Draft · 2018-11-27 · Draft 48 Keywords: reptile, body-mass index, movement ecology, parasite, gps tracking, host-parasite 49 interaction, spatial behaviour, sleepy lizard, Tiliqua

Draft

Endure your parasites: sleepy lizard movement is notaffected by their ectoparasites

Journal: Canadian Journal of Zoology

Manuscript ID cjz-2017-0352.R1

Manuscript Type: Article

Date Submitted by theAuthor: 18-Mar-2018

Complete List of Authors: Taggart, Patrick; The University of Adelaide, School of Animaland Veterinary Science; University of AdelaideLeu, Stephan; Macquarie University Department of BiologicalSciencesSpiegel, Orr; Ben Gurion University of the Negev, MitraniDepartment of Desert EcologyGodfery, Stephanie; University of Otago, Department ofZoologySih, Andrew; University of California, Davis, EnvironmentalScience and Policy Dept.Bull, C; Flinders University, College of Science and Engineering

Is your manuscript invitedfor consideration in a

Special Issue?:Not applicable (regular submission)

Keyword: reptile, body-mass index, movement ecology, parasite, gpstracking, host-parasite interaction, spatial behaviour

https://mc06.manuscriptcentral.com/cjz-pubs

Canadian Journal of Zoology

Draft

1 Endure your parasites: sleepy lizard movement is not affected by their ectoparasites

2

3 Patrick L Taggarta,b, Stephan T Leua,c, Orr Spiegeld, Stephanie S Godfreye, Andrew Sihf and C Michael Bulla

4

5 aCollege of Science and Engineering, Flinders University

6 bcurrent address: School of Animal and Veterinary Science, University of Adelaide

7 ccurrent address: Department of Biological Sciences, Macquarie University

8 dSchool of Zoology, Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel

9 eDepartment of Zoology, University of Otago, Dunedin, New Zealand

10 fDepartment of Environmental Science and Policy, University of California, Davis, U.S.A.

11

12

13 Corresponding author:

14 Patrick Taggart

15 Email: [email protected]

16 Address: University of Adelaide, Veterinary Health Centre, Building E40 Mudla Wirra Rd,

17 Roseworthy SA, 5371

18

19 Author contributions:

20 PLT wrote, drafted and revised manuscript; STL conducted statistical analyses, jointly wrote,

21 drafted and revised manuscript, jointly conceived research questions, and co-supervised project;

22 OS drafted and revised manuscript; SSG collected data, drafted and revised manuscript; AS

23 drafted and revised manuscript; CMB primary supervisor of project, sourced funding and jointly

24 conceived research questions.

25

Page 1 of 26



mailto:[email protected]

Draft

26

27

28 Short title:

29 Ectoparasites does not affect lizard movement

30

31 Abstract

32 Movement is often used to indicate host vigour, as it has various ecological and evolutionary

33 implications, and has been shown to be affected by parasites. We investigate the relationship

34 between tick load and movement in the Australian sleepy lizard (Tiliqua rugosa Grey, 1825)

35 using high resolution GPS tracking. This allowed us to track individuals across the entire activity

36 season. We hypothesised that tick load negatively affects host movement (mean distance moved

37 day-1). We used a multivariate statistical model informed by the ecology and biology of the host

38 and parasite, their host-parasite relationship, and known host movement patterns. This allowed

39 us to quantify the effects of ticks on lizard movement above and beyond effects of other factors

40 such as time in the activity season, lizard body condition and stress. We did not find any support

41 for our hypothesis. Instead, our results provide evidence that lizard movement is strongly driven

42 by internal state (sex and body condition independent of tick load), and by external factors

43 (environmental conditions). We suggest that the sleepy lizard has largely adapted to natural

44 levels of tick infection in this system. Our results conform to host-parasite arms race theory,

45 which predicts varying impacts of parasites on hosts in natural systems.

46

47

Page 2 of 26



Draft

48 Keywords: reptile, body-mass index, movement ecology, parasite, gps tracking, host-parasite

49 interaction, spatial behaviour, sleepy lizard, Tiliqua rugosa

50

51

52 Introduction

53 Parasites, by definition, gain their energy resources from their hosts, which negatively impacts

54 the host’s health and vigour (Beyer and Grossman 1997; Fenner and Bull 2008). Such impacts in

55 host-parasite relationships drive selection for host defence mechanisms against parasites and

56 parasite specialisation to overcome host defences, resulting in a host-parasite arm race (Dawkins

57 and Krebs 1979). These opposing selection pressures are key drivers of evolutionary and

58 ecosystem processes within host-parasite interactions (Thomas et al. 1999; Brockhurst et al.

59 2004; Nunn et al. 2004; Martin et al. 2011). The impact parasites have on their hosts varies

60 substantially, ranging from behavioural changes, for instance when parasites manipulate their

61 hosts to increase transmission probability (Poulin 2010, 2013), to population growth regulation

62 (Anderson 1978), substantial reductions in host health (e.g., paralysis (Hamat et al. 2017)),

63 reduced reproductive success (Marzal et al. 2005) and increased mortality (Fraser et al. 2016).

64 Movement, is a measure often used to indicate host vigour (Angelier et al. 2007; Warren et al.

65 2011) and has been shown to be affected by parasite load (Fenner and Bull 2008). Here we

66 investigate the relationship between ectoparasites and vigour in a lizard-tick system.

67

68 Animal movement drives various ecological processes like dispersal, foraging and species-co-

69 existence (Nathan et al. 2008; Hansson and Åkesson 2014). Thus, understanding how parasites

70 affect movement of their host is essential for broad topics in ecology, with implications for

Page 3 of 26



Draft

71 management and conservation (Poulin 2010; Kays et al. 2015; Dougherty et al. 2017).

72 Furthermore, a feedback loop between host movement and parasite infection may exist. Before

73 infection, increased host movement can increase the probability of parasite exposure, particularly

74 for parasites that have a free-living infectious stage in the off-host environment or socially

75 transmitted parasites. For example, dispersal events have been suggested to increase parasite and

76 disease spread in both animals and humans (White et al. 2000; Russell et al. 2005; Viboud et al.

77 2006). Post-infection, parasites can decrease movement by reducing host vigour. For example,

78 O’Dwyer et al. (2014) found a trend towards reduced movements in intertidal snails infected

79 with philophthalmid trematode parasites. Similarly, lizards with experimentally reduced parasite

80 burdens have been shown to move around for longer time periods during observation sessions

81 (Main and Bull 2000; Fenner and Bull 2008). Conversely, there is also evidence that host

82 manipulation by parasites can enhance certain movements or overall activity (Poulin 2013), for

83 example risk taking behaviour in rodents towards cats (Médoc and Beisel 2008) or risky displays

84 by fish infested by trophically transmitted parasites (Lafferty and Morris 1996; Sato et al. 2012).

85 Understanding this feedback between parasite load and host movement within a species (Leung

86 and Koprivnikar 2016) can provide valuable insights into the host-parasite relationship. If there

87 is a negative feedback loop (e.g., where increased movement increases parasite loads, but higher

88 parasite load decreases movement), the system would tend towards intermediate parasite loads,

89 whereas a positive feedback loop (e.g., where increased movement increases parasite loads and

90 parasites cause hosts to move more (e.g., with parasite manipulation of host behaviour (Berdoy

91 et al. 2000), the feedback loop can result in an explosive accumulation of parasites in some hosts.

92

Page 4 of 26



Draft

93 Many of the above mentioned relationships and examples (but not all) have been demonstrated in

94 the laboratory, but not the field. Here, we aim to fill this gap by working in-situ to upscale these

95 findings to a natural population of sleepy lizards (Tiliqua rugosa Gray, 1825). We focus on the

96 effect of parasite load on individual host movement. Working in South Australia, we tracked the

97 movement of free-ranging sleepy lizards infested with two species of parasitic ticks. Previous

98 studies of the effect of parasites in this system supported two complementary hypotheses. Bull

99 and Burzacott (1993) reported a positive relationship between lizard body condition and tick

100 load, and suggested that high quality home ranges simultaneously benefit both lizards and ticks,

101 driving this relationship, and supporting the favourable patch hypothesis. Later, Main and Bull

102 (2000) found evidence for the ‘reduced vigor’ hypothesis by experimentally manipulating lizard

103 tick loads and showing that tick infestation reduced endurance and daily movements in sleepy

104 lizards. While this work demonstrates that heavy tick loads can have a negative effect on lizards,

105 we still do not know whether this effect is present under natural conditions where infestation

106 levels are generally lower, and lizards can compensate for lower vigor by obtaining greater

107 energy resources in high quality home ranges.

108

109 Due to the wealth of knowledge surrounding the sleepy lizard – tick system, and following the

110 results of Main and Bull (2000), it is unlikely that ticks increase lizard movement to facilitate

111 their transmission. Consequently, we formulate three predictions: 1) under natural conditions

112 ticks negatively affects lizard movement activity; 2) the effects of ticks on lizard movement

113 increases with decreasing host body condition (Beldomenico et al. 2008; Beldomenico and

114 Begon 2010), such as later in the season when individuals experience particularly harsh

115 environmental conditions (hot and dry) and food resources have dried out; and 3) tick abundance

Page 5 of 26



Draft

116 in the environment, and hence tick load on the lizard, increases over time, further enhancing the

117 influence of ticks on lizard movement later in the season. To test these predictions, we followed

118 the movement behaviour of 55 adult sleepy lizards using GPS tracking devices and

119 simultaneously quantified their naturally occurring tick loads on a fortnightly basis. We then

120 built a multivariate statistical model that incorporated tick-host relationships known from

121 previous studies in this species (and which we detail below) to address these predictions.

122

123

124 Methods

125 Study site

126 This study was conducted in a 1.5km2 study site near Bundy Bore station in South Australia’s

127 mid-north (33°54’16”S, 139°20’43”E) (Figure 1(A)). This semi-arid region receives an average

128 annual rainfall of about 250 mm (Bull and Baghurst 1998) and its vegetation consists primarily

129 of chenopod shrubland (Bull and Pamula 1998; Leu et al. 2011a). Average monthly high

130 temperatures in winter are ~15 oC and in summer ~32 oC. Winter temperatures regularly drop

131 below 10 oC and summer temperatures regularly climb above 40 oC. In 2009, when our study

132 was conducted, monthly winter and summer temperatures were above average, but annual

133 rainfall equal to the average for the region.

134

135 Host parasite system

136 The host species, the sleepy lizard is a large scincid lizard, with adult snout to vent lengths of up

137 to 350 mm and body mass of up to 1000 g (Bull and Pamula 1996). Adult lizards maintain stable

138 home ranges of approximately 4 ha, with little spatial shift over time (Bull 1995; Bull and

Page 6 of 26



Draft

139 Baghurst 1998; Bull and Freake 1999; Kerr and Bull 2006a). Lizards in the focal population use

140 bushes (Maireana sedifolia, and other species), coarse woody or human (corrugated iron sheets)

141 debris, wombat and rabbit warrens as overnight refuges and to avoid adverse weather conditions

142 (Kerr et al. 2003; Leu et al. 2011a).

143

144 At our study site, the sleepy lizard is parasitised by two ectoparasitic tick species, Amblyomma

145 limbatum Neumann, 1899 and Bothriocroton hydrosauri Denny, 1843 (Figure 1(C)). Both tick

146 species are three host ticks, with each life stage, larvae, nymphs and adults, requiring a host

147 individual. All life stages (except adult males) engorge with blood and or lymph while they are

148 attached to their host (Bull and Sharrad 1980). When ticks are fully engorged they detach from

149 their hosts, mostly in the lizards refuges, to moult into the next life stage (Petney et al. 1983;

150 Chilton and Bull 1993b; Kerr and Bull 2006b). Favourable environmental conditions in these

151 host refuges reduce the risks of desiccation for both tick species (Bull et al. 1988; Chilton and

152 Bull 1993a, b). Larvae and nymphs take some time to moult (8-72 days depending on

153 temperature and relative humidity) into the next life stage, during which they are not infectious,

154 before attaching to the next host individual (Chilton et al. 2000). Both tick species rely on host

155 movement and their reuse of refuges to attach to subsequent hosts and do not actively seek out

156 hosts (Petney et al. 1983). Subsequent hosts may be the same or different individuals that use the

157 same refuge. Sleepy lizards repeatedly use a set of overnight refuges, where ticks are transmitted.

158 Furthermore, Leu et al. (2010b) and Wohlfiel et al. (2013) showed that transmission networks

159 based on asynchronous refuge sharing among neighbours predicts individual tick load.

160

161 Movement and tick infestation

Page 7 of 26



Draft

162 We captured most of the resident adult lizards (29 males and 26 females) in our study site, and

163 observed their movement throughout the main activity period (Bull and Baghurst 1998),

164 September 2009 – January 2010. We attached a data logger to the dorsal surface of each lizard’s

165 tail using surgical adhesive tape (Leu et al. 2010b) (Figure 1(B)). Data loggers included a GPS

166 unit to record lizard movement, a radio transmitter to allow the recapture of lizards and regular

167 tick counts, and a step counter. GPS locations were recorded every 10 mins when lizards were

168 active and moving, that is when they took at least one step in the previous 10 min. The GPS

169 devices had a median horizontal accuracy of +/- 6m (Leu et al. 2010a). Each data logger (GPS,

170 radio transmitter and step counter units) weighed 37 g, representing 4.9 % of an average 750 g

171 adult lizard body mass. Based on the GPS data, we calculated for each individual, the total

172 distance moved per day (sum of all distances between consecutive locations on that day) and

173 subsequently for each lizard the mean daily distance moved per fortnight (Figure 1(D)). Sleepy

174 lizard movement has previously been quantified by a number of different measures, including

175 steps taken per day (Leu and Bull 2016); daily activity (total number of 10 min intervals during

176 which lizards were recorded active divided by the number of days observed) and daily distance

177 moved (total distance moved in metres divided by the number of days observed) (Leu et al.

178 2016a); movement activity (distance between consecutive GPS locations) and duration (Leu et

179 al. 2011b); and cotton spooling, sprint speed, endurance, active/inactive behaviour (Main and

180 Bull 2000), all of which measure different aspects of lizard activity. Here our focus was on the

181 mean daily distance moved by lizards on active days, similar to previous studies (Leu et al.

182 2016a), hence we excluded days on which lizards did not move, i.e. either no steps were taken

183 and hence no GPS locations were recorded, or the total daily distance travelled was 0 m.

184

Page 8 of 26



Draft

185 Individual lizards could be identified by their unique radio transmitter frequency. We recaptured

186 lizards once every fortnight to record tick loads (number, species and life stage), take

187 morphological measurements and download GPS data. Lizard handling periods (a standard two

188 hour period) were excluded from the GPS data, and lizards released at their point of capture.

189 Handling periods accounted for the period of time lizards were actively handled (~30 mins) and

190 the period of time over which lizard movement was expected to be impacted as a consequence of

191 handling (~1 hour (Kerr et al. 2004)). Larvae, nymphs and adult females have an increasing

192 impact on the hosts because they draw increasing amounts of blood and lymph. Males do not

193 engorge with blood. Following Main (1999), we quantified the differential impact of each life

194 stage and converted each lizard’s fortnightly tick load to a larvae equivalent index. One nymph

195 equalled ten larvae, one adult female equalled 250 larvae, and one adult male equalled one larva

196 (Main 1999). Fortnightly morphological measurements included lizard mass (to the nearest 5 g),

197 and snout-vent length (to the nearest 5 mm). We log transformed these fortnightly morphological

198 measures and used them in a mass on snout-vent length linear regression. We then used the

199 unstandardized residuals from this regression analysis as a measure of each lizard’s fortnightly

200 body condition (Moore et al. 2009). These morphological measures were regressed in one

201 analysis including all lizards and all fortnights, as this provided a better perspective of body

202 condition across the activity season and population.

203

204 The first time lizards were captured we also determined each individual’s sex by assessing head

205 morphology, males have substantially longer jaws and broader heads than females (Bull and

206 Pamula 1996). Male sex was confirmed by gently everting their hemipenes where possible.

207

Page 9 of 26



Draft

208 After the study we removed all data loggers and released all lizards. We observed no damage or

209 irritation to the skin where data loggers were attached and lizards naturally shed their skin in the

210 following months. All lizards were treated using procedures formally approved by the Flinders

211 University Animal Welfare Committee in compliance with the Australian Code of Practice for

212 the Use of Animals for Scientific Purposes and conducted under the South Australian

213 Department of Environment and Natural Resources Permit to Undertake Scientific Research.

214

215 Model construction

216 We constructed our multivariate statistical model informed by previous knowledge of the

217 ecology of the sleepy lizard and its ecto-parasitic ticks as well as their host-parasite relationship.

218 This allowed us to investigate the effects of ticks on lizard movement above and beyond effects

219 of other factors that are known to affect movement. For each fortnight, we calculated the mean

220 daily distance travelled by each lizard, which was our dependent variable. This variable was

221 square root transformed to meet model assumptions (normality, homoscedasticity). Then, to test

222 our first prediction we included total larvae equivalent as a measure of tick load and as one of

223 our independent variables into the model.

224

225 Lizard movement activity first increases, then decreases over the course of the season (Firth and

226 Belan 1998; Kerr and Bull 2006c). At the beginning and end of the activity season lizard

227 movement is low due to cool ambient temperatures and low food resources, respectively. We

228 accounted for this non-linear relationship by including a time variable, fortnight, and fortnight2

229 into our model (both continuous). Lizard sex was included as a fixed categorical factor, because

230 male sleepy lizards have been demonstrated to have higher levels of movement activity than

Page 10 of 26



Draft

231 females (Leu et al. 2011b). Movement activity also varies with body condition (Leu and Bull

232 2016), which we included in our model. We also included a random intercept and slope for each

233 individual, (fortnight | lizard id), to account for individual differences in movement activity

234 (intercept), as well as individual differences in movement change over time (slope).

235

236 In the sleepy lizard, tick load is known to vary with body condition (Bull and Burzacott 1993),

237 and hence we included the interaction term total larvae equivalent x body condition to test our

238 second prediction that the impact of ticks on lizard movement increases with decreasing host

239 condition. Lastly, we included the interaction term total larvae equivalent x fortnight to test our

240 third prediction that lizard movement is more strongly affected late in the season because tick

241 abundance in the environment and hence tick load on the lizard increases over time. All

242 continuous independent variables were centred and scaled to allow us to determine their

243 importance relative to one another in the statistical model. We used the “vif.mer” function (Frank

244 2011) to check for collinearity using variance inflation factors (VIFs), which were generally very

245 low (< 1.07) and below a pre-selected threshold (Zuur et al. (2010) suggest a cut-off of 3). As

246 expected, VIFs were higher for fortnight (20.33) and fortnight2 (20.36), due to the polynomial

247 structure of the model, with a clear relationship between the two variables.

248

249 Our final model was as follows:

250 √mean daily distance = sex + fortnight + fortnight2 + total larvae equivalent + body condition +

251 total larvae equivalent x body condition + total larvae equivalent x fortnight + (fortnight | lizard

252 id)

253

Page 11 of 26



Draft

254 All statistical analyses and data manipulation was conducted in IMB SPSS Statistics 22 (Corp

255 2013), or R (R Core Team 2016), using mostly lme4 (Bates et al. 2015), lmerTest (Kuznetsova et

256 al. 2016), and ggplot2 (Wickham 2009) to produce graphs.

257

258

259 Results

260 Over the duration of this study, the mean number of location records per lizard was 1728 (95%

261 CI 1617-1838), with lizards monitored for a mean of 5.42 (95% CI 5.08-5.8) fortnights over the

262 duration of the season. On average lizards moved a mean daily distance of 215 m (95% CI 203-

263 228m) in a given fortnight over the duration of the season. Mean daily distance moved in a given

264 fortnight ranged from 3-745m/day. Lizard tick loads varied by fortnight, ranging from 0-801

265 larval units, with a mean of 64 (95% CI 47-81).

266

267 We found no evidence to support any of our three predictions on tick load effects; tick load did

268 not affect lizard movement; the effect of tick load on lizard movement was not host condition

269 dependent; and the effect of tick load on lizard movement did not depend on the time in the

270 lizard activity season (Table 1). Our model accounted for lizard ID and fortnight as random

271 effects (intercept and slope (fortnight | lizard id)). Lizard ID explained twice as much of the

272 variance in the intercept than fortnight in the slope (0.15 and 0.08, respectively), suggesting

273 lizards differed from each other more strongly (intercept) than their observed temporal variation

274 (slope) over the duration of the season. Consistent with previous studies, lizard movement varied

275 over the course of the activity season (Table 1). Fortnight had a positive effect, and fortnight2 a

276 negative effect, implying lower movement at the start of the season and that lizards also walked

Page 12 of 26



Draft

277 less as the season progressed. Lizard movement increased with body condition (Table 1). Female

278 lizards additionally moved shorter mean distances per day than males (Table 1 and Figure 2).

279 Interaction effects included in the model were not significant (Table 1).

280

281

282 Discussion

283 Summary of main result

284 Our study is based on a rich dataset using GPS data loggers attached to lizards which recorded

285 host movement in greater detail than previous studies investigating similar questions, and

286 combining this with repeated measures of lizard tick loads. In contrast to our predictions, we

287 found no evidence that tick load negatively affected lizard travel distance, and the effect of tick

288 load was not dependent on host condition or the time in the lizard activity season. Our

289 interpretation of this results is that lizards have adapted to the natural level of tick infection, and

290 that the average intensity (number of individual ticks on a given lizard) of ticks is too low to

291 have a measurable effect on movement patterns. Chronic stress is known to negatively impact

292 immune system function (Dhabhar 2014). Consequently, we predicted that lizard vigour

293 (movement) is more strongly affected during periods of chronic stress, for example at low body

294 conditions or high tick loads, which may occur later in the season due to harsh environmental

295 conditions, poorer quality food resources and increased tick abundance in the environment.

296 However, we found no evidence for these predictions. A possible explanation for the lack of tick

297 effects shown in our study could be that both ticks and lizards are affected by environmental

298 conditions in a similar manner (both negatively affected by harsh and positively affected by good

299 environmental conditions). For example, increased temperature and decreased humidity shortens

Page 13 of 26



Draft

300 the time to desiccation for ticks and decreases tick survival when waiting for the next host (Bull

301 et al. 1988; Chilton and Bull 1993a, b). Similarly, temperature and humidity have been shown to

302 influence sleepy lizard activity and movement (Firth and Belan 1998; Kerr and Bull 2006c).

303

304 Comparison with earlier findings from the sleepy lizard – tick model system

305 Our results, demonstrating no impact of ticks on the movement activity of lizard hosts, contrast

306 previous work by Main and Bull (2000) who experimentally manipulated lizard tick load in the

307 field and laboratory, and monitored lizard movements and activity levels. The conclusion of their

308 study was that ticks reduced the daily distance lizards moved, home range size, endurance and

309 sprint speed, and altered basking/active behaviour. However, these effects could not be

310 consistently identified across study sites, lizard age classes, or years (Main and Bull 2000). We

311 argue that increasing tick loads beyond natural levels may increase the level of stress

312 experienced by lizards in a similar manner to a reduction in body condition and food resource

313 availability, as ticks consume host resources. Thus, although the results of Main and Bull (2000)

314 are not consistent with ours or across seasons and sites in their study, their conclusions conform

315 to our original predictions. Whilst the work by Main and Bull (2000) suggests that under some

316 circumstances ticks can impact on lizard movement, our results suggest that overall, under

317 natural conditions, this is not the case.

318

319 Parasite effects on movement in other systems

320 Several studies in other systems also found no effect of parasites on host movement (Hillegass et

321 al. 2010; Mayer et al. 2015; Nelson et al. 2015). For example, Mayer et al. (2015) found no

322 difference in swimming speed of adult keelback snakes (Tropidonophis mairi Jan, 1863) before

Page 14 of 26



Draft

323 and after the removal of the gastrointestinal nematode Tanqua anomala Linstow, 1904.

324 Similarly, Nelson et al. (2015) found no relationship between a range of movement measures

325 (distance travelled per minute, voluntary movements, movements towards prey items, sprint

326 speed and endurance) and lungworm (Rhabdias hylae Johnston and Simpson, 1942) infection in

327 multiple anuran species (Rhinella marina Linnaeus, 1758, Cyclorana Australis Gray, 1842,

328 Limnodynastes convexiusculus MacLeay, 1877 and Litoria nasuta Gray, 1842). The host-parasite

329 arms race clearly suggests that the effects of parasites on their hosts in natural systems are on a

330 scale from large to small (Dawkins and Krebs 1979). Consequently, studies reporting no impact

331 of parasites on hosts are important as they suggest that in some systems hosts have adapted to

332 natural levels of parasite infection (Mayer et al. 2015; Nelson et al. 2015), at least with regard to

333 the measure of interest.

334

335 Nevertheless there are studies which have clearly shown that parasites can reduce host

336 movements. For example, Garrido and Pérez-Mellado (2013) found haemogregarine blood

337 parasite infestation to reduce sprint speed in wild caught Lilford’s wall lizards (Podarcis lilfordi

338 Günther, 1874). They suggested that this is the result of the blood parasite reducing haemoglobin

339 concentrations and subsequently lowering the lizards’ capacity for oxygen transportation to

340 muscle tissues, thereby reducing sprint speed (Garrido and Pérez-Mellado 2013). In contrast,

341 there are also reports of parasites increasing host movement. Médoc and Beisel (2008) found that

342 acanthocephalan (Polymorphus minutus Goeze, 1782) infection increased both swimming speed

343 and distance covered at the beginning of an escape attempt (first 0.5 sec) in male amphipods

344 (Gammarus roeseli Gervais, 1835). They suggested that the acanthocephalan parasite

345 manipulates its host’s vigour to enhance its transmission to a suitable avian definitive host by

Page 15 of 26



Draft

346 increasing the ability of G. roeseli to avoid other crustacean predators (Médoc and Beisel 2008).

347 Together the relationships in these studies suggest that the effect of parasite infections on host

348 movement is largely dependent on the host-parasite system and type of movement/behaviour

349 being observed or measured.

350

351 Other drivers of sleepy lizard movement

352 Whilst tick load did not appear to impact movement activity in the sleepy lizard, we did identify

353 three other drivers of lizard movement. Lizard movement changed over the lizard activity season

354 (change over time) in a non-linear fashion. This is consistent with earlier findings (Kerr and Bull

355 2006c; Spiegel et al. 2015) and is driven by ambient conditions, predominately temperature

356 (Firth and Belan 1998) and food availability. Movement activity in the sleepy lizard increases

357 from the beginning of the active season, when nights are cold and lizards bask longer to reach

358 suitable body temperatures before they start moving. Following its peak, movement activity

359 decreases again towards the end of the activity season, as ambient temperatures become too hot

360 and food availability is scarce. Lizard sex was a second major driver of movement, with males

361 moving greater daily distances than females and covering larger home ranges (Spiegel et al.

362 accepted). Leu et al. (2011b) found similar differences between sexes and suggested that greater

363 male movement is due to males seeking extra pair matings and/or additional foraging

364 opportunities because costs of pair living are higher for males, or due to males investing more in

365 home range guarding activities such as patrolling it more or marking their home range more

366 frequently than females (Leu et al. 2016b; Sih et al. 2017). Thirdly, we showed movement to

367 increase with lizard body condition. Earlier studies suggested that lizard body condition and tick

368 loads are related (Bull and Burzacott 1993). Here, we included lizard body condition, tick load

Page 16 of 26



Draft

369 and their interaction in our model and did not find a significant interaction effect nor any

370 collinearity between these variables. Based on this, we were able to distinguish between the

371 effect of both factors and clearly identified body condition as a strong driver of movement in the

372 sleepy lizard. Whilst not demonstrated in our study, social interactions are also known to

373 influence behaviour and movement in the sleepy lizard, for example in monogamous lizard pairs,

374 where pair partners remain in close proximity for extended periods of time (Leu et al. 2015), or

375 in social networks, where individuals actively associate with or avoid individual neighbours (Leu

376 et al. 2010a; Spiegel et al. 2016). Ectoparasitic ticks also have the potential to influence the

377 relationship between social interaction and movement both directly, by changing host behaviours

378 such as the frequency of refuge reuse (Leu et al. 2010b), or indirectly by affecting how

379 conspecific lizards respond to infected individuals (Bull 1990).

380

381 Concluding remarks

382 Parasites are defined as having adverse impacts on their hosts. Nevertheless, several studies have

383 described exceptions to this (Mayer et al. 2015; Nelson et al. 2015; Taggart 2016; Taggart and

384 Schultz 2016). Here, we describe another exception in a lizard-tick host-parasite system which

385 extends our understanding of the complex relationships between hosts, parasites and the

386 environment by including detailed previous knowledge, frequently unavailable in other systems.

387 In this study, we chose a single index of vigour, however, it is possible that other aspects of the

388 lizards’ movement are impacted by tick loads, such as area covered or duration of activity, as

389 well as non-movement measures. The host-parasite arms race clearly predicts that some parasites

390 have little or no impact on their host while others have great impacts depending on the state of

Page 17 of 26



Draft

391 the arms race (Dawkins and Krebs 1979). This system is another example of a parasite having

392 little effect on its host.

393

394

395 Acknowledgments:

396 This research was supported by funds from the Australian Research Council. We thank Ron and

397 Leona Clark for allowing access to their land, Geoff Cottrell for maintaining the data loggers,

398 and Dale Burzacott for logistical support at the field site. We also extend our thanks to Jana

399 Bradley and Caroline Wohlfeil their tireless efforts collecting the data. O.S. was supported by the

400 Israeli Ministry of Science and Technology 'Shamir Postdoctoral Fellowship'.

401

402

403 References:

404 Anderson, R. 1978. The regulation of host population growth by parasitic species. Parasitology, 76(2): 405 119-157.406 Angelier, F., Shaffer, S.A., Weimerskirch, H., Trouvé, C., and Chastel, O. 2007. Corticosterone and 407 foraging behavior in a pelagic seabird. Physiol. Biochem. Zool. 80(3): 283-292.408 Bates, D., Maechler, M., Bolker, B., and Walker, S. 2015. Fitting Linear Mixed-Effects Models Using lme4. 409 Journal of Statistical Software, 67(1): 1-48.410 Beldomenico, P.M., and Begon, M. 2010. Disease spread, susceptibility and infection intensity: vicious 411 circles? Trends. Ecol. Evol. 25(1): 21-27.412 Beldomenico, P.M., Telfer, S., Gebert, S., Lukomski, L., Bennett, M., and Begon, M. 2008. Poor condition 413 and infection: a vicious circle in natural populations. Proc. R. Soc. Lond. B Biol. Sci. 275(1644): 1753-414 1759.415 Berdoy, M., Webster, J.P., and Macdonald, D. 2000. Fatal attraction in rats infected with Toxoplasma 416 gondii. Proc. R. Soc. Lond. B Biol. Sci. 267(1452): 1591-1594.417 Beyer, A.B., and Grossman, M. 1997. Tick paralysis in a red wolf. J. Wildl. Dis. 33(4): 900-902.418 Brockhurst, M.A., Rainey, P.B., and Buckling, A. 2004. The effect of spatial heterogeneity and parasites 419 on the evolution of host diversity. Proc. R. Soc. Lond. B Biol. Sci. 271(1534): 107-111.420 Bull, C. 1990. Comparisons of displaced and retained partners in a monogamous lizard. Wildl. Res. 17(2): 421 135-140.422 Bull, C. 1995. Population ecology of the sleepy lizard, Tiliqua rugosa, at Mt Mary, South Australia. Aust. J. 423 Ecol. 20(3): 393-402.

Page 18 of 26



Draft

424 Bull, C., Chilton, N., and Sharrad, R. 1988. Risk of predation for two reptile tick species. Exp. Appl. Acarol. 425 5(1-2): 93-99.426 Bull, C., and Sharrad, R. 1980. Seasonal activity of the reptile tick Aponomma Hydrosauri (Denny) (Acari: 427 Ixodidae) in experimental enclosures. Aust. J. Entomol. 19(1): 47-52.428 Bull, C.M., and Baghurst, B.C. 1998. Home range overlap of mothers and their offspring in the sleepy 429 lizard, Tiliqua rugosa. Behav. Ecol. Sociobiol. 42(5): 357-362.430 Bull, C.M., and Burzacott, D. 1993. The impact of tick load on the fitness of their lizard hosts. Oecologia, 431 96(3): 415-419.432 Bull, C.M., and Freake, M.J. 1999. Home-range fidelity in the Australian sleepy lizard, Tiliqua rugosa. 433 Aust. J. Zool. 47(2): 125-132.434 Bull, C.M., and Pamula, Y. 1996. Sexually dimorphic head sizes and reproductive success in the sleepy 435 lizard Tiliqua rugosa. J. Zool. 240(3): 511-521.436 Bull, C.M., and Pamula, Y. 1998. Enhanced vigilance in monogamous pairs of the lizard, Tiliqua rugosa. 437 Behav. Ecol. 9(5): 452-455.438 Chilton, N.B., Andrews, R.H., and Bull, C.M. 2000. Influence of temperature and relative humidity on the 439 moulting success of Amblyomma limbatum and Aponomma hydrosauri (Acari: Ixodidae) larvae and 440 nymphs. Int. J. Parasitol. 30(9): 973-979.441 Chilton, N.B., and Bull, C.M. 1993a. A comparison of the off-host survival times of larvae and nymphs of 442 two species of reptile ticks. Int. J. Parasitol., 23(5): 693-696.443 Chilton, N.B., and Bull, C.M. 1993b. Interspecific differences in microhabitat choice by two species of 444 Australian reptile tick. Int. J. Parasitol. 23(8): 1045-1051.445 Corp, I. 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.446 Dawkins, R., and Krebs, J.R. 1979. Arms races between and within species. Proc. R. Soc. Lond. B Biol. Sci. 447 205(1161): 489-511.448 Dhabhar, F.S. 2014. Effects of stress on immune function: the good, the bad, and the beautiful. 449 Immunologic Research, 58(2-3): 193-210.450 Dougherty, E.R., Seidel, D.P., Carlson, C.J., Speigel, O., and Getz, W.M. 2017. Going through the motions: 451 incorporating movement analyses into disease research. bioRxiv: 237891.452 Fenner, A.L., and Bull, C.M. 2008. The impact of nematode parasites on the behaviour of an Australian 453 lizard, the gidgee skink Egernia stokesii. Ecol. Res. 23(5): 897-903.454 Firth, B.T., and Belan, I. 1998. Daily and seasonal rhythms in selected body temperatures in the 455 Australian lizard Tiliqua rugosa (Scincidae): field and laboratory observations. Physiol. Biochem. Zool. 456 71(3): 303-311.457 Frank, A.F. 2011. R-hacks/mer-utils.R. https://github.com/aufrank/R-hacks/blob/master/mer-utils.R. 458 Accessed459 21 Oct 2017.460 Fraser, T.A., Charleston, M., Martin, A., Polkinghorne, A., and Carver, S. 2016. The emergence of 461 sarcoptic mange in Australian wildlife: an unresolved debate. Parasites & Vectors, 9(1): 316.462 Garrido, M., and Pérez-Mellado, V. 2013. Sprint speed is related to blood parasites, but not to 463 ectoparasites, in an insular population of lacertid lizards. Can. J. Zool. 92(1): 67-72.464 Hamat, N.A.B., Salahuddin, Z., and Salim, R. 2017. Facial nerve paralysis due to intra aural tick 465 infestation: a case report. IMC Journal of Medical Science, 11(1): 29-31.466 Hansson, L.-A., and Åkesson, S. 2014. Animal movement across scales. Oxford University Press.467 Hillegass, M.A., Waterman, J.M., and Roth, J.D. 2010. Parasite removal increases reproductive success in 468 a social African ground squirrel. Behav. Ecol. 21(4): 696-700.469 Kays, R., Crofoot, M.C., Jetz, W., and Wikelski, M. 2015. Terrestrial animal tracking as an eye on life and 470 planet. Science, 348(6240): aaa2478.

Page 19 of 26



Draft

471 Kerr, G.D., and Bull, C.M. 2006a. Exclusive core areas in overlapping ranges of the sleepy lizard, Tiliqua 472 rugosa. Behav. Ecol. 17(3): 380-391.473 Kerr, G.D., and Bull, C.M. 2006b. Interactions between climate, host refuge use, and tick population 474 dynamics. Parasitol. Res. 99(3): 214-222.475 Kerr, G.D., and Bull, C.M. 2006c. Movement patterns in the monogamous sleepy lizard (Tiliqua rugosa): 476 effects of gender, drought, time of year and time of day. J. Zool. 269(2): 137-147.477 Kerr, G.D., Bull, C.M., and Burzacott, D. 2003. Refuge sites used by the scincid lizard Tiliqua rugosa. 478 Austral Ecology, 28(2): 152-160.479 Kerr, G.D., Bull, C.M., and Mackay, D. 2004. Human disturbance and stride frequency in the sleepy lizard 480 (Tiliqua rugosa): implications for behavioral studies. Journal of Herpetology, 38(4): 519-526.481 Kuznetsova, A., Brockhoff, P.B., and Christensen, R.H.B. 2016. lmerTest: Tests in linear mixed effects 482 models. Version 2.0-33.483 Lafferty, K.D., and Morris, A.K. 1996. Altered behavior of parasitized killifish increases susceptibility to 484 predation by bird final hosts. Ecology, 77(5): 1390-1397.485 Leu, S.T., Bashford, J., Kappeler, P.M., and Bull, C.M. 2010a. Association networks reveal social 486 organization in the sleepy lizard. Anim. Behav. 79(1): 217-225.487 Leu, S.T., and Bull, C.M. 2016. Artificial water point for livestock influences spatial ecology of a native 488 lizard species. PloS One, 11(1): e0147433.489 Leu, S.T., Burzacott, D., Whiting, M.J., and Bull, C.M. 2015. Mate Familiarity Affects Pairing Behaviour in 490 a Long-Term Monogamous Lizard: Evidence from Detailed Bio-Logging and a 31-Year Field Study. 491 Ethology, 121(8): 760-768.492 Leu, S.T., Farine, D.R., Wey, T.W., Sih, A., and Bull, C.M. 2016a. Environment modulates population social 493 structure: experimental evidence from replicated social networks of wild lizards. Anim. Behav. 111: 23-494 31.495 Leu, S.T., Jackson, G., Roddick, J.F., and Bull, C.M. 2016b. Lizard movement tracks: variation in path re-496 use behaviour is consistent with a scent-marking function. PeerJ, 4: e1844.497 Leu, S.T., Kappeler, P.M., and Bull, C.M. 2010b. Refuge sharing network predicts ectoparasite load in a 498 lizard. Behav. Ecol. Sociobiol. 64(9): 1495-1503.499 Leu, S.T., Kappeler, P.M., and Bull, C.M. 2011a. The influence of refuge sharing on social behaviour in the 500 lizard Tiliqua rugosa. Behav. Ecol. Sociobiol. 65(4): 837-847.501 Leu, S.T., Kappeler, P.M., and Bull, C.M. 2011b. Pair-living in the absence of obligate biparental care in a 502 lizard: trading-off sex and food? Ethology, 117(9): 758-768.503 Leung, T.L., and Koprivnikar, J. 2016. Nematode parasite diversity in birds: the role of host ecology, life 504 history and migration. J. Anim. Ecol. 85(6): 1471-1480.505 Main, A. 1999. Ecological impact of ticks on their Australian lizard hosts. Chapter 5: Behaviour in the 506 field. PhD thesis. Flinders University.507 Main, A.R., and Bull, C.M. 2000. The impact of tick parasites on the behaviour of the lizard Tiliqua 508 rugosa. Oecologia, 122(4): 574-581.509 Martin, S.J., Helanterä, H., and Drijfhout, F.P. 2011. Is parasite pressure a driver of chemical cue diversity 510 in ants? Proc. R. Soc. Lond. B Biol. Sci. 278(1705): 496-503.511 Marzal, A., De Lope, F., Navarro, C., and Møller, A.P. 2005. Malarial parasites decrease reproductive 512 success: an experimental study in a passerine bird. Oecologia, 142(4): 541-545.513 Mayer, M., Brown, G.P., Zimmermann, B., and Shine, R. 2015. High infection intensities, but negligible 514 fitness costs, suggest tolerance of gastrointestinal nematodes in a tropical snake. Austral Ecology, 40(6): 515 683-692.516 Médoc, V., and Beisel, J.-N. 2008. An acanthocephalan parasite boosts the escape performance of its 517 intermediate host facing non-host predators. Parasitology, 135(8): 977-984.

Page 20 of 26



Draft

518 Moore, J.A., Daugherty, C.H., Godfrey, S.S., and Nelson, N.J. 2009. Seasonal monogamy and multiple 519 paternity in a wild population of a territorial reptile (tuatara). Biol. J. Linn. Soc. 98(1): 161-170.520 Nathan, R., Getz, W.M., Revilla, E., Holyoak, M., Kadmon, R., Saltz, D., and Smouse, P.E. 2008. A 521 movement ecology paradigm for unifying organismal movement research. Proc. Natl. Acad. Sci. U. S. A. 522 105(49): 19052-19059.523 Nelson, F.B., Brown, G.P., Dubey, S., and Shine, R. 2015. The effects of a nematode lungworm (Rhabdias 524 hylae) on its natural and invasive anuran hosts. J. Parasitol. 101(3): 290-296.525 Nunn, C.L., Altizer, S., Sechrest, W., Jones, K.E., Barton, R.A., and Gittleman, J.L. 2004. Parasites and the 526 evolutionary diversification of primate clades. Am. Nat. 164(S5): S90-S103.527 O’Dwyer, K., Kamiya, T., and Poulin, R. 2014. Altered microhabitat use and movement of littorinid 528 gastropods: the effects of parasites. Mar. Biol. 161(2): 437-445.529 Petney, T.N., Andrews, R.H., and Bull, C.M. 1983. Movement and host finding by unfed nymphs of two 530 Australian reptile ticks. Aust. J. Z. 31(5): 717-721.531 Poulin, R. 2010. Parasite manipulation of host behavior: an update and frequently asked questions. 532 Stud. 41: 151-186.533 Poulin, R. 2013. Parasite manipulation of host personality and behavioural syndromes. J. Exp. Biol. 534 216(1): 18-26.535 R Core Team. 2016. In R: A language and environment for statistical computing. R Foundation for 536 Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.537 Russell, C.A., Smith, D.L., Childs, J.E., and Real, L.A. 2005. Predictive spatial dynamics and strategic 538 planning for raccoon rabies emergence in Ohio. PLoS. Biol. 3(3): e88.539 Sato, T., Egusa, T., Fukushima, K., Oda, T., Ohte, N., Tokuchi, N., Watanabe, K., Kanaiwa, M., Murakami, 540 I., and Lafferty, K.D. 2012. Nematomorph parasites indirectly alter the food web and ecosystem function 541 of streams through behavioural manipulation of their cricket hosts. Ecol. Lett. 15(8): 786-793.542 Sih, A., Spiegel, O., Godfrey, S., Leu, S., and Bull, C.M. 2017. Integrating social networks, animal 543 personalities, movement ecology and parasites: a framework with examples from a lizard. Anim. Behav. 544 136: 195-205545 Spiegel, O., Leu, S.T., Sih, A., and Bull, C.M. 2016. Socially interacting or indifferent neighbours? 546 Randomization of movement paths to tease apart social preference and spatial constraints. Methods in 547 Ecol. Evol. 7(8): 971-979.548 Spiegel, O., Leu, S.T., Sih, A., Godfrey, S.S., and Bull, C.M. 2015. When the going gets tough: behavioural 549 type-dependent space use in the sleepy lizard changes as the season dries. Proc. R. Soc. Lond. B Biol. Sci. 550 282(1819): 20151768.551 Spiegel, O., Sih, A., Leu, S.T., and Bull, C.M. 2017. Where should we meet? Mapping social network 552 interactions of sleepy lizards shows sex-dependent social network structure. Anim. Behav. 136: 207-215553 Taggart, P.L. 2016. Are blood haemoglobin concentrations a reliable indicator of parasitism and 554 individual condition in New Holland honeyeaters (Phylidonyris novaehollandiae)? Trans. R. Soc. S. Aust. 555 140(1): 17-27.556 Taggart, P.L., and Schultz, D. 2016. Do avian ticks (Ixodes hirsti) influence host phenotype in New Holland 557 honeyeaters (Phylidonyris novaehollandiae)? Trans. R. Soc. S. Aust. 141(1): 48-56.558 Thomas, F., Poulin, R., de Meeüs, T., Guégan, J.-F., and Renaud, F. 1999. Parasites and ecosystem 559 engineering: what roles could they play? Oikos, 84(1): 167-171.560 Viboud, C., Bjørnstad, O.N., Smith, D.L., Simonsen, L., Miller, M.A., and Grenfell, B.T. 2006. Synchrony, 561 waves, and spatial hierarchies in the spread of influenza. Science, 312(5772): 447-451.562 Warren, B.S., Wakshlag, J.J., Maley, M., Farrell, T.J., Struble, A.M., Panasevich, M.R., and Wells, M.T. 563 2011. Use of pedometers to measure the relationship of dog walking to body condition score in obese 564 and non-obese dogs. Br. J. Nutr. 106(S1): S85-S89.

Page 21 of 26



Draft

565 White, A., Watt, A.D., Hails, R.S., and Hartley, S.E. 2000. Patterns of spread in insect-pathogen systems: 566 the importance of pathogen dispersal. Oikos, 89(1): 137-145.567 Wickham, H. 2009. ggplot2: Elegant graphics for data analysis. Springer-Verlag New York.568 Wohlfiel, C.K., Leu, S.T., Godfrey, S.S., and Bull, C.M. 2013. Testing the robustness of transmission 569 network models to predict ectoparasite loads. One lizard, two ticks and four years. Int. J. Parasitol: 570 Parasites and Wildlife, 2: 271-277.571 Zuur, A.F., Ieno, E.N., and Elphick, C.S. 2010. A protocol for data exploration to avoid common statistical 572 problems. Methods in Ecology and Evolution, 1(1): 3-14.573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

Page 22 of 26



Draft

594 Table 1: Linear mixed model summary showing the effects of sleepy lizard (Tiliqua rugosa) sex,

595 body condition, tick load and time in the activity season on the average daily movement of

596 lizards.

597 SE = standard error

598 DF = degrees of freedom

Fixed-effects Estimate SE DF T-value P-valueInterceptMale 15.12 0.27 84.18 56.70 <0.001SexFemale -2.03 0.39 86.33 -5.22 <0.001Fortnight 6.93 0.86 283.19 8.09 <0.001Fortnight2 -7.66 0.86 217.16 -8.96 <0.001Total larvae equivalent -0.10 0.19 226.12 -0.54 0.590Body condition 0.87 0.19 194.89 4.45 <0.001Total larvae equivalent x body condition -0.08 0.25 282.65 -0.31 0.760Total larvae equivalent x fortnight 0.06 0.21 217.08 0.26 0.790

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

Page 23 of 26



Draft

615

616

617

618 Figure 1: Sleepy lizards (Tiliqua rugosa) and our study site: (A) the location of our study site in

619 South Australia; (B) adult sleepy lizard with GPS unit and radio transmitter taped to tail, note

620 permanent marker pen adjacent lizard for scale; (C) ticks attached to lizard; and (D) movement

621 tracks of a single tagged lizard over the duration of the activity season (green lines), showing

622 example of our dependent variable, the total daily distance moved (red line). Map source data

623 were provided by the Department of Environment, Water and Natural Resources.

624

625 Figure 2: Effect of lizard sex on the mean daily movement of male and female sleepy lizards

626 (Tiliqua rugosa).

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

Page 24 of 26



Draft

642

643

644

645

646

647 Figure 1:

648

649

Page 25 of 26



Draft

650

651

652

653

654

655

656

657 Figure 2:

658

Page 26 of 26



Documents

Draft · 2018-11-27 · Draft 48 Keywords: reptile, body-mass index, movement ecology, parasite, gps tracking, host-parasite 49 interaction, spatial behaviour, sleepy lizard, Tiliqua