Since N-linked carbohydrates are added exclusively in the ER lumen,
glycosylation is a good assay for translocation. Two of its limitations are obvious: if
your protein contains no N-linked carbohydrate addition sites (Asn-X-Ser/Thr) you
cannot use this assay. Since not all N-linked sites are used, and since a given site can
be found to be used in some but not all copies of a protein, failure to get glycosylated
does not necessarily mean the domain was not translocated. On the other hand, if a
domain is glycosylated, you can be essentially certain that it was translocated. Thus it
is an assay that is very specific but not necessarily sensitive -- high possibility of false
negatives. I typically immunoprecipitate first and endo H second rather than the
reverse. Carry out immunoprecipitation as usual to the stage of washed protein A -
beads ready to add SDS gel buffer. At this point, add 50 ul 0.1 M Na citrate pH 5.5,
0.1% SDS and vortex, then heat to 100°C for 2 minutes, spin out the beads and divide
the supernate into two aliquots. To one aliquot add 0.25 ul (a tiny bit) of
endoglycosidase H (concentration 1 unit/ml); to the other aliquot add nothing. Incubate
both at 37°C for 6 to 12 hours. Then dry both down in a speed vac, and add SDS gel
loading buffer as usual to process samples for PAGE.
WARNING: Unless you want to get tremendous variation in lane width, be sure to add 25
ul of citrate buffer to all samples for the gel, and dry them all down before adding SDS
gel sample buffer, e.g. put all samples in the same salts and volume.
If you want to endo H digest total products, simply take an aliquot of total
translation products (e.g. 1 or 2 ul), dilute into 50 ul of citrate buffer pH 5.5, heat, cool
to room temp, digest with endo H at 37°C, speed vac to dryness and dissolve in SDS
gel sample buffer.