Work

Editor's evaluation: Lifespan extension in female mice by early, transient exposure to adult female olfactory cues

Year: 2022
Type: peer-review
Abstract: Article Figures and data Abstract Editor's evaluation eLife digest Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Metrics Abstract Several previous lines of research have suggested, indirectly, that mouse lifespan is particularly susceptible to endocrine or nutritional signals in the first few weeks of life, as tested by manipulations of litter size, growth hormone levels, or mutations with effects specifically on early-life growth rate. The pace of early development in mice can also be influenced by exposure of nursing and weanling mice to olfactory cues. In particular, odors of same-sex adult mice can in some circumstances delay maturation. We hypothesized that olfactory information might also have a sex-specific effect on lifespan, and we show here that the lifespan of female mice can be increased significantly by odors from adult females administered transiently, that is from 3 days until 60 days of age. Female lifespan was not modified by male odors, nor was male lifespan susceptible to odors from adults of either sex. Conditional deletion of the G protein Gαo in the olfactory system, which leads to impaired accessory olfactory system function and blunted reproductive priming responses to male odors in females, did not modify the effect of female odors on female lifespan. Our data provide support for the idea that very young mice are susceptible to influences that can have long-lasting effects on health maintenance in later life, and provide a potential example of lifespan extension by olfactory cues in mice. Editor's evaluation This valuable study provides solid evidence for a new intervention, exposure to male vs. female olfactory cues, with an impact on female mouse lifespan. This is interesting to the field of aging research, especially since most described pro-longevity interventions to date tend to work better in male mice. https://doi.org/10.7554/eLife.84060.sa0 Decision letter Reviews on Sciety eLife's review process eLife digest The environment that animals are exposed to early in life can influence their subsequent rate of development, reproduction and aging. Experiments done in rodents have shown that social stimuli such as odours from the same sex or opposite sex individuals can affect the age at which sexual maturity is reached. Variations in age of sexual maturity are directly correlated with median lifespans of mice, with strong associations observed between later sexual maturity and longer lifespans in female mice. Detailed experiments exposing female or male mice to scents from mice of the same or another sex strongly suggest that growing up smelling the same sex can delay sexual maturity, while scents from another sex can hasten it. Interestingly, mice that lacked the cells that sense odours do not change their age of sexual maturity in response to scents from the opposite sex. This ability to steer one's developmental timeline depending on environmental cues may allow animals to prepare for future environments. But can it also influence an animal's lifespan? To answer this question, Garratt et al. observed the lifespans of female and male mice under different conditions. Mice were exposed to same-sex or other-sex odours, in the form of urine or soiled bedding, from day 3 to day 60 of their lives. The results showed that female mice exposed to odours from other females exhibited an increased lifespan, as compared to those not exposed to scents, while those exposed to odours from males did not show any change in their lifespan. In striking contrast, male mice exposed to odours from either sex showed no variation in their lifespans. The impairment of a particular type of odour-sensing neuron in mice did not change these results, making it likely that another neuron type is responsible for the changes in lifespan observed in the female mice. These experiments elegantly demonstrate that exposure to certain sensory information, in this case scent, can change how long mammals live. While similar effects involving smells are unlikely to influence lifespan in humans, it is possible that other types of sensory information affect our health and how we age. Introduction Interventions in the first few weeks of postnatal life seem to have particular power to produce life-long changes in mouse physiology, with effects on late-life illnesses and on survival. Bartke's laboratory, for example, has shown that the exceptionally long lifespan of the Ames dwarf mice, in which longevity appears to reflect lower levels of growth hormone (GH) produced by the anterior pituitary, can be reduced back to that of non-mutant controls by transient exposure to daily GH injections started at 2 weeks of age and discontinued 6 weeks later (Panici et al., 2010). Furthermore, these early-life GH injections block the development of multi-modal stress resistance shown by skin-derived fibroblasts of Ames dwarf mice (Panici et al., 2010), and also prevent the lower levels of hypothalamic inflammatory characteristics of unmanipulated Ames mice (Sadagurski et al., 2015b). Many other characteristics of Ames dwarf mice, plausibly connected to their longevity, insulin sensitivity (Dominici et al., 2002), and cognitive health (Kinney et al., 2001; Sun et al., 2005), are also blocked by early-life GH treatments when tested at 18 months of age, including changes in fat, muscle, plasma irisin and GPLD1, liver, and brain (Li et al., 2022). Conversely, lifespan can be increased in genetically normal mice by pre-weaning manipulation of litter size – mice weaned from litters in which litter size has been increased to 12 pups/nursing mother live longer than those in control litters of 8 pups (Sun et al., 2009). Genetic selection experiments have also provided relevant data, by showing extended lifespan in non-inbred stocks of mice that were selected, for 17 or more generations, for exceptionally slow growth in the first 10 days of life, or between 26 and 56 days of age (Miller et al., 2000). Age at sexual maturity is another early-life developmental factor that has been associated with lifespan. Across strains of mice, age at sexual maturity correlates positively with median lifespan (Yuan et al., 2012), and genetic manipulations that delay sexual maturity have been associated with longer lifespan in females (Wang et al., 2018). Age at sexual maturity in rodents is also influenced by environmental cues, in particular olfactory cues from conspecifics. Female mice that are exposed to male odors (typically soiled bedding or male urine) during development show hastened sexual maturity (Drickamer, 1983; Vandenbergh, 1973). In contrast, exposing females to the odors of group-housed adult virgin females can delay sexual maturity (Vandenbergh, 1973; Drickamer, 1977). In males, exposure to females or their odors has been shown to be associated with increased masses of reproductive organs (testes, seminal vesicles) early in life, also indicating earlier sexual maturity in males (Vandenbergh, 1971). These odor priming effects are assumed to be adaptive, allowing animals to modulate their timing of early-life reproduction to suit predicted future environments. In one previous study, hastened female sexual maturity was associated with an increase in mortality over the first 180 days of life and reduced litter sizes when breeding, suggesting a cost to this developmental change (Drickamer, 1988). However, in this prior study females were exposed directly to males on the day of their first estrous, which was earlier in females exposed to male odors producing a confounding effect that could contribute to mortality differences. The effects of odors on sexual maturity are initiated, at least partly, by detection of odorants through the vomeronasal organ, one of the two olfactory sub-systems in mice. Female mice that have had their vomeronasal organ surgically removed do not show hastened sexual maturity in response to male odors (Lomas and Keverne, 1982). Vomeronasal sensory neurons (VSNs) express hundreds of different sensory receptors, but these receptors largely use one of two different types of G protein for signal transduction (Munger et al., 2009). The G protein Gαo is used for signal transduction in VSNs of the basal layer of the vomeronasal organ. Cre-mediated deletion of Gαo in the olfactory system has been shown to inhibit females from showing hastened sexual maturity in response to male odors (Oboti et al., 2014), indicating that this subset of VSNs is important in driving changes in sexual maturity, at least in response to opposite sex cues in females. With these precedents in mind, we tested the idea that olfactory signals, presented to mice from 3 days of age but discontinued at 60 days, would have long-lasting effects that influence age at death. We initiated odor exposure early in life, prior to weaning, because a previous study had shown that very early exposure influences female body weight during development (Cowley and Wise, 1972). In particular, we speculated that odor from adult female mice would lead to lifespan extension in female (but not in male) mice, and vice versa. We also hypothesized that exposure to opposite-sex odors would reduce survival given the hastening of sexual maturity. Here, we show that transient exposure to odor from adult females starting at 3 days of age can extend lifespan of female mice. Results Survival Our principal hypothesis was that early-life exposure to sex-specific olfactory cues indicative of social environment would influence lifespan of mice. A secondary hypothesis was that the response to those cues would depend on function of Gαo-expressing neurons in the vomeronasal organ, since these have been shown to mediate early-life priming responses to opposite-sex urine, at least in female mice (Oboti et al., 2014). Mice were exposed either to same-sex or to opposite-sex odors starting from 3 days of age, with daily exposure through day 60 of life. They were housed in same-sex cages from the age of weaning, that is from 19 or 20 days. Figure 1 shows survival curves for female and for male mice subjected either to odors from adult males ('MU') or from adult females ('FU'), or exposed to water as a control (ZU, for zero urine). Data from male and female mice were evaluated separately. Table 1 shows median age, and age at 90th percentile, for each combination of sex and treatment, pooling across genotypes. In female mice, FU led to an 8% increase in median lifespan, compared to the ZU control group, and a 9% increase in the age at 90th percentile. The log-rank test, comparing all three treatment groups, showed that treatment led to significant differences among the groups (p=0.04), and a follow-up test showed that for females the FU mice were longer-lived than the ZU controls (p=0.01 by log-rank test). Male odors did not produce a lifespan change in female mice compared to controls (p=0.6). Treatment did not modify lifespan in the male mice; log-rank p=0.8 for all treatment groups taken together. Figure 1 Download asset Open asset Survival curves for female (left panels) and male mice (right panels) exposed to male, female, or no odors daily from 2 to 60 days of age. Top row: pooled across genotypes. Middle row: wild-type (WT) mice. Bottom row: mutant mice. Cox regression analysis (see Table 2) shows no significant interaction between genotype and odor treatment in either sex. See the Supplementary data for age at death of each mouse and its genotype, sex, and treatment group. Figure 1—source data 1 Individual-level survival data in days. https://cdn.elifesciences.org/articles/84060/elife-84060-fig1-data1-v2.xlsx Download elife-84060-fig1-data1-v2.xlsx Table 1 Median and 90th percentile survival statistics for mice treated with same-sex or opposite-sex odors from 3 to 60 days of age. Female mice, pooled across genotypeTreatmentCountMedian% Changep(90)% ChangeFemale odor (FU)59 (30 WT)62188239Male odor (MU)58 (30 WT)58527925No odor (ZU)57 (27 WT)576(NA)754(NA)Male mice, pooled across genotypeTreatmentCountMedian% Changep(90)% ChangeFemale odor (FU)60 (30 WT)737−1932−1Male odor (MU)59 (29 WT)7724879−7No odor (ZU)59 (29 WT)741(NA)944(NA) To see if the effect of odor was dependent on Gαo expression in the vomeronasal organ, survival data were analyzed by Cox regression, with two factors: treatment (MU, FU, or ZU) and genotype (WT for wild-type [Gnao1fx/fx, Omp+/+] and mutant for Ompcre positive mice [Gnao1fx/fx, Ompcre/+]), with a [treatment × genotype] interaction term. A significant interaction would indicate that the two genotypes of mice have responded differently to the odor treatment. Table 2 shows the significance levels for each of the three terms, calculated separately for female and for male mice. The interaction term did not achieve significance in either sex, implying that the odor effect did not depend on genotype. In addition, there was no independent effect of genotype on survival. Only the odor treatment had a significant effect on survival, and only in female mice (p=0.04), consistent with the results of the log-rank tests. The Cox regression calculation also confirmed the inference that FU females lived longer than ZU females (p=0.015 without adjustment for multiple comparisons, and p=0.045 with Sidak adjustment), and that no other pairwise comparison between treatment groups had a significant effect in either sex. Within each genotype and sex individual comparison (e.g. WT females, mutant females, WT males, mutant males), there was no significant difference in survival between odor treatment groups using a log-rank test. This further emphasizes the lack of any strong genotype-specific effect and that the effects of the odor treatment are more clearly apparent when both genotypes are included within an analysis due to the increased sample size. Table 2 Significance tests for predictors in Cox regression, calculated separately for each sex. Sex of mouseTreatmentGenotypeInteractionFemaleΧ2 = 6.5p=0.039Χ2 = 0.4p=0.5Χ2 = 0.12p=0.9MaleΧ2 = 0.52p=0.8Χ2 = 0.23p=0.6Χ2 = 0.31p=0.5 To evaluate the effects of early-life odor exposure on survival to unusually high ages, we used the method of Wang and Allison (Wang et al., 2004), which calculates a Fisher's exact test statistic on the proportion of mice in each treatment that remain alive at the 90th percentile of their joint survival distribution, the method also employed by the mouse Intervention Testing Program consortium (Macchiarini, 2021). In females, the comparison of FU to control ZU mice produced a significance value of p=0.004, showing that FU-treated females were more likely to reach old age than ZU-treated females. None of the other paired comparisons reached significance in female mice, and none did so in male mice. Functional testing Each mouse surviving to 22 months of age was evaluated using a series of tests for age-sensitive physiological function. The data were analyzed separately for each sex, using a two-factor ANOVA (treatment, genotype, interaction). Because there was no significant interaction between odor treatment and genotype for any measured trait, the data were pooled across genotypes for analysis of odor effects. Figure 2 (top row) shows core body temperature at 22 months. For female mice, exposure to female odors leads to a significantly higher body temperature compared to the ZU control group (p=0.007). There are no significant differences between the MU group and FU mice or ZU controls, and there are no significant differences among any of the groups for male mice. Figure 2 also shows (bottom row) levels of non-fasting glucose measured at 12 months of age. There were no effects of treatment in female mice, but glucose levels were significantly lower in male ZU mice than in the FU or MU mice (Figure 2c; p=0.004 for ANOVA, and p=0.01 for each pairwise comparison to ZU male mice). There were no treatment effects, or significant interaction terms, for mean forepaw grip strength or for rotarod performance (mean of three trials) in either sex (Figure 2—figure supplement 1). Figure 2 with 1 supplement see all Download asset Open asset Measurement of core body temperature at 22 months (top row) and of plasma non-fasting glucose at 12 months (bottom row) in female (left) and male (right) mice of the indicated odor treatment groups. Each symbol represents one mouse, with color indicating genotype. The black line indicates the mean of each group. ZU = control mice, FU = those exposed to female odors, MU those that were exposed to male odors. Full data matched to each genotype and treatment group is available in the Supplementary data. Figure 2—source data 1 Individual data for temperature, grip strength, and rotarod function. https://cdn.elifesciences.org/articles/84060/elife-84060-fig2-data1-v2.xlsx Download elife-84060-fig2-data1-v2.xlsx Figure 2—source data 2 Individual data for glucose levels. https://cdn.elifesciences.org/articles/84060/elife-84060-fig2-data2-v2.xlsx Download elife-84060-fig2-data2-v2.xlsx Figure 3 shows weight trajectories for each of the odor treatment groups separately in both sexes. There were no significant differences at any age, except that in female mice, ZU females were marginally heavier than mice in the MU or ZU control groups at 6 months of age; p=0.03, and p=0.04 if an MU mouse weighing 16 g is considered an outlier. Because these significance tests were not adjusted for multiple comparisons and because there were no large or significant effects in odor at any other age, we suspect that this difference at 6 months is a chance effect. Body weight changes with odor treatment were also uninfluenced by genotype, and the genotypes of mice had similar body weights across the experiment (Figure 3—figure supplement 1). Figure 3 with 1 supplement see all Download asset Open asset Weight as a function of age in female and male mice. The genotypes are pooled in the figure with separate figures provided in Figure 3—figure supplement 1. N=60 in each group through month 3, and then diminished to 32–39 mice (females) or to 48–49 mice (males) by month 18. Standard error of the mean (SEM) values are plotted, but are in most cases too small to see; SEM was <0.3 g at younger ages and typically 0.7–1.3 g at 18 months. Figure 3—source data 1 Individual level data for body weight at each time point. https://cdn.elifesciences.org/articles/84060/elife-84060-fig3-data1-v2.xlsx Download elife-84060-fig3-data1-v2.xlsx Discussion Our results add to the growing evidence that late life health, including risks of lethal diseases, can be influenced by non-genetic factors, such as litter size (Sun et al., 2009; Sadagurski et al., 2015a) and GH levels (Panici et al., 2010; Sadagurski et al., 2015b), to which a newborn or pre-weanling mouse is exposed transiently. The 'window of opportunity', or age at which a mouse is susceptible to such influences, is not known, and may differ depending on the specific form of environmental variation, but it is noteworthy that GH treatments can prevent lifespan extension of Ames dwarf mice when these are initiated at 2 weeks of age, but fail in Snell dwarf (Vergara et al., 2004) and Ames dwarf animals (Bartke, unpublished results) when started at 4 weeks of age. The basis for this transient susceptibility to factors that set the tempo of late-life decline is unknown, but seems likely to involve epigenetic modulation of events that are sensitive to nutritional and neuro-endocrine stimuli. Fibroblast cultures from adult mice of the Ames and Snell dwarf mutant strains are resistant, in culture, to multiple forms of lethal injury (Murakami et al., 2003; Salmon et al., 2005; Salmon et al., 2008), unless the Ames mice had been subjected to early-life GH treatment (Panici et al., 2010). In this context it is noteworthy that fibroblasts derived from Snell dwarf and control mice do not differ in stress resistance et al., at this age GH levels are largely derived from the and are not by the of the newborn mouse GH exposure in the to mouse (Sadagurski et al., 2015b), as as availability by the size of the nursing litter (Sadagurski et al., also the hypothalamic inflammatory in mice 18 months of age or Ames mice treated with GH also that controls in levels, of and in of and its irisin by muscle, and plasma levels of GPLD1, and brain levels of factors and (Li et al., showing that these early-life environmental factors can have effects on physiological of high to health maintenance and The data here another for early-life modulation of late-life and mortality by manipulation of social odors. We to study this because of data showing effects of early exposure to same-sex and opposite-sex odors on the pace of mouse development and sexual maturation. We hypothesized that exposure to same-sex which development in some circumstances (Drickamer, Drickamer, might also delay the rate of and we that female mice exposed early to odors from adult females had longer lifespan than females exposed to Male odor did not have this effect on female The FU-treated female mice also the decline in core body temperature, there was no difference in grip strength or rotarod other of late-life physiological Male mice, in contrast, did not lifespan or performance effects in response to either same-sex or opposite-sex odors. We do not the for the lifespan effect in the FU-treated female mice, or we no such effect in male these are we to in future We that the genotype of the mice exposed to odor was not the same as the genotype genetically from which urine and bedding was The use of a genetically mouse as odor would mean that the odor animals would be exposed to and odorants produced by strains of mice. We this as a strength because different mouse strains produce different olfactory et al., and exposing mice to odors from a of genotypes would increase the of exposure to an that influences lifespan. However, different mouse strains also express different olfactory receptors et al., and could show different physiological responses to bedding from different responses in mice to odors from the same or different of mouse has been as similar et al., et al., 2018). In possible that different lifespan responses may if mice are exposed to odors from their genotype. The effects of odor exposure on lifespan were not modified by deletion of Gαo in the olfactory This G protein is important in responses to and in basal VSNs et al., and has been shown to inhibit priming responses to male odors in females, including changes in sexual maturity (Oboti et al., 2014). We this experiment using this because this deletion is to inhibit priming effects, but not affect normal body development or as has been observed with manipulations that olfactory system function et al., research with this has largely on responses to male odors, but in this study we observed changes in lifespan in response to female odors. of female mice to female odors has been shown to a different subset of VSNs compared to male including VSNs that use a different G protein in signal transduction et al., 2018). effects of female odors on female survival may be by a subset of VSNs different from those that were in this is also possible that effects could be by detection of odorants by the olfactory our results were not significantly by the olfactory genetic manipulation employed in this it is possible that a different sensory might be in these effects. we the that with the bedding, than the urine odor and bedding odor these effects. this be a factor that only influences female lifespan, and can do so when females are exposed only over the first 60 days of life. in sexual maturity with exposure to odors can be initiated by exposure to the urine of a weaning, effects on female body weight have been observed when odor exposure prior to (Cowley and Wise, which is we an extended treatment to the chance of effects on survival. with a treatment time weaning, and with only urine or specific would to the signals and developmental that are important subsequent changes in survival. of a different genetic of impaired for example of the layer of the or a olfactory system mouse would to the of the olfactory in these effects. of the effects of sensory on aging and lifespan in back at least to the work of and in and and in the a of sensory including and have as important of aging across et al., et al., and of and to for example, the of and influence of including and daily (Miller et al., et al., of these have on how of through and can modulate and lifespan and et al., et al., 2014). there is no that to aging in and to be to in mice, do provide that to be and and such as are for aging effects in and et al., et al., 2014). seems to test these same influence lifespan in mice, either or in the sensory cues of the by our as we this is the first that lifespan can be in a by olfactory signals, or factors in soiled bedding and the work that very young mice, and animals of other are to of the social environment and their development to their as a slow the rate of aging and health in later life. Methods Mice We an experiment on changes in sexual maturity with urine and bedding exposure using mice. The of the mice used in the experiment were female are and male are The of the test mice were are and are mice used in the lifespan experiment were on a of the and mouse They a of the that had been at and These were with mice a of that expression under the control of olfactory protein (Li et al., research has shown that expression is strongly reduced in the of mice, and Gαo protein is in the sensory and accessory olfactory et al., contrast, expression is in the olfactory and in brain et al., their development and these mice have been as a at and have been used to show that mice show in of reproductive (Oboti et al., and et al., et al., in to diminished responses to specific
Source: eLife
Language: English
Subfield: Aging
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4319435841
Cites:
Cited by:
Related to: 10