A primary research article can safely be considered influential when it has been cited >1100 times. This is certainly the case for Richard R. (Randy) Hardy et al.’s (1) “Resolution and characterization of pro-B and pre–pro-B cell stages in normal mouse bone marrow.” In fact, it continues to influence how immune system replenishment is studied and described some 20 y later.

By 1991, the existence of hematopoietic stem cells (HSCs) had been known for some time. HSCs that reside in adult bone marrow give rise to lymphocytes and other blood cells. Large dividing and small nondividing “pre-B” cells had also been discovered in bone marrow and were isolated using mAbs. As indicated by the name, pre-B cells were known to be the immediate precursors of B lymphocytes. The Ig gene rearrangement process needed to assemble Ag receptors on these lymphocytes was thought to operate in a stepwise fashion. Furthermore, a probable sequence for these events was predicted from studies of human and murine tumors.

However, serious technical limitations restricted progress in this very important field, leaving huge gaps in our knowledge of the process. Newly formed B lymphocytes were considered to arise from a specialized “assembly line,” but little was known about the steps between HSC and pre-B stages. Few developmental milestones or cell isolation procedures were available to allow probing of molecular events. As one important example, it was not formally demonstrated how Ag receptor genes were rearranged in normal cells. Transgenic and knockout models were starting to become available, but it was difficult to exploit them in assessing the functional significance of particular lymphocyte genes. A “roadmap” of B lymphocyte lineage differentiation, especially one pertaining to early progenitors, was seriously needed.

Pre-B cells were defined as having detectable cytoplasmic μ IgH chains but no κ/λ L chains or surface Igs (2). Before the Hardy et al. paper, pre-B cells could only be enumerated by two-color staining with difficult-to-standardize polyclonal Abs and tedious microscopy. Hardy et al. (1) identified particular mAbs that differentially stained B lymphocyte lineage cells and used them for four-color flow cytometry, a method that was far from routine at the time. Relatively abundant small pre-B cells (designated fraction D) were identified as B220+CD43sIgM, whereas more primitive cells were resolved and designated fractions A, B, and C.

No fixation was required, so cells in the Hardy fractions could be sorted in viable form and cocultured on a supporting stromal cell line. This approach revealed that fraction A cells acquired properties of fraction B, and then B assumed fraction C characteristics (1). The small pre-B–containing fraction D generated sIgM+ B cells in short-term cultures. Cumbersome thymidine labeling methods had previously provided kinetic information relating to B lymphopoiesis, but Hardy et al. directly demonstrated proliferation of particular progenitor cells and responsiveness to IL-7. As might be expected from previous studies, fraction D cells were postmitotic, whereas the most proliferative progenitors were found to be in a CD24/30F1hi subset of fraction C. Thus, the tremendous expansion required for daily generation of large numbers of lymphocytes preferentially occurred in this small population of progenitors.

At the time of these discoveries, new approaches were being developed to dissect the hematopoietic microenvironment in bone marrow. Dexter and Whitlock–Witte culture conditions supported long-term generation of myeloid and lymphoid cells, respectively (3). That line of investigation led to the generation of many hematopoiesis-supporting stromal cell lines. Hardy et al. exploited this knowledge to show that the various stages of B lymphopoiesis differed with respect to stromal cell contact and soluble factor dependence.

As another first, Hardy et al. (1) demonstrated stepwise rearrangement of Ig genes in normal bone marrow cells. All Ig loci in fraction A were in germline configuration, whereas partial DH–JH rearrangement of H chain genes had occurred by the time progenitors reached the fraction C stage. Preceding the development of single-cell rearrangement assays, Hardy et al. learned that VH–DHJH rearrangements did not involve both alleles. Finally, progenitors were shown to rearrange VL–JL gene segments while transitioning from C to D stages.

Although incompletely settled even today, nothing was known at that time about how the process of B cell formation is initiated. Hardy et al. (1) predicted that the least differentiated B220+CD43+BP-1CD24 “fraction A” was probably a mixture of cells, as was subsequently shown to be the case. That is, other laboratories extended the Hardy et al. approach by demonstrating that fraction A also included progenitors destined to become other blood cell types. It is now a routine finding that any subset of stem/progenitor cells can be subdivided when informative markers become available.

Hardy et al. used refreshingly simple A, B, C nomenclature for fractions and depicted changes in levels of marker expression as progressive (Fig. 8 in Ref. 1), which correctly described the process as a continuum rather than a series of abrupt changes. It is now clear that many important differentiation events are not closely synchronized. For example, most progenitors that have recently lost nonlymphoid options would be designated by surface markers as common lymphoid progenitors or pre–pro-B cells. However, some cells phenotypically classified as HSCs are already biased to become lymphocytes (4).

If nature abhors a vacuum, Randy Hardy was the ideal person to fill it. After receiving his Ph.D. from the California Institute of Technology, he trained as a postdoctoral fellow in the famed Herzenberg Laboratory at Stanford University. The timing and location was perfect, as new mAbs and fluorochrome technology were being used in flow cytometry to discover new B cell subsets. Hardy’s interest in one of these subsets, Ly-1 (B1), inspired him to learn more about the origins of “conventional” (B2) B cells. This work was continued after Hardy joined the Fox Chase Cancer Center in 1987 and coincided with the description of SCID mice by Bosma and colleagues (5). Because SCID mice lacked B and pre-B cells in bone marrow, Hardy was able to focus on the more primitive pro-B cells, find distinctive markers for them, and construct the now famous “Hardy scheme.”

The approaches described in this classic paper have driven discovery of ever smaller populations of progenitors and inspired differentiation diagrams for T, NK, and dendritic cells. Reflecting on this paper reminds us of how far the field has moved, as well as how much remains to be learned (6). This is especially the case for human B lymphopoiesis, where a pre-B growth factor that is of importance equivalent to IL-7 in the mouse stubbornly resists discovery.

Abbreviation used in this article:

HSC

hematopoietic stem cell.

1
Hardy
R. R.
,
Carmack
C. E.
,
Shinton
S. A.
,
Kemp
J. D.
,
Hayakawa
K.
.
1991
.
Resolution and characterization of pro-B and pre–pro-B cell stages in normal mouse bone marrow
.
J. Exp. Med.
173
:
1213
1225
.
2
Raff
M. C.
,
Megson
M.
,
Owen
J. J.
,
Cooper
M. D.
.
1976
.
Early production of intracellular IgM by B-lymphocyte precursors in mouse
.
Nature
259
:
224
226
.
3
Kincade
P. W.
1987
.
Experimental models for understanding B lymphocyte formation
.
Adv. Immunol.
41
:
181
267
.
4
Muller-Sieburg
C. E.
,
Sieburg
H. B.
,
Bernitz
J. M.
,
Cattarossi
G.
.
2012
.
Stem cell heterogeneity: implications for aging and regenerative medicine
.
Blood
119
:
3900
3907
.
5
Schuler
W.
,
Weiler
I. J.
,
Schuler
A.
,
Phillips
R. A.
,
Rosenberg
N.
,
Mak
T. W.
,
Kearney
J. F.
,
Perry
R. P.
,
Bosma
M. J.
.
1986
.
Rearrangement of antigen receptor genes is defective in mice with severe combined immune deficiency
.
Cell
46
:
963
972
.
6
Hardy
R. R.
,
Kincade
P. W.
,
Dorshkind
K.
.
2007
.
The protean nature of cells in the B lymphocyte lineage
.
Immunity
26
:
703
714
.

The author has no financial conflicts of interest.