Combining Top6 protein depletion with microSol IEFprefractionation and narrow pH range gels

We previously showed that prefractionating complex proteomes including unde-pleted serum using microSol IEF followed by analysis of fractions on very narrow pH range gels could isolate albumin in a single very narrow pH range fraction and could substantially expand the number of protein spots that could be reproducibly resolved [25]. Hence, in this study we evaluated the utility of combining Top-6 protein depletion with subsequent microSol IEF fractionation and analysis of each fraction on very narrow pH range gels. However, only a moderate further increase

Tab. 1 Proteins identified from silver-stained 2-D gels after MARS depletion

Spot Primary I Da|

1 Cemloplasmin (4557485, l)b'

2 Serum albumin (28592, 39)

Albumin precursor (4502027, 16)

4 Complement component lq, B chain (11038662, 7)

5 Complement component lq, A chain (7705753, 3)

6 Complement component lq, B chain

6a Complement component 3 (40786791, 6)

7 Complement component 3 (40786791, 15)

8 P-2-glycoprotein I precursor (4557327, 18)

9 P-2-glycoprotein I precursor (4557327, 8)

10 Complement component 4A preproprotein (14577919, 7)

11 Cemloplasmin (1620909, 1)

12 Transferrin (4557871, 38)

Other IDs

Complement C8-a chain precursor (25757820, 10) Peptidoglycan recognition protein L precursor (21361845, 8) Heparin cofactorll (123055, 7) a-2-macroglobulin (177872, 4) Complement component 3 (40786791, 4) Hemopexin (11321561, 1) Carboxypeptidase B2 (13937897, 2) Histidine-rich glycoprotein precursor (4504489, 10) Complement C4 binding protein a (4502503, 8) Inter-a-trypsin inhibitor heavy chain HI precursor (2851501, 9)

Complement component 3 (40786791, 1) Desmoglein 1 preproprotein (4503401, 1)

Migration inhibitoiy factor-related protein 14 variant E (7417329, 1)

Kallistatin precursor (1708609, 2) a-2-macroglobulin (177872, 1)

a-2-macroglobulin (224053, 4)

Ig heavy chain (106378, 2)

Coagulation factor XII precursor (182292, 1)

Tab. 1 Continued

Spot no.

Primary IDa|

Other IDs

13

Transferrin (4557871, 37)

a-2-macroglobulin (224053, 3)

Ig heavy chain (106378, 3)

Coagulation factor XII precursor (182292, 2)

14

Presemm amyloid P component (337758, 4)c'

Serum albumin precursor (4502027, 2)

Ig heavy chain V-region (197102, 1)

15

Preserum amyloid P component (337758, 5)c'

Antipneumococcal antibody 7C5 light chain variable region (21311315, 2)

16

Apolipoprotein M (22091452, 7)

Apolipoprotein A-I precursor (4557321, 3)

17

Hemopexin (11321561, 11)

Fibrinogen-a A (223918, 2)

18

Carboxypeptidase N (4503011, 3)c)

19

Fibrinogen-P chain precursor (399492, 17)

Hemopexin (11321561, 1)

Complement factor I precursor (116133, 1)

20

Complement C4A precursor (2144577, 3)

Complement factor H-related protein 2 (1064908, 2)

a) For spots where multiple proteins were identified, primary identifications were defined by the protein with the highest ion current for matched peptides.

b) Numbers in parentheses refer to the NCBI accession number followed by the number of unique peptides identified. MS/ MS spectra for all reported peptides identified have been manually validated.

c) Indicates proteins with expected concentrations of <0.1 mg/ml_.

Top 6 Protein Depletion

ZOOM IEF Fractionation

ln-gel Trypsin Digestion +

Protein Identification

Fig. 5 Scheme for detection of low-abundance proteins using major protein depletion followed by a multidimensional downstream separation strategy. To identify the maximum number of low-abundance proteins in plasma or serum samples, major protein depletion should be coupled with multiple downstream fractionation techniques, such as solution IEF and 1-D SDS-PAGE prior to LC-MS/MS to increase detection sensitivity.

in the number of reproducibly resolved spots was obtained when Top-6 depleted plasma was fractionated into seven pH ranges. Apparently due to the very wide dynamic range of concentrations even after depleting six of the most abundant spots, only about 2000-3000 total spots were resolved on a series of seven slightly overlapping narrow pH range 2-D gels (data not shown). Since this strategy increased the number of 2-D gels required to survey a complete proteome, this modest improvement in resolution did not justify the much higher workload. In contrast, microSol IEF fractionation followed by narrow pH range gels could reproducibly resolve more than 8000 protein spots when working with human cancer cell lysates due to the smaller dynamic range of protein abundance (data not shown).

Analysis of Top-6 depleted serum and plasma using protein array pixelation

The effectiveness of major protein depletion on enhancing detection of lower abundance proteins by non2-D gel methods was also assessed. Fig. 5 summarizes a scheme for a promising new protein profiling method that combines multiple dimensions of protein separations with one or more dimensions of peptide separations prior to MS/MS. After major protein depletion, the unbound fraction is alkylated and further fractionated by microSol IEF. Each microSol IEF fraction is then separated by 1-D SDS-PAGE and each lane is cut into uniform slices. The result is a 2-D protein array where each point or pixel on the array contains a group of proteins within a specific p I and Mr range.

To evaluate the effectiveness of protein depletion on detection of proteins by protein array pixelation, duplicate aliquots of the HUPO BDCA02-HEP human plasma were either not depleted or depleted of the top six proteins using the MARS HPLC column. Both samples were then separated into seven pH range fractions on a ZOOM-IEF fractionator, and fraction 3, pH 4.9-5.4, from each of the two samples

Tab. 2 Effects of Top-6 depletion on identification of proteins using the protein array pixelation methoda)

Number of peptide/proteinb)

Not depleted

Depleted

>4

37

54

4

4

1

3

7

12

2

11

8

1

50

85

Total

109

160

a) Number of protein identifications obtained from pixelation of a 1-D gel lane containing fraction 3 (pH 4.9-5.4). Redundant protein identifications within each gel lane have been deleted.

b) Peptides were filtered using the following criteria: Xcorr > 1.9 (z = 1), 2.2 (z = 2), 3.75 (z = 3) and ACn > 0.1, and Rsp < 4.

a) Number of protein identifications obtained from pixelation of a 1-D gel lane containing fraction 3 (pH 4.9-5.4). Redundant protein identifications within each gel lane have been deleted.

b) Peptides were filtered using the following criteria: Xcorr > 1.9 (z = 1), 2.2 (z = 2), 3.75 (z = 3) and ACn > 0.1, and Rsp < 4.

was separated on 1-D SDS-PAGE until the tracking dye had migrated 4cm on a minigel. The lane was sliced into 20 uniform slices prior to trypsin digestion and LC-MS/MS analysis on a Thermo LCQ-XP + IT mass spectrometer. The results comparing depletion with nondepletion for fraction 3 are summarized in Tab. 2. Substantially more unique proteins were identified after major protein depletion. Further analysis of the 66 proteins common to both samples indicates that 39 proteins were identified with more unique peptides in the depleted sample, 12 proteins have more peptides in the undepleted sample, and 15 proteins did not show any changes in the number of peptides identified. Of the 12 proteins identified by more peptides in the undepleted sample, three proteins were among those depleted by the MARS column, i.e., haptoglobin, albumin, and a-1-antitrypsin. The remainder of the 12 proteins from the undepleted sample had only one or two additional peptides being identified. In contrast, 30 of the 39 proteins identified by more peptides in the depleted sample had at least three additional matched peptides compared with the nondepleted sample. Hence, by depleting the major proteins, we increased the proportion of lesser abundance proteins identified in this single ZOOM-IEF fraction, leading to substantially more proteins identified and increased sequence coverage for identified proteins. However, even after major protein depletion, most of the high-quality MS/MS spectra data acquired matched high- or medium-abundance blood proteins. Since comparisons in this experiment were based on only one of seven fractions, the above differences could be extrapolated by about sevenfold if all fractions from depleted and nondepleted samples were compared.

Fig. 6 shows a representative gel from microSol IEF separation of ~200 mL depleted or nondepleted human plasma. Because Top-6 protein depletion reduced the total protein by about seven-fold, volumes ofdepleted sample fractions loaded on the gels were increased approximately seven-fold relative to nondepleted samples for comparison. Due to the sample simplification that occurred after major protein de-

Fig. 6 ZOOM-IEF fractionation of plasma proteins. Nondepleted (N), or depleted (D) BDCA02 human plasma (about 200 mL each) representing 16.5 or 2.4 mg total protein, respectively, were separated using the ZOOM-IEF fractionator into seven discrete pH pools (-700-800 mL final fraction volumes), run on 10% Bis-Tris 1-D gels and stained with colloidal CBB. All aliquots loaded to the gel (-1.7 mL for nondepleted sample, and -12 mL for depleted sample) were equivalent to about 35 mg of the original plasma protein.

pletion, much larger volumes of some fractions could be applied to 1-D SDS gels before band distortion due to overloading occurred. However, the next most abundant proteins limited 1-D gel loads and therefore limited detection of even lower abundance proteins. When protein array pixelation was applied to a complete human plasma sample proteome after depletion, the most abundant proteins detected as estimated based on sequence coverage are summarized in Tab. 3.

Was this article helpful?

0 0

Post a comment