Abstract
Context:
Pigment epithelium-derived factor (PEDF) was recently implicated as a metabolic regulatory protein because plasma concentration was increased in obese or insulin resistant adults. To our knowledge, circulating PEDF values in children have not been reported. Because PEDF is a predictor of metabolic health in adults, it may have a similar impact on metabolic profiles in children.
Objective:
The objective of the study was to determine whether PEDF in normal-weight (NW) and overweight/obese (OW) children and young adults varies with age, sex, or body composition or is associated with clinical markers of metabolic disease.
Setting:
Volunteers were tested at the University of Oklahoma Health Sciences Center.
Participants:
Ninety-one NW (8–30 yr old) and 105 OW (8–35 yr old) males and females participated in the study.
Main Outcome Measures:
Body composition, blood pressure, arterial compliance, fasting plasma PEDF, glucose, insulin, (used for homeostasis model assessment of insulin resistance), triglycerides, cholesterol (total, low density lipoprotein, and high density lipoprotein), and C-reactive protein.
Results:
PEDF was 60% higher in the OW vs. NW participants but did not differ between males and females. PEDF was positively correlated with body mass, body mass index, fat and lean mass, fasting insulin, and homeostasis model assessment of insulin resistance in both the NW and OW groups. Multiple regression models revealed that fat and lean mass were significant predictors of circulating PEDF levels independent of age, sex, and body mass index category.
Conclusions:
Plasma PEDF is elevated in OW youth and is positively associated with insulin resistance. These findings suggest that PEDF may play a role in the development of cardiometabolic dysfunction in youth.
Pigment epithelium-derived factor (PEDF) is an antiangiogenic, antiinflammatory, and antioxidant protein derived mostly from the liver and adipose tissue (1–3) and associated with insulin resistance and diabetic complications (2, 4–6). Animal models demonstrated that experimental increases of PEDF promoted insulin resistance in liver and peripheral tissues, whereas neutralization of PEDF restored insulin sensitivity (2). Those results and recent work with adipocytes in culture suggest that PEDF may act as a signal through which excess adipose tissue promotes insulin resistance and metabolic dysfunction (2, 3).
Prior reports showed that PEDF is positively associated with several metabolic risk factors in adults. In Japanese adults, PEDF concentration was positively correlated with body mass index (BMI), waist circumference, fasting triglycerides, glucose, and insulin and inversely correlated with high-density lipoprotein (HDL)-cholesterol (7). Subsequently, Jenkins et al. (5) reported that serum PEDF was higher in adults with type 1 or type 2 diabetes (4) compared with age-matched adults without diabetes; in people with type 2 diabetes, PEDF was positively correlated with BMI and low-density lipoprotein (LDL)-cholesterol (4). Furthermore, in recent studies serum PEDF declined in obese adults after weight loss (6) and increased in normal-weight adults during short-term energy surplus (8). Collectively, the available evidence suggests that PEDF may be involved in mediating the effect of obesity on metabolic dysfunction because its production varies with energy balance in humans, and it was shown to regulate insulin action in an animal model.
To date, all of the data linking PEDF to metabolic health in humans is from studies of adults, and those studies showed the most striking differences in PEDF concentration are between normal-weight healthy people compared with people with diabetes or obesity (4, 5, 7, 9). To our knowledge the concentration of PEDF in children and young adults has not been reported. If the circulating PEDF level is positively related to body fat in otherwise healthy children and young adults as it is in middle-age and older adults, this finding would support the potential role of PEDF in development of metabolic disease and would merit further investigation. The primary goal of the current investigation, therefore, was to test the hypothesis that circulating PEDF is positively associated with body fatness and insulin resistance in children and young adults. We investigated whether circulating PEDF differs with age, sex, and body composition in young people without diabetes or overt cardiovascular disease. We also assessed whether PEDF could be used as an early marker of future disease risk by measuring the strength of association of PEDF with clinical outcomes, including insulin resistance, blood lipids, C-reactive protein, blood pressure, and arterial compliance.
Materials and Methods
Participants
Data compiled from 196 children and young adults aged 8–35 yr were included in the current analyses. Participants did not have diabetes or other metabolic, endocrine, or cardiovascular health conditions and were not using medications that would interfere with study outcomes. Participants were classified as either normal weight (NW) or overweight/obese (OW) using BMI criteria as described in the Supplemental Methods, published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org.
Protocol and procedures
Informed written consent and assent were obtained in accordance with the University of Oklahoma Health Sciences Center Institutional Review Board. Height and weight were measured, and BMI was calculated (Supplemental Methods). After a medical examination and history, body composition was measured using dual-energy x-ray absorptiometry, and blood pressure, arterial compliance, and analysis of fasting plasma or serum were performed as described in the Supplemental Methods.
Statistical analyses
The data are presented as mean ± se. Comparisons between sexes and obesity categories were conducted using a Student's t test or a Mann-Whitney U test. Comparisons between age groups were performed using ANOVA or Kruskal-Wallis tests and Spearman's correlation coefficients were used to conduct correlations. A detailed description of the statistical methods is provided in Supplemental Methods.
Results
Participant anthropometric, biochemical, and cardiovascular outcomes are shown in Table 1. PEDF was 60% higher in the OW group compared with the NW group (P < 0.001). All anthropometric, biochemical, and cardiovascular variables except age and height differed between the NW and OW groups, with less favorable cardiometabolic factors in the OW group (Table 1).
Table 1.
Body composition and markers of cardiometabolic disease risk in normal-weight and overweight participants
| Normal weight | Overweight | |
|---|---|---|
| Age | 15.5 ± 0.6 | 16.1 ± 0.5 |
| Sex (male/female) | 50/41 | 57/48 |
| Anthropometric measures | ||
| Height (cm)a | 160 ± 1 | 164 ± 1 |
| Body mass (kg)a | 52.5 ± 1.5 | 84.4 ± 2.0 |
| BMI (kg/m2)a | 20.1 ± 0.3 | 31.0 ± 0.5 |
| Fat mass (kg)a | 13.7 ± 0.6 | 35.0 ± 1.3 |
| Lean mass (kg) | 35.1 ± 1.2 | 45.0 ± 1.1 |
| Trunk fat mass (kg)a | 6.0 ± 0.3 | 18.1 ± 0.8 |
| Appendicular lean mass (kg) | 16.2 ± 0.6 | 21.9 ± 0.6 |
| Plasma/serum biochemical measures | ||
| PEDF (ng/ml)a | 3714 ± 186 | 5945 ± 344 |
| Glucose (mg/dl) | 84 ± 1 | 86 ± 1 |
| Insulin (μU/ml)a | 6.9 ± 0.6 | 17.7 ± 1.1 |
| Triglycerides (mg/dl)a | 71 ± 3 | 103 ± 5 |
| Total cholesterol (mg/dl)a | 155 ± 3 | 180 ± 6 |
| HDL cholesterol (mg/dl)a | 47 ± 1 | 44 ± 1 |
| LDL cholesterol (mg/dl)a | 94 ± 2 | 115 ± 5 |
| CRP (mg/liter)a | 0.8 ± 0.1 | 2.8 ± 0.4 |
| HOMA-IRa | 1.4 ± 0.1 | 3.8 ± 0.2 |
| Cardiovascular measures | ||
| Systolic blood pressure (mm Hg) | 109 ± 1 | 118 ± 1 |
| Diastolic blood pressure (mm Hg)a | 58 ± 1 | 61 ± 1 |
| LAEI (ml/mm Hg × 10)a | 15.5 ± 0.4 | 17.5 ± 0.6 |
| SAEI (ml/mm Hg × 100)a | 8.0 ± 0.3 | 9.6 ± 0.3 |
Data are mean ± se. All between-group comparisons were significant (P < 0.05) with the exception of age and height. CRP, C-reactive protein; HOMA-IR, homeostasis model of assessment-insulin resistance; LAEI, large artery elastic index; SAEI, small artery elastic index.
P values from Mann-Whitney U test.
PEDF concentration was not significantly different between males and females (data not shown), so sex was not considered in subsequent analyses. To assess whether PEDF varied with age and interacted with the BMI category, participants were separated into age groups combining all of the adults (≥18 yr old) and distributing children 8–18 yr old into four similarly sized groups (i.e. quartiles). There was not an age group-by-BMI group interaction using a factorial ANOVA (P = 0.270); however, the main effect of BMI (PEDF was significantly higher in OW vs. NW) was evident in all age groups (Fig. 1A). A similar approach was used for homeostasis model assessment of insulin resistance (HOMA-IR) and revealed consistently higher values across age (Fig. 1B). Additionally, OW participants had greater total body and appendicular lean mass, and greater total body and trunk fat mass compared to their NW counterparts across all age groups (Supplemental Fig. 1).
Fig. 1.
Increased plasma PEDF (A) and HOMA-IR (B) in overweight (closed bars) vs. normal weight (open bars) participants. Relationship between fat mass (FM; C), lean mass (LM; D), HOMA-IR (E), and serum PEDF concentration in normal weight (open circles) and overweight (closed circles) participants is shown. The regression lines and corresponding R2 values shown are for the combined groups. Values are expressed as mean ± se. *, Significant difference between overweight and normal weight people within the age group (P < 0.05). ALM, Appendicular lean mass; TFM, trunk fat mass.
To assess whether PEDF was associated with components of body composition, univariate correlations between PEDF and lean and fat mass was performed after adjusting the body composition outcomes for height (expressed as kilograms lean or fat mass per square meter and referred to as the lean or fat mass index). Within both the NW and OW groups, PEDF was significantly positively correlated with body mass, BMI, fat mass index, insulin, and HOMA-IR. In addition, PEDF was positively correlated with lean mass index, trunk fat mass index, appendicular lean mass index, C-reactive protein, and large artery elasticity index in the OW but not the NW group (Supplemental Table 1). PEDF was positively associated with all four components of body composition included in the multivariate modeling [total body lean mass and fat mass (Fig. 1, C and D), trunk fat mass, and appendicular lean mass (Supplemental Fig. 2)]. However, the strongest bivariate correlate of PEDF was HOMA-IR (Fig. 1E). This association was independent of BMI category and persisted (P < 0.001), even after adjustment for other plasma markers of metabolic syndrome (triglycerides, HDL, and fasting glucose) and inflammation (C-reactive protein). Like PEDF, HOMA-IR was positively correlated (P < 0.01) with trunk fat mass index (r = 0.683), total body fat mass index (r = 0.673), appendicular lean mass index (r = 0.451), and total lean mass index (r = 0.422).
Discussion
The current study is the first to our knowledge to report values for circulating PEDF in NW and OW children and young adults without overt metabolic or vascular disease. We found that PEDF was 60% higher in OW young people. Circulating PEDF level did not differ between sexes but was positively associated with both fat mass and lean mass, particularly within the OW group. PEDF was highest in children approximately 12–14 yr old and tended to decline (nonsignificant) with age. Additionally, circulating PEDF was strongly correlated with fasting insulin and HOMA-IR. These results are consistent with the proposal that PEDF may have a role in mediating, or at least serving as a marker of, obesity-related insulin resistance.
We found that PEDF concentration was positively associated with total body and abdominal fat mass in children and young adults. This finding is consistent with prior studies in adults that demonstrated that PEDF concentration was increased in obesity and declined after weight loss (8). The expanded adipose tissue volume in obese individuals results in changes in circulating adipokines that are posited to exert metabolic effects on other tissues (reviewed in Ref. 10); it now appears that PEDF is another candidate adipokine that plays a role in obesity-mediated insulin resistance. The current study demonstrates that PEDF is as highly abundant in children as it is in adults and may therefore exert similar effects across ages.
Body fat content is positively correlated with HOMA-IR in adolescents (11). Although many factors impact this association, PEDF was demonstrated as a plausible mediator of insulin resistance in mice and elevated in adults with hyperglycemia or impaired glucose tolerance (2, 6, 12, 13). Our data revealed that both PEDF and HOMA-IR are increased in OW vs. NW youth and are positively related to each other, independent of BMI category, with PEDF being the best predictor of HOMA-IR. Interestingly, insulin seemed to be the mediating factor in this relationship because PEDF and fasting glucose were not significantly correlated. A recent study revealed that treatment with insulin increased PEDF secretion from differentiated human primary adipocytes (3). Those results combined with the current and prior data suggest that elevated circulating PEDF in overweight people may result from the effects of hyperinsulinemia on adipose tissue and that PEDF in turn could promote insulin resistance in a feed-forward manner.
In addition to the liver and adipose tissue, recent human and cell culture studies demonstrate that skeletal muscle is also a source for circulating PEDF, particularly after contractions or exercise (14). Analysis of gene transcripts showed that PEDF mRNA was present and increased by 40 and 80% in the vastus lateralis and trapezius muscles, respectively, after 11 wk of strength training (14). Those data may also explain our observation of a significant positive correlation between plasma PEDF and both total body and appendicular lean mass, the latter of which is predominantly skeletal muscle. We found that this association persisted, even when controlling for fat mass, age, and sex. Because the OW participants had significantly greater appendicular lean mass, this could be another potential source of the elevated PEDF levels observed in the OW group. Collectively, the emerging data on PEDF suggest that it may act in ways that are similar to IL-6 or other novel myokines (15). The predominant role of IL-6 was long thought to be as a proinflammatory cytokine because its concentration is often elevated in humans who are obese or have related metabolic disorders (16, 17). However, it is now recognized that IL-6 is produced by skeletal muscles during exercise, released into circulation, and acts as a signal to organs like the liver to coordinate metabolism (18). Similarly, irisin was recently shown to be produced by muscle and communicate with adipose cells (19). Whether this corollary applies to PEDF is not yet clear.
A potential limitation of the current study is that we did not identify a mechanism linking PEDF with insulin resistance. Nevertheless, the association between PEDF and HOMA-IR provides a foundation for future research to investigate the mechanisms through which PEDF may result in insulin resistance. Strengths of our study include its relatively large sample size, the detailed body composition measures, and the novel evaluation of circulating PEDF in a young generally healthy cohort.
In conclusion, the current study demonstrates that circulating PEDF is increased in OW children and young adults compared with their NW peers. We also found the PEDF is positively correlated with several cardiometabolic risk factors (BMI, fat mass, fasting insulin and HOMA-IR) and show for the first time that PEDF is positively correlated with lean body mass independent of fat mass. The later finding corroborates another recent study (14) and may reveal an understudied source of circulating PEDF. Thus, the current data support a role for PEDF as a potential mediator, or at least a marker, of insulin resistance in children and young adults.
Acknowledgments
This work was supported by National Institutes of Health Grant P20 RR024215 from the Center of Biomedical Research Excellence Program of the National Center for Research Resources, the Oklahoma Center for Advancement of Science and Technology, the Endocrine Fellows Foundation Marilyn Fishman Grant for Diabetes Research, and the Lawson Wilkins Pediatric Endocrine Society Clinical Scholars Award.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- BMI
- Body mass index
- HDL
- high-density lipoprotein
- HOMA-IR
- homeostasis model assessment of insulin resistance
- LDL
- low-density lipoprotein
- NW
- normal weight
- OW
- overweight
- PEDF
- pigment epithelium-derived factor.
References
- 1. Matsumoto K, Ishikawa H, Nishimura D, Hamasaki K, Nakao K, Eguchi K. 2004. Antiangiogenic property of pigment epithelium derived factor in hepatocellular carcinoma. Hepatology 40:252–259 [DOI] [PubMed] [Google Scholar]
- 2. Crowe S, Wu LE, Economou C, Turpin SM, Matzaris M, Hoehn KL, Hevener AL, James DE, Duh EJ, Watt MJ. 2009. Pigment epithelium-derived factor contributes to insulin resistance in obesity. Cell Metab 10:40–47 [DOI] [PubMed] [Google Scholar]
- 3. Famulla S, Lamers D, Hartwig S, Passlack W, Horrighs A, Cramer A, Lehr S, Sell H, Eckel J. 2011. Pigment epithelium-derived factor (PEDF) is one of the most abundant proteins secreted by human adipocytes and induces insulin resistance and inflammatory signaling in muscle and fat cells. Int J Obes (Lond) 35:762–772 [DOI] [PubMed] [Google Scholar]
- 4. Jenkins A, Zhang SX, Gosmanova A, Aston C, Dashti A, Baker MZ, Lyons T, Ma JX. 2008. Increased serum pigment epithelium derived factor levels in type 2 diabetes patients. Diabetes Res Clin Pract 82:e5–e7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Jenkins AJ, Zhang SX, Rowley KG, Karschimkus CS, Nelson CL, Chung JS, O'Neal DN, Januszewski AS, Croft KD, Mori TA, Dragicevic G, Harper CA, Best JD, Lyons TJ, Ma JX. 2007. Increased serum pigment epithelium-derived factor is associated with microvascular complications, vascular stiffness and inflammation in type 1 diabetes. Diabet Med 24:1345–1351 [DOI] [PubMed] [Google Scholar]
- 6. Sabater M, Moreno-Navarrete JM, Ortega FJ, Pardo G, Salvador J, Ricart W, Frühbeck G, Fernández-Real JM. 2010. Circulating pigment epithelium-derived factor levels are associated with insulin resistance and decrease after weight loss. J Clin Endocrinol Metab 95:4720–4728 [DOI] [PubMed] [Google Scholar]
- 7. Yamagishi S, Adachi H, Abe A, Yashiro T, Enomoto M, Furuki K, Hino A, Jinnouchi Y, Takenaka K, Matsui T, Nakamura K, Imaizumi T. 2006. Elevated serum levels of pigment epithelium-derived factor in the metabolic syndrome. J Clin Endocrinol Metab 91:2447–2450 [DOI] [PubMed] [Google Scholar]
- 8. Franck N, Gummesson A, Jernås M, Glad C, Svensson PA, Guillot G, Rudemo M, Nyström FH, Carlsson LM, Olsson B. 2011. Identification of adipocyte genes regulated by caloric intake. J Clin Endocrinol Metab 96:E413–E418 [DOI] [PubMed] [Google Scholar]
- 9. Wang P, Smit E, Brouwers MC, Goossens GH, van der Kallen CJ, van Greevenbroek MM, Mariman EC. 2008. Plasma pigment epithelium-derived factor is positively associated with obesity in Caucasian subjects, in particular with the visceral fat depot. Eur J Endocrinol 159:713–718 [DOI] [PubMed] [Google Scholar]
- 10. Kershaw EE, Flier JS. 2004. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 89:2548–2556 [DOI] [PubMed] [Google Scholar]
- 11. Tershakovec AM, Kuppler KM, Zemel BS, Katz L, Weinzimer S, Harty MP, Stallings VA. 2003. Body composition and metabolic factors in obese children and adolescents. Int J Obes Relat Metab Disord 27:19–24 [DOI] [PubMed] [Google Scholar]
- 12. Nakamura K, Yamagishi S, Adachi H, Matsui T, Kurita Y, Imaizumi T. 2010. Serum levels of pigment epithelium-derived factor (PEDF) are an independent determinant of insulin resistance in patients with essential hypertension. Int J Cardiol 143:96–98 [DOI] [PubMed] [Google Scholar]
- 13. Yang S, Li Q, Zhong L, Song Y, Tian B, Cheng Q, Qing H, Xia W, Luo M, Mei M. 2011. Serum pigment epithelium-derived factor is elevated in women with polycystic ovary syndrome and correlates with insulin resistance. J Clin Endocrinol Metab 96:831–836 [DOI] [PubMed] [Google Scholar]
- 14. Norheim F, Raastad T, Thiede B, Rustan AC, Drevon CA, Haugen F. 2011. Proteomic identification of secreted proteins from human skeletal muscle cells and expression in response to strength training. Am J Physiol Endocrinol Metab 301:E1013–E1021 [DOI] [PubMed] [Google Scholar]
- 15. Pedersen BK, Febbraio MA. 2012. Muscles, exercise and obesity: skeletal muscle as a secretory organ. Nat Rev Endocrinol 8:457–465 [DOI] [PubMed] [Google Scholar]
- 16. Pickup JC, Mattock MB, Chusney GD, Burt D. 1997. NIDDM as a disease of the innate immune system: association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 40:1286–1292 [DOI] [PubMed] [Google Scholar]
- 17. Mohamed-Ali V, Goodrick S, Rawesh A, Katz DR, Miles JM, Yudkin JS, Klein S, Coppack SW. 1997. Subcutaneous adipose tissue releases interleukin-6, but not tumor necrosis factor-α, in vivo. J Clin Endocrinol Metab 82:4196–4200 [DOI] [PubMed] [Google Scholar]
- 18. Steensberg A, van Hall G, Osada T, Sacchetti M, Saltin B, Klarlund Pedersen B. 2000. Production of interleukin-6 in contracting human skeletal muscles can account for the exercise-induced increase in plasma interleukin-6. J Physiol 529(Pt 1):237–242 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Boström P, Wu J, Jedrychowski MP, Korde A, Ye L, Lo JC, Rasbach KA, Boström EA, Choi JH, Long JZ, Kajimura S, Zingaretti MC, Vind BF, Tu H, Cinti S, Højlund K, Gygi SP, Spiegelman BM. 2012. A PGC1-α-dependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature 481:463–468 [DOI] [PMC free article] [PubMed] [Google Scholar]