Sex-independent senescence in a cooperatively breeding mammal

1. Researchers studying mammals have frequently interpreted earlier or faster rates of ageing 28 in males as resulting from polygyny and the associated higher costs of reproductive 29 competition. 30 2. Yet few studies conducted on wild populations have compared sex-specific senescence 31 trajectories outside of polygynous species, making it difficult to make generalised inferences 32 on the role of reproductive competition in driving senescence, particularly when other 33 differences between males and females might also contribute to sex-specific changes in 34 performance across lifespan. 35 3. Here, we examine age-related variation in body mass, reproductive output and survival in 36 dominant male and female meerkats, Suricata suricatta . Meerkats are socially monogamous 37 cooperative breeders where a single dominant pair virtually monopolize reproduction in each 38 group and subordinate group members help to rear offspring produced by breeders. 39 4. In contrast to many polygynous societies, we find that neither the onset nor the rate of 40 senescence in body mass or reproductive output show clear differences between males and 41 females. Both sexes also display similar patterns of age-related survival across lifespan, but 42 unlike most wild vertebrates, survival senescence (increases in annual mortality with rising 43 age) was absent in dominants of both sexes, and as a result, the fitness costs of senescence were 44 entirely attributable to declines in reproductive output from mid- to late-life. 45 5. We suggest that the potential for intrasexual competition to increase rates of senescence in 46 females – who are hormonally masculinised and frequently aggressive – is offset by their 47 ability to maintain longer tenures of dominance than males, and that these processes combined 48 lead to similar patterns of senescence in both sexes. 49 6. Our results stress the need to consider the form and intensity of sexual competition as well as other sex-specific features of life history when investigating the operation of senescence in 51 wild populations.


INTRODUCTION
with higher rates of senescence in females (see Sharp    Body mass records were obtained early in the morning prior to foraging by enticing individuals 192 onto an electronic balance with crumbs of boiled egg as a reward. As stated above, subordinates 193 that never acquire a dominance position but remain in the study population until their death 194 have usually died before they reach three years of age. Moreover, for a further sizeable Statistical analyses consisted of three steps. In the first step, we examined age-related 207 variation in body mass, reproductive output and mortality, with models parameterised so that 208 we could directly compare sex differences in ageing trajectories. In the second step, we 209 extended the best supported models from step one to incorporate effects of partner age and 210 thereby examine whether senescence declines were conflated by partner effects. In the third 211 step, we used information on age-related changes in reproduction and survival to calculate the  The body mass dataset partitioned individual lifespans into 4-month periods, with body 218 mass then calculated as the mean daily morning mass within each period. This allowed us the 219 highest possible resolution of sampling without losing full periods of weights due to pregnancy, 220 which we excluded. Specifically, we excluded any pregnancy weights by back-casting 70 days 221 from the day of birth or litter loss, thus removing any weight increases due to gestational growth 222 (Fig. S1). On average, this resulted in 24.4 ± 0.6 mass records/female/period and 43.1 ± 0.7 For the reproductive output dataset, we instead partitioned individual lifespans into 6-225 month periods. Preliminary analyses suggested that doing so reduced the number of zeroes in 226 the dataset and therefore improved the fit of models compared to shorter time intervals.

227
Reproductive output was defined as the number of offspring that were produced by a male or 228 female within each 6-month period that survived to nutritional independence at 3 months of 229 age. 78.0% of pups that survive to nutritional independence go on to reach adulthood at one 230 year of age (n = 2040 pups between January 1994 and July 1998). Parentage was assigned 231 through genetic analysis of 18 microsatellites derived from tissue samples taken from the tip 232 of individuals tails (Nielsen, 2012), and where genetic data were missing, maternity could be 233 inferred from field observations where we were certain only a single female had given birth.

236
Age-related variation in many vertebrate traits often takes the form an initial early-life 237 increase, a mid-life plateau, and a later-life decline. To capture this pattern for body mass and 238 reproductive output in meerkats, we fitted a series of mixed effects models that included 239 chronological age either as a quadratic function (a linear and quadratic age term) or as a 240 threshold function (usually where linear slopes are estimated on either side of each fitted 241 threshold age). Threshold functions are generally better equipped to reliably recover the full 242 age-dependence of trait change but do so at the expense of additional parameters. We 243 implemented our models in a Bayesian framework using the brms package (Bürkner, 2018).

244
This offers a distinct advantage over a frequentist treatment, for while the former generates a 245 posterior distribution for threshold parameters upon which other model terms are conditioned, 246 frequentist analyses must fit multiple models and secondarily estimate the position of any 247 thresholds (with associated confidence) through likelihood profiling (Ulm 1989). We modelled 248 body mass using a Gaussian error distribution, and reproductive output using a zero-inflated negative binomial distribution with a single zero-inflation parameter applied to all observations 250 (zi ~ 1).

251
In order to test for sex differences in ageing patterns we adopted a similar approach to 252 Tompkins and Anderson (2019) and fitted six models for each trait (Table S1). In model 1, age 253 was included as a quadratic function, and males and females were assumed to follow the same 254 age trajectory. In model 2 the linear and quadratic age terms of model 1 were each interacted 255 with a covariate for sex to allow for male and female age trajectories to differ. We then 256 specified four forms of threshold model. For body mass, this included two thresholds for each 257 sex, one in early life (first threshold age = TSEX,1) and one in mid to late life (second threshold 258 = TSEX,2). For reproductive output we only fit a single threshold in mid to late life as preliminary 259 model fitting found no evidence for an additional threshold in early life. Threshold models then 260 differed in the extent to which they forced males and females to have sex-specific slopes on 261 age across lifespan, and/or sex-specific thresholds (Table S1). In the most advanced threshold 262 model for body mass, model 6 (from which other models were derived), different threshold 263 ages and different slopes across age were parameterised for males and females, such that the 264 body mass of meerkat individual i at age j was parameterised as: where for thresholds TSEX,t, (Ageij -TSEX,t)+ = (Ageij -TSEX,t)*I(Ageij ≥ TSEX,t). I(Ageij 271 ≥ TSEX,t) is an indicator function equalling 1 when Ageij ≥ TSEX,t, and 0 otherwise. Thus, β4 and 272 β5 are the difference in the slope of each response variable on age after the first threshold age 273 relative to the slopes before the first threshold age (for females and males, respectively). The step function 'switches' β4 and β5 terms off for ages ≤ TSEX,1 and on for ages > than TSEX,1 in 275 each sex, where sex is a dummy variable with females coded as "0" and males as "1".

276
Additional population-level "fixed" effects (βkXk…) included the age at first dominance 277 (AFD), AFD:sex, longevity, longevity:sex, total rainfall, group size, season, and dominance 278 status. Total rainfall during each period was calculated from onsite rain gauge data; on days  All continuous parameters were z-score transformed prior to model fitting, apart from age.

296
Within each model, we used 95% Bayesian credible intervals (BCI) drawn from the posterior compared using k-fold cross-validation ('k-Fold IC') using subset number k = 10. This method 300 divides the data into ten subsets and validates the results of the nine subsets for each missing 301 dataset. For each model, we also calculated the Bayesian equivalent of the R 2 using bayes_R2 302 function in brms (Gelman, Goodrich, Gabry, & Ali, 2017).

304
Partner age effects 305 To assess the extent to which reproductive declines in one sex might contribute to or 306 partly explain reproductive declines in the other sex, we modelled the relationship between the 307 age of a dominant female and the age of her male partner across her period of tenure, and vice 308 versa. Partner age was fitted as the response variable in general additive models (gam), with 309 the age of the focal individual included as a sex-specific smoother function in each (6 knots).

310
As this preliminary analysis hinted at a linear increase in partner age with the age of the focal To test for sex differences in longevity and survival across age we performed both semi-322 parametric and parametric survival modelling with the survival and flexsurv packages (Jackson fates, 63 females and 92 males who disappeared during the study with their fate being 325 unknown, and 9 females and 7 males who were still alive at the end of the study. Individuals 326 of unknown fate, who either disappeared during the study or were still alive at the end of study 327 sampling period, can still be incorporated into the analysis through censoring. However, two 328 key assumptions of censoring are that it is random with respect the individuals affected, and 329 independent of the process of mortality such that individuals do not experience a change in 330 mortality risk due to being censored. This is unlikely to be the case in meerkats, where where is the age for which reproductive value is being calculated, w is the age at last 359 reproduction, lx is survival at age x and mx is reproductive output at age x. Further details are 360 provided in the Supporting Information. reproductive declines were absent. For the latter, reproductive output was held constant from 369 the age of peak reproductive output and RV was estimated as above. CRS is then [(RVobserved -1). Model comparisons highlighted that age-related variation in body mass was best described 375 by threshold models that partitioned lifespan into three stages: an early-life increase, a mid-life 376 plateau, and later-life senescence (Table 1). In the best fitting model (model 3), males and 377 females were parameterised to share common slopes and common threshold ages, suggesting 378 that both the onset and rate of senescence were independent of sex (Fig. 1, Table 2).   Fig. 2a, 2b). In the best supported model (model 5), males and females shared a common 395 threshold age term, but were given separate slopes on age which took the form of a linear 396 increase from mid-life to late life, followed a subsequent period of reproductive decline.
reproductive output that drove this trend, with estimates for the later-life slope showing no 399 apparent difference between males and females ( Table 2). As for body mass then, both the 400 onset and rate of senescence in reproductive output were independent of sex (Fig. 2). Based Although analyses of the raw data provided some suggestion that older dominant 409 individuals were more likely to be paired with an older partner (Fig. S2), the effect was weak  that even under the conservative assumption that censored individuals are exposed to only a 422 small increase in mortality risk compared to non-censored individuals, the sex difference in lifespan no longer held (Fig. 3). Moreover, when we excise individuals that were censored 424 before the end of the study, or if we treat them as having immediately died, no sex difference 425 was apparent (Fig. 3). In the scenario where censoring is associated with reduced risk the effect 426 of sex remains stable.

427
Parametric modelling of our survival data revealed that the pattern of survival in 428 meerkats was best described by a log-normal distribution (Fig. S4). The log-normal distribution  (Table S5). However, as sample 435 sizes decrease later in life the power with which to detect senescence declines. Annual mortality 436 derived from our log-normal survival model tracks mortality probabilities well with reasonable 437 sample sizes up to around 8 years (Fig. 4A, B), after which we are unlikely to be able to detect 438 senescent trends. While the parametric models revealed no difference in the log mean TABLE 1. Comparison of models investigating sex-specificity of senescence in meerkats. Models are ranked according to k-fold IC, with the 453 lowest k-fold IC taken as the best-supported model (bold). Threshold models differed in the extent to which they allowed males and females to 454 have common or distinct sex-specific estimates for threshold ages (T) and/or slopes on age (as detailed in the main text). 'params' refers to the 455 number of parameters estimated by each model. All models included additional population-level ('fixed') and group-level ('random') terms as 456 described in the main text.

FIGURE 1: (a) Age-related variation in body mass in female (red) and male (blue) meerkats
Our study finds that in meerkats, the form and rate of senescence across three 546 components of life history are similar in males and females. We show that the onset and rate 547 of senescence in body mass and reproductive output were largely independent of sex, with trait 548 values peaking between 4 and 6 years of age and declining at similar rates thereafter. Age-549 related survival probability was also unaffected by sex, but unlike the former two traits, we survival are more severe, as reflected in their shortened tenures. Taken together, we suggest 585 that the potential for intrasexual competition to increase rates of senescence in females is offset 586 by their ability to maintain longer tenures of dominance than males, and that these processes 587 combined lead to similar patterns of senescence in both sexes. Or, put differently, the realised 588 costs of competition on fitness are not divergent enough to have led to the evolution of sex 589 differences in senescence trajectories in meerkats. 590 We found that body mass and reproductive output senesced in parallel. For female 591 meerkats, the fitness consequences of reduced body mass have already been well described  However, we do not find convincing support for partner age effects in our study, and by 603 implication it is likely that intrinsic physiological declines in males and females are mostly 604 responsible for the observed reproductive declines in either sex.

605
Our study also detected a strong age-independent contribution to body mass variation in  use this to guide interpretation of any results. In our case, even a minor increase in hazard for 663 censored individuals (a reasonably conservative assumption) lead to a loss of significance and 664 a reduction in the effect size of the sex term on longevity, a result that was also found when 665 individuals that disappeared prior to study end were truncated rather than censored. Thus, with 666 censoring-induced bias considered, we do not find strong support for differences in longevity  In concert with other studies, our results emphasize that broad categorisation into 685 mating systems will likely only get us so far in understanding sex differences in ageing in wild 686 vertebrates, for within mating systems and within species, sex differences in the degree to 687 which males and females compete for reproductive opportunities, and the manner in which 688 they do so, vary widely. In order to better understand why ageing rates differ so widely within 689 and between species in the wild, and in particular, between the sexes, it will therefore be