The paper by Butcher et al. (1) presented here is one of my favorites because of its importance to the field of mucosal immunology. One may ask how a paper with a somewhat restricted aim, which used a lectin peanut agglutinin (PNA) that others had already shown to bind Peyer’s patch germinal centers, could truly be a classic. It has been cited more than 250 times since its publication; however, this number, although excellent, is somewhat modest. I encourage you to fully read this paper for its novelty and discover how it helped change the field of B cell commitment to IgA in mucosal inductive sites, and thus affected the entire field.

To fully appreciate this paper, you should realize the Peyer’s patches, which are a major component of the gut-associated lymphoreticular tissues (GALT), are major sites for Ag uptake for induction of mucosal secretory IgA (S-IgA) Ab responses. The Peyer’s patches have a lymphoepithelium containing M cells for Ag uptake. In addition, Peyer’s patches exhibit prominent B cell zones with germinal centers (usually five or more per Peyer’s patch). We now know, of course, that μ→α B cell switching occurs in high frequency in the B cell germinal centers of Peyer’s patches, and this classic paper by Butcher et al. certainly provided the most convincing evidence for this feature. This is not a trivial point, since Abs of the IgA isotype represent over two-thirds of all Igs produced in mammals.

To fully appreciate the contribution by Butcher et al., we should step back in time and appreciate the unique methods that allowed for such wonderful conclusions, that Peyer’s patch germinal centers contain enormous numbers of B cell blasts that are switching to the IgA isotype, i.e., they are surface IgA-positive (sIgA+) B cells. As indicated, it had already been shown that PNA binds to Peyer’s patch B cells in germinal centers (2). However, this earlier study did not tell us much about the B cells themselves. One should also appreciate that the two-laser flow cytometer was just being developed in the Herzenberg Laboratory at Stanford University. The authors had access to this new development before it was marketed by Becton Dickinson as a FACS II (see Materials and Methods of Ref.1). This allowed the use of green color (FITC) and two red colors (Texas Red and RITC) simultaneously. The new dye, Texas Red, was used to fluorochrome label PNA, and the advantages of the dye allowed separation of germinal center cells into PNAhigh, PNAlow, and PNAneg B cells. As it turned out, the PNAhigh B cells were the germinal center cells, while the others were present in the B cell zone. This differentiation of PNAhigh vs PNAlow is now commonplace in studies in mice, rabbits, rats, and humans.

You should also keep in mind while reading this paper that the use of rabbit or goat anti-mouse isotype-specific Abs often was not completely purified. mAbs were just being developed and used; however, in 1982, we still often used goat or rabbit anti-α Abs, which were absorbed with other mouse isotypes (usually myelomas). You should judge for yourself whether the conclusions drawn for numbers of sIgA+ B cells in GALT germinal centers are accurate. In doing this, you should note the extensive absorbance conducted with the rabbit anti-α Ab over columns made from myeloma Igs of other isotypes/both L chains. The authors backed up their results with the use of a newly developed rat anti-mouse α mAb and essentially made similar findings (see Table II of Ref.1).

Another interesting aspect of the methods used in this paper involves the actual isolation of Peyer’s patch lymphoid cells. The authors used a mechanical method (pushing Peyer’s patches through sterile wire mesh), which is notorious for allowing isolation of B lymphocytes but not APCs like dendritic cells and clusters of T cells. Therefore, when you read about the comparison of percentages of germinal center B cell blasts (PNAhigh cells) as determined by morphometric analysis with isolated lymphocytes, you should appreciate that this indeed may not be a truly representative population. Nevertheless, the numbers of PNAhigh cells in both situations were similar, and the authors were correct in making the conclusions that they did based on their data.

I will not go figure-by-figure with the reader; however, please appreciate the need to literally count thousands of cells in sections like those in Fig. 2 with those shown in the adjoining FACS profiles. Fig. 3 is quite important, since it shows that PNAhigh germinal center B cells contain high numbers of large blasts. A key point in studies in 1982 (as they are today) was whether sIg on cells that stained positive for a particular isotype was actually produced by that cell or had been acquired passively, in the case of sIgA from the ubiquitous mucosal IgA environment. The results shown in Table III were derived from use of an allotype-specific mAb and clearly show that the sIgA+ B cells actually produced the IgA.

I really liked the Butcher et al. Discussion for several reasons. First, the authors gave credit to others for discovery of PNA binding to germinal center cells and then proceeded to highlight the results of this important study. The overall purpose was again stated (as it was in their Introduction). Quite simply said, the purpose was to examine the surface phenotype of Peyer’s patch germinal center lymphoid cells. As we have said throughout this summary, the authors were able to conclude that Peyer’s patch germinal centers contain mostly B cells that are actively dividing (B cell blasts). Quite remarkably, the germinal center B cells, which of course lack sIgD, contain up to 70% sIgA+ B cells. One has to conclude that B cells in Peyer’s patch germinal centers are actively dividing and are committing (switching) to sIgA in very high frequency when compared with germinal centers in peripheral lymphoid tissues where most B cells are switching to IgG subclasses. The authors continue their discussion by indicating that this study does not shed light on the reason why sIgA+ B cells migrate to effector sites before differentiating into IgA-producing plasma cells. That largely remains a question even today, although the mechanisms for homing of lymphocytes to and from GALT and on to mucosal effector sites are reasonably well understood.

In summary, the authors presented a logical series of experiments that were built on recent knowledge in 1982 that the lectin PNA binds to B cells in germinal centers. Their studies made extensive use of existing and recently developed Ab reagents, some of which were quite notable at the time. They also nicely combined in situ immunohistochemistry with in vitro cell suspensions for two-laser flow cytometry for differentiation of B lymphocytes and other cell types in Peyer’s patches. The result was a terrific study strongly implying that mucosal inductive sites indeed have germinal centers where very high proportions of B cells are switching to the IgA isotype. Authors and chapter writers still quote (as they should) this paper when the point is made that mucosal inductive sites, and in particular the germinal centers, are the locale of isotype switching to IgA, which of course is the major Ab isotype present in mucosal secretions.

1
Butcher, E. C., R. V. Rouse, R. L. Coffman, C. N. Nottenburg, R. R. Hardy, I. L. Weissman.
1982
. Surface phenotype of Peyer’s patch germinal center cells: implications for the role of germinal centers in B cell differentiation.
J. Immunol.
129
:
2698
-2707.
2
Rose, M. L., M. S. C. Birbeck, V. J. Wallis, J. A. Forrester, A. J. S. Davies.
1980
. Peanut lectin-binding properties of germinal centers in mouse lymphoid tissue.
Nature
284
:
364
-366.