Human mesenchymal stromal cells from adipose tissue of the neck

Mesenchymal stromal cells (MSC) have been introduced into the field of tissue-engineered airway transplantation. Since patients with extensive tracheal defects often require an open tracheotomy, this study investigated if MSC could be obtained from the adipose tissue of the neck during this procedure. Cells were isolated by plastic adherence from the adipose tissue of 8 patients. Cell isolates were analyzed for (i) proliferation, (ii) the expression of CD marker molecules and (iii) multilineage differentiation. The isolated spindle-shaped cells showed a high proliferation capacity and the flow cytometric analysis revealed a distinct population meeting the criteria for MSC. Using classical MSC cultivation protocols the characterized cells showed adipogenic, chondrogenic and osteogenic differentiation for all analyzed cell isolates. This study was able to demonstrate that sufficient amounts of stem/progenitor cells can be easily isolated from adipose tissue of the neck obtained during open tracheotomy. These cells may be a source for future tracheal replacement therapies.


Introduction
Airway stenosis can lead to atelectasis and severe obstructive pneumonia. To secure airway function tracheotomy is often necessary. Moreover, tracheomalacia due to trauma, severe inXammation or malignant diseases may require extensive surgical treatment. Various therapeutic approaches for stenosis and obstruction of the trachea or bronchi have been reported, including bronchoscopic dilatation, laser ablation, airway stenting and surgical bronchoplasty [1,2]. The current gold standard in surgical treatment of severe stenosis or defects is resection with end-to-end anastomosis. However, large airway defects (>5 cm) still present a major problem for clinicians because of the absence of eVective methods of treatment. Recent studies have focused on the replacement of the cervical trachea or main bronchi based on the concept of in situ-tissue engineering [3,4]. Attempts to provide synthetic grafts for application in humans have been associated with poor epithelization, exposure of the prosthesis matrix and poor functional outcome [5,6]. Allograft transplants such as aorta, functioning as a biologic incubator, may allow host cells to migrate into the airway transplant in vivo [7]. In particular, after malignant or infectious diseases lifelong immunosuppressive medication is contraindicated, but it is necessary to avoid rejection of allogenic material. This restricts the clinical application of allografts. Therefore, autologous transplants are favored for replacement strategies. Autologous bone marrow mesenchymal stem cells have recently been introduced into the Weld of tissueengineered airway transplantation. Besides autologous epithelial cells and chondrocytes, bone marrow-derived mesenchymal stem cells were used for the complete replacement of the left bronchus in a bioengineered human trachea [8]. Bone marrow mesenchymal stem cells have been found to diVerentiate along multiple mesenchymal lineages [9][10][11]. Even though it is commonly accepted that multipotent mesenchymal stem cells exist, there is clear evidence that isolation via plastic adherence results in heterogeneous populations including subsets of stem/progenitor cells with a contingently diVerent stem cell potential. To point out this heterogeneity multipotent Wbroblast colonyforming cells have been reclassiWed as mesenchymal stromal cells (MSC) [11]. Besides bone marrow, other sources of adult stem cells have been characterized, posting the hypothesis that stem cells capable of multilineage diVerentiation might reside in any post natal-organ [12]. Adiposederived MSC possess a particular relevance for tissue engineering settings because these cells can be harvested and handled easily and eVectively without donor site morbidity. In rats regeneration of the trachea using a bioengineered scaVold with adipose-derived MSC has already been reported [13].
This study investigated whether adult MSC can be retrieved during a tracheotomy from the adipose tissue of the neck. The cells were isolated from the adipose biopsies and characterized by their adherence to plastic culture dishes, their Wbroblastoid morphology, their ability to proliferate in culture and by means of Xow cytometric analysis. The stem/progenitor cell isolates showed characteristics of multipotent adult stem cells, and given the appropriate culture conditions, demonstrated multilineage diVerentiation into adipogenic, osteogenic and chondrogenic cells. Furthermore, there is strong evidence that these adipose tissuederived cells meet the requirements for MSC [11] and may therefore be applicable for clinical tissue-engineered airway transplantation in the future.

Isolation of human stem/progenitor cells
The isolation of human stem/progenitor cells from the adipose tissue surrounding the trachea was carried out with samples of eight patients, aged between 55 and 78 years, in accordance with the patients` informed consent (approval by the ethics committee, University of Lübeck, no. 05-218) and given indication for surgical tracheotomy. The biopsy was removed during the classic procedure of an open tracheotomy which was performed under general anesthesia in the operating room of the intensive care unit. Approximately 2-3 g of adipose tissue were removed from the neck as a routine procedure and immediately transferred to the laboratory for further processing. Adipose tissue samples were digested with basal medium supplemented with 1 mg/ml collagenase A (Sigma, München, Germany) for 45 min at 37°C, sieved, centrifuged, and washed with basal medium. The number of cells was determined and the cell suspension was plated onto 75 cm 2 tissue culture Xasks (TPP, Trasadingen, Switzerland). The number of cells obtained directly after isolation was around 4 £ 10 5 cells. Non adherent cells were removed by the Wrst medium change after 2 days. Single colonies of adherent Wbroblastlike cells were Wrst visible after 72 h of cultivation. All cultivations were performed at 37°C and 5% CO 2 .
To analyze adipogenic and osteogenic diVerentiation, isolated stem/progenitor cells were diVerentiated via monolayer protocols [10,15]. Adipogenic and osteogenic induction of the stem/progenitor cells was performed at 80-90% conXuence. To induce osteogenic diVerentiation cells were treated with osteogenic medium for 25 days. Osteogenic medium consisted of basal medium supplemented with 0.1 M dexamethasone (Merk, Darmstadt, Germany), 10 mM -glycerolphosphate (Sigma, München, Germany) and 300 M ascorbic acid (Sigma, München, Germany). To induce adipogenic diVerentiation cells were treated with adipogenic induction medium and adipogenic maintenance medium for 25 days. Induction medium consisted of basal medium supplemented with 0.5 mM 3-isobutyl-1-methylxanthine (IBMX Sigma, München, Germany), 1 M dexamethasone (Merk, Darmstadt, Germany), 200 M indomethacin (Sigma, München, Germany) and 2 M insulin (Sigma, München, Germany). Following a four-day induction period, the adipogenic induction medium was replaced with adipogenic maintenance medium consisting of basal medium supplemented with 2 M insulin for 3 days. This cycle was repeated three times and ultimately followed by a four-day period of adipogenic maintenance culture.
Lipid accumulation during adipogenic diVerentiation was demonstrated by Sudan III staining. Cells were washed with PBS followed by staining with a 0.2% solution of Sudan III (Sigma, München, Germany) in 70% ethanol. Alkaline Phosphatase (AP) activity of stem/progenitor cells diVerentiating along the osteogenic lineage was demonstrated using the AP staining kit (Sigma, München, Germany).

Quantitative analysis of histochemical staining
To analyze the diVerentiation of human adipose tissuederived stem/progenitor cells from the neck by AP or Sudan III staining, ten areas of 0.77 mm 2 for osteogenic diVerentiation and ten areas of 0.235 mm 2 for adipogenic diVerentiation were quantiWed per sample per day. The stained areas were measured in relation to the total area of cells using ImageJ software (NIH, Bethesda, MD, USA) and quantiWed in percent. Chondrogenic diVerentiation was analyzed by measuring Alcian blue-positive stained areas in relation to the total area of the sectioned MMB using ImageJ software (NIH, Bethesda, MD, USA) and quantiWed in percent.

RT-PCR analysis
Stem/progenitor cells diVerentiated via monolayer or MMB were collected at diVerent time points, washed twice with PBS and total RNA was isolated using a standardized RNA Isolation Kit (Macherey & Nagel, Düren, Germany). The RNA concentrations were determined by measuring the absorbance at 260 and 280 nm. Samples of 500 ng RNA were reverse transcribed using oligo-dT primer and Superscript II reverse transcriptase following the manufacturer's recommendations (Invitrogen, Paisley, U.K.). Aliquots of 1 l from the reverse transcriptase reactions were used for ampliWcation of transcripts using primers speciWc for the analyzed genes and Taq [21] and GAPDH (5Ј-CCG CAT CTT CTT TTG CGT CGC -3Ј, 5Ј-GCA ACT GTG AGG AGG GGA GAT TCA G -3Ј, 55°C, 1100 bp). Electrophoretic separation of PCR products was carried out on 2% agarose gels [2% (w/v) agarose (Roth, Karlsruhe, Germany)], 0.7 ng/ml ethidium bromide (Roth, Karlsruhe, Germany). The fragments were analyzed by computer-assisted densitometry in relation to GAPDH gene expression. The densitometric values of each marker were calculated in relation to GAPDH. From these values the highest of each marker was taken as 100%. Distilled water and no-RT reactions were always included as a negative control.

Statistical analysis
Statistical analysis was performed using SigmaPlot 2000 software (Systat, Erkrath, Germany) and calculated according to the student's t test. Samples were analyzed at least in three independent experiments (n = 3).

Successful establishment of stem/progenitor cells after tracheotomy
During an open tracheotomy stem/progenitor cells were isolated from subcutaneous adipose tissue of eight patients, aged between 55 and 78 years. Plastic adherent cell populations from adipose tissue of the neck exhibited a typical spindleshaped morphology (Fig. 1a). The Wrst colonies of Wbroblastlike cells could be observed after 72 h, with a 100% successful rate of stem/progenitor cell isolation. All cell isolates displayed stable growth characteristics, and STRO-1 could be detected in all isolates (Fig. 1b). Optimizing cell culture, an optimal cell growth and maintenance was achieved using a plating density of 1 £ 10 4 cells per cm 2 . During further passages, a daily doubling index of 2.02 was determined (Fig. 1c). The cells did not diVerentiate spontaneously during culture expansion into any morphologically identiWable cell type.
Multilineage diVerentiation of tracheotomy-derived stem/progenitor cells During multilineage diVerentiation and characterization of adipose tissue-derived cell cultures, the isolates were regularly analyzed by histochemical and immunoXuorescence staining for adipogenic, osteogenic and chondrogenic cell types. RT-PCR analysis conWrmed the multilineage diVerentiation of the isolated human stem/progenitor cells. We found that the amount of diVerentiated cells increased up to almost 70-90% after 25 days of cultivation in lineage-speciWc induction medium.

Adipogenic diVerentiation
To analyze the adipogenic diVerentiation, the stem/progenitor cells were induced in the expanded monolayer cultures by treatment with IBMX, dexamethasone, insulin and indomethacin. Morphologic changes as well as the formation of neutral lipid vacuoles were apparent as early as 1 week after induction and visualized by staining with Sudan III (Fig. 3a). Sudan III positivity was only observed after application of adipogenic medium. Initially, no staining was detectable in the stem/progenitor cell cultures. After 25 days of adipogenic induction, almost 70% of all cells were positive for the lipid staining and eventually Wlled with lipid vacuoles (Fig. 3a; 0 day vs. 25 days: p · 0.001).

Osteogenic diVerentiation
To assess the osteogenic diVerentiation, the stem/progenitor cells were induced by dexamethasone, glycerol phosphate and ascorbic acid in the presence of 10% FCS. As early as 1 week after induction, aggregates of cells showed expression of alkaline phosphatase (AP) which increased during the ongoing diVerentiation, resulting in almost 100% APpositive cells on day 25 ( Fig. 4b; 0 day vs. 25 days: p · 0.001). Immunostaining revealed the expression of the osteogenic marker molecules osteopontin and bone sialoprotein (Fig. 4b). Both molecules could be detected after 9 days of osteogenic diVerentiation and the number of positive cells increased during further cultivation, indicating osteogenic diVerentiation. Before application of the osteogenic medium, no protein expression of bone sialoprotein was observed in the undiVerentiated cell cultures by immunostaining, whereas a basal expression of osteopontin could be detected (data not shown).
These results were conWrmed at the gene expression level by RT-PCR analysis (Fig. 4c). Osteopontin was already expressed in undiVerentiated cell cultures, as it has Fig. 1 Hematoxylin-eosin (HE) staining of proliferating adipose tissue-derived cells from the neck (passage 4) isolated by plastic adherence after tracheotomy demonstrates a spindle-shaped morphology (a). The cells stained positive for STRO-1 (b) and showed a stable daily doubling index of 2.02 during further passages (c). Nuclei are stained blue using DAPI been previously demonstrated for MSC [22]. However, osteocalcin demonstrated an increasing expression after 9 days of osteogenic diVerentiation and continued to be expressed at a high level until the end of the cultivation period, indicating formation of osteoblasts.

Chondrogenic diVerentiation
To promote chondrogenic diVerentiation, we gently centrifuged the isolated cells to form a pelleted micromass, and cultured the micromass bodies (MMB) in a medium containing TGF-3 . DiVerentiation of tracheotomy-derived stem/progenitor cells along the chondrogenic lineage, using the classical MMB protocol, was characterized by Alcian blue-staining ( Figure 5 a; 0 day vs. 25 days: p · 0.001). All MMBs cultured in chondrogenic medium stained positive for Alcian blue after 18 days of cultivation demonstrating a proteoglycan-rich extracellular matrix. Immunostaining revealed the expression of collagen type II, the major component of hyaline cartilage (Fig. 5b). Collagen Type X, which is expressed by hypertrophic chondrocytes, was detected at the end of the 25-day diVerentiation period. Before application of chondrogenic medium, no protein expression of the analyzed chondrogenic marker molecules could be found in the undiVerentiated cell cultures by immunostaining (data not shown).
RT-PCR analysis conWrmed chondrogenic diVerentiation and expression of collagen type II and demonstrated the expression of aggrecan, the major proteoglycan of cartilage tissue (Fig. 5c). Taken together, these results demonstrate that all analyzed tracheotomy-derived stem/progenitor cells were able to diVerentiate along the chondrogenic lineage.

Discussion
Cells used in tissue engineering can be derived from numerous sources, including primary tissues, cell lines and stem cells. Stem/progenitor cells contribute to the homeostasis and regeneration of tissue after trauma and disease [23,24]. Many sources of adult stem cells have been characterized, posting the hypothesis that stem cells capable of multilineage diVerentiation might reside in any post natalorgan [12]. Among them, bone marrow MSC have been found to diVerentiate along multiple mesenchymal lineages [9][10][11] and have currently been used for the clinical transplantation of a tissue-engineered airway [8,25]. However, it is necessary to obtain these cells by general anesthetic procedures, bone marrow biopsies or even operations. These methods may complicate the routine clinical scenario.
Ideally, cells used for clinical transplantation should be (1) easy to harvest, (2) highly proliferative, (3) autologous and non-immunogenic and (4) able to diVerentiate into a variety of cell types with specialized functions [26]. Most patients undergoing extensive airway or head and neck surgery require open tracheotomy to secure airway passage until recovery of swallowing and respiratory function is reached. In this study we investigated if stem/progenitor cells applicable for tissue engineering of the airway function could be easily isolated from adipose tissue of the neck during an open tracheotomy.
Characteristics that deWne mesenchymal stem/progenitor cells include high proliferative potential, ability to generate primary colony-forming unit-Wbroblast and the ability to diVerentiate into bone, fat and cartilage. The multilineage potential of adult stem cells, however, has often been considered an attribute of heterogenic cell isolates [27,28]. Therefore, minimal criteria have been denominated to deWne human MSC [11]. First, the cells are plastic adherent and second, they express the CD marker molecules CD105 (endoglin), CD73 (5Ј-nucleotidase) as well as CD90 (Thy1) and do not express the leukocyte marker molecule CD45 and the marker molecule CD34 for primitive hematopoietic progenitor cells. However, current studies demonstrated that MSC cannot be distinguished from Wbroblasts by Xow cytometric analysis using a panel of common marker molecules [29]. Furthermore, various markers that are used to purify MSC from marrow aspirates are rapidly down regulated following ex vivo expansion [30] or have been attributed to diVerent subpopulations [31]. Therefore, the third criterion plays a pivotal role: MSC must show the multilineage capacity to diVerentiate into adipogenic, osteogenic and chondrogenic cells.
In this study we were able to demonstrate that proliferating cells could be easily isolated from adipose tissue surrounding the trachea by adherence to plastic culture dishes. The isolated cells showed a spindle-shaped morphology and expressed CD marker molecules characteristic for MSC [11,30] and adipose tissue-derived stem cells [32]. Moreover, these cells were capable of diVerentiating eYciently into adipogenic, osteogenic and chondrogenic cell types under speciWc culture conditions in vitro. In addition, it has been shown that multipotential adipose tissuederived stem cells accelerated neovascularization and epithelialization in an animal model of a tissue-engineered trachea [13]. Recently, there has also been evidence that Wbroblasts, closely related to MSC populations [33,34] enhanced epithelial diVerentiation and proliferation in an applied tracheal prosthesis [35].

Conclusion
This study was able to verify the hypothesis that stem/progenitor cells can be easily isolated from adipose tissue of the neck during tracheotomy. The stem/progenitor cell isolates showed characteristics of multipotent adult stem cells, similar to those of bone marrow-derived MSC, and given the appropriate culture conditions demonstrated multilineage diVerentiation into adipogenic, osteogenic and chondrogenic cells. Furthermore, there is strong evidence that these adipose tissue-derived stem cells from the neck meet the requirements for MSC [11]. Thus, additional operative interventions to obtain bone marrow MSC through multiple time-intensive general anesthetic procedures, biopsies and operations may no longer be necessary to receive MSC-derived tissue-engineered airway transplantation. Direct sources of autologous, highly proliferative, multipotent stem cells for clinical transplantation which can be safely extracted during a one-time procedure will be pivotal for regenerative medicine in the future. To safely apply stem/progenitor cells in future medicine it will also be necessary to understand the integration and interaction of these cells with surrounding cellular structures [36]. The goal of future studies is to optimize the diVerentiation and application of adipose tissue-derived stem/ progenitor cells from the neck for the reconstructive surgery of tracheal or bronchial airway. Ethical conduct of research The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.