Open access peer-reviewed chapter

Characterizing Osteosarcoma Through PTEN and PI3K: What p53 and Rb1 Can’t Tell Us

By Matthew G. Cable and R. Lor Randall

Submitted: April 8th 2016Reviewed: November 25th 2016Published: April 26th 2017

DOI: 10.5772/67071

Downloaded: 558

Abstract

Attention has been given to the fact that overall survival of osteosarcoma has plateaued over the last 30 years despite the addition of chemotherapy regimens. Elucidating the involvement of p53 and Rb1 in osteosarcoma has not yielded many novel treatments, but recent studies have started to characterize how the PTEN and the PI3K pathway can contribute to osteosarcoma. PTEN is a tumor suppressor that regulates a variety of signal transduction pathways and cellular processes, mainly by antagonizing PI3K activity and shutting down the PI3K/Akt pathway. Loss of PTEN function with concurrent PI3K activation has been detected frequently in a multitude of cancers, including osteosarcoma. This chapter aims to characterize PTEN and the PI3K/Akt pathway in osteosarcoma, their effects on primary bone tumor behavior, and potential therapeutic targets.

Keywords

  • osteosarcoma
  • phosphatase and tensin homolog
  • phosphatidylinositol 3-kinase
  • therapeutic applications

1. Introduction

Osteosarcoma (OS) is a malicious cancer that affects predominantly children and adolescents, and is the most common primary sarcoma of bone. After the advents of adjuvant multi-agent chemotherapy, in combination with surgery, the 5-year survival rate for OS has increased from 40 to 76% in children under 15 and from 56 to 66% in adolescents 15–19 years old [1]. Despite these advances, the prognosis for the 20% of patients that present with stage IV disease remains poor, and survival rates have plateaued [2]. In addition, with roughly 60% of cases occurring in just the second decade of life, the societal cost of OS exceeds that of many other cancers.

OS also represents a unique entity among pediatric cancers, with cases arising de novo and already exhibiting high-grade pathology, heterogeneous karyotypes, and frequent genomic mutations. This genomic instability is further characterized by unusually high numbers of chromosomal structural variants and not single nucleotide mutations [3]. An OS genome can often contain over 200 of these structural variants, making it the most disordered among childhood cancers [4] (Figure 1).

Figure 1.

CIRCOS plots of osteosarcoma tumors. Representative CIRCOS plots of validated mutations and chromosomal lesions in diagnostic and metastatic osteosarcoma tumors from different patients. Loss of heterozygosity (orange), gain (red), and loss (blue) is shown. Intra-chromosomal (green lines) and inter-chromosomal (purple lines) translocations are indicated. Sequence mutations in RefSeq genes included silent single nucleotide variations (green), nonsense and missense single nucleotide variations (brown), splice-site mutations (dark blue), and insertion/deletion mutations (red). An additional track was added to the innermost ring of the plot showing the density of single nucleotide variations to highlight regions adjacent to structural variations characteristic of kataegis. Adapted from Figure 1A, Chen et al. [3].

To overcome the stagnation in survival rates, the cellular etiology and biology of OS need to be more completely understood. Toward this end, molecular targets that actively modulate essential cell processes such as cell cycle regulation, migration, mitosis, metabolism, and apoptosis have been studied to develop potential therapies. The most well-known and frustrating examples are the frequent inactivation mutations of cell cycle regulator gene tumor protein 53 (TP53) or tumor suppressor gene retinoblastoma protein 1 (Rb1), and attempts in translating these targets into applicable therapies have been met with much difficulty. Recent advances in cell signaling have broadly identified tyrosine kinase receptors (TKRs) as prominent targets for cancer therapies, with many receptors confluencing on the second messenger phosphatidylinositol (3,4,5)-triphosphate (PIP3), and it is activation by phosphatidylinositol-4,5-bisphosphonate 3-kinase (PI3K). More importantly, PI3K and its inhibitor, phosphatase and tensin homolog (PTEN), may play significant roles in OS and represent therapeutic targets [4]. Investigations have also been centered on the effects of PTEN on osteoclastogenesis, and how the bone microenvironment may facilitate tumor expansion with PTEN loss. Therapeutic targets are expanding as strategies focus on restoring normal PTEN function and inhibiting PI3K pathway activation.

2. The PI3K pathway and PTEN

Class I PI3K is a family of heterodimeric signal transduction enzymes that phosphorylates the 3′ hydroxyl group of the inositol ring on phosphatidylinositol-4,5-bisphosphate (PIP2) to PIP3. The implications of that biochemical mouthful are that PI3K activates one of the most influential pathways in cell cycle regulation and cell proliferation, the PI3K/Akt pathway (Figure 2).

Figure 2.

PI3K/Akt signaling pathway. Illustration reproduced courtesy of Cell Signaling Technology, Inc. (www.cellsignal.com).

The PI3K/Akt pathway was first identified when attempting to characterize insulin signaling and discovering the tyrosine kinase receptor type-1 insulin-like growth factor receptor (IGF1R) [5]. IGFR1 is one of the many tyrosine kinase receptors (TKRs) in the cell membrane, in addition to G protein-coupled receptors, that can instigate signaling through the canonical PI3K pathway. Although there are three classes of PI3Ks, class I is the most involved in oncogenesis and divided into class IA (PI3K alpha, PI3K beta, and PI3K delta) and class IB (PI3K gamma) [6]. All class I PI3Ks form a heterodimer consisting of a catalytic and regulatory subunit. The catalytic subunits forming class IA PI3Ks are p110 alpha, beta, and delta, and encodedby the genes PIK3CA, PIK3CB, and PIK3CD, respectively. The regulatory subunits for class IA PI3Ks are p85 alpha, p85 beta, and p55 gamma encoded by PIK3R1, PIK3R2, and PIK3R3, respectively [7]. Class IB PI3K is formed exclusively from the catalytic subunit p110 gamma (encoded by PIK3CG) and and the regulatory subunit p101 (encoded by PIK3R5). Class IA PI3Ks can be activated by TKRs including IGFR1, platelet-derived growth factor (PDGF), and epidermal growth factor receptor (EGFR). Class IB PI3Ks are activated only by G protein-coupled receptors.

Once activated, PI3K converts PIP2 to PIP3 [8]. PIP3 acts as a second messenger by recruiting and activating proteins containing a pleckstrin homology domain to the plasma membrane, notably the phosphoinositide-dependent kinase 1 (PDK1) and serine-threonine kinase Akt (also known as protein kinase B) [9]. Akt is activated via two phosphorylation sites, and the first occurs at threonine 308 by none other than PDK1. The second phosphorylation required for Akt activation occurs at serine 473 by a number of kinases including PDK1 or even Akt itself [7]. Once activated, Akt translocates to the cell cytoplasm or nucleus to set in motion a number of downstream effects to inhibit apoptosis and induce protein synthesis.

One of the most important downstream targets of Akt is mammalian target of rapamycin (mTOR), a 289 kDa serine/threonine kinase that drives one of the two complexes in mammalian cells, mTORC1 or mTORC2. mTORC1 has been implicated as a driver of many cancers, and mTORC2 can create a positive feedback loop by phosphorylating and activating Akt at serine 473 [10]. mTOR is activated when Akt phosphorylates and inactivates a regulatory protein in the mTORC1 complex called proline-rich Akt substrate of 40 kDa (PRAS40). Akt also inhibits the tumor suppressor protein tuberous sclerosis protein 2 (TSC2) which can result in activation of mTOR [11]. The activation of mTORC1 promotes protein synthesis and cellular proliferation by phosphorylation of eukaryotic translation initiation factor 4E-binding protein 1 (4EBP1) and ribosomal protein S6 kinase polypeptide 1 (S6K1) which in turn phosphorylates ribosomal protein S6 (RPS6). The majority of mTOR inhibitors (including rapamycin derivatives) function by inhibiting mTORC1.

In addition to mTOR, important downstream targets of the PI3K/Akt pathway include promotion of cyclin-dependent kinase 4 (CDK4) to support cell cycle progression [12] and activation of nuclear factor-kappa B (NF-κB) which initiates transcription of many target genes including those seen in drug-resistant malignancies [13]. Akt also inhibits Bcl-2-associated death promoter (BAD), caspase-9, and forkhead box O3 (FOXO3), while activating cyclic AMP response element-binding protein (CREB), all of which serve to prevent apoptosis [1416].

The grandiosity and arborous nature of the of the PI3K/Akt pathway emphasize the current attention being given to phosphatase and tensin homolog (PTEN) (Figure 3), a lipid phosphatase that directly antagonizes PI3K during the initiation steps of the PI3K/Akt pathway [17]. PTEN is a 200 kb gene located on chromosome 10q23.3 [18], which encodes a 60-kDa dual-specificity phosphatase that cleaves phosphate groups from phospholipids (including phosphatidylinositols) as well as proteins (serine, threonine, and tyrosine residues including those on TKRs) [19]. PTEN functions as a tumor suppressor by dephosphorylating and inactivating the second messenger PIP3 to PIP2, cogently shutting down the PI3K/Akt pathway and promoting cell cycle regulation and apoptosis [20]. The function of PTEN is further supported by phosphorylation at tyrosine 336 by RAK, which prevents ubiquitin-mediated proteosomal degradation [21]. PTEN also has tumor suppressor functions outside of the canonical PI3K pathway and regulates a variety of cellular processes and signal transduction pathways, such as controlling cell proliferation through cyclin D1 levels [22, 23]. A 576 amino acid translational variant of PTEN has been discovered, termed PTEN long, that can be secreted and enter other cells, enabling tumor suppressor effects in a paracrine-like manner [24]. The C-terminus of PTEN is thought to be essential in maintaining heterochromatin structure and genomic stability [25].

Figure 3.

A ribbon diagram of the PTEN tumor suppressor. The phosphatase domain and the C2 domains are shown in yellow and green ribbons, respectively. Reproduced from Das et al. [98].

3. Relevance of PTEN and PI3K to cancer processes

OS is already prone for structural genomic variations and chromothripsis, and whether this is due to selective loss of the TP53 gene or an intrinsic feature of OS itself is beyond the scope of this paper. What this intrinsic genomic instability does allow for is the selection of cancer cells that consistently develop predictable mutations that are beneficial for tumor survival [3]. This phenomenon of localized hypermutation in a cancer genome is called kataegis. Two of these predictable mutations in OS are the frequent loss of PTEN and activation of the PI3K/Akt pathway [26, 27]. The PI3K/Akt pathway is activated in a multitude of cancers, as its overall effects are to promote cellular proliferation and survival while reducing apoptosis. Akt upregulates expression of murine double minute 2 (MDM2) which further inhibits release of tumor protein 53 (p53) [28] and can result in very aggressive phenotypes since p53 is already stunted in many OS cancers. p53 itself has been shown to be a potent dual inhibitor of mTORC1 and mTORC2 in OS cells, further supporting a highly active PI3K/Akt pathway in OS tumors [29].

Loss of PTEN function has been detected frequently in many different forms of cancer including breast, prostate, lung, gastric, colon, and skin cancer, as well as endometrial carcinoma. The frequency of mono-allelic mutations of the PTEN locus is estimated to be 50–80% in sporadic tumors, with complete loss generally associated with advanced malignancies and metastases [30]. In fact, the PTEN protein was initially referred to as MMAC1 for “mutated in multiple advanced cancers” [18]. PTEN deletion mutations were first identified in canine OS cell lines [31], and retrospectively, it is interesting to note that chromosomal loss of 10q has been a frequent occurrence in over 50% of human OS tumor samples analyzed in some studies [32]. Specific mutations in PTEN have been identified in 44% of pediatric OS [3].

The genetic disruption of PTEN often leads to unchecked PI3K/Akt signaling [33]. In breast cancer cell lines, a functional PTEN can directly inhibit cell growth [21]. More recently, the PI3K/Akt pathway and PTEN have risen to the forefront of OS research since the surprise finding by Perry et al. that identified biologically and clinically relevant alterations in the PI3K/Akt pathway in 24% of OS samples [4]. Furthermore, when comparing the same OS tumors to a murine model of OS (with conditional deletions of TP53 and Rb1 in the preosteoblast), both contained somatic mutations in PTEN and PIK3R1.

4. How is PTEN lost in cancer?

There are numerous methods by which the PI3K/Akt pathway is activated in malignancies, often the consequence of upregulation via a multitude of TKRs and G-protein-coupled receptors. With PTEN having a pronounced effect on the PI3K/Akt pathway, PTEN function is accordingly inhibited in many ways. Recurrent PTEN germline mutations often involve exon 5, which codes for its phosphatase domain, and missense and nonsense somatic mutations also can occur in exons 5–8 [34]. Cowden’s disease (also known as multiple hamartoma syndrome) is an autosomal dominant condition with increased risk of thyroid, breast, uterine, and kidney cancers that results from similar somatic inactivating mutations of PTEN [35]. The end result is a functional loss of heterozygosity, which is sufficient to produce oncogenesis, as PTEN is essential for embryonic development, and homozygous loss results in embryonic lethal phenotype in mice [36].

At the epigenetic level, PTEN is inhibited by aberrant promoter hypermethylation of CpG islands, which is a poor prognostic indicator in numerous cancer types including breast, colorectal, uterine, and malignant melanoma [3740]. While hypermethylation has been identified in several soft tissue sarcomas and in murine models of OS, it has only been studied in human OS tumors in vitro [41]. PTEN protein translation can be negatively regulated by many microRNAs including miR-92a, miR-17, miR-128, and miR-130/131 [4244]. Theoretically, structural features of PTEN itself could prove to be potential sites for compromising protein stability. Post-translational modification such as phosphorylation at tyrosine 336 could be abrogated, thus promoting protein degradation. Protein localization to the cell membrane or nucleus could be affected if the C2 or PDZ-binding domains of PTEN were impaired. Although the majority of inactivating mutations affect the phosphatase domain of PTEN, preventing it from cleaving PIP3 to PIP2, there are several ways that PTEN expression and dysregulation can be imparted in malignancies.

5. PTEN in bone and osteosarcoma

PTEN is frequently deactivated through deletions in human OS tumor samples [26]. A sleeping beauty forward genetic screen by Moriarity et al. supports the role of PTEN loss as a key driver in osteosarcomagenesis in mice, with a concordant enrichment of genes involved in the PI3K/Akt pathway [45]. The PI3K/Akt pathway is already recognized as a common effector for RTK-activating mutations in cancer [46]. The loss of PTEN only further promotes this, but what factors in the bone microenvironment that facilitate tumor expansion in the absence of PTEN is unknown. It would be very un-Darwinian for OS, multiple myeloma, and bone metastases to all have PTEN derangements for unrelated reasons. The effects of PTEN on osteoclastogenesis and receptor activator of nuclear factor-kappa-B ligand (RANKL) may be the commonality among these cancers.

There is a vicious cycle that begins when a malignancy metastasizes or originates in bone that allows it to propagate and induce osteolysis. Metastatic tumor cells can secrete interleukin-1 (IL-1) and parathyroid hormone-related protein (PTHrP), which stimulate RANKL production and secretion from osteoblasts. IL-1 also stimulates osteoblasts to secrete IL-6 via the PI3K/Akt pathway, and IL-6 is a potent stimulator of osteoclastogenesis [47]. OS cells also directly produce RANKL. RANKL binding to the RANK receptor on osteoclast precursors activates osteoclast differentiation and osteolysis. The increased bone resorption results in the release of bone morphogenetic proteins (BMPs) and insulin-like growth factor 1 (IGF-1) that further attract and stimulate the growth of cancer cells in bone, thus initiating a vicious cycle of tumor expansion and osteolysis [48, 49].

PI3K/Akt acts as a cog in the wheel of this vicious cycle, being activated by RANKL and resulting in downstream expression of NFATc1, a key transcription factor of osteoclastogenesis [50]. This gives PTEN the opportunity to prevent bone resorption by inhibiting the PI3K/Akt pathway, thus suppressing RANKL-induced osteoclast differentiation and stopping the vicious cycle [51]. This would support the loss of PTEN in aggressive bony malignancies, preventing stimulation of osteoblasts by IL-1 and osteoclasts by RANKL. Murine models with PTEN deletions have shown increased osteoblast proliferation and bone mass [52], in addition to increased osteoclast differentiation [53]. Fibroblast growth factor 18 (FGF18) is another stimulator of bone growth, and with all FGF receptors being RTK’s, its effects are mediated via the PI3K/Akt pathway and can be inhibited by PTEN [54]. Vascular endothelial growth factor (VEGF) and EGFR are both upregulated in OS and are key activators of the PI3K/Akt pathway [26, 55].

The finding that WNT5A may phosphorylate Akt is further support for the bone microenvironment being conducive to tumor growth [56]. WNT5A is expressed on osteoblast-precursor cells and activates RTK-like orphan receptors (Ror) on osteoclast precursors, culminating in increased expression of RANK on osteoclasts and enhancing RANKL osteoclastogenesis [57]. The result of PTEN loss in OS would be twofold in promoting the effects of WNT5A: (1) further increasing RANKL production by osteoblasts and (2) enabling WNT5A-related Akt activation, both increasing osteoclastogenesis and osteolysis.

6. PTEN in osteosarcoma and metastasis

WNT5A also provides segue into the realm of OS metastases. Metastatic melanoma cell motility and invasiveness are increased through WNT5A and Akt. Akt is increased in metastatic OS specimens, and inhibiting Akt in mice decreases pulmonary metastases [7]. WNT5A is even implicated in helping to initiate the epithelial to mesenchymal transition (EMT), a key event in malignant cancers developing metastatic potential [58, 59]. The effect of PTEN loss in promoting EMT has been shown in prostate, colorectal, and OS cancer cells [6062]; however, the role of EMT in mesenchymal tumors including OS is still a topic of debate and a focus of current research.

IL-6 is also increased in OS tissues and can promote ICAM-1 expression and cell motility in OS in vitro, possibly correlating to metastatic potential. These effects of IL-6 in OS can be negated by Akt inhibition [63], suggesting that PTEN could also negate this effect in bone by preventing IL-1 from stimulating IL-6 through PI3K/Akt. Another promoter of metastasis is the chemokine CXCL12 and its receptor CXCR4, both induced by IL-1 and strongly linked to bone metastasis [64]. PTEN is involved in the negative regulation of CXCR4, and in prostate cancer, PTEN loss induces CXCL12/CXCR4 expression [65].

Focal adhesion kinase (FAK) deserves special recognition as a direct target of PTEN outside of the canonical PI3K pathway, as it is a target of the protein phosphatase (not lipophosphatase) domain of PTEN [66]. Activation of FAK by phosphorylation (pFAK) promotes proliferation and invasion of tumor cells and increases matrix metalloproteinases that can degrade the extracellular matrix. FAK and pFAK are overexpressed in human OS samples and independently predict overall and metastasis-free survival [67]. This effect may be reversed in OS cells with PTEN transfections, which exhibit decreased migration and adhesion capabilities and concomitant downregulation of pFAK and MMP-9, further supporting the loss of PTEN in OS [68]. Just as FAK can be prognostic in human OS, the presence of PTEN in tumor resections is significantly associated with improved survival prognosis [69]. Unfortunately neither can be correlated with response to current chemotherapy regimens.

7. Therapeutic applications targeting PI3K/Akt pathway and PTEN in osteosarcoma

There is an abundance of convincing evidence supporting a key role for PTEN in OS, but from a therapeutic standpoint, this assumes that there is causality and not just correlation between PTEN loss and poor patient prognosis. Two major therapeutic strategies are restoring normal PTEN function and inhibiting the PI3K pathway. Comparing these two strategies could also aid in distinguishing how PTEN serves as a tumor suppressor beyond the canonical PI3K pathway.

Many chemotherapy agents already target various levels of the PI3K/Akt pathway, and several are in various phases of clinical trials involving OS. These include small molecule inhibitors of PI3K, Akt, and numerous mTOR inhibitors (Table 1). Countless PI3K small molecule inhibitors have been developed, but several have specifically been effective in OS including GSK458, LY294002, BYL719, and BKM120 (Buparlisib) [4, 7072]. Pictilisib (GDC-0941) is a pan PI3K inhibitor that has entered phase I clinical trials for advanced solid tumors [73]. Aminopeptidase N (also known as CD13) is a surface receptor activated by IL-6 that can stimulate PI3K and is involved in tumor invasion. An aminopeptidase N inhibitor, ubenimex, is currently used for acute myeloid leukemia treatments and may help prevent OS metastases [74]. The Akt inhibitors perifosine and MK-2206 both exert anti-OS activity in vitro [75, 76]. Inhibitors of the mTOR complex are the most numerous of this pathway, being extensively studied and developed since the discovery of the mTORC1 inhibitor rapamycin in 1975. Many new inhibitors of the mTORC1/2 complex are currently being developed, but several in particular have been tested on OS either in vivo or are in various phases of clinical trials: temsirolimus [77], ridaforolimus [78], everolimus [79], XL388 [80], and NVP-BEZ235 [81]. Apitolisib (GDC-0980) has entered phase I clinical trials for patients with advanced solid tumors and could precede future studies involving OS patients [82].

AgentTargetTherapies
GSK458PI3KPhase I trial for advanced solid tumors
LY294002PI3KEfficacy against OS cells in vitro
BYL719 (Alpelisib)PI3K alphaPhase I for advanced solid tumors
BKM120 (Buparlisib)PI3KPhase Ib for advanced solid tumors
GDC0941 (Pictilisib)PI3KPhase I for advanced solid tumors
Bestatin (Ubenimex)Aminopeptidase NEfficacy against OS cells in vitro
KRX0401 (Perifosine)PI3K, AktEfficacy against OS cells in vitro
MK-2206AktPhase I for advanced solid tumors and metastatic breast cancer
Rapamycin (Sirolimus)mTORC1Phase II for soft tissue sarcoma and osteosarcoma
CCI779 (Temsirolimus)mTORC1Phase II for soft tissue sarcoma and recurrent/refractory sarcoma
MK8669 (Ridaforolimus)mTORC1Phase II and III for advanced soft tissue sarcoma and osteosarcoma
RAD001 (Everolimus)mTORC1Phase II for advanced osteosarcoma
XL388mTORC1/2Efficacy against OS cells in vitro
NVP-BEZ235 (Dactolisib)PI3K, mTORC1/2Phase I for advanced solid tumors
Efficacy in vivo against OS in mice
GDC0980 (Apitolisib)PI3K, mTORC1/2Phase I for advanced solid tumors
SorafenibRTKPhase II for advanced osteosarcoma

Table 1.

Current agents targeting PI3K/Akt pathway with potential against osteosarcoma.

“Advanced” denotes tumors that have metastasized, recurred, failed prior chemotherapies, or are not surgically resectable.


With current chemotherapy regimens for OS having reached seemingly maximum efficacy, attention has been generous in identifying new agents to increase PTEN function in OS. MicroRNA-21 (miR-21) is one of the first mammalian microRNA’s discovered that happens to be highly expressed in bone marrow and post-transcriptionally regulates a number of tumor suppressors including PTEN. miR-21 is overexpressed, suppresses PTEN in OS, and has been identified as a potential therapeutic target [83]. MicroRNAs are small non-coding pieces of RNA, similar to small interfering RNA (siRNAs) that bind to complimentary pieces of messenger RNA, effectively preventing translation of the mRNA into proteins. In multiple myeloma, miR-21 inhibitors have been used in vivo using murine models to cause tumor suppression, with tumors exhibiting increased PTEN and decreased p-Akt levels [84]. Targeting microRNAs could show promise, as numerous microRNAs are specifically upregulated in OS cell lines [44]. In addition to miR-21, miR-17 and miR-221 are also increased and can inhibit PTEN in OS cells and tissues, potentially being therapeutic targets [42, 85]. Further incentive for therapeutic inhibition of miR-221 is its association with cisplatin resistance, an agent frequently included in chemotherapy regimens for OS.

Targeted molecular therapy using RTK inhibitors has been fruitful in many cancers. EGFR causes activation of Akt in OS, and resistance to EGFR inhibitors can be seen in tumors with PTEN deletions leading to unchecked Akt activity. This is encouraging for possible combination therapies that restore PTEN function and inhibit either the PI3K pathway or RTKs. One RTK inhibitor in particular, sorafenib, may have interactions with PTEN. Sorafenib is a small molecule inhibitor of many RTKs including platelet-derived growth factor receptors (PDGFR), VEGF, EGFR, and Raf family kinases and is currently used in advanced renal cell carcinoma, thyroid cancer, and hepatocellular carcinoma. In thyroid cancers in vivo, sorafenib reversed tumor growth that had been attributed to PTEN loss [86]. Interactions between sorafenib and PTEN can also explain why acquired resistance to sorafenib in hepatocellular carcinomas is partly due to miR-21-mediated inhibition of PTEN [87]. Notably, sorafenib is the first targeted chemotherapy agent used for treatment of OS. Sorafenib did demonstrate some activity as a single agent in patients with unresectable OS, with median progression free and overall survival of 4 and 7 months, respectively [88]. Additional effect was seen with combination of sorafenib and the mTOR inhibitor everolimus [78], and further improvement could be achieved with combination therapy specifically targeting PTEN or miR-21.

Demethylating agents may be able to restore PTEN function by removing hypermethylated promoter regions in cancer cells. In many cancerous processes, methyl groups are added throughout the genome preferentially in the promoter regions of tumor suppressor genes, to the five position of cytosine of a CpG dinucleotide (i.e., where a guanine is preceded by a cytosine). These 5-methylcytosines act as roadblocks on a cell’s DNA, preventing transcription of tumor suppressor genes. Although hypermethylation occurs in many cancers, it is difficult to show that this occurs in OS in vivo. 5-Azacytidine is a commonly used demethylating agent approved for treatment of myelodysplastic syndrome. It has been shown in human OS in vitro that the PTEN gene promoter is hypermethylated and that 5-azacytidine treatments activate PTEN expression [89]. Initial uses of 5-azacytidine as an isolated agent in OS were disappointing [90]; however, recent investigations show a role for combination therapies targeting PTEN through epigenetic regulation [91].

Many other specific activators of PTEN have been recently identified, but at this time, they remain tested only in vitro. Tepoxalin, a 5-lipogenase inhibitor, appears to increase PTEN activity by preventing its alkylation or oxidation in canine OS cell lines [92]. Evodiamine, derived from the fruit of the Evodia rutaecarpa plant, inhibits human OS cell proliferation by increasing both protein and gene levels of PTEN [93]. Celecoxib, a cyclooxygenase-2 inhibitor, inhibits hepatocarcinoma tumor growth and angiogenesis in mice by concurrently increasing PTEN and decreasing PI3K levels [94]. Similarly, using a VEGF inhibitor in combination with celecoxib may prove beneficial in human OS [95]. Even caffeine has been shown to induce PTEN activation and many other tumor suppressor genes, while decreasing expression of IL-6 and matrix metallopeptidase 2 [96, 97]. Overall there is much promise for improving OS treatments by honing our understanding of PTEN and the PI3K pathway (Table 2).

AgentTargetEffect on PTEN
None yet availablemiR17, miR21, miR221Reduced inhibition of PTEN by miR
5-AzacytidineDemethylating agentIncreases PTEN in human OS cells
Tepoxalin5-Lipogenase inhibitorIncreases PTEN in canine OS cells
EvodiamineUnknownIncreases PTEN in human OS cells
CelecoxibCyclooxygenase-2 inhibitorIncreases PTEN and decreases PI3K in vitro
CaffeineUnknownActivates PTEN, decreases expression of IL-6 and MMP2

Table 2.

Potential future agents for targeting PTEN in osteosarcoma.

8. Conclusion

The stagnation of current chemotherapy regimens has forced us to look beyond the usual players in oncogenesis. The mentality behind the recent targeted developments against OS involving PI3K and PTEN could be expanded to benefit many tumor types. We have been able to see through the genetic chaos of OS, finding predictability in the form of kataegis and the seemingly random mutations that converge on the PI3K/Akt pathway. Significant inroads still need to be made to clinically validate what has been proven on the bench and in animal models, but sarcomatologists are optimistic that potential therapies and improved patient survival lie within the PI3K and PTEN axis.

© 2017 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite and reference

Link to this chapter Copy to clipboard

Cite this chapter Copy to clipboard

Matthew G. Cable and R. Lor Randall (April 26th 2017). Characterizing Osteosarcoma Through PTEN and PI3K: What p53 and Rb1 Can’t Tell Us, Osteosarcoma - Biology, Behavior and Mechanisms, Kanya Honoki and Kurt Richard Weiss, IntechOpen, DOI: 10.5772/67071. Available from:

chapter statistics

558total chapter downloads

More statistics for editors and authors

Login to your personal dashboard for more detailed statistics on your publications.

Access personal reporting

Related Content

This Book

Next chapter

The Use of Molecular Pathway Inhibitors in the Treatment of Osteosarcoma

By Adel Mahjoub, Jared A. Crasto, Jonathan Mandell, Mitchell S. Fourman, Rashmi Agarwal and Kurt R. Weiss

Related Book

Frontiers in Guided Wave Optics and Optoelectronics

Edited by Bishnu Pal

First chapter

Frontiers in Guided Wave Optics and Optoelectronics

By Bishnu Pal

We are IntechOpen, the world's leading publisher of Open Access books. Built by scientists, for scientists. Our readership spans scientists, professors, researchers, librarians, and students, as well as business professionals. We share our knowledge and peer-reveiwed research papers with libraries, scientific and engineering societies, and also work with corporate R&D departments and government entities.

More About Us