• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Aug 17, 2010; 107(33): 14781–14786.
Published online Aug 2, 2010. doi:  10.1073/pnas.1003512107
PMCID: PMC2930432
Medical Sciences

De novo generation of white adipocytes from the myeloid lineage via mesenchymal intermediates is age, adipose depot, and gender specific

Abstract

It is generally assumed that white adipocytes arise from resident adipose tissue mesenchymal progenitor cells. We challenge this paradigm by defining a hematopoietic origin for both the de novo development of a subset of white adipocytes in adults and a previously uncharacterized adipose tissue resident mesenchymal progenitor population. Lineage and cytogenetic analysis revealed that bone marrow progenitor (BMP)-derived adipocytes and adipocyte progenitors arise from hematopoietic cells via the myeloid lineage in the absence of cell fusion. Global gene expression analysis indicated that the BMP-derived fat cells are bona fide adipocytes but differ from conventional white or brown adipocytes in decreased expression of genes involved in mitochondrial biogenesis and lipid oxidation, and increased inflammatory gene expression. The BMP-derived adipocytes accumulate with age, occur in higher numbers in visceral than in subcutaneous fat, and in female versus male mice. BMP-derived adipocytes may, therefore, account in part for adipose depot heterogeneity and detrimental changes in adipose metabolism and inflammation with aging and adiposity.

Keywords: bone marrow, hematopoietic, stem cell

Adipose tissue is the primary site of energy storage in the body and an important endocrine organ that participates in the regulation of energy intake and expenditure. Changes in adiposity associated with aging, weight gain, and gender can alter regional fat distribution, and adipose tissue metabolism and inflammation (13). Such changes, particularly visceral versus subcutaneous fat accumulation and inflammation, have been linked to obesity-related comorbidities such as type 2 diabetes and cardiovascular disease (3, 4).

The generation of new adipocytes from progenitor cells has been a topic of great interest, in terms of understanding normal adipose tissue development and turnover and the expansion of adipose tissue that occurs with obesity. Two types of adipocytes are present in mammals; white adipocytes, the primary function of which is lipid storage, and brown adipocytes, which store lipids and also oxidize fatty acids for heat production (5). It is generally accepted that white adipocytes arise solely from progenitors residing in fat stroma (6, 7), whereas myoblastic precursors serve as the source of brown adipocytes (8). Not all white adipocyte progenitors are equivalent, as distinct subpopulations have been identified in fat from different body locations (9). Functional variations between different white adipose depots may reflect depot-specific differences in the progenitors from which adipocytes are generated (10, 11). Despite the recognition of distinct progenitor populations in adipose tissue, it has been assumed that all white adipocytes and their progenitors arise solely from resident cells of mesenchymal origin.

We previously reported that a subpopulation of adipocytes in white and brown fat tissue arise from bone marrow progenitor (BMP) cells (12). These results challenge the paradigm of a resident mesenchymal origin for all white adipocytes, and question the developmental origin of adipose stromal progenitor subpopulations. Here we report the detailed analysis of BMP-derived adipocytes and stromal cells in adipose tissue to clarify the lineage origin of these unique cells. Our data define a myeloid origin for both the de novo development of a subset of white adipocytes and a previously uncharacterized adipose tissue resident mesenchymal progenitor population. The BMP-derived adipocytes accumulated over time in a depot- and gender-specific manner with higher numbers in visceral versus subcutaneous fat and in female rather than male mice. This differential accumulation is particularly interesting in view of global gene expression patterns in BMP-derived adipocytes showing decreased expression of mitochondrial and peroxisomal genes related to organelle biogenesis and lipid oxidation, and increased expression of inflammatory cytokine genes. Thus, differential accumulation of BMP-derived adipocytes may explain, in part, adipose depot heterogeneity and detrimental changes in adipose function with aging, adiposity, and gender. The results also highlight the need for a comprehensive examination of adipose stromal populations to identify subpopulations that may be unfavorable for regenerative medicine applications.

Results

De Novo Generation of Adipocytes from Bone Marrow Hematopoietic/Myeloid Progenitor Cells via Differentiation.

Our previous work demonstrated a bone marrow origin for a population of adipocytes with variable morphology in white and brown adipose depots (Fig. 1A) (12). As there is no direct evidence that bone marrow (BM) mesenchymal cells enter the circulation, our observations suggested a hematopoietic origin for BMP-derived adipocytes. To evaluate this idea, competitive BM transplantations were performed using labeled male BM subpopulations isolated based on expression of CD45 (pan-leukocyte marker) or hematopoietic lineage (lin, combination of CD11b, Gr-1, B220, CD5, and Ter119) markers and transplanted into unlabeled female recipients. Labeled adipocytes were detected only in mice receiving labeled hematopoietic (CD45+) BM cells (Fig. 1B). Likewise, labeled adipocytes were present only in mice receiving labeled lin+ BM early posttransplantation, presumably derived from more committed short-term repopulating cells (Fig. 1B). Over time, labeled adipocytes were also detected in recipients of labeled lin− BM. The data indicate that BMP-derived adipocytes are generated from lineage-committed hematopoietic progenitor cells that arise from lin− BM over time. Similar BM reconstitution experiments indicated that BMP-derived adipocyte precursors expresses the Sca-1 and c-kit surface markers (Fig. S1).

Fig. 1.
BM hematopoietic progenitors generate adipocytes via myeloid intermediates. (A) Representative photomicrographs of LacZ-expressing adipocytes in adipose tissue sections or the adipocyte fraction from mice transplanted with BM from LacZ-expressing male ...

Given the promiscuity of myeloid cells and their ability to generate phenotypes of skeletal muscle (13), vascular endothelium (14), and liver (15), we hypothesized that BM myeloid cells would give rise to BMP-derived adipocytes. To test this hypothesis, we looked for labeled adipocytes in adipose tissue from LysMcreROSAflox/STOP mice in which LacZ expression is restricted to the myeloid lineage (13, 16). A substantial number of labeled adipocytes were detected in fat tissue from the LysMcreROSAflox/STOP mice (Fig. 1C). Therefore, BMP-derived adipocytes arise from myeloid progenitor cells. Further cytometric and visual examination was performed to ensure that lineage analysis results were not due to stromal contamination of adipocyte preparations (Fig. S2). Moreover, these data indicate that BMP-derived adipocytes are generated in a non-BM transplant model, and thus are not an artifact of myeloablative injury.

To confirm that BMP-derived fat cells arise via differentiation of BMP cells, we performed cytogenetic analysis to assess the chromosome content of genetically labeled fat cells arising from male donor marrow. Probes were generated to portions of the SRY (Y chromosome) and dystrophin (X chromosome) genes. Analysis revealed that all BMP-derived fat cells contained single copies of X and Y chromosome each (Fig. 1D). Thus, the BMP-derived fat cells were derived entirely from the transplanted male BM cells and arise from differentiation of BMP cells rather than fusion of progenitors with resident adipose tissue cells.

To further substantiate a myeloid origin for BMP-derived adipocytes, we examined whether BM myeloid cells could differentiate into adipocytes in vitro. Murine myeloid (CD11b-positive) BM cells were isolated by antibody/magnetic bead separation from whole BM. Cells were cultured in three-dimensional matrices composed of either Matrigel or fibrin (17). Liquid media was layered on the matrices and supplemented with adipogenic inducers. Within 14 d, cells containing lipid were present in both Matrigel (Fig. 1E) or fibrin (Fig. S3A) matrices. In Matrigel the cells exhibited a large multilocular appearance, whereas in fibrin cells exhibited a smaller unilocular phenotype. RT-PCR analysis of RNA from the adipocytes confirmed perilipin expression (Fig. S3B). Therefore, BM myeloid cells have the capacity to undergo adipogenesis.

Tissue Resident “Mesenchymal” Adipocyte Progenitor Cells Are Derived From a Myeloid Lineage.

The generation of adipocytes from hematopoietic progenitors raises the issue of whether tissue-resident adipocyte precursors also arise from hematopoietic sources. To examine this possibility, we isolated adipose stroma, which contains the adipose precursors, from LysMcreROSAflox/STOP mice. LacZ-expressing cells were analyzed for expression of CD45 and CD11b (Fig. 2A). Analyses revealed a myeloid-derived LacZ+ population double negative for the hematopoietic and myeloid markers (Fig. 2B), which was capable of adipogenic differentiation in vitro (Fig. 2C). The results demonstrate that a subset of tissue-resident adipocyte progenitor cells arise from myeloid intermediates but lack hematopoietic markers and resemble mesenchymal stroma.

Fig. 2.
Myeloid-derived adipocyte progenitor cells lacking hematopoietic markers are present in adipose tissue. Gonadal white adipose tissue was taken from LysMcreROSAflox/STOP mice and digested with collagenase. Stroma was recovered by centrifugation, loaded ...

Preliminary examination of the adipogenic potential of different adipocyte progenitor populations indicated that ≈70% of BM mesenchymal (CD45−) cells differentiate into adipocytes within 3 d of treatment with adipogenic inducing agents (Fig. S4A). Adipose stromal mesenchymal cells also undergo adipogenic conversion within 3 d of induction, but only 40% of the cells differentiated (Fig. S4B). Lipid accumulation was not evident in BM myeloid cells until day 8 postinduction, and less than 1% of the cells became adipocytes (Fig. S4C). Adipogenic differentiation of stromal myeloid-derived adipocyte progenitors was also evident by 8 d postinduction, with ≈40% of the cells generating adipocytes (Fig. S4D). At this time, it is not clear whether these differences in adipogenic efficiency and differentiation rate reflect intrinsic differences between these populations or reflect population heterogeneity or different culture conditions (e.g., monolayer versus 3D matrix).

BMP-Derived Adipocytes Accumulate with Age in a Depot- and Gender-Specific Manner.

Functional variations between different adipose depots are due in part to niche-specific differences in progenitors from which mature adipocytes are generated (911). Likewise, changes in adipose tissue function with aging or associated with gender may also be due to the differential accumulation of adipocytes arising from distinct progenitor subpopulations. Therefore, we examined the accumulation of BMP-derived adipocytes over time, their distribution in various fat depots, and their presence in male versus female mice. Flow cytometry of samples at various times posttransplantation showed a progressive increase in BMP-derived adipocyte number over time (Fig. 3A). Absolute numbers of BMP-derived adipocytes varied between transplants becauses of transplant efficiency and degree of engraftment, but levels invariably increased with time. A similar time-dependent increase is seen in the appearance of BMP-derived adipocytes from lin− BM in Fig. 1B. The highest numbers of BMP-derived adipocytes were consistently seen in gonadal and perirenal fat, with lower numbers in subcutaneous and intrascapular fat (Fig. 3B). Significantly higher numbers of BMP-derived adipocytes were present in fat from female than male mice (Fig. 3C). The data show that BMP-derived adipocytes accumulate over time primarily in females and in visceral versus subcutaneous fat depots.

Fig. 3.
BMP-derived adipocytes accumulate with age in a depot and gender-specific manner. (A) Gonadal adipose tissue was harvested from female recipient mice at 4, 8, and 12 wk after GFP BM transplantation. Adipocyte fraction was isolated, and GFP-expressing ...

BMP-Derived Adipocytes Are Distinct from Conventional White and Brown Adipocytes.

We used global gene expression analysis to determine whether the BMP-derived fat cells were bona fide adipocytes, and to assess their relationship to conventional white and brown adipocytes as well as circulating (Pb-Myeloid) and stromal (St-Myeloid) cells. These populations were isolated from the same genetically labeled mice in two separate experiments, permiting us to evaluate different populations from the same mice. Complimentary DNA from each sample was hybridized to Affymetrix murine whole-genome arrays. Examination of the frequency of expression values after normalization of the data revealed consistent values between all samples (Fig. S5). Principal component analysis showed that white, brown, and BMP-derived adipocytes were more closely related to each other than to circulating or stromal myeloid cells (Fig. 4A). Supervised hierarchical analysis of 26 genes associated with the adipocyte phenotype revealed that all were expressed in BMP-derived fat cells (Fig. 4B). Many of the genes including perilipin, PPARγ, C/EBPα, Glut4, fatty acid synthase, and hormone-sensitive lipase were expressed at comparable levels in BMP-derived fat cells and white or brown adipocytes. Notable exceptions included uncoupling protein 1, which was highly expressed only in brown adipocytes, and leptin, which was more highly expressed in white adipocytes than brown or BMP-derived adipocytes. Fifteen of these genes were not detected in circulating myeloid cells, and five genes were absent in stromal myeloid cells. Of those genes expressed in myeloid cells, levels were generally lower than in adipocytes. Expression of the myeloid/dendritic markers CD11b and CD11c was high in myeloid cells but minimal in adipocytes. Relative levels of leptin, CD11b, and CD11c expression were verified by quantitative PCR (Fig. S6A). Perilipin expression was verified by RT-PCR and fluorescent deconvolution microscopy (Fig. S6 B and C). Previous semiquantitative RT-PCR results for various adipocyte-related genes also confirm the gene array data (12).

Fig. 4.
BMP-derived adipocytes are distinct from conventional white and brown adipocytes. BMP-derived adipocytes were purified from transplanted recipient mice by flow cytometry. GFP/LacZ-negative white adipocytes (two samples) were purified from gonadal fat ...

When only white, brown and BMP-derived adipocytes were subjected to principal component analysis, each set of duplicate samples segregated to a different region of the diagram (Fig. 4C). This is consistent with the functional distinction between white and brown adipocytes, and indicates that BMP-derived adipocytes are also a unique adipocyte subpopulation. Unsupervised hierarchical clustering revealed a group of ≈530 genes the expression of which was decreased more than 2-fold (P ≤ 0.05) in BMP-derived adipocytes compared with either white or brown adipocytes. This group of genes was enriched for factors involved in mitochondrial and peroxisomal biogenesis and lipid oxidation. Figure 4D shows a supervised hierarchical cluster of 46 of these genes, the expression of which was 2.6- to 21-fold lower in the BMP-derived fat cells. Clustering also highlighted a group of 340 genes the expression of which was 2-fold or greater than that of white and brown adipocytes. These genes fell into several categories, including a small cluster of inflammatory genes and chemotactic factors with expression levels 3.6- to 14-fold higher in BMP-derived adipocytes than white or brown fat cells (Fig. 4D). This group included interleukin-6, -15, and -16, CXCL9, CX3CL1, stem cell factor, and stromal cell-derived factor-1. The elevated expression of these cytokines in BMP-derived adipocytes was confirmed by quantitative PCR (Fig. S6A). Overall, the expression data reveal a detrimental phenotype for the BMP-derived adipocytes characterized by low leptin expression (Fig. 4B), low mitochondrial/peroxisomal content and oxidative capacity, and elevated inflammatory cytokine production (Fig. 4D).

Discussion

The studies reported here highlight a nonresident origin for a subset of white adipocytes, and define a hematopoietic origin for these cells and a population of tissue-resident adipocyte progenitors with mesenchymal characteristics. Our work definitively demonstrates that hematopoietic progenitors generate BMP-derived adipocytes and adipocyte precursors via myeloid intermediates. These conclusions are supported by our previous work (12) and that of Ogawa et al. (18) and Marra et al. (19) reporting the detection of BM-derived adipocytes in mice transplanted with whole bone marrow or hematopoietic stem cells.

In this regard, it is interesting to note that Koh et al. (20) failed to detect genetically labeled adipocytes in a similar but not identical BM transplant model. Their negative results may be due to their use of BM donor mice in which GFP expression was under control of the chicken β-actin gene promoter. Expression from this promoter can be poor or absent in adipose tissue of transgenic mice (e.g., strains B6.129(ICR)-Tg(ACTB-ECFP)CK6Nagy/J and 129-Tg(ACTB-YFP)7AC5Nagy/J from Jackson Laboratories). In addition, GFP appears to be particularly soluble or “leaky” in Koh et al.'s donor mice (21), and can be lost without substantial fixation before detection (22). Also, EGFP levels in BMP-derived adipocytes may be lower than levels in conventional adipocytes, making it difficult to detect by confocal microscopy against the background of brighter BM-derived cells in adipose tissue.

As our cytogenetic data rule out fusion as a mechanism for the production of BMP-derived adipocytes, we conclude that the generation of these cells from BM hematopoietic cells represents a unique developmental pathway for production of fat cells that involves transdifferentiation (across lineage) of myeloid intermediates to bona fide adipocytes. We propose a model in which hematopoietic stem cells (CD45+) in the BM give rise to myeloid lineage cells (CD45+/CD11b+), which traffic via the circulation to adipose tissue, where they lose their hematopoietic markers, taking on the appearance of resident mesenchymal adipocyte precursors (CD45-/CD11b−) (Fig. 5). Terminal differentiation of these cells results in de novo generation of BMP-derived adipocytes. Alternatively, hematopoietic stem cells (HSCs) that reside in adipose tissue (2325) might also serve as a resident source of white adipocytes. However, there are arguments against this alternative model. First, adipose HSCs are incapable of reconstituting long-term hematopoiesis (23). Second, adipocytes are generally negative regulators of hematopoiesis (26). Third, Cao et al. (27) have demonstrated that bona fide hematopoiesis in the adult is restricted to foci in BM and spleen, even though HSCs are present in nonmedullary tissues. Thus, we favor BM as the source for BMP-derived adipocytes and their precursors.

Fig. 5.
Model showing generation of myeloid-derived progenitor cells from hematopoietic stem cells in BM. The myeloid-derived progenitors traffic to fat tissue where they loose hematopoietic marker expression and transdifferentiate into BMP-derived white adipocytes. ...

In light of our data showing generation of adipocytes from BM myeloid cells, it is interesting to consider the ability of macrophages and other myeloid cells to express certain adipocyte biomarkers including FABP4 (28), perilipin (29, 30), adipophilin (2931), PPARγ (32), and hormone-sensitive lipase (33). Macrophages also respond to thiazolidinediones, potent inducers of adipogenesis, by up-regulating adipocyte factors such as FABP4 (28). Foam cells, lipid-laden macrophages common in atherosclerotic plaques, share many of these features (34). It is not surprising, then, that BM myeloid cells serve as a source for bona fide adipocytes. Moreover, the elevated expression of inflammatory cytokines by these unique adipocytes may reflect their origin from cells involved in immune function and inflammation. Despite these features, there are crucial distinctions between adipocytes and macrophages. For example, adipocytes do not express macrophage or foam/dendritic cell markers like CD11b and CD11c (confirmed by our analysis). Likewise, macrophages/foam cells do not express the complete cadre of adipocyte-related factors such as adiponectin (35) (Fig. 4B), nor do they exhibit insulin-stimulated glucose uptake. Furthermore, the origin of lipids stored in macrophages/foam cells and adipocytes are different (36). These distinctions are underscored by our global gene expression results showing white, brown, and BMP-derived adipocytes to be dissimilar from myeloid cells.

Another important observation from these studies is the depot-specific accumulation of BMP-derived adipocytes. Distinct populations of adipocytes and adipocyte progenitors are present in fat depots from different anatomical locations and are responsible in part for the differential function of individual depots (10, 11). Our results further define this process by revealing a preferential accumulation of BMP-derived adipocytes in visceral fat depots. Some of the unique properties of distinct adipose depots may reflect the differential accumulation and distinctive phenotype of BMP-derived adipocytes.

The impact of BMP-derived adipocytes on depot heterogeneity probably increases as they accumulate over time. Interestingly, aging is associated with increases in visceral fat with concomitant loss of lower body subcutaneous adipose tissue (2). This pattern of fat redistribution, coupled with the preferential accumulation of BMP-derived adipocytes in abdominal depots, suggests an increasing impact of these cells on visceral adipose function and body-wide physiology as subjects age. Our data also reveal a preferential accumulation of BMP-derived adipocytes in female versus male mice. This may be an especially important consideration in humans, among whom females generally exhibit greater adiposity than males (1).

The age, gender, and depot-specific accumulation of BMP-derived adipocytes is especially significant, given the unique and possibly detrimental phenotype of these cells. Particularly notable is the decreased expression in BMP-derived adipocytes of leptin, an adipokine that participates in the regulation of satiety and energy expenditure. Accrual of BMP-derived adipocytes may lower leptin levels over time, leading to increased food intake and decreased energy expenditure. Elevated expression of inflammatory cytokines by BMP-derived adipocytes may also have an impact on metabolism. Interleukin-6 can directly decrease insulin sensitivity via inhibition of insulin receptor signaling (37), and other cytokines may elicit recruitment of inflammatory cells to adipose tissue, further exacerbating insulin resistance. In this context, it is interesting to note that BMP-derived adipocytes preferentially accumulate in visceral fat, which is recognized for lower leptin production (38) higher numbers of macrophages, and greater inflammatory cytokine production than subcutaneous fat (4). We also noted lower expression of genes related to mitochondrial and peroxisomal biogenesis and lipid oxidation in BMP-derived adipocytes, suggesting lower oxidative capacity compared with conventional adipocytes.

The differential accumulation and unique phenotype of BMP-derived adipocytes provides an explanation not only for adipose depot heterogeneity and differences related to age and gender, but an explanation specifically for the unfavorable changes in adipose tissue function that accompany increased visceral adiposity and aging. Further studies will be needed to fully appreciate the impact of BMP-derived adipocytes on adipose tissue development and function. Finally, a careful investigation of myeloid-derived adipocyte progenitors is warranted, given their potential detrimental impact on regenerative medicine applications involving adipose stroma.

Methods

BM Transplantation.

All procedures and treatments were approved by the Institutional Animal Care and Use Committee. Transplantations were performed as previously described (12).

Competitive BM transplants were performed with BM from either GFP- or LacZ-expressing male donor mice separated into subpopulations based on the expression of specific cell surface markers by antibody-labeled magnetic bead separation. Two rounds of magnetic separation were performed to improve separation efficiency, which was greater than 90% (Fig. S7).

The separated, labeled BM cells were combined with whole BM from WT mice to ensure survival of recipients. Engraftment of the labeled cell subpopulations was greater than 90% for all transplants, as determined by flow cytometry of PBMCs and whole BM (Fig. S8).

Fractionation of Adipose Tissue and Preparation of PBMCs.

Adipocyte and stromal/vascular fractions were prepared by collagenase digestion and flotation/differential centrifugation, as previously described (12).

Flow Cytometry.

BM cells or stromal/vascular cells were stained with antibodies to the cell surface markers indicated in the figure legends. Gating strategies included dead cell discrimination with propidium iodide, exclusion of red blood cells with Ter199, and doublet discrimination. Controls included unstained cells and cell suspensions incubated with APC- or PE-conjugated isotype-matched control antibodies. Cells were analyzed on a CyAn ADP flow cytometer (Beckman Coulter).

Adipocytes were analyzed or sorted using a Legacy Moflo cell sorter with Summit 4.3 software (Beckman Coulter). GFP-expressing or LacZ (detected with FDG)-expressing cells were collected with a 530/40 bandpass filter, autofluorescence or PE at 570/40, and APC at 675/40. Sort mode was set to Purify 1.

Cytogenetic Analysis of Chromosome Number.

BAC clones encompassing regions of the SRY gene in mouse chromosome Y and the dystrophin gene in mouse chromosome X were subjected to amplification and labeled by nick translation with SpectrumRed-conjugated dUTPs for SRY and with SpectrumGreen-conjugated dUTPs for dystrophin. The labeled probes were tested in normal mouse cell lines to verify hybridization efficiency and chromosomal mapping (Fig. S9).

GFP- and LacZ-expressing adipocytes were isolated by flow cytometry. The cells were treated with 0.3% Nonidet P-40 to disperse clumps, pelleted, and dissolved in 3:1 methanol:acetic fixative. The suspension was dropped on etched slides and allowed to dry overnight. The slides were washed and hybridized as previously described (39). A total of 100 interphase nuclei were analyzed in each sample. Images were captured using CytoVision software (Applied Imaging).

Gene Microarray Analysis.

RNA was prepared with PicoPure RNA Isolation Kit reagents (Arcturus Bioscience). CDNA generated from amplified RNA was hybridized to Affymetrix. Mouse 430v2 chips for 16 h at 45 °C. Arrays were washed and stained on a GeneChip Fluidics Station 450 and scanned with a GeneChip 7G Scanner.

Data files were imported into Partek Genomics Suite software (Partek) and normalized using the RMA algorithm. Principal component analysis was performed on all 45,000 transcripts. All data sets were subjected to multiway ANOVA, and transcripts with signals differing between groups by more than 1.25-fold at a value of P ≤ 0.05 were selected for further study. Supervised hierarchical cluster analysis was performed with the k-means algorithim. The data discussed here have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE19757 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE19757).

3D Cell Culture.

BM cells were separated into myeloid (CD11b+) and nonmyeloid (CD11b−) subpopulations by magnetic bead separation. For Matrigel, the cells were pelleted and resuspended in cold Matrigel at 106 cells/mL. A 1-mL quantity of the suspension was transferred to individual wells of a 12-well plate and placed in a tissue culture incubator to warm and solidify. Fibrin matrices containing BM cells were prepared as previously described (17).

DMEM containing 10% FCS was gently added on top of the solid matrices. The medium was replaced with fresh medium containing 3 mM isobutylmethylxanthine, 1 μg/mL insulin, and 1 μM dexamethasone for 48 h. Every 48 h thereafter, the medium was replaced with medium containing 1 μg/mL insulin.

Supplementary Material

Supporting Information:

Acknowledgments

This research was supported by National Institutes of Health R01 Grant DK078966 (to D.J.K.), an American Heart Association Grant in Aid (to S.M.M.), and RO1 Grant DK059767 and P30 Grant DK048520 (to J.E.F.).

Footnotes

*This Direct Submission article had a prearranged editor.

The authors declare no conflict of interest.

Data deposition: The gene array data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE19757 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE19757).

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1003512107/-/DCSupplemental.

References

1. Geer EB, Shen W. Gender differences in insulin resistance, body composition, and energy balance. Gend Med. 2009;6(Suppl 1):60–75. [PMC free article] [PubMed]
2. Kuk JL, Saunders TJ, Davidson LE, Ross R. Age-related changes in total and regional fat distribution. Ageing Res Rev. 2009;8:339–348. [PubMed]
3. Stevens J. Obesity, fat patterning and cardiovascular risk. Adv Exp Med Biol. 1995;369:21–27. [PubMed]
4. Hamdy O, Porramatikul S, Al-Ozairi E. Metabolic obesity: The paradox between visceral and subcutaneous fat. Curr Diabetes Rev. 2006;2:367–373. [PubMed]
5. Park KW, Halperin DS, Tontonoz P. Before they were fat: Adipocyte progenitors. Cell Metab. 2008;8:454–457. [PubMed]
6. Ailhaud G, Grimaldi P, Négrel R. Cellular and molecular aspects of adipose tissue development. Annu Rev Nutr. 1992;12:207–233. [PubMed]
7. Tang W, et al. White fat progenitor cells reside in the adipose vasculature. Science. 2008;322:583–586. [PMC free article] [PubMed]
8. Kajimura S, et al. Initiation of myoblast to brown fat switch by a PRDM16-C/EBP-beta transcriptional complex. Nature. 2009;460:1154–1158. [PMC free article] [PubMed]
9. Perrini S, Cignarelli A, Ficarella R, Laviola L, Giorgino F. Human adipose tissue precursor cells: A new factor linking regulation of fat mass to obesity and type 2 diabetes? Arch Physiol Biochem. 2009;115:218–226. [PubMed]
10. Tchkonia T, et al. Fat depot-specific characteristics are retained in strains derived from single human preadipocytes. Diabetes. 2006;55:2571–2578. [PubMed]
11. Tchkonia T, et al. Identification of depot-specific human fat cell progenitors through distinct expression profiles and developmental gene patterns. Am J Physiol Endocrinol Metab. 2007;292:E298–E307. [PubMed]
12. Crossno JT, Jr, Majka SM, Grazia T, Gill RG, Klemm DJ. Rosiglitazone promotes differentiation of bone marrow-derived circulating progenitor cells to multilocular adipocytes in adipose tissue. J Clin Invest. 2006;116:3220–3228. [PMC free article] [PubMed]
13. Camargo FD, et al. Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat Med. 2003;9:1520–1527. [PubMed]
14. Bailey AS, et al. Myeloid lineage progenitors give rise to vascular endothelium. Proc Natl Acad Sci USA. 2006;103:13156–13161. [PMC free article] [PubMed]
15. Willenbring H, et al. Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nat Med. 2004;10:744–748. [PubMed]
16. Clausen BE, Burkhardt C, Reith W, Renkawitz R, Förster I. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic Res. 1999;8:265–277. [PubMed]
17. Gérard C, Blouin K, Tchernof A, Doillon CJ. Adipogenesis in nonadherent and adherent bone marrow stem cells grown in fibrin gel and in the presence of adult plasma. Cells Tissues Organs. 2008;187:186–198. [PubMed]
18. Sera Y, et al. Hematopoietic stem cell origin of adipocytes. Exp Hematol. 2009;37:1108–1120. [PMC free article] [PubMed]
19. Tomiyama K, et al. Characterization of transplanted green fluorescent protein+ bone marrow cells into adipose tissue. Stem Cells. 2008;26:330–338. [PMC free article] [PubMed]
20. Koh YJ, et al. Bone marrow-derived circulating progenitor cells fail to transdifferentiate into adipocytes in adult adipose tissues in mice. J Clin Invest. 2007;117:3684–3695. [PMC free article] [PubMed]
21. Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett. 1997;407:313–319. [PubMed]
22. Jockusch H, Voigt S, Eberhard D. Localization of GFP in frozen sections from unfixed mouse tissues: Immobilization of a highly soluble marker protein by formaldehyde vapor. J Histochem Cytochem. 2003;51:401–404. [PubMed]
23. Corre J, et al. Human subcutaneous adipose cells support complete differentiation but not self-renewal of hematopoietic progenitors. J Cell Physiol. 2006;208:282–288. [PubMed]
24. Han J, et al. Adipose tissue is an extramedullary reservoir for functional hematopoietic stem and progenitor cells. Blood. 2010;115:957–964. [PubMed]
25. Miñana MD, Carbonell-Uberos F, Mirabet V, Marín S, Encabo A. IFATS collection: Identification of hemangioblasts in the adult human adipose tissue. Stem Cells. 2008;26:2696–2704. [PubMed]
26. Naveiras O, et al. Bone-marrow adipocytes as negative regulators of the haematopoietic microenvironment. Nature. 2009;460:259–263. [PMC free article] [PubMed]
27. Cao Y-A, et al. Shifting foci of hematopoiesis during reconstitution from single stem cells. Proc Natl Acad Sci USA. 2004;101:221–226. [PMC free article] [PubMed]
28. Pelton PD, Zhou L, Demarest KT, Burris TP. PPARgamma activation induces the expression of the adipocyte fatty acid binding protein gene in human monocytes. Biochem Biophys Res Commun. 1999;261:456–458. [PubMed]
29. Persson J, Degerman E, Nilsson J, Lindholm MW. Perilipin and adipophilin expression in lipid loaded macrophages. Biochem Biophys Res Commun. 2007;363:1020–1026. [PubMed]
30. Tanigawa K, et al. Expression of adipose differentiation-related protein (ADRP) and perilipin in macrophages infected with Mycobacterium leprae. FEMS Microbiol Lett. 2008;289:72–79. [PubMed]
31. Paul A, Chang BH, Li L, Yechoor VK, Chan L. Deficiency of adipose differentiation-related protein impairs foam cell formation and protects against atherosclerosis. Circ Res. 2008;102:1492–1501. [PMC free article] [PubMed]
32. Marx N, Sukhova G, Murphy C, Libby P, Plutzky J. Macrophages in human atheroma contain PPARgamma: Differentiation-dependent peroxisomal proliferator-activated receptor gamma(PPARgamma) expression and reduction of MMP-9 activity through PPARgamma activation in mononuclear phagocytes in vitro. Am J Pathol. 1998;153:17–23. [PMC free article] [PubMed]
33. Yeaman SJ. Hormone-sensitive lipase—new roles for an old enzyme. Biochem J. 2004;379:11–22. [PMC free article] [PubMed]
34. Webb NR, Moore KJ. Macrophage-derived foam cells in atherosclerosis: Lessons from murine models and implications for therapy. Curr Drug Targets. 2007;8:1249–1263. [PubMed]
35. Luo N, et al. Macrophage adiponectin expression improves insulin sensitivity and protects against inflammation and atherosclerosis. Diabetes. 2010;59:791–799. [PMC free article] [PubMed]
36. Vainio S, Ikonen E. Macrophage cholesterol transport: A critical player in foam cell formation. Ann Med. 2003;35:146–155. [PubMed]
37. Senn JJ, Klover PJ, Nowak IA, Mooney RA. Interleukin-6 induces cellular insulin resistance in hepatocytes. Diabetes. 2002;51:3391–3399. [PubMed]
38. Woods SC, Gotoh K, Clegg DJ. Gender differences in the control of energy homeostasis. Exp Biol Med (Maywood) 2003;228:1175–1180. [PubMed]
39. Varella-Garcia M, et al. The detection of chromosomal aneusomy by fluorescence in situ hybridization in sputum predicts lung cancer incidence. Cancer Prev Res (Phila Pa) 2010;3:447–453. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...