REVIEW Modulation of the immune response and tumor ... - Nature

2 downloads 16 Views 190KB Size Report
Cancer Immunology Program, Cardinal Bernardin Cancer Center, Loyola University Chicago, 2160 ... Keywords: Ras; leukemia; immune escape; tumor growth.
Leukemia (1999) 13, 502–513  1999 Stockton Press All rights reserved 0887-6924/99 $12.00 http://www.stockton-press.co.uk/leu

REVIEW Modulation of the immune response and tumor growth by activated Ras S Weijzen, MP Velders and WM Kast Cancer Immunology Program, Cardinal Bernardin Cancer Center, Loyola University Chicago, 2160 South First Avenue, Maywood IL 60153, USA

As a result of its transforming abilities, activated Ras is expressed in a great number of cancers. The ras mutation frequency varies between 95% in pancreatic cancer and 5% in breast cancer. In leukemia, the highest frequency (30%) is found in acute myeloid leukemia. The presence of ras mutations has been correlated with a poor prognosis and negative clinical outcome. This suggests that mutated Ras activates mechanisms, which favor tumor growth, enhance the metastatic capacity of tumors or modulate tumor-specific immune responses. Several new functions of Ras, such as downregulation of major histocompatibility complex molecules, upregulation of certain cytokines, growth factors and degradative enzymes have been uncovered in the last decade. Additionally, mutated Ras can also serve as a primary target for the development of immunotherapy or drug therapy. This review will discuss the mechanisms by which Ras expressing tumors are able to evade destruction by the immune system and enhance their growth and metastatic potential. It will further elaborate on the attempts to develop successful immunotherapy and drug therapy targeting Ras expressing tumors. Keywords: Ras; leukemia; immune escape; tumor growth enhancement; metastatic potential

Introduction Mutations in the ras proto-oncogene have been implicated in a large variety of human malignancies.1 High frequencies of ras mutations have been found in pancreas, colorectal and thyroid carcinomas. In myeloid leukemia, the overall frequency of ras mutations is approximately 30%.1 These mutations result in a constitutively active Ras protein, which disrupts the normally tightly regulated Ras-dependent signal transduction pathways. This disruption can have several consequences, including transformation. The fact that deregulation of Ras activity is a widespread phenomenon in a large number of cancers would suggest that the presence of activated Ras can result in a growth advantage, by stimulating tumor growth and/or by modulating the immune response against the tumor cells. Indeed, studies investigating the effect of the ras mutations on tumor growth characteristics, potential to metastasize and prognosis suggests a correlation between Ras expression and clinical outcome (Table 1). A number of studies have attempted to determine if Ras mutations have an influence on prognosis, disease-free and overall survival in several forms of leukemia. These studies indicated that the presence of ras mutations in leukemia had a negative impact on survival. Particularly in myelodysplastic syndrome (MDS), N-ras mutations were associated with a poor prognosis and an increased risk at progression to acute myeloid leukemia (AML).2–4 A minor number of studies claim that ras mutations have no effect on clinical outcome in myeloid leukemia.5–7

Correspondence: WM Kast; Fax: 17083273238 Received 9 November 1998; accepted 15 December 1998

However, in these studies a smaller number of ras-positive patients was analyzed, which reduces the power needed to determine the prognostic significance of ras mutations in hematological malignancies. Comparable studies have been performed in other types of cancer. In breast cancer, more invasive tumors express higher levels of Ras than less aggressive tumors.8 In colorectal cancer, a comparison of the K-ras mutation rate between patients with and without recurrent disease showed that 71% of patients with recurrent disease demonstrated a K-ras mutation vs only 25% of patients who were disease-free 5 years after surgery, a highly significant difference.9 In lung carcinoma, mutations in K-ras have also been related to poorer prognosis and shorter survival.10–12 When benign pancreatic adenomas were compared with pancreatic carcinomas only the latter exhibited 100% K-ras mutations. No mutations were found in benign lesions, suggesting that activated Ras is correlated with a malignant phenotype.13 In other studies K-ras mutations in pancreatic and colorectal carcinomas were not found to correlate with decreased patient survival.14–16 These studies analyzed a smaller number of patients with Ras expressing tumors and as a consequence the observed trend toward a poorer prognosis in patients with Ras mutations was found not to be significant. The correlation between the presence of activated Ras in these tumors and their poorer prognosis can be linked to several mechanisms initiated by Ras. This review will discuss the current knowledge on the modulation of tumor growth and immune escape mechanisms by activated Ras and possible strategies to develop treatments targeting tumors expressing activated Ras. Ras characteristics The three human ras genes encode 21 kDa proteins: H-Ras, K-Ras and N-Ras. The ras genes show a high degree of homology in their sequence: the first 164 amino acids are homologous and the cysteine, located four amino acids from the Cterminus is identical in all three proteins. Ras proteins are expressed in most adult and fetal tissues, although the level of expression may vary among different tissues and different Ras protein isoforms.17 For example, H-Ras expression is very high in skin and skeletal muscle, whereas expression of K-Ras is high in gut and thymus and N-Ras expression is highest in testis and thymus.18 Ras is a protein, which can bind to guanosine triphosphate (GTP). This activates the Ras protein and hydrolysis of GTP into guanosine diphosphate (GDP) and a phosphate will inactivate Ras. The activity of Ras is regulated by two sets of proteins: guanine–nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs).19 GEFs, like Sos1/2, stimulate the release of the Ras bound GDP.20 This results in binding of GTP to Ras because of a natural excess of GTP in the

Review S Weijzen et al

Table 1

503

Studies analyzing the correlation between ras mutations and their clinical effect

Cancer type

Ras gene

Number analyzed

Mutation frequency (%)

Clinical effect

Ref.

MDS

N, K, H

75

36/75 (48)

MDS

N

70

6/70 (9)

MDS

N

50

21/50 (42)

43% progressed to AML

MDS

N

220

20/220 (9)

Increased risk of progression to AML. Shorter survival

101

MDS

N, K, H

51

5/51 (10)

Possible shorter survival in patients with codon 12 mutations (P = 0.1)

102

MDS AML JCML

N, K, H

120

H-ras 8/120 (7) AML: K-ras 2/67 (3) MDS/JCML: N-ras 2/41 (5)

AML AML

H N

30 40

13/30 (43) 8/40 (20)

Increase in risk of progression to AML. Decreased 10 year survival

2

Ras mutation correlated with disease progression Once mutation emerges a shorter survival

7

No clinical effect

3

6

Risk of relapse after remission Decrease in disease-free and overall survival

103 4

AML

N

55

8/55 (15)

No clinical effect

104

AMM Colorectal

N K

43 78

2/43 (4.6) 26/78 (33)

No clinical effect Decrease in survival if mutation was Glu → Val

5 105

Colorectal

K

229

93/229 (41)

Decrease in survival in stage II patients

106

Colorectal Colorectal

K K

63 191

20/63 (32) 62/191 (32)

No difference in prognosis No prognostic effect

15 16

Pancreatic Lung

K K

19 84

19/19 (100) 5/84 (6)

All tumors with K-ras mutations are malignant Decreased survival (P = 0.08)

13 12

Lung

N, K, H

66

K-ras 11/66 (17) N-ras/H-ras 8/66 (12)

Lung

K

69

19/69 (28)

cell. GAP proteins, like neurofibromin 1 (NF1), accelerate the hydrolysis of GTP into GDP.19 Therefore, the GAP proteins are negative regulators of Ras functions. This balance of GTP binding and hydrolysis tightly regulates the functions of Ras and disturbances of this balance have been implicated in many types of cancer.

Shorter survival Shorter overall survival. Strong unfavorable prognostic factor

107 10

depending on the balance of downstream effectors activated by Ras.18 Other downstream kinase molecules are the c-jun N-terminal kinase (JNK).25 A schematic representation of the Ras signaling pathway is presented in Figure 1. Ras and cancer

Functions of Ras Although Ras functions are complex and only partially understood, it is known that Ras can only be functional when it is associated with the plasma membrane. Cytokines such as interleukin-2 (IL-2)21 and growth factors like epidermal growth factor (EGF) or the platelet-derived growth factor (PDGF)22 signal via Ras. Upon ligand binding, the guanine–nucleotide exchange factor Sos1/2 translocates to the membrane and activates Ras.22 Consequently, Ras activates the appropriate effector proteins. Downstream effectors of Ras proteins are, among others, phosphatidyl inositol-3-kinase and Raf-1.23 The function of both proteins is not completely clear. It is believed that Ras activation causes Raf-1 translocation to the plasma membrane, where Raf-1 kinase activity is activated. Raf-1 is then thought to activate other kinases like ERK-1 and ERK-2.24 In turn, these proteins will activate other MAP kinases that are transported to the nucleus, where they regulate activity of several transcription factors like AP-1 (the jun/fos complex).22 The transcriptional activation of various target genes results in control of cell growth, cell differentiation or apoptosis,

The most common ras mutations have been found in codons 12, 13 and 61, which lock the GTP binding protein into a permanent activation state.1 These mutations are mostly substitutions, which render Ras proteins resistant to GAPs and consequently prevent hydrolysis of GTP into GDP.18 This continuously activated Ras protein can autonomously stimulate cell growth or differentiation by stimulating its downstream effectors. Apart from direct mutations in the ras gene, other events have been reported to affect the regulation of Ras. Constitutive activation of GEFs leads to a continuous state of activation for Ras proteins without a mutation in the ras gene itself. Furthermore, loss of GAPs, which normally catalyze the hydrolysis of Ras-GTP into Ras-GDP, can result in constitutive association of Ras with GTP and therefore activation of Ras. An example of this deregulation will be discussed in relation to the development of leukemia. Another factor important for Ras regulation is the upstream signal resulting in Ras activation. Overexpression or truncation of certain growth factor receptors, like the EGF receptor and the PDGF receptor, that act through the Ras signaling pathway, may also result in con-

Review S Weijzen et al

504

Figure 1

Schematic representation of the normal signaling pathway through Raf.

tinuous activation of Ras. These activating events have been reported in breast tumors,26 and in pancreatic cancer.14 Ras activation in leukemia In several forms of leukemia Ras was found to be deregulated as the result of different events. The most common mechanism is a mutation in certain regions of the ras gene, as mentioned before. The frequency of the mutations found in leukemia patients varies between the different forms of leukemia (Table 1). The highest ras mutation rate is found in MDS and AML with a frequency of approximately 30%.1 Disturbance of upstream or regulator proteins of Ras can also affect Ras activation (Figure 2) and have been linked to cancer development, ie deregulation of GAPs. Loss of NF1 in certain forms of leukemia was described to correlate with upregulation of Ras activity.27 Neurofibromatosis patients, who have lost one wild-type NF1 allele, are more likely to develop myeloid leukemia, especially juvenile chronic myeloid leukemia (JCML) by loss of the remaining wild-type NF1 gene.28

Figure 2

Another mechanism by which upstream proteins can upregulate Ras activity is through disruption of the receptors that activate the Ras signaling pathway. In chronic myelomonocytic leukemia (CMML) PDGF receptor (PDGF-R) is affected by the t(5;12) translocation, which is found in a subset of CMML patients. This translocation fuses PDGF-R to Tel, a transcription factor. It is speculated that Tel dimerizes the PDGF-R, resulting in continuous activation of this receptor in the absence of a ligand and consequently in a constitutive activation of Ras.29 Furthermore, wild-type Ras can be permanently activated through the BCR-ABL fusion protein. BCR-ABL is fused as a result of the t(9;22) translocation, which is found in a subset of chronic myeloid leukemia (CML) patients. It has been shown that BCR-ABL can activate Ras through permanent activation of Sos1.30 Downstream of the Ras signaling pathway, Bcl-2 is induced. Bcl-2 is involved in inhibition of programmed cell death (apoptosis) and Ras-induced Bcl-2 prevents apoptosis in hematopoietic cells, resulting in tumor growth.

Deregulation of Ras activation in different forms of leukemia.

Review S Weijzen et al

Immune responses against Ras expressing tumors Immune responses observed against tumors are mostly based on the activation of T cells. As opposed to B cells, which preferentially recognize surface antigens, T cells can monitor the inside of the cells through interaction between the T cell receptor (TCR) and peptides presented by MHC molecules. These peptides result from degradation of intracellular proteins or extracellular proteins (reviewed in Ref. 31). T cells are divided into two major T cell populations. Cytotoxic T lymphocytes (CTLs) are CD8+ and recognize peptides in complex with MHC class I molecules. Under the appropriate circumstances, CTLs will be activated and are able to kill the cell expressing the peptide. T helper cells, which are CD4+, recognize peptides on the cell surface associated with MHC class II. Upon activation of T helper cells, they will secrete cytokines which can result in lysis of the target cell directly (TNF-␣) or cytokines which will attract or activate nearby CTLs (IFN-␥ and specific interleukins). Each T cell has its own specificity determined by the TCR and only non-self peptides presented in self MHC are able to activate the CTL or T helper cell through the TCR. Peptides recognized by CTLs are called CTL epitopes and peptides recognized by T helper cells are called T helper epitopes.

Ras-specific T cell responses in mice Since in some tumors Ras is activated by a mutation in one of its genes, a tumor-specific mutant protein arises. Theoretically, mutated Ras proteins are a source for unique T cell epitopes, which are not present in the wild-type Ras proteins. Considerable effort has been put into the identification of these epitopes, since they represent the targets for possible immunotherapy in patients expressing mutated Ras proteins. CD4+ T cell responses were first found when mice were immunized with mutant Ras peptides or mutated Ras protein.32–34 This T cell reactivity was only observed when mice were immunized with the mutant ras peptides. No cross reactivity against the wild-type ras could be detected.34 The peptide immunization resulted in a T helper 1 response as measured by production of IL-2, IFN-␥, TNF and GM-CSF and was MHC class II restricted. Moreover, the CD4+ T cells demonstrated cytolytic activity against targets loaded with the mutant Ras peptide and tumor cells transfected with K-ras encoding the corresponding point mutation.32 In 1993 the first CTL epitope in the Ras protein was identified when mice were immunized with DNA encoding for NRas with and without a mutation at codon 61. CTLs isolated from the mice immunized with the mutated peptide only recognized the mutant Ras protein. The CTL epitope recognized was mapped to amino acids 60–67, with the mutated amino acid at position 61.35 More recent studies identified murine T cell epitopes in K-Ras mutated at codon 12. T cell epitopes were mapped to K-Ras4–12 (Val12) and K-Ras5–17 (Val12), which were recognized by CTL and T helper cells respectively.32,36 Important to notice is that K-Ras4–17 (Val12) contains overlapping epitopes that activate CTLs as well as T helper cells. Immunization with such peptides could result in a more comprehensive tumor-specific immune response. T cells specific for these epitopes demonstrated the capacity to cross-react against target cells pulsed with the peptide. They were also able to recognize tumor cells transfected with Kras containing the corresponding mutation.32,36 This is very

important, since it demonstrates that the epitope can be endogenously processed and presented by tumor cells. The immunogenicity of mutated H-ras was investigated by in vitro immunization of normal murine lymphocytes with a set of 36 overlapping peptides containing the Gly-Leu substitution at position 6137 and in vivo by immunization of mice with the mutated Ras protein33 or mutant Ras peptides.38 CTLs obtained after both immunizations were shown to specifically lyse target cells expressing the mutated H-Ras protein37,38 and 90% of peptide vaccinated mice remained tumor-free after challenge with Ras expressing tumor cells.38 These data indicate that in mouse models mutated Ras proteins can serve as tumor antigens, which both T helper cells as well as CTLs can recognize.

Ras-specific T cell responses in humans In humans the cellular immune response against the mutated Ras protein has also been examined (reviewed in Ref. 39). CD4+ T cell responses were determined in Ras-positive pancreatic and colon cancer patients. It was shown that 44% of the pancreatic cancer patients and 8% of the colon cancer patients were able to respond to the Ras peptide containing a mutation at position 12. In contrast, T helper cells obtained from healthy individuals did not respond to either the mutated Ras peptides or wild-type Ras protein.40 Several CD4+ T cell responses could be generated after in vitro immunization with mutant Ras peptides of normal human lymphocytes.41–43 These T helper cells were restricted by HLA-DR, -DP or -DQ, dependent on the peptide used for in vitro immunization. In one study cytokine profiles were determined to be characteristic of the T helper 1 phenotype, which plays a role in cellular immunity, and CD4+ T cells were found to have cytotoxic capacities.44 Thus, CD4+ T cells may also contribute to the elimination of Ras expressing tumor cells. In vitro immunization with mutant Ras peptides has also resulted in CD8+ T cell responses. Stimulation of normal human lymphocytes with N-ras peptides reflecting a codon 61 mutation induced CTLs restricted by HLA-A2.45 Another study showed Ras-specific CD8+ responses in a patient with a Ras expressing colon carcinoma. CTL clones obtained from this patient killed a colon carcinoma cell line expressing the same ras mutation as the patient’s tumor did. The cytotoxic activity was restricted by HLA-B12.46 These studies all show that Ras expressing tumors can elicit an immune response against mutated Ras, but these responses are not present in all patients bearing Ras+ tumors, nor are they successful at eradicating the tumor. So far, there is no proof that the immune system can play a role in the prevention of spontaneous tumors. However, Ras-mediated immune escape mechanisms or tumor growth enhancement may explain the development of progressive tumors in the presence of an immune response against mutated Ras. This could also be an explanation for the fact that in many tumors constitutive activation of Ras is correlated with poorer prognosis. These mechanisms can be divided into three categories: escape from immune-mediated destruction, modulation of tumor cell growth and modulation of the metastatic capacity. These mechanisms will be discussed in detail below (Table 2).

505

Review S Weijzen et al

506

Table 2

Ras-mediated changes in gene or protein expression

Increased expression or activity

Decreased expression or activity

Cathepsin L and B TGF-␤

MHC molecules immunodominant CTL epitope

TNF-␣

TNF-␣ receptor

IL-1 and IL-6 GM-CSF and G-CSF

Fas

osteopontin LFA-1

Immune modulation by activated Ras

MHC downregulation by Ras Low MHC expression levels on the cell surface will prevent effective recognition and tumor cell lysis by CTLs. As some oncogenes have been linked to MHC downregulation, such as c-myc in melanoma47 and N-myc in neuroblastoma,48 several studies have investigated the effect of Ras on MHC expression. Mouse fibroblasts were transfected with activated H-ras and increased expression of the Ras protein resulted in downregulation of all MHC loci.49 When a highly immunogenic murine leukemia cell line was transfected with activated K-ras and injected into mice, it was no longer able to induce a CTL response.50 These same results were found in 3T3 fibroblasts transfected with mutant H-ras: downregulation of the MHC resulted in decreased immunogenicity.51 Data from breast cancer patients support the correlation between Ras expression and MHC class I downregulation. In Ras-positive breast tumors, the level of Ras expression was inversely correlated with MHC class I expression levels. Furthermore, low HLA expression and high Ras expression was shown to correlate with increased levels of invasiveness.8 Experiments determining the reason for this MHC downregulation showed that Ras affected neither promotor activity nor transcriptional activity of the MHC genes. This suggests that post-translational mechanisms may be involved. One explanation might be a decrease in transporter associated with antigen processing (TAP) expression.49 TAP is responsible for transport of peptides into the endoplasmic reticulum (ER). In the ER peptides will form a complex with MHC class I molecules, which subsequently will be transported to the cell surface. Decreased TAP expression can result in a reduced availability of intracellular peptides in the ER and consequently a decreased number of MHC–peptide complexes on the cell surface. Other explanations for a decrease in MHC expression on the cell surface could be inhibition of proper MHC production or improper folding of MHC–peptide complexes in the ER.49 In addition to CTLs, a second set of immune cells, which are able to kill tumor cells under some circumstances, are natural killer (NK) cells. Since NK cells can recognize cells without MHC expression, Ras-induced MHC downregulation may increase susceptibility to NK-mediated killing. This hypothesis was not supported by two studies that determined the susceptibility of ras-transfected cells to lysis by NK cells. Both untransfected and ras-transfected cells were equally sensitive to lysis by NK cells.51,52 In summary, downregulation of MHC class I by Ras greatly impairs the efficiency of recog-

nition by CTLs and does not increase the sensitivity of tumor cells to NK activity. The effect of Ras on MHC expression may have important consequences for the effectiveness of immunotherapies and needs to be considered when designing immunotherapy for Ras expressing tumors.

The effect of Ras on antigen processing As mentioned above, CTLs can recognize tumor-derived peptides presented in complex with MHC class I. If the antigen processing pathways are altered by Ras, recognition and subsequent killing of the tumor cells by CTLs may be influenced. An effect of Ras on antigen processing has been observed in an adenovirus tumor model, containing two CTL epitopes. The immunodominant epitope was no longer recognized by a specific CTL clone when activated Ras was transfected into the cell. However, a CTL clone against the subdominant epitope, which exhibits a lower capacity to activate CTLs than the immunodominant peptide, was still able to recognize and kill the tumor cell in the presence of Ras. This interesting observation could not be explained by low expression levels of MHC class I or low expression of the protein which contains the immunodominant epitope.53 These data suggest that Ras selectively affects the antigen processing pathways: the cell surface expression of the immunodominant peptide is downregulated, resulting in the inability of CTLs to recognize and kill the tumor cells. The mechanisms of action in this model have not yet been addressed. If this Ras-mediated regulation of CTL epitope expression is a more general phenomenon, it could have implications for the development of CTLmediated immunotherapy of cancer. Most T cell-based immunotherapeutic strategies use tumor-derived CTL epitopes to induce a tumor-specific immune response. Therefore, downregulation of the expression of this particular CTL epitope could abolish the effectiveness of the immunotherapy.

Ras-mediated regulation of cytokines, growth factors and their receptors Several studies have investigated possible effects of Ras on immunoregulatory cytokines and/or growth factors and have uncovered immune downregulation through regulation of cytokine expression or growth factors. A growth factor involved in immune suppression mediated by Ras is transforming growth factor-␤ (TGF-␤). Ras was found to increase TGF-␤ production in tumor cells, which turned non-tumorigenic tumor cells into tumorigenic tumor cells.54 Since TGF-␤ is known to suppress T cell activation, Rasinduced TGF-␤ production and subsequent downregulation of CTL responses could explain the observed tumor outgrowth. A cytokine receptor, implicated in Ras-induced evasion of CTL-mediated killing, is the tumor necrosis factor-␣ (TNF) receptor. TNF-␣, which is produced by CTLs, can cause apoptosis of several tumors in vivo. Ras-mediated downregulation of TNF-␣ receptors results in resistance to killing by TNF-␣ and outgrowth of a tumor.55 Another mechanism by which CTLs attempt to kill tumor cells is by Fas–Fas ligand interaction. Fas is a transmembrane protein expressed by tumor cells. Fas ligand, expressed by CTLs, can bind to Fas, thereby inducing apoptosis of the tumor cells. Ras expressing tumor cells are shown to have a marked decrease in Fas cell surface expression, which renders tumor cells highly resistant to apoptosis.56

Review S Weijzen et al

Modulation of tumor growth To enhance tumor growth Ras was shown to increase production of several growth factors and growth promoting cytokines. Several studies have observed an increase in IL-1 and IL-6 mRNA expression and stability mediated by activated Ras.57,58 In some models, upregulation of granulocyte–macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) was also correlated with mutated Ras expression. In melanoma, it has been shown that enhanced production of IL-1 and IL-6 mediated by Ras results in a growth advantage for the tumor cells. The effect of enhanced GM-CSF production is paradoxical. On the one hand, injection of GM-CSF in humans with myeloid leukemia leads to a decrease in neutrophil recovery time after chemotherapy and increases sensitivity to some chemotherapeutic agents.59 Furthermore, administration of GM-CSF is now used as an adjuvant immunotherapy in order to enhance the presentation capacity of professional antigen presenting cells, resulting in more efficient tumor-specific immune responses.60–62 Thus, Ras-mediated increases in GM-CSF production might be detrimental to the tumor. On the other hand, activated Ras induces hypersensitivity to GM-CSF in JCML, when the negative regulator NFI is lost. This results in aberrant growth of hematopoietic cells. This Ras-mediated hypersensitivity to GM-CSF is thought to be a major step in the development of JCML.63,64 Another growth factor linked to activated Ras is vascular endothelial growth factor (VEGF). VEGF regulates the formation of blood vessels during embryonic development as well as endothelial cell permeability, motility and proliferation in vivo and is necessary for tumor growth beyond 2 mm3. Constitutive Ras activation has been directly correlated to upregulation of VEGF mRNA and consequently tumor progression.65,66 Enhancement of the metastatic potential Metastatic capacity of a tumor is dependent on a variety of factors. These factors are all involved in interaction between the tumor cells and their surrounding. Some of these factors are altered in the presence of activated Ras and are discussed below.

Ras-mediated regulation of adhesion molecules Adhesion molecules play an important role in cell–cell and cell–matrix interactions. Alteration in the expression of these molecules can enhance the metastatic potential of tumor cells. Expression of lymphocyte function-associated antigen-1 (LFA-1), an adhesion molecule which plays an important role in T cell–target interactions, was shown to be upregulated by activated Ras in EBV-infected human B lymphoblasts.67 As LFA-1 is the ligand for intercellular adhesion molecule-1 (ICAM-1), upregulation of LFA-1 has been reported to facilitate invasion into other tissues expressing ICAM-1 molecules. For certain cell types, LFA-1 expression has been correlated with their capacity to invade and metastasize into surrounding tissues.68 Therefore, increase of LFA-1 expression levels mediated by Ras may influence metastatic potential of tumor cells. In metastatic melanoma constitutive Ras activation was directly correlated with an increased production of TNF-␣.57

Increase in TNF-␣ production is associated with elevated expression of ICAM-1. Increased expression of ICAM-1 has been reported to result in increased binding between LFA-1 and ICAM-1 and consequently in enhanced interactions between T cells and cancer cells.69 However, in melanoma, elevated ICAM-1 expression is correlated with increased risk of metastasis.57 Thus, activated Ras may result in increased tumor metastatic potential through upregulation of TNF-␣ and ICAM-1. The increased binding to and enhanced production of two other adhesion molecules, osteopontin and laminin, have also been associated with the presence of activated Ras. When the overexpression of these adhesion molecules was blocked and binding was abrogated, tumorigenicity of metastatic Ras expressing tumor cells was reduced.70 This indicated that in this model Ras is responsible for the metastatic potential of the tumor cells through regulation of osteopontin and laminin.

Increase in proteolytic enzymes and decrease in inhibitors of these enzymes A very important factor that also contributes to a metastatic capacity of tumor cells is their ability to degrade and invade through basement membranes. Several degradative enzymes, such as cathepsin L and B, play an important role in invasion of tissue. Normally these enzymes are located in the lysosomes. Overexpression in malignant cells results in secretion of these degradative enzymes.71 Several in vitro and in vivo studies have demonstrated that the levels of cathepsin L and B were positively correlated with the level of Ras and the metastatic potential of the tumor cells.72–75 Moreover, in one study the activity of cathepsin inhibitors was shown to be decreased in the metastatic tumor cells.72 Increased production of these enzymes and a decrease of the activity of their inhibitors will result in an increased potential to invade surrounding tissues and blood vessels in order to metastasize to other organs. Activated Ras as a tumor antigen Ras expressing tumors can initiate many changes in cells, which aim at avoiding destruction by the immune system and enhance tumor growth. On the other hand, mutated Ras also provides the immune system with a unique protein which contains several immunogenic T helper and CTL epitopes. Therefore, immunotherapy using Ras peptide vaccination in humans has recently been explored. Mutant Ras peptides have been used to immunize patients harboring a ras mutation in their tumor cells. The first attempt to vaccinate patients with pancreatic cancer involved the vaccination with large amounts of autologous antigen presenting cells pulsed with K-Ras5–21. Five patients with pancreatic carcinoma and identified ras mutations participated in this study,76 none of whom showed any T cell responsiveness against any of the Ras mutant peptides before vaccination. Two out of five patients responded with a transient T cell proliferation after two and three vaccinations. These responses were restricted to the mutant Ras peptide in the first patient, while in the second patient both wild-type Ras and mutant Ras peptides were able to elicit T cell proliferation.76 The vaccination did not result in an increased survival in the responding patients. However, this might be explained by the large tumor burden and the deterioration of the immune sys-

507

Review S Weijzen et al

508

tem in the terminal stage of the disease. At this stage of disease, vaccination may come too late and this situation is therefore not ideal to test the efficacy of the vaccination. Since no adverse effects were observed in the five patients, this vaccination method is planned to be tested in patients with minimal residual disease.76 The same vaccination regimen was repeated recently in a patient with pancreatic carcinoma resulting in the generation of Ras-specific cytotoxic CD4+ and CD8+ T lymphocytes. These clones were able to efficiently kill a tumor cell line, derived from the patients’ ascites and the HLA restriction for CTL-mediated lysis appeared to be HLA-B35.77 In this protocol, a K-Ras peptide was used, which contained epitopes that could be recognized by CD8+ and CD4+ T cells. This might ensure a more adequate immune response, since it is known that T helper cells may be essential for activating and maintaining the CTL responses in vivo.78 In a phase I clinical trial eight patients were treated with a K-Ras peptide, in which the mutation matched the mutation detected in their primary carcinoma. The results showed that three patients developed a T cell-mediated immune response against the K-Ras peptide, consisting of a CD4+ T cell response, a CD8+ T cell response or both.79 The finding that the CD8+ T cell response in one of the responding patients was restricted by HLA-A2, led to the discovery of two HLAA2 epitopes in K-Ras, ie K-Ras5–14 (Asp12) and K-Ras4–12 (Val12).79,80 This patient also developed a T helper response after peptide vaccination, indicating the presence of a second, overlapping T helper epitope.79 Recently, five phase I and II clinical trials have been initiated treating carcinoma patients, whose tumors express a ras mutation at codon 12.39,81 These different clinical trials use mutant peptides with and without GM-CSF and IL-2, whole Ras proteins and transfer of Ras-specific T cells. Preliminary results of a phase II clinical trial, which has recruited 45 patients so far, using Ras peptide vaccination and GM-CSF as adjuvant, show that in approximately 50% of patients with operable pancreatic carcinoma a T cellmediated immune response was induced.81 The vaccination method used in these pilot studies might be improved by using pure dendritic cells (DCs) loaded with the peptide of interest. DCs are the most potent stimulators of native T cells and can very efficiently activate dormant T cells.82 In the mixture of cells used in the studies mentioned DCs are present, but most likely represent only a small proportion of the total number of cells. Furthermore, this approach will prevent the possible tolerizing effect of peptides. Vaccination with peptide in adjuvant can sometimes lead to enhancement of tumor growth instead of protection. This effect was shown to be mediated by induction of tolerance of T cells.83,84 Tolerance can be avoided when peptides are introduced in the host by a carrier, ie loaded on to DCs.85 The limited results obtained with peptide vaccination in pancreatic carcinoma patients are difficult to interpret, since the numbers of patients are small. As more clinical trials come to an end, results will show if immunotherapy as tested can result in a clinical benefit for the patients with Ras expressing tumors. As mentioned before, the success of these vaccination strategies is often dependent on the immune status of the patients. Furthermore, in most cases the tumor burden is too large for any immune response to result in tumor regression after peptide vaccination. Another factor influencing the success of immunotherapy is the MHC expression in the treated patients. Since Ras has been shown to downregulate MHC expression in some cases, it would be sensible to determine the levels of MHC molecules on the patients’ tumor. When

this level is very low, CTL epitopes will not be efficiently presented and T cell-mediated immunotherapy will therefore be ineffective. Treatment with interferon-␥ of the patients with low MHC expression might increase levels of MHC molecules on the tumor cell surface, which could increase the efficacy of immunotherapy. In leukemia, T cell-based immunotherapy remains a largely unexplored area of research. In practice only a certain subset of patients would be eligible for this kind of therapy. These would be patients who are not severely immune suppressed by previous treatments such as aggressive chemotherapy and bone marrow transplantation. Ras-specific immunotherapy might be effective in myeloid leukemia patients who have undergone induction chemotherapy. In these patients, the leukemia has been reduced to minimal residual disease. This might be the best opportunity to start immunotherapy successfully. Blocking Ras functions Treatment of advanced forms of cancer mostly consists of chemotherapy and/or radiotherapy. The success of chemotherapy is dependent on the sensitivity of the tumor tissue to the drugs. However, in most tumors this sensitivity is progressively lost and the effect of chemotherapy is reduced. One of the mechanisms behind resistance to chemotherapy may be Ras dependent. It has been reported that in melanoma the activated N-Ras can evoke chemoresistance both in vitro and in vivo by blocking apoptosis.86 It is therefore essential to develop new kinds of therapy, which target tumor cells expressing activated Ras. Mutated Ras was described above as a target for immunotherapy. Another approach to the treatment of these kinds of tumors has focussed on blocking Ras function.

Farnesyl transferase inhibitors One of the most successful drugs to block Ras activity is the farnesyl transferase inhibitor (FTI). Ras is produced as a precursor molecule that requires several post-translational modifications, which increase its hydrophobicity and result in association with the plasma membrane. Farnesylation is one of those modifications and is the only one that is necessary for Ras transforming activity.87 The enzyme involved in this process is farnesyl protein transferase (FPTase). It catalyzes the transfer of a farnesyl group to a cysteine residue located near the C-terminus of the protein. This farnesylation is dependent on the CAAX motive, which represents the cysteine (C) to be farnesylated, two nonpolar hydrophobic aliphatic amino acids (A) and preferably methionine, serine or phenylalanine at the last position (X). FPTase can be inhibited by a series of drugs called farnesyl transferase inhibitors. These inhibitors specifically block the farnesylation of Ras proteins, antagonize its cell transforming activities88 and therefore can be used as antitumor agents. Several groups (ie natural product inhibitors, bisubstrate derivatives and peptidomimetics) of FTIs have been described and some of them seem to be very effective and well tolerated.87,89 Peptide analogues were designed to inhibit FPTase activity on the basis of the last four amino acids of the Ras protein. These inhibitors have been reported to be very specific, since the minimal concentration at which they are effective is fairly low. The most promising FTI, L-744,832 was tested in mouse

Review S Weijzen et al

models. L744,832 was shown to inhibit growth in vitro in 70% of all tumor cell lines at low concentrations. When Hras transgenic mice with substantial spontaneous mammary and salivary tumors were treated, tumors regressed completely within 2 weeks in 100% of the mice. The effectiveness of this new anti-tumor drug was compared to that of doxorubicin, a traditional chemotherapeutic agent. At its maximal tolerated dose, treatment with doxorubicin of these tumors only slowed down tumor growth and provoked severe systemic toxicity. When L-744,832 treatment was abrogated, tumor growth restarted in all mice, but a second cycle of L-744,832 treatment of the grown tumors again resulted in further regression after 2–6 weeks in two out of three mice. No systemic toxicity was observed.90 The previously mentioned studies have all been performed in mice or cells expressing mutated H-ras. However, when comparable studies were performed in cells transformed by K-ras or N-ras, FTI appeared to be less potent at inhibiting Ras farnesylation.91,92 FTIs cannot only be used as treatment for tumors expressing mutated Ras, but also for tumors in which Ras is deregulated through other mechanisms. A report on the use of FTI in tumor cell lines obtained from neurofibromatosis patients, which lack NF1 and therefore have constitutive Ras activation, showed that the malignant growth properties could be blocked.93 These results are very encouraging, but there are two concerns which need to be addressed. The first is whether these compounds will have an effect on cells expressing wild-type Ras. The second concern is the effect FTI might have on T cell signaling, since Ras is known to play an important role in signal transduction via the TCR. Before these drugs can be applied in a clinical setting, these issues must be addressed with great care. A more general remark is that these kind of drugs will probably act as cytostatic agents, since most results show they are effective at accomplishing tumor regression, but unable to completely eradicate tumors. However, treatment with FTI is likely to result in a dramatic decrease in tumor burden, which is the perfect situation for the start of any immunotherapy. Theoretically, treatment with FTI might increase the chance of success of immunotherapy not only by reducing tumor size, but also by suppressing the effects of Ras on the immune recognition of tumor cells. Combinations of immunotherapy or conventional chemotherapy with FPTase inhibitor treatment could result in a very efficient therapy for tumors expressing activated Ras, if effects of FTI on T cell activation can be avoided. However, the possibility of tumor cells developing resistance to FTI as an attempt to escape growth inhibition needs to be considered. A cell line resistant to FPTase inhibitors has already been described: the transformed phenotype and growth potential of a ras-transformed cell line were not affected by treatment with a concentration of an FTI 30-fold higher than that sufficient to inhibit farnesylation in other ras-transformed cells.94 The resistance was caused by a reduced ability of the compound to inhibit farnesylation of Ras. This was not due to increased multiple drug resistance gene activity, decreased FTI accumulation or mutations in the FPTase. Thus, the mechanism of action behind this phenomenon remains unknown.94

Antisense ras A second agent used to block mutated Ras activity is antisense Ras RNA. This technique has been used in only a very limited

number of tumors. In human lung cancer cells and pancreatic cancer cells, introduction of antisense K-Ras RNA was shown to block K-Ras specifically, whereas expression of H-Ras and N-Ras and the wild-type K-Ras remained unchanged. The reduced expression of K-Ras resulted in a decrease in in vitro colony formation and tumorigenicity of the human tumor cells in nude mice. However, cell viability was not completely abrogated and although the rate of tumor growth was diminished, tumors continued to grow in vivo.95–98

Neutralizing anti-Ras antibody Recently a neutralizing, intracellular anti-Ras antibody was engineered and tested in human cancer cells. In vitro intracellular introduction of this antibody resulted in a 70% decrease in H-Ras expression and consequent induction of apoptosis in ras-transformed cells, but not in untransformed cells. Furthermore, intratumoral injection of the antibody in a nude mouse promoted sustained, but not permanent, regression of a tumor expressing mutated K-ras.99

Reovirus therapy A recent development in the search for Ras expressing tumor therapy is the application of reoviruses. Human reovirus requires an activated Ras signaling pathway to infect a cell. In vitro experiments showed that within 24 h after infection with the reovirus, the host protein synthesis is shut down and eventually the virus will lyse the host cell. Therefore, reovirus was used to infect immunocompetent mice with an established Ras expressing tumor to determine the effect of infection in vivo. After multiple injections of the reovirus, complete regression of the Ras expressing tumor was observed in 66% of the mice. The presence of neutralizing antibodies against the reovirus did not eliminate the therapeutic effect and the reovirus remained confined to the tumor mass.100 The advantage of this therapeutic approach is that this virus can infect all cells with an activated ras pathway, regardless of the reason for Ras activation: it is not specific for a certain mutation or a certain isoform of Ras. Since reovirus infection in humans is believed to be asymptomatic, this approach might prove to be a very potent anticancer therapeutic.

General conclusions The observation that in some types of cancer, activated Ras expression results in a poorer prognosis and a decreased 5year survival can be linked to several mechanisms. These mechanisms can lead to increased growth and metastatic potential and evasion of recognition and destruction by the immune system. The expression of MHC molecules, CTL epitopes, cytokines, growth factors and their receptors and adhesion molecules are involved in modulation of the cellular interactions between tumor cells themselves, between tumor cells and the surrounding tissue and between tumor cells and cells of the immune system. In leukemia, most studies so far have focussed on determining the mutation rate of the ras genes in the different forms of leukemia. Several reports have only recently shown a possible correlation between leukemia prognosis and the presence of ras mutations. The mechanisms described above may be responsible for these differences in

509

Review S Weijzen et al

510

prognosis. The relative contribution of each mechanism in the specific forms of leukemia remains to be investigated. Some critical remarks should be made regarding the studies reviewed here. It is important to realize that, even though there are strong indications for the greater part of these Rasrelated mechanisms, most of these studies only tested their hypothesis on a limited number of cell lines or malignancies. It would be interesting to determine which of these mechanisms can be generalized. Furthermore, the majority of these studies were performed in vitro. Ras-transfected cell lines were used to determine which effects Ras might have in an in vivo situation. In vivo, however, malignancies most likely develop as a consequence of multiple mutational events. The exact role of activated Ras in the development and maintenance of tumors needs to be further elucidated. The few in vivo observations which have been related to activated Ras may well be the result of complex interactions between different intercellular proteins deregulated by several tumorigenic mutations. Several mechanisms have been proposed to explain the Ras-mediated effects. In some studies Ras activation influenced mRNA expression levels and stability of certain proteins, in other studies post-translational interference of Ras is suggested to be involved. This area of research could provide us with more insight into Ras pathways, which may lead to the discovery of new possibilities to block specific Rasinduced effects. Immunotherapeutic strategies with Ras epitopes as a target for CTL-mediated destruction are currently being explored. The final results of clinical trials using antigen processing cells loaded with Ras peptides are not available as yet. However, preliminary results are encouraging, since peptide vaccination can induce a Ras-specific immune response in 50% of Raspositive cancer patients. Drug therapy, using farnesyltransferase inhibitors is very efficient at achieving tumor regression of Ras expressing tumors. When the possible side-effects of FTI on untransformed cells and T cells have been elucidated, FTI treatment might prove to be a good treatment option in patients with Ras-positive tumors. Miscellaneous Ras-directed therapies, such as use of the reovirus and intracellular ras-specific antibody are very elegant ways to block activated Ras. Clinical trials should supply information about toxicity and safety of these kinds of therapies. Furthermore, the effectivity in humans should be determined in a clinical trial. It would be interesting to determine whether farnesyl transferase inhibitors or other Ras antagonists are able to revert the mechanisms that result in increased metastatic ability and evasion of immune responses. In that case, Ras expressing tumor cells, treated with Ras antagonists, might become more sensitive to the effects of chemo- and immunotherapy, resulting in better prognosis and higher survival rates in patients with Ras expressing tumors, including, but not limited to, several forms of leukemia.

Acknowledgements We would like to thank Dr Giuseppina Nucifora for her support and Dr Lucio Miele for critically reading the manuscript. This review was partly supported by NIH grant R01 CA/AI 78399–01 (WMK), a Yamanouchi Europe Fellowship (SW) and a CRI fellowship (MPV).

References 1 Bos JL. Ras oncogenes in human cancer: a review. Cancer Res 1989; 49: 4682–4689. 2 Padua RA, Guinn BA, Al-Sabah AI, Smith M, Taylor C, Pettersson T, Ridge S, Carter G, White D, Oscier D, Chevert S, West R. RAS, FMS and p53 mutations and poor clinical outcome in myelodysplasias: a 10-year follow-up. Leukemia 1998; 12: 887–892. 3 De Souza Fernandez T, Menezes de Souza J, Macedo Silva ML, Tabak D, Abdelhay E. Correlation of N-ras point mutations with specific chromosomal abnormalities in primary myelodysplastic syndrome. Leukemia Res 1998; 22: 125–134. 4 De Melo MB, Lorand-Metze I, Lima CS, Saad ST, Costa FF. N-ras gene point mutations in Brazilian acute myelogenous leukemia patients correlate with a poor prognosis. Leuk Lymphoma 1997; 24: 309–317. 5 Wang JC, Chen C. N-Ras oncogene mutations in patients with agnogenic myeloid metaplasia in leukemic transformation. Leukemia Res 1998; 22: 639–643. 6 Sheng XM, Kawamura M, Ohnishi H, Ida K, Hanada R, Kojima S, Kobayashi M, Bessho F, Yanagisawa M, Hayashi Y. Mutations of the RAS genes in childhood acute myeloid leukemia, myelodysplastic syndrome and juvenile chronic myelocytic leukemia. Leukemia Res 1997; 21: 697–701. 7 Misawa S, Horiike S, Kaneko H, Kashima K. Genetic aberrations in the development and subsequent progression of myelodysplastic syndrome. Leukemia 1997; 11 (Suppl. 3): 533–535. 8 Solana R, Romero J, Alonso C, Pena J. MHC class I antigen expression is inversely related with tumor malignancy and ras oncogene product (p21ras) levels in human breast tumors. Inv Met 1992; 12: 210–217. 9 Benhatter J, Losi L, Chaubert P, Givel JC, Costa J. Prognostic significance of K-ras mutations in colorectal cancer. Gastroent 1993; 104: 1044–1048. 10 Slebos RJ, Kibbelaar RE, Dalesio O, Kooistra A, Stam J, Meijer CJ, Wagenaar SS, Vanderschueren RG, Van Zandwijk N, Mooi WJ, Bos JL, Rodenhuis S. K-ras oncogene activation as a prognostic marker in adenocarcinoma of the lung. New Engl J Med 1990; 323: 561–565. 11 Stahel RA. Antigens, receptors and dominant oncogenes and the prognosis of non-small cell lung cancer. Lung Cancer 1994; 11 (Suppl. 3): S31–S38. 12 Wang Y, Lee H, Chen S, Yang S, Chen C. Analysis of K-ras gene mutations in lung carcinomas: correlation with gender, histological subtypes, and clinical outcome. J Cancer Res Clin Oncol 1998; 124: 517–522. 13 Bartsch D, Bastian D, Barth P, Schudy A, Nies C, Kisker O, Wagner H, Rothmund M. K-ras oncogene mutations indicate malignancy in cystic tumors of the pancreas. Ann Surg 1998; 228: 79–86. 14 Friess H, Berberat P, Schilling M, Kunz J, Korc M, Buchler MW. Pancreatic cancer: the potential clinical relevance of alterations in growth factors and their receptors. J Mol Med 1996; 74: 35–42. 15 Iniesta P, de Juan C, Caldes T, Vega F, Lopez J, Frenandez C, Sanchez A, Torres A, Balibrea J, Benito M. Genetic abnormalities and microsatellite instability in colorectal cancer. Cancer Det Prev 1998; 22: 383–395. 16 Kressner U, Bjorheim J, Westring S, Wahlberg SS, Pahlman L, Glimelius B, Lindmark G, Lindblom A, Borresen-Dale A. Ki-ras mutations and prognosis in colorectal cancer. Eur J Cancer 1998; 34: 518–521. 17 Leon J, Guerrero I, Pellicer A. Differential expression of the ras gene family in mice. Mol Cell Biol 1987; 7: 1535–1540. 18 Lowy DR, Willumsen BM. Function and regulation of ras. Ann Rev Biochem 1993; 62: 851–891. 19 Wittinghofer A, Scheffzek K, Ahmadian MR. The interaction of Ras with GTPase-activating proteins. FEBS Lett 1997; 410: 63– 67. 20 Prendergast GC, Gibbs JB. Ras regulatory interactions: novel targets for anti-cancer intervention? Bioessays 1994; 16: 187–191. 21 Gomez J, Martinez C, Fernandez B, Garcia A, Rebollo A. Ras activation leads to cell proliferation or apoptotic cell death upon interleukin-2 stimulation or lymphokine deprivation, respectively. Eur J Immunol 1997; 27: 1610–1618. 22 Clark GJ, Der CJ. Aberrant function of the Ras signal transduction

Review S Weijzen et al

23 24 25 26

27

28 29

30 31 32

33 34 35 36

37

38

39 40

41 42

43

44

pathway in human breast cancer. Br Cancer Res Trt 1995; 35: 133–144. Rommel C, Hafen E. Ras-a versatile cellular switch. Curr Opin Genet Dev 1998; 8: 412–418. Hall A. A biochemical function for ras – at last. Science 1994; 264: 1413–1414. Chen CY, Liou J, Forman LW, Faller DV. Differential regulation of discrete apoptotic pathways by Ras. J Biol Chem 1998; 273: 16700–16709. Toi M, Hamada Y, Nakamura T, Mukaida H, Suehiro S, Wada T, Toge T, Niimoto M, Hattori T. Immunocytochemical and biochemical analysis of epidermal growth factor receptor expression in human breast cancer tissues: relationship to estrogen receptor and lymphatic invasion. Int J Cancer 1989; 43: 220–225. Basu TN, Gutmann DH, Fletcher JA, Glover TW, Collins FS, Downward J. Aberrant regulation of ras proteins in malignant tumour cells from type 1 neurofibromatosis patients. Nature 1992; 356: 713–715. Stiller CA, Chessells JM, Fitchett M. Neurofibromatosis and childhood leukemia/lymphoma: a population-based UKCCSG study. Br J Cancer 1994; 70: 969–972. Satoh T, Fantl WJ, Escobedo JA, Williams LT, Kaziro Y. Plateletderived growth factor receptor mediates activation of ras through different signaling pathways in different cell types. Mol Cell Biol 1993; 13: 3706–3713. Sanchez-Garcia I, Martin-Zanca D. Regulation of Bcl-2 gene expression by BCR-ABL is mediated by Ras. J Mol Biol 1997; 267: 225–228. Neefjes JJ, Momburg F. Cell biology of antigen presentation. Curr Opin Immunol 1993; 5: 27–34. Abrams SI, Dobrzanski MJ, Wells DT, Stanziale SF, Zaremba S, Masuelli, L, Kantor JA, Schlom J, Masuelle L, Kantor JL, Schlom J. Peptide-specific activation of cytolytic CD4+ T lymphocytes against tumor cells bearing mutated epitopes of K-ras p21. Eur J Immunol 1995; 25: 2588–2597. Cheever MA, Chen W, Disis ML, Takahashi M, Peace DJ. T-cell immunity to oncogenic proteins including mutated ras and chimeric bcr-abl. Ann NY Acad Sci 1993; 690: 101–112. Peace DJ, Chen W, Nelson H, Cheever MA. T cell recognition of transforming proteins encoded by mutated ras proto-oncogenes. J Immunol 1991; 146: 2059–2065. Skipper J, Stauss HJ. Identification of two cytotoxic T lymphocyterecognized epitopes in the Ras protein. J Exp Med 1993; 177: 1493–1498. Abrams SI, Stanziale SF, Lunin SD, Zaremba S, Schlom J. Identification of overlapping epitopes in mutant ras oncogene peptides that activate CD4+ and CD8+ T cell responses. Eur J Immunol 1996; 26: 435–443. Peace DJ, Smith JW, Chen W, You SG, Cosand WL, Blake J, Cheever MA. Lysis of ras oncogene-transformed cells by specific cytotoxic T lymphocytes elicited by primary in vitro immunization with mutated ras peptide. J Exp Med 1994; 179: 473–479. Fenton RG, Taub DD, Kwak LW, Smith MR, Longo DL. Cytotoxic T-cell response and in vivo protection against tumor cells harboring activated ras proto-oncogenes. J Natl Cancer Inst 1993; 85: 1294–1302. Abrams SI, Hand PH, Tsang KY, Schlom J. Mutant ras epitopes as targets for cancer vaccines. Semin Oncol 1996; 23: 118–134. Qin H, Chen W, Takahashi M, Disis ML, Byrd DR, McCahill L, Bertram KA, Fenton RG, Peace DJ, Cheever MA. CD4+ T-cell immunity to mutated ras protein in pancreatic and colon cancer patients. Cancer Res 1995; 55: 2984–2987. Jung S, Schluesener HJ. Human T lymphocytes recognize a peptide of single point-mutated oncogenic ras proteins. J Exp Med 1991; 173: 273–276. Gedde-Dahl T 3rd, Eriksen JA, Thorsby E, Gaudernack G. T-cell responses against products of oncogenes: generation and characterization of human T cell clones specific for p21 ras-derived synthetic peptides. Hum Immunol 1992; 33: 266–274. Fossum B, Gedde-Dahl T 3rd, Hansen T, Eriksen JA, Thorsby E, Gaudernack G. Overlapping epitopes encompassing a point mutation (12 Gly → Arg) in p21 ras can be recognized by HLADR, -DP and -DQ restricted T cells. Eur J Immunol 1993; 23: 2687–2691. Tsang KY, Nieroda CA, DeFilippi R, Chung YK, Yamaue H,

45

46

47 48 49 50

51

52

53

54

55

56 57

58

59

60

61

Greiner JW, Schlom J. Induction of human cytotoxic T cell lines directed against point-mutated p21 ras-derived synthetic peptides. Vaccine Res 1994; 3: 183–193. Van Elsas A, Nijman HW, Van der Minne CE, Mourer JS, Kast WM, Melief CJM, Schrier PI. Induction and characterization of cytotoxic T-lymphocytes recognizing a mutated p21 ras peptide presented by HLA-A*0201. Int J Cancer 1995; 61: 389–396. Fossum B, Olsen AC, Thorsby E, Gaudernack G. CD8+ T cells from a patient with colon carcinoma, specific for a mutant p21Ras-derived peptide (Gly13 → Asp), are cytotoxic towards a carcinoma cell line harbouring the same mutation. Cancer Immunol Immunother 1995; 40: 165–172. Versteeg R, Noordermeer IA, Kruse-Wolters M, Ruiter DJ, Schrier PI. c-myc down-regulates class I HLA expression in human melanomas. EMBO J 1988; 7: 1023–1029. Bernards R, Dessain SK, Weinberg RA. N-myc amplification causes down-modulation of MHC class I antigen expression in neuroblastoma. Cell 1986; 47: 667–674. Lohmann S, Wollscheid U, Huber C, Seliger B. Multiple levels of MHC class I down-regulation by ras oncogenes. Scand J Immunol 1996; 43: 537–544. Testorelli C, Bussini S, De Filippi R, Marelli O, Orlando L, Greiner JW, Grohmann U, Tentori L, Giuliani A, Bonmassar E, Graziani G. Dacarbazine-induced immunogenicity of a murine leukemia is attenuated in cells transfected with mutated K-ras gene. J Exp Clin Cancer Res 1997; 16: 15–22. Ehrlich T, Wishniak O, Isakov N, Cohen O, Segal S, Rager-Zisman B, Gopas J. The effect of H-ras expression on tumorigenicity and immunogenicity of Balb/c 3T3 fibroblasts. Immunol Lett 1993; 39: 3–8. Van Elsas A, Van Deursen E, Wielders R, Van den Berg-Bakker CA, Schrier PI. ras oncogene activation does not induce sensitivity to natural killer cell-mediated lysis in human melanoma. J Invest Dermatol 1994; 103 (Suppl.): 117S–121S. Toes REM, Offringa R, Blom RJJ, Brandt RMP, Van der Eb AJ, Melief CJM, Kast WM. An adenovirus type 5 early region 1Bencoded CTL epitope-mediating tumor eradication by CTL clones is down-modulated by an activated ras oncogene. J Immunol 1995; 154: 3396–3405. Offringa R, Feltkamp MCW, Voordouw AC, Lowik C, Van der Eb AJ, Melief CJM, Kast WM. Cells transformed by adenovirus type 5 E1A and activated ras evade destruction by T-cell immunity. Thesis, Leiden University, The Netherlands, 1991; 131–147. Fernandez A, Chen PW, Aggarwal BB, Ananthaswamy HN. Resistance of Ha-ras oncogene-induced progressor tumor variants to tumor necrosis factor and interferon-␥. Lymph Cyt Res 1992; 11: 79–85. Fenton RG, Hixon JA, Wright PW, Brooks AD, Sayers TJ. Inhibition of Fas(CD95) expression and Fas-mediated apoptosis by oncogenic ras. Cancer Res 1998; 58: 3391–3400. Castelli C, Sensi M, Lupetti R, Mortarini R, Panceri P, Anichini A, Parmiani G. Expression of interleukin 1␣, interleukin 6, and tumor necrosis factor ␣ genes in human melanoma clones is associated with that of mutated N-RAS oncogene. Cancer Res 1994; 54: 4785–4790. Demetri GD, Ernst TJ, Pratt ES, Zenzie BW, Rheinwald JG, Griffin JD. Expression of ras oncogenes in cultured human cells alters the transcriptional and posttranscriptional regulation of cytokine genes. J Clin Invest 1990; 86: 1261–1269. Hoelzer D, Seipelt G. Granulocyte colony-stimulating factor and granulocyte–macrophage colony-stimulating factor in the treatment of myeloid leukemia. Curr Opin Hematol 1995; 2: 196– 203. Samanci A, Qing Y, Fagerberg J, Strigard K, Smith G, Ruden U, Wahren B, Mellstedt H. Pharmacological administration of granulocyte/macrophage colony-stimulating factor is of significant importance for the induction of a strong humoral and cellular response in patients immunized with recombinant carcinoembryonic antigen. Cancer Immunol Immunother 1998; 47: 131– 142. Nagai E, Ogawa T, Kielian T, Ikubo A, Suzuki T. Irradiated tumor cells adenovirally engineered to secrete granulocyte/macrophage colony-stimulating factor establish antitumor immunity and eliminate pre-existing tumors in syngeneic mice. Cancer Immunol Immunother 1998; 47: 72–80.

511

Review S Weijzen et al

512

62 Jager E, Ringhoffer M, Dienes H-P, Arand M, Karbach J, Jager D, Ilsemann C, Hagedorn M, Oesch F, Knuth A. Granulocyte– macrophage colony-stimulating factor enhances immune responses to melanoma-associated peptides in vivo. Int J Cancer 1996; 67: 54–62. 63 Bollag G, Clapp DW, Shih S, Adler F, Zhang YY, Thompson P, Lange BJ, Freedman MH, McCormick F, Jacks T, Shannon K. Loss of NF1 results in activation of the Ras signaling pathway and leads to aberrant growth in haematopoietic cells. Nature Gen 1996; 12: 144–148. 64 Largaespada DA, Brannan CI, Jenkins NA, Copeland NG. NF1 deficiency causes Ras-mediated granulocyte/macrophage colony stimulating factor hypersensitivity and chronic myeloid leukaemia. Nature Gen 1996; 12: 137–143. 65 White FC, Benehacene A, Scheele JS, Kamps M. VEGF mRNA is stabilized by ras and tyrosine kinase oncogenes, as well as by UV radiation – evidence for divergent stabilization pathways. Growth Factors 1997; 14: 199–212. 66 Larcher F, Robles AI, Duran H, Murillas R, Quintanilla M, Cano A, Conti CJ, Jorcano JL. Up-regulation of vascular endothelial growth factor/vascular permeability factor in mouse skin carcinogenesis correlates with malignant progression state and activated H-ras expression levels. Cancer Res 1996; 56: 5391–5396. 67 Endo K, Borer CH, Tsujimoto Y. Modulation of LFA-1 surface antigen expression by activated H-ras oncogene in EBV-infected human B-lymphoblast cells. Oncogene 1991; 6: 1391–1396. 68 Roosien FF, De Kuiper PE, De Rijk D, Roos E. Invasive and metastatic capacity of revertants of LFA-1-deficient mutant T-cell hybridomas. Cancer Res 1990; 50: 3509–3513. 69 Springer TA. Adhesion receptors of the immune system. Nature 1990; 346: 425–434. 70 Behrend EI, Craig AM, Wilson SM, Denhardt DT, Chambers AF. Reduced malignancy of ras-transformed NIH 3T3 cells expressing antisense osteopontin RNA. Cancer Res 1994; 54: 832–837. 71 Achkar C, Gong QM, Frankfater A, Bajkowski AS. Differences in targeting and secretion of cathepsins B and L by BALB/3T3 fibroblasts and Moloney murine sarcoma virus-transformed BALB/3T3 fibroblasts. J Biol Chem 1990; 265: 13650–13654. 72 Chambers AF, Colella R, Denhardt DT, Wilson SM. Increased expression of cathepsins L and B and decreased activity of their inhibitors in metastatic, ras-transformed NIH 3T3 cells. Mol Carc 1992; 5: 238–245. 73 Yan Z, Deng X, Chen M, Xu Y, Ahram M, Sloane BF, Friedman E. Oncogenic c-Ki-ras but not oncogenic c-Ha-ras up-regulates CEA expression and disrupts basolateral polarity in colon epithelial cells. J Biol Chem 1997; 272: 27902–27907. 74 Donatien PD, Diment SL, Boissy RE, Orlow SJ. Melanosomal and lysosomal alterations in murine melanocytes following transfection with the v-rasHa oncogene. Int J Cancer 1996; 66: 557–563. 75 Kim K, Cai J, Shuja S, Kuo T, Murnane MJ. Presence of activated ras correlates with increased cysteine proteinase activities in human colorectal carcinomas. Int J Cancer 1998; 79: 324–333. 76 Gjertsen MK, Bakka A, Breivik J, Saeterdal I, Solheim BG, Soreide O, Thorsby E, Gaudernack G. Vaccination with mutant ras peptides and induction of T-cell responsiveness in pancreatic carcinoma patients carrying the corresponding RAS mutation. Lancet 1995; 346: 1399–1400. 77 Gjertsen MK, Bjorheim J, Saeterdal I, Myklebust AT, Gaudernack G. Cytotoxic CD4+ and CD8+ T lymphocytes, generated by mutant p21-ras (12val) peptide vaccination of a patient recognize 12val-dependent nested epitopes present within the vaccine peptide and kill autologous tumour cells carrying this mutation. Int J Cancer 1997; 72: 784–790. 78 Walter EA, Greenberg PD, Gilbert MJ, Finch RJ, Watanabe KS, Thomas ED, Riddell SR. Reconstruction of cellular immunity against CMV in recipients of allogeneic bone marrow by adoptive transfer of T cell clones from the donor. New Engl J Med 1995; 333: 1038–1044. 79 Abrams SI, Khleif SN, Bergmannleitner ES, Kantor JA, Chung Y, Hamilton JM, Schlom J. Generation of stable CD4(+) and CD8(+) T cell lines from patients immunized with ras oncogene-derived peptides reflecting codon 12. Cell Immunol 1997; 182: 137–151. 80 Bergmann-Leitner ES, Kantor JA, Shupert WL, Schlom J, Abrams SI. Identification of a human CD8+ T lymphocyte neo-epitope

81 82 83

84

85

86

87 88 89 90

91 92

93 94

95 96

97

98

99

100

created by a ras codon 12 mutation which is restricted by the HLA-A2 allele. Cell Immunol 1998; 187: 103–116. Gjertsen MK, Gaudernack G. Mutated ras peptides as vaccines in immunotherapy of cancer. Vox Sang 1998; 74 (Suppl. 2): 489–495. Cella M, Sallusto F, Lanzavecchia A. Origin, maturation and antigen presenting function of dendritic cells. Curr Opin Immunol 1997; 9: 10–16. Toes REM, Blom RJJ, Offringa R, Kast WM, Melief CJM. Enhanced tumor outgrowth after peptide vaccination. Functional deletion of tumor-specific CTL induced by peptide vaccination can lead to the inability to reject tumors. J Immunol 1996; 156: 3911–3918. Toes REM, Offringa R, Blom RJJ, Melief CJM, Kast WM. Peptide vaccination can lead to enhance tumor growth through specific T-cell tolerance induction. Proc Natl Acad Sci USA 1996; 93: 7855–7860. Toes REM, Van der Voort EIH, Schoenberger SP, Drijfhout JW, Van Bloois L, Storm G, Kast WM, Offringa R, Melief CJM. Enhancement of tumor outgrowth through CTL tolerization after peptide vaccination is avoided by peptide presentation on dendritic cells. J Immunol 1998; 160: 4449–4456. Jansen B, Schlagbauer-Wadl H, Eichler HG, Wolff K, Van Elsas A, Schrier Pl, Pehamberger H. Activated N-ras contributes to the chemoresistance of human melanoma in severe combined immunodeficiency (SCID) mice by blocking apoptosis. Cancer Res 1997; 57: 362–365. Perrin D, Halazy S, Hill BT. Inhibitors of the Ras signal transduction pathway as potential antitumour agents. J Enzyme Inhibition 1996; 11: 77–95. Gibbs JB, Oliff A, Kohl NE. Farnesyltransferase inhibitors: Ras research yields a potential cancer therapeutic. Cell 1994; 77: 175–178. Qian Y, Sebti SM, Hamilton AD. Farnesyltransferase as a target for anticancer drug design. Biopolymers 1997; 43: 25–41. Kohl NE, Omer CA, Conner MW, Anthony NJ, Davide JP, deSolms SJ, Giuliani EA, Gomez RP, Graham SL, Hamilton K, Gibbs J. Inhibition of farnesyltransferase induces regression of mammary and salivary carcinomas in ras transgenic mice. Nature Med 1995; 1: 792–797. Cox AD, Der CJ. Farnesyltransferase inhibitors and cancer treatment: targeting simply Ras? Biochim Biophys Acta 1997; 1333: F51–71. Mangues R, Corral T, Kohl NE, Symmans WF, Lu S, Malumbres M, Gibbs JB, Oliff A, Pellicer A. Antitumor effect of a farnesyl protein transferase inhibitor in mammary and lymphoid tumors overexpressing N-ras in transgenic mice. Cancer Res 1998; 58: 1253–1259. Yan N, Ricca C, Fletcher J, Glover T, Seizinger BR, Manne V. Farnesyltransferase inhibitors block the neurofibromatosis type I (NF1) malignant phenotype. Cancer Res 1995; 55: 3569–3575. Prendergast GC, Davide JP, Lebowitz PF, Wechsler-Reya R, Kohl NE. Resistance of a variant ras-transformed cell line to phenotypic reversion by farnesyl transferase inhibitors. Cancer Res 1996; 56: 2626–2632. Mukhopadhyay T, Tainsky M, Cavender AC, Roth JA. Specific inhibition of K-ras expression and tumorigenicity of lung cancer cells by antisense RNA. Cancer Res 1991; 51: 1744–1748. Zhang Y, Mukhopadhyay T, Donehower LA, Georges RN, Roth JA. Retroviral vector-mediated transduction of K-ras antisense RNA into human lung cancer cells inhibits expression of the malignant phenotype. Hum Gene Ther 1993; 4: 451–460. Alemany R, Ruan S, Kataoka M, Koch PE, Mukhopadhyay T, Cristiano RJ, Roth JA, Zhang WW. Growth inhibitory effect of antiK-ras adenovirus on lung cancer cells. Cancer Gene Ther 1996; 3: 296–301. Aoki K, Yoshida T, Sugimura T, Terada M. Liposome-mediated in vivo gene transfer of antisense K-ras construct inhibits pancreatic tumor dissemination in the murine peritoneal cavity. Cancer Res 1995; 55: 3810–3816. Cochet O, Kenigsberg M, Delumeau I, Virone-Oddos A, Multon M, Fridman WH, Schweighoffer F, Teillaud J, Tocque B. Intracellular expression of an antibody fragment-neutralizing p21 ras promotes tumor regression. Cancer Res 1998; 58: 1170–1176. Coffey MS, Strong JE, Forsyth PA, Lee PWK. Reovirus therapy for

Review S Weijzen et al

101

102

103

104

tumors with activated Ras pathway. Science 1998; 282: 1332– 1334. Paquette RL, Landaw EM, Pierre RV, Kahan J, Lubbert M, Lazcano O, Isaac G, McCormick F, Koeffler HP. N-ras mutations are associated with poor prognosis and increased risk of leukemia in myelodysplastic syndrome. Blood 1993; 82: 590–599. Constantinidou M, Chalevelakis G, Economopoulos T, Koffa M, Liloglou T, Anastassiou C, Yalouris A, Spandidos DA, Raptis S. Codon 12 ras mutations in patients with myelodysplastic syndrome: incidence and prognostic value. Ann Hematol 1997; 74: 11–14. Fengru L, Suyun L, Jinhai R, Junping W, Shirong X, Runsheng L, Ergu Y. Correlated flow cytometric analysis of H-ras p21 and DNA ploidy in acute myelogenous leukemia. J Tongji Med University 1996; 16: 75–77. Radich JP, Kopecky KJ, Willman CL, Weick J, Head D, Appelbaum F, Collins SJ. N-ras mutations in adult de novo acute

myelogenous leukemia: prevalence and clinical significance. Blood 1990; 76: 801–807. 105 Al-Mulla F, Going J, Sowden ETHH, Winter A, Pickford IR, Birnie GD. Heterogeneity of mutant versus wild-type Ki-ras in primary and metastatic colorectal carcinomas, and association of codon12 valine with early mortality. J Path 1998; 185: 130–138. 106 Ahnen DJ, Feigl P, Quan G, Fenoglio-Preiser C, Lovato LC, Bunn PA Jr, Stemmerman G, Wells JD, MacDonald JS, Meyskens FL, Jr. Ki-ras mutation and p53 overexpression predict the clinical behavior of colorectal cancer: a Southwest Oncology Group study. Cancer Res 1998; 58: 1149–1158. 107 Mitsudomi T, Steinberg SM, Oie HK, Mulshine JL, Phelps R, Viallet J, Pass HI, Minna JD, Gazdar AF. ras gene mutations in nonsmall cell lung cancers are associated with shortened survival irrespective of treatment intent. Cancer Res 1991; 51: 4999– 5002.

513

Suggest Documents