**5. Pathophysiological functions of TGF-βs in the nervous system**

In addition to their role in normal functioning of the nervous system, TGF-βs have also been implicated in different mechanisms under pathophysiological conditions. TGF-βs have been shown to be injury-related proteins. They were suggested to be neuroprotective following ischemia as well as in various neurological diseases. Furthermore, TGF-βs were implicated in glial scar formation. Finally, their role in brain tumor formation will be discussed.

#### **5.1 Neuroprotective function in brain ischemia**

136 Neuroscience – Dealing with Frontiers

The potential involvement of TGF-β in central reproductive regulation is also an emerging topic. Gonadotropin-releasing hormone neurons in the preoptic area contain TGF-β receptors as well as SMAD2/3 suggesting that they are fully capable of responding directly to TGF-β1 stimulation (Prevot et al., 2000). Subsequent double-labeling experiments showed that astrocytes in the preoptic area expressed TGF-β1 mRNA and that GnRH perikarya were often found in close association with TGF-β1 mRNA-expressing cells (Bouret et al., 2004). Incubation of preoptic explants with TGF-β1 caused a significant, dose-dependent decrease in GnRH mRNA expression in individual neurons. This effect was inhibited by addition of the soluble form of TGF-β-RII to the incubation medium (Bouret et al., 2004). These results support that astrocyte-derived TGF-β1 may directly influence GnRH expression and/or

Intracisternal administration of TGF-β induces an increase in fat oxidation while intracisternal administration of anti-TGF-β antibody partially inhibits an increase in fat oxidation during treadmill running in rats indicating a regulatory role of TGF-β in the brain on fat oxidation during exercise (Fujikawa et al., 2007). Since TGF-β3 increased noradrenaline levels in the paraventricular and ventromedial hypothalamic nuclei and chemical lesion of noradrenaline input to these nuclei completely abolished the regulatory effect of TGF-β on fat oxidation it has been suggested that TGF-β in the brain enhances fat oxidation via noradrenergic neurons in the paraventricular and ventromedial hypothalamic nuclei (Fujikawa et al., 2007). An inhibitor of FA oxidation could induce an activation of TGF-β in the CSF suggesting that shortage of energy derived from fatty acids leads to the

TGF-β1 and 3 also co-localize with arginine vasopressin in magnocellular neurons of the supraoptic and paraventricular nuclei of the hypothalamus suggesting that TGF-β secreted by the neurohypophysis might regulate the proliferation and secretion of certain anterior pituitary cells (Fevre-Montange et al., 2004). So far, the regulatory role of TGF-β on hormone secretion, gene transcription, and cellular growth of prolactin-producing cells has been shown. TGF-β inhibited the transcriptional activity of the estrogen receptor although estrogens had no effect on TGF-β-specific Smad protein transcriptional activity (Giacomini et al., 2009). A diurnal pattern of expression of TGF-β as well as SMAD3 was found in the suprachiasmatic and paraventricular nuclei of young animals, a rhythm that was not onserved in older mice suggesting a diurnal and age-dependent function of the TGF-β system in these nuclei (Beynon et al., 2009). These expressional data indicate numerous yet unexplored functions of TGF-βs in the hypothalamus. Therefore, promising future

investigations on the role of TGF-βs in endocrine regulations are eagerly awaited.

In addition to their role in normal functioning of the nervous system, TGF-βs have also been implicated in different mechanisms under pathophysiological conditions. TGF-βs have been shown to be injury-related proteins. They were suggested to be neuroprotective following ischemia as well as in various neurological diseases. Furthermore, TGF-βs were implicated in glial scar formation. Finally, their role in brain tumor formation will be

**5. Pathophysiological functions of TGF-βs in the nervous system** 

secretion in vivo by acting on the perikarya of GnRH neurons.

activation of TGF-β (Fujikawa et al., 2011).

discussed.

A number of studies have documented that TGF-β1 levels are enhanced in the brain following cerebral ischemia (Dhandapani & Brann, 2003). The expressions of other isoforms are also enhanced albeit with a different pattern around a focal lesion. Following middle cerebral artery occlusion, TGF-β1 expression was elevated in the penumbra around the lesion site while TGF-β2 and 3 expressions showed increases in particular cortical layers throughout the ipsilateral cerebral cortex (Vincze et al., 2010).

TGF-β1 administered into the brain reduced the infarct size dose dependently in experimental models of ischemia including permanent occlusion of the left middle cerebral artery by microbipolar electrocoagulation in mice (Prehn et al., 1993), autologous clot embolus injection into the right internal carotid artery in rabbit (Gross et al., 1993), and transient middle cerebral artery occlusion in rat (Zhu et al., 2002). In turn, antagonizing the endogenous action of TGF-β1 with a soluble TGF-β type II receptor resulted in a dramatic increase in infarct area (Ruocco et al., 1999). An intracortical injection of the soluble antagonsi in rats subjected to a 30-minute reversible cerebral focal ischemia aggravated the volume of infarction (Ruocco et al., 1999). These results suggest that, in response to an ischemic insult, brain tissue responds by the synthesis of TGF-β1, which is involved in the limitation of the extent of the injury.

Despite the accumulating knowledge on the signal transduction of TGF-βs, the signaling pathway mediating its protective effect is not fully understood. Bad is a proapoptotic member of the Bcl-2 family and is inactivated on phosphorylation via mitogen-activated protein kinase (MAPK). A gradual activation of extracellular signal-regulated kinase 1/2 and MAPK-activated protein kinase-1 and a concomitant increase in Bad phosphorylation in mouse brains after adenovirus-mediated TGF-β1 transduction under nonischemic and ischemic conditions induced by transient middle cerebral artery occlusion (Zhu et al., 2002). Consistent with these effects, the ischemia-induced increases in Bad protein level and caspase-3 activation were suppressed in TGF-β1-transduced brain. Consequently, DNA fragmentation, ischemic lesions, and neurological deficiency were significantly reduced. Furthermore, inhibitors of the MAPK signal transduction pathway abolished the neuroprotective activity of TGF-β1 in staurosporine-induced apoptosis, indicating that activation of MAPK is necessary for the antiapoptotic effect of TGF-β1 (Zhu et al., 2002). These data suggest that TGF-β1 regulates the expression and ratio of apoptotic (Bad) and antiapoptotic proteins, creating an environment favorable for cell survival of death-inducing insults (Dhandapani & Brann, 2003).

#### **5.2 TGF-β in astrogliosis**

The physiological role of astrogliosis remains controversial with respect to the beneficial or detrimental influence of reactive astrocytes on CNS recovery. On the one hand, the very dense network of processes built up in the scar by reactive astrocytes suggests that the scar tissue may fulfill important functions as a barrier isolating and protecting the intact tissue from the lesions, from which toxic molecules could be released. On the other hand, molecules expressed in lesion scars on the astroglial cell surface or secreted molecules render the reactive astrocyte a less favorable substrate, which could be inhibitory to neuritic outgrowth. Nevertheless, scar formation in the nervous system begins within hours after

Transforming Growth Factor Beta in the Central Nervous System 139

Paradoxically, many brain tumors escape from normal TGF-β inhibitory control. High-grade human gliomas secrete TGF-β and can activate latent TGF-β (Sasaki et al., 2001). Yet, they are resistant to its growth inhibitory effects. In fact, they develop mechanisms that change the anti-proliferative influence of TGF-β into oncogenic cues. Thus, TGF-β is involved in tumor progression (Aigner & Bogdahn, 2008). The dominant hypothesis of TGF-β's pathogenetic association with malignant transformation has been predicated upon acquisition of resistance to its growth inhibitory effects. However, the lack of obvious correlation with TGF-β receptor expression between hyperdiploid glioblastoma multiforme and TGF-β-inhibited glioblastoma cultures suggests the existence of intrinsically opposed regulatory mechanisms influenced by TGF-β (Jennings & Pietenpol, 1998). The mechanism of conversion might be explained either by the loss of a putative tumor suppressor gene, which mediates TGF-β's inhibition of growth or by enhancement of an active oncogenic pathway among hyperdiploid glioblastoma multiforme. Experimental evidence supports the involvement of several factors including inactivating mutation/loss of the TbetaR type II, alterations in post-receptor signal transmission or the cyclin/cyclin dependent kinase system. The expression of the Smad2 and Smad3 proteins is lowered in many glioma cell lines. The phosphorylation and nuclear translocation of Smad2 and Smad3 are also impaired (Zhang et al., 2006). The loss of p15(INK4B) may also explain the selective loss of growth inhibition by TGF-β in gliomas to form a more aggressive tumor phenotype (Rich et al., 1999). TGF-β also induced expression of Sox2, a stemness gene, and this induction was mediated by Sox4, a direct TGF-β target gene. Inhibitors of TGF-β signaling drastically deprived tumorigenicity of glioblastoma cells identifying the relevance of the TGF-β-Sox4- Sox2 pathway, too (Ikushima et al., 2009). Among TGF-β isoforms, TGF-β2 has been identified as the most important factor in the progression of malignant gliomas. TGF-β2, originally described as "glioblastoma-derived T-cell suppressor factor", was particularly associated with the immuno-suppressed status of patients with glioblastoma. Furthermore, elevated TGF-β2 levels in tumors and in the plasma of patients have been associated with

advanced disease stage and poor prognosis (Hau et al., 2011).

High-grade gliomas are the most common primary tumors in the central nervous system (CNS) in adults. Despite efforts to improve treatment by combination therapies (neurosurgery, radio- and chemotherapy), high-grade glioma patients still have a grim prognosis, indicating an urgent need for new therapeutic approaches. Since TGF-β is intimately involved in the regulation of several processes characteristic of human malignant glioma including excessive proliferation, infiltrative growth, angiogenesis and suppression of anti-tumor immune surveillance, TGF-β promises to become a novel target for the experimental therapy of human malignant glioma (Platten et al., 2001). Several in vitro paradigms and rodent glioma models have been used to demonstrate that the antagonism of TGF-β holds promise for the treatment of glioblastoma, employing antisense strategies, inhibition of pro-TGF-β processing, scavenging TGF-β by decorin, or blocking TGF-β activity by specific TGF-β receptor I kinase antagonists (Naumann et al., 2008; Wick et al., 2006). Among these possibilites, the antisense oligonucleotide trabedersen (AP 12009) that specifically blocks TGF-β2 mRNA has the highest potential at present to treat gliobastomas (Hau et al., 2011). In three phase I/II studies and a randomized, active-controlled dosefinding phase IIb study, trabedersen treatment of high-grade glioma patients with recurrent or refractory tumor disease led to long-lasting tumor responses and so far promising

traumatic injury and is characterized primarily by reactive astrocytes depositing proteoglycans that inhibit regeneration. A fundamental question in CNS repair has been the identity of the initial molecular mediator that triggers glial scar formation. Recent evidence suggests that one of the gliosis signaling molecules is TGF-β. TGF-β up-regulates the biosynthesis of keratin sulphate and chondroitin sulphate (Yin et al., 2009). Local injection of TGF-β antagonists into cerebral wounds reduces glial scarring (Lagord et al., 2002). Inhibition of the TGF-β receptor pathway abolished the fibrinogen-induced effects on glial scar formation in vivo and in vitro (Schachtrup et al., 2011) pointing to TGF-β as a molecular link between vascular permeability and scar formation. Furthermore, TGF-β expression increases immediately after injury e.g. in the injured segment in an animal model using an impactor (Wang et al., 2009). There are, however, differences in the expression pattern of individual TGF-β isoforms. Levels of TGF-β1 mRNA were most elevated over the acute inflammatory phase after transection of the dorsal funiculi in the spinal cord, while TGF-β2 mRNA levels were raised locally about the wound, particularly in astrocytes and neovascular endothelial cells, over the subacute period of scarring. TGF-β protein production also increased after injury. Both TGF-β1 and TGF-β2 were found in hematogenous inflammatory cells, while TGF-β1 was also neuron-associated, and high levels of TGF-β2 were localized to multiple cell types in the wound, including reactive astrocytes, during the period of glial/collagen scar formation (Lagord et al., 2002). More recently, TGF-β levels were also reported in the human spinal cord after traumatic injury. Sections from human spinal cords from 4 control patients and from 14 patients who died at different time points after traumatic spinal cord injury were investigated immunohistochemically. In control cases, TGF-β1 was confined to occasional blood vessels, intravascular monocytes and some motoneurons, whereas TGF-β2 was only found in intravascular monocytes (Buss et al., 2008). After traumatic spinal cord injury, TGF-β1 immunoreactivity was dramatically upregulated by 2 days after injury and was detected within neurons, astrocytes and invading macrophages. The staining was most intense over the first weeks after injury but gradually declined by 1 year. TGF-β2 immunoreactivity was first detected 24 days after injury. It was located in macrophages and astrocytes and remained elevated for up to 1 year. In white matter tracts undergoing Wallerian degeneration, there was no induction of either isoform (Buss et al., 2008). The conclusion from these studies is that TGF-β1 modulates the acute inflammatory and neural responses and formation of the glial scar, while the later induction of TGF-β2 may indicate a role in the maintenance of the scar.

#### **5.3 The role of TGF-βs in brain tumor formation**

TGF-β reveals anti-proliferative control on most cell types including an inhibitory effect on the proliferation of normal astrocytes. Thus, upon TGF-β treatment, primary rat astrocytes show a significant decrease in DNA synthesis upon thymidine incorporation with a cell cycle arrest in the G(1) phase and the expression of the cyclin-dependent kinase inhibitor (CdkI) p15(INK4B) is up-regulated (Rich et al., 1999). The SMAD signal transduction pathway is likely to be involved as Smad3 null mouse astrocytes show a loss of both TGF-βmediated inhibition of growth (Rich et al., 1999). Analysis of Smad3 null mouse astrocytes showed a significant loss of both TGF-β-mediated growth inhibition and p15(INK4B) induction compared with wild-type mouse astrocytes.

traumatic injury and is characterized primarily by reactive astrocytes depositing proteoglycans that inhibit regeneration. A fundamental question in CNS repair has been the identity of the initial molecular mediator that triggers glial scar formation. Recent evidence suggests that one of the gliosis signaling molecules is TGF-β. TGF-β up-regulates the biosynthesis of keratin sulphate and chondroitin sulphate (Yin et al., 2009). Local injection of TGF-β antagonists into cerebral wounds reduces glial scarring (Lagord et al., 2002). Inhibition of the TGF-β receptor pathway abolished the fibrinogen-induced effects on glial scar formation in vivo and in vitro (Schachtrup et al., 2011) pointing to TGF-β as a molecular link between vascular permeability and scar formation. Furthermore, TGF-β expression increases immediately after injury e.g. in the injured segment in an animal model using an impactor (Wang et al., 2009). There are, however, differences in the expression pattern of individual TGF-β isoforms. Levels of TGF-β1 mRNA were most elevated over the acute inflammatory phase after transection of the dorsal funiculi in the spinal cord, while TGF-β2 mRNA levels were raised locally about the wound, particularly in astrocytes and neovascular endothelial cells, over the subacute period of scarring. TGF-β protein production also increased after injury. Both TGF-β1 and TGF-β2 were found in hematogenous inflammatory cells, while TGF-β1 was also neuron-associated, and high levels of TGF-β2 were localized to multiple cell types in the wound, including reactive astrocytes, during the period of glial/collagen scar formation (Lagord et al., 2002). More recently, TGF-β levels were also reported in the human spinal cord after traumatic injury. Sections from human spinal cords from 4 control patients and from 14 patients who died at different time points after traumatic spinal cord injury were investigated immunohistochemically. In control cases, TGF-β1 was confined to occasional blood vessels, intravascular monocytes and some motoneurons, whereas TGF-β2 was only found in intravascular monocytes (Buss et al., 2008). After traumatic spinal cord injury, TGF-β1 immunoreactivity was dramatically upregulated by 2 days after injury and was detected within neurons, astrocytes and invading macrophages. The staining was most intense over the first weeks after injury but gradually declined by 1 year. TGF-β2 immunoreactivity was first detected 24 days after injury. It was located in macrophages and astrocytes and remained elevated for up to 1 year. In white matter tracts undergoing Wallerian degeneration, there was no induction of either isoform (Buss et al., 2008). The conclusion from these studies is that TGF-β1 modulates the acute inflammatory and neural responses and formation of the glial scar, while the later induction of TGF-β2 may indicate a role in the

TGF-β reveals anti-proliferative control on most cell types including an inhibitory effect on the proliferation of normal astrocytes. Thus, upon TGF-β treatment, primary rat astrocytes show a significant decrease in DNA synthesis upon thymidine incorporation with a cell cycle arrest in the G(1) phase and the expression of the cyclin-dependent kinase inhibitor (CdkI) p15(INK4B) is up-regulated (Rich et al., 1999). The SMAD signal transduction pathway is likely to be involved as Smad3 null mouse astrocytes show a loss of both TGF-βmediated inhibition of growth (Rich et al., 1999). Analysis of Smad3 null mouse astrocytes showed a significant loss of both TGF-β-mediated growth inhibition and p15(INK4B)

maintenance of the scar.

**5.3 The role of TGF-βs in brain tumor formation** 

induction compared with wild-type mouse astrocytes.

Paradoxically, many brain tumors escape from normal TGF-β inhibitory control. High-grade human gliomas secrete TGF-β and can activate latent TGF-β (Sasaki et al., 2001). Yet, they are resistant to its growth inhibitory effects. In fact, they develop mechanisms that change the anti-proliferative influence of TGF-β into oncogenic cues. Thus, TGF-β is involved in tumor progression (Aigner & Bogdahn, 2008). The dominant hypothesis of TGF-β's pathogenetic association with malignant transformation has been predicated upon acquisition of resistance to its growth inhibitory effects. However, the lack of obvious correlation with TGF-β receptor expression between hyperdiploid glioblastoma multiforme and TGF-β-inhibited glioblastoma cultures suggests the existence of intrinsically opposed regulatory mechanisms influenced by TGF-β (Jennings & Pietenpol, 1998). The mechanism of conversion might be explained either by the loss of a putative tumor suppressor gene, which mediates TGF-β's inhibition of growth or by enhancement of an active oncogenic pathway among hyperdiploid glioblastoma multiforme. Experimental evidence supports the involvement of several factors including inactivating mutation/loss of the TbetaR type II, alterations in post-receptor signal transmission or the cyclin/cyclin dependent kinase system. The expression of the Smad2 and Smad3 proteins is lowered in many glioma cell lines. The phosphorylation and nuclear translocation of Smad2 and Smad3 are also impaired (Zhang et al., 2006). The loss of p15(INK4B) may also explain the selective loss of growth inhibition by TGF-β in gliomas to form a more aggressive tumor phenotype (Rich et al., 1999). TGF-β also induced expression of Sox2, a stemness gene, and this induction was mediated by Sox4, a direct TGF-β target gene. Inhibitors of TGF-β signaling drastically deprived tumorigenicity of glioblastoma cells identifying the relevance of the TGF-β-Sox4- Sox2 pathway, too (Ikushima et al., 2009). Among TGF-β isoforms, TGF-β2 has been identified as the most important factor in the progression of malignant gliomas. TGF-β2, originally described as "glioblastoma-derived T-cell suppressor factor", was particularly associated with the immuno-suppressed status of patients with glioblastoma. Furthermore, elevated TGF-β2 levels in tumors and in the plasma of patients have been associated with advanced disease stage and poor prognosis (Hau et al., 2011).

High-grade gliomas are the most common primary tumors in the central nervous system (CNS) in adults. Despite efforts to improve treatment by combination therapies (neurosurgery, radio- and chemotherapy), high-grade glioma patients still have a grim prognosis, indicating an urgent need for new therapeutic approaches. Since TGF-β is intimately involved in the regulation of several processes characteristic of human malignant glioma including excessive proliferation, infiltrative growth, angiogenesis and suppression of anti-tumor immune surveillance, TGF-β promises to become a novel target for the experimental therapy of human malignant glioma (Platten et al., 2001). Several in vitro paradigms and rodent glioma models have been used to demonstrate that the antagonism of TGF-β holds promise for the treatment of glioblastoma, employing antisense strategies, inhibition of pro-TGF-β processing, scavenging TGF-β by decorin, or blocking TGF-β activity by specific TGF-β receptor I kinase antagonists (Naumann et al., 2008; Wick et al., 2006). Among these possibilites, the antisense oligonucleotide trabedersen (AP 12009) that specifically blocks TGF-β2 mRNA has the highest potential at present to treat gliobastomas (Hau et al., 2011). In three phase I/II studies and a randomized, active-controlled dosefinding phase IIb study, trabedersen treatment of high-grade glioma patients with recurrent or refractory tumor disease led to long-lasting tumor responses and so far promising

Transforming Growth Factor Beta in the Central Nervous System 141

full-length huntingtin with an expanded glutamine repeat (Battaglia et al., 2011). These data suggest that serum TGF-β1 levels are potential biomarkers of HD development during the asymptomatic phase of the disease, and raise the possibility that strategies aimed at rescuing

Additional neurodegenerative diseases were also associated with an alteration of the TGFβs. Using immunohistochemistry, the expression of TGF-β2 appeared in neurofibrillary tangle bearing neurons and tangle-bearing glial cells in progressive supranuclear palsy and in neurons with age-related neurofibrillary tangle formation (Lippa et al., 1995). Widespread staining of reactive astrocytes for TGF-β2 was observed in all degenerative diseases. TGF-β1 and -3 staining was not selectively altered in these diseases (Lippa et al., 1995). These data suggest that the induction of TGF-β2 may be an intrinsic part of the processes that underlie neurofibrillary tangle formation and reactive gliosis in a variety of neurodegenerative

Activation of the TGF-β pathway was identified as the underlying mechanism behind the epileptogenic effect of albumin following the compromise of the blood brain barrier (Cacheaux et al., 2009). TGF-β1 resulted in epileptiform activity similar to that after exposure to albumin. Coimmunoprecipitation revealed binding of albumin to TGF-β receptor II, and Smad2 phosphorylation confirmed downstream activation of this pathway. Transcriptome profiling demonstrated similar expression patterns after blood brain barrier breakdown, albumin, and TGF-β1 exposure, including modulation of genes associated with the TGF-β pathway, early astrocytic activation, inflammation, and reduced inhibitory transmission. Importantly, TGF-β pathway blockers suppressed most albumin-induced transcriptional changes and prevented the generation of epileptiform activity. Based on these data, the TGF-β pathway was suggested to be a novel putative epileptogenic signaling cascade and therapeutic target for the prevention of injury-induced epilepsy (Cacheaux et

TGF-βs are a class of growth factors and cytokines with a special biochemistry and range of actions thoughout the organs of the body. Our knowledge on their roles in the central nervous system is accumulating fast in the last years. Neverthless, their neurochemistry is not well described, and often we can only anticipate that their synthesis, activation, and signal transduction pathways is similar to that in other tissues. The potential differences are to be determinded in future studies. Our understanding of the physiological functions of endogenous TGF-βs is also limited despite recent significant progress in the field. An increasing body of evidence suggests the otherwise logical assumption that TGF-βs play a role in the development of the nervous tissue. Recent studies revelased that TGF-βs are also involved in the physiological functions of the adult nervous system as well. So far, the best established functions include synaptic transmission and neuronal plasticity. Somewhat surprisingly, however, the direct involvement of TGF-βs in neuroendocrine functions has also been supported. The experiments often did not differentiate between the 3 different isoforms of TGF-β. Therefore, future studies are needed to elaborate their specific functions. Based on differences in the distribution of TGF-β1, TGF-β2, and TGF-β3, they possess

TGF-β1 levels in the brain may influence the progression of HD.

diseases.

al., 2009).

**6. Conclusion** 

separate neural functions.

survival data. On the basis of these data the currently ongoing phase III study SAPHIRRE was initiated (Hau et al., 2011). In addition, TGF-β inhibition may also be used as a supplementary treatment as it can enhance the therapeutic efficacy of glioma-associated antigen vaccines (Ueda et al., 2009).
