Abstract
Postprandial dysmetabolism in type 2 diabetes (T2D) is exacerbated by prolonged sitting and may trigger inflammation and oxidative stress. It is unknown what impact countermeasures to prolonged sitting have on the postprandial lipidome.
In this study, we investigated the effects of regular interruptions to sitting, compared with prolonged sitting, on the postprandial plasma lipidome.
Randomized crossover experimental trial.
Participants underwent three 7-hour conditions: uninterrupted sitting (SIT); light-intensity walking interruptions (LW); and simple resistance activity interruptions (SRA).
Baseline (fasting) and 7-hour (postprandial) plasma samples from 21 inactive overweight/obese adults with T2D were analyzed for 338 lipid species using mass spectrometry.
Using mixed model analysis (controlling for baseline outcome variable, gender, body mass index, and condition order), the percentage change in lipid species (baseline to 7 hours) was compared between conditions with Benjamini–Hochberg correction.
Thirty-seven lipids were different between conditions (P < 0.05). Compared with SIT, postprandial elevations in diacylglycerols, triacylglycerols, and phosphatidylethanolamines were attenuated in LW and SRA. Plasmalogens and lysoalkylphosphatidylcholines were reduced in SIT, compared with attenuated reductions or elevations in LW and SRA. Phosphatidylserines were elevated with LW, compared with reductions in SIT and SRA.
Compared with SIT, LW and SRA were associated with reductions in lipids associated with inflammation; increased concentrations of lipids associated with antioxidant capacity; and differential changes in species associated with platelet activation. Acutely interrupting prolonged sitting time may impart beneficial effects on the postprandial plasma lipidome of adults with T2D. Evidence on longer-term intervention is needed.
Lifestyle interventions are recommended as front-line therapy for the management of type 2 diabetes (T2D), including moderate-to-vigorous aerobic and resistance exercises (1). Despite this, many with T2D remain inactive and spend a large proportion of their waking hours in sedentary behaviors (sitting) (2, 3). Detrimental associations of sedentary behaviors with increased risk of cardiometabolic diseases, including T2D, have been consistently observed (4, 5). Importantly, the manner in which sedentary time is accumulated seems to be important, in which frequent interruptions in sitting are beneficially associated with biological markers of metabolic risk, compared with an equal volume of sitting accumulated in prolonged uninterrupted bouts (6).
From a mechanistic perspective, inflammation provides a potential link between sedentary time and chronic diseases, as high overall sitting time and television viewing time are detrimentally associated with key inflammatory markers (7–10). A likely mechanism by which exercise and active breaks in sitting may act to reduce inflammation and disease risk is by lowering postprandial glucose excursions and altering postprandial lipid responses.
Plasma lipids for cardiovascular risk prediction are routinely measured in the fasting state (11). However, except in the early morning, typical Western lifestyles are mostly spent in a postprandial state. Transient postprandial glucose and lipid excursions (postprandial dysmetabolism) promote oxidative stress and inflammation, which create an environment conducive to the development of metabolic and cardiovascular disease (12). It is, therefore, important to understand factors that influence the duration and magnitude of postprandial hyperlipidemia. Furthermore, lifestyle modifications that can practically and effectively attenuate postprandial lipid excursions are likely to be important in preventing and managing disease.
There are >500 lipid species present in plasma that, in addition to their role as energy substrates, are important inflammatory and signaling mediators (13). Patients with T2D exhibit an altered fasting plasma lipidome compared with healthy controls, demonstrating dysregulation of several lipid classes that are mechanistically linked to oxidative stress and inflammation, and that may be both a cause and a consequence of insulin resistance (14–16). Recent studies have demonstrated that meal composition and an acute exercise bout can modulate plasma and tissue lipidome profiles (15, 17, 18). However, the potential benefits of interrupting prolonged sitting on the postprandial lipidome have yet to be explored.
This ancillary study was conducted on a subset of participants from a larger trial that has been previously described (19). The main study showed that interrupting sitting time with light-intensity activities lowered postprandial glucose, insulin, and triglyceride levels, compared with uninterrupted sitting (SIT) (19). The aim of the current study was to determine the effects of light-intensity breaks in sitting on the postprandial plasma lipidome in adults with T2D. We hypothesized that interrupting sitting time would differentially alter postprandial plasma lipidome profiles, in comparison with SIT.
Materials and Methods
Study overview
This trial was approved by the Alfred Hospital Human Ethics Committee (Melbourne, VIC 3004, Australia), and was carried out in accordance with the Declaration of Helsinki. Participants provided signed informed consent. The study is registered with the Australian New Zealand Clinical Trials Registry (ACTRN12613000576729).
Participant characteristics, screening, and testing procedures for the main study have been described in Supplemental Material and previously (19). Of the 24 participants in the main study, 21 (13 men, 8 women) overweight/obese adults with T2D were included in the current investigation. Three participants from the main study were excluded due to lack of available plasma samples for the baseline or 7-hour time points. The resulting unbalanced order of intervention was adjusted for in the statistical model.
Trial conditions
Details of the trial conditions have been described in Supplemental Material and previously (19). Each experimental condition commenced upon starting the breakfast meal (0 hours); participants consumed lunch at 3.5 hours. Fasting (−1-hour) and postprandial (7-hour) blood samples were analyzed. The three trial conditions were as follows:
SIT. Participants sat upright in a comfortable chair throughout the experimental period and were instructed to minimize excessive movement, only rising to attend the lavatory.
Light-intensity walking interruption (LW). Participants rose from the seated position every 30 minutes and completed a 3-minute bout of light walking on a treadmill (3.2 km/h, zero gradient), and then returned to the seated position. This procedure was undertaken on 12 occasions (no break was undertaken at lunch or at 7 hours), totaling 36 minutes of activity.
Simple resistance activity interruption (SRA). Identical to the LW condition, except with bouts of simple resistance activities, totaling 36 minutes. Each 3 minutes was divided into a total of nine 20-second movement segments, alternating between body weight–resisted half-squats, calf raises, and knee raises with a gluteal contraction.
Lipidomic analysis
The lipidomic analysis method used was semiquantitative, performed as previously described (20) with some minor adjustments (see Supplemental Material). The nomenclature used when referring to individual lipid species is based on LIPID Metabolites and Pathways (21), with recent revisions (22). For a number of the lipids that contain two fatty acid chains, the mass spectrometry–based measurements do not directly determine the constituent double bonds, but rather the sum of the number of carbons and the sum of the number of double bonds across both fatty acids. These species are denoted as the combined length and number of double bonds (e.g., 36:4).
Statistical analyses
Normality of the outcome variables was checked using histograms and the number of severe outliers. A severe outlier was defined as value <25th percentile minus three interquartile range or >75th percentile plus three interquartile range (23). All models were adjusted for potentially important covariates explaining residual outcome variance [sex and body mass index (BMI)], baseline outcome values, and period effects (order of condition). Sensitivity analyses including additional covariates (statin use, waist:hip ratio) were performed to test the robustness of the results. Plots of residuals vs predicted values indicated that the data were normally distributed and homoschedastic.
Linear mixed models accounting for dependency in the data (repeated measures) were used to evaluate the differential effects of the trial conditions on percentage change in lipid species and lipid classes from baseline to postintervention.
All P values obtained were corrected for multiple comparisons using the false discovery rate method of Benjamini–Hochberg (BH) approach (24). Post hoc, pairwise analyses were performed using postestimation commands of the linear mixed model, and the resultant pairwise P values were corrected by the Dunn–Sidak approach. A corrected P value of <0.05 was considered to be statistically significant. All statistical analyses were performed using Stata 14.1 for Windows (StataCorp LP, College Station, TX).
Results
The 21 participants included in the current analysis had a mean (±standard deviation) age of 63 ± 6 years and BMI 32.7 ± 3.4 kg/m2; 20 participants were taking metformin medication. Detailed participant characteristics are provided in Supplemental Material and previously (19). Respiratory exchange ratio for SIT (0.85 ± 0.02), LW (0.83 ± 0.01), and SRA (0.81 ± 0.01) indicate increased fat oxidation in SRA and LW compared with SIT, and SRA compared with LW (P value for difference = 0.049). Controlling for treatment order, no significant difference was observed in fasting lipid concentrations between conditions for any lipid species (P > 0.05), indicating reproducibility of the lipidomic profile at baseline. Before correction for multiple comparisons, the percent change from fasting (baseline) to 7-hour postprandial plasma concentrations of 4 of 24 lipid classes and subclasses [diacylglycerol (DG), triacylglycerol (TG), alkenylphosphatidylcholine [PC(P)], and phosphatidylserine (PS)] was significantly different between conditions (Supplemental Table 1). Significance was lost after BH correction, but strong trends remained for TGs (P = 0.06), PS (P = 0.06), and PC(P) (P = 0.08; Fig. 1). Before correction, 85 of 338 lipid species were significantly different between conditions, and 37 remained significant following BH correction (Supplemental Table 2). These included species in the DG, TG, PC(P), PS, phosphatidylethanolamine (PE), and lysoalkylphosphatidylcholine [LPC(O)] lipid classes and subclasses.
Postprandial plasma lipid class and subclass percentage change from fasting baseline. Data are presented as mean ± standard error of the mean. Following BH correction, no significant differences (P < 0.05) were observed across treatment conditions (linear mixed model analysis). However, three lipid classes showed strong trends, as indicated.
Diacylglycerol and triacylglycerol
All postprandial DG and TG species were elevated over the day compared with baseline fasting values (Figs. 2 and 3). Of the 21 DG and 42 TG species, 9 and 16 species, respectively, were significantly different between conditions following BH correction. Post hoc pairwise comparisons showed attenuated elevations for 9 and 16 species in LW, and 8 and 16 species in SRA, for DG and TG species, respectively, compared with SIT. No significant differences were observed between LW and SRA.
Postprandial plasma diacylglycerol lipid species percentage change from fasting baseline. Data are presented as mean ± standard error of the mean for diacylglycerol species that were significantly different (P < 0.05) across treatment conditions in the linear mixed model analysis, following BH correction. *Significant compared with SIT (post hoc pairwise analysis with Dunn–Sidak correction; P < 0.05). Mean ± standard error of the mean fasting concentration for each diacylglycerol species is indicated on the right (µM).
Postprandial plasma triacylglycerol lipid species percentage change from fasting baseline. Data are presented as mean ± standard error of the mean for triacylglycerol species that were significantly different (P < 0.05) across treatment conditions in the linear mixed model analysis, following BH correction. *Significant compared with SIT (post hoc pairwise analysis with Dunn-Sidak correction; P < 0.05). Mean ± standard error of the mean fasting concentration for each triacylglycerol species is indicated on the right (µM).
The difference between the activity and SIT conditions appeared to be largely driven by lipids containing saturated fatty acids (SFAs), particularly palmitic acid (16:0). All 9 DG and 16 TGs that were significantly different between conditions contained at least one SFA; 5 DGs and 10 TGs contained at least one 16:0.
Alkenylphosphatidylcholine
Postprandial PC(P) (plasmalogen) species were reduced in SIT, compared with baseline (Fig. 4). By contrast, PC(P) species were generally elevated or unchanged in LW and elevated or reduced to a lesser extent (compared with SIT) in SRA. Of the 13 PC(P) species detected, PC(P) 30:0, 32:1, 34:1, 34:2, and 36:2 were significantly different between conditions following BH correction. Pairwise, all five species were significantly different in LW, and PC(P) 34:1 and 36:2 in SRA, compared with SIT. No significant differences were observed between LW and SRA.
Postprandial plasma PC(P), PE, PS, and LPC(O) lipid species percentage change from fasting baseline. Data are presented as mean ± standard error of the mean for species that were significantly different (P < 0.05) across treatment conditions in the linear mixed model analysis, following BH correction. *Significant compared with SIT; #significant compared with SRA (post hoc pairwise analysis with Dunn–Sidak correction; P < 0.05). Mean ± standard error of the mean fasting concentration for each lipid species is indicated on the right (µM).
Phosphatidylethanolamine
Postprandial PE species were generally elevated or unchanged in all conditions, compared with baseline (Fig. 4). Of the 15 PE species detected, PE 32:1, 34:2, 36:2, and 36:3 were significantly different between conditions following BH correction. Pairwise, PE 34:2, 36:2, and 36:3 showed significantly attenuated elevations in LW, and all 4 were attenuated in SRA, compared with SIT. No significant differences were observed between LW and SRA.
Phosphatidylserine
Postprandial PS species were elevated in LW, compared with a reduction in SIT and SRA (Fig. 4). Of the three PS species detected, only PS 38:4 was significantly different between conditions following BH correction, although PS 36:1 and PS 40:6 showed strong trends (P < 0.08, Supplemental Table 2). Pairwise, PS 38:4 was significantly elevated in LW, compared with reduced postprandial concentrations in SIT and SRA. No significant differences were observed between SIT and SRA.
Lysoalkylphosphatidylcholine
LPC(O) species were reduced in SIT, compared with being either elevated or unchanged in LW, or reduced to a lesser extent in SRA (Fig. 4). Of the nine LPC(O) species detected, only LPC(O) 18:1 was significantly different between conditions following BH correction, although five others (16:0, 18:0, 22:1, 24:2, and 24:1) showed moderate to strong trends (P < 0.20, Supplemental Table 2). Pairwise, LPC(O) 18:1 was unchanged in LW, compared with reduced postprandial concentrations in SIT. No significant differences were observed between SIT and SRA, or LW and SRA.
Sensitivity analyses
Waist:hip ratio was added as a covariate in the statistical model as an alternative method of assessing fat mass, in addition to BMI. Waist:hip ratio and BMI were moderately correlated (r = −0.42). Addition of waist:hip ratio to the model identified the same lipid species as different between conditions as the original model, as well as three additional species (following BH correction). These additional species included trihexosylceramide 22:0 and PS 36:1, which were significantly elevated in LW, compared with reductions in SIT and SRA; and the postprandial elevation in DG 16:0/22:6 was significantly attenuated in LW and SRA, compared with SIT. Including statin use as a covariate in the model additionally identified trihexosylceramide 22:0, but moderated cholesteryl ester 18:1 and PC(P) 36:2 to insignificance following BH correction.
Discussion
This study demonstrates that, compared with prolonged sitting, light-intensity activity interruptions are associated with changes in the postprandial plasma lipidome in T2D patients. Following correction for multiple comparisons, lipid species in the DG, TG, PC(P), PE, PS, and LPC(O) lipid classes and subclasses were significantly different between conditions. The type of activity that was performed during each break (LW vs SRA) appeared to induce differential postprandial effects, including lower respiratory exchange ratio (higher fat oxidation) in the SRA condition. This study extends upon previous experimental trials, which have shown benefits for traditional cardiometabolic risk biomarkers with breaks in sitting time. Our findings provide new insight into the potential candidate mechanisms driving the benefits of interrupting prolonged sitting and support this approach as a lifestyle-based treatment strategy for patients with T2D.
Diacylglycerol and triacylglycerol
Previous studies have demonstrated that frequent active breaks from sitting are as or more effective at reducing postprandial plasma triglyceride levels, compared with a continuous bout of exercise (25, 26). Taken together, these previous studies and our current results suggest that the type, intensity, frequency, and duration of activity breaks, and their timing relative to meals, may be important in attenuating postprandial triglyceride responses. In the current study, the significant between-condition effect for TGs as a lipid class (P = 0.004) was attenuated after correction for multiple comparisons (P = 0.057). In pairwise comparisons, TGs were reduced in LW and SRA, compared with SIT (P < 0.05). This contrasts somewhat with the findings from the main trial, in which SRA, but not LW, breaks significantly lowered postprandial plasma triglyceride responses in comparison with SIT (19). These differences between our current and previously reported results are most likely due to the different statistical and analytical methods used to assess plasma TG content (see Supplemental Material).
The DG and TG responses observed in the current analysis are likely to be driven by the major constituents of the standardized meals. Differences between the LW and SRA conditions, compared with SIT, could indicate differences in both DG and TG entry into the circulation and how quickly they are removed. As little as one day of inactivity with high levels of sitting time can reduce whole-body insulin action, even when energy intake is reduced to match expenditure (27). Therefore, increased entry of lipids into the plasma in the SIT condition may result from an acute increase in insulin resistance of the adipose tissue, which would impair insulin-mediated inhibition of lipolysis. Increased levels of fatty acids in the circulation, in combination with acute insulin resistance in the liver, could act to increase hepatic TG synthesis and release of TG-rich very-low-density lipoprotein (16). Moreover, lack of muscle activity in the SIT condition may lead to a reduction in lipoprotein lipase activity, thus reducing muscle-mediated uptake of fatty acids for adenosine triphosphate production and adipose-mediated uptake for storage (28).
The constituent fatty acids incorporated into circulating DG and TG species have important implications for cardiometabolic health. Several of the plasma DGs and TGs that have previously been shown to be positively associated with insulin resistance and excess weight (29) were attenuated in the LW and SRA conditions in the current study, compared with SIT. We also observed that several DG and TG species that were differentially altered in the activity conditions contained the SFA palmitic acid (16:0). Emerging evidence suggests that individual fatty acids within a given class have distinct associations with inflammation and oxidative stress (30). A recent study showed that fasting levels of 14:0 and 16:0 SFAs were positively associated with markers of inflammation, possibly due to stimulation of Toll-like receptors and activation of proinflammatory nuclear factor κB pathways. In contrast, 18:0 SFA was inversely associated with markers of inflammation (30). Therefore, the attenuated increase in 16:0 SFAs observed in the activity conditions in this study could indicate reduced postprandial inflammation (Table 1).
Summary of Changes in the Postprandial Lipidome, Compared With Fasting, in the Prolonged Sitting and Frequent Activity Conditions
| Pathways | Key Lipid Class/Subclass | SIT | LW | SRA |
|---|---|---|---|---|
| Proinflammatory | Diacylglycerol | ↑↑ | ↑ | ↑ |
| Triacylglycerol | ↑↑ | ↑ | ↑ | |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Anti-inflammatory | Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Antioxidant capacity | Alkenylphosphatidylcholine | ↓ | ↑/↔ | ↔ |
| Platelet activation | Phosphatidylserine | ↓ | ↑ | ↓ |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Pathways | Key Lipid Class/Subclass | SIT | LW | SRA |
|---|---|---|---|---|
| Proinflammatory | Diacylglycerol | ↑↑ | ↑ | ↑ |
| Triacylglycerol | ↑↑ | ↑ | ↑ | |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Anti-inflammatory | Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Antioxidant capacity | Alkenylphosphatidylcholine | ↓ | ↑/↔ | ↔ |
| Platelet activation | Phosphatidylserine | ↓ | ↑ | ↓ |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
Summary of Changes in the Postprandial Lipidome, Compared With Fasting, in the Prolonged Sitting and Frequent Activity Conditions
| Pathways | Key Lipid Class/Subclass | SIT | LW | SRA |
|---|---|---|---|---|
| Proinflammatory | Diacylglycerol | ↑↑ | ↑ | ↑ |
| Triacylglycerol | ↑↑ | ↑ | ↑ | |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Anti-inflammatory | Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Antioxidant capacity | Alkenylphosphatidylcholine | ↓ | ↑/↔ | ↔ |
| Platelet activation | Phosphatidylserine | ↓ | ↑ | ↓ |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Pathways | Key Lipid Class/Subclass | SIT | LW | SRA |
|---|---|---|---|---|
| Proinflammatory | Diacylglycerol | ↑↑ | ↑ | ↑ |
| Triacylglycerol | ↑↑ | ↑ | ↑ | |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Anti-inflammatory | Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
| Antioxidant capacity | Alkenylphosphatidylcholine | ↓ | ↑/↔ | ↔ |
| Platelet activation | Phosphatidylserine | ↓ | ↑ | ↓ |
| Phosphatidylethanolamine | ↑↑ | ↑ | ↑ | |
| Lysoalkylphosphatidylcholine | ↓ | ↑/↔ | ↓ |
Alkenylphosphatidylcholine
Plasmalogens are thought to act as endogenous antioxidants, and a decrease in concentration of these species is suggested to relate to higher oxidative stress (31, 32). Indeed, plasmalogens are negatively associated with T2D and cardiovascular disease (14, 15, 31, 33). Postprandial dysmetabolism leading to high plasma lipids may promote oxidative stress and reduce the antioxidative potential of the phospholipid pool (12, 15). In the present analysis, postprandial PC(P) species were generally reduced in the SIT condition; elevated or unchanged in LW; and elevated or reduced to a lesser extent (compared with SIT) in SRA. Statin use has been shown to normalize plasma plasmalogen levels in patients with metabolic syndrome; however, additionally controlling for statin use in the statistical model only attenuated the effect for one plasmalogen species—PC(P) 36:2 (15). These results suggest a potentially beneficial effect of active breaks in sitting for antioxidative capacity in the postprandial state (Table 1). The mechanisms behind this benefit are not clear, but could relate to reduced breakdown of PC(P) plasmalogens due to greater clearance of proinflammatory SFAs in the activity conditions, and therefore lower levels of oxidative stress. These findings suggest that some of the benefit from breaking up prolonged sitting in those with T2D may be via improved antioxidant capacity.
Phosphatidylethanolamine
Several PE species showed attenuated elevations in the LW and SRA conditions, relative to SIT. High plasma concentrations of PE and a decreased phosphatidylcholine:PE ratio are positively associated with obesity, prediabetes, and T2D (14, 20). However, plasma PE were not reported to be significantly associated with insulin resistance or overweight/obesity (29). PE may also play a role in thrombus formation, in which it can act both synergistically with and independently of PS to catalyze the formation of thrombin from prothrombin (34) (Table 1).
Phosphatidylserine
When platelets are activated, they release platelet-derived microparticles, which are important in inflammatory processes and play a central role in thrombosis in part due to the function of PS, which are contained on their cell surface (35). It has been suggested that a sedentary lifestyle may affect exercise-induced platelet function in a prothrombotic manner (36, 37). Compared with decreases in SIT and SRA, we observed a consistent increase in plasma PS species in the LW condition, possibly due to greater shear stress–mediated platelet activation and release of PS-containing platelet-derived microparticles (Table 1). This is a novel finding; however, hemostasis and thrombus formation are regulated via complex signaling pathways, and thus the physiological outcome of an increase in PS species in the LW condition is unclear.
Lysoalkylphosphatidylcholine
Platelet-activating factor is a proinflammatory, prothrombotic signaling lipid, whereas LPC(O) opposes the activities of platelet-activating factor (38). Several LPC(O) species are negatively associated with obesity, insulin resistance, and T2D (14, 29, 39). In the current analysis, only LPC(O) 18:1 remained significant following BH correction; however, similar trends were observed for almost all species within this lipid subclass. LPC(O) species tended to decrease or remain unchanged postprandially in SIT and SRA, whereas several species tended to increase or remain unchanged in the LW condition. Although no strong conclusions can be drawn from these data, the overall trends suggest a potentially beneficial effect of LW breaks on LPC(O) species, indicating a possible anti-inflammatory and antithrombotic effect (Table 1).
Strengths and limitations
Key strengths of our study include the focus on men and women with overt T2D who are at high risk of postprandial dysmetabolism and could derive substantial benefits from lifestyle interventions that help to manage exaggerated postprandial lipid excursions. Participants were their own controls, which reduced variance and permitted a smaller sample size. We also incorporated standardized trial condition lead-in periods, strict behavioral supervision, and standardized feeding of a typical Western diet during experimental conditions. False positives were minimized by the use of mixed model analysis that accounted for repeated measures and adjusted for potential confounders and correction for multiple testing with the false discovery rate method by Benjamini and Hochberg (24). Limitations of the study include the acute nature of the experimental protocol. Further experimental insights into the chronic effects of breaking up sedentary time on the postprandial plasma lipidome are required. The size of the cohort, although sufficient to detect significant differences between conditions, was relatively small. A larger cohort may uncover a greater number of differentially regulated lipid classes, subclasses, and species and would allow for more in-depth analysis of the potential effects of other covariates (such as ethnicity, visceral adiposity, and presence of the metabolic syndrome or its components). The results from this study may not be generalizable to other populations, e.g., nondiabetic, nonobese, less well-controlled T2D (insulin-dependent), those with diabetic complications or other comorbidities, and those of non-Caucasian ethnicity. Moreover, the generalizability of this highly controlled laboratory-based trial to free-living settings is not certain, and the behaviors implemented in this study may not reflect habitual behaviors. For example, the SIT condition is unlikely to represent normal daily sitting patterns and exaggerate prolonged sitting. The standardized meals reflecting a Western diet will inevitably vary in daily life settings, and how such variations interact with physical activity is also unknown. Finally, future studies should incorporate concurrent measures of inflammatory and oxidative stress markers, which could help to clarify the mechanisms by which prolonged sitting may affect health outcomes via alterations in the postprandial plasma lipidome.
Summary
Much of what is known about the link between the plasma lipidome and disease is understood within the context of the fasted state. However, increasing evidence suggests that measurement of postprandial glucose, lipids, inflammation, and oxidative stress may provide important clinical information concerning susceptibility to disease onset and progression (40). Patients with T2D are vulnerable to postprandial oxidative stress and may benefit from lifestyle interventions aimed at improving antioxidant capacity and attenuating postprandial dysmetabolism. This study builds on recent findings that the postprandial lipidome can be modified by alterations in diet and exercise, and it poses a number of questions for future research. We have shown that regular active interruptions in sitting may be of benefit in reducing postprandial proinflammatory lipids and increasing concentrations of lipids with antioxidant capacity in adults with well-controlled noninsulin-dependent T2D. Moreover, we have observed differential changes in lipid species linked to platelet activation. The majority (n = 20) of participants in this study were taking metformin, and 15 were on statins, indicating a potentially beneficial effect of breaks in sitting in addition to medication. These areas of investigation will be important to target in future studies and could further elucidate the mechanisms driving the cardiometabolic effects linked to reducing and breaking up prolonged sitting.
Abbreviations:
- BH
Benjamini–Hochberg
- BMI
body mass index
- DG
diacylglycerol
- LPC(O)
lysoalkylphosphatidylcholine
- LW
light-intensity walking interruption
- PC(P)
alkenylphosphatidylcholine
- PE
phosphatidylethanolamine
- PS
phosphatidylserine
- SFA
saturated fatty acid
- SIT
uninterrupted sitting
- SRA
simple resistance activity interruption
- T2D
type 2 diabetes
- TG
triacylglycerol.
Acknowledgments
This work was supported by National Health and Medical Research Council (NHMRC) Project Grant 1081734 and the Victorian Government Opperational Infrastructure Support scheme. P.C.D. is supported by an Australian postgraduate award and a Baker Bright Sparks top-up scholarship. B.A.K., P.J.M., N.O., and D.W.D. are supported by the NHMRC Fellowships scheme. M.S.G. is supported by a Flack Fellowship.
Clinical trial registry: ClinicalTrials.gov no. ACTRN12613000576729 (registered 21 May 2013).
Disclosure Summary: The authors have nothing to disclose.
References
Author notes
Address all correspondence and requests for reprints to: Megan S. Grace, PhD, Baker Heart and Diabetes Institute, Level 4 Alfred Centre, 99 Commercial Road, Melbourne, VIC 3004, Australia. E-mail: [email protected].
