| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Pediatric Hemato-Oncology, Safra Children’s Hospital and
b Department of Obstetrics and Gynecology, The Sheba Medical Center, Tel-Hashomer, Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel;
c Department of Molecular Cell Biology and
d Department of Complex Systems, Weizmann Institute of Science, Rehovot, Israel;
e Department of Obstetrics and Gynecology, Rambam Medical Center, Bruce Rappaport Institute of Technology, Technion-Israel Institute of Technology, Haifa, Israel
Key Words. Hematopoietic stem cells • Stemness • Peripheral blood • Cord blood • Gene expression
Correspondence: Amos Toren, M.D., Ph.D., Department of Pediatric Hematology-Oncology, Sheba Medical Center, Tel-Hashomer, Israel 52621. Telephone: 972-3-5303037; Fax: 972-3-5303031; e-mail: amost{at}post.tau.ac.il
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Umbilical cord blood (CB) and mobilized peripheral blood (PB) are relatively new sources of HSCs that have been increasingly used in clinical transplantations. These two sources show considerable differences in their proliferative capacity, engraftment kinetics, and differentiation potential [1]. Differences in these biological properties may stem from the expression of a different set of genes in these two groups of HSCs. Despite these differences, these stem cells, similar to stem cells derived from other tissues, share many biological properties. This may result from "stemness" gene expression profile characteristic of stem cells.
Recent studies using different DNA microarray technologies (oligonucleotide-based or cDNA arrays) for the analysis of gene expression profiles of HSCs were performed mainly in mice using antibodies reactive to CD34 with a combination of other markers to enrich for stem cells. Some studies focused on the differences between primitive HSCs, the more differentiated progenitors, and the mature cells. Other studies concentrated on the differences and similarities in gene expression between stem cells from different tissues (hematopoietic, neuronal, embryonic, etc.) [2–9].
The CD34 antigen was usually served for the isolation of HSCs both for the purpose of stem cell transplantation and for laboratory studies. However, recent studies show that the CD34– fraction also has a repopulating capacity and includes cells that are precursors of CD34+ cells [10]. The CD133 antigen (also known as AC133 or prominin-1) was found to be coexpressed with CD34 but also found in CD34–CD38–Lin– precursors [11, 12]. A small, rare population within the CD34–Lin– cells that expresses CD133 has a high progenitor activity and was capable of giving rise to CD34+ cells [11], and coexpression of CD133+ and CD34+ led to a higher clonogenic capacity compared with the CD133–CD34+ cells [13, 14]. Another recent study [15] pointed out that among the CD34+/CD38– cells, the fraction of slow dividing cells that is associated with primitive function and self-renewal expresses high levels of the CD133 gene in contrast with the fast dividing cells. CD133 may have a central role in the asymmetric division that is believed to characterize true stemness [15]. Altogether, these studies suggest that CD133+ may provide a more appropriate marker to enrich stem cells and therefore was used as an isolation target in our study.
The aim of this study was to identify specific genes that are upregulated in progenitors and HSCs originating from CB and mobilized PB and may represent the stemness genes in HSCs. Because the CD133+ cell fraction used in this study is far from being a pure HSC fraction, the term stemness is used in this work merely for simplicity, and many of the stemness genes defined in the study are in fact stem/progenitor or progenitor rather than stem cell–proper genes. In addition, we looked for genes differentially expressed in HSCs from either PB or CB.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two separate pools of CB cells were obtained: One pool contained cells derived from 5 units (average 75 ml/U of CB), and the second pool was derived from 15 units of CB (average 40 ml/U).
PB specimens were collected, by pheresis using CobeSpectra Aphersis System (Gambro BCT, Planegg-Martinsried, Germany, http://www.cobebct.com/cps/rde/xchg/SID-3916ADC2-7F00F89F/gambro-bct/hs.xsl/index.htm), from two healthy adult donors after mobilization with granulocyte colony-stimulating factor 10 µg/kg per day for 4 days.
CD133+ Cell Enrichment
CD133+ cells were enriched from both sources, CB and PB, using the Midimacs direct magnetic labeling system according to the manufacturer’s protocol (Miltenyi Biotec, Bergisch Gladbach, Germany, http://www.miltenyibiotec.com). Briefly, 1 x 109 mononuclear (CB or PB) cells were incubated with anti-CD133+ MicroBeads and with FcR-blocking antibody for 30 minutes at 4°C. The cells were then washed and applied to a positive selection on Macs column placed in a magnetic field (Miltenyi Biotec). The cells were allowed to pass through the column. The column was washed and then removed from the magnetic field, and the positive cell fraction was eluted. The purity of the positive fraction was evaluated by fluorescence-activated cell sorting (FACS) analysis (purity 85%). The nonselected cells (CD 133–) served as the differentiated cells.
For each comparative analysis, FACS, microarray and quantitative real-time polymerase chain reaction (QRT-PCR), two samples of CB 133+, two samples of CB CD 133–, two samples of PB 133+, and two samples of PB CD 133– were used. The precipitated cells were stored in Trizol at –70°C until RNA isolation.
RNA Isolation
Total RNA was isolated from each sample using TRIZOL (catalog no. 15596; Gibco BRL Life Technologies, Invitrogen, Carls-bad, CA, http://www.invitrogen.com).
Array Processing, Sample Preparation, and Hybridization All experiments were performed using Affymetrix Human Hu133A oligonucleotide arrays containing 22,215 probe sets (PSs) as described in the Affymetrix human_datasheet.pdf (http://www.affymetrix.com/support/technical/datasheets/human_datasheet.pdf) (Affymetrix, Santa Clara, CA, http://www.affymetrix.com/index.affx). Total RNA from each sample was used to prepare biotinylated target cRNA, with minor modifications from the manufacturer’s recommendations (Affymetrix Expression Manual). Briefly, 10 µg of mRNA was used to generate first-strand cDNA using a T7-linked oligo(dT) primer. After second-strand synthesis, in vitro transcription was performed with biotinylated uridine triphosphate and cytidine triphosphate (Enzo Diagnostics, Farmingdale, NY, http://www.enzo.com/index_flash.asp), resulting in approximately 100-fold amplification of RNA. A complete description of the procedures is available in the Affymetrix Expression Manual. The target cDNA generated from each sample was processed per the manufacturer’s recommendation using an Affymetrix GeneChip Instrument System (Affymetrix Expression Manual). Briefly, spike controls were added to 10 µg of fragmented cDNA before overnight hybridization. Arrays were then washed and stained with streptavidin-phycoerythrin before being scanned on an Affymetrix GeneChip scanner. A complete description of these procedures is available in the Affymetrix Expression Manual.
Additionally, the quality and the amount of the total RNA were analyzed using an agarose gel. After scanning, array images were assessed by eye to confirm scanner alignment and the absence of significant bubbles or scratches on the chip surface. Ratios of 3'/5' for GAPDH and beta-actin were confirmed to be within acceptable limits (1.05–1.18), and BioB spike controls were found to be present on all the arrays, with BioC, BioD, and CreX also present in increasing intensity. When scaled to a target intensity of 150 (using Affymetrix MAS 5.0 array analysis software), scaling factors for all arrays were within acceptable limits (0.855–1.533), as were background, Q value, and mean intensities. Details of quality control measures can be found in supplemental online Table 1.
Data Analysis
Genes were filtered using Mas 5 algorithm results. A list of 14,025 valid PSs, representing PSs with signals higher than 20 and detected as present in at least one sample, was obtained (supplemental online Table 2). CD133+ and CD133– samples were compared. Additional filtering excluded upregulated genes in CD133+ samples with signals lower than 20 or detected as absent and downregulated genes with baseline, CD133–, signals lower than 20, and detected as absent. Each comparison generated a list of active genes representing genes changed by at least twofold (log ratio [LR] 
1) and detected as an increase (p = .0025) or marginal increase (p = .003) in the two duplicates or changed by at least twofold (LR 
–1) and detected as a decrease (p = .0025) or marginal decrease (p = .003) in the two duplicates in at least one of the comparisons. Hierarchical clustering was performed using Spotfire Decision Site for Functional Genomics (Somerville, MA, http://www.spotfire.com). Genes were classified into functional groups using the GO Annotation Tool [16]. Over-representation calculations were done using Ease [17]. Functional classifications with an Ease score lower than 0.05 were marked as over-represented.
QRT-PCR QRT-PCR assays were developed to determine the level of expression of the genes FZD6, HOXA9, FLT3, MLLT3, TIE, HLF, SPINK2, and MEIS1 using the Sybr Green method. All QRT-PCR reactions were performed on a 7900HT ABI platform (Perkin-Elmer/Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com). Primers (Danyel Biotech, Rehovot, Israel, http://www.danyel.co.il) were designed according to Primer-Express software. Forward and reverse primers were designed in different exons to eliminate DNA contamination. Sybr Green Master Mix was purchased from QuanTitect (SYBR green PCR kit; Qiagen, Hilden, Germany, http://www1.qiagen.com/SelectCountry.aspx). The relative initial amount of mRNA of a particular gene was extrapolated from a standard curve. For standard curve determination, we used a pool of all the samples (eight samples), serially diluted in four log2 steps and run in parallel to the samples. The total volume of each reaction was 20 µl, containing 300 nM of each forward and reverse primer and 125 ng of cDNA. Appropriate negative controls were run for each reaction. PCR amplification included a first step of 10 minutes at 95°C of denaturation, followed by 45 cycles of amplification at 95°C for 15 seconds and 60°C for 60 seconds. All of the reactions were performed in triplicate. The relative amounts obtained were normalized to the housekeeping gene beta-actin, whose level of expression was not changed significantly according to the microarray data (data not shown). The ratio was calculated by dividing each gene expression signal by that of the internal standard (beta-actin) in each sample. The sequences of the primers are as follows: FZD6 (forward primer), 5'-TCCCGCAGTATAGAACATTCCA-3'; FZD6 (reverse primer), 5'-GCCAGGCCAGTGTCAGTAATATC-3' ; HOXA9 (forward primer), 5'-CGGTGATTTAGGTAGTTTCCTGTTG-3' ; H0XA9 (reverse primer), 5'-GTAATGAAGGCAGTTCGTGCTG-3' ; FLT3 (forward primer), 5'-GATGCAGAAGAAGCGATGTATCA-3'; FLT3 (reverse primer), 5'-AGGTGTGAGGACATTCCGAAAC-3'; MLLT3 (forward primer), 5'-GGTAGAGCTTCACAGAAGGTTAATGA-3'; MLLT3 (reverse primer), 5'-AAGGTTCACGATCTGCTGCAG-3'; TIE1(forward primer), 5'-AGCCAGGAAGGCCTATGTGA-3'; TIE1 (reverse primer), 5'-ATCAATGCCCGCGTAAGTG-3'; HLF (forward primer), 5'-TGATCAAGAAAGCTCGCAAAGTC-3'; HLF(reverse primer), 5'-CCCAGTACTTGTCATCCTTCAGG-3'; SPINK2 (forward primer), 5'-AGAACGCCAAACTGCTCTCAG-3'; SPINK2 (reverse primer), 5'-AAAGTGTCTGGGACATCCTGGTA-3'; MEIS1 (forward primer), 5'-GCCCGGAGAAGAATAGTGCA-3'; MEIS1(reverse primer),5'-TTGATTGCCTGCTCGGTTG-3'; beta-actin (forward primer), 5'-CCTGGCACCCAGCACAAT-3'; beta-actin (reverse primer), 5'-GCCGATCCACACGGAGTACT-3';
Comparisons with Other Stem Cell Profiles We compared the HSC populations studied by Ivanova et al. [2], Ramalho-Santos et al. [3], and Georgantas et al. [9] with the HSCs studied by us. Ivanova et al. [2] compared murine bone marrow (BM) and fetal liver HSC populations with the differentiated cell population data. Valid genes were determined as described above, resulting in a list of 18,334 PSs. A total of 5,410 PSs (supplemental online Table 3) met the criteria described above for active genes in at least one of the comparisons. Of these PSs, 3,365 are represented as human orthologs on the HU133A array. This list describes 1,905 genes. Ramalho-Santos et al. [3] described 1,977 PSs overexpressed in murine BM HSCs compared with the differentiated cells. Of these PSs, 1,863 are represented as human orthologs on the HU133A array. This list describes 922 genes.
Georgantas et al. [9] described 214 upregulated PSs in human CD34+CD38– cells from CB and PB HSCs. The HSC lists from these three studies were compared with the 244 upregulated genes obtained from our data (supplemental online Table 3). Georgantas et al. [9] described 605 and 889 upregulated PSs in PB and CB, respectively; 291 and 603 of these were represented on the Hu133A array used in this study. These lists describe 276 and 562 genes. The upregulated genes from the two different tissues were compared with the 218 and 304 upregulated genes obtained from our data for PB and CB, respectively (supplemental online Table 3).
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An unsupervised hierarchical clustering of 14,025 valid PSs out of the total 22,215 PSs present on the microarray (Fig. 1A
) showed a clear distinction between the CD133-expressing and -nonexpressing cells, each type clustered together on different sides of the tree. The comparison between the CD133+ samples and CD133– samples identified a list of 584 differentially expressed PSs, of which 297 were upregulated and 287 were downregulated (Fig. 1C, supplemental online Table 2). CB CD133+ cells were compared with CB CD133– cells, and 1,689 PSs were identified as differentially expressed by at least twofold. Of these PSs, 789 (672 genes) were upregulated and 900 (708 genes) were downregulated (Fig. 1D, supplemental online Table 2). PB CD133+ cells were compared with PB CD133– cells. A total of 1,336 PSs differentiate between the two populations by at least twofold; 690 (596 genes) PSs were upregulated and 646 (495 genes) were downregulated (Fig. 1D, supplemental online Table 2).
|
Of these genes, a list of 132 that were upregulated by at least five are presented in Table 1. The genes common to Ivanova et al. [2], Ramalho-Santos et al. [3], and Georgantas et al. [9] are marked in specific lanes.
|
, supplemental online Tables 4B, 4C).
To further validate the hybridization results, QRT-PCR analysis was performed on eight genes selected from the 244 stemness genes. A good correlation between the two methods was observed, as shown in Figure 2 and supplemental online Table 6.
|
, Tables 2 and 3, supplemental online Table 3).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have found a set of genes that are specific to CD133+ cells from each source (supplemental online Table 10) and a set of genes that are common to both sources (Table 1 and supplemental online Table 5).
In contrast to previous studies that used CD34+ cells for the study of HSCs, we used CD133+ cells because they may represent more primitive cells [10].
The selection process resulted in an enriched population of CD133+ cells (85% purity, which is not very high, especially for these types of expression assays). Nevertheless, an indirect evidence for the efficacy of this method to isolate the real HSCs is suggested by the over-representation of genes associated with the immune system in the CD133– fraction representing differentiated cells (supplemental online Tables 7 and 8, supplemental online Fig. 1) compared with the CD133+ cells.
Two hundred forty-four genes were upregulated by at least twofold in the gene expression profiles of HSCs from CB and mobilized PB compared with the differentiated cells in our study. We compared this list of genes to HSC-specific gene lists provided by Ivanova et al. [2] and Ramalho-Santos et al. [3], who performed a similar analysis mainly in mice, and found 65 and 33 common genes, respectively. A comparison with the list of genes provided by Georgantas et al. [9], who studied human HSCs, yielded 24 common stemness genes (Fig. 3).
Among the common, probably more conserved genes that were identified in the above three studies and our study, we focused on two functional groups that contain 16 genes that have a role in cell development and in cell growth and maintenance. These groups were found to be over-represented (supplemental online Fig. 2, supplemental online Table 9) in relation to other functional groups. Some of these genes also have characteristics of growth factor receptors (receptor tyrosine kinase and c-mpl), and some are transcription factors, among them homeobox genes and transforming growth factor (TGF)-ß–targeted genes. These 16 genes have an important role in hematopoiesis as well as in the development of hematopoietic malignancies.
Growth Factor Receptors
C-kit and flt3, which were both highly expressed in HSCs, are involved in the early stages of normal hematopoiesis. Mutations in these receptors have been linked to several hematological malignancies. Flt3 is the most commonly mutated gene in myeloid leukemia, whereas c-kit mutations are strongly linked to the development of mast cells malignancies, acute leukemia, and myelodysplastic syndrome (MDS) [18]. Tie-1 has been shown to play an important role in angiogenesis and hematopoiesis. Over-expression of Tie-1 was documented in acute myeloid leukemia (AML) and MDS and is linked to increased vascularity noticed in the BM of patients with hematological malignancies [18, 19].
C-mpl, the thrombopoietin receptor, has a known physiological role in HSC production and function[4]. Recent studies showed that its expression in AML blasts is indicative of a relatively poor prognosis [20].
Transcription Factors
The homeobox genes encode proteins that play an important role in the regulation of normal hematopoiesis. They regulate genes that are involved in differentiation, self-renewal, and proliferation of HSCs. They are expressed in early progenitors and are down-regulated during differentiation [21, 22]. Three homeobox genes (HOXA9, HOXA5, and HOXA10) were significantly upregulated in CD133-positive cells in our analysis. In addition to the role of HOXA9 in normal hematopoiesis, it was found to be involved in leukomogenesis as a direct target of the mixed-lineage leukemia gene (MLL). Binding of the HOXA9 and HOXA7 gene products, together with upregulation of the HOX cofactor Meis 1, leads to leukemic transformation [23]. Both MLL and MEIS1 were found to be upregulated in HSCs in our study. T-cell acute lymphoblastic leukemia (ALL) with MLL rearrangement consistently demonstrated overexpression of HOXA9, HOXA10, HOXC6, and the Meis 1 HOX coregulator [24]. HOXA5 is an important regulator of lineage commitment and has a role in the molecular mechanism that directs a multipotent HSC toward myeloid differentiation [25]. A study of defined subsets of AML found overexpression of HOX genes in poor prognosis cases. There was a correlation between event-free survival and expression of several HOX genes, including HOXA5, suggesting that dysregulation of these key hematopoietic genes is important in the process of leukomogenesis and is correlated with prognosis [26].
HOXA10 is expressed in primitive HSCs and has a role in their normal development, whereas its expression is downregulated during differentiation. Overexpression of HOXA10 leads to impaired differentiation and increased proliferation of hematopoietic progenitors and, together with other cofactors, can cause malignant transformation [27].
Another gene of interest encoding a transcription factor is GATA-2, which plays a crucial role in transcriptional regulation of early hematopoiesis by blocking differentiation and allowing self-renewal [28]. Recent studies showed that GATA-2 interaction with promyelocytic leukemia zinc finger (PLZF) might have a role in the pathogenesis of APL, whereas insertional mutation in GATA-2 may play a role in AML [29, 30].
TGF-ß plays an important role in regulating the balance between proliferation and differentiation in hematopoietic cells and is a potent inhibitor of cell-cycle progression and maintains HSC quiescence [31].
Two of TGF-ß target genes were upregulated in our stemness gene sets: TGIF2 (TGF-ß–induced factor 2), a member of homeo-domain factors that plays a role in directing cellular proliferation and differentiation [32], and TSC22 (TGF-ß–stimulated gene), which encodes a transcription factor expressed at varying levels in all fetal and adult tissues [33]. Although the precise role of these two genes in hematopoiesis and leukomogenesis is unknown, a recent study suggested an important role of TGF-ß pathway in the pathogenesis of lymphoid malignancy. Loss of Smad3 protein, one of the cytoplasmic intermediates involved in the transduction of signals in the activation process of TGF-ß, is a specific feature of childhood T-cell ALL [34].
Another group includes genes that are associated with cell development. Hepatic leukemia factor (HLF) is a transcription factor that has been linked to malignancies of the lymphoid system. The chromosomal translocation t(17; 19) in early B-lineage acute leukemia results in the fusion protein E2A-HLF, and the ectopic expression of HLF promotes malignant transformation by interfering with apoptosis [35]. Overexpression of the ectropic viral integration site 1 (EVI1) gene is a consistent feature of the 3q21q26 syndrome, an association of myeloid leukemia/myelodysplastic syndrome with a specific chromosomal rearrangement such as t(3; 3)(q21q26) or inv(3)(q21q26) [36].
The high mobility group AT-hook 2 (HMGA2) transcription factor was found to be involved in ALL with a t(9; 12)(p22q14). The rearrangement was associated with overexpression of an HMGA2 mRNA, suggesting its contribution to the malignant process in leukemia with 12q translocation [37].
The fourth group of genes is associated with cell growth and maintenance. MYCN plays a crucial role in transcriptional regulation of early hematopoiesis by blocking differentiation, and its downregulation leads to progress through differentiation [38]. Its role in leukemogenesis was suggested by McElwaine et al. [39]. The MYB gene located on chromosome 6q21–q23 encodes proteins that are critical for hematopoietic cell proliferation and development. Deletions of the long arm of chromosome 6 (6q-) are frequently found in hematopoietic malignancies, including lymphoblastic leukemias, non-Hodgkin’s lymphomas, and myeloid leukemias [40]. Also, MYB was found to be one of the genes that synergies with CBFB-MYH11 in the pathogenesis of AML [41]. The FHL1 (four and a half LIM domain 1) gene is overexpressed in MLL knockout cells and in human t(4; 11) cell line, suggesting an important role of this gene in leukomogenesis [42].
CB and mobilized PB show considerable differences in their biological properties. We identified genes that are differentially expressed in CB and PB CD133+ cells.
Two hundred eighteen genes were upregulated in the PB CD133+ cells (supplemental online Table 10B), and 304 genes were upregulated in the CB CD133+ cells (supplemental online Table 10A). Specific functional categories were significantly over-represented in CB and PB. The first over-represented group in PB includes genes that are associated with cell growth and maintenance (51 genes); the second belongs to nucleic acid metabolism (48 genes), and the third to protein metabolism (26 genes) (supplemental online Table 10B). The first group in CB belongs to the nucleobase, nucleoside, nucleotide, and nucleic acid metabolism (74 genes). The second group includes genes that are associated with cell growth and maintenance (66 genes), and the third group includes genes that belong to protein metabolism (38 genes) (supplemental online Table 10A).
The differences in gene expression profile between CB and PB can help in identifying the molecular mechanisms underlying the functional and biological differences of these two sources of HSCs.
Five other groups analyzed the differential gene expression between human HSCs of different sources. Bhatia et al. [8] analyzed differential gene expression of HSCs derived from early stages of in utero human hematopoiesis compared with adult mobilized peripheral blood. Steidl et al. [5] and Graf et al. [6] studied the genetic expression profiles of BM HSCs compared with PB HSCs. Ng et al. [7] compared the gene expression profiles of HSCs derived from BM, PB, and CB.
All of these studies based their isolation process on the expression of CD34, as opposed to our study, which used CD133 as the selection marker. In addition, different microarrays containing smaller numbers of PSs were used in the three other studies [5–7], whereas suppressive subtractive hybridization was used in the fourth study [8]. Most important is that while the other studies compare the expression profile of HSCs from various sources with each other, we compared the HSCs from each source to their corresponding differentiated CD133– cells in the same source. An analysis that is based on the comparison of one source of HSCs to the other and not to differentiated cells may lack stem cell specificity because it may include genes that are over-represented in a certain tissue and not specifically by the stem cells of this tissue. The strategy used by us as well as by Ivanova et al. [2] and Ramalho-Santos et al. [3], who studied mainly the murine HSCs, identifies the genes most probably characteristic of the true HSCs. One potential weakness must be acknowledged when using a heterogeneous differentiated CD133– as a reference population. This heterogeneity could complicate the interpretation of the patterns of gene expression. For example, T cells might differentially upregulate the same genes identified with the CD133+ cells, but monocytes might downregulate those same genes. Depending on the proportion of those cells in the CD133– fraction, the clustering algorithms could identify those genes as stem cell or not.
Very recently, Georgantas et al. [9] published the results of gene expression analysis in three different sources of human HSCs. The stem cell population analyzed by this group seems to be ideal (CD34+/CD38–/Lin–) because it is considered to represent the more primitive/true stem cells. They also used the same Affymetrix microarray that was used by us and compared the HSCs to the more differentiated population as done by others [2, 3] and us. Twenty-four genes that were upregulated in the CD34+/CD38–/Lin– cells were also upregulated in our study. This is a surprisingly low overlap. One must bear in mind three possible limitations of this study. First, as mentioned before, there might be an important HSC population within the CD34–/CD133+ cells. Second, the stringent purification of HSCs yielded only a minimal amount of RNA for hybridization to the microarrays. This necessitated amplification of the RNA, which may result in false representation of some of the analyzed genes. Third, in an attempt to define the genes relevant to true human HSCs, the authors in this study used a reference cell population that was more differentiated than the stem cell fraction (CD34+/38–/Lin– vs. CD34+/38+/Lin+) but remained within the same stem/progenitor compartment. By necessity, the authors should have discarded several genes that might be important for stem/progenitor cells and are significantly downregulated in differentiated cells but expressed at rather even levels throughout the stem/progenitor compartment, including the reference population. Many of these discarded genes may be present among the stemness genes of our work, which used less-stringent criteria for defining stemness and a totally differentiated reference cell population.
The leukemic cell shares a well-known HSC characteristic: the capacity for unlimited self-renewal. Recent studies suggest that the leukemic cell originates from a normal self-renewing HSC by accumulating mutations that lead to leukemic transformation [43, 44]. We showed that genes that encode growth factor receptors and transcription factors, which are involved in cell development or cell growth and have an important role in normal hematopoiesis, are upregulated in the stemness genes. These genes are known to be overexpressed or mutated in hematological malignancies. This is suggestive of a pivotal role of these genes; on the one hand, they control normal hematopoiesis, but when mutated, overexpressed, or underexpressed, they have the potential to direct the HSCs toward malignancy. This suggests that stem cell genes are potential leukemic genes and also supports the existence of a leukemic stem cell. Interestingly, the recent finding that AC133+ cells were present in various malignancies like brain tumors and leukemias supports this contention and characterizes the so-called malignant stem cell [45–47].
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Langer, M. D. Radmacher, A. S. Ruppert, S. P. Whitman, P. Paschka, K. Mrozek, C. D. Baldus, T. Vukosavljevic, C.-G. Liu, M. E. Ross, et al. High BAALC expression associates with other molecular prognostic markers, poor outcome, and a distinct gene-expression signature in cytogenetically normal patients younger than 60 years with acute myeloid leukemia: a Cancer and Leukemia Group B (CALGB) study Blood, June 1, 2008; 111(11): 5371 - 5379. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
