†These authors contributed equally.
Academic Editors: Zahid Pranjol and Georgios Giamas
Background: Protein kinase G type II (PKG II) is a serine/threonine-protein kinase that was originally isolated from the small intestinal mucosa with primary functions in the secretion of small intestinal mucosal cells, secretion of renin and aldosterone, and chondrocyte activities. Recent studies have shown that PKG II exerts anti-tumor effects, while a previous study by our group confirmed that PKG II inhibited the proliferation and migration of cancer cells. Interestingly, PKG II, which was typically bound to the intracellular side of the membrane, was detected in the serum and cell culture medium as a diagnostic biomarker of tumor growth. Thus, the aim of the present study was to elucidate the function and the targets of PKG II, and the mechanism underlying the secretion of this kinase. Methods: Construction of peptides and plasmids, RNA interference, Immunoelectron microscopy, Co-immunoprecipitation, N-glycosylation assay and Isolation of the Golgi apparatus were applied to investigate the secretory mechanism, and the targets and function of PKG II. Results: PKG II was secreted by enterochromaffin (EC) cells, which were components of the endocrine system in the gastrointestinal tract. Myristoylation of glycine 2 and the N-terminal sequence, especially the amino acids 3–30, acted as a signal peptide to induce the secretion of PKG II via the conventional protein secretory pathway. Moreover, recombinant PKG II inhibited the epidermal growth factor (EGF)-induced activation of the EGF receptor via phosphorylating the T406 of the extracellular domain and blocked EGF-triggered proliferation of various cancer cells. Conclusions: These results revealed a correlation between the endocrine system and the secretion of protein kinase, suggesting a novel protein secretory pathway. The resuls also indicated that secreted PKG II was a potential diagnostic biomarker and an inhibitor of tumor.
Protein Kinases are a heterogeneous family of enzymes modulating several biological pathways, including cell survival, differentiation and apoptosis [1]. As a serine/threonine-protein kinase, Protein kinase G type II (PKG II) was first detected as a membrane-bound receptor of cyclic guanosine monophosphate (cGMP) in the intestinal epithelium [2] and later found to bind to the cellular membrane via a myristoyl moiety attached to the N-terminus of the amino acid (aa) chain [3]. PKG II has been implicated in several physiological functions, including intestinal secretion, bone growth, and learning and memory [4], and more recently in regulation of the sodium channels of intestinal epithelial cells and mechanical signaling of osteoblasts [5, 6, 7]. In addition, PKG II plays a role in regulating cell proliferation and apoptosis [8, 9], which might be potentially associated with tumorigenesis. Our group confirmed that PKG II could inhibit the proliferation and migration of gastric cancer cells by blocking epidermal growth factor (EGF)-induced activation of the epidermal growth factor receptor (EGFR) [10, 11]. Surprisingly, an investigation of the mechanism underlying the influence of PKG II on EGFR found that PKG II could be secreted into the extracellular fluid and PKG II serum levels of cancer patients were significantly downregulated, and the above results indicated that the PKG II could be used as a tumor diagnostic marker. However, since PKG II is an intracellular protein without a signal peptide, the effect and roles of secreted PKG II remain unclear. Therefore, the aim of the present study was to clarify the mechanism of secretion and the extracellular effects and targets of PKG II.
The gastric cancer cell lines AGS and HGC-27, colon cancer cell line SW480,
hepatic cancer cell line HepG2, breast cancer cell line MCF-7, human embryonic
kidney cell line 293, and fibroblast-like cell line COS-7 derived from monkey
kidney tissue were obtained from the Institute of Biochemistry and Cell Biology,
Shanghai Institute for Biological Sciences, Chinese Academy of Sciences
(Shanghai, China). The gastric mucosal epithelial cell line GES-1 and the gastric
cancer cell lines SGC-7901 and BGC-823 were got from Jiangsu University (Jiangsu,
China). F-12K medium (Kaighn’s modification of Ham’s F-12 medium), Dulbecco’s
modified Eagle’s medium (DMEM), and fetal bovine serum (FBS) were purchased from
Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Plasmids encoding the
complementary DNA (cDNA) of mutant EGFRs were constructed in our laboratory.
Adenoviral vectors encoding the cDNA of PKG II (Ad-PKG II) were kind gifts from
Dr. Gerry Boss and Dr. Renate Pilz of the University of California, San Diego
(San Diego, CA, USA). Recombinant glutathione S-transferase (GST)-tagged PKG II
(GST-PKG II) was obtained from SignalChem Lifesciences Corporation (Richmond, BC,
Canada). Antibodies (Abs) against
Female nude BALB/c mice (age, 6 weeks) were purchased from the Animal Center of Yangzhou University (Yangzhou, Jiangsu, China) and housed in the Animal Center of Jiangsu University in compliance with the Guide for the Care and Use of Laboratory Animals (NIH,76 FR 91; May 11, 2011). The protocols of the mouse experiments were approved by the Institutional Animal Care and Use Committee of Jiangsu University. Human serum samples were collected from healthy volunteers (n = 24), patients with gastric cancer (n = 42), and patients with colorectal cancer (n = 20) at the Affiliated Hospital of Jiangsu University (Zhenjiang, China). The study protocol involving human subjects was approved by the Medical Ethics Committee of Jiangsu University and written informed consent was obtained from each subject prior to participation in this study.
The cells were cultured in DMEM or F-12K medium supplied with 10% FBS under a
humidified atmosphere of 5% CO
The culture medium was harvested into polypropylene tubes and centrifuged at
2000
Protein samples were separated based on molecular size by electrophoresis using
8–12% agarose gels and then transferred onto polyvinylidene fluoride membranes.
The membranes were blocked with 5% (w/v) non-fat milk in Tris-buffered saline
with 0.1% Tween® 20 detergent for 1 h at room temperature, then
incubated with primary Abs at 4
Briefly, 2000–5000 cells in 100
Tumor growth was induced by subcutaneously injecting the flank of the nude mice
with 2
The cells were plated on glass coverslips in the wells of a 24-well plate,
treated, and then fixed with 4% paraformaldehyde for 12 h at 4
Tissues were fixed in 4% paraformaldehyde and 0.05% glutaraldehyde solution in
0.1 M sodium phosphate buffer (pH 7.0) for 2 h, rinsed with 0.1 M phosphate
buffer (pH 7.0), dehydrated via a graded ethanol series, and embedded in LR White
(London Resin Company Limited, Stansted, Essex, England, UK). After ultraviolet
radiation for 72 h at –25
Cells grown on a 10-cm culture dish were washed two times with cold PBS and
lysed with cold immunoprecipitation assay buffer (Thermo Fisher Scientific,
Inc.). The supernatant was obtained by centrifugation of the lysate at 12,000
The cells were treated and then scraped into 1 mL of cold PBS, centrifuged at
600
The GA was extracted from HGC-27 cells (5
Serum samples were collected from healthy female BALB/c mice intraperitoneally
injected with 10 mg/kg of Brefeldin A for 3 days. In addition, serum samples were
collected from healthy female BALB/c mice intraperitoneally injected with 40
mg/kg of DBA once every 3 days, 6 times in total. For both groups, blood samples
were collected at 48 h after the last injection by puncturing the posterior
venous plexus of the eyeball, stored at room temperature for 2 h, and then at 4
Site mutation Gly 2
For the deletion plasmids PKG II-Del(3–8), Del(3–14), Del(3–20), Del(3–26), Del(3–30), and Del(3–34), the HindIII and XbaI sequences were included to the 5′-ends of the upstream and downstream primers. For the PKG II-Del(21–32) deletion plasmid, the HindIII and XbaI sequences were added to the 5′-end of the upstream primer for the shorter segment and the 5′-end of the downstream primer for the longer segment. The following primers were used for the deletion plasmids in Table 1. The PCR amplifications were performed according to the protocol of the Site-Directed Mutagenesis kit. The correct plasmids and pcDNA3.1-c-flag plasmid were digested with the restriction endonucleases HindIII and XbaI, then the products were collected for T4 ligation, and identified by digestion with HindIII and XbaI and sequenced.
PKG II deletants | Primer sequence |
Del(3–8) | F: 5′-AAGCTTCCTGAGCAACATGGGAAAGCATTC-3′ |
Del(3–14) | F: 5′-AAGCTTCCTGAGCAACATGGGAGACGGCCA-3′ |
Del(3–20) | F:5′-AAGCTTGAGCAACATGGGACTCAGCAATGAA GCCCTG-3′ |
Del(3–26) | F:5′-AAGCTTGAGCAACATGGGACGGAGCAAAGTGGCAGAG-3′ |
Del(3–30) | F:5′-AAGCTTGAGCAACATGGGAGCAGAGCTGGAGCGCGAG-3′ |
Del(3–34) | F:5′-AAGCTTGAGCAACATGGGACGCGAGGTGAAGAGGAAG-3′ |
R: 5′-TCTAGACTCTCCTGTCAGAAGTCCTTATCC-3′ | |
Del(21–32) | F: 5′-AAGCTTATATCGAATTCCGTTGCTGTC-3′ |
R: 5′-GTTCCCTGACTGGCCGTC-3′(short) | |
F: 5′-CTGGAGCGCGAGGTGAAG-3′ | |
R: 5′-TCTAGACTCTCCTGTCAGAAGTCCTTATCC-3′(long) |
Short-interfering RNAs (siRNAs) against SRP54 and a negative control were synthesized by RiboBio Co., Ltd. (Guangzhou, China). The nucleotide sequences of three siRNAs against SRP54 were GAAATGAACAGGAGTCAAT, GCAAGAGGATCGGGTGTAT, and GATCCTGTCATCATTGCTT reapectively. Transfection with siRNA was performed using Lipofectamine™ 2000 reagent. After 4–6 h, the siRNA/lipid complexes were removed and the cells were maintained in complete medium for 48–72 h. Afterward, the cells were harvested and the cell lysates were collected for WB analysis to detect the siRNA-induced inhibition of SRP54 protein expression.
The peptides of PKG II-Myr-(2–30)-Flag (Myr-GNGSVKPKHSKHPDGHSGNLTTDALRNKVDYKDD-DDK) and PKG II-(2–30)-Flag (GNGSVKPKHSKHPDGHSGNLTTDALRNKVDYKDDDDK) were synthesized by ChinaPeptides Co., Ltd. (Hangzhou, China).
The first 30 aa of PKG II were ligated into the plasmid pcDNA3.1-C-eGFP (6143 bp) to construct the peptide PKG II-(1–30)-GFP with the cloning sites HindIII and BamHI. The RhoA sequence was ligated to the plasmid pcDNA3.1+C-DYK (5438 bp) to construct the RhoA-Flag plasmid. The first 30 aa of PKG II were ligated to the N- terminus of RhoA-Flag to construct the peptide PKG II-(1–30)-RhoA-Flag. The above plasmids were constructed by Nanjing Genscript Biotechnology Co., Ltd. (Nanjing, China).
The cells were serum-starved for 12 h, incubated with Alexa Flour 488-EGF, then fixed with 2% paraformaldehyde and washed with PBS prior to Confocal Microscopy and Flow Cytometry. For the detailed steps, please refer to Thomas’ paper [14].
The cells were seeded into the wells of a six-well plate at a density of 1
About 200
The reaction mixture, which included 20
The data were expressed as the mean
PKG II was first identified as a membrane-bound cGMP receptor in the intestinal epithelium. Interestingly, during the investigation of PKG II,we identified that the cells containing PKG II are similar with the intestinal endocrine cells of human and mouse (Fig. 1). This finding was especially interesting because it remained unclear whether PKG II was only present within endocrine cells or was actually secreted. Chromogranin A (CGA) is often used as a biomarker of enterochromaffin (EC) cells, which are the most common type of enteroendocrine cells in the gastrointestinal (GI) tract [15]. Based on the characteristics and location of these cells, the relationship between PKG II and chromogranin A was investigated. The confocal microscopy results showed that there is colocalization of PKG II and CGA in the same cell (Fig. 2A), hinting that PKG II might be secreted by EC cells and convey endocrinological activities. Furthermore, the results of immunoelectron microscopy confirmed that cells in GI tissues had secretory vesicles that contained CGA, while other secretory vesicles contained PKG II (Fig. 2B).
PKG II is located in the GI tract. There were at least a group of cells containing high-level PKG II in the human and mouse gastrointestinal mucosal tissues by Immunohistochemistry.
The secretion of PKG II was a physiological phenomenon, and its
secretion level was related to gastric and colon cancer. (A) Co-location of PKG
II and Chromogranin A (CGA) was sporadically distributed cells in the
gastrointestinal tract, detected by Confocal Microscopy. (B) The secretory
vesicles of the cells contained PKG II (up panel) and CGA (bottom panel) were
detected by Electro-Microscopy, and the positively labeled vesicles were
indicated by red arrows. (C) PKG II was detected in the serum of normal subjects
(n = 24); patients with colorectal (n = 20) or gastric (n = 42) cancer. All blood
samples were taken from the patients before treatment, and PKG II in the serum
samples were detected by the ELISA kit according to manufacturer’s
specifications. The levels represent the median
PKG II was detected in the serum of healthy volunteers and cancer patients, but
the levels were comparatively downregulated in the serum of patients with colon
or gastric cancer (754.1
PKG II was found in the culture medium of different cells and
the blood from the mouse. (A) PKG II was detectable in the culture medium of
gastric mucosal epithelial cell line GES-1 and gastric cancer cell lines AGS and
HGC-27 infected with the virus vector encoding cDNA of PKG II (Ad-PKG II) by
ELISA assay. (B) Detection of PKG II in the serum of mouse and the concentration
of PKG II was increased in the serum of the mouse injected with Ad-PKG II
(10
As PKG II has no classical signal peptide, as predicted by the SignalP 6.0 Prediction Server (Fig. 4), PKG II might be secreted via a non-classical secretory pathway, such as the lysosome or exosome secretory pathway. But the Confocal Microscopy results showed that PKG II did not colocalize with lysosome-associated membrane protein-1 (LAMP1) (Fig. 5A) and the results of WB detection showed that PKG II was absent in the exosomes extracted from HGC-27 cells infected with Ad-PKG II (Fig. 5B). The above results indicated that PKG II was not secreted by lysosomes or the exosome secretory pathways. However, in Ad-PKG II-infected cells, PKG II was located surrounding the nucleus or accumulated in one side of the nucleus, similar to the location of the endoplasmic reticulum (ER) and Golgi (Fig. 5C, middle). Furthermore, Confocal Microscopy results showed that PKG II was co-localized with both glucose-6-phosphatase (G-6-Pase) and Golgi matrix protein 130 kD (GM130), which were biomarkers of the ER and Golgi, respectively [16, 17], indicating that PKG II was located in the ER and Golgi both in the cell lines and small intestine tissues (Fig. 5C,D). To further clarify that PKG II was indeed located within the ER cavity and not attached to the membrane surface, Co-IP was performed to detect the binding between PKG II and glucose-regulated protein 78 (GRP78), which acts as the master of the unfolded protein response process in the lumen of the ER [18] (Fig. 5E). Meanwhile, N-glycosylation of PKG II, an essential post-translational modification in the ER cavity, was also detected by enzymatic digestion [19] (Fig. 5F). The results showed that PKG II was not only bound to GRP78, but also N-glycosylated (Fig. 5E,F), confirming that PKG II was located within the ER cavity. PKG II was also detected in the Golgi Extract by WB (Fig. 5G). In addition, brefeldin A (BFA), an inhibitor of protein transport from the ER to Golgi [20], decreased the PKG II content in cell culture medium (Fig. 5H) and mouse serum (Fig. 5I). These results indicated that PKG II was likely secreted via the conventional ER-Golgi pathway [21].
Prediction of Signal peptide in PKG II-AA chain by SignalP-6.0. There was no signal peptide for PKG II.
PKG II could be secreted by the conventional secretion pathway.
(A) The location of PKG II and LAMP1 (a biomarker of lysome) in the HGC-27 cells
infected Ad-PKG II, detected by Confocal Microscopy. (B) The detection of PKG II
in the exosomes by Western blotting. (C) Co-location of PKG II with G-6-Pase and
GM130 in Ad-PKG II infected HGC-27 cells, detected by Confocal Microscopy. (D)
Co-location of PKG II with G-6-Pase and GM130 in human small intestine tissue
detected by Confocal Microscopy. (E) The binding of PKG II with GRP78, detected
by Co-IP. (F) Evidence of N-glycosylation-modulation of PKG II. The molecular
size of native PKG II but not recombinant GST-PKG II was decreased by PNGase F
which could remove almost all N-linked oligosaccharides. (G) Location of PKG II
in isolated Golgi apparatus, detected by Western blotting. (H) Result of ELISA
assay showing the down-regulation of PKG II content in cell culture medium by
ER-Golgi transport inhibitor Brefeldin A (BFA).HGC-27 cells were infected with
Ad-PKG II for 24 h and treated with BFA (50 ng/mL) for another 24 h. (*p
The N-terminal signal peptide directs the newly synthesized aa chain to enter the ER cavity for secretion via the conventional protein secretory pathway [22]. However, PKG II has no signal peptide. Reportedly, N-terminal myristoylation is associated with anchoring of PKG II to the cellular membrane [23]. In this experiment, the results of Co-IP and confocal microscopy confirmed the binding and co-localization of PKG II and N-myristoyl-transferase 1 (NMT1), which was an enzyme that regulates N-terminal myristoylation [24] (Fig. 6A,B). PKG II expression in mouse serum was downregulated by the NMT1 inhibitor DBA [25] (Fig. 6C), confirming the role of myristoylation modulation in the secretion of PKG II in vivo.
Glycine 2 (G2) myristoylation and N-terminal peptide
(3rd–30th aa) of PKG II were the key factors for inducing PKG II to
enter the ER-Golgi secretion pathway. (A) The interaction between NMT1 and PKG
II, detected by Co-IP. (B) Co-location between PKG II and NMT1 in HGC-27 cells
infected with Ad-PKG II, detected by Confocal Microscopy. (C) Detection of PKG II
in mouse serum and the content was decreased by treatment with NMT1 inhibitor DBA
(40 mg/kg, once every 3 days, 6 times in total). (*p
Myristoylation of PKG II occurs through modification of G2 by the covalent attachment of myristate, while mutation of G2 to alanine (G2A) blocks this modification [23]. The results of this experiment showed that PKG II was not detected in the cell culture medium and had no longer co-localized with G-6-Pase and GM130 in cells with the nonmyristoylated PKG II mutant (G2A) (Fig. 6D,E), demonstrating that PKG II secretion was dependent on G2 myristoylation. Considering that not all G2 myristoylation-modulated proteins were secreted, we inferred that other key aa residues might induce PKG II to enter the ER. Thus, several truncated forms of PKG II were constructed [i.e., PKG II-Del(3–8), Del(3–14), Del(3–20), Del(3–26), Del(3–30), Del(3–34), and Del(21–32)]. Notably, most of these truncated forms [i.e., Del(3–8), Del(3–14), Del(3–20), Del(3–26), and Del(21–32)] had no obvious effect on the secretion of PKG II. However, the truncated forms Del(3–30) and Del(3–34) abolished the secretion of PKG II (Fig. 6F), confirming that aa 3–30 were important for secretion of PKG II. These results indicated that both G2 myristoylation and aa 3–30 were key factors for the secretion of PKG II via the conventional secretory pathway.
To further clarify the mechanism underlying the entry of PKG II into the ER, the relationship between PKG II and signal recognition particle 54 (SRP54) was investigated. SRP54 is key to inducing newly synthesized peptides/aa chains to enter the ER [26]. The Co-IP results showed that PKG II could bind with SRP54 (Fig. 6G) and PKG II secretion was decreased by silencing of SRP54 with siRNA (Fig. 6H). The Co-IP results showed that binding between SRP54 and PKG II was decreased by both G2 mutation (PKG II-G2A-Flag plasmid) and deletion of aa 3–30 (PKG II-Del 3rd-30th-Flag plasmid) (Fig. 6I). Furthermore, there is binding for PKG II and SRP54 both in the cell-tansfected with plasmid coding 1–30 of PKG II (PKG II-1st-30th-Flag) or the synthesized peptides containing aa 2–30 with G2 myristoylation (PKG II-Myr-2nd-30th-Flag), but the binding is decreased when there is only 2–30 aa without G2 myristoylation (PKG II-2nd-30th-Flag peptides) (Fig. 6J). Finally, Therefore, it appeared that both G2 myristoylation and the sequence of aa 3–30 acted as a “signal peptide” to induce newly synthesized PKG II-aa modified to enter the ER by binding with SRP54. Finally, insertion of aa 1–30 of PKG II into the non-secretory proteins GFP and RhoA promoted secretion of the corresponding proteins (Fig. 6K), further confirming the above mechanism of secretion.
To determine the extracellular function of PKG II, recombinant PKG II was used to imitate secreted PKG II. In vitro, the results of the CCK-8 assay showed that recombinant PKG II significantly inhibited EGF-trigged proliferation of multiple types of cancer cells (Fig. 7A). In vivo, recombinant PKG II with 8-pCPT-cGMP significantly inhibited the growth of transplanted tumor cells (Fig. 7B), indicating that PKG II plays an anti-cancer role upon secretion. Furthermore, recombinant PKG II was found to block the binding of fluorescently labeled EGF to EGFR, which happened on the outside of the membrane (Fig. 7C,D).
The secreted PKG II could block the EGF-induced activation of
EGFR. (A) SGC-7901, SW480, MCF-7, and HepG2 cells were respectively seeded in
96-well plates for 24 h. After serum-starved for 12 h, the cells were incubated
with recombinant PKG II at indicated concentrations and treated with/without 250
To further confirm the extracellular function of PKG II, recombinant PKG II was added to the culture medium of AGS and HGC-27 cells and its inhibitory effect on EGFR was assessed by WB analysis. The WB results showed that the EGF-induced phosphorylation of EGFR, ERK, and Akt were decreased by pretreatment with recombinant PKG II plus 8-pCPT-cGMP for 1 h (Fig. 7E). Meanwhile, the results of an in vitro reaction system showed that incubation of the isolated membranes with recombinant PKG II plus 8-pCPT-cGMP inhibited EGF-stimulated phosphorylation of EGFR-Y1068 (Fig. 7F). Furthermore, the isolated membrane reaction system with the use of poly (Glu, Tyr) as a substrate to detect EGFR activity [27] revealed that recombinant PKG II also inhibited EGF-induced phosphorylation of the tyrosine residue of the Glu/Tyr polymer (Fig. 7G), indicating that secreted PKG II could block EGF-induced tyrosine phosphorylation/activation of EGFR.
To further investigate the interaction between PKG II and EGFR, prokaryotic expression vectors encoding for different domain segments EGFR or PKG II were constructed. The results of the pull-down assay showed that the PKG II bound to the fragment containing aa 331–645 of the extracellular domain of EGFR (Fig. 8A,C). Meanwhile, the results also confirmed that the fragments containing aa 1–176 and 1–285 of PKG II bound to EGFR (Fig. 8B,D). The above results showed that PKG II could act on EGFR from the outside of the cell, thereby serving as a secretory protein kinase. Therefore, the His-tagged recombinant extracellular fragment of EGFR was incubated with recombinant GST-tagged PKG II in the in vitro reaction system. The results showed the Ser/Thr phosphorylation of the EGFR fragment was increased in the presence of recombinant PKG II (Fig. 8E, left), and that recombinant PKG II bound to the extracellular fragment of EGFR (Fig. 8E, right), indicating that recombinant PKG II could induce Ser/Thr phosphorylation of the recombinant extracellular fragment of EGFR.
The secreted PKG II could bind with EGFR and induced the
phosphorylation of T406 on EGFR. (A,B) Identification of the binding domains of
EGFR and PKG II. The prokaryotic expression vectors coding cDNA of different
fragments of EGFR and PKG II were constructed, and the fragments were expressed
in E. coli. The Pull-down method was applied for detecting the binding between
different fragments of PKG II and EGFR in vitro. (C) Schematic of EGFR
fragments/domains. (D) Schematic of PKG II fragments/domains. (E) Recombinant PKG
II increased the p-Ser/Thr of a recombinant extracellular fragment of EGFR
through binding with the fragment in vitro. The reaction mixtures
included 570 nM His-extracellular fragment (1-645aa) of EGFR (His-EGFR), 150 nM GST-PKG II
or 400
The Group-based Prediction System (version 2.1) [28] predicted that threonine 406 (T406), threonine 727 (T727), and serine 945 (S945) were potential PKG II-specific phosphorylation sites of EGFR. Hence, various mutants of EGFR (i.e., T406P, T727P, and S945P) were constructed using a Site-Directed Mutagenesis Kit with the plasmid p-cDNA3.1-Flag-EGFR. COS-7 cells were transfected with the corresponding plasmids and the mutant EGFRs expressed by the cells were isolated by Co-IP and then incubated with recombinant PKG II plus 8-pCPT-cGMP in an in vitro reaction mixture. The results showed that the mutations of T727P and S945P had no effect on phosphorylation of Ser/Thr of EGFR. However, the mutation of T406P decreased PKG II-induced Ser/Thr phosphorylation of EGFR (Fig. 8F). Besides, recombinant PKG II inhibited EGF-induced phosphorylation of Tyr1068/Tyr1173 of the T727P and S945P mutants, but not the T406P mutant (Fig. 8G), indicating that T406 was the key site which was phosphorylated by secreted PKG II to block the activation of EGFR.
Most protein kinases are located within the cell bound to the membrane or distributed within the cytoplasm. However, some protein kinases were recently found to be secreted the outside of the cell via different protein secretory pathways in order to phosphorylate extracellular proteins [29, 30, 31]. Research on the phosphorylation of extracellular proteins has been ongoing for more than 100 years [32]. Recently, Klement et al. reported that phosphorylation of extracellular proteins was an important modification with potentially important biological significance [33]. However, the protein kinases responsible for phosphorylation of extracellular proteins remained unclear. Starting in about the year 2012, several studies identified that several kinases (i.e., family with sequence similarity 20, member C, vertebrate lonesome kinase, and pyruvate kinase isozyme M2) were secreted to the outside of the cell for phosphorylation of extracellular proteins [29, 30, 31, 34], thereby representing a new era of research on the phosphorylation of extracellular proteins.
PKG II was first described in 1983 as an intracellular membrane-bound serine/threonine protein kinase [2, 3, 35]. In the present study, PKG II was detected in cell culture medium as well as the serum of healthy volunteers with lower concentrations in the serum of cancer patients. These findings present convincing evidence that PKG II is also a secreted protein kinase. Most proteins are secreted via the conventional protein secretory pathway. However, proteins which are secreted via this pathway usually have an N terminal signal peptide composed of 15–30 aa that can bind with SRP54 and the SRP54 receptor to direct newly synthesized aa chains to enter the ER for synthesis and then mature in the GA, prior to secretion via secretory vesicles [21]. However, this theory cannot explain the secretion of all proteins because those without signal peptides might be secreted via the unconventional protein secretory pathway [36, 37].
Since PKG II has no signal peptide, the origin and the mechanism underlying the secretion of PKG II remain unclear. Usually, the core of the signal peptide contains a long stretch of about 5–16 hydrophobic aa residues [38], which is absent in PKG II. However, the second aa of the PKG II chain is a glycine residue, which can undergo myristoylation to increase the hydrophobicity of the N-terminus [23]. Myristoylation is the NMT-mediated modulation on the N-terminal glycine in the aa chain [24]. Previous studies have shown that myristoylation also regulates the intracellular location and the secretion of some proteins. For example, myristoylated adenylate kinase 2 of the protozoan Plasmodium falciparum can penetrate the plasma membrane and move to the outside of the cell [39]. In addition, secretion of extracellular protein kinase A by tumor cells is inhibited by NMT [40], while myristoylation promotes the translocation of protein kinase C in human and animal serum across the cell membrane [41, 42]. Therefore, the ability of N-terminal myristoylation to induce PKG II to enter the ER- GA secretory pathway was investigated and the results confirmed that G2 myristoylation was closely related to the secretion of PKG II.
The prevalence of proteins with a myristoylated N-terminus in eukaryotes is estimated at 0.5–3% of the cellular proteome, depending on the species and predictive model [43, 44]. This study is the first to show that the N-terminale myristoylated protein without a signal peptide can be secreted via the conventional secretory pathway. So, there must be other factors regulating the secretion, such as the aa sequence of PKG II. In this study, truncation of the aa sequence confirmed that aa 3–30 were necessary for the secretion of PKG II, suggesting that myristoylation of G2 and aa 3–30 acted as a “signal peptide” to induce the secretion of PKG II. This finding also provided a clue about the mechanism underlying the secretion of other proteins without signal peptides.
Another interesting finding was that PKG II was secreted by EC cells, which were important components of the enteroendocrine cell system in the GI tract, and could secrete/release more than one GI hormone or peptide [45]. EC cells were a subset of chromaffin cells that contained many vesicles stained by potassium dichromate [46, 47] and were first discovered in the adrenal medulla [48]. Later, EC cells were also found in the GI tract and could secrete 5-hydroxytryptaphane (serotonin), cholecystokinin, and secretin [49], which regulated the motility of the GI tract [50]. The results of this study confirmed that PKG II was secreted by EC cells, indicating a potential correlation between the endocrine system and the secretion of protein kinases. Further investigations of the effects and targets of the secreted PKG II confirmed that the N-terminus of PKG II (aa 1–176) bound to the C-terminus of the extracellular fragment of EGFR (aa 331–645), blocking the activation of EGFR through phosphoralyting its T406.
In summary, this study confirmed that PKG II was secreted by the EC cells in gastrointestinal mucose via the conventional protein secretory pathway without the present of typical “signal peptide” which was required for the secretion, and the myristoylated G2 and aa 3–30 of the amino acid chain of PKG II acted as a “signal peptide” to initiate the secretion of the kinase (Fig. 9). These results revealed a new protein secretion pattern and a correlation between the endocrine system and the secretion of protein kinases. It was also confirmed that the level of secreted PKG II in blood was related to tumogenesis and the secreted PKG II could block the activation of EGFR through phosphoralyting its therione 406 and inhibit the growth of tumor cells. Our results not only confirmed the phenomenon and mechanism of PKG II, but also affirmed that secreted PKG II has an anti-cancer effect and might be useful for the diagnosis of colon and gastric cancer, indicating that detection of PKG II in the blood and body fluids, similar with hormones, is predictive of tumorigenesis.
Schematic diagram of PKGII secretion and targets.
PKG II, Protein kinase G type II; EC, enterochromaffin; EGF, epidermal growth factor; CGMP, cyclic guanosine monophosphate; EGFR, epidermal growth factor receptor; WB, Western blot; BSA, bovine serum albumin; Co-IP, Co-immunoprecipitation; GA, Golgi apparatus; ELISA, enzyme-linked immunosorbent assay; siRNAs, Short-interfering RNAs; GI, gastrointestinal; ROC, Receiver operating characteristic; ER, endoplasmic reticulum; NMT1, N-myristoyl-transferase 1; SRP, signal recognition particle.
YW and YCC designed the research study. MW, ZBW, JP, MLZ, TL, XYY, HQ, and YJZ performed the research. XYL, LJ, YT provided help and advice on the research. MW and TL analyzed the data. YW and YCC wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.
Studies involving animals include a statement on ethics approval. The study protocol involving human subjects was approved by the Medical Ethics Committee of Jiangsu University and written informed consent was obtained from each subject prior to participation in this study (attached in the Supplementary materials). Animal license number: SYXK(Su)-2013-0036.
We thank LetPub (www.letpub.com) for linguistic assistance and pre-submission expert review. And we also thank Gerry Boss and Renate Pilz of the University of California (San Diego, CA, USA) for providing the adenoviral constructs.
This study was supported by grants from the National Natural Science Foundation of China (grant number 31771564 and 81201959); the Natural science fund for colleges and universities in Jiangsu Province (grant number 17KJB310001); Funding from the Health and Health Commission of Jiangsu Province (grant number LGY2018025); “333 project” scientific research project (grant number BRA2019172).
The authors declare no conflict of interest.