That depends. What exactly are you trying to do? Short or long term studies? Mouse vs other species?

Assuming that you're working in mice then what about using one of the macrophage reporter strains either as the tumour host or by adoptive transfer? eg Mac-green, Mac-blue, CX3CR1-GFP, LysM-crexGFP. note that not all strains are equal and each of these has distinct patterns of reporter gene expression in different monocyte/mac populations.

I agree with Andrew. It strongly depends on the experiment, time span for study and the question being asked. The ideal method is adoptively transfer macrophages from strains like Mac-green, Mac-blue, CX3CR1-GFP etc preferably bone marrow chimera. In Mac-Green (cfms-GFP) mouse dendritic cells, macrophages, neutrophils and monocytes are GFP positive. This expression gets a bit restricted in Mac-blue mice. In CX3CR1-GFP monocytes are positive and most probably monocyte derived macrophages should be positive. So if you just want peritoneal macrophages use Mac green as most macrophages populations are positive but if you want to look long term for monocyte derived macrophages then CX3CR1 or LysM-CreGFP maybe the best. Cautionary note: Macrophages are not a uniform population and the subpopulation you use will determine the outcome of the experiment, so choose wisely.

If you want to label you cells for MRI follow up, maybe can you check Riou et al paper published in NMR in biomedicine (2012) untitled MRI assessment of the intra-carotid route for macrophage delivery after transient cerebral ischemi or Cai et al paper published in Nanomedicine: Nanotechnology, Biology and Medicine (2012) untitled Magnetic resonance imaging of superparamagnetic iron oxide-labeled macrophage infiltrates in acute-phase renal ischemia-reperfusion mouse model.

Thanks All

Is it possible to transfer the macrophages with GFP and label them with SPION, and then image with MRI, stain with prussian blue, and read the GFP with confocal microscope?