Thymic Involution in Viable Motheaten (meυ) Mice is Associated with a Loss of Intrathymic Precursor Activity

Mice homozygous for the viable motheaten (meυ) allele manifest abnormalities in thymocytopoiesis, are severely immunodeficient, and develop autoimmune disorders early in life. Premature thymic involution occurs in meυ/meυ mice, and their bone marrow prothymocytes are unable to repopulate the thymus of adoptive recipients following intravenous (i.v.) transfer. However, analysis of thymocytopoiesis following intrathymic (i.t.) adoptive transfer of bone marrow from meυ/meυ mice demonstrates the presence of normal numbers of prothymocytes. To investigate intrathymic development in meυ/meυ mice, we determined intrathymic precursor cell number and activity. Dual labeling analyses showed that an involuted meυ/meυ thymus is relatively enriched (fivefold) in CD4– CD8– thymocytes (intrathymic precursor phenotype) compared with wild-type (+/+) thymus. However, thymocytes from meυ/meυ mice were deficient in precursor activity when adoptively transferred i.t. into irradiated recipients. Thymocytes recovered from the involuted thymus of aged or steroid-treated normal mice also displayed reduced precursor activity. However, the phenotypic profile of thymocyte subsets from steroid-treated mice was enriched in single positive cells (mature phenotype) and was distinctly different from the subset distribution of thymocytes in meυ/meυ and aged mice. These results suggest that intrathymic precursor activity in meυ/meυ mice is decreased, and may be reflective of decreased prothymocyte seeding to the thymus in vivo, In addition, the results suggest that the thymic involution in meυ/meυ mice is not due solely to effects of corticosteroids.


ormal periphe
al T-cell functions (Green and Shultz, 1975; Shultz et al., 1984;  Greiner et al., 1986; Shultz and Sidman, 1987;  Shultz, 1988).Furthermore, a number of auto- immune disorders develop that include hypergamma-globulinemia, autoantibody production, and immune complex glomerulonep,hritis (Kincade, 1981; Shultz and Sidman, 1987; Shultz,  1988).The usual cause of death is a characteristic hemorrhagic macrophagic pneumonitis that is *Corresponding author.Present address: The Jackson Laboratory, Bar Harbor, Maine 04609.believed to be autoimmune in origin (Shultz and  Sidman, 1987; Shultz, 1988).This autoimmune syndrome can be transferred to irradiated recipi- ents using meV/me bone marrow stem cells (Kincade, 1981; Shultz, 1988).

Developmental abnormalities are evident in cells that express the enzyme terminal deoxynucleotidyl transferase (TdT; presumptive Tand B-lymphoid stem cells) in me and me"/me bone marrow and thymus (Landreth et al., 1981;  Greiner et al., 1986).These bone marrow and thymus TdT/-cell developmental defects appear to be associated with the defects in thymocytopoiesis and may have a causal relationship to the peripheral Tand B-lymphocyte abnormalities.Moreover, at least some of the TdT /- (Landreth et  al., 1981; Greiner et al., 1986) and B-cell progenitor cell (McCoy et al., 1985; Greiner et al., 1986;  Kincade, 1987) developmental abnormalities, when analyzed in selective in vitro assay systems (Medlock et al., 1987; Hayashi et al., 1988), appear to be secondary to bone marrow micro- environmental defects.

Studies on thymocytopoiesis in me and meV/me mice have demonstrated a premature thymic involution, commencing as early as 4 weeks of age (Greiner et al., 1986; Shultz and Sid-  man 1987; Shultz, 1988).Furthermore, we have shown that both me and meV/me bone marrow prothymocytes fail to repopulate the thymus of irradiated recipients after intravenous (i.v.)   transfer, but generate normal numbers of thymocytes following intrathymic (i.t.) adoptive trans- fer (Greiner et al., 1986).Mixing of meV/me bone marrow with normal bone marrow restores the ability of meV/me prothymocytes to generate thymocytes after i.v.injection (Komschlies et al.,  1987), again suggesting that a marrow micro- environmental defect may underlie certain of the early developmental defects observed in lymphopoiesis in me/me and meV/me mice.

The apparent prothymocyte homing defect in me and meV/me bone marrow cells may have in vivo consequences, since postnatal mainten- ance of normal thymocytopoiesis is de endent upon a low, but continual, seeding of the thymus by bone marrow hemopoietic precursors (Scollay and Shortman, 1984; Scollay et al., 1986).Further- more, the ability of bone marrow precursors to generate thymoc}tes following entry into the thymus appears to be self-limiting, and will cease within 4 to 5 weeks following thymic entry (Goldschneider et al., 1986; Scollay et al., 1988).

Because me and meV/me bone marrow pro- thymocytes appear to be unable to repopulate the thymus of irradiated recipients following i.v.t ansfer, we have postulated that this defect may result in an early loss of intrathymic pre- cursors in me and meV/me mice, leading to premature thymic involution (Greiner et al.,   1986; Komschlies et al., 1987).

We have investigated this hypothesis in the present study using the i.t.adoptive transfer assay to quantitate irectly the intrathymic precursor activity prior to and during thymic invol- ution in meV/me mice.Our results indicate that the pool of intrathymic precursors in meV/me mice decreases early in life, in association with the onset of thymic involution.Furthermore, the thymic involution observed in meV/me mice does not appear to be similar to that induced follow- ing administration of corticosteroids.


RESULTS


Phenotypic Characteristics of me'/me Thymocytes During Thymic Involution

Investigations of murine

ubpopulations have revealed a large diversity of cell sub- sets (Scollay
t al. 1988; Wilson et al., 1988).Four major phenotypic populations have been exten- sively characterized, and are identified as CD4-CD8-(double negative, DN), CD4/CD8 / (double positive, DP), and CD4/CD8 or CD4-CD8 / (single positive, SP).Because the thymus of meV/me mice has been shown to undergo premature thymic involution, commencing as  early as 4 weeks of age (Greiner et al., 1986;   Shultz and Sidman, 1987; Shultz, 1988), we exam- ined the phenotypic characteristics of the thymocytes prior to and during this involution process.

As shown in Table 1 and Figure 1B, the loss of cellularity in the meV/me thymus is evident as early as 4 weeks of age compared with untreated control B6+/+ mice.The loss of cellularity, vari- able between 4 and 7 weeks of age, is accompanied bya loss in the absolute numbers of DN, DP, and SP thymocytes compared with weanling B6 wild-type mice, with losses of 95%, 99.5 %, and 98.4%, respectively, in the most severely involuted thymuses.The relative pro- portions of each unique thymocyte subset, how- ever, were not as severely altered (Fig. 1B).There was a relative increase in the proportion of DN thymocytes in the most severely involuted thy- mus, approaching 30% of the total thymocyte population as compared to that of approximately 5% DN cells in the thymus of control B6+/+ mice.In the severely involuted thymuses (less than 13x106 total thymocytes), the DP and SP populations were approximately equal, each comprising 33-49% of the remaining cells present, whereas in normal B6+/+ mice, the DP and SP populations normally comprise approxi- mately 80% and 15%, respectively, of a thymocyte population (Scollay et al., 1988).However, during the involution process, as thymus cellu- larity decreased, but prior to the most severely involuted state, the relative proportions of each B -..=


FIGURE

Dual labeling analysis of normal and involuted thymuses with anti-CD4 and anti-CD8 monoclonal antibodies. (A)

Normal thymus with a cellularity of 250106 cells.Percentage of thymocytes in each quadrant: (CD4/CD8"), 10.5%; 2 (CD4+CD8/), 81.3%; 3 (CD4-CD8-), r.4%; and 4 (CD4-CD8/), 4.8%. (B) Involuted me"/me thymus with a cellularity of 13x106 cells.

Percentage of thymocytes in each quadrant: 1, 28.3%; 2, 47.8%;.3, 14.0%; and 4, 9.9%. (C) Involuted thymus from an 18-month-old normal B6 wild-type mouse with a cellularity of 42106 cells.Percentage of thymocytes in each quadrant: 1, 9.0%; 2, 83.9%; 3, 3.9%; and 4, 3.2%. (D) Involuted thymus with a cellularity of 8x106 cells from a steroid-treated 4-week-old normal B6 mouse.Percentage of thymocytes in each quadrant: 1, 50.9%; 2, 12.3%; 3, 17.4%; and 4, 19.4%.

of the major thymocyte subsets approximated that observed in normal mice (Table 1).


Intrathymic Precursor Activity of me'/me Thymocytes During Thymic Involution

The relative proportion of DN thymocytes was increased in the involuted meV/me thymus (Table 1 and Figure 1B), although the absolute number of DN thymocytes was decreased (Table 1).The increased percentage of DN thymocytes may be reflective of a compensatory enrichment in intra- thymic precursors (i.e., DN thymocytes that pos- sess the ability to repopulate the thymus of irradiated recipients upon adoptive transfer).

Alternatively, the DN thymocytes remaining in the involuted thymuses of meV/me mice may be CD3 / DN thymocytes with no precursor activity upon adoptive transfer (Scollay et al., 1988).To investigate these various possibilities, we used the i.t.adoptive transfer assay (Goldschneider et  al., 1986; Scollay et al., 1988; Shortman et al.,  1988) to quantitate the intrathymic precursor activity of thymocytes in meV/me mice.This assay system was used because intrathymic precursors poorly repopulate the thymus following i.v.adoptive transfer (Kadish and Basch, 1977;   Fowlkes et al., 1985; Scollay et al., 1988),.andintrathymic precursor activity is readily quant- ified using the ,i.t.adoptive transfer assay (Goldschneider et al., 1986; Scollay et al., 1988).Thymocytes recovered from meV/me mice with involuted thymuses were relatively deficient, on a per cell basis, in their ability to repopulate the thymus of adoptive recipients as compared to thymocytes recovered from B6+/+ mice.The i.t.transfer of 1.2 to 1.9x106 thymocytes from meV/me mice with 2.1+1.0x106cells/thymus resulted in the generation of <1x106 (nondetectable) (n=6) donor-origin cells on day 14, and the transfer of 1x 06 or 2.5x106 thymocytes from B6+/+ mice resulted in the generation of 3.9+1.0x106 (n=4) and 8.0+4.3x106 (n=3) donor-origin cells, respect- ively, on day 14.In addition, it was of interest to compare the "total" precursor population in the thymus of meV/me mice with that O f normal B6+/+ mice during the involution process.To compare this directly, the relative intrathymic precursor activity of each thymus, which takes into account the total cellularity of the thymus, hence the total precursor pool within the thymus, was calculated (see Materials and Methods).

Thymocytes from 4-week-old me'/me mice clustered into two distinct groups when the rela- tive intrathymic precursor activity was calcu- lated (Fig. 2A).In one group, the relative intra- thymic precursor activity was equivalent to that of age-matched B6+/-littermates.In" the second group, the relative intrathymic precursor activity was decreased in relation to that of B6+/litter- mates (Fig. 2A).In 5-, 6-, and 7-week-old meV/me mice, the relative intrathymic precursor activity was reduced in relation to that of age-matched B6+/-littermates, approaching nondetectable levels of precursor activity in 60% (3/5) of 6- week-old meV/me ice and 100% (3/3) of 7-week-old meV/me v mice.No differences were observed in the relative intrathymic precursor activity of B6+/littermates between 4 and 7 weeks of age (Fig. 2A), which, as observed in bone marrow prothymocyte activity (Komschlies  et al., 1987), were comparable to that of wild-type B6+/+ mice (data not shown).Furthermore, in 3-week-old meV/me mice, we observed a decrease in intrathymic precursor activity prior to detect- able thymic involution, suggesting that the loss of intrathymic precursor activity precedes thymic involution (data not shown).

Due to the vari bility in intrathymic precursor activity observed in involuting meV/me thymuses, especially in 4-week-old meV/me mice, we deter- mined whether the extent of thymic involution in meV/me mice (loss in total cellularity) was related to the level of intrathymic precursor activity.As shown in Fig. 2B, there is a strong linear corre- lation (r2=0.872) between the cellularity of the meV/me thymus and the relative intrathymic pre- cursor activity detectable during involution; as thymic cellularity decreased, the amount of intrathymic precursor activity correspondingly decreased.

Phenotypic Characteristics of Thymocytes in Aged B6 Wild-Type Mice During Thymic Involution Motheaten (me and meV/me mice have expected lifespans of 22 and 61 days, respectively (Shultz and Sidman, 1987; Shultz, 1988), and we have previously suggested (Medlock et al., 1986)  that these mice may be representative of an accel- erated aging process in the immune system.Therefore, it was of interest to compare directly Age (Weeks)   Thymus Cell Number (xl0 FIGURE 2 Thymocytes from meV/me and age-matched B6+/-littermate mice (Ly 5.2) were tested for precursor activity following intra hymic adoptive transfer into irradiated B6-Ly5.1 congenic recipients.Relative intrathymic precursor.activitywas calculated as described (see Materials and Methods).Values represent data from individual thymuses. (A) Relative intrathymic precursor activity normalized to that of control B6+/-littermates as a function of age.Open squares represent me'/md thymuses and x's represent age-matched B6+/-littermate thymuses. (B) Relative intrathymic precursor activity as a function of thymus cellularity (r2=0.872).

the thymic involution of meV/me mice with that observed in normal mice during aging.

As observed in an meV/me involuting thymus, there is a significant decrease in the absolute number of each of the four major thymocyte sub- sets (DN, 76%; DP, 92%; SP 80%) in the severely involuted thymuses from aged normal mice as compared to thymocyte subsets present in wean- ling (4-6-week-old) B6+/+ mice (Table 2).However, the relative proportion of each of the major thymocyte subsets was consistent in aging B6+/+ mice through 14-15 months of age, although the cellularity of the older 14-16-month thymuses was decreased 60 to 90% with respect to that of weaning B6+/+ mice (Table 2).The staining profile of a typical older thymus (cellularity of 42x 106) is shown in Fig. 1C.

Intrathymic Precursor Activity of Thymocytes in Aged Mice During Thymic Involution Thymocytes from weanling (1-month-old) B6+/+ mice were comparable to that of weanling B6+/- littermate mice in intrathymic precursor activity (Figs.2A and 3A).By 11 to 13 months of age, however, the intrathymic precursor activity of B6+/+ mice was decreased by 57 to 87% as com- pared to that of 1-month-old B6+/+ mice and continued to decrease with time to essentially nondetectable levels (98% decrease) of intra- thymic precursor activity in 100% (4/4) of 16-17- month-old mice (Fig. 3A).As observed in involu- ting meV/me thymuses, the decrease in thymus cellularity exhibited a strong linear correlation (r2=0.900)with the loss of intrathymic precursor activity with age (Fig. 3B).

Prothymocyte Activity in the Bone Marrow of Aged Mice Viable motheaten (meV/mev) mice with severely involuted thymuses still contain normal numbers of bone marrow prothymocytes as quantitiated by the i.t.adoptive transfer assay system, but no detectable intrathymic precursor activity Greiner et al., 1986; Komschlies et al., 1987; Fig.  2).Therefore, we determined whether the loss of intrathymic precursor activity in B6+/+ aged aThymocytes from various aged C57BL/6J+/+ mice analyzed for expression of CD4 and CD8 antigens using two-color immunofluorescence staining (see Materials and Methods).Data represent individual thymuses except for the 1-month-old group that represents the percentage+standard deviation of three individual thymuses.bThe percentage of the CD4-CD8-subset excludes sIg cells in the thymus.

mice is associated with a decrease in bone mar- row prothymocyte activity.Bone marrow was recovered from 12-month-old B6+/+ mice (57 to 87% decrease in intrathymic precursor activity) and injected both i.v. and i.t.into young (4-6- week-.old)irradiated adoptive recipients to deter- mine their relative one marrow prothymocyte activity.As shown in Table 3, in contrast to that observed in meV/me mice and as shown pre- viously by other investigators (Hirokawa et al.,  1986), the bone marrow of aged mice is able to repopulate the thymus of young adoptive recipi- ents following i.v., as well as i.t., adoptive trans- fer.Furthermore, on a per cell basis, bone mar- row from 1-year-old donors does not differ significantly from that of weanling (1-month-old) bone marrow in relative prothymocyte activity.For the i.v.adoptive transfer system, the relative prothymocyte activity is 219+109 (mean+S.D.) units for young donors and 421+147 units for aged ones.Similar results are observed for the i.t.adoptive transfer system in which the relative prothymocyte activity is 10,410+7,563 units for young bone marrow and 16,245+8,189 units for aged marrow.

To investigate further the possible mechanisms that account for thymic involution in aged mice, especially since bone marrow precursor activity in older mice is comparable to that of young mice (Table 3), we determined the ability of young and older (12-month-old) bone marrow to repopulate the thymuses of older (10-month-old) irradiated recipients.Surprisingly, there were no significant differences (p>0.1 for the i.v.assay; p>0.5 for the i.t.assay) in the number of thymocytes generated by either young or old marrow on a per cell basis in older recipients as compared to that of young recipients.The relative prothymocyte activities calculated for both adoptive transfer assays are 477+127 units (i.v.) and 15,912+3,917 units (i.t.)   for young bone marrow and 372+58 units (i.v.) and 14,070+9,478 units (i.t.) f r aged marrow.Phenotypic Characteristics of Thymocytes in Corticosteroid-Treated Mice Following Thymic Involution To investigate whether the thymic involution observed in meV/me mice resembled that of corticosteroid-induced thymic involution, nor- mal B6+/+ mice were injected with dexa- methasone (0.5 mg), and the thymus analyzed 48 h later.Steroid treatment resulted in a 97% decrease in cellularity, and a decrease in the absolute numbers of each of the four major thymocyte subsets (Table 4).The proportions of each of the thymus subsets that were present, however, were different (Fig. 1D) from that observed in an meV/me involuted thymus (Fig. 1B).The proportion of DN thymocytes increased to 19.3+3.1%,similar to that of meV/me thymo- cytes (range 5.5-32%), but the perc

ursor Activ
ty of Thymocytes in Steroid-Treated Mice

Thymocytes recovered from steroid-treated B6+/+ mice were tested for their intrathymic precursor activity in the i.t.adoptive transfer assay.

The thymocyte repopulating activity of steroid- treated thymocytes (1.1+0.1 units) was decreased more than 99% when compared to the relative thymocyte repopulating activity of age-matched untreated B6+/+ (392.2+185.8units) mice.

Prothymocyte Activity in the Bone Marrow of


Steroid-Treated Mice

The high sensitivity of intrathymic precursors to steriods was surprising in light of previously published results detailing the relative steroid resistance of bone marrow prothymocytes (Greiner et al., 1982).To confirm this result, bone marrow cells were recovered from the same steriod-treated mice that were used as donors of thymocytes (see Table 4), and were injected both i.t. and i.v.into adoptive recipients.As expected, on a per cell basis and as previously reported injected with dexamethasone (0.5 mg) and the thymuses recovered for analysis 48 h later.Thymocytes from these mice analyzed for expression of CD4 and CD8 antigens using two-color immunofluorescence staining (see Materials and Methods).Data represent the percentage+standard deviation for four mice in the dexamethasone-treated group and for three mice in the untreated group.bThe percentage for CD4-CD8-thymocytes excludes surface Ig cells in the thymus. (Greiner et al., 1982), the number of thymocytes generated by equivalent doses of untreated and steroid-treated bone marrow was essentially equal (Table 5).However, because the cellularity of the steroid-treated bone marrow was reduced approximately 60%, the total number of prothymocytes in steroid-treated marrow was approximately 40% of that of normal marrow.

Although variable, when the same bone marrow was analyzed by the i.t.assay, similar results

were observed.By using a standard cell dose of 2.0x105 cells/recipient, the total prothymocyte activity per leg (femur plus tibia) detectable in the marrow of steroid-treated mice was approximately 30% of that of untreated mice (Table 5).


Expression of CD3 and Surface

Immunoglobulin in the DN Populatio The DN thymocyte population is the only thymo- cyte subset demonstrated to be able to repopulate the thymus of adoptive recipients in both the i.v. and i.t.adoptive transfer systems (Scollay et al.,  1988).However, CD4-CD8-thymocytes that express CD3 do not generate thymocytes upon i.t.injection into adoptive recipients (Crispe et  al., 1987; Scollay et al., 1988; Shortman et al.,  1988).In addition, approximately 1% of cells in the normal thymus are surface Ig / B lymphocytes (Andreu-Sanchez et al., 1990).Surface Ig / cells could be contaminants in the DN (CD4-CD8-) population, and would also fail to generate thymocytes upon i.t.adoptive transfer into irradiated recipients.Therefore, DN thymocytes that expressed either CD3 or sIg were quantit- ated in the thymus of normal B6+/+ 4-6-week mice as well as in involuted thymuses (Table 6).

Even though the percentage of DN thymocytes is greater in the severely involuted thymuses (<25x 106 thymocytes), the proportion of CD3 / or sIg / DN cells approximates that observed in normal thymuses (approximately 50%).


DISCUSSION

In the present study, we have demonstrated that meV/me involuted thymuses display a loss in all four of their major phenotypic thymocyte subse s (DN, DP, SP) by 4 to 7 weeks of age.Prior to this loss of thymocytes, a decrease in the intrathymic precursor activity was observed.This decrease was evident as early as 3 weeks of age in meV/me mice whose thymuses have not lost cellularity.At 4 weeks of age,, approximately half of the meV/me mice examined had reduced intrathymic precursor activity, and the majority of the ani- mals studied at 6 to 7 weeks of age demonstrated reduced activity.Although a relative increase in the percentage of DN thymocytes was observed in the involuting thymus of meV/me mice, there was a decrease in the absolute number of DN cells relative to that of normal age-matched B6+/+ mice.Because only DN thymocytes are able to generate the other three phenotypic thymocyte subsets (DP and SP) in vivo in adoptive recipients (Scollay et al., 1988; Shortman et  al., 1990), our observations would suggest hat the intrathymic precursors, although reduced in absolute numbers, may be relatively enriched in involuting meV/me thymuses.However, the DN thymocytes that were present in the severely involuted meV/me thymuses were deficient in their ability to generate thymocytes upon trans- fer to irradiated adoptive recipients.

It has been calculated that as few as lx103 puri- fied DN thymocytes can generate detectable levels of donor-origin thymocytes in an irradiated recipient's thymus, while up to 6x105 DN thymocytes (2x107 unfractionated thymocytes) are required for saturation repopulation of the thymus of an irradiated recipient (Scollay et  al., 1988).We injected in our experiments at least 5x105 thymocytes from each involuted thymus into the irradiated recipient's thymus.Because, in the severely involuted thymus, at least 20% of the thymocytes were DN (Table 1), at least 100,000 DN cells were injected into the thymus of each of the adoptive recipients.Our failure to find evidence of intrathymic precursor activity following injection of approximately 100-fold more DN thymocytes than should have been necessary to generate thymocytes in adoptive recipients (and approximately fourfold more DN thymocytes on a per cell basis than that present in normal thymus that gave high levels of repopulation in adoptive recipients) suggests that at a maximum, less than 1% of the DN thymocytes in the severely involuted meV/me thy- muses could have had precursor activity.

The DN thymocytes that are present in normal mice, however, are composed of at least 11 differ- ent subsets, each with different functional capabilities and kinetics of thymocyte repopulation in adoptive recipients (Scollay et al., 1988).Also contained within these various thymocyte sub- sets are DN thymocytes that express either CD3 or surface Ig and are unable to function as intra- thymic precursors in adoptive transfer assay systems (Scollay et al., 1988).In addition, variations in the DN CD3 / population with respect to mouse strains and to the class of T-cell receptor expressed occurs (Shortman et al., 1990).Conse- quently, we analyzed the DN thymocyte popu- lation in the involuted thymuses to determine whether there was an increase in the percentage of DN cells expressing CD3 using the 2Cll monoclonal antibody that detects a portion of the CD3 complex expressed by both alpha/betaand gamma/delta-expressing T cells.An increase in Thymocytes from normal and involuted thymuses analyzed for expression of CD4, CD8, CD3, and surface Ig using two-color immunofluorescence staining (see Materials and Methods).Data, unless otherwise marked, represent mean+standard deviations of three more.thymuses.bDue to variability among single group, data given for individual thymuses.

the proportion of DN CD3-expressing cells would indicate an increase in the percentage of CD4-CD8-thymocytes that would be unable to generate thymocytes in adoptive recipients.We have demonstrated that the involution process in the meV/me mouse does not affect the relative proportion of these cells in the DN subset (Table 6).

In light of our findings, it appears that the thymic involution in meV/me mice may in part be due to the CD4-CD8-CD3-sIg-subset being unable to generate thymocytes, as demonstrated by their transfer to adoptive recipients.To deter- mine whether these precursors are arrested in development, it will be of interest in future studies to examine additional time points follow- ing adoptive transfer of the donor thymocytes to confirm the absence in involuted meV/me thy- muses of each of the developmentally distinct populations that have been described.However, based on our inability to demonstrate any intra- thymic precursor activity at day 14 following transfer to irradiated recipients, and because of the large excess of DN cells injected, it seems highly unlikely that we have missed a significant population of DN thymocytes in involuting meV/me thymus that possesses precursor activity.

Alternatively, we have previously suggested (Komschlies et al., 1987) that the bone marrow of meV/me mice is deficient in an accessory cell (or factor) that permits repopulation of the thymus of irradiated recipients following i.v.adoptive transfer.Addition of normal bone marrow to meV/me bone marrow restores its i.v.thymus repopulating capacity.Viable motheaten (meV/mev) bone marrow (as are normal thymocytes), however, is able to repopulate the thymus of irradiated recipients following i.t.injection without additional accessory cell (or factor) requirements.It has been shown previously that a normal thymus contains a popu- lation of bone marrow derived "accessory" den- dritic cells of myeloid origin (Fowlkes and  Pardoll, 1989; Robey et al., 1990; Spangrude and Scollay, 1990; von Boehmer, 1990) that are important in the clonal selection, survival, and differentiation of thymocytes (Robey et al., 1990;  von Boehmer, 1990).Codevelopment of such myeloid stem cells following bone marrow i.t.injection appears to be requisite for the sub- sequent development of intrathymic precursors and donor-origin thymocytes (Spangrude and Scollay, 1990).Because numerous defects in myeloid cells have been described in meV/me mice in vivo and in vitro (McCoy et al., 1982, 1984;   Medlock et al., 1987; Hayashi et al., 1988; Shultz,  1988; Van Zant and Shultz, 1989), the lack of these bone marrow derived dendritic cells in an meV/me thymus may result in the absence, death, and/or functional inactivity of meV/me intra- thymic precursors.However, cell mixture experiments of meV/me thymocytes with normal bone marrow cells failed to "rescue" or provide evi- dence for suppression of the thymus repopulating capacity of meV/me DN thymocytes following i.t.adoptive transfer into irradiated recipients (unpublished observations).These results sug- gest that meV/me DN thymocytes, if they in fact are intrathymic precursors, may be anergic to such regulatory signals.It does not rule out, however, that these positive signals may be required only at specific, perhaps early, stages of precursor development.The setup of our assay system, in which "accessory" cells are not added until thymic involution (hence loss of intra- thymic precursor activity) has already occurred, may be too late to rescue the intrathymic precur- sor population.

It as also of interest in these studies to com- pare the phenotypic and functional character- istics of thymocytes in an involuting meV/me thy- mus with other models of thymic involution in an attempt to gain insight into the possible mech- anisms involved.Thymic involution has been observed in pregnancy, infection, surgery, drugs (including cyclosporin A), steroids, malnutrition, malignancy, and aging (Clarke and Kendall,  1989).Of special interest in our studies was the involution in meV/me thymus with that of thy- muses in aged mice, because we have suggested previously (Medlock et al., 1986) that the meV/me mouse may represent an accelerated model of aging in the lymphopoietic system.Thymic involution in normal mice begins soon after puberty, while the animal is still increasing in body weight (Santisteban, 1960).There is a decrease in thymic weight with age, a loss of cor- tical thymocytes, an increase of thymic adipose tissue, and a general decrease in thymic function.In addition, the thymic atrophy is reflected in the peripheral tissues by a decrease in T-cell func- tional responses, and an increase in a CD4 / Pgp-1 / population reflective of memory T cells (Nagelkerken et al., 1991).

As predicted, by 12 to 18 months of age, the thymus cellularity in B6+/+ mice is reduced by approximately 90% in comparison to that of young (4-week-old) B6 mice.Our phenotypic studies have demonstrated that in an involuting aged thymus, essentially normal ratios of the four phenotypic subsets of thymocytes are pre- sent.Therefore, in the in vivo studies, equivalent numbers of young and aged DN thymocytes were injected into the recipient thymus.Even though this is the case, the 17-month-old thymus is severely deficient in intrathymic precursor activity, and displays essentially no ability to repopulate the thymus of irradiated recipients at the cell doses examined (3-5106 cells injected).One 18-month-old thymus, however, did exhibit a variation in the proportions of the major thymocyte subsets (24.3% DN, 43.6% DP, 32.1% SP).When the

as also
tained for CD3 or sIg, it was observed that in this involuted thymus, 80% of the CD4-CD8-thymocytes expressed either CD3 or Ig.

It appears, however, that the loss of intra- thymic precursor activity with age in mice is not due to a defect in the bone marrow prothymocyte population.We found that on a per cell basis, bone marrow from older mice is as efficient as young marrow in thymus repopulating capacity.Contradictory results exist in regards to the study of the aging process and the immune sys- tem.CFU-s activity' and basal hemopoietic activity do not appear to be affected by age (Tyan, 1982; Williams et al., 1986).However, other investigators find prothymocyte and thy- mic activity to be impaired in aged mice (Kay,   1984; Hirokawa et al., 1986).We observed the opposite; the defect is not readily apparent in the thymus, as older irradiated recipients

eadily supp
rted the generation of donor-origin thymocytes from both young and older marrow as well as did younger irradiated recipients.It must be cautioned that the irradiation model of repopulation may not directly represent the in vivo situ- ation in regard to thymus function.Thus, irradiation may induce or reactivate "accessory cells" or factors that promote thymocyte regener- ation that are quiescent in intact aged host,s.For example, an irradiated thymus may be induced to secrete increased levels of chemotactic factors (Harr et al., 1989) such as thymotaxin (Imhof et  al., 1988; Deugnier et al., 1989).However, it is most likely not thymotaxin, because this com- pound has been found to be beta-2 microglobulin (Dargemont et al., 1989), and beta-2 microglobu- lin-deficient mutants have normal development of the thymus (Koller et al., 1990; Zijlstra et al.,  1990).Altern

ively other, as y
t undescribed, factors may have a role in the irradiation induced ability of thymus in aged mice to support thymo- cyte production in the adoptive transfer system.

Similarly, we have provided evidence that the thymic involution in meV/me mice is not due to increased levels of steroids.Although essentially all intrathymic precursor activity is eliminated by a single dose of dexamethasone, we observed sig- nificant differences in the phenotypic character- istics of the remaining thymocyte po

lations.Thus, in thymuses fr
m steroid-treated mice, a large proportion of the remaining lymphoid cells were SP, and in thymuses from either meV/me or aged mice, a much lower proportion of the lymphoid cells were SP.The characteristics of the thymocyte populations that were present follow- ing steroid treatment were similar to those previously reported by numerous investigators (Reichert et al., 1986; Vliet et al., 1986).Of interest in this study was the unexpect