SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Objective

Human abdominal subcutaneous white adipose tissue (SAT) is composed of two different subcompartments: a “superficial” SAT (SSAT), located between the skin and a fibrous-fascia plane; and a deeper SAT, located under this fibrous fascia plane, indicated as “deep” SAT (DSAT).

Design and Methods

In order to investigate whether SSAT and DSAT have different molecular and morphological features, paired SSAT/DSAT biopsies were collected from 10 female obese patients and used for microarray and morphologic analysis. The stroma-vascular fraction cells were also isolated from both depots and cultured in vitro to assess the lipid accumulation rate.

Results

SSAT and DSAT displayed different patterns of gene expression, mainly for metabolic and inflammatory genes, respectively. Detailed gene expression analysis indicated that several metabolic genes, including adiponectin, are preferentially expressed in SSAT, whereas inflammatory genes are over-expressed in DSAT. Despite a similar lipid accumulation rate in vitro, in vivo SSAT showed a significant adipocyte hypertrophy together with a significantly lower inflammatory infiltration and vascular vessel lumen mean size, when compared to DSAT.

Conclusions

These data show that, SSAT and DSAT are functionally and morphologically different and emphasize the importance of considering independent these two adipose depots when investigating SAT biology and obesity complications.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The increasing evidence of the complexity of fat metabolic patterns emphasizes the importance of studying adipose tissue at anatomic, cellular and molecular levels. Compared to tissues with a confined localization in the body, the adipose tissue, or “adipose organ” as firstly defined by Cinti [1, 2], is composed by a multiple series of well-defined depots, mainly located at two corporal levels in mammals: superficial (or subcutaneous, composed by white adipocytes and, in the interscapular region, by brown adipocytes) and visceral compartments (mainly composed by white adipocytes) [3].

In humans, the subcutaneous adipose tissue (SAT) of the abdominal region is in continuity with the dermal tissue and it is usually considered as a unique, homogeneous fat depot. However, in this region SAT is composed by two anatomically different layers, located, respectively, above and below a connective plane named fascia superficialis, or Scarpa's fascia and usually indicated as superficial “thigh” SAT (SSAT) and subfascial, or deep, “thigh” adipose tissue (DSAT) [5, 6]. The fascia is clearly visible by magnetic resonance imaging and ultrasonography [7]. The layering of SAT was reconsidered after the advent of liposuction surgical techniques [3] and SAT subcompartments were analyzed in detail by CT scanning techniques, both in lean and obese subjects [4]. The SSAT is often very thin in humans but the relative amount is depending on body mass index, fat mass, gender, and ethnicity [9]. The DSAT is predominant posteriorly at the L4-L5 interspace and the anterior and posterior DSAT compartments are contiguous. The DSAT extends superiorly at least to the inferior aspect of the ribs and inferiorly in a contiguous fashion to the lower crease of the buttocks. In some individuals, the DSAT extends inferiorly into the upper leg region. At the level of the L4-L5 interspace, the posterior-lateral portion of the fascia is often penetrated by what appears to be a neurovascular bundle from the dorsal spine [7]. Furthermore, the amount of DSAT is highly correlated to fasting insulin, especially in men [5, 10]. An hypothesis for a “thermo-insulator role” of SSAT and a “metabolic-mode role” for DSAT has already been proposed, based on the studies on swine adipose tissue [11, 12] suggesting differences in lipogenic and lipolitic activities [13]. At gross anatomy level, SSAT appears macroscopically lamellar, i.e., with adipocytes organized in vertically oriented and closely spaced septa dividing the adipose tissue in regular tightly packed lobules, whereas DSAT is less defined and characterized by a loose areolar morphology, i.e., by adipocytes distributed in an irregular manner [9]. A different embryologic origin of cells composing the two subcutaneous adipose layers was suggested in swine, based on proliferation, enzymatic, fatty acid composition, and morphologic differences [5, 9, 13]. However, these data are not available for SSAT and DSAT adipocytes in humans.

In normal weight healthy subjects, we have previously shown an independent and unexpected metabolic role for DSAT [14, 15] and a recently published study provided evidence that, in morbidly obese patients, the degree of DSAT inflammation is related to the severity of liver disease (NASH and fibro inflammatory hepatic lesions) [16].

In this study, we employed microarray and morphological techniques to better investigate SSAT and DSAT in a cohort of severely obese patients, aiming to further identify their specific molecular and morphological differences.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Subjects and adipose tissue biopsies

The present study is in accordance with the Declaration of Helsinki and it was approved by the Ethical Committee of Istituto Auxologico Italiano. A signed informed consent was obtained from each enrolled patient. Surgical biopsies of whole abdominal SAT were collected per-operatively from 10 obese female patients (BMI: 37.5 ± 9.9 kg/m2; age: 48.2 ± 10.6 years; mean ± standard deviation), during dermolipectomy procedures, a postbariatric surgical removal of adipose tissue and excessive skin [17, 18]. Each collected biopsy was weighed and stored in 1 ml of DMEM (Invitrogen Corporation, Jefferson City, MO) supplemented with 2.5% Bovine Serum Albumin (BSA, Sigma, St. Louis, MO) per 1 g of tissue. The collected biopsy was immediately transferred to the laboratory and processed.

Tissue, cell isolation and culture conditions

The two layers composing SAT as well as the fibrous fascia were clearly visible (Supporting Information Figure 1). The skin was always present in the sample in order to properly orientate the biopsy and DSAT/SSAT paired samples were isolated and processed independently from each collected sample, for a total of 20 biopsies. A fragment of the whole adipose tissue biopsy was immediately frozen in liquid nitrogen for RNA extraction (see below), another fragment was formalin-fixed and the remaining material was digested with 1 mg/ml collagenase type 2 (Sigma, St. Louis, MO) for at least 1 h at 37°C under shaking. The digested tissue was then filtered through a sterile gauze and a nylon filter (BD Bioscience 1 Becton Drive Franklin Lakes, NJ). The stroma-vascular fraction (SVF) cells were isolated by centrifugation and then treated with a buffer containing 154 mM NH4Cl, 10 mM KHCO3, and 0.1 mM EDTA for lysis of red blood cells. SVF cells were plated and cultured in a medium containing a 1:1 mixture of Ham's F12/DMEM (Invitrogen Corporation, Jefferson City, MO) supplemented with 10% foetal bovine serum (Sigma, St. Louis, MO) until confluence. At confluence, cells were differentiated into mature adipocytes using NH-AdipoDiff medium (Miltenyi Biotec Bergisch Gladbach, Germany) for 10 days. Intracellular triglyceride storage levels were assessed by AdipoRed staining, according to the manufacturer protocol (Lonza, Milan, Italy).

image

Figure 1. Panel A: Heat map of significant genes in SSAT and DSAT (n = 6). Each row represents the expression profile of an mRNA across significant differentially expressed genes and each column represents a sample. Red and green colors, respectively, represent either higher or lower expression levels of the mRNA (median-centered). Panel B: Microarray average signal ratio for some “inflammatory” and “metabolic” genes in SSAT vs. DSAT (n = 6). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Isolation of RNA from SSAT/DSAT

Approximately 500 mg of frozen adipose tissue was homogenized in RLT buffer (Qiagen) and transferred into a 2-ml centrifuge tubes. RNA from the SSAT and DSAT was extracted using the RNeasy Mini Kit (Qiagen) according to the manufacturer protocol and samples were then treated with the RNase-Free DNase Set (Qiagen). Concentration and quality of the extracted RNA were determined by the NanoDropH ND-1000 spectrophotometer (NanoDrop Technologies, USA) and RNA integrity verified by gel-electrophoresis.

Gene expression analysis

A total of 12 samples, corresponding to paired SSAT/DSAT biopsies from six patients, were independently analyzed by microarray. Gene expression profile was evaluated using HumanHT-12 v3 BeadChips whole-genome hybridization assay (Illumina, San Diego, CA), which is based upon fluorescence detection of biotin-labeled cRNA. Each array contains full-length 50-mer probes representing more than 48,000 well-annotated RefSeq transcripts, including >25,400 unique, curated, and up-to-date genes derived from the National Center for Biotechnology Information Reference Sequence database. Initially, 300 ng of total RNA was converted to cDNA, followed by an in vitro transcription step to generate Biotin-16-UTP-labeled cRNA using the Ambion Illumina Total Prep RNA Amplification Kit (Ambion, Austin, TX) as per manufacturer's instructions. The labelled probes were mixed with hybridization reagents and hybridized overnight to the HumanHT-12 v3 BeadChips (Illumina, San Diego, CA). Following washing and staining, the BeadChips were imaged using the Illumina BeadArray Reader (Illumina, San Diego, CA) to measure fluorescence intensity for each probe. One BeadChip with 12 arrays was used. Slide was immediately scanned and first quality check, background correction, and normalization of the data were done by Beadstudio expression module v 3.2.7 (Illumina). Differential gene expression was assessed using Genome Studio software (Illumina). P-values were corrected for multiple testing using Benjamini and Hochberg False Discovery Rates. A differential expression score ≥13 was considered statistically significant (P < 0.05).

Heat map of gene expression profiles was obtained by GEDAS (Gene Expression Data Analysis Suite; http://sourceforge.net/projects/gedas) software, an extension of Cluster 3.0 (Eisen Lab's Cluster and Tree View software). The microarray data are MIAME compliant [19].

Functional annotations

Significant gene-lists, based on Illumina-IDs, were uploaded to DAVID (Database for Annotation, Visualization, and Integrated Discovery) [20, 21], analyzed with Ingenuity Pathway Analysis (IPA; Ingenuity Systems®, Mountain View, CA, http://www.ingenuity.com, www.ingenuity.com), and ToppGene suite open access-software [22]. The networks, reported in Supporting Information figures, were generated through the use of Ingenuity Pathways Analysis (Ingenuity Systems®, www.ingenuity.com).

The Functional Analysis of a network identified the biological functions that were most significant to the genes in the network. Fischer's exact test was used to calculate a P-value determining the probability that each biological function assigned to that network is because of the chance alone.

Validations of microarray data by real time quantitative PCR (RTqPCR)

The cDNAs were obtained by reverse-transcription with SuperScript III (Invitrogen, Carlsbad, CA) from 1 μg of total RNA for all collected samples (20 samples of paired SSAT/DSAT correspondingt to 10 enrolled patients). At room temperature, 1 μg of the total RNA was treated with DNaseI for 15 min. The reaction was incubated at 25°C for 5 min, 42°C for 30 min, and 85°C for 5 min, then stored at −20°C until use. We selected some differentially expressed genes underwent expression-level verification using RTqPCR. This was done using the 7300 Real Time PCR System with TaqMan probes (Applied Biosystems, Foster City, CA). Data were analyzed using the SDS v.1.4 software (Software Diversified Systems, Spring Lake Park, MN). Cycle threshold (Ct) was defined as the cycle number at which a significant increase in the fluorescence signal was firstly detected.

The mRNA levels were normalized to ribosomal protein large P0 (RPLP0) a very well suited, normalizing gene for mRNA quantification, producing a delta Ct value. The relative quantization was then obtained by the 2−ΔΔCt method and indicated by arbitrary units (AU).

Characterization of adipose tissue morphology

For morphological studies, adipose tissue samples (20 samples of paired SSAT/DSAT biopsies of the 10 enrolled patients) were fixed in 4% buffered formalin overnight at 4°C, washed and paraffin embedded. The size and density of adipocytes as well as blood vessels in the adipose tissues were determined by haematoxylin/eosin stained sections using Image-Pro Plus 4.0 software. For each sample, five randomly selected sectional areas, measuring at least 200 cell diameters, were analyzed and blood vessels counted and measured using an image analysis system (Leica QWin image analysis and processing). The estimation of macrophage number was performed as previously described [23]. In brief, 5-μm sections were mounted on glass slides, deparaffinized in xylol and stained for CD68 using anti-CD68 monoclonal mouse antihuman antibody (Dako, dilution 1:100), using standard immuno-histochemistry methods. Macrophages were identified in the adipose parenchyma (CD68 within blood vessels were excluded), when cytoplasmic staining for CD68 was present along with an identifiable mononuclear nucleus, and presented as the number per 100 adipocytes (% macrophages). Fibrosis was estimated by Masson's trichrome staining kit (Bioptica, Italy) and a score was assigned for 10 randomly observed fields (0, 1, 2 for a fibrosis area covering, respectively, the 5%, 15%, or > 25% of the observed field) and expressed as mean ± SD. Images were acquired using an optical microscope coupled with a digital camera (Leica DMR280) and were analyzed using the software Leica QWin image analysis and processing.

Statistical analyses

Data are shown as means ± standard errors of the mean unless stated otherwise. Statistical analysis was performed using SPSS version 12.0 (Chicago, IL) and Graph Pad (SAS). For genes analyzed by RTqPCR, significant differences were determined by Wilcoxon nonparametric paired test. The correlations between mRNA levels and microarray average signals of the different transcripts were examined by the nonparametric Spearman's rank correlation test. P-values < 0.05 were considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Superficial SAT showed a lamellar and dense structure, with adipocytes organized in parallel layers of a compact appearance. In contrast DSAT, in agreement with previous observations [9], showed a more loose and areolar appearance (Supporting Information Figure 1). The fibrous fascia superficialis was visible in all collected samples and was discarded during separation of SSAT and DSAT, as well as the skin layer.

Global gene expression in SSAT and DSAT

In order to identify differentially expressed genes between SSAT and DSAT, we performed a microarray analysis in paired SSAT/DSAT samples of six morbidly obese patients. A total of 37,800 genes, corresponding to 48,796 probes, were correctly detected after the scanning procedure and 14,415 genes had a significant average signal detection (P-value < 0.05). A total of 7422 genes had an average detection signal with a P-value < 0.001. Using Genome studio we performed differential expression analysis and obtained a list of 137 genes (corresponding to 347 probes) overexpressed in SSAT and 120 genes (corresponding to 250 probes) overexpressed in DSAT. The significant differentially expressed genes without an associated gene symbol were 17.5% in SSAT and 34.1% in DSAT. Significant genes were also analyzed by a Heat Map, a two-dimensional representation of gene expression data, in which each row represents the expression profile of mRNA across significant differentially expressed genes and each column represents a sample. A separated pattern of expression is clearly visible (Figure 1A).

Analyzing the SSAT/DSAT average signal (AVG) ratio fold-change, an over-expression of lipid-adipocyte specific genes in SSAT was observed, with an exception for serum amyloid protein genes SAA1, SAA2, and SAA4, acute phase inflammatory genes. adiponectin (ADIPOQ), adiponectin receptor 2 (ADIPOR2) and caveolin 2 (CAV2) were significantly overexpressed by SSAT, whereas leptin receptor gene (LEPR), apolipoprotein C1 (APOC1) adrenergic alpha 1B receptor (ADRAB1), adenosin A2a receptor (ADORA2A), inducible prostaglandin-endoperoxide synthase type 2 (PTGS2 or COX2), interleukin 1 receptor antagonist (IL1RN) genes were overexpressed by DSAT (Figure 1B).

Functional annotations of significant genes

Functional annotations of SSAT/DSAT overexpressed genes were based on Illumina ID using three different annotating tools (i.e., that are based on different algorithms and integrating systems): 1) the open source DAVID (the Database for Annotation, Visualization and Integrated Discovery), which provides a comprehensive set of functional annotations behind large list of genes, is widely used to understand if there are differentially regulated genes that fall into discrete functional groups ([20, 21];2) the IPA-Ingenuity systems (a tool that helps researchers to model, analyze and understand over-represented pathways of “omics” data helping to identify significant biological pathways by comparing the overlap of differentially expressed genes and all genes in the human genome with the gene sets associated with well-known pathways); and 3) the ToppGene Suite (ToppGene and TopFun integrate a vast number of genomic data from humans and can be used for the analysis of gene functional enrichment and for the prioritization of disease gene candidates, respectively, deriving the probability that each annotation is related to the gene in question) [22]. Based on these methods, the annotation “Lipid Metabolism” was the only one common to both SSAT and DSAT, despite a higher level of significance for SSAT. The genes preferentially expressed by SSAT were annotated as “Cytoplasm, Mitochondria, intracellular organelle” and as “Lipid metabolism, Small molecules Biochemistry, Molecular Transport" by DAVID annotation tool (Table 1) and as “Protein Trafficking, Carbohydrate Metabolism” by IPA-Ingenuity annotation system (Supporting Information Table 1). The SSAT genes prioritized by ToppGene suite were mainly “adipocyte-specific” genes, as reported in the Supporting Information table (Supporting Information Table 2, upper panel). The first gene identified by this system is the ADIPOQ (also well known as adiponectin) and it represents the gene that best describe the SSAT depot.

Table 1. SSAT annotation clusters
Annotation cluster 1Enrichment score: 1.88  
CategoryTermCountP-Value
GOTERM_BP_FATGO:0006006∼glucose metabolic process60.002
GOTERM_BP_FATGO:0019318∼hexose metabolic process60.006
GOTERM_BP_FATGO:0005996∼monosaccharide metabolic process60.012
GOTERM_BP_FATGO:0006091∼generation of precursor metabolites and energy50.013
    
Annotation cluster 2Enrichment score: 1.68  
CategoryTermCountP-Value
GOTERM_BP_FATGO:0030258∼lipid modification66.70E-05
GOTERM_BP_FATGO:0006635∼fatty acid beta-oxidation30.012
KEGG_PATHWAYhsa03320:PPAR signaling pathway40.016
GOTERM_BP_FATGO:0009062∼fatty acid catabolic process30.020
GOTERM_BP_FATGO:0019395∼fatty acid oxidation30.024
GOTERM_BP_FATGO:0034440∼lipid oxidation30.024
GOTERM_BP_FATGO:0046395∼carboxylic acid catabolic process40.03
GOTERM_BP_FATGO:0016054∼organic acid catabolic process40.031
GOTERM_BP_FATGO:0006631∼fatty acid metabolic process50.034
SP_PIR_KEYWORDSfatty acid metabolism30.049
    
Annotation cluster 3Enrichment score: 1.66  
CategoryTermCountP-Value
SP_PIR_KEYWORDScell cycle control30.013
INTERPROIPR013763:Cyclin-related30.021
INTERPROIPR006670:Cyclin30.023
SMARTSM00385:CYCLIN30.031
    
Annotation cluster 4Enrichment score: 1.50  
CategoryTermCountP-Value
GOTERM_BP_FATGO:0009719∼response to endogenous stimulus108.95E−11
GOTERM_BP_FATGO:0009725∼response to hormone stimulus90.0019
GOTERM_BP_FATGO:0010033∼response to organic substance120.0048
GOTERM_BP_FATGO:0032870∼cellular response to hormone stimulus50.0093
GOTERM_BP_FATGO:0043434∼response to peptide hormone stimulus50.015
GOTERM_MF_FATGO:0019900∼kinase binding50.02
GOTERM_BP_FATGO:0043627∼response to estrogen stimulus40.02
GOTERM_BP_FATGO:0048545∼response to steroid hormone stimulus50.03
    
Annotation cluster 5Enrichment score: 1.4  
CategoryTermCountP-Value
GOTERM_BP_FATGO:0006928∼cell motion80.028
GOTERM_BP_FATGO:0016477∼cell migration60.028
GOTERM_BP_FATGO:0051674∼localization of cell60.042

DSAT overexpressed genes were mainly annotated as “Response to external stimulus, Inflammatory response, Defence response” and as “Cellular Movement, cell to cell signalling interactions” by DAVID annotating system (Table 2) and as “Cell death” by IPA Ingenuity annotation system (Supporting Information Table 3). The ToppGene suite for gene prioritization identified the first gene that better represents the DSAT as APOC-1 gene, also known as apolipoprotein C1 (Supporting Information Table 2, lower panel). The first eight genes prioritized for SSAT and DSAT by ToppGene suite are reported as annexe (Supporting Information Table 2).

Table 2. DSAT annotation clusters
Annotation cluster 1Enrichment score: 5.67P-Value
CategoryTerm 
GOTERM_BP_FATGO:0009611∼response to wounding1.68E−08
GOTERM_BP_FATGO:0006954∼inflammatory response8.93E−08
GOTERM_BP_FATGO:0006952∼defense response6.09E−11
   
Annotation cluster 2Enrichment score: 2.94 
CategoryTermP-Value
GOTERM_CC_FATGO:0005730∼nucleolus2.05E−11
GOTERM_CC_FATGO:0043233∼organelle lumen8.15E−10
GOTERM_CC_FATGO:0031974∼membrane-enclosed lumen8.27E−10
GOTERM_CC_FATGO:0070013∼intracellular organelle lumen2.69E−10
GOTERM_CC_FATGO:0031981∼nuclear lumen9.68E−09
GOTERM_CC_FATGO:0043228∼non-membrane-bounded organelle0.014
GOTERM_CC_FATGO:0043232∼intracellular non-membrane-bounded organelle0.014
   
Annotation cluster 3Enrichment score: 2.37 
CategoryTermP-Value
KEGG_PATHWAYhsa04610:Complement and coagulation cascades9.89E−08
SP_PIR_KEYWORDSInnate immunity2.27E−09
GOTERM_BP_FATGO:0006956∼complement activation8.00E−10
GOTERM_BP_FATGO:0002541∼activation of plasma proteins involved in acute inflammatory Response9.77E−09
GOTERM_BP_FATGO:0002526∼acute inflammatory response1.31E−10
SP_PIR_KEYWORDSComplement pathway2.29E−12
SP_PIR_KEYWORDSImmune response3.12E−12
GOTERM_BP_FATGO:0006958∼complement activation, classical pathway3.36E−12
GOTERM_BP_FATGO:0050778∼positive regulation of immune response4.14E−11
GOTERM_BP_FATGO:0002455∼humoral immune response mediated by circulating immunoglobulin5.12E−10
GOTERM_BP_FATGO:0002253∼activation of immune response0.001
   
Annotation cluster 4Enrichment score: 2.30P-Value
CategoryTerm 
SP_PIR_KEYWORDSLysosome4.51E−12
GOTERM_CC_FATGO:0005764∼lysosome0.007
GOTERM_CC_FATGO:0000323∼lytic vacuole0.007
GOTERM_CC_FATGO:0005773∼vacuole0.022
   
Annotation cluster 5Enrichment score: 2.18P-Value
CategoryTerm 
GOTERM_BP_FATGO:0009991∼response to extracellular stimulus0.0021
GOTERM_BP_FATGO:0031667∼response to nutrient levels0.0070

The SSAT gene network, identified by IPA-Ingenuity systems based on regulatory relationships between the genes significant in this dataset, was mainly characterized by “metabolic genes” with insulin receptor as a central hub (Supporting Information Figure 2A). By the same method, the most significant gene network identified for DSAT included a large majority of “inflammatory genes”, with a central hub for fibrinogen (Supporting Information Figure 2B).

image

Figure 2. Expression levels of PPAR gamma, CEBPα, CEBPβ, Adiponectin, Hypoxia inducible factor-1 alpha (HIF-1α), CD68, monocyte chemotactic protein 1 (CCL2), and tumor necrosis factor alpha (TNFα) genes are reported in SSAT and DSAT. Data were normalized on RPLP0 expression levels and indicated as arbitrary units (AU). * P < 0.05; **P < 0.01.

Download figure to PowerPoint

Validations by RTqPCR

We randomly selected eight significant differentially expressed metabolic/inflammatory genes for validation by RTqPCR. The selected genes included some upregulated in SSAT vs. DSAT (CEBPα, CEBPβ, PPARγ, Adiponectin) as well as some upregulated in DSAT vs. SSAT (HIF1α, CD68, CCL2, TNFα). Expression levels fold changes were concordant with the microarray data fold changes (Rho2 = 0.51, P-value < 0.001) (Figure 2).

In vitro differentiation rate in SSAT and DSAT SVF cells

SVF cells were isolated from the two adipose depots (SSAT-SVF and DSAT-SVF), cultured until confluence and then differentiated in vitro into mature adipose cells. Multiple lipid droplets were clearly visible starting from the third day of culture. The lipid accumulation rate was then evaluated and the amount of intracellular lipids was not significantly different after 10 days of in vitro differentiation between SSAT-SVF and DSAT-SVF (Supporting Information Figure 3).

image

Figure 3. Adipocyte mean size (μm), inflammatory infiltration (percentage of CD68 immunopositive cells), fibrosis score, mean lumen vessel size area (μm2), and vascular vessel density in SSAT (black bars) and DSAT (grey bars) are shown. A representative picture of SSAT and DSAT (40× magnification) is shown and large vessels in DSAT indicated by arrows.

Download figure to PowerPoint

Morphologic differences in SSAT and DSAT

The morphology of each collected SSAT/DSAT paired samples was studied and the adipocyte hypertrophy, inflammation (by CD68+ cells estimation), fibrosis and vascularization were estimated. A significant adipocyte hypertrophy was observed in SSAT compared to DSAT (P < 0.01) whereas DSAT contained more CD68 immune-positive cells than SSAT in the parenchyma (P < 0.001) (Figure 3). Moreover, the two depots had comparable levels of overall fibrosis (Figure 3). The number of vascular vessels was significantly higher in SSAT than DSAT, despite a significantly smaller mean lumen size (Figure 3).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In this study we show that, in obese patients, human abdominal SSAT and DSAT subcompartments are different both at molecular and morphologic levels. The molecular profile observed was mainly “metabolic” for SSAT: this depot was characterized by adipocyte hypertrophy and a lower level of inflammatory cells infiltration. DSAT displayed an “inflammatory” molecular profile probably because of the significantly higher inflammatory cells infiltration. Moreover, in SSAT and DSAT a similar degree of fibrosis but a different vessel density and mean size was observed. Our study is in agreement with the recently published data, suggesting a specific role for DSAT in obesity hepatic complications [16]. In addition, thanks to microarray analysis, we have been able to identify potential markers of these two depots that could be taken into account when studying human adipose tissue and obesity complications. Indeed, despite a global metabolic role for SSAT, the SAA protein genes were preferentially expressed in this compartment suggesting a possible link between SSAT and renal disease. Taken together with previous reports [16], the data presented herein further support the hypothesis that SSAT and DSAT have different roles in the appearance of obesity complications and should be deeply investigated in the future. We must underline that, as clinical data of the enrolled patients were unavailable, these results presented herein are based exclusively on computational analysis and thus they must be validated by clinical observations in larger cohorts of morbidly obese patients.

The results presented herein are also in agreement with the recently published data on superficial and DSAT in swine [24], also showing that lipid metabolism features, typical of SSAT, and inflammatory-immunologic profile of DSAT, are regulated by methylation. The existence of an epigenetic regulation of SSAT/DSAT-gene expression has yet to be confirmed in humans.

The different characteristics of the two adipose layers here described may be relevant for the analysis of some previously published data related to fat removal in the abdomen with different surgical techniques, liposuction and dermolipectomy. Liposuction and dermolipectomy are the most frequently used surgical techniques to address obesity. While liposuction of the abdomen have contrasting effects on insulin sensitivity and in decreasing vascular risk factors [25, 26], dermolipectomy procedures seem to improve glucose handling and to decrease levels of some inflammatory circulating markers (i.e., TNFα and IL6) [27]. Indeed, it is tempting to speculate that during liposuction a less amount of DSAT could be removed accounting for a modest metabolic effect. Conversely, by dermolipectomy a larger amount of DSAT removal might be linked to the observed metabolic improvements [27]. Further studies in larger populations are certainly needed to confirm these findings.

In addition, morphologic differences observed for DSAT and SSAT may also have implications for site-specific differences in adipose cell biology. A lower in vitro lipolytic activity has already been demonstrated for DSAT, indicating that these two adipose depots are metabolically different [28]. In agreement with our previously reported data for SSAT/DSAT adipose precursors in lean subjects [14, 15], in this study we did not observe differences in in vitro triglycerides accumulation rate in precursor cells isolated from the two depots of obese patients. These findings may suggest that the increase of SSAT and DSAT is not determined by a different functional activity of precursor cells per se but it is likely affected by environmental “extra-adipocyte” factors such as vascularization, innervations, and extra cellular matrix remodeling.

We here propose that, when studying human subcutaneous abdominal adipose tissue, both for gene expression and protein content/secretion or morphology, is important to clearly indicate the collection of either SSAT or DSAT (or eventually both depots without distinction). Indeed, as clearly demonstrated by the present study, sampling of abdominal adipose tissue from above or below the fascia could lead to a great variability in the results and might explain some conflicting data in the literature.

In summary, we have further characterized the molecular and morphological differences of abdominal SSAT and DSAT in morbidly obese patients. The physiological relevance and the applicability of all presented findings are crucial to expand the knowledge of obesity complications and to improve translational research.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The authors would like to thank Laura Ermellino and Anna Rita Grindati for technical assistance with adipose tissue samples processing as well as Dr. M. Besozzi for the free access to anatomical pathology laboratory facilities at the Istituto Auxologico Italiano, Cusano Milanino, Italy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
oby20417-sup-0001-SuppInfo.ppt689KSupporting Information Figures
oby20417-sup-0002-SuppInfo.doc153KSupporting Information Tables

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.