paraformaldehyde in 0.05M
NaCacodylate Buffer, pH 7.0. They were washed in a 0.1M NaCacodylate buffer rinse, six times over one hour, post fixed in NaCacodylate buffered 2% osmium tetroxide for 2 hours at room temperature, dehydrated in an acetone series and slowly infiltrated with Spurrs low- viscosity embedding resin. Ninety nm gold sections of the tissue were cut on a Reichert/AO
47
Ultracut microtome with a Diatome diamond knife and mounted onto 200 mesh Ni grids.
They were stained with 4% uranyl acetate and 3% lead citrate then viewed in a HT-7700
Hitachi Microscope at 80kV. Images were modified for brightness and contrast by adjusting levels, sized, and placed together to produce figures using Adobe Photoshop 8.0.
RESULTS
HWA feeding sites
Crawlers most consistently insert at the base of hemlock needles (Figure 1A, C).
However, under conditions of heavy infestations, when populations reach two or more HWA per needle, they may insert their stylet bundle into the needle itself or into the woody stem
(Figure 1D, E). Those that settle on the needle insert their stylets through stomatal openings at inconsistent distances from the stem, as viewed by pulling the crawler away from the surface of the plant and observing the stylet enter the stomata. Crawlers that insert their stylet bundle into the woody stem may survive and produce eggs. This was observed visually, by seeing live ovisacs on live HWA settled on woody stem tissue of infested material. Hemlock buds occur throughout the plant, positioned near the characteristic feeding site of HWA, at the needle bases. The presence of buds does not deter HWA feeding; conversely, multiple (three or more) HWA may insert near bud bases (Figure 1B).
External morphology
There are distinct morphological differences between HWA crawlers and HWA adults. Crawlers are smaller (body length: 0.357 ± .036 mm, n=10), have conspicuous antennae (antenna length: 98.680 ± 8.363 µm, n=10), stylet bundle is not yet extended from
48
the labium, and they lack prominent wax pores (Figure 2A-B). Crawlers have obvious sensilla or setae at the tips of their antennae and legs. Adults are larger and have comparatively smaller antennae, drawn up close to their body. Their stylet bundle extends from the labium (it is unlikely that they are able to withdraw it in a manner that other hemipterans can withdraw stylets) and prominent wax pores on the dorsal sclerites of its body (Figure 2C). During SEM sample preparation, stylets separated but remained outside the adelgid since they are not withdrawn once they are fully extended. They have inconspicuous setae or sensilla, relative to crawlers. Both crawlers and adult have three ocelli per side. Adults possess a three-pronged ovipositor at their posterior end. The winged sexupara are morphologically very distinct. They are equipped with wings and compound eyes with at least 77 ommatidia per eye.
Labium and stylet bundle morphology
The HWA labium possesses several short sensilla at the proximal end and subapically
(Fig. 3A). From it, the four stylets arise. The stylet bundle of HWA is composed of four pieces: two maxillary stylets inside two mandibular stylets (Fig. 3B, Fig. 4C). The tip of the insect’s mandibular stylet has at least five ridge-like dentitions, pointing towards the labium and body (Fig. 3C). A food canal, salivary canal, and neural canals are evident in the stylet bundle cross-section, observed with TEM (Fig. 4C). The diameter of the food canal is
0.89+0.14, with a range of 0.48-1.0 µm (n=11); the diameter of the salivary canal is
0.45+0.08, with a range of 0.25-0.54 µm (n=11). The maxillary stylets interlock with one another to form these canals. We did not measure the distance from the labium of each
49
image; therefore, these figures represent images at various locations along the length of the stylet bundle.
Sheath material
Sheath material is observed outside plant tissue, extending from the tip of the labium to the insertion point, approximately 200 µm (Figure 5). The sheath material adheres to the plant, beginning distal to the stem in reference to the abscission layer, then traveling across the abscission layer to the insertion point on the pulvini. Sheath material internal to the plant was not observed using SEM.
Western populations of HWA
Based on external morphological observations, individuals from the population of
HWA collected in Olympia, WA, appear slightly more robust than their eastern counterparts
(Figure 6A-B). However, morphologically, there were no obvious quantitative differences between populations. Figure 6E-F shows antennal structures of both specimens, although we also observed tarsi and stylet bundle. Typical insertion points for both populations are alike, occuring on the adaxial side of the hemlock needle, proximal to the twig with respect to the abscission layer (Figure 6C-D).
DISCUSSION
Our observations corroborate descriptions by Young et al. (1995) for the most consistent feeding sites for HWA, which is on the adaxial side of eastern hemlock needle, proximal to the twig with respect to the abscission layer. However, our studies also indicate that in events of extensive infestation (where settled individuals per needle base is greater
50
than two), alternate insertion points are used, which may or may not result in continued survival. HWA which insert their stylets into the stomata of a hemlock needle do not survive beyond their first instar, evident by the complete mortality observed (quantitative data not recorded; however, in each case (n = >20), the HWA was not alive). In all of these cases, the individual may be teased off the needle and it is apparent that the stylet bundle inserts directly through the stomata. Alternatively, HWA that settle on the woody stem of hemlock material do survive, as is evident from the production of a woolly ovisac. While not a lucrative insertion point for HWA, stomatal insertions are not completely unusual among
Aphidiodea. One aphid, Schizolachnus orientalis (Takahashi), inserts into the stomata of pine needles to reach its feeding site (Klingauf 1987b). Likewise, stem settling is typical behavior for the closely related balsam woolly adelgid (Adelges piceae (Ratz.)). Young et al.
(1995) reported that the HWA stylet destination and feeding site is the xylem ray parenchyma cells, presumably their primary or possibly exclusive feeding site. These storage cells are found within the stem, likely in reach of the stylets of stem-settled HWA.
Therefore, the alternate insertion point on the stem of a hemlock still allows access to the xylem ray parenchyma cells while HWA that settle on the needle, inserting through the stomata, are a greater distance from their internal feeding site, and likely unable to reach the appropriate cells. An inability to access food undoubtedly results in death for any sessile insect. A similar phenomenon occurs with the feeding biology of the silverleaf whitefly,
Bemisia argentifolii Bellows and Perring. If a whitefly crawler does not successfully penetrate an appropriate site on host tissues, it is unable to reach the internal veins on which it feeds (Cohen et al. 1998). While these observations reveal possible alternative settling sites
51
for HWA during severe infestations, understanding survival and fecundity subsequent to these insertion sites calls for more research.
Morphological differences between the HWA crawler and adult correspond with the biological activities that occur during those time periods. The crawler is the host-seeking stage of the insect, and is therefore responsible for determining a suitable stylet insertion site
(McClure 1989). This is likely the reason we see conspicuous antennae and sensilla on crawlers, possibly used in olfaction, gustation, or tactile reception during the process of host- plant selection. After a suitable insertion site is determined, the crawler settles and is sessile for the remainder of its life. As it develops, it is no longer necessary that they be receptive to cues from the host-plant, likely the reason we observe inconspicuous antennae and sensilla in adult HWA.
The ratio of antennal length to body length for HWA crawlers was 0.278 ± 0.031
(n=10). We would have liked to compare this to the antennal to body ratio of adult HWA, but all observed samples did not lend themselves for this purpose. In some cases, body length could be measured, but the antennae were at awkward angles. In other cases, antennal length could be measured, but the body length was obscured by angle. More samples could be prepped of adult HWA to quantify this relationship; our results are based on a qualitative comparison only.
As passive dispersers, HWA does not employ long-range host finding of hemlock.
Therefore, visual cues are not necessary for their detection of the host-plant, a process which is exclusively close-range. Adult and crawler stages of HWA both lack compound eyes and have three ocelli per side likely attributable to their passive dispersion habits. These ocelli
52
may allow HWA to sense light intensity, and therefore could be involved in the preference to settle on the underside of branches. It is unknown if day length is used to cue HWA to enter and exit aestivation, but temperature is known to play a primary role (Salom et al. 2001).
Being equipped with only ocelli contrasts with what we know of aphids, which use a combination of visual and olfactory cues to locate a host plant, and thus use compound eyes and olfaction sensilla or plates (Kennedy et al. 1959, Chapman et al. 1981, Bromley and
Anderson 1982, Backus 1988, Pickett et al. 1992, Park and Hardie 2003). Unlike its apterate counterparts, winged HWA sexuparae are involved in long-range host finding. This generation leaves its hemlock host, searching for a spruce where a sexual generation occurs.
This occurs across Adelgidae; the development of winged adults on a secondary non-spruce species is indicative of a holocyclic life cycle which alternates between Picea and another conifer in other genera, including Abies, Larix, Pinus, Pseudotsuga, and Tsuga (Annand
1928, Carter 1971, Havill and Footit 2007). In eastern North America, a suitable spruce does not exist, and the winged sexuparae presumably leave their hemlock host in search of a spruce and subsequently die. The well-developed compound eyes on the sexuparae suggest that visual cues may be involved in the long-range search for a suitable spruce in the native range of HWA.
The gross morphological observations of HWA mouthparts are similar to what is known of other homopterans (Hemiptera) (Backus and McLean 1982, Backus 1985, 1988,
Mora et al. 2001). The apex of the HWA labium possesses several short sensilla, reminiscent to apical labial sensilla found in aphids (Adams and Fyfe 1970, Forbes 1977, Wensler 1977,
Tjallingii 1978, Harris and Childress 1981), which should be explored in further detail. In
53
aphids, these sensilla are used strictly as mechanoreceptors during host acceptance (Wensler
1977, Tjallingii 1978). After selecting an insertion point, aphids enter host tissue using an alternating sawing motion of their four individual stylets to pierce the plant cuticle (Pollard
1973). The stylet bundle insertion behavior of HWA is unknown; however, the ridge-like apical dentitions of the mandibular stylet suggest that a similar behavior may be employed.
Conversely, these serrations may be considered adaptive for holding onto internal host tissue and/or serving as a fulcrum for maxillary stylet movements amongst hemipterans that do not produce a salivary flange (Cobben 1978, Cohen 2000, Boyd et al. 2002). Dentitions are observed in other members of Hemiptera (de Zoeten 1968, Cohen 1990, Boyd et al. 2003,
Leopold et al. 2003) and, at least in predaceous heteropterans, thought to initiate penetration of the stylet bundle, grasp prey, and mechanically degrade prey (Cohen 1995). For HWA, it is possible that these dentitions are used to initiate entry into the plant, anchor the sessile insect to the plant, and/or mechanically macerate plant tissues. Cohen (1990) describes the independent movement of mandibular stylets from maxillary stylets in predatory Hemiptera to facilitate cutting, rasping, and maceration of solid structures. Generally, the mandibular stylet serrations of phytophagous insects are not as deeply serrated and occur in fewer numbers than those found in predaceous insects (Faucheux 1975, Cobben 1978, Cohen
1990). The shallow dentitions seen in HWA may indicate that limited mechanical maceration occurs, suggesting the use of chemical maceration of plant tissues (i.e., by the injection of extra-oral digestive enzymes). This issue deserves further attention, especially in light of the implications the use of extra-oral digestion has on host response and host-plant resistance.
54
Mittler (1957) was the first to use electron microscopy to study aphid stylet bundle morphology (Parrish 1967). It is a powerful tool that has since been used to study the feeding biology of numerous hemipterans. Studies using electron microscopy have been done for HWA previously (Young et al. 1995), but they did not reveal stylet cross-section or analyze stylet morphology specifically.
The stylet bundle morphology of HWA is typical of other hemipterans, specifically aphids, with interlocking maxillary stylets forming separate food and salivary canals (Fig.
4A) (Mittler 1957, Parrish 1967, Cobben 1979, Klingauf 1987a, Cohen 1990, Miles 1999).
The stylet bundle is enclosed by the labium (Cohen 1990). The TEM cross-section image of
HWA stylets confirms the external morphologically-based hypothesis of Young et al. (1995) in regards to the components and structure of the stylet bundle. SEM micrographs did not indicate the presence of two distinct canals (Young et al. 1995). In addition, the stylet structure of HWA is similar to the stylet bundle morphology of the balsam woolly adelgid
(Fig. 4B), with interlocking maxillary stylets, forming salivary and food canals. Canal widths are comparable to those found in aphids (Klingauf 1987a). Because serial sections were taken, we did not measure the distance of each cross-section from the labium for each sample, and therefore these measurements represent the stylet bundle at different points along the length of the stylet bundle. For 10 of the 11 images, cross-sections of the food canal ranged from 0.86 to 1.0 µm. The outlier measurement was 0.48 µm, possibly at a point more distal to the HWA body. Similarly, 10 of the 11 measurements for the salivary canal cross-sections were 0.43 to 0.54 µm, with an outlier measuring 0.24 µm. These drastically smaller measurements likely represented a distal portion of the stylet bundle, as the food and
55
salivary canals in aphids are tapered with the proximal part approximately three times broader than the distal part (Pollard 1973).
The stylet bundle of HWA is more than three times the length of a crawler's body, more than three time the length of an established nymph's body, and longer than the body length of an adult (Young et al. 1995). In addition, our observations reveal that the food canal is 0.48-1.0 µm in diameter. The movement of materials through a passage with these dimensions is under constraint by Poiseuille's Law, which predicts the relationship between sucking force and: 1) stylet length, 2) stylet diameter, and 3) fluid viscosity necessary for flow (Mittler 1967, Kingsolver and Daniel 1993, Cohen 1995). The pressure gradient necessary for liquid flow through the food canal is described by Poiseuille's Law (Nobel
1983) as:
3 -1 where P = pressure difference (Pa), Xv = volume flow of fluid (m s ), n = dynamic viscosity
(Pa·s), C = length of food canal (m), R = radius of the food canal (m). Based on this equation, smaller insects must generate higher pressure differences to overcome the resistance of the feeding apparatus (Novotny and Wilson 1997). Given the measurements of the stylet bundle and the small body size of HWA, we can assume that a pressure differential or sucking force must be used to overcome the limitations for HWA as well. In addition, this has implications for use of digestive enzymes, which would decrease viscosity of materials in the food canal. Predaceous heteropterans chemically macerate their prey using digestive enzymes, then augment the mixture with watery saliva, further reducing the viscosity of the
56
slurry (Cohen 2000). This is a critical step in digestion of their prey, as highly viscous liquids would be difficult to transport in a long and narrow food canal (Mittler 1967, Cohen
1998).
Until this research, there was no evidence for the presence of sensory receptors associated with stylets for HWA. Our observations of the internal stylet bundle show there are canals in mandibular stylets containing neurons, indicating a sensory function. The neural canals of the mandibular stylets are similar to those found in aphids and other homopterans, which contain two dendrites used as proprioreceptors to detect the movement and position of the stylet bundle (Parrish 1967, Saxena and Chada 1971, Forbes and Raine
1973, Wensler 1974, Forbes 1966, 1977, Backus and McLean 1982, Foster et al. 1983,
Leopold et al. 2003). Neural canals are not present in the maxillary stylets, indicating that
HWA uses single stylet innervation, which optimizes sensation for the mandibular stylets- ahead probing method. As previously discussed in this section, we suggest that this probing method is used by HWA based on mandibular stylet dentitions. Single stylet innervation, in addition to longer mandibular than maxillary stylets and few labial sensilla, is typical of all sternorrhynchous homopterans (Backus 1988). Compounded by a lack of sensilla on the stylet bundle tip, it seems highly likely that HWA use their dendrites as proprioreceptors.
Evidence of sheath production in HWA was expected, based on the generalization that all homopterans, and more specifically sternorrhynchans, are stylet-sheath feeders (Miles
1968, Pollard 1973, Backus 1988) and specific observations of HWA feeding by Young et al.
(1995) describe a sheath material. The beaded appearance of the HWA salivary sheath outside the plant tissue corresponds to descriptions of aphid salivary sheath production in
57
unconfined spaces (Miles 1987). Production of salivary sheath material outside the plant tissue is found in other hemipterans in the form of a salivary flange. This flange, used as a fulcrum for stylet movement, is produced on the surface of the plant material and assists in anchoring the insect to its host (Saxena 1963, Cohen 1998, Cohen 2000). We do not observe a flange in this sense since HWA is not settled directly above its insertion site. Rather, the stylet distance between the labium tip and insertion point is covered in sheath, and appears to adhere to the plant surface. Sheath material, composed of lipoproteins, sticks reasonably well to a plant surface, thought to prevent the stylet bundle from slipping during insertion
(Pollard 1973). This may have implications for the purpose the salivary sheath serves to
HWA, especially since they are piercing the plant cuticle at an oblique angle from a distance and stabilization of the stylet bundle is crucial, similar to findings by Cohen (1990). It is evident from these studies that SEM is not satisfactory for determining internal sheath production and internal feeding sites by HWA.
Western North American populations of HWA do not appear morphologically distinct from the eastern North American population, save for a slightly more stout appearance
(qualitative observations only). Phylogenetically, the two populations diverge from different lineages, as the eastern population shares a single haplotype with populations in southern
Japan (Havill et al. 2006). Although no morphological differences were detected superficially, disparity might be documented if a revision of the family Adelgidae were done.
The family is in critical need of a taxonomic reworking, including morphological studies between species and among the different generations of the same species (Havill and Footit
58
2007). Currently, reliable keys to distinguish species do not exist, and our identification of
HWA is confined to the host tree on which it is found.
In conclusion, these studies indicate that: 1) HWA may insert stylets at alternative sites as a result of high population densities, 2) morphological differences between HWA ages and generations correspond with their respective biology, 3) dentitions of the mandibular stylets may be used to insert stylet bundle, anchor to the stem, and/or to mechanically macerate plant tissue, 4) HWA uses separate food and salivary canals to feed and the length and diameter of the food canal indicates that either great force is used or extra- oral digestion is employed, 5) neural canals in the mandibular stylets indicate single stylet innervation is occurring, likely used in proprioreception, 6) sheath material is produced outside of plant tissue and has a number of possible functions, namely to assist tissue penetration and for stylet stabilization, and 7) apart from western populations appearing slightly more robust, there are no conspicuous structural differences between the western and eastern populations of HWA in North America.
Despite these studies advancing what we currently know, there is much left to be studied in regards to the biology of HWA and the group Adelgidae (Havill and Footit 2007).
Their small size is somewhat deceptive, as these are complex organisms with complicated life cycles. Cholodkovsky (1915) reported in his studies on the group that the complexity of adelgids, particularly their parthenogenesis and polymorphism, "attain such a high degree here as nowhere else in the animal kingdom." In light of their rise as one of the most significant pests in forest entomology, it has become increasingly important that we study and understand their basic biology for use in management techniques. As these results
59
indicate, there is a great deal regarding the basic biology of HWA that is deserving of future research attention.
LITERATURE CITED
Adams, J.B., and F.Q. Fyfe. 1970. Stereoscope views of some aphid mouthparts. Can. J. Zool. 48: 1033-1034.
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac. Entomol. 1: 79-82.
Annand, P.N. 1928. A contribution toward a monograph of the Adelginae (Phylloxeridae) of North America. Stanford University, Palo Alto, Calif.
Arnaud, G. 1918. Les Astérinées. Annales de l'Ecole d'Agriculture. Montpelier 1918: 52-65.
Backus, E.A. 1985. Anatomical and sensory mechanisms of planthopper and leafhopper feeding behavior, pp. 163-194. In Nault, L.R. and J.G. Rodriguez (eds.), Leafhoppers and planthoppers. Wiley, New York, NY.
Backus, E.A. 1988. Sensory systems and behaviours which mediate hemipteran plant- feeding: a taxonomic overview. J. Insect Physiol. 34: 151-165.
Backus, E.A., and D.L. McLean. 1982. The sensory systems and feeding behavior of leafhoppers. I. The aster leafhopper, Macrosteles fascifrons Stal (Homoptera: Cicadellidae). J. Morphol. 172: 361-379.
Bernays, E.A., and R.F. Chapman. 1994. Host-plant selection by phytophagous insects. Chapman & Hall, New York, NY.
Boyd, D.W., Jr., A.C. Cohen, and D.R. Alverson. 2002. Digestive enzymes and stylet morphology of Deraeocoris nebulosus (Hemiptera: Miridae), a predacious plant bug. Ann. Entomol. Soc. Am. 95: 395-401.
Bromley, A.K., and M. Anderson. 1982. An electrophysiological study of olfaction in the aphid Nasonovia ribis-nigri. Entomol. Exp. Appl. 32: 101-110.
Butin, E., E. Preisser, and J. Elkinton. 2007. Factors affecting settlement rate of the hemlock woolly adelgid, Adelges tsugae, on eastern hemlock, Tsuga canadensis. Ag. For. Entomol. 9: 215-219.
60
Carter, C. I. 1971. Conifer woolly aphids (Adelgidae) in Britain. Bull. For. Comm. Lond. 42.
Chapman, R.F., E.A. Bernays, and S.J. Simpson. 1981. Attraction and repulsion of the aphid, Cavariella aegopodii, by plant foods. J. Chem. Ecol. 7: 881-889.
Cholodkovsky, N.A. 1915. Chermes injurious to conifers. Petrograd: Dep. Agric. Central Board Land Admin. Agric.
Cobben, R.H. 1978. Evolutionary trends in Heteroptera. Part II. Mouthpart-structures and feeding strategies. Meded. Landbouwhogesch. Wageningen.
Cobben, R.H. 1979. On the origins of feeding habits of the Hemiptera (Insecta): a reply to Merrill Sweet. Ann. Entomol. Soc. Am. 72: 711-715.
Cohen, A.C. 1990. Feeding adaptations of some predaceous Hemiptera. Ann. Entomol. Soc. Am. 83: 1215-1223.
Cohen, A.C. 1995. Extra-oral digestion in predaceous terrestrial Arthropoda. Annu. Rev. Entomol. 40: 85-103.
Cohen, A.C. 1998. Biochemical and morphological dynamics and predatory feeding habits in terrestrial Heteroptera, pp. 21-32. In J. R. Ruberson and M. Coll (eds.), Predaceous Heteroptera: Implications for biological control. Thomas Say Publications in Entomology. Entomological Society of America, Lanham, MD.
Cohen, A.C. 2000. How carnivorous bugs feed, pp. 563-570. In C. W. Schaefer and A. R. Panizzi (eds.), Heteroptera of economic importance. CRC, Boca Raton, FL.
Cohen, A.C., C. Chu, and T.J. Henneberry. 1998. Feeding biology of the silverleaf whitefly (Homoptera: Aleyrodidae). Chinese J. Entomol. 18: 65-82. de Zoeten, G.A. 1968. Application of scanning microscopy in the study of virus transmission of aphids. J. Virol. 2: 745-751.
Faucheux, M.M.J. 1975. Relations entre l'ultrastructure des stylets et mandibalaires et maxillaires et la prise de nourriture chez les Insectes Hémiptérès. Compte R. Hebd. Seances Acad. Sci. Ser. D. Sci. Nat. 281: 41-44.
Fidgen, J.G., D.E. Legg, and S.M. Salom. 2006. Binomial sequential sampling plan for hemlock woolly adelgid (Hemiptera: Adelgidae) sistens’ infesting individual eastern hemlock trees. J. Econ. Entomol. 99: 1500-1508.
61
Forbes, A.R. 1966. Electron microscope evidence for nerves in the mandibular stylets of the green peach aphid. Nature 212: 726.
Forbes, A.R. 1977. The mouthparts and feeding mechanism of aphids, pp. 83-103. In K.F. Harris and K. Maramorosch (eds.), Aphids as Virus Vectors. Academic Press, New York, NY.
Forbes, A.R., and D.B. Mullick. 1970. The stylets of the balsam woolly aphid, Adelges piceae (Homoptera: Adelgidae). Can. Ent. 102: 1074-1082.
Forbes, A.R., and J. Raine. 1973. The stylets of the six-spotted leafhopper, Macrosteles fascifrons (Homoptera: Cicadellidae). Can. Entomol. 105: 559-567.
Foster, S., L.J. Goodman, and J.G. Duckett. 1983. Ultrastructure of sensory receptors on the labium of the rice brown planthopper. Cell Tissue Res. 230: 353-366.
Hale, F.A. 2004. The hemlock woolly adelgid: a threat to hemlock in Tennessee. The University of Tennessee: Agricultural Extension Service. SP503-G.
Harris, K.F., and S.A. Childress. 1981. Preliminary observations on the morphology of apical sensory pegs on aphid labia. Phytopathol. 71: 897.
Havill, N.P., and R.G. Footit. 2007. Biology and evolution of Adelgidae. Annu. Rev. Entomol. 52: 325-349.
Havill, N.P., M.E. Montgomery, G. Yu, S. Shiyake, and A. Caccone. 2006. Mitochondrial DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic speciation and pinpoints the source of the introduction to eastern North America. Ann. Entomol. Soc. Am. 99: 195-203.
Havill, N.P., M.E. Montgomery, and R.G. Foottit. 2008. Molecular ecology of hemlock woolly adelgid, its hosts, and its natural enemies. In B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Jetton, R.M., F.P. Hain, W.S. Dvorak, and J. Frampton. 2008. Infestation rate of hemlock woolly adelgid (Hemiptera: Adelgidae) among three North American hemlock (Tsuga) species following artificial inoculation. J. Entomol. Sci. 43: 438-442.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
62
Kennedy, J.S., C.O. Booth, and W.J.S. Kershaw. 1959. Host finding by aphids in the field. I. Gynoparae of Myzus persicae (Sulzer). Ann. Appl. Biol. 47: 410-423.
Kingsolver, J.G., and T.L. Daniel. 1993. Mechanics of fluid feeding in insects, pp. 149- 161. In C.W. Schaefer, and R.A.B. Leschem (eds.), Functional morphology of insect feeding. Proc. Thomas Say Publications in Entomology, Entomol. Soc. Am., Lanham, MD.
Klingauf, F.A. 1987a. Feeding, adaptation and excretion, pp. 225-253. In A. K. Minks and P. Harrewijn (eds.), Aphids: Their biology, natural enemies and control, Volume 2A. Elsevier, Amsterdam, The Netherlands.
Klingauf, F.A. 1987b. Host plant finding and acceptance, pp. 209-223. In A. K. Minks and P. Harrewijn (eds.), Aphids: Their biology, natural enemies and control, Volume 2A. Elsevier, Amsterdam, The Netherlands.
Kloft, W. 1955. Untersuchungen an der Rinde von Weisstannen (Abies pectinata) bei Befall durch Dreyfusia (Adelges) piceae Ratz. Z. Angew. Ent. 37: 340-348.
Lambdin, P., J. Grant, D. Paulsen, and A. Saxton. 2006. Invasion of the hemlock woolly adelgid on eastern hemlock in the Great Smoky Mountains National Park. J. Tennessee Acad. Sci. 81: 49-54.
Leopold, R.A., T.P. Freeman, J.S. Buckner, and D.R. Nelson. 2003. Mouthpart morphology and stylet penetration of host plants by the glassywinged sharpshooter, Homalodisca coagulata, (Homoptera: Cicadellidae). Arthropod Struct. Dev. 32: 189-199.
MacKenzie, D.S. 2002. Perennial ground covers. Timber Press, Portland, OR.
McClure, M.S. 1987. Biology and control of hemlock woolly adelgid. Bull. Conn. Agric. Exp. Stn. 851.
McClure, M.S. 1989. Evidence of a polymorphic life cycle in the hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae). Ann. Entomol. Soc. Am. 82: 50-54.
McClure, M.S. 1990. Role of wind, birds, deer, and humans in dispersal of hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 19: 36-43.
McClure, M.S. 1991. Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis. Environ. Entomol. 20: 258-264.
McClure, M.S., and C.A.S.-J. Cheah. 1999. Reshaping the ecology of invading populations of hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae), in eastern North America. Biol. Invasions 1: 247–254.
63
McClure, M.S. 2001. Biological control of hemlock woolly adelgid in the eastern United States. FHTET-2000-08. USDA Forest Service, Forest Health Technology Enterprise Team, Morgantown, WV. 10 pp.
McLean, D.L., and M.G. Kinsey. 1965. Identification of electrically recorded curve patterns associated with aphid salivation and ingestion. Nature, London 205: 1130-1131.
Miles, P.W. 1959a. Secretion of two types of saliva by an aphid. Nature 183: 756.
Miles, P.W. 1959b. The salivary secretions of a plant-sucking bug, Oncopeltus fasciatus (Dall.) (Heteroptera: Lygaeidae). I. The types of secretion and their roles during feeding. J. Insect Physiol. 3: 243-255.
Miles, P.W. 1965. Studies on the salivary physiology of plant bugs: the salivary secretion of aphids. J. Insect Physiol. 11:1261-1268.
Miles, P.W. 1968. Insect secretions in plants. Ann. Rev. Phytopath. 6:137-164.
Miles, P.W. 1972. The saliva of Hemiptera. Adv. Insect Physiol. 9: 183-255.
Miles, P.W. 1987. Feeding process of Aphidoidea in relation to effects on their food plants, pp. 321-339. In A. K. Minks and P. Harrewijn (eds.), Aphids: Their biology, natural enemies and control, Volume 2A. Elsevier, Amsterdam, The Netherlands.
Miles, P.W. 1999. Aphid saliva. Biol. Rev. Camb. Philos. Soc. 74: 41-85.
Mittler, T.E. 1957. Studies on the feeding and nutrition of Tuberolachnus salignus (Gmelin). I. The uptake of phloem sap. J. Exp. Biol. 34: 334-341.
Mittler, T.E. 1967. Flow relationships for hemipterous stylets. Ann. Entomol. Soc. Am. 60: 1112-1114.
Mittler, T.E. 1988. Application of artificial feeding techniques for aphids, pp. 145-170. In A. K. Minks and P. Harrewijn (eds.), Aphids: Their biology, natural enemies and control, Volume 2B. Elsevier, Amsterdam, The Netherlands.
Mora, R., A. Retana, and A.M. Espinoza. 2001. External morphology of Tagosodes orizicolus (Homoptera: Delphacidae) revealed by scanning electron microscopy. Ann. Entomol. Soc. Am. 94: 438– 448.
Nobel, P.S. 1983. Biophysical plant physiology and ecology. W.H. Freeman, San Fransisco, CA.
64
Novotny, V., and M.R. Wilson. 1997. Why are there no small species among xylem-sucking insects? Evol. Ecol. 11:419-437.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
Parrish, W.B. 1967. The origin, morphology, and innervation of aphid stylets (Homoptera). Ann. Entomol. Soc. Am. 60: 273-276.
Park, K.C., and J. Hardie. 2003. Electroantennogram responses of aphid nymphs to plant volatiles. Physiol. Entomol. 28: 215-220.
Pickett, J.A., L.J. Wadhams, and C.M. Woodcock. 1992. The chemical ecology of aphids. Ann. Rev. Entomol. 37: 67-90.
Pirone, P.P. 1978. Diseases and pests of ornamental plants, 5th ed. Wiley, New York, NY.
Pollard, D.G. 1958. Feeding of the cotton aphid (Aphis gossypii Glover). Empire Cotton Growers Review 35: 244-253.
Pollard, D.G. 1969. Directional control of the stylets in phytophagous Hemiptera. Proc R Entomol. Soc. Lond. Ser. A Gen. Entomol. 44:173-185.
Pollard, D.G. 1971. Some aspects of plant penetration by Myzus persicae (Sulz.) nymphs (Homoptera, Aphididae). Bull. Entomol. Res. 61:315-324.
Pollard, D.G. 1973. Plant penetration by feeding aphids (Hemiptera, Aphidoidea): A review. Bull. Entomol. Res. 62:631-714.
Salom, S.M., A.A. Sharov, W.T. Mays, and J.W. Neal, Jr. 2001. Evaluation of aestival diapauses in hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 30: 877- 882.
Saxena, K.N. 1963. Mode of ingestion of a heteropterous insect Dysdercus koenigii (F.) (Pyrrhocoridae). J. Insect Physiol. 9:47-71.
Saxena, P.N., and H.L. Chada. 1971. The greenbug, Schizaphis graminum. III. Digestive system. Ann. Entomol. Soc. Am. 64: 1031-1038.
Smith, J.J.B. 1985. Feeding mechanisms, pp. 33-85. In G.A. Kerkut and L.I. Gilbert (eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology, vol. 4. Pergammon Press, Oxford.
65
Speight, M.R. 2007. Sampling insects from trees: shoots, stems, and trunks, pp. 77-115. In S.R. Leather (ed.), Insect sampling in forest ecosystems. Blackwell Science Ltd, Oxford.
Tjallingii, W.F. 1978. Mechanoreceptors of the aphid labium. Entomol. Exp. Appl. 24: 531- 537.
Tjallingii, W.F. 1994. Sieve element acceptance by aphids. Eur. J. Entomol. 91: 47-52.
Tjallingii, W.F., and T. Hogen Esch. 1993. Fine structure of aphid stylet routes in plant tissues in correlation with EPG signals. Physiol. Entomol. 18: 317-328.
Townsend, L., and L. Rieske-Kinney. 2006. Meeting the threat of the hemlock woolly adelgid. University of Kentucky: Cooperative Extension Service. ENTFACT-452.
USDA Forest Service. 2011. Counties with established HWA populations 2010. http://na.fs.fed.us/fhp/hwa/maps/2010.pdf.
Wensler, R.J. 1974. Sensory innervation monitoring movement and position in the mandibular stylets of the aphid, Brevicoryne brassicae. J. Morphol. 143: 349-364.
Wensler, R.J. 1977. The fine structure of distal receptors on the labium of the aphid, Brevicoryne brassicae L. (Homoptera). Cell Tissue Res. 181: 409-422.
Young, R.F., Shields, K.S. and G.P. Berlyn. 1995. Hemlock woolly adelgid (Homoptera: Adelgidae): stylet bundle insertion and feeding sites. Ann. Entomol. Soc. Am. 88: 827-835.
66
Figure 1. Feeding sites of the hemlock woolly adelgid. A) Woolly masses indicate settled HWA at the base of each needle, the usual feeding site for HWA; scale, 5 mm. B) HWA settled near the base of buds; scale, 1 mm. C) Sistentes crawlers recently settled on new growth; scale, 1 mm. D) Unconventional insertion site on the stomatal bands of the needle; scale, 0.1 mm. E) Unconventional insertion site on the woody stem; scale, 1 mm.
67
wp
ov
Figure 2. Hemlock woolly adelgid external morphology. A) Lateral view of progrediens crawler. B) Ventral view of progrediens crawler. C) Lateral view of adult. D) Head and thorax of a sexupara, showing compound eye. ant, antenna; ce, compound eye; la, labium; sty, stylet; ov, ovipositor; wp, wax pores. Scale, 100 µm.
68
Figure 3. Labium and stylet morphology. A) Distal segment of labium, showing sensilla on labium and four stylets emerging; scale, 10 µm. B) Tip of stylet bundle, showing mandibular stylets enclosing over the interior maxillary stylets; scale, 1 µm. C) Tip of maxillary stylet with five distinct dentitions. Scale, 1 µm. la, labium; sens, sensilla; sty, stylet.
69
Figure 4. Stylet cross-sections. A) Diagram representing the black bean aphid, Aphis fabae (taken from Miles (1999)). B) Transmission electron micrograph of balsam woolly adelgid, Adelges piceae (taken from Forbes and Mullick 1970). C) Transmission electron micrograph of hemlock woolly adelgid, A. tsugae. FC, food canal; Mds, mandibular stylet; Mxs, maxillary stylet; NT, neural canals of mandibles; SC, salivary canal. Scale, 1 µm.
70
Figure 5. Adaxial surface of needle petiole and pulvini with settled HWA; inset showing sheath-enclosed stylets lying on the surface of the plant, traversing the abscission layer, and inserting into the pulvini. Insertion site is indicated by the circle. Scale, 100 µm. abs, abscission layer; HWA, hemlock woolly adelgid; n, needle; p, pulvinus; sty, stylets.
71
Figure 6. Western North American HWA population compared to eastern North American population HWA. A) Adult western HWA. B) Adult eastern HWA. C) Settled western HWA, showing stylets entering between needle and stem. D) Settled eastern HWA, showing stylets entering between needle and stem. Scale, 100 µm. E) Antenna tip of western HWA, showing terminal sensilla. F) Antenna tip of eastern HWA, showing terminal sensilla. Scale, 10 µm. abs, abscission layer; n, needle; p, pulvinus; sens, sensilla; sty, stylets.
72
Chapter III
Fine Structure of Labial, Tarsal, and Antennal Sensilla and Setae of Sistens Crawlers of the Hemlock Woolly Adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae)
ABSTRACT
The fine structure of sensilla and setae located on the labium, antennae, and tarsi of sistens crawlers of the hemlock woolly adelgid (Hemiptera: Adelgidae) are described.
Possible functions of individual structures are suggested on morphological grounds and compared to closely related families, Aphididae and Phylloxeridae. Sensilla of the four- segmented labium appear to be strict mechanosensors, possibly involved in plant surface contact during host exploration and acceptance. Spatulate setae on tarsi possibly function in adhesion and one sensillum appears to function as a mechanoreceptor. Sensilla on antennae appear to be mechanoreceptors and are present on all three antennal segments. Near the tip of the flagellum, a sensorium is located subapically and may play a role in olfaction. A thorough understanding of hemlock woolly adelgid sensilla function has implications for the cues used in host acceptance and may increase our understanding of host-plant resistance in hemlocks. Further investigations should be done to confirm these interpretations.
INTRODUCTION
The motile stage of the hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera:
Adelgidae), known as the crawler, exhibits an array of behaviors during host searching that may be involved in acquisition of sensory information about appropriate sites to initiate feeding. They display incessant antennation, frequent taps of the labial apex on the host
73
surface, and maintain continuous tarsal contact with the host surface. Similar behaviors are found in other Hemipterans: antennation in Aphididae and Phylloxeridae (Backus, 1988;
Storer et al., 1996), labial tapping in Aphididae, Aleyrodidae, Pseudococcidae, and
Lygaeidae (Feir & Beck, 1963; Noldus et al., 1986; Walker, 1987; Rü et al., 1995), and tarsal contact-induced stylet penetration in Aphididae (Snodgrass, 1935; Powell et al., 1999).
Observations of these behaviors in A. tsugae crawlers suggest the labium, antenna, or tarsi may be used to obtain information about a potential host and the appropriate site to initiate feeding. Further observations have indicated a consistent initiation of stylet sheath formation at a specific site, on the adaxial side of the hemlock needle pulvinus (Young et al. 1995).
Stylet bundle insertion almost always occurs at this location; cues used by A. tsugae to locate it are unknown. Given this precision, A. tsugae must be obtaining sensory information to identify the site with such specificity. Sensilla or setae on all aforementioned appendages were seen in scanning electron micrographs of A. tsugae at comparatively low magnifications, but have never been investigated thoroughly (Oten et al., 2010). Sensory organs used by A. tsugae in host selection and acceptance are unknown.
Adelges tsugae Annand (Hemiptera: Adelgidae) is an inasive forst pest in eastern
North America with a complex life cycle consisting of two parthenogenetic generations per year that feed exclusively on hemlocks. The spring generation, present from March to June, is called progrediens, followed by the sistens generation, present from June until the following March (McClure, 1989; Gray & Salom, 1996). When progrediens mature in late spring, they produce up to 300 eggs within cottony white ovisacs. A large portion of these progeny become wingless adult progrediens, which are morphologically similar to their
74
mothers and remain on hemlock to aestivate through the summer. In October, they break aestivation and commence feeding (McClure, 1989). The remaining portion, a third generation, becomes winged adult sexuparae that leave hemlock, migrating to a primary spruce host (Picea spp.). In the eastern U.S., a suitable host spruce is not available and this generation presumably departs from its hemlock host and does not survive (McClure, 1987;
McClure 1989).
Host plant selection by phytophagous insects is governed by phagostimulants and deterrents recognized by the insect (Chapman, 2003). Sensilla used to sense these cues can be found on various parts of the body and in various forms. Functions are assigned to sensilla and setae of these appendages on aphids and other hemipterans. Sensilla have been found on the labial apex in several hemipteran families: Aphididae (Adams & Fyfe, 1970;
Forbes, 1977; Wensler, 1977; Tjallingii, 1978; Harris & Childress, 1981), Aleyrodidae
(Walker, 1987; Walker and Gordh, 1989), Pseudococcidae (Rü et al. 1995), and Miridae
(Avé et al., 1978). Except for the aphids, sensilla in these families function as chemoreceptors or mechano-chemoreceptors. In the aphids, apical labial sensilla are strictly mechanoreceptors, and they have olfactory receptors on antennal sensilla in addition to chemosensilla internal to the stylet food canal (Miles, 1958; Bromley et al., 1979; Anderson
& Bromley, 1987). Aphids antennate while walking using rhinaria to detect odors of the host boundary layer and trichoid sensilla at the tip of antennae for contact-chemosensory function
(Bromley et al, 1980; Pickett et al., 1992). They are thought to play roles in sensing plant olfactory cues and possibly alarm and sex pheromones (Backus, 1988). In addition, the tarsi of Aphididae utilize claws and attachment devices at the distal end of the tibia to move about
75
on rough host surfaces (Kennedy, 1986; Dixon et al., 1990). In silverleaf whitefly, Bemesia argentifolii Bellows and Perring (Hemiptera: Aleyrodidae), physical characteristics of the cotton surface act as cues for feeding sites, likely using surface contact with legs or antennae
(Cohen et al., 1996). Given their phylogenetic proximity and similar biology, the model for comparative studies of A. tsugae morphological characters is Aphididae, which is a comparatively well-studied family.
In the present study, our objective is to understand sensilla and setae function in A. tsugae by characterizing its fine structure. We hypothesize that we will determine the presence of olfactory, gustatory, and contact sensilla among the three appendages observed
(i.e., labium, tarsi, and antennae). Understanding their function will assist in understanding cue determination used in host selection and acceptance, in turn a significant component in understanding resistance mechanisms in some hemlock.
MATERIALS AND METHODS
Branches of infested eastern hemlock (Tsuga canadensis (L.) Carrière) were field collected in Crossnore and Laurel Springs, NC, and transported to Raleigh, NC, trimmed, and ends kept in water at 22ºC and approximately 65% humidity. Sites were selected based on our assessibility to infested hemlocks. Newly emerged sistens crawlers (Fig. 1) were shaken from the branches onto white filter paper. The sistens generation of A. tsugae develops on hemlock from late fall through spring, they are fecund and reproduce rapidly, and have the ability to locate an insertion point quickly. Thus, sistens are our generation of interest for this study. Crawlers were fixed in 3.0% glutaraldehyde in 0.1 M Sorenson’s buffer, pH 7.0,
76
for 48 h, then dehydrated in ethanol baths of 30%, 50%, 70%, 95% and 100% (3X) for 1 h each at 4ºC. Samples were critical point dried, mounted, and coated with gold-palladium at a deposition thickness of 20-25Å on four sides and the top. Samples were examined with a
JEOL JSM-5900LV scanning electron microscope (JEOL, Tokyo, Japan) at an accelerating voltage of 15-20 kV. Sensilla were categorized based on morphology and putative function was determined as suggested by Shields (2010) and Bromley et al. (1979, 1980).
RESULTS
Labium
The labium of the hemlock woolly adelgid (Fig. 2-4) is a four-segmented tubular structure. In the crawler stage (Fig. 1), it lies in the medial ventral concavity between coxae and reaches slightly beyond the metacoxae. The cuticle of all segments of the labium is smooth. The basal segment, segment I, possesses three sensilla trichoidea (Fig. 2.A). A bilaterally symmetrical pair occurs at the anterior region of the segment and a single medial sensillum trichoidea occurs near the posterior margin of the segment. High magnification
SEM indicates all three sensilla trichoidea are in a pit and lack a pore at the tip, indicative of strict mechanoreceptors (Fig. 2.B-C) (Iwasaki et al., 1999). Segment II appears to be the first movable segment of the A. tsugae labium (Fig. 3.A). All micrographs (n=17) show attachment to the A. tsugae body at its basal end only. A row of five bilaterally symmetrically arranged sensilla occur on opposite sides of the labial groove. On segment III of the A. tsugae labium, two sensilla occur on each side of the labial groove, one proximal and one peripheral to the groove. All of the sensilla on segments II and III exhibit the
77
morphological characteristics of mechanoreceptors: sensilla arise from a distinct pit and lack a pore at the tip and pores along the surface. Segment IV, the distal segment of the labium, possesses two fields of sensilla (Fig. 4.A). Six long (11.56 – 13.78 µm) subapical sensilla trichodea form a ring around the labium. Five shorter apical labial sensilla trichodea form a sensory field in close proximity to the stylet bundle opening (Fig. 4.B-C). Sensilla 4 and 5 are close to the stylet bundle and leaning towards it, seen touching the stylet bundle in some micrographs.
Tarsi
The tarsi are two-segmented bearing an ungues, or claw, on the distal end (Fig. 5.A).
The first segment is longer and annulated in appearance and possesses four pairs of setae or sensilla. It possesses a pair of long, knobbed setae near the distal end and four pairs of other setae or sensilla (Fig. 5.B). Pairs 1, 3 and 4 appear flexible and curved and 3 is spatulate or flatted on the apical ends (Fig. 5.C). Pair 2 is shorter, straight, seated in a pit and has no obvious pore at its tip (Fig. 5.D). The second tarsal segment is in the form of an ungues and a pair of setae arises subapically, also spatulate at its tip.
Antennae
The antenna is three-segmented, consisting of the scape, pedicel, and flagellum, and measures 101.78 ± 9.11 µm (n=16) in total length, exclusive of sensilla (Fig. 6.A). In comparison to the sessile adults of the same generation, the antennae are well-developed and elongated in the crawler stage. The scape has a pair of trichoid sensilla, one on the ventral surface and one on the anterior surface of the segment, both seated in a pit and lacking an observable pore at its tip. The pedicel also has a pair of trichoid sensilla, one on the ventral
78
surface and the other on the posterior surface of the segment (Fig. 6.B). They are also seated in pits and lack a visible terminal pore. The morphology indicates they likely function as mechanoreceptors. The flagellum is imbricate and 72.86 ± 4.77 µm (n=9) in length. It has a long terminal sensillum (36.90 ± 4.36 µm, n=7) and four secondary sensilla at the distal end
(Fig. 6.C). A pair of short pegs, approximately 0.5 µm in height and half as wide arises near the antennal tip (Fig. 6.E). A single sensillum arises on the posterior side of the flagellum, segregated from the field of terminal sensilla. A sensorium lays subapically on the posterior edge of the flagellum between this disjunct sensillum and the antennal tip (Fig. 6.D). The sensorium is surrounded with a cuticular fringe.
DISCUSSION
Taxonomic work on Adelgidae is difficult because several distinct morphs occur throughout a complex life cycle (Sano et al., 2011). This research focuses on the morphology of the crawlers because this stage is responsible for detecting cues of the host- plant, affecting crawler survival and in turn infestation success of a hemlock tree. Our studies corroborate much of the original, yet vague description of A. tsugae, which was based on a population infesting western hemlock (Tsuga heterophylla (Raf.) Sarg.) in the Pacific
Northwest (Annand, 1924). In these studies, we observe A. tsugae infesting eastern hemlock
(T. canadensis (L.) Carr.) in the eastern United States, a population originated from A. tsugae population in southern Japan, likely infesting T. sieboldii Carr. (Havill et al. 2006). In the original description, sensilla or setae were not described in detail nor ascribed a function.
79
Based on these morphological studies, all 33 observed sensilla of the A. tsugae labium function as mechanoreceptors. This is reminiscent of the aphid group, whose apical labial sensilla function strictly as mechanoreceptors (Wensler, 1977; Tjallingii, 1978). In aphids, stylet penetration is initially based on substrate color and texture and chemoreception occurs after short probing of the host plant and internal gustation (Miles, 1958; Ibbotson &
Kennedy, 1959; Pelletier, 1990; Powell et al., 2006). There is the possibility that some of these external sensilla of A. tsugae function as contact chemo-mechanoreceptors, but the images shown here do not support that theory. Given their biological similarities to aphids, we theorize that A. tsugae uses a similar host acceptance tool, and probes the plant briefly, sensing its host with internal chemoreceptors. However, more research should be done to substantiate this conjecture. For example, studies using transmission electron microscopy and/or electrical penetration graphs should be completed as complementary to this work.
Sensilla 4 and 5 of the labial tip are shorter and observed leaning in toward the stylet bundle as it emerges from the labium (Fig. 4.B-C). This may be used to record the position of the stylet bundle as it maneuvers, similar to aphids. At the apex of the aphid labium, eight pairs of sensilla record the position of the labium and the stylets (Tjallingii, 1978). Because they come into direct contact with the plant surface, they may also be used to assess the physical structure of the host plant surface prior to stylet insertion (Dunn, 1978). Moreover, in sheath feeding hemipterans, the salivary flange encases these terminal sensilla and it molds to the sensilla (Backus, 1985). Therefore, it has been suggested that the tip sensilla in sheath- feeders detects stimuli within the sheath, rather than the plant surface directly (Backus,
1988). In addition to short pegs at the tip of the labium, aphids possess neurons within the
80
stylets used as proprioreceptors to sense twisting and turning of the stylet bundle (Forbes,
1966; Wensler, 1974; Backus and McLean, 1982).
All segments of the A. tsugae labium are bare and lack teeth, contrasting with the cowpea aphid (Aphis craccivora Koch), where segment II of the labium possesses rows of teeth, thought to play a role in securing the telescoped segments together when they shortened during feeding (Ibbotson & Kennedy, 1959; Klingauf, 1987). Given the distance between A. tsugae and its insertion point between the pulvilli and the stem, it is likely unnecessary for condensing the labium. Similarly, aphids whose hosts have rough bark typically have longer labia to reach the bottom of crevices and do not condense their labium upon feeding (Heie, 1987).
We expected the apical labial sensilla to function as mechanoreceptors based on our knowledge of aphid feeding. However, there are distinct differences in biology between these groups (i.e., A. tsugae is sessile, aphids are motile; A. tsugae feeds on parenchyma cells, aphids feed on xylem) that give us cause to confirm the functions we assign here. It is possible that A. tsugae may be using internal chemosensory structures to find feeding sites; however, behavioral, morphological, or physiological evidence is not yet there to support that claim (Backus, 1988).
The original description of A. tsugae depicts tarsi with “two long setae arising from the first tarsal segment and two knobbed setae on the second” (Annand, 1924). We observed four additional pairs of setae and/or sensilla. The tarsal claw of A. tsugae is typical of aphids and other adelgids and is well-developed in tree-inhabiting aphids living on rough bark, like
A. tsugae (Cumming, 1968; Heie, 1987; Dixon et al., 1990; Beutel & Gorb, 2001). Anderson
81
and Bromley (1987) speculated that sensilla on the aphid tarsi may have chemosensory function but there is no published ultrastructural or electrophysiological evidence to support this. We suspect these extensions play roles in either adhesion or host selection.
The pair of knobbed setae on the first tarsal segment is the longest of all tarsal setae and has been seen touching the smooth surface of an inverted glass slide during visual observations at a stage microscope. Such setae were observed in other adelgids, a pair in the pine leaf adelgid (Pineus pinifoliae (Fitch)) and three per tarsus in the spruce gall adelgid
(Adelges lariciatus (Patch)), but the authors of these studies did not speculate about the function of the setae (Underwood, 1955; Cumming, 1968). These setae may act as adhesive organs, mechanoreceptors, or contact chemoreceptors, but these are conjectures based upon its morphology and contact with the plant surface. We suggest they may be involved in adhesion based on visual observations of surface contact during inverted walking and the morphological similarity to the adhesive tenant hairs of some springtails (Blottner &
Eisenbeis, 1984). Further work with ultrastructure may help elucidate their function.
Setae pairs 1 and 3 appear to be spatulate empodial setae based on their morphology.
This character is seen in related groups: the Japanese oak dwarflouse has two dorsoapical setae which are long and conspicuously capitate and the genus description for
Daktulosphaira (Hemiptera: Phylloxeridae) describes knobbed setae on the tarsi, called
‘digituli’ (Shimer, 1867; Miyazaki & Teramoto, 1991). The term ‘digitule’ is used amongst researchers of the Coccoidea in reference to “a pair of normally capitate setae at the inner base of the tarsal claws and at the outer distal margin of the tarsus” (Nichols, 1989). This description certainly fits the structure we see here. Furthermore, the occurrence of spatulate
82
empodial setae is believed to occur in aphid ancestors, but is a character now lost amongst present-day Aphididae (Zhang & Qiao, 1998). None of these descriptions make conjectures about the function of these spatulate empodial setae or digitules.
Pair 2 is morphologically similar to sensilla observed on the other appendages of A. tsugae. They are rigid, come to a blunt tip without a visible pore even at 15,000X and are seated in a pit. Therefore, it may be acting as a mechanoreceptor, but this needs to be corroborated by further ultrastructural analysis.
Within Adelgidae, antennal morphological variation occurs depending on the generational form; they can have three, four, or five segments, each segment possessing a sensorium. The scape and pedicel do not differ between forms (Stoetzel, 1998). Our examinations corroborate the original description of A. tsugae crawler antennae, described as three-segmented, with the third segment approximately four times the length of the second
(Annand, 1924).
The pedicel of aphids possesses usually paired coeloconic sensilla and a campaniform sensillum, used to monitor movement of the flagellum (Dunn, 1978). The setiform sensilla we see here are likely analogous to the aphid coeloconic sensilla, a type of trichoid sensilla, and we did not observe the presence of a campaniform sensillum. In the grape phylloxera, three trichoid sensilla arise from the pedicel and two from the scape (Huber et al., 2009).
On a morphological basis, the six sensilla at the antenna apex likely function as mechanoreceptors. In casual behavioral observations, contact with an object by the antennae causes the crawler to change direction, supporting the theory that these work as mechanoreceptors. Antennal contact with the host surface and short terminal sensilla occur
83
in aphids (Dunn, 1978). These four sensilla have a pore at the tip, seen by crystal violet staining and SEM (Slifer et al., 1964; Bromley et al., 1979). We did not observe a pore at the tip of A. tsugae sensilla with SEM, but further studies at higher magnification and possibly with sections of the sensillum examined by transmission electron microscopy (TEM) might be required to substantiate this. Several distinct sensilla of undescribed function are observed at the tip of the flagellum of grape phylloxera and five minute sensilla occur on the Japanese oak dwarflouse (Miyazaki & Teramoto, 1991; Huber et al., 2009). The pair of short pegs observed near the antennal tip is undescribed in other aphids or phylloxerids and their function unknown.
The sensorium observed on the antenna of the A. tsugae crawlers is likely analogous to the primary rhinaria in aphids and referred to by Stoetzel (1998) in alate female Adelgidae.
In Stoetzel’s description, each segment of the flagellum possesses a sensorium. Bromley et al. (1979, 1980) describes primary and secondary rhinaria of chemosensory function in aphids, and Hardie et al. (1994) determined their sensitivity to sex pheromones and plant volatiles. The same term is applied to at least two phylloxerids: the Japanese oak dwarflouse,
Phylloxera kunugi Shinji, and the grape phylloxera, Daktulosphaira vitifoliae Fitch
(Miyazaki & Teramoto, 1991; Huber et al., 2009). In other phylloxerid taxonomic work, the term ‘sensorium’ is applied to identical structures (Duncan, 1922). We choose to follow the terminology used by Stoetzel which is applied specifically to Adelgidae and is an accepted term across Aphidioidea, whereas ‘rhinaria’ is specific only to the family Aphididae
(Nichols, 1989). Moreover, the term ‘sensorium’ is also used in descriptions of the pine leaf adelgid (Underwood, 1955).
84
A cuticular fringe surrounds the sensorium, resembling finger-like projections. In aphids, this fringe is thought to act as a shielding sieve, keeping particles out (Dunn, 1978).
The grape phylloxera possesses a rhinarium which is not surrounded by a cuticular fringe, but located slightly basal of the terminal sensilla, a location similar to the sensorium of A. tsugae (Huber et al., 2009). The sensorium in A. tsugae is well situated on the antenna to be directed toward the plant surface, supplementary supporting evidence that the structure is likely used to detect olfactory cues.
The majority of the sensilla observed on appendages of A. tsugae appear to be mechanoreceptors, contrasting with the multitude of chemoreceptors interspersed on many aphids. From this, we draw two implications. First, the A. tsugae sistens and progrediens generations are flightless and passive dispersers. They are transported between host plants via wind, phoresy on macrovertebrates, or human movement (McClure, 1990). Therefore, landing on an appropriate host tree, a hemlock, is entirely serendipitous. Conversely, aphids are active dispersers, using olfactory and visual cues to locate a host plant (Kennedy et al.,
1959; Chapman et al., 1981; Bromley & Anderson, 1982; Backus, 1988; Pickett et al., 1992;
Park & Hardie, 2003). Subsequent host acceptance is accomplished by contact chemoreception internal to the stylet food canal (Miles, 1958). The passive mode of transportation by A. tsugae suggests that olfactory cues are not useful in long-range host location. Therefore, the single pair of sensorium on the A. tsugae antenna is likely sufficient for detecting the limited olfactory cues likely needed. Similar trends are seen in related insects. Apterous aphids lack numerous antennal placoid sensilla that alate aphids possess
(Bromley et al., 1979). Also, alate grape phylloxera have a secondary rhinaria while the
85
wingless form has one (Huber et al., 2009). Wingless generations of aphids, phylloxerids, and A. tsugae simply do not need structures highly sensitive to long-range olfactory cues.
The second implication deals with the methodology used and indicates the need for continued research. SEM is a powerful tool, but should be corroborated by complementary studies. We were unable to observe any terminal pores on sensilla, a character that would indicate contact-chemoreception is being used. Observations using TEM would be a beneficial accompaniment to this work and provide additional information about sensilla function. Results would depict internal structures of the sensilla trichoidea, especially the presence or absence of pores and innervating neurons and dendrites. In our minimal attempts to do so, we were met with challenges that prevented positive results. Sections of minute sensilla are difficult to obtain since they cannot be located before sectioning with a light microscope (Bromley et al., 1979). In addition to TEM work, electrophysiological and behavioral studies, such as electroantennogram studies and/or electrical penetration graphs, would be substantial to verify our morphologically-based functional hypotheses outlined here.
Finally, these data were based off field collections from two sites in North Carolina,
Crossnore and Laurel Springs. Broader specimen collections across the eastern U.S. range of
A. tsugae followed by SEM observations of sensilla and setae will provide more representative data to confirm our conclusions. Based on their clonal reproduction and their infestation of the same host species throughout the range, we may assume that there is little variation in sensilla and setae between populations occupying different parts of the range, but further observations should be done to reflect the population at the regional and national
86
levels. Variability within the eastern U.S. population is not an unsupported assumption, as phenotypic plasticity has been documented between the northern and southern region A. tsugae populations with respect to coldhardiness (Skinner et al., 2003).
LITERATURE CITED
Adams, J.B. and Fyfe, F.Q., 1970. Stereoscope views of some aphid mouthparts. Canadian Journal of Zoology 48, 1033-1034.
Anderson, M. and Bromley, A.K., 1987. Sensory system. In: Minks, A.K. and Harrewijn, P. (Eds), Aphids: Their Biology, Natural Enemies and Control. Elsevier, Amsterdam, pp. 153-162.
Annand, P.N., 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pacific Entomologist 1, 79-82.
Avé , D.J., Frazier, L. and Hatfield, L.D., 1978. Contact chemoreception in the tarnished plant bug Lygus lineolaris. Entomologia Experimentalis et Applicata 24, 217-227.
Backus, E.A., 1985. Anatomical and sensory mechanisms of planthopper and leafhopper feeding behavior. In: Nault, L.R. and Rodriguez, J.G. (Eds.), Leafhoppers and Planthoppers. Wiley, New York, pp. 163-194.
Backus, E.A., 1988. Sensory systems and behaviours which mediate hemipteran plant- feeding: a taxonomic overview. Journal of Insect Physiology 34, 151-165.
Backus, E.A. and McLean, D.L., 1982. The sensory systems and feeding behavior of leafhoppers. I. The aster leafhopper, Macrosteles fascifrons Stal (Homoptera: Cicadellidae). Journal of Morphology 172, 361-379.
Beutel, R.G. and Gorb, S.N., 2001. Ultrastructure of attachment specializations of hexapods (Arthropoda): evolutionary patterns inferred from a revised ordinal phylogeny. Zoological Systematics and Evolutionary Research 39, 177-207.
Blottner, D. and Eisenbeis, G., 1984. Ultrastructure of long tibiotarsal spatula-hairs in Tomocerus flavescens (Collembola: Tomoceridae). Annales de la Société Royale Zoologique de Belgique 114, 51-57.
87
Bromley, A.K. and Anderson, M., 1982. An electrophysiological study of olfaction in the aphid Nasonovia ribis-nigri. Entomologia Experimentalis et Applicata 32, 101-110.
Bromley, A.K., Dunn, J.A. and Anderson, M., 1979. Ultrastructure of the antennal sensilla of aphids. I. Coelonic and placoid sensilla. Cell and Tissue Research 203, 427-442.
Bromley, A.K., Dunn, J.A. and Anderson, M., 1980. Ultrastructure of the antennal sensilla of aphids. I. Trichoid, chordotonal and campaniform sensilla. Cell and Tissue Research 205, 493-511.
Chapman, R.F., 2003. Contact chemoreception in feeding by phytophagous insects. Annual Review of Entomology 48, 455-484.
Chapman, R.F., Bernays, E.A. and Simpson, S.J., 1981. Attraction and repulsion of the aphid, Cavariella aegopodii, by plant foods. Journal of Chemical Ecology 7, 881-889.
Cohen, A.C., Chu, C., Henneberry, T.J., Freeman, T., Buckner, J. and Nelson, D., 1996. Cotton leaf surface features serve as behavioral cues to silverleaf whiteflies. Southwestern Entomologist 21, 377-385.
Cumming, M.E.P., 1968. The life history and morphology of Adelges lariciatus (Homoptera: Phylloxeridae). Canadian Entomologist 100, 113-126.
Dixon, A.F.G., Croghan, P.C. and Gowing, R.P., 1990. The mechanism by which aphids adhere to smooth surfaces. Journal of Experimental Biology 152, 243-253.
Duncan, C.D., 1922. The North American species of Phylloxera infesting oak and chestnut (Hemiptera: Phylloxeridae). Canadian Entomologist 54, 275-276.
Dunn, J.A., 1978. Antennal sensilla of vegetable aphids. Entomologia Experimentalis et Applicata 24, 348-349.
Feir, D. and Beck, S.D., 1963. Feeding behavior of the large milkweed bug, Oncopeltus fasciatus. Annals of the Entomological Society of America 56, 224-229.
Forbes, A.R., 1966. Electron microscope evidence for nerves in the mandibular stylets of the green peach aphid. Nature 212, 726.
Forbes, A.R., 1977. The mouthparts and feeding mechanisms of aphids. In: Harris, K.F. and Maramorosch, K. (Eds.), Aphids as Virus Vectors. Academic Press, New York, pp. 83-103.
88
Gray, D.R., and Salom, S.M., 1996. Biology of the hemlock woolly adelgid in the southern Appalachians, pp. 16-25. In: S.M. Salom, T.C. Tigner, and R.C. Rheardon (Eds.), Proceedings of the First Hemlock Woolly Adelgid Review, 12 October 1995. USDA Forest Service, FHTET96-10, Morgantown, WV, pp. 16-25.
Hardie, J., Visser, J.H. and Piron, P.G.M., 1994. Perception of volatiles associated with sex and food by different adult forms of the black bean aphid, Aphis fabae. Physiological Entomology 19, 278-284.
Harris, K.F. and Childress, S.A., 1981. Preliminary observations on the morphology of apical sensory pegs on aphid labia. Phytopathology 71, 897.
Havill, N.P., Montgomery, M.E., Yu, G., Shiyake, S. and Caccone, A., 2006. Mitochondrial DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic speciation and pinpoints the source of the introduction to eastern North America. Annals of the Entomological Society of America 99, 195-203.
Heie, O.E., 1987. Morphological structures and adaptations. In: Minks, A.K. and Harrewijn, P. (Eds), Aphids: Their Biology, Natural Enemies and Control. Elsevier, Amsterdam, pp. 393-400.
Huber, L., Michaelis, P., Eisenbeis, G. and Forneck, A., 2009. Antennal sensilla of grape phylloxera (Daktulosphaira vitifoliae Fitch). Acta Horticulturae 816, 91-96.
Ibbotson, A. and Kennedy, J.S., 1959. Interaction between walking and probing in Aphis fabae. Journal of Experimental Biology 36, 377-390.
Iwasaki, M., Itoh, T. and Tominaga, Y., 1999. Mechano- and phonoreceptors. In: Eguchi, E. and Tominaga, Y. (Eds.), Atlas of Arthropod Sensory Receptors: Dynamic Morphology in Relation to Function. Springer, Tokyo, pp. 177-190.
Kennedy, C.E.J., 1986. Attachment may be a basis for specialization in oak aphids. Ecological Entomology 11, 291-300.
Kennedy, J.S., Booth, C.O. and Kershaw, W.J.S., 1959. Host finding by aphids in the field. I. Gynoparae of Myzus persicae (Sulzer). Annals of Applied Biology 47, 410-423.
Klingauf, F.A., 1987. Host plant finding and acceptance. In: Minks, A.K. and Harrewijn, P. (Eds), Aphids: Their Biology, Natural Enemies and Control. Elsevier, Amsterdam, pp. 209- 223.
McClure, M.S., 1987. Biology and control of hemlock woolly adelgid. Bulletin of the Connecticut Agriculture Experiment Station 851.
89
McClure, M.S., 1989. Evidence of a polymorphic life cycle in the hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae). Annals of the Entomological Society of America 82, 50-54.
McClure, M.S., 1990. Role of wind, birds, deer, and humans in dispersal of hemlock woolly adelgid (Homoptera: Adelgidae). Environmental Entomology 19, 36-43.
Miles, P.W., 1958. Contact chemoreception in some Heteroptera, including chemoreception internal to the stylet food canal. Journal of Insect Physiology 2, 338-347.
Miyazaki, M. and Teramoto, N., 1991. Morphology and bionomics of the Japanese oak dwarflouse Phylloxera kunugi (Homoptera: Aphidinea: Phylloxeridae). Entomologia Generalis 16, 201-206.
Nichols, S.W., 1989. The Torre-Bueno Glossary of Entomology. 7th edition. New York Entomological Society and the American Museum of Natural History, New York.
Noldus, L.P.J.J., Rumei, X. and van Lenteren, J.C., 1986. The parasite-host relationship between Encarsia formosa (Gahan) (Hymenoptera, Aphelinidae) and Trialeurodes vaporariorum (Westwood) (Homoptera: Aleyrodidae). XIX Feeding-site selection by the greenhouse whitefly. Journal of Applied Entomology 101, 492–507.
Oten, K.F., Cohen, A.C., Strider, J.B. and Hain, F.P., 2010. Investigating host resistance using scanning electron microscopy. In: Onken, B. and Rheardon, R. (Eds.), Proceedings, Fifth Symposium on Hemlock Woolly Adelgid in the Eastern United States, 17-19 August 2010, Asheville, NC. USDA Forest Service, Forest Health Technology Enterprise Team, FHTET-2010-07, Morganton, WV, pp. 122-128.
Park, K.C. and Hardie, J., 2003. Electroantennogram responses of aphid nymphs to plant volatiles. Physiological Entomology 28, 215-220.
Pelletier, Y., 1990. The role of the color of the substratum on the initiation of the probing behavior in Myzus persicae (Sulzer) and Macrosiphum euphorbiae (Thomas) (Homoptera: Aphididae). Canadian Journal of Zoology 68, 694-698.
Pickett, J.A., Wadhams, L.J. and Woodcock, C.M., 1992. The chemical ecology of aphids. Annual Review of Entomology 37, 67-90.
Powell, G., Maniar, S.P., Pickett, J.A. and Hardie, J., 1999. Aphid responses to non-host epicuticular lipids. Entomologia Experimentalis et Applicata 91, 115-123.
Powell, G., Tosh, C.R. and Hardie, J., 2006. Host plant selection by aphids: behavioral, evolutionary, and applied perspectives. Annual Review of Entomology 51, 309-330.
90
Rü, B.L., Renard, S., Allo, M.-R., Le Lannic, J. and Rolland, J.P., 1995. Ultrastructure of sensory receptors on the labium of the cassava mealybug, Phenacoccus manihoti Matile Ferrero. Entomologia Experimentalis et Applicata 77, 31-36.
Sano, M., Havill, N.P. and Ozaki, K., 2011. Taxonomic identity of a galling adelgid (Hemiptera: Adelgidae) from three spruce species in Central Japan. Entomological Science 14, 94-99.
Shields, V.D.C., 2010. High resolution ultrastructural investigation of insect sensory organs using field emission scanning electron microscopy. In: Méndez-Vilas, A. and Díaz , J. (Eds.), Microscopy: Science, Technology, Applications and Education. Formatex, Badajoz, Spain, pp. 321-328.
Shimer, H., 1867. On a new genus of Aphidae. Transaction American Entomological Society 1, 283-285.
Skinner, M., Parker, B.L., Gouli, S. and Ashikaga, T., 2003. Regional responses of hemlock woolly adelgid (Homoptera: Adelgidae) to low temperatures. Environmental Entomology 32, 523-528.
Slifer, E.H., Sekhon, S.S. and Lees, A.D., 1964. The sense organs on the antennal flagellum of aphids (Homoptera), with special reference to the plate organs. Quarterly Journal of Microscopical Science 105, 21-26.
Snodgrass, R.E., 1935. Principles of Insect Morphology. McGraw-Hill Book Company, New York.
Stoetzel, M.B., 1998. Antennal and other characters useful in identification of the Aphidoidea (Homoptera). Proceedings of the Entomological Society of Washington 100, 588-593.
Storer, J.R., Powell, G. and Hardie, J., 1996. Settling responses of aphids in air permeated with non-host plant volatiles. Entomologia Experimentalis et Applicata 80, 76-78.
Tjallingii, W.F., 1978. Mechanoreceptors of the aphid labium. Entomologia Experimentalis et Applicata 24, 531-537.
Underwood, G.R., 1955. The external morphology of the five forms of Pineus pinifoliae (Fitch) (Homoptera: Phylloxeridae). Canadian Entomologist 87, 201-209.
Walker, G.P., 1987. Probing and oviposition behavior of the bayberry whitefly (Homoptera: Aleyrodidae) on young and mature lemon leaves. Annals of the Entomological Society of America 80, 524-529.
91
Walker, G.P. and Gordh, G., 1989. The occurrence of apical labial sensilla in the Aleyrodidae and evidence for a contact chemosensory function. Entomologia Experimentalis et Applicata 51, 215-224.
Wensler, R.J., 1974. Sensory innervation monitoring movement and position in the mandibular stylets of the aphid, Brerricoryne brassicae. Journal of Morphology 143, 349- 364.
Wensler, R.J., 1977. The fine structure of distal receptors on the labium of the aphid, Brevicoryne brassicae L. (Homoptera). Cell and Tissue Research 181, 409-422.
Young, R.F., Shields, K.S. and Berlyn, G.P., 1995. Hemlock woolly adelgid (Homoptera: Adelgidae): stylet bundle insertion and feeding sites. Annals of the Entomological Society of America 88, 827-835.
Zhang, G. and Qiao, G., 1998. The phylogenetic analysis of Saltusaphidinae (Homoptera: Drepanosiphidae). Insect Science 5, 301-309.
92
Figure 1. Scanning electron micrograph of the hemlock woolly adelgid crawler, ventral view, anterior directed upward; scale, 100 µm. ant, antenna; la, labium.
93
Figure 2. Scanning electron micrographs of the hemlock woolly adelgid labium, segment I. A) Overview of segment, showing an anterior pair of sensilla and single medial posterior sensillum; scale, 10 µm. B) Close up of anterior sensillum, part of the pair; scale, 1 µm. C) Close up of posterior single sensillum; scale, 1 µm.
94
Figure 3. Scanning electron micrographs of the hemlock woolly adelgid labium segments II and III. A) Labial segment II, showing five sensilla; scale, 10 µm. B) Labial segment III, showing 2 sensilla; scale, 10 µm. lg, labial groove.
95
Figure 4. Scanning electron micrographs of the hemlock woolly adelgid labium tip, segment IV. A) Apical aspect, showing ring of six subapical sensilla trichodea, labial groove at top of micrograph; scale, 10 µm. B) Apical labial sensilla trichodea, labial groove at top of micrograph; scale, 1 µm. C) Lateral view of apical labial sensilla, labial groove at bottom of micrograph; scale, 1 µm. lg, labial groove; sst, subapical sensilla trichodea; sty, stylet bundle.
96
Figure 5. Scanning electron micrograph of the hemlock woolly adelgid tarsus. A) Overview of tarsus; scale, 10 µm. B) Tip of knobbed seta tip; scale, 1 µm. C) Tip of pairs 1 (forefront) and 2 (background); scale, 1 µm. D) Setae 2; scale, 1 µm. ks, knobbed seta.
97
Figure 6. Scanning electron micrographs of the hemlock woolly adelgid antenna. A) Ventral aspect of entire antenna; scale, 10 µm. B) Scape and pedicel, arrows pointing at sensilla; scale, 10 µm. C) Antennal tip; scale, 10 µm. D) Sensorium; scale, 1 µm. E) Pegs near antenna tip; scale, 1 µm. fl, flagellum; pe, pedicel; sc, scape.
98
Chapter IV
Trophically-related Enzymes of the Hemlock Woolly Adelgid, Adelges tsugae (Hemiptera: Adelgidae)
ABSTRACT
Extra-oral digestion is employed by many insects and has far-reaching implications for their success in feeding. Several characteristics of the feeding biology of the hemlock woolly adelgid (HWA), Adelges tsugae, suggest that salivary enzymes may be injected into their host prior to ingestion. These studies investigate trophically-related enzymes used by
HWA, specifically trypsin-like protease, amylase-like enzyme, peroxidase, and polyphenol oxidase. Using whole body homogenate, the presence of all four of these enzymes were established, although their specific source (e.g., salivary glands, alimentary canal) could not be discerned. Lygus lineolaris, the tarnished plant bug, served as a positive control used in comparisons throughout these studies. HWA had four times lower trypsin-like activity
(0.259 BAEE U µg protein-1 min-1) and eight times lower amylase-like activity (0.1088 mU protein-1 min-1) than L. lineolaris. Despite occuring in lower activity levels, the presence of trypsin-like enzyme and amylase-like enzyme suggests that, if injected into the plant, it is likely used for digestion of insoluble plant proteins and starches, respectively. HWA had almost twice as much peroxidase (0.2323 mU µg protein-1 min-1) and three times as much polyphenol oxidase activity (0.0569 abs µg protein-1 min-1) compared to L. lineolaris.
Moreover, utilization of oxidases (i.e., peroxidase and polyphenol oxidase) by HWA is suggestive of a detoxification event of plant defenses. Peroxidases may counteract increased levels of H2O2 (which occurs in infested hemlock), or conversely, the action of salivary
99
polyphenol oxidase on plant phenolics has been shown to generate H2O2. The implications for insect-plant interactions and the potential for developing resistant hemlocks are discussed.
Further work should focus on determining the origin of these trophically-related enzymes, establishing the presence of other digestive enzymes used by HWA, further characterizing these enzymes, and investigating the implications for host-plant response.
INTRODUCTION
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an invasive pest of eastern hemlock, Tsuga canadensis (L.) Carriére, and
Carolina hemlock, Tsuga caroliniana Engelm., in the eastern U.S. Following infestation, hemlocks exhibit signs of water stress, marked by needle drop, bud abortion, and inhibition of new growth (McClure 1991, 1992, Walker-Lane 2009). Internally, production of abnormal wood occurs (Walker-Lane 2009), reminiscent of the abrupt transition from earlywood to latewood in gymnosperms (Pallardy 2008). A healthy hemlock can die in as few as four to more than 10 years following infestation (McClure 1987, 1991). Despite being close to infested trees in addition to the formerly widely-accepted notion that hemlocks in the eastern U.S. were entirely susceptible to HWA, anecdotal evidence has suggested that some inividual and stands of hemlocks still remain healthy (Oten et al. 2011). It is now acknowledged that there are varying degrees of susceptibility to HWA, both inter- and intraspecifically. Infested hemlock susceptible to HWA exhibit increased levels of hydrogen peroxide, a suggested hypersensitive response to HWA feeding (Radville et al. 2011).
However, it is unknown if the response is caused by or a result of health decline or tree
100
stress. It has been suggested that the injection of 'toxic' saliva is causing the decline in tree health as seen in aphid-plant systems (Miles 1990, McClure 1995, Young et al. 1995, Shields et al. 1996), but there is no evidence to support that claim.
The current knowledge of digestive enzymes used by HWA is lacking. Young et al.
(1995) suggest the possibility of pectinase in the saliva of HWA, supported only by evidence of a predominantly intercellular pathway within plant tissue and indication of protein in the saliva via staining with Napthol yellow-S. The authors acknowledged that their data do not confirm the presence of any enzymes, and suggested that research should continue in this area. Using a starch iodine test of HWA whole-body homogenate, Kaur (2009) detected a weak presence of amylase in HWA homogenate, but it was surmised to originate from microorganisms of the gut. Further investigations into the presence of digestive enzymes should be done to better understand the feeding biology of HWA.
HWA feeds primarily on the xylem ray parenchyma cells of its host hemlock (Young et al. 1995), which is consistent with feeding sites documented throughout the family
Adelgidae (Auclair 1963, Pollard 1973). The closely related balsam woolly adelgid, Adelges piceae (Ratz.), has been documented feeding on the phloem tissue of new growth (Balch
1952), but this has not been observed in HWA.
HWA belongs to the superfamily Aphidiodea. Based on their close relationship, aphid feeding biology is often used to make hypotheses or generalizations regarding HWA feeding. Aphids produce two types of saliva during feeding (Miles 1959). The first is salivary sheath material that hardens upon extrusion. It surrounds the stylet bundle and is thought to stabilize the labium during insertion, act as a fulcrum for stylet maneuvering,
101
protect the insect against host-plant defenses, and/or enable stylet bundle reinsertion following a molt (Cohen 1990, Miles 1999). The second salivary excretion is watery saliva, which may contain digestive enzymes for extra-oral digestion, establish and maintain feeding sites, suppress plant defenses, and/or induce changes in plant physiology (Miles 1999, Will et al. 2007, Mutti et al. 2008). Both saliva types have been documented in HWA and deserve further research attention (Young et al. 1995, Oten et al. 2011).
Salivary proteins have been studied with direct techniques and/or salivary gland extracts of several aphid species (Cooper et al. 2011 and references therein). Attention to the composition of aphid saliva has increased in recent years due to their role in insect-plant interactions (Miles 1999, Cherqui and Tjallingii 2000, Will et al. 2007, Cooper et al. 2010).
Obtaining watery saliva from hemipterans for analytical research can be done by several methods. Perhaps the most direct method is to collect it from the insect mouth parts themselves (Miles 1964), augmented by surface application of pilocarpine to the insect's abdomen and/or treating insects with a stream of warm air (Binnington and Schotz 1973,
Madhusudhan et al. 1994). However, this method has not been effective for the collection of aphid saliva (Miles 1999), and therefore likely inappropriate for collection of HWA saliva. It is also possible to stimulate feeding on a feeding substrate which then can be assessed for digestion (Adams and McAllan 1956). A similar method involves inducing aphid probing into water, resulting in a dilute solution that can be analyzed spectrophotometrically or electrophoretically (Miles and Harrewijn 1991, Madhusudhan and Miles 1998). This method is unsuitable for collection of HWA saliva because to date, attempts to prompt HWA probing into water or artificial media has been unsuccessful. Insect salivation has been stimulated by
102
starving them (Miles and Slowiak 1970) and with a novel brush-induced approach, used on glassy-winged and smoke tree sharpshooters (Alhaddad et al. 2011). The sessile behavior and small size of HWA make it difficult to follow either of these methods. In some cases, saliva can be obtain by administering electric stimulation (Pereira et al. 1996, Corzo et al.
2001), but again, the small size of HWA is a limiting factor for this approach. Baptiste
(1941) and Madhusudhan et al. (1994) suggest using salivary gland homogenate to perform tests in scenarios when the aforementioned methods are inapplicable, but we have been unsuccessful in attempts to dissect the salivary glands from the minute HWA bodies
(crawlers are less than 0.5 mm in length (Young et al. 1995)). In select situations, whole- body homogenate may be used to obtain watery saliva (Adams and McAllan 1958, Adams and Drew 1963). This is typically used for rapid enzyme assessment once the presence of an enzyme has been established (Miles 1999), but we utilize this approach to overcome the difficulties of isolating the saliva or salivary glands of HWA.
The objective of this study is to establish the presence or absence of trophically- related enzymes in HWA whole-body homogenate. We hypothesize that we will detect four enzymes, protease, amylase, peroxidase, and polyphenol oxidase, in HWA. Therefore, we targeted these enzymes for our survey approach to enzyme determination. While we cannot discern the location of the enzyme using whole-body homogenate, the presence or absence of specific enzymes has implications for plant-insect interactions and therefore variation in hemlock susceptibility. In addition, several discussions (Adams and McAllan 1956, Hori
1971, Cohen 1990) have explained that the presence of certain enzymes is indicative of
103
feeding behaviors. Understanding the enzymes used by HWA should help to clarify some of the vagueness currently surrounding its feeding habits.
MATERIALS AND METHODS
Sample storage
Adult, actively-feeding HWA used in these enzyme surveys were collected from live eastern hemlock material obtained from Laurel Springs, NC, in January 2011. The ovisac was teased off with a paintbrush and ten individuals per vial were stored at -80°C. Similarly, neonate Lygus lineolaris (Palisot de Beauvois) (Hemiptera: Miridae) were stored ten individuals per vial at -80°C for use as positive controls in these assays.
Purification of HWA samples
To prepare HWA samples for the amylase-like enzyme, peroxidase, and polyphenol oxidase (PPO) assays, one vial containing ten of the frozen HWA was thawed at room temperature and homogenized in 300 µL phosphate buffered saline (PBS). To partially purify the homogenate, the sample was centrifuged and supernatant was filtered through a
3,000 MW Centrifugal Filter Unit (Millipore Corporation, Billerica, MA). This was centrifuged at 14,000 rpm for 15 min. The filter was inverted and inserted into a clean vial and centrifuged at 1,000 rpm for two min to recover products retained in the filter. The resulting concentrate was augmented with 100 µL PBS to bring the volume near 130 µL.
Protease assay
General protease activity was determined using the EnzChek® protease assay kit
(Invitrogen, Eugene, OR). A casein substrate was prepared by mixing 0.2 mL PBS to one
104
vial of the lyophilized substrate. The solution was mixed, left at room temperature for 5 m, and added to 19.8 mL 1X digestion buffer, resulting in a ten µg mL-1 BODIPY® casein substrate. An authentic trypsin standard was prepared by diluting trypsin, Type II-S, from porcine pancreas (Sigma-Aldrich®, St. Louis, MOS) in PBS at final concentrations of 2.32 mg mL-1, 3.25 mg mL-1, and 4.0 mg mL-1. The positive control was derived from insects known to possess the enzymes in question, one and ten L. lineolaris were thawed at room temperature and homogenized in 100 µL PBS with a motorized pestle tissue grinder (L. lineolaris has confirmed trypsin-like protease activity (Zeng et al. 2002)). To prepare the
HWA samples (crude extracts), one and ten HWA were thawed and homogenized in 100 µL
PBS. In each well of a standard 96-well plate, 100 µL of the casein substrate was mixed with
10 µL of the standards and samples. Analyzed samples included three repetitions each of: negative control, deionized water (diH2O), trypsin standards, one and ten L. lineolaris in
PBS, and one and ten HWA in PBS. Fluorescence, which is proportional to enzyme activity, was measured with a FilterMax F5 multi-mode microplate reader (Molecular Devices,
Sunnyvale, CA) at an excitation wavelength of 485 nm and an emission wavelength of 535 nm at 37°C.
Amylase assay
For determination of amylase-like activity, we used the EnzChek® Ultra Amylase
Assay Kit (Invitrogen, Eugene, OR). Starch substrate was prepared by mixing one vial of the lyophilized substrate with 50mM sodium acetate buffer (pH 4.0). Solution was vortexed for
20 s and left at room temperature for 5 min with occasional mixing. Next, 900 µL of PBS was added, the solution was mixed, then kept at room temperature in the dark until used. An
105
amylase standard curve was prepared by serially diluting α-amylase from Aspergillus oryzae
(Moyashimon) (Sigma-Aldrich: Fluka BioChemika, Buchs, Switzerland). Final concentrations used in the assay were: 2.5 mU mL-1, 5 mU mL-1, 10 mU mL-1, and 20 mU mL-1. For the positive control, four and ten L. lineolaris were thawed and homogenized in
120 µL PBS with a motorized pestle tissue grinder (L. lineolaris has confirmed α-amylase activity (Zeng and Cohen 2000)). In each well of a standard 96-well plate, 50 µL of the starch substrate was mixed with 50 µL of the sample. Analyzed samples included two repetitions each of: negative control (diH2O), the amylase standard curve, four and ten L. lineolaris in PBS, and ten purified HWA in PBS. Fluorescence of the digestion products from the substrate was measured by a FilterMax F5 multi-mode microplate reader (Molecular
Devices, Sunnyvale, CA) at an excitation wavelength of 485 nm and an emission wavelength of 535 nm at 37°C.
Peroxidase assay
Peroxidase assays were performed using the Amplex® Red Hydrogen Peroxide/
Peroxidase Assay Kit (Invitrogen, Eugene, OR). Hydrogen peroxide (H2O2) substrate was prepared by dissolving one vial of Amplex® Red reagent in 60 µL dimethylsulfoxide. 50 µL of this solution was mixed with 500 µL 10 mM H2O2 and 4.45 mL 1X Reaction buffer. A peroxidase standard curve was prepared by serially diluting 10 U mL-1 horseradish peroxidase to final concentrations of: 0.3 mU mL-1, 0.6 mU mL-1, 1.2 mU mL-1, and 2.4 mU mL-1. Although literature does not indicate the presence of peroxidase in L. lineolaris, we included the samples for consistency and as a control. Four and ten L. lineolaris were thawed and homogenized in 120 µL PBS with a motorized pestle tissue grinder. In each well
106
of a standard 96-well plate, 50 µL of the H2O2 substrate was mixed with 50 µL of the sample.
Samples included two repetitions each of: negative control (diH2O), the horseradish peroxidase standard curve, four and ten L. lineolaris in PBS, and purified HWA in PBS.
Fluorescence of the digestion products from the substrate was measured by a FilterMax F5 multi-mode microplate reader (Molecular Devices, Sunnyvale, CA) at an excitation wavelength of 535 nm and an emission wavelength of 595 nm at 37°C.
Polyphenol oxidase assay
To determine the presence of PPO activity in HWA, we modified methods used by
Lee et al. (2008). Samples were prepared by homogenizing four, ten, and ten L. lineolaris in separate vials in 120 µL PBS with a motorized pestle tissue grinder. Two purified HWA samples were prepared, and then 50 µL of each sample were pipetted into wells, along with ten µL of PBS or 1.0 mg mL-1 phenylthiourea (PTU), a polyphenol oxidase inhibitor.
Samples included two repetitions each of: diH2O (with ten µL of PBS), four L. lineolaris
(with ten µL of PBS), ten L. lineolaris (with ten µL of PBS), ten L. lineolaris (with ten µL
PTU), ten HWA (with ten µL of PBS), ten HWA (with ten µL PTU). Samples were incubated at 37°C for 20 min to allow inhibition of PPO by PTU. Then, 50 µL 5mM L-Dopa
(substrate) was added. Absorbance at 450 nm was measured from the substrate by a
FilterMax F5 multi-mode microplate reader (Molecular Devices, Sunnyvale, CA) at 37°C.
Quantifying activity per unit of protein
To quantify the amount of protein in the samples, we performed a Bradford protein assay. Quick Start™ Bradford Dye Reagent, 1X (Bio-Rad Laboratories, Inc., Hercules, CA) was prepared by diluting in a 1:4 ratio with diH2O. The solution was filtered through a
107
Whatman #1 filter to remove particles. Protein standards were prepared by serially diluting bovine serum albumin (Sigma-Aldrich, St. Louis, MO) in concentrations of 0.5, 0.25, 0.125,
0.063, 0.031, 0.016, and 0.008 mg mL-1. Ten μl of each standard and samples were pipetted into microplate wells. Two-hundred µl diluted reagent was added, shaken for 5 s, and incubated at room temperature for 5 min. Absorbance at 595 nm was measured with a
FilterMax F5 multi-mode microplate reader (Molecular Devices, Sunnyvale, CA) at room temperature.
Using the standard curve, amount of protein for each sample was calculated. These were converted to units of activity per µg of protein (as measured by the Bradford assay) per minute. Data was analyzed for enzyme activity units at one h for amylase and PPO and at the asymptote of enzyme activity for trypsin-like enzyme (30 min) and peroxidase (15 min).
RESULTS
In whole-body homogenate of HWA, all four of the enzymes we tested (trypsin-like protease, amylase-like enzyme, peroxidase, and PPO) were present. There were clear differences in the enzymatic activities between HWA and L. lineolaris. The kinetic for
HWA enzymatic activity for trypsin-like activity reached its asymptote at 30 min, whereas the L. lineolaris comparison continued to increase after more than 1 h (Fig. 1). Trypsin-like activity in HWA (0.259 BAEE U µg protein-1 min-1) was approximately four times lower than in L. lineolaris (0.8405 BAEE U µg protein-1 min-1) (Fig. 2). Amylase-like activity in L. lineolaris was almost eight times higher (0.7748 mU µg protein-1 min-1) than HWA (0.1088 mU µg protein-1 min-1) (Fig. 3, 4). At 15 min, HWA peroxidase enzyme activity had reached
108
an asymptote, whereas L. lineolaris continued to increase after 1 h (Fig. 5). The amount present in HWA (0.2323 mU µg protein-1 min-1) is almost twice as much relative to peroxidase of L. lineolaris (0.1283 mU µg protein-1 min-1) (Fig. 6). Investigations into PPO- like activity reveal that HWA (without PTU) is approximately three times more active in
HWA (0.0569 abs µg protein-1 min-1) than L. lineolaris (0.017 abs µg protein-1 min-1), which did not exhibit a kinetics curve (Fig. 7, 8). The sample that contained PTU (PPO inhibitor) was almost 18 times less active than the sample without PTU (0.00323 abs µg protein-1 min-1).
DISCUSSION
Using whole-body homogenate in our survey for trophically-related enzymes of
HWA, we detected trypsin-like enzyme, amylase-like enzyme, peroxidase, and polyphenol oxidase (PPO). Although these enzymes could be present in the salivary gland complex, midgut, or hemolymph of the insect, their occurrence is nevertheless revealing of the feeding habits of HWA. Generally, the presence or absence of specific enzymes is indicative of a consumer's ability to digest plant or prey materials (Baptist 1941, Adams and McAllan 1956,
Strong and Kruitwagen 1968, Tingey and Pillemer 1977, Cohen 1990, 1998, Agusti and
Cohen 2000).
After several trials using crude whole-body homogenate of HWA to detect amylase- like activity, we were observing a lower emission reading in the HWA sample than in the negative control, diH2O. If there was no activity, the sample fluorescence emission should have been equivalent to the diH2O readings. The HWA homogenate sample was dark green
109
in appearance, relative to the nearly colorless L. lineolaris samples or translucent standards.
We hypothesized that the opaque coloration could be blocking fluorescence emitted as a result of starch degradation. Therefore, we opted to purify the samples under the assumption that a more translucent solution would enable more accurate readings by the plate reader.
This was also a beneficial step to include in light of our limitation to use whole-body homogenate because additional materials in the sample, apart from the salivary gland complex or alimentary canal, might have inhibitors or color elements that could possibly mask the enzyme activity. We used 3,000 molecular weight filters because all of the enzymes we were testing for had molecular weights greater than 3,000. After this procedure produced successful results for our amylase-like enzyme assays, we applied it to the remaining assays, peroxidase and PPO.
Lygus lineolaris is a major agricultural pest that feeds on at least 130 economically important plants, including cotton, cereals, fruits, vegetables, and alfalfa (Snodgrass et al.
1984, Young 1986, Butts and Lamb 1991). The use of this insect species for comparisons with HWA was useful, given its accessibility, its comparable biomass to HWA as a neonate, and its well-characterized use of digestive enzymes in the literature. Trypsin-like protease has been partially characterized from their salivary gland complex, indicating that this enzyme is used extra-orally in a highly efficient feeding behavior (Zeng and Cohen, Zeng et al. 2000). Many salivary enzymes are similar between homopterans and heteropterans (Miles
1964, 1965).
Our assays confirm that trypsin-like-enzyme activity is present in HWA, enabling it to digest protein either extra-orally and/or in the alimentary canal. Protease activity from the
110
salivary glands is adaptive quality for zoophagy (Cohen 1993, 1998), but is also found in insects with phytophagous feeding habits (Agusti and Cohen 2000). Protease has been detected in both the salivary glands and gut of the green peach aphid, Myzus persicae
(Sulzer) (Bramstedt 1948). Accordingly, it is possible that the trypsin-like activity of HWA is in the salivary glands and/or gut. If used extra-orally by HWA, it may be used to digest structural proteins of the plant that are otherwise insoluble (Hori 1970, 1971). Whether it is derived from the salivary glands or the alimentary canal, it seems likely that the presence of protease indicates HWA is utilizing protein-containing food, rather than depending upon free amino acids as is the case for sap-feeding insects such as aphids.
The presence of protease in HWA is not unique, as it occurs in several other homopeterans. For example, protease was detected in the salivary gland and gut of the green peach aphid, as mentioned above, and in the gut and filter chamber of Schizolachnus spp.
(Bramstedt 1948), and in the gut of the pea aphid, Acyrthosiphon pisum (Harr.) (Srivastava and Auclair 1963). Trypsin-like and cathepsin B-like proteases also occur in the gut of the rice brown plant hopper, Nilaparvata lugens (Stål) and is suggested to occur in many the sap- sucking homopterans (Foissac et al. 2002).
The presence of trypsin-like-enzyme activity in HWA has implications for host response and host-plant resistance. When herbivorous insects feed, some plants respond defensively. The wound-response pathway is one of these responses. Triggering of this pathway, which could be done when HWA insert their stylets into host tissues, ultimately results in expression of many genes, including protease inhibitor genes (Dietrich et al. 1999).
This response is systemic; for example, damage of a single tomato leaf induces protease
111
inhibitors in the non-wounded leaves of the same plant (Green and Ryan 1972, Ryan 2000).
Similarly, feeding by HWA is implicated in inducing a hypothesized systemic response of hemlock following infestation (Walker-Lane 2009, Radville et al. 2011). If HWA feeding triggers the wound-response pathway, we would expect resistant hemlocks to successfully execute the physiological processes of the pathway, effectively inhibiting protease activity of
HWA while susceptible hemlocks would be unable to do so. This pathway may be triggered not only by mechanical wounding, but by physiological elicitors in the saliva. Elicitors that induce chemical changes of plants have been identified in the saliva of several insects, such as caterpillars (Mattiacci et al. 1995, Halitschke et al. 2001, Musser et al. 2002). This suggests that the response of hemlock to HWA feeding may be triggered not only by mechanical wounding, but by physiological elicitors in the saliva. Future research should focus on this interaction to understand if HWA is triggering the wound-response pathway in hemlocks via mechanical or salivary wounding. To test this, trees could be mechanically damaged (in an attempt to mimic HWA feeding) and regurgitant or HWA homogenate applied to the wound. The resulting response could then be examined. This approach has been used to mimic herbivore wounding in other systems (Halitschke et al. 2001,
Gouinguené et al. 2003) and may be successful if applied to this system.
Protease inhibition by plants can be induced (as discussed above) or constitutive
(Schoonhoven et al. 2005, Jongsma and Beekwilder 2008). It is effective in conferring resistance of some host-plants to insects closely related to HWA. For example, the grape phylloxera, Viteus vitifoliae (Fitch), has cathepsin-like protease in its saliva which is secreted into its host-plant (Blagowestschenski 1955). It is accepted that this enzyme is inactivated by
112
the high oxidation-reduction potential of resistant grape vines. In artificial diet bioassays fortified with protease inhibitors, the potato aphid (Macrosiphum euphorbiae (Thomas)) exhibited nymphal mortality and did not reproduce (Rahbé et al. 2003, Azzouz et al. 2005).
The protease inhibitors used in these assays were those that are inserted into crops by genetic engineering to produce varieties resistant to the potato aphid. Likewise, transgenic rice plants containing Soya bean Kunitz trypsin inhibitor have partial resistance to the rice brown plant hopper which, as mentioned above, has proteases in its gut (Foissac et al. 2002).
Indication of amylase-like activity in HWA is also suggestive of their feeding behavior. As a macerating enzyme, amylase-like enzymes are significant in breaking down and liquefying plant tissues to be utilized by insects (Bateman and Millar 1968, Laurema et al. 1985, Cohen and Wheeler 1998, Agusti and Cohen 2000). Ray parenchyma cells, the targeted feeding site of HWA (Young et al. 1995), contain starch (Wiedenhoeft and Miller
2005). It is suggested that these starches are likely ingested by HWA; therefore, amylase is the necessary enzyme needed to digest starches to convert them to sugars (van der Maarel et al. 2002).
While the amylase-like activity in HWA is lower than in L. lineolaris, it is possible that there is low activity of amylase in the saliva, which plays a minor role in digestion. It is possible that small quantites are contained within the salivary glands (for use in extra-oral digestion) at any one point in time. Because this assay took a "snapshot" of the amylase-like enzyme present, its actual employment in HWA feeding throughout its life could be substantially higher. Conversely, a minor amount of amylase could be adequate for the starch-consuming function necessary to HWA survival. Weak amylase activity was observed
113
in certain aphids by Duspiva (1953), who deduced that they likely do not play a significant role in extra-oral digestion. Amylase activity has been determined in the salivary gland, gut, and filter chamber of many aphid species (Auclair 1963 and references therein, Srivastava
1987). Perhaps one of the earliest detections of amylase activity was indicated in the salivary glands of Aphis rumicis L. (Davidson 1923), supporting the theory by Zweigwelt (1914) that aphids must be digesting plant starches. In addition, the amylase has been demonstrated to be present in larval midguts of many insects including members of Orthoptera, Hymenoptera,
Diptera, Lepidoptera and Coleoptera (Terra and Ferreira 1994).
HWA is an efficient pest of hemlocks, capable of locating a specific insertion site without complex vision or the capability of flight. When they find this feeding site, they are sessile for the remainder of their life (Annand 1924, Young et al. 1995). Therefore, it is critical that HWA not only locate a feeding site that will remain suitable throughout the course of their life, but they must be able to efficiently exploit the host tissues for nutrition.
Ray parenchyma cells, the feeding site of HWA, are thin-walled and typically 15 µm high,
10 µm wide, and 150–250 µm long (Wiedenhoeft and Miller 2005, Pallardy 2008) (Fig. 9).
Their function is primarily the synthesis, storage, and lateral transport of biochemicals and to a lesser extent, water, to the internal parts of the stem (Koning 1994, Wiedenhoeft and Miller
2005). Ray parenchyma cells contain starch, oils, tannins, and crystals, which must be the targeted items for ingestion by HWA (Mauseth 2009). If starches are being ingested by
HWA, which is reasonable to assume based on the dietary need for sucrose of all aphids
(Srivastava 1987), then we hypothesize that extra-oral digestion via injection of amylase-like activity into the plant, may be occurring. Starch granule size varies based on host plant;
114
granules of examined agriculture crops vary between 2 and 100 µm in diameter (Robyt
1998). This size may be a prohibitive factor for ingestion of intact granules. HWA stylets are long and narrow, they reach over 1 mm in length (Young et al. 1995) and the food canal is less than 1 µm in diameter (see Chapter 2). This canal is subject to Poiseuille's Law, which predicts the relationship between sucking force and: 1) stylet length, 2) stylet diameter, and 3) fluid viscosity necessary for flow (Mittler 1967, Kingsolver and Daniel
1993, Cohen 1995). The pressure gradient necessary for liquid flow through the food canal is described by Poiseuille's Law (Nobel 1983) as:
3 -1 where P = pressure difference (Pa), Xv = volume flow of fluid (m s ), n = dynamic viscosity
(Pa·s), C = length of food canal (m), R = radius of the food canal (m). Transporting starch granules through this canal would likely require extra-oral digestive enzymes, i.e. amylase- like enzymes, to break starches down prior to ingestion. The biology of HWA, specifically the presence of amylase-like activity in whole-body homogenate, is suggestive that extra-oral digestion is occurring, despite our inability to confirm this in the lab.
As with protease inhibitors produced by plants in response to herbivore attack, amylase inhibitors may also be produced across the plant kingdom (Marshall 1975, Garcia-
Olmedo et al. 1987). In some cases, inhibitors may be bifunctional, capable of inhibiting protease and amylase activity (Ryan 1990). In a similar fashion to protease inhibitors, amylase inhibition may occur in resistant plants or be used in the development of a transgenic hemlock resistant to HWA.
115
Peroxidase-like activity and PPO were detected in HWA in a higher amounts than was observed in L. lineolaris. We hypothesize PPO activity is characteristic of HWA stylet sheath material. This assumption is based on two established data. First, L. lineolaris does not secrete a stylet sheath (Smith 1926, Miles 1972); therefore, if PPO is present primarily in the sheath material of HWA, we would expect much lower concentrations in L. lineolaris, which is what is observed. Second, PPO is readily detected in the sheath material of many aphids (Miles 1965, Miles and Slowiak 1970, Urbanska et al. 1994, 1998).
Prior to our comparative studies between L. lineolaris and HWA, presence of peroxidase and PPO in L. lineolaris has not been demonstrated. While they could not truly be a 'positive control' for these assays, they provide a suitable comparison between HWA and a non-sheath-feeder. The milkweed bug, Oncopeltus fasciatus (Dallas), is likely the most thoroughly researched plant-sucking insect in terms of sheath material analysis (Miles 1960,
1967, 1999). Future investigations into HWA sheath material enzymes could be done with comparisons with O. fasciatus.
The presence of oxidases suggests a detoxification role against plant defenses if it is present in the saliva (Miles 1999). From a biochemical standpoint, this is a major contribution to the suggested reaction of hemlock to HWA feeding. Infestation by HWA results in elevated hydrogen peroxide (H2O2) levels (Radville et al. 2011), a defense mechanism exhibited across multiple plant species (Heath 2000, Liu et al. 2010). This suggests that H2O2 may be countered by or induced by salivary injections of HWA.
Peroxidases may counteract H2O2, but the action of salivary PPO on plant phenolics has been shown to generate H2O2 (Jiang and Miles 1993, Madhusudhan and Miles 1998).
116
Alternatively, if production of H2O2 does not occur across the genus and susceptible hemlocks are the only hemlocks with increased H2O2 levels following infestation, implying a negative reaction to HWA feeding, then oxidases may be involved in this as well.
Peroxidases require H2O2 as an electron acceptor (Miles 1999). It is possible that an increased level of H2O2 is enabling increased level of injected peroxidases to detoxify complementary plant defenses of hemlock in synergistic activities.
Systemic damage may result from injection of oxidases into host tissue. For example,
Madhusudhan and Miles (1998) suggest that necrosis of alfalfa susceptible to the spotted alfalfa aphid, Therioaphis maculata Buckton, is possibly caused by the injection of salivary oxidases into the plant. It is suggested that the necrotic reaction occurs more rapidly than a naturally-occurring reducing system can accommodate (Miles 1999). If this is true in the
HWA system, then possibly the systemic reaction seen in hemlock is an exhibition of the tree's inability to compensate for the injection of oxidases, while resistant trees are capable of counteracting the effects or do not respond as extremely as susceptible trees.
PPO present in whole-body homogenate does not entirely indicate that it is salivary in origin. The hemolymph of insects also has PPO activity used in immunity. During cellular encapsulation, an invading organism is surrounded by multiple layers of haemocytes, which may then become melanized (under the action of PPO), leading to the death of the invader
(Rowley et al. 1990, Ourth and Renis 1993, Hagen et al. 1994). However, we suggest that the observed activity of PPO in HWA homogenate primarily originates from the salivary gland complex of HWA and is augmented by minimal amount of hemolymph-derived PPO, comparable to the quantity observed in L. lineolaris. This statement is based on the
117
relatively higher levels of PPO activity in HWA (without PTU) to L. lineolaris, which had significantly lower activity altogether (Fig. 8). Because L. lineolaris does not produce a stylet sheath, the PPO in those samples is likely a product of the hemolymph, watery saliva, or both. Therefore, HWA, which is of comparable biomass to the neonate L. lineolaris, is likely exhibiting activity from more than one source. Moreover, PPO is a product of the accessory glands throughout the Hemiptera (Miles 1960, 1964, Miles and Helliwell 1961) and occurs in the sheath material and watery saliva of many studied Aphidoidea (Miles 1965,
Peng and Miles 1988, Miles and Harrewijn 1991, Leszczynski et al. 1996, Madhusudhan and
Miles 1998, Urbanska et al. 1998).
While it is possible that these enzymes from whole-body homogenate originate in part or wholly from microorganisms within HWA, there is considerable precedence of each of these enzymes being of insect origin. However, Kaur (2009) suggested the detection of minor amylase-like activity in HWA originated from endosymbionts. Shields and Hirth
(2005) suggest at least four endosymbiotic bacteria occur intercellularly and extracellularly in HWA hemolymph and microorganisms have been observed in many homopterans
(Douglas 1989, Campbell 1990, Clark et al. 1992). We support the hypothesis that the enzymes we detected were of insect origin based on what has been detected in the salivary glands and alimentary canals of other hemipterans.
One of the next steps for these studies is to determine the origin of these enzymes.
Ideally, saliva can be collected from insects for such assays, typically by allowing the insect to feed in artificial media or by stimulating salivation. After various attempts to stimulate penetration of artificial media by HWA, we were not successful. A close relative of HWA,
118
the balsam woolly adelgid (BWA), Adelges piceae (Ratz.), will penetrate artificial media in a lab setting. Given their similar biology, it may be appropriate to assay secretions of BWA for postulations that can be applied towards better understanding the feeding biology of
HWA. Alternatively, the salivary gland complex can be dissected. Again, this was attempted for use in these assays, but the small size of HWA (approximately 1 mm (Young et al. 1995)) has complicated the successful removal of salivary glands. The extraction procedure is even more difficult given that the glands must be removed without any portion of the alimentary canal. If we cannot be certain that we have extracted the salivary glands without any portion of the alimentary canal, conclusions cannot be drawn as to the true origin of the enzymatic activity. Isolation of watery saliva and/or sheath material from the alimentary canal will be essential in further discerning enzymatic source and also discerning the production of HWA-derived enzymes from enzymes produced by endosymbionts. To determine if our detection of PPO is salivary or hemolymph derived, it would be idyllic to induce HWA stylet insertion into L-DOPA gel. If PPO is in the secreted saliva, we can expect the gel to darken surrounding their feeding sites.
Clearly, whether present in the salivary glands or the gut of the insect, HWA is equipped with digestive enzymes that help them to consume plant material. This research provides a knowledge base from which many more studies can stem. Characterization of these enzymes will be useful in understanding the biochemical feeding of this pest. This should be done by challenging the enzymes against specific inhibitors, assessing the optimal pH of activity, and/or sequencing (Zeng et al. 2002). To further understand HWA feeding, enzyme assays should be expanded to include others that are known to be significant in
119
feeding. Pectinases are common in the salivary glands of many phytophagous insects, including at least 24 aphid species (Adams and McAllen 1956, 1958), and are important in degradation of the middle lamella between plant cells. Aphids can insert their stylets into single sieve tube or epidermal cells without damage and their stylets frequently pass between cells to reach the target tissue (Pollard 1973). Pectinase is useful in maneuvering of the stylets between those cells that are held together by pectin complexes to ultimately reach their feeding site (Bateman and Millar 1968, Agusti and Cohen 2000). We may also expect to find evidence of α-glucosidase or possibly invertase activity, common enzymes in the gut of phytophagous insects used in digestion of sucrose, maltose, and trehalose (Miles 1999,
Agusti and Cohen 2000).
From the perspective of host-plant resistance, resistant and susceptible hemlocks should be examined for the presence of enzyme inhibitors specific to the enzymes detected in
HWA in this study. In addition, there is potential for the development of a transgenic hemlock that includes an enzyme inhibitor to decrease HWA populations and/or prohibit reproduction, similar to those transgenic crops developed for resistance to aphids (Rahbé et al. 2003, Azzouz et al. 2005). Nonetheless, this research adds to our knowledge of HWA feeding behaviors and should be expanded to further increase our understanding and capabilities to manage them.
ACKNOWLEDGMENTS
We thank Leo Magalhaes for donating Lygus lineolaris as positive controls, Wes
Watson and the Medical and Veterinary Entomology Lab at NCSU for use of analytical
120
equipment, and J.C. Domec for assistance imaging the xylem ray parenchyma cells. This research was funded by the United States Department of Agriculture Forest Service,
Northern Research Station and Forest Health Protection in Region 8.
LITERATURE CITED
Adams, J.B., and M. Drew. 1963. A cellulose-hydrolyzing factor in aphid saliva. Can. J. Zool. 41: 489-496.
Adams, J.B., and J.W. McAllan. 1956. Pectinase in the saliva of Myzus persicae (Sulz.). Can. J. Zool. 34: 541-542.
Adams, J.B., and J.W. McAllan. 1958. Pectinase in certain insects. Can. J. Zool. 36: 305- 308.
Agusti, N., and A.C. Cohen. 2000. Lygus hesperus and L. lineolaris (Meniptera: Miridae), phytophages, zoophages, or omnivores: evidence of feeding adaptations suggested by the salivary and midgut digestive enzymes. J. Entomol. Sci. 35: 176-186.
Alhaddad, H., T.A. Coudron, E.A. Backus, and F. Schreiber. 2011. Comparative behavioral and protein study of salivary secretions in Homalodisca spp. sharpshooters (Hemiptera: Cicadellidae: Cicadellinae). Ann. Entomol. Soc. Am. 104: 543-552.
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac. Entomol. 1: 79-82.
Auclair, J.L. 1963. Aphid feeding and nutrition. Annu. Rev. Entomol. 8: 439-490.
Azzouz, H., A. Cherqui, E.D.M. Campan, Y. Rahbé, G. Duport, L. Jouanin, L. Kaiser, and P. Giordanengo. 2005. Effects of plant protease inhibitors, oryzacystatin I and soybean Bowman–Birk inhibitor, on the aphid Macrosiphum euphorbiae (Homoptera, Aphididae) and its parasitoid Aphelinus abdominalis (Hymenoptera, Aphelinidae). J. Insect Physiol. 51: 75- 86.
Balch, R.E. 1952. Studies on the balsam woolly aphid, Adelges piceae (Ratz.) and its effects on balsam fir, Abies balsamea (L.) Mill. Can. Dept. Agr., Publ. No. 867.
Baptist, B.A. 1941. The morphology and physiology of the salivary glands of Hemiptera- Heteroptera. Q. J. Microscop. Sci. 83: 91-139.
121
Bateman, D.F., and R.L. Millar. 1968. Pectic enzymes in tissue degradation. Annu. Rev. Plant Physiol. 19: 119-146.
Binnington K.C, and M. Schotz M. 1973. Automatic collection of tick saliva. J. Australian Entomol. Soc. 12: 78-79.
Blagowestschenski, A.W. 1955. Die biochemischen Grundlagen des Evolutionsprozesses der Pflanzen. Akademie-Verlig, Berlin, Germany.
Bramstedt, F. 1948. Uber die Verdauungsphysiologie der Aphiden. Zeitschrift für Naturforschung 3b: 14-24.
Butts, R.A., and R.J. Lamb. 1991. Seasonal abundance of three Lygus species (Heteroptera: Miridae) in oilseed rape and alfalfa in Alberta. J. Econ. Entomol. 84: 450-456.
Campbell, B.C. 1990. On the role of microbial symbiotes in herbivorous insects, pp. 1-44. In E.A. Bernays (ed.), Insect-plant interactions, vol. I. CRC Press, Inc., Boca Raton, FL.
Cherqui, A., and W.F. Tjallingii. 2000. Salivary proteins of aphids, a pilot study on identification, separation and immunolocalisation. J. Insect Physiol. 46: 1177-1186.
Clark, M.A., L. Baumann, M.A. Munson, P. Bauman, B.C. Campbell, J.E. Duffus, L.S. Osborne, and N.A. Moran. 1992. The eubacterial endosymbionts of whiteflies (Homoptera: Aleyrodoidea) constitute a lineage distinct from the endosymbionts of aphids and mealybugs. Curr. Microbiol. 25: 119-123.
Cohen, A.C. 1990. Feeding adaptations of some predaceous Hemiptera. Ann. Entomol. Soc. Am. 83: 1215-1223.
Cohen, A.C. 1993. Organization of digestion and preliminary characterization of salivary trypsin-like enzymes in a predaceous heteropteran, Zelus renardii. J. Insect Physiol. 39: 823- 829.
Cohen, A.C. 1995. Extra-oral digestion in predaceous terrestrial Arthropoda. Annu. Rev. Entomol. 40: 85-103.
Cohen, A.C. 1998. Solid-to-liquid feeding: the inside(s) story of extra-oral digestion in predaceous Arthropoda. Am. Entomol. 44: 103-117.
Cohen, A.C., and A.G. Wheeler, Jr. 1998. Role of saliva in the highly destructive fourlined plant bug (Hemiptera: Miridae: Mirinae). Ann. Entomol. Soc. Am. 91: 94-100.
122
Cooper, W.R., J.W. Dillwith, and G.J. Puterka. 2010. Salivary proteins of Russian wheat aphid (Diuraphis noxia Kurdjumov, Hemiptera,: Aphididae). Environ. Entomol. 39: 223-231.
Cooper, W.R., J.W. Dillwith, and G.J. Puterka. 2011. Comparisons of salivary proteins from five aphid (Hemiptera: Aphididae) species. Environ. Entomol. 40: 151-156.
Corzo, G., S. Adachi-Akahane, T. Nagao, Y. Kusui, ad T. Nakajima. 2001. Novel peptides from assassin bugs (Hemiptera: Reduviidae): isolation, chemical and biological characterization. FEBS Lett. 499: 256-261.
Davidson, J. 1923. Biological studies of Aphis rumicis L. The penetration of plant tissues and the source of food supply in aphids. Ann. Appl. Biol. 10: 35-54.
Dietrich, R.A., K. Lawton, L. Friedrich, R. Cade, M. Willits, and K. Maleck. 1999. Induced plant defense responses: scientific and commercial development possibilities. In D.J. Chadwick, and J.A. Goode (eds.), Insect-plant interactions and induced plant defence. Wiley, Chichester, England.
Douglas, A.E. 1989. Mycetocyte symbiosis in insects. Biol. Rev. 64: 409-434.
Duspiva, F. 1953. Der Kohlenhydratumsatz im Verdauungstrakt der Rhynchoten, ein Beitrag zum Problem der stofflichen Wechselbeziehungen zwischen saugenden Insekten und ihren Wirtspflanzen. Mitt. biol. Zen Anst., Beri., 75: 82-89.
Foissac, X., M.G. Edwards, J.P. Du, A.M.R. Gatehouse, and J.A. Gatehouse. 2002. Putative protein digestion in a sap-sucking homopteran plant pest (rice brown plant hopper; Nilaparvata lugens: Delphacidae) - identification of trypsin-like and cathepsin B-like proteases. Insect Biochem. Molec. Biol. 32: 967-978.
Garcia-Olmedo, F., G. Salcedo, R. Sanchez-Monge, L. Gomez, J. Royo, and P. Carbonero. 1987. Plant proteinaceous inhibitors of proteinases and αamylases. Oxf. Surv. Plant Mol. Cell Biol. 4: 275-334
Gouinguené, S., H. Alborn, and T.C.J. Turlings. 2003. Induction of volatile emissions in maize by different larval instars of Spodoptera littoralis. J. Chem. Ecol. 29: 145-162.
Green, T.R., and C.A. Ryan. 1972. Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175: 776-777.
Hagen, H.E., J. Grunewald, J., and P.J. Hamm. 1994. Induction of the prophenoloxidase- activating system of Simulium (Diptera: Simuliidae) following Onchocerca (Nematoda: Filaroidea) infection. Parasitol. 109: 649–655.
123
Halitschke, R., U. Schittko, G. Pohnert, W. Boland, and I.T. Baldwin. 2001. Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. III. Fatty acid-amino acid conjugates in herbivore oral secretions are necessary and sufficient for herbivore-specific plant responses. Plant Physiol. 125: 711-717.
Heath, M.C. 2000. Hypersensitive response-related death. Plant Molec. Biol. 44: 321-334.
Hori, K. 1970. Some properties of amylase in the salivary gland of Lygus disponsi (Hemiptera). J. Insect Physiol. 16: 373-386.
Hori, K. 1971. Studies on the feeding habits of Lygus disponsi Linnavuori (Hemiptera: Miridae) and the injury to its host plants. I. Histological observations of the injury. Appl. Entomol. Zool. 6: 84-90.
Jiang, Y., and P.W. Miles. 1993. Generation of H2O2 during enzymic oxidation of catechin. Phytochem. 33: 29-34.
Jongsma, M.A., and J. Beekwilder. 2008. Plant protease inhibitors: functional evolution for defense, pp. 235-252. In A. Schaller (ed.), Induced plant resistance to herbivory. Springer, Dordecht, Netherlands.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
Kingsolver, J.G., and T.L. Daniel. 1993. Mechanics of fluid feeding in insects, pp. 149- 161. In C.W. Schaefer, and R.A.B. Leschem (eds.), Functional morphology of insect feeding. Proc. Thomas Say Publications in Entomology, Entomol. Soc. Am., Lanham, MD.
Koning, R.E. 1994. Secondary growth. Plant physiology information website. http://plantphys.info/plant_biology/secondary.shtml.
Laurema, S., A. Varis, and H. Miettinen. 1985. Studies on enzymes in the salivary glands of Lygus rugulipennis (Hemiptera, Miridae). Insect Biochem. 15: 211-224.
Lee, K.P., S.J. Simpson, and K. Wilson. 2008. Dietary protein-quality influences melanization and immune function in an insect. Funct. Ecol. 22: 1052-1061.
Leszczynski, B., A. Urbanska, H. Matok, I. Leskowska, B. Jozwiak, M. Wozniak, and S. Niraz. 1996. Metabolism of cereal xenobiotics by grain aphid. Exp. Toxic Pathol. (Suppl. 8) 48: 238-243.
124
Liu, X., C.E. Williams, J.A. Nemacheck, H. Wang, S. Subramanyam, C. Zheng, and M.- S. Chen. 2010. Reactive oxygen species are involved in plant defense against a gall midge. Plant Physiol. 152: 985-999.
Madhusudhan, V.V., and P.W. Miles. 1998. Mobility of salivary components as a possible reason for differences in the responses of alfalfa to the spotted alfalfa aphid and pea aphid. Entomol. Exp. Appl. 86: 25-39.
Madhusudhan, V.V., G.S. Taylor, and P.W. Miles. 1994. The detection of salivary enzymes of phytophagous Hemiptera: a compilation of methods. Ann. Appl. Biol. 124: 405- 412.
Marshall, J.J. 1975. α-Amylase inhibitors from plants, pp. 244-266. In A. Jeanes, and J. Hodge (eds.), Physiological effects of food carbohydrates. CS Symposium Series, American Chemical Society, Washington, DC.
Mattiacci, L., M. Dicke, and M.A. Posthumus. 1995. β-Glucosidase: an elicitor of herbivore-inducible plant odors that attract host-searching parasitic wasps. Proc. Nat. Acad. Sci. U.S.A. 92: 2036-2046.
Mauseth, J.D. 2009. Botany: an introduction to plant biology, 4th ed. Jones and Bartlett Publishers, LLC, Subury, MA.
McClure, M.S. 1987. Biology and control of hemlock woolly adelgid. Bull. Conn. Agric. Exp. Stn. 851.
McClure, M.S. 1991. Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis. Environ. Entomol. 20: 258-264.
McClure, M.S. 1992. Hemlock woolly adelgid. Am. Nurseryman 175: 82-89.
McClure, M.S. 1995. Managing hemlock woolly adelgid in ornamental landscapes. Conn. AES Bull. 925.
Miles, P.W. 1959. Secretion of two types of saliva by an aphid. Nature 183: 756.
Miles, P.W. 1960. The salivary secretions of a plant-sucking bug, Oncopeltus fasciatus (Dall.) (Heteroptera: Lygaeidae). II. Physical and chemical properties. J. Insect. Physiol. 4: 209-219.
Miles, P.W. 1964. Studies on the salivary physiology of plant-bugs: oxidase activity in the salivary apparatus and saliva. J. Insect Physiol. 10: 121-129.
125
Miles, P.W. 1965. Studies on the salivary physiology of plant bugs: the salivary secretion of aphids. J. Insect Physiol. 11: 1261-1268.
Miles, P.W. 1967. The physiological division of labour in the salivary glands of Oncopeltus fasciatus (Dall.) (Heteroptera: Lygaeidae). Aust. J. Biol. Sci. 20: 785-797.
Miles, P.W. 1972. The saliva of Hemiptera. Adv. Insect Physiol. 9: 183-255.
Miles, P.W. 1990. Aphid salivary secretions and their involvement in plant toxicoses, pp. 131-147. In R.K. Campbell, and R.D. Eikenbary, Aphid-plant genotype interactions. Elsevier, Amsterdam, The Netherlands.
Miles, P.W. 1999. Aphid saliva. Biol. Rev. Camb. Philos. Soc. 74: 41-85.
Miles, P.W., and P. Harrewijn, P. 1991. Discharge by aphids of soluble secretions into dietary sources. Entomol. Exp. Appl. 59: 123-134.
Miles, P.W., and A. Helliwell. 1961. Oxidase activity in the saliva of a plant bug. Nature 192: 374.
Miles, P.W., and D. Slowiak. 1970. Transport of whole protein molecules from blood to saliva of a plant-bug. Experientia 26: 611-612.
Mittler, T.E. 1967. Flow relationships for hemipterous stylets. Ann. Entomol. Soc. Am. 60: 1112-1114.
Musser, R.O., S.M. Hum-Musser, H. Eichenseer, M. Peiffer, G. Ervin, J.B. Murphy, and G.W. Felton. 2002. Caterpillar saliva beats plant defences - a new weapon emerges in the evolutionary arms race between plants and herbivores. Nature 416: 599-600.
Mutti, N.S., J. Louis, L.K. Pappan, K. Pappan, K. Begum, M.S.C. Ming-Shun, Y. Park, N. Dittmer, J. Marshall, J.C. Reese, and G.R. Reeck. 2008. A protein from the salivary glands of the pea aphid, Acyrthosiphon pisum, is essential in feeding on a host plant. Proc. Nat. Acad. Sci. U.S.A. 105: 9965-9969.
Nobel, P.S. 1983. Biophysical plant physiology and ecology. W.H. Freeman, San Fransisco, CA.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
126
Ourth, D.D., and H.E. Renis. 1993. Antiviral melanisation reaction of Heliothis virescens haemolymph against DNA and RNA viruses in vitro. Comp. Biochem. Physiol. 105B: 719– 723.
Pallardy, S.G. 2008. Physiology of woody plants, 3rd ed. Academic Press, Burlington, MA.
Peng, Z., and P.W. Miles. 1988. Studies on the salivary physiology of plant-bugs: function of the catechol oxidase of the rose aphid. J. Insect Physiol. 34: 1027-1033.
Pereira, M.H., M.E.L. Souza, A.P. Vargas, M.S. Martins, C.M. Penido, and L. Diotaiuti. 1996. Anticoagulant activity of Triatoma infestans and Panstrongylus megistus saliva (Hemiptera/Triatominae). Acta Trop. 61: 255–261.
Pollard, D.G. 1973. Plant penetration by feeding aphids (Hemiptera, Aphidoidea): a review. Bull. Entomol. Res. 62: 631-714.
Radville, L., A. Chaves, and E.L. Preisser. 2011. Variation in plant defense against invasive herbivores: Evidence for a hypersensitive response in eastern hemlocks (Tsuga canadensis). J. Chem. Ecol. 37: 592-597.
Rahbé, Y., C. Deraison, M. Bonadé-Bottino, C. Girard, Christiane Nardon, and L. Jouanin. 2003. Effects of the cysteine protease inhibitor oryzacystatin (OC-I) on different aphids and reduced performance of Myzus persicae on OC-I expressing transgenic oilseed rape. Plant Sci. 164: 441-450.
Robyt, J.F. 1998. Essentials of carbohydrate chemistry. Springer, New York.
Rowley, A.F., J.L. Brookman, and N.A. Ratcliffe. 1990. Possible involvement of the prophenoloxidase system of the locust, Locusta migratoria, in antimicrobial activity. J. Invertebr. Pathol. 56: 31–38.
Ryan, C.A. 1990. Protease inhibitors in plants: genes for improving defenses against insects and pathogens. Annu. Rev. Phytopathol. 28: 425-449.
Ryan, C.A. 2000. The systemin signaling pathway: differential activation of plant defensive genes. Biochim. Biophys. Acta 1477: 112-121.
Schoonhoven, L.M., J.J.A. van Loon, and M. Dicke. 2005. Insect-plant biology. Oxford University Press, New York.
127
Shields, K.S., and R.T. Hirth. 2005. Bacterial endosymbionts of Adelges tsugae Annand: potential targets for biocontrol? In B. Onken, and R. C. Reardon (compilers), Third Symposium on hemlock woolly adelgid in the eastern United States, 2005 February 1-3, Asheville, NC. USDA Forest Service, FHTET-2005-01, Morgantown, WV: 357-359.
Shields, K.S., R.F. Young, and G.P. Berlyn. 1996. Hemlock woolly adelgid feeding mechanisms. In S. M. Salom, T. C. Tigner, and R. C. Reardon (eds.), Proceedings of the first hemlock woolly adelgid review, 1995 October 12, Charlottesville, VA. USDA Forest Service, FHTET 96-10, Morgantown, WV: 36-41.
Smith, K.M. 1926. A comparative study of the feeding methods of certain Hemiptera and of the resulting effects of plant tissue, with special reference to the potato plant. Ann. Appl. Biol. 13: 109-136.
Snodgrass, G.L., W.P. Scott, and J.W. Smith. 1984. Host-plants and seasonal distribution of the tarnished plant bug (Hemiptera: Miridae) in the Delta of Arkansas, Louisiana, and Mississippi. Environ. Entomol. 13: 110-116.
Srivastava, P.N. 1987. Nutritional physiology, pp. 99-122. In A. K. Minks and P. Harrewijn (eds.), Aphids: Their biology, natural enemies and control, Volume 2A. Elsevier, Amsterdam, The Netherlands.
Srivastava, P.N., and J.L. Auclair. 1963. Characteristics and nature of proteases from the alimentary canal of the pea aphid,Acyrthosiphon pisum (Harr.) (Homoptera, Aphididae). J. Insect Physiol. 9: 469-474.
Strong, F.E., and E.C. Kruitwagen. 1968. Polygalacturonase in the salivary apparatus of Lygus hesperus (Hemiptera). J. Insect Physiol. 14: 1113-1119.
Terra, W.R., and C. Ferreira. 1994. Insect digestive enzymes: properties, compartmentalization and function. Comp. Biochem. Physiol. 109B: 1–62.
Tingey, W.M., and E.A. Pillemer. 1977. Lygus bugs: crop resistance and physiological nature of feeding injury. Bull. Ent. Soc. Am. 23: 277-287.
Urbanska, A., W.F. Tjallingii, A.F.G. Dixon, and B. Leszczynski. 1998. Phenol oxidising enzymes in the grain aphid's saliva. Entomol. Exp. Appl. 86: 197-203.
Urbanska, A., W.F. Tjallingii, and B. Leszczynski. 1994. Application of agarose-sucrose gels for investigation of aphid salivary enzymes, pp. 81-87. In K. Hamilton et al. (eds), Aphids and other homopterous insects, vol. 4. Polish Academy of Science, Skierniewice.
128
van der Maarel, M.J.E.C., B. ven der Veen, J.C.M. Uitdehaag, H. Leemhuis, and L. Dijkhuizen. 2002. Properties and applications of starch-converting enzymes of the α- amylase family. J. Biotech. 94: 137-155.
Walker-Lane, L.N. 2009. The effect of hemlock woolly adelgid infestation on water relations of Carolina and eastern Hemlock. M.S. thesis. North Carolina State University, Raleigh.
Wiedenhoeft, A.C., and R.B. Miller. 2005. Structure and function of wood, pp. 9-33. In R.M. Rowell (ed.), Wood chemistry and wood composites. CRC Press, Boca Raton, FL.
Will, T., W.F. Tjallingii, A. Thönnessen, and A.J.E. van Bel. 2007. Molecular sabotage of plant defense by aphid saliva. Proc. Nat. Acad. Sci. U.S.A. 104: 10536-10541.
Young, O.P. 1986. Host plants of the tarnished plant bug, Lygus lineolaris (Heteroptera: Miridae). Ann. Entomol. Soc. 79: 747-762.
Young, R.F., Shields, K.S., and G.P. Berlyn. 1995. Hemlock woolly adelgid (Homoptera: Adelgidae): stylet bundle insertion and feeding sites. Ann. Entomol. Soc. Am. 88: 827-835.
Zeng, F., and A.C. Cohen. 2000. Partial characterization of α-amylase in the salivary glands of Lygus hesperus and L. lineolaris. Comp. Biochem. Phys. B 126: 9-16.
Zeng, F., Y. Zhu, and A. Cohen. 2002. Partial characterization of trypsin-like protease and molecular cloning of a trypsin-like precursor cDNA in salivary glands of Lygus lineolaris. Comp. Biochem. Phys. B 131: 453-463.
Zweigelt, F. 1914. Beiträge zur Kenntnis des Saugphänomens der Blattläuse und der Reaktionen der Pflanzenzellen. Zentr. Bakteriol., Parasitenk. 42: 265-335.
129
Figure 1. Enzyme kinetics of trypsin-like assay.
130
0.9 0.8405 0.8
0.7
0.6
0.5
0.4
0.3 0.259
0.2 BAEE U/µg protein/min BAEE U/µg
0.1
0 A. tsugae L. lineolaris Species
Figure 2. Trypsin-like activity per µg protein per minute of A. tsugae and L. lineolaris.
131
Figure 3. Enzyme kinetics of amylase-like enzyme assay.
132
0.9
0.8 0.7748
0.7
0.6
0.5
0.4
0.3 mU/µg protein/min mU/µg
0.2 0.1088 0.1
0 A. tsugae L. lineolaris Species
Figure 4. Amylase-like activity per µg protein per minute of A. tsugae and L. lineolaris.
133
Figure 5. Enzyme kinetics of peroxidase assay.
134
0.25 0.2323
0.2
0.15 0.1283
0.1 mU/µg protein/min mU/µg
0.05
0 A. tsugae L. lineolaris
Figure 6. Peroxidase activity per µg protein per minute of A. tsugae and L. lineolaris.
135
Figure 7. Enzyme kinetics of polyphenol oxidase assay.
136
0.06 0.0569
0.05
0.04
0.03
0.02 0.017 Abs/µg protein/min Abs/µg
0.01 0.00323
0 A. tsugae A. tsugae (with PTU) L. lineolaris Species
Figure 8. Polyphenol oxidase activity per µg protein per minute for A. tsugae, A. tsugae incubated with PTU for 20 min, and L. lineolaris.
137
ray parenchyma
Figure 9. Radial view of xylem ray parenchyma cells of eastern hemlock, Tsuga canadensis.
138
Chapter V
Biophysical Characteristics of Six Hemlock Species' (Tsuga spp.) Surfaces: Implications for Resistance to the Hemlock Woolly Adelgid, Adelges tsugae (Hemiptera: Adelgidae)
ABSTRACT
Characteristics of the plant surface significantly affect host-plant selection by phytophagous insects. Surface morphology of six hemlock species (Tsuga spp.) and a hybrid was investigated using low-temperature scanning electron microscopy. Observations focused on trichome presence and placement, wax crystalline micromorphology, and cuticle thickness. These characteristics were studied in the context of species-level host-plant resistance to Adelges tsugae (Hemiptera: Adelgidae), an exotic insect responsible for massive mortality of eastern hemlock (Tsuga canadensis) and Carolina hemlock (T. caroliniana) in the eastern United States. Hemlock species in the native range of the insect do not succumb to infestations and intermittent survivors of eastern and Carolina hemlocks show signs of innate resistance. Mechanisms of inter- and intraspecific variation in resistance are unknown.
We found non-glandular tichomes present in both resistant and susceptible hemlocks, indicating that trichomes are likely not involved in resistance to A. tsugae. Cuticle thickness is significantly thinner at the stylet bundle insertion point of A. tsugae than on the side and outside of the pulvinus. It is possible that a thinner cuticle more easily facilitates tissue penetration by the stylets, which may be the reason the insect is specific in its insertion site.
In addition, cuticle thickness at this point is significantly different between species, with the hybrid, Carolina, southern Japanese, and northern Japanese hemlocks possessing the thickest cuticles of observed species. Chinese hemlock is significantly thinner than the hybrid, but
139
not than the other three species mentioned. Eastern, western, and mountain hemlocks constitute the group with the thinnest cuticle. Cuticle thickness may be one level of resistance to HWA feeding. Epicuticular wax morphology appears to be smoother in susceptible species, and thus may be involved in insect-plant interactions by facilitating A. tsugae feeding, especially sheath adhesion to the plant surface. In addition, wax morphology is typically linked to chemistry, which could be involved in host plant selections and/or resistance mechanisms. More samples should be analyzed to corroborate these results.
INTRODUCTION
The plant surface is the primary interface between an herbivore and its host and consequently plays an important role in host-plant selection and plant defense. Its physical features are involved in these behavioral processes of major pest insects. For example, the initial decision for many aphids to insert their stylets involves analysis of the texture and color of the plant surface (Ibbotson & Kennedy 1959; Klingauf 1987; Pelletier 1990).
Because of their role in behavioral acceptance by phytophagous insects, physical features are also understood to confer resistance in some plant species.
Trichomes and hairs of the plant surface are understood to act as a resistant trait against herbivorous insects (Johnson 1975; Levin 1973). Through mechanical and chemical functions, they can affect movement, attachment, shelter, feeding, and survival of insects and are one of the important morphological parameters of plant resistance to insects (Johnson
1975; Ram et al. 2005). For example, a hairy plant surface can prevent or slow insect movement, but a bald surface can be slippery and cause insects to fall off (Southwood 1986).
140
Conversely, trichome absence can increase susceptibility to herbivore damage as well; experimental removal of hairs from willow leaves increased accessibility for leaf beetle larvae (Rowell-Rahier and Pasteels 1982). Resistance of mustard varieties, Brassica juncea
L., to the mustard aphid, Lipaphis erysimi (Kalt.), is correlated to hairiness of the leaf surface
(Lal et al. 1999). Glandular trichomes may secrete repellent substances or secondary metabolites that can trap insects and hooked trichomes can impale pests (Levin 1973; Duffey
1986; Musetti and Neal 1997). The latter characteristic has been exploited for bed bug management -- the underside of bean leaves possess hooked hairs and are placed upside down in areas infested with bed bugs. Wandering insects become entangled in the trichomes and subsequently die from starvation (Richardson 1943). Pubescent wheat varieties have also been found to be less suitable for oviposition and larval feeding by the cereal leaf beetle,
Oulema melanopus L., thus initiating a breeding program for resistant varieties based on morphological characters (Gallun 1966; McDaniel and Janke 1973).
Epicuticular waxes make up the outermost layer of a plant, thus the point of contact between an herbivore and its host which can influence subsequent host acceptance or rejection (Tomaszewski 2004; Daoust et al. 2010). One of the primary physical defenses against herbivores is the production of thick and/or tough tissues which deter initiation of feeding, such as stylet penetration (Larsson 2002; Krokene et al. 2008). Thick cuticular layers and epicuticular waxes confer resistance to a variety of different plants by making it difficult for an herbivore to penetrate, consume, or digest plant material (Van and Guan
1959; Patanakamjorn and Pathak 1967; Blum 1968; Schalk et al. 1986; Southwood 1986;
Bodnaryk and Lamb 1991; Dussourd 1993; Yang et al. 1993; Peter and Shanower 2001; Ram
141
et al. 2005). For example, a thick cortex (the layer just below the epidermis) on the stem of wild tomato deters feeding by the potato aphid, Macrosiphum euphoribiae (Thomas) (Quiras et al. 1977). The thickness of the cuticle is highly variable among plants and can range from less than one µm to 15 µm. In shade-tolerant plants, the cuticle is generally thinner than those exposed to full sun (Romberger et al. 1993; Kozlowski and Pallardy 1997). It is also key in desiccation prevention; pine needles are covered in a thick cuticle, which makes them markedly more resistant to drought than most trees (Eglinton and Hamilton 1967; Raven et al. 1986).
At the microscopic level, the structure of the waxes varies and can be a useful taxonomic tool (Raven et al. 1986; Romberger et al. 1993). Moreover, their physical make- up can influence host-plant acceptance by a phytophagous pest. The morphology and abundance of the crystalline waxes in poplar hybrids confers resistance to the aphid
Chaitophorus leucomelas (Alfaro-Tapia et al. 2007).
Microscopic observations of unaltered plant surface morphology require appropriate preparation techniques. Tissue preparation for conventional scanning electron microcopy
(SEM) uses chemical fixation and critical point drying of sample tissue. This often degrades the epicuticular waxes of the plant surface. Cryofixation, followed by low-temperature SEM
(LT-SEM) studies, can be used to preserve plant surface details and is considered superior to conventional fixation methods (Parsons et al.1974; Eveling and McCall 1983; Sargent 1983).
This research investigates physical features of hemlock (Tsuga spp.) that may be involved in host-plant resistance to the hemlock woolly adelgid (HWA), Adelges tsugae
Annand. Native to parts of Asia and northwestern North America (Havill et al. 2006), HWA
142
is an invasive pest first detected in the eastern U.S. in the early 1950s (Souto et al. 1996). Its range has since spread to 18 eastern states and continues to expand (USDA - Forest Service
2011). Of the nine species of hemlock worldwide, the two species in eastern North America, eastern and Carolina hemlock (Tsuga canadensis (L.) Carrière and T. caroliniana Engelm.), are the only species thought to succumb to infestations. However, in the wake of large-scale destruction, anecdotal evidence suggested surviving individuals or stands of eastern and
Carolina hemlocks may be less susceptible. This has since been corroborated by research and distinct individuals have been selected and continue to be pursued as putatively resistant to the adelgid (Caswell et al. 2008; Jetton et al. 2008; Kaur 2009; Ingwell and Preisser 2010;
Oten et al. 2011) (Table 1). The ability for HWA to establish large populations and overcome its host is dependent on its abilities to successfully penetrate and feed on host tissues, the xylem ray parenchyma cells (Young et al. 1995).
The objective of this research is to elucidate the role of physical characteristics of resistant hemlocks to HWA. We discuss morphological characteristics of different hemlock species using light microscopy and LT-SEM. No descriptions of hemlock wax structure, cuticle thickness, or trichomes have been made previously. We hypothesize that resistant hemlock will have a higher density of trichomes than susceptible hemlock species, that the cuticle is thinner in susceptible hemlock species, and that wax structure will correlate to susceptiblity level. Understanding the hemlock surface has implications for mechanisms of host-plant resistance and may advance resistant hemlock breeding efforts.
143
MATERIALS AND METHODS
Plant material
We observed samples from seven species and a hybrid of variable resistance levels:
Tsuga canadensis (uninfested and infested samples), T. caroliniana, T. chinensis (Franchet)
E. Pritzel, T. sieboldii Carrière, T. diversifolia Masters, T. heterophylla Sargent, T. mertensiana (Bong.), and T. chinensis x T. caroliniana hybrid. Hemlock material was collected from several sources based on availability and HWA infestation status (Table 1).
Samples collected locally were clipped from trees, placed in large bags, and transported to
North Carolina State University in Raleigh. For samples that were shipped, clipped ends were wrapped in moist paper towels, placed in Ziploc® bags, and shipped overnight on ice.
Transportation from Raleigh to LT-SEM facilities in Beltsville, MD occurred in a cooler with ice.
Low-temperature scanning electron microscopy (LT-SEM)
Low-temperature SEM observations were performed using an S-4700 field emission scanning electron microscope (Hitachi High Technologies America, Inc., Pleasanton, CA) equipped with a Quorum CryoPrep PP2000 (Energy Bean Sciences, East Grandby, CT) cryotransfer system. To prepare specimens, hemlock needles were excised at the abscission layer and stems cut directly above that point. They were mounted with the abscission layer up in cross-section arrangement on flat specimen holders consisting of 16x30mm copper plates that contained a thin layer of Tissue Tek (OCT Compound, Ted Pella, Inc., Redding,
CA), which acted as the cyro-adhesive upon freezing. The samples were frozen conductively, in a Styrofoam box, by placing the plates on the surface of a pre-cooled
144
° (-196 C) brass bar whose lower half was submerged in liquid nitrogen (LN2). After 20-30 s, the holders containing the frozen samples were transferred to a LN2 Dewar for future use or cryotransferred under vacuum to the cold stage in the pre-chamber of the cryotransfer system. Removal of any surface contamination (condensed water vapor) took place in the cryotransfer system by etching the frozen specimens for 10-15 min by raising the temperature of the stage to -90°C. Following etching, the temperature was lowered below
-130°C, and a magnetron sputter head equipped with a platinum target, was used to coat the specimens with a very fine layer of platinum. The specimens were transferred to a pre-cooled
(-140°C) cryostage in the SEM for observation. An accelerating voltage of 5kV was used to view the specimens. Images were captured using a 4pi Analysis System (Durham, NC).
Images were modified for brightness and contrast by adjusting levels, sized, and placed together to produce figures using Adobe Photoshop 8.0.
Measurements and analyses
To measure cuticle thickness, LT-SEM micrographs were taken at three positions on the pulvinus: at the HWA stylet insertion point (Fig. 1A), on the side where the HWA body would typically rest, and on the outside (Fig. 1B). Thickness varied slightly; to compensate for variability, up to three points from each image were measured for cuticle thickness (µm).
Statistical analyses were performed for each location (i.e. insertion point, side, outside) for T. canadensis, T. caroliniana, T. chinensis, T. diversifolia, T. heterophylla, T. mertensiana. For the analysis at the stylet bundle insertion point only, T. sieboldii and the T. chinensis x T. caroliniana hybrid were added (measurements at the side and outside were not taken for these samples). To stay within logicial contrainsts of the study, a single individual was
145
sampled for each species and at least five cross-sections were prepared. Cross-sections were taken from the most recently produce woody material of the stem.
Analysis of variance was performed using the Fit Least Squares procedure of JMP 9.0
Pro (SAS Institute 2010). The main effects of hemlock species and position were tested for cuticle thickness. All pair-wise least square means were compared by the Tukey HSD procedure.
To examine the ultrastructure of epicuticular waxes, LT-SEM micrographs were taken of both sides of the needle, above and below the abscission layer, at up to 5000x. Wax structures were categorized as suggested by Barthlott et al. (1998).
RESULTS
Trichomes
Trichomes were observed on seven of the eight hemlock species. The only species lacking trichomes is T. sieboldii, which has a smooth, hairless surface (Fig. 2D). Tsuga canadensis, T. diversifolia, T. heterophylla, and T. mertensiana have trichomes encircling the stem (Figs. 2A, 2E, 2F, 2G). In T. heterophylla and T. mertensiana, the two western North
American species, we observed relatively long hairs as compared to all other samples observed. Tsuga caroliniana, T. chinensis, and T. chinensis x T. caroliniana are mostly smooth with a conspicuous area of trichomes that extend longitudinally along the length of the stem (Figs. 2B, 2C, 2H).
146
Cuticle thickness
Electron micrographs of the HWA insertion point for each hemlock species studied are shown in Figure 3. ANOVA was performed on the cuticle thickness measurements from the six hemlock species from which multiple locations on the pulvinus were sampled; the effects for species, location, and location x species were all significant. The main effects of species and location are presented in Fig. 4 and Fig 5, respectively. Mean cuticle thickness for all measurement locations (insertion point, side, outside) are clustered in three distinct groups: T. canadensis, T. heterophylla, and T. mertensiana are grouped with the thinnest overall cuticle; T. caroliniana and T. diversifolia are grouped with intermediate measurements; and T. chinensis is isolated with the thickest mean cuticle (Fig. 4). In addition, a comparison of cuticle thickness averaged across species indicates that the cuticle is thickest at the side and outside measurement points relative to the insertion point (Fig. 5).
Analysis of cuticle thickness at the insertion point only indicates that the T. chinensis x T. caroliniana hybrid has the thickest cuticle at the insertion point, but it is not statistically different from T. caroliniana, T. sieboldii, and T. diversifolia (Figure 6). Relative to its parent species, the hybrid is similar in thickness to T. caroliniana but is thicker than T. chinensis. Tsuga canadensis (infested) has the thinnest cuticle at the insertion point, but is not statistically significant from T. canadensis (uninfested), and the two western North
American species, T. heterophylla and T. mertensiana. At the pulvinus side, T. chinensis has the thickest cuticle of all other species observed (Fig. 7). The second thickest group consisted of T. caroliniana, T. diversofolia, and T. mertensiana. Tsuga mertensiana was not statistically different from T. heterophylla at α = 0.05. Tsuga canadensis (infested) had the
147
thinnest cuticle at the side measurement, but was not statistically different from T. heterophylla. At the needle outside, a similar trend was observed, with T. chinensis exhibiting the thickest wax (Fig. 7). Tsuga caroliniana and T. diversifolia are grouped with the second thickest wax at the outside point, but T. caroliniana is not statistically different from T. canadensis (infested), T. heterophylla, and T. mertensiana at α = 0.05. The interaction between species and location was significant at α = 0.05.
Wax structure
Most of the samples have tubule crystalloids on their cuticle surface (Fig. 8A), but several species presented exceptions to this. Tsuga canadensis has variable wax structure on the needle petiole of the adaxial side (the side of HWA stylet insertion); wax was relatively smooth in two samples (Fig. 8B), had minimal tubules in another, and conspicuous tubules but only in the trough of cuticular ridges (Fig. 8C). The abaxial side of the needle is smooth in all samples observed. Similarly, T. caroliniana has mostly smooth layers on the adaxial side of the pulvini with a few patches of tubules; on the petiole, some samples exhibited dense coverage with tubules and some were mostly smooth with scattered platelets (Fig. 8D).
On the abaxial side, the pulvini had tubules or were smooth and the petiole was covered in tubules. Samples of T. chinensis show most surfaces covered by tubule crystalloids; the adaxial side of the pulvinus and petiole have patches of smooth layer or film with a tiny bump-like texture (Fig. 8E). On the adaxial side of the needle, T. sieboldii is covered with clustered tubules. On the abaxial side of the pulvinus, T. diversifolia has an irregular surface, most similar to granules (Fig. 8F). On the adaxial side, pulvini are smooth or may have polygonal platelets (Fig. 8G). Tsuga heterophylla samples were also diverse and the only
148
sample in which both new and old growth was observed. On the abaxial side, samples of new growth were laden with tubule crystalloid wax and old growth exhibits smooth layers.
Samples of Tsuga mertensiana were covered in tubule crystalloid waxes on the abaxial and adaxial sides of the pulvini and the petioles. Tsuga chinensis x T. caroliniana exhibits unique crystalloids, similar to parallel oriented platelets but with more texture than are typically seen on platelets (Fig. 8H).
DISCUSSION
Based on these observations, trichomes are unlikely to account for the documented differences in resistance among hemlock species. These results are contrary to what we predicted. We expected trichomes to be related to hemlock susceptibility to HWA by one of two mechanisms: that the presence of trichomes would enable these small insects to overcome an 'attachment hurdle' by providing a substrate easier to grasp (Southwood 1973), or that trichome density or glandular trichomes deter HWA movement and access to feeding sites (Levin 1973; Duffey 1986; Carter 1982; Hoffman and McEvoy 1985). However, no correlation was found. There were differences in trichome density, placement, and length on the species studied, but they did not relate to degree of susceptibility to HWA. Tsuga canadensis has the most trichomes, yet HWA is clearly able to overcome them to feed successfully. Conversely, the surface of T. sieboldii, the hemlock species from which the invasive eastern U.S. population of HWA originates (Havill et al. 2006), is entirely smooth, indicating that trichomes do not facilitate feeding or act as a stimulant either. Unmistakably,
HWA successfully feeds on hemlocks with and without a dense covering of trichomes. It is
149
interesting to note the similarity in trichome arrangement of T. chinensis and T. caroliniana.
In both species, and the T. chinensis x T. caroliniana hybrid, the surface lacks trichomes except for a tangential band. Although geographically isolated from one another (see Table
1), phylogenetic studies indicate they are closely related species (Havill et al. 2008), indicating that trichome characteristics are likely commonly derived characters. Therefore, it is highly unlikely that a co-evolutionary history between HWA and trichome development in hemlocks occurred because we see such drastic difference between hemlocks that were able to co-evolve with HWA and those that could not (i.e., T. canadensis, T. caroliniana).
Based on overall cuticle thickness, T. chinensis stands apart from other species with the thickest cuticle (Fig. 4); it is also considered the most resistant of all hemlock species
(Table 1). Tsuga caroliniana and T. diversifolia, are second in overall thickness measurements and T. canadensis, T. heterophylla, and T. mertensiana are grouped third.
These characteristics correspond to the former's close phylogenetic relationship with T. chinensis (Havill et al. 2008). Tsuga caroliniana is nested within a clade consisting of the
Asian hemlocks, including T. chinensis and T. diversifolia. We suspect if we had successfully obtained measurements at all points for T. sieboldii, then it, too, would not be significantly different from this grouping. Tsuga heterophylla and T. mertensiana are close relatives to one another, and T. canadensis appears to be in a clade all its own, sister to Asian hemlocks. All three of these species are distinct from the Asian clade, corresponding to their placement as the group with the thinnest overall cuticle.
A thin cuticle may have implications for response to HWA. Generally, the Tsuga genus is the most shade tolerant of all Pinaceae genera (Farjon 1990) and shade tolerant
150
plants typically have a thinner cuticle than those requiring full sun (Romberger et al. 1993;
Kozlowski and Pallardy 1997). The thin cuticle observed in the T. canadensis group may be an adaptation it its riparian habitat but cause it to be more susceptible to water stress and drought symptoms, a consequence of HWA infestation (Walker-Lane 2009). However, T. heterophylla and T. mertensiana are not different from T. canadensis and they do not exhibit a hypersensitive response to HWA infestation. The generally thin cuticle of the Tsuga genus can be contrasted to needles of the genus Pinus, which are covered in a thick cuticle, making them markedly more resistant to drought than most trees (Eglinton and Hamilton 1967;
Raven et al. 1986). It is clear that a thick cuticle protects trees from desiccation and may offer additional protection from HWA attack.
Cuticle thickness measurements reveal that in all hemlock species, the cuticle is thinnest on the adaxial side of the pulvinus, where HWA consistently insert their stylets (Fig.
1A, 1B, Fig. 5). A thin cuticle may translate to a minimized obstacle for HWA to successfully penetrate host-plant tissue, and therefore, could be the motive for insertion at that specific location. No other theories have been formally developed to explain why HWA would target that specific spot. Cues used to recognize that exact location remain unknown, although sensory sensilla on the tip of HWA labium are probably strictly mechanoreceptors,
(K.L.F.Oten, A.C. Cohen, and F.P. Hain (submitted)), as they are in aphids (Forbes 1977;
Tjallingii 1978). These results not only insinuate host selection behavior by HWA and possible mechanisms of antixenosis resistance, but it can also be used in ongoing work to develop an artificial diet for HWA (Cohen et al. 2008; Kaur 2009). This data will be useful
151
to assess membrane potential in an artificial feeding delivery system and in the process, corroborate this research by evaluating artificial membranes of varying thicknesses.
In addition, there was an interaction between hemlock species and the location, indicating a relationship between cuticle thickness at each location based on species. The overall trend of cuticle thickness is not the same for each of the hemlock species observed.
This suggests an independent development of cuticle thickness possibly as a defense mechanism or as an adaptation to its particular habitat.
Interestingly, the T. chinensis x T. caroliniana hybrid, although not significantly different from T. diversifolia, T.sieboldii, and T. caroliniana, has the thickest cuticle when looking at the HWA stylet insertion point only (Fig. 6). Tsuga diversifolia and T. sieboldii are both considered moderately resistant to HWA, and T. caroliniana is considered more resistant to HWA than T. canadensis (Table 1). This resistance has all been exhibited by a lower infestation rate in these species comparable to T. canadensis. While the effects of cuticle thickness on HWA feeding are unknown, we theorize that this may be involved in the resistance exhibited by the pure species and the hybrid, which also has low levels of HWA settling when artificially infested (Montgomery et al. 2009). It is possible that thick cuticle may deter high levels of infestation, enabling the tree to better cope with feeding. Further investigations are necessary to substantiate this hypothesis. If cuticle thickness remains a significant factor in resistance, it could be useful as a selective trait or screening tool for resistance breeding programs.
We expected our investigations into the micromorphology of epicuticular wax to be revealing for the interactions of HWA with its host. Surface waxes can be diverse and
152
systematically significant (Raven et al. 1986; Romberger et al. 1993; Barthlott et al. 1998) and they are an essential element for interaction between the plant and its environment
(including herbivorous insects). We initiated these studies to determine associations between structure of the epicuticular waxes and resistance to HWA. Across the hemlock samples, tubules were the most common morphological characteristic observed and were seen in all samples. Tubules are one of the most common structures found in all plant species (Barthlott et al. 1998).
Characteristics of the crystalline structure of hemlock epicuticular waxes exhibit a trend that calls for more exploration. Although highly variable, samples of T. canadensis had higher occurrence of smooth wax than the other species observed (Fig. 8B). While tubules were still present, they occurred infrequently or in the troughs of cuticular ridges. A smooth surface could have implications for plant-insect interactions, possibly to aid adhesion of the stylet sheath to the surface of the plant. As the stylets of HWA extend toward the insertion point on the pulvinus, the stylet sheath, a product of the salivary glands, is exuded in a bead- like manner, solidifies upon extrusion, and encloses the stylet bundle. The sheath is produced both outside and within the plant (Fig. 1A). When outside of host tissue, it can adhere to the plant surface, as exhibited in HWA and other Hemiptera. Many sheath-feeding hemipterans use sheath material to create an external flange adhering to the surface of the plant (Miles 1999), thought to assist labial tip stabilization and act as a fulcrum for stylet maneuvering (Pollard 1973; Cohen 1990). The flange in most hemipterans is deposited directly between the body of the insect and its insertion site into the host plant (Miles 1999;
Leopold et al. 2003). Therefore, it spans just a short distance. However, the stylet bundle of
153
HWA extends outside of the plant for ~200 µm, a much greater distance. Its interaction with the surface is perhaps more critical to feeding success. There have been no studies that indicate an influence of surface micromorphology on sheath material adhesion abilities, but perhaps a smooth surface, as commonly seen in T. canadensis, facilitates sheath adhesion and thus influences stylet penetration by HWA. The sheath may stick reasonably well to the plant surface, but some aphids are able to remove it themselves following exploratory probes
(Sylvester 1962; Nault and Gyrisco 1966). Its adhesion properties are limited (Miles 1999), and a smooth surface could influence these abilities in HWA feeding.
There are many factors, both internal and external, that affect wax production and quality and suggest that these studies should be expanded. Of all these factors, the age of the leaf is perhaps the most important; mature leaves lose their ability to produce wax
(Romberger et al. 1993). Old leaves can generally be expected to have thinner layers of wax, increased transpiration rates, large amounts of minerals lost by leaching, and low resistance to pathogens and mechanical damage by herbivorous insects (Romberger et al. 1993;
Kozlowski and Pallardy 1997). Leaf age is compounded by extended exposure to detrimental environmental conditions. Dust particles act as abrasives against this sensitive layer over time, compounding the effect of needle age on wax quality (Romberger et al.
1993) and light intensity, humidity, water availability, and air pollutants, are all contributors to the quantity and quality of wax produced (Weete et al. 1978). Generally, environmental conditions may have little impact on chemical composition, but surface wax structures, such as size, configuration, and distribution can be modified considerably (Baker 1974). During these studies, an effort was made to sample the most recently produced woody tissue of
154
mature trees. Since we could not control for tree age given our wide range of samples sources, we opted to select for tissue and tree maturity. When we compared the most recently produced woody tissue to new growth of T. heterophylla, the new growth contained a considerably higher amount of tubule crystalloid wax. However, it is well-understood that
HWA readily insert their stylets into new growth despite these increased wax structures
(Smith-Fiola et al. 2004), indicating that an abundance of tubule crystalloids do not confer resistance to HWA feeding. In addition, environmental damage to the cuticular waxes near the insertion point of HWA stylets may be negligible, as the area targeted for stylet insertion is in a crevice between the pulvinus and the stem and is protected. Epicuticular waxes are also influenced by the level of nutrition. In Douglas-fir that are fertilized with N and K, larger proportions of tubular wax to scale-like wax are formed (Chiu et al. 1992). Similarly, when Scotch pine has unbalanced mineral nutrition, both coverage and structure of wax deteriorates within a year. Deficiencies in Ca, Mg, and K and an excess in N decrease wax coverage over epistomatal chambers and alter wax structures from tube-like structures to be fused and net-like (Ylimartimo et al. 1994). Clearly, tree nutrient supplementation is a factor in wax characteristics. Fertilization histories of our samples were not determined, but all samples used came from an arboretum setting and were therefore presumably well-managed.
However, the proximity to urban settings may have exposed them to high levels of air pollutants. Finally, there is probably a substantial amount of intraspecific variation in wax structure (Wirthensohn et al. 1999), indicating that observing numerous samples would be the best. Because only a single tree was used for these comparative studies, these results may not be representative of a species. We would have preferred standard tree ages, standard
155
environmental conditions, and increased repetition for these studies (25-30 trees), but were limited by accessibility to material and the time-consuming nature of the methods. In favor of observing as many species as we were able to, we opted to keep the study feasible. The process of preparing samples followed by observing and image capturing was a time- consuming expenditure. At any rate, we may expect the trend we observed of a thinner cuticle at the stylet insertion point than on the side and outside of the pulvinus to stay consistent with increased sample sizes, as these comparisons were made from the same samples and are not attempting to serve as representatives across a species.
We observed unique characteristics of the epicuticular wax morphology in a few samples that could be artifacts. Specifically, the bumpy texture of T. chinensis (Fig. 8E) appears similar in morphology to granular ice deposits described by Jeffree and Sandford
(1982). Similarly, the amorphous structures of T. diversifolia are best described as granules
(Fig. 8F); however Barthlott et al. (1998) indicate that often contaminants on the surface of the plant may be misinterpreted as granules. In most observations, the consistent and unique structures, such as tubules, correlate with distinct epicuticular wax classification and terminology and are thus treated as characteristics of the plant surface.
It is widely accepted that wax morphology depends mainly on chemical composition
(Jeffree et al. 1976; Baker 1982; Jeffree 1986; Barthlott et al. 1998; Jetter et al. 2006;
Dragota and Riederer 2007). Tubule crystalloid waxes, as observed here within Tsuga, are dominant among gymnosperms (e.g. Ginkgo, Taxus, Picea, Pinus, Abies) and contain considerable amounts of nonacosane-10-ol (Jeffree 2006). While our studies were strictly focused on micromorphology of the hemlock surface, it may have implications for further
156
work. Primarily, unique wax crystalloids were observed in the T. chinensis x T. caroliniana hybrid. If this morphology is linked to a chemical composition unique to the hybrid, it may contribute to its resistance to HWA. Epicuticular wax composition is responsible for resistance to pest insects in some plant species, such as white spruce, [Picea glauca
(Moench) Voss] resistance to spruce budworm, [Choristoneura fumiferana (Clemens)]
(Daoust et al. 2010). Therefore, the chemical composition of epicuticular waxes across hemlock species should be closely scrutinized to determine if chemistry is associated with interspecific differences in host-plant resistance. Kaur (2009) acknowledged intraspecific variation in epicuticular wax chemistry of T. caroliniana linked to resistant provenances of the species; therefore, we may expect to see not only interspecific chemical variation in waxes, but intraspecific variation in all of the species upon thorough analysis.
ACKNOWLEDGMENTS
We thank Susan Bentz and Richard Olson (U.S. National Arboretum), Glenn Kohler
(Washington Dept. of Natural Resources), and David Marin and Kunso Kim (Morton
Arboretum) for hemlock samples. This research was funded by the United States
Department of Agriculture Forest Service, Northern Research Station and Forest Health
Protection in Region 8.
157
LITERATURE CITED
Alfaro-Tapia, A., Verdugo, J.A., Astudillo, L.A., and Ramírez, C.C. 2007. Effect of
epicuticular waxes of poplar hybrids on the aphid Chaitophorus leucomelas
(Hemiptera: Aphididae). J. Appl. Entomol. 131: 486-492.
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac.
Entomol. 1: 79-82.
Baker, E.A. 1974. The influence of environment on leaf wax deposition in Brassica oleracea
var. gemmifera. New Phytol. 73: 955-966.
Baker, E.A. 1982. Chemistry and morphology of plant epicuticular waxes. In The plant
cuticle. Edited by D.F. Cutler, K.L. Alvin, and C.E. Price. Academic Press, London,
U.K. pp. 139-166.
Barthlott, W., Neinhuis, C., Cutler, D., Ditsch, F., Meusel, I., Theisen, I., and Wilhelmi, H.
1998. Classification and terminology of plant epicuticular waxes. Bot. J. Linn. Soc.
126: 237-260.
Bentz, S.E., Riedel, L.G.H., Pooler, M.R., and Townsend, A.M. 2002. Hybridization and
self-compatibility in controlled pollinations of eastern North American and Asian
hemlock (Tsuga) species. J. Arboric. 28: 200-205.
Blum, A. 1968. Anatomical phenomena in seedlings of sorghum varieties resistant to the
sorghum shoot fly (Atherigona varia soccata). Crop. Sci. 8: 388-391.
Bodnaryk, R.P., and Lamb, R.J. 1991. Mechanisms of resistance to the flea beetle,
Phyllotreta cruciferae (Goeze), in mustard seedlings, Sinapis alba L. Can. J. Plant
Sci. 71: 13-20.
158
Carter, C.I. 1982. Susceptibility of Tilia species to the aphid Eucallipterus tilae. In
Proceedings of the 5th International Symposium on Insect-Plant Relationships,
Wageningen, The Netherlands, 1-4 March 1982. Edited by J.H Visser, and A.K.
Minks. Pudoc, Wageningen, The Netherlands. pp. 421-423.
Caswell, T., Casagrande, R., Maynard, B., and Preisser, E. 2008. Production and evaluation
of eastern hemlocks potentially resistant to the hemlock woolly adelgid. In Fourth
Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14
February 2008. Edited by B. Onken, and R. Rheardon. USDA Forest Service,
FHTET-2008-01, Morgantown, W.V.
Chiu, S.-T., Anton, L.H., Ewers, F.W., Hammerschmidt, R., and Pregitzer, K.S. 1992.
Effects of fertilization on epicuticular wax morphology of needle leaves of Douglas-
fir, Pseudotsuga menziesii (Pinaceae). Am. J. Bot. 79: 149-154.
Cohen, A.C. 1990. Feeding adaptations of some predaceous Hemiptera. Ann. Entomol. Soc.
Am. 83: 1215-1223.
Cohen, A.C., Cheah, C.A.S.-J., Strider, J., and Hain, F. 2008. Diet development for hemlock
woolly adelgids and their predators, pp. 151-156. In Fourth symposium on hemlock
woolly adelgid in the eastern United States, Hartford, Connecticut, 12-14 February
2008. Compiled by B. Onken, and R. Reardon. FHTET-2008-01, USDA Forest
Service, Morgantown, W.V. pp. 151-156.
Daoust, S.P., Mader, B.J., Bauce, E., Despland, E., Dussutour, A., and Albert, P.J. 2010.
Influence of epicuticular-wax composition on the feeding pattern of a phytophagous
insect: implications for host resistance. Can. Entomol. 142: 261-270.
159
Del Tredici, P., and Kitajima, A. 2004. Introduction and cultivation of Chinese hemlock
(Tsuga chinensis) and its resistance to hemlock woolly adelgid (Adelges tsugae). J.
Arboric. 30: 282-287.
Dragota, S., and Riederer, M. 2007. Epicuticular wax crystals of Wollemia nobilis:
morphology and chemical composition. Ann. Bot. 100: 225-231.
Duffey, S.S. 1986. Plant glandular trichomes: their partial role in defence against insects. In
Insects and the plant surface. Edited by B.E. Juniper, and T.R.E. Southwood. Arnold,
London, U.K. pp. 151-172.
Dussourd, D.E. 1993. Foraging with finesse: caterpillar adaptations for circumventing plant
defenses. In Caterpillars: Ecological and evolutionary constraints on foraging. Edited
by N.E. Stamp, and T.M. Casey. Chapman and Hall, New York, N.Y. pp. 92–123.
Eglinton, G., and Hamilton, R.J. 1967. Leaf epicuticular waxes. Science 156: 1322-1355.
Eveling, D.W., and McCall, R.D. 1983. An evaluation of methods for preparing easily
damaged cuticular surfaces of plants for scanning electron microscopy. J. Microsc.
129: 113-122.
Farjon, A. 1990. Pinaceae. Drawings and descriptions of the genera Abies, Cedrus,
Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea.
Koeltz Scientific Books, Konigstein, Germany. 330 p.
Forbes, A.R. 1977. The mouthparts and feeding mechanisms of aphids. In Aphids as virus
vectors. Edited by K.F. Harris, and K. Maramorosch, K. Academic Press, New York,
N.Y. pp. 83-103.
160
Gallun, R.L. 1966. Resistance of small grains to the cereal leaf beetle. J. Econ. Entomol. 59:
827-829.
Havill, N.P., Montgomery, M.E., Yu, G., Shiyake, S., and Caccone, A. 2006. Mitochondrial
DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic
speciation and pinpoints the source of the introduction to eastern North America.
Ann. Entomol. Soc. Am. 99: 195-203.
Havill, N.P., Campbell, C.S., Vining, T.F., LePage, B., Bayer, R.J., and Donoghue, M.J.
2008. Phylogeny and biogeography of Tsuga (Pinaceae) inferred from nuclear
ribosomal ITS and chloroplast DNA sequence data. Syst. Bot. 33: 478-489.
Hoffman, G.D., and McEvoy, P.B. 1985. The mechanisms of trichome resistance in
Anaphalis margaritacea to the meadow spittlebug Philaenus spumarius. Entomol.
Exp. Appl. 39: 123-129.
Ibbotson, A., and Kennedy, J.S. 1959. Interaction between walking and probing in Aphis
fabae. J. Exp. Biol. 36: 377–390.
Ingwell, L.L., and Preisser, E.L. 2010. Using citizen science programs to identify host
resistance in pest-invaded forests. Conserv. Biol. 25: 182-188.
Jeffree, C.E. 1986. The cuticle, epicuticular waxes and trichomes of plants, with reference to
their structure, functions and evolution. In Insects and the plant surface. Edited by
B.E. Juniper, and S.R. Southwood. Edward Arnold, London, U.K. pp. 23-63.
Jeffree, C.E. 2006. The fine structure of the plant cuticle. In Biology of the plant cuticle.
Edited by M. Riederer, and C. Müller. Blackwell Publishing, Oxford, U.K. pp. 79-86.
161
Jeffree, C.E., and Sandford, A.P. 1982. Crystalline structure of plant epicuticular waxes
demonstrated by cryostage scanning electron microscopy. New Phytol. 91: 549-559.
Jeffree, C.E., Baker, E.A., and Holloway, P.J. 1976. Origins of the fine structure of plant
epicuticular waxes. In Microbiology of aerial plant surfaces. Edited by C.H.
Dickinson, and T.F. Preece. Academic Press, London, U.K. pp. 119-158.
Jetter, R., Kunst, L., and Samuels, L. 2006. Composition of plant cuticular waxes. In Biology
of the plant cuticle. Edited by M. Riederer and C. Müller. Blackwell Publishing,
Oxford, U.K. pp. 145-181.
Jetton, R.M., Hain, F.P., Dvorak, W.S., and Frampton, J. 2008. Infestation rate of hemlock
woolly adelgid (Hemiptera: Adelgidae) among three North American hemlock
(Tsuga) species following artificial inoculation. J. Entomol. Sci. 43: 438-442.
Johnson, H.B. 1975. Plant pubescence: an ecological perspective. Bot. Rev. 41: 233-258.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges
tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and
epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.Sc.
thesis, Department of Entomology, North Carolina State University, Raleigh, N.C.
Klingauf, F.A. 1987. Host plant finding and acceptance. In Aphids: their biology, natural
enemies and control. Edited by A.K. Minks, and P. Harrewijn. Elsevier, Amsterdam,
The Netherlands. pp. 209-223.
Kozlowski, T.T., and S.G. Pallardy. 1997. Physiology of woody plants. 2nd ed. Academic
Press, San Diego, CA.
162
Krokene, P., Nagy, N.E., and Krekling, T. 2008. Traumatic resin ducts and polyphenolic
parenchyma cells in conifers. In Induced plant resistance to herbivory. Edited by A.
Schaller. Springer, Amsterdam, The Netherlands. pp. 147-169.
Lal, M.N., Singh, S.S., and Singh, V.P.. 1999. Reaction of mustard aphid, Lipaphis erysimi
(Kalt.) to morphological characters of mustard, Brassica juncea L. J. Entomol. Res.
23: 221-223.
Larsson, L. 2002. Resistance in trees to insects – an overview of mechanisms and
interactions. In Mechanisms and deployment of resistance in trees to insects. Edited
by M.R. Wagner, K.M. Clancy, F. Lieutier, and T.D. Paine. Kluwer Academic
Publishers, Dordrecht, The Netherlands. pp. 1-30.
Leopold, R.A., Freeman, T.P., Buckner, J.S., and Nelson, D.R. 2003. Mouthpart morphology
and stylet penetration of host plants by the glassywinged sharpshooter, Homalodisca
coagulata, (Homoptera: Cicadellidae). Arthropod Struct. Dev. 32: 189-199.
Levin, D.A. 1973. The role of trichomes in plant defense. Q. Rev. Biol. 48: 3-15.
McClure, M.S. 1992. Hemlock woolly adelgid. Am. Nurserym. 175: 82-89.
McClure, M.S., Cheah, C.A., and Tigner, T.C. 2000. Is Pseudoscymnus tsugae the solution to
the hemlock woolly adelgid problem? An early perspective. In Proceedings:
Symposium on Sustainable Management of Hemlock Ecosystems in Eastern North
America, 22–24 June 1999, Durham, NH. GTR-NE-267. Edited by K.A. McManus,
K.S. Shields, and D.R. Souto. USDA Forest Service, Northeastern Research Station,
Newtown Square, PA. pp. 89–96.
McDaniel, M.E., and Janke, G.D. 1973. Leaf pubescence in oats. Crop Sci. 13: 68-69.
163
Miles, P.W. 1999. Aphid saliva. Biol. Rev. Camb. Philos. Soc. 74: 41-85.
Montgomery, M.E., Bentz, S.E., and Olsen, R.T. 2009. Evaluation of hemlock (Tsuga)
species and hybrids for resistance to Adelges tsugae (Hemiptera: Adelgidae) using
artificial infestation. J. Econ. Entomol. 102: 1247-1254.
Musetti, L., and Neal, J.J. 1997. Resistance to the pink potato aphid, Macrosiphum
euphorbiae, in two accessions of Lycopersicon hirsutum f. glabratum. Entomol. Exp.
Appl. 84: 137–146.
Nault, L.R., and Gyrisco, G.G. 1966. Relation of the feeding process of the pea aphid to the
inoculation of the pea enation mosaic virus. Ann. Entomol. Soc. Am. 59: 1185-1197.
Oten, K.L.F., Walker-Lane, L.N., Jetton, R.M., Kaur, N., Smith, B., Frampton, J., Cohen,
A.C., and Hain, F.P. 2011. Developing hemlocks resistant to the hemlock woolly
adelgid. North American Forest Insect Work Conference, 9-12 May 2011, Portland,
OR. Available from http://www.threatenedforests.com/research/wp-
content/uploads/2011/08/NAFIWC-2011.pdf [accessed 11 August 2011].
Parsons, E., Bole, B., Hall, D.J., and Thomas, W.D.E. 1974. A comparative survey of
techniques for preparing plant surfaces for the scanning electron microscope. J.
Microsc. 101: 59-75.
Patanakamjorn, S., and Pathak, M.D. 1967. Varietal resistance of rice to the Asiatic rice
borer, Chile suppressalis (Lepidoptera: Crambidae), and its association with various
plant characters. Ann. Entomol. Soc. Am. 60: 287-292.
164
Pelletier, Y. 1990. The role of the color of the substratum on the initiation of the probing
behavior in Myzus persicae (Sulzer) and Macrosiphum euphorbiae (Thomas)
(Homoptera: Aphididae). Can. J. Zool. 68: 694-698.
Peter, A.J., and Shanower, T.G. 2001. Role of plant surface in resistance to insect herbivores.
In Insects and plant defence dynamics. Edited by T.N. Ananthakrishnan. Science
Publishers, Enfield, N.H. pp. 107-132.
Pollard, D.G. 1973. Plant penetration by feeding aphids (Hemiptera, Aphidoidea): a review.
Bull. Entomol. Res. 62: 631-714.
Quiras, C.F., Stevens, M.A., Rick, C.M., and Kok-Yokomi, M.K. 1977. Resistance in tomato
to the pink form of the potato aphid, Macrosiphum euphorbiae (Thomas): the role of
anatomy, epidermal hairs and foliage composition. J. Am. Soc. Hortic. Sci. 102: 166-
171.
Ram, P., Singh, R., and Dhaliwal, G.S. 2005. Biophysical bases of resistance in plants to
insects. In Host plant resistance to insects: Concepts and applications. Edited by G.S.
Dhaliwal, and R. Singh. Panima Publishing Corporation, New Delhi, India. pp. 42-83.
Raven, P.H., Evert, R.F., and Eichorn, S.E. 1986. Biology of plants, 4th edition. Worth
Publishers, Inc., New York, N.Y.
Richardson, H.H. 1943. The action of bean leaves against the bed bug. J. Econ. Entomol. 36:
543-545.
Romberger, J.A, Hejnowicz, Z., and Hill, J.F. 1993. Plant structure: function and
development. Springer-Verlag, Berlin, Germany.
165
Rowell-Rahier, M., and Pasteels, J.M. 1982. The significance of salicin for a Salix-feeder,
Phratora (Phyllodecta) vitellinae. In Proceedings from the 5th International
Symposium on Insect-Plant Relationships. Edited by J.H. Visser, and A.K. Minks.
Pudoc, Wageningen, pp. 73-79.
Sargent, J.A. 1983. The preparation of leaf surfaces for scanning electron microscopy: a
comparative study. J. Microsc. 129: 103-110.
SAS Institute Inc. 2010. New features in JMP® 9. SAS Institute Inc, Cary, N.C.
Schalk, J.M., Jones, A., and Dukes, P.D. 1986. Factors associated with resistance in recently
developed sweetpotato cultivars and germplasm to the banded cucumber beetle,
Diabrotica alteata LeConte. J. Agric. Entomol. 3: 329-334.
Smith-Fiola, D., Hamilton, G.C., and Lashomb, J. 2004. The hemlock woolly adelgid: life
cycle, monitoring, and pest management in New Jersey. Fact Sheet751, Rutgers
Cooperative Research and Extension, New Jersey Agricultural Experiment Station,
New Brunswick, N.J.
Southwood, T.R.E. 1973. The insect-plant relationship - an evolutionary perspective. Symp.
Roy. Entomol. Soc. 6: 3-20.
Southwood, R. 1986. Plant surfaces and insects – an overview. In Insects and the plant
surface. Edited by B. Juniper, and S.R. Southwood. Arnold, London, U.K. pp. 1–22.
Souto, D., Luther, T., and Chianese, B. 1996. Past and current status of HWA in eastern and
Carolina hemlock stands. In Proceedings of the First Hemlock Woolly Adelgid
Review. Edited by S.M. Salom, T.C. Tignor, and R.C. Reardon. USDA Forest
Service, Morgantown, W.V. pp. 9-15.
166
Sylvester, E.S. 1962. Mechanisms of plant virus transmission by aphids. In Biological
transmission of disease agents. Edited by K. Maramorosch). Academic Press, New
York, NY. pp. 11-31.
Tjallingii, W.F. 1978. Mechanoreceptors of the aphid labium. Entomol. Exp. Appl. 24: 531-
537.
Tomaszewski, D. 2004. The wax layer and its morphological variability in four European
Salix species. Flora 199: 320-326.
USDA Forest Service. 2011. Counties with established HWA populations 2010.
http://na.fs.fed.us/fhp/hwa/maps/2010.pdf.
Van, T.K., and Guan, G.K. 1959. The resistance of Oryzae ridleyi Hook to paddy stem borer
(Chilo suppressalis Walk.) attack. Malayan Agric. J. 42: 207-210.
Walker-Lane, L.N. 2009. The effect of hemlock woolly adelgid infestation on water relations
of Carolina and eastern Hemlock. M.Sc. thesis, Department of Forestry and
Environmental Resources, North Carolina State University, Raleigh, N.C.
Weete, J.D., Leek, G.L., Peterson, C.M., Currie, H.E., and Branch, W.D. 1978. Lipid and
surface wax synthesis in water stressed cotton leaves. Plant Physiol. 62: 675-677.
Wirthensohn, M.G., Collins, G., Jones, G.P., and Sedgley, M. 1999. Variability in waxiness
of Eucalyptus gunnii foliage for floriculture. Sci. Hortic. 82: 279-288.
Yang, G., Espelie, K.E., Todd, J.W., Culbreath, A.K., Pittmann, R.N., and Demski, J.W.
1993. Cuticular lipids from wild and cultivated peanuts and the relative resistance of
these peanut species to fall armyworm and thrips. J. Agric. Food Chem. 41: 814-818.
167
Ylimartimo, A., Pääkkonen, E., Holopainen, T., and Rita, H. 1994. Unbalanced nutrient
status and epicuticular wax of Scots pine needles. Can. J. For. Res. 24: 522-532.
Young, R.F., Shields, K.S., and Berlyn, G.P. 1995. Hemlock woolly adelgid (Homoptera:
Adelgidae): stylet bundle insertion and feeding sites. Ann. Entomol. Soc. Am. 88:
827-835.
168
Table 1. Hemlock (Tsuga) species used in studies: resistance documented in literature, native range, and source of samples.
Resistance to hemlock woolly Species adelgid Rangej Sample source T. canadensis Mostly susceptible, Northeast Schenck Forest variation occursb,c North Raleigh, NC America T. canadensis -- -- US National (infested) Arboretum Beltsville, MD T. caroliniana Mostly susceptible, Blue Ridge US National less susceptible than Mountains, Arboretum T. canadensisb, Southern Beltsville, MD variation occursd Appalachians T. chinensis Highly resistante,i Southeast US National China Arboretum Beltsville, MD T. sieboldii Resistante,f Southern Morton Arboretum Japan Lisle, IL T. diversifolia Resistante,f,g,i Northern US National Japan Arboretum Beltsville, MD T. heterophylla Resistante,h,i Northwest Priest Point Park North Olympia, WA America T. mertensiana Resistante,i Northwest Montair at Somerset North Hill America Olympia, WA T. chinensis x Resistante -- US National carolinianaa Arboretum Beltsville, MD abred by U.S. National Arboretum (Bentz et al. 2002), bJetton et al. 2008, cCaswell et al. 2008, dKaur 2009, eMontgomery et al. 2009, fMcClure et al. 2000, gMcClure 1992, hAnnand 1924, iDel Tredici and Kitajima 2004, jFarjon 1990
169
Figure 1. Hemlock woolly adelgid insertion site in relation to other sites measured on pulvini. A) Settled hemlock woolly adelgid, first instar, showing adaxial side of the needle and pulvinus. HWA, hemlock woolly adelgid; ins, insertion point; sty, stylet bundle. Scale, 100 µm. B) Three locations measured on hemlock pulvini for cuticle thickness. ins, insertion point; out, outside. Scale, 200 µm.
170
Figure 2. Cross-sections of hemlock stem. A) Eastern hemlock, Tsuga canadensis. Scale, 200 µm. B) Carolina hemlock, T. caroliniana. Scale, 200 µm. C) Chinese hemlock, T. chinensis. Scale, 200 µm. D) Southern Japanese hemlock, T. sieboldii. Scale, 500 µm. E) Northern Japanese hemlock, T. diversifolia. Scale, 200 µm. F) Western hemlock, T. heterophylla. Scale, 500 µm. G) Mountain hemlock, T. mertensiana. Scale, 500 µm. H) Chinese-Carolina hybrid, T. chinensis x T. caroliniana. Scale, 200 µm.
171
Figure 3. Hemlock cuticle at hemlock woolly adelgid stylet insertion point. A) Eastern hemlock, Tsuga canadensis. B) Carolina hemlock, T. caroliniana. C) Chinese hemlock, T chinensis. D) Southern Japanese hemlock, T. sieboldii. E) Northern Japanese hemlock, T. diversifolia. F) Western hemlock, T. heterophylla. G) Mountain hemlock, T. mertensiana. H) Chinese-Carolina hybrid, T. chinensis x T. caroliniana. Scale, 10 µm.
172
4.5
4 c
3.5 b b 3 a a 2.5 a 2
1.5 Cuticle Thickness (µm) Thickness Cuticle 1
0.5
0 T. canadensis T. caroliniana T. chinensis T. diversifolia T. heterophylla T. mertensiana (infested) Hemlock Species
Figure 4. Effect of hemlock species on cuticle thickness. Columns with different letters are significantly different at α = 0.05, LSMeans Tukey HSD Procedure.
173
4
3.5 b b
3
2.5
2 a
1.5
1 Cuticle Thickness (µm) Thickness Cuticle
0.5
0 Insertion point Side Outside Position
Figure 5. Effect of location on cuticle thickness. Columns with different letters are significantly different at α = 0.05, LSMeans Tukey HSD Procedure.
174
3 c
2.5 cd acd cd ad 2 abd ab ab b 1.5
1 Insertion Point Point (µm) Insertion Cuticle Thickness at at Thickness Cuticle 0.5
0
Hemlock Species
Figure 6. Effect of species on cuticle thickness at insertion point. Columns with different letters are significantly different at α = 0.05, LSMeans Tukey HSD Procedure.
175
6 Insertion point 5 Side Outside
4
3
2 Cuticle Thickness (µm) Thickness Cuticle
1
0 T. canadensis T. caroliniana T. chinensis T. diversifolia T. heterophylla T. mertensiana (infested) Species
Figure 7. Cuticle thickness across all measurement points for six species. Standard error bars shown. ANOVA performed at α = 0.05, LSMeans Tukey HSD Procedure.
176
Figure 8. Wax structures. A) Tubules (shown: T. mertensiana). Scale, 5µm. B) Smooth layer (shown: T. canadensis). Scale, 5 µm. C) Tubules in trough of cuticular ridges (T. canadensis). Scale, 2 µm. D) Scattered platelets (T. caroliniana). Scale, 2 µm. E) Bump- like texture (T. chinensis). Scale, 5 µm. F) Irregular, granule-like (T. diversifolia). Scale, 5 µm. G) Polygonal platelets (T. diversifolia). Scale, 5 µm. H) Parallel-like platelets (T. chinensis x caroliniana). Scale, 5 µm.
177
Chapter VI
Chemical Composition and Variability of Epicuticular Lipids of Eight Hemlock (Tsuga spp.) Taxa and their Relationship to Hemlock Woolly Adelgid (Adelges tsugae) Host-Plant Selection
ABSTRACT
Phytophagous insects that contact the plant surface often receive cues from surface waxes. These cues may be physical or chemical and may be involved in resistance of a plant to a pest. The hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae) is an invasive pest that threatens survival of eastern and Carolina hemlocks in the eastern U.S.
Hemlocks in eastern Asia and northwestern North America are considered resistant to HWA, but the mechanism of resistance is unknown. The objective of this study was to assess the variability in epicuticular lipids of hemlocks between susceptible and resistant hemlock on the species level. The epicuticular lipids of eight hemlock (Tsuga spp.) taxa (seven species, one hybrid) were analyzed by gas chromatography with mass spectrometry to determine qualitative differences. Results indicate a high degree of inter- and intraspecific variation in chemical profiles. A peak of interest is identified, called peak 5, that is present in many resistant species. This peak should be the focus of further research, including identification of the compound and investigating its relationship to HWA host selection behaviors.
INTRODUCTION
The plant surface is covered in lipids, or waxes, that are important to the mediation of plant-insect interactions. As the first point of contact between a herbivore and the host-plant,
178
characteristics of the plant cuticle may form the basis of acceptance or rejection by the herbivore (Eigenbrode and Espelie 1995, Huang et al. 2003, Daoust et al. 2010). Chemical components are significant in governing behavioral responses of insects and serve as attractants, deterrents, or can even be toxic to herbivores (Chapman 1977, Müller 2008).
These include a wide array of allelochemicals, most commonly alkanes, primary and secondary alcohols, and ketones (Eigenbrode 1996, Powell et al. 1999). Most plants have a unique blend of compounds, and are therefore useful in host-plant selection by many phytophagous insects (Bernays and Chapman 1994). Because of this correlation between host chemistry and acceptance by an herbivore, the chemistry of epicuticular waxes can confer host-plant resistance to certain species or varieties (Müller 2008).
Host-plant resistance to insect pests conferred by wax chemistry has been documented across a number of taxa. Alfalfa (Medicago sativa L.) selected for resistance to the spotted alfalfa aphid, Therioaphis maculata (Buckton.), have as much as 50% more wax esters than susceptible varieties (Bergman et al. 1991). White spruce (Picea glauca
(Moench) Voss) resistant to the spruce budworm (Choristoneura fumiferana (Clemens)) have significantly higher concentrations of α-pinene and myrcene in epicuticular waxes compared to its susceptible counterpart. When these waxes are removed, the spruce budworm feeds more readily on the resistant individuals, suggesting that chemical components of the wax play a critical role in host selection (Daoust et al. 2010). Wax chemistry of different rice varieties is responsible for increased resistance to the brown planthopper, Nilaparvata lugens
(Stål) (Woodhead and Padgham 1988). Similarly, applying waxes extracted from a resistant
179
rice cultivar to the surface of a susceptible variety deterred feeding of the brown planthopper,
Nilaparvata lugens (Stål) (Woodhead and Padgham 1988).
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an exotic pest that threatens the survival of eastern (Tsuga canadensis (L.)
Carrière) and Carolina hemlocks (T. caroliniana Engelm.) in the eastern U.S. The motile first instar of HWA, called the crawler, is responsible for determining a suitable feeding site where they remain sessile for the remainder of their life. Crawlers are apterous and fundamentally blind (Annand 1928), and it is unknown how this feeding site is ultimately selected. Specific cues must be involved in this process, as HWA consistently inserts in a precise location, on the adaxial side of the hemlock needle, proximal to the twig with respect to the abscission layer (Young et al. 1995). Long-range dispersal of crawlers is entirely passive and occurs via wind, phoresy on macro-vertebrates, or by human movement
(McClure 1990). Therefore, HWA crawlers do not participate in long range host-finding and their interaction with the wax-laden plant surface is likely critical to their acceptance of a host plant. It is unknown if the chemistry of epicuticular lipids of hemlocks affect host-plant selection behavior of HWA.
HWA infests and kills most hemlocks in the eastern U.S., but in its native range, eastern Asia and the Pacific Northwest, it is described as a minor pest and does not cause hemlock mortality (Bentz et al. 2002). Since its introduction near Richmond, VA, in the early 1950s (Souto et al. 1996), it has been widely accepted that T. canadensis and T. caroliniana were entirely and exclusively susceptible to HWA. However, in the wake of significant hemlock mortality, anecdotal evidence suggested individuals or stands of T.
180
canadensis and T. caroliniana survive despite close proximity to HWA-infested trees and therefore may be less susceptible. This has since been corroborated by research; T. caroliniana is likely less susceptible than T. canadensis and distinct populations of putatively resistant T. canadensis have been selected and continue to be pursued for the development of a resistant hemlock (Jetton et al. 2008, Preisser et al. 2008, Kaur 2009, Oten et al. 2011).
Relative susceptibility is typically based on degree of infestation, which indicates the extent of host acceptance and/or post-feeding fitness of HWA populations. Given the significance of epicuticular waxes in host acceptance by many herbivores, understanding the chemical components of hemlock epicuticular waxes is a significant step towards understanding resistance mechanisms and increasing our knowledge of host selection behavior of HWA.
Several studies on hemlock chemistry have taken place, but most of them have focused on volatile composition. Volatiles of T. canadensis, which increase in quantity and change in composition following HWA infestation, were identified as mostly monoterpenes
(Broeckling and Salom 2003). Lagalante and Montgomery (2003) identified 51 volatile terpenoids, with most of the interspecific variation attributed to two components. These components were correlated with levels of resistance to HWA: elevated levels of α-humulene was identified as a possible feeding deterrent and isobornyl acetate was identified as a potential attractant (Lagalante and Montgomery 2003, Iosue 2008). Hemlock volatiles are important olfactory cues for Laricobius nigrinus Fender, a predator of HWA that occurs in the Pacific Northwest and is being targeted for biological control of HWA in the eastern U.S.
(Zilahi-Balogh et al. 2003, Wallin et al. 2011). The role of volatiles in HWA host-plant selection is unknown, but morphology of the HWA antennae suggests that olfactory
181
perception is occurring to some degree (K.L.F.Oten, A.C. Cohen, and F.P. Hain (submitted); see Chapter 3). Another analysis of hemlock chemistry based on entire needles was performed at the species level. Authors hypothesized that high levels of N and K enhance palatability to the adelgid, while high levels of Ca and P seem to deter severe infestations
(Pontius et al. 2006). Intraspecific variation in the foliar chemistry has also been documented which may be linked to hemlock resistance to HWA (Ingwell et al. 2009). Internal chemistry could be linked to host acceptance, a hypothesis based on our knowledge of aphid feeding biology. As part of the host acceptance process, many aphids probe the plant, secrete a minor amount of saliva, and ingest a small portion of tissue. Chemoreceptors internal to the stylet food canal communicate the suitability of the host, subsequently resulting in acceptance or rejection of the plant by the investigating aphid (Miles 1958, 1999).
In addition to volatiles and internal plant chemistry that may be linked to host acceptance by HWA, the components of epicuticular lipids could also be involved. Only one previous study has focused on the epicuticular wax composition of hemlock. Kaur (2009) investigated the chemical profiles of T. caroliniana waxes from several different provenances. Higher levels of n-hexacosanol correlated with high levels of HWA infestation as determined by artificial infestation. N-hexacosanol affects host-plant selection of many phytophagous insects. It is a feeding stimulant to silkworm moth larvae, Bombyx mori L.
(Mori 1982), but delays feeding of the black bean aphid, Aphis fabae Scopoli, on wheat
(Powell et al. 1999). Assays have not been performed to confirm the role, if any, of n- hexacosanol on HWA behavior.
182
The purpose of this study was to identify the chemical components of epicuticular waxes of different species of hemlock with varying levels of resistance to HWA using gas chromatography with mass spectrometry (GC/MS). Our hypothesis was that a component within the waxes, possibly n-hexacosanol, stimulates feeding by HWA and will correlate with our understanding of host susceptibility at the species level. Our long-term objective is to understand the mechanism of host-plant resistance among hemlocks and apply this knowledge to selection and screening methods among native populations or within hemlock breeding programs.
METHODS AND MATERIALS
A total of 45 samples were collected representing eight hemlock taxa (Table 1).
Hemlock samples that were collected by others and shipped to our lab were cut from the tree, wrapped in moist paper towels, sealed in a Ziploc bag, and shipped overnight on ice. All samples were devoid of HWA infestations.
From each sample, approximately 3 g material was cut. The cut ends of the branchlets were held with hemostats and dipped twice for 30 s in 50 mL CH2Cl2 at room temperature in a pre-weighed clean vial. Samples were allowed to evaporate overnight in a fume hood and the vials, now full of waxes, were weighed. The weight of the empty vial was subtracted from this weight to determine the wax load. Depending on the species, this yielded 2-15 mg of epicuticular waxes. We did not compare the wax load between species because samples did not contain equal amounts of stem to leaf material; therefore, any
183
comparisons would not be made from standard material. Samples were lidded and kept at
4°C in darkness until analyzed.
We also performed analyses on the same individual trees to assess the effect of time between cutting the material and wax extraction. For this, we used three T. canadensis in the
Schenck Forest (Raleigh, NC). Time period one correlated with wax extractions taken directly from the hemlock in the field. Following this initial extraction, material was cut and transported to our lab and maintained under the same conditions as the other samples. Time period two was extracted from the material one week following removal from the tree, and time period three correlated to two weeks after removal from the tree.
Samples were prepared for GC/MS analysis by adding CH2Cl2 to the extracted wax at a ratio of 1 mg:1 mL. Three-hundred µL of the reconstituted sample was pippetted into scintillation vials with a Teflon-coated cap and evaporated under a Nitrogen (N) stream.
After complete evaporation, we added 200 µL pyridine (Sigma-Aldrich) and 200 µL N,O-
Bis(trimethylsilyl) trifluoroacetamide (BSTFA) and incubated at 70°C for 40 min in a heat block to derivitise the material. Then, 500 µL of CH2Cl2 was added for a final volume of
900 µL. The chemical composition was analyzed with a DB-5 capillary column (30m) using helium gas as the carrier. An initial temperature of 40°C was set then subsequently increased by 40°C m-1 to 200°C, raised by 3°C m-1 to 320°C, then decreased 30° m-1 back down to
40°C. The chemical profiles that were obtained were qualitatively compared with one another, with specific reference to differences between resistant and suscpetible hemlocks at the species level. Because results were not quantitative, statistical analyses were not performed.
184
RESULTS
Hemlock species for which multiple waxes were identified indicated significant intraspecific variability occurs. Peaks were numbered for identification purposes (Table 2).
General trends emerged when profiles were examined across all species and within species.
All hemlock taxa examined have chemicals represented by peaks 1 through 4 (Fig. 1). Seven of the 14 T. canadensis samples contained only these four compounds, while six of the samples possessed a fifth peak, peak 6 (Fig. 2). One T. canadensis had peaks 1-6 (Fig. 3).
All T. chinensis had peaks 1-5 (Fig. 4), but only one of the seven samples had only these five peaks. Four of the seven T. chinensis samples had moderately sized peak 7 and one sample had peak 8. Four of the five T. caroliniana also had peaks 1-5, similar to T. chinensis (Fig.
5); two of these also had peak 7 (Fig. 6). One sample of T. caroliniana had only peaks 1-4, similar to seven of the fourteen T. canadensis. One sample of T. caroliniana had many small peaks (Fig. 7). All three samples of T. sieboldii had peaks 1-5, a minor peak 6, and numerous small peaks between retention times 35-38 min retention time (Fig. 8). Tsuga diversifolia has peaks 1-5, 8, and one of the two samples also had peaks 6 and 11 (Fig. 9).
The single sample of T. heterophylla had peaks 1-5, consistent with T. chinensis (Fig. 10).
The single sample of T. mertensiana had peaks 1-6, and additional peaks near retention times
22 min, 24 min, and between 35 and 46 min (Fig. 11). The T. chinensis x T. caroliniana hybrid had peaks 1-5, similar to T. chinensis, T. heterophylla, and four of the five T. caroliniana (Fig. 12).
Our results for the changes in chemical composition of waxes at different amounts of elapsed time (i.e., zero, one, and two weeks) after removal from the tree revealed that sample
185
composition does not change for at least two weeks following removal from the living tree
(Fig. 13). The chemicals identified in the waxes extracted directly from the tree corresponded with the chemicals on weeks one and two.
DISCUSSION
Although there was variation of the hemlock waxes both inter- and intraspecifically, it is unclear if any of the chemicals detected in the epicuticular waxes play a role in host- selection behavior by HWA. Perhaps the most intriguing peak is 5, which is found in all samples of T. chinensis, four of the seven T. caroliniana, all T. sieboldii, all T. diversifolia, T. heterophylla, T. mertensiana, and the T. canadensis x T. caroliniana hybrid. Only one sample of T. canadensis, the most susceptible species, contained this unidentified compound.
We suggest that this peak be scrutinized further, as it may be involved in host-plant selection and/or resistance to HWA. Phylogenetic relationships cannot explain this presence or absence of this characteristic, as T. heterophylla and T. mertensiana both carry the compound but T. canadensis does not. Tsuga canadensis is a sister species to the clade that includes the
Asian hemlocks and T. caroliniana. Tsugae heterophylla and T. mertensiana are in a separate clade altogether (Havill et. al 2008b). Therefore, if it was a characteristic specific to the Asian clade, we would expect the peak to be absent in T. heterophylla and T. mertensiana. We plan to continue analyzing this compound, by verifying its identity and quantifying it amongst the species that contain it. Preliminary identifications results determined by the WILEY NILT library did not give consistent or accurate suggestions.
186
Following identification of the compound, it will be important to determine its role on HWA behavior, which can be accomplished using behavioral bioassays with the pure compound.
We obtained samples of the 'bulletproof stand', a group of putatively resistant T. canadensis that researchers from the University of Rhode Island have selected and continue to assess for resistance to HWA (samples A-16, A-17). In addition, another putatively resistant sample from New Jersey (samples A-14, A-15) was analyzed. These samples were selected for inclusion in this study because they can represent intraspecific variation in resistnace to HWA. Both of these samples lacked peak 5, similar with the known susceptible
T. canadensis. If peak 5 is involved in some level of host resistance, as suggested above, then these hemlocks, if truly resistant, are employing another mechanism.
Host chemistry is detected by herbivorous insects in a predetermined order: (1) volatiles offer long-range host finding, (2) insects come into contact with surface waxes, directly affecting insect-plant interactions; and, (3) components internal to the plant can further influence host acceptance following mouthpart insertion (Bernays and Chapman
1994). Passively-dispersed insects, like HWA crawlers, likely do not utilize long-range volatile cues because they lack the means to pursue them. As a logical conclusion, their encounter with surface lipids of hemlocks is likely significant to the process of host-plant selection. As this research continues, we expect that one of these chemicals, possibly peak 5, is involved in some behavioral effects.
Perception of plant surface chemicals is a widespread phenomenon within Hemiptera.
From a broad perspective, sternorrhynchans (to which HWA belongs) detect volatiles rising from the plant surface but also use gustatory cues to 'taste' external chemicals during
187
antennation (Backus 1988). This suggests that HWA is using volatile cues, as suggested by
Lagalante and Montgomery (2003), and chemical cues from the plant surface, as suggested by Kaur (2009), in host selection processes. This assumption is supported by their behavior on the plant surface. Our casual observations reveal that HWA conspicuously antennates, with its antennae often making contact with the surface on which they walk. It is not clear if they are using mechanoreceptors or gustatory receptors (or both) to gather knowledge of the host during this process. In some Hemiptera, chemicals of the plant surface are detected by secreting a salivary droplet onto the surface then drawing it back into the stylets to contact gustatory sensilla internal to the stylet food canal (Miles 1958). In the Aleyrodidae
(Hemiptera), apical labial sensilla have a mechano-chemosensory function, likely used to discriminate between hosts before probing (Walker and Gordh 1989). The composition of the plant surface has long been accepted to influence settling behavior of aphids (Hemiptera:
Aphididae) (Pickett et al. 1992). Contact chemoreception of sensilla on the labial tip occurs in the tarnished plant bug, Lygus lineolaris (Palisot de Beauvois) (Hemiptera: Miridae) (Avé et al. 1978). It seems highly likely that HWA uses the host surface in a similar fashion, but needs to be examined further.
Often, the same chemical may function as a stimulant to one species and a deterrent to another, even if they are closely related. For example, a chemical of the leaf-surface of
Malus spp. is a probing stimulant for the green apple aphid, Aphis pomi (DeGeer), and the apple grass aphid, Rhopalosiphum insertum (Walker), but a deterrent for the green peach aphid, Myzus persicae (Sulzer), and raspberry aphid, Amphorophora agathonica Hottes
(Montgomery and Arn 1974). It has been suggested that HWA is actually a group of several
188
species or subspecies (Havill et al. 2008a). More work should be done to determine if the subgroups within the HWA species complex respond similarly or differently to epicuticular chemicals.
We did not expect differences between freshly collected material and material that was kept in the lab for two weeks before extraction, and accordingly, there were no compositional differences between them (Fig. 13). This is significant because it confirms that a delay between material collection and wax extraction is not detrimental to the integrity of the wax. Several of our samples were extracted several weeks after collection, but we have confidence that it does not affect the results. It is not uncommon for there to be a delay between material collection and chemical analysis in other types of studies. For example, in a terpenoid analysis of hemlocks, Ingwell et al. (2009) processed materials within a month after collection. While we cannot speak for other analytical methods, epicuticular wax analyses are consistent for at least two weeks post-cutting.
Based on our knowledge of HWA and related insects' biology, it seems highly likely that HWA uses cues from the host surfaces in its host-plant selection processes. These studies have confirmed inter- and intraspecific differences in epicuticular wax chemistry, but they should be expanded to identify these compounds and related to HWA behavior. This is but a step toward a more complete understanding of resistance mechanisms, and the presence of a peak that may be related to resistance calls for more investigation.
189
ACKNOWLEDGMENTS
I am grateful to the many hemlock sample contributors: Susan Bentz and Richard
Olsen (National Arboretum), Laura Radville and Evan Preisser (University of Rhode Island),
Glenn Kohler (Washington State Department of Natural Resources), Kunso Kim and David
Marin (Morton Arboretum), Matt Gocke (NC Botanical Gardens), and Linda Randolph
(USDA - Forest Service, Pisgah National Forest). I also thank John Strider for his time spent analyzing samples, troubleshooting GC/MS equipment, and training me in the procedure.
LITERATURE CITED
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac. Entomol. 1: 79-82.
Avé, D., J.L. Frazier, and L.D. Hatfield. 1978. Contact chemoreception in the tarnished plant bug Lygus lineolaris. Ent. Exp. Appl. 24: 17-27.
Bentz, S.E., L.G.H. Riedel, M.R. Pooler, and A.M. Townsend. 2002. Hybridization and self-compatibility in controlled pollinations of eastern North American and Asian hemlock (Tsuga) species. J. Arboric.28: 200-205.
Berenbaum, M.R. 1995. The chemistry of defence: theory and practice. Proceedings of the National Academy of Sciences of the United States of America, 92: 2-8.
Bergman, D.K., J.W. Dillwith, A.A. Zarrabi, J.L. Caddel, and R.C. Berberet. 1991. Epicuticular lipids of alfalfa relative to its susceptibility to spotted alfalfa aphids (Homoptera: Aphididae). Environ. Entomol. 20: 781-85
Bernays, E.A., and R.F. Chapman. 1994. Host-plant selection by phytophagous insects. Chapman & Hall, New York, NY.
Broeckling, C.D., and S.M. Salom. 2003. Volatile emission of eastern hemlock, Tsuga canadensis, and the influence of hemlock woolly adelgid. Phytochemistry 62: 175-180.
Chapman, R.F. 1977. The role of the leaf surface in food selection by acridids and other insects. Colloq. Int. C.N.R.S. 265: 133-149.
190
Chapman, R.F. 2003. Contact chemoreception in feeding by phytophagous insects. Annu. Rev. Entomol. 48: 455-484.
Chapman, R.F., Bernays, E.A. and Simpson, S.J., 1981. Attraction and repulsion of the aphid, Cavariella aegopodii, by plant foods. J. Chem. Ecol. 7: 881-889.
Daost, S.P., B.J. Mader, E. Bauce, E. Despland, A. Dussutour and P.J. Albert. 2010. Influence of epicuticular-wax composition on the feeding pattern of a phytophagous insect: Implications for host resistance. Can. Entomol. 142: 261-270.
Eigenbrode, S.D. 1996. Plant surface waxes and insect behaviour, pp. 201-221. In: G. Kerstiens (ed.), Plant Cuticles. An Integrated Functional Approach. Bios Scientific Publishers, Oxford.
Eigenbrode, S.D., and K.E. Espelie. 1995. Effects of plant epicuticular lipids on insect herbivores. Annu. Rev. Entomol. 40: 171-194.
Havill, N.P., M.E. Montgomery, and R.G. Foottit. 2008a. Molecular ecology of hemlock woolly adelgid, its hosts, and its natural enemies. In B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Havill, N.P., C.S. Campbell, T.F. Vining, B. LePage, R.J. Bayer, and M.J. Donoghue. 2008b. Phylogeny and biogeography of Tsuga (Pinaceae) inferred from nuclear ribosomal ITS and chloroplast DNA sequence data. Syst. Bot. 33: 478-489.
Huang, J., H.J. McAuslane, and G.S. Nuesly. 2003. Effect of leaf surface extraction on palatibility of romaine lettuce to Diabrotica balteata. Ent. Exp. Appl. 106: 227-234.
Ingwell, L., J. Brady, M. Fitzpatrick, B. Maynard, R. Casagrande, and E. Preisser. 2009. Intraspecific variation in Tsuga canadensis foliar chemistry. Northeast. Nat. 16: 585- 594.
Iosue, F.V. 2008. The role of phytochemicals and leaf cushion morphology in host resistance to the hemlock woolly adelgid (Adelges tsugae). M.S. thesis, Villanova University, Villanova, PA.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
191
Lagalante, A.F., and M. Montgomery. 2003. Analysis of terpenoids from hemlock (Tsuga) species by solid-phase microextraction/gas chromatography/ion-trap mass spectrometry. J. Agric. Food Chem. 51: 2115-2120.
McClure, M.S., C.A. Cheah, and T.C. Tigner. 2000. Is Pseudoscymnus tsugae the solution to the hemlock woolly adelgid problem? An early perspective, pp 89–96. In McManus, K.A., K.S. Shields, and D.R. Souto (Eds.). Proceedings: Symposium on Sustainable Management of Hemlock Ecosystems in Eastern North America, 22–24 June 1999, Durham, NH. GTR- NE-267. USDA Forest Service, Northeastern Research Station, Newtown Square, PA.
Miles, P.W. 1958. Contact chemorecpetion in some Heteroptera, including chemoreception internal to the stylet food canal. J. Ins. Physiol. 2: 338-347.
Miles, P.W. 1999. Aphid saliva. Biol. Rev. Camb. Philos. Soc. 74: 41-85.
Montgomery, M.E., and H. Arn. 1974. Feeding response of Aphis pomi, Myzus persicae, and Amphorophora agathonica to phlorizin. J. Insect Physiol. 20: 413-421.
Müller, C. 2003. Resistance at the plant cuticle, pp. 107-130. In A. Schaller (ed.), Induced plant resistance to herbivory. Springer, Dordecht, Netherlands.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
Pickett, J.A., L.J. Wadhams, C.M. Woodcock, and J. Hardie. 1992. The chemical ecology of aphids. Annu. Rev. Entomol. 37: 67-90.
Pontius, J.A., R.A. Hallett, and J.C. Jenkins. 2006. Foliar chemistry linked to infestation and susceptibility to hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 35: 112-120.
Powell, G., S.P. Maniar, J.A. Pickett, and J. Hardie. 1999. Aphid responses to non-host epicuticular lipids. Ent. Exp. Appl. 91: 115-223.
Preisser, E., A. Lodge, D. Orwig, and J. Elkington. 2008. Range expansion and population dynamics of co-occurring invasive herbivores. Biological Invasions 10: 201-213.
Souto, D., T. Luther, and B. Chianese. 1996. Past and current status of HWA in eastern and Carolina hemlock stands, pp. 9-15. In S.M. Salom, T.C. Tignor, and R.C. Reardon (eds.). Proceedings of the First Hemlock Woolly Adelgid Review. USDA Forest Service, Morgantown, WV.
192
Walker, G.P., and G. Gordh. 1989. The occurrence of apical labial sensilla in the Aleyrodidae and evidence for a contact chemosensory function. Entomol. Exp. Appl. 51: 215-224.
Wallin, K.F., T.M. Latty, and D.W. Ross. 2011. Orientation behavior of the predator Laricobius nigrinus (Coleoptera: Derodontidae) to hemlock woolly adelgid and host tree odors in a multi-chambered olfactometer. Enriron. Entomol. 40: 864-872.
Woodhead, S., and D.E. Padgham. 1988. The effect of plant surface characteristics on resistance of rice to the brown planthopper, Nilaparvata lugens. Entomol. Exp. Appl. 47: 15- 22.
Young, R.F., Shields, K.S. and G.P. Berlyn. 1995. Hemlock woolly adelgid (Homoptera: Adelgidae): stylet bundle insertion and feeding sites. Ann. Entomol. Soc. Am. 88: 827-835.
Zilahi-Balogh, G.M.G., L.M. Humble, A.B. Lamb, S.M. Salom, and L.T. Kok. 2003. Seasonal abundance and synchrony between Laricobius nigrinus Fender (Coleoptera: Derodontidae) and its prey, the hemlock woolly adelgid, Adelges tsugae (Homoptera: Adelgidae) in British Columbia. Can. Entomol. 135: 103-115.
193
Table 1. Hemlock species and sources analyzed with GC/MS. Wax Date load Code Species Source Date Cut extracted per 3 g (mg) Tsuga canadensis, eastern hemlock Tsuga Hemlock Bluffsb, HB27 A-3a 9 Feb 2010 24 Feb 2010 3.6 canadensis (Cary, NC) Tsuga Hemlock Bluffs, HB28 A-4 9 Feb 2010 24 Feb 2010 3.6 canadensis (Cary, NC) Tsuga Hemlock Bluffs, HB56 A-5 9 Feb 2010 24 Feb 2010 3.8 canadensis (Cary, NC) Tsuga Hemlock Bluffs, HB57 A-6 9 Feb 2010 24 Feb 2010 4.0 canadensis (Cary, NC) Tsuga Linville Falls, PL1 A-7 8 Feb 2010 24 Feb 2010 8.5 canadensis (Linville, NC) Tsuga Linville Falls, PL2 A-8 8 Feb 2010 24 Feb 2010 4.1 canadensis (Linville, NC) Tsuga 'putatively resistant' 10 Mar A-14 11 Mar 2010 3.4 canadensis (Walpack, NJ) 2010 Tsuga 'putatively resistant' 10 Mar A-15 11 Mar 2010 3.5 canadensis (Walpack, NJ) 2010 Tsuga 'bulletproof stand' 10 Mar A-16 11 Mar 2010 5.0 canadensis (Madison, CT) 2010 Tsuga 'bulletproof stand' 10 Mar A-17 11 Mar 2010 3.9 canadensis (Madison, CT) 2010 Tsuga 10 Mar A-18 'susceptible' (Michigan) 11 Mar 2010 2.9 canadensis 2010 Tsuga 10 Mar A-19 'susceptible' (Michigan) 11 Mar 2010 2.3 canadensis 2010 Tsuga Linville River Nursery, A-20 5 Mar 2010 15 Mar 2010 4.7 canadensis plot 1 (Crossnore, NC) Tsuga Morton Arboretum, A-21 1 Feb 2010 15 Mar 2010 4.2 canadensis 288-2003 (Lisle, IL) Tsuga National Arboretum A-31 8 Apr 2010 22 Apr 2010 4.0 canadensis (Beltsville, MD) Tsuga caroliniana, Carolina hemlock Tsuga Linville Falls, PL A-9 8 Feb 2010 24 Feb 2010 11.8 caroliniana (Linville, NC) Tsuga Linville Falls, overlook A-10 8 Feb 2010 24 Feb 2010 9.6 caroliniana (Linville, NC)
194
Table 1 Continued Tsuga Morton Arboretum, A-24 1 Feb 2010 15 Mar 2010 14.2 caroliniana 239-81 (Lisle, IL) Tsuga Linville River Nursery, A-25 5 Mar 2010 15 Mar 2010 3.3 caroliniana plot 1 (Crossnore, NC) Tsuga National Arboretum A-32 8 Apr 2010 22 Apr 2010 6.0 caroliniana (Beltsville, MD) Tsuga chinensis, Chinese hemlock Tsuga Morton Arboretum A-11 1 Feb 2010 11 Mar 2010 6.9 chinensis (Lisle, IL) Tsuga National Arboretum, A-26 8 Apr 2010 22 Apr 2010 3.1 chinensis 001 (Beltsville, MD) Tsuga National Arboretum, A-27 8 Apr 2010 22 Apr 2010 5.0 chinensis 002 (Beltsville, MD) Tsuga National Arboretum, A-28 8 Apr 2010 22 Apr 2010 4.2 chinensis 003 (Beltsville, MD) Tsuga National Arboretum, A-29 8 Apr 2010 22 Apr 2010 5.1 chinensis 004 (Beltsville, MD) Tsuga 'var. oblong', Nat. A-30 8 Apr 2010 22 Apr 2010 5.0 chinensis Arbor. (Beltsville, MD) Tsuga sieboldii, southern Japanese hemlock Tsuga Raulston Arboretum A-2 9 Feb 2010 24 Feb 2010 3.7 sieboldii (Raleigh, NC) Tsuga Morton Arboretum, A-22 1 Feb 2010 15 Mar 2010 9.1 sieboldii 602-82 (Lisle, IL) Tsuga National Arboretum A-34 8 Apr 2010 22 Apr 2010 5.5 sieboldii (Beltsville, MD) Tsuga diversifolia, northern Japanese hemlock Tsuga Morton Arboretum, A-23 1 Feb 2010 15 Mar 2010 9.9 diversifolia 146-95 (Lisle, IL) Tsuga National Arboretum A-33 8 Apr 2010 22 Apr 2010 14.3 diversifolia (Beltsville, MD) Taxa with a single sample Tsuga Priest Point Park A-13 1 Feb 2010 11 Mar 2010 5.4 heterophylla (Olympia, WA) Tsuga Montair at Somerset A-12 1 Feb 2010 11 Mar 2010 7.6 mertensiana Hill (Olympia, WA) T. chinensis x National Arboretum A-45 8 Oct 2010 17 Oct 2010 10.3 T. caroliniana (Beltsville, MD)
195
Table 1 Continued B- tree ID - # ----- Time Elapse Study ----- weeks Tsuga Schenk Forest, #1 Extracted from tree 25 B-1-0 5.5 canadensis (Raleigh, NC) March 2010 Tsuga Schenk Forest, #2 Extracted from tree 25 B-2-0 4.6 canadensis (Raleigh, NC) March 2010 Tsuga Schenk Forest, #3 Extracted from tree 25 B-3-0 10.0 canadensis (Raleigh, NC) March 2010 Tsuga Schenk Forest, #1 25 Mar B-1-1 1 Apr 2010 2.9 canadensis (Raleigh, NC) 2010 Tsuga Schenk Forest, #2 25 Mar B-2-1 1 Apr 2010 3.6 canadensis (Raleigh, NC) 2010 Tsuga Schenk Forest, #3 25 Mar B-3-1 1 Apr 2010 3.2 canadensis (Raleigh, NC) 2010 Tsuga Schenk Forest, #1 25 Mar B-1-2 8 Apr 2010 3.7 canadensis (Raleigh, NC) 2010 Tsuga Schenk Forest, #2 25 Mar B-2-2 8 Apr 2010 3.8 canadensis (Raleigh, NC) 2010 Tsuga Schenk Forest, #3 25 Mar B-3-2 8 Apr 2010 2.3 canadensis (Raleigh, NC) 2010 a Code designates individual samples used to track identity of the samples throughout the study, A = samples used in comparative analysis, B = samples used in time elapse study b Hemlock Bluffs material were collected from clones of Hemlock Bluff trees at NC Botanical Gardens (Chapel Hill, NC), code is associated with mark trees at Hemlock Bluffs (Cary, NC).
196
Table 2. Peak identification numbers, associated retention times, and the summary of the species containing the peak.
Retention Time Peak ID Samples Containing Peak (min) 1 between 28-29 all 34 samples 2 between 31-32 all 34 samples 3 between 32-33 all 34 samples 4 near 33 all 34 samples all T. chinensis (6 samples), 1 of 15 T. canadensis, 4 of 5 T. caroliniana, all T. sieboldii (3 samples), all 5 between 31-32 T. diversifolia (2 samples), all T. heterophylla (1 sample), all T. mertensiana (1 sample), all T. canadensis x T. caroliniana hybrid (1 sample) 6 of 15 T. canadensis, 2 of 5 T. caroliniana, 1 of 2 6 near 15 T. diversifolia, 1 of 1 T. mertensiana 7 between 37-38 4 of 6 T. chinensis 8 near 11 1 of 6 T. chinensis, all T. diversifolia (2 samples) 9 9 1 of 15 T. canadensis 10 12 1 of 15 T. canadensis 11 near 19 1 of 2 T. diversifolia
197
1
4 2 3
Figure 1. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-4 (Sample shown: A-14).
198
1
2 4 6 3
Figure 2. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-4 and a small peak 6 (Sample shown: A-16).
199
1
2 5 3 4 6
Figure 3. Chemical profile of eastern hemlock, Tsuga canadensis, indicating peaks 1-6 (Sample shown: A-18).
200
1
2 3 5 4
Figure 4. Chemical profile of Chinese hemlock, Tsuga chinensis, indicating peaks 1-5 (Sample shown: A-27).
201
1
5 3 2 4
Figure 5. Chemical profile of Carolina hemlock, Tsuga caroliniana, indicating peaks 1-5 (Sample shown: A-24).
202
6
Figure 6. Chemical profile of Carolina hemlock, Tsuga caroliniana, indicating peak 6 (Sample shown: A-25).
203
Figure 7. Chemical profile of Carolina hemlock, Tsuga caroliniana, with multiple additional peaks (Sample shown: A-10).
204
Figure 8. Chemical profile of southern Japanese hemlock, Tsuga sieboldii, with numerous small peaks between 35 and 38 min (Sample shown: A-2).
205
8 6 11
Figure 9. Chemical profile of northern Japanese hemlock, Tsuga diversifolia, with peaks 1-5, 6, 8, and 11 (Sample shown: A-33).
206
Figure 10. Chemical profile of western hemlock, Tsuga heterophylla (Sample shown: A-13).
207
Figure 11. Chemical profile of mountain hemlock, Tsuga metensiana (Sample shown: A-12).
208
Figure 12. Chemical profile of the hemlock hybrid, Tsuga chinensis x T. caroliniana (Sample shown: A-45).
209
Figure 13. Effect of the period of time since cutting material on epicuticular wax composition (T. canadensis). A) Wax extracted directly from the tree; B) Wax extracted one week after cutting; C) Wax extracted two weeks after cutting.
210
Chapter VII
Infestation of Hemlock Woolly Adelgid among Three North American Hemlock (Tsuga) Species and their Growth Rates
ABSTRACT
This field study was established following a greenhouse investigation by Jetton et al.
(2008) suggesting lower hemlock woolly adelgid, Adelges tsugae Annand, infestation rates occurr on Carolina hemlock (Tsuga caroliniana Engelm.) than on eastern hemlock (T. canadensis (L.) Carr.). Both hemlock species have been thought to be equally and entirely susceptible to infestations of the exotic pest, but recent investigations have called this generalization into question. Three hemlock species (eastern, Carolina, and western (T. heterophylla Sargent) were planted in a block design in a field at the Linville River Nursery in Crossnore, NC. Western hemlock, which occurs in northwestern North America, is in the native range of the adelgid and considered resistant. Treatments were randomized among the blocks. Two blocks were purposefully infested with the adelgid using a novel "rain-down infestation" technique, one block was chemically treated to deter infestation, and two blocks were left untreated. Throughout all blocks within the study, only hemlocks in the artificially infested blocks were colonized by the hemlock woolly adelgid. Our results indicate that eastern hemlock is 36 times more likely to become infested than western hemlock and
Carolina hemlock is less than half as likely to become infested as western hemlock.
Infestation rates were expressed as the average percentage of the species that were infested in the two artificially infested blocks. Eastern hemlock had 93.75% infestation, Carolina hemlock had 11.15% infestation, and western hemlock had 29.09% infestation. Overall
211
adelgid infestations decreased in the subsequent years, likely due to high adelgid mortality caused by low lethal temperatures. During the growing season of each year (June-
September) tree height was assessed on a monthly basis. The results indicate that eastern and western hemlocks have the highest initial growth rates (3.88 - 4.59 in month-1, 4.46 - 6.35 in month-1, respectively) but the growth rate of western hemlock decreases after the first year
(3.14 - 4.02 in month-1 in 2009, 0.57 - 5.51 in month-1 in 2010) possibly given incompatible or non-optimal growing conditions. While Carolina hemlock incurs lower adelgid infestation rates, it also has the slowest growth rate of the three species (1.27 - 4.15 in month-1 across all three years), which may hinder its acceptability as a replacement for eastern hemlock in the landscape.
INTRODUCTION
Hemlock trees (Tsuga spp.) are long-lived, shade tolerant, and drought-susceptible trees within Pinaceae. Typical forest types for hemlock growth are cool, humid and receive high amounts of precipitation. They may regenerate under their own canopy and often germinate on nurse logs, litter, moss beds, and humus. The genus consists of nine species that occur in three distinct regions of the northern hemisphere, four of which are located in two distinct regions in North America. Both regions in North America contain two hemlock species each. The eastern region extends west to southern Ontario, northern Michigan,
Wisconsin, and eastern Minnesota, and south to the southern Appalachians, reaching the northern parts of Georgia and Alabama. Eastern hemlock, Tsuga canadensis (L.) Carrière, is ubiquitous throughout this range. The second species in the region is Carolina hemlock, T.
212
caroliniana Engelm., which is found in a limited portion of the southern Appalachians. The second region of hemlocks occurs in the western part of the country. Western hemlock, T. heterophylla Sargent, extends from northern California to central Alaska east to the northern
Rocky Mountain region from central Idaho to northern British Columbia from 600-1800 m in elevation. Mountain hemlock, T. mertensiana (Bong.), occurs primarily on moist, north- facing slopes from Alaska to the southern Sierra Nevadas, occupying coastal regions in its northern range and montane regions to the south. The third concentration of hemlocks occurs in eastern Asia. Three species occur in China and the Himalayans: Chinese hemlock
(T. chinensis (Franchet) E. Pritzel), Himalayan hemlock (T. dumosa (D Don) Eichler), and
Forrest hemlock (T. forrestii Downie). Two species are found in Japan, southern Japanese hemlock (T. sieboldii Carriére) and northern Japanese hemlock (T. diversifolia Masters)
(Farjon 1990, Eckenwalder 2009).
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an exotic insect causing extensive mortality of eastern and Carolina hemlocks throughout eastern North America. It was first reported near Richmond, Virginia, in 1951
(Stoetzel 2002). Since then, its range has extended to 18 states, from Georgia to Maine, often killing up to 90% or more of the native hemlocks in its path (Hale 2004, Townsend and
Rieske-Kinney 2006, USDA - Forest Service 2011). HWA is native to the hemlock regions in northwest North America and eastern Asia. HWA infests hemlocks in both regions of its native range, but is described as a minor pest there (Keen 1938, Furniss and Carolin 1977,
Bentz et al. 2002). The destructive population that now occurs in the eastern U.S. is known to originate from populations infesting T. sieboldii (Havill et al. 2006).
213
Following initial infestation, hemlocks decline in health, manifested by a lack of new growth, bud abortion, and needle drop (McClure 1991). Tree death can occur in as little as 4 years, but in some cases they may survive longer than 10 years (McClure 1987, 1991). Since the arrival of HWA to the eastern U.S., it has been widely accepted that eastern and Carolina hemlocks were entirely susceptible to infestations. However, in the wake of large-scale hemlock mortality, anecdotal evidence suggests surviving individuals or stands of eastern and Carolina hemlocks may be less susceptible. Intraspecific resistance of eastern and
Carolina hemlocks has since been corroborated by research and distinct individuals have been selected and continue to be pursued in breeding programs (Caswell et al. 2008, Kaur
2009, Ingwell and Preisser 2010, Oten et al. 2011). The interspecific resistance of Asian and western North American hemlocks to HWA is well documented. However, a recent investigation implies that interspecific resistance may occur between the two eastern North
American hemlocks. Greenhouse studies by Jetton et al. (2008b) demonstrated that eastern hemlock incurred significantly higher numbers of live feeding HWA than Carolina and western hemlocks.
Attempts to manage populations of HWA include a diverse collection of approaches.
Control can be achieved with systemic insecticides; however, pesticide applications are expensive, provide ephemeral protection, and are harmful to non-target organisms (Dilling et al. 2009), restricting their use to ornamental and forest trees with high recreational value
(Cowles et al. 2006). Therefore, another approach is highly desirable. Alternatives to pesticides have largely focused on biological control. The absence of effective predators in the eastern U.S. (Wallace and Hain 2000) has resulted in the emphasis being placed on a
214
classical biological control approach (Cheah et al. 2004). At least six coleopterans have been released or are being explored for release as predators of HWA (Sasaji and McClure 1997,
Cheah and McClure 1998, Cheah and McClure 2000, Cheah et al. 2004, 2005, Flowers et al.
2006, Lamb et al. 2006, Grant 2008, Salom et al. 2008). The efficacy of these releases on hemlock health has not yet been documented, but it has been suggested that targeting releases in hemlock populations with low adelgid susceptibility may be the most effective approach
(Jetton et al. 2008b). Hypothetically, trees with decreased susceptibility could withstand low
HWA populations for an extended period of time, increasing the likelihood of predator establishment.
Several resistant breeding programs have been initiated to develop a hemlock with low susceptibility to HWA. At the National Arboretum, Carolina hemlock has been bred with several Asian hemlock species, resulting in progeny with increased resistance to HWA
(Montgomery et al. 2009). Other programs are concentrating on developing a resistant hemlock that is entirely derived from the native gene pool. The process includes identifying putatively resistant hemlocks in nature and propagating them for resistance assessments. To date, several distinct individuals and stands have been identified and the search continues
(Ingwell and Preisser 2010, Oten et al. 2011). A more complete understanding of both inter- and intraspecific resistance of hemlocks to HWA will surely enhance such breeding efforts.
In addition, it is critical that an optimal approach for artificial infestation is developed to allow researchers to challenge putatively resistant hemlock. Gene conservation of hemlocks through ex situ conservaion banks and long-term seed storage will also be a significant
215
contribution in the search for native resistance. Such programs will preserve a variety of genetic material from which resistance can be screened in coming years (Jetton et al. 2008a).
The objectives of this research were to: (1) corroborate greenhouse studies by Jetton et al. (2008b) indicating Carolina and western hemlocks become less infested than eastern hemlock, (2) assess a new infestation technique that could be used in resistance challenges, and (3) compare growth rates of three hemlock species following infestation by HWA. Our hypotheses were that eastern hemlock would exhibit the highest level of HWA infestation relative to Carolina and western hemlocks and that Carolina hemlock would have the slowest growth rate relative to eastern and western hemlocks.
MATERIALS AND METHODS
Seedlings used for this study were donated by the NCSU Christmas Tree Genetics
Program and included three hemlock species (eastern, Carolina, and western), two to three years in age. Trees were planted in a field at the Linville River Nursery (Crossnore, NC) in a block design, five blocks total. Trees were fertilized with 14-14-14 Osmocote® per the recommended application rate on 16 April 2008, when the study began. Three treatments were randomly assigned to the five blocks, resulting in two blocks being artificially infested and two blocks to be naturally infested although this did not occur, and one block to be treated with an insecticide. This block was treated with Talstar®, an insecticide that prevents
HWA infestation. It was not feasible to treat individual trees within a block which would be needed for a complete randomized block design; therefore, treatements were assigned randomly to the entire block. To artificially infest blocks, PVC pipe frames were built and
216
placed around an entire block. Chicken wire was stretched over the frame and fastened to the frame with zip ties, covering the entire block. Locally-collected infested hemlock branches were clipped and fastened to the chicken wire using zip ties, effectively creating a "rain- down infestation" of HWA crawlers onto the trees. On 14 April 2009, blocks were randomly thinned to prevent crowding.
Trees in the artificial infestation blocks were assessed for HWA infestation three times throughout this study using a binary scale: HWA absent (0) or HWA present (1). This included 97 trees total across the 2 blocks, 36 Carolina hemlock, 34 eastern hemlock, and 27 western hemlock. Following the 2009 thinning, there were 58 trees total: 22 Carolina hemlock, 22 eastern hemlock, and 14 western hemlock. Assessments were done 31 July
2008, 1 June 2009, and 2 May 2011. Infestation data were analyzed for each year (i.e., 2008,
2009, 2011) using the logistic regression procedure (PROC LOGISTIC) in SAS 9.2 (SAS
Institute 2008) with western hemlock as the reference species.
Between April 2008 to April 2009, diameter and height measurements were taken monthly for each tree across all blocks. From May 2009 to June 2011, measurements were taken quarterly. Growth data were analyzed with a Robust Regression (PROC
ROBUSTREG) procedure using the Huber Method in SAS 9.2. Much of the hemlock growth occurred during a short period during summer months (June - September), and during the remainder of the year, there was no observable increase in height. Therefore, the growth rates were estimated using covariates for each of the growing seasons (i.e. 2008, 2009, 2010).
These three covariates increased linearly during the months June - September, leveling off and remaining constant for the remainder of the study. This model was necessary as opposed
217
to least squares to avoid sensitivity to apparent outliers, which actually represent seasonal growth. Diameter data was disregarded given multiple unexplainable irregularities in the data. It is unknown what caused the abnormalities, but any number of factors may be involved, including malfunction of the electronic calipers caused by moisture or transcript errors.
RESULTS
In 2008, HWA infestation had a strong species effect (p<0.0001 at α=0.05) and no block effect (p=0.9485 at α=0.05). Eastern hemlock was 36 times more likely to be infested as western hemlock and Carolina hemlock was less than half as likely to become infested as western hemlock (Fig. 1). Infestation rates were expressed as the average percentage of the species that were infested in the two artificially infested blocks. In 2008, eastern hemlock had 93.75% infestation, Carolina hemlock had 11.15% infestation, and western hemlock had
29.09% infestation. In 2009 and 2011, the overall infestation rate dropped substantially.
Infestation data was not recorded in 2010. Frequency of infestation was 42.27% in 2008,
28.81% in 2009, and 12.07% in 2011 (Fig. 2). Infestation analysis by species was not performed on years 2009 and 2011 because some trees were infested one year and not the next (and vice versa). This, coupled with the absence of infestation in other blocks despite exposure to incoming HWA, call for a more thorough investigation to assess neighborhood effects of newly infested trees with previously infested trees.
For the growth data, there was a treatment effect in 2008, but in 2009 and 2010, there only only block effects. In 2008, western hemlock had the highest monthly growth rate
218
during the growing season of June-September (4.46 - 6.35 in month-1), closely followed by eastern hemlock (3.88 - 4.59 in month-1) (Fig. 3). Carolina hemlock had the slowest overall monthly growth (1.28 - 2.35 in month-1). In 2009, eastern hemlock surpassed western hemlock with the highest average monthly growth (4.93 - 6.09 in month-1), which was followed by western hemlock (3.13 - 4.02 in month-1) and Carolina hemlock (2.07 - 4.16 in month-1) (Fig. 4). In 2010, we observed the same trend: eastern with 4.73 - 6.72 in month-1, western with 0.57 - 5.51 in month-1, and Carolina with 2.47 - 3.86 in month-1 (Fig. 5). When incorporating the infestation data with growth rate, there was no effect of infestation status on tree growth.
DISCUSSION
In this field trial, the level of HWA infestation of eastern hemlock was significantly higher than Carolina or western hemlocks, effectively corroborating the greenhouse studies by Jetton et al. (2008b). It is reasonable to assume, based on both of these studies, that
Carolina hemlock is less likely to get infested than eastern hemlock in natural conditions as well. The mechanism of this phenomenon is unknown, but could be linked to innate resistance or decreased susceptibility of Carolina hemlock to HWA. Carolina hemlock is phylogenetically more closely related to resistant Asian hemlocks than it is to its neighbor species, eastern hemlock (Havill et al. 2008); therefore, the incidence of lesser infestation rate may be a genetic characteristic that the group shares.
In the study by Jetton et al. (2008b), the authors found no significant difference between Carolina and western hemlocks. The results from this study indicate that Carolina
219
hemlock is less than half as likely to get infested as Western hemlock. Several things could explain this event. First, Carolina hemlock is a smaller tree and thus a smaller target for falling HWA. They were also part of understory-like conditions created by the larger eastern and western hemlocks. Because their canopies were not exposed to the same extent that the other species were, there could have been less occurrence of HWA contact. Finally, these results could reflect accurate variation in resistance, indicating Carolina hemlock is less likely to get infested than eastern and western hemlocks under natural conditions. Research could be expanded to determine the cause of this, but the evidence remains that of the two eastern North American hemlock species, Carolina hemlock appears to be less suitable as a host to HWA than eastern hemlock.
In the second year of the study (2009), there were significantly lower infestation rates throughout the study than were observed in the first year (2008). In 2011, there were also decreased infestation rates, with respect to 2009. We obtained daily minimum temperatures from a nearby weather station (Spruce Pine, NC, Mitchell County) from 2008 to 2011 (State
Climate Office of North Carolina, NC State University 2011). On 22 and 23 November
2008, daily minimum temperatures were -13°C; on 22 and 23 December 2008, daily minimums were -14°C; between 15 and 18 January 2009, daily minimums ranged from -14 to -18°C; from 4 to 6 February 2009, daily minimums were -16 to -17°C; and on 3 and 4
March 2009, daily minimums were -14 to -15°C. In short, during the first winter season following the initiation of this study, there were five periods (equal to 13 days) when the daily minimum temperatures were -13 to -18°C. Based on laboratory data, HWA mortality increases as temperature decreases. After only a few hours at -20°C, HWA survival is less
220
than 11% (Parker et al. 1999). Moreover, populations from the southern region of the range lose their coldhardiness earlier in the season than northern populations of HWA, with only
14% survival at -15°C, (Skinner et al. 2003). In addition, the plots were in an exposed field, with no protection from wind, ice, or snow. It is feasible that lower infestations of HWA in
2009 and 2011 are a result from mortality caused by low lethal winter temperatures.
HWA success is related to individual plant health, low winter temperatures, and environmental growing conditions in addition to innate resistance of the tree (McClure and
Cheah 2002, Pontius et al. 2006). The phenomenon of low winter temperatures being related to high HWA mortality is well-documented in the literature. Twice during the past 20 years severe winter cold has greatly reduced HWA abundance: once in Virginia (1985) and once throughout the entire eastern North American range (1994) (Souto et al. 1996). Initial laboratory studies support this observation, and an apparent lack of cold-hardiness may limit or prevent the spread into northern New England (Gouli et al. 2000, Parker et al. 1998,
1999).
In the first year of this study, 100% of eastern hemlocks were infested in one of the blocks. Given the successful rate of infestation in the first year (prior to the low temperatures), this "rain-down" infestation technique is appropriate for infesting trees on a plot-size scale. We recommend ensuring that tree canopies are equally exposed to decrease the possibility that the crown of another tree blocks infestation occurrence of a smaller tree.
While this study indicates that infestation is possible at a high rate, it should be investigated more thoroughly to determine optimal conditions or infestation rates.
221
Upon initial inspection, the growth data indicates that a large majority of tree growth occurred during early summer, when new growth was visible on the trees. This corresponds with the period HWA enters aestivation for the summer and early fall months. It is accepted that these crawlers seek insertion sites on the newly produced foliage. We analyzed just these growing months for the mean growth per month over the course of this period for our growth rate comparisons between species.
Hemlock growth rate was not affected by HWA infestation status, but it was strongly affected by species. In the first year, western hemlock had the highest growth rate, but quickly dropped in the second year and even further in the third. This could be attributed to several things. Perhaps the most probable response is that western hemlock is not native to western North Carolina, and therefore the environmental conditions affecting growth may not be optimal. In western North America, western hemlock occurs from 600 to 1800 m in elevation (Farjon 1990). Crossnore, NC, where the plots were maintained, is at approximately 1100 m; therefore, an elevational difference is likely not the issue. The annual precipitation in the native range of western hemlock is 500-3800 mm (Farjon 1990), while the average annual precipitation in Crossnore is typically 1422-1524 mm (State Climate
Office of North Carolina, NC State University 2011). We cautiously assume that rainfall was not the factor because this range fits well into the range of its native range. Climate, soil type, and many other factors may play a significant role in tree health. Western conditions are described as cool to cold and wet lowland to montane conifer forests with specific soil pH requirements (Farjon 1990, Eckenwalder 2009).
222
Carolina hemlock, on the other hand, began with a low growth rate and maintained this throughout the study. It is likely that this is the typical growth habit of the species, with a slow growth rate relative to eastern and western hemlocks. The largest known individual
Carolina hemlock was 30.2 m tall and 1.3 m in diameter. The largest eastern hemlock was
50.3 m tall and 1.6 m in diameter and western hemlocks are recorded to range from 53 to 76 m tall and 2.2 to 2.8 m in diameter (Eckenwalder 2009). Typically, Carolina hemlock is a much smaller tree with average sizes ranging from 20-25 m in height, compared with 30 to
40 and 60 to 70 m for eastern and western hemlocks, respectively (Farjon 1990).
These results reflect the absolute growth rate of the three hemlock species and therefore may not accurately represent growth rate in relation to initial tree size. To account for this, an improved method should be applied to the data, such as calculating relative growth rate or including initial height as a covariate within the model. Using relative growth rate to analyze growth rate is popular among forest researchers and considered to be ecologically significant and useful in plant comparisons (Pearcy et al. 1989). There is also precedence for comparing hemlock growth rates between species using this analysis (Evans
2008).
The collective evidence that Carolina hemlock is less susceptible to HWA infestations than eastern hemlock is mounting. This knowledge is invaluable for several reasons. Researchers can benefit from this when studying interspecific variation of hemlock susceptibility or plant characteristics or by using Carolina hemlock as a comparative model with eastern hemlock. Material from Carolina hemlock may be more attainable for many researchers in the southern Appalachian region and thus may be more appropriate to use in
223
studies than material obtained elsewhere. Growers and landscapers may select Carolina hemlock in an appropriate landscape setting as an alternative to eastern hemlock. However, given its overall slow growth habit, it may or may not be an optimal selection. This knowledge is also helpful in the pursuit of a resistant hemlock. Involving Carolina hemlock in a breeding program, either to select for highly resistant individuals or quickly growing individuals or to hybridize the species with another, will advance efforts to develop a suitable hemlock resistant to HWA.
ACKNOWLEDGMENTS
We thank John Frampton for donating the hemlock trees for use in this study, Anne
Margaret Braham for technical assistance, Jerry Moody for applying pesticides, and the many people who assisted in taking measurements in the field, particularly John Strider and Micah
Gardner.
LITERATURE CITED
Bentz, S.E., L.G.H. Riedel, M.R. Pooler, and A.M. Townsend. 2002. Hybridization and self-compatibility in controlled pollinations of eastern North American and Asian hemlock (Tsuga) species. J. Arboric. 28: 200-205.
Caswell, T., R. Casagrande, B. Maynard, and E. Preisser. 2008. Production and evaluation of eastern hemlocks potentially resistant to the hemlock woolly adelgid. In: B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Cheah, C.A.A.-J., and M.S. McClure. 1998. Life history and development of Pseudoscymnustsugae (Coleoptera: Coccinellidae), a new predator of the hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 27: 1531-1536.
224
Cheah, C.A.S.-J., and M.S. McClure. 2000. Seasonal synchrony of life cycles between the exotic predator, Pseudoscymnus tsugae (Coleoptera: Coccinellidae) and its prey, the hemlock woolly adelgid Adelges tsugae (Homoptera: Adelgidae). Agricultural and Forest Entomology 2: 241-251.
Cheah, C.A.S.-J., M.E. Montgomery, S. Salom, B.L. Parker, S. Costa, and M. Skinner. 2004. Biological control of hemlock woolly adelgid. USDA Forest Service, FHTET-2004-04. Morgantown, WV. 22 pp.
Cheah, C.A.S.-J., M.A. Mayer, D. Palmer, T. Scudder, and R. Chianese. 2005. Assessments of biological control of hemlock woolly adelgid with Sasajiscymnus tsugae in Connecticut and New Jersey, pp. 116-130. In B. Onken, and R. Rheardon (eds.), Third Symposium on Hemlock Woolly Adelgid in the Eastern United States, 1-3 February 2005, Asheville, NC. USDA Forest Service, FHTET 2005-01, Morgantown, WV.
Cowles, R.S., Montgomery, M.E., and C.A.S.-J. Cheah. 2006. Activity and residues of imidacloprid applied to soil and tree trunks to control hemlock woolly adelgid (Hemiptera: Adelgidae) in forests. J. Econ. Entomol. 99: 1258-1267.
Dilling, C., P. Lambdin, J. Grant, and R. Rhea. 2009. Community response of insects associated with eastern hemlock to imidacloprid and horticultural oil treatments. Environ. Entomol. 38: 53–66.
Eckenwalder, J.E. 2009. Conifers of the World. Timber Press, Portland, OR.
Evans, A.M. 2008. Growth and infestation by hemlock woolly adelgid of two exotic hemlock species in a New England forest. J. Sustainable For. 26: 223-240.
Farjon, A. 1990. Pinaceae. Drawings and descriptions of the genera Abies, Cedrus, Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea. Koeltz Scientific Books, Konigstein.
Flowers, R.W., S.M. Salom, and L.T. Kok. 2006. Competitive interactions among two specialist predators and a generalist predator of hemlock woolly adelgid, Adelges tsugae (Hemiptera: Adelgidae) in south-western Virginia. Agric. For. Entomol. 8: 253-262.
Furniss, R.L., and V.M. Carolin. 1977. Western Forest Insects. USDA For. Serv. Misc. Publ. 1339.
Gouli, V., B.L. Parker, and M. Skinner. 2000. Haemocytes of the hemlock woolly adelgid Adelges tsugae Annand (Hom., Adelgidae) and changes after exposure to low temperatures. J. Appl. Ent. 124: 201-206.
225
Grant, J.F. 2008. Establishing Sasajiscymnus tsugae in the South, pp. 39-44. In: B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Hale, F.A. 2004. The hemlock woolly adelgid: A threat to hemlock in Tennessee. The Univ. of Tennessee: Agric. Extension Serv. SP503-G.
Havill, N.P., Montgomery, M.E., Yu, G., Shiyake, S., and Caccone, A. 2006. Mitochondrial DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic speciation and pinpoints the source of the introduction to eastern North America. Ann. Entomol. Soc. Am. 99: 195-203.
Havill, N.P., C.S. Campbell, T.F. Vining, B. LePage, R.J. Bayer, and M.J. Donoghue. 2008. Phylogeny and biogeography of Tsuga (Pinaceae) inferred from nuclear ribosomal ITS and chloroplast DNA sequence data. Syst. Bot. 33: 478-489.
Ingwell, L.L., and E.L. Preisser. 2010. Using citizen science programs to identify host resistance in pest-invaded forests. Conserv. Biol. 25: 182-188.
Jetton, R.M., W.S. Dvorak, and W.A. Whittier. 2008a. Ecological and genetic factors that define the natural distribution of Carolina hemlock in the southeastern United States and their role in ex situ conservation. Forest Ecol. Manag. 255: 3212–3221.
Jetton, R.M., F.P. Hain, W.S. Dvorak, and J. Frampton. 2008b. Infestation rate of hemlock woolly adelgid (Hemiptera: Adelgidae) among three North American hemlock (Tsuga) species following artificial inoculation. J. Entomol. Sci. 43: 438-442.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
Keen, F.P. 1938. Insect enemies of western forests. USDA Miscellaneous Publication 273.
Lamb, A.B., S.M. Salom, L.T. Kok, and D.L. Mausel. 2006. Confined field release of Laricobius nigrinus (Coleoptera: Derodontidae), a predator of the hemlock woolly adelgid, Adelges tsugae (Hemiptera: Adelgidae), in Virginia. Can. J. For. Res. 36: 369-375.
McClure, M.S. 1987. Biology and control of hemlock woolly adelgid. Bull. Conn. Agric. Exp. Stn. 851.
226
McClure, M.S. 1991. Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis. Environ. Entomol. 20: 258-264.
McClure, M.S., and C.A.S-J. Cheah. 2002. Important mortality factors in the life cycle of hemlock woolly adelgid, Adelges tsugae Annand (Homoptera: Adelgidae) in the northeastern United States, pp. 13-22. In B. Onken, R. Reardon, and J. Lashomb (eds.), Proceedings: Hemlock woolly adelgid in the eastern United States, 12 October 2002, Charlottesville, VA. FHTET 96-10. Morgantown, WV: U.S. Department of Agriculture, Forest Service, Forest Health Technology Enterprise Team.
Montgomery, M.E., S.E. Bentz, and R.T. Olson. 2009. Evaluation of hemlock (Tsuga) species and hybrids for resistance to Adelges tsugae (Hemiptera: Adelgidae) using artificial infestation. J. Econ. Entomol. 102: 1247-1254.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
Parker, B.L., M. Skinner, S. Gouli, T. Ashikaga, and H.B. Teillon. 1998. Survival of hemlock woolly adelgid (Homoptera: Adelgidae) at low temperatures. Forest Sci. 44: 414- 420.
Parker, B.L., M. Skinner, S. Gouli, T. Ashikaga, and H.B. Teillon. 1999. Low lethal temperature for hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 28: 1085-1091.
Pearcy, R.W., J. Ehleringer, H.A. Mooney, and P.W. Rundel. 1989. Plant physiological ecology. Chapman and Hall, London, England.
Pontius, J.A., R.A. Hallett, and J.C. Jenkins. 2006. Foliar chemistry linked to infestation and susceptibility to hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 35: 112-120.
Salom, S., L. Kok, A. Lamb, C. Jubb, and B. Onken. 2008. Biological control of hemlock woolly adelgid: what is it going to take to make it work?, pp. 11-17. In B. Onken, and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008, Hartford, CT.. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
SAS Institute Inc. 2008. SAS/ETS user’s guide, version 9.2. Cary, NC: SAS Institute Inc.
227
Sasaji, H., and M.S. McClure. 1997. Description and distribution of Pseudoscymnus tsugae sp. nov. (Coleoptera: Coccinellidae), an important predator of hemlock woollyadelgid in Japan. Ann. Entomol. Soc. Am. 90: 563-568.
Skinner, M., B.L. Parker, S. Gouli, and T. Ashikaga. 2003. Regional responses of hemlock woolly adelgid (Homoptera: Adelgidae) to low temperatures. Environ. Entomol. 32: 523-528.
Souto, D., T. Luther, and B. Chianese. 1996. Past and current status of HWA in eastern and Carolina hemlock stands. pp. 9-15 In: S.M. Salom, T.C. Tigner, and R.C. Reardon (eds.), Proceedings of the The First Hemlock Woolly Adelgid Review. USDA Forest Service. FHTET 96-10. Morgantown, WV.
State Climate Office of North Carolina, NC State University. CRONOS [internet database] available at http://www.nc-climate.ncsu.edu/cronos/. Accessed October 11, 2011.
Stoetzel, M.B. 2002. History of the introduction of Adelges tsugae based on voucher specimens in the Smithsonian Institute National Collection of Insects, p. 12. In B. Onken, R. Rheadon, and J. Lashomb (eds.). Proceedings: Hemlock Woolly Adelgid in the Eastern United States Symposium, 5-7 February 2002, East Brunswick, NJ,. USDA Forest Service, Morgantown, WV.
Townsend, L., and Rieske-Kinney, L. 2006. Meeting the threat of the hemlock woolly adelgid. Univ. of Kentucky: Coop. Extension Serv. ENTFACT-452.
USDA Forest Service. 2011. Counties with established HWA populations 2010. http://na.fs.fed.us/fhp/hwa/maps/2010.pdf.
Wallace, M.S., and F.P. Hain. 2000. Field surveys and evaluation of native and established predators of the hemlock woolly adelgid (Homoptera: Adelgidae) in the southeastern United States. Environ. Entomol. 29: 638-644.
228
100 a 90 Block 1 80 Block 2 70 Average 60 50 40 30 b
Percent (%)Infestation Percent 20 c 10 0 Eastern Carolina Western Species
Figure 1. Percent infestation level of Adelges tsugae for each species per block (2008). Columns with different letters are significantly different at α = 0.05.
229
45
40
35
30
25
(%) 20
15
10
5 Percent Infestation Across Blocks Species, Across Infestation Percent 0 Year 1 (2008) Year 2 (2009) Year 3 (2011) Year
Figure 2. Percent infestation across species and block by year.
230
7
6 September September
- 5
4 Carolina
3 Eastern (in./month) Western 2
1 Mean Monthly Monthly JuneGrowth,Mean 0 Infested (A) Infested (B) Insecticide No treatment (A) No treatment (B) Treatment Block
Figure 3. Mean monthly growth by species and block for the growing period in 2008.
231
8
7
6
September September -
5
4 Carolina Eastern (in./month) 3 Western 2
1 Mean Monthly Monthly June Growth, Mean 0 Infested (A) Infested (B) Insecticide No treatment (A) No treatment (B) Treatment Block
Figure 4. Mean monthly growth by species and block for the growing period in 2009.
232
8
7
6
September September -
5
4 Carolina Eastern (in./month) 3 Western 2
1 Mean Monthly Monthly JuneGrowth,Mean 0 Infested (A) Infested (B) Insecticide No treatment (A) No treatment (B) Treatment Block
Figure 5. Mean monthly growth by species and block for the growing period in 2010.
233
Chapter VIII
Comparative Vegetative Propagation of Seven Hemlock Species (Tsuga spp.) by Hardwood Cuttings
ABSTRACT
This study compared the health of seven hemlock species following vegetative propagation by hardwood cuttings. Cuttings of seven species were obtained from several arboreta and from natural populations. Cuttings were trimmed to a length of 6-10 cm and dipped in 1:1 Dip'n Grow® to water for 5 s, stuck in a 3:2 perlite to peat medium, and maintained in a mist house. At four months after rooting in a mist house, cuttings were qualitatively assessed with a 0-3 scale where 0 = dead, 1 = conspicuous health decline, 2 = healthy, no new growth/no decline, and 3 = cutting healthy, new growth apparent. The highest health rates (i.e., percentage of cuttings with a rating of 2 or 3) were: T. canadensis
(77%), T. caroliniana (63%), T. heterophylla (3%), T. mertensiana (67%), T. chinensis
(71%), T. sieboldii (78%), and T. diversifolia (58%). Tsuga sieboldii, T. canadensis, T. chinensis, and T. caroliniana exhibited the greatest health, and T. heterophylla had the worst.
An effect for source locations of the material was also determined, but we could not conclude a reason for this. Source treatement (e.g., if they were shipped or not) and tree age did not correlate with the data. This study has demonstrated a method of vegetatively propagating hemlock species with a relatively high rate of success (60-70% on average), for all hemlock taxa except for T. heterophylla. These methods may be useful for applications to breeding programs, genetic conservation, comparative studies, investigations into resistance mechanisms, or to the nursery industry.
234
INTRODUCTION
The hemlock woolly adelgid (HWA), Adelges tsugae Annand (Hemiptera:
Adelgidae), is an invasive pest that threatens the survival of two hemlocks in eastern North
America: eastern hemlock, Tsuga canadensis (L.) Carr., and Carolina hemlock, T. caroliniana Engelm. The mortality HWA has caused in forest ecosystems is tremendous, ranging from 80-90% in most areas where the insect occurs (Hale 2004, Townsend and
Rieske-Kinney 2006). In addition, the ornamental industry, which boasts over 270 cultivars and $34 million in planting stock between North Carolina and Tennessee alone, has been virtually eliminated (Cregg 2004, DLIA 2011).
In recent years, there has been increased interest in the propagation of hemlocks for many reasons. As breeding programs become established to develop a hemlock resistant to
HWA, researchers must rely on optimal methods to clone putatively resistant individuals.
Grafting would offer similar results, but rooted cutting techniques are generally more cost effective and less time consuming (Jetton et al. 2005). Vegetative propagation techniques are also useful to the nursery industry, for applications to rearing HWA and their predators, for the genetic conservation of hemlock, and to control genetic effects for research purposes.
Hemlock breeding programs were developed in light of anecdotal evidence of intraspecific variation in resistance to HWA, which has since been corroborated by research
(Caswell et al. 2008, Kaur 2009, Ingwell and Preisser 2010, Oten et al. 2011). As extensive hemlock mortality occurs, some trees have survived. The source of this resistance is currently unknown. The Asian hemlocks [Chinese hemlock (T. chinensis (Franchet) E.
Pritzel), northern Japanese hemlock (T. diversifolia Masters), southern Japanese hemlock (T.
235
sieboldii Carrière), Himalayan hemlock (T. dumosa), and Forrest hemlock (T. forrestii)] and western North American hemlocks [western hemlock (T. heterophylla Sargent) and mountain hemlock (T. mertensiana (Bong.)] are within the native range of HWA and exhibit resistance to infestations (Annand 1924, McClure 1992, McClure et al. 2000, Del Tredici and Kitajima
2004, Montgomery et al. 2009) (Table 1). Therefore, to truly understand resistance mechanisms at work, it is helpful to compare susceptible eastern North American hemlocks compared to resistant species. Successful vegetative propagation can be useful in research to understand hemlock host resistance mechanisms.
Information regarding methods for vegetative propagation for T. canadensis, T. heterophylla, and T. caroliniana is available (Doran 1941, 1952, Flint and Jesinger 1971,
Fordham 1971, Swartley 1984, Waxman 1985, Del Tredici 1985, Hartmann et al. 1990,
Packee 1991, Wigman and Woods 2000, Jetton et al. 2005). However, there is minimal information regarding vegetative propagation of the five Asian species and mountain hemlock. The ability to do so is beneficial in comparative studies across all hemlock species.
The objective of this study was to vegetatively propagate seven hemlock species with methods adapted from Ingwell and Preisser (2010), Jetton et al. (2005), and personal communications with propagators. We compared cutting health for each species. Our hypothesis was that these methods would be suitable for the vegetative propagation of all hemlock species.
236
MATERIALS AND METHODS
Material from seven hemlock species (T. canadensis, T. caroliniana, T. heterophylla,
T. mertensiana, T. chinensis, T. sieboldii, and T. diversifolia) were collected from 1-5 sources per species (Table 1). Hardwood cuttings were taken from the lower boughs of the trees between 1 and 5 February 2010. Freshly cut ends were wrapped in moist paper towels and transported or shipped overnight inside Ziploc® bags on ice to Raleigh, NC where they were stored at 4°C until 11 and 12 February, when the rooted cutting trial began.
Cuttings were flat cut from branch tips at a length of 6-10 cm and the basal end was dipped for 5 s in a 1:1 solution of Dip'nGrow® (Dip'n Grow, Inc., Clackamas, Oregon), a commercially available rooting hormone consisting of 1.0% indole-3butyric acid (IBA) and
0.5% 1-naphthalenacetic acid (NAA). Cutting tips were allowed to air dry for 15 min, and were monitored throughout this period. If tips began to dry out, they were lightly misted with water to maintain moisture. Cuttings were then inserted into moist growing medium, 3 perlite : 2 peat (v/v) in individual 164 mL Ray Leach SC-10 Super Cells (Stuewe & Sons,
Inc., Corvallis, OR) at a depth of 2 cm.
The cuttings were maintained in a mist house at the North Carolina State University
Horticultural Field Laboratory (Raleigh, NC). The daily air temperature was maintained between 23-26°C and the nightly temperature was maintained between 20-23°C with a natural photoperiod. Soil moisture was maintained with an intermittent mist deployed by a
Grower Junior (McConkey Co., Mt. Puyallup, WA) overhead boom irrigation system. Mist delivery rate began at 53 mL·m-2 and was subsequently reduced bi-weekly after two months.
237
Cuttings were moved to the Upper Mountain Research Station (Laurel Springs, NC) in the fall of 2010.
The experiment was a block design consisting of 72 blocks with 15 cuttings randomized per block. Each cutting within the block represented a unique species/source.
Two rows of T. canadensis cuttings from Schenck Forest (Raleigh, NC) surrounded the entire experiment, serving as border rows. Cutting health was assessed on 3 June 2010 by qualitatively categorizing cuttings as: 0 (dead), 1 (conspicuous health decline), 2 (healthy, no new growth/no decline), or 3 (very healthy, new growth apparent) (Fig. 1).
Analysis of variance was performed using the Ordinal Logistic procedure of JMP 9.0
Pro (SAS Institute 2010). The main effects were determined for cutting health based on hemlock species, source of the sample, and block were tested for qualitative health ranking.
All pair-wise comparisons among least square means were analyzed by the Least Squares
Means Tukey HSD procedure.
RESULTS
Our analysis revealed a significant effect of hemlock species and sample source (see
Table 1) on the health of the cuttings after approximately four months at α = 0.05. There was no effect of block on hemlock health. The species with the best overall health ratings were T. sieboldii (2.34 ± 0.13 mean health rating), T. canadensis (2.21 ± 0.06 mean health rating), T. chinensis (2.04 ± 0.14 mean health rating), and T. caroliniana (1.94 ± 0.08 mean health rating) (Fig. 2). Tsuga mertensiana cutting health (1.28 ± 0.31 mean health rating) was not significantly different from T. canadensis, T. chinensis, and T. caroliniana, but was lower
238
than T. sieboldii. A third grouping consisted of T. caroliniana, T. mertensiana, and T. diversifolia (1.49 ± 0.14 mean health rating). The inclusion of T. mertensiana in the second and third clusters may be due to the limited materials for this species (n=16). Tsuga heterophylla had the poorest health overall (0.12 ± 0.13 mean health rating). Figure 4 displays the number of cuttings assigned per species/source combination. Survival rates were determined using the cuttings that received a 2 or 3 health rating. Survival rates were:
T. canadensis (77%), T. caroliniana (63%), T. heterophylla (3%), T. mertensiana (67%), T. chinensis (71%), T. sieboldii (78%), and T. diversifolia (58%).
A significant effect (α = 0.05) for source indicated that cuttings obtained from the
Schenck Forest (Raleigh, NC), Washington State (Olympia, WA), the Linville River Nursery
(Crossnore, NC), Morton Arboretum (Lisle, IL), and J.C. Raulston Arboretum (Raleigh, NC) had the best cutting health (Fig. 3). J.C. Raulston Arboretum and the NC Botanical Gardens
(Chapel Hill, NC) were not significantly different from one another and the lowest health ratings came from cuttings obtained from the NC Botanical Gardens and Pisgah National
Forest (Linville, NC).
DISCUSSION
My results indicate that the methods used here are not optimal for all hemlock species, specifically T. heterophylla which had extremely poor health (Fig. 2). Generally, it may be more difficult to root cuttings of T. heterophylla than the other species. Both T. canadensis and T. caroliniana belonged to the healthiest group after four months, indicating that this method could be useful in the vegetative propagation of these threatened species and
239
consequently for restoration, conservation, or breeding efforts. Tsuga sieboldii responded the best to the conditions and may be a good candidate for future studies where a resistant Asian species is desired for comparisons. It could also be useful in research applications, especially to compare plant-insect interactions on T. sieboldii, the ancestral host species of the invasive population of HWA in the eastern U.S. (Havill et al. 2006). This vegetative propagation method could be useful to develop clonal material.
The statistical significance determined between sample sources was unexpected (Fig.
3). There were no trends associated with the less successful sources. We considered the possibility that shipping may have compromised samples, but samples from Washington
State and Morton Arboretum in IL were among the healthier samples as estimated based on average cutting health (Fig. 3). We also considered variation in tree age (younger plants are generally more successful for rooted cuttings (Del Tredici 1995)), but again, the most successful group consisted of some of the youngest trees and mature trees. While we do not fully understand the cause of the differences in health between sources, it does confirm that shipping samples on ice does not compromise the material for purposes of vegetative propagation by cuttings.
The health of T. canadensis (77%) and T. caroliniana (63%) were similar to rooting success reported by Doran (1952) for both species. However, Doran (1952) reports only
10.7% survival for T. sieboldii without a hormone treatment and 50% rooting success when using IBA. Perhaps inclusion of NAA enhances rooting success, as we observed the greatest success for this species in these studies which used both rooting hormones.
240
The health of T. canadensis and T. caroliniana in this study contrasts with research by Jetton et al. (2005) who determined lower rooting success for T. caroliniana (10%) than
T. canadensis (41%) for softwood cuttings. The authors recognized that softwood cuttings may result in poorer success compared with hardwood cuttings. Because there was no significant difference between the health of T. canadensis and T. caroliniana using hardwood cuttings as there is in softwood cuttings, we infer that T. caroliniana in particular performs better when rooted as a hardwood. Jetton et al. (2005) also reported that T. caroliniana may need additional time to root, as they reported cutting survival beyond six months without definitive rooting success. Therefore, our designated health rating of 2 and 3 indicative of a
"healthy" and likely successfully rooted tree may not be correct. A health rating of 2 may not necessarily result in eventual rooting. A thorough investigation of cutting health at a later date would help confirm this hypothesis.
ACKNOWLEDGMENTS
We are grateful to Josh Steiger for facilitating the use of the mist house, to those who contributed samples and shipped them, especially Kunso Kim and David Marin (Morton
Arboretum), Glenn Kohler (Washington State Department of Natural Resources), and Matt
Gocke (NC Botanical Garden). In addition, we extend thanks to Linda Randolph (Pisgah
National Forest) for administering a permit for hemlock bough collections. We are also grateful to the many people who gave their time to stick the cuttings, especially John Strider,
Micah Gardner, Zac Arcaro, Will Blankenship, Matt Parker, and Jenna Beck.
241
LITERATURE CITED
Annand, P.N. 1924. A new species of Adelges (Hemiptera: Phylloxeridae). Pan-Pac. Entomol. 1: 79-82.
Caswell, T., R. Casagrande, B. Maynard, and E. Preisser. 2008. Production and evaluation of eastern hemlocks potentially resistant to the hemlock woolly adelgid. In: B. Onken and R. Rheardon (eds.), Fourth Symposium on the Hemlock Woolly Adelgid in the Eastern United States, 12-14 February 2008. USDA Forest Service, FHTET-2008-01, Morgantown, WV.
Cregg, B. 2004. The hemlocks. The Michigan Landscape: Conifer Corner. 3 pp.
Del Tredici, P. 1985. Propagation of Tsuga canadensis cultivars: hardwood versus softwood cuttings. Proc. Intl. Plant Prop. Soc. 35: 565-569.
Del Tredici, P., and A. Kitajima. 2004. Introduction and cultivation of Chinese hemlock (Tsuga chinensis) and its resistance to hemlock woolly adelgid (Adelges tsugae). J. Arboriculture 30: 282-287.
DLIA (Discover Life in America). 2011. < http://www.discoverlife.org/>. Accessed 14 May 2011.
Doran, W.L. 1941. Propagation of hemlock by cuttings. Am. Nurseryman 74: 18-19.
Doran, W.L. 1952. The vegetative propagation of hemlock. J. For. 50: 126-129.
Farjon, A. 1990. Pinaceae. Drawings and descriptions of the genera Abies, Cedrus, Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea. Koeltz Scientific Books, Konigstein. 330 p.
Flint, H.L., and R. Jesinger. 1971. Rooting cuttings of Canada hemlock. Plant Prop. 17: 5- 9.
Fordham, A.J. 1971. Canadian hemlock variants and their propagation. Proc. Intl. Plant Prop. Soc. 21: 470-475.
Hale, F.A. 2004. The hemlock woolly adelgid: a threat to hemlock in Tennessee. The University of Tennessee: Agricultural Extension Service. SP503-G.
Hartmann, H.T., D.E. Kester, and F.T. Davies, Jr. 1990. Plant propagation: Principles and practices. Prentice Hall, Englewood Cliffs, NJ.
242
Havill, N.P., M.E. Montgomery, G. Yu, S. Shiyake, and A. Caccone. 2006. Mitochondrial DNA from hemlock woolly adelgid (Hemiptera: Adelgidae) suggests cryptic speciation and pinpoints the source of the introduction to eastern North America. Ann. Entomol. Soc. Am. 99: 195-203.
Ingwell, L.L., and E.L. Preisser. 2010. Using citizen science programs to identify host resistance in pest-invaded forests. Conserv. Biol. 25: 182-188.
Jetton, R.M., J. Frampton, and F.P. Hain. 2005. Vegetative propagation of mature eastern and Carolina hemlocks by rooted softwood cuttings. HortScience 40: 1469-1473.
Jetton, R.M., F.P. Hain, W.S. Dvorak, and J. Frampton. 2008. Infestation rate of hemlock woolly adelgid (Hemiptera: Adelgidae) among three North American hemlock (Tsuga) species following artificial inoculation. J. Entomol. Sci. 43: 438-442.
Kaur, N. 2009. Developing artificial rearing techniques for hemlock woolly adelgid, Adelges tsugae and balsam woolly adelgid, Adelges piceae; artificial infestation and epicuticular wax study of Carolina hemlock, Tsuga caroliniana, provenances. M.S. Thesis. North Carolina State University, Raleigh.
McClure, M.S. 1992. Hemlock woolly adelgid. Am. Nurseryman 175: 82-89.
McClure, M.S., Cheah, C.A., and Tigner, T.C. 2000. Is Pseudoscymnus tsugae the solution to the hemlock woolly adelgid problem? An early perspective, pp. 89-96. In: K.A. McManus, K.S. Shields, and D.R. Souto (eds.), Proceedings: Symposium on Sustainable Management of Hemlock Ecosystems in Eastern North America, 22–24 June 1999, Durham, NH. GTR-NE-267. USDA Forest Service, Northeastern Research Station, Newtown Square, PA.
Montgomery, M.E., S.E. Bentz, and R.T. Olson. 2009. Evaluation of hemlock (Tsuga) species and hybrids for resistance to Adelges tsugae (Hemiptera: Adelgidae) using artificial infestation. J. Econ. Entomol. 102: 1247-1254.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
Packee, E.C. 1991. Tsuga heterophylla (Raf.) Sarg. western hemlock. In: R.M.Burns and B.H. Honkala (eds.). Silvics of North America. vol. 1. Conifers. USDA - Forest Service. Agr. Hdbk. 654.
SAS Institute Inc. 2010. New features in JMP® 9. SAS Institute Inc, Cary, N.C.
243
Swartley, J.C. 1984. The cultivated hemlocks. Timber Press, Portland, OR.
Townsend, L. and L. Rieske-Kinney. 2006. Meeting the threat of the hemlock woolly adelgid. University of Kentucky: Cooperative Extension Service. ENTFACT-452.
Waxman, S. 1985. Clonal differences in propagating conifers. Proc. Intl. Plant Prop. Soc. 35: 555-559.
Wigmore, B.G., and J.H. Woods. 2000. Cultural procedures for propagation of rooted cuttings of Sitka spruce, western hemlock, and douglas fir in British Columbia. Res. Br., B.C. Min. For., Victoria, B.C. Work. Pap. 46.
244
Table 1. Hemlock (Tsuga) species used in comparative vegetative propagation: resistance to the hemlock woolly adelgid documented in literature, native range, and source(s) of sample. Species Resistance to hemlock Rangei Sample source(s) woolly adelgid T. canadensis Mostly susceptible, Northeast 1) Morton Arboretum, Eastern hemlock variation occursa,b North Lisle, IL America 2) Schenck Forest, Raleigh, NC 3) Linville River Nursery, Crossnore, NC 4) Pisgah National Forest, Linville, NC 5) NC Botanical Gardens, Chapel Hill T. caroliniana Mostly susceptible, less Blue Ridge 1) Morton Arboretum, Carolina hemlock susceptible than T. Mountains, Lisle, IL canadensisa, variation Southern 2) Linville River occursc Appalachians Nursery, Crossnore, NC 3) Pisgah National Forest, Linville, NC T. heterophylla Resistantd,e,f Northwest 1) Priest Point Park, Western hemlock North Olympia, WA America 2) Linville River Nursery, Crossnore, NC T. mertensiana Resistantd,f Northwest 1) Montair at Mountain hemlock North Somerset Hill, America Olympia, WA T. chinensis Highly resistantd,f Southeast 1) Morton Arboretum, Chinese hemlock China Lisle, IL T. sieboldii Resistantd,g Southern 1) Morton Arboretum, Southern Japanese Japan Lisle, IL hemlock 2) J.C. Raulston Arboretum, Raleigh, NC T. diversifolia Resistantd,g,h,f Northern 1) Morton Arboretum, Northern Japanese Japan Lisle, IL hemlock aJetton et al. 2008, bCaswell et al. 2008, cKaur 2009, dMontgomery et al. 2009, eAnnand 1924, fMcClure et al. 2000, gMcClure 1992, hDel Tredici and Kitajima 2004, iFarjon 1990
245
Figure 1. Images of hemlock cuttings, indicating health rating where: A) 0 = dead; B) 1 = conspicuous health decline; C) 2 = healthy, no new growth/no decline; and D) 3 = cutting healthy, new growth apparent.
246
3
2.5 a ab
ab abc 2
c bc 1.5
1 Mean Health Rating Health Mean
0.5 d
0 T. sieboldii T. T. chinensis T. T. T. T. canadensis caroliniana diversifolia mertensiana heterophylla Species
Figure 2. Mean health of cuttings by species at approximately four months post-rooting. Health rating: 0 = dead; 1 = conspicuous health decline; 2 = healthy, no new growth/no decline; 3 = cutting healthy, new growth apparent. Columns with different letters are significantly different at α = 0.05, LSMeans Tukey HSD Procedure.
247
2.5 a a a
2 a ab
1.5 bc
1
c Mean Health Rating Health Mean
0.5
0 Schenck Washington Linville Morton Raulston NC Botanical Pisgah Forest, NC State River Arboretum, Arboretum, Gardens, NC National Nursery, NC IL NC Forest, NC Source of Samples
Figure 3. Mean health of cuttings by sample source at approximately four months post- rooting. Health rating: 0 = dead; 1 = conspicuous health decline; 2 = healthy, no new growth/no decline; 3 = cutting healthy, new growth apparent. Columns with different letters are significantly different at α = 0.05, LSMeans Tukey HSD Procedure.
248
Health ratings: 3 2 1 0 70
60 50 40 30 20
Cutting Quantity Cutting 10 0
Species and Source
Figure 4. Number of cuttings per species/source for each health rating at approximately four months post-rooting. Health rating: 0 = dead; 1 = conspicuous health decline; 2 = healthy, no new growth/no decline; 3 = cutting healthy, new growth apparent.
249
APPENDICES
250
Appendix A
Establishment of a Hemlock Common-Garden in Western North Carolina
ABSTRACT
The establishment of a hemlock common-garden will be a valuable tool to future research involving hemlocks (Tsuga spp.) and/or the hemlock woolly adelgid (Adelges tsugae Annand). This appendix establishes written documentation for the origins, ages, and history of the hemlock trees in a hemlock common-garden in western North Carolina and discusses its significance for researchers in the region. The garden consists of all four hemlock species that occur in North America: 28 T. canadensis, 130 T. caroliniana, 11 T. heterophylla, and 4 T. mertensiana. The primary advantage to the common-garden approach is that environmental effects are minimized, thus isolating the effects of genotype. This common-garden will remain unprotected from adelgid infestations to preserve its viability as a source for research materials.
INTRODUCTION
During the research discussed throughout this dissertation, a concern inundated several studies -- there were recurring concessions that hemlock materials used in comparative analyses were not similar in age or from similar environmental conditions with respect to one another. For example, in the case of the vegetative propagation study, our sample sources varied greatly, with hemlock cuttings obtained from the eastern, western, and midwestern parts of the United States. The health of the source tree, influenced greatly by
251
environmental conditions, likely affects its ability to reproduce vegetatively. In addition, the ages of the trees from which samples were taken ranged considerably, from less than 10 years old to full reproductive maturity. Age is considered one of the most significant factors in vegetative propagation success; it is generally accepted, particularly for conifers, that younger plants reproduce more reliably than older plants (Del Tredici 1995). Consequently, the variability of our cutting sources could be involved in effects for which we did not account. In another portion of this dissertation, we address the effect of tree age and environmental conditions to the micromorphology of epicuticular waxes of hemlock needles
(see Chapter 5). Confidence in our results could have been strengthened had our samples been obtained from trees that were comparable in age and/or from similar environments.
Although these scenarios describe personal accounts of our inability to obtain material from the same sources, the problem is well-known to many researchers who regularly face the struggle of collecting materials from comparable conditions. The establishment of a common-garden is one way to overcome the obstacle of obtaining such material.
A common-garden approach is similar to that of a botanical garden or arboretum (an arboretum is described as a botanical garden that features only trees and shrubs). The biggest difference, especially in the context of this discussion, is that botanical gardens serve the general public through education and/or recreation. In a well-known study, Ulrich (1981) quantifies the psychological effects that natural settings have on the human psyche.
Botanical gardens and arboreta offer this as an irreplaceable value to their visitors (Schroeder
1993). They also serve other functions; acting as demonstration plantings to assess the suitability of a plant species or variety in local environmental conditions (Diller 1976), or to
252
contribute to the conservation of genetic resources (Assembly of Life Sciences 1978). In addition, research is often fundamental and crucial to botanical gardens and arboreta (Steere
1969, Dosmann 2006), which may be surreptitious from the perspective of an everyday visitor. In contrast, a common-garden is established for the primary purpose of addressing the needs of research and may be planted with an experimental design in mind. It can serve as a resource of material or function as a research plot for comparative analyses. Regardless, the idea is the same -- that research is its primary purpose.
Historically, arboreta have played a valuable role in studying the hemlock and the hemlock woolly adelgid (HWA; Adelges tsugae Annand). For example, the Arnold
Arboretum (Jamaica Plain, MA) has demonstrated to be an invaluable resource in developing an evolutionary context for understanding hemlock resistance to HWA (Havill and
Montgomery 2008). The National Arboretum (Beltsville, MD) has served as the pioneering grounds for breeding a resistant hemlock hybrid (Montgomery et al. 2009). Additionally, many of our own samples were collected from arboreta nationwide, some being shipped overnight by accommodating arboretum staff. We have no doubt that an easily accessible common-garden will provide similar support for future HWA research. Moreover, it will establish a base from which material of a similar environment can be obtained. In addition, the ages of the individual trees will be known, thus allowing the researcher to pull from a variety of tree ages or from trees of equal age.
The objective of this appendix is to establish written documentation for the origins, ages, and history of the hemlock trees that were part of the initial planting of the hemlock common-garden in western North Carolina.
253
METHODS
Most of the hemlock seedlings were offered following unsuccessful grafting attempts.
Rootstock of the failed grafts, T. caroliniana and T. canadensis, were transported from the
North Carolina State University Horticultural Field Laboratory (Raleigh, NC) to a greenhouse at the Upper Mountain Research Station (Laurel Springs, NC) over several trips.
Approximate ages of the seedlings were three years. The eastern North American species were augmented with western North American species. Three-year old T. heterophylla were ordered from Burnt Ridge Nursery and Orchards (Onalaska, WA) on 28 May 2009 and two- year old T. mertensiana were ordered from Hansen's Northwest Native Plants (Salem, OR) on 22 May 2009. All trees were maintained in one-gallon pots at the Upper Mountain
Research Station greenhouse in Laurel Springs.
On 25 June 2009, the common-garden was established in a field that was pre-treated with herbicide to remove weeds and grasses (Fig. 1). Holes were manually dug with post- hole diggers at a distance of 5 ft. between each hole. Trees were assigned and planted randomly using Research Randomizer, online randomizing software (Urbaniak and Plous
2009). The initial planting consisted of 24 T. canadensis, 125 T. caroliniana, 15 T. heterophylla, and 9 T. mertensiana. Each tree was labeled with a tag indicating its species designation. A month later, trees that had suffered mortality (possibly due to transplant shock) were removed and replaced with hemlocks that remained in the greenhouse (these had been in the greenhouse in 1 gallon pots). Following the replacement plantings, there were
173 total trees within the hemlock common-garden. Species count was: 28 T. canadensis,
254
130 T. caroliniana, 11 T. heterophylla, and 4 T. mertensiana (Fig. 2). The common-garden is monitored by staff at the Upper Mountain Research Station.
DISCUSSION
A major advantage for using a common-garden approach to study plant-insect interactions is that it minimizes the environmental variation that contributes to fitness, thus more efficiently enabling quantification of the effects of genotypic variation (Nuismer and
Gandon 2008, Moloney et al. 2009). With hemlocks, we understand that there is a high degree of variation in plant resistance, both inter- and intra-specifically. There has been some deliberation as to what confers resistance to hemlocks in eastern Asia and northwestern
North America, but it is often attributed to a complex of natural enemies, the scattered distribution of the hemlocks, and innate host-plant resistance (McClure 1992, Montgomery and Lyon 1996). In the eastern U.S., there is no effective natural enemy of HWA present
(Wallace and Hain 2000). When resistance is exhibited by an individual hemlock, particularly T. canadensis or T. caroliniana, there is uncertainty whether the resistance is caused by environmental factors or tree genotype (Pontius et al. 2006, Preisser et al. 2008,
Oten et al. 2011). Ideally, HWA resistance should be assessed under standardized conditions in order to isolate the effect of innate plant resistance. Using material in a common-garden setting is ideal to interpreting the basis of resistance because several factors (especially environmental conditions) are standardized. In addition, the same genotypes can be included in studies through time.
255
Hemlocks in the vicinity of the common-garden are infested with HWA. Currently, there is no plan to chemically treat the garden to protect the trees from infestation.
Therefore, it is appropriate to assume that they will eventually become infested as well. We have opted against proactive chemical treatment because we view this garden as an ongoing experiment, with the potential for future analyses (e.g., probability of HWA infestation, probability of survival, etc.). In addition, researchers should have the opportunity to gather untreated plant material from this garden for investigations of plant-insect interactions in the field or in the lab. There is still much to understand regarding host-plant selection, species interactions, and the chemical ecology of hemlocks. If the trees are chemically treated to prevent infestation, then many of these potential studies are effectively eliminated.
Conversely, trees may become infested and decline in health or die.
We decided to incorporate T. mertensiana in this common-garden in order to include representatives of all four North American hemlock species. We are uncertain if these trees will survive for an extended period of time. Within the first month, five of the nine originally-planted T. mertensiana had died and the surviving four did not appear to be in optimal health. During our pursuit of multiple hemlock species for other research projects, we approached numerous arboreta in the northeastern U.S. about obtaining samples. None of the arboreta or botanical gardens we contacted had T. mertensiana within their landscape, and several contacts added that they had tried to grow it, but were unsuccessful. The inability to grow T. mertensiana in the eastern U.S. is attributed to a mis-match of environmental conditions. In its native range, which extends from Alaska to the southern
Sierra Nevadas, it is primarily found on moist, north-facing slopes. The common-garden is
256
at lower latitude than its documented range and the site is neither categorized as moist nor on a north-facing slope. However, literature indicates that T. mertensiana is unique in that it can tolerate sites with increased radiation and water deficiencies (Farjon 1990). Perhaps the conditions of the common-garden will be tolerable to a few of the T. mertensiana individuals.
This hemlock common-garden will unquestionably be a significant contribution to the available resources for ongoing and future research on hemlocks and/or HWA. The establishment of this garden will be particularly beneficial to researchers in North Carolina and the surrounding areas. The garden is approximately three hours from Raleigh, NC, enabling day-trip access to all four hemlock species of North America for researchers at
North Carolina State University. Moreover, because they are grown in similar environmental conditions, the sample source will be a consistent factor and ideal for research needs. The obstacle of obtaining ideal hemlock material may be overcome through accessibility to this newly established hemlock common-garden.
ACKNOWLEDGMENTS
We are grateful to John Frampton and Anne Margaret Braham for donating and facilitating access to T. canadensis and T. caroliniana. We also thank Les Miller and the many staff at the Upper Mountain Research Station for prepping the site and assisting in the planting. Finally, we are grateful to the many assistants throughout the planting process.
257
LITERATURE CITED
Assembly of Life Sciences (U.S.). 1978. Conservation of germplasm resources: an imperative. National Academy of Sciences, Washington, D.C.
Del Tredici, P. 1995. Shoots from roots: a horticultural review. Arnoldia 55: 11-19.
Diller, O.D. 1976. The role of arboreta in selecting trees for the modern landscape. In: 52nd International Society of Arboriculture Convention, St Louis, MO, 9 August 1976.
Dosmann, M.S. 2006. Research in the garden: averting the collections crisis. Bot. Rev. 72: 207-234.
Farjon, A. 1990. Pinaceae. Drawings and descriptions of the genera Abies, Cedrus, Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea. Koeltz Scientific Books, Konigstein.
Havill, N.P., and M.E. Montgomery. 2008. The role of arboreta in studying the evolution of host resistance to the hemlock woolly adelgid. Arnoldia 65: 2-9.
McClure, M.S. 1992. Hemlock woolly adelgid. Am. Nurseryman 175: 82-89.
Moloney, K.A., C. Holzapfel, K. Tielbörger, F. Jeltsch, and F.M. Schurr. 2009. Rethinking the common garden in invasion research. Perspect. Plant. Ecol. 11: 311-320.
Montgomery, M.E., and S.M. Lyon. 1996. Natural enemies of adelgids in North America: their prospect for biological control of Adelges tsugae (Homoptera: Adelgidae). In S. M. Salom, T. C. Tigner, and R. C. Reardon [eds.], Proceedings of the first hemlock woolly adelgid review; 1995 October 12; Charlottesville, VA. FHTET 96-10. Morgantown, WV: U.S. Department of Agriculture, Forest Service, Forest Health Technology Enterprise Team: 89-102.
Montgomery, M.E., S.E. Bentz, and R.T. Olson. 2009. Evaluation of hemlock (Tsuga) species and hybrids for resistance to Adelges tsugae (Hemiptera: Adelgidae) using artificial infestation. J. Econ. Entomol. 102: 1247-1254.
Nuismer, S.L., and S. Gandon. 2008. Moving beyond common-garden and transplant designs: insight into the causes of local adaptation in species interaction. Amer. Nat. 171: 658-668.
Oten, K.L.F., L.N. Walker-Lane, R.M. Jetton, N. Kaur, B. Smith, J. Frampton, A.C. Cohen, and F.P. Hain. 2011. Developing hemlocks resistant to the hemlock woolly adelgid http://www.threatenedforests.com/research/wp-content/uploads/2011/08/NAFIWC-2011.pdf.
258
Pontius, J.A., R.A. Hallett, and J.C. Jenkins. 2006. Foliar chemistry linked to infestation and susceptibility to hemlock woolly adelgid (Homoptera: Adelgidae). Environ. Entomol. 35: 112-120.
Preisser, E.L, A.G. Lodge, D.A. Orwig, and J.S. Elkinton. 2008. Range expansion and population dynamics of co-occurring invasive herbivores. Biol. Invasions 10: 201-213.
Schroeder, H.W. 1991. Preferences and meaning of arboretum landscapes: combining quantitative and qualitative data. J. Environ. Psychol. 11: 231–248.
Steere, W.C. 1969. Research as a function of a botanical garden. Longwood Program Semin. 1: 43-47.
Ulrich, R.S. 1981. Natural versus urban scenes: some psychophysiological effects. Environ. Behav. 13: 523-556.
Urbaniak, G.C., and S. Plous, S. 2009. Research Randomizer (Version 3.0). Computer software. http://www.randomizer.org/.
Wallace, M.S., and F.P. Hain. 2000. Field surveys and evaluation of native and established predators of the hemlock woolly adelgid (Homoptera: Adelgidae) in the southeastern United States. Environ. Entomol. 29: 638-644.
259
Figure 1. Hemlock common-garden at the Upper Mountain Research Station (Laurel Springs, NC), 25 June 2009.
260
CH EH CH CH CH CH CH CH CH CH CH CH CH MH CH EH CH CH CH CH CH CH WH CH EH EH EH CH CH CH EH CH CH EH CH CH CH EH CH CH CH CH EH CH EH CH CH MH CH CH CH CH CH EH EH EH CH CH CH EH MH CH WH CH CH CH CH EH CH CH CH CH CH CH EH CH CH CH CH CH CH CH CH CH CH CH CH CH EH WH CH CH CH CH WH CH EH CH EH CH CH WH CH EH CH EH CH CH WH WH CH CH CH CH CH CH CH CH CH CH WH CH CH CH CH WH CH CH EH CH CH CH CH CH EH CH CH EH CH CH EH CH CH CH CH CH CH CH CH CH MH CH CH CH CH EH CH WH CH CH CH WH CH CH EH CH CH MH CH EH CH CH CH
Figure 2. Diagram of the hemlock common-garden in Laurel Springs, NC. Left margin represents garden edge nearest to Hwy 88. CH, Carolina hemlock; EH, eastern hemlock; WH, western hemlock.
261
DISSERTATION CONCLUSIONS
The hemlock woolly adelgid (HWA), Adelges tsugae Annand, is a major pest that has and will continue to cause devastation to forest and landscape hemlocks in the eastern United
States. The research I present in this dissertation is a significant contribution toward a more complete understanding of the plant-insect interactions that are occurring between HWA and hemlocks. This is the groundwork knowledge that will contribute to continuing efforts to manage HWA, particularly by means of host-plant resistance.
Several pieces of data across the chapters fit together to support the hypothesis that extra-oral digestion is being used by HWA. The cross-section of the stylet bundle, shown in
Chapter 2, confirms the presence of a salivary canal and food canal similar to that in aphids
(which are recognized extra-oral digestion feeders). While a salivary canal is necessary for the secretions of a stylet sheath, which is understood to be used by HWA according to previous studies and my own investigations, it also facilitates secretion of watery saliva potentially filled with enzymes. Therefore, on a purely morphological basis, HWA is capable of secreting extra-oral digestive enzymes. Moreover, the width of the salivary canal is approximately 1 µm at its widest measurement. This suggests that starch granules, which are found in ray parenchyma cells and known to range from 2-50 µm amongst studied plant species, would likely need to be broken down via extra-oral digestion prior to ingestion.
Food canal width and stylet length has implications for the viscosity of the fluid that is being ingested, according to Poiseuille's Law. Digestion enzymes would help to break down large molecules incapable of being transported through the food canal, or they would decrease its viscosity to enable ingestion. In addition, our studies on HWA sensilla indicate that
262
mechanoreceptors are located at the labial apex (Chapter 3). If HWA uses some form of chemoreception to assess its host prior to host acceptance, then it possibly occurs internal to the stylet food canal, as in aphids. If this is true, then HWA could secrete saliva and ingest a small amount prior to complete stylet bundle insertion (a behavioral step that occurs in aphid host-plant selection). Finally, our investigations into trophically-related enzymes in Chapter
4 indicate that a trypsin-like enzyme, amylase-like enzyme, peroxidase, and polyphenol oxidase are all used by HWA is some fashion. While we cannot ascertain that any of these enzymes are actually being injected into the plant, our knowledge of its presence in HWA homogenate fits in with other biological details to support our hypothesis that extra-oral digestion is being used. As suggested above, starch granules would likely need to be broken down prior to ingestion, which amylase-like activity could make possible.
Injection of salivary enzymes into the hemlock has implications for host response and management. The enzymes may be responsible for the hypersensitive response and abnormal wood production that occur in susceptible hemlocks following HWA infestation.
The elicitor of such a response should be investigated further. It also means that it can be used in management, especially if resistant hemlocks are shown to use enzyme inhibitors.
Specific inhibitors, once identified, can be bred into or genetically inserted into susceptible hemlocks, resulting in a hemlock with some level of resistance to HWA.
The comparisons between resistant and susceptible hemlocks at the species level include both morphological and chemical differences. Based on low temperature-scanning electron microscopy studies (Chapter 5), the cuticular layer at the stylet bundle insertion point is significantly thinner than at other points, suggesting that HWA selects that location
263
to lessen the efforts needed to penetrate the tissue. Of the hemlocks studied, eastern, western, and mountain hemlocks exhibit the thinnest cuticle at the insertion point.
Significantly thicker cuticles are observed on Carolina, Chinese, northern Japanese, and southern Japanese hemlocks. The latter group represents species that exhibit more resistance to HWA than eastern hemlock. Infestation rate of Carolina hemlock is lower than eastern and western hemlocks, as discussed in Chapter 7. In addition, the Chinese by Carolina hybrid, which exhibits resistance, has the thickest cuticle at the insertion point. This characteristic aligns with our knowledge of resistance to HWA, with the exception that western and mountain hemlocks have cuticle thickness similar to eastern hemlock. It is possible that cuticle thickness is one of a few mechanisms used in defense against HWA, and western and mountain are employing another mechanism. To complement the morphological studies, a chemical analysis across multiple hemlock species is described in Chapter 6.
Although there is significant variation inter- and intraspecifically, one peak, which we call peak 5, occurs in almost all of the species considered more resistant to HWA than the highly susceptible eastern hemlock. This peak may appear trivial based on its quantity in some chemical profiles, but it is accepted that an insect at the plant surface is sensitive to the most minor amounts of chemical cues and therefore, HWA is likely able to detect seemingly insignificant amounts. This peak should be identified and correlated with HWA behavior. If it is involved in host-plant selection by HWA, then again, it could be selected for or bred into a resistant hemlock for use in restoration.
In order to accomplish intraspecific resistant breeding, an important step will be identifying resistant individuals from native stands. Following selection, they must be
264
assessed for resistance, which can be done by cloning the individual and challenging it with
HWA infestations. In addition, if a characteristic is selected for in an interspecific breeding program, it will be essential to screen for that characteristic, a step which would also be aided with successful vegetative propagation techniques. The study I present in Chapter 8 assesses the health of hardwood cuttings at the species level following attempts to root them. The methods I use are likely suitable for the vegetative propagation for all hemlocks, with the exception of western hemlock. Therefore, rooted cuttings are a viable option for cloning select hemlocks, including those that are putatively resistant or those of a known resistant species for which specific characters are desired. Lastly, I discuss the establishment of a common-garden at the Upper Mountain Research Station (Laurel Springs, NC) that will undoubtedly be a useful contribution to ongoing and future research on hemlocks and the hemlock woolly adelgid.
265