Phenotypic Mapping of The Chicken Embryonic Thymic Microenvironment Developing Within an Organ Culture System

The chicken thymic microenvironment, as it developed in an embryonic thymus organ culture system, was phenotypically mapped using a panel of mAb defining both epithelial and nonepithelial stromal cell antigens. We have previously reported that thymocyte proliferation and differentiation will proceed for up to 6–8 days in thymus organ culture, hence demonstrating the functional integrity of the thymic microenvironment in vitro. During this time, the stromal component reflected that of the normal embryo with cortical and medullary epithelial areas readily identifiable by both morphology and surface-antigen expression. An abundance of subcapsular and cortical epithelial antigens was detected in the cultured thymus, particularly those normally expressed by the epithelium lining the capsule, trabeculae, and vascular regions (type epithelium) in the adult and embryonic thymus. Medullary epithelial antigens developed in organ culture, although were present in lower frequency than observed in the age-matched embryonic thymus. MHC class II expression by both epithelial and nonepithelial cells was maintained at high levels throughout the culture period. With increasing time in culture, the ratio of epithelial to nonepithelial cells decreased, concurrent with a decrease in thymocyte frequency and suggestive of a bidirectional interaction between these two cell types. Thus, a functionally intact thymic microenvironment appears to be maintained in embryonic thymus organ culture, a model that is currently being exploited to assess the role of stromal antigens, as defined by our mAb, in the process of thymopoiesis.


urring prior
o T-cell TcR expression on immature thymocytes, while equally important, are less well-defined.Clearly, there is a need to further define and characterize thymic stromal- cell subsets/surface antigens and their specific role in the various stages of T-cell differentiation.Monoclonal antibodies (mAb) raised against thymic stromal elements of the mouse, rat, and human have revealed a complex phenotypic pro- file, indicating the antigenically distinct regions within and between the basic cortical-medullary definition of the thymic stroma (Haynes, 1984;

INTRODUCTION

The thymus is comprised of a heterogeneous array of stromal cells and cytokines, both stromal-and thymocyte-derived, that constitute a microenvironment essential for the production of a competent T-cell repertoire (reviewed in   Boyd and Hugo, 1991; Carding et al., 1991).An essential step in T-cell differentiation involves the interaction between stromal cell MHC antigens and T-cell receptor (TcR) on developing mAb reactive with chicken stromal elements (Boyd et al., 1992), including MHC class II- specific reagents (Guillemot et al., 1984; Boyd et  al., 1992), have delineated a similar stromal het- erogeneity, demonstrating the conserved struc- tural and antigenic nature of the thymus between the avian and mammalian species.Amongst the mAb produced in our laboratory (Boyd et al.,  1992) are those specific for epithelium lining the capsule, trabeculae, and perivascular regions (type I epithelium, as defined by electron microscopy [van de Wijngaert et al., 1984]).Such reagents are distinct from those that label both the subcapsular and medullary epithelium described in various mammalian species (Haynes  et al., 1984; Colic et al., 1988; Godfrey et al., 1990;  Izon and Boyd, 1990), in addition to the chicken (Boyd et al., 1992).Takeuchi et al. (1991) have recently described a mAb reacting only with sub- capsular epithelium in the human thymus; how- ever, on close examination, it appears that this mAb also shows significant, although weak, reac- tivity with cortical epithelium.The type I epi- thelium-specific determinants facilitate further analysis of this region of the thymus and are of particular interest as they are selectively deficient in L200 chickens that develop autoimmune scleroderma (Boyd et al., 1991).

Thymic stromal cell-reactive mAb provide a means with which to not only map the thymic microenvironment, but to examine the functional contribution of mAb-defined thymic stromal sub- sets to thymocyte differentiation.One method by which this may be achieved is through addition of mAb to embryonic thymus organ culture (ETOC) and monitoring the subsequent effects upon thymocyte development.In a previous report (Davidson et al., 1992), we detailed the growth kinetics and phenotypic development of thymocytes in chicken ETOC.In this model, as for murine fetal thymus organ culture (FOTC)   (reviewed in Jenkinson and Owen, 1990), thymo- cyte proliferation and differentiation of both and ,dTcR lineage cells proceed in a manner par- alleling that in the embryo.It follows, therefore, that the essential components of the thymic microenvironment must develop and be func- tionally maintained in ETOC.In one of very few studies, van Vliet et al. (1985) have shown, through the use of specific mAb, that both corti- cal and medullary epithelial cells are represented in cultured fetal mouse thymus.Similarly, rat thymus fragments cultured in vitro are composed of cells reactive with mAb that normally stain either cortical or medullary epithelium (Kendall  et al., 1988).Hence, as a prelude to the functional assessment of thymic stromal subsets in ETOC, their in vitro development was mapped using a panel of mAb (Boyd et al., 1992) and compared to that in ovo.The phenotypic development of stro- mal subsets in the normal chicken embryo have recently been described (Wilson et al., 1992).


RESULTS

The results are categorized according to cell-type antigens recognized by the mAb: epithelial cel

antigens
antigens shared between epithelial and nonepithelial cells; and nonepithelial cell antigens (Tables 1-3 and Figs.1-3, respectively).Thymic epithelium-reactive mAb have, in some instances, been categorized according to CTES (Clusters of Thymic Epithelium Staining) guide- lines as defined in Kampinga et al. (1989).

Epithelial Cell Antigens (Table 1) All epithelial tissue (as defined by double-labe- ling with an anticytok ratin reagent) expressed an antigen defined by MUI-54 (CTES XX.a) dur- ing thymic development in ovo.Both of these panepithelial reagents defined cortical-and med- ullarylike epithelium in the cultured thymus, particularly from days 2-8.Subsets of flattened type I epithelium lining the subcapsule, subtra- beculae, and perivascular regions were defined by MUI-53 in the embryonic thymus, but showed an unusual distribution in ETOC.At days 4-6 ETOC, as much as half of the epithelium was stained with this mAb, including some indentifi- able subcapsular epithelium (Fig. la).By days 10-12, the majority of epithelium expressed this marker.Such reactivity revealed a lack of type I epithelial organization in the cultured thymus when compared to that observed in the normal embryo (Fig. lc).Isolated cortical epithelial cells of the embryo/adult thymus were defined by MUI-52 (Fig. l g).A marked increase in the fre- quency of epithelial expression of this molecule was observed in ETOC.Isolated epithelial cells stained with MUI-52 at the onset of culture (10E), approximately 50% by days 4-6 ETOC (Fig.   aThymus sections screened with each mAb at 2-day intervals from days 0-12 ETOC and from 10E to day posthatch.bmAb are classified according to CTES nomenclature where the appropriate category has been defined (Kampinga et al., 1989).CResults are similar between these time points and hence grouped for simplicity.dNo change in mAb distribution >14E but for increased extent with developmental age.Abbreviations: C: cortex; Ep: epithelium; ETOC: embryonic thymus organ culture; KNC: keratin-negative cells; M: medulla; PV: perivascular; SC: subcapsular; STb: subtrabecular.and the majority by days 10-12.Isolated, or small clusters of, medullary epithelial cells are defined by the mAb MUI-62 (CTES II) in the normal adult thymus, the pattern first observed beginning at 10-12E.These cells were present throughout the ETOC period, although at a reduced frequency compared to the in ovo thymus.A phenotypic relationship between subcapsular and medullary epithelium is demonstrated by the mAb MUI-58 (CTES V.c) in the adult thymus.A subset of these cells express this marker during embryogenesis, increasing to the adult distribution by hatching (Fig. lk).In ETO , MUI-58 reacted with epi- thelial cells, including subcapsular epithelium at days 2-4 (Fig. li), the frequency of reactive cells increasing with time in culture to encompass most epithelium by days 10-12.

The combined reactivity of these mAb and the anticytokeratin reagent, facilitated the mapping of epithelial cell development in ETOC.Cortical- and medullarylike epithelial areas, recognized by their morphology and antigenic profile, were clearly identifiable until day 6 ETOC (e.g., Fig.  lb).Thereafter, epithelial cell regions appeared to condense such that by days 10-12 dense epi- thelial areas were surrounded by keratin-nega- tive tissue, the latter comprising at least half of the lobe (data not shown).Additionally, it is interesting to note the reactivity of MUI-53 and MUI-52 with a subset of nonepithelial (keratin- negative) cells at this stage, staining that has not been previously noted in vivo.

Antigens Shared Between Epithelial and Nonepithelial Cells (Table 2) Scattered epithelial and nonepithelial cells (reticular fibroblasts and macrophages) in the cortex, medulla, and within the trabeculae and perivascular spaces are defined by MUI-66,  although reactivity is predominantly directed toward clusters of medullary epithelial cells (Fig. 2c).Such distribution was detectable from 14E through to the adult.At the onset of culture (10E), all epithelium expressed this marker and remained positive throughout the culture period, including both cortical-and medullarylike epi- thelium (Fig. 2a).The majority of keratin-nega- tive cells also strongly expressed this molecule in ETOC, present as isolated cells or small clusters within epithelial areas and as large areas around the epithelium.This nonepithelial reactivity expanded as the proportion of keratin-negative tissue increased with time in culture.In the embryo, isolated medullary epithelial clusters and keratin-negative cells scattered throughout the thymus were stained with MUI-72 and MUI- 80 (Fig. 2k), although, the keratin-negative reac- tivity of MUI-72 was restricted to the medulla, and MUI-80 stained isolated epithelial cells, showing extensive granular reactivity.In cul- tured thymus lobes, isolated, medullarylike epi- thelium and keratin-negative cells were recognized by these markers; however, the determinant defined by MUI-72 was not detectable until days 4-6 of culture.MUI-80 also showed a granular reactivity over much of the lobe (Fig. 2i).A monomorphic MHC class II determinant is recognized by MUI-78  (Boyd et  al., 1992), associated with isolated epithelial cells in the thymic cortex and the majority of epithelial and nonepithelial cells in the medulla of the adult (Fig. 2g).This pattern of expression was noted from 14-16E, and prior to this was associ- ated with isolated epithelial and nonepithelial cells, as cortical-medullary delineation is not morphologically defined until 14E.Up to day 6 ETOC, this distribution of MHC class II expression developed and was maintained m

h like that in the embryo (Fig. 2e).By days 10-12,   howeve
, the majority of epithelial and nonepi- thelial cells were MHC class II positive, and, in most cases, MUI-78 showed confluent reactivity over the entire lobe with a granular appearance characteristic of a secreted molecule (data not shown).

Nonepithelial Cell Antigens (Table 3) Connective tissue fibroblasts associated with the capsule, trabecular, perivascular spaces, within keratin-negative areas, and isolated cells in the medulla are recognized by MUI-56.At the onset of culture, there was a perithymic distribution of this connective tissue.Infrequent nonepithelial cells within epithelial areas also expressed this marker.At days 4-6 ETOC, MUI-56 / keratin- negative cells were present both within and around epithelial areas (Fig. 3a), similar to the embryo (Fig. 3b), and expanded to form at least half of the tissue by days 10-12.Macrophages (M(I)) primarily located in the adult thymus med- ulla, keratin-negative areas, and trabeculae are recognized by MUI-79.These cells were unde- tectable at the onset of ETOC, but present at low frequency by days 4-6, increasing by days 10-12.Medullary, but not cortical, vascular endo- thelium defined by MUI-82 in the adult and observed from 14-16E was absent throughout the ETOC period.

Thymocyte development was examined in cul- tured thymus sections in relation to the development of the stroma.MUI-36 reacted with B cells and a subset of thymocytes and M(I) located in the medulla of the adult thymus (Fig. 3f) (Boyd et   al., 1992; A. G. Bean, N. J. Davidson, H. A. Ward   and R. L. Boyd, in preparation).In the 10E thymus, isolated cells expressed this marker, located within and around the lobe.During the organ culture period, the majority of thymocytes expressed this marker (Fig. 3d), a significant increase over that in the embryo (peaking at 61% by day 8 ETOC in contrast to 28% at 16E; data not shown).A subset of M(I) was stained with MUI- 36 in ETOC; however, no B cells were detectable in this system as staining with goat anti-chicken Ig (reactive with heavy and light chain) was aThymus sections screened with each ,mAb at 2-day intervals from days 0-12 ETOC and from 10E to day posthatch.bResults similar between these time points and hence grouped for simplicity.CNo change in mAb distribution >14E but for somewhat increased extent with increasing developmental age.aKNC reactivity present from 16E.Abbrevia

ons: C: cor
ex; Ep: epithelium; ETOC: embryonic thymus organ culture; KNC/KNA: keratin-negative cells/areas; M: medulla; PVS: perivascular space; Tb: trabeculae.Ep and KNA aThymus sections screened with each mAb at 2-day intervals from days 0-12 ETOC and from 10E to day posthatch.bResults similar between these time points and hence grouped for simplicity.CNo change in mAb distribution 14E but for somewhat increased extent with increasing developmental age.

Abbreviations: C: cortex; Cap: capsule; CT: connective tissue; Ep: epithelium; ETOC: embryonic thymus organ culture; KNC/KNA: keratin-negative cells/areas; M: medulla; M(1): macrophages; PVS: perivascular space; Tb: trabeculae; Th: thymocytes.negative at all time points tested.MUI-83 is an mAb reacting primarily with thymocytes (95% from 16E) (Fig. 3j) and a subpopulation of peripheral T cells in the adult chicken (Boyd et al.,   1992; A. G. Bean, N. J. Davidson and R. L. Boyd,   submitted).MUI-83 / thymocytes were present at the onset of culture and expanded rapidly, peaking at approximately 85-90% by day 6 ETOC and were primarily located within the corticallike epithelial areas of the lobe, as in the embryo (Fig. 3h).In contrast to the embryonic thymus, the fre- quency of MUI-83 / thymocytes had decreased markedly by days 10-12 ETOC (50% compared to 96% at 20E) and they were scattered throughout the epithelial and nonepithelial (keratin- negative) areas of the lobe.Section staining for the CD3, CD4, and CD8 antigens demonstrated a similar developmental trend for thymocytes in ETOC, correlating with that observed previously (Davidson et al., 1992).


DISCUSSION

ETOC is a model whereby the functional signifi- cance of thymic stromal-cell antigens may be assessed by their presence (or absence), and, more definitively, by blocking molecular inter- actions through addition of purified mAb and assessing the subsequent effects on developing thymocytes.Such a model may also be used to analyze mechanisms of thymocyte selection with respect to antigens present during embryonic life.We have previously shown that thymocytes proliferate and differentiate in chicken ETOC in a manner analogous to that in the embryo (Davidson et al., 1992), implying that the stromal microenvironment is sufficiently intact and func- tional in vitro.As a prologue to identifying stro- mal antigens of functional importance, the main- tenance and development of stromal subsets defined by mAb was delineated in ETOC and compared to normal embryonic development.

Both cortical and medullary epithelial cells were present in ETOC for up to 8 days, similar to that in the embryo, as identified morphologically by the anticytokeratin reagent and MUI-54, the latter a pan epithelial marker in the thymus and possibly a marker of endodermal epithelium (Wilson et al. 1992).Furthermore, epithelial cells in ETOC expressed antigens normally present in the cortex and/or medulla of the embryonic thymus, although their distribution/organization was not always identical.Similarly, in murine FTOC, cortical and medullary epithelial cells have been detected using specific mAb (van Vliet et al., 1985; D. I. Godfrey, G. A. Waanders, D. J.

Izon, C. L. Tucek and R. L. Boyd, in preparation).Antigens present on type I epithelium, which normally lines the capsule, trabeculae, and peri- vascular spaces (the latter predominantly in the medulla), were present at a much higher frequency in ETOC compared to the embryo and lacked the organization seen in the latter.This disorganization, observed beyond day 4 ETOC, may be due to the lack of a functional vascular system, as demonstrated by an absence of medul- lary vascular endothelial cells defined by MUI- 82.During normal thymic histogenesis, mes- enchymal tissue accompanies vascular pen- etration of the epithelial rudiment, resulting in the formation of trabeculae and perivascular spaces (Le Douarin and Jotereau, 1975).It would appear that as these structures do not form in ETOC, the epithelium normally associated with them is not restricted to defined regions.The extensive expression of the normally infrequent stromal cell molecule detected by MUI-52 sug- gests either a high incidence of this cell type or that this molecule is upregulated in vitro.In sup- port of the latter, high levels of expression are also noted in epithelial monolayer culture (R. L. Boyd and T. W. Wilson, unpublished observations).Additionally, this marker is expressed by bursal follicle-associated epi- thelium from 15E to hatching, a time when lymphoid precursors are recruited to, and localize in, the bursal follicles (Wilson and Boyd,   1991).

Medullary epithelial components, defined by MUI-62, were present throughout the ETOC period, although at a lower frequency compared to the embryo.A similar trend was noted for epi- thelial cells/clusters defined by MUI-66, MUI-72,  and MUI-80, which normally include medullary epithelium in their distribution.This was particularly noticeable by days 10-12, when these molecules were largely restricted to keratin-nega- tiv cells.It may be, therefore, that medullary components are underdeveloped in ETOC and that the extensive reactivity seen with such mark- ers as MUI-52, -51, -53, and -70 is due to an over- representation of subcapsular/subtrabecular and cortical epithelium.Similarly, the sub- capsular/corticallike reactivity of MUI-58 is more frequent in ETOC.This mAb demonstrates the phenotypic relationship between subcapsular and medullary epithelium (Boyd et al., 1992), a relationship conserved between birds and mam- mals (Haynes, 1984; Colic et al., 1988; Godfrey et  al., 1990; Izon and Boyd, 1990).As the correct development of the medulla appears to require the presence of mature, TcR / thymocytes (Shores  et al., 1991), a low frequency of CD4 / mature cells in cultured thymus (Davidson et al., 1992) may result in underdevelopment of the medulla.Alternatively, epithelial cells not normally expressing cortical antigens may do so in ETOC.

Lampert and Ritte (1988) have proposed that both cortical and medullary epithelium are derived from a resident epithelial stem cell.MUI- 66 may be a marker for such stem cells as it is a pan epithelial marker at 10E, becoming restric- ted, with increasing developmental age, to iso- lated cortical epithelium and a subset of medul- lary epithelial clusters of a morphologically less differentiated nature, being round-bodied.It also stains the least differentiated epithelial cell layers in the skin (Boyd et al., 1992).The maintenance of its pan epithelial reactivity in ETOC suggests that epithelial maturation may be slowed.This is also consistent with the concept that cortical epi- thelium is of a simple or less differentiated form than medullary epithelium on the basis of expression of different isoforms of cytokeratin (Colic et al., 1989; reviewed Brekelmans and van   Ewijk, 1990) and mAb reactive with cortical epi- thelium showed increased reactivity/distribution in ETOC.

Additionally, MUI-66, -72, and -80, as well as MUI-56 (probably a marker of mesodermally derived tissue (Wilson et al., 1992), demonstrated the increasing proportion of keratin-negative cells in ETOC, which comprised at least half of the lobe by day 12, consisting of fibroblasts, and, to a lesser extent, M(I) and dendritic cells (Boyd et  al., 1992).Although M are present in the 10E thymus (Oliver and Le Douarin, 1984), nei her MUI-79 nor MUI-36, both reactive with a subset of M (as yet there is no chicken pan-macro- phage reagent available), showed a significantly increased frequency of M(I) reactivity over that seen in the embryo.The proportion of keratin- negative tissue in ETOC expanded with time in culture, often present as large, circular areas both within and around the epithelial regions, predominating by days 10-12 ETOC.Similarly, although to a lesser extent, keratin-negative cells, including reticular fibroblasts, were observed within and around epithelial areas in murine FTOC (van Vliet et al., 1985).

There was extensive expression of MHC class II molecules, as defined by MUI-78, in ETOC.Up to day 6 ETOC, MUI-78 showed cortical and medullary reactivity comparable to that in the embryo.In the later stages of culture, however, confluent reactivity was often seen over the entire cultured lobe area consisting of at least 50% nonepithelial cells.It is difficult to make a comparison with MHC class II expression in murine FTOC as reports are conflicting, some authors claiming it is low (Lo and Sprent, 1986), and others have observed strong development in vitro (Jenkinson et al., 1981).A granular staining pattern was often observed with MUI-78, sugges- tive of secreted MHC II.This may account for a twofold higher proportion of MUI-78 / thymocytes in ETOC (data not shown) that may acquire MHC molecules from their surrounds (Sharrow  et al., 1981).Whether the expression of MHC class II by most nonepithelial cells toward the end of the culture period is also due to passive acquisition is not known.It is possible that these nonepithelial cells, which are predominantly MUI-56 / fibroblasts, may express class II deter- minants in response to locally secreted cytokines such as ,-interferon, previously demonstrated to regulate the class II expression of cultured epi- thelial cells (Berrih et al., 1985), although there is as yet no direct evidence for the presence of -interferon in ETOC.

An antigen shared between stromal cells, B cells, and thymocytes is defined by MUI-36.Expression of this marker was upregulated on thymocytes during ETOC (half of which co- expressed TcR/CD3 complex; data not shown).Some reactivity could be attributed to M, but not to B cells, which do not appear to develop in ETOC.This suggests that either B-cell precursors were not present in the first wave of precursor cells entering the thymus at 6.5E (Coltey et al.,  1987) nd/or the lack of suitability of the embryonic thymus in promoting B-cell development and/or maintenance in vitro.Such data are also consistent with the paucity of B cells in the nor- mal embryonic thymus (N.J. Davidson and A. G.   Bean, unpublished data).

In summary, cortical and medullary epithelial and nonepithelial cell antigens, including MHC class II, developed and were maintained in ETOC, reflecting that in the embryo within a lim- ited time frame.However, the data suggest that epithelial maturation may be slowed in vitro, whereas conditions appear permissive for the maintenance of nonepit.helialcells, perhaps as their requirements are less stringent than those of epithelial cells.Concurrent with the decreasing proportion of epithelial areas and an increasing proportion of keratin-negative tissue was a decline in thymocyte frequency, demonstrated by MUI-83 reactivity, and as previously docu- mented using mAb reactive with standard T-cell determinants (Davidson et al., 1992).It has been similarly demonstrated during culture of rat thymic fragments that a type of dedifferentiation of epithelial cells is coincident with a depletion of thymocytes (Kendall et al., 1988).In this model, after days 6-9 of culture, all epithelial cells within the thymic fragment possess a uniform and unusual morphology that, following in vivo reconstitution of the fragment

rediffere
tiates" into distinct cortical and medullary compartments.It would be particu- larly interesting to determine if such a pattern of cellular reorganization would occur following grafting of the cultured chicken thymus into embryonic recipients.This data, combined with that presented here

, would suggest that an inti- m
te relationship exists between developing T cells and the maintenance and/or maturation of epithelial cells, a conclusion that is supported by the recent finding that normal T cells are able to restore the thymic stromal microenvironment of scid mice (Shores et al., 1991; Surth et al., 1992).

Beyond day 8 ETOC, the constraints of the sys- tem increasingly effect cellular development, particularly that of the epithelium.Such con- straints include a decreased efficiency of nutrient/gaseous e change with increasing lobe size (Jenkinson and Owen, 1990) and a lack of second-wave thymocyte and stromal precursors (M and dendritic cells) at 12E (Oliver and Le Douarin, 1984; Coltey et al., 1987), the interaction of which with the developing stroma appears necessary for continued growth, as has been demonstrated in scid mice (Shores et al., 1991).

Hence, with respect to both stromal (this study) and thymocyte development (Davidson et  al., 1992), thymopoiesis in chicken ETOC reflects that in the embryo within a limited time frame (0-8 days) and represents a useful model with which to examine the functional potential of thymic microenvironmental components.Further- more, the thymic development described in this study is reminiscent of that in murine FTOC (van  Vliet et al., 1985; D. I. Godfrey, G. A. Waanders, D. J. Izon, C. L. Tucek and R. L. Boyd, in   preparation) and consistent with the high degree of similarity of thymic events between the two species (reviewed in Cooper et al., 1991).This model is currently proving useful in identifying functionally relevant thymic stromal molecules and to address the issue of the symbiotic relation- ship between developing thymocytes and strom- al-cell subsets.


MATERIALS AND METHODS


Chickens

Closed colony Australorp/White Leghorn F1 hybrid chicks and embryos were obtained from Research Poultry Farm (Research, Australia).Chicks were kept under standard animal house conditions and eggs maintained in a humidified incubator at 39C, age estimated by the duration of incubation.

bryonic Thymus Organ Cultu
e

The method for ETOC has been previously detailed (Davidson et al., 1992).Briefly, thymus lobes from 10-day-old embryos were isolated under sterile conditions and 5-8 lobes cultured on 0.45/m polysulfone filter membranes (Gelman Sciences, Michigan) supported by gelatin sponge (Upjohn, Kalamazoo, Michigan) in RPMI-1640 tissue-culture medium (supplemented with 10% [v/v] heat-inactivated fetal calf serum and 2 mM L-glutamine) for 2-12 days in a humidified, 42C incubator.Medium was replaced after 6 days of culture.Tissue Section Staining Tissue sections of 10E thymus lobes cultured for 0, 2, 4,