Trp250-hK2 is defective in intracellular trafficking and activates the unfolded protein response Eun Ju Choi1, Sei Mee Yoon1, Suman Lee2* and Jinu Lee1* 1
College of Pharmacy, Yonsei Institute of Pharmaceutical Sciences, Yonsei University, 162-1 Songdo-dong, Yeonsu-gu, Incheon 406-840, Korea 2 Division of Structural and Functional Genomics, Center for Genome Science, National Institute of Health, Osong, Chungcheongbuk-do 363-951, Korea
hK2, a member of the kallikrein protease family encoded by KLK2, is expressed exclusively in prostate and is a putative adjunct tumor marker for prostate cancer screening. The T allele of rs198977, a single nucleotide polymorphism in exon 5 of KLK2, codes for W-hK2 and is associated with lower serum hK2 levels and higher risk of prostate cancer than the C allele encoding R-hK2. To elucidate the mechanism that underlies this SNP’s function, we transfected plasmids expressing R-hK2 or W-hK2 into PC3, HeLa and HEK293A cells and measured the hK2 level in cell lysates and conditioned media. The level of W-hK2 was lower than R-hK2 in conditioned media but was not different from R-hK2 in cell lysates. W-hK2 was hardly colocalized with Golgi-targeted fluorescent protein whereas R-hK2 colocalized. Reporter assays related to the unfolded protein response (UPR) and phospho-eIF2a immunoblot showed that W-hK2 increased UPR activity more than R-hK2. These results indicated that W-hK2 had a defect in cellular trafficking from the ER to the Golgi complex due to its misfolding and that it activated the UPR, suggesting a mechanism to explain the association of the T allele with higher prostate cancer risk.
Introduction Prostate-specific antigen (PSA) is encoded by KLK3, a member of the human kallikrein gene family. Although PSA is an important tumor marker for prostate cancer screening, its positive predictive value is low, and researchers have tried to find adjunct biomarkers (Catalona et al. 1993; Gann et al. 1995). Another member of the kallikrein family, KLK2, encodes hK2 and represents one such putative adjunct tumor marker to PSA for prostate cancer screening (Nam et al. 2000). As hK2 is expressed exclusively in prostate (Chapdelaine et al. 1988), elevated serum hK2 levels are associated with proliferative diseases of the prostate such as prostate cancer and benign prostate hypertrophy (Haese et al. 2000). There have been several epidemiological studies on a single nucleotide polymorphism (SNP), rs198977 Communicated by: Shuh Narumiya *Correspondence: [email protected] (JL) or smnl93@ gmail.com; [email protected] (SL)
512
Genes to Cells (2015) 20, 512–520
(C/T), located in exon 5 of the KLK2 gene (Nam et al. 2003, 2006; Klein et al. 2010; Kohli et al. 2010; Savblom et al. 2014). The T allele results in Trp250hK2 (hereafter W-hK2), whereas the common C allele codes for Arg250-hK2 (hereafter R-hK2). Serum hK2 levels are threefold lower in subjects with the TT genotype compared to those with the CC genotype (Nam et al. 2006; Klein et al. 2010). The cutoff value of serum hK2 concentration in prostate cancer screening thus needs to be adjusted according to the genotype of this SNP. The physiological role of PSA and hK2 is in male fertility. After ejaculation, semenogelins in semen form a coagulum to immobilize sperm in it. PSA is secreted from the prostate and mixed into the semen at a high concentration as an inactive precursor. As hK2 activates PSA, PSA gradually cleaves semenogelins, leading to the release of sperm from the coagulum, called semen liquefaction (Robert & Gagnon 1999). A recent report showed the association of seminal plasma concentration of hK2 and rs198977 in young, healthy men (Savblom et al. 2014). When this
DOI: 10.1111/gtc.12242 © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd
Mechanism for the function of a SNP, rs198977
function of hK2 is taken into consideration, the T allele of rs198977 may also be a risk factor of male infertility. The cellular mechanism underlying this important functional SNP has not been elucidated. The T allele of rs198977 has been reported as a risk factor for prostate cancer. This allele was found to be more abundant in the patient group with prostate cancer than in the control group (Nam et al. 2003, 2006; Klein et al. 2010). How the T allele promotes cancer development is another important question. The accumulation of unfolded or misfolded proteins in ER is referred to as ER stress. To handle this stress, cells activate several signal transduction pathways called unfolded protein response (UPR) (Ron & Walter 2007). The activation of the UPR retards cellular protein translation and enhances folding and degradation of unfolded proteins in ER. The role of ER stress in tumorigenesis is controversial (Ma & Hendershot 2004). There are many reports that prolonged UPR activates various apoptotic pathways of cancer cells (Breckenridge et al. 2003). Conversely, cells can become resistant to apoptosis via many signaling pathways, including the activation of nuclear factor kappa B and the degradation of p53 by ER stress (Pahl & Baeuerle 1995; Qu et al. 2004; Li et al. 2011). If any member of the pro-apoptotic signaling pathways in cells with chronic ER stress is disarmed by acquiring a mutation, chronic ER stress could become a key promoter of cancer. In addition, the activation of UPR is essential for many types of tumor cells to survive hypoxic stress and fold key signaling molecules (Davenport et al. 2008). In this study, we report decreased cellular secretion of W-hK2 compared to R-hK2 due to a defect in intracellular transport to the Golgi complex. We also show increased activity of the UPR in cells expressing W-hK2 compared to R-hK2, suggesting a possible mechanism of its enhancement of tumorigenesis.
to normalize the hK2-mycHis level to the loaded protein amount and the transfection efficiency, levels of actin and NPT II were also assayed, respectively. The levels of W-hK2-mycHis in CM were fourfold lower than those of R-hK2-mycHis in PC3 cells and HEK293A cells. In HeLa cells, just a twofold lower level was detected in the CM (Fig. 1A–C). As we obtained the similar results from the experiments with plasmids expressing hK2 without a tag and hK2 with amino-terminal Ypet, a YFP variant, we excluded the misleading effect of the carboxy-terminal tag (Fig. 2). To gain insight into the mechanism of this difference in expression, the levels of intracellular hK2-mycHis and KLK2-mycHis transcript were measured. The intracellular levels of hK2-mycHis were not significantly different according to genotype in all three tested cell lines. No significant difference in KLK2-mycHis transcript level was observed between the two genotypes (Fig. 3). Mechanism of the difference in hK2 level in CM according to genotype
Results
Based on these results, we hypothesized that W-hK2mycHis is more unstable after secretion than R-hK2mycHis and/or that W-hK2-mycHis has a defect in intracellular trafficking. To test the stability hypothesis, we prepared the CM containing Ypet-W-hK2 and transferred it to PC3-R or PC3-W cells, followed by further incubation for 18 h. Ypet-hK2 could be distinguished from hK2-mycHis due to its higher molecular weight. If the stability hypothesis was correct, the Ypet-W-hK2 level would be reduced more readily in PC3-W than PC3-R cells. Figure 4A shows the scheme of this experiment. The immunoblot showed that the amount of YpetW-hK2 did not change on further incubation in PC3-R or PC3-W (Fig. 4B). These results enabled us to reject the hypothesis that W-hK2 is less stable than R-hK2. The alternative hypothesis was that tryptophan at the 250th residue of hK2 impaired intracellular trafficking of hK2.
Expression of hK2 in conditioned media (CM) and cell lysates (CL)
Intracellular localization of hK2
To investigate the expression level of hK2 in CM and CL, we transfected PC3, HEK293A and HeLa cells with pR- or pW-hK2-mycHis. hK2 was tagged with mycHis at the carboxy terminus to differentiate hK2 expressed by transfected DNA and endogenous hK2. CM and CL samples were prepared and used in immunoblotting with an anti-His antibody. In order
To examine whether W-hK2 could pass through the ER and be transported to the Golgi complex, we transfected HEK293A cells with pmTurquoise2-Golgi combined with pYpet-R- or pYpet-W-hK2. Cells were transferred to glass-bottom confocal dishes and examined with laser scanning confocal microscope. When Ypet-R-hK2 was expressed, most mTurquoise2
© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd
Genes to Cells (2015) 20, 512–520
513
E J Choi et al. Table 1 Genetic and functional differences of R-hK2 and W-hK2 Allele of rs198977 250th amino acid of hK2
C
T
Arginine (R-hK2)
Tryptophan (W-hK2)
Allele frequency Intracellular production of hK2 Cellular disposal of hK2 Intracellular degradation Secretion Expression level of hK2 Intracellular In conditioned media Trafficking to Golgi complex Prostate cancer risk Unfolded protein response
>* >† > > > > † > > > >
College of Pharmacy, Yonsei Institute of Pharmaceutical Sciences, Yonsei University, 162-1 Songdo-dong, Yeonsu-gu, Incheon 406-840, Korea 2 Division of Structural and Functional Genomics, Center for Genome Science, National Institute of Health, Osong, Chungcheongbuk-do 363-951, Korea
hK2, a member of the kallikrein protease family encoded by KLK2, is expressed exclusively in prostate and is a putative adjunct tumor marker for prostate cancer screening. The T allele of rs198977, a single nucleotide polymorphism in exon 5 of KLK2, codes for W-hK2 and is associated with lower serum hK2 levels and higher risk of prostate cancer than the C allele encoding R-hK2. To elucidate the mechanism that underlies this SNP’s function, we transfected plasmids expressing R-hK2 or W-hK2 into PC3, HeLa and HEK293A cells and measured the hK2 level in cell lysates and conditioned media. The level of W-hK2 was lower than R-hK2 in conditioned media but was not different from R-hK2 in cell lysates. W-hK2 was hardly colocalized with Golgi-targeted fluorescent protein whereas R-hK2 colocalized. Reporter assays related to the unfolded protein response (UPR) and phospho-eIF2a immunoblot showed that W-hK2 increased UPR activity more than R-hK2. These results indicated that W-hK2 had a defect in cellular trafficking from the ER to the Golgi complex due to its misfolding and that it activated the UPR, suggesting a mechanism to explain the association of the T allele with higher prostate cancer risk.
Introduction Prostate-specific antigen (PSA) is encoded by KLK3, a member of the human kallikrein gene family. Although PSA is an important tumor marker for prostate cancer screening, its positive predictive value is low, and researchers have tried to find adjunct biomarkers (Catalona et al. 1993; Gann et al. 1995). Another member of the kallikrein family, KLK2, encodes hK2 and represents one such putative adjunct tumor marker to PSA for prostate cancer screening (Nam et al. 2000). As hK2 is expressed exclusively in prostate (Chapdelaine et al. 1988), elevated serum hK2 levels are associated with proliferative diseases of the prostate such as prostate cancer and benign prostate hypertrophy (Haese et al. 2000). There have been several epidemiological studies on a single nucleotide polymorphism (SNP), rs198977 Communicated by: Shuh Narumiya *Correspondence: [email protected] (JL) or smnl93@ gmail.com; [email protected] (SL)
512
Genes to Cells (2015) 20, 512–520
(C/T), located in exon 5 of the KLK2 gene (Nam et al. 2003, 2006; Klein et al. 2010; Kohli et al. 2010; Savblom et al. 2014). The T allele results in Trp250hK2 (hereafter W-hK2), whereas the common C allele codes for Arg250-hK2 (hereafter R-hK2). Serum hK2 levels are threefold lower in subjects with the TT genotype compared to those with the CC genotype (Nam et al. 2006; Klein et al. 2010). The cutoff value of serum hK2 concentration in prostate cancer screening thus needs to be adjusted according to the genotype of this SNP. The physiological role of PSA and hK2 is in male fertility. After ejaculation, semenogelins in semen form a coagulum to immobilize sperm in it. PSA is secreted from the prostate and mixed into the semen at a high concentration as an inactive precursor. As hK2 activates PSA, PSA gradually cleaves semenogelins, leading to the release of sperm from the coagulum, called semen liquefaction (Robert & Gagnon 1999). A recent report showed the association of seminal plasma concentration of hK2 and rs198977 in young, healthy men (Savblom et al. 2014). When this
DOI: 10.1111/gtc.12242 © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd
Mechanism for the function of a SNP, rs198977
function of hK2 is taken into consideration, the T allele of rs198977 may also be a risk factor of male infertility. The cellular mechanism underlying this important functional SNP has not been elucidated. The T allele of rs198977 has been reported as a risk factor for prostate cancer. This allele was found to be more abundant in the patient group with prostate cancer than in the control group (Nam et al. 2003, 2006; Klein et al. 2010). How the T allele promotes cancer development is another important question. The accumulation of unfolded or misfolded proteins in ER is referred to as ER stress. To handle this stress, cells activate several signal transduction pathways called unfolded protein response (UPR) (Ron & Walter 2007). The activation of the UPR retards cellular protein translation and enhances folding and degradation of unfolded proteins in ER. The role of ER stress in tumorigenesis is controversial (Ma & Hendershot 2004). There are many reports that prolonged UPR activates various apoptotic pathways of cancer cells (Breckenridge et al. 2003). Conversely, cells can become resistant to apoptosis via many signaling pathways, including the activation of nuclear factor kappa B and the degradation of p53 by ER stress (Pahl & Baeuerle 1995; Qu et al. 2004; Li et al. 2011). If any member of the pro-apoptotic signaling pathways in cells with chronic ER stress is disarmed by acquiring a mutation, chronic ER stress could become a key promoter of cancer. In addition, the activation of UPR is essential for many types of tumor cells to survive hypoxic stress and fold key signaling molecules (Davenport et al. 2008). In this study, we report decreased cellular secretion of W-hK2 compared to R-hK2 due to a defect in intracellular transport to the Golgi complex. We also show increased activity of the UPR in cells expressing W-hK2 compared to R-hK2, suggesting a possible mechanism of its enhancement of tumorigenesis.
to normalize the hK2-mycHis level to the loaded protein amount and the transfection efficiency, levels of actin and NPT II were also assayed, respectively. The levels of W-hK2-mycHis in CM were fourfold lower than those of R-hK2-mycHis in PC3 cells and HEK293A cells. In HeLa cells, just a twofold lower level was detected in the CM (Fig. 1A–C). As we obtained the similar results from the experiments with plasmids expressing hK2 without a tag and hK2 with amino-terminal Ypet, a YFP variant, we excluded the misleading effect of the carboxy-terminal tag (Fig. 2). To gain insight into the mechanism of this difference in expression, the levels of intracellular hK2-mycHis and KLK2-mycHis transcript were measured. The intracellular levels of hK2-mycHis were not significantly different according to genotype in all three tested cell lines. No significant difference in KLK2-mycHis transcript level was observed between the two genotypes (Fig. 3). Mechanism of the difference in hK2 level in CM according to genotype
Results
Based on these results, we hypothesized that W-hK2mycHis is more unstable after secretion than R-hK2mycHis and/or that W-hK2-mycHis has a defect in intracellular trafficking. To test the stability hypothesis, we prepared the CM containing Ypet-W-hK2 and transferred it to PC3-R or PC3-W cells, followed by further incubation for 18 h. Ypet-hK2 could be distinguished from hK2-mycHis due to its higher molecular weight. If the stability hypothesis was correct, the Ypet-W-hK2 level would be reduced more readily in PC3-W than PC3-R cells. Figure 4A shows the scheme of this experiment. The immunoblot showed that the amount of YpetW-hK2 did not change on further incubation in PC3-R or PC3-W (Fig. 4B). These results enabled us to reject the hypothesis that W-hK2 is less stable than R-hK2. The alternative hypothesis was that tryptophan at the 250th residue of hK2 impaired intracellular trafficking of hK2.
Expression of hK2 in conditioned media (CM) and cell lysates (CL)
Intracellular localization of hK2
To investigate the expression level of hK2 in CM and CL, we transfected PC3, HEK293A and HeLa cells with pR- or pW-hK2-mycHis. hK2 was tagged with mycHis at the carboxy terminus to differentiate hK2 expressed by transfected DNA and endogenous hK2. CM and CL samples were prepared and used in immunoblotting with an anti-His antibody. In order
To examine whether W-hK2 could pass through the ER and be transported to the Golgi complex, we transfected HEK293A cells with pmTurquoise2-Golgi combined with pYpet-R- or pYpet-W-hK2. Cells were transferred to glass-bottom confocal dishes and examined with laser scanning confocal microscope. When Ypet-R-hK2 was expressed, most mTurquoise2
© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd
Genes to Cells (2015) 20, 512–520
513
E J Choi et al. Table 1 Genetic and functional differences of R-hK2 and W-hK2 Allele of rs198977 250th amino acid of hK2
C
T
Arginine (R-hK2)
Tryptophan (W-hK2)
Allele frequency Intracellular production of hK2 Cellular disposal of hK2 Intracellular degradation Secretion Expression level of hK2 Intracellular In conditioned media Trafficking to Golgi complex Prostate cancer risk Unfolded protein response
>* >† > > > > † > > > >
