The increased volume of serum or plasma that can be loaded onto 2-D gels after depleting six ofthe most abundant proteins should result in improved capacity to detect lower abundance proteins by this method. To evaluate the extent to which lower abundance proteins could actually be detected, silver-stained 2-D gels of high loads of depleted serum and plasma (Fig. 4) were compared to a 100 mg load of unfractionated serum or plasma. Many ofthe moderate and low-intensity protein spots that were detected on the depleted sample gels were actually detectable on the unfractionated sample gel. However, a few new spots that were below the detection limit on the unfractionated gel could be seen on the more heavily loaded post-depletion gels. These spots as well as several spots from the MARS spin column unbound fraction that were tentatively identified as incompletely depleted albumin and transferrin were excised (Fig. 4), digested with trypsin, and identified using LC-MS/MS. As anticipated, the spots in the serum albumin region (spots 2, 3) and the transferrin region (spots 12, 13) were identified as these proteins, which indicated they were incompletely depleted when the prototype spin column was used. Since the same antibodies were used by the same manufacturer to produce both products, it is most likely that this difference in efficiency was due to either an overloading ofthe spin column or, more likely, simply a difference in purification format, i.e., it is well known that batch purifications tend to be less effective than column chromatography using the same resin.
Although most of the analyzed samples were moderate or very faint silver-stained spots, more than one protein was identified for some spots (Tab. 1) due to the high-sensitivity of the linear IT mass spectrometer. Surprisingly, most of the new "low-abundance" spots that appeared at high protein loads were apparently minor forms of major proteins such as ceruloplasmin and complement proteins. The lowest abundance proteins detected were amyloid P serum component, at an expected protein concentration of 28 mg/mL , and carboxypeptidase N, with an expected plasma concentration of 30 mg/mL , indicating that the combination of depleting six major proteins and high-resolution broad-range 2-D gels are not sufficient for detection of proteins in the ng/mL range.
Was this article helpful?