Abstract
As a common procedure in dentistry for replacing a missing tooth, allogenic tooth transplantation has encountered many difficulties in the clinical application because of immunological rejection. It is hypothesized that immature dendritic cell injection might be a potential alternative method to avoid or alleviate immunological rejection in allogenic tooth transplantation. To test this hypothesis, a mouse model of allogenic and autogeneic tooth transplantation was to established test the immunosuppressive effect of immature dendritic cells (imDCs) derived from donor bone marrows on transplant rejection in allogenic tooth transplantation. 2 × 106 imDCs generated with 50 U/ml GM-CSF were injected to each recipient mouse by two ways: tail vein injection 7 days before transplantation or regional dermal injection at day 0 and day 3 after transplantation. Groups of autogeneic tooth transplantation and allogenic tooth transplantation without any treatment were set as control groups. The effects were evaluated with histopathology and immunohistochemistry. We found there was no obvious rejection in autogeneic tooth transplantation group; tail intravenous injection group showed obviously alleviated rejection while local injection group and none-treatment allogenic tooth transplantation group both showed severe rejection. Our results suggested that the rejection of allogenic tooth transplantation could be alleviated by tail vein injection of donor bone marrow-derived imDCs though it could not be completely eliminated. The clinical application of imDCs in allogenic tooth transplantation still needs further deep research.
Keywords: Allogenic tooth transplantation, immature dendritic cells, immune tolerance, T cells
Introduction
Autogenic tooth transplantation is now a common procedure in dentistry for replacing a missing tooth or the treatment of the complete luxation of teeth. Tooth transplantation offers one of the fastest and most economically feasible means of replacing a missing tooth in patients for whom implants and other prosthetic replacements are contraindicated for various reasons [1]. Besides, tooth extraction in orthodontics surgeries could provide abundant autogenic tooth supply for transplantation. However, the clinical application of allogenic tooth transplantation is restricted because of the immune rejection in the recipient which may induce inflammatory response and transplantation failure [2,3].
Dendritic cells (DCs) are the most powerful antigen-presenting cells in the body and play an important role in both autoimmunity and immune tolerance [4], previous studies have demonstrated that DCs can not only initiate the immunogenic pathway but also participate in tolerogenic pathway depending on their maturational state and their location [5]. It is generally believed that immature DCs modulate tolerance and mature DCs stimulated by inflammatory signals facilitate the activity of T cells and lead to inflammation [6,7]. Recent studies have found that imDCs derived from bone marrow could induce immune tolerance and avoid or reduce the degree of transplant rejection in a variety of organ grafting experiments, such as allogeneic skin graft [8], renal allograft [9] and corneal allograft [10], which made it a potential alternative method to avoid or alleviate immunological rejection in allogenic tooth transplantation, and up to now there still have no references in this area.
In the current study, we investigated the effect of imDCs purified from mouse bone marrow on tolerance induction in a mouse model of allogenic tooth transplantation.
Materials and methods
Mice
Male C57BL/6 (H-2b) and BALB/c (H-2d) mice between 6 to 8-week-old and weighing 18-22 g were purchased from Laboratory Animal Center of Chongqing Medical University. BALB/c mice served as tooth donors and C57BL/6 mice as allograft recipients of heterotopic tooth transplants. All experimental procedures were approved by the Ethical Committee of the Animal Center of Chongqing Medical University.
Reagents
Recombinant mouse granulocyte macrophage colony stimulating factor (GM-CSF), rat anti-mouse CD4 monoclonal antibody, rat anti-mouse CD8 monoclonal antibody and Histostain-SP Kit (Rabbit) were purchased from PeproTech (Rocky Hill, NJ, USA). Fetal calf serum (FCS) and mouse percoll were purchased from Gibco/BRL (Gaithersburg, MD, USA). PE anti-mouse CD11c, FITC anti-mouse CD80, PE anti-mouse CD86 and PE anti-mouse I-Ek were purchased from eBioscience (San Diego, CA, USA).
Generation of bone marrow-derived dendritic cells
DCs were prepared as described previously [11,12]. Femurs and tibiae were isolated from the hind limbs of C57BL/b mice with muscle tissue removed, then placed into 70% ethanol for 5 min for disinfection and washed with phosphate-buffered saline (PBS) in sterile petri dishes. Both ends of the bone were cut and the marrow was flushed out with RPMI 1640 medium. The marrow was disintegrated, filtered through a nylon screen and centrifuged in percoll. The cells were separated into three layers apparently and the middle lamella was extracted and washed. Cells were suspended and adjusted to 2 × 106 cells/ml in RPMI 1640 medium containing 10% FCS and 50 U/ml GM-CSF (based on the protophase experiment). 4 ml cell suspension was added to each well of six-well plates. Half of the medium was replaced every two days. Cell morphology was observed and recorded using a microscope (Olympus, Tokyo, Japan). Cells were harvested after 7 days culture and used for tail intravenous injection or local subcutaneous injection.
Flow cytometry
After 7 days of culture, the phenotype of DCs was examined by flow cytometric analysis. Cells were harvested by centrifugation and resuspended in phosphate-buffered saline (PBS) supplemented with 0.1% FCS and 0.02% sodium azide and then incubated with PE anti-mouse CD11c, FITC anti-mouse CD80, PE anti-mouse CD86 and PE anti-mouse I-Ek at 4°C for 30 minutes. After washing with PBS, cells were analyzed with the FACS Calibur flow cytometer (Becton Dickinson Immuno-cytometry Systems, San Jose, CA, USA) and data were analyzed using Cell Quest software.
Mouse model of allogenic and autogenetic tooth transplantation
Tooth transplant model was established as previously reported [13]. Lower incisors were extracted with a pair of dental forceps from C57BL/b and Balb/c donor mice respectively in aseptic condition with modification under anesthesia by an intraperitoneal injection of chloral hydrate (the maximum dose of 350 mg/kg), and the roots and pulp floor were resected with a surgical knife. After washed several times with physiological saline, the coronal portion of the resected samples without the periodontal tissue was immediately transplanted into the subcutaneously tissue right behind the right ear of Balb/c mice after sterilized with 75% ethanol, and the section was sutured with a nylon suture.
There were two sources for lower incisors used for transplantation: from C57BL/b mouse as allotransplantation donor and from Balb/c mouse as autotransplantation donor. Recipients were all Balb/c mice.
DC treatment and group sets
The recipients were divided into 4 groups with 10 Balb/c mice in each group as follows: group A, tail intravenous injection group, each recipient received intravenous injections of 2 × 106 imDCs diluted in 0.01 moL PBS 7 days before allogenic tooth transplantation; group B, local injection group, every recipient received local subcutaneous injection of 2 × 106 imDCs diluted in 0.01 moL PBS around the transplanted tooth at day 0 and day 3 after transplantation; group C, allogenic tooth transplantation group, recipients in this group did not receive any other treatment; group D, autogenic tooth transplantation group without any treatment either.
Histopathological analysis
Seven days after transplantation, grafts were extracted and then fixed with 4% paraformaldehyde for 48 hours. After decalcification in EDTA demineralization fluid for 1 week, the specimens were embedded in paraffin, and sagittal sections of transplants with surrounding skin tissues were cut into sections of 4 µm. Sections were stained with hematoxylin and eosin (H & E) for histopathological observation under a microscope (Olympus, Tokyo, Japan).
Immunohistochemical analysis
Sections from each group were dewaxed ,rehydrated, and endogenous peroxidase activity was blocked using 3% hydrogen peroxide in methanol. Sections were then pre-incubated with goat serum to reduce nonspecific staining and incubated with primary antibodies (including rat anti-mouse CD8 mAb and rat anti-mouse CD4 mAb) at 4°C overnight. After washing with PBS for 3 times, sections were incubated with biotinylated secondary antibodies for 15 minutes at 37°C and then washed in PBS before incubating with streptavidin-horseradish peroxidase (SA-HRP) conjugate for 15 minutes at 37°C. Then sections were again washed in PBS and the antibody-HRP complex was visualized by incubation with diaminobenzidine (DAB) for 5 minutes. Slides were briefly counterstained in hematoxylin, dehydrated and mounted Analysis of immunohistochemical results was carried out semi-quantitatively by using a microscope (Olympus, Tokyo, Japan) and Image Pro Plus software 6.0 (Media Cybernetics, CA, USA).
The cells stained pale brown was defined as positive staining, and 5 fields around the transplanted teeth were randomly chosen and observed under the high power microscope (× 400). The number of positive cells and total cells were counted using Image-Pro Plus 6.0 system and the positive percentage were calculated. Analysis of variance and q test (Newman-Keuls method) were used to calculate the statistically difference between the positive rates of CD4+T cells and CD8+T cells in each group by SPSS 19.0 statistical software. P < 0.05 was regarded as significantly different.
Results
Preparation and characterization of DCs
Non-adherent cells were removed 3 hours after cell seeded. The adherent cells were small, round in shape with a clear border. In a typical dissection about 1~2 × 107 leucocytes were isolated from each mouse. Complete medium with GM-CSF was added and increase cell volume and aggregated growth of the adherent cells could be seen after 2 days of culture (Figure 1A). On day 4, semi-adherent DC colony formed and small protrusions could be seen occasionally on its surface (Figure 1B). After 7 days of culture, the colony was reduced and DCs became semi-suspension and were released from the colony (Figure 1C).
Figure 1.

Morphological appearance of DCs under a microscope (original magnification × 100). A: At 2th day, cells show volume increase tendency, some cells formed small clusters composed of four to six; B: At 4th day, half-adherent DCs colony formed, small ecphyma on the surface could be seen; C: At 7th day, colony reduced, DCs were semi-suspension-like after release from the colony. Bar = 30 μm
After 7 days of culture, cell phenotype was analyzed by staining with fluorescein labeled antibodies and flow cytometry. Results showed high expression of CD11c, low expression of CD86 and MHCII while none-expression of CD80, which represented the phenotype of immature DCs (Figure 2).
Figure 2.
Flow cytometric detection of DC phenotype.
Clinical evaluation
On the first postoperative day, all recipient mice were alive and the graft sites were healing well with normal temperature and none swelling or infiltration existed. On postoperative day 3, tail intravenous injection group (group A) and autogenic tooth transplantation group (group D) showed continuous well healing, while the skin color of local injection group (group B) and allogenic tooth transplantation group (group C) turned to dark red with infiltration and high skin temperature than left ear was detected. On postoperative day 7, the graft sites of tail intravenous injection group (group A) healed well with normal skin temperature and none swelling, occasionally a few transparent scab could be found (Figure 3A left). The local injection group (group B) displayed congestion, red scar, liquid infiltration and higher skin temperature at the graft sites (Figure 3B left). The graft sites of the allogenic tooth transplantation group (group C) displayed purple skin, oncotic auricle, apparent higher skin temperature and there was a severe rejection occurred in one recipient mouse with allogenic teeth expelled out (Figure 3C left). The graft sites of autogenic tooth transplantation group (group D) showed no swelling, infiltration or crusting (Figure 3D left).
Figure 3.

Macroscopic observation on the 7th day after transplantation. A: Tail intravenous injection group. The transplanted place healed well. Tissue surrounded tooth compactly, slight hyperemia. B: Local injection group. The transplanted place hyperemia, incrustation. Tissue surrounded tooth flaccidly, hyperemia and hemorrhage. C: Allogenic tooth transplantation group. The skin of transplanted place present prunosus. Transplanted tooth exposed to the surrounding skin, obviously hyperemia and hemorrhage. D: Autogenic tooth transplantation group. The transplanted place healing well, without hyperemia and hemorrhage. Tissue surrounded tooth compactly, without obviously hyperemia and hemorrhage.
Further observation of the grafts after extraction from transplantation sites was executed. In group A, transplanted teeth were encapsulated tightly with skin tissue which displayed slight hyperemia (Figure 3A right). Hyperemia and hemorrhage could be seen in both group B and group C while the latter seemed more severe. The transplanted teeth were loose encapsulated by skin tissue in these two groups and in group C half of transplanted teeth were partly outside of the skin (Figure 3B right & Figure 3C right). In group D, the skin wrapped transplanted teeth tightly with no significant hyperemia and edema (Figure 3D right).
Histopathology results
Group A: skin and dental pulp around the transplanted sites were infiltrated with neutrophils and some lymphocyte. Angiectasis and hyperemia could be seen around the transplanted teeth. Local cell degeneration and mononuclear cell infiltration could be seen (Figure 4A).
Figure 4.

The HE staining histologic characteristics of experimental groups (× 400). A: Tail intravenous injection group. B: Local injection group. C: Allogenic tooth transplantation group. D: Autogenic tooth transplantation group. Bar = 100 μm
Group B: skin tissue surrounded the transplanted teeth were infiltrated with neutrophils, monocytes and a large number of lymphocytes. There was obvious angiectasis, hyperemia, and visible necrotic olistherozone (Figure 4B).
Group C: surrounding skin and dental pulp were infiltrated with mononuclear cells, plasma cells and a large number of lymphocytes. There was pulp necrosis and small abscess formation in dental pulp cavity. Angiectasis, hyperemia and hemorrhage could be seen in the tissues around the transplanted teeth with red blood cell infiltration (Figure 4C).
Group D: skin around the transplanted site showed normal morphology, with mildly dilated capillary vessel and neutrophil infiltration, few lymphocytes appeared. There was visible neutrophils infiltration in dental pulp cavity (Figure 4D).
Immunohistochemical results
Distribution of CD4+ and CD8+ cells were investigated for the spicemens in all groups. Brown labeling cells showed in the soft tissue around the transplanted teeth. The expression of CD4+ and CD8+ cells in group B and group C was stronger than group A. In group D, there is no obvious expression of CD4 and CD8 (Figure 5).
Figure 5.

The infiltration of CD4+T cells and CD8+T cells in every experimental group (IHC, × 400). A; Left, CD4 in tail intravenous injection group; right, CD8 in tail intravenous injection group. B: Left, CD4 in local injection group; right, CD8 in Local injection group. C: Left, CD4 in allogenic tooth transplantation group; right, CD8 in allogenic tooth transplantation group. D: Left, CD4 in autogenic tooth transplantation group; right, CD8 in autogenic tooth transplantation group. Bar = 50 μm
We analyzed the positive rates of CD4+T cells and CD8+T cells in each group. Results (Table 1) showed that rates of CD4+T cells and CD8+T cells were the lowest in autogenic tooth transplantation group (group D). The tail intravenous injection group (group A) has lower positive rates lower than both local injection group (group B) and allogenic tooth transplantation group (group C). There is no significant difference in the positive rates of CD4+T cells and CD8+T cells between local injection group (group B) and allogenic tooth transplantation group (group C).
Table 1.
Positive rate of CD4+T cells and CD8+T cells in each group (x̅ ± s)
| Group | CD4% | CD8% | CD4/CD8 |
|---|---|---|---|
| Tail intravenous injection group | 30.71 ± 1.83* | 17.34 ± 2.94* | 1.79* |
| Local injection group | 44.83 ± 2.32 | 23.87 ± 4.51 | 1.91 |
| Allogenic tooth transplantation group | 43.94 ± 2.06 | 23.04 ± 4.03 | 2.03 |
| Autogenic tooth transplantation group | 5.99 ± 3.24* | 3.62 ± 2.52* | 1.65* |
means differences between this group and allogenic tooth transplantation group were statistically significant.
Discussion
Allogenetic tooth transplantation as a common procedure for patients with a missing tooth usually causes rejection which may lead to graft failure and significantly hamper the clinical application. There have been several methods for inhibiting allogenetic tooth transplantation rejection up to now, including histocompatibility match test [14], root canal treatment before transplantation [15], removement of periodontal membrane of transplant teeth [16], freezing treatment of transplant teeth, irradiation of allogenetic teeth, treatment with fluorinated fluid [16] et al. These methods can extend the survival time of allogenetic tooth to various degrees though different imperfections still exist. For instance, histocompatibility match test limits the sources of transplant teeth; with root canal treatment the tooth may become fragile and change color because of the removement of dental pulp; removement of periodontal membrane will affect normal periodontal attachment, consequently affect the function of tooth; other methods such as irradiating and frozen processing can extend the survival time of allogenetic tooth, but fail to provide a fundamental solution to immune rejection. In general, there is still no widely used method to suppress the immune rejection in the practical application of allogenetic tooth transplantation.
As a professional antigen-presenting cell (APC), besides its immune stimulating function, DC can also play a central role in the development of T cell tolerance [17-19]. It is reported that the type and intensity of the immune response induced by DCs depends on their maturational state and their location. Mature DCs can stimulate T cell proliferation and induce specific T lymphocyte reaction [20]; imDCs with strong ability of antigen presenting could not active T cell due to the low expression of cell surface molecule CD40, CD80, CD86 and MHC which are necessary for T cell activation, consequently induce immune tolerance. Previous literatures demonstrated that imDCs derived from bone marrow could induce immune tolerance and avoid or reduce the degree of transplant rejection in a variety of organ grafting experiments, such as allogeneic skin graft [8], renal allograft [9] and corneal allograft [10], which made it a potential alternative method to avoid or alleviate immunological rejection in allogenic tooth transplantation, and up to now there still have no references in this area.
Here we compared the effect of imDCs inducing tolerance in allogenic tooth transplantation in a mouse model with two different imDCs treatment: tail intravenous injection and local subcutaneous injection. Results showed that tail intravenous injection apparently inhibited rejection while no obvious suppression effects on allogenic tooth transplantation in local subcutaneous injection. This suggested that different infusion methods affected clinical outcomes of imDCs treatment and this might because the number and paths of cell arrived lymph node are different depending on the infusion methods of imDCs. Morse et al [21] found that intracutaneously or subcutaneously injected DC might be partly transferred out of body and subcutaneously injected DC could not be detected in the local lymph node, this might explain why subcutaneous injection of imDCs had no effect on rejection in allogenic tooth transplantation in our study.
Through the result we found that tail intravenous injection of imDC could alleviate immunological rejection in allogenic tooth transplantation but the rejection still could not be completely suppressed. The changes in biological characters of the imDC, maturation during the migration, the type and the dose of derived cells as well as infusion time might be affecting factors of clinical effects. Besides, other tolerance mechanisms may be involved in the development of immune tolerance in allogenic tooth transplantation, which might affect imDC’s ability in inducing complete immune tolerance.
In conclusion, we investigated the effect of imDCs purified from mouse bone marrow on tolerance induction in a mouse model of allogenic tooth transplantation and results suggested that the rejection of allogenic tooth transplantation could be alleviated by tail vein injection of donor bone marrow-derived imDCs though it could not be completely eliminated. Subcutaneous injection of imDCs had no obvious suppression effects on rejection in allogenic tooth transplantation. Our study provided an alternative solution to suppress immunological rejection induced by allogenic tooth transplantation; however the clinical application still needs further deep research on the dose, injection method, time and possible complications of imDC infusion.
Acknowledgements
This project was supported by the Science Foundation of The Chong Qing Education Commission (KJ100302) and Program for Innovation Team Building at Institutions of Higher Education in Chongqing in 2013.
Disclosure of conflict of interest
None.
References
- 1.Kallu R, Vinckier F, Politis C, Mwalili S, Willems G. Tooth transplantations: a descriptive retrospective study. Int J Oral Maxillofac Surg. 2005;34:745–755. doi: 10.1016/j.ijom.2005.03.009. [DOI] [PubMed] [Google Scholar]
- 2.Zaleski M, Obersztyn A, Rymaszewska-Kossakowska T, Jedrzejczyk J, Sankowski A. Immunizing properties of allogeneic dental pulp grafts in rabbits. Transplantation. 1967;5:589–596. doi: 10.1097/00007890-196707000-00003. [DOI] [PubMed] [Google Scholar]
- 3.Goldstein BH, Tissot RG, Laskin DM, Cohen C. Histocompatibility and tooth transplantation in the rabbit. Oral Surg Oral Med Oral Pathol. 1975;39:929–933. doi: 10.1016/0030-4220(75)90113-9. [DOI] [PubMed] [Google Scholar]
- 4.Arnold B. Levels of peripheral T cell tolerance. Transpl Immunol. 2002;10:109–114. doi: 10.1016/s0966-3274(02)00056-4. [DOI] [PubMed] [Google Scholar]
- 5.Steinman RM, Nussenzweig MC. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. Proc Natl Acad Sci U S A. 2002;99:351–358. doi: 10.1073/pnas.231606698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Morel PA, Turner MS. Dendritic cells and the maintenance of self-tolerance. Immunol Res. 2011;50:124–129. doi: 10.1007/s12026-011-8217-y. [DOI] [PubMed] [Google Scholar]
- 7.Carreno LJ, Gonzalez PA, Bueno SM, Riedel CA, Kalergis AM. Modulation of the dendritic cell-T-cell synapse to promote pathogen immunity and prevent autoimmunity. Immunotherapy. 2011;3:6–11. doi: 10.2217/imt.11.38. [DOI] [PubMed] [Google Scholar]
- 8.Xin H, Yang W, Wang Q, You B, Tong Y, Peng Y. Immune tolerance of skin allograft transplantation induced by immature dendritic cells of a third party carrying donor antigens in mice. Transplant Proc. 2013;45:552–557. doi: 10.1016/j.transproceed.2012.07.161. [DOI] [PubMed] [Google Scholar]
- 9.Wang T, Xu L, Li H, Huang ZY, Zhang SP, Miao B, Na N. Immature CD4(+) dendritic cells conditioned with donor kidney antigen prolong renal allograft survival in rats. Chin Med J (Engl) 2012;125:2530–2537. [PubMed] [Google Scholar]
- 10.Gao XW, Fu Y, Li WJ, Du AJ, Li X, Zhao XD. Mechanism of immune tolerance induced by donor derived immature dendritic cells in rat high-risk corneal transplantation. Int J Ophthalmol. 2013;6:269–275. doi: 10.3980/j.issn.2222-3959.2013.03.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N, Schuler G. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods. 1999;223:77–92. doi: 10.1016/s0022-1759(98)00204-x. [DOI] [PubMed] [Google Scholar]
- 12.Zhang GX, Kishi M, Xu H, Rostami A. Mature bone marrow-derived dendritic cells polarize Th2 response and suppress experimental autoimmune encephalomyelitis. Mult Scler. 2002;8:463–468. doi: 10.1191/1352458502ms857oa. [DOI] [PubMed] [Google Scholar]
- 13.Riviere GR. Influence of serum on cultured murine tooth organ allotransplants. J Dent Res. 1984;63:1072–1074. doi: 10.1177/00220345840630081301. [DOI] [PubMed] [Google Scholar]
- 14.Ivanyi D, Kominek J. Tooth allografts in children matched for HLA. Transplantation. 1977;23:255–260. [PubMed] [Google Scholar]
- 15.Schwartz O, Andreasen JO. Allotransplantation and autotransplantation of mature teeth in monkeys: the influence of endodontic treatment. J Oral Maxillofac Surg. 1988;46:672–681. doi: 10.1016/0278-2391(88)90110-3. [DOI] [PubMed] [Google Scholar]
- 16.Shulman LB. Allogenic tooth transplantation. J Oral Surg. 1972;30:395–409. [PubMed] [Google Scholar]
- 17.Belz GT, Heath WR, Carbone FR. The role of dendritic cell subsets in selection between tolerance and immunity. Immunol Cell Biol. 2002;80:463–468. doi: 10.1046/j.1440-1711.2002.01116.x. [DOI] [PubMed] [Google Scholar]
- 18.Shklovskaya E, Fazekas de St Groth B. Balancing tolerance and immunity: the role of dendritic cell and T cell subsets. Methods Mol Biol. 2007;380:25–46. doi: 10.1007/978-1-59745-395-0_2. [DOI] [PubMed] [Google Scholar]
- 19.Steinman RM. The control of immunity and tolerance by dendritic cell. Pathol Biol (Paris) 2003;51:59–60. doi: 10.1016/s0369-8114(03)00096-8. [DOI] [PubMed] [Google Scholar]
- 20.Pereira SR, Faca VM, Gomes GG, Chammas R, Fontes AM, Covas DT, Greene LJ. Changes in the proteomic profile during differentiation and maturation of human monocyte-derived dendritic cells stimulated with granulocyte macrophage colony stimulating factor/interleukin-4 and lipopolysaccharide. Proteomics. 2005;5:1186–1198. doi: 10.1002/pmic.200400988. [DOI] [PubMed] [Google Scholar]
- 21.Morse MA, Coleman RE, Akabani G, Niehaus N, Coleman D, Lyerly HK. Migration of human dendritic cells after injection in patients with metastatic malignancies. Cancer Res. 1999;59:56–58. [PubMed] [Google Scholar]

