Alleviation of neurotoxicity by microglial human Siglec-11

Abstract

Sialic acid-binding Ig superfamily lectins (Siglecs) are members of the Ig superfamily that recognize sialic acid residues of glycoproteins. Siglec-11 is a recently identified human-specific CD33-related Siglec that binds to alpha2,8-linked polysialic acids and is expressed on microglia, the brain resident innate immune cells. Polysialylated neuronal cell adhesion molecule (PSA-NCAM) is a putative ligand of Siglec-11. We observed gene transcription and protein expression of Siglec-11 splice variant 2 in human brain tissue samples by RT-PCR and Western blot analysis. Siglec-11 was detected on microglia in human brain tissue by immunohistochemistry. Human Siglec-11 splice variant 2 was ectopically expressed by a lentiviral vector system in cultured murine microglial cells. Stimulation of Siglec-11 by cross-linking suppressed the lipopolysaccharides (LPS)-induced gene transcription of the proinflammatory mediators interleukin-1beta and nitric oxide synthase-2 in microglia. Furthermore, phagocytosis of apoptotic neuronal material was reduced in Siglec-11 transduced microglia. Expression of PSA-NCAM was detected on microglia and neurons by immunohistochemistry and RT-PCR. Coculture of microglia transduced with Siglec-11 and neurons demonstrated neuroprotective function of Siglec-11. The neuroprotective effect of Siglec-11 was dependent on polysialic acid (PSA) residues on neurons, but independent on PSA on microglia. Thus, data demonstrate that human Siglec-11 ectopically expressed on murine microglia interacts with PSA on neurons, reduces LPS-induced gene transcription of proinflammatory mediators, impairs phagocytosis and alleviates microglial neurotoxicity.

Figure 1.

Detection of the splice variant 2 of Siglec-11 in human brain tissue. A, Gene transcript analysis of three different human brain tissues samples (H. brain 1, 2, and 3) by RT-PCR. Amplification with three different pairs of primers (primer pair 1, 2, and 3) from distinct exons resulted in PCR products of 103 bp, 352 bp, and 294 bp, respectively. All three PCR products were sequenced and showed gene sequence of Siglec-11 splice variant 2. Control, RNA of tissue sample 1 without reverse transcription. M. brain, cDNA from mouse brain tissue. Marker, 100 bp ladder. B, Western blot analysis of protein lysates from distinct human brain tissue samples (brain 1, 2, and 3), normal 293 cells (293), and 293 cells transduced with the Siglec-11 vector (293-Siglec-11). Siglec-11 protein was detected in the brain tissue and the positive control (293-Siglec-11). Antibody directed against β-actin, which labeled a 42 kDa protein, was applied as control. C, Immunofluorescence analysis of Siglec-11 in human brain tissue samples. Brain tissue sections were double immunolabeled with antibodies directed against Siglec-11 and Iba1. Siglec-11 expression was detected in microglial cells by confocal laser microscopy. Background staining was determined by control antibodies (control). Scale bars, 15 μm.

Figure 2.

Reduced phagocytosis of apoptotic neuronal material and suppression of proinflammatory gene transcription by Siglec-11 transduction in murine microglia. A, Representative images showing murine microglia transduced with the control (left image) or Siglec-11 vector (right image) and stained with antibodies directed against Iba1 (labeled with green fluorescence dye). Scale bars, 10 μm. B, Suppression of proinflammatory gene transcription by Siglec-11 transduction in murine microglia. Primary brain-derived microglial cells transduced with a flag-tagged Siglec-11 (flag-Siglec-11) vector or the GFP control vector (Control vector) were cultured on plates precoated with antibodies directed against the flag-tag (Flag Ab) or the isotype control antibody (Control Ab). Microglial cells were either unstimulated or stimulated with 500 ng/ml LPS. Gene transcripts of microglial cells were studied by real-time RT-PCR. Cross-linking of Siglec-11 prevented the LPS-induced increase in IL-1β and NOS2 gene transcription in Siglec-11, but not in control vector transduced microglia. Data are shown as mean ± SEM and as relative increase in LPS-induced gene transcription; number of independent experiments n = 4. *p < 0.05, ANOVA followed by Bonferroni multiple-comparison test.

Figure 3.

Detection of PSA-NCAM expression on cultured neurons and microglia. A, Hippocampal neurons were cultured from embryonic brain tissue. Neurons were double immunostained with specific antibody recognizing the PSA-containing form of NCAM and β-tubulin-III-specific antibody. Neurons were either untreated or treated with EndoN for 5 h before fixation. PSA-NCAM was detected on untreated neuronal cell bodies and neurites, but was absent on EndoN-treated neurons. Scale bar, 50 μm. B, Microglial cells were obtained from postnatal mixed glial cultures. Microglia were immunostained with monoclonal anti-PSA-NCAM antibody. Microglial cells were identified by double labeling with antibodies directed against Iba1. No immunostaining of PSA-NCAM was detected after removal of polysialic acids by EndoN. Scale bar, 50 μm. C, RT-PCR was performed using microglial and neuronal cDNAs. NCAM gene transcripts were detected in both microglia and neurons. Amplification of GAPDH was applied to confirm the integrity of the materials. Neg. control, Microglial mRNA samples without reverse transcription.

Figure 4.

Neuroprotection of microglial Siglec-11 in microglia–neuron cocultures dependent on PSA on neurons. A, Microglial cells were lentivirally transduced with the Siglec-11 (Siglec-11) or control vector (Control). PSA was removed from neurons and/or microglia by EndoN treatment. Cells were washed and cocultured for 48 h. Cells were fixed and double-immunostained with antibodies directed against β-tubulin-III (green signal) and Iba-1 (red signal). Confocal images were obtained by laser scanning microscopy. Neurite density was higher in cocultures with Siglec-11-expressing microglia and neurons having PSA residues on their glycoproteins. Scale bars, 50 μm. B, Statistics analysis of neurite density. The number of neurites was determined and values were normalized to those of cocultures with control vector transduced microglia. Data are shown as mean ± SEM; number of independent experiments n = 5, *p < 0.05, ANOVA followed by Bonferroni multiple-comparison test. C, Statistics analysis of neuronal cell body survival. The number of survived neuronal cell body in each treatment was counted and values were normalized to those of cocultures with control vector transduced microglia. Data are shown as mean ± SEM; number of independent experiments n = 5, *p < 0.05, ANOVA followed by Bonferroni multiple-comparison test.