Enhanced Activity of Variant Phospholipase C-δ1 Protein (R257H) Detected in Patients With Coronary Artery Spasm
Background— We recently demonstrated that phospholipase C (PLC)–δ1 activity in cultured skin fibroblasts obtained from patients with coronary spastic angina (CSA) is enhanced. We tested the hypothesis that structural abnormality in PLC-δ1 isoform is a cause of the enhanced activity.
Methods and Results— Sequence analysis of the cDNA coding for PLC-δ1 obtained from fibroblasts revealed that one conversion of guanine to adenine (A) was present at nucleotide position 864 in one CSA patient, resulting in the amino acid replacement of arginine 257 by histidine (R257H). The incidence of 864A/A in genomic DNA, analyzed by single-strand conformation polymorphism, was greater in patients with CSA than in male control subjects (6 of 57 patients with CSA versus 1 of 62 control subjects, P<0.05). The activity of the variant PLC-δ1 protein under free calcium concentration between 10−8 and 10−7 mol/L was 2-fold higher than that of the wild-type protein. Baseline intracellular calcium concentration ([Ca2+]i) in human embryonic kidney 293 cells transfected with the variant PLC-δ1 was higher than that in cells with the wild type. The peak increase in [Ca2+]i in response to acetylcholine at 10−6 and 10−5 mol/L was greater in the cells with the variant PLC-δ1 than in those with the wild type.
Conclusions— These findings indicate that the R257H variant in the PLC-δ1 gene detected in patients with CSA is associated with enhancement of enzyme activity, and they describe a novel mechanism for the enhanced coronary vasomotility in CSA.
Received December 7, 2001; revision received February 15, 2002; accepted February 15, 2002.
Coronary artery spasm plays important roles in the pathogenesis of variant angina1,2⇓ and the other acute coronary syndromes.3,4⇓ We and other investigators have shown that the basal vasomotor tone of the entire coronary artery system of Japanese patients with variant angina is enhanced.5–7⇓⇓ In addition, the coronary artery constrictor response to diverse constrictor stimuli is enhanced, 8 and occlusive constriction is readily induced when exposed to such a stimulus. On the other hand, the incidences of both migraine and Raynaud’s phenomenon in patients with variant angina are high as compared with those in control subjects.9 Also, enhanced esophageal motility in patients with variant angina has been reported.10 These results strongly suggest the presence of a generalized disorder of smooth muscle contraction in coronary spastic angina (CSA).
Ito et al,11 Katsumata et al,12 and Kandabashi et al13 have investigated the intracellular mechanism for spasm using swine coronary spasm models. They indicate that the protein kinase C–mediated pathway and enhanced myosin light chain phosphorylation play an important role in the enhanced constrictor response of the coronary artery smooth muscle to autocoids such as serotonin and histamine. Recently, we focused on phospholipase C (PLC) as a pivotal enzyme in the pathogenesis of coronary artery spasm and demonstrated that the enzyme activity in the cultured skin fibroblast obtained from the patients with CSA was enhanced. A major PLC isozyme detected in the membrane fraction of the skin fibroblast was the δ1 isoform, but there was no difference in the amount of δ1 isoform protein between the control subjects and the patients with CSA.14 Because PLC-δ isozymes are more sensitive to Ca2+ than the other isozymes, the initial transient increase in intracellular free Ca2+ concentration ([Ca2+]i) induced by receptor/G-protein–mediated PLC activation may induce the prolonged activation of PLC-δ1 in a positive-feedback fashion and result in hypercontraction of the smooth muscle cells, especially in CSA patients showing enhanced PLC-δ1 activity. In the present study, we tested the hypothesis that enhanced PLC activity in patients with CSA is associated with abnormalities in the primary structure and function of PLC-δ1.
The ethics committee of our institution approved the study protocol. Written informed consent was obtained from all patients before the study. This study population included 85 patients with CSA (57 men and 28 women with a mean age of 60±10 years) and 142 control subjects without hypertension or any history suggestive of angina pectoris (62 men and 80 women with a mean age of 52±8 years). Three of the 142 control subjects were admitted to our hospital for treatment of supraventricular tachycardias, and the other 139 control subjects were healthy volunteers. Coronary spasm, defined as total or subtotal occlusion or severe vasoconstriction of the coronary artery associated with chest pain and ischemic ECG change, was induced with intracoronary injection of acetylcholine (ACh) in all CSA patients. After intracoronary injection of isosorbide dinitrate, the coronary arteriograms revealed normal or almost normal coronary arteries with diameter stenosis <50% of the lumen diameter.
Human skin fibroblasts were prepared by explant method as described previously.14 Human skin fibroblasts and embryonic kidney (HEK) 293 cells were cultured in DMEM supplemented with 10% FCS at 37°C under 5% CO2 and 95% air.
Isolation of mRNA and Sequencing
Total RNA was extracted from the skin fibroblasts, and complementary DNA (cDNA) was synthesized by Superscript II (Life Technologies) using oligo(dT) as a primer. The PLC-δ1 cDNA was amplified by polymerase chain reaction (PCR) using 3 sets of primers (sense 5′-TGCTGTCGCTACTCAAGTGAGTCC-3′, antisense 5′-AGGAAGCCGTCCTTGGTCATCTGC-3′, nucleotide position 37 to 921; sense 5′-TACGAGCCCAGCGAGACTACCAAG-3′, antisense 5′-ACGCCATCTCGTAGAAGGCCTGTC-3′, position 866 to 1658; and sense 5′-GCTCAGGCTAGCACAGGAGCTCTC-3′, antisense 5′-ATGTGGACAGAGGGCCCAGCCCACT-3′, position 1549 to 2419). The PCR products were subcloned into pGEM-T Easy Vector (Promega) and then sequenced by the dideoxy-mediated chain-termination method.
Isolation and Analysis of Genomic DNA
Genomic DNA was isolated from the whole blood using the QIAamp DNA Blood Kit (QIAGEN). Exon 5 of the PLC-δ1 gene was amplified by PCR using a set of primers (sense 5′-CTGTCGGTGGATCAGTTAGTGACG-3′, antisense 5′-AGGCTCTCACCAGTCTCGCTG-3′). Direct sequencing and single-strand conformation polymorphism (SSCP) analysis were performed as reported previously.15
Mutagenesis of Human PLC-δ1
Human PLC-δ1 cDNA (pRSETAplc) was a generous gift from Dr Hitoshi Yagisawa (Himeji Institute of Technology, Hyogo, Japan), with permission from Dr K. King.16 To construct the variant human PLC-δ1, in which the arginine residue at position 257 was replaced with histidine (R257H), the cDNA fragment was amplified from the fibroblasts of the patient with CSA by PCR with a set of primers (sense 5′-TGGAGATCGACCGCACCTTCGCCG-3′, antisense 5′-CATGTCCTGGTAGACACGGCGGTG-3′). The PCR product was subcloned into pGEM-T Easy Vector and inserted into the corresponding sites of pRSETAplc (pRSETAplcV). The variant pRSETAplc sequence was confirmed by sequencing.
Protein Expression and Purification
To express the wild-type and variant PLC-δ1 as glutathione S-transferase (GST) fusion proteins, pRSETAplc and pRSETAplcV were transfected into Escherichia coli BL21 (DE3) pLysS. The recombinant proteins were purified with glutathione Sepharose (Amersham Pharmacia Biotech) and were used to measure the PLC activity in vitro. The molecular mass and purity of the recombinant PLC-δ1 proteins were confirmed by SDS-PAGE.
Assay for PLC Activity
Phosphatidylinositol 4,5-bisphosphate (PIP2) hydrolysis was determined in a reaction mixture (50 μL) consisting of 50 mmol/L 2-(N-Morpholine) ethanesulfonic acid-NaOH (pH 6.5), 50 μmol/L PIP2 (25 000 dpm [3H]PIP2), 3 mmol/L EGTA, and 1 pmol wild-type or variant recombinant PLC-δ1. CaCl2 to the assay mixture to give the indicated [Ca2+]. Assays were performed for 10 minutes at 37°C, and inositol trisphosphate (IP3) was measured.
Measurement of [Ca2+]i
HEK293 cells were transfected with muscarine M1 receptor cDNA (a generous gift from Dr Tomohiro Kurosaki at Kansai Medical University, Osaka, Japan17) and the constructed plasmid DNA of either the wild-type or variant PLC-δ1, or the empty vector. The expression of the wild-type and variant PLC-δ1 proteins in HEK293 cells was examined by Western blot analysis using anti–PLC-δ1 monoclonal antibody. After loading with 5 μmol/L fura-2–acetoxymethyl ester, [Ca2+]i in response to ACh at 10−6 and 10−5 mol/L was measured at excitation wavelengths of 340 and 380 nm and emission wavelength of 510 nm, as described previously.18
All data are mean±SEM. Categorical variables were compared by χ2 analysis. An unpaired ttest for comparison of 2 variables, ANOVA for repeated measures, and 2-way ANOVA for multiple comparisons were used for statistical analysis. Differences were considered significant when P values were <0.05.
Sequencing of PLC-δ1 cDNA From Fibroblasts
As shown in Figure 1A, we analyzed the nucleotide position 37 to 2419 of the PLC-δ1 cDNA obtained from fibroblasts of 3 patients with CSA and 3 control subjects, in which its coding region is the nucleotide position between 95 and 2365. Then, we compared the sequence of the PLC-δ1 cDNA obtained from fibroblasts with that of the PLC-δ1 cDNA previously reported (GenBank accession No. U09117).16 According to the sequence of 3 parts of the PCR products, 8 different nucleotide sites from the PLC-δ1 cDNA reported were found at nucleotide positions 89 (A to G), 214 (G to C), 764 (C to T), 864 (G to A), 884 (A to G), 1744 (A to C), 1866 (A to G), and 2376 (C to G). Nucleotide replacement at the 7 sites other than position 864 was found in all 6 subjects and pRSETAplc used in this study and was identical to the sequence of genomic DNA reported,19 indicating that the 7 sites of PLC-δ1 cDNA were reported incorrectly. As shown in the left panel of Figure 1B, 864G, which is identical to the sequence of cDNA reported, was found in 3 control subjects and 2 patients with CSA. However, the 864G-to-A conversion (864G-A) was found in one patient with CSA (right panel). This 864G-A variant results in the amino acid replacement of arginine 257 by histidine in human PLC-δ1.
Detection of 864G-A Variant in Genomic DNA
To confirm the 864G-A variant of genomic DNA, we amplified genomic DNA corresponding to exon 5 of the human PLC-δ1 gene obtained from 85 patients with CSA and 142 control subjects by PCR and examined the presence of this variant by the SSCP method. As shown in Figure 2A, the PCR products were demonstrated as 3 patterns of bands, and each corresponded to the 864G/G normal homozygote, 864G/A heterozygote, and 864A/A homozygote of the human PLC-δ1 gene. This G-to-A conversion detected in cDNA from the fibroblasts was confirmed by direct sequencing of the PCR products of genomic DNA in all patients (Figure 2B). As shown in the Table, the 864A/A homozygote, heterozygote, and 864G/G normal homozygote were present in 8 (9.4%), 25 (29.4%), and 52 (61.2%) cases in the 85 patients with CSA, respectively. In the 142 control subjects, the 864A/A homozygote, heterozygote, and 864G/G normal homozygote were present in 6 (4.2%), 55 (38.7%), and 81 (57.1%) cases, respectively. The incidence of the 864A/A homozygote variant was significantly greater in male patients with CSA than in male control subjects (6 of 57 patients with CSA versus 1 of 62 control subjects, P<0.05). However, it was similar between female patients with CSA and female controls.
PIP2-PLC Activities of Wild-Type and Variant GST–PLC-δ1 Fusion Proteins
To examine the activities of PLC-δ1 protein, we constructed wild-type and variant GST–PLC-δ1 fusion proteins and measured the [Ca2+]-dependent PIP2 hydrolysis activity. SDS-PAGE showed that there was a single compound at 86 kDa in lanes of wild-type and variant GST–PLC-δ1 fusion proteins, respectively (Figure 3A), and the molecular mass of these peptides was completely identical to that of fusion proteins. As shown in Figure 3B, the activities of both wild-type and variant PLC-δ1 were increased in a [Ca2+]-dependent manner (both P<0.05 by ANOVA for repeated measures), and the [Ca2+] dependency was significantly greater in variant PLC-δ1 than in wild-type PLC-δ1 (P<0.05 by 2-way ANOVA). The activity of the variant PLC-δ1 was significantly higher by 80±15% at 10−8 mol/L [Ca2+], by 77±9% at 10−7.5 mol/L [Ca2+], and by 42±3% at 10−7 mol/L [Ca2+] than that of the wild type (n=4, all P<0.05), despite no difference in the enzyme activity at baseline. At 10−6.5 and 10−6 mol/L [Ca2+], the activity of the variant PLC-δ1 became similar to that of the wild type.
Effect of Variant PLC-δ1 on the Response of [Ca2+]i to ACh
Western blot analysis showed that there was a single immunoreactive compound at 86 kDa that was identical to the wild-type and variant PLC-δ1 (Figure 4A), and the amount of PLC-δ1 protein expressed in HEK293 cells was similar between the wild types and variant. HEK293 cells per se showed no [Ca2+]i response to ACh, whereas those expressing muscarine M1 receptor showed a constant [Ca2+]i response to ACh. Figure 4B illustrates the representative waveforms of [Ca2+]i in response to ACh. ACh at 10−6 and 10−5 mol/L both caused a rapid transient increase in [Ca2+]i followed by a lower but sustained phase of the increase. It is noted that in HEK293 cells transfected with the variant PLC-δ1, the transient increase in [Ca2+]i was augmented and the sustained phase was prolonged. As shown in Figure 4C, [Ca2+]i at baseline was 17±4 nmol/L in the wild type and 33±7 nmol/L in the variant PLC-δ1 (P<0.05). The peak increase in [Ca2+]i from the baseline after ACh at 10−5 mol/L was higher than that after ACh at 10−6 mol/L in both types of cells (both P<0.05) (Figure 4D). The peak increase in [Ca2+]i from the baseline after ACh was significantly greater in the variant PLC-δ1 than in the wild type (42±4 versus 28±3 nmol/L at 10−6 mol/L ACh and 74±8 versus 45±8 nmol/L at 10−5 mol/L ACh, both P<0.05).
On the basis of previous clinical findings,5–7,9,10⇓⇓⇓⇓ it was suggested that a generalized disorder of smooth muscle contraction due to an abnormal, impaired intracellular signaling is present in patients with CSA. We recently demonstrated that PLC activity in cultured skin fibroblasts obtained from patients with CSA was enhanced, and a major PLC isozyme detected in the membrane fraction was the δ1 isoform. In the present study, we found a single base variant (864G-A) in the entire coding region of the PLC-δ1 gene in human CSA. This 864G-A variant resulted in the amino acid replacement of arginine 257 by histidine. It is important to note that this 864G-A variant of PLC-δ1 is associated with significantly enhanced enzyme activity. To the best of our knowledge, the present variant of PLC-δ1 gene seems to be the only variant related to the enhancement of the activity in humans. Shimohama et al20 previously reported missense mutation in the PH domain of the human PLC-δ1 gene, but it was associated with a remarkable loss of function. In the spontaneously hypertensive rat, a variation in PLC-δ1 sequence was reported, but its involvement in the function remains unclear.21 The site of the 864G-A variant corresponds to the fourth lobe of the EF hand domain of PLC-δ1. Nakashima et al22 reported that the EF hand domain may not play a role in the regulation of Ca2+ sensitivity of PLC-δ1. Crystallographic analysis also showed that the EF hand domain does not bind Ca2+ but rather serves as a flexible link between the PH domain and the rest of the enzyme.23,24⇓ However, it was recently demonstrated that the EF hand domain of PLC-δ1 binds Ca2+ and it is necessary for the efficient interaction of the PH domain with PIP2.25 Thus, the 864G-A variant of PLC-δ1 found in this study seems to contribute to the altered enzyme activity induced by Ca2+.
PLC-δ isozyme is more sensitive to Ca2+ than the other isozymes. An increase in [Ca2+]i within the physiological range (10 to 1000 nmol/L) is sufficient to stimulate PLC-δ1.26 Therefore, the initial transient increase in [Ca2+]i induced by IP3 may in turn contribute to the prolonged activation of PLC-δ1 in a positive-feedback fashion.25 Furthermore, [Ca2+]i at baseline also may be regulated by PLC-δ1, given that [Ca2+]i at baseline per se ranges within the levels that activate PLC-δ1. We tested our hypothesis by using HEK293 cells transfected with muscarine M1 receptor and ACh as an agonist. We used ACh because it is widely used in provocation tests for coronary spasm27 and it binds muscarine M1 receptor linked to the Gq–PLC-β pathway. The results showed that the peak increase in [Ca2+]i in response to ACh was enhanced and the sustained phase that followed was prolonged in the cells transfected with the variant PLC-δ1. In addition, [Ca2+]i at baseline was elevated in the cells with the variant PLC-δ1. These results seem to be consistent with the pathogenesis of CSA in that the basal vascular tone and the vasoconstrictor response to stimuli were both enhanced. We previously demonstrated that PLC activity was positively correlated not only with basal coronary artery tone but with the maximal and averaged constrictor responses of the coronary artery to ACh, and the higher PLC activity was, the greater the basal tone and constrictor response to ACh were. This suggests the role of this variant PLC-δ1 protein in the genesis of coronary spasm.
Vascular endothelium has been shown to play important roles in the regulation of vascular tone. Recently, Nakayama et al28 reported that the endothelial NO synthase allele/T784C, A922G, and T1468A homozygote mutation was present in 3 (2%) of 174 patients with CSA and in none of 161 control subjects. This mutation resulted in the reduction of NO synthesis and may predispose the patients with this mutation to coronary spasm.28 NO deficiency may contribute to the enhanced constrictor response of the coronary artery. The variant of PLC-δ1 detected in the present study could be another mechanism for the increased coronary vasomotility through the enhanced contractility of the vascular smooth muscle cells of the coronary arteries, especially in males. It should be emphasized, however, that this PLC-δ1 variant was present in only ≈10% of the CSA patients. Other factors or a combination of factors may contribute to or play a more important role in the pathogenesis of coronary spasm. It also should be pointed out that the incidence of the present gene variant in female CSA patients (n=28) was not different from that in female control subjects (n=80). The relatively small number of female CSA patients may be related to the similarity of the incidence. CSA was known to occur much more frequently in males than in females. Thus, in females, an additional mechanism or factor may be required for the manifestation of the enhanced activity of the present variant PLC-δ1. Further studies on the mechanism for the difference in incidence by sex are required.
This study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture (C03268107), and Grant 10C-5 from the Ministry of Health and Welfare (Tokyo, Japan).
- ↵Oliva PB, Potts DE, Pluss RG. Coronary artery spasm in Prinzmetal angina: documentation by coronary angiography. N Engl J Med. 1973; 288: 745–751.
- ↵Yasue H, Touyama M, Shimamoto M, et al. Role of autonomic nervous system in the pathogenesis of Prinzmetal’s variant form of angina. Circulation. 1974; 50: 534–539.
- ↵Oliva PB, Breckinridge JC. Arteriographic evidence of coronary artery spasm in acute myocardial infarction. Circulation. 1977; 56: 366–374.
- ↵Maseri A, L’Abbate A, Baroldi G, et al. Coronary vasospasm as a possible cause of myocardial infarction: a conclusion derived from the study of preinfarction angina. N Engl J Med. 1978; 299: 1271–1277.
- ↵Okumura K, Yasue H, Matsuyama K, et al. Diffuse disorder of coronary artery vasomotility in patients with coronary spastic angina. Hyperreactivity to the constrictor effects of acetylcholine and the dilator effects of nitroglycerin. J Am Coll Cardiol. 1996; 27: 45–52.
- ↵Hoshio A, Kotake H, Mashiba H. Significance of coronary artery tone in patients with vasospastic angina. J Am Coll Cardiol. 1989; 14: 604–609.
- ↵Kuga T, Egashira K, Inou T, et al. Correlation of basal coronary artery tone with constrictive response to ergonovine in patients with variant angina. J Am Coll Cardiol. 1993; 22: 144–150.
- ↵Kaski JC, Crea F, Meran D, et al. Local coronary supersensitivity to diverse vasoconstrictive stimuli in patients with variant angina. Circulation. 1986; 74: 1255–1265.
- ↵Miller D, Waters DD, Warnica W, et al. Is variant angina the coronary manifestation of a generalized vasospastic disorder? N Engl J Med. 1981; 304: 763–766.
- ↵Rasmussen K, Ravnsbaek J, Funch-Jensen P, et al. Oesophageal spasm in patients with coronary artery spasm. Lancet. 1986; 1: 174–176.
- ↵Ito A, Shimokawa H, Nakaide R, et al. Role of protein kinase C-mediated pathway in the pathogenesis of coronary artery spasm in a swine model. Circulation. 1994; 90: 2425–2431.
- ↵Katsumata N, Shimokawa H, Seto M, et al. Enhanced myosin light chain phosphorylations as a central mechanism for coronary artery spasm in a swine model with interleukin-1β. Circulation. 1997; 96: 4357–4363.
- ↵Kandabashi T, Shimokawa H, Miyata K, et al. Inhibition of myosin phosphatase by upregulated rho-kinase plays a key role for coronary artery spasm in a porcine model with interleukin-1β. Circulation. 2000; 101: 1319–1323.
- ↵Okumura K, Osanai T, Kosugi T, et al. Enhanced phospholipase C activity in the cultured skin fibroblast obtained from patients with coronary spastic angina: possible role for enhanced vasoconstrictor response. J Am Coll Cardiol. 2000; 36: 1847–1852.
- ↵Orita M, Iwahara H, Kanazawa H, et al. Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc Natl Acad Sci U S A. 1989; 86: 2766–2770.
- ↵Cheng HF, Jiang MJ, Chen CL, et al. Cloning and identification of amino acid residues of human phospholipase Cδ1 essential for catalysis. J Biol Chem. 1995; 270: 5495–5505.
- ↵Takata M, Sabe H, Hata A, et al. Tyrosine kinases Lyn and Syk regulate B cell receptor-coupled Ca2+ mobilization through distinct pathways. EMBO J. 1994; 13: 1341–1349.
- ↵Osanai T, Dunn MJ. Phospholipase C responses in cells from spontaneously hypertensive rats. Hypertension. 1992; 19: 446–455.
- ↵Ishikawa S, Takahashi T, Ogawa M, et al. Genomic structure of the human PLCD1 (phospholipase C δ1) locus on 3p22–>p21.3. Cytogenet Cell Genet. 1997; 78: 58–60.
- ↵Shimohama S, Kamiya S, Fujii M, et al. Mutation in the pleckstrin homology domain of the human phospholipase C-δ1 gene is associated with loss of function. Biochem Biophys Res Commun. 1998; 245: 722–728.
- ↵Yagisawa H, Tanase H, Nojima H. Phospholipase C-δ gene of the spontaneously hypertensive rat harbors point mutations causing amino acid substitutions in a catalytic domain. J Hypertens. 1991; 9: 997–1004.
- ↵Nakashima S, Banno Y, Watanabe T, et al. Deletion and site-directed mutagenesis of EF-hand domain of phospholipase C-δ1: effects on its activity. Biochem Biophys Res Commun. 1995; 211: 365–369.
- ↵Essen LO, Perisic O, Cheung R, et al. Crystal structure of a mammalian phosphoinositide-specific phospholipase Cδ. Nature. 1996; 380: 595–602.
- ↵Williams RL, Katan M. Structural views of phosphoinositide-specific phospholipase C: signaling the way ahead. Structure. 1996; 4: 1387–1394.
- ↵Yamamoto T, Takeuchi H, Kanematsu T, et al. Involvement of EF hand motifs in the Ca2+-dependent binding of the pleckstrin homology domain to phosphoinositides. Eur J Biochem. 1999; 265: 481–490.
- ↵Kim YH, Park TJ, Lee YH, et al. Phospholipase C-δ1 is activated by capacitative calcium entry that follows phospholipase C-β activation upon bradykinin stimulation. J Biol Chem. 1999; 274: 26127–26134.
- ↵Okumura K, Yasue H, Matsuyama K, et al. Sensitivity and specificity of intracoronary injection of acetylcholine for the induction of coronary artery spasm. J Am Coll Cardiol. 1988; 12: 883–888.
- ↵Nakayama M, Yasue H, Yoshimura M, et al. T−786-C mutation in the 5′-flanking region of the endothelial nitric oxide synthase gene is associated with coronary spasm. Circulation. 1999; 99: 2864–2870.