The Picture of the Prostatic Lymphokine Network Is Becoming Increasingly Complex

Abstract

The immunology of the prostate has recently developed into a new field of research in urology. Although we do not yet understand why the leukocyte population increases, we know that most resected prostate tissue shows signs of an inflammatory reaction. Different types of inflammation exist, and must be distinguished carefully according to distribution and location of leukocytes and histology of the surrounding tissue. This article reviews recent findings and discusses the complex mechanisms involved in the prostatic inflammatory response. The roles of estrogen, interleukin (IL)-6, IL-8, IL-15, and IL-17 are examined.

Neglected by most and underestimated in its significance, the immunology of the prostate has recently developed into a new field of research in urology. In this context the research of lymphokines has gained particular importance. One must differentiate between immunocompetence of prostatic cells on the one hand and infiltrating or resident leukocytes on the other. Even though investigation of the immunological function of the prostate is still at a very early stage, we now know that both prostatic epithelial and stromal cells express members of the Toll-like receptor family and are therefore capable of recognizing foreign incoming antigens. Although the functional consequences of antigen binding of prostatic cells via Toll-like receptors are largely undefined, preliminary data suggest that both cell types respond with increased production of proinflammatory cytokines. Another point of interest is the alteration associated with the frequently occurring chronic inflammation process accompanying both atrophic and benign hyperplastic changes. This review summarizes recent data and discusses the complex regulatory interaction between cells of the prostate and leukocytes.

Leukocytes

The prostate is populated by small numbers of αβ T cells, B lymphocytes, macrophages, and mast cells.^1,2 T lymphocytes populate the prostate as early as the twelfth week of gestation. Their incidence reaches a peak between weeks 21 and 30 and decreases thereafter.³ Although the number of leukocytes declines during weeks 31 to 40 of gestation, the normal prostate still harbors 7.3 ± 3.3 CD3⁺ T lymphocytes per square millimeter, two thirds of which are CD8⁺ T cells. In the normal prostate, the T lymphocytes are evenly distributed throughout the interstitium and between the epithelial cells.¹ The number of T lymphocytes generally increases with age. In specimens taken from 50-year-old patients, for example, up to 55 T lymphocytes per square millimeter have been counted without the presence of clustering or pathologic histological alteration.

Inflammation

Although we do not yet understand why and how often the leukocytic population increases, it is well accepted that most resected prostate tissue shows signs of an inflammatory reaction. Different types of inflammation exist, and must be distinguished carefully according to distribution and location of leukocytes and histology of the surrounding tissue. The first type, formerly designated as post-inflammatory atrophy by McNeal and colleagues,⁴ chronic prostatitis by Bennett and colleagues,⁵ and lymphocytic prostatitis by Blumenfeld and colleagues,⁶ was recently termed proliferative inflammatory atrophy (PIA).⁷ PIA is characterized by discrete foci of proliferative glandular epithelium with the morphological appearance of simple atrophy or postatrophic hyperplasia occurring in association with inflammation. The key features of PIA are the presence of two distinct cell layers of epithelial cells, mononuclear and/or polymorphonuclear inflammatory cells (80%–85% of which are CD3⁺ T cells) in both the epithelial and the stromal compartments, and stromal atrophy with variable amounts of fibrosis (Figure 1).

Figure 1.

Proliferative inflammatory atrophy. Immunohistochemistry using anti-CD3.

A second type of inflammation occurs with benign prostatic hyperplasia (BPH). An association of BPH with inflammatory processes was noted as early as 1937.⁸ Kohnen and colleagues reported that 98% of 162 analyzed BPH specimens had an inflammatory infiltrate.⁹ It has been shown that BPH specimens are often infiltrated by leukocytes and that the majority of infiltrating T lymphocytes are CD4⁺ memory T lymphocytes.1,2,9–12 Bierhoff¹³ described a “scattered” type of inflammation in BPH, characterized by significantly increased diffuse infiltrates of T lymphocytes in fibroblastic, fibromuscular, and smooth-muscular stromal nodules, but decreased infiltration of mesenchymal nodules when compared to the surrounding stroma (Figure 2). He concluded that lymphocyte infiltrates of the stromal nodules must be strictly separated from inflammatory changes frequently accompanying BPH. Functional testing of these BPH-tissue-infiltrating T cells showed that they display the functional features of activated T lymphocytes,² although it is still unclear whether this BPH-associated immunoreaction is triggered by foreign antigens, autoantigens, or both.

Figure 2.

“Scattered” type of inflammation. Immunohistochemistry using anti-CD3.

However, the incidence of chronic inflammatory prostatic diseases of noninfectious origin has been shown to be eight times as high as that of bacterial prostatitis.¹⁴ Taguchi and colleagues first used the term autoimmune prostatitis in 1985¹⁵ after showing that the prostate became strongly infiltrated after neonatal thymectomy. Ten years later Zisman and colleagues¹⁶ detected anti-prostate-specific antigen (PSA) antibodies in the sera of BPH patients and concluded that BPH may represent an organ-limited autoimmune condition without polyclonal B-cell activation. Animal models for the induction of autoimmune- or antigen-independent prostatitis have now been established^17,18 that suggest, that at least some determinants of normal prostatic proteins are not tolerated by the immune system (Figure 3).

Figure 3.

(A) Prostatic T cells and syngeneic benign prostatic hyperplasia (BPH) stromal cells. (B) Prostatic T cells attacking syngeneic BPH epithelial cells. Raster electron tunneling microscopy.

Normal human prostatic proteins such as prostate-specific antigen and prostate-specific membrane antigen have recently been used successfully as vaccines (Figure 4).^19–21 Furthermore, functional experiments in mice,¹⁷ dogs,²² and rats,^23–25 as well as in humans,²⁶ suggest that prostatic inflammation is at least partly under hormonal influence and may be caused by a decrease in androgens and a simultaneous increase in estrogens. In addition to this possible hormonal influence, evidence exists that, at least in rats, genetic background and age are also associated with the susceptibility to lymphoid infiltration of the prostate.²⁷ It has been speculated that growth factors released by these infiltrating leukocytes might alter growth of neighboring stromal cells and thus contribute to prostatic hyperplasia.^2,13

Figure 4.

Kinetics of epithelial cell killing by prostatic T lymphocytes. Parallel time lapse microscopy of autologous epithelial cell killing by prostatic T lymphocytes derived from a prostatic specimen with benign prostatic hyperplasia.

More recently, Kramer and colleagues demonstrated a direct relationship between T-cell infiltration, lymphokines and stromal hyperproliferation in BPH. These investigators were able to show that hyperplastic tissues and, to a lesser degree, normal prostate tissues, exhibited significant amounts of interferon (IFN)-γ mRNA. However, in contrast to normal prostate tissues, benign hyperplastic tissues also contained considerable amounts of interleukin (IL)-2 and IL-4 mRNA (ratio: 10:13). Screening for the major source of lymphokine production revealed that T-cell lines generated from the tissue expressed high amounts of IFN-γ, IL-2, and IL-4 mRNA and also small amounts of fibroblast growth factor (FGF)-2 mRNA, irrespective of their CD4-to-CD8 ratio (Figure 5).²⁸

Figure 5.

Normally rare T-cell types in benign prostatic hyperplasia (BPH). Flow cytometry of 3 different BPH-tissue derived T-cell lines using triple staining. Cells were gated for anti-CD3 reactive cells and reactivity with the two other markers is shown. The normally rare T-cell subsets are best indicated using the combination anti-CD4 (x-axis) and CD8 (y-axis) of i) BPH-T-cell-line 2 (left lower corner shows substantial numbers of double negative T-cells) and of ii) BPH-T-cell-line 3 (shows a shift of all cells to the right indicating CD4 and CD8 positive “double positive” T cells. A double positive and double negative phenotype is normally seen only on a subset of thymocytes.

Expression of IFN-γ, IL-2, and IL-4 mRNA in benign hyperplastic tissues suggests that the disease is associated with both classical types of T-cell responses, the type 1 and type 2 immune response.^29,30 Type 1 T lymphocytes are sources of IFN-γ, IL-2, and IL-10 and are involved in cell-mediated inflammatory reactions. Type 2 T lymphocytes are negative for IFN-γ and positive for IL-4, IL-5, and IL-13 and are found in association with strong antibody and allergic responses.^31,32 Most benign hyperplastic tissues and all polyclonal, long-term-cultured-tissue-derived T-cell lines revealed no such dichotomy and exhibited Southern Blot bands when probed for IFN-γ, IL-2, and IL-4. Preliminary results using intracellular cytokine detection confirmed these findings and revealed substantial numbers of IFN-γ and IL-2-positive prostatic T lymphocytes that coexpress Th2 cytokines (IL-4 and IL-13). One explanation is that IFN-γ influences differentiation and that the combination of TGF (transforming growth factor)-β1 and IFN-γ keeps CD4-positive T cells in a proliferating, IL-2-secreting T_H0 state³³ with a memory phenotype similar to that described for BPH T cells.

In our recent study,²⁸ marked differences between normal stromal cells and prostatic stromal cells taken from BPH were shown. Normal prostatic stromal-cell lines were a functionally heterogeneous population that did not respond to IL-2, IL-7, and IFN-γ but revealed a significant dose-dependent inhibition of the proliferation by IL-4. In contrast, the majority of polyclonal BPH prostatic stromal-cell lines responded to IL-2, IL-7, and IFN-γ with increased proliferation, which remained relatively high even after inhibition by IL-4.

Pro-Inflammatory Cytokine Expression

A number of different theories may explain the causes of leukocytic inflammation. Among other explanations, the effect of hormones on the regulation of pro-inflammatory genes has been addressed by several studies (Figure 6).

Figure 6.

Intraprostatic lymphokine network. IL, interleukin; IFN, interferon; TGF, transforming growth factor; FGF, fibroblast growth factor; GM-CSF, granulocyte macrophage colony stimulating factor; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor; SCF, stem cell factor.

Estrogen

A variety of evidence suggests that estrogen plays an important role in susceptibility to inflammation and regulation of IFN-γ production.³⁴ For example, estradiol (E2) was shown to increase activity or mRNA expression of IFN-γ promoter in lymphocytes.³⁴ More recently, treatment of rats with E2 resulted in upregulation of pro-inflammatory transcripts of macrophage inflammatory protein (MIP)-2, inducible nitric oxide synthase, IL-6, IL-1γ, and tumor necrosis factor (TNF)-γ in the prostate after as little as 4 days, well before cellular inflammation occurs. Histological signs of inflammation became evident only after 2 to 4 weeks of E2 exposure (capsule implantation of solid E2). Inflammatory cells were predominantly T lymphocytes and were consistently detected first in prostatic interstitium and then in the acini. Among the various upregulated pro-inflammatory transcripts, the most dramatic results were shown for C-X-C chemokine and MIP-2, followed by IL-6. Only a small increase (1.5–3 times) was detected in IL-1β and TNF-α. In addition, modest upregulation of IL-4, IL-5, and IL-10 transcripts, but no change in IL-2, IL-12, or cyclooxygenase (COX)-2 expression, was shown in E2-treated animals.²⁵

Interleukin-6

In the context of chronic inflammation and pro-inflammatory cytokine expression, the role of IL-6, in particular, has been studied in detail, specifically in the prostate. IL-6 is one of the major physiological mediators of acute phase reaction and is a pleiotropic cytokine influencing antigen-specific immune responses and inflammatory reactions. It has been shown to play a role in hematopoiesis, immune response, inflammation, bone metabolism, and neural development. Recognized sources of IL-6 include fibroblasts, activated macrophages or monocytes, activated T and B cells, endothelial cells, stromal cells, and a variety of cancer cells. In the prostate, IL-6 is secreted by both normal and neoplastic prostatic epithelial cells and can act as a growth factor for normal prostatic epithelial cells as well as prostate cancer cells.^35,36

Several of the commonly used prostate cancer lines (PC3, DU145, LNCaP) express high-affinity receptors for IL-6, and PC3 and DU145 cell lines secrete IL-6.^37,38 An autocrine growth factor loop has therefore been suggested.³⁹ Furthermore, IL-6 protein concentrations are increased (~18-fold) in localized prostate cancers when compared to normal prostate tissue. Increased expression of IL-6 receptor is correlated with increased proliferation in vivo as assessed by Ki67 immunohistochemistry.

Interleukin-8

Another cytokine with important pro-inflammatory properties is IL-8. It is often induced along with IL-6 in response to numerous stimuli in vitro, including chemical and microbial stimuli and selected cytokines. IL-8 has chemotactic properties, attracting neutrophils and mononuclear cells into sites of inflammation. It is a primary inflammatory cytokine, secreted by monocytes, mitogen-stimulated T lymphocytes, neutrophils, eosinophils, fibroblasts, synovial cells, endothelial cells, mesothelial cells, epithelial cells, and keratinocytes.

Ferrer and colleagues reported production of IL-8 by several prostatic cell lines in vitro as well as of prostatic cancer cells in vivo. Epithelial cells in normal prostate and BPH specimens were negative for IL-8.⁴⁰ Another interesting finding is that IL-8 mRNA expression was upregulated as much as five-fold in peripheral blood lymphocytes of patients with carcinoma of the prostate.⁴¹ IL-8 produced by prostatic epithelial cells can act as a paracrine inducer of fibroblast growth factor (FGF)-2 (a potent growth factor for prostatic stromal cells) production by prostatic stromal cells in vitro.⁴² Enzyme-linked immunosorbent assays performed on normal and benign hyperplastic tissues have revealed substantial quantities of IL-8 in the normal prostate and elevated levels in the hyperplastic prostate. Thus, the authors concluded stromal proliferation might be controlled, at least in part, by IL-8 secreted by prostatic epithelial cells.⁴³ Although the data regarding IL-8 production by normal and/or BPH epithelial cells are conflicting both studies showed that large quantities of IL-8 can be expressed intraprostatically.

Interleukin-15

Another important pro-inflammatory cytokine that has been studied in the human prostate is IL-15. It was first identified in the supernatant of kidney epithelial cells by its ability to replace IL-2.⁴⁴ In contrast to IL-2, which is mainly produced by activated T cells, IL-15 is expressed constitutively in a wide variety of tissues and cell types, but not in normal resting or activated T lymphocytes. IL-15 plays an important role in the homeostasis of lymphocytes and acts as a potent chemoattractant, inducing locomotion and migration of human T lymphocytes,⁴² thereby contributing to protective immune responses against microbial pathogens.

Handisurya and colleagues demonstrated that in the normal prostate, IL-15 was strongly expressed by smooth-muscle cells, weakly expressed by endothelial cells, and very weakly expressed (often only irregularly or not at all) by epithelial cells.⁴⁵ However, in BPH specimens, increased IL-15 expression was frequently found in luminal secretory epithelial cells. Because IL-15 is a potent stimulator of BPH-T lymphocyte proliferation, the authors suggested that prostatic IL-15 might play an important role in lymphocytic infiltration and the maintenance of chronic inflammation in BPH tissue.⁴⁵ Theoretically, these lymphocytes will generate inflammatory conditions characterized by increased IFN-γ production. Handisurya and colleagues also showed that elevated IFN-γ will augment IL-15 production by prostatic cells, which might result in increased chemoattractant activity and in the recruitment of more T lymphocytes by facilitating transendothelial migration into perivascular tissues.

Interleukin-17

We recently identified another very potent player in the prostate, namely IL-17, a T lymphocyte-derived proinflammatory cytokine. It is expressed primarily by activated prostatic T lymphocytes and strongly augments the production of the other members of the pro-inflammatory cytokine family, which is why it is called the local fine tuner of immune response. In our study, IL-17 had enormous capacities in stimulating IL-6, IL-8, IL-1α and β protein and mRNA production multifold. In this respect IL-17 was much more powerful than for example LPS or the protein kinase activator PMA. Although expression of IL-17 mRNA in human prostate is frequent, it is not produced by prostatic cells or only in very small quantities. The vast majority of prostatic IL-17 is T-cell derived and it has been shown that only activated BPH-T-lymphocytes produce and secrete IL-17.⁴⁶

The Network

It is well established that TGF-β inhibits growth, whereas FGF-2 stimulates growth. Less well known is that substantial amounts of both growth-regulating cytokines are produced by T lymphocytes.²⁸ An additional pathway by which T cells can influence the balance existing between cells and growth factors is via the natural antagonist of TGF-β, IFN-γ. As was recently shown,⁴⁷ TGF-β1 mRNA synthesis by prostatic stromal cells is stimulated by IFN-γ, whereas the uptake of TGF-β1 by prostatic stromal cells is inhibited. As discussed above, IFN-γ simultaneously stimulates production of IL-15, thereby augmenting the influx of further T cells. These T cells are activated by an unknown antigen and respond by producing lymphokines such as IL-4 and IL-13. Both lymphokines have been shown to induce the production of 3β-hydroxydehydrogenase/isomerase type 1 (3β-HSD) by prostatic epithelial cells. 3β-HSD is known to catalyze an essential step in the formation of active androgens and estrogens from dehydroepiandrosterone. ^48,49

Main Points.

• Most resected prostate tissue shows signs of an inflammatory reaction. Different types of inflammation exist, and must be distinguished carefully according to distribution and location of leukocytes and histology of the surrounding tissue.

• A direct relationship between T-cell infiltration and stromal proliferation has been demonstrated.

• A variety of evidence suggests that estrogen plays an important role in susceptibility to inflammation and regulation of IFN-γ production.

• The pro-inflammatory cytokines interleukin (IL)-6, IL-8, IL-15, and IL-17 also play important roles in the complex mechanisms underlying prostate inflammation.