SEARCH

SEARCH BY CITATION

Keywords:

  • bladder cancer;
  • invasion;
  • PI-3 kinase;
  • PTEN

Abstract

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

Three of the studies described in this section relate to bladder cancer. The first of these concerns the PI-3 kinase pathway, which has been a topic of interest in cancer in general. The authors from Sacramento suggest that it may regulate cancer cell invasion, and hope that this may lead to translational therapeutic uses.

Another study describes the pharmacological characteristics of Ro115-1240, which is a selective alpha1A/1L adrenoceptor partial agonist, a compound which may have a future in treating stress urinary incontinence.

OBJECTIVES

To investigate the role of the phosphatidylinositol (PI)-3 kinase pathway in the invasion of bladder cancer cell lines, and to assess the activation of this pathway in primary human bladder tumours.

MATERIALS AND METHODS

Human bladder cancer cells were treated with pathway specific inhibitors or were transfected with PI-3 kinase pathway components. The invasion of cultured bladder cancer cells was analysed by an invasion assay. Bladder cancer cells lines and primary human bladder tumours were analysed for pathway activation by western blotting.

RESULTS

A specific inhibitor of PI-3 kinase enzyme activity, Ly294002, potently suppressed the invasive properties of three highly invasive bladder tumour cell lines. Restoration of the PTEN gene to invasive UM-UC-3 bladder tumour cells or expression of a dominant-negative version of the PI-3 kinase target, Akt, also potently inhibited invasion, indicating a central role for the PI-3 kinase/Akt pathway in this process. In addition, 55% of primary tumours from patients with bladder cancer had markedly high levels of phosphorylated Akt.

CONCLUSION

Pharmacological or biochemical inhibition of the PI-3 kinase pathway drastically reduced the invasive capacity of bladder cancer cell lines; over half of primary human bladder tumours had high Akt phosphorylation, suggesting that the aberrant activation of this pathway may contribute to the invasion of a significant subset of bladder cancers.


Abbreviations
PI

phosphatidylinositol

PDK1

phosphoinositide-dependent protein kinase 1

FBS

fetal bovine serum

GFP

green fluorescent protein

LOH

loss of heterozygosity

DN

dominant negative

PIP-3

phosphatidylinositol 3,4,5 trisphosphate

HA

haemagglutinin

DAPI

4’,6-diamidino-2-phenylindone.

INTRODUCTION

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

Most bladder cancers (≈ 80%) present as superficial disease [1] confined to the bladder mucosa (Ta) or lamina propria (T1) with no muscle invasion. While superficial tumours are amenable to bladder-sparing therapy, at least half of these tumours will recur. Among those that recur 10–20% will progress to muscle-invasive disease. Tumours that have penetrated the lamina propria (T1) progress to muscle-invasive disease more often (25–40%) than those that have not (Ta, 3–4%) [2]. The 20% of tumours that have invaded muscle at the time of diagnosis (T2–T4) often progress more rapidly, and patients have a poor prognosis [1].

Numerous studies reported a strong correlation between tumour grade, depth of invasion and patient survival [3–5]. Indeed, a recent study of 184 patients with muscle-invasive bladder cancer found depth of invasion to be the most significant prognostic factor in determining patient survival [4].

While it is clear that the stage of bladder cancer invasion is critical to the patients’ outcome, there is still much to be learned about how and why a subpopulation of superficial tumours progress to muscle-invasive cancer. As existing therapies for muscle-invasive disease still result in nearly 50% mortality there is much interest in obtaining markers that will identify those superficial tumours that are at risk of progression. Moreover, an in-depth understanding of the molecular mechanisms underlying bladder cancer invasion will ultimately facilitate the development of more reliable drugs and treatment strategies that can be used to suppress the progression of the disease.

Cytogenetic analyses of bladder tumours show a spectrum of molecular changes, yielding important clues about tumour initiation and progression. Loss of heterozygosity (LOH) at 10q is a common event in the later stages of bladder cancer and is found predominantly in muscle-invasive tumours [6–8].

PTEN (MMAC1/TEP1), a tumour-suppressor gene that maps to chromosome 10q23, is a dual-specificity lipid phosphatase that antagonises the phosphatidylinositol (PI)-3 kinase pathway by dephosphorylating its end product, the lipid second-messenger phosphatidylinositol 3,4,5 trisphosphate (PIP3) (Fig. 1). Mutations and deletions in PTEN have been found in numerous cancers, including glioblastomas, prostate and endometrial carcinomas [9,10]. Recent studies indicate that PTEN is mutated or deleted in a significant proportion (14–23%) of invasive bladder cancers with LOH at 10q [7,11]. Bladder tumours have also been reported in patients with Cowden disease, an autosomal dominant cancer-predisposition syndrome resulting from germ line mutations in PTEN[12,13]. In addition, primary human bladder tumours of all stages overexpress PI-3 kinase and have significantly higher PI-3 kinase activity (5–20-fold) than adjacent normal epithelium [14]. These findings indicate that the PI-3 kinase pathway is aberrantly activated in bladder tumours and may be involved in bladder tumorigenesis and/or progression.

image

Figure 1. A model for the contribution of the PI-3 kinase/Akt pathway to bladder cancer. PI-3 kinase in tumour cells may be aberrantly activated by various mechanisms, including overexpression of the catalytic subunit p110, PTEN loss, or through autocrine signalling loops. PI-3 kinase produces the phosphorylated lipid second messenger PIP3 that in turn activates Akt. Akt kinase activity leads to the phosphorylation and regulation of target proteins that regulate cellular survival, migration, invasion and resistance to chemotherapeutic agents.

Download figure to PowerPoint

The PI-3 kinase pathway regulates key processes that are relevant for tumour growth and progression, including cell-cycle progression, survival, motility, invasion and angiogenesis (Fig. 1) [15,16]. The serine/threonine kinase Akt is a major downstream target of PI-3 kinase and is important in each of these processes [17]. Akt is recruited to the plasma membrane after PI-3 kinase activation through the interaction of its PH domain with PIP3. Then PIP3 binding alters the conformation of Akt, allowing phosphorylation of the activation loop threonine residue (Thr308) by phosphoinositide-dependent protein kinase 1 (PDK1) [16]. This event then triggers autophosphorylation in the hydrophobic motif (Ser473) [18], rendering Akt fully active. Akt has recently emerged as the critical downstream mediator of motility in growth factor-stimulated cells, in cells expressing activated alleles of the small GTPases Rac and Cdc42, and in PTEN-deficient mouse embryo fibroblasts [19]. Akt is also essential for the growth of PTEN-null embryonic stem cells as aggressive teratomas in mice [20]. Studies with human tumours have found that Akt phosphorylation is increased in advanced prostate, breast and ovarian cancers, indicating that Akt may be important in progression in these tumours [21]. As the acquisition of invasive behaviour is a crucial event in bladder cancer progression, we investigated the role of the PI-3 kinase pathway in this process in bladder cancer cell lines and examined primary human bladder tumours.

MATERIALS AND METHODS

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

Primary human bladder cancer specimens and normal tissue were obtained from patients who underwent surgery at the UC Davis Cancer Center. Of the 23 bladder specimens, five were from cystectomy and 18 from transurethral resections. For the cystectomies, immediately after the bladder was removed it was taken to pathology where the tumour was removed and snap-frozen. With the specimens retrieved through transurethral resection, tissue was removed from the bladder at the start of the procedure, a scalpel used to trim off the surface where cautery was applied, then the specimen placed in saline on ice and immediately transported to the pathology laboratory, where it was snap-frozen. Tissue samples were stored at −80°C. Slides from each tumour sample were reviewed by a pathologist to establish the stage and grade. Each tumour sample contained ≥ 90% tumour cells, as determined by histological examination. Tumour samples were weighed and solubilized in tissue protein extraction reagent (Pierce, Rockford, IL). Insoluble matter was removed by microcentrifugation and the protein content determined using a Coomassie protein assay reagent (Pierce).

The human bladder cancer cell lines RT4, J82, T24 and UM-UC-3 were obtained from American Type Culture Collection. Cells were maintained in RPMI containing 10% fetal bovine serum (FBS). The immortalized normal human urothelial cell line SV-HUC-1 was obtained from the same source and maintained in Ham's F12 medium with 7% FBS.

CELL ASSAYS

For phospho-Akt and phospho-Erk1/2 assays, cells were seeded in 12-well dishes, grown to ≈ 60% confluency and serum-starved overnight. Cells were then treated without (DMSO only) or with pathway inhibitors (dissolved in DMSO), and were lysed in Laemmli sample buffer. Proteins in lysates were resolved by 8% SDS-PAGE, transferred to nitrocellulose and blotted with antibodies to phospho-Akt (S473) and phospho-Erk1/2 (T202/Y204; Cell Signalling Technology, Beverly, MA). Blots were then stripped with stripping buffer (Pierce) and re-probed with antibodies to Akt or Erk1/2, or actin (clone AC-15; Sigma Chem Co., St Louis, MO) or anti-cytokeratin-18 (Progen Industries Ltd., Qld, Australia). For cell viability assays cells were grown to ≈ 60% confluence, serum-starved for 16 h and treated with either DMSO or increasing concentrations of LY294002 or U0126 for 72 h. Thiazolyl blue was added to the media (to 0.5 mg/mL) and cells incubated for 3 h. The medium was removed and converted dye was solubilized with acidic isopropanol (0.1 mol/L HCl in isopropanol). The absorbance of the converted dye was measured at 570 nm. Concentrations of pathway inhibitors used for the subsequent invasion assays were: T24; 10 µmol/L Ly294002, 5 µmol/L U0126: UM-UC-3; 10 µmol/L Ly294002, 5 µmol/L U0126: J82; 25 µmol/L Ly294002, 12.5 µmol/L U0126.

For the transfection experiments, UM-UC-3 cells were transfected using Fugene 6 (Roche, Basel, Switzerland) according to the manufacturer's instructions. Cells in 100 mm dishes were cotransfected with 10 µg of pAdTrack-CMV [22] for green fluorescent protein (GFP) expression together with 5 µg of pBabe puromycin resistance vector or rat HA epitope tagged Akt (K179M/T308A/S473A; [23]) in pAdTrack-CMV, together with 5 µg of pBabe puromycin resistance vector, or pGFP-PTEN (pGZ21 δ xZ; [24]) together with 5 µg of pBabe puromycin resistance vector [24]. Cells were subcultured at a 1 : 3 dilution 24 h after transfection and maintained for 3 days in 1 µg/mL puromycin-containing medium to eliminate untransfected cells. For Western blotting, cells were lysed in Laemmli sample buffer and lysates processed by SDS-PAGE, followed by electrophoretic transfer to nitrocellulose. PTEN was detected with anti-PTEN (♯9522, Cell Signalling Technology) and haemagglutinin (HA) epitope-tagged dominant-negative (DN) (HA)Akt was detected with anti-HA antibody (Boehringer Mannheim, Germany). Blots were visualized by enhanced chemiluminescence using West-Femto chemicals (Pierce) and an imaging station. For invasion assays, cells were serum starved in RPMI, 0.1% BSA during the last 24 h of puromycin selection to exclude any effects of growth suppression on cell invasion.

INVASION ASSAYS

Invasion was assayed using Biocoat Matrigel Invasion Chambers (24-well size, Becton-Dickinson, Franklin Lakes, NJ) according to the directions of the manufacturer. All experiments were carried out in triplicate. Cells were serum-starved in RPMI, 0.1% BSA and seeded into the upper well of the invasion chamber in RPMI, 0.1% BSA. RPMI, 10% FBS was added to the bottom well of the chamber to serve as a chemoattractant. For pharmacological inhibition assays 10 000 cells/well were seeded in the presence (both top and bottom chamber) or absence of inhibitors (treated with DMSO only), and invasion assessed after 24 h by staining membranes with Diff-Quik cell staining kit (Fisher, Hamilton Thorne Biociences, Beverly, MA). At least five independent fields were counted for each chamber. For transfection experiments, 10 000 puromycin-selected cells were seeded into each chamber. Invasion was assessed after 24 h by staining membranes with Diff-Quik; again, at least five independent fields were counted for each chamber. Invasion of GFP transfected cells was normalized to 1 and results for others were reported relative to GFP cells (relative invasion).

In each case, cells were also seeded into 12-well plates and onto cover slips. Cover slips were stained with 4’,6-diamidino-2-phenylindone (DAPI) to visualize nuclei, and then examined by fluorescence microscopy for any evidence of nuclear changes indicative of apoptosis (condensed or fragmented nuclei). Cells in 12-wells were counted and viability assessed using the thiazolyl blue assay, as described above. None of the treatments affected cell number or viability during the 24-h period of the invasion assay.

RESULTS

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

The three cultured cell line models of invasive bladder cancer chosen, T24, J82 and UMUC-3, are widely used and representative of advanced bladder cancer, harbouring either PTEN mutation (T24 [11]), or deletions (J82, UM-UC-3 [8,11]). As controls we selected SV-HUC-1 cells, an immortalized but not tumorigenic urothelial cell line [25], and RT4 cells, a transformed but not tumorigenic or invasive cell line widely used as a model of superficial bladder cancer [26]. First it was confirmed that the invasive cells exhibit constitutive activation of the PI-3 kinase pathway by blotting equivalent amounts of lysate from each cell line with a phospho-specific antibody (Ser473) directed toward the activated form of Akt. Phosphorylation of Akt at Ser473 occurs in response to PI-3 kinase activation, and correlates precisely with Akt kinase activity [21]. Figure 2 shows that the invasive cell lines T24, J82 and UM-UC-3 all contain phosphorylated Akt after overnight serum starvation, while the normal cell line SV-HUC-1 and the noninvasive cell line RT4 had very low levels of phospho-Akt. Because the Ras/Erk1/2 pathway has also been implicated in invasion in some cell types [27–31] the activation of this pathway was examined using a phospho-specific antibody (T202/Y204) that recognises the activated, dually phosphorylated forms of p42 (Erk2) and p44 (Erk1). Varying levels of phospho-Erk were found in all five cell lines, but the differences did not correlate with invasive ability. Particularly notable was the SV-HUC-1 cells; despite a high degree of Erk1/2 activation, these cells are not invasive. These observations, summarized in Table 1, indicate that the aberrant activation of the PI-3 kinase pathway but not the Erk1/2 pathway correlates with the invasive potential of bladder tumour cells.

image

Figure 2. The phospho-Akt and phospho-Erk content of urothelial cell lines. Lysates from serum-starved cells were blotted with antibodies to phosphorylated and unphosphorylated forms of Akt and Erk, as indicated.

Download figure to PowerPoint

Table 1.  Correlation of pAkt content and invasiveness of urothelial cell lines
Cell linePTENpErkpAktInvasion
SV-HUC-1Normal + + +
RT4Normal + +
J82Deleted + + + + + +
T24Mutated + + + + + +
UM-UC-3Deleted + + + + +

PHARMACOLOGICAL INHIBITION OF THE PI-3 KINASE PATHWAY

Bladder cancer cell invasion by the model cell lines was assessed using the Matrigel invasion-chamber assay, which measures two important factors that contribute to tumour malignancy, motility and invasiveness. Cells must first degrade a reconstituted basement membrane that occludes the pores of a filter and then move through these cleared pores to adhere to the other side. As expected, the SV-HUC-1 and RT4 cells were completely noninvasive using this assay, while the T24, J82 and UM-UC-3 cells were invasive (Fig. 3B). A highly specific pharmacological inhibitor of PI-3 kinase catalytic activity, LY294002 [32], was used as the initial approach to assess the importance of the PI-3 kinase pathway in invasion. As a control the U0126 inhibitor of MEK1/2 was also included, a necessary component of the Erk pathway. As both PI-3 kinase and Erk pathways have roles in cellular growth and viability [33,34] these inhibitors may have indirect effects in this assay by affecting these properties. Hence, for each cell line an optimal concentration had to be established for each inhibitor that maximally suppressed its target pathway, but maintained cellular viability over the course of a 24-h invasion assay. Thus, we constructed a dose–response curve for each drug and examined both pathway inhibition and effects on cellular viability. In the experiment shown in Fig. 3A lysates from inhibitor-treated cells were blotted with antiphospho-Akt (left panels) or antiphospho-Erk1/2 (right panels). While LY294002 inhibited Akt phosphorylation as expected, it had no effect on Erk1/2 phosphorylation. Likewise, U0126 inhibited Erk1/2 phosphorylation as expected but had no effect on Akt phosphorylation. Only at the highest inhibitor concentrations were there any deleterious effects on cellular viability (data not shown).

image

Figure 3. Inhibition of invasion by Ly294002. (A) Cell lines were treated with increasing concentrations of Ly294002 or U0126, and equivalent amounts of lysates blotted with antibodies to phosphorylated forms of Akt and Erk, as indicated. (B) Cells were treated with DMSO (open bars) or with optimal concentrations of Ly294002 (green bars) or U0126 (red hatched bars), and the effect on invasion measured.

Download figure to PowerPoint

Concentrations of each inhibitor were chosen for each cell line that inhibited the target pathway by > 90% while maintaining > 90% cellular viability. Importantly, in parallel experiments, these inhibitor concentrations were determined to have no effect on cell number or proliferation during the period used for the invasion assay (data not shown). Figure 3B shows the effect of Ly294002 and U0126 on cellular invasion. While inhibition of the Erk1/2 pathway had minimal effects on the ability of the three cell lines to invade, inhibition of the PI-3 kinase pathway potently inhibited invasion by all three lines.

BIOCHEMICAL MANIPULATION OF THE PI-3 KINASE PATHWAY

While the results shown in Fig. 3 provide strong evidence for an essential role of the PI-3 kinase pathway in bladder cancer cell invasion, we sought to confirm these results independently by inhibiting the PI-3 kinase pathway with the expression of two negative regulators, the antagonist PTEN or a dominant negative version of Akt (K179M/T308A/S473A). UM-UC-3 cells were cotransfected with pAdTrackCMV (for GFP expression) with a vector encoding puromycin resistance (pBabe) or with DN-(HA)Akt in pADTrackCMV [23], with pBabe or with a vector (pGZ21 δ xZ, 24) expressing a GFP-PTEN fusion protein along with pBabe. Cells were placed in puromycin-containing media for 3 days to select cells expressing both plasmids, and then the same number of cells seeded into invasion chambers in triplicate. Puromycin selection eliminates untransfected cells, resulting in a pure population of expressing cells. As the expression of PTEN or DN-Akt in UM-UC-3 cells may inhibit their growth [35], cells were serum-starved during the last 24 h of puromycin selection to suppress growth, excluding contributions of growth suppression in the invasion assay. This approach was used by Tamura et al.[24] to examine the effects of PTEN expression on invasion of PTEN-null glioblastoma cells. After selection, transfected cells were also seeded onto coverslips and stained with DAPI, which revealed no evidence of condensed or fragmented nuclei (morphological changes indicative of apoptosis [36,37] in the GFP, GFP-PTEN or GFP/DN-(HA)Akt transfectants (data not shown).

In addition, in parallel experiments, it was determined that transfection of PTEN or DN-Akt had no significant effect on cellular number or cellular viability of selected cells during the 24-h period of the invasion assay (data not shown). In a separate study, adenoviral-mediated overexpression of PTEN in UM-UC-3 cells reduced cell numbers by 40% at ≥ 4 days after infection [35]. Increasing levels of PTEN were associated with more growth inhibition. While there was no growth inhibition by PTEN during the 24-h period used it is likely that the expression levels of PTEN achieved did not approach those obtained with adenoviral-mediated overexpression.

Figure 4 shows the expression of GFP, PTEN and DN-(HA)Akt in UM-UC-3 cells visualized by Western blotting. The expression of PTEN abolished Akt phosphorylation while expression of DN-(HA)Akt reduced Akt phosphorylation by 70%. The extent of inhibition of invasion by PTEN and DN-(HA)Akt was quantified, with GFP invading cells normalized to 1 (Figure 4B). The expression of PTEN inhibited the invasive properties of UM-UC-3 cells by 95% while expression of DN-(HA)Akt inhibited invasion by 91%. This experiment was repeated at least three times and these results are representative. These results confirm the observations from pharmacological inhibition, and together indicate that the PI-3 kinase pathway is essential in the invasive behaviour of bladder cancer cells. These results further show that activation of a specific pathway component, Akt, is necessary for efficient invasion.

imageimage

Figure 4. Inhibition of UM-UC-3 invasion by PTEN and DN Akt. (A) Cells were transfected with pADTrack-CMV together with pBabe puromycin resistance vector (GFP) or with GFP/DN-(HA)Akt or GFP-PTEN as indicated. Total cell lysates were blotted with either anti-PTEN, anti-HA, anti p-Akt or anti-actin. (B) Cells were transfected with GFP, GFP-PTEN or GFP-DN-(HA)Akt together with pBabe puromycin resistance as indicated, seeded into invasion chambers in triplicate, and invasion measured. The relative number (SE) of invading cells is plotted for each condition.

Thus we sought to determine if Akt phosphorylation was elevated in primary human bladder tumours. Twenty tumours of various stage/grade and three normal tissue samples were examined for evidence of Akt phosphorylation by immunoblotting lysates with antiphosphoAkt (S473). Samples were simultaneously probed with anticytokeratin 18 to normalize for epithelial content. As shown in Fig. 5, 11 of 20 (55%) tumour samples had significantly higher Akt phosphorylation than normal controls, indicating that Akt is highly phosphorylated in a subset of primary human bladder tumours. Positive samples included both superficial tumours (five of 11) of all grades and invasive (six of 11) tumours of low and high grade. As the activation of Akt is necessary for efficient bladder cancer cell invasion, and tumour invasion correlates strongly with poor patient prognosis, these results suggest that activation of the PI-3 kinase/Akt pathway may be important in the progression of some bladder tumours.

image

Figure 5. Phospho-Akt content was higher in a subset of bladder tumour samples. Lysates of surgically excised tumours or normal bladder tissue (N) were blotted with antibodies to phospho-Akt (S473) or to cytokeratin 18 (Ck18) to normalize for epithelial content. Stage and grade are indicated above each sample. LG, low grade; IG, intermediate grade; HG, high grade.

Download figure to PowerPoint

DISCUSSION

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

We assessed the possibility that the PI-3 kinase pathway actively contributes to bladder cancer progression by regulating invasive behaviour. The PI-3 kinase pathway has been shown to increase the proliferation, survival and motility of various cell types, and has been implicated in the vascularization and invasion of several solid tumour types [15]. Accumulating evidence indicates a potential role for the PI-3 kinase pathway in bladder cancer. Mutations and deletions in the negative regulator PTEN have been identified in a significant population of muscle-invasive bladder tumours, and p110 overexpression and elevated PI-3 kinase activity are found in both superficial and invasive primary human bladder tumours.

These observations suggest that the aberrant activation of the PI-3 kinase pathway by any number of mechanisms could contribute to the transition of superficial tumours to muscle-invasive cancer. The present study tested this hypothesis by examining the effect of pathway inhibition on the invasive properties of cultured bladder tumour cell lines. The highly invasive T24, J82 and UM-UC-3 bladder tumour cell lines harbour mutations or deletions of PTEN, and each had a high level of PI-3 kinase activity, as assessed by Ly294002-inhibitable Akt phosphorylation. Ly294002 potently suppressed the invasive properties of each of these cell lines, and reconstitution of intact PTEN into UM-UC-3 cells also efficiently suppressed invasion. This indicates that the 3-phosphorylated lipid products of PI-3 kinase are necessary for cellular invasion. A DN version of Akt significantly inhibited the invasive properties, indicating that Akt kinase activity is necessary for this process. Finally, more than half of the bladder tumours examined had aberrant phosphorylation of Akt.

While increased cellular motility mediated by PI-3 kinase has been implicated in the invasive properties of several tumour types, a role for Akt activation in tumour invasion has only recently been appreciated. DN-Akt suppresses the invasive behaviour of the human fibrosarcoma cell line HT1080 and the murine Lewis lung carcinoma cell line H-59 [38,39]. In addition, constitutively active Akt increases the invasiveness of pancreatic tumour cells [40], and induces the invasive behaviour of immortalized mouse mammary epithelial cells [41].

While the PI-3 kinase/Akt pathway is clearly important in tumour invasion, several other pathways are known to be involved in invasion. Elevated expression of focal adhesion kinase has been found in invasive breast, colon and oral tumours [42,43], and over-expression is correlated with increased malignancy [44]. c-Jun N-terminal kinase activation downstream of the focal adhesion kinase is required for invasion [45]. Inhibition of phospholipase C-γ, a downstream target of activated growth factor receptors, blocks prostate [46], glioma [47] and bladder cancer cell invasion [48]. The Ras/Erk1/2 pathway is involved in invasion in many cell types, including breast cancer cells [27,28], fibroblasts [29] and prostate cancer cells [30], and activation of this pathway is sufficient to induce invasive behaviour of mouse mammary epithelial cells, both in vitro and in vivo[31].

However, in the present study, inhibition of the Erk1/2 pathway only marginally affected invasion of the J82, T24 and UM-UC-3 cells, and in a separate study, urothelial expression of an activated Ha-Ras in transgenic mice yielded urothelial hyperplasia and superficial tumours but did not result in invasive tumours [49]. While the present results strongly implicate the PI-3 kinase/Akt pathway, four of 10 invasive tumours tested had little or no Akt phosphorylation, indicating that invasion can occur in the absence of aberrant PI-3 kinase/Akt pathway activation.

Recent studies indicate a role for the PI-3 kinase/Akt pathway in therapeutic resistance in numerous tumours, including lung, ovary, pancreas [50–52] and most recently, bladder [35]. Over-expression of PTEN and diminution of Akt phosphorylation restored doxorubicin sensitivity to the doxorubicin-resistant UM-UC-6dox bladder tumour cells. The studies outlined here suggest that the PI-3 kinase/Akt pathway could be a target in the development of bladder cancer therapies. As the pathway appears to be involved in several cellular processes it will be important to identify biochemical and gene targets of PI-3 kinase/Akt activation that specifically lead to invasion and chemoresistance.

The present results suggest that the PI-3 kinase/Akt pathway could actively contribute to the process by which superficial TCC progresses to muscle invasive disease. They also point to the possibility that phospho-Akt might be used as a marker for those superficial tumours that are at risk of progression to invasive tumours. Indeed, immunoblotting experiments showed that of nine superficial tumours tested, five had high Akt phosphorylation. While the pathological variables of tumour grade and stage provide important prognostic information [5,53], much focus is now on identifying biological markers that will fully predict the malignant potential of tumours. Current markers with prognostic value include the tumour suppressors p53 [54,55] and Rb [53,56]. Future studies with a clinical follow-up will seek to determine if those superficial tumours that have increased Akt phosphorylation are those that progress, and whether Akt phosphorylation provides additional or independent information.

ACKNOWLEDGEMENTS

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

We gratefully acknowledge Dr Kenneth M. Yamada for kindly providing the PTEN expression plasmid and Dr Paola Marignani for providing the pBabe plasmid. This work was supported by a grant from the California Tobacco-Related Disease Research Program (11KT-0236) to C.S.

REFERENCES

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