The Effects of European Fire Ants on Blacklegged Ticks in Acadia National Park

The University of Maine DigitalCommons@UMaine

Honors College

Spring 2019

Lucy Guarnieri University of Maine

Follow this and additional works at: https://digitalcommons.library.umaine.edu/honors

Part of the Biology Commons

Recommended Citation Guarnieri, Lucy, "The Effects of European Fire Ants on Blacklegged Ticks in Acadia National Park" (2019). Honors College. 519. https://digitalcommons.library.umaine.edu/honors/519

This Honors Thesis is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Honors College by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected]. THE EFFECTS OF EUROPEAN FIRE ANTS ON BLACKLEGGED TICKS IN

ACADIA NATIONAL PARK

Lucy Guarnieri

A Thesis Submitted in Partial Fulfillment of the Requirements for a Degree with Honors (Biology)

The Honors College

University of Maine

May 2019

Advisory Committee: Allison Gardner, Assistant Professor of Arthropod Vector Biology, Advisor Eleanor Groden, Professor of Entomology Samantha Jones, Adjunct Assistant Professor in Art and Preceptor in the Honors College Danielle Levesque, Assistant Professor of Mammalogy and Mammalian Health Anne Lichtenwalner, Associate Professor of Animal and Veterinary Sciences ABSTRACT

The blacklegged tick (Ixodes scapularis) and the European fire ant (Myrmica rubra) are notable invasive pests in Maine, especially across coastal and southern regions. I. scapularis is the primary vector of the bacterium Borrelia burgdorferi, which is the causative agent of Lyme disease, and Maine currently has one of the highest Lyme disease incidence rates in the U.S. Ticks have many natural predators, including ants.

This study investigates the effects of M. rubra on I. scapularis abundance and pathogen infection prevalence in Acadia National Park (ANP). I collected ticks by drag-sampling at eight ant-infested sites and eight control sites in ANP. I hypothesized that I. scapularis abundance would be lower at treatment sites, due to predation by M. rubra, but found no significant difference. In addition, I conducted laboratory bioassays to measure M. rubra aggression against different life stages of I. scapularis. I found that M. rubra behaves more aggressively towards I. scapularis adults than towards nymphs and larvae. Finally, I extracted DNA from a subset of collected nymphs and tested the samples for B. burgdorferi to determine nymphal infection prevalence (NIP) at control and treatment sites. I hypothesized a decreased NIP at treatment sites due to reduced abundance and host-encounter frequency of the small mammal reservoir hosts of B. burgdorferi, and again found no significant difference between control and treatment sites. DEDICATION

To my Grandmama.

iii ACKNOWLEDGEMENTS

Thank you to my friends and co-workers Alyssa Marini, Troy Cloutier, and

Cassie Steele for doing a large part of the tick collection. It is well worth acknowledging that your morale never dropped in spite of several stings which were suffered among the group. A big thank you to Ann Bryant for lending your time and expertise during the pathogen testing process. A very special thank you to my friend and mentor Sara

McBride—it was always a joy working with you, even at our least-favorite sites. I am so proud of what we accomplished together in Acadia and I cherish the memories of our time there. To my advisor, Allie Gardner, thank you for providing me with this research opportunity and for supporting me from start to finish. You have helped me to grow more than I thought was possible. And finally, to my mom, dad, Beth, and all my family and friends, thank you for your endless love and support!

iv TABLE OF CONTENTS

LIST OF FIGURES ...... vi

INTRODUCTION ...... 1

METHODS ...... 11

Field Sites ...... 11

Drag Sampling ...... 13

Small Mammal Track Plates ...... 14

Aggression Bioassay ...... 15

RESULTS ...... 19

BIBLIOGRAPHY ...... 31

AUTHOR’S BIOGRAPHY ...... 36

v LIST OF FIGURES

Figure 1: Submissions of I. scapularis to Maine Medical Center Research Institute

(MMCRI) by town, 1989-2013………………………………………………….…8

Figure 2: M. rubra distribution in Maine, 2005……………………………………...…..9

Figure 3: Confirmed Lyme disease cases in the U.S., 2001 and 2017…………...….…..10

Figure 4: Map of drag sampling sites in Acadia National Park………………….……..13

Figure 5: Small mammal track plate…………………………………………………….15

Figure 6: Ant nest boxes…………………………………………………………..……..17

Figure 7: Nymphs collected per hour at M. rubra and control sites……………..……...20

Figure 8: M. rubra aggression scores against ticks………………………………..……22

Figure 9: Gel electrophoresis showing positive bands for Borrelia DNA……………….23

Figure 10: Nymphal infection prevalence at control and M. rubra sites……...……...….24

vi INTRODUCTION

Invasive species are changing ecosystems worldwide at an unprecedented scale.

Recent research has underlined the connection between human-driven environmental change, biological invasion, and the resultant loss of biodiversity (e.g. Chapin et al. 2000;

Wiens and Graham 2005; Wiens et al. 2010). The rate at which species respond to environmental change—the rate at which they adapt—determines their spread into new habitats (Holt 2003). Counterbalancing adaptation is the theory of niche conservatism: species tend to retain their ecological characteristics, fulfilling similar roles over time

(Wiens and Graham 2005). Therefore, the potential range that an organism can occupy depends on similarity between the new habitat and the organism’s native habitat, as well as the potential for the species to adapt and establish itself successfully in a new niche.

Arthropods are particularly good at this and are recognized as a highly successful group among biological invaders. Owing to their rapid regeneration time and high reproductive capacity, arthropod populations are well-equipped to adapt when faced with changing environmental conditions. Furthermore, they are able to disperse rapidly by flight, wind, and host transportation (Sanders et al. 2010). Humans often directly facilitate the spread of these invaders to novel habitats. This usually occurs inadvertently (e.g. emerald ash borer transportation in firewood), though there have been various cases of intentional introduction as well, the gypsy moth being one of the most striking examples. Invasive arthropods such as these have severe effects on animal and human health, agriculture, forestry, and habitat biodiversity (Sanders et al. 2010).

1 Climate change and invasive species greatly impact the ecology of vector-borne diseases. By disrupting the intricate relationships between pathogens, vectors, vector hosts, and the environment, climate change and invasive species have the potential to significantly alter vector-borne disease complexes (Shope 1991; Githeko et al. 2000).

Rising average temperature alone is already causing pronounced change in species distribution. After analyzing distribution data for 99 species of butterflies, birds and herbs, Parmesan and Yohe (2003) reported an average range shift of 6.1km per decade northward. Other climate models have reiterated the potential for geographic expansion of epidemiologically relevant arthropod species, including the blacklegged tick, Ixodes scapularis (Ogden et al. 2005), and mosquitos Aedes aegypti and Aedes albopictus (Liu et al. 2019). Furthermore, range shift will lead to increased encounters between invasive species. This raises the question: under what circumstances will one invasive species facilitate/inhibit the survival of another?

This study investigates the interaction between I. scapularis and the European fire ant, Myrmica rubra: two invasive arthropods that cohabit deciduous forest in coastal New

England. Myrmica rubra were accidentally introduced from Europe in the early 1900s, though the precise date and cause of invasion remain unclear (Groden et al. 2005). Ixodes scapularis and M. rubra occupy similar geographic range in the state of Maine (Figure 1;

Figure 2). Both thrive in deciduous forests, where leaf litter provides protection against desiccation. Ecological models have predicted that climate change will cause future range expansion of both species (Ogden et al. 2005; Soucy et al. 2018; Bertelsmeier 2013).

Ixodes scapularis was first confirmed to transmit Borrelia burgdorferi, the bacterial spirochete that causes Lyme disease, in 1982 (Burgdorfer and Gage 1986;

2 Piesman and Sinsky 1988). Since then, the blacklegged tick has become recognized as a serious threat to public health. Without treatment, Lyme disease can cause many symptoms including facial paralysis, arthritis, joint pain, and severe chronic headaches

(CDC 2018). Ixodes scapularis populations have become well established across the

Northeast and Upper Midwest during the 21st century (Eisen et al. 2016). Coupled with increasing tick abundance, states in these regions have experienced increasing rates of

Lyme disease (Figure 3). These states account for over 95% of cases reported in the U.S. each year. In Maine, 1,769 cases of Lyme were reported in 2017, and statewide incidence reached over 100 confirmed cases per 100,000 population, which was a dramatic increase from previous years and well above the national incidence rate of 9.1 (CDC 2018). In addition to B. burgdorferi, I. scapularis is a vector of the pathogens that cause

Anaplasmosis and Babesiosis (i.e., Anaplasma phagocytophilum and Babesia microti), of which there were 662 and 117 reported cases in Maine in 2017, respectively (CDC 2018).

Furthermore, as surveillance for Lyme is conducted passively, the CDC estimates that reported cases may represent as little as 10% of actual cases (CDC 2018).

Ixodes scapularis have a two-generation life cycle with four life stages: eggs, larvae, nymphs, and adults (CDC 2018). During spring and summer, larvae hatch from eggs and take a blood meal from a bird or a small mammal, typically from a white footed mouse or a chipmunk. They then molt into nymphs and enter dormancy for the winter.

Nymphs emerge in the spring of the second year, take a blood meal from another small mammal, and molt into adults. In the fall, adult females take a final blood meal from a large mammal, such as a deer or a person, before they mate and lay eggs. Borrelia burgdorferi infects many vertebrates, including small mammals and birds, and it may be

3 acquired by I. scapularis larvae and nymphs while they feed on these hosts. Once a larva or a nymph becomes infected, it can transmit the pathogen to other host animals during subsequent feedings. Humans can only acquire the bacterium from a biting nymph or adult, and nymphs present the greatest threat because they are very small and often go unnoticed.

Social insect pests, i.e. ant, wasp and termite pests, are attributed to huge amounts of economic damage (Rust 2012). In fact, a single ant species, the red imported fire ant

Solenopsis invicta, has become one of the most important invasive pests in the U.S. and accounts for up to 18% of the studies on invasive insects (Kenis et al. 2008). Though less aggressive than S. invicta, M. rubra are still a formidable nuisance. First reported on MDI as early as the late 1960s, M. rubra are now infested in very high densities within isolated areas of the island (Groden et al. 2005). Infestations on MDI reach average densities of

1.4 nests/m2, nearly ten times higher than the peak density of this species in its native range (Groden et al. 2005). Their stings are painful, there are no effective control measures to mitigate them, and their presence reduces property value (Arevalo and

Groden 2007). Myrmica rubra inflict substantial damage on native species of vertebrates

(Defisher 2013) and invertebrates (McPhee et al. 2012), as well. Highly adept at displacing native ant species through aggression and competitive exclusion of resources,

M. rubra have led to a steep decline in ant diversity in infested areas (Ouellette 2010;

Garnas et al. 2014). These ants also homogenize plant and insect species composition by tending to homopterans such as aphids, as honeydew secretions provide them with an important source of carbohydrates (McPhee et al. 2012).

4 Ants have been cited as predators of ticks more than any other arthropod group

(Samish and Alekseev 2001); however, the relationships between ants and ticks have not been widely investigated (Zingg et al. 2018). Much of the documentation on this subject comes from laboratory bioassays. Ants are not considered obligate predators of ticks under natural conditions, thus their role in regulating tick populations has been regarded as minimal (Samish et al. 2008). Nevertheless, several field studies have produced evidence for a negative correlation between tick abundance and the presence and/or density of ant nests (Burns and Melancon 1977; Zingg et al. 2018). Previous literature indicates that ants prey on adult females more than other life stages, though there are also numerous examples of predation on nymphs, larvae and eggs (Samish and Alekseev

2001). These data most likely vary based on the species of the ticks and ants in question, as well as particular experimental conditions. To our knowledge, there has been no attempt to determine whether M. rubra, in particular, plays a role in tick predation.

Due to the importance of small mammals as reservoir hosts in the I. scapularis life cycle and the B. burgdorferi transmission cycle, any effect of M. rubra on small mammals could impact I. scapularis and their rates of infection. Change in small mammal behavior in infested areas may include active avoidance of M. rubra on the ground (Bruce Connery, personal communication). If small mammals do indeed spend less time foraging in M. rubra infested habitat, presumably host-encounter frequency between small mammals and ticks would be reduced, thereby lowering nymphal infection prevalence (NIP). Castellanos et al. (2016) showed that suppression of S. invicta was associated with an increase in small mammal abundance as well as abundance of off-host ticks. However, NIP depends on more factors than strictly host abundance. For example,

5 the amount of time each host spends foraging or grooming may also factor in determining

NIP. A second alternate hypothesis is that M. rubra will reduce the diversity of small mammal hosts, thereby inhibiting the dilution effect (Ostfeld 2000) and increasing NIP.

The complexity of these interactions make it difficult to predict a result; however, given the volume of literature describing ant predation on ticks, as well as adverse effects on small mammals, I proposed that M. rubra would likely reduce at least one of these two elements in the Lyme disease system.

Maine has received particular attention with regard to tick-borne disease due to the rapid expansion of I. scapularis in south-north and east-west gradients across the state, the sheer magnitude of Lyme disease incidence, and the risk for human-tick encounters during outdoor recreation. Ticks are highly sensitive to moisture (Subak 2003;

Randolph 2004) because their high surface area to volume ratio makes them prone to desiccation; thus, summer rainfall and humidity, relatively high winter temperatures, and insulating snow cover presumably create suitable conditions for the tick along the Maine coast and, increasingly, further inland. Other hypothesized mechanisms favoring the spread of I. scapularis include forest fragmentation, creating ideal habitat for mice and deer (Allan et al. 2003), as well as tick dispersal by birds (Ogden 2006; Ogden et al.

2008; Rand et al. 2007). Outdoor recreation is popular and an important source of revenue for the state, especially during the summer months when nymphs are active, which elevates the risk of exposure to the ticks and the pathogens they transmit. Tick- borne disease incidence is highest in southern and coastal regions, including Mount

Desert Island (MDI; Figure 3), which is home to Acadia National Park (ANP). In ANP, risk of tick-borne disease exposure has raised particular concern (Johnson et al. 2017).

6 ANP is one of the most heavily trafficked areas in the state, with up to 3,500,000 annual visitors. The vast majority of these visitors come during the summer months during peak nymphal activity. On top of that, most park guests travel from out of state and may be less informed about Lyme disease risk.

Keeping in mind the risk of Lyme disease associated with I. scapularis, the ecosystem damage associated with M. rubra, as well as the likelihood that climate change will favor range expansion for both species as previously discussed, the importance of management practices cannot be overstated. This thesis investigates the effects of M. rubra on I. scapularis abundance and NIP in ANP, in order to better inform management strategies of both species. Two hypothesis were tested: (1) due to predation, average I. scapularis abundance will be lower in M. rubra-infested areas than in areas where the ants are not present, and (2) NIP will be lower in infested areas due to reduced small mammal abundance and thus reduced host-encounter frequency between ticks and the small mammal reservoir hosts of B. burgdorferi.

Figure 1: Submissions of I. scapularis to MMCRI, 1989-2013. Source: Maine Land Trust Network

Figure 2: Distribution of M. rubra in Maine. Note similarities to I. scapularis distribution above. Source: Groden et al. 2005

Figure 3: Lyme disease cases in 2001 (top) and 2017 (bottom). Source: http://www.cdc.gov/lyme/datasurveillance

10 METHODS

Field Sites

I relied on maps and personal communication with Dr. Ellie Groden to locate M. rubra infestations, and selected sites to represent the full range of M. rubra on the east side of

MDI (Table 1; Figure 4). Considering the time and resources available for this project, it was necessary to minimize the range of sites geographically. Limiting sites to the east side also kept habitat and climate features, such as forest type, humidity, and temperature—all of which are important factors in I. scapularis survival as discussed previously—relatively consistent across sites. Due to the high density of M. rubra nests, it was possible to determine presence by turning over logs and inspecting the vegetation for ants. Infested sites were completely saturated with M. rubra: leaf litter, understory plants and trees were covered with workers, and nests could be found under nearly every fallen log. All treatment sites were thus infested; though we did not measure density quantitatively, we estimate that average density was 1.24 nests/m2 based on previous measurements in ANP by Groden et al. (2005). We assumed that probing logs and vegetation was sufficient to confirm absence of M. rubra at control sites, as they occur only at such high densities in this area. Native ants were present at many of the control sites, which served as a secondary indication of M. rubra absence. When ants of any species were observed at a control site, we collected specimens in 70% ethanol and checked for identifying features of M. rubra [i.e. propodeum with two spurs, post-petiole, and a slightly angled scape (Arevalo and Groden 2007)] with a light microscope to confirm a negative identification.

11 I aimed to pair control sites with treatment sites to account for local variation in I. scapularis presence, which required locating areas that were as physically close to treatment sites as possible without the presence of M. rubra nests. Pairs were also matched for habitat features—including plant species, canopy cover, and leaf litter—to control for microclimate conditions as well as the efficiency of drag sampling. We took advantage of natural and artificial barriers such as roads and streams when possible, which permitted the placement of some control sites within 50m of respective treatment sites. All control sites were located within 1.25km of treatment sites except in one case, where the infestation was sufficiently widespread to warrant further location of the control (Lakewood and Old Farm Rd.).

Table 1: Drag sampling locations (* indicate sites included in pathogen testing)

Site ID Location Latitude Longitude Control 1 Lakewood Rd. 44.41272 -68.26913 MR 1 Old Farm Rd. 44.3731 -68.19398 Control 2 Hull's Cove Visitor Center 44.40919 -68.25005 MR 2 Hull's Cove Visitor Center 44.40993 -68.24862 Control 3 Bear Brook Picnic Area 44.35991 -68.19917 MR 3 Bear Brook Picnic Area 44.3614 -68.19765 Control 4 Park HQ* 44.37427 -68.26047 MR 4 McFarland Hill* 44.37684 -68.2567 Control 5 Otter Cliff Rd.* 44.33756 -68.20241 MR 5 Miller Gardens Greenhouse* 44.32905 -68.20025 Control 6 Blackwoods Campground 44.3092 -68.20291 MR 6 Fabbri Picnic Area 44.31471 -68.19507 Control 7 Schooner Head Rd. 44.34322 -68.17958 MR 7 Schooner Head Rd. 44.35394 -68.18452 Control 8 Great Head Trail 44.33469 -68.17761 MR 8 Great Head Trail 44.33238 -68.18051

12 Mount Desert Island

Field sites

Treatment 6 Control 6 Treatment 8 Control 8

Control 7 Treatment 7 Control 3 Treatment 3 Treatment 1 Control 4 Treatment 4 Treatment 5 Control 5 Treatment 2 Control 2 Control 1

Figure 4: Map of drag sampling sites. We located 8 M. rubra infestations and paired each with a control site. Sites were paired to account for local variation in I. scapularis.

Drag Sampling

To test the hypothesis that M. rubra presence lowers I. scapularis abundance, we

collected ticks at the eight paired sites by the method of drag sampling (Falco and Fish

1992), whereby a 1 m2 square cloth attached to a wooden rod was pulled across the

13 ground to pick up questing ticks. Dragging was conducted for a period of one man-hour per site, or 1 hour divided by the number of collectors, with the unit of measurement being ticks collected per hour. During dragging, the cloth was inspected for ticks approximately every 30 seconds. No adult ticks were found. Nymphs were removed by hand and transferred into Eppendorf tubes with 70% ethanol. To account for time spend checking the cloth and removing nymphs, an additional 5 minutes were added to each drag period. Larvae were removed with a lint roller at the end of the period, and the lint sheets were stored in Ziploc bags.

Drag sampling took place on July 3, 5, 9, and 10, and August 20 and 21, 2018. Every site was sampled once in July and once in August to represent peak activity of nymphs and larvae, respectively. Paired sites were sampled on the same day. Weather conditions were sunny to partly sunny, with temperatures ranging from 65 to 85 °F.

Small Mammal Track Plates

To test the hypothesis that M. rubra changes small mammal behavior, and thus host- encounter frequency, we deployed track plates to record small mammal prints at control and M. rubra sites. Plates were assembled with instructions from the Cary Institute of

Ecosystem Studies (available at caryinstitute.org). Each plate was taped to a piece of plywood and baited with a small pile of sunflower seeds (Figure 5). Five plates were established in transect at each site on August 20, and they were collected the following day. Plates were not analyzed due to time constraints.

Figure 5: Small mammal track plate with seed bait.

Aggression Bioassay

To test the hypothesis that M. rubra preys on I. scapularis, predation was simulated with a laboratory aggression assay. Two M. rubra nests were collected from Orono, ME

(44.887369, -68.668525) for use in the assay. Each nest contained workers, queens and brood; queens were counted postmortem at the end of the assay. Nests were housed in separate plastic boxes, which were connected with tubing to smaller “arenas” (Figure 6).

We coated the inside walls with Fluon to prevent climbing. Ants were fed with frozen tuna and gauze soaked in 25% sucrose solution, which was provided every two days.

Moisture was provided by pipetting water onto a sponge.

I tested M. rubra aggression against I. scapularis adults, nymphs, and larvae, as well dog tick (Dermacentor variabilis) adults. All ticks were shipped from the Oklahoma State

15 University Tick Rearing Facility and refrigerated inside vials with damp cotton to prevent desiccation. Ant nests were tested alternatingly in coordination with the feeding schedule, which ensured that ants had been starved for at least 48 hours before use in a trial. Before each trial, ants were removed from the arena, the entrance was blocked, and inside surfaces were wiped with a damp sponge and dried with a paper towel to remove lingering pheromones. Ants were gently replaced inside the arena and the entrance was unblocked to allow them to move freely during the trial. Nymphs and adult ticks were tested in groups of three. Due to their small size and low rate of encounter by the ants, larvae were tested in groups of six. Ticks were placed in the center of the arena, and the following behaviors were tallied over 10 minutes: antennation (probing tick with antennae), threat (lunging with mandibles open), biting, carrying/dragging, and stinging.

Each of the tick treatment groups was tested five times for a total of 20 trials. Aggression scores were calculated according to Garnas et al. (2007), with an equation modeled after

De Vroey and Pasteels (1978):

Score = (1 x threats) + (2 x bites) + (2 x carrying) + (3 x stings)

Figure 6: Ant nest boxes with connected arenas.

Pathogen Testing

To test the hypothesis that M. rubra presence lowers NIP, we tested 46 nymphs from our

M. rubra and control sites for B. burgdorferi DNA. Sites were chosen to achieve a balanced number of nymphs (23 each) among control and M. rubra groups. Protocols were supplied by the Hitchner 240 Lichtenwalner Laboratory. Nymphs were bisected with a scalpel and digested with proteinase K, and DNA was extracted, washed and eluted with a QIAGEN DNeasy® Blood & Tissue Kit. One blank was included in each

17 round as a negative control. Borrelia DNA was amplified by primary and nested PCR reactions targeting the 16s ribosomal species specific gene, with forward primer:

5’-ATGCACACTTGGTGTTAACTA-3’, reverse primer (primary and nested):

5’-GACTTATCACCGGCAGTCTTA-3’, and nested forward primer:

5’-CTGGGGAGTATGCTCGCAAGA-3’.

Positive controls included B. burgdorferi DNA (ATCC 35210D-5 Strain 31, genomic

DNA) diluted 1:10,000 in nuclease free water and three positive tick DNA samples from previous projects. Thermocycler incubation consisted of initial denaturing for 2 minutes at 95°C, 35 cycles of denature for 30 seconds at 95°C, anneal for 30 seconds at 55°C, and extension for 1 minute at 72°C, and a final extension for 10 minutes at 72°C. Products were mixed with loading dye, added to a 2% agarose gel along with a small molecular weight ladder, and separated with a 130 v current. Once DNA products were separated, positives were detected with UV imaging.

18 RESULTS

We collected a total of 232 I. scapularis nymphs and 3778 larvae. No adults were encountered, which was expected as we timed our sampling to account for larval and nymphal activity, prior to adult emergence in the fall. Data was visualized in RStudio

(version 1.1.456, 2009-2018 RStudio, Inc.) with a boxplot using the ggplot package, and analysis was completed with a two-way ANOVA, also in RStudio. In accordance with the I. scapularis life cycle, there was a significant drop in nymph abundance from July to

August (P=0.0045, Figure 7). Treatment, however, had almost no effect on nymph abundance (P=0.40, Figure 7). There was a large outlier in the July M. rubra group and a small outlier in the August M. rubra group; nevertheless, the results within each pair of treatments are extremely similar.

19 30

Treatment 20 Control M. rubra Nymphs per hour

July August Month of collection

Figure 7: Nymph abundance by treatment and by time of collection. Note the large outlier in the July M. rubra group. Treatment was not a significant predictor for the number of nymphs collected per hour (P= 0.40).

Aggression scores from the bioassay were calculated in MS Excel, visualized in RStudio with ggplot, analyzed with a two-way ANOVA (Figure 8). We did not initially plan to test the effect of ant nest: however, we found that the degree of aggression was influenced by nest and decided to include it as a variable in the analysis. In comparing I. scapularis adults and D. variabilis adults, species had a marginally significant effect on

20 aggression score (P=0.17). In comparing I. scapularis adults, nymphs and larvae, life stage had a slightly significant effect on aggression score (P=0.053). The effect of ant nest was also slightly significant (P=0.063). Although some aspects of the ants’ behavior were lost in this method of measurement, the ants consistently displayed greater aggression towards the adult life stages, especially towards the D. variabilis adults, and this was reflected in the analysis.

21 60

40 Treatment

D. variabilis adult I. scapularis adult I. scapularis larva I. scapularis nymph Aggression Score 20

1 2 Nest

Figure 8: M. rubra aggression by treatment and by nest. Note that aggression scores were always zero for treatments with nymphs and larvae. Species (i.e. D. variabilis or I. scapularis) was not a significant predictor of aggression (P=0.17). Life stage (i.e. adult, nymph or larva) was a marginally significant predictor of aggression (P=0.053).

22 Eleven nymphs tested positive for B. burgdorferi (Figure 8). Six of the positives were from control sites (Park HQ and Otter Cliff Rd); the other five were from M. rubra sites

(McFarland Hill and Miller Gardens). NIP was calculated by dividing the number of positives for each treatment by the sample size of 23 nymphs. NIP was 0.2609 and

0.2174 for control and M. rubra sites, respectfully (Figure 9). Data was analyzed with a

Z-test for ratio comparison in MS Excel. The P value was not small enough to reject the null hypothesis with 95% confidence (P= 0.73 > α = 0.05), therefore we were unable to claim that NIP differed according to treatment. Although the sample size of 46 was too small to afford a powerful result regardless of the outcome, our results confirmed high B. burgdorferi prevalence at both treatment types.

Figure 9: Images from UV scanning. Each lane contains a ladder and a positive control. Eleven samples tested positive.

Figure 10: Nymphal infection prevalence for control and M. rubra sites. Due to small sample size, the difference between treatments was not significant (P= 0.730).

24 DISCUSSION

The goals of this project were to (1) test the hypothesis that M. rubra decreases I. scapularis abundance by drag sampling for nymphs at 8 pairs of field sites, (2) design a laboratory bioassay to provide a possible explanation for the results of the field study on the basis of aggression and/or predation, (3) test the hypothesis that M. rubra decreases

NIP by extracting DNA from collected nymphs and performing a screen for B. burgdorferi DNA, and (4) measure the abundance of small mammals at field sites using track plates, thus providing a mechanism for the effects of M. rubra on NIP on the basis of changes in host-encounter frequency. The first three objectives were accomplished in their entirety. The fourth was terminated after data collection due to time constraints. I found that M. rubra did not have a significant effect on I. scapularis nymph abundance nor NIP.

Contrary to my first hypothesis, the results from drag collection indicate that M. rubra presence has very little to no effect on I. scapularis nymph abundance. Larvae were omitted from data analysis because they were clustered in patches, unlike nymphs which were distributed relatively homogeneously throughout the sites; however, total larval counts ranged from 0-560 for control sites and from 0-743 for M. rubra sites. The fact that M. rubra never displayed aggressive behavior (i.e. threatening, biting, carrying, stinging) towards nymphs or larvae in the bioassay supports the interpretation that the ants do not prey on these life stages. While the seasonality of the I. scapularis life cycle prevented us from measuring adult abundance, one would expect predation on adult ticks to reduce the number of eggs laid by females, thus reducing the abundance of larvae and

25 nymphs the following year. Considering these observations, I suggest that predation by

M. rubra is opportunistic at best, and unlikely to have any inhibitory effect on I. scapularis population growth. That being said, all of our sites were located in deciduous forest on the east side of ANP, where I. scapularis abundance is comparatively high (Sara

McBride, unpublished data), and it stands to reason that the effect of M. rubra presence could be more pronounced in areas where I. scapularis are less prolific.

It is important to highlight that the results of the laboratory bioassay reflect aggression under experimental conditions. Our goal was to test how fire ant aggression varied in response to different life stages of ticks, and the laboratory design enabled us to test this variable while controlling for factors such as moisture, substrate, pheromone presence, and food for the ants. Of course, the behavior of these species may differ under more natural circumstances, and the extent to which the aggression scores represent natural behavior is difficult to assess. One of the biggest impediments of the assay was the lack of a positive control, which limits our ability to place the aggression scoring system in the context of natural predation. We attempted to use waxworms for this purpose, but the size difference between the worms and the ticks brought the viability of their role as a control organism into question. In trial, the ants swarmed the worms which made it difficult to tally behaviors in the same manner as the tick treatments. Forgoing the control was not entirely detrimental as we were still able to draw comparisons among the tick treatment groups, but it would be worth investigating how these aggression scores compare to those of an organism on which M. rubra readily preys.

The increased aggression towards adult ticks, and dog ticks in particular, may be explained by several mechanisms. All hard-bodied ticks have a sclerotized exoskeleton,

26 which is their primary defense against predators. Burtis and Pflueger (2017) presented 26 different species of spiders and centipedes with I. scapularis nymphs and engorged larvae, and found that engorged larvae were targeted significantly more often than nymphs, and the two most consistent predators were large, wandering spiders. The authors reasoned that smaller predators may be unable to break through the chitinous exoskeleton of the tick. Several other studies have indicated that arthropod predators target engorged ticks more readily than flat ticks under laboratory conditions (reviewed by Samish and Alekessv 2001); not only are engorged ticks easier to penetrate, they are high in nutrients including protein, which may facilitate digestion (Burtis and Pflueger

2017). Although we only tested flat ticks, the same principle of energy reward may be applicable: being the largest tick among the treatments, D. variabilis have the lowest surface-area to volume ratio, and perhaps the fire ants perceived them as a more valuable prey candidate. Myrmica rubra rely almost exclusively on honeydew secretion for carbohydrates, and protein is typically obtained from predation on a variety of other invertebrates (Groden et al. 2005). Presumably, bountiful food availability reduces M. rubra aggression (Garnas et al. 2007), and seems doubtful that a tick would be targeted for consumption unless it were engorged and the habitat were otherwise devoid of protein sources for the ants. It is also worthy to note that ant queens often have significant influence over worker behavior, predominantly with pheromones (Hölldobler and Wilson

1977; Vienne et al. 2010; Vander Meer and Alonso 2002). Nest 1 was generally more aggressive than Nest 2 in trials (Figure 7), and Nest 1 had more queens by chance (both nests contained an unquantified amount of brood), suggesting queens as a possible determinant in the outcome of the assay.

27 With regard to B. burgdorferi, we could not produce sufficient evidence to conclude that M. rubra presence alters NIP. In fact, NIP between control and M. rubra sites was nearly identical (Figure 9). With such a small sample size, our results may not be an accurate testament to the actual NIP in these areas. However, they do confirm the presence of B. burgdorferi at four of our sites (Table 1), and suggest that M. rubra does not affect host-encounter frequency to the extent that we anticipated. Although we were unable to complete a detailed analysis of the track plates, every site had at least one disturbed bait accompanied by mouse footprints, and I estimate that a large number of plates from both treatment groups had fifty or more prints. The plates are an excellent tool for presence/absence measurements; however, they do not provide information about individual mice and their behavior. For instance, one mouse could have left fifty prints on one plate, or ten individuals could have visited the plate separately. We also found that these flat prints differed slightly from impressions in the ground, which added to the difficulty of identifying them. In the future, I would recommend deploying trail cameras in addition to track plates, which would allow abundance of mice to be calculated directly.

Jeffrey Garnas (2004) trapped white footed mice (Peromyscus leucopus) at M. rubra and control sites in ANP, and though the result was not statistically significant, fewer mice were captured at M. rubra sites. Unfortunately, M. rubra have killed animals in live traps (Bruce Connery, personal communication), which presents challenges for tick/small mammal surveillance in ANP and other areas where these species coincide.

Several studies have, however, documented the effects of a second fire ant species, the southern imported fire ant (Solenopsis invicta), on various small mammals (reviewed by

28 Allen et al. 2004). Lechner and Ribble (1996) and Ferris et al. (1998) both found negative correlations between small mammal capture and S. invicta density at select spatial scales.

Capture rates for northern pygmy mice, Baiomys taylori (Killion and Grant 1993; Killion et al. 1995, Castellanos et al. 2016), as well as fulvous harvest mice, Reithrodontomys fulvescens, and cotton rats, Sigmodon hispidus (Castellanos et al. 2016), specifically, have been shown to decrease in the presence of S. invicta. Drops in abundances such as these may be caused by nonconsumptive effects, such as changes in dispersal and foraging, in addition to animal injury and/or predation (Abrams 1984; Fischhoff 2018).

Indeed, Abrams produced evidence that indirect effects between species may have a greater impact on food web ecology than predation. Solenopsis invicta may cause small mammals to relocate their burrows (Killion et al. 1995) and adopt new foraging patterns

(Holtcamp et al. 1997, Orrock and Danielson 2004). Holtcamp et al. (1997) found that in the presence of S. invicta, deer mice increased the frequency of their foraging trips and were more likely to forage in rich patches of food, essentially increasing their feeding efficiency. The authors suggest this may serve as an energy tradeoff to make up for the exertions of ant evasion. Interestingly, Orrock and Danielson (2004) found that S. invicta had the opposite effect on old field mice, Peromyscus polionotus, for seed removal was significantly lowered. Myrmica rubra and S. invicta are both considered highly aggressive in comparison to native ant species; however, while single stings are commonplace for M. rubra (Arevalo and Groden 2007), S. invicta are known to wait until many individuals have climbed onto a victim before delivering multiple stings simultaneously. Presumably the former is associated with a greater risk of anaphylaxis; perhaps this is one reason decreases in small mammal abundance have been attributed to

29 S. invicta and less so to M. rubra. Additionally, as M. rubra activity is highest during the day (Eleanor Groden, personal communication), perhaps nocturnal mammals such as mice are able to forage in infested habitats at night without running serious risk of injury.

Blacklegged ticks are known to quest both diurnally and nocturnally (Durden 1996), so while small mammals may not be present during the day while ants are foraging, they would still find opportunities for host attachment at night.

Myrmica rubra and I. scapularis each cause severe ecological and economic damage in their invaded range, and both species pose serious threats to human health.

Here I have investigated the possible impacts of M. rubra presence on I. scapularis nymph abundance, as well as nymphal infection prevalence through effects on small mammal hosts, in Acadia National Park. Myrmica rubra presence did not have a significant effect on I. scapularis abundance nor NIP. I suggest that predation by M. rubra is possible, though unlikely to occur under natural circumstances. Further research will be required to understand the cascading effects of M. rubra on small mammals— specifically their foraging behavior—and how this may impact host encounter rates and the transmission of B. burgdorferi. This research may be informative for the management of M. rubra and I. scapularis in ANP and other areas of cohabitation.

30 BIBLIOGRAPHY

Abrams, P. A. (1984). Foraging Time Optimization and Interactions in Food Webs. The American Naturalist, 124(1), 80–96.

Allan, B. F., Keesing, F., & Ostfeld, R. S. (2003). Effect of Forest Fragmentation on Lyme Disease Risk. Conservation Biology, 17(1), 267–272.

ALLEN, C. R., Epperson, D. M., & GARMESTANI, A. S. (2004). Red Imported Fire Ant Impacts on Wildlife: A Decade of Research. Https://Doi.Org/10.1674/0003- 0031(2004)152[0088:RIFAIO]2.0.CO;2, 152(1), 88–103.

Alejandro Arevalo, H., and Groden, E. (2007). European fire ant - Myrmica rubra Linnaeus. Retrieved April 19, 2019, from University of Florida website: http://entnemdept.ufl.edu/creatures/urban/ants/Myrmica_ruba.htm

Alonso, L., & Vander Meer, R. (2002). Queen primer pheromone affects conspecific fire ant ( Solenopsis invicta ) aggression. Behavioral Ecology and Sociobiology, 51(2), 122–130.

BERTELSMEIER, C., LUQUE, G. M., & COURCHAMP, F. (2013). Increase in Quantity and Quality of Suitable Areas for Invasive Species as Climate Changes. Conservation Biology, 27(6), 1458–1467.

Burgdorfer, W., & Gage, K. L. (1986). Susceptibility of the black-legged tick, Ixodes scapularis, to the Lyme disease spirochete, Borrelia burgdorferi. Zentralblatt Fur Bakteriologie, Mikrobiologie, Und Hygiene. Series A, Medical Microbiology, Infectious Diseases, Virology, Parasitology, 263(1–2), 15–20.

Burns, E. C., & Melancon, D. G. (1977). Effect of Imported Fire Ant (Hymenoptera: Formicidae) Invasion on Lone Star Tick (Acarina: Ixodidae) Populations. Journal of Medical Entomology, 14(2), 247–249.

Burtis, J. C., & Pflueger, C. (2017). Interactions between soil-dwelling arthropod predators and Ixodes scapularis under laboratory and field conditions. Ecosphere, 8(8), e01914.

Castellanos, A. A., Medeiros, M. C. I., Hamer, G. L., Morrow, M. E., Eubanks, M. D., Teel, P. D., … Light, J. E. (2016). Decreased small mammal and on-host tick abundance in association with invasive red imported fire ants ( Solenopsis invicta ). Biology Letters, 12(9), 20160463.

31 Chapin III, F. S., Zavaleta, E. S., Eviner, V. T., Naylor, R. L., Vitousek, P. M., Reynolds, H. L., … Díaz, S. (2000). Consequences of changing biodiversity. Nature, 405(6783), 234–242.

De Vroey, C., & Pasteels, J. M. (1978). Agonistic behaviour ofMyrmica rubra L. Insectes Sociaux, 25(3), 247–265.

DeFisher, L. E., & Bonter, D. N. (2013). Effects of Invasive European Fire Ants (Myrmica rubra) on Herring Gull (Larus argentatus) Reproduction. PLoS ONE, 8(5), e64185.

Durden, L. A., Vogel, G. N., & Oliver, J. H. (1996). Nocturnal Questing by Adult Blacklegged Ticks, Ixodes scapularis (Acari: Ixodidae). The Journal of Parasitology, 82(1), 174.

Eisen, R. J., Eisen, L., & Beard, C. B. (2016). County-Scale Distribution of Ixodes scapularis and Ixodes pacificus (Acari: Ixodidae) in the Continental United States. Journal of Medical Entomology, 53(2), 349–386.

Falco, R. C., & Fish, D. (1992). A comparison of methods for sampling the deer tick,Ixodes dammini, in a Lyme disease endemic area. Experimental & Applied Acarology, 14(2), 165–173.

Ferris, D. K., Killion, M. J., Ferris, K. P., Grant, W. E., & Vinson, S. B. (n.d.). Influence of Relative Abundance of Red Imported Fire Ants (Solenopsis Invicta) on Small Mammal Captures. The Southwestern Naturalist, Vol. 43, pp. 97–100.

Fischhoff, I. R., Burtis, J. C., Keesing, F., & Ostfeld, R. S. (2018). Tritrophic interactions between a fungal pathogen, a spider predator, and the blacklegged tick. Ecology and Evolution, 8(16), 7824–7834.

Garnas, J. (2004) European Fire Ants on Mount Desert Island, Maine: Population Structure, Mechanisms of Competition and Community Impacts of Myrmica Rubra L. (Hymenoptera: Formicidae). Electronic Theses and Dissertations. 378. https://digitalcommons.library.umaine.edu/etd/378

Garnas, J., Groden, E., & Drummond, F. A. (2014). Mechanisms of Competitive Displacement of Native Ant Fauna by Invading <I>Myrmica rubra</I> (Hymenoptera: Formicidae) Populations. Environmental Entomology, 43(6), 1496–1506.

Githeko, A. K., Lindsay, S. W., Confalonieri, U. E., & Patz, J. A. (2000). Climate change and vector-borne diseases: a regional analysis. Bulletin of the World Health Organization, 78(9), 1136–1136.

32 Groden, E., Drummond, F. A., Garnas, J., & Franceour, A. (2005). Distribution of an Invasive Ant, <I>Myrmica rubra</I> (Hymenoptera: Formicidae), in Maine. Journal of Economic Entomology, 98(6), 1774–1784.

Holt, R. D. (2003) On the evolutionary ecology of species' ranges. Evolutionary Ecology Research, 5: 159–178.

Holtcamp, W. N., Grant, W. E., & Vinson, S. B. (1997). Patch use under predation hazard: effect of the red imported fire ant on deer mouse foraging behavior. Ecology, 78(1), 308–318.

Hölldobler, B., & Wilson, E. O. (1977). The number of queens: An important trait in ant evolution. Naturwissenschaften, 64(1), 8–15.

Johnson, T. L., Graham, C. B., Boegler, K. A., Cherry, C. C., Maes, S. E., Pilgard, M. A., … Eisen, R. J. (2017). Prevalence and Diversity of Tick-Borne Pathogens in Nymphal Ixodes scapularis (Acari: Ixodidae) in Eastern National Parks. Journal of Medical Entomology, 54(3), 742–751.

Kenis, M., Auger-Rozenberg, M.-A., Roques, A., Timms, L., Péré, C., Cock, M. J. W., … Lopez-Vaamonde, C. (2009). Ecological effects of invasive alien insects. Biological Invasions, 11(1), 21–45.

Killion, M. J., & Grant, W. E. (1993). Scale Effects in Assessing the Impact of Imported Fire Ants on Small Mammals. The Southwestern Naturalist, 38(4), 393.

Killion, M. J., Grant, W. E., & Vinson, S. B. (1995). Response of Baiomys taylori to Changes in Density of Imported Fire Ants. Journal of Mammalogy, 76(1), 141– 147.

Lechner, K. A., & Ribble, D. O. (n.d.). Behavioral Interactions between Red Imported Fire Ants (Solenopsis invicta) and Three Rodent Species of South Texas. The Southwestern Naturalist, Vol. 41, pp. 123–128.

Liu, B., Gao, X., Ma, J., Jiao, Z., Xiao, J., Hayat, M. A., & Wang, H. (2019). Modeling the present and future distribution of arbovirus vectors Aedes aegypti and Aedes albopictus under climate change scenarios in Mainland China. Science of The Total Environment, 664, 203–214.

Lyme Disease Maps: Most Recent Year | Lyme Disease | CDC. (n.d.). Retrieved March 12, 2019, from https://www.cdc.gov/lyme/datasurveillance/maps-recent.html

McPhee, K., Garnas, J., Drummond, F., & Groden, E. (2012). Homopterans and an Invasive Red Ant, Myrmica rubra (L.), in Maine. Environmental Entomology, 41(1), 59–71.

33 Ogden, N. H., Bigras-Poulin, M., O’Callaghan, C. J., Barker, I. K., Lindsay, L. R., Maarouf, A., … Charron, D. (2005). A dynamic population model to investigate effects of climate on geographic range and seasonality of the tick Ixodes scapularis. International Journal for Parasitology, 35(4), 375–389.

Ogden, N. H., Lindsay, L. R., Hanincová, K., Barker, I. K., Bigras-Poulin, M., Charron, D. F., … Thompson, R. A. (2008). Role of migratory birds in introduction and range expansion of Ixodes scapularis ticks and of Borrelia burgdorferi and Anaplasma phagocytophilum in Canada. Applied and Environmental Microbiology, 74(6), 1780–1790.

Orrock, J. L., & Danielson, B. J. (2004). Rodents balancing a variety of risks: invasive fire ants and indirect and direct indicators of predation risk. Oecologia, 140(4), 662–667.

Ostfeld, R. S., & Keesing, F. (2000). Biodiversity series: The function of biodiversity in the ecology of vector-borne zoonotic diseases. Canadian Journal of Zoology, 78(12), 2061–2078.

Ouellette, G. D., Drummond, F. A., Choate, B., & Groden, E. (2010). Ant Diversity and Distribution in Acadia National Park, Maine. Environmental Entomology, 39(5), 1447–1456.

Parmesan, C., & Yohe, G. (2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature, 421(6918), 37–42.

Piesman, J., & Sinsky, R. J. (1988). Ability of Ixodes scapularis, Dermacentor variabilis, and Amblyomma americanum (Acari: Ixodidae) to Acquire, Maintain, and Transmit Lyme Disease Spirochetes (Borrelia burgdorferi). Journal of Medical Entomology, 25(5), 336–339.

Rand, P. W., Lacombe, E. H., Dearborn, R., Cahill, B., Elias, S., Lubelczyk, C. B., … Smith, R. P. (2007). Passive Surveillance in Maine, an Area Emergent for Tick- Borne Diseases. Journal of Medical Entomology, 44(6), 1118–1129.

RANDOLPH, S. E. (2004). Tick ecology: processes and patterns behind the epidemiological risk posed by ixodid ticks as vectors. Parasitology, 129(7), S37– S65.

Rust, M. K., & Su, N.-Y. (2012). Managing Social Insects of Urban Importance. Annual Review of Entomology, 57(1), 355–375. https://doi.org/10.1146/annurev-ento- 120710-100634

Samish, M., & Alekseev, E. (2001). Arthropods as Predators of Ticks (Ixodoidea). Journal of Medical Entomology, 38(1), 1–11. https://doi.org/10.1603/0022-2585- 38.1.1

Samish, M., Ginsberg, H., & Glazer, I. (2008). Anti-tick biological control agents: Assessment and future perspectives. In Ticks: Biology, Disease and Control.

SANDERS, C. J., MELLOR, P. S., & WILSON, A. J. (2010). Invasive arthropods. Revue Scientifique et Technique de l’OIE, 29(2), 273–286.

Shope, R. (1991). Global Climate Change and Infectious Diseases. Environmental Health Perspectives, 96, 171.

Soucy, J.-P. R., Slatculescu, A. M., Nyiraneza, C., Ogden, N. H., Leighton, P. A., Kerr, J. T., & Kulkarni, M. A. (2018). High-Resolution Ecological Niche Modeling of Ixodes scapularis Ticks Based on Passive Surveillance Data at the Northern Frontier of Lyme Disease Emergence in North America. Vector-Borne and Zoonotic Diseases, 18(5), 235–242.

Subak, S. (2003). Effects of Climate on Variability in Lyme Disease Incidence in the Northeastern United States. American Journal of Epidemiology, 157(6), 531–538.

Vienne, C., Errard, C., & Lenoir, A. (2010). Influence of the Queen on Worker Behaviour and Queen Recognition Behaviour in Ants. Ethology, 104(5), 431–446.

Wiens, J. J., Ackerly, D. D., Allen, A. P., Anacker, B. L., Buckley, L. B., Cornell, H. V., … Stephens, P. R. (2010). Niche conservatism as an emerging principle in ecology and conservation biology. Ecology Letters, 13(10), 1310–1324.

Wiens, J. J., & Graham, C. H. (2005). Niche Conservatism: Integrating Evolution, Ecology, and Conservation Biology. Annual Review of Ecology, Evolution, and Systematics, 36(1), 519–539.

Zingg, S., Dolle, P., Voordouw, M. J., & Kern, M. (2018). The negative effect of wood ant presence on tick abundance. Parasites & Vectors, 11(1), 164.

35 AUTHOR’S BIOGRAPHY

I am the daughter Susan Holmes and Frank Guarneri and the older sister of two brothers,

Vann and Martin. I grew up in an old white farm house in Belgrade, Maine, where I learned how to climb trees, catch frogs and raise chickens. I attended Messalonskee High

School and will forever be a proud Eagle. I will graduate from the University of Maine with a bachelor’s degree in Biology and a minor in Spanish. I enjoy triple jumping, dancing, baking, playing with my cats, and doing just about anything outside. I credit the time I spent outdoors as a kid with my passion for studying insects. After I graduate, I plan to apply to the Peace Corps.