TURKISH JOURNAL OF ONCOLOGY

Summary

OBJECTIVE
Endoplasmic reticulum (ER) stress, which occurs as a result of the accumulation of misfolded or unfolded proteins, has been observed in many cancers. To re-establish the ER homeostasis, pathways of unfolded protein response (UPR) are activated. The effect of ER stress-activated UPR pathways on leukemogenesis has not been elucidated. In this study, we aimed to find out whether activated UPR pathways were involved in acute myeloid leukemia (AML).

METHODS
Expression levels of the 2 UPR components, binding immunoglobulin protein (BiP), and X-box protein 1 (XBP1), as well as the HLA-B were analyzed in 100 newly diagnosed AML patients using quantitative and qualitative polymerase chain reaction techniques.

RESULTS
Fifteen of the 100 newly diagnosed AML patients were shown to carry the spliced XBP1 (XBP1s) variant. Furthermore, in patients with acute promyelocytic leukemia (APL), a subtype of AML, expression levels of XBP1, and BiP were significantly higher than in non-APL patients (p<0.001 and p<0.015, respectively). It was determined that 6 of 15 (40%) AML patients carrying XBP1s variant were positive for myeloperoxidase.

CONCLUSION
Our study provides adequate evidences for the involvement of activated UPR pathways in some of the AML patients by showing the presence XBP1s variant and increased expression levels for XBP1 and BiP. Thus, agents targeting XBP1s might serve as a new approach for anti-cancer therapy.

Summary

Introduction

Proteins are the principal actors of cellular function. Proper regulation of protein synthesis is utmost importance for the maintenance of cellular homeostasis and survival of cells in the presence of external and internal insults. Cancer leads to a number of cytotoxic conditions caused by tumor cells such as hypoxia, nutritional deficiency, and changes in pH, which results in the disruption of normal protein synthesis and accumulation of unfolded or misfolded proteins in the endoplasmic reticulum (ER), called ER stress. Cells have acquired mechanisms to cope with such aberrant protein synthesis and to re-establish the ER homeostasis. In tumor cells, various genetic, transcriptional, and metabolic abnormalities cause persistent ER stress. It creates adverse microenvironments affecting the functions, fate, and survival of cells.[1]

The unfolded protein response (UPR) is cellular stress response mechanism designed to reinstate correct protein folding and to prevent the accumulation of unfolded proteins in the ER lumen. Cancer cells stimulate UPR signaling pathways, consisting of three main tracks called protein kinase RNA-like ER kinase, activating transcription factor 6, and inositol-requiring enzyme 1 (IRE1), to overcome the ER stress.[2] Activation of IRE1α regulates several genes involved in ER homeostasis, cell survival, autophagy, apoptosis, angiogenesis, metastasis, drug resistance, and lipid metabolism.[3]

IRE1α cleaves X-box protein 1 (XBP1) mRNA when the cell enters into ER stress, resulting in the formation of XBP1s, a highly active transcription factor. The unspliced isoform is constitutively expressed and is called XBP1u.[4] The XBP1s involved in the UPR signal pathway, allow the survival of the ER stress in the cell by inducing genes included in ER chaperones and ERrelated degradation mechanism.[5-7] In addition, ER chaperone protein (Binding immunoglobulin protein [BiP]/GRP78) is one of the most active components of cancer cells and has been reported to be overexpressed in many different types of cancer. In ER stress, BiP contributes to re-establishment of homeostasis through the correction of unfolded or misfolded proteins.[8]

XBP1s expression levels have been shown to increase in many cancers, such as multiple myeloma, breast cancer, hepatocellular carcinoma, and glioblastoma. This has been associated with cancer progression. There are reports stating increased expression of XBP1s in B-acute lymphoblastic leukemia (B-ALL) and in acute promyelocytic leukemia (APL) cells; in the latter, the PML-RARA fusion protein being responsible for the activation of UPR.[9,10] In addition, XBP1 has many functions in pre-B-ALL cells and activates proapoptotic pathways when the cell loses its vital survival and proliferation functions.[3]

Acute myeloid leukemia (AML) is a cancer of the hematopoietic system characterized by the malignant transformation of myeloid progenitor cells into blasts which interfere with normal hematopoiesis through accumulating in the bone marrow and peripheral blood.[11,12]

To investigate the role of UPR involvement in AML, we analyzed the expression levels of 2 UPR components, BiP and XBP1, as well as HLA-B, which is known to play a role in ER pathways and is discussed to be involved in the pathogenesis of AML.[13,14]

Introduction

Methods

Patient Samples
A total of 100 patients including 23 pediatric patients (12 female and 11 male) with diagnosed AML at Istanbul University Istanbul Faculty of Medicine, Division of pediatric hematology and oncology, and 77 adult (33 female and 44 male) patients with diagnosed AML at Istanbul University, Cerrahpaşa Medical Faculty, Haematology Department were included in this study. These patients are consecutive patients at diagnosis. Clinical and laboratory characteristics of the patients are presented in Table 1. In addition, bone marrow or peripheral blood samples were obtained from 8 of these 100 patients at 33rd day of treatment. Commercial total RNA samples (Clontech USA, cat no: 636591) (aged 22-85 years) were used as healthy control.

Table 1 Clinical and laboratory characteristics of AML patients

Quantitative and Qualitative Polymerase Chain Reaction (PCR)
RNA isolation from the AML samples was performed using PureLink, RNA Mini Kit (Invitrogen). cDNA synthesis from the RNAs of the resulting patient samples was performed using the Transcriptor High Fidelity cDNA Synthesis Kit (Roche). cDNA synthesis conditions were adjusted to be 1 h at 37°C and 5 min at 94°C. Quantitative real-time PCR method was performed to determine the expression levels of XBP1, HLA-B, and BiP genes. LightCycler 480 instrument and Light Cycler 480 SYBR Green I kit were used for quantitation. Primers were designed for HLA-B and BiP genes, and the 18S rRNA gene was used as a reference gene. Primers designed specifically for XBP1 are capable of distinguishing between XBP1s and XBP1u. To observe the XBP1s variant, the qualitative PCR conditions were set at one cycle of 94°C for 2 min after preliminary denaturation, 35 cycles of 30 s at 94°C, 30 s at 57°C, 30 s at 72°C, and final extension for 5 min at 72°C. The primers used for XBP1 are as follows: Forward, 5"CCTTGTAGTTGAGAACCAGG-3", and reverse, 5"-GGGGCTTGGTATATATGTGG-3". The primers generate two PCR products of 442 (XBP1u) and 416 (XBP1s) bps, respectively. PCR products were also run on a 3% agarose gel (Fig. 1).

Fig. 1. Sample 1, 2, 5 are X-box protein 1 (XBP1) unsplicing variants (XBP1u) and 3, 4 XBP1 splicing variants (XBPs) variants in acute myeloid leukaemia cases.
M: Marker Puc Mix 8.

Statistical Analysis
Data obtained in the study were statistically analyzed using the SPSS 16.0 Software (SPSS, Inc., Chicago, IL, USA). Independent sample t-test was used for comparing parametric data, and Kruskal-Wallis and Mann- Whitney U tests were used to compare non-parametric data. Correlation was studied using the Pearson and Spearman"s tests. Chi-square test was used to compare ratios. P<0.05 was considered statistically significant.

Methods

Results

In our study, XBP1s variant was found in 15 out of 100 patients with AML (%15) (Table 2). Two of these patients were carried t(15; 17), one patient was positive for inv(16), another one case carried FLT3-ITD mutation, and the rest 11 cases did not display any of these specific changes (Fig. 2).

Table 2 Distribution of XBP1 variants in AML cases

Fig. 2. Pie chart of distribution of X-box protein 1 (XBP1) variants in acute myeloid leukemia (AML) patients. (a) Distribution of XBP1u variants in AML patients (b) Distribution of XBP1s variants in AML patients.
t: Translocation; inv: Inversion; FLT3-ITD: Fms-like tyrosine kinase 3 internal tandem duplication

Expression levels of XBP1, BiP, and HLA-B genes were analyzed to determine impact of the ER-mediated pathway and were compared with other clinical parameters in AML such as age, sex, and leukocyte count. There was no significant correlation in between sex (Appendix 1), age, and leukocyte count compared with XBP1, BiP, and HLA-B expressions (Table 3).

APPENDIX 1 Correlation between HLA-B, BiP, XBP1 genes and gender in AML patients

Table 3 Correlation analysis between HLA-B, BiP, XBP1 expressions and patient leukocyte count and ages

A significant correlation (p<0.001) was observed between BiP and XBP1 expressions. HLA-B and XBP1 expressions were not significantly correlated (p=0.14), while there was a robust correlation between the expression levels of HLA-B and BiP (p=0.019) (Table 3). When BiP and XBP1 expression levels were analyzed against the presence of t(15;17), t(8;21), inv(16), and FLT3-ITD mutations, BiP and XBP1 levels were found to be significantly higher in carriers of t(15;17) (p<0.001) and p=0.015, respectively) (Table 4) (Appendix 2). HLA-B and BiP expression levels did not correlate significantly among XBP1 expression (p=0.52 and p=0.19, respectively) (Table 3).

APPENDIX 2. XBP1 gene expression level in t(15;17) positive and negative AML patients.
XBP1: X-box protein 1; t: Translocation; AML: Acute myeloid leukemia

Table 4 Comparison of HLA-BiP and XBP1 expressions between t(15;17), t(8;21), inv(16), and FLT3 positive and negative cases

Furthermore, XBP1u expression was found higher than XBP1s between patients with XBP1s variants and without XBP1s variants (p=0.003). No significant results were found when age, gender, leukocyte count, and blast rate were compared statistically in XBP1s and XBP1u cases. The results are listed in Table 5. XBP1s variants in two AML patients with t(15;17) and one patient inv(16) were not observed in the 33^rd day samples of the patients, although they were still positive for both t(15; 17) and inv(16). When the patients carrying the XBP1s variant were examined, 6 out of 15 patients were found to be myeloperoxidase (MPO) positive patients.

Table 5 Comparison of age, sex, blast, and leukocyte values of AML patients with XBP1u and XBP1s

Mechanisms that activate the UPR in leukemia cells have not yet been fully disclosed. When the ER stress activates the UPR pathways, IRE1α cuts XBP1 and forms the XBP1s variant. Our results of XBP1s variant rates (15%) which we easily detected by quantitative PCR is similar to those of Schardt et al. (16.2%) in AML patients. Unlike Schardt et al.,[15] however, we could identify XBP1s variant in patients with inv[16]. Our results confirm the findings of the latter study and serve as preliminary evidence for the relationship between AML and UPR. Expression levels of BiP and XBP1s involved in UPR were observed to be higher in the t(15;17) positive patients. This positive correlation between BiP and XBP1 levels indicates that UPR pathways are active in APL. Vice versa HLA-B expression levels were not significantly elevated in patients with AML. When proteins are improperly folded or unfolded in the cell, they tend to accumulate and activate the ER stress pathways.[7] MPO expression, as measured by cytochemistry, in leukemia blasts is widely accepted as a gold marker for the diagnosis of AML. Expression of MPO provides critical information regarding the phenotype of AML cells.[16]

We thought that, in patients with high MPO levels, ER stress pathways could be highly stimulated due to the accumulation of these MPO granules. There was an increase in the MPO levels in the 6 AML cases, in which found the XBP1s variant.

This finding suggests that the accumulation of MPO can trigger ER stress and activate UPR pathways, especially IRE1-XBP1 pathway. In the Ph + ALL subclasses, imatinib treatment reduced XBP1 expression.[9] Since XBP1 is more associated with survival, XBP1 expression is also decreased when cell apoptosis is induced by bortezomib treatment in patients with multiple myeloma. [17] In our cohort, we could show the disappearance of XBP1s variants in 2 patients with t(15;17) and one with inv(16) positivity. This indicates that anti-cancer treatment might induce XBP1 cleavage by activating UPR,[8] but that XBP1 expression is likely to decrease as different UPR pathways can be selected. The unsignificant results might be related to low cohort number. We hope that this study will be forward for other research.

Results

Discussion

AML is a malignant disease that is caused by uncontrolled clonal proliferation of precursor hematopoietic cells. While several genetic and environmental factors have been accused to involve in the pathogenesis of AML, the etiology is not fully understood yet.[11,12] Earlier studies have demonstrated an association between cancer and ER stress components.[15] Knowing this we analyzed the expression levels of 2 important UPR components, XBP1 and BiP genes, in patients with AML. Mechanisms that activate the UPR in leukemia cells have not yet been fully disclosed. When the ER stress activates the UPR pathways, IRE1α cuts XBP1 and forms the XBP1s variant. Our results of XBP1s variant rates (15%) which we easily detected by quantitative PCR is similar to those of Schardt et al. (16.2%) in AML patients. Unlike Schardt et al.,[15] however, we could identify XBP1s variant in patients with inv[16]. Our results confirm the findings of the latter study and serve as preliminary evidence for the relationship between AML and UPR. Expression levels of BiP and XBP1s involved in UPR were observed to be higher in the t(15;17) positive patients. This positive correlation between BiP and XBP1 levels indicates that UPR pathways are active in APL. Vice versa HLA-B expression levels were not significantly elevated in patients with AML. When proteins are improperly folded or unfolded in the cell, they tend to accumulate and activate the ER stress pathways.[7] MPO expression, as measured by cytochemistry, in leukemia blasts is widely accepted as a gold marker for the diagnosis of AML. Expression of MPO provides critical information regarding the phenotype of AML cells.[16] We thought that, in patients with high MPO levels, ER stress pathways could be highly stimulated due to the accumulation of these MPO granules. There was an increase in the MPO levels in the 6 AML cases, in which found the XBP1s variant. This finding suggests that the accumulation of MPO can trigger ER stress and activate UPR pathways, especially IRE1-XBP1 pathway.

In the Ph + ALL subclasses, imatinib treatment reduced XBP1 expression.[9] Since XBP1 is more associated with survival, XBP1 expression is also decreased when cell apoptosis is induced by bortezomib treatment in patients with multiple myeloma.[17] In our cohort, we could show the disappearance of XBP1s variants in 2 patients with t(15;17) and one with inv(16) positivity. This indicates that anti-cancer treatment might induce XBP1 cleavage by activating UPR,[8] but that XBP1 expression is likely to decrease as different UPR pathways can be selected. The unsignificant results might be related to low cohort number. We hope that this study will be forward for other research.

Discussion

Conclusion

In this study, we have found that pathways of UPR are active in t(15;17)-positive AML, and we have shown that these pathways may be related to AML pathogenesis. In ER stress, XBP1 involves a strong UPR pathway more associated with cell survival.[18] Our findings indicate that XBP1 might play a facilitating role in the pathogenesis of AML. The mechanism of function of XBP1 in setting of AML is an area of interest for developing new target specific therapeutic strategies which involve UPR. Therapeutic approaches targeting UPR in cancer includes enhancing misfolded protein synthesis in ER to overload the UPR which results in more severe ER stress to cause cell death or inhibiting UPR adaptive and survival pathways for increasing susceptibility to anticancer therapy.[8,19]

Peer-review: Externally peer-reviewed.

Conflict of Interest: All authors declared no conflict of interest.

Ethics Committee Approval: The study was approved by the Istanbul University Istanbul Faculty of Medicine Clinical Research Ethics Committee (no: 21, date: 22/12/2017). Financial Support: The project was supported by the Scientific Research Projects Coordination Unit of Istanbul University (Project number: 574).

Authorship contributions: Concept - N.A., N.E.B., Ş.P.; Design - N.A., N.E.B., Z.E.; Supervision - N.A., S.S.E.; Funding - N.A.; Materials - M.C.A., A.Ü., T.E., A.S., Z.K., Ş.P.; Data collection and/or processing - M.C.A., A.Ü., T.E., A.S., Z.K.; Data analysis and/or interpretation - N.E.B., Ş.P., S.S.E., Z.E.; Literature search - M.C.A., A.Ü., T.E., A.S., Z.K., Ş.P., Z.E.; Writing - N.A., N.E.B., Ş.P., M.C.A.; Critical review - S.S.E., M.C.A., A.Ü.

Conclusion