SEARCH

SEARCH BY CITATION

Keywords:

  • bone marrow stromal cells;
  • cell culture;
  • contractility;
  • bladder;
  • smooth muscle

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To assess the potential use of bone marrow stromal cell (BMSC)-seeded biodegradable scaffold for bladder regeneration in a canine model, by characterizing BMSCs and comparing them to bladder smooth muscle cells (SMCs) by immunohistochemistry, growth capability, and contractility.

MATERIALS AND METHODS

Bone marrow was taken by direct needle aspiration from the femurs of five beagle dogs for the in vitro study. Mononuclear cells were isolated by Ficoll-Paque® density gradient centrifugation and cultivated in medium 199 with 10% fetal bovine serum. BMSCs were characterized by cell proliferation, in vitro contractility, immunohistochemical analysis, and the growth pattern on small intestinal submucosa (SIS) scaffolds compared to bladder SMC cultures from the same dogs. Another six dogs had a hemicystectomy and bladder augmentation with BMSC-seeded (two), bladder cells including urothelial cells plus SMC-seeded SIS (two) and unseeded SIS scaffolds (two). The six dogs were followed for 10 weeks after augmentation.

RESULTS

In vitro BMSCs had a significant contractile response to calcium-ionophore, with a mean (sem) 36 (2)%, relative contraction (P < 0.01), which was similar to bladder SMCs but markedly different from fibroblasts. BMSCs also expressed α-smooth muscle actin by immunohistochemical staining and Western blotting, but did not express desmin or myosin. In vivo, both BMSC-seeded and bladder cell-seeded SIS grafts had solid smooth-muscle bundle formation throughout the graft.

CONCLUSIONS

BMSCs had a similar cell proliferation, histological appearance and contractile phenotype as primary cultured bladder SMCs. SIS supported three-dimensional growth of BMSCs in vitro, and BMSC-seeded SIS scaffold promoted bladder regeneration in a canine model. BMSCs may serve as an alternative cell source in urological tissue engineering.


Abbreviations
BMSC

bone marrow stromal cell

SM(C)

smooth muscle (cell)

SIS

small intestinal submucosa

XXT

2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-arboxanilide

DAPI

4′,6-diamidino-2-phenylindole, HCl.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The success of urological tissue engineering in creating suitable grafts for transplantation depends on a biodegradable scaffold that can support cell growth, and the availability of bladder cells for scaffold seeding. The ideal scaffold supports cell growth and acts as a matrix, allowing three-dimensional cell growth and seeding of different cell populations. Small intestinal submucosa (SIS) derived from porcine small intestine serves as a suitable scaffold; it contains many components required for normal cell growth, differentiation, and functioning, including: collagen, glycoproteins, proteoglycans, and functional growth factors [1,2], and preliminary work showed that these factors promoted cell growth and differentiation both in vivo and in vitro[3,4]. SIS also induced in vivo tissue-specific bladder regeneration in rats and dogs [5–7]. Additionally, the layered cell co-culture of urothelial cells and smooth muscle cells (SMCs) seeded together on SIS promote in vitro cell growth and differentiation [3].

The second necessary factor for in vitro graft creation is a readily available cell line capable of regenerating the desired tissue without problems of rejection, apoptosis and malignant transformation. In certain clinical situations, e.g. bladder exstrophy, bladder agenesis or bladder cancer, normal cell lines are unobtainable from the host bladder. Therefore, an alternative cell line for bladder SMC would be useful.

Bone marrow stromal cells (BMSCs) are obtained by a simple bone marrow aspiration and can be expanded in vitro. They are not stem cells, but have many features of stem cells and can differentiate into several mature cell types, including cardiac muscles [8], pneumocysts [9], hepatocytes [10], neural cells [11], skeleton muscle [12], bone [13], cartilage [14] and skin [15].

In the current study, BMSCs were evaluated as an alternative to replace bladder SMCs when native bladder muscle tissue is unavailable. The potential of BMSCs to differentiate into cells with bladder SMC characteristics were assessed in vitro and in vivo. We focused on the expression of SMC biomarkers, contractile function, growth patterns, and regenerative abilities of BMSC after being seeded on SIS.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Institutional Review Board and Institutional Animal Care & Use committee approvals for this project were obtained for animal surgery, including the collection of bone marrow and bladder tissue samples. Eleven adult male beagles weighting 10–12 kg were used; five provided bone marrow for in vitro characterization of BMSCs and six were used for the in vivo bladder regeneration study.

Five dogs were anaesthetized with 15–20 mg/kg sodium pentobarbital i.v.; approximately 15 mL of bone marrow was taken from one femur of each dog by needle aspiration and collected in a heparinized 50 mL test tube. Aspirated bone marrow was mixed with an equal volume of medium 199 (Invitrogen, Grandland, NY). A total of 30 mL bone marrow suspension was layered on top of 10 mL of Ficoll-Paque® (Amersham Pharmacia Biotech AB, Uppsala, Sweden) and centrifuged at 400 g for 35 min at 4°C. Mononuclear cells were collected from the interphase and plated in modified medium 199 supplemented with 10% fetal bovine serum (FBS, Invitrogen) at a 106 cells/mL. For controls, primary cultures of bladder SMCs were established from bladder biopsies from the same dogs and cultured in medium 199 plus 10% FBS. Cells were maintained at 37°C under 5% CO2. Non-adherent cells, including haematopoietic cells, were removed by washing in PBS 2 days after primary culture. The medium was changed every other day until the cells reached 95% confluence. Dog bladder SMCs and BMSCs were evaluated in parallel for cell proliferation, cell contractility, Western blotting, immunohistochemical assays and in vitro cell growth.

For cell proliferation studies, BMSCs were plated in triplicate at 2500 cells/mL in medium 199 containing 10% FBS. Aliquots of 200 µL cell suspensions were distributed into each well of 96-well tissue-culture plates. Bladder SMCs were plated in parallel as controls. The medium was changed every 2 days. Cell proliferation was determined using a 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-arboxanilide (XXT) cell proliferation assay kit II (Roche Molecular Biochemicals, IN, USA) according to the manufacturer's suggestion. The results were expressed as the multiple of the induction cell number, with the absorbance read at 450 nm with a plate reader (uQuant, Biotech Instruments Inc, Winooski, VT) at 1, 2, 3, 5, and 7 days of incubation.

To measure cell contractility in vitro, collagen lattice contraction was assayed in BMSC, bladder SMC, and human palmar aponeurosis fibroblasts, as previously reported [16]. Briefly, each type of cell was mixed with the soluble stabilized type I collagen (1 mg/mL, Upstate USA Inc, Charlottesville, VI, USA) at 1.5 × 105 cells/mL to create a cell-collagen solution. An aliquot (250 µL) of the cell-collagen solution was placed onto a 35-mm tissue culture plate (MIDSCI, St. Louis, MO, USA) and maintained for 5 days. The cell-collagen lattice was then mechanically released from the underlying plastic. The diameters of the cell-collagen lattices were measured before and after releasing, and the relative change calculated by dividing the lattice diameters at 10 min after release by the initial diameters. Each experiment consisted of at least three cell-collagen lattices. Lattices released in the presence of 10% FBS served as positive controls; lattices released under serum-free conditions served as negative controls. In addition, BMSCs, bladder SMCs and fibroblasts were treated with a Ca2+-ionophore (A23187, Sigma, St. Louis, MO, USA; 10−5m), to differentiate bladder SMCs from fibroblasts as previously reported [16].

To prepare protein samples for Western blot analysis, both types of cells were lysed with a buffer consisting of 10 mm Tris-HCl (pH 7.4), 150 mm NaCl, 5 mm EDTA, and 1% (v/v) Triton X-100 in the presence of a proteinase inhibitor cocktail (CompleteTM; Roche) at 100 µL/106 cells. They were then incubated on ice for 20 min. Soluble proteins were collected from supernatants after centrifugation at 12 000 g for 20 min at 4°C and stored at −80°C until use. Protein concentration was determined using the BCA Protein Assay reagents (Pierce, Rockford, IL, USA). From each sample, 20 µg of cellular protein was loaded onto 10% SDS-PAGE gel and transferred to polyvinylidene difluoride membranes (NUNC Life Science, Rochester, NY, USA). The membranes were incubated in 5% nonfat dry milk (Bio-Rad) in 25 mm Tris-HCl, pH 8.0, 125 mm NaCl, 0.05% Tween®20, to block nonspecific binding, and probed with a 1 : 400 dilution of mouse antihuman α-SM actin (Sigma) at 4°C overnight. The membranes were then incubated with 1 : 3000 dilution of peroxidase-labelled goat antimouse IgG (KPL, Gaithersburg, MD, USA) secondary antibody. The presence of protein bands was detected by an Enhanced Chemiluminescence Assay kit (Pierce).

BMSCs and bladder SMCs were stained by immunofluorescence using cells cultured on cover-slips. After fixation cells were stained with monoclonal primary antibodies against α-SM actin (1 : 1000), desmin (1 : 20), and SM myosin (1 : 100; all Sigma) as previously described [16], followed by rhodamine-conjugated goat antimouse secondary antibody. The cells were also were stained with the nucleic acid stain 4′,6-diamidino-2-phenylindole, HCl (DAPI).

For in vitro cell-seeded SIS, BMSCs were seeded and grown on the SIS mucosal side of cell-culture discs (vivo SIStm Inserts, COOK® Biotech. Inc., IN, USA) and placed in 12-well tissue culture plates in the presence of medium 199. The cells were maintained on SIS for 3, 7, 14, and 28 days. As a control, bladder SMCs were seeded on SIS disks under the same conditions as BMSCs. The cell-seeded SIS constructs were harvested, fixed with 10% formalin, and embedded in paraffin-wax overnight. Sections cut at 5-µm were stained for haematoxylin and eosin, Masson trichrome, and α-SM actin expression.

For bladder augmentation surgery, six adult male beagle dogs were divided equally into three groups: unseeded SIS (control), BMSC-seeded SIS and bladder urothelial cells plus SMCs. Harvesting, isolation and cultivation of BMSCs and SMCs is described above. Bladder urothelial cells were harvested from bladder biopsies and cultured in keratinocyte serum-free medium (Gibco/Invitrogen, Grand Island, NY, USA). This serum-free medium was supplemented with epidermal growth factor (5 ng/mL), bovine pituitary extract (50 µg/mL) and cholera toxin (30 ng/mL; all Gibco/Invitrogen) as described previously [3–4]. The BMSCs or bladder cells (urothelial cells plus SMCs) were cultured on the mucosal side of the SIS scaffolds in medium 199 with 10% FBS for 14 days. Six dogs had a hemi-cystectomy (40–50% of the bladder removed), followed either by bladder augmentation with SIS alone, with BMSC-seeded SIS, or bladder cell-seeded SIS. The scaffolds (5 × 7 cm) were anastomosed to the native bladder edge in a watertight manner with 5/0 chromic sutures. The edges of the graft were marked with nonabsorbable sutures for later graft identification. All the dogs were killed and the grafted bladders harvested 10 weeks after surgery. The graft sizes were measured when the bladders were filled with 20 mL of normal saline. The entire vesico-urethral complex, including the grafted area, was removed. The grafts were microscopically examined and pathological bladder changes noted. The harvested bladders were fixed in 10% formalin overnight, bisected longitudinally and horizontally, and embedded in paraffin wax for standard histopathology. The paraffin wax-embedded sections were stained with haematoxylin and eosin, Masson trichrome and by immunohistochemistry.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Primary cultures of BMSCs were successfully established from the aspirated bone marrow. Mononuclear cells isolated from Ficoll-Paque density gradients started to adhere and grow in colonies during the first 2 days. Each colony consisted of four to six cells, which were identified by phase-contrast microscopy (Fig. 1A). The BMSCs remained dormant for 3–5 days; they then began to multiply rapidly and form uniform spindle-shaped cells (Fig. 1B) morphologically similar to bladder SMCs. BMSCs showed proliferative potential and growth patterns similar to bladder SMCs (Fig. 2).

image

Figure 1. Phase-contrast microscopic appearances of the culture dog BMSCs. (A) colonies consisted of several cells on day 2. (B) the spindle-shaped cells on day 7. × 100.

Download figure to PowerPoint

image

Figure 2. Comparison of the cell proliferation of BMSCs (green dashed line) and bladder SMCs (red line) using bioreduction of XTT. Each point is the mean (sem) from three different experiments and is expressed as the multiple of the first day of cell proliferation. There was no difference in cell proliferation between BMSCs and bladder SMCs at each time during the 7 days of culture. *P < 0.05, Student's t-test.

Download figure to PowerPoint

Over 95% of cultured BMSCs stained positively for the SMC marker, α-SM actin; BMSCs expressed prominent α-SM actin-labelled thick filaments in the cytoplasm. These cells also showed more vigorous expression of α-SM actin than bladder SMCs (Fig. 3A,B). Western blotting further confirmed α-SM actin positivity in the cultured BMSCs. The intensity of α-SM actin was greater in the BMSCs than in the SMCs (Fig. 4). However, none of the BMSCs stained positively with antidesmin or antibodies against SM myosin, although SM myosin and desmin were both consistently present in cultured bladder SMCs (Table 1).

image

Figure 3. Double staining of α-SM actin and nucleic acid stain DAPI on both cultured bladder SMCs (A) and BMSCs (B). Nuclear labeling with DAPI is shown in blue and α-SM actin expression in red. Both cell types showed positivity for α-SM actin expression, BMSC showing brighter expression of α-SM actin than bladder SMC.

Download figure to PowerPoint

image

Figure 4. Western blot analysis of α-SM actin. Cellular proteins extracted from cultured BMSCs and bladder SMCs were size separated by SDA-PAGE and transferred to polyvinylidene difluoride membranes for detecting α-SM actin. The patterns of α-SM actin expression were similar in these two cell types, but the intensity was greater in BMSCs than in bladder SMCs.

Download figure to PowerPoint

Table 1.  Immunohistochemistry for SM markers in dog BMSCs and bladder SMCs (% cells stained)
Cell linesBladder SMCDog BMSC
α-SM actin >95>95
Desmin1–250
SM myosin1–250

Cell-collagen contractility was used to determine the contractile response of BMSCs, bladder SMCs and fibroblasts. The collagen-lattice diameter was reduced by all cell types within 10 min of being released. Prompt contraction followed the release of attached cell-collagen lattices in the presence of serum, with a rapid mean (sem) contraction of 37 (1)% for BMSCs, 27 (1)% for SMCs and 40 (4)% for fibroblasts. In the absence of serum (negative control), the relative contraction was 20 (1)% for BMSCs, 6% (0.3)% for SMCs, and 9 (4)% for fibroblasts. In the presence of 10 µm Ca2+-ionophore, the lattice showed a relatively rapid contraction of 36 (2)% for BMSCs 17% (1)% for SMCs, and 11 (2)% for fibroblasts. The contractility response of BMSCs in serum and Ca2+-ionophore conditions was significantly higher than in serum-free medium (P < 0.01), as for bladder SMCs (Fig. 5). However, there was no significant difference between the contractility response of fibroblasts in Ca2+-ionophore and serum-free conditions (Fig. 5). Carbachol, phenylephrine and KCl caused no significant contractile response compared to vehicle control for dog BMSCs, bladder SMCs and fibroblasts (data not shown).

image

Figure 5. Ca2+-ionophore promotes significant bladder SMC and BMSC contractility with collagen type I lattices, but not fibroblast contraction. Data are means (sem) of three sets of lattices. *P < 0.01, Student's t-test.

Download figure to PowerPoint

Growth pattern of cells on SIS matrix

Histologically, one to two cell layers of BMSCs were attached to the SIS matrix by 3 days. Over the next few days of incubation, BMSCs proliferated and formed several layers on the SIS matrix surface. Cell-matrix penetration occurred at 7 days. By 14 days there were many cell layers on the SIS surface, with evidence of increased matrix penetration. A cell-seeded SIS graft with several cell layers of matrix penetration was formed by day 28 (Fig. 6A). Dog bladder SMCs grew on SIS scaffolds in a similar pattern. Immunohistochemistry showed that surface and penetrating BMSCs and bladder SMCs expressed α-SM actin at all times (Fig. 6B).

image

Figure 6. Photomicrograph of cross sections of cell-seeded SIS when the cells seeded and penetrated the SIS matrix on day 28. (A) detection of BMSCs (red) grown on SIS scaffolds (blue) with intense cell penetration. Masson trichrome staining. (B) cultured BMSCs expressed α-SM actin.

Download figure to PowerPoint

Augmented bladder

A complete urothelial lining was evident in the augmented bladders of all six dogs. The graft expanded very well in a normal bladder-shaped pattern and the size of the graft remained stable in two of the six dogs. The two with well-expanded grafts came from one dog with a bladder-cell-seeded SIS and one with BMSC-seeded SIS graft. Moderate graft shrinkages were observed in the other four dogs, including two dogs with SIS alone, one with a bladder-cell-seeded graft and one with a BMSC-seeded SIS graft. A tiny calcification formation (≈ 0.2 × 0.3 cm) was found at the centre of the grafts in three of the six dogs; one from each group of grafted dogs. Small free-floating stones were found only in the two dogs with the unseeded SIS grafts. Histologically, the BMSC-seeded grafts had prominent SMC regeneration throughout the entire graft. SMC regeneration in the BMSC augmented bladder included SMC proliferation and muscle bundle formation at the edges and at the centre of the grafts. This was shown by α-SM actin staining (Fig. 7A). The bladder-cell-seeded grafts also showed SM regeneration with muscle bundle formation throughout the entire graft (Fig. 7B). By contrast, the unseeded control group showed sufficient SMC regeneration only at the edges of the graft, adjacent to the native tissue, and the graft centres contained few SMCs and lacked SM bundle formation (Fig. 7C). Also, there were some irregular bluish dystrophic calcifications in the submucosa of unseeded graft on Masson trichrome staining.

image

Figure 7. Immunohistochemistry for α-SM actin. The augmented bladders expressed α-SM actin in the enteric surface of the graft transplanted 10 weeks after cystectomy. Both BMSC-seeded (A) and bladder cell-seeded SIS scaffolds (B) showed SM regeneration with muscle bundle formation at the centre of the grafted tissue. (C) Some cells showed α-SM actin at the centre of the grafted bladder but lack of bundle formation in the controls.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Tissue-engineering technology provides an approach for rebuilding a functional urinary bladder for patients with congenital or acquired bladder defects. Various biodegradable scaffolds have been introduced for bladder reconstruction, and most appear to do well when seeded with cells before transplantation. Currently, the required urothelium and SMC for seeding are obtained by bladder biopsy and then expanded in vitro to give sufficient numbers to engineer the graft [17–19]. However, pre-existing fibrotic, inflammatory, or even malignant changes limit the use of the patients’ bladder cells, and make them unsuitable for culture and reimplantation. Recently, the limited usability of neuropathic bladder SMCs for tissue engineering; compared to normal bladder SMCs in culture, neuropathic SMCs showed an irregular and elevated cell proliferation, reduced cell adhesion, and inferior contractility [20]. Therefore, finding alternative cell sources for engineering functional bladders in the clinical setting is important.

Several alternative cell sources, including embryonic stem cells, bone marrow stem cells, BMSCs, and adult stem cells from other tissues have been investigated for use in tissue engineering. BMSC are attractive alternatives as they are obtained from the individual receiving the graft, thereby eliminating possible graft rejection. Also, cell procurement and expansion in vitro are feasible. Despite the heterogeneous population of cells obtained by crude cell aspiration, cells with sufficient plasticity were obtained, and which provided evidence of bladder tissue regeneration in the present study.

It is difficult to characterize BMSCs based only on their morphology in culture, because they have phenotypes similar to bladder SMCs and fibroblasts. However, previous data show that BMSCs have a strong α-SM actin expression equivalent to that of SM [21,22].

Interestingly, in the present study, BMSCs contained higher molecular levels of α-SM actin than did bladder SMC, confirmed by immunohistochemistry and Western blot analysis. Conversely, BMSCs did not express SM myosin and desmin. These results indicate that BMSCs and bladder SMCs are very similar, but further investigation is required to determine if BMSCs are progenitor cells or if they will differentiate into bladder SMC.

The cells with α-SM actin positivity in their cytoskeleton have potential for contractility. Cai et al. isolated BMSCs in lapine and canine models [22], and reported strong contractile capabilities of BMSCs after seeding them on a collagen-glycosaminoglycan analogue of extracellular matrix. The BMSCs showed a high content of muscle actin isoform, which increased with passaging, in the same cell population [21]. The authors concluded that the BMSCs serve as a mesenchymal stem cell for musculoskeletal tissues, including fibroblasts and muscle cells, which have contractile capabilities. To further determine whether BMSCs are similar to fibroblasts or bladder SMCs, BMSC contractility was tested using a similar lattice-contractility assay. Because bladder SMCs and fibroblasts both have α-SM actin in their cytoskeleton, one way to discriminate between the cell lines is to test their response to a Ca2+-ionophore. Bladder SMCs contract rapidly to a Ca2+-ionophore, whereas fibroblasts do not contract above their baseline level even in the presence of calcium. This strong contractile response to Ca2+-ionophore is currently the most reliable method of distinguishing a bladder SMC from a fibroblast [16]. In the present study, the same Ca2+-ionophore induced a significant contraction of bladder SMCs and BMSCs cultures, but not of fibroblasts. Therefore, this contractile lattice model established physiological evidence for the hypothesis that BMSCs have a contractile character similar to bladder SMCs. These data further verify that BMSCs resemble bladder SMCs more than they resemble fibroblasts.

Finally, BMSC had the same pattern of deep SIS penetration as previously noted with bladder SMCs, especially in the presence of urothelial cells. In the in vivo dog model, although four of the six dogs had some graft shrinkage, the remaining two, one with BMSC-seeded and one with bladder cell-seeded grafts, had excellent graft expansion. Also, the augmented bladder with BMSC-seeded grafts had better smooth bundle formation throughout the graft. This is comparable to bladder cell-seeded grafts after 10 weeks. However, the large unseeded grafts showed less SM regeneration and had stone formation within the augmented bladders. The current dog model is an example of the potential of tissue engineering for future clinical studies. It completes the circle, starting with obtaining progenitor cells from an individual to expanding, differentiating, and seeding them in vitro onto a scaffold, to finally transplanting the seeded graft into the patient to repair the primary defect. BMSCs could be especially valuable in tissue engineering because the cells are readily available and bypassing the problems of graft rejection.

This preliminary study provides in vitro evidence that cells obtained from bone marrow have biological characteristics similar to bladder SMCs. BMSC-seeded SIS scaffolds appeared to promote bladder regeneration with SMC bundle formation in a dog model. Further investigation is needed to determine the extent and mechanisms of BMSC differentiation into bladder SMC in vivo.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

This work was supported by a grant from the University of Oklahoma Health Sciences Center, College of Medicine Alumni Association Award.

CONFLICT OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

None declared. Source of funding: University of Oklahoma Health Sciences Center, College of Medicine Alumni Association Award.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES