Introduction

Research on the causes of the rapid rise in obesity over the past few decades has revealed unique environmental influences, one of which is a diet high in fat [1]. Achieving energy balance and oxidizing fat is essential for weight maintenance and the prevention of fat mass gain. The amount and type of dietary fatty acids (FA) in the diet give rise to physiological differences in terms of their effects on whole-body metabolism, insulin sensitivity, inflammation, and lipid profiles with important health consequences. In general, saturated FAs (SFAs) are considered to be detrimental to health [2–6], whereas polyunsaturated FAs (PUFAs) are considered to be beneficial to health [7–12]. Different types of PUFAs exist, such as the omega-6 (n-6) and omega-3 (n-3) PUFAs; however, they are a fairly heterogeneous group that share a number of common metabolic and physiologic effects. Notably, though, the quality of dietary PUFA can be important, in particular the long-chain polyunsaturated fatty acids (LC-PUFAs) of the n-3 series EPA (eicosapentaenoic acid; C20:5n-3) and DHA (docosahexaenoic acid; C22:6n-3), which are abundant in cold-water fish. It is also important to discriminate between the n-3 LC-PUFAs EPA and DHA and their precursor ALA (α-linolenic acid; C18:3n-3), as n-3 LC-PUFAs usually exert much stronger effects.

The adverse outcomes of SFA consumption on health have long been established [5, 13–18]. Diets rich in SFAs increase total and low-density lipoprotein (LDL) cholesterol, blood pressure, insulin resistance, and chronic disease, which can be attributed to the chronic inflammation caused by consuming SFAs [5]. Moreover, diets high in SFAs appear to have a negative impact on fat oxidation (Fox) which has led researchers to label SFAs as obesogenic [13]. Studies have reported lower Fox, diet-induced thermogenesis (DIT) [19], and energy expenditure (EE) with SFAs compared to diets rich in monounsaturated FAs (MUFAs) or PUFAs [14, 16, 20–22]. Even acute ingestion of SFA-rich HF meals results in smaller increases in DIT [19], postprandial Fox [21, 22], and total EE, compared to MUFA- and PUFA-rich HF meals [17, 23]. Since an inverse relationship between dietary Fox and BMI as well as Fox and body fat percentage exists [18], SFAs may play a greater role in human obesity than other dietary FAs.

Numerous studies demonstrate the benefits of increasing PUFA intake or replacing SFAs with PUFAs in the diet [15, 17, 24, 25]. Consuming high amounts of PUFAs have been associated with decreased risk of chronic disease by decreasing total and LDL cholesterol, elevating HDL cholesterol, and decreasing markers of inflammation and coagulation potential [15, 26, 27]. Moreover, in human studies, PUFA consumption has been shown to elicit moderate body fat-lowering effects [25, 28, 29]. High PUFA consumption may potentially lead to better weight maintenance and positively affect EE because it has been shown to result in higher RMRs, DIT, and rates of Fox when compared to consuming meals high in SFAs [30]. We also recently showed that acute ingestion of a PUFA-rich HF meal induced a greater DIT in normal-weight women compared with SFA- or MUFA-rich HF meals [31]. Taken together, PUFA consumption seems to be beneficial for a number of metabolic and chronic disease risk factors. However, to our knowledge, no studies have specifically examined the effects of a longer-term PUFA consumption on acute metabolic responses to HF meals rich in SFAs. Most studies on substrate oxidation examine the type of dietary fatty acid by directly comparing different test meals or diets. No one has used the same test meal in a pre-/post-design to see if altering daily intake can change the way the body responds to an identical meal.

The purpose of this study was to determine whether a moderate-fat diet, rich in PUFAs, could compensate for the detrimental metabolic effects of acute SFA-rich HF meal consumption compared to a control diet. We studied metabolic responses to two SFA-rich HF meals (breakfast and lunch) before and after either a 7-day PUFA-rich diet or control diet in order to assess metabolic responses to the SFA-rich meals in both the fasted and fed states. We hypothesized that the PUFA-rich diet would lead to higher fasting RMR and Fox and that postprandial Fox and DIT would be increased following the 7-day PUFA-rich diet versus the control diet.

Participants and methods

This 7-day feeding study was a randomized, single-blinded, placebo-controlled, parallel trial and was designed to test the metabolic responses to two SFA-rich HF meals before and after a 7-day diet. This study was registered at clinicaltrials.gov (ID: NCT02246933). The study protocol consisted of a screening visit, a 3-day lead-in diet, a pre-diet visit, a 7-day feeding protocol (either PUFA-rich diet or control diet), and a post-diet visit (Table 1). For all testing procedures, participants reported to the Human Nutrition Lab (HNL) following an overnight fast (10–15 h) and without performing vigorous exercise for at least 12 h. The institutional review board approved the study, and procedures followed were in accordance with the Helsinki Declaration of 1975 as revised in 1983. Written consent was obtained prior to beginning the study.

Table 1 Timeline of study visits and measurements taken at each visit
Full size table

Thirty-two participants were randomly assigned to either the PUFA-rich diet (n = 16) or control diet (n = 16). However, six participants either dropped out of the study or were excluded from further participation due to poor compliance with the diet during the study (all in the control group). Thus, twenty-six (n = 13 men and n = 13 women) normal-weight (BMI = 18–24.9 kg/m2), sedentary (perform <3 h/week of structured exercise) adults between the ages of 18–35 years completed the study in its entirety. Exclusion criteria were history of chronic, metabolic, or endocrine disease, history of surgery that could affect digestion or hormone signaling, gastrointestinal disorders, and dyslipidemia (based on fasting blood sample drawn at screening visit). Individuals who were on a medically prescribed diet, took medications/supplements, experienced body weight changes (3 months prior to testing), and used tobacco were also excluded. Lastly, women who were pregnant, planning on becoming pregnant, or lactating were excluded.

Protocol

Screening visit and 3-day lead-in diet

All participants completed a screening visit. At the screening visit, height, weight, 5 mL fasting blood draw (for blood lipids), and RMR were obtained. The 5 mL of blood was collected into a Vacutainer and immediately centrifuged at 3000 rpm for 15 min at 4 °C. The plasma sample was then transported to Covenant Health Laboratory (Lubbock, TX, USA) where a lipid panel analysis was performed (UniCel DxC 800 Synchron Clinical Systems, Beckman Coulter, Inc., Brea, CA, USA) on the fasting sample. Resting metabolic rate (kcal/d) was measured for 30 min with a metabolic cart (TrueOne 2400, Parvo Medics, Sandy, UT, USA). Participants were instructed to remain motionless without sleeping while respiratory gases were collected. Thirty minutes of respiratory gases were collected, but only the final 20 min of data were used to calculate RMR using the Weir equation [32]. RMR ranged from 1193 to 2171 kcals/day for men and ranged from 1049 to 1652 kcals/day for women. Estimated total daily energy needs were calculated as participant’s RMR*1.65 (1.65 was used to represent an average US physical activity factor) [33] and used to calculate total daily energy needs for the 3-day lead-in diet, 7-day PUFA-rich diet, 7-day control diet, and the SFA-rich HF meals. The diet (kcals/day) was meant to keep participants in energy balance throughout the duration of the study. Once participants qualified for the study (based on a normal lipid profile: fasting total cholesterol <200 g/dL, HDL cholesterol >40 mg/dL, LDL cholesterol <100 mg/dL, and/or triglycerides <150 mg/dL), they were randomized into one of the two treatment conditions: PUFA-rich diet or control diet. For allocation of participants, a computer-generated list of random numbers was used to allocate participants to either the PUFA-rich diet or control diet. The participant was blinded as to which diet they were receiving.

After completion of the screening visit, participants were scheduled for the pre-diet visit. For 3 days prior to the pre-diet visit, participants were provided with a lead-in diet that is representative of the standard American diet (Table 2). Average total daily energy needs for participants in the PUFA-rich diet were 2573 ± 5.86 kcals/day (men: 2914 ± 7.80 kcals/day; women: 2232 ± 6.88 kcals/day). Average total energy needs for participants in the control diet were 2585 ± 6.62 kcals/day (men: 2967 ± 6.03; women: 2203 ± 6.12 kcals/day). The lead-in diet provided approximately 29, 31, and 40 % of energy at breakfast, lunch, dinner + snacks, respectively. However, participants could consume meals and snacks in any order they chose (except for breakfast which was consumed in the HNL each morning), as long as they ate all the food that was provided each day. Finally, no additional foods or caloric beverages were permitted. Participants kept a food and physical activity log to help ensure compliance.

Table 2 Nutrient breakdown for each diet
Full size table

Pre-diet visit

Following a 3-day lead-in diet, participants completed the pre-diet visit (~9 h). For women, the pre-diet visit occurred during the follicular phase of the participant’s menstrual cycle (days three through nine). Height, body weight, body composition, and waist and hip circumference were measured. Body composition was measured using air displacement plethysmography (BodPod, Cosmed USA, Inc., Concord, CA, USA). After anthropometric data were recorded, RMR was measured as described in the screening visit. Respiratory gases were used to calculate RMR using the Weir equation [32]. Calibration gas (Airgas Specialty Gases, Inc., Lenexa, KS, USA) was used for O2 and CO2 analyzer calibration. This was conducted before each of the study visits and on an hourly basis during testing. Additionally, a 3L syringe flowmeter was used for calibration of flow/volume measurement before each of the study visits.

After baseline measurements were taken, a 5 mL fasting blood sample was obtained (t = 0) to asses blood lipids. The 5 mL of blood was collected into a Vacutainer and immediately centrifuged at 3000 rpm for 15 min at 4 °C. The plasma sample was transported to Covenant Health Laboratory (Lubbock, TX, USA) for lipid panel analyses (UniCel DxC 800 Synchron Clinical Systems, Beckman Coulter, Inc., Brea, CA, USA). Next, the participant ingested the first of two identical SFA-rich HF liquid meals (Table 3). The first meal was ingested at 0800 h, and the second identical meal was ingested at 1200 h. Each shake provided 35 % of total daily energy needs. Average kcals/shake provided to participants in the PUFA-rich diet was 901 ± 3.47 kcals (men: 1020 ± 4.62 kcals; women: 781 ± 4.07 kcals). Average kcals/shake provided to participants in the control diet was 905 ± 3.92 kcals (men: 1038 ± 3.57 kcals; women: 771 ± 3.62 kcals). Participants were instructed to finish the entire meal within 5 min. This included a 118 mL water rinse to guarantee all of the liquid meal was ingested. The “base” was original milk chocolate, ready-to-drink shake (Ensure®, Abbott Nutrition, Abbott Laboratories, Inc., Columbus, OH, USA), soy lecithin (Lecithin Granules Non-GMO, NOW® Foods, Bloomingdale, IL, USA), and powdered chocolate drink mix with no sugar added (Nesquik®, Nestle® Inc., Glendale, CA, USA). The SFA-rich meal had “base” plus butter (majority of fat is palmitic acid-C16:0), red palm oil (majority of oil is palmitic acid-C16:0), and coconut oil (majority of oil is lauric acid-C12:0), with a total volume of 282 mL.

Table 3 Nutrient breakdown for each high saturated fat test meal
Full size table

Following meal ingestion, respiratory gases were measured to determine EE, fuel utilization (respiratory exchange ratio (RER) and fat and carbohydrate oxidation), and DIT every 30 min for the next 4 h. Specifically, data were collected for a 20-min period followed by a 10-min break. The first 5 min of data collection for each of the 20-min periods was discarded to ensure the participant was in a steady state and re-acclimated to the hood. During the 10-min break, participants remained sedentary. Participants were also provided with 118 mL of water every hour to provide additional fluid for the participant over the 8-h testing period. At 1200 h the second SFA-rich HF meal was ingested (5 min for meal consumption), and respiratory gases and blood samples were collected every 30 min for another 4-h period (as described for the first meal).

7-day diet

Following the pre-diet visit, participants began the 7-day diet (either a PUFA-rich diet or control diet, Table 2). The diet (kcals/day) was designed to keep participants in energy balance throughout the duration of the study. Average total daily energy needs for participants in the PUFA-rich diet were 2573 ± 5.86 kcals/day (men: 2914 ± 7.80 kcals/day; women: 2232 ± 6.88 kcals/day). Average total energy needs for participants in the control diet were 2585 ± 6.62 kcals/day (men: 2967 ± 6.03; women: 2203 ± 6.12 kcals/day). The PUFA-rich diet and control diet provided approximately 29, 31, and 40 % of energy at breakfast, lunch, dinner + snacks, respectively. The percentage of energy provided by each macronutrient was the same for both the PUFA-rich diet and the control diet (50 % carbohydrate, 15 % protein, 35 % fat). The control diet had the same macronutrient and FA breakdown as the 3-day lead-in diet (7 % of energy provided by PUFA (all from omega-6 PUFAs), 15 % of energy provided by MUFA, and 13 % of energy provided by SFA). For the PUFA-rich diet, 21 % of energy provided by PUFA, 9 % of energy provided by MUFA, and 5 % of energy provided by SFA. The breakdown of PUFA was approximately 12.5 % of total energy as omega-6 PUFA (nearly all linoleic acid) and 8.5 % of total energy as omega-3 PUFA (nearly all α-linolenic acid) along with very small amounts eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). We achieved this level of PUFAs through whole foods such as walnuts, wild-caught Alaskan keta salmon, tuna, flax seed oil, grapeseed oil, and canola oil. For the PUFA-rich diet, participants were supplemented with fish oil supplements in conjunction with the diet described above to receive an additional ~3 g/day of combined EPA (2157 mg/day) and DHA (843 mg/day) (GNC Ultra Triple Strength Omega 1000 EPA & DHA, Pittsburgh, Pennsylvania, USA). We wanted to supplement the PUFA-rich diet with fish oil pills to provide additional EPA and DHA. This was done to help elicit a strong metabolic response in just 7 days and show proof of principle with high dietary PUFA intake. Participants were instructed to ingest four pills daily: (1) with breakfast (2) with lunch, (3) with mid-afternoon snack, and (4) with dinner. Since we did use a whole foods approach, other components in the diet varied between the two diets. The PUFA-rich diet provided 35.9 ± 2.0 g/day dietary fiber, 143.4 ± 8.2 mg/day dietary cholesterol, 10.4 ± 0.6 g/day SFA, and 0.01 ± 0.0 g/day trans FA, whereas the control diet provided an average of 19.5 ± 0.8 g/day dietary fiber, 191.4 ± 8.8 mg/day dietary cholesterol, 36.2 ± 1.4 g/day SFA, and 3.3 ± 0.2 g/day trans FA.

Participants received their first meal on the evening after completing the pre-diet visit. On days one through seven of the diet, all participants arrived at the HNL in a fasted state and consumed breakfast between 0600 h and 1000 h. Following the meal, participants were provided with their remaining food and beverages for the day. All participants were instructed to eat all of the food and abstain from any other food or any beverages besides water. All food items were weighed and prepared by study personnel. Each day when the participants arrived at the HNL, they were asked whether they consumed everything that was provided from the day before and whether they ate anything that was not provided by research personnel.

Post-diet visit

After the 7-day diet, participants reported to the HNL for the post-diet visit. Participants completed another 9-h visit where the exact same study procedures and measurements for the pre-diet visit had occurred.

Calculations

Respiratory gases were used to calculate EE using the Weir equation [32] and macronutrient oxidation using equations developed by Frayn [34]: fat (g/min) = (1.67*VO2 (L/min))–(1.67*VCO2 (L/min)) and carbohydrate (g/min) 0 = (4.56*VCO2 (L/min))–(3.21*VO2 (L/min)). For these calculations, the first 5 min of each 20-min segment was discarded to allow participants to enter into a steady state. Therefore, for every 30-min postprandial period, we used 15 min of data collection. Additionally, the metabolic cart was calibrated against methanol burns throughout the duration of the study. The percentage recoveries from each burn were used to develop correction factors for the corresponding metabolic cart data from each study visit. The average correction factors were 98.0 ± 0.01 % for O2 and 96.1 ± 0.01 % for CO2. The precisions of the corrections were 98 and 96 % for O2 and CO2, respectively. The respiratory gases were used to calculate postprandial macronutrient oxidation using equations by Frayn [34] and EE using the Weir equation [32]. DIT was calculated from EE (postprandial EE subtracted by baseline EE).

Statistical analyses

The SAS version 9.2 statistical package (SAS Institute Inc, Cary, NC, USA) was used for all data analyses. Since this study has never been done before, we looked at data from SFA meal challenges [31] and determined the sample size needed to see a 10 % change (our determined effect size) in our outcomes. To observe a 10 % increase in fat oxidation and EE during the SFA meal challenges from visit 1 to 2, a sample size of 14 total participants was needed (assuming 80 % power and an alpha of 0.05) for our PUFA group. Once 26 participants completed the study (determined sample size), participant testing was terminated. Descriptive statistics including mean, range, and standard deviation were calculated for all outcome variables. A Student’s t test for paired comparisons was used to determine significant differences between pre- and post-diet for both the PUFA group and the control group values for anthropometrics, blood pressure, and fasting blood lipids. Change in EE RER, Fox, and carbohydrate oxidation (CHOox) was calculated as time-point value minus baseline value. In order to isolate the meal response and make it independent from potential influence or bias from fasting differences, we calculated metabolic meal responses as change from fasting/baseline any time there were fasting differences from pre- to post-diet. A two-way repeated measures ANOVA was used to test for differences in EE, DIT, RER, and substrate oxidation for the two treatment conditions (PUFA vs. control diet). Additionally, incremental area under the curve (iAUC) was calculated for RER, Fox, and CHOox and compared using an ANOVA. When significance was found, post hoc analyses were done using a Tukey’s test. Statistical significance was set at p < 0.05. Data are presented as mean ± SE unless specified otherwise.

Results

Participants

Twenty-eight participants were recruited to participate in this study. However, two participants were excluded from further participation due to poor compliance (in the control group). Thus, twenty-six men and women adhered to the entire study protocol and were included in data analyses (PUFA-rich diet: n = 8 men and n = 8 women; control diet: n = 5 men and n = 5 women). We also had a higher dropout rate in the control group which resulted in an uneven number of participants in each group at the end of the study. Participant characteristics and blood lipids from pre- to post-diet are presented in Table 4. In the PUFA-rich diet, there were no significant changes in height, body weight, BMI, body fat %, waist circumference, hip circumference, or WHR between study visits. There was a slight decrease in body weight (66.6 ± 10.7–66.2 ± 10.9 kg, p < 0.001) and subsequently BMI (21.9 ± 2.1–21.7 ± 2.2 kg/m2, p < 0.001) for the control diet from pre- to post-diet visits. There were significant decreases in total cholesterol, triglycerides, non-HDL, LDL cholesterol, VLDL cholesterol, and cholesterol/HDL ratio from pre- to post-diet visit (p < 0.01) in the PUFA-rich diet and for triglycerides, LDL and VLDL cholesterol in the control diet (p < 0.05). Furthermore, at baseline, there were no significant differences for EE (16.3 ± 0.8 vs. 16.8 ± 0.8 kcal/20 min), RER (0.83 ± 0.0 vs. 0.83 ± 0.0), Fox (0.94 ± 0.07 vs. 1.05 ± 0.09 g/20 min), or CHOox (1.97 ± 0.13 vs. 1.97 ± 0.13 g/20 min) between the PUFA-rich diet and control diet, respectively.

Table 4 Participant characteristics
Full size table

Fasting measurements

To assess any changes induced by the 7-day diet, fasting RMR and fasting substrate utilization [RER, Fox, and carbohydrate oxidation (CHOox)] were analyzed (Table 5). For RMR, there were no significant differences from pre- to post-diet in either the PUFA-rich diet or control diet, ns. A two-way ANOVA revealed that the PUFA-rich diet significantly increased fasting RER from pre- to post-diet (p = 0.02), whereas the control diet showed no difference (ns). Similarly, the PUFA-rich diet significantly decreased fasting Fox from pre- to post-diet (p = 0.03), while there were no significant differences from pre- to post-diet in the control diet (ns). Finally, the PUFA-rich diet significantly increased fasting CHOox from pre- to post-diet (p = 0.02), whereas the control diet showed no difference (ns).

Table 5 Fasting energy expenditure and substrate oxidation measurements
Full size table

Metabolic meal responses

To assess metabolic responses to the two SFA-rich HF meal challenges (provided at breakfast and lunch) before and after the 7-day diet, change in EE, DIT, RER, Fox, and CHOox was analyzed. The time course for meal responses for change in EE from pre- to post-diet visits is shown in Fig. 1. For change in EE, there was a significant main effect of time in both the PUFA-rich and control diets (p < 0.001), but no visit effect (ns) or time × visit interactions (ns). Furthermore, the iAUC for EE for the PUFA-rich diet showed no differences from pre- to post-diet (118.9 ± 20.6–126.9 ± 14.1 kcal/8h, ns). Not surprisingly, the iAUC for EE for the control diet showed no differences from pre- to post-diet (145.9 ± 13.0–136.9 ± 12.3 kcal/8h, ns). Lastly, we also calculated DIT and found no differences from pre- to post-diet in either the PUFA-rich diet (2.7 ± 0.1–2.9 ± 0.1 kcals/8h) or the control diet (3.4 ± 0.3–3.0 ± 0.3 kcals/8h).

Fig. 1
figure 1

Time course of meal responses for change in energy expenditure in the PUFA-rich diet and control diet from pre- to post-diet. Values are presented as mean ± SE. A two-way repeated measures ANOVA revealed a significant main effect of time in both the PUFA-rich diet (a, n = 16) and control diet (b, n = 10) (p < 0.001), but no visit effect (ns) or time × visit interactions (ns) for change in EE. The iAUC for EE showed no change from pre- to post-diet in both the PUFA-rich diet (ns, c) and the control diet (ns, d). iAUC incremental area under the curve, EE energy expenditure, M meals occurred at time zero and 240 min

Full size image

Substrate oxidation

The time course for meal responses for RER for pre- and post-diet visits for the PUFA-rich and control diets are shown in Fig. 2. There was a significant main effect of time (p < 0.001) and visit (p < 0.001), but no visit × time interaction (ns). After the 7-day PUFA-rich diet, change in RER was significantly lower versus the pre-diet visit (p < 0.001). Similarly, there was a significant decrease in RER iAUC from pre- to post-intervention for the PUFA-rich diet (0.7 ± 0.3 to −0.5 ± 0.4 units/8h, p < 0.01). Alternatively, the control diet had no change in the time course of RER between visits or RER iAUC from pre- to post-intervention (0.5 ± 0.3–0.5 ± 0.4 units/8h, ns).

Fig. 2
figure 2

Time course of meal responses for RER in the PUFA-rich diet and control diet from pre- to post-diet. Values are presented as mean ± SE. A two-way repeated measures ANOVA revealed a significant main effect of time (p < 0.001) and visit (p < 0.001), but no visit × time interaction (ns). After the 7 day PUFA-rich diet (n = 16), change in RER was significantly lower versus the pre-diet visit (p < 0.001, a). The control diet (n = 10) had no change in the time course of RER between visits (ns, b). Similarly, there was a significant decrease in RER iAUC from pre- to post-intervention for the PUFA-rich diet (p < 0.01, c). The control diet had no change in RER iAUC from pre- to post-intervention (ns, d). iAUC incremental area under the curve, M meals occurred at time zero and 240 min, RER respiratory exchange ratio

Full size image

For change in Fox, there was a significant main effect of time (p < 0.001) and visit (p < 0.001), but no time × visit interaction (ns). Change in Fox was significantly increased at the post-diet visit compared to the pre-diet visit for the PUFA-rich diet with no differences in the control diet (Fig. 3). Further, the iAUC for Fox increased from pre- to post-diet for the PUFA-rich diet (Fig. 3c: 1.0 ± 1.9–9.1 ± 1.9 g/8h, p < 0.001) with no differences in the control diet (Fig. 3d: 2.9 ± 2.3–4.8 ± 2.2 g/8h, ns).

Fig. 3
figure 3

Meal responses for the change in fat oxidation in PUFA-rich diet and control diet from pre- to post-diet. Values are presented as mean ± SE. Asterisk indicates significant difference from pre- to post-diet of p < 0.05. A two-way repeated measures ANOVA revealed there was a significant main effect of time (p < 0.001) and visit (p < 0.001), but no visit × time interaction (ns). After the 7-day PUFA-rich diet (n = 16), Fox was significantly higher versus the pre-diet (p < 0.001, a). The control diet (n = 10) had no change in the time course of Fox between visits (ns, b). The iAUC for Fox significantly increased from pre- to post-diet for the PUFA-rich diet (p < 0.001, c) with no differences in the control diet (ns, d). Fox fat oxidation, iAUC incremental area under the curve, M meals occurred at time zero and 240 min

Full size image

Similarly, for change in CHOox for the PUFA-rich diet, there was a significant main effect of time (p < 0.001) and visit (p < 0.001), but no time × visit interaction (ns). Change in CHOox was significantly decreased from pre- to post-diet for the PUFA-rich diet (p < 0.001) (Fig. 4). Similarly, the iAUC for CHOox decreased from pre- to post-diet (Fig. 4c: 28.2 ± 6.6–10.8 ± 5.9 g/8h, p = 0.01) in the PUFA-rich diet with no change in the control diet (Fig. 4d: 28.5 ± 5.9–24.3 ± 5.5 g/8h, ns).

Fig. 4
figure 4

Meal responses for the change in carbohydrate oxidation in PUFA-rich diet and control diet from pre- to post-diet. Values are presented as mean ± SE. Asterisk indicates significant difference from pre- to post-diet of p < 0.05. A two-way repeated measures ANOVA revealed there was a significant main effect of time (p < 0.001) and visit (p < 0.001), but no visit × time interaction (ns). After the 7-day PUFA-rich diet (n = 16), CHOox was significantly lower versus the pre-diet (p < 0.001, a). The control diet (n = 10) had no change in the time course of CHOox between visits (ns, b). The iAUC for CHOox decreased from pre- to post-diet (p = 0.01, c) in the PUFA-rich diet with no change in the control diet (ns, d). CHOox carbohydrate oxidation, iAUC incremental area under the curve, M meals occurred at time zero and 240 min

Full size image

Finally, we looked at the absolute postprandial values of substrate oxidation in addition to the change data prevented in previous paragraphs. The Fox data for the PUFA group are shown in Supplemental Materials. The statistical significance remained with postprandial Fox being higher (p = 0.01) and CHOox lower (p = 0.05) at the post-diet versus pre-diet visits for the PUFA group. This shows that despite lower fasting Fox after the 7-day PUFA-rich diet, postprandial Fox was higher following the SFA-rich test meals.

Discussion

Body weight gain results from a chronic imbalance of energy intake and energy expenditure, and diets high in fat have been identified as one of the factors associated with body weight gain. However, there is considerable evidence that not all types of fats influence physiological responses in the human body in the same way which has led to certain fats being labeled as obesogenic, while others are not [35]. In particular, HF diets, which emphasize PUFA, do not seem to lead to obesity, while high SFA diets are considered obesogenic [36]. Several studies have reported increased RMRs, DIT, and rates of Fox for PUFA-rich meals when compared to SFA-rich meals; however, to our knowledge, no study has specifically examined the effects of longer-term PUFA consumption on acute metabolic responses to HF meals rich in SFA. For the first time, we showed that a PUFA-rich diet was able to offer some metabolic protection to the body from the SFA-rich HF meals through higher postprandial Fox rates. Interestingly, the 7-day PUFA-rich diet actually increased fasting RER and CHOox and reduced fasting Fox compared to the control diet. Conversely, postprandial EE did not differ from pre- to post-PUFA-rich diet, and there were no differences in RMR as a result of the 7-day PUFA-rich diet. These findings indicate that the substrate oxidation responses to high daily PUFA intake differ for the fasting versus postprandial state in normal-weight adults.

As mentioned above, previous studies on the impact of dietary FA composition have primarily assessed acute substrate oxidation and DIT responses to meal challenges. Regardless of whether investigators have looked at meal responses or diet responses, however, studies are done that directly compare the metabolic response between different FA compositions (e.g., high PUFA vs. high SFA test meals). Until this study, no one had actually studied meal responses to the exact same meal (rich in SFAs) before and after manipulating the FA composition of the daily diet. While this study is unique, we can still draw some comparisons to previous studies. Our higher postprandial Fox following the PUFA-rich diet supports findings of previous literature showing that Fox is higher following unsaturated fat vs. SFA meal consumption [37–39]. Conversely, our PUFA-rich diet was not strong enough to alter the EE response following the SFA-rich meal challenges. We have previously shown that DIT is higher following a PUFA-rich meal when compared to SFA- and MUFA-rich meals [31]; however, that again was a direct comparison between meals of varying FA composition and was different from this study design. It is possible that PUFAs may elicit a stronger DIT when directly compared to a meal high in SFAs, but high PUFA intake is unable to alter DIT after a SFA-rich meal.

We were surprised to see that the PUFA-rich diet resulted in lower fasting Fox (higher CHOox). Since this study was not a mechanistic study, we do not know the exact mechanism by which PUFAs increased fasting CHOox; however, we can speculate on why this occurred. Since PUFA regulation of gene expression is widespread, several mechanisms may be involved. One mechanism may involve PUFA suppression of de novo lipogenesis. PUFA-rich diets in rodents have been shown to suppress a variety of genes involved in fatty acid synthesis, including stearoyl-CoA desaturase-1, acetyl CoA carboxylase, and fatty acid synthase in the liver [40–45]. A decrease in lipogenic genes and thus de novo lipogenesis may result in higher oxidation of carbohydrates in the fasted state.

The higher postprandial Fox after the high PUFA-rich diet was likely greater exogenous fat oxidation from the HF meal, so any PUFA suppression of de novo lipogenesis would not have had any effect on the postprandial Fox rates. Yet another possibility is that PUFAs have the ability to control the transcriptional activity of nuclear receptors and thereby the transcription rate of specified genes related to lipid and carbohydrate metabolism [46]. In addition to PPARs, PUFAs can bind to several nuclear receptors including liver X receptor (LXR) [47]. Kase et al. [48, 49] showed that chronic activation of LXRs may affect glucose uptake and oxidation. Considering the transcriptional control exerted by LXR on ChREBP [50], it would be interesting to determine whether PUFAs effects on CHOox are triggered, at least in part by ChREBP through an LXR-dependent pathway. Lastly, it is possible that there is another unidentified mechanism by which PUFAs are increasing carbohydrate oxidation, and future research in this area is warranted.

Since this study employed an outpatient feeding protocol, we wanted to look at fasting levels to see the chronic diet effect, but we also wanted to isolate and study the SFA-rich meal response before and after the high PUFA diet. If one does not control for differences that may exist at fasting, then any differences seen postprandially could be an artifact of those fasting/baseline differences. Since we found differences in our fasting measures of RER, Fox, and CHOox from pre- to post-diet in the PUFA group, we looked at change from baseline/fasting for the meal responses. This was done to isolate our meal response and not allow it to be influenced or biased by the differences in fasting substrate oxidation that occurred because of the diet. However, we also analyzed the absolute values for the postprandial response before and after the PUFA-rich diet and found that our significant differences remained (Fox was higher for post- versus pre-PUFA-rich diet; p = 0.01, Fig. 1s of supplemental documents). Therefore, both the change data and absolute data show greater Fox to the SFA-rich meals after the 7-day PUFA-rich diet.

Other results of this study that merit discussion are the blood lipid data. It has been well documented that increasing PUFA intake can improve blood lipids and decrease cardiovascular disease risk. In order for participants to be included in this study, they could not have dyslipidemia from the fasting blood sample results. Therefore, we did not expect to see much change in blood lipids in these normal lipidemic adults. Yet, in just 7 days, participants in the PUFA-rich group were able to significantly and drastically improve their lipid profile as previously demonstrated by Hodson et al. [51]. We saw an average decrease of 30.9 mg/dL in total cholesterol, 18.9 mg/dL in triglycerides, 31.9 mg/dL in LDL cholesterol, and 4.1 mg/dL in VLDL cholesterol. It is important to note that the PUFA-rich diet also had more dietary fiber (35.9 ± 2.0 vs. 19.5 ± 0.8 g/day), less dietary cholesterol (191.4 ± 8.8 vs. 143.4 ± 8.2 mg/day), less trans FA (0.01 ± 0.0 vs. 3.3 ± 0.2 g/day), and less SFA (10.4 ± 0.6 vs. 36.2 ± 1.4 g/day) compared to the control diet. Therefore, we do not know whether it was the increased PUFA content, higher fiber, lower trans FA and SFA, or a combination of all of these that elicited the drastic improvements. Regardless of which component(s) of the diet caused these results, this specific PUFA-rich diet provided to participants for 7 days drastically improved their already normal lipid profile and could be used as a therapeutic tool for rapid changes in blood lipids.

There were some limitations in the current study. The physical activity status and meal compliance for the day before visits were self-reported which could impact the outcomes tested in this study. Further, our participants were free-living, so there was a possibility for participants to consume other foods and beverages throughout the day. However, participants had to come to the HNL every day to consume breakfast, and they were reminded to only eat and drink what was provided to them for the remainder of the day. Another limitation of this study is that we used a whole foods approach to our feeding study. By using a whole foods approach, it is impossible to discern whether changes we saw from the 7-day PUFA feeding period can be attributed to increases in one specific PUFA, one food source of PUFA, a combination of different types of PUFAs or a combination of higher fiber, lower trans-fat, etc., in the PUFA-rich diet group. Another limitation of this study is the relatively high contribution of PUFAs to the diet (21 % of total energy). While this may not be feasible to consume on a daily basis for most adults, we chose to provide such a high amount of PUFAs for 7 day to stress the body and look for a metabolic response. Future studies are needed to examine different doses from PUFAs in order to see what amount and type of PUFA is necessary to elicit a positive metabolic response. Lastly, the study sample was comprised of apparently healthy men and women with a normal BMI, so these results cannot be extrapolated to other populations.

Conclusions

The present findings indicate that a moderate-fat diet rich in PUFAs can compensate for the detrimental metabolic effects of eating two SFA-rich HF meals. Consuming a 7-day diet rich in PUFAs led to greater increases in fasting RER, but increased postprandial fat oxidation following two SFA-rich HF meals. Therefore, people may be protected from the negative effects from eating occasional high SFA meals if they are consuming more PUFAs on a regular basis. Importantly, we cannot overstate the role of energy balance in order to prevent fat mass accumulation. The impact of substrate oxidation on metabolism and body weight stores is assuming that a person is in energy balance. An energy surplus will override any potential metabolic differences that exist based on the type of fatty acid being consumed. Future studies are needed to investigate long-term consumption of various dosages and types of PUFAs and their impact on altering energy intake in a variety of populations as well as the mechanism by which PUFAs are altering fasting and postprandial energy expenditure and substrate oxidation.