Recruitment and differentiation of intramuscular preadipocytes in stromal-vascular cell cultures derived from neonatal pig semitendinosus muscles 1

Transcription

1 Recruitment and differentiation of intramuscular preadipocytes in stromal-vascular cell cultures derived from neonatal pig semitendinosus muscles 1 G. J. Hausman 2 and S. Poulos USDA-ARS, Richard B. Russell Agricultural Research Center, Animal Physiology Research Unit, Athens, GA ABSTRACT: The present study examined the influence of dexamethasone (DEX) treatment on preadipocyte recruitment and expression of CCAAT/enhancing binding protein-α (C/EBPα) and peroxisome proliferator-activated receptor-γ (PPARγ) proteins in stromalvascular (SV) cell cultures derived from neonatal subcutaneous adipose tissue and semitendinosus muscles. One adipose tissue SV cell culture and one semitendinosus muscle SV cell culture were established from each of six young pigs (5 to 7 d of age). Conventional SV cell-culture procedures were used to digest adipose and muscle tissue and to harvest and culture adipose and muscle SV cells. Muscles were digested after the removal of all visible connective tissue from the excised muscle. One hour after seeding, muscle SV cell cultures were rinsed and refed new media to remove debris and insoluble muscle protein. The SV cell cultures were double-stained for lipid and the AD-3 antibody, a preadipocyte marker, at 1, 3, and 6 d and were double-stained for lipid and C/EBPα or PPARγ at d 6. Preadipocytes were randomly distributed and not clustered after 1 d in muscle and adipose SV cultures. Regardless of treatment, relative and absolute fat cell numbers were lower (P < 0.05) in muscle than in adipose-sv cell cultures. The DEX treatments produced similar magnitudes of increase in relative and absolute preadipocytes and adipocytes in muscle- and adipose-sv cultures. Several extracellular matrix substrata had no influence on adipogenesis in muscle-sv cell cultures. These studies indicate that muscle-sv cultures are characterized by a low number of adipocytes under basal conditions and a low number of glucocorticoid-responsive preadipocytes. Key Words: Adipose Tissue, Cell Culture, Cell Differentiation, Fat Cells, Skeletal Muscle, Transcription Factors 2004 American Society of Animal Science. All rights reserved. J. Anim. Sci : Introduction Marbling, or intramuscular adipose tissue, enhances juiciness, flavor, and overall desirability of meat and has been the focus of many studies attempting to improve meat quality (for review, see Wood et al., 1999). The influence of dietary conjugated linoleic acids on carcass traits and meat quality in growing pigs demonstrates the potential to increase intramuscular fat deposition while decreasing or maintaining subcutaneous fat deposition (Ostrowska 1 Mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the USDA and does not imply its approval to the exclusion of other products that may be suitable. 2 Correspondence: 950 College Station Road (phone: ; fax: ; ghausman@saa.ars.usda.gov). Received August 5, Accepted October 16, et al., 1999; Tischendorf et al., 2002; Wiegand et al., 2002). Indirect evidence indicated that conjugated linoleic acid feeding in pigs may induce the development or recruitment of intramuscular preadipocytes from stromal-vascular (SV) cells (Meadus et al., 2002). However, intramuscular preadipocyte development per se was not examined because analysis was restricted to adipocyte marker gene expression in muscle samples (Meadus et al., 2002). Nevertheless, these studies indicate that preadipocyte development in s.c. and i.m. depots may be regulated differently. The recruitment or developmental regulation of intramuscular preadipocytes in pig muscle-sv cell cultures has not been examined. Furthermore, the development of s.c. and i.m. preadipocytes has never been compared despite the marked differences in the ontogeny and regulation of s.c. and i.m. adipocyte cellularity in pigs (see review of Allen, 1976). We adapted the collagenase digestion protocol for adipose tissue stromal-vascular cells to whole neona- 429

2 430 Hausman and Poulos tal semitendinosus muscles (referred to herein as muscle-sv cell cultures) to quantify the number of intramuscular preadipocytes and their response to adipogenic agents. This approach permits the direct comparison of the response of muscle-sv and s.c. adipose- SV cells to adipocyte differentiation-inducing agents. We propose that muscle-sv cell cultures will contain fewer preadipocytes and will respond less to conventional adipogenic agents than s.c. adipose-sv cultures. Materials and Methods Postnatal pigs were killed at 5 to 7 d of age via intraperitoneal injections of 3 g of sodium pentothal solution followed by exsanguination. Stromal-vascular cells were isolated from subcutaneous adipose tissue and from semitendinosus muscles. Subcutaneous adipose tissue and semitendinosus muscles were both aseptically isolated and all visible connective tissue was removed. Tissues were finely minced and subjected to a 2-h digestion at 37 C in a shaking water bath. The digestion buffer included 100 mm HEPES (Sigma Aldrich, St. Louis, MO) buffer containing 120 mm NaCl, 50 mm KCl, 5 mm D-glucose, 1.5% type- V BSA (Sigma Aldrich), 1 mm CaCl 2,andatype-II collagenase (Sigma Aldrich) solution. The type-ii collagenase solution used to isolate stromal-vascular cells from subcutaneous adipose tissue contained approximately 6,250 collagen digestion units, whereas the solution used for semitendinosus muscle stromalvascular cells contained approximately 12,500 collagen digestion units. Semitendinosus muscle digesta were centrifuged at 500 g for 5 min. The remaining procedure was similar to a previously described method for isolating stromal-vascular cells from subcutaneous adipose tissue and was used for both semitendinosus muscle and subcutaneous adipose tissue (Hausman, 2000). Briefly, digesta were passed through sterile 180- m and 20- m sterile nylon mesh filters to isolate digested cells. Cells were rinsed with Dulbecco s modified Eagle s medium (DMEM) and subjected to centrifugation twice at 1,500 g for 10 min. Viable cells were counted using 0.4% trypan blue and a hemacytometer. Subcutaneous adipose tissue stromal-vascular cells were plated at a density of cells per 35-mm culture dish, and semitendinosus muscle cells were plated at a density of cells per 35-mm culture dish in DMEM supplemented with 10 ml/l fetal bovine serum (FBS; Sigma Aldrich) or 10 ml/l FBS and 80 nm dexamethasone (DEX; Sigma Aldrich). On d 3, media were aspirated, cells rinsed with DMEM and switched to media containing either 5 U/mL insulin, 5 g/ml transferrin, and 5 ng/ml selenium (ITS) or ITS+DEX (ITS + 20 nm DEX). Thus, three treatment protocols were used in this experiment:1)fbs+80nm DEX on d 0 to 3, followed by ITSond3to6;2)FBSond0to3,followedbyITS+ 20 nm DEX on d 3 to 6; and 3) FBS on d 0 to 3, followed by ITS on d 3 to 6, which represents the control or no DEX treatment. Semitendinosus muscle-sv cultures were routinely aspirated 1 h after plating and 2 ml of fresh media was added to each dish. A media change after 1 h of seeding and plating of muscle-sv cells was necessary to remove insoluble myofibrillar proteins and other insoluble debris. Additionally, preadipocytes and fibroblasts attach much earlier than myoblasts; thus, an h-1 rinse would favor preadipocyte and fibroblast attachment but preclude extensive myoblast attachment. Because we did not routinely rinse adipose tissue-sv cultures after 1 h, we examined the influence that an h-1 rinse had on adipose-sv cultures on several substrata in these studies. Adipose- and muscle-sv cells were harvested and cultured from the same pig in six experiments. In an additional four experiments, only muscle-sv cells were harvested and cultured. In preliminary studies, differentiating preadipocytes in muscle-sv cell cultures did not remain firmly attached, so laminin, fibronectin (FN), and type-iv collagen precoated dishes (BD Biosciences, Bedford, MA) were used with muscle-sv cells in four studies and laminin was used in an additional three studies of muscle-sv cells. In several of these studies, adipose- SV cells were also seeded and plated in FN and laminin dishes. Pig serum was used with muscle and adipose-sv cells in laminin-coated dishes (i.e., 5% pig serum ± 80 nm DEX on d 0 to 3, followed by ITS on d 3 to 6). Immunocytochemistry. We immunostained with the AD-3 monoclonal antibody to identify or mark preadipocytes (Hausman and Richardson, 1998). Cultures were also immunostained with the 5.1H11 monoclonal antibody (Developmental Studies Hybridoma Bank, Univ. Iowa, Iowa City) to identify myoblasts and myotubes because this antibody recognizes a cell surface antigen on myoblasts and myotubes. Cultures were fixed and reacted with AD-3 (1/200 of affinity purified IgG) or 5.1H11 (1/50 of affinity purified IgG) and stained with an ExtrAvidin peroxidase staining kit (Sigma Chemical Co.) as described by Hausman and Richardson (1998). Mouse peroxisome proliferator-activated receptor-γ (PPARγ) antibody and CCAAT/enhancing binding protein-α (C/EBPα) antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The epitope for the C/EBPα antibody (catalogue No. sc-61) corresponds to amino acids of rat C/EBPα, and this antibody reacts with the mouse, rat, and human protein. The epitope for the monoclonal PPARγ antibody (catalogue No. sc-7273) maps at the carboxy terminus of human PPARγ, and this antibody reacts with the mouse, rat, and human protein. Considerable Western blot and immunocytochemical analyses indicate that the C/EBPα and PPARγ antibodies also react with pig proteins (Hausman, 2000, 2003; Tchoukalova et al., 2000). Staining for the C/EBPα Santa Cruz antibody is blocked by a 2-h preincubation with control peptide (Santa Cruz

3 Porcine intramuscular preadipocytes 431 Figure 1. Muscle-SV cultures from d 1 double-stained for AD-3 and lipid and lightly counterstained with hematoxylin. Immunoreactivity was visualized by using a peroxidase staining kit. Lipid appears as dark-staining droplets in the cytosol. Note the single AD-3-reactive preadipocyte (white arrow) and the myoblast-like cell (black arrow). Also, note the variety of cell morphologies (stars), including some poorly attached cells, 250. Biotechnology) for the C/EBPα antibody (Hausman, 2000). Use of an unrelated primary antibody or second antibody alone showed little to no peroxidase staining. Immunocytochemistry was performed as described by Yu and Hausman (1998). Briefly, cultures were washed three times with 0.01 M PBS and fixed with 4% paraformaldehyde for 30 min. The cultures for C/ EBPα and PPARγ staining were permeabilized with PBS containing 3% Triton X-100 for 15 min and then incubated with either anti-c/ebpα antibodies (1/50) or anti-pparγ antibody (1/200). Reactivity was visualized by using an Extravidin peroxidase staining kit (Sigma Chemical Co.). The kits were used as recommended by suppliers. Double-staining simply involved lipid staining (no additional fixation) after peroxidase staining for either PPARγ, C/EBPα, AD-3, or the 5.1H11 antigen. On d 3 and 6, cultures were reacted for the C/EBPα, PPARγ, and the AD-3 antigen. Cultures were only stained for the 5.1H11 antibody on d 6. Three dishes of each treatment were stained at each time point. Evaluation of Immunoreactive Cells, Fat Cells, Fat Cell Clusters, Myotubes, and Total Cell Number. Cultures were routinely stained for lipid and counterstained as detailed elsewhere (Hausman, 1981). Three photomicrographs of each vessel were used for total cell counting. Fat cells were counted in photomicrographs of 2.2-mm 2 microscopic fields, whereas immunoreactive cells were counted in photographs of 1.3- mm 2 microscopic fields and 10 to 15 photographs of each dish were counted. Fat cell clusters and myotubes were counted in 6-mm 2 and 3.5-mm 2 microscopic fields, respectively, and six to eight microscopic fields of each dish were counted. Statistics. Data from the comparisons of adiposeand muscle-sv cultures (depots) were analyzed by a two-way ANOVA for main effects of depot, DEX treatment, and treatment depot interactions (SAS Inst. Inc., Cary, NC). Data from the comparisons of different substrata were analyzed by a two-way ANOVA for main effects of substrata, DEX treatment, and substrata treatment interactions. Differences between means were determined by least squares contrasts (SAS Inst. Inc.). Results Morphological Observations. In general, muscle-sv cells attached and spread slower than adipose-sv cells, resulting in some poorly attached and spread cellsinmuscle-svculturesatd1(figure1).throughout the culture period, muscle-sv cells remained morphologically distinct from adipose-sv cells in regards to the spreading and establishment of uniform monolayers. Preadipocytes (AD-3+ cells) were randomly distributed as isolated cells in adipose-sv and muscle-

4 432 Hausman and Poulos Figure 2. Muscle-SV cultures treated with either insulin, transferrin, and selenium (ITS)+dexamethasone (DEX; A), fetal bovine serum (FBS)+DEX (B), or pig serum+dex (C). Cultures were double-stained for AD-3 and lipid and lightly counterstained with hematoxylin on d 6. Immunoreactivity was visualized by using a peroxidase staining kit. Lipid appears as dark-staining droplets in the cytosol. Note that AD-3-reactive preadipocytes (A, white arrows) in ITS+DEX-treated cultures have little lipid. Clusters of preadipocytes with lipid are evident in both FBS+DEX and pig serum+dex-treated cultures (B and C, white arrows), A, B, C, 250. SV cultures at d 1 (Figure 1). Preadipocyte clusters were evident in muscle-sv and adipose-sv cultures by d 3, but they were fewer and larger in muscle-sv cultures. This divergence in number of preadipocyte clusters continued because the number of preadipocyte/fat cell clusters at d 6 was much greater in adipose-sv control (ITS) cultures than in muscle-sv control cultures (i.e., 4.7 ± 0.3 vs ± 0.4, means ± SEM of four to five experiments; t-test, P < 0.05). Also, the percentage of fat cells on d 6 was lower in muscle- SV than in adipose-sv control cultures (Table 1). Regardless of treatment media, a small number of myoblasts and small myotubes were observed on d 6 in muscle-sv cultures, and, as expected, myotubes were never observed in adipose-sv cultures. Immunocytochemical Analysis of Preadipocyte Development in Muscle-SV and Adipose-SV Cultures Treated from d 3 to 6 with Insulin (ITS) or ITS+DEX. Preadipocytes had little lipid and were clustered in muscle-sv cultures following treatment with ITS+DEX (Figure 2A). There were significant (P < 0.001) main effects of depot and DEX treatment and significant treatment depot interactions for the percentage of fat cells, the percentage and number of preadipocytes, the percentage and number of C/EBPαreactive cells, and the number of PPARγ-reactive cells (Table 1). The main effect of depot and treatment was significant (P < 0.001) for fat cell number, but the treatment depot interaction was not significant (Table 1). Treatment with ITS+DEX increased the absolute and relative number of preadipocytes approximately threefold in muscle-sv cultures and fivefold in adipose-sv cultures (Table 1). The absolute and relative number of C/EBPα-reactive cells were increased similarly by ITS+DEX in muscle-sv and adipose-sv cultures, whereas the absolute and relative number of fat cells was increased in adipose-sv cultures but not in muscle-sv cultures (Table 1). Treatment with ITS+DEX increased PPARγ-reactive cells to a much greater degree in adipose-sv cultures than in muscle-sv cultures (Table 1). The absolute and relative numbers of fat cells, preadipocytes, and C/ EBPα- and PPARγ-reactive cells after ITS+DEX treatment (Table 1) were 3 ± 0.4-fold higher in adipose-sv than in muscle-sv cultures. Fat Cell Cluster and Fat Cell Numbers in Muscle- SV and Adipose-SV Cultures Treated from d 0 to 3 with FBS or FBS+DEX Followed by ITS from d 3 to

5 Porcine intramuscular preadipocytes 433 Table 1. Number of preadipocytes (AD-3+), fat cells, CCAAT/enhancing binding protein (C/EBP)α+ cells, and peroxisome proliferator-activated receptor (PPAR)-γ+ cells in adipose tissue and muscle stromal-vascular (SV) cell cultures after seeding and plating in fetal bovine serum for 3 d followed by insulin, transferrin, and selenium (ITS) + dexamethasone (DEX) or ITS alone from d3to6 a Source of SV cells Adipose tissue Muscle Item Treatment ITS ITS+DEX ITS ITS+DEX AD-3+ cells/unit area 6 ± 1.1 bc 38 ± 1.3 d 2.7 ± 1.1 b 9.5 ± 1.3 c AD-3+ cells, % 1.3 ± 0.3 bc 8.6 ± 0.4 d 0.63 ± 0.3 b 1.96 ± 0.4 c Fat cells/unit area 6 ± 1.3 b 12.7 ± 1.5 c 1.1 ± 1.3 d 3.7 ± 1.5 db Fat cells, % 1.3 ± 0.3 b 2.5 ± 0.3 c 0.3 ± 0.25 d 0.86 ± 0.3 db C/EBPα+ cells/unit area 6.2 ± 1.8 bc 33 ± 1.8 d 2.1 ± 1.8 b 9.6 ± 1.8 c C/EBPα+ cells, % 1.5 ± 0.4 bc 6.5 ± 0.4 d 0.33 ± 0.4 b 2.1 ± 0.4 c PPARγ+ cells/unit area 5.4 ± 3 b 38 ± 3 c 1.3 ± 2.6 b 6.5 ± 2.6 b a Values are least squares means ± SEM for four to six experiments or replications for the AD-3 and fat cell counts and two to three experiments for the PPARγ and C/EBPα counts with three to four dishes in each experiment. Cultures were reacted for the AD-3 antigen, PPARγ, or C/EBPα after treatment for 3 d with either ITS or ITS+DEX, following seeding and plating in FBS from d 0 to 3. b,c,d Means in a row that do not have a common superscript differ (P < 0.05). 6. There were significant (P < 0.001) main effects of depot and DEX treatment and significant treatment depot interactions for the percentage and number of fat cells and fat cell cluster number (Table 2). Fat cell numbers and percentages and fat cell cluster numbers were increased by FBS+DEX-ITS in adipose-sv and muscle-sv cultures (Table 2; Figure 2B). Treatment with FBS+DEX induced 8 ± 1.4- and 6 ± 1.5- fold increases in fat cell cluster number in adipose- SV cultures and muscle-sv cultures, respectively. However, fat cell numbers and percentages and fat cell cluster numbers were three- to fivefold higher in adipose-sv than in muscle-sv cultures after FBS+DEX treatment (Table 2). Treatment with FBS ± DEX and pig serum (5%) ± DEX treatments induced a similar degree of fat cell cluster development in muscle-sv cultures (Figure 2C). In addition, several substrata were used to promote attachment in muscle-sv cultures (Figure 3) because fat cells in DEX-treated muscle-sv cultures on un- coated substrata were not attached well and tended to round up during the serum-free ITS period. The main effect of treatment was significant (P < 0.001), but there was no main effect of substrata and no substrata treatment interaction. These studies showed that DEX treatment increased fat cell cluster development in muscle-sv cultures to a similar degree on various substrata (Figure 3). In another attempt to improve cell attachment in muscle-sv cultures, muscle-sv and adipose-sv cultures were treated with ITS+DEX from d 3 to 6 following FBS+DEX (FBS+DEX-ITS+DEX). Compared to FBS+DEX-ITS, FBS+DEX-ITS+DEX treatment increased fat cell clusters in muscle-sv cultures by 87 ± 25% (6.9 ± 0.7 vs ± 0.5, means ± SEM of two experiments). In contrast, treatment of adipose-sv cultures with FBS+DEX-ITS+DEX had no influence on fat cell cluster number as has been reported (data not shown; Yu et al., 1997). All fat cells in DEX-treated muscle-sv and adipose-sv cultures were positive for C/EBPα (not Table 2. Absolute and relative number of fat cells and fat cell clusters in adipose tissue and muscle stromal-vascular (SV) cell cultures after seeding and plating in fetal bovine serum (FBS) + dexamethasone (DEX) for 3 d followed by insulin, transferrin, and selenium (ITS) alone from d3to6 a Source of SV cells Adipose tissue Muscle Item Treatment FBS, ITS FBS+DEX, ITS FBS, ITS FBS+DEX, ITS Fat cells/unit area 5.1 ± 1.6 bc 48 ± 1.6 d 1.5 ± 1.6 b 9 ± 1.6 c Fat cells, % 1.2 ± 0.7 b 11.5 ± 0.7 c 0.35 ± 0.7 b 2.4 ± 0.7 b Fat cell clusters/unit area 4.9 ± 2.3 b 41 ± 2.3 c 1.3 ± 2.3 b 7.1 ± 2.3 b a Values are least squares means ± SEM for four experiments or replications with three to four dishes in each replicate. Cultures were stained for lipid after seeding and plating in FBS alone or FBS+DEX from d 0 to 3 followed by treatment with ITS from d3to6. b,c,d Means within a row that do not share a common superscript are different (P < 0.05).

6 434 Hausman and Poulos Discussion Figure 3. Fat cell cluster numbers in fetal bovine serum (FBS) control (black bars) and FBS+dexamethasone (DEX)-treated (gray bars) muscle-sv cell cultures on various substrata. Cultures were stained for lipid after 3 d with insulin, transferrin, and selenium (ITS) following FBS+DEX from d0to3.values are least squares means ± SEM for four to six experiments or replications for fat cell cluster counts with three to four dishes in each experiment. a,b Bars with different letters indicate that FBS+DEX treatment increased fat cell cluster number on each substratum (P < 0.05). shown), AD-3, and PPARγ (Figures 2B,C and 4). Staining for lipid after reacting for either C/EBPα or PPARγ showed that lipid accretion was only evident in cells with C/EBPα- and PPARγ-reactive nuclei (Figure 4). In one of these studies, adipose-sv cells were rinsed early and treated with FBS-ITS+DEX. In two of these studies, adipose-sv cells were seeded and plated on laminin, rinsed early, and treated with FBS+DEX- ITS. As expected, early rinsing and/or laminin substrata had no adverse influence on the response of adipose-sv cultures to DEX treatment. Morphological and Cytochemical Analysis of Myogenesis in Muscle-SV cultures. Immunocytochemistry with the myoblast/myotube-specific monoclonal antibody 5.1.H11 delineated myoblasts and small myotubes from preadipocytes and other cells in muscle- SV cultures (Figure 5). Staining for lipid after reacting with the 5.1.H11 antibody showed that myoblasts and myotubes were devoid of lipid (Figure 5). Furthermore, myoblasts and myotubes were not reactive for the AD-3 antigen (data not shown). The percentage of total nuclei that were myotube nuclei was less than 1% on uncoated and FN substrata. Total Cell Number. Total cell number per unit of area was not influenced by treatment or source of SV cells. For instance, total cell numbers at d 6 ranged from 461 ± 20 to 414 ± 15 for adipose-sv cultures and from 448 ± 41 to 360 ± 50 for muscle-sv cultures (means ± SEM of four studies). These studies are the first to demonstrate the development of porcine intramuscular preadipocytes in vitro. Intramuscular adipose tissue-sv cell cultures were derived from older animals in studies of bovine intramuscular preadipocytes (Sato et al., 1996; Torii et al., 1998). Therefore, the current studies are also the first to demonstrate the development of preadipocytes in earlier stages of intramuscular adipose tissue development. Although treatment with ITS+DEX produced magnitudes of increases in C/EBPα- and PPARγ- reactive cells and preadipocytes in muscle and adipose-sv cultures, the absolute number of fat cells and resulting proportions of preadipocytes, fat cells, and C/EBPα-reactive cells were much lower in muscle- SV cultures. The extent of adipogenesis in ITS+DEXtreated muscle-sv cultures is obviously related to the remarkably low proportions of preadipocytes and adipocytes in basal (no DEX) muscle-sv cultures. Very low proportions of preadipocytes and adipocytes were also evident in basal cultures of adipose-sv cells derived from 50-d fetal pigs (Hausman, 2003). In 50-d fetal adipose-sv cultures, DEX induced a significant but very small increase in preadipocyte number whereas fat cell number remained unchanged. Dexamethasone induced a greater increase in preadipocyte number in 75-d fetal adipose-sv cultures but failed to increase fat cell number as in 50-d fetal adipose- SV cultures (Hausman, 2003). Therefore, muscle-sv cultures and early fetal adipose-sv cultures are similar in several ways, including an inability of DEX to increase fat cell number. Regardless, the diminished adipogenesis in muscle-sv cultures relative to young pig adipose-sv cultures may be attributable, in part, to lower proportions of preadipocytes/adipocytes in basal cultures and lower numbers of DEX-recruitable cells. Ligand binding studies indicate that glucocorticoidreceptor affinity and number increase with differentiation in pig adipose-sv cell cultures (Chen et al., 1995). Furthermore, reduced dexamethasone-driven adipogenesis in fetal adipose-sv cell cultures was associated with low-glucocorticoid-receptor affinity and number (Chen et al., 1995). Therefore, reduced dexamethasone-driven adipogenesis in muscle-sv cultures may be attributable to lower glucocorticoid-receptor affinity and number relative to adipose-sv cultures. However, examination of glucocorticoid receptor gene expression and/or ligand binding studies in muscle- SV cultures are necessary to support or substantiate this possibility. The very low proportions of preadipocytes/adipocytes in muscle-sv cell cultures may, in part, be due to the presence of myogenic cells (Mesires and Doumit, 2002; present study), which are apparently nonadipogenic. Furthermore, adipose-sv cell cultures on laminin substrata without serum contained preadipocytes (high laminin affinity) and DEX-recruitable cells (in-

7 Porcine intramuscular preadipocytes 435 Figure 4. Double-staining for lipid and peroxisome proliferator-activated receptor (PPAR)-γ at d 6 in fetal bovine serum (FBS)+dexamethasone (DEX)-treated muscle-sv cell cultures. Immunoreactivity was visualized by using a peroxidase staining kit. Lipid appears as the dark staining droplets (black arrows) in the cytosol. Note that PPARγreactive nuclei distinguish preadipocytes (white arrows) from unreactive or unstained cells ( ). Furthermore, PPARγ nuclear reactivity (white arrows) was associated with lipid accretion (black arrows). Stars ( ) indicate unreactive or unstained cells, 400. termediate laminin affinity; Yu and Hausman, 1998), whereas fetal muscle-sv cell cultures contained only myoblasts and preadipocytes under the same conditions (high laminin affinity; our unpublished observations). Therefore, the absence of a population of adipogenic (DEX-recruitable) cells coupled with the presence of nonadipogenic (myoblasts) cells may account for the low proportions of adipogenic cells in muscle- SV cultures. Several features distinguish the present study from studies of bovine intramuscular muscle cell cultures (Sato et al., 1996; Torii et al., 1998). For instance, preadipocyte lipid filling was dependent on either octanoate (Sato et al., 1996) or high levels of a thiazolidinedione (Torii et al., 1998) in bovine intramuscular- SV cell cultures. Analysis of preadipocyte differentiation in those studies did not include early markers of differentiation, in contrast to the present study, which precluded evaluation of preadipocyte recruitment per se (Sato et al., 1996; Torii et al., 1998). Furthermore, in those studies (Sato et al., 1996; Torii et al., 1998) and studies of cloned bovine intramuscular preadipocytes (Nakajima et al., 2002a,b; Aso et al., 1995), differentiation was not evaluated cytochemically for lipogenic enzymes and transcription factors. It is critical to demonstrate that lipid-accreting cells are, in fact, expressing lipogenic enzymes and transcription factors like PPARγ and C/EBPα, as we did in the present study. Furthermore, localization of transcription factors like PPARγ within the cell is critical to functionality because the activation of PPARγ involves translocation from the cytosol to the nucleus (Inoue et al., 2001; Jiang et al., 2000). Western blots for PPARγ in whole cytosol preparations cannot identify or localize cellular or intracellular sources of PPARγ immunoreactivity (Torii et al., 1998). In this regard, we clearly

8 436 Hausman and Poulos Figure 5. Muscle-SV control culture on fibronectin substratum immunostained for the 5.1H11 antigen followed by lipid staining and light counterstaining with hematoxylin on d 6. Immunoreactivity was visualized by using a peroxidase staining kit. Lipid appears as dark-staining droplets in the cytosol. Note the 5.1H11 immunoreactive small myotubes (white arrows) associated with nonreactive small adipocytes (black arrow). Nonreactive cells are indicated by a star, 200. demonstrated, herein, that nuclear PPARγ immunoreactivity in intramuscular preadipocytes was associated with the onset of lipid accretion. Implications A cell culture system was developed to culture intramuscular preadipocytes derived from neonatal pig muscle. Dexamethasone enhanced the development of both subcutaneous and intramuscular preadipocytes. Cultures of subcutaneous and intramuscular preadipocytes derived from the same pig can be used to identify agents or compounds that can differentially influence intramuscular and subcutaneous adipocyte development. Indirect influences of these agents can also be examined by treating subcutaneous and intramuscular preadipocyte cultures with serum from pigs treated with these agents. Identification of such agents will allow the development of treatment protocols to increase marbling deposition, while decreasing or maintaining subcutaneous fat deposition. Literature Cited Allen, C. E Cellularity of adipose tissue in meat animals. Fed. Proc. 35: Aso, H., H. Abe, I. Nakajima, K. Ozutsumi, T. Yamaguchi, Y. Takamori, A. Kodama, F. B. Hoshino, and S. Takano A preadipocyte clonal line from bovine intramuscular adipose tissue:

9 Porcine intramuscular preadipocytes 437 nonexpression of GLUT-4 protein during adipocyte differentiation. Biochem. Biophys. Res. Commun. 213: Chen, N. X., B. D. White, and G. J. Hausman Glucocorticoid receptor binding in porcine preadipocytes during development. J. Anim. Sci. 73: Hausman, G. J Techniques for studying adipocytes. Stain Technol. 56: Hausman, G. J Dexamethasone induced preadipocyte recruitment and expression of CCAAT/enhancing protein alpha and peroxisome proliferator activated receptor-gamma proteins in porcine stromal-vascular (S-V) cell cultures obtained before and after the onset of fetal adipogenesis. Gen. Comp. Endocrinol. 133: Hausman, G. J., and R. L. Richardson Newly recruited and pre-existing preadipocytes in cultures of porcine stromal-vascular cells: morphology, expression of extracellular matrix components, and lipid accretion. J. Anim. Sci. 76: Hausman, G. J Techniques for studying adipocytes. Stain Technol. 56: Hausman, G. J The influence of dexamethasone and insulin on expression of CCAAT/enhancer binding protein isoforms during preadipocyte differentiation in porcine stromal-vascular cell cultures: Evidence for very early expression of C/EBPα. J. Anim. Sci. 78: Inoue-Murayama, M., Y. Sugimoto, Y. Niimi, and H. Aso Type XVIII collagen is newly transcribed during bovine adipogenesis. Differentiation 65: Inoue, K., Y. Kawahito, Y. Tsubouchi, M. Kohno, R. Yoshimura, T. Yoshikawa, and H. Sano Expression of peroxisome proliferator-activated receptor gamma in renal cell carcinoma and growth inhibition by its agonists. Biochem. Biophys. Res. Commun. 287: Jiang, W. G., A. Redfern, R. P. Bryce, and R. E. Mansel Peroxisome proliferator activated receptor-gamma (PPARgamma) mediates the action of gamma linolenic acid in breast cancer cells. Prostaglandins Leukot. Essent. Fatty Acids 62: Meadus, W. J., R. MacInnis, and M. E. Dugan Prolonged dietary treatment with conjugated linoleic acid stimulates porcine muscle peroxisome proliferator activated receptor gamma and glutamine-fructose aminotransferase gene expression in vivo. J. Mol. Endocrinol. 28: Mesires, N. T., and M. E. Doumit Satellite cell proliferation and differentiation during postnatal growth of porcine skeletal muscle. Am. J. Physiol. Cell Physiol. 282:C899 C906. Nakajima, I., S. Muroya, R. Tanabe, and K. Chikuni. 2002a. Extracellular matrix development during differentiation into adipocytes with a unique increase in type V and VI collagen. Biol. Cell 94: Nakajima, I., S. Muroya, R. Tanabe, and K. Chikuni. 2002b. Positive effect of collagen V and VI on triglyceride accumulation during differentiation in cultures of bovine intramuscular adipocytes. Differentiation 70: Ostrowska, E., M. Muralitharan, R. F. Cross, D. E. Bauman, and F. R. Dunshea Dietary conjugated linoleic acids increase lean tissue and decrease fat deposition in growing pigs. J. Nutr. 129: Sato, K., N. Nakanishi, and M. Mitsumoto Culture conditions supporting adipose conversion of stromal-vascular cells from bovine intramuscular adipose tissues. J. Vet. Med. Sci. 58: Tchoukalova, Y. D., D. B. Hausman, R. G. Dean, and G. J. Hausman Enhancing effect of troglitazone on porcine adipocyte differentiation in primary culture: A comparison with dexamethasone. Obes. Res. 8: Tischendorf, F., F. Schone, U. Kirchheim, and G. Jahreis Influence of a conjugated linoleic acid mixture on growth, organ weights, carcass traits and meat quality in growing pigs. J. Anim. Physiol. Anim. Nutr. 86: Torii, S. I., T. Kawada, K. Matsuda, T. Matsui, T. Ishihara, and H. Yano Thiazolidinedione induces the adipose differentiation of fibroblast-like cells resident within bovine skeletal muscle. Cell Biol. Int. 22: Wiegand, B. R., J. C. Sparks, F. C. Parrish, Jr., and D. R. Zimmerman Duration of feeding conjugated linoleic acid influences growth performance, carcass traits, and meat quality of finishing barrows. J. Anim. Sci. 80: Wood, J. D., M. Enser, A. V. Fisher, G. R. Nute, R. I. Richardson, and P. R. Sheard Manipulating meat quality and composition. Proc. Nutr. Soc. 58: Yu, Z. K., J. T. Wright, and G. J. Hausman Preadipocyte recruitment in stromal vascular cultures after depletion of committed preadipocytes by immunocytotoxicity. Obes. Res. 5:9 15. Yu, Z. K., and G. J. Hausman Preadipocyte screening by laminin in porcine stromal vascular cell cultures. Obes. Res. 6: