Histopathology of the Thymus

Abstract

The thymus is a primary lymphoid organ that manifests dynamic physiological changes as animals age in addition to being exquisitely sensitive to stress and toxic insult. It is typically the first lymphoid tissue to respond to immunotoxic xenobiotics, with the first change being loss of cortical lymphocytes by apoptosis. This is followed by removal of the apoptotic cellular debris and, in the absence of recovery, may lead to loss of the corticomedullary demarcation and organ atrophy. Nonneoplastic proliferative changes include focal lymphoid hyperplasia and proliferation of medullary epithelial cells, often with formation of ribbons, cords, or tubules. Thymomas are relatively rare tumors that exhibit a wide spectrum of morphologic types but do not metastasize. Thymic lymphomas are common in some mouse strains and can become leukemic with hematogenous spread throughout the body.

Introduction

The immune status of the thymus, as reflected in the histological appearance, and/or changes in the relative organ weight, varies according to factors such as the age and genetic background, the adequacy of nutrition, the stress levels, and hormonal status of the animal, in addition to, and potentially interfering with, the interpretation of exposure to xenobiotics. According to Kuper: “The primary limitation of histopathology is that a tissue section represents a static time point in a dynamic process. Therefore, the dynamics of the immune system should be carefully considered in the histopathologic assessment of immunotoxicity” (Kuper et al., 2002). In addition to the thymus not being a static histologic entity, the interchangeable use of interpretive and descriptive terms has led to overlapping terminology for the same morphological change, potentially causing confusion and poor communication. For example, when there is a decrease in thymic size and cellularity, it is always appropriate to use descriptive terminology such as “reduced numbers of cortical lymphocytes” and “increased numbers of tingible body macrophages.” However, similar or even identical changes may be diagnosed as “atrophy” or “involution,” especially in the evaluation of tissues from older animals in chronic studies. Normal development, histology, and function of the thymus have been reported previously (Kuper et al., 1992; Pearse, 2006). The use of standardized descriptive nomenclature with respect to thymic pathology is addressed elsewhere (Haley et al., 2005; Elmore, 2006). This paper focuses on morphological features of nonproliferative and proliferative lesions of the rodent thymus.

Nonproliferative Morphological Changes in the Thymus

Factors Effecting Thymic Cellularity

A variety of factors and conditions result in an alteration in the cellular density and cellular composition of the thymus. Most commonly recognized is a decrease in lymphyocytes resulting from a range of background physiological influences and the immunosuppressive effect of xenobiotics.

Normal age-associated decreases in cellularity are termed involution, whereas induced reductions such as from inadequate nutrition, stress, or toxicity represent thymic atrophy. The histological appearance of the thymus under these varied conditions is similar, since the end point is the reduction in cortical lymphocytes, and shrinkage of the thymic lobules (Schuurman et al., 1994). Increased numbers of apoptotic bodies and tingible body macrophages are followed by lymphocyte depletion and increased prominence of interlobular septae and eventually an inverse of the normal histological picture, where the medulla has a higher cellular density than the cortex. Factors such as stress and toxicity can simultaneously be superimposed on the normal ageing process of lymphocyte reduction. Consequently, the distinction between atrophy and involution in older animals can be problematic. In such situations, the best course of action is to (1) assess changes in cell density and compartment size, (2) take into consideration comparison with concurrent untreated controls, (3) factor in the temporal sequence of events, (4) dose-response findings, (5) and the totality of clinical findings including histological changes in other tissue and from there deduce the appropriateness of interpreting the change atrophy versus involution.

Age Associated Effects/Involution

The age of the animal plays a large part in the level of cellularity of the thymus and its overall histological appearance. Physiological involution reflects the change in function of the thymus from lymphocyte production to recirculation. In the intact rodent, involution is a normal, gradual, and, irreversible aging change that begins at puberty and is considered to be associated with increased circulating levels of sex steroids. Gonadectomy will delay involution in both sexes (Grossman, 1985). Greenstein et al. (1987) showed that orchidectomy restored the thymus and raised the total white cell count in 18-month-old rats in which the thymus had virtually disappeared. Similar thymic regeneration was achieved in intact old rats with subcutaneous implants of lutenizing hormone releasing hormone (LHRH).

Histologically, involution is characterized by a reduction in the size of the thymus with a decrease in cortical lymphocytes, thinning and irregularity of the cortex, and loss of corticomedullary demarcation (Figures 1 and 2). At the corticomedullary junction, there is an increase in perivascular spaces, and perivascular B lymphocyte and plasma cell populations, which may form lymphofollicular structures with prominent germinal centers. There is infiltration by adipose tissue in the connective tissue capsule and septae (Figure 3). In the medulla, epithelial cells become progressively more prominent. The microscopic changes in epithelial cells that accompany involution can show considerable pleomorphic variation in distribution, architectural arrangement, and cytological appearance. Undoubtedly, there is some degree of cell proliferation as the epithelial cells are sometimes arranged in cords or ribbons and may form tubules or cysts lined by cuboidal to squamous epithelium. Epithelial changes are more prominent in females than in males and in rats (Figure 3A) more so than mice.

Species, strain, and sex differences occur in the evolution of age-dependent thymic changes. Complete involution typically does not occur in any species, including humans; there is usually some stroma with remnants almost entirely composed of epithelial cords and tubules. Changes are much less pronounced in the mouse when compared with the rat. In mice, thymic cyst formation (Figure 4) becomes more prominent with age (Khosla and Ovalle, 1986). Strain differences can be significant. For example, the thymuses of aged female Brown Norway rats consists mostly of epithelial cords or tubules with few lymphocytes, whereas epithelial components are more scarce in aged females of the Wistar or WAG strain. In ageing female NZB X SJL mice, cellular depletion of the cortex is accompanied by follicular expansion of small lymphocytes comprising a subset of sex-dependent mature T cells, and the emergence of B cells and plasma cells in the medulla (Kuper et al., 1990).

Nutrition

General undernutrition, and specific deficiencies of Vitamin B6, amino acids, fatty acids, and minerals such as zinc cause immunosuppression and a decrease in thymic weight (Robson and Schwartz, 1975; Corman, 1985; Mittal et al., 1988; Good and Lorenz, 1992). In addition, feed and/or water restriction are significant stressors, resulting in secondary immunosuppression via elevated adrenocortical hormone levels (Levine et al., 1993).

Stress

Stress causes elevated circulating levels of glucocorticosteroids mediated by the hypothalamus-pituitary-adrenal axis. The thymus is the most sensitive of the lymphoid tissues to changes in adrenocortical hormone levels and a decrease in thymic weight occurs due to loss of cortical lymphocytes. The splenic white pulp and lymph nodes can be similarly affected, although to a lesser degree. The initial response of apoptosis of cortical lymphocytes can be seen within hours of treatment with the synthetic glucocorticoid dexamethasone, followed by removal of apoptotic debris by macrophages (Figure 5). Such changes due to acute debilitating disease are common and may be prominent in animals sacrificed in a moribund condition. Known environmental stressors are social ranking within gang-housing systems, immobilization, as well as excessive changes in temperature or humidity and restriction of access to food and water (Gamallo et al., 1986; Kioukia-Fougia et al., 2002; Dal-Zotto et al., 2003; Engler and Stefanski, 2003). These changes are usually reversible on removal of the stressor.

Steroid Hormone Levels

The thymus is the lymphoid organ that shows the largest response to hormonal fluctuations. Increased levels of sex steroids have a profound effect on the thymus, ultimately resulting in involution, commencing at the onset of puberty. Studies in mice have shown that cells in the thymus staining with CD8α (Lyt-2) monoclonal antibody are particularly sensitive to sex hormone action (Greaves, 2000). Generally, estrogen reduces and androgens maintain CD8α (Lyt-2) cells (Ahmed et al., 1985). In pregnant females, an early increase in thymic weight is followed by a marked reduction in cellularity of the cortex. The level of cellularity returns to normal once pregnancy is over (Schuurman et al., 1994). Increased levels of progesterone during pregnancy have a negative effect on thymic weight, whereas increased prolactin occurring during lactation has a stimulatory effect on the thymus. Thyroxine and growth hormone also have a stimulatory effect and decreased levels of growth hormone have been associated with a reduction in thymic weight.

Immunotoxicity

Immunotoxicity refers to the potentially harmful effects that physical, chemical, or other agents have on the immune system (Koller, 1987). The dynamics of the immune system, with its ongoing cellular proliferation and differentiation, lymphocyte trafficking, and gene amplification, make it highly susceptible to toxic insults particularly in the thymus and bone marrow where rapid cell turnover occurs. Since generation of T cells by the thymus is particularly important early in life, immunotoxicants may show their effects in particular, and in lower doses, during the prenatal and the early postnatal period (Schuurman et al., 1994).

The consequences of immune dysfunction are immunosuppression or immunoenhancement, and either type of immunomodulation can provoke hypersensitivity or autoimmunity. Immunotoxic reactions manifest themselves most commonly as immunosuppression (Gopinath, 1996) characterized by selective or generalized depression of the lymphoid organs. However, there may be dramatic species differences in response to given immunotoxicants (Haley, 2003).

The most sensitive indicator of immunosuppression, particularly in short-term studies, is a decrease in relative thymus weight, which may or may not have a corresponding detectible morphological finding of decreased cellularity in the cortex, and less commonly the medulla or both. However, high doses of an immunotoxicant can cause overt toxicity or a decrease in food and/or water consumption and severe stress resulting in nonspecific (secondary) inhibitory effects on the immune system. A decrease in thymus:body weight ratio, therefore, cannot be used as stand-alone criterion of immunosuppression. In some cases a dose-response relationship as well as changes in other lymphoid tissues may be of some help in deciding whether thymic atrophy is a direct effect of immunosupression or nonspecific response to stress (Greaves, 2000).

The thymus is especially sensitive to exposure to immune system toxicants, and often there is a clear dose-associated decrease in size of the thymus secondary to apoptosis of cortical lymphocytes. The histological changes present are dependent upon the dose of the immunotoxicant and when, in the dynamic process, the thymus is examined (Figures 6–9). Following apoptosis of cortical lymphocytes and their removal by macrophages, a decrease in cortical cellularity, characterized by thinning and loss of the cortex and blurring of normal corticomedullary demarcation are seen. At a sufficiently high dose of an immunotoxicant, there may be degeneration of epithelial cells (Figure 7) and epithelial cell proliferation with development of glandular structures containing eosinophilic material (Figure 10).

The immune system has a high regenerative capacity, and, depending upon the degree of toxicity, can recover in a relatively short period of time following a toxic insult. Therefore the time between chemical insult and analysis is also an important consideration in immunotoxicity testing (Schuurman et al., 1994).

For interpretation of immunopathology, it is important to remember that the thymus is not a fixed histological entity. The level of cellularity will vary according to a number of background factors. Not all effects on the thymus seen in regulatory studies are due to the immunomodulatory effects of the test material and the use of age-matched control animals is critical. To assist with the evaluation of thymic weight data, it is recommended that an historical database should be developed and maintained for each species and strain used and should include the age, weight and sex of the animal from which the data are collected (Haley, 2003).

For a more in-depth discussion of immunotoxicology, the following references should be consulted: De Waal et al. (1997); Kuper et al. (1995); Schuurman et al. (1991); Luster et al. (1988); Koller (1987); Dean and Thurmond (1987); and Vos et al. (1998).

Other Nonproliferative Lesions of the Thymus

Other non-proliferative lesions are uncommon. Occasional developmental anomalies resulting in ectopic thyroid, parathyroid, or thymic tissue as well as intrathyroid cysts have been reported (Pearse, 2006). Inflammatory and vascular lesions are rare in the rodent thymus. Primary inflammation by extension from adjacent tissues is always possible. Multifocal dystrophic mineralization (Figure 11A) secondary to renal or parathyroid disease may be observed. Hemorrhage (Figure 11B and 11C) and fibrosis (Figure 11D) have been seen occasionally with the former, possibly a consequence of esophageal perforation during gavage administration of a test agent.

Proliferative Morphological Changes in the Thymus

Hyperplastic Changes

Epithelial Hyperplasia (Epithelial Tubules or Cords)

Epithelial hyperplasia may be present as an age-associated lesion with relatively high incidence in some strains of rodent. It may be focal (Figure 12) or diffuse and is particularly common in rats, occurring more frequently and with a higher secretory activity in females. Component cells are cuboidal to columnar, often form tubules or cords, are occasionally ciliated, and may have an admixture of secretory (goblet) cells. Tubules may contain variable amounts of eosinophilic secretory material (Figure 13). Unusual cellular forms may be seen (Figure 14). Since epithelial hyperplasia is seen in greater frequency as animals age, it is often present in thymuses that are undergoing involution (Figure 15). Furthermore, the pleomorphic variations of thymic epithelial hyperplasia are similar to the spectrum of cellular forms of thymoma; consequently, extensive hyperplastic lesions may be difficult to distinguish from early benign thymomas. Treatment-related epithelial hyperplasia has been seen with diethylstilbesterol administration in mice. Epithelial cyst formation has been recorded in rats treated with exogenous estrogen.

Lymphoid Hyperplasia

Lymphoid hyperplasia occurs in older rodents, especially mice greater than 6 months old. It is more common in females, may be unilateral or bilateral and focal or diffuse (Figure 16). It consists of proliferations of pleomorphic lymphoid cells, often near the corticomedullary junction. When focal it is typically well demarcated and may resemble a lymphoid follicle. Lymphoid follicle development has been reported in female NZB × SJL hybrid mice (Dumont and Robert, 1980). Epithelial and lymphoid hyperplasia may occur together and, because of its prevalence in older rodents, it may be seen in thymuses undergoing involution.

Atypical Hyperplasia

The histological features of this preneoplasitic precursor to treatment-induced thymic lymphoma in mice is described in the section on lymphoma below.

Neoplasia

Thymoma

This neoplasm of thymic epithelial cells is characterized by a variable admixed population of lymphoid cells (Rosai and Levine, 1976). Neoplastic epithelial cells may be localized or dispersed among lymphoid cells (Figures 17 and 18). It is likely that the gradation between epithelial hyperplasia and thymoma may present diagnostic challenges, especially for the more diffuse proliferative lesions. While it has been reported that thymomas in rodents can be divided into 3 main histological groups, depending on the relative preponderance of epithelial and lymphoid cells (Greaves, 2000), the many morphological variations seem to defy predetermined categories.

The neoplastic epithelial cells in thymomas can vary considerably in their arrangement and appearance. Thymomas may be epidermoid (Figure 19), forming nodules of nonkeratinizing squamous epithelium; consist of squamoid cells with some areas of keratinization (Figure 20); form papillary lesions with cystic areas (Figure 21); form ribbons, cords or tubules (Figure 22); and can occur as spindeloid cells (Figure 23). Some less common types include thymomas with endocrine (adenoid) growth patterns (Figure 24), a neuroendocrine phenotype (Figure 25), and a myoid thymoma (Figure 26) with skeletal muscle differentiation.

Benign thymomas are usually solitary, encapsulated, or discrete lesions confined to the thymus (Figure 27). In the malignant tumors, there is invasion of adjacent tissues (Figures 19, 21, 28, and 29), but metastasis is rare. For many cases, the distinction between benign and malignant thymoma may not be possible to ascertain with confidence.

Thymomas are uncommon in most conventional strains of rats and mice (Greaves, 2000). A low background incidence is seen in the F344/N rat and B6C3F1 mouse according to the NTP historical database (Table 1), and benign tumors are more commonly seen than malignant tumors.

In rats, however, the incidence varies considerably with strain. Spontaneous thymomas, occurring with an incidence of 97% and 36% in male and female inbred Wistar/Neuherberg rats, have been described by Murray et al. (1985). Thymoma is common in Buffalo rats, in which the microscopic and ultrastructural appearance is reported to resemble thymoma in man (Matsuyama et al., 1975). Urethane has been reported to induce thymoma in both F344/N and Buffalo rats.

Thymic Lymphoma

Neoplasms of the thymus of mice are most commonly T-cell lymphomas of thymic lymphocyte origin. T-cell lymphoma can occur spontaneously in young mice <3 months of age (Frith et al., 1985). In B6C3F1 mice spontaneously occurring thymic lymphomas are rare but can readily be induced by chemicals, viruses, irradiation, and in some types of tumor suppressor gene knockout mice. (Dunnick et al., 1997; Ward, 2006).

Most spontaneous and induced lymphoblastic lymphomas in the mouse arise unilaterally in the thymus (Frith et al., 1985). The initial gross lesion is a reduction in thymic size (thymic atrophy). Cortical atrophy is characterized by loss of cortical lymphocytes with thinning of the cortex. There is maintenance of normal lobular architecture. Subsequent to cortical atrophy, a preneoplastic stage of thymic lymphoma called atypical hyperplasia has been described in chemically treated B6C3F1 and p53-deficient mice (Dunnick et al., 1997). Atypical hyperplasia may be unilateral (Figure 30) or bilateral. There is a diffuse change with loss of the normal corticomedullary demarcation. Normal architecture is replaced by sheets of large, atypical lymphocytes and fewer admixed small lymphocytes (Figure 31). This preneoplastic lesion can be differentiated from lymphoma by the heterogenous cell population, variable mitotic index, and failure of lymphocytes to extend beyond the capsule of the thymus.

Ultimately, there is enlargement of the affected thymus lobe by nodular lymphocytic proliferation which progresses to generalized involvement of the thymus and mediastinum and finally dissemination in the blood with multiorgan involvement. (leukemic phase). Histologically, there is obliteration of the normal thymic architecture by broad sheets of a uniform population of lymphoblasts, which may extend through the thymic capsule with infiltration into surrounding tissues (Figures 32–34). Component cells are large, with moderate amounts of cytoplasm, vesicular, sometimes irregular nuclei and prominent central nucleoli. Mitoses are common as well as tingible body macrophages. Thymic lymphoma is rare in F344/N rats. It can be differentiated from mononuclear cell leukemia (Figure 35) by the lack of splenic involvement.