Galectin-1 enhances TNFα-induced inflammatory responses in Sertoli cells through activation of MAPK signalling

Abstract

Galectin-1 (Gal-1) is a pleiotropic lectin involved in the modulation of immune responses. Using a model of rat experimental autoimmune orchitis (EAO), we investigated the role of Gal-1 in testicular inflammation. EAO is characterized by leukocytic infiltrates in the interstitium, damage of spermatogenesis and production of inflammatory mediators like TNFα and MCP1 causing infertility. In normal rat testis Gal-1 was mainly expressed in Sertoli cells and germ cells. In the inflamed testis, Gal-1 expression was significantly downregulated most likely due to germ cell loss. Analyses of lectin binding and expression of glucosaminyl- and sialyltransferases indicated that the glycan composition on the cell surface of Sertoli and peritubular cells becomes less favourable for Gal-1 binding under inflammatory conditions. In primary Sertoli cells Gal-1 expression was found to be upregulated after TNFα challenge. Pretreatment with Gal-1 synergistically and specifically enhanced TNFα-induced expression of MCP1, IL-1α, IL-6 and TNFα in Sertoli cells. Combined stimulation of Sertoli cells with Gal-1 and TNFα enhanced the phosphorylation of MAP kinases as compared to TNFα or Gal-1 alone. Taken together, our data show that Gal-1 modulates inflammatory responses in Sertoli cells by enhancing the pro-inflammatory activity of TNFα via stimulation of MAPK signalling.

Figure 1

In normal rat testes Gal-1 is expressed mainly in Sertoli cells and germ cells but not in macrophages. Hematoxylin-eosin (HE) staining in cryostat sections from normal (a), adjuvant control (b) and EAO (c) rat testes. Localization of Gal-1 (Alexa 546, orange) in normal (d,g,j), adjuvant control (e,h,k) and EAO (f,i,l,m,n) testis. Vimentin (Alexa 488, green) was used as a marker of Sertoli cells (d,e,f). Insets show Gal-1 (Alexa 546, orange) stained in germ cells (thin arrow) and Sertoli cells (thick arrow) (d,f). Staining of Gal-1 and CD68 (Alexa 488, green) or CD163 (Alexa 488, green) in the region of granulomas (m,n). Testicular macrophages were stained with CD68 and CD163 antibodies. Gal-1 was expressed in some CD68 macrophages (m) found around granulomas (thick arrow), but not in CD163 macrophages (n) (thin arrow).

Figure 2

Changes in the expression of Gal-1 in EAO testes are due to germ cell loss. Western blot (a) and densitometric (b) analysis of Gal-1 expression in testes from untreated, adjuvant control and EAO animals. Gal-1 mRNA relative expression (RE) was normalized to three house keeping genes (β-actin, Hprt and 18 s rRNA) (c) or Sertoli cell marker Sox9 (d). The blots were cropped and the full-length blots are presented in the supplementary data (Supplementary Fig. 1); (n = 5, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 3

Expression analysis of transferases involved in glycan biosynthesis and lectin-FITC binding in EAO testes. (a) Schematic representation of N- and O-glycan biosynthesis. St6gal1 (b), Mgat5 (c) and Gcnt1 (d) relative mRNA expression was normalized to β-actin, Hprt and 18s rRNA (n = 5). (e) SNA-FITC binds stronger to testicular sections from EAO rats; (**P < 0.01, ***P < 0.001).

Figure 4

Influence of TNFα stimulation on the glycan profile of Sertoli cells and peritubular cells. (a) Flow cytometric analysis of cell-surface glycans in Sertoli cells and peritubular cells after stimulation with 25 ng/ml TNFα was detected by staining cells with FITC-labeled lectins (SNA, MAA, L-PHA, or PNA) (black filled histograms) or without (open histograms). Numbers in the upper-right corner represent the median of fluorescence intensity (black filled histograms). The binding of FITC-labelled lectins to Sertoli cells (b) or peritubular cells (c) was quantified as relative median fluorescence intensity (rMFI); (rMFI = (MFI with lectin – MFI without lectin)/MFI without lectin) (n = 3–5, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 5

Analysis of Gal-1 expression in primary Sertoli cells. Western blot (a) and densitometric (b) analysis of Gal-1 expression in primary Sertoli cells after TNFα stimulation. The blots were cropped and entire/uncut blots are presented in the supplementary data (Supplementary Fig. 3) (n = 5, *P < 0.05).

Figure 6

Gal-1 and TNFα act synergistically on the expression of pro-inflammatory mediators in Sertoli cells. Primary Sertoli cells were pretreated with Gal-1 (5 µg/ml) for 2 h, and then stimulated with TNFα (25 ng/ml) for 6 h. Lactose (Lac, 50 mM) was used as an inhibitor of Gal-1 binding. Relative mRNA expression of IL-1α (a), TNFα (b), MCP1 (c), TGFβ2 (d), and IL-6 (e) was normalized to Hprt; (n = 3–5, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 7

Effects of Gal-1 and TNFα on MAPK phosphorylation in Sertoli cells. (a) Isolated Sertoli cells were pretreated with Gal-1 (5 μg/ml; 2 h) and then stimulated with TNFα (25 ng/ml) for 0–120 min. Subsequently, phosphorylation of MAP kinases ERK1/2, p38 and JNK was investigated by Western blotting. Blots are representative of at least three independent experiments. The blots were cropped and the full-length blots are presented in Supplementary Fig. 4. Densitometric analyses of Fig. 7a are shown in Fig. 7b,c; (n = 3, *P < 0.05).

Figure 8

Lactose abrogates effects of Gal-1 on MAPK phosphorylation in TNFα treated Sertoli cells. (a) Gal-1 was pre-incubated with lactose (50 mM) for 5 min prior to addition to Sertoli cells. After 2 h Sertoli cells were stimulated with TNFα (25 ng/ml) for the indicated times. Subsequently phosphorylation of MAP kinases ERK1/2, p38 and JNK was investigated by Western blotting. Blots are representative of at least three independent experiments. The blots were cropped and the entire/uncut blots are presented in the Supplementary Fig. 6. Densitometric analyses of Fig. 8a are shown in Fig. 8b–g. Stimulation with TNFα was for 15 min (b,d,f) or 30 min (c,e,g); (n = 3–4, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 9

Inhibitors of p38 and JNK MAP kinases reverse the synergistic effect of Gal-1 and TNFα on expression of inflammatory mediators in Sertoli cells. Primary Sertoli cells were pretreated with Gal-1 (5 µg/ml) for 1 h prior to addition of p38 inhibitor SB 203580 (5 mM; p38i) and JNK inhibitor SP600125 (20 mM; JNKi) for 1 h. Afterwards Sertoli cells were stimulated with TNFα (25 ng/ml) for 6 h. Relative mRNA expression of IL–1α (a), TNFα (b), MCP1 (c), TGFβ2 (d) and IL-6 (e) was normalized to Hprt; (n = 4–7, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 10

Schematic diagram illustrating induction of EAO in Wistar rats by active immunization with testicular homogenate (TH) in complete Freund’s adjuvant followed by i.v. injection of inactivated Bordetella pertussis bacteria. Control adjuvant animals received saline instead of TH.