FLT3 ligand and not TSLP is the key regulator of IL-7–independent B-1 and B-2 B lymphopoiesis

Human immunodeficiencies have uncovered the key role of cytokine signaling pathways in T lymphocyte development, facilitating development of cell replacement and gene therapies.^1,2  In contrast, little is known about the cytokine pathways that might be perturbed in immunodeficiencies with a B-cell phenotype

In mice, fetal and adult B-cell development has been suggested to be regulated by overlapping yet largely distinct cytokine ligands and receptors.^3-9  Specifically, whereas conventional B-2 B lymphopoiesis in the adult bone marrow (BM) is lost in the absence of only interleukin-7 (IL-7), fetal and early postnatal B lymphopoiesis, in particular B-1 B-cell genesis, seems to be largely driven by IL-7–independent mechanisms.^10-12  Importantly, such IL-7–independent B lymphopoiesis is sufficient to sustain a pool of peripheral B cells and normal immunoglobulin levels through life.^3,13 

Notably, also human B-cell development appears to be largely or entirely IL-7–independent, as X-linked severe combined immunodeficiency caused by a loss of function mutation in the IL-2 receptor gamma chain (IL-2Rg), as well as IL-7 receptor alpha chain (IL-7Rα) deficiency, are accompanied by severely defective T lymphopoiesis but apparently normal B-cell numbers.^14-16  Thus, it could have considerable clinical implications to establish which alternative cytokines might be involved in regulating IL-7–independent B lymphopoiesis.

Several previous lines of evidence strongly implicated thymic stromal lymphopoietin (TSLP) as a primary regulator of IL-7–independent fetal B lymphopoiesis.^8,9  TSLP shares with IL-7 the IL-7Rα as a ligand-binding receptor subunit. However, whereas IL-7–dependent signaling in addition depends on association with the IL-2Rg, shared with multiple other lymphokines,¹⁴  TSLP-mediated signaling rather acts through an IL-2Rg–like TSLP receptor (TSLPR) unit.^17,18  Accordingly, and in contrast to IL-2Rg and IL-7–deficient (Il2rg^−/− and Il7^−/−) mice, IL-7Rα–deficient (Il7r^−/−) mice lack IL-7 as well as TSLP-mediated signaling. One of the major reasons that TSLP has been implicated as an important regulator of IL-7–independent B lymphopoiesis is the 10-fold further reduction in B-cell progenitors and mature B cells observed in adult Il7r^−/− mice compared with Il7-deficient mice.^8,9  However, few direct comparisons of Il7^−/− and Il7r^−/− mice have been made and largely performed on mice of mixed genetic backgrounds.^11,19,20  In support of a unique role of TSLP in embryonic rather than adult B lymphopoiesis, fetal but not adult conventional pro-B cells have been shown to be TSLP responsive,^8,9  and also the residual compartment of B-1 B-cell progenitors in adult BM sustains TSLP responsiveness.⁶  Whereas a lack of an important role of TSLP alone in adult B lymphopoiesis has been established in TSLPR-deficient (Tpte2^−/−) mice,^6,21,22  its proposed role in fetal B-cell development was based on comparative studies in postnatal Il7r- and Il7-deficient mice and differences in TSLP responsiveness between fetal and adult B-cell progenitors.^6,8,9  Notably, more recent and direct analysis of Tpte2 and Il7 double-deficient embryos failed to confirm an important role of TSLP in IL-7–independent fetal B lymphopoiesis.²³  As human IL7R-deficient patients have normal B-cell numbers,^1,24  although B lymphopoiesis in these patients has not been investigated in detail and only in infants, this further questions the importance of TSLP in conventional fetal B lymphopoiesis. However, these findings do not rule out an important role of TSLP in other aspects of IL-7–independent B lymphopoiesis, neither in adult B lymphopoiesis, and in particular not in B-1 B lymphopoiesis which has been demonstrated to be largely IL-7–independent.³  On the contrary, the potential key role of TSLP in IL-7–independent B-cell development has been reinforced²⁵  and therefore remains controversial.

Previous findings have raised but not addressed several key questions relating to the proposed role of TSLP and other cytokines in postnatal and fetal IL-7–independent B lymphopoiesis. First, what, if any, is the role of TSLP in postnatal B-2 lymphopoiesis, as implicated by Il7r^−/− mice having a more severe B-cell phenotype than Il7^−/− mice? Second, as mice double-deficient in Il7 and Tpte2 sustain significant fetal B-cell genesis and postnatal B cells, other identified or novel growth factors might play essential roles in promoting IL-7–independent B lymphopoiesis^8,9,21 ; and although TSLP might not play a primary role in IL-7–independent B lymphopoiesis, it could be involved in conjunction with other cytokines. Third, TSLP might as implicated play a unique role in IL-7–independent regulation of distinct B-1 B progenitors.^6,25  Importantly, as B-1 B cells can be maintained through peripheral expansion,²⁶  normal levels of B-1 B cells do not rule out a deficiency in B-1 B lymphopoiesis in the fetal liver or BM.

Previous studies of mice double-deficient in FMS-like tyrosine kinase receptor (FLT3 also called FLK2²⁷ ) or its ligand (FLT3L) and IL-7Rα (Flt3^−/−Il7r^−/− and Flt3l^−/−Il7r^−/− mice) have suggested important roles for IL-7, TSLP, and FLT3L in postnatal B-cell development.^8,28  However, the relative importance and interdependence of these cytokines in B-cell development could not be established, as the crossing of Flt3l^−/− and Il7r^−/− mice precluded dissection of the relative roles of IL-7 and TSLP in FLT3L-indpendent B lymphopoiesis.

Herein, we explored B lymphopoiesis in Tpte2^−/− as well as Il7^−/−Tpte2^−/− mice. In the absence of evidence for any important role of TSLP in IL-7–dependent or –independent B-1 and B-2 B lymphopoiesis, we also investigated whether lack of TSLP function might be involved in the almost complete loss of fetal B-1 and postnatal B-2 B-cell development reported in mice double-deficient in FLT3 and IL-7Rα signaling.^8,28  However, whereas TSLPR deficiency failed to have any further effect on postnatal B-cell development in Flt3l^−/− or Il7^−/− mice, we demonstrate that a combined loss of FLT3L and IL-7 is sufficient to lose all discernable stages of neonatal and adult B lymphopoiesis. For the first time, we also identified FLT3L and IL-7, and not TSLP, as essential regulators of the recently identified B-1–specified B-cell progenitors.⁶  Thus, we here unequivocally establish that FLT3L is the main cytokine driving IL-7–independent B-1 as well as B-2 B lymphopoiesis.

Flt3l^−/− mice²⁹  were on a pure C57BL/6 background. Il7r^−/− mice²⁰  were from The Jackson Laboratory (Bar Harbor, ME) and had been backcrossed for a total of 10 generations with C57BL/6 mice, as had Il7^−/−10  mice. WT C57BL/6 mice from The Jackson Laboratory were used as controls in the Il7r^−/− and Il7^−/− experiments. Tpte2^−/− mice²²  were on a mixed 129/SvJ/C57BL/6 background, and WT littermates were used as controls. Double-deficient mice were generated through intercrossing the aforementioned strains except for the mice double-deficient in FLT3L and IL-7Rα as they were obtained by cross breeding of Flt3l^−/− (pure C57BL/6) mice with Il7r^−/− mice backcrossed 5 generations to C57BL/6 background.²⁸  All experiments were approved by the Ethical Committee at Lund University.

Antibodies used for cell surface staining were as follows: 2.4G2 (CD16/32), RA3-6B2 (B220), RB6-8C5 (Gr1), M1/70 (CD11b, Mac1), 53-7 (CD5), 53-6.7 (CD8α), LY-76 (Ter-119), ID3 (CD19), AA4.1 (AA4.1), S7 (CD43), R6-60.2 (IgM), AL-21 (Ly6c), PK136 (NK1.1), H59-597 (TCRβ), GL3 (TCRγδ) (all from BD Biosciences PharMingen, San Diego, CA), and MB19-1 (CD19) from (eBioscience, San Diego, CA).

B-cell progenitors were defined as pre-pro-B (B220⁺CD43⁺AA4.1⁺CD19⁻), pro-B (B220⁺CD43⁺AA4.1⁺CD19⁺), pre-B (B220⁺CD43⁻IgM⁻), and imm-B/mature-B (imm-B, B220⁺CD43⁻IgM⁺). B-1 progenitors were analyzed in perinatal mice at the day of birth and defined as Lineage⁻ (anti-Ly6c, -CD8α, -CD11b, -Gr1, -IgM, -TCRβ, -TCRγδ, -NK1.1, and -Ter119), AA4.1⁺, CD19⁺, and B220^lo/neg.⁶  In the case of pre-B and imm-B/mature-B, 7 amino-actinomycin D (Sigma-Aldrich, St Louis, MO) was used to exclude dead cells and for the B-1 B progenitor staining ToPro-1 (Invitrogen, Carlsbad, CA) was used to as a viability dye. Data acquisition was performed on a FACSCalibur (BD Biosciences, San Jose, CA). Data analysis was carried out using FlowJo software (TreeStar, Ashland, OR).

All results are expressed as means (SD). The statistical significance between 2 groups of mice with deletion of single versus multiple signaling pathways was determined using a one-tailed Student's t test as removal of additional signaling pathways was hypothesized to result in the same or a more severe B-cell phenotype. Aspin-Welch correction was used in cases of unequal variances between groups.

Multiple previous studies implicated a considerably more severe B-cell phenotype in Il7r^−/− than Il7^−/− mice,^{3,8,9,11,19,30}  indirectly providing evidence for a role of TSLP in IL-7–independent lymphopoiesis. However, the only study that compared B-cell progenitors in the 2 mouse strains side by side and on identical genetic backgrounds did indeed fail to confirm any significant difference in B-cell generation in Il7r^−/− and Il7^−/− mice.¹²  In addition, direct comparison of embryonic B-cell development in Il7^−/− and Tpte2^−/−Il7^−/− fetuses revealed only small differences in immature-B (imm-B) cells in fetal liver, and no significant differences in B-cell progenitors.²³  In an effort to resolve the apparent discrepancy between previous studies and to potentially reveal a function of TSLP in postnatal B lymphopoiesis in the absence of IL-7, we first investigated the progression from pro-B to imm/mature-B cells in the BM of age-matched Il7^−/− and Il7r^−/− mice, both backcrossed for more than 10 generations to a C57BL/6 background. Because both IL-7Rα– and IL-7–deficient mice have an age-progressive loss of B-cell progenitors,^3,8,31  we chose to study 3-week-old mice in which both mouse strains, although severely reduced in numbers, sustain all stages from pro-B to mature-B cells. Whereas BM cellularity was not significantly different, spleen cellularity was further reduced in Il7r^−/− mice compared with Il7^−/− mice (1.6-fold; Figure 1A). Further, Il7r^−/− mice revealed reductions in frequencies as well as numbers of pro-B, pre-B, and imm/mature-B cells in the BM (Figure 1B), and numbers of mature B cells in the spleen (Figure 1C), compared with Il7^−/− mice.

These findings indicate, through an indirect approach, a potential significant, although limited, role of TSLP in adult IL-7–independent B-cell development. However, to more directly investigate the role of TSLP in IL-7–independent postnatalB lymphopoiesis, we next studied B-cell development in Tpte2^−/− and Il7^−/−Tpte2^−/− mice. Previous studies of Tpte2^−/− mice failed to establish a role of TSLP alone in adult conventional B-cell lymphopoiesis.^21,22  Here, we confirmed and extended these findings by a careful staging of pro-B to imm/mature-B cells^30,32  and found no effect of the TSLPR deficiency alone at any B-cell stage in steady-state adult BM (Figure 2A) or spleen (Figure 2B). In mice double-deficient in IL-7 and TSLPR, small (nonsignificant) differences in pro-B- and pre-B-cell numbers as well as a significant reduction in imm-B and splenic B cells were observed compared with IL-7–deficient mice (Figure 2C,D), resembling at least to some degree the further reduction in Il7r^−/− mice compared with Il7^−/− mice, confirming a significant but very modest role of TSLP in promoting postnatal B lymphopoiesis in the absence of IL-7.

Strikingly, previous studies of Flt3l (or Flt3 receptor) and Il7r double-deficient mice, which lack FLT3L, IL-7, as well as TSLP-mediated signaling, failed to demonstrate detectable numbers of committed B-cell progenitors or mature B cells.^8,28  To clarify the relative role of these 3 cytokines in this complete loss of B lineage development, we generated and examined Flt3l^−/−Il7^−/− as well as Flt3l^−/−Tpte2^−/− double-deficient mice. Notably, Flt3l^−/−Il7^−/− as Flt3l^−/−Il7r^−/− mice were indistinguishable in that they both sustained little or no B-cell progenitors or mature B cells (Figure 3A,B), whereas, in striking contrast, the B-cell phenotype of Flt3l^−/−Tpte2^−/− mice was indistinguishable from that of Flt3l^−/− mice (Figure 3C,D), establishing a primary role of FLT3L, and not TSLP in IL-7–independent B lymphopoiesis.

B-1 B cells represent a distinct population of embryonic and early postnatally derived self-replenishing B cells, proposed to be regulated through mechanisms distinct from those of conventional B-2 B cells.^7,33,34  Recently, a population of B-1 B progenitors (lin⁻AA4.1⁺CD19⁺B220^−/low) was identified in fetal liver, which is sustained at low levels also in postnatal BM.⁶  These B-1 B progenitors were found to be highly TSLP-responsive; and although present at normal numbers in the BM of adult Tpte2^−/− mice,⁶  a potential important role of TSLP in regulation of B-1 B progenitor has been proposed.²⁵  However, the role of TSLP or other cytokines in regulating IL-7–independent B-1 B progenitors in late fetal development or perinatal BM has not been investigated. Here, B-1 B-cell progenitors were found to be present at normal levels in the BM of newborn Tpte2^−/− mice (Figure 4A). Furthermore, IL-7–independent generation of B-1 B-cell progenitors was also not driven by TSLP, as no further reduction was observed in Tpte2^−/−Il7^−/− mice compared with Il7^−/− mice (Figure 4B). In contrast, B-1 B progenitor numbers were reduced 4-fold and 40-fold in Flt3l^−/− and Il7^−/− mice, respectively, and were undetectable in Flt3l^−/−Il7^−/− mice (Figure 4C,D), demonstrating the critical importance and interdependence of FLT3L and IL-7 also in regulation of B-1 cell progenitors, and an inability of TSLP to rescue any fetal B-1 or B-2 B-cell development in the absence of FLT3L– and IL-7–mediated signaling.

Previous studies have proposed TSLP as the key regulator of fetal and early postnatal IL-7–independent B lymphopoiesis.^8,9  Although this conclusion was cast into doubt as more recent studies of fetal B-cell development in mice deficient in IL-7 and TSLPR failed to uncover an essential role of TSLP in IL-7–dependent and –independent fetal B lymphopoiesis,²³  the potential importance of TSLP in B lymphopoiesis has been reiterated.²⁵  The establishment of which cytokine(s) might be responsible for IL-7–independent B lymphopoiesis in mice could have important implications for a better understanding of which pathways might regulate human B lymphopoiesis, as this has been suggested to be primarily or even entirely IL-7–independent, although studies performed on patients have been rather restricted.^14,16,24 

Whereas previous claims in support of a key role of TSLP in IL-7–independent mouse B lymphopoiesis were based on indirect evidence through comparative analysis of B-cell development in Il7- and Il7r-deficient mice and in vitro TSLP responsiveness of B-cell progenitors,^6,8,9  the present studies in mice deficient in either one or multiple cytokines provided direct evidence for FLT3L rather than TSLP being essential for IL-7–independent B lymphopoiesis. Thus, although our comparative studies of Il7^−/− and Il7r^−/− and Il7^−/− and Il7^−/−Tpte2^−/− mice established a limited role of TSLP in IL-7–independent B lymphopoiesis, the importance of TSLP appears to be rather marginal compared with that of FLT3L. Although Tpte2-deficient BM cells and splenocytes have been shown to be unresponsive to stimulation with TSLP ligand,^21,22  it cannot be completely ruled out that there might be alternative TSLP receptor signaling subunits at some stages of B-cell development; and, if so, the importance of TSLP in B lymphopoiesis might have been underestimated in our studies. However, importantly, in the absence of FLT3L and IL-7, TSLP was regardless unable to rescue any discernable B-1 or B-2 cell development. Our findings of a lack of a key role of TSLP in mouse B lymphopoiesis are in agreement with the findings in IL-7Rα deficient patients.^16,24 

Herein, we also identified, for the first time, the role of specific cytokine receptor-ligand pairs in regulation of distinct B-1 B progenitors and found that TSLP was dispensable not only for B-2 B lymphopoiesis but also for IL-7–dependent and -independent regulation of B-1 B progenitors. In contrast, IL-7 and FLT3L play the same critical role in regulation of B-1 B progenitors as of conventional B-2 B progenitors, in that IL-7 appears to be the primary regulator, whereas FLT3L is absolutely required for any IL-7–independent B-1 and B-2 B lymphopoiesis to occur, further demonstrating that TSLP has no ability to rescue early B-cell development in the absence of these 2 regulators of B lymphopoiesis. Thus, these findings demonstrate that, among the investigated cytokines, FLT3L and IL-7 are sufficient and required to drive each of the 2 developmentally distinct pathways of B-1 and B-2 B lymphopoiesis. Importantly, as human severe combined immunodeficiency patients with an IL-2Rg or IL-7Rα deficiency have normal numbers of B cells but abrogated T-cell development, these findings warrant a careful investigation of a potential involvement of FLT3 receptor or ligand dysfunction in human B-cell deficiencies. FLT3L has been shown to facilitate the in vitro generation of human B cells from cord blood progenitors, but it remains to be established to what degree FLT3L is physiologically involved in regulating normal human B lymphopoiesis.³⁵ 

The authors thank Lilian Wittman for expert technical assistance, James Ihle (St Jude Children's Research Hospital, Memphis, TN) for generously providing Tpte2^−/− mice, Ana Cumano (Pasteur Institute, Paris, France) for kindly providing Il7^−/− mice, and Tobias Ryden for help with the statistical analyses.

This study was supported by grants from the Swedish Research Council, Juvenile Diabetes Research Foundation (JDRF; New York, NY), and the Göran Gustafsson Foundation. S.E.W.J. and this project are supported by a Medical Research Council (MRC, London, UK) Strategic Appointment award. The Lund Stem Cell Center is supported by a Center of Excellence grant from the Swedish Foundation for Strategic Research (Stockholm, Sweden).

Contribution: C.T.J., E.S., S.K., C.B., and S.E.W.J. designed and conceptualized the research, analyzed the data, and wrote the manuscript; and C.T.J., S.K., C.B., M.C., and A.L. did the phenotypic characterization of different knockout mice.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Sten Eirik W. Jacobsen, Haematopoietic Stem Cell Laboratory, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Headington, Oxford OX3 9DS, United Kingdom; e-mail: sten.jacobsen@imm.ox.ac.uk.