B-cell precursors differentiated from cord blood CD34⁺ cells are more immature than those derived from granulocyte colony-stimulating factor-mobilized peripheral blood CD34⁺ cells

Abstract

Umbilical cord blood (CB) has been widely used instead of bone marrow (BM) and peripheral blood (PB) for stem cell transplantation (SCT). However, problems of sustained immunodeficiency after CB transplantation remain to be resolved. To elucidate the mechanism of immunodeficiency, we compared the characteristics of B cells differentiated in vitro from CD34⁺ cells of CB with those of PB. Purified CD34⁺ cells from CB and PB were cultured on murine stroma cell-line MS-5 with stem cell factor and granulocyte colony-stimulating factor for 6 weeks. The B-cell precursors (pre-B cells) that differentiated in this culture system, were analysed as to their immunoglobulin heavy chain (IgH) variable region gene repertoire and the expression of B-cell differentiation-related genes. CD10⁺ CD19⁺ pre-B cells were differentiated from both PB and CB. Although the usages of IgH gene segments in pre-B cells differentiated from CB and PB were similar, the N region was significantly shorter in CB-derived than PB-derived cells. Productive rearrangements were significantly fewer in cells of CB than PB in the third week. Among a number of B-cell differentiation-related genes, the terminal deoxynucleotidyl transferase (TdT) gene was not expressed in CB-derived cells during the culture. These results indicated that immature features of pre-B cells from CB, such as lack of TdT expression, and a short N region and few productive rearrangements in the IgH gene, might cause the delay in mature B-cell production.

Introduction

Umbilical cord blood (CB), as an alternative to bone marrow (BM) or peripheral blood (PB), has increasingly been used as a source of haematopoietic stem cells for transplantation in the treatment of haematological disorders.^1–5 The advantages of CB transplantation are substantial, and might be based on differences between fetal and adult haematopoietic stem cells. Notably, the relative immaturity of lymphocytes in CB has been thought to reduce the risk and severity of graft-versus-host disease (GVHD), while this immunological property in CB causes a prolonged immunodeficient state after CB transplantation.^6,7 The mechanism of immunological reconstitution after CB transplantation has been examined from a variety of viewpoints in experimental models as well as clinical studies.^8–14 However, attention has mainly focused on the characteristics of T cells. Information on B-cell differentiation from CB stem cells is therefore relatively limited.

In B-cell development, the most immature B-cell precursors, designated pro-B cells, are CD10⁺ CD19⁻ CD22⁻ CD34⁺, and their immunoglobulin heavy chain (IgH) genes remain germline or affect only D–J_H joining. At the next stage, prepre-B/early pre-B cells express CD19, and their IgH genes are rearranged. A hallmark of pre-B cells is the expression of the µ-chain in their cytoplasm (cµ). At this stage, CD34 is no longer present, and CD22 is expressed. Late pre-B cells have rearranged IgL genes, and production of IgL causes the expression of the surface µ-chain (sµ), which is characteristic of B cells.¹⁵ The development of rapid cloning and sequencing techniques has resulted in a substantial accumulation of immunoglobulin variable region gene sequences at various stages of B-cell development, and has revealed stage-specific trends in the usage of V, D and J genes, the degree of N nucleotide addition, and the rate of somatic mutations.^16–27

Recently, a novel long-term culture system using the murine bone marrow stroma cell line, MS-5, in which human CD34⁺ cells differentiate to CD19⁺ cells in combination with stem cell factor (SCF) and granulocyte colony-stimulating factor (G-CSF), has been developed.^28–30 To compare the characteristics of pre-B cells differentiated from CD34⁺ cells of CB with those of PB, we analysed the IgH complementarity-determining region (CDR)-3 gene repertoire and the expression of B-cell differentiation-related genes of cultured cells from CB and PB.

Materials and methods

Cells

Human CB was collected after full-term deliveries with informed consent approved by the Review Board of Tokai Cord Blood Bank. Mononuclear cells (MNC) were separated by Ficoll–Hypaque (Pharmacia LKB, Uppsala, Sweden) density gradient centrifugation after depletion of phagocytes with silica. G-CSF-mobilized PB was obtained from harvesting pools when enough CD34⁺ cells were collected for the patient to receive stem cell transplantation three times after informed consent was obtained. MNC were separated as above. MNC from CB and G-CSF-mobilized PB were incubated with fluorescein isothiocyanate (FITC)-conjugated anti-human CD38 and Cy5-conjugated anti-human CD34 antibodies (Coulter, Hialeah, FL), then subjected to flow cytometer analysis (EPICS ELITE; Coulter). Expression profile of the CD38 antigen in CD34⁺ populations was the same between CB and PB MNC (Fig. 1). CD34⁺ cells from CB and PB were separated from MNC by using Dynabeads M-450 conjugated with anti-CD34 monoclonal antibody and DETACHaBEAD (Dynal, Oslo, Norway) according to the manufacturer's instructions. Each separated aliquot was confirmed to contain over 90% CD34⁺ cells and less than 0·1% CD19⁺ cells by flow cytometry (Fig. 1).

Figure 1.

Flow cytometric analysis of cultured cells from CD34⁺ cells purified from CB and PB. Expression profile of the CD38 antigen in CD34⁺ populations was the same between CB and PB MNC. Both separated aliquots from CB and PB contained over 90% CD34⁺ cells and less than 0·1% CD19⁺ cells. CD10⁺ CD19⁺ pre-B cells were differentiated from both CB and PB.

Cell culture

The murine BM stroma cell line MS-5 was maintained in α-minimum essential medium (α-MEM; Gibco BRL, Gaithersburg, MD) supplemented with 20% horse serum (Nichimen America, Los Angeles, CA). MS-5 cells were prepared at a concentration of 5 × 10⁴ cells/ml/well in a six-well tissue plate (Becton Dickinson, Franklin Lakes, NJ) 1 day prior to the seeding of CD34⁺ cells. CD34⁺ cells from CB and PB were plated at a density of 1 × 10³ cells/ml onto the MS-5 feeders in α-MEM supplemented with 10% fetal calf serum (FCS), 100 ng/ml recombinant human SCF and 10 ng/ml recombinant G-CSF. SCF and G-CSF were generous gifts from Kirin Brewery Co. Ltd. (Tokyo, Japan). All cultures were performed in a humidified incubator containing 5% CO₂ in air at 37°, with a change of medium every week. Non-adherent floating cells were harvested after gentle agitation at every medium change. After 3, 4 and 6 weeks, cells were harvested using Trypsin–ethylenediaminetetraacetic acid (Gibco BRL), and subjected to phenotype analysis and extraction of high molecular weight DNA and total RNA.

Phenotype analysis of cultured cells

Cultured cells were incubated with FITC-conjugated anti-human CD10, phycoerythrin (PE)-conjugated anti-human CD19 and Cy5-conjugated anti-human CD34 antibodies (Coulter, Hialeah, FL), then subjected to flow cytometer analysis.

Analysis of IgH CDR-3 repertoire

High molecular weight DNA was extracted from cultured cells by a standard method. The IgH CDR-3 sequences were amplified by polymerase chain reaction (PCR) using a set of primers corresponding to the consensus sequences of the 3′ ends of the V_H and J_H genes (consensus V_H and consensus J_H, Table 1) as previously reported.^23,26 Amplified products were separated on 8% polyacrylamide gels, and fragments spanning the 50–250 bp region were isolated. The fragments were cloned into pT7Blue T-vector (Novagen, Madison, WI), then transfected into Escherichia coli strain DH5α. Recombinant colonies were randomly selected from PCR-amplified libraries, then plasmid DNA was prepared by using a QIAprep Spin Miniprep Kit (Qiagen Inc, Chatsworth, CA). DNA sequencing was performed on a DNA sequencer (310; Applied Biosystems, Foster City, CA) using a BigDye terminator cycle sequencing kit (Applied Biosystems). The sequence data of each clone were assigned to the V_H, D, J_H and N regions according to the published germline sequences.^31–33 According to the terminology of Kabat et al.³⁴, the position of IgH CDR-3 is defined as beginning from amino acid residues Y95. The frame sequence of the CDR-3 was started at TA(T/C)-T(T/A)C-TGT, corresponding to Y95-Y96-C97 at the 3′ end of the V_H gene segments, then rearranged V_H–D–J_H sequences were determined to be productive or not. In abortive V_H–D–J_H rearrangements, the CDR-3 sequences were started after TA(T/C)T(T/A)CTGT and ended before TGG, corresponding to amino acid residue 103. Statistical analyses were performed by a modified Student's t-test and Fisher's exact test.

Table 1.

Sequences of PCR primers

Primer	Sequence (5′–3′)
consensus VH	GAGGACACGGC(C/T)(G/C)TGTATTACTG
consensus JH	CT(C/T)ACCTGA(G/A)GAGAC(G/A)GTGACC
RAG1 F	CAGCGTTTTGCTGAGCTCCT
RAG1 R	TGGCTTTCCAGAGAGTCCTC
RAG2 F	TTCTTGGCATACCAGCAG
RAG2 R	CTATTTGCTTCTGCACTG
TdT F	CTTCTTTTCCCATAAGTTCATCAC
VpreB F	CATGCTGTTTGTCTACTGCACAG
VpreB R	TGCAGTGGGTTCCATTTCTTCC
λ5F	GGCCGCGGCATGTGTTTGGCAGC
λ5R	ATCGATAGGTCACCGTCAAGATT
MB1 F	CATGCCTGGGGGTCCAGGAGTCCTC
MB1 R	CTCACGGCTTCTCCAGCTGGACATC
PAX5 F	ATGGATTTAGAGAAAAATTATCCGA
PAX5 R	GGATTTTGGCGTTTATATTCAGCGA
IL-7R F	GACAATTCTAGGTACAACTTTTGGC
IL-7R R	GATAGATGACACTCAGGTCAAAAGG
CXCR4 F	CTGAGAAGCATGACGGACAA
CXCR4 R	TGGAGTGTGACAGCTTGGAG
FLT3 F	TGTCGAGCAGTACTCTAAACATG
FLT3 R	CTTTCAGCATTTTGACGGCAACC
Cµ F	ACGAGCAGCGTGGCCGTTGG
Cµ R	GGTGGGACGAAGACGCTCAC
Actin F	TCACTCATGAAGATCCTCA
Actin R	TTCGTGGATGCCACAGGAC

Analysis of the expression of B-cell differentiation-related genes

Total RNA was extracted from cultured cells using a QIAamp RNA Blood Mini Kit (Qiagen). A cDNA was synthesized from each RNA using a random primer and Moloney murine leukaemia virus reverse transcriptase (Super-Script II; Gibco BRL) according to the manufacturer's recommendations. Using this cDNA, we analysed the expression of several B-cell differentiation-related genes (RAG1, RAG2, TdT, VpreB, λ5, MB1, PAX5, IL7-R, CXCR4, FLT3 and IgM) by reverse transcription (RT)-PCR. The primer sequences for PCR are listed in Table 1. We used HALO1 and UOCB1 cell lines, which show the phenotype of early pre-B, as a control. These cell lines were kindly obtained from Dr Kazuma Ohyashiki (Tokyo Medical college, Tokyo, Japan) and Dr Toshiya Inaba (Hiroshima University, Hiroshima, Japan), respectively. Amplified products were subjected to electrophoresis on agarose gel, and stained with ethidium bromide. The intensity of the amplified products was measured by the densitometer. The relative expression level of each product to the CD34⁺ cells from CB was calculated after correction according to each respective expression level of the actin gene.

Immunocytochemistry of TdT and IgM

Cytospin slide preparations were made from cultured cells. Immunostaining was performed as previously described.³⁵ In brief, specimens were completely dried and fixed with periodate–lysine–paraformaldehyde (0·01 m NaIO₄, 0·075 m phosphate buffer, 2% paraformaldehyde, pH 6·2) for 20 min at 4°. After a wash with phosphate-buffered saline, anti-TdT monoclonal antibody (DAKO, Carpinteria, CA) or anti-IgM polyclonal antibody (DAKO) was dropped on the specimens and incubated overnight at 4° for TdT, and for 20 min at room temperature for IgM. For the detection of antibody, a LSAB2 kit (DAKO) was used according to the manufacturer's instructions.

Results

Differentiation of pre-B cells from CB and PB CD34⁺ cells

We examined the expression of CD10, CD19 and CD34 antigens on cultured cells in the 3rd, 4th and 6th weeks. After 3 weeks of culture, cells derived from both CB and PB expressed CD10 and CD19 antigens, but not CD34 antigen. In addition, CD19-expressing cells increased in number during culture for 6 weeks consistent with previous reports. However, more CD19-expressing cells emerged in the culture from PB than CB CD34⁺ cells (Fig. 1 and Table 2). These results indicated that CD34⁺ cells from both CB and PB differentiated into pre-B cells under our culture conditions.

Table 2.

CD10- and CD19-positive cells during the culture

	CD10+/19+ cells (%)
	3rd week	4th week	6th week
CB ex-1	29·9	20·9	27·6
ex-2	20·9	30·7	26·1
PB ex-1	19·9	27·4	35·1
ex-2	31·5	40·2	46·2

Each two independent samples from CB and PB were cultured in vitro, and CD10- and CD19-positive cells were analysed by flow cytometry.

Comparison of the IgH CDR-3 gene repertoires of pre-B cells differentiated from CB and PB CD34⁺ cells

We analysed IgH CDR-3 sequences derived from the DNA of pre-B cells differentiated from CB and PB CD34⁺ cells in the 3rd, 4th and 6th weeks of culture. We amplified IgH CDR-3 genes of each cultured cell type and the purified CD34⁺ cells from CB and PB. Amplified products from cultured cells revealed broad smear bands on the polyacrylamide gel, while those from the purified CD34⁺ cells did not show any bands, indicating that almost all B cells in which the IgH gene was rearranged were eliminated in the purified CD34⁺ cells (data not shown). The smear bands amplified from cultured cells were isolated and cloned in the vector. Randomly selected clones were sequenced. First, we examined whether rearrangements of the IgH genes were potentially productive. In the pre-B cells from CB, 12 out of the 47 rearrangements (25·5%) were in-frame at week 3, this percentage increasing to 40·7% at week 4 and 52·5% at week 6. On the other hand, in the pre-B cells from PB, over half of the rearrangements were already in-frame early in the culture (75·0%, 58·8% and 58·3% in the 3rd, 4th and 6th weeks, respectively). At the 3rd week, the ratio of in-frame rearrangements in the pre-B cells from CB was significantly lower than for those from PB (P = 0·0003 by Fisher's exact test) (Table 3).

Table 3.

The ratio of the productive/non-productive rearrangements in the IgH gene

	3 weeks	4 weeks	6 weeks
CB	12/35* (25·5%)	11/16 (40·7%)	21/19 (52·5%)
PB	15/5* (75·0%)	10/7 (58·8%)	14/10 (58·3%)

^*At the 3rd week, the ratio of in-frame rearrangements in the pre-B cells from CB was significantly lower than PB (P = 0·0003 by Fisher's exact test).

Next, we assigned the IgH sequences to published germline V_H, D and J_H genes, then determined the N region. Usage of D and J_H genes was the same among culture periods and among the sources of CD34⁺ cells (Fig. 2). D3 and D6 family genes were preferentially used both in CB- and PB-derived cells, followed by D1 and D2 family genes. Preferential use of the D7-27 gene was not observed. Of the J_H genes, the J_H4 gene was most frequently used in both cells, followed by the J_H6 gene. This is consistent with the results for B cells of CB as well as for PB B cells. The N region at the V_H–D junction in the pre-B cells from CB and PB measured 6·6 ± 4·2, 9·0 ± 5·1 and 6·5 ± 4·6 in the former, and 10·4 ± 7·3, 9·5 ± 8·0 and 10·2 ± 6·8 in the latter, in the 3rd, 4th and 6th weeks, respectively. At weeks 3 and 6, the N-region was significantly longer in the cells from PB than from CB (P = 0·016 in the 3rd week and P = 0·015 in the 6th week) (Fig. 3). In the 4th week, no significant difference was found, although this might simply be due to the small number of clones analysed. In contrast, there were no significant differences in the lengths of CDR3 and the N region at the D–J_H junction among the cells at any time during the experiment (Fig. 3).

Figure 2.

Usage of the D and J_H genes in pre-B cells from CB and PB CD34⁺ cells. D3 and D6 family genes, and J4 genes were preferentially used during the culture both from CB and PB CD34⁺ cells. ND indicates that D gene was not determined in the rearranged sequences.

Figure 3.

Comparison of the length of N region and CDR-3. The lengths of N region at the V_H–D junction in the pre-B cells from CB and PB were significantly shorter than from PB at the 3rd and 6th week. In contrast, there were no significant differences in the lengths of CDR3 and the N region at the D–J_H junction among them at any cultured periods.

Expression of B-cell differentiation-related genes

We analysed the expression levels of several B-cell differentiation-related genes by RT-PCR (Fig. 4). To compare the expression level of each cultured cell type, we estimated the relative level to the CD34⁺ cells from CB in each gene after the correction according to the level of actin gene (Fig. 5). RAG1 and RAG2 genes were not expressed in the CD34⁺ cells purified from either CB or PB, although they came to be expressed in both cells during culture. TdT was not expressed in the CD34⁺ cells either, although it came to be expressed in cultured cells from PB. Although expression levels of VpreB, λ5 and FLT3 genes of CD34⁺ cells from PB were seen to be relatively higher than those from CB on the agarose gel, the difference was not significant after correction according to the level of actin gene (Fig. 5).

Figure 4.

Expression of B-cell differentiation-related genes. RAG1 and RAG2 genes were not expressed in the purified CD34⁺ cells both from CB and PB, while they came to be expressed in both cultured cells. Although TdT was also not expressed in both CD34⁺ cells, it came to be expressed in cultured cells from PB. However, in cultured cells from CB, its expression was not observed throughout the 6 weeks.

Figure 5.

Relative expression level of B-cell differentiation-related genes. Expression level of each gene was determined by the densitometer, then corrected according to the expression level of actin gene. In each gene, the ratio to the CD34⁺ cells from CB is shown.

Immunocytochemistry of TdT and IgM

Since the difference in the TdT gene expression was most notable, we examined the expression of the TdT product in comparison with that of IgM. As shown in Fig. 6, IgM protein was stained in the cytoplasm (cµ) of cultured cells from both CB and PB. However, cultured cells from CB in the 3rd week did not express cµ protein. This might be because productive rearrangements of the IgH gene were fewer in the CB-derived cells in the 3rd week. TdT was stained in the nuclei of the cultured cells from PB, but not in those from CB, which is consistent with the results of RT-PCR analysis.

Figure 6.

Immunocytochemistry of TdT and IgM. IgM protein was stained in the cytoplasm of both cultured cells from CB and PB. However, all of the cultured cells from CB at the 3rd week did not represent the cµ protein. TdT was stained in the nuclei of the cultured cells from PB, but not in those from CB in accordance with the results by RT-PCR analysis. HALO1 and UOCB1 cell lines were used for positive controls.

Discussion

It has been well established that the murine stroma cell line MS-5 supports the growth of human pre-B cells, and enhances their differentiation from CB CD34⁺ cells in combination with SCF and G-CSF.^29,30 We took advantage of this culture system to analyse differences in the early stage of B-cell development between CB and PB CD34⁺ cells in vitro. Since it has been reported that the expression of CD38 antigen influences the production of B-cell progenitor from CD34⁺ cells in vitro,³⁶ it is important to analyse the expression profiles of CD38 antigen between CB and PB CD34⁺ cells. Although we could not analyse these by using the purified CD34⁺ cells owing to the limited number of purified cells, we confirmed that there was no significant difference in the expression profiles of CD38 antigen in the CD34⁺ populations of CB and PB MNC. We believe, therefore, that the purified CD34⁺ cells held the same profile.

Essentially, the same developmental capacities were demonstrated on phenotypical analysis and the expression of B-cell development-related genes except for the TdT. Therefore, we focused on the IgH gene repertoires in pre-B cells derived from CB and PB CD34⁺ cells. A substantial accumulation of immunoglobulin gene sequences at various stages of B-cell development has revealed stage-specific trends. The human germline D and J_H segments have been completely mapped and sequenced. There are 27 functional D genes and six functional J_H genes.^32,33 In Epstein–Barr virus-transformed fetal B-cell lines, D7-27–J_H1 joining was frequently observed, suggesting that the initial D–J_H joining might frequently involve these genes, as they are located immediately adjacent to each other.³⁷ However, preferential use of the D7-27 gene was not observed, even early in the culture. It has been reported that the D7-27 gene contributes to one-half of the first trimester of fetal life.³⁸ Furthermore, in an analysis of 893 rearranged IgH sequences, the D7-27 gene was found in only 0·5% of the rearrangements.³² The D7-27 gene, therefore, might be essentially involved in the first D–J_H joining. Since the cells derived in our culture system had the pre-B phenotype, secondary or more rearrangements might have already occurred.

From immature to mature B cells, the J_H4 gene is most frequently used in the rearranged IgH genes. It is suggested that recombination signal sequences (RSS) can influence rearrangement frequency. The J_H4 RSS is most closely consistent with the consensus sequence, followed by the RSS of J_H6, J_H5, J_H3, J_H2 and J_H1. This order is parallel to the pattern of the J_H usage in mature B cells.^17,27,38 The present results obtained with cultured cells also support this order.

Rearrangements of IgH genes are highly regulated developmentally.^39–41 During B-cell development, the IgH gene is assembled in two steps. The D and J_H genes are joined first, and then V_H is joined to D–J_H. If V_H–D–J_H joining is in-frame (productive), cµ protein is expressed, followed by sµ protein. After the expression of sµ protein, affinity maturation occurs through the stimulation of antigens. However, in the present in vitro differentiation system, we could exclude the effect of exogenous stimuli on the IgH genes. Therefore, the IgH gene repertoire of pre-B cells generated by this system might simply reflect the recombination machinery. It was especially notable that productive rearrangements were significantly fewer in the pre-B cells from CB than from PB in the 3rd week, though RAG1 and RAG2 genes were equally expressed in both cells. Although the contamination of mature B cells in the purified CD34⁺ cells from PB should be considered, this is less probable because of the results showing that the purified CD34⁺ cells contained less than 0·1% CD19⁺ cells and the PCR-amplification of IgH CDR-3 genes of the purified CD34⁺ cells gave no product. If V_H–D–J_H joining is non-productive, V_H–V_H replacement on the same allele or V_H–D–J_H joining on the other allele would occur to increase the chance of productive rearrangements for further maturation. Since productive rearrangements were increasing to the level in PB-derived cells during the culture, this difference might reflect the small number of recombination trials on the IgH gene of CB-derived cells. Alternatively, a short N region at the V_H–D junction might influence the chance of productive rearrangement (discussed below). However, it is still possible that the effect of SCF and G-CSF on the signal-transduction pathway for B-cell development is weaker in CD34⁺ cells from CB than in those from PB.

Another notable difference in the IgH gene repertoires was that the N region of the V_H–D junction was shorter in CB-derived cells. The addition of this region to the V_H–D and D–J_H junctions of the IgH gene is mediated by TdT, and is developmentally regulated.^42–49 The N region at the D–J_H junction was found in less than 5% of murine fetal liver cells, but in 5–23% of newborn spleen and liver cells and in 64–73% of adult spleen cells.^46–48 In human B cells, the N region at the D–J_H junction was found in 68%, 86% and 91–100% of fetal, neonatal and adult lymphoid tissue, respectively.^44,45 These observations therefore suggest that lack of the N region at the D–J_H junction is representative of the D–J_H recombination occurring during the period of fetal development when TdT activity may have been absent. In our analysis, expression of TdT was not detected in the CB-derived cells during the culture by either RT-PCR or immunostaining, though the N region was observed at both the V_H–D and D–J_H junctions. It has been demonstrated that interleukin-7 (IL-7) and its receptor complex are essential for up-regulating the expression of TdT.⁵⁰ Furthermore, the expression of IL-7 receptor in the CB-derived cells was the same as that in the PB-derived cells by RT-PCR analysis. Therefore, we examined the effect on the expression of TdT of adding 10 ng/ml IL-7 to the culture. However, no gene expression was observed during culture for 6 weeks (data not shown). On the other hand, it has been reported that the expression of TdT was observed in the cultured cells from CB by using the S17 stroma cells.⁵¹ In addition, we could not analyse the phenotypic difference in detail among the CD34⁺ cells from CB and PB. Therefore, it should be considered that the different expression of TdT might depend on the character of stroma cells or the phenotypic difference of the CD34⁺ cells from CB and PB. Recently, IgH gene repertoires in human IgH minilocus transgenic TdT^−/− mice have been reported.⁵² This report demonstrated that most of the rearrangements in TdT^−/− mice lacked the N region, but 3% showed evidence of a short N region even in the absence of TdT. Therefore, there might be a very weak expression of TdT in the cultured cells from CB in our culture conditions or some other mechanism involved, considering that in some cell lines, the N region is added without TdT.^43,44 However, one must take into account the expression of TdT and recombination machinery of the IgH gene, since TdT adds the N region during the joining phase of recombination. That is, alternative explanations for the shorter N region and fewer productive rearrangements in CB-derived cells are possible. First, the small number of recombination trials and a lack of TdT expression could be the cause of the shorter N region. Alternatively, a lack of TdT reduced the chance of productive rearrangements. Which explanation is correct remains to be elucidated; however, our results demonstrated the immaturity of CB CD34⁺ cells for B-cell development and the possibility of an immature B-cell repertoire at the early stage of post-CB transplantation.