Table of Contents

Contact Us

Head of Institute: Prof. Ido Braslavsky

Administrative manager: Rakefet Kalev

Office Address:
Institute of Biochemistry, Food Science and Nutrition,
Robert H. Smith Faculty of Agriculture, Food and Environment,
The Hebrew University of Jerusalem, 
Herzl 229, Rehovot 7610001, ISRAEL

Tel: +972 - (0)8-9489385
Fax: +972 - (0)8-9363208
Email Address:

Prof. Efrat Monsonego Ornan

Ultra-processed food targets bone quality via endochondral ossification

Ultra-processed foods are those food products which undergo repeated stages of processing and contain non-dietary ingredients.  Culturally these foods are generally known as “junk food” and are typically pre-packaged and sold for mass consumption.  The increasing prevalence of these products around the world have directly contributed to increased obesity as well as other mental and metabolic impacts on people across the globe- and of all ages.  Much of the world has experienced a trend towards increased consumption of these products known to be easily accessible, relatively inexpensive and ready to eat out of the package.  Children are known to be particularly large consumers of these products with estimates that as much as 70 percent of their caloric consumption comes from ultra-processed foods.  And while many studies have reflected on the overall negative impact of these products, limited research has focused on their direct developmental effects on children- particularly in their younger years.

Our paper provides the first comprehensive analysis for how these foods impact skeletal development.  The study surveyed lab rodents whose skeletons were in the post embryonic stages of growth. The rodents who were subjected to ultra-processed foods were shown to suffer from growth retardation and a direct impact on their bone strength in ways that were atypical for rodents of that age who would typically be healthy.  Under histological examination, the researchers detected high levels of cartilage buildup in the rodents’ growth plates, the “engine” of bone growth.

Consumption of an ultra-processed diet leads to growth retardation and to alterations in bone architecture and biomechanical properties. The control group, which received a standard diet for growing rats, was compared to the UPF + CSD group that received a diet based on UPF and a caloric soft drink. a Body weight. b Total length from nose to tail. c Femur length at 6 and 9 weeks of age was measured using SkyScan software. d Daily caloric intake (kcal·d−1 per rat). eo Femur µCT analyses at 9 weeks of age. eh Trabecular parameters: bone volume fraction (BV/TV), trabecular number (Tb. N), trabecular thickness (Tb. Th), and trabecular separation (Tb.Sp). il Cortical parameters: cortical area fraction (Ct.Ar/Tt. Ar), average cortical thickness (Ct. Th), medullary area (Ma. Ar) and bone mineral density (BMD). mo Bone porosity parameters: cortical porosity (Ct. Po), pore number (Po. N) and total pore volume (Po. V). p Light microscopy images of cross-sections of rat cortical bone representing cortical bone porosity. q Biomechanical properties: stiffness (N·mm–1), yield (N), fracture load (N), max load (N) and energy to fracture (N·mm), assessed by three-point bending test. r Representative three-point bending load-displacement curves of the control and UPF + CSD bones. s Representative 3D images of cross-sections of femur bone visualized by Amira software, with arrows indicating cortical thickness. Values are expressed as the mean ± SD, n = 8. *P < 0.05 compared to control

RNA sequence analysis of the growth plates demonstrated an imbalance in extracellular matrix formation and degradation and impairment of proliferation, differentiation and mineralization processes.

Consumption of an ultra-processed diet modifies the growth plate (GP) transcriptome. a Pie chart representing 302 upregulated (pink) and downregulated (green) genes [ultra-processed food (UPF) vs. control (CN) groups]. b Heatmap representing the expression signature of differentially expressed genes clustered according to different skeletal pathways. c Heatmap representing the expression signature of differentially expressed genes clustered according to general pathways. d Pie chart representing the frequency of differentially expressed genes in the mouse genome using MGI phenotype analysis. 

The next stage sought to investigate what specific ingredients in ultra-processed foods are suspected for impeding proper growth with the objective of determining whether increased fat and carbohydrate content are the damaging factors or perhaps decreased levels of calcium, phosphorus. While these decreased levels likely have negative impact on overall nutrition, they are not the main factors for explaining the highly damaging effect of these foods on the developing skeleton.

Ultra-processed food (UPF) macronutrients (fats and sugar) are not responsible for alterations in bone architecture and biomechanical properties. The control and UPF groups were compared to a group receiving a high-fat diet based on the addition of corn oil (Corn) and a group fed a high-sucrose diet including a control diet + drinking solution with 10% sucrose (sucrose). ahFemur µCT analyses. ad Trabecular parameters: bone volume fraction (BV/TV), trabecular number (Tb. N), trabecular thickness (Tb. Th), and trabecular separation (Tb.Sp). eh Cortical bone parameters: cortical area fraction (Ct.Ar/Tt. Ar), average cortical thickness (Ct. Th), medullary area (Ma. Ar) and bone mineral density (BMD). i Growth plates (GPs) from the control, UPF + CSD, corn and sucrose groups were stained with safranin-O. j Quantification of the relative ratio of the zones in the GP. kRepresentative 3D images of transverse sections of femoral trabecular and cortical areas visualized by Amira software. Values are expressed as the mean ± SD, n = 8. Different letters denote significant differences at P < 0.05 between groups

We then sought to analyze how specific eating habits might impact on people who consume these foods. Since 70 percent of a child’s food consumption comes from these foods, we worked to replicate this type of food intake.  We divided the rodents weekly nutritional intake where thirty percent was from a ‘controlled’ diet and the remaining 70 percent consisted of ultra-processed foods. We found that the rodents experienced moderate damage to their bone density but lesser indications of cartilage buildup in the growth plates.  Our conclusion was that even with a more limited intake, ultra-processed foods have a definite negative impact on skeletal growth.

The study serves as the first evidence of Ultra processed food direct impact on rodent skeleton development during the key stages of early growth and are believed to be even more damaging that the previous metabolic damage which is known to be associated with these products.  These findings are critical because we know that children and adolescents consume these foods on a very regular basis to the extent that 50 percent of children in the US are ingesting them each and every day. Professor Carlos Monteiro, one of the world’s leading experts on nutrition has said that there is no such thing as a healthy ultra-processed food and he’s clearly right.  Even if we reduce fats, carbs nitrates and other known harmful substances, these foods still possess their damaging attributes.  Every part of the body is prone to this damage and certainly those systems that remain in the critical stages of development.



The reason that vertebrates do not look like, say, jellyfish is their skeletal system, which supports their body and protects vital organs while enabling efficient mobility. These functions depend on a precise and optimally-designed overall body proportioning determined by the size of the skeleton, which in turn is determined by the rate and duration of bone growth. Longitudinal bone growth occurs via the chondrocytes at the growth plates (GP), through a very tightly regulated process of endochondral ossification. The chondrocytes differentiate and produce cartilage anlagen (endochondral) that is sequentially calcified, degraded and replaced by bone (ossification). When the system is functioning correctly, our legs are of equal length, as are our arms. However, disrupted regulation is manifested in skeletal diseases, malformations and height disarrays.

My interest in GP metabolism began while working towards my PhD, when my research dealt with the effect of growth hormone (GH) on chondrocyte proliferation and differentiation. After completing my PhD, I continued my training at the Weizmann Institute of Science by joining the effort to understand the basis for inherited human skeletal disorders caused by mutation in FGF receptor 3 (FGFR3) leading to Acondroplasia, the most common form of human genetic dwarfism. In October 2006, I joined the Institute of Biochemistry and Nutrition at the Faculty of Agriculture of the Hebrew University.

During these years, we (see the list of the lab, young talented students) uncovered new roles for FGF in chondrocytes and in mutant mouse models. We described the interactions between the cytokine receptor family, and FGFR signaling pathway. I also initiated a new topic of research into the role of matrix metalloproteases (MMPs) in skeleton development. The MMPs are a family of enzymes that are involved in proteolytic cleavage of matrix components. These enzymes can be modulated by diet (Ca, Zn, Cu) and are subject to complex transcriptional regulation, indicating a genetic-nutritional interrelation during bone development. We showed, for the first time, the perivascular localization of cMMP9 and proved its role in vascularization of the avian GP. This project was conducted by several students in my lab and generated several publications, grants, and synergistic international collaborations. In collaboration with Dr. Dalit Sela-Donenfeld, we questioned whether by degrading the extracellular matrix, the MMPs mediate delamination and migration of neural crest cells from the neural tube.


The main MMPs involved in bone resorption and endochondral ossification. During bone resorption MMP-9 and MT1-MMP are secreted by osteoclasts. However, these enzymes do not participate in matrix resorption per se but rather in osteoclasts migration and invasion. In contrast, MMP-13 is secreted from peri-osteoclastic osteocytes, translocate from the osteocytes into the resorption site where it participate in resorption. After osteoclast withdrawal, bone lining cells enter the lacuna and secrete MMPs (probably MMP-2, 13, and MT1-MMP) which remove demineralized collagen that was not digested by the osteoclasts. This “cleaning” is prerequisite for the subsequent deposition of a first layer of collagenous proteins in the resorption pit. MMP-9 and MMP-13 are the most important MMPs in endochondral ossification. MMP-13 is secreted from hypertrophic chondrocytes and MMP-9 from osteoclasts and endothelial cells

MMPs expression in NCC (A-B) Immunolabeling of (A) transverse section obtained from embryos of 25 somites (stage 15) and (B) neural crest migrated out from a neural tube (6-8 somites, stages 8-9) explants, using MMP9 (green) and HNK1 (red) antibodies. Blue staining represents DAPI staining. White arrowheads indicate MMP9 and HNK1 positive cells.  

A continuing focus of interest of my lab is the effect of mechanical loading on the process of bone development. This is a multifaceted project, combining an in-vivo approach, in which we apply external loads on young fast-growing bones, with cell biology methods to understand the transduction of mechano-signaling in chondrocytes. Using various validation techniques, we identified load-dependent modifications in the major signaling pathways in the growth plate. These modifications reduce the number of chondrocytes in the GP, and therefore the latter's width and the final length of the bone. The possible involvement of the cilia in transducing mechano-signaling was further studied in tissue cultured chondrocytes by confocal microscopy and gene knock-down methods combined with loading protocols of cells in silicon chambers. This project involve collaboration with Prof. Nicolai Miosge dealing with the role of mechanical forces in tissue engineering of cartilage and mesenchymal cells. Within this research we also studied the effects of loading on bone phenotype in terms of architecture and mechanical properties. This research is carried out in collaboration with Prof. Ron Shahar, and has enabled us to purchase a micro-CT that significantly expands our scientific capabilities, combining our diverse expertise and joint interest in bone biology.

(A) Electron micrograph of the primary cilium (Ax), the distal (Dc), and proximal (Pc) centriole. Bar 500 nm. (B) Cross section of the proximal centriole. Bar 100 nm. (C) schematic presentation of the primary cilium with membrane signaling molecules that were described in it. (D) human articular chondrocytes and (F) CPC stained with acetylated α-tubulin (tb,green) Abs to detect primary cilium (arrows), phalloidin (ph,red) and DAPI (DAPI,blue). (E) Tissue from late stage of human OA exhibits surface fissures and cell clusters. Breaks in the tidemark are filled with blood vessels, and the bone marrow is visible underneath the OA tissue.

The broad aim of my research is to understand the role of nutritional and metabolic factors in bone development. The connection between under-nutrition and growth retardation is well documented. A trend of increasing height is recognized in countries where proper nutrition, exercise and fitness are more equally distributed. Children from developing countries with Protein-energy malnutrition (PEM) had significantly lower body weight and height than healthy subjects. Furthermore, animal studies clearly demonstrated the deleterious effects of PEM on linear growth. But what is the reciprocal? What if you eat too much? Are you going to over grow? Yes, it is recognized that overweight children are taller than age-matched, normal-weight children. This tall stature is often associated with an accelerated skeletal maturation and an earlier timing of puberty, but not with the final adult height. Thus it is clear that factors such as vitamin D and calcium are just the tip of the iceberg, and the role of nutritional factors in bone development has yet to be elucidated.

In the last few years, the connection between an organism's metabolic status and mature skeletal homeostasis has revolutionized the field of bone research, revealing the effect of leptin on bone density, and suggesting an endocrine loop in which a bone-specific protein (osteocalcin) affects insulin secretion and sensitivity. Despite these advances, other predictable link between organism's metabolic status and skeletal development has not been addressed. My research vision and activities aim to fill this gap. In this line, the projects we are conducting in the lab are dealing with:

The effect of different fatty acids on bone development. We have shown the role of W-3 in the processes of bone modeling and remodeling, and its positive effects in the process of young bone development. For this project, in collaboration with Prof. Betty Schwartz, we are using the fat-1 transgenic mice, which are genetically modified to endogenously synthesize W-3 PUFA. We further studying the differential effects of high fat diets based on SFA, MUFA or PUFA on bone elongation and quality, and on the metabolic consequences of these diets. Our results suggest that these fatty acids are not only energy sources but may serve as signaling molecules in bone cells.


The effects of cytokines secreted from the adipose tissue on bone development and properties. In this project we are studying the effect of fat-derived cytokines on skeletal phenotype in-vivo by local or peripheral injections and by using mice models and in-vitro by chondrocytic or osteoblastic cell lines. By this combined approach we will get a new insight to the cross-talk between the adipose the cartilaginous and bone tissues during developmental stages. So far we have studied the role of leptin and adiponectin on embryonic bone development, and initiated an in-vivo model of obesity and metabolic syndrome during the early postnatal period by blocking the signaling of leptin.

IL-1RI k.o mice have narrower growth plates and higher proteoglycan content. Tibiae from C57BL/6 and IL1-RI k.o mice were subjected to histological analysis. (A) Masson’s Trichrome staining of the growth plate. Arrows showing the proliferative zone (PZ) and hypertrophic zone (HZ). (B) Length of whole growth plates (GP), proliferative (PZ), and hypertrophic (HZ) zones were measured. (C) Safranin O stains proteoglycans in red/orange. (D,E)  In-situ Hybridization analyses with probes for (D) collagen type II (E) collagen type X.

In a fascinating project we compared the plasticity of different mesenchymal cells in terms of differentiation and mineralization. Taking advantage of our expertise in chondrogenesis we were able to demonstrate that vascular smooth muscle cells undergo similar differentiation routs which results in the pathological condition of vascular calcification. We conducted a series of experiments in young rats (during their growth period) inducing arteries calcification by calcitriol injections. Using this model, and tissue culture approaches, we demonstrated the involvement of glucose metabolism in the physiological (in cartilage) and pathological (in vasculature) calcification. We further elucidated the underlying mechanisms regulating these events (through BGP and HIF-1α) by knocking down these genes in-vitro and in-vivo, using the Si-RNA approach.

Proposed mechanism of action of BGP on glucose metabolism and mineralization.    Vitamin D stimulates bone BGP production, which diffuses into the neighboring growth-plate and is released into the circulation.    At the level of chondrocyte and VSMC, BGP stabilizes the hypoxia sensitive HIF-1α subunit leading to an increased rate of glucose uptake.   Inside the cell, via HIF-1 stabilization, BGP channels glucose into the glycolytic pathway by increasing the expression of key enzymes PFK and PDK1, switching off the mitochondrial oxidative phosphorylation and stimulating less efficient, but much faster conversion of glucose into energy.  Concomitant with elevation of glycolysis, BGP suppresses the expression of key gluconeogenesis enzymes, G6Pase and PEPCK.  Altered metabolic status translates into augmented chondro-osteogenic differentiation state, contributing to the development of cartilage and vascular calcification.

In our most recent project we explore the effect of the western diet on bone development.  In recent years obesity has become a public health crisis in the western society has tripled. As children's body weights have increased, so has their consumption of fast foods and soft drinks. The proportion of foods that children consume from restaurants and fast food outlets increased by nearly 300% between 1977 and 1996. Children’s soft drink consumption has also increased during those years. Junk foods, fast foods and snack foods are high in fat and refined sugars and don’t provide appropriate levels of fiber, protein, vitamins and minerals for the growth requirement of growing children. Thus, beside obesity, diet based on junk food may cause malnutrition. We examine the influence of childhood obesity and malnutrition, caused by unhealthy and unbalanced diet on the development and quality of the post-natal bone while the main focus is on the growth plates, as well as on the metabolic state of the body. We use young rats as our experimental model, and discovered that rats fed junk food diet, had impairments in bone growth and structure, which could result from the deficiency of nutrients in the junk food diet.

            The metabolic regulation of skeletal development is of great importance from both basic and clinical points of view. From a practical standpoint, these concerns are related to the dramatic rise in childhood obesity, which is now reaching epidemic proportions. From a more basic standpoint, the fact that adipocytes, chondrocytes and osteoblasts all originate from the same progenitors (mesenchymal stem cells) and are able to trans-differentiate into one another, raises enthralling scientific possibilities. Just imagine that by manipulating these progenitor cells, ideally via nutritional or mechanical strategies, we might shift their programmed differentiation from fat-forming to bone-forming cells: if this vision would become reality, we could help solve two major health concerns in the world today, obesity and osteoporosis.