by Harwood Academic Publishers GmbH Printed in Singapore Two Distinct Pathways of B-Cell Development in Peyer’s Patches

The developmental biology of sheep ileal and jejunal Peyer's patches (PP) was investigated using corticosteroids to deplete immature B lymphocytes. During a 7-day treatment with dexamethasone, ileal PP follicular (iPf)B-cell proliferation was arrested and most iPfB-cells died. This resulted in follicular involution with the survival of mesenchymal cells. No iPfB-cell proliferation was detected in follicular remnants for 4 weeks postdexamethasone treatment, and during a subsequent 3-month period, there was limited iPfB-cell proliferation that resulted in a partial regeneration of follicles. Ileal PP involution was also associated with a severe B lymphopenia that persisted for over 14 weeks and was characterized by the survival of primarily isotype-switched and CD5 sIgM B-cells in blood. In contrast, the size of jejunal PP follicles was reduced following dexamethasone treatment, but intrafolli-cular B-cell proliferation was not arrested. Furthermore, within 4 weeks, the jejunal PP follicles had recovered in size and cellularity and there was no disruption in IgA plasma-cell production. Thus, dexamethasone selectively depleted iPfB-cells and revealed that the ileal and jejunal PPs contain functionally distinct B-cell populations. The partial regeneration of the iPfB-cell population indicated that either an intrafollicular, corticosteroid-resistant B-stem cell existed or that ileal PP follicles can be repopulated by circulating B-cells. Finally, the association between ileal PP involution and the absence of circulating, CD5-B-cells confirmed that this lymphoid tissue provides an essential environment for conventional sIgM B-cell development.


opulation ind
cated that either an intrafollicular, corticosteroid-resistant B-stem cell existed or that ileal PP follicles can be repopulated by circulating B-cells.Finally, the association between ileal PP involution and the absence of circulating, CD5-B-cells confirmed that this lymphoid tissue provides an essential environment for conventional sIgM B-cell development.

INTRODUCTION

Investigations in rodents have established the con- cept that after birth, the bone marrow functions as the primary site of B-cell generation (Claman et al.,  1966; Mitchell and Miller, 1968; Osmond, 1980), and investigations in both rodents and rabbits re- vealed that the Peyer's patches (PPs) function as a secondary, antigen-dependent lymphoid tissue that plays a major role in mucosal immunity (Pollard and   Sharon, 1970; Craig and Cebra, 1971; Husband and   Gowans, 1978).Consistent with this hypothesis, lymphoid follicles in the PPs develop postnatally in these species (Hummel, 1935; Waksman et alo, 1973;   "Corresponding author.Present address: Veterinary Infectious Disease Organisation, 120 Veterinary Road, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5E3.

Abe and Ito, 1977).However, in humans (Cornes,  1965), sheep (Reynolds and Morris, 1983), pigs (Chapman et al., 1974), and other species (Carlens,  1928), the lymphoid ollicles of the PPs are well- developed prior to birth, and in PPs of fetal sheep, there is a high level of B-cell proliferation (Reynolds  and Morris, 1983).This fetal developmental sug- gests that B-cell development in the PPs is, to some extent, antigen-independent and T-cell- independent.Thus, PPs may function as a site of both antigen-dependent and antigen-independent B-cell development.

Located along the small intestine of sheep are two distinct types of PPs: the ileal PP and the jejunal PP.These two PPs differ markedly in their life history, histology, lymphocyte composition, B-cell differen- tiation, and their role in the development of the humoral immune system.During ontogeny, lymphoid follicles develop first in the jejunal PPs, but in 264 P.J. GRIEBEL et al.   late-term fetuses, the follicular B-cells proliferate at MATERIAL AND METHODS a high level in both PPs (Reynolds and Morris,  1983).However, B-cell development in the two Reagents PPs appears to diverge at birth (Griebel et al., Dexamethasone  (9a-fluoro-16a-methylpredniso-1992).Jejunal PPs are active throughout the life of lone) was purchased from Sigma (St.Louis); the animal and are a site for T-cell-dependent 5-bromo-2-deoxyuridine (BrdU)and mouse mono- antigen responses with the production of both clonal (mAb) anti-BrdU (BMC 9318)were purchased IgG1 and IgA PCs (Gerber et al., 1986; Griebel from  Boehringer-Mannheim (Mannheim, Ger-  and Ferrari, 1995).In contrast, in the ileal PP, the many).The biotinylated-, fluorescein isothiocyanate production of B-cells (Griebel and Ferrari, 1994) (FITC-), and phycoerythrin (PE-) conjugated, and diversification of the immunoglobulin (Ig) isotype-specific goat anti-mouse Ig reagents were receptor repertoire (Reynaud et al., 1995) are purchased from Southern Biotechnology (Birming- antigen-independent and T-cell-independent and ham, AL).The PE-conjugated rat anti-mouse CD8 the lymphoid follicles involute following sexual was purchased from Caltag (San Francisco).Soluble, maturity (Reynolds and Morris, 1983).In young recombinant fusion protein of murine CD40L-CD8 sheep, the ileal PP is the primary source of B-cells (mCD40L-CD8; Lane et al., 1993) was a generous for all lymphoid tissues and provides an essentials gift from Peter Lane (Basel Institute for Immunol- environment for B-cell development (Gerber et al., ogy, Basel, Switzerland) and was previously shown 1986; Reynolds et al., 1991).to react with sheep CD40 (Griebel and Ferrari, Immature lymphocytes, such as cortical thymo-1995).The anti-IgM (PIg45A), anti-IgG1 (Blg715A), cytes (Ishidate and Metcalf, 1963) and B-cells in anti-IgA (Blg312D3), anti-X Ig LC (Blg501E), and lymphoid follicles of the chicken bursa of Fabricius anti-Ig LC (Blg43) mAbs were purchased from (Glick, 1957), display a high corticosteroid sensitiv-VMRD Inc. (Pullman, WA).The pan-B cell (Du2- ity.In contrast, mature lymphocyte populations in 104; Press et al., 1993), anti-CD4 (17D-13; Maddox secondary lymphoid tissues are much less sensitive et al., 1985), anti-CD5 (STla; Beya et al., 1986), to the cytolytic effects of corticosteroid (Cupps and anti-CD8 (E95; Ezaki et al., 1987), anti-,3 T-cell  Fauci, 1982).The sheep ileal PP B-cells display receptor (127; Mackay et al., 1989), anti-CD44 (25- many of the characteristics of a primary lymphoid 32; Mackay et al., 1988), and anti-CD45R (Mackay tissue and should display a high corticosteroid et al., 1987) mAbs were obtained from hybridomas sensitivity relative to other gut-associated lym-maintained at the Basel Institute for Immunology, phoid tissues.Previous analyses of the V-J junc-Basel.The anti-vimentin (VIM13.2)mAbwas pur- tional diversity of rearranged Ig light-chain (LC) chased from Sigma and the rabbit anti-human CD3 genes suggested that the B-cell population in each was purchased from Dakopatts (Glostrup, Den- ileal PP follicle was oligoclonal, arising from a mark).limited number of B-cell immigrants during fetal development (Reynaud et al., 1991).In vivo (Rey-Animals and Dexamethasone Treatment nolds, 1986) and in vitro (Griebel and Ferrari,  1994) analyses of iPfB-cell production and death All experiments were conducted using 4-5-week- further support the idea that a closed population old suckling lambs or 144-day gestation (148-day of iPfB-cells is maintained by self-renewing pro-gestation period) fetuses, of either sex (Versuchbeliferation. Therefore, it was postulated that if trieb Sennweid, Olsburg, Switzerland).Dexametha- corticosteroid treatment induced B-cell death and sone was dissolved in dimethysulfoxide (Fulka arrested proliferation, then follicular involution Chemie, Fuchs, Switzerland) and prior to intrave- should follow.A dexamethasone-treatment proto-nous (iv) injection diluted to a final concentration of col was developed that induced thymic involution 1 mg/ml in 37C phosphate-buffered saline (PBS).and arrested B lymphopoiesis in the ileal PP of Six groups of lambs were used for dexamethasone young sheep.The consequences of dexamethasone treatment: 3 lambs were used to investigate the treatment were then compared for the ileal and response of ileal PP follicles and the thymus to 0.02, jejunal PPs that contain functionally and pheno-0.2, and 2.0 mg dexamethasone/kg body weight typically distinct B-cell populations (Hein et al., (BW) injected for 3 consecutive days; 3 lambs were 1989; Griebel and Ferrari, 1995).

used to evaluate the response of ileal PP follicles and the thymus to dexamethasone following daily injec- tions of 2 mg/kg BW for 3, 5, or 7 days; 4 groups of 4 lambs were used to evaluate the responses of the PPs, thymus, and blood lymphocyte populations following daily injections of 2 mg/kg BW dexa- methasone for 7 consecutive days.One group of 4 lambs was used to study the responses of lymphoid tissues during each of the following posttreatment intervals: days

10; days
1-28; days 1-52; and days 1-98.

of 20 mg/kg BW 30 min prior to collecting tissues.This procedure resulted in a detectable level of BrdU incorporation in 40-45% iPfB-cells and 8-10% of thymocytes (Griebel and Ferrari, 1995).Immuno- peroxidase detection of BrdU incorporated in tissue sections was performed as previously described (Griebel and Ferrari, 1994).


RESULTS

Tissue Collection, Cell Isolation, Immunohistochemistry (IHC), and Flow Cytometry Blood collected in EDTA was used to determine total white cell counts, differential counts of leukocytes, and to isolate mononuclear cells with a discontinu- ous Percoll gradient (Griebel and Ferrari, 1995).Cell suspensions were prepared from lymphoid follicles of the PP and other tissues as described previously (Griebel and Ferrari, 1995; Griebel et al., 1994).Tissues for histology were first fixed in phosphatebuffered formaldehyde (12 %) prepared in methanol and then dehydrated in graded ethanol before em- bedding in Technovit 7100 medium (Heraeus Kulzer, Wehrheim, Germany).Tissue sections, 1-1.5 m thick, were mounted on precleaned glass slides, heated at 70C for 1 hr and then stained for 3 min with 1% threonine-acetate (Fluka) prepared in dis- tilled H20.Tissues for IHC were placed in cryo- molds (Tissue-Tek II; Lab-Tek Products, Nunc Inc., Naperville, IL) and mucosal surfaces were covered with a thin slice of liver before embedding in O.C.T. compound (Miles Lab.Inc., Naperville, IL) and freezing on dry ice.The methods for indirect label- ing of cell suspensions for flow cytometric analysis (FACScan; Becton Dickinson, Mountain View, CA), cell sorting (FACStar Plus, Becton Dickinson), and in

rect immunoperoxidase staining of
frozen tissue sections have previously been described in detail (Griebel et al., 1994; Griebel and F

rari, 1995).To quantitate lymphocyte subpopulations in blood,
the total number of blood mononuclear cells/ml blood was multiplied by the percent mononuclear cells labeled by the appropriate mAb and detected with flow cytometric analyses.


BrdU Incorporation and Detection

BrdU was dissolved in 60C PBS, cooled to room temperature, and injected iv at a final concentration Dexamethasone-Induced Involution of Primary Lymphoid Tissues

Preliminary experiments were completed to deter- mine if dexamethasone induced involution of primary lymphoid tissues in young lambs.The thymus was used as a control organ because of its well- characterized corticosteroid sensitivity in mice (Ishidate and Metcalf, 1963; Clamen et al., 1971).The effect of dexamethasone treatment on the thymus and ileal PP was first evaluated with 0.02, 0.2, and 2 mg dexamethasone/kg BW administered for 3 consecutive days.Ileal PP histology and thymic weights were evaluated with tissues collected 24 hr after the last treatment.A marked reduction in thymic weight (40-60% decrease) and ileal PP follicular size and cellularity was observed at all doses of dexamethasone, but with 2 mg/kg, few follicular B cells were seen on tissue sections (data not shown).Three lamb were then injected with 2 mg dexamethasone/kg BW for 3, 5, and 7 days, and tissues were collected 24 hr after each treatment and 30 min after injecting BrdU.Few BrdU cells were detected in ileal PP follicles following dexametha- sone treatment for 3 days, and no detectable BrdU incorporation was observed following the 7-day treatment (data not shown).Thus, a 7-day treatment with 2 mg dexamethasone/kg BW was chosen to study the long-term effects of arrested iPfB-cell proliferation.This dexamethasone treatment regime resulted in a marked reduction in thymic cortex with a relative increase in the medullary region (Fig. l b), but did not arrest proliferation of cortical thymocytes (Fig. 2b).Thymic weights for untreated, age- matched lambs were 58.6 + 7.2 g (mean + S.D. of values from 5 lambs), but during the first 2 weeks postdexamethasone, the average thymic weights were 12.8 + 3.2 g (n 5 lambs).The effects of dexamethasone on thymic architecture and thymocyte proliferation were no longer evident 4-5 weeks posttreatment (Fig. 2c).FIGURE 1.The histology of the thymus, jejunal PP, and ileal PP are compared for age-matched untreated (control) lambs and lambs treated for 7 days with 2 mg dexamethasone/kg BW (dex).Thymus: a (control) and b (dex) with the cortex (c) and medulla (m) indicated.Jejunal PP: c (control) and d (dex) wi

the lymphoid follicles (f), dome region (d)
interfollicular area (if), and lymphatic sinuses (s) indicated.Ileal PP: e (control) and f (dex) with lymphoid follicles (f), dome region (d), interfollicular area (if), and lymphatic sinuses (s) indicated.Ileal PP follicle: g (control) and h (dex) with the follicular capsule (cs) and "nests" of lymphocytes (closed triangles) distributed among nonlymphoid stromal cells (open triangles) indicated.Magnification: a-f (bar [in a] 300 m); g and h (bar [in h] 20 gm).


Depletion of Ileal and Jejunal PP Follicles

Normal ileal PP follicles are enclosed by a thin fibrous capsule (2-3 cells thick), and in the outer follicle, there are numerous "nests" of lymphocytes distributed among nonlymphoid stromal cells (Fig. l g).The majorityof lymphocytes in the outer folli- cle incorporate BrdU during a 30-min period (Fig. 2g).Dexamethasone treatment for 7 days resulted in a dramatic decrease in ileal PP follicular size (>90%) that was also associated with a reduction in the size and cellularity of the dome and interfol- licular regions (Fig. lf).Follicular remnants were suspended within dilated lymphatic sinuses (Fig. lf) and no cell proliferation could be detected in the follicular remnants following BrdU injection (Fig. 2h).The capsule of involuted follicles was thickened and disorganized, lymphocytes were absent within the follicle, but there were many stromal cells with large nuclei and abundant, vacu- olated cytoplasm (Fig. lh).IHC confirmed that most cells within the follicular remnants were mes- enchymal cells (Fig. 3a), including numerous mac- rophages (Fig. 3b).A small number of CD4 T- cells (Fig. 3c) and CD45R cells (Fig. 3d) were also detected within follicular remnants.The IHC stain- ing for sIgM B-cells was difficult to interpret be- cause an intense reticular staining pattern, suggestive of extracellular Ig, was observed within follicular remnants.Dexamethasone treatment also reduced the size and cellularity of the jejunal PP follicles but did not deplete intrafollicular lymphocytes (Fig. l d) and did not arrest the proliferation of these cells (Fig. 2e).Thus, dexamethasone treat- ment selectively depleted most iPfB-cells, but many stromal cells, macrophages, and some CD4 T- cells survived.

FIGURE 2. Cell proliferation, detected by BrdU incorporation, was analyzed in the thymus, jejunal PP, and ileal PP for age-matched, untreated lambs (control) and for lambs day (D.lpostdex) and 3 weeks (D.21postdex) after treatment for 7 days with 2 mg dexamethasone/kg BW.Thymus: a (control), b (D.lpostdex), and (D.21postdex) with the cortex (c) and medulla (m) indicated.BrdU incorporation was evident in the cortex both before and after dexamethasone treatment.Jejunal PP: d (control) and e (D.lpostdex), and f (D.21postdex) with the lymphoid follicles (f), dome region (d), and interfollicular area (if) indicated.BrdU inc

poration was prominent in the crypt epithelium and lymphoid
ollicles both before and after dexamethasone treatment.Ileal PP: g (control), h (D.lpostdex), and (D.21postdex) with lymphoid follicles (f) dome region (d), interfollicular area (if), and muscularis (ms) indicated.No BrdU incorporation was detected within the follicles following dexamethasone treatment, but BrdU incorporation in the crypt epithelium remained unchanged.Magnification: a-i (bar [in a] 350 tm).Differential Regeneration of Ileal and Jejunal PP Follicles Ileal PP follicles are characterized by a high mitotic rate, 15-20x greater than that of the thymus (Reynolds, 1987).This B-cell proliferation occurs primarily in the outer follicle with few cells proliferating in the central follicle, dome region, or inter- follicular region (Fig. 2g).Thus, BrdU incorporation FIGURE 3. Immunohistochemical characterization of cells in ileal PP lymphoid follicles following treat- ment with 2 mg dexamethasone/kg BW for 7 days.a: The stain for mes- enchymal cells (vimentin) revealed a network of cells in the lamina propria of the villi, numerous cells within the dome (d) and interfollicular region (if), the majority of cells within the follicular remants (f), and cells scat- tered throughout the muscularis (ms).b: The stain for macrophages (Du2- 66) revealed scattered cells (solid tri- angles) in the dome region (d) and many cells (solid triangles) scattered throughout the follicular remnants (f).c: The stain for CD4 T cells revealed cells scattered throughout the lamina propria, interfollicular re- gion (if), dome (d), and rare cells (solid triangles) within the follicular remnant (f).d: The stain for B cells (CD45R) revealed numerous cells within the dome (d), few cells in the interfollicular region (if), and rare cells (solid triangles) surviving within the follicular remnant (f).Magnifica- provided a sensitive method to detect B-

ll production in ileal PP follicles.There was no d
tectable BrdU incorporation in the ileal PP follicular rem- nants for 3 weeks after dexamethasone treatment (Figs. 2h and 2i).IHC analysis of 10-20 serial sections did not reveal BrdU incorporation in the follicles despite consistent BrdU incorporation in mucosal crypt epithelium and cells in the interfolli- cular region.Thus, dexamethasone induced a com- plete arrest of iPfB-cell proliferation, but small foci of proliferating iPfB-cells reappeared between 28-35 days postdexamethasone (Fig. 4a) and the size of this proliferating population increased during the following 2 months (Figs 4b and 4c).However, the regenerating follicles were heterogeneous in size and misshapen when compared with ileal PP folli- cles from age-matched control lambs (Fig. 2g) and there was no further increase in follicular size between days 54-98 posttreatment (Fig. 4c).BrdU incorporation in normal jejunal PPs revealed a high level of cell proliferation in mucosal crypt epithe- lium and the lymphoid follicles, but few cells proliferating in the dome or interfollicular area (Fig. 2d).Dexamethasone treatment did not inhibit cell proliferation in the crypt epithelium or the lymphoid follicles despite a marked reduction in follicle size (Fig. 2e).Following dexamethasone treatment, the jejunal PP follicles rapidly increased in size and FIGURE 4. Cell proliferation, de- tected by BrdU incorporation, in the ileal and jejunal PPs following treat- ment for 7 days with 2 mg dexamethasone/kg BW (postdex).Ileal PP: a (D.32mpostdex); b (D.54-postdex); and c: (D.98upostdex).BrdU incorporation was consistently detected in villus crypt epithelium and scattered cells in the interfollicu- lar region (if).Foci of proliferating cells were detected in follicles (f) on D.32 posttreatment, but the size of this proliferating population did not increase beyond D.54 posttreatment.The muscularis (ms) is indicated be- neath the follicles.Jejunal PPs: d (D.32Mpostdex); e (D.54mpostdex); and f: (D.98--postdex).A high level of BrdU incorporation was evident within the lymphoid follicles (f) at all times posttreatment and was associ- ated with a gradual increase in follicle size.The pattern of BrdU incorpora- tion in the villus crypt epithelium, interfollicular region (if), and dome region (d) remained constant.Magni- fication: a-f (bar [in a] 350 tm).cellularity (Figs 2f, 4d, e a d f) until they were similar to follicles in untreated lambs (Fig. 2d).


B-cell Phenotype and Function in Regenerating PPs

The marked variation in follicle size in the regenerating ileal PP (Figs 4a, b, and c) suggested that an altered B-cell population was developing within the follicles.Therefore, we examined the normal varia- tion in ileal and jejunal PP follicular size us ng a method to release intact follicles from PPs of young lambs (Griebel and Ferrari, 1995) and by examining serial tissue sections of PPs from a 144-day fetal lamb injected for 1 hr with 20 mg BrdU/kg BW.From these analyses, it was apparent that lymphoid follicles in the ileal and jejunal PPs of untreated, 6-10-week-old lambs (Figs 5a and 5b) and a fetal lamb (Figs 5c and 5d) varied 4-5-fold in size and also varied in shape.Similar size and shape varia- tion was observed with both histology and mechan- ically isolated follicles and the variation in follicle size occ rred despite cell proliferation in all fetal follicles (Figs 5c and 5d).Thus, variable follicle size in the regenerating ileal PP may reflect the normal variation in PP follicular size and the process(es) responsible for variable follicle size during fetal development.The phenotype of cells isolated from FIGURE 5. Lymphoid follicles in the PPs of foetuses and young lambs varied markedly in size.A: Lymphoid follicles (F) mechanically isolated from the ileal PP of a 6-week-old lamb display considerable variation in size and shape that was independent of the presence or absence of an attached dome region (D).Follicles viewed with phase-contrast micros- copy.B: Lymphoid follicles (F) iso- lated from the jejunal PPs display a similar variation in size and shape independent of an attached dome re- gion (D).Follicles viewed with phasecontrast microscopy.C: All follicles (F) in the fetal (144-day gestation) ileal PP had incorporated BrdU but varied in size.BrdU incorporation was also evident in the crypt epithe- lium of the mucosa (m).These obser- vations were based on the examination of 45 serial tissue sec- tions stained for BrdU.d: Fetal

ejunal PP displayed consistent BrdU
ncor- poration within follicles (F) and crypt epithelium (m) but a marked varia- tion in follicular size and shape.regenerating ileal and jejunal PP follicles was also analyzed.These analyses did not reveal significant differences between cells in regenerating follicles and cells isolated from PP follicles of untreated age-matched controls (Table 1).

To evaluated if dexamethasone treatment altered B-cell function, we examined the production of IgA and IgG1 PCs in the PPs.The jejunal PPs contain numerous IgA PCs in the lamina propria adjacent to villus crypts, in the interfollicular region, the dome FACS analyses of cell suspensions prepared from lymphoid follicles, Data presented the S.D. of values from 5, 6-12-week-old lambs.CData presented the S.D. of values for lambs analyzed between days 42-98 posttreatment.Data pooled after analyses showed significant differences, The CD40 molecule detected using soluble CD40 ligand-CD8 detected with PE-conjugated rat anti-mouse CD8.Values the /o sIgM B cells that coexpressed CD5.The B-cell population defined by subtracting CD5 cells (T-cells) from the total CD44 population.The light-scatter gates used for data collection excluded macrophages and stromal cells from the analyses.region, and within the follicle (Fig. 6d).IgA PCs are absent in ileal PP follicles, but otherwise have a distribution similar to the jejunal PPs (Fig. 6a).

Following dexamethasone treatment, there was no change in the distribution or apparent frequency of IgA PCs in either the ileal (Figs.6b and 6c) or jejunal PPs (Figs. 6e and 6f; Table 1).Similar observations were made for IgG1 PCs that were located primarily in the dome region, in the follicles, and in the interfollicular regions of the jejunal PPs and within FIGURE 6. Immunoperoxidase staining for IgA revealed a similar distribution of IgA PCs in the PPs in age-matched, untreated lambs (control) and lambs treated for 7 days with 2 mg examethasone/kg BW (postdex), Ileal PP: a (control); b (D.21--postdex); (D.74--postdex).IgA PCs in the ileal PP were located primarily in the lamina propria, adjacent to the villus crypt, and rare PCs are seen in the interfollicular (if) and dome (d) region but never in follicles (f).Jejunal PPs: d (control); e (D.21--postdex); f (D.74--postdex).IgA PCs in the jejunal PP were numerous in the lamina propria, adjacent to the villus crypt, and numerous in the dome region (d), in the follicle (f), and the interfollicular region (if).f(i) inset: Higher magnification of area outlined in f shows intense cytoplasmic staining of PCs in the lamina propria and surface staining of the mucosal epithelium.Magnification: a-f (bar [in a] 150 m). the dome and interfollicular region of the ileal PP (data not shown; Table 1).The observations on follicle structure, B-cell phenotype, and PP function indicated that the distinct functions performed by ileal and jejunal PPs remained intact following dexamethasone treatment.

Depletion of T Lymphocytes in Blood An acute depletion of greater than 50% blood T-cells began during and progressed for 1 week following dexamethasone treatment (Fig. 7b).This T lymphopenia involved all T-cell subsets and persisted for approximately 6 weeks.The recovery of circulating T-cell numbers to pretreatment levels was influenced most by an increase in the predominant T6 TcR-cell population.This occurred between 6-8 weeks posttreatment and followed thymic re- generation by approximately 3 weeks (Fig. 7).Flow cytometric analysis of CD44 expression on C

hi cells (
-cells) indicated that thymic involution and regeneration were linked to changes in blood T-cell populations.The subpopulation of CD44]CD5hi cells disappeared during dexamethasone treatment (Fig. 8; day 01 postdexamethasone) and did not reappear until 6 weeks posttreatment (Fig. 8; day 40 postdexamethasone).The reappearance of CD44]CD5hi cells followed thymic regeneration (Fig. 7a) and preceded the recovery of circulating T-cells (Fig. 7b).Thus, the temporal order of these events was consistent with the thymus regenerating the circulating T-cell population.


Depletion of B Lymphocytes in Blood

Dexamethasone treatment induced an acute and prolonged decline in blood B-cell numbers and the composition of this B-cell population was markedly changed (Fig. 9).In young lambs, approximately 90% blood B-cells (CD40 /) express sIgM (Fig. 9a) and less than 10% sIgM/B cells express detectable levels of CD5 (Fig. 8; pretreatment).One week after dexamethasone treatment less than 20% sIgM/B cells remained and >90% B cells expressed CD5 (Fig. 8; day 01 postdexamethasone).Furthermore, the relative contribution of isotype-switched B cells to the total B-cell population increased to 50% following dexamethasone treatment (Figs. 9a and   9b).The number of circulating sIgG1 and sIgA B cells remained relatively constant following dexam- ethasone treatment, but these B-cell populations varied widely among individual animals (Fig. 8b).FIGURE 7. Dexamethasone treatment (dex) induced a marked decrease in thymic weight and the number of T lymphocytes in blood, a: Thymic weights were 20-35% of the normal weight immediately after dexamethasone treatment (day 0), but 4-5 weeks later there was a complete recovery of thymic mass.Each data point represents the weight of a single thymus, as a percentage of BW, and the range of normal thymic weights is depicted by the rectangle, b: The total number of T lymphocytes (CD5hi) and all T-lymphocyte subsets (CD4, CD8, 73 TcR [gamma/delta]) declined markedly following dexamethasone treatment.The total number of T lymphocytes returned to pretreatment levels 8 weeks after dexamethasone treatment with 73 T lymphocytes contributing the most to the T-lymphocyte population.Data presented are the mean + S.D. of values from varying numbers of lambs (days -7 and 0: n 16; day 98: n 2).

Specificity of the STla mAb for CD5 on B cells was assessed in two ways.First, CD5hiDu2-104 cells and CD5Du2-104 cells were sorted with a FAC- Star cell sorter and cytospots prepared of the two populations.IHC analyses of cytospots revealed that CD5hiDu2-104 were 98% CD3 (T-cells) with no detectable staining for IgM or Ig LC.In contrast, FIGURE 8. Phenotypic analysis of B and T lymphocytes present in blood before and after dexamethasone treatment (2 mg/kg for 7 days).In young lambs (pretreatment), few B-cells expressed detectable levels of CD5, but immediately following dexametha- sone treatment (postdexamethasone: day 01), most sIgM B-cells expressed a low level of CD5.The predominance of CD5 IgM B-cells persisted for 14 weeks posttreatment (postdexamethasone day 98) with a minor subpopulation of CD5-sIgM B-cells.The changes in CD5 +IgM/B-cell number were reflected in the lo h CD5 CD44 population.In young lambs (pretreatment), T-cells h lo (CD5 ) were clearly divided into CD44 (32.4 + 3.8 '/o of T-cells) and CD44 hi subpopulations (66.9 + 5.7% of T-cells).During dexamethasone treatment, the CD44CD5 hi subpopulation (postdexamethasone day 01) was markedly reduced (7.4 + 1.9% of T-cells) and this reduction persisted for at least 3 weeks (postdexamethasone day 20).However, 6 weeks after dexam- ethasone treatment (postdexamethasone day 40), the relative lo h, frequency of the CD44 CDS subpopulation had recovered (38.6 + 4.7% of T-cells) to a level comparable to pretreatment lambs.The X axis presents sIgM expression (column 1) or CD44 expression (column 2) and the Y-axis presents CD5 expression for the dot scatter plots.

CD51Du2-104 cells were 2% CD3 /, 83% sIgM /, and 95% Ig LC /.Second, the STla mAb immuno- precipitated a 67-kD protein from iPfB-cells cocul- tured with J558L cells transfected with murine CD40 ligand, but no protein was immunoprecipi- tated from normal iPfB-cells (data not shown).Few iPfB-cells bind STla mAb (Table 1), but 60-70%

iPfB-cells bind a detectable level of the STla mAb following coculture with CD40 ligand (Griebel and  Ferrari, 1995).Thus, present experiments indicate that in blood, the most steroid-sensitive population of B-cells were CD5-sIgM and, in the absence of a functional ileal PP, this population remained se- verely depleted.A small number of CD5-sIgM B-cells were evident between 9-10 weeks posttreatment and this population increased a little during the next 4 weeks (Fig. 8; day 98 postdexamethasone; Fig. 9a).Thus, a return in iPfB-cell proliferation was followed 4 weeks later by the appearance of a limited number of CD5-sIgM B-cells in blood (Figs.4a and 9a).The prolonged B lymphopaenia was in marked contrast to the recovery in circulating T-cell number (Fig. 8).Finally, a transient neutro- philia during dexamethasone treatment was the only alteration in blood polymorphonuclear leuko- cyte, monocyte, and red blood cell populations observed throughout these experiments (data not shown).These observations, together with thymic regeneration, indicate that hematopoiesis in bone marrow was not altered by dexamethasone.Thus, it seems unlikely that altered B lymphopoiesis in the bone marrow could explain the prolonged B lymphopenia in blood.


DISCUSSION

The present experiments clearly demonstrated that, as expected for a population of immature lymphocytes, the B-cells in lymphoid follicles of the ileal PP were very corticosteroid-sensitive (Fig. 1).Dexa- methasone treatment completely disrupted the function of the ileal PP (Fig. 2h), but had relatively minor effects on the jejunal PP (Figs. 2e, 2f, 6e and 6f).Extensive follicular involution in the ileal PP after treatment with dexamethasone for 72 hr (data not shown) was consistent with a direct cytolytic effect on iPfB-cells.However, it remains to be determined if dexamethasone also acted indirectly by disrupting signals that supported iPfB-cell growth despite the survival of many mesenchymal cells in the follicular remnants (Fig. 3).Clearly, corticosteroid treatment inhibited much more effectively the T-cell-independent proliferation of iPfB- cells than the T-cell-dependent development of B-cells in the jejunal PPs (Griebel and Ferrari, 1994;  Griebel and Ferrari, 1995).Thus, B-cell development in these two PP was shown to be distinct.The absence of detectable iPfB-cell proliferation for a 4-week period did not result in a permanent arrest of iPfB-cell development (Fig. 4).Thus, sus- Days Post-Dexamethasone FIGURE 9.There was a sharp decline in total B-cell numbers in blood following dexamethasone treatment (2 mg/kg for 7 days) and this B lymphopenia persisted throughout the experiment, a: One week following dexamethasone treatment, the total number of B-cells (CD40 /) in blood had declined to 30% of pretreatment values.The sIgM B-cells (sIgM accounted for 90% of B-cells in blood prior to dexamethasone treatment, but less than 50% follow- ing dexamethasone tr atment.Finally, CD5 sIgM B-cells (CD5 sIgM were a minor population in young lambs, but more than 90% sIgM B-cells expressed a detectable level of CD5 fol- lowing dexamethasone treatment.A detectable level of CD5- sIgM B-cells were apparent 80 days posttreatment and then grad- ually increased, b: Dexamethasone treatment did not significantly alter the number of sIgG B-cells (sIgG or sIgA B-cells (sIgA in blood.Considerable variation in isotype-switched B-cell number was noted within and among experimental groups.Following dex- amethasone treatment, the isotype-switched B-cells comprised 50% of the total B-cell population.Data presented are the mean + S.D. of values from varying numbers of lambs (days -7 and 0: n 16; day 98: n 2; see Fig. 7a).

tained self-renewing proliferation is not essential to maintain the iPfB-cell population as was previously suggested by in vitro experiments (Griebel and  Ferrari, 1994).The involuted follicles may have been repopulated by B-cells surviving within the ileal PP follicles or dome region (Fig. 3d) or B-cells circulating in the blood (Fig. 9).If either intrafollic- ular or dome region B-cells were the source of the proliferating iPfB-cells, this would imply the exist- ence of a functionally distinct, steroid-resistant stem B-cell that can give rise to the rapidly dividing, steroid-sensitive iPfB-cells.Alternatively, it could b postulated that, as for the thymus, the bone marrow functions as the primary source of B-cell progenitors that then undergo further development in the unique microenvironment of the ileal PP follicles.

The present experiments could not determine the source of B-cells in regenerating ileal PP follicles.However, it was evident that the regenerating ileal PP maintained the unique function of this tissue.This conclusion was supported by the analyses of follicle structure (Fig. 5), B-cell phenotype (Table 1), and PP function (Fig. 6).Finally, the limited regen- eration of ileal PP follicles (Fig. 4c) may also reflect normal ileal PP development because ileal PP invo- lution usually begins at 3-4 months of age (Rey- nolds and Morris, 1983).

It is difficult to explain the prolonged quiescent period in B-cell production if the ileal PP follicles were regenerated by intrafollicular B-cells.The de- tection of BrdU incorporation in individual mucosal epithelial cells and interfollicular cells (Figs. 2h and  2i) indicated that limited sensitivity of BrdU detec- tion could not explain the absence of proliferating iPfB-cells.Alternatively, despite the survival of fol- licular architecture (Fig. 3), the dexamethasone treatment may have disrupted nonlymphoid stromal-cell functions th