Significant changes in the composition of the precursor B-cell compartment in children less than 2 years old


  • How to cite this article: Piatosa B, Birbach M, Siewiera K, Ussowicz M, Kalwak K, Drabko K, Rekawek A, Tkaczyk K, Kurowski PN. Significant Changes in the Composition of the Precursor B-Cell Compartment in Children Less Than 2 Years Old. Cytometry Part B 2013; 84B: 179–186.


Background: Defects in early B lymphocyte maturation in bone marrow (BM) compose a characteristic feature of many primary immune deficiencies associated with agammaglobulinemia. To date, only limited data on the composition of the precursor B-cell compartment in BM is available. The aim of this study was to define normal age-related ranges of total B-cell content and distribution of precursor B—cell stages in BM for the future use in clinical diagnostics.

Methods: Four color flow cytometry was used to analyze the composition of the B-cell compartment in specimens from 59 hematologically healthy children, aged 14 days to 16 years, assigned to six age groups: neonates less than 1 month old, infants >1–12 months old, children >1–2 years old, >2–5 years old, >5–10 years old, and older than 10 years.

Results: Analysis of the composition of the B-cell compartment revealed significant age-related variation in the distribution of individual B-cell maturation stages, most seriously affecting children during first 2 years of life, with the shift from domination of the earliest stages, to gradually increasing content of mature B-cells. Significantly higher proportions of pro-B lymphocytes were observed in neonates than in any other age group.

Conclusion: Physiological age-related variation in the precursor B-cell compartment composition affects most seriously very young children below the age of 2 years. Proper interpretation of immunophenotyping results performed in cases of suspected early B-cell differentiation defect requires application of adequate reference data. © 2013 International Clinical Cytometry Society

Development of human B-lymphocytes is characterized by progression through a series of checkpoints defined primarily by rearrangement and expression of immunoglobulin genes (1). The maturation process, accompanied by changes in cell immunophenotype, is initiated in bone marrow (BM) and ends in peripheral lymphoid organs, when the ability to produce full antibody repertoire is achieved (2–7).

Antibody deficiencies (AD) comprise the largest group of primary immunodeficiencies (PID). Patients may present at various ages with an increased susceptibility to infections due to reduced or absent serum immunoglobulins (8). A minor proportion of AD is caused by defects in early B lymphocyte maturation. An increasing awareness among the medical community of consequences of untreated defects leads to referral of very young children for diagnostics. According to the European Society of Immune Deficiencies, definite diagnosis of a primary immune deficiency is based on detection of a disease-causing gene mutation. It is, however, time-consuming and cost-ineffective to sequence all potentially mutated genes producing similar clinical phenotype. Instead, a common approach is to search first for a potentially defective protein and/or B—cell maturation defect (9). Interpretation of the precursor B-cell compartment analysis performed for diagnosis of suspected early B-cell maturation defects requires reference to normal age-related data, as the composition of progenitor cells in BM significantly differs between childhood and adulthood (2,10–15).

The aim of this study was to describe the physiological changes in the B lymphocyte compartment occurring during early childhood and to define normal age-related ranges of the B—cell content, as well as individual precursor B—cell compartment composition in BM for future use in clinical diagnostics. The strategy for bone marrow analysis was adapted with minor changes from the study by Noordzij et al. (14).


Patient Cohort

The study included 59 hematologically healthy children; among them 45 children, aged 14 days to 10 years who underwent open cardiac surgery in the Department of Cardiac Surgery, Children's Memorial Health Institute for the correction of congenital cardiac defects. After sternotomy, a BM biopsy was taken with a surgical spoon and immediately transferred to heparinized tube (Vacutainer, Becton Dickinson) with 2 ml of physiological saline. Only patients with normal blood differential, with no evidence of infections or other organ dysfunctions were considered eligible. Fourteen other children aged 1.0–16.4 years were qualified as BM donors in bone marrow transplantation units in Wrocław or Lublin. BM was harvested from iliac crests under general anesthesia, according to local standard operating procedure. About 1 ml sample was transferred to heparinized tube (Vacutainer, Becton Dickinson) and sent by overnight courier service to laboratory performing flow analysis. All samples were harvested after obtaining a written consent from parents of donors and donors themselves, if older than 16 years.

The study was approved by an institutional review board at the Children's Memorial Health Institute, Warsaw, Poland and conducted in accordance with Helsinki Declaration.

Flow Cytometry

Specimens harvested from sternum were vortexed vigorously, and then filtered through a nylon mesh to eliminate any residual debris. The obtained cell suspensions, as well as specimens from iliac crest, were rinsed four times with phosphate buffered saline (PBS) supplemented with 2% bovine serum albumin (BSA) to eliminate any cell bound immunoglobulins, and 0.1% sodium azide to prevent internalization of extracellular markers. 100 μl aliquots of cell suspension (5 × 106/ml) were incubated for 15 min at room temperature with combinations of optimally titrated monoclonal antibodies used for staining of membrane bound antigens (Table 1). After erythrocyte lysis with BD FACS Lysing solution (Becton Dickinson), performed according to manufacturer's recommendations, cell suspensions were washed with PBS supplemented with 0.1% sodium azide. Further processing depended on type of labeling. Samples stained only for extracellular antigens were analyzed directly by flow cytometry. Cell labelings involving intracellular staining of CD79a, Igμ, VpreB (CD179a), and terminal deoxynucleotidyl transferase (TdT) were first stained for extracellular antigens. After membrane permeabilization with BDPerm II (Becton Dickinson) cells were washed once with PBS supplemented with 0.5% bovine serum albumin (BSA) and 0.1% sodium azide. Subsequent intracellular staining was performed at 4°C for 30 min. Cells were ready for acquisition after additional washing with PBS supplemented with 0.5% BSA and 0.1% sodium azide. At least 50,000 events from the lymphocyte gate defined by low forward and side scatter (tubes 1 and 2) and 5,000 events from the B-cell gate defined by high expression of CD19 and low side scatter (tubes 3–8) were acquired to four color FACS Calibur flow cytometer (Becton Dickinson), equipped with blue laser (488 nm/15 mW) and red diode (635 nm). The cytometer was routinely calibrated monthly with CaliBRITE (Becton Dickinson, cat. no. 340486) and CaliBRITE APC beads (Becton Dickinson, cat. no. 340487). All samples were acquired as list mode files (format FCS 2.0) using Cellquest software v.3.3 and standard instrument settings as supplied by the manufacturer. Lyse-wash settings were used, without any custom modifications.

Table 1. Combinations of Monoclonal Antibodies Used for Quadruple Immunostaining to Describe the Precursor B-Cell Compartment and the Lymph Gate in Bone Marrow of Healthy Children
  1. “c” before the name of monoclonal antibody denominates intracellular staining. Tubes 1 and 2 was used to analyze the composition of the lymphocyte gate defined on cells with low forward and side scatter. Tubes3–9 were used to define the composition of the B cell compartment.


Statistical Analysis

Mann-Whitney test was used to evaluate statistical differences between age groups.


Considering age-related variation in B lymphocyte content and composition in peripheral blood (PB) (16), patients were divided into six age groups: (1) neonates less than 1 month old, (2) infants >1–12 months old, (3) children >1–2 years old, (4) >2–5 years old, (5) >5–10 years old, and (6) older than 10 years. Results of the analyzed parameters in specimens harvested from healthy BM donors and patients submitted to cardiac procedures were found to be comparable for patients2–5 and5–10 years old; therefore, we assumed that further analysis might be carried in groups assigned depending on patient's age, not the source of material (see Supporting Information Fig. S1).

To describe the composition of lymphocyte gate defined by low side and forward scatter, we analyzed the age-related distribution of CD22+ B lymphocytes, CD19+ B lymphocytes, CD3+ T lymphocytes, CD3CD16.56+ NK cells, as well as contamination of the lymphocyte gate with myeloid CD13.33+ and erythroid CD45CD71+ cells (Fig. 1). We found, that similar to PB, the composition of lymphocyte gate demonstrated significant age-related changes in B lymphocyte content (Table 2). This difference was most prominent between neonates and infants, and referred both to B lymphocytes defined by CD22, as well as CD19 expression. Apart from the youngest children less than 1 month old, who had significantly less B lymphocytes than any other patient group, gradual reduction in B-cell content with age was observed. The proportion of T lymphocytes increased gradually with age, while no significant age-related changes within the NK cell population were found. The myeloid contamination of the lymphocyte gate, although significantly varying between age groups during first 2 years of life, remained low in all analyzed groups (Table 2). Erythroid contamination was significantly greater in neonates than in children under 2 years, but above that age the contamination of the lymphocyte gate with erythroid cells remained almost invariable.

Table 2. Age Related Changes in Composition of Lymphocyte Gate
Age groupPopulation NCD22+CD19+CD3+CD16.56+CD3-CD13.33+CD71+CD45-
  • Median and 9–95 percentile ranges are presented. Significant difference between the analysed and the younger age group is marked by asterisk:

  • *

    P < 0.05,

  • **

    P < 0.01.

<1 month69.
>1–12 months562.7**64.3***26.8**
>12 months to 2 years744.5**37.712.3*3.16.9**31.2**
>2–5 years2134.0*31.2**38.6
>5–10 years1233.928.521.3*
>10 years826.3*19.1**

Gating sequence for the definition of B-cell maturation stages is presented on Figure 1. Considering that basophils demonstrate both CD22 and CD13+ (17), the population of B lymphocytes was defined as CD22-positive cells with low side scatter (2, 14) adjusted to minimize the contamination with basophils (Fig. 1). Moreover, we considered as pro-B only cells with positive expression of CD34, but lacking expression of CD19 (Fig. 1). CD19+ B lymphocytes were defined as cells with positive expression of CD19 and low side scatter. Differential expression of surface CD19, CD10, CD20, IgM, IgD, and intracellular TdT and IgM allowed distinction of pro-B, pre-B-I, pre-B-II, and immature, and mature B-lymphocytes (Fig. 1, Table 3) (3, 14). More detailed description of precursor B-cell stages included differential expression of intracellular CD79a, TDT, CD179 (V-pre-B), and IgM (Table 3). Pro-B were defined as CD22+CD34+CD19, pre-B-I as CD19+CD34+TdT+, pre-B-II as CD19+cIgM+sIgM, immature B as CD19+IgM+IgD, while mature B lymphocytes demonstrated CD19+IgM+IgD+ phenotype (14, 18) (Fig. 1). Detailed data on the calculation strategy are included in supplementary material.

Figure 1.

A. Gating strategy in BM. (1) The lymphocyte gate was defined based on forward and side scatter characteristics, lymphocyte gate was composed of (2) B lymphocytes, (3) T lymphocytes, and (4) NK cells. B. The gate was contaminated by a varying proportion of cells from (5) erythroid and (6) myeloid lineages. (7) Total B cell pool was defined on cells expressing CD22 and low scatter characteristics. (8) The gate was adjusted to minimize contamination with myeloid cells. (9) Pro-B lymphocytes were characterized by CD22+CD34+CD19 phenotype, and could be subdivided into three stages: Stage 1 defined as: (10) CD22+CD34+CD79, (11) Stage 3 defined as CD22+CD19TdT+, and Stage 2, which was calculated as a difference between the total pro-B cell pool and sum of Stages 1 and 3. C. (12) CD19+ lymphocytes were described within a gate defined by positive expression of CD19 and low side scatter. Pre-BI lymphocytes were characterized by expression of (13) TdT and (14) CD34. Within pre-BI, Stage 4 was defined as cells with (15) high expression of CD10, negative expression of CD20 and (16) positive expression of CD34, while Stage 5 was calculated as the difference between total pre-BI and Stage 4. (17) Pre-BII were defined as CD19+ B lymphocytes with positive expression of cytoplasmic μ heavy chain, but negative expression of surface IgM. (18) Stage 6 of pre-BII differed from Stage 7 by high expression of cytoplasmic CD179a. The population (20) defined by positive expression of CD10 and CD20 contained cells from Stage 7 of pre-BII and immature B lymphocytes. Immature B lymphocytes (21, 23) differed from mature B (19, 22) by positive expression of CD10 and lack of surface IgD.

Table 3. Distribution of Precursor B (CD22+) Cell Stages in Bone Marrow of Healthy Children
  1. Essential phenotypic differences and median and 5–95 percentile ranges are presented, including significant differences between the analyzed and younger age group marked by asterisk:

  2. a

    P < 0.05,

  3. b

    P < 0.01. “c” and “s” are used to discriminate between intracellular and surface expression of IgM.

<1 month626.
>1–12 months51.0***11.430.4**40.6*
0.6−5.91.4−4.30.3−2.72.8−4.41.8−9.09.0 16.823.2−34.4
>1–2 years74.7**4.923.8*16.7*33.8
>2–5 years214.
>5–10 years125.
>10 years84.
CD10   ++++++
CD179a   ++++

Analysis of the composition of the B-cell compartment within the total pool of CD22+ B lymphocytes revealed significant age-related variation in the distribution of individual maturation stages. During first 2 years of life significant changes in the proportions of pro-B, pre-B-II and immature B stages, and gradually increasing with age proportion of mature B lymphocytes within the CD22+ pool were observed. The most important changes occurred at the transition from neonates to infants (Fig. 2).

Figure 2.

Age-related distribution of B cell maturation stages within the whole B cell compartment (top panel) and the precursor B cell compartment (bottom panel).

Considering that mature B lymphocytes may compose a significant proportion of the total B-cell population, due to either PB contamination during harvesting procedure or physiological recirculation of mature cells (11, 19), we have excluded them from our calculations to analyze the composition of the precursor B-cell compartment. During the first 2 years of life, significant differences in the distribution of major B lymphocyte subsets in BM were observed (Table 3). The proportion of pro-B lymphocytes appeared to be highly variable in early life, accounting for median 36.3% of the whole precursor B-cell compartment in neonates, significantly more than among infants 1—12 months old (median 5.2%, P = 0.0303). This proportion significantly changed in children 1—2 years old (median 11.8%, P = 0.0479) and remained almost invariable in older children. The variability of the pro-B cell compartment during the first 2 years of life depended mostly on Stage 1 (CD22+CD79aCD34+CD19TdT). The proportion of pre-B-I lymphocytes was almost stable except for infants, who had significantly less cells from this population than neonates or older children. Both pre-B-II and immature B-lymphocytes were significantly less numerous in neonates than in infants (median 24.1 vs. 40.5%, P = 0.0086 and median 13.6 vs. 40.6%, P = 0.0303, respectively). Only minor differences in the distribution of individual B-cell maturation stages were observed in children older than 2 years (Table 3, Fig. 2).


Over the last 20 years, multiparameter flow cytometry became an essential tool for the diagnosis and monitoring the response to therapy in a wide spectrum of diseases, including primary immunodeficiencies and chronic lymphoproliferative disorders (20–22). By evaluation of presence and absence of specific antigens, flow cytometry allows identification of cells from different cell lineages, determination of their maturation stage, differentiation of normal and abnormal cell populations, and provides other information that may be of prognostic value (20).

Primary AD comprise the largest group of inherited disorders of the immune system. Several genetic defects were identified as affecting differentiation at various B-cell stages or cooperation of B lymphocytes with other cell types, although in many cases the underlying mechanism is not yet known. The maturation block in normal precursor B-cell differentiation was found in X-linked agammaglobulinemia (XLA) due to mutations in BTK (14, 23) and in autosomal recessive agammaglobulinemia due to mutations in IGHM (24), CD79A (25), BLNK (26), L14.1 (27), or CD79B (28). Despite lack of a defined gene defect, a proportion of patients with common variable immune deficiency (CVID) demonstrate a similar block in B-cell development as observed in XLA (29). Besides the T cell defect, patients with severe combined immune deficiency (SCID) due to mutations in RAG or DCLRE1C, formerly known as Artemis, demonstrate a complete block at the transition from pre-B-I to pre-B-II stage (30, 31). Replacement therapy, necessary to restore the ability to resolve recurrent bacterial infections in patients with AD (32), is insufficient for patients with SCID, who generally die within the first 2 years of life unless treated with hematopoietic stem cell transplantation or therapy that replaces the function of the affected gene (33). The clinical outcome significantly depends on fast diagnosis based on combined results of clinical and laboratory evaluation.

There are no generally accepted reference ranges describing normal distribution of precursor B lymphocytes, as BM examinations are usually not performed in healthy people. Few earlier studies uniformly reported that the number of precursor B lymphocytes is significantly higher among children than adults, and that their number declines with age (2, 10, 12). It has been emphasized, that special caution should be taken as far as children under 4 years of age are concerned, because significantly higher percentage of immature B-cells may suggest the presence of B-lineage malignancy (12). Recently, it was found that major changes in the B-cell compartment occur during first 2 years of life (15). Considering that most children referred for diagnostics of severe conditions associated with lack of ability to produce normal amount of antibodies are less than 2 years of age, it is highly probable that the interpretation of their results of precursor B-cell compartment analysis might be inadequate according to the currently available data (12, 14, 15). We therefore analyzed the composition of B lymphocyte compartment in BM specimens of healthy children with no evidence of hematological disorders, dissected into four groups under the age of five, when most prominent changes in B-cell composition in PB have been observed (16), and additionally in children5–10 and above 10 years old.

A substantial contamination of the lymphocyte gate in all specimens, mostly with erythroid cells observed previously (12, 34), significantly affected the distribution of the lymphocyte subsets within the lymphocyte gate defined by low forward and side scatter characteristics. Within the lymphocyte gate, T and B lymphocytes were present in almost similar percentages among neonates and children above 5 years, as reported by Rego (12), but CD22-positive B-lymphocytes composed about twenty times greater population in infants than T-lymphocytes, and these proportions declined with age.

B lymphocytes and their subsets composed the major lymphoid population in children less than 5 years old, with the percentage decreasing with age (2, 10, 12, 15). These changes correlated with the age-related distribution of B lymphocytes in peripheral blood (15, 16). Greatest changes in the composition of the precursor B-cell compartment were observed at the transition from neonates to infants and later, during the first 2 years of life, with the shift from domination of the earliest stages to gradually increasing content of the mature B-cells. Considering low probability of contamination with PB cells during the harvesting procedure, as most specimens originated from sternum biopsies, this observation brings proof of physiological, age-related accumulation of mature B-cell population in BM.

Antibody responses slowly improve with age (35, 36). Neonatal antibody responses are delayed in onset, reach lower peak levels, are of shorter duration, and differ in distribution of isotypes from the antibody responses observed among older children (36). Apart from the presence of maternal antibodies (36), the reduced antibody responses in human infants might be attributed to significantly lower B-cell counts among neonates than among older children (16), but also significantly lower B-cell content in BM of neonates than of older children, as shown in this study.

B-cell maturation depends on activity of specific transcription factors, as well as age, type of disease and applied therapy (10, 13–15, 18, 37). Physiological changes in the process of B lymphocyte maturation demonstrated in this study correlate with the observation that most children with agammaglobulinemia are referred for diagnosis during the first 2 years of life. They lack or demonstrate very low peripheral B lymphocyte counts, but differ in the precursor B-cell maturation block (9). Although the genetic defect has not been identified in all patients, the immunophenotyping results guide selection of the right candidate gene and help to institute the best therapy (38).

In conclusion, physiological age-related variation in the B-cell content and the precursor B-cell compartment composition affecting most seriously very young children below the age of 2 years indicates that proper interpretation of immunophenotyping among children with suspected early B-cell differentiation defects requires application of adequate reference data. Normal reference values produced as result of this study will significantly facilitate the interpretation of flow data and efficient implementation of the results to management of primary and secondary AD.


The authors are grateful to all children and their parents for consent and participation in the study.