| Sign In to gain access to subscriptions and/or personal tools. |
Neuroendocrine-Immune Surveillance of Osteosarcoma: Emerging Hypothesis
1 Division of Oral Biology and Medicine, CHS 63-090, UCLA School of Dentistry, Los Angeles, CA 90095-1668; Correspondence: *corresponding author, Chiappelli{at}dent.ucla.edu
Osteosarcoma is a bone-forming cancer predominantly found in children and adolescents more often than in adults. Osteosarcoma of the gnathic apparatus is relatively rare in the young population, and this condition becomes a concern of clinical dentists for predominantly the middle-aged and aging patient groups. Osteosarcomas are invaded by lymphocytes, which exhibit signs of activation. The immune processes that are engaged within the malignant bone matrix involve the production of cytokines, which regulate the process of apoptotic programed cell death. This paper discusses the mechanisms by which apoptosis of osteosarcoma cells is modulated by the neuroendocrine-immune system, and potential physiological implications.
Key Words: osteosarcoma • neuroendocrine-immune • apoptosis
Osteogenic sarcoma (osteosarcoma, OS) is a bone-forming cancer, and the most common primary malignancy of the skeleton, and is located predominantly in the metaphysis, the growing terminals of the bone. The natural history of OS has not changed over time, and fewer than 30% of patients with localized resectable primary OS tumors, and who are treated by surgery alone, can be expected to survive free of relapse. However, novel treatment interventions that involve aggressive chemotherapy to supplement surgery increasingly ensure close to a 75% five-year survival rate (Schajowicz et al., 1995; Murphey et al., 1997). The site of primary OS is a significant prognostic factor. Axial skeleton OS tumors have the greatest risk of progression and death and must be resected surgically in toto. While the prevalence of OS tumors of the cranial skeleton is rare, it still represents one of the most common sarcomas of the head and neck regions. Distant metastasis, albeit rare, can involve lung, cervical lymph nodes, spinal column, and brain (Clark et al., 1983; Odell, 1996; Smeele et al., 1997; Chen et al., 1999). Maxillary or mandibular OS is associated with a typical symptomatology. Patients complain of swelling and pain on the average of three to four months before consulting a physician or a dentist. Typically, OS involves the body of the mandible (51% prevalence), with osteolytic lesions, or the alveolar ridge of the maxilla (49% prevalence), with osteoblastic lesions (Clark et al., 1983; Chen et al., 1999). The literature to date is mixed in terms of the ratio of mandibular to maxillary OS metastases, although it is clear that these patterns of metastasis must follow the pathways of lymphatic drainage. The vessels of the incisor and canine teeth drain into lymphatics that pass forward to join the anterior facial lymphatics that drain into the submental and the submandibular nodes, and eventually the deep ring of cervical lymph nodes (Trieb et al., 1998). Chemotherapy and immune-directed therapies for patients with OS have been proposed on the basis of recent developments in our understanding of tumor surveillance. Immunotherapy for cancer in general and OS in particular can be traced back about one century, with attempts to use a prepared immune serum (Bieling et al., 1996; Smeele et al., 1997; Bielack et al., 1999; Bacci et al., 2000; Chiappelli et al., 2001a). Today, immune-based therapy rests on the use of various immunopotentiating agents, such as whole bacterial cells, bacterial cell wall fractions (e.g., lipopolysaccharide, LPS), cytokines, and thymic humoral factors. Our limited understanding of the fundamental systemic and molecular events that control and regulate immune processes renders the therapeutic efficiency of immunotherapy for OS and for all other tumors erratic and unreliable. The general consensus from the literature is that beneficial effects from immune therapies can be obtained only in conjunction with other procedures, and that, alone, immunotherapeutic approaches are not sufficient to ensure cure, increase survival time, or improve the quality of life of patients with OS. Furthermore, the consensus states that a concerted effort must drive to a more comprehensive understanding of the underlying cellular and molecular modulation of immune processes if novel and improved modes of intervention for OS are to be developed (Ben-Efraim, 1996).
Recent improvements in outcome have been obtained by the combining of neoadjuvant chemotherapy with radical surgery, despite the observation that a substantial number of patients appear to be resistant to chemotherapy. Concerted efforts have focused upon the development and testing of alternative approaches (Ioannides and Whiteside, 1993; Trieb et al., 1998). Research suggests that OS cells may escape natural killer immunosurveillance. Phenotypic characterization of this immune cell subpopulation, which consists of 1-5% of the peripheral blood mononuclear cells, is negative for the cluster of differentiation (CD) #3, positive for the Fc receptor, CD16, and for the NK receptor, CD56 (CD3-CD16+CD56+). Experiments performed in vitro have indicated that human OS-established cell lines, such as U2 OS and Saos-2, are resistant to natural cell-mediated cytotoxicity (Tarozzi et al., 1995). By contrast, immunohistochemical studies of paraffin sections of high-grade OS biopsies have established that OS is heavily infiltrated by CD3+ (33/35, 95%) and with cytotoxic CD8+ T-lymphocytes (24/35, 68%). Close to one-third of these lymphocytes expressed the marker of activation, HLA-Dr, which signifies that HLA-Dr is significantly overexpressed (p < 0.05) in OS sections, compared with normal control bone and with non-malignant bone tumors. Indeed, HLA-Dr expression in OS-infiltrating lymphocytes was positively correlated (p < 0.05) with the severity and the duration of the symptoms (p < 0.05). It is possible and even probable that increased HLA-Dr expression in OS-infiltrating lymphocytes is indicative of actively ongoing immune activity against the tumor (Trieb et al., 1998). It is possible and even probable that certain cytokines within the OS tumor may result in altered programed cell death (apoptosis) by OS tumor cells. Apoptosis is a major element in physiologic cellular control and is directed by immune production of a specific set of cytokines. Cell death in T-cells, one of the principal mechanisms for maintaining peripheral tolerance and for limiting an ongoing immune response, is initiated by antigen engagement of the T-cell receptor, and mediated through Fas/Fas ligand (FasL) interactions. One pathway of apoptosis engages the plasma membrane, another apoptotic pathway engages the mitochondrion. The plasma membrane can be engaged either by specific receptors (e.g., "death domain, TNF-receptor family [TRAIL and others], Fas-Fas-Ligand interaction), or by perforin-like pores induced by Granzyme-B. These independent plasma membrane pathways for apoptosis converge in the activation of cysteinyl aspartic acid proteases (caspases), which act in a cascade. Upstream caspases are caspase-2 and caspase-8, which initiate the cascade that leads to caspase-3 activation. The mitochondrial pathway can be engaged by U.V. irradiation, steroids, chemicals, or toxic pharmacological agents. It results in a change in the mitochondrial membrane potential, increased porosity of the membrane, and leakage of certain contents from the mitochondrion, including cytochrome C. Cytochrome C in the cytoplasmic compartment engages the apoptosis activating factor 1 (APAF-1), which participates in caspase 9 activation. This pathway joins the plasma membrane pathway when caspase 9 activates caspase 3. Caspase 3 activates the nuclear nuclease (caspase-dependent nuclease, CDN), which is responsible for the characteristic DNA fragmentation signature of apoptotic cell death. During apoptosis, an early and ubiquitous event is the exposure of phosphatidylserine, such as Annexin V, a calcium-dependent phospholipid binding protein, at the cell surface (Hengartner, 2000; Chiappelli et al., 2002; Prolo et al., 2002). These elements of the literature, taken together with the observation that, in chemosensitive leukemic T-cells and certain solid tumors, drugs, including adriamycin, CDDP, cyclophosphamide, methotrexate, or vincristine, either alone or in combination, mediate the induction of apoptosis by activation of the CD95/APO-1/Fas system, led to test this response experimentally in OS. Analysis of the data showed that these cytotoxic drugs induced many of the biochemical and morphological alterations characteristic of apoptosis in certain OS cell lines (e.g., HOS, SaOS), but by a pathway distinct from the traditional CD95/APO-1Fas system (Fellenberg et al., 2000; Seki et al., 2000).
As a test of the hypothesis that chemotherapeutic drugs may act upon OS tumor cells by altering cell replication, rather than or in addition to inducing programed cell death, the effects of the common chemotherapeutic drug colcemid were tested upon cell-cycle traversal of the transformed human OS cell line (MG63). Colcemid treatment (50 ng/mL) arrested the growth of MG63 cells, but failed to alter c-myc, p53, and p21/WAF gene expression, which together act to foster progression through the cell cycle. Analysis of the data showed that colcemid led to a significant increase in mitochondrial markers of apoptosis (e.g., bcl-2, bax-
An emerging body of evidence suggests an alternative approach to modulate apoptotic cell death in OS cell populations. Research in mind-body interaction has established that psychoneuroendocrine and immune system responses are intimately intertwined, and that these interactions have profound relevance to cancer as well as to oral biology and medicine (Chiappelli et al., 2001a). Sympathetic and cranial innervations play a significant role in the modulation of neuro-immunosurveillance of a variety of cancers, and may have an impact on cell growth, replication, and apoptosis in OS. Autonomic sympathetic nerve fibers, which are found in the periosteum and the endosteum, and often the cortical bone as well, are associated with the neurovasculature. In the instance of the cranial and the facial skeletons, autonomic fibers arise from the stellar and the superior cervical ganglions. Indeed, research has established that superior cervical ganglionectomy in experimental animals (female Wistar rats) results in a significantly hampered mineralization of the mandibular bone over a period of a month (Ladizesky et al., 2000). Cranial nerves also play a critical immunomodulatory role in the head and neck region. Glossopharyngeal afferents convey sensory information to the central nervous system. We have shown that the glossopharyngeal nerve may also act as a neural channel, perhaps in part via its parasympathetic fibers, for information about localized immune reactions in the soft palate, Waldeyer ring of tonsils, and posterior aspects of the tongue to reach the central nervous system (Romeo et al., 2001). It is now clear that apoptotic programed cell death induced by the T-cell receptor and mediated through Fas/FasL interactions is down-played by the neuropeptides, VIP and PACAP. This effect is mediated by specific receptors (VPAC1, VPAC2), whose concerted effect results in the reduction specifically of FasL expression at the protein and mRNA levels (Delgado and Ganea, 2000a,b). It is possible and even probable that neuropeptides produced at the nerve endings of the terminal branches of the trigeminal cranial nerve (e.g., infra-alveolar branches of V2 and V3) and the sensory terminals of the periosteum modulate apoptosis of OS-infiltrating lymphocytes and OS tumor cells. It is also possible that endogenous or exogenous endocrine products, such as glucocorticoids (e.g., circulating cortisol, pharmacotherapeutic administration of the synthetic glucocorticoid, dexamethasone [DEX]), often used to contain inflammatory reactions, swelling, and pain reported by patients, could play a significant role in OS cells’ programed cell death. It is now well-established that glucocorticoids modulate immune cells and OS cells by receptor-mediated mechanisms. Steroids regulate the expression of certain cytokine mRNAs and their production (e.g., interleukin [IL]-6, IL-2) (Chiappelli et al., 1994), and IL-2 or IL-6 (100 IU/mL) and other cytokines lead to a significant increase in the expression and the binding characteristics (number, kDa) of the glucocorticoid receptors in lymphocytes and in established OS cell lines (e.g., Sa-Os2). In fact, the number and binding of the glucocorticoid receptors were significantly decreased in OS cells constitutively endowed with high levels of receptor (e.g., MG-63 cells), and whose growth medium was depleted of IL-6 with specific anti-IL-6 antibody (100 ng/mL). Therefore, we inferred and demonstrated that IL-2 and other cytokines act in concert with steroid hormones to regulate specific responses in OS (Sartori et al., 1998; Angeli et al., 1999, 2002; Masera et al., 2000; Dovio et al., 2001). It is interesting to note that the basal number of type II glucocorticoid receptor in Sa-Os2 cells (17,800 ± 4600) (Sartori et al., 1998) lies in the range of the human cutaneous T-cell lymphoma line, Hut-78 (22,000 sites per cell) (Chiappelli et al., 1994). Following treatment with IL-2 (100 IU/mL, 20 hrs), the number of type II glucocorticoid receptors in the SaOS line rose to about 35,000 ± 6800 (Sartori et al., 1998), which lies in the range of the human acute T-lymphoblastic leukemia line, MOLT (31,000 sites per cell) (Chiappelli et al., 1994). The physiological relevance of this outcome is confirmed by experimental data, which together show that increased binding of steroids in OS cells alters bone remodeling processes by decreasing bone formation and increasing bone resorption. Pro-inflammatory cytokines, such as IL-1β and IL-6, may be involved in bone resorption by activating immature osteoclasts, and by hindering mitogenic effects on osteoblasts. Cortisol decreases the expression of IL-6 as well as IL-1β in human osteoblast-like cells. Taken together, these lines of evidence may shed light on the biphasic pattern of bone loss in the course of therapy in patients with elevated levels of circulating IL-6 who are given exogenous glucocorticoids (Swolin-Eide and Ohlsson, 1998; Dovio et al., 2001).
Cytokines may alter the balance between bone resorption and bone formation by several other pathways. For instance, growth hormone administration increases bone resorption and results in enhanced bone formation. Treatment of human osteoblasts in vitro with growth hormone yields a significant increase in IL-6 production, which mediates increased bone resorption (Swolin and Ohlsson, 1996). The expression of osteoprotegerin, the soluble receptor for the osteoprotegerin-Ligand, which mediates the signal for osteoclast differentiation, is significantly increased on osteoblastic cells in vitro (e.g., MG-63) by IL-1
Based on these elements of the literature, we have formulated the hypothesis that a cytokine-induced increase in glucocorticoid receptor could also modulate OS in situ by inducing OS tumor cells to engage in apoptosis. Based upon the observations that cytokines, including IL-2, interferon-
Increased cytokine gene expression and production associated with immune activation by OS-infiltrating lymphocytes may lead to a rise in the maturation of T-cell-mediated anti-OS cytotoxicity, and in the generation of lymphokine-activated killer cells directed against OS. Immune surveillance of OS may be modulated in situ by the neural and the endocrine systems. Indeed, leptin, which is elevated in subjects with OS (Kushlinskii et al., 2000), plays a significant role in regulating circadian neuroendocrine (e.g., cortisol) and cytokine (e.g., IL-6, IFN) production (Prolo et al., 1998). Taken together, these avenues of basic research suggest novel interventions for patients with mandibular and maxillary OS, or with OS outside of the confines of the head and neck region. Translational research (e.g., RBT’s) for neuroendocrine-immune control of OS must now be developed and tested.
Received for publication December 2, 2002. Accepted for publication February 27, 2003.
Journal of Dental Research, Vol. 82, No. 6,
417-421 (2003)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
), confirming the growing body of literature that supports apoptotic cell death of OS cells mediated by chemotherapeutic agents (
(IFN), and IL-10, acted to increase the number and binding efficiency of glucocorticoid receptors significantly in Sa-Os2 cultures (
