Application of Two-Dimensional Gel Electrophoresis in Combination with Mass Spectrometry in the Study of Hormone Proteoforms

Hormone is a category of important endocrine regulatory proteins in human endocrine systems. Clarification of hormone proteoforms directly leads to understanding of its biological roles. Two-dimensional gel electrophoresis (2DGE) in combination with mass spectrometry (MS) plays important roles in identification of hormone proteoforms such as human growth hormone (hGH) proteoforms and human prolactin (hPRL) proteoforms. This book chapter will review the hormone proteoforms focusing on hGH and hPRL, the methodology of hormone proteoform study, and future perspective of human hormone proteoform study to find biomarkers for in-depth understanding of molecular mechanisms, and individualized and precise diagnosis, therapy, and prognostic assessment of hormone-related diseases.


Introduction
Hormone is a category of important endocrine regulatory proteins in the human endocrine systems. Hormone is a chemical message substance synthesized by highly differentiated endocrine cells and directly secreted into the blood, which has high biological activities and transmits information in the body as a messenger. It is a vital substance in human life and plays important roles in regulating physiological processes such as metabolism, growth, and development.
Notably, human growth hormone (hGH) and human prolactin (hPRL) are two key hormones in human body. The hGH, also known as somatotropin, is produced in the acidophilic somatotroph cells of the anterior pituitary gland and is a 191 amino acid single chain polypeptide, which is released into the blood circulation and takes part in the hypothalamic-anterior pituitary-skeletal muscle axis system to regulate growth and development in human body [1,2]. The synthesis and release of hGHs are affected by multiple complex feedback mechanisms, and the This book chapter will mainly review the hormone proteoforms focusing on hGH and hPRL, the methodology of hormone proteoform study, and future perspective to find effective and potential hormone proteoform biomarkers for in-depth understanding of molecular mechanisms, and individualized and precise diagnosis, therapy, and prognostic assessment of hormone-related diseases.

Tissues and protein extraction
The human control pituitary tissue samples were post-mortem tissues, and human pituitary adenoma tissue samples were obtained from neurosurgery. The detailed information of those tissue samples were described previously [15,16]. The collected tissues were immediately frozen in liquid nitrogen, and stored at −80°C until used. The protein extraction of pituitary control and adenoma tissues was performed as described previously [17,18]. In brief, the contaminated blood in each tissue sample was washed with 0.9% NaCl (3 mL, 5×). The proteins were extracted with protein extraction buffer that consists of 2 mol/L thiourea, 7 mol/L urea, 40 g/L CHAPS, 100 mmol/L dithiothreitol (DTT), 5 mol/L immobilized pH gradient (IPG) buffer pH 3-10 NL, and a trace of bromophenol blue, followed by centrifugation (15,000×g, 15 min, 4°C). The supernatant was collected as protein sample, and its protein concentration was determined with a Bio-Rad 2D Quant kit (Bio-Rad) [15].

2DGE and 2DGE-based Western blot
Each protein sample was first separated by IEF with pH 3-10 NL IPG strips (180 mm × 3 mm × 0.5 mm), under the IEF condition that was a gradient from 0 to 250 V within 1 h (125 Vh), a gradient from 125 to 1000 V within 1 h (500 Vh), a gradient from 1000 to 8000 V within 1 h (4000 Vh), a step-and-hold at 8000 V for 4 h (32,000 Vh), and a step-and-hold at 500 V for 0.5 h (250 Vh) to achieve a total of 36,875 Vh within ~7.5 h [19]. After IEF, proteins were reduced with DTT, and alkylated with iodoacetamide, and then were separated with the 12% SDS-PAGE resolving gel (250 mm × 215 mm × 1.0 mm) in the Tris-glycine-SDS electrophoresis buffer that contained 192 mmol/L glycine, 25 mmol/L Tris-base, and 1 g/L SDS with an electrophoresis condition (250 V, 360 min, 25°C) [20]. The 2DGE-separated proteins were visualized with silver-staining [21].
After 2DGE, the proteins were transferred onto a polyvinylidene fluoride (PVDF) membrane, blocked with bovine serum albumin (BSA), and incubated with hormone antibodies and secondary antibody. The proteins on the membrane were visualized with 5-bromo-4-chloro-3-indolyl phosphate [15].
The silver-stained 2D gels and 2DE-Western blot images were digitized and analyzed with Discovery Series PDQuest 2D Gel Analysis software [22,23].

MALDI-TOF MS
The tryptic peptide mixture was purified with ZipTipC18, and analyzed with a Perspective Biosystems MALDI-TOF Voyager DE-RP MS (Framingham, MA, USA). The PMF data were obtained to search Swiss-Prot database with PeptIdent software for protein identification. The detailed procedure was described previously [22].

LC-ESI-Q-IT MS
The tryptic peptide mixture was purified with ZipTipC18, and analyzed with LC-ESI-Q-IT MS on an LCQ Deca mass spectrometer (Thermo Finnigan, San Jose, CA, USA). The MS/MS data were obtained to search Swiss-Prot database for protein identification. The detailed procedure was described previously [22].

MALDI-TOF-TOF MS
The tryptic peptide mixture was purified with ZipTipC18, and analyzed with MALDI-TOF-TOF MS on Perspective Biosystems UltraFlex III MALDI-TOF-TOF (Bruker Daltonics). The MS and MS/MS data were obtained to search Swiss-Prot database for protein identification. The detailed procedure was described previously [15].

MS-identification of open reading frames (ORFs) of hGH and hPRL proteoforms within 2DGE pattern, and confirmed with Western blot
Over 1000 protein spots were found in each pituitary 2D gel (n = 30). MS is an effective method to identify ORFs of a gene across 2DGE map, generally cannot identify proteoforms with the common procedure because MS only detects a partial amino acid sequence but not its complete sequence of a protein. However, each proteoform has its specific pI and Mr, which can be separated with 2DGE. 2DGE coupled with MS can effectively array and identify proteoforms that are derived from the same gene. Thus, 24 hGH proteoforms ( Figure 1A) [1] and 6 hPRL proteoforms ( Figure 1B) [15] were MS-characterized within pituitary 2DGE map.
Some hGH proteoforms and hPRL proteoforms within 2DGE map (Figure 1) were also validated with Western immunoblot. For example, four hPRL proteoforms were confirmed with 2DGE-Western blot (Figure 2). It demonstrated that 2DGE-MS was an effective and reliable approach to detect and identify proteoforms derived from hGH and hPRL genes. The other hGH proteoforms and hPRL proteoforms were not be validated with immnuoblot, which might be due to no-reactivity of antibody to a specific proteoform.
The hGH proteoforms or hPRL proteoforms in each 2D spot were identified with MS including PMF and MS/MS. For example, the protein in spot 6 in Figure 1A was identified as hGH isoform 1 (P01241) with MALDI-TOF PMF data (Figure 3). The protein in spot v6 in Figure 1B was identified as hPRL (P01236) with MALDI-TOF-TOF PMF data (Figure 4), and MS/MS data ( Figure 5).
Therefore, 2DGE-MS clearly identified 24 hGH proteoforms and 6 hPRL proteoforms with different pI-M r distributions on a 2DGE pattern. Furthermore, Application of Two-Dimensional Gel Electrophoresis in Combination with Mass Spectrometry… DOI: http://dx.doi.org /10.5772/intechopen.82524 one investigated the reasons to form those hormone proteoforms, including signal peptide, splicing variation, and PTMs such as deamination, phosphorylation, and glycosylation.

MS-determination of signal peptide contained in each hGH or hPRL proteoform in human pituitaries
If signal peptide is contained in the sequence of hGH or hPRL, then it means that hormone is a prohormone (Figure 6). It is necessary to determine whether signal  peptide is contained in each hGH proteoform or hPRL proteoform. MS can identify the characteristic tryptic peptide sequences before and after removal of signal peptide in hGH or hPRL ( Table 1). For 24 hGH proteoforms in human pituitaries, checking all MS data of those 24 hGH proteoforms, the tryptic peptide sequence FPTIPLSR (position 27-34) (Figure 3) was detected for each hGH proteoform, but no any characteristic tryptic peptide sequence before removal of signal peptide was found ( Table 1). It clearly demonstrated that no signal peptide sequence was contained in those 24 hGH proteoforms in the 2DGE pattern ( Figure 1A). For 6 hPRL proteoforms in human pituitaries, checking all MS data of those 6 hPRL proteoforms, the tryptic peptide sequence MNIKGSPWK (position 1-9) (Figure 4) was detected for each hPRL proteoform, but no any characteristic tryptic peptide sequence after removal of signal peptide was found ( Table 1). It clearly demonstrated that signal peptide sequence was contained in those 6 hPRL proteoforms in the 2DGE pattern ( Figure 1B).

MS-identification of splicing variants of hGH in human pituitaries
MS is an effective method to identify splicing variants of hGH. hGH splicing variant 2 is derived from deletion of positions 58-72 from hGH, hGH splicing variant 3 is derived from deletion of positions 111-148 from hGH, and hGH splicing variant 4 is derived from deletion of positions 117-162 from hGH (Figure 6). The characteristic tryptic peptides for each hGH splicing variant can be theoretically obtained and observed by MS ( Table 2). The MS data of 24 hGH proteoforms were checked one-by-one. The results showed that hGH splicing variant 2 existed in spot 9, hGH splicing variant 3 existed in spots 2 and 23, and hGH splicing variant 4 existed in spot 13. hGH variant 1 existed in the rest 20 spots.
However, hPRL does not have splicing variants. The Mr of hPRL proteoforms was 26.0 kDa for v1, 26.1 kDa for v4, 25.9 kDa for v5, and 25.9 kDa for v6 in 2DGE pattern (Figure 1), which was almost equal to the Mr (25.9 kDa) of hPRL prohormone. Moreover, the Mr of hPRL proteoforms was 26.4 kDa for v2, and 27.9 kDa for v3, which is slightly greater than Mr (25.9 kDa) of hPRL prohormone. These data further confirmed that those six hPRL proteoforms were not generated from mature hPRL (position 29-227) but       from hPRL prohormone (positions 1-227) in human pituitary tissues. These data coupled with MS data clearly demonstrated six hPRL proteoforms were not derived from splicing.

MS/MS-identification of PTMs in hormone proteoforms
MS/MS was an effective method to identify the PTMs and their modification sites of hGH and hPRL to clarify the reasons of formation of hGH proteoforms and hPRL proteoforms. MS/MS identified phosphorylation at sites Ser-77, Ser-132, and Ser-176 in many hGH proteoforms ( Table 3). A representative MS/MS spectrum was shown to determine phosphorylation site Ser-77 in hGH proteoform 6 (Spot 6) (Figure 7). Also, deamidation was found in many hGH proteoforms. In addition, there would be other PTMs in hGH proteoforms that need to be further characterized.  For hPRL proteoforms, bioinformatics including NetPhos 3.1 Server (http:// www.cbs.dtu.dk/services/NetPhos) [24,25] predicted 14 pS sites, 5 pT sites, and 3 pY sites in the hPRL (Table 4), NetNGlyc 1.0 Server (http://www.cbs.dtu.dk/ services/NetNGlyc) [26] predicted ten significantly N-glycosylated sites ( Table  5), and NetOGlyc 4.0 Server (http://www.cbs.dtu.dk/services/NetOGlyc) [27] predicted six significantly O-glycosylated sites in the hPRL ( Table 6) in human pituitaries. These predicted PTM sites in hPRL proteoforms provided clues and needed to be confirmed with MS/MS in future studies.

Conclusions
hGH and hPRL are two important hormones in human endocrine systems, which are synthesized in the pituitary gland and secreted into the circulation system. Clarification of hGH proteoforms and hPRL proteoforms in human pituitary is essential step to elucidate their biological functions. Alternative splicing and PTMs are two important factors to cause proteoforms. 2DGE effectively presented 24 hGH proteoforms and 6 hPRL proteoforms with different pI-M r distributions in 2DGE pattern of pituitary tissue proteome. MS/MS effectively identified their splicing variants and PTMs: (i) 24 hGH proteoforms in pituitary removed their signal peptide, whereas 6 hPRL proteoforms in human pituitary did not remove their signal peptide.
(ii) 24 hGH proteoforms in human pituitary are derived from 4 types of alternative splicing variants, whereas 6 hPRL proteoforms do not exist any alternative splicing variants. (iii) PTMs pSer-77, pSer-132, and pSer-176 were identified in some of 24 hGH proteoforms, whereas although no PTMs were identified in hPRL proteoforms with MS/MS. However, phosphorylation, N-glycosylation, and O-glycosylation have been predicted with bioinformatics in hPRL proteoforms. Deamidation was presented in both hGH proteoforms and hPRL proteoforms. Therefore, 2DGE coupled with MS plays crucial roles in detection, identification, and quantification of hormone (hGH and hPRL) proteoforms, which benefits insight into the molecular mechanisms and discovery of effective biomarkers of hormone-related diseases.
© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. work and financial support. Z.T. participated in analysis of references and wrote partial manuscript.