Blood-Tissue barriers are found in the brain, eye, testis, and peripheral nerves.  These are formed of non-fenestrated endothelial cells and their interaction with glial processes in the brain and eye and astrocyte-like glial cells in the peripheral nerve ganglia, and interaction with Sertoli cells in the testis.  These are NOT absolute barriers to Ig but only relative barriers; most permit about 0.1% of circulating IgG to enter the brain; there is no barrier in the meninges  Within the brain parenchyma itself, some areas are more permeable than others such as occurs in the periventricular neural tissues. So despite dogma, a small portion of IgG enters these tissues.  References supporting these statements can be found in Chapter 11 of Preclinical Safety Evaluation of Biopharmaceuticals, Joy A. Cavagnaro, Ed, Wiley Publications, 2008 (Rojko J, et al., PhysiologicIgG biodistribution, transport and clearance: Implications for monoclonal antibody products, p241-276). 

Stuart, antibodies do not normally cross the BBB.  In fact, getting them to cross for therapeutic use has long been a holy grail for Pharma.   Here is a short write-up in Nature about some engineered IgG's that have therapeutic potential.

http://www.nature.com/news/2011/110525/full/news.2011.319.html#B1

My understanding about CNS IgGs in disease states is that increased BBB leakiness allows their passage.  But I am less knowledgable about this.

As far as I know brain nad testes are immune-privledged zones and IgGs will not migrate there under normal circumstances

Blood-Tissue barriers are found in the brain, eye, testis, and peripheral nerves.  These are formed of non-fenestrated endothelial cells and their interaction with glial processes in the brain and eye and astrocyte-like glial cells in the peripheral nerve ganglia, and interaction with Sertoli cells in the testis.  These are NOT absolute barriers to Ig but only relative barriers; most permit about 0.1% of circulating IgG to enter the brain; there is no barrier in the meninges  Within the brain parenchyma itself, some areas are more permeable than others such as occurs in the periventricular neural tissues. So despite dogma, a small portion of IgG enters these tissues.  References supporting these statements can be found in Chapter 11 of Preclinical Safety Evaluation of Biopharmaceuticals, Joy A. Cavagnaro, Ed, Wiley Publications, 2008 (Rojko J, et al., PhysiologicIgG biodistribution, transport and clearance: Implications for monoclonal antibody products, p241-276). 

William Hall: I cannot find reference to this: " most permit about 0.1% of circulating IgG to enter the brain; there is no barrier in the meninges " in Chapter 11 of the reference you give. Can you point me in the right direction here? Thanks.