Galectin-4 is a tandem-repeat galectin with two distinct carbohydrate recognition domains (CRD). Galectin-4 is expressed mainly in the alimentary tract and is proposed to function as a lipid raft and adherens junction stabilizer by its glycan cross-linking capacity. Galectin-4 plays divergent roles in cancer and inflammatory conditions, either promoting or inhibiting each disease progression, depending on the specific pathological condition. The study of galectin-4's ligand-binding profile may help decipher its roles under specific conditions. Here we present the X-ray structures of human galectin-4 N-terminal CRD (galectin-4N) bound to different saccharide ligands. Galectin-4's overall fold and its core interactions to lactose are similar to other galectin CRDs. Galectin-4N recognises the sulfate cap of 3'-sulfated glycans by a weak interaction through Arg45 and two water-mediated hydrogen bonds via Trp84 and Asn49. When galectin-4N interacts with the H-antigen mimic, 2'-fucosyllactose, an interaction is formed between the ring oxygen of fucose and Arg45. The extended binding site of galectin-4N may not be well suited to the A/B-antigen determinants, α-GalNAc/α-Gal, specifically due to clashes with residue Phe47. Overall, galectin-4N favours sulfated glycans whilst galectin-4C prefers blood group determinants. However, the two CRDs of galectin-4 can, to a less extent, recognise each other's ligands.
Hallqvist A, Wagenius G, Rylander H, Brodin O, Holmberg E, et al.
Lung cancer (Amsterdam, Netherlands) •
Several attempts to increase the locoregional control in locally advanced lung cancer including concurrent chemotherapy, accelerated fractionation and dose escalation have been made during the last years. As the EGFR directed antibody cetuximab has shown activity concurrent with radiotherapy in squamous cell carcinoma of the head and neck, as well as in stage IV NSCLC combined with chemotherapy, we wanted to investigate radiotherapy with concurrent cetuximab in locally advanced NSCLC, a tumour type often over expressing the EGF-receptor. Between February 2006 and August 2007 75 patients in stage III NSCLC with good performance status (PS 0 or 1) and adequate lung function (FEV1>1.0) were enrolled in this phase II study at eight institutions. Treatment consisted of 2 cycles of induction chemotherapy, docetaxel 75 mg/m² and cisplatin 75 mg/m² with 3 weeks interval. An initial dose of cetuximab 400 mg/m² was given before start of 3D-CRT to 68 Gy with 2 Gy per fraction in 7 weeks concurrent with weekly cetuximab 250 mg/m². TOXICITY was scored weekly during radiotherapy (CTC 3.0), and after treatment the patients were followed every third month with CT-scans, toxicity scoring and QLQ. Seventy-one patients were eligible for analysis as four were incorrectly enrolled. adenocarcinoma 49%, squamous cell carcinoma 39% and other NSCLC 12%. The majority had PS 0 (62.5%), median age 62.2 (42-81), 50% were women and 37% had a pre-treatment weight loss>5%. esophagitis grade 1-2: 72%; grade 3: 1.4%. Hypersensitivity reactions grade 3-4: 5.6%. Febrile neutropenia grade 3-4: 15.4%. Skin reactions grade 1-2: 74%; grade 3: 4.2%. Diarrhoea grade 1-2: 38%; grade 3: 11.3%. Pneumonitis grade 1-2: 26.8%; grade 3: 4.2%; grade 5: 1.4%. The median follow-up was 39 months for patients alive and the median survival was 17 months with a 1-, 2- and 3-year OS of 66%, 37% and 29% respectively. Until now local or regional failure has occurred in 20 patients and 22 patients have developed distant metastases. Weight loss, PS and stage were predictive for survival in univariate as well as in multivariate analysis. Induction chemotherapy followed by concurrent cetuximab and RT to 68 Gy is clearly feasible with promising survival. TOXICITY, e.g. pneumonitis and esophagitis is low compared to most schedules with concurrent chemotherapy. This treatment strategy should be evaluated in a randomised manner vs. concurrent chemoradiotherapy to find out if it is a valid treatment option.
This study was designed to examine the role of selectins and CD18 in leukocyte recruitment in hepatic injury induced by tumor necrosis factor-alpha (TNF-alpha) and galactosamine (Gal) in vivo. Intravital fluorescence microscopy of the hepatic microcirculation was used to quantify leukocyte-endothelium interactions provoked by 24 h of systemic TNF-alpha/Gal challenge in rats. Hepatic injury was evaluated with liver enzymes. When administered after 24 h of TNF-alpha/Gal challenge, fucoidan, a selectin-function inhibitor, reduced leukocyte rolling by 69%, whereas firm adhesion was unaltered. In contrast, passive immunization against CD18 decreased leukocyte adhesion by 60%, whereas rolling remained unchanged. Notably, when administered prior to TNF-alpha/Gal, fucoidan attenuated both leukocyte rolling and adhesion, by 57 and 69%, respectively. Pretreatment with an anti-CD18 antibody decreased TNF-alpha/Gal-induced rolling and firm adhesion by 25 and 90%, respectively. Moreover, pretreatment with fucoidan and the anti-CD18 antibody both protected against TNF-alpha/Gal-induced increases in liver enzymes. For example, the pretreatments reduced alanine aminotransferase by 59 and 87%, respectively. Our data suggest that TNF-alpha/Gal-induced leukocyte rolling is selectin-mediated and a precondition for CD18-dependent firm adhesion in hepatic venules. Thus, reducing leukocyte recruitment by inhibition of selectins or CD18 may be useful to control TNF-alpha-induced liver injury.
Linomide is an immunomodulator that ameliorates several autoimmune and inflammatory diseases. We assessed the effect of Linomide on microvascular perfusion failure, leukocyte recruitment and hepatocellular injury induced by tumor necrosis factor alpha (TNF-alpha) and D-Galactosamine (Gal). After 3 days of Linomide pretreatment (1, 10 and 100 mg/kg/day), rats were challenged with TNF-alpha/Gal for 24 h. Microvascular perfusion, leukocyte-endothelium interactions in hepatic postsinusoidal venules and leukocyte sequestration in sinusoids were evaluated using intravital microscopy. Liver enzymes were measured spectrophotometrically. Challenge with TNF-alpha/Gal significantly reduced sinusoidal perfusion, and increased leukocyte rolling, adhesion and liver enzymes. Interestingly, pretreatment with Linomide (10 and 100 mg/kg/day) significantly reduced TNF-alpha/Gal-induced leukocyte rolling by 65 and 63%, and leukocyte adhesion by 87 and 84%, respectively. Moreover, Linomide (10 and 100 mg/kg/day) decreased sinusoidal sequestration of leukocytes by 71 and 51%, and markedly improved sinusoidal perfusion. Moreover, Linomide reduced aspartate aminotransferase by 87-97%, and alanine aminotransferase by 79-96%. However, Linomide had no protective effect when administered concomitantly with TNF-alpha/Gal. These data demonstrate a dose-dependent inhibitory effect of Linomide on perfusion failure, leukocyte recruitment and hepatocellular injury provoked by TNF-alpha. Indeed, these findings suggest that Linomide may be an effective substance for protection of the liver in sepsis.
Isoglobotetraosylceramide (GalNAc(beta 1-3)Gal(alpha 1-3)Gal(beta 1-4)Glc (beta 1-1)Cer), the major glycolipid species in dimethylhydrazine-induced rat tumors of colorectal origin, was not detected in epithelial cells of normal colon but was present in the non-epithelial stroma and could be extracted from each of nine tumors studied. Monoclonal antibodies produced against isoglobotetraosylceramide detected this and another novel rat tumor-associated glycolipid not present in epithelial cells nor in non-epithelial stroma of normal rat colon (Brodin, T., Thurin, J., Strömberg, N., Karlsson, K.-A. and Sjögren, H.O. (1985) Eur. J. Immunol. 16, 951-956). This novel glycolipid was present in 8/9 of the studied tumors and was also present in two in vitro cell clones. These were originally obtained from a W49/T4 colon tumor isograft. The novel glycolipid was characterized by mass spectrometry, 1H-NMR, and methylation analysis as a hybrid between the isoglobo- and neolacto-series, with the structure GalNAc(beta 1-3)Gal(alpha 1-3)Gal(beta 1-4)GlcNA(beta 1-3)Gal (beta 1-4)Glc(beta 1-1)Cer.
Brodin T, Thurin J, Strömberg N, Karlsson KA, Sjögren HO
European journal of immunology •
Two species of neutral glycosphingolipids purified from rat colon carcinoma tissue, isoglobotetraosylceramide [GalNAc(beta 1----3)Gal(alpha 1----3)Gal(beta 1----3)Glc(beta 1----1)Cer] and a related 6-sugar "analogue" were inserted into liposomes together with lipid A (from bacterial lipopolysaccharide) and used for immunization of mice and monoclonal antibody production. The yield of hybridomas producing glycolipid-specific antibody was 5-10% using a high-dose booster schedule with liposome-inserted glycolipid. In contrast the frequency was below 0.1% (no glycolipid-binding antibodies were found) when using the previously described method of immunizing with glycolipid coated on the surface of acid-treated S. minnesota. Monoclonal antibodies were screened on the purified glycolipids used for immunization and selected for differential reactivity to the two glycolipids. A diversity of specificities was demonstrated by binding to the purified antigens, in a thin-layer chromatogram binding assay and in binding tests to tumor and normal target cells.
