Accumulating evidence suggests that the nutritional environment experienced by an individual during fetal and early infant development has long-lasting impacts on their metabolic health(Reference Langley-Evans1). In the context of the global epidemic of obesity and nutritional excess, there has been considerable interest in determining the effects of maternal overnutrition on the metabolic health of the offspring. The majority of these studies have utilised animal models and have consistently reported that maternal high fat feeding during pregnancy has detrimental effects on the metabolic health of both the mother and her offspring(Reference Ribaroff, Wastnedge and Drake2,Reference Ainge, Thompson and Ozanne3) . As a result, excess maternal fat consumption has been implicated as a key contributor to metabolic programming of long-term health and disease risk.
There is increasing evidence, however, that the impact of a high-fat diet on the metabolic health of the offspring depends not only on the amount of fat in the diet but also on the fatty acid composition(Reference Muhlhausler and Ailhaud4,Reference Draycott, Liu and Daniel5) . There has been particular interest in the role of two classes of PUFA, due to the substantive increase in the amounts of n-6 PUFA, predominately linoleic acid (LA), being consumed in the diets of many Western countries over the past few decades(Reference Blasbalg, Hibbeln and Ramsden6,Reference Sioen, van Lieshout and Eilander7) . This increase in the intake of LA has not been accompanied by substantial changes in the consumption of n-3 PUFA such as α-linolenic acid (ALA) and has therefore resulted in increases in the ratio of n-6:n-3 PUFA consumed in the diets of many Western countries(Reference Blasbalg, Hibbeln and Ramsden6,Reference Ailhaud, Massiera and Weill8) .
The increasing dominance of n-6 over n-3 PUFA in modern Western diets has considerable biological significance, since the n-6 and n-3 fatty acid families utilise the same enzymes for production of longer chain bioactive derivatives such as AA (n-6), EPA (n-3), DPA (n-3) and DHA (n-3) and also compete for incorporation into cell membranes. As a result, excess consumption of LA leads to a decrease in the production and incorporation of n-3 fatty acids through simple substrate competition, and this effect is exacerbated when total dietary PUFA is high(Reference Tu, Cook-Johnson and James9,Reference Gibson, Neumann and Lien10) . The n-3 and n-6 long-chain PUFA derivatives also have opposing physiological actions, with those from the n-3 family predominately exhibiting anti-inflammatory properties (e.g. via the suppression of the pro-inflammatory transcription factor NF-κB and activation of the anti-inflammatory transcription factor PPARγ(Reference Calder11)) and those from the n-6 family exhibiting more pro-inflammatory and pro-adipogenic properties(Reference Massiera, Saint-Marc and Seydoux12). This has led to the hypothesis that the increasing ratio of n-6:n-3 fatty acids in modern Western diets may have negative effects on conditions characterised by low-grade inflammation, including obesity and the metabolic syndrome, and may potentially be contributing to an intergenerational cycle of obesity(Reference Ailhaud, Massiera and Weill8).
Data from observational studies in humans and animal models provide supporting evidence that suggests high intakes of n-6 PUFA during pregnancy could have negative effects on metabolic health of the progeny(Reference Massiera, Barbry and Guesnet13–Reference Much, Brunner and Vollhardt15). However, the results of these studies have not been consistent. The results of pre-clinical studies are also limited by the use of diets with much higher n-6:n-3 PUFA ratios and/or absolute PUFA contents than those encountered in typical human diets. Furthermore, offspring often continue to have access to the same diet as their mother so that any effects observed cannot be clearly attributed to dietary fatty acid exposure during the gestation and lactation periods(Reference Muhlhausler, Miljkovic and Fong16–Reference Muhlhausler, Gibson and Makrides18). The aim of this study, therefore, was to investigate the effects of feeding a maternal dietary LA:ALA ratio similar to that of the Western diet (9:1)(Reference Blasbalg, Hibbeln and Ramsden6), compared with a proposed ‘ideal’ ratio of 1:1·5(Reference Gibson, Muhlhausler and Makrides19,Reference Lands20) on growth and lipogenic gene expression in the offspring. Since total dietary PUFA intake also influences PUFA metabolism(Reference Tu, Cook-Johnson and James9,Reference Gibson, Neumann and Lien10) , we also investigated the effect of feeding each dietary fat ratio at either 18 % fat, w/w (in line with dietary recommendations(21)) or at a higher fat level of 36 % fat, w/w. A rat model was utilised to achieve the study objectives by allowing for tight control of dietary manipulation as well as invasive end points.
Materials and methods
Animals
All animal procedures were performed in accordance with the Animals (Scientific Procedures) Act 1986 under Home Office licence and were approved by the Animal Ethics Committee of the University of Nottingham, UK. Virgin female Wistar rats (n 30; 75–100 g) were housed on wood shavings in individually ventilated cages under a 12 h light–12 h dark cycle at a temperature of 20–22°C and had ad libitum access to food and water throughout the experiment. Female rats were allowed to acclimatise to the unit for 1–2 weeks, during which time they were fed standard laboratory chow (2018 Teklad Global 18 % Protein Rodent Diet, Harlan Laboratories). After acclimatisation, a tail vein blood sample was taken from each animal for the determination of fatty acid status. The rats were then randomly assigned to one of the four dietary groups (details provided below). Animals were maintained on their allocated diet for a 4-week ‘pre-feeding’ period, after which they were mated. Conception was confirmed by the presence of a semen plug, and this was recorded as day 0 of pregnancy. Animals were housed in individual cages and remained on their respective diets throughout pregnancy and lactation.
Litters were standardised to eight pups within 24 h of birth (four males and four females, where possible). At 1 and 2 weeks of age, one randomly selected male and one randomly selected female from each litter were culled via cervical dislocation and exsanguination for blood and tissue collection. At 3 weeks of age, the remaining offspring were weaned and dams were then euthanised by CO2 asphyxiation and cervical dislocation for collection of maternal blood and tissues. All dams were weighed and had feed intake measured daily throughout the experiment, and offspring body weight (BW) was measured weekly.
Diets
Diets were designed to provide either a high (9:1, high LA) or low (1:1·5, low LA) ratio of LA (cis/cis isomer):ALA, achieved by altering the amounts of flaxseed and sunflower oil included in the fat component of the feed. The levels of SFA and MUFA were comparable in all diets, achieved by adjusting the amounts of coconut (saturated fat source) and macadamia (monounsaturated fat source) oils in the diets. For each level of LA, diets were developed to contain either 18 % fat (w/w), in line with government recommendations(21), or 36 % fat (w/w) to highlight any additive effects (38·6 v. 63·5 % of dietary energy, respectively). This resulted in four experimental diets (n 6–9 per dietary group): high LA (18 % fat), high LA (36 % fat), low LA (18 % fat) and low LA (36 % fat). The list of ingredients and final fatty acid composition of the four experimental diets have been reported previously(Reference Draycott, Liu and Daniel5).
Blood sample and tissue collection
Blood samples were collected from dams prior to the start of the experiment and after the 4-week ‘feed-in’ period (tail vein sample) and at the end of lactation (via cardiac puncture following CO2 asphyxiation and cervical dislocation). Truncal blood samples were also collected from one randomly selected male and one randomly selected female at 1 and 2 weeks of age. In all cases, samples of whole blood (approximately 30 µl) from non-fasted animals were spotted onto a PUFAcoat™ dried blood spot collection paper(Reference Liu, Muhlhausler and Gibson22), allowed to dry at room temperature and stored at −20°C for subsequent fatty acid analysis. Maternal tissues were weighed, and samples of whole liver, retroperitoneal and gonadal adipose tissues were collected. Offspring body and organ weights were measured, and whole liver samples were collected from one randomly selected male and female pup at both time points. At 2 weeks of age, samples of gonadal and retroperitoneal fat were also collected from one male and one female pup per litter. All tissue samples were snap-frozen in liquid N2 and stored at −80°C until determination of gene expression by quantitative reverse transcriptase PCR.
Fatty acid methylation and analysis
Fatty acid composition in maternal and fetal blood was determined as previously described(Reference Liu, Muhlhausler and Gibson22). Briefly, whole dried blood spot samples were directly transesterified with 2 ml of 1 % H2SO4 in methanol, and the fatty acid methyl esters were extracted with heptane. Samples were separated and analysed by a Hewlett-Packard 6890 GC equipped with a capillary column (30 m × 0·25 mm) coated with 70 % cyanopropyl polysilphenylene-siloxane (BPX-70; 0·25 µm film thickness) which was fitted with a flame ionisation detector. Fatty acid methyl esters were identified in unknown samples based on the comparison of retention times with an external lipid standard (Standard 463, Nu-check prep Inc.) using Agilent Chemstation software (Agilent Technologies Australia Pty Ltd). Individual fatty acid content was calculated based on peak area and response factors normalised to total fatty acid content and expressed as a percentage of total fatty acids.
Isolation of RNA and cDNA synthesis and quantitative reverse transcription real-time PCR
RNA was isolated from crushed snap-frozen samples of approximately 25 mg of liver using the Roche High Pure Tissue kit (Roche Diagnostics Ltd). Adipose RNA was extracted, after homogenisation of approximately 100 mg of tissue with MagNA lyser green beads and instrument (Roche Diagnostics Ltd), using the RNeasy Mini Kit (QIAGEN Ltd). RNA concentration was determined using a Nanodrop 2000 (Thermo Scientific), and RNA quality was evaluated by agarose gel electrophoresis. cDNA was synthesised using a RevertAid™ reverse transcriptase kit (Thermo Fisher Scientific) with random hexamer primers.
Lipogenic pathway and adipokine target genes were chosen based on previous data from our laboratory that indicated that these genes were sensitive to changes in the maternal diet(Reference Erhuma, Salter and Sculley23) and included Pparg, sterol regulatory element-binding protein (variant 1c; Srebf1), fatty acid synthase (Fasn), lipoprotein lipase (Lpl) and leptin, with β-actin as the housekeeper. Primer efficiency ranged from 85 to 108 % and sequences have previously been published elsewhere(Reference Draycott, Liu and Daniel5). Adipocyte and hepatic gene expression was quantified using SYBR Green (Roche Diagnostics) in a Light-Cycler 480 (Roche Diagnostics). Samples were analysed against a standard curve of a serially diluted cDNA pool to produce quantitative data, and expression was normalised to the housekeeping gene using LightCycler® 480 software (version 1.5.1) as previously described(Reference Rhinn, Scherman and Escriou24). The expression of the housekeeper gene was not different between treatment groups.
Determination of liver DNA, protein and glycogen content
For determination of DNA and protein content of liver samples, approximately 100 mg of frozen crushed sample was added to 1 ml of 0·05m trisodium citrate buffer. Samples were homogenised and centrifuged at 2500 rpm for 10 min at 4°C. Supernatant was used for further analyses. DNA concentration (ug/ml) was measured using a Hoechst fluorimetric method, and protein content (ug/well), modified for a ninety-six-well plate format, was measured as described by Lowry et al. (Reference Lowry, Rosebrough and Farr25). Measurements were normalised to the exact amount of tissue used for measurements. Liver glycogen was measured using the Colorimetric Glycogen Assay Kit II (Abcam Ltd) according to the manufacturer’s instructions.
Statistical analysis
Data are presented as mean values with their standard errors. Data were analysed using the Statistical Package for Social Sciences (version 24, SPSS Inc.). The effect of maternal dietary fatty acid ratio and maternal dietary fat content on maternal dependent variables was assessed using a two-way ANOVA, with dietary LA:ALA ratio and dietary fat content as factors and dams were used as the unit of analysis. Where longitudinal data were analysed, as with maternal feed, protein and energy intakes, the impact of maternal dietary LA:ALA ratio and maternal dietary fat content was analysed using a two-way repeated-measures ANOVA. Offspring data were analysed using a two-way ANOVA, with maternal dietary LA:ALA ratio and fat content as factors; where there was no overall effect of sex, male and female offspring data were combined. Where data were not normally distributed, analyses were performed on log10 transformed data. A value of P < 0·05 was considered to be statistically significant.
Results
Maternal dietary intakes
There were no differences in feed intake of dams between treatment groups before or during pregnancy. During lactation, dams receiving the 36 % fat diets had a lower average daily feed intake than those receiving the 18 % fat diets, irrespective of dietary LA:ALA ratio (P < 0·001; Fig. 1(a)). Energy intake was similar between groups throughout the experiment (Fig. 1(b)). Protein intake prior to and during pregnancy was affected by both dietary LA:ALA ratio and fat content (P < 0·05; Fig. 1(c)); however, these effects were small and inconsistent. During lactation, protein intake was affected by dietary fat content only (P < 0·001; Fig. 1(c)), such that mothers receiving 36 % fat diets consumed 24 % less protein on average compared with those consuming 18 % fat diets, irrespective of dietary LA:ALA ratio. As expected, all dams consumed more food, energy and protein during lactation than before and during pregnancy regardless of dietary group (P < 0·001).
Fig. 1. Maternal average daily (a) feed intake, (b) energy intake and (c) protein intake during pre-feeding, pregnancy and lactation fed on either a high-linoleic acid (LA) (18 % fat) diet (
), high-LA (36 % fat) diet (
), low-LA (18 % fat) diet (
) and a low-LA (36 % fat) diet (
). Values are means with their standard errors (n 6–9 per group). The effects of dietary fatty acid ratio and dietary fat content were determined using a two-way repeated-measures ANOVA. *** Significant effect of dietary fat content (P < 0·001). † Significant interaction between dietary fat content and fatty acid ratio (P < 0·05).
Maternal fatty acid profile
There were no differences in the proportions of either SFA, MUFA, n-6 (Fig. 2(a)) or n-3 PUFA (Fig. 2(b)) in whole blood samples collected from the dams prior to the commencement of dietary intervention. After 4 weeks on their respective diets, the blood fatty acid profiles were significantly different between treatment groups and largely reflected the composition of the experimental diets. Thus, dams fed on high-LA diets had higher proportions of LA (1·2-fold) and AA (1·4-fold) compared with those consuming a low-LA diet (P < 0·001; Fig. 2(c)). Conversely, dams fed the low-LA diets had a 5·5-fold higher proportion of ALA and an 8·5-fold higher proportion of EPA compared with those consuming a high-LA diet (P < 0·001; Fig. 2(d)). These changes were independent of the total fat content of the diet. DPA and DHA levels after the 4-week pre-feeding period were influenced by both dietary LA:ALA ratio and total fat content. Thus, the relative proportions of DPA were higher in dams fed the low-LA compared with high-LA diets (P < 0·001) and marginally higher in dams consuming the 18 v. 36 % fat diets (P < 0·05). DHA proportions were also higher in the low-LA group (P < 0·001) but, unlike DPA, were modestly but significantly higher in dams consuming the 36 % fat v. 18 % fat diets (P < 0·05; Fig. 2(d)). Total blood MUFA proportions were higher (1·3-fold) in dams consuming the low-LA diet, irrespective of dietary fat content (P < 0·001; Fig. 2(c)).
Fig. 2. Maternal whole blood fatty acid profile at (a/b) baseline, (c/d) after 4 weeks on experimental diet and (d/e) at the end of lactation (3 weeks post-partum).Values are means with their standard errors (n 6–9 per group). The effects of dietary fatty acid ratio and dietary fat content were determined using a two-way ANOVA (* P < 0·05, *** P < 0·001). † Significant interaction effect (P < 0·05). , High linoleic acid (LA) (18 % fat); , high LA (36 % fat); 
, low LA (18 % fat); 
, low LA (36 % fat). AA, arachidonic acid; ALA, α-linolenic acid.
The blood fatty acid profile of the dams at the end of lactation, after a further 6 weeks on their respective experimental diets, was similar to those observed after the first 4 weeks of dietary intervention. A notable difference, however, was that at this time point, relative proportions of DHA, as a percentage of total lipids, were not different between dietary groups (Fig. 2(f)). LA (1·5-fold), AA (1·8-fold) and total n-6 (1·5-fold) were all higher in dams consuming a high-LA diet irrespective of dietary fat content (P < 0·001; Fig. 2(e)). Conversely, total n-3 levels were 3-fold higher in dams consuming a low LA diet, irrespective of dietary fat content (P < 0·001). The proportions of ALA were also higher in the groups consuming the low-LA diets and in rats consuming the 36 v. 18 % fat diets in the low-LA group only (P < 0·05; Fig. 2(f)). DPA proportions were higher in the groups consuming the low-LA diets; however, unlike ALA, DPA proportions were lower, rather than higher, in dams consuming the 36 % fat diets in the low-LA group only (P < 0·001; Fig. 2(f)). EPA proportions were higher in groups consuming a low-LA diet compared with those consuming a high-LA diet (P < 0·001; Fig. 2(f)). EPA proportions were also affected by total dietary fat content and were lower in dams consuming a 36 % fat diet compared with an 18 % fat diet (P < 0·001; Fig. 2(f)). Maternal blood total MUFA levels at the end of lactation were 1·4-fold higher in the dams consuming a low-LA diet irrespective of dietary fat content (P < 0·001; Fig. 2(e)).
Maternal weight, body composition and gene expression
There were no significant differences in dam BW between dietary groups prior to the commencement of the dietary intervention or at any time during the experiment (data not shown). Dams consuming the 36 % fat diets had heavier lungs relative to BW at the end of lactation compared with those consuming the 18 % fat diets, independent of the LA:ALA ratio (P < 0·05). There were no differences in the relative weight of the heart, liver, brain, kidney, gonadal or retroperitoneal fat pads between experimental groups (Table 1).
Table 1. Maternal organ weights and gene expression*
(Mean values with their standard errors; n 6–9 per dietary group)
LA, linoleic acid; BW, body weight, Fasn, fatty acid synthase; Lpl, lipoprotein lipase; Srebf1, sterol regulatory element-binding protein; Lep, leptin.
* Effect of dietary fatty acid ratio and dietary fat content was assessed using a two-way ANOVA.
Significant effect of dietary fat content: † P < 0·05, †† P < 0·01.
‡ Although not statistically significant, there was some evidence that maternal Srebpf1 expression was influenced by the LA (P = 0·08) and fat content (P = 0·06) of the diet.
Analysis of mRNA expression of lipogenic genes indicated that hepatic (3-fold) and gonadal fat (7-fold) expression of Fasn was higher in dams consuming an 18 % fat diet, compared with those on a 36 % fat diet, irrespective of dietary fatty acid ratio (P < 0·01). The mRNA expression of Lpl, Pparg and Srebf1 was not, however, affected by either dietary fat content or ratio in either hepatic or gonadal fat tissues (Table 1). Expression of leptin mRNA in gonadal adipose tissue was not significantly different between treatment groups.
Birth outcomes and offspring body weights
There were no differences between dietary groups in terms of litter size or sex ratio of pups (Table 2). Birth weight was lower in offspring of dams fed a 36 % fat v. 18 % fat diets, independent of the dietary LA:ALA ratio (Table 2). The lower BW in offspring of dams fed the 36 % fat diet persisted during the sucking period such that offspring of dams fed the 36 % fat diets remained lighter than offspring of dams fed on 18 % fat diets at both 1 and 2 weeks of age; again this was independent of dietary LA:ALA ratio (P < 0·001; Table 3).
Table 2. Birth outcomes*
(Mean values with their standard errors)
LA, linoleic acid.
* Effect of dietary fatty acid ratio and dietary fat content was assessed using a two-way ANOVA.
† Significant effect of maternal dietary fat content (P < 0·05).
Table 3. Offspring organ weights and hepatic gene expression*
(Mean values with their standard errors)
LA, linoleic acid; BW, body weight, Fasn, fatty acid synthase; Lpl, lipoprotein lipase; Srebf1, sterol regulatory element-binding protein.
a,b Mean values in a row with unlike superscript letters were significantly different (P < 0·05) (n 4–9 per dietary group). All comparisons are made within sex group.
* Two-way ANOVA was used to analyse results with maternal dietary fatty acid ratio and maternal dietary fat content as factors.
Offspring fatty acid profile
At 1 week of age, proportions of AA (2·1-fold) were lower in the offspring of the low-LA compared with high-LA dams (P < 0·001), and in offspring of dams consuming the 36 % fat v. 18 % fat diets (1·4-fold; P < 0·001; Fig. 3(a)). Blood ALA proportions were 5·9-fold higher in offspring of dams in the low-LA groups compared with high-LA groups (P < 0·001; Fig. 3(b)). Offspring EPA and DPA proportions were also higher in the low-LA group compared with the high-LA group. Blood EPA was also influenced by total dietary fat content, but only in offspring of dams fed the low-LA diet, in which EPA levels were lower in offspring of dams fed the 36 % fat diets compared with the 18 % fat diets (EPA, P < 0·001; DPA, P < 0·01; Fig. 3(b)). DHA proportions were not different between groups at 1 week of age (Fig. 3(b)). MUFA proportions were higher (1·2-fold) in offspring of dams in the low-LA groups (P < 0·001), consistent with the pattern in maternal blood. However, unlike maternal MUFA, offspring MUFA levels were also affected by maternal dietary fat content and were 1·2-fold higher in offspring of dams fed the 36 % fat v. 18 % fat diets (P < 0·001; Fig. 3(a)). At 1 week of age, offspring of dams in the 36 % fat diet groups also had lower blood proportions of SFA, irrespective of LA:ALA ratio of the maternal diet (P < 0·01; Fig. 3(a)).
Fig. 3. Offspring whole blood fatty acid profile at (a/b) 1 week of age and (c/d) at 2 weeks of age. Values are means with their standard errors (n 11–17 per group). The effects of maternal dietary fatty acid ratio, maternal dietary fat content and sex were determined using a three-way ANOVA. No effect of sex was found for any of the fatty acids measured, and so male and female samples were combined for further analysis. Significant difference (* P < 0·05, *** P < 0·001). † Significant interaction effect (P < 0·05). , High linoleic acid (LA) (18 % fat); , high LA (36 % fat); , low LA (18 % fat); , low LA (36 % fat). AA, arachidonic acid; ALA, α-linolenic acid.
The fatty acid profiles of the offspring at 2 weeks of age were similar to those observed at 1 week. Thus, blood AA (1·9-fold) and total n-6 (1·6-fold) proportions were lower (Fig. 3(c)) and ALA (6·3-fold), EPA (4·7-fold), DPA (2·4-fold), and total n-3 PUFA (3-fold) proportions (Fig. 3(d)) were higher in offspring of dams in the low-LA group compared with high-LA groups, irrespective of maternal dietary fat content (P < 0·001). Proportions of LA were higher in offspring of dams fed the 36 % fat diets compared with those fed 18 % fat diets in the high-LA group only (P < 0·05; Fig. 3(c)), while EPA and DPA proportions were lower in the 36 % compared with the 18 % fat diet groups, independent of the dietary LA:ALA ratio (P < 0·001; Fig. 3(d)). Unlike findings at 1 week of age, the DHA levels in 2-week-old offspring of dams consuming a 36 % fat diet were lower (P < 0·05) when compared with 18 % fat groups, irrespective of maternal dietary fatty acid ratio. As at 1 week, SFA proportions were lower (1·2-fold) in offspring of dams fed a 36 % fat diet, independent of the LA:ALA ratio (P < 0·001). MUFA proportions were 1·2-fold higher in offspring of dams fed the low-LA diets and 1·2-fold higher in offspring of dams who consumed a 36 % fat v. 18 % diet (P < 0·001; Fig. 3(c)).
Offspring organ weight and liver composition
At 1 week of age, heart weight relative to BW was higher in female offspring of dams receiving a 36 % fat diet compared with the 18 % fat diet, independent of the dietary LA:ALA ratio (P < 0·05). There were no differences in the relative weight of lung or kidney at 1 week of age and no differences in the relative weight of the heart, lung, liver, gonadal or retroperitoneal fat pads in the offspring at 2 weeks of age between treatment groups in either male or female offspring (Table 3).
Liver weight at 1 week appeared to be influenced by the LA:ALA ratio of the diet to a greater extent than total fat level, at least in males. Thus, male offspring of dams consuming the high-LA diets had increased liver weights compared with offspring of dams receiving a low-LA diet (P < 0·01), irrespective of total dietary fat content. The glycogen content of the livers was also higher in male offspring of dams consuming the high-LA diets at 1 week of age (P < 0·05). No effect of maternal diet on offspring liver protein or DNA concentration was observed (Table 4). These differences were not present in females at 1 week of age, and no differences in glycogen content were observed at 2 weeks of age in male offspring. DNA concentration in females at 2 weeks of age was marginally increased (1·1-fold) in offspring exposed to a 36 % fat diet, irrespective of maternal dietary fatty acid ratio (P < 0·05).
Table 4. Offspring liver composition*
(Mean values with their standard errors)
LA, linoleic acid.
a,b Mean values in a row with unlike superscript letters were significantly different (P < 0·05) (n 4–9 per dietary group). All comparisons are made within sex group.
* Two-way ANOVA was used to analyse results with maternal dietary fatty acid ratio and maternal dietary fat content as factors.
Hepatic gene expression
At 1 week of age, hepatic Fasn expression was influenced by maternal dietary intervention in a sex-specific manner. Thus, in males, Fasn expression was higher in offspring of dams consuming a 36 % fat diet irrespective of maternal LA:ALA ratio (P < 0·05). In female offspring, however, Fasn expression was higher in offspring of dams consuming a low-LA diet, independent of dietary fat content (P < 0·05). Hepatic Lpl mRNA expression in male offspring at 1 week of age was also influenced by maternal dietary fat content, with higher expression in offspring of dams consuming a 36 % fat diet v. 18 % fat diet (P < 0·05). In female offspring, hepatic Srebf expression, similar to that of Fasn, was higher in offspring of dams consuming a low-LA diet at 1 and 2 weeks of age (P < 0·01). Female hepatic expression of Pparg was lower in offspring of dams consuming a low LA diet at 2 weeks of age (P < 0·05). There were no differences in the expression of Fasn or Lpl in female offspring or expression of any hepatic genes in male offspring at this time point (Table 3).
Discussion
This study has demonstrated that altering the fat content and/or LA:ALA ratio of the maternal diet during pregnancy and lactation resulted in significant alteration in the circulating fatty acid profile of dams in the absence of any significant effects on maternal BW or body composition. Exposure to a 36 % fat diet during gestation and lactation was, however, associated with lower offspring BW from birth, which persisted to 2 weeks of age. This suggests that increased dietary fat intake during pregnancy and lactation can compromise growth of the progeny, irrespective of the type of fat consumed. In addition, alterations in the fat content and/or composition of the maternal diet had transient effects on offspring body composition and hepatic gene expression, effects which were also sex-specific.
Maternal fatty acid profiles after 4 weeks on the experimental diets largely reflected dietary composition, confirming that the dietary intervention had the desired effect on maternal circulating fatty acid composition. These changes persisted after a further 6 weeks of exposure to the diets and, as expected, the dietary LA:ALA ratio had a greater impact on the maternal blood n-6 and n-3 status than total dietary fat content. Consistent with previous studies(Reference Draycott, Liu and Daniel5,Reference Tu, Cook-Johnson and James9,Reference Blank, Neumann and Makrides26,Reference Brenna, Salem and Sinclair27) , decreasing the dietary LA:ALA ratio resulted in substantial increases in relative maternal ALA and EPA levels but only a very modest increase in DHA proportions after a 4-week exposure, and no difference compared with the higher LA:ALA ratio after 10 weeks. Interestingly, and independent of dietary LA:ALA ratio, dams appeared to be more efficient at converting DPA to DHA when total dietary fat load was higher. One possibility could be that this is simply a result of the higher amount of substrate (i.e. ALA) available for conversion to the longer chain derivatives such as DPA and DHA in diets containing higher total fat levels. This effect did not, however, persist after a further 6 weeks of dietary exposure, at which point EPA and DPA were lower in dams consuming a low-LA 36 % fat diet compared with a low-LA 18 % fat diet. This may be a result of saturation of the PUFA metabolic pathway when total fat, and therefore PUFA, levels were higher(Reference Gibson, Neumann and Lien10,Reference Tu, Muhlhausler and Yelland28) . This apparent decrease in capacity to convert ALA through to EPA and DHA during consumption of a 36 % fat diet coincides with the decreased protein intake observed in these groups. It is possible that the lower consumption of protein in rats fed on the 36 % fat diets may have contributed to reduced conversion of ALA, since previous studies have shown reduced desaturase, particularly Δ6-desaturase, expression in the mammary gland(Reference Bautista, Rodriguez-Gonzalez and Torres29) and liver(Reference Mercuri, de Tomas and Itarte30) of rats exposed to a low-protein diet. Maternal whole blood MUFA proportions appeared to be influenced by dietary LA:ALA ratio; however, this is most likely a result of the slightly higher MUFA content of the low-LA diets.
Offspring fatty acid profiles at 1 and 2 weeks of age largely reflected maternal profiles with maternal dietary LA:ALA ratio exhibiting the strongest effect on offspring circulating fatty acid proportions. However, the total fat content of the maternal diet appeared to have a greater influence on the blood fatty acid composition of the offspring as opposed to that observed in the dams. Of particular interest was the finding that the proportion of both EPA and DPA in offspring at 1 week of age was higher in the low-LA (18 % fat) v. the low-LA (36 % fat) group and that this effect persisted at 2 weeks of age despite ALA levels being increased in the low-LA (36 % fat) group at this time point. DHA was not different between groups at 1 week of age but was lower in offspring exposed to a 36 % fat diet at 2 weeks of age. As with the maternal fatty acid profiles, this again may be a result of saturation of the PUFA metabolic pathway at higher total PUFA intakes and is in line with findings from numerous studies, both human and animal, that indicate that simply increasing the quantity of substrate, that is, ALA, is not an effective strategy for increasing concentrations of its long-chain derivatives, in particular DHA(Reference Blank, Neumann and Makrides26,Reference Brenna, Salem and Sinclair27,Reference Mantzioris, James and Gibson31,Reference Chan, McDonald and Gerrard32) .
The total dietary fat content of the maternal diet also had a influence on the proportion of SFA in the offspring, such that offspring of dams consuming 36 % fat diets exhibited lower SFA proportions than offspring of dams consuming the 18 % fat diets. Unlike the fetus, where fatty acid composition is largely related to maternal dietary intake, during suckling, offspring fatty acid composition is largely determined by the composition of the milk, which may not fully reflect maternal fatty acid intakes. In a study by Mohammad et al. (Reference Mohammad, Sunehag and Haymond33), for example, women consuming diets with a higher total fat content (55 v. 25 %en) exhibited reduced SFA concentrations (C6 : 0–C14 : 0) in breast milk but not in maternal plasma. While milk composition was not assessed in the present study, this raises the possibility that SFA content of the milk may have been lower in those dams consuming the 36 % fat diets, which could in turn explain the lower SFA status of the offspring. Alternatively, it may be that increasing the fat content of the diets resulted in an increased conversion of SFA to MUFA, since high fat feeding has been associated with increased expression of the enzyme responsible for conversion of SFA to MUFA, stearoyl-CoA desaturase 1(Reference Hu, Qing and Chen34) and could therefore be the reason for the observed effect of fat content on offspring MUFA levels in this study. It is important to note, however, that circulating fatty acid profiles are a product of both dietary fatty acid intake and tissue fatty acid production and release. Whilst the collection of blood samples from animals in the fed state suggests that the dietary fraction of fatty acids would provide a greater contribution to the fatty acid profile of both dams and offspring, the influence of hepatic synthesis of fatty acids should not be overlooked as a contributor to the observed differences.
Despite significant shifts in maternal fatty acid profiles and increased fat content of the 36 % fat diets, we saw no differences in maternal BW or fat deposition. This is consistent with our previous study(Reference Draycott, Liu and Daniel5) and is likely a result of the reduced feed intakes of the dams to compensate for the increased energy density of the higher fat diets, a phenomenon consistently seen with dietary intervention trials using rodents(Reference Keesey and Hirvonen35). Despite the lack of an effect on maternal weight gain and fat deposition, BW was reduced in offspring of dams receiving a 36 % fat diet, irrespective of maternal dietary LA:ALA ratio. This phenotype was consistent across sexes and persisted from birth to 2 weeks of age. Variable results have been reported in this regard with some studies reporting no effects(Reference Khan, Taylor and Dekou36–Reference Dyrskog, Gregersen and Hermansen38) or increased weight(Reference Samuelsson, Matthews and Argenton39). This finding was, however, consistent with many other studies that reported decreased fetal(Reference Taylor, Khan and Lakasing40,Reference Mark, Sisala and Connor41) , birth(Reference Howie, Sloboda and Kamal42) and weaning weight(Reference Cerf, Muller and Du Toit43) in offspring of dams exposed to a 36 % fat diet during gestation and lactation periods. The differential effects of different 36 % fat diets on offspring growth are likely due to differences in composition of the diet as well as periods of exposure between studies(Reference Ainge, Thompson and Ozanne3). In those studies that have reported lower offspring weights in offspring fed a high-fat diet, lower protein intakes in dams consuming a high-fat diet have been cited as a likely contributing factor. Further to this, protein-restricted diets have been associated with impaired mammary gland development(Reference Bautista, Rodriguez-Gonzalez and Torres29,Reference Moretto, Ballen and Goncalves44) leading to impaired milk synthesis(Reference Moretto, Ballen and Goncalves44), and this may also have contributed to reduced offspring growth observed during the suckling period. It is important to note however that the reduction in protein intake in dams consuming a 36 % fat diet in the present study was more modest (10–25 %) than those typically used in low-protein diet studies (approximately 50 % reduction)(Reference Langley and Jackson45–Reference Malandro, Beveridge and Kilberg48).
The lower Fasn expression in the liver and adipose tissue of dams exposed to a 36 % fat diet is consistent with the established role of this enzyme in suppressing lipogenesis in times of energy excess(Reference Geelen, Blazquez and Geelen49). Surprisingly, this change did not appear to be mediated through changes in maternal Srebf1 mRNA expression, a known regulator of Fasn expression(Reference Griffin and Sul50). It is important to note that since only mRNA expression was measured, we cannot comment on any differences in protein expression or activity of this transcription factor, although mRNA and protein levels have been shown to be closely correlated(Reference Erhuma, Salter and Sculley23). Following this up at the protein level is a major priority for future study. In the offspring, however, hepatic Fasn expression was not down-regulated by exposure to a maternal 36 % fat diet but was actually higher in male offspring of dams consuming the 36 % fat compared with the 18 % fat diets at 1 week of age and was accompanied by an increase in Lpl expression. In female offspring, however, hepatic Fasn and Sbrepf1 expression at 1 week were influenced by maternal dietary fatty acid ratio, rather than total fat content, with both genes up-regulated in offspring of dams fed the low-LA diets. In both cases, the up-regulation of Sbref1, Fasn and Lpl genes would be expected to be associated with an up-regulation of both lipogenesis and fatty acid uptake. It is worth mentioning that differences in hepatic expression of lipogenic genes in male offspring were consistently associated with maternal dietary fat content, whereas differences in female hepatic expression were consistently associated with maternal dietary fatty acid ratio. This suggests that female offspring are more sensitive to changes in the types of maternal dietary fat, whereas male offspring are more sensitive to gross maternal fat consumption. Sex-specific effects associated with the programming of disease hypothesis have been frequently reported(Reference Gabory, Roseboom and Moore51). The mechanism by which sex influences these effects, however, remains to be elucidated within a larger perspective, as well as within the context of this study.
We found no evidence that these alterations in hepatic gene expression translated to increases in liver weight; however, whether there was any effect on hepatic fat content remains to be determined. In both male and female offspring, relative liver weight was increased in offspring of dams fed the high-LA diet. With the aim of further elucidating the source of this increased weight, we measured liver DNA, protein and glycogen composition. Similar to liver weight, glycogen levels were increased in offspring of dams fed the high-LA diets. This increase in glycogen, however, was not sufficient enough to completely account for the differences observed in liver weight but may be a contributing factor. Consideration of DNA and protein content of the tissue did not indicate significant changes to cell size or number. More detailed analysis is required to further elucidate the mechanism by which high maternal dietary n-6 may impact upon offspring liver physiology.
The majority of the hepatic mRNA expression differences, as well as gross differences in liver weight and composition, appeared to be transient and were no longer present at 2 weeks of age. A notable exception was the lower expression of Srebf1 mRNA and higher expression of Pparg in females of dams exposed to a high-LA diet compared with the low-LA diet, with a similar trend observed in males. Although found in relatively low concentrations in the liver, activation of Pparg has been shown to increase hepatic lipid storage and is elevated in models of hepatic steatosis(Reference Inoue, Ohtake and Motomura52). As such, decreased Pparg expression can alleviate some of the symptoms of hepatic steatosis leading to a reduced liver weight in conjunction with a reduction in hepatic TAG content(Reference Gavrilova, Haluzik and Matsusue53). Thus, our finding that female offspring of dams exposed to a high-LA diet tended towards to have an increased liver weight at 1 week of age followed by increased hepatic Pparg expression at 2 weeks of age may suggest that the increase in Pparg expression is a potential response to the increased liver growth observed a week earlier. Alternatively, parallels may be drawn to the effect of low-protein diets where fluctuations between an increased and decreased lipogenic capacity, chiefly mediated by altered Srebf1 expression, occur in early life only to settle into a pattern of up-regulated lipogenesis at a later life stage(Reference Erhuma, Salter and Sculley23). Further studies would be needed to directly evaluate this hypothesis.
In conclusion, we have demonstrated that exposure to a 36 % fat diet during gestation and lactation is associated with persistent growth restriction in both male and female offspring irrespective of maternal dietary fatty acid composition. Growth restriction has been associated with a plethora of metabolic disturbances later in life(Reference Barker54–Reference Holemans, Aerts and Van Assche56), and transient alterations in gene expression have been suggested as a mechanism for programming changes in metabolic processes within tissues as well as the morphology of the tissues themselves(Reference Langley-Evans1). In this study, offspring are still exposed to the experimental diets via the dams milk and further studies in offspring at older ages are required to assess whether the changes in growth, hepatic gene expression and liver weights in the present study are associated with phenotypic changes that persist once offspring are no longer exposed directly to the altered diet composition. In addition, analysis of lipogenic pathway and adipokines targets at the protein level, as well as whole transcriptome analysis, may yield useful information about their regulation and the extent to which these experimental diets programme other metabolic and regulatory pathways in the liver. Finally, the longevity of these perturbations into later life, especially when presented with secondary metabolic challenges such as ageing, prolonged high fat feeding or in the case of female offspring, pregnancy, remains to be elucidated.
Acknowledgements
The authors gratefully acknowledge the staff at the Bio-Support Unit (University of Nottingham) for help and advice with animal procedures and Zoe Daniel for assistance with molecular analyses.
B. S. M. is supported by a Career Development Fellowship from the National Health and Medical Research Council of Australia (APP1083009).
S. C. L.-E., B. S. M. and M. J. E. participated in study design. S. A. V. D. carried out the study (assisted by G. G.), data analysis and preparation of the manuscript which was revised and approved by S. C. L.-E., B. S. M., M. J. E. and G. G.
The authors declare that there are no conflicts of interest.
