Effects of inbreeding and elevated rearing temperatures on strategic sperm investment
Data files
Jun 07, 2024 version files 122.74 KB
-
Analysis_1.csv
19.23 KB
-
Analysis_2.csv
8.94 KB
-
Analysis_in_method.csv
21.81 KB
-
Chung_et_al_guppy_sperm.R
5.80 KB
-
Raw_data.xlsx
59.82 KB
-
README.md
7.14 KB
Abstract
Males often strategically adjust the number of available sperm based on the social context (i.e., sperm priming response), but it remains unclear how environmental and genetic factors shape this adjustment. In freshwater ecosystems, high ambient temperatures often lead to isolated pools of hotter water in which inbreeding occurs. Higher water temperatures and inbreeding can impair fish development, potentially disrupting sperm production. We used guppies (Poecilia reticulata) to investigate how developmental temperature (26 °C, 30 °C) and male inbreeding status (inbred, outbred) influence their sperm priming response. We also tested if sperm priming was affected by whether the female was a relative (sister), and whether she was inbred or outbred.
We ran two separate analyses to address our research questions. First, we ran a linear mixed model (LMM) to investigate the effects of male inbreeding status (inbred, outbred), rearing temperature (warm, control), and social environment (no female, unrelated outbred female, related female) and all three two-way interactions on sperm priming response (i.e., the total number of sperm produced in 7 days). Second, we noted that inbred males might produce fewer sperm than outbred males in the presence of a related female due to their sister being inbred and therefore of lower quality (e.g., less fecund). To test whether the observed effect of male inbreeding status was confounded by the related female’s inbreeding status, we ran an additional LMM exclusively for inbred males. We separated the effects of inbreeding status of the female and her genetic relatedness to the male by considering three types of female (inbred related, inbred unrelated, outbred unrelated) that inbred males encountered.
There was no effect of rearing temperature; male inbreeding status alone determined the number of available sperm in response to female presence, her inbreeding status and her relatedness. Inbred males produced significantly more sperm in the presence of an unrelated, outbred female than when no female was present. Conversely, outbred males did not alter the number of sperm available in response to female presence or relatedness. Moreover, inbred males produced marginally more sperm when exposed to an unrelated female that was outbred rather than inbred, but there was no difference when exposed to an inbred female that was unrelated versus related. Together, a sperm priming response was only observed in inbred males when exposed to an outbred female. Outbred females in our study were larger than inbred females, suggesting that inbred males strategically allocated ejaculate resources towards females in better condition.
https://doi.org/10.5061/dryad.wstqjq2vt
Description of the data and file structure
– Variables description for csv. files –
For all files (“Analysis_in_method”, “Analysis_1”, “Analysis_2”):
Male_id: male identity
Adult_age: male’s age at testing (days)
Male_size: male’s body size at testing (mm)
Brood_id: male family identity
Male_breeding_status: male inbreeding status (“Inbreed” or “Outbreed”)
Male_temp: male rearing temperature before maturation (“Control” = 26°C; “Warm” = 30°C)
sperm_count: male sperm count. The cell was left empty if no sperm was stripped from the male.
For both “Analysis_in_method” and “Analysis_1”:
Social_environ: male social environment (“No_female” = no female; “Nonrelated_outbred” = a non-related outbred female; “Related” = a related female)
For both “Analysis_1” and “Analysis_2”:
Female_size: female body size (mm). The cell was left empty if the social environment (“Social_environ”) indicated the absence of a female (“No_female”).
For “Analysis_in_method” only:
Likelihood_of_producing_sperm: The likelihood that we extracted any sperm from a male (“1” = sperm extracted successfully, “0” = no sperm extracted)
For “Analysis_2” only:
Female_type: Female inbreeding status and relatedness (“Inbred_nonrelated” = a non-related inbred female, “Inbred_related” = a related inbred female, “Outbred_nonrelated” = a non-related outbred female)
For “Raw data”:
“Male ID”: Male identity
“Male standard length (mm)”: male body size (the snout tip to the base of caudal fin)
“Male block ID”, “Male brood ID”, “Male breeding status”, “Crosstype”: we randomly selected and paired two outbred families to create “blocks” of inbred and outbred fish (e.g., families A and B for block 1, families C and D for block 2). Within each block a male and a female from either the same family (AA or BB) or different families (AB or BA) were paired as above. The column “Crosstype” indicated their genetic cross type (“AA”, “AB”, “BA”, “BB”). Please see figure 1 in Chung et al (2024) Behavioral Ecology for details.
“Male rearing temperature”: male rearing temperature before maturation (Control = 26°C; Warm = 30°C).
“Date of male maturity”, “Start date of the sperm priming phase”: Timeline for the experiment.
“Social environment”: No_female = no female; Nonrelated_outbred_female = a non-related outbred female; Nonorelated_inbred_female = a non-related inbred female; Related = a related female
“Female standard length (mm)”: female body size (the snout tip to the base of caudal fin). The cell was marked as “NA” (not available) if there was no female in the social environment (see above).
“Dilute volume of sperm solution (µL)”: the volume of saline solution (0.9% NaCl) placed to dilute extracted sperm for a concentration suitable for counting. The cell was marked as “NA” (not available) if we failed to extract any sperm from the male.
“Subsample 1”, “Subsample 2”, “Subsample 3”, “Subsample 4”, “Subsample 5”: the number of sperm in five randomly selected subsamples on the 20-micron capillary slide for each male. The cell was marked as “NA” (not available) if we failed to extract any sperm from the male.
“Average (sperm count in field)”: The mean sperm number of the five subsamples (see above). The cell was marked as “NA” (not available) if we failed to extract any sperm from the male.
“total sperm count”: The overall sperm count from a male, calculated using the dilution factor of the sperm solution, the mean sperm number averaged from five subsamples, and the visible field of the slide under the microscope. The cell was marked as “NA” (not available) if we failed to extract any sperm from the male.
– Analyses in R studio –
The analysis can be re-run using the R code called “Chung_et_al_guppy_sperm”. For each response variable (the likelihood of extracting sperm from a male, and their sperm count), we used mixed models (glmmTMB or lme4 package) including brood ID as a random factor to account for repeated measurements of the same family. Male size and age at testing were both standardised (mean = 0, SD = 1) and considered as separate covariates in the models. For all analyses (see below), we included male breeding status and rearing temperature and their interaction in initial models. A non-significant interaction was removed from the final model to interpret the main effects. Full model outputs (for both initial and final models) are provided in the supplementary material. Wald chi-square tests were performed to calculate the P value using the Anova function in the car package (R version 4.4.0). We used type III sums of squares for models with the interaction term and type II sums of squares for models without the interaction. Results are presented as mean ± SE. The significance level is set at alpha = 0.05 (two tailed). We ran Tukey’s post hoc pairwise tests (emmeans package) if the treatment effect was significant. Any deviations from this approach are specified below.
A. Method: Likelihood that we extracted any sperm from a male (using dataset: “Analysis_in_method.csv”).
We considered the likelihood that we extracted any sperm from a male (“Likelihood_of_producing_sperm”: 1 = yes, 0 = no) as the response variable and treated Male_breeding_status (Inbreed, Outbreed) and Male_temp (Warm, Control) as separate fixed factors using a generalised linear mixed model (GLMM) with binomial error.
B. Main results: Effect of social environment on sperm priming response (using dataset: “Analysis_1.csv”)
We considered power-transformed sperm count (“sperm_count”) as the response variable. We included male breeding status (“Male_breeding_status”: Inbreed, Outbreed), male rearing temperature (“Male_temp”: Warm, Control), and social environment (“Social_environ”: No_female = no female, Nonrelated_outbred = non-related outbred female, Related = related female) as individual fixed factors and included their two-way interactions in the initial model. We ran linear mixed models (LMM) and checked the normality of model residuals using a Q-Q plot and the Shapiro-Wilk test (shapiro.test function) and the homogeneity of variances using Levene’s Test (leveneTest function).
C. Main result: Effect of female inbreeding status and female relatedness on inbred male’s sperm priming response (using dataset: “Analysis_2.csv”)
The models are very similar to the analysis (B), except for two changes. First, we only considered the sperm count (“sperm_count”) of inbred males, so we did not consider “Male_breeding_status” as a fixed factor. Second, we considered “Female_type (Inbred_nonrelated = non-related inbred female, Inbred_related = related inbred female, Outbred_nonrelated = non-related outbred female)” as a fixed factor instead of “Social_environ”. We considered the interaction between male rearing temperature (“Male_temp”) and “Female_type” in the initial model.