The effects of paternal dietary fat versus sugar on offspring body composition and anxiety-related behavior

Increasing evidence suggests that the pre-conception parental environment has long-term consequences for offspring health and disease susceptibility. Though much of the work in this field concentrates on maternal influences, there is growing understanding that fathers also play a significant role in affecting offspring phenotypes. In this study, we investigate effects of altering the proportion of dietary fats and carbohydrates on paternal and offspring body composition and anxiety-related behavior in C57Bl/6-JArc mice. We show that in an iso-caloric context, greater dietary fat increased body fat and reduced anxiety-like behavior of studs, whereas increased dietary sucrose had no significant effect. These dietary effects were not reflected in offspring traits, rather, we found sex-specific effects that differed between offspring body composition and behavioral traits. This finding is consistent with past paternal effect studies, where transgenerational effects have been shown to be more prominent in one sex over the other. Here, male offspring of fathers fed high-fat diets were heavier at 10 weeks of age due to increased lean body mass, whereas paternal diet had no significant effect on female offspring body fat or lean mass. In contrast, paternal dietary sugar appeared to have the strongest effects on male offspring behavior, with male offspring of high-sucrose fathers spending less time in the closed arms of the elevated plus maze. Both high-fat and high-sugar paternal diets were found to reduce anxiety-like behavior of female offspring, although this effect was only evident when offspring were fed a control diet. This study provides new understanding of the ways in which diet can shape the behavior of fathers and their offspring and contribute to the development of dietary guidelines to improve obesity and mental health conditions, such as anxiety.


Introduction
Obesity is a major public health crisis affecting many countries worldwide.It is connected to a wide range of comorbidities including a higher susceptibility for developing type 2 diabetes [1], cancer [2], cardiovascular disease [3], and a variety of mental illnesses [4,5].Studies have consistently found associations between hypercaloric diets (such as high-fat, or Western diets), poor metabolic health, and increased risk of anxiety and depressive-like behavior [6,7].It appears that overnutrition, obesity and mental disorders share a strong link, although the underlying mechanisms remain unclear.In recent years, the impact of the paternal environment and reproductive outcomes on offspring health are becoming more widely acknowledged [8,9].However, there are still gaps in our understanding of how specific macronutrients, and their balance within the diet, can affect male fertility and offspring health.Hence, there is an urgent need for targeted studies assessing dietary influences on male reproduction and paternally mediated programming.
Multiple neuro-pathologies have been associated with poor metabolic health and diet-induced obesity [4,10].Rodents fed a high-fat diet (HFD) have been shown to develop anxiety, depressive-like behavior and metabolic complications [11,12].Meanwhile, studies investigating non-stressed rats that were submitted to a high-sucrose diet have found an increase in anxiety-related phenotypes [13].Furthermore, a combination of both high fat and high sucrose appears to result in poorer memory performance and cognitive decline [14,15].
Although many mental disorders have been shown to have high heritability, and therefore a significant genetic component, many of the anxiety-related loci identified do not seem to fully explain the development of these conditions and phenotypic differences can occur even in the absence of genetically transmitted variation [16,17].It is likely that social and environmental factors also contribute to the risk of developing an anxiety disorder [18].Increasing evidence suggests that the perinatal environment may be particularly crucial in determining long-term consequences on health and disease susceptibility [19,20].Though much of the work in this field focuses on maternal influences, there is growing understanding that fathers also play a significant role in affecting offspring phenotypes [21,22].Aside from direct paternal care, epigenetic inheritance has gained traction in recent years as a means of passing down paternally driven phenotypes independent of the genome [23,24].Processes included in this form of regulation include DNA methylation, histone modifications, and non-coding RNAs.Furthermore, the influence of paternal environments on offspring may persist across multiple generations, indicating a transgenerational inheritance [25,26].
There are several studies that demonstrate how paternal experiences and phenotypes can be passed on through epigenetic mechanisms and other similar processes.Paternal HFD exposure prior to conception can affect offspring metabolic traits, including body composition, glucose intolerance, and adiponectin and leptin gene expression [27].Socially defeated fathers in rodent studies have been shown to transmit depressive and anxiogenic behaviors, with more distinct effects observed in male offspring [28].Paternal psychological stress from toxins and other traumatic exposures have been shown to reprogram offspring metabolic outcomes, such as hepatic gluconeogenesis [29], memory and cognition [30,31], and other behavioral outcomes [32][33][34].Just as metabolism and behavior can be linked within a generation, the metabolism and behavior of male mice may also be linked to their descendants across generations [35].
Along with paternal influences, and offspring environments, another factor that could interact to produce certain phenotypes is the sex of the individual.Sex differences occur in most non-communicable diseases, including obesity and metabolic syndrome [22,36,37], yet the potential role of non-genetic paternal effects in mediating this (and vice versa) is still understudied.A few instances of paternal effects have been shown to vary depending on the offspring's sex in complex ways [38].Studies show that paternal influences can have opposing effects on the same trait in sons compared to daughters [33], only be present in either sex [39], or affect them in different magnitudes [40].
Another important consideration that many studies in this field do not fully explain is whether the anxiogenic effect resulting from dietinduced obesity is driven by the hypercaloric nature of these diets or the specific components (e.g., fat or carbohydrate) per se.In this study, we focus on exploring how varying the carbohydrate and fat proportions within paternal diets (but keeping energy content similar) can affect metabolism, reproductive outcomes and anxiety-related behavior in the studs.Additionally, we investigate how this can influence offspring body composition and behavior.To do this, we assessed offspring body composition at 10 weeks of age and employed the elevated plus maze and open field maze-two common tools widely used in research to evaluate anxiety-associated responses, general activity, and locomotion.
Here, we show that high fat diets in male studs led to the most obese phenotype.We observed that increasing the percentage of sucrose versus starch in the paternal diet had little influence on body weight, fat mass, or fasting insulin levels, while increasing the percentage of dietary fat (while keeping energy density constant) resulted in increased body weight, fat mass, and fasting insulin levels.Meanwhile, in the offspring, sex specific interactions were observed in a variety of health outcomes including body composition and behavior, as influenced by paternal dietary manipulations.

Ethical approval
C57Bl/6-JArc mice were used in this study.All animal protocols were approved by Sydney University's Animal Ethics Committee (Protocol number 2019/1609) to ensure they were in line with the NSW Animal Research Act 1985 and the Australian code for the care and use of animals for scientific purposes 8th edition (2013).

Experimental diets
All diets are detailed in Table S1.

Stud diets
The studs were fed 1 of 3 diets (control, high-sucrose, high-fat) that were matched for metabolizable energy content (i.e., isocaloric) as much as was practically possible through the addition of non-digestible cellulose.The source of protein was casein, fat was soybean oil and carbohydrates were a mixture of sucrose, wheat starch and dextrinized starch.

Ø Control
The control diet (CD; SF18-025, Specialty Feeds, Glen Forrest, WA, Australia) was derived from a standard rodent AIN-93 G formulation, with 18.5 % protein, 64 % carbohydrate and 17.5 % fat making up the total calorie content of 14.4 kJ/g.The carbohydrate component consisted of 84.3 % starch and 15.7 % sucrose.

Ø High Sucrose
The high-sucrose diet (HSucrose; SF19-202, Specialty Feeds, Glen Forrest, WA, Australia) was derived from a standard rodent AIN-93 G formulation, with 18.5 % protein, 64 % carbohydrate and 17.5 % fat making up the total calorie content of 14.7 kJ/g.The carbohydrate component consisted of 52.8 % starch and 47.2 % sucrose.

Ø High Fat
The high-fat diet (HFat; SF19-203, Specialty Feeds, Glen Forrest, WA, Australia) was derived from a standard rodent AIN-93 G formulation, with 18.5 % protein, 42.3 % carbohydrate and 39.2 % fat making up the total calorie content of 14.7 kJ/g.The carbohydrate component consisted of 84.2 % starch and 15.8 % sucrose.

Ø Control
The control diet fed to pups was the same as the control diet fed to studs (CD; SF18-025, Specialty Feeds, Glen Forrest, WA, Australia).

Ø Western
A high-calorie, high-fat, and high-sucrose "Western" diet (WD; SF18-050, Specialty Feeds, Glen Forrest, WA, Australia) was used as an obesogenic diet for pups in this study.The Western diet designed for this study was again derived from a standard rodent AIN-93 G, with 10 % Protein, 50 % Carbohydrate, and 40 % Fat, making up the total calorie content of 17.3 kJ/g.The fat component consisted of 84.7 % lard and 15.3 % soybean oil.The carbohydrate component consisted of 60.5 % starch and 39.5 % sucrose.

Dam diet
Dams were fed a non-purified brown chow diet.
T. Freire et al.

General husbandry and cage allocations
All animals in this study were housed in cages at the Charles Perkins Centre, University of Sydney with an ambient temperature of 22 • C and constant access to water.Fifty-four C57Bl/6-JArc 3-4-week-old studs were purchased from Animal Resources Centre (Perth, WA) in two cohorts, and housed 3 per cage until mating.They were started on experimental diets immediately upon arrival, so the fathers were exposed to treatments during adolescence, and over multiple rounds of spermatogenesis.Studs were divided into 2 cohorts.The first cohort were not mated and were only tested on the elevated plus maze (n = 6).The second cohort of males were mated after behavior testing on both the elevated plus maze and open field test.From the 2nd cohort studs offspring, 2 pups of each sex were restricted to either a Control or Western diet from 3 weeks of age, yielding n = 8-10/sex/ diet group.From up to a week prior to behavioral testing at 15 weeks of age, animals were temporarily moved to the Charles Perkins Centre Animal Behavioural Facility for acclimatization.Food intake and body weights continued to be measured once a week.
Thirty-six 16-week-old female C57Bl/6-JArc mice that had successfully raised a litter of pups (proven fertile) were purchased from Animal Resources Centre (Perth, WA) and housed individually in pre-allocated mating cages.Dams were allowed to acclimatize to the facility for two weeks prior to mating and exposed to bedding of their assigned mate to prime them for breeding.

Breeding protocol
Studs were housed individually during the mating period to prevent fighting with fellow males (n = 18).For each breeding cycle, studs were paired with their allocated female overnight for four consecutive nights (n = 12 mating pairs per diet) and returned to their home cages during the day to minimize access to dam food.Although this protocol meant that studs had access to brown chow during mating, it was considered more important that dams were never exposed to experimental diets.To confirm successful copulation, the stage of estrous cycle and presence of vaginal plugs was recorded.During gestation, dams were housed in individual cages and body weights were measured twice weekly to detect pregnancies.The date and number of any offspring was recorded.To prevent infant cannibalism, litter body weights were not measured until three days post birth.Litters were continually monitored until they were weaned at 3 weeks of age, where the sex of the offspring was recorded.

Body composition
Body weights were measured weekly for all mice, unless specified.Body composition (fat mass and lean mass) was assessed using an EchoMRI 900 (EchoMRI, TX, USA) at 18 weeks of age in studs (n = 12/ sex/group) and 10 weeks of age in offspring (n = 8-10/sex/group).Organ weights were measured at cull including liver, gonadal and subcutaneous white adipose tissue, brown adipose tissue, and quadriceps.

Glucose and insulin metabolism
Glucose tolerance tests were performed on male and female offspring (n = 8-10/sex/group) at 10 weeks of age.Mice were fasted for 4 h prior to testing.Basal blood samples were obtained by tail tipping and blood glucose measured using a clinical glucometer (Accu-Chek Performa, Roche Diagnostics Australia Pty Ltd).Glucose (2 g kg-1 lean mass) was then administered via oral gavage.Blood was collected at baseline, 15, 30, 45, 60 and 90 min from the original tail wound and serial tail tipping was not required.The incremental area under the curve (iAUC) was calculated.The iAUC indicates the time taken to clear a bolus dose of glucose from the bloodstream.Blood that was collected from the basal timepoint after 4 h of fasting from 10-week-old pups were also used for the analysis of insulin levels using an enzyme-linked immunosorbent assay (ELISA) following manufacturer's instructions (Crystal Chem IL) (n = 8-10/sex/treatment group).

Behavioral analysis
Behavioral traits of both studs and offspring were assessed at weeks of age using an Elevated Plus Maze (EPM) and Open Field (OF) Test.Both techniques have been widely validated previously and exploit the natural aversion of rodents to exposed fields [41][42][43][44].They are, therefore, based on an anxiety-promoting agent, such as an unprotected, elevated field or an open area, the anxiety level being expressed by the number of entries into, and the length of time spent in the aversive area.

Elevated plus maze
The elevated plus maze consisted of 2 closed arms (30 × 5 cm), open arms (30 × 5 cm), and a central zone (5 × 5 cm).The maze was elevated 50 cm above the ground.The light intensity on the open arms of the maze was 700-750 lx, and in the closed arms was 175-225 lx.Each mouse was placed into the center of the maze facing the left open arm for consistency at the beginning of recording using Logitech webcam software (Logitech; Lausanne, Switzerland).Mice were exposed to the elevated plus maze for a total of 5 min and the apparatus was cleaned thoroughly between animals.The data were analyzed using TopScan Image Analysis Software (CleverSys Inc; VA, USA) for the parameters of percentage time spent in the open and closed arms of the maze and entries into the arms.An entry was defined by crossing the dividing line between an arm and the center platform with all four feet.Other parameters measured were total distance travelled (in mm), velocity (the animal's average speed when moving in a straight line throughout the maze), speed (separated into fast or slow speed thresholds; slow speed animal moving at < 20 mm/s, fast speed = animal moving at >20 mm/ s,) and motion (separated into 'fast' or 'slow' detectable motion thresholds defined by the program, where a value of 0 indicates the animal is static; slow motion = animal moving from 0 to 0.05, fast motion = animal moving at >0.05).

Open field maze
As an additional way to measure anxiety related behavior and locomotor activity, an open field test was used.The test was performed by placing a mouse into the center of an open-field arena with a white base (40 × 40 cm) and black walls bordering the outside (20 cm) and allowing the mouse to explore for 5 min.Bright overhead lighting was approximately 225 lx inside the arenas.Activity in the open-field was quantified by a computer-operated TopScan Image Analysis Software (CleverSys Inc; VA, USA).Total distance (locomotor activity), movement time (in seconds), movement speed (cm/s), and center distance (the distance traveled in the center of the arena) were recorded.

Statistical analysis
Data are presented as means ± SEM and statistical significance determined when P < 0.05.Stud data was analyzed using ANOVA in GraphPad Prism (v9.0.2;GraphPad Software Inc., La Jolla, CA, USA).Post hoc analysis was performed using Tukey's multiple comparisons test unless otherwise stated in the relevant figure legends.Offspring data was analyzed using mixed effect models in R (RStudio version 1.4.1717;Vienna, Austria) with lme4 [45] and LmerTest [46] packages.Stud diet, Pup diet, Sex, and their interactions were included as fixed effects.Dam was included as a random variable to account for litter effects.Litter size at weaning and body weight were initially included as covariates in all models and then removed if shown to not have a significant effect.Correlation between parameters was assessed by Spearman rank correlation in RStudio version 1.4.1717.All other statistical parameters are included in the figure legends.
T. Freire et al.

Increased dietary fat, but not carbohydrates, promoted adiposity and increased entries into the open arms of an elevated plus maze
High-fat-fed studs were heavier than both control and high-sucrose studs when measured at 18 weeks of age, 14 weeks after commencement on each allocated diet (Fig. 1A, Table S2).The increased body weight observed in high-fat studs was attributed to elevated fat mass, which was significantly higher than both control and high-sucrose studs (Fig. 1B, Table S2), whereas there was no significant difference in lean mass across groups (Fig. 1C, Table S2).Total energy intake was not statistically significant across the three diets (Fig. 1D) but was positively correlated to body weight (R2= 0.0988.P = 0.0206).No significant differences were found in the mating success and average number of pups per litter between paternal diet groups (Fig. 1E, F).However, a stud by sex interaction was observed in the ratio of male to female offspring at weaning (Fig. 1G, p = 0.0472).A significant diet effect was observed in the number of entries into the open arms of the elevated plus maze (EPM) (Fig. 1H, Table S2), where males fed a high-fat diet had increased entries compared to both control (p = 0.0179) and high-sucrose (p = 0.0195) groups, indicating increased boldness.Despite this, no significant effects were detected in the number of entries into the closed arms (Figure S1A, Table S2) and center zone of the EPM (Figure S1B, Table S2).Additionally, no significant differences were found in terms of time spent by the groups in these three regions (Fig. 1I; Figure S1C,D; Table S2).There were also no significant effects of stud diet on total distance travelled (Fig. 1J; Table S2), velocity (Fig. 1K; Table S2), time travelling in fast or slow speed (Figure S1E,F; Table S2), and fast and slow motion in the EPM (Figure S1G,H; Table S2).
No statistically significant effects of changes in dietary fat or carbohydrate were detected in a wide range of outcomes measured in open field (OF) maze (Figure S2; Table S3).No significant effects were found in the total distance travelled in the maze, velocity, entries into and time spent in either the outer or center zones, time travelling in fast or slow speed, and fast and slow motion in this arena.

Sex specific responses to paternal and offspring diets are observed in body composition of offspring at 10 weeks of age
Total energy intake consumed up to 10 weeks of age was influenced by pup diet and sex (interaction effect p < 0.0001), but no statistically significant influence of paternal diet was detected (Fig. 2A, Table S4).We found that at 10 weeks of age, although all WD-fed mice had higher energy intakes compared to CD mice, the magnitude of this effect was more prominent in females (Fig. 2A; Table S4).Both paternal and offspring diets impacted various body composition measures for both sexes when measured at 10 weeks of age, but to different degrees.When looking at offspring body weights, we found an interactive effect between stud diets and sex (p = 0.0136), as well as pup diets and sex (p = 0.0097; Table S4).Male offspring of fathers fed high-fat diets were heavier when fed either a control or Western diet, whereas paternal diet appeared to have little influence on the weight of female offspring at 10 weeks (Fig. 2C,D).The distinction between CD and WD groups was more defined in females, where WD mice were heavier (Fig. 2C; Table S4).Looking at fat mass, we detected a main effect of pup diet (p = 0.0135) and sex (p = 0.0282), but no effect of stud diet (Fig. 2E,F; Table S4).For lean mass, stud diet x sex (p = 0.0020) and stud diet x pup diet interactions were observed (p = 0.0215; Fig. 2G,H; Table S4).As for body weight, effects of paternal diet on lean mass were stronger in male offspring than female offspring.A pup diet x sex interaction was found when looking at offspring body weight change between 3 and 10 weeks of age (p = 0.0002).The lean mass of pups from high-fat fathers was greater than control pups regardless of whether they were fed a control or Western diet.However, the effect of a paternal high-sugar diet depended on the pup on the pup diet, where increased lean mass was observed on a Western diet only (Fig. 2G,H).
In terms of fasting insulin, we detected a main effect of pup diet (p = 0.0002) and sex (p < 0.0001; Fig. 2K,L; Table S4).No significant effects of stud diet were detected on outcomes of glucose tolerance tests of offspring at 10 weeks of age.No significant differences were observed for the incremental area under the curve (iAUC) levels of the pups, a measure of glucose tolerance (Fig. 2N,P; Table S4).

Interactions between paternal diet, pup diet and sex were observed in offspring body composition at 10 weeks of age and anxiety related responses at 15 weeks of age
A statistically significant three-way stud diet x pup diet x sex interaction was observed in the amount of time spent in the closed arms of the EPM (p = 0.0412; Fig. 3A,B; Table S5).Control-fed female offspring from control fed studs spent the smallest percentage of time in the open arms (Fig. 3C,D; Table S5), and inversely, the largest percentage of time in the closed arms.In males, pups from high-sucrose fathers spent more time in the open arms, and the least in the closed arms, regardless of pup diet.No differences were found between groups for the time spent in the  decision-making center zone (Fig. 3E,F; Table S5).Similar trends were observed in open arm entries, although it did not reach significance (Figure S3A,B; Table S5).No significant effects were detected in the number of entries into the closed arms of the maze either (Figure S3C,D; Table S5).

Differences in movement were observed in offspring in an elevated plus maze at 15 weeks of age
Aside from anxiety-related responses, the elevated plus maze was also used to observe movement patterns of the offspring.An interaction between stud diet and pup diet was found in the motion of the offspring (p = 0.0200; Fig. 4A,B; Table S6).In CD mice, pups from control-fed fathers spent the least amount of time in fast motion in the EPM.In WD animals, pups from high-sucrose fathers spent more time in fast motion than the other groups., while offspring of high-fat fed males spent the least percentage of time in fast motion.A main pup diet effect was found in the speed of the animals (p = 0.0200; Fig. 4C,D; Table S6), the total distance travelled (p = 0.0062; Fig. 4E,F; Table S6), velocity (Fig. 4G,H; Table S6), where those fed a control diet in adulthood were observed to cover larger distances, have increased velocity and spent more time where speed was >20 mm/s when moving around the EPM.Removing body weight from the statistical model also led to a statistically significant sex effect being detected in these outcomes (Table S6).
An open field (OF) maze was used to test behavior and movement patterns of 15-week-old male mice.Contrary to the results found in the elevated plus maze, no significant effects were found as an effect of stud diet, in a wide range of outcomes measured (Figure S5; Table S7).However, pup diet and sex were shown to have main effects on the total distance travelled in the maze (Figure S5A,B; Table S7) and velocity (Figure S5C,D; Table S7).Only sex influenced the speed of the animals in the OF (Figure S5I-L; Table S7).An interactive effect between pup diet and sex was also found to be present in the motion of the offspring (Figure S5M-P; Table S7).

Correlations between stud and offspring traits
In control-fed male offspring, EPM distance and velocity negatively correlated with studs' EPM distance and velocity (Fig. 5A).In females, EPM closed bouts negatively correlated with stud bouts in OF, and positively correlated with fast motion in EPM.EPM fast motion was negatively correlated with studs closed bout in EPM (Fig. 5B).The distance travelled and velocity in the EPM of WD males were negatively correlated with stud energy intake, time spent in open arms and fast motion (Fig. 5C).For these same animals, entries into the closed arms negatively correlated with stud distance travelled and velocity, fast motion, percentage time in the open arms of the EPM, and positively correlated with stud percentage time in the open arms.In WD females, the percentage time spent in open arms negatively correlated with stud body weight, energy intake and lean mass (Fig. 5D).Their entries into closed arms positively correlated with stud body weight and fat mass, while percentage time in closed arms positively correlated with stud energy intake.

Discussion
In this study, we set out to isolate the effects of dietary fat and carbohydrate in paternal diet, without the confounder of energy density.Using two behavioral assays that exploit the natural aversion of rodents to exposed fields-namely, the elevated plus maze and open field maze we sought to characterize the effect of our dietary manipulation on both paternal and offspring behavior.Increasing the percentage of dietary fat (while keeping energy density and protein content constant) resulted in increased body weight and fat mass in male studs.Despite finding minimal effects of diet on stud behavior, we observed marked, sexspecific differences in offspring behavior and locomotion, influenced by both paternal and current diet.This included a three-way stud diet x pup diet x sex interaction, which was observed in the amount of time spent in the closed arms and an interaction between stud diet and pup diet in the motion of the offspring in an EPM.Further correlations found between stud body composition, behavioral outcomes and offspring behavior suggest that paternal effects are highly interactive, depending on factors such as offspring environment and sex, within and across generations.
We initially investigated how changing the percentage of dietary fat and sucrose to starch ratio in the diet of 15-week-old male mice can affect their body composition.We observed that increasing the percentage of sucrose in the paternal diet had little influence on body weight, fat mass, while high fat diets in male studs led to the most obesogenic effect.Additionally, no effects of the diets used in our study were found on mating success, litter size, or offspring sex ratio.Since fat is a calorie-dense macronutrient, it is often considered as a crucial constituent of obesogenic diets.However, studies that explore the association of high fat diets and development of cardiometabolic disorders often produce contradictory findings [47].In animal studies, an obese model is often achieved through an excess of total energy derived from typically palatable, energy rich diets such as "Western" diets that are high in fat and carbohydrate.In addition to dietary fat, the ratio and source of dietary carbohydrate is also an important predictor of metabolic health [48].Surprisingly, we found that a three-fold increase in dietary sucrose had no significant effects on body composition or metabolic health of studs.This supports recent findings explaining how it is not necessarily a high percentage of certain carbohydrates, but an equal combination of monosaccharides (typically a 50:50 combination of glucose and fructose) that promote deleterious metabolic outcomes [49].
Unlike what has been shown in past studies [27,50,51], the obesogenic high fat diet of the fathers did not result in an increased risk of lipid accumulation and susceptibility to obesity in their offspring in this study.We observed that male offspring from high fat fed fathers had elevated body weights at 10 weeks, but this was instead attributed to an increased lean mass rather than fat mass, contrasting to previous studies that have reported fat mass changes in offspring [52].In fact, it was the male pups from chow-fed fathers that were found to have the greatest amounts of fat mass among the three stud diet groups, irrespective of adult diet.Meanwhile, despite the lack of effects observed in studs, increased sucrose content in the paternal diet led to increased lean mass of pups fed a western diet at 10 weeks of age.The mechanism underlying this effect is unknown as no statistically significant difference in energy intake was detected in the offspring.
It could be possible that the disparity of our findings compared to past studies are due to the paternal diet compositions used.High energy diets have been associated with the declined functioning of the male reproduction and consequently affects the testicular physiology, disrupting its metabolism and bioenergetic capacity [53].This has been associated with various mechanisms, such as sperm transfer RNA-derived small RNAs (tsRNAs) exhibiting changes in expression profiles and RNA modifications [54].Keeping the energy content constant in all three diets could have maintained a type of paternal influence in the offspring, without leading to a higher susceptibility to obesity and metabolic decline.
Much of the current literature suggest that high-fat diets increase the susceptibility of an individual to not only metabolic disorders such as obesity, but also mood disorders such as depression and anxiety [7,12].In mammals, the proposed mechanisms behind this include decreased serotonin levels [55], altered dopamine transmission [56,57], suppression of cAMP/PKA signaling in the hypothalamus [58], and increased glucocorticoid receptor expression along with increased stress-induced corticosterone release [59].There is also a potential link with inflammatory pathways [60], through the release of inflammatory molecules such as cytokines, that can interact with all these mechanisms [61].On the other hand, increasing the sucrose content of the carbohydrate component within the diet has been shown to limit activation of the stress system by activating the reward circuits in the brain [62], but conversely can also promote stress-driven emotional and addictive behaviors and anxiety [63,64].Overall, most research find that obesogenic diets are associated with increased anxiety, although there is a large amount of unexplained variation among studies, with some studies reporting opposite trends [65].
In this study, we observed that high-fat studs had increased open arm entries, which could indicate a predisposition to risk-taking behavior.We found that males fed high-fat diets entered open arms more frequently than the control and high-sucrose diets, the opposite to most studies that show rodents on HFDs enter closed arms more, and open arms less often [7,12].This difference may be due to a variety of factors such as animal strain [12], or age they were started on diets [7].However, many of the findings that link high-fat diets to a decline in mental condition are confounded by the high caloric nature of the high-fat diets used, meaning it is not clear which dietary factor is driving behavioral change-calories or macronutrients.Contrastingly, we have used diets with similar caloric contents and have instead focused on manipulating the proportions and source of fat and carbohydrates between the three diets.This suggests that it is most likely the energy excess in high-fat and Western diets, not increased fat content per se, that results in an increased susceptibility to anxiety-like behavior as observed in previous studies.
We then chose to perform the same tests on the studs' male and female offspring at the same 15-week timepoint.Interestingly, although we found that HSucrose diets did not affect stud behavior, their offspring behaved differently to controls.It was observed that offspring fed WD (high in fat, sucrose, and energy density) post-weaning were generally spending more time in the closed arms, indicating an increased anxietyrelated response similarly seen in other studies [10,66].Additionally, within the offspring fed control diets, female pups of both HFat and HSucrose studs spent less time in closed arms, indicating reduced anxiety.Interestingly, the stud control diet in this study has a lower amount of the methyl donor choline (Table 4.1; 1.5 g/kg) in comparison to HSucrose and HFat diets (2.1 g/kg).This could suggest a link to previous reports, where the levels of methyl donors in the paternal diet led to changes in memory-related gene expression, neural function, and anxiety-like behavior in offspring [67,68].However, further evidence is required to confirm these findings.
An interaction between stud diet and pup diet was found in the movement of the offspring, suggesting a more erratic behavior in offspring from high-sucrose fed fathers and those on Western diet in adulthood.Though erratic behavior has previously been reported to have resulted from high-sucrose diets [69], a paternal effect is lesser known.It is also interesting that influences of paternal diet are reflected in their offspring but not in the studs themselves.The mechanisms that mediated these paternal effects are unknown, but could include epigenetics [70,71], hypothalamic-pituitary-adrenal (HPA) axis dysregulation [72], inflammation [61], and the alteration of signaling pathways [58,59].Other studies also suggest that the hippocampus plays a crucial role in influencing performance in behavioral tests and anxiety-related behavior [73].
In this study, we found that the nature of paternal effects that manifested in offspring depended on the offspring's current environment, with different effects observed in control-fed and Western diet-fed offspring.In addition, a direct effect of pup diet was found in general activity of offspring, where those fed a control diet in adulthood were observed to cover larger distances and have increased velocity and speed when moving around the maze.These results were also mirrored in the open field maze.Offspring on Western diets displayed a higher tendency towards anxiety-like behavior; something that is in line with much of the literature in this field [15,74,75].Since the Western diets used in this portion of the study were higher in caloric density (17.3 kJ/g vs 14.7 kJ/g in SD), these results support the idea that anxiety-related behavior is impacted more by increased energy content than specific macronutrient composition.
Sex is also a key factor that defines the consequence of paternal diet on their offspring.Aside from the well-recognized sex differences in body weight and morphology [76,77], neurophysiological responses [78] and hormonally mediated changes of the inflammatory response [79][80][81], there are also studies indicating the sex-specific nature of parental, and even grand paternal inheritance [82][83][84].Some studies have demonstrated that female offspring and grand offspring, but not males, can exhibit anxiety-like behavior and social deficits [85,86].It has been proposed that males can transmit such effects to subsequent generations, even when they are not displaying the phenotype [86].Compatible with this, we found that offspring displayed differences in behavior reflecting paternal diets, even when we did not detect a significant difference between the studs themselves.Indeed, previous studies have also shown that parental effects may be more distinct in one offspring sex over the other [87].These have been attributed to differential epigenetic processes in the placenta, noting the abundance of X-linked genes involved and the early unequal gene expression by the sex chromosomes between males and females [88,89], which can have further consequences for the development of certain traits in offspring.
We explored associations between stud body composition and behavior outcomes and offspring behavior and found several effects that were correlated between fathers and offspring.It has previously been reported that phenotypic changes can alternate in presence or direction from one generation to the next, from father to offspring to grand offspring.This has been shown in the context of paternal stress regulation through corticosterone [33], emphasizing that environmental manipulations can alter outcomes such as anxiety and depression-related behavior across multiple generations.In some studies, paternal effects have also been observed to reduce offspring performance in the next generation, especially under highly variable environments [90].Moreover, most of the significant correlations we found in this study stemmed from the Western diet-fed offspring groups, for both sexes.
Altogether, this work highlights the complex interactions between paternal transmission of obesity-associated phenotypes, sex, and offspring metabolism.Our findings show that paternal diet can impact offspring via a network of direct and context-dependent effects, including offspring sex and diet.Appreciating these interactions may go some way towards helping to understand some of the apparent contradictions in previous studies and help to frame further research.

Fig. 1 .
Fig. 1.Increased dietary fat, but not carbohydrates, promoted adiposity and increased entries into the open arms of an elevated plus maze.Several parameters were measured in 18-week-old males including (A) Total energy intake from 3 to 8 weeks of age, (B) body weight, (C) fat mass and (D) lean mass.Other mating results for studs and paired dam were also recorded, including (E) mating attempts taken to produce a litter of pups, (F) number of pups per litter at weaning, and (G) number of pups per litter separated by sex indicating both males and females.Using an elevated plus maze, other traits were measured at 15 weeks of age, including (H) number of entries into open arms, (I) Percentage of time spent in open arms, (J) Total distance travelled in maze and (K) Velocity (n = 18 animals/group).All bars indicate means ± SEM.Significant p-values are provided on the relevant figures.

Fig. 2 .
Fig. 2. Sex specific responses to paternal and offspring diets are observed in body composition of offspring at 10 weeks of age.Average cumulative food intake was measured up to 10 weeks of age for (A) Females and (B) Males (n = 4-5 cages/sex/treatment group).Body weight at 10 weeks of age for (C) female and (D) male offspring.Fat mass at 10 weeks of age for (E) female and (F) male offspring.Lean mass at 10 weeks of age for (G) female and (H) male offspring.Body weight change between 3 and 10 weeks of age for (I) female and (J) male offspring.Fasting insulin for (K) female and (L) male offspring after 4 hrs of fasting.Blood glucose levels and iAUC levels from an oGTT for (M-N) female and (O-P) male offspring (n = 8-10 animals/sex/group).All bars indicate means ± SEM.Significant p-values are provided on the relevant figures.

Fig. 3 .
Fig. 3. Interactions between paternal diet, offspring diet and sex were observed in offspring anxiety related responses at 15 weeks of age.Using an elevated plus maze, several parameters were measured including percentage of time spent in closed arms for (A) female and (B) male offspring, percentage of time spent in open arms for (C) female and (D) male offspring and percentage of time spent in center zone for (E) female and (F) male offspring.(n = 8-10 animals/ sex/group) at 15 weeks of age.All bars indicate means ± SEM.Significant p-values are provided on the relevant figures.

Fig. 4 .
Fig. 4. Differences in movement were observed in offspring in an elevated plus maze at 15 weeks of age.Using an elevated plus maze, several parameters were measured including Percentage of time when motion was >0.0500 (fast) for (A) female and (B) male offspring, Percentage of time when speed was >20 mm/s (fast) for (C) female and (D) male offspring, total distance travelled around the maze for (E) female and (F) male offspring and velocity of movements for (G) female and (H) male offspring (n = 8-10 animals/ sex/group) at 15 weeks of age.All bars indicate means ± SEM.Significant p-values are provided on the relevant figures.

T
.Freire et al.

Fig. 5 .
Fig. 5. Correlations between stud and offspring traits.Correlogram of stud metabolic outcomes and offspring behavioral parameters for (A) Male and (B) Female CD fed offspring and (C) Male and (D) Female WD-fed offspring.Circle size and color indicate correlation strength (Spearman's rho).Presence of a circle indicates a significant (p < 0.05) Spearman correlation.