Upregulation of SPINK2 in acute myeloid leukemia

Abstract Objectives Acute myeloid leukemia (AML) is a highly heterogeneous disease. Although patients can be classified into risk groups based on their genetic changes, the prognosis of disease within these categories varies widely. This situation raises the need to search for new molecular markers related to AML. Serine peptidase inhibitor Kazal type 2 (SPINK2) has recently been reported to be upregulated in AML and associated with poor outcomes by meta-analysis and in a limited number of AML patients. Methods We analyzed SPINK2 mRNA expression in 62 patients (45 adult and 17 pediatric) with AML and 11 cell lines using quantitative Real-Time PCR (qRT-PCR). SPINK2 protein level was determined using ELISA in cell lines. Results We found that the expression of SPINK2 mRNA and protein levels in AML cell lines (HL60 and NB4) have increased compared to other cell lines (K562, Jurkat and NALM6, MCF7, HeLa, HUVEC, hFOB, 293T, U87). SPINK2 mRNA expression was upregulated in patients with AML compared to controls (p=0.004) and significantly lower in t(8;21)-positive patients compared to negative patients (p=0.0006). Conclusions Our results suggest that SPINK2 serves an important role in AML development. Further studies are needed to evaluate SPINK2 expression in AML patients with t(8.21) and investigate to clarify its prognostic value in various subgroups of AML.

SPINK2 is one of the differentially expressed genes in primitive hematopoietic stem cells obtained from the umbilical cord blood [12]. SPINK2 is significantly upregulated in the CD33+ bone marrow (BM) blast cells of a patient with AML compared to normal BM CD33+ cells in a case report [13]. It has been shown that SPINK2 is upregulated gene in AML in data from the Gene Expression Omnibus (GEO) database [14]. The results of this study showed that elevated SPINK2 expression was associated with poor outcomes [14]. This data need to be confirmed by further studies. This study aimed to determine the expression level of SPINK2 and to evaluate its prognostic value in AML.

Patient samples
In this retrospective study, 62 patients who were diagnosed at Istanbul University, Istanbul Medical Faculty, Pediatric Hematology Oncology and Department and Istanbul University-Cerrahpasa, Cerrahpasa Medical Faculty, Department of Internal Medicine, Division of Hematology were included. Blood or bone marrow samples were obtained for routine analysis with the informed consent from 17 pediatric (under 18 years) and 45 adult (18-80 years) patients with AML at diagnosis between 2013 and 2016. SPINK2 expression analysis was performed using leftover RNA samples of patients from routine fusion gene analysis. The control group consisted of three pediatric (under 18 years) and five adults (26-67 years old), who were bone marrow transplantation donors. Apart from healthy control, nine individuals without leukemia who underwent bone marrow biopsy due to the pre-diagnosis of leukemia were also used as another group of controls. The approval of Istanbul University Istanbul Medical Faculty Clinical Research Ethics Committee (reference number: 537, 24/05/2017) was obtained for this study. This study was conducted in accordance with the 1964 Declaration of Helsinki and its subsequent amendments. The demographic characteristics of the patients are shown in Table 1.
Determination of SPINK2 mRNA expression in patients with AML and cell lines using qRT-PCR RNA samples were extracted from cells using PureLink, RNA Mini Kit (Invitrogen, Carlsbad, CA, USA) following the protocol of the kit.
Transcriptor High Fidelity cDNA Synthesis Kit (Roche Diagnostic, Mannheim, Germany) was used for RT-PCR according to the manufacturer's instructions. cDNA was synthesized from 500 ng total RNA in 20 µL reaction volume.
A real-time quantitative PCR device (LightCycler, Roche Diagnostic, Mannheim, Germany) was used for quantitation. TATA binding protein (TBP) housekeeping gene was used as a control gene. Primer sequences and PCR product sizes are given in Table 2. SYBR Green master kit (Roche Diagnostic, Mannheim, Germany) and 1 µL cDNA sample were used for the 20 µL PCR reaction. The PCR conditions were as follows 60 s predenaturation at 95°C, 10 s at 95°C, 20 s at 68°C and 10 s at 72°C (45 cycles). After amplification, the program was applied for melting curve analysis for 0 s at 95°C, 10 s at 65°C and 0 s at 95°C. The differences in SPINK2 expression in patients were determined using ΔΔCt method according to the healthy control sample [15]. mRNA expression levels were indicated as arbitrary units (AU).

Statistical analysis
The data were analyzed using the SPSS Statistics Program Version 20.0 (IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp.) and GraphPad Prism 7.0. Kolmogorov-Smirnov and Shapiro-Wilk tests was used to determine normality. Independent sample T-test for parametric data and Kruskal-Wallis and Mann-Whitney U tests were used for non-parametric data. To investigate the correlation, Spearman's tests were used. Receiver operating characteristic curve (ROC) and the area under curve (AUC) were conducted to assess the value of SPINK2 expression for discriminating patients with AML from controls. Overall survival (OS) was defined as time from disease diagnosis to death from any cause or censoring for patients alive at their last known date of contact. The Kaplan-Meier analysis and log rank test were used to determine patients' survival rates. p<0.05 was considered statistically significant for all statistical analyses.
SPINK2 mRNA was correlated with intracellular and extracellular protein (p=0.0002, R=0.935 and p=0.046, R=0.610 respectively). There was also a statistically significant correlation between the intracellular and secreted protein levels (p=0.002, R=0.817).
SPINK2 expression and its prognostic significance in patients SPINK2 mRNA levels in patients with AML (median: 1.62 AU, range: 0.0032-28.67 AU) were significantly higher than in controls (median: 0.058 AU, range: 0.00017-1.51 AU), which showed that SPINK2 expression was significantly up-regulated in AML (p=0.004) (Figure 1). We also used nine samples from individuals without leukemia who underwent bone marrow biopsy due to the pre-diagnosis of leukemia as another group of control apart from healthy control samples. The median value of SPINK2 mRNA expression was 0.28 AU (range: 0.035-1.15 AU). We observed that SPINK2 was significantly increased in AML, even compared with nine samples without leukemia (p=0.022).
Additionally, ROC curves were generated to evaluate the diagnostic value of SPINK2 expression in AML. ROC analysis revealed that SPINK2 is a well diagnostic marker with the AUC value of 0.82 [95% confidence interval (CI): 0.685-0.946] (p=0.004). The result is shown in Figure 2.
The comparisons of SPINK2 expression according to the clinical and laboratory characteristics of patients with AML are given in Table 3. There was no significant relationship between SPINK2 expression and age, gender, white blood cell count and French American British (FAB) groups in patients (Table 3).
We categorized the patients (n=50) into two groups (SPINK2 high and SPINK2 low ) based on the median expression level of SPINK2 mRNA (1.62 AU) to evaluate the prognostic value of SPINK2 expression. According to Kaplan-Meier analysis, the 7-year overall survival rates were 27.9 ± 8.4%, for SPINK2 high and 40.6 ± 11.6% for SPINK2 low patients. There was no significant difference in OS between SPINK2 high and SPINK2 low patients (log-rank p= 0.40) (Figure 3). Median survival times were 15 (range: 1-92) months for SPINK2 high and 20.5 (range: 3-89) months for SPINK2 low patients.

Discussion
Acute myeloid leukemia (AML) is a heterogeneous disease that results in abnormal hematopoiesis, block in differentiation, excessive proliferation of myeloblasts, and disruption of normal blood cell production [17]. Although the presence of various mutations is used as a prognostic factor for Figure 1: The median expression level of SPINK2 mRNA in healthy controls and patients with AML. SPINK2 mRNA levels were determined using cDNA derived from 25 ng total RNA. The relative mRNA expression levels were calculated using the comparative CT method. The long horizontal lines above the dots represent the median. The short horizontal lines above the dots represent the median interquartile range. AML, acute myeloid leukemia.
p-Values marked with bold indicate statistically significant differences between the groups. AML, acute myeloid leukemia; WBC, white blood cell; FAB, French American British; M-M, patient's FAB, groups; NA, not avaiable. the classification of the patients, molecular subgroups of patients need further improvement. This situation raises the need to search for new molecular markers related to AML. It has been reported that the SPINK2 is overexpressed in hematopoietic progenitor stem cells [18]. He et al. identified SPINK2 as one of the differentially expressed genes (DEG) when comparing CD34+ and CD133+ hematopoietic progenitor stem cells and CD34+ cells purified from umbilical cord blood [12]. Transcriptome analysis in CD33+BM blast cells purified from a patient with AML transformed from myelodysplastic syndrome showed that the expression of SPINK2 was higher than control CD33+ BM cells [13].
In a recent study, to the identification of pivotal genes in AML, microarray data from Gene Expression Omnibus (GEO) analyzed via bioinformatics analysis [14]. They revealed that the expression level of SPINK2 was the most altered among all upregulated genes. They validated this result using qRT-PCR in six controls and 12 patients with AML and simultaneously, in GEPIA (Gene Expression Profilling Interactive Analysis) and Oncomine data. They found that elevated SPINK2 expression was associated with poor outcomes [14]. Barresi et al. found that SPINK2 was overexpressed in NUP98+ patients from the data GEO database and 358 RNAseq AML samples from TARGET data. They also confirmed in pediatric patients with primary induction failure using qRT-PCR [19].
In our study, we found that SPINK2 mRNA was upregulated in pediatric, adult and total patients with AML, which was consistent with the results of the studies by Barresi et al. and Xue et al. [14,19]. The ROC analysis showed that SPINK2 expression might serve as a potential biomarker for distinguishing AML from controls.
SPINK2 expression was evaluated in pediatric, adult and total AML patients with t(15;17), t(8;21) and inv (16) mutations. We found that SPINK2 expression significantly decreased in t(8;21) positive patients compared to negative patients for both total and adult patients with AML. This is the first study to show a statistically significant association between t(8;21) and low SPINK2 expression. t(8;21) is known to be associated with a favorable prognosis [3]. However, Xue et al. reported that high SPINK2 expression is associated with poor prognosis [14]. We suggest that the high survival rate of SPINK2 low patients might be related to t(8;21) positivity. The association between SPINK2 expression and t(8;21) should be validated in large patient cohorts.
In this study, we were only analyzed SPINK2 mRNA expression since we did not have protein samples from the patients. We could not detect the reflection of the SPINK2 mRNA upregulation to the protein levels in patients with AML. Therefore, we analyzed SPINK2 expression at both mRNA and protein levels in 11 cell lines to investigate the correlation between SPINK2 mRNA and protein levels. SPINK2 mRNA levels were determined by qRT-PCR and the intracellular and secreted protein levels were determined by ELISA in cell lines. The expression of SPINK2 mRNA in AML cell lines was significantly increased compared to other cell lines. It was also significantly increased in leukemia cell lines compared to non-leukemic cell lines. To understand different implications of this molecule in other hematopoietic neoplasms, SPINK2 expression should also be investigated in acute lymphoblastic leukemia, myelodysplastic syndrome and myeloproliferative neoplasms.
The intracellular SPINK2 protein level was found to be significantly increased in AML cell lines compared to other cell lines. A statistically significant correlation was found between SPINK2 mRNA levels and intracellular and secreted protein levels. There was also a statistically significant correlation between the levels of intracellular protein and secreted protein. The correlation between SPINK2 mRNA and protein levels suggests that the elevated SPINK2 mRNA expression results in increased protein activity.
Xue et al. analyzed the effect of SPINK2 on the prognosis of patients with AML using the GEPIA (106 patients) and UALCAN datasets (163 patients). They showed that SPINK2 high patients had shorter survival times than SPIN-K2 low patients [14]. In this study, there was no significant difference between SPINK2 high and SPINK2 low patients. Since we were able to perform survival analysis in a limited number of patients, no statistically significant difference could be found.
This study has some limitations. First, survival analysis was performed in a limited number of patients, since the outcomes of some patients were not available. Secondly, SPINK2 expression was not analyzed in different molecular subtypes of AML because of mutation profiles of the patients were not available.
Our findings showed that SPINK2 is upregulated in patients with AML and SPINK2 expression is significantly low in t(8;21)-positive AML subgroup. It should be evaluated in large patient cohorts to clarify the prognostic significance of SPINK2 expression. Further studies are required to investigate the potential role of SPINK2 in the development of AML and SPINK2 targeting therapies.
Research funding: This study was funded by Scientific Research Projects Coordination Unit of Istanbul University (Project numbers TYL-2017-26523 and FBG-2018-30048). Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission. Competing interests: Authors state no conflict of interest.
Informed consent: Informed consent was obtained from all individuals included in this study. Ethical approval: The Istanbul University Medical Faculty Clinical Research Ethics Committee gave approval to the study's procedures (reference number: 537, 24/05/2017). All subjects were informed of the study and voluntarily agreed to participate, providing a written consent.