Prolactin Proteoform Pattern Changed in Human Pituitary Adenoma Relative to Control Pituitary Tissues

PRL gene-encoded prolactin is synthesized in the ribosome in the pituitary and then secretes into blood circulation to reach its target organ and exerts its biological roles, for example, involving in production, growth, development, immunoregulation, and metabolism. Multiple post-translational modifications and other unknown factors might be involved in this process to cause different prolactin proteoforms with differential isoelectric point (p I ) and relative mass ( M r ). Pituitary adenomas are the common disease occurring in pituitary organ to affect the endocrine system. Two-dimensional gel electrophoresis (2DGE) was used to separate prolactin proteoforms according to their p I and M r , followed by identification with Western blot and mass spectrometry (MS) analyses. Six prolactin proteoforms were identified in control pituitary tissues, and this prolactin proteoform pattern was significantly changed in different hormone subtypes of nonfunctional pituitary adenomas (NF − , LH + , FSH + , and LH + / FSH + ) and prolactinomas (PRL + ). Further, bioinformatics analysis revealed that different prolactin proteoforms might bind to different short- or long-PRL receptor-mediated signaling pathways. These findings clearly demonstrated that prolactin proteoform pattern existed in human pituitary and changed in different subtypes of pituitary adenomas. It is the scientific data to in-depth study prolactin functions, and to discover the prolactin proteoform biomarkers for PRL-related adenomas. controls, with 2DGE and MS. These findings provide the scientific data to in-depth study the PRL functions and to discover PRL proteoform biomarker for PRL-related adenomas.


Introduction
Prolactin (PRL) is a multifunctional hormone which is synthesized and secreted by pituitary [1]. Human PRL gene is located on chromosome 6 [2]. The secretory mode of PRL is autocrine and paracrine [3], and the secretion of PRL is pulsating and circadian rhythm [4]. The concentration of PRL in human serum has a certain reference range, and when its concentration is too high or too low, it will have a certain impact on the body. Dopamine can inhibit the secretion of

Pituitary tissue samples and preparation of protein samples
Eight human post-mortem control pituitary tissues, five PRL-positive prolactinoma tissues, three non-hormone expressed nonfunctional pituitary adenoma (NF-NFPA) tissues, three luteinizing hormone (LH)-positive NFPA tissues, three follicle-stimulating hormone (FSH)-positive NFPA tissues, and three LH/FSH-both positive NFPA tissues were used to extract proteins, with the previously described procedure [21,22]. The extracted protein of each tissue sample was used for 2DGE and MS analysis.

2DGE
A amount (70 μg) of proteins was diluted into 350 μL of protein sample buffer (7 mol/L urea, 2 mol/L thiourea, 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 rehydration (18 h, 20°C) of precast IPG strips pH 3-10 NL (180 x 3 × 0.5 mm) in 18-cm IPG strip holder on an IPGphor instrument, and IEF (25°C) with parameters (Gradient 250 V and 1 h for 125 Vh, gradient at 1000 V and 1 h for 500 Vh, gradient at 8000 V and 1 h for 4000 Vh, step-and-hold at 8000 and 4 h for 32,000 Vh, step-and-hold at 500 V and 0.5 h for 250 Vh to achieve a total of 36,875 Vh). After IEF, the proteins on IPG strip were reduced (15 min) with DTT, and alkylated with iodoacetamide, followed by separation with 12% SDS-PAGE (250 × 215 × 1.0 mm) in an Ettan DALT II system (GE Healthcare, up to 12 gels at a time) with a constant voltage (250 V, 360 min). All 2DGE-arrayed proteins were stained with silver-staining [23], and then digitized and analyzed with Discovery Series PDQuest 2D Gel Analysis software [24,25]. Each sample was performed for 3-5 times.

In-gel digestion with trypsin and MS identification of PRL
The proteins in each Western blot-positive spot was performed in-gel digestion with trypsin, purification of tryptic peptides with ZipTipC18, followed by analysis with three types of MS instruments, including MALDI-TOF MS [24], LC-ESI-Q-IT MS [24], and MALDI-TOF-TOF MS [9]. The detailed procedure was described previously [9,24]. The obtained peptide mass fingerprint (PMF) and tandem mass spectrometry (MS/MS) data were used to search Swiss-Prot human database for protein determination and PTM analysis.

The amino acid sequences of human PRL prohormone and mature PRL
In human pituitary, the PRL prohormone is synthesized in the ribosome, with 227 amino acids (position 1-227; 25.9 kDa), containing a signal peptide (position 1-28) ( Table 1), which was assigned with Swiss-Prot accession No. P01236. However, the mature human PRL only contains 199 amino acids (position 29-227; 22.9 kDa), which removed the signal peptide (position 1-28), and secreted into the circulation system to bind to its target organ for exerting PRL function.

PRL proteoform pattern in human pituitaries
The PRL proteoform pattern was found in human pituitaries. Qian et al. [9] found six PRL proteoforms with 2DGE in human pituitaries and then verified four of six PRL proteoforms with 2DGE-based Western blot in human pituitaries (Figures 1 and 2). The pI and M r of these PRL proteoforms are slightly different. Each PRL proteoform was digested with trypsin, and followed by MS and MS/MS analysis (Figures 3 and 4). The characteristic tryptic peptide are calculated to determine whether the signal peptide (position 1-28) in each PRL proteoform ( Table 2), which was compared to the observed ions of each PRL proteoform. It found all PRL proteoforms all contained the tryptic peptide sequence MNIKGSPWK (position 1-9), which clearly demonstrated that six PRL proteoforms are all PRL prohormone, but not mature PRL.

PRL proteoform changes in human pituitary adenomas compared to controls
The PRL proteoform pattern changed in different subtypes of pituitary adenomas compared to control pituitaries ( adenoma relative to control pituitaries was decreased or unchanged. The proportional ratio of six PRL proteoforms among five subtypes of pituitary adenomas was changed ( Table 4 and Figure 5). In FSH + /LH + and PRL + pituitary adenomas, the proportion of PRL proteoform v1 is the largest. In FSH + pituitary adenoma, the proportion of PRL proteoform v5 is the largest. The PRL proteoform changes suggest their scientific merit for clinical application.

Bioinformatics prediction of potential factors to form PRL proteoforms and pathway networks
PRL is a hormone which is secreted by pituitary gland. PRL has a variety of biological functions. Only when it reaches a specific target organ and binds to its receptor can it play its biological function (Figure 6). PRL can bind to short PRL receptor or long PRL receptor and then plays its biological functions. The long or short PRL receptors definitely bind to different PRL proteoforms. PRL proteoforms are definitely derived from a PRL gene undergoing splicing, transcription, translation, PTMs, translocation/re-distribution, and interaction with other molecules, etc. Therefore, phosphorylation sites in hPRL (position 1-227) were predicted with NetPhos 3.1 Server with a score more than 0.5. It obtained 22 statistically significantly phosphorylation sites in hPRL (position 1-227). N-glycosylation sites in hPRL (position 1-227) were predicted with NetNGlyc 1.0 Server with score more than 0.5. It obtained 10 statistically significant N-glycosylation sites in hPRL (position 1-227). O-glycosylation sites in hPRL (position 1-227) were predicted with NetOGlyc 4.0 Server with score more than 0.5. It obtained six statistically significant O-glycosylation sites in hPRL (position 1-227). These data suggest that PTMs such as phosphorylation and glycosylation might be the important reason to cause the PRL proteoforms.

Potential clinical application of PRL proteoform pattern
Prolactin synthesized in the ribosome in the pituitary secretes into blood circulation to reach its target organ and exert its biological roles, which is closely associated with multiple physiological and pathological processes, including pituitary adenomas. This study found six PRL proteoforms with different with differential isoelectric point (pI) and relative mass (M r ) in control pituitary tissues, which were identified with 2DGE coupled with Western blot and MS. This prolactin     proteoform pattern was significantly changed among different hormone-subtypes of nonfunctional pituitary adenoma (NF − , LH + , FSH + , and LH + /FSH + ) and prolactinoma (PRL + ) tissues. This result suggests the potentially important clinical value of serum PRL proteoforms. The reason is that pituitary tissues are impossible to obtain for clinical diagnosis, and prolactin must secrete into blood to exert its biological roles, we strongly believe serum PRL proteoforms exist and the serum PRL proteoform pattern changes among different pituitary adenomas. Therefore, we will further analyze serum PRL proteoform pattern changes among different subtypes of pituitary adenomas, and develop the PRL proteoform pattern as biomarker for prediction, diagnosis, or prognostic assessment of pituitary adenoma occurrence, progression, and prognosis.

Conclusions
Six PRL proteoforms were identified in human pituitary tissue with 2DGE and MS analyses, and four of six PRL proteoforms were validated with 2DGE-based Western blot, MS, and MS/MS analyses. There were significant differences in PRL proteoform pattern among five different subtypes of pituitary adenomas (LH + , NF − , FSH + , FSH + /LH + , and PRL + ) (P < 0.05). Moreover, MS analysis revealed that six PRL proteoforms are PRL prohormone. PRL proteoforms might be derived from PTMs such as phosphorylation, deamidation, and glycosylation. Further, different PRL proteoforms might bind to different PRL receptors to produce different physiological functions. These findings provide scientific basis for in-depth understanding of pituitary adenomas, and help develop biomarkers for treatment of pituitary adenoma patients. The serum PRL proteoform pattern has important clinical application value for prediction, diagnosis, and prognostic assessment of pituitary adenomas.