I can't help you with ER from hippocampus, but there are protocols for isolation of ER from pancreas in the field of protein translocation, mostly for dog pancreas. The most famous protocol (I think) is from Walter&Blobel (1983). Maybe this protocol could provide a basis for optimization for preparation of hippocampus-ER.

Short version (all steps at 4°C, buffer pH 7,6): Prepare pancreas (dog), homogenize quickly with knife until it looks like baby food, then weigh and add 3 volumes of buffer (50mM HEPES, 50mM KAc, 6mM MgAc, 1mM EDTA, 1mM DTT, 250mM Sucrose, protease inhibitors), then homogenize further using a Dounce homogenisator. Pellet the debris (ER stays in the supernatant) in two steps (1.000g/10min + 6.000g/10min) and filter supernatant trough a house-hold tea sieve. Then centrifuge the solution through a sucrose cushion (Cushion buffer: 50mM HEPES, 50mM KAc, 6mM MgAc, 1mM EDTA, 1mM DTT, 1,6M Sucrose, protease inhibitors). After 20.000g/3,5h, the pellet (whitish-reddish) contains the ER. Resolve in a buffer containing 50mM HEPES and 250mM sucrose and homogenize. Clear sticky pellet contains lost ribosomes.

Quantification: Dilute in 2%SDS solution and measure absorption at 280nm. 1 OD280 = 1eq/µl (Walter et al., 1981).

Thankyou very much for this! This is somewhat similar to the methods I have trialed. I will see if I can modify my protocol to get a better outcome

Brain is somehow different from other organs as expected. I think you could pellet microsome from light membrane fractions. After that you could make a dis-continuous gradient to separate different membrane structures. After confirm by western blots of each fraction from gradient, you will use the fractions enriched ER markers like Bip or Grp78. There is a good protocol from this paper. You began from the so-called light membrane fraction. 

http://www.ncbi.nlm.nih.gov/pubmed/16617342