Association of endothelial proliferation with the magnitude of weight loss during calorie restriction.

Korybalska K1 Swora-Cwynar E2, Łuczak J1, Kanikowska A2, Czepulis N1, Rutkowski R1, Bręborowicz A1, Grzymisławski M2 and Witowski J1

1 Department of Pathophysiology and 2 Department of Internal Medicine, Metabolic Diseases, and Dietetics; Poznań University of Medical Sciences, Poznań, Poland



Objectives Substantial weight loss through intense dietary regimens is thought to ameliorate endothelial dysfunction in obesity. It is less clear whether similar improvements can be achieved with modest dietary interventions. This study aimed to identify the parameters of endothelial cell status in obesity that are affected by mild calorie restriction.

Methods Human umbilical vein endothelial cells (EA.hy926line) in culture were exposed pairwise to serum from 57 individuals with simple obesity (BMI[30 kg/m2) collected before and after 8-week dietary intervention with energy deficit of 300–500 kcal/day.

Results Analysis of endothelial transcriptome suggested that the intervention could impact on endothelial cell growth. Cell proliferation was measured with the MTT test and verified by [3H]-thymidine incorporation. The participants were categorized according to a change in proliferation over time. Significant decrease in endothelial cell proliferation correlated with the extent of weight loss in men, but not in women. This effect corresponded with changes in serum levels of leptin and adiponectin, but was not related to serum concentrations of several known angiogenic mediators (VEGF, MCP-1, TSP-1, MMP-9, angiopoietin-2).

Conclusion Direction and magnitude of changes in serum induced endothelial cell proliferation identifies patients with the greatest weight loss in response to modest calorie restriction. 

KEYWORDS: Cell proliferation; Endothelium; Obesity; Thrifty phenotype; Weight loss

PMID: 27245991



The growing prevalence of obesity is a well-recognized health problem worldwide. Many detrimental effects associated with obesity are thought to be related to endothelial cell dysfunction (Fig. 1). However, the exact pathophysiological context that impairs endothelial homeostasis in obesity is only partly understood. It is not always clear whether the effects observed in endothelial cells represent adaptive or maladaptive responses to obesity-associated metabolic signals. The complexity of the problem is further exacerbated by frequent co-existence of other diseases, such as diabetes, dyslipidemia, and hypertension, that have a clear negative impact on endothelial cells. Also, the exact mechanisms underlying favorable effects of calorie restriction remain partially obscured. In this respect, it is a matter of debate whether the benefits of dietary regimens toward endothelial cells are related primarily to a negative energy balance or rather to a decrease in body fat mass.

The lack of a universal biomarker of endothelial cell function adds to difficulties in untangling endothelial pathophysiology in obesity. The markers used in clinical practice and research include parameters as different as flow-mediated vasodilation (assessed by ultrasound) or the release of endothelial cell-derived molecules (measured by specific immunoassays). Although these parameters appear to improve in response to significant weight loss by drastic dietary restriction or bariatric surgery, they show less convincing effects and large variability following mild-to-moderate calorie restriction. This type o intervention is of particular practical relevance as most individuals attempting to lose weight would find it difficult to sustain a very low-calorie diet.

Here, we created an in vitro model that recapitulated the exposure of vascular endothelial cells to hypocaloric environment. We hypothesized that it would enable us to identify new parameters of endothelial cell function that could be useful for predicting and/or monitoring the response to mild dietary intervention. To this end, we obtained samples of serum from individuals with obesity (BMI>30 kg/m2) but without other apparent diseases, who were undergoing moderate calorie restriction. The intervention lasted 8 weeks and conveyed a predicted energy deficit of 300-500 kcal/d. Samples of serum were drawn under standardized conditions before and after the intervention and then added to endothelial cells in culture (Fig.2). The maintenance of cells in vitro usually requires the addition of some serum (typically from calf fetuses). Here, we supplemented culture medium with 20% patient’s serum. The cells used were EA.hy926, which are human umbilical vein endothelial cells (HUVEC) from a well-characterized permanent cell line. The use of a reference cell line eliminated some variability in cell responses that would be associated with the use of primary cultures from different donors. Thus, the only significant variable left was a change in serum properties as a result of dietary intervention.


As a first step we looked into global gene expression profiles in endothelial cells assessed whether the intervention had any consistent impact on gene expression in the endothelium. We were surprised to find that not a single gene emerged as differently and consistently regulated. We have therefore analysed the data using a gene set enrichment analysis (GSEA), which determines whether the expression of a set of genes (involved in the same biological pathway and/or acting together) rather than of a single gene differs between two biological states. We reasoned that even a small change in the expression of a single gene (in fact, too small to qualify it  formally as significant), might in fact bear biological importance if occurring together with an equally small change in all genes of a given pathway. Indeed, GSEA showed some consistent patterns in the expression of genes involved in the regulation of cell cycle and proliferation.

Therefore, we went on to compare endothelial cell growth in the presence of sera drawn before and after the intervention. This analysis revealed that – for the whole population of our test subjects – the average cell proliferation capacity rose by slightly more than 10%. A similar increase has previously been observed in endothelial cells treated with sera from rhesus monkey (Macaca mulatta) subjected to a long-term 30% calorie restriction and interpreted as an evidence of improved cell function (J Gerontol A Biol Sci Med Sci 2013;68:235-249).

When we had a closer look at individual data, however, it became clear that the effect was not the same in all participants. While in the majority of them the intervention did not change or increased the ability of serum to stimulate cell proliferation, there was a smaller group of individuals in whom the dietary restriction led to a reduction in cell proliferation (by more than 10%). Curiously, there was proportionally twice as many men than women in this group. Even more strikingly, there was a significant correlation in men, but not in women, between the degree of change in endothelial cell proliferation and the magnitude of weight loss. It meant that men whose endothelial cell proliferation declined over time lost significantly more weight than the others.

We hypothesized that in some individuals (men rather than women) the dietary restriction led to such a change in their humoral milieu, which promoted simultaneously a more extensive weight loss and a decline in cell proliferation. As angiogenesis and adipose tissue expansion appear to be intimately linked, we checked the levels of known modulators o angiogenesis, such as VEGF. It turned out, however, that the intervention did not significantly change serum concentrations of those. We have then looked at leptin and adiponectin, as these adipokines can also modulate endothelial cell proliferation and their levels typically differ between the sexes (ref).Indeed, we found that changes in endothelial cell proliferation correlated with changes in the relative levels of leptin and adiponectin and with the magnitude of weight loss. As a result, the greatest decrease in weight occurred in men whose sera drawn after the intervention exhibited the most substantial changes in leptin and adiponectin and had a decreased potential to stimulate endothelial cell growth.

One can look at these results as the cellular manifestation of how different metabolic phenotypes impact on the response to calorie restriction or fasting. As cell proliferation is an energy-consuming process, its magnitude depends on how much resources could be spent on it. Thus, individuals with a ‘‘thrifty’’ phenotype (more likely women, probably due to their different hormonal make-up) tend to conserve energy, so that they lose less weight during calorie restriction but can allocate more energy for cell proliferation. By contrast, individuals with a ‘‘spendthrift’’ phenotype lose weight more easily but have less energy to spend on cell proliferation (Fig. 3).

The importance of this study is to find a new endothelial cell biomarker, which is sensitive enough to distinguish the effect of caloric restriction measured by the magnitude of weight loss. The limitation of this marker is that serum-induced endothelial cell growth is sex-dependent. Changes in cell proliferation identify the magnitude of weight loss after the intervention only in men, rather than in women. There are some data from other studies that could support this hypothesis (Diabetes 2015;64:2859-2867). However, the exact mechanisms that determine and govern the metabolic phenotype in either sexes remain to be fully elucidated.



  1. Rosen ED Spiegelman BM (2014) What we talk about when we talk about fat. Cell 156:20-44
  2. Avogaro A de Kreutzenberg SV (2005) Mechanisms of endothelial dysfunction in obesity. Clin Chim Acta 360:9-26
  3. Reinhardt M, Thearle MS, Ibrahim M, Hohenadel MG, Bogardus C, Krakoff J, Votruba SB (2015) A human thrifty phenotype associated with less weight loss during caloric restriction. Diabetes 64:2859-2867



This study was supported by National Science Centre Poland (grant no. N N404 151340) Królewska 57, 30-081 Cracow, Poland.