RAG1 and RAG2 in developing rabbit appendix subpopulations.

The appendix of young rabbits is a site of primary heavy chain variable region-gene diversification and B cell selection. Appendix cells from 6- to 9-wk-old rabbits were stained and sorted for surface CD43 and IgM. We found that the CD43+IgM- and double-negative CD43-IgM- cells contained RAG1 transcripts and RAG2 protein. The presence of RAG gene products in appendix raised the possibility that pro-/pre-B cells were present in young rabbit appendix. Although an early suggestion that RAG2 plays a role in variable region-gene diversification by gene conversion in chicken bursa was not supported by studies of RAG2 protein in this tissue, we produced anti-rabbit RAG2 Abs to determine whether RAG2 protein was present in rabbit appendix, where cells that recently underwent gene conversion are found. We detected RAG2 protein in the four subpopulations of rabbit appendix lymphocytes, distinguished by surface CD43 and IgM markers. The appearance of RAG gene products during different stages of B cell maturation may reflect the function of the young rabbit appendix as a site of both B cell development and diversification.

Rabbit IgH sequences in appendix germinal centers: VH diversification by gene conversion-like and hypermutation mechanisms.

Although the rabbit IgH locus contains approximately 100 VH genes, the majority of B cells rearrange VH1. To produce a primary repertoire containing a sufficient number of protective antibodies, rearranged VH1-DH-JH sequences may diversify within rabbit B cells in an organ that functions like a chicken bursa, sheep ileal Peyer's patch, or both. It was suggested many years ago that the rabbit appendix could be a bursal equivalent. To reexamine this possibility, we analyzed rearranged heavy chain variable region sequences in B cells from light and dark zones of appendix germinal centers from 6-week-old rabbits. Our findings indicate that antibody diversification occurs by gene conversion-like and somatic hypermutation mechanisms in appendix germinal centers of young rabbits.

Sea star factor blocks development of T-dependent antibody secreting clones by preventing lymphokine secretion.

Sea star factor (SSF), a protein of 39 kDa purified from macrophage-like coelomocytes of the echinoderm Asterias forbesi, has potent immunosuppressive effects on T-dependent but not T-independent antibody responses in vivo. SSF at a concentration of 0.5 microgram/ml markedly inhibits T-dependent antibody production in vitro by fluorescein (Flu)-specific B cells responding in clonal microculture to antigenic stimulation with Flu-conalbumin via the conalbumin-specific T cells D10.G4.1 (D10). At this concentration of SSF, Ig secretion induced by a T cell-independent stimulus, lipopolysaccharide (LPS), is not affected. Inhibition of antibody production in T-dependent microcultures by SSF can be completely overcome in a dose-dependent fashion by addition of lymphokine-rich supernatants from stimulated cultures of D10 cells. The possibility that SSF suppresses production of requisite cytokine growth factors from T cells was substantiated by the finding that SSF diminishes concentrations of stimulatory cytokines detectable in supernatants from antigen-stimulated cultures. Nevertheless, levels of intracytoplasmic mRNA for IL-4 and IL-5 are not detectably altered by concentrations of SSF that suppress antibody production. Furthermore, when cultures of D10 cells stimulated in the presence of SSF are subjected to freezing and thawing to release intracytoplasmic lymphokines, total levels of stimulatory cytokines are not lower than those in cultures without SSF. These results suggest that SSF inhibits antibody responses by limiting the availability of lymphokines produced by helper T cells. The mechanism for this inhibition may involve either direct effects of SSF on T cells or a block in effective T cell-B cell interaction.

The appendix functions as a mammalian bursal equivalent in the developing rabbit.

In this paper we present genomic DNA sequence and histological evidence that the appendix is a site of diversification of the rabbit's primary antibody repertoire. By 6 weeks after birth, the B cell follicular regions of the rabbit appendix and the distribution of the resident lymphoid cells bear a strong morphological resemblance to similar regions within two primary lymphoid tissues, the chicken bursa and the sheep ileal Peyer's patch. However, similarities between the rabbit appendix, chicken bursa and sheep ileal Peyer's patch end as these animals reach adulthood. The rabbit appendix undergoes morphological and cellular distribution changes as it matures taking on the appearance of a secondary lymphoid tissue, while the sheep ileal Peyer's patch and the chicken bursa both involute. We determined DNA sequences of PCR amplified rearranged variable region genes from germinal center B cells of 6 week old rabbits isolated from several different appendix dark zones and light zones. There was a trend toward a higher degree of diversification from the germ-line VH gene DNA sequence in dark zones than light zones. It is likely that both gene conversion and somatic hypermutation are responsible for the nucleotide changes we observed. Our findings suggest that the rabbit appendix functions as a mammalian bursal equivalent early in development. As the rabbit matures, the appendix appears to evolve into a secondary lymphoid tissue resembling secondary GALT in appearance and possibly in function.

Systemic complications of fluid resuscitation.

Fluid administration in critically ill individuals is frequently a major component of their therapy. There are important effects on blood pressure and maintenance of cardiac output and oxygen delivery, as detailed elsewhere in this text. There are also potentially negative side effects of this therapy, which have been less well defined. Edema of the gastrointestinal tract has been well described, primarily with crystalloid infusions. Gastrointestinal edema may have very complicated effects on albumin kinetics, fluid flux, and ion flux. It may lead to development of ileus. Increased nasogastric tube output may be incorrectly construed as unremitting obstruction rather than a result of the aforementioned changes and increased crystalloid loads. The relationships of intestinal edema to intestinal absorptive function and diarrhea are less clear. At present, changes in type of fluid infusion or correction of serum albumin level to normal cannot be uniformly recommended. The myocardium, although showing evidence of edema with crystalloid infusion, may appear to benefit from colloidal, osmotically active suspensions in the all too few studies that have been done. To date, there is no study giving evidence of clinically different outcome using a variety of fluids that cause, reduce, or prevent this edema. The presence or absence of myocardial edema may be important in patients who demonstrate decreased ventricular function during sepsis or other disorders in which aggressive fluid administration is routine. Edema of the skin has been associated primarily with decreased oxygen tension. Other studies have shown an association with impaired wound healing or increased risk of infection. A direct causal relationship can only be inferred. We are left with a sense that aggressive fluid resuscitation with crystalloid, although improving oxygen delivery, may have other deleterious effects on organ systems, such as the gastrointestinal tract, myocardium, and integument. The edema resulting from crystalloid administration may lessen or negate the benefits of increased oxygen delivery. Care needs to be taken in interpreting any alteration in organ function with respect to the fluid type and volume being administered. An alternative choice of therapy is lacking at present. The role of colloid has not been as well investigated as that of crystalloid and further study is warranted before any benefits become clear.

Molecular genetic features reflecting the preference for isotype switching to IgA expression by Peyer's patch germinal center B cells.

It has proven difficult to evaluate the functional potential of germinal center (GC) B cells, including those from Peyer's patches (PP), by either in vivo or in vitro methods. Thus, rather than assess secreted Ig product as an indicator of functional potential we have instead sought to detect mRNAs related to the various Ig heavy chains in GC B cells from PP by in situ hybridization. We have found that the GCs of PP contain the vast majority of B cells with easily detectable levels of mRNA alpha. These levels are intermediate between those of small resting B cells and plasmablasts. When PP B cells are enriched for cells bearing GC markers, approximately 50% contain mRNA mu and 40% mRNA alpha. Similar enrichment for sIgA+ B cells gave 50% of cells with easily detectable mRNA alpha and few if any positive for mRNA mu. The sizes of these mRNAs were similar to those encoding the membrane and secretory form of mu and alpha chains. No C alpha germ-line transcripts could be detected by Northern analyses using a probe for sequences 5' to the alpha switch regions. Finally, GC and sIgA+ cells from PP also showed the absence of a portion of their genomic DNA for CH genes 5' of C alpha. Thus, it seems likely that most of the GC cells expressing mRNA alpha have undergone conventional VDJ recombination to C alpha at the DNA level in order to switch to the expression of IgA. Our findings reflect the extraordinary preference for switching to IgA by GC cells in PP.

The preference for switching to IgA expression by Peyer's patch germinal center B cells is likely due to the intrinsic influence of their microenvironment.

This study was aimed at determining whether the chronically activated physiologic state of Peyer's patch (PP) tissue is primarily responsible for the IgA isotype preference expressed by PP germinal centers (GC) and memory B cells. We have used reovirus type 1/Lang to stimulate acute, de novo GC reactions in lymph nodes (LN) or PP to test the possibility that the surface (s)IgA component of enteric responses is peculiar to the local gut microenvironment whether or not PP are in a state of chronic activation. GC were raised in PP of germ-free mice by oral administration of virus and in lymph nodes (LN) of conventionally reared mice by local parenteral infection. Transient GC reactions were found to develop with similar time courses in both PP and LN after both primary and secondary local infections with reovirus. sIgA+ B cells, which were the major non-sIgM+/sIgD+ population found to arise in GC of PP, were not found in the LN. In LN, sIgG1+ B cells comprised the predominant non-IgM/IgD bearing population that arise after local infection. Lymphoid fragment cultures of PP initiated in vitro as early as 5 days after in vivo infection contained detectable secreted reovirus-specific IgA, whereas IgG1 but no IgA was found in LN cultures. Northern blot analysis to detect C alpha and C gamma 1 germline transcripts further substantiated a site-related bias in the expression of non-IgM/IgD isotypes that was manifest within a few days after infection. In summation, these observations taken together suggest that the preference for generating sIgA+ B cells in PP may be the result of intrinsic features of their gut microenvironment.

Use of Peyer's patch and lymph node fragment cultures to compare local immune responses to Morganella morganii.

Lymphoid tissue fragment cultures were established to analyze the differentiative processes among B cells in Peyer's patches (PP) and peripheral lymph nodes (PLN), especially those in germinal centers. PP cultures from both conventionally reared mice and formerly germ-free mice colonized with Morganella morganii could be maintained for greater than 12 days with continued B-cell division, especially among cells binding high levels of peanut agglutinin, a characteristic of germinal center cells. PLN cultures from conventionally reared mice injected with a heat-killed vaccine of M. morganii could be maintained for the same amount of time. Over this period, PP cultures continued to secrete immunoglobulin A (IgA) as well as smaller amounts of IgM. PP cultures from formerly germ-free mice colonized with M. morganii showed net increases of IgA antiphosphocholine (anti-PC) antibodies with avidities as high as those of the prototypic T15 monoclonal antibody. Similar PLN fragment cultures from conventionally reared mice given footpad injections of M. morganii showed net increases of IgM and IgG anti-PC antibodies in the culture fluid. Thus, although M. morganii stimulated lymphoid tissues in vivo to produce an anti-PC response in vitro when given by either the oral or the parenteral route, the antibody isotypes differed between PP and PLN fragment cultures. Fragment culturing may offer a complementary and simpler way to detect a local secretory IgA response than does either measuring IgA antibody in secretions or detecting IgA antibody in the cytoplasm of plasma cells in the lamina propria of gastrointestinal or respiratory tissue.