Plasma skeletal muscle myosin phenotypes identified by immunoblotting are associated with pulmonary embolism occurrence in young adults.

Abstract

Background:

Purified skeletal muscle myosin (SkM) binds factor Xa and is procoagulant. The molecular forms of SkM in human plasma have not been characterized.

Method:

Human plasma SkM heavy chain (HC) isoforms of different molecular weights were detected by a newly developed immunoblotting protocol. In this pilot study, the distribution of SkM HC antigen isoforms in plasmas of healthy subjects and young adult patients with venous thrombosis was analyzed.

Results:

Multiple SkM HC antigen bands were detected in human plasmas, corresponding to full-length SkM HC, heavy meromyosin, or the S1 fragment. Plasma immunoblots of healthy subjects displayed three major phenotypes: Type I with two primary bands for full-length SkM and heavy meromyosin, and two lesser bands including S1 fragment (54%); Type II with bands primarily for full-length SkM HC (34%); and Type III with only a band for the S1 fragment (12%). Plasma SkM HC antigen Type II phenotype was associated with an increased occurrence of isolated pulmonary embolism in younger patients, respectively (≤50 years old).

Conclusions:

Three SkM HC antigen phenotypes were identified in human plasma by immunoblotting, and Type II phenotype was correlated with the occurrence of isolated pulmonary embolisms in younger patients.

1. Introduction

Skeletal muscle myosin (SkM) was recently reported to be prothrombotic ex vivo in flowing human blood and to promote thrombin generation in whole blood, platelet rich plasma, and platelet poor plasma by binding clotting factors Xa and Va, thereby activating prothrombin and yielding thrombin [1–3]. Further, another study reported SkM supports both the procoagulant anticoagulant actions of activated protein C, thus indicating that SkM may contribute to negative feedback downregulation of thrombin generation [4]. SkM was also reported to accelerate fibrinolysis as a cofactor of tissue type plasminogen activator [5]. As an in vivo proof of concept experiment, SkM was shown to promote hemostasis (i.e., reduce bleeding) in a murine-acquired hemophilia A tail bleeding model wherein mice received anti-factor VIII antibodies [6]. This suggests that SkM may express either prohemostatic or prothrombotic activities in vivo, depending on the context.

Myosins are a large family of motor proteins sharing common features of ATP hydrolysis, actin binding, and the potential for kinetic energy transduction [7]. The conventional myosins consist of a dimer of heterotrimers [7], and three isoforms of SkM have been extensively studied (i.e., intact SkM at 520 kDa, heavy meromyosin (HMM) at 350 kDa, and the S1 fragment at 130 kDa). Following electrophoresis in an SDS gel, dissociated SkM heavy chains appear as bands at 240–260 kDa, 160 kDa, and 95 kDa, corresponding to full length heavy chain, HMM, and the S1 fragment, respectively [8–12].

SkM was originally isolated from muscle cells but is broadly found in the body, even in plasma. Plasma SkM heavy chain levels were reported to be elevated in patients with muscle damage (e.g., rhabdomyolysis) [13–16]. Plasma levels of SkM are frequently elevated in polymyositis and dermatomyositis patients [13], and polymyositis or dermatomyositis are associated with increased venous thrombosis risk [17]. However, since almost all previous studies concerning SkM focused primarily on muscle and not plasma SkM, there currently is no information about what forms of SkM circulate in the blood.

Here, we first detected SkM isoforms in human plasma by immunoblotting and discovered three distinct phenotypes of SkM isoforms in plasma from different individuals. We then investigated the association of these phenotypes with venous thromboembolism (VTE) and pulmonary embolism (PE) in this pilot study.

2. Material and Methods

2.1. Materials

Human factor Va and factor Xa were purchased from Hematologic Technologies Inc. (Essex Junction, VT). Prothrombin and chromogenic substrate Pefachrome® TH were from Enzyme Research Laboratories (South Bend, IN). Rabbit SkM was purchased from Cytoskeleton, Inc. (Denver, CO). Beads containing immobilized bovine chymotrypsin were from Princeton Separations, Inc. (Adelphia, NJ). Monoclonal antibodies (28D4, 22D4, 28G12, 31A3, 21H11, 33D8, 35C11,46B12, 35B6, 37E3) were produced against SkM peptides (amino acid residuals 815–830) in SJL mice following standard antibody production protocol at Scripps Antibody Core Facility. Hybridomas of monoclonal antibodies against SkM heavy chain (MF30, MF20, S22, F59, BF35) were purchased from Developmental Studies Hybridoma Bank at University of Iowa (Iowa City, IA) and then used to obtain antibodies that were purified using a protein G column. Goat polyclonal antibodies against SkM heavy chain (EBU016) were from Kerafast, Inc. (Boston, MA). Rabbit polyclonal antibodies against SkM heavy chain (ab91506) and polyclonal serum against SkM heavy and light chains (PA1–28037) were obtained from Abcam (Cambridge, MA) and Fisher Scientific (Pittsburgh, PA), respectively. IRDdye 680RD donkey anti-mouse IgG, IRDdye 800CW donkey anti-rabbit IgG, and IRDdye 680RD donkey anti-goat IgG were purchased from LiCor Biosciences (Lincoln, NE). Fatty acid-free and protease-free bovine serum albumin (BSA) was purchased from Calbiochem (San Diego, CA)

2.2. SkM cleavage by chymotrypsin

SkM (20 uL, 10 mg/mL) in 25 mmol/L PIPES-NaOH pH 7.0, 1.25 mol/L KCl, 2.5% sucrose, and 0.5% dextran solution was digested by 1 μL of chymotrypsin immobilized on beads for 5 or 10 minutes. Chymotrypsin-beads were removed by centrifugation (1 minute, 8000 g) and the supernatant was immediately added to 4× SDS loading buffer. The result of the digestion was analyzed by SDS-PAGE using BioRad TGX 420% Gels (Bio-Rad). The protein bands were stained using InstantBlue™ (Expedeon Inc, San Diego, CA).

2.3. The Scripps Venous Thrombosis Registry

Plasma samples (N=105 healthy donors and N=105 VTE patients including 99 Caucasian and 6 non-Caucasian in each cohort) were obtained from the Scripps Venous Thrombosis Registry which is a case-control study of risk factors for VTE [18, 19]. Inclusion criteria for this study included age at thrombosis < 55 years, > 3 months since diagnosis of acute thrombosis, and a life expectancy of at least three years. Age matched (± 2 years) healthy controls were recruited through the General Clinical Research Center’s (GCRC) blood donation program. For all patients, the diagnosis of VTE and pulmonary embolism (PE) was confirmed by objective methods (phlebography, compression duplex ultrasonography, pulmonary computerized tomography, or perfusion ventilation lung scan). Blood was collected in the GCRC at least three months after VTE diagnosis and after 12 hours of fasting. EDTA-plasma was prepared and stored at −70ºC. Participants in the blood donation program had normal CBC. The protocol was approved by the Institutional Review Board of Scripps Clinic and subjects provided written informed consent. Subjects with non-Caucasian ancestry (N=6 healthy donors and N=6 VTE patients) were excluded from the analyses due to potential differences among races.

2.4. Immunoblot assay for SkM isomers in plasma

Anti-SkM heavy chain antibodies were screened for utility in immunoblotting to identify SkM heavy chain isoforms in plasma. BioRad TGX 4–20% Gels were used for SDS-PAGE of plasma samples (2 μL of neat or of 5 times-diluted plasma samples). Then, the proteins were transferred to FL-PVDF membranes using a semi-dry system (Trans-Blot Turbo Transfer System, BioRad). The membrane was then blocked using LiCor blocking buffer in phosphate buffered saline, and anti-SkM heavy chain monoclonal or polyclonal antibodies (1 to 10 mg/L) were incubated with the membrane for 1 hour at room temperature (primary antibody). After washing three times with 50 mmol/L tris buffered saline at pH 7.4 containing 0.1% Tween 20, IRDye-conjugated donkey anti-mouse, anti-rabbit, or anti-goat antibodies were incubated with the membrane for 1 hour at room temperature, followed by three washes with 50 mmol/L Tris buffered saline at pH 7.4 containing 0.1% Tween 20. The signal of bound IRDye-conjugated secondary antibodies to the primary anti-myosin antibodies on the membrane was detected by the LiCor imaging system.

2.5. Statistical analysis

Chi-square tests and nonparametrical Friedman tests were done using the Prism™ 7.01 software (Graph Pad Software Inc., San Diego, CA).

3. Results

3.1. Rabbit SkM cleavage by chymotrypsin

Rabbit SkM was digested by chymotrypsin. The intact SkM heavy chain band was seen as a 240 kDa band in SDS-PAGE at time 0 (no digestion, Figure 1A). After 5 minutes of digestion, 150 kDa, 95 kDa and other lower molecular weight (MW) bands appeared without the 240 kDa band (Figure 1A). The chymotrypsin digestion pattern, as described previously [8, 20], indicated that 150 kDa and 95 kDa bands were the heavy chain of HMM and the S1 fragment, respectively.

Figure 1. SDS-PAGE profiles of SkM antigen bands following chymotrypsin cleavage and immunoblots of the SkM antigen in human plasma.

(A) SDS-polyacrylamide (PAGE) profiles of chymotryptic digests of rabbit skeletal muscle myosin in the presence of EDTA are shown. The protein bands were stained by Coomassie blue. Times of digestion, 0 and 5 minutes, are indicated. A one μg SkM sample was loaded for 0 minutes (no digestion) and 10 μg SkM samples were loaded for the 5-minute digestion. MWs were estimated with the Bio-Rad Precision Plus Protein Standard.

(B) Immunoblot profiles for the SkM heavy chain antigen bands in two individuals’ plasma (a and b) (2 μl samples) are shown. The SkM heavy chain antigen bands were detected using the F59 monoclonal antibody. MWs estimated with the Bio-Rad Precision Plus Protein Standard.

(C) Immunoblot profiles for 6 different subjects’ plasma samples including two for each of the three phenotypes (I, II, and III) of the SkM heavy chain (donors 69–74 also seen in Figure 1S) are shown, with the phenotype labels I, II, or III above the corresponding lanes. MWs were estimated with the Bio-Rad Precision Plus Protein Standard. SkM plasma phenotypes were defined based on the presence or absence of different MW SkM isoforms on SDS gel immunoblots as follows: Type I - two primary bands observed at 240 kDa and160 kDa, and two lesser bands at 300 kDa and 95 kDa; Type II – bands primarily at ~300 kDa and 260 kDa with negligible to no bands at 160 kDa and 95 kDa; and Type III – only one major band at 95 kDa with a very small to negligible amount of any other band.

3.2. Screening of antibodies for immunoblotting for SkM in plasma

Fifteen monoclonal antibodies against the SkM heavy chain (MF30, MF20, S22, F59, BF35, 28D4, 22D4, 28G12, 31A3, 21H11, 33D8, 35C11,46B12, 35B6 and 37E3), one rabbit polyclonal antibody (ab91506), one goat polyclonal antibody (EB16), and one polyclonal serum against SkM heavy and light chains (PA1–28037) were screened to detect SkM in plasma by immunoblotting. One anti-heavy chain monoclonal antibody (F59), which recognizes a region in the S1 domain [21], clearly detected SkM bands in plasma under non-reduced conditions (Figure 1B and Supplemental Figure S1). The SkM bands had MWs of 240–260 kDa,160 kDa, and 95 kDa, along with higher MWs of approximately 300 kDa and above. An anti-heavy chain polyclonal serum (Pierce) also detected bands of 240–260 kDa and above 300 kDa, but the signal was much stronger for the latter. Under reduced conditions, two other anti-heavy chain monoclonal antibodies (28D4 and 22D4), which recognize a sequence in the S1 fragment, detected the same three major bands in pooled human plasma (data not shown).

3.3. SkM isoforms in human plasma and SkM plasma phenotypes

When analyzing individual donors (N=105 healthy donors and N=105 VTE patients) for plasma SkM isoforms using immunoblots with the F59 monoclonal antibody, three phenotypes (designated Types I, II, and III) of SkM band patterns were defined in plasma based on the presence or absence of different MW SkM isoforms on SDS gel immunoblots (Figure 1S). Figure 1C shows SkM band patterns for two donors of each phenotype in which Type I contains two primary bands observed at 240 kDa and 160 kDa, and two lesser bands at 300 kDa and 95 kDa. Type II contains bands primarily at ~300 kDa and 260 kDa with negligible to no bands at 160K and 95 kDa. Type III contains only one major band at 95KDa with a very small to negligible amount of any other band. Plasma SkM phenotype assignment for each of 10 healthy individuals whose plasma was analyzed twice remained the same upon repeated analyses (N=10 healthy control donors) (Figure S1, donors # 12, 23, 25, 27, 31, 33, 36, 38, 48, and 50). The plasma SkM phenotype for each of 18 individuals remained constant after two independent blood drawings (N=18 subjects, data not shown). Phenotypes I and II were also stable and did not change even after 24 hours incubation of plasma at room temperature, showing that there was no conversion in vitro between phenotypes (e.g., Type I to III) (data not shown). These results indicate that the SkM plasma phenotypes are not a result of experimental artifacts.

3.4. Distribution of SkM plasma phenotypes for normal subjects, VTE patients, and pulmonary embolism (PE) patients

The distribution of Types I, II, and III among Caucasian controls (N=99 healthy donors) was 53.5%, 34.3%, and 12.1%, respectively (Table 1A). When plasma SkM phenotypes in Caucasian VTE patients (≤ 55 years old) were analyzed (N=99 healthy donors and N=99 VTE patients), there was no association of plasma SkM phenotypes with the occurrence of VTE (Table 1A). There were no differences in the distribution of plasma SkM phenotypes by age or gender for the controls (p=0.22 and 0.86, respectively) (Supplement Table S1). Because it is generally thought that the younger the subject, the more likely it is that genetic factors contribute to VTE risk among younger subgroups. SkM phenotype among controls was not associated with age (Supplement Table S1A). So we analyzed 82 VTE patients aged ≤ 50 in whom hereditary factors are more likely [22]. When plasma SkM phenotypes were compared between subgroups of VTE patients with PE incidence and VTE patients without PE incidence (those with only isolated deep vein thrombosis (DVT)), Type III phenotype was associated with a lower PE occurrence in comparison to Type I and II phenotypes (p=0.017) (Table 1B). When the patient subgroup with isolated PE (i.e., PE patients without any DVT) was analyzed, 90% of isolated PE patients (9 of 10 patients) displayed the Type II phenotype. In comparison, only 34% of subjects in the control group and 29% of patients in the isolated DVT group displayed the Type II phenotype. Thus, isolated PE occurrence is associated with the Type II phenotype (p=0.0009 and p=0.0003 in comparison to the control and isolated DVT groups, respectively).

Table 1A.

Distribution of SkM plasma phenotypes for Caucasian venous thrombosis (VTE) patients and controls (age at a first episode of VTE ≤ 55 years old)

SkM phenotype	Control N (%)	VTE N (%)
I	53 (53.5)	49 (49.5)
II	34 (34.3)	35 (35.4)
III	12 (12.1)	15 (15.2)
total	99 (100)	99 (100)

Table 1B.

Distribution of SkM plasma phenotypes for Caucasian deep vein thrombosis (DVT) and pulmonary embolism (PE) (age of initial thrombosis occurrence ≤ 50 years)

SkM phenotype	control N (%)	isolated DVT N (%)		PE, N (%)
			DVT+PE	isolated PE	all PE
I	37 (55.2)	29 (52.7)	12 (70.6)	1 (10)	13 (48.1)
II	23 (34.3)	16 (29.1)	5 (29.4)	9 (90)	14 (51.9)
III	7 (10.4)	10 (18.2)	0 (0)	0 (0)	0 (0)
total	67 (100)	55 (100)	17 (100)	10 (100)	27 (100)

4. Discussion

4.1. SkM isoforms in plasma

We have developed new immunoblotting assays for detecting plasma SkM heavy chain antigen and showed, for the first time, that plasma contains multiple isoforms of SkM. Adapting the information about SkM heavy chain fragments, it appears that the 240–260 kDa bands correspond to the full-length heavy chain, indicating that the intact SkM heavy chain is circulating in plasma. The 160 kDa band and the 95 kDa band appear to correspond to heavy meromyosin (HMM) and the S1 fragment, respectively [8–11], indicating that HMM and the S1 fragment are also circulating in plasma. In addition, in some subjects there were higher MW bands of approximately 300 kDa and above which might be SkM complexes or aggregates.

4.2. The three plasma SkM phenotypes

Plasmas from individual blood donors displayed three distinct MW band patterns, here termed plasma SkM phenotypes (Types I, II and III) based on the presence or absence of different MW SkM heavy chain isoforms. Type I is the most common phenotype and contains two primary bands observed at 240 kDa and 160 kDa, and two lesser bands at 300kDa and 95 kDa. Type II is the second most common phenotype containing bands at ~300 kDa and 260 kDa. Type III, observed for 15% of control subjects, is a relatively rare phenotype containing essentially only one major SkM band at 95 kDa with negligible amounts of any other bands. Although it might be possible that these phenotypes were artificially caused by certain blood drawing circumstances or by timing of the immunoblot assays, our studies showed that phenotype assignment was the same for repeated blood draws from 18 subjects and was independent of time after blood drawing, making this concern very unlikely.

Skeletal muscle proteins are disassembled during metabolic turnover accompanied with rerelease of myosin by the ubiquitin-proteasome pathway and cysteine proteases [23–26]. The presence of multiple isoforms of SkM heavy chain in plasma indicates that the released SkM may be cleaved by proteases before release or during circulation in plasma. Chymotrypsin cleaves SkM to generate HMM and the S1 fragment [8, 9]. Lysosomal proteases of the cathepsin family or neutral proteases can also cleave the SkM heavy chain at several different sites, yielding HMM and S1 fragments with slightly different MWs than their chymotrypsin-cleaved counterparts [10, 12]. Although the detailed mechanisms for generating SkM isoforms observed in plasma need to be further investigated in the future, there may be differences in these proteolytic proteases for each SkM plasma phenotype. For instance, subjects with the Type II phenotype may lack the protease activity for generating HMM and the S1 fragment. In contrast, in Type III subjects, rapid proteolysis of intact SkM and HMM heavy chains could occur, producing only the S1 fragment in plasma. It is also possible that subjects with each SkM phenotype have different mutations in the SkM heavy chain which alter susceptibility to different proteolytic enzymes. The slightly different MWs of HMM and S1 species between Type I and Type II may support this mutation hypothesis. In any case, further investigation is warranted into the mechanisms for generating the circulating SkM isoforms.

4.3. The association of SkM plasma phenotypes with pulmonary embolism (PE)

There are no reports about the clinical relevance of circulating plasma SkM in thrombotic diseases, although full-length SkM is quite prothrombotic ex vivo [1]. PE is a very dangerous, life-threatening form of venous thromboembolism, and the third most common cause of cardiovascular death worldwide after stroke and heart attack [27, 28]. Identifying patients with high or low risk for PE occurrence is important in determining appropriate anticoagulant usage. Here, our data showed that one plasma SkM phenotype, namely Type III, is associated with lower occurrence of any PE in young adults (age ≤ 50 years). Furthermore, 90% of isolated PE patients displayed the Type II phenotype for which plasma consists mainly full-length SkM with minimum cleavage. The SkM plasma phenotype might become a useful biomarker for helping to identify risk and for determining the appropriate VTE prophylaxis; however, replications with larger cohorts, ideally prospective studies, are required to validate the findings of this pilot study.

The association of SkM plasma Type III phenotype with reduced occurrence of any PE or of Type II phenotype with isolated PE may be related to SkM’s procoagulant, anticoagulant, and/or profibrinolytic activities [1–5]. Type III phenotype plasma contains only the S1 fragment and not HMM or full-length SkM, suggesting that the S1 fragment is related to less procoagulant activity and/or to more fibrinolytic activity compared with full-length SkM. Type II phenotype’s association with isolated PE may imply that full length SkM’s procoagulant activity in plasma may augment thrombosis in the pulmonary vasculature. Clearly the associations of SkM isoforms of different compositions with SkM’s procoagulant, anticoagulant, and profibrinolytic activities need to be determined and then linked to more extensive SkM phenotype data in future studies.

Several clinical co-morbidities such as heart disease and cancer have been proposed as possible causes contributing to isolated PE [29, 30]. However, the clinical characteristics and risks of isolated PE for patients have not been adequately investigated. Since no obvious association of the Type II phenotype with DVT was found, the plasma SkM Type II phenotype may be a unique risk factor for isolated PE. Extensive replication studies, especially for associations with PE, seem very well warranted.

4.4. Limitations of this study

There are some limitations to this study. We may not have been able to adequately distinguish cardiac myosin from SkM due to the high homology of the monoclonal antibody’s recognition site [21]. However, plasma levels of cardiac myosin (<0.1 nmol/L) [31–35] are less than 1% of the average levels of SkM in plasma (4–20 nmol/L) [1, 11, 13]. Even when cardiac myosin levels are elevated as in the case of a myocardial infarction, it is less than 1 nmol/L [31–36]. Thus, cardiac myosin in plasma is not likely to influence the bands of SkM. For the finding about the association of SkM plasma phenotypes with PE, the small number of PE samples (27 PE subjects of 81 VTE subjects under 50 years old and 10 of those 27 with isolated PE) also limits the value of the clinical data here. The influence of background co-morbidities (e.g., hypertension, smoking, diabetes, peripheral vascular disease, etc.) related to VTE or PE and other confounding factors on the associations of SkM phenotypes with PE subtype could not be evaluated due to our small cohort. Determination of the associations between plasma SkM phenotypes with age was also limited by the lack of subjects aged >55 years old. In addition, physical activity is also a factor that might also affect the distribution of the phenotypes. Due to such limitations, replication of our findings, ideally in prospective studies, will be required using much larger PE and VTE cohorts to definitively establish clinical significance.

4.5. Other Implications

In this paper, we focused on the association of SkM plasma phenotypes with only thrombotic events. However, since SkM is a component of muscle, metabolic muscle diseases may also be associated with plasma SkM plasma phenotypes. Future clinical studies about the association of SkM plasma phenotypes with a variety of other diseases, including disease occurrence and/or prognosis, are warranted.

5. Conclusion

Immunoblotting detected different isoforms of plasma SkM with different MWs and characterized human subjects as having different SkM plasma phenotypes. This information about plasma SkM phenotypes may aid in evaluation of risks for increased blood clotting, including PE. Patients for whom such risk analyses may also be useful include those with trauma-induced coagulopathy and those with a hereditary or acquired bleeding/thrombotic disease.

Supplementary Material

1

Figure S1. SDS-PAGE immunoblots of the SkM antigen in plasma from Scripps Venous Thrombosis Registry subjects using a monoclonal antibody against the SkM heavy chain. Immunoblots of the SkM antigen in 220 plasmas (2 μL plasma /lane) (N=105 healthy controls and N=105 VTE patients from the Registry, plus healthy donors) were obtained. Phenotype assignment (I, II, or III) based on the distribution of bands is shown on blot under each lane where “C” or “V” denotes control or venous thrombosis subjects. Stars indicate Caucasian pulmonary embolism (PE) patients with a first episode at age ≤ 50 years (N=27). MW denotes molecular weights for markers from Bio-Rad Precision Plus Protein Standard.

Table S1. The distribution of SkM plasma phenotypes by (A) age or (B) gender in control group.

• Immunoblotting detects multiple skeletal muscle myosin (SkM) isoforms in human plasma

• Plasma contains SkM heavy chain, heavy meromyosin, and subfragment-1-like fragments

• Three principal phenotypes of SkM heavy chain antigen are found in human plasma

• SkM phenotypes might may be linked positively or negatively with pulmonary embolism

Abbreviations:

SkM

skeletal muscle myosin

HMM

heavy meromyosin

MW

molecular weight

VTE

venous thromboembolism

PE

pulmonary embolism

DVT

deep vein thrombosis