Cervical Thymus in the Mouse James Dooley, Matthew Erickson, Geoffrey O

Cervical Thymus in the Mouse James Dooley, Matthew Erickson, Geoffrey O. Gillard and Andrew G. Farr This information is current as J Immunol 2006; 176:6484-6490; ; of September 29, 2021. doi: 10.4049/jimmunol.176.11.6484 http://www.jimmunol.org/content/176/11/6484 Downloaded from References This article cites 29 articles, 8 of which you can access for free at: http://www.jimmunol.org/content/176/11/6484.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/ • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 29, 2021 Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Cervical Thymus in the Mouse1

James Dooley,* Matthew Erickson,* Geoffrey O. Gillard,† and Andrew G. Farr2*†

Although thymic ectopy has long been recognized in humans, the functional activity or potential immunological signiﬁcance of this thymic tissue is unknown. In this study, we describe murine thymic ectopy, cervical thymic tissue that possesses the same general organization as the thoracic thymus, that is able to support T cell differentiation, and that can export T cells to the periphery. Unexpectedly, the pattern of autoantigen expression by ectopic thymic tissue differs from that of the thoracic thymus, raising the possibility that these two thymic environments may project different versions of self. The Journal of Immunology, 2006, 176: 6484–6490.

he epithelial component of the thymic environment dis- strains of mice to display reductions in peripheral lymphocyte cel- plays considerable heterogeneity and is organized into lularity following neonatal thymectomy (11, 12). More recently,

T compartments with discrete functional, morphological, increased incidence of thymic ectopy has been reported to occur in Downloaded from and phenotypic characteristics (reviewed in Ref. 1). Epithelial cells autoimmune-prone NOD (13) and rats subjected to elevated di- of the cortical compartment, which support positive selection, are etary iodine (14). Despite the relevance of ectopic thymic tissue to phenotypically distinct from medullary epithelial cells, which are several issues of thymus biology (organogenesis, thymic epithe- thought to contribute to negative selection and may support post- lium differentiation, “extra-thymic” T cell production, and auto- selection maturation of single-positive thymocytes. immunity), and the potential to model human thymic ectopy, mu- During thymic organogenesis, endoderm from the third pharyn- rine cervical thymic tissue has received surprisingly little attention. http://www.jimmunol.org/ geal pouch separates from the pharyngeal tube and gives rise to We report here that the cervical thymic tissue in mice resembles both the thymus and parathyroid glands. Nonoverlapping domains the thoracic counterpart in many respects. However, variability in of endoderm that give rise to thymus and parathyroid gland have autoantigen expression among individual cervical thymic samples been identiﬁed by the expression pattern of Foxn1 and Gcm2, tran- compared with their thoracic counterparts raises the possibility that scription factors that deﬁne thymic and parathyroid domains, the range self-Ags that are projected may not be equivalent at these respectively (2). An obligate role for Foxn1, a member of the fork- two sites. head transcription family, in thymic organogenesis is well-estab- lished (3), although the role of Foxn1 in this process is not clear (4, Materials and Methods

5). Recently, Aire has been shown to play a critical, but ill-defined Mice by guest on September 29, 2021 role in expression of a spectrum of tissue-restricted Ags (TRA)3 by BALB/c, C57BL/6 and BALB/c nude mice were obtained from Charles medullary thymic epithelium (MTEC), which in turn impacts the River Laboratories. RAG-GFP mice (15) were obtained from Dr. P. Fink contribution of the thymus to self-tolerance (6). (University of Washington, Seattle, WA). Foxp3-GFP (16), OT-2 (17), and Although thymic tissue normally has a mediastinal location, ec- RipOVA (18) mice were obtained from Dr. A. Rudensky (University of topic cervical thymic tissue in humans has been reported (7, 8). Washington). All mice were used in accordance with protocols approved by the Institutional Animal Care and Use Committee at the University of This tissue has gained attention clinically because it presents as a Washington. Tissue samples were from neonatal 8-wk-old mice. cervical tissue mass that can obstruct breathing or can give rise to ectopic thymic tumors. Some investigators have considered cervi- Human tissue samples cal thymic tissue to be rare (9), but it has also been suggested that The use of the human tissue sections was approved by the subject review the incidence of cervical thymic tissue in adult humans may be board at the University of Washington. Ͼ50% (10). Abs and reagents Histological demonstration of what appeared to be cervical thymic tissue in mice was suggested to explain the failure of some Primary Abs for immunohistochemistry or flow cytometry have been described previously (5, 19). Immunohistochemistry † *Department of Biological Structure and Department of Immunology, University of Immunohistology techniques were performed as previously described (5, 19). Washington School of Medicine, Seattle, WA 98195 Received for publication February 8, 2006. Accepted for publication March 16, 2006. Analysis of gene expression The costs of publication of this article were defrayed in part by the payment of page Thoracic and cervical thymic TRA expression survey in pooled samples. charges. This article must therefore be hereby marked advertisement in accordance Thoracic thymic samples and pools of cervical thymic samples (from two with 18 U.S.C. Section 1734 solely to indicate this fact. to three mice) were obtained from 3-wk-old BALB/c mice. Samples were 1 This work was supported by National Institutes of Health (NIH) Grants AI 24137 homogenized and total RNA extracted using the Absolute RNA Miniprep and AI50957. G.O.G. was supported in part by training grants from the NIH and the kit per the manufacturer’s protocol (Stratagene). cDNA was synthesized Cancer Research Institute. using the Omniscript RT kit (Qiagen), starting with equal amounts of total 2 Address correspondence and reprint requests to Dr. Andrew G. Farr, Department of RNA from each of the samples. Control samples of cDNA without reverse Biological Structure, University of Washington, Box 357420, Seattle, WA 98195- transcriptase (RT) (no RT) were also made to test for nonspecific real-time 7420. E-mail address: [email protected] PCR products. cDNA samples were mixed with TRA-specific primers and 3 Abbreviations used in this paper: TRA, tissue-restricted Ag; MTEC, medullary thy- SYBR Green master mix (Applied Biosystems) and the reactions were mic epithelial cell; Ct, cycle threshold; RT, reverse transcriptase; IRBP, interphoto- assayed with a 7300 Real-Time PCR machine (Applied Biosystems). Prim- receptor retinoid-binding protein ers for the following molecules were generated: Aire, Ep-Cam, e-cadherin

ϩ recoverin, pancreatic polypeptide, insulin, c-reactive protein, rhodopsin, tral region of the tissue that was relatively devoid of ER-TR5 interphotoreceptor retinoid-binding protein (IRBP), H-K ATPase (␤- cells (Fig. 1g). This is contrasted by the ER-TR7ϩ cells that were chain), and green cone opsin (sequences available on request). Gene ex- typically investing blood vessels and scattered throughout the tho- pression was evaluated by the ⌬–⌬ cycle threshold (⌬⌬Ct) method, where the expression of a TRA by each sample was normalized to Ep-cam, a racic thymus (Fig. 1h). reliable marker for MTEC. An average ⌬Ct value was determined for each Consistent with their thymic character, cervical thymic tissue of the target genes in both thoracic and cervical samples and these values also expressed Foxn1 and pre-T␣ (Fig. 1j). Approximately 50% of were then used to calculate differences in ⌬Ct values for the two sample types after normalization to levels of Ep-cam to control for relative abun- the BALB/c mice (25 of 47) examined displayed cervical thymic dance of medullary epithelium in the samples. To validate the SYBR Green tissue, while the incidence was reduced in C57BL/6 mice (33%; 8 5 PCR products, a dissociation step was done to verify the Tm (annealing of 24), with cellularity ranging from around 10 to upwards of 5 ϫ temperature) of the SYBR Green PCR product after the PCR were run. 105 cells/organoid (Fig. 1k). The incidence of cervical thymic tis- Individual thymic and cervical sample preps. Total RNA from individual sue reported here is a conservative estimate because mice were cervical and thoracic thymi from 3-wk-old BALB/c mice was isolated (RNEasy kit; Qiagen). Only sample pairs from mice that had Ͼ200 ng of considered positive only if medial thymic tissue was identified. cervical total RNA were amplified. Total RNA from both sample types was The frequency of cervical thymic tissue juxtaposed with thyroid then amplified according to Ref. 20, starting with 200 ng of cervical RNA, and parathyroid tissue in the mouse prompted an evaluation of ␮ 200 ng of thoracic RNA, and a third sample of 2 g of thoracic RNA. normal human parathyroid in the medical histology collection at cDNA was synthesized from the amplified RNA and used as a template for SYBR Green real-time relative quantitation as described above, using the University of Washington. Two of three of human parathyroid equivalent amounts of cDNA. Because “No RT” controls were not gener- tissue samples contained lymphatic tissue judged to be thymus ated from amplified RNA, the SYBR-Green RT-PCR results were vali- (lacked lymphatic sinuses, afferent lymphatic vessels, or cortical Downloaded from dated by the dissociation temperatures (Tm) of the amplicons. follicular structures, and displayed multicellular structures consid- To confirm the linearity of mRNA amplification by the protocol used here, we amplified different amounts of thoracic thymic RNA and interro- ered to be Hassall’s bodies) and resembled the human sample de- gated the resulting cDNA with SYBR Green real-time PCR. Selected target scribed by Wu et al. (21). One of the samples is demonstrated in genes showed Ͻ1-fold variation among the samples after one or two Fig. 1, l and m. The frequency of occurrence in this small uns- rounds of amplification (data not shown). elected sample population suggests that the incidence of “occult” Grafting procedures. Although under ketamine/xylazine anesthesia, mice thymic ectopy in humans may be significant. were laparotomized to expose the left kidney and thoracic or cervical thy- http://www.jimmunol.org/ mic tissue was implanted under the kidney capsule, using aseptic tech- nique. Four weeks after grafting, graft and host tissues were evaluated with immunohistochemistry and flow cytometry. Cervical thymic tissue supports thymocyte differentiation ϫ 5 In vitro mitogenesis assay. In 96-well flat-bottom plates, 1 10 thymo- Because previous characterization of cervical thymus was based cytes from thoracic or cervical BALB/c thymi were cultured with 2 ϫ 105 irradiated syngeneic spleen cells in the presence or absence of anti-CD28 on morphology, we wanted to formally assess the capacity of this (1 ␮g/ml; eBiosciences) and graded concentrations of anti-CD3 Abs (15– tissue to support thymocyte development. Initial examination of 500 ng/ml; eBiosciences). Medium was RPMI 1640 with 10% FBS. Cells thymocyte development in cervical thymic tissue, as assessed by were cultured for 72 h at 37°C in humidified air containing 5% CO .1␮Ci 2 the relative sizes of double-negative, double-positive, and single-

of tritiated thymidine was present for the last 16 h of culture. Cells were by guest on September 29, 2021 harvested and thymidine incorporation was measured with a scintillation positive thymocyte populations, was very similar to that of tho- counter. racic thymus (Fig. 2a). Furthermore, thymocytes recovered from pooled thoracic or cervical thymic tissue responded equivalently to mitogenic stimulation by anti-CD28 and anti-CD3 Abs (Fig. 2b), Results ␣␤ Ectopic thymic tissue occurs frequently in the cervical region indicating that comparable programs of thymocyte development were supported at both sites. We performed histological evaluation of tissues in the cervical To evaluate thymocyte selection in the cervical thymus in more region to identify “islands” of thymic tissue encapsulated by con- detail, we analyzed transgenic mice bearing the MHC class II- nective tissue. We also used immunohistochemistry to screen se- restricted, OVA-specific OTII TCR transgene (17). Fig. 2c dem- rial sections of cervical tissue blocks for CD3 expression and sec- onstrates that the accumulation of mature single-positive thymo- tions of tissue from RAG-GFP mice to locate sites of RAG cytes was similar, although the frequency of double-positive expression in this region. These initial experiments focused atten- thymocytes was reduced in this pool of cervical thymocytes. The tion to areas lateral to the groove formed by the juxtaposition of extent of negative selection in mice coexpressing the OTII TCR the trachea and esophagus (Fig. 1a). Cervical thymic tissue was with an OVA-encoding a transgene driven by the rat insulin pro- rarely bilateral and there was considerable variability in localiza- moter directing the expression of the cognate Ag, OVA (18) was tion as to left or right side location and in cephalic-caudal posi- similar in cervical and thymic tissue. These data indicate that the tioning, which ranged from immediately above the sternum to association with the thyroid (Fig. 1b) and parathyroid (Fig. 1c) cervical thymic environment can support positive selection and is glands. Additional discrete lymphatic tissue located lateral to these permissive for negative selection if the selecting ligand is present. structures were identified as either additional ectopic thymic tissue As another assessment of the organization of the cervical thymus, we used Foxp3-GFP mice to compare the distribution of reg- or lymph nodes. Because of this heterogeneity, lateral structures ϩ ϩ were omitted from subsequent analyses. ulatory Foxp3 CD4 cells within the cervical and thoracic thy- In agreement with Ref. 13, immunohistochemical analyses of mus. We have previously shown that thymocytes expressing the cervical tissue showed stromal organization that was very sim- Foxp3 are localized to the medullary compartment in the thoracic ϩ ilar to thoracic thymus, with discrete cortical and medullary com- thymus (16). As shown in Fig. 2d, Foxp3 thymocytes were partments. Expression patterns of CD40, CD80, CD86, and MHC highly restricted to the medullary compartment in both cervical class II in the cervical thymus were unremarkable (data not and thoracic environments. These data indicate that the medullary shown). However, the mesenchymal fibroblastic component of the compartment of the cervical thymic provides the necessary envi- cervical thymus (as defined by ER-TR7) was less integrated with ronment for the development of regulatory T cells and that these the MTEC compartment than in the thoracic thymus. As shown in cells occupy the same thymic compartment in cervical and thoracic Fig. 1i, ER-TR7 reactivity in the cervical thymus occupied a cen- thymi. 6486 CERVICAL THYMUS Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. a, Cervical thymic tissue is often associated with the trachea and esophagus. The most reproducible location for cervical thymic tissue (3) was lateral to the groove between the trachea (1) and the underlying esophagus. There was considerable variation in the relative cephalocaudal axis; ectopic thymic tissue could be found adjacent to the sterum (2) or toward the larynx (4) and associated thyroid and parathyroid tissue. b, Thymic tissue more cephalic, surrounded by thyroid tissue. c, Thymic tissue more cephalic, adjacent to parathyroid tissue and surrounded by thyroid tissue. d–i, Cervical and thoracic thymic tissue have similar stromal organization. Thoracic (d, f, and h) and cervical (e, g, and i) thymi were reacted with Abs that identify cortical epithelium, ER-TR4, (d and e), medullary epithelium, ER-TR5, (f and g) and the ﬁbroblastic component, ER-TR7 (h and i). C, cortex; M, medulla; T, trachea. Arrow in g and i indicates a discrete medullary region that is enriched in ER-TR7ϩ components. j, RT-PCR demonstration that cervical thymic tissue expresses Foxn1 and pre-T␣, two hallmark features of functional thymic tissue. k, Cellularity of cervical thymic tissue. Incidence of cervical thymic tissue detection (as deﬁned in the text) is indicated below the x-axis label. l and m, One of two archival histology slides of human parathyroid gland (PT) adjacent to cervical thymic tissue (T). A Hassall’s body is shown in m.

The cervical thymus can contribute to the peripheral T cell pool The representation of self-Ags in cervical and thoracic thymic To formally demonstrate that the cervical thymic tissue can sup- tissue is not equivalent port the export of T cells to peripheral lymphatic tissue, we grafted To assess the expression of TRA by cervical thymus, we per- cervical thymic tissue under the kidney capsule of nude mice. Fig. formed real-time PCR on cDNA from paired cervical and thoracic 3, a–f, depicts one of six cervical thymic grafts processed to dem- thymus samples from individual mice. Because it was not feasible onstrate that normal thymic organization persisted in the grafts for to isolate MTEC from individual samples, TRA expression by at least one month after transplant, with appropriate cortical and whole thymus samples was normalized to expression levels of Ep- medullary epithelial compartmentalization and typical distribution cam (preferentially expressed by MTEC (22). As depicted in Fig. of thymocytes and dendritic cells. Furthermore, the peripheral 4, the average values for Aire and TRA expression by at least 6 in- lymph nodes (Fig. 3g) or spleen (data not shown) from nude mice dependent thoracic (T) and cervical (C) thymic samples were approx- bearing these grafts clearly contain T cells one month after initi- imately equivalent. In contrast to the modest SD of the thoracic sam- ation of the grafts. These data indicate that thymocytes from the ples, the range of ⌬Ct values of TRA expression by individual cervical thymus can contribute to the peripheral T cell pool, al- cervical samples was rather large, on the order of six cycles. though at this time we do not know the relative contribution of To assess this variability more directly, equivalent amounts of steady-state output from the grafted tissue and peripheral homeo- cervical and thoracic thymic RNA from a second set of ﬁve static proliferation. individual mice were ampliﬁed and processed identically, then The Journal of Immunology 6487 Downloaded from http://www.jimmunol.org/

FIGURE 2. Cervical thymic tissue support the development of ␣␤-lineage T cells. a, Thoracic and cervical thymic tissue display comparable representation of thymocyte subsets. Upper two panels, Expression of CD4 and CD8 by thoracic and cervical thymocytes; lower two panels, the expression of

CD3. Numbers indicate percentages in indicated quadrant or portion of histogram. b, Thymocytes from thoracic and cervical thymus tissue respond by guest on September 29, 2021 equivalently to ligand-induced proliferation. Thoracic and cervical thymocyte proliferation in the presence of anti-CD28 Abs alone was 155 and 204 cpm, respectively. c, Selection efﬁciency is comparable in thoracic and cervical thymus. Pools of cervical and thoracic tissue from OT-2 or OT-2/RIPOVA mice were processed for ﬂow cytometric analyses. Panels on the left depict the representation of thymocytes that are V␣2ϩ. d, Cervical thymic tissue contains Foxp3 cells in the appropriate location, indicating a mature organization of the medulla. Ctx, cortex; M, medulla.

FIGURE 3. Cervical thymic tissue can be transplanted and exports T cells to the periphery. a–f, Typical thymic architecture is maintained a month after transplanting under the kidney capsule. a, E-cadherin (ECCD-2); b, Ep-CAM (G8.8); c, 6C3; d, MHC class II (M5114); e, N418; f, CD3 (500A2). g, The output of thoracic and cervical thymic grafts in nude mice recipients. 6488 CERVICAL THYMUS

cervical or thoracic thymic samples was used as an index of variability of target gene expression. Consistent with the data pre- sented in Fig. 4, 2 of 32 thoracic and 21 of 39 cervical thymic samples displayed a 2-fold or greater variance from their cohorts. Comparison of values obtained with different amounts of thoracic thymic RNA indicated the starting amount of RNA was not a sig- niﬁcant contribution to the variability (1 of 30 samples with a 2-␮g sample vs 2 of 32 samples with 200-ng samples had a 2-fold or greater variance from their cohorts), further supporting the con- clusion that increased variability of TRA expression reﬂects the biology of cervical thymic tissue.

Discussion FIGURE 4. Expression of TRAs by cervical and thoracic thymi. Ex- In this report, we have revisited cervical thymic tissue in the mouse pression of TRA by cervical (C) and thoracic (T) thymi. Horizontal bars are and provide formal demonstration that it functionally resembles average values; vertical bars represent the SD of at least six samples. A, the thoracic thymus in terms of stromal cell composition and or- Aire; B, recoverin; C, pancreatic polypeptide, D, rhodopsin; E, IRBP; F, ganization, and the abilities to support thymocyte development, to H/K ATPase ␤; G, green cone opsin. export T cells to the periphery, and to contribute to central toler- Downloaded from ance. In contrast to a previous report that correlated thymic ectopy subjected to the same type of analysis. Table I displays the relative and autoimmune disease in the NOD mice (13), we found that quantity values obtained, where TRA expression was normalized thymic ectopy was a feature of commonly used strains of mice that to Ep-cam expression and differences in ⌬Ct values (⌬⌬Ct) among do not display elevated incidence of autoimmunity.

Table I. Relative quantity of gene expression in thoracic and cervical thymus samples http://www.jimmunol.org/

Target Thoracic Thymus, 2 ␮g of RNA Thoracic Thymus, 200 ng of RNA Cervical Thymus, 200 ng of RNA

Aire 0.99b 1.00 1.00 0.90 2.40 0.35 1.76 2.67 0.69 1.00 2.04 5.97 1.34 0.01 C-reactive protein ND 1.00 1.00

ND 1.79 LOST by guest on September 29, 2021 ND 1.68 0.17 ND 3.23 3.79 ND 0.01 Green cone opsin ND 1.00 1.00 ND 1.79 0.05 ND 3.42 2.68 ND 2.23 1.71 ND 0.02 H/K ATPase ␤ 0.49 1.00 1.00 0.61 0.68 0.02 1.00 2.12 4.81 1.10 1.79 0.01 0.56 0.01 Interphotoreceptor retinoid-binding protein 1.00 1.00 1.00 0.36 0.68 0.08 1.00 2.12 1.43 0.59 1.79 2.49 0.32 0.01 Pancreatic polypeptide 0.74 1.00 1.00 1.00 1.33 3.57 1.35 2.20 1.75 1.32 2.04 5.19 0.93 0.03 Recoverin 0.61 1.00 1.00 0.61 4.25 0.44 1.00 5.14 1.52 0.80 1.38 6.10 0.67 0.01 Rhodopsin 0.84 1.00 1.00 0.25 1.52 0.06 1.00 2.84 0.69 0.64 2.67 1.98 0.71 0.02

b The relative quantity (RO) values from SYBR green real-time PCR of equivalent samples of cervical and thoracic thymus for nine target genes where expression of the target molecule has been normalized to expression of Ep-cam and the representation of medullary TE. The RQ software arbitrarily assigns a value of 1 to one ⌬CT of each group for comparative purposes. This variability of RQ, ⌬⌬CT for each target gene, represents 1/2⌬⌬CT. Values that fall outside a 2-fold value difference range within the group are in bold font. The Journal of Immunology 6489

The occurrence of thymic ectopy is significant in several re- may be due to intrinsic differences in the epithelial composition of spects. First, the presence of extrathoracic sites of “thymic” T cell individual cervical thymic lobes that could in turn affect the spec- development represents a potential confounder for studies of ex- trum of TRAs they express. Takase et al. (27) recently reported trathymic T cell development, particularly those based on thoracic that expression of a subset of TRA in the human thymus displayed thymectomy models (23, 24). The contribution of ectopic thymic considerable individual variation. The relative contributions of ge- tissue to the peripheral T cell pool could be affected by strain- netic polymorphism and the number of MTEC analyzed to the dependent prevalence of thymic ectopy and relative contribution of variability of TRA expression in the human thymus is presently these two sources of T cells to subsequent peripheral homeostatic unclear. proliferation. Based on the premise that perturbed expression of TRAs can The occurrence of cervical thymic tissue in mice is also impor- contribute to autoimmunity (28–30), a smaller ectopic thymus tant in the context of thymic organogenesis. Ectopic thymic tissue with fewer MTEC may display a spectrum of TRAs that does not in humans has been widely considered to reflect third pharyngeal totally overlap the TRA profile of the thoracic thymus. This may pouch endoderm that is specified to a thymic fate but fails to mi- lead to export of T cells by the cervical thymus that have been grate to the appropriate location (7). If that is the basis for cervical vetted by a different projection of self than that of the majority of thymic tissue in mice, commingling of epithelial populations des- the peripheral T cell pool derived from the thoracic thymus. Al- tined to have thymic and parathyroid fates would be expected to though not necessarily sufficient to initiate autoimmune disease, by occur, because the ectopic thymic tissue is found along the path placing additional pressure on peripheral tolerance mechanisms, T likely to be taken by derivatives of pharyngeal pouch endoderm cells exported from ectopic thymic tissue may represent a previ- during development. However, the distribution of markers of thy- ously unrecognized susceptibility factor for autoimmunity. In this Downloaded from mic and parathyroid epithelium, Foxn1 and Gcm2, respectively, do context, it is interesting that human thymic ectopy has been shown not appear to overlap during murine embryogenesis (2). Although to modify some of the clinical parameters of myasthenia gravis and this may simply reflect assay sensitivity, the lack of Foxn1 expres- is correlated with lower remission rates following treatment (8). sion in cervical regions of the developing embryo where Foxn1ϩ It remains to be determined whether the differences in TRA thymic tissue later develops raises the possibility that progenitor expression between cervical and thoracic thymus reflect intrinsic

epithelial cells could become specified to a thymic fate (and ex- differences in these two thymic environments or whether they re- http://www.jimmunol.org/ press Foxn1) some time after their migration to a cervical location. flect a common property of thymic epithelium that becomes evi- The mechanism underlying the organogenesis of cervical thymic dent as the thymic environment is scaled down. In either case, tissue is an important question that warrants examination. The re- these subtle differences in the cervical thymic environment may producible thymic ectopy in mice represents an opportunity to fol- have significant immunological consequences. low the organogenesis of ectopic thymus and to clarify its embry- ological origins, issues that are not readily addressed in humans. Disclosures Finally, ectopic thymic tissue may have a bearing on the issue of The authors have no financial conflict of interest. self-tolerance. It is now generally recognized that medullary epi- by guest on September 29, 2021 thelial cells express a remarkable spectrum of self-Ags that are References normally considered to be unique or specific to a particular tissue 1. Boyd, R. L., C. L. Tucek, D. I. Godfrey, D. J. Izon, T. J. Wilson, N. J. Davidson, or organ and that their expression represents an important mech- A. G. Bean, H. M. Ladyman, M. A. Ritter, and P. Hugo. 1993. The thymic anism to effect self-tolerance. We have shown here that cervical microenvironment. Immunol. Today 14: 445–459. 2. Gordon, J., A. R. Bennett, C. C. Blackburn, and N. R. Manley. 2001. Gcm2 and thymic tissue collectively expresses a spectrum of TRA that ap- Foxn1 mark early parathyroid- and thymus-specific domains in the developing pears similar to the thoracic thymus. However, assessment of in- third pharyngeal pouch. Mech. Dev. 103: 141–143. dividual cervical thymic “lobes” revealed considerable heteroge- 3. Nehls, M., D. Pfeifer, M. Schorpp, H. Hedrich, and T. Boehm. 1994. New member of the winged-helix protein family disrupted in mouse and rat nude mutations. neity in levels of TRA expression. Because individual TRAs are Nature 372: 103–107. expressed by relatively rare medullary epithelial cells (25) or dis- 4. Nehls, M., B. Kyewski, M. Messerle, R. Waldschutz, K. Schuddekopf, crete subsets of epithelial cells (19), scaling thymic size (and the A. J. Smith, and T. Boehm. 1996. Two genetically separable steps in the differentiation of thymic epithelium. Science 272: 886–889. number of MTEC) could have an impact on the range or levels of 5. Dooley, J., M. Erickson, H. Roelink, and A. G. Farr. 2005. Nude thymic rudiment TRA expression. lacking functional Foxn1 resembles respiratory epithelium. Dev. Dyn. 233: 1605–1612. Analyses of TRA expression by isolated MTEC from the tho- 6. Anderson, M. S., E. S. Venanzi, L. Klein, Z. Chen, S. P. Berzins, S. J. Turley, racic thymus has been based on relatively large pools of cells and H. von Boehmer, R. Bronson, A. Dierich, C. Benoist, and D. Mathis. 2002. represent a profile average of many cells. Although we have com- Projection of an immunological self shadow within the thymus by the aire protein. Science 298: 1395–1401. pared equivalent amounts of cervical and thoracic thymic RNA, 7. Tovi, F., and A. J. Mares. 1978. The aberrant cervical thymus: embryology, the thoracic sample represents the activity of many more MTEC pathology, and clinical implications. Am. J. Surg. 136: 631–637. and thus also presents an average expression of TRAs by many 8. Ashour, M. 1995. Prevalence of ectopic thymic tissue in myasthenia gravis and its clinical significance. J. Thorac. Cardiovasc. Surg. 109: 632–635. cells. Consistent with the immunohistochemical analysis of TRA 9. Buyukyavuz, I., S. Otcu, I. Karnak, Z. Akcoren, and M. E. Senocak. 2002. Ec- expression, we have found that a low frequency of MTEC express topic thymic tissue as a rare and confusing entity. Eur. J. Pediatr. Surg. 12: TRA when individual or small numbers of thoracic MTEC was 327–329. 10. Zielinski, M., J. Kuzdzal, A. Szlubowski, and J. Soja. 2004. Comparison of late assessed by RT-PCR (26). These data suggest that the variability results of basic transsternal and extended transsternal thymectomies in the treat- of TRA expression observed in the cervical thymic samples could ment of myasthenia gravis. Ann. Thorac. Surg. 78: 253–258. reflect the small size of this tissue and the low numbers of MTEC 11. Law, L. W., T. B. Dunn, N. Trainin, and R. H. Levey. 1964. Studies of thymic function. Wistar. Inst. Symp. Monogr. 2: 105–120. that they contain. It is likely that the smaller absolute number of 12. Miller, J. F. A. P., A. H. E. Marshall, and R. G. White. 1962. The immunological MTEC in the cervical thymus cannot provide a significant aver- significance of the thymus. Adv. Immunol. 2: 111–162. 13. Many, M. C., H. A. Drexhage, and J. F. Denef. 1993. High frequency of thymic aging effect in terms of TRA expression. This interpretation is ectopy in thyroids from autoimmune prone nonobese diabetic female mice. Lab. consistent with the data in Fig. 4, where the average levels of Invest. 69: 364–367. individual TRA expression by multiple cervical thymi approxi- 14. Mooij, P., H. J. de Wit, and H. A. Drexhage. 1994. A high iodine intake in Wistar rats results in the development of a thyroid-associated ectopic thymic tissue and mate those of thoracic thymus. These differences may reflect tem- is accompanied by a low thyroid autoimmune reactivity. Immunology 81: poral oscillations in the production of these TRAs by MTEC or 309–316. 6490 CERVICAL THYMUS

15. Nagaoka, H., G. Gonzalez-Aseguinolaza, M. Tsuji, and M. C. Nussenzweig. 23. Dejbakhsh-Jones, S., L. Jerabek, I. L. Weissman, and S. Strober. 1995. Extra- 2000. Immunization and infection change the number of recombination activating thymic maturation of ␣␤ T cells from hemopoietic stem cells. J. Immunol. 155: gene (RAG)-expressing B cells in the periphery by altering immature lymphocyte 3338–3344. production. J. Exp. Med. 191: 2113–2120. 24. Sato, K., K. Ohtsuka, K. Hasegawa, S. Yamagiwa, H. Watanabe, H. Asakura, and 16. Fontenot, J. D., J. P. Rasmussen, L. M. Williams, J. L. Dooley, A. G. Farr, and T. Abo. 1995. Evidence for extrathymic generation of intermediate T cell recep- A. Y. Rudensky. 2005. Regulatory T cell lineage specification by the forkhead tor cells in the liver revealed in thymectomized, irradiated mice subjected to bone transcription factor foxp3. Immunity 22: 329–341. J. Exp. Med. 17. Barnden, M. J., J. Allison, W. R. Heath, and F. R. Carbone. 1998. Defective TCR marrow transplantation. 182: 759–767. expression in transgenic mice constructed using cDNA-based ␣- and ␤-chain 25. Smith, K. M., D. C. Olson, R. Hirose, and D. Hanahan. 1997. Pancreatic gene genes under the control of heterologous regulatory elements. Immunol. Cell Biol. expression in rare cells of thymic medulla: evidence for functional contribution 76: 34–40. to T cell tolerance. Int. Immunol. 9: 1355–1365. 18. Kurts, C., R. M. Sutherland, G. Davey, M. Li, A. M. Lew, E. Blanas, 26. Gillard, G. O., and A. Farr. 2006. Features of medullary thymic epithelium im- F. R. Carbone, J. F. Miller, and W. R. Heath. 1999. CD8 T cell ignorance or plicate postnatal development in maintaining epithelial heterogeneity and TRA tolerance to islet antigens depends on antigen dose. Proc. Natl. Acad. Sci. USA expression. J. Immunol. In press. 96: 12703–12707. 19. Dooley, J. L., M. Erickson, and A. G. Farr. 2005. An organized medullary epi- 27. Takase, H., C. R. Yu, R. M. Mahdi, D. C. Douek, G. B. Dirusso, F. M. Midgley, thelial structure in the normal thymus coordinately expresses molecules of respi- R. Dogra, G. Allende, E. Rosenkranz, A. Pugliese, et al. 2005. Thymic expression ratory epithelium and phenotypically resembles the epithelial rudiment of nude of peripheral tissue antigens in humans: a remarkable variability among individ- mice. J. Immunol. 175: 4331-4337. uals. Int. Immunol. 17: 1131–1140. 20. Scherer, A., A. Krause, J. R. Walker, S. E. Sutton, D. Seron, F. Raulf, and 28. Egwuagu, C. E., P. Charukamnoetkanok, and I. Gery. 1997. Thymic expression M. P. Cooke. 2003. Optimized protocol for linear RNA amplification and appli- of autoantigens correlates with resistance to autoimmune disease. J. Immunol. cation to gene expression profiling of human renal biopsies. BioTechniques 34: 159: 3109–3112. 546–550, 552–544, 556. 29. Hanahan, D. 1998. Peripheral-antigen-expressing cells in thymic medulla: factors 21. Wu, S. L., D. Gupta, and J. Connelly. 2001. Adult ectopic thymus adjacent to thyroid and parathyroid. Arch. Pathol. Lab. Med. 125: 842–843. in self-tolerance and autoimmunity. Curr. Opin. Immunol. 10: 656–662. 22. Farr, A., A. Nelson, J. Truex, and S. Hosier. 1991. Epithelial heterogeneity in the 30. Derbinski, J., A. Schulte, B. Kyewski, and L. Klein. 2001. Promiscuous gene Downloaded from murine thymus: a cell surface glycoprotein expressed by subcapsular and med- expression in medullary thymic epithelial cells mirrors the peripheral self. Nat. ullary epithelium. J. Histochem. Cytochem. 39: 645–653. Immunol. 2: 1032–1039. Vol. 176, No. 11 http://www.jimmunol.org/ by guest on September 29, 2021