You are here

| Neural Stem Cells

Do Glial Cells Aid NSC Therapeutics through Cell Fusion?



"Embryonic Stem Cell-Derived Neural Stem Cells Fuse with Microglia and Mature Neurons"

Recent studies into the transplantation of neural stem cells (NSCs) in animal models have suggested that this may represent a novel strategy in combating loss of function in human brain disorders. However, the proposed mechanisms by which this is accomplished are many; neuron replacement, supply of trophic factors, modulation of inflammation, stimulation of angiogenesis and neuroprotection, amongst others (Lindvall and Kokaia). Importantly, inflammation activates innate immune cells, such as microglia, which are known to fuse with mature resident neurons (Ackman et al.) possibly exerting a protective role (Alvarez-Dolado 2007) as fusion is enhanced by inflammation and tissue damage (Espejel et al., Johansson et al. and Nygren et al.). Now, in a study published in Stem Cells, researchers from the group of Zaal Kokaia at the Laboratory of Stem Cells and Restorative Neurology, University Hospital/Lund Stem Cell Center, Sweden have demonstrated that microglial cells do fuse with mature neurons and that they also mediate the fusion of NSCs with mature neurons (Cusulin and Monni et al.).

The authors cultured cortical cells from green fluorescent protein (GFP)+ mouse embryos, which contained around 80% Microtubule-associated protein 2 (Map2)+ neurons, 11% glial fibrillary acidic protein (Gfap)+ astrocytes and 5.5% Iba1+ microglia and, to this, added red fluorescent protein (RFP)+ NSCs derived from embryonic stem cells (ESCs). After three days, fluorescence analysis found around 5% of cells to be GFP+RFP+ with the majority of these cells expressing neuron-specific markers (NeuN, v-Glut1 or Map2) and the appearance of differentiated mature neurons. The discovery of two nuclei in 28% of the GFP+RFP+ cells further suggested that cells had fused. This was also observed when PKH26-labelled rat primary cortical cultures were used, with some double-positive cells also positive for the microglial marker Iba1. Selective removal of microglia from PKH26-labelled rat primary cortical cultures using a microglia-specific toxin (Mac1-saporin) reduced the level of fusion at three days after NSC addition from 3.5% to negligible levels, while re-introduction of microglia restored the fusion level back to normal. GFP+ NSCs were then co-cultured with RFP+ microglia which gave rise to 27% GFP+RFP+ cells after 2 days of culture. Fusion was confirmed through BrdU labelling of GFP+ microglia before co-culture, which led to the appearance of fused GFP+RFP+BrdU+ cells after NSC co-culture which increased in number with increased time in culture. Overall, this suggests that NSCs and microglia fuse in vitro.

Fluorescence-activated cell sorting (FACS) was then used to separate GFP+RFP+, GFP+ and RFP+ cells from co-culture experiments. All separated cell fractions continued to proliferate after replating, with the fused cell fraction being passaged and replated numerous times and remaining GFP+RFP+. Fused cells expressed NSC markers (Nestin, Sox2, Mash1) as well as microglial markers (CD11b, F4/80) and could be differentiated into neurons (positive for Map2, Gamma-aminobutyric acid (Gaba), BIII-tubulin or Doublecortin (Dcx)) or Gfap-expressing cells when differentiated under a neural or glial protocol, respectively. Functional analysis of fused cells using whole-cell patch-clamp recordings found that fused cells were electrophysiologically similar to NSCs, and less similar to microglia, although some characteristics were altered in the fused cells suggesting that the microglial influence may alter the membrane properties of fused microglia-NSCs. To explore this, fused cells were exposed to LPS (Lipopolysaccharide), which in microglia leads to the increased production of pro-inflammatory molecules (Inducible Nitric oxide synthases (iNOS) and Tumor necrosis factor alpha (TNFa)) or Interleukin (IL)-4 and IL-13 which leads to arginase and CD206 expression. mRNA analysis indeed found that the fused cells had responded to LPS in a microglial-specific fashion, although levels of the tested factors were lower in the fused cells, again suggesting the retention of some microglial properties in the fused cells. Mechanisms mediating cell fusion were then explored by analysing the requirement of phosphatidylserine (PS), which is required for myoblast (van den Eijnde et al.) and macrophage (Helming and Gordon) fusion as well as trophoblast formation (Huppertz et al.), through analysis of AnnexinV which binds to PS and expression of the scavenger receptor CD36, which is required on at least one of the cells for fusion to occur (Helming et al.). This found that NSCs constitutively exposed PS and the microglia expressed CD36 and, additionally, that the masking of PS by AnnexinV led to a 50% decrease in fusion rates.

Next the ability of the NSC-microglia fused cells to fuse with CellVue (fluorescent dye)-labelled mouse primary cortical cells was analysed. Co-culture of these cells followed by FACS found a population of 2.7% GFP+RFP+CellVue+ cells and exposure of these to cortical differentiation medium for 1 week led to the induction of a mature neuronal phenotype, of which some were Map2+, suggesting that the fused cells could fuse with mature neurons. Subsequently, co-culture experiments with rat fetal cortical cultures and GFP+ mouse microglia were undertaken which demonstrated the appearance of cells which expressed both GFP and neuronal markers (Map2, NeuN, and v-Glut1) after only 1 day. LPS-mediated activation of microglia increased the percentage of these cells by 87% compared to non-activated controls, while IL-4 treatment of co-culture experiments between mouse fetal cortex-derived GFP+ neurospheres and mouse RFP+ microglia led to an increase in the percentage of fused GFP+RFP+ cells of 4.4- and 3.7-fold at 3 and 6 days respectively.

Lastly, in vivo analysis was undertaken through the implantation of GFP+ mouse NSCs into the cerebral cortex of new-born rat pups. At three weeks, two populations were observed; the first was located in superficial cortical layers and showed an immature phenotype, but the second had characteristics of mature cortical pyramidal neurons and expressed NeuN. Few GFP+ cells were positive for Iba1. The use of species-specific antibodies (M2 and M6) which recognise antigens on neurons and glia of mouse origin only demonstrated that all GFP+ cells with mature pyramidal neuron morphology in rats grafted with GFP+ NSCs from mice were negative for M2/M6. Further analysis using an injected retrograde tracer (Lumafluor RedBeads) further indicated that any labelled pyramidal neurons were of host origin and had not differentiated from grafted NSCs. However, 17% of GFP+ pyramidal neurons had two distinct nuclei suggesting a fusion event between host neurons and grafted NSCs.

Overall the authors suggest that this work describes the first evidence of the fusion of ESC-derived NSCs with mature neurons, mediated partly through the interaction of PS from NSCs and CD36 receptor of microglia. In vitro analysis shows that fused NSC-microglial cells can go on to fuse with mature neurons but additional work will be required to show that this definitively happens during the in vivo fusion events observed. This work suggests that transplantation of NSCs for therapeutic reasons may rely heavily on the status of the microglial cell population; the presence and activation status of these cells may have an important influence on clinical outcome. Indeed, this data may give sufficient proof for the introduction the fusion of NSCs and activated microglia prior to transplantation.



Ackman JB et al.
Fusion of microglia with pyramidal neurons after retroviral infection.
J Neurosci 2006;26:11413–11422.

Alvarez-Dolado M.
Cell fusion: Biological perspectives and potential for regenerative medicine.
Front Biosci 2007;12:1–12.

Espejel S et al.
Radiation damage increases Purkinje neuron heterokaryons in neonatal cerebellum.
Ann Neurol 2009;66:100–109.

Helming L et al.
The scavenger receptor CD36 plays a role in cytokine-induced macrophage fusion.
J Cell Sci 2009;122:453–459.

Helming L, Gordon S.
Molecular mediators of macrophage fusion.
Trends Cell Biol 2009;19:514–522.

Huppertz B et al.
Trophoblast fusion: Fusogenic proteins, syncytins and ADAMs, and other prerequisites for syncytial fusion.
Micron 2006;37:509–517.

Johansson CB et al.
Extensive fusion of haematopoietic cells with Purkinje neurons in response to chronic inflammation.
Nat Cell Biol 2008;10:575–583.

Lindvall O, Kokaia Z.
Stem cells in human neurodegenerative disorders— Time for clinical translation?
J Clin Invest 2010;120:29–40.

Nygren JM et al.
Myeloid and lymphoid contribution to non-haematopoietic lineages through irradiation-induced heterotypic cell fusion.
Nat Cell Biol 2008;10:584–592.

Van den Eijnde SM et al.
Transient expression of phosphatidylserine at cell–cell contact areas is required for myotube formation.
J Cell Sci 2001;114: 3631–3642.



Study originally appeared in Stem Cells.

STEM CELLS correspondent Stuart P Atkinson reports on those studies appearing in current journals that are destined to make an impact on stem cell research and clinical studies.