The characteristic expression pattern of BMI-1 and SALL4 genes in placenta tissue and cord blood

Introduction SALL4 and BMI-1 are important factors in hematopoiesis. Placental tissue (PT) and umbilical cord blood (CB) are rich in hematopoietic stem/progenitor cells (HSCs/HPCs), but their SALL4 and BMI-1 expression levels remain unknown. Methods Real-time PCR was used to determine the expression level of these genes in PT and CB from ten cases, and ten healthy donors were used as controls. Results A significantly higher BMI-1 and SALL4 gene expression level was found in PT (median: 17.548 and 34.362, respectively) than in cord blood mononuclear cells (CBMCs) (median: 2.071 and 11.300, respectively) (P = 0.0001 and P = 0.007) and healthy peripheral blood mononuclear cells (PBMCs) (median: 0.259 and 0.384, respectively) (P = 0.001 and P <0.0001), and their expression level was lower in PBMCs than in CBMCs (P = 0.029 and P = 0.002). A positive correlation between the BMI-1 and SALL4 genes was found in the PT and CB groups, while there was no significant correlation between these genes in the healthy group. There was also no significant correlation between the expression level of each gene in PT and CB. Conclusions These results describe the characteristic features of the BMI-1 and SALL4 gene expression pattern in placental tissue and cord blood. Placental tissue with higher expression level of both genes may be considered as a potential resource for SALL4-related HPC expansion.

Sal-like protein 4 (SALL4) and B cell-specific MLV integration site 1 (BMI-1) are important factors in hematopoiesis and are expressed in hematopoietic stem/ progenitor cells (HSCs/HPCs) and myeloid leukemia cells [7]. Placental tissue (PT) and CB comprise rich HSCs/HPCs; however, little is known about the difference in the expression level of SALL4 and BMI-1 in PT and CB.
SALL4, a newly identified zinc-finger transcription factor that is a member of the SALL gene family, was originally cloned based on its sequence homology to Drosophila spalt (sal) [7,8]. This protein plays important roles in maintaining embryonic stem cells (ESC) pluripotency and HSC/HPC self-renewal properties, and it has been recently proposed for use in CB expansion [9]. Recently, it was demonstrated that BMI-1 is a direct SALL4 target gene. BMI-1 is a member of the polycomb group of proteins, and it was initially identified in * Correspondence: yangqiuli@hotmail.com † Equal contributors 1 Drosophila as a repressor of homeotic genes [7,10]. This protein is highly expressed in purified HSCs, its expression declines with differentiation [7,11], and it plays an essential role in regulating adult, self-renewing HSCs/HPCs [7,[11][12][13]. The induction of SALL4 expression is associated with increased levels of histone H3-K4 and H3-K79 methylation in the BMI-1 promoter, indicating a novel connection between SALL4 and the polycomb group proteins [7].
In this study, we sought to characterize the expression pattern of the BMI-1 and SALL4 genes in PT and CB.

Samples
Ten placental tissue samples were obtained from fullterm deliveries, and umbilical cord blood was obtained at the same time from full-term healthy babies with the mothers' consent. All human tissue and cell samples were obtained with consent from the human subjects. Peripheral blood samples from ten healthy donors with informed consent were used as control. All of the procedures were conducted according to the guidelines of the Medical Ethics Committees of the Health Bureau of the Guangdong Province of China, and ethical approval was obtained from the Ethics Committee of Medical School of Jinan University for this study.

Quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR)
Mononuclear cells were isolated from cord blood samples (CBMCs) and healthy peripheral blood (PBMCs) by Ficoll-Hypaque gradient centrifugation. The placental tissue (approximately 100 mg) was obtained by curettage of the central portion of the placenta (decidua in majority). RNA isolation and cDNA synthesis were performed according to the manufacturer's protocol [14].
The expression level of BMI-1, SALL4, and the β2 microglobulin (β2-MG) reference gene was determined by SYBR Green I real-time PCR as previously described [15].

Statistical analyses
Differences in mRNA expression between two groups were analyzed using the Mann-Whitney test. Data are presented as the median. Spearman's rank correlation analysis was used to analyze the SALL4 and BMI-1 mRNA levels in different samples. Differences were considered statistically significant with a P <0.05.

Results and Discussion
Due to its wide availability, low cost and lack of ethical concerns, CB provides an attractive source of stem cells for investigational and therapeutic uses [1][2][3]. Recently, in the promising field of regenerative medicine, human perinatal stem cells that could be isolated from normally discarded human placentae are of great interest as potential stem cells with clinical applications. Perinatal stem cells are an ideal cell source in terms of availability, the fewer number of ethical concerns, less DNA damage, and so on [16,17]. The biological characteristics of CBderived HSCs/HPCs were characterized in numerous studies [4,6]. SALL4 may act as a critical regulator of the fate of hematopoietic cells. In normal bone marrow, SALL4 is selectively expressed in primitive hematopoietic precursors and is rapidly downregulated following differentiation. Therefore, the SALL4 gene expression level should be positively correlated with stem cells [7]. However, little is known about the difference in the expression level of SALL4 in CB; thus, it may be interesting to determine the expression characteristics of this gene in PT collected from the same case as the CB to compare expression pattern differences. In this study, we found that the SALL4 expression level was higher in PT Figure 1 Expression level of SALL4 and BMI-1. Expression level of the SALL4 (A) and BMI-1 (B) genes in placental tissue (PT), cord blood and peripheral blood mononuclear cells (CBMCs and PBMCs, respectively) from healthy individuals (HI). A significant difference was found among the three groups.
(median: 34.362) than that in CBMCs (median: 11.300; P = 0.007) and PBMCs (median: 0.384; P < 0.0001), and a lower SALL4 expression level was found in PBMCs than CBMCs ( Figure 1A; P = 0.002). It has been reported that SALL4-expanded HSCs/HPCs retain multilineage repopulation and long-term engraftment abilities, which are clinically significant [18]. Thus, SALL4 could be used to stimulate the large scale ex vivo expansion of HSCs/ HPCs, and it is particularly interesting whether SALL4rich placental tissue is able to be used as a resource for HSC/HPC expansion.
Because BMI-1 is a direct SALL4 target gene [7], BMI-1 expression may be correlated with that of SALL4. Our results also demonstrated that the BMI-1 gene expression level was significantly higher in PT (median: 17.548) than that in CBMCs (median: 2.071; P = 0.0001) and PBMCs (median: 0.259; P = 0.001), and its expression level was lower in PBMCs than that in CBMCs ( Figure 1B; P = 0.029). A positive correlation between the BMI-1 and SALL4 genes was found in the PT and CB groups (rs = 0.648, P = 0.043 and rs = 0.721, P = 0.019, respectively), while there was no significant correlation between these genes in the healthy group (rs = -0.212, P = 0.556) (Figure 2). This result further supports the finding of a relationship between the expression pattern of BMI-1 and SALL4 in HSC/HPC-rich CB and placental tissue, while PBMCs contain only limited BMI-1 and SALL4 expression, which may reflect their expression in different levels of HPCs and myeloid cells [7]. Moreover, this result may indicate that the PT niche contains hematopoietic transcription factors that are able to enhance HPC expansion. It is thought that the expression level of genes in PT and CB derived from the same case may demonstrate a positive correlation; however, interestingly, the expression levels of BMI-1 and SALL4 were not significantly correlated between the PT and CB (Figure 3). This result may be related to a difference in the HSC/HPC numbers in CB or may imply that placental tissue-derived HSCs/HPCs have a different SALL4 and BMI-1 expression pattern. Further analysis is needed to compare the expression levels of these genes and the percentages of HSCs/HPCs in CB and PT.

Conclusions
We determined the expression characteristics of BMI-1 and SALL4 in placental tissue and cord blood. The results of this study may contribute to a better understanding of the expression characteristics and correlation of SALL4 and BMI-1 in placenta tissue. Placental tissue with higher expression level of both genes may be considered as a potential resource for SALL4-related HPC expansion.