INFLUENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE) ______
A Dissertation presented to the Faculty of the Graduate School at the University of Missouri-Columbia ______
In Partial Fulfillment of the Requirements for the Degree
Doctor of Philosophy
______
by
TAMRA REALL LINCOLN
Dr. Richard M. Houseman, Dissertation Supervisor
MAY 2015
The undersigned, appointed by the dean of the Graduate School, have examined the dissertation entitled
INFLUENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE)
Presented by Tamra Reall Lincoln, a candidate for the degree of doctor of philosophy, and hereby certify that, in their opinion, it is worthy of acceptance.
Professor Richard M. Houseman, Dissertation Supervisor
Professor Jeanne D. Mihail
Professor Bruce A. Barrett
Research Professor Mark R. Ellersieck
…………………………. to mom
Acknowledgments
Although one name is listed as the author of this dissertation, there are many people who contributed to its completion and have my absolute gratitude.
Many thanks to my advisor, Dr. Richard M. Houseman, who provided invaluable assistance with experimental designs and with editing manuscripts. More importantly, he allowed me the freedom to try new things. Dr. Houseman, you made these projects possible and provided me with an incredible journey. I also thank my committee members, Drs. Bruce Barrett, Jeanne Mihail, and Mark
Ellersieck. I thank Dr. Barrett for thoughtful questions and teaching mentorship. I thank Dr. Mihail (and her assistant, Susan Taylor) for opening the door of the fungal kingdom to me while teaching me fungal isolation, culturing and management techniques and guiding me in research and teaching. I also thank
Dr. Ellersieck for exceptional statistical assistance and encouragement.
Thank you, Dr. Debbie Finke, for the use of your lab, equipment and mentorship. My appreciation goes to the BCIRL (USDA-ARS) and Dr. Cindy
Goodman for microscope and camera assistance and the conversations that led to my future entomological studies. Thank you, Ben Puttler, for bringing me fungi to identify or simply appreciate. Thank you, Kris Simpson, for enthusiastically assisting with each and every outreach event and systematic question. To all my professors for teaching me “the trade,” thank you! My gratitude also sincerely goes to the Division of Plant Sciences and Dr. Houseman for providing financial support of my research and education.
ii
I am indebted to all the Columbia homeowners who permitted me to make holes in their yard to take soil samples, and to the community members who brought their children to the many outreach events so we could explore science and insects together. Another community I would like to express appreciation for is the Entomological Society of America for giving me a forum to share my research, to become engaged in national and international entomological conversations, and to participate in leadership opportunities.
Finally, yet so importantly, my deep gratitude goes to my husband, children, parents, siblings, fellow graduate students, and dear friends who helped me in the field, the lab, who listened to countless hours of explanations of my projects, put up with late nights and busy semesters, read and reread this dissertation, and with patience and understanding supported my goals. I also thank my God who opened doors and presented ever brighter tomorrows— tomorrows made more beautiful with an increasing understanding of the often delicate, and always intricate, world of insects.
iii
Table of Contents
Acknowledgments…………………………………………………………………ii
List of Figures……………………..………………………………………………vi
List of Tables………………………..……………………………………………xix
Abstract…………………………………..………………………………………xxiii
Chapter One: LITERATURE REVIEW……….………………………………....1
Introduction…………………………………………………………………1
Study Organisms……………………….………………………………….3
Entomopathogen Pathology………………………………………..…....9
Research Objectives……………………………………………..…...…12
Chapter Two: COMPARING THE RELATIVE FREQUENCY OF OCCURRENCE AND FREQUENCY OF MORTALITY OF TWO ENTOMOPATHOGENIC FUNGI, BEAUVERIA AND METARHIZIUM (ASCOMYCOTA: HYPOCREALES), FROM FORESTED AND URBAN HABITATS IN MISSOURI………………….……14
Introduction………………………………………………………….……14
Methods…………………………………………………………...………18
Results……………………………………………………………….……23
Discussion……………………………………………………………...... 27
Chapter Three: VIRULENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE) ………………….…31
Introduction………………………………………………………….……31
Methods…………………………………………………………….….…37
Results………………………………………………………………....…42
Discussion……………………………………………………….….……46
iv
Chapter Four: INFLUENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON NEST SITE SELECTION BY FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE) …………49
Introduction…………………………………………………………………49
Methods…………………………………………….………………………54
Results………………………………………………………………………60
Discussion…………………………………………….……………………63
Chapter Five: FUNGISTATIC EFFECTS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE) FOUNDING PAIR EXOCRINE GLANDS ON ENTOMOPATHOGENIC FUNGAL GERMINATION……………………………………………………………………69
Introduction…………………………………………………………………69
Methods……………………………………………………….……………73
Results……………………………………………………………………...78
Discussion.…………………………………………………………………79
Literature Cited………………………………………………….…………………83
Figures…………………………………………………………………………… . 98
Tables……………………………………………………………………………...167
VITA………………………………………………………………..………………184
v
List of Figures Figure 1 Two entomopathogenic fungi, Beauveria bassiana sensu lato
(Balsamo) Vuillemin (A) and Metarhizium anisopliae s.l. (Metschnikoff) Sorokin
(B), found in soils of forested habitats and urban areas that were historically forested around Columbia, Missouri…………………………………... ……..…….98
Figure 2 Termite infestation risk in the United States. ………….…………..…….99
Figure 3 Subterranean termite colony cycle. The colony starts with a founding reproductive pair (king and queen) who build a nest chamber, lay eggs and care for the initial emerging nymphs. The nymphs develop into workers or soldiers, or sometimes supplementary reproductives. A mature colony produces winged reproductives (alates) which will leave the colony in a mass emergence, called a swarm, enabling new colonies to begin. ………………….. ………………….….100
Figure 4 Beauveria bassiana (Balsamo) Vuillemin conidial structures (A);
Conidiognous cells (B); and Conidia (C). Image: Hoog 1972 ………………..…101
Figure 5 Metarhizium anisopliae (Metschnikoff) Sorokin, Sporodochium (a),
Conidiophores (b), Conidia (c). Image: Hoog et al. 2000. …………………...….102
Figure 6 Metarhizium anisopliae (Metschnikoff) Sorokin sensu lato emerging from a Reticulitermes flavipes (Kollar) dealated founding reproductive. ………103
Figure 7 Infection pathway of entomopathogenic fungi. These fungi invade their insect hosts through the cuticle by recognizing the host surface, adhering to the host cuticle, germinating and penetrating the cuticle, entering the hemocoel, producing toxins, and ultimately causing the death of the host (A). The fungal mycelia then reemerge through the intercuticular regions to sporulate outside the cadaver (B). ………………………………………………………………………….104
vi
Figure 8 Landscapes where soil was sampled along a temporal gradient of urban development in forested and previously forested urban habitats. Three undeveloped forested habitats, three previously forested 10 yr subdivisions, and three previously forested 20 yr subdivisions were sampled to determine relative frequency of occurrence and relative frequency of mortality of entomopathogenic fungi in the soil. ………………………………………………………………..…….105
Figure 9 Guilford Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition
(B)……………………………………………………………………………….…….106
Figure 10 Monterey Hills Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition
(B)……………………………………………………………………………………..107
Figure 11 Grindstone Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition
(B)……………………………………………………………………………………...108
Figure 12 Eastland Hills. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for
vii
entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B). …………………………..109
Figure 13 Timber Ridge. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B). …………………………..110
Figure 14 Stoneridge Estates. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B).……………………..111
Figure 15 Lake Woodrail. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B).………………………….112
Figure 16 Parkade North. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B). ……………………..…..113
Figure 17 Wynfield Meadows. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B). ………….………..114
viii
Figure 18 Nine study areas in Columbia, Missouri where soil was collected to bait for entomopathogenic fungi. 1) Guilford Forest; 2) Monterey Hills Forest; 3)
Grindstone Forest; 4) Eastland Hills; 5) Timber Ridge; 6) Stoneridge Estates; 7)
Lake Woodrail; 8) Parkade North; 9) Wynfield Meadows ……………….……..115
Figure 19 Collecting the soil. ……………………………………………………..116
Figure 20 Soil fungi experimental set-up. Soil from each location (A) was homogenized and placed into two 100 ml containers (B). Ten Galleria mellonella
(Linnaeus) larvae were placed into each container (C). One container from each location was randomly selected and placed in one of two incubating temperatures (15oC or 25oC) with 24 hr darkness, 80% ± 3 RH (D)…………..117
Figure 21 Soil containing Galleria mellonella (Linnaeus) from undisturbed forest and historically forested subdivisions, incubated at two temperatures (15°C and
25°C) (A), was checked weekly for cadavers (B). Cadavers were removed, surface sterilized, and placed into individual moisture chambers (C). Cadavers were placed again into the original incubating temperature and allowed to sporulate (D). Beauveria bassiana sensu lato (E) and Metarhizium anisopliae
(Metschnikoff) Sorokin sensu lato (F) were two entomopathogenic fungi isolated from these soils. ……………………………………………………………………..118
Figure 22 Soil sample results: Guilford Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC.
The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). …………………………………………………..119
ix
Figure 23 Soil sample results: Monterey Hills Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). ………………………………....120
Figure 24 Soil sample results: Grindstone Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). …………………………..……..121
Figure 25 Soil sample results: Eastland Hills (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). No fungi were found at the locations marked in blue. ……………………………………………….…..……..122
Figure 26 Soil sample results: Timber Ridge (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B)…………………………..……..123
Figure 27 Soil sample results: Stoneridge Estates (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). No fungi were found at the locations marked in blue………………………..…………………………..……..124
x
Figure 28 Soil sample results: Lake Woodrail (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). No fungi were found at the location marked in blue………………………….…………………………..……..125
Figure 29 Soil sample results: Parkade North (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B)..…………………………..……..126
Figure 30 Soil sample results: Wynfield Meadows (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures,
15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium
(green) where found at 15oC (A) and 25oC (B). No fungi were found at the location marked in blue……….………………….…………………………..……..127
Figure 31 Relative frequency of occurrence of two soil fungi, Metarhizium and
Beauveria from forest and previously forested subdivisions around Columbia,
Missouri. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05)
………………………………………………………………………………………...128
Figure 32 Relative frequency of occurrence of two entomopathogenic fungi,
Beauveria and Metarhizium, in three habitats. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).…………….…………………………..……..129
xi
Figure 33 Relative frequency of occurrence of two soil fungi, Metarhizium and
Beauveria from forest and previously forested subdivisions around Columbia,
Missouri. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
………………………………………………………………………………………..130
Figure 34 Relative frequency of occurrence of Beauveria and Metarhizium fungi in forests and subdivisions of various ages. Forested habitats, 10 yr old previously forested subdivisions, and 20 yr old previously forested subdivisions were sampled to determine relative frequency of occurrence of entomopathogenic fungi…..…………………….…………………………..……..131
Figure 35 Relative frequency of mortality of Galleria mellonella (Linnaeus) by incubation temperature and fungus. Data are shown as mean ± standard error.
Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05). ……………………………...…………………………..……..132
Figure 36 Relative frequency of mortality of Galleria mellonella (Linnaeus) baits by entomopathogenic fungi, Beauveria and Metarhizium, in three habitats types.
Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).……..……..133
Figure 37 Relative frequency of mortality of Galleria mellonella (Linnaeus) by fungus and habitat type. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤
0.05). ……………………..……………………….…………………………..……..134
xii
Figure 38 Relative frequency of mortality of Galleria mellonella (Linnaeus) baits by entomopathogenic fungi, Beauveria and Metarhizium. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05)..………………………..……..135
Figure 39 Relative frequency of occurrence of Beauveria and Metarhizium according to habitat type, incubation temperature and fungi. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).………………………..……..136
Figure 40 Relative frequency of mortality of Galleria mellonella (Linnaeus) by incubation temperature and fungus. Data are shown as mean ± standard error.
Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05). ………………………….…………..…………………..……..137
Figure 41 Relative frequency of mortality of Galleria mellonella (Linnaeus) by
Beauveria and Metarhizium fungi through time in forest and subdivisions of various ages. Study areas used to sample soil along a temporal gradient of urban development in both forested and historically forested urban habitats. Forested habitats, 10 yr old previously forested subdivisions, and 20 yr old previously forested subdivisions were sampled to determine relative frequency of occurrence and of entomopathogenic fungi.….…………………………..……..138
Figure 42 Subterranean termite infested logs (A) were collected from the forest at Capen Park in Columbia, Missouri. Individual colonies were placed into swarming chambers (B). Reticulitermes flavipes (Kollar) built swarming tubes (C) and swarmed nearly a month after collection….…………………………..……..139
xiii
Figure 43 In 6-well dishes with Q5 filter paper moistened with dH2O in each chamber, sexed and paired Reticulitermes flavipes (Kollar) were placed in a holding cell for a day before treatment (A). Termites were placed into treatment cells (B) for 30 minutes and then transferred into clean chambers (C) for recovery and observation. The 6-well plates with termites were then placed into an environmental chamber at15oC, 80% ± 3% RH, 24h dark and checked every 3-4 days. ………………………….…………………………..……………………..…..140
Figure 44 Fungi used in this bioassay: forest Beauveria (BF), urban Beauveria
(BU), forest Metarhizium (MF) and urban Metarhizium (MU)….………..……..141
Figure 45 VanTieghem cells plus water agar are ½ cm pieces of sterile silicon tubing (1/4 inch inner diameter) placed in the center of a water agar plate (100 mm x 15mm). ………………………….……………………………………..……..142
Figure 46 Reticulitermes flavipes (Kollar) were monitored every 3-4 days for 60 days. Mortality for each pair, time to lay eggs and maximum number of eggs found during the study were recorded each time. Cadavers (B) were removed, surface sterilized and placed individually into 1.5 ml Eppendorf tubes containing
Potato Dextrose Agar (PDA) (C) to verify the emerging, sporulating fungus. If the termites survived (D), the date of first eggs and number of eggs (E) contained in the cell were recorded. ………………………….…………………………..……..143
Figure 47 No-choice virulence bioassay testing the virulence of spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban
Metarhizium (MU) or no spores as a control (C) towards R. flavipes…..……..144
xiv
Figure 48 No-choice virulence bioassay Lifetest results examining days to death for each member of the pair for 50% of the population of R. flavipes after exposure to spores from forest Beauveria (BF), urban Beauveria (BU), forest
Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control
(C)……………………………………………………………………………………..145
Figure 49 No-choice virulence bioassay PROC LIFETEST results examining days to death for each member of the pair for 50% of the population of R. flavipes after exposure to spores from forest Beauveria (BF), urban Beauveria
(BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). This figure shows colony response differences. ..…………..……...146
Figure 50 Time until half (T50) of Reticulitermes flavipes (Kollar) founding pairs laid eggs. BF=forest Beauveria, BU=urban Beauveria, MF=forest Metarhizium,
MU=urban Metarhizium, and C=control (no spores)….…………………..….…..147
Figure 51 Beauveria bassiana (Balsamo) Vuillemin sensu lato spores on a hemocytometer. Urban (A) and forest (B) collected fungi are similar sizes.…..148
Figure 52 Metarhizium anisopliae (Metschnikoff) Sorokin sensu lato spores on a hemocytometer. Urban (A) and forest (B) collected fungi are different sizes….149
Figure 53 Reticulitermes flavipes (Kollar) founding pairs were initially placed in the interstitial area with access to four chambers (A). Green filled circles are an example of the random combinations of two fungal treated chambers with the white cell representing control chambers. Termites are shown in the various ways found when observed: (B) together in a control chamber, (C) together in a
xv
fungal treated chamber, (D) separated into two chambers—could both be fungal treated or control or one of each, or (E) in the interstitial area.…………..……..150
Figure 54 Treatment chamber preference comparisons for Reticulitermes flavipes (Kollar). Forest Beauveria (BF), forest Metarhizium (MF) and urban
Metarhizium (MU). Sorted by treatment. α=0.05…………………………..……..151
Figure 55 Mortality comparisons of Reticulitermes flavipes (Kollar) founding pairs after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). ………………………….……….………………..……..152
Figure 56 Mortality comparisons of Reticulitermes flavipes (Kollar) founding pair individuals [d1 (A) and d2 (B)] after exposure to spores from forest Beauveria
(BF), forest Metarhizium (MF) and urban Metarhizium (MU)……………..……..153
Figure 57 Nesting site selection study examining the virulence of spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU) towards Reticulitermes flavipes (Kollar). d1= death of first individual of the founding pair. d2= death of second individual of the founding pair. ……..…….154
Figure 58 Nesting site selection by treatment. PROC Lifetest results examining days to death for 50% of Reticulitermes flavipes (Kollar) founding pairs after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). ……………………….…………………………………..155
Figure 59 Nesting site selection by treatment. PROC Lifetest results examining days to death days to death for 50% of each individual of the Reticulitermes flavipes (Kollar) founding pair for 50% after exposure to spores from forest
Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). ……..156
xvi
Figure 60 Cumulative mortality (cd) and mortality at each observation (td). Forest
Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU)..……..157
Figure 61 Nesting site selection by colony. PROC Lifetest results examining days to death for each member of the Reticulitermes flavipes (Kollar) founding pair for 50% after exposure to spores from forest Beauveria (BF), forest
Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control
(C)……………………………………………………………………………………..158
Figure 62 Head and abdomen regions and associated glands found on
Reticulitermes. (Figure adapted from Costa-Leonardo and Haifig 2010)……..159
Figure 63 Reticulitermes flavipes (Kollar) were sexed morphologically using the characteristics taken from Lainé and Wright (2003): Reticulitermes spp. females have an enlarged seventh sternite (A) that covers the eighth and ninth sternites whereas the male’s seventh sternite is similar in size to the sixth sternite (B); the male also has a pair of styli on the ninth sternite (C). …….……………………160
Figure 64 A solution containing spores of Beauveria or Metarhizium (A) was mixed with Reticulitermes flavipes (Kollar) head supernatant to achieve 0%
(controls) and 10% solution (10% head to 90% spore solution). ………………161
Figure 65 After the elapsed incubation time [Metarhizium 29 hr (B), Beauveria 31 hr (A)], spore germination images were digitally taken using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera.
Images were taken at 400x magnification. Spore germination tubes (C and D) are shown for Beauveria and Metarhizium…………………………………………….162
xvii
Figure 66 Metarhizium spore germination remained the same when exposed to
10% Reticulitermes flavipes (Kollar) head or abdomen extract (B) compared to the control (A). Images were digitally taken using CellSense software on a Nikon
Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification. …………………………………………..……..163
Figure 67 Metarhizium spore germination means after exposure to
Reticulitermes flavipes (Kollar) abdomen or head extracts. Values are shown as mean (columns) and standard error………………………………………………..164
Figure 68 Beauveria spore germination decreased when exposed to 10%
Reticulitermes flavipes (Kollar) head extract (B) compared to the control (A).
Images were digitally taken using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification. ………………………….……………………….……………..……..165
Figure 69 Beauveria spore germination means after exposure to Reticulitermes flavipes (Kollar) abdomen or head extracts. Values are shown as mean (columns) and standard error………………………….……………..…………………..……..166
xviii
List of Tables
Table 1 Descriptive statistics for subdivisions in Columbia, MO where soil samples were taken to bait for entomopathogenic fungi. (* = information updated from Botch 2013) ………………………….……………..…………………..……..167
Table 2 Relative frequency of occurrence. Total sampling locations where
Beauveria and Metarhizium were found when baited from Galleria mellonella
(Linnaeus) larvae at 15oC and 25oC. Soil was taken in forested, 10 yr previously forested, and 20 yr previously forested subdivisions. Ten homes or plots were sampled within each habitat. Neither Beauveria or Metarhizium emerged from control soil samples. Gu=Guilford; MH=Monterey Hills; Gr=Grindstone;
EH=Eastland Hills; TR=Timber Ridge; SE=Stoneridge Estates; LW=Lake
Woodrail; PN=Parkade North; WM=Wynfield Meadows..………………..……..168
Table 3 Relative frequency of mortality. Soil fungi emerging from Galleria mellonella (Linnaeus) larvae exposed to soil from forest, 10 yr urban, and 20 yr urban historically forested landscapes incubated at two temperatures (15oC and
25oC ). The top table shows the number of G. mellonella) exhibiting Beauveria bassiana (Balsamo) Vuillemin infections and the bottom table shows the number of waxworms exhibiting Metarhizium anisopliae (Metschnikoff) Sorokin infections at each sample location. ……….………….……………..…………………..……..169
Table 4 Relative frequency of occurrence and frequency of mortality of Galleria mellonella (Linnaeus) larvae exposed to fungi found in soils from forests, 10 yr subdivisions and 20 yr subdivisions in Columbia, Missouri. ANOVA performed using a link logit with a binomial distribution. Type=habitat (forest, 10 yr, 20 yr);
xix
nDF=Numerator Degrees of Freedom; dDF= Denominator Degrees of Freedom; ns=non-significant; * = p ≤ 0.05; ** = p ≤ 0.01; † = p ≤ 0.10……………..……..170
Table 5 The USDA-ARS Collection of Entomopathogenic Fungal Cultures
(ARSEF) Collection information for fungi used in this study. ...…………..……..171
Table 6 No-choice virulence study overall statistical significance. Forest
Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban
Metarhizium (MU). Significant values in bold, α=0.05...…………………..……..172
Table 7 No-choice virulence bioassay testing the virulence of spores from forest
Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban
Metarhizium (MU) or no spores as a control (C). d1=death of first individual of the founding pair. d2=death of second individual of the founding pair. Significant values in bold, α=0.05. *=significant; **=very significant. ………………..……..173
Table 8 Percentage of Reticulitermes flavipes (Kollar) founding pair individuals killed by exposure to Beauveria and Metarhizium in forest and urban habitats.
………………………….……………..…………………………………….…..……..174
Table 9 No-choice virulence bioassay T50 results examining days to death for each member of the pair for 50% of the population of Reticulitermes flavipes
(Kollar) after exposure to spores from forest Beauveria (BF), urban Beauveria
(BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). A: d1=death of first member of the pair. B: d2=death of second member of the pair. Significant values in bold, α=0.05. *=significant; **=very significant. ………………………….……………..………………….………..……..175
xx
Table 10 No-choice virulence bioassay T50 results examining the days until the first egg was laid for 50% of the Reticulitermes flavipes (Kollar) population when exposed to spores from forest Beauveria (BF), urban Beauveria (BU), forest
Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C).
Significant values in bold, α=0.05. *=significant; **=very significant. …..……..176
Table 11 Percentage of observations of Reticulitermes flavipes (Kollar) found in control (C) or fungus (F) chambers or interstitial area (I). ……………….……..177
Table 12 Choice virulence study overall statistical significance. Reticulitermes flavipes (Kollar) founding pairs were more likely to die when exposed to urban fungi than forest fungi. Significant values in bold, α=0.05………………..……..178
Table 13 Choice virulence bioassay testing the virulence of spores from forest
Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). These tables contrast the mortality of Reticulitermes flavipes (Kollar) exposed to the various fungi. A: td=Overall mortality. B: d1=death of first individual of the founding pair. C:d2=death of second individual of the founding pair. D:td=Overall mortality by colony. Significant values in bold, α=0.05, *=significant. T50 results in blue with the first number being the mean number of days for 50% of founding pair individuals to die. ………………………….……………..……………………..179
Table 14 Metarhizium spores germinated 9.29 times more often than Beauveria
(p=<0.0001). ………………………….……………..………………………..……..180
Table 15 Metarhizium spore germination results when exposed to Reticulitermes flavipes (Kollar) body extracts…………….……………..…………………..……..181
xxi
Table 16 Analysis of Beauveria and Metarhizium spore germination when exposed to Reticulitermes flavipes (Kollar) founding pairs body extracts from head and abdomen..…………………….………………..………………….……..182
Table 17 Beauveria spore germination results when exposed to Reticulitermes flavipes (Kollar) body extracts.………….……………..….………………………..183
xxii
Abstract
In forest habitats, termites break down woody debris and assist in nutrient cycling of carbon and nitrogen. However, infestations in suburban habitats have given subterranean termites a high priority for pest management. Biological control of subterranean termites has been unsuccessful using entomopathogenic fungi, such as
Beauveria and Metarhizium. Previous biocontrol research focused mostly on the worker caste and exploiting social behaviors. Due to many social and physiological defenses, large termite colonies survive fungal invasions. This research focused on founding pairs of primary reproductives when colonies are smallest and there are only two termites.
Because most termite colonies invading newer homes in Missouri are started by
Reticulitermes flavipes primary kings and queens, control of founding pairs by preventing establishment may be an important method of termite control. Using locally collected Beauveria and Metarhizium, this research compared mortality and sublethal effects on founding pairs by exposure to fungal spores from forest and urban habitats.
Behavior of founding pairs in the presence of any spores and fungistatic effects of founding pair head and abdomen extracts on spore germination were also examined.
Forest Beauveria spores were more virulent than Metarhizium spores when founding pairs were examined in a no-choice bioassay. Sublethal effects on egg laying were also observed. When founding pairs where given a choice to nest in areas with or without spores, Metarhizium spores were more virulent. Metarhizium germination was not affected in the presence of imago body extracts. Beauveria germination decreased in the presence of imago head extracts but increased in the presence of imago abdomen extracts.
xxiii
Chapter One
LITERATURE REVIEW
Introduction
In the early 1800s, the French silk industry was in trouble. Silk worms were dying and when they did, they were covered with a white powdery substance. No one knew what the substance was or what was causing the death of domesticated silkworms. In 1835, after investigating for over twenty years,
Agostino Bassi discovered the white substance to be a parasitic fungal muscardine disease, Beauveria bassiana (Balsamo) Vuillemin (Vega et al. 2009).
Green muscardine disease, or Metarhizium anisopliae (Metschnikoff) Sorokin was discovered soon afterward and the causal fungus has received much attention as a biological control agent for a wide range of insects (Rath 2000)
(Figure 1). Entomopathogenic fungi are now well known natural enemies to many insects (Shah and Pell 2003) and research continues to discover the biological and ecological implications of these organisms in the environment.
The dark, warm, moist environmental conditions created in the chambers of subterranean termite nests are consistent with the growth requirements for entomopathogenic fungi (Kramm et al. 1982). Combined with the social habits of subterranean termites, it was thought that this would allow entomopathogenic fungi to be a potentially effective form of biological control. In 1994, Bio-Blast®, an insecticide containing spores of M. anisopliae, was marketed as a highly effective treatment for subterranean termite control showing as great as 100% mortality in laboratory experiments when treating as few as 5-10% of the termites
1 within a colony (Rath 1995, Rath 2000). Infections of baits by M. anisopliae killed lab colonies of termites fast, but not too fast, so the spores could be transferred to nestmates through social interactions. However, in field trials, Bio-Blast failed to cause the same mortality. The failure was due to a combination of subterranean termite defensive mechanisms, such as behavioral avoidance of the fungus, production of antimicrobial compounds, allogrooming, gut antifungal activity, cellular and humoral immunity, necrophagy, corpse avoidance, and microbial competition (Chouvenc and Su 2010).
Beauveria bassiana and M. anisopliae abundance varies between different ecosystems with B. bassiana found more commonly in forests and M. anisopliae found more commonly in agriculture fields and disturbed environments
(Sanchez-Pena et al. 2011). Sooker et al. (2008) found M. anisopliae isolates were recovered more frequently from soils at 25-30°C while more B. bassiana isolates were recovered at 15°C. Physiological properties and densities of insect hosts can also influence the distribution and frequency of occurrence of entomopathogenic fungi (Rodrigues et al. 2005). In a comparative survey of
Reticulitermes subterranean termite communities in undeveloped forests and urban habitats of Missouri, Pinzon and Houseman (2009) found that different species were most abundant in forest habitats compared to urban habitats. The biotic and abiotic factors related to the shift in dominant subterranean termite species when habitats are converted from forests to urban are not completely understood, but may be related in part to biotic factors like shifts in pathogens such as entomopathogenic fungi.
2
Study Organisms
Termites. Historically, taxonomists placed termites in their own order,
Isoptera, but termites were classified within the order Blattodea. Dictyoptera, a taxonomic group combining Isoptera with two closely related orders, Blattodea
(cockroaches) and Mantodea (mantises) (Nalepa and Bandi 2000), was based on shared autoapomorphic traits including eggs encased within an ootheca and perforation in the head capsule’s tentorium (Inward et al. 2007). The disputed phylogenetic relationships between these groups was examined molecularly by
Inward et al. (2007) who found that Isoptera is nested within the order Blattodea and that termites are a sister taxonomic group to the cockroach genus
Cryptocercus. This implies that Blattodea is a paraphyletic order and termites are eusocial cockroaches (Inward et al. 2007). The authors recommended that
Isoptera be demoted to a family within Blattodea.
The demise of the Isopteran order did not go undisputed. Lo et al. (2007) argued that Isoptera is a well studied monophyletic group, and at the very least, the families within Isoptera should retain their family status. Eggleton et al. (2007) proposed another nomenclature. While still eliminating the order Isoptera, the name Termitoidae has been suggested at an epifamily level allowing for a distinction of the monophyletic termite taxon within Blattodea, but leaving the established family names untouched (Eggleton et al. 2007). Still, despite the evidence and arguments, the order name, Isoptera, remains in use in the literature.
Seven extant families of termites are commonly recognized:
Mastotermitidae (the most basal termite family), Termopsidae (the dampwood 3 termite), Hodotermitidae (the harvester termites), Kalotermitidae (the drywood termites), Rhinotermitidae (the subterranean termites), Serritermitidae, and
Termitidae (the higher termites) (Engel et al. 2009, Lo and Eggleton 2011).
Stolotermitidae, Archeorhinotermitidae and Stylotermitidae are clades elevated to family status by Engel et al. (2009).
Approximately 3000 species of termites are recorded worldwide (Engel et al. 2009). However, only 80 species are considered serious pests (Edwards and
Mill 1986, Rust and Su 2012). In a recent review by Evans et al. (2013), only 28 termite species were reported as invasive. Thirteen pest species are drywood termites, and the remaining are subterranean termites. Coptotermes,
Reticulitermes, and Cryptotermes are genera of particular importance due to distribution and seriousness of damage (Rust and Su 2012).
All termites consume and break down cellulose as their primary source of nutrition (Noirot and Noirot-Timothee 1969, Breznak and Brune 1994). Termites feed primarily on wood, most species requiring the use of protozoan or bacterial gut microbial symbionts, to hydrolyze lignocelluloses and convert them into digestible sugars (Thorne 1997, Brune and Ohkuma 2011). In forest ecosystems termites break down woody debris and thus assist in the nutrient cycling of organic carbon and nitrogen (Slaytor and Chappell 1994, Holt and Lepage 2000).
In urban ecosystems, infestations in man-made structures have made termites a high priority for pest management. Estimated worldwide termite control and repair costs are $40 billion, with subterranean termites accountable for approximately 80% of the amount (Rust and Su 2012). In the United States
4 alone, subterranean termites cause an estimated $1.2 billion of damage annually to structures (Su and Scheffrahn 1998). There is almost no place in the United
State that is free from the risk of termite infestation. In Columbia, Missouri, there is a moderate to heavy risk (Figure 2).
Termites are eusocial insects, defined as insects with a reproductive division of labor, overlapping generations, and cooperative caring of the young
(Wilson 1971). The division of labor is found in a distinctive caste system, generally characterized by workers, soldiers and reproductive adults. Termites are hemimetabolous, developing in three distinct stages, egg, nymph and the adult, as compared to holometabolous insects that undergo complete metamorphosis and have a true larval stage (Figure 3). However, in the termite literature, apterous immatures are often referred to as larvae, while immatures with wing pads who are differentiating towards reproductive development are called nymphs (Thorne 1996).
Subterranean termites. Subterranean termites build decentralized underground nests that consist of loosely interconnected tunnel systems leading to feeding sites at the soil surface (Noirot 1970). The tunnel walls are lined with a mixture of soil, feces and saliva to maintain optimal levels of humidity and to create a microclimate necessary for survival (Lee and Wood 1971).
Incipient subterranean termite colonies are formed from synchronized mating events called swarming (Lepage and Darlington 2000). Winged reproductive termites gather and disperse simultaneously from established nests, shed their wings, find a mate, and then locate a suitable site for starting a new
5 colony (Shellman-Reeve 1997). A copularium is formed and eggs are laid, tended first by the primary reproductive pair alone and later by their developing worker offspring. Reticulitermes subterranean termites usually begin egg-laying
20 days after copulation (Weesner 1970). Unlike the social hymenoptera, if a subterranean termite reproductive king or queen dies later, it will be replaced by a secondary reproductive king or queen (Weesner 1965).
Typically, subterranean termites invade structures through foundation cracks, conduits, expansion joints, or by constructing shelter tubes on foundation walls. Their cryptic behavior may go unnoticed even when thorough inspections are performed. Subterranean termites maintain a soil connection due to the need for moisture. Mature subterranean termite colonies are large and cause the greatest amount of structural damage (Su and Schefferahn 1998). They are also accountable for a greater proportion of treatment and repair costs (Rust and Su
2012). Important subterranean termite species in North America include
Reticulitermes flavipes (Kollar) and Coptotermes formosanus Shiraki.
In Missouri, there are four species of termites, all of which are subterranean and in the genus Reticulitermes (Pinzon and Houseman 2009). In a comparative survey of Reticulitermes subterranean termite communities in
Missouri, Pinzon and Houseman (2009) found that different termite species were the most abundant in forest habitats versus urban habitats. Reticulitermes hageni
(Banks) colonies were more frequently collected in forested habitats and R. flavipes were more common in urban habitats. Reticulitermes virginicus (Banks) and Reticulitermes tibialis (Banks) were also found but in smaller frequencies.
6
Biotic and abiotic factors related to the shift in dominant subterranean termite species as habitats are converted from forests to urban are not completely understood, but may be related in part to biotic factors, including the presence of pathogens such as entomopathogenic fungi.
Reticulitermes flavipes prefers wetter conditions than R. hageni and is more frequently found in urban landscapes near homes (Botch 2013). The soil near homes is often more moist due to landscape coverings and irrigation common in urban settings (Botch 2013). Reticulitermes hageni is more frequently found in the drier conditions of the forest.
Entomopathogenic Fungi. Most entomopathogenic fungi are found in the order Hypocreales, Division Ascomycota including the common genera:
Aspergillus (Micheli), Beauveria (Balsamo), Culicinoces, Hirsutella (Patouillard),
Metarhizium (Metschnikoff), Nomurea (Yasuda), Isaria (=Paecilomyces)
(Samson), Tolypocladium and Lecanicillium (=Verticillium) (Gams and Zare)
(Inglis et al. 2001). Often, these fungi are soil borne opportunistic pathogens of the arthropods they infect (Meyling 2007, Sun et al. 2008).
Beauveria and Metarhizium are known to be facultatively necrotrophic and pathogenic toward arthropods, and are found in soils worldwide (Roberts and
Hajek 1992, Charnley and Collins 2007). They can also occur as saprotrophs and plant endophytes (Vega et al. 2008, Behie et al. 2012).
Beauvaria bassiana and M. anisopliae are anamorphs, meaning they reproduce asexually. Each fungus has multiple strains, but the fungi are known by the aforementioned anamorphic names. The exact phylogeny has not been
7 established for these fungi and both B. bassiana and M. anisopliae are considered species complexes (Li et al. 2001, Zimmerman 2007, Bischoff et al.
2009). Recently, Beauveria and Metarhizium were revised and, in both cases, previously recognized varieties of M. anisopliae and B. bassiana were upgraded to species level (Li et al. 2001, Bischoff et al. 2009).
Beauveria. In 1972, de Hoog recognized three species of Beauveria: B. bassiana, B. brongniartii, and B. alba (Limber Saccas). Beauveria amorpha
Samson & Evan, B. caledonica Bisset & Widden, B. malawiensis Rehner et al. were recognized over the next three decades (Rehner et al. 2011). It has also been argued that B. bassiana and B. brongniartii contain cryptic phylogenetic species complexes in which they redescribed B. bassiana and B. brogniartii and described novel species: B. asiantica, B. australis, B. kipukae, B. pseudobassiana, B. sungii and B. varroae (Rehner et al. 2011).
The species Beauveria bassiana is characterized as “cottony, powdery, velutinous, or wooly, white or yellow or rarely pink, margin white, reverse uncolored or yellow and occasionally with pink to red soluble pigment that discolors the medium, at times producing synnemata or irregular synnema-like structures” (Rehner et al. 2011). Beauvaria bassiana, alone, has a known teleomorph, Cordyceps bassiana. The teleomorph has only been collected in
Asia from the cadaver of a carpenterworm (Lepidoptera: Cossidae) larva in China
(Li et al. 2001). The conidiogenous cells are 1.5-5.5 µm, globose to flask-shaped and conidia are found on an indeterminate, denticulate rachis, and are
8 characterized morphologically by one-celled, terminal holoblastic conidia (Figure
4).
Metarhizium. As with Beauveria, this genus was originally classified based on morphological characteristics and has a complicated taxonomy
(Bidochka and Small 2005). The columns of dry, green columnar conidia rest in chains on the phialides (Hoog et al. 2000, Zimmerman 2007) (Figure 5). Two species, M. anisopliae and M. flavoviride, and two morphological forms of
Metarhizium anisopliae were described by Tulloch (1976). Metarhizium anisopliae var. anisopliae has a short spore (5-8 µm) while M. anisopliae var. majus has a long conidial form (10-16 µm) (Zimmerman 2007).
Based on a multigene phylogenetic approach, Bischoff et al. (2009) proposed to elevate nine varieties of terminal taxa in the M. anisopliae complex to species: M. anisopliae, M. guizhouense, M. pingshaense, M. acridum stat. nov., M. lepidiotae stat. nov. and M. majus stat. nov. Additionally, two new species were described, M. globosum and M. robertsii in the same paper.
Entomopathogen Pathology
More than 100 mycoinsecticide products are commercially available
(Jaronski 2010). Due to the broad spectrum of insects killed by Beauveria and
Metarhizium, they have been exploited for their insecticidal potential. Each has commercial products marketed towards pestiferous arthropods, such as flies, thrips, orthopterans, beetles, and termites (Rath 2000) (Figure 6).
Entomopathogenic fungi invade their insect hosts through the cuticle rather than through the alimentary tract (Roberts and Humber 1981). The fungi
9 infect arthropods by first recognizing the host surface, adhering to the host cuticle, germinating and then penetrating the cuticle surface using both mechanical pressure from the appressorium and cuticle-degrading enzymatic action, entering the hemocoel, producing toxins and ultimately causing the death of the host (Roberts and Humber 1981, Bidochka and Small 2005, Ortiz-Urquiza and Keyhani 2013). The fungal mycelia then reemerge through the intercuticular regions to sporulate outside the cadaver (Figure 7).
Successful invasion by these fungi is due to the susceptibility of the insects and to abiotic factors that affect the pathogenicity of the fungal agent
(Ferron 1978). As biological organisms, fungi have optimal conditions in which they successfully act as pathogens. Temperature, soil type, soil moisture, and relative humidity are all factors that must be taken into consideration when applying entomopathogenic controls (Ferron 1978).
Both solitary and social insects are affected by entomopathogenic fungi and both groups have developed immune defenses against fungal invaders.
However, social insects have additional behavioral and physiological defense mechanisms that protect them from the effects of these fungi (Roy et al. 2006,
Chouvenc et al. 2011b, Rosengaus et al. 2011). Social interactions such as allogrooming, trophallactic feeding, signaling and avoidance, and removal or burial of infected nestmates increase the chances of colony survival (Rosengaus et al. 1999a, Chouvenc et al. 2011b). Although subterranean termites may become contaminated externally on the cuticle by spores in the soil environment, nestmates ingest the spores during grooming activities but do not become
10 infected due to inhibition effects (Chouvenc et al. 2009b). Additionally, individual immune responses are more effective when living in a group due to an effect called social immunity, or herd immunity (Traniello et al. 2002, Rosengaus et al.
2011). Adaptive immunity has been shown in ant colonies where allogrooming increases after the colony is exposed to fungus and individual ants were less likely to die because of the increased grooming by nestmates (Walker and
Hughes 2009). Termites may also have additional mechanisms to inhibit infection. In a recent study by Rosengaus et al. (2014), the microbial symbiotic community within the termite gut assisted infected termites to overcome a fungal attack by metabolizing fungal mycelia.
Traditional methods of subterranean termite control are toxic and expensive and alternative methods may be necessary. Subterranean termite biological control is very attractive, yet it still is not a reality (Chouvenc et al.
2011b). Chouvenc et al. (2011b) explain in their review of the past fifty years of attempted biological termite control that more basic biological and ecological research is necessary to understand new avenues of fungus-termite interactions and implications for control. Most of the research performed on subterranean termite control with entomopathogenic fungi has been on mature colonies with large populations, potentially in the millions. In contrast, m research focuses on incipient colonies of recently emerged primary reproductives. In this colony stage, there are only two individuals, the founding king and queen of a future colony. Because most subterranean termite colonies that infest homes in
Missouri subdivisions are started by founding pairs of primary reproductives
11
(Botch 2013), control of R. flavipes founding pairs by preventing establishment using entomopathogenic fungi may be an important method of urban subterranean termite control to consider.
The ecological influence of Beauveria and Metarhizium on R. flavipes is not well understood, and this research seeks to answer questions about this relationship. This research looks at differences in fungal frequency and virulence
(mortality of termites) in forest and urban habitats, as well as the influence of these fungi on R. flavipes in a laboratory setting.
Research Objectives
This study has four objectives. The first objective explores the presence- absence and relative frequency of occurrence of Beauveria and Metarhizium, between soils in forested and urban habitats. Soils were compared between oak- hickory forests, and ten year old or twenty year old subdivisions that were previously forested. The second objective examines the virulence of Beauveria and Metarhizium from forested and urban habitats on founding pairs of R. flavipes. Using a no-choice bioassay, the ability of the founding pair to survive exposure to entomopathogenic fungi was determined, and sublethal effects associated with female fecundity were measured. Objective three examines the behavioral responses of founding pairs of R. flavipes when exposed to fungal spores of Beauveria and Metarhizium using a choice bioassay to determine whether founding pairs perceive the presence or virulence of entomopathogenic fungi and avoid areas where spores are found. The fourth objective explores whether fungistatic compounds from glands on the head or in the rectum of R.
12 flavipes are present and active against Beauveria and Metarhizium. By answering these questions, perhaps alternatives for safe treatments may be found and new methods to prevent infestation may be developed.
13
Chapter Two
COMPARING THE RELATIVE FREQUENCY OF OCCURRENCE AND FREQUENCY OF MORTALITY OF TWO ENTOMOPATHOGENIC FUNGI, BEAUVERIA AND METARHIZIUM (ASCOMYCOTA: HYPOCREALES), FROM FORESTED AND URBAN HABITATS IN MISSOURI
Introduction
Entomopathogenic fungi are natural enemies to a wide range of insects
(Roy et al. 2006). When the sporulating fungi come into contact with an insect, the spores germinate and penetrate the insect cuticle, invade the hemocoel, produce toxins, and eventually cause death of the insect host. Beauveria bassiana (Balsamo) Vuillemin and Metarhizium anisopliae (Metschnikoff) Sorokin
(Ascomycota: Hypocreales), two common entomopathogenic fungi, are found in soils throughout the world and although these fungi are used extensively for biological control purposes, little is known of their ecology (Meyling et al. 2009).
Beauveria bassiana was first identified in 1835 as a living entity capable of causing death of insects (Vega et al. 2009). After B. bassiana kills its insect host, the cadaver is coated by white mycelium and white powdery conidia. The one- celled, terminal holoblastic conidia are globose and found on indeterminate, denticulate rachises (Figures 1A, 4). Metarhizium anisopliae emerges from insect cadavers, first with white mycelium that is soon covered in usually olive green conidia found in columns resting in chains on the phialides (Hoog et al. 2000,
Zimmerman 2007) (Figures 1B, 5). Both fungi were originally described morphologically but were recently revised and split into multiple species using molecular methods. The fungi in this chapter were identified morphologically as
14
Beauveria bassiana sensu lato (s.l.) and Metarhizium anisopliae s.l. (Hoog 1972,
Hoog et al. 2000).
Studies have examined entomopathogenic fungal preferences on meadows, orchards, forests, organic and conventional agricultural lands, temperate regions, field margins, pastures, and chaparral areas (Vänninen 1996,
Chandler et al. 1997, Vänninen et al. 1989, Klingen et al. 2002, Keller et al. 2003,
Rodrigues et al. 2005, Meyling and Eilenberg 2006, Meyling et al. 2009,
Sánchez-Peña et al. 2011, Stranne 2014). These studies have found that fungal community differences in soil environments exist in different habitats with less disturbed habitats often correlating with higher populations of fungi (Meyling et al.
2009). Studies in Canada found B. bassiana was isolated more frequently from soils in northern locations compared to M. anisopliae (Bidochka et al. 1998).
Using the Galleria bait method (Zimmerman 1986), the Bidochka et al. (1998) study found B. bassiana was isolated more frequently from larvae baited in soils incubated at 8 and 15°C, while M. anisopliae was isolated most frequently at
25°C.
To assess fungal frequency of occurrence in habitats, a sensitive detection method was developed by Zimmerman called the Galleria bait method
(Zimmerman 1986, Meyling 2007). This method introduces waxworm larvae,
Galleria mellonella (Linnaeus) (Lepidoptera: Pyralidae), to soil where existing spores may attach to the larvae. Galleria mellonella is a parasite of bee colonies that feeds on honeycomb, pollen, propolis and honey (Paddock 1918, Nielson and Brister 1979). The larvae of this moth do not have contact with soil naturally
15 and therefore, have no evolutionary basis for immune defenses to pathogens found in soils, making them a beneficial model for soil pathogen studies
(Greenwood and Rebek 2009). Last instar larvae are about 2 cm long and weigh approximately 250 mg (Ramarao et al. 2012). Also, larvae have a high tolerance for temperature variation allowing for infection studies to be conducted between
15°C to at least 37°C (Ramarao et al. 2012).
One habitat not examined in previous entomopathogenic fungal habitat comparison studies is urban habitats. Urbanization converts indigenous habitat to anthropogenic land use (McIntyre et al. 2001). This process changes the existing habitat by removing indigenous soil and vegetation, altering resource availability, creating habitat fragmentation and potentially causing localized extinction events by eliminating native biota (McIntyre 2000, Turner et al. 2001, McKinney 2002).
The effect of the drastic alteration of the indigenous habitat by devegetation, paving, and the process of building structures on large parcels of land when creating urban habitats exceeds that of logging and traditional farming land uses
(Marzluff and Ewing 2001).
Most studies on urbanization show a decline of native species while also showing an increase in abundance and species richness for many non-native species (McKinney 2006). This trend has been found in insects (McIntyre 2000), mammals (Mackin-Rogalska et al. 1988) and birds (Marzluff 2001). Botch (2013) found that urbanization of historically forested habitats first eliminated subterranean termite populations when vegetation and soil were removed, and then altered species composition in ten and twenty year old subdivisions.
16
Whereas R. hageni typically dominated forested habitats prior to urbanization, R. flavipes was most frequently found in urban habitats (Pinzon and Houseman
2009, Botch 2013).
Meyling and Eilenberg (2007) called for more studies using local isolates that interact with local host populations to understand the ecology of indigenous fungal populations. Research with local B. bassiana and M. anisopliae isolates is important to establish groundwork for ecological interaction studies with the subterranean termites, Reticulitermes flavipes (Kollar) which lives in forest and urban habitats.
The overall goal of this study was to examine how B. bassiana s.l. and M. anisopliae s.l. populations respond to urbanization (Figure 8). The specific objectives of the study are: 1) how relative frequency of occurrence of fungal populations change as subdivisions develop; 2) whether differences exist in relative frequency of mortality of waxworms exposed to fungal spores in the soil from different habitats; and 3) whether there are temperature preference differences at which fungi from these soils are most virulent. It was expected that
B. bassiana s.l. and M. anisopliae s.l. would be found. Different frequencies of theses fungi were expected. Greater frequencies of M. anisopliae s.l. in urban habitats were expected while B. bassiana s.l. was expected to be more frequent from undisturbed forest habitats. It was also anticipated that B. bassiana s.l. would be found more often at 15oC than 25oC and M. anisopliae s.l. would be found more often at 25oC than 15oC.
17
Methods
Study areas
Study locations in and near Columbia, Missouri, were selected using aerial photography from the Boone County Assessor’s Office Parcel Information Viewer
(http://maps.showmeboone.com/viewers/AS_ParcelMapping_v1/), The Center for
Applied Research and Environmental Systems (CARES, http://www.cares.missouri.edu/), Google Earth and personal observations. Many of these study sites were also examined for subterranean termite and habitat comparison by Botch (2013). Aerial photography from Missouri Spatial Data
Information Service (http://msdis.missouri.edu), CARES, and Boone County
Assessor's Office Parcel Information Viewer were used to select forests and subdivisions in his study. Urban landscapes were considered previously forested when over 59% of the area had been forested prior to urbanization. Historical and contemporary aerial photography show the forested habitat 30 years ago in each area and the forest or urban development of each area (Figures 9-17).
Soil collection
Soil was sampled from nine habitats—subdivisions and forests—in
Columbia, Missouri. The study areas chosen fit into one of three habitat types: 1) forested habitats undeveloped for at least 30 years; 2) Ten year old urban subdivisions built on historically forest habitats; or 3) Twenty year old urban subdivision built on historically forest habitats. Soil from three different forests, 10 year old subdivisions, and 20 year old subdivisions were sampled.
18
Forest locations tested were the Guilford forest, Monterey Hills and
Grindstone forest. In these oak-hickory forests of at least 30 years in age, an eight by eight grid measuring total average length and width of 115 m2 was laid out and each grid square was given a sequential number. Sample locations were chosen using a random number generator (https://www.random.org/) to randomly select which squares were sampled. GPS coordinates of the corresponding grid square were established using Google Earth and located in the field using a handheld GPS unit while on horseback or on foot.
Ten year subdivisions were Eastland Hills, Timber Ridge and Stoneridge
Estates. Twenty year subdivisions were Lake Woodrail, Parkade North and
Wynfield Meadows (Figure 18). Subdivisions built within the appropriate age ranges were chosen using data from the Boone County Assessors website. The average age of all homes in the ten year old subdivisions was 10.9 years old.
The average age of all homes in the twenty year old subdivisions was 20.9 years old (Table 1). Homeowners were contacted and asked for permission to take the soil samples. When permission was granted, samples were taken in the back yard, as close to the center of the yard as possible.
Soil samples were randomly taken using a paper circle divided into six sectors that was spun and placed on the ground (Figure 19). Three previously selected random numbers (https://www.random.org/) determined the sectors chosen and three paces were taken away from the circle. At each of the three sublocations within each sample location, leaf litter or turf was removed; a 12.7 cm (5 inch) deep and 6.4 cm (2 ½ inch) wide soil sample was extracted by a
19 trowel, and placed into a plastic 3.8 L (1 gallon) sized zipper sealed bag (Figure
20 A). The trowel was disinfected between each location within each habitat with
80% EtOH. This process occurred at each of the three sublocations and the samples were combined into one bag.
Ten locations were sampled in each forest and in each subdivision for a total of ninety locations. Sampling took place over a two year period (2011 and
2012) during the months of August and September. Soil samples were refrigerated at 4oC until ready for use and were used in the study within three months of collection. Soil type and pH were not tested as other studies showed that the occurrence of B. bassiana and M. anisopliae were not related to these abiotic factors (Vänninen 1996, Bidochka et al. 1998).
Bait Preparation
The soil subsamples were thoroughly mixed within each sample’s bag, then spread on individual sterile trays and allowed to sit in a closed room where the samples could dry. Soil samples were then sifted through a 2 mm sieve.
Each sieve was cleaned between samples with 80% EtOH. Eighty ml of soil from each of the ninety locations was placed into two sterile 100 ml plastic containers.
Containers were sealed with lids containing six small holes for air flow (Figure 20
B). The two containers from each sample location were randomly assigned to the temperature (15oC or 25oC) at which it would be studied by flipping a coin. Five ml of sterile dH2O were added to the soil in each container and mixed to homogenize.
20
Fungi were isolated from the soil using the Galleria bait method
(Zimmermann 1986, Meyling 2007). Galleria mellonella larvae were purchased from Timberline Fisheries. Before shipping, the larvae were cold treated to prevent webbing (as per conversation with a Timberline representative). Upon arrival, waxworms were observed for 24 hours before using them in the study to assure their health. Waxworms have seven to nine instars, based on food quality and temperature (Esperk et al. 2007). Ten last-instar waxworms were placed into each container with the soil and the lid was attached (Figure 20 C). Each container was inverted and placed in a randomly selected environmental chamber at 15oC or 25oC, 24 hours darkness and approximately 80% relative humidity (RH) (Figure 20 D). Relative humidity was maintained by placing an open container of sterile water into the environmental chamber.
Soil samples were shaken daily for the first seven days to ensure waxworm exposure to the soil. Each sample was also checked, larvae counted and cadavers removed weekly for nine weeks (Figure 21 B). Cadavers were then removed and surface sterilized in 1% sodium hypochlorite (NaClO) for three minutes, rinsed in sterile dH2O and dried on sterile filter paper (Fisher Scientific
Q5). Cadavers were placed individually into moisture chambers created using 50 mm petri dishes containing sterile cotton moistened with 1 ml of sterile dH2O and the edges sealed with parafilm (Figure 21 C).
Soil used for control containers was sterilized by autoclaving twice.
Otherwise, control soils for forests, ten-year urban and twenty-year urban habitats were handled in the same manner as treatment samples. During
21 waxworm cadaver inspections, controls were also handled first so as to avoid contamination.
Fungal Identification
When fungal mycelia and spores emerged from cadavers, the fungi were identified (Figure 21 E and F) by observing cadavers under a dissecting microscope at 30-40x magnification. Beauveria bassiana s.l. and M. anisopliae s.l. fungi were identified morphologically based on the morphology of spores and mycelia. Beauveria bassiana s.l. morphological identifications were based on
Hoog (1972) and Metarhizium anisopliae s.l. morphological identifications were based on Hoog et al. (2000).
Analysis
Data were analyzed using PROC GLIMMIX using link logit and the distribution was bionomial using SAS® software (version 9.3) (α ≤ 0.05). A split plot design was used with the main plot being habitat type (forest, 10 year urban,
20 year urban), then by fungus (B. bassiana s.l. and M. anisopliae s.l.) as the subplot, incubation temperature (15°C and 25°C) as the sub-sub-plot and all possible interactions. The F statistic for habitat type was subdivision within habitat type. Main effects interaction was determined by residual means squared.
A logit test examined dichotomous outcomes using the “estimates” (or odds) results when the analysis was performed by SAS software. The odds are a test to determine the probability (α ≤ 0.05) that a certain event will occur and can be compared between treatments.
22
Data for relative frequency of occurrence of B. bassiana s.l. and M. anisopliae s.l. was considered to be number of locations where each species was found (Table 2). Relative frequency of mortality was determined by the number of waxworms killed at each sample location (Table 3). Results for each temperature were compared for each fungus and habitat type.
Results
Relative Frequency of Occurrence
No waxworm larvae in control soils from forest or urban habitats were infected with B. bassiana s.l. or M. anisopliae s.l. at either incubation temperature. Fungi were isolated from 86 of the 90 (95.56%, p= <0.0001, F value=104.61) locations sampled (Figures 22-30). Significant differences (α ≤
0.05) were found in the overall relative frequency of occurrence of the two fungi
(B. bassiana s.l. and M. anisopliae s.l.) (Figure 31) as well as in the interaction between temperature and fungi (Table 4). The probability of finding fungi was
76.58% in forest soil (p=0.0193), 67.74% (p=0.1075) in 20 yr urban soil and
48.72% (p=0.8711) in 10 yr urban soil (Figure 32). Fungi were 3.6 times
(p=0.0447) more likely to be recovered from forest soil than 10 yr urban soil and
1.6 times (p=0.4429) more likely from forest soil than from 20 yr urban soil. Fungi were 2.2 times (p=0.1659) more likely to be recovered from 20 yr urban soil than
10 year urban soil.
Metarhizium anisopliae s.l. was found more frequently (p=<0.0001) than
B. bassiana s.l. Beauveria bassiana s.l. was isolated from only 36.67% sample locations while M. anisopliae s.l. was isolated from 95.56% of sample locations.
23
Beauveria bassiana s.l. was predominately recovered from forest soils while M. anisopliae s.l. was recovered from 100% of forest and 20 yr urban sample locations but only in 26 out of 30 (86.67%) of the 10 yr urban sample locations.
Beauveria bassiana s.l. infections of baits occurred 2.7 times more often in the forest compared to urban habitats; 4.0 times more often in forest than 10 yr urban (p=0.0181); and 2.9 times more often in forest than 20 yr urban
(p=0.0517). Beauveria bassiana s.l. was found only 1.4 times more often in 20 yr urban than 10 yr urban (p=0.5679). Statistically significant differences were not found between habitats with M. anisopliae s.l. infections of waxworm bait.
However, M. anisopliae s.l. occurred 3.0 times (p=0.1482) more often in forests than 10 yr urban, 1.2 times (p=0.8466) more often in 20 yr urban habitats than forests and 3.5 times (p=0.1039) more often in 20 yr urban than in 10 yr urban habitats (Figure 33, 34).
Relative Frequency of Mortality
None of the waxworm larvae exposed to control soils from forest or urban habitats were infected with B. bassiana s.l. or M. anisopliae s.l. at either incubation temperature and were not included in the analysis. A total of 29.76%
(1048 out of 3522) of waxworm larvae exposed to the treatment soils exhibited B. bassiana s.l. or M. anisopliae s.l. fungal infections (Table 3). Significant differences (α ≤ 0.05) were found in the relative frequency of mortality of the two fungi between temperatures (Figure 35). There was also a significant interaction between temperature and fungi (Table 4). Overall significant differences (α =
0.10) were found for habitat type (Figure 36) and interaction between habitat type
24 and fungi (Figure 37, Table 4). Infections of baits were 2.7 times (p=0.0476) more likely to occur from forest soil than 10 year urban soil, 2.1 times (p=0.1152) more likely from forest soil than 20 year urban soil and 1.3 times (p=0.5940) more likely from 20 year urban soil than 10 year urban soil (Figures 37 and 42).
The most frequent fungus causing mortality of G. mellonella larvae was M. anisopliae s.l. (49.74% of the total waxworms exhibited infection) (Figure 38).
Only 6.13% of waxworms exhibited B. bassiana s.l. colonization. Differences in
M. anisopliae s.l. rates of infection were not found between habitats with data from temperatures combined. Differences in B. bassiana s.l. colonization rates were significant when comparing the forest soil to both 10 year and 20 year soils with combined data 15oC and 25oC incubation temperatures: waxworms in forest soil were 3.9 times (p=0.0432) more likely to exhibit B. bassiana s.l. infection than 20 year urban soil and 4.3 times (p=0.0335) more than 10 year urban soil.
Waxworms exhibited B. bassiana s.l. colonization at the same rate for 10 and 20 year soils (p=0.9145).
Incubation Temperature Effects
Fungi were found when soils were incubated at 15oC and 25oC from soil of each habitat (Figure 39, Table 3). Significant temperature effects on relative frequency of occurrence and frequency of mortality were found when comparing fungal colonization of waxworm baits (p=0.0075, F value=9.06, Figure 40). Fungi were found more often in forest soils (p=0.0193) at both temperatures but found equally as often in 10 year (p=0.8711) and 20 year (p=0.1075) soils. At 15oC, fungi were isolated from 59.44% of locations sampled. Beauveria bassiana s.l.
25 was isolated from 32.22% and M. anisopliae s.l. from 86.67% of locations sampled. At 25oC, fungi were isolated from 50.56% of locations sampled.
Beauveria bassiana s.l. was isolated from 7.78% and M. anisopliae s.l. was isolated from 91.11% of locations sampled.
More waxworms were killed by M. anisopliae s.l. than B. bassiana s.l. and differences between incubation temperatures and differences between habitat types were found. No differences were found in fungal frequency of Metarhizium anisopliae s.l. across all habitats and temperatures (Figure 39). Significant differences were found in frequency of mortality of waxworms exposed to
Metarhizium anisopliae s.l. across habitats and temperatures (Figure 40).
Beauveria bassiana s.l. was found most frequently in forest soil at 15oC.
Beauveria bassiana s.l. was rarely found when baits were incubated at 25oC in all habitat types, or in 10 or 20 year urban habitats at 15oC. Metarhizium anisopliae s.l. was had the highest numbers of waxworm mortality at 25oC in forest soil (p=0.0031). Metarhizium anisopliae s.l. frequency of mortality in forest and 20 year urban habitats at 25oC were similar. Metarhizium anisopliae s.l. frequency of mortality in forest and 20 year urban habitats at 15oC were also statistically similar. The 10 year urban M. anisopliae s.l. frequency of mortality numbers decrease at both temperatures, although at 15oC the difference was not significant. Beauveria bassiana s.l. in forest soil had the highest waxworm mortality at 15oC. There was a decrease in total waxworms killed in 10 year urban soils at both temperatures, with a slight, although not significant, increase in 20 year urban soils. (Figures 34 and 41).
26
Discussion
When determining the relative frequency of occurrence of B. bassiana s.l., habitat and incubation temperature should be considered. The analyses of relative frequency of occurrence of fungi showed in forest and urban habitats no significant difference in temperature and habitat type for M. anisopliae s.l. recovery. However, B. bassiana s.l. was found more frequently in forested habitats than urban habitats when baits were incubated at cooler temperatures.
Botch (2013) showed that forest soils are relatively cool and dry while urban soils are typically warm and moist in comparison. While forest fauna relies on precipitation for soil moisture, urban habitats are often irrigated (Oke et al.
1989). Trees trap heat in their canopy (Stewart and Thom 1973) and intercept precipitation (Rutter et al. 1975) leading to cooler, drier soils in areas of dense tree populations. Suburban habitats include buildings that not only reflect heat, but also add heat to the surrounding due to the climate controls within (Oke et al.
1989).
Entomopathogenic fungi in different habitats adapt to soil conditions and arthropod hosts found there. Rangel et al. (2006) found that in conditions of nutritional stress and non-preferred carbon sources, M. anisopliae tolerance to
UV-B increased and conidial yield decrease while culturing on preferred carbon sources decreased UV-B tolerance and increased conidial yield. Rangel et al.
(2008) demonstrated also that stress, such as osmotic, oxidative, UV-A, heat, and nutritive stressors, can influence germination, adhesion and virulence of M. anisopliae conidia. Temperature tolerance influences the growth of
27 enotomopathogenic fungi as Beauveria generally grows in cooler conditions
(Fernandes et al. 2008).
According to weather records from the National Weather Service, the mean temperature for April and May (R. flavipes swarming season) in Columbia,
Missouri for the years 2001-2011 was 15.7oC. The mean temperature for August
(Reticulitermes hageni swarming season) in Columbia, Missouri for the years
2001-2011 was 24.8oC. As B. bassiana s.l. infections of baits were more common at 15°C, infection of soil-dwelling arthropods including termite founding pairs may be more likely to occur in cooler soils and seasons such as spring and autumn. However, M. anisopliae s.l. infections of baits may be almost as likely in both warm and cool soils, meaning spring, summer and autumn (Figure 41).
The relative frequency of mortality of waxworms exposed to B. bassiana s.l. was highest in forest soil, decreasing significantly in 10 yr urban habitats with a slight increase in 20 yr urban habitats. The relative frequency of mortality of waxworms exposed to M. anisopliae s.l. decreased after urbanization (sometime before 10 yrs) but recovered after 20 years. At 15oC, M. anisopliae s.l. relative frequency of mortality of waxworms from 20 yr urban habitats was higher than the relative frequency of mortality in forest habitats. At 25oC, the relative frequency of mortality of waxworms exposed to M. anisopliae s.l. from forest habitats and 20 year urban habitats were similar (Figure 41).
Bidochka et al. (2001) performed a study on M. anisopliae comparing isolates from agricultural and forested habitats. They concluded that M. anisopliae from agricultural habitats exhibited heat-tolerance and stronger UV
28 light resistance compared to forested M. anisopliae isolates and that these abiotic conditions may select for a genetic isolate group that can survive in that habitat (Bidochka et al. 2001). When trees and other herbaceous material are removed from the habitat as a result of anthropomorphic changes associated with urbanization, abiotic environmental changes occur, such as increased light exposure resulting in higher UV radiation exposure and increased soil temperatures.
Beauveria bassiana s.l. infections of baits drop significantly in frequency of occurrence and relative frequency of mortality, in 10 yr urban soils and do not recover quickly, although the number of infections of baits is higher in the 20 yr urban soils. Beauveria bassiana s.l. prefers cooler soil and may not adjust to warmer conditions in 10 yr urban soils where UV radiation and temperature are higher. In contrast, M. anisopliae s.l. prefers warmer temperatures and although the frequency of infections of baits dropped significantly in 10 yr urban habitats, the period of time immediately after urbanization begins, M. anisopliae s.l. frequency returns in 20 yr urban habitats and has a similar number of infections of baits as the forested habitat. At 15oC incubation temperatures, M. anisopliae s.l. infections of baits increased in 20 yr urban habitats in contrast to forested habitat. Theoretically, as urban habitats age, successional increase of fauna could lead to an increase of arthropod host availability which could increase fungal frequency and arthropod mortality. Virulence levels of entomopathogenic fungi can be altered by serial passaging on various host sources (Vandenberg and Cantone 2004).
29
Pinzon and Houseman (2009) found differences in frequencies of termite species in forest versus urban habitats with R. flavipes dominate in urban landscapes (82.4%) and R. hageni (56.4%) most common in the forest. Botch
(2013) confirmed higher frequencies of R. flavipes in urban habitats. He found that subterranean termite colonies were absent in the transitionary period when forests were cleared and houses built. Using microsatellites to assess the frequency of specific allele groupings, Botch (2013) determined that while forests had simple and extended subterranean termite families, infestations in young urban habitats were primarily started by spring-swarming, R. flavipes incipient founding pairs (simple families).
Spring can be a precarious time to swarm, especially in the forest. Cool springtime temperatures are optimal for B. bassiana s.l. and the forest had the highest frequency of entomopathogenic fungi. Ten year old urban habitats had the lowest frequency of fungi and the fungi found had lower levels of frequency of mortality. Interactions between R. flavipes and entomopathogenic fungi may explain why R. flavipes is less successful in forests and more successful in invading young urban habitats. Different abiotic conditions in urban areas
(Bidochka et al. 2001), as well as possibly fewer host species for fungi to passage through to maintain virulence (Fargues and Robert 1983), and possibly less competition explain the greater potential for successful establishment of R. flavipes founding pairs in young urban habitats.
30
Chapter Three
VIRULENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE)
Introduction
Beauveria bassiana (Balsamo) Vuillemin and Metarhizium anisopliae
(Metschnikoff) Sorokin (Ascomycota: Hypocraeles) are cosmopolitan entomopathogenic fungi that reside in the soil and infect soil dwelling arthropods, including subterranean termites (Figure 6). These fungi have been exploited for their biological control properties (Roy et al. 2006) and are unique among entomopathogens because of their mode of entry. These fungi invade their host directly through the cuticle rather than through the lining of the alimentary tract
(Charnley and Collins 2007). Often, the site of invasion is through the spiracles, at intersegmental folds, or between mouthparts (Hajek and St. Leger 1994,
Clarkson and Charnley 1996, Ortiz-Urquiza and Keyhani 2013). Additionally,
Beauveria and Metarhizium can affect all life stages of their hosts and have both a parasitic phase that kills the host and a saprotrophic phase that grows on the cadaver (Charnley and Collins 2007).
Beauveria and Metarhizium occur in habitats of all kinds including meadows, orchards, forests, organic and conventional agricultural lands, temperate regions, field margins, pastures, and chaparral areas (Vänninen et al.
1989, Vänninen 1996, Chandler et al. 1997, Klingen et al. 2002, Keller et al.
2003, Meyling and Eilenberg 2006, Meyling et al. 2009, Rodrigues et al. 2005,
Sánchez-Peña et al. 2011, Stranne 2014). Beauveria, a white sporulating
31 muscardine fungus, is found more frequently in forests and cooler temperatures
(Rodrigues et al. 2005, Sanchez-Pena et al. 2011). Metarhizium, a green sporulating muscardine fungus, is found in many soil types and at both cool and warm temperatures (Vänninen 1996, Meyling and Eilenberg 2007, Sooker er al.
2008).
When forests are cleared to make room for urbanization, the landscape is drastically altered (McIntyre et al. 2001). The top layer of soil, as well as flora and fauna are removed, eliminating some species (McIntyre 2000, Turner et al. 2001,
McKinney 2002, Botch 2013). The existing flora and fauna are altered to varying degrees, depending on the characteristics of the organisms (Friese et al. 1997).
As sensitive, yet resilient organisms, fungi are among those affected by these anthropogenic disturbances as discussed in the previous chapter (Chapter Two).
Fungi are able to reinvade disturbed areas, but the availability is dependent on sources of fungal inocula (Friese et al. 1997). Differences in the relative frequency of occurrence of B. bassiana and M. anisopliae in forested and previously forested urban habitats were found in this study. Beauveria was more likely to be found in the forest while Metarhizium was found equally in the forest and urban areas. Differences in the relative frequency of mortality of Galleria mellonella (Linnaeus) baits in forested and previously forested urban habitats were also found in this study. More baits were infected by forest Beauveria than urban Beauveria. Baits were as likely to be infected by Metarhizium from the forest as Metarhizium from 20 year urban habitats, but were less likely to be infected by Metarhizium from 10 year urban habitats.
32
In Missouri, there are four species of termites, all of which are subterranean termites in the genus Reticulitermes (Pinzon and Houseman 2009).
In a comparative survey of Reticulitermes subterranean termite communities in undeveloped forests and urban habitats of Missouri, Pinzon and Houseman
(2009) found that different termite species are more abundant in undeveloped forest habitats compared to urban habitats. Reticulitermes hageni (Banks) were more abundant in undeveloped forested habitats (56.4% of samples) and
Reticulitermes flavipes (Kollar) were more abundant in urban habitats (82.4% of samples) (Pinzon and Houseman 2009). Reticulitermes virginicus and R. tibialis were also found in Missouri, but in smaller frequencies (Pinzon and Houseman
2009).
Botch (2013) found that suburban development changed the landscape, eliminated the termite fauna, and allowed for different species to become dominant. Incipient R. flavipes colonies established in new subdivisions at a much higher rate than other species of termites (Botch 2013). Factors associated with shifts in dominant subterranean termite species as habitats are converted from forests to suburban may be related in part to biotic factors including the presence of pathogens such as entomopathogenic fungi (Calleri et al. 2005).
Botch (2013) also found temperature differences in forested and suburban habitats. Forest soils were cooler and drier compared to the warmer and moister urban soils. Seasonal temperature variation may also contribute to the interactions between fungal entomopathogens and termite hosts. Reticulitermes flavipes forages more actively during cool, wet seasons (Houseman et al. 2001)
33 and swarms in the spring when temperatures are cooler (Weesner 1970). This is in contrast to R. hageni that forages more actively during warm, dry seasons
(Houseman et al. 2001) and swarms in late summer or early fall when temperatures are warmer (Weesner 1970, Botch 2013).
As eusocial organisms, termites engage in many behaviors that were originally thought to increase the likelihood of fungal epizootic episodes if entomopathogenic spores were introduced into the colony. The classic biological control hypothesis was that if a few spore-exposed termites returned to their colony and engaged in social behaviors, such as allogrooming and trophallaxis, spores would spread to other colony members (Rath 1995, Chouvenc and Su
2010). It was also hypothesized that the entomopathogenic fungus would self- perpetuate in the warm, moist homeostatic nest conditions with numerous resources and eventually wipe out the entire colony (Rath 1995). Rath (1995) reported impressive results from lab studies of 100% death in a colony with only
5-10% initial termite exposure. A Metarhizium based biological control product, called Bio-Blast, was soon developed for commercial use (Rath 1995). However, in the field, this product failed, termite colonies were not eliminated and the biological control of termites remained elusive (Chouvenc et al. 2011b).
As it turns out, subterranean termites are well-equipped for the pathogen- laden environment they live in (Rosengaus et al. 2011). While originally thought to enhance the virulence of a fungal attack, termite social behaviors actually decrease susceptibility to entomopathogenic fungi. Termites groom each other
(Rosengaus et al. 1999b, Raina et al. 2003, Matsuura 2011), line their nests with
34 fungistatic fecal material (Rosengaus et al. 1998a, Chouvenc et al. 2013), possess antimicrobial glandular secretions (Rosengaus et al. 2004), possess fungistatic saliva (Matsuura 2011), inhibit the viability of spores by consuming during allogrooming (Chouvenc et al. 2009b, Rosengaus et al. 2011, Yanagawa et al. 2011a), have pathogen alarm behaviors to warn nestmates of the presence of lethal spores (Rosengaus et al. 1999a), and engage in sanitary behaviors such as burial and cannibalism of infected nestmates (Babayan and Schneider 2012).
Additionally, Rosengaus et al. (2014) recently discovered that the symbiotic microorganisms living within the termites may also enhance the termites’ ability to metabolize the mycelium of an attacking fungus.
Individually, termites have physiological mechanisms to counter entomopathogenic fungi, including cellular and humoral immune responses
(Rosengaus et al. 2007, Calleri et al. 2007, Chouvenc et al. 2009a). Termites also have the ability to sense fungal volatiles (Hussain et al. 2010b, Yanagawa et al. 2011b, Mburu et al. 2013). Large colony size also increases the likelihood of survival when exposed to entomopathogenic fungi (Rosengaus et al. 1998b).
Most biological control research for termites has been done using mature stages of colony development with fungal exposure to the worker caste.
However, it is important to remember that when termite colonies reproduce by synchronized mating events called swarming (Weesner 1960, Lepage and
Darlington 2000), winged individuals leave the protection of the nest en mass
(Shellman-Reeve 1997). These alates are exposed to environmental factors after flying away from the nest and landing on the pathogen-laden ground. After they
35 shed their wings, they seek a mate and a location to enter the soil and begin an incipient colony (Weesner 1960, Noirot 1989). A copularium is formed and eggs are laid which are tended first by these primary reproductives and later by the developing workers (Weesner 1970). It is during this time that the founding pair of primary reproductives and their incipient colony are exposed to entomopathogenic fungi and have the greatest risk of being eliminated.
Fungi have likely shaped the habitat preferences of subterranean termites.
Understanding the effects of entomopathogenic fungi on mortality and sublethal effects on subterranean termite founding pairs may be important. This study will examine the effects of entomopathogenic fungus on the founding pairs of R. flavipes. Meyling and Eilenberg (2007) emphasize that studies of fungal interactions ought to be carried out using fungal isolates collected at a local level and from various habitats in the ecosystem. Since these isolates represent genotypes that local host populations interact with, important ecological information may be discovered (Meyling and Eilenberg 2007). Local Beauveria and Metarhizium isolates from forested and previously forested suburban habitats were collected and used in this study to learn how local subterranean termite founding pairs respond to the fungi they encounter when they leave their parent colony and engage in their colony founding behaviors.
This study tests the virulence and sublethal effects of four fungi, including isolates of Beauveria and Metarhizium from forest and urban soils on the founding pairs of R. flavipes. The specific objectives of the study are: 1) to assess the frequency of mortality (virulence) of R. flavipes when exposed to
36
Beauveria and Metarhizium spores from forest and urban habitats; 2) to assess sublethal effects of R. flavipes when exposed to Beauveria and Metarhizium spores from forest and urban habitats, measured by days until eggs laid, number of eggs in first clutch size, and mean clutch size.
Methods
Collection and swarm handling of subterranean termites
Subterranean termite infested logs were collected from the forest near
Capen Park in Columbia, Missouri during late spring (April 2014). Infested logs were more than 10 m apart and likely to be different colonies (Vargo and
Hussender 2009). Subterranean termites were morphologically identified as
Reticulitermes flavipes using keys by Scheffrahn and Su (1994). Some of the termites in the infested logs possessed wing buds, indicating they were ready to molt to the winged alate stage. Infested logs were cut into lengths less than 76 cm to fit into swarming chambers in the laboratory at room temperature and humidity (Figure 42). Log segments were placed into swarming chambers created using 121 L (32 gal) trash bins. Air flow was controlled by creating openings in the lid that were covered by mesh and then placing a removable 7 x
13 cm flaps over the opening. Humidity inside the chambers was elevated using a clean moistened towel (Clorox Handi Wipes) in a dish [Mainstays 1.1 L (4.5 cups) food storage set] with the sides open and top and bottom closed, on the bottom of the swarming chamber.
Swarm chambers were observed daily for swarming termites. Three colonies of R. flavipes swarmed nearly a month later (May 7, May 13, and May
37
15, 2014), and were placed into large 15 cm x 20 mm petri dishes with moistened filter papers. One filter paper disk moistened with 5 ml dH2O adhering to the lid and the other filter paper was left dry and placed on the bottom of the petri dish.
Founding pairs consisting of one male and one female were created after swarmers were collected and their wings were shed. Termite gender was identified behaviorally. Female R. flavipes flag their abdomen until antennated by males. In contrast, males moved about quickly, searching for other termites to follow. When they contact another termite, they follow it (called tandem running) by placing their antennae on the abdomen of the individual in front of them and keep continuous contact as they follow (DeHeer and Vargo 2006). One male and one female R. flavipes were randomly collected by capturing a female and a male that were engaged in tandem running and placing the pair in a single 35 mm x 15 mm holding cell for 24 hours before being exposed to fungal spores
(Figure 43 A).
Fungal culture and spore solution
Fungal spores used in this experiment were from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU)
(ARSEF collection numbers: BHF 12456, BHU 12457, MHF 12452, MHU 12453,
Figure 44, Table 5) at a concentration of approximately 1 x 107. Beauveria bassiana s.l. and M. anisopliae s.l. fungi were identified morphologically using a dissecting microscope at 30-40x magnification based on the morphology of spores and mycelia. Beauveria bassiana s.l. morphological identifications were
38 based on Hoog (1972) and Metarhizium anisopliae s.l. morphological identifications were based on Hoog et al. (2000).
Fungi were passaged through Galleria mellonella (Linnaeus) (Lepidoptera:
Pyralidae) larvae to maximize the fungal virulence (Fargues and Roberts 1983).
Passaging is performed by exposing insects to a fungus and harvesting the resulting conidia after insect death. Surface sterilized G. mellonella were exposed to forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium
(MF) and urban Metarhizium (MU) spores and died (Figure 44). Sporulation occurred on waxworm surface and the spores were purified.
To purify each fungus, a small number of spores were removed from the cadaver and placed on individual 100 mm x 15mm Potato Dextrose Agar (PDA)
(Oxoid, England) plates. Fungi were allowed to grow under sterile conditions and mycelium from the edge of the mycelial mat was taken and placed in vanTieghem cells + water agar to remove any impurities. VanTieghem cells are 1 cm pieces of sterile silicon tubing placed in the center of a water agar plate (100 mm x 15mm) (Figure 45). As mycelia extend beyond the vanTieghem cell, it grows at a different rate than other organisms and can be purified. Clean mycelia may then be extracted and plated on PDA where the fungus grew and sporulated.
A flame-sterilized glass rod was used to dislodge spores of each isolate.
The hydrophobic conidia were put into solution with a 0.01% aqueous solution of
Tween 80 (Sigma Aldrich Co., USA). The spore solution was aspirated with a pipette, placed into a sterile glass test tube with enough 0.01% Tween 80
39 aqueous solution to create a concentration of approximately 1x107 spores per ml for each fungus. Spore concentration was determined using a hemocytometer
(Improved Neubauer Bright Line, Hausser Scientific) at 400x magnification.
Spore solutions were kept at 4oC and were used in the experiment within 7 days.
Application of Treatment
Twenty R. flavipes imago pairs from the same colony were exposed to each of four treatments [exposure to forest Beauveria (BF), urban Beauveria
(BU), forest Metarhizium (MF) and urban Metarhizium (MU) spores] and a treatment of no spores. This was repeated with three colonies. Individual pairs were placed into one of the chambers of 6-well plates (3.5 cm internal diameter) temporarily prior to exposure to fungal spores. Each chamber contained moistened filter paper (Fisher Scientific, Q5) on the bottom.
A second, 6-well plate was prepared and filter paper in each cell was inoculated with 100 µl of a randomly selected spore solution from one of the fungal varieties. Founding pairs were moved from the 6-well plate where they were being held temporarily into the plate containing treated filter paper and were exposed for 30 minutes before being placed into a third, clean 6-well plate containing clean moistened filter paper (0.5 ml dH2O) (Figure 43). Controls were exposed to 0.01% aqueous solution of Tween 80 without spores. Plates containing founding pairs were placed into an environmental chamber at 15oC and ~80% ± 3% RH, 24h dark. Weather records from the National Weather
Service show the mean temperature for April and May (R. flavipes swarming season) in Columbia, Missouri for the years 2001-2011 was 15.7oC.
40
Termite mortality for each individual within the pair, time to lay eggs and maximum number of eggs laid during the study were recorded every 3-4 days for
60 days. The filter paper in the chambers was moistened when needed to confirm that mortality was due to fungal infection. When dead termites were discovered, they were removed, surface sterilized in 1% sodium hypochlorite
(NaClO) for three minutes, rinsed in sterile dH2O and dried on sterile filter paper
(Fisher Scientific, Q5). Cadavers were placed individually into 1.5 ml Eppendorf tubes containing Potato Dextrose Agar (PDA). The plastic top of the Eppendorf tube was removed and replaced with Parafilm allowing air flow and verification of emerging, sporulating fungus. (Figure 46).
Statistical Analysis
Data were analyzed using PROC GENMOD using link logit and the distribution was binomial using SAS® software (version 9.3) (α ≤ 0.05). A split plot design was used with the main plot being treatment (each fungi and control), then the subplot was colony and all possible interactions were tested. Individual colony responses to treatments should be considered when analyzing the results but are taken into account in this study by this SAS procedure.
A PROC LIFETEST and Wilcoxon rank statistic were calculated using
SAS software (α ≤ 0.05). Lifetesting is a nonparametric survival analysis method measuring the time to an event. Time to an event is typically not normally distributed and this model can handle censoring without modification (SAS software support: lifetest). Days until death for each member of the termite pair
(d1=days until the death of the first termite; d2=days until death of the second
41 termite); days until first eggs (fe=first egg) and maximum number of eggs laid in the 60 day period were recorded (me=max egg). We looked at the overall T50 for death which is the time it takes for 50% of the population to die, and the T50 for first eggs which is the time it takes for 50% of the population to lay eggs.
Termites that survived the entire 60 days were censored, meaning they were removed, from the death data.
Results
Virulence
Reticulitermes flavipes engaged in natural behaviors in this study, including allogrooming, trophallactic feeding, mating, egg laying and egg grooming. Founding pairs were significantly affected (α ≤ 0.05) when exposed to fungal spores compared to the controls (p=<0.0001, X2=87.11). Significant differences (α ≤ 0.05) were found in mortality between the first to die and the second of the pair to die, as well as sublethal effects. There were also significant differences in the time until first egg and how many eggs were laid.
Ninety-nine out of 120 (82.5%) control founding pairs survived the 60 day evaluation period. Only 21.7 % of the first termite in the pairs not exposed to spores died compared to 88.3% of the first individuals of a pair exposed to
Beauveria spores and 73.3% of the first individuals of a pair exposed to
Metarhizium spores. In contrast, only 134 out of 480 (27.9%) fungal exposed founding pairs survived. A total of 73.8% of exposed pairs had at least one individual die and 68.3% of exposed pairs had both individuals die (Table 6).
Overall, founding pairs exposed to fungal spores were significantly more likely to
42 die than controls (p=<0.0001). This also held true when Beauveria and
Metarhizium were statistically separated (p=<0.0001). Overall, higher mortality occurred among founding pairs exposed to forest fungi compared to urban fungi, especially among the first termite of a pair (p=0.0160) (Table 7). Beauveria was more virulent than Metarhizium in causing mortality towards the first and second individuals of a pair (d1: p=0.0160, d2: p=0.0009). Forest Beauveria was more virulent than urban Beauveria against the first of the termite pair (d1: p= 0.0415: d2: p=0.1869) (Table 8). Forest and urban Metarhizium virulence were not significantly different (d1: p=0.4593; d2: p=0.1233). (Figure 47)
The first individual in founding pairs exposed to forest Beauveria was 2.8 times more likely to die than those exposed to urban Beauveria (p=0.0415) or forest Metarhizium (p=0.0415), and 5.1 times more likely than urban Metarhizium
(p=0.0008). Mortality of the first individual in founding pairs (d1) exposed to urban
Beauveria was not significantly different from the mortality of the first individual in founding pairs exposed to forest Metarhizium (p=1.0000). Mortality of the first individual in founding pairs exposed to urban Beauveria was not significantly different from the mortality of the first individual in founding pairs exposed to urban Metarhizium (p=0.1233). Mortality of the first individual in founding pairs exposed to forest Metarhizium was not significantly different from the mortality of the first individual in founding pairs exposed to urban Metarhizium (p=0.1233).
The second individual in founding pairs exposed to forest Beauveria was
3.1 times more likely to die than the second individual in founding pairs exposed to forest Metarhizium (p=0.0094) and 4.1 times more likely to die than the second
43 individual in founding pairs exposed to urban Metarhizium (p=0.0011). Mortality of the second individual in founding pairs exposed to forest Beauveria was not significantly different from the mortality of the second individual in founding pairs exposed to urban Beauveria (p=0.1869). (Table 8 B)
The T50 varied between treatments. Because 50% of the control population did not die, the results for comparison to controls is not precise, but still highly significant. The T50 for the first dead founder exposed to forest
Beauveria or urban Beauveria was 28 days. The T50 for the first dead founder exposed to forest Metarhizium was 31 days and 39 days for urban Metarhizium.
The T50 for the second dead founder to appear was 28 days for both forest
Beauveria and urban Beauveria. The T50 for the second dead founder exposed to forest Metarhizium was 39 days and urban Metarhizium was 46 days. (Table 9,
Figure 48)
Colony variations were also examined. In colony one, the first individual in founding pairs died sooner than the first individual in founding pairs in colonies two and three (p=<0.0001). In colony one, the second individual in founding pairs died sooner than the second individual in founding pairs in colony two
(p=0.0302). Figure 49 shows individual colony differences for time to death for individuals in founding pairs (d1 and d2).
Sublethal effects: First egg
Slightly more than half (50.41%) of female founders exposed to spores laid eggs. A significant difference (α ≤ 0.05) was found between female founders exposed to fungal treatments and control founding females in the number of days
44 until the first eggs were laid (p=<0.0001). Female founders in the control laid eggs on average 10 days earlier than female founders exposed to fungi.
However, these differences were not significant (α ≤ 0.05).
The time until first eggs laid by unexposed females in controls (T50) was
28 days. On average, female founders exposed to forest Metarhizium laid eggs by 34.5 days, urban Metarhizium by 38 days, urban Beauveria by 38.5 days.
Female founders exposed to forest Beauveria did not have a T50 because they died before 50% of the population could lay eggs, but had a minimum period of
39 days for those that did lay eggs (Table 10). Contrasting treatments show that exposure to forest Beauveria significantly delayed egg laying (p=0.0025). All other contrasts were not statistically significant (Figure 50). Fertility of eggs laid was zero as no eggs were observed to hatch during the entire period of the study.
Sublethal effects: Maximum egg clutch size
The mean number of eggs laid by unexposed female founders was 4.1.
The mean number of eggs laid by exposed female founders was 2.3 for forest
Beauveria, 3.0 for urban Beauveria, 3.4 for forest Metarhizium, and 3.6 for urban
Metarhizium. All female founders exposed to fungi laid fewer eggs than unexposed founders. Female founders exposed to Beauveria spores laid the fewest eggs. Reticulitermes flavipes female founders that were exposed to either fungus from forest soils laid fewer eggs than those exposed to fungi from urban soils. Female founders exposed to forest Beauveria laid significantly fewer eggs than females who were not exposed (p=<0.0001). Female founders exposed to
45 forest Beauveria also laid significantly fewer eggs than females who were exposed to forest Metarhizium (p=0.0174) and urban Metarhizium (p=0.0011).
Female founders exposed to urban Beauveria (p=0.0048) and forest Metarhizium
(p=0.0254) also laid significantly fewer eggs than females who were not exposed.
The mean number of eggs laid by female founders exposed to urban
Metarhizium and females who were not exposed were not statistically different
(p=0.1804).
Discussion
The entomopathogenic fungi found in the soils of Columbia, Missouri, are virulent towards founding pairs of subterranean termites in the genus
Reticulitermes. As such colonies swarm and founding pairs establish incipient colonies, they must deal with the threat of pathogens found in the soil environment of both forest and urban habitats.
While Beauveria is less common in the forest and in urban habitats than
Metarhizium, its effectiveness as a pathogen is greater than Metarhizium.
Beauveria from both habitats killed founding pairs of R. flavipes in laboratory bioassays faster than Metarhizium in this study. This may be due to the interaction between grooming and spore size. Fungal spores get lodged in intersegmental areas and are hard to remove via grooming. Swarming is a dangerous time and often damage can happen to the swarmers. Antennae are important in grooming and, if damaged, can hinder the ability of termites to successfully remove spores from each other, thereby reducing the chance of survival in the presence of pathogens (Yanagawa et al. 2009).
46
Beauveria spores are much smaller than those of Metarhizium (Figures 51 and 52) and are likely more difficult for founding pairs to groom off each other thus allowing for spore germination, cuticle penetration and toxin production.
Spore size differences also exist between forest and urban Metarhizium and the smaller size of these spores from the forest soils may explain why they were more virulent than the Metarhizium with larger spores from urban soils in this study.
Sublethal effects, in this study, showed that exposure to Beauveria increased the time until females laid eggs. Calleri et al. (2007) also found
Zootermopsis angusticollis Hagen females had significant delays in oviposition when exposed to pathogen attacks during early colony foundation periods.
Trade-offs between immunity and reproduction may occur during this critical phase in colony foundation (Calleri et al. 2007). There was also an overall decrease in the number of eggs laid when female founders were exposed to
Beauveria. Most female founders exposed to forest Beauveria died before laying any eggs. In contrast, female R. flavipes exposed to urban Metarhizium laid the same number of eggs on average as unexposed females.
Mated pairs were observed to engage in social behaviors such as allogrooming and trophallaxis, however founding pairs do not have the same protections as established colonies. A copularium must be created in the pathogen-laden wood-soil interface. The copularium is created with a protective coating of fungistatic fecal material. The founding pair cares for each other and the eggs without the assistance of offspring so if one individual dies, the colony
47 will not survive. In contrast, in a large colony, kings and queens can be replaced when they die. When the colony population is only two, these individuals are not replaced. Calleri et al. (2006) found that incipient pairs of Zootermopsis angusticollis invest more in disease resistance than brood production, showing the importance of pathogen management in colony establishment.
Beauveria and Metarhizium are both found commonly in forest soils.
However, Beauveria is rarely present in urban soils and urban Metarhizium is the least virulent. Survival of R. flavipes founding pairs may be enhanced by escaping Beauveria in forest soils when they land in urban soils. This Beauveria escape hypothesis predicts that R. flavipes escapes the pressure of the more virulent forest fungi and becomes more successful at establishing in urban soils that contain less virulent fungi.
Additional testing may find other sublethal effects, such as egg size and hatching rate in fungal exposed R. flavipes (Matsuura and Kobayashi 2007).
Fungal volatile studies could examine R. flavipes founding pairs’ response to fungal volatiles from other soils. Field studies also need to be considered as conditions in the laboratory differ from the natural conditions founding pairs experience when swarming and building copularia.
Understanding how urbanization enhances the survival of some
Reticulitermes species and leads to successful establishment of new colonies may provide novel strategies for control. Learning how to utilize the more virulent forest fungi in urban soils may lead to strategies that disrupt successful establishment by founding pairs of R. flavipes.
48
Chapter Four
INFLUENCE OF ENTOMOPATHOGENIC FUNGI FROM FOREST AND URBAN HABITATS ON NEST SITE SELECTION BY FOUNDING PAIRS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE)
Introduction
The cryptic nature of the subterranean termites in the genus
Reticulitermes (Rhinotermitidae) has made it difficult to define the biology behind specific behaviors (Thorne et al. 1999, Lainé and Wright 2003). The reproductive behaviors of Reticulitermes flavipes (Kollar) were previously assumed to be primarily due budding (Shellman-Reeve 1997, Thorne 1998, Myles 1999, Thorne et al. 1999), in which a colony develops as a result of becoming isolated from their home colony (Vargo and Husseneder 2009). Yet, recent research indicates that many R. flavipes colonies are started by founding pairs of imagoes (Bulmer et al. 2001, Vargo 2003, DeHeer and Vargo 2004, Vargo and Carlson 2006,
Botch 2013).
Termite reproductive imagoes are winged and fully sclerotized with compound eyes. They are produced at certain times of the year after a colony reaches maturity, approximately two years after colony foundation (Lainé and
Wright 2003). Emergence holes or tubes are created by the worker caste
(Emerson 1938, Nutting 1969) and winged imagoes gather near the holes to disperse during annual synchronized mating flight called swarming (Weesner
1960, Lepage and Darlington 2000, Bordereau and Pasteels 2011). Raised emergence holes may facilitate imago take off or reduce the risk of attack by
49 surface predators (Nutting 1969). Soldiers guard the exits while imagoes emerge and exits are then plugged by workers (Emerson 1938, Nutting 1969). Soldiers may also drive off or attack dealated imagoes that try to reenter the nest (Nutting
1969). The timing of alate flights is dependent on internal and climatic conditions, often involving temperature, precipitation events and humidity (Nutting 1969,
Lainé and Wright 2003).
After leaving the nest, imagoes shed their wings and seek a mate (Nutting
1969, Shellman-Reeve 1997). It is believed that females release a sex pheromone with abdomen raised in a flagged position to call males to her position (Bordereau and Pasteels 2011). Males locate the females and antennate the posterior pleural membranes of the female and the pair follow each other in tandem, with the female in the lead and the male maintaining constant antennal contact behind, to select a nest site (Shellman-Reeve 1997, Bordereau and
Pasteels 2011). If contact is broken, the female stops and resumes the calling position until contact is reinstated (Nutting 1969, Bordereau and Pasteels 2011).
Most founding pairs do not survive due to predation, climatic conditions, and competition for mates or nest site selection (Nutting 1969, Scheffrahn et al.
1998).
General termite nesting preferences include cracks, crevices, holes abandoned by wood-boring beetles, and other small protected locations from which to begin further excavations (Nutting 1969, Schefferahn et al. 1998).
Emerson (1938) noted that imagoes do not leave nest site selection up to chance. They determine whether a nest site is suitable or not by visually orienting
50 toward trees or other wooden objects detecting volatiles from freshly exposed wood or fermenting sap, or honing in on the condition of wood and the associated volatiles from decomposing fungi (Nutting 1969). de Lima et al. (2006) examined substrate type preference for Coptotermes gestroi (Wasmann)
(Rhinotermitidae), another soil-nesting wood-feeding termite, and found these termites can distinguish differences among substrate types and continue their search if a nest site is not suitable. Shellman-Reeve (1994) reported that founding pairs of Zootermopsis nevadensis (Hagen) (Termopsidae) dispersed to logs with higher quantities of total cambium-nitrogen compared to logs with increased amounts of wood xylem. Schefferahn et al. (1998) found the site preferences of Cryptotermes brevis (Walker) (Kalotermitidae) to be inside wood block crevices. For Reticulitermes, Weesner (1965) commented that most species in this genus are capable of establishing in any workable substrate that has appropriate moisture and temperature conditions.
Site selection may also include risk avoidance through detection of volatiles from pathogenic substances (Mburu et al. 2013). Termites are known to detect the presence of entomopathogenic fungi, exhibit alarm behavior and avoid pathogens such as Beauveria bassiana (Balsamo) Vuillemin and Metarhizium anisopliae (Metschnikoff) Sorokin (Ascomycota: Hypocreales) (Rosengaus et al.
1999a, Hussain et al. 2010a, Chouvenc et al. 2011b, Yanagawa et al. 2012).
Volatiles from Beauveria and Metarhizium were shown to cause Co. formosanus
Shiraki workers to position themselves further from the more virulent fungi and closer to less virulent fungi (Hussain et al. 2010b). This same study also showed
51
Co. formosanus workers responded more strongly to Metarhizium isolates than
Beauveria isolates (Hussain et al. 2010b). Mburu et al. (2013) examined the response of Macrotermes michaelseni (Sjöstedt) workers to isolates of Beauveria and Metarhizium and found this termite was able to detect virulence levels with high sensitivity. Yanagawa et al. (2012) found that the spores of highly virulent pathogens attached more strongly to the surface of Co. formosanus cuticle and that their volatiles were more strongly avoided than less virulent fungal isolates of the same species.
In addition to humoral and cellular immunity (Rosengaus et al. 2007,
Chouvenc et al. 2009a), termites have behavioral defenses against pathogens.
Many termites line their nests with fungistatic fecal material (Rosengaus et al.
1998a), groom each another and their eggs with fungistatic saliva (Matsuura et al. 2007, Hamilton et al. 2011), and benefit from symbiotic microorganisms that assist in metabolizing the fungal mycelium (Rosengaus et al. 2014).
Allogrooming, the removal of pathogens by nestmates, is important in established colonies and has been shown to be critical to the success of founding pairs when establishing a new colony (Rosengaus et al. 1999b, Raina et al. 2003, Matsuura 2011). Grooming removes spores and inhibits future germination after passing through the alimentary tract (Shimizu and Yamaji 2003,
Chouvenc et al. 2009b).
When a forest is cleared to make room for a new urban landscape, the existing flora and fauna are altered, including the elimination of termite colonies
(Botch 2013). The relative frequency of occurrence and virulence of
52 entomopathogenic fungi in the soil is also influenced. Forest habitats may contain more virulent fungi found in greater frequency than new urban habitats because of the existing flora and fauna established in the habitat. As previously reported,
Beauveria from forest soils was most virulent and also produced strongest sublethal effects. Reticulitermes flavipes founding pairs died fastest when exposed to fungi from forest soils, fewer females laid eggs, and of those that did laid fewer eggs than females in any other treatments (Chapter Three).
Botch (2013) concluded that new urban habitats are initially invaded by incipient R. flavipes colonies that begin from founding pairs. This may occur because after forests are cleared and subdivisions built, soil fungal frequency and virulence may be lower and therefore more favorable for establishments of subterranean termite colonies.
This study examined the behavioral response of R. flavipes founding pairs when allowed to choose between nest sites containing Beauveria or Metarhizium spores from forest and urban habitats and sites containing no spores. The specific objectives of the study are: 1) to determine the frequency of mortality
(virulence) of founding pairs when given the choice of exposure to Beauveria or
Metarhizium spores from forest and urban habitats and sites containing no spores and 2) to determine R. flavipes founding pair behavior when given a choice to enter sites containing Beauveria or Metarhizium spores from forest and urban habitats, whether the founding pairs enter spore-containing chambers, avoid spore-containing chambers, or are indifferent. It was anticipated that
53 founding pairs would avoid nest sites with fungal spores, and also avoid forest soil fungal spores more than urban soil fungal spores.
Methods
Collection and swarm handling of subterranean termites
Subterranean termite infested logs were collected from the forest near
Capen Park in Columbia, Missouri, during late spring (April 2014). Infested logs were more than 10 m apart and likely to be different colonies (Vargo and
Hussender 2009). Subterranean termites were morphologically identified as
Reticulitermes flavipes using keys by Scheffrahn and Su (1994). Some of the termites in the infested logs possessed wing buds, indicating they were ready to molt to the winged alate stage. Infested logs were cut into lengths less than 76 cm to fit into swarming chambers in the laboratory at room temperature and humidity (Figure 42). Log segments were placed into swarming chambers created using 121 L (32 gal) trash bins. Air flow was controlled by creating openings in the lid that were covered by mesh and then placing a removable 7 x
13 cm flaps over the opening. Humidity inside the chambers was elevated using a clean moistened towel (Clorox Handi Wipes) in a dish [Mainstays 1.1L (4.5 cup) food storage set] with the sides open and top and bottom closed, on the bottom of the swarming chamber.
Swarm chambers were observed daily for swarming termites. Three colonies of R. flavipes swarmed nearly a month later (May 7, May 13, and May
15, 2014), and were placed into large 15 cm x 20 mm petri dishes with moistened
54 filter papers. One filter paper disk moistened with 5 ml dH2O adhered to the lid and the other filter paper was left dry and placed on the bottom of the petri dish.
Founding pairs consisting of one male and one female were created after swarmers were collected and their wings were shed. Termite gender was identified behaviorally. Female R. flavipes flag their abdomen until antennated by males. In contrast, males moved about quickly, searching for other termites to follow. When they contact another termite, they follow it (called tandem running) by placing their antennae on the abdomen of the individual in front of them and keep continuous contact as they follow (DeHeer and Vargo 2006). One male and one female R. flavipes were randomly collected by capturing a female and a male that were engaged in tandem running and placing the pair in a single 35 mm x 15 mm holding cell for 24 hours before being exposed to fungal spores
(Figure 43 A).
Fungal culture and spore solution
Fungal spores used in this experiment were from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU) (ARSEF collection numbers: BHF 12456, MHF 12452, MHU 12453, Figure 44, Table 5) at a concentration of approximately 1 x 107. Beauveria bassiana s.l. and M. anisopliae s.l. fungi were identified morphologically using a dissecting microscope at 30-40x magnification based on the morphology of spores and mycelia. Beauveria bassiana s.l. morphological identifications were based on
Hoog (1972) and Metarhizium anisopliae s.l. morphological identifications were based on Hoog et al. (2000).
55
Fungi were passaged through Galleria mellonella (Linnaeus) (Lepidoptera:
Pyralidae) larvae to maximize the fungal virulence (Fargues and Robert1983).
Passaging is performed by exposing insects to a fungus and harvesting the resulting conidia after insect death. Surface sterilized G. mellonella were exposed to forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium
(MF) and urban Metarhizium (MU) spores and died. Sporulation occurred on waxworm surface and the spores were purified.
To purify each fungus, a small number of spores were removed from the cadaver and placed on individual 100 mm x 15mm Potato Dextrose Agar (PDA)
(Oxoid, England) plates. Fungi were allowed to grow under sterile conditions and mycelium from the edge of the mycelial mat was taken and placed in vanTieghem cells + water agar to remove any impurities. VanTieghem cells are
½ cm pieces of sterile silicon tubing placed in the center of a water agar plate
(100 mm x 15mm) (Figure 45). As mycelia extend beyond the vanTieghem cell, it grows at a different rate than other organisms and can be purified. Clean mycelia may then be extracted and plated on PDA where the fungus grew and sporulated.
A flame-sterilized glass rod was used to dislodge spores of each isolate.
The hydrophobic conidia were put into solution with a 0.01% aqueous solution of
Tween 80 (Sigma Aldrich Co., USA). The spore solution was aspirated with a pipette, placed into a sterile glass test tube with enough 0.01% Tween 80 aqueous solution to create a concentration of approximately 1x107 spores per ml for each fungus. Spore concentration was determined using a hemocytometer
56
(Improved Neubauer Bright Line, Hausser Scientific) at 400x magnification.
Spore solutions were kept at 4oC and were used in the experiment within 7 days.
Material Preparation
Four adjacent chambers of a single 6-well plate and the interstitial area between these chambers (Fisher Scientific Cat. No. 12-556-004, 3.5 cm Internal diameter) were used as nest chambers containing treatments. Two-millimeter diameter holes were opened between the interstitial area into each of the four chambers so founding pairs could move freely between all four chambers. Holes were created by heating a small metal rod using a propane blow torch and gently pressing the hot metal through the plastic wall between the interstitial area and each chamber. The hole was created as close to the level of the floor of the interstitial area as possible (Figure 53 A).
Filter paper circles (3.5 cm, Q5, Fisher Scientific Cat. No. 09-801BB) were placed in the bottom of each of the chambers. Two fungal treated chambers and two control chambers were randomized using a random sequence generator
(www.random.org/sequences/) (Figure 53). Combinations include forest
Beauveria and control (no spores), forest Metarhizium and control (no spores), and urban Metarhizium and control (no spores). Filter papers were inoculated with 100 ml of approximately 1x107 spore solution in 0.01% aqueous solution of
Tween 80 (Sigma Aldrich Co., USA) or 0.01% aqueous solution of Tween 80 without spores for the controls.
57
Application of Treatment
A founding pair of R. flavipes termites from one of three different colonies was randomly chosen and placed into the interstitial area between the four nest chambers. The pairs were allowed to move about unrestricted between the nest chambers and interstitial area. They were checked every 3-4 days and sterile dH2O was added as needed to maintain favorable moisture levels. Founding pairs remained in the arena during the entire study period of 45 days. As the average spring temperature when R. flavipes typically swarms is 15oC, the 6-well plates containing founding pairs were placed in a 15oC environmental chamber
(~80% +/- 3% RH, 24h dark). The location of each termite (in control chamber, fungal treated chamber or interstitial area) and deaths of either termite in the pair were noted at each check. Presence/absence of eggs was also noted at the end of the 45 day period.
Statistical Analysis
Behavior. Results were analyzed using PROC GLIMMIX using link logit with a binomial distribution using SAS® software (version 9.3) (α ≤ 0.05). The main design was a randomized complete block (by colony) in which the main plot contained colony and fungal treatment (forest Beauveria, urban Metarhizium, forest Metarhizium)*chamber (control, fungus, or interstitial). Of thirteen observation periods, some observations in period 6-13 had no termites in particular chambers and were given an observation count of zero which a logit analysis does not perform. Therefore, the behavioral analysis was performed for observation periods 1-5. Data for interstitial chamber were combined with control
58 data to examine the position of the termites relative to fungal treated chambers.
Colony with treatment and chamber represented the denominator of F for the main plot effects of treatment, chamber, and treatment*chamber interactions.
The residual means squared was the denominator of F for subplot effects.
Statistical tests were carried out using SAS software (α ≤ 0.05).
Virulence. Time to the death of the first and second individuals of founding pairs were analyzed using PROC GENMOD using link logit with a binomial distribution using SAS software (α ≤ 0.05). A split plot design was used with the main plot being treatment (forest Beauveria, forest Metarhizium, and urban Metarhizium) and all possible interactions.
Virulence was also analyzed using PROC LIFETEST and the Wilcoxon rank statistic using SAS software (α ≤ 0.05). Lifetesting is a nonparametric survival analysis method measuring the time to an event. This method was used because time to an event is typically not normally distributed and this model can handle censoring without modification (SAS software support: lifetest). Days until death for each member of the founding pair (overall days until termite death; d1=days until the death of the first individual of founding pairs; d2=days until death of the second individual of founding pairs) were used as events. We looked at the overall T50 for death (time in which it takes for 50% of the population to die). A split plot design was used with the main being the treatment (forest
Beauveria, forest Metarhizium, and urban Metarhizium), then by colony and all possible interactions. Termites that survived the entire 45 days were censored from the death data.
59
Virulence was also analyzed using PROC GLIMMIX using link logit with a binomial distribution using SAS software (α ≤ 0.05). The main design was a randomized complete block (by termite colony) with repeated measurements.
The colony was the block. We did not have treatment as the factorial (no control, fungi, or interstitial subplots) and only had overall treatment. Some blocks of total and cumulative death results for specific observation periods were zero. Since a logit analysis does not perform analyses with zeros, observation periods were trimmed as follows: total death observation periods 2-12 and cumulative death observation periods 2-13. The colony within treatment was the denominator of F for treatment. The subplot contained time and interactions of treatment x time.
The residual means squared was the denominator of F for subplot effects.
Results
Behavior results
Reticulitermes flavipes were found in spore treated and chambers without spores more often than interstitial areas (forest Beauveria: p=<0.0001; forest
Metarhizium: p=0.0020; urban Metarhizium: p=0.0044). Founding pairs initially entered chambers without spores more frequently than with forest Beauveria.
Later, pairs moved to chambers with fungal spores more frequently (observation
1: p=0.3096, 2: p=0.0946, 3: p=0.2062, 4: p=0.8475, 5: p=0.0707) (Figure 54
BF). Most R. flavipes pairs visited multiple chambers while few pairs remained in the same chamber each. Reticulitermes flavipes avoided fungal chambers with forest Metarhizium and urban Metarhizium, choosing the chamber without spores significantly more often (observation forest Metarhizium 1: p=0.0056, forest
60
Metarhizium 3: p=0.0034, forest Metarhizium 4: p=0.0075, urban Metarhizium 4: p=0.0426, urban Metarhizium5: p=0.0183) (Figure 54 B and C).
There were no differences in preference of founding pairs between chambers containing forest Metarhizium and urban Metarhizium. However, founding pairs did not avoid Beauveria fungal chambers (Table 11). Founding pairs were found in forest Beauveria chambers more often than forest
Metarhizium fungal chambers on the third observation (p=0.0240) (Figure 54).
Founding pairs were also found in forest Beauveria more frequently than forest
Metarhizium (p=0.0450) and urban Metarhizium (p=0.0013) on the fifth observation.
Virulence results
After Reticulitermes flavipes were introduced to interstitial areas, they were observed to enter and exit chambers without difficulty. Copularia were made under the filter paper in 11.7% (17 out of 145) of the chambers. Egg laying occurred with 17.9% (26 of 145) of female founders.
Most (92.6%) founding pairs died in this study. Of the urban Metarhizium exposed founding pairs, 95.6% died, 82.2% of forest Metarhizium exposed founding pairs died, and 77.8% of the founding pairs exposed to forest Beauveria died. Out of 145 founding pairs, 93.3% had at least one mate die and 91.9% had both individuals die. Overall, R. flavipes were significantly more likely to die when exposed to the urban fungus (urban Metarhizium) than the forest fungi (forest
Metarhizium + forest Beauveria) (p=0.0369) (Table 12, Figure 55). They were more likely to die when exposed to urban Metarhizium than forest Metarhizium
61
(p=0.0704), and more likely to die when exposed to Metarhizium spores (urban
Metarhizium + forest Metarhizium) than Beauveria spores (forest Beauveria)
(p=0.0625) (Table 12, Figure 55). Results were similar when days until the death of the first individual of founding pairs (d1) and days until death of the second individual of founding pairs (d2) were analyzed individually, although results were not statistically significant. (Table 12, Figure 56, 57).
The time until half of the population (T50) of founding pairs died varied between treatments of spore exposure. When both individuals in each pair were included (overall T50), founding pairs exposed to urban Metarhizium died faster
(17 days) than founding pairs exposed to Beauveria (26 days) or forest
Metarhizium (24 days) (p=<0.0001) (Table 13 A, Figure 58). This pattern was the same when days until the death of the first individual (d1) and days until death of the second individual (d2) of founding pairs were analyzed individually (Table 13
B and C, Figure 59).
Curves showing total death at each observation showed peaks and valleys of founder mortality. Founding pairs exposed to forest Beauveria had lower mortality at the beginning which increased between the fourth and fifth observations (p=0.0473). Mortality increased among founding pairs exposed to forest Metarhizium between observations 5 and 6 (p=0.0153), with the highest mortality of founding pairs at observation eight (p=0.0482). Urban Metarhizium mortality was the highest at the fifth observation. Founding pairs in forest
Metarhizium had fewer deaths at the beginning, increased significantly between observations five and six (p=0.0336), peaked with the highest deaths at the
62 eighth observation (p=0.0473). Founding pairs in urban Metarhizium died faster in the beginning with a significant increase between observation three and four
(p=0.0038) and a decrease between observations five and six (p=0.0078) (Figure
60).
On average, founding pairs from colony 3 died faster (18 days) than founders from colonies 1 (22 days) and 2 (28 days) (Table 13 D, Figure 61).
Discussion
The ephemeral nature of swarming and cryptic behavior of subterranean termites make studying nest site selection of founding pairs difficult. However, as this stage of colony development could be a target for termite colony control, especially in new urban habitats, more attention to this ephemeral stage is needed. Understanding the effects of entomopathogenic fungi, such as
Metarhizium and Beauveria, on nest site selection and establishment behavior of
R. flavipes founding pairs is one such area needing attention.
When allowed to move about between chambers, R. flavipes founding pairs died soonest when exposed to spores from urban Metarhizium (MU).
Founding pairs had a choice to enter a chamber with spores, or not, and were not forced to remain exposed for a time period before moving to a clean chamber as was the case in other studies. Spores were constantly available during the entire length of the bioassay, similar to natural conditions in the environment.
However, unlike natural conditions, founding pairs did not always stay together and were sometimes found in different chambers. This could affect the amount of time devoted to defensive social behaviors, such as allogrooming.
63
In previous studies, R. flavipes imago behavior is affected by the presence of entomopathogenic fungi, although in different degrees of significance
(Rosengaus et al. 1999a, Hussain et al. 2010a, 2010b, Yanagawa et al. 2011a,
2012). In Metarhizium spore treatments, R. flavipes spent more time avoiding the chambers containing fungal spores, choosing instead to remain in non-fungal chambers or interstitial areas. This finding agrees with studies that show recognition of Metarhizium spores and avoidance behaviors in relation to volatiles and virulence levels (Hussain et al. 2010a, 2010b). However,
Reticulitermes founding pairs in this study did not avoid the chambers with
Beauveria spores. This finding may support studies that showed less aggressive behavior between nestmates when treated with Beauveria volatile substances compared to nestmates treated with Metarhizium volatile substances (Yanagawa et al. 2011a).
As spores were constantly available for exposure throughout the length of the observation period, it is possible that spore volatile signatures changed over time and could affect the choices made by founding pairs. This could explain the increased movement toward forest Beauveria treatment chambers between observations three and four (Figure 54 A). However, R. flavipes avoided
Metarhizium the entire length of the study.
Fungal volatile profiles can be influenced by the media on which the fungus is cultured and whether the fungus has passed through the host or not
(Hussain et al. 2010a). Virulence levels of fungi also changed depending on whether Beauveria or Metarhizium were cultured on PDA or insect-passaged
64 cultures grown on PDA (Hussain et al. 2010a). Insect-passaged fungi produced conidia with high virulence while conidia from a PDA culture also have high virulence but had a different volatile profile (Hussain et al. 2010a). Culture media can also influence germination time, UV sensitivity, cold activity and thermotolerance of pathogenic fungi (Rangel et al. 2006, 2008, Fernandes et al.
2008).
Termite workers respond differently to nestmates when spores are only attached to their cuticle, as compared to when the fungus has invaded the cuticle and the termite is infected (Yanagawa et al. 2011b). When volatile odors of
Metarhizium were present, Coptotermes formosanus Shiraki increased nestmate grooming and aggressive behavior but there were no changes in cannibalism and burial behavior (Yanagawa et al. 2011b). Yanagawa et al. (2011b) suggested that cannibalism and burial behavior could be induced by volatile signals initiated after the infection of nestmates.
Fungi were passaged through Galleria mellonella (Linnaeus) (Lepidoptera:
Pyralidae) larvae to maximize the fungal virulence (Fargues and Robert 1983).
While there may be differences in R. flavipes response if the fungi were passaged through R. flavipes in this study, the spores founding pairs encounter in natural environments would not necessarily have been passaged through R. flavipes termites, but more likely would have passaged through other surface dwelling insects.
Although the founding pairs in Metarhizium treatments avoided the chambers with spores in the first weeks of the study, they died sooner than their
65 counterparts in Beauveria treatments that entered chambers with spores equally as often as chambers without spores. It is possible founding pairs were entering
Metarhizium fungal chambers at times other than when they were observed and therefore being exposed.
In previous work, R. flavipes died faster when exposed to Beauveria compared to R. flavipes exposed to Metarhizium. Additionally, R. flavipes were more likely to die when exposed to forest fungi versus urban fungi (forest
Beauveria and forest Metarhizium). However, in the no-choice bioassay, founding pairs were confined to a single recovery chamber where they engaged in nesting and colony foundation behaviors such as allogrooming, trophallaxis, constructing a copularium, mating, laying eggs, and egg grooming. Founding pairs exposed to spores in no-choice chambers were constantly together and
27.9% of founders exposed to fungal spores survived, compared to only 7.4% surviving where they had to choose between multiple chambers. In situations when founding pairs could move individually between chambers, they were sometimes separated from one another. Fewer pairs settled down to construct copularia and even fewer pairs laid eggs. The ability to wander and become separated with the inherent consequences of less time spent in social behaviors such as allogrooming, increased the risk of mortality.
Allogrooming is essential to the success of colony foundation (Matsuura et al. 2002). Metarhizium germinates faster than Beauveria (chapter five) and has more direct appresorial pressure with which to penetrate the insect cuticle
(Hussain and Tain 2013). Rapid germination and hyphal growth rates may give
66 fungal isolates an advantage for successful infection if not removed quickly by a nestmate (Hajek and St. Leger 1994, Lui et al. 2003). The size of the spores may also play a role in grooming success and potentially affect survivorship. Spore size has been correlated with increased virulence (Smith 1977, Altre et al. 1999), but spore size has not been studied in regard to grooming efficiency. Urban
Metarhizium has larger spores than forest Metarhizium or Beauveria from either forest or urban habitats and thus may be easier to remove from cuticular folds.
Founding pairs exposed to urban Metarhizium in nesting site choice assays had the longest T50 (39 days) compared to Beauveria T50 (28 days) in both choice and no-choice virulence assays. Founding pairs in no-choice virulence assay may have achieved greater survival by more easily removing larger spores. Reticulitermes flavipes exposed to Beauveria may have experienced similar mortality rates in choice and no-choice assays because of the difficulty in removing small spores. Grooming efficiency in relation to spore size has not been studied but appears to be important.
Other factors could be involved in nest site selection, such as substrate color, substance texture, tree species, etc. An interesting study of the pharaoh ant by Pontieri et al. (2014) found that when selecting a nest site, Monomorium pharaonis (Linnaeus) prefer infected over uninfected nest sites. The authors hypothesize that the ants possibly do this to immunize the colony against pathogens there (Pontieri et al. 2014). Another reason could be familiarity with volatile signatures from the nest in which founding pairs emerged. Calleri et al.
(2007) found that immunity of Zootermopsis angusticollis varied during periods of
67 colony foundation: pairing, copulation and oocycte maturation, and oviposition.
The greatest level of immunity occurred in early stages but declined for females later on as they shifted from internal resources to reproduction (Calleri et al.
2007). If founding pairs haven’t settled quickly after pairing, physiological defenses may decrease and thus increase the risk of mortality, especially if founding pairs are separated while exposed to pathogenic fungi.
The results of this assay, compared to the no-choice virulence assays, show the need for additional research with the importance of allogrooming, especially field studies because of the many variables inherent in the founding and reproductive process of R. flavipes.
68
Chapter Five
FUNGISTATIC EFFECTS OF RETICULITERMES FLAVIPES (RHINOTERMITIDAE) FOUNDING PAIR EXOCRINE GLANDS ON ENTOMOPATHOGENIC FUNGAL GERMINATION
Introduction
Subterranean termites live in microbial-rich environments that are favorable to potential pathogens (Rosengaus et al. 1998b, Chouvenc et al.
2009b). While some of the pathogens sharing the termite habitat are known to be lethal, such as the soil residing entomopathogenic fungi Beauveria bassiana
(Balsamo) Vuillemin and Metarhizium anisopliae (Metschnikoff) Sorokin
(Ascomycota: Hypocreales) (Zoberi 1995, Rath 2000, Wright et al. 2005,
Yanagawa et al. 2008, 2009), termites continue to thrive in soil environments
(Rosengaus et al. 1998a, Chouvenc et al. 2011a). Their ability to succeed despite the pathogen threat lies not only in individual immune responses, but also behavioral and biochemical defenses, as well as immunity on a colony level
(Rosengaus et al. 1998b, 1999a, Calleri et al. 2007, Chouvenc et al. 2009b,
Yanagawa et al. 2011a).
In addition to avoiding pathogenic substances, an essential defensive behavior of this eusocial insect is allogrooming, or the grooming of nestmates, which removes pathogens, including fungal spores (Rosengaus et al. 1998b,
Shimizu and Yamaji 2003). While grooming, glandular secretions containing β-
1,3-glucanases and termicins (Siderhurst et al. 2005b, Rosengaus et al. 2014) are left on the cuticle of nestmates protecting against pathogens (Rosengaus et al. 1998a, Matsuura et al. 2007, Chouvenc et al. 2009b). Termicins are antifungal
69 peptides (Lamberty et al. 2001) secreted by the salivary glands and β-1,3- glucanases may originate from gut symbionts (Siderhurst et al. 2005a,
Rosengaus et al. 2014) also known for their role in cellulose digestion (Cleveland
1923, Zhou et al. 2007, Brune 2014). β-1,3-glucanases and termicins may synergistically weaken fungal cell walls so cell membranes are damaged, thus protecting the termites before the fungus can even penetrate the cuticle (Bulmer et al. 2010, Hamilton and Bulmer 2012).
Spores are consumed in the process of allogrooming. Germination of ingested spores is inhibited, not only inside the termite gut (Chouvenc et al.
2009b), but fecal matter also continues to inhibit unconsumed spores in the nest
(Rosengaus et al. 1998a). Fungistatic effects have also been shown in sternal glands of Zootermopsis angusticollis (Hagen) (Rosengaus et al. 2004) and the frontal gland in Nasutitermes (Rosengaus et al. 2000). Little is known of the functioning of frontal and mandibular glands (Noirot 1969, Miura et al. 1999,
Chouvenc et al. 2009b) and Chouvenc et al. (2009b) postulates the possibility of a synergistic effect between the mandibular or labial glands and salivary glands, aiding in the inhibition of fungal spores in the gut.
Noirot (1969) distinguishes three main adaptations of external secretions from exocrine glands: nutrition, defense, and chemical signals. The number of exocrine glands in termites is debated, from 17 (Gonçalves et al. 2010) to 19 distinct known glands (Křížková et al. 2014). However, there are five principle exocrine glands that are considered: frontal, mandibular, salivary, sternal and tergal glands (Costa-Leonardo and Haifig 2010). Not all glands are found in all
70 termite species and not all glands are found in each caste (Costa-Leonardo and
Haifig 2010) but the imago caste has the highest diversity of exocrine glands
(Leis and Sbrenna 1983, Křížková et al. 2014, Šobotník et al. 2010). Frontal, mandibular, salivary and sternal glands are found in Reticulitermes imagoes while tergal glands are not (Costa-Leonardo and Haifig 2010) (Figure 62). The salivary and sternal glands have been studied, but little is known of the functioning of the frontal and mandibular glands (Noirot 1969, Miura et al. 1999,
Chouvenc et al. 2009b). Additionally, the fungistatic nature of the distal region of the abdomen is not known.
Reticulitermes imago heads contain the frontal and mandibular glands
(Noirot 1969, Costa-Leonardo and Haifig 2010). The frontal gland is located behind the brain and the exterior opening as a simple rounded pore at the top of the head (Noirot 1969, Šobotník et al. 2010). Chemical defensive compounds have been found in Prorhinotermes (Rhinotermitidae). The amount peaks at the time of swarming and stops when soldier offspring appear (Šobotník et al. 2010).
The nitrocompounds and sesquiterpenes found in the glandular secretion are thought to be an irritant and may make imagoes unpalatable during their dispersal flight, thus reducing the risk of predation (Šobotník et al. 2010).
Mandibular glands are small, paired ovoid structures located at the base of the mandible between the maxillae and the mandible (Costa-Leonardo and
Haifig 2010, Noirot 1969). The large glandular cells contain intracellular reservoirs from which the glandular secretions emerge (Noirot 1969). The function of this gland is not well established (Noirot 1969, Costa-Leonardo and
71
Haifig 2010) but may be related to defense or caste regulation (Costa-Leonardo and Haifig 2010).
While sternal gland secretions have been associated with trail pheromones and fungistatic properties (Rosengaus et al. 2004), little information is available regarding organs in the distal region of the abdomen. The rectal gland is not listed by Křížková et al. (2014) as a distinct exocrine gland and
Snodgrass (1935) notes the rectal gland is not demonstrated to be a true gland.
Rectal glands, also known as rectal papillae and rectal pads, are composed of up to six longitudinal thickenings on the inner wall of the rectum, forming a chamber as each papilla terminates before reaching the anus (Snodgrass 1935, Noirot and Noirot-Timothee 1969). Although the function of the rectal gland is not known
(Snodgrass 1935), Wigglesworth (1932) postulated the organ’s function is to reabsorb precious water from fecal matter passing through the alimentary tract before it is deposited outside the body. However, fecal matter from
Reticulitermes is fluid and therefore water reabsorption may not be the main function of this organ.
This study aims to elucidate the fungistatic effects of extracts from the head and organs in the distal region of the abdomen of the eastern subterranean termite, Reticulitermes flavipes (Kollar) (Figure 62), previously unexposed to
Metarhizium or Beauveria and without the aid of synergistic microorganisms. The specific objective of the study is to determine if there is a change in germination of Beauveria or Metarhizium spores when exposed to R. flavipes head and abdomen extracts.
72
Methods
Collection and swarm handling of subterranean termites
Subterranean termite infested logs were collected from forests near
Capen Park in Columbia, Missouri during late spring (April 2014). Infested logs were more than 10 m apart and likely to be different colonies (Vargo and
Hussender 2009). Subterranean termites were morphologically identified as
Reticulitermes flavipes using keys by Scheffrahn and Su (1994). Some of the termites in infested logs possessed wing buds, indicating they were ready to molt to the winged alate stage. Infested logs were cut into lengths less than 76 cm to fit into swarming chambers in the laboratory at room temperature and humidity
(Figure 42). Log segments were placed into swarming chambers created using
121 L (32 gal) trash bins. Air flow was controlled by creating openings in the lid that were covered by mesh and then placing a removable 7 x 13 cm flaps over the opening. Humidity inside the chambers was elevated using a clean moistened towel (Clorox Handi Wipes) in a dish [Mainstays 1.1 L (4.5 cups) food storage set] with the sides open and top and bottom closed, on the bottom of the swarming chamber.
Swarm chambers were observed daily for swarming termites. Three colonies of R. flavipes swarmed nearly a month later (May 7, May 13, and May
15, 2014), and were placed into large 15 cm x 20 mm petri dishes with moistened filter papers. One filter paper disk moistened with 5 ml dH2O adhering to the lid and the other filter paper was left dry and placed on the bottom of the petri dish.
Founding pairs consisting of one male and one female were created after swarmers were collected and their wings were shed. Termite gender was
73 identified behaviorally. Female R. flavipes flag their abdomen until antennated by males. In contrast, males moved about quickly, searching for other termites to follow. When they contact another termite, they follow it (called tandem running) by placing their antennae on the abdomen of the individual in front of them and keep continuous contact as they follow (DeHeer and Vargo 2006). One male and one female R. flavipes were randomly collected by capturing a female and a male that were engaged in tandem running and placing the pair in a single 35 mm x 15 mm holding cell for 24 hours before being exposed to fungal spores
(Figure 43 A).
Fungal culture and spore solution
Fungal spores used to examine fungistatic secretions were Beauveria and
Metarhizium isolated from forest soil (ARSEF collection numbers: BHF 12456 and MHF 12452, Table 5). Beauveria bassiana s.l. and M. anisopliae s.l. fungi were identified morphologically using a dissecting microscope at 30-40x magnification based on the morphology of spores and mycelia. Beauveria bassiana s.l. morphological identifications were based on Hoog (1972) and
Metarhizium anisopliae s.l. morphological identifications were based on Hoog et al. (2000).
Germination rates of these fungi in the presence of R. flavipes cuticular and glandular matter were reported by Rosengaus et al. (2004). At 15oC,
Beauveria germinates at 31 hours and Metarhizium germinates at 29 hours. In pilot studies, higher germination rates of Beauveria spores occurred, but the germination tubes extended at such a rapid rate that the ability to observe the
74 germination of all spores was obscured, therefore an earlier germination time of
31 hours was chosen.
Fungi were passaged through Galleria mellonella (Linnaeus) (Lepidoptera:
Pyralidae) larvae to maximize the fungal virulence (Fargues and Robert 1983).
Passaging is performed by exposing insects to a fungus and harvesting the resulting conidia after insect death. Surface sterilized G. mellonella were exposed to forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium
(MF) and urban Metarhizium (MU) spores and died. Sporulation occurred on waxworm surface and the spores were purified.
To purify each fungus, a small number of spores were removed from the cadaver and placed on individual 100 mm x 15mm Potato Dextrose Agar (PDA)
(Oxoid, England) plates. Fungi were allowed to grow under sterile conditions and mycelium from the edge of the mycelial mat was taken and placed in vanTieghem cells + water agar to remove any impurities. VanTieghem cells are
½ cm pieces of sterile silicon tubing placed in the center of a water agar plate
(100 mm x 15mm) (Figure 45). As mycelia extend beyond the vanTieghem cell, it grows at a different rate than other organisms and can be purified. Clean mycelia may then be extracted and plated on PDA where the fungus grew and sporulated.
A flame-sterilized glass rod was used to dislodge spores of each isolate.
The hydrophobic conidia were put into solution with a 0.01% aqueous solution of
Tween 80 (Sigma Aldrich Co., USA). The spore solution was aspirated with a pipette, placed into a sterile glass test tube with enough 0.01% Tween 80
75 aqueous solution to create a concentration of approximately 1x107 spores per ml for each fungus. Spore concentration was determined using a hemocytometer
(Improved Neubauer Bright Line, Hausser Scientific) at 400x magnification.
Spore solutions were kept at 4oC and were used in the experiment within 7 days.
Single R. flavipes founding pairs from swarm chambers in the lab were isolated in chambers within sterile 6-well plates (Fisher Scientific Cat. No. 12-
556-004, 3.5 cm internal diameter). Chambers contained a filter paper (3.5 cm,
Q5, Fisher Scientific Cat. No. 09-801BB) moistened with dH2O. After 60 days, each surviving pair was euthanized by placing in -80oC freezer to assure the death of any symbiotic microorganisms inside the gut of founding pairs.
Individual R. flavipes founders were sexed morphologically using the characteristics taken from Lainé and Wright (2003): Reticulitermes spp. females have an enlarged seventh sternite that covers the eighth and ninth sternites, whereas the male seventh sternite is similar in size to the sixth sternite. Males also have a pair of styli on the ninth sternite (Figure 63).
Three females and three males were randomly selected from each of three colonies and were surface sterilized in 1% NaClO for three minutes, rinsed in sterile dH2O, and dried on sterile filter paper. Heads were removed at the base using a flame sterilized razor and homogenized in 100 μl sterile 0.01% aqueous solution of Tween 80. This mixture was centrifuged at 1000 RPM for 10 minutes to concentrate the cuticular matter and the supernatant was removed and placed into a sterile 1.5 ml Eppendorf tube and held overnight at 4oC. Under sterile conditions, a solution containing spores of Beauveria or Metarhizium was mixed
76 with R. flavipes head supernatant to achieve 0% (controls) and 10% solution
(10% head to 90% spore solution). Abdomen solutions were created by excising the tip of the abdomen from the seventh sternite and following the same methods to create supernatant. (Figures 64 A and B)
Potato dextrose agar (PDA) plates were inoculated with forty μl of
Beauveria or Metarhizium spore and R. flavipes extract (head or abdomen) solution, aliquoted into four-ten μl drops placed on each quarter of the plate
(Figure 64 C). After applying the solution, the plates were tipped gently allowing the solution droplet to spread. Inoculated plates were held in an environmental chamber at 15oC, 24 hr darkness, and 80 ± 3% R.H.
After the elapsed incubation time (Metarhizium 29 hr, Beauveria 31 hr), digital images (Figure 65) were taken of inoculated plates using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification. This microscope, camera, and software are the property of USDA-ARS Biological Control of Insect
Research Laboratory and the use of this equipment is gratefully acknowledged.
Spore germination percent was assessed by counting the number of spores with germination tubes as a percentage of the total number of spores in each field of view. Ten to twelve fields of view were counted per sample. The number of germinated spores was divided by the total number of spores in each field of view to calculate a mean percent of germinated spores per field of view.
Given the high magnification, depth of field was limited. Only spores in focus that
77 were clear and recognizable were counted. Objects that were out of focus due to being out of the depth of field were not counted.
Statistical Analysis
Data for percent germination were analyzed using PROC GLIMMIX using link logit with a binomial distribution and blocked by colony. A split plot design was used with the main plots being fungus and spore concentration. The subplot was body segment region (head or abdomen) as the body regions were taken from the same experimental unit. This was treated as a repeated measurement in space. The residual means squared was the denominator of F for subplot effects. All statistical tests were carried out using SAS® software (version 9.3) (α
≤ 0.05).
Results
Metarhizium spores germinated 9.29 times more often than Beauveria
(p=<0.0001) (Table 14). Germination rates for Metarhizium spores were similar on control (Abdomen: 90.78%, Head: 89.27%) and treatment plates (Abdomen:
89.18%, p=0.1885, Head: 88.08%, p=0.2526) (Tables 15 and 16). Differences in spore germination were found between abdomen control plates and head treatment plates (p=0.0239). The abdomen control germination was 1.33 times greater than germination on plates treated with head supernatant. (Figures 66,
67)
Beauveria germination results were similar (Abdomen: 8.59%, Head:
10.00%) on control plates. However, Beauveria spore germination rate decreased to 6.94% when exposed to head supernatant (p=0.0004). Abdomen
78 supernatant increased germination to 14.10% (p=0.0003). Beauveria spore germination rate when exposed to abdomen supernatant was 2.17 times greater than the Beauveria spore germination rate (p=<0.0001) when exposed to head supernatant. (Table 17, Figures 68, 69)
Discussion
Beauveria and Metarhizium are important pathogens and have exerted their evolutionary influence on R. flavipes (Rosengaus et al. 1998b). This is shown in the many defenses of R. flavipes in behavior, biochemical responses and immune responses (Rosengaus et al. 1998b, Shimizu and Yamaji 2003,
Chouvenc et al. 2008, Bulmer et al. 2010, Hamilton and Bulmer 2012) and they benefit internally from fungistatic compounds from symbiotic gut microorganisms
(Siderhurst et al. 2005a, Rosengaus et al. 2014). The mandibular and frontal glands may have fungistatic effects in the imago caste and this, in addition to the growing understanding of the fungistatic nature of many termite exocrine glands, shows how R. flavipes is so successful despite living in locations with high pathogen populations.
It was expected that germination rates would decrease in the presence of
R. flavipes head and abdomen extracts. Only Beauveria germination decreased when exposed to R. flavipes head extracts. No significant change was noted between germination rates of Metarhizium spores when exposed to treatments with head or abdomen extracts. Beauveria spore germination unexpectedly increased in the presence of supernatant from the abdomen tip.
79
It appears that Metarhizium spores are not affected by compounds in the head or abdomen of R. flavipes founders. Perhaps R. flavipes utilizes other mechanisms to defend against fungal pathogens. Behavioral mechanisms may be effective against this fungus. Reticulitermes flavipes have been shown to sense Metarhizium volatiles (Hussain et al. 2010b) and avoid the spores.
Allogrooming may also be important for inhibiting Metarhizium spore germination especially as these spores are large in size.
Metarhizium germination rate may have been fast enough to overcome fungistatic effects of R. flavipes head and abdomen extracts in this study.
Alternatively, the spores may not have been susceptible to chemical compounds produced by the glands associated with heads or abdomens, or they may not have been susceptible to the concentration used in this study.
Beauveria spores are much smaller than Metarhizium spores and may be more difficult to remove when allogrooming so chemical defense mechanisms may have evolved. Beauveria spores were more susceptible to chemical compounds produced by mandibular or frontal glands. It is possible that a chemical in glandular extracts enhanced termicins and β-1,3-glucanase in breaking down fungal cell walls.
Beauveria spore germination increased when exposed to abdomen tip extracts. The spores may have germinated faster due to more favorable conditions, such as changes in carbon or nitrogen levels. Alternatively, germination inhibitors, chemicals that prevent germination until conditions are
80 more favorable, could have diminished effects due to fewer overall spores being in the solution (Feofilova et al. 2012).
Beauveria spore germination rate was very low overall (mean= 9.53%).
Future experiments should consider ways to achieve higher germination rates by finding optimal conditions. While many studies use PDA for culturing fungi, it may not be the best agar for Beauveria. An experiment examining Beauveria spore germination rates on various agars (such as recipes found in Kiraly et al. 1970), as well as different agar concentrations, is recommended. This has been done for Metarhizium (Ibrahim et al. 2002, Rangel et al. 2008). Some fungi delay germination when spore concentration is high (Feofilova et al. 2012) since autoinhibitors accumulate in high concentrations of sister spores. Autoinhibitors, or self-inhibitors are self-produced chemical signals produced by spores in response to an overcrowded environment (Allen 1955, Macko and Staples 1973).
This can be overcome by washing the spores with water as many times as necessary to remove the inhibitors (Feofilova et al. 2012). This should be considered in future studies.
There is much to be learned regarding the interactions between entomopathogenic fungi and subterranean termites. Reticulitermes flavipes, particularly founding pairs, have many defenses against pathogenic fungi in their environment and may use various defenses, such as allogrooming, avoidance of pathogens, and humoral and cellular immunity. Reticulitermes flavipes may employ fungal specific defenses such as behavioral defense mechanisms for commonly encountered, larger fungal spores, like Metarhizium, and rely more on
81 biochemical immunity defenses for less frequently encountered and smaller fungal genera such as Beauveria.
82
Literature Cited
Allen, P. J. 1955. The role of a self-inhibitor in the germination of rust uredospores. Phytopathology 45: 259-266.
Altre, J., J. Vandenberg, and F. Cantone. 1999. Pathogenicity of Paecilomyces fumosoroseus Isolates to Diamondback Moth, Plutella xylostella: Correlation with Spore Size, Germination Speed, and Attachment to Cuticle. Journal of Invertebrate Pathology 73: 332-338.
Babayan, S. A., and D. S. Schneider. 2012. Immunity in society: diverse solutions to common problems. PLOS Biology 10: e1001297.
Behie, S., P. Zelisko, and M. Bidochka. 2012. Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science 336: 1576- 1577.
Bidochka, M. J., A. M. Kamp, T. M. Lavender, J. Dekoning, and J. A. De Croos. 2001. Habitat association in two genetic groups of the insect-pathogenic fungus Metarhizium anisopliae: uncovering cryptic species? Applied and Environmental Microbiology 67: 1335-1342.
Bidochka, M. J., and C. L. Small. 2005. Phylogeography of Metarhizium, an insect pathogenic fungus. In: F.E. Vega and M. Blackwell (Eds.) Insect- Fungal Associations: Ecology and Evolution: Ecology and Evolution: 28.
Bidochka, M. J., J. E. Kasperski, and G. A. Wild. 1998. Occurrence of the entomopathogenic fungi Metarhizium anisopliae and Beauveria bassiana in soils from temperate and near-northern habitats. Canadian Journal of Botany 76: 1198-1204.
Bischoff, J. F., S. A. Rehner, and R. A. Humber. 2009. A multilocus phylogeny of the Metarhizium anisopliae lineage. Mycologia 101: 512-530.
Bordereau, C., and J. M. Pasteels. 2011. Pheromones and chemical ecology of dispersal and foraging in termites, pp. 279-320, Biology of Termites: A Modern Synthesis. Springer.
Botch, P. S. 2013. 'Historical Landscape Interactions and Patterns of Invasion By Subterranean Termites (Isoptera: Reticulitermes) In Subdivisions Of Different Ages. Dissertation. University of Missouri.
Breznak, J. A., and A. Brune. 1994. Role of microorganisms in the digestion of lignocellulose by termites. Annual Review of Entomology 39: 453-487.
Brune, A. 2014. Symbiotic digestion of lignocellulose in termite guts. Nature Reviews Microbiology 12: 168-180.
83
Brune, A., and M. Ohkuma. 2011. Role of the termite gut microbiota in symbiotic digestion, pp. 439-475, Biology of termites: a modern synthesis. Springer.
Bulmer, M. S., E. S. Adams, and J. F. Traniello. 2001. Variation in colony structure in the subterranean termite Reticulitermes flavipes. Behavioral Ecology and Sociobiology 49: 236-243.
Bulmer, M., F. Lay, and C. Hamilton. 2010. Adaptive evolution in subterranean termite antifungal peptides. Insect Molecular Biology 19: 669-674.
Calleri II, D. V., R. B. Rosengaus, and J. F. Traniello. 2005. Disease and colony foundation in the dampwood termite Zootermopsis angusticollis: the survival advantage of nestmate pairs. Naturwissenschaften 92: 300-304.
Calleri, D. V., E. M. Reid, R. B. Rosengaus, E. L. Vargo, and J. F. Traniello. 2006. Inbreeding and disease resistance in a social insect: effects of heterozygosity on immunocompetence in the termite Zootermopsis angusticollis. Proceedings of the Royal Society B: Biological Sciences 273: 2633-2640.
Calleri, D. V., R. B. Rosengaus, and J. F. Traniello. 2007. Immunity and reproduction during colony foundation in the dampwood termite, Zootermopsis angusticollis. Physiological Entomology 32: 136-142.
Chandler, D., D. Hay, and A. Reid. 1997. Sampling and occurrence of entomopathogenic fungi and nematodes in UK soils. Applied Soil Ecology 5: 133-141.
Charnley, A., and S. Collins. 2007. 10 Entomopathogenic Fungi and Their Role in Pest Control. Environmental and Microbial Relationships 4: 159.
Chouvenc, T., N.-Y. Su, and M. L. Elliott. 2008. Interaction between the subterranean termite Reticulitermes flavipes (Isoptera: Rhinotermitidae) and the entomopathogenic fungus Metarhizium anisopliae in foraging arenas. Journal of Economic Entomology 101: 885-893.
Chouvenc, T., N.-Y. Su, and A. Robert. 2009a. Cellular encapsulation in the eastern subterranean termite, Reticulitermes flavipes (Isoptera), against infection by the entomopathogenic fungus Metarhizium anisopliae. Journal of Invertebrate Pathology 101: 234-241.
Chouvenc, T., N.-Y. Su, and A. Robert. 2009b. Inhibition of Metarhizium anisopliae in the alimentary tract of the eastern subterranean termite Reticulitermes flavipes. Journal of Invertebrate Pathology 101: 130-136.
84
Chouvenc, T., and N.-Y. Su. 2010. Apparent synergy among defense mechanisms in subterranean termites (Rhinotermitidae) against epizootic events: Limits and potential for biological control. Journal of Economic Entomology 103: 1327-1337.
Chouvenc, T., M. L. Elliott, and N.-Y. Su. 2011a. Rich Microbial Community Associated with the Nest Material of Reticulitermes flavipes (Isoptera: Rhinotermitidae). Florida Entomologist 94: 115-116.
Chouvenc, T., N.-Y. Su, and J. K. Grace. 2011b. Fifty years of attempted biological control of termites–Analysis of a failure. Biological Control 59: 69-82
Chouvenc, T., C. A. Efstathion, M. L. Elliott, and N.-Y. Su. 2013. Extended disease resistance emerging from the faecal nest of a subterranean termite. Proceedings of the Royal Society B: Biological Sciences 280: 20131885.
Clarkson, J. M., and A. K. Charnley. 1996. New insights into the mechanisms of fungal pathogenesis in insects. Trends in Microbiology 4: 197-203.
Cleveland, L. R. 1923. Symbiosis between termites and their intestinal protozoa. Proceedings of the National Academy of Sciences of the United States of America 9: 424.
Costa-Leonardo, A. M., and I. Haifig. 2010. Chapter Twenty-One-Pheromones and Exocrine Glands in Isoptera. Vitamins & Hormones 83: 521-549. de Lima, M. M., T. T. Gonçalves, O. DeSouza, and R. Reis-Jr. 2006. Nesting site selection by Coptotermes gestroi (Insecta: Isoptera). Sociobiology 48.
DeHeer, C. J., and E. L. Vargo. 2004. Colony genetic organization and colony fusion in the termite Reticulitermes flavipes as revealed by foraging patterns over time and space. Molecular Ecology 13: 431-441.
DeHeer, C. J., and E. L. Vargo. 2006. An indirect test of inbreeding depression in the termites Reticulitermes flavipes and Reticulitermes virginicus. Behavioral Ecology and Sociobiology 59: 753-761.
Edwards, R., and A. E. Mill. 1986. 261pp. Termites in buildings. Their biology and control, Rentokil Ltd. East Grinstead, UK.
Eggleton, P., G. Beccaloni, and D. Inward. 2007. Response to Lo et al. Biology Letters 3: 564-565.
Emerson, A. E. 1938. Termite nests--a study of the phylogeny of behavior. Ecological Monographs: 247-284.
85
Engel, M. S., D. A. Grimaldi, and K. Krishna. 2009. Termites (Isoptera): their phylogeny, classification, and rise to ecological dominance. American Museum Novitates: 1-27.
Esperk, T., T. Tammaru, and S. Nylin. 2007. Intraspecific variability in number of larval instars in insects. Journal of Economic Entomology 100: 627-645.
Evans, T. A., B. T. Forschler, and J. K. Grace. 2013. Biology of invasive termites: a worldwide review. Annual Review of Entomology 58: 455-474.
Fargues, J. F., and P. H. Robert. 1983. Effects of passaging through scarabeid hosts on virulence and host specificity of two strains of the entomopathogenic hyphomycete Metarhizium anisopliae. Canadian Journal of Microbiology 29: 576-583.
Feofilova, E., A. Ivashechkin, A. Alekhin, and Y. E. Sergeeva. 2012. Fungal spores: Dormancy, germination, chemical composition, and role in biotechnology (review). Applied Biochemistry and Microbiology 48: 1-11.
Fernandes, E. K., D. E. Rangel, A. M. Moraes, V. R. Bittencourt, and D. W. Roberts. 2008. Cold activity of Beauveria and Metarhizium, and thermotolerance of Beauveria. Journal of Invertebrate Pathology 98: 69- 78.
Ferron, P. 1978. Biological control of insect pests by entomogenous fungi. Annual Review of Entomology 23: 409-442.
Friese, C., S. Morris, and M. Allen. 1997. Disturbance in natural ecosystems: scaling from fungal diversity to ecosystem functioning. The Mycota IV. Environmental and Microbial Relationships. Springer, Berlin: 47-65.
Gonçalves, T. T., O. DeSouza, and J. Billen. 2010. A novel exocrine structure of the bicellular unit type in the thorax of termites. Acta Zoologica 91: 193- 198.
Greenwood, C. M. and E. J. Rebek. 2009. Detection, Conservation, and Augmentation of Naturally Occurring Beneficial Nematodes for Natural Pest Suppression Oklahoma State University Cooperative Extension Service Fact Sheet EPP7670. Retrieved from http://osufacts.okstate.edu
Hajek, A., and R. St. Leger. 1994. Interactions between fungal pathogens and insect hosts. Annual Review of Entomology 39: 293-322.
Hamilton, C., F. Lay, and M. S. Bulmer. 2011. Subterranean termite prophylactic secretions and external antifungal defenses. Journal of Insect Physiology 57: 1259-1266.
86
Hamilton, C., and M. S. Bulmer. 2012. Molecular antifungal defenses in subterranean termites: RNA interference reveals in vivo roles of termicins and GNBPs against a naturally encountered pathogen. Developmental & Comparative Immunology 36: 372-377.
Holt, J. A., and M. Lepage. 2000. Termites and soil properties, pp. 389-407, Termites: evolution, sociality, symbioses, ecology. Springer.
Hoog, G. d. 1972. The genera Beauveria, Isaria, Tritirachium and Acrodontium gen. nov. Studies in Mycology 1: 1-41.
Hoog, G. S. d., J. Guarro, J. Gené, and M. Figueras. 2000. Atlas of clinical fungi, Centraalbureau voor Schimmelcultures (CBS). 1-1126, pg 734.
Houseman, R., R. Gold, and B. Pawson. 2001. Resource partitioning in two sympatric species of subterranean termites, Reticulitermes flavipes and Reticulitermes hageni (Isoptera: Rhinotermitidae). Environmental Entomology 30: 673-685.
Hussain, A., M.-Y. Tian, Y.-R. He, and Y.-Y. Lei. 2010a. Differential fluctuation in virulence and VOC profiles among different cultures of entomopathogenic fungi. Journal of Invertebrate Pathology 104: 166-171.
Hussain, A., M.-Y. Tian, Y.-R. He, J. M. Bland, and W.-X. Gu. 2010b. Behavioral and electrophysiological responses of Coptotermes formosanus Shiraki towards entomopathogenic fungal volatiles. Biological Control 55: 166- 173.
Hussain, A., and M.-Y. Tian. 2013. Germination pattern and inoculum transfer of entomopathogenic fungi and their role in disease resistance among Coptotermes formosanus (Isoptera: Rhinotermitidae). International Journal of Agriculture and Biology 15: 319-324.
Ibrahim, L., T. M. Butt, and P. Jenkinson. 2002. Effect of artificial culture media on germination, growth, virulence and surface properties of the entomopathogenic hyphomycete Metarhizium anisopliae. Mycological Research 106: 705-715.
Inglis, G. D., M. S. Goettel, T. M. Butt, and H. Strasser. 2001. Use of hyphomycetous fungi for managing insect pests. pp. 23-69, In T. M. Butt, C. Jackson, N. Magan (eds.) Fungi as biocontrol agents. CABI Publishing.
Inward, D., G. Beccaloni, and P. Eggleton. 2007. Death of an order: a comprehensive molecular phylogenetic study confirms that termites are eusocial cockroaches. Biology Letters 3: 331-335.
87
Jaronski, S. T. 2010. Ecological factors in the inundative use of fungal entomopathogens, pp. 159-185, The ecology of fungal entomopathogens. Springer.
Keller, S., P. Kessler, and C. Schweizer. 2003. Distribution of insect pathogenic soil fungi in Switzerland with special reference to Beauveria brongniartii and Metharhizium anisopliae. BioControl 48: 307-319.
Kiraly, Z., Z. Klement, F. Solymosy, J. Voros. 1970. Methods in Plant Pathology. Akademiai Kiado, Budapest, Hungary.
Klingen, I., J. Eilenberg, and R. Meadow. 2002. Effects of farming system, field margins and bait insect on the occurrence of insect pathogenic fungi in soils. Agriculture, Ecosystems & Environment 91: 191-198.
Kramm, K. R., D. F. West, and P. G. Rockenbach. 1982. Termite pathogens: transfer of the entomopathogen Metarhizium anisopliae between Reticulitermes sp. termites. Journal of Invertebrate Pathology 40: 1-6.
Křížková, B., T. Bourguignon, B. Vytisková, and J. Šobotník. 2014. The clypeal gland: A new exocrine gland in termite imagoes (Isoptera: Serritermitidae, Rhinotermitidae, Termitidae). Arthropod structure & development 43: 537- 542.
Lainé, L., and D. Wright. 2003. The life cycle of Reticulitermes spp.(Isoptera: Rhinotermitidae): what do we know? Bulletin of Entomological Research 93: 267-278.
Lamberty, M., D. Zachary, R. Lanot, C. Bordereau, A. Robert, J. A. Hoffmann, and P. Bulet. 2001. Insect immunity constitutive expression of a cysteine- rich antifungal and a linear antibacterial peptide in a termite insect. Journal of Biological Chemistry 276: 4085-4092.
Lee, K. E., and T. G. Wood. 1971. Termites and soils. pp. 251, London, UK, Academic Press.
Leis, M., and G. Sbrenna. 1983. Epidermal glands and integument of different castes of Kalotermes flavicollis (Isoptera, Calotermitidae): a comparative study [Termites]. Redia giornale di zoologia.
Lepage, M., and J. P. Darlington. 2000. Population dynamics of termites, pp. 333-361, Termites: Evolution, sociality, symbioses, ecology. Springer.
Li, Z., C. Li, B. Huang, and M. Fan. 2001. Discovery and demonstration of the teleomorph of Beauveria bassiana (Bals.) Vuill., an important entomogenous fungus. Chinese Science Bulletin 46: 751-753.
88
Liu, H., M. Skinner, M. Brownbridge, and B. L. Parker. 2003. Characterization of Beauveria bassiana and Metarhizium anisopliae isolates for management of tarnished plant bug, Lygus lineolaris (Hemiptera: Miridae). Journal of Invertebrate Pathology 82: 139-147.
Lo, N., and P. Eggleton. 2011. Termite phylogenetics and co-cladogenesis with symbionts, pp. 27-50, Biology of termites: a modern synthesis. Springer.
Lo, N., M. S. Engel, S. Cameron, C. A. Nalepa, G. Tokuda, D. Grimaldi, O. Kitade, K. Krishna, K.-D. Klass, and K. Maekawa. 2007. Save Isoptera: A comment on Inward et al. Biology Letters 3: 562-563.
Mackin-Rogalska, R., J. Pinowski, J. Solon, and Z. Wojcik. 1988. Changes in vegetation, avifauna, and small mammals in a suburban habitat. Polish Ecological Studies 14: 293-330.
Macko, V., and R. Staples. 1973. Regulation of uredospore germination and germ tube development. Bulletin of the Torrey Botanical Club: 223-229.
Marzluff, J. M. 2001. Worldwide urbanization and its effects on birds, pp. 19-47, Avian ecology and conservation in an urbanizing world. Springer.
Marzluff, J. M., and K. Ewing. 2001. Restoration of fragmented landscapes for the conservation of birds: a general framework and specific recommendations for urbanizing landscapes. Restoration Ecology 9: 280- 292.
Matsuura, K. 2011. Sexual and asexual reproduction in termites, pp. 255-277, Biology of termites: a modern synthesis. Springer.
Matsuura, K., and N. Kobayashi. 2007. Size, hatching rate, and hatching period of sexually and asexually produced eggs in the facultatively parthenogenetic termite Reticulitermes speratus (Isoptera: Rhinotermitidae). Applied Entomology and Zoology 42: 241.
Matsuura, K., M. Fujimoto, K. Goka, and T. Nishida. 2002. Cooperative colony foundation by termite female pairs: altruism for survivorship in incipient colonies. Animal Behaviour 64: 167-173.
Matsuura, K., T. Tamura, N. Kobayashi, T. Yashiro, and S. Tatsumi. 2007. The antibacterial protein lysozyme identified as the termite egg recognition pheromone. PLoS One 2: e813.
Mburu, D., N. Maniania, and A. Hassanali. 2013. Comparison of volatile blends and nucleotide sequences of two Beauveria bassiana isolates of different virulence and repellency towards the termite Macrotermes michealseni. Journal of Chemical Ecology 39: 101-108.
89
McIntyre, N. E. 2000. Ecology of urban arthropods: a review and a call to action. Annals of the Entomological Society of America 93: 825-835.
McIntyre, N., J. Rango, W. Fagan, and S. Faeth. 2001. Ground arthropod community structure in a heterogeneous urban environment. Landscape and Urban Planning 52: 257-274.
McKinney, M. L. 2002. Urbanization, Biodiversity, and Conservation. Bioscience 52: 883-890.
McKinney, M. L. 2006. Urbanization as a major cause of biotic homogenization. Biological conservation 127(3), 247-260.
Meyling, N. V. 2007. Methods for isolation of entomopathogenic fungi from the soil environment-laboratory manual. pp. 1-18, Laboratory manual. 03 Sep 2007.
Meyling, N. V. and J. Eilenberg. 2006. Occurrence and distribution of soil borne entomopathogenic fungi within a single organic agroecosystem. Agriculture, ecosystems & environment 113(1), 336-341.
Meyling, N. V., and J. Eilenberg. 2007. Ecology of the entomopathogenic fungi Beauveria bassiana and Metarhizium anisopliae in temperate agroecosystems: potential for conservation biological control. Biological Control 43: 145-155.
Meyling, N. V., M. Lübeck, E. P. Buckley, J. Eilenberg, and S. A. Rehner. 2009. Community composition, host range and genetic structure of the fungal entomopathogen Beauveria in adjoining agricultural and seminatural habitats. Molecular Ecology 18: 1282-1293.
Miura, T., A. Kamikouchi, M. Sawata, H. Takeuchi, S. Natori, T. Kubo, and T. Matsumoto. 1999. Soldier caste-specific gene expression in the mandibular glands of Hodotermopsis japonica (Isoptera: Termopsidae). Proceedings of the National Academy of Sciences 96: 13874-13879.
Myles, T. G. 1999. Review of secondary reproduction in termites (Insecta: Isoptera) with comments on its role in termite ecology and social evolution. Sociobiology 33: 1-43.
Nalepa, C. A., and C. Bandi. 2000. Characterizing the ancestors: paedomorphosis and termite evolution, pp. 53-75, Termites: Evolution, sociality, symbioses, ecology. Springer.
National Weather Service. Monthly and Seasonal Temperatures for Columbia, MO. http://www.crh.noaa.gov/lsx/?n=cli_archive (Accessed 3/12/2015)
90
Nielsen, R. A., and C. Brister. 1979. Greater wax moth: behavior of larvae. Annals of the Entomological Society of America 72: 811-815.
Noirot, C. 1969. Glands and secretions, pp. 89-123. In K. Krishna and F. M Weesner (eds.). Biology of termites, vol. 1, Academic Press.
Noirot, C. 1970. The nests of termites. pp. 73-125. In K. Krishna and F. M Weesner (eds.). Biology of termites, vol. 1, Academic Press.
Noirot, C. 1989. Social structure in termite societies. Ethology Ecology & Evolution 1: 1-17.
Noirot, C., and C. Noirot-Timothée. 1969. The digestive system. pp. 49-88. In K. Krishna and F. M Weesner (eds.). Biology of termites, vol. 1, Academic Press.
Nutting, W. 1969. Flight and colony foundation. pp. 233-282. In K. Krishna and F. M Weesner (eds.). Biology of termites, vol. 1, Academic Press.
Oke, T. R., J. Crowther, K. McNaughton, J. Monteith, and B. Gardiner. 1989. The micrometeorology of the urban forest [and discussion]. Philosophical Transactions of the Royal Society B: Biological Sciences 324: 335-349.
Ortiz-Urquiza, A., and N. O. Keyhani. 2013. Action on the surface: entomopathogenic fungi versus the insect cuticle. Insects 4: 357-374.
Paddock, F. B. 1918. The beemoth or waxworm, pp. 1-46, Texas Agricultural Experiment Station.
Pinzon, O. P., and R. M. Houseman. 2009. Species Diversity and Intraspecific Genetic Variation of Reticulitermes (Isoptera: Rhinotermitidae) Subterranean Termites in Woodland and Urban Environments of Missouri. Annals of the Entomological Society of America 102: 868-880.
Pontieri, L., S. Vojvodic, R. Graham, J. S. Pedersen, and T. A. Linksvayer. 2014. Ant colonies prefer infected over uninfected nest sites. PloS One 9: e111961.
Raina, A. K., Y. I. Park, and Z. Hruska. 2003. Ecdysone agonist halofenozide affects corpora allata and reproductive physiology of the Formosan subterranean termite, Coptotermes formosanus. Journal of Insect Physiology 49: 677-683.
Ramarao, N., C. Nielsen-Leroux, and D. Lereclus. 2012. The insect Galleria mellonella as a powerful infection model to investigate bacterial pathogenesis. Journal of Visualized Experiments.
91
Rangel, D. E. N., D. G. Alston, and D. W. Roberts. 2008. Effects of physical and nutritional stress conditions during mycelial growth on conidial germination speed, adhesion to host cuticle, and virulence of Metarhizium anisopliae, an entomopathogenic fungus. Mycological Research 112: 1355-1361.
Rangel, D. E., A. J. Anderson, and D. W. Roberts. 2006. Growth of Metarhizium anisopliae on non-preferred carbon sources yields conidia with increased UV-B tolerance. Journal of Invertebrate Pathology 93: 127-134.
Rath, A. 1995. Termite eats fungus-fungus eat termites. Pest Control 63: 42-43.
Rath, A. C. 2000. The use of entomopathogenic fungi for control of termites. Biocontrol Science and Technology 10: 563-581.
Rehner, S. A., A. M. Minnis, G.-H. Sung, J. J. Luangsa-ard, L. Devotto, and R. A. Humber. 2011. Phylogeny and systematics of the anamorphic, entomopathogenic genus Beauveria. Mycologia 103: 1055-1073.
Roberts, D. W., and A. E. Hajek. 1992. Entomopathogenic fungi as bioinsecticides, pp. 144-159, Frontiers in Industrial Mycology. Springer.
Roberts, D. W., and R. A. Humber. 1981. Entomogenous fungi. Biology of Conidial Fungi 2: 201-236.
Rodrigues, S., R. Peveling, P. Nagel, and S. Keller. 2005. The natural distribution of the entomopathogenic soil fungus Metarhizium anisopliae in different regions and habitat types in Switzerland. Insect Pathogens and Insect Parasitic Nematodes Melolontha 28: 185-188.
Rosengaus, R. B., M. R. Guldin, and J. F. Traniello. 1998a. Inhibitory effect of termite fecal pellets on fungal spore germination. Journal of Chemical Ecology 24: 1697-1706.
Rosengaus, R. B., A. B. Maxmen, L. E. Coates, and J. F. A. Traniello. 1998b. Disease resistance: a benefit of sociality in the dampwood termite Zootermopsis angusticollis (Isoptera: Termopsidae). Behavioral Ecology and Sociobiology 44: 125-134.
Rosengaus, R., C. Jordan, M. Lefebvre, and J. Traniello. 1999a. Pathogen alarm behavior in a termite: a new form of communication in social insects. Naturwissenschaften 86: 544-548.
Rosengaus, R. B., J. F. Traniello, T. Chen, J. J. Brown, and R. D. Karp. 1999b. Immunity in a social insect. Naturwissenschaften 86: 588-591.
Rosengaus, R. B., M. L. Lefebvre, and J. F. Traniello. 2000. Inhibition of fungal spore germination by Nasutitermes: evidence for a possible antiseptic role of soldier defensive secretions. Journal of Chemical Ecology 26: 21-39.
92
Rosengaus, R., J. Traniello, M. Lefebvre, and A. Maxmen. 2004. Fungistatic activity of the sternal gland secretion of the dampwood termite Zootermopsis angusticollis. Insectes Sociaux 51: 259-264.
Rosengaus, R. B., T. Cornelisse, K. Guschanski, and J. F. Traniello. 2007. Inducible immune proteins in the dampwood termite Zootermopsis angusticollis. Naturwissenschaften 94: 25-33.
Rosengaus, R. B., J. F. Traniello, and M. S. Bulmer. 2011. Ecology, behavior and evolution of disease resistance in termites, pp. 165-191, Biology of termites: a modern synthesis. Springer.
Rosengaus, R. B., K. F. Schultheis, A. Yalonetskaya, M. S. Bulmer, W. S. DuComb, R. W. Benson, J. P. Thottam, and V. Godoy-Carter. 2014. Symbiont-derived beta-1,3-glucanases in a social insect: mutualism beyond nutrition. Frontiers in Microbiology 5.
Roy, H. E., D. Steinkraus, J. Eilenberg, A. Hajek, and J. K. Pell. 2006. Bizarre interactions and endgames: entomopathogenic fungi and their arthropod hosts. Annual Review of Entomology 51: 331-357.
Rust, M. K., and N.-Y. Su. 2012. Managing social insects of urban importance. Annual Review of Entomology 57: 355-375.
Rutter, A., A. Morton, and P. Robins. 1975. A predictive model of rainfall interception in forests. II. Generalization of the model and comparison with observations in some coniferous and hardwood stands. Journal of Applied Ecology: 367-380.
SAS Institute Inc. 2014. Cary, NC: SAS Institute Inc.
Sánchez-Peña, S. R., J. S.-J. Lara, and R. F. Medina. 2011. Occurrence of entomopathogenic fungi from agricultural and natural ecosystems in Saltillo, México, and their virulence towards thrips and whiteflies. Journal of Insect Science 11: 1.
Scheffrahn, R. H., and N.-Y. Su. 1994. Keys to soldier and winged adult termites (Isoptera) of Florida. Florida Entomologist: 460-474.
Scheffrahn, R. H., N.-Y. Su, J. Krecek, A. Van Liempt, B. Maharajh, and G. S. Wheeler. 1998. Prevention of colony foundation by Cryptotermes brevis and remedial control of drywood termites (Isoptera: Kalotermitidae) with selected chemical treatments. Journal of Economic Entomology 91: 1387- 1396.
Shah, P., and J. Pell. 2003. Entomopathogenic fungi as biological control agents. Applied Microbiology and Biotechnology 61: 413-423.
93
Shellman-Reeve, J. S. 1994. Limited nutrients in a dampwood termite: nest preference, competition and cooperative nest defence. Journal of Animal Ecology: 921-932.
Shellman-Reeve, J. S. 1997. The spectrum of eusociality in termites. pp. 52-93, In. J. Choe and B Crespi (eds.) The evolution of social behavior in insects and arachnids, Cambridge University Press.
Shimizu, S., and M. Yamaji. 2003. Effect of density of the termite, Reticulitermes speratus Kolbe (Isoptera: Rhinotermitidae), on the susceptibilities to Metarhizium anisopliae. Applied Entomology and Zoology 38: 125-130.
Siderhurst, M., D. James, T. Blunt, and L. Bjostad. 2005a. Antimicrobial activity of norharmane against the entomopathogenic fungus Metarhizium anisopliae (Metsch.) and the caste and phylogenetic distribution of this defense in termites (Insecta: Isoptera). Sociobiology 46: 563-577.
Siderhurst, M. S., D. M. James, C. D. Rithner, D. L. Dick, and L. B. Bjostad. 2005b. Isolation and characterization of norharmane from Reticulitermes termites (Isoptera: Rhinotermitidae). Journal of Economic Entomology 98: 1669-1678.
Slaytor, M., and D. J. Chappell. 1994. Nitrogen metabolism in termites. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 107: 1-10.
Smith, G. 1977. Aspergillus fumigatus: A possible relationship between spore size and virulence for mice. Journal of General Microbiology 102: 413-415.
Snodgrass, R. E. 1935. pp. 667. Principles of Insect Morphology. Cornell University Press.
Šobotník, J., A. Jirošová, and R. Hanus. 2010. Chemical warfare in termites. Journal of Insect Physiology 56: 1012-1021.
Sookar, P., S. Bhagwant, and E. Awuor Ouna. 2008. Isolation of entomopathogenic fungi from the soil and their pathogenicity to two fruit fly species (Diptera: Tephritidae). Journal of Applied Entomology 132: 778- 788.
Stewart, J., and A. Thom. 1973. Energy budgets in pine forest. Quarterly Journal of the Royal Meteorological Society 99: 154-170.
Stranne, S. 2014. Species composition and abundance of entomopathogenic Metarhizium fungi in soils of a forest, pasture and agricultural field in Sweden. Bachelor’s Programme--Independent Project, Swedish University of Agricultural Sciences.
94
Su, N.-Y., and R. H. Scheffrahn. 1998. A review of subterranean termite control practices and prospects for integrated pest management programmes. Integrated Pest Management Reviews 3: 1-13.
Sun, B.-D., H.-y. Yu, A. J. Chen, and X.-Z. Liu. 2008. Insect-associated fungi in soils of field crops and orchards. Crop Protection 27: 1421-1426.
Survival Analysis: LIFETEST procedure. SAS Software. From http://support.sas.com/rnd/app/stat/procedures/SurvivalAnalysis.html#lifet est (accessed 25 March 2015).
Thorne, B. 1996. Termite terminology. Sociobiology 28: 253-263.
Thorne, B. L. 1997. Evolution of eusociality in termites. Annual Review of Ecology, Evolution, and Systematics: 27-54.
Thorne, B. L. 1998. Biology of Subterranean Termites of the Genus Reticulitermes, pp. 1-30, NPCA Research Report on Subterranean Termites. National Pest Control Association, Dunn Loring, Virginia.
Thorne, B., J. Traniello, E. Adams, and M. Bulmer. 1999. Reproductive dynamics and colony structure of subterranean termites of the genus Reticulitermes (Isoptera Rhinotermitidae): a review of the evidence from behavioral, ecological, and genetic studies. Ethology Ecology & Evolution 11: 149- 169.
Traniello, J. F., R. B. Rosengaus, and K. Savoie. 2002. The development of immunity in a social insect: evidence for the group facilitation of disease resistance. Proceedings of the National Academy of Sciences 99: 6838- 6842.
Tulloch, M. 1976. The genus Metarhizium. Transactions of the British Mycological Society 66: 407-411.
Turner, M. G., R. H. Gardner, and R. V. O’Neill. 2001. Landscape ecology in theory and practice: patterns and process. Springer-Verlag, New York.
Vandenberg, J. D., and F. A. Cantone. 2004. Effect of serial transfer of three strains of Paecilomyces fumosoroseus on growth in vitro, virulence, and host specificity. Journal of Invertebrate Pathology 85: 40-45.
Vänninen, I. 1996. Distribution and occurrence of four entomopathogenic fungi in Finland: effect of geographical location, habitat type and soil type. Mycological Research 100: 93-101.
Vänninen, I., G. B. Husberg, and M. T. Hokkanen. 1989. Occurrence of entomopathogenic fungi and entomoparasitic nematodes in cultivated soils in Finland. Acta Entomologica Fennica 53.
95
Vargo, E. L. 2003. Genetic structure of Reticulitermes flavipes and R. virginicus (Isoptera: Rhinotermitidae) colonies in an urban habitat and tracking of colonies following treatment with hexaflumuron bait. Environmental Entomology 32: 1271-1282.
Vargo, E. L., and C. Husseneder. 2009. Biology of subterranean termites: insights from molecular studies of Reticulitermes and Coptotermes. Annual Review of Entomology 54: 379-403.
Vargo, E. L., and J. R. Carlson. 2006. Comparative study of breeding systems of sympatric subterranean termites (Reticulitermes flavipes and R. hageni) in central North Carolina using two classes of molecular genetic markers. Environmental Entomology 35: 173-187.
Vega, F. E., F. Posada, M. C. Aime, M. Pava-Ripoll, F. Infante, and S. A. Rehner. 2008. Entomopathogenic fungal endophytes. Biological Control 46: 72-82.
Vega, F. E., M. S. Goettel, M. Blackwell, D. Chandler, M. A. Jackson, S. Keller, M. Koike, N. K. Maniania, A. Monzon, and B. H. Ownley. 2009. Fungal entomopathogens: new insights on their ecology. Fungal Ecology 2: 149- 159.
Walker, T. N., and W. O. Hughes. 2009. Adaptive social immunity in leaf-cutting ants. Biology Letters 5: 446-448.
Weesner, F. M. 1960. Evolution and biology of the termites. Annual Review of Entomology 5: 153-170.
Weesner, F. M. 1965. The Termites of the United States: a handbook. National Pest Control Association.
Weesner, F. M. 1970. Termites of the Nearctic region. pp. 477-525, In K. Krishna and F. M. Weesner (eds.) Biology of Termites vol. 2, Academic Press.
Wigglesworth, V. 1932. Memoirs: On the Function of the so-called ‘Rectal Glands’ of Insects. Quarterly Journal of Microscopical Science 2: 131-150.
Wilson, E. O. 1971. The insect societies. Belknap Press of Harvard University Press. Cambridge, Massachusetts. 562 p.
Wright, M. S., A. K. Raina, and A. R. Lax. 2005. A strain of the fungus Metarhizium anisopliae for controlling subterranean termites. Journal of Economic Entomology 98: 1451-1458.
Yanagawa, A., F. Yokohari, and S. Shimizu. 2008. Defense mechanism of the termite, Coptotermes formosanus Shiraki, to entomopathogenic fungi. Journal of Invertebrate Pathology 97: 165-170.
96
Yanagawa, A., F. Yokohari, and S. Shimizu. 2009. The role of antennae in removing entomopathogenic fungi from cuticle of the termite, Coptotermes formosanus. Journal of Insect Science 9: 6.
Yanagawa, A., N. Fujiwara-Tsujii, T. Akino, T. Yoshimura, T. Yanagawa, and S. Shimizu. 2011a. Behavioral changes in the termite, Coptotermes formosanus (Isoptera), inoculated with six fungal isolates. Journal of Invertebrate Pathology 107: 100-106.
Yanagawa, A., N. Fujiwara-Tsujii, T. Akino, T. Yoshimura, T. Yanagawa, and S. Shimizu. 2011b. Musty odor of entomopathogens enhances disease- prevention behaviors in the termite Coptotermes formosanus. Journal of Invertebrate Pathology 108: 1-6.
Yanagawa, A., N. Fujiwara-Tsujii, T. Akino, T. Yoshimura, T. Yanagawa, and S. Shimizu. 2012. Odor aversion and pathogen-removal efficiency in grooming behavior of the termite Coptotermes formosanus. PloS One 7: e47412.
Zhou, X., J. A. Smith, F. M. Oi, P. G. Koehler, G. W. Bennett, and M. E. Scharf. 2007. Correlation of cellulase gene expression and cellulolytic activity throughout the gut of the termite Reticulitermes flavipes. Gene 395: 29-39.
Zimmermann, G. 1986. The ‘Galleria bait method’ for detection of entomopathogenic fungi in soil. Journal of Applied Entomology 102: 213- 215.
Zimmermann, G. 2007. Review on safety of the entomopathogenic fungi Beauveria bassiana and Beauveria brongniartii. Biocontrol Science and Technology 17: 553-596.
Zoberi, M. H. 1995. Metarhizium anisopliae, a fungal pathogen of Reticulitermes flavipes (Isoptera: Rhinotermitidae). Mycologia: 354-359.
97
Figures
Figure 1 Two entomopathogenic fungi, Beauveria bassiana sensu lato (Balsamo) Vuillemin (A) and Metarhizium anisopliae s.l. (Metschnikoff) Sorokin (B), found in soils of forested habitats and urban areas that were historically forested around Columbia, Missouri.
98
Figure 2 Termite infestation risk in the United States. (Image source: USDA FS Home & Garden Bulletin 64)
99
Figure 3 Subterranean termite colony cycle. The colony starts with a founding reproductive pair (king and queen) who build a nest chamber, lay eggs and care for the initial emerging nymphs. The nymphs develop into workers or soldiers, or sometimes supplementary reproductives. A mature colony produces winged reproductives (alates) which will leave the colony in a mass emergence, called a swarm, enabling new colonies to begin. (Image source: http://www.ipmpestcontrol.com.au/termite-pest-control.php)
100
Figure 4 Beauveria bassiana (Balsamo) Vuillemin conidial structures (A); Conidiognous cells (B); and Conidia (C). Image: Hoog 1972
101
Figure 5 Metarhizium anisopliae (Metschnikoff) Sorokin, Sporodochium (a), Conidiophores (b), Conidia (c). Image: Hoog et al. 2000
102
Figure 6 Metarhizium anisopliae (Metschnikoff) Sorokin sensu lato emerging from a Reticulitermes flavipes (Kollar) dealated founding reproductive.
103
Figure 7 Infection pathway of entomopathogenic fungi. These fungi invade their insect hosts through the cuticle by recognizing the host surface, adhering to the host cuticle, germinating and penetrating the cuticle, entering the hemocoel, producing toxins, and ultimately causing the death of the host (A). The fungal mycelia then reemerge through the intercuticular regions to sporulate outside the cadaver (B). (Image sources: (A) www.nature.com/nrmicro/journal/v5/n5/fig_tab/nrmicro1638_F1.html; (B) organicsoiltechnology.com)
104
Figure 8 Landscapes where soil was sampled along a temporal gradient of urban development in forested and previously forested urban habitats. Three undeveloped forested habitats, three previously forested 10 yr subdivisions, and three previously forested 20 yr subdivisions were sampled to determine relative frequency of occurrence and relative frequency of mortality of entomopathogenic fungi in the soil.
105
Figure 9 Guilford Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition (B). Photo Source: Boone County Assessor website.
106
Figure 10 Monterey Hills Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition (B). Photo Source: Boone County Assessor website.
107
Figure 11 Grindstone Forest. Historical and contemporary land cover of a forested landscape in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area has been forested since before 1980 (A) and is currently in a similar condition (B). Photo Source: Boone County Assessor website.
108
Figure 12 Eastland Hills. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B). Photo Source: Boone County Assessor website.
109
Figure 13 Timber Ridge. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B). Photo Source: Boone County Assessor website.
110
Figure 14 Stoneridge Estates. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 2002 (B). Photo Source: Boone County Assessor website.
111
Figure 15 Lake Woodrail. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B). Photo Source: Boone County Assessor website.
112
Figure 16 Parkade North. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B). Photo Source: Boone County Assessor website.
113
Figure 17 Wynfield Meadows. Historical and contemporary land cover of a previously forested subdivision in Columbia, MO where soil sampling occurred to bait for entomopathogenic fungi. Historical aerial photographs show that this area was forested in 1980 (A) and was developed by 1992 (B). Photo Source: Boone County Assessor website.
114
Figure 18 Nine study areas in Columbia, Missouri where soil was collected to bait for entomopathogenic fungi. 1) Guilford Forest; 2) Monterey Hills Forest; 3) Grindstone Forest; 4) Eastland Hills; 5) Timber Ridge; 6) Stoneridge Estates; 7) Lake Woodrail; 8) Parkade North; 9) Wynfield Meadows. (Image source: Google Maps)
115
Figure 19 Collecting the soil. (Image source: http://www.dreamlakehomes.com/listing/3-bedroom-charmer-with-beautifully- treed-large-backyard/)
116
Figure 20 Soil fungi experimental set-up. Soil from each location (A) was homogenized and placed into two 100 ml containers (B). Ten Galleria mellonella (Linnaeus) larvae were placed into each container (C). One container from each location was randomly selected and placed in one of two incubating temperatures (15oC or 25oC) with 24 hr darkness, 80% ± 3 RH (D).
117
Figure 21 Soil containing Galleria mellonella (Linnaeus) from undisturbed forest and historically forested subdivisions, incubated at two temperatures (15°C and 25°C) (A), was checked weekly for cadavers (B). Cadavers were removed, surface sterilized, and placed into individual moisture chambers (C). Cadavers were placed again into the original incubating temperature and allowed to sporulate (D). Beauveria bassiana sensu lato (E) and Metarhizium anisopliae (Metschnikoff) Sorokin sensu lato (F) were two entomopathogenic fungi isolated from these soils.
118
Figure 22 Soil sample results: Guilford Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). (Map source: Google Maps)
119
Figure 23 Soil sample results: Monterey Hills Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). (Map source: Google Maps)
120
Figure 24 Soil sample results: Grindstone Forest. Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). (Map source: Google Maps)
121
Figure 25 Soil sample results: Eastland Hills (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). No fungi were found at the locations marked in blue. (Map source: Google Maps)
122
Figure 26 Soil sample results: Timber Ridge (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). (Map source: Google Maps)
123
Figure 27 Soil sample results: Stoneridge Estates (10 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). No fungi were found at the locations marked in blue. (Map source: Google Maps)
124
Figure 28 Soil sample results: Lake Woodrail (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). No fungi were found at the location marked in blue. (Map source: Google Maps)
125
Figure 29 Soil sample results: Parkade North (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). (Map source: Google Maps)
126
Figure 30 Soil sample results: Wynfield Meadows (20 year). Relative frequency of occurrence and location of Beauveria and Metarhizium at two temperatures, 15oC and 25oC. The figures show whether Beauveria (purple) and Metarhizium (green) where found at 15oC (A) and 25oC (B). No fungi were found at the location marked in blue. (Map source: Google Maps)
127
Figure 31 Relative frequency of occurrence of two soil fungi, Metarhizium and Beauveria from forest and previously forested subdivisions around Columbia, Missouri. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
128
Figure 32 Relative frequency of occurrence of two entomopathogenic fungi, Beauveria and Metarhizium, in three habitats. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
129
Figure 33 Relative frequency of occurrence of two soil fungi, Metarhizium and Beauveria from forest and previously forested subdivisions around Columbia, Missouri. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
130
Figure 34 Relative frequency of occurrence of Beauveria and Metarhizium fungi in forests and subdivisions of various ages. Forested habitats, 10 yr old previously forested subdivisions, and 20 yr old previously forested subdivisions were sampled to determine relative frequency of occurrence of entomopathogenic fungi.
131
Figure 35 Relative frequency of mortality of Galleria mellonella (Linnaeus) by incubation temperature and fungus. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
132
Figure 36 Relative frequency of mortality of Galleria mellonella (Linnaeus) baits by entomopathogenic fungi, Beauveria and Metarhizium, in three habitats types. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
133
Figure 37 Relative frequency of mortality of Galleria mellonella (Linnaeus) by fungus and habitat type. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
134
Figure 38 Relative frequency of mortality of Galleria mellonella (Linnaeus) baits by entomopathogenic fungi, Beauveria and Metarhizium. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
135
Figure 39 Relative frequency of occurrence of Beauveria and Metarhizium according to habitat type, incubation temperature and fungi. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
136
Figure 40 Relative frequency of mortality of Galleria mellonella (Linnaeus) by incubation temperature and fungus. Data are shown as mean ± standard error. Bars within individual graphs that share a letter in common are not significantly different (α ≤ 0.05).
137
Figure 41 Relative frequency of mortality of Galleria mellonella (Linnaeus) by Beauveria and Metarhizium fungi through time in forest and subdivisions of various ages. Study areas used to sample soil along a temporal gradient of urban development in both forested and historically forested urban habitats. Forested habitats, 10 yr old previously forested subdivisions, and 20 yr old previously forested subdivisions were sampled to determine relative frequency of occurrence and of entomopathogenic fungi.
138
Figure 42 Subterranean termite infested logs (A) were collected from the forest at Capen Park in Columbia, Missouri. Individual colonies were placed into swarming chambers (B). Reticulitermes flavipes (Kollar) built swarming tubes (C) and swarmed nearly a month after collection.
139
Figure 43 In 6-well dishes with Q5 filter paper moistened with dH2O in each chamber, sexed and paired Reticulitermes flavipes (Kollar) were placed in a holding cell for a day before treatment (A). Termites were placed into treatment cells (B) for 30 minutes and then transferred into clean chambers (C) for recovery and observation. The 6-well plates with termites were then placed into an environmental chamber at15oC, 80% ± 3% RH, 24h dark and checked every 3-4 days.
140
Figure 44 Fungi used in this bioassay: forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU).
141
Figure 45 VanTieghem cells plus water agar are ½ cm pieces of sterile silicon tubing (1/4 inch inner diameter) placed in the center of a water agar plate (100 mm x 15mm).
142
Figure 46 Reticulitermes flavipes (Kollar) were monitored every 3-4 days for 60 days. Mortality for each pair, time to lay eggs and maximum number of eggs found during the study were recorded each time. Cadavers (B) were removed, surface sterilized and placed individually into 1.5 ml Eppendorf tubes containing Potato Dextrose Agar (PDA) (C) to verify the emerging, sporulating fungus. If the termites survived (D), the date of first eggs and number of eggs (E) contained in the cell were recorded.
143
Figure 47 No-choice virulence bioassay testing the virulence of spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C) towards Reticulitermes flavipes (Kollar). d1= death of first member of the pair. d2=death of second member of the pair. d1 and d2 were analyzed separately. α=0.05.
144
Figure 48 No-choice virulence bioassay Lifetest results examining days to death for each member of the pair for 50% of the population of R. flavipes after exposure to spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C).
145
Figure 49 No-choice virulence bioassay PROC LIFETEST results examining days to death for each member of the pair for 50% of the population of R. flavipes after exposure to spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). This figure shows colony response differences.
146
Figure 50 Time until half (T50) of Reticulitermes flavipes (Kollar) founding pairs laid eggs. BF=forest Beauveria, BU=urban Beauveria, MF=forest Metarhizium, MU=urban Metarhizium, and C=control (no spores).
147
Figure 51 Beauveria bassiana (Balsamo) Vuillemin sensu lato spores on a hemocytometer. Urban (A) and forest (B) collected fungi are similar sizes.
148
Figure 52 Metarhizium anisopliae (Metschnikoff) Sorokin sensu lato spores on a hemocytometer. Urban (A) and forest (B) collected fungi are different sizes.
149
Figure 53 Reticulitermes flavipes (Kollar) founding pairs were initially placed in the interstitial area with access to four chambers (A). Green filled circles are an example of the random combinations of two fungal treated chambers with the white cell representing control chambers. Termites are shown in the various ways found when observed: (B) together in a control chamber, (C) together in a fungal treated chamber, (D) separated into two chambers—could both be fungal treated or control or one of each, or (E) in the interstitial area.
150
Figure 54 Treatment chamber preference comparisons for Reticulitermes flavipes (Kollar). Forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). Sorted by treatment. α=0.05.
151
Figure 55 Mortality comparisons of Reticulitermes flavipes (Kollar) founding pairs after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU).
152
Figure 56 Mortality comparisons of Reticulitermes flavipes (Kollar) founding pair individuals [d1 (A) and d2 (B)] after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU).
153
Figure 57 Nesting site selection study examining the virulence of spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU) towards Reticulitermes flavipes (Kollar). d1= death of first individual of the founding pair. d2= death of second individual of the founding pair.
154
Figure 58 Nesting site selection by treatment. PROC Lifetest results examining days to death for 50% of Reticulitermes flavipes (Kollar) founding pairs after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU).
155
Figure 59 Nesting site selection by treatment. PROC Lifetest results examining days to death days to death for 50% of each individual of the Reticulitermes flavipes (Kollar) founding pair for 50% after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU).
156
Figure 60 Cumulative mortality (cd) and mortality at each observation (td). Forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU).
157
Figure 61 Nesting site selection by colony. PROC Lifetest results examining days to death for each member of the Reticulitermes flavipes (Kollar) founding pair for 50% after exposure to spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C).
158
Figure 62 Head and abdomen regions and associated glands found on Reticulitermes. (Figure adapted from Costa-Leonardo and Haifig 2010)
159
Figure 63 Reticulitermes flavipes (Kollar) were sexed morphologically using the characteristics taken from Lainé and Wright (2003): Reticulitermes spp. females have an enlarged seventh sternite (A) that covers the eighth and ninth sternites whereas the male’s seventh sternite is similar in size to the sixth sternite (B); the male also has a pair of styli on the ninth sternite (C). (Image source: R.M. Houseman)
160
Figure 64 A solution containing spores of Beauveria or Metarhizium (A) was mixed with Reticulitermes flavipes (Kollar) head supernatant to achieve 0% (controls) and 10% solution (10% head to 90% spore solution).
161
Figure 65 After the elapsed incubation time [Metarhizium 29 hr (B), Beauveria 31 hr (A)], spore germination images were digitally taken using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification. Spore germination tubes (C and D) are shown for Beauveria and Metarhizium.
162
Figure 66 Metarhizium spore germination remained the same when exposed to 10% Reticulitermes flavipes (Kollar) head or abdomen extract (B) compared to the control (A). Images were digitally taken using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification.
163
Figure 67 Metarhizium spore germination means after exposure to Reticulitermes flavipes (Kollar) abdomen or head extracts. Values are shown as mean (columns) and standard error (whiskers).
164
Figure 68 Beauveria spore germination decreased when exposed to 10% Reticulitermes flavipes (Kollar) head extract (B) compared to the control (A). Images were digitally taken using CellSense software on a Nikon Eclipse TS100 inverted microscope with an Olympus XC30 camera. Images were taken at 400x magnification.
165
Figure 69 Beauveria spore germination means after exposure to Reticulitermes flavipes (Kollar) abdomen or head extracts. Values are shown as mean (columns) and standard error (whiskers).
166
Tables
Table 1 Descriptive statistics for subdivisions in Columbia, MO where soil samples were taken to bait for entomopathogenic fungi. (* = information updated from Botch 2013)
167
Table 2 Relative frequency of occurrence. Total sampling locations where Beauveria and Metarhizium were found when baited from Galleria mellonella (Linnaeus) larvae at 15oC and 25oC. Soil was taken in forested, 10 yr previously forested, and 20 yr previously forested subdivisions. Ten homes or plots were sampled within each habitat. Neither Beauveria or Metarhizium emerged from control soil samples. Gu=Guilford; MH=Monterey Hills; Gr=Grindstone; EH=Eastland Hills; TR=Timber Ridge; SE=Stoneridge Estates; LW=Lake Woodrail; PN=Parkade North; WM=Wynfield Meadows.
168
Table 3 Relative frequency of mortality. Soil fungi emerging from Galleria mellonella (Linnaeus) larvae exposed to soil from forest, 10 yr urban, and 20 yr urban historically forested landscapes incubated at two temperatures (15oC and 25oC ). The top table shows the number of G. mellonella) exhibiting Beauveria bassiana (Balsamo) Vuillemin infections and the bottom table shows the number of waxworms exhibiting Metarhizium anisopliae (Metschnikoff) Sorokin infections at each sample location.
169
Table 4 Relative frequency of occurrence and frequency of mortality of Galleria mellonella (Linnaeus) larvae exposed to fungi found in soils from forests, 10 yr subdivisions and 20 yr subdivisions in Columbia, Missouri. ANOVA performed using a link logit with a binomial distribution. Type=habitat (forest, 10 yr, 20 yr); nDF=Numerator Degrees of Freedom; dDF= Denominator Degrees of Freedom; ns=non-significant; * = p ≤ 0.05; ** = p ≤ 0.01; † = p ≤ 0.10
170
Table 5 The USDA-ARS Collection of Entomopathogenic Fungal Cultures (ARSEF) Collection information for fungi used in this study.
171
Table 6 No-choice virulence study overall statistical significance. Forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU). Significant values in bold, α=0.05.
172
Table 7 No-choice virulence bioassay testing the virulence of spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). d1=death of first individual of the founding pair. d2=death of second individual of the founding pair. Significant values in bold, α=0.05. *=significant; **=very significant.
173
Table 8 Percentage of Reticulitermes flavipes (Kollar) founding pair individuals killed by exposure to Beauveria and Metarhizium in forest and urban habitats.
Control Forest Urban
Total founding pairs d1 21.7% 88.3% 73.3% killed with Beauveria d2 11.7% 83.3% 73.3% Total founding pairs d1 21.7% 73.3% 60.0% killed with Metarhizium d2 11.7% 63.3% 55.0%
174
Table 9 No-choice virulence bioassay T50 results examining days to death for each member of the pair for 50% of the population of Reticulitermes flavipes (Kollar) after exposure to spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). T50 results in blue with the first number being the mean number of days for 50% of founding pair individuals to die. T50 values include 95% confidence intervals. A: d1=death of first member of the pair. B: d2=death of second member of the pair. Significant values in bold, α=0.05. *=significant; **=very significant.
175
Table 10 No-choice virulence bioassay T50 results examining the days until the first egg was laid for 50% of the Reticulitermes flavipes (Kollar) population when exposed to spores from forest Beauveria (BF), urban Beauveria (BU), forest Metarhizium (MF) and urban Metarhizium (MU) or no spores as a control (C). T50 results in blue with the first number being the mean number of days for 50% of founding females to lay eggs. T50 values include 95% confidence intervals. Significant values in bold, α=0.05. *=significant; **=very significant.
176
Table 11 Percentage of observations of Reticulitermes flavipes (Kollar) found in control (C) or fungus (F) chambers or interstitial area (I).
177
Table 12 Choice virulence study overall statistical significance. Reticulitermes flavipes (Kollar) founding pairs were more likely to die when exposed to urban fungi than forest fungi. Significant values in bold, α=0.05.
178
Table 13 Choice virulence bioassay testing the virulence of spores from forest Beauveria (BF), forest Metarhizium (MF) and urban Metarhizium (MU). These tables contrast the mortality of Reticulitermes flavipes (Kollar) exposed to the various fungi. A: td=Overall mortality. B: d1=death of first individual of the founding pair. C: d2=death of second individual of the founding pair. D: td=Overall mortality by colony. T50 results in blue with the first number being the mean number of days for 50% of founding pair individuals to die. T50 values include 95% confidence intervals. Significant values in bold, α=0.05, *=significant.
179
Table 14 Metarhizium spores germinated 9.29 times more often than Beauveria (p=<0.0001). Data were analyzed using SAS® Software.
180
Table 15 Metarhizium spore germination results when exposed to Reticulitermes flavipes (Kollar) body extracts.
181
Table 16 Analysis of Beauveria and Metarhizium spore germination when exposed to Reticulitermes flavipes (Kollar) founding pairs body extracts from head and abdomen. Data were analyzed using SAS® Software.
182
Table 17 Beauveria spore germination results when exposed to Reticulitermes flavipes (Kollar) body extracts.
183
VITA
Tamra Reall Lincoln was born in Provo, UT to Shela and Michael Reall. She graduated from Hickman High School in Columbia, Missouri. In 1995, she began her Bachelor’s of Science degree in horticulture at Brigham Young University. After graduating, she worked in landscaping, with the public as a master gardener, and as a mom to three wonderful children. Crossing the country many times and visiting nearly every state, she found many things to appreciate, especially the diversity of insect life. In 2010, she returned to school to begin a PhD at the University of Missouri in the Division of Plant Sciences— Entomology Program Area. While there, she completed a minor in college teaching and a graduate certificate in science outreach. She looks forward to many years of discovering the intricacies of social insects and sharing the knowledge with others.
184